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.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_PCI_COMPONENT
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 PCI/PCI interrupt routing info messages:
64 acpi.debug_layer=0x400000
65 Enable AML "Debug" output, i.e., stores to the Debug
66 object while interpreting AML:
67 acpi.debug_layer=0xffffffff acpi.debug_level=0x2
68 Enable all messages related to ACPI hardware:
69 acpi.debug_layer=0x2 acpi.debug_level=0xffffffff
70
71 Some values produce so much output that the system is
72 unusable. The "log_buf_len" parameter may be useful
73 if you need to capture more output.
74
75 acpi_enforce_resources= [ACPI]
76 { strict | lax | no }
77 Check for resource conflicts between native drivers
78 and ACPI OperationRegions (SystemIO and SystemMemory
79 only). IO ports and memory declared in ACPI might be
80 used by the ACPI subsystem in arbitrary AML code and
81 can interfere with legacy drivers.
82 strict (default): access to resources claimed by ACPI
83 is denied; legacy drivers trying to access reserved
84 resources will fail to bind to device using them.
85 lax: access to resources claimed by ACPI is allowed;
86 legacy drivers trying to access reserved resources
87 will bind successfully but a warning message is logged.
88 no: ACPI OperationRegions are not marked as reserved,
89 no further checks are performed.
90
91 acpi_force_table_verification [HW,ACPI]
92 Enable table checksum verification during early stage.
93 By default, this is disabled due to x86 early mapping
94 size limitation.
95
96 acpi_irq_balance [HW,ACPI]
97 ACPI will balance active IRQs
98 default in APIC mode
99
100 acpi_irq_nobalance [HW,ACPI]
101 ACPI will not move active IRQs (default)
102 default in PIC mode
103
104 acpi_irq_isa= [HW,ACPI] If irq_balance, mark listed IRQs used by ISA
105 Format: <irq>,<irq>...
106
107 acpi_irq_pci= [HW,ACPI] If irq_balance, clear listed IRQs for
108 use by PCI
109 Format: <irq>,<irq>...
110
111 acpi_mask_gpe= [HW,ACPI]
112 Due to the existence of _Lxx/_Exx, some GPEs triggered
113 by unsupported hardware/firmware features can result in
114 GPE floodings that cannot be automatically disabled by
115 the GPE dispatcher.
116 This facility can be used to prevent such uncontrolled
117 GPE floodings.
118 Format: <byte>
119
120 acpi_no_auto_serialize [HW,ACPI]
121 Disable auto-serialization of AML methods
122 AML control methods that contain the opcodes to create
123 named objects will be marked as "Serialized" by the
124 auto-serialization feature.
125 This feature is enabled by default.
126 This option allows to turn off the feature.
127
128 acpi_no_memhotplug [ACPI] Disable memory hotplug. Useful for kdump
129 kernels.
130
131 acpi_no_static_ssdt [HW,ACPI]
132 Disable installation of static SSDTs at early boot time
133 By default, SSDTs contained in the RSDT/XSDT will be
134 installed automatically and they will appear under
135 /sys/firmware/acpi/tables.
136 This option turns off this feature.
137 Note that specifying this option does not affect
138 dynamic table installation which will install SSDT
139 tables to /sys/firmware/acpi/tables/dynamic.
140
141 acpi_no_watchdog [HW,ACPI,WDT]
142 Ignore the ACPI-based watchdog interface (WDAT) and let
143 a native driver control the watchdog device instead.
144
145 acpi_rsdp= [ACPI,EFI,KEXEC]
146 Pass the RSDP address to the kernel, mostly used
147 on machines running EFI runtime service to boot the
148 second kernel for kdump.
149
150 acpi_os_name= [HW,ACPI] Tell ACPI BIOS the name of the OS
151 Format: To spoof as Windows 98: ="Microsoft Windows"
152
153 acpi_rev_override [ACPI] Override the _REV object to return 5 (instead
154 of 2 which is mandated by ACPI 6) as the supported ACPI
155 specification revision (when using this switch, it may
156 be necessary to carry out a cold reboot _twice_ in a
157 row to make it take effect on the platform firmware).
158
159 acpi_osi= [HW,ACPI] Modify list of supported OS interface strings
160 acpi_osi="string1" # add string1
161 acpi_osi="!string2" # remove string2
162 acpi_osi=!* # remove all strings
163 acpi_osi=! # disable all built-in OS vendor
164 strings
165 acpi_osi=!! # enable all built-in OS vendor
166 strings
167 acpi_osi= # disable all strings
168
169 'acpi_osi=!' can be used in combination with single or
170 multiple 'acpi_osi="string1"' to support specific OS
171 vendor string(s). Note that such command can only
172 affect the default state of the OS vendor strings, thus
173 it cannot affect the default state of the feature group
174 strings and the current state of the OS vendor strings,
175 specifying it multiple times through kernel command line
176 is meaningless. This command is useful when one do not
177 care about the state of the feature group strings which
178 should be controlled by the OSPM.
179 Examples:
180 1. 'acpi_osi=! acpi_osi="Windows 2000"' is equivalent
181 to 'acpi_osi="Windows 2000" acpi_osi=!', they all
182 can make '_OSI("Windows 2000")' TRUE.
183
184 'acpi_osi=' cannot be used in combination with other
185 'acpi_osi=' command lines, the _OSI method will not
186 exist in the ACPI namespace. NOTE that such command can
187 only affect the _OSI support state, thus specifying it
188 multiple times through kernel command line is also
189 meaningless.
190 Examples:
191 1. 'acpi_osi=' can make 'CondRefOf(_OSI, Local1)'
192 FALSE.
193
194 'acpi_osi=!*' can be used in combination with single or
195 multiple 'acpi_osi="string1"' to support specific
196 string(s). Note that such command can affect the
197 current state of both the OS vendor strings and the
198 feature group strings, thus specifying it multiple times
199 through kernel command line is meaningful. But it may
200 still not able to affect the final state of a string if
201 there are quirks related to this string. This command
202 is useful when one want to control the state of the
203 feature group strings to debug BIOS issues related to
204 the OSPM features.
205 Examples:
206 1. 'acpi_osi="Module Device" acpi_osi=!*' can make
207 '_OSI("Module Device")' FALSE.
208 2. 'acpi_osi=!* acpi_osi="Module Device"' can make
209 '_OSI("Module Device")' TRUE.
210 3. 'acpi_osi=! acpi_osi=!* acpi_osi="Windows 2000"' is
211 equivalent to
212 'acpi_osi=!* acpi_osi=! acpi_osi="Windows 2000"'
213 and
214 'acpi_osi=!* acpi_osi="Windows 2000" acpi_osi=!',
215 they all will make '_OSI("Windows 2000")' TRUE.
216
217 acpi_pm_good [X86]
218 Override the pmtimer bug detection: force the kernel
219 to assume that this machine's pmtimer latches its value
220 and always returns good values.
221
222 acpi_sci= [HW,ACPI] ACPI System Control Interrupt trigger mode
223 Format: { level | edge | high | low }
224
225 acpi_skip_timer_override [HW,ACPI]
226 Recognize and ignore IRQ0/pin2 Interrupt Override.
227 For broken nForce2 BIOS resulting in XT-PIC timer.
228
229 acpi_sleep= [HW,ACPI] Sleep options
230 Format: { s3_bios, s3_mode, s3_beep, s4_nohwsig,
231 old_ordering, nonvs, sci_force_enable, nobl }
232 See Documentation/power/video.rst for information on
233 s3_bios and s3_mode.
234 s3_beep is for debugging; it makes the PC's speaker beep
235 as soon as the kernel's real-mode entry point is called.
236 s4_nohwsig prevents ACPI hardware signature from being
237 used during resume from hibernation.
238 old_ordering causes the ACPI 1.0 ordering of the _PTS
239 control method, with respect to putting devices into
240 low power states, to be enforced (the ACPI 2.0 ordering
241 of _PTS is used by default).
242 nonvs prevents the kernel from saving/restoring the
243 ACPI NVS memory during suspend/hibernation and resume.
244 sci_force_enable causes the kernel to set SCI_EN directly
245 on resume from S1/S3 (which is against the ACPI spec,
246 but some broken systems don't work without it).
247 nobl causes the internal blacklist of systems known to
248 behave incorrectly in some ways with respect to system
249 suspend and resume to be ignored (use wisely).
250
251 acpi_use_timer_override [HW,ACPI]
252 Use timer override. For some broken Nvidia NF5 boards
253 that require a timer override, but don't have HPET
254
255 add_efi_memmap [EFI; X86] Include EFI memory map in
256 kernel's map of available physical RAM.
257
258 agp= [AGP]
259 { off | try_unsupported }
260 off: disable AGP support
261 try_unsupported: try to drive unsupported chipsets
262 (may crash computer or cause data corruption)
263
264 ALSA [HW,ALSA]
265 See Documentation/sound/alsa-configuration.rst
266
267 alignment= [KNL,ARM]
268 Allow the default userspace alignment fault handler
269 behaviour to be specified. Bit 0 enables warnings,
270 bit 1 enables fixups, and bit 2 sends a segfault.
271
272 align_va_addr= [X86-64]
273 Align virtual addresses by clearing slice [14:12] when
274 allocating a VMA at process creation time. This option
275 gives you up to 3% performance improvement on AMD F15h
276 machines (where it is enabled by default) for a
277 CPU-intensive style benchmark, and it can vary highly in
278 a microbenchmark depending on workload and compiler.
279
280 32: only for 32-bit processes
281 64: only for 64-bit processes
282 on: enable for both 32- and 64-bit processes
283 off: disable for both 32- and 64-bit processes
284
285 alloc_snapshot [FTRACE]
286 Allocate the ftrace snapshot buffer on boot up when the
287 main buffer is allocated. This is handy if debugging
288 and you need to use tracing_snapshot() on boot up, and
289 do not want to use tracing_snapshot_alloc() as it needs
290 to be done where GFP_KERNEL allocations are allowed.
291
292 amd_iommu= [HW,X86-64]
293 Pass parameters to the AMD IOMMU driver in the system.
294 Possible values are:
295 fullflush - enable flushing of IO/TLB entries when
296 they are unmapped. Otherwise they are
297 flushed before they will be reused, which
298 is a lot of faster
299 off - do not initialize any AMD IOMMU found in
300 the system
301 force_isolation - Force device isolation for all
302 devices. The IOMMU driver is not
303 allowed anymore to lift isolation
304 requirements as needed. This option
305 does not override iommu=pt
306
307 amd_iommu_dump= [HW,X86-64]
308 Enable AMD IOMMU driver option to dump the ACPI table
309 for AMD IOMMU. With this option enabled, AMD IOMMU
310 driver will print ACPI tables for AMD IOMMU during
311 IOMMU initialization.
312
313 amd_iommu_intr= [HW,X86-64]
314 Specifies one of the following AMD IOMMU interrupt
315 remapping modes:
316 legacy - Use legacy interrupt remapping mode.
317 vapic - Use virtual APIC mode, which allows IOMMU
318 to inject interrupts directly into guest.
319 This mode requires kvm-amd.avic=1.
320 (Default when IOMMU HW support is present.)
321
322 amijoy.map= [HW,JOY] Amiga joystick support
323 Map of devices attached to JOY0DAT and JOY1DAT
324 Format: <a>,<b>
325 See also Documentation/input/joydev/joystick.rst
326
327 analog.map= [HW,JOY] Analog joystick and gamepad support
328 Specifies type or capabilities of an analog joystick
329 connected to one of 16 gameports
330 Format: <type1>,<type2>,..<type16>
331
332 apc= [HW,SPARC]
333 Power management functions (SPARCstation-4/5 + deriv.)
334 Format: noidle
335 Disable APC CPU standby support. SPARCstation-Fox does
336 not play well with APC CPU idle - disable it if you have
337 APC and your system crashes randomly.
338
339 apic= [APIC,X86] Advanced Programmable Interrupt Controller
340 Change the output verbosity while booting
341 Format: { quiet (default) | verbose | debug }
342 Change the amount of debugging information output
343 when initialising the APIC and IO-APIC components.
344 For X86-32, this can also be used to specify an APIC
345 driver name.
346 Format: apic=driver_name
347 Examples: apic=bigsmp
348
349 apic_extnmi= [APIC,X86] External NMI delivery setting
350 Format: { bsp (default) | all | none }
351 bsp: External NMI is delivered only to CPU 0
352 all: External NMIs are broadcast to all CPUs as a
353 backup of CPU 0
354 none: External NMI is masked for all CPUs. This is
355 useful so that a dump capture kernel won't be
356 shot down by NMI
357
358 autoconf= [IPV6]
359 See Documentation/networking/ipv6.rst.
360
361 show_lapic= [APIC,X86] Advanced Programmable Interrupt Controller
362 Limit apic dumping. The parameter defines the maximal
363 number of local apics being dumped. Also it is possible
364 to set it to "all" by meaning -- no limit here.
365 Format: { 1 (default) | 2 | ... | all }.
366 The parameter valid if only apic=debug or
367 apic=verbose is specified.
368 Example: apic=debug show_lapic=all
369
370 apm= [APM] Advanced Power Management
371 See header of arch/x86/kernel/apm_32.c.
372
373 arcrimi= [HW,NET] ARCnet - "RIM I" (entirely mem-mapped) cards
374 Format: <io>,<irq>,<nodeID>
375
376 ataflop= [HW,M68k]
377
378 atarimouse= [HW,MOUSE] Atari Mouse
379
380 atkbd.extra= [HW] Enable extra LEDs and keys on IBM RapidAccess,
381 EzKey and similar keyboards
382
383 atkbd.reset= [HW] Reset keyboard during initialization
384
385 atkbd.set= [HW] Select keyboard code set
386 Format: <int> (2 = AT (default), 3 = PS/2)
387
388 atkbd.scroll= [HW] Enable scroll wheel on MS Office and similar
389 keyboards
390
391 atkbd.softraw= [HW] Choose between synthetic and real raw mode
392 Format: <bool> (0 = real, 1 = synthetic (default))
393
394 atkbd.softrepeat= [HW]
395 Use software keyboard repeat
396
397 audit= [KNL] Enable the audit sub-system
398 Format: { "0" | "1" | "off" | "on" }
399 0 | off - kernel audit is disabled and can not be
400 enabled until the next reboot
401 unset - kernel audit is initialized but disabled and
402 will be fully enabled by the userspace auditd.
403 1 | on - kernel audit is initialized and partially
404 enabled, storing at most audit_backlog_limit
405 messages in RAM until it is fully enabled by the
406 userspace auditd.
407 Default: unset
408
409 audit_backlog_limit= [KNL] Set the audit queue size limit.
410 Format: <int> (must be >=0)
411 Default: 64
412
413 bau= [X86_UV] Enable the BAU on SGI UV. The default
414 behavior is to disable the BAU (i.e. bau=0).
415 Format: { "0" | "1" }
416 0 - Disable the BAU.
417 1 - Enable the BAU.
418 unset - Disable the BAU.
419
420 baycom_epp= [HW,AX25]
421 Format: <io>,<mode>
422
423 baycom_par= [HW,AX25] BayCom Parallel Port AX.25 Modem
424 Format: <io>,<mode>
425 See header of drivers/net/hamradio/baycom_par.c.
426
427 baycom_ser_fdx= [HW,AX25]
428 BayCom Serial Port AX.25 Modem (Full Duplex Mode)
429 Format: <io>,<irq>,<mode>[,<baud>]
430 See header of drivers/net/hamradio/baycom_ser_fdx.c.
431
432 baycom_ser_hdx= [HW,AX25]
433 BayCom Serial Port AX.25 Modem (Half Duplex Mode)
434 Format: <io>,<irq>,<mode>
435 See header of drivers/net/hamradio/baycom_ser_hdx.c.
436
437 blkdevparts= Manual partition parsing of block device(s) for
438 embedded devices based on command line input.
439 See Documentation/block/cmdline-partition.rst
440
441 boot_delay= Milliseconds to delay each printk during boot.
442 Values larger than 10 seconds (10000) are changed to
443 no delay (0).
444 Format: integer
445
446 bootconfig [KNL]
447 Extended command line options can be added to an initrd
448 and this will cause the kernel to look for it.
449
450 See Documentation/admin-guide/bootconfig.rst
451
452 bert_disable [ACPI]
453 Disable BERT OS support on buggy BIOSes.
454
455 bgrt_disable [ACPI][X86]
456 Disable BGRT to avoid flickering OEM logo.
457
458 bttv.card= [HW,V4L] bttv (bt848 + bt878 based grabber cards)
459 bttv.radio= Most important insmod options are available as
460 kernel args too.
461 bttv.pll= See Documentation/admin-guide/media/bttv.rst
462 bttv.tuner=
463
464 bulk_remove=off [PPC] This parameter disables the use of the pSeries
465 firmware feature for flushing multiple hpte entries
466 at a time.
467
468 c101= [NET] Moxa C101 synchronous serial card
469
470 cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection.
471 Sometimes CPU hardware bugs make them report the cache
472 size incorrectly. The kernel will attempt work arounds
473 to fix known problems, but for some CPUs it is not
474 possible to determine what the correct size should be.
475 This option provides an override for these situations.
476
477 carrier_timeout=
478 [NET] Specifies amount of time (in seconds) that
479 the kernel should wait for a network carrier. By default
480 it waits 120 seconds.
481
482 ca_keys= [KEYS] This parameter identifies a specific key(s) on
483 the system trusted keyring to be used for certificate
484 trust validation.
485 format: { id:<keyid> | builtin }
486
487 cca= [MIPS] Override the kernel pages' cache coherency
488 algorithm. Accepted values range from 0 to 7
489 inclusive. See arch/mips/include/asm/pgtable-bits.h
490 for platform specific values (SB1, Loongson3 and
491 others).
492
493 ccw_timeout_log [S390]
494 See Documentation/s390/common_io.rst for details.
495
496 cgroup_disable= [KNL] Disable a particular controller
497 Format: {name of the controller(s) to disable}
498 The effects of cgroup_disable=foo are:
499 - foo isn't auto-mounted if you mount all cgroups in
500 a single hierarchy
501 - foo isn't visible as an individually mountable
502 subsystem
503 {Currently only "memory" controller deal with this and
504 cut the overhead, others just disable the usage. So
505 only cgroup_disable=memory is actually worthy}
506
507 cgroup_no_v1= [KNL] Disable cgroup controllers and named hierarchies in v1
508 Format: { { controller | "all" | "named" }
509 [,{ controller | "all" | "named" }...] }
510 Like cgroup_disable, but only applies to cgroup v1;
511 the blacklisted controllers remain available in cgroup2.
512 "all" blacklists all controllers and "named" disables
513 named mounts. Specifying both "all" and "named" disables
514 all v1 hierarchies.
515
516 cgroup.memory= [KNL] Pass options to the cgroup memory controller.
517 Format: <string>
518 nosocket -- Disable socket memory accounting.
519 nokmem -- Disable kernel memory accounting.
520
521 checkreqprot [SELINUX] Set initial checkreqprot flag value.
522 Format: { "0" | "1" }
523 See security/selinux/Kconfig help text.
524 0 -- check protection applied by kernel (includes
525 any implied execute protection).
526 1 -- check protection requested by application.
527 Default value is set via a kernel config option.
528 Value can be changed at runtime via
529 /sys/fs/selinux/checkreqprot.
530 Setting checkreqprot to 1 is deprecated.
531
532 cio_ignore= [S390]
533 See Documentation/s390/common_io.rst for details.
534 clk_ignore_unused
535 [CLK]
536 Prevents the clock framework from automatically gating
537 clocks that have not been explicitly enabled by a Linux
538 device driver but are enabled in hardware at reset or
539 by the bootloader/firmware. Note that this does not
540 force such clocks to be always-on nor does it reserve
541 those clocks in any way. This parameter is useful for
542 debug and development, but should not be needed on a
543 platform with proper driver support. For more
544 information, see Documentation/driver-api/clk.rst.
545
546 clock= [BUGS=X86-32, HW] gettimeofday clocksource override.
547 [Deprecated]
548 Forces specified clocksource (if available) to be used
549 when calculating gettimeofday(). If specified
550 clocksource is not available, it defaults to PIT.
551 Format: { pit | tsc | cyclone | pmtmr }
552
553 clocksource= Override the default clocksource
554 Format: <string>
555 Override the default clocksource and use the clocksource
556 with the name specified.
557 Some clocksource names to choose from, depending on
558 the platform:
559 [all] jiffies (this is the base, fallback clocksource)
560 [ACPI] acpi_pm
561 [ARM] imx_timer1,OSTS,netx_timer,mpu_timer2,
562 pxa_timer,timer3,32k_counter,timer0_1
563 [X86-32] pit,hpet,tsc;
564 scx200_hrt on Geode; cyclone on IBM x440
565 [MIPS] MIPS
566 [PARISC] cr16
567 [S390] tod
568 [SH] SuperH
569 [SPARC64] tick
570 [X86-64] hpet,tsc
571
572 clocksource.arm_arch_timer.evtstrm=
573 [ARM,ARM64]
574 Format: <bool>
575 Enable/disable the eventstream feature of the ARM
576 architected timer so that code using WFE-based polling
577 loops can be debugged more effectively on production
578 systems.
579
580 clearcpuid=BITNUM [X86]
581 Disable CPUID feature X for the kernel. See
582 arch/x86/include/asm/cpufeatures.h for the valid bit
583 numbers. Note the Linux specific bits are not necessarily
584 stable over kernel options, but the vendor specific
585 ones should be.
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 cma=nn[MG]@[start[MG][-end[MG]]]
594 [ARM,X86,KNL]
595 Sets the size of kernel global memory area for
596 contiguous memory allocations and optionally the
597 placement constraint by the physical address range of
598 memory allocations. A value of 0 disables CMA
599 altogether. For more information, see
600 include/linux/dma-contiguous.h
601
602 cmo_free_hint= [PPC] Format: { yes | no }
603 Specify whether pages are marked as being inactive
604 when they are freed. This is used in CMO environments
605 to determine OS memory pressure for page stealing by
606 a hypervisor.
607 Default: yes
608
609 coherent_pool=nn[KMG] [ARM,KNL]
610 Sets the size of memory pool for coherent, atomic dma
611 allocations, by default set to 256K.
612
613 com20020= [HW,NET] ARCnet - COM20020 chipset
614 Format:
615 <io>[,<irq>[,<nodeID>[,<backplane>[,<ckp>[,<timeout>]]]]]
616
617 com90io= [HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers)
618 Format: <io>[,<irq>]
619
620 com90xx= [HW,NET]
621 ARCnet - COM90xx chipset (memory-mapped buffers)
622 Format: <io>[,<irq>[,<memstart>]]
623
624 condev= [HW,S390] console device
625 conmode=
626
627 console= [KNL] Output console device and options.
628
629 tty<n> Use the virtual console device <n>.
630
631 ttyS<n>[,options]
632 ttyUSB0[,options]
633 Use the specified serial port. The options are of
634 the form "bbbbpnf", where "bbbb" is the baud rate,
635 "p" is parity ("n", "o", or "e"), "n" is number of
636 bits, and "f" is flow control ("r" for RTS or
637 omit it). Default is "9600n8".
638
639 See Documentation/admin-guide/serial-console.rst for more
640 information. See
641 Documentation/networking/netconsole.rst for an
642 alternative.
643
644 uart[8250],io,<addr>[,options]
645 uart[8250],mmio,<addr>[,options]
646 uart[8250],mmio16,<addr>[,options]
647 uart[8250],mmio32,<addr>[,options]
648 uart[8250],0x<addr>[,options]
649 Start an early, polled-mode console on the 8250/16550
650 UART at the specified I/O port or MMIO address,
651 switching to the matching ttyS device later.
652 MMIO inter-register address stride is either 8-bit
653 (mmio), 16-bit (mmio16), or 32-bit (mmio32).
654 If none of [io|mmio|mmio16|mmio32], <addr> is assumed
655 to be equivalent to 'mmio'. 'options' are specified in
656 the same format described for ttyS above; if unspecified,
657 the h/w is not re-initialized.
658
659 hvc<n> Use the hypervisor console device <n>. This is for
660 both Xen and PowerPC hypervisors.
661
662 If the device connected to the port is not a TTY but a braille
663 device, prepend "brl," before the device type, for instance
664 console=brl,ttyS0
665 For now, only VisioBraille is supported.
666
667 console_msg_format=
668 [KNL] Change console messages format
669 default
670 By default we print messages on consoles in
671 "[time stamp] text\n" format (time stamp may not be
672 printed, depending on CONFIG_PRINTK_TIME or
673 `printk_time' param).
674 syslog
675 Switch to syslog format: "<%u>[time stamp] text\n"
676 IOW, each message will have a facility and loglevel
677 prefix. The format is similar to one used by syslog()
678 syscall, or to executing "dmesg -S --raw" or to reading
679 from /proc/kmsg.
680
681 consoleblank= [KNL] The console blank (screen saver) timeout in
682 seconds. A value of 0 disables the blank timer.
683 Defaults to 0.
684
685 coredump_filter=
686 [KNL] Change the default value for
687 /proc/<pid>/coredump_filter.
688 See also Documentation/filesystems/proc.rst.
689
690 coresight_cpu_debug.enable
691 [ARM,ARM64]
692 Format: <bool>
693 Enable/disable the CPU sampling based debugging.
694 0: default value, disable debugging
695 1: enable debugging at boot time
696
697 cpuidle.off=1 [CPU_IDLE]
698 disable the cpuidle sub-system
699
700 cpuidle.governor=
701 [CPU_IDLE] Name of the cpuidle governor to use.
702
703 cpufreq.off=1 [CPU_FREQ]
704 disable the cpufreq sub-system
705
706 cpufreq.default_governor=
707 [CPU_FREQ] Name of the default cpufreq governor or
708 policy to use. This governor must be registered in the
709 kernel before the cpufreq driver probes.
710
711 cpu_init_udelay=N
712 [X86] Delay for N microsec between assert and de-assert
713 of APIC INIT to start processors. This delay occurs
714 on every CPU online, such as boot, and resume from suspend.
715 Default: 10000
716
717 cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver
718 Format:
719 <first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>]
720
721 crashkernel=size[KMG][@offset[KMG]]
722 [KNL] Using kexec, Linux can switch to a 'crash kernel'
723 upon panic. This parameter reserves the physical
724 memory region [offset, offset + size] for that kernel
725 image. If '@offset' is omitted, then a suitable offset
726 is selected automatically.
727 [KNL, X86-64] Select a region under 4G first, and
728 fall back to reserve region above 4G when '@offset'
729 hasn't been specified.
730 See Documentation/admin-guide/kdump/kdump.rst for further details.
731
732 crashkernel=range1:size1[,range2:size2,...][@offset]
733 [KNL] Same as above, but depends on the memory
734 in the running system. The syntax of range is
735 start-[end] where start and end are both
736 a memory unit (amount[KMG]). See also
737 Documentation/admin-guide/kdump/kdump.rst for an example.
738
739 crashkernel=size[KMG],high
740 [KNL, X86-64] range could be above 4G. Allow kernel
741 to allocate physical memory region from top, so could
742 be above 4G if system have more than 4G ram installed.
743 Otherwise memory region will be allocated below 4G, if
744 available.
745 It will be ignored if crashkernel=X is specified.
746 crashkernel=size[KMG],low
747 [KNL, X86-64] range under 4G. When crashkernel=X,high
748 is passed, kernel could allocate physical memory region
749 above 4G, that cause second kernel crash on system
750 that require some amount of low memory, e.g. swiotlb
751 requires at least 64M+32K low memory, also enough extra
752 low memory is needed to make sure DMA buffers for 32-bit
753 devices won't run out. Kernel would try to allocate at
754 at least 256M below 4G automatically.
755 This one let user to specify own low range under 4G
756 for second kernel instead.
757 0: to disable low allocation.
758 It will be ignored when crashkernel=X,high is not used
759 or memory reserved is below 4G.
760
761 cryptomgr.notests
762 [KNL] Disable crypto self-tests
763
764 cs89x0_dma= [HW,NET]
765 Format: <dma>
766
767 cs89x0_media= [HW,NET]
768 Format: { rj45 | aui | bnc }
769
770 dasd= [HW,NET]
771 See header of drivers/s390/block/dasd_devmap.c.
772
773 db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port
774 (one device per port)
775 Format: <port#>,<type>
776 See also Documentation/input/devices/joystick-parport.rst
777
778 ddebug_query= [KNL,DYNAMIC_DEBUG] Enable debug messages at early boot
779 time. See
780 Documentation/admin-guide/dynamic-debug-howto.rst for
781 details. Deprecated, see dyndbg.
782
783 debug [KNL] Enable kernel debugging (events log level).
784
785 debug_boot_weak_hash
786 [KNL] Enable printing [hashed] pointers early in the
787 boot sequence. If enabled, we use a weak hash instead
788 of siphash to hash pointers. Use this option if you are
789 seeing instances of '(___ptrval___)') and need to see a
790 value (hashed pointer) instead. Cryptographically
791 insecure, please do not use on production kernels.
792
793 debug_locks_verbose=
794 [KNL] verbose self-tests
795 Format=<0|1>
796 Print debugging info while doing the locking API
797 self-tests.
798 We default to 0 (no extra messages), setting it to
799 1 will print _a lot_ more information - normally
800 only useful to kernel developers.
801
802 debug_objects [KNL] Enable object debugging
803
804 no_debug_objects
805 [KNL] Disable object debugging
806
807 debug_guardpage_minorder=
808 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this
809 parameter allows control of the order of pages that will
810 be intentionally kept free (and hence protected) by the
811 buddy allocator. Bigger value increase the probability
812 of catching random memory corruption, but reduce the
813 amount of memory for normal system use. The maximum
814 possible value is MAX_ORDER/2. Setting this parameter
815 to 1 or 2 should be enough to identify most random
816 memory corruption problems caused by bugs in kernel or
817 driver code when a CPU writes to (or reads from) a
818 random memory location. Note that there exists a class
819 of memory corruptions problems caused by buggy H/W or
820 F/W or by drivers badly programing DMA (basically when
821 memory is written at bus level and the CPU MMU is
822 bypassed) which are not detectable by
823 CONFIG_DEBUG_PAGEALLOC, hence this option will not help
824 tracking down these problems.
825
826 debug_pagealloc=
827 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this parameter
828 enables the feature at boot time. By default, it is
829 disabled and the system will work mostly the same as a
830 kernel built without CONFIG_DEBUG_PAGEALLOC.
831 Note: to get most of debug_pagealloc error reports, it's
832 useful to also enable the page_owner functionality.
833 on: enable the feature
834
835 debugfs= [KNL] This parameter enables what is exposed to userspace
836 and debugfs internal clients.
837 Format: { on, no-mount, off }
838 on: All functions are enabled.
839 no-mount:
840 Filesystem is not registered but kernel clients can
841 access APIs and a crashkernel can be used to read
842 its content. There is nothing to mount.
843 off: Filesystem is not registered and clients
844 get a -EPERM as result when trying to register files
845 or directories within debugfs.
846 This is equivalent of the runtime functionality if
847 debugfs was not enabled in the kernel at all.
848 Default value is set in build-time with a kernel configuration.
849
850 debugpat [X86] Enable PAT debugging
851
852 decnet.addr= [HW,NET]
853 Format: <area>[,<node>]
854 See also Documentation/networking/decnet.rst.
855
856 default_hugepagesz=
857 [HW] The size of the default HugeTLB page. This is
858 the size represented by the legacy /proc/ hugepages
859 APIs. In addition, this is the default hugetlb size
860 used for shmget(), mmap() and mounting hugetlbfs
861 filesystems. If not specified, defaults to the
862 architecture's default huge page size. Huge page
863 sizes are architecture dependent. See also
864 Documentation/admin-guide/mm/hugetlbpage.rst.
865 Format: size[KMG]
866
867 deferred_probe_timeout=
868 [KNL] Debugging option to set a timeout in seconds for
869 deferred probe to give up waiting on dependencies to
870 probe. Only specific dependencies (subsystems or
871 drivers) that have opted in will be ignored. A timeout of 0
872 will timeout at the end of initcalls. This option will also
873 dump out devices still on the deferred probe list after
874 retrying.
875
876 dfltcc= [HW,S390]
877 Format: { on | off | def_only | inf_only | always }
878 on: s390 zlib hardware support for compression on
879 level 1 and decompression (default)
880 off: No s390 zlib hardware support
881 def_only: s390 zlib hardware support for deflate
882 only (compression on level 1)
883 inf_only: s390 zlib hardware support for inflate
884 only (decompression)
885 always: Same as 'on' but ignores the selected compression
886 level always using hardware support (used for debugging)
887
888 dhash_entries= [KNL]
889 Set number of hash buckets for dentry cache.
890
891 disable_1tb_segments [PPC]
892 Disables the use of 1TB hash page table segments. This
893 causes the kernel to fall back to 256MB segments which
894 can be useful when debugging issues that require an SLB
895 miss to occur.
896
897 stress_slb [PPC]
898 Limits the number of kernel SLB entries, and flushes
899 them frequently to increase the rate of SLB faults
900 on kernel addresses.
901
902 disable= [IPV6]
903 See Documentation/networking/ipv6.rst.
904
905 hardened_usercopy=
906 [KNL] Under CONFIG_HARDENED_USERCOPY, whether
907 hardening is enabled for this boot. Hardened
908 usercopy checking is used to protect the kernel
909 from reading or writing beyond known memory
910 allocation boundaries as a proactive defense
911 against bounds-checking flaws in the kernel's
912 copy_to_user()/copy_from_user() interface.
913 on Perform hardened usercopy checks (default).
914 off Disable hardened usercopy checks.
915
916 disable_radix [PPC]
917 Disable RADIX MMU mode on POWER9
918
919 radix_hcall_invalidate=on [PPC/PSERIES]
920 Disable RADIX GTSE feature and use hcall for TLB
921 invalidate.
922
923 disable_tlbie [PPC]
924 Disable TLBIE instruction. Currently does not work
925 with KVM, with HASH MMU, or with coherent accelerators.
926
927 disable_cpu_apicid= [X86,APIC,SMP]
928 Format: <int>
929 The number of initial APIC ID for the
930 corresponding CPU to be disabled at boot,
931 mostly used for the kdump 2nd kernel to
932 disable BSP to wake up multiple CPUs without
933 causing system reset or hang due to sending
934 INIT from AP to BSP.
935
936 perf_v4_pmi= [X86,INTEL]
937 Format: <bool>
938 Disable Intel PMU counter freezing feature.
939 The feature only exists starting from
940 Arch Perfmon v4 (Skylake and newer).
941
942 disable_ddw [PPC/PSERIES]
943 Disable Dynamic DMA Window support. Use this if
944 to workaround buggy firmware.
945
946 disable_ipv6= [IPV6]
947 See Documentation/networking/ipv6.rst.
948
949 disable_mtrr_cleanup [X86]
950 The kernel tries to adjust MTRR layout from continuous
951 to discrete, to make X server driver able to add WB
952 entry later. This parameter disables that.
953
954 disable_mtrr_trim [X86, Intel and AMD only]
955 By default the kernel will trim any uncacheable
956 memory out of your available memory pool based on
957 MTRR settings. This parameter disables that behavior,
958 possibly causing your machine to run very slowly.
959
960 disable_timer_pin_1 [X86]
961 Disable PIN 1 of APIC timer
962 Can be useful to work around chipset bugs.
963
964 dis_ucode_ldr [X86] Disable the microcode loader.
965
966 dma_debug=off If the kernel is compiled with DMA_API_DEBUG support,
967 this option disables the debugging code at boot.
968
969 dma_debug_entries=<number>
970 This option allows to tune the number of preallocated
971 entries for DMA-API debugging code. One entry is
972 required per DMA-API allocation. Use this if the
973 DMA-API debugging code disables itself because the
974 architectural default is too low.
975
976 dma_debug_driver=<driver_name>
977 With this option the DMA-API debugging driver
978 filter feature can be enabled at boot time. Just
979 pass the driver to filter for as the parameter.
980 The filter can be disabled or changed to another
981 driver later using sysfs.
982
983 driver_async_probe= [KNL]
984 List of driver names to be probed asynchronously.
985 Format: <driver_name1>,<driver_name2>...
986
987 drm.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>]
988 Broken monitors, graphic adapters, KVMs and EDIDless
989 panels may send no or incorrect EDID data sets.
990 This parameter allows to specify an EDID data sets
991 in the /lib/firmware directory that are used instead.
992 Generic built-in EDID data sets are used, if one of
993 edid/1024x768.bin, edid/1280x1024.bin,
994 edid/1680x1050.bin, or edid/1920x1080.bin is given
995 and no file with the same name exists. Details and
996 instructions how to build your own EDID data are
997 available in Documentation/admin-guide/edid.rst. An EDID
998 data set will only be used for a particular connector,
999 if its name and a colon are prepended to the EDID
1000 name. Each connector may use a unique EDID data
1001 set by separating the files with a comma. An EDID
1002 data set with no connector name will be used for
1003 any connectors not explicitly specified.
1004
1005 dscc4.setup= [NET]
1006
1007 dt_cpu_ftrs= [PPC]
1008 Format: {"off" | "known"}
1009 Control how the dt_cpu_ftrs device-tree binding is
1010 used for CPU feature discovery and setup (if it
1011 exists).
1012 off: Do not use it, fall back to legacy cpu table.
1013 known: Do not pass through unknown features to guests
1014 or userspace, only those that the kernel is aware of.
1015
1016 dump_apple_properties [X86]
1017 Dump name and content of EFI device properties on
1018 x86 Macs. Useful for driver authors to determine
1019 what data is available or for reverse-engineering.
1020
1021 dyndbg[="val"] [KNL,DYNAMIC_DEBUG]
1022 module.dyndbg[="val"]
1023 Enable debug messages at boot time. See
1024 Documentation/admin-guide/dynamic-debug-howto.rst
1025 for details.
1026
1027 nopku [X86] Disable Memory Protection Keys CPU feature found
1028 in some Intel CPUs.
1029
1030 module.async_probe [KNL]
1031 Enable asynchronous probe on this module.
1032
1033 early_ioremap_debug [KNL]
1034 Enable debug messages in early_ioremap support. This
1035 is useful for tracking down temporary early mappings
1036 which are not unmapped.
1037
1038 earlycon= [KNL] Output early console device and options.
1039
1040 When used with no options, the early console is
1041 determined by stdout-path property in device tree's
1042 chosen node or the ACPI SPCR table if supported by
1043 the platform.
1044
1045 cdns,<addr>[,options]
1046 Start an early, polled-mode console on a Cadence
1047 (xuartps) serial port at the specified address. Only
1048 supported option is baud rate. If baud rate is not
1049 specified, the serial port must already be setup and
1050 configured.
1051
1052 uart[8250],io,<addr>[,options]
1053 uart[8250],mmio,<addr>[,options]
1054 uart[8250],mmio32,<addr>[,options]
1055 uart[8250],mmio32be,<addr>[,options]
1056 uart[8250],0x<addr>[,options]
1057 Start an early, polled-mode console on the 8250/16550
1058 UART at the specified I/O port or MMIO address.
1059 MMIO inter-register address stride is either 8-bit
1060 (mmio) or 32-bit (mmio32 or mmio32be).
1061 If none of [io|mmio|mmio32|mmio32be], <addr> is assumed
1062 to be equivalent to 'mmio'. 'options' are specified
1063 in the same format described for "console=ttyS<n>"; if
1064 unspecified, the h/w is not initialized.
1065
1066 pl011,<addr>
1067 pl011,mmio32,<addr>
1068 Start an early, polled-mode console on a pl011 serial
1069 port at the specified address. The pl011 serial port
1070 must already be setup and configured. Options are not
1071 yet supported. If 'mmio32' is specified, then only
1072 the driver will use only 32-bit accessors to read/write
1073 the device registers.
1074
1075 meson,<addr>
1076 Start an early, polled-mode console on a meson serial
1077 port at the specified address. The serial port must
1078 already be setup and configured. Options are not yet
1079 supported.
1080
1081 msm_serial,<addr>
1082 Start an early, polled-mode console on an msm serial
1083 port at the specified address. The serial port
1084 must already be setup and configured. Options are not
1085 yet supported.
1086
1087 msm_serial_dm,<addr>
1088 Start an early, polled-mode console on an msm serial
1089 dm port at the specified address. The serial port
1090 must already be setup and configured. Options are not
1091 yet supported.
1092
1093 owl,<addr>
1094 Start an early, polled-mode console on a serial port
1095 of an Actions Semi SoC, such as S500 or S900, at the
1096 specified address. The serial port must already be
1097 setup and configured. Options are not yet supported.
1098
1099 rda,<addr>
1100 Start an early, polled-mode console on a serial port
1101 of an RDA Micro SoC, such as RDA8810PL, at the
1102 specified address. The serial port must already be
1103 setup and configured. Options are not yet supported.
1104
1105 sbi
1106 Use RISC-V SBI (Supervisor Binary Interface) for early
1107 console.
1108
1109 smh Use ARM semihosting calls for early console.
1110
1111 s3c2410,<addr>
1112 s3c2412,<addr>
1113 s3c2440,<addr>
1114 s3c6400,<addr>
1115 s5pv210,<addr>
1116 exynos4210,<addr>
1117 Use early console provided by serial driver available
1118 on Samsung SoCs, requires selecting proper type and
1119 a correct base address of the selected UART port. The
1120 serial port must already be setup and configured.
1121 Options are not yet supported.
1122
1123 lantiq,<addr>
1124 Start an early, polled-mode console on a lantiq serial
1125 (lqasc) port at the specified address. The serial port
1126 must already be setup and configured. Options are not
1127 yet supported.
1128
1129 lpuart,<addr>
1130 lpuart32,<addr>
1131 Use early console provided by Freescale LP UART driver
1132 found on Freescale Vybrid and QorIQ LS1021A processors.
1133 A valid base address must be provided, and the serial
1134 port must already be setup and configured.
1135
1136 ec_imx21,<addr>
1137 ec_imx6q,<addr>
1138 Start an early, polled-mode, output-only console on the
1139 Freescale i.MX UART at the specified address. The UART
1140 must already be setup and configured.
1141
1142 ar3700_uart,<addr>
1143 Start an early, polled-mode console on the
1144 Armada 3700 serial port at the specified
1145 address. The serial port must already be setup
1146 and configured. Options are not yet supported.
1147
1148 qcom_geni,<addr>
1149 Start an early, polled-mode console on a Qualcomm
1150 Generic Interface (GENI) based serial port at the
1151 specified address. The serial port must already be
1152 setup and configured. Options are not yet supported.
1153
1154 efifb,[options]
1155 Start an early, unaccelerated console on the EFI
1156 memory mapped framebuffer (if available). On cache
1157 coherent non-x86 systems that use system memory for
1158 the framebuffer, pass the 'ram' option so that it is
1159 mapped with the correct attributes.
1160
1161 linflex,<addr>
1162 Use early console provided by Freescale LINFlexD UART
1163 serial driver for NXP S32V234 SoCs. A valid base
1164 address must be provided, and the serial port must
1165 already be setup and configured.
1166
1167 earlyprintk= [X86,SH,ARM,M68k,S390]
1168 earlyprintk=vga
1169 earlyprintk=sclp
1170 earlyprintk=xen
1171 earlyprintk=serial[,ttySn[,baudrate]]
1172 earlyprintk=serial[,0x...[,baudrate]]
1173 earlyprintk=ttySn[,baudrate]
1174 earlyprintk=dbgp[debugController#]
1175 earlyprintk=pciserial[,force],bus:device.function[,baudrate]
1176 earlyprintk=xdbc[xhciController#]
1177
1178 earlyprintk is useful when the kernel crashes before
1179 the normal console is initialized. It is not enabled by
1180 default because it has some cosmetic problems.
1181
1182 Append ",keep" to not disable it when the real console
1183 takes over.
1184
1185 Only one of vga, efi, serial, or usb debug port can
1186 be used at a time.
1187
1188 Currently only ttyS0 and ttyS1 may be specified by
1189 name. Other I/O ports may be explicitly specified
1190 on some architectures (x86 and arm at least) by
1191 replacing ttySn with an I/O port address, like this:
1192 earlyprintk=serial,0x1008,115200
1193 You can find the port for a given device in
1194 /proc/tty/driver/serial:
1195 2: uart:ST16650V2 port:00001008 irq:18 ...
1196
1197 Interaction with the standard serial driver is not
1198 very good.
1199
1200 The VGA and EFI output is eventually overwritten by
1201 the real console.
1202
1203 The xen output can only be used by Xen PV guests.
1204
1205 The sclp output can only be used on s390.
1206
1207 The optional "force" to "pciserial" enables use of a
1208 PCI device even when its classcode is not of the
1209 UART class.
1210
1211 edac_report= [HW,EDAC] Control how to report EDAC event
1212 Format: {"on" | "off" | "force"}
1213 on: enable EDAC to report H/W event. May be overridden
1214 by other higher priority error reporting module.
1215 off: disable H/W event reporting through EDAC.
1216 force: enforce the use of EDAC to report H/W event.
1217 default: on.
1218
1219 ekgdboc= [X86,KGDB] Allow early kernel console debugging
1220 ekgdboc=kbd
1221
1222 This is designed to be used in conjunction with
1223 the boot argument: earlyprintk=vga
1224
1225 This parameter works in place of the kgdboc parameter
1226 but can only be used if the backing tty is available
1227 very early in the boot process. For early debugging
1228 via a serial port see kgdboc_earlycon instead.
1229
1230 edd= [EDD]
1231 Format: {"off" | "on" | "skip[mbr]"}
1232
1233 efi= [EFI]
1234 Format: { "debug", "disable_early_pci_dma",
1235 "nochunk", "noruntime", "nosoftreserve",
1236 "novamap", "no_disable_early_pci_dma" }
1237 debug: enable misc debug output.
1238 disable_early_pci_dma: disable the busmaster bit on all
1239 PCI bridges while in the EFI boot stub.
1240 nochunk: disable reading files in "chunks" in the EFI
1241 boot stub, as chunking can cause problems with some
1242 firmware implementations.
1243 noruntime : disable EFI runtime services support
1244 nosoftreserve: The EFI_MEMORY_SP (Specific Purpose)
1245 attribute may cause the kernel to reserve the
1246 memory range for a memory mapping driver to
1247 claim. Specify efi=nosoftreserve to disable this
1248 reservation and treat the memory by its base type
1249 (i.e. EFI_CONVENTIONAL_MEMORY / "System RAM").
1250 novamap: do not call SetVirtualAddressMap().
1251 no_disable_early_pci_dma: Leave the busmaster bit set
1252 on all PCI bridges while in the EFI boot stub
1253
1254 efi_no_storage_paranoia [EFI; X86]
1255 Using this parameter you can use more than 50% of
1256 your efi variable storage. Use this parameter only if
1257 you are really sure that your UEFI does sane gc and
1258 fulfills the spec otherwise your board may brick.
1259
1260 efi_fake_mem= nn[KMG]@ss[KMG]:aa[,nn[KMG]@ss[KMG]:aa,..] [EFI; X86]
1261 Add arbitrary attribute to specific memory range by
1262 updating original EFI memory map.
1263 Region of memory which aa attribute is added to is
1264 from ss to ss+nn.
1265
1266 If efi_fake_mem=2G@4G:0x10000,2G@0x10a0000000:0x10000
1267 is specified, EFI_MEMORY_MORE_RELIABLE(0x10000)
1268 attribute is added to range 0x100000000-0x180000000 and
1269 0x10a0000000-0x1120000000.
1270
1271 If efi_fake_mem=8G@9G:0x40000 is specified, the
1272 EFI_MEMORY_SP(0x40000) attribute is added to
1273 range 0x240000000-0x43fffffff.
1274
1275 Using this parameter you can do debugging of EFI memmap
1276 related features. For example, you can do debugging of
1277 Address Range Mirroring feature even if your box
1278 doesn't support it, or mark specific memory as
1279 "soft reserved".
1280
1281 efivar_ssdt= [EFI; X86] Name of an EFI variable that contains an SSDT
1282 that is to be dynamically loaded by Linux. If there are
1283 multiple variables with the same name but with different
1284 vendor GUIDs, all of them will be loaded. See
1285 Documentation/admin-guide/acpi/ssdt-overlays.rst for details.
1286
1287
1288 eisa_irq_edge= [PARISC,HW]
1289 See header of drivers/parisc/eisa.c.
1290
1291 elanfreq= [X86-32]
1292 See comment before function elanfreq_setup() in
1293 arch/x86/kernel/cpu/cpufreq/elanfreq.c.
1294
1295 elfcorehdr=[size[KMG]@]offset[KMG] [IA64,PPC,SH,X86,S390]
1296 Specifies physical address of start of kernel core
1297 image elf header and optionally the size. Generally
1298 kexec loader will pass this option to capture kernel.
1299 See Documentation/admin-guide/kdump/kdump.rst for details.
1300
1301 enable_mtrr_cleanup [X86]
1302 The kernel tries to adjust MTRR layout from continuous
1303 to discrete, to make X server driver able to add WB
1304 entry later. This parameter enables that.
1305
1306 enable_timer_pin_1 [X86]
1307 Enable PIN 1 of APIC timer
1308 Can be useful to work around chipset bugs
1309 (in particular on some ATI chipsets).
1310 The kernel tries to set a reasonable default.
1311
1312 enforcing [SELINUX] Set initial enforcing status.
1313 Format: {"0" | "1"}
1314 See security/selinux/Kconfig help text.
1315 0 -- permissive (log only, no denials).
1316 1 -- enforcing (deny and log).
1317 Default value is 0.
1318 Value can be changed at runtime via
1319 /sys/fs/selinux/enforce.
1320
1321 erst_disable [ACPI]
1322 Disable Error Record Serialization Table (ERST)
1323 support.
1324
1325 ether= [HW,NET] Ethernet cards parameters
1326 This option is obsoleted by the "netdev=" option, which
1327 has equivalent usage. See its documentation for details.
1328
1329 evm= [EVM]
1330 Format: { "fix" }
1331 Permit 'security.evm' to be updated regardless of
1332 current integrity status.
1333
1334 failslab=
1335 fail_page_alloc=
1336 fail_make_request=[KNL]
1337 General fault injection mechanism.
1338 Format: <interval>,<probability>,<space>,<times>
1339 See also Documentation/fault-injection/.
1340
1341 floppy= [HW]
1342 See Documentation/admin-guide/blockdev/floppy.rst.
1343
1344 force_pal_cache_flush
1345 [IA-64] Avoid check_sal_cache_flush which may hang on
1346 buggy SAL_CACHE_FLUSH implementations. Using this
1347 parameter will force ia64_sal_cache_flush to call
1348 ia64_pal_cache_flush instead of SAL_CACHE_FLUSH.
1349
1350 forcepae [X86-32]
1351 Forcefully enable Physical Address Extension (PAE).
1352 Many Pentium M systems disable PAE but may have a
1353 functionally usable PAE implementation.
1354 Warning: use of this parameter will taint the kernel
1355 and may cause unknown problems.
1356
1357 ftrace=[tracer]
1358 [FTRACE] will set and start the specified tracer
1359 as early as possible in order to facilitate early
1360 boot debugging.
1361
1362 ftrace_dump_on_oops[=orig_cpu]
1363 [FTRACE] will dump the trace buffers on oops.
1364 If no parameter is passed, ftrace will dump
1365 buffers of all CPUs, but if you pass orig_cpu, it will
1366 dump only the buffer of the CPU that triggered the
1367 oops.
1368
1369 ftrace_filter=[function-list]
1370 [FTRACE] Limit the functions traced by the function
1371 tracer at boot up. function-list is a comma separated
1372 list of functions. This list can be changed at run
1373 time by the set_ftrace_filter file in the debugfs
1374 tracing directory.
1375
1376 ftrace_notrace=[function-list]
1377 [FTRACE] Do not trace the functions specified in
1378 function-list. This list can be changed at run time
1379 by the set_ftrace_notrace file in the debugfs
1380 tracing directory.
1381
1382 ftrace_graph_filter=[function-list]
1383 [FTRACE] Limit the top level callers functions traced
1384 by the function graph tracer at boot up.
1385 function-list is a comma separated list of functions
1386 that can be changed at run time by the
1387 set_graph_function file in the debugfs tracing directory.
1388
1389 ftrace_graph_notrace=[function-list]
1390 [FTRACE] Do not trace from the functions specified in
1391 function-list. This list is a comma separated list of
1392 functions that can be changed at run time by the
1393 set_graph_notrace file in the debugfs tracing directory.
1394
1395 ftrace_graph_max_depth=<uint>
1396 [FTRACE] Used with the function graph tracer. This is
1397 the max depth it will trace into a function. This value
1398 can be changed at run time by the max_graph_depth file
1399 in the tracefs tracing directory. default: 0 (no limit)
1400
1401 fw_devlink= [KNL] Create device links between consumer and supplier
1402 devices by scanning the firmware to infer the
1403 consumer/supplier relationships. This feature is
1404 especially useful when drivers are loaded as modules as
1405 it ensures proper ordering of tasks like device probing
1406 (suppliers first, then consumers), supplier boot state
1407 clean up (only after all consumers have probed),
1408 suspend/resume & runtime PM (consumers first, then
1409 suppliers).
1410 Format: { off | permissive | on | rpm }
1411 off -- Don't create device links from firmware info.
1412 permissive -- Create device links from firmware info
1413 but use it only for ordering boot state clean
1414 up (sync_state() calls).
1415 on -- Create device links from firmware info and use it
1416 to enforce probe and suspend/resume ordering.
1417 rpm -- Like "on", but also use to order runtime PM.
1418
1419 gamecon.map[2|3]=
1420 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad
1421 support via parallel port (up to 5 devices per port)
1422 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5>
1423 See also Documentation/input/devices/joystick-parport.rst
1424
1425 gamma= [HW,DRM]
1426
1427 gart_fix_e820= [X86-64] disable the fix e820 for K8 GART
1428 Format: off | on
1429 default: on
1430
1431 gcov_persist= [GCOV] When non-zero (default), profiling data for
1432 kernel modules is saved and remains accessible via
1433 debugfs, even when the module is unloaded/reloaded.
1434 When zero, profiling data is discarded and associated
1435 debugfs files are removed at module unload time.
1436
1437 goldfish [X86] Enable the goldfish android emulator platform.
1438 Don't use this when you are not running on the
1439 android emulator
1440
1441 gpt [EFI] Forces disk with valid GPT signature but
1442 invalid Protective MBR to be treated as GPT. If the
1443 primary GPT is corrupted, it enables the backup/alternate
1444 GPT to be used instead.
1445
1446 grcan.enable0= [HW] Configuration of physical interface 0. Determines
1447 the "Enable 0" bit of the configuration register.
1448 Format: 0 | 1
1449 Default: 0
1450 grcan.enable1= [HW] Configuration of physical interface 1. Determines
1451 the "Enable 0" bit of the configuration register.
1452 Format: 0 | 1
1453 Default: 0
1454 grcan.select= [HW] Select which physical interface to use.
1455 Format: 0 | 1
1456 Default: 0
1457 grcan.txsize= [HW] Sets the size of the tx buffer.
1458 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0.
1459 Default: 1024
1460 grcan.rxsize= [HW] Sets the size of the rx buffer.
1461 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0.
1462 Default: 1024
1463
1464 gpio-mockup.gpio_mockup_ranges
1465 [HW] Sets the ranges of gpiochip of for this device.
1466 Format: <start1>,<end1>,<start2>,<end2>...
1467
1468 hardlockup_all_cpu_backtrace=
1469 [KNL] Should the hard-lockup detector generate
1470 backtraces on all cpus.
1471 Format: 0 | 1
1472
1473 hashdist= [KNL,NUMA] Large hashes allocated during boot
1474 are distributed across NUMA nodes. Defaults on
1475 for 64-bit NUMA, off otherwise.
1476 Format: 0 | 1 (for off | on)
1477
1478 hcl= [IA-64] SGI's Hardware Graph compatibility layer
1479
1480 hd= [EIDE] (E)IDE hard drive subsystem geometry
1481 Format: <cyl>,<head>,<sect>
1482
1483 hest_disable [ACPI]
1484 Disable Hardware Error Source Table (HEST) support;
1485 corresponding firmware-first mode error processing
1486 logic will be disabled.
1487
1488 highmem=nn[KMG] [KNL,BOOT] forces the highmem zone to have an exact
1489 size of <nn>. This works even on boxes that have no
1490 highmem otherwise. This also works to reduce highmem
1491 size on bigger boxes.
1492
1493 highres= [KNL] Enable/disable high resolution timer mode.
1494 Valid parameters: "on", "off"
1495 Default: "on"
1496
1497 hlt [BUGS=ARM,SH]
1498
1499 hpet= [X86-32,HPET] option to control HPET usage
1500 Format: { enable (default) | disable | force |
1501 verbose }
1502 disable: disable HPET and use PIT instead
1503 force: allow force enabled of undocumented chips (ICH4,
1504 VIA, nVidia)
1505 verbose: show contents of HPET registers during setup
1506
1507 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET
1508 registers. Default set by CONFIG_HPET_MMAP_DEFAULT.
1509
1510 hugetlb_cma= [HW] The size of a cma area used for allocation
1511 of gigantic hugepages.
1512 Format: nn[KMGTPE]
1513
1514 Reserve a cma area of given size and allocate gigantic
1515 hugepages using the cma allocator. If enabled, the
1516 boot-time allocation of gigantic hugepages is skipped.
1517
1518 hugepages= [HW] Number of HugeTLB pages to allocate at boot.
1519 If this follows hugepagesz (below), it specifies
1520 the number of pages of hugepagesz to be allocated.
1521 If this is the first HugeTLB parameter on the command
1522 line, it specifies the number of pages to allocate for
1523 the default huge page size. See also
1524 Documentation/admin-guide/mm/hugetlbpage.rst.
1525 Format: <integer>
1526
1527 hugepagesz=
1528 [HW] The size of the HugeTLB pages. This is used in
1529 conjunction with hugepages (above) to allocate huge
1530 pages of a specific size at boot. The pair
1531 hugepagesz=X hugepages=Y can be specified once for
1532 each supported huge page size. Huge page sizes are
1533 architecture dependent. See also
1534 Documentation/admin-guide/mm/hugetlbpage.rst.
1535 Format: size[KMG]
1536
1537 hung_task_panic=
1538 [KNL] Should the hung task detector generate panics.
1539 Format: 0 | 1
1540
1541 A value of 1 instructs the kernel to panic when a
1542 hung task is detected. The default value is controlled
1543 by the CONFIG_BOOTPARAM_HUNG_TASK_PANIC build-time
1544 option. The value selected by this boot parameter can
1545 be changed later by the kernel.hung_task_panic sysctl.
1546
1547 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC)
1548 terminal devices. Valid values: 0..8
1549 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs.
1550 If specified, z/VM IUCV HVC accepts connections
1551 from listed z/VM user IDs only.
1552
1553 hv_nopvspin [X86,HYPER_V] Disables the paravirt spinlock optimizations
1554 which allow the hypervisor to 'idle' the
1555 guest on lock contention.
1556
1557 keep_bootcon [KNL]
1558 Do not unregister boot console at start. This is only
1559 useful for debugging when something happens in the window
1560 between unregistering the boot console and initializing
1561 the real console.
1562
1563 i2c_bus= [HW] Override the default board specific I2C bus speed
1564 or register an additional I2C bus that is not
1565 registered from board initialization code.
1566 Format:
1567 <bus_id>,<clkrate>
1568
1569 i8042.debug [HW] Toggle i8042 debug mode
1570 i8042.unmask_kbd_data
1571 [HW] Enable printing of interrupt data from the KBD port
1572 (disabled by default, and as a pre-condition
1573 requires that i8042.debug=1 be enabled)
1574 i8042.direct [HW] Put keyboard port into non-translated mode
1575 i8042.dumbkbd [HW] Pretend that controller can only read data from
1576 keyboard and cannot control its state
1577 (Don't attempt to blink the leds)
1578 i8042.noaux [HW] Don't check for auxiliary (== mouse) port
1579 i8042.nokbd [HW] Don't check/create keyboard port
1580 i8042.noloop [HW] Disable the AUX Loopback command while probing
1581 for the AUX port
1582 i8042.nomux [HW] Don't check presence of an active multiplexing
1583 controller
1584 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX
1585 controllers
1586 i8042.notimeout [HW] Ignore timeout condition signalled by controller
1587 i8042.reset [HW] Reset the controller during init, cleanup and
1588 suspend-to-ram transitions, only during s2r
1589 transitions, or never reset
1590 Format: { 1 | Y | y | 0 | N | n }
1591 1, Y, y: always reset controller
1592 0, N, n: don't ever reset controller
1593 Default: only on s2r transitions on x86; most other
1594 architectures force reset to be always executed
1595 i8042.unlock [HW] Unlock (ignore) the keylock
1596 i8042.kbdreset [HW] Reset device connected to KBD port
1597
1598 i810= [HW,DRM]
1599
1600 i8k.ignore_dmi [HW] Continue probing hardware even if DMI data
1601 indicates that the driver is running on unsupported
1602 hardware.
1603 i8k.force [HW] Activate i8k driver even if SMM BIOS signature
1604 does not match list of supported models.
1605 i8k.power_status
1606 [HW] Report power status in /proc/i8k
1607 (disabled by default)
1608 i8k.restricted [HW] Allow controlling fans only if SYS_ADMIN
1609 capability is set.
1610
1611 i915.invert_brightness=
1612 [DRM] Invert the sense of the variable that is used to
1613 set the brightness of the panel backlight. Normally a
1614 brightness value of 0 indicates backlight switched off,
1615 and the maximum of the brightness value sets the backlight
1616 to maximum brightness. If this parameter is set to 0
1617 (default) and the machine requires it, or this parameter
1618 is set to 1, a brightness value of 0 sets the backlight
1619 to maximum brightness, and the maximum of the brightness
1620 value switches the backlight off.
1621 -1 -- never invert brightness
1622 0 -- machine default
1623 1 -- force brightness inversion
1624
1625 icn= [HW,ISDN]
1626 Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]]
1627
1628 ide-core.nodma= [HW] (E)IDE subsystem
1629 Format: =0.0 to prevent dma on hda, =0.1 hdb =1.0 hdc
1630 .vlb_clock .pci_clock .noflush .nohpa .noprobe .nowerr
1631 .cdrom .chs .ignore_cable are additional options
1632 See Documentation/ide/ide.rst.
1633
1634 ide-generic.probe-mask= [HW] (E)IDE subsystem
1635 Format: <int>
1636 Probe mask for legacy ISA IDE ports. Depending on
1637 platform up to 6 ports are supported, enabled by
1638 setting corresponding bits in the mask to 1. The
1639 default value is 0x0, which has a special meaning.
1640 On systems that have PCI, it triggers scanning the
1641 PCI bus for the first and the second port, which
1642 are then probed. On systems without PCI the value
1643 of 0x0 enables probing the two first ports as if it
1644 was 0x3.
1645
1646 ide-pci-generic.all-generic-ide [HW] (E)IDE subsystem
1647 Claim all unknown PCI IDE storage controllers.
1648
1649 idle= [X86]
1650 Format: idle=poll, idle=halt, idle=nomwait
1651 Poll forces a polling idle loop that can slightly
1652 improve the performance of waking up a idle CPU, but
1653 will use a lot of power and make the system run hot.
1654 Not recommended.
1655 idle=halt: Halt is forced to be used for CPU idle.
1656 In such case C2/C3 won't be used again.
1657 idle=nomwait: Disable mwait for CPU C-states
1658
1659 ieee754= [MIPS] Select IEEE Std 754 conformance mode
1660 Format: { strict | legacy | 2008 | relaxed }
1661 Default: strict
1662
1663 Choose which programs will be accepted for execution
1664 based on the IEEE 754 NaN encoding(s) supported by
1665 the FPU and the NaN encoding requested with the value
1666 of an ELF file header flag individually set by each
1667 binary. Hardware implementations are permitted to
1668 support either or both of the legacy and the 2008 NaN
1669 encoding mode.
1670
1671 Available settings are as follows:
1672 strict accept binaries that request a NaN encoding
1673 supported by the FPU
1674 legacy only accept legacy-NaN binaries, if supported
1675 by the FPU
1676 2008 only accept 2008-NaN binaries, if supported
1677 by the FPU
1678 relaxed accept any binaries regardless of whether
1679 supported by the FPU
1680
1681 The FPU emulator is always able to support both NaN
1682 encodings, so if no FPU hardware is present or it has
1683 been disabled with 'nofpu', then the settings of
1684 'legacy' and '2008' strap the emulator accordingly,
1685 'relaxed' straps the emulator for both legacy-NaN and
1686 2008-NaN, whereas 'strict' enables legacy-NaN only on
1687 legacy processors and both NaN encodings on MIPS32 or
1688 MIPS64 CPUs.
1689
1690 The setting for ABS.fmt/NEG.fmt instruction execution
1691 mode generally follows that for the NaN encoding,
1692 except where unsupported by hardware.
1693
1694 ignore_loglevel [KNL]
1695 Ignore loglevel setting - this will print /all/
1696 kernel messages to the console. Useful for debugging.
1697 We also add it as printk module parameter, so users
1698 could change it dynamically, usually by
1699 /sys/module/printk/parameters/ignore_loglevel.
1700
1701 ignore_rlimit_data
1702 Ignore RLIMIT_DATA setting for data mappings,
1703 print warning at first misuse. Can be changed via
1704 /sys/module/kernel/parameters/ignore_rlimit_data.
1705
1706 ihash_entries= [KNL]
1707 Set number of hash buckets for inode cache.
1708
1709 ima_appraise= [IMA] appraise integrity measurements
1710 Format: { "off" | "enforce" | "fix" | "log" }
1711 default: "enforce"
1712
1713 ima_appraise_tcb [IMA] Deprecated. Use ima_policy= instead.
1714 The builtin appraise policy appraises all files
1715 owned by uid=0.
1716
1717 ima_canonical_fmt [IMA]
1718 Use the canonical format for the binary runtime
1719 measurements, instead of host native format.
1720
1721 ima_hash= [IMA]
1722 Format: { md5 | sha1 | rmd160 | sha256 | sha384
1723 | sha512 | ... }
1724 default: "sha1"
1725
1726 The list of supported hash algorithms is defined
1727 in crypto/hash_info.h.
1728
1729 ima_policy= [IMA]
1730 The builtin policies to load during IMA setup.
1731 Format: "tcb | appraise_tcb | secure_boot |
1732 fail_securely"
1733
1734 The "tcb" policy measures all programs exec'd, files
1735 mmap'd for exec, and all files opened with the read
1736 mode bit set by either the effective uid (euid=0) or
1737 uid=0.
1738
1739 The "appraise_tcb" policy appraises the integrity of
1740 all files owned by root.
1741
1742 The "secure_boot" policy appraises the integrity
1743 of files (eg. kexec kernel image, kernel modules,
1744 firmware, policy, etc) based on file signatures.
1745
1746 The "fail_securely" policy forces file signature
1747 verification failure also on privileged mounted
1748 filesystems with the SB_I_UNVERIFIABLE_SIGNATURE
1749 flag.
1750
1751 ima_tcb [IMA] Deprecated. Use ima_policy= instead.
1752 Load a policy which meets the needs of the Trusted
1753 Computing Base. This means IMA will measure all
1754 programs exec'd, files mmap'd for exec, and all files
1755 opened for read by uid=0.
1756
1757 ima_template= [IMA]
1758 Select one of defined IMA measurements template formats.
1759 Formats: { "ima" | "ima-ng" | "ima-sig" }
1760 Default: "ima-ng"
1761
1762 ima_template_fmt=
1763 [IMA] Define a custom template format.
1764 Format: { "field1|...|fieldN" }
1765
1766 ima.ahash_minsize= [IMA] Minimum file size for asynchronous hash usage
1767 Format: <min_file_size>
1768 Set the minimal file size for using asynchronous hash.
1769 If left unspecified, ahash usage is disabled.
1770
1771 ahash performance varies for different data sizes on
1772 different crypto accelerators. This option can be used
1773 to achieve the best performance for a particular HW.
1774
1775 ima.ahash_bufsize= [IMA] Asynchronous hash buffer size
1776 Format: <bufsize>
1777 Set hashing buffer size. Default: 4k.
1778
1779 ahash performance varies for different chunk sizes on
1780 different crypto accelerators. This option can be used
1781 to achieve best performance for particular HW.
1782
1783 init= [KNL]
1784 Format: <full_path>
1785 Run specified binary instead of /sbin/init as init
1786 process.
1787
1788 initcall_debug [KNL] Trace initcalls as they are executed. Useful
1789 for working out where the kernel is dying during
1790 startup.
1791
1792 initcall_blacklist= [KNL] Do not execute a comma-separated list of
1793 initcall functions. Useful for debugging built-in
1794 modules and initcalls.
1795
1796 initrd= [BOOT] Specify the location of the initial ramdisk
1797
1798 initrdmem= [KNL] Specify a physical address and size from which to
1799 load the initrd. If an initrd is compiled in or
1800 specified in the bootparams, it takes priority over this
1801 setting.
1802 Format: ss[KMG],nn[KMG]
1803 Default is 0, 0
1804
1805 init_on_alloc= [MM] Fill newly allocated pages and heap objects with
1806 zeroes.
1807 Format: 0 | 1
1808 Default set by CONFIG_INIT_ON_ALLOC_DEFAULT_ON.
1809
1810 init_on_free= [MM] Fill freed pages and heap objects with zeroes.
1811 Format: 0 | 1
1812 Default set by CONFIG_INIT_ON_FREE_DEFAULT_ON.
1813
1814 init_pkru= [X86] Specify the default memory protection keys rights
1815 register contents for all processes. 0x55555554 by
1816 default (disallow access to all but pkey 0). Can
1817 override in debugfs after boot.
1818
1819 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver
1820 Format: <irq>
1821
1822 int_pln_enable [X86] Enable power limit notification interrupt
1823
1824 integrity_audit=[IMA]
1825 Format: { "0" | "1" }
1826 0 -- basic integrity auditing messages. (Default)
1827 1 -- additional integrity auditing messages.
1828
1829 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option
1830 on
1831 Enable intel iommu driver.
1832 off
1833 Disable intel iommu driver.
1834 igfx_off [Default Off]
1835 By default, gfx is mapped as normal device. If a gfx
1836 device has a dedicated DMAR unit, the DMAR unit is
1837 bypassed by not enabling DMAR with this option. In
1838 this case, gfx device will use physical address for
1839 DMA.
1840 forcedac [X86-64]
1841 With this option iommu will not optimize to look
1842 for io virtual address below 32-bit forcing dual
1843 address cycle on pci bus for cards supporting greater
1844 than 32-bit addressing. The default is to look
1845 for translation below 32-bit and if not available
1846 then look in the higher range.
1847 strict [Default Off]
1848 With this option on every unmap_single operation will
1849 result in a hardware IOTLB flush operation as opposed
1850 to batching them for performance.
1851 sp_off [Default Off]
1852 By default, super page will be supported if Intel IOMMU
1853 has the capability. With this option, super page will
1854 not be supported.
1855 sm_on [Default Off]
1856 By default, scalable mode will be disabled even if the
1857 hardware advertises that it has support for the scalable
1858 mode translation. With this option set, scalable mode
1859 will be used on hardware which claims to support it.
1860 tboot_noforce [Default Off]
1861 Do not force the Intel IOMMU enabled under tboot.
1862 By default, tboot will force Intel IOMMU on, which
1863 could harm performance of some high-throughput
1864 devices like 40GBit network cards, even if identity
1865 mapping is enabled.
1866 Note that using this option lowers the security
1867 provided by tboot because it makes the system
1868 vulnerable to DMA attacks.
1869 nobounce [Default off]
1870 Disable bounce buffer for untrusted devices such as
1871 the Thunderbolt devices. This will treat the untrusted
1872 devices as the trusted ones, hence might expose security
1873 risks of DMA attacks.
1874
1875 intel_idle.max_cstate= [KNL,HW,ACPI,X86]
1876 0 disables intel_idle and fall back on acpi_idle.
1877 1 to 9 specify maximum depth of C-state.
1878
1879 intel_pstate= [X86]
1880 disable
1881 Do not enable intel_pstate as the default
1882 scaling driver for the supported processors
1883 passive
1884 Use intel_pstate as a scaling driver, but configure it
1885 to work with generic cpufreq governors (instead of
1886 enabling its internal governor). This mode cannot be
1887 used along with the hardware-managed P-states (HWP)
1888 feature.
1889 force
1890 Enable intel_pstate on systems that prohibit it by default
1891 in favor of acpi-cpufreq. Forcing the intel_pstate driver
1892 instead of acpi-cpufreq may disable platform features, such
1893 as thermal controls and power capping, that rely on ACPI
1894 P-States information being indicated to OSPM and therefore
1895 should be used with caution. This option does not work with
1896 processors that aren't supported by the intel_pstate driver
1897 or on platforms that use pcc-cpufreq instead of acpi-cpufreq.
1898 no_hwp
1899 Do not enable hardware P state control (HWP)
1900 if available.
1901 hwp_only
1902 Only load intel_pstate on systems which support
1903 hardware P state control (HWP) if available.
1904 support_acpi_ppc
1905 Enforce ACPI _PPC performance limits. If the Fixed ACPI
1906 Description Table, specifies preferred power management
1907 profile as "Enterprise Server" or "Performance Server",
1908 then this feature is turned on by default.
1909 per_cpu_perf_limits
1910 Allow per-logical-CPU P-State performance control limits using
1911 cpufreq sysfs interface
1912
1913 intremap= [X86-64, Intel-IOMMU]
1914 on enable Interrupt Remapping (default)
1915 off disable Interrupt Remapping
1916 nosid disable Source ID checking
1917 no_x2apic_optout
1918 BIOS x2APIC opt-out request will be ignored
1919 nopost disable Interrupt Posting
1920
1921 iomem= Disable strict checking of access to MMIO memory
1922 strict regions from userspace.
1923 relaxed
1924
1925 iommu= [X86]
1926 off
1927 force
1928 noforce
1929 biomerge
1930 panic
1931 nopanic
1932 merge
1933 nomerge
1934 soft
1935 pt [X86]
1936 nopt [X86]
1937 nobypass [PPC/POWERNV]
1938 Disable IOMMU bypass, using IOMMU for PCI devices.
1939
1940 iommu.strict= [ARM64] Configure TLB invalidation behaviour
1941 Format: { "0" | "1" }
1942 0 - Lazy mode.
1943 Request that DMA unmap operations use deferred
1944 invalidation of hardware TLBs, for increased
1945 throughput at the cost of reduced device isolation.
1946 Will fall back to strict mode if not supported by
1947 the relevant IOMMU driver.
1948 1 - Strict mode (default).
1949 DMA unmap operations invalidate IOMMU hardware TLBs
1950 synchronously.
1951
1952 iommu.passthrough=
1953 [ARM64, X86] Configure DMA to bypass the IOMMU by default.
1954 Format: { "0" | "1" }
1955 0 - Use IOMMU translation for DMA.
1956 1 - Bypass the IOMMU for DMA.
1957 unset - Use value of CONFIG_IOMMU_DEFAULT_PASSTHROUGH.
1958
1959 io7= [HW] IO7 for Marvel based alpha systems
1960 See comment before marvel_specify_io7 in
1961 arch/alpha/kernel/core_marvel.c.
1962
1963 io_delay= [X86] I/O delay method
1964 0x80
1965 Standard port 0x80 based delay
1966 0xed
1967 Alternate port 0xed based delay (needed on some systems)
1968 udelay
1969 Simple two microseconds delay
1970 none
1971 No delay
1972
1973 ip= [IP_PNP]
1974 See Documentation/admin-guide/nfs/nfsroot.rst.
1975
1976 ipcmni_extend [KNL] Extend the maximum number of unique System V
1977 IPC identifiers from 32,768 to 16,777,216.
1978
1979 irqaffinity= [SMP] Set the default irq affinity mask
1980 The argument is a cpu list, as described above.
1981
1982 irqchip.gicv2_force_probe=
1983 [ARM, ARM64]
1984 Format: <bool>
1985 Force the kernel to look for the second 4kB page
1986 of a GICv2 controller even if the memory range
1987 exposed by the device tree is too small.
1988
1989 irqchip.gicv3_nolpi=
1990 [ARM, ARM64]
1991 Force the kernel to ignore the availability of
1992 LPIs (and by consequence ITSs). Intended for system
1993 that use the kernel as a bootloader, and thus want
1994 to let secondary kernels in charge of setting up
1995 LPIs.
1996
1997 irqchip.gicv3_pseudo_nmi= [ARM64]
1998 Enables support for pseudo-NMIs in the kernel. This
1999 requires the kernel to be built with
2000 CONFIG_ARM64_PSEUDO_NMI.
2001
2002 irqfixup [HW]
2003 When an interrupt is not handled search all handlers
2004 for it. Intended to get systems with badly broken
2005 firmware running.
2006
2007 irqpoll [HW]
2008 When an interrupt is not handled search all handlers
2009 for it. Also check all handlers each timer
2010 interrupt. Intended to get systems with badly broken
2011 firmware running.
2012
2013 isapnp= [ISAPNP]
2014 Format: <RDP>,<reset>,<pci_scan>,<verbosity>
2015
2016 isolcpus= [KNL,SMP,ISOL] Isolate a given set of CPUs from disturbance.
2017 [Deprecated - use cpusets instead]
2018 Format: [flag-list,]<cpu-list>
2019
2020 Specify one or more CPUs to isolate from disturbances
2021 specified in the flag list (default: domain):
2022
2023 nohz
2024 Disable the tick when a single task runs.
2025
2026 A residual 1Hz tick is offloaded to workqueues, which you
2027 need to affine to housekeeping through the global
2028 workqueue's affinity configured via the
2029 /sys/devices/virtual/workqueue/cpumask sysfs file, or
2030 by using the 'domain' flag described below.
2031
2032 NOTE: by default the global workqueue runs on all CPUs,
2033 so to protect individual CPUs the 'cpumask' file has to
2034 be configured manually after bootup.
2035
2036 domain
2037 Isolate from the general SMP balancing and scheduling
2038 algorithms. Note that performing domain isolation this way
2039 is irreversible: it's not possible to bring back a CPU to
2040 the domains once isolated through isolcpus. It's strongly
2041 advised to use cpusets instead to disable scheduler load
2042 balancing through the "cpuset.sched_load_balance" file.
2043 It offers a much more flexible interface where CPUs can
2044 move in and out of an isolated set anytime.
2045
2046 You can move a process onto or off an "isolated" CPU via
2047 the CPU affinity syscalls or cpuset.
2048 <cpu number> begins at 0 and the maximum value is
2049 "number of CPUs in system - 1".
2050
2051 managed_irq
2052
2053 Isolate from being targeted by managed interrupts
2054 which have an interrupt mask containing isolated
2055 CPUs. The affinity of managed interrupts is
2056 handled by the kernel and cannot be changed via
2057 the /proc/irq/* interfaces.
2058
2059 This isolation is best effort and only effective
2060 if the automatically assigned interrupt mask of a
2061 device queue contains isolated and housekeeping
2062 CPUs. If housekeeping CPUs are online then such
2063 interrupts are directed to the housekeeping CPU
2064 so that IO submitted on the housekeeping CPU
2065 cannot disturb the isolated CPU.
2066
2067 If a queue's affinity mask contains only isolated
2068 CPUs then this parameter has no effect on the
2069 interrupt routing decision, though interrupts are
2070 only delivered when tasks running on those
2071 isolated CPUs submit IO. IO submitted on
2072 housekeeping CPUs has no influence on those
2073 queues.
2074
2075 The format of <cpu-list> is described above.
2076
2077 iucv= [HW,NET]
2078
2079 ivrs_ioapic [HW,X86-64]
2080 Provide an override to the IOAPIC-ID<->DEVICE-ID
2081 mapping provided in the IVRS ACPI table. For
2082 example, to map IOAPIC-ID decimal 10 to
2083 PCI device 00:14.0 write the parameter as:
2084 ivrs_ioapic[10]=00:14.0
2085
2086 ivrs_hpet [HW,X86-64]
2087 Provide an override to the HPET-ID<->DEVICE-ID
2088 mapping provided in the IVRS ACPI table. For
2089 example, to map HPET-ID decimal 0 to
2090 PCI device 00:14.0 write the parameter as:
2091 ivrs_hpet[0]=00:14.0
2092
2093 ivrs_acpihid [HW,X86-64]
2094 Provide an override to the ACPI-HID:UID<->DEVICE-ID
2095 mapping provided in the IVRS ACPI table. For
2096 example, to map UART-HID:UID AMD0020:0 to
2097 PCI device 00:14.5 write the parameter as:
2098 ivrs_acpihid[00:14.5]=AMD0020:0
2099
2100 js= [HW,JOY] Analog joystick
2101 See Documentation/input/joydev/joystick.rst.
2102
2103 nokaslr [KNL]
2104 When CONFIG_RANDOMIZE_BASE is set, this disables
2105 kernel and module base offset ASLR (Address Space
2106 Layout Randomization).
2107
2108 kasan_multi_shot
2109 [KNL] Enforce KASAN (Kernel Address Sanitizer) to print
2110 report on every invalid memory access. Without this
2111 parameter KASAN will print report only for the first
2112 invalid access.
2113
2114 keepinitrd [HW,ARM]
2115
2116 kernelcore= [KNL,X86,IA-64,PPC]
2117 Format: nn[KMGTPE] | nn% | "mirror"
2118 This parameter specifies the amount of memory usable by
2119 the kernel for non-movable allocations. The requested
2120 amount is spread evenly throughout all nodes in the
2121 system as ZONE_NORMAL. The remaining memory is used for
2122 movable memory in its own zone, ZONE_MOVABLE. In the
2123 event, a node is too small to have both ZONE_NORMAL and
2124 ZONE_MOVABLE, kernelcore memory will take priority and
2125 other nodes will have a larger ZONE_MOVABLE.
2126
2127 ZONE_MOVABLE is used for the allocation of pages that
2128 may be reclaimed or moved by the page migration
2129 subsystem. Note that allocations like PTEs-from-HighMem
2130 still use the HighMem zone if it exists, and the Normal
2131 zone if it does not.
2132
2133 It is possible to specify the exact amount of memory in
2134 the form of "nn[KMGTPE]", a percentage of total system
2135 memory in the form of "nn%", or "mirror". If "mirror"
2136 option is specified, mirrored (reliable) memory is used
2137 for non-movable allocations and remaining memory is used
2138 for Movable pages. "nn[KMGTPE]", "nn%", and "mirror"
2139 are exclusive, so you cannot specify multiple forms.
2140
2141 kgdbdbgp= [KGDB,HW] kgdb over EHCI usb debug port.
2142 Format: <Controller#>[,poll interval]
2143 The controller # is the number of the ehci usb debug
2144 port as it is probed via PCI. The poll interval is
2145 optional and is the number seconds in between
2146 each poll cycle to the debug port in case you need
2147 the functionality for interrupting the kernel with
2148 gdb or control-c on the dbgp connection. When
2149 not using this parameter you use sysrq-g to break into
2150 the kernel debugger.
2151
2152 kgdboc= [KGDB,HW] kgdb over consoles.
2153 Requires a tty driver that supports console polling,
2154 or a supported polling keyboard driver (non-usb).
2155 Serial only format: <serial_device>[,baud]
2156 keyboard only format: kbd
2157 keyboard and serial format: kbd,<serial_device>[,baud]
2158 Optional Kernel mode setting:
2159 kms, kbd format: kms,kbd
2160 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud]
2161
2162 kgdboc_earlycon= [KGDB,HW]
2163 If the boot console provides the ability to read
2164 characters and can work in polling mode, you can use
2165 this parameter to tell kgdb to use it as a backend
2166 until the normal console is registered. Intended to
2167 be used together with the kgdboc parameter which
2168 specifies the normal console to transition to.
2169
2170 The name of the early console should be specified
2171 as the value of this parameter. Note that the name of
2172 the early console might be different than the tty
2173 name passed to kgdboc. It's OK to leave the value
2174 blank and the first boot console that implements
2175 read() will be picked.
2176
2177 kgdbwait [KGDB] Stop kernel execution and enter the
2178 kernel debugger at the earliest opportunity.
2179
2180 kmac= [MIPS] korina ethernet MAC address.
2181 Configure the RouterBoard 532 series on-chip
2182 Ethernet adapter MAC address.
2183
2184 kmemleak= [KNL] Boot-time kmemleak enable/disable
2185 Valid arguments: on, off
2186 Default: on
2187 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y,
2188 the default is off.
2189
2190 kprobe_event=[probe-list]
2191 [FTRACE] Add kprobe events and enable at boot time.
2192 The probe-list is a semicolon delimited list of probe
2193 definitions. Each definition is same as kprobe_events
2194 interface, but the parameters are comma delimited.
2195 For example, to add a kprobe event on vfs_read with
2196 arg1 and arg2, add to the command line;
2197
2198 kprobe_event=p,vfs_read,$arg1,$arg2
2199
2200 See also Documentation/trace/kprobetrace.rst "Kernel
2201 Boot Parameter" section.
2202
2203 kpti= [ARM64] Control page table isolation of user
2204 and kernel address spaces.
2205 Default: enabled on cores which need mitigation.
2206 0: force disabled
2207 1: force enabled
2208
2209 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs.
2210 Default is 0 (don't ignore, but inject #GP)
2211
2212 kvm.enable_vmware_backdoor=[KVM] Support VMware backdoor PV interface.
2213 Default is false (don't support).
2214
2215 kvm.mmu_audit= [KVM] This is a R/W parameter which allows audit
2216 KVM MMU at runtime.
2217 Default is 0 (off)
2218
2219 kvm.nx_huge_pages=
2220 [KVM] Controls the software workaround for the
2221 X86_BUG_ITLB_MULTIHIT bug.
2222 force : Always deploy workaround.
2223 off : Never deploy workaround.
2224 auto : Deploy workaround based on the presence of
2225 X86_BUG_ITLB_MULTIHIT.
2226
2227 Default is 'auto'.
2228
2229 If the software workaround is enabled for the host,
2230 guests do need not to enable it for nested guests.
2231
2232 kvm.nx_huge_pages_recovery_ratio=
2233 [KVM] Controls how many 4KiB pages are periodically zapped
2234 back to huge pages. 0 disables the recovery, otherwise if
2235 the value is N KVM will zap 1/Nth of the 4KiB pages every
2236 minute. The default is 60.
2237
2238 kvm-amd.nested= [KVM,AMD] Allow nested virtualization in KVM/SVM.
2239 Default is 1 (enabled)
2240
2241 kvm-amd.npt= [KVM,AMD] Disable nested paging (virtualized MMU)
2242 for all guests.
2243 Default is 1 (enabled) if in 64-bit or 32-bit PAE mode.
2244
2245 kvm-arm.vgic_v3_group0_trap=
2246 [KVM,ARM] Trap guest accesses to GICv3 group-0
2247 system registers
2248
2249 kvm-arm.vgic_v3_group1_trap=
2250 [KVM,ARM] Trap guest accesses to GICv3 group-1
2251 system registers
2252
2253 kvm-arm.vgic_v3_common_trap=
2254 [KVM,ARM] Trap guest accesses to GICv3 common
2255 system registers
2256
2257 kvm-arm.vgic_v4_enable=
2258 [KVM,ARM] Allow use of GICv4 for direct injection of
2259 LPIs.
2260
2261 kvm-intel.ept= [KVM,Intel] Disable extended page tables
2262 (virtualized MMU) support on capable Intel chips.
2263 Default is 1 (enabled)
2264
2265 kvm-intel.emulate_invalid_guest_state=
2266 [KVM,Intel] Enable emulation of invalid guest states
2267 Default is 0 (disabled)
2268
2269 kvm-intel.flexpriority=
2270 [KVM,Intel] Disable FlexPriority feature (TPR shadow).
2271 Default is 1 (enabled)
2272
2273 kvm-intel.nested=
2274 [KVM,Intel] Enable VMX nesting (nVMX).
2275 Default is 0 (disabled)
2276
2277 kvm-intel.unrestricted_guest=
2278 [KVM,Intel] Disable unrestricted guest feature
2279 (virtualized real and unpaged mode) on capable
2280 Intel chips. Default is 1 (enabled)
2281
2282 kvm-intel.vmentry_l1d_flush=[KVM,Intel] Mitigation for L1 Terminal Fault
2283 CVE-2018-3620.
2284
2285 Valid arguments: never, cond, always
2286
2287 always: L1D cache flush on every VMENTER.
2288 cond: Flush L1D on VMENTER only when the code between
2289 VMEXIT and VMENTER can leak host memory.
2290 never: Disables the mitigation
2291
2292 Default is cond (do L1 cache flush in specific instances)
2293
2294 kvm-intel.vpid= [KVM,Intel] Disable Virtual Processor Identification
2295 feature (tagged TLBs) on capable Intel chips.
2296 Default is 1 (enabled)
2297
2298 l1tf= [X86] Control mitigation of the L1TF vulnerability on
2299 affected CPUs
2300
2301 The kernel PTE inversion protection is unconditionally
2302 enabled and cannot be disabled.
2303
2304 full
2305 Provides all available mitigations for the
2306 L1TF vulnerability. Disables SMT and
2307 enables all mitigations in the
2308 hypervisors, i.e. unconditional L1D flush.
2309
2310 SMT control and L1D flush control via the
2311 sysfs interface is still possible after
2312 boot. Hypervisors will issue a warning
2313 when the first VM is started in a
2314 potentially insecure configuration,
2315 i.e. SMT enabled or L1D flush disabled.
2316
2317 full,force
2318 Same as 'full', but disables SMT and L1D
2319 flush runtime control. Implies the
2320 'nosmt=force' command line option.
2321 (i.e. sysfs control of SMT is disabled.)
2322
2323 flush
2324 Leaves SMT enabled and enables the default
2325 hypervisor mitigation, i.e. conditional
2326 L1D flush.
2327
2328 SMT control and L1D flush control via the
2329 sysfs interface is still possible after
2330 boot. Hypervisors will issue a warning
2331 when the first VM is started in a
2332 potentially insecure configuration,
2333 i.e. SMT enabled or L1D flush disabled.
2334
2335 flush,nosmt
2336
2337 Disables SMT and enables the default
2338 hypervisor mitigation.
2339
2340 SMT control and L1D flush control via the
2341 sysfs interface is still possible after
2342 boot. Hypervisors will issue a warning
2343 when the first VM is started in a
2344 potentially insecure configuration,
2345 i.e. SMT enabled or L1D flush disabled.
2346
2347 flush,nowarn
2348 Same as 'flush', but hypervisors will not
2349 warn when a VM is started in a potentially
2350 insecure configuration.
2351
2352 off
2353 Disables hypervisor mitigations and doesn't
2354 emit any warnings.
2355 It also drops the swap size and available
2356 RAM limit restriction on both hypervisor and
2357 bare metal.
2358
2359 Default is 'flush'.
2360
2361 For details see: Documentation/admin-guide/hw-vuln/l1tf.rst
2362
2363 l2cr= [PPC]
2364
2365 l3cr= [PPC]
2366
2367 lapic [X86-32,APIC] Enable the local APIC even if BIOS
2368 disabled it.
2369
2370 lapic= [X86,APIC] "notscdeadline" Do not use TSC deadline
2371 value for LAPIC timer one-shot implementation. Default
2372 back to the programmable timer unit in the LAPIC.
2373
2374 lapic_timer_c2_ok [X86,APIC] trust the local apic timer
2375 in C2 power state.
2376
2377 libata.dma= [LIBATA] DMA control
2378 libata.dma=0 Disable all PATA and SATA DMA
2379 libata.dma=1 PATA and SATA Disk DMA only
2380 libata.dma=2 ATAPI (CDROM) DMA only
2381 libata.dma=4 Compact Flash DMA only
2382 Combinations also work, so libata.dma=3 enables DMA
2383 for disks and CDROMs, but not CFs.
2384
2385 libata.ignore_hpa= [LIBATA] Ignore HPA limit
2386 libata.ignore_hpa=0 keep BIOS limits (default)
2387 libata.ignore_hpa=1 ignore limits, using full disk
2388
2389 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume
2390 when set.
2391 Format: <int>
2392
2393 libata.force= [LIBATA] Force configurations. The format is comma
2394 separated list of "[ID:]VAL" where ID is
2395 PORT[.DEVICE]. PORT and DEVICE are decimal numbers
2396 matching port, link or device. Basically, it matches
2397 the ATA ID string printed on console by libata. If
2398 the whole ID part is omitted, the last PORT and DEVICE
2399 values are used. If ID hasn't been specified yet, the
2400 configuration applies to all ports, links and devices.
2401
2402 If only DEVICE is omitted, the parameter applies to
2403 the port and all links and devices behind it. DEVICE
2404 number of 0 either selects the first device or the
2405 first fan-out link behind PMP device. It does not
2406 select the host link. DEVICE number of 15 selects the
2407 host link and device attached to it.
2408
2409 The VAL specifies the configuration to force. As long
2410 as there's no ambiguity shortcut notation is allowed.
2411 For example, both 1.5 and 1.5G would work for 1.5Gbps.
2412 The following configurations can be forced.
2413
2414 * Cable type: 40c, 80c, short40c, unk, ign or sata.
2415 Any ID with matching PORT is used.
2416
2417 * SATA link speed limit: 1.5Gbps or 3.0Gbps.
2418
2419 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7].
2420 udma[/][16,25,33,44,66,100,133] notation is also
2421 allowed.
2422
2423 * [no]ncq: Turn on or off NCQ.
2424
2425 * [no]ncqtrim: Turn off queued DSM TRIM.
2426
2427 * nohrst, nosrst, norst: suppress hard, soft
2428 and both resets.
2429
2430 * rstonce: only attempt one reset during
2431 hot-unplug link recovery
2432
2433 * dump_id: dump IDENTIFY data.
2434
2435 * atapi_dmadir: Enable ATAPI DMADIR bridge support
2436
2437 * disable: Disable this device.
2438
2439 If there are multiple matching configurations changing
2440 the same attribute, the last one is used.
2441
2442 memblock=debug [KNL] Enable memblock debug messages.
2443
2444 load_ramdisk= [RAM] List of ramdisks to load from floppy
2445 See Documentation/admin-guide/blockdev/ramdisk.rst.
2446
2447 lockd.nlm_grace_period=P [NFS] Assign grace period.
2448 Format: <integer>
2449
2450 lockd.nlm_tcpport=N [NFS] Assign TCP port.
2451 Format: <integer>
2452
2453 lockd.nlm_timeout=T [NFS] Assign timeout value.
2454 Format: <integer>
2455
2456 lockd.nlm_udpport=M [NFS] Assign UDP port.
2457 Format: <integer>
2458
2459 lockdown= [SECURITY]
2460 { integrity | confidentiality }
2461 Enable the kernel lockdown feature. If set to
2462 integrity, kernel features that allow userland to
2463 modify the running kernel are disabled. If set to
2464 confidentiality, kernel features that allow userland
2465 to extract confidential information from the kernel
2466 are also disabled.
2467
2468 locktorture.nreaders_stress= [KNL]
2469 Set the number of locking read-acquisition kthreads.
2470 Defaults to being automatically set based on the
2471 number of online CPUs.
2472
2473 locktorture.nwriters_stress= [KNL]
2474 Set the number of locking write-acquisition kthreads.
2475
2476 locktorture.onoff_holdoff= [KNL]
2477 Set time (s) after boot for CPU-hotplug testing.
2478
2479 locktorture.onoff_interval= [KNL]
2480 Set time (s) between CPU-hotplug operations, or
2481 zero to disable CPU-hotplug testing.
2482
2483 locktorture.shuffle_interval= [KNL]
2484 Set task-shuffle interval (jiffies). Shuffling
2485 tasks allows some CPUs to go into dyntick-idle
2486 mode during the locktorture test.
2487
2488 locktorture.shutdown_secs= [KNL]
2489 Set time (s) after boot system shutdown. This
2490 is useful for hands-off automated testing.
2491
2492 locktorture.stat_interval= [KNL]
2493 Time (s) between statistics printk()s.
2494
2495 locktorture.stutter= [KNL]
2496 Time (s) to stutter testing, for example,
2497 specifying five seconds causes the test to run for
2498 five seconds, wait for five seconds, and so on.
2499 This tests the locking primitive's ability to
2500 transition abruptly to and from idle.
2501
2502 locktorture.torture_type= [KNL]
2503 Specify the locking implementation to test.
2504
2505 locktorture.verbose= [KNL]
2506 Enable additional printk() statements.
2507
2508 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver
2509 Format: <irq>
2510
2511 loglevel= All Kernel Messages with a loglevel smaller than the
2512 console loglevel will be printed to the console. It can
2513 also be changed with klogd or other programs. The
2514 loglevels are defined as follows:
2515
2516 0 (KERN_EMERG) system is unusable
2517 1 (KERN_ALERT) action must be taken immediately
2518 2 (KERN_CRIT) critical conditions
2519 3 (KERN_ERR) error conditions
2520 4 (KERN_WARNING) warning conditions
2521 5 (KERN_NOTICE) normal but significant condition
2522 6 (KERN_INFO) informational
2523 7 (KERN_DEBUG) debug-level messages
2524
2525 log_buf_len=n[KMG] Sets the size of the printk ring buffer,
2526 in bytes. n must be a power of two and greater
2527 than the minimal size. The minimal size is defined
2528 by LOG_BUF_SHIFT kernel config parameter. There is
2529 also CONFIG_LOG_CPU_MAX_BUF_SHIFT config parameter
2530 that allows to increase the default size depending on
2531 the number of CPUs. See init/Kconfig for more details.
2532
2533 logo.nologo [FB] Disables display of the built-in Linux logo.
2534 This may be used to provide more screen space for
2535 kernel log messages and is useful when debugging
2536 kernel boot problems.
2537
2538 lp=0 [LP] Specify parallel ports to use, e.g,
2539 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses
2540 lp=reset first parallel port). 'lp=0' disables the
2541 lp=auto printer driver. 'lp=reset' (which can be
2542 specified in addition to the ports) causes
2543 attached printers to be reset. Using
2544 lp=port1,port2,... specifies the parallel ports
2545 to associate lp devices with, starting with
2546 lp0. A port specification may be 'none' to skip
2547 that lp device, or a parport name such as
2548 'parport0'. Specifying 'lp=auto' instead of a
2549 port specification list means that device IDs
2550 from each port should be examined, to see if
2551 an IEEE 1284-compliant printer is attached; if
2552 so, the driver will manage that printer.
2553 See also header of drivers/char/lp.c.
2554
2555 lpj=n [KNL]
2556 Sets loops_per_jiffy to given constant, thus avoiding
2557 time-consuming boot-time autodetection (up to 250 ms per
2558 CPU). 0 enables autodetection (default). To determine
2559 the correct value for your kernel, boot with normal
2560 autodetection and see what value is printed. Note that
2561 on SMP systems the preset will be applied to all CPUs,
2562 which is likely to cause problems if your CPUs need
2563 significantly divergent settings. An incorrect value
2564 will cause delays in the kernel to be wrong, leading to
2565 unpredictable I/O errors and other breakage. Although
2566 unlikely, in the extreme case this might damage your
2567 hardware.
2568
2569 ltpc= [NET]
2570 Format: <io>,<irq>,<dma>
2571
2572 lsm.debug [SECURITY] Enable LSM initialization debugging output.
2573
2574 lsm=lsm1,...,lsmN
2575 [SECURITY] Choose order of LSM initialization. This
2576 overrides CONFIG_LSM, and the "security=" parameter.
2577
2578 machvec= [IA-64] Force the use of a particular machine-vector
2579 (machvec) in a generic kernel.
2580 Example: machvec=hpzx1
2581
2582 machtype= [Loongson] Share the same kernel image file between different
2583 yeeloong laptop.
2584 Example: machtype=lemote-yeeloong-2f-7inch
2585
2586 max_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory greater
2587 than or equal to this physical address is ignored.
2588
2589 maxcpus= [SMP] Maximum number of processors that an SMP kernel
2590 will bring up during bootup. maxcpus=n : n >= 0 limits
2591 the kernel to bring up 'n' processors. Surely after
2592 bootup you can bring up the other plugged cpu by executing
2593 "echo 1 > /sys/devices/system/cpu/cpuX/online". So maxcpus
2594 only takes effect during system bootup.
2595 While n=0 is a special case, it is equivalent to "nosmp",
2596 which also disables the IO APIC.
2597
2598 max_loop= [LOOP] The number of loop block devices that get
2599 (loop.max_loop) unconditionally pre-created at init time. The default
2600 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead
2601 of statically allocating a predefined number, loop
2602 devices can be requested on-demand with the
2603 /dev/loop-control interface.
2604
2605 mce [X86-32] Machine Check Exception
2606
2607 mce=option [X86-64] See Documentation/x86/x86_64/boot-options.rst
2608
2609 md= [HW] RAID subsystems devices and level
2610 See Documentation/admin-guide/md.rst.
2611
2612 mdacon= [MDA]
2613 Format: <first>,<last>
2614 Specifies range of consoles to be captured by the MDA.
2615
2616 mds= [X86,INTEL]
2617 Control mitigation for the Micro-architectural Data
2618 Sampling (MDS) vulnerability.
2619
2620 Certain CPUs are vulnerable to an exploit against CPU
2621 internal buffers which can forward information to a
2622 disclosure gadget under certain conditions.
2623
2624 In vulnerable processors, the speculatively
2625 forwarded data can be used in a cache side channel
2626 attack, to access data to which the attacker does
2627 not have direct access.
2628
2629 This parameter controls the MDS mitigation. The
2630 options are:
2631
2632 full - Enable MDS mitigation on vulnerable CPUs
2633 full,nosmt - Enable MDS mitigation and disable
2634 SMT on vulnerable CPUs
2635 off - Unconditionally disable MDS mitigation
2636
2637 On TAA-affected machines, mds=off can be prevented by
2638 an active TAA mitigation as both vulnerabilities are
2639 mitigated with the same mechanism so in order to disable
2640 this mitigation, you need to specify tsx_async_abort=off
2641 too.
2642
2643 Not specifying this option is equivalent to
2644 mds=full.
2645
2646 For details see: Documentation/admin-guide/hw-vuln/mds.rst
2647
2648 mem=nn[KMG] [KNL,BOOT] Force usage of a specific amount of memory
2649 Amount of memory to be used in cases as follows:
2650
2651 1 for test;
2652 2 when the kernel is not able to see the whole system memory;
2653 3 memory that lies after 'mem=' boundary is excluded from
2654 the hypervisor, then assigned to KVM guests.
2655
2656 [X86] Work as limiting max address. Use together
2657 with memmap= to avoid physical address space collisions.
2658 Without memmap= PCI devices could be placed at addresses
2659 belonging to unused RAM.
2660
2661 Note that this only takes effects during boot time since
2662 in above case 3, memory may need be hot added after boot
2663 if system memory of hypervisor is not sufficient.
2664
2665 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel
2666 memory.
2667
2668 memchunk=nn[KMG]
2669 [KNL,SH] Allow user to override the default size for
2670 per-device physically contiguous DMA buffers.
2671
2672 memhp_default_state=online/offline
2673 [KNL] Set the initial state for the memory hotplug
2674 onlining policy. If not specified, the default value is
2675 set according to the
2676 CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE kernel config
2677 option.
2678 See Documentation/admin-guide/mm/memory-hotplug.rst.
2679
2680 memmap=exactmap [KNL,X86] Enable setting of an exact
2681 E820 memory map, as specified by the user.
2682 Such memmap=exactmap lines can be constructed based on
2683 BIOS output or other requirements. See the memmap=nn@ss
2684 option description.
2685
2686 memmap=nn[KMG]@ss[KMG]
2687 [KNL] Force usage of a specific region of memory.
2688 Region of memory to be used is from ss to ss+nn.
2689 If @ss[KMG] is omitted, it is equivalent to mem=nn[KMG],
2690 which limits max address to nn[KMG].
2691 Multiple different regions can be specified,
2692 comma delimited.
2693 Example:
2694 memmap=100M@2G,100M#3G,1G!1024G
2695
2696 memmap=nn[KMG]#ss[KMG]
2697 [KNL,ACPI] Mark specific memory as ACPI data.
2698 Region of memory to be marked is from ss to ss+nn.
2699
2700 memmap=nn[KMG]$ss[KMG]
2701 [KNL,ACPI] Mark specific memory as reserved.
2702 Region of memory to be reserved is from ss to ss+nn.
2703 Example: Exclude memory from 0x18690000-0x1869ffff
2704 memmap=64K$0x18690000
2705 or
2706 memmap=0x10000$0x18690000
2707 Some bootloaders may need an escape character before '$',
2708 like Grub2, otherwise '$' and the following number
2709 will be eaten.
2710
2711 memmap=nn[KMG]!ss[KMG]
2712 [KNL,X86] Mark specific memory as protected.
2713 Region of memory to be used, from ss to ss+nn.
2714 The memory region may be marked as e820 type 12 (0xc)
2715 and is NVDIMM or ADR memory.
2716
2717 memmap=<size>%<offset>-<oldtype>+<newtype>
2718 [KNL,ACPI] Convert memory within the specified region
2719 from <oldtype> to <newtype>. If "-<oldtype>" is left
2720 out, the whole region will be marked as <newtype>,
2721 even if previously unavailable. If "+<newtype>" is left
2722 out, matching memory will be removed. Types are
2723 specified as e820 types, e.g., 1 = RAM, 2 = reserved,
2724 3 = ACPI, 12 = PRAM.
2725
2726 memory_corruption_check=0/1 [X86]
2727 Some BIOSes seem to corrupt the first 64k of
2728 memory when doing things like suspend/resume.
2729 Setting this option will scan the memory
2730 looking for corruption. Enabling this will
2731 both detect corruption and prevent the kernel
2732 from using the memory being corrupted.
2733 However, its intended as a diagnostic tool; if
2734 repeatable BIOS-originated corruption always
2735 affects the same memory, you can use memmap=
2736 to prevent the kernel from using that memory.
2737
2738 memory_corruption_check_size=size [X86]
2739 By default it checks for corruption in the low
2740 64k, making this memory unavailable for normal
2741 use. Use this parameter to scan for
2742 corruption in more or less memory.
2743
2744 memory_corruption_check_period=seconds [X86]
2745 By default it checks for corruption every 60
2746 seconds. Use this parameter to check at some
2747 other rate. 0 disables periodic checking.
2748
2749 memtest= [KNL,X86,ARM,PPC] Enable memtest
2750 Format: <integer>
2751 default : 0 <disable>
2752 Specifies the number of memtest passes to be
2753 performed. Each pass selects another test
2754 pattern from a given set of patterns. Memtest
2755 fills the memory with this pattern, validates
2756 memory contents and reserves bad memory
2757 regions that are detected.
2758
2759 mem_encrypt= [X86-64] AMD Secure Memory Encryption (SME) control
2760 Valid arguments: on, off
2761 Default (depends on kernel configuration option):
2762 on (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=y)
2763 off (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=n)
2764 mem_encrypt=on: Activate SME
2765 mem_encrypt=off: Do not activate SME
2766
2767 Refer to Documentation/virt/kvm/amd-memory-encryption.rst
2768 for details on when memory encryption can be activated.
2769
2770 mem_sleep_default= [SUSPEND] Default system suspend mode:
2771 s2idle - Suspend-To-Idle
2772 shallow - Power-On Suspend or equivalent (if supported)
2773 deep - Suspend-To-RAM or equivalent (if supported)
2774 See Documentation/admin-guide/pm/sleep-states.rst.
2775
2776 meye.*= [HW] Set MotionEye Camera parameters
2777 See Documentation/admin-guide/media/meye.rst.
2778
2779 mfgpt_irq= [IA-32] Specify the IRQ to use for the
2780 Multi-Function General Purpose Timers on AMD Geode
2781 platforms.
2782
2783 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when
2784 the BIOS has incorrectly applied a workaround. TinyBIOS
2785 version 0.98 is known to be affected, 0.99 fixes the
2786 problem by letting the user disable the workaround.
2787
2788 mga= [HW,DRM]
2789
2790 min_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory below this
2791 physical address is ignored.
2792
2793 mini2440= [ARM,HW,KNL]
2794 Format:[0..2][b][c][t]
2795 Default: "0tb"
2796 MINI2440 configuration specification:
2797 0 - The attached screen is the 3.5" TFT
2798 1 - The attached screen is the 7" TFT
2799 2 - The VGA Shield is attached (1024x768)
2800 Leaving out the screen size parameter will not load
2801 the TFT driver, and the framebuffer will be left
2802 unconfigured.
2803 b - Enable backlight. The TFT backlight pin will be
2804 linked to the kernel VESA blanking code and a GPIO
2805 LED. This parameter is not necessary when using the
2806 VGA shield.
2807 c - Enable the s3c camera interface.
2808 t - Reserved for enabling touchscreen support. The
2809 touchscreen support is not enabled in the mainstream
2810 kernel as of 2.6.30, a preliminary port can be found
2811 in the "bleeding edge" mini2440 support kernel at
2812 https://repo.or.cz/w/linux-2.6/mini2440.git
2813
2814 mitigations=
2815 [X86,PPC,S390,ARM64] Control optional mitigations for
2816 CPU vulnerabilities. This is a set of curated,
2817 arch-independent options, each of which is an
2818 aggregation of existing arch-specific options.
2819
2820 off
2821 Disable all optional CPU mitigations. This
2822 improves system performance, but it may also
2823 expose users to several CPU vulnerabilities.
2824 Equivalent to: nopti [X86,PPC]
2825 kpti=0 [ARM64]
2826 nospectre_v1 [X86,PPC]
2827 nobp=0 [S390]
2828 nospectre_v2 [X86,PPC,S390,ARM64]
2829 spectre_v2_user=off [X86]
2830 spec_store_bypass_disable=off [X86,PPC]
2831 ssbd=force-off [ARM64]
2832 l1tf=off [X86]
2833 mds=off [X86]
2834 tsx_async_abort=off [X86]
2835 kvm.nx_huge_pages=off [X86]
2836
2837 Exceptions:
2838 This does not have any effect on
2839 kvm.nx_huge_pages when
2840 kvm.nx_huge_pages=force.
2841
2842 auto (default)
2843 Mitigate all CPU vulnerabilities, but leave SMT
2844 enabled, even if it's vulnerable. This is for
2845 users who don't want to be surprised by SMT
2846 getting disabled across kernel upgrades, or who
2847 have other ways of avoiding SMT-based attacks.
2848 Equivalent to: (default behavior)
2849
2850 auto,nosmt
2851 Mitigate all CPU vulnerabilities, disabling SMT
2852 if needed. This is for users who always want to
2853 be fully mitigated, even if it means losing SMT.
2854 Equivalent to: l1tf=flush,nosmt [X86]
2855 mds=full,nosmt [X86]
2856 tsx_async_abort=full,nosmt [X86]
2857
2858 mminit_loglevel=
2859 [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this
2860 parameter allows control of the logging verbosity for
2861 the additional memory initialisation checks. A value
2862 of 0 disables mminit logging and a level of 4 will
2863 log everything. Information is printed at KERN_DEBUG
2864 so loglevel=8 may also need to be specified.
2865
2866 module.sig_enforce
2867 [KNL] When CONFIG_MODULE_SIG is set, this means that
2868 modules without (valid) signatures will fail to load.
2869 Note that if CONFIG_MODULE_SIG_FORCE is set, that
2870 is always true, so this option does nothing.
2871
2872 module_blacklist= [KNL] Do not load a comma-separated list of
2873 modules. Useful for debugging problem modules.
2874
2875 mousedev.tap_time=
2876 [MOUSE] Maximum time between finger touching and
2877 leaving touchpad surface for touch to be considered
2878 a tap and be reported as a left button click (for
2879 touchpads working in absolute mode only).
2880 Format: <msecs>
2881 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices
2882 reporting absolute coordinates, such as tablets
2883 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices
2884 reporting absolute coordinates, such as tablets
2885
2886 movablecore= [KNL,X86,IA-64,PPC]
2887 Format: nn[KMGTPE] | nn%
2888 This parameter is the complement to kernelcore=, it
2889 specifies the amount of memory used for migratable
2890 allocations. If both kernelcore and movablecore is
2891 specified, then kernelcore will be at *least* the
2892 specified value but may be more. If movablecore on its
2893 own is specified, the administrator must be careful
2894 that the amount of memory usable for all allocations
2895 is not too small.
2896
2897 movable_node [KNL] Boot-time switch to make hotplugable memory
2898 NUMA nodes to be movable. This means that the memory
2899 of such nodes will be usable only for movable
2900 allocations which rules out almost all kernel
2901 allocations. Use with caution!
2902
2903 MTD_Partition= [MTD]
2904 Format: <name>,<region-number>,<size>,<offset>
2905
2906 MTD_Region= [MTD] Format:
2907 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>]
2908
2909 mtdparts= [MTD]
2910 See drivers/mtd/parsers/cmdlinepart.c
2911
2912 multitce=off [PPC] This parameter disables the use of the pSeries
2913 firmware feature for updating multiple TCE entries
2914 at a time.
2915
2916 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration
2917
2918 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock]
2919
2920 boundary - index of last SLC block on Flex-OneNAND.
2921 The remaining blocks are configured as MLC blocks.
2922 lock - Configure if Flex-OneNAND boundary should be locked.
2923 Once locked, the boundary cannot be changed.
2924 1 indicates lock status, 0 indicates unlock status.
2925
2926 mtdset= [ARM]
2927 ARM/S3C2412 JIVE boot control
2928
2929 See arch/arm/mach-s3c2412/mach-jive.c
2930
2931 mtouchusb.raw_coordinates=
2932 [HW] Make the MicroTouch USB driver use raw coordinates
2933 ('y', default) or cooked coordinates ('n')
2934
2935 mtrr_chunk_size=nn[KMG] [X86]
2936 used for mtrr cleanup. It is largest continuous chunk
2937 that could hold holes aka. UC entries.
2938
2939 mtrr_gran_size=nn[KMG] [X86]
2940 Used for mtrr cleanup. It is granularity of mtrr block.
2941 Default is 1.
2942 Large value could prevent small alignment from
2943 using up MTRRs.
2944
2945 mtrr_spare_reg_nr=n [X86]
2946 Format: <integer>
2947 Range: 0,7 : spare reg number
2948 Default : 1
2949 Used for mtrr cleanup. It is spare mtrr entries number.
2950 Set to 2 or more if your graphical card needs more.
2951
2952 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card
2953
2954 netdev= [NET] Network devices parameters
2955 Format: <irq>,<io>,<mem_start>,<mem_end>,<name>
2956 Note that mem_start is often overloaded to mean
2957 something different and driver-specific.
2958 This usage is only documented in each driver source
2959 file if at all.
2960
2961 nf_conntrack.acct=
2962 [NETFILTER] Enable connection tracking flow accounting
2963 0 to disable accounting
2964 1 to enable accounting
2965 Default value is 0.
2966
2967 nfsaddrs= [NFS] Deprecated. Use ip= instead.
2968 See Documentation/admin-guide/nfs/nfsroot.rst.
2969
2970 nfsroot= [NFS] nfs root filesystem for disk-less boxes.
2971 See Documentation/admin-guide/nfs/nfsroot.rst.
2972
2973 nfsrootdebug [NFS] enable nfsroot debugging messages.
2974 See Documentation/admin-guide/nfs/nfsroot.rst.
2975
2976 nfs.callback_nr_threads=
2977 [NFSv4] set the total number of threads that the
2978 NFS client will assign to service NFSv4 callback
2979 requests.
2980
2981 nfs.callback_tcpport=
2982 [NFS] set the TCP port on which the NFSv4 callback
2983 channel should listen.
2984
2985 nfs.cache_getent=
2986 [NFS] sets the pathname to the program which is used
2987 to update the NFS client cache entries.
2988
2989 nfs.cache_getent_timeout=
2990 [NFS] sets the timeout after which an attempt to
2991 update a cache entry is deemed to have failed.
2992
2993 nfs.idmap_cache_timeout=
2994 [NFS] set the maximum lifetime for idmapper cache
2995 entries.
2996
2997 nfs.enable_ino64=
2998 [NFS] enable 64-bit inode numbers.
2999 If zero, the NFS client will fake up a 32-bit inode
3000 number for the readdir() and stat() syscalls instead
3001 of returning the full 64-bit number.
3002 The default is to return 64-bit inode numbers.
3003
3004 nfs.max_session_cb_slots=
3005 [NFSv4.1] Sets the maximum number of session
3006 slots the client will assign to the callback
3007 channel. This determines the maximum number of
3008 callbacks the client will process in parallel for
3009 a particular server.
3010
3011 nfs.max_session_slots=
3012 [NFSv4.1] Sets the maximum number of session slots
3013 the client will attempt to negotiate with the server.
3014 This limits the number of simultaneous RPC requests
3015 that the client can send to the NFSv4.1 server.
3016 Note that there is little point in setting this
3017 value higher than the max_tcp_slot_table_limit.
3018
3019 nfs.nfs4_disable_idmapping=
3020 [NFSv4] When set to the default of '1', this option
3021 ensures that both the RPC level authentication
3022 scheme and the NFS level operations agree to use
3023 numeric uids/gids if the mount is using the
3024 'sec=sys' security flavour. In effect it is
3025 disabling idmapping, which can make migration from
3026 legacy NFSv2/v3 systems to NFSv4 easier.
3027 Servers that do not support this mode of operation
3028 will be autodetected by the client, and it will fall
3029 back to using the idmapper.
3030 To turn off this behaviour, set the value to '0'.
3031 nfs.nfs4_unique_id=
3032 [NFS4] Specify an additional fixed unique ident-
3033 ification string that NFSv4 clients can insert into
3034 their nfs_client_id4 string. This is typically a
3035 UUID that is generated at system install time.
3036
3037 nfs.send_implementation_id =
3038 [NFSv4.1] Send client implementation identification
3039 information in exchange_id requests.
3040 If zero, no implementation identification information
3041 will be sent.
3042 The default is to send the implementation identification
3043 information.
3044
3045 nfs.recover_lost_locks =
3046 [NFSv4] Attempt to recover locks that were lost due
3047 to a lease timeout on the server. Please note that
3048 doing this risks data corruption, since there are
3049 no guarantees that the file will remain unchanged
3050 after the locks are lost.
3051 If you want to enable the kernel legacy behaviour of
3052 attempting to recover these locks, then set this
3053 parameter to '1'.
3054 The default parameter value of '0' causes the kernel
3055 not to attempt recovery of lost locks.
3056
3057 nfs4.layoutstats_timer =
3058 [NFSv4.2] Change the rate at which the kernel sends
3059 layoutstats to the pNFS metadata server.
3060
3061 Setting this to value to 0 causes the kernel to use
3062 whatever value is the default set by the layout
3063 driver. A non-zero value sets the minimum interval
3064 in seconds between layoutstats transmissions.
3065
3066 nfsd.nfs4_disable_idmapping=
3067 [NFSv4] When set to the default of '1', the NFSv4
3068 server will return only numeric uids and gids to
3069 clients using auth_sys, and will accept numeric uids
3070 and gids from such clients. This is intended to ease
3071 migration from NFSv2/v3.
3072
3073 nmi_debug= [KNL,SH] Specify one or more actions to take
3074 when a NMI is triggered.
3075 Format: [state][,regs][,debounce][,die]
3076
3077 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels
3078 Format: [panic,][nopanic,][num]
3079 Valid num: 0 or 1
3080 0 - turn hardlockup detector in nmi_watchdog off
3081 1 - turn hardlockup detector in nmi_watchdog on
3082 When panic is specified, panic when an NMI watchdog
3083 timeout occurs (or 'nopanic' to not panic on an NMI
3084 watchdog, if CONFIG_BOOTPARAM_HARDLOCKUP_PANIC is set)
3085 To disable both hard and soft lockup detectors,
3086 please see 'nowatchdog'.
3087 This is useful when you use a panic=... timeout and
3088 need the box quickly up again.
3089
3090 These settings can be accessed at runtime via
3091 the nmi_watchdog and hardlockup_panic sysctls.
3092
3093 netpoll.carrier_timeout=
3094 [NET] Specifies amount of time (in seconds) that
3095 netpoll should wait for a carrier. By default netpoll
3096 waits 4 seconds.
3097
3098 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths
3099 emulation library even if a 387 maths coprocessor
3100 is present.
3101
3102 no5lvl [X86-64] Disable 5-level paging mode. Forces
3103 kernel to use 4-level paging instead.
3104
3105 nofsgsbase [X86] Disables FSGSBASE instructions.
3106
3107 no_console_suspend
3108 [HW] Never suspend the console
3109 Disable suspending of consoles during suspend and
3110 hibernate operations. Once disabled, debugging
3111 messages can reach various consoles while the rest
3112 of the system is being put to sleep (ie, while
3113 debugging driver suspend/resume hooks). This may
3114 not work reliably with all consoles, but is known
3115 to work with serial and VGA consoles.
3116 To facilitate more flexible debugging, we also add
3117 console_suspend, a printk module parameter to control
3118 it. Users could use console_suspend (usually
3119 /sys/module/printk/parameters/console_suspend) to
3120 turn on/off it dynamically.
3121
3122 novmcoredd [KNL,KDUMP]
3123 Disable device dump. Device dump allows drivers to
3124 append dump data to vmcore so you can collect driver
3125 specified debug info. Drivers can append the data
3126 without any limit and this data is stored in memory,
3127 so this may cause significant memory stress. Disabling
3128 device dump can help save memory but the driver debug
3129 data will be no longer available. This parameter
3130 is only available when CONFIG_PROC_VMCORE_DEVICE_DUMP
3131 is set.
3132
3133 noaliencache [MM, NUMA, SLAB] Disables the allocation of alien
3134 caches in the slab allocator. Saves per-node memory,
3135 but will impact performance.
3136
3137 noalign [KNL,ARM]
3138
3139 noaltinstr [S390] Disables alternative instructions patching
3140 (CPU alternatives feature).
3141
3142 noapic [SMP,APIC] Tells the kernel to not make use of any
3143 IOAPICs that may be present in the system.
3144
3145 noautogroup Disable scheduler automatic task group creation.
3146
3147 nobats [PPC] Do not use BATs for mapping kernel lowmem
3148 on "Classic" PPC cores.
3149
3150 nocache [ARM]
3151
3152 noclflush [BUGS=X86] Don't use the CLFLUSH instruction
3153
3154 nodelayacct [KNL] Disable per-task delay accounting
3155
3156 nodsp [SH] Disable hardware DSP at boot time.
3157
3158 noefi Disable EFI runtime services support.
3159
3160 noexec [IA-64]
3161
3162 noexec [X86]
3163 On X86-32 available only on PAE configured kernels.
3164 noexec=on: enable non-executable mappings (default)
3165 noexec=off: disable non-executable mappings
3166
3167 nosmap [X86,PPC]
3168 Disable SMAP (Supervisor Mode Access Prevention)
3169 even if it is supported by processor.
3170
3171 nosmep [X86,PPC]
3172 Disable SMEP (Supervisor Mode Execution Prevention)
3173 even if it is supported by processor.
3174
3175 noexec32 [X86-64]
3176 This affects only 32-bit executables.
3177 noexec32=on: enable non-executable mappings (default)
3178 read doesn't imply executable mappings
3179 noexec32=off: disable non-executable mappings
3180 read implies executable mappings
3181
3182 nofpu [MIPS,SH] Disable hardware FPU at boot time.
3183
3184 nofxsr [BUGS=X86-32] Disables x86 floating point extended
3185 register save and restore. The kernel will only save
3186 legacy floating-point registers on task switch.
3187
3188 nohugeiomap [KNL,X86,PPC] Disable kernel huge I/O mappings.
3189
3190 nosmt [KNL,S390] Disable symmetric multithreading (SMT).
3191 Equivalent to smt=1.
3192
3193 [KNL,X86] Disable symmetric multithreading (SMT).
3194 nosmt=force: Force disable SMT, cannot be undone
3195 via the sysfs control file.
3196
3197 nospectre_v1 [X86,PPC] Disable mitigations for Spectre Variant 1
3198 (bounds check bypass). With this option data leaks are
3199 possible in the system.
3200
3201 nospectre_v2 [X86,PPC_FSL_BOOK3E,ARM64] Disable all mitigations for
3202 the Spectre variant 2 (indirect branch prediction)
3203 vulnerability. System may allow data leaks with this
3204 option.
3205
3206 nospec_store_bypass_disable
3207 [HW] Disable all mitigations for the Speculative Store Bypass vulnerability
3208
3209 noxsave [BUGS=X86] Disables x86 extended register state save
3210 and restore using xsave. The kernel will fallback to
3211 enabling legacy floating-point and sse state.
3212
3213 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended
3214 register states. The kernel will fall back to use
3215 xsave to save the states. By using this parameter,
3216 performance of saving the states is degraded because
3217 xsave doesn't support modified optimization while
3218 xsaveopt supports it on xsaveopt enabled systems.
3219
3220 noxsaves [X86] Disables xsaves and xrstors used in saving and
3221 restoring x86 extended register state in compacted
3222 form of xsave area. The kernel will fall back to use
3223 xsaveopt and xrstor to save and restore the states
3224 in standard form of xsave area. By using this
3225 parameter, xsave area per process might occupy more
3226 memory on xsaves enabled systems.
3227
3228 nohlt [BUGS=ARM,SH] Tells the kernel that the sleep(SH) or
3229 wfi(ARM) instruction doesn't work correctly and not to
3230 use it. This is also useful when using JTAG debugger.
3231
3232 no_file_caps Tells the kernel not to honor file capabilities. The
3233 only way then for a file to be executed with privilege
3234 is to be setuid root or executed by root.
3235
3236 nohalt [IA-64] Tells the kernel not to use the power saving
3237 function PAL_HALT_LIGHT when idle. This increases
3238 power-consumption. On the positive side, it reduces
3239 interrupt wake-up latency, which may improve performance
3240 in certain environments such as networked servers or
3241 real-time systems.
3242
3243 nohibernate [HIBERNATION] Disable hibernation and resume.
3244
3245 nohz= [KNL] Boottime enable/disable dynamic ticks
3246 Valid arguments: on, off
3247 Default: on
3248
3249 nohz_full= [KNL,BOOT,SMP,ISOL]
3250 The argument is a cpu list, as described above.
3251 In kernels built with CONFIG_NO_HZ_FULL=y, set
3252 the specified list of CPUs whose tick will be stopped
3253 whenever possible. The boot CPU will be forced outside
3254 the range to maintain the timekeeping. Any CPUs
3255 in this list will have their RCU callbacks offloaded,
3256 just as if they had also been called out in the
3257 rcu_nocbs= boot parameter.
3258
3259 noiotrap [SH] Disables trapped I/O port accesses.
3260
3261 noirqdebug [X86-32] Disables the code which attempts to detect and
3262 disable unhandled interrupt sources.
3263
3264 no_timer_check [X86,APIC] Disables the code which tests for
3265 broken timer IRQ sources.
3266
3267 noisapnp [ISAPNP] Disables ISA PnP code.
3268
3269 noinitrd [RAM] Tells the kernel not to load any configured
3270 initial RAM disk.
3271
3272 nointremap [X86-64, Intel-IOMMU] Do not enable interrupt
3273 remapping.
3274 [Deprecated - use intremap=off]
3275
3276 nointroute [IA-64]
3277
3278 noinvpcid [X86] Disable the INVPCID cpu feature.
3279
3280 nojitter [IA-64] Disables jitter checking for ITC timers.
3281
3282 no-kvmclock [X86,KVM] Disable paravirtualized KVM clock driver
3283
3284 no-kvmapf [X86,KVM] Disable paravirtualized asynchronous page
3285 fault handling.
3286
3287 no-vmw-sched-clock
3288 [X86,PV_OPS] Disable paravirtualized VMware scheduler
3289 clock and use the default one.
3290
3291 no-steal-acc [X86,PV_OPS,ARM64] Disable paravirtualized steal time
3292 accounting. steal time is computed, but won't
3293 influence scheduler behaviour
3294
3295 nolapic [X86-32,APIC] Do not enable or use the local APIC.
3296
3297 nolapic_timer [X86-32,APIC] Do not use the local APIC timer.
3298
3299 noltlbs [PPC] Do not use large page/tlb entries for kernel
3300 lowmem mapping on PPC40x and PPC8xx
3301
3302 nomca [IA-64] Disable machine check abort handling
3303
3304 nomce [X86-32] Disable Machine Check Exception
3305
3306 nomfgpt [X86-32] Disable Multi-Function General Purpose
3307 Timer usage (for AMD Geode machines).
3308
3309 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to
3310 shutdown the other cpus. Instead use the REBOOT_VECTOR
3311 irq.
3312
3313 nomodule Disable module load
3314
3315 nopat [X86] Disable PAT (page attribute table extension of
3316 pagetables) support.
3317
3318 nopcid [X86-64] Disable the PCID cpu feature.
3319
3320 norandmaps Don't use address space randomization. Equivalent to
3321 echo 0 > /proc/sys/kernel/randomize_va_space
3322
3323 noreplace-smp [X86-32,SMP] Don't replace SMP instructions
3324 with UP alternatives
3325
3326 nordrand [X86] Disable kernel use of the RDRAND and
3327 RDSEED instructions even if they are supported
3328 by the processor. RDRAND and RDSEED are still
3329 available to user space applications.
3330
3331 noresume [SWSUSP] Disables resume and restores original swap
3332 space.
3333
3334 no-scroll [VGA] Disables scrollback.
3335 This is required for the Braillex ib80-piezo Braille
3336 reader made by F.H. Papenmeier (Germany).
3337
3338 nosbagart [IA-64]
3339
3340 nosep [BUGS=X86-32] Disables x86 SYSENTER/SYSEXIT support.
3341
3342 nosmp [SMP] Tells an SMP kernel to act as a UP kernel,
3343 and disable the IO APIC. legacy for "maxcpus=0".
3344
3345 nosoftlockup [KNL] Disable the soft-lockup detector.
3346
3347 nosync [HW,M68K] Disables sync negotiation for all devices.
3348
3349 nowatchdog [KNL] Disable both lockup detectors, i.e.
3350 soft-lockup and NMI watchdog (hard-lockup).
3351
3352 nowb [ARM]
3353
3354 nox2apic [X86-64,APIC] Do not enable x2APIC mode.
3355
3356 cpu0_hotplug [X86] Turn on CPU0 hotplug feature when
3357 CONFIG_BOOTPARAM_HOTPLUG_CPU0 is off.
3358 Some features depend on CPU0. Known dependencies are:
3359 1. Resume from suspend/hibernate depends on CPU0.
3360 Suspend/hibernate will fail if CPU0 is offline and you
3361 need to online CPU0 before suspend/hibernate.
3362 2. PIC interrupts also depend on CPU0. CPU0 can't be
3363 removed if a PIC interrupt is detected.
3364 It's said poweroff/reboot may depend on CPU0 on some
3365 machines although I haven't seen such issues so far
3366 after CPU0 is offline on a few tested machines.
3367 If the dependencies are under your control, you can
3368 turn on cpu0_hotplug.
3369
3370 nps_mtm_hs_ctr= [KNL,ARC]
3371 This parameter sets the maximum duration, in
3372 cycles, each HW thread of the CTOP can run
3373 without interruptions, before HW switches it.
3374 The actual maximum duration is 16 times this
3375 parameter's value.
3376 Format: integer between 1 and 255
3377 Default: 255
3378
3379 nptcg= [IA-64] Override max number of concurrent global TLB
3380 purges which is reported from either PAL_VM_SUMMARY or
3381 SAL PALO.
3382
3383 nr_cpus= [SMP] Maximum number of processors that an SMP kernel
3384 could support. nr_cpus=n : n >= 1 limits the kernel to
3385 support 'n' processors. It could be larger than the
3386 number of already plugged CPU during bootup, later in
3387 runtime you can physically add extra cpu until it reaches
3388 n. So during boot up some boot time memory for per-cpu
3389 variables need be pre-allocated for later physical cpu
3390 hot plugging.
3391
3392 nr_uarts= [SERIAL] maximum number of UARTs to be registered.
3393
3394 numa_balancing= [KNL,X86] Enable or disable automatic NUMA balancing.
3395 Allowed values are enable and disable
3396
3397 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA.
3398 'node', 'default' can be specified
3399 This can be set from sysctl after boot.
3400 See Documentation/admin-guide/sysctl/vm.rst for details.
3401
3402 ohci1394_dma=early [HW] enable debugging via the ohci1394 driver.
3403 See Documentation/core-api/debugging-via-ohci1394.rst for more
3404 info.
3405
3406 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands
3407 Rather than timing out after 20 ms if an EC
3408 command is not properly ACKed, override the length
3409 of the timeout. We have interrupts disabled while
3410 waiting for the ACK, so if this is set too high
3411 interrupts *may* be lost!
3412
3413 omap_mux= [OMAP] Override bootloader pin multiplexing.
3414 Format: <mux_mode0.mode_name=value>...
3415 For example, to override I2C bus2:
3416 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100
3417
3418 oprofile.timer= [HW]
3419 Use timer interrupt instead of performance counters
3420
3421 oprofile.cpu_type= Force an oprofile cpu type
3422 This might be useful if you have an older oprofile
3423 userland or if you want common events.
3424 Format: { arch_perfmon }
3425 arch_perfmon: [X86] Force use of architectural
3426 perfmon on Intel CPUs instead of the
3427 CPU specific event set.
3428 timer: [X86] Force use of architectural NMI
3429 timer mode (see also oprofile.timer
3430 for generic hr timer mode)
3431
3432 oops=panic Always panic on oopses. Default is to just kill the
3433 process, but there is a small probability of
3434 deadlocking the machine.
3435 This will also cause panics on machine check exceptions.
3436 Useful together with panic=30 to trigger a reboot.
3437
3438 page_alloc.shuffle=
3439 [KNL] Boolean flag to control whether the page allocator
3440 should randomize its free lists. The randomization may
3441 be automatically enabled if the kernel detects it is
3442 running on a platform with a direct-mapped memory-side
3443 cache, and this parameter can be used to
3444 override/disable that behavior. The state of the flag
3445 can be read from sysfs at:
3446 /sys/module/page_alloc/parameters/shuffle.
3447
3448 page_owner= [KNL] Boot-time page_owner enabling option.
3449 Storage of the information about who allocated
3450 each page is disabled in default. With this switch,
3451 we can turn it on.
3452 on: enable the feature
3453
3454 page_poison= [KNL] Boot-time parameter changing the state of
3455 poisoning on the buddy allocator, available with
3456 CONFIG_PAGE_POISONING=y.
3457 off: turn off poisoning (default)
3458 on: turn on poisoning
3459
3460 panic= [KNL] Kernel behaviour on panic: delay <timeout>
3461 timeout > 0: seconds before rebooting
3462 timeout = 0: wait forever
3463 timeout < 0: reboot immediately
3464 Format: <timeout>
3465
3466 panic_print= Bitmask for printing system info when panic happens.
3467 User can chose combination of the following bits:
3468 bit 0: print all tasks info
3469 bit 1: print system memory info
3470 bit 2: print timer info
3471 bit 3: print locks info if CONFIG_LOCKDEP is on
3472 bit 4: print ftrace buffer
3473 bit 5: print all printk messages in buffer
3474
3475 panic_on_taint= Bitmask for conditionally calling panic() in add_taint()
3476 Format: <hex>[,nousertaint]
3477 Hexadecimal bitmask representing the set of TAINT flags
3478 that will cause the kernel to panic when add_taint() is
3479 called with any of the flags in this set.
3480 The optional switch "nousertaint" can be utilized to
3481 prevent userspace forced crashes by writing to sysctl
3482 /proc/sys/kernel/tainted any flagset matching with the
3483 bitmask set on panic_on_taint.
3484 See Documentation/admin-guide/tainted-kernels.rst for
3485 extra details on the taint flags that users can pick
3486 to compose the bitmask to assign to panic_on_taint.
3487
3488 panic_on_warn panic() instead of WARN(). Useful to cause kdump
3489 on a WARN().
3490
3491 crash_kexec_post_notifiers
3492 Run kdump after running panic-notifiers and dumping
3493 kmsg. This only for the users who doubt kdump always
3494 succeeds in any situation.
3495 Note that this also increases risks of kdump failure,
3496 because some panic notifiers can make the crashed
3497 kernel more unstable.
3498
3499 parkbd.port= [HW] Parallel port number the keyboard adapter is
3500 connected to, default is 0.
3501 Format: <parport#>
3502 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation,
3503 0 for XT, 1 for AT (default is AT).
3504 Format: <mode>
3505
3506 parport= [HW,PPT] Specify parallel ports. 0 disables.
3507 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] }
3508 Use 'auto' to force the driver to use any
3509 IRQ/DMA settings detected (the default is to
3510 ignore detected IRQ/DMA settings because of
3511 possible conflicts). You can specify the base
3512 address, IRQ, and DMA settings; IRQ and DMA
3513 should be numbers, or 'auto' (for using detected
3514 settings on that particular port), or 'nofifo'
3515 (to avoid using a FIFO even if it is detected).
3516 Parallel ports are assigned in the order they
3517 are specified on the command line, starting
3518 with parport0.
3519
3520 parport_init_mode= [HW,PPT]
3521 Configure VIA parallel port to operate in
3522 a specific mode. This is necessary on Pegasos
3523 computer where firmware has no options for setting
3524 up parallel port mode and sets it to spp.
3525 Currently this function knows 686a and 8231 chips.
3526 Format: [spp|ps2|epp|ecp|ecpepp]
3527
3528 pause_on_oops=
3529 Halt all CPUs after the first oops has been printed for
3530 the specified number of seconds. This is to be used if
3531 your oopses keep scrolling off the screen.
3532
3533 pcbit= [HW,ISDN]
3534
3535 pcd. [PARIDE]
3536 See header of drivers/block/paride/pcd.c.
3537 See also Documentation/admin-guide/blockdev/paride.rst.
3538
3539 pci=option[,option...] [PCI] various PCI subsystem options.
3540
3541 Some options herein operate on a specific device
3542 or a set of devices (<pci_dev>). These are
3543 specified in one of the following formats:
3544
3545 [<domain>:]<bus>:<dev>.<func>[/<dev>.<func>]*
3546 pci:<vendor>:<device>[:<subvendor>:<subdevice>]
3547
3548 Note: the first format specifies a PCI
3549 bus/device/function address which may change
3550 if new hardware is inserted, if motherboard
3551 firmware changes, or due to changes caused
3552 by other kernel parameters. If the
3553 domain is left unspecified, it is
3554 taken to be zero. Optionally, a path
3555 to a device through multiple device/function
3556 addresses can be specified after the base
3557 address (this is more robust against
3558 renumbering issues). The second format
3559 selects devices using IDs from the
3560 configuration space which may match multiple
3561 devices in the system.
3562
3563 earlydump dump PCI config space before the kernel
3564 changes anything
3565 off [X86] don't probe for the PCI bus
3566 bios [X86-32] force use of PCI BIOS, don't access
3567 the hardware directly. Use this if your machine
3568 has a non-standard PCI host bridge.
3569 nobios [X86-32] disallow use of PCI BIOS, only direct
3570 hardware access methods are allowed. Use this
3571 if you experience crashes upon bootup and you
3572 suspect they are caused by the BIOS.
3573 conf1 [X86] Force use of PCI Configuration Access
3574 Mechanism 1 (config address in IO port 0xCF8,
3575 data in IO port 0xCFC, both 32-bit).
3576 conf2 [X86] Force use of PCI Configuration Access
3577 Mechanism 2 (IO port 0xCF8 is an 8-bit port for
3578 the function, IO port 0xCFA, also 8-bit, sets
3579 bus number. The config space is then accessed
3580 through ports 0xC000-0xCFFF).
3581 See http://wiki.osdev.org/PCI for more info
3582 on the configuration access mechanisms.
3583 noaer [PCIE] If the PCIEAER kernel config parameter is
3584 enabled, this kernel boot option can be used to
3585 disable the use of PCIE advanced error reporting.
3586 nodomains [PCI] Disable support for multiple PCI
3587 root domains (aka PCI segments, in ACPI-speak).
3588 nommconf [X86] Disable use of MMCONFIG for PCI
3589 Configuration
3590 check_enable_amd_mmconf [X86] check for and enable
3591 properly configured MMIO access to PCI
3592 config space on AMD family 10h CPU
3593 nomsi [MSI] If the PCI_MSI kernel config parameter is
3594 enabled, this kernel boot option can be used to
3595 disable the use of MSI interrupts system-wide.
3596 noioapicquirk [APIC] Disable all boot interrupt quirks.
3597 Safety option to keep boot IRQs enabled. This
3598 should never be necessary.
3599 ioapicreroute [APIC] Enable rerouting of boot IRQs to the
3600 primary IO-APIC for bridges that cannot disable
3601 boot IRQs. This fixes a source of spurious IRQs
3602 when the system masks IRQs.
3603 noioapicreroute [APIC] Disable workaround that uses the
3604 boot IRQ equivalent of an IRQ that connects to
3605 a chipset where boot IRQs cannot be disabled.
3606 The opposite of ioapicreroute.
3607 biosirq [X86-32] Use PCI BIOS calls to get the interrupt
3608 routing table. These calls are known to be buggy
3609 on several machines and they hang the machine
3610 when used, but on other computers it's the only
3611 way to get the interrupt routing table. Try
3612 this option if the kernel is unable to allocate
3613 IRQs or discover secondary PCI buses on your
3614 motherboard.
3615 rom [X86] Assign address space to expansion ROMs.
3616 Use with caution as certain devices share
3617 address decoders between ROMs and other
3618 resources.
3619 norom [X86] Do not assign address space to
3620 expansion ROMs that do not already have
3621 BIOS assigned address ranges.
3622 nobar [X86] Do not assign address space to the
3623 BARs that weren't assigned by the BIOS.
3624 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be
3625 assigned automatically to PCI devices. You can
3626 make the kernel exclude IRQs of your ISA cards
3627 this way.
3628 pirqaddr=0xAAAAA [X86] Specify the physical address
3629 of the PIRQ table (normally generated
3630 by the BIOS) if it is outside the
3631 F0000h-100000h range.
3632 lastbus=N [X86] Scan all buses thru bus #N. Can be
3633 useful if the kernel is unable to find your
3634 secondary buses and you want to tell it
3635 explicitly which ones they are.
3636 assign-busses [X86] Always assign all PCI bus
3637 numbers ourselves, overriding
3638 whatever the firmware may have done.
3639 usepirqmask [X86] Honor the possible IRQ mask stored
3640 in the BIOS $PIR table. This is needed on
3641 some systems with broken BIOSes, notably
3642 some HP Pavilion N5400 and Omnibook XE3
3643 notebooks. This will have no effect if ACPI
3644 IRQ routing is enabled.
3645 noacpi [X86] Do not use ACPI for IRQ routing
3646 or for PCI scanning.
3647 use_crs [X86] Use PCI host bridge window information
3648 from ACPI. On BIOSes from 2008 or later, this
3649 is enabled by default. If you need to use this,
3650 please report a bug.
3651 nocrs [X86] Ignore PCI host bridge windows from ACPI.
3652 If you need to use this, please report a bug.
3653 routeirq Do IRQ routing for all PCI devices.
3654 This is normally done in pci_enable_device(),
3655 so this option is a temporary workaround
3656 for broken drivers that don't call it.
3657 skip_isa_align [X86] do not align io start addr, so can
3658 handle more pci cards
3659 noearly [X86] Don't do any early type 1 scanning.
3660 This might help on some broken boards which
3661 machine check when some devices' config space
3662 is read. But various workarounds are disabled
3663 and some IOMMU drivers will not work.
3664 bfsort Sort PCI devices into breadth-first order.
3665 This sorting is done to get a device
3666 order compatible with older (<= 2.4) kernels.
3667 nobfsort Don't sort PCI devices into breadth-first order.
3668 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size)
3669 tuning and use the BIOS-configured MPS defaults.
3670 pcie_bus_safe Set every device's MPS to the largest value
3671 supported by all devices below the root complex.
3672 pcie_bus_perf Set device MPS to the largest allowable MPS
3673 based on its parent bus. Also set MRRS (Max
3674 Read Request Size) to the largest supported
3675 value (no larger than the MPS that the device
3676 or bus can support) for best performance.
3677 pcie_bus_peer2peer Set every device's MPS to 128B, which
3678 every device is guaranteed to support. This
3679 configuration allows peer-to-peer DMA between
3680 any pair of devices, possibly at the cost of
3681 reduced performance. This also guarantees
3682 that hot-added devices will work.
3683 cbiosize=nn[KMG] The fixed amount of bus space which is
3684 reserved for the CardBus bridge's IO window.
3685 The default value is 256 bytes.
3686 cbmemsize=nn[KMG] The fixed amount of bus space which is
3687 reserved for the CardBus bridge's memory
3688 window. The default value is 64 megabytes.
3689 resource_alignment=
3690 Format:
3691 [<order of align>@]<pci_dev>[; ...]
3692 Specifies alignment and device to reassign
3693 aligned memory resources. How to
3694 specify the device is described above.
3695 If <order of align> is not specified,
3696 PAGE_SIZE is used as alignment.
3697 A PCI-PCI bridge can be specified if resource
3698 windows need to be expanded.
3699 To specify the alignment for several
3700 instances of a device, the PCI vendor,
3701 device, subvendor, and subdevice may be
3702 specified, e.g., 12@pci:8086:9c22:103c:198f
3703 for 4096-byte alignment.
3704 ecrc= Enable/disable PCIe ECRC (transaction layer
3705 end-to-end CRC checking).
3706 bios: Use BIOS/firmware settings. This is the
3707 the default.
3708 off: Turn ECRC off
3709 on: Turn ECRC on.
3710 hpiosize=nn[KMG] The fixed amount of bus space which is
3711 reserved for hotplug bridge's IO window.
3712 Default size is 256 bytes.
3713 hpmmiosize=nn[KMG] The fixed amount of bus space which is
3714 reserved for hotplug bridge's MMIO window.
3715 Default size is 2 megabytes.
3716 hpmmioprefsize=nn[KMG] The fixed amount of bus space which is
3717 reserved for hotplug bridge's MMIO_PREF window.
3718 Default size is 2 megabytes.
3719 hpmemsize=nn[KMG] The fixed amount of bus space which is
3720 reserved for hotplug bridge's MMIO and
3721 MMIO_PREF window.
3722 Default size is 2 megabytes.
3723 hpbussize=nn The minimum amount of additional bus numbers
3724 reserved for buses below a hotplug bridge.
3725 Default is 1.
3726 realloc= Enable/disable reallocating PCI bridge resources
3727 if allocations done by BIOS are too small to
3728 accommodate resources required by all child
3729 devices.
3730 off: Turn realloc off
3731 on: Turn realloc on
3732 realloc same as realloc=on
3733 noari do not use PCIe ARI.
3734 noats [PCIE, Intel-IOMMU, AMD-IOMMU]
3735 do not use PCIe ATS (and IOMMU device IOTLB).
3736 pcie_scan_all Scan all possible PCIe devices. Otherwise we
3737 only look for one device below a PCIe downstream
3738 port.
3739 big_root_window Try to add a big 64bit memory window to the PCIe
3740 root complex on AMD CPUs. Some GFX hardware
3741 can resize a BAR to allow access to all VRAM.
3742 Adding the window is slightly risky (it may
3743 conflict with unreported devices), so this
3744 taints the kernel.
3745 disable_acs_redir=<pci_dev>[; ...]
3746 Specify one or more PCI devices (in the format
3747 specified above) separated by semicolons.
3748 Each device specified will have the PCI ACS
3749 redirect capabilities forced off which will
3750 allow P2P traffic between devices through
3751 bridges without forcing it upstream. Note:
3752 this removes isolation between devices and
3753 may put more devices in an IOMMU group.
3754 force_floating [S390] Force usage of floating interrupts.
3755 nomio [S390] Do not use MIO instructions.
3756 norid [S390] ignore the RID field and force use of
3757 one PCI domain per PCI function
3758
3759 pcie_aspm= [PCIE] Forcibly enable or disable PCIe Active State Power
3760 Management.
3761 off Disable ASPM.
3762 force Enable ASPM even on devices that claim not to support it.
3763 WARNING: Forcing ASPM on may cause system lockups.
3764
3765 pcie_ports= [PCIE] PCIe port services handling:
3766 native Use native PCIe services (PME, AER, DPC, PCIe hotplug)
3767 even if the platform doesn't give the OS permission to
3768 use them. This may cause conflicts if the platform
3769 also tries to use these services.
3770 dpc-native Use native PCIe service for DPC only. May
3771 cause conflicts if firmware uses AER or DPC.
3772 compat Disable native PCIe services (PME, AER, DPC, PCIe
3773 hotplug).
3774
3775 pcie_port_pm= [PCIE] PCIe port power management handling:
3776 off Disable power management of all PCIe ports
3777 force Forcibly enable power management of all PCIe ports
3778
3779 pcie_pme= [PCIE,PM] Native PCIe PME signaling options:
3780 nomsi Do not use MSI for native PCIe PME signaling (this makes
3781 all PCIe root ports use INTx for all services).
3782
3783 pcmv= [HW,PCMCIA] BadgePAD 4
3784
3785 pd_ignore_unused
3786 [PM]
3787 Keep all power-domains already enabled by bootloader on,
3788 even if no driver has claimed them. This is useful
3789 for debug and development, but should not be
3790 needed on a platform with proper driver support.
3791
3792 pd. [PARIDE]
3793 See Documentation/admin-guide/blockdev/paride.rst.
3794
3795 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at
3796 boot time.
3797 Format: { 0 | 1 }
3798 See arch/parisc/kernel/pdc_chassis.c
3799
3800 percpu_alloc= Select which percpu first chunk allocator to use.
3801 Currently supported values are "embed" and "page".
3802 Archs may support subset or none of the selections.
3803 See comments in mm/percpu.c for details on each
3804 allocator. This parameter is primarily for debugging
3805 and performance comparison.
3806
3807 pf. [PARIDE]
3808 See Documentation/admin-guide/blockdev/paride.rst.
3809
3810 pg. [PARIDE]
3811 See Documentation/admin-guide/blockdev/paride.rst.
3812
3813 pirq= [SMP,APIC] Manual mp-table setup
3814 See Documentation/x86/i386/IO-APIC.rst.
3815
3816 plip= [PPT,NET] Parallel port network link
3817 Format: { parport<nr> | timid | 0 }
3818 See also Documentation/admin-guide/parport.rst.
3819
3820 pmtmr= [X86] Manual setup of pmtmr I/O Port.
3821 Override pmtimer IOPort with a hex value.
3822 e.g. pmtmr=0x508
3823
3824 pm_debug_messages [SUSPEND,KNL]
3825 Enable suspend/resume debug messages during boot up.
3826
3827 pnp.debug=1 [PNP]
3828 Enable PNP debug messages (depends on the
3829 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time
3830 via /sys/module/pnp/parameters/debug. We always show
3831 current resource usage; turning this on also shows
3832 possible settings and some assignment information.
3833
3834 pnpacpi= [ACPI]
3835 { off }
3836
3837 pnpbios= [ISAPNP]
3838 { on | off | curr | res | no-curr | no-res }
3839
3840 pnp_reserve_irq=
3841 [ISAPNP] Exclude IRQs for the autoconfiguration
3842
3843 pnp_reserve_dma=
3844 [ISAPNP] Exclude DMAs for the autoconfiguration
3845
3846 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration
3847 Ranges are in pairs (I/O port base and size).
3848
3849 pnp_reserve_mem=
3850 [ISAPNP] Exclude memory regions for the
3851 autoconfiguration.
3852 Ranges are in pairs (memory base and size).
3853
3854 ports= [IP_VS_FTP] IPVS ftp helper module
3855 Default is 21.
3856 Up to 8 (IP_VS_APP_MAX_PORTS) ports
3857 may be specified.
3858 Format: <port>,<port>....
3859
3860 powersave=off [PPC] This option disables power saving features.
3861 It specifically disables cpuidle and sets the
3862 platform machine description specific power_save
3863 function to NULL. On Idle the CPU just reduces
3864 execution priority.
3865
3866 ppc_strict_facility_enable
3867 [PPC] This option catches any kernel floating point,
3868 Altivec, VSX and SPE outside of regions specifically
3869 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()).
3870 There is some performance impact when enabling this.
3871
3872 ppc_tm= [PPC]
3873 Format: {"off"}
3874 Disable Hardware Transactional Memory
3875
3876 print-fatal-signals=
3877 [KNL] debug: print fatal signals
3878
3879 If enabled, warn about various signal handling
3880 related application anomalies: too many signals,
3881 too many POSIX.1 timers, fatal signals causing a
3882 coredump - etc.
3883
3884 If you hit the warning due to signal overflow,
3885 you might want to try "ulimit -i unlimited".
3886
3887 default: off.
3888
3889 printk.always_kmsg_dump=
3890 Trigger kmsg_dump for cases other than kernel oops or
3891 panics
3892 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
3893 default: disabled
3894
3895 printk.devkmsg={on,off,ratelimit}
3896 Control writing to /dev/kmsg.
3897 on - unlimited logging to /dev/kmsg from userspace
3898 off - logging to /dev/kmsg disabled
3899 ratelimit - ratelimit the logging
3900 Default: ratelimit
3901
3902 printk.time= Show timing data prefixed to each printk message line
3903 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
3904
3905 processor.max_cstate= [HW,ACPI]
3906 Limit processor to maximum C-state
3907 max_cstate=9 overrides any DMI blacklist limit.
3908
3909 processor.nocst [HW,ACPI]
3910 Ignore the _CST method to determine C-states,
3911 instead using the legacy FADT method
3912
3913 profile= [KNL] Enable kernel profiling via /proc/profile
3914 Format: [<profiletype>,]<number>
3915 Param: <profiletype>: "schedule", "sleep", or "kvm"
3916 [defaults to kernel profiling]
3917 Param: "schedule" - profile schedule points.
3918 Param: "sleep" - profile D-state sleeping (millisecs).
3919 Requires CONFIG_SCHEDSTATS
3920 Param: "kvm" - profile VM exits.
3921 Param: <number> - step/bucket size as a power of 2 for
3922 statistical time based profiling.
3923
3924 prompt_ramdisk= [RAM] List of RAM disks to prompt for floppy disk
3925 before loading.
3926 See Documentation/admin-guide/blockdev/ramdisk.rst.
3927
3928 prot_virt= [S390] enable hosting protected virtual machines
3929 isolated from the hypervisor (if hardware supports
3930 that).
3931 Format: <bool>
3932
3933 psi= [KNL] Enable or disable pressure stall information
3934 tracking.
3935 Format: <bool>
3936
3937 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to
3938 probe for; one of (bare|imps|exps|lifebook|any).
3939 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports
3940 per second.
3941 psmouse.resetafter= [HW,MOUSE]
3942 Try to reset the device after so many bad packets
3943 (0 = never).
3944 psmouse.resolution=
3945 [HW,MOUSE] Set desired mouse resolution, in dpi.
3946 psmouse.smartscroll=
3947 [HW,MOUSE] Controls Logitech smartscroll autorepeat.
3948 0 = disabled, 1 = enabled (default).
3949
3950 pstore.backend= Specify the name of the pstore backend to use
3951
3952 pt. [PARIDE]
3953 See Documentation/admin-guide/blockdev/paride.rst.
3954
3955 pti= [X86-64] Control Page Table Isolation of user and
3956 kernel address spaces. Disabling this feature
3957 removes hardening, but improves performance of
3958 system calls and interrupts.
3959
3960 on - unconditionally enable
3961 off - unconditionally disable
3962 auto - kernel detects whether your CPU model is
3963 vulnerable to issues that PTI mitigates
3964
3965 Not specifying this option is equivalent to pti=auto.
3966
3967 nopti [X86-64]
3968 Equivalent to pti=off
3969
3970 pty.legacy_count=
3971 [KNL] Number of legacy pty's. Overwrites compiled-in
3972 default number.
3973
3974 quiet [KNL] Disable most log messages
3975
3976 r128= [HW,DRM]
3977
3978 raid= [HW,RAID]
3979 See Documentation/admin-guide/md.rst.
3980
3981 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes
3982 See Documentation/admin-guide/blockdev/ramdisk.rst.
3983
3984 random.trust_cpu={on,off}
3985 [KNL] Enable or disable trusting the use of the
3986 CPU's random number generator (if available) to
3987 fully seed the kernel's CRNG. Default is controlled
3988 by CONFIG_RANDOM_TRUST_CPU.
3989
3990 ras=option[,option,...] [KNL] RAS-specific options
3991
3992 cec_disable [X86]
3993 Disable the Correctable Errors Collector,
3994 see CONFIG_RAS_CEC help text.
3995
3996 rcu_nocbs= [KNL]
3997 The argument is a cpu list, as described above,
3998 except that the string "all" can be used to
3999 specify every CPU on the system.
4000
4001 In kernels built with CONFIG_RCU_NOCB_CPU=y, set
4002 the specified list of CPUs to be no-callback CPUs.
4003 Invocation of these CPUs' RCU callbacks will be
4004 offloaded to "rcuox/N" kthreads created for that
4005 purpose, where "x" is "p" for RCU-preempt, and
4006 "s" for RCU-sched, and "N" is the CPU number.
4007 This reduces OS jitter on the offloaded CPUs,
4008 which can be useful for HPC and real-time
4009 workloads. It can also improve energy efficiency
4010 for asymmetric multiprocessors.
4011
4012 rcu_nocb_poll [KNL]
4013 Rather than requiring that offloaded CPUs
4014 (specified by rcu_nocbs= above) explicitly
4015 awaken the corresponding "rcuoN" kthreads,
4016 make these kthreads poll for callbacks.
4017 This improves the real-time response for the
4018 offloaded CPUs by relieving them of the need to
4019 wake up the corresponding kthread, but degrades
4020 energy efficiency by requiring that the kthreads
4021 periodically wake up to do the polling.
4022
4023 rcutree.blimit= [KNL]
4024 Set maximum number of finished RCU callbacks to
4025 process in one batch.
4026
4027 rcutree.dump_tree= [KNL]
4028 Dump the structure of the rcu_node combining tree
4029 out at early boot. This is used for diagnostic
4030 purposes, to verify correct tree setup.
4031
4032 rcutree.gp_cleanup_delay= [KNL]
4033 Set the number of jiffies to delay each step of
4034 RCU grace-period cleanup.
4035
4036 rcutree.gp_init_delay= [KNL]
4037 Set the number of jiffies to delay each step of
4038 RCU grace-period initialization.
4039
4040 rcutree.gp_preinit_delay= [KNL]
4041 Set the number of jiffies to delay each step of
4042 RCU grace-period pre-initialization, that is,
4043 the propagation of recent CPU-hotplug changes up
4044 the rcu_node combining tree.
4045
4046 rcutree.use_softirq= [KNL]
4047 If set to zero, move all RCU_SOFTIRQ processing to
4048 per-CPU rcuc kthreads. Defaults to a non-zero
4049 value, meaning that RCU_SOFTIRQ is used by default.
4050 Specify rcutree.use_softirq=0 to use rcuc kthreads.
4051
4052 rcutree.rcu_fanout_exact= [KNL]
4053 Disable autobalancing of the rcu_node combining
4054 tree. This is used by rcutorture, and might
4055 possibly be useful for architectures having high
4056 cache-to-cache transfer latencies.
4057
4058 rcutree.rcu_fanout_leaf= [KNL]
4059 Change the number of CPUs assigned to each
4060 leaf rcu_node structure. Useful for very
4061 large systems, which will choose the value 64,
4062 and for NUMA systems with large remote-access
4063 latencies, which will choose a value aligned
4064 with the appropriate hardware boundaries.
4065
4066 rcutree.rcu_min_cached_objs= [KNL]
4067 Minimum number of objects which are cached and
4068 maintained per one CPU. Object size is equal
4069 to PAGE_SIZE. The cache allows to reduce the
4070 pressure to page allocator, also it makes the
4071 whole algorithm to behave better in low memory
4072 condition.
4073
4074 rcutree.jiffies_till_first_fqs= [KNL]
4075 Set delay from grace-period initialization to
4076 first attempt to force quiescent states.
4077 Units are jiffies, minimum value is zero,
4078 and maximum value is HZ.
4079
4080 rcutree.jiffies_till_next_fqs= [KNL]
4081 Set delay between subsequent attempts to force
4082 quiescent states. Units are jiffies, minimum
4083 value is one, and maximum value is HZ.
4084
4085 rcutree.jiffies_till_sched_qs= [KNL]
4086 Set required age in jiffies for a
4087 given grace period before RCU starts
4088 soliciting quiescent-state help from
4089 rcu_note_context_switch() and cond_resched().
4090 If not specified, the kernel will calculate
4091 a value based on the most recent settings
4092 of rcutree.jiffies_till_first_fqs
4093 and rcutree.jiffies_till_next_fqs.
4094 This calculated value may be viewed in
4095 rcutree.jiffies_to_sched_qs. Any attempt to set
4096 rcutree.jiffies_to_sched_qs will be cheerfully
4097 overwritten.
4098
4099 rcutree.kthread_prio= [KNL,BOOT]
4100 Set the SCHED_FIFO priority of the RCU per-CPU
4101 kthreads (rcuc/N). This value is also used for
4102 the priority of the RCU boost threads (rcub/N)
4103 and for the RCU grace-period kthreads (rcu_bh,
4104 rcu_preempt, and rcu_sched). If RCU_BOOST is
4105 set, valid values are 1-99 and the default is 1
4106 (the least-favored priority). Otherwise, when
4107 RCU_BOOST is not set, valid values are 0-99 and
4108 the default is zero (non-realtime operation).
4109
4110 rcutree.rcu_nocb_gp_stride= [KNL]
4111 Set the number of NOCB callback kthreads in
4112 each group, which defaults to the square root
4113 of the number of CPUs. Larger numbers reduce
4114 the wakeup overhead on the global grace-period
4115 kthread, but increases that same overhead on
4116 each group's NOCB grace-period kthread.
4117
4118 rcutree.qhimark= [KNL]
4119 Set threshold of queued RCU callbacks beyond which
4120 batch limiting is disabled.
4121
4122 rcutree.qlowmark= [KNL]
4123 Set threshold of queued RCU callbacks below which
4124 batch limiting is re-enabled.
4125
4126 rcutree.qovld= [KNL]
4127 Set threshold of queued RCU callbacks beyond which
4128 RCU's force-quiescent-state scan will aggressively
4129 enlist help from cond_resched() and sched IPIs to
4130 help CPUs more quickly reach quiescent states.
4131 Set to less than zero to make this be set based
4132 on rcutree.qhimark at boot time and to zero to
4133 disable more aggressive help enlistment.
4134
4135 rcutree.rcu_idle_gp_delay= [KNL]
4136 Set wakeup interval for idle CPUs that have
4137 RCU callbacks (RCU_FAST_NO_HZ=y).
4138
4139 rcutree.rcu_idle_lazy_gp_delay= [KNL]
4140 Set wakeup interval for idle CPUs that have
4141 only "lazy" RCU callbacks (RCU_FAST_NO_HZ=y).
4142 Lazy RCU callbacks are those which RCU can
4143 prove do nothing more than free memory.
4144
4145 rcutree.rcu_kick_kthreads= [KNL]
4146 Cause the grace-period kthread to get an extra
4147 wake_up() if it sleeps three times longer than
4148 it should at force-quiescent-state time.
4149 This wake_up() will be accompanied by a
4150 WARN_ONCE() splat and an ftrace_dump().
4151
4152 rcutree.sysrq_rcu= [KNL]
4153 Commandeer a sysrq key to dump out Tree RCU's
4154 rcu_node tree with an eye towards determining
4155 why a new grace period has not yet started.
4156
4157 rcuperf.gp_async= [KNL]
4158 Measure performance of asynchronous
4159 grace-period primitives such as call_rcu().
4160
4161 rcuperf.gp_async_max= [KNL]
4162 Specify the maximum number of outstanding
4163 callbacks per writer thread. When a writer
4164 thread exceeds this limit, it invokes the
4165 corresponding flavor of rcu_barrier() to allow
4166 previously posted callbacks to drain.
4167
4168 rcuperf.gp_exp= [KNL]
4169 Measure performance of expedited synchronous
4170 grace-period primitives.
4171
4172 rcuperf.holdoff= [KNL]
4173 Set test-start holdoff period. The purpose of
4174 this parameter is to delay the start of the
4175 test until boot completes in order to avoid
4176 interference.
4177
4178 rcuperf.kfree_rcu_test= [KNL]
4179 Set to measure performance of kfree_rcu() flooding.
4180
4181 rcuperf.kfree_nthreads= [KNL]
4182 The number of threads running loops of kfree_rcu().
4183
4184 rcuperf.kfree_alloc_num= [KNL]
4185 Number of allocations and frees done in an iteration.
4186
4187 rcuperf.kfree_loops= [KNL]
4188 Number of loops doing rcuperf.kfree_alloc_num number
4189 of allocations and frees.
4190
4191 rcuperf.nreaders= [KNL]
4192 Set number of RCU readers. The value -1 selects
4193 N, where N is the number of CPUs. A value
4194 "n" less than -1 selects N-n+1, where N is again
4195 the number of CPUs. For example, -2 selects N
4196 (the number of CPUs), -3 selects N+1, and so on.
4197 A value of "n" less than or equal to -N selects
4198 a single reader.
4199
4200 rcuperf.nwriters= [KNL]
4201 Set number of RCU writers. The values operate
4202 the same as for rcuperf.nreaders.
4203 N, where N is the number of CPUs
4204
4205 rcuperf.perf_type= [KNL]
4206 Specify the RCU implementation to test.
4207
4208 rcuperf.shutdown= [KNL]
4209 Shut the system down after performance tests
4210 complete. This is useful for hands-off automated
4211 testing.
4212
4213 rcuperf.verbose= [KNL]
4214 Enable additional printk() statements.
4215
4216 rcuperf.writer_holdoff= [KNL]
4217 Write-side holdoff between grace periods,
4218 in microseconds. The default of zero says
4219 no holdoff.
4220
4221 rcutorture.fqs_duration= [KNL]
4222 Set duration of force_quiescent_state bursts
4223 in microseconds.
4224
4225 rcutorture.fqs_holdoff= [KNL]
4226 Set holdoff time within force_quiescent_state bursts
4227 in microseconds.
4228
4229 rcutorture.fqs_stutter= [KNL]
4230 Set wait time between force_quiescent_state bursts
4231 in seconds.
4232
4233 rcutorture.fwd_progress= [KNL]
4234 Enable RCU grace-period forward-progress testing
4235 for the types of RCU supporting this notion.
4236
4237 rcutorture.fwd_progress_div= [KNL]
4238 Specify the fraction of a CPU-stall-warning
4239 period to do tight-loop forward-progress testing.
4240
4241 rcutorture.fwd_progress_holdoff= [KNL]
4242 Number of seconds to wait between successive
4243 forward-progress tests.
4244
4245 rcutorture.fwd_progress_need_resched= [KNL]
4246 Enclose cond_resched() calls within checks for
4247 need_resched() during tight-loop forward-progress
4248 testing.
4249
4250 rcutorture.gp_cond= [KNL]
4251 Use conditional/asynchronous update-side
4252 primitives, if available.
4253
4254 rcutorture.gp_exp= [KNL]
4255 Use expedited update-side primitives, if available.
4256
4257 rcutorture.gp_normal= [KNL]
4258 Use normal (non-expedited) asynchronous
4259 update-side primitives, if available.
4260
4261 rcutorture.gp_sync= [KNL]
4262 Use normal (non-expedited) synchronous
4263 update-side primitives, if available. If all
4264 of rcutorture.gp_cond=, rcutorture.gp_exp=,
4265 rcutorture.gp_normal=, and rcutorture.gp_sync=
4266 are zero, rcutorture acts as if is interpreted
4267 they are all non-zero.
4268
4269 rcutorture.n_barrier_cbs= [KNL]
4270 Set callbacks/threads for rcu_barrier() testing.
4271
4272 rcutorture.nfakewriters= [KNL]
4273 Set number of concurrent RCU writers. These just
4274 stress RCU, they don't participate in the actual
4275 test, hence the "fake".
4276
4277 rcutorture.nreaders= [KNL]
4278 Set number of RCU readers. The value -1 selects
4279 N-1, where N is the number of CPUs. A value
4280 "n" less than -1 selects N-n-2, where N is again
4281 the number of CPUs. For example, -2 selects N
4282 (the number of CPUs), -3 selects N+1, and so on.
4283
4284 rcutorture.object_debug= [KNL]
4285 Enable debug-object double-call_rcu() testing.
4286
4287 rcutorture.onoff_holdoff= [KNL]
4288 Set time (s) after boot for CPU-hotplug testing.
4289
4290 rcutorture.onoff_interval= [KNL]
4291 Set time (jiffies) between CPU-hotplug operations,
4292 or zero to disable CPU-hotplug testing.
4293
4294 rcutorture.read_exit= [KNL]
4295 Set the number of read-then-exit kthreads used
4296 to test the interaction of RCU updaters and
4297 task-exit processing.
4298
4299 rcutorture.read_exit_burst= [KNL]
4300 The number of times in a given read-then-exit
4301 episode that a set of read-then-exit kthreads
4302 is spawned.
4303
4304 rcutorture.read_exit_delay= [KNL]
4305 The delay, in seconds, between successive
4306 read-then-exit testing episodes.
4307
4308 rcutorture.shuffle_interval= [KNL]
4309 Set task-shuffle interval (s). Shuffling tasks
4310 allows some CPUs to go into dyntick-idle mode
4311 during the rcutorture test.
4312
4313 rcutorture.shutdown_secs= [KNL]
4314 Set time (s) after boot system shutdown. This
4315 is useful for hands-off automated testing.
4316
4317 rcutorture.stall_cpu= [KNL]
4318 Duration of CPU stall (s) to test RCU CPU stall
4319 warnings, zero to disable.
4320
4321 rcutorture.stall_cpu_block= [KNL]
4322 Sleep while stalling if set. This will result
4323 in warnings from preemptible RCU in addition
4324 to any other stall-related activity.
4325
4326 rcutorture.stall_cpu_holdoff= [KNL]
4327 Time to wait (s) after boot before inducing stall.
4328
4329 rcutorture.stall_cpu_irqsoff= [KNL]
4330 Disable interrupts while stalling if set.
4331
4332 rcutorture.stall_gp_kthread= [KNL]
4333 Duration (s) of forced sleep within RCU
4334 grace-period kthread to test RCU CPU stall
4335 warnings, zero to disable. If both stall_cpu
4336 and stall_gp_kthread are specified, the
4337 kthread is starved first, then the CPU.
4338
4339 rcutorture.stat_interval= [KNL]
4340 Time (s) between statistics printk()s.
4341
4342 rcutorture.stutter= [KNL]
4343 Time (s) to stutter testing, for example, specifying
4344 five seconds causes the test to run for five seconds,
4345 wait for five seconds, and so on. This tests RCU's
4346 ability to transition abruptly to and from idle.
4347
4348 rcutorture.test_boost= [KNL]
4349 Test RCU priority boosting? 0=no, 1=maybe, 2=yes.
4350 "Maybe" means test if the RCU implementation
4351 under test support RCU priority boosting.
4352
4353 rcutorture.test_boost_duration= [KNL]
4354 Duration (s) of each individual boost test.
4355
4356 rcutorture.test_boost_interval= [KNL]
4357 Interval (s) between each boost test.
4358
4359 rcutorture.test_no_idle_hz= [KNL]
4360 Test RCU's dyntick-idle handling. See also the
4361 rcutorture.shuffle_interval parameter.
4362
4363 rcutorture.torture_type= [KNL]
4364 Specify the RCU implementation to test.
4365
4366 rcutorture.verbose= [KNL]
4367 Enable additional printk() statements.
4368
4369 rcupdate.rcu_cpu_stall_ftrace_dump= [KNL]
4370 Dump ftrace buffer after reporting RCU CPU
4371 stall warning.
4372
4373 rcupdate.rcu_cpu_stall_suppress= [KNL]
4374 Suppress RCU CPU stall warning messages.
4375
4376 rcupdate.rcu_cpu_stall_suppress_at_boot= [KNL]
4377 Suppress RCU CPU stall warning messages and
4378 rcutorture writer stall warnings that occur
4379 during early boot, that is, during the time
4380 before the init task is spawned.
4381
4382 rcupdate.rcu_cpu_stall_timeout= [KNL]
4383 Set timeout for RCU CPU stall warning messages.
4384
4385 rcupdate.rcu_expedited= [KNL]
4386 Use expedited grace-period primitives, for
4387 example, synchronize_rcu_expedited() instead
4388 of synchronize_rcu(). This reduces latency,
4389 but can increase CPU utilization, degrade
4390 real-time latency, and degrade energy efficiency.
4391 No effect on CONFIG_TINY_RCU kernels.
4392
4393 rcupdate.rcu_normal= [KNL]
4394 Use only normal grace-period primitives,
4395 for example, synchronize_rcu() instead of
4396 synchronize_rcu_expedited(). This improves
4397 real-time latency, CPU utilization, and
4398 energy efficiency, but can expose users to
4399 increased grace-period latency. This parameter
4400 overrides rcupdate.rcu_expedited. No effect on
4401 CONFIG_TINY_RCU kernels.
4402
4403 rcupdate.rcu_normal_after_boot= [KNL]
4404 Once boot has completed (that is, after
4405 rcu_end_inkernel_boot() has been invoked), use
4406 only normal grace-period primitives. No effect
4407 on CONFIG_TINY_RCU kernels.
4408
4409 rcupdate.rcu_task_ipi_delay= [KNL]
4410 Set time in jiffies during which RCU tasks will
4411 avoid sending IPIs, starting with the beginning
4412 of a given grace period. Setting a large
4413 number avoids disturbing real-time workloads,
4414 but lengthens grace periods.
4415
4416 rcupdate.rcu_task_stall_timeout= [KNL]
4417 Set timeout in jiffies for RCU task stall warning
4418 messages. Disable with a value less than or equal
4419 to zero.
4420
4421 rcupdate.rcu_self_test= [KNL]
4422 Run the RCU early boot self tests
4423
4424 rdinit= [KNL]
4425 Format: <full_path>
4426 Run specified binary instead of /init from the ramdisk,
4427 used for early userspace startup. See initrd.
4428
4429 rdrand= [X86]
4430 force - Override the decision by the kernel to hide the
4431 advertisement of RDRAND support (this affects
4432 certain AMD processors because of buggy BIOS
4433 support, specifically around the suspend/resume
4434 path).
4435
4436 rdt= [HW,X86,RDT]
4437 Turn on/off individual RDT features. List is:
4438 cmt, mbmtotal, mbmlocal, l3cat, l3cdp, l2cat, l2cdp,
4439 mba.
4440 E.g. to turn on cmt and turn off mba use:
4441 rdt=cmt,!mba
4442
4443 reboot= [KNL]
4444 Format (x86 or x86_64):
4445 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] \
4446 [[,]s[mp]#### \
4447 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \
4448 [[,]f[orce]
4449 Where reboot_mode is one of warm (soft) or cold (hard) or gpio
4450 (prefix with 'panic_' to set mode for panic
4451 reboot only),
4452 reboot_type is one of bios, acpi, kbd, triple, efi, or pci,
4453 reboot_force is either force or not specified,
4454 reboot_cpu is s[mp]#### with #### being the processor
4455 to be used for rebooting.
4456
4457 refscale.holdoff= [KNL]
4458 Set test-start holdoff period. The purpose of
4459 this parameter is to delay the start of the
4460 test until boot completes in order to avoid
4461 interference.
4462
4463 refscale.loops= [KNL]
4464 Set the number of loops over the synchronization
4465 primitive under test. Increasing this number
4466 reduces noise due to loop start/end overhead,
4467 but the default has already reduced the per-pass
4468 noise to a handful of picoseconds on ca. 2020
4469 x86 laptops.
4470
4471 refscale.nreaders= [KNL]
4472 Set number of readers. The default value of -1
4473 selects N, where N is roughly 75% of the number
4474 of CPUs. A value of zero is an interesting choice.
4475
4476 refscale.nruns= [KNL]
4477 Set number of runs, each of which is dumped onto
4478 the console log.
4479
4480 refscale.readdelay= [KNL]
4481 Set the read-side critical-section duration,
4482 measured in microseconds.
4483
4484 refscale.scale_type= [KNL]
4485 Specify the read-protection implementation to test.
4486
4487 refscale.shutdown= [KNL]
4488 Shut down the system at the end of the performance
4489 test. This defaults to 1 (shut it down) when
4490 rcuperf is built into the kernel and to 0 (leave
4491 it running) when rcuperf is built as a module.
4492
4493 refscale.verbose= [KNL]
4494 Enable additional printk() statements.
4495
4496 relax_domain_level=
4497 [KNL, SMP] Set scheduler's default relax_domain_level.
4498 See Documentation/admin-guide/cgroup-v1/cpusets.rst.
4499
4500 reserve= [KNL,BUGS] Force kernel to ignore I/O ports or memory
4501 Format: <base1>,<size1>[,<base2>,<size2>,...]
4502 Reserve I/O ports or memory so the kernel won't use
4503 them. If <base> is less than 0x10000, the region
4504 is assumed to be I/O ports; otherwise it is memory.
4505
4506 reservetop= [X86-32]
4507 Format: nn[KMG]
4508 Reserves a hole at the top of the kernel virtual
4509 address space.
4510
4511 reservelow= [X86]
4512 Format: nn[K]
4513 Set the amount of memory to reserve for BIOS at
4514 the bottom of the address space.
4515
4516 reset_devices [KNL] Force drivers to reset the underlying device
4517 during initialization.
4518
4519 resume= [SWSUSP]
4520 Specify the partition device for software suspend
4521 Format:
4522 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>}
4523
4524 resume_offset= [SWSUSP]
4525 Specify the offset from the beginning of the partition
4526 given by "resume=" at which the swap header is located,
4527 in <PAGE_SIZE> units (needed only for swap files).
4528 See Documentation/power/swsusp-and-swap-files.rst
4529
4530 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to
4531 read the resume files
4532
4533 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up.
4534 Useful for devices that are detected asynchronously
4535 (e.g. USB and MMC devices).
4536
4537 hibernate= [HIBERNATION]
4538 noresume Don't check if there's a hibernation image
4539 present during boot.
4540 nocompress Don't compress/decompress hibernation images.
4541 no Disable hibernation and resume.
4542 protect_image Turn on image protection during restoration
4543 (that will set all pages holding image data
4544 during restoration read-only).
4545
4546 retain_initrd [RAM] Keep initrd memory after extraction
4547
4548 rfkill.default_state=
4549 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm,
4550 etc. communication is blocked by default.
4551 1 Unblocked.
4552
4553 rfkill.master_switch_mode=
4554 0 The "airplane mode" button does nothing.
4555 1 The "airplane mode" button toggles between everything
4556 blocked and the previous configuration.
4557 2 The "airplane mode" button toggles between everything
4558 blocked and everything unblocked.
4559
4560 rhash_entries= [KNL,NET]
4561 Set number of hash buckets for route cache
4562
4563 ring3mwait=disable
4564 [KNL] Disable ring 3 MONITOR/MWAIT feature on supported
4565 CPUs.
4566
4567 ro [KNL] Mount root device read-only on boot
4568
4569 rodata= [KNL]
4570 on Mark read-only kernel memory as read-only (default).
4571 off Leave read-only kernel memory writable for debugging.
4572
4573 rockchip.usb_uart
4574 Enable the uart passthrough on the designated usb port
4575 on Rockchip SoCs. When active, the signals of the
4576 debug-uart get routed to the D+ and D- pins of the usb
4577 port and the regular usb controller gets disabled.
4578
4579 root= [KNL] Root filesystem
4580 See name_to_dev_t comment in init/do_mounts.c.
4581
4582 rootdelay= [KNL] Delay (in seconds) to pause before attempting to
4583 mount the root filesystem
4584
4585 rootflags= [KNL] Set root filesystem mount option string
4586
4587 rootfstype= [KNL] Set root filesystem type
4588
4589 rootwait [KNL] Wait (indefinitely) for root device to show up.
4590 Useful for devices that are detected asynchronously
4591 (e.g. USB and MMC devices).
4592
4593 rproc_mem=nn[KMG][@address]
4594 [KNL,ARM,CMA] Remoteproc physical memory block.
4595 Memory area to be used by remote processor image,
4596 managed by CMA.
4597
4598 rw [KNL] Mount root device read-write on boot
4599
4600 S [KNL] Run init in single mode
4601
4602 s390_iommu= [HW,S390]
4603 Set s390 IOTLB flushing mode
4604 strict
4605 With strict flushing every unmap operation will result in
4606 an IOTLB flush. Default is lazy flushing before reuse,
4607 which is faster.
4608
4609 sa1100ir [NET]
4610 See drivers/net/irda/sa1100_ir.c.
4611
4612 sbni= [NET] Granch SBNI12 leased line adapter
4613
4614 sched_debug [KNL] Enables verbose scheduler debug messages.
4615
4616 schedstats= [KNL,X86] Enable or disable scheduled statistics.
4617 Allowed values are enable and disable. This feature
4618 incurs a small amount of overhead in the scheduler
4619 but is useful for debugging and performance tuning.
4620
4621 sched_thermal_decay_shift=
4622 [KNL, SMP] Set a decay shift for scheduler thermal
4623 pressure signal. Thermal pressure signal follows the
4624 default decay period of other scheduler pelt
4625 signals(usually 32 ms but configurable). Setting
4626 sched_thermal_decay_shift will left shift the decay
4627 period for the thermal pressure signal by the shift
4628 value.
4629 i.e. with the default pelt decay period of 32 ms
4630 sched_thermal_decay_shift thermal pressure decay pr
4631 1 64 ms
4632 2 128 ms
4633 and so on.
4634 Format: integer between 0 and 10
4635 Default is 0.
4636
4637 skew_tick= [KNL] Offset the periodic timer tick per cpu to mitigate
4638 xtime_lock contention on larger systems, and/or RCU lock
4639 contention on all systems with CONFIG_MAXSMP set.
4640 Format: { "0" | "1" }
4641 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1"
4642 1 -- enable.
4643 Note: increases power consumption, thus should only be
4644 enabled if running jitter sensitive (HPC/RT) workloads.
4645
4646 security= [SECURITY] Choose a legacy "major" security module to
4647 enable at boot. This has been deprecated by the
4648 "lsm=" parameter.
4649
4650 selinux= [SELINUX] Disable or enable SELinux at boot time.
4651 Format: { "0" | "1" }
4652 See security/selinux/Kconfig help text.
4653 0 -- disable.
4654 1 -- enable.
4655 Default value is 1.
4656
4657 apparmor= [APPARMOR] Disable or enable AppArmor at boot time
4658 Format: { "0" | "1" }
4659 See security/apparmor/Kconfig help text
4660 0 -- disable.
4661 1 -- enable.
4662 Default value is set via kernel config option.
4663
4664 serialnumber [BUGS=X86-32]
4665
4666 shapers= [NET]
4667 Maximal number of shapers.
4668
4669 simeth= [IA-64]
4670 simscsi=
4671
4672 slram= [HW,MTD]
4673
4674 slab_nomerge [MM]
4675 Disable merging of slabs with similar size. May be
4676 necessary if there is some reason to distinguish
4677 allocs to different slabs, especially in hardened
4678 environments where the risk of heap overflows and
4679 layout control by attackers can usually be
4680 frustrated by disabling merging. This will reduce
4681 most of the exposure of a heap attack to a single
4682 cache (risks via metadata attacks are mostly
4683 unchanged). Debug options disable merging on their
4684 own.
4685 For more information see Documentation/vm/slub.rst.
4686
4687 slab_max_order= [MM, SLAB]
4688 Determines the maximum allowed order for slabs.
4689 A high setting may cause OOMs due to memory
4690 fragmentation. Defaults to 1 for systems with
4691 more than 32MB of RAM, 0 otherwise.
4692
4693 slub_debug[=options[,slabs][;[options[,slabs]]...] [MM, SLUB]
4694 Enabling slub_debug allows one to determine the
4695 culprit if slab objects become corrupted. Enabling
4696 slub_debug can create guard zones around objects and
4697 may poison objects when not in use. Also tracks the
4698 last alloc / free. For more information see
4699 Documentation/vm/slub.rst.
4700
4701 slub_memcg_sysfs= [MM, SLUB]
4702 Determines whether to enable sysfs directories for
4703 memory cgroup sub-caches. 1 to enable, 0 to disable.
4704 The default is determined by CONFIG_SLUB_MEMCG_SYSFS_ON.
4705 Enabling this can lead to a very high number of debug
4706 directories and files being created under
4707 /sys/kernel/slub.
4708
4709 slub_max_order= [MM, SLUB]
4710 Determines the maximum allowed order for slabs.
4711 A high setting may cause OOMs due to memory
4712 fragmentation. For more information see
4713 Documentation/vm/slub.rst.
4714
4715 slub_min_objects= [MM, SLUB]
4716 The minimum number of objects per slab. SLUB will
4717 increase the slab order up to slub_max_order to
4718 generate a sufficiently large slab able to contain
4719 the number of objects indicated. The higher the number
4720 of objects the smaller the overhead of tracking slabs
4721 and the less frequently locks need to be acquired.
4722 For more information see Documentation/vm/slub.rst.
4723
4724 slub_min_order= [MM, SLUB]
4725 Determines the minimum page order for slabs. Must be
4726 lower than slub_max_order.
4727 For more information see Documentation/vm/slub.rst.
4728
4729 slub_nomerge [MM, SLUB]
4730 Same with slab_nomerge. This is supported for legacy.
4731 See slab_nomerge for more information.
4732
4733 smart2= [HW]
4734 Format: <io1>[,<io2>[,...,<io8>]]
4735
4736 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices
4737 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port
4738 smsc-ircc2.ircc_sir= [HW] SIR base I/O port
4739 smsc-ircc2.ircc_fir= [HW] FIR base I/O port
4740 smsc-ircc2.ircc_irq= [HW] IRQ line
4741 smsc-ircc2.ircc_dma= [HW] DMA channel
4742 smsc-ircc2.ircc_transceiver= [HW] Transceiver type:
4743 0: Toshiba Satellite 1800 (GP data pin select)
4744 1: Fast pin select (default)
4745 2: ATC IRMode
4746
4747 smt [KNL,S390] Set the maximum number of threads (logical
4748 CPUs) to use per physical CPU on systems capable of
4749 symmetric multithreading (SMT). Will be capped to the
4750 actual hardware limit.
4751 Format: <integer>
4752 Default: -1 (no limit)
4753
4754 softlockup_panic=
4755 [KNL] Should the soft-lockup detector generate panics.
4756 Format: 0 | 1
4757
4758 A value of 1 instructs the soft-lockup detector
4759 to panic the machine when a soft-lockup occurs. It is
4760 also controlled by the kernel.softlockup_panic sysctl
4761 and CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC, which is the
4762 respective build-time switch to that functionality.
4763
4764 softlockup_all_cpu_backtrace=
4765 [KNL] Should the soft-lockup detector generate
4766 backtraces on all cpus.
4767 Format: 0 | 1
4768
4769 sonypi.*= [HW] Sony Programmable I/O Control Device driver
4770 See Documentation/admin-guide/laptops/sonypi.rst
4771
4772 spectre_v2= [X86] Control mitigation of Spectre variant 2
4773 (indirect branch speculation) vulnerability.
4774 The default operation protects the kernel from
4775 user space attacks.
4776
4777 on - unconditionally enable, implies
4778 spectre_v2_user=on
4779 off - unconditionally disable, implies
4780 spectre_v2_user=off
4781 auto - kernel detects whether your CPU model is
4782 vulnerable
4783
4784 Selecting 'on' will, and 'auto' may, choose a
4785 mitigation method at run time according to the
4786 CPU, the available microcode, the setting of the
4787 CONFIG_RETPOLINE configuration option, and the
4788 compiler with which the kernel was built.
4789
4790 Selecting 'on' will also enable the mitigation
4791 against user space to user space task attacks.
4792
4793 Selecting 'off' will disable both the kernel and
4794 the user space protections.
4795
4796 Specific mitigations can also be selected manually:
4797
4798 retpoline - replace indirect branches
4799 retpoline,generic - google's original retpoline
4800 retpoline,amd - AMD-specific minimal thunk
4801
4802 Not specifying this option is equivalent to
4803 spectre_v2=auto.
4804
4805 spectre_v2_user=
4806 [X86] Control mitigation of Spectre variant 2
4807 (indirect branch speculation) vulnerability between
4808 user space tasks
4809
4810 on - Unconditionally enable mitigations. Is
4811 enforced by spectre_v2=on
4812
4813 off - Unconditionally disable mitigations. Is
4814 enforced by spectre_v2=off
4815
4816 prctl - Indirect branch speculation is enabled,
4817 but mitigation can be enabled via prctl
4818 per thread. The mitigation control state
4819 is inherited on fork.
4820
4821 prctl,ibpb
4822 - Like "prctl" above, but only STIBP is
4823 controlled per thread. IBPB is issued
4824 always when switching between different user
4825 space processes.
4826
4827 seccomp
4828 - Same as "prctl" above, but all seccomp
4829 threads will enable the mitigation unless
4830 they explicitly opt out.
4831
4832 seccomp,ibpb
4833 - Like "seccomp" above, but only STIBP is
4834 controlled per thread. IBPB is issued
4835 always when switching between different
4836 user space processes.
4837
4838 auto - Kernel selects the mitigation depending on
4839 the available CPU features and vulnerability.
4840
4841 Default mitigation:
4842 If CONFIG_SECCOMP=y then "seccomp", otherwise "prctl"
4843
4844 Not specifying this option is equivalent to
4845 spectre_v2_user=auto.
4846
4847 spec_store_bypass_disable=
4848 [HW] Control Speculative Store Bypass (SSB) Disable mitigation
4849 (Speculative Store Bypass vulnerability)
4850
4851 Certain CPUs are vulnerable to an exploit against a
4852 a common industry wide performance optimization known
4853 as "Speculative Store Bypass" in which recent stores
4854 to the same memory location may not be observed by
4855 later loads during speculative execution. The idea
4856 is that such stores are unlikely and that they can
4857 be detected prior to instruction retirement at the
4858 end of a particular speculation execution window.
4859
4860 In vulnerable processors, the speculatively forwarded
4861 store can be used in a cache side channel attack, for
4862 example to read memory to which the attacker does not
4863 directly have access (e.g. inside sandboxed code).
4864
4865 This parameter controls whether the Speculative Store
4866 Bypass optimization is used.
4867
4868 On x86 the options are:
4869
4870 on - Unconditionally disable Speculative Store Bypass
4871 off - Unconditionally enable Speculative Store Bypass
4872 auto - Kernel detects whether the CPU model contains an
4873 implementation of Speculative Store Bypass and
4874 picks the most appropriate mitigation. If the
4875 CPU is not vulnerable, "off" is selected. If the
4876 CPU is vulnerable the default mitigation is
4877 architecture and Kconfig dependent. See below.
4878 prctl - Control Speculative Store Bypass per thread
4879 via prctl. Speculative Store Bypass is enabled
4880 for a process by default. The state of the control
4881 is inherited on fork.
4882 seccomp - Same as "prctl" above, but all seccomp threads
4883 will disable SSB unless they explicitly opt out.
4884
4885 Default mitigations:
4886 X86: If CONFIG_SECCOMP=y "seccomp", otherwise "prctl"
4887
4888 On powerpc the options are:
4889
4890 on,auto - On Power8 and Power9 insert a store-forwarding
4891 barrier on kernel entry and exit. On Power7
4892 perform a software flush on kernel entry and
4893 exit.
4894 off - No action.
4895
4896 Not specifying this option is equivalent to
4897 spec_store_bypass_disable=auto.
4898
4899 spia_io_base= [HW,MTD]
4900 spia_fio_base=
4901 spia_pedr=
4902 spia_peddr=
4903
4904 split_lock_detect=
4905 [X86] Enable split lock detection
4906
4907 When enabled (and if hardware support is present), atomic
4908 instructions that access data across cache line
4909 boundaries will result in an alignment check exception.
4910
4911 off - not enabled
4912
4913 warn - the kernel will emit rate limited warnings
4914 about applications triggering the #AC
4915 exception. This mode is the default on CPUs
4916 that supports split lock detection.
4917
4918 fatal - the kernel will send SIGBUS to applications
4919 that trigger the #AC exception.
4920
4921 If an #AC exception is hit in the kernel or in
4922 firmware (i.e. not while executing in user mode)
4923 the kernel will oops in either "warn" or "fatal"
4924 mode.
4925
4926 srbds= [X86,INTEL]
4927 Control the Special Register Buffer Data Sampling
4928 (SRBDS) mitigation.
4929
4930 Certain CPUs are vulnerable to an MDS-like
4931 exploit which can leak bits from the random
4932 number generator.
4933
4934 By default, this issue is mitigated by
4935 microcode. However, the microcode fix can cause
4936 the RDRAND and RDSEED instructions to become
4937 much slower. Among other effects, this will
4938 result in reduced throughput from /dev/urandom.
4939
4940 The microcode mitigation can be disabled with
4941 the following option:
4942
4943 off: Disable mitigation and remove
4944 performance impact to RDRAND and RDSEED
4945
4946 srcutree.counter_wrap_check [KNL]
4947 Specifies how frequently to check for
4948 grace-period sequence counter wrap for the
4949 srcu_data structure's ->srcu_gp_seq_needed field.
4950 The greater the number of bits set in this kernel
4951 parameter, the less frequently counter wrap will
4952 be checked for. Note that the bottom two bits
4953 are ignored.
4954
4955 srcutree.exp_holdoff [KNL]
4956 Specifies how many nanoseconds must elapse
4957 since the end of the last SRCU grace period for
4958 a given srcu_struct until the next normal SRCU
4959 grace period will be considered for automatic
4960 expediting. Set to zero to disable automatic
4961 expediting.
4962
4963 ssbd= [ARM64,HW]
4964 Speculative Store Bypass Disable control
4965
4966 On CPUs that are vulnerable to the Speculative
4967 Store Bypass vulnerability and offer a
4968 firmware based mitigation, this parameter
4969 indicates how the mitigation should be used:
4970
4971 force-on: Unconditionally enable mitigation for
4972 for both kernel and userspace
4973 force-off: Unconditionally disable mitigation for
4974 for both kernel and userspace
4975 kernel: Always enable mitigation in the
4976 kernel, and offer a prctl interface
4977 to allow userspace to register its
4978 interest in being mitigated too.
4979
4980 stack_guard_gap= [MM]
4981 override the default stack gap protection. The value
4982 is in page units and it defines how many pages prior
4983 to (for stacks growing down) resp. after (for stacks
4984 growing up) the main stack are reserved for no other
4985 mapping. Default value is 256 pages.
4986
4987 stacktrace [FTRACE]
4988 Enabled the stack tracer on boot up.
4989
4990 stacktrace_filter=[function-list]
4991 [FTRACE] Limit the functions that the stack tracer
4992 will trace at boot up. function-list is a comma separated
4993 list of functions. This list can be changed at run
4994 time by the stack_trace_filter file in the debugfs
4995 tracing directory. Note, this enables stack tracing
4996 and the stacktrace above is not needed.
4997
4998 sti= [PARISC,HW]
4999 Format: <num>
5000 Set the STI (builtin display/keyboard on the HP-PARISC
5001 machines) console (graphic card) which should be used
5002 as the initial boot-console.
5003 See also comment in drivers/video/console/sticore.c.
5004
5005 sti_font= [HW]
5006 See comment in drivers/video/console/sticore.c.
5007
5008 stifb= [HW]
5009 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]]
5010
5011 sunrpc.min_resvport=
5012 sunrpc.max_resvport=
5013 [NFS,SUNRPC]
5014 SunRPC servers often require that client requests
5015 originate from a privileged port (i.e. a port in the
5016 range 0 < portnr < 1024).
5017 An administrator who wishes to reserve some of these
5018 ports for other uses may adjust the range that the
5019 kernel's sunrpc client considers to be privileged
5020 using these two parameters to set the minimum and
5021 maximum port values.
5022
5023 sunrpc.svc_rpc_per_connection_limit=
5024 [NFS,SUNRPC]
5025 Limit the number of requests that the server will
5026 process in parallel from a single connection.
5027 The default value is 0 (no limit).
5028
5029 sunrpc.pool_mode=
5030 [NFS]
5031 Control how the NFS server code allocates CPUs to
5032 service thread pools. Depending on how many NICs
5033 you have and where their interrupts are bound, this
5034 option will affect which CPUs will do NFS serving.
5035 Note: this parameter cannot be changed while the
5036 NFS server is running.
5037
5038 auto the server chooses an appropriate mode
5039 automatically using heuristics
5040 global a single global pool contains all CPUs
5041 percpu one pool for each CPU
5042 pernode one pool for each NUMA node (equivalent
5043 to global on non-NUMA machines)
5044
5045 sunrpc.tcp_slot_table_entries=
5046 sunrpc.udp_slot_table_entries=
5047 [NFS,SUNRPC]
5048 Sets the upper limit on the number of simultaneous
5049 RPC calls that can be sent from the client to a
5050 server. Increasing these values may allow you to
5051 improve throughput, but will also increase the
5052 amount of memory reserved for use by the client.
5053
5054 suspend.pm_test_delay=
5055 [SUSPEND]
5056 Sets the number of seconds to remain in a suspend test
5057 mode before resuming the system (see
5058 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG
5059 is set. Default value is 5.
5060
5061 svm= [PPC]
5062 Format: { on | off | y | n | 1 | 0 }
5063 This parameter controls use of the Protected
5064 Execution Facility on pSeries.
5065
5066 swapaccount=[0|1]
5067 [KNL] Enable accounting of swap in memory resource
5068 controller if no parameter or 1 is given or disable
5069 it if 0 is given (See Documentation/admin-guide/cgroup-v1/memory.rst)
5070
5071 swiotlb= [ARM,IA-64,PPC,MIPS,X86]
5072 Format: { <int> | force | noforce }
5073 <int> -- Number of I/O TLB slabs
5074 force -- force using of bounce buffers even if they
5075 wouldn't be automatically used by the kernel
5076 noforce -- Never use bounce buffers (for debugging)
5077
5078 switches= [HW,M68k]
5079
5080 sysctl.*= [KNL]
5081 Set a sysctl parameter, right before loading the init
5082 process, as if the value was written to the respective
5083 /proc/sys/... file. Both '.' and '/' are recognized as
5084 separators. Unrecognized parameters and invalid values
5085 are reported in the kernel log. Sysctls registered
5086 later by a loaded module cannot be set this way.
5087 Example: sysctl.vm.swappiness=40
5088
5089 sysfs.deprecated=0|1 [KNL]
5090 Enable/disable old style sysfs layout for old udev
5091 on older distributions. When this option is enabled
5092 very new udev will not work anymore. When this option
5093 is disabled (or CONFIG_SYSFS_DEPRECATED not compiled)
5094 in older udev will not work anymore.
5095 Default depends on CONFIG_SYSFS_DEPRECATED_V2 set in
5096 the kernel configuration.
5097
5098 sysrq_always_enabled
5099 [KNL]
5100 Ignore sysrq setting - this boot parameter will
5101 neutralize any effect of /proc/sys/kernel/sysrq.
5102 Useful for debugging.
5103
5104 tcpmhash_entries= [KNL,NET]
5105 Set the number of tcp_metrics_hash slots.
5106 Default value is 8192 or 16384 depending on total
5107 ram pages. This is used to specify the TCP metrics
5108 cache size. See Documentation/networking/ip-sysctl.rst
5109 "tcp_no_metrics_save" section for more details.
5110
5111 tdfx= [HW,DRM]
5112
5113 test_suspend= [SUSPEND][,N]
5114 Specify "mem" (for Suspend-to-RAM) or "standby" (for
5115 standby suspend) or "freeze" (for suspend type freeze)
5116 as the system sleep state during system startup with
5117 the optional capability to repeat N number of times.
5118 The system is woken from this state using a
5119 wakeup-capable RTC alarm.
5120
5121 thash_entries= [KNL,NET]
5122 Set number of hash buckets for TCP connection
5123
5124 thermal.act= [HW,ACPI]
5125 -1: disable all active trip points in all thermal zones
5126 <degrees C>: override all lowest active trip points
5127
5128 thermal.crt= [HW,ACPI]
5129 -1: disable all critical trip points in all thermal zones
5130 <degrees C>: override all critical trip points
5131
5132 thermal.nocrt= [HW,ACPI]
5133 Set to disable actions on ACPI thermal zone
5134 critical and hot trip points.
5135
5136 thermal.off= [HW,ACPI]
5137 1: disable ACPI thermal control
5138
5139 thermal.psv= [HW,ACPI]
5140 -1: disable all passive trip points
5141 <degrees C>: override all passive trip points to this
5142 value
5143
5144 thermal.tzp= [HW,ACPI]
5145 Specify global default ACPI thermal zone polling rate
5146 <deci-seconds>: poll all this frequency
5147 0: no polling (default)
5148
5149 threadirqs [KNL]
5150 Force threading of all interrupt handlers except those
5151 marked explicitly IRQF_NO_THREAD.
5152
5153 topology= [S390]
5154 Format: {off | on}
5155 Specify if the kernel should make use of the cpu
5156 topology information if the hardware supports this.
5157 The scheduler will make use of this information and
5158 e.g. base its process migration decisions on it.
5159 Default is on.
5160
5161 topology_updates= [KNL, PPC, NUMA]
5162 Format: {off}
5163 Specify if the kernel should ignore (off)
5164 topology updates sent by the hypervisor to this
5165 LPAR.
5166
5167 torture.disable_onoff_at_boot= [KNL]
5168 Prevent the CPU-hotplug component of torturing
5169 until after init has spawned.
5170
5171 torture.ftrace_dump_at_shutdown= [KNL]
5172 Dump the ftrace buffer at torture-test shutdown,
5173 even if there were no errors. This can be a
5174 very costly operation when many torture tests
5175 are running concurrently, especially on systems
5176 with rotating-rust storage.
5177
5178 tp720= [HW,PS2]
5179
5180 tpm_suspend_pcr=[HW,TPM]
5181 Format: integer pcr id
5182 Specify that at suspend time, the tpm driver
5183 should extend the specified pcr with zeros,
5184 as a workaround for some chips which fail to
5185 flush the last written pcr on TPM_SaveState.
5186 This will guarantee that all the other pcrs
5187 are saved.
5188
5189 trace_buf_size=nn[KMG]
5190 [FTRACE] will set tracing buffer size on each cpu.
5191
5192 trace_event=[event-list]
5193 [FTRACE] Set and start specified trace events in order
5194 to facilitate early boot debugging. The event-list is a
5195 comma separated list of trace events to enable. See
5196 also Documentation/trace/events.rst
5197
5198 trace_options=[option-list]
5199 [FTRACE] Enable or disable tracer options at boot.
5200 The option-list is a comma delimited list of options
5201 that can be enabled or disabled just as if you were
5202 to echo the option name into
5203
5204 /sys/kernel/debug/tracing/trace_options
5205
5206 For example, to enable stacktrace option (to dump the
5207 stack trace of each event), add to the command line:
5208
5209 trace_options=stacktrace
5210
5211 See also Documentation/trace/ftrace.rst "trace options"
5212 section.
5213
5214 tp_printk[FTRACE]
5215 Have the tracepoints sent to printk as well as the
5216 tracing ring buffer. This is useful for early boot up
5217 where the system hangs or reboots and does not give the
5218 option for reading the tracing buffer or performing a
5219 ftrace_dump_on_oops.
5220
5221 To turn off having tracepoints sent to printk,
5222 echo 0 > /proc/sys/kernel/tracepoint_printk
5223 Note, echoing 1 into this file without the
5224 tracepoint_printk kernel cmdline option has no effect.
5225
5226 ** CAUTION **
5227
5228 Having tracepoints sent to printk() and activating high
5229 frequency tracepoints such as irq or sched, can cause
5230 the system to live lock.
5231
5232 traceoff_on_warning
5233 [FTRACE] enable this option to disable tracing when a
5234 warning is hit. This turns off "tracing_on". Tracing can
5235 be enabled again by echoing '1' into the "tracing_on"
5236 file located in /sys/kernel/debug/tracing/
5237
5238 This option is useful, as it disables the trace before
5239 the WARNING dump is called, which prevents the trace to
5240 be filled with content caused by the warning output.
5241
5242 This option can also be set at run time via the sysctl
5243 option: kernel/traceoff_on_warning
5244
5245 transparent_hugepage=
5246 [KNL]
5247 Format: [always|madvise|never]
5248 Can be used to control the default behavior of the system
5249 with respect to transparent hugepages.
5250 See Documentation/admin-guide/mm/transhuge.rst
5251 for more details.
5252
5253 tsc= Disable clocksource stability checks for TSC.
5254 Format: <string>
5255 [x86] reliable: mark tsc clocksource as reliable, this
5256 disables clocksource verification at runtime, as well
5257 as the stability checks done at bootup. Used to enable
5258 high-resolution timer mode on older hardware, and in
5259 virtualized environment.
5260 [x86] noirqtime: Do not use TSC to do irq accounting.
5261 Used to run time disable IRQ_TIME_ACCOUNTING on any
5262 platforms where RDTSC is slow and this accounting
5263 can add overhead.
5264 [x86] unstable: mark the TSC clocksource as unstable, this
5265 marks the TSC unconditionally unstable at bootup and
5266 avoids any further wobbles once the TSC watchdog notices.
5267 [x86] nowatchdog: disable clocksource watchdog. Used
5268 in situations with strict latency requirements (where
5269 interruptions from clocksource watchdog are not
5270 acceptable).
5271
5272 tsc_early_khz= [X86] Skip early TSC calibration and use the given
5273 value instead. Useful when the early TSC frequency discovery
5274 procedure is not reliable, such as on overclocked systems
5275 with CPUID.16h support and partial CPUID.15h support.
5276 Format: <unsigned int>
5277
5278 tsx= [X86] Control Transactional Synchronization
5279 Extensions (TSX) feature in Intel processors that
5280 support TSX control.
5281
5282 This parameter controls the TSX feature. The options are:
5283
5284 on - Enable TSX on the system. Although there are
5285 mitigations for all known security vulnerabilities,
5286 TSX has been known to be an accelerator for
5287 several previous speculation-related CVEs, and
5288 so there may be unknown security risks associated
5289 with leaving it enabled.
5290
5291 off - Disable TSX on the system. (Note that this
5292 option takes effect only on newer CPUs which are
5293 not vulnerable to MDS, i.e., have
5294 MSR_IA32_ARCH_CAPABILITIES.MDS_NO=1 and which get
5295 the new IA32_TSX_CTRL MSR through a microcode
5296 update. This new MSR allows for the reliable
5297 deactivation of the TSX functionality.)
5298
5299 auto - Disable TSX if X86_BUG_TAA is present,
5300 otherwise enable TSX on the system.
5301
5302 Not specifying this option is equivalent to tsx=off.
5303
5304 See Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
5305 for more details.
5306
5307 tsx_async_abort= [X86,INTEL] Control mitigation for the TSX Async
5308 Abort (TAA) vulnerability.
5309
5310 Similar to Micro-architectural Data Sampling (MDS)
5311 certain CPUs that support Transactional
5312 Synchronization Extensions (TSX) are vulnerable to an
5313 exploit against CPU internal buffers which can forward
5314 information to a disclosure gadget under certain
5315 conditions.
5316
5317 In vulnerable processors, the speculatively forwarded
5318 data can be used in a cache side channel attack, to
5319 access data to which the attacker does not have direct
5320 access.
5321
5322 This parameter controls the TAA mitigation. The
5323 options are:
5324
5325 full - Enable TAA mitigation on vulnerable CPUs
5326 if TSX is enabled.
5327
5328 full,nosmt - Enable TAA mitigation and disable SMT on
5329 vulnerable CPUs. If TSX is disabled, SMT
5330 is not disabled because CPU is not
5331 vulnerable to cross-thread TAA attacks.
5332 off - Unconditionally disable TAA mitigation
5333
5334 On MDS-affected machines, tsx_async_abort=off can be
5335 prevented by an active MDS mitigation as both vulnerabilities
5336 are mitigated with the same mechanism so in order to disable
5337 this mitigation, you need to specify mds=off too.
5338
5339 Not specifying this option is equivalent to
5340 tsx_async_abort=full. On CPUs which are MDS affected
5341 and deploy MDS mitigation, TAA mitigation is not
5342 required and doesn't provide any additional
5343 mitigation.
5344
5345 For details see:
5346 Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
5347
5348 turbografx.map[2|3]= [HW,JOY]
5349 TurboGraFX parallel port interface
5350 Format:
5351 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7>
5352 See also Documentation/input/devices/joystick-parport.rst
5353
5354 udbg-immortal [PPC] When debugging early kernel crashes that
5355 happen after console_init() and before a proper
5356 console driver takes over, this boot options might
5357 help "seeing" what's going on.
5358
5359 uhash_entries= [KNL,NET]
5360 Set number of hash buckets for UDP/UDP-Lite connections
5361
5362 uhci-hcd.ignore_oc=
5363 [USB] Ignore overcurrent events (default N).
5364 Some badly-designed motherboards generate lots of
5365 bogus events, for ports that aren't wired to
5366 anything. Set this parameter to avoid log spamming.
5367 Note that genuine overcurrent events won't be
5368 reported either.
5369
5370 unknown_nmi_panic
5371 [X86] Cause panic on unknown NMI.
5372
5373 usbcore.authorized_default=
5374 [USB] Default USB device authorization:
5375 (default -1 = authorized except for wireless USB,
5376 0 = not authorized, 1 = authorized, 2 = authorized
5377 if device connected to internal port)
5378
5379 usbcore.autosuspend=
5380 [USB] The autosuspend time delay (in seconds) used
5381 for newly-detected USB devices (default 2). This
5382 is the time required before an idle device will be
5383 autosuspended. Devices for which the delay is set
5384 to a negative value won't be autosuspended at all.
5385
5386 usbcore.usbfs_snoop=
5387 [USB] Set to log all usbfs traffic (default 0 = off).
5388
5389 usbcore.usbfs_snoop_max=
5390 [USB] Maximum number of bytes to snoop in each URB
5391 (default = 65536).
5392
5393 usbcore.blinkenlights=
5394 [USB] Set to cycle leds on hubs (default 0 = off).
5395
5396 usbcore.old_scheme_first=
5397 [USB] Start with the old device initialization
5398 scheme (default 0 = off).
5399
5400 usbcore.usbfs_memory_mb=
5401 [USB] Memory limit (in MB) for buffers allocated by
5402 usbfs (default = 16, 0 = max = 2047).
5403
5404 usbcore.use_both_schemes=
5405 [USB] Try the other device initialization scheme
5406 if the first one fails (default 1 = enabled).
5407
5408 usbcore.initial_descriptor_timeout=
5409 [USB] Specifies timeout for the initial 64-byte
5410 USB_REQ_GET_DESCRIPTOR request in milliseconds
5411 (default 5000 = 5.0 seconds).
5412
5413 usbcore.nousb [USB] Disable the USB subsystem
5414
5415 usbcore.quirks=
5416 [USB] A list of quirk entries to augment the built-in
5417 usb core quirk list. List entries are separated by
5418 commas. Each entry has the form
5419 VendorID:ProductID:Flags. The IDs are 4-digit hex
5420 numbers and Flags is a set of letters. Each letter
5421 will change the built-in quirk; setting it if it is
5422 clear and clearing it if it is set. The letters have
5423 the following meanings:
5424 a = USB_QUIRK_STRING_FETCH_255 (string
5425 descriptors must not be fetched using
5426 a 255-byte read);
5427 b = USB_QUIRK_RESET_RESUME (device can't resume
5428 correctly so reset it instead);
5429 c = USB_QUIRK_NO_SET_INTF (device can't handle
5430 Set-Interface requests);
5431 d = USB_QUIRK_CONFIG_INTF_STRINGS (device can't
5432 handle its Configuration or Interface
5433 strings);
5434 e = USB_QUIRK_RESET (device can't be reset
5435 (e.g morph devices), don't use reset);
5436 f = USB_QUIRK_HONOR_BNUMINTERFACES (device has
5437 more interface descriptions than the
5438 bNumInterfaces count, and can't handle
5439 talking to these interfaces);
5440 g = USB_QUIRK_DELAY_INIT (device needs a pause
5441 during initialization, after we read
5442 the device descriptor);
5443 h = USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL (For
5444 high speed and super speed interrupt
5445 endpoints, the USB 2.0 and USB 3.0 spec
5446 require the interval in microframes (1
5447 microframe = 125 microseconds) to be
5448 calculated as interval = 2 ^
5449 (bInterval-1).
5450 Devices with this quirk report their
5451 bInterval as the result of this
5452 calculation instead of the exponent
5453 variable used in the calculation);
5454 i = USB_QUIRK_DEVICE_QUALIFIER (device can't
5455 handle device_qualifier descriptor
5456 requests);
5457 j = USB_QUIRK_IGNORE_REMOTE_WAKEUP (device
5458 generates spurious wakeup, ignore
5459 remote wakeup capability);
5460 k = USB_QUIRK_NO_LPM (device can't handle Link
5461 Power Management);
5462 l = USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL
5463 (Device reports its bInterval as linear
5464 frames instead of the USB 2.0
5465 calculation);
5466 m = USB_QUIRK_DISCONNECT_SUSPEND (Device needs
5467 to be disconnected before suspend to
5468 prevent spurious wakeup);
5469 n = USB_QUIRK_DELAY_CTRL_MSG (Device needs a
5470 pause after every control message);
5471 o = USB_QUIRK_HUB_SLOW_RESET (Hub needs extra
5472 delay after resetting its port);
5473 Example: quirks=0781:5580:bk,0a5c:5834:gij
5474
5475 usbhid.mousepoll=
5476 [USBHID] The interval which mice are to be polled at.
5477
5478 usbhid.jspoll=
5479 [USBHID] The interval which joysticks are to be polled at.
5480
5481 usbhid.kbpoll=
5482 [USBHID] The interval which keyboards are to be polled at.
5483
5484 usb-storage.delay_use=
5485 [UMS] The delay in seconds before a new device is
5486 scanned for Logical Units (default 1).
5487
5488 usb-storage.quirks=
5489 [UMS] A list of quirks entries to supplement or
5490 override the built-in unusual_devs list. List
5491 entries are separated by commas. Each entry has
5492 the form VID:PID:Flags where VID and PID are Vendor
5493 and Product ID values (4-digit hex numbers) and
5494 Flags is a set of characters, each corresponding
5495 to a common usb-storage quirk flag as follows:
5496 a = SANE_SENSE (collect more than 18 bytes
5497 of sense data, not on uas);
5498 b = BAD_SENSE (don't collect more than 18
5499 bytes of sense data, not on uas);
5500 c = FIX_CAPACITY (decrease the reported
5501 device capacity by one sector);
5502 d = NO_READ_DISC_INFO (don't use
5503 READ_DISC_INFO command, not on uas);
5504 e = NO_READ_CAPACITY_16 (don't use
5505 READ_CAPACITY_16 command);
5506 f = NO_REPORT_OPCODES (don't use report opcodes
5507 command, uas only);
5508 g = MAX_SECTORS_240 (don't transfer more than
5509 240 sectors at a time, uas only);
5510 h = CAPACITY_HEURISTICS (decrease the
5511 reported device capacity by one
5512 sector if the number is odd);
5513 i = IGNORE_DEVICE (don't bind to this
5514 device);
5515 j = NO_REPORT_LUNS (don't use report luns
5516 command, uas only);
5517 l = NOT_LOCKABLE (don't try to lock and
5518 unlock ejectable media, not on uas);
5519 m = MAX_SECTORS_64 (don't transfer more
5520 than 64 sectors = 32 KB at a time,
5521 not on uas);
5522 n = INITIAL_READ10 (force a retry of the
5523 initial READ(10) command, not on uas);
5524 o = CAPACITY_OK (accept the capacity
5525 reported by the device, not on uas);
5526 p = WRITE_CACHE (the device cache is ON
5527 by default, not on uas);
5528 r = IGNORE_RESIDUE (the device reports
5529 bogus residue values, not on uas);
5530 s = SINGLE_LUN (the device has only one
5531 Logical Unit);
5532 t = NO_ATA_1X (don't allow ATA(12) and ATA(16)
5533 commands, uas only);
5534 u = IGNORE_UAS (don't bind to the uas driver);
5535 w = NO_WP_DETECT (don't test whether the
5536 medium is write-protected).
5537 y = ALWAYS_SYNC (issue a SYNCHRONIZE_CACHE
5538 even if the device claims no cache,
5539 not on uas)
5540 Example: quirks=0419:aaf5:rl,0421:0433:rc
5541
5542 user_debug= [KNL,ARM]
5543 Format: <int>
5544 See arch/arm/Kconfig.debug help text.
5545 1 - undefined instruction events
5546 2 - system calls
5547 4 - invalid data aborts
5548 8 - SIGSEGV faults
5549 16 - SIGBUS faults
5550 Example: user_debug=31
5551
5552 userpte=
5553 [X86] Flags controlling user PTE allocations.
5554
5555 nohigh = do not allocate PTE pages in
5556 HIGHMEM regardless of setting
5557 of CONFIG_HIGHPTE.
5558
5559 vdso= [X86,SH]
5560 On X86_32, this is an alias for vdso32=. Otherwise:
5561
5562 vdso=1: enable VDSO (the default)
5563 vdso=0: disable VDSO mapping
5564
5565 vdso32= [X86] Control the 32-bit vDSO
5566 vdso32=1: enable 32-bit VDSO
5567 vdso32=0 or vdso32=2: disable 32-bit VDSO
5568
5569 See the help text for CONFIG_COMPAT_VDSO for more
5570 details. If CONFIG_COMPAT_VDSO is set, the default is
5571 vdso32=0; otherwise, the default is vdso32=1.
5572
5573 For compatibility with older kernels, vdso32=2 is an
5574 alias for vdso32=0.
5575
5576 Try vdso32=0 if you encounter an error that says:
5577 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed!
5578
5579 vector= [IA-64,SMP]
5580 vector=percpu: enable percpu vector domain
5581
5582 video= [FB] Frame buffer configuration
5583 See Documentation/fb/modedb.rst.
5584
5585 video.brightness_switch_enabled= [0,1]
5586 If set to 1, on receiving an ACPI notify event
5587 generated by hotkey, video driver will adjust brightness
5588 level and then send out the event to user space through
5589 the allocated input device; If set to 0, video driver
5590 will only send out the event without touching backlight
5591 brightness level.
5592 default: 1
5593
5594 virtio_mmio.device=
5595 [VMMIO] Memory mapped virtio (platform) device.
5596
5597 <size>@<baseaddr>:<irq>[:<id>]
5598 where:
5599 <size> := size (can use standard suffixes
5600 like K, M and G)
5601 <baseaddr> := physical base address
5602 <irq> := interrupt number (as passed to
5603 request_irq())
5604 <id> := (optional) platform device id
5605 example:
5606 virtio_mmio.device=1K@0x100b0000:48:7
5607
5608 Can be used multiple times for multiple devices.
5609
5610 vga= [BOOT,X86-32] Select a particular video mode
5611 See Documentation/x86/boot.rst and
5612 Documentation/admin-guide/svga.rst.
5613 Use vga=ask for menu.
5614 This is actually a boot loader parameter; the value is
5615 passed to the kernel using a special protocol.
5616
5617 vm_debug[=options] [KNL] Available with CONFIG_DEBUG_VM=y.
5618 May slow down system boot speed, especially when
5619 enabled on systems with a large amount of memory.
5620 All options are enabled by default, and this
5621 interface is meant to allow for selectively
5622 enabling or disabling specific virtual memory
5623 debugging features.
5624
5625 Available options are:
5626 P Enable page structure init time poisoning
5627 - Disable all of the above options
5628
5629 vmalloc=nn[KMG] [KNL,BOOT] Forces the vmalloc area to have an exact
5630 size of <nn>. This can be used to increase the
5631 minimum size (128MB on x86). It can also be used to
5632 decrease the size and leave more room for directly
5633 mapped kernel RAM.
5634
5635 vmcp_cma=nn[MG] [KNL,S390]
5636 Sets the memory size reserved for contiguous memory
5637 allocations for the vmcp device driver.
5638
5639 vmhalt= [KNL,S390] Perform z/VM CP command after system halt.
5640 Format: <command>
5641
5642 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic.
5643 Format: <command>
5644
5645 vmpoff= [KNL,S390] Perform z/VM CP command after power off.
5646 Format: <command>
5647
5648 vsyscall= [X86-64]
5649 Controls the behavior of vsyscalls (i.e. calls to
5650 fixed addresses of 0xffffffffff600x00 from legacy
5651 code). Most statically-linked binaries and older
5652 versions of glibc use these calls. Because these
5653 functions are at fixed addresses, they make nice
5654 targets for exploits that can control RIP.
5655
5656 emulate [default] Vsyscalls turn into traps and are
5657 emulated reasonably safely. The vsyscall
5658 page is readable.
5659
5660 xonly Vsyscalls turn into traps and are
5661 emulated reasonably safely. The vsyscall
5662 page is not readable.
5663
5664 none Vsyscalls don't work at all. This makes
5665 them quite hard to use for exploits but
5666 might break your system.
5667
5668 vt.color= [VT] Default text color.
5669 Format: 0xYX, X = foreground, Y = background.
5670 Default: 0x07 = light gray on black.
5671
5672 vt.cur_default= [VT] Default cursor shape.
5673 Format: 0xCCBBAA, where AA, BB, and CC are the same as
5674 the parameters of the <Esc>[?A;B;Cc escape sequence;
5675 see VGA-softcursor.txt. Default: 2 = underline.
5676
5677 vt.default_blu= [VT]
5678 Format: <blue0>,<blue1>,<blue2>,...,<blue15>
5679 Change the default blue palette of the console.
5680 This is a 16-member array composed of values
5681 ranging from 0-255.
5682
5683 vt.default_grn= [VT]
5684 Format: <green0>,<green1>,<green2>,...,<green15>
5685 Change the default green palette of the console.
5686 This is a 16-member array composed of values
5687 ranging from 0-255.
5688
5689 vt.default_red= [VT]
5690 Format: <red0>,<red1>,<red2>,...,<red15>
5691 Change the default red palette of the console.
5692 This is a 16-member array composed of values
5693 ranging from 0-255.
5694
5695 vt.default_utf8=
5696 [VT]
5697 Format=<0|1>
5698 Set system-wide default UTF-8 mode for all tty's.
5699 Default is 1, i.e. UTF-8 mode is enabled for all
5700 newly opened terminals.
5701
5702 vt.global_cursor_default=
5703 [VT]
5704 Format=<-1|0|1>
5705 Set system-wide default for whether a cursor
5706 is shown on new VTs. Default is -1,
5707 i.e. cursors will be created by default unless
5708 overridden by individual drivers. 0 will hide
5709 cursors, 1 will display them.
5710
5711 vt.italic= [VT] Default color for italic text; 0-15.
5712 Default: 2 = green.
5713
5714 vt.underline= [VT] Default color for underlined text; 0-15.
5715 Default: 3 = cyan.
5716
5717 watchdog timers [HW,WDT] For information on watchdog timers,
5718 see Documentation/watchdog/watchdog-parameters.rst
5719 or other driver-specific files in the
5720 Documentation/watchdog/ directory.
5721
5722 watchdog_thresh=
5723 [KNL]
5724 Set the hard lockup detector stall duration
5725 threshold in seconds. The soft lockup detector
5726 threshold is set to twice the value. A value of 0
5727 disables both lockup detectors. Default is 10
5728 seconds.
5729
5730 workqueue.watchdog_thresh=
5731 If CONFIG_WQ_WATCHDOG is configured, workqueue can
5732 warn stall conditions and dump internal state to
5733 help debugging. 0 disables workqueue stall
5734 detection; otherwise, it's the stall threshold
5735 duration in seconds. The default value is 30 and
5736 it can be updated at runtime by writing to the
5737 corresponding sysfs file.
5738
5739 workqueue.disable_numa
5740 By default, all work items queued to unbound
5741 workqueues are affine to the NUMA nodes they're
5742 issued on, which results in better behavior in
5743 general. If NUMA affinity needs to be disabled for
5744 whatever reason, this option can be used. Note
5745 that this also can be controlled per-workqueue for
5746 workqueues visible under /sys/bus/workqueue/.
5747
5748 workqueue.power_efficient
5749 Per-cpu workqueues are generally preferred because
5750 they show better performance thanks to cache
5751 locality; unfortunately, per-cpu workqueues tend to
5752 be more power hungry than unbound workqueues.
5753
5754 Enabling this makes the per-cpu workqueues which
5755 were observed to contribute significantly to power
5756 consumption unbound, leading to measurably lower
5757 power usage at the cost of small performance
5758 overhead.
5759
5760 The default value of this parameter is determined by
5761 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT.
5762
5763 workqueue.debug_force_rr_cpu
5764 Workqueue used to implicitly guarantee that work
5765 items queued without explicit CPU specified are put
5766 on the local CPU. This guarantee is no longer true
5767 and while local CPU is still preferred work items
5768 may be put on foreign CPUs. This debug option
5769 forces round-robin CPU selection to flush out
5770 usages which depend on the now broken guarantee.
5771 When enabled, memory and cache locality will be
5772 impacted.
5773
5774 x2apic_phys [X86-64,APIC] Use x2apic physical mode instead of
5775 default x2apic cluster mode on platforms
5776 supporting x2apic.
5777
5778 x86_intel_mid_timer= [X86-32,APBT]
5779 Choose timer option for x86 Intel MID platform.
5780 Two valid options are apbt timer only and lapic timer
5781 plus one apbt timer for broadcast timer.
5782 x86_intel_mid_timer=apbt_only | lapic_and_apbt
5783
5784 xen_512gb_limit [KNL,X86-64,XEN]
5785 Restricts the kernel running paravirtualized under Xen
5786 to use only up to 512 GB of RAM. The reason to do so is
5787 crash analysis tools and Xen tools for doing domain
5788 save/restore/migration must be enabled to handle larger
5789 domains.
5790
5791 xen_emul_unplug= [HW,X86,XEN]
5792 Unplug Xen emulated devices
5793 Format: [unplug0,][unplug1]
5794 ide-disks -- unplug primary master IDE devices
5795 aux-ide-disks -- unplug non-primary-master IDE devices
5796 nics -- unplug network devices
5797 all -- unplug all emulated devices (NICs and IDE disks)
5798 unnecessary -- unplugging emulated devices is
5799 unnecessary even if the host did not respond to
5800 the unplug protocol
5801 never -- do not unplug even if version check succeeds
5802
5803 xen_legacy_crash [X86,XEN]
5804 Crash from Xen panic notifier, without executing late
5805 panic() code such as dumping handler.
5806
5807 xen_nopvspin [X86,XEN]
5808 Disables the qspinlock slowpath using Xen PV optimizations.
5809 This parameter is obsoleted by "nopvspin" parameter, which
5810 has equivalent effect for XEN platform.
5811
5812 xen_nopv [X86]
5813 Disables the PV optimizations forcing the HVM guest to
5814 run as generic HVM guest with no PV drivers.
5815 This option is obsoleted by the "nopv" option, which
5816 has equivalent effect for XEN platform.
5817
5818 xen_scrub_pages= [XEN]
5819 Boolean option to control scrubbing pages before giving them back
5820 to Xen, for use by other domains. Can be also changed at runtime
5821 with /sys/devices/system/xen_memory/xen_memory0/scrub_pages.
5822 Default value controlled with CONFIG_XEN_SCRUB_PAGES_DEFAULT.
5823
5824 xen_timer_slop= [X86-64,XEN]
5825 Set the timer slop (in nanoseconds) for the virtual Xen
5826 timers (default is 100000). This adjusts the minimum
5827 delta of virtualized Xen timers, where lower values
5828 improve timer resolution at the expense of processing
5829 more timer interrupts.
5830
5831 nopv= [X86,XEN,KVM,HYPER_V,VMWARE]
5832 Disables the PV optimizations forcing the guest to run
5833 as generic guest with no PV drivers. Currently support
5834 XEN HVM, KVM, HYPER_V and VMWARE guest.
5835
5836 nopvspin [X86,XEN,KVM]
5837 Disables the qspinlock slow path using PV optimizations
5838 which allow the hypervisor to 'idle' the guest on lock
5839 contention.
5840
5841 xirc2ps_cs= [NET,PCMCIA]
5842 Format:
5843 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]]
5844
5845 xive= [PPC]
5846 By default on POWER9 and above, the kernel will
5847 natively use the XIVE interrupt controller. This option
5848 allows the fallback firmware mode to be used:
5849
5850 off Fallback to firmware control of XIVE interrupt
5851 controller on both pseries and powernv
5852 platforms. Only useful on POWER9 and above.
5853
5854 xhci-hcd.quirks [USB,KNL]
5855 A hex value specifying bitmask with supplemental xhci
5856 host controller quirks. Meaning of each bit can be
5857 consulted in header drivers/usb/host/xhci.h.
5858
5859 xmon [PPC]
5860 Format: { early | on | rw | ro | off }
5861 Controls if xmon debugger is enabled. Default is off.
5862 Passing only "xmon" is equivalent to "xmon=early".
5863 early Call xmon as early as possible on boot; xmon
5864 debugger is called from setup_arch().
5865 on xmon debugger hooks will be installed so xmon
5866 is only called on a kernel crash. Default mode,
5867 i.e. either "ro" or "rw" mode, is controlled
5868 with CONFIG_XMON_DEFAULT_RO_MODE.
5869 rw xmon debugger hooks will be installed so xmon
5870 is called only on a kernel crash, mode is write,
5871 meaning SPR registers, memory and, other data
5872 can be written using xmon commands.
5873 ro same as "rw" option above but SPR registers,
5874 memory, and other data can't be written using
5875 xmon commands.
5876 off xmon is disabled.
1 accept_memory= [MM]
2 Format: { eager | lazy }
3 default: lazy
4 By default, unaccepted memory is accepted lazily to
5 avoid prolonged boot times. The lazy option will add
6 some runtime overhead until all memory is eventually
7 accepted. In most cases the overhead is negligible.
8 For some workloads or for debugging purposes
9 accept_memory=eager can be used to accept all memory
10 at once during boot.
11
12 acpi= [HW,ACPI,X86,ARM64,RISCV64,EARLY]
13 Advanced Configuration and Power Interface
14 Format: { force | on | off | strict | noirq | rsdt |
15 copy_dsdt }
16 force -- enable ACPI if default was off
17 on -- enable ACPI but allow fallback to DT [arm64,riscv64]
18 off -- disable ACPI if default was on
19 noirq -- do not use ACPI for IRQ routing
20 strict -- Be less tolerant of platforms that are not
21 strictly ACPI specification compliant.
22 rsdt -- prefer RSDT over (default) XSDT
23 copy_dsdt -- copy DSDT to memory
24 For ARM64 and RISCV64, ONLY "acpi=off", "acpi=on" or
25 "acpi=force" are available
26
27 See also Documentation/power/runtime_pm.rst, pci=noacpi
28
29 acpi_apic_instance= [ACPI,IOAPIC,EARLY]
30 Format: <int>
31 2: use 2nd APIC table, if available
32 1,0: use 1st APIC table
33 default: 0
34
35 acpi_backlight= [HW,ACPI]
36 { vendor | video | native | none }
37 If set to vendor, prefer vendor-specific driver
38 (e.g. thinkpad_acpi, sony_acpi, etc.) instead
39 of the ACPI video.ko driver.
40 If set to video, use the ACPI video.ko driver.
41 If set to native, use the device's native backlight mode.
42 If set to none, disable the ACPI backlight interface.
43
44 acpi_force_32bit_fadt_addr [ACPI,EARLY]
45 force FADT to use 32 bit addresses rather than the
46 64 bit X_* addresses. Some firmware have broken 64
47 bit addresses for force ACPI ignore these and use
48 the older legacy 32 bit addresses.
49
50 acpica_no_return_repair [HW, ACPI]
51 Disable AML predefined validation mechanism
52 This mechanism can repair the evaluation result to make
53 the return objects more ACPI specification compliant.
54 This option is useful for developers to identify the
55 root cause of an AML interpreter issue when the issue
56 has something to do with the repair mechanism.
57
58 acpi.debug_layer= [HW,ACPI,ACPI_DEBUG]
59 acpi.debug_level= [HW,ACPI,ACPI_DEBUG]
60 Format: <int>
61 CONFIG_ACPI_DEBUG must be enabled to produce any ACPI
62 debug output. Bits in debug_layer correspond to a
63 _COMPONENT in an ACPI source file, e.g.,
64 #define _COMPONENT ACPI_EVENTS
65 Bits in debug_level correspond to a level in
66 ACPI_DEBUG_PRINT statements, e.g.,
67 ACPI_DEBUG_PRINT((ACPI_DB_INFO, ...
68 The debug_level mask defaults to "info". See
69 Documentation/firmware-guide/acpi/debug.rst for more information about
70 debug layers and levels.
71
72 Enable processor driver info messages:
73 acpi.debug_layer=0x20000000
74 Enable AML "Debug" output, i.e., stores to the Debug
75 object while interpreting AML:
76 acpi.debug_layer=0xffffffff acpi.debug_level=0x2
77 Enable all messages related to ACPI hardware:
78 acpi.debug_layer=0x2 acpi.debug_level=0xffffffff
79
80 Some values produce so much output that the system is
81 unusable. The "log_buf_len" parameter may be useful
82 if you need to capture more output.
83
84 acpi_enforce_resources= [ACPI]
85 { strict | lax | no }
86 Check for resource conflicts between native drivers
87 and ACPI OperationRegions (SystemIO and SystemMemory
88 only). IO ports and memory declared in ACPI might be
89 used by the ACPI subsystem in arbitrary AML code and
90 can interfere with legacy drivers.
91 strict (default): access to resources claimed by ACPI
92 is denied; legacy drivers trying to access reserved
93 resources will fail to bind to device using them.
94 lax: access to resources claimed by ACPI is allowed;
95 legacy drivers trying to access reserved resources
96 will bind successfully but a warning message is logged.
97 no: ACPI OperationRegions are not marked as reserved,
98 no further checks are performed.
99
100 acpi_force_table_verification [HW,ACPI,EARLY]
101 Enable table checksum verification during early stage.
102 By default, this is disabled due to x86 early mapping
103 size limitation.
104
105 acpi_irq_balance [HW,ACPI]
106 ACPI will balance active IRQs
107 default in APIC mode
108
109 acpi_irq_nobalance [HW,ACPI]
110 ACPI will not move active IRQs (default)
111 default in PIC mode
112
113 acpi_irq_isa= [HW,ACPI] If irq_balance, mark listed IRQs used by ISA
114 Format: <irq>,<irq>...
115
116 acpi_irq_pci= [HW,ACPI] If irq_balance, clear listed IRQs for
117 use by PCI
118 Format: <irq>,<irq>...
119
120 acpi_mask_gpe= [HW,ACPI]
121 Due to the existence of _Lxx/_Exx, some GPEs triggered
122 by unsupported hardware/firmware features can result in
123 GPE floodings that cannot be automatically disabled by
124 the GPE dispatcher.
125 This facility can be used to prevent such uncontrolled
126 GPE floodings.
127 Format: <byte> or <bitmap-list>
128
129 acpi_no_auto_serialize [HW,ACPI]
130 Disable auto-serialization of AML methods
131 AML control methods that contain the opcodes to create
132 named objects will be marked as "Serialized" by the
133 auto-serialization feature.
134 This feature is enabled by default.
135 This option allows to turn off the feature.
136
137 acpi_no_memhotplug [ACPI] Disable memory hotplug. Useful for kdump
138 kernels.
139
140 acpi_no_static_ssdt [HW,ACPI,EARLY]
141 Disable installation of static SSDTs at early boot time
142 By default, SSDTs contained in the RSDT/XSDT will be
143 installed automatically and they will appear under
144 /sys/firmware/acpi/tables.
145 This option turns off this feature.
146 Note that specifying this option does not affect
147 dynamic table installation which will install SSDT
148 tables to /sys/firmware/acpi/tables/dynamic.
149
150 acpi_no_watchdog [HW,ACPI,WDT]
151 Ignore the ACPI-based watchdog interface (WDAT) and let
152 a native driver control the watchdog device instead.
153
154 acpi_rsdp= [ACPI,EFI,KEXEC,EARLY]
155 Pass the RSDP address to the kernel, mostly used
156 on machines running EFI runtime service to boot the
157 second kernel for kdump.
158
159 acpi_os_name= [HW,ACPI] Tell ACPI BIOS the name of the OS
160 Format: To spoof as Windows 98: ="Microsoft Windows"
161
162 acpi_rev_override [ACPI] Override the _REV object to return 5 (instead
163 of 2 which is mandated by ACPI 6) as the supported ACPI
164 specification revision (when using this switch, it may
165 be necessary to carry out a cold reboot _twice_ in a
166 row to make it take effect on the platform firmware).
167
168 acpi_osi= [HW,ACPI] Modify list of supported OS interface strings
169 acpi_osi="string1" # add string1
170 acpi_osi="!string2" # remove string2
171 acpi_osi=!* # remove all strings
172 acpi_osi=! # disable all built-in OS vendor
173 strings
174 acpi_osi=!! # enable all built-in OS vendor
175 strings
176 acpi_osi= # disable all strings
177
178 'acpi_osi=!' can be used in combination with single or
179 multiple 'acpi_osi="string1"' to support specific OS
180 vendor string(s). Note that such command can only
181 affect the default state of the OS vendor strings, thus
182 it cannot affect the default state of the feature group
183 strings and the current state of the OS vendor strings,
184 specifying it multiple times through kernel command line
185 is meaningless. This command is useful when one do not
186 care about the state of the feature group strings which
187 should be controlled by the OSPM.
188 Examples:
189 1. 'acpi_osi=! acpi_osi="Windows 2000"' is equivalent
190 to 'acpi_osi="Windows 2000" acpi_osi=!', they all
191 can make '_OSI("Windows 2000")' TRUE.
192
193 'acpi_osi=' cannot be used in combination with other
194 'acpi_osi=' command lines, the _OSI method will not
195 exist in the ACPI namespace. NOTE that such command can
196 only affect the _OSI support state, thus specifying it
197 multiple times through kernel command line is also
198 meaningless.
199 Examples:
200 1. 'acpi_osi=' can make 'CondRefOf(_OSI, Local1)'
201 FALSE.
202
203 'acpi_osi=!*' can be used in combination with single or
204 multiple 'acpi_osi="string1"' to support specific
205 string(s). Note that such command can affect the
206 current state of both the OS vendor strings and the
207 feature group strings, thus specifying it multiple times
208 through kernel command line is meaningful. But it may
209 still not able to affect the final state of a string if
210 there are quirks related to this string. This command
211 is useful when one want to control the state of the
212 feature group strings to debug BIOS issues related to
213 the OSPM features.
214 Examples:
215 1. 'acpi_osi="Module Device" acpi_osi=!*' can make
216 '_OSI("Module Device")' FALSE.
217 2. 'acpi_osi=!* acpi_osi="Module Device"' can make
218 '_OSI("Module Device")' TRUE.
219 3. 'acpi_osi=! acpi_osi=!* acpi_osi="Windows 2000"' is
220 equivalent to
221 'acpi_osi=!* acpi_osi=! acpi_osi="Windows 2000"'
222 and
223 'acpi_osi=!* acpi_osi="Windows 2000" acpi_osi=!',
224 they all will make '_OSI("Windows 2000")' TRUE.
225
226 acpi_pm_good [X86]
227 Override the pmtimer bug detection: force the kernel
228 to assume that this machine's pmtimer latches its value
229 and always returns good values.
230
231 acpi_sci= [HW,ACPI,EARLY] ACPI System Control Interrupt trigger mode
232 Format: { level | edge | high | low }
233
234 acpi_skip_timer_override [HW,ACPI,EARLY]
235 Recognize and ignore IRQ0/pin2 Interrupt Override.
236 For broken nForce2 BIOS resulting in XT-PIC timer.
237
238 acpi_sleep= [HW,ACPI] Sleep options
239 Format: { s3_bios, s3_mode, s3_beep, s4_hwsig,
240 s4_nohwsig, old_ordering, nonvs,
241 sci_force_enable, nobl }
242 See Documentation/power/video.rst for information on
243 s3_bios and s3_mode.
244 s3_beep is for debugging; it makes the PC's speaker beep
245 as soon as the kernel's real-mode entry point is called.
246 s4_hwsig causes the kernel to check the ACPI hardware
247 signature during resume from hibernation, and gracefully
248 refuse to resume if it has changed. This complies with
249 the ACPI specification but not with reality, since
250 Windows does not do this and many laptops do change it
251 on docking. So the default behaviour is to allow resume
252 and simply warn when the signature changes, unless the
253 s4_hwsig option is enabled.
254 s4_nohwsig prevents ACPI hardware signature from being
255 used (or even warned about) during resume.
256 old_ordering causes the ACPI 1.0 ordering of the _PTS
257 control method, with respect to putting devices into
258 low power states, to be enforced (the ACPI 2.0 ordering
259 of _PTS is used by default).
260 nonvs prevents the kernel from saving/restoring the
261 ACPI NVS memory during suspend/hibernation and resume.
262 sci_force_enable causes the kernel to set SCI_EN directly
263 on resume from S1/S3 (which is against the ACPI spec,
264 but some broken systems don't work without it).
265 nobl causes the internal blacklist of systems known to
266 behave incorrectly in some ways with respect to system
267 suspend and resume to be ignored (use wisely).
268
269 acpi_use_timer_override [HW,ACPI,EARLY]
270 Use timer override. For some broken Nvidia NF5 boards
271 that require a timer override, but don't have HPET
272
273 add_efi_memmap [EFI,X86,EARLY] Include EFI memory map in
274 kernel's map of available physical RAM.
275
276 agp= [AGP]
277 { off | try_unsupported }
278 off: disable AGP support
279 try_unsupported: try to drive unsupported chipsets
280 (may crash computer or cause data corruption)
281
282 ALSA [HW,ALSA]
283 See Documentation/sound/alsa-configuration.rst
284
285 alignment= [KNL,ARM]
286 Allow the default userspace alignment fault handler
287 behaviour to be specified. Bit 0 enables warnings,
288 bit 1 enables fixups, and bit 2 sends a segfault.
289
290 align_va_addr= [X86-64]
291 Align virtual addresses by clearing slice [14:12] when
292 allocating a VMA at process creation time. This option
293 gives you up to 3% performance improvement on AMD F15h
294 machines (where it is enabled by default) for a
295 CPU-intensive style benchmark, and it can vary highly in
296 a microbenchmark depending on workload and compiler.
297
298 32: only for 32-bit processes
299 64: only for 64-bit processes
300 on: enable for both 32- and 64-bit processes
301 off: disable for both 32- and 64-bit processes
302
303 alloc_snapshot [FTRACE]
304 Allocate the ftrace snapshot buffer on boot up when the
305 main buffer is allocated. This is handy if debugging
306 and you need to use tracing_snapshot() on boot up, and
307 do not want to use tracing_snapshot_alloc() as it needs
308 to be done where GFP_KERNEL allocations are allowed.
309
310 allow_mismatched_32bit_el0 [ARM64,EARLY]
311 Allow execve() of 32-bit applications and setting of the
312 PER_LINUX32 personality on systems where only a strict
313 subset of the CPUs support 32-bit EL0. When this
314 parameter is present, the set of CPUs supporting 32-bit
315 EL0 is indicated by /sys/devices/system/cpu/aarch32_el0
316 and hot-unplug operations may be restricted.
317
318 See Documentation/arch/arm64/asymmetric-32bit.rst for more
319 information.
320
321 amd_iommu= [HW,X86-64]
322 Pass parameters to the AMD IOMMU driver in the system.
323 Possible values are:
324 fullflush - Deprecated, equivalent to iommu.strict=1
325 off - do not initialize any AMD IOMMU found in
326 the system
327 force_isolation - Force device isolation for all
328 devices. The IOMMU driver is not
329 allowed anymore to lift isolation
330 requirements as needed. This option
331 does not override iommu=pt
332 force_enable - Force enable the IOMMU on platforms known
333 to be buggy with IOMMU enabled. Use this
334 option with care.
335 pgtbl_v1 - Use v1 page table for DMA-API (Default).
336 pgtbl_v2 - Use v2 page table for DMA-API.
337 irtcachedis - Disable Interrupt Remapping Table (IRT) caching.
338
339 amd_iommu_dump= [HW,X86-64]
340 Enable AMD IOMMU driver option to dump the ACPI table
341 for AMD IOMMU. With this option enabled, AMD IOMMU
342 driver will print ACPI tables for AMD IOMMU during
343 IOMMU initialization.
344
345 amd_iommu_intr= [HW,X86-64]
346 Specifies one of the following AMD IOMMU interrupt
347 remapping modes:
348 legacy - Use legacy interrupt remapping mode.
349 vapic - Use virtual APIC mode, which allows IOMMU
350 to inject interrupts directly into guest.
351 This mode requires kvm-amd.avic=1.
352 (Default when IOMMU HW support is present.)
353
354 amd_pstate= [X86,EARLY]
355 disable
356 Do not enable amd_pstate as the default
357 scaling driver for the supported processors
358 passive
359 Use amd_pstate with passive mode as a scaling driver.
360 In this mode autonomous selection is disabled.
361 Driver requests a desired performance level and platform
362 tries to match the same performance level if it is
363 satisfied by guaranteed performance level.
364 active
365 Use amd_pstate_epp driver instance as the scaling driver,
366 driver provides a hint to the hardware if software wants
367 to bias toward performance (0x0) or energy efficiency (0xff)
368 to the CPPC firmware. then CPPC power algorithm will
369 calculate the runtime workload and adjust the realtime cores
370 frequency.
371 guided
372 Activate guided autonomous mode. Driver requests minimum and
373 maximum performance level and the platform autonomously
374 selects a performance level in this range and appropriate
375 to the current workload.
376
377 amd_prefcore=
378 [X86]
379 disable
380 Disable amd-pstate preferred core.
381
382 amijoy.map= [HW,JOY] Amiga joystick support
383 Map of devices attached to JOY0DAT and JOY1DAT
384 Format: <a>,<b>
385 See also Documentation/input/joydev/joystick.rst
386
387 analog.map= [HW,JOY] Analog joystick and gamepad support
388 Specifies type or capabilities of an analog joystick
389 connected to one of 16 gameports
390 Format: <type1>,<type2>,..<type16>
391
392 apc= [HW,SPARC]
393 Power management functions (SPARCstation-4/5 + deriv.)
394 Format: noidle
395 Disable APC CPU standby support. SPARCstation-Fox does
396 not play well with APC CPU idle - disable it if you have
397 APC and your system crashes randomly.
398
399 apic= [APIC,X86,EARLY] Advanced Programmable Interrupt Controller
400 Change the output verbosity while booting
401 Format: { quiet (default) | verbose | debug }
402 Change the amount of debugging information output
403 when initialising the APIC and IO-APIC components.
404 For X86-32, this can also be used to specify an APIC
405 driver name.
406 Format: apic=driver_name
407 Examples: apic=bigsmp
408
409 apic_extnmi= [APIC,X86,EARLY] External NMI delivery setting
410 Format: { bsp (default) | all | none }
411 bsp: External NMI is delivered only to CPU 0
412 all: External NMIs are broadcast to all CPUs as a
413 backup of CPU 0
414 none: External NMI is masked for all CPUs. This is
415 useful so that a dump capture kernel won't be
416 shot down by NMI
417
418 autoconf= [IPV6]
419 See Documentation/networking/ipv6.rst.
420
421 apm= [APM] Advanced Power Management
422 See header of arch/x86/kernel/apm_32.c.
423
424 apparmor= [APPARMOR] Disable or enable AppArmor at boot time
425 Format: { "0" | "1" }
426 See security/apparmor/Kconfig help text
427 0 -- disable.
428 1 -- enable.
429 Default value is set via kernel config option.
430
431 arcrimi= [HW,NET] ARCnet - "RIM I" (entirely mem-mapped) cards
432 Format: <io>,<irq>,<nodeID>
433
434 arm64.nobti [ARM64] Unconditionally disable Branch Target
435 Identification support
436
437 arm64.nomops [ARM64] Unconditionally disable Memory Copy and Memory
438 Set instructions support
439
440 arm64.nomte [ARM64] Unconditionally disable Memory Tagging Extension
441 support
442
443 arm64.nopauth [ARM64] Unconditionally disable Pointer Authentication
444 support
445
446 arm64.nosme [ARM64] Unconditionally disable Scalable Matrix
447 Extension support
448
449 arm64.nosve [ARM64] Unconditionally disable Scalable Vector
450 Extension support
451
452 ataflop= [HW,M68k]
453
454 atarimouse= [HW,MOUSE] Atari Mouse
455
456 atkbd.extra= [HW] Enable extra LEDs and keys on IBM RapidAccess,
457 EzKey and similar keyboards
458
459 atkbd.reset= [HW] Reset keyboard during initialization
460
461 atkbd.set= [HW] Select keyboard code set
462 Format: <int> (2 = AT (default), 3 = PS/2)
463
464 atkbd.scroll= [HW] Enable scroll wheel on MS Office and similar
465 keyboards
466
467 atkbd.softraw= [HW] Choose between synthetic and real raw mode
468 Format: <bool> (0 = real, 1 = synthetic (default))
469
470 atkbd.softrepeat= [HW]
471 Use software keyboard repeat
472
473 audit= [KNL] Enable the audit sub-system
474 Format: { "0" | "1" | "off" | "on" }
475 0 | off - kernel audit is disabled and can not be
476 enabled until the next reboot
477 unset - kernel audit is initialized but disabled and
478 will be fully enabled by the userspace auditd.
479 1 | on - kernel audit is initialized and partially
480 enabled, storing at most audit_backlog_limit
481 messages in RAM until it is fully enabled by the
482 userspace auditd.
483 Default: unset
484
485 audit_backlog_limit= [KNL] Set the audit queue size limit.
486 Format: <int> (must be >=0)
487 Default: 64
488
489 bau= [X86_UV] Enable the BAU on SGI UV. The default
490 behavior is to disable the BAU (i.e. bau=0).
491 Format: { "0" | "1" }
492 0 - Disable the BAU.
493 1 - Enable the BAU.
494 unset - Disable the BAU.
495
496 baycom_epp= [HW,AX25]
497 Format: <io>,<mode>
498
499 baycom_par= [HW,AX25] BayCom Parallel Port AX.25 Modem
500 Format: <io>,<mode>
501 See header of drivers/net/hamradio/baycom_par.c.
502
503 baycom_ser_fdx= [HW,AX25]
504 BayCom Serial Port AX.25 Modem (Full Duplex Mode)
505 Format: <io>,<irq>,<mode>[,<baud>]
506 See header of drivers/net/hamradio/baycom_ser_fdx.c.
507
508 baycom_ser_hdx= [HW,AX25]
509 BayCom Serial Port AX.25 Modem (Half Duplex Mode)
510 Format: <io>,<irq>,<mode>
511 See header of drivers/net/hamradio/baycom_ser_hdx.c.
512
513 bert_disable [ACPI]
514 Disable BERT OS support on buggy BIOSes.
515
516 bgrt_disable [ACPI,X86,EARLY]
517 Disable BGRT to avoid flickering OEM logo.
518
519 blkdevparts= Manual partition parsing of block device(s) for
520 embedded devices based on command line input.
521 See Documentation/block/cmdline-partition.rst
522
523 boot_delay= [KNL,EARLY]
524 Milliseconds to delay each printk during boot.
525 Only works if CONFIG_BOOT_PRINTK_DELAY is enabled,
526 and you may also have to specify "lpj=". Boot_delay
527 values larger than 10 seconds (10000) are assumed
528 erroneous and ignored.
529 Format: integer
530
531 bootconfig [KNL,EARLY]
532 Extended command line options can be added to an initrd
533 and this will cause the kernel to look for it.
534
535 See Documentation/admin-guide/bootconfig.rst
536
537 bttv.card= [HW,V4L] bttv (bt848 + bt878 based grabber cards)
538 bttv.radio= Most important insmod options are available as
539 kernel args too.
540 bttv.pll= See Documentation/admin-guide/media/bttv.rst
541 bttv.tuner=
542
543 bulk_remove=off [PPC] This parameter disables the use of the pSeries
544 firmware feature for flushing multiple hpte entries
545 at a time.
546
547 c101= [NET] Moxa C101 synchronous serial card
548
549 cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection.
550 Sometimes CPU hardware bugs make them report the cache
551 size incorrectly. The kernel will attempt work arounds
552 to fix known problems, but for some CPUs it is not
553 possible to determine what the correct size should be.
554 This option provides an override for these situations.
555
556 carrier_timeout=
557 [NET] Specifies amount of time (in seconds) that
558 the kernel should wait for a network carrier. By default
559 it waits 120 seconds.
560
561 ca_keys= [KEYS] This parameter identifies a specific key(s) on
562 the system trusted keyring to be used for certificate
563 trust validation.
564 format: { id:<keyid> | builtin }
565
566 cca= [MIPS,EARLY] Override the kernel pages' cache coherency
567 algorithm. Accepted values range from 0 to 7
568 inclusive. See arch/mips/include/asm/pgtable-bits.h
569 for platform specific values (SB1, Loongson3 and
570 others).
571
572 ccw_timeout_log [S390]
573 See Documentation/arch/s390/common_io.rst for details.
574
575 cgroup_disable= [KNL] Disable a particular controller or optional feature
576 Format: {name of the controller(s) or feature(s) to disable}
577 The effects of cgroup_disable=foo are:
578 - foo isn't auto-mounted if you mount all cgroups in
579 a single hierarchy
580 - foo isn't visible as an individually mountable
581 subsystem
582 - if foo is an optional feature then the feature is
583 disabled and corresponding cgroup files are not
584 created
585 {Currently only "memory" controller deal with this and
586 cut the overhead, others just disable the usage. So
587 only cgroup_disable=memory is actually worthy}
588 Specifying "pressure" disables per-cgroup pressure
589 stall information accounting feature
590
591 cgroup_no_v1= [KNL] Disable cgroup controllers and named hierarchies in v1
592 Format: { { controller | "all" | "named" }
593 [,{ controller | "all" | "named" }...] }
594 Like cgroup_disable, but only applies to cgroup v1;
595 the blacklisted controllers remain available in cgroup2.
596 "all" blacklists all controllers and "named" disables
597 named mounts. Specifying both "all" and "named" disables
598 all v1 hierarchies.
599
600 cgroup_favordynmods= [KNL] Enable or Disable favordynmods.
601 Format: { "true" | "false" }
602 Defaults to the value of CONFIG_CGROUP_FAVOR_DYNMODS.
603
604 cgroup.memory= [KNL] Pass options to the cgroup memory controller.
605 Format: <string>
606 nosocket -- Disable socket memory accounting.
607 nokmem -- Disable kernel memory accounting.
608 nobpf -- Disable BPF memory accounting.
609
610 checkreqprot= [SELINUX] Set initial checkreqprot flag value.
611 Format: { "0" | "1" }
612 See security/selinux/Kconfig help text.
613 0 -- check protection applied by kernel (includes
614 any implied execute protection).
615 1 -- check protection requested by application.
616 Default value is set via a kernel config option.
617 Value can be changed at runtime via
618 /sys/fs/selinux/checkreqprot.
619 Setting checkreqprot to 1 is deprecated.
620
621 cio_ignore= [S390]
622 See Documentation/arch/s390/common_io.rst for details.
623
624 clearcpuid=X[,X...] [X86]
625 Disable CPUID feature X for the kernel. See
626 arch/x86/include/asm/cpufeatures.h for the valid bit
627 numbers X. Note the Linux-specific bits are not necessarily
628 stable over kernel options, but the vendor-specific
629 ones should be.
630 X can also be a string as appearing in the flags: line
631 in /proc/cpuinfo which does not have the above
632 instability issue. However, not all features have names
633 in /proc/cpuinfo.
634 Note that using this option will taint your kernel.
635 Also note that user programs calling CPUID directly
636 or using the feature without checking anything
637 will still see it. This just prevents it from
638 being used by the kernel or shown in /proc/cpuinfo.
639 Also note the kernel might malfunction if you disable
640 some critical bits.
641
642 clk_ignore_unused
643 [CLK]
644 Prevents the clock framework from automatically gating
645 clocks that have not been explicitly enabled by a Linux
646 device driver but are enabled in hardware at reset or
647 by the bootloader/firmware. Note that this does not
648 force such clocks to be always-on nor does it reserve
649 those clocks in any way. This parameter is useful for
650 debug and development, but should not be needed on a
651 platform with proper driver support. For more
652 information, see Documentation/driver-api/clk.rst.
653
654 clock= [BUGS=X86-32, HW] gettimeofday clocksource override.
655 [Deprecated]
656 Forces specified clocksource (if available) to be used
657 when calculating gettimeofday(). If specified
658 clocksource is not available, it defaults to PIT.
659 Format: { pit | tsc | cyclone | pmtmr }
660
661 clocksource= Override the default clocksource
662 Format: <string>
663 Override the default clocksource and use the clocksource
664 with the name specified.
665 Some clocksource names to choose from, depending on
666 the platform:
667 [all] jiffies (this is the base, fallback clocksource)
668 [ACPI] acpi_pm
669 [ARM] imx_timer1,OSTS,netx_timer,mpu_timer2,
670 pxa_timer,timer3,32k_counter,timer0_1
671 [X86-32] pit,hpet,tsc;
672 scx200_hrt on Geode; cyclone on IBM x440
673 [MIPS] MIPS
674 [PARISC] cr16
675 [S390] tod
676 [SH] SuperH
677 [SPARC64] tick
678 [X86-64] hpet,tsc
679
680 clocksource.arm_arch_timer.evtstrm=
681 [ARM,ARM64,EARLY]
682 Format: <bool>
683 Enable/disable the eventstream feature of the ARM
684 architected timer so that code using WFE-based polling
685 loops can be debugged more effectively on production
686 systems.
687
688 clocksource.verify_n_cpus= [KNL]
689 Limit the number of CPUs checked for clocksources
690 marked with CLOCK_SOURCE_VERIFY_PERCPU that
691 are marked unstable due to excessive skew.
692 A negative value says to check all CPUs, while
693 zero says not to check any. Values larger than
694 nr_cpu_ids are silently truncated to nr_cpu_ids.
695 The actual CPUs are chosen randomly, with
696 no replacement if the same CPU is chosen twice.
697
698 clocksource-wdtest.holdoff= [KNL]
699 Set the time in seconds that the clocksource
700 watchdog test waits before commencing its tests.
701 Defaults to zero when built as a module and to
702 10 seconds when built into the kernel.
703
704 cma=nn[MG]@[start[MG][-end[MG]]]
705 [KNL,CMA,EARLY]
706 Sets the size of kernel global memory area for
707 contiguous memory allocations and optionally the
708 placement constraint by the physical address range of
709 memory allocations. A value of 0 disables CMA
710 altogether. For more information, see
711 kernel/dma/contiguous.c
712
713 cma_pernuma=nn[MG]
714 [KNL,CMA,EARLY]
715 Sets the size of kernel per-numa memory area for
716 contiguous memory allocations. A value of 0 disables
717 per-numa CMA altogether. And If this option is not
718 specified, the default value is 0.
719 With per-numa CMA enabled, DMA users on node nid will
720 first try to allocate buffer from the pernuma area
721 which is located in node nid, if the allocation fails,
722 they will fallback to the global default memory area.
723
724 numa_cma=<node>:nn[MG][,<node>:nn[MG]]
725 [KNL,CMA,EARLY]
726 Sets the size of kernel numa memory area for
727 contiguous memory allocations. It will reserve CMA
728 area for the specified node.
729
730 With numa CMA enabled, DMA users on node nid will
731 first try to allocate buffer from the numa area
732 which is located in node nid, if the allocation fails,
733 they will fallback to the global default memory area.
734
735 cmo_free_hint= [PPC] Format: { yes | no }
736 Specify whether pages are marked as being inactive
737 when they are freed. This is used in CMO environments
738 to determine OS memory pressure for page stealing by
739 a hypervisor.
740 Default: yes
741
742 coherent_pool=nn[KMG] [ARM,KNL,EARLY]
743 Sets the size of memory pool for coherent, atomic dma
744 allocations, by default set to 256K.
745
746 com20020= [HW,NET] ARCnet - COM20020 chipset
747 Format:
748 <io>[,<irq>[,<nodeID>[,<backplane>[,<ckp>[,<timeout>]]]]]
749
750 com90io= [HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers)
751 Format: <io>[,<irq>]
752
753 com90xx= [HW,NET]
754 ARCnet - COM90xx chipset (memory-mapped buffers)
755 Format: <io>[,<irq>[,<memstart>]]
756
757 condev= [HW,S390] console device
758 conmode=
759
760 con3215_drop= [S390,EARLY] 3215 console drop mode.
761 Format: y|n|Y|N|1|0
762 When set to true, drop data on the 3215 console when
763 the console buffer is full. In this case the
764 operator using a 3270 terminal emulator (for example
765 x3270) does not have to enter the clear key for the
766 console output to advance and the kernel to continue.
767 This leads to a much faster boot time when a 3270
768 terminal emulator is active. If no 3270 terminal
769 emulator is used, this parameter has no effect.
770
771 console= [KNL] Output console device and options.
772
773 tty<n> Use the virtual console device <n>.
774
775 ttyS<n>[,options]
776 ttyUSB0[,options]
777 Use the specified serial port. The options are of
778 the form "bbbbpnf", where "bbbb" is the baud rate,
779 "p" is parity ("n", "o", or "e"), "n" is number of
780 bits, and "f" is flow control ("r" for RTS or
781 omit it). Default is "9600n8".
782
783 See Documentation/admin-guide/serial-console.rst for more
784 information. See
785 Documentation/networking/netconsole.rst for an
786 alternative.
787
788 uart[8250],io,<addr>[,options]
789 uart[8250],mmio,<addr>[,options]
790 uart[8250],mmio16,<addr>[,options]
791 uart[8250],mmio32,<addr>[,options]
792 uart[8250],0x<addr>[,options]
793 Start an early, polled-mode console on the 8250/16550
794 UART at the specified I/O port or MMIO address,
795 switching to the matching ttyS device later.
796 MMIO inter-register address stride is either 8-bit
797 (mmio), 16-bit (mmio16), or 32-bit (mmio32).
798 If none of [io|mmio|mmio16|mmio32], <addr> is assumed
799 to be equivalent to 'mmio'. 'options' are specified in
800 the same format described for ttyS above; if unspecified,
801 the h/w is not re-initialized.
802
803 hvc<n> Use the hypervisor console device <n>. This is for
804 both Xen and PowerPC hypervisors.
805
806 { null | "" }
807 Use to disable console output, i.e., to have kernel
808 console messages discarded.
809 This must be the only console= parameter used on the
810 kernel command line.
811
812 If the device connected to the port is not a TTY but a braille
813 device, prepend "brl," before the device type, for instance
814 console=brl,ttyS0
815 For now, only VisioBraille is supported.
816
817 console_msg_format=
818 [KNL] Change console messages format
819 default
820 By default we print messages on consoles in
821 "[time stamp] text\n" format (time stamp may not be
822 printed, depending on CONFIG_PRINTK_TIME or
823 `printk_time' param).
824 syslog
825 Switch to syslog format: "<%u>[time stamp] text\n"
826 IOW, each message will have a facility and loglevel
827 prefix. The format is similar to one used by syslog()
828 syscall, or to executing "dmesg -S --raw" or to reading
829 from /proc/kmsg.
830
831 consoleblank= [KNL] The console blank (screen saver) timeout in
832 seconds. A value of 0 disables the blank timer.
833 Defaults to 0.
834
835 coredump_filter=
836 [KNL] Change the default value for
837 /proc/<pid>/coredump_filter.
838 See also Documentation/filesystems/proc.rst.
839
840 coresight_cpu_debug.enable
841 [ARM,ARM64]
842 Format: <bool>
843 Enable/disable the CPU sampling based debugging.
844 0: default value, disable debugging
845 1: enable debugging at boot time
846
847 cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver
848 Format:
849 <first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>]
850
851 cpuidle.off=1 [CPU_IDLE]
852 disable the cpuidle sub-system
853
854 cpuidle.governor=
855 [CPU_IDLE] Name of the cpuidle governor to use.
856
857 cpufreq.off=1 [CPU_FREQ]
858 disable the cpufreq sub-system
859
860 cpufreq.default_governor=
861 [CPU_FREQ] Name of the default cpufreq governor or
862 policy to use. This governor must be registered in the
863 kernel before the cpufreq driver probes.
864
865 cpu_init_udelay=N
866 [X86,EARLY] Delay for N microsec between assert and de-assert
867 of APIC INIT to start processors. This delay occurs
868 on every CPU online, such as boot, and resume from suspend.
869 Default: 10000
870
871 cpuhp.parallel=
872 [SMP] Enable/disable parallel bringup of secondary CPUs
873 Format: <bool>
874 Default is enabled if CONFIG_HOTPLUG_PARALLEL=y. Otherwise
875 the parameter has no effect.
876
877 crash_kexec_post_notifiers
878 Run kdump after running panic-notifiers and dumping
879 kmsg. This only for the users who doubt kdump always
880 succeeds in any situation.
881 Note that this also increases risks of kdump failure,
882 because some panic notifiers can make the crashed
883 kernel more unstable.
884
885 crashkernel=size[KMG][@offset[KMG]]
886 [KNL,EARLY] Using kexec, Linux can switch to a 'crash kernel'
887 upon panic. This parameter reserves the physical
888 memory region [offset, offset + size] for that kernel
889 image. If '@offset' is omitted, then a suitable offset
890 is selected automatically.
891 [KNL, X86-64, ARM64, RISCV, LoongArch] Select a region
892 under 4G first, and fall back to reserve region above
893 4G when '@offset' hasn't been specified.
894 See Documentation/admin-guide/kdump/kdump.rst for further details.
895
896 crashkernel=range1:size1[,range2:size2,...][@offset]
897 [KNL] Same as above, but depends on the memory
898 in the running system. The syntax of range is
899 start-[end] where start and end are both
900 a memory unit (amount[KMG]). See also
901 Documentation/admin-guide/kdump/kdump.rst for an example.
902
903 crashkernel=size[KMG],high
904 [KNL, X86-64, ARM64, RISCV, LoongArch] range could be
905 above 4G.
906 Allow kernel to allocate physical memory region from top,
907 so could be above 4G if system have more than 4G ram
908 installed. Otherwise memory region will be allocated
909 below 4G, if available.
910 It will be ignored if crashkernel=X is specified.
911 crashkernel=size[KMG],low
912 [KNL, X86-64, ARM64, RISCV, LoongArch] range under 4G.
913 When crashkernel=X,high is passed, kernel could allocate
914 physical memory region above 4G, that cause second kernel
915 crash on system that require some amount of low memory,
916 e.g. swiotlb requires at least 64M+32K low memory, also
917 enough extra low memory is needed to make sure DMA buffers
918 for 32-bit devices won't run out. Kernel would try to allocate
919 default size of memory below 4G automatically. The default
920 size is platform dependent.
921 --> x86: max(swiotlb_size_or_default() + 8MiB, 256MiB)
922 --> arm64: 128MiB
923 --> riscv: 128MiB
924 --> loongarch: 128MiB
925 This one lets the user specify own low range under 4G
926 for second kernel instead.
927 0: to disable low allocation.
928 It will be ignored when crashkernel=X,high is not used
929 or memory reserved is below 4G.
930
931 cryptomgr.notests
932 [KNL] Disable crypto self-tests
933
934 cs89x0_dma= [HW,NET]
935 Format: <dma>
936
937 cs89x0_media= [HW,NET]
938 Format: { rj45 | aui | bnc }
939
940 csdlock_debug= [KNL] Enable or disable debug add-ons of cross-CPU
941 function call handling. When switched on,
942 additional debug data is printed to the console
943 in case a hanging CPU is detected, and that
944 CPU is pinged again in order to try to resolve
945 the hang situation. The default value of this
946 option depends on the CSD_LOCK_WAIT_DEBUG_DEFAULT
947 Kconfig option.
948
949 dasd= [HW,NET]
950 See header of drivers/s390/block/dasd_devmap.c.
951
952 db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port
953 (one device per port)
954 Format: <port#>,<type>
955 See also Documentation/input/devices/joystick-parport.rst
956
957 debug [KNL,EARLY] Enable kernel debugging (events log level).
958
959 debug_boot_weak_hash
960 [KNL,EARLY] Enable printing [hashed] pointers early in the
961 boot sequence. If enabled, we use a weak hash instead
962 of siphash to hash pointers. Use this option if you are
963 seeing instances of '(___ptrval___)') and need to see a
964 value (hashed pointer) instead. Cryptographically
965 insecure, please do not use on production kernels.
966
967 debug_locks_verbose=
968 [KNL] verbose locking self-tests
969 Format: <int>
970 Print debugging info while doing the locking API
971 self-tests.
972 Bitmask for the various LOCKTYPE_ tests. Defaults to 0
973 (no extra messages), setting it to -1 (all bits set)
974 will print _a_lot_ more information - normally only
975 useful to lockdep developers.
976
977 debug_objects [KNL,EARLY] Enable object debugging
978
979 debug_guardpage_minorder=
980 [KNL,EARLY] When CONFIG_DEBUG_PAGEALLOC is set, this
981 parameter allows control of the order of pages that will
982 be intentionally kept free (and hence protected) by the
983 buddy allocator. Bigger value increase the probability
984 of catching random memory corruption, but reduce the
985 amount of memory for normal system use. The maximum
986 possible value is MAX_PAGE_ORDER/2. Setting this
987 parameter to 1 or 2 should be enough to identify most
988 random memory corruption problems caused by bugs in
989 kernel or driver code when a CPU writes to (or reads
990 from) a random memory location. Note that there exists
991 a class of memory corruptions problems caused by buggy
992 H/W or F/W or by drivers badly programming DMA
993 (basically when memory is written at bus level and the
994 CPU MMU is bypassed) which are not detectable by
995 CONFIG_DEBUG_PAGEALLOC, hence this option will not
996 help tracking down these problems.
997
998 debug_pagealloc=
999 [KNL,EARLY] When CONFIG_DEBUG_PAGEALLOC is set, this parameter
1000 enables the feature at boot time. By default, it is
1001 disabled and the system will work mostly the same as a
1002 kernel built without CONFIG_DEBUG_PAGEALLOC.
1003 Note: to get most of debug_pagealloc error reports, it's
1004 useful to also enable the page_owner functionality.
1005 on: enable the feature
1006
1007 debugfs= [KNL,EARLY] This parameter enables what is exposed to
1008 userspace and debugfs internal clients.
1009 Format: { on, no-mount, off }
1010 on: All functions are enabled.
1011 no-mount:
1012 Filesystem is not registered but kernel clients can
1013 access APIs and a crashkernel can be used to read
1014 its content. There is nothing to mount.
1015 off: Filesystem is not registered and clients
1016 get a -EPERM as result when trying to register files
1017 or directories within debugfs.
1018 This is equivalent of the runtime functionality if
1019 debugfs was not enabled in the kernel at all.
1020 Default value is set in build-time with a kernel configuration.
1021
1022 debugpat [X86] Enable PAT debugging
1023
1024 default_hugepagesz=
1025 [HW] The size of the default HugeTLB page. This is
1026 the size represented by the legacy /proc/ hugepages
1027 APIs. In addition, this is the default hugetlb size
1028 used for shmget(), mmap() and mounting hugetlbfs
1029 filesystems. If not specified, defaults to the
1030 architecture's default huge page size. Huge page
1031 sizes are architecture dependent. See also
1032 Documentation/admin-guide/mm/hugetlbpage.rst.
1033 Format: size[KMG]
1034
1035 deferred_probe_timeout=
1036 [KNL] Debugging option to set a timeout in seconds for
1037 deferred probe to give up waiting on dependencies to
1038 probe. Only specific dependencies (subsystems or
1039 drivers) that have opted in will be ignored. A timeout
1040 of 0 will timeout at the end of initcalls. If the time
1041 out hasn't expired, it'll be restarted by each
1042 successful driver registration. This option will also
1043 dump out devices still on the deferred probe list after
1044 retrying.
1045
1046 delayacct [KNL] Enable per-task delay accounting
1047
1048 dell_smm_hwmon.ignore_dmi=
1049 [HW] Continue probing hardware even if DMI data
1050 indicates that the driver is running on unsupported
1051 hardware.
1052
1053 dell_smm_hwmon.force=
1054 [HW] Activate driver even if SMM BIOS signature does
1055 not match list of supported models and enable otherwise
1056 blacklisted features.
1057
1058 dell_smm_hwmon.power_status=
1059 [HW] Report power status in /proc/i8k
1060 (disabled by default).
1061
1062 dell_smm_hwmon.restricted=
1063 [HW] Allow controlling fans only if SYS_ADMIN
1064 capability is set.
1065
1066 dell_smm_hwmon.fan_mult=
1067 [HW] Factor to multiply fan speed with.
1068
1069 dell_smm_hwmon.fan_max=
1070 [HW] Maximum configurable fan speed.
1071
1072 dfltcc= [HW,S390]
1073 Format: { on | off | def_only | inf_only | always }
1074 on: s390 zlib hardware support for compression on
1075 level 1 and decompression (default)
1076 off: No s390 zlib hardware support
1077 def_only: s390 zlib hardware support for deflate
1078 only (compression on level 1)
1079 inf_only: s390 zlib hardware support for inflate
1080 only (decompression)
1081 always: Same as 'on' but ignores the selected compression
1082 level always using hardware support (used for debugging)
1083
1084 dhash_entries= [KNL]
1085 Set number of hash buckets for dentry cache.
1086
1087 disable_1tb_segments [PPC,EARLY]
1088 Disables the use of 1TB hash page table segments. This
1089 causes the kernel to fall back to 256MB segments which
1090 can be useful when debugging issues that require an SLB
1091 miss to occur.
1092
1093 disable= [IPV6]
1094 See Documentation/networking/ipv6.rst.
1095
1096 disable_radix [PPC,EARLY]
1097 Disable RADIX MMU mode on POWER9
1098
1099 disable_tlbie [PPC]
1100 Disable TLBIE instruction. Currently does not work
1101 with KVM, with HASH MMU, or with coherent accelerators.
1102
1103 disable_ddw [PPC/PSERIES,EARLY]
1104 Disable Dynamic DMA Window support. Use this
1105 to workaround buggy firmware.
1106
1107 disable_ipv6= [IPV6]
1108 See Documentation/networking/ipv6.rst.
1109
1110 disable_mtrr_cleanup [X86,EARLY]
1111 The kernel tries to adjust MTRR layout from continuous
1112 to discrete, to make X server driver able to add WB
1113 entry later. This parameter disables that.
1114
1115 disable_mtrr_trim [X86, Intel and AMD only,EARLY]
1116 By default the kernel will trim any uncacheable
1117 memory out of your available memory pool based on
1118 MTRR settings. This parameter disables that behavior,
1119 possibly causing your machine to run very slowly.
1120
1121 disable_timer_pin_1 [X86,EARLY]
1122 Disable PIN 1 of APIC timer
1123 Can be useful to work around chipset bugs.
1124
1125 dis_ucode_ldr [X86] Disable the microcode loader.
1126
1127 dma_debug=off If the kernel is compiled with DMA_API_DEBUG support,
1128 this option disables the debugging code at boot.
1129
1130 dma_debug_entries=<number>
1131 This option allows to tune the number of preallocated
1132 entries for DMA-API debugging code. One entry is
1133 required per DMA-API allocation. Use this if the
1134 DMA-API debugging code disables itself because the
1135 architectural default is too low.
1136
1137 dma_debug_driver=<driver_name>
1138 With this option the DMA-API debugging driver
1139 filter feature can be enabled at boot time. Just
1140 pass the driver to filter for as the parameter.
1141 The filter can be disabled or changed to another
1142 driver later using sysfs.
1143
1144 reg_file_data_sampling=
1145 [X86] Controls mitigation for Register File Data
1146 Sampling (RFDS) vulnerability. RFDS is a CPU
1147 vulnerability which may allow userspace to infer
1148 kernel data values previously stored in floating point
1149 registers, vector registers, or integer registers.
1150 RFDS only affects Intel Atom processors.
1151
1152 on: Turns ON the mitigation.
1153 off: Turns OFF the mitigation.
1154
1155 This parameter overrides the compile time default set
1156 by CONFIG_MITIGATION_RFDS. Mitigation cannot be
1157 disabled when other VERW based mitigations (like MDS)
1158 are enabled. In order to disable RFDS mitigation all
1159 VERW based mitigations need to be disabled.
1160
1161 For details see:
1162 Documentation/admin-guide/hw-vuln/reg-file-data-sampling.rst
1163
1164 driver_async_probe= [KNL]
1165 List of driver names to be probed asynchronously. *
1166 matches with all driver names. If * is specified, the
1167 rest of the listed driver names are those that will NOT
1168 match the *.
1169 Format: <driver_name1>,<driver_name2>...
1170
1171 drm.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>]
1172 Broken monitors, graphic adapters, KVMs and EDIDless
1173 panels may send no or incorrect EDID data sets.
1174 This parameter allows to specify an EDID data sets
1175 in the /lib/firmware directory that are used instead.
1176 An EDID data set will only be used for a particular
1177 connector, if its name and a colon are prepended to
1178 the EDID name. Each connector may use a unique EDID
1179 data set by separating the files with a comma. An EDID
1180 data set with no connector name will be used for
1181 any connectors not explicitly specified.
1182
1183 dscc4.setup= [NET]
1184
1185 dt_cpu_ftrs= [PPC,EARLY]
1186 Format: {"off" | "known"}
1187 Control how the dt_cpu_ftrs device-tree binding is
1188 used for CPU feature discovery and setup (if it
1189 exists).
1190 off: Do not use it, fall back to legacy cpu table.
1191 known: Do not pass through unknown features to guests
1192 or userspace, only those that the kernel is aware of.
1193
1194 dump_apple_properties [X86]
1195 Dump name and content of EFI device properties on
1196 x86 Macs. Useful for driver authors to determine
1197 what data is available or for reverse-engineering.
1198
1199 dyndbg[="val"] [KNL,DYNAMIC_DEBUG]
1200 <module>.dyndbg[="val"]
1201 Enable debug messages at boot time. See
1202 Documentation/admin-guide/dynamic-debug-howto.rst
1203 for details.
1204
1205 early_ioremap_debug [KNL,EARLY]
1206 Enable debug messages in early_ioremap support. This
1207 is useful for tracking down temporary early mappings
1208 which are not unmapped.
1209
1210 earlycon= [KNL,EARLY] Output early console device and options.
1211
1212 When used with no options, the early console is
1213 determined by stdout-path property in device tree's
1214 chosen node or the ACPI SPCR table if supported by
1215 the platform.
1216
1217 cdns,<addr>[,options]
1218 Start an early, polled-mode console on a Cadence
1219 (xuartps) serial port at the specified address. Only
1220 supported option is baud rate. If baud rate is not
1221 specified, the serial port must already be setup and
1222 configured.
1223
1224 uart[8250],io,<addr>[,options[,uartclk]]
1225 uart[8250],mmio,<addr>[,options[,uartclk]]
1226 uart[8250],mmio32,<addr>[,options[,uartclk]]
1227 uart[8250],mmio32be,<addr>[,options[,uartclk]]
1228 uart[8250],0x<addr>[,options]
1229 Start an early, polled-mode console on the 8250/16550
1230 UART at the specified I/O port or MMIO address.
1231 MMIO inter-register address stride is either 8-bit
1232 (mmio) or 32-bit (mmio32 or mmio32be).
1233 If none of [io|mmio|mmio32|mmio32be], <addr> is assumed
1234 to be equivalent to 'mmio'. 'options' are specified
1235 in the same format described for "console=ttyS<n>"; if
1236 unspecified, the h/w is not initialized. 'uartclk' is
1237 the uart clock frequency; if unspecified, it is set
1238 to 'BASE_BAUD' * 16.
1239
1240 pl011,<addr>
1241 pl011,mmio32,<addr>
1242 Start an early, polled-mode console on a pl011 serial
1243 port at the specified address. The pl011 serial port
1244 must already be setup and configured. Options are not
1245 yet supported. If 'mmio32' is specified, then only
1246 the driver will use only 32-bit accessors to read/write
1247 the device registers.
1248
1249 liteuart,<addr>
1250 Start an early console on a litex serial port at the
1251 specified address. The serial port must already be
1252 setup and configured. Options are not yet supported.
1253
1254 meson,<addr>
1255 Start an early, polled-mode console on a meson serial
1256 port at the specified address. The serial port must
1257 already be setup and configured. Options are not yet
1258 supported.
1259
1260 msm_serial,<addr>
1261 Start an early, polled-mode console on an msm serial
1262 port at the specified address. The serial port
1263 must already be setup and configured. Options are not
1264 yet supported.
1265
1266 msm_serial_dm,<addr>
1267 Start an early, polled-mode console on an msm serial
1268 dm port at the specified address. The serial port
1269 must already be setup and configured. Options are not
1270 yet supported.
1271
1272 owl,<addr>
1273 Start an early, polled-mode console on a serial port
1274 of an Actions Semi SoC, such as S500 or S900, at the
1275 specified address. The serial port must already be
1276 setup and configured. Options are not yet supported.
1277
1278 rda,<addr>
1279 Start an early, polled-mode console on a serial port
1280 of an RDA Micro SoC, such as RDA8810PL, at the
1281 specified address. The serial port must already be
1282 setup and configured. Options are not yet supported.
1283
1284 sbi
1285 Use RISC-V SBI (Supervisor Binary Interface) for early
1286 console.
1287
1288 smh Use ARM semihosting calls for early console.
1289
1290 s3c2410,<addr>
1291 s3c2412,<addr>
1292 s3c2440,<addr>
1293 s3c6400,<addr>
1294 s5pv210,<addr>
1295 exynos4210,<addr>
1296 Use early console provided by serial driver available
1297 on Samsung SoCs, requires selecting proper type and
1298 a correct base address of the selected UART port. The
1299 serial port must already be setup and configured.
1300 Options are not yet supported.
1301
1302 lantiq,<addr>
1303 Start an early, polled-mode console on a lantiq serial
1304 (lqasc) port at the specified address. The serial port
1305 must already be setup and configured. Options are not
1306 yet supported.
1307
1308 lpuart,<addr>
1309 lpuart32,<addr>
1310 Use early console provided by Freescale LP UART driver
1311 found on Freescale Vybrid and QorIQ LS1021A processors.
1312 A valid base address must be provided, and the serial
1313 port must already be setup and configured.
1314
1315 ec_imx21,<addr>
1316 ec_imx6q,<addr>
1317 Start an early, polled-mode, output-only console on the
1318 Freescale i.MX UART at the specified address. The UART
1319 must already be setup and configured.
1320
1321 ar3700_uart,<addr>
1322 Start an early, polled-mode console on the
1323 Armada 3700 serial port at the specified
1324 address. The serial port must already be setup
1325 and configured. Options are not yet supported.
1326
1327 qcom_geni,<addr>
1328 Start an early, polled-mode console on a Qualcomm
1329 Generic Interface (GENI) based serial port at the
1330 specified address. The serial port must already be
1331 setup and configured. Options are not yet supported.
1332
1333 efifb,[options]
1334 Start an early, unaccelerated console on the EFI
1335 memory mapped framebuffer (if available). On cache
1336 coherent non-x86 systems that use system memory for
1337 the framebuffer, pass the 'ram' option so that it is
1338 mapped with the correct attributes.
1339
1340 linflex,<addr>
1341 Use early console provided by Freescale LINFlexD UART
1342 serial driver for NXP S32V234 SoCs. A valid base
1343 address must be provided, and the serial port must
1344 already be setup and configured.
1345
1346 earlyprintk= [X86,SH,ARM,M68k,S390,UM,EARLY]
1347 earlyprintk=vga
1348 earlyprintk=sclp
1349 earlyprintk=xen
1350 earlyprintk=serial[,ttySn[,baudrate]]
1351 earlyprintk=serial[,0x...[,baudrate]]
1352 earlyprintk=ttySn[,baudrate]
1353 earlyprintk=dbgp[debugController#]
1354 earlyprintk=pciserial[,force],bus:device.function[,baudrate]
1355 earlyprintk=xdbc[xhciController#]
1356 earlyprintk=bios
1357
1358 earlyprintk is useful when the kernel crashes before
1359 the normal console is initialized. It is not enabled by
1360 default because it has some cosmetic problems.
1361
1362 Append ",keep" to not disable it when the real console
1363 takes over.
1364
1365 Only one of vga, serial, or usb debug port can
1366 be used at a time.
1367
1368 Currently only ttyS0 and ttyS1 may be specified by
1369 name. Other I/O ports may be explicitly specified
1370 on some architectures (x86 and arm at least) by
1371 replacing ttySn with an I/O port address, like this:
1372 earlyprintk=serial,0x1008,115200
1373 You can find the port for a given device in
1374 /proc/tty/driver/serial:
1375 2: uart:ST16650V2 port:00001008 irq:18 ...
1376
1377 Interaction with the standard serial driver is not
1378 very good.
1379
1380 The VGA output is eventually overwritten by
1381 the real console.
1382
1383 The xen option can only be used in Xen domains.
1384
1385 The sclp output can only be used on s390.
1386
1387 The bios output can only be used on SuperH.
1388
1389 The optional "force" to "pciserial" enables use of a
1390 PCI device even when its classcode is not of the
1391 UART class.
1392
1393 edac_report= [HW,EDAC] Control how to report EDAC event
1394 Format: {"on" | "off" | "force"}
1395 on: enable EDAC to report H/W event. May be overridden
1396 by other higher priority error reporting module.
1397 off: disable H/W event reporting through EDAC.
1398 force: enforce the use of EDAC to report H/W event.
1399 default: on.
1400
1401 edd= [EDD]
1402 Format: {"off" | "on" | "skip[mbr]"}
1403
1404 efi= [EFI,EARLY]
1405 Format: { "debug", "disable_early_pci_dma",
1406 "nochunk", "noruntime", "nosoftreserve",
1407 "novamap", "no_disable_early_pci_dma" }
1408 debug: enable misc debug output.
1409 disable_early_pci_dma: disable the busmaster bit on all
1410 PCI bridges while in the EFI boot stub.
1411 nochunk: disable reading files in "chunks" in the EFI
1412 boot stub, as chunking can cause problems with some
1413 firmware implementations.
1414 noruntime : disable EFI runtime services support
1415 nosoftreserve: The EFI_MEMORY_SP (Specific Purpose)
1416 attribute may cause the kernel to reserve the
1417 memory range for a memory mapping driver to
1418 claim. Specify efi=nosoftreserve to disable this
1419 reservation and treat the memory by its base type
1420 (i.e. EFI_CONVENTIONAL_MEMORY / "System RAM").
1421 novamap: do not call SetVirtualAddressMap().
1422 no_disable_early_pci_dma: Leave the busmaster bit set
1423 on all PCI bridges while in the EFI boot stub
1424
1425 efi_no_storage_paranoia [EFI,X86,EARLY]
1426 Using this parameter you can use more than 50% of
1427 your efi variable storage. Use this parameter only if
1428 you are really sure that your UEFI does sane gc and
1429 fulfills the spec otherwise your board may brick.
1430
1431 efi_fake_mem= nn[KMG]@ss[KMG]:aa[,nn[KMG]@ss[KMG]:aa,..] [EFI,X86,EARLY]
1432 Add arbitrary attribute to specific memory range by
1433 updating original EFI memory map.
1434 Region of memory which aa attribute is added to is
1435 from ss to ss+nn.
1436
1437 If efi_fake_mem=2G@4G:0x10000,2G@0x10a0000000:0x10000
1438 is specified, EFI_MEMORY_MORE_RELIABLE(0x10000)
1439 attribute is added to range 0x100000000-0x180000000 and
1440 0x10a0000000-0x1120000000.
1441
1442 If efi_fake_mem=8G@9G:0x40000 is specified, the
1443 EFI_MEMORY_SP(0x40000) attribute is added to
1444 range 0x240000000-0x43fffffff.
1445
1446 Using this parameter you can do debugging of EFI memmap
1447 related features. For example, you can do debugging of
1448 Address Range Mirroring feature even if your box
1449 doesn't support it, or mark specific memory as
1450 "soft reserved".
1451
1452 efivar_ssdt= [EFI; X86] Name of an EFI variable that contains an SSDT
1453 that is to be dynamically loaded by Linux. If there are
1454 multiple variables with the same name but with different
1455 vendor GUIDs, all of them will be loaded. See
1456 Documentation/admin-guide/acpi/ssdt-overlays.rst for details.
1457
1458
1459 eisa_irq_edge= [PARISC,HW]
1460 See header of drivers/parisc/eisa.c.
1461
1462 ekgdboc= [X86,KGDB,EARLY] Allow early kernel console debugging
1463 Format: ekgdboc=kbd
1464
1465 This is designed to be used in conjunction with
1466 the boot argument: earlyprintk=vga
1467
1468 This parameter works in place of the kgdboc parameter
1469 but can only be used if the backing tty is available
1470 very early in the boot process. For early debugging
1471 via a serial port see kgdboc_earlycon instead.
1472
1473 elanfreq= [X86-32]
1474 See comment before function elanfreq_setup() in
1475 arch/x86/kernel/cpu/cpufreq/elanfreq.c.
1476
1477 elfcorehdr=[size[KMG]@]offset[KMG] [PPC,SH,X86,S390,EARLY]
1478 Specifies physical address of start of kernel core
1479 image elf header and optionally the size. Generally
1480 kexec loader will pass this option to capture kernel.
1481 See Documentation/admin-guide/kdump/kdump.rst for details.
1482
1483 enable_mtrr_cleanup [X86,EARLY]
1484 The kernel tries to adjust MTRR layout from continuous
1485 to discrete, to make X server driver able to add WB
1486 entry later. This parameter enables that.
1487
1488 enable_timer_pin_1 [X86]
1489 Enable PIN 1 of APIC timer
1490 Can be useful to work around chipset bugs
1491 (in particular on some ATI chipsets).
1492 The kernel tries to set a reasonable default.
1493
1494 enforcing= [SELINUX] Set initial enforcing status.
1495 Format: {"0" | "1"}
1496 See security/selinux/Kconfig help text.
1497 0 -- permissive (log only, no denials).
1498 1 -- enforcing (deny and log).
1499 Default value is 0.
1500 Value can be changed at runtime via
1501 /sys/fs/selinux/enforce.
1502
1503 erst_disable [ACPI]
1504 Disable Error Record Serialization Table (ERST)
1505 support.
1506
1507 ether= [HW,NET] Ethernet cards parameters
1508 This option is obsoleted by the "netdev=" option, which
1509 has equivalent usage. See its documentation for details.
1510
1511 evm= [EVM]
1512 Format: { "fix" }
1513 Permit 'security.evm' to be updated regardless of
1514 current integrity status.
1515
1516 early_page_ext [KNL,EARLY] Enforces page_ext initialization to earlier
1517 stages so cover more early boot allocations.
1518 Please note that as side effect some optimizations
1519 might be disabled to achieve that (e.g. parallelized
1520 memory initialization is disabled) so the boot process
1521 might take longer, especially on systems with a lot of
1522 memory. Available with CONFIG_PAGE_EXTENSION=y.
1523
1524 failslab=
1525 fail_usercopy=
1526 fail_page_alloc=
1527 fail_make_request=[KNL]
1528 General fault injection mechanism.
1529 Format: <interval>,<probability>,<space>,<times>
1530 See also Documentation/fault-injection/.
1531
1532 fb_tunnels= [NET]
1533 Format: { initns | none }
1534 See Documentation/admin-guide/sysctl/net.rst for
1535 fb_tunnels_only_for_init_ns
1536
1537 floppy= [HW]
1538 See Documentation/admin-guide/blockdev/floppy.rst.
1539
1540 forcepae [X86-32]
1541 Forcefully enable Physical Address Extension (PAE).
1542 Many Pentium M systems disable PAE but may have a
1543 functionally usable PAE implementation.
1544 Warning: use of this parameter will taint the kernel
1545 and may cause unknown problems.
1546
1547 fred= [X86-64]
1548 Enable/disable Flexible Return and Event Delivery.
1549 Format: { on | off }
1550 on: enable FRED when it's present.
1551 off: disable FRED, the default setting.
1552
1553 ftrace=[tracer]
1554 [FTRACE] will set and start the specified tracer
1555 as early as possible in order to facilitate early
1556 boot debugging.
1557
1558 ftrace_boot_snapshot
1559 [FTRACE] On boot up, a snapshot will be taken of the
1560 ftrace ring buffer that can be read at:
1561 /sys/kernel/tracing/snapshot.
1562 This is useful if you need tracing information from kernel
1563 boot up that is likely to be overridden by user space
1564 start up functionality.
1565
1566 Optionally, the snapshot can also be defined for a tracing
1567 instance that was created by the trace_instance= command
1568 line parameter.
1569
1570 trace_instance=foo,sched_switch ftrace_boot_snapshot=foo
1571
1572 The above will cause the "foo" tracing instance to trigger
1573 a snapshot at the end of boot up.
1574
1575 ftrace_dump_on_oops[=2(orig_cpu) | =<instance>][,<instance> |
1576 ,<instance>=2(orig_cpu)]
1577 [FTRACE] will dump the trace buffers on oops.
1578 If no parameter is passed, ftrace will dump global
1579 buffers of all CPUs, if you pass 2 or orig_cpu, it
1580 will dump only the buffer of the CPU that triggered
1581 the oops, or the specific instance will be dumped if
1582 its name is passed. Multiple instance dump is also
1583 supported, and instances are separated by commas. Each
1584 instance supports only dump on CPU that triggered the
1585 oops by passing 2 or orig_cpu to it.
1586
1587 ftrace_dump_on_oops=foo=orig_cpu
1588
1589 The above will dump only the buffer of "foo" instance
1590 on CPU that triggered the oops.
1591
1592 ftrace_dump_on_oops,foo,bar=orig_cpu
1593
1594 The above will dump global buffer on all CPUs, the
1595 buffer of "foo" instance on all CPUs and the buffer
1596 of "bar" instance on CPU that triggered the oops.
1597
1598 ftrace_filter=[function-list]
1599 [FTRACE] Limit the functions traced by the function
1600 tracer at boot up. function-list is a comma-separated
1601 list of functions. This list can be changed at run
1602 time by the set_ftrace_filter file in the debugfs
1603 tracing directory.
1604
1605 ftrace_notrace=[function-list]
1606 [FTRACE] Do not trace the functions specified in
1607 function-list. This list can be changed at run time
1608 by the set_ftrace_notrace file in the debugfs
1609 tracing directory.
1610
1611 ftrace_graph_filter=[function-list]
1612 [FTRACE] Limit the top level callers functions traced
1613 by the function graph tracer at boot up.
1614 function-list is a comma-separated list of functions
1615 that can be changed at run time by the
1616 set_graph_function file in the debugfs tracing directory.
1617
1618 ftrace_graph_notrace=[function-list]
1619 [FTRACE] Do not trace from the functions specified in
1620 function-list. This list is a comma-separated list of
1621 functions that can be changed at run time by the
1622 set_graph_notrace file in the debugfs tracing directory.
1623
1624 ftrace_graph_max_depth=<uint>
1625 [FTRACE] Used with the function graph tracer. This is
1626 the max depth it will trace into a function. This value
1627 can be changed at run time by the max_graph_depth file
1628 in the tracefs tracing directory. default: 0 (no limit)
1629
1630 fw_devlink= [KNL,EARLY] Create device links between consumer and supplier
1631 devices by scanning the firmware to infer the
1632 consumer/supplier relationships. This feature is
1633 especially useful when drivers are loaded as modules as
1634 it ensures proper ordering of tasks like device probing
1635 (suppliers first, then consumers), supplier boot state
1636 clean up (only after all consumers have probed),
1637 suspend/resume & runtime PM (consumers first, then
1638 suppliers).
1639 Format: { off | permissive | on | rpm }
1640 off -- Don't create device links from firmware info.
1641 permissive -- Create device links from firmware info
1642 but use it only for ordering boot state clean
1643 up (sync_state() calls).
1644 on -- Create device links from firmware info and use it
1645 to enforce probe and suspend/resume ordering.
1646 rpm -- Like "on", but also use to order runtime PM.
1647
1648 fw_devlink.strict=<bool>
1649 [KNL,EARLY] Treat all inferred dependencies as mandatory
1650 dependencies. This only applies for fw_devlink=on|rpm.
1651 Format: <bool>
1652
1653 fw_devlink.sync_state =
1654 [KNL,EARLY] When all devices that could probe have finished
1655 probing, this parameter controls what to do with
1656 devices that haven't yet received their sync_state()
1657 calls.
1658 Format: { strict | timeout }
1659 strict -- Default. Continue waiting on consumers to
1660 probe successfully.
1661 timeout -- Give up waiting on consumers and call
1662 sync_state() on any devices that haven't yet
1663 received their sync_state() calls after
1664 deferred_probe_timeout has expired or by
1665 late_initcall() if !CONFIG_MODULES.
1666
1667 gamecon.map[2|3]=
1668 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad
1669 support via parallel port (up to 5 devices per port)
1670 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5>
1671 See also Documentation/input/devices/joystick-parport.rst
1672
1673 gamma= [HW,DRM]
1674
1675 gart_fix_e820= [X86-64,EARLY] disable the fix e820 for K8 GART
1676 Format: off | on
1677 default: on
1678
1679 gather_data_sampling=
1680 [X86,INTEL,EARLY] Control the Gather Data Sampling (GDS)
1681 mitigation.
1682
1683 Gather Data Sampling is a hardware vulnerability which
1684 allows unprivileged speculative access to data which was
1685 previously stored in vector registers.
1686
1687 This issue is mitigated by default in updated microcode.
1688 The mitigation may have a performance impact but can be
1689 disabled. On systems without the microcode mitigation
1690 disabling AVX serves as a mitigation.
1691
1692 force: Disable AVX to mitigate systems without
1693 microcode mitigation. No effect if the microcode
1694 mitigation is present. Known to cause crashes in
1695 userspace with buggy AVX enumeration.
1696
1697 off: Disable GDS mitigation.
1698
1699 gcov_persist= [GCOV] When non-zero (default), profiling data for
1700 kernel modules is saved and remains accessible via
1701 debugfs, even when the module is unloaded/reloaded.
1702 When zero, profiling data is discarded and associated
1703 debugfs files are removed at module unload time.
1704
1705 goldfish [X86] Enable the goldfish android emulator platform.
1706 Don't use this when you are not running on the
1707 android emulator
1708
1709 gpio-mockup.gpio_mockup_ranges
1710 [HW] Sets the ranges of gpiochip of for this device.
1711 Format: <start1>,<end1>,<start2>,<end2>...
1712 gpio-mockup.gpio_mockup_named_lines
1713 [HW] Let the driver know GPIO lines should be named.
1714
1715 gpt [EFI] Forces disk with valid GPT signature but
1716 invalid Protective MBR to be treated as GPT. If the
1717 primary GPT is corrupted, it enables the backup/alternate
1718 GPT to be used instead.
1719
1720 grcan.enable0= [HW] Configuration of physical interface 0. Determines
1721 the "Enable 0" bit of the configuration register.
1722 Format: 0 | 1
1723 Default: 0
1724 grcan.enable1= [HW] Configuration of physical interface 1. Determines
1725 the "Enable 0" bit of the configuration register.
1726 Format: 0 | 1
1727 Default: 0
1728 grcan.select= [HW] Select which physical interface to use.
1729 Format: 0 | 1
1730 Default: 0
1731 grcan.txsize= [HW] Sets the size of the tx buffer.
1732 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0.
1733 Default: 1024
1734 grcan.rxsize= [HW] Sets the size of the rx buffer.
1735 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0.
1736 Default: 1024
1737
1738 hardened_usercopy=
1739 [KNL] Under CONFIG_HARDENED_USERCOPY, whether
1740 hardening is enabled for this boot. Hardened
1741 usercopy checking is used to protect the kernel
1742 from reading or writing beyond known memory
1743 allocation boundaries as a proactive defense
1744 against bounds-checking flaws in the kernel's
1745 copy_to_user()/copy_from_user() interface.
1746 on Perform hardened usercopy checks (default).
1747 off Disable hardened usercopy checks.
1748
1749 hardlockup_all_cpu_backtrace=
1750 [KNL] Should the hard-lockup detector generate
1751 backtraces on all cpus.
1752 Format: 0 | 1
1753
1754 hashdist= [KNL,NUMA] Large hashes allocated during boot
1755 are distributed across NUMA nodes. Defaults on
1756 for 64-bit NUMA, off otherwise.
1757 Format: 0 | 1 (for off | on)
1758
1759 hcl= [IA-64] SGI's Hardware Graph compatibility layer
1760
1761 hd= [EIDE] (E)IDE hard drive subsystem geometry
1762 Format: <cyl>,<head>,<sect>
1763
1764 hest_disable [ACPI]
1765 Disable Hardware Error Source Table (HEST) support;
1766 corresponding firmware-first mode error processing
1767 logic will be disabled.
1768
1769 hibernate= [HIBERNATION]
1770 noresume Don't check if there's a hibernation image
1771 present during boot.
1772 nocompress Don't compress/decompress hibernation images.
1773 no Disable hibernation and resume.
1774 protect_image Turn on image protection during restoration
1775 (that will set all pages holding image data
1776 during restoration read-only).
1777
1778 hibernate.compressor= [HIBERNATION] Compression algorithm to be
1779 used with hibernation.
1780 Format: { lzo | lz4 }
1781 Default: lzo
1782
1783 lzo: Select LZO compression algorithm to
1784 compress/decompress hibernation image.
1785
1786 lz4: Select LZ4 compression algorithm to
1787 compress/decompress hibernation image.
1788
1789 highmem=nn[KMG] [KNL,BOOT,EARLY] forces the highmem zone to have an exact
1790 size of <nn>. This works even on boxes that have no
1791 highmem otherwise. This also works to reduce highmem
1792 size on bigger boxes.
1793
1794 highres= [KNL] Enable/disable high resolution timer mode.
1795 Valid parameters: "on", "off"
1796 Default: "on"
1797
1798 hlt [BUGS=ARM,SH]
1799
1800 hostname= [KNL,EARLY] Set the hostname (aka UTS nodename).
1801 Format: <string>
1802 This allows setting the system's hostname during early
1803 startup. This sets the name returned by gethostname.
1804 Using this parameter to set the hostname makes it
1805 possible to ensure the hostname is correctly set before
1806 any userspace processes run, avoiding the possibility
1807 that a process may call gethostname before the hostname
1808 has been explicitly set, resulting in the calling
1809 process getting an incorrect result. The string must
1810 not exceed the maximum allowed hostname length (usually
1811 64 characters) and will be truncated otherwise.
1812
1813 hpet= [X86-32,HPET] option to control HPET usage
1814 Format: { enable (default) | disable | force |
1815 verbose }
1816 disable: disable HPET and use PIT instead
1817 force: allow force enabled of undocumented chips (ICH4,
1818 VIA, nVidia)
1819 verbose: show contents of HPET registers during setup
1820
1821 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET
1822 registers. Default set by CONFIG_HPET_MMAP_DEFAULT.
1823
1824 hugepages= [HW] Number of HugeTLB pages to allocate at boot.
1825 If this follows hugepagesz (below), it specifies
1826 the number of pages of hugepagesz to be allocated.
1827 If this is the first HugeTLB parameter on the command
1828 line, it specifies the number of pages to allocate for
1829 the default huge page size. If using node format, the
1830 number of pages to allocate per-node can be specified.
1831 See also Documentation/admin-guide/mm/hugetlbpage.rst.
1832 Format: <integer> or (node format)
1833 <node>:<integer>[,<node>:<integer>]
1834
1835 hugepagesz=
1836 [HW] The size of the HugeTLB pages. This is used in
1837 conjunction with hugepages (above) to allocate huge
1838 pages of a specific size at boot. The pair
1839 hugepagesz=X hugepages=Y can be specified once for
1840 each supported huge page size. Huge page sizes are
1841 architecture dependent. See also
1842 Documentation/admin-guide/mm/hugetlbpage.rst.
1843 Format: size[KMG]
1844
1845 hugetlb_cma= [HW,CMA,EARLY] The size of a CMA area used for allocation
1846 of gigantic hugepages. Or using node format, the size
1847 of a CMA area per node can be specified.
1848 Format: nn[KMGTPE] or (node format)
1849 <node>:nn[KMGTPE][,<node>:nn[KMGTPE]]
1850
1851 Reserve a CMA area of given size and allocate gigantic
1852 hugepages using the CMA allocator. If enabled, the
1853 boot-time allocation of gigantic hugepages is skipped.
1854
1855 hugetlb_free_vmemmap=
1856 [KNL] Requires CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP
1857 enabled.
1858 Control if HugeTLB Vmemmap Optimization (HVO) is enabled.
1859 Allows heavy hugetlb users to free up some more
1860 memory (7 * PAGE_SIZE for each 2MB hugetlb page).
1861 Format: { on | off (default) }
1862
1863 on: enable HVO
1864 off: disable HVO
1865
1866 Built with CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP_DEFAULT_ON=y,
1867 the default is on.
1868
1869 Note that the vmemmap pages may be allocated from the added
1870 memory block itself when memory_hotplug.memmap_on_memory is
1871 enabled, those vmemmap pages cannot be optimized even if this
1872 feature is enabled. Other vmemmap pages not allocated from
1873 the added memory block itself do not be affected.
1874
1875 hung_task_panic=
1876 [KNL] Should the hung task detector generate panics.
1877 Format: 0 | 1
1878
1879 A value of 1 instructs the kernel to panic when a
1880 hung task is detected. The default value is controlled
1881 by the CONFIG_BOOTPARAM_HUNG_TASK_PANIC build-time
1882 option. The value selected by this boot parameter can
1883 be changed later by the kernel.hung_task_panic sysctl.
1884
1885 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC)
1886 terminal devices. Valid values: 0..8
1887 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs.
1888 If specified, z/VM IUCV HVC accepts connections
1889 from listed z/VM user IDs only.
1890
1891 hv_nopvspin [X86,HYPER_V,EARLY]
1892 Disables the paravirt spinlock optimizations
1893 which allow the hypervisor to 'idle' the guest
1894 on lock contention.
1895
1896 i2c_bus= [HW] Override the default board specific I2C bus speed
1897 or register an additional I2C bus that is not
1898 registered from board initialization code.
1899 Format:
1900 <bus_id>,<clkrate>
1901
1902 i8042.debug [HW] Toggle i8042 debug mode
1903 i8042.unmask_kbd_data
1904 [HW] Enable printing of interrupt data from the KBD port
1905 (disabled by default, and as a pre-condition
1906 requires that i8042.debug=1 be enabled)
1907 i8042.direct [HW] Put keyboard port into non-translated mode
1908 i8042.dumbkbd [HW] Pretend that controller can only read data from
1909 keyboard and cannot control its state
1910 (Don't attempt to blink the leds)
1911 i8042.noaux [HW] Don't check for auxiliary (== mouse) port
1912 i8042.nokbd [HW] Don't check/create keyboard port
1913 i8042.noloop [HW] Disable the AUX Loopback command while probing
1914 for the AUX port
1915 i8042.nomux [HW] Don't check presence of an active multiplexing
1916 controller
1917 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX
1918 controllers
1919 i8042.notimeout [HW] Ignore timeout condition signalled by controller
1920 i8042.reset [HW] Reset the controller during init, cleanup and
1921 suspend-to-ram transitions, only during s2r
1922 transitions, or never reset
1923 Format: { 1 | Y | y | 0 | N | n }
1924 1, Y, y: always reset controller
1925 0, N, n: don't ever reset controller
1926 Default: only on s2r transitions on x86; most other
1927 architectures force reset to be always executed
1928 i8042.unlock [HW] Unlock (ignore) the keylock
1929 i8042.kbdreset [HW] Reset device connected to KBD port
1930 i8042.probe_defer
1931 [HW] Allow deferred probing upon i8042 probe errors
1932
1933 i810= [HW,DRM]
1934
1935 i915.invert_brightness=
1936 [DRM] Invert the sense of the variable that is used to
1937 set the brightness of the panel backlight. Normally a
1938 brightness value of 0 indicates backlight switched off,
1939 and the maximum of the brightness value sets the backlight
1940 to maximum brightness. If this parameter is set to 0
1941 (default) and the machine requires it, or this parameter
1942 is set to 1, a brightness value of 0 sets the backlight
1943 to maximum brightness, and the maximum of the brightness
1944 value switches the backlight off.
1945 -1 -- never invert brightness
1946 0 -- machine default
1947 1 -- force brightness inversion
1948
1949 ia32_emulation= [X86-64]
1950 Format: <bool>
1951 When true, allows loading 32-bit programs and executing 32-bit
1952 syscalls, essentially overriding IA32_EMULATION_DEFAULT_DISABLED at
1953 boot time. When false, unconditionally disables IA32 emulation.
1954
1955 icn= [HW,ISDN]
1956 Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]]
1957
1958
1959 idle= [X86,EARLY]
1960 Format: idle=poll, idle=halt, idle=nomwait
1961 Poll forces a polling idle loop that can slightly
1962 improve the performance of waking up a idle CPU, but
1963 will use a lot of power and make the system run hot.
1964 Not recommended.
1965 idle=halt: Halt is forced to be used for CPU idle.
1966 In such case C2/C3 won't be used again.
1967 idle=nomwait: Disable mwait for CPU C-states
1968
1969 idxd.sva= [HW]
1970 Format: <bool>
1971 Allow force disabling of Shared Virtual Memory (SVA)
1972 support for the idxd driver. By default it is set to
1973 true (1).
1974
1975 idxd.tc_override= [HW]
1976 Format: <bool>
1977 Allow override of default traffic class configuration
1978 for the device. By default it is set to false (0).
1979
1980 ieee754= [MIPS] Select IEEE Std 754 conformance mode
1981 Format: { strict | legacy | 2008 | relaxed }
1982 Default: strict
1983
1984 Choose which programs will be accepted for execution
1985 based on the IEEE 754 NaN encoding(s) supported by
1986 the FPU and the NaN encoding requested with the value
1987 of an ELF file header flag individually set by each
1988 binary. Hardware implementations are permitted to
1989 support either or both of the legacy and the 2008 NaN
1990 encoding mode.
1991
1992 Available settings are as follows:
1993 strict accept binaries that request a NaN encoding
1994 supported by the FPU
1995 legacy only accept legacy-NaN binaries, if supported
1996 by the FPU
1997 2008 only accept 2008-NaN binaries, if supported
1998 by the FPU
1999 relaxed accept any binaries regardless of whether
2000 supported by the FPU
2001
2002 The FPU emulator is always able to support both NaN
2003 encodings, so if no FPU hardware is present or it has
2004 been disabled with 'nofpu', then the settings of
2005 'legacy' and '2008' strap the emulator accordingly,
2006 'relaxed' straps the emulator for both legacy-NaN and
2007 2008-NaN, whereas 'strict' enables legacy-NaN only on
2008 legacy processors and both NaN encodings on MIPS32 or
2009 MIPS64 CPUs.
2010
2011 The setting for ABS.fmt/NEG.fmt instruction execution
2012 mode generally follows that for the NaN encoding,
2013 except where unsupported by hardware.
2014
2015 ignore_loglevel [KNL,EARLY]
2016 Ignore loglevel setting - this will print /all/
2017 kernel messages to the console. Useful for debugging.
2018 We also add it as printk module parameter, so users
2019 could change it dynamically, usually by
2020 /sys/module/printk/parameters/ignore_loglevel.
2021
2022 ignore_rlimit_data
2023 Ignore RLIMIT_DATA setting for data mappings,
2024 print warning at first misuse. Can be changed via
2025 /sys/module/kernel/parameters/ignore_rlimit_data.
2026
2027 ihash_entries= [KNL]
2028 Set number of hash buckets for inode cache.
2029
2030 ima_appraise= [IMA] appraise integrity measurements
2031 Format: { "off" | "enforce" | "fix" | "log" }
2032 default: "enforce"
2033
2034 ima_appraise_tcb [IMA] Deprecated. Use ima_policy= instead.
2035 The builtin appraise policy appraises all files
2036 owned by uid=0.
2037
2038 ima_canonical_fmt [IMA]
2039 Use the canonical format for the binary runtime
2040 measurements, instead of host native format.
2041
2042 ima_hash= [IMA]
2043 Format: { md5 | sha1 | rmd160 | sha256 | sha384
2044 | sha512 | ... }
2045 default: "sha1"
2046
2047 The list of supported hash algorithms is defined
2048 in crypto/hash_info.h.
2049
2050 ima_policy= [IMA]
2051 The builtin policies to load during IMA setup.
2052 Format: "tcb | appraise_tcb | secure_boot |
2053 fail_securely | critical_data"
2054
2055 The "tcb" policy measures all programs exec'd, files
2056 mmap'd for exec, and all files opened with the read
2057 mode bit set by either the effective uid (euid=0) or
2058 uid=0.
2059
2060 The "appraise_tcb" policy appraises the integrity of
2061 all files owned by root.
2062
2063 The "secure_boot" policy appraises the integrity
2064 of files (eg. kexec kernel image, kernel modules,
2065 firmware, policy, etc) based on file signatures.
2066
2067 The "fail_securely" policy forces file signature
2068 verification failure also on privileged mounted
2069 filesystems with the SB_I_UNVERIFIABLE_SIGNATURE
2070 flag.
2071
2072 The "critical_data" policy measures kernel integrity
2073 critical data.
2074
2075 ima_tcb [IMA] Deprecated. Use ima_policy= instead.
2076 Load a policy which meets the needs of the Trusted
2077 Computing Base. This means IMA will measure all
2078 programs exec'd, files mmap'd for exec, and all files
2079 opened for read by uid=0.
2080
2081 ima_template= [IMA]
2082 Select one of defined IMA measurements template formats.
2083 Formats: { "ima" | "ima-ng" | "ima-ngv2" | "ima-sig" |
2084 "ima-sigv2" }
2085 Default: "ima-ng"
2086
2087 ima_template_fmt=
2088 [IMA] Define a custom template format.
2089 Format: { "field1|...|fieldN" }
2090
2091 ima.ahash_minsize= [IMA] Minimum file size for asynchronous hash usage
2092 Format: <min_file_size>
2093 Set the minimal file size for using asynchronous hash.
2094 If left unspecified, ahash usage is disabled.
2095
2096 ahash performance varies for different data sizes on
2097 different crypto accelerators. This option can be used
2098 to achieve the best performance for a particular HW.
2099
2100 ima.ahash_bufsize= [IMA] Asynchronous hash buffer size
2101 Format: <bufsize>
2102 Set hashing buffer size. Default: 4k.
2103
2104 ahash performance varies for different chunk sizes on
2105 different crypto accelerators. This option can be used
2106 to achieve best performance for particular HW.
2107
2108 init= [KNL]
2109 Format: <full_path>
2110 Run specified binary instead of /sbin/init as init
2111 process.
2112
2113 initcall_debug [KNL] Trace initcalls as they are executed. Useful
2114 for working out where the kernel is dying during
2115 startup.
2116
2117 initcall_blacklist= [KNL] Do not execute a comma-separated list of
2118 initcall functions. Useful for debugging built-in
2119 modules and initcalls.
2120
2121 initramfs_async= [KNL]
2122 Format: <bool>
2123 Default: 1
2124 This parameter controls whether the initramfs
2125 image is unpacked asynchronously, concurrently
2126 with devices being probed and
2127 initialized. This should normally just work,
2128 but as a debugging aid, one can get the
2129 historical behaviour of the initramfs
2130 unpacking being completed before device_ and
2131 late_ initcalls.
2132
2133 initrd= [BOOT,EARLY] Specify the location of the initial ramdisk
2134
2135 initrdmem= [KNL,EARLY] Specify a physical address and size from which to
2136 load the initrd. If an initrd is compiled in or
2137 specified in the bootparams, it takes priority over this
2138 setting.
2139 Format: ss[KMG],nn[KMG]
2140 Default is 0, 0
2141
2142 init_on_alloc= [MM,EARLY] Fill newly allocated pages and heap objects with
2143 zeroes.
2144 Format: 0 | 1
2145 Default set by CONFIG_INIT_ON_ALLOC_DEFAULT_ON.
2146
2147 init_on_free= [MM,EARLY] Fill freed pages and heap objects with zeroes.
2148 Format: 0 | 1
2149 Default set by CONFIG_INIT_ON_FREE_DEFAULT_ON.
2150
2151 init_pkru= [X86] Specify the default memory protection keys rights
2152 register contents for all processes. 0x55555554 by
2153 default (disallow access to all but pkey 0). Can
2154 override in debugfs after boot.
2155
2156 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver
2157 Format: <irq>
2158
2159 int_pln_enable [X86] Enable power limit notification interrupt
2160
2161 integrity_audit=[IMA]
2162 Format: { "0" | "1" }
2163 0 -- basic integrity auditing messages. (Default)
2164 1 -- additional integrity auditing messages.
2165
2166 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option
2167 on
2168 Enable intel iommu driver.
2169 off
2170 Disable intel iommu driver.
2171 igfx_off [Default Off]
2172 By default, gfx is mapped as normal device. If a gfx
2173 device has a dedicated DMAR unit, the DMAR unit is
2174 bypassed by not enabling DMAR with this option. In
2175 this case, gfx device will use physical address for
2176 DMA.
2177 strict [Default Off]
2178 Deprecated, equivalent to iommu.strict=1.
2179 sp_off [Default Off]
2180 By default, super page will be supported if Intel IOMMU
2181 has the capability. With this option, super page will
2182 not be supported.
2183 sm_on
2184 Enable the Intel IOMMU scalable mode if the hardware
2185 advertises that it has support for the scalable mode
2186 translation.
2187 sm_off
2188 Disallow use of the Intel IOMMU scalable mode.
2189 tboot_noforce [Default Off]
2190 Do not force the Intel IOMMU enabled under tboot.
2191 By default, tboot will force Intel IOMMU on, which
2192 could harm performance of some high-throughput
2193 devices like 40GBit network cards, even if identity
2194 mapping is enabled.
2195 Note that using this option lowers the security
2196 provided by tboot because it makes the system
2197 vulnerable to DMA attacks.
2198
2199 intel_idle.max_cstate= [KNL,HW,ACPI,X86]
2200 0 disables intel_idle and fall back on acpi_idle.
2201 1 to 9 specify maximum depth of C-state.
2202
2203 intel_pstate= [X86,EARLY]
2204 disable
2205 Do not enable intel_pstate as the default
2206 scaling driver for the supported processors
2207 active
2208 Use intel_pstate driver to bypass the scaling
2209 governors layer of cpufreq and provides it own
2210 algorithms for p-state selection. There are two
2211 P-state selection algorithms provided by
2212 intel_pstate in the active mode: powersave and
2213 performance. The way they both operate depends
2214 on whether or not the hardware managed P-states
2215 (HWP) feature has been enabled in the processor
2216 and possibly on the processor model.
2217 passive
2218 Use intel_pstate as a scaling driver, but configure it
2219 to work with generic cpufreq governors (instead of
2220 enabling its internal governor). This mode cannot be
2221 used along with the hardware-managed P-states (HWP)
2222 feature.
2223 force
2224 Enable intel_pstate on systems that prohibit it by default
2225 in favor of acpi-cpufreq. Forcing the intel_pstate driver
2226 instead of acpi-cpufreq may disable platform features, such
2227 as thermal controls and power capping, that rely on ACPI
2228 P-States information being indicated to OSPM and therefore
2229 should be used with caution. This option does not work with
2230 processors that aren't supported by the intel_pstate driver
2231 or on platforms that use pcc-cpufreq instead of acpi-cpufreq.
2232 no_hwp
2233 Do not enable hardware P state control (HWP)
2234 if available.
2235 hwp_only
2236 Only load intel_pstate on systems which support
2237 hardware P state control (HWP) if available.
2238 support_acpi_ppc
2239 Enforce ACPI _PPC performance limits. If the Fixed ACPI
2240 Description Table, specifies preferred power management
2241 profile as "Enterprise Server" or "Performance Server",
2242 then this feature is turned on by default.
2243 per_cpu_perf_limits
2244 Allow per-logical-CPU P-State performance control limits using
2245 cpufreq sysfs interface
2246
2247 intremap= [X86-64,Intel-IOMMU,EARLY]
2248 on enable Interrupt Remapping (default)
2249 off disable Interrupt Remapping
2250 nosid disable Source ID checking
2251 no_x2apic_optout
2252 BIOS x2APIC opt-out request will be ignored
2253 nopost disable Interrupt Posting
2254
2255 iomem= Disable strict checking of access to MMIO memory
2256 strict regions from userspace.
2257 relaxed
2258
2259 iommu= [X86,EARLY]
2260 off
2261 force
2262 noforce
2263 biomerge
2264 panic
2265 nopanic
2266 merge
2267 nomerge
2268 soft
2269 pt [X86]
2270 nopt [X86]
2271 nobypass [PPC/POWERNV]
2272 Disable IOMMU bypass, using IOMMU for PCI devices.
2273
2274 iommu.forcedac= [ARM64,X86,EARLY] Control IOVA allocation for PCI devices.
2275 Format: { "0" | "1" }
2276 0 - Try to allocate a 32-bit DMA address first, before
2277 falling back to the full range if needed.
2278 1 - Allocate directly from the full usable range,
2279 forcing Dual Address Cycle for PCI cards supporting
2280 greater than 32-bit addressing.
2281
2282 iommu.strict= [ARM64,X86,S390,EARLY] Configure TLB invalidation behaviour
2283 Format: { "0" | "1" }
2284 0 - Lazy mode.
2285 Request that DMA unmap operations use deferred
2286 invalidation of hardware TLBs, for increased
2287 throughput at the cost of reduced device isolation.
2288 Will fall back to strict mode if not supported by
2289 the relevant IOMMU driver.
2290 1 - Strict mode.
2291 DMA unmap operations invalidate IOMMU hardware TLBs
2292 synchronously.
2293 unset - Use value of CONFIG_IOMMU_DEFAULT_DMA_{LAZY,STRICT}.
2294 Note: on x86, strict mode specified via one of the
2295 legacy driver-specific options takes precedence.
2296
2297 iommu.passthrough=
2298 [ARM64,X86,EARLY] Configure DMA to bypass the IOMMU by default.
2299 Format: { "0" | "1" }
2300 0 - Use IOMMU translation for DMA.
2301 1 - Bypass the IOMMU for DMA.
2302 unset - Use value of CONFIG_IOMMU_DEFAULT_PASSTHROUGH.
2303
2304 io7= [HW] IO7 for Marvel-based Alpha systems
2305 See comment before marvel_specify_io7 in
2306 arch/alpha/kernel/core_marvel.c.
2307
2308 io_delay= [X86,EARLY] I/O delay method
2309 0x80
2310 Standard port 0x80 based delay
2311 0xed
2312 Alternate port 0xed based delay (needed on some systems)
2313 udelay
2314 Simple two microseconds delay
2315 none
2316 No delay
2317
2318 ip= [IP_PNP]
2319 See Documentation/admin-guide/nfs/nfsroot.rst.
2320
2321 ipcmni_extend [KNL,EARLY] Extend the maximum number of unique System V
2322 IPC identifiers from 32,768 to 16,777,216.
2323
2324 irqaffinity= [SMP] Set the default irq affinity mask
2325 The argument is a cpu list, as described above.
2326
2327 irqchip.gicv2_force_probe=
2328 [ARM,ARM64,EARLY]
2329 Format: <bool>
2330 Force the kernel to look for the second 4kB page
2331 of a GICv2 controller even if the memory range
2332 exposed by the device tree is too small.
2333
2334 irqchip.gicv3_nolpi=
2335 [ARM,ARM64,EARLY]
2336 Force the kernel to ignore the availability of
2337 LPIs (and by consequence ITSs). Intended for system
2338 that use the kernel as a bootloader, and thus want
2339 to let secondary kernels in charge of setting up
2340 LPIs.
2341
2342 irqchip.gicv3_pseudo_nmi= [ARM64,EARLY]
2343 Enables support for pseudo-NMIs in the kernel. This
2344 requires the kernel to be built with
2345 CONFIG_ARM64_PSEUDO_NMI.
2346
2347 irqfixup [HW]
2348 When an interrupt is not handled search all handlers
2349 for it. Intended to get systems with badly broken
2350 firmware running.
2351
2352 irqpoll [HW]
2353 When an interrupt is not handled search all handlers
2354 for it. Also check all handlers each timer
2355 interrupt. Intended to get systems with badly broken
2356 firmware running.
2357
2358 isapnp= [ISAPNP]
2359 Format: <RDP>,<reset>,<pci_scan>,<verbosity>
2360
2361 isolcpus= [KNL,SMP,ISOL] Isolate a given set of CPUs from disturbance.
2362 [Deprecated - use cpusets instead]
2363 Format: [flag-list,]<cpu-list>
2364
2365 Specify one or more CPUs to isolate from disturbances
2366 specified in the flag list (default: domain):
2367
2368 nohz
2369 Disable the tick when a single task runs.
2370
2371 A residual 1Hz tick is offloaded to workqueues, which you
2372 need to affine to housekeeping through the global
2373 workqueue's affinity configured via the
2374 /sys/devices/virtual/workqueue/cpumask sysfs file, or
2375 by using the 'domain' flag described below.
2376
2377 NOTE: by default the global workqueue runs on all CPUs,
2378 so to protect individual CPUs the 'cpumask' file has to
2379 be configured manually after bootup.
2380
2381 domain
2382 Isolate from the general SMP balancing and scheduling
2383 algorithms. Note that performing domain isolation this way
2384 is irreversible: it's not possible to bring back a CPU to
2385 the domains once isolated through isolcpus. It's strongly
2386 advised to use cpusets instead to disable scheduler load
2387 balancing through the "cpuset.sched_load_balance" file.
2388 It offers a much more flexible interface where CPUs can
2389 move in and out of an isolated set anytime.
2390
2391 You can move a process onto or off an "isolated" CPU via
2392 the CPU affinity syscalls or cpuset.
2393 <cpu number> begins at 0 and the maximum value is
2394 "number of CPUs in system - 1".
2395
2396 managed_irq
2397
2398 Isolate from being targeted by managed interrupts
2399 which have an interrupt mask containing isolated
2400 CPUs. The affinity of managed interrupts is
2401 handled by the kernel and cannot be changed via
2402 the /proc/irq/* interfaces.
2403
2404 This isolation is best effort and only effective
2405 if the automatically assigned interrupt mask of a
2406 device queue contains isolated and housekeeping
2407 CPUs. If housekeeping CPUs are online then such
2408 interrupts are directed to the housekeeping CPU
2409 so that IO submitted on the housekeeping CPU
2410 cannot disturb the isolated CPU.
2411
2412 If a queue's affinity mask contains only isolated
2413 CPUs then this parameter has no effect on the
2414 interrupt routing decision, though interrupts are
2415 only delivered when tasks running on those
2416 isolated CPUs submit IO. IO submitted on
2417 housekeeping CPUs has no influence on those
2418 queues.
2419
2420 The format of <cpu-list> is described above.
2421
2422 iucv= [HW,NET]
2423
2424 ivrs_ioapic [HW,X86-64]
2425 Provide an override to the IOAPIC-ID<->DEVICE-ID
2426 mapping provided in the IVRS ACPI table.
2427 By default, PCI segment is 0, and can be omitted.
2428
2429 For example, to map IOAPIC-ID decimal 10 to
2430 PCI segment 0x1 and PCI device 00:14.0,
2431 write the parameter as:
2432 ivrs_ioapic=10@0001:00:14.0
2433
2434 Deprecated formats:
2435 * To map IOAPIC-ID decimal 10 to PCI device 00:14.0
2436 write the parameter as:
2437 ivrs_ioapic[10]=00:14.0
2438 * To map IOAPIC-ID decimal 10 to PCI segment 0x1 and
2439 PCI device 00:14.0 write the parameter as:
2440 ivrs_ioapic[10]=0001:00:14.0
2441
2442 ivrs_hpet [HW,X86-64]
2443 Provide an override to the HPET-ID<->DEVICE-ID
2444 mapping provided in the IVRS ACPI table.
2445 By default, PCI segment is 0, and can be omitted.
2446
2447 For example, to map HPET-ID decimal 10 to
2448 PCI segment 0x1 and PCI device 00:14.0,
2449 write the parameter as:
2450 ivrs_hpet=10@0001:00:14.0
2451
2452 Deprecated formats:
2453 * To map HPET-ID decimal 0 to PCI device 00:14.0
2454 write the parameter as:
2455 ivrs_hpet[0]=00:14.0
2456 * To map HPET-ID decimal 10 to PCI segment 0x1 and
2457 PCI device 00:14.0 write the parameter as:
2458 ivrs_ioapic[10]=0001:00:14.0
2459
2460 ivrs_acpihid [HW,X86-64]
2461 Provide an override to the ACPI-HID:UID<->DEVICE-ID
2462 mapping provided in the IVRS ACPI table.
2463 By default, PCI segment is 0, and can be omitted.
2464
2465 For example, to map UART-HID:UID AMD0020:0 to
2466 PCI segment 0x1 and PCI device ID 00:14.5,
2467 write the parameter as:
2468 ivrs_acpihid=AMD0020:0@0001:00:14.5
2469
2470 Deprecated formats:
2471 * To map UART-HID:UID AMD0020:0 to PCI segment is 0,
2472 PCI device ID 00:14.5, write the parameter as:
2473 ivrs_acpihid[00:14.5]=AMD0020:0
2474 * To map UART-HID:UID AMD0020:0 to PCI segment 0x1 and
2475 PCI device ID 00:14.5, write the parameter as:
2476 ivrs_acpihid[0001:00:14.5]=AMD0020:0
2477
2478 js= [HW,JOY] Analog joystick
2479 See Documentation/input/joydev/joystick.rst.
2480
2481 kasan_multi_shot
2482 [KNL] Enforce KASAN (Kernel Address Sanitizer) to print
2483 report on every invalid memory access. Without this
2484 parameter KASAN will print report only for the first
2485 invalid access.
2486
2487 keep_bootcon [KNL,EARLY]
2488 Do not unregister boot console at start. This is only
2489 useful for debugging when something happens in the window
2490 between unregistering the boot console and initializing
2491 the real console.
2492
2493 keepinitrd [HW,ARM] See retain_initrd.
2494
2495 kernelcore= [KNL,X86,IA-64,PPC,EARLY]
2496 Format: nn[KMGTPE] | nn% | "mirror"
2497 This parameter specifies the amount of memory usable by
2498 the kernel for non-movable allocations. The requested
2499 amount is spread evenly throughout all nodes in the
2500 system as ZONE_NORMAL. The remaining memory is used for
2501 movable memory in its own zone, ZONE_MOVABLE. In the
2502 event, a node is too small to have both ZONE_NORMAL and
2503 ZONE_MOVABLE, kernelcore memory will take priority and
2504 other nodes will have a larger ZONE_MOVABLE.
2505
2506 ZONE_MOVABLE is used for the allocation of pages that
2507 may be reclaimed or moved by the page migration
2508 subsystem. Note that allocations like PTEs-from-HighMem
2509 still use the HighMem zone if it exists, and the Normal
2510 zone if it does not.
2511
2512 It is possible to specify the exact amount of memory in
2513 the form of "nn[KMGTPE]", a percentage of total system
2514 memory in the form of "nn%", or "mirror". If "mirror"
2515 option is specified, mirrored (reliable) memory is used
2516 for non-movable allocations and remaining memory is used
2517 for Movable pages. "nn[KMGTPE]", "nn%", and "mirror"
2518 are exclusive, so you cannot specify multiple forms.
2519
2520 kgdbdbgp= [KGDB,HW,EARLY] kgdb over EHCI usb debug port.
2521 Format: <Controller#>[,poll interval]
2522 The controller # is the number of the ehci usb debug
2523 port as it is probed via PCI. The poll interval is
2524 optional and is the number seconds in between
2525 each poll cycle to the debug port in case you need
2526 the functionality for interrupting the kernel with
2527 gdb or control-c on the dbgp connection. When
2528 not using this parameter you use sysrq-g to break into
2529 the kernel debugger.
2530
2531 kgdboc= [KGDB,HW] kgdb over consoles.
2532 Requires a tty driver that supports console polling,
2533 or a supported polling keyboard driver (non-usb).
2534 Serial only format: <serial_device>[,baud]
2535 keyboard only format: kbd
2536 keyboard and serial format: kbd,<serial_device>[,baud]
2537 Optional Kernel mode setting:
2538 kms, kbd format: kms,kbd
2539 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud]
2540
2541 kgdboc_earlycon= [KGDB,HW,EARLY]
2542 If the boot console provides the ability to read
2543 characters and can work in polling mode, you can use
2544 this parameter to tell kgdb to use it as a backend
2545 until the normal console is registered. Intended to
2546 be used together with the kgdboc parameter which
2547 specifies the normal console to transition to.
2548
2549 The name of the early console should be specified
2550 as the value of this parameter. Note that the name of
2551 the early console might be different than the tty
2552 name passed to kgdboc. It's OK to leave the value
2553 blank and the first boot console that implements
2554 read() will be picked.
2555
2556 kgdbwait [KGDB,EARLY] Stop kernel execution and enter the
2557 kernel debugger at the earliest opportunity.
2558
2559 kmac= [MIPS] Korina ethernet MAC address.
2560 Configure the RouterBoard 532 series on-chip
2561 Ethernet adapter MAC address.
2562
2563 kmemleak= [KNL,EARLY] Boot-time kmemleak enable/disable
2564 Valid arguments: on, off
2565 Default: on
2566 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y,
2567 the default is off.
2568
2569 kprobe_event=[probe-list]
2570 [FTRACE] Add kprobe events and enable at boot time.
2571 The probe-list is a semicolon delimited list of probe
2572 definitions. Each definition is same as kprobe_events
2573 interface, but the parameters are comma delimited.
2574 For example, to add a kprobe event on vfs_read with
2575 arg1 and arg2, add to the command line;
2576
2577 kprobe_event=p,vfs_read,$arg1,$arg2
2578
2579 See also Documentation/trace/kprobetrace.rst "Kernel
2580 Boot Parameter" section.
2581
2582 kpti= [ARM64,EARLY] Control page table isolation of
2583 user and kernel address spaces.
2584 Default: enabled on cores which need mitigation.
2585 0: force disabled
2586 1: force enabled
2587
2588 kunit.enable= [KUNIT] Enable executing KUnit tests. Requires
2589 CONFIG_KUNIT to be set to be fully enabled. The
2590 default value can be overridden via
2591 KUNIT_DEFAULT_ENABLED.
2592 Default is 1 (enabled)
2593
2594 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs.
2595 Default is 0 (don't ignore, but inject #GP)
2596
2597 kvm.eager_page_split=
2598 [KVM,X86] Controls whether or not KVM will try to
2599 proactively split all huge pages during dirty logging.
2600 Eager page splitting reduces interruptions to vCPU
2601 execution by eliminating the write-protection faults
2602 and MMU lock contention that would otherwise be
2603 required to split huge pages lazily.
2604
2605 VM workloads that rarely perform writes or that write
2606 only to a small region of VM memory may benefit from
2607 disabling eager page splitting to allow huge pages to
2608 still be used for reads.
2609
2610 The behavior of eager page splitting depends on whether
2611 KVM_DIRTY_LOG_INITIALLY_SET is enabled or disabled. If
2612 disabled, all huge pages in a memslot will be eagerly
2613 split when dirty logging is enabled on that memslot. If
2614 enabled, eager page splitting will be performed during
2615 the KVM_CLEAR_DIRTY ioctl, and only for the pages being
2616 cleared.
2617
2618 Eager page splitting is only supported when kvm.tdp_mmu=Y.
2619
2620 Default is Y (on).
2621
2622 kvm.enable_vmware_backdoor=[KVM] Support VMware backdoor PV interface.
2623 Default is false (don't support).
2624
2625 kvm.nx_huge_pages=
2626 [KVM] Controls the software workaround for the
2627 X86_BUG_ITLB_MULTIHIT bug.
2628 force : Always deploy workaround.
2629 off : Never deploy workaround.
2630 auto : Deploy workaround based on the presence of
2631 X86_BUG_ITLB_MULTIHIT.
2632
2633 Default is 'auto'.
2634
2635 If the software workaround is enabled for the host,
2636 guests do need not to enable it for nested guests.
2637
2638 kvm.nx_huge_pages_recovery_ratio=
2639 [KVM] Controls how many 4KiB pages are periodically zapped
2640 back to huge pages. 0 disables the recovery, otherwise if
2641 the value is N KVM will zap 1/Nth of the 4KiB pages every
2642 period (see below). The default is 60.
2643
2644 kvm.nx_huge_pages_recovery_period_ms=
2645 [KVM] Controls the time period at which KVM zaps 4KiB pages
2646 back to huge pages. If the value is a non-zero N, KVM will
2647 zap a portion (see ratio above) of the pages every N msecs.
2648 If the value is 0 (the default), KVM will pick a period based
2649 on the ratio, such that a page is zapped after 1 hour on average.
2650
2651 kvm-amd.nested= [KVM,AMD] Control nested virtualization feature in
2652 KVM/SVM. Default is 1 (enabled).
2653
2654 kvm-amd.npt= [KVM,AMD] Control KVM's use of Nested Page Tables,
2655 a.k.a. Two-Dimensional Page Tables. Default is 1
2656 (enabled). Disable by KVM if hardware lacks support
2657 for NPT.
2658
2659 kvm-arm.mode=
2660 [KVM,ARM,EARLY] Select one of KVM/arm64's modes of
2661 operation.
2662
2663 none: Forcefully disable KVM.
2664
2665 nvhe: Standard nVHE-based mode, without support for
2666 protected guests.
2667
2668 protected: nVHE-based mode with support for guests whose
2669 state is kept private from the host.
2670
2671 nested: VHE-based mode with support for nested
2672 virtualization. Requires at least ARMv8.3
2673 hardware.
2674
2675 Defaults to VHE/nVHE based on hardware support. Setting
2676 mode to "protected" will disable kexec and hibernation
2677 for the host. "nested" is experimental and should be
2678 used with extreme caution.
2679
2680 kvm-arm.vgic_v3_group0_trap=
2681 [KVM,ARM,EARLY] Trap guest accesses to GICv3 group-0
2682 system registers
2683
2684 kvm-arm.vgic_v3_group1_trap=
2685 [KVM,ARM,EARLY] Trap guest accesses to GICv3 group-1
2686 system registers
2687
2688 kvm-arm.vgic_v3_common_trap=
2689 [KVM,ARM,EARLY] Trap guest accesses to GICv3 common
2690 system registers
2691
2692 kvm-arm.vgic_v4_enable=
2693 [KVM,ARM,EARLY] Allow use of GICv4 for direct
2694 injection of LPIs.
2695
2696 kvm_cma_resv_ratio=n [PPC,EARLY]
2697 Reserves given percentage from system memory area for
2698 contiguous memory allocation for KVM hash pagetable
2699 allocation.
2700 By default it reserves 5% of total system memory.
2701 Format: <integer>
2702 Default: 5
2703
2704 kvm-intel.ept= [KVM,Intel] Control KVM's use of Extended Page Tables,
2705 a.k.a. Two-Dimensional Page Tables. Default is 1
2706 (enabled). Disable by KVM if hardware lacks support
2707 for EPT.
2708
2709 kvm-intel.emulate_invalid_guest_state=
2710 [KVM,Intel] Control whether to emulate invalid guest
2711 state. Ignored if kvm-intel.enable_unrestricted_guest=1,
2712 as guest state is never invalid for unrestricted
2713 guests. This param doesn't apply to nested guests (L2),
2714 as KVM never emulates invalid L2 guest state.
2715 Default is 1 (enabled).
2716
2717 kvm-intel.flexpriority=
2718 [KVM,Intel] Control KVM's use of FlexPriority feature
2719 (TPR shadow). Default is 1 (enabled). Disable by KVM if
2720 hardware lacks support for it.
2721
2722 kvm-intel.nested=
2723 [KVM,Intel] Control nested virtualization feature in
2724 KVM/VMX. Default is 1 (enabled).
2725
2726 kvm-intel.unrestricted_guest=
2727 [KVM,Intel] Control KVM's use of unrestricted guest
2728 feature (virtualized real and unpaged mode). Default
2729 is 1 (enabled). Disable by KVM if EPT is disabled or
2730 hardware lacks support for it.
2731
2732 kvm-intel.vmentry_l1d_flush=[KVM,Intel] Mitigation for L1 Terminal Fault
2733 CVE-2018-3620.
2734
2735 Valid arguments: never, cond, always
2736
2737 always: L1D cache flush on every VMENTER.
2738 cond: Flush L1D on VMENTER only when the code between
2739 VMEXIT and VMENTER can leak host memory.
2740 never: Disables the mitigation
2741
2742 Default is cond (do L1 cache flush in specific instances)
2743
2744 kvm-intel.vpid= [KVM,Intel] Control KVM's use of Virtual Processor
2745 Identification feature (tagged TLBs). Default is 1
2746 (enabled). Disable by KVM if hardware lacks support
2747 for it.
2748
2749 l1d_flush= [X86,INTEL,EARLY]
2750 Control mitigation for L1D based snooping vulnerability.
2751
2752 Certain CPUs are vulnerable to an exploit against CPU
2753 internal buffers which can forward information to a
2754 disclosure gadget under certain conditions.
2755
2756 In vulnerable processors, the speculatively
2757 forwarded data can be used in a cache side channel
2758 attack, to access data to which the attacker does
2759 not have direct access.
2760
2761 This parameter controls the mitigation. The
2762 options are:
2763
2764 on - enable the interface for the mitigation
2765
2766 l1tf= [X86,EARLY] Control mitigation of the L1TF vulnerability on
2767 affected CPUs
2768
2769 The kernel PTE inversion protection is unconditionally
2770 enabled and cannot be disabled.
2771
2772 full
2773 Provides all available mitigations for the
2774 L1TF vulnerability. Disables SMT and
2775 enables all mitigations in the
2776 hypervisors, i.e. unconditional L1D flush.
2777
2778 SMT control and L1D flush control via the
2779 sysfs interface is still possible after
2780 boot. Hypervisors will issue a warning
2781 when the first VM is started in a
2782 potentially insecure configuration,
2783 i.e. SMT enabled or L1D flush disabled.
2784
2785 full,force
2786 Same as 'full', but disables SMT and L1D
2787 flush runtime control. Implies the
2788 'nosmt=force' command line option.
2789 (i.e. sysfs control of SMT is disabled.)
2790
2791 flush
2792 Leaves SMT enabled and enables the default
2793 hypervisor mitigation, i.e. conditional
2794 L1D flush.
2795
2796 SMT control and L1D flush control via the
2797 sysfs interface is still possible after
2798 boot. Hypervisors will issue a warning
2799 when the first VM is started in a
2800 potentially insecure configuration,
2801 i.e. SMT enabled or L1D flush disabled.
2802
2803 flush,nosmt
2804
2805 Disables SMT and enables the default
2806 hypervisor mitigation.
2807
2808 SMT control and L1D flush control via the
2809 sysfs interface is still possible after
2810 boot. Hypervisors will issue a warning
2811 when the first VM is started in a
2812 potentially insecure configuration,
2813 i.e. SMT enabled or L1D flush disabled.
2814
2815 flush,nowarn
2816 Same as 'flush', but hypervisors will not
2817 warn when a VM is started in a potentially
2818 insecure configuration.
2819
2820 off
2821 Disables hypervisor mitigations and doesn't
2822 emit any warnings.
2823 It also drops the swap size and available
2824 RAM limit restriction on both hypervisor and
2825 bare metal.
2826
2827 Default is 'flush'.
2828
2829 For details see: Documentation/admin-guide/hw-vuln/l1tf.rst
2830
2831 l2cr= [PPC]
2832
2833 l3cr= [PPC]
2834
2835 lapic [X86-32,APIC,EARLY] Enable the local APIC even if BIOS
2836 disabled it.
2837
2838 lapic= [X86,APIC] Do not use TSC deadline
2839 value for LAPIC timer one-shot implementation. Default
2840 back to the programmable timer unit in the LAPIC.
2841 Format: notscdeadline
2842
2843 lapic_timer_c2_ok [X86,APIC,EARLY] trust the local apic timer
2844 in C2 power state.
2845
2846 libata.dma= [LIBATA] DMA control
2847 libata.dma=0 Disable all PATA and SATA DMA
2848 libata.dma=1 PATA and SATA Disk DMA only
2849 libata.dma=2 ATAPI (CDROM) DMA only
2850 libata.dma=4 Compact Flash DMA only
2851 Combinations also work, so libata.dma=3 enables DMA
2852 for disks and CDROMs, but not CFs.
2853
2854 libata.ignore_hpa= [LIBATA] Ignore HPA limit
2855 libata.ignore_hpa=0 keep BIOS limits (default)
2856 libata.ignore_hpa=1 ignore limits, using full disk
2857
2858 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume
2859 when set.
2860 Format: <int>
2861
2862 libata.force= [LIBATA] Force configurations. The format is a comma-
2863 separated list of "[ID:]VAL" where ID is PORT[.DEVICE].
2864 PORT and DEVICE are decimal numbers matching port, link
2865 or device. Basically, it matches the ATA ID string
2866 printed on console by libata. If the whole ID part is
2867 omitted, the last PORT and DEVICE values are used. If
2868 ID hasn't been specified yet, the configuration applies
2869 to all ports, links and devices.
2870
2871 If only DEVICE is omitted, the parameter applies to
2872 the port and all links and devices behind it. DEVICE
2873 number of 0 either selects the first device or the
2874 first fan-out link behind PMP device. It does not
2875 select the host link. DEVICE number of 15 selects the
2876 host link and device attached to it.
2877
2878 The VAL specifies the configuration to force. As long
2879 as there is no ambiguity, shortcut notation is allowed.
2880 For example, both 1.5 and 1.5G would work for 1.5Gbps.
2881 The following configurations can be forced.
2882
2883 * Cable type: 40c, 80c, short40c, unk, ign or sata.
2884 Any ID with matching PORT is used.
2885
2886 * SATA link speed limit: 1.5Gbps or 3.0Gbps.
2887
2888 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7].
2889 udma[/][16,25,33,44,66,100,133] notation is also
2890 allowed.
2891
2892 * nohrst, nosrst, norst: suppress hard, soft and both
2893 resets.
2894
2895 * rstonce: only attempt one reset during hot-unplug
2896 link recovery.
2897
2898 * [no]dbdelay: Enable or disable the extra 200ms delay
2899 before debouncing a link PHY and device presence
2900 detection.
2901
2902 * [no]ncq: Turn on or off NCQ.
2903
2904 * [no]ncqtrim: Enable or disable queued DSM TRIM.
2905
2906 * [no]ncqati: Enable or disable NCQ trim on ATI chipset.
2907
2908 * [no]trim: Enable or disable (unqueued) TRIM.
2909
2910 * trim_zero: Indicate that TRIM command zeroes data.
2911
2912 * max_trim_128m: Set 128M maximum trim size limit.
2913
2914 * [no]dma: Turn on or off DMA transfers.
2915
2916 * atapi_dmadir: Enable ATAPI DMADIR bridge support.
2917
2918 * atapi_mod16_dma: Enable the use of ATAPI DMA for
2919 commands that are not a multiple of 16 bytes.
2920
2921 * [no]dmalog: Enable or disable the use of the
2922 READ LOG DMA EXT command to access logs.
2923
2924 * [no]iddevlog: Enable or disable access to the
2925 identify device data log.
2926
2927 * [no]logdir: Enable or disable access to the general
2928 purpose log directory.
2929
2930 * max_sec_128: Set transfer size limit to 128 sectors.
2931
2932 * max_sec_1024: Set or clear transfer size limit to
2933 1024 sectors.
2934
2935 * max_sec_lba48: Set or clear transfer size limit to
2936 65535 sectors.
2937
2938 * [no]lpm: Enable or disable link power management.
2939
2940 * [no]setxfer: Indicate if transfer speed mode setting
2941 should be skipped.
2942
2943 * [no]fua: Disable or enable FUA (Force Unit Access)
2944 support for devices supporting this feature.
2945
2946 * dump_id: Dump IDENTIFY data.
2947
2948 * disable: Disable this device.
2949
2950 If there are multiple matching configurations changing
2951 the same attribute, the last one is used.
2952
2953 load_ramdisk= [RAM] [Deprecated]
2954
2955 lockd.nlm_grace_period=P [NFS] Assign grace period.
2956 Format: <integer>
2957
2958 lockd.nlm_tcpport=N [NFS] Assign TCP port.
2959 Format: <integer>
2960
2961 lockd.nlm_timeout=T [NFS] Assign timeout value.
2962 Format: <integer>
2963
2964 lockd.nlm_udpport=M [NFS] Assign UDP port.
2965 Format: <integer>
2966
2967 lockdown= [SECURITY,EARLY]
2968 { integrity | confidentiality }
2969 Enable the kernel lockdown feature. If set to
2970 integrity, kernel features that allow userland to
2971 modify the running kernel are disabled. If set to
2972 confidentiality, kernel features that allow userland
2973 to extract confidential information from the kernel
2974 are also disabled.
2975
2976 locktorture.acq_writer_lim= [KNL]
2977 Set the time limit in jiffies for a lock
2978 acquisition. Acquisitions exceeding this limit
2979 will result in a splat once they do complete.
2980
2981 locktorture.bind_readers= [KNL]
2982 Specify the list of CPUs to which the readers are
2983 to be bound.
2984
2985 locktorture.bind_writers= [KNL]
2986 Specify the list of CPUs to which the writers are
2987 to be bound.
2988
2989 locktorture.call_rcu_chains= [KNL]
2990 Specify the number of self-propagating call_rcu()
2991 chains to set up. These are used to ensure that
2992 there is a high probability of an RCU grace period
2993 in progress at any given time. Defaults to 0,
2994 which disables these call_rcu() chains.
2995
2996 locktorture.long_hold= [KNL]
2997 Specify the duration in milliseconds for the
2998 occasional long-duration lock hold time. Defaults
2999 to 100 milliseconds. Select 0 to disable.
3000
3001 locktorture.nested_locks= [KNL]
3002 Specify the maximum lock nesting depth that
3003 locktorture is to exercise, up to a limit of 8
3004 (MAX_NESTED_LOCKS). Specify zero to disable.
3005 Note that this parameter is ineffective on types
3006 of locks that do not support nested acquisition.
3007
3008 locktorture.nreaders_stress= [KNL]
3009 Set the number of locking read-acquisition kthreads.
3010 Defaults to being automatically set based on the
3011 number of online CPUs.
3012
3013 locktorture.nwriters_stress= [KNL]
3014 Set the number of locking write-acquisition kthreads.
3015
3016 locktorture.onoff_holdoff= [KNL]
3017 Set time (s) after boot for CPU-hotplug testing.
3018
3019 locktorture.onoff_interval= [KNL]
3020 Set time (s) between CPU-hotplug operations, or
3021 zero to disable CPU-hotplug testing.
3022
3023 locktorture.rt_boost= [KNL]
3024 Do periodic testing of real-time lock priority
3025 boosting. Select 0 to disable, 1 to boost
3026 only rt_mutex, and 2 to boost unconditionally.
3027 Defaults to 2, which might seem to be an
3028 odd choice, but which should be harmless for
3029 non-real-time spinlocks, due to their disabling
3030 of preemption. Note that non-realtime mutexes
3031 disable boosting.
3032
3033 locktorture.rt_boost_factor= [KNL]
3034 Number that determines how often and for how
3035 long priority boosting is exercised. This is
3036 scaled down by the number of writers, so that the
3037 number of boosts per unit time remains roughly
3038 constant as the number of writers increases.
3039 On the other hand, the duration of each boost
3040 increases with the number of writers.
3041
3042 locktorture.shuffle_interval= [KNL]
3043 Set task-shuffle interval (jiffies). Shuffling
3044 tasks allows some CPUs to go into dyntick-idle
3045 mode during the locktorture test.
3046
3047 locktorture.shutdown_secs= [KNL]
3048 Set time (s) after boot system shutdown. This
3049 is useful for hands-off automated testing.
3050
3051 locktorture.stat_interval= [KNL]
3052 Time (s) between statistics printk()s.
3053
3054 locktorture.stutter= [KNL]
3055 Time (s) to stutter testing, for example,
3056 specifying five seconds causes the test to run for
3057 five seconds, wait for five seconds, and so on.
3058 This tests the locking primitive's ability to
3059 transition abruptly to and from idle.
3060
3061 locktorture.torture_type= [KNL]
3062 Specify the locking implementation to test.
3063
3064 locktorture.verbose= [KNL]
3065 Enable additional printk() statements.
3066
3067 locktorture.writer_fifo= [KNL]
3068 Run the write-side locktorture kthreads at
3069 sched_set_fifo() real-time priority.
3070
3071 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver
3072 Format: <irq>
3073
3074 loglevel= [KNL,EARLY]
3075 All Kernel Messages with a loglevel smaller than the
3076 console loglevel will be printed to the console. It can
3077 also be changed with klogd or other programs. The
3078 loglevels are defined as follows:
3079
3080 0 (KERN_EMERG) system is unusable
3081 1 (KERN_ALERT) action must be taken immediately
3082 2 (KERN_CRIT) critical conditions
3083 3 (KERN_ERR) error conditions
3084 4 (KERN_WARNING) warning conditions
3085 5 (KERN_NOTICE) normal but significant condition
3086 6 (KERN_INFO) informational
3087 7 (KERN_DEBUG) debug-level messages
3088
3089 log_buf_len=n[KMG] [KNL,EARLY]
3090 Sets the size of the printk ring buffer, in bytes.
3091 n must be a power of two and greater than the
3092 minimal size. The minimal size is defined by
3093 LOG_BUF_SHIFT kernel config parameter. There
3094 is also CONFIG_LOG_CPU_MAX_BUF_SHIFT config
3095 parameter that allows to increase the default size
3096 depending on the number of CPUs. See init/Kconfig
3097 for more details.
3098
3099 logo.nologo [FB] Disables display of the built-in Linux logo.
3100 This may be used to provide more screen space for
3101 kernel log messages and is useful when debugging
3102 kernel boot problems.
3103
3104 lp=0 [LP] Specify parallel ports to use, e.g,
3105 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses
3106 lp=reset first parallel port). 'lp=0' disables the
3107 lp=auto printer driver. 'lp=reset' (which can be
3108 specified in addition to the ports) causes
3109 attached printers to be reset. Using
3110 lp=port1,port2,... specifies the parallel ports
3111 to associate lp devices with, starting with
3112 lp0. A port specification may be 'none' to skip
3113 that lp device, or a parport name such as
3114 'parport0'. Specifying 'lp=auto' instead of a
3115 port specification list means that device IDs
3116 from each port should be examined, to see if
3117 an IEEE 1284-compliant printer is attached; if
3118 so, the driver will manage that printer.
3119 See also header of drivers/char/lp.c.
3120
3121 lpj=n [KNL]
3122 Sets loops_per_jiffy to given constant, thus avoiding
3123 time-consuming boot-time autodetection (up to 250 ms per
3124 CPU). 0 enables autodetection (default). To determine
3125 the correct value for your kernel, boot with normal
3126 autodetection and see what value is printed. Note that
3127 on SMP systems the preset will be applied to all CPUs,
3128 which is likely to cause problems if your CPUs need
3129 significantly divergent settings. An incorrect value
3130 will cause delays in the kernel to be wrong, leading to
3131 unpredictable I/O errors and other breakage. Although
3132 unlikely, in the extreme case this might damage your
3133 hardware.
3134
3135 ltpc= [NET]
3136 Format: <io>,<irq>,<dma>
3137
3138 lsm.debug [SECURITY] Enable LSM initialization debugging output.
3139
3140 lsm=lsm1,...,lsmN
3141 [SECURITY] Choose order of LSM initialization. This
3142 overrides CONFIG_LSM, and the "security=" parameter.
3143
3144 machvec= [IA-64] Force the use of a particular machine-vector
3145 (machvec) in a generic kernel.
3146 Example: machvec=hpzx1
3147
3148 machtype= [Loongson] Share the same kernel image file between
3149 different yeeloong laptops.
3150 Example: machtype=lemote-yeeloong-2f-7inch
3151
3152 max_addr=nn[KMG] [KNL,BOOT,IA-64] All physical memory greater
3153 than or equal to this physical address is ignored.
3154
3155 maxcpus= [SMP,EARLY] Maximum number of processors that an SMP kernel
3156 will bring up during bootup. maxcpus=n : n >= 0 limits
3157 the kernel to bring up 'n' processors. Surely after
3158 bootup you can bring up the other plugged cpu by executing
3159 "echo 1 > /sys/devices/system/cpu/cpuX/online". So maxcpus
3160 only takes effect during system bootup.
3161 While n=0 is a special case, it is equivalent to "nosmp",
3162 which also disables the IO APIC.
3163
3164 max_loop= [LOOP] The number of loop block devices that get
3165 (loop.max_loop) unconditionally pre-created at init time. The default
3166 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead
3167 of statically allocating a predefined number, loop
3168 devices can be requested on-demand with the
3169 /dev/loop-control interface.
3170
3171 mce [X86-32] Machine Check Exception
3172
3173 mce=option [X86-64] See Documentation/arch/x86/x86_64/boot-options.rst
3174
3175 md= [HW] RAID subsystems devices and level
3176 See Documentation/admin-guide/md.rst.
3177
3178 mdacon= [MDA]
3179 Format: <first>,<last>
3180 Specifies range of consoles to be captured by the MDA.
3181
3182 mds= [X86,INTEL,EARLY]
3183 Control mitigation for the Micro-architectural Data
3184 Sampling (MDS) vulnerability.
3185
3186 Certain CPUs are vulnerable to an exploit against CPU
3187 internal buffers which can forward information to a
3188 disclosure gadget under certain conditions.
3189
3190 In vulnerable processors, the speculatively
3191 forwarded data can be used in a cache side channel
3192 attack, to access data to which the attacker does
3193 not have direct access.
3194
3195 This parameter controls the MDS mitigation. The
3196 options are:
3197
3198 full - Enable MDS mitigation on vulnerable CPUs
3199 full,nosmt - Enable MDS mitigation and disable
3200 SMT on vulnerable CPUs
3201 off - Unconditionally disable MDS mitigation
3202
3203 On TAA-affected machines, mds=off can be prevented by
3204 an active TAA mitigation as both vulnerabilities are
3205 mitigated with the same mechanism so in order to disable
3206 this mitigation, you need to specify tsx_async_abort=off
3207 too.
3208
3209 Not specifying this option is equivalent to
3210 mds=full.
3211
3212 For details see: Documentation/admin-guide/hw-vuln/mds.rst
3213
3214 mem=nn[KMG] [HEXAGON,EARLY] Set the memory size.
3215 Must be specified, otherwise memory size will be 0.
3216
3217 mem=nn[KMG] [KNL,BOOT,EARLY] Force usage of a specific amount
3218 of memory Amount of memory to be used in cases
3219 as follows:
3220
3221 1 for test;
3222 2 when the kernel is not able to see the whole system memory;
3223 3 memory that lies after 'mem=' boundary is excluded from
3224 the hypervisor, then assigned to KVM guests.
3225 4 to limit the memory available for kdump kernel.
3226
3227 [ARC,MICROBLAZE] - the limit applies only to low memory,
3228 high memory is not affected.
3229
3230 [ARM64] - only limits memory covered by the linear
3231 mapping. The NOMAP regions are not affected.
3232
3233 [X86] Work as limiting max address. Use together
3234 with memmap= to avoid physical address space collisions.
3235 Without memmap= PCI devices could be placed at addresses
3236 belonging to unused RAM.
3237
3238 Note that this only takes effects during boot time since
3239 in above case 3, memory may need be hot added after boot
3240 if system memory of hypervisor is not sufficient.
3241
3242 mem=nn[KMG]@ss[KMG]
3243 [ARM,MIPS,EARLY] - override the memory layout
3244 reported by firmware.
3245 Define a memory region of size nn[KMG] starting at
3246 ss[KMG].
3247 Multiple different regions can be specified with
3248 multiple mem= parameters on the command line.
3249
3250 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel
3251 memory.
3252
3253 memblock=debug [KNL,EARLY] Enable memblock debug messages.
3254
3255 memchunk=nn[KMG]
3256 [KNL,SH] Allow user to override the default size for
3257 per-device physically contiguous DMA buffers.
3258
3259 memhp_default_state=online/offline/online_kernel/online_movable
3260 [KNL] Set the initial state for the memory hotplug
3261 onlining policy. If not specified, the default value is
3262 set according to the
3263 CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE kernel config
3264 option.
3265 See Documentation/admin-guide/mm/memory-hotplug.rst.
3266
3267 memmap=exactmap [KNL,X86,EARLY] Enable setting of an exact
3268 E820 memory map, as specified by the user.
3269 Such memmap=exactmap lines can be constructed based on
3270 BIOS output or other requirements. See the memmap=nn@ss
3271 option description.
3272
3273 memmap=nn[KMG]@ss[KMG]
3274 [KNL, X86,MIPS,XTENSA,EARLY] Force usage of a specific region of memory.
3275 Region of memory to be used is from ss to ss+nn.
3276 If @ss[KMG] is omitted, it is equivalent to mem=nn[KMG],
3277 which limits max address to nn[KMG].
3278 Multiple different regions can be specified,
3279 comma delimited.
3280 Example:
3281 memmap=100M@2G,100M#3G,1G!1024G
3282
3283 memmap=nn[KMG]#ss[KMG]
3284 [KNL,ACPI,EARLY] Mark specific memory as ACPI data.
3285 Region of memory to be marked is from ss to ss+nn.
3286
3287 memmap=nn[KMG]$ss[KMG]
3288 [KNL,ACPI,EARLY] Mark specific memory as reserved.
3289 Region of memory to be reserved is from ss to ss+nn.
3290 Example: Exclude memory from 0x18690000-0x1869ffff
3291 memmap=64K$0x18690000
3292 or
3293 memmap=0x10000$0x18690000
3294 Some bootloaders may need an escape character before '$',
3295 like Grub2, otherwise '$' and the following number
3296 will be eaten.
3297
3298 memmap=nn[KMG]!ss[KMG,EARLY]
3299 [KNL,X86] Mark specific memory as protected.
3300 Region of memory to be used, from ss to ss+nn.
3301 The memory region may be marked as e820 type 12 (0xc)
3302 and is NVDIMM or ADR memory.
3303
3304 memmap=<size>%<offset>-<oldtype>+<newtype>
3305 [KNL,ACPI,EARLY] Convert memory within the specified region
3306 from <oldtype> to <newtype>. If "-<oldtype>" is left
3307 out, the whole region will be marked as <newtype>,
3308 even if previously unavailable. If "+<newtype>" is left
3309 out, matching memory will be removed. Types are
3310 specified as e820 types, e.g., 1 = RAM, 2 = reserved,
3311 3 = ACPI, 12 = PRAM.
3312
3313 memory_corruption_check=0/1 [X86,EARLY]
3314 Some BIOSes seem to corrupt the first 64k of
3315 memory when doing things like suspend/resume.
3316 Setting this option will scan the memory
3317 looking for corruption. Enabling this will
3318 both detect corruption and prevent the kernel
3319 from using the memory being corrupted.
3320 However, its intended as a diagnostic tool; if
3321 repeatable BIOS-originated corruption always
3322 affects the same memory, you can use memmap=
3323 to prevent the kernel from using that memory.
3324
3325 memory_corruption_check_size=size [X86,EARLY]
3326 By default it checks for corruption in the low
3327 64k, making this memory unavailable for normal
3328 use. Use this parameter to scan for
3329 corruption in more or less memory.
3330
3331 memory_corruption_check_period=seconds [X86,EARLY]
3332 By default it checks for corruption every 60
3333 seconds. Use this parameter to check at some
3334 other rate. 0 disables periodic checking.
3335
3336 memory_hotplug.memmap_on_memory
3337 [KNL,X86,ARM] Boolean flag to enable this feature.
3338 Format: {on | off (default)}
3339 When enabled, runtime hotplugged memory will
3340 allocate its internal metadata (struct pages,
3341 those vmemmap pages cannot be optimized even
3342 if hugetlb_free_vmemmap is enabled) from the
3343 hotadded memory which will allow to hotadd a
3344 lot of memory without requiring additional
3345 memory to do so.
3346 This feature is disabled by default because it
3347 has some implication on large (e.g. GB)
3348 allocations in some configurations (e.g. small
3349 memory blocks).
3350 The state of the flag can be read in
3351 /sys/module/memory_hotplug/parameters/memmap_on_memory.
3352 Note that even when enabled, there are a few cases where
3353 the feature is not effective.
3354
3355 memtest= [KNL,X86,ARM,M68K,PPC,RISCV,EARLY] Enable memtest
3356 Format: <integer>
3357 default : 0 <disable>
3358 Specifies the number of memtest passes to be
3359 performed. Each pass selects another test
3360 pattern from a given set of patterns. Memtest
3361 fills the memory with this pattern, validates
3362 memory contents and reserves bad memory
3363 regions that are detected.
3364
3365 mem_encrypt= [X86-64] AMD Secure Memory Encryption (SME) control
3366 Valid arguments: on, off
3367 Default: off
3368 mem_encrypt=on: Activate SME
3369 mem_encrypt=off: Do not activate SME
3370
3371 Refer to Documentation/virt/kvm/x86/amd-memory-encryption.rst
3372 for details on when memory encryption can be activated.
3373
3374 mem_sleep_default= [SUSPEND] Default system suspend mode:
3375 s2idle - Suspend-To-Idle
3376 shallow - Power-On Suspend or equivalent (if supported)
3377 deep - Suspend-To-RAM or equivalent (if supported)
3378 See Documentation/admin-guide/pm/sleep-states.rst.
3379
3380 mfgpt_irq= [IA-32] Specify the IRQ to use for the
3381 Multi-Function General Purpose Timers on AMD Geode
3382 platforms.
3383
3384 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when
3385 the BIOS has incorrectly applied a workaround. TinyBIOS
3386 version 0.98 is known to be affected, 0.99 fixes the
3387 problem by letting the user disable the workaround.
3388
3389 mga= [HW,DRM]
3390
3391 microcode.force_minrev= [X86]
3392 Format: <bool>
3393 Enable or disable the microcode minimal revision
3394 enforcement for the runtime microcode loader.
3395
3396 min_addr=nn[KMG] [KNL,BOOT,IA-64] All physical memory below this
3397 physical address is ignored.
3398
3399 mini2440= [ARM,HW,KNL]
3400 Format:[0..2][b][c][t]
3401 Default: "0tb"
3402 MINI2440 configuration specification:
3403 0 - The attached screen is the 3.5" TFT
3404 1 - The attached screen is the 7" TFT
3405 2 - The VGA Shield is attached (1024x768)
3406 Leaving out the screen size parameter will not load
3407 the TFT driver, and the framebuffer will be left
3408 unconfigured.
3409 b - Enable backlight. The TFT backlight pin will be
3410 linked to the kernel VESA blanking code and a GPIO
3411 LED. This parameter is not necessary when using the
3412 VGA shield.
3413 c - Enable the s3c camera interface.
3414 t - Reserved for enabling touchscreen support. The
3415 touchscreen support is not enabled in the mainstream
3416 kernel as of 2.6.30, a preliminary port can be found
3417 in the "bleeding edge" mini2440 support kernel at
3418 https://repo.or.cz/w/linux-2.6/mini2440.git
3419
3420 mitigations=
3421 [X86,PPC,S390,ARM64,EARLY] Control optional mitigations for
3422 CPU vulnerabilities. This is a set of curated,
3423 arch-independent options, each of which is an
3424 aggregation of existing arch-specific options.
3425
3426 Note, "mitigations" is supported if and only if the
3427 kernel was built with CPU_MITIGATIONS=y.
3428
3429 off
3430 Disable all optional CPU mitigations. This
3431 improves system performance, but it may also
3432 expose users to several CPU vulnerabilities.
3433 Equivalent to: if nokaslr then kpti=0 [ARM64]
3434 gather_data_sampling=off [X86]
3435 kvm.nx_huge_pages=off [X86]
3436 l1tf=off [X86]
3437 mds=off [X86]
3438 mmio_stale_data=off [X86]
3439 no_entry_flush [PPC]
3440 no_uaccess_flush [PPC]
3441 nobp=0 [S390]
3442 nopti [X86,PPC]
3443 nospectre_bhb [ARM64]
3444 nospectre_v1 [X86,PPC]
3445 nospectre_v2 [X86,PPC,S390,ARM64]
3446 reg_file_data_sampling=off [X86]
3447 retbleed=off [X86]
3448 spec_rstack_overflow=off [X86]
3449 spec_store_bypass_disable=off [X86,PPC]
3450 spectre_bhi=off [X86]
3451 spectre_v2_user=off [X86]
3452 srbds=off [X86,INTEL]
3453 ssbd=force-off [ARM64]
3454 tsx_async_abort=off [X86]
3455
3456 Exceptions:
3457 This does not have any effect on
3458 kvm.nx_huge_pages when
3459 kvm.nx_huge_pages=force.
3460
3461 auto (default)
3462 Mitigate all CPU vulnerabilities, but leave SMT
3463 enabled, even if it's vulnerable. This is for
3464 users who don't want to be surprised by SMT
3465 getting disabled across kernel upgrades, or who
3466 have other ways of avoiding SMT-based attacks.
3467 Equivalent to: (default behavior)
3468
3469 auto,nosmt
3470 Mitigate all CPU vulnerabilities, disabling SMT
3471 if needed. This is for users who always want to
3472 be fully mitigated, even if it means losing SMT.
3473 Equivalent to: l1tf=flush,nosmt [X86]
3474 mds=full,nosmt [X86]
3475 tsx_async_abort=full,nosmt [X86]
3476 mmio_stale_data=full,nosmt [X86]
3477 retbleed=auto,nosmt [X86]
3478
3479 mminit_loglevel=
3480 [KNL,EARLY] When CONFIG_DEBUG_MEMORY_INIT is set, this
3481 parameter allows control of the logging verbosity for
3482 the additional memory initialisation checks. A value
3483 of 0 disables mminit logging and a level of 4 will
3484 log everything. Information is printed at KERN_DEBUG
3485 so loglevel=8 may also need to be specified.
3486
3487 mmio_stale_data=
3488 [X86,INTEL,EARLY] Control mitigation for the Processor
3489 MMIO Stale Data vulnerabilities.
3490
3491 Processor MMIO Stale Data is a class of
3492 vulnerabilities that may expose data after an MMIO
3493 operation. Exposed data could originate or end in
3494 the same CPU buffers as affected by MDS and TAA.
3495 Therefore, similar to MDS and TAA, the mitigation
3496 is to clear the affected CPU buffers.
3497
3498 This parameter controls the mitigation. The
3499 options are:
3500
3501 full - Enable mitigation on vulnerable CPUs
3502
3503 full,nosmt - Enable mitigation and disable SMT on
3504 vulnerable CPUs.
3505
3506 off - Unconditionally disable mitigation
3507
3508 On MDS or TAA affected machines,
3509 mmio_stale_data=off can be prevented by an active
3510 MDS or TAA mitigation as these vulnerabilities are
3511 mitigated with the same mechanism so in order to
3512 disable this mitigation, you need to specify
3513 mds=off and tsx_async_abort=off too.
3514
3515 Not specifying this option is equivalent to
3516 mmio_stale_data=full.
3517
3518 For details see:
3519 Documentation/admin-guide/hw-vuln/processor_mmio_stale_data.rst
3520
3521 <module>.async_probe[=<bool>] [KNL]
3522 If no <bool> value is specified or if the value
3523 specified is not a valid <bool>, enable asynchronous
3524 probe on this module. Otherwise, enable/disable
3525 asynchronous probe on this module as indicated by the
3526 <bool> value. See also: module.async_probe
3527
3528 module.async_probe=<bool>
3529 [KNL] When set to true, modules will use async probing
3530 by default. To enable/disable async probing for a
3531 specific module, use the module specific control that
3532 is documented under <module>.async_probe. When both
3533 module.async_probe and <module>.async_probe are
3534 specified, <module>.async_probe takes precedence for
3535 the specific module.
3536
3537 module.enable_dups_trace
3538 [KNL] When CONFIG_MODULE_DEBUG_AUTOLOAD_DUPS is set,
3539 this means that duplicate request_module() calls will
3540 trigger a WARN_ON() instead of a pr_warn(). Note that
3541 if MODULE_DEBUG_AUTOLOAD_DUPS_TRACE is set, WARN_ON()s
3542 will always be issued and this option does nothing.
3543 module.sig_enforce
3544 [KNL] When CONFIG_MODULE_SIG is set, this means that
3545 modules without (valid) signatures will fail to load.
3546 Note that if CONFIG_MODULE_SIG_FORCE is set, that
3547 is always true, so this option does nothing.
3548
3549 module_blacklist= [KNL] Do not load a comma-separated list of
3550 modules. Useful for debugging problem modules.
3551
3552 mousedev.tap_time=
3553 [MOUSE] Maximum time between finger touching and
3554 leaving touchpad surface for touch to be considered
3555 a tap and be reported as a left button click (for
3556 touchpads working in absolute mode only).
3557 Format: <msecs>
3558 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices
3559 reporting absolute coordinates, such as tablets
3560 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices
3561 reporting absolute coordinates, such as tablets
3562
3563 movablecore= [KNL,X86,IA-64,PPC,EARLY]
3564 Format: nn[KMGTPE] | nn%
3565 This parameter is the complement to kernelcore=, it
3566 specifies the amount of memory used for migratable
3567 allocations. If both kernelcore and movablecore is
3568 specified, then kernelcore will be at *least* the
3569 specified value but may be more. If movablecore on its
3570 own is specified, the administrator must be careful
3571 that the amount of memory usable for all allocations
3572 is not too small.
3573
3574 movable_node [KNL,EARLY] Boot-time switch to make hotplugable memory
3575 NUMA nodes to be movable. This means that the memory
3576 of such nodes will be usable only for movable
3577 allocations which rules out almost all kernel
3578 allocations. Use with caution!
3579
3580 MTD_Partition= [MTD]
3581 Format: <name>,<region-number>,<size>,<offset>
3582
3583 MTD_Region= [MTD] Format:
3584 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>]
3585
3586 mtdparts= [MTD]
3587 See drivers/mtd/parsers/cmdlinepart.c
3588
3589 mtdset= [ARM]
3590 ARM/S3C2412 JIVE boot control
3591
3592 See arch/arm/mach-s3c/mach-jive.c
3593
3594 mtouchusb.raw_coordinates=
3595 [HW] Make the MicroTouch USB driver use raw coordinates
3596 ('y', default) or cooked coordinates ('n')
3597
3598 mtrr=debug [X86,EARLY]
3599 Enable printing debug information related to MTRR
3600 registers at boot time.
3601
3602 mtrr_chunk_size=nn[KMG,X86,EARLY]
3603 used for mtrr cleanup. It is largest continuous chunk
3604 that could hold holes aka. UC entries.
3605
3606 mtrr_gran_size=nn[KMG,X86,EARLY]
3607 Used for mtrr cleanup. It is granularity of mtrr block.
3608 Default is 1.
3609 Large value could prevent small alignment from
3610 using up MTRRs.
3611
3612 mtrr_spare_reg_nr=n [X86,EARLY]
3613 Format: <integer>
3614 Range: 0,7 : spare reg number
3615 Default : 1
3616 Used for mtrr cleanup. It is spare mtrr entries number.
3617 Set to 2 or more if your graphical card needs more.
3618
3619 multitce=off [PPC] This parameter disables the use of the pSeries
3620 firmware feature for updating multiple TCE entries
3621 at a time.
3622
3623 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card
3624
3625 netdev= [NET] Network devices parameters
3626 Format: <irq>,<io>,<mem_start>,<mem_end>,<name>
3627 Note that mem_start is often overloaded to mean
3628 something different and driver-specific.
3629 This usage is only documented in each driver source
3630 file if at all.
3631
3632 netpoll.carrier_timeout=
3633 [NET] Specifies amount of time (in seconds) that
3634 netpoll should wait for a carrier. By default netpoll
3635 waits 4 seconds.
3636
3637 nf_conntrack.acct=
3638 [NETFILTER] Enable connection tracking flow accounting
3639 0 to disable accounting
3640 1 to enable accounting
3641 Default value is 0.
3642
3643 nfs.cache_getent=
3644 [NFS] sets the pathname to the program which is used
3645 to update the NFS client cache entries.
3646
3647 nfs.cache_getent_timeout=
3648 [NFS] sets the timeout after which an attempt to
3649 update a cache entry is deemed to have failed.
3650
3651 nfs.callback_nr_threads=
3652 [NFSv4] set the total number of threads that the
3653 NFS client will assign to service NFSv4 callback
3654 requests.
3655
3656 nfs.callback_tcpport=
3657 [NFS] set the TCP port on which the NFSv4 callback
3658 channel should listen.
3659
3660 nfs.delay_retrans=
3661 [NFS] specifies the number of times the NFSv4 client
3662 retries the request before returning an EAGAIN error,
3663 after a reply of NFS4ERR_DELAY from the server.
3664 Only applies if the softerr mount option is enabled,
3665 and the specified value is >= 0.
3666
3667 nfs.enable_ino64=
3668 [NFS] enable 64-bit inode numbers.
3669 If zero, the NFS client will fake up a 32-bit inode
3670 number for the readdir() and stat() syscalls instead
3671 of returning the full 64-bit number.
3672 The default is to return 64-bit inode numbers.
3673
3674 nfs.idmap_cache_timeout=
3675 [NFS] set the maximum lifetime for idmapper cache
3676 entries.
3677
3678 nfs.max_session_cb_slots=
3679 [NFSv4.1] Sets the maximum number of session
3680 slots the client will assign to the callback
3681 channel. This determines the maximum number of
3682 callbacks the client will process in parallel for
3683 a particular server.
3684
3685 nfs.max_session_slots=
3686 [NFSv4.1] Sets the maximum number of session slots
3687 the client will attempt to negotiate with the server.
3688 This limits the number of simultaneous RPC requests
3689 that the client can send to the NFSv4.1 server.
3690 Note that there is little point in setting this
3691 value higher than the max_tcp_slot_table_limit.
3692
3693 nfs.nfs4_disable_idmapping=
3694 [NFSv4] When set to the default of '1', this option
3695 ensures that both the RPC level authentication
3696 scheme and the NFS level operations agree to use
3697 numeric uids/gids if the mount is using the
3698 'sec=sys' security flavour. In effect it is
3699 disabling idmapping, which can make migration from
3700 legacy NFSv2/v3 systems to NFSv4 easier.
3701 Servers that do not support this mode of operation
3702 will be autodetected by the client, and it will fall
3703 back to using the idmapper.
3704 To turn off this behaviour, set the value to '0'.
3705
3706 nfs.nfs4_unique_id=
3707 [NFS4] Specify an additional fixed unique ident-
3708 ification string that NFSv4 clients can insert into
3709 their nfs_client_id4 string. This is typically a
3710 UUID that is generated at system install time.
3711
3712 nfs.recover_lost_locks=
3713 [NFSv4] Attempt to recover locks that were lost due
3714 to a lease timeout on the server. Please note that
3715 doing this risks data corruption, since there are
3716 no guarantees that the file will remain unchanged
3717 after the locks are lost.
3718 If you want to enable the kernel legacy behaviour of
3719 attempting to recover these locks, then set this
3720 parameter to '1'.
3721 The default parameter value of '0' causes the kernel
3722 not to attempt recovery of lost locks.
3723
3724 nfs.send_implementation_id=
3725 [NFSv4.1] Send client implementation identification
3726 information in exchange_id requests.
3727 If zero, no implementation identification information
3728 will be sent.
3729 The default is to send the implementation identification
3730 information.
3731
3732 nfs4.layoutstats_timer=
3733 [NFSv4.2] Change the rate at which the kernel sends
3734 layoutstats to the pNFS metadata server.
3735
3736 Setting this to value to 0 causes the kernel to use
3737 whatever value is the default set by the layout
3738 driver. A non-zero value sets the minimum interval
3739 in seconds between layoutstats transmissions.
3740
3741 nfsd.inter_copy_offload_enable=
3742 [NFSv4.2] When set to 1, the server will support
3743 server-to-server copies for which this server is
3744 the destination of the copy.
3745
3746 nfsd.nfs4_disable_idmapping=
3747 [NFSv4] When set to the default of '1', the NFSv4
3748 server will return only numeric uids and gids to
3749 clients using auth_sys, and will accept numeric uids
3750 and gids from such clients. This is intended to ease
3751 migration from NFSv2/v3.
3752
3753 nfsd.nfsd4_ssc_umount_timeout=
3754 [NFSv4.2] When used as the destination of a
3755 server-to-server copy, knfsd temporarily mounts
3756 the source server. It caches the mount in case
3757 it will be needed again, and discards it if not
3758 used for the number of milliseconds specified by
3759 this parameter.
3760
3761 nfsaddrs= [NFS] Deprecated. Use ip= instead.
3762 See Documentation/admin-guide/nfs/nfsroot.rst.
3763
3764 nfsroot= [NFS] nfs root filesystem for disk-less boxes.
3765 See Documentation/admin-guide/nfs/nfsroot.rst.
3766
3767 nfsrootdebug [NFS] enable nfsroot debugging messages.
3768 See Documentation/admin-guide/nfs/nfsroot.rst.
3769
3770 nmi_backtrace.backtrace_idle [KNL]
3771 Dump stacks even of idle CPUs in response to an
3772 NMI stack-backtrace request.
3773
3774 nmi_debug= [KNL,SH] Specify one or more actions to take
3775 when a NMI is triggered.
3776 Format: [state][,regs][,debounce][,die]
3777
3778 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels
3779 Format: [panic,][nopanic,][num]
3780 Valid num: 0 or 1
3781 0 - turn hardlockup detector in nmi_watchdog off
3782 1 - turn hardlockup detector in nmi_watchdog on
3783 When panic is specified, panic when an NMI watchdog
3784 timeout occurs (or 'nopanic' to not panic on an NMI
3785 watchdog, if CONFIG_BOOTPARAM_HARDLOCKUP_PANIC is set)
3786 To disable both hard and soft lockup detectors,
3787 please see 'nowatchdog'.
3788 This is useful when you use a panic=... timeout and
3789 need the box quickly up again.
3790
3791 These settings can be accessed at runtime via
3792 the nmi_watchdog and hardlockup_panic sysctls.
3793
3794 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths
3795 emulation library even if a 387 maths coprocessor
3796 is present.
3797
3798 no4lvl [RISCV,EARLY] Disable 4-level and 5-level paging modes.
3799 Forces kernel to use 3-level paging instead.
3800
3801 no5lvl [X86-64,RISCV,EARLY] Disable 5-level paging mode. Forces
3802 kernel to use 4-level paging instead.
3803
3804 noalign [KNL,ARM]
3805
3806 noaltinstr [S390,EARLY] Disables alternative instructions
3807 patching (CPU alternatives feature).
3808
3809 noapic [SMP,APIC,EARLY] Tells the kernel to not make use of any
3810 IOAPICs that may be present in the system.
3811
3812 noautogroup Disable scheduler automatic task group creation.
3813
3814 nocache [ARM,EARLY]
3815
3816 no_console_suspend
3817 [HW] Never suspend the console
3818 Disable suspending of consoles during suspend and
3819 hibernate operations. Once disabled, debugging
3820 messages can reach various consoles while the rest
3821 of the system is being put to sleep (ie, while
3822 debugging driver suspend/resume hooks). This may
3823 not work reliably with all consoles, but is known
3824 to work with serial and VGA consoles.
3825 To facilitate more flexible debugging, we also add
3826 console_suspend, a printk module parameter to control
3827 it. Users could use console_suspend (usually
3828 /sys/module/printk/parameters/console_suspend) to
3829 turn on/off it dynamically.
3830
3831 no_debug_objects
3832 [KNL,EARLY] Disable object debugging
3833
3834 nodsp [SH] Disable hardware DSP at boot time.
3835
3836 noefi [EFI,EARLY] Disable EFI runtime services support.
3837
3838 no_entry_flush [PPC,EARLY] Don't flush the L1-D cache when entering the kernel.
3839
3840 noexec [IA-64]
3841
3842 noexec32 [X86-64]
3843 This affects only 32-bit executables.
3844 noexec32=on: enable non-executable mappings (default)
3845 read doesn't imply executable mappings
3846 noexec32=off: disable non-executable mappings
3847 read implies executable mappings
3848
3849 no_file_caps Tells the kernel not to honor file capabilities. The
3850 only way then for a file to be executed with privilege
3851 is to be setuid root or executed by root.
3852
3853 nofpu [MIPS,SH] Disable hardware FPU at boot time.
3854
3855 nofsgsbase [X86] Disables FSGSBASE instructions.
3856
3857 nofxsr [BUGS=X86-32] Disables x86 floating point extended
3858 register save and restore. The kernel will only save
3859 legacy floating-point registers on task switch.
3860
3861 nohalt [IA-64] Tells the kernel not to use the power saving
3862 function PAL_HALT_LIGHT when idle. This increases
3863 power-consumption. On the positive side, it reduces
3864 interrupt wake-up latency, which may improve performance
3865 in certain environments such as networked servers or
3866 real-time systems.
3867
3868 no_hash_pointers
3869 [KNL,EARLY]
3870 Force pointers printed to the console or buffers to be
3871 unhashed. By default, when a pointer is printed via %p
3872 format string, that pointer is "hashed", i.e. obscured
3873 by hashing the pointer value. This is a security feature
3874 that hides actual kernel addresses from unprivileged
3875 users, but it also makes debugging the kernel more
3876 difficult since unequal pointers can no longer be
3877 compared. However, if this command-line option is
3878 specified, then all normal pointers will have their true
3879 value printed. This option should only be specified when
3880 debugging the kernel. Please do not use on production
3881 kernels.
3882
3883 nohibernate [HIBERNATION] Disable hibernation and resume.
3884
3885 nohlt [ARM,ARM64,MICROBLAZE,MIPS,PPC,SH] Forces the kernel to
3886 busy wait in do_idle() and not use the arch_cpu_idle()
3887 implementation; requires CONFIG_GENERIC_IDLE_POLL_SETUP
3888 to be effective. This is useful on platforms where the
3889 sleep(SH) or wfi(ARM,ARM64) instructions do not work
3890 correctly or when doing power measurements to evaluate
3891 the impact of the sleep instructions. This is also
3892 useful when using JTAG debugger.
3893
3894 nohugeiomap [KNL,X86,PPC,ARM64,EARLY] Disable kernel huge I/O mappings.
3895
3896 nohugevmalloc [KNL,X86,PPC,ARM64,EARLY] Disable kernel huge vmalloc mappings.
3897
3898 nohz= [KNL] Boottime enable/disable dynamic ticks
3899 Valid arguments: on, off
3900 Default: on
3901
3902 nohz_full= [KNL,BOOT,SMP,ISOL]
3903 The argument is a cpu list, as described above.
3904 In kernels built with CONFIG_NO_HZ_FULL=y, set
3905 the specified list of CPUs whose tick will be stopped
3906 whenever possible. The boot CPU will be forced outside
3907 the range to maintain the timekeeping. Any CPUs
3908 in this list will have their RCU callbacks offloaded,
3909 just as if they had also been called out in the
3910 rcu_nocbs= boot parameter.
3911
3912 Note that this argument takes precedence over
3913 the CONFIG_RCU_NOCB_CPU_DEFAULT_ALL option.
3914
3915 noinitrd [RAM] Tells the kernel not to load any configured
3916 initial RAM disk.
3917
3918 nointremap [X86-64,Intel-IOMMU,EARLY] Do not enable interrupt
3919 remapping.
3920 [Deprecated - use intremap=off]
3921
3922 nointroute [IA-64]
3923
3924 noinvpcid [X86,EARLY] Disable the INVPCID cpu feature.
3925
3926 noiotrap [SH] Disables trapped I/O port accesses.
3927
3928 noirqdebug [X86-32] Disables the code which attempts to detect and
3929 disable unhandled interrupt sources.
3930
3931 noisapnp [ISAPNP] Disables ISA PnP code.
3932
3933 nojitter [IA-64] Disables jitter checking for ITC timers.
3934
3935 nokaslr [KNL,EARLY]
3936 When CONFIG_RANDOMIZE_BASE is set, this disables
3937 kernel and module base offset ASLR (Address Space
3938 Layout Randomization).
3939
3940 no-kvmapf [X86,KVM,EARLY] Disable paravirtualized asynchronous page
3941 fault handling.
3942
3943 no-kvmclock [X86,KVM,EARLY] Disable paravirtualized KVM clock driver
3944
3945 nolapic [X86-32,APIC,EARLY] Do not enable or use the local APIC.
3946
3947 nolapic_timer [X86-32,APIC,EARLY] Do not use the local APIC timer.
3948
3949 nomca [IA-64] Disable machine check abort handling
3950
3951 nomce [X86-32] Disable Machine Check Exception
3952
3953 nomfgpt [X86-32] Disable Multi-Function General Purpose
3954 Timer usage (for AMD Geode machines).
3955
3956 nomodeset Disable kernel modesetting. Most systems' firmware
3957 sets up a display mode and provides framebuffer memory
3958 for output. With nomodeset, DRM and fbdev drivers will
3959 not load if they could possibly displace the pre-
3960 initialized output. Only the system framebuffer will
3961 be available for use. The respective drivers will not
3962 perform display-mode changes or accelerated rendering.
3963
3964 Useful as error fallback, or for testing and debugging.
3965
3966 nomodule Disable module load
3967
3968 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to
3969 shutdown the other cpus. Instead use the REBOOT_VECTOR
3970 irq.
3971
3972 nopat [X86,EARLY] Disable PAT (page attribute table extension of
3973 pagetables) support.
3974
3975 nopcid [X86-64,EARLY] Disable the PCID cpu feature.
3976
3977 nopku [X86] Disable Memory Protection Keys CPU feature found
3978 in some Intel CPUs.
3979
3980 nopti [X86-64,EARLY]
3981 Equivalent to pti=off
3982
3983 nopv= [X86,XEN,KVM,HYPER_V,VMWARE,EARLY]
3984 Disables the PV optimizations forcing the guest to run
3985 as generic guest with no PV drivers. Currently support
3986 XEN HVM, KVM, HYPER_V and VMWARE guest.
3987
3988 nopvspin [X86,XEN,KVM,EARLY]
3989 Disables the qspinlock slow path using PV optimizations
3990 which allow the hypervisor to 'idle' the guest on lock
3991 contention.
3992
3993 norandmaps Don't use address space randomization. Equivalent to
3994 echo 0 > /proc/sys/kernel/randomize_va_space
3995
3996 noreplace-smp [X86-32,SMP] Don't replace SMP instructions
3997 with UP alternatives
3998
3999 noresume [SWSUSP] Disables resume and restores original swap
4000 space.
4001
4002 nosbagart [IA-64]
4003
4004 no-scroll [VGA] Disables scrollback.
4005 This is required for the Braillex ib80-piezo Braille
4006 reader made by F.H. Papenmeier (Germany).
4007
4008 nosgx [X86-64,SGX,EARLY] Disables Intel SGX kernel support.
4009
4010 nosmap [PPC,EARLY]
4011 Disable SMAP (Supervisor Mode Access Prevention)
4012 even if it is supported by processor.
4013
4014 nosmep [PPC64s,EARLY]
4015 Disable SMEP (Supervisor Mode Execution Prevention)
4016 even if it is supported by processor.
4017
4018 nosmp [SMP,EARLY] Tells an SMP kernel to act as a UP kernel,
4019 and disable the IO APIC. legacy for "maxcpus=0".
4020
4021 nosmt [KNL,MIPS,PPC,S390,EARLY] Disable symmetric multithreading (SMT).
4022 Equivalent to smt=1.
4023
4024 [KNL,X86,PPC] Disable symmetric multithreading (SMT).
4025 nosmt=force: Force disable SMT, cannot be undone
4026 via the sysfs control file.
4027
4028 nosoftlockup [KNL] Disable the soft-lockup detector.
4029
4030 nospec_store_bypass_disable
4031 [HW,EARLY] Disable all mitigations for the Speculative
4032 Store Bypass vulnerability
4033
4034 nospectre_bhb [ARM64,EARLY] Disable all mitigations for Spectre-BHB (branch
4035 history injection) vulnerability. System may allow data leaks
4036 with this option.
4037
4038 nospectre_v1 [X86,PPC,EARLY] Disable mitigations for Spectre Variant 1
4039 (bounds check bypass). With this option data leaks are
4040 possible in the system.
4041
4042 nospectre_v2 [X86,PPC_E500,ARM64,EARLY] Disable all mitigations
4043 for the Spectre variant 2 (indirect branch
4044 prediction) vulnerability. System may allow data
4045 leaks with this option.
4046
4047 no-steal-acc [X86,PV_OPS,ARM64,PPC/PSERIES,RISCV,EARLY] Disable
4048 paravirtualized steal time accounting. steal time is
4049 computed, but won't influence scheduler behaviour
4050
4051 nosync [HW,M68K] Disables sync negotiation for all devices.
4052
4053 no_timer_check [X86,APIC] Disables the code which tests for
4054 broken timer IRQ sources.
4055
4056 no_uaccess_flush
4057 [PPC,EARLY] Don't flush the L1-D cache after accessing user data.
4058
4059 novmcoredd [KNL,KDUMP]
4060 Disable device dump. Device dump allows drivers to
4061 append dump data to vmcore so you can collect driver
4062 specified debug info. Drivers can append the data
4063 without any limit and this data is stored in memory,
4064 so this may cause significant memory stress. Disabling
4065 device dump can help save memory but the driver debug
4066 data will be no longer available. This parameter
4067 is only available when CONFIG_PROC_VMCORE_DEVICE_DUMP
4068 is set.
4069
4070 no-vmw-sched-clock
4071 [X86,PV_OPS,EARLY] Disable paravirtualized VMware
4072 scheduler clock and use the default one.
4073
4074 nowatchdog [KNL] Disable both lockup detectors, i.e.
4075 soft-lockup and NMI watchdog (hard-lockup).
4076
4077 nowb [ARM,EARLY]
4078
4079 nox2apic [X86-64,APIC,EARLY] Do not enable x2APIC mode.
4080
4081 NOTE: this parameter will be ignored on systems with the
4082 LEGACY_XAPIC_DISABLED bit set in the
4083 IA32_XAPIC_DISABLE_STATUS MSR.
4084
4085 noxsave [BUGS=X86] Disables x86 extended register state save
4086 and restore using xsave. The kernel will fallback to
4087 enabling legacy floating-point and sse state.
4088
4089 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended
4090 register states. The kernel will fall back to use
4091 xsave to save the states. By using this parameter,
4092 performance of saving the states is degraded because
4093 xsave doesn't support modified optimization while
4094 xsaveopt supports it on xsaveopt enabled systems.
4095
4096 noxsaves [X86] Disables xsaves and xrstors used in saving and
4097 restoring x86 extended register state in compacted
4098 form of xsave area. The kernel will fall back to use
4099 xsaveopt and xrstor to save and restore the states
4100 in standard form of xsave area. By using this
4101 parameter, xsave area per process might occupy more
4102 memory on xsaves enabled systems.
4103
4104 nps_mtm_hs_ctr= [KNL,ARC]
4105 This parameter sets the maximum duration, in
4106 cycles, each HW thread of the CTOP can run
4107 without interruptions, before HW switches it.
4108 The actual maximum duration is 16 times this
4109 parameter's value.
4110 Format: integer between 1 and 255
4111 Default: 255
4112
4113 nptcg= [IA-64] Override max number of concurrent global TLB
4114 purges which is reported from either PAL_VM_SUMMARY or
4115 SAL PALO.
4116
4117 nr_cpus= [SMP,EARLY] Maximum number of processors that an SMP kernel
4118 could support. nr_cpus=n : n >= 1 limits the kernel to
4119 support 'n' processors. It could be larger than the
4120 number of already plugged CPU during bootup, later in
4121 runtime you can physically add extra cpu until it reaches
4122 n. So during boot up some boot time memory for per-cpu
4123 variables need be pre-allocated for later physical cpu
4124 hot plugging.
4125
4126 nr_uarts= [SERIAL] maximum number of UARTs to be registered.
4127
4128 numa=off [KNL, ARM64, PPC, RISCV, SPARC, X86, EARLY]
4129 Disable NUMA, Only set up a single NUMA node
4130 spanning all memory.
4131
4132 numa_balancing= [KNL,ARM64,PPC,RISCV,S390,X86] Enable or disable automatic
4133 NUMA balancing.
4134 Allowed values are enable and disable
4135
4136 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA.
4137 'node', 'default' can be specified
4138 This can be set from sysctl after boot.
4139 See Documentation/admin-guide/sysctl/vm.rst for details.
4140
4141 ohci1394_dma=early [HW,EARLY] enable debugging via the ohci1394 driver.
4142 See Documentation/core-api/debugging-via-ohci1394.rst for more
4143 info.
4144
4145 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands
4146 Rather than timing out after 20 ms if an EC
4147 command is not properly ACKed, override the length
4148 of the timeout. We have interrupts disabled while
4149 waiting for the ACK, so if this is set too high
4150 interrupts *may* be lost!
4151
4152 omap_mux= [OMAP] Override bootloader pin multiplexing.
4153 Format: <mux_mode0.mode_name=value>...
4154 For example, to override I2C bus2:
4155 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100
4156
4157 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration
4158
4159 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock]
4160
4161 boundary - index of last SLC block on Flex-OneNAND.
4162 The remaining blocks are configured as MLC blocks.
4163 lock - Configure if Flex-OneNAND boundary should be locked.
4164 Once locked, the boundary cannot be changed.
4165 1 indicates lock status, 0 indicates unlock status.
4166
4167 oops=panic [KNL,EARLY]
4168 Always panic on oopses. Default is to just kill the
4169 process, but there is a small probability of
4170 deadlocking the machine.
4171 This will also cause panics on machine check exceptions.
4172 Useful together with panic=30 to trigger a reboot.
4173
4174 page_alloc.shuffle=
4175 [KNL] Boolean flag to control whether the page allocator
4176 should randomize its free lists. The randomization may
4177 be automatically enabled if the kernel detects it is
4178 running on a platform with a direct-mapped memory-side
4179 cache, and this parameter can be used to
4180 override/disable that behavior. The state of the flag
4181 can be read from sysfs at:
4182 /sys/module/page_alloc/parameters/shuffle.
4183
4184 page_owner= [KNL,EARLY] Boot-time page_owner enabling option.
4185 Storage of the information about who allocated
4186 each page is disabled in default. With this switch,
4187 we can turn it on.
4188 on: enable the feature
4189
4190 page_poison= [KNL,EARLY] Boot-time parameter changing the state of
4191 poisoning on the buddy allocator, available with
4192 CONFIG_PAGE_POISONING=y.
4193 off: turn off poisoning (default)
4194 on: turn on poisoning
4195
4196 page_reporting.page_reporting_order=
4197 [KNL] Minimal page reporting order
4198 Format: <integer>
4199 Adjust the minimal page reporting order. The page
4200 reporting is disabled when it exceeds MAX_PAGE_ORDER.
4201
4202 panic= [KNL] Kernel behaviour on panic: delay <timeout>
4203 timeout > 0: seconds before rebooting
4204 timeout = 0: wait forever
4205 timeout < 0: reboot immediately
4206 Format: <timeout>
4207
4208 panic_on_taint= [KNL,EARLY]
4209 Bitmask for conditionally calling panic() in add_taint()
4210 Format: <hex>[,nousertaint]
4211 Hexadecimal bitmask representing the set of TAINT flags
4212 that will cause the kernel to panic when add_taint() is
4213 called with any of the flags in this set.
4214 The optional switch "nousertaint" can be utilized to
4215 prevent userspace forced crashes by writing to sysctl
4216 /proc/sys/kernel/tainted any flagset matching with the
4217 bitmask set on panic_on_taint.
4218 See Documentation/admin-guide/tainted-kernels.rst for
4219 extra details on the taint flags that users can pick
4220 to compose the bitmask to assign to panic_on_taint.
4221
4222 panic_on_warn=1 panic() instead of WARN(). Useful to cause kdump
4223 on a WARN().
4224
4225 panic_print= Bitmask for printing system info when panic happens.
4226 User can chose combination of the following bits:
4227 bit 0: print all tasks info
4228 bit 1: print system memory info
4229 bit 2: print timer info
4230 bit 3: print locks info if CONFIG_LOCKDEP is on
4231 bit 4: print ftrace buffer
4232 bit 5: print all printk messages in buffer
4233 bit 6: print all CPUs backtrace (if available in the arch)
4234 bit 7: print only tasks in uninterruptible (blocked) state
4235 *Be aware* that this option may print a _lot_ of lines,
4236 so there are risks of losing older messages in the log.
4237 Use this option carefully, maybe worth to setup a
4238 bigger log buffer with "log_buf_len" along with this.
4239
4240 parkbd.port= [HW] Parallel port number the keyboard adapter is
4241 connected to, default is 0.
4242 Format: <parport#>
4243 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation,
4244 0 for XT, 1 for AT (default is AT).
4245 Format: <mode>
4246
4247 parport= [HW,PPT] Specify parallel ports. 0 disables.
4248 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] }
4249 Use 'auto' to force the driver to use any
4250 IRQ/DMA settings detected (the default is to
4251 ignore detected IRQ/DMA settings because of
4252 possible conflicts). You can specify the base
4253 address, IRQ, and DMA settings; IRQ and DMA
4254 should be numbers, or 'auto' (for using detected
4255 settings on that particular port), or 'nofifo'
4256 (to avoid using a FIFO even if it is detected).
4257 Parallel ports are assigned in the order they
4258 are specified on the command line, starting
4259 with parport0.
4260
4261 parport_init_mode= [HW,PPT]
4262 Configure VIA parallel port to operate in
4263 a specific mode. This is necessary on Pegasos
4264 computer where firmware has no options for setting
4265 up parallel port mode and sets it to spp.
4266 Currently this function knows 686a and 8231 chips.
4267 Format: [spp|ps2|epp|ecp|ecpepp]
4268
4269 pata_legacy.all= [HW,LIBATA]
4270 Format: <int>
4271 Set to non-zero to probe primary and secondary ISA
4272 port ranges on PCI systems where no PCI PATA device
4273 has been found at either range. Disabled by default.
4274
4275 pata_legacy.autospeed= [HW,LIBATA]
4276 Format: <int>
4277 Set to non-zero if a chip is present that snoops speed
4278 changes. Disabled by default.
4279
4280 pata_legacy.ht6560a= [HW,LIBATA]
4281 Format: <int>
4282 Set to 1, 2, or 3 for HT 6560A on the primary channel,
4283 the secondary channel, or both channels respectively.
4284 Disabled by default.
4285
4286 pata_legacy.ht6560b= [HW,LIBATA]
4287 Format: <int>
4288 Set to 1, 2, or 3 for HT 6560B on the primary channel,
4289 the secondary channel, or both channels respectively.
4290 Disabled by default.
4291
4292 pata_legacy.iordy_mask= [HW,LIBATA]
4293 Format: <int>
4294 IORDY enable mask. Set individual bits to allow IORDY
4295 for the respective channel. Bit 0 is for the first
4296 legacy channel handled by this driver, bit 1 is for
4297 the second channel, and so on. The sequence will often
4298 correspond to the primary legacy channel, the secondary
4299 legacy channel, and so on, but the handling of a PCI
4300 bus and the use of other driver options may interfere
4301 with the sequence. By default IORDY is allowed across
4302 all channels.
4303
4304 pata_legacy.opti82c46x= [HW,LIBATA]
4305 Format: <int>
4306 Set to 1, 2, or 3 for Opti 82c611A on the primary
4307 channel, the secondary channel, or both channels
4308 respectively. Disabled by default.
4309
4310 pata_legacy.opti82c611a= [HW,LIBATA]
4311 Format: <int>
4312 Set to 1, 2, or 3 for Opti 82c465MV on the primary
4313 channel, the secondary channel, or both channels
4314 respectively. Disabled by default.
4315
4316 pata_legacy.pio_mask= [HW,LIBATA]
4317 Format: <int>
4318 PIO mode mask for autospeed devices. Set individual
4319 bits to allow the use of the respective PIO modes.
4320 Bit 0 is for mode 0, bit 1 is for mode 1, and so on.
4321 All modes allowed by default.
4322
4323 pata_legacy.probe_all= [HW,LIBATA]
4324 Format: <int>
4325 Set to non-zero to probe tertiary and further ISA
4326 port ranges on PCI systems. Disabled by default.
4327
4328 pata_legacy.probe_mask= [HW,LIBATA]
4329 Format: <int>
4330 Probe mask for legacy ISA PATA ports. Depending on
4331 platform configuration and the use of other driver
4332 options up to 6 legacy ports are supported: 0x1f0,
4333 0x170, 0x1e8, 0x168, 0x1e0, 0x160, however probing
4334 of individual ports can be disabled by setting the
4335 corresponding bits in the mask to 1. Bit 0 is for
4336 the first port in the list above (0x1f0), and so on.
4337 By default all supported ports are probed.
4338
4339 pata_legacy.qdi= [HW,LIBATA]
4340 Format: <int>
4341 Set to non-zero to probe QDI controllers. By default
4342 set to 1 if CONFIG_PATA_QDI_MODULE, 0 otherwise.
4343
4344 pata_legacy.winbond= [HW,LIBATA]
4345 Format: <int>
4346 Set to non-zero to probe Winbond controllers. Use
4347 the standard I/O port (0x130) if 1, otherwise the
4348 value given is the I/O port to use (typically 0x1b0).
4349 By default set to 1 if CONFIG_PATA_WINBOND_VLB_MODULE,
4350 0 otherwise.
4351
4352 pata_platform.pio_mask= [HW,LIBATA]
4353 Format: <int>
4354 Supported PIO mode mask. Set individual bits to allow
4355 the use of the respective PIO modes. Bit 0 is for
4356 mode 0, bit 1 is for mode 1, and so on. Mode 0 only
4357 allowed by default.
4358
4359 pause_on_oops=<int>
4360 Halt all CPUs after the first oops has been printed for
4361 the specified number of seconds. This is to be used if
4362 your oopses keep scrolling off the screen.
4363
4364 pcbit= [HW,ISDN]
4365
4366 pci=option[,option...] [PCI,EARLY] various PCI subsystem options.
4367
4368 Some options herein operate on a specific device
4369 or a set of devices (<pci_dev>). These are
4370 specified in one of the following formats:
4371
4372 [<domain>:]<bus>:<dev>.<func>[/<dev>.<func>]*
4373 pci:<vendor>:<device>[:<subvendor>:<subdevice>]
4374
4375 Note: the first format specifies a PCI
4376 bus/device/function address which may change
4377 if new hardware is inserted, if motherboard
4378 firmware changes, or due to changes caused
4379 by other kernel parameters. If the
4380 domain is left unspecified, it is
4381 taken to be zero. Optionally, a path
4382 to a device through multiple device/function
4383 addresses can be specified after the base
4384 address (this is more robust against
4385 renumbering issues). The second format
4386 selects devices using IDs from the
4387 configuration space which may match multiple
4388 devices in the system.
4389
4390 earlydump dump PCI config space before the kernel
4391 changes anything
4392 off [X86] don't probe for the PCI bus
4393 bios [X86-32] force use of PCI BIOS, don't access
4394 the hardware directly. Use this if your machine
4395 has a non-standard PCI host bridge.
4396 nobios [X86-32] disallow use of PCI BIOS, only direct
4397 hardware access methods are allowed. Use this
4398 if you experience crashes upon bootup and you
4399 suspect they are caused by the BIOS.
4400 conf1 [X86] Force use of PCI Configuration Access
4401 Mechanism 1 (config address in IO port 0xCF8,
4402 data in IO port 0xCFC, both 32-bit).
4403 conf2 [X86] Force use of PCI Configuration Access
4404 Mechanism 2 (IO port 0xCF8 is an 8-bit port for
4405 the function, IO port 0xCFA, also 8-bit, sets
4406 bus number. The config space is then accessed
4407 through ports 0xC000-0xCFFF).
4408 See http://wiki.osdev.org/PCI for more info
4409 on the configuration access mechanisms.
4410 noaer [PCIE] If the PCIEAER kernel config parameter is
4411 enabled, this kernel boot option can be used to
4412 disable the use of PCIE advanced error reporting.
4413 nodomains [PCI] Disable support for multiple PCI
4414 root domains (aka PCI segments, in ACPI-speak).
4415 nommconf [X86] Disable use of MMCONFIG for PCI
4416 Configuration
4417 check_enable_amd_mmconf [X86] check for and enable
4418 properly configured MMIO access to PCI
4419 config space on AMD family 10h CPU
4420 nomsi [MSI] If the PCI_MSI kernel config parameter is
4421 enabled, this kernel boot option can be used to
4422 disable the use of MSI interrupts system-wide.
4423 noioapicquirk [APIC] Disable all boot interrupt quirks.
4424 Safety option to keep boot IRQs enabled. This
4425 should never be necessary.
4426 ioapicreroute [APIC] Enable rerouting of boot IRQs to the
4427 primary IO-APIC for bridges that cannot disable
4428 boot IRQs. This fixes a source of spurious IRQs
4429 when the system masks IRQs.
4430 noioapicreroute [APIC] Disable workaround that uses the
4431 boot IRQ equivalent of an IRQ that connects to
4432 a chipset where boot IRQs cannot be disabled.
4433 The opposite of ioapicreroute.
4434 biosirq [X86-32] Use PCI BIOS calls to get the interrupt
4435 routing table. These calls are known to be buggy
4436 on several machines and they hang the machine
4437 when used, but on other computers it's the only
4438 way to get the interrupt routing table. Try
4439 this option if the kernel is unable to allocate
4440 IRQs or discover secondary PCI buses on your
4441 motherboard.
4442 rom [X86] Assign address space to expansion ROMs.
4443 Use with caution as certain devices share
4444 address decoders between ROMs and other
4445 resources.
4446 norom [X86] Do not assign address space to
4447 expansion ROMs that do not already have
4448 BIOS assigned address ranges.
4449 nobar [X86] Do not assign address space to the
4450 BARs that weren't assigned by the BIOS.
4451 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be
4452 assigned automatically to PCI devices. You can
4453 make the kernel exclude IRQs of your ISA cards
4454 this way.
4455 pirqaddr=0xAAAAA [X86] Specify the physical address
4456 of the PIRQ table (normally generated
4457 by the BIOS) if it is outside the
4458 F0000h-100000h range.
4459 lastbus=N [X86] Scan all buses thru bus #N. Can be
4460 useful if the kernel is unable to find your
4461 secondary buses and you want to tell it
4462 explicitly which ones they are.
4463 assign-busses [X86] Always assign all PCI bus
4464 numbers ourselves, overriding
4465 whatever the firmware may have done.
4466 usepirqmask [X86] Honor the possible IRQ mask stored
4467 in the BIOS $PIR table. This is needed on
4468 some systems with broken BIOSes, notably
4469 some HP Pavilion N5400 and Omnibook XE3
4470 notebooks. This will have no effect if ACPI
4471 IRQ routing is enabled.
4472 noacpi [X86] Do not use ACPI for IRQ routing
4473 or for PCI scanning.
4474 use_crs [X86] Use PCI host bridge window information
4475 from ACPI. On BIOSes from 2008 or later, this
4476 is enabled by default. If you need to use this,
4477 please report a bug.
4478 nocrs [X86] Ignore PCI host bridge windows from ACPI.
4479 If you need to use this, please report a bug.
4480 use_e820 [X86] Use E820 reservations to exclude parts of
4481 PCI host bridge windows. This is a workaround
4482 for BIOS defects in host bridge _CRS methods.
4483 If you need to use this, please report a bug to
4484 <linux-pci@vger.kernel.org>.
4485 no_e820 [X86] Ignore E820 reservations for PCI host
4486 bridge windows. This is the default on modern
4487 hardware. If you need to use this, please report
4488 a bug to <linux-pci@vger.kernel.org>.
4489 routeirq Do IRQ routing for all PCI devices.
4490 This is normally done in pci_enable_device(),
4491 so this option is a temporary workaround
4492 for broken drivers that don't call it.
4493 skip_isa_align [X86] do not align io start addr, so can
4494 handle more pci cards
4495 noearly [X86] Don't do any early type 1 scanning.
4496 This might help on some broken boards which
4497 machine check when some devices' config space
4498 is read. But various workarounds are disabled
4499 and some IOMMU drivers will not work.
4500 bfsort Sort PCI devices into breadth-first order.
4501 This sorting is done to get a device
4502 order compatible with older (<= 2.4) kernels.
4503 nobfsort Don't sort PCI devices into breadth-first order.
4504 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size)
4505 tuning and use the BIOS-configured MPS defaults.
4506 pcie_bus_safe Set every device's MPS to the largest value
4507 supported by all devices below the root complex.
4508 pcie_bus_perf Set device MPS to the largest allowable MPS
4509 based on its parent bus. Also set MRRS (Max
4510 Read Request Size) to the largest supported
4511 value (no larger than the MPS that the device
4512 or bus can support) for best performance.
4513 pcie_bus_peer2peer Set every device's MPS to 128B, which
4514 every device is guaranteed to support. This
4515 configuration allows peer-to-peer DMA between
4516 any pair of devices, possibly at the cost of
4517 reduced performance. This also guarantees
4518 that hot-added devices will work.
4519 cbiosize=nn[KMG] The fixed amount of bus space which is
4520 reserved for the CardBus bridge's IO window.
4521 The default value is 256 bytes.
4522 cbmemsize=nn[KMG] The fixed amount of bus space which is
4523 reserved for the CardBus bridge's memory
4524 window. The default value is 64 megabytes.
4525 resource_alignment=
4526 Format:
4527 [<order of align>@]<pci_dev>[; ...]
4528 Specifies alignment and device to reassign
4529 aligned memory resources. How to
4530 specify the device is described above.
4531 If <order of align> is not specified,
4532 PAGE_SIZE is used as alignment.
4533 A PCI-PCI bridge can be specified if resource
4534 windows need to be expanded.
4535 To specify the alignment for several
4536 instances of a device, the PCI vendor,
4537 device, subvendor, and subdevice may be
4538 specified, e.g., 12@pci:8086:9c22:103c:198f
4539 for 4096-byte alignment.
4540 ecrc= Enable/disable PCIe ECRC (transaction layer
4541 end-to-end CRC checking). Only effective if
4542 OS has native AER control (either granted by
4543 ACPI _OSC or forced via "pcie_ports=native")
4544 bios: Use BIOS/firmware settings. This is the
4545 the default.
4546 off: Turn ECRC off
4547 on: Turn ECRC on.
4548 hpiosize=nn[KMG] The fixed amount of bus space which is
4549 reserved for hotplug bridge's IO window.
4550 Default size is 256 bytes.
4551 hpmmiosize=nn[KMG] The fixed amount of bus space which is
4552 reserved for hotplug bridge's MMIO window.
4553 Default size is 2 megabytes.
4554 hpmmioprefsize=nn[KMG] The fixed amount of bus space which is
4555 reserved for hotplug bridge's MMIO_PREF window.
4556 Default size is 2 megabytes.
4557 hpmemsize=nn[KMG] The fixed amount of bus space which is
4558 reserved for hotplug bridge's MMIO and
4559 MMIO_PREF window.
4560 Default size is 2 megabytes.
4561 hpbussize=nn The minimum amount of additional bus numbers
4562 reserved for buses below a hotplug bridge.
4563 Default is 1.
4564 realloc= Enable/disable reallocating PCI bridge resources
4565 if allocations done by BIOS are too small to
4566 accommodate resources required by all child
4567 devices.
4568 off: Turn realloc off
4569 on: Turn realloc on
4570 realloc same as realloc=on
4571 noari do not use PCIe ARI.
4572 noats [PCIE, Intel-IOMMU, AMD-IOMMU]
4573 do not use PCIe ATS (and IOMMU device IOTLB).
4574 pcie_scan_all Scan all possible PCIe devices. Otherwise we
4575 only look for one device below a PCIe downstream
4576 port.
4577 big_root_window Try to add a big 64bit memory window to the PCIe
4578 root complex on AMD CPUs. Some GFX hardware
4579 can resize a BAR to allow access to all VRAM.
4580 Adding the window is slightly risky (it may
4581 conflict with unreported devices), so this
4582 taints the kernel.
4583 disable_acs_redir=<pci_dev>[; ...]
4584 Specify one or more PCI devices (in the format
4585 specified above) separated by semicolons.
4586 Each device specified will have the PCI ACS
4587 redirect capabilities forced off which will
4588 allow P2P traffic between devices through
4589 bridges without forcing it upstream. Note:
4590 this removes isolation between devices and
4591 may put more devices in an IOMMU group.
4592 force_floating [S390] Force usage of floating interrupts.
4593 nomio [S390] Do not use MIO instructions.
4594 norid [S390] ignore the RID field and force use of
4595 one PCI domain per PCI function
4596
4597 pcie_aspm= [PCIE] Forcibly enable or ignore PCIe Active State Power
4598 Management.
4599 off Don't touch ASPM configuration at all. Leave any
4600 configuration done by firmware unchanged.
4601 force Enable ASPM even on devices that claim not to support it.
4602 WARNING: Forcing ASPM on may cause system lockups.
4603
4604 pcie_ports= [PCIE] PCIe port services handling:
4605 native Use native PCIe services (PME, AER, DPC, PCIe hotplug)
4606 even if the platform doesn't give the OS permission to
4607 use them. This may cause conflicts if the platform
4608 also tries to use these services.
4609 dpc-native Use native PCIe service for DPC only. May
4610 cause conflicts if firmware uses AER or DPC.
4611 compat Disable native PCIe services (PME, AER, DPC, PCIe
4612 hotplug).
4613
4614 pcie_port_pm= [PCIE] PCIe port power management handling:
4615 off Disable power management of all PCIe ports
4616 force Forcibly enable power management of all PCIe ports
4617
4618 pcie_pme= [PCIE,PM] Native PCIe PME signaling options:
4619 nomsi Do not use MSI for native PCIe PME signaling (this makes
4620 all PCIe root ports use INTx for all services).
4621
4622 pcmv= [HW,PCMCIA] BadgePAD 4
4623
4624 pd_ignore_unused
4625 [PM]
4626 Keep all power-domains already enabled by bootloader on,
4627 even if no driver has claimed them. This is useful
4628 for debug and development, but should not be
4629 needed on a platform with proper driver support.
4630
4631 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at
4632 boot time.
4633 Format: { 0 | 1 }
4634 See arch/parisc/kernel/pdc_chassis.c
4635
4636 percpu_alloc= [MM,EARLY]
4637 Select which percpu first chunk allocator to use.
4638 Currently supported values are "embed" and "page".
4639 Archs may support subset or none of the selections.
4640 See comments in mm/percpu.c for details on each
4641 allocator. This parameter is primarily for debugging
4642 and performance comparison.
4643
4644 pirq= [SMP,APIC] Manual mp-table setup
4645 See Documentation/arch/x86/i386/IO-APIC.rst.
4646
4647 plip= [PPT,NET] Parallel port network link
4648 Format: { parport<nr> | timid | 0 }
4649 See also Documentation/admin-guide/parport.rst.
4650
4651 pmtmr= [X86] Manual setup of pmtmr I/O Port.
4652 Override pmtimer IOPort with a hex value.
4653 e.g. pmtmr=0x508
4654
4655 pmu_override= [PPC] Override the PMU.
4656 This option takes over the PMU facility, so it is no
4657 longer usable by perf. Setting this option starts the
4658 PMU counters by setting MMCR0 to 0 (the FC bit is
4659 cleared). If a number is given, then MMCR1 is set to
4660 that number, otherwise (e.g., 'pmu_override=on'), MMCR1
4661 remains 0.
4662
4663 pm_debug_messages [SUSPEND,KNL]
4664 Enable suspend/resume debug messages during boot up.
4665
4666 pnp.debug=1 [PNP]
4667 Enable PNP debug messages (depends on the
4668 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time
4669 via /sys/module/pnp/parameters/debug. We always show
4670 current resource usage; turning this on also shows
4671 possible settings and some assignment information.
4672
4673 pnpacpi= [ACPI]
4674 { off }
4675
4676 pnpbios= [ISAPNP]
4677 { on | off | curr | res | no-curr | no-res }
4678
4679 pnp_reserve_irq=
4680 [ISAPNP] Exclude IRQs for the autoconfiguration
4681
4682 pnp_reserve_dma=
4683 [ISAPNP] Exclude DMAs for the autoconfiguration
4684
4685 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration
4686 Ranges are in pairs (I/O port base and size).
4687
4688 pnp_reserve_mem=
4689 [ISAPNP] Exclude memory regions for the
4690 autoconfiguration.
4691 Ranges are in pairs (memory base and size).
4692
4693 ports= [IP_VS_FTP] IPVS ftp helper module
4694 Default is 21.
4695 Up to 8 (IP_VS_APP_MAX_PORTS) ports
4696 may be specified.
4697 Format: <port>,<port>....
4698
4699 possible_cpus= [SMP,S390,X86]
4700 Format: <unsigned int>
4701 Set the number of possible CPUs, overriding the
4702 regular discovery mechanisms (such as ACPI/FW, etc).
4703
4704 powersave=off [PPC] This option disables power saving features.
4705 It specifically disables cpuidle and sets the
4706 platform machine description specific power_save
4707 function to NULL. On Idle the CPU just reduces
4708 execution priority.
4709
4710 ppc_strict_facility_enable
4711 [PPC,ENABLE] This option catches any kernel floating point,
4712 Altivec, VSX and SPE outside of regions specifically
4713 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()).
4714 There is some performance impact when enabling this.
4715
4716 ppc_tm= [PPC,EARLY]
4717 Format: {"off"}
4718 Disable Hardware Transactional Memory
4719
4720 preempt= [KNL]
4721 Select preemption mode if you have CONFIG_PREEMPT_DYNAMIC
4722 none - Limited to cond_resched() calls
4723 voluntary - Limited to cond_resched() and might_sleep() calls
4724 full - Any section that isn't explicitly preempt disabled
4725 can be preempted anytime.
4726
4727 print-fatal-signals=
4728 [KNL] debug: print fatal signals
4729
4730 If enabled, warn about various signal handling
4731 related application anomalies: too many signals,
4732 too many POSIX.1 timers, fatal signals causing a
4733 coredump - etc.
4734
4735 If you hit the warning due to signal overflow,
4736 you might want to try "ulimit -i unlimited".
4737
4738 default: off.
4739
4740 printk.always_kmsg_dump=
4741 Trigger kmsg_dump for cases other than kernel oops or
4742 panics
4743 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
4744 default: disabled
4745
4746 printk.console_no_auto_verbose=
4747 Disable console loglevel raise on oops, panic
4748 or lockdep-detected issues (only if lock debug is on).
4749 With an exception to setups with low baudrate on
4750 serial console, keeping this 0 is a good choice
4751 in order to provide more debug information.
4752 Format: <bool>
4753 default: 0 (auto_verbose is enabled)
4754
4755 printk.devkmsg={on,off,ratelimit}
4756 Control writing to /dev/kmsg.
4757 on - unlimited logging to /dev/kmsg from userspace
4758 off - logging to /dev/kmsg disabled
4759 ratelimit - ratelimit the logging
4760 Default: ratelimit
4761
4762 printk.time= Show timing data prefixed to each printk message line
4763 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
4764
4765 processor.max_cstate= [HW,ACPI]
4766 Limit processor to maximum C-state
4767 max_cstate=9 overrides any DMI blacklist limit.
4768
4769 processor.nocst [HW,ACPI]
4770 Ignore the _CST method to determine C-states,
4771 instead using the legacy FADT method
4772
4773 profile= [KNL] Enable kernel profiling via /proc/profile
4774 Format: [<profiletype>,]<number>
4775 Param: <profiletype>: "schedule", "sleep", or "kvm"
4776 [defaults to kernel profiling]
4777 Param: "schedule" - profile schedule points.
4778 Param: "sleep" - profile D-state sleeping (millisecs).
4779 Requires CONFIG_SCHEDSTATS
4780 Param: "kvm" - profile VM exits.
4781 Param: <number> - step/bucket size as a power of 2 for
4782 statistical time based profiling.
4783
4784 prompt_ramdisk= [RAM] [Deprecated]
4785
4786 prot_virt= [S390] enable hosting protected virtual machines
4787 isolated from the hypervisor (if hardware supports
4788 that).
4789 Format: <bool>
4790
4791 psi= [KNL] Enable or disable pressure stall information
4792 tracking.
4793 Format: <bool>
4794
4795 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to
4796 probe for; one of (bare|imps|exps|lifebook|any).
4797 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports
4798 per second.
4799 psmouse.resetafter= [HW,MOUSE]
4800 Try to reset the device after so many bad packets
4801 (0 = never).
4802 psmouse.resolution=
4803 [HW,MOUSE] Set desired mouse resolution, in dpi.
4804 psmouse.smartscroll=
4805 [HW,MOUSE] Controls Logitech smartscroll autorepeat.
4806 0 = disabled, 1 = enabled (default).
4807
4808 pstore.backend= Specify the name of the pstore backend to use
4809
4810 pti= [X86-64] Control Page Table Isolation of user and
4811 kernel address spaces. Disabling this feature
4812 removes hardening, but improves performance of
4813 system calls and interrupts.
4814
4815 on - unconditionally enable
4816 off - unconditionally disable
4817 auto - kernel detects whether your CPU model is
4818 vulnerable to issues that PTI mitigates
4819
4820 Not specifying this option is equivalent to pti=auto.
4821
4822 pty.legacy_count=
4823 [KNL] Number of legacy pty's. Overwrites compiled-in
4824 default number.
4825
4826 quiet [KNL,EARLY] Disable most log messages
4827
4828 r128= [HW,DRM]
4829
4830 radix_hcall_invalidate=on [PPC/PSERIES]
4831 Disable RADIX GTSE feature and use hcall for TLB
4832 invalidate.
4833
4834 raid= [HW,RAID]
4835 See Documentation/admin-guide/md.rst.
4836
4837 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes
4838 See Documentation/admin-guide/blockdev/ramdisk.rst.
4839
4840 ramdisk_start= [RAM] RAM disk image start address
4841
4842 random.trust_cpu=off
4843 [KNL,EARLY] Disable trusting the use of the CPU's
4844 random number generator (if available) to
4845 initialize the kernel's RNG.
4846
4847 random.trust_bootloader=off
4848 [KNL,EARLY] Disable trusting the use of the a seed
4849 passed by the bootloader (if available) to
4850 initialize the kernel's RNG.
4851
4852 randomize_kstack_offset=
4853 [KNL,EARLY] Enable or disable kernel stack offset
4854 randomization, which provides roughly 5 bits of
4855 entropy, frustrating memory corruption attacks
4856 that depend on stack address determinism or
4857 cross-syscall address exposures. This is only
4858 available on architectures that have defined
4859 CONFIG_HAVE_ARCH_RANDOMIZE_KSTACK_OFFSET.
4860 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
4861 Default is CONFIG_RANDOMIZE_KSTACK_OFFSET_DEFAULT.
4862
4863 ras=option[,option,...] [KNL] RAS-specific options
4864
4865 cec_disable [X86]
4866 Disable the Correctable Errors Collector,
4867 see CONFIG_RAS_CEC help text.
4868
4869 rcu_nocbs[=cpu-list]
4870 [KNL] The optional argument is a cpu list,
4871 as described above.
4872
4873 In kernels built with CONFIG_RCU_NOCB_CPU=y,
4874 enable the no-callback CPU mode, which prevents
4875 such CPUs' callbacks from being invoked in
4876 softirq context. Invocation of such CPUs' RCU
4877 callbacks will instead be offloaded to "rcuox/N"
4878 kthreads created for that purpose, where "x" is
4879 "p" for RCU-preempt, "s" for RCU-sched, and "g"
4880 for the kthreads that mediate grace periods; and
4881 "N" is the CPU number. This reduces OS jitter on
4882 the offloaded CPUs, which can be useful for HPC
4883 and real-time workloads. It can also improve
4884 energy efficiency for asymmetric multiprocessors.
4885
4886 If a cpulist is passed as an argument, the specified
4887 list of CPUs is set to no-callback mode from boot.
4888
4889 Otherwise, if the '=' sign and the cpulist
4890 arguments are omitted, no CPU will be set to
4891 no-callback mode from boot but the mode may be
4892 toggled at runtime via cpusets.
4893
4894 Note that this argument takes precedence over
4895 the CONFIG_RCU_NOCB_CPU_DEFAULT_ALL option.
4896
4897 rcu_nocb_poll [KNL]
4898 Rather than requiring that offloaded CPUs
4899 (specified by rcu_nocbs= above) explicitly
4900 awaken the corresponding "rcuoN" kthreads,
4901 make these kthreads poll for callbacks.
4902 This improves the real-time response for the
4903 offloaded CPUs by relieving them of the need to
4904 wake up the corresponding kthread, but degrades
4905 energy efficiency by requiring that the kthreads
4906 periodically wake up to do the polling.
4907
4908 rcutree.blimit= [KNL]
4909 Set maximum number of finished RCU callbacks to
4910 process in one batch.
4911
4912 rcutree.do_rcu_barrier= [KNL]
4913 Request a call to rcu_barrier(). This is
4914 throttled so that userspace tests can safely
4915 hammer on the sysfs variable if they so choose.
4916 If triggered before the RCU grace-period machinery
4917 is fully active, this will error out with EAGAIN.
4918
4919 rcutree.dump_tree= [KNL]
4920 Dump the structure of the rcu_node combining tree
4921 out at early boot. This is used for diagnostic
4922 purposes, to verify correct tree setup.
4923
4924 rcutree.gp_cleanup_delay= [KNL]
4925 Set the number of jiffies to delay each step of
4926 RCU grace-period cleanup.
4927
4928 rcutree.gp_init_delay= [KNL]
4929 Set the number of jiffies to delay each step of
4930 RCU grace-period initialization.
4931
4932 rcutree.gp_preinit_delay= [KNL]
4933 Set the number of jiffies to delay each step of
4934 RCU grace-period pre-initialization, that is,
4935 the propagation of recent CPU-hotplug changes up
4936 the rcu_node combining tree.
4937
4938 rcutree.jiffies_till_first_fqs= [KNL]
4939 Set delay from grace-period initialization to
4940 first attempt to force quiescent states.
4941 Units are jiffies, minimum value is zero,
4942 and maximum value is HZ.
4943
4944 rcutree.jiffies_till_next_fqs= [KNL]
4945 Set delay between subsequent attempts to force
4946 quiescent states. Units are jiffies, minimum
4947 value is one, and maximum value is HZ.
4948
4949 rcutree.jiffies_till_sched_qs= [KNL]
4950 Set required age in jiffies for a
4951 given grace period before RCU starts
4952 soliciting quiescent-state help from
4953 rcu_note_context_switch() and cond_resched().
4954 If not specified, the kernel will calculate
4955 a value based on the most recent settings
4956 of rcutree.jiffies_till_first_fqs
4957 and rcutree.jiffies_till_next_fqs.
4958 This calculated value may be viewed in
4959 rcutree.jiffies_to_sched_qs. Any attempt to set
4960 rcutree.jiffies_to_sched_qs will be cheerfully
4961 overwritten.
4962
4963 rcutree.kthread_prio= [KNL,BOOT]
4964 Set the SCHED_FIFO priority of the RCU per-CPU
4965 kthreads (rcuc/N). This value is also used for
4966 the priority of the RCU boost threads (rcub/N)
4967 and for the RCU grace-period kthreads (rcu_bh,
4968 rcu_preempt, and rcu_sched). If RCU_BOOST is
4969 set, valid values are 1-99 and the default is 1
4970 (the least-favored priority). Otherwise, when
4971 RCU_BOOST is not set, valid values are 0-99 and
4972 the default is zero (non-realtime operation).
4973 When RCU_NOCB_CPU is set, also adjust the
4974 priority of NOCB callback kthreads.
4975
4976 rcutree.nocb_nobypass_lim_per_jiffy= [KNL]
4977 On callback-offloaded (rcu_nocbs) CPUs,
4978 RCU reduces the lock contention that would
4979 otherwise be caused by callback floods through
4980 use of the ->nocb_bypass list. However, in the
4981 common non-flooded case, RCU queues directly to
4982 the main ->cblist in order to avoid the extra
4983 overhead of the ->nocb_bypass list and its lock.
4984 But if there are too many callbacks queued during
4985 a single jiffy, RCU pre-queues the callbacks into
4986 the ->nocb_bypass queue. The definition of "too
4987 many" is supplied by this kernel boot parameter.
4988
4989 rcutree.qhimark= [KNL]
4990 Set threshold of queued RCU callbacks beyond which
4991 batch limiting is disabled.
4992
4993 rcutree.qlowmark= [KNL]
4994 Set threshold of queued RCU callbacks below which
4995 batch limiting is re-enabled.
4996
4997 rcutree.qovld= [KNL]
4998 Set threshold of queued RCU callbacks beyond which
4999 RCU's force-quiescent-state scan will aggressively
5000 enlist help from cond_resched() and sched IPIs to
5001 help CPUs more quickly reach quiescent states.
5002 Set to less than zero to make this be set based
5003 on rcutree.qhimark at boot time and to zero to
5004 disable more aggressive help enlistment.
5005
5006 rcutree.rcu_delay_page_cache_fill_msec= [KNL]
5007 Set the page-cache refill delay (in milliseconds)
5008 in response to low-memory conditions. The range
5009 of permitted values is in the range 0:100000.
5010
5011 rcutree.rcu_divisor= [KNL]
5012 Set the shift-right count to use to compute
5013 the callback-invocation batch limit bl from
5014 the number of callbacks queued on this CPU.
5015 The result will be bounded below by the value of
5016 the rcutree.blimit kernel parameter. Every bl
5017 callbacks, the softirq handler will exit in
5018 order to allow the CPU to do other work.
5019
5020 Please note that this callback-invocation batch
5021 limit applies only to non-offloaded callback
5022 invocation. Offloaded callbacks are instead
5023 invoked in the context of an rcuoc kthread, which
5024 scheduler will preempt as it does any other task.
5025
5026 rcutree.rcu_fanout_exact= [KNL]
5027 Disable autobalancing of the rcu_node combining
5028 tree. This is used by rcutorture, and might
5029 possibly be useful for architectures having high
5030 cache-to-cache transfer latencies.
5031
5032 rcutree.rcu_fanout_leaf= [KNL]
5033 Change the number of CPUs assigned to each
5034 leaf rcu_node structure. Useful for very
5035 large systems, which will choose the value 64,
5036 and for NUMA systems with large remote-access
5037 latencies, which will choose a value aligned
5038 with the appropriate hardware boundaries.
5039
5040 rcutree.rcu_min_cached_objs= [KNL]
5041 Minimum number of objects which are cached and
5042 maintained per one CPU. Object size is equal
5043 to PAGE_SIZE. The cache allows to reduce the
5044 pressure to page allocator, also it makes the
5045 whole algorithm to behave better in low memory
5046 condition.
5047
5048 rcutree.rcu_nocb_gp_stride= [KNL]
5049 Set the number of NOCB callback kthreads in
5050 each group, which defaults to the square root
5051 of the number of CPUs. Larger numbers reduce
5052 the wakeup overhead on the global grace-period
5053 kthread, but increases that same overhead on
5054 each group's NOCB grace-period kthread.
5055
5056 rcutree.rcu_kick_kthreads= [KNL]
5057 Cause the grace-period kthread to get an extra
5058 wake_up() if it sleeps three times longer than
5059 it should at force-quiescent-state time.
5060 This wake_up() will be accompanied by a
5061 WARN_ONCE() splat and an ftrace_dump().
5062
5063 rcutree.rcu_resched_ns= [KNL]
5064 Limit the time spend invoking a batch of RCU
5065 callbacks to the specified number of nanoseconds.
5066 By default, this limit is checked only once
5067 every 32 callbacks in order to limit the pain
5068 inflicted by local_clock() overhead.
5069
5070 rcutree.rcu_unlock_delay= [KNL]
5071 In CONFIG_RCU_STRICT_GRACE_PERIOD=y kernels,
5072 this specifies an rcu_read_unlock()-time delay
5073 in microseconds. This defaults to zero.
5074 Larger delays increase the probability of
5075 catching RCU pointer leaks, that is, buggy use
5076 of RCU-protected pointers after the relevant
5077 rcu_read_unlock() has completed.
5078
5079 rcutree.sysrq_rcu= [KNL]
5080 Commandeer a sysrq key to dump out Tree RCU's
5081 rcu_node tree with an eye towards determining
5082 why a new grace period has not yet started.
5083
5084 rcutree.use_softirq= [KNL]
5085 If set to zero, move all RCU_SOFTIRQ processing to
5086 per-CPU rcuc kthreads. Defaults to a non-zero
5087 value, meaning that RCU_SOFTIRQ is used by default.
5088 Specify rcutree.use_softirq=0 to use rcuc kthreads.
5089
5090 But note that CONFIG_PREEMPT_RT=y kernels disable
5091 this kernel boot parameter, forcibly setting it
5092 to zero.
5093
5094 rcutree.enable_rcu_lazy= [KNL]
5095 To save power, batch RCU callbacks and flush after
5096 delay, memory pressure or callback list growing too
5097 big.
5098
5099 rcuscale.gp_async= [KNL]
5100 Measure performance of asynchronous
5101 grace-period primitives such as call_rcu().
5102
5103 rcuscale.gp_async_max= [KNL]
5104 Specify the maximum number of outstanding
5105 callbacks per writer thread. When a writer
5106 thread exceeds this limit, it invokes the
5107 corresponding flavor of rcu_barrier() to allow
5108 previously posted callbacks to drain.
5109
5110 rcuscale.gp_exp= [KNL]
5111 Measure performance of expedited synchronous
5112 grace-period primitives.
5113
5114 rcuscale.holdoff= [KNL]
5115 Set test-start holdoff period. The purpose of
5116 this parameter is to delay the start of the
5117 test until boot completes in order to avoid
5118 interference.
5119
5120 rcuscale.kfree_by_call_rcu= [KNL]
5121 In kernels built with CONFIG_RCU_LAZY=y, test
5122 call_rcu() instead of kfree_rcu().
5123
5124 rcuscale.kfree_mult= [KNL]
5125 Instead of allocating an object of size kfree_obj,
5126 allocate one of kfree_mult * sizeof(kfree_obj).
5127 Defaults to 1.
5128
5129 rcuscale.kfree_rcu_test= [KNL]
5130 Set to measure performance of kfree_rcu() flooding.
5131
5132 rcuscale.kfree_rcu_test_double= [KNL]
5133 Test the double-argument variant of kfree_rcu().
5134 If this parameter has the same value as
5135 rcuscale.kfree_rcu_test_single, both the single-
5136 and double-argument variants are tested.
5137
5138 rcuscale.kfree_rcu_test_single= [KNL]
5139 Test the single-argument variant of kfree_rcu().
5140 If this parameter has the same value as
5141 rcuscale.kfree_rcu_test_double, both the single-
5142 and double-argument variants are tested.
5143
5144 rcuscale.kfree_nthreads= [KNL]
5145 The number of threads running loops of kfree_rcu().
5146
5147 rcuscale.kfree_alloc_num= [KNL]
5148 Number of allocations and frees done in an iteration.
5149
5150 rcuscale.kfree_loops= [KNL]
5151 Number of loops doing rcuscale.kfree_alloc_num number
5152 of allocations and frees.
5153
5154 rcuscale.minruntime= [KNL]
5155 Set the minimum test run time in seconds. This
5156 does not affect the data-collection interval,
5157 but instead allows better measurement of things
5158 like CPU consumption.
5159
5160 rcuscale.nreaders= [KNL]
5161 Set number of RCU readers. The value -1 selects
5162 N, where N is the number of CPUs. A value
5163 "n" less than -1 selects N-n+1, where N is again
5164 the number of CPUs. For example, -2 selects N
5165 (the number of CPUs), -3 selects N+1, and so on.
5166 A value of "n" less than or equal to -N selects
5167 a single reader.
5168
5169 rcuscale.nwriters= [KNL]
5170 Set number of RCU writers. The values operate
5171 the same as for rcuscale.nreaders.
5172 N, where N is the number of CPUs
5173
5174 rcuscale.scale_type= [KNL]
5175 Specify the RCU implementation to test.
5176
5177 rcuscale.shutdown= [KNL]
5178 Shut the system down after performance tests
5179 complete. This is useful for hands-off automated
5180 testing.
5181
5182 rcuscale.verbose= [KNL]
5183 Enable additional printk() statements.
5184
5185 rcuscale.writer_holdoff= [KNL]
5186 Write-side holdoff between grace periods,
5187 in microseconds. The default of zero says
5188 no holdoff.
5189
5190 rcuscale.writer_holdoff_jiffies= [KNL]
5191 Additional write-side holdoff between grace
5192 periods, but in jiffies. The default of zero
5193 says no holdoff.
5194
5195 rcutorture.fqs_duration= [KNL]
5196 Set duration of force_quiescent_state bursts
5197 in microseconds.
5198
5199 rcutorture.fqs_holdoff= [KNL]
5200 Set holdoff time within force_quiescent_state bursts
5201 in microseconds.
5202
5203 rcutorture.fqs_stutter= [KNL]
5204 Set wait time between force_quiescent_state bursts
5205 in seconds.
5206
5207 rcutorture.fwd_progress= [KNL]
5208 Specifies the number of kthreads to be used
5209 for RCU grace-period forward-progress testing
5210 for the types of RCU supporting this notion.
5211 Defaults to 1 kthread, values less than zero or
5212 greater than the number of CPUs cause the number
5213 of CPUs to be used.
5214
5215 rcutorture.fwd_progress_div= [KNL]
5216 Specify the fraction of a CPU-stall-warning
5217 period to do tight-loop forward-progress testing.
5218
5219 rcutorture.fwd_progress_holdoff= [KNL]
5220 Number of seconds to wait between successive
5221 forward-progress tests.
5222
5223 rcutorture.fwd_progress_need_resched= [KNL]
5224 Enclose cond_resched() calls within checks for
5225 need_resched() during tight-loop forward-progress
5226 testing.
5227
5228 rcutorture.gp_cond= [KNL]
5229 Use conditional/asynchronous update-side
5230 primitives, if available.
5231
5232 rcutorture.gp_exp= [KNL]
5233 Use expedited update-side primitives, if available.
5234
5235 rcutorture.gp_normal= [KNL]
5236 Use normal (non-expedited) asynchronous
5237 update-side primitives, if available.
5238
5239 rcutorture.gp_sync= [KNL]
5240 Use normal (non-expedited) synchronous
5241 update-side primitives, if available. If all
5242 of rcutorture.gp_cond=, rcutorture.gp_exp=,
5243 rcutorture.gp_normal=, and rcutorture.gp_sync=
5244 are zero, rcutorture acts as if is interpreted
5245 they are all non-zero.
5246
5247 rcutorture.irqreader= [KNL]
5248 Run RCU readers from irq handlers, or, more
5249 accurately, from a timer handler. Not all RCU
5250 flavors take kindly to this sort of thing.
5251
5252 rcutorture.leakpointer= [KNL]
5253 Leak an RCU-protected pointer out of the reader.
5254 This can of course result in splats, and is
5255 intended to test the ability of things like
5256 CONFIG_RCU_STRICT_GRACE_PERIOD=y to detect
5257 such leaks.
5258
5259 rcutorture.n_barrier_cbs= [KNL]
5260 Set callbacks/threads for rcu_barrier() testing.
5261
5262 rcutorture.nfakewriters= [KNL]
5263 Set number of concurrent RCU writers. These just
5264 stress RCU, they don't participate in the actual
5265 test, hence the "fake".
5266
5267 rcutorture.nocbs_nthreads= [KNL]
5268 Set number of RCU callback-offload togglers.
5269 Zero (the default) disables toggling.
5270
5271 rcutorture.nocbs_toggle= [KNL]
5272 Set the delay in milliseconds between successive
5273 callback-offload toggling attempts.
5274
5275 rcutorture.nreaders= [KNL]
5276 Set number of RCU readers. The value -1 selects
5277 N-1, where N is the number of CPUs. A value
5278 "n" less than -1 selects N-n-2, where N is again
5279 the number of CPUs. For example, -2 selects N
5280 (the number of CPUs), -3 selects N+1, and so on.
5281
5282 rcutorture.object_debug= [KNL]
5283 Enable debug-object double-call_rcu() testing.
5284
5285 rcutorture.onoff_holdoff= [KNL]
5286 Set time (s) after boot for CPU-hotplug testing.
5287
5288 rcutorture.onoff_interval= [KNL]
5289 Set time (jiffies) between CPU-hotplug operations,
5290 or zero to disable CPU-hotplug testing.
5291
5292 rcutorture.read_exit= [KNL]
5293 Set the number of read-then-exit kthreads used
5294 to test the interaction of RCU updaters and
5295 task-exit processing.
5296
5297 rcutorture.read_exit_burst= [KNL]
5298 The number of times in a given read-then-exit
5299 episode that a set of read-then-exit kthreads
5300 is spawned.
5301
5302 rcutorture.read_exit_delay= [KNL]
5303 The delay, in seconds, between successive
5304 read-then-exit testing episodes.
5305
5306 rcutorture.shuffle_interval= [KNL]
5307 Set task-shuffle interval (s). Shuffling tasks
5308 allows some CPUs to go into dyntick-idle mode
5309 during the rcutorture test.
5310
5311 rcutorture.shutdown_secs= [KNL]
5312 Set time (s) after boot system shutdown. This
5313 is useful for hands-off automated testing.
5314
5315 rcutorture.stall_cpu= [KNL]
5316 Duration of CPU stall (s) to test RCU CPU stall
5317 warnings, zero to disable.
5318
5319 rcutorture.stall_cpu_block= [KNL]
5320 Sleep while stalling if set. This will result
5321 in warnings from preemptible RCU in addition to
5322 any other stall-related activity. Note that
5323 in kernels built with CONFIG_PREEMPTION=n and
5324 CONFIG_PREEMPT_COUNT=y, this parameter will
5325 cause the CPU to pass through a quiescent state.
5326 Given CONFIG_PREEMPTION=n, this will suppress
5327 RCU CPU stall warnings, but will instead result
5328 in scheduling-while-atomic splats.
5329
5330 Use of this module parameter results in splats.
5331
5332
5333 rcutorture.stall_cpu_holdoff= [KNL]
5334 Time to wait (s) after boot before inducing stall.
5335
5336 rcutorture.stall_cpu_irqsoff= [KNL]
5337 Disable interrupts while stalling if set.
5338
5339 rcutorture.stall_gp_kthread= [KNL]
5340 Duration (s) of forced sleep within RCU
5341 grace-period kthread to test RCU CPU stall
5342 warnings, zero to disable. If both stall_cpu
5343 and stall_gp_kthread are specified, the
5344 kthread is starved first, then the CPU.
5345
5346 rcutorture.stat_interval= [KNL]
5347 Time (s) between statistics printk()s.
5348
5349 rcutorture.stutter= [KNL]
5350 Time (s) to stutter testing, for example, specifying
5351 five seconds causes the test to run for five seconds,
5352 wait for five seconds, and so on. This tests RCU's
5353 ability to transition abruptly to and from idle.
5354
5355 rcutorture.test_boost= [KNL]
5356 Test RCU priority boosting? 0=no, 1=maybe, 2=yes.
5357 "Maybe" means test if the RCU implementation
5358 under test support RCU priority boosting.
5359
5360 rcutorture.test_boost_duration= [KNL]
5361 Duration (s) of each individual boost test.
5362
5363 rcutorture.test_boost_interval= [KNL]
5364 Interval (s) between each boost test.
5365
5366 rcutorture.test_no_idle_hz= [KNL]
5367 Test RCU's dyntick-idle handling. See also the
5368 rcutorture.shuffle_interval parameter.
5369
5370 rcutorture.torture_type= [KNL]
5371 Specify the RCU implementation to test.
5372
5373 rcutorture.verbose= [KNL]
5374 Enable additional printk() statements.
5375
5376 rcupdate.rcu_cpu_stall_ftrace_dump= [KNL]
5377 Dump ftrace buffer after reporting RCU CPU
5378 stall warning.
5379
5380 rcupdate.rcu_cpu_stall_notifiers= [KNL]
5381 Provide RCU CPU stall notifiers, but see the
5382 warnings in the RCU_CPU_STALL_NOTIFIER Kconfig
5383 option's help text. TL;DR: You almost certainly
5384 do not want rcupdate.rcu_cpu_stall_notifiers.
5385
5386 rcupdate.rcu_cpu_stall_suppress= [KNL]
5387 Suppress RCU CPU stall warning messages.
5388
5389 rcupdate.rcu_cpu_stall_suppress_at_boot= [KNL]
5390 Suppress RCU CPU stall warning messages and
5391 rcutorture writer stall warnings that occur
5392 during early boot, that is, during the time
5393 before the init task is spawned.
5394
5395 rcupdate.rcu_cpu_stall_timeout= [KNL]
5396 Set timeout for RCU CPU stall warning messages.
5397 The value is in seconds and the maximum allowed
5398 value is 300 seconds.
5399
5400 rcupdate.rcu_exp_cpu_stall_timeout= [KNL]
5401 Set timeout for expedited RCU CPU stall warning
5402 messages. The value is in milliseconds
5403 and the maximum allowed value is 21000
5404 milliseconds. Please note that this value is
5405 adjusted to an arch timer tick resolution.
5406 Setting this to zero causes the value from
5407 rcupdate.rcu_cpu_stall_timeout to be used (after
5408 conversion from seconds to milliseconds).
5409
5410 rcupdate.rcu_cpu_stall_cputime= [KNL]
5411 Provide statistics on the cputime and count of
5412 interrupts and tasks during the sampling period. For
5413 multiple continuous RCU stalls, all sampling periods
5414 begin at half of the first RCU stall timeout.
5415
5416 rcupdate.rcu_exp_stall_task_details= [KNL]
5417 Print stack dumps of any tasks blocking the
5418 current expedited RCU grace period during an
5419 expedited RCU CPU stall warning.
5420
5421 rcupdate.rcu_expedited= [KNL]
5422 Use expedited grace-period primitives, for
5423 example, synchronize_rcu_expedited() instead
5424 of synchronize_rcu(). This reduces latency,
5425 but can increase CPU utilization, degrade
5426 real-time latency, and degrade energy efficiency.
5427 No effect on CONFIG_TINY_RCU kernels.
5428
5429 rcupdate.rcu_normal= [KNL]
5430 Use only normal grace-period primitives,
5431 for example, synchronize_rcu() instead of
5432 synchronize_rcu_expedited(). This improves
5433 real-time latency, CPU utilization, and
5434 energy efficiency, but can expose users to
5435 increased grace-period latency. This parameter
5436 overrides rcupdate.rcu_expedited. No effect on
5437 CONFIG_TINY_RCU kernels.
5438
5439 rcupdate.rcu_normal_after_boot= [KNL]
5440 Once boot has completed (that is, after
5441 rcu_end_inkernel_boot() has been invoked), use
5442 only normal grace-period primitives. No effect
5443 on CONFIG_TINY_RCU kernels.
5444
5445 But note that CONFIG_PREEMPT_RT=y kernels enables
5446 this kernel boot parameter, forcibly setting
5447 it to the value one, that is, converting any
5448 post-boot attempt at an expedited RCU grace
5449 period to instead use normal non-expedited
5450 grace-period processing.
5451
5452 rcupdate.rcu_task_collapse_lim= [KNL]
5453 Set the maximum number of callbacks present
5454 at the beginning of a grace period that allows
5455 the RCU Tasks flavors to collapse back to using
5456 a single callback queue. This switching only
5457 occurs when rcupdate.rcu_task_enqueue_lim is
5458 set to the default value of -1.
5459
5460 rcupdate.rcu_task_contend_lim= [KNL]
5461 Set the minimum number of callback-queuing-time
5462 lock-contention events per jiffy required to
5463 cause the RCU Tasks flavors to switch to per-CPU
5464 callback queuing. This switching only occurs
5465 when rcupdate.rcu_task_enqueue_lim is set to
5466 the default value of -1.
5467
5468 rcupdate.rcu_task_enqueue_lim= [KNL]
5469 Set the number of callback queues to use for the
5470 RCU Tasks family of RCU flavors. The default
5471 of -1 allows this to be automatically (and
5472 dynamically) adjusted. This parameter is intended
5473 for use in testing.
5474
5475 rcupdate.rcu_task_ipi_delay= [KNL]
5476 Set time in jiffies during which RCU tasks will
5477 avoid sending IPIs, starting with the beginning
5478 of a given grace period. Setting a large
5479 number avoids disturbing real-time workloads,
5480 but lengthens grace periods.
5481
5482 rcupdate.rcu_task_lazy_lim= [KNL]
5483 Number of callbacks on a given CPU that will
5484 cancel laziness on that CPU. Use -1 to disable
5485 cancellation of laziness, but be advised that
5486 doing so increases the danger of OOM due to
5487 callback flooding.
5488
5489 rcupdate.rcu_task_stall_info= [KNL]
5490 Set initial timeout in jiffies for RCU task stall
5491 informational messages, which give some indication
5492 of the problem for those not patient enough to
5493 wait for ten minutes. Informational messages are
5494 only printed prior to the stall-warning message
5495 for a given grace period. Disable with a value
5496 less than or equal to zero. Defaults to ten
5497 seconds. A change in value does not take effect
5498 until the beginning of the next grace period.
5499
5500 rcupdate.rcu_task_stall_info_mult= [KNL]
5501 Multiplier for time interval between successive
5502 RCU task stall informational messages for a given
5503 RCU tasks grace period. This value is clamped
5504 to one through ten, inclusive. It defaults to
5505 the value three, so that the first informational
5506 message is printed 10 seconds into the grace
5507 period, the second at 40 seconds, the third at
5508 160 seconds, and then the stall warning at 600
5509 seconds would prevent a fourth at 640 seconds.
5510
5511 rcupdate.rcu_task_stall_timeout= [KNL]
5512 Set timeout in jiffies for RCU task stall
5513 warning messages. Disable with a value less
5514 than or equal to zero. Defaults to ten minutes.
5515 A change in value does not take effect until
5516 the beginning of the next grace period.
5517
5518 rcupdate.rcu_tasks_lazy_ms= [KNL]
5519 Set timeout in milliseconds RCU Tasks asynchronous
5520 callback batching for call_rcu_tasks().
5521 A negative value will take the default. A value
5522 of zero will disable batching. Batching is
5523 always disabled for synchronize_rcu_tasks().
5524
5525 rcupdate.rcu_tasks_rude_lazy_ms= [KNL]
5526 Set timeout in milliseconds RCU Tasks
5527 Rude asynchronous callback batching for
5528 call_rcu_tasks_rude(). A negative value
5529 will take the default. A value of zero will
5530 disable batching. Batching is always disabled
5531 for synchronize_rcu_tasks_rude().
5532
5533 rcupdate.rcu_tasks_trace_lazy_ms= [KNL]
5534 Set timeout in milliseconds RCU Tasks
5535 Trace asynchronous callback batching for
5536 call_rcu_tasks_trace(). A negative value
5537 will take the default. A value of zero will
5538 disable batching. Batching is always disabled
5539 for synchronize_rcu_tasks_trace().
5540
5541 rcupdate.rcu_self_test= [KNL]
5542 Run the RCU early boot self tests
5543
5544 rdinit= [KNL]
5545 Format: <full_path>
5546 Run specified binary instead of /init from the ramdisk,
5547 used for early userspace startup. See initrd.
5548
5549 rdrand= [X86,EARLY]
5550 force - Override the decision by the kernel to hide the
5551 advertisement of RDRAND support (this affects
5552 certain AMD processors because of buggy BIOS
5553 support, specifically around the suspend/resume
5554 path).
5555
5556 rdt= [HW,X86,RDT]
5557 Turn on/off individual RDT features. List is:
5558 cmt, mbmtotal, mbmlocal, l3cat, l3cdp, l2cat, l2cdp,
5559 mba, smba, bmec.
5560 E.g. to turn on cmt and turn off mba use:
5561 rdt=cmt,!mba
5562
5563 reboot= [KNL]
5564 Format (x86 or x86_64):
5565 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] | d[efault] \
5566 [[,]s[mp]#### \
5567 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \
5568 [[,]f[orce]
5569 Where reboot_mode is one of warm (soft) or cold (hard) or gpio
5570 (prefix with 'panic_' to set mode for panic
5571 reboot only),
5572 reboot_type is one of bios, acpi, kbd, triple, efi, or pci,
5573 reboot_force is either force or not specified,
5574 reboot_cpu is s[mp]#### with #### being the processor
5575 to be used for rebooting.
5576
5577 refscale.holdoff= [KNL]
5578 Set test-start holdoff period. The purpose of
5579 this parameter is to delay the start of the
5580 test until boot completes in order to avoid
5581 interference.
5582
5583 refscale.lookup_instances= [KNL]
5584 Number of data elements to use for the forms of
5585 SLAB_TYPESAFE_BY_RCU testing. A negative number
5586 is negated and multiplied by nr_cpu_ids, while
5587 zero specifies nr_cpu_ids.
5588
5589 refscale.loops= [KNL]
5590 Set the number of loops over the synchronization
5591 primitive under test. Increasing this number
5592 reduces noise due to loop start/end overhead,
5593 but the default has already reduced the per-pass
5594 noise to a handful of picoseconds on ca. 2020
5595 x86 laptops.
5596
5597 refscale.nreaders= [KNL]
5598 Set number of readers. The default value of -1
5599 selects N, where N is roughly 75% of the number
5600 of CPUs. A value of zero is an interesting choice.
5601
5602 refscale.nruns= [KNL]
5603 Set number of runs, each of which is dumped onto
5604 the console log.
5605
5606 refscale.readdelay= [KNL]
5607 Set the read-side critical-section duration,
5608 measured in microseconds.
5609
5610 refscale.scale_type= [KNL]
5611 Specify the read-protection implementation to test.
5612
5613 refscale.shutdown= [KNL]
5614 Shut down the system at the end of the performance
5615 test. This defaults to 1 (shut it down) when
5616 refscale is built into the kernel and to 0 (leave
5617 it running) when refscale is built as a module.
5618
5619 refscale.verbose= [KNL]
5620 Enable additional printk() statements.
5621
5622 refscale.verbose_batched= [KNL]
5623 Batch the additional printk() statements. If zero
5624 (the default) or negative, print everything. Otherwise,
5625 print every Nth verbose statement, where N is the value
5626 specified.
5627
5628 regulator_ignore_unused
5629 [REGULATOR]
5630 Prevents regulator framework from disabling regulators
5631 that are unused, due no driver claiming them. This may
5632 be useful for debug and development, but should not be
5633 needed on a platform with proper driver support.
5634
5635 relax_domain_level=
5636 [KNL, SMP] Set scheduler's default relax_domain_level.
5637 See Documentation/admin-guide/cgroup-v1/cpusets.rst.
5638
5639 reserve= [KNL,BUGS] Force kernel to ignore I/O ports or memory
5640 Format: <base1>,<size1>[,<base2>,<size2>,...]
5641 Reserve I/O ports or memory so the kernel won't use
5642 them. If <base> is less than 0x10000, the region
5643 is assumed to be I/O ports; otherwise it is memory.
5644
5645 reservetop= [X86-32,EARLY]
5646 Format: nn[KMG]
5647 Reserves a hole at the top of the kernel virtual
5648 address space.
5649
5650 reset_devices [KNL] Force drivers to reset the underlying device
5651 during initialization.
5652
5653 resume= [SWSUSP]
5654 Specify the partition device for software suspend
5655 Format:
5656 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>}
5657
5658 resume_offset= [SWSUSP]
5659 Specify the offset from the beginning of the partition
5660 given by "resume=" at which the swap header is located,
5661 in <PAGE_SIZE> units (needed only for swap files).
5662 See Documentation/power/swsusp-and-swap-files.rst
5663
5664 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to
5665 read the resume files
5666
5667 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up.
5668 Useful for devices that are detected asynchronously
5669 (e.g. USB and MMC devices).
5670
5671 retain_initrd [RAM] Keep initrd memory after extraction. After boot, it will
5672 be accessible via /sys/firmware/initrd.
5673
5674 retbleed= [X86] Control mitigation of RETBleed (Arbitrary
5675 Speculative Code Execution with Return Instructions)
5676 vulnerability.
5677
5678 AMD-based UNRET and IBPB mitigations alone do not stop
5679 sibling threads from influencing the predictions of other
5680 sibling threads. For that reason, STIBP is used on pro-
5681 cessors that support it, and mitigate SMT on processors
5682 that don't.
5683
5684 off - no mitigation
5685 auto - automatically select a migitation
5686 auto,nosmt - automatically select a mitigation,
5687 disabling SMT if necessary for
5688 the full mitigation (only on Zen1
5689 and older without STIBP).
5690 ibpb - On AMD, mitigate short speculation
5691 windows on basic block boundaries too.
5692 Safe, highest perf impact. It also
5693 enables STIBP if present. Not suitable
5694 on Intel.
5695 ibpb,nosmt - Like "ibpb" above but will disable SMT
5696 when STIBP is not available. This is
5697 the alternative for systems which do not
5698 have STIBP.
5699 unret - Force enable untrained return thunks,
5700 only effective on AMD f15h-f17h based
5701 systems.
5702 unret,nosmt - Like unret, but will disable SMT when STIBP
5703 is not available. This is the alternative for
5704 systems which do not have STIBP.
5705
5706 Selecting 'auto' will choose a mitigation method at run
5707 time according to the CPU.
5708
5709 Not specifying this option is equivalent to retbleed=auto.
5710
5711 rfkill.default_state=
5712 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm,
5713 etc. communication is blocked by default.
5714 1 Unblocked.
5715
5716 rfkill.master_switch_mode=
5717 0 The "airplane mode" button does nothing.
5718 1 The "airplane mode" button toggles between everything
5719 blocked and the previous configuration.
5720 2 The "airplane mode" button toggles between everything
5721 blocked and everything unblocked.
5722
5723 rhash_entries= [KNL,NET]
5724 Set number of hash buckets for route cache
5725
5726 ring3mwait=disable
5727 [KNL] Disable ring 3 MONITOR/MWAIT feature on supported
5728 CPUs.
5729
5730 riscv_isa_fallback [RISCV,EARLY]
5731 When CONFIG_RISCV_ISA_FALLBACK is not enabled, permit
5732 falling back to detecting extension support by parsing
5733 "riscv,isa" property on devicetree systems when the
5734 replacement properties are not found. See the Kconfig
5735 entry for RISCV_ISA_FALLBACK.
5736
5737 ro [KNL] Mount root device read-only on boot
5738
5739 rodata= [KNL,EARLY]
5740 on Mark read-only kernel memory as read-only (default).
5741 off Leave read-only kernel memory writable for debugging.
5742 full Mark read-only kernel memory and aliases as read-only
5743 [arm64]
5744
5745 rockchip.usb_uart
5746 [EARLY]
5747 Enable the uart passthrough on the designated usb port
5748 on Rockchip SoCs. When active, the signals of the
5749 debug-uart get routed to the D+ and D- pins of the usb
5750 port and the regular usb controller gets disabled.
5751
5752 root= [KNL] Root filesystem
5753 Usually this a a block device specifier of some kind,
5754 see the early_lookup_bdev comment in
5755 block/early-lookup.c for details.
5756 Alternatively this can be "ram" for the legacy initial
5757 ramdisk, "nfs" and "cifs" for root on a network file
5758 system, or "mtd" and "ubi" for mounting from raw flash.
5759
5760 rootdelay= [KNL] Delay (in seconds) to pause before attempting to
5761 mount the root filesystem
5762
5763 rootflags= [KNL] Set root filesystem mount option string
5764
5765 rootfstype= [KNL] Set root filesystem type
5766
5767 rootwait [KNL] Wait (indefinitely) for root device to show up.
5768 Useful for devices that are detected asynchronously
5769 (e.g. USB and MMC devices).
5770
5771 rootwait= [KNL] Maximum time (in seconds) to wait for root device
5772 to show up before attempting to mount the root
5773 filesystem.
5774
5775 rproc_mem=nn[KMG][@address]
5776 [KNL,ARM,CMA] Remoteproc physical memory block.
5777 Memory area to be used by remote processor image,
5778 managed by CMA.
5779
5780 rw [KNL] Mount root device read-write on boot
5781
5782 S [KNL] Run init in single mode
5783
5784 s390_iommu= [HW,S390]
5785 Set s390 IOTLB flushing mode
5786 strict
5787 With strict flushing every unmap operation will result
5788 in an IOTLB flush. Default is lazy flushing before
5789 reuse, which is faster. Deprecated, equivalent to
5790 iommu.strict=1.
5791
5792 s390_iommu_aperture= [KNL,S390]
5793 Specifies the size of the per device DMA address space
5794 accessible through the DMA and IOMMU APIs as a decimal
5795 factor of the size of main memory.
5796 The default is 1 meaning that one can concurrently use
5797 as many DMA addresses as physical memory is installed,
5798 if supported by hardware, and thus map all of memory
5799 once. With a value of 2 one can map all of memory twice
5800 and so on. As a special case a factor of 0 imposes no
5801 restrictions other than those given by hardware at the
5802 cost of significant additional memory use for tables.
5803
5804 sa1100ir [NET]
5805 See drivers/net/irda/sa1100_ir.c.
5806
5807 sched_verbose [KNL,EARLY] Enables verbose scheduler debug messages.
5808
5809 schedstats= [KNL,X86] Enable or disable scheduled statistics.
5810 Allowed values are enable and disable. This feature
5811 incurs a small amount of overhead in the scheduler
5812 but is useful for debugging and performance tuning.
5813
5814 sched_thermal_decay_shift=
5815 [KNL, SMP] Set a decay shift for scheduler thermal
5816 pressure signal. Thermal pressure signal follows the
5817 default decay period of other scheduler pelt
5818 signals(usually 32 ms but configurable). Setting
5819 sched_thermal_decay_shift will left shift the decay
5820 period for the thermal pressure signal by the shift
5821 value.
5822 i.e. with the default pelt decay period of 32 ms
5823 sched_thermal_decay_shift thermal pressure decay pr
5824 1 64 ms
5825 2 128 ms
5826 and so on.
5827 Format: integer between 0 and 10
5828 Default is 0.
5829
5830 scftorture.holdoff= [KNL]
5831 Number of seconds to hold off before starting
5832 test. Defaults to zero for module insertion and
5833 to 10 seconds for built-in smp_call_function()
5834 tests.
5835
5836 scftorture.longwait= [KNL]
5837 Request ridiculously long waits randomly selected
5838 up to the chosen limit in seconds. Zero (the
5839 default) disables this feature. Please note
5840 that requesting even small non-zero numbers of
5841 seconds can result in RCU CPU stall warnings,
5842 softlockup complaints, and so on.
5843
5844 scftorture.nthreads= [KNL]
5845 Number of kthreads to spawn to invoke the
5846 smp_call_function() family of functions.
5847 The default of -1 specifies a number of kthreads
5848 equal to the number of CPUs.
5849
5850 scftorture.onoff_holdoff= [KNL]
5851 Number seconds to wait after the start of the
5852 test before initiating CPU-hotplug operations.
5853
5854 scftorture.onoff_interval= [KNL]
5855 Number seconds to wait between successive
5856 CPU-hotplug operations. Specifying zero (which
5857 is the default) disables CPU-hotplug operations.
5858
5859 scftorture.shutdown_secs= [KNL]
5860 The number of seconds following the start of the
5861 test after which to shut down the system. The
5862 default of zero avoids shutting down the system.
5863 Non-zero values are useful for automated tests.
5864
5865 scftorture.stat_interval= [KNL]
5866 The number of seconds between outputting the
5867 current test statistics to the console. A value
5868 of zero disables statistics output.
5869
5870 scftorture.stutter_cpus= [KNL]
5871 The number of jiffies to wait between each change
5872 to the set of CPUs under test.
5873
5874 scftorture.use_cpus_read_lock= [KNL]
5875 Use use_cpus_read_lock() instead of the default
5876 preempt_disable() to disable CPU hotplug
5877 while invoking one of the smp_call_function*()
5878 functions.
5879
5880 scftorture.verbose= [KNL]
5881 Enable additional printk() statements.
5882
5883 scftorture.weight_single= [KNL]
5884 The probability weighting to use for the
5885 smp_call_function_single() function with a zero
5886 "wait" parameter. A value of -1 selects the
5887 default if all other weights are -1. However,
5888 if at least one weight has some other value, a
5889 value of -1 will instead select a weight of zero.
5890
5891 scftorture.weight_single_wait= [KNL]
5892 The probability weighting to use for the
5893 smp_call_function_single() function with a
5894 non-zero "wait" parameter. See weight_single.
5895
5896 scftorture.weight_many= [KNL]
5897 The probability weighting to use for the
5898 smp_call_function_many() function with a zero
5899 "wait" parameter. See weight_single.
5900 Note well that setting a high probability for
5901 this weighting can place serious IPI load
5902 on the system.
5903
5904 scftorture.weight_many_wait= [KNL]
5905 The probability weighting to use for the
5906 smp_call_function_many() function with a
5907 non-zero "wait" parameter. See weight_single
5908 and weight_many.
5909
5910 scftorture.weight_all= [KNL]
5911 The probability weighting to use for the
5912 smp_call_function_all() function with a zero
5913 "wait" parameter. See weight_single and
5914 weight_many.
5915
5916 scftorture.weight_all_wait= [KNL]
5917 The probability weighting to use for the
5918 smp_call_function_all() function with a
5919 non-zero "wait" parameter. See weight_single
5920 and weight_many.
5921
5922 skew_tick= [KNL,EARLY] Offset the periodic timer tick per cpu to mitigate
5923 xtime_lock contention on larger systems, and/or RCU lock
5924 contention on all systems with CONFIG_MAXSMP set.
5925 Format: { "0" | "1" }
5926 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1"
5927 1 -- enable.
5928 Note: increases power consumption, thus should only be
5929 enabled if running jitter sensitive (HPC/RT) workloads.
5930
5931 security= [SECURITY] Choose a legacy "major" security module to
5932 enable at boot. This has been deprecated by the
5933 "lsm=" parameter.
5934
5935 selinux= [SELINUX] Disable or enable SELinux at boot time.
5936 Format: { "0" | "1" }
5937 See security/selinux/Kconfig help text.
5938 0 -- disable.
5939 1 -- enable.
5940 Default value is 1.
5941
5942 serialnumber [BUGS=X86-32]
5943
5944 sev=option[,option...] [X86-64] See Documentation/arch/x86/x86_64/boot-options.rst
5945
5946 shapers= [NET]
5947 Maximal number of shapers.
5948
5949 show_lapic= [APIC,X86] Advanced Programmable Interrupt Controller
5950 Limit apic dumping. The parameter defines the maximal
5951 number of local apics being dumped. Also it is possible
5952 to set it to "all" by meaning -- no limit here.
5953 Format: { 1 (default) | 2 | ... | all }.
5954 The parameter valid if only apic=debug or
5955 apic=verbose is specified.
5956 Example: apic=debug show_lapic=all
5957
5958 simeth= [IA-64]
5959 simscsi=
5960
5961 slab_debug[=options[,slabs][;[options[,slabs]]...] [MM]
5962 Enabling slab_debug allows one to determine the
5963 culprit if slab objects become corrupted. Enabling
5964 slab_debug can create guard zones around objects and
5965 may poison objects when not in use. Also tracks the
5966 last alloc / free. For more information see
5967 Documentation/mm/slub.rst.
5968 (slub_debug legacy name also accepted for now)
5969
5970 slab_max_order= [MM]
5971 Determines the maximum allowed order for slabs.
5972 A high setting may cause OOMs due to memory
5973 fragmentation. For more information see
5974 Documentation/mm/slub.rst.
5975 (slub_max_order legacy name also accepted for now)
5976
5977 slab_merge [MM]
5978 Enable merging of slabs with similar size when the
5979 kernel is built without CONFIG_SLAB_MERGE_DEFAULT.
5980 (slub_merge legacy name also accepted for now)
5981
5982 slab_min_objects= [MM]
5983 The minimum number of objects per slab. SLUB will
5984 increase the slab order up to slab_max_order to
5985 generate a sufficiently large slab able to contain
5986 the number of objects indicated. The higher the number
5987 of objects the smaller the overhead of tracking slabs
5988 and the less frequently locks need to be acquired.
5989 For more information see Documentation/mm/slub.rst.
5990 (slub_min_objects legacy name also accepted for now)
5991
5992 slab_min_order= [MM]
5993 Determines the minimum page order for slabs. Must be
5994 lower or equal to slab_max_order. For more information see
5995 Documentation/mm/slub.rst.
5996 (slub_min_order legacy name also accepted for now)
5997
5998 slab_nomerge [MM]
5999 Disable merging of slabs with similar size. May be
6000 necessary if there is some reason to distinguish
6001 allocs to different slabs, especially in hardened
6002 environments where the risk of heap overflows and
6003 layout control by attackers can usually be
6004 frustrated by disabling merging. This will reduce
6005 most of the exposure of a heap attack to a single
6006 cache (risks via metadata attacks are mostly
6007 unchanged). Debug options disable merging on their
6008 own.
6009 For more information see Documentation/mm/slub.rst.
6010 (slub_nomerge legacy name also accepted for now)
6011
6012 slram= [HW,MTD]
6013
6014 smart2= [HW]
6015 Format: <io1>[,<io2>[,...,<io8>]]
6016
6017 smp.csd_lock_timeout= [KNL]
6018 Specify the period of time in milliseconds
6019 that smp_call_function() and friends will wait
6020 for a CPU to release the CSD lock. This is
6021 useful when diagnosing bugs involving CPUs
6022 disabling interrupts for extended periods
6023 of time. Defaults to 5,000 milliseconds, and
6024 setting a value of zero disables this feature.
6025 This feature may be more efficiently disabled
6026 using the csdlock_debug- kernel parameter.
6027
6028 smp.panic_on_ipistall= [KNL]
6029 If a csd_lock_timeout extends for more than
6030 the specified number of milliseconds, panic the
6031 system. By default, let CSD-lock acquisition
6032 take as long as they take. Specifying 300,000
6033 for this value provides a 5-minute timeout.
6034
6035 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices
6036 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port
6037 smsc-ircc2.ircc_sir= [HW] SIR base I/O port
6038 smsc-ircc2.ircc_fir= [HW] FIR base I/O port
6039 smsc-ircc2.ircc_irq= [HW] IRQ line
6040 smsc-ircc2.ircc_dma= [HW] DMA channel
6041 smsc-ircc2.ircc_transceiver= [HW] Transceiver type:
6042 0: Toshiba Satellite 1800 (GP data pin select)
6043 1: Fast pin select (default)
6044 2: ATC IRMode
6045
6046 smt= [KNL,MIPS,S390,EARLY] Set the maximum number of threads
6047 (logical CPUs) to use per physical CPU on systems
6048 capable of symmetric multithreading (SMT). Will
6049 be capped to the actual hardware limit.
6050 Format: <integer>
6051 Default: -1 (no limit)
6052
6053 softlockup_panic=
6054 [KNL] Should the soft-lockup detector generate panics.
6055 Format: 0 | 1
6056
6057 A value of 1 instructs the soft-lockup detector
6058 to panic the machine when a soft-lockup occurs. It is
6059 also controlled by the kernel.softlockup_panic sysctl
6060 and CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC, which is the
6061 respective build-time switch to that functionality.
6062
6063 softlockup_all_cpu_backtrace=
6064 [KNL] Should the soft-lockup detector generate
6065 backtraces on all cpus.
6066 Format: 0 | 1
6067
6068 sonypi.*= [HW] Sony Programmable I/O Control Device driver
6069 See Documentation/admin-guide/laptops/sonypi.rst
6070
6071 spectre_bhi= [X86] Control mitigation of Branch History Injection
6072 (BHI) vulnerability. This setting affects the
6073 deployment of the HW BHI control and the SW BHB
6074 clearing sequence.
6075
6076 on - (default) Enable the HW or SW mitigation
6077 as needed.
6078 off - Disable the mitigation.
6079
6080 spectre_v2= [X86,EARLY] Control mitigation of Spectre variant 2
6081 (indirect branch speculation) vulnerability.
6082 The default operation protects the kernel from
6083 user space attacks.
6084
6085 on - unconditionally enable, implies
6086 spectre_v2_user=on
6087 off - unconditionally disable, implies
6088 spectre_v2_user=off
6089 auto - kernel detects whether your CPU model is
6090 vulnerable
6091
6092 Selecting 'on' will, and 'auto' may, choose a
6093 mitigation method at run time according to the
6094 CPU, the available microcode, the setting of the
6095 CONFIG_MITIGATION_RETPOLINE configuration option,
6096 and the compiler with which the kernel was built.
6097
6098 Selecting 'on' will also enable the mitigation
6099 against user space to user space task attacks.
6100
6101 Selecting 'off' will disable both the kernel and
6102 the user space protections.
6103
6104 Specific mitigations can also be selected manually:
6105
6106 retpoline - replace indirect branches
6107 retpoline,generic - Retpolines
6108 retpoline,lfence - LFENCE; indirect branch
6109 retpoline,amd - alias for retpoline,lfence
6110 eibrs - Enhanced/Auto IBRS
6111 eibrs,retpoline - Enhanced/Auto IBRS + Retpolines
6112 eibrs,lfence - Enhanced/Auto IBRS + LFENCE
6113 ibrs - use IBRS to protect kernel
6114
6115 Not specifying this option is equivalent to
6116 spectre_v2=auto.
6117
6118 spectre_v2_user=
6119 [X86] Control mitigation of Spectre variant 2
6120 (indirect branch speculation) vulnerability between
6121 user space tasks
6122
6123 on - Unconditionally enable mitigations. Is
6124 enforced by spectre_v2=on
6125
6126 off - Unconditionally disable mitigations. Is
6127 enforced by spectre_v2=off
6128
6129 prctl - Indirect branch speculation is enabled,
6130 but mitigation can be enabled via prctl
6131 per thread. The mitigation control state
6132 is inherited on fork.
6133
6134 prctl,ibpb
6135 - Like "prctl" above, but only STIBP is
6136 controlled per thread. IBPB is issued
6137 always when switching between different user
6138 space processes.
6139
6140 seccomp
6141 - Same as "prctl" above, but all seccomp
6142 threads will enable the mitigation unless
6143 they explicitly opt out.
6144
6145 seccomp,ibpb
6146 - Like "seccomp" above, but only STIBP is
6147 controlled per thread. IBPB is issued
6148 always when switching between different
6149 user space processes.
6150
6151 auto - Kernel selects the mitigation depending on
6152 the available CPU features and vulnerability.
6153
6154 Default mitigation: "prctl"
6155
6156 Not specifying this option is equivalent to
6157 spectre_v2_user=auto.
6158
6159 spec_rstack_overflow=
6160 [X86,EARLY] Control RAS overflow mitigation on AMD Zen CPUs
6161
6162 off - Disable mitigation
6163 microcode - Enable microcode mitigation only
6164 safe-ret - Enable sw-only safe RET mitigation (default)
6165 ibpb - Enable mitigation by issuing IBPB on
6166 kernel entry
6167 ibpb-vmexit - Issue IBPB only on VMEXIT
6168 (cloud-specific mitigation)
6169
6170 spec_store_bypass_disable=
6171 [HW,EARLY] Control Speculative Store Bypass (SSB) Disable mitigation
6172 (Speculative Store Bypass vulnerability)
6173
6174 Certain CPUs are vulnerable to an exploit against a
6175 a common industry wide performance optimization known
6176 as "Speculative Store Bypass" in which recent stores
6177 to the same memory location may not be observed by
6178 later loads during speculative execution. The idea
6179 is that such stores are unlikely and that they can
6180 be detected prior to instruction retirement at the
6181 end of a particular speculation execution window.
6182
6183 In vulnerable processors, the speculatively forwarded
6184 store can be used in a cache side channel attack, for
6185 example to read memory to which the attacker does not
6186 directly have access (e.g. inside sandboxed code).
6187
6188 This parameter controls whether the Speculative Store
6189 Bypass optimization is used.
6190
6191 On x86 the options are:
6192
6193 on - Unconditionally disable Speculative Store Bypass
6194 off - Unconditionally enable Speculative Store Bypass
6195 auto - Kernel detects whether the CPU model contains an
6196 implementation of Speculative Store Bypass and
6197 picks the most appropriate mitigation. If the
6198 CPU is not vulnerable, "off" is selected. If the
6199 CPU is vulnerable the default mitigation is
6200 architecture and Kconfig dependent. See below.
6201 prctl - Control Speculative Store Bypass per thread
6202 via prctl. Speculative Store Bypass is enabled
6203 for a process by default. The state of the control
6204 is inherited on fork.
6205 seccomp - Same as "prctl" above, but all seccomp threads
6206 will disable SSB unless they explicitly opt out.
6207
6208 Default mitigations:
6209 X86: "prctl"
6210
6211 On powerpc the options are:
6212
6213 on,auto - On Power8 and Power9 insert a store-forwarding
6214 barrier on kernel entry and exit. On Power7
6215 perform a software flush on kernel entry and
6216 exit.
6217 off - No action.
6218
6219 Not specifying this option is equivalent to
6220 spec_store_bypass_disable=auto.
6221
6222 spia_io_base= [HW,MTD]
6223 spia_fio_base=
6224 spia_pedr=
6225 spia_peddr=
6226
6227 split_lock_detect=
6228 [X86] Enable split lock detection or bus lock detection
6229
6230 When enabled (and if hardware support is present), atomic
6231 instructions that access data across cache line
6232 boundaries will result in an alignment check exception
6233 for split lock detection or a debug exception for
6234 bus lock detection.
6235
6236 off - not enabled
6237
6238 warn - the kernel will emit rate-limited warnings
6239 about applications triggering the #AC
6240 exception or the #DB exception. This mode is
6241 the default on CPUs that support split lock
6242 detection or bus lock detection. Default
6243 behavior is by #AC if both features are
6244 enabled in hardware.
6245
6246 fatal - the kernel will send SIGBUS to applications
6247 that trigger the #AC exception or the #DB
6248 exception. Default behavior is by #AC if
6249 both features are enabled in hardware.
6250
6251 ratelimit:N -
6252 Set system wide rate limit to N bus locks
6253 per second for bus lock detection.
6254 0 < N <= 1000.
6255
6256 N/A for split lock detection.
6257
6258
6259 If an #AC exception is hit in the kernel or in
6260 firmware (i.e. not while executing in user mode)
6261 the kernel will oops in either "warn" or "fatal"
6262 mode.
6263
6264 #DB exception for bus lock is triggered only when
6265 CPL > 0.
6266
6267 srbds= [X86,INTEL,EARLY]
6268 Control the Special Register Buffer Data Sampling
6269 (SRBDS) mitigation.
6270
6271 Certain CPUs are vulnerable to an MDS-like
6272 exploit which can leak bits from the random
6273 number generator.
6274
6275 By default, this issue is mitigated by
6276 microcode. However, the microcode fix can cause
6277 the RDRAND and RDSEED instructions to become
6278 much slower. Among other effects, this will
6279 result in reduced throughput from /dev/urandom.
6280
6281 The microcode mitigation can be disabled with
6282 the following option:
6283
6284 off: Disable mitigation and remove
6285 performance impact to RDRAND and RDSEED
6286
6287 srcutree.big_cpu_lim [KNL]
6288 Specifies the number of CPUs constituting a
6289 large system, such that srcu_struct structures
6290 should immediately allocate an srcu_node array.
6291 This kernel-boot parameter defaults to 128,
6292 but takes effect only when the low-order four
6293 bits of srcutree.convert_to_big is equal to 3
6294 (decide at boot).
6295
6296 srcutree.convert_to_big [KNL]
6297 Specifies under what conditions an SRCU tree
6298 srcu_struct structure will be converted to big
6299 form, that is, with an rcu_node tree:
6300
6301 0: Never.
6302 1: At init_srcu_struct() time.
6303 2: When rcutorture decides to.
6304 3: Decide at boot time (default).
6305 0x1X: Above plus if high contention.
6306
6307 Either way, the srcu_node tree will be sized based
6308 on the actual runtime number of CPUs (nr_cpu_ids)
6309 instead of the compile-time CONFIG_NR_CPUS.
6310
6311 srcutree.counter_wrap_check [KNL]
6312 Specifies how frequently to check for
6313 grace-period sequence counter wrap for the
6314 srcu_data structure's ->srcu_gp_seq_needed field.
6315 The greater the number of bits set in this kernel
6316 parameter, the less frequently counter wrap will
6317 be checked for. Note that the bottom two bits
6318 are ignored.
6319
6320 srcutree.exp_holdoff [KNL]
6321 Specifies how many nanoseconds must elapse
6322 since the end of the last SRCU grace period for
6323 a given srcu_struct until the next normal SRCU
6324 grace period will be considered for automatic
6325 expediting. Set to zero to disable automatic
6326 expediting.
6327
6328 srcutree.srcu_max_nodelay [KNL]
6329 Specifies the number of no-delay instances
6330 per jiffy for which the SRCU grace period
6331 worker thread will be rescheduled with zero
6332 delay. Beyond this limit, worker thread will
6333 be rescheduled with a sleep delay of one jiffy.
6334
6335 srcutree.srcu_max_nodelay_phase [KNL]
6336 Specifies the per-grace-period phase, number of
6337 non-sleeping polls of readers. Beyond this limit,
6338 grace period worker thread will be rescheduled
6339 with a sleep delay of one jiffy, between each
6340 rescan of the readers, for a grace period phase.
6341
6342 srcutree.srcu_retry_check_delay [KNL]
6343 Specifies number of microseconds of non-sleeping
6344 delay between each non-sleeping poll of readers.
6345
6346 srcutree.small_contention_lim [KNL]
6347 Specifies the number of update-side contention
6348 events per jiffy will be tolerated before
6349 initiating a conversion of an srcu_struct
6350 structure to big form. Note that the value of
6351 srcutree.convert_to_big must have the 0x10 bit
6352 set for contention-based conversions to occur.
6353
6354 ssbd= [ARM64,HW,EARLY]
6355 Speculative Store Bypass Disable control
6356
6357 On CPUs that are vulnerable to the Speculative
6358 Store Bypass vulnerability and offer a
6359 firmware based mitigation, this parameter
6360 indicates how the mitigation should be used:
6361
6362 force-on: Unconditionally enable mitigation for
6363 for both kernel and userspace
6364 force-off: Unconditionally disable mitigation for
6365 for both kernel and userspace
6366 kernel: Always enable mitigation in the
6367 kernel, and offer a prctl interface
6368 to allow userspace to register its
6369 interest in being mitigated too.
6370
6371 stack_guard_gap= [MM]
6372 override the default stack gap protection. The value
6373 is in page units and it defines how many pages prior
6374 to (for stacks growing down) resp. after (for stacks
6375 growing up) the main stack are reserved for no other
6376 mapping. Default value is 256 pages.
6377
6378 stack_depot_disable= [KNL,EARLY]
6379 Setting this to true through kernel command line will
6380 disable the stack depot thereby saving the static memory
6381 consumed by the stack hash table. By default this is set
6382 to false.
6383
6384 stacktrace [FTRACE]
6385 Enabled the stack tracer on boot up.
6386
6387 stacktrace_filter=[function-list]
6388 [FTRACE] Limit the functions that the stack tracer
6389 will trace at boot up. function-list is a comma-separated
6390 list of functions. This list can be changed at run
6391 time by the stack_trace_filter file in the debugfs
6392 tracing directory. Note, this enables stack tracing
6393 and the stacktrace above is not needed.
6394
6395 sti= [PARISC,HW]
6396 Format: <num>
6397 Set the STI (builtin display/keyboard on the HP-PARISC
6398 machines) console (graphic card) which should be used
6399 as the initial boot-console.
6400 See also comment in drivers/video/console/sticore.c.
6401
6402 sti_font= [HW]
6403 See comment in drivers/video/console/sticore.c.
6404
6405 stifb= [HW]
6406 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]]
6407
6408 strict_sas_size=
6409 [X86]
6410 Format: <bool>
6411 Enable or disable strict sigaltstack size checks
6412 against the required signal frame size which
6413 depends on the supported FPU features. This can
6414 be used to filter out binaries which have
6415 not yet been made aware of AT_MINSIGSTKSZ.
6416
6417 stress_hpt [PPC,EARLY]
6418 Limits the number of kernel HPT entries in the hash
6419 page table to increase the rate of hash page table
6420 faults on kernel addresses.
6421
6422 stress_slb [PPC,EARLY]
6423 Limits the number of kernel SLB entries, and flushes
6424 them frequently to increase the rate of SLB faults
6425 on kernel addresses.
6426
6427 sunrpc.min_resvport=
6428 sunrpc.max_resvport=
6429 [NFS,SUNRPC]
6430 SunRPC servers often require that client requests
6431 originate from a privileged port (i.e. a port in the
6432 range 0 < portnr < 1024).
6433 An administrator who wishes to reserve some of these
6434 ports for other uses may adjust the range that the
6435 kernel's sunrpc client considers to be privileged
6436 using these two parameters to set the minimum and
6437 maximum port values.
6438
6439 sunrpc.svc_rpc_per_connection_limit=
6440 [NFS,SUNRPC]
6441 Limit the number of requests that the server will
6442 process in parallel from a single connection.
6443 The default value is 0 (no limit).
6444
6445 sunrpc.pool_mode=
6446 [NFS]
6447 Control how the NFS server code allocates CPUs to
6448 service thread pools. Depending on how many NICs
6449 you have and where their interrupts are bound, this
6450 option will affect which CPUs will do NFS serving.
6451 Note: this parameter cannot be changed while the
6452 NFS server is running.
6453
6454 auto the server chooses an appropriate mode
6455 automatically using heuristics
6456 global a single global pool contains all CPUs
6457 percpu one pool for each CPU
6458 pernode one pool for each NUMA node (equivalent
6459 to global on non-NUMA machines)
6460
6461 sunrpc.tcp_slot_table_entries=
6462 sunrpc.udp_slot_table_entries=
6463 [NFS,SUNRPC]
6464 Sets the upper limit on the number of simultaneous
6465 RPC calls that can be sent from the client to a
6466 server. Increasing these values may allow you to
6467 improve throughput, but will also increase the
6468 amount of memory reserved for use by the client.
6469
6470 suspend.pm_test_delay=
6471 [SUSPEND]
6472 Sets the number of seconds to remain in a suspend test
6473 mode before resuming the system (see
6474 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG
6475 is set. Default value is 5.
6476
6477 svm= [PPC]
6478 Format: { on | off | y | n | 1 | 0 }
6479 This parameter controls use of the Protected
6480 Execution Facility on pSeries.
6481
6482 swiotlb= [ARM,IA-64,PPC,MIPS,X86,EARLY]
6483 Format: { <int> [,<int>] | force | noforce }
6484 <int> -- Number of I/O TLB slabs
6485 <int> -- Second integer after comma. Number of swiotlb
6486 areas with their own lock. Will be rounded up
6487 to a power of 2.
6488 force -- force using of bounce buffers even if they
6489 wouldn't be automatically used by the kernel
6490 noforce -- Never use bounce buffers (for debugging)
6491
6492 switches= [HW,M68k,EARLY]
6493
6494 sysctl.*= [KNL]
6495 Set a sysctl parameter, right before loading the init
6496 process, as if the value was written to the respective
6497 /proc/sys/... file. Both '.' and '/' are recognized as
6498 separators. Unrecognized parameters and invalid values
6499 are reported in the kernel log. Sysctls registered
6500 later by a loaded module cannot be set this way.
6501 Example: sysctl.vm.swappiness=40
6502
6503 sysrq_always_enabled
6504 [KNL]
6505 Ignore sysrq setting - this boot parameter will
6506 neutralize any effect of /proc/sys/kernel/sysrq.
6507 Useful for debugging.
6508
6509 tcpmhash_entries= [KNL,NET]
6510 Set the number of tcp_metrics_hash slots.
6511 Default value is 8192 or 16384 depending on total
6512 ram pages. This is used to specify the TCP metrics
6513 cache size. See Documentation/networking/ip-sysctl.rst
6514 "tcp_no_metrics_save" section for more details.
6515
6516 tdfx= [HW,DRM]
6517
6518 test_suspend= [SUSPEND]
6519 Format: { "mem" | "standby" | "freeze" }[,N]
6520 Specify "mem" (for Suspend-to-RAM) or "standby" (for
6521 standby suspend) or "freeze" (for suspend type freeze)
6522 as the system sleep state during system startup with
6523 the optional capability to repeat N number of times.
6524 The system is woken from this state using a
6525 wakeup-capable RTC alarm.
6526
6527 thash_entries= [KNL,NET]
6528 Set number of hash buckets for TCP connection
6529
6530 thermal.act= [HW,ACPI]
6531 -1: disable all active trip points in all thermal zones
6532 <degrees C>: override all lowest active trip points
6533
6534 thermal.crt= [HW,ACPI]
6535 -1: disable all critical trip points in all thermal zones
6536 <degrees C>: override all critical trip points
6537
6538 thermal.off= [HW,ACPI]
6539 1: disable ACPI thermal control
6540
6541 thermal.psv= [HW,ACPI]
6542 -1: disable all passive trip points
6543 <degrees C>: override all passive trip points to this
6544 value
6545
6546 thermal.tzp= [HW,ACPI]
6547 Specify global default ACPI thermal zone polling rate
6548 <deci-seconds>: poll all this frequency
6549 0: no polling (default)
6550
6551 threadirqs [KNL,EARLY]
6552 Force threading of all interrupt handlers except those
6553 marked explicitly IRQF_NO_THREAD.
6554
6555 topology= [S390,EARLY]
6556 Format: {off | on}
6557 Specify if the kernel should make use of the cpu
6558 topology information if the hardware supports this.
6559 The scheduler will make use of this information and
6560 e.g. base its process migration decisions on it.
6561 Default is on.
6562
6563 topology_updates= [KNL, PPC, NUMA]
6564 Format: {off}
6565 Specify if the kernel should ignore (off)
6566 topology updates sent by the hypervisor to this
6567 LPAR.
6568
6569 torture.disable_onoff_at_boot= [KNL]
6570 Prevent the CPU-hotplug component of torturing
6571 until after init has spawned.
6572
6573 torture.ftrace_dump_at_shutdown= [KNL]
6574 Dump the ftrace buffer at torture-test shutdown,
6575 even if there were no errors. This can be a
6576 very costly operation when many torture tests
6577 are running concurrently, especially on systems
6578 with rotating-rust storage.
6579
6580 torture.verbose_sleep_frequency= [KNL]
6581 Specifies how many verbose printk()s should be
6582 emitted between each sleep. The default of zero
6583 disables verbose-printk() sleeping.
6584
6585 torture.verbose_sleep_duration= [KNL]
6586 Duration of each verbose-printk() sleep in jiffies.
6587
6588 tp720= [HW,PS2]
6589
6590 tpm_suspend_pcr=[HW,TPM]
6591 Format: integer pcr id
6592 Specify that at suspend time, the tpm driver
6593 should extend the specified pcr with zeros,
6594 as a workaround for some chips which fail to
6595 flush the last written pcr on TPM_SaveState.
6596 This will guarantee that all the other pcrs
6597 are saved.
6598
6599 tpm_tis.interrupts= [HW,TPM]
6600 Enable interrupts for the MMIO based physical layer
6601 for the FIFO interface. By default it is set to false
6602 (0). For more information about TPM hardware interfaces
6603 defined by Trusted Computing Group (TCG) see
6604 https://trustedcomputinggroup.org/resource/pc-client-platform-tpm-profile-ptp-specification/
6605
6606 tp_printk [FTRACE]
6607 Have the tracepoints sent to printk as well as the
6608 tracing ring buffer. This is useful for early boot up
6609 where the system hangs or reboots and does not give the
6610 option for reading the tracing buffer or performing a
6611 ftrace_dump_on_oops.
6612
6613 To turn off having tracepoints sent to printk,
6614 echo 0 > /proc/sys/kernel/tracepoint_printk
6615 Note, echoing 1 into this file without the
6616 tp_printk kernel cmdline option has no effect.
6617
6618 The tp_printk_stop_on_boot (see below) can also be used
6619 to stop the printing of events to console at
6620 late_initcall_sync.
6621
6622 ** CAUTION **
6623
6624 Having tracepoints sent to printk() and activating high
6625 frequency tracepoints such as irq or sched, can cause
6626 the system to live lock.
6627
6628 tp_printk_stop_on_boot [FTRACE]
6629 When tp_printk (above) is set, it can cause a lot of noise
6630 on the console. It may be useful to only include the
6631 printing of events during boot up, as user space may
6632 make the system inoperable.
6633
6634 This command line option will stop the printing of events
6635 to console at the late_initcall_sync() time frame.
6636
6637 trace_buf_size=nn[KMG]
6638 [FTRACE] will set tracing buffer size on each cpu.
6639
6640 trace_clock= [FTRACE] Set the clock used for tracing events
6641 at boot up.
6642 local - Use the per CPU time stamp counter
6643 (converted into nanoseconds). Fast, but
6644 depending on the architecture, may not be
6645 in sync between CPUs.
6646 global - Event time stamps are synchronize across
6647 CPUs. May be slower than the local clock,
6648 but better for some race conditions.
6649 counter - Simple counting of events (1, 2, ..)
6650 note, some counts may be skipped due to the
6651 infrastructure grabbing the clock more than
6652 once per event.
6653 uptime - Use jiffies as the time stamp.
6654 perf - Use the same clock that perf uses.
6655 mono - Use ktime_get_mono_fast_ns() for time stamps.
6656 mono_raw - Use ktime_get_raw_fast_ns() for time
6657 stamps.
6658 boot - Use ktime_get_boot_fast_ns() for time stamps.
6659 Architectures may add more clocks. See
6660 Documentation/trace/ftrace.rst for more details.
6661
6662 trace_event=[event-list]
6663 [FTRACE] Set and start specified trace events in order
6664 to facilitate early boot debugging. The event-list is a
6665 comma-separated list of trace events to enable. See
6666 also Documentation/trace/events.rst
6667
6668 trace_instance=[instance-info]
6669 [FTRACE] Create a ring buffer instance early in boot up.
6670 This will be listed in:
6671
6672 /sys/kernel/tracing/instances
6673
6674 Events can be enabled at the time the instance is created
6675 via:
6676
6677 trace_instance=<name>,<system1>:<event1>,<system2>:<event2>
6678
6679 Note, the "<system*>:" portion is optional if the event is
6680 unique.
6681
6682 trace_instance=foo,sched:sched_switch,irq_handler_entry,initcall
6683
6684 will enable the "sched_switch" event (note, the "sched:" is optional, and
6685 the same thing would happen if it was left off). The irq_handler_entry
6686 event, and all events under the "initcall" system.
6687
6688 trace_options=[option-list]
6689 [FTRACE] Enable or disable tracer options at boot.
6690 The option-list is a comma delimited list of options
6691 that can be enabled or disabled just as if you were
6692 to echo the option name into
6693
6694 /sys/kernel/tracing/trace_options
6695
6696 For example, to enable stacktrace option (to dump the
6697 stack trace of each event), add to the command line:
6698
6699 trace_options=stacktrace
6700
6701 See also Documentation/trace/ftrace.rst "trace options"
6702 section.
6703
6704 trace_trigger=[trigger-list]
6705 [FTRACE] Add a event trigger on specific events.
6706 Set a trigger on top of a specific event, with an optional
6707 filter.
6708
6709 The format is is "trace_trigger=<event>.<trigger>[ if <filter>],..."
6710 Where more than one trigger may be specified that are comma deliminated.
6711
6712 For example:
6713
6714 trace_trigger="sched_switch.stacktrace if prev_state == 2"
6715
6716 The above will enable the "stacktrace" trigger on the "sched_switch"
6717 event but only trigger it if the "prev_state" of the "sched_switch"
6718 event is "2" (TASK_UNINTERUPTIBLE).
6719
6720 See also "Event triggers" in Documentation/trace/events.rst
6721
6722
6723 traceoff_on_warning
6724 [FTRACE] enable this option to disable tracing when a
6725 warning is hit. This turns off "tracing_on". Tracing can
6726 be enabled again by echoing '1' into the "tracing_on"
6727 file located in /sys/kernel/tracing/
6728
6729 This option is useful, as it disables the trace before
6730 the WARNING dump is called, which prevents the trace to
6731 be filled with content caused by the warning output.
6732
6733 This option can also be set at run time via the sysctl
6734 option: kernel/traceoff_on_warning
6735
6736 transparent_hugepage=
6737 [KNL]
6738 Format: [always|madvise|never]
6739 Can be used to control the default behavior of the system
6740 with respect to transparent hugepages.
6741 See Documentation/admin-guide/mm/transhuge.rst
6742 for more details.
6743
6744 trusted.source= [KEYS]
6745 Format: <string>
6746 This parameter identifies the trust source as a backend
6747 for trusted keys implementation. Supported trust
6748 sources:
6749 - "tpm"
6750 - "tee"
6751 - "caam"
6752 If not specified then it defaults to iterating through
6753 the trust source list starting with TPM and assigns the
6754 first trust source as a backend which is initialized
6755 successfully during iteration.
6756
6757 trusted.rng= [KEYS]
6758 Format: <string>
6759 The RNG used to generate key material for trusted keys.
6760 Can be one of:
6761 - "kernel"
6762 - the same value as trusted.source: "tpm" or "tee"
6763 - "default"
6764 If not specified, "default" is used. In this case,
6765 the RNG's choice is left to each individual trust source.
6766
6767 tsc= Disable clocksource stability checks for TSC.
6768 Format: <string>
6769 [x86] reliable: mark tsc clocksource as reliable, this
6770 disables clocksource verification at runtime, as well
6771 as the stability checks done at bootup. Used to enable
6772 high-resolution timer mode on older hardware, and in
6773 virtualized environment.
6774 [x86] noirqtime: Do not use TSC to do irq accounting.
6775 Used to run time disable IRQ_TIME_ACCOUNTING on any
6776 platforms where RDTSC is slow and this accounting
6777 can add overhead.
6778 [x86] unstable: mark the TSC clocksource as unstable, this
6779 marks the TSC unconditionally unstable at bootup and
6780 avoids any further wobbles once the TSC watchdog notices.
6781 [x86] nowatchdog: disable clocksource watchdog. Used
6782 in situations with strict latency requirements (where
6783 interruptions from clocksource watchdog are not
6784 acceptable).
6785 [x86] recalibrate: force recalibration against a HW timer
6786 (HPET or PM timer) on systems whose TSC frequency was
6787 obtained from HW or FW using either an MSR or CPUID(0x15).
6788 Warn if the difference is more than 500 ppm.
6789 [x86] watchdog: Use TSC as the watchdog clocksource with
6790 which to check other HW timers (HPET or PM timer), but
6791 only on systems where TSC has been deemed trustworthy.
6792 This will be suppressed by an earlier tsc=nowatchdog and
6793 can be overridden by a later tsc=nowatchdog. A console
6794 message will flag any such suppression or overriding.
6795
6796 tsc_early_khz= [X86,EARLY] Skip early TSC calibration and use the given
6797 value instead. Useful when the early TSC frequency discovery
6798 procedure is not reliable, such as on overclocked systems
6799 with CPUID.16h support and partial CPUID.15h support.
6800 Format: <unsigned int>
6801
6802 tsx= [X86] Control Transactional Synchronization
6803 Extensions (TSX) feature in Intel processors that
6804 support TSX control.
6805
6806 This parameter controls the TSX feature. The options are:
6807
6808 on - Enable TSX on the system. Although there are
6809 mitigations for all known security vulnerabilities,
6810 TSX has been known to be an accelerator for
6811 several previous speculation-related CVEs, and
6812 so there may be unknown security risks associated
6813 with leaving it enabled.
6814
6815 off - Disable TSX on the system. (Note that this
6816 option takes effect only on newer CPUs which are
6817 not vulnerable to MDS, i.e., have
6818 MSR_IA32_ARCH_CAPABILITIES.MDS_NO=1 and which get
6819 the new IA32_TSX_CTRL MSR through a microcode
6820 update. This new MSR allows for the reliable
6821 deactivation of the TSX functionality.)
6822
6823 auto - Disable TSX if X86_BUG_TAA is present,
6824 otherwise enable TSX on the system.
6825
6826 Not specifying this option is equivalent to tsx=off.
6827
6828 See Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
6829 for more details.
6830
6831 tsx_async_abort= [X86,INTEL,EARLY] Control mitigation for the TSX Async
6832 Abort (TAA) vulnerability.
6833
6834 Similar to Micro-architectural Data Sampling (MDS)
6835 certain CPUs that support Transactional
6836 Synchronization Extensions (TSX) are vulnerable to an
6837 exploit against CPU internal buffers which can forward
6838 information to a disclosure gadget under certain
6839 conditions.
6840
6841 In vulnerable processors, the speculatively forwarded
6842 data can be used in a cache side channel attack, to
6843 access data to which the attacker does not have direct
6844 access.
6845
6846 This parameter controls the TAA mitigation. The
6847 options are:
6848
6849 full - Enable TAA mitigation on vulnerable CPUs
6850 if TSX is enabled.
6851
6852 full,nosmt - Enable TAA mitigation and disable SMT on
6853 vulnerable CPUs. If TSX is disabled, SMT
6854 is not disabled because CPU is not
6855 vulnerable to cross-thread TAA attacks.
6856 off - Unconditionally disable TAA mitigation
6857
6858 On MDS-affected machines, tsx_async_abort=off can be
6859 prevented by an active MDS mitigation as both vulnerabilities
6860 are mitigated with the same mechanism so in order to disable
6861 this mitigation, you need to specify mds=off too.
6862
6863 Not specifying this option is equivalent to
6864 tsx_async_abort=full. On CPUs which are MDS affected
6865 and deploy MDS mitigation, TAA mitigation is not
6866 required and doesn't provide any additional
6867 mitigation.
6868
6869 For details see:
6870 Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
6871
6872 turbografx.map[2|3]= [HW,JOY]
6873 TurboGraFX parallel port interface
6874 Format:
6875 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7>
6876 See also Documentation/input/devices/joystick-parport.rst
6877
6878 udbg-immortal [PPC] When debugging early kernel crashes that
6879 happen after console_init() and before a proper
6880 console driver takes over, this boot options might
6881 help "seeing" what's going on.
6882
6883 uhash_entries= [KNL,NET]
6884 Set number of hash buckets for UDP/UDP-Lite connections
6885
6886 uhci-hcd.ignore_oc=
6887 [USB] Ignore overcurrent events (default N).
6888 Some badly-designed motherboards generate lots of
6889 bogus events, for ports that aren't wired to
6890 anything. Set this parameter to avoid log spamming.
6891 Note that genuine overcurrent events won't be
6892 reported either.
6893
6894 unknown_nmi_panic
6895 [X86] Cause panic on unknown NMI.
6896
6897 unwind_debug [X86-64,EARLY]
6898 Enable unwinder debug output. This can be
6899 useful for debugging certain unwinder error
6900 conditions, including corrupt stacks and
6901 bad/missing unwinder metadata.
6902
6903 usbcore.authorized_default=
6904 [USB] Default USB device authorization:
6905 (default -1 = authorized (same as 1),
6906 0 = not authorized, 1 = authorized, 2 = authorized
6907 if device connected to internal port)
6908
6909 usbcore.autosuspend=
6910 [USB] The autosuspend time delay (in seconds) used
6911 for newly-detected USB devices (default 2). This
6912 is the time required before an idle device will be
6913 autosuspended. Devices for which the delay is set
6914 to a negative value won't be autosuspended at all.
6915
6916 usbcore.usbfs_snoop=
6917 [USB] Set to log all usbfs traffic (default 0 = off).
6918
6919 usbcore.usbfs_snoop_max=
6920 [USB] Maximum number of bytes to snoop in each URB
6921 (default = 65536).
6922
6923 usbcore.blinkenlights=
6924 [USB] Set to cycle leds on hubs (default 0 = off).
6925
6926 usbcore.old_scheme_first=
6927 [USB] Start with the old device initialization
6928 scheme (default 0 = off).
6929
6930 usbcore.usbfs_memory_mb=
6931 [USB] Memory limit (in MB) for buffers allocated by
6932 usbfs (default = 16, 0 = max = 2047).
6933
6934 usbcore.use_both_schemes=
6935 [USB] Try the other device initialization scheme
6936 if the first one fails (default 1 = enabled).
6937
6938 usbcore.initial_descriptor_timeout=
6939 [USB] Specifies timeout for the initial 64-byte
6940 USB_REQ_GET_DESCRIPTOR request in milliseconds
6941 (default 5000 = 5.0 seconds).
6942
6943 usbcore.nousb [USB] Disable the USB subsystem
6944
6945 usbcore.quirks=
6946 [USB] A list of quirk entries to augment the built-in
6947 usb core quirk list. List entries are separated by
6948 commas. Each entry has the form
6949 VendorID:ProductID:Flags. The IDs are 4-digit hex
6950 numbers and Flags is a set of letters. Each letter
6951 will change the built-in quirk; setting it if it is
6952 clear and clearing it if it is set. The letters have
6953 the following meanings:
6954 a = USB_QUIRK_STRING_FETCH_255 (string
6955 descriptors must not be fetched using
6956 a 255-byte read);
6957 b = USB_QUIRK_RESET_RESUME (device can't resume
6958 correctly so reset it instead);
6959 c = USB_QUIRK_NO_SET_INTF (device can't handle
6960 Set-Interface requests);
6961 d = USB_QUIRK_CONFIG_INTF_STRINGS (device can't
6962 handle its Configuration or Interface
6963 strings);
6964 e = USB_QUIRK_RESET (device can't be reset
6965 (e.g morph devices), don't use reset);
6966 f = USB_QUIRK_HONOR_BNUMINTERFACES (device has
6967 more interface descriptions than the
6968 bNumInterfaces count, and can't handle
6969 talking to these interfaces);
6970 g = USB_QUIRK_DELAY_INIT (device needs a pause
6971 during initialization, after we read
6972 the device descriptor);
6973 h = USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL (For
6974 high speed and super speed interrupt
6975 endpoints, the USB 2.0 and USB 3.0 spec
6976 require the interval in microframes (1
6977 microframe = 125 microseconds) to be
6978 calculated as interval = 2 ^
6979 (bInterval-1).
6980 Devices with this quirk report their
6981 bInterval as the result of this
6982 calculation instead of the exponent
6983 variable used in the calculation);
6984 i = USB_QUIRK_DEVICE_QUALIFIER (device can't
6985 handle device_qualifier descriptor
6986 requests);
6987 j = USB_QUIRK_IGNORE_REMOTE_WAKEUP (device
6988 generates spurious wakeup, ignore
6989 remote wakeup capability);
6990 k = USB_QUIRK_NO_LPM (device can't handle Link
6991 Power Management);
6992 l = USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL
6993 (Device reports its bInterval as linear
6994 frames instead of the USB 2.0
6995 calculation);
6996 m = USB_QUIRK_DISCONNECT_SUSPEND (Device needs
6997 to be disconnected before suspend to
6998 prevent spurious wakeup);
6999 n = USB_QUIRK_DELAY_CTRL_MSG (Device needs a
7000 pause after every control message);
7001 o = USB_QUIRK_HUB_SLOW_RESET (Hub needs extra
7002 delay after resetting its port);
7003 p = USB_QUIRK_SHORT_SET_ADDRESS_REQ_TIMEOUT
7004 (Reduce timeout of the SET_ADDRESS
7005 request from 5000 ms to 500 ms);
7006 Example: quirks=0781:5580:bk,0a5c:5834:gij
7007
7008 usbhid.mousepoll=
7009 [USBHID] The interval which mice are to be polled at.
7010
7011 usbhid.jspoll=
7012 [USBHID] The interval which joysticks are to be polled at.
7013
7014 usbhid.kbpoll=
7015 [USBHID] The interval which keyboards are to be polled at.
7016
7017 usb-storage.delay_use=
7018 [UMS] The delay in seconds before a new device is
7019 scanned for Logical Units (default 1).
7020
7021 usb-storage.quirks=
7022 [UMS] A list of quirks entries to supplement or
7023 override the built-in unusual_devs list. List
7024 entries are separated by commas. Each entry has
7025 the form VID:PID:Flags where VID and PID are Vendor
7026 and Product ID values (4-digit hex numbers) and
7027 Flags is a set of characters, each corresponding
7028 to a common usb-storage quirk flag as follows:
7029 a = SANE_SENSE (collect more than 18 bytes
7030 of sense data, not on uas);
7031 b = BAD_SENSE (don't collect more than 18
7032 bytes of sense data, not on uas);
7033 c = FIX_CAPACITY (decrease the reported
7034 device capacity by one sector);
7035 d = NO_READ_DISC_INFO (don't use
7036 READ_DISC_INFO command, not on uas);
7037 e = NO_READ_CAPACITY_16 (don't use
7038 READ_CAPACITY_16 command);
7039 f = NO_REPORT_OPCODES (don't use report opcodes
7040 command, uas only);
7041 g = MAX_SECTORS_240 (don't transfer more than
7042 240 sectors at a time, uas only);
7043 h = CAPACITY_HEURISTICS (decrease the
7044 reported device capacity by one
7045 sector if the number is odd);
7046 i = IGNORE_DEVICE (don't bind to this
7047 device);
7048 j = NO_REPORT_LUNS (don't use report luns
7049 command, uas only);
7050 k = NO_SAME (do not use WRITE_SAME, uas only)
7051 l = NOT_LOCKABLE (don't try to lock and
7052 unlock ejectable media, not on uas);
7053 m = MAX_SECTORS_64 (don't transfer more
7054 than 64 sectors = 32 KB at a time,
7055 not on uas);
7056 n = INITIAL_READ10 (force a retry of the
7057 initial READ(10) command, not on uas);
7058 o = CAPACITY_OK (accept the capacity
7059 reported by the device, not on uas);
7060 p = WRITE_CACHE (the device cache is ON
7061 by default, not on uas);
7062 r = IGNORE_RESIDUE (the device reports
7063 bogus residue values, not on uas);
7064 s = SINGLE_LUN (the device has only one
7065 Logical Unit);
7066 t = NO_ATA_1X (don't allow ATA(12) and ATA(16)
7067 commands, uas only);
7068 u = IGNORE_UAS (don't bind to the uas driver);
7069 w = NO_WP_DETECT (don't test whether the
7070 medium is write-protected).
7071 y = ALWAYS_SYNC (issue a SYNCHRONIZE_CACHE
7072 even if the device claims no cache,
7073 not on uas)
7074 Example: quirks=0419:aaf5:rl,0421:0433:rc
7075
7076 user_debug= [KNL,ARM]
7077 Format: <int>
7078 See arch/arm/Kconfig.debug help text.
7079 1 - undefined instruction events
7080 2 - system calls
7081 4 - invalid data aborts
7082 8 - SIGSEGV faults
7083 16 - SIGBUS faults
7084 Example: user_debug=31
7085
7086 userpte=
7087 [X86,EARLY] Flags controlling user PTE allocations.
7088
7089 nohigh = do not allocate PTE pages in
7090 HIGHMEM regardless of setting
7091 of CONFIG_HIGHPTE.
7092
7093 vdso= [X86,SH,SPARC]
7094 On X86_32, this is an alias for vdso32=. Otherwise:
7095
7096 vdso=1: enable VDSO (the default)
7097 vdso=0: disable VDSO mapping
7098
7099 vdso32= [X86] Control the 32-bit vDSO
7100 vdso32=1: enable 32-bit VDSO
7101 vdso32=0 or vdso32=2: disable 32-bit VDSO
7102
7103 See the help text for CONFIG_COMPAT_VDSO for more
7104 details. If CONFIG_COMPAT_VDSO is set, the default is
7105 vdso32=0; otherwise, the default is vdso32=1.
7106
7107 For compatibility with older kernels, vdso32=2 is an
7108 alias for vdso32=0.
7109
7110 Try vdso32=0 if you encounter an error that says:
7111 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed!
7112
7113 vector= [IA-64,SMP]
7114 vector=percpu: enable percpu vector domain
7115
7116 video= [FB,EARLY] Frame buffer configuration
7117 See Documentation/fb/modedb.rst.
7118
7119 video.brightness_switch_enabled= [ACPI]
7120 Format: [0|1]
7121 If set to 1, on receiving an ACPI notify event
7122 generated by hotkey, video driver will adjust brightness
7123 level and then send out the event to user space through
7124 the allocated input device. If set to 0, video driver
7125 will only send out the event without touching backlight
7126 brightness level.
7127 default: 1
7128
7129 virtio_mmio.device=
7130 [VMMIO] Memory mapped virtio (platform) device.
7131
7132 <size>@<baseaddr>:<irq>[:<id>]
7133 where:
7134 <size> := size (can use standard suffixes
7135 like K, M and G)
7136 <baseaddr> := physical base address
7137 <irq> := interrupt number (as passed to
7138 request_irq())
7139 <id> := (optional) platform device id
7140 example:
7141 virtio_mmio.device=1K@0x100b0000:48:7
7142
7143 Can be used multiple times for multiple devices.
7144
7145 vga= [BOOT,X86-32] Select a particular video mode
7146 See Documentation/arch/x86/boot.rst and
7147 Documentation/admin-guide/svga.rst.
7148 Use vga=ask for menu.
7149 This is actually a boot loader parameter; the value is
7150 passed to the kernel using a special protocol.
7151
7152 vm_debug[=options] [KNL] Available with CONFIG_DEBUG_VM=y.
7153 May slow down system boot speed, especially when
7154 enabled on systems with a large amount of memory.
7155 All options are enabled by default, and this
7156 interface is meant to allow for selectively
7157 enabling or disabling specific virtual memory
7158 debugging features.
7159
7160 Available options are:
7161 P Enable page structure init time poisoning
7162 - Disable all of the above options
7163
7164 vmalloc=nn[KMG] [KNL,BOOT,EARLY] Forces the vmalloc area to have an
7165 exact size of <nn>. This can be used to increase
7166 the minimum size (128MB on x86). It can also be
7167 used to decrease the size and leave more room
7168 for directly mapped kernel RAM.
7169
7170 vmcp_cma=nn[MG] [KNL,S390,EARLY]
7171 Sets the memory size reserved for contiguous memory
7172 allocations for the vmcp device driver.
7173
7174 vmhalt= [KNL,S390] Perform z/VM CP command after system halt.
7175 Format: <command>
7176
7177 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic.
7178 Format: <command>
7179
7180 vmpoff= [KNL,S390] Perform z/VM CP command after power off.
7181 Format: <command>
7182
7183 vsyscall= [X86-64,EARLY]
7184 Controls the behavior of vsyscalls (i.e. calls to
7185 fixed addresses of 0xffffffffff600x00 from legacy
7186 code). Most statically-linked binaries and older
7187 versions of glibc use these calls. Because these
7188 functions are at fixed addresses, they make nice
7189 targets for exploits that can control RIP.
7190
7191 emulate Vsyscalls turn into traps and are emulated
7192 reasonably safely. The vsyscall page is
7193 readable.
7194
7195 xonly [default] Vsyscalls turn into traps and are
7196 emulated reasonably safely. The vsyscall
7197 page is not readable.
7198
7199 none Vsyscalls don't work at all. This makes
7200 them quite hard to use for exploits but
7201 might break your system.
7202
7203 vt.color= [VT] Default text color.
7204 Format: 0xYX, X = foreground, Y = background.
7205 Default: 0x07 = light gray on black.
7206
7207 vt.cur_default= [VT] Default cursor shape.
7208 Format: 0xCCBBAA, where AA, BB, and CC are the same as
7209 the parameters of the <Esc>[?A;B;Cc escape sequence;
7210 see VGA-softcursor.txt. Default: 2 = underline.
7211
7212 vt.default_blu= [VT]
7213 Format: <blue0>,<blue1>,<blue2>,...,<blue15>
7214 Change the default blue palette of the console.
7215 This is a 16-member array composed of values
7216 ranging from 0-255.
7217
7218 vt.default_grn= [VT]
7219 Format: <green0>,<green1>,<green2>,...,<green15>
7220 Change the default green palette of the console.
7221 This is a 16-member array composed of values
7222 ranging from 0-255.
7223
7224 vt.default_red= [VT]
7225 Format: <red0>,<red1>,<red2>,...,<red15>
7226 Change the default red palette of the console.
7227 This is a 16-member array composed of values
7228 ranging from 0-255.
7229
7230 vt.default_utf8=
7231 [VT]
7232 Format=<0|1>
7233 Set system-wide default UTF-8 mode for all tty's.
7234 Default is 1, i.e. UTF-8 mode is enabled for all
7235 newly opened terminals.
7236
7237 vt.global_cursor_default=
7238 [VT]
7239 Format=<-1|0|1>
7240 Set system-wide default for whether a cursor
7241 is shown on new VTs. Default is -1,
7242 i.e. cursors will be created by default unless
7243 overridden by individual drivers. 0 will hide
7244 cursors, 1 will display them.
7245
7246 vt.italic= [VT] Default color for italic text; 0-15.
7247 Default: 2 = green.
7248
7249 vt.underline= [VT] Default color for underlined text; 0-15.
7250 Default: 3 = cyan.
7251
7252 watchdog timers [HW,WDT] For information on watchdog timers,
7253 see Documentation/watchdog/watchdog-parameters.rst
7254 or other driver-specific files in the
7255 Documentation/watchdog/ directory.
7256
7257 watchdog_thresh=
7258 [KNL]
7259 Set the hard lockup detector stall duration
7260 threshold in seconds. The soft lockup detector
7261 threshold is set to twice the value. A value of 0
7262 disables both lockup detectors. Default is 10
7263 seconds.
7264
7265 workqueue.unbound_cpus=
7266 [KNL,SMP] Specify to constrain one or some CPUs
7267 to use in unbound workqueues.
7268 Format: <cpu-list>
7269 By default, all online CPUs are available for
7270 unbound workqueues.
7271
7272 workqueue.watchdog_thresh=
7273 If CONFIG_WQ_WATCHDOG is configured, workqueue can
7274 warn stall conditions and dump internal state to
7275 help debugging. 0 disables workqueue stall
7276 detection; otherwise, it's the stall threshold
7277 duration in seconds. The default value is 30 and
7278 it can be updated at runtime by writing to the
7279 corresponding sysfs file.
7280
7281 workqueue.cpu_intensive_thresh_us=
7282 Per-cpu work items which run for longer than this
7283 threshold are automatically considered CPU intensive
7284 and excluded from concurrency management to prevent
7285 them from noticeably delaying other per-cpu work
7286 items. Default is 10000 (10ms).
7287
7288 If CONFIG_WQ_CPU_INTENSIVE_REPORT is set, the kernel
7289 will report the work functions which violate this
7290 threshold repeatedly. They are likely good
7291 candidates for using WQ_UNBOUND workqueues instead.
7292
7293 workqueue.cpu_intensive_warning_thresh=<uint>
7294 If CONFIG_WQ_CPU_INTENSIVE_REPORT is set, the kernel
7295 will report the work functions which violate the
7296 intensive_threshold_us repeatedly. In order to prevent
7297 spurious warnings, start printing only after a work
7298 function has violated this threshold number of times.
7299
7300 The default is 4 times. 0 disables the warning.
7301
7302 workqueue.power_efficient
7303 Per-cpu workqueues are generally preferred because
7304 they show better performance thanks to cache
7305 locality; unfortunately, per-cpu workqueues tend to
7306 be more power hungry than unbound workqueues.
7307
7308 Enabling this makes the per-cpu workqueues which
7309 were observed to contribute significantly to power
7310 consumption unbound, leading to measurably lower
7311 power usage at the cost of small performance
7312 overhead.
7313
7314 The default value of this parameter is determined by
7315 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT.
7316
7317 workqueue.default_affinity_scope=
7318 Select the default affinity scope to use for unbound
7319 workqueues. Can be one of "cpu", "smt", "cache",
7320 "numa" and "system". Default is "cache". For more
7321 information, see the Affinity Scopes section in
7322 Documentation/core-api/workqueue.rst.
7323
7324 This can be changed after boot by writing to the
7325 matching /sys/module/workqueue/parameters file. All
7326 workqueues with the "default" affinity scope will be
7327 updated accordignly.
7328
7329 workqueue.debug_force_rr_cpu
7330 Workqueue used to implicitly guarantee that work
7331 items queued without explicit CPU specified are put
7332 on the local CPU. This guarantee is no longer true
7333 and while local CPU is still preferred work items
7334 may be put on foreign CPUs. This debug option
7335 forces round-robin CPU selection to flush out
7336 usages which depend on the now broken guarantee.
7337 When enabled, memory and cache locality will be
7338 impacted.
7339
7340 writecombine= [LOONGARCH,EARLY] Control the MAT (Memory Access
7341 Type) of ioremap_wc().
7342
7343 on - Enable writecombine, use WUC for ioremap_wc()
7344 off - Disable writecombine, use SUC for ioremap_wc()
7345
7346 x2apic_phys [X86-64,APIC,EARLY] Use x2apic physical mode instead of
7347 default x2apic cluster mode on platforms
7348 supporting x2apic.
7349
7350 xen_512gb_limit [KNL,X86-64,XEN]
7351 Restricts the kernel running paravirtualized under Xen
7352 to use only up to 512 GB of RAM. The reason to do so is
7353 crash analysis tools and Xen tools for doing domain
7354 save/restore/migration must be enabled to handle larger
7355 domains.
7356
7357 xen_emul_unplug= [HW,X86,XEN,EARLY]
7358 Unplug Xen emulated devices
7359 Format: [unplug0,][unplug1]
7360 ide-disks -- unplug primary master IDE devices
7361 aux-ide-disks -- unplug non-primary-master IDE devices
7362 nics -- unplug network devices
7363 all -- unplug all emulated devices (NICs and IDE disks)
7364 unnecessary -- unplugging emulated devices is
7365 unnecessary even if the host did not respond to
7366 the unplug protocol
7367 never -- do not unplug even if version check succeeds
7368
7369 xen_legacy_crash [X86,XEN,EARLY]
7370 Crash from Xen panic notifier, without executing late
7371 panic() code such as dumping handler.
7372
7373 xen_msr_safe= [X86,XEN,EARLY]
7374 Format: <bool>
7375 Select whether to always use non-faulting (safe) MSR
7376 access functions when running as Xen PV guest. The
7377 default value is controlled by CONFIG_XEN_PV_MSR_SAFE.
7378
7379 xen_nopvspin [X86,XEN,EARLY]
7380 Disables the qspinlock slowpath using Xen PV optimizations.
7381 This parameter is obsoleted by "nopvspin" parameter, which
7382 has equivalent effect for XEN platform.
7383
7384 xen_nopv [X86]
7385 Disables the PV optimizations forcing the HVM guest to
7386 run as generic HVM guest with no PV drivers.
7387 This option is obsoleted by the "nopv" option, which
7388 has equivalent effect for XEN platform.
7389
7390 xen_no_vector_callback
7391 [KNL,X86,XEN,EARLY] Disable the vector callback for Xen
7392 event channel interrupts.
7393
7394 xen_scrub_pages= [XEN]
7395 Boolean option to control scrubbing pages before giving them back
7396 to Xen, for use by other domains. Can be also changed at runtime
7397 with /sys/devices/system/xen_memory/xen_memory0/scrub_pages.
7398 Default value controlled with CONFIG_XEN_SCRUB_PAGES_DEFAULT.
7399
7400 xen_timer_slop= [X86-64,XEN,EARLY]
7401 Set the timer slop (in nanoseconds) for the virtual Xen
7402 timers (default is 100000). This adjusts the minimum
7403 delta of virtualized Xen timers, where lower values
7404 improve timer resolution at the expense of processing
7405 more timer interrupts.
7406
7407 xen.balloon_boot_timeout= [XEN]
7408 The time (in seconds) to wait before giving up to boot
7409 in case initial ballooning fails to free enough memory.
7410 Applies only when running as HVM or PVH guest and
7411 started with less memory configured than allowed at
7412 max. Default is 180.
7413
7414 xen.event_eoi_delay= [XEN]
7415 How long to delay EOI handling in case of event
7416 storms (jiffies). Default is 10.
7417
7418 xen.event_loop_timeout= [XEN]
7419 After which time (jiffies) the event handling loop
7420 should start to delay EOI handling. Default is 2.
7421
7422 xen.fifo_events= [XEN]
7423 Boolean parameter to disable using fifo event handling
7424 even if available. Normally fifo event handling is
7425 preferred over the 2-level event handling, as it is
7426 fairer and the number of possible event channels is
7427 much higher. Default is on (use fifo events).
7428
7429 xirc2ps_cs= [NET,PCMCIA]
7430 Format:
7431 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]]
7432
7433 xive= [PPC]
7434 By default on POWER9 and above, the kernel will
7435 natively use the XIVE interrupt controller. This option
7436 allows the fallback firmware mode to be used:
7437
7438 off Fallback to firmware control of XIVE interrupt
7439 controller on both pseries and powernv
7440 platforms. Only useful on POWER9 and above.
7441
7442 xive.store-eoi=off [PPC]
7443 By default on POWER10 and above, the kernel will use
7444 stores for EOI handling when the XIVE interrupt mode
7445 is active. This option allows the XIVE driver to use
7446 loads instead, as on POWER9.
7447
7448 xhci-hcd.quirks [USB,KNL]
7449 A hex value specifying bitmask with supplemental xhci
7450 host controller quirks. Meaning of each bit can be
7451 consulted in header drivers/usb/host/xhci.h.
7452
7453 xmon [PPC,EARLY]
7454 Format: { early | on | rw | ro | off }
7455 Controls if xmon debugger is enabled. Default is off.
7456 Passing only "xmon" is equivalent to "xmon=early".
7457 early Call xmon as early as possible on boot; xmon
7458 debugger is called from setup_arch().
7459 on xmon debugger hooks will be installed so xmon
7460 is only called on a kernel crash. Default mode,
7461 i.e. either "ro" or "rw" mode, is controlled
7462 with CONFIG_XMON_DEFAULT_RO_MODE.
7463 rw xmon debugger hooks will be installed so xmon
7464 is called only on a kernel crash, mode is write,
7465 meaning SPR registers, memory and, other data
7466 can be written using xmon commands.
7467 ro same as "rw" option above but SPR registers,
7468 memory, and other data can't be written using
7469 xmon commands.
7470 off xmon is disabled.
7471