1 accept_memory= [MM]
2 Format: { eager | lazy
3 default: lazy
4 By default, unaccepted
5 avoid prolonged boot t
6 some runtime overhead
7 accepted. In most case
8 For some workloads or
9 accept_memory=eager ca
10 at once during boot.
11
12 acpi= [HW,ACPI,X86,ARM64,RIS
13 Advanced Configuration
14 Format: { force | on |
15 copy_dsdt |
16 force -- enable ACPI i
17 on -- enable ACPI but
18 off -- disable ACPI if
19 noirq -- do not use AC
20 strict -- Be less tole
21 strictly ACPI
22 rsdt -- prefer RSDT ov
23 copy_dsdt -- copy DSDT
24 nospcr -- disable cons
25 default _seria
26 For ARM64, ONLY "acpi=
27 "acpi=nospcr" are avai
28 For RISCV64, ONLY "acp
29 are available
30
31 See also Documentation
32
33 acpi_apic_instance= [ACPI,IOAPIC,E
34 Format: <int>
35 2: use 2nd APIC table,
36 1,0: use 1st APIC tabl
37 default: 0
38
39 acpi_backlight= [HW,ACPI]
40 { vendor | video | nat
41 If set to vendor, pref
42 (e.g. thinkpad_acpi, s
43 of the ACPI video.ko d
44 If set to video, use t
45 If set to native, use
46 If set to none, disabl
47
48 acpi_force_32bit_fadt_addr [ACPI,EARLY
49 force FADT to use 32 b
50 64 bit X_* addresses.
51 bit addresses for forc
52 the older legacy 32 bi
53
54 acpica_no_return_repair [HW, ACPI]
55 Disable AML predefined
56 This mechanism can rep
57 the return objects mor
58 This option is useful
59 root cause of an AML i
60 has something to do wi
61
62 acpi.debug_layer= [HW,ACPI,ACPI_
63 acpi.debug_level= [HW,ACPI,ACPI_
64 Format: <int>
65 CONFIG_ACPI_DEBUG must
66 debug output. Bits in
67 _COMPONENT in an ACPI
68 #define _COMPONENT
69 Bits in debug_level co
70 ACPI_DEBUG_PRINT state
71 ACPI_DEBUG_PRINT((
72 The debug_level mask d
73 Documentation/firmware
74 debug layers and level
75
76 Enable processor drive
77 acpi.debug_layer=0
78 Enable AML "Debug" out
79 object while interpret
80 acpi.debug_layer=0
81 Enable all messages re
82 acpi.debug_layer=0
83
84 Some values produce so
85 unusable. The "log_bu
86 if you need to capture
87
88 acpi_enforce_resources= [ACPI]
89 { strict | lax | no }
90 Check for resource con
91 and ACPI OperationRegi
92 only). IO ports and me
93 used by the ACPI subsy
94 can interfere with leg
95 strict (default): acce
96 is denied; legacy driv
97 resources will fail to
98 lax: access to resourc
99 legacy drivers trying
100 will bind successfully
101 no: ACPI OperationRegi
102 no further checks are
103
104 acpi_force_table_verification [HW,AC
105 Enable table checksum
106 By default, this is di
107 size limitation.
108
109 acpi_irq_balance [HW,ACPI]
110 ACPI will balance acti
111 default in APIC mode
112
113 acpi_irq_nobalance [HW,ACPI]
114 ACPI will not move act
115 default in PIC mode
116
117 acpi_irq_isa= [HW,ACPI] If irq_balan
118 Format: <irq>,<irq>...
119
120 acpi_irq_pci= [HW,ACPI] If irq_balan
121 use by PCI
122 Format: <irq>,<irq>...
123
124 acpi_mask_gpe= [HW,ACPI]
125 Due to the existence o
126 by unsupported hardwar
127 GPE floodings that can
128 the GPE dispatcher.
129 This facility can be u
130 GPE floodings.
131 Format: <byte> or <bit
132
133 acpi_no_auto_serialize [HW,ACPI]
134 Disable auto-serializa
135 AML control methods th
136 named objects will be
137 auto-serialization fea
138 This feature is enable
139 This option allows to
140
141 acpi_no_memhotplug [ACPI] Disable memo
142 kernels.
143
144 acpi_no_static_ssdt [HW,ACPI,EARLY
145 Disable installation o
146 By default, SSDTs cont
147 installed automaticall
148 /sys/firmware/acpi/tab
149 This option turns off
150 Note that specifying t
151 dynamic table installa
152 tables to /sys/firmwar
153
154 acpi_no_watchdog [HW,ACPI,WDT]
155 Ignore the ACPI-based
156 a native driver contro
157
158 acpi_rsdp= [ACPI,EFI,KEXEC,EARLY]
159 Pass the RSDP address
160 on machines running EF
161 second kernel for kdum
162
163 acpi_os_name= [HW,ACPI] Tell ACPI BI
164 Format: To spoof as Wi
165
166 acpi_rev_override [ACPI] Override the
167 of 2 which is mandated
168 specification revision
169 be necessary to carry
170 row to make it take ef
171
172 acpi_osi= [HW,ACPI] Modify list
173 acpi_osi="string1"
174 acpi_osi="!string2"
175 acpi_osi=!*
176 acpi_osi=!
177
178 acpi_osi=!!
179
180 acpi_osi=
181
182 'acpi_osi=!' can be us
183 multiple 'acpi_osi="st
184 vendor string(s). Not
185 affect the default sta
186 it cannot affect the d
187 strings and the curren
188 specifying it multiple
189 is meaningless. This
190 care about the state o
191 should be controlled b
192 Examples:
193 1. 'acpi_osi=! acpi_
194 to 'acpi_osi="Win
195 can make '_OSI("W
196
197 'acpi_osi=' cannot be
198 'acpi_osi=' command li
199 exist in the ACPI name
200 only affect the _OSI s
201 multiple times through
202 meaningless.
203 Examples:
204 1. 'acpi_osi=' can m
205 FALSE.
206
207 'acpi_osi=!*' can be u
208 multiple 'acpi_osi="st
209 string(s). Note that
210 current state of both
211 feature group strings,
212 through kernel command
213 still not able to affe
214 there are quirks relat
215 is useful when one wan
216 feature group strings
217 the OSPM features.
218 Examples:
219 1. 'acpi_osi="Module
220 '_OSI("Module Dev
221 2. 'acpi_osi=!* acpi
222 '_OSI("Module Dev
223 3. 'acpi_osi=! acpi_
224 equivalent to
225 'acpi_osi=!* acpi
226 and
227 'acpi_osi=!* acpi
228 they all will mak
229
230 acpi_pm_good [X86]
231 Override the pmtimer b
232 to assume that this ma
233 and always returns goo
234
235 acpi_sci= [HW,ACPI,EARLY] ACPI S
236 Format: { level | edge
237
238 acpi_skip_timer_override [HW,ACPI,EARL
239 Recognize and ignore I
240 For broken nForce2 BIO
241
242 acpi_sleep= [HW,ACPI] Sleep option
243 Format: { s3_bios, s3_
244 s4_nohwsig,
245 sci_force_en
246 See Documentation/powe
247 s3_bios and s3_mode.
248 s3_beep is for debuggi
249 as soon as the kernel'
250 s4_hwsig causes the ke
251 signature during resum
252 refuse to resume if it
253 the ACPI specification
254 Windows does not do th
255 on docking. So the def
256 and simply warn when t
257 s4_hwsig option is ena
258 s4_nohwsig prevents AC
259 used (or even warned a
260 old_ordering causes th
261 control method, with r
262 low power states, to b
263 of _PTS is used by def
264 nonvs prevents the ker
265 ACPI NVS memory during
266 sci_force_enable cause
267 on resume from S1/S3 (
268 but some broken system
269 nobl causes the intern
270 behave incorrectly in
271 suspend and resume to
272
273 acpi_use_timer_override [HW,ACPI,EARLY
274 Use timer override. Fo
275 that require a timer o
276
277 add_efi_memmap [EFI,X86,EARLY] Includ
278 kernel's map of availa
279
280 agp= [AGP]
281 { off | try_unsupporte
282 off: disable AGP suppo
283 try_unsupported: try t
284 (may crash com
285
286 ALSA [HW,ALSA]
287 See Documentation/soun
288
289 alignment= [KNL,ARM]
290 Allow the default user
291 behaviour to be specif
292 bit 1 enables fixups,
293
294 align_va_addr= [X86-64]
295 Align virtual addresse
296 allocating a VMA at pr
297 gives you up to 3% per
298 machines (where it is
299 CPU-intensive style be
300 a microbenchmark depen
301
302 32: only for 32-bit pr
303 64: only for 64-bit pr
304 on: enable for both 32
305 off: disable for both
306
307 alloc_snapshot [FTRACE]
308 Allocate the ftrace sn
309 main buffer is allocat
310 and you need to use tr
311 do not want to use tra
312 to be done where GFP_K
313
314 allow_mismatched_32bit_el0 [ARM64,EARL
315 Allow execve() of 32-b
316 PER_LINUX32 personalit
317 subset of the CPUs sup
318 parameter is present,
319 EL0 is indicated by /s
320 and hot-unplug operati
321
322 See Documentation/arch
323 information.
324
325 amd_iommu= [HW,X86-64]
326 Pass parameters to the
327 Possible values are:
328 fullflush - Deprecated
329 off - do not ini
330 the system
331 force_isolation - Forc
332 devi
333 allo
334 requ
335 does
336 force_enable - Force e
337 to be b
338 option
339 pgtbl_v1 - Use v1
340 pgtbl_v2 - Use v2
341 irtcachedis - Disable
342
343 amd_iommu_dump= [HW,X86-64]
344 Enable AMD IOMMU drive
345 for AMD IOMMU. With th
346 driver will print ACPI
347 IOMMU initialization.
348
349 amd_iommu_intr= [HW,X86-64]
350 Specifies one of the f
351 remapping modes:
352 legacy - Use legac
353 vapic - Use virtu
354 to inject
355 This mode
356 (Default
357
358 amd_pstate= [X86,EARLY]
359 disable
360 Do not enable amd_ps
361 scaling driver for t
362 passive
363 Use amd_pstate with
364 In this mode autonom
365 Driver requests a de
366 tries to match the s
367 satisfied by guarant
368 active
369 Use amd_pstate_epp d
370 driver provides a hi
371 to bias toward perfo
372 to the CPPC firmware
373 calculate the runtim
374 frequency.
375 guided
376 Activate guided auto
377 maximum performance
378 selects a performanc
379 to the current workl
380
381 amd_prefcore=
382 [X86]
383 disable
384 Disable amd-pstate p
385
386 amijoy.map= [HW,JOY] Amiga joystic
387 Map of devices attache
388 Format: <a>,<b>
389 See also Documentation
390
391 analog.map= [HW,JOY] Analog joysti
392 Specifies type or capa
393 connected to one of 16
394 Format: <type1>,<type2
395
396 apc= [HW,SPARC]
397 Power management funct
398 Format: noidle
399 Disable APC CPU standb
400 not play well with APC
401 APC and your system cr
402
403 apic= [APIC,X86,EARLY] Advan
404 Change the output verb
405 Format: { quiet (defau
406 Change the amount of d
407 when initialising the
408 For X86-32, this can a
409 driver name.
410 Format: apic=driver_na
411 Examples: apic=bigsmp
412
413 apic_extnmi= [APIC,X86,EARLY] Exter
414 Format: { bsp (default
415 bsp: External NMI is
416 all: External NMIs ar
417 backup of CPU 0
418 none: External NMI is
419 useful so that a
420 shot down by NMI
421
422 autoconf= [IPV6]
423 See Documentation/netw
424
425 apm= [APM] Advanced Power M
426 See header of arch/x86
427
428 apparmor= [APPARMOR] Disable or
429 Format: { "0" | "1" }
430 See security/apparmor/
431 0 -- disable.
432 1 -- enable.
433 Default value is set v
434
435 arcrimi= [HW,NET] ARCnet - "RIM
436 Format: <io>,<irq>,<no
437
438 arm64.no32bit_el0 [ARM64] Unconditiona
439 32 bit applications.
440
441 arm64.nobti [ARM64] Unconditionall
442 Identification support
443
444 arm64.nomops [ARM64] Unconditionall
445 Set instructions suppo
446
447 arm64.nomte [ARM64] Unconditionall
448 support
449
450 arm64.nopauth [ARM64] Unconditionall
451 support
452
453 arm64.nosme [ARM64] Unconditionall
454 Extension support
455
456 arm64.nosve [ARM64] Unconditionall
457 Extension support
458
459 ataflop= [HW,M68k]
460
461 atarimouse= [HW,MOUSE] Atari Mouse
462
463 atkbd.extra= [HW] Enable extra LEDs
464 EzKey and similar keyb
465
466 atkbd.reset= [HW] Reset keyboard du
467
468 atkbd.set= [HW] Select keyboard c
469 Format: <int> (2 = AT
470
471 atkbd.scroll= [HW] Enable scroll whe
472 keyboards
473
474 atkbd.softraw= [HW] Choose between sy
475 Format: <bool> (0 = re
476
477 atkbd.softrepeat= [HW]
478 Use software keyboard
479
480 audit= [KNL] Enable the audit
481 Format: { "0" | "1" |
482 0 | off - kernel audit
483 enabled until the
484 unset - kernel audit i
485 will be fully enab
486 1 | on - kernel audit
487 enabled, storing a
488 messages in RAM un
489 userspace auditd.
490 Default: unset
491
492 audit_backlog_limit= [KNL] Set the aud
493 Format: <int> (must be
494 Default: 64
495
496 bau= [X86_UV] Enable the BA
497 behavior is to disable
498 Format: { "0" | "1" }
499 0 - Disable the BAU.
500 1 - Enable the BAU.
501 unset - Disable the BA
502
503 baycom_epp= [HW,AX25]
504 Format: <io>,<mode>
505
506 baycom_par= [HW,AX25] BayCom Paral
507 Format: <io>,<mode>
508 See header of drivers/
509
510 baycom_ser_fdx= [HW,AX25]
511 BayCom Serial Port AX.
512 Format: <io>,<irq>,<mo
513 See header of drivers/
514
515 baycom_ser_hdx= [HW,AX25]
516 BayCom Serial Port AX.
517 Format: <io>,<irq>,<mo
518 See header of drivers/
519
520 bert_disable [ACPI]
521 Disable BERT OS suppor
522
523 bgrt_disable [ACPI,X86,EARLY]
524 Disable BGRT to avoid
525
526 blkdevparts= Manual partition parsi
527 embedded devices based
528 See Documentation/bloc
529
530 boot_delay= [KNL,EARLY]
531 Milliseconds to delay
532 Only works if CONFIG_B
533 and you may also have
534 values larger than 10
535 erroneous and ignored.
536 Format: integer
537
538 bootconfig [KNL,EARLY]
539 Extended command line
540 and this will cause th
541
542 See Documentation/admi
543
544 bttv.card= [HW,V4L] bttv (bt848 +
545 bttv.radio= Most important insmod
546 kernel args too.
547 bttv.pll= See Documentation/admi
548 bttv.tuner=
549
550 bulk_remove=off [PPC] This parameter
551 firmware feature for f
552 at a time.
553
554 c101= [NET] Moxa C101 synchr
555
556 cachesize= [BUGS=X86-32] Override
557 Sometimes CPU hardware
558 size incorrectly. The
559 to fix known problems,
560 possible to determine
561 This option provides a
562
563 carrier_timeout=
564 [NET] Specifies amount
565 the kernel should wait
566 it waits 120 seconds.
567
568 ca_keys= [KEYS] This parameter
569 the system trusted key
570 trust validation.
571 format: { id:<keyid> |
572
573 cca= [MIPS,EARLY] Override
574 algorithm. Accepted v
575 inclusive. See arch/mi
576 for platform specific
577 others).
578
579 ccw_timeout_log [S390]
580 See Documentation/arch
581
582 cgroup_disable= [KNL] Disable a partic
583 Format: {name of the c
584 The effects of cgroup_
585 - foo isn't auto-mount
586 a single hierarchy
587 - foo isn't visible as
588 subsystem
589 - if foo is an optiona
590 disabled and corresp
591 created
592 {Currently only "memor
593 cut the overhead, othe
594 only cgroup_disable=me
595 Specifying "pressure"
596 stall information acco
597
598 cgroup_no_v1= [KNL] Disable cgroup c
599 Format: { { controller
600 [,{ controll
601 Like cgroup_disable, b
602 the blacklisted contro
603 "all" blacklists all c
604 named mounts. Specifyi
605 all v1 hierarchies.
606
607 cgroup_favordynmods= [KNL] Enable or D
608 Format: { "true" | "fa
609 Defaults to the value
610
611 cgroup.memory= [KNL] Pass options to
612 Format: <string>
613 nosocket -- Disable so
614 nokmem -- Disable kern
615 nobpf -- Disable BPF m
616
617 checkreqprot= [SELINUX] Set initial
618 Format: { "0" | "1" }
619 See security/selinux/K
620 0 -- check protection
621 any implied ex
622 1 -- check protection
623 Default value is set v
624 Value can be changed a
625 /sys/fs/selinu
626 Setting checkreqprot t
627
628 cio_ignore= [S390]
629 See Documentation/arch
630
631 clearcpuid=X[,X...] [X86]
632 Disable CPUID feature
633 arch/x86/include/asm/c
634 numbers X. Note the Li
635 stable over kernel opt
636 ones should be.
637 X can also be a string
638 in /proc/cpuinfo which
639 instability issue. How
640 in /proc/cpuinfo.
641 Note that using this o
642 Also note that user pr
643 or using the feature w
644 will still see it. Thi
645 being used by the kern
646 Also note the kernel m
647 some critical bits.
648
649 clk_ignore_unused
650 [CLK]
651 Prevents the clock fra
652 clocks that have not b
653 device driver but are
654 by the bootloader/firm
655 force such clocks to b
656 those clocks in any wa
657 debug and development,
658 platform with proper d
659 information, see Docum
660
661 clock= [BUGS=X86-32, HW] gett
662 [Deprecated]
663 Forces specified clock
664 when calculating getti
665 clocksource is not ava
666 Format: { pit | tsc |
667
668 clocksource= Override the default c
669 Format: <string>
670 Override the default c
671 with the name specifie
672 Some clocksource names
673 the platform:
674 [all] jiffies (this is
675 [ACPI] acpi_pm
676 [ARM] imx_timer1,OSTS,
677 pxa_timer,time
678 [X86-32] pit,hpet,tsc;
679 scx200_hrt on
680 [MIPS] MIPS
681 [PARISC] cr16
682 [S390] tod
683 [SH] SuperH
684 [SPARC64] tick
685 [X86-64] hpet,tsc
686
687 clocksource.arm_arch_timer.evtstrm=
688 [ARM,ARM64,EARLY]
689 Format: <bool>
690 Enable/disable the eve
691 architected timer so t
692 loops can be debugged
693 systems.
694
695 clocksource.verify_n_cpus= [KNL]
696 Limit the number of CP
697 marked with CLOCK_SOUR
698 are marked unstable du
699 A negative value says
700 zero says not to check
701 nr_cpu_ids are silentl
702 The actual CPUs are ch
703 no replacement if the
704
705 clocksource-wdtest.holdoff= [KNL]
706 Set the time in second
707 watchdog test waits be
708 Defaults to zero when
709 10 seconds when built
710
711 cma=nn[MG]@[start[MG][-end[MG]]]
712 [KNL,CMA,EARLY]
713 Sets the size of kerne
714 contiguous memory allo
715 placement constraint b
716 memory allocations. A
717 altogether. For more i
718 kernel/dma/contiguous.
719
720 cma_pernuma=nn[MG]
721 [KNL,CMA,EARLY]
722 Sets the size of kerne
723 contiguous memory allo
724 per-numa CMA altogethe
725 specified, the default
726 With per-numa CMA enab
727 first try to allocate
728 which is located in no
729 they will fallback to
730
731 numa_cma=<node>:nn[MG][,<node>:nn[MG]]
732 [KNL,CMA,EARLY]
733 Sets the size of kerne
734 contiguous memory allo
735 area for the specified
736
737 With numa CMA enabled,
738 first try to allocate
739 which is located in no
740 they will fallback to
741
742 cmo_free_hint= [PPC] Format: { yes |
743 Specify whether pages
744 when they are freed.
745 to determine OS memory
746 a hypervisor.
747 Default: yes
748
749 coherent_pool=nn[KMG] [ARM,KNL,EARLY
750 Sets the size of memor
751 allocations, by defaul
752
753 com20020= [HW,NET] ARCnet - COM2
754 Format:
755 <io>[,<irq>[,<nodeID>[
756
757 com90io= [HW,NET] ARCnet - COM9
758 Format: <io>[,<irq>]
759
760 com90xx= [HW,NET]
761 ARCnet - COM90xx chips
762 Format: <io>[,<irq>[,<
763
764 condev= [HW,S390] console devi
765 conmode=
766
767 con3215_drop= [S390,EARLY] 3215 cons
768 Format: y|n|Y|N|1|0
769 When set to true, drop
770 the console buffer is
771 operator using a 3270
772 x3270) does not have t
773 console output to adva
774 This leads to a much f
775 terminal emulator is a
776 emulator is used, this
777
778 console= [KNL] Output console d
779
780 tty<n> Use the virtual consol
781
782 ttyS<n>[,options]
783 ttyUSB0[,options]
784 Use the specified seri
785 the form "bbbbpnf", wh
786 "p" is parity ("n", "o
787 bits, and "f" is flow
788 omit it). Default is
789
790 See Documentation/admi
791 information. See
792 Documentation/networki
793 alternative.
794
795 <DEVNAME>:<n>.<n>[,options]
796 Use the specified seri
797 The addressing uses DE
798 device, followed by th
799 and the serial port in
800 as documented for the
801
802 The mapping of the ser
803 can be viewed with:
804
805 $ ls -d /sys/bus/seria
806 /sys/bus/serial-base/d
807
808 In the above example,
809 console=00:04:0.0. Not
810 way will only get adde
811 is ready. The use of a
812 the console may be des
813
814 uart[8250],io,<addr>[,options]
815 uart[8250],mmio,<addr>[,option
816 uart[8250],mmio16,<addr>[,opti
817 uart[8250],mmio32,<addr>[,opti
818 uart[8250],0x<addr>[,options]
819 Start an early, polled
820 UART at the specified
821 switching to the match
822 MMIO inter-register ad
823 (mmio), 16-bit (mmio16
824 If none of [io|mmio|mm
825 to be equivalent to 'm
826 the same format descri
827 the h/w is not re-init
828
829 hvc<n> Use the hypervisor con
830 both Xen and PowerPC h
831
832 { null | "" }
833 Use to disable console
834 console messages disca
835 This must be the only
836 kernel command line.
837
838 If the device connected to the
839 device, prepend "brl," before
840 console=brl,ttyS0
841 For now, only VisioBraille is
842
843 console_msg_format=
844 [KNL] Change console m
845 default
846 By default we print me
847 "[time stamp] text\n"
848 printed, depending on
849 `printk_time' param).
850 syslog
851 Switch to syslog forma
852 IOW, each message will
853 prefix. The format is
854 syscall, or to executi
855 from /proc/kmsg.
856
857 consoleblank= [KNL] The console blan
858 seconds. A value of 0
859 Defaults to 0.
860
861 coredump_filter=
862 [KNL] Change the defau
863 /proc/<pid>/coredump_f
864 See also Documentation
865
866 coresight_cpu_debug.enable
867 [ARM,ARM64]
868 Format: <bool>
869 Enable/disable the CPU
870 0: default value, disa
871 1: enable debugging at
872
873 cpcihp_generic= [HW,PCI] Generic port
874 Format:
875 <first_slot>,<last_slo
876
877 cpuidle.off=1 [CPU_IDLE]
878 disable the cpuidle su
879
880 cpuidle.governor=
881 [CPU_IDLE] Name of the
882
883 cpufreq.off=1 [CPU_FREQ]
884 disable the cpufreq su
885
886 cpufreq.default_governor=
887 [CPU_FREQ] Name of the
888 policy to use. This go
889 kernel before the cpuf
890
891 cpu_init_udelay=N
892 [X86,EARLY] Delay for
893 of APIC INIT to start
894 on every CPU online, s
895 Default: 10000
896
897 cpuhp.parallel=
898 [SMP] Enable/disable p
899 Format: <bool>
900 Default is enabled if
901 the parameter has no e
902
903 crash_kexec_post_notifiers
904 Run kdump after runnin
905 kmsg. This only for th
906 succeeds in any situat
907 Note that this also in
908 because some panic not
909 kernel more unstable.
910
911 crashkernel=size[KMG][@offset[KMG]]
912 [KNL,EARLY] Using kexe
913 upon panic. This param
914 memory region [offset,
915 image. If '@offset' is
916 is selected automatica
917 [KNL, X86-64, ARM64, R
918 under 4G first, and fa
919 4G when '@offset' hasn
920 See Documentation/admi
921
922 crashkernel=range1:size1[,range2:size2
923 [KNL] Same as above, b
924 in the running system.
925 start-[end] where star
926 a memory unit (amount[
927 Documentation/admin-gu
928
929 crashkernel=size[KMG],high
930 [KNL, X86-64, ARM64, R
931 above 4G.
932 Allow kernel to alloca
933 so could be above 4G i
934 installed. Otherwise m
935 below 4G, if available
936 It will be ignored if
937 crashkernel=size[KMG],low
938 [KNL, X86-64, ARM64, R
939 When crashkernel=X,hig
940 physical memory region
941 crash on system that r
942 e.g. swiotlb requires
943 enough extra low memor
944 for 32-bit devices won
945 default size of memory
946 size is platform depen
947 --> x86: max(swiotlb
948 --> arm64: 128MiB
949 --> riscv: 128MiB
950 --> loongarch: 128Mi
951 This one lets the user
952 for second kernel inst
953 0: to disable low allo
954 It will be ignored whe
955 or memory reserved is
956
957 cryptomgr.notests
958 [KNL] Disable crypto s
959
960 cs89x0_dma= [HW,NET]
961 Format: <dma>
962
963 cs89x0_media= [HW,NET]
964 Format: { rj45 | aui |
965
966 csdlock_debug= [KNL] Enable or disabl
967 function call handling
968 additional debug data
969 in case a hanging CPU
970 CPU is pinged again in
971 the hang situation. T
972 option depends on the
973 Kconfig option.
974
975 dasd= [HW,NET]
976 See header of drivers/
977
978 db9.dev[2|3]= [HW,JOY] Multisystem j
979 (one device per port)
980 Format: <port#>,<type>
981 See also Documentation
982
983 debug [KNL,EARLY] Enable ker
984
985 debug_boot_weak_hash
986 [KNL,EARLY] Enable pri
987 boot sequence. If ena
988 of siphash to hash poi
989 seeing instances of '(
990 value (hashed pointer)
991 insecure, please do no
992
993 debug_locks_verbose=
994 [KNL] verbose locking
995 Format: <int>
996 Print debugging info w
997 self-tests.
998 Bitmask for the variou
999 (no extra messages), s
1000 will print _a_lot_ mo
1001 useful to lockdep dev
1002
1003 debug_objects [KNL,EARLY] Enable ob
1004
1005 debug_guardpage_minorder=
1006 [KNL,EARLY] When CONF
1007 parameter allows cont
1008 be intentionally kept
1009 buddy allocator. Bigg
1010 of catching random me
1011 amount of memory for
1012 possible value is MAX
1013 parameter to 1 or 2 s
1014 random memory corrupt
1015 kernel or driver code
1016 from) a random memory
1017 a class of memory cor
1018 H/W or F/W or by driv
1019 (basically when memor
1020 CPU MMU is bypassed)
1021 CONFIG_DEBUG_PAGEALLO
1022 help tracking down th
1023
1024 debug_pagealloc=
1025 [KNL,EARLY] When CONF
1026 enables the feature a
1027 disabled and the syst
1028 kernel built without
1029 Note: to get most of
1030 useful to also enable
1031 on: enable the featur
1032
1033 debugfs= [KNL,EARLY] This para
1034 userspace and debugfs
1035 Format: { on, no-moun
1036 on: All functions
1037 no-mount:
1038 Filesystem is
1039 access APIs a
1040 its content.
1041 off: Filesystem is
1042 get a -EPERM
1043 or directorie
1044 This is equiv
1045 debugfs was n
1046 Default value is set
1047
1048 debugpat [X86] Enable PAT debu
1049
1050 default_hugepagesz=
1051 [HW] The size of the
1052 the size represented
1053 APIs. In addition, t
1054 used for shmget(), mm
1055 filesystems. If not
1056 architecture's defaul
1057 sizes are architectur
1058 Documentation/admin-g
1059 Format: size[KMG]
1060
1061 deferred_probe_timeout=
1062 [KNL] Debugging optio
1063 deferred probe to giv
1064 probe. Only specific
1065 drivers) that have op
1066 of 0 will timeout at
1067 out hasn't expired, i
1068 successful driver reg
1069 dump out devices stil
1070 retrying.
1071
1072 delayacct [KNL] Enable per-task
1073
1074 dell_smm_hwmon.ignore_dmi=
1075 [HW] Continue probing
1076 indicates that the dr
1077 hardware.
1078
1079 dell_smm_hwmon.force=
1080 [HW] Activate driver
1081 not match list of sup
1082 blacklisted features.
1083
1084 dell_smm_hwmon.power_status=
1085 [HW] Report power sta
1086 (disabled by default)
1087
1088 dell_smm_hwmon.restricted=
1089 [HW] Allow controllin
1090 capability is set.
1091
1092 dell_smm_hwmon.fan_mult=
1093 [HW] Factor to multip
1094
1095 dell_smm_hwmon.fan_max=
1096 [HW] Maximum configur
1097
1098 dfltcc= [HW,S390]
1099 Format: { on | off |
1100 on: s390 zlib h
1101 level 1 and
1102 off: No s390 zli
1103 def_only: s390 zlib h
1104 only (compr
1105 inf_only: s390 zlib h
1106 only (decom
1107 always: Same as 'on
1108 level alway
1109
1110 dhash_entries= [KNL]
1111 Set number of hash bu
1112
1113 disable_1tb_segments [PPC,EARLY]
1114 Disables the use of 1
1115 causes the kernel to
1116 can be useful when de
1117 miss to occur.
1118
1119 disable= [IPV6]
1120 See Documentation/net
1121
1122 disable_radix [PPC,EARLY]
1123 Disable RADIX MMU mod
1124
1125 disable_tlbie [PPC]
1126 Disable TLBIE instruc
1127 with KVM, with HASH M
1128
1129 disable_ddw [PPC/PSERIES,EARLY]
1130 Disable Dynamic DMA W
1131 to workaround buggy f
1132
1133 disable_ipv6= [IPV6]
1134 See Documentation/net
1135
1136 disable_mtrr_cleanup [X86,EARLY]
1137 The kernel tries to a
1138 to discrete, to make
1139 entry later. This par
1140
1141 disable_mtrr_trim [X86, Intel and AMD
1142 By default the kernel
1143 memory out of your av
1144 MTRR settings. This
1145 possibly causing your
1146
1147 disable_timer_pin_1 [X86,EARLY]
1148 Disable PIN 1 of APIC
1149 Can be useful to work
1150
1151 dis_ucode_ldr [X86] Disable the mic
1152
1153 dma_debug=off If the kernel is comp
1154 this option disables
1155
1156 dma_debug_entries=<number>
1157 This option allows to
1158 entries for DMA-API d
1159 required per DMA-API
1160 DMA-API debugging cod
1161 architectural default
1162
1163 dma_debug_driver=<driver_name>
1164 With this option the
1165 filter feature can be
1166 pass the driver to fi
1167 The filter can be dis
1168 driver later using sy
1169
1170 reg_file_data_sampling=
1171 [X86] Controls mitiga
1172 Sampling (RFDS) vulne
1173 vulnerability which m
1174 kernel data values pr
1175 registers, vector reg
1176 RFDS only affects Int
1177
1178 on: Turns ON the
1179 off: Turns OFF the
1180
1181 This parameter overri
1182 by CONFIG_MITIGATION_
1183 disabled when other V
1184 are enabled. In order
1185 VERW based mitigation
1186
1187 For details see:
1188 Documentation/admin-g
1189
1190 driver_async_probe= [KNL]
1191 List of driver names
1192 matches with all driv
1193 rest of the listed dr
1194 match the *.
1195 Format: <driver_name1
1196
1197 drm.edid_firmware=[<connector>:]<file
1198 Broken monitors, grap
1199 panels may send no or
1200 This parameter allows
1201 in the /lib/firmware
1202 An EDID data set will
1203 connector, if its nam
1204 the EDID name. Each c
1205 data set by separatin
1206 data set with no conn
1207 any connectors not ex
1208
1209 dscc4.setup= [NET]
1210
1211 dt_cpu_ftrs= [PPC,EARLY]
1212 Format: {"off" | "kno
1213 Control how the dt_cp
1214 used for CPU feature
1215 exists).
1216 off: Do not use it, f
1217 known: Do not pass th
1218 or userspace, only th
1219
1220 dump_apple_properties [X86]
1221 Dump name and content
1222 x86 Macs. Useful for
1223 what data is availabl
1224
1225 dyndbg[="val"] [KNL,DYNAMIC_
1226 <module>.dyndbg[="val"]
1227 Enable debug messages
1228 Documentation/admin-g
1229 for details.
1230
1231 early_ioremap_debug [KNL,EARLY]
1232 Enable debug messages
1233 is useful for trackin
1234 which are not unmappe
1235
1236 earlycon= [KNL,EARLY] Output ea
1237
1238 When used with no opt
1239 determined by stdout-
1240 chosen node or the AC
1241 the platform.
1242
1243 cdns,<addr>[,options]
1244 Start an early, polle
1245 (xuartps) serial port
1246 supported option is b
1247 specified, the serial
1248 configured.
1249
1250 uart[8250],io,<addr>[,options
1251 uart[8250],mmio,<addr>[,optio
1252 uart[8250],mmio32,<addr>[,opt
1253 uart[8250],mmio32be,<addr>[,o
1254 uart[8250],0x<addr>[,options]
1255 Start an early, polle
1256 UART at the specified
1257 MMIO inter-register a
1258 (mmio) or 32-bit (mmi
1259 If none of [io|mmio|m
1260 to be equivalent to '
1261 in the same format de
1262 unspecified, the h/w
1263 the uart clock freque
1264 to 'BASE_BAUD' * 16.
1265
1266 pl011,<addr>
1267 pl011,mmio32,<addr>
1268 Start an early, polle
1269 port at the specified
1270 must already be setup
1271 yet supported. If 'm
1272 the driver will use o
1273 the device registers.
1274
1275 liteuart,<addr>
1276 Start an early consol
1277 specified address. Th
1278 setup and configured.
1279
1280 meson,<addr>
1281 Start an early, polle
1282 port at the specified
1283 already be setup and
1284 supported.
1285
1286 msm_serial,<addr>
1287 Start an early, polle
1288 port at the specified
1289 must already be setup
1290 yet supported.
1291
1292 msm_serial_dm,<addr>
1293 Start an early, polle
1294 dm port at the specif
1295 must already be setup
1296 yet supported.
1297
1298 owl,<addr>
1299 Start an early, polle
1300 of an Actions Semi So
1301 specified address. Th
1302 setup and configured.
1303
1304 rda,<addr>
1305 Start an early, polle
1306 of an RDA Micro SoC,
1307 specified address. Th
1308 setup and configured.
1309
1310 sbi
1311 Use RISC-V SBI (Super
1312 console.
1313
1314 smh Use ARM semihosting c
1315
1316 s3c2410,<addr>
1317 s3c2412,<addr>
1318 s3c2440,<addr>
1319 s3c6400,<addr>
1320 s5pv210,<addr>
1321 exynos4210,<addr>
1322 Use early console pro
1323 on Samsung SoCs, requ
1324 a correct base addres
1325 serial port must alre
1326 Options are not yet s
1327
1328 lantiq,<addr>
1329 Start an early, polle
1330 (lqasc) port at the s
1331 must already be setup
1332 yet supported.
1333
1334 lpuart,<addr>
1335 lpuart32,<addr>
1336 Use early console pro
1337 found on Freescale Vy
1338 A valid base address
1339 port must already be
1340
1341 ec_imx21,<addr>
1342 ec_imx6q,<addr>
1343 Start an early, polle
1344 Freescale i.MX UART a
1345 must already be setup
1346
1347 ar3700_uart,<addr>
1348 Start an early, polle
1349 Armada 3700 serial po
1350 address. The serial p
1351 and configured. Optio
1352
1353 qcom_geni,<addr>
1354 Start an early, polle
1355 Generic Interface (GE
1356 specified address. Th
1357 setup and configured.
1358
1359 efifb,[options]
1360 Start an early, unacc
1361 memory mapped framebu
1362 coherent non-x86 syst
1363 the framebuffer, pass
1364 mapped with the corre
1365
1366 linflex,<addr>
1367 Use early console pro
1368 serial driver for NXP
1369 address must be provi
1370 already be setup and
1371
1372 earlyprintk= [X86,SH,ARM,M68k,S390
1373 earlyprintk=vga
1374 earlyprintk=sclp
1375 earlyprintk=xen
1376 earlyprintk=serial[,t
1377 earlyprintk=serial[,0
1378 earlyprintk=ttySn[,ba
1379 earlyprintk=dbgp[debu
1380 earlyprintk=pciserial
1381 earlyprintk=xdbc[xhci
1382 earlyprintk=bios
1383
1384 earlyprintk is useful
1385 the normal console is
1386 default because it ha
1387
1388 Append ",keep" to not
1389 takes over.
1390
1391 Only one of vga, seri
1392 be used at a time.
1393
1394 Currently only ttyS0
1395 name. Other I/O port
1396 on some architectures
1397 replacing ttySn with
1398 earlyprintk=s
1399 You can find the port
1400 /proc/tty/driver/seri
1401 2: uart:ST166
1402
1403 Interaction with the
1404 very good.
1405
1406 The VGA output is eve
1407 the real console.
1408
1409 The xen option can on
1410
1411 The sclp output can o
1412
1413 The bios output can o
1414
1415 The optional "force"
1416 PCI device even when
1417 UART class.
1418
1419 edac_report= [HW,EDAC] Control how
1420 Format: {"on" | "off"
1421 on: enable EDAC to re
1422 by other higher prior
1423 off: disable H/W even
1424 force: enforce the us
1425 default: on.
1426
1427 edd= [EDD]
1428 Format: {"off" | "on"
1429
1430 efi= [EFI,EARLY]
1431 Format: { "debug", "d
1432 "nochunk",
1433 "novamap",
1434 debug: enable misc de
1435 disable_early_pci_dma
1436 PCI bridges while in
1437 nochunk: disable read
1438 boot stub, as chunkin
1439 firmware implementati
1440 noruntime : disable E
1441 nosoftreserve: The EF
1442 attribute may cause t
1443 memory range for a me
1444 claim. Specify efi=no
1445 reservation and treat
1446 (i.e. EFI_CONVENTIONA
1447 novamap: do not call
1448 no_disable_early_pci_
1449 on all PCI bridges wh
1450
1451 efi_no_storage_paranoia [EFI,X86,EARL
1452 Using this parameter
1453 your efi variable sto
1454 you are really sure t
1455 fulfills the spec oth
1456
1457 efivar_ssdt= [EFI; X86] Name of an
1458 that is to be dynamic
1459 multiple variables wi
1460 vendor GUIDs, all of
1461 Documentation/admin-g
1462
1463
1464 eisa_irq_edge= [PARISC,HW]
1465 See header of drivers
1466
1467 ekgdboc= [X86,KGDB,EARLY] Allo
1468 Format: ekgdboc=kbd
1469
1470 This is designed to b
1471 the boot argument: ea
1472
1473 This parameter works
1474 but can only be used
1475 very early in the boo
1476 via a serial port see
1477
1478 elanfreq= [X86-32]
1479 See comment before fu
1480 arch/x86/kernel/cpu/c
1481
1482 elfcorehdr=[size[KMG]@]offset[KMG] [P
1483 Specifies physical ad
1484 image elf header and
1485 kexec loader will pas
1486 See Documentation/adm
1487
1488 enable_mtrr_cleanup [X86,EARLY]
1489 The kernel tries to a
1490 to discrete, to make
1491 entry later. This par
1492
1493 enable_timer_pin_1 [X86]
1494 Enable PIN 1 of APIC
1495 Can be useful to work
1496 (in particular on som
1497 The kernel tries to s
1498
1499 enforcing= [SELINUX] Set initial
1500 Format: {"0" | "1"}
1501 See security/selinux/
1502 0 -- permissive (log
1503 1 -- enforcing (deny
1504 Default value is 0.
1505 Value can be changed
1506 /sys/fs/selinux/enfor
1507
1508 erst_disable [ACPI]
1509 Disable Error Record
1510 support.
1511
1512 ether= [HW,NET] Ethernet car
1513 This option is obsole
1514 has equivalent usage.
1515
1516 evm= [EVM]
1517 Format: { "fix" }
1518 Permit 'security.evm'
1519 current integrity sta
1520
1521 early_page_ext [KNL,EARLY] Enforces p
1522 stages so cover more
1523 Please note that as s
1524 might be disabled to
1525 memory initialization
1526 might take longer, es
1527 memory. Available wit
1528
1529 failslab=
1530 fail_usercopy=
1531 fail_page_alloc=
1532 fail_make_request=[KNL]
1533 General fault injecti
1534 Format: <interval>,<p
1535 See also Documentatio
1536
1537 fb_tunnels= [NET]
1538 Format: { initns | no
1539 See Documentation/adm
1540 fb_tunnels_only_for_i
1541
1542 floppy= [HW]
1543 See Documentation/adm
1544
1545 forcepae [X86-32]
1546 Forcefully enable Phy
1547 Many Pentium M system
1548 functionally usable P
1549 Warning: use of this
1550 and may cause unknown
1551
1552 fred= [X86-64]
1553 Enable/disable Flexib
1554 Format: { on | off }
1555 on: enable FRED when
1556 off: disable FRED, th
1557
1558 ftrace=[tracer]
1559 [FTRACE] will set and
1560 as early as possible
1561 boot debugging.
1562
1563 ftrace_boot_snapshot
1564 [FTRACE] On boot up,
1565 ftrace ring buffer th
1566 /sys/kernel/tracing/s
1567 This is useful if you
1568 boot up that is likel
1569 start up functionalit
1570
1571 Optionally, the snaps
1572 instance that was cre
1573 line parameter.
1574
1575 trace_instance=foo,sc
1576
1577 The above will cause
1578 a snapshot at the end
1579
1580 ftrace_dump_on_oops[=2(orig_cpu) | =<
1581 ,<instance>=2(orig_
1582 [FTRACE] will dump th
1583 If no parameter is pa
1584 buffers of all CPUs,
1585 will dump only the bu
1586 the oops, or the spec
1587 its name is passed. M
1588 supported, and instan
1589 instance supports onl
1590 oops by passing 2 or
1591
1592 ftrace_dump_on_oops=f
1593
1594 The above will dump o
1595 on CPU that triggered
1596
1597 ftrace_dump_on_oops,f
1598
1599 The above will dump g
1600 buffer of "foo" insta
1601 of "bar" instance on
1602
1603 ftrace_filter=[function-list]
1604 [FTRACE] Limit the fu
1605 tracer at boot up. fu
1606 list of functions. Th
1607 time by the set_ftrac
1608 tracing directory.
1609
1610 ftrace_notrace=[function-list]
1611 [FTRACE] Do not trace
1612 function-list. This l
1613 by the set_ftrace_not
1614 tracing directory.
1615
1616 ftrace_graph_filter=[function-list]
1617 [FTRACE] Limit the to
1618 by the function graph
1619 function-list is a co
1620 that can be changed a
1621 set_graph_function fi
1622
1623 ftrace_graph_notrace=[function-list]
1624 [FTRACE] Do not trace
1625 function-list. This
1626 functions that can be
1627 set_graph_notrace fil
1628
1629 ftrace_graph_max_depth=<uint>
1630 [FTRACE] Used with th
1631 the max depth it will
1632 can be changed at run
1633 in the tracefs tracin
1634
1635 fw_devlink= [KNL,EARLY] Create de
1636 devices by scanning t
1637 consumer/supplier rel
1638 especially useful whe
1639 it ensures proper ord
1640 (suppliers first, the
1641 clean up (only after
1642 suspend/resume & runt
1643 suppliers).
1644 Format: { off | permi
1645 off -- Don't create
1646 permissive -- Create
1647 but use it on
1648 up (sync_stat
1649 on -- Create device
1650 to enforce pr
1651 rpm -- Like "on", bu
1652
1653 fw_devlink.strict=<bool>
1654 [KNL,EARLY] Treat all
1655 dependencies. This on
1656 Format: <bool>
1657
1658 fw_devlink.sync_state =
1659 [KNL,EARLY] When all
1660 probing, this paramet
1661 devices that haven't
1662 calls.
1663 Format: { strict | ti
1664 strict -- Default. Co
1665 probe success
1666 timeout -- Give up wa
1667 sync_state()
1668 received thei
1669 deferred_prob
1670 late_initcall
1671
1672 gamecon.map[2|3]=
1673 [HW,JOY] Multisystem
1674 support via parallel
1675 Format: <port#>,<pad1
1676 See also Documentatio
1677
1678 gamma= [HW,DRM]
1679
1680 gart_fix_e820= [X86-64,EARLY] disabl
1681 Format: off | on
1682 default: on
1683
1684 gather_data_sampling=
1685 [X86,INTEL,EARLY] Con
1686 mitigation.
1687
1688 Gather Data Sampling
1689 allows unprivileged s
1690 previously stored in
1691
1692 This issue is mitigat
1693 The mitigation may ha
1694 disabled. On systems
1695 disabling AVX serves
1696
1697 force: Disable AVX t
1698 microcode mit
1699 mitigation is
1700 userspace wit
1701
1702 off: Disable GDS m
1703
1704 gcov_persist= [GCOV] When non-zero
1705 kernel modules is sav
1706 debugfs, even when th
1707 When zero, profiling
1708 debugfs files are rem
1709
1710 goldfish [X86] Enable the gold
1711 Don't use this when y
1712 android emulator
1713
1714 gpio-mockup.gpio_mockup_ranges
1715 [HW] Sets the ranges
1716 Format: <start1>,<end
1717 gpio-mockup.gpio_mockup_named_lines
1718 [HW] Let the driver k
1719
1720 gpt [EFI] Forces disk wit
1721 invalid Protective MB
1722 primary GPT is corrup
1723 GPT to be used instea
1724
1725 grcan.enable0= [HW] Configuration of
1726 the "Enable 0" bit of
1727 Format: 0 | 1
1728 Default: 0
1729 grcan.enable1= [HW] Configuration of
1730 the "Enable 0" bit of
1731 Format: 0 | 1
1732 Default: 0
1733 grcan.select= [HW] Select which phy
1734 Format: 0 | 1
1735 Default: 0
1736 grcan.txsize= [HW] Sets the size of
1737 Format: <unsigned int
1738 Default: 1024
1739 grcan.rxsize= [HW] Sets the size of
1740 Format: <unsigned int
1741 Default: 1024
1742
1743 hardened_usercopy=
1744 [KNL] Under CONFIG_HA
1745 hardening is enabled
1746 usercopy checking is
1747 from reading or writi
1748 allocation boundaries
1749 against bounds-checki
1750 copy_to_user()/copy_f
1751 on Perform hardened user
1752 off Disable hardened user
1753
1754 hardlockup_all_cpu_backtrace=
1755 [KNL] Should the hard
1756 backtraces on all cpu
1757 Format: 0 | 1
1758
1759 hashdist= [KNL,NUMA] Large hash
1760 are distributed acros
1761 for 64-bit NUMA, off
1762 Format: 0 | 1 (for of
1763
1764 hd= [EIDE] (E)IDE hard dr
1765 Format: <cyl>,<head>,
1766
1767 hest_disable [ACPI]
1768 Disable Hardware Erro
1769 corresponding firmwar
1770 logic will be disable
1771
1772 hibernate= [HIBERNATION]
1773 noresume Don't check i
1774 present durin
1775 nocompress Don't compres
1776 no Disable hiber
1777 protect_image Turn on image
1778 (that will se
1779 during restor
1780
1781 hibernate.compressor= [HIBERNATION]
1782 used with hib
1783 Format: { lzo
1784 Default: lzo
1785
1786 lzo: Select L
1787 compress/deco
1788
1789 lz4: Select L
1790 compress/deco
1791
1792 highmem=nn[KMG] [KNL,BOOT,EARLY] forc
1793 size of <nn>. This wo
1794 highmem otherwise. Th
1795 size on bigger boxes.
1796
1797 highres= [KNL] Enable/disable
1798 Valid parameters: "on
1799 Default: "on"
1800
1801 hlt [BUGS=ARM,SH]
1802
1803 hostname= [KNL,EARLY] Set the h
1804 Format: <string>
1805 This allows setting t
1806 startup. This sets th
1807 Using this parameter
1808 possible to ensure th
1809 any userspace process
1810 that a process may ca
1811 has been explicitly s
1812 process getting an in
1813 not exceed the maximu
1814 64 characters) and wi
1815
1816 hpet= [X86-32,HPET] option
1817 Format: { enable (def
1818 verbose }
1819 disable: disable HPET
1820 force: allow force en
1821 VIA, nVidia)
1822 verbose: show content
1823
1824 hpet_mmap= [X86, HPET_MMAP] Allo
1825 registers. Default s
1826
1827 hugepages= [HW] Number of HugeTL
1828 If this follows hugep
1829 the number of pages o
1830 If this is the first
1831 line, it specifies th
1832 the default huge page
1833 number of pages to al
1834 See also Documentatio
1835 Format: <integer> or
1836 <node>:<integ
1837
1838 hugepagesz=
1839 [HW] The size of the
1840 conjunction with huge
1841 pages of a specific s
1842 hugepagesz=X hugepage
1843 each supported huge p
1844 architecture dependen
1845 Documentation/admin-g
1846 Format: size[KMG]
1847
1848 hugetlb_cma= [HW,CMA,EARLY] The si
1849 of gigantic hugepages
1850 of a CMA area per nod
1851 Format: nn[KMGTPE] or
1852 <node>:nn[KMG
1853
1854 Reserve a CMA area of
1855 hugepages using the C
1856 boot-time allocation
1857
1858 hugetlb_free_vmemmap=
1859 [KNL] Requires CONFIG
1860 enabled.
1861 Control if HugeTLB Vm
1862 Allows heavy hugetlb
1863 memory (7 * PAGE_SIZE
1864 Format: { on | off (d
1865
1866 on: enable HVO
1867 off: disable HVO
1868
1869 Built with CONFIG_HUG
1870 the default is on.
1871
1872 Note that the vmemmap
1873 memory block itself w
1874 enabled, those vmemma
1875 feature is enabled.
1876 the added memory bloc
1877
1878 hung_task_panic=
1879 [KNL] Should the hung
1880 Format: 0 | 1
1881
1882 A value of 1 instruct
1883 hung task is detected
1884 by the CONFIG_BOOTPAR
1885 option. The value sel
1886 be changed later by t
1887
1888 hvc_iucv= [S390] Number of z/V
1889 terminal devi
1890 hvc_iucv_allow= [S390] Comma-separat
1891 If specified,
1892 from listed z
1893
1894 hv_nopvspin [X86,HYPER_V,EARLY]
1895 Disables the paravirt
1896 which allow the hyper
1897 on lock contention.
1898
1899 i2c_bus= [HW] Override the
1900 or register a
1901 registered fr
1902 Format:
1903 <bus_id>,<clk
1904
1905 i2c_touchscreen_props= [HW,ACPI,X86]
1906 Set device-properties
1907 touchscreen, to e.g.
1908 mounted touchscreens.
1909 submit a drivers/plat
1910 adding a DMI quirk fo
1911
1912 Format:
1913 <ACPI_HW_ID>:<prop_na
1914 Where <val> is one of
1915 Omit "=<val>" entirel
1916 Unsigned number
1917 Anything else
1918
1919 Examples (split over
1920 i2c_touchscreen_props
1921 touchscreen-inverted-
1922
1923 i2c_touchscreen_props
1924 touchscreen-size-y=10
1925 firmware-name=gsl1680
1926
1927 i8042.debug [HW] Toggle i8042 deb
1928 i8042.unmask_kbd_data
1929 [HW] Enable printing
1930 (disabled by def
1931 requires that i8
1932 i8042.direct [HW] Put keyboard por
1933 i8042.dumbkbd [HW] Pretend that con
1934 keyboard and can
1935 (Don't attempt t
1936 i8042.noaux [HW] Don't check for
1937 i8042.nokbd [HW] Don't check/crea
1938 i8042.noloop [HW] Disable the AUX
1939 for the AUX port
1940 i8042.nomux [HW] Don't check pres
1941 controller
1942 i8042.nopnp [HW] Don't use ACPIPn
1943 controllers
1944 i8042.notimeout [HW] Ignore timeout c
1945 i8042.reset [HW] Reset the contro
1946 suspend-to-ram t
1947 transitions, or
1948 Format: { 1 | Y | y |
1949 1, Y, y: always reset
1950 0, N, n: don't ever r
1951 Default: only on s2r
1952 architectures force r
1953 i8042.unlock [HW] Unlock (ignore)
1954 i8042.kbdreset [HW] Reset device con
1955 i8042.probe_defer
1956 [HW] Allow deferred p
1957
1958 i810= [HW,DRM]
1959
1960 i915.invert_brightness=
1961 [DRM] Invert the sens
1962 set the brightness of
1963 brightness value of 0
1964 and the maximum of th
1965 to maximum brightness
1966 (default) and the mac
1967 is set to 1, a bright
1968 to maximum brightness
1969 value switches the ba
1970 -1 -- never invert br
1971 0 -- machine default
1972 1 -- force brightnes
1973
1974 ia32_emulation= [X86-64]
1975 Format: <bool>
1976 When true, allows loa
1977 syscalls, essentially
1978 boot time. When false
1979
1980 icn= [HW,ISDN]
1981 Format: <io>[,<membas
1982
1983
1984 idle= [X86,EARLY]
1985 Format: idle=poll, id
1986 Poll forces a polling
1987 improve the performan
1988 will use a lot of pow
1989 Not recommended.
1990 idle=halt: Halt is fo
1991 In such case C2/C3 wo
1992 idle=nomwait: Disable
1993
1994 idxd.sva= [HW]
1995 Format: <bool>
1996 Allow force disabling
1997 support for the idxd
1998 true (1).
1999
2000 idxd.tc_override= [HW]
2001 Format: <bool>
2002 Allow override of def
2003 for the device. By de
2004
2005 ieee754= [MIPS] Select IEEE St
2006 Format: { strict | le
2007 Default: strict
2008
2009 Choose which programs
2010 based on the IEEE 754
2011 the FPU and the NaN e
2012 of an ELF file header
2013 binary. Hardware imp
2014 support either or bot
2015 encoding mode.
2016
2017 Available settings ar
2018 strict accept binari
2019 supported by
2020 legacy only accept l
2021 by the FPU
2022 2008 only accept 2
2023 by the FPU
2024 relaxed accept any bi
2025 supported by
2026 emulated accept any b
2027 if binary mod
2028
2029 The FPU emulator is a
2030 encodings, so if no F
2031 been disabled with 'n
2032 'legacy' and '2008' s
2033 'relaxed' straps the
2034 2008-NaN, whereas 'st
2035 legacy processors and
2036 MIPS64 CPUs.
2037
2038 The setting for ABS.f
2039 mode generally follow
2040 except where unsuppor
2041
2042 ignore_loglevel [KNL,EARLY]
2043 Ignore loglevel setti
2044 kernel messages to th
2045 We also add it as pri
2046 could change it dynam
2047 /sys/module/printk/pa
2048
2049 ignore_rlimit_data
2050 Ignore RLIMIT_DATA se
2051 print warning at firs
2052 /sys/module/kernel/pa
2053
2054 ihash_entries= [KNL]
2055 Set number of hash bu
2056
2057 ima_appraise= [IMA] appraise integr
2058 Format: { "off" | "en
2059 default: "enforce"
2060
2061 ima_appraise_tcb [IMA] Deprecated. U
2062 The builtin appraise
2063 owned by uid=0.
2064
2065 ima_canonical_fmt [IMA]
2066 Use the canonical for
2067 measurements, instead
2068
2069 ima_hash= [IMA]
2070 Format: { md5 | sha1
2071 | sha512 |
2072 default: "sha1"
2073
2074 The list of supported
2075 in crypto/hash_info.h
2076
2077 ima_policy= [IMA]
2078 The builtin policies
2079 Format: "tcb | apprai
2080 fail_securel
2081
2082 The "tcb" policy meas
2083 mmap'd for exec, and
2084 mode bit set by eithe
2085 uid=0.
2086
2087 The "appraise_tcb" po
2088 all files owned by ro
2089
2090 The "secure_boot" pol
2091 of files (eg. kexec k
2092 firmware, policy, etc
2093
2094 The "fail_securely" p
2095 verification failure
2096 filesystems with the
2097 flag.
2098
2099 The "critical_data" p
2100 critical data.
2101
2102 ima_tcb [IMA] Deprecated. Us
2103 Load a policy which m
2104 Computing Base. This
2105 programs exec'd, file
2106 opened for read by ui
2107
2108 ima_template= [IMA]
2109 Select one of defined
2110 Formats: { "ima" | "i
2111 "ima-sigv2
2112 Default: "ima-ng"
2113
2114 ima_template_fmt=
2115 [IMA] Define a custom
2116 Format: { "field1|...
2117
2118 ima.ahash_minsize= [IMA] Minimum file
2119 Format: <min_file_siz
2120 Set the minimal file
2121 If left unspecified,
2122
2123 ahash performance var
2124 different crypto acce
2125 to achieve the best p
2126
2127 ima.ahash_bufsize= [IMA] Asynchronous
2128 Format: <bufsize>
2129 Set hashing buffer si
2130
2131 ahash performance var
2132 different crypto acce
2133 to achieve best perfo
2134
2135 init= [KNL]
2136 Format: <full_path>
2137 Run specified binary
2138 process.
2139
2140 initcall_debug [KNL] Trace initcalls
2141 for working out where
2142 startup.
2143
2144 initcall_blacklist= [KNL] Do not exe
2145 initcall functions.
2146 modules and initcalls
2147
2148 initramfs_async= [KNL]
2149 Format: <bool>
2150 Default: 1
2151 This parameter contro
2152 image is unpacked asy
2153 with devices being pr
2154 initialized. This sho
2155 but as a debugging ai
2156 historical behaviour
2157 unpacking being compl
2158 late_ initcalls.
2159
2160 initrd= [BOOT,EARLY] Specify
2161
2162 initrdmem= [KNL,EARLY] Specify a
2163 load the initrd. If a
2164 specified in the boot
2165 setting.
2166 Format: ss[KMG],nn[KM
2167 Default is 0, 0
2168
2169 init_on_alloc= [MM,EARLY] Fill newly
2170 zeroes.
2171 Format: 0 | 1
2172 Default set by CONFIG
2173
2174 init_on_free= [MM,EARLY] Fill freed
2175 Format: 0 | 1
2176 Default set by CONFIG
2177
2178 init_pkru= [X86] Specify the def
2179 register contents for
2180 default (disallow acc
2181 override in debugfs a
2182
2183 inport.irq= [HW] Inport (ATI XL a
2184 Format: <irq>
2185
2186 int_pln_enable [X86] Enable power li
2187
2188 integrity_audit=[IMA]
2189 Format: { "0" | "1" }
2190 0 -- basic integrity
2191 1 -- additional integ
2192
2193 intel_iommu= [DMAR] Intel IOMMU dr
2194 on
2195 Enable intel iommu dr
2196 off
2197 Disable intel iommu d
2198 igfx_off [Default Off]
2199 By default, gfx is ma
2200 device has a dedicate
2201 bypassed by not enabl
2202 this case, gfx device
2203 DMA.
2204 strict [Default Off]
2205 Deprecated, equivalen
2206 sp_off [Default Off]
2207 By default, super pag
2208 has the capability. W
2209 not be supported.
2210 sm_on
2211 Enable the Intel IOMM
2212 advertises that it ha
2213 translation.
2214 sm_off
2215 Disallow use of the I
2216 tboot_noforce [Default Off]
2217 Do not force the Inte
2218 By default, tboot wil
2219 could harm performanc
2220 devices like 40GBit n
2221 mapping is enabled.
2222 Note that using this
2223 provided by tboot bec
2224 vulnerable to DMA att
2225
2226 intel_idle.max_cstate= [KNL,HW,ACPI,
2227 0 disables inte
2228 1 to 9 specify maxim
2229
2230 intel_pstate= [X86,EARLY]
2231 disable
2232 Do not enable intel
2233 scaling driver for
2234 active
2235 Use intel_pstate dr
2236 governors layer of
2237 algorithms for p-st
2238 P-state selection a
2239 intel_pstate in the
2240 performance. The w
2241 on whether or not t
2242 (HWP) feature has b
2243 and possibly on the
2244 passive
2245 Use intel_pstate as
2246 to work with generi
2247 enabling its intern
2248 used along with the
2249 feature.
2250 force
2251 Enable intel_pstate
2252 in favor of acpi-cp
2253 instead of acpi-cpu
2254 as thermal controls
2255 P-States informatio
2256 should be used with
2257 processors that are
2258 or on platforms tha
2259 no_hwp
2260 Do not enable hardw
2261 if available.
2262 hwp_only
2263 Only load intel_pst
2264 hardware P state co
2265 support_acpi_ppc
2266 Enforce ACPI _PPC p
2267 Description Table,
2268 profile as "Enterpr
2269 then this feature i
2270 per_cpu_perf_limits
2271 Allow per-logical-C
2272 cpufreq sysfs inter
2273
2274 intremap= [X86-64,Intel-IOMMU,E
2275 on enable Interr
2276 off disable Inter
2277 nosid disable Sourc
2278 no_x2apic_optout
2279 BIOS x2APIC o
2280 nopost disable Inter
2281 posted_msi
2282 enable MSIs d
2283
2284 iomem= Disable strict checki
2285 strict regions from userspac
2286 relaxed
2287
2288 iommu= [X86,EARLY]
2289 off
2290 force
2291 noforce
2292 biomerge
2293 panic
2294 nopanic
2295 merge
2296 nomerge
2297 soft
2298 pt [X86]
2299 nopt [X86]
2300 nobypass [PPC/POWERNV]
2301 Disable IOMMU bypass,
2302
2303 iommu.forcedac= [ARM64,X86,EARLY] Con
2304 Format: { "0" | "1" }
2305 0 - Try to allocate a
2306 falling back to the
2307 1 - Allocate directly
2308 forcing Dual Addres
2309 greater than 32-bit
2310
2311 iommu.strict= [ARM64,X86,S390,EARLY
2312 Format: { "0" | "1" }
2313 0 - Lazy mode.
2314 Request that DMA un
2315 invalidation of har
2316 throughput at the c
2317 Will fall back to s
2318 the relevant IOMMU
2319 1 - Strict mode.
2320 DMA unmap operation
2321 synchronously.
2322 unset - Use value of
2323 Note: on x86, strict
2324 legacy driver-specifi
2325
2326 iommu.passthrough=
2327 [ARM64,X86,EARLY] Con
2328 Format: { "0" | "1" }
2329 0 - Use IOMMU transla
2330 1 - Bypass the IOMMU
2331 unset - Use value of
2332
2333 io7= [HW] IO7 for Marvel-b
2334 See comment before ma
2335 arch/alpha/kernel/cor
2336
2337 io_delay= [X86,EARLY] I/O delay
2338 0x80
2339 Standard port 0x80 ba
2340 0xed
2341 Alternate port 0xed b
2342 udelay
2343 Simple two microsecon
2344 none
2345 No delay
2346
2347 ip= [IP_PNP]
2348 See Documentation/adm
2349
2350 ipcmni_extend [KNL,EARLY] Extend th
2351 IPC identifiers from
2352
2353 irqaffinity= [SMP] Set the default
2354 The argument is a cpu
2355
2356 irqchip.gicv2_force_probe=
2357 [ARM,ARM64,EARLY]
2358 Format: <bool>
2359 Force the kernel to l
2360 of a GICv2 controller
2361 exposed by the device
2362
2363 irqchip.gicv3_nolpi=
2364 [ARM,ARM64,EARLY]
2365 Force the kernel to i
2366 LPIs (and by conseque
2367 that use the kernel a
2368 to let secondary kern
2369 LPIs.
2370
2371 irqchip.gicv3_pseudo_nmi= [ARM64,EARL
2372 Enables support for p
2373 requires the kernel t
2374 CONFIG_ARM64_PSEUDO_N
2375
2376 irqfixup [HW]
2377 When an interrupt is
2378 for it. Intended to g
2379 firmware running.
2380
2381 irqpoll [HW]
2382 When an interrupt is
2383 for it. Also check al
2384 interrupt. Intended t
2385 firmware running.
2386
2387 isapnp= [ISAPNP]
2388 Format: <RDP>,<reset>
2389
2390 isolcpus= [KNL,SMP,ISOL] Isolat
2391 [Deprecated - use cpu
2392 Format: [flag-list,]<
2393
2394 Specify one or more C
2395 specified in the flag
2396
2397 nohz
2398 Disable the tick wh
2399
2400 A residual 1Hz tick
2401 need to affine to h
2402 workqueue's affinit
2403 /sys/devices/virtua
2404 by using the 'domai
2405
2406 NOTE: by default th
2407 so to protect indiv
2408 be configured manua
2409
2410 domain
2411 Isolate from the ge
2412 algorithms. Note th
2413 is irreversible: it
2414 the domains once is
2415 advised to use cpus
2416 balancing through t
2417 It offers a much mo
2418 move in and out of
2419
2420 You can move a proc
2421 the CPU affinity sy
2422 <cpu number> begins
2423 "number of CPUs in
2424
2425 managed_irq
2426
2427 Isolate from being
2428 which have an inter
2429 CPUs. The affinity
2430 handled by the kern
2431 the /proc/irq/* int
2432
2433 This isolation is b
2434 if the automaticall
2435 device queue contai
2436 CPUs. If housekeepi
2437 interrupts are dire
2438 so that IO submitte
2439 cannot disturb the
2440
2441 If a queue's affini
2442 CPUs then this para
2443 interrupt routing d
2444 only delivered when
2445 isolated CPUs submi
2446 housekeeping CPUs h
2447 queues.
2448
2449 The format of <cpu-li
2450
2451 iucv= [HW,NET]
2452
2453 ivrs_ioapic [HW,X86-64]
2454 Provide an override t
2455 mapping provided in t
2456 By default, PCI segme
2457
2458 For example, to map I
2459 PCI segment 0x1 and P
2460 write the parameter a
2461 ivrs_ioapic=1
2462
2463 Deprecated formats:
2464 * To map IOAPIC-ID de
2465 write the parameter
2466 ivrs_ioapic[1
2467 * To map IOAPIC-ID de
2468 PCI device 00:14.0
2469 ivrs_ioapic[1
2470
2471 ivrs_hpet [HW,X86-64]
2472 Provide an override t
2473 mapping provided in t
2474 By default, PCI segme
2475
2476 For example, to map H
2477 PCI segment 0x1 and P
2478 write the parameter a
2479 ivrs_hpet=10@
2480
2481 Deprecated formats:
2482 * To map HPET-ID deci
2483 write the parameter
2484 ivrs_hpet[0]=
2485 * To map HPET-ID deci
2486 PCI device 00:14.0
2487 ivrs_ioapic[1
2488
2489 ivrs_acpihid [HW,X86-64]
2490 Provide an override t
2491 mapping provided in t
2492 By default, PCI segme
2493
2494 For example, to map U
2495 PCI segment 0x1 and P
2496 write the parameter a
2497 ivrs_acpihid=
2498
2499 Deprecated formats:
2500 * To map UART-HID:UID
2501 PCI device ID 00:14
2502 ivrs_acpihid[
2503 * To map UART-HID:UID
2504 PCI device ID 00:14
2505 ivrs_acpihid[
2506
2507 js= [HW,JOY] Analog joyst
2508 See Documentation/inp
2509
2510 kasan_multi_shot
2511 [KNL] Enforce KASAN (
2512 report on every inval
2513 parameter KASAN will
2514 invalid access.
2515
2516 keep_bootcon [KNL,EARLY]
2517 Do not unregister boo
2518 useful for debugging
2519 between unregistering
2520 the real console.
2521
2522 keepinitrd [HW,ARM] See retain_i
2523
2524 kernelcore= [KNL,X86,PPC,EARLY]
2525 Format: nn[KMGTPE] |
2526 This parameter specif
2527 the kernel for non-mo
2528 amount is spread even
2529 system as ZONE_NORMAL
2530 movable memory in its
2531 event, a node is too
2532 ZONE_MOVABLE, kernelc
2533 other nodes will have
2534
2535 ZONE_MOVABLE is used
2536 may be reclaimed or m
2537 subsystem. Note that
2538 still use the HighMem
2539 zone if it does not.
2540
2541 It is possible to spe
2542 the form of "nn[KMGTP
2543 memory in the form of
2544 option is specified,
2545 for non-movable alloc
2546 for Movable pages. "
2547 are exclusive, so you
2548
2549 kgdbdbgp= [KGDB,HW,EARLY] kgdb
2550 Format: <Controller#>
2551 The controller # is t
2552 port as it is probed
2553 optional and is the n
2554 each poll cycle to th
2555 the functionality for
2556 gdb or control-c on t
2557 not using this parame
2558 the kernel debugger.
2559
2560 kgdboc= [KGDB,HW] kgdb over c
2561 Requires a tty driver
2562 or a supported pollin
2563 Serial only format:
2564 keyboard only format
2565 keyboard and serial
2566 Optional Kernel mode
2567 kms, kbd format: kms
2568 kms, kbd and serial
2569
2570 kgdboc_earlycon= [KGDB,HW,EARL
2571 If the boot console p
2572 characters and can wo
2573 this parameter to tel
2574 until the normal cons
2575 be used together with
2576 specifies the normal
2577
2578 The name of the early
2579 as the value of this
2580 the early console mig
2581 name passed to kgdboc
2582 blank and the first b
2583 read() will be picked
2584
2585 kgdbwait [KGDB,EARLY] Stop ker
2586 kernel debugger at th
2587
2588 kmac= [MIPS] Korina etherne
2589 Configure the RouterB
2590 Ethernet adapter MAC
2591
2592 kmemleak= [KNL,EARLY] Boot-time
2593 Valid arguments: on,
2594 Default: on
2595 Built with CONFIG_DEB
2596 the default is off.
2597
2598 kprobe_event=[probe-list]
2599 [FTRACE] Add kprobe e
2600 The probe-list is a s
2601 definitions. Each def
2602 interface, but the pa
2603 For example, to add a
2604 arg1 and arg2, add to
2605
2606 kprobe_event=p,
2607
2608 See also Documentatio
2609 Boot Parameter" secti
2610
2611 kpti= [ARM64,EARLY] Control
2612 user and kernel addre
2613 Default: enabled on c
2614 0: force disabled
2615 1: force enabled
2616
2617 kunit.enable= [KUNIT] Enable execut
2618 CONFIG_KUNIT to be se
2619 default value can be
2620 KUNIT_DEFAULT_ENABLED
2621 Default is 1 (enabled
2622
2623 kvm.ignore_msrs=[KVM] Ignore guest ac
2624 Default is 0 (don't i
2625
2626 kvm.eager_page_split=
2627 [KVM,X86] Controls wh
2628 proactively split all
2629 Eager page splitting
2630 execution by eliminat
2631 and MMU lock contenti
2632 required to split hug
2633
2634 VM workloads that rar
2635 only to a small regio
2636 disabling eager page
2637 still be used for rea
2638
2639 The behavior of eager
2640 KVM_DIRTY_LOG_INITIAL
2641 disabled, all huge pa
2642 split when dirty logg
2643 enabled, eager page s
2644 the KVM_CLEAR_DIRTY i
2645 cleared.
2646
2647 Eager page splitting
2648
2649 Default is Y (on).
2650
2651 kvm.enable_vmware_backdoor=[KVM] Supp
2652 Default is
2653
2654 kvm.nx_huge_pages=
2655 [KVM] Controls the so
2656 X86_BUG_ITLB_MULTIHIT
2657 force : Always depl
2658 off : Never deplo
2659 auto : Deploy work
2660 X86_BUG_ITL
2661
2662 Default is 'auto'.
2663
2664 If the software worka
2665 guests do need not to
2666
2667 kvm.nx_huge_pages_recovery_ratio=
2668 [KVM] Controls how ma
2669 back to huge pages.
2670 the value is N KVM wi
2671 period (see below).
2672
2673 kvm.nx_huge_pages_recovery_period_ms=
2674 [KVM] Controls the ti
2675 back to huge pages. I
2676 zap a portion (see ra
2677 If the value is 0 (th
2678 on the ratio, such th
2679
2680 kvm-amd.nested= [KVM,AMD] Control nes
2681 KVM/SVM. Default is 1
2682
2683 kvm-amd.npt= [KVM,AMD] Control KVM
2684 a.k.a. Two-Dimensiona
2685 (enabled). Disable by
2686 for NPT.
2687
2688 kvm-arm.mode=
2689 [KVM,ARM,EARLY] Selec
2690 operation.
2691
2692 none: Forcefully disa
2693
2694 nvhe: Standard nVHE-b
2695 protected guest
2696
2697 protected: nVHE-based
2698 state is k
2699
2700 nested: VHE-based mod
2701 virtualizatio
2702 hardware.
2703
2704 Defaults to VHE/nVHE
2705 mode to "protected" w
2706 for the host. "nested
2707 used with extreme cau
2708
2709 kvm-arm.vgic_v3_group0_trap=
2710 [KVM,ARM,EARLY] Trap
2711 system registers
2712
2713 kvm-arm.vgic_v3_group1_trap=
2714 [KVM,ARM,EARLY] Trap
2715 system registers
2716
2717 kvm-arm.vgic_v3_common_trap=
2718 [KVM,ARM,EARLY] Trap
2719 system registers
2720
2721 kvm-arm.vgic_v4_enable=
2722 [KVM,ARM,EARLY] Allow
2723 injection of LPIs.
2724
2725 kvm-arm.wfe_trap_policy=
2726 [KVM,ARM] Control whe
2727 KVM VMs. Traps are al
2728 CPU architecture.
2729
2730 trap: set WFE instruc
2731
2732 notrap: clear WFE ins
2733
2734 kvm-arm.wfi_trap_policy=
2735 [KVM,ARM] Control whe
2736 KVM VMs. Traps are al
2737 CPU architecture.
2738
2739 trap: set WFI instruc
2740
2741 notrap: clear WFI ins
2742
2743 kvm_cma_resv_ratio=n [PPC,EARLY]
2744 Reserves given percen
2745 contiguous memory all
2746 allocation.
2747 By default it reserve
2748 Format: <integer>
2749 Default: 5
2750
2751 kvm-intel.ept= [KVM,Intel] Control K
2752 a.k.a. Two-Dimensiona
2753 (enabled). Disable by
2754 for EPT.
2755
2756 kvm-intel.emulate_invalid_guest_state
2757 [KVM,Intel] Control w
2758 state. Ignored if kvm
2759 as guest state is nev
2760 guests. This param do
2761 as KVM never emulates
2762 Default is 1 (enabled
2763
2764 kvm-intel.flexpriority=
2765 [KVM,Intel] Control K
2766 (TPR shadow). Default
2767 hardware lacks suppor
2768
2769 kvm-intel.nested=
2770 [KVM,Intel] Control n
2771 KVM/VMX. Default is 1
2772
2773 kvm-intel.unrestricted_guest=
2774 [KVM,Intel] Control K
2775 feature (virtualized
2776 is 1 (enabled). Disab
2777 hardware lacks suppor
2778
2779 kvm-intel.vmentry_l1d_flush=[KVM,Inte
2780 CVE-2018-3620.
2781
2782 Valid arguments: neve
2783
2784 always: L1D cache flu
2785 cond: Flush L1D on
2786 VMEXIT and VM
2787 never: Disables the
2788
2789 Default is cond (do L
2790
2791 kvm-intel.vpid= [KVM,Intel] Control K
2792 Identification featur
2793 (enabled). Disable by
2794 for it.
2795
2796 l1d_flush= [X86,INTEL,EARLY]
2797 Control mitigation fo
2798
2799 Certain CPUs are vuln
2800 internal buffers whic
2801 disclosure gadget und
2802
2803 In vulnerable process
2804 forwarded data can be
2805 attack, to access dat
2806 not have direct acces
2807
2808 This parameter contro
2809 options are:
2810
2811 on - enable t
2812
2813 l1tf= [X86,EARLY] Control m
2814 affected CPUs
2815
2816 The kernel PTE invers
2817 enabled and cannot be
2818
2819 full
2820 Provides all
2821 L1TF vulnerab
2822 enables all m
2823 hypervisors,
2824
2825 SMT control a
2826 sysfs interfa
2827 boot. Hyperv
2828 when the firs
2829 potentially i
2830 i.e. SMT enab
2831
2832 full,force
2833 Same as 'full
2834 flush runtime
2835 'nosmt=force'
2836 (i.e. sysfs c
2837
2838 flush
2839 Leaves SMT en
2840 hypervisor mi
2841 L1D flush.
2842
2843 SMT control a
2844 sysfs interfa
2845 boot. Hyperv
2846 when the firs
2847 potentially i
2848 i.e. SMT enab
2849
2850 flush,nosmt
2851
2852 Disables SMT
2853 hypervisor mi
2854
2855 SMT control a
2856 sysfs interfa
2857 boot. Hyperv
2858 when the firs
2859 potentially i
2860 i.e. SMT enab
2861
2862 flush,nowarn
2863 Same as 'flus
2864 warn when a V
2865 insecure conf
2866
2867 off
2868 Disables hype
2869 emit any warn
2870 It also drops
2871 RAM limit res
2872 bare metal.
2873
2874 Default is 'flush'.
2875
2876 For details see: Docu
2877
2878 l2cr= [PPC]
2879
2880 l3cr= [PPC]
2881
2882 lapic [X86-32,APIC,EARLY] E
2883 disabled it.
2884
2885 lapic= [X86,APIC] Do not use
2886 value for LAPIC timer
2887 back to the programma
2888 Format: notscdeadline
2889
2890 lapic_timer_c2_ok [X86,APIC,EAR
2891 in C2 power state.
2892
2893 libata.dma= [LIBATA] DMA control
2894 libata.dma=0 Dis
2895 libata.dma=1 PAT
2896 libata.dma=2 ATA
2897 libata.dma=4 Com
2898 Combinations also wor
2899 for disks and CDROMs,
2900
2901 libata.ignore_hpa= [LIBATA] Igno
2902 libata.ignore_hpa=0
2903 libata.ignore_hpa=1
2904
2905 libata.noacpi [LIBATA] Disables use
2906 when set.
2907 Format: <int>
2908
2909 libata.force= [LIBATA] Force config
2910 separated list of "[I
2911 PORT and DEVICE are d
2912 or device. Basically
2913 printed on console by
2914 omitted, the last POR
2915 ID hasn't been specif
2916 to all ports, links a
2917
2918 If only DEVICE is omi
2919 the port and all link
2920 number of 0 either se
2921 first fan-out link be
2922 select the host link.
2923 host link and device
2924
2925 The VAL specifies the
2926 as there is no ambigu
2927 For example, both 1.5
2928 The following configu
2929
2930 * Cable type: 40c, 80
2931 Any ID with matchin
2932
2933 * SATA link speed lim
2934
2935 * Transfer mode: pio[
2936 udma[/][16,25,33,44
2937 allowed.
2938
2939 * nohrst, nosrst, nor
2940 resets.
2941
2942 * rstonce: only attem
2943 link recovery.
2944
2945 * [no]dbdelay: Enable
2946 before debouncing a
2947 detection.
2948
2949 * [no]ncq: Turn on or
2950
2951 * [no]ncqtrim: Enable
2952
2953 * [no]ncqati: Enable
2954
2955 * [no]trim: Enable or
2956
2957 * trim_zero: Indicate
2958
2959 * max_trim_128m: Set
2960
2961 * [no]dma: Turn on or
2962
2963 * atapi_dmadir: Enabl
2964
2965 * atapi_mod16_dma: En
2966 commands that are n
2967
2968 * [no]dmalog: Enable
2969 READ LOG DMA EXT co
2970
2971 * [no]iddevlog: Enabl
2972 identify device dat
2973
2974 * [no]logdir: Enable
2975 purpose log directo
2976
2977 * max_sec_128: Set tr
2978
2979 * max_sec_1024: Set o
2980 1024 sectors.
2981
2982 * max_sec_lba48: Set
2983 65535 sectors.
2984
2985 * [no]lpm: Enable or
2986
2987 * [no]setxfer: Indica
2988 should be skipped.
2989
2990 * [no]fua: Disable or
2991 support for devices
2992
2993 * dump_id: Dump IDENT
2994
2995 * disable: Disable th
2996
2997 If there are multiple
2998 the same attribute, t
2999
3000 load_ramdisk= [RAM] [Deprecated]
3001
3002 lockd.nlm_grace_period=P [NFS] Assig
3003 Format: <integer>
3004
3005 lockd.nlm_tcpport=N [NFS] Assign
3006 Format: <integer>
3007
3008 lockd.nlm_timeout=T [NFS] Assign
3009 Format: <integer>
3010
3011 lockd.nlm_udpport=M [NFS] Assign
3012 Format: <integer>
3013
3014 lockdown= [SECURITY,EARLY]
3015 { integrity | confide
3016 Enable the kernel loc
3017 integrity, kernel fea
3018 modify the running ke
3019 confidentiality, kern
3020 to extract confidenti
3021 are also disabled.
3022
3023 locktorture.acq_writer_lim= [KNL]
3024 Set the time limit in
3025 acquisition. Acquisi
3026 will result in a spla
3027
3028 locktorture.bind_readers= [KNL]
3029 Specify the list of C
3030 to be bound.
3031
3032 locktorture.bind_writers= [KNL]
3033 Specify the list of C
3034 to be bound.
3035
3036 locktorture.call_rcu_chains= [KNL]
3037 Specify the number of
3038 chains to set up. Th
3039 there is a high proba
3040 in progress at any gi
3041 which disables these
3042
3043 locktorture.long_hold= [KNL]
3044 Specify the duration
3045 occasional long-durat
3046 to 100 milliseconds.
3047
3048 locktorture.nested_locks= [KNL]
3049 Specify the maximum l
3050 locktorture is to exe
3051 (MAX_NESTED_LOCKS).
3052 Note that this parame
3053 of locks that do not
3054
3055 locktorture.nreaders_stress= [KNL]
3056 Set the number of loc
3057 Defaults to being aut
3058 number of online CPUs
3059
3060 locktorture.nwriters_stress= [KNL]
3061 Set the number of loc
3062
3063 locktorture.onoff_holdoff= [KNL]
3064 Set time (s) after bo
3065
3066 locktorture.onoff_interval= [KNL]
3067 Set time (s) between
3068 zero to disable CPU-h
3069
3070 locktorture.rt_boost= [KNL]
3071 Do periodic testing o
3072 boosting. Select 0 t
3073 only rt_mutex, and 2
3074 Defaults to 2, which
3075 odd choice, but which
3076 non-real-time spinloc
3077 of preemption. Note
3078 disable boosting.
3079
3080 locktorture.rt_boost_factor= [KNL]
3081 Number that determine
3082 long priority boostin
3083 scaled down by the nu
3084 number of boosts per
3085 constant as the numbe
3086 On the other hand, th
3087 increases with the nu
3088
3089 locktorture.shuffle_interval= [KNL]
3090 Set task-shuffle inte
3091 tasks allows some CPU
3092 mode during the lockt
3093
3094 locktorture.shutdown_secs= [KNL]
3095 Set time (s) after bo
3096 is useful for hands-o
3097
3098 locktorture.stat_interval= [KNL]
3099 Time (s) between stat
3100
3101 locktorture.stutter= [KNL]
3102 Time (s) to stutter t
3103 specifying five secon
3104 five seconds, wait fo
3105 This tests the lockin
3106 transition abruptly t
3107
3108 locktorture.torture_type= [KNL]
3109 Specify the locking i
3110
3111 locktorture.verbose= [KNL]
3112 Enable additional pri
3113
3114 locktorture.writer_fifo= [KNL]
3115 Run the write-side lo
3116 sched_set_fifo() real
3117
3118 logibm.irq= [HW,MOUSE] Logitech B
3119 Format: <irq>
3120
3121 loglevel= [KNL,EARLY]
3122 All Kernel Messages w
3123 console loglevel will
3124 also be changed with
3125 loglevels are defined
3126
3127 0 (KERN_EMERG)
3128 1 (KERN_ALERT)
3129 2 (KERN_CRIT)
3130 3 (KERN_ERR)
3131 4 (KERN_WARNING)
3132 5 (KERN_NOTICE)
3133 6 (KERN_INFO)
3134 7 (KERN_DEBUG)
3135
3136 log_buf_len=n[KMG] [KNL,EARLY]
3137 Sets the size of the
3138 n must be a power of
3139 minimal size. The min
3140 LOG_BUF_SHIFT kernel
3141 is also CONFIG_LOG_CP
3142 parameter that allows
3143 depending on the numb
3144 for more details.
3145
3146 logo.nologo [FB] Disables display
3147 This may be used to p
3148 kernel log messages a
3149 kernel boot problems.
3150
3151 lp=0 [LP] Specify paral
3152 lp=port[,port...] lp=none,parpo
3153 lp=reset first paralle
3154 lp=auto printer drive
3155 specified in
3156 attached prin
3157 lp=port1,port
3158 to associate
3159 lp0. A port s
3160 that lp devic
3161 'parport0'. S
3162 port specific
3163 from each por
3164 an IEEE 1284-
3165 so, the drive
3166 See also head
3167
3168 lpj=n [KNL]
3169 Sets loops_per_jiffy
3170 time-consuming boot-t
3171 CPU). 0 enables autod
3172 the correct value for
3173 autodetection and see
3174 on SMP systems the pr
3175 which is likely to ca
3176 significantly diverge
3177 will cause delays in
3178 unpredictable I/O err
3179 unlikely, in the extr
3180 hardware.
3181
3182 lsm.debug [SECURITY] Enable LSM
3183
3184 lsm=lsm1,...,lsmN
3185 [SECURITY] Choose ord
3186 overrides CONFIG_LSM,
3187
3188 machtype= [Loongson] Share the
3189 different yeeloong la
3190 Example: machtype=lem
3191
3192 maxcpus= [SMP,EARLY] Maximum n
3193 will bring up during
3194 the kernel to bring u
3195 bootup you can bring
3196 "echo 1 > /sys/device
3197 only takes effect dur
3198 While n=0 is a specia
3199 which also disables t
3200
3201 max_loop= [LOOP] The number of
3202 (loop.max_loop) unconditionally pre-c
3203 number is configured
3204 of statically allocat
3205 devices can be reques
3206 /dev/loop-control int
3207
3208 mce [X86-32] Machine Chec
3209
3210 mce=option [X86-64] See Document
3211
3212 md= [HW] RAID subsystems
3213 See Documentation/adm
3214
3215 mdacon= [MDA]
3216 Format: <first>,<last
3217 Specifies range of co
3218
3219 mds= [X86,INTEL,EARLY]
3220 Control mitigation fo
3221 Sampling (MDS) vulner
3222
3223 Certain CPUs are vuln
3224 internal buffers whic
3225 disclosure gadget und
3226
3227 In vulnerable process
3228 forwarded data can be
3229 attack, to access dat
3230 not have direct acces
3231
3232 This parameter contro
3233 options are:
3234
3235 full - Enable M
3236 full,nosmt - Enable M
3237 SMT on v
3238 off - Uncondit
3239
3240 On TAA-affected machi
3241 an active TAA mitigat
3242 mitigated with the sa
3243 this mitigation, you
3244 too.
3245
3246 Not specifying this o
3247 mds=full.
3248
3249 For details see: Docu
3250
3251 mem=nn[KMG] [HEXAGON,EARLY] Set t
3252 Must be specified, ot
3253
3254 mem=nn[KMG] [KNL,BOOT,EARLY] Forc
3255 of memory Amount of m
3256 as follows:
3257
3258 1 for test;
3259 2 when the kernel is
3260 3 memory that lies af
3261 the hypervisor, then
3262 4 to limit the memory
3263
3264 [ARC,MICROBLAZE] - th
3265 high memory is not af
3266
3267 [ARM64] - only limits
3268 mapping. The NOMAP re
3269
3270 [X86] Work as limitin
3271 with memmap= to avoid
3272 Without memmap= PCI d
3273 belonging to unused R
3274
3275 Note that this only t
3276 in above case 3, memo
3277 if system memory of h
3278
3279 mem=nn[KMG]@ss[KMG]
3280 [ARM,MIPS,EARLY] - ov
3281 reported by firmware.
3282 Define a memory regio
3283 ss[KMG].
3284 Multiple different re
3285 multiple mem= paramet
3286
3287 mem=nopentium [BUGS=X86-32] Disable
3288 memory.
3289
3290 memblock=debug [KNL,EARLY] Enable me
3291
3292 memchunk=nn[KMG]
3293 [KNL,SH] Allow user t
3294 per-device physically
3295
3296 memhp_default_state=online/offline/on
3297 [KNL] Set the initial
3298 onlining policy. If n
3299 set according to the
3300 CONFIG_MEMORY_HOTPLUG
3301 option.
3302 See Documentation/adm
3303
3304 memmap=exactmap [KNL,X86,EARLY] Enabl
3305 E820 memory map, as s
3306 Such memmap=exactmap
3307 BIOS output or other
3308 option description.
3309
3310 memmap=nn[KMG]@ss[KMG]
3311 [KNL, X86,MIPS,XTENSA
3312 Region of memory to b
3313 If @ss[KMG] is omitte
3314 which limits max addr
3315 Multiple different re
3316 comma delimited.
3317 Example:
3318 memmap=100M@2
3319
3320 memmap=nn[KMG]#ss[KMG]
3321 [KNL,ACPI,EARLY] Mark
3322 Region of memory to b
3323
3324 memmap=nn[KMG]$ss[KMG]
3325 [KNL,ACPI,EARLY] Mark
3326 Region of memory to b
3327 Example: Exclude memo
3328 memmap=64K$0
3329 or
3330 memmap=0x100
3331 Some bootloaders may
3332 like Grub2, otherwise
3333 will be eaten.
3334
3335 memmap=nn[KMG]!ss[KMG,EARLY]
3336 [KNL,X86] Mark specif
3337 Region of memory to b
3338 The memory region may
3339 and is NVDIMM or ADR
3340
3341 memmap=<size>%<offset>-<oldtype>+<new
3342 [KNL,ACPI,EARLY] Conv
3343 from <oldtype> to <ne
3344 out, the whole region
3345 even if previously un
3346 out, matching memory
3347 specified as e820 typ
3348 3 = ACPI, 12 = PRAM.
3349
3350 memory_corruption_check=0/1 [X86,EARL
3351 Some BIOSes seem to c
3352 memory when doing thi
3353 Setting this option w
3354 looking for corruptio
3355 both detect corruptio
3356 from using the memory
3357 However, its intended
3358 repeatable BIOS-origi
3359 affects the same memo
3360 to prevent the kernel
3361
3362 memory_corruption_check_size=size [X8
3363 By default it checks
3364 64k, making this memo
3365 use. Use this parame
3366 corruption in more or
3367
3368 memory_corruption_check_period=second
3369 By default it checks
3370 seconds. Use this pa
3371 other rate. 0 disabl
3372
3373 memory_hotplug.memmap_on_memory
3374 [KNL,X86,ARM] Boolean
3375 Format: {on | off (de
3376 When enabled, runtime
3377 allocate its internal
3378 those vmemmap pages c
3379 if hugetlb_free_vmemm
3380 hotadded memory which
3381 lot of memory without
3382 memory to do so.
3383 This feature is disab
3384 has some implication
3385 allocations in some c
3386 memory blocks).
3387 The state of the flag
3388 /sys/module/memory_ho
3389 Note that even when e
3390 the feature is not ef
3391
3392 memtest= [KNL,X86,ARM,M68K,PPC
3393 Format: <integer>
3394 default : 0 <disable>
3395 Specifies the number
3396 performed. Each pass
3397 pattern from a given
3398 fills the memory with
3399 memory contents and r
3400 regions that are dete
3401
3402 mem_encrypt= [X86-64] AMD Secure M
3403 Valid arguments: on,
3404 Default: off
3405 mem_encrypt=on:
3406 mem_encrypt=off:
3407
3408 Refer to Documentatio
3409 for details on when m
3410
3411 mem_sleep_default= [SUSPEND] Def
3412 s2idle - Suspend-To-
3413 shallow - Power-On Su
3414 deep - Suspend-To-
3415 See Documentation/adm
3416
3417 mfgptfix [X86-32] Fix MFGPT ti
3418 the BIOS has incorrec
3419 version 0.98 is known
3420 problem by letting th
3421
3422 mga= [HW,DRM]
3423
3424 microcode.force_minrev= [X86]
3425 Format: <bool>
3426 Enable or disable the
3427 enforcement for the r
3428
3429 mini2440= [ARM,HW,KNL]
3430 Format:[0..2][b][c][t
3431 Default: "0tb"
3432 MINI2440 configuratio
3433 0 - The attached scre
3434 1 - The attached scre
3435 2 - The VGA Shield is
3436 Leaving out the scree
3437 the TFT driver, and t
3438 unconfigured.
3439 b - Enable backlight.
3440 linked to the kernel
3441 LED. This parameter i
3442 VGA shield.
3443 c - Enable the s3c ca
3444 t - Reserved for enab
3445 touchscreen support i
3446 kernel as of 2.6.30,
3447 in the "bleeding edge
3448 https://repo.or.cz/w/
3449
3450 mitigations=
3451 [X86,PPC,S390,ARM64,E
3452 CPU vulnerabilities.
3453 arch-independent opti
3454 aggregation of existi
3455
3456 Note, "mitigations" i
3457 kernel was built with
3458
3459 off
3460 Disable all o
3461 improves syst
3462 expose users
3463 Equivalent to
3464
3465
3466
3467
3468
3469
3470
3471
3472
3473
3474
3475
3476
3477
3478
3479
3480
3481
3482
3483
3484
3485
3486 Exceptions:
3487
3488
3489
3490
3491 auto (default)
3492 Mitigate all
3493 enabled, even
3494 users who don
3495 getting disab
3496 have other wa
3497 Equivalent to
3498
3499 auto,nosmt
3500 Mitigate all
3501 if needed. T
3502 be fully miti
3503 Equivalent to
3504
3505
3506
3507
3508
3509 mminit_loglevel=
3510 [KNL,EARLY] When CONF
3511 parameter allows cont
3512 the additional memory
3513 of 0 disables mminit
3514 log everything. Infor
3515 so loglevel=8 may als
3516
3517 mmio_stale_data=
3518 [X86,INTEL,EARLY] Con
3519 MMIO Stale Data vulne
3520
3521 Processor MMIO Stale
3522 vulnerabilities that
3523 operation. Exposed da
3524 the same CPU buffers
3525 Therefore, similar to
3526 is to clear the affec
3527
3528 This parameter contro
3529 options are:
3530
3531 full - Enable m
3532
3533 full,nosmt - Enable m
3534 vulnerab
3535
3536 off - Uncondit
3537
3538 On MDS or TAA affecte
3539 mmio_stale_data=off c
3540 MDS or TAA mitigation
3541 mitigated with the sa
3542 disable this mitigati
3543 mds=off and tsx_async
3544
3545 Not specifying this o
3546 mmio_stale_data=full.
3547
3548 For details see:
3549 Documentation/admin-g
3550
3551 <module>.async_probe[=<bool>] [KNL]
3552 If no <bool> value is
3553 specified is not a va
3554 probe on this module.
3555 asynchronous probe on
3556 <bool> value. See als
3557
3558 module.async_probe=<bool>
3559 [KNL] When set to tru
3560 by default. To enable
3561 specific module, use
3562 is documented under <
3563 module.async_probe an
3564 specified, <module>.a
3565 the specific module.
3566
3567 module.enable_dups_trace
3568 [KNL] When CONFIG_MOD
3569 this means that dupli
3570 trigger a WARN_ON() i
3571 if MODULE_DEBUG_AUTOL
3572 will always be issued
3573 module.sig_enforce
3574 [KNL] When CONFIG_MOD
3575 modules without (vali
3576 Note that if CONFIG_M
3577 is always true, so th
3578
3579 module_blacklist= [KNL] Do not load
3580 modules. Useful for
3581
3582 mousedev.tap_time=
3583 [MOUSE] Maximum time
3584 leaving touchpad surf
3585 a tap and be reported
3586 touchpads working in
3587 Format: <msecs>
3588 mousedev.xres= [MOUSE] Horizontal sc
3589 reporting absolute co
3590 mousedev.yres= [MOUSE] Vertical scre
3591 reporting absolute co
3592
3593 movablecore= [KNL,X86,PPC,EARLY]
3594 Format: nn[KMGTPE] |
3595 This parameter is the
3596 specifies the amount
3597 allocations. If both
3598 specified, then kerne
3599 specified value but m
3600 own is specified, the
3601 that the amount of me
3602 is not too small.
3603
3604 movable_node [KNL,EARLY] Boot-time
3605 NUMA nodes to be mova
3606 of such nodes will be
3607 allocations which rul
3608 allocations. Use with
3609
3610 MTD_Partition= [MTD]
3611 Format: <name>,<regio
3612
3613 MTD_Region= [MTD] Format:
3614 <name>,<region-number
3615
3616 mtdparts= [MTD]
3617 See drivers/mtd/parse
3618
3619 mtouchusb.raw_coordinates=
3620 [HW] Make the MicroTo
3621 ('y', default) or coo
3622
3623 mtrr=debug [X86,EARLY]
3624 Enable printing debug
3625 registers at boot tim
3626
3627 mtrr_chunk_size=nn[KMG,X86,EARLY]
3628 used for mtrr cleanup
3629 that could hold holes
3630
3631 mtrr_gran_size=nn[KMG,X86,EARLY]
3632 Used for mtrr cleanup
3633 Default is 1.
3634 Large value could pre
3635 using up MTRRs.
3636
3637 mtrr_spare_reg_nr=n [X86,EARLY]
3638 Format: <integer>
3639 Range: 0,7 : spare re
3640 Default : 1
3641 Used for mtrr cleanup
3642 Set to 2 or more if y
3643
3644 multitce=off [PPC] This parameter
3645 firmware feature for
3646 at a time.
3647
3648 n2= [NET] SDL Inc. RISCom
3649
3650 netdev= [NET] Network devices
3651 Format: <irq>,<io>,<m
3652 Note that mem_start i
3653 something different a
3654 This usage is only do
3655 file if at all.
3656
3657 netpoll.carrier_timeout=
3658 [NET] Specifies amoun
3659 netpoll should wait f
3660 waits 4 seconds.
3661
3662 nf_conntrack.acct=
3663 [NETFILTER] Enable co
3664 0 to disable accounti
3665 1 to enable accountin
3666 Default value is 0.
3667
3668 nfs.cache_getent=
3669 [NFS] sets the pathna
3670 to update the NFS cli
3671
3672 nfs.cache_getent_timeout=
3673 [NFS] sets the timeou
3674 update a cache entry
3675
3676 nfs.callback_nr_threads=
3677 [NFSv4] set the total
3678 NFS client will assig
3679 requests.
3680
3681 nfs.callback_tcpport=
3682 [NFS] set the TCP por
3683 channel should listen
3684
3685 nfs.delay_retrans=
3686 [NFS] specifies the n
3687 retries the request b
3688 after a reply of NFS4
3689 Only applies if the s
3690 and the specified val
3691
3692 nfs.enable_ino64=
3693 [NFS] enable 64-bit i
3694 If zero, the NFS clie
3695 number for the readdi
3696 of returning the full
3697 The default is to ret
3698
3699 nfs.idmap_cache_timeout=
3700 [NFS] set the maximum
3701 entries.
3702
3703 nfs.max_session_cb_slots=
3704 [NFSv4.1] Sets the ma
3705 slots the client will
3706 channel. This determi
3707 callbacks the client
3708 a particular server.
3709
3710 nfs.max_session_slots=
3711 [NFSv4.1] Sets the ma
3712 the client will attem
3713 This limits the numbe
3714 that the client can s
3715 Note that there is li
3716 value higher than the
3717
3718 nfs.nfs4_disable_idmapping=
3719 [NFSv4] When set to t
3720 ensures that both the
3721 scheme and the NFS le
3722 numeric uids/gids if
3723 'sec=sys' security fl
3724 disabling idmapping,
3725 legacy NFSv2/v3 syste
3726 Servers that do not s
3727 will be autodetected
3728 back to using the idm
3729 To turn off this beha
3730
3731 nfs.nfs4_unique_id=
3732 [NFS4] Specify an add
3733 ification string that
3734 their nfs_client_id4
3735 UUID that is generate
3736
3737 nfs.recover_lost_locks=
3738 [NFSv4] Attempt to re
3739 to a lease timeout on
3740 doing this risks data
3741 no guarantees that th
3742 after the locks are l
3743 If you want to enable
3744 attempting to recover
3745 parameter to '1'.
3746 The default parameter
3747 not to attempt recove
3748
3749 nfs.send_implementation_id=
3750 [NFSv4.1] Send client
3751 information in exchan
3752 If zero, no implement
3753 will be sent.
3754 The default is to sen
3755 information.
3756
3757 nfs4.layoutstats_timer=
3758 [NFSv4.2] Change the
3759 layoutstats to the pN
3760
3761 Setting this to value
3762 whatever value is the
3763 driver. A non-zero va
3764 in seconds between la
3765
3766 nfsd.inter_copy_offload_enable=
3767 [NFSv4.2] When set to
3768 server-to-server copi
3769 the destination of th
3770
3771 nfsd.nfs4_disable_idmapping=
3772 [NFSv4] When set to t
3773 server will return on
3774 clients using auth_sy
3775 and gids from such cl
3776 migration from NFSv2/
3777
3778 nfsd.nfsd4_ssc_umount_timeout=
3779 [NFSv4.2] When used a
3780 server-to-server copy
3781 the source server. I
3782 it will be needed aga
3783 used for the number o
3784 this parameter.
3785
3786 nfsaddrs= [NFS] Deprecated. Us
3787 See Documentation/adm
3788
3789 nfsroot= [NFS] nfs root filesy
3790 See Documentation/adm
3791
3792 nfsrootdebug [NFS] enable nfsroot
3793 See Documentation/adm
3794
3795 nmi_backtrace.backtrace_idle [KNL]
3796 Dump stacks even of i
3797 NMI stack-backtrace r
3798
3799 nmi_debug= [KNL,SH] Specify one
3800 when a NMI is trigger
3801 Format: [state][,regs
3802
3803 nmi_watchdog= [KNL,BUGS=X86] Debugg
3804 Format: [panic,][nopa
3805 Valid num: 0 or 1
3806 0 - turn hardlockup d
3807 1 - turn hardlockup d
3808 rNNN - configure the
3809
3810 When panic is specifi
3811 timeout occurs (or 'n
3812 watchdog, if CONFIG_B
3813 To disable both hard
3814 please see 'nowatchdo
3815 This is useful when y
3816 need the box quickly
3817
3818 These settings can be
3819 the nmi_watchdog and
3820
3821 no387 [BUGS=X86-32] Tells t
3822 emulation library eve
3823 is present.
3824
3825 no4lvl [RISCV,EARLY] Disable
3826 Forces kernel to use
3827
3828 no5lvl [X86-64,RISCV,EARLY]
3829 kernel to use 4-level
3830
3831 noalign [KNL,ARM]
3832
3833 noapic [SMP,APIC,EARLY] Tell
3834 IOAPICs that may be p
3835
3836 noautogroup Disable scheduler aut
3837
3838 nocache [ARM,EARLY]
3839
3840 no_console_suspend
3841 [HW] Never suspend th
3842 Disable suspending of
3843 hibernate operations.
3844 messages can reach va
3845 of the system is bein
3846 debugging driver susp
3847 not work reliably wit
3848 to work with serial a
3849 To facilitate more fl
3850 console_suspend, a pr
3851 it. Users could use c
3852 /sys/module/printk/pa
3853 turn on/off it dynami
3854
3855 no_debug_objects
3856 [KNL,EARLY] Disable o
3857
3858 nodsp [SH] Disable hardware
3859
3860 noefi [EFI,EARLY] Disable E
3861
3862 no_entry_flush [PPC,EARLY] Don't flu
3863
3864 noexec32 [X86-64]
3865 This affects only 32-
3866 noexec32=on: enable n
3867 read doesn't
3868 noexec32=off: disable
3869 read implies
3870
3871 no_file_caps Tells the kernel not
3872 only way then for a f
3873 is to be setuid root
3874
3875 nofpu [MIPS,SH] Disable har
3876
3877 nofsgsbase [X86] Disables FSGSBA
3878
3879 nofxsr [BUGS=X86-32] Disable
3880 register save and res
3881 legacy floating-point
3882
3883 no_hash_pointers
3884 [KNL,EARLY]
3885 Force pointers printe
3886 unhashed. By default
3887 format string, that p
3888 by hashing the pointe
3889 that hides actual ker
3890 users, but it also ma
3891 difficult since unequ
3892 compared. However, i
3893 specified, then all n
3894 value printed. This o
3895 debugging the kernel.
3896 kernels.
3897
3898 nohibernate [HIBERNATION] Disable
3899
3900 nohlt [ARM,ARM64,MICROBLAZE
3901 busy wait in do_idle(
3902 implementation; requi
3903 to be effective. This
3904 sleep(SH) or wfi(ARM,
3905 correctly or when doi
3906 the impact of the sle
3907 useful when using JTA
3908
3909 nohugeiomap [KNL,X86,PPC,ARM64,EA
3910
3911 nohugevmalloc [KNL,X86,PPC,ARM64,EA
3912
3913 nohz= [KNL] Boottime enable
3914 Valid arguments: on,
3915 Default: on
3916
3917 nohz_full= [KNL,BOOT,SMP,ISOL]
3918 The argument is a cpu
3919 In kernels built with
3920 the specified list of
3921 whenever possible. Th
3922 the range to maintain
3923 in this list will hav
3924 just as if they had a
3925 rcu_nocbs= boot param
3926
3927 Note that this argume
3928 the CONFIG_RCU_NOCB_C
3929
3930 noinitrd [RAM] Tells the kerne
3931 initial RAM disk.
3932
3933 nointremap [X86-64,Intel-IOMMU,E
3934 remapping.
3935 [Deprecated - use int
3936
3937 noinvpcid [X86,EARLY] Disable t
3938
3939 noiotrap [SH] Disables trapped
3940
3941 noirqdebug [X86-32] Disables the
3942 disable unhandled int
3943
3944 noisapnp [ISAPNP] Disables ISA
3945
3946 nokaslr [KNL,EARLY]
3947 When CONFIG_RANDOMIZE
3948 kernel and module bas
3949 Layout Randomization)
3950
3951 no-kvmapf [X86,KVM,EARLY] Disab
3952 fault handling.
3953
3954 no-kvmclock [X86,KVM,EARLY] Disab
3955
3956 nolapic [X86-32,APIC,EARLY] D
3957
3958 nolapic_timer [X86-32,APIC,EARLY] D
3959
3960 nomce [X86-32] Disable Mach
3961
3962 nomfgpt [X86-32] Disable Mult
3963 Timer usage (for AMD
3964
3965 nomodeset Disable kernel modese
3966 sets up a display mod
3967 for output. With nomo
3968 not load if they coul
3969 initialized output. O
3970 be available for use.
3971 perform display-mode
3972
3973 Useful as error fallb
3974
3975 nomodule Disable module load
3976
3977 nonmi_ipi [X86] Disable using N
3978 shutdown the other cp
3979 irq.
3980
3981 nopat [X86,EARLY] Disable P
3982 pagetables) support.
3983
3984 nopcid [X86-64,EARLY] Disabl
3985
3986 nopku [X86] Disable Memory
3987 in some Intel CPUs.
3988
3989 nopti [X86-64,EARLY]
3990 Equivalent to pti=off
3991
3992 nopv= [X86,XEN,KVM,HYPER_V,
3993 Disables the PV optim
3994 as generic guest with
3995 XEN HVM, KVM, HYPER_V
3996
3997 nopvspin [X86,XEN,KVM,EARLY]
3998 Disables the qspinloc
3999 which allow the hyper
4000 contention.
4001
4002 norandmaps Don't use address spa
4003 echo 0 > /proc/sys/ke
4004
4005 noreplace-smp [X86-32,SMP] Don't re
4006 with UP alternatives
4007
4008 noresume [SWSUSP] Disables res
4009 space.
4010
4011 no-scroll [VGA] Disables scroll
4012 This is required for
4013 reader made by F.H. P
4014
4015 nosgx [X86-64,SGX,EARLY] Di
4016
4017 nosmap [PPC,EARLY]
4018 Disable SMAP (Supervi
4019 even if it is support
4020
4021 nosmep [PPC64s,EARLY]
4022 Disable SMEP (Supervi
4023 even if it is support
4024
4025 nosmp [SMP,EARLY] Tells an
4026 and disable the IO AP
4027
4028 nosmt [KNL,MIPS,PPC,S390,EA
4029 Equivalent to smt=1.
4030
4031 [KNL,X86,PPC] Disable
4032 nosmt=force: Force di
4033 via the
4034
4035 nosoftlockup [KNL] Disable the sof
4036
4037 nospec_store_bypass_disable
4038 [HW,EARLY] Disable al
4039 Store Bypass vulnerab
4040
4041 nospectre_bhb [ARM64,EARLY] Disable
4042 history injection) vu
4043 with this option.
4044
4045 nospectre_v1 [X86,PPC,EARLY] Disab
4046 (bounds check bypass)
4047 possible in the syste
4048
4049 nospectre_v2 [X86,PPC_E500,ARM64,E
4050 for the Spectre varia
4051 prediction) vulnerabi
4052 leaks with this optio
4053
4054 no-steal-acc [X86,PV_OPS,ARM64,PPC
4055 Disable paravirtualiz
4056 is computed, but won'
4057
4058 nosync [HW,M68K] Disables sy
4059
4060 no_timer_check [X86,APIC] Disables t
4061 broken timer IRQ sour
4062
4063 no_uaccess_flush
4064 [PPC,EARLY] Don't flu
4065
4066 novmcoredd [KNL,KDUMP]
4067 Disable device dump.
4068 append dump data to v
4069 specified debug info.
4070 without any limit and
4071 so this may cause sig
4072 device dump can help
4073 data will be no longe
4074 is only available whe
4075 is set.
4076
4077 no-vmw-sched-clock
4078 [X86,PV_OPS,EARLY] Di
4079 scheduler clock and u
4080
4081 nowatchdog [KNL] Disable both lo
4082 soft-lockup and NMI w
4083
4084 nowb [ARM,EARLY]
4085
4086 nox2apic [X86-64,APIC,EARLY] D
4087
4088 NOTE: this parameter
4089 LEGACY_XAPIC_DISABLED
4090 IA32_XAPIC_DISABLE_ST
4091
4092 noxsave [BUGS=X86] Disables x
4093 and restore using xsa
4094 enabling legacy float
4095
4096 noxsaveopt [X86] Disables xsaveo
4097 register states. The
4098 xsave to save the sta
4099 performance of saving
4100 xsave doesn't support
4101 xsaveopt supports it
4102
4103 noxsaves [X86] Disables xsaves
4104 restoring x86 extende
4105 form of xsave area. T
4106 xsaveopt and xrstor t
4107 in standard form of x
4108 parameter, xsave area
4109 memory on xsaves enab
4110
4111 nr_cpus= [SMP,EARLY] Maximum n
4112 could support. nr_cp
4113 support 'n' processor
4114 number of already plu
4115 runtime you can physi
4116 n. So during boot up
4117 variables need be pre
4118 hot plugging.
4119
4120 nr_uarts= [SERIAL] maximum numb
4121
4122 numa=off [KNL, ARM64, PPC, RIS
4123 Disable NUMA, Only se
4124 spanning all memory.
4125
4126 numa_balancing= [KNL,ARM64,PPC,RISCV,
4127 NUMA balancing.
4128 Allowed values are en
4129
4130 numa_zonelist_order= [KNL, BOOT] Sele
4131 'node', 'default' can
4132 This can be set from
4133 See Documentation/adm
4134
4135 ohci1394_dma=early [HW,EARLY] en
4136 See Documentation/cor
4137 info.
4138
4139 olpc_ec_timeout= [OLPC] ms delay when
4140 Rather than timing ou
4141 command is not proper
4142 of the timeout. We h
4143 waiting for the ACK,
4144 interrupts *may* be l
4145
4146 omap_mux= [OMAP] Override bootl
4147 Format: <mux_mode0.mo
4148 For example, to overr
4149 omap_mux=i2c2_scl.i2c
4150
4151 onenand.bdry= [HW,MTD] Flex-OneNAND
4152
4153 Format: [die0_boundar
4154
4155 boundary - index of l
4156 The remain
4157 lock - Configure
4158 Once locke
4159 1 indicate
4160
4161 oops=panic [KNL,EARLY]
4162 Always panic on oopse
4163 process, but there is
4164 deadlocking the machi
4165 This will also cause
4166 Useful together with
4167
4168 page_alloc.shuffle=
4169 [KNL] Boolean flag to
4170 should randomize its
4171 used to enable/disabl
4172 the flag can be read
4173 /sys/module/page_allo
4174 This parameter is onl
4175
4176 page_owner= [KNL,EARLY] Boot-time
4177 Storage of the inform
4178 each page is disabled
4179 we can turn it on.
4180 on: enable the featur
4181
4182 page_poison= [KNL,EARLY] Boot-time
4183 poisoning on the budd
4184 CONFIG_PAGE_POISONING
4185 off: turn off poisoni
4186 on: turn on poisoning
4187
4188 page_reporting.page_reporting_order=
4189 [KNL] Minimal page re
4190 Format: <integer>
4191 Adjust the minimal pa
4192 reporting is disabled
4193
4194 panic= [KNL] Kernel behaviou
4195 timeout > 0: seconds
4196 timeout = 0: wait for
4197 timeout < 0: reboot i
4198 Format: <timeout>
4199
4200 panic_on_taint= [KNL,EARLY]
4201 Bitmask for condition
4202 Format: <hex>[,nouser
4203 Hexadecimal bitmask r
4204 that will cause the k
4205 called with any of th
4206 The optional switch "
4207 prevent userspace for
4208 /proc/sys/kernel/tain
4209 bitmask set on panic_
4210 See Documentation/adm
4211 extra details on the
4212 to compose the bitmas
4213
4214 panic_on_warn=1 panic() instead of WA
4215 on a WARN().
4216
4217 panic_print= Bitmask for printing
4218 User can chose combin
4219 bit 0: print all task
4220 bit 1: print system m
4221 bit 2: print timer in
4222 bit 3: print locks in
4223 bit 4: print ftrace b
4224 bit 5: print all prin
4225 bit 6: print all CPUs
4226 bit 7: print only tas
4227 *Be aware* that this
4228 so there are risks of
4229 Use this option caref
4230 bigger log buffer wit
4231
4232 parkbd.port= [HW] Parallel port nu
4233 connected to, default
4234 Format: <parport#>
4235 parkbd.mode= [HW] Parallel port ke
4236 0 for XT, 1 for AT (d
4237 Format: <mode>
4238
4239 parport= [HW,PPT] Specify para
4240 Format: { 0 | auto |
4241 Use 'auto' to force t
4242 IRQ/DMA settings dete
4243 ignore detected IRQ/D
4244 possible conflicts).
4245 address, IRQ, and DMA
4246 should be numbers, or
4247 settings on that part
4248 (to avoid using a FIF
4249 Parallel ports are as
4250 are specified on the
4251 with parport0.
4252
4253 parport_init_mode= [HW,PPT]
4254 Configure VIA paralle
4255 a specific mode. This
4256 computer where firmwa
4257 up parallel port mode
4258 Currently this functi
4259 Format: [spp|ps2|epp|
4260
4261 pata_legacy.all= [HW,LIBATA]
4262 Format: <int>
4263 Set to non-zero to pr
4264 port ranges on PCI sy
4265 has been found at eit
4266
4267 pata_legacy.autospeed= [HW,LIBATA]
4268 Format: <int>
4269 Set to non-zero if a
4270 changes. Disabled by
4271
4272 pata_legacy.ht6560a= [HW,LIBATA]
4273 Format: <int>
4274 Set to 1, 2, or 3 for
4275 the secondary channel
4276 Disabled by default.
4277
4278 pata_legacy.ht6560b= [HW,LIBATA]
4279 Format: <int>
4280 Set to 1, 2, or 3 for
4281 the secondary channel
4282 Disabled by default.
4283
4284 pata_legacy.iordy_mask= [HW,LIBATA]
4285 Format: <int>
4286 IORDY enable mask. S
4287 for the respective ch
4288 legacy channel handle
4289 the second channel, a
4290 correspond to the pri
4291 legacy channel, and s
4292 bus and the use of ot
4293 with the sequence. B
4294 all channels.
4295
4296 pata_legacy.opti82c46x= [HW,LIBATA]
4297 Format: <int>
4298 Set to 1, 2, or 3 for
4299 channel, the secondar
4300 respectively. Disabl
4301
4302 pata_legacy.opti82c611a= [HW,L
4303 Format: <int>
4304 Set to 1, 2, or 3 for
4305 channel, the secondar
4306 respectively. Disabl
4307
4308 pata_legacy.pio_mask= [HW,LIBATA]
4309 Format: <int>
4310 PIO mode mask for aut
4311 bits to allow the use
4312 Bit 0 is for mode 0,
4313 All modes allowed by
4314
4315 pata_legacy.probe_all= [HW,LIBATA]
4316 Format: <int>
4317 Set to non-zero to pr
4318 port ranges on PCI sy
4319
4320 pata_legacy.probe_mask= [HW,LIBATA]
4321 Format: <int>
4322 Probe mask for legacy
4323 platform configuratio
4324 options up to 6 legac
4325 0x170, 0x1e8, 0x168,
4326 of individual ports c
4327 corresponding bits in
4328 the first port in the
4329 By default all suppor
4330
4331 pata_legacy.qdi= [HW,LIBATA]
4332 Format: <int>
4333 Set to non-zero to pr
4334 set to 1 if CONFIG_PA
4335
4336 pata_legacy.winbond= [HW,LIBATA]
4337 Format: <int>
4338 Set to non-zero to pr
4339 the standard I/O port
4340 value given is the I/
4341 By default set to 1 i
4342 0 otherwise.
4343
4344 pata_platform.pio_mask= [HW,LIBATA]
4345 Format: <int>
4346 Supported PIO mode ma
4347 the use of the respec
4348 mode 0, bit 1 is for
4349 allowed by default.
4350
4351 pause_on_oops=<int>
4352 Halt all CPUs after t
4353 the specified number
4354 your oopses keep scro
4355
4356 pcbit= [HW,ISDN]
4357
4358 pci=option[,option...] [PCI,EARLY] v
4359
4360 Some options
4361 or a set of d
4362 specified in
4363
4364 [<domain>:]<b
4365 pci:<vendor>:
4366
4367 Note: the fir
4368 bus/device/fu
4369 if new hardwa
4370 firmware chan
4371 by other kern
4372 domain is lef
4373 taken to be z
4374 to a device t
4375 addresses can
4376 address (this
4377 renumbering i
4378 selects devic
4379 configuration
4380 devices in th
4381
4382 earlydump dump PCI conf
4383 changes anyth
4384 off [X86] don't p
4385 bios [X86-32] forc
4386 the hardware
4387 has a non-sta
4388 nobios [X86-32] disa
4389 hardware acce
4390 if you experi
4391 suspect they
4392 conf1 [X86] Force u
4393 Mechanism 1 (
4394 data in IO po
4395 conf2 [X86] Force u
4396 Mechanism 2 (
4397 the function,
4398 bus number. T
4399 through ports
4400 See http://wi
4401 on the config
4402 noaer [PCIE] If the
4403 enabled, this
4404 disable the u
4405 nodomains [PCI] Disable
4406 root domains
4407 nommconf [X86] Disable
4408 Configuration
4409 check_enable_amd_mmconf [X86]
4410 properly conf
4411 config space
4412 nomsi [MSI] If the
4413 enabled, this
4414 disable the u
4415 noioapicquirk [APIC] Disabl
4416 Safety option
4417 should never
4418 ioapicreroute [APIC] Enable
4419 primary IO-AP
4420 boot IRQs. Th
4421 when the syst
4422 noioapicreroute [APIC] Disabl
4423 boot IRQ equi
4424 a chipset whe
4425 The opposite
4426 biosirq [X86-32] Use
4427 routing table
4428 on several ma
4429 when used, bu
4430 way to get th
4431 this option i
4432 IRQs or disco
4433 motherboard.
4434 rom [X86] Assign
4435 Use with caut
4436 address decod
4437 resources.
4438 norom [X86] Do not
4439 expansion ROM
4440 BIOS assigned
4441 nobar [X86] Do not
4442 BARs that wer
4443 irqmask=0xMMMM [X86] Set a b
4444 assigned auto
4445 make the kern
4446 this way.
4447 pirqaddr=0xAAAAA [X86]
4448 of the PIRQ t
4449 by the BIOS)
4450 F0000h-100000
4451 lastbus=N [X86] Scan al
4452 useful if the
4453 secondary bus
4454 explicitly wh
4455 assign-busses [X86] Always
4456 numbers ourse
4457 whatever the
4458 usepirqmask [X86] Honor t
4459 in the BIOS $
4460 some systems
4461 some HP Pavil
4462 notebooks. Th
4463 IRQ routing i
4464 noacpi [X86] Do not
4465 or for PCI sc
4466 use_crs [X86] Use PCI
4467 from ACPI. O
4468 is enabled by
4469 please report
4470 nocrs [X86] Ignore
4471 If you need t
4472 use_e820 [X86] Use E82
4473 PCI host brid
4474 for BIOS defe
4475 If you need t
4476 <linux-pci@vge
4477 no_e820 [X86] Ignore
4478 bridge window
4479 hardware. If
4480 a bug to <linu
4481 routeirq Do IRQ routin
4482 This is norma
4483 so this optio
4484 for broken dr
4485 skip_isa_align [X86] do not
4486 handle more p
4487 noearly [X86] Don't d
4488 This might he
4489 machine check
4490 is read. But
4491 and some IOMM
4492 bfsort Sort PCI devi
4493 This sorting
4494 order compati
4495 nobfsort Don't sort PC
4496 pcie_bus_tune_off Disab
4497 tuning and us
4498 pcie_bus_safe Set every dev
4499 supported by
4500 pcie_bus_perf Set device MP
4501 based on its
4502 Read Request
4503 value (no lar
4504 or bus can su
4505 pcie_bus_peer2peer Set e
4506 every device
4507 configuration
4508 any pair of d
4509 reduced perfo
4510 that hot-adde
4511 cbiosize=nn[KMG] The f
4512 reserved for
4513 The default v
4514 cbmemsize=nn[KMG] The f
4515 reserved for
4516 window. The d
4517 resource_alignment=
4518 Format:
4519 [<order of ali
4520 Specifies ali
4521 aligned memor
4522 specify the d
4523 If <order of
4524 PAGE_SIZE is
4525 A PCI-PCI bri
4526 windows need
4527 To specify th
4528 instances of
4529 device, subve
4530 specified, e.
4531 for 4096-byte
4532 ecrc= Enable/disabl
4533 end-to-end CR
4534 OS has native
4535 ACPI _OSC or
4536 bios: Use BIO
4537 the default.
4538 off: Turn ECR
4539 on: Turn ECRC
4540 hpiosize=nn[KMG] The f
4541 reserved for
4542 Default size
4543 hpmmiosize=nn[KMG] The f
4544 reserved for
4545 Default size
4546 hpmmioprefsize=nn[KMG] The f
4547 reserved for
4548 Default size
4549 hpmemsize=nn[KMG] The f
4550 reserved for
4551 MMIO_PREF win
4552 Default size
4553 hpbussize=nn The minimum a
4554 reserved for
4555 Default is 1.
4556 realloc= Enable/disabl
4557 if allocation
4558 accommodate r
4559 devices.
4560 off: Turn rea
4561 on: Turn real
4562 realloc same as reall
4563 noari do not use PC
4564 noats [PCIE, Intel-
4565 do not use PC
4566 pcie_scan_all Scan all poss
4567 only look for
4568 port.
4569 big_root_window Try to add a
4570 root complex
4571 can resize a
4572 Adding the wi
4573 conflict with
4574 taints the ke
4575 disable_acs_redir=<pci_dev>[;
4576 Specify one o
4577 specified abo
4578 Each device s
4579 redirect capa
4580 allow P2P tra
4581 bridges witho
4582 this removes
4583 may put more
4584 config_acs=
4585 Format:
4586 <ACS flags>@<p
4587 Specify one o
4588 specified abo
4589 and separated
4590 capabilities
4591 unchanged bas
4592 flags.
4593
4594 ACS Flags is
4595 bit-0 : ACS
4596 bit-1 : ACS
4597 bit-2 : ACS
4598 bit-3 : ACS
4599 bit-4 : ACS
4600 bit-5 : ACS
4601 bit-6 : ACS
4602 Each bit can
4603 '0' – for
4604 '1' – for
4605 'x' – unc
4606 For example,
4607 pci=config_
4608 would configu
4609 ACS to enable
4610 Translation B
4611 Validation un
4612 or firmware s
4613
4614 Note: this ma
4615 and may put m
4616 force_floating [S390] Force
4617 nomio [S390] Do not
4618 norid [S390] ignore
4619 one PCI domai
4620
4621 pcie_aspm= [PCIE] Forcibly enabl
4622 Management.
4623 off Don't touch ASPM conf
4624 configuration done by
4625 force Enable ASPM even on d
4626 WARNING: Forcing ASPM
4627
4628 pcie_ports= [PCIE] PCIe port serv
4629 native Use native PCIe servi
4630 even if the platform
4631 use them. This may c
4632 also tries to use the
4633 dpc-native Use native PC
4634 cause conflic
4635 compat Disable native PCIe s
4636 hotplug).
4637
4638 pcie_port_pm= [PCIE] PCIe port powe
4639 off Disable power managem
4640 force Forcibly enable power
4641
4642 pcie_pme= [PCIE,PM] Native PCIe
4643 nomsi Do not use MSI for na
4644 all PCIe root ports u
4645
4646 pcmv= [HW,PCMCIA] BadgePAD
4647
4648 pd_ignore_unused
4649 [PM]
4650 Keep all power-domain
4651 even if no driver has
4652 for debug and develop
4653 needed on a platform
4654
4655 pdcchassis= [PARISC,HW] Disable/E
4656 boot time.
4657 Format: { 0 | 1 }
4658 See arch/parisc/kerne
4659
4660 percpu_alloc= [MM,EARLY]
4661 Select which percpu f
4662 Currently supported v
4663 Archs may support sub
4664 See comments in mm/pe
4665 allocator. This para
4666 and performance compa
4667
4668 pirq= [SMP,APIC] Manual mp-
4669 See Documentation/arc
4670
4671 plip= [PPT,NET] Parallel po
4672 Format: { parport<nr>
4673 See also Documentatio
4674
4675 pmtmr= [X86] Manual setup of
4676 Override pmtimer IOPo
4677 e.g. pmtmr=0x508
4678
4679 pmu_override= [PPC] Override the PM
4680 This option takes ove
4681 longer usable by perf
4682 PMU counters by setti
4683 cleared). If a number
4684 that number, otherwis
4685 remains 0.
4686
4687 pm_debug_messages [SUSPEND,KNL]
4688 Enable suspend/resume
4689
4690 pnp.debug=1 [PNP]
4691 Enable PNP debug mess
4692 CONFIG_PNP_DEBUG_MESS
4693 via /sys/module/pnp/p
4694 current resource usag
4695 possible settings and
4696
4697 pnpacpi= [ACPI]
4698 { off }
4699
4700 pnpbios= [ISAPNP]
4701 { on | off | curr | r
4702
4703 pnp_reserve_irq=
4704 [ISAPNP] Exclude IRQs
4705
4706 pnp_reserve_dma=
4707 [ISAPNP] Exclude DMAs
4708
4709 pnp_reserve_io= [ISAPNP] Exclude I/O
4710 Ranges are in pairs (
4711
4712 pnp_reserve_mem=
4713 [ISAPNP] Exclude memo
4714 autoconfiguration.
4715 Ranges are in pairs (
4716
4717 ports= [IP_VS_FTP] IPVS ftp
4718 Default is 21.
4719 Up to 8 (IP_VS_APP_MA
4720 may be specified.
4721 Format: <port>,<port>
4722
4723 possible_cpus= [SMP,S390,X86]
4724 Format: <unsigned int
4725 Set the number of pos
4726 regular discovery mec
4727
4728 powersave=off [PPC] This option dis
4729 It specifically disab
4730 platform machine desc
4731 function to NULL. On
4732 execution priority.
4733
4734 ppc_strict_facility_enable
4735 [PPC,ENABLE] This opt
4736 Altivec, VSX and SPE
4737 allowed (eg kernel_en
4738 There is some perform
4739
4740 ppc_tm= [PPC,EARLY]
4741 Format: {"off"}
4742 Disable Hardware Tran
4743
4744 preempt= [KNL]
4745 Select preemption mod
4746 none - Limited to con
4747 voluntary - Limited t
4748 full - Any section th
4749 can be preempt
4750 contended spin
4751 explicitly pre
4752
4753 print-fatal-signals=
4754 [KNL] debug: print fa
4755
4756 If enabled, warn abou
4757 related application a
4758 too many POSIX.1 time
4759 coredump - etc.
4760
4761 If you hit the warnin
4762 you might want to try
4763
4764 default: off.
4765
4766 printk.always_kmsg_dump=
4767 Trigger kmsg_dump for
4768 panics
4769 Format: <bool> (1/Y/
4770 default: disabled
4771
4772 printk.console_no_auto_verbose=
4773 Disable console logle
4774 or lockdep-detected i
4775 With an exception to
4776 serial console, keepi
4777 in order to provide m
4778 Format: <bool>
4779 default: 0 (auto_verb
4780
4781 printk.devkmsg={on,off,ratelimit}
4782 Control writing to /d
4783 on - unlimited loggin
4784 off - logging to /dev
4785 ratelimit - ratelimit
4786 Default: ratelimit
4787
4788 printk.time= Show timing data pref
4789 Format: <bool> (1/Y/
4790
4791 proc_mem.force_override= [KNL]
4792 Format: {always | ptr
4793 Traditionally /proc/p
4794 overridden without re
4795 restrict that. Can be
4796 - 'always': tradition
4797 - 'ptrace': only allo
4798 - 'never': never all
4799 If not specified, def
4800
4801 processor.max_cstate= [HW,ACPI]
4802 Limit processor to ma
4803 max_cstate=9 override
4804
4805 processor.nocst [HW,ACPI]
4806 Ignore the _CST metho
4807 instead using the leg
4808
4809 profile= [KNL] Enable kernel p
4810 Format: [<profiletype
4811 Param: <profiletype>:
4812 [defaults to
4813 Param: "schedule" - p
4814 Param: "kvm" - profil
4815 Param: <number> - ste
4816 statistical t
4817
4818 prompt_ramdisk= [RAM] [Deprecated]
4819
4820 prot_virt= [S390] enable hosting
4821 isolated from the hyp
4822 that). If enabled, th
4823 might be overridden e
4824 Layout Randomization
4825 Format: <bool>
4826
4827 psi= [KNL] Enable or disab
4828 tracking.
4829 Format: <bool>
4830
4831 psmouse.proto= [HW,MOUSE] Highest PS
4832 probe for; one of (ba
4833 psmouse.rate= [HW,MOUSE] Set desire
4834 per second.
4835 psmouse.resetafter= [HW,MOUSE]
4836 Try to reset the devi
4837 (0 = never).
4838 psmouse.resolution=
4839 [HW,MOUSE] Set desire
4840 psmouse.smartscroll=
4841 [HW,MOUSE] Controls L
4842 0 = disabled, 1 = ena
4843
4844 pstore.backend= Specify the name of t
4845
4846 pti= [X86-64] Control Page
4847 kernel address spaces
4848 removes hardening, bu
4849 system calls and inte
4850
4851 on - unconditionall
4852 off - unconditionall
4853 auto - kernel detects
4854 vulnerable to
4855
4856 Not specifying this o
4857
4858 pty.legacy_count=
4859 [KNL] Number of legac
4860 default number.
4861
4862 quiet [KNL,EARLY] Disable m
4863
4864 r128= [HW,DRM]
4865
4866 radix_hcall_invalidate=on [PPC/PSERI
4867 Disable RADIX GTSE fe
4868 invalidate.
4869
4870 raid= [HW,RAID]
4871 See Documentation/adm
4872
4873 ramdisk_size= [RAM] Sizes of RAM di
4874 See Documentation/adm
4875
4876 ramdisk_start= [RAM] RAM disk image
4877
4878 random.trust_cpu=off
4879 [KNL,EARLY] Disable t
4880 random number generat
4881 initialize the kernel
4882
4883 random.trust_bootloader=off
4884 [KNL,EARLY] Disable t
4885 passed by the bootloa
4886 initialize the kernel
4887
4888 randomize_kstack_offset=
4889 [KNL,EARLY] Enable or
4890 randomization, which
4891 entropy, frustrating
4892 that depend on stack
4893 cross-syscall address
4894 available on architec
4895 CONFIG_HAVE_ARCH_RAND
4896 Format: <bool> (1/Y/
4897 Default is CONFIG_RAN
4898
4899 ras=option[,option,...] [KNL] RAS-spe
4900
4901 cec_disable [X86]
4902 Disable the C
4903 see CONFIG_RA
4904
4905 rcu_nocbs[=cpu-list]
4906 [KNL] The optional ar
4907 as described above.
4908
4909 In kernels built with
4910 enable the no-callbac
4911 such CPUs' callbacks
4912 softirq context. Inv
4913 callbacks will instea
4914 kthreads created for
4915 "p" for RCU-preempt,
4916 for the kthreads that
4917 "N" is the CPU number
4918 the offloaded CPUs, w
4919 and real-time workloa
4920 energy efficiency for
4921
4922 If a cpulist is passe
4923 list of CPUs is set t
4924
4925 Otherwise, if the '='
4926 arguments are omitted
4927 no-callback mode from
4928 toggled at runtime vi
4929
4930 Note that this argume
4931 the CONFIG_RCU_NOCB_C
4932
4933 rcu_nocb_poll [KNL]
4934 Rather than requiring
4935 (specified by rcu_noc
4936 awaken the correspond
4937 make these kthreads p
4938 This improves the rea
4939 offloaded CPUs by rel
4940 wake up the correspon
4941 energy efficiency by
4942 periodically wake up
4943
4944 rcutree.blimit= [KNL]
4945 Set maximum number of
4946 process in one batch.
4947
4948 rcutree.do_rcu_barrier= [KNL]
4949 Request a call to rcu
4950 throttled so that use
4951 hammer on the sysfs v
4952 If triggered before t
4953 is fully active, this
4954
4955 rcutree.dump_tree= [KNL]
4956 Dump the structure of
4957 out at early boot. T
4958 purposes, to verify c
4959
4960 rcutree.gp_cleanup_delay= [KNL]
4961 Set the number of jif
4962 RCU grace-period clea
4963
4964 rcutree.gp_init_delay= [KNL]
4965 Set the number of jif
4966 RCU grace-period init
4967
4968 rcutree.gp_preinit_delay= [KNL]
4969 Set the number of jif
4970 RCU grace-period pre-
4971 the propagation of re
4972 the rcu_node combinin
4973
4974 rcutree.jiffies_till_first_fqs= [KNL]
4975 Set delay from grace-
4976 first attempt to forc
4977 Units are jiffies, mi
4978 and maximum value is
4979
4980 rcutree.jiffies_till_next_fqs= [KNL]
4981 Set delay between sub
4982 quiescent states. Un
4983 value is one, and max
4984
4985 rcutree.jiffies_till_sched_qs= [KNL]
4986 Set required age in j
4987 given grace period be
4988 soliciting quiescent-
4989 rcu_note_context_swit
4990 If not specified, the
4991 a value based on the
4992 of rcutree.jiffies_ti
4993 and rcutree.jiffies_t
4994 This calculated value
4995 rcutree.jiffies_to_sc
4996 rcutree.jiffies_to_sc
4997 overwritten.
4998
4999 rcutree.kthread_prio= [KNL,BOOT]
5000 Set the SCHED_FIFO pr
5001 kthreads (rcuc/N). Th
5002 the priority of the R
5003 and for the RCU grace
5004 rcu_preempt, and rcu_
5005 set, valid values are
5006 (the least-favored pr
5007 RCU_BOOST is not set,
5008 the default is zero (
5009 When RCU_NOCB_CPU is
5010 priority of NOCB call
5011
5012 rcutree.nocb_nobypass_lim_per_jiffy=
5013 On callback-offloaded
5014 RCU reduces the lock
5015 otherwise be caused b
5016 use of the ->nocb_byp
5017 common non-flooded ca
5018 the main ->cblist in
5019 overhead of the ->noc
5020 But if there are too
5021 a single jiffy, RCU p
5022 the ->nocb_bypass que
5023 many" is supplied by
5024
5025 rcutree.nohz_full_patience_delay= [KN
5026 On callback-offloaded
5027 disturbing RCU unless
5028 reached the specified
5029 Defaults to zero. La
5030 at five seconds. All
5031 to the nearest value
5032
5033 rcutree.qhimark= [KNL]
5034 Set threshold of queu
5035 batch limiting is dis
5036
5037 rcutree.qlowmark= [KNL]
5038 Set threshold of queu
5039 batch limiting is re-
5040
5041 rcutree.qovld= [KNL]
5042 Set threshold of queu
5043 RCU's force-quiescent
5044 enlist help from cond
5045 help CPUs more quickl
5046 Set to less than zero
5047 on rcutree.qhimark at
5048 disable more aggressi
5049
5050 rcutree.rcu_delay_page_cache_fill_mse
5051 Set the page-cache re
5052 in response to low-me
5053 of permitted values i
5054
5055 rcutree.rcu_divisor= [KNL]
5056 Set the shift-right c
5057 the callback-invocati
5058 the number of callbac
5059 The result will be bo
5060 the rcutree.blimit ke
5061 callbacks, the softir
5062 order to allow the CP
5063
5064 Please note that this
5065 limit applies only to
5066 invocation. Offloade
5067 invoked in the contex
5068 scheduler will preemp
5069
5070 rcutree.rcu_fanout_exact= [KNL]
5071 Disable autobalancing
5072 tree. This is used b
5073 possibly be useful fo
5074 cache-to-cache transf
5075
5076 rcutree.rcu_fanout_leaf= [KNL]
5077 Change the number of
5078 leaf rcu_node structu
5079 large systems, which
5080 and for NUMA systems
5081 latencies, which will
5082 with the appropriate
5083
5084 rcutree.rcu_min_cached_objs= [KNL]
5085 Minimum number of obj
5086 maintained per one CP
5087 to PAGE_SIZE. The cac
5088 pressure to page allo
5089 whole algorithm to be
5090 condition.
5091
5092 rcutree.rcu_nocb_gp_stride= [KNL]
5093 Set the number of NOC
5094 each group, which def
5095 of the number of CPUs
5096 the wakeup overhead o
5097 kthread, but increase
5098 each group's NOCB gra
5099
5100 rcutree.rcu_kick_kthreads= [KNL]
5101 Cause the grace-perio
5102 wake_up() if it sleep
5103 it should at force-qu
5104 This wake_up() will b
5105 WARN_ONCE() splat and
5106
5107 rcutree.rcu_resched_ns= [KNL]
5108 Limit the time spend
5109 callbacks to the spec
5110 By default, this limi
5111 every 32 callbacks in
5112 inflicted by local_cl
5113
5114 rcutree.rcu_unlock_delay= [KNL]
5115 In CONFIG_RCU_STRICT_
5116 this specifies an rcu
5117 in microseconds. Thi
5118 Larger delays increas
5119 catching RCU pointer
5120 of RCU-protected poin
5121 rcu_read_unlock() has
5122
5123 rcutree.sysrq_rcu= [KNL]
5124 Commandeer a sysrq ke
5125 rcu_node tree with an
5126 why a new grace perio
5127
5128 rcutree.use_softirq= [KNL]
5129 If set to zero, move
5130 per-CPU rcuc kthreads
5131 value, meaning that R
5132 Specify rcutree.use_s
5133
5134 But note that CONFIG_
5135 this kernel boot para
5136 to zero.
5137
5138 rcutree.enable_rcu_lazy= [KNL]
5139 To save power, batch
5140 delay, memory pressur
5141 big.
5142
5143 rcutree.rcu_normal_wake_from_gp= [KNL
5144 Reduces a latency of
5145 maintains its own tra
5146 does not interact wit
5147 use a call_rcu[_hurry
5148 normal grace period.
5149
5150 How to enable it:
5151
5152 echo 1 > /sys/module/
5153 or pass a boot parame
5154
5155 Default is 0.
5156
5157 rcuscale.gp_async= [KNL]
5158 Measure performance o
5159 grace-period primitiv
5160
5161 rcuscale.gp_async_max= [KNL]
5162 Specify the maximum n
5163 callbacks per writer
5164 thread exceeds this l
5165 corresponding flavor
5166 previously posted cal
5167
5168 rcuscale.gp_exp= [KNL]
5169 Measure performance o
5170 grace-period primitiv
5171
5172 rcuscale.holdoff= [KNL]
5173 Set test-start holdof
5174 this parameter is to
5175 test until boot compl
5176 interference.
5177
5178 rcuscale.kfree_by_call_rcu= [KNL]
5179 In kernels built with
5180 call_rcu() instead of
5181
5182 rcuscale.kfree_mult= [KNL]
5183 Instead of allocating
5184 allocate one of kfree
5185 Defaults to 1.
5186
5187 rcuscale.kfree_rcu_test= [KNL]
5188 Set to measure perfor
5189
5190 rcuscale.kfree_rcu_test_double= [KNL]
5191 Test the double-argum
5192 If this parameter has
5193 rcuscale.kfree_rcu_te
5194 and double-argument v
5195
5196 rcuscale.kfree_rcu_test_single= [KNL]
5197 Test the single-argum
5198 If this parameter has
5199 rcuscale.kfree_rcu_te
5200 and double-argument v
5201
5202 rcuscale.kfree_nthreads= [KNL]
5203 The number of threads
5204
5205 rcuscale.kfree_alloc_num= [KNL]
5206 Number of allocations
5207
5208 rcuscale.kfree_loops= [KNL]
5209 Number of loops doing
5210 of allocations and fr
5211
5212 rcuscale.minruntime= [KNL]
5213 Set the minimum test
5214 does not affect the d
5215 but instead allows be
5216 like CPU consumption.
5217
5218 rcuscale.nreaders= [KNL]
5219 Set number of RCU rea
5220 N, where N is the num
5221 "n" less than -1 sele
5222 the number of CPUs.
5223 (the number of CPUs),
5224 A value of "n" less t
5225 a single reader.
5226
5227 rcuscale.nwriters= [KNL]
5228 Set number of RCU wri
5229 the same as for rcusc
5230 N, where N is the num
5231
5232 rcuscale.scale_type= [KNL]
5233 Specify the RCU imple
5234
5235 rcuscale.shutdown= [KNL]
5236 Shut the system down
5237 complete. This is us
5238 testing.
5239
5240 rcuscale.verbose= [KNL]
5241 Enable additional pri
5242
5243 rcuscale.writer_holdoff= [KNL]
5244 Write-side holdoff be
5245 in microseconds. The
5246 no holdoff.
5247
5248 rcuscale.writer_holdoff_jiffies= [KNL
5249 Additional write-side
5250 periods, but in jiffi
5251 says no holdoff.
5252
5253 rcutorture.fqs_duration= [KNL]
5254 Set duration of force
5255 in microseconds.
5256
5257 rcutorture.fqs_holdoff= [KNL]
5258 Set holdoff time with
5259 in microseconds.
5260
5261 rcutorture.fqs_stutter= [KNL]
5262 Set wait time between
5263 in seconds.
5264
5265 rcutorture.fwd_progress= [KNL]
5266 Specifies the number
5267 for RCU grace-period
5268 for the types of RCU
5269 Defaults to 1 kthread
5270 greater than the numb
5271 of CPUs to be used.
5272
5273 rcutorture.fwd_progress_div= [KNL]
5274 Specify the fraction
5275 period to do tight-lo
5276
5277 rcutorture.fwd_progress_holdoff= [KNL
5278 Number of seconds to
5279 forward-progress test
5280
5281 rcutorture.fwd_progress_need_resched=
5282 Enclose cond_resched(
5283 need_resched() during
5284 testing.
5285
5286 rcutorture.gp_cond= [KNL]
5287 Use conditional/async
5288 primitives, if availa
5289
5290 rcutorture.gp_exp= [KNL]
5291 Use expedited update-
5292
5293 rcutorture.gp_normal= [KNL]
5294 Use normal (non-exped
5295 update-side primitive
5296
5297 rcutorture.gp_sync= [KNL]
5298 Use normal (non-exped
5299 update-side primitive
5300 of rcutorture.gp_cond
5301 rcutorture.gp_normal=
5302 are zero, rcutorture
5303 they are all non-zero
5304
5305 rcutorture.irqreader= [KNL]
5306 Run RCU readers from
5307 accurately, from a ti
5308 flavors take kindly t
5309
5310 rcutorture.leakpointer= [KNL]
5311 Leak an RCU-protected
5312 This can of course re
5313 intended to test the
5314 CONFIG_RCU_STRICT_GRA
5315 such leaks.
5316
5317 rcutorture.n_barrier_cbs= [KNL]
5318 Set callbacks/threads
5319
5320 rcutorture.nfakewriters= [KNL]
5321 Set number of concurr
5322 stress RCU, they don'
5323 test, hence the "fake
5324
5325 rcutorture.nocbs_nthreads= [KNL]
5326 Set number of RCU cal
5327 Zero (the default) di
5328
5329 rcutorture.nocbs_toggle= [KNL]
5330 Set the delay in mill
5331 callback-offload togg
5332
5333 rcutorture.nreaders= [KNL]
5334 Set number of RCU rea
5335 N-1, where N is the n
5336 "n" less than -1 sele
5337 the number of CPUs.
5338 (the number of CPUs),
5339
5340 rcutorture.object_debug= [KNL]
5341 Enable debug-object d
5342
5343 rcutorture.onoff_holdoff= [KNL]
5344 Set time (s) after bo
5345
5346 rcutorture.onoff_interval= [KNL]
5347 Set time (jiffies) be
5348 or zero to disable CP
5349
5350 rcutorture.read_exit= [KNL]
5351 Set the number of rea
5352 to test the interacti
5353 task-exit processing.
5354
5355 rcutorture.read_exit_burst= [KNL]
5356 The number of times i
5357 episode that a set of
5358 is spawned.
5359
5360 rcutorture.read_exit_delay= [KNL]
5361 The delay, in seconds
5362 read-then-exit testin
5363
5364 rcutorture.shuffle_interval= [KNL]
5365 Set task-shuffle inte
5366 allows some CPUs to g
5367 during the rcutorture
5368
5369 rcutorture.shutdown_secs= [KNL]
5370 Set time (s) after bo
5371 is useful for hands-o
5372
5373 rcutorture.stall_cpu= [KNL]
5374 Duration of CPU stall
5375 warnings, zero to dis
5376
5377 rcutorture.stall_cpu_block= [KNL]
5378 Sleep while stalling
5379 in warnings from pree
5380 any other stall-relat
5381 in kernels built with
5382 CONFIG_PREEMPT_COUNT=
5383 cause the CPU to pass
5384 Given CONFIG_PREEMPTI
5385 RCU CPU stall warning
5386 in scheduling-while-a
5387
5388 Use of this module pa
5389
5390
5391 rcutorture.stall_cpu_holdoff= [KNL]
5392 Time to wait (s) afte
5393
5394 rcutorture.stall_cpu_irqsoff= [KNL]
5395 Disable interrupts wh
5396
5397 rcutorture.stall_gp_kthread= [KNL]
5398 Duration (s) of force
5399 grace-period kthread
5400 warnings, zero to dis
5401 and stall_gp_kthread
5402 kthread is starved fi
5403
5404 rcutorture.stat_interval= [KNL]
5405 Time (s) between stat
5406
5407 rcutorture.stutter= [KNL]
5408 Time (s) to stutter t
5409 five seconds causes t
5410 wait for five seconds
5411 ability to transition
5412
5413 rcutorture.test_boost= [KNL]
5414 Test RCU priority boo
5415 "Maybe" means test if
5416 under test support RC
5417
5418 rcutorture.test_boost_duration= [KNL]
5419 Duration (s) of each
5420
5421 rcutorture.test_boost_interval= [KNL]
5422 Interval (s) between
5423
5424 rcutorture.test_no_idle_hz= [KNL]
5425 Test RCU's dyntick-id
5426 rcutorture.shuffle_in
5427
5428 rcutorture.torture_type= [KNL]
5429 Specify the RCU imple
5430
5431 rcutorture.verbose= [KNL]
5432 Enable additional pri
5433
5434 rcupdate.rcu_cpu_stall_ftrace_dump= [
5435 Dump ftrace buffer af
5436 stall warning.
5437
5438 rcupdate.rcu_cpu_stall_notifiers= [KN
5439 Provide RCU CPU stall
5440 warnings in the RCU_C
5441 option's help text.
5442 do not want rcupdate.
5443
5444 rcupdate.rcu_cpu_stall_suppress= [KNL
5445 Suppress RCU CPU stal
5446
5447 rcupdate.rcu_cpu_stall_suppress_at_bo
5448 Suppress RCU CPU stal
5449 rcutorture writer sta
5450 during early boot, th
5451 before the init task
5452
5453 rcupdate.rcu_cpu_stall_timeout= [KNL]
5454 Set timeout for RCU C
5455 The value is in secon
5456 value is 300 seconds.
5457
5458 rcupdate.rcu_exp_cpu_stall_timeout= [
5459 Set timeout for exped
5460 messages. The value
5461 and the maximum allow
5462 milliseconds. Please
5463 adjusted to an arch t
5464 Setting this to zero
5465 rcupdate.rcu_cpu_stal
5466 conversion from secon
5467
5468 rcupdate.rcu_cpu_stall_cputime= [KNL]
5469 Provide statistics on
5470 interrupts and tasks
5471 multiple continuous R
5472 begin at half of the
5473
5474 rcupdate.rcu_exp_stall_task_details=
5475 Print stack dumps of
5476 current expedited RCU
5477 expedited RCU CPU sta
5478
5479 rcupdate.rcu_expedited= [KNL]
5480 Use expedited grace-p
5481 example, synchronize_
5482 of synchronize_rcu().
5483 but can increase CPU
5484 real-time latency, an
5485 No effect on CONFIG_T
5486
5487 rcupdate.rcu_normal= [KNL]
5488 Use only normal grace
5489 for example, synchron
5490 synchronize_rcu_exped
5491 real-time latency, CP
5492 energy efficiency, bu
5493 increased grace-perio
5494 overrides rcupdate.rc
5495 CONFIG_TINY_RCU kerne
5496
5497 rcupdate.rcu_normal_after_boot= [KNL]
5498 Once boot has complet
5499 rcu_end_inkernel_boot
5500 only normal grace-per
5501 on CONFIG_TINY_RCU ke
5502
5503 But note that CONFIG_
5504 this kernel boot para
5505 it to the value one,
5506 post-boot attempt at
5507 period to instead use
5508 grace-period processi
5509
5510 rcupdate.rcu_task_collapse_lim= [KNL]
5511 Set the maximum numbe
5512 at the beginning of a
5513 the RCU Tasks flavors
5514 a single callback que
5515 occurs when rcupdate.
5516 set to the default va
5517
5518 rcupdate.rcu_task_contend_lim= [KNL]
5519 Set the minimum numbe
5520 lock-contention event
5521 cause the RCU Tasks f
5522 callback queuing. Th
5523 when rcupdate.rcu_tas
5524 the default value of
5525
5526 rcupdate.rcu_task_enqueue_lim= [KNL]
5527 Set the number of cal
5528 RCU Tasks family of R
5529 of -1 allows this to
5530 dynamically) adjusted
5531 for use in testing.
5532
5533 rcupdate.rcu_task_ipi_delay= [KNL]
5534 Set time in jiffies d
5535 avoid sending IPIs, s
5536 of a given grace peri
5537 number avoids disturb
5538 but lengthens grace p
5539
5540 rcupdate.rcu_task_lazy_lim= [KNL]
5541 Number of callbacks o
5542 cancel laziness on th
5543 cancellation of lazin
5544 doing so increases th
5545 callback flooding.
5546
5547 rcupdate.rcu_task_stall_info= [KNL]
5548 Set initial timeout i
5549 informational message
5550 of the problem for th
5551 wait for ten minutes.
5552 only printed prior to
5553 for a given grace per
5554 less than or equal to
5555 seconds. A change in
5556 until the beginning o
5557
5558 rcupdate.rcu_task_stall_info_mult= [K
5559 Multiplier for time i
5560 RCU task stall inform
5561 RCU tasks grace perio
5562 to one through ten, i
5563 the value three, so t
5564 message is printed 10
5565 period, the second at
5566 160 seconds, and then
5567 seconds would prevent
5568
5569 rcupdate.rcu_task_stall_timeout= [KNL
5570 Set timeout in jiffie
5571 warning messages. Di
5572 than or equal to zero
5573 A change in value doe
5574 the beginning of the
5575
5576 rcupdate.rcu_tasks_lazy_ms= [KNL]
5577 Set timeout in millis
5578 callback batching for
5579 A negative value will
5580 of zero will disable
5581 always disabled for s
5582
5583 rcupdate.rcu_tasks_rude_lazy_ms= [KNL
5584 Set timeout in millis
5585 Rude asynchronous cal
5586 call_rcu_tasks_rude()
5587 will take the default
5588 disable batching. Ba
5589 for synchronize_rcu_t
5590
5591 rcupdate.rcu_tasks_trace_lazy_ms= [KN
5592 Set timeout in millis
5593 Trace asynchronous ca
5594 call_rcu_tasks_trace(
5595 will take the default
5596 disable batching. Ba
5597 for synchronize_rcu_t
5598
5599 rcupdate.rcu_self_test= [KNL]
5600 Run the RCU early boo
5601
5602 rdinit= [KNL]
5603 Format: <full_path>
5604 Run specified binary
5605 used for early usersp
5606
5607 rdrand= [X86,EARLY]
5608 force - Override the
5609 advertisement
5610 certain AMD p
5611 support, spec
5612 path).
5613
5614 rdt= [HW,X86,RDT]
5615 Turn on/off individua
5616 cmt, mbmtotal, mbmloc
5617 mba, smba, bmec.
5618 E.g. to turn on cmt a
5619 rdt=cmt,!mba
5620
5621 reboot= [KNL]
5622 Format (x86 or x86_64
5623 [w[arm] | c[o
5624 [[,]s[mp]####
5625 [[,]b[ios] |
5626 [[,]f[orce]
5627 Where reboot_mode is
5628 (pref
5629 reboo
5630 reboot_type is
5631 reboot_force is
5632 reboot_cpu is s
5633 to be
5634
5635 refscale.holdoff= [KNL]
5636 Set test-start holdof
5637 this parameter is to
5638 test until boot compl
5639 interference.
5640
5641 refscale.lookup_instances= [KNL]
5642 Number of data elemen
5643 SLAB_TYPESAFE_BY_RCU
5644 is negated and multip
5645 zero specifies nr_cpu
5646
5647 refscale.loops= [KNL]
5648 Set the number of loo
5649 primitive under test.
5650 reduces noise due to
5651 but the default has a
5652 noise to a handful of
5653 x86 laptops.
5654
5655 refscale.nreaders= [KNL]
5656 Set number of readers
5657 selects N, where N is
5658 of CPUs. A value of
5659
5660 refscale.nruns= [KNL]
5661 Set number of runs, e
5662 the console log.
5663
5664 refscale.readdelay= [KNL]
5665 Set the read-side cri
5666 measured in microseco
5667
5668 refscale.scale_type= [KNL]
5669 Specify the read-prot
5670
5671 refscale.shutdown= [KNL]
5672 Shut down the system
5673 test. This defaults
5674 refscale is built int
5675 it running) when refs
5676
5677 refscale.verbose= [KNL]
5678 Enable additional pri
5679
5680 refscale.verbose_batched= [KNL]
5681 Batch the additional
5682 (the default) or nega
5683 print every Nth verbo
5684 specified.
5685
5686 regulator_ignore_unused
5687 [REGULATOR]
5688 Prevents regulator fr
5689 that are unused, due
5690 be useful for debug a
5691 needed on a platform
5692
5693 relax_domain_level=
5694 [KNL, SMP] Set schedu
5695 See Documentation/adm
5696
5697 reserve= [KNL,BUGS] Force kern
5698 Format: <base1>,<size
5699 Reserve I/O ports or
5700 them. If <base> is l
5701 is assumed to be I/O
5702
5703 reserve_mem= [RAM]
5704 Format: nn[KNG]:<alig
5705 Reserve physical memo
5706 other subsystems can
5707 used for systems that
5708 line will try to rese
5709 soft reboots. Note, i
5710 location. For example
5711 or if booting a diffe
5712 places the kernel at
5713 was from a previous b
5714 different location.
5715 Any subsystem using t
5716 that the contents of
5717 boot, as there may be
5718 located at the same l
5719
5720 The format is size:al
5721 12 megabytes of 4096
5722
5723 reserve_mem=12M:4096:
5724
5725 reservetop= [X86-32,EARLY]
5726 Format: nn[KMG]
5727 Reserves a hole at th
5728 address space.
5729
5730 reset_devices [KNL] Force drivers t
5731 during initialization
5732
5733 resume= [SWSUSP]
5734 Specify the partition
5735 Format:
5736 {/dev/<dev> | PARTUUI
5737
5738 resume_offset= [SWSUSP]
5739 Specify the offset fr
5740 given by "resume=" at
5741 in <PAGE_SIZE> units
5742 See Documentation/po
5743
5744 resumedelay= [HIBERNATION] Delay (
5745 read the resume files
5746
5747 resumewait [HIBERNATION] Wait (i
5748 Useful for devices th
5749 (e.g. USB and MMC dev
5750
5751 retain_initrd [RAM] Keep initrd mem
5752 be accessible via /sy
5753
5754 retbleed= [X86] Control mitigat
5755 Speculative Code Exec
5756 vulnerability.
5757
5758 AMD-based UNRET and I
5759 sibling threads from
5760 sibling threads. For
5761 cessors that support
5762 that don't.
5763
5764 off - no mit
5765 auto - automa
5766 auto,nosmt - automa
5767 disabl
5768 the fu
5769 and ol
5770 ibpb - On AMD
5771 window
5772 Safe,
5773 enable
5774 on Int
5775 ibpb,nosmt - Like "
5776 when S
5777 the al
5778 have S
5779 unret - Force
5780 only e
5781 system
5782 unret,nosmt - Like u
5783 is not
5784 system
5785
5786 Selecting 'auto' will
5787 time according to the
5788
5789 Not specifying this o
5790
5791 rfkill.default_state=
5792 0 "airplane mode". All
5793 etc. communication is
5794 1 Unblocked.
5795
5796 rfkill.master_switch_mode=
5797 0 The "airplane mode" b
5798 1 The "airplane mode" b
5799 blocked and the previ
5800 2 The "airplane mode" b
5801 blocked and everythin
5802
5803 ring3mwait=disable
5804 [KNL] Disable ring 3
5805 CPUs.
5806
5807 riscv_isa_fallback [RISCV,EARLY]
5808 When CONFIG_RISCV_ISA
5809 falling back to detec
5810 "riscv,isa" property
5811 replacement propertie
5812 entry for RISCV_ISA_F
5813
5814 ro [KNL] Mount root devi
5815
5816 rodata= [KNL,EARLY]
5817 on Mark read-only kernel
5818 off Leave read-only kerne
5819 full Mark read-only kernel
5820 [arm64]
5821
5822 rockchip.usb_uart
5823 [EARLY]
5824 Enable the uart passt
5825 on Rockchip SoCs. Whe
5826 debug-uart get routed
5827 port and the regular
5828
5829 root= [KNL] Root filesystem
5830 Usually this a a bloc
5831 see the early_lookup_
5832 block/early-lookup.c
5833 Alternatively this ca
5834 ramdisk, "nfs" and "c
5835 system, or "mtd" and
5836
5837 rootdelay= [KNL] Delay (in secon
5838 mount the root filesy
5839
5840 rootflags= [KNL] Set root filesy
5841
5842 rootfstype= [KNL] Set root filesy
5843
5844 rootwait [KNL] Wait (indefinit
5845 Useful for devices th
5846 (e.g. USB and MMC dev
5847
5848 rootwait= [KNL] Maximum time (i
5849 to show up before att
5850 filesystem.
5851
5852 rproc_mem=nn[KMG][@address]
5853 [KNL,ARM,CMA] Remotep
5854 Memory area to be use
5855 managed by CMA.
5856
5857 rw [KNL] Mount root devi
5858
5859 S [KNL] Run init in sin
5860
5861 s390_iommu= [HW,S390]
5862 Set s390 IOTLB flushi
5863 strict
5864 With strict flushing
5865 in an IOTLB flush. De
5866 reuse, which is faste
5867 iommu.strict=1.
5868
5869 s390_iommu_aperture= [KNL,S390]
5870 Specifies the size of
5871 accessible through th
5872 factor of the size of
5873 The default is 1 mean
5874 as many DMA addresses
5875 if supported by hardw
5876 once. With a value of
5877 and so on. As a speci
5878 restrictions other th
5879 cost of significant a
5880
5881 sa1100ir [NET]
5882 See drivers/net/irda/
5883
5884 sched_verbose [KNL,EARLY] Enables v
5885
5886 schedstats= [KNL,X86] Enable or d
5887 Allowed values are en
5888 incurs a small amount
5889 but is useful for deb
5890
5891 sched_thermal_decay_shift=
5892 [Deprecated]
5893 [KNL, SMP] Set a deca
5894 pressure signal. Ther
5895 default decay period
5896 signals(usually 32 ms
5897 sched_thermal_decay_s
5898 period for the therma
5899 value.
5900 i.e. with the default
5901 sched_thermal_decay_s
5902 1
5903 2
5904 and so on.
5905 Format: integer betwe
5906 Default is 0.
5907
5908 scftorture.holdoff= [KNL]
5909 Number of seconds to
5910 test. Defaults to ze
5911 to 10 seconds for bui
5912 tests.
5913
5914 scftorture.longwait= [KNL]
5915 Request ridiculously
5916 up to the chosen limi
5917 default) disables thi
5918 that requesting even
5919 seconds can result in
5920 softlockup complaints
5921
5922 scftorture.nthreads= [KNL]
5923 Number of kthreads to
5924 smp_call_function() f
5925 The default of -1 spe
5926 equal to the number o
5927
5928 scftorture.onoff_holdoff= [KNL]
5929 Number seconds to wai
5930 test before initiatin
5931
5932 scftorture.onoff_interval= [KNL]
5933 Number seconds to wai
5934 CPU-hotplug operation
5935 is the default) disab
5936
5937 scftorture.shutdown_secs= [KNL]
5938 The number of seconds
5939 test after which to s
5940 default of zero avoid
5941 Non-zero values are u
5942
5943 scftorture.stat_interval= [KNL]
5944 The number of seconds
5945 current test statisti
5946 of zero disables stat
5947
5948 scftorture.stutter_cpus= [KNL]
5949 The number of jiffies
5950 to the set of CPUs un
5951
5952 scftorture.use_cpus_read_lock= [KNL]
5953 Use use_cpus_read_loc
5954 preempt_disable() to
5955 while invoking one of
5956 functions.
5957
5958 scftorture.verbose= [KNL]
5959 Enable additional pri
5960
5961 scftorture.weight_single= [KNL]
5962 The probability weigh
5963 smp_call_function_sin
5964 "wait" parameter. A
5965 default if all other
5966 if at least one weigh
5967 value of -1 will inst
5968
5969 scftorture.weight_single_wait= [KNL]
5970 The probability weigh
5971 smp_call_function_sin
5972 non-zero "wait" param
5973
5974 scftorture.weight_many= [KNL]
5975 The probability weigh
5976 smp_call_function_man
5977 "wait" parameter. Se
5978 Note well that settin
5979 this weighting can pl
5980 on the system.
5981
5982 scftorture.weight_many_wait= [KNL]
5983 The probability weigh
5984 smp_call_function_man
5985 non-zero "wait" param
5986 and weight_many.
5987
5988 scftorture.weight_all= [KNL]
5989 The probability weigh
5990 smp_call_function_all
5991 "wait" parameter. Se
5992 weight_many.
5993
5994 scftorture.weight_all_wait= [KNL]
5995 The probability weigh
5996 smp_call_function_all
5997 non-zero "wait" param
5998 and weight_many.
5999
6000 skew_tick= [KNL,EARLY] Offset th
6001 xtime_lock contention
6002 contention on all sys
6003 Format: { "0" | "1" }
6004 0 -- disable. (may be
6005 1 -- enable.
6006 Note: increases power
6007 enabled if running ji
6008
6009 security= [SECURITY] Choose a l
6010 enable at boot. This
6011 "lsm=" parameter.
6012
6013 selinux= [SELINUX] Disable or
6014 Format: { "0" | "1" }
6015 See security/selinux/
6016 0 -- disable.
6017 1 -- enable.
6018 Default value is 1.
6019
6020 serialnumber [BUGS=X86-32]
6021
6022 sev=option[,option...] [X86-64] See D
6023
6024 shapers= [NET]
6025 Maximal number of sha
6026
6027 show_lapic= [APIC,X86] Advanced P
6028 Limit apic dumping. T
6029 number of local apics
6030 to set it to "all" by
6031 Format: { 1 (default)
6032 The parameter valid i
6033 apic=verbose is speci
6034 Example: apic=debug s
6035
6036 slab_debug[=options[,slabs][;[options
6037 Enabling slab_debug a
6038 culprit if slab objec
6039 slab_debug can create
6040 may poison objects wh
6041 last alloc / free. Fo
6042 Documentation/mm/slub
6043 (slub_debug legacy na
6044
6045 slab_max_order= [MM]
6046 Determines the maximu
6047 A high setting may ca
6048 fragmentation. For mo
6049 Documentation/mm/slub
6050 (slub_max_order legac
6051
6052 slab_merge [MM]
6053 Enable merging of sla
6054 kernel is built witho
6055 (slub_merge legacy na
6056
6057 slab_min_objects= [MM]
6058 The minimum number of
6059 increase the slab ord
6060 generate a sufficient
6061 the number of objects
6062 of objects the smalle
6063 and the less frequent
6064 For more information
6065 (slub_min_objects leg
6066
6067 slab_min_order= [MM]
6068 Determines the minimu
6069 lower or equal to sla
6070 Documentation/mm/slub
6071 (slub_min_order legac
6072
6073 slab_nomerge [MM]
6074 Disable merging of sl
6075 necessary if there is
6076 allocs to different s
6077 environments where th
6078 layout control by att
6079 frustrated by disabli
6080 most of the exposure
6081 cache (risks via meta
6082 unchanged). Debug opt
6083 own.
6084 For more information
6085 (slub_nomerge legacy
6086
6087 slram= [HW,MTD]
6088
6089 smart2= [HW]
6090 Format: <io1>[,<io2>[
6091
6092 smp.csd_lock_timeout= [KNL]
6093 Specify the period of
6094 that smp_call_functio
6095 for a CPU to release
6096 useful when diagnosin
6097 disabling interrupts
6098 of time. Defaults to
6099 setting a value of ze
6100 This feature may be m
6101 using the csdlock_deb
6102
6103 smp.panic_on_ipistall= [KNL]
6104 If a csd_lock_timeout
6105 the specified number
6106 system. By default,
6107 take as long as they
6108 for this value provid
6109
6110 smsc-ircc2.nopnp [HW] Don't us
6111 smsc-ircc2.ircc_cfg= [HW] Device c
6112 smsc-ircc2.ircc_sir= [HW] SIR base
6113 smsc-ircc2.ircc_fir= [HW] FIR base
6114 smsc-ircc2.ircc_irq= [HW] IRQ line
6115 smsc-ircc2.ircc_dma= [HW] DMA chan
6116 smsc-ircc2.ircc_transceiver= [HW] Tra
6117 0: Toshiba Sa
6118 1: Fast pin s
6119 2: ATC IRMode
6120
6121 smt= [KNL,MIPS,S390,EARLY]
6122 (logical CPUs) to use
6123 capable of symmetric
6124 be capped to the actu
6125 Format: <integer>
6126 Default: -1 (no limit
6127
6128 softlockup_panic=
6129 [KNL] Should the soft
6130 Format: 0 | 1
6131
6132 A value of 1 instruct
6133 to panic the machine
6134 also controlled by th
6135 and CONFIG_BOOTPARAM_
6136 respective build-time
6137
6138 softlockup_all_cpu_backtrace=
6139 [KNL] Should the soft
6140 backtraces on all cpu
6141 Format: 0 | 1
6142
6143 sonypi.*= [HW] Sony Programmabl
6144 See Documentation/adm
6145
6146 spectre_bhi= [X86] Control mitigat
6147 (BHI) vulnerability.
6148 deployment of the HW
6149 clearing sequence.
6150
6151 on - (default) En
6152 needed. Thi
6153 both syscall
6154 vmexit - On systems w
6155 available, e
6156 ONLY. On su
6157 protected fr
6158 may still be
6159 off - Disable the
6160
6161 spectre_v2= [X86,EARLY] Control m
6162 (indirect branch spec
6163 The default operation
6164 user space attacks.
6165
6166 on - unconditionall
6167 spectre_v2_use
6168 off - unconditionall
6169 spectre_v2_use
6170 auto - kernel detects
6171 vulnerable
6172
6173 Selecting 'on' will,
6174 mitigation method at
6175 CPU, the available mi
6176 CONFIG_MITIGATION_RET
6177 and the compiler with
6178
6179 Selecting 'on' will a
6180 against user space to
6181
6182 Selecting 'off' will
6183 the user space protec
6184
6185 Specific mitigations
6186
6187 retpoline - r
6188 retpoline,generic - R
6189 retpoline,lfence - L
6190 retpoline,amd - a
6191 eibrs - E
6192 eibrs,retpoline - E
6193 eibrs,lfence - E
6194 ibrs - u
6195
6196 Not specifying this o
6197 spectre_v2=auto.
6198
6199 spectre_v2_user=
6200 [X86] Control mitigat
6201 (indirect branch spec
6202 user space tasks
6203
6204 on - Uncondition
6205 enforced by
6206
6207 off - Uncondition
6208 enforced by
6209
6210 prctl - Indirect br
6211 but mitigat
6212 per thread.
6213 is inherite
6214
6215 prctl,ibpb
6216 - Like "prctl
6217 controlled
6218 always when
6219 space proce
6220
6221 seccomp
6222 - Same as "pr
6223 threads wil
6224 they explic
6225
6226 seccomp,ibpb
6227 - Like "secco
6228 controlled
6229 always when
6230 user space
6231
6232 auto - Kernel sele
6233 the availab
6234
6235 Default mitigation: "
6236
6237 Not specifying this o
6238 spectre_v2_user=auto.
6239
6240 spec_rstack_overflow=
6241 [X86,EARLY] Control R
6242
6243 off - Dis
6244 microcode - Ena
6245 safe-ret - Ena
6246 ibpb - Ena
6247 ker
6248 ibpb-vmexit - Iss
6249 (cl
6250
6251 spec_store_bypass_disable=
6252 [HW,EARLY] Control Sp
6253 (Speculative Store By
6254
6255 Certain CPUs are vuln
6256 a common industry wid
6257 as "Speculative Store
6258 to the same memory lo
6259 later loads during sp
6260 is that such stores a
6261 be detected prior to
6262 end of a particular s
6263
6264 In vulnerable process
6265 store can be used in
6266 example to read memor
6267 directly have access
6268
6269 This parameter contro
6270 Bypass optimization i
6271
6272 On x86 the options ar
6273
6274 on - Uncondition
6275 off - Uncondition
6276 auto - Kernel dete
6277 implementat
6278 picks the m
6279 CPU is not
6280 CPU is vuln
6281 architectur
6282 prctl - Control Spe
6283 via prctl.
6284 for a proce
6285 is inherite
6286 seccomp - Same as "pr
6287 will disabl
6288
6289 Default mitigations:
6290 X86: "prctl"
6291
6292 On powerpc the option
6293
6294 on,auto - On Power8 a
6295 barrier on
6296 perform a s
6297 exit.
6298 off - No action.
6299
6300 Not specifying this o
6301 spec_store_bypass_dis
6302
6303 split_lock_detect=
6304 [X86] Enable split lo
6305
6306 When enabled (and if
6307 instructions that acc
6308 boundaries will resul
6309 for split lock detect
6310 bus lock detection.
6311
6312 off - not enabled
6313
6314 warn - the kernel
6315 about appli
6316 exception o
6317 the default
6318 detection o
6319 behavior is
6320 enabled in
6321
6322 fatal - the kernel
6323 that trigge
6324 exception.
6325 both featur
6326
6327 ratelimit:N -
6328 Set system
6329 per second
6330 0 < N <= 10
6331
6332 N/A for spl
6333
6334
6335 If an #AC exception i
6336 firmware (i.e. not wh
6337 the kernel will oops
6338 mode.
6339
6340 #DB exception for bus
6341 CPL > 0.
6342
6343 srbds= [X86,INTEL,EARLY]
6344 Control the Special R
6345 (SRBDS) mitigation.
6346
6347 Certain CPUs are vuln
6348 exploit which can lea
6349 number generator.
6350
6351 By default, this issu
6352 microcode. However,
6353 the RDRAND and RDSEED
6354 much slower. Among o
6355 result in reduced thr
6356
6357 The microcode mitigat
6358 the following option:
6359
6360 off: Disable mitig
6361 performance i
6362
6363 srcutree.big_cpu_lim [KNL]
6364 Specifies the number
6365 large system, such th
6366 should immediately al
6367 This kernel-boot para
6368 but takes effect only
6369 bits of srcutree.conv
6370 (decide at boot).
6371
6372 srcutree.convert_to_big [KNL]
6373 Specifies under what
6374 srcu_struct structure
6375 form, that is, with a
6376
6377 0: Never.
6378 1: At ini
6379 2: When r
6380 3: Decide
6381 0x1X: Above
6382
6383 Either way, the srcu_
6384 on the actual runtime
6385 instead of the compil
6386
6387 srcutree.counter_wrap_check [KNL]
6388 Specifies how frequen
6389 grace-period sequence
6390 srcu_data structure's
6391 The greater the numbe
6392 parameter, the less f
6393 be checked for. Note
6394 are ignored.
6395
6396 srcutree.exp_holdoff [KNL]
6397 Specifies how many na
6398 since the end of the
6399 a given srcu_struct u
6400 grace period will be
6401 expediting. Set to z
6402 expediting.
6403
6404 srcutree.srcu_max_nodelay [KNL]
6405 Specifies the number
6406 per jiffy for which t
6407 worker thread will be
6408 delay. Beyond this li
6409 be rescheduled with a
6410
6411 srcutree.srcu_max_nodelay_phase [KNL]
6412 Specifies the per-gra
6413 non-sleeping polls of
6414 grace period worker t
6415 with a sleep delay of
6416 rescan of the readers
6417
6418 srcutree.srcu_retry_check_delay [KNL]
6419 Specifies number of m
6420 delay between each no
6421
6422 srcutree.small_contention_lim [KNL]
6423 Specifies the number
6424 events per jiffy will
6425 initiating a conversi
6426 structure to big form
6427 srcutree.convert_to_b
6428 set for contention-ba
6429
6430 ssbd= [ARM64,HW,EARLY]
6431 Speculative Store Byp
6432
6433 On CPUs that are vuln
6434 Store Bypass vulnerab
6435 firmware based mitiga
6436 indicates how the mit
6437
6438 force-on: Unconditio
6439 for both k
6440 force-off: Unconditio
6441 for both k
6442 kernel: Always ena
6443 kernel, an
6444 to allow u
6445 interest i
6446
6447 stack_guard_gap= [MM]
6448 override the default
6449 is in page units and
6450 to (for stacks growin
6451 growing up) the main
6452 mapping. Default valu
6453
6454 stack_depot_disable= [KNL,EARLY]
6455 Setting this to true
6456 disable the stack dep
6457 consumed by the stack
6458 to false.
6459
6460 stacktrace [FTRACE]
6461 Enabled the stack tra
6462
6463 stacktrace_filter=[function-list]
6464 [FTRACE] Limit the fu
6465 will trace at boot up
6466 list of functions. Th
6467 time by the stack_tra
6468 tracing directory. No
6469 and the stacktrace ab
6470
6471 sti= [PARISC,HW]
6472 Format: <num>
6473 Set the STI (builtin
6474 machines) console (gr
6475 as the initial boot-c
6476 See also comment in d
6477
6478 sti_font= [HW]
6479 See comment in driver
6480
6481 stifb= [HW]
6482 Format: bpp:<bpp1>[:<
6483
6484 strict_sas_size=
6485 [X86]
6486 Format: <bool>
6487 Enable or disable str
6488 against the required
6489 depends on the suppor
6490 be used to filter out
6491 not yet been made awa
6492
6493 stress_hpt [PPC,EARLY]
6494 Limits the number of
6495 page table to increas
6496 faults on kernel addr
6497
6498 stress_slb [PPC,EARLY]
6499 Limits the number of
6500 them frequently to in
6501 on kernel addresses.
6502
6503 sunrpc.min_resvport=
6504 sunrpc.max_resvport=
6505 [NFS,SUNRPC]
6506 SunRPC servers often
6507 originate from a priv
6508 range 0 < portnr < 10
6509 An administrator who
6510 ports for other uses
6511 kernel's sunrpc clien
6512 using these two param
6513 maximum port values.
6514
6515 sunrpc.svc_rpc_per_connection_limit=
6516 [NFS,SUNRPC]
6517 Limit the number of r
6518 process in parallel f
6519 The default value is
6520
6521 sunrpc.pool_mode=
6522 [NFS]
6523 Control how the NFS s
6524 service thread pools.
6525 you have and where th
6526 option will affect wh
6527 Note: this parameter
6528 NFS server is running
6529
6530 auto the serve
6531 automatic
6532 global a single
6533 percpu one pool
6534 pernode one pool
6535 to global
6536
6537 sunrpc.tcp_slot_table_entries=
6538 sunrpc.udp_slot_table_entries=
6539 [NFS,SUNRPC]
6540 Sets the upper limit
6541 RPC calls that can be
6542 server. Increasing th
6543 improve throughput, b
6544 amount of memory rese
6545
6546 suspend.pm_test_delay=
6547 [SUSPEND]
6548 Sets the number of se
6549 mode before resuming
6550 /sys/power/pm_test).
6551 is set. Default value
6552
6553 svm= [PPC]
6554 Format: { on | off |
6555 This parameter contro
6556 Execution Facility on
6557
6558 swiotlb= [ARM,PPC,MIPS,X86,S39
6559 Format: { <int> [,<in
6560 <int> -- Number of I/
6561 <int> -- Second integ
6562 areas with t
6563 to a power o
6564 force -- force using
6565 wouldn't be
6566 noforce -- Never use
6567
6568 switches= [HW,M68k,EARLY]
6569
6570 sysctl.*= [KNL]
6571 Set a sysctl paramete
6572 process, as if the va
6573 /proc/sys/... file. B
6574 separators. Unrecogni
6575 are reported in the k
6576 later by a loaded mod
6577 Example: sysctl.vm.sw
6578
6579 sysrq_always_enabled
6580 [KNL]
6581 Ignore sysrq setting
6582 neutralize any effect
6583 Useful for debugging.
6584
6585 tcpmhash_entries= [KNL,NET]
6586 Set the number of tcp
6587 Default value is 8192
6588 ram pages. This is us
6589 cache size. See Docum
6590 "tcp_no_metrics_save"
6591
6592 tdfx= [HW,DRM]
6593
6594 test_suspend= [SUSPEND]
6595 Format: { "mem" | "st
6596 Specify "mem" (for Su
6597 standby suspend) or "
6598 as the system sleep s
6599 the optional capabili
6600 The system is woken f
6601 wakeup-capable RTC al
6602
6603 thash_entries= [KNL,NET]
6604 Set number of hash bu
6605
6606 thermal.act= [HW,ACPI]
6607 -1: disable all activ
6608 <degrees C>: override
6609
6610 thermal.crt= [HW,ACPI]
6611 -1: disable all criti
6612 <degrees C>: override
6613
6614 thermal.off= [HW,ACPI]
6615 1: disable ACPI therm
6616
6617 thermal.psv= [HW,ACPI]
6618 -1: disable all passi
6619 <degrees C>: override
6620 value
6621
6622 thermal.tzp= [HW,ACPI]
6623 Specify global defaul
6624 <deci-seconds>: poll
6625 0: no polling (defaul
6626
6627 threadirqs [KNL,EARLY]
6628 Force threading of al
6629 marked explicitly IRQ
6630
6631 topology= [S390,EARLY]
6632 Format: {off | on}
6633 Specify if the kernel
6634 topology information
6635 The scheduler will ma
6636 e.g. base its process
6637 Default is on.
6638
6639 torture.disable_onoff_at_boot= [KNL]
6640 Prevent the CPU-hotpl
6641 until after init has
6642
6643 torture.ftrace_dump_at_shutdown= [KNL
6644 Dump the ftrace buffe
6645 even if there were no
6646 very costly operation
6647 are running concurren
6648 with rotating-rust st
6649
6650 torture.verbose_sleep_frequency= [KNL
6651 Specifies how many ve
6652 emitted between each
6653 disables verbose-prin
6654
6655 torture.verbose_sleep_duration= [KNL]
6656 Duration of each verb
6657
6658 tpm_suspend_pcr=[HW,TPM]
6659 Format: integer pcr i
6660 Specify that at suspe
6661 should extend the spe
6662 as a workaround for s
6663 flush the last writte
6664 This will guarantee t
6665 are saved.
6666
6667 tpm_tis.interrupts= [HW,TPM]
6668 Enable interrupts for
6669 for the FIFO interfac
6670 (0). For more informa
6671 defined by Trusted Co
6672 https://trustedcomput
6673
6674 tp_printk [FTRACE]
6675 Have the tracepoints
6676 tracing ring buffer.
6677 where the system hang
6678 option for reading th
6679 ftrace_dump_on_oops.
6680
6681 To turn off having tr
6682 echo 0 > /proc/sys/k
6683 Note, echoing 1 into
6684 tp_printk kernel cmdl
6685
6686 The tp_printk_stop_on
6687 to stop the printing
6688 late_initcall_sync.
6689
6690 ** CAUTION **
6691
6692 Having tracepoints se
6693 frequency tracepoints
6694 the system to live lo
6695
6696 tp_printk_stop_on_boot [FTRACE]
6697 When tp_printk (above
6698 on the console. It ma
6699 printing of events du
6700 make the system inope
6701
6702 This command line opt
6703 to console at the lat
6704
6705 trace_buf_size=nn[KMG]
6706 [FTRACE] will set tra
6707
6708 trace_clock= [FTRACE] Set the cloc
6709 at boot up.
6710 local - Use the per C
6711 (converted in
6712 depending on
6713 in sync betwe
6714 global - Event time s
6715 CPUs. May be
6716 but better fo
6717 counter - Simple coun
6718 note, some co
6719 infrastructur
6720 once per even
6721 uptime - Use jiffies
6722 perf - Use the same c
6723 mono - Use ktime_get_
6724 mono_raw - Use ktime_
6725 stamps.
6726 boot - Use ktime_get_
6727 Architectures may add
6728 Documentation/trace/f
6729
6730 trace_event=[event-list]
6731 [FTRACE] Set and star
6732 to facilitate early b
6733 comma-separated list
6734 also Documentation/tr
6735
6736 trace_instance=[instance-info]
6737 [FTRACE] Create a rin
6738 This will be listed i
6739
6740 /sys/kernel/t
6741
6742 Events can be enabled
6743 via:
6744
6745 trace_instanc
6746
6747 Note, the "<system*>:
6748 unique.
6749
6750 trace_instanc
6751
6752 will enable the "sche
6753 the same thing would
6754 event, and all events
6755
6756 trace_options=[option-list]
6757 [FTRACE] Enable or di
6758 The option-list is a
6759 that can be enabled o
6760 to echo the option na
6761
6762 /sys/kernel/traci
6763
6764 For example, to enabl
6765 stack trace of each e
6766
6767 trace_options=s
6768
6769 See also Documentatio
6770 section.
6771
6772 trace_trigger=[trigger-list]
6773 [FTRACE] Add a event
6774 Set a trigger on top
6775 filter.
6776
6777 The format is is "tra
6778 Where more than one t
6779
6780 For example:
6781
6782 trace_trigger="sche
6783
6784 The above will enable
6785 event but only trigge
6786 event is "2" (TASK_UN
6787
6788 See also "Event trigg
6789
6790
6791 traceoff_on_warning
6792 [FTRACE] enable this
6793 warning is hit. This
6794 be enabled again by e
6795 file located in /sys/
6796
6797 This option is useful
6798 the WARNING dump is c
6799 be filled with conten
6800
6801 This option can also
6802 option: kernel/trace
6803
6804 transparent_hugepage=
6805 [KNL]
6806 Format: [always|madvi
6807 Can be used to contro
6808 with respect to trans
6809 See Documentation/adm
6810 for more details.
6811
6812 trusted.source= [KEYS]
6813 Format: <string>
6814 This parameter identi
6815 for trusted keys impl
6816 sources:
6817 - "tpm"
6818 - "tee"
6819 - "caam"
6820 - "dcp"
6821 If not specified then
6822 the trust source list
6823 first trust source as
6824 successfully during i
6825
6826 trusted.rng= [KEYS]
6827 Format: <string>
6828 The RNG used to gener
6829 Can be one of:
6830 - "kernel"
6831 - the same value as t
6832 - "default"
6833 If not specified, "de
6834 the RNG's choice is l
6835
6836 trusted.dcp_use_otp_key
6837 This is intended to b
6838 trusted.source=dcp an
6839 instead of the DCP UN
6840
6841 trusted.dcp_skip_zk_test
6842 This is intended to b
6843 trusted.source=dcp an
6844 blob key is all zeros
6845 having this key zero'
6846 scenarios.
6847
6848 tsc= Disable clocksource s
6849 Format: <string>
6850 [x86] reliable: mark
6851 disables clocksource
6852 as the stability chec
6853 high-resolution timer
6854 virtualized environme
6855 [x86] noirqtime: Do n
6856 Used to run time disa
6857 platforms where RDTSC
6858 can add overhead.
6859 [x86] unstable: mark
6860 marks the TSC uncondi
6861 avoids any further wo
6862 [x86] nowatchdog: dis
6863 in situations with st
6864 interruptions from cl
6865 acceptable).
6866 [x86] recalibrate: fo
6867 (HPET or PM timer) on
6868 obtained from HW or F
6869 Warn if the differenc
6870 [x86] watchdog: Use T
6871 which to check other
6872 only on systems where
6873 This will be suppress
6874 can be overridden by
6875 message will flag any
6876
6877 tsc_early_khz= [X86,EARLY] Skip earl
6878 value instead. Useful
6879 procedure is not reli
6880 with CPUID.16h suppor
6881 Format: <unsigned int
6882
6883 tsx= [X86] Control Transac
6884 Extensions (TSX) feat
6885 support TSX control.
6886
6887 This parameter contro
6888
6889 on - Enable TSX
6890 mitigations f
6891 TSX has been
6892 several previ
6893 so there may
6894 with leaving
6895
6896 off - Disable TSX
6897 option takes
6898 not vulnerabl
6899 MSR_IA32_ARCH
6900 the new IA32_
6901 update. This
6902 deactivation
6903
6904 auto - Disable TSX
6905 otherwise e
6906
6907 Not specifying this o
6908
6909 See Documentation/adm
6910 for more details.
6911
6912 tsx_async_abort= [X86,INTEL,EARLY] Co
6913 Abort (TAA) vulnerabi
6914
6915 Similar to Micro-arch
6916 certain CPUs that sup
6917 Synchronization Exten
6918 exploit against CPU i
6919 information to a disc
6920 conditions.
6921
6922 In vulnerable process
6923 data can be used in a
6924 access data to which
6925 access.
6926
6927 This parameter contro
6928 options are:
6929
6930 full - Enable T
6931 if TSX i
6932
6933 full,nosmt - Enable T
6934 vulnerab
6935 is not d
6936 vulnerab
6937 off - Uncondit
6938
6939 On MDS-affected machi
6940 prevented by an activ
6941 are mitigated with th
6942 this mitigation, you
6943
6944 Not specifying this o
6945 tsx_async_abort=full.
6946 and deploy MDS mitiga
6947 required and doesn't
6948 mitigation.
6949
6950 For details see:
6951 Documentation/admin-g
6952
6953 turbografx.map[2|3]= [HW,JOY]
6954 TurboGraFX parallel p
6955 Format:
6956 <port#>,<js1>,<js2>,<
6957 See also Documentatio
6958
6959 udbg-immortal [PPC] When debugging
6960 happen after console_
6961 console driver takes
6962 help "seeing" what's
6963
6964 uhash_entries= [KNL,NET]
6965 Set number of hash bu
6966
6967 uhci-hcd.ignore_oc=
6968 [USB] Ignore overcurr
6969 Some badly-designed m
6970 bogus events, for por
6971 anything. Set this p
6972 Note that genuine ove
6973 reported either.
6974
6975 unknown_nmi_panic
6976 [X86] Cause panic on
6977
6978 unwind_debug [X86-64,EARLY]
6979 Enable unwinder debug
6980 useful for debugging
6981 conditions, including
6982 bad/missing unwinder
6983
6984 usbcore.authorized_default=
6985 [USB] Default USB dev
6986 (default -1 = authori
6987 0 = not authorized, 1
6988 if device connected t
6989
6990 usbcore.autosuspend=
6991 [USB] The autosuspend
6992 for newly-detected US
6993 is the time required
6994 autosuspended. Devic
6995 to a negative value w
6996
6997 usbcore.usbfs_snoop=
6998 [USB] Set to log all
6999
7000 usbcore.usbfs_snoop_max=
7001 [USB] Maximum number
7002 (default = 65536).
7003
7004 usbcore.blinkenlights=
7005 [USB] Set to cycle le
7006
7007 usbcore.old_scheme_first=
7008 [USB] Start with the
7009 scheme (default 0 = o
7010
7011 usbcore.usbfs_memory_mb=
7012 [USB] Memory limit (i
7013 usbfs (default = 16,
7014
7015 usbcore.use_both_schemes=
7016 [USB] Try the other d
7017 if the first one fail
7018
7019 usbcore.initial_descriptor_timeout=
7020 [USB] Specifies timeo
7021 USB_REQ_GET_DESCRIPTO
7022 (default 5000 = 5.0 s
7023
7024 usbcore.nousb [USB] Disable the USB
7025
7026 usbcore.quirks=
7027 [USB] A list of quirk
7028 usb core quirk list.
7029 commas. Each entry ha
7030 VendorID:ProductID:Fl
7031 numbers and Flags is
7032 will change the built
7033 clear and clearing it
7034 the following meaning
7035 a = USB_QUIRK
7036 descr
7037 a 255
7038 b = USB_QUIRK
7039 corre
7040 c = USB_QUIRK
7041 Set-I
7042 d = USB_QUIRK
7043 handl
7044 strin
7045 e = USB_QUIRK
7046 (e.g
7047 f = USB_QUIRK
7048 more
7049 bNumI
7050 talki
7051 g = USB_QUIRK
7052 durin
7053 the d
7054 h = USB_QUIRK
7055 high
7056 endpo
7057 requi
7058 micro
7059 calcu
7060 (bInt
7061 Devic
7062 bInte
7063 calcu
7064 varia
7065 i = USB_QUIRK
7066 handl
7067 reque
7068 j = USB_QUIRK
7069 gener
7070 remot
7071 k = USB_QUIRK
7072 Power
7073 l = USB_QUIRK
7074 (Devi
7075 frame
7076 calcu
7077 m = USB_QUIRK
7078 to be
7079 preve
7080 n = USB_QUIRK
7081 pause
7082 o = USB_QUIRK
7083 delay
7084 p = USB_QUIRK
7085 (Redu
7086 reque
7087 Example: quirks=0781:
7088
7089 usbhid.mousepoll=
7090 [USBHID] The interval
7091
7092 usbhid.jspoll=
7093 [USBHID] The interval
7094
7095 usbhid.kbpoll=
7096 [USBHID] The interval
7097
7098 usb-storage.delay_use=
7099 [UMS] The delay in se
7100 scanned for Logical U
7101 Optionally the delay
7102 suffix with "ms".
7103 Example: delay_use=25
7104
7105 usb-storage.quirks=
7106 [UMS] A list of quirk
7107 override the built-in
7108 entries are separated
7109 the form VID:PID:Flag
7110 and Product ID values
7111 Flags is a set of cha
7112 to a common usb-stora
7113 a = SANE_SENS
7114 of se
7115 b = BAD_SENSE
7116 bytes
7117 c = FIX_CAPAC
7118 devic
7119 d = NO_READ_D
7120 READ_
7121 e = NO_READ_C
7122 READ_
7123 f = NO_REPORT
7124 comma
7125 g = MAX_SECTO
7126 240 s
7127 h = CAPACITY_
7128 repor
7129 secto
7130 i = IGNORE_DE
7131 devic
7132 j = NO_REPORT
7133 comma
7134 k = NO_SAME (
7135 l = NOT_LOCKA
7136 unloc
7137 m = MAX_SECTO
7138 than
7139 not o
7140 n = INITIAL_R
7141 initi
7142 o = CAPACITY_
7143 repor
7144 p = WRITE_CAC
7145 by de
7146 r = IGNORE_RE
7147 bogus
7148 s = SINGLE_LU
7149 Logic
7150 t = NO_ATA_1X
7151 comma
7152 u = IGNORE_UA
7153 w = NO_WP_DET
7154 mediu
7155 y = ALWAYS_SY
7156 even
7157 not o
7158 Example: quirks=0419:
7159
7160 user_debug= [KNL,ARM]
7161 Format: <int>
7162 See arch/arm/Kconfig.
7163 1 - undefine
7164 2 - system c
7165 4 - invalid
7166 8 - SIGSEGV
7167 16 - SIGBUS f
7168 Example: user_debug=3
7169
7170 userpte=
7171 [X86,EARLY] Flags con
7172
7173 nohigh = do n
7174 HIGHM
7175 of CO
7176
7177 vdso= [X86,SH,SPARC]
7178 On X86_32, this is an
7179
7180 vdso=1: enable VDSO (
7181 vdso=0: disable VDSO
7182
7183 vdso32= [X86] Control the 32-
7184 vdso32=1: enable 32-b
7185 vdso32=0 or vdso32=2:
7186
7187 See the help text for
7188 details. If CONFIG_C
7189 vdso32=0; otherwise,
7190
7191 For compatibility wit
7192 alias for vdso32=0.
7193
7194 Try vdso32=0 if you e
7195 dl_main: Assertion `(
7196
7197 video= [FB,EARLY] Frame buff
7198 See Documentation/fb/
7199
7200 video.brightness_switch_enabled= [ACP
7201 Format: [0|1]
7202 If set to 1, on recei
7203 generated by hotkey,
7204 level and then send o
7205 the allocated input d
7206 will only send out th
7207 brightness level.
7208 default: 1
7209
7210 virtio_mmio.device=
7211 [VMMIO] Memory mapped
7212
7213 <size>@<basead
7214 where:
7215 <size> :=
7216
7217 <baseaddr> :=
7218 <irq> :=
7219
7220 <id> :=
7221 example:
7222 virtio_mmio.d
7223
7224 Can be used multiple
7225
7226 vga= [BOOT,X86-32] Select
7227 See Documentation/arc
7228 Documentation/admin-g
7229 Use vga=ask for menu.
7230 This is actually a bo
7231 passed to the kernel
7232
7233 vm_debug[=options] [KNL] Availab
7234 May slow down system
7235 enabled on systems wi
7236 All options are enabl
7237 interface is meant to
7238 enabling or disabling
7239 debugging features.
7240
7241 Available options are
7242 P Enable page s
7243 - Disable all o
7244
7245 vmalloc=nn[KMG] [KNL,BOOT,EARLY] Forc
7246 exact size of <nn>. T
7247 the minimum size (128
7248 It can also be used t
7249 for directly mapped k
7250 not exist on many oth
7251 loongarch, arc, csky,
7252 parisc, m64k, powerpc
7253
7254 vmcp_cma=nn[MG] [KNL,S390,EARLY]
7255 Sets the memory size
7256 allocations for the v
7257
7258 vmhalt= [KNL,S390] Perform z/
7259 Format: <command>
7260
7261 vmpanic= [KNL,S390] Perform z/
7262 Format: <command>
7263
7264 vmpoff= [KNL,S390] Perform z/
7265 Format: <command>
7266
7267 vsyscall= [X86-64,EARLY]
7268 Controls the behavior
7269 fixed addresses of 0x
7270 code). Most statical
7271 versions of glibc use
7272 functions are at fixe
7273 targets for exploits
7274
7275 emulate Vsyscalls
7276 reasonabl
7277 readable.
7278
7279 xonly [default]
7280 emulated
7281 page is n
7282
7283 none Vsyscalls
7284 them quit
7285 might bre
7286
7287 vt.color= [VT] Default text col
7288 Format: 0xYX, X = for
7289 Default: 0x07 = light
7290
7291 vt.cur_default= [VT] Default cursor s
7292 Format: 0xCCBBAA, whe
7293 the parameters of the
7294 see VGA-softcursor.tx
7295
7296 vt.default_blu= [VT]
7297 Format: <blue0>,<blue
7298 Change the default bl
7299 This is a 16-member a
7300 ranging from 0-255.
7301
7302 vt.default_grn= [VT]
7303 Format: <green0>,<gre
7304 Change the default gr
7305 This is a 16-member a
7306 ranging from 0-255.
7307
7308 vt.default_red= [VT]
7309 Format: <red0>,<red1>
7310 Change the default re
7311 This is a 16-member a
7312 ranging from 0-255.
7313
7314 vt.default_utf8=
7315 [VT]
7316 Format=<0|1>
7317 Set system-wide defau
7318 Default is 1, i.e. UT
7319 newly opened terminal
7320
7321 vt.global_cursor_default=
7322 [VT]
7323 Format=<-1|0|1>
7324 Set system-wide defau
7325 is shown on new VTs.
7326 i.e. cursors will be
7327 overridden by individ
7328 cursors, 1 will displ
7329
7330 vt.italic= [VT] Default color fo
7331 Default: 2 = green.
7332
7333 vt.underline= [VT] Default color fo
7334 Default: 3 = cyan.
7335
7336 watchdog timers [HW,WDT] For informat
7337 see Documentation/wat
7338 or other driver-speci
7339 Documentation/watchdo
7340
7341 watchdog_thresh=
7342 [KNL]
7343 Set the hard lockup d
7344 threshold in seconds.
7345 threshold is set to t
7346 disables both lockup
7347 seconds.
7348
7349 workqueue.unbound_cpus=
7350 [KNL,SMP] Specify to
7351 to use in unbound wor
7352 Format: <cpu-list>
7353 By default, all onlin
7354 unbound workqueues.
7355
7356 workqueue.watchdog_thresh=
7357 If CONFIG_WQ_WATCHDOG
7358 warn stall conditions
7359 help debugging. 0 di
7360 detection; otherwise,
7361 duration in seconds.
7362 it can be updated at
7363 corresponding sysfs f
7364
7365 workqueue.cpu_intensive_thresh_us=
7366 Per-cpu work items wh
7367 threshold are automat
7368 and excluded from con
7369 them from noticeably
7370 items. Default is 100
7371
7372 If CONFIG_WQ_CPU_INTE
7373 will report the work
7374 threshold repeatedly.
7375 candidates for using
7376
7377 workqueue.cpu_intensive_warning_thres
7378 If CONFIG_WQ_CPU_INTE
7379 will report the work
7380 intensive_threshold_u
7381 spurious warnings, st
7382 function has violated
7383
7384 The default is 4 time
7385
7386 workqueue.power_efficient
7387 Per-cpu workqueues ar
7388 they show better perf
7389 locality; unfortunate
7390 be more power hungry
7391
7392 Enabling this makes t
7393 were observed to cont
7394 consumption unbound,
7395 power usage at the co
7396 overhead.
7397
7398 The default value of
7399 the config option CON
7400
7401 workqueue.default_affinity_scope=
7402 Select the default af
7403 workqueues. Can be on
7404 "numa" and "system".
7405 information, see the
7406 Documentation/core-ap
7407
7408 This can be changed a
7409 matching /sys/module/
7410 workqueues with the "
7411 updated accordingly.
7412
7413 workqueue.debug_force_rr_cpu
7414 Workqueue used to imp
7415 items queued without
7416 on the local CPU. Th
7417 and while local CPU i
7418 may be put on foreign
7419 forces round-robin CP
7420 usages which depend o
7421 When enabled, memory
7422 impacted.
7423
7424 writecombine= [LOONGARCH,EARLY] Con
7425 Type) of ioremap_wc()
7426
7427 on - Enable writeco
7428 off - Disable writec
7429
7430 x2apic_phys [X86-64,APIC,EARLY] U
7431 default x2apic cluste
7432 supporting x2apic.
7433
7434 xen_512gb_limit [KNL,X86-64,X
7435 Restricts the kernel
7436 to use only up to 512
7437 crash analysis tools
7438 save/restore/migratio
7439 domains.
7440
7441 xen_emul_unplug= [HW,X
7442 Unplug Xen emulated d
7443 Format: [unplug0,][un
7444 ide-disks -- unplug p
7445 aux-ide-disks -- unpl
7446 nics -- unplug networ
7447 all -- unplug all emu
7448 unnecessary -- unplug
7449 unnecessary e
7450 the unplug pr
7451 never -- do not unplu
7452
7453 xen_legacy_crash [X86,XEN,EARL
7454 Crash from Xen panic
7455 panic() code such as
7456
7457 xen_mc_debug [X86,XEN,EARLY]
7458 Enable multicall debu
7459 Enabling this feature
7460 bit, so it should onl
7461 debug data in case of
7462
7463 xen_msr_safe= [X86,XEN,EARLY]
7464 Format: <bool>
7465 Select whether to alw
7466 access functions when
7467 default value is cont
7468
7469 xen_nopv [X86]
7470 Disables the PV optim
7471 run as generic HVM gu
7472 This option is obsole
7473 has equivalent effect
7474
7475 xen_no_vector_callback
7476 [KNL,X86,XEN,EARLY] D
7477 event channel interru
7478
7479 xen_scrub_pages= [XEN]
7480 Boolean option to con
7481 to Xen, for use by ot
7482 with /sys/devices/sys
7483 Default value control
7484
7485 xen_timer_slop= [X86-64,XEN,EARLY]
7486 Set the timer slop (i
7487 timers (default is 10
7488 delta of virtualized
7489 improve timer resolut
7490 more timer interrupts
7491
7492 xen.balloon_boot_timeout= [XEN]
7493 The time (in seconds)
7494 in case initial ballo
7495 Applies only when run
7496 started with less mem
7497 max. Default is 180.
7498
7499 xen.event_eoi_delay= [XEN]
7500 How long to delay EOI
7501 storms (jiffies). Def
7502
7503 xen.event_loop_timeout= [XEN]
7504 After which time (jif
7505 should start to delay
7506
7507 xen.fifo_events= [XEN]
7508 Boolean parameter to
7509 even if available. No
7510 preferred over the 2-
7511 fairer and the number
7512 much higher. Default
7513
7514 xirc2ps_cs= [NET,PCMCIA]
7515 Format:
7516 <irq>,<irq_mask>,<io>
7517
7518 xive= [PPC]
7519 By default on POWER9
7520 natively use the XIVE
7521 allows the fallback f
7522
7523 off Fallback to
7524 controller
7525 platforms.
7526
7527 xive.store-eoi=off [PPC]
7528 By default on POWER10
7529 stores for EOI handli
7530 is active. This optio
7531 loads instead, as on
7532
7533 xhci-hcd.quirks [USB,KNL]
7534 A hex value specifyin
7535 host controller quirk
7536 consulted in header d
7537
7538 xmon [PPC,EARLY]
7539 Format: { early | on
7540 Controls if xmon debu
7541 Passing only "xmon" i
7542 early Call xmon as
7543 debugger is c
7544 on xmon debugger
7545 is only calle
7546 i.e. either "
7547 with CONFIG_X
7548 rw xmon debugger
7549 is called onl
7550 meaning SPR r
7551 can be writte
7552 ro same as "rw"
7553 memory, and o
7554 xmon commands
7555 off xmon is disab
Linux® is a registered trademark of Linus Torvalds in the United States and other countries.
TOMOYO® is a registered trademark of NTT DATA CORPORATION.