1 acpi= [HW,ACPI,X86,ARM64]
2 Advanced Configuration and Power Interface
3 Format: { force | on | off | strict | noirq | rsdt |
5 force -- enable ACPI if default was off
6 on -- enable ACPI but allow fallback to DT [arm64]
7 off -- disable ACPI if default was on
8 noirq -- do not use ACPI for IRQ routing
9 strict -- Be less tolerant of platforms that are not
10 strictly ACPI specification compliant.
11 rsdt -- prefer RSDT over (default) XSDT
12 copy_dsdt -- copy DSDT to memory
13 For ARM64, ONLY "acpi=off", "acpi=on" or "acpi=force"
16 See also Documentation/power/runtime_pm.rst, pci=noacpi
18 acpi_apic_instance= [ACPI, IOAPIC]
20 2: use 2nd APIC table, if available
21 1,0: use 1st APIC table
24 acpi_backlight= [HW,ACPI]
25 { vendor | video | native | none }
26 If set to vendor, prefer vendor-specific driver
27 (e.g. thinkpad_acpi, sony_acpi, etc.) instead
28 of the ACPI video.ko driver.
29 If set to video, use the ACPI video.ko driver.
30 If set to native, use the device's native backlight mode.
31 If set to none, disable the ACPI backlight interface.
33 acpi_force_32bit_fadt_addr
34 force FADT to use 32 bit addresses rather than the
35 64 bit X_* addresses. Some firmware have broken 64
36 bit addresses for force ACPI ignore these and use
37 the older legacy 32 bit addresses.
39 acpica_no_return_repair [HW, ACPI]
40 Disable AML predefined validation mechanism
41 This mechanism can repair the evaluation result to make
42 the return objects more ACPI specification compliant.
43 This option is useful for developers to identify the
44 root cause of an AML interpreter issue when the issue
45 has something to do with the repair mechanism.
47 acpi.debug_layer= [HW,ACPI,ACPI_DEBUG]
48 acpi.debug_level= [HW,ACPI,ACPI_DEBUG]
50 CONFIG_ACPI_DEBUG must be enabled to produce any ACPI
51 debug output. Bits in debug_layer correspond to a
52 _COMPONENT in an ACPI source file, e.g.,
53 #define _COMPONENT ACPI_EVENTS
54 Bits in debug_level correspond to a level in
55 ACPI_DEBUG_PRINT statements, e.g.,
56 ACPI_DEBUG_PRINT((ACPI_DB_INFO, ...
57 The debug_level mask defaults to "info". See
58 Documentation/firmware-guide/acpi/debug.rst for more information about
59 debug layers and levels.
61 Enable processor driver info messages:
62 acpi.debug_layer=0x20000000
63 Enable AML "Debug" output, i.e., stores to the Debug
64 object while interpreting AML:
65 acpi.debug_layer=0xffffffff acpi.debug_level=0x2
66 Enable all messages related to ACPI hardware:
67 acpi.debug_layer=0x2 acpi.debug_level=0xffffffff
69 Some values produce so much output that the system is
70 unusable. The "log_buf_len" parameter may be useful
71 if you need to capture more output.
73 acpi_enforce_resources= [ACPI]
75 Check for resource conflicts between native drivers
76 and ACPI OperationRegions (SystemIO and SystemMemory
77 only). IO ports and memory declared in ACPI might be
78 used by the ACPI subsystem in arbitrary AML code and
79 can interfere with legacy drivers.
80 strict (default): access to resources claimed by ACPI
81 is denied; legacy drivers trying to access reserved
82 resources will fail to bind to device using them.
83 lax: access to resources claimed by ACPI is allowed;
84 legacy drivers trying to access reserved resources
85 will bind successfully but a warning message is logged.
86 no: ACPI OperationRegions are not marked as reserved,
87 no further checks are performed.
89 acpi_force_table_verification [HW,ACPI]
90 Enable table checksum verification during early stage.
91 By default, this is disabled due to x86 early mapping
94 acpi_irq_balance [HW,ACPI]
95 ACPI will balance active IRQs
98 acpi_irq_nobalance [HW,ACPI]
99 ACPI will not move active IRQs (default)
102 acpi_irq_isa= [HW,ACPI] If irq_balance, mark listed IRQs used by ISA
103 Format: <irq>,<irq>...
105 acpi_irq_pci= [HW,ACPI] If irq_balance, clear listed IRQs for
107 Format: <irq>,<irq>...
109 acpi_mask_gpe= [HW,ACPI]
110 Due to the existence of _Lxx/_Exx, some GPEs triggered
111 by unsupported hardware/firmware features can result in
112 GPE floodings that cannot be automatically disabled by
114 This facility can be used to prevent such uncontrolled
118 acpi_no_auto_serialize [HW,ACPI]
119 Disable auto-serialization of AML methods
120 AML control methods that contain the opcodes to create
121 named objects will be marked as "Serialized" by the
122 auto-serialization feature.
123 This feature is enabled by default.
124 This option allows to turn off the feature.
126 acpi_no_memhotplug [ACPI] Disable memory hotplug. Useful for kdump
129 acpi_no_static_ssdt [HW,ACPI]
130 Disable installation of static SSDTs at early boot time
131 By default, SSDTs contained in the RSDT/XSDT will be
132 installed automatically and they will appear under
133 /sys/firmware/acpi/tables.
134 This option turns off this feature.
135 Note that specifying this option does not affect
136 dynamic table installation which will install SSDT
137 tables to /sys/firmware/acpi/tables/dynamic.
139 acpi_no_watchdog [HW,ACPI,WDT]
140 Ignore the ACPI-based watchdog interface (WDAT) and let
141 a native driver control the watchdog device instead.
143 acpi_rsdp= [ACPI,EFI,KEXEC]
144 Pass the RSDP address to the kernel, mostly used
145 on machines running EFI runtime service to boot the
146 second kernel for kdump.
148 acpi_os_name= [HW,ACPI] Tell ACPI BIOS the name of the OS
149 Format: To spoof as Windows 98: ="Microsoft Windows"
151 acpi_rev_override [ACPI] Override the _REV object to return 5 (instead
152 of 2 which is mandated by ACPI 6) as the supported ACPI
153 specification revision (when using this switch, it may
154 be necessary to carry out a cold reboot _twice_ in a
155 row to make it take effect on the platform firmware).
157 acpi_osi= [HW,ACPI] Modify list of supported OS interface strings
158 acpi_osi="string1" # add string1
159 acpi_osi="!string2" # remove string2
160 acpi_osi=!* # remove all strings
161 acpi_osi=! # disable all built-in OS vendor
163 acpi_osi=!! # enable all built-in OS vendor
165 acpi_osi= # disable all strings
167 'acpi_osi=!' can be used in combination with single or
168 multiple 'acpi_osi="string1"' to support specific OS
169 vendor string(s). Note that such command can only
170 affect the default state of the OS vendor strings, thus
171 it cannot affect the default state of the feature group
172 strings and the current state of the OS vendor strings,
173 specifying it multiple times through kernel command line
174 is meaningless. This command is useful when one do not
175 care about the state of the feature group strings which
176 should be controlled by the OSPM.
178 1. 'acpi_osi=! acpi_osi="Windows 2000"' is equivalent
179 to 'acpi_osi="Windows 2000" acpi_osi=!', they all
180 can make '_OSI("Windows 2000")' TRUE.
182 'acpi_osi=' cannot be used in combination with other
183 'acpi_osi=' command lines, the _OSI method will not
184 exist in the ACPI namespace. NOTE that such command can
185 only affect the _OSI support state, thus specifying it
186 multiple times through kernel command line is also
189 1. 'acpi_osi=' can make 'CondRefOf(_OSI, Local1)'
192 'acpi_osi=!*' can be used in combination with single or
193 multiple 'acpi_osi="string1"' to support specific
194 string(s). Note that such command can affect the
195 current state of both the OS vendor strings and the
196 feature group strings, thus specifying it multiple times
197 through kernel command line is meaningful. But it may
198 still not able to affect the final state of a string if
199 there are quirks related to this string. This command
200 is useful when one want to control the state of the
201 feature group strings to debug BIOS issues related to
204 1. 'acpi_osi="Module Device" acpi_osi=!*' can make
205 '_OSI("Module Device")' FALSE.
206 2. 'acpi_osi=!* acpi_osi="Module Device"' can make
207 '_OSI("Module Device")' TRUE.
208 3. 'acpi_osi=! acpi_osi=!* acpi_osi="Windows 2000"' is
210 'acpi_osi=!* acpi_osi=! acpi_osi="Windows 2000"'
212 'acpi_osi=!* acpi_osi="Windows 2000" acpi_osi=!',
213 they all will make '_OSI("Windows 2000")' TRUE.
216 Override the pmtimer bug detection: force the kernel
217 to assume that this machine's pmtimer latches its value
218 and always returns good values.
220 acpi_sci= [HW,ACPI] ACPI System Control Interrupt trigger mode
221 Format: { level | edge | high | low }
223 acpi_skip_timer_override [HW,ACPI]
224 Recognize and ignore IRQ0/pin2 Interrupt Override.
225 For broken nForce2 BIOS resulting in XT-PIC timer.
227 acpi_sleep= [HW,ACPI] Sleep options
228 Format: { s3_bios, s3_mode, s3_beep, s4_nohwsig,
229 old_ordering, nonvs, sci_force_enable, nobl }
230 See Documentation/power/video.rst for information on
232 s3_beep is for debugging; it makes the PC's speaker beep
233 as soon as the kernel's real-mode entry point is called.
234 s4_nohwsig prevents ACPI hardware signature from being
235 used during resume from hibernation.
236 old_ordering causes the ACPI 1.0 ordering of the _PTS
237 control method, with respect to putting devices into
238 low power states, to be enforced (the ACPI 2.0 ordering
239 of _PTS is used by default).
240 nonvs prevents the kernel from saving/restoring the
241 ACPI NVS memory during suspend/hibernation and resume.
242 sci_force_enable causes the kernel to set SCI_EN directly
243 on resume from S1/S3 (which is against the ACPI spec,
244 but some broken systems don't work without it).
245 nobl causes the internal blacklist of systems known to
246 behave incorrectly in some ways with respect to system
247 suspend and resume to be ignored (use wisely).
249 acpi_use_timer_override [HW,ACPI]
250 Use timer override. For some broken Nvidia NF5 boards
251 that require a timer override, but don't have HPET
253 add_efi_memmap [EFI; X86] Include EFI memory map in
254 kernel's map of available physical RAM.
257 { off | try_unsupported }
258 off: disable AGP support
259 try_unsupported: try to drive unsupported chipsets
260 (may crash computer or cause data corruption)
263 See Documentation/sound/alsa-configuration.rst
266 Allow the default userspace alignment fault handler
267 behaviour to be specified. Bit 0 enables warnings,
268 bit 1 enables fixups, and bit 2 sends a segfault.
270 align_va_addr= [X86-64]
271 Align virtual addresses by clearing slice [14:12] when
272 allocating a VMA at process creation time. This option
273 gives you up to 3% performance improvement on AMD F15h
274 machines (where it is enabled by default) for a
275 CPU-intensive style benchmark, and it can vary highly in
276 a microbenchmark depending on workload and compiler.
278 32: only for 32-bit processes
279 64: only for 64-bit processes
280 on: enable for both 32- and 64-bit processes
281 off: disable for both 32- and 64-bit processes
283 alloc_snapshot [FTRACE]
284 Allocate the ftrace snapshot buffer on boot up when the
285 main buffer is allocated. This is handy if debugging
286 and you need to use tracing_snapshot() on boot up, and
287 do not want to use tracing_snapshot_alloc() as it needs
288 to be done where GFP_KERNEL allocations are allowed.
290 amd_iommu= [HW,X86-64]
291 Pass parameters to the AMD IOMMU driver in the system.
293 fullflush - enable flushing of IO/TLB entries when
294 they are unmapped. Otherwise they are
295 flushed before they will be reused, which
297 off - do not initialize any AMD IOMMU found in
299 force_isolation - Force device isolation for all
300 devices. The IOMMU driver is not
301 allowed anymore to lift isolation
302 requirements as needed. This option
303 does not override iommu=pt
305 amd_iommu_dump= [HW,X86-64]
306 Enable AMD IOMMU driver option to dump the ACPI table
307 for AMD IOMMU. With this option enabled, AMD IOMMU
308 driver will print ACPI tables for AMD IOMMU during
309 IOMMU initialization.
311 amd_iommu_intr= [HW,X86-64]
312 Specifies one of the following AMD IOMMU interrupt
314 legacy - Use legacy interrupt remapping mode.
315 vapic - Use virtual APIC mode, which allows IOMMU
316 to inject interrupts directly into guest.
317 This mode requires kvm-amd.avic=1.
318 (Default when IOMMU HW support is present.)
320 amijoy.map= [HW,JOY] Amiga joystick support
321 Map of devices attached to JOY0DAT and JOY1DAT
323 See also Documentation/input/joydev/joystick.rst
325 analog.map= [HW,JOY] Analog joystick and gamepad support
326 Specifies type or capabilities of an analog joystick
327 connected to one of 16 gameports
328 Format: <type1>,<type2>,..<type16>
331 Power management functions (SPARCstation-4/5 + deriv.)
333 Disable APC CPU standby support. SPARCstation-Fox does
334 not play well with APC CPU idle - disable it if you have
335 APC and your system crashes randomly.
337 apic= [APIC,X86] Advanced Programmable Interrupt Controller
338 Change the output verbosity while booting
339 Format: { quiet (default) | verbose | debug }
340 Change the amount of debugging information output
341 when initialising the APIC and IO-APIC components.
342 For X86-32, this can also be used to specify an APIC
344 Format: apic=driver_name
345 Examples: apic=bigsmp
347 apic_extnmi= [APIC,X86] External NMI delivery setting
348 Format: { bsp (default) | all | none }
349 bsp: External NMI is delivered only to CPU 0
350 all: External NMIs are broadcast to all CPUs as a
352 none: External NMI is masked for all CPUs. This is
353 useful so that a dump capture kernel won't be
357 See Documentation/networking/ipv6.rst.
359 show_lapic= [APIC,X86] Advanced Programmable Interrupt Controller
360 Limit apic dumping. The parameter defines the maximal
361 number of local apics being dumped. Also it is possible
362 to set it to "all" by meaning -- no limit here.
363 Format: { 1 (default) | 2 | ... | all }.
364 The parameter valid if only apic=debug or
365 apic=verbose is specified.
366 Example: apic=debug show_lapic=all
368 apm= [APM] Advanced Power Management
369 See header of arch/x86/kernel/apm_32.c.
371 arcrimi= [HW,NET] ARCnet - "RIM I" (entirely mem-mapped) cards
372 Format: <io>,<irq>,<nodeID>
374 arm64.nobti [ARM64] Unconditionally disable Branch Target
375 Identification support
377 arm64.nopauth [ARM64] Unconditionally disable Pointer Authentication
382 atarimouse= [HW,MOUSE] Atari Mouse
384 atkbd.extra= [HW] Enable extra LEDs and keys on IBM RapidAccess,
385 EzKey and similar keyboards
387 atkbd.reset= [HW] Reset keyboard during initialization
389 atkbd.set= [HW] Select keyboard code set
390 Format: <int> (2 = AT (default), 3 = PS/2)
392 atkbd.scroll= [HW] Enable scroll wheel on MS Office and similar
395 atkbd.softraw= [HW] Choose between synthetic and real raw mode
396 Format: <bool> (0 = real, 1 = synthetic (default))
398 atkbd.softrepeat= [HW]
399 Use software keyboard repeat
401 audit= [KNL] Enable the audit sub-system
402 Format: { "0" | "1" | "off" | "on" }
403 0 | off - kernel audit is disabled and can not be
404 enabled until the next reboot
405 unset - kernel audit is initialized but disabled and
406 will be fully enabled by the userspace auditd.
407 1 | on - kernel audit is initialized and partially
408 enabled, storing at most audit_backlog_limit
409 messages in RAM until it is fully enabled by the
413 audit_backlog_limit= [KNL] Set the audit queue size limit.
414 Format: <int> (must be >=0)
417 bau= [X86_UV] Enable the BAU on SGI UV. The default
418 behavior is to disable the BAU (i.e. bau=0).
419 Format: { "0" | "1" }
422 unset - Disable the BAU.
424 baycom_epp= [HW,AX25]
427 baycom_par= [HW,AX25] BayCom Parallel Port AX.25 Modem
429 See header of drivers/net/hamradio/baycom_par.c.
431 baycom_ser_fdx= [HW,AX25]
432 BayCom Serial Port AX.25 Modem (Full Duplex Mode)
433 Format: <io>,<irq>,<mode>[,<baud>]
434 See header of drivers/net/hamradio/baycom_ser_fdx.c.
436 baycom_ser_hdx= [HW,AX25]
437 BayCom Serial Port AX.25 Modem (Half Duplex Mode)
438 Format: <io>,<irq>,<mode>
439 See header of drivers/net/hamradio/baycom_ser_hdx.c.
441 blkdevparts= Manual partition parsing of block device(s) for
442 embedded devices based on command line input.
443 See Documentation/block/cmdline-partition.rst
445 boot_delay= Milliseconds to delay each printk during boot.
446 Values larger than 10 seconds (10000) are changed to
451 Extended command line options can be added to an initrd
452 and this will cause the kernel to look for it.
454 See Documentation/admin-guide/bootconfig.rst
457 Disable BERT OS support on buggy BIOSes.
459 bgrt_disable [ACPI][X86]
460 Disable BGRT to avoid flickering OEM logo.
462 bttv.card= [HW,V4L] bttv (bt848 + bt878 based grabber cards)
463 bttv.radio= Most important insmod options are available as
465 bttv.pll= See Documentation/admin-guide/media/bttv.rst
468 bulk_remove=off [PPC] This parameter disables the use of the pSeries
469 firmware feature for flushing multiple hpte entries
472 c101= [NET] Moxa C101 synchronous serial card
474 cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection.
475 Sometimes CPU hardware bugs make them report the cache
476 size incorrectly. The kernel will attempt work arounds
477 to fix known problems, but for some CPUs it is not
478 possible to determine what the correct size should be.
479 This option provides an override for these situations.
482 [NET] Specifies amount of time (in seconds) that
483 the kernel should wait for a network carrier. By default
484 it waits 120 seconds.
486 ca_keys= [KEYS] This parameter identifies a specific key(s) on
487 the system trusted keyring to be used for certificate
489 format: { id:<keyid> | builtin }
491 cca= [MIPS] Override the kernel pages' cache coherency
492 algorithm. Accepted values range from 0 to 7
493 inclusive. See arch/mips/include/asm/pgtable-bits.h
494 for platform specific values (SB1, Loongson3 and
497 ccw_timeout_log [S390]
498 See Documentation/s390/common_io.rst for details.
500 cgroup_disable= [KNL] Disable a particular controller
501 Format: {name of the controller(s) to disable}
502 The effects of cgroup_disable=foo are:
503 - foo isn't auto-mounted if you mount all cgroups in
505 - foo isn't visible as an individually mountable
507 {Currently only "memory" controller deal with this and
508 cut the overhead, others just disable the usage. So
509 only cgroup_disable=memory is actually worthy}
511 cgroup_no_v1= [KNL] Disable cgroup controllers and named hierarchies in v1
512 Format: { { controller | "all" | "named" }
513 [,{ controller | "all" | "named" }...] }
514 Like cgroup_disable, but only applies to cgroup v1;
515 the blacklisted controllers remain available in cgroup2.
516 "all" blacklists all controllers and "named" disables
517 named mounts. Specifying both "all" and "named" disables
520 cgroup.memory= [KNL] Pass options to the cgroup memory controller.
522 nosocket -- Disable socket memory accounting.
523 nokmem -- Disable kernel memory accounting.
525 checkreqprot [SELINUX] Set initial checkreqprot flag value.
526 Format: { "0" | "1" }
527 See security/selinux/Kconfig help text.
528 0 -- check protection applied by kernel (includes
529 any implied execute protection).
530 1 -- check protection requested by application.
531 Default value is set via a kernel config option.
532 Value can be changed at runtime via
533 /sys/fs/selinux/checkreqprot.
534 Setting checkreqprot to 1 is deprecated.
537 See Documentation/s390/common_io.rst for details.
540 Prevents the clock framework from automatically gating
541 clocks that have not been explicitly enabled by a Linux
542 device driver but are enabled in hardware at reset or
543 by the bootloader/firmware. Note that this does not
544 force such clocks to be always-on nor does it reserve
545 those clocks in any way. This parameter is useful for
546 debug and development, but should not be needed on a
547 platform with proper driver support. For more
548 information, see Documentation/driver-api/clk.rst.
550 clock= [BUGS=X86-32, HW] gettimeofday clocksource override.
552 Forces specified clocksource (if available) to be used
553 when calculating gettimeofday(). If specified
554 clocksource is not available, it defaults to PIT.
555 Format: { pit | tsc | cyclone | pmtmr }
557 clocksource= Override the default clocksource
559 Override the default clocksource and use the clocksource
560 with the name specified.
561 Some clocksource names to choose from, depending on
563 [all] jiffies (this is the base, fallback clocksource)
565 [ARM] imx_timer1,OSTS,netx_timer,mpu_timer2,
566 pxa_timer,timer3,32k_counter,timer0_1
567 [X86-32] pit,hpet,tsc;
568 scx200_hrt on Geode; cyclone on IBM x440
576 clocksource.arm_arch_timer.evtstrm=
579 Enable/disable the eventstream feature of the ARM
580 architected timer so that code using WFE-based polling
581 loops can be debugged more effectively on production
584 clearcpuid=BITNUM[,BITNUM...] [X86]
585 Disable CPUID feature X for the kernel. See
586 arch/x86/include/asm/cpufeatures.h for the valid bit
587 numbers. Note the Linux specific bits are not necessarily
588 stable over kernel options, but the vendor specific
590 Also note that user programs calling CPUID directly
591 or using the feature without checking anything
592 will still see it. This just prevents it from
593 being used by the kernel or shown in /proc/cpuinfo.
594 Also note the kernel might malfunction if you disable
597 cma=nn[MG]@[start[MG][-end[MG]]]
599 Sets the size of kernel global memory area for
600 contiguous memory allocations and optionally the
601 placement constraint by the physical address range of
602 memory allocations. A value of 0 disables CMA
603 altogether. For more information, see
604 kernel/dma/contiguous.c
608 Sets the size of kernel per-numa memory area for
609 contiguous memory allocations. A value of 0 disables
610 per-numa CMA altogether. And If this option is not
611 specificed, the default value is 0.
612 With per-numa CMA enabled, DMA users on node nid will
613 first try to allocate buffer from the pernuma area
614 which is located in node nid, if the allocation fails,
615 they will fallback to the global default memory area.
617 cmo_free_hint= [PPC] Format: { yes | no }
618 Specify whether pages are marked as being inactive
619 when they are freed. This is used in CMO environments
620 to determine OS memory pressure for page stealing by
624 coherent_pool=nn[KMG] [ARM,KNL]
625 Sets the size of memory pool for coherent, atomic dma
626 allocations, by default set to 256K.
628 com20020= [HW,NET] ARCnet - COM20020 chipset
630 <io>[,<irq>[,<nodeID>[,<backplane>[,<ckp>[,<timeout>]]]]]
632 com90io= [HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers)
636 ARCnet - COM90xx chipset (memory-mapped buffers)
637 Format: <io>[,<irq>[,<memstart>]]
639 condev= [HW,S390] console device
642 console= [KNL] Output console device and options.
644 tty<n> Use the virtual console device <n>.
648 Use the specified serial port. The options are of
649 the form "bbbbpnf", where "bbbb" is the baud rate,
650 "p" is parity ("n", "o", or "e"), "n" is number of
651 bits, and "f" is flow control ("r" for RTS or
652 omit it). Default is "9600n8".
654 See Documentation/admin-guide/serial-console.rst for more
656 Documentation/networking/netconsole.rst for an
659 uart[8250],io,<addr>[,options]
660 uart[8250],mmio,<addr>[,options]
661 uart[8250],mmio16,<addr>[,options]
662 uart[8250],mmio32,<addr>[,options]
663 uart[8250],0x<addr>[,options]
664 Start an early, polled-mode console on the 8250/16550
665 UART at the specified I/O port or MMIO address,
666 switching to the matching ttyS device later.
667 MMIO inter-register address stride is either 8-bit
668 (mmio), 16-bit (mmio16), or 32-bit (mmio32).
669 If none of [io|mmio|mmio16|mmio32], <addr> is assumed
670 to be equivalent to 'mmio'. 'options' are specified in
671 the same format described for ttyS above; if unspecified,
672 the h/w is not re-initialized.
674 hvc<n> Use the hypervisor console device <n>. This is for
675 both Xen and PowerPC hypervisors.
677 If the device connected to the port is not a TTY but a braille
678 device, prepend "brl," before the device type, for instance
680 For now, only VisioBraille is supported.
683 [KNL] Change console messages format
685 By default we print messages on consoles in
686 "[time stamp] text\n" format (time stamp may not be
687 printed, depending on CONFIG_PRINTK_TIME or
688 `printk_time' param).
690 Switch to syslog format: "<%u>[time stamp] text\n"
691 IOW, each message will have a facility and loglevel
692 prefix. The format is similar to one used by syslog()
693 syscall, or to executing "dmesg -S --raw" or to reading
696 consoleblank= [KNL] The console blank (screen saver) timeout in
697 seconds. A value of 0 disables the blank timer.
701 [KNL] Change the default value for
702 /proc/<pid>/coredump_filter.
703 See also Documentation/filesystems/proc.rst.
705 coresight_cpu_debug.enable
708 Enable/disable the CPU sampling based debugging.
709 0: default value, disable debugging
710 1: enable debugging at boot time
712 cpuidle.off=1 [CPU_IDLE]
713 disable the cpuidle sub-system
716 [CPU_IDLE] Name of the cpuidle governor to use.
718 cpufreq.off=1 [CPU_FREQ]
719 disable the cpufreq sub-system
721 cpufreq.default_governor=
722 [CPU_FREQ] Name of the default cpufreq governor or
723 policy to use. This governor must be registered in the
724 kernel before the cpufreq driver probes.
727 [X86] Delay for N microsec between assert and de-assert
728 of APIC INIT to start processors. This delay occurs
729 on every CPU online, such as boot, and resume from suspend.
732 cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver
734 <first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>]
736 crashkernel=size[KMG][@offset[KMG]]
737 [KNL] Using kexec, Linux can switch to a 'crash kernel'
738 upon panic. This parameter reserves the physical
739 memory region [offset, offset + size] for that kernel
740 image. If '@offset' is omitted, then a suitable offset
741 is selected automatically.
742 [KNL, X86-64] Select a region under 4G first, and
743 fall back to reserve region above 4G when '@offset'
744 hasn't been specified.
745 See Documentation/admin-guide/kdump/kdump.rst for further details.
747 crashkernel=range1:size1[,range2:size2,...][@offset]
748 [KNL] Same as above, but depends on the memory
749 in the running system. The syntax of range is
750 start-[end] where start and end are both
751 a memory unit (amount[KMG]). See also
752 Documentation/admin-guide/kdump/kdump.rst for an example.
754 crashkernel=size[KMG],high
755 [KNL, X86-64] range could be above 4G. Allow kernel
756 to allocate physical memory region from top, so could
757 be above 4G if system have more than 4G ram installed.
758 Otherwise memory region will be allocated below 4G, if
760 It will be ignored if crashkernel=X is specified.
761 crashkernel=size[KMG],low
762 [KNL, X86-64] range under 4G. When crashkernel=X,high
763 is passed, kernel could allocate physical memory region
764 above 4G, that cause second kernel crash on system
765 that require some amount of low memory, e.g. swiotlb
766 requires at least 64M+32K low memory, also enough extra
767 low memory is needed to make sure DMA buffers for 32-bit
768 devices won't run out. Kernel would try to allocate at
769 at least 256M below 4G automatically.
770 This one let user to specify own low range under 4G
771 for second kernel instead.
772 0: to disable low allocation.
773 It will be ignored when crashkernel=X,high is not used
774 or memory reserved is below 4G.
777 [KNL] Disable crypto self-tests
782 cs89x0_media= [HW,NET]
783 Format: { rj45 | aui | bnc }
785 csdlock_debug= [KNL] Enable debug add-ons of cross-CPU function call
786 handling. When switched on, additional debug data is
787 printed to the console in case a hanging CPU is
788 detected, and that CPU is pinged again in order to try
789 to resolve the hang situation.
790 0: disable csdlock debugging (default)
791 1: enable basic csdlock debugging (minor impact)
792 ext: enable extended csdlock debugging (more impact,
796 See header of drivers/s390/block/dasd_devmap.c.
798 db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port
799 (one device per port)
800 Format: <port#>,<type>
801 See also Documentation/input/devices/joystick-parport.rst
803 ddebug_query= [KNL,DYNAMIC_DEBUG] Enable debug messages at early boot
805 Documentation/admin-guide/dynamic-debug-howto.rst for
806 details. Deprecated, see dyndbg.
808 debug [KNL] Enable kernel debugging (events log level).
811 [KNL] Enable printing [hashed] pointers early in the
812 boot sequence. If enabled, we use a weak hash instead
813 of siphash to hash pointers. Use this option if you are
814 seeing instances of '(___ptrval___)') and need to see a
815 value (hashed pointer) instead. Cryptographically
816 insecure, please do not use on production kernels.
819 [KNL] verbose locking self-tests
821 Print debugging info while doing the locking API
823 Bitmask for the various LOCKTYPE_ tests. Defaults to 0
824 (no extra messages), setting it to -1 (all bits set)
825 will print _a_lot_ more information - normally only
826 useful to lockdep developers.
828 debug_objects [KNL] Enable object debugging
831 [KNL] Disable object debugging
833 debug_guardpage_minorder=
834 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this
835 parameter allows control of the order of pages that will
836 be intentionally kept free (and hence protected) by the
837 buddy allocator. Bigger value increase the probability
838 of catching random memory corruption, but reduce the
839 amount of memory for normal system use. The maximum
840 possible value is MAX_ORDER/2. Setting this parameter
841 to 1 or 2 should be enough to identify most random
842 memory corruption problems caused by bugs in kernel or
843 driver code when a CPU writes to (or reads from) a
844 random memory location. Note that there exists a class
845 of memory corruptions problems caused by buggy H/W or
846 F/W or by drivers badly programing DMA (basically when
847 memory is written at bus level and the CPU MMU is
848 bypassed) which are not detectable by
849 CONFIG_DEBUG_PAGEALLOC, hence this option will not help
850 tracking down these problems.
853 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this parameter
854 enables the feature at boot time. By default, it is
855 disabled and the system will work mostly the same as a
856 kernel built without CONFIG_DEBUG_PAGEALLOC.
857 Note: to get most of debug_pagealloc error reports, it's
858 useful to also enable the page_owner functionality.
859 on: enable the feature
861 debugfs= [KNL] This parameter enables what is exposed to userspace
862 and debugfs internal clients.
863 Format: { on, no-mount, off }
864 on: All functions are enabled.
866 Filesystem is not registered but kernel clients can
867 access APIs and a crashkernel can be used to read
868 its content. There is nothing to mount.
869 off: Filesystem is not registered and clients
870 get a -EPERM as result when trying to register files
871 or directories within debugfs.
872 This is equivalent of the runtime functionality if
873 debugfs was not enabled in the kernel at all.
874 Default value is set in build-time with a kernel configuration.
876 debugpat [X86] Enable PAT debugging
878 decnet.addr= [HW,NET]
879 Format: <area>[,<node>]
880 See also Documentation/networking/decnet.rst.
883 [HW] The size of the default HugeTLB page. This is
884 the size represented by the legacy /proc/ hugepages
885 APIs. In addition, this is the default hugetlb size
886 used for shmget(), mmap() and mounting hugetlbfs
887 filesystems. If not specified, defaults to the
888 architecture's default huge page size. Huge page
889 sizes are architecture dependent. See also
890 Documentation/admin-guide/mm/hugetlbpage.rst.
893 deferred_probe_timeout=
894 [KNL] Debugging option to set a timeout in seconds for
895 deferred probe to give up waiting on dependencies to
896 probe. Only specific dependencies (subsystems or
897 drivers) that have opted in will be ignored. A timeout of 0
898 will timeout at the end of initcalls. This option will also
899 dump out devices still on the deferred probe list after
903 Format: { on | off | def_only | inf_only | always }
904 on: s390 zlib hardware support for compression on
905 level 1 and decompression (default)
906 off: No s390 zlib hardware support
907 def_only: s390 zlib hardware support for deflate
908 only (compression on level 1)
909 inf_only: s390 zlib hardware support for inflate
911 always: Same as 'on' but ignores the selected compression
912 level always using hardware support (used for debugging)
915 Set number of hash buckets for dentry cache.
917 disable_1tb_segments [PPC]
918 Disables the use of 1TB hash page table segments. This
919 causes the kernel to fall back to 256MB segments which
920 can be useful when debugging issues that require an SLB
924 Limits the number of kernel SLB entries, and flushes
925 them frequently to increase the rate of SLB faults
929 See Documentation/networking/ipv6.rst.
932 [KNL] Under CONFIG_HARDENED_USERCOPY, whether
933 hardening is enabled for this boot. Hardened
934 usercopy checking is used to protect the kernel
935 from reading or writing beyond known memory
936 allocation boundaries as a proactive defense
937 against bounds-checking flaws in the kernel's
938 copy_to_user()/copy_from_user() interface.
939 on Perform hardened usercopy checks (default).
940 off Disable hardened usercopy checks.
943 Disable RADIX MMU mode on POWER9
945 radix_hcall_invalidate=on [PPC/PSERIES]
946 Disable RADIX GTSE feature and use hcall for TLB
950 Disable TLBIE instruction. Currently does not work
951 with KVM, with HASH MMU, or with coherent accelerators.
953 disable_cpu_apicid= [X86,APIC,SMP]
955 The number of initial APIC ID for the
956 corresponding CPU to be disabled at boot,
957 mostly used for the kdump 2nd kernel to
958 disable BSP to wake up multiple CPUs without
959 causing system reset or hang due to sending
962 disable_ddw [PPC/PSERIES]
963 Disable Dynamic DMA Window support. Use this
964 to workaround buggy firmware.
967 See Documentation/networking/ipv6.rst.
969 disable_mtrr_cleanup [X86]
970 The kernel tries to adjust MTRR layout from continuous
971 to discrete, to make X server driver able to add WB
972 entry later. This parameter disables that.
974 disable_mtrr_trim [X86, Intel and AMD only]
975 By default the kernel will trim any uncacheable
976 memory out of your available memory pool based on
977 MTRR settings. This parameter disables that behavior,
978 possibly causing your machine to run very slowly.
980 disable_timer_pin_1 [X86]
981 Disable PIN 1 of APIC timer
982 Can be useful to work around chipset bugs.
984 dis_ucode_ldr [X86] Disable the microcode loader.
986 dma_debug=off If the kernel is compiled with DMA_API_DEBUG support,
987 this option disables the debugging code at boot.
989 dma_debug_entries=<number>
990 This option allows to tune the number of preallocated
991 entries for DMA-API debugging code. One entry is
992 required per DMA-API allocation. Use this if the
993 DMA-API debugging code disables itself because the
994 architectural default is too low.
996 dma_debug_driver=<driver_name>
997 With this option the DMA-API debugging driver
998 filter feature can be enabled at boot time. Just
999 pass the driver to filter for as the parameter.
1000 The filter can be disabled or changed to another
1001 driver later using sysfs.
1003 driver_async_probe= [KNL]
1004 List of driver names to be probed asynchronously.
1005 Format: <driver_name1>,<driver_name2>...
1007 drm.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>]
1008 Broken monitors, graphic adapters, KVMs and EDIDless
1009 panels may send no or incorrect EDID data sets.
1010 This parameter allows to specify an EDID data sets
1011 in the /lib/firmware directory that are used instead.
1012 Generic built-in EDID data sets are used, if one of
1013 edid/1024x768.bin, edid/1280x1024.bin,
1014 edid/1680x1050.bin, or edid/1920x1080.bin is given
1015 and no file with the same name exists. Details and
1016 instructions how to build your own EDID data are
1017 available in Documentation/admin-guide/edid.rst. An EDID
1018 data set will only be used for a particular connector,
1019 if its name and a colon are prepended to the EDID
1020 name. Each connector may use a unique EDID data
1021 set by separating the files with a comma. An EDID
1022 data set with no connector name will be used for
1023 any connectors not explicitly specified.
1028 Format: {"off" | "known"}
1029 Control how the dt_cpu_ftrs device-tree binding is
1030 used for CPU feature discovery and setup (if it
1032 off: Do not use it, fall back to legacy cpu table.
1033 known: Do not pass through unknown features to guests
1034 or userspace, only those that the kernel is aware of.
1036 dump_apple_properties [X86]
1037 Dump name and content of EFI device properties on
1038 x86 Macs. Useful for driver authors to determine
1039 what data is available or for reverse-engineering.
1041 dyndbg[="val"] [KNL,DYNAMIC_DEBUG]
1042 <module>.dyndbg[="val"]
1043 Enable debug messages at boot time. See
1044 Documentation/admin-guide/dynamic-debug-howto.rst
1047 nopku [X86] Disable Memory Protection Keys CPU feature found
1050 <module>.async_probe [KNL]
1051 Enable asynchronous probe on this module.
1053 early_ioremap_debug [KNL]
1054 Enable debug messages in early_ioremap support. This
1055 is useful for tracking down temporary early mappings
1056 which are not unmapped.
1058 earlycon= [KNL] Output early console device and options.
1060 When used with no options, the early console is
1061 determined by stdout-path property in device tree's
1062 chosen node or the ACPI SPCR table if supported by
1065 cdns,<addr>[,options]
1066 Start an early, polled-mode console on a Cadence
1067 (xuartps) serial port at the specified address. Only
1068 supported option is baud rate. If baud rate is not
1069 specified, the serial port must already be setup and
1072 uart[8250],io,<addr>[,options]
1073 uart[8250],mmio,<addr>[,options]
1074 uart[8250],mmio32,<addr>[,options]
1075 uart[8250],mmio32be,<addr>[,options]
1076 uart[8250],0x<addr>[,options]
1077 Start an early, polled-mode console on the 8250/16550
1078 UART at the specified I/O port or MMIO address.
1079 MMIO inter-register address stride is either 8-bit
1080 (mmio) or 32-bit (mmio32 or mmio32be).
1081 If none of [io|mmio|mmio32|mmio32be], <addr> is assumed
1082 to be equivalent to 'mmio'. 'options' are specified
1083 in the same format described for "console=ttyS<n>"; if
1084 unspecified, the h/w is not initialized.
1088 Start an early, polled-mode console on a pl011 serial
1089 port at the specified address. The pl011 serial port
1090 must already be setup and configured. Options are not
1091 yet supported. If 'mmio32' is specified, then only
1092 the driver will use only 32-bit accessors to read/write
1093 the device registers.
1096 Start an early, polled-mode console on a meson serial
1097 port at the specified address. The serial port must
1098 already be setup and configured. Options are not yet
1102 Start an early, polled-mode console on an msm serial
1103 port at the specified address. The serial port
1104 must already be setup and configured. Options are not
1107 msm_serial_dm,<addr>
1108 Start an early, polled-mode console on an msm serial
1109 dm port at the specified address. The serial port
1110 must already be setup and configured. Options are not
1114 Start an early, polled-mode console on a serial port
1115 of an Actions Semi SoC, such as S500 or S900, at the
1116 specified address. The serial port must already be
1117 setup and configured. Options are not yet supported.
1120 Start an early, polled-mode console on a serial port
1121 of an RDA Micro SoC, such as RDA8810PL, at the
1122 specified address. The serial port must already be
1123 setup and configured. Options are not yet supported.
1126 Use RISC-V SBI (Supervisor Binary Interface) for early
1129 smh Use ARM semihosting calls for early console.
1137 Use early console provided by serial driver available
1138 on Samsung SoCs, requires selecting proper type and
1139 a correct base address of the selected UART port. The
1140 serial port must already be setup and configured.
1141 Options are not yet supported.
1144 Start an early, polled-mode console on a lantiq serial
1145 (lqasc) port at the specified address. The serial port
1146 must already be setup and configured. Options are not
1151 Use early console provided by Freescale LP UART driver
1152 found on Freescale Vybrid and QorIQ LS1021A processors.
1153 A valid base address must be provided, and the serial
1154 port must already be setup and configured.
1158 Start an early, polled-mode, output-only console on the
1159 Freescale i.MX UART at the specified address. The UART
1160 must already be setup and configured.
1163 Start an early, polled-mode console on the
1164 Armada 3700 serial port at the specified
1165 address. The serial port must already be setup
1166 and configured. Options are not yet supported.
1169 Start an early, polled-mode console on a Qualcomm
1170 Generic Interface (GENI) based serial port at the
1171 specified address. The serial port must already be
1172 setup and configured. Options are not yet supported.
1175 Start an early, unaccelerated console on the EFI
1176 memory mapped framebuffer (if available). On cache
1177 coherent non-x86 systems that use system memory for
1178 the framebuffer, pass the 'ram' option so that it is
1179 mapped with the correct attributes.
1182 Use early console provided by Freescale LINFlexD UART
1183 serial driver for NXP S32V234 SoCs. A valid base
1184 address must be provided, and the serial port must
1185 already be setup and configured.
1187 earlyprintk= [X86,SH,ARM,M68k,S390]
1191 earlyprintk=serial[,ttySn[,baudrate]]
1192 earlyprintk=serial[,0x...[,baudrate]]
1193 earlyprintk=ttySn[,baudrate]
1194 earlyprintk=dbgp[debugController#]
1195 earlyprintk=pciserial[,force],bus:device.function[,baudrate]
1196 earlyprintk=xdbc[xhciController#]
1198 earlyprintk is useful when the kernel crashes before
1199 the normal console is initialized. It is not enabled by
1200 default because it has some cosmetic problems.
1202 Append ",keep" to not disable it when the real console
1205 Only one of vga, efi, serial, or usb debug port can
1208 Currently only ttyS0 and ttyS1 may be specified by
1209 name. Other I/O ports may be explicitly specified
1210 on some architectures (x86 and arm at least) by
1211 replacing ttySn with an I/O port address, like this:
1212 earlyprintk=serial,0x1008,115200
1213 You can find the port for a given device in
1214 /proc/tty/driver/serial:
1215 2: uart:ST16650V2 port:00001008 irq:18 ...
1217 Interaction with the standard serial driver is not
1220 The VGA and EFI output is eventually overwritten by
1223 The xen output can only be used by Xen PV guests.
1225 The sclp output can only be used on s390.
1227 The optional "force" to "pciserial" enables use of a
1228 PCI device even when its classcode is not of the
1231 edac_report= [HW,EDAC] Control how to report EDAC event
1232 Format: {"on" | "off" | "force"}
1233 on: enable EDAC to report H/W event. May be overridden
1234 by other higher priority error reporting module.
1235 off: disable H/W event reporting through EDAC.
1236 force: enforce the use of EDAC to report H/W event.
1239 ekgdboc= [X86,KGDB] Allow early kernel console debugging
1242 This is designed to be used in conjunction with
1243 the boot argument: earlyprintk=vga
1245 This parameter works in place of the kgdboc parameter
1246 but can only be used if the backing tty is available
1247 very early in the boot process. For early debugging
1248 via a serial port see kgdboc_earlycon instead.
1251 Format: {"off" | "on" | "skip[mbr]"}
1254 Format: { "debug", "disable_early_pci_dma",
1255 "nochunk", "noruntime", "nosoftreserve",
1256 "novamap", "no_disable_early_pci_dma" }
1257 debug: enable misc debug output.
1258 disable_early_pci_dma: disable the busmaster bit on all
1259 PCI bridges while in the EFI boot stub.
1260 nochunk: disable reading files in "chunks" in the EFI
1261 boot stub, as chunking can cause problems with some
1262 firmware implementations.
1263 noruntime : disable EFI runtime services support
1264 nosoftreserve: The EFI_MEMORY_SP (Specific Purpose)
1265 attribute may cause the kernel to reserve the
1266 memory range for a memory mapping driver to
1267 claim. Specify efi=nosoftreserve to disable this
1268 reservation and treat the memory by its base type
1269 (i.e. EFI_CONVENTIONAL_MEMORY / "System RAM").
1270 novamap: do not call SetVirtualAddressMap().
1271 no_disable_early_pci_dma: Leave the busmaster bit set
1272 on all PCI bridges while in the EFI boot stub
1274 efi_no_storage_paranoia [EFI; X86]
1275 Using this parameter you can use more than 50% of
1276 your efi variable storage. Use this parameter only if
1277 you are really sure that your UEFI does sane gc and
1278 fulfills the spec otherwise your board may brick.
1280 efi_fake_mem= nn[KMG]@ss[KMG]:aa[,nn[KMG]@ss[KMG]:aa,..] [EFI; X86]
1281 Add arbitrary attribute to specific memory range by
1282 updating original EFI memory map.
1283 Region of memory which aa attribute is added to is
1286 If efi_fake_mem=2G@4G:0x10000,2G@0x10a0000000:0x10000
1287 is specified, EFI_MEMORY_MORE_RELIABLE(0x10000)
1288 attribute is added to range 0x100000000-0x180000000 and
1289 0x10a0000000-0x1120000000.
1291 If efi_fake_mem=8G@9G:0x40000 is specified, the
1292 EFI_MEMORY_SP(0x40000) attribute is added to
1293 range 0x240000000-0x43fffffff.
1295 Using this parameter you can do debugging of EFI memmap
1296 related features. For example, you can do debugging of
1297 Address Range Mirroring feature even if your box
1298 doesn't support it, or mark specific memory as
1301 efivar_ssdt= [EFI; X86] Name of an EFI variable that contains an SSDT
1302 that is to be dynamically loaded by Linux. If there are
1303 multiple variables with the same name but with different
1304 vendor GUIDs, all of them will be loaded. See
1305 Documentation/admin-guide/acpi/ssdt-overlays.rst for details.
1308 eisa_irq_edge= [PARISC,HW]
1309 See header of drivers/parisc/eisa.c.
1312 See comment before function elanfreq_setup() in
1313 arch/x86/kernel/cpu/cpufreq/elanfreq.c.
1315 elfcorehdr=[size[KMG]@]offset[KMG] [IA64,PPC,SH,X86,S390]
1316 Specifies physical address of start of kernel core
1317 image elf header and optionally the size. Generally
1318 kexec loader will pass this option to capture kernel.
1319 See Documentation/admin-guide/kdump/kdump.rst for details.
1321 enable_mtrr_cleanup [X86]
1322 The kernel tries to adjust MTRR layout from continuous
1323 to discrete, to make X server driver able to add WB
1324 entry later. This parameter enables that.
1326 enable_timer_pin_1 [X86]
1327 Enable PIN 1 of APIC timer
1328 Can be useful to work around chipset bugs
1329 (in particular on some ATI chipsets).
1330 The kernel tries to set a reasonable default.
1332 enforcing [SELINUX] Set initial enforcing status.
1334 See security/selinux/Kconfig help text.
1335 0 -- permissive (log only, no denials).
1336 1 -- enforcing (deny and log).
1338 Value can be changed at runtime via
1339 /sys/fs/selinux/enforce.
1342 Disable Error Record Serialization Table (ERST)
1345 ether= [HW,NET] Ethernet cards parameters
1346 This option is obsoleted by the "netdev=" option, which
1347 has equivalent usage. See its documentation for details.
1351 Permit 'security.evm' to be updated regardless of
1352 current integrity status.
1357 fail_make_request=[KNL]
1358 General fault injection mechanism.
1359 Format: <interval>,<probability>,<space>,<times>
1360 See also Documentation/fault-injection/.
1363 Format: { initns | none }
1364 See Documentation/admin-guide/sysctl/net.rst for
1365 fb_tunnels_only_for_init_ns
1368 See Documentation/admin-guide/blockdev/floppy.rst.
1370 force_pal_cache_flush
1371 [IA-64] Avoid check_sal_cache_flush which may hang on
1372 buggy SAL_CACHE_FLUSH implementations. Using this
1373 parameter will force ia64_sal_cache_flush to call
1374 ia64_pal_cache_flush instead of SAL_CACHE_FLUSH.
1377 Forcefully enable Physical Address Extension (PAE).
1378 Many Pentium M systems disable PAE but may have a
1379 functionally usable PAE implementation.
1380 Warning: use of this parameter will taint the kernel
1381 and may cause unknown problems.
1384 [FTRACE] will set and start the specified tracer
1385 as early as possible in order to facilitate early
1388 ftrace_dump_on_oops[=orig_cpu]
1389 [FTRACE] will dump the trace buffers on oops.
1390 If no parameter is passed, ftrace will dump
1391 buffers of all CPUs, but if you pass orig_cpu, it will
1392 dump only the buffer of the CPU that triggered the
1395 ftrace_filter=[function-list]
1396 [FTRACE] Limit the functions traced by the function
1397 tracer at boot up. function-list is a comma-separated
1398 list of functions. This list can be changed at run
1399 time by the set_ftrace_filter file in the debugfs
1402 ftrace_notrace=[function-list]
1403 [FTRACE] Do not trace the functions specified in
1404 function-list. This list can be changed at run time
1405 by the set_ftrace_notrace file in the debugfs
1408 ftrace_graph_filter=[function-list]
1409 [FTRACE] Limit the top level callers functions traced
1410 by the function graph tracer at boot up.
1411 function-list is a comma-separated list of functions
1412 that can be changed at run time by the
1413 set_graph_function file in the debugfs tracing directory.
1415 ftrace_graph_notrace=[function-list]
1416 [FTRACE] Do not trace from the functions specified in
1417 function-list. This list is a comma-separated list of
1418 functions that can be changed at run time by the
1419 set_graph_notrace file in the debugfs tracing directory.
1421 ftrace_graph_max_depth=<uint>
1422 [FTRACE] Used with the function graph tracer. This is
1423 the max depth it will trace into a function. This value
1424 can be changed at run time by the max_graph_depth file
1425 in the tracefs tracing directory. default: 0 (no limit)
1427 fw_devlink= [KNL] Create device links between consumer and supplier
1428 devices by scanning the firmware to infer the
1429 consumer/supplier relationships. This feature is
1430 especially useful when drivers are loaded as modules as
1431 it ensures proper ordering of tasks like device probing
1432 (suppliers first, then consumers), supplier boot state
1433 clean up (only after all consumers have probed),
1434 suspend/resume & runtime PM (consumers first, then
1436 Format: { off | permissive | on | rpm }
1437 off -- Don't create device links from firmware info.
1438 permissive -- Create device links from firmware info
1439 but use it only for ordering boot state clean
1440 up (sync_state() calls).
1441 on -- Create device links from firmware info and use it
1442 to enforce probe and suspend/resume ordering.
1443 rpm -- Like "on", but also use to order runtime PM.
1445 fw_devlink.strict=<bool>
1446 [KNL] Treat all inferred dependencies as mandatory
1447 dependencies. This only applies for fw_devlink=on|rpm.
1451 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad
1452 support via parallel port (up to 5 devices per port)
1453 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5>
1454 See also Documentation/input/devices/joystick-parport.rst
1458 gart_fix_e820= [X86-64] disable the fix e820 for K8 GART
1462 gcov_persist= [GCOV] When non-zero (default), profiling data for
1463 kernel modules is saved and remains accessible via
1464 debugfs, even when the module is unloaded/reloaded.
1465 When zero, profiling data is discarded and associated
1466 debugfs files are removed at module unload time.
1468 goldfish [X86] Enable the goldfish android emulator platform.
1469 Don't use this when you are not running on the
1472 gpt [EFI] Forces disk with valid GPT signature but
1473 invalid Protective MBR to be treated as GPT. If the
1474 primary GPT is corrupted, it enables the backup/alternate
1475 GPT to be used instead.
1477 grcan.enable0= [HW] Configuration of physical interface 0. Determines
1478 the "Enable 0" bit of the configuration register.
1481 grcan.enable1= [HW] Configuration of physical interface 1. Determines
1482 the "Enable 0" bit of the configuration register.
1485 grcan.select= [HW] Select which physical interface to use.
1488 grcan.txsize= [HW] Sets the size of the tx buffer.
1489 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0.
1491 grcan.rxsize= [HW] Sets the size of the rx buffer.
1492 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0.
1495 gpio-mockup.gpio_mockup_ranges
1496 [HW] Sets the ranges of gpiochip of for this device.
1497 Format: <start1>,<end1>,<start2>,<end2>...
1499 hardlockup_all_cpu_backtrace=
1500 [KNL] Should the hard-lockup detector generate
1501 backtraces on all cpus.
1504 hashdist= [KNL,NUMA] Large hashes allocated during boot
1505 are distributed across NUMA nodes. Defaults on
1506 for 64-bit NUMA, off otherwise.
1507 Format: 0 | 1 (for off | on)
1509 hcl= [IA-64] SGI's Hardware Graph compatibility layer
1511 hd= [EIDE] (E)IDE hard drive subsystem geometry
1512 Format: <cyl>,<head>,<sect>
1515 Disable Hardware Error Source Table (HEST) support;
1516 corresponding firmware-first mode error processing
1517 logic will be disabled.
1519 highmem=nn[KMG] [KNL,BOOT] forces the highmem zone to have an exact
1520 size of <nn>. This works even on boxes that have no
1521 highmem otherwise. This also works to reduce highmem
1522 size on bigger boxes.
1524 highres= [KNL] Enable/disable high resolution timer mode.
1525 Valid parameters: "on", "off"
1530 hpet= [X86-32,HPET] option to control HPET usage
1531 Format: { enable (default) | disable | force |
1533 disable: disable HPET and use PIT instead
1534 force: allow force enabled of undocumented chips (ICH4,
1536 verbose: show contents of HPET registers during setup
1538 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET
1539 registers. Default set by CONFIG_HPET_MMAP_DEFAULT.
1541 hugetlb_cma= [HW,CMA] The size of a CMA area used for allocation
1542 of gigantic hugepages.
1545 Reserve a CMA area of given size and allocate gigantic
1546 hugepages using the CMA allocator. If enabled, the
1547 boot-time allocation of gigantic hugepages is skipped.
1549 hugepages= [HW] Number of HugeTLB pages to allocate at boot.
1550 If this follows hugepagesz (below), it specifies
1551 the number of pages of hugepagesz to be allocated.
1552 If this is the first HugeTLB parameter on the command
1553 line, it specifies the number of pages to allocate for
1554 the default huge page size. See also
1555 Documentation/admin-guide/mm/hugetlbpage.rst.
1559 [HW] The size of the HugeTLB pages. This is used in
1560 conjunction with hugepages (above) to allocate huge
1561 pages of a specific size at boot. The pair
1562 hugepagesz=X hugepages=Y can be specified once for
1563 each supported huge page size. Huge page sizes are
1564 architecture dependent. See also
1565 Documentation/admin-guide/mm/hugetlbpage.rst.
1569 [KNL] Should the hung task detector generate panics.
1572 A value of 1 instructs the kernel to panic when a
1573 hung task is detected. The default value is controlled
1574 by the CONFIG_BOOTPARAM_HUNG_TASK_PANIC build-time
1575 option. The value selected by this boot parameter can
1576 be changed later by the kernel.hung_task_panic sysctl.
1578 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC)
1579 terminal devices. Valid values: 0..8
1580 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs.
1581 If specified, z/VM IUCV HVC accepts connections
1582 from listed z/VM user IDs only.
1584 hv_nopvspin [X86,HYPER_V] Disables the paravirt spinlock optimizations
1585 which allow the hypervisor to 'idle' the
1586 guest on lock contention.
1589 Do not unregister boot console at start. This is only
1590 useful for debugging when something happens in the window
1591 between unregistering the boot console and initializing
1594 i2c_bus= [HW] Override the default board specific I2C bus speed
1595 or register an additional I2C bus that is not
1596 registered from board initialization code.
1600 i8042.debug [HW] Toggle i8042 debug mode
1601 i8042.unmask_kbd_data
1602 [HW] Enable printing of interrupt data from the KBD port
1603 (disabled by default, and as a pre-condition
1604 requires that i8042.debug=1 be enabled)
1605 i8042.direct [HW] Put keyboard port into non-translated mode
1606 i8042.dumbkbd [HW] Pretend that controller can only read data from
1607 keyboard and cannot control its state
1608 (Don't attempt to blink the leds)
1609 i8042.noaux [HW] Don't check for auxiliary (== mouse) port
1610 i8042.nokbd [HW] Don't check/create keyboard port
1611 i8042.noloop [HW] Disable the AUX Loopback command while probing
1613 i8042.nomux [HW] Don't check presence of an active multiplexing
1615 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX
1617 i8042.notimeout [HW] Ignore timeout condition signalled by controller
1618 i8042.reset [HW] Reset the controller during init, cleanup and
1619 suspend-to-ram transitions, only during s2r
1620 transitions, or never reset
1621 Format: { 1 | Y | y | 0 | N | n }
1622 1, Y, y: always reset controller
1623 0, N, n: don't ever reset controller
1624 Default: only on s2r transitions on x86; most other
1625 architectures force reset to be always executed
1626 i8042.unlock [HW] Unlock (ignore) the keylock
1627 i8042.kbdreset [HW] Reset device connected to KBD port
1631 i8k.ignore_dmi [HW] Continue probing hardware even if DMI data
1632 indicates that the driver is running on unsupported
1634 i8k.force [HW] Activate i8k driver even if SMM BIOS signature
1635 does not match list of supported models.
1637 [HW] Report power status in /proc/i8k
1638 (disabled by default)
1639 i8k.restricted [HW] Allow controlling fans only if SYS_ADMIN
1642 i915.invert_brightness=
1643 [DRM] Invert the sense of the variable that is used to
1644 set the brightness of the panel backlight. Normally a
1645 brightness value of 0 indicates backlight switched off,
1646 and the maximum of the brightness value sets the backlight
1647 to maximum brightness. If this parameter is set to 0
1648 (default) and the machine requires it, or this parameter
1649 is set to 1, a brightness value of 0 sets the backlight
1650 to maximum brightness, and the maximum of the brightness
1651 value switches the backlight off.
1652 -1 -- never invert brightness
1653 0 -- machine default
1654 1 -- force brightness inversion
1657 Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]]
1659 ide-core.nodma= [HW] (E)IDE subsystem
1660 Format: =0.0 to prevent dma on hda, =0.1 hdb =1.0 hdc
1661 .vlb_clock .pci_clock .noflush .nohpa .noprobe .nowerr
1662 .cdrom .chs .ignore_cable are additional options
1663 See Documentation/ide/ide.rst.
1665 ide-generic.probe-mask= [HW] (E)IDE subsystem
1667 Probe mask for legacy ISA IDE ports. Depending on
1668 platform up to 6 ports are supported, enabled by
1669 setting corresponding bits in the mask to 1. The
1670 default value is 0x0, which has a special meaning.
1671 On systems that have PCI, it triggers scanning the
1672 PCI bus for the first and the second port, which
1673 are then probed. On systems without PCI the value
1674 of 0x0 enables probing the two first ports as if it
1677 ide-pci-generic.all-generic-ide [HW] (E)IDE subsystem
1678 Claim all unknown PCI IDE storage controllers.
1681 Format: idle=poll, idle=halt, idle=nomwait
1682 Poll forces a polling idle loop that can slightly
1683 improve the performance of waking up a idle CPU, but
1684 will use a lot of power and make the system run hot.
1686 idle=halt: Halt is forced to be used for CPU idle.
1687 In such case C2/C3 won't be used again.
1688 idle=nomwait: Disable mwait for CPU C-states
1692 Allow force disabling of Shared Virtual Memory (SVA)
1693 support for the idxd driver. By default it is set to
1696 ieee754= [MIPS] Select IEEE Std 754 conformance mode
1697 Format: { strict | legacy | 2008 | relaxed }
1700 Choose which programs will be accepted for execution
1701 based on the IEEE 754 NaN encoding(s) supported by
1702 the FPU and the NaN encoding requested with the value
1703 of an ELF file header flag individually set by each
1704 binary. Hardware implementations are permitted to
1705 support either or both of the legacy and the 2008 NaN
1708 Available settings are as follows:
1709 strict accept binaries that request a NaN encoding
1710 supported by the FPU
1711 legacy only accept legacy-NaN binaries, if supported
1713 2008 only accept 2008-NaN binaries, if supported
1715 relaxed accept any binaries regardless of whether
1716 supported by the FPU
1718 The FPU emulator is always able to support both NaN
1719 encodings, so if no FPU hardware is present or it has
1720 been disabled with 'nofpu', then the settings of
1721 'legacy' and '2008' strap the emulator accordingly,
1722 'relaxed' straps the emulator for both legacy-NaN and
1723 2008-NaN, whereas 'strict' enables legacy-NaN only on
1724 legacy processors and both NaN encodings on MIPS32 or
1727 The setting for ABS.fmt/NEG.fmt instruction execution
1728 mode generally follows that for the NaN encoding,
1729 except where unsupported by hardware.
1731 ignore_loglevel [KNL]
1732 Ignore loglevel setting - this will print /all/
1733 kernel messages to the console. Useful for debugging.
1734 We also add it as printk module parameter, so users
1735 could change it dynamically, usually by
1736 /sys/module/printk/parameters/ignore_loglevel.
1739 Ignore RLIMIT_DATA setting for data mappings,
1740 print warning at first misuse. Can be changed via
1741 /sys/module/kernel/parameters/ignore_rlimit_data.
1743 ihash_entries= [KNL]
1744 Set number of hash buckets for inode cache.
1746 ima_appraise= [IMA] appraise integrity measurements
1747 Format: { "off" | "enforce" | "fix" | "log" }
1750 ima_appraise_tcb [IMA] Deprecated. Use ima_policy= instead.
1751 The builtin appraise policy appraises all files
1754 ima_canonical_fmt [IMA]
1755 Use the canonical format for the binary runtime
1756 measurements, instead of host native format.
1759 Format: { md5 | sha1 | rmd160 | sha256 | sha384
1763 The list of supported hash algorithms is defined
1764 in crypto/hash_info.h.
1767 The builtin policies to load during IMA setup.
1768 Format: "tcb | appraise_tcb | secure_boot |
1769 fail_securely | critical_data"
1771 The "tcb" policy measures all programs exec'd, files
1772 mmap'd for exec, and all files opened with the read
1773 mode bit set by either the effective uid (euid=0) or
1776 The "appraise_tcb" policy appraises the integrity of
1777 all files owned by root.
1779 The "secure_boot" policy appraises the integrity
1780 of files (eg. kexec kernel image, kernel modules,
1781 firmware, policy, etc) based on file signatures.
1783 The "fail_securely" policy forces file signature
1784 verification failure also on privileged mounted
1785 filesystems with the SB_I_UNVERIFIABLE_SIGNATURE
1788 The "critical_data" policy measures kernel integrity
1791 ima_tcb [IMA] Deprecated. Use ima_policy= instead.
1792 Load a policy which meets the needs of the Trusted
1793 Computing Base. This means IMA will measure all
1794 programs exec'd, files mmap'd for exec, and all files
1795 opened for read by uid=0.
1798 Select one of defined IMA measurements template formats.
1799 Formats: { "ima" | "ima-ng" | "ima-sig" }
1803 [IMA] Define a custom template format.
1804 Format: { "field1|...|fieldN" }
1806 ima.ahash_minsize= [IMA] Minimum file size for asynchronous hash usage
1807 Format: <min_file_size>
1808 Set the minimal file size for using asynchronous hash.
1809 If left unspecified, ahash usage is disabled.
1811 ahash performance varies for different data sizes on
1812 different crypto accelerators. This option can be used
1813 to achieve the best performance for a particular HW.
1815 ima.ahash_bufsize= [IMA] Asynchronous hash buffer size
1817 Set hashing buffer size. Default: 4k.
1819 ahash performance varies for different chunk sizes on
1820 different crypto accelerators. This option can be used
1821 to achieve best performance for particular HW.
1825 Run specified binary instead of /sbin/init as init
1828 initcall_debug [KNL] Trace initcalls as they are executed. Useful
1829 for working out where the kernel is dying during
1832 initcall_blacklist= [KNL] Do not execute a comma-separated list of
1833 initcall functions. Useful for debugging built-in
1834 modules and initcalls.
1836 initrd= [BOOT] Specify the location of the initial ramdisk
1838 initrdmem= [KNL] Specify a physical address and size from which to
1839 load the initrd. If an initrd is compiled in or
1840 specified in the bootparams, it takes priority over this
1842 Format: ss[KMG],nn[KMG]
1845 init_on_alloc= [MM] Fill newly allocated pages and heap objects with
1848 Default set by CONFIG_INIT_ON_ALLOC_DEFAULT_ON.
1850 init_on_free= [MM] Fill freed pages and heap objects with zeroes.
1852 Default set by CONFIG_INIT_ON_FREE_DEFAULT_ON.
1854 init_pkru= [X86] Specify the default memory protection keys rights
1855 register contents for all processes. 0x55555554 by
1856 default (disallow access to all but pkey 0). Can
1857 override in debugfs after boot.
1859 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver
1862 int_pln_enable [X86] Enable power limit notification interrupt
1864 integrity_audit=[IMA]
1865 Format: { "0" | "1" }
1866 0 -- basic integrity auditing messages. (Default)
1867 1 -- additional integrity auditing messages.
1869 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option
1871 Enable intel iommu driver.
1873 Disable intel iommu driver.
1874 igfx_off [Default Off]
1875 By default, gfx is mapped as normal device. If a gfx
1876 device has a dedicated DMAR unit, the DMAR unit is
1877 bypassed by not enabling DMAR with this option. In
1878 this case, gfx device will use physical address for
1880 strict [Default Off]
1881 With this option on every unmap_single operation will
1882 result in a hardware IOTLB flush operation as opposed
1883 to batching them for performance.
1884 sp_off [Default Off]
1885 By default, super page will be supported if Intel IOMMU
1886 has the capability. With this option, super page will
1889 By default, scalable mode will be disabled even if the
1890 hardware advertises that it has support for the scalable
1891 mode translation. With this option set, scalable mode
1892 will be used on hardware which claims to support it.
1893 tboot_noforce [Default Off]
1894 Do not force the Intel IOMMU enabled under tboot.
1895 By default, tboot will force Intel IOMMU on, which
1896 could harm performance of some high-throughput
1897 devices like 40GBit network cards, even if identity
1899 Note that using this option lowers the security
1900 provided by tboot because it makes the system
1901 vulnerable to DMA attacks.
1903 intel_idle.max_cstate= [KNL,HW,ACPI,X86]
1904 0 disables intel_idle and fall back on acpi_idle.
1905 1 to 9 specify maximum depth of C-state.
1909 Do not enable intel_pstate as the default
1910 scaling driver for the supported processors
1912 Use intel_pstate as a scaling driver, but configure it
1913 to work with generic cpufreq governors (instead of
1914 enabling its internal governor). This mode cannot be
1915 used along with the hardware-managed P-states (HWP)
1918 Enable intel_pstate on systems that prohibit it by default
1919 in favor of acpi-cpufreq. Forcing the intel_pstate driver
1920 instead of acpi-cpufreq may disable platform features, such
1921 as thermal controls and power capping, that rely on ACPI
1922 P-States information being indicated to OSPM and therefore
1923 should be used with caution. This option does not work with
1924 processors that aren't supported by the intel_pstate driver
1925 or on platforms that use pcc-cpufreq instead of acpi-cpufreq.
1927 Do not enable hardware P state control (HWP)
1930 Only load intel_pstate on systems which support
1931 hardware P state control (HWP) if available.
1933 Enforce ACPI _PPC performance limits. If the Fixed ACPI
1934 Description Table, specifies preferred power management
1935 profile as "Enterprise Server" or "Performance Server",
1936 then this feature is turned on by default.
1938 Allow per-logical-CPU P-State performance control limits using
1939 cpufreq sysfs interface
1941 intremap= [X86-64, Intel-IOMMU]
1942 on enable Interrupt Remapping (default)
1943 off disable Interrupt Remapping
1944 nosid disable Source ID checking
1946 BIOS x2APIC opt-out request will be ignored
1947 nopost disable Interrupt Posting
1949 iomem= Disable strict checking of access to MMIO memory
1950 strict regions from userspace.
1965 nobypass [PPC/POWERNV]
1966 Disable IOMMU bypass, using IOMMU for PCI devices.
1968 iommu.forcedac= [ARM64, X86] Control IOVA allocation for PCI devices.
1969 Format: { "0" | "1" }
1970 0 - Try to allocate a 32-bit DMA address first, before
1971 falling back to the full range if needed.
1972 1 - Allocate directly from the full usable range,
1973 forcing Dual Address Cycle for PCI cards supporting
1974 greater than 32-bit addressing.
1976 iommu.strict= [ARM64] Configure TLB invalidation behaviour
1977 Format: { "0" | "1" }
1979 Request that DMA unmap operations use deferred
1980 invalidation of hardware TLBs, for increased
1981 throughput at the cost of reduced device isolation.
1982 Will fall back to strict mode if not supported by
1983 the relevant IOMMU driver.
1984 1 - Strict mode (default).
1985 DMA unmap operations invalidate IOMMU hardware TLBs
1989 [ARM64, X86] Configure DMA to bypass the IOMMU by default.
1990 Format: { "0" | "1" }
1991 0 - Use IOMMU translation for DMA.
1992 1 - Bypass the IOMMU for DMA.
1993 unset - Use value of CONFIG_IOMMU_DEFAULT_PASSTHROUGH.
1995 io7= [HW] IO7 for Marvel-based Alpha systems
1996 See comment before marvel_specify_io7 in
1997 arch/alpha/kernel/core_marvel.c.
1999 io_delay= [X86] I/O delay method
2001 Standard port 0x80 based delay
2003 Alternate port 0xed based delay (needed on some systems)
2005 Simple two microseconds delay
2010 See Documentation/admin-guide/nfs/nfsroot.rst.
2012 ipcmni_extend [KNL] Extend the maximum number of unique System V
2013 IPC identifiers from 32,768 to 16,777,216.
2015 irqaffinity= [SMP] Set the default irq affinity mask
2016 The argument is a cpu list, as described above.
2018 irqchip.gicv2_force_probe=
2021 Force the kernel to look for the second 4kB page
2022 of a GICv2 controller even if the memory range
2023 exposed by the device tree is too small.
2025 irqchip.gicv3_nolpi=
2027 Force the kernel to ignore the availability of
2028 LPIs (and by consequence ITSs). Intended for system
2029 that use the kernel as a bootloader, and thus want
2030 to let secondary kernels in charge of setting up
2033 irqchip.gicv3_pseudo_nmi= [ARM64]
2034 Enables support for pseudo-NMIs in the kernel. This
2035 requires the kernel to be built with
2036 CONFIG_ARM64_PSEUDO_NMI.
2039 When an interrupt is not handled search all handlers
2040 for it. Intended to get systems with badly broken
2044 When an interrupt is not handled search all handlers
2045 for it. Also check all handlers each timer
2046 interrupt. Intended to get systems with badly broken
2050 Format: <RDP>,<reset>,<pci_scan>,<verbosity>
2052 isolcpus= [KNL,SMP,ISOL] Isolate a given set of CPUs from disturbance.
2053 [Deprecated - use cpusets instead]
2054 Format: [flag-list,]<cpu-list>
2056 Specify one or more CPUs to isolate from disturbances
2057 specified in the flag list (default: domain):
2060 Disable the tick when a single task runs.
2062 A residual 1Hz tick is offloaded to workqueues, which you
2063 need to affine to housekeeping through the global
2064 workqueue's affinity configured via the
2065 /sys/devices/virtual/workqueue/cpumask sysfs file, or
2066 by using the 'domain' flag described below.
2068 NOTE: by default the global workqueue runs on all CPUs,
2069 so to protect individual CPUs the 'cpumask' file has to
2070 be configured manually after bootup.
2073 Isolate from the general SMP balancing and scheduling
2074 algorithms. Note that performing domain isolation this way
2075 is irreversible: it's not possible to bring back a CPU to
2076 the domains once isolated through isolcpus. It's strongly
2077 advised to use cpusets instead to disable scheduler load
2078 balancing through the "cpuset.sched_load_balance" file.
2079 It offers a much more flexible interface where CPUs can
2080 move in and out of an isolated set anytime.
2082 You can move a process onto or off an "isolated" CPU via
2083 the CPU affinity syscalls or cpuset.
2084 <cpu number> begins at 0 and the maximum value is
2085 "number of CPUs in system - 1".
2089 Isolate from being targeted by managed interrupts
2090 which have an interrupt mask containing isolated
2091 CPUs. The affinity of managed interrupts is
2092 handled by the kernel and cannot be changed via
2093 the /proc/irq/* interfaces.
2095 This isolation is best effort and only effective
2096 if the automatically assigned interrupt mask of a
2097 device queue contains isolated and housekeeping
2098 CPUs. If housekeeping CPUs are online then such
2099 interrupts are directed to the housekeeping CPU
2100 so that IO submitted on the housekeeping CPU
2101 cannot disturb the isolated CPU.
2103 If a queue's affinity mask contains only isolated
2104 CPUs then this parameter has no effect on the
2105 interrupt routing decision, though interrupts are
2106 only delivered when tasks running on those
2107 isolated CPUs submit IO. IO submitted on
2108 housekeeping CPUs has no influence on those
2111 The format of <cpu-list> is described above.
2115 ivrs_ioapic [HW,X86-64]
2116 Provide an override to the IOAPIC-ID<->DEVICE-ID
2117 mapping provided in the IVRS ACPI table. For
2118 example, to map IOAPIC-ID decimal 10 to
2119 PCI device 00:14.0 write the parameter as:
2120 ivrs_ioapic[10]=00:14.0
2122 ivrs_hpet [HW,X86-64]
2123 Provide an override to the HPET-ID<->DEVICE-ID
2124 mapping provided in the IVRS ACPI table. For
2125 example, to map HPET-ID decimal 0 to
2126 PCI device 00:14.0 write the parameter as:
2127 ivrs_hpet[0]=00:14.0
2129 ivrs_acpihid [HW,X86-64]
2130 Provide an override to the ACPI-HID:UID<->DEVICE-ID
2131 mapping provided in the IVRS ACPI table. For
2132 example, to map UART-HID:UID AMD0020:0 to
2133 PCI device 00:14.5 write the parameter as:
2134 ivrs_acpihid[00:14.5]=AMD0020:0
2136 js= [HW,JOY] Analog joystick
2137 See Documentation/input/joydev/joystick.rst.
2140 When CONFIG_RANDOMIZE_BASE is set, this disables
2141 kernel and module base offset ASLR (Address Space
2142 Layout Randomization).
2145 [KNL] Enforce KASAN (Kernel Address Sanitizer) to print
2146 report on every invalid memory access. Without this
2147 parameter KASAN will print report only for the first
2152 kernelcore= [KNL,X86,IA-64,PPC]
2153 Format: nn[KMGTPE] | nn% | "mirror"
2154 This parameter specifies the amount of memory usable by
2155 the kernel for non-movable allocations. The requested
2156 amount is spread evenly throughout all nodes in the
2157 system as ZONE_NORMAL. The remaining memory is used for
2158 movable memory in its own zone, ZONE_MOVABLE. In the
2159 event, a node is too small to have both ZONE_NORMAL and
2160 ZONE_MOVABLE, kernelcore memory will take priority and
2161 other nodes will have a larger ZONE_MOVABLE.
2163 ZONE_MOVABLE is used for the allocation of pages that
2164 may be reclaimed or moved by the page migration
2165 subsystem. Note that allocations like PTEs-from-HighMem
2166 still use the HighMem zone if it exists, and the Normal
2167 zone if it does not.
2169 It is possible to specify the exact amount of memory in
2170 the form of "nn[KMGTPE]", a percentage of total system
2171 memory in the form of "nn%", or "mirror". If "mirror"
2172 option is specified, mirrored (reliable) memory is used
2173 for non-movable allocations and remaining memory is used
2174 for Movable pages. "nn[KMGTPE]", "nn%", and "mirror"
2175 are exclusive, so you cannot specify multiple forms.
2177 kgdbdbgp= [KGDB,HW] kgdb over EHCI usb debug port.
2178 Format: <Controller#>[,poll interval]
2179 The controller # is the number of the ehci usb debug
2180 port as it is probed via PCI. The poll interval is
2181 optional and is the number seconds in between
2182 each poll cycle to the debug port in case you need
2183 the functionality for interrupting the kernel with
2184 gdb or control-c on the dbgp connection. When
2185 not using this parameter you use sysrq-g to break into
2186 the kernel debugger.
2188 kgdboc= [KGDB,HW] kgdb over consoles.
2189 Requires a tty driver that supports console polling,
2190 or a supported polling keyboard driver (non-usb).
2191 Serial only format: <serial_device>[,baud]
2192 keyboard only format: kbd
2193 keyboard and serial format: kbd,<serial_device>[,baud]
2194 Optional Kernel mode setting:
2195 kms, kbd format: kms,kbd
2196 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud]
2198 kgdboc_earlycon= [KGDB,HW]
2199 If the boot console provides the ability to read
2200 characters and can work in polling mode, you can use
2201 this parameter to tell kgdb to use it as a backend
2202 until the normal console is registered. Intended to
2203 be used together with the kgdboc parameter which
2204 specifies the normal console to transition to.
2206 The name of the early console should be specified
2207 as the value of this parameter. Note that the name of
2208 the early console might be different than the tty
2209 name passed to kgdboc. It's OK to leave the value
2210 blank and the first boot console that implements
2211 read() will be picked.
2213 kgdbwait [KGDB] Stop kernel execution and enter the
2214 kernel debugger at the earliest opportunity.
2216 kmac= [MIPS] Korina ethernet MAC address.
2217 Configure the RouterBoard 532 series on-chip
2218 Ethernet adapter MAC address.
2220 kmemleak= [KNL] Boot-time kmemleak enable/disable
2221 Valid arguments: on, off
2223 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y,
2226 kprobe_event=[probe-list]
2227 [FTRACE] Add kprobe events and enable at boot time.
2228 The probe-list is a semicolon delimited list of probe
2229 definitions. Each definition is same as kprobe_events
2230 interface, but the parameters are comma delimited.
2231 For example, to add a kprobe event on vfs_read with
2232 arg1 and arg2, add to the command line;
2234 kprobe_event=p,vfs_read,$arg1,$arg2
2236 See also Documentation/trace/kprobetrace.rst "Kernel
2237 Boot Parameter" section.
2239 kpti= [ARM64] Control page table isolation of user
2240 and kernel address spaces.
2241 Default: enabled on cores which need mitigation.
2245 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs.
2246 Default is 0 (don't ignore, but inject #GP)
2248 kvm.enable_vmware_backdoor=[KVM] Support VMware backdoor PV interface.
2249 Default is false (don't support).
2251 kvm.mmu_audit= [KVM] This is a R/W parameter which allows audit
2256 [KVM] Controls the software workaround for the
2257 X86_BUG_ITLB_MULTIHIT bug.
2258 force : Always deploy workaround.
2259 off : Never deploy workaround.
2260 auto : Deploy workaround based on the presence of
2261 X86_BUG_ITLB_MULTIHIT.
2265 If the software workaround is enabled for the host,
2266 guests do need not to enable it for nested guests.
2268 kvm.nx_huge_pages_recovery_ratio=
2269 [KVM] Controls how many 4KiB pages are periodically zapped
2270 back to huge pages. 0 disables the recovery, otherwise if
2271 the value is N KVM will zap 1/Nth of the 4KiB pages every
2272 minute. The default is 60.
2274 kvm-amd.nested= [KVM,AMD] Allow nested virtualization in KVM/SVM.
2275 Default is 1 (enabled)
2277 kvm-amd.npt= [KVM,AMD] Disable nested paging (virtualized MMU)
2279 Default is 1 (enabled) if in 64-bit or 32-bit PAE mode.
2282 [KVM,ARM] Select one of KVM/arm64's modes of operation.
2284 nvhe: Standard nVHE-based mode, without support for
2287 protected: nVHE-based mode with support for guests whose
2288 state is kept private from the host.
2289 Not valid if the kernel is running in EL2.
2291 Defaults to VHE/nVHE based on hardware support.
2293 kvm-arm.vgic_v3_group0_trap=
2294 [KVM,ARM] Trap guest accesses to GICv3 group-0
2297 kvm-arm.vgic_v3_group1_trap=
2298 [KVM,ARM] Trap guest accesses to GICv3 group-1
2301 kvm-arm.vgic_v3_common_trap=
2302 [KVM,ARM] Trap guest accesses to GICv3 common
2305 kvm-arm.vgic_v4_enable=
2306 [KVM,ARM] Allow use of GICv4 for direct injection of
2309 kvm_cma_resv_ratio=n [PPC]
2310 Reserves given percentage from system memory area for
2311 contiguous memory allocation for KVM hash pagetable
2313 By default it reserves 5% of total system memory.
2317 kvm-intel.ept= [KVM,Intel] Disable extended page tables
2318 (virtualized MMU) support on capable Intel chips.
2319 Default is 1 (enabled)
2321 kvm-intel.emulate_invalid_guest_state=
2322 [KVM,Intel] Enable emulation of invalid guest states
2323 Default is 0 (disabled)
2325 kvm-intel.flexpriority=
2326 [KVM,Intel] Disable FlexPriority feature (TPR shadow).
2327 Default is 1 (enabled)
2330 [KVM,Intel] Enable VMX nesting (nVMX).
2331 Default is 0 (disabled)
2333 kvm-intel.unrestricted_guest=
2334 [KVM,Intel] Disable unrestricted guest feature
2335 (virtualized real and unpaged mode) on capable
2336 Intel chips. Default is 1 (enabled)
2338 kvm-intel.vmentry_l1d_flush=[KVM,Intel] Mitigation for L1 Terminal Fault
2341 Valid arguments: never, cond, always
2343 always: L1D cache flush on every VMENTER.
2344 cond: Flush L1D on VMENTER only when the code between
2345 VMEXIT and VMENTER can leak host memory.
2346 never: Disables the mitigation
2348 Default is cond (do L1 cache flush in specific instances)
2350 kvm-intel.vpid= [KVM,Intel] Disable Virtual Processor Identification
2351 feature (tagged TLBs) on capable Intel chips.
2352 Default is 1 (enabled)
2354 l1tf= [X86] Control mitigation of the L1TF vulnerability on
2357 The kernel PTE inversion protection is unconditionally
2358 enabled and cannot be disabled.
2361 Provides all available mitigations for the
2362 L1TF vulnerability. Disables SMT and
2363 enables all mitigations in the
2364 hypervisors, i.e. unconditional L1D flush.
2366 SMT control and L1D flush control via the
2367 sysfs interface is still possible after
2368 boot. Hypervisors will issue a warning
2369 when the first VM is started in a
2370 potentially insecure configuration,
2371 i.e. SMT enabled or L1D flush disabled.
2374 Same as 'full', but disables SMT and L1D
2375 flush runtime control. Implies the
2376 'nosmt=force' command line option.
2377 (i.e. sysfs control of SMT is disabled.)
2380 Leaves SMT enabled and enables the default
2381 hypervisor mitigation, i.e. conditional
2384 SMT control and L1D flush control via the
2385 sysfs interface is still possible after
2386 boot. Hypervisors will issue a warning
2387 when the first VM is started in a
2388 potentially insecure configuration,
2389 i.e. SMT enabled or L1D flush disabled.
2393 Disables SMT and enables the default
2394 hypervisor mitigation.
2396 SMT control and L1D flush control via the
2397 sysfs interface is still possible after
2398 boot. Hypervisors will issue a warning
2399 when the first VM is started in a
2400 potentially insecure configuration,
2401 i.e. SMT enabled or L1D flush disabled.
2404 Same as 'flush', but hypervisors will not
2405 warn when a VM is started in a potentially
2406 insecure configuration.
2409 Disables hypervisor mitigations and doesn't
2411 It also drops the swap size and available
2412 RAM limit restriction on both hypervisor and
2417 For details see: Documentation/admin-guide/hw-vuln/l1tf.rst
2423 lapic [X86-32,APIC] Enable the local APIC even if BIOS
2426 lapic= [X86,APIC] Do not use TSC deadline
2427 value for LAPIC timer one-shot implementation. Default
2428 back to the programmable timer unit in the LAPIC.
2429 Format: notscdeadline
2431 lapic_timer_c2_ok [X86,APIC] trust the local apic timer
2434 libata.dma= [LIBATA] DMA control
2435 libata.dma=0 Disable all PATA and SATA DMA
2436 libata.dma=1 PATA and SATA Disk DMA only
2437 libata.dma=2 ATAPI (CDROM) DMA only
2438 libata.dma=4 Compact Flash DMA only
2439 Combinations also work, so libata.dma=3 enables DMA
2440 for disks and CDROMs, but not CFs.
2442 libata.ignore_hpa= [LIBATA] Ignore HPA limit
2443 libata.ignore_hpa=0 keep BIOS limits (default)
2444 libata.ignore_hpa=1 ignore limits, using full disk
2446 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume
2450 libata.force= [LIBATA] Force configurations. The format is comma-
2451 separated list of "[ID:]VAL" where ID is
2452 PORT[.DEVICE]. PORT and DEVICE are decimal numbers
2453 matching port, link or device. Basically, it matches
2454 the ATA ID string printed on console by libata. If
2455 the whole ID part is omitted, the last PORT and DEVICE
2456 values are used. If ID hasn't been specified yet, the
2457 configuration applies to all ports, links and devices.
2459 If only DEVICE is omitted, the parameter applies to
2460 the port and all links and devices behind it. DEVICE
2461 number of 0 either selects the first device or the
2462 first fan-out link behind PMP device. It does not
2463 select the host link. DEVICE number of 15 selects the
2464 host link and device attached to it.
2466 The VAL specifies the configuration to force. As long
2467 as there's no ambiguity shortcut notation is allowed.
2468 For example, both 1.5 and 1.5G would work for 1.5Gbps.
2469 The following configurations can be forced.
2471 * Cable type: 40c, 80c, short40c, unk, ign or sata.
2472 Any ID with matching PORT is used.
2474 * SATA link speed limit: 1.5Gbps or 3.0Gbps.
2476 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7].
2477 udma[/][16,25,33,44,66,100,133] notation is also
2480 * [no]ncq: Turn on or off NCQ.
2482 * [no]ncqtrim: Turn off queued DSM TRIM.
2484 * nohrst, nosrst, norst: suppress hard, soft
2487 * rstonce: only attempt one reset during
2488 hot-unplug link recovery
2490 * dump_id: dump IDENTIFY data.
2492 * atapi_dmadir: Enable ATAPI DMADIR bridge support
2494 * disable: Disable this device.
2496 If there are multiple matching configurations changing
2497 the same attribute, the last one is used.
2499 memblock=debug [KNL] Enable memblock debug messages.
2501 load_ramdisk= [RAM] [Deprecated]
2503 lockd.nlm_grace_period=P [NFS] Assign grace period.
2506 lockd.nlm_tcpport=N [NFS] Assign TCP port.
2509 lockd.nlm_timeout=T [NFS] Assign timeout value.
2512 lockd.nlm_udpport=M [NFS] Assign UDP port.
2515 lockdown= [SECURITY]
2516 { integrity | confidentiality }
2517 Enable the kernel lockdown feature. If set to
2518 integrity, kernel features that allow userland to
2519 modify the running kernel are disabled. If set to
2520 confidentiality, kernel features that allow userland
2521 to extract confidential information from the kernel
2524 locktorture.nreaders_stress= [KNL]
2525 Set the number of locking read-acquisition kthreads.
2526 Defaults to being automatically set based on the
2527 number of online CPUs.
2529 locktorture.nwriters_stress= [KNL]
2530 Set the number of locking write-acquisition kthreads.
2532 locktorture.onoff_holdoff= [KNL]
2533 Set time (s) after boot for CPU-hotplug testing.
2535 locktorture.onoff_interval= [KNL]
2536 Set time (s) between CPU-hotplug operations, or
2537 zero to disable CPU-hotplug testing.
2539 locktorture.shuffle_interval= [KNL]
2540 Set task-shuffle interval (jiffies). Shuffling
2541 tasks allows some CPUs to go into dyntick-idle
2542 mode during the locktorture test.
2544 locktorture.shutdown_secs= [KNL]
2545 Set time (s) after boot system shutdown. This
2546 is useful for hands-off automated testing.
2548 locktorture.stat_interval= [KNL]
2549 Time (s) between statistics printk()s.
2551 locktorture.stutter= [KNL]
2552 Time (s) to stutter testing, for example,
2553 specifying five seconds causes the test to run for
2554 five seconds, wait for five seconds, and so on.
2555 This tests the locking primitive's ability to
2556 transition abruptly to and from idle.
2558 locktorture.torture_type= [KNL]
2559 Specify the locking implementation to test.
2561 locktorture.verbose= [KNL]
2562 Enable additional printk() statements.
2564 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver
2567 loglevel= All Kernel Messages with a loglevel smaller than the
2568 console loglevel will be printed to the console. It can
2569 also be changed with klogd or other programs. The
2570 loglevels are defined as follows:
2572 0 (KERN_EMERG) system is unusable
2573 1 (KERN_ALERT) action must be taken immediately
2574 2 (KERN_CRIT) critical conditions
2575 3 (KERN_ERR) error conditions
2576 4 (KERN_WARNING) warning conditions
2577 5 (KERN_NOTICE) normal but significant condition
2578 6 (KERN_INFO) informational
2579 7 (KERN_DEBUG) debug-level messages
2581 log_buf_len=n[KMG] Sets the size of the printk ring buffer,
2582 in bytes. n must be a power of two and greater
2583 than the minimal size. The minimal size is defined
2584 by LOG_BUF_SHIFT kernel config parameter. There is
2585 also CONFIG_LOG_CPU_MAX_BUF_SHIFT config parameter
2586 that allows to increase the default size depending on
2587 the number of CPUs. See init/Kconfig for more details.
2589 logo.nologo [FB] Disables display of the built-in Linux logo.
2590 This may be used to provide more screen space for
2591 kernel log messages and is useful when debugging
2592 kernel boot problems.
2594 lp=0 [LP] Specify parallel ports to use, e.g,
2595 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses
2596 lp=reset first parallel port). 'lp=0' disables the
2597 lp=auto printer driver. 'lp=reset' (which can be
2598 specified in addition to the ports) causes
2599 attached printers to be reset. Using
2600 lp=port1,port2,... specifies the parallel ports
2601 to associate lp devices with, starting with
2602 lp0. A port specification may be 'none' to skip
2603 that lp device, or a parport name such as
2604 'parport0'. Specifying 'lp=auto' instead of a
2605 port specification list means that device IDs
2606 from each port should be examined, to see if
2607 an IEEE 1284-compliant printer is attached; if
2608 so, the driver will manage that printer.
2609 See also header of drivers/char/lp.c.
2612 Sets loops_per_jiffy to given constant, thus avoiding
2613 time-consuming boot-time autodetection (up to 250 ms per
2614 CPU). 0 enables autodetection (default). To determine
2615 the correct value for your kernel, boot with normal
2616 autodetection and see what value is printed. Note that
2617 on SMP systems the preset will be applied to all CPUs,
2618 which is likely to cause problems if your CPUs need
2619 significantly divergent settings. An incorrect value
2620 will cause delays in the kernel to be wrong, leading to
2621 unpredictable I/O errors and other breakage. Although
2622 unlikely, in the extreme case this might damage your
2626 Format: <io>,<irq>,<dma>
2628 lsm.debug [SECURITY] Enable LSM initialization debugging output.
2631 [SECURITY] Choose order of LSM initialization. This
2632 overrides CONFIG_LSM, and the "security=" parameter.
2634 machvec= [IA-64] Force the use of a particular machine-vector
2635 (machvec) in a generic kernel.
2636 Example: machvec=hpzx1
2638 machtype= [Loongson] Share the same kernel image file between
2639 different yeeloong laptops.
2640 Example: machtype=lemote-yeeloong-2f-7inch
2642 max_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory greater
2643 than or equal to this physical address is ignored.
2645 maxcpus= [SMP] Maximum number of processors that an SMP kernel
2646 will bring up during bootup. maxcpus=n : n >= 0 limits
2647 the kernel to bring up 'n' processors. Surely after
2648 bootup you can bring up the other plugged cpu by executing
2649 "echo 1 > /sys/devices/system/cpu/cpuX/online". So maxcpus
2650 only takes effect during system bootup.
2651 While n=0 is a special case, it is equivalent to "nosmp",
2652 which also disables the IO APIC.
2654 max_loop= [LOOP] The number of loop block devices that get
2655 (loop.max_loop) unconditionally pre-created at init time. The default
2656 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead
2657 of statically allocating a predefined number, loop
2658 devices can be requested on-demand with the
2659 /dev/loop-control interface.
2661 mce [X86-32] Machine Check Exception
2663 mce=option [X86-64] See Documentation/x86/x86_64/boot-options.rst
2665 md= [HW] RAID subsystems devices and level
2666 See Documentation/admin-guide/md.rst.
2669 Format: <first>,<last>
2670 Specifies range of consoles to be captured by the MDA.
2673 Control mitigation for the Micro-architectural Data
2674 Sampling (MDS) vulnerability.
2676 Certain CPUs are vulnerable to an exploit against CPU
2677 internal buffers which can forward information to a
2678 disclosure gadget under certain conditions.
2680 In vulnerable processors, the speculatively
2681 forwarded data can be used in a cache side channel
2682 attack, to access data to which the attacker does
2683 not have direct access.
2685 This parameter controls the MDS mitigation. The
2688 full - Enable MDS mitigation on vulnerable CPUs
2689 full,nosmt - Enable MDS mitigation and disable
2690 SMT on vulnerable CPUs
2691 off - Unconditionally disable MDS mitigation
2693 On TAA-affected machines, mds=off can be prevented by
2694 an active TAA mitigation as both vulnerabilities are
2695 mitigated with the same mechanism so in order to disable
2696 this mitigation, you need to specify tsx_async_abort=off
2699 Not specifying this option is equivalent to
2702 For details see: Documentation/admin-guide/hw-vuln/mds.rst
2704 mem=nn[KMG] [KNL,BOOT] Force usage of a specific amount of memory
2705 Amount of memory to be used in cases as follows:
2708 2 when the kernel is not able to see the whole system memory;
2709 3 memory that lies after 'mem=' boundary is excluded from
2710 the hypervisor, then assigned to KVM guests.
2712 [X86] Work as limiting max address. Use together
2713 with memmap= to avoid physical address space collisions.
2714 Without memmap= PCI devices could be placed at addresses
2715 belonging to unused RAM.
2717 Note that this only takes effects during boot time since
2718 in above case 3, memory may need be hot added after boot
2719 if system memory of hypervisor is not sufficient.
2721 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel
2725 [KNL,SH] Allow user to override the default size for
2726 per-device physically contiguous DMA buffers.
2728 memhp_default_state=online/offline
2729 [KNL] Set the initial state for the memory hotplug
2730 onlining policy. If not specified, the default value is
2731 set according to the
2732 CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE kernel config
2734 See Documentation/admin-guide/mm/memory-hotplug.rst.
2736 memmap=exactmap [KNL,X86] Enable setting of an exact
2737 E820 memory map, as specified by the user.
2738 Such memmap=exactmap lines can be constructed based on
2739 BIOS output or other requirements. See the memmap=nn@ss
2742 memmap=nn[KMG]@ss[KMG]
2743 [KNL, X86, MIPS, XTENSA] Force usage of a specific region of memory.
2744 Region of memory to be used is from ss to ss+nn.
2745 If @ss[KMG] is omitted, it is equivalent to mem=nn[KMG],
2746 which limits max address to nn[KMG].
2747 Multiple different regions can be specified,
2750 memmap=100M@2G,100M#3G,1G!1024G
2752 memmap=nn[KMG]#ss[KMG]
2753 [KNL,ACPI] Mark specific memory as ACPI data.
2754 Region of memory to be marked is from ss to ss+nn.
2756 memmap=nn[KMG]$ss[KMG]
2757 [KNL,ACPI] Mark specific memory as reserved.
2758 Region of memory to be reserved is from ss to ss+nn.
2759 Example: Exclude memory from 0x18690000-0x1869ffff
2760 memmap=64K$0x18690000
2762 memmap=0x10000$0x18690000
2763 Some bootloaders may need an escape character before '$',
2764 like Grub2, otherwise '$' and the following number
2767 memmap=nn[KMG]!ss[KMG]
2768 [KNL,X86] Mark specific memory as protected.
2769 Region of memory to be used, from ss to ss+nn.
2770 The memory region may be marked as e820 type 12 (0xc)
2771 and is NVDIMM or ADR memory.
2773 memmap=<size>%<offset>-<oldtype>+<newtype>
2774 [KNL,ACPI] Convert memory within the specified region
2775 from <oldtype> to <newtype>. If "-<oldtype>" is left
2776 out, the whole region will be marked as <newtype>,
2777 even if previously unavailable. If "+<newtype>" is left
2778 out, matching memory will be removed. Types are
2779 specified as e820 types, e.g., 1 = RAM, 2 = reserved,
2780 3 = ACPI, 12 = PRAM.
2782 memory_corruption_check=0/1 [X86]
2783 Some BIOSes seem to corrupt the first 64k of
2784 memory when doing things like suspend/resume.
2785 Setting this option will scan the memory
2786 looking for corruption. Enabling this will
2787 both detect corruption and prevent the kernel
2788 from using the memory being corrupted.
2789 However, its intended as a diagnostic tool; if
2790 repeatable BIOS-originated corruption always
2791 affects the same memory, you can use memmap=
2792 to prevent the kernel from using that memory.
2794 memory_corruption_check_size=size [X86]
2795 By default it checks for corruption in the low
2796 64k, making this memory unavailable for normal
2797 use. Use this parameter to scan for
2798 corruption in more or less memory.
2800 memory_corruption_check_period=seconds [X86]
2801 By default it checks for corruption every 60
2802 seconds. Use this parameter to check at some
2803 other rate. 0 disables periodic checking.
2805 memtest= [KNL,X86,ARM,PPC] Enable memtest
2807 default : 0 <disable>
2808 Specifies the number of memtest passes to be
2809 performed. Each pass selects another test
2810 pattern from a given set of patterns. Memtest
2811 fills the memory with this pattern, validates
2812 memory contents and reserves bad memory
2813 regions that are detected.
2815 mem_encrypt= [X86-64] AMD Secure Memory Encryption (SME) control
2816 Valid arguments: on, off
2817 Default (depends on kernel configuration option):
2818 on (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=y)
2819 off (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=n)
2820 mem_encrypt=on: Activate SME
2821 mem_encrypt=off: Do not activate SME
2823 Refer to Documentation/virt/kvm/amd-memory-encryption.rst
2824 for details on when memory encryption can be activated.
2826 mem_sleep_default= [SUSPEND] Default system suspend mode:
2827 s2idle - Suspend-To-Idle
2828 shallow - Power-On Suspend or equivalent (if supported)
2829 deep - Suspend-To-RAM or equivalent (if supported)
2830 See Documentation/admin-guide/pm/sleep-states.rst.
2832 meye.*= [HW] Set MotionEye Camera parameters
2833 See Documentation/admin-guide/media/meye.rst.
2835 mfgpt_irq= [IA-32] Specify the IRQ to use for the
2836 Multi-Function General Purpose Timers on AMD Geode
2839 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when
2840 the BIOS has incorrectly applied a workaround. TinyBIOS
2841 version 0.98 is known to be affected, 0.99 fixes the
2842 problem by letting the user disable the workaround.
2846 min_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory below this
2847 physical address is ignored.
2849 mini2440= [ARM,HW,KNL]
2850 Format:[0..2][b][c][t]
2852 MINI2440 configuration specification:
2853 0 - The attached screen is the 3.5" TFT
2854 1 - The attached screen is the 7" TFT
2855 2 - The VGA Shield is attached (1024x768)
2856 Leaving out the screen size parameter will not load
2857 the TFT driver, and the framebuffer will be left
2859 b - Enable backlight. The TFT backlight pin will be
2860 linked to the kernel VESA blanking code and a GPIO
2861 LED. This parameter is not necessary when using the
2863 c - Enable the s3c camera interface.
2864 t - Reserved for enabling touchscreen support. The
2865 touchscreen support is not enabled in the mainstream
2866 kernel as of 2.6.30, a preliminary port can be found
2867 in the "bleeding edge" mini2440 support kernel at
2868 https://repo.or.cz/w/linux-2.6/mini2440.git
2871 [X86,PPC,S390,ARM64] Control optional mitigations for
2872 CPU vulnerabilities. This is a set of curated,
2873 arch-independent options, each of which is an
2874 aggregation of existing arch-specific options.
2877 Disable all optional CPU mitigations. This
2878 improves system performance, but it may also
2879 expose users to several CPU vulnerabilities.
2880 Equivalent to: nopti [X86,PPC]
2882 nospectre_v1 [X86,PPC]
2884 nospectre_v2 [X86,PPC,S390,ARM64]
2885 spectre_v2_user=off [X86]
2886 spec_store_bypass_disable=off [X86,PPC]
2887 ssbd=force-off [ARM64]
2890 tsx_async_abort=off [X86]
2891 kvm.nx_huge_pages=off [X86]
2892 no_entry_flush [PPC]
2893 no_uaccess_flush [PPC]
2896 This does not have any effect on
2897 kvm.nx_huge_pages when
2898 kvm.nx_huge_pages=force.
2901 Mitigate all CPU vulnerabilities, but leave SMT
2902 enabled, even if it's vulnerable. This is for
2903 users who don't want to be surprised by SMT
2904 getting disabled across kernel upgrades, or who
2905 have other ways of avoiding SMT-based attacks.
2906 Equivalent to: (default behavior)
2909 Mitigate all CPU vulnerabilities, disabling SMT
2910 if needed. This is for users who always want to
2911 be fully mitigated, even if it means losing SMT.
2912 Equivalent to: l1tf=flush,nosmt [X86]
2913 mds=full,nosmt [X86]
2914 tsx_async_abort=full,nosmt [X86]
2917 [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this
2918 parameter allows control of the logging verbosity for
2919 the additional memory initialisation checks. A value
2920 of 0 disables mminit logging and a level of 4 will
2921 log everything. Information is printed at KERN_DEBUG
2922 so loglevel=8 may also need to be specified.
2925 [KNL] When CONFIG_MODULE_SIG is set, this means that
2926 modules without (valid) signatures will fail to load.
2927 Note that if CONFIG_MODULE_SIG_FORCE is set, that
2928 is always true, so this option does nothing.
2930 module_blacklist= [KNL] Do not load a comma-separated list of
2931 modules. Useful for debugging problem modules.
2934 [MOUSE] Maximum time between finger touching and
2935 leaving touchpad surface for touch to be considered
2936 a tap and be reported as a left button click (for
2937 touchpads working in absolute mode only).
2939 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices
2940 reporting absolute coordinates, such as tablets
2941 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices
2942 reporting absolute coordinates, such as tablets
2944 movablecore= [KNL,X86,IA-64,PPC]
2945 Format: nn[KMGTPE] | nn%
2946 This parameter is the complement to kernelcore=, it
2947 specifies the amount of memory used for migratable
2948 allocations. If both kernelcore and movablecore is
2949 specified, then kernelcore will be at *least* the
2950 specified value but may be more. If movablecore on its
2951 own is specified, the administrator must be careful
2952 that the amount of memory usable for all allocations
2955 movable_node [KNL] Boot-time switch to make hotplugable memory
2956 NUMA nodes to be movable. This means that the memory
2957 of such nodes will be usable only for movable
2958 allocations which rules out almost all kernel
2959 allocations. Use with caution!
2961 MTD_Partition= [MTD]
2962 Format: <name>,<region-number>,<size>,<offset>
2964 MTD_Region= [MTD] Format:
2965 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>]
2968 See drivers/mtd/parsers/cmdlinepart.c
2970 multitce=off [PPC] This parameter disables the use of the pSeries
2971 firmware feature for updating multiple TCE entries
2974 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration
2976 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock]
2978 boundary - index of last SLC block on Flex-OneNAND.
2979 The remaining blocks are configured as MLC blocks.
2980 lock - Configure if Flex-OneNAND boundary should be locked.
2981 Once locked, the boundary cannot be changed.
2982 1 indicates lock status, 0 indicates unlock status.
2985 ARM/S3C2412 JIVE boot control
2987 See arch/arm/mach-s3c/mach-jive.c
2989 mtouchusb.raw_coordinates=
2990 [HW] Make the MicroTouch USB driver use raw coordinates
2991 ('y', default) or cooked coordinates ('n')
2993 mtrr_chunk_size=nn[KMG] [X86]
2994 used for mtrr cleanup. It is largest continuous chunk
2995 that could hold holes aka. UC entries.
2997 mtrr_gran_size=nn[KMG] [X86]
2998 Used for mtrr cleanup. It is granularity of mtrr block.
3000 Large value could prevent small alignment from
3003 mtrr_spare_reg_nr=n [X86]
3005 Range: 0,7 : spare reg number
3007 Used for mtrr cleanup. It is spare mtrr entries number.
3008 Set to 2 or more if your graphical card needs more.
3010 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card
3012 netdev= [NET] Network devices parameters
3013 Format: <irq>,<io>,<mem_start>,<mem_end>,<name>
3014 Note that mem_start is often overloaded to mean
3015 something different and driver-specific.
3016 This usage is only documented in each driver source
3020 [NETFILTER] Enable connection tracking flow accounting
3021 0 to disable accounting
3022 1 to enable accounting
3025 nfsaddrs= [NFS] Deprecated. Use ip= instead.
3026 See Documentation/admin-guide/nfs/nfsroot.rst.
3028 nfsroot= [NFS] nfs root filesystem for disk-less boxes.
3029 See Documentation/admin-guide/nfs/nfsroot.rst.
3031 nfsrootdebug [NFS] enable nfsroot debugging messages.
3032 See Documentation/admin-guide/nfs/nfsroot.rst.
3034 nfs.callback_nr_threads=
3035 [NFSv4] set the total number of threads that the
3036 NFS client will assign to service NFSv4 callback
3039 nfs.callback_tcpport=
3040 [NFS] set the TCP port on which the NFSv4 callback
3041 channel should listen.
3044 [NFS] sets the pathname to the program which is used
3045 to update the NFS client cache entries.
3047 nfs.cache_getent_timeout=
3048 [NFS] sets the timeout after which an attempt to
3049 update a cache entry is deemed to have failed.
3051 nfs.idmap_cache_timeout=
3052 [NFS] set the maximum lifetime for idmapper cache
3056 [NFS] enable 64-bit inode numbers.
3057 If zero, the NFS client will fake up a 32-bit inode
3058 number for the readdir() and stat() syscalls instead
3059 of returning the full 64-bit number.
3060 The default is to return 64-bit inode numbers.
3062 nfs.max_session_cb_slots=
3063 [NFSv4.1] Sets the maximum number of session
3064 slots the client will assign to the callback
3065 channel. This determines the maximum number of
3066 callbacks the client will process in parallel for
3067 a particular server.
3069 nfs.max_session_slots=
3070 [NFSv4.1] Sets the maximum number of session slots
3071 the client will attempt to negotiate with the server.
3072 This limits the number of simultaneous RPC requests
3073 that the client can send to the NFSv4.1 server.
3074 Note that there is little point in setting this
3075 value higher than the max_tcp_slot_table_limit.
3077 nfs.nfs4_disable_idmapping=
3078 [NFSv4] When set to the default of '1', this option
3079 ensures that both the RPC level authentication
3080 scheme and the NFS level operations agree to use
3081 numeric uids/gids if the mount is using the
3082 'sec=sys' security flavour. In effect it is
3083 disabling idmapping, which can make migration from
3084 legacy NFSv2/v3 systems to NFSv4 easier.
3085 Servers that do not support this mode of operation
3086 will be autodetected by the client, and it will fall
3087 back to using the idmapper.
3088 To turn off this behaviour, set the value to '0'.
3090 [NFS4] Specify an additional fixed unique ident-
3091 ification string that NFSv4 clients can insert into
3092 their nfs_client_id4 string. This is typically a
3093 UUID that is generated at system install time.
3095 nfs.send_implementation_id =
3096 [NFSv4.1] Send client implementation identification
3097 information in exchange_id requests.
3098 If zero, no implementation identification information
3100 The default is to send the implementation identification
3103 nfs.recover_lost_locks =
3104 [NFSv4] Attempt to recover locks that were lost due
3105 to a lease timeout on the server. Please note that
3106 doing this risks data corruption, since there are
3107 no guarantees that the file will remain unchanged
3108 after the locks are lost.
3109 If you want to enable the kernel legacy behaviour of
3110 attempting to recover these locks, then set this
3112 The default parameter value of '0' causes the kernel
3113 not to attempt recovery of lost locks.
3115 nfs4.layoutstats_timer =
3116 [NFSv4.2] Change the rate at which the kernel sends
3117 layoutstats to the pNFS metadata server.
3119 Setting this to value to 0 causes the kernel to use
3120 whatever value is the default set by the layout
3121 driver. A non-zero value sets the minimum interval
3122 in seconds between layoutstats transmissions.
3124 nfsd.nfs4_disable_idmapping=
3125 [NFSv4] When set to the default of '1', the NFSv4
3126 server will return only numeric uids and gids to
3127 clients using auth_sys, and will accept numeric uids
3128 and gids from such clients. This is intended to ease
3129 migration from NFSv2/v3.
3131 nmi_backtrace.backtrace_idle [KNL]
3132 Dump stacks even of idle CPUs in response to an
3133 NMI stack-backtrace request.
3135 nmi_debug= [KNL,SH] Specify one or more actions to take
3136 when a NMI is triggered.
3137 Format: [state][,regs][,debounce][,die]
3139 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels
3140 Format: [panic,][nopanic,][num]
3142 0 - turn hardlockup detector in nmi_watchdog off
3143 1 - turn hardlockup detector in nmi_watchdog on
3144 When panic is specified, panic when an NMI watchdog
3145 timeout occurs (or 'nopanic' to not panic on an NMI
3146 watchdog, if CONFIG_BOOTPARAM_HARDLOCKUP_PANIC is set)
3147 To disable both hard and soft lockup detectors,
3148 please see 'nowatchdog'.
3149 This is useful when you use a panic=... timeout and
3150 need the box quickly up again.
3152 These settings can be accessed at runtime via
3153 the nmi_watchdog and hardlockup_panic sysctls.
3155 netpoll.carrier_timeout=
3156 [NET] Specifies amount of time (in seconds) that
3157 netpoll should wait for a carrier. By default netpoll
3160 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths
3161 emulation library even if a 387 maths coprocessor
3164 no5lvl [X86-64] Disable 5-level paging mode. Forces
3165 kernel to use 4-level paging instead.
3167 nofsgsbase [X86] Disables FSGSBASE instructions.
3170 [HW] Never suspend the console
3171 Disable suspending of consoles during suspend and
3172 hibernate operations. Once disabled, debugging
3173 messages can reach various consoles while the rest
3174 of the system is being put to sleep (ie, while
3175 debugging driver suspend/resume hooks). This may
3176 not work reliably with all consoles, but is known
3177 to work with serial and VGA consoles.
3178 To facilitate more flexible debugging, we also add
3179 console_suspend, a printk module parameter to control
3180 it. Users could use console_suspend (usually
3181 /sys/module/printk/parameters/console_suspend) to
3182 turn on/off it dynamically.
3184 novmcoredd [KNL,KDUMP]
3185 Disable device dump. Device dump allows drivers to
3186 append dump data to vmcore so you can collect driver
3187 specified debug info. Drivers can append the data
3188 without any limit and this data is stored in memory,
3189 so this may cause significant memory stress. Disabling
3190 device dump can help save memory but the driver debug
3191 data will be no longer available. This parameter
3192 is only available when CONFIG_PROC_VMCORE_DEVICE_DUMP
3195 noaliencache [MM, NUMA, SLAB] Disables the allocation of alien
3196 caches in the slab allocator. Saves per-node memory,
3197 but will impact performance.
3201 noaltinstr [S390] Disables alternative instructions patching
3202 (CPU alternatives feature).
3204 noapic [SMP,APIC] Tells the kernel to not make use of any
3205 IOAPICs that may be present in the system.
3207 noautogroup Disable scheduler automatic task group creation.
3209 nobats [PPC] Do not use BATs for mapping kernel lowmem
3210 on "Classic" PPC cores.
3214 noclflush [BUGS=X86] Don't use the CLFLUSH instruction
3216 nodelayacct [KNL] Disable per-task delay accounting
3218 nodsp [SH] Disable hardware DSP at boot time.
3220 noefi Disable EFI runtime services support.
3222 no_entry_flush [PPC] Don't flush the L1-D cache when entering the kernel.
3227 On X86-32 available only on PAE configured kernels.
3228 noexec=on: enable non-executable mappings (default)
3229 noexec=off: disable non-executable mappings
3232 Disable SMAP (Supervisor Mode Access Prevention)
3233 even if it is supported by processor.
3236 Disable SMEP (Supervisor Mode Execution Prevention)
3237 even if it is supported by processor.
3240 This affects only 32-bit executables.
3241 noexec32=on: enable non-executable mappings (default)
3242 read doesn't imply executable mappings
3243 noexec32=off: disable non-executable mappings
3244 read implies executable mappings
3246 nofpu [MIPS,SH] Disable hardware FPU at boot time.
3248 nofxsr [BUGS=X86-32] Disables x86 floating point extended
3249 register save and restore. The kernel will only save
3250 legacy floating-point registers on task switch.
3252 nohugeiomap [KNL,X86,PPC,ARM64] Disable kernel huge I/O mappings.
3254 nosmt [KNL,S390] Disable symmetric multithreading (SMT).
3255 Equivalent to smt=1.
3257 [KNL,X86] Disable symmetric multithreading (SMT).
3258 nosmt=force: Force disable SMT, cannot be undone
3259 via the sysfs control file.
3261 nospectre_v1 [X86,PPC] Disable mitigations for Spectre Variant 1
3262 (bounds check bypass). With this option data leaks are
3263 possible in the system.
3265 nospectre_v2 [X86,PPC_FSL_BOOK3E,ARM64] Disable all mitigations for
3266 the Spectre variant 2 (indirect branch prediction)
3267 vulnerability. System may allow data leaks with this
3270 nospec_store_bypass_disable
3271 [HW] Disable all mitigations for the Speculative Store Bypass vulnerability
3274 [PPC] Don't flush the L1-D cache after accessing user data.
3276 noxsave [BUGS=X86] Disables x86 extended register state save
3277 and restore using xsave. The kernel will fallback to
3278 enabling legacy floating-point and sse state.
3280 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended
3281 register states. The kernel will fall back to use
3282 xsave to save the states. By using this parameter,
3283 performance of saving the states is degraded because
3284 xsave doesn't support modified optimization while
3285 xsaveopt supports it on xsaveopt enabled systems.
3287 noxsaves [X86] Disables xsaves and xrstors used in saving and
3288 restoring x86 extended register state in compacted
3289 form of xsave area. The kernel will fall back to use
3290 xsaveopt and xrstor to save and restore the states
3291 in standard form of xsave area. By using this
3292 parameter, xsave area per process might occupy more
3293 memory on xsaves enabled systems.
3295 nohlt [ARM,ARM64,MICROBLAZE,SH] Forces the kernel to busy wait
3296 in do_idle() and not use the arch_cpu_idle()
3297 implementation; requires CONFIG_GENERIC_IDLE_POLL_SETUP
3298 to be effective. This is useful on platforms where the
3299 sleep(SH) or wfi(ARM,ARM64) instructions do not work
3300 correctly or when doing power measurements to evalute
3301 the impact of the sleep instructions. This is also
3302 useful when using JTAG debugger.
3304 no_file_caps Tells the kernel not to honor file capabilities. The
3305 only way then for a file to be executed with privilege
3306 is to be setuid root or executed by root.
3308 nohalt [IA-64] Tells the kernel not to use the power saving
3309 function PAL_HALT_LIGHT when idle. This increases
3310 power-consumption. On the positive side, it reduces
3311 interrupt wake-up latency, which may improve performance
3312 in certain environments such as networked servers or
3316 Force pointers printed to the console or buffers to be
3317 unhashed. By default, when a pointer is printed via %p
3318 format string, that pointer is "hashed", i.e. obscured
3319 by hashing the pointer value. This is a security feature
3320 that hides actual kernel addresses from unprivileged
3321 users, but it also makes debugging the kernel more
3322 difficult since unequal pointers can no longer be
3323 compared. However, if this command-line option is
3324 specified, then all normal pointers will have their true
3325 value printed. Pointers printed via %pK may still be
3326 hashed. This option should only be specified when
3327 debugging the kernel. Please do not use on production
3330 nohibernate [HIBERNATION] Disable hibernation and resume.
3332 nohz= [KNL] Boottime enable/disable dynamic ticks
3333 Valid arguments: on, off
3336 nohz_full= [KNL,BOOT,SMP,ISOL]
3337 The argument is a cpu list, as described above.
3338 In kernels built with CONFIG_NO_HZ_FULL=y, set
3339 the specified list of CPUs whose tick will be stopped
3340 whenever possible. The boot CPU will be forced outside
3341 the range to maintain the timekeeping. Any CPUs
3342 in this list will have their RCU callbacks offloaded,
3343 just as if they had also been called out in the
3344 rcu_nocbs= boot parameter.
3346 noiotrap [SH] Disables trapped I/O port accesses.
3348 noirqdebug [X86-32] Disables the code which attempts to detect and
3349 disable unhandled interrupt sources.
3351 no_timer_check [X86,APIC] Disables the code which tests for
3352 broken timer IRQ sources.
3354 noisapnp [ISAPNP] Disables ISA PnP code.
3356 noinitrd [RAM] Tells the kernel not to load any configured
3359 nointremap [X86-64, Intel-IOMMU] Do not enable interrupt
3361 [Deprecated - use intremap=off]
3365 noinvpcid [X86] Disable the INVPCID cpu feature.
3367 nojitter [IA-64] Disables jitter checking for ITC timers.
3369 no-kvmclock [X86,KVM] Disable paravirtualized KVM clock driver
3371 no-kvmapf [X86,KVM] Disable paravirtualized asynchronous page
3375 [X86,PV_OPS] Disable paravirtualized VMware scheduler
3376 clock and use the default one.
3378 no-steal-acc [X86,PV_OPS,ARM64] Disable paravirtualized steal time
3379 accounting. steal time is computed, but won't
3380 influence scheduler behaviour
3382 nolapic [X86-32,APIC] Do not enable or use the local APIC.
3384 nolapic_timer [X86-32,APIC] Do not use the local APIC timer.
3386 noltlbs [PPC] Do not use large page/tlb entries for kernel
3387 lowmem mapping on PPC40x and PPC8xx
3389 nomca [IA-64] Disable machine check abort handling
3391 nomce [X86-32] Disable Machine Check Exception
3393 nomfgpt [X86-32] Disable Multi-Function General Purpose
3394 Timer usage (for AMD Geode machines).
3396 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to
3397 shutdown the other cpus. Instead use the REBOOT_VECTOR
3400 nomodule Disable module load
3402 nopat [X86] Disable PAT (page attribute table extension of
3403 pagetables) support.
3405 nopcid [X86-64] Disable the PCID cpu feature.
3407 norandmaps Don't use address space randomization. Equivalent to
3408 echo 0 > /proc/sys/kernel/randomize_va_space
3410 noreplace-smp [X86-32,SMP] Don't replace SMP instructions
3411 with UP alternatives
3413 nordrand [X86] Disable kernel use of the RDRAND and
3414 RDSEED instructions even if they are supported
3415 by the processor. RDRAND and RDSEED are still
3416 available to user space applications.
3418 noresume [SWSUSP] Disables resume and restores original swap
3421 no-scroll [VGA] Disables scrollback.
3422 This is required for the Braillex ib80-piezo Braille
3423 reader made by F.H. Papenmeier (Germany).
3427 nosep [BUGS=X86-32] Disables x86 SYSENTER/SYSEXIT support.
3429 nosgx [X86-64,SGX] Disables Intel SGX kernel support.
3431 nosmp [SMP] Tells an SMP kernel to act as a UP kernel,
3432 and disable the IO APIC. legacy for "maxcpus=0".
3434 nosoftlockup [KNL] Disable the soft-lockup detector.
3436 nosync [HW,M68K] Disables sync negotiation for all devices.
3438 nowatchdog [KNL] Disable both lockup detectors, i.e.
3439 soft-lockup and NMI watchdog (hard-lockup).
3443 nox2apic [X86-64,APIC] Do not enable x2APIC mode.
3445 cpu0_hotplug [X86] Turn on CPU0 hotplug feature when
3446 CONFIG_BOOTPARAM_HOTPLUG_CPU0 is off.
3447 Some features depend on CPU0. Known dependencies are:
3448 1. Resume from suspend/hibernate depends on CPU0.
3449 Suspend/hibernate will fail if CPU0 is offline and you
3450 need to online CPU0 before suspend/hibernate.
3451 2. PIC interrupts also depend on CPU0. CPU0 can't be
3452 removed if a PIC interrupt is detected.
3453 It's said poweroff/reboot may depend on CPU0 on some
3454 machines although I haven't seen such issues so far
3455 after CPU0 is offline on a few tested machines.
3456 If the dependencies are under your control, you can
3457 turn on cpu0_hotplug.
3459 nps_mtm_hs_ctr= [KNL,ARC]
3460 This parameter sets the maximum duration, in
3461 cycles, each HW thread of the CTOP can run
3462 without interruptions, before HW switches it.
3463 The actual maximum duration is 16 times this
3465 Format: integer between 1 and 255
3468 nptcg= [IA-64] Override max number of concurrent global TLB
3469 purges which is reported from either PAL_VM_SUMMARY or
3472 nr_cpus= [SMP] Maximum number of processors that an SMP kernel
3473 could support. nr_cpus=n : n >= 1 limits the kernel to
3474 support 'n' processors. It could be larger than the
3475 number of already plugged CPU during bootup, later in
3476 runtime you can physically add extra cpu until it reaches
3477 n. So during boot up some boot time memory for per-cpu
3478 variables need be pre-allocated for later physical cpu
3481 nr_uarts= [SERIAL] maximum number of UARTs to be registered.
3483 numa_balancing= [KNL,ARM64,PPC,RISCV,S390,X86] Enable or disable automatic
3485 Allowed values are enable and disable
3487 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA.
3488 'node', 'default' can be specified
3489 This can be set from sysctl after boot.
3490 See Documentation/admin-guide/sysctl/vm.rst for details.
3492 ohci1394_dma=early [HW] enable debugging via the ohci1394 driver.
3493 See Documentation/core-api/debugging-via-ohci1394.rst for more
3496 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands
3497 Rather than timing out after 20 ms if an EC
3498 command is not properly ACKed, override the length
3499 of the timeout. We have interrupts disabled while
3500 waiting for the ACK, so if this is set too high
3501 interrupts *may* be lost!
3503 omap_mux= [OMAP] Override bootloader pin multiplexing.
3504 Format: <mux_mode0.mode_name=value>...
3505 For example, to override I2C bus2:
3506 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100
3508 oops=panic Always panic on oopses. Default is to just kill the
3509 process, but there is a small probability of
3510 deadlocking the machine.
3511 This will also cause panics on machine check exceptions.
3512 Useful together with panic=30 to trigger a reboot.
3515 [KNL] Boolean flag to control whether the page allocator
3516 should randomize its free lists. The randomization may
3517 be automatically enabled if the kernel detects it is
3518 running on a platform with a direct-mapped memory-side
3519 cache, and this parameter can be used to
3520 override/disable that behavior. The state of the flag
3521 can be read from sysfs at:
3522 /sys/module/page_alloc/parameters/shuffle.
3524 page_owner= [KNL] Boot-time page_owner enabling option.
3525 Storage of the information about who allocated
3526 each page is disabled in default. With this switch,
3528 on: enable the feature
3530 page_poison= [KNL] Boot-time parameter changing the state of
3531 poisoning on the buddy allocator, available with
3532 CONFIG_PAGE_POISONING=y.
3533 off: turn off poisoning (default)
3534 on: turn on poisoning
3536 panic= [KNL] Kernel behaviour on panic: delay <timeout>
3537 timeout > 0: seconds before rebooting
3538 timeout = 0: wait forever
3539 timeout < 0: reboot immediately
3542 panic_print= Bitmask for printing system info when panic happens.
3543 User can chose combination of the following bits:
3544 bit 0: print all tasks info
3545 bit 1: print system memory info
3546 bit 2: print timer info
3547 bit 3: print locks info if CONFIG_LOCKDEP is on
3548 bit 4: print ftrace buffer
3549 bit 5: print all printk messages in buffer
3551 panic_on_taint= Bitmask for conditionally calling panic() in add_taint()
3552 Format: <hex>[,nousertaint]
3553 Hexadecimal bitmask representing the set of TAINT flags
3554 that will cause the kernel to panic when add_taint() is
3555 called with any of the flags in this set.
3556 The optional switch "nousertaint" can be utilized to
3557 prevent userspace forced crashes by writing to sysctl
3558 /proc/sys/kernel/tainted any flagset matching with the
3559 bitmask set on panic_on_taint.
3560 See Documentation/admin-guide/tainted-kernels.rst for
3561 extra details on the taint flags that users can pick
3562 to compose the bitmask to assign to panic_on_taint.
3564 panic_on_warn panic() instead of WARN(). Useful to cause kdump
3567 crash_kexec_post_notifiers
3568 Run kdump after running panic-notifiers and dumping
3569 kmsg. This only for the users who doubt kdump always
3570 succeeds in any situation.
3571 Note that this also increases risks of kdump failure,
3572 because some panic notifiers can make the crashed
3573 kernel more unstable.
3575 parkbd.port= [HW] Parallel port number the keyboard adapter is
3576 connected to, default is 0.
3578 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation,
3579 0 for XT, 1 for AT (default is AT).
3582 parport= [HW,PPT] Specify parallel ports. 0 disables.
3583 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] }
3584 Use 'auto' to force the driver to use any
3585 IRQ/DMA settings detected (the default is to
3586 ignore detected IRQ/DMA settings because of
3587 possible conflicts). You can specify the base
3588 address, IRQ, and DMA settings; IRQ and DMA
3589 should be numbers, or 'auto' (for using detected
3590 settings on that particular port), or 'nofifo'
3591 (to avoid using a FIFO even if it is detected).
3592 Parallel ports are assigned in the order they
3593 are specified on the command line, starting
3596 parport_init_mode= [HW,PPT]
3597 Configure VIA parallel port to operate in
3598 a specific mode. This is necessary on Pegasos
3599 computer where firmware has no options for setting
3600 up parallel port mode and sets it to spp.
3601 Currently this function knows 686a and 8231 chips.
3602 Format: [spp|ps2|epp|ecp|ecpepp]
3604 pata_legacy.all= [HW,LIBATA]
3606 Set to non-zero to probe primary and secondary ISA
3607 port ranges on PCI systems where no PCI PATA device
3608 has been found at either range. Disabled by default.
3610 pata_legacy.autospeed= [HW,LIBATA]
3612 Set to non-zero if a chip is present that snoops speed
3613 changes. Disabled by default.
3615 pata_legacy.ht6560a= [HW,LIBATA]
3617 Set to 1, 2, or 3 for HT 6560A on the primary channel,
3618 the secondary channel, or both channels respectively.
3619 Disabled by default.
3621 pata_legacy.ht6560b= [HW,LIBATA]
3623 Set to 1, 2, or 3 for HT 6560B on the primary channel,
3624 the secondary channel, or both channels respectively.
3625 Disabled by default.
3627 pata_legacy.iordy_mask= [HW,LIBATA]
3629 IORDY enable mask. Set individual bits to allow IORDY
3630 for the respective channel. Bit 0 is for the first
3631 legacy channel handled by this driver, bit 1 is for
3632 the second channel, and so on. The sequence will often
3633 correspond to the primary legacy channel, the secondary
3634 legacy channel, and so on, but the handling of a PCI
3635 bus and the use of other driver options may interfere
3636 with the sequence. By default IORDY is allowed across
3639 pata_legacy.opti82c46x= [HW,LIBATA]
3641 Set to 1, 2, or 3 for Opti 82c611A on the primary
3642 channel, the secondary channel, or both channels
3643 respectively. Disabled by default.
3645 pata_legacy.opti82c611a= [HW,LIBATA]
3647 Set to 1, 2, or 3 for Opti 82c465MV on the primary
3648 channel, the secondary channel, or both channels
3649 respectively. Disabled by default.
3651 pata_legacy.pio_mask= [HW,LIBATA]
3653 PIO mode mask for autospeed devices. Set individual
3654 bits to allow the use of the respective PIO modes.
3655 Bit 0 is for mode 0, bit 1 is for mode 1, and so on.
3656 All modes allowed by default.
3658 pata_legacy.probe_all= [HW,LIBATA]
3660 Set to non-zero to probe tertiary and further ISA
3661 port ranges on PCI systems. Disabled by default.
3663 pata_legacy.probe_mask= [HW,LIBATA]
3665 Probe mask for legacy ISA PATA ports. Depending on
3666 platform configuration and the use of other driver
3667 options up to 6 legacy ports are supported: 0x1f0,
3668 0x170, 0x1e8, 0x168, 0x1e0, 0x160, however probing
3669 of individual ports can be disabled by setting the
3670 corresponding bits in the mask to 1. Bit 0 is for
3671 the first port in the list above (0x1f0), and so on.
3672 By default all supported ports are probed.
3674 pata_legacy.qdi= [HW,LIBATA]
3676 Set to non-zero to probe QDI controllers. By default
3677 set to 1 if CONFIG_PATA_QDI_MODULE, 0 otherwise.
3679 pata_legacy.winbond= [HW,LIBATA]
3681 Set to non-zero to probe Winbond controllers. Use
3682 the standard I/O port (0x130) if 1, otherwise the
3683 value given is the I/O port to use (typically 0x1b0).
3684 By default set to 1 if CONFIG_PATA_WINBOND_VLB_MODULE,
3687 pata_platform.pio_mask= [HW,LIBATA]
3689 Supported PIO mode mask. Set individual bits to allow
3690 the use of the respective PIO modes. Bit 0 is for
3691 mode 0, bit 1 is for mode 1, and so on. Mode 0 only
3695 Halt all CPUs after the first oops has been printed for
3696 the specified number of seconds. This is to be used if
3697 your oopses keep scrolling off the screen.
3702 See header of drivers/block/paride/pcd.c.
3703 See also Documentation/admin-guide/blockdev/paride.rst.
3705 pci=option[,option...] [PCI] various PCI subsystem options.
3707 Some options herein operate on a specific device
3708 or a set of devices (<pci_dev>). These are
3709 specified in one of the following formats:
3711 [<domain>:]<bus>:<dev>.<func>[/<dev>.<func>]*
3712 pci:<vendor>:<device>[:<subvendor>:<subdevice>]
3714 Note: the first format specifies a PCI
3715 bus/device/function address which may change
3716 if new hardware is inserted, if motherboard
3717 firmware changes, or due to changes caused
3718 by other kernel parameters. If the
3719 domain is left unspecified, it is
3720 taken to be zero. Optionally, a path
3721 to a device through multiple device/function
3722 addresses can be specified after the base
3723 address (this is more robust against
3724 renumbering issues). The second format
3725 selects devices using IDs from the
3726 configuration space which may match multiple
3727 devices in the system.
3729 earlydump dump PCI config space before the kernel
3731 off [X86] don't probe for the PCI bus
3732 bios [X86-32] force use of PCI BIOS, don't access
3733 the hardware directly. Use this if your machine
3734 has a non-standard PCI host bridge.
3735 nobios [X86-32] disallow use of PCI BIOS, only direct
3736 hardware access methods are allowed. Use this
3737 if you experience crashes upon bootup and you
3738 suspect they are caused by the BIOS.
3739 conf1 [X86] Force use of PCI Configuration Access
3740 Mechanism 1 (config address in IO port 0xCF8,
3741 data in IO port 0xCFC, both 32-bit).
3742 conf2 [X86] Force use of PCI Configuration Access
3743 Mechanism 2 (IO port 0xCF8 is an 8-bit port for
3744 the function, IO port 0xCFA, also 8-bit, sets
3745 bus number. The config space is then accessed
3746 through ports 0xC000-0xCFFF).
3747 See http://wiki.osdev.org/PCI for more info
3748 on the configuration access mechanisms.
3749 noaer [PCIE] If the PCIEAER kernel config parameter is
3750 enabled, this kernel boot option can be used to
3751 disable the use of PCIE advanced error reporting.
3752 nodomains [PCI] Disable support for multiple PCI
3753 root domains (aka PCI segments, in ACPI-speak).
3754 nommconf [X86] Disable use of MMCONFIG for PCI
3756 check_enable_amd_mmconf [X86] check for and enable
3757 properly configured MMIO access to PCI
3758 config space on AMD family 10h CPU
3759 nomsi [MSI] If the PCI_MSI kernel config parameter is
3760 enabled, this kernel boot option can be used to
3761 disable the use of MSI interrupts system-wide.
3762 noioapicquirk [APIC] Disable all boot interrupt quirks.
3763 Safety option to keep boot IRQs enabled. This
3764 should never be necessary.
3765 ioapicreroute [APIC] Enable rerouting of boot IRQs to the
3766 primary IO-APIC for bridges that cannot disable
3767 boot IRQs. This fixes a source of spurious IRQs
3768 when the system masks IRQs.
3769 noioapicreroute [APIC] Disable workaround that uses the
3770 boot IRQ equivalent of an IRQ that connects to
3771 a chipset where boot IRQs cannot be disabled.
3772 The opposite of ioapicreroute.
3773 biosirq [X86-32] Use PCI BIOS calls to get the interrupt
3774 routing table. These calls are known to be buggy
3775 on several machines and they hang the machine
3776 when used, but on other computers it's the only
3777 way to get the interrupt routing table. Try
3778 this option if the kernel is unable to allocate
3779 IRQs or discover secondary PCI buses on your
3781 rom [X86] Assign address space to expansion ROMs.
3782 Use with caution as certain devices share
3783 address decoders between ROMs and other
3785 norom [X86] Do not assign address space to
3786 expansion ROMs that do not already have
3787 BIOS assigned address ranges.
3788 nobar [X86] Do not assign address space to the
3789 BARs that weren't assigned by the BIOS.
3790 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be
3791 assigned automatically to PCI devices. You can
3792 make the kernel exclude IRQs of your ISA cards
3794 pirqaddr=0xAAAAA [X86] Specify the physical address
3795 of the PIRQ table (normally generated
3796 by the BIOS) if it is outside the
3797 F0000h-100000h range.
3798 lastbus=N [X86] Scan all buses thru bus #N. Can be
3799 useful if the kernel is unable to find your
3800 secondary buses and you want to tell it
3801 explicitly which ones they are.
3802 assign-busses [X86] Always assign all PCI bus
3803 numbers ourselves, overriding
3804 whatever the firmware may have done.
3805 usepirqmask [X86] Honor the possible IRQ mask stored
3806 in the BIOS $PIR table. This is needed on
3807 some systems with broken BIOSes, notably
3808 some HP Pavilion N5400 and Omnibook XE3
3809 notebooks. This will have no effect if ACPI
3810 IRQ routing is enabled.
3811 noacpi [X86] Do not use ACPI for IRQ routing
3812 or for PCI scanning.
3813 use_crs [X86] Use PCI host bridge window information
3814 from ACPI. On BIOSes from 2008 or later, this
3815 is enabled by default. If you need to use this,
3816 please report a bug.
3817 nocrs [X86] Ignore PCI host bridge windows from ACPI.
3818 If you need to use this, please report a bug.
3819 routeirq Do IRQ routing for all PCI devices.
3820 This is normally done in pci_enable_device(),
3821 so this option is a temporary workaround
3822 for broken drivers that don't call it.
3823 skip_isa_align [X86] do not align io start addr, so can
3824 handle more pci cards
3825 noearly [X86] Don't do any early type 1 scanning.
3826 This might help on some broken boards which
3827 machine check when some devices' config space
3828 is read. But various workarounds are disabled
3829 and some IOMMU drivers will not work.
3830 bfsort Sort PCI devices into breadth-first order.
3831 This sorting is done to get a device
3832 order compatible with older (<= 2.4) kernels.
3833 nobfsort Don't sort PCI devices into breadth-first order.
3834 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size)
3835 tuning and use the BIOS-configured MPS defaults.
3836 pcie_bus_safe Set every device's MPS to the largest value
3837 supported by all devices below the root complex.
3838 pcie_bus_perf Set device MPS to the largest allowable MPS
3839 based on its parent bus. Also set MRRS (Max
3840 Read Request Size) to the largest supported
3841 value (no larger than the MPS that the device
3842 or bus can support) for best performance.
3843 pcie_bus_peer2peer Set every device's MPS to 128B, which
3844 every device is guaranteed to support. This
3845 configuration allows peer-to-peer DMA between
3846 any pair of devices, possibly at the cost of
3847 reduced performance. This also guarantees
3848 that hot-added devices will work.
3849 cbiosize=nn[KMG] The fixed amount of bus space which is
3850 reserved for the CardBus bridge's IO window.
3851 The default value is 256 bytes.
3852 cbmemsize=nn[KMG] The fixed amount of bus space which is
3853 reserved for the CardBus bridge's memory
3854 window. The default value is 64 megabytes.
3857 [<order of align>@]<pci_dev>[; ...]
3858 Specifies alignment and device to reassign
3859 aligned memory resources. How to
3860 specify the device is described above.
3861 If <order of align> is not specified,
3862 PAGE_SIZE is used as alignment.
3863 A PCI-PCI bridge can be specified if resource
3864 windows need to be expanded.
3865 To specify the alignment for several
3866 instances of a device, the PCI vendor,
3867 device, subvendor, and subdevice may be
3868 specified, e.g., 12@pci:8086:9c22:103c:198f
3869 for 4096-byte alignment.
3870 ecrc= Enable/disable PCIe ECRC (transaction layer
3871 end-to-end CRC checking).
3872 bios: Use BIOS/firmware settings. This is the
3876 hpiosize=nn[KMG] The fixed amount of bus space which is
3877 reserved for hotplug bridge's IO window.
3878 Default size is 256 bytes.
3879 hpmmiosize=nn[KMG] The fixed amount of bus space which is
3880 reserved for hotplug bridge's MMIO window.
3881 Default size is 2 megabytes.
3882 hpmmioprefsize=nn[KMG] The fixed amount of bus space which is
3883 reserved for hotplug bridge's MMIO_PREF window.
3884 Default size is 2 megabytes.
3885 hpmemsize=nn[KMG] The fixed amount of bus space which is
3886 reserved for hotplug bridge's MMIO and
3888 Default size is 2 megabytes.
3889 hpbussize=nn The minimum amount of additional bus numbers
3890 reserved for buses below a hotplug bridge.
3892 realloc= Enable/disable reallocating PCI bridge resources
3893 if allocations done by BIOS are too small to
3894 accommodate resources required by all child
3896 off: Turn realloc off
3898 realloc same as realloc=on
3899 noari do not use PCIe ARI.
3900 noats [PCIE, Intel-IOMMU, AMD-IOMMU]
3901 do not use PCIe ATS (and IOMMU device IOTLB).
3902 pcie_scan_all Scan all possible PCIe devices. Otherwise we
3903 only look for one device below a PCIe downstream
3905 big_root_window Try to add a big 64bit memory window to the PCIe
3906 root complex on AMD CPUs. Some GFX hardware
3907 can resize a BAR to allow access to all VRAM.
3908 Adding the window is slightly risky (it may
3909 conflict with unreported devices), so this
3911 disable_acs_redir=<pci_dev>[; ...]
3912 Specify one or more PCI devices (in the format
3913 specified above) separated by semicolons.
3914 Each device specified will have the PCI ACS
3915 redirect capabilities forced off which will
3916 allow P2P traffic between devices through
3917 bridges without forcing it upstream. Note:
3918 this removes isolation between devices and
3919 may put more devices in an IOMMU group.
3920 force_floating [S390] Force usage of floating interrupts.
3921 nomio [S390] Do not use MIO instructions.
3922 norid [S390] ignore the RID field and force use of
3923 one PCI domain per PCI function
3925 pcie_aspm= [PCIE] Forcibly enable or disable PCIe Active State Power
3928 force Enable ASPM even on devices that claim not to support it.
3929 WARNING: Forcing ASPM on may cause system lockups.
3931 pcie_ports= [PCIE] PCIe port services handling:
3932 native Use native PCIe services (PME, AER, DPC, PCIe hotplug)
3933 even if the platform doesn't give the OS permission to
3934 use them. This may cause conflicts if the platform
3935 also tries to use these services.
3936 dpc-native Use native PCIe service for DPC only. May
3937 cause conflicts if firmware uses AER or DPC.
3938 compat Disable native PCIe services (PME, AER, DPC, PCIe
3941 pcie_port_pm= [PCIE] PCIe port power management handling:
3942 off Disable power management of all PCIe ports
3943 force Forcibly enable power management of all PCIe ports
3945 pcie_pme= [PCIE,PM] Native PCIe PME signaling options:
3946 nomsi Do not use MSI for native PCIe PME signaling (this makes
3947 all PCIe root ports use INTx for all services).
3949 pcmv= [HW,PCMCIA] BadgePAD 4
3953 Keep all power-domains already enabled by bootloader on,
3954 even if no driver has claimed them. This is useful
3955 for debug and development, but should not be
3956 needed on a platform with proper driver support.
3959 See Documentation/admin-guide/blockdev/paride.rst.
3961 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at
3964 See arch/parisc/kernel/pdc_chassis.c
3966 percpu_alloc= Select which percpu first chunk allocator to use.
3967 Currently supported values are "embed" and "page".
3968 Archs may support subset or none of the selections.
3969 See comments in mm/percpu.c for details on each
3970 allocator. This parameter is primarily for debugging
3971 and performance comparison.
3974 See Documentation/admin-guide/blockdev/paride.rst.
3977 See Documentation/admin-guide/blockdev/paride.rst.
3979 pirq= [SMP,APIC] Manual mp-table setup
3980 See Documentation/x86/i386/IO-APIC.rst.
3982 plip= [PPT,NET] Parallel port network link
3983 Format: { parport<nr> | timid | 0 }
3984 See also Documentation/admin-guide/parport.rst.
3986 pmtmr= [X86] Manual setup of pmtmr I/O Port.
3987 Override pmtimer IOPort with a hex value.
3990 pm_debug_messages [SUSPEND,KNL]
3991 Enable suspend/resume debug messages during boot up.
3994 Enable PNP debug messages (depends on the
3995 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time
3996 via /sys/module/pnp/parameters/debug. We always show
3997 current resource usage; turning this on also shows
3998 possible settings and some assignment information.
4004 { on | off | curr | res | no-curr | no-res }
4007 [ISAPNP] Exclude IRQs for the autoconfiguration
4010 [ISAPNP] Exclude DMAs for the autoconfiguration
4012 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration
4013 Ranges are in pairs (I/O port base and size).
4016 [ISAPNP] Exclude memory regions for the
4018 Ranges are in pairs (memory base and size).
4020 ports= [IP_VS_FTP] IPVS ftp helper module
4022 Up to 8 (IP_VS_APP_MAX_PORTS) ports
4024 Format: <port>,<port>....
4026 powersave=off [PPC] This option disables power saving features.
4027 It specifically disables cpuidle and sets the
4028 platform machine description specific power_save
4029 function to NULL. On Idle the CPU just reduces
4032 ppc_strict_facility_enable
4033 [PPC] This option catches any kernel floating point,
4034 Altivec, VSX and SPE outside of regions specifically
4035 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()).
4036 There is some performance impact when enabling this.
4040 Disable Hardware Transactional Memory
4043 Select preemption mode if you have CONFIG_PREEMPT_DYNAMIC
4044 none - Limited to cond_resched() calls
4045 voluntary - Limited to cond_resched() and might_sleep() calls
4046 full - Any section that isn't explicitly preempt disabled
4047 can be preempted anytime.
4049 print-fatal-signals=
4050 [KNL] debug: print fatal signals
4052 If enabled, warn about various signal handling
4053 related application anomalies: too many signals,
4054 too many POSIX.1 timers, fatal signals causing a
4057 If you hit the warning due to signal overflow,
4058 you might want to try "ulimit -i unlimited".
4062 printk.always_kmsg_dump=
4063 Trigger kmsg_dump for cases other than kernel oops or
4065 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
4068 printk.devkmsg={on,off,ratelimit}
4069 Control writing to /dev/kmsg.
4070 on - unlimited logging to /dev/kmsg from userspace
4071 off - logging to /dev/kmsg disabled
4072 ratelimit - ratelimit the logging
4075 printk.time= Show timing data prefixed to each printk message line
4076 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
4078 processor.max_cstate= [HW,ACPI]
4079 Limit processor to maximum C-state
4080 max_cstate=9 overrides any DMI blacklist limit.
4082 processor.nocst [HW,ACPI]
4083 Ignore the _CST method to determine C-states,
4084 instead using the legacy FADT method
4086 profile= [KNL] Enable kernel profiling via /proc/profile
4087 Format: [<profiletype>,]<number>
4088 Param: <profiletype>: "schedule", "sleep", or "kvm"
4089 [defaults to kernel profiling]
4090 Param: "schedule" - profile schedule points.
4091 Param: "sleep" - profile D-state sleeping (millisecs).
4092 Requires CONFIG_SCHEDSTATS
4093 Param: "kvm" - profile VM exits.
4094 Param: <number> - step/bucket size as a power of 2 for
4095 statistical time based profiling.
4097 prompt_ramdisk= [RAM] [Deprecated]
4099 prot_virt= [S390] enable hosting protected virtual machines
4100 isolated from the hypervisor (if hardware supports
4104 psi= [KNL] Enable or disable pressure stall information
4108 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to
4109 probe for; one of (bare|imps|exps|lifebook|any).
4110 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports
4112 psmouse.resetafter= [HW,MOUSE]
4113 Try to reset the device after so many bad packets
4116 [HW,MOUSE] Set desired mouse resolution, in dpi.
4117 psmouse.smartscroll=
4118 [HW,MOUSE] Controls Logitech smartscroll autorepeat.
4119 0 = disabled, 1 = enabled (default).
4121 pstore.backend= Specify the name of the pstore backend to use
4124 See Documentation/admin-guide/blockdev/paride.rst.
4126 pti= [X86-64] Control Page Table Isolation of user and
4127 kernel address spaces. Disabling this feature
4128 removes hardening, but improves performance of
4129 system calls and interrupts.
4131 on - unconditionally enable
4132 off - unconditionally disable
4133 auto - kernel detects whether your CPU model is
4134 vulnerable to issues that PTI mitigates
4136 Not specifying this option is equivalent to pti=auto.
4139 Equivalent to pti=off
4142 [KNL] Number of legacy pty's. Overwrites compiled-in
4145 quiet [KNL] Disable most log messages
4150 See Documentation/admin-guide/md.rst.
4152 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes
4153 See Documentation/admin-guide/blockdev/ramdisk.rst.
4155 ramdisk_start= [RAM] RAM disk image start address
4157 random.trust_cpu={on,off}
4158 [KNL] Enable or disable trusting the use of the
4159 CPU's random number generator (if available) to
4160 fully seed the kernel's CRNG. Default is controlled
4161 by CONFIG_RANDOM_TRUST_CPU.
4163 randomize_kstack_offset=
4164 [KNL] Enable or disable kernel stack offset
4165 randomization, which provides roughly 5 bits of
4166 entropy, frustrating memory corruption attacks
4167 that depend on stack address determinism or
4168 cross-syscall address exposures. This is only
4169 available on architectures that have defined
4170 CONFIG_HAVE_ARCH_RANDOMIZE_KSTACK_OFFSET.
4171 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
4172 Default is CONFIG_RANDOMIZE_KSTACK_OFFSET_DEFAULT.
4174 ras=option[,option,...] [KNL] RAS-specific options
4177 Disable the Correctable Errors Collector,
4178 see CONFIG_RAS_CEC help text.
4181 The argument is a cpu list, as described above.
4183 In kernels built with CONFIG_RCU_NOCB_CPU=y, set
4184 the specified list of CPUs to be no-callback CPUs.
4185 Invocation of these CPUs' RCU callbacks will be
4186 offloaded to "rcuox/N" kthreads created for that
4187 purpose, where "x" is "p" for RCU-preempt, and
4188 "s" for RCU-sched, and "N" is the CPU number.
4189 This reduces OS jitter on the offloaded CPUs,
4190 which can be useful for HPC and real-time
4191 workloads. It can also improve energy efficiency
4192 for asymmetric multiprocessors.
4195 Rather than requiring that offloaded CPUs
4196 (specified by rcu_nocbs= above) explicitly
4197 awaken the corresponding "rcuoN" kthreads,
4198 make these kthreads poll for callbacks.
4199 This improves the real-time response for the
4200 offloaded CPUs by relieving them of the need to
4201 wake up the corresponding kthread, but degrades
4202 energy efficiency by requiring that the kthreads
4203 periodically wake up to do the polling.
4205 rcutree.blimit= [KNL]
4206 Set maximum number of finished RCU callbacks to
4207 process in one batch.
4209 rcutree.dump_tree= [KNL]
4210 Dump the structure of the rcu_node combining tree
4211 out at early boot. This is used for diagnostic
4212 purposes, to verify correct tree setup.
4214 rcutree.gp_cleanup_delay= [KNL]
4215 Set the number of jiffies to delay each step of
4216 RCU grace-period cleanup.
4218 rcutree.gp_init_delay= [KNL]
4219 Set the number of jiffies to delay each step of
4220 RCU grace-period initialization.
4222 rcutree.gp_preinit_delay= [KNL]
4223 Set the number of jiffies to delay each step of
4224 RCU grace-period pre-initialization, that is,
4225 the propagation of recent CPU-hotplug changes up
4226 the rcu_node combining tree.
4228 rcutree.use_softirq= [KNL]
4229 If set to zero, move all RCU_SOFTIRQ processing to
4230 per-CPU rcuc kthreads. Defaults to a non-zero
4231 value, meaning that RCU_SOFTIRQ is used by default.
4232 Specify rcutree.use_softirq=0 to use rcuc kthreads.
4234 But note that CONFIG_PREEMPT_RT=y kernels disable
4235 this kernel boot parameter, forcibly setting it
4238 rcutree.rcu_fanout_exact= [KNL]
4239 Disable autobalancing of the rcu_node combining
4240 tree. This is used by rcutorture, and might
4241 possibly be useful for architectures having high
4242 cache-to-cache transfer latencies.
4244 rcutree.rcu_fanout_leaf= [KNL]
4245 Change the number of CPUs assigned to each
4246 leaf rcu_node structure. Useful for very
4247 large systems, which will choose the value 64,
4248 and for NUMA systems with large remote-access
4249 latencies, which will choose a value aligned
4250 with the appropriate hardware boundaries.
4252 rcutree.rcu_min_cached_objs= [KNL]
4253 Minimum number of objects which are cached and
4254 maintained per one CPU. Object size is equal
4255 to PAGE_SIZE. The cache allows to reduce the
4256 pressure to page allocator, also it makes the
4257 whole algorithm to behave better in low memory
4260 rcutree.jiffies_till_first_fqs= [KNL]
4261 Set delay from grace-period initialization to
4262 first attempt to force quiescent states.
4263 Units are jiffies, minimum value is zero,
4264 and maximum value is HZ.
4266 rcutree.jiffies_till_next_fqs= [KNL]
4267 Set delay between subsequent attempts to force
4268 quiescent states. Units are jiffies, minimum
4269 value is one, and maximum value is HZ.
4271 rcutree.jiffies_till_sched_qs= [KNL]
4272 Set required age in jiffies for a
4273 given grace period before RCU starts
4274 soliciting quiescent-state help from
4275 rcu_note_context_switch() and cond_resched().
4276 If not specified, the kernel will calculate
4277 a value based on the most recent settings
4278 of rcutree.jiffies_till_first_fqs
4279 and rcutree.jiffies_till_next_fqs.
4280 This calculated value may be viewed in
4281 rcutree.jiffies_to_sched_qs. Any attempt to set
4282 rcutree.jiffies_to_sched_qs will be cheerfully
4285 rcutree.kthread_prio= [KNL,BOOT]
4286 Set the SCHED_FIFO priority of the RCU per-CPU
4287 kthreads (rcuc/N). This value is also used for
4288 the priority of the RCU boost threads (rcub/N)
4289 and for the RCU grace-period kthreads (rcu_bh,
4290 rcu_preempt, and rcu_sched). If RCU_BOOST is
4291 set, valid values are 1-99 and the default is 1
4292 (the least-favored priority). Otherwise, when
4293 RCU_BOOST is not set, valid values are 0-99 and
4294 the default is zero (non-realtime operation).
4296 rcutree.rcu_nocb_gp_stride= [KNL]
4297 Set the number of NOCB callback kthreads in
4298 each group, which defaults to the square root
4299 of the number of CPUs. Larger numbers reduce
4300 the wakeup overhead on the global grace-period
4301 kthread, but increases that same overhead on
4302 each group's NOCB grace-period kthread.
4304 rcutree.qhimark= [KNL]
4305 Set threshold of queued RCU callbacks beyond which
4306 batch limiting is disabled.
4308 rcutree.qlowmark= [KNL]
4309 Set threshold of queued RCU callbacks below which
4310 batch limiting is re-enabled.
4312 rcutree.qovld= [KNL]
4313 Set threshold of queued RCU callbacks beyond which
4314 RCU's force-quiescent-state scan will aggressively
4315 enlist help from cond_resched() and sched IPIs to
4316 help CPUs more quickly reach quiescent states.
4317 Set to less than zero to make this be set based
4318 on rcutree.qhimark at boot time and to zero to
4319 disable more aggressive help enlistment.
4321 rcutree.rcu_idle_gp_delay= [KNL]
4322 Set wakeup interval for idle CPUs that have
4323 RCU callbacks (RCU_FAST_NO_HZ=y).
4325 rcutree.rcu_kick_kthreads= [KNL]
4326 Cause the grace-period kthread to get an extra
4327 wake_up() if it sleeps three times longer than
4328 it should at force-quiescent-state time.
4329 This wake_up() will be accompanied by a
4330 WARN_ONCE() splat and an ftrace_dump().
4332 rcutree.rcu_unlock_delay= [KNL]
4333 In CONFIG_RCU_STRICT_GRACE_PERIOD=y kernels,
4334 this specifies an rcu_read_unlock()-time delay
4335 in microseconds. This defaults to zero.
4336 Larger delays increase the probability of
4337 catching RCU pointer leaks, that is, buggy use
4338 of RCU-protected pointers after the relevant
4339 rcu_read_unlock() has completed.
4341 rcutree.sysrq_rcu= [KNL]
4342 Commandeer a sysrq key to dump out Tree RCU's
4343 rcu_node tree with an eye towards determining
4344 why a new grace period has not yet started.
4346 rcuscale.gp_async= [KNL]
4347 Measure performance of asynchronous
4348 grace-period primitives such as call_rcu().
4350 rcuscale.gp_async_max= [KNL]
4351 Specify the maximum number of outstanding
4352 callbacks per writer thread. When a writer
4353 thread exceeds this limit, it invokes the
4354 corresponding flavor of rcu_barrier() to allow
4355 previously posted callbacks to drain.
4357 rcuscale.gp_exp= [KNL]
4358 Measure performance of expedited synchronous
4359 grace-period primitives.
4361 rcuscale.holdoff= [KNL]
4362 Set test-start holdoff period. The purpose of
4363 this parameter is to delay the start of the
4364 test until boot completes in order to avoid
4367 rcuscale.kfree_rcu_test= [KNL]
4368 Set to measure performance of kfree_rcu() flooding.
4370 rcuscale.kfree_rcu_test_double= [KNL]
4371 Test the double-argument variant of kfree_rcu().
4372 If this parameter has the same value as
4373 rcuscale.kfree_rcu_test_single, both the single-
4374 and double-argument variants are tested.
4376 rcuscale.kfree_rcu_test_single= [KNL]
4377 Test the single-argument variant of kfree_rcu().
4378 If this parameter has the same value as
4379 rcuscale.kfree_rcu_test_double, both the single-
4380 and double-argument variants are tested.
4382 rcuscale.kfree_nthreads= [KNL]
4383 The number of threads running loops of kfree_rcu().
4385 rcuscale.kfree_alloc_num= [KNL]
4386 Number of allocations and frees done in an iteration.
4388 rcuscale.kfree_loops= [KNL]
4389 Number of loops doing rcuscale.kfree_alloc_num number
4390 of allocations and frees.
4392 rcuscale.nreaders= [KNL]
4393 Set number of RCU readers. The value -1 selects
4394 N, where N is the number of CPUs. A value
4395 "n" less than -1 selects N-n+1, where N is again
4396 the number of CPUs. For example, -2 selects N
4397 (the number of CPUs), -3 selects N+1, and so on.
4398 A value of "n" less than or equal to -N selects
4401 rcuscale.nwriters= [KNL]
4402 Set number of RCU writers. The values operate
4403 the same as for rcuscale.nreaders.
4404 N, where N is the number of CPUs
4406 rcuscale.perf_type= [KNL]
4407 Specify the RCU implementation to test.
4409 rcuscale.shutdown= [KNL]
4410 Shut the system down after performance tests
4411 complete. This is useful for hands-off automated
4414 rcuscale.verbose= [KNL]
4415 Enable additional printk() statements.
4417 rcuscale.writer_holdoff= [KNL]
4418 Write-side holdoff between grace periods,
4419 in microseconds. The default of zero says
4422 rcutorture.fqs_duration= [KNL]
4423 Set duration of force_quiescent_state bursts
4426 rcutorture.fqs_holdoff= [KNL]
4427 Set holdoff time within force_quiescent_state bursts
4430 rcutorture.fqs_stutter= [KNL]
4431 Set wait time between force_quiescent_state bursts
4434 rcutorture.fwd_progress= [KNL]
4435 Enable RCU grace-period forward-progress testing
4436 for the types of RCU supporting this notion.
4438 rcutorture.fwd_progress_div= [KNL]
4439 Specify the fraction of a CPU-stall-warning
4440 period to do tight-loop forward-progress testing.
4442 rcutorture.fwd_progress_holdoff= [KNL]
4443 Number of seconds to wait between successive
4444 forward-progress tests.
4446 rcutorture.fwd_progress_need_resched= [KNL]
4447 Enclose cond_resched() calls within checks for
4448 need_resched() during tight-loop forward-progress
4451 rcutorture.gp_cond= [KNL]
4452 Use conditional/asynchronous update-side
4453 primitives, if available.
4455 rcutorture.gp_exp= [KNL]
4456 Use expedited update-side primitives, if available.
4458 rcutorture.gp_normal= [KNL]
4459 Use normal (non-expedited) asynchronous
4460 update-side primitives, if available.
4462 rcutorture.gp_sync= [KNL]
4463 Use normal (non-expedited) synchronous
4464 update-side primitives, if available. If all
4465 of rcutorture.gp_cond=, rcutorture.gp_exp=,
4466 rcutorture.gp_normal=, and rcutorture.gp_sync=
4467 are zero, rcutorture acts as if is interpreted
4468 they are all non-zero.
4470 rcutorture.irqreader= [KNL]
4471 Run RCU readers from irq handlers, or, more
4472 accurately, from a timer handler. Not all RCU
4473 flavors take kindly to this sort of thing.
4475 rcutorture.leakpointer= [KNL]
4476 Leak an RCU-protected pointer out of the reader.
4477 This can of course result in splats, and is
4478 intended to test the ability of things like
4479 CONFIG_RCU_STRICT_GRACE_PERIOD=y to detect
4482 rcutorture.n_barrier_cbs= [KNL]
4483 Set callbacks/threads for rcu_barrier() testing.
4485 rcutorture.nfakewriters= [KNL]
4486 Set number of concurrent RCU writers. These just
4487 stress RCU, they don't participate in the actual
4488 test, hence the "fake".
4490 rcutorture.nocbs_nthreads= [KNL]
4491 Set number of RCU callback-offload togglers.
4492 Zero (the default) disables toggling.
4494 rcutorture.nocbs_toggle= [KNL]
4495 Set the delay in milliseconds between successive
4496 callback-offload toggling attempts.
4498 rcutorture.nreaders= [KNL]
4499 Set number of RCU readers. The value -1 selects
4500 N-1, where N is the number of CPUs. A value
4501 "n" less than -1 selects N-n-2, where N is again
4502 the number of CPUs. For example, -2 selects N
4503 (the number of CPUs), -3 selects N+1, and so on.
4505 rcutorture.object_debug= [KNL]
4506 Enable debug-object double-call_rcu() testing.
4508 rcutorture.onoff_holdoff= [KNL]
4509 Set time (s) after boot for CPU-hotplug testing.
4511 rcutorture.onoff_interval= [KNL]
4512 Set time (jiffies) between CPU-hotplug operations,
4513 or zero to disable CPU-hotplug testing.
4515 rcutorture.read_exit= [KNL]
4516 Set the number of read-then-exit kthreads used
4517 to test the interaction of RCU updaters and
4518 task-exit processing.
4520 rcutorture.read_exit_burst= [KNL]
4521 The number of times in a given read-then-exit
4522 episode that a set of read-then-exit kthreads
4525 rcutorture.read_exit_delay= [KNL]
4526 The delay, in seconds, between successive
4527 read-then-exit testing episodes.
4529 rcutorture.shuffle_interval= [KNL]
4530 Set task-shuffle interval (s). Shuffling tasks
4531 allows some CPUs to go into dyntick-idle mode
4532 during the rcutorture test.
4534 rcutorture.shutdown_secs= [KNL]
4535 Set time (s) after boot system shutdown. This
4536 is useful for hands-off automated testing.
4538 rcutorture.stall_cpu= [KNL]
4539 Duration of CPU stall (s) to test RCU CPU stall
4540 warnings, zero to disable.
4542 rcutorture.stall_cpu_block= [KNL]
4543 Sleep while stalling if set. This will result
4544 in warnings from preemptible RCU in addition
4545 to any other stall-related activity.
4547 rcutorture.stall_cpu_holdoff= [KNL]
4548 Time to wait (s) after boot before inducing stall.
4550 rcutorture.stall_cpu_irqsoff= [KNL]
4551 Disable interrupts while stalling if set.
4553 rcutorture.stall_gp_kthread= [KNL]
4554 Duration (s) of forced sleep within RCU
4555 grace-period kthread to test RCU CPU stall
4556 warnings, zero to disable. If both stall_cpu
4557 and stall_gp_kthread are specified, the
4558 kthread is starved first, then the CPU.
4560 rcutorture.stat_interval= [KNL]
4561 Time (s) between statistics printk()s.
4563 rcutorture.stutter= [KNL]
4564 Time (s) to stutter testing, for example, specifying
4565 five seconds causes the test to run for five seconds,
4566 wait for five seconds, and so on. This tests RCU's
4567 ability to transition abruptly to and from idle.
4569 rcutorture.test_boost= [KNL]
4570 Test RCU priority boosting? 0=no, 1=maybe, 2=yes.
4571 "Maybe" means test if the RCU implementation
4572 under test support RCU priority boosting.
4574 rcutorture.test_boost_duration= [KNL]
4575 Duration (s) of each individual boost test.
4577 rcutorture.test_boost_interval= [KNL]
4578 Interval (s) between each boost test.
4580 rcutorture.test_no_idle_hz= [KNL]
4581 Test RCU's dyntick-idle handling. See also the
4582 rcutorture.shuffle_interval parameter.
4584 rcutorture.torture_type= [KNL]
4585 Specify the RCU implementation to test.
4587 rcutorture.verbose= [KNL]
4588 Enable additional printk() statements.
4590 rcupdate.rcu_cpu_stall_ftrace_dump= [KNL]
4591 Dump ftrace buffer after reporting RCU CPU
4594 rcupdate.rcu_cpu_stall_suppress= [KNL]
4595 Suppress RCU CPU stall warning messages.
4597 rcupdate.rcu_cpu_stall_suppress_at_boot= [KNL]
4598 Suppress RCU CPU stall warning messages and
4599 rcutorture writer stall warnings that occur
4600 during early boot, that is, during the time
4601 before the init task is spawned.
4603 rcupdate.rcu_cpu_stall_timeout= [KNL]
4604 Set timeout for RCU CPU stall warning messages.
4606 rcupdate.rcu_expedited= [KNL]
4607 Use expedited grace-period primitives, for
4608 example, synchronize_rcu_expedited() instead
4609 of synchronize_rcu(). This reduces latency,
4610 but can increase CPU utilization, degrade
4611 real-time latency, and degrade energy efficiency.
4612 No effect on CONFIG_TINY_RCU kernels.
4614 rcupdate.rcu_normal= [KNL]
4615 Use only normal grace-period primitives,
4616 for example, synchronize_rcu() instead of
4617 synchronize_rcu_expedited(). This improves
4618 real-time latency, CPU utilization, and
4619 energy efficiency, but can expose users to
4620 increased grace-period latency. This parameter
4621 overrides rcupdate.rcu_expedited. No effect on
4622 CONFIG_TINY_RCU kernels.
4624 rcupdate.rcu_normal_after_boot= [KNL]
4625 Once boot has completed (that is, after
4626 rcu_end_inkernel_boot() has been invoked), use
4627 only normal grace-period primitives. No effect
4628 on CONFIG_TINY_RCU kernels.
4630 But note that CONFIG_PREEMPT_RT=y kernels enables
4631 this kernel boot parameter, forcibly setting
4632 it to the value one, that is, converting any
4633 post-boot attempt at an expedited RCU grace
4634 period to instead use normal non-expedited
4635 grace-period processing.
4637 rcupdate.rcu_task_ipi_delay= [KNL]
4638 Set time in jiffies during which RCU tasks will
4639 avoid sending IPIs, starting with the beginning
4640 of a given grace period. Setting a large
4641 number avoids disturbing real-time workloads,
4642 but lengthens grace periods.
4644 rcupdate.rcu_task_stall_timeout= [KNL]
4645 Set timeout in jiffies for RCU task stall warning
4646 messages. Disable with a value less than or equal
4649 rcupdate.rcu_self_test= [KNL]
4650 Run the RCU early boot self tests
4654 Run specified binary instead of /init from the ramdisk,
4655 used for early userspace startup. See initrd.
4658 force - Override the decision by the kernel to hide the
4659 advertisement of RDRAND support (this affects
4660 certain AMD processors because of buggy BIOS
4661 support, specifically around the suspend/resume
4665 Turn on/off individual RDT features. List is:
4666 cmt, mbmtotal, mbmlocal, l3cat, l3cdp, l2cat, l2cdp,
4668 E.g. to turn on cmt and turn off mba use:
4672 Format (x86 or x86_64):
4673 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] \
4675 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \
4677 Where reboot_mode is one of warm (soft) or cold (hard) or gpio
4678 (prefix with 'panic_' to set mode for panic
4680 reboot_type is one of bios, acpi, kbd, triple, efi, or pci,
4681 reboot_force is either force or not specified,
4682 reboot_cpu is s[mp]#### with #### being the processor
4683 to be used for rebooting.
4685 refscale.holdoff= [KNL]
4686 Set test-start holdoff period. The purpose of
4687 this parameter is to delay the start of the
4688 test until boot completes in order to avoid
4691 refscale.loops= [KNL]
4692 Set the number of loops over the synchronization
4693 primitive under test. Increasing this number
4694 reduces noise due to loop start/end overhead,
4695 but the default has already reduced the per-pass
4696 noise to a handful of picoseconds on ca. 2020
4699 refscale.nreaders= [KNL]
4700 Set number of readers. The default value of -1
4701 selects N, where N is roughly 75% of the number
4702 of CPUs. A value of zero is an interesting choice.
4704 refscale.nruns= [KNL]
4705 Set number of runs, each of which is dumped onto
4708 refscale.readdelay= [KNL]
4709 Set the read-side critical-section duration,
4710 measured in microseconds.
4712 refscale.scale_type= [KNL]
4713 Specify the read-protection implementation to test.
4715 refscale.shutdown= [KNL]
4716 Shut down the system at the end of the performance
4717 test. This defaults to 1 (shut it down) when
4718 refscale is built into the kernel and to 0 (leave
4719 it running) when refscale is built as a module.
4721 refscale.verbose= [KNL]
4722 Enable additional printk() statements.
4724 refscale.verbose_batched= [KNL]
4725 Batch the additional printk() statements. If zero
4726 (the default) or negative, print everything. Otherwise,
4727 print every Nth verbose statement, where N is the value
4731 [KNL, SMP] Set scheduler's default relax_domain_level.
4732 See Documentation/admin-guide/cgroup-v1/cpusets.rst.
4734 reserve= [KNL,BUGS] Force kernel to ignore I/O ports or memory
4735 Format: <base1>,<size1>[,<base2>,<size2>,...]
4736 Reserve I/O ports or memory so the kernel won't use
4737 them. If <base> is less than 0x10000, the region
4738 is assumed to be I/O ports; otherwise it is memory.
4740 reservetop= [X86-32]
4742 Reserves a hole at the top of the kernel virtual
4747 Set the amount of memory to reserve for BIOS at
4748 the bottom of the address space.
4750 reset_devices [KNL] Force drivers to reset the underlying device
4751 during initialization.
4754 Specify the partition device for software suspend
4756 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>}
4758 resume_offset= [SWSUSP]
4759 Specify the offset from the beginning of the partition
4760 given by "resume=" at which the swap header is located,
4761 in <PAGE_SIZE> units (needed only for swap files).
4762 See Documentation/power/swsusp-and-swap-files.rst
4764 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to
4765 read the resume files
4767 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up.
4768 Useful for devices that are detected asynchronously
4769 (e.g. USB and MMC devices).
4771 hibernate= [HIBERNATION]
4772 noresume Don't check if there's a hibernation image
4773 present during boot.
4774 nocompress Don't compress/decompress hibernation images.
4775 no Disable hibernation and resume.
4776 protect_image Turn on image protection during restoration
4777 (that will set all pages holding image data
4778 during restoration read-only).
4780 retain_initrd [RAM] Keep initrd memory after extraction
4782 rfkill.default_state=
4783 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm,
4784 etc. communication is blocked by default.
4787 rfkill.master_switch_mode=
4788 0 The "airplane mode" button does nothing.
4789 1 The "airplane mode" button toggles between everything
4790 blocked and the previous configuration.
4791 2 The "airplane mode" button toggles between everything
4792 blocked and everything unblocked.
4794 rhash_entries= [KNL,NET]
4795 Set number of hash buckets for route cache
4798 [KNL] Disable ring 3 MONITOR/MWAIT feature on supported
4801 ro [KNL] Mount root device read-only on boot
4804 on Mark read-only kernel memory as read-only (default).
4805 off Leave read-only kernel memory writable for debugging.
4808 Enable the uart passthrough on the designated usb port
4809 on Rockchip SoCs. When active, the signals of the
4810 debug-uart get routed to the D+ and D- pins of the usb
4811 port and the regular usb controller gets disabled.
4813 root= [KNL] Root filesystem
4814 See name_to_dev_t comment in init/do_mounts.c.
4816 rootdelay= [KNL] Delay (in seconds) to pause before attempting to
4817 mount the root filesystem
4819 rootflags= [KNL] Set root filesystem mount option string
4821 rootfstype= [KNL] Set root filesystem type
4823 rootwait [KNL] Wait (indefinitely) for root device to show up.
4824 Useful for devices that are detected asynchronously
4825 (e.g. USB and MMC devices).
4827 rproc_mem=nn[KMG][@address]
4828 [KNL,ARM,CMA] Remoteproc physical memory block.
4829 Memory area to be used by remote processor image,
4832 rw [KNL] Mount root device read-write on boot
4834 S [KNL] Run init in single mode
4836 s390_iommu= [HW,S390]
4837 Set s390 IOTLB flushing mode
4839 With strict flushing every unmap operation will result in
4840 an IOTLB flush. Default is lazy flushing before reuse,
4844 See drivers/net/irda/sa1100_ir.c.
4846 sbni= [NET] Granch SBNI12 leased line adapter
4848 sched_verbose [KNL] Enables verbose scheduler debug messages.
4850 schedstats= [KNL,X86] Enable or disable scheduled statistics.
4851 Allowed values are enable and disable. This feature
4852 incurs a small amount of overhead in the scheduler
4853 but is useful for debugging and performance tuning.
4855 sched_thermal_decay_shift=
4856 [KNL, SMP] Set a decay shift for scheduler thermal
4857 pressure signal. Thermal pressure signal follows the
4858 default decay period of other scheduler pelt
4859 signals(usually 32 ms but configurable). Setting
4860 sched_thermal_decay_shift will left shift the decay
4861 period for the thermal pressure signal by the shift
4863 i.e. with the default pelt decay period of 32 ms
4864 sched_thermal_decay_shift thermal pressure decay pr
4868 Format: integer between 0 and 10
4871 scftorture.holdoff= [KNL]
4872 Number of seconds to hold off before starting
4873 test. Defaults to zero for module insertion and
4874 to 10 seconds for built-in smp_call_function()
4877 scftorture.longwait= [KNL]
4878 Request ridiculously long waits randomly selected
4879 up to the chosen limit in seconds. Zero (the
4880 default) disables this feature. Please note
4881 that requesting even small non-zero numbers of
4882 seconds can result in RCU CPU stall warnings,
4883 softlockup complaints, and so on.
4885 scftorture.nthreads= [KNL]
4886 Number of kthreads to spawn to invoke the
4887 smp_call_function() family of functions.
4888 The default of -1 specifies a number of kthreads
4889 equal to the number of CPUs.
4891 scftorture.onoff_holdoff= [KNL]
4892 Number seconds to wait after the start of the
4893 test before initiating CPU-hotplug operations.
4895 scftorture.onoff_interval= [KNL]
4896 Number seconds to wait between successive
4897 CPU-hotplug operations. Specifying zero (which
4898 is the default) disables CPU-hotplug operations.
4900 scftorture.shutdown_secs= [KNL]
4901 The number of seconds following the start of the
4902 test after which to shut down the system. The
4903 default of zero avoids shutting down the system.
4904 Non-zero values are useful for automated tests.
4906 scftorture.stat_interval= [KNL]
4907 The number of seconds between outputting the
4908 current test statistics to the console. A value
4909 of zero disables statistics output.
4911 scftorture.stutter_cpus= [KNL]
4912 The number of jiffies to wait between each change
4913 to the set of CPUs under test.
4915 scftorture.use_cpus_read_lock= [KNL]
4916 Use use_cpus_read_lock() instead of the default
4917 preempt_disable() to disable CPU hotplug
4918 while invoking one of the smp_call_function*()
4921 scftorture.verbose= [KNL]
4922 Enable additional printk() statements.
4924 scftorture.weight_single= [KNL]
4925 The probability weighting to use for the
4926 smp_call_function_single() function with a zero
4927 "wait" parameter. A value of -1 selects the
4928 default if all other weights are -1. However,
4929 if at least one weight has some other value, a
4930 value of -1 will instead select a weight of zero.
4932 scftorture.weight_single_wait= [KNL]
4933 The probability weighting to use for the
4934 smp_call_function_single() function with a
4935 non-zero "wait" parameter. See weight_single.
4937 scftorture.weight_many= [KNL]
4938 The probability weighting to use for the
4939 smp_call_function_many() function with a zero
4940 "wait" parameter. See weight_single.
4941 Note well that setting a high probability for
4942 this weighting can place serious IPI load
4945 scftorture.weight_many_wait= [KNL]
4946 The probability weighting to use for the
4947 smp_call_function_many() function with a
4948 non-zero "wait" parameter. See weight_single
4951 scftorture.weight_all= [KNL]
4952 The probability weighting to use for the
4953 smp_call_function_all() function with a zero
4954 "wait" parameter. See weight_single and
4957 scftorture.weight_all_wait= [KNL]
4958 The probability weighting to use for the
4959 smp_call_function_all() function with a
4960 non-zero "wait" parameter. See weight_single
4963 skew_tick= [KNL] Offset the periodic timer tick per cpu to mitigate
4964 xtime_lock contention on larger systems, and/or RCU lock
4965 contention on all systems with CONFIG_MAXSMP set.
4966 Format: { "0" | "1" }
4967 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1"
4969 Note: increases power consumption, thus should only be
4970 enabled if running jitter sensitive (HPC/RT) workloads.
4972 security= [SECURITY] Choose a legacy "major" security module to
4973 enable at boot. This has been deprecated by the
4976 selinux= [SELINUX] Disable or enable SELinux at boot time.
4977 Format: { "0" | "1" }
4978 See security/selinux/Kconfig help text.
4983 apparmor= [APPARMOR] Disable or enable AppArmor at boot time
4984 Format: { "0" | "1" }
4985 See security/apparmor/Kconfig help text
4988 Default value is set via kernel config option.
4990 serialnumber [BUGS=X86-32]
4993 Maximal number of shapers.
5001 Enable merging of slabs with similar size when the
5002 kernel is built without CONFIG_SLAB_MERGE_DEFAULT.
5005 Disable merging of slabs with similar size. May be
5006 necessary if there is some reason to distinguish
5007 allocs to different slabs, especially in hardened
5008 environments where the risk of heap overflows and
5009 layout control by attackers can usually be
5010 frustrated by disabling merging. This will reduce
5011 most of the exposure of a heap attack to a single
5012 cache (risks via metadata attacks are mostly
5013 unchanged). Debug options disable merging on their
5015 For more information see Documentation/vm/slub.rst.
5017 slab_max_order= [MM, SLAB]
5018 Determines the maximum allowed order for slabs.
5019 A high setting may cause OOMs due to memory
5020 fragmentation. Defaults to 1 for systems with
5021 more than 32MB of RAM, 0 otherwise.
5023 slub_debug[=options[,slabs][;[options[,slabs]]...] [MM, SLUB]
5024 Enabling slub_debug allows one to determine the
5025 culprit if slab objects become corrupted. Enabling
5026 slub_debug can create guard zones around objects and
5027 may poison objects when not in use. Also tracks the
5028 last alloc / free. For more information see
5029 Documentation/vm/slub.rst.
5031 slub_max_order= [MM, SLUB]
5032 Determines the maximum allowed order for slabs.
5033 A high setting may cause OOMs due to memory
5034 fragmentation. For more information see
5035 Documentation/vm/slub.rst.
5037 slub_min_objects= [MM, SLUB]
5038 The minimum number of objects per slab. SLUB will
5039 increase the slab order up to slub_max_order to
5040 generate a sufficiently large slab able to contain
5041 the number of objects indicated. The higher the number
5042 of objects the smaller the overhead of tracking slabs
5043 and the less frequently locks need to be acquired.
5044 For more information see Documentation/vm/slub.rst.
5046 slub_min_order= [MM, SLUB]
5047 Determines the minimum page order for slabs. Must be
5048 lower than slub_max_order.
5049 For more information see Documentation/vm/slub.rst.
5051 slub_merge [MM, SLUB]
5052 Same with slab_merge.
5054 slub_nomerge [MM, SLUB]
5055 Same with slab_nomerge. This is supported for legacy.
5056 See slab_nomerge for more information.
5059 Format: <io1>[,<io2>[,...,<io8>]]
5061 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices
5062 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port
5063 smsc-ircc2.ircc_sir= [HW] SIR base I/O port
5064 smsc-ircc2.ircc_fir= [HW] FIR base I/O port
5065 smsc-ircc2.ircc_irq= [HW] IRQ line
5066 smsc-ircc2.ircc_dma= [HW] DMA channel
5067 smsc-ircc2.ircc_transceiver= [HW] Transceiver type:
5068 0: Toshiba Satellite 1800 (GP data pin select)
5069 1: Fast pin select (default)
5072 smt [KNL,S390] Set the maximum number of threads (logical
5073 CPUs) to use per physical CPU on systems capable of
5074 symmetric multithreading (SMT). Will be capped to the
5075 actual hardware limit.
5077 Default: -1 (no limit)
5080 [KNL] Should the soft-lockup detector generate panics.
5083 A value of 1 instructs the soft-lockup detector
5084 to panic the machine when a soft-lockup occurs. It is
5085 also controlled by the kernel.softlockup_panic sysctl
5086 and CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC, which is the
5087 respective build-time switch to that functionality.
5089 softlockup_all_cpu_backtrace=
5090 [KNL] Should the soft-lockup detector generate
5091 backtraces on all cpus.
5094 sonypi.*= [HW] Sony Programmable I/O Control Device driver
5095 See Documentation/admin-guide/laptops/sonypi.rst
5097 spectre_v2= [X86] Control mitigation of Spectre variant 2
5098 (indirect branch speculation) vulnerability.
5099 The default operation protects the kernel from
5102 on - unconditionally enable, implies
5104 off - unconditionally disable, implies
5106 auto - kernel detects whether your CPU model is
5109 Selecting 'on' will, and 'auto' may, choose a
5110 mitigation method at run time according to the
5111 CPU, the available microcode, the setting of the
5112 CONFIG_RETPOLINE configuration option, and the
5113 compiler with which the kernel was built.
5115 Selecting 'on' will also enable the mitigation
5116 against user space to user space task attacks.
5118 Selecting 'off' will disable both the kernel and
5119 the user space protections.
5121 Specific mitigations can also be selected manually:
5123 retpoline - replace indirect branches
5124 retpoline,generic - google's original retpoline
5125 retpoline,amd - AMD-specific minimal thunk
5127 Not specifying this option is equivalent to
5131 [X86] Control mitigation of Spectre variant 2
5132 (indirect branch speculation) vulnerability between
5135 on - Unconditionally enable mitigations. Is
5136 enforced by spectre_v2=on
5138 off - Unconditionally disable mitigations. Is
5139 enforced by spectre_v2=off
5141 prctl - Indirect branch speculation is enabled,
5142 but mitigation can be enabled via prctl
5143 per thread. The mitigation control state
5144 is inherited on fork.
5147 - Like "prctl" above, but only STIBP is
5148 controlled per thread. IBPB is issued
5149 always when switching between different user
5153 - Same as "prctl" above, but all seccomp
5154 threads will enable the mitigation unless
5155 they explicitly opt out.
5158 - Like "seccomp" above, but only STIBP is
5159 controlled per thread. IBPB is issued
5160 always when switching between different
5161 user space processes.
5163 auto - Kernel selects the mitigation depending on
5164 the available CPU features and vulnerability.
5167 If CONFIG_SECCOMP=y then "seccomp", otherwise "prctl"
5169 Not specifying this option is equivalent to
5170 spectre_v2_user=auto.
5172 spec_store_bypass_disable=
5173 [HW] Control Speculative Store Bypass (SSB) Disable mitigation
5174 (Speculative Store Bypass vulnerability)
5176 Certain CPUs are vulnerable to an exploit against a
5177 a common industry wide performance optimization known
5178 as "Speculative Store Bypass" in which recent stores
5179 to the same memory location may not be observed by
5180 later loads during speculative execution. The idea
5181 is that such stores are unlikely and that they can
5182 be detected prior to instruction retirement at the
5183 end of a particular speculation execution window.
5185 In vulnerable processors, the speculatively forwarded
5186 store can be used in a cache side channel attack, for
5187 example to read memory to which the attacker does not
5188 directly have access (e.g. inside sandboxed code).
5190 This parameter controls whether the Speculative Store
5191 Bypass optimization is used.
5193 On x86 the options are:
5195 on - Unconditionally disable Speculative Store Bypass
5196 off - Unconditionally enable Speculative Store Bypass
5197 auto - Kernel detects whether the CPU model contains an
5198 implementation of Speculative Store Bypass and
5199 picks the most appropriate mitigation. If the
5200 CPU is not vulnerable, "off" is selected. If the
5201 CPU is vulnerable the default mitigation is
5202 architecture and Kconfig dependent. See below.
5203 prctl - Control Speculative Store Bypass per thread
5204 via prctl. Speculative Store Bypass is enabled
5205 for a process by default. The state of the control
5206 is inherited on fork.
5207 seccomp - Same as "prctl" above, but all seccomp threads
5208 will disable SSB unless they explicitly opt out.
5210 Default mitigations:
5211 X86: If CONFIG_SECCOMP=y "seccomp", otherwise "prctl"
5213 On powerpc the options are:
5215 on,auto - On Power8 and Power9 insert a store-forwarding
5216 barrier on kernel entry and exit. On Power7
5217 perform a software flush on kernel entry and
5221 Not specifying this option is equivalent to
5222 spec_store_bypass_disable=auto.
5224 spia_io_base= [HW,MTD]
5230 [X86] Enable split lock detection or bus lock detection
5232 When enabled (and if hardware support is present), atomic
5233 instructions that access data across cache line
5234 boundaries will result in an alignment check exception
5235 for split lock detection or a debug exception for
5240 warn - the kernel will emit rate-limited warnings
5241 about applications triggering the #AC
5242 exception or the #DB exception. This mode is
5243 the default on CPUs that support split lock
5244 detection or bus lock detection. Default
5245 behavior is by #AC if both features are
5246 enabled in hardware.
5248 fatal - the kernel will send SIGBUS to applications
5249 that trigger the #AC exception or the #DB
5250 exception. Default behavior is by #AC if
5251 both features are enabled in hardware.
5253 If an #AC exception is hit in the kernel or in
5254 firmware (i.e. not while executing in user mode)
5255 the kernel will oops in either "warn" or "fatal"
5258 #DB exception for bus lock is triggered only when
5262 Control the Special Register Buffer Data Sampling
5265 Certain CPUs are vulnerable to an MDS-like
5266 exploit which can leak bits from the random
5269 By default, this issue is mitigated by
5270 microcode. However, the microcode fix can cause
5271 the RDRAND and RDSEED instructions to become
5272 much slower. Among other effects, this will
5273 result in reduced throughput from /dev/urandom.
5275 The microcode mitigation can be disabled with
5276 the following option:
5278 off: Disable mitigation and remove
5279 performance impact to RDRAND and RDSEED
5281 srcutree.counter_wrap_check [KNL]
5282 Specifies how frequently to check for
5283 grace-period sequence counter wrap for the
5284 srcu_data structure's ->srcu_gp_seq_needed field.
5285 The greater the number of bits set in this kernel
5286 parameter, the less frequently counter wrap will
5287 be checked for. Note that the bottom two bits
5290 srcutree.exp_holdoff [KNL]
5291 Specifies how many nanoseconds must elapse
5292 since the end of the last SRCU grace period for
5293 a given srcu_struct until the next normal SRCU
5294 grace period will be considered for automatic
5295 expediting. Set to zero to disable automatic
5299 Speculative Store Bypass Disable control
5301 On CPUs that are vulnerable to the Speculative
5302 Store Bypass vulnerability and offer a
5303 firmware based mitigation, this parameter
5304 indicates how the mitigation should be used:
5306 force-on: Unconditionally enable mitigation for
5307 for both kernel and userspace
5308 force-off: Unconditionally disable mitigation for
5309 for both kernel and userspace
5310 kernel: Always enable mitigation in the
5311 kernel, and offer a prctl interface
5312 to allow userspace to register its
5313 interest in being mitigated too.
5315 stack_guard_gap= [MM]
5316 override the default stack gap protection. The value
5317 is in page units and it defines how many pages prior
5318 to (for stacks growing down) resp. after (for stacks
5319 growing up) the main stack are reserved for no other
5320 mapping. Default value is 256 pages.
5322 stack_depot_disable= [KNL]
5323 Setting this to true through kernel command line will
5324 disable the stack depot thereby saving the static memory
5325 consumed by the stack hash table. By default this is set
5329 Enabled the stack tracer on boot up.
5331 stacktrace_filter=[function-list]
5332 [FTRACE] Limit the functions that the stack tracer
5333 will trace at boot up. function-list is a comma-separated
5334 list of functions. This list can be changed at run
5335 time by the stack_trace_filter file in the debugfs
5336 tracing directory. Note, this enables stack tracing
5337 and the stacktrace above is not needed.
5341 Set the STI (builtin display/keyboard on the HP-PARISC
5342 machines) console (graphic card) which should be used
5343 as the initial boot-console.
5344 See also comment in drivers/video/console/sticore.c.
5347 See comment in drivers/video/console/sticore.c.
5350 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]]
5352 sunrpc.min_resvport=
5353 sunrpc.max_resvport=
5355 SunRPC servers often require that client requests
5356 originate from a privileged port (i.e. a port in the
5357 range 0 < portnr < 1024).
5358 An administrator who wishes to reserve some of these
5359 ports for other uses may adjust the range that the
5360 kernel's sunrpc client considers to be privileged
5361 using these two parameters to set the minimum and
5362 maximum port values.
5364 sunrpc.svc_rpc_per_connection_limit=
5366 Limit the number of requests that the server will
5367 process in parallel from a single connection.
5368 The default value is 0 (no limit).
5372 Control how the NFS server code allocates CPUs to
5373 service thread pools. Depending on how many NICs
5374 you have and where their interrupts are bound, this
5375 option will affect which CPUs will do NFS serving.
5376 Note: this parameter cannot be changed while the
5377 NFS server is running.
5379 auto the server chooses an appropriate mode
5380 automatically using heuristics
5381 global a single global pool contains all CPUs
5382 percpu one pool for each CPU
5383 pernode one pool for each NUMA node (equivalent
5384 to global on non-NUMA machines)
5386 sunrpc.tcp_slot_table_entries=
5387 sunrpc.udp_slot_table_entries=
5389 Sets the upper limit on the number of simultaneous
5390 RPC calls that can be sent from the client to a
5391 server. Increasing these values may allow you to
5392 improve throughput, but will also increase the
5393 amount of memory reserved for use by the client.
5395 suspend.pm_test_delay=
5397 Sets the number of seconds to remain in a suspend test
5398 mode before resuming the system (see
5399 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG
5400 is set. Default value is 5.
5403 Format: { on | off | y | n | 1 | 0 }
5404 This parameter controls use of the Protected
5405 Execution Facility on pSeries.
5408 [KNL] Enable accounting of swap in memory resource
5409 controller if no parameter or 1 is given or disable
5410 it if 0 is given (See Documentation/admin-guide/cgroup-v1/memory.rst)
5412 swiotlb= [ARM,IA-64,PPC,MIPS,X86]
5413 Format: { <int> | force | noforce }
5414 <int> -- Number of I/O TLB slabs
5415 force -- force using of bounce buffers even if they
5416 wouldn't be automatically used by the kernel
5417 noforce -- Never use bounce buffers (for debugging)
5422 Set a sysctl parameter, right before loading the init
5423 process, as if the value was written to the respective
5424 /proc/sys/... file. Both '.' and '/' are recognized as
5425 separators. Unrecognized parameters and invalid values
5426 are reported in the kernel log. Sysctls registered
5427 later by a loaded module cannot be set this way.
5428 Example: sysctl.vm.swappiness=40
5430 sysfs.deprecated=0|1 [KNL]
5431 Enable/disable old style sysfs layout for old udev
5432 on older distributions. When this option is enabled
5433 very new udev will not work anymore. When this option
5434 is disabled (or CONFIG_SYSFS_DEPRECATED not compiled)
5435 in older udev will not work anymore.
5436 Default depends on CONFIG_SYSFS_DEPRECATED_V2 set in
5437 the kernel configuration.
5439 sysrq_always_enabled
5441 Ignore sysrq setting - this boot parameter will
5442 neutralize any effect of /proc/sys/kernel/sysrq.
5443 Useful for debugging.
5445 tcpmhash_entries= [KNL,NET]
5446 Set the number of tcp_metrics_hash slots.
5447 Default value is 8192 or 16384 depending on total
5448 ram pages. This is used to specify the TCP metrics
5449 cache size. See Documentation/networking/ip-sysctl.rst
5450 "tcp_no_metrics_save" section for more details.
5454 test_suspend= [SUSPEND][,N]
5455 Specify "mem" (for Suspend-to-RAM) or "standby" (for
5456 standby suspend) or "freeze" (for suspend type freeze)
5457 as the system sleep state during system startup with
5458 the optional capability to repeat N number of times.
5459 The system is woken from this state using a
5460 wakeup-capable RTC alarm.
5462 thash_entries= [KNL,NET]
5463 Set number of hash buckets for TCP connection
5465 thermal.act= [HW,ACPI]
5466 -1: disable all active trip points in all thermal zones
5467 <degrees C>: override all lowest active trip points
5469 thermal.crt= [HW,ACPI]
5470 -1: disable all critical trip points in all thermal zones
5471 <degrees C>: override all critical trip points
5473 thermal.nocrt= [HW,ACPI]
5474 Set to disable actions on ACPI thermal zone
5475 critical and hot trip points.
5477 thermal.off= [HW,ACPI]
5478 1: disable ACPI thermal control
5480 thermal.psv= [HW,ACPI]
5481 -1: disable all passive trip points
5482 <degrees C>: override all passive trip points to this
5485 thermal.tzp= [HW,ACPI]
5486 Specify global default ACPI thermal zone polling rate
5487 <deci-seconds>: poll all this frequency
5488 0: no polling (default)
5491 Force threading of all interrupt handlers except those
5492 marked explicitly IRQF_NO_THREAD.
5496 Specify if the kernel should make use of the cpu
5497 topology information if the hardware supports this.
5498 The scheduler will make use of this information and
5499 e.g. base its process migration decisions on it.
5502 topology_updates= [KNL, PPC, NUMA]
5504 Specify if the kernel should ignore (off)
5505 topology updates sent by the hypervisor to this
5508 torture.disable_onoff_at_boot= [KNL]
5509 Prevent the CPU-hotplug component of torturing
5510 until after init has spawned.
5512 torture.ftrace_dump_at_shutdown= [KNL]
5513 Dump the ftrace buffer at torture-test shutdown,
5514 even if there were no errors. This can be a
5515 very costly operation when many torture tests
5516 are running concurrently, especially on systems
5517 with rotating-rust storage.
5519 torture.verbose_sleep_frequency= [KNL]
5520 Specifies how many verbose printk()s should be
5521 emitted between each sleep. The default of zero
5522 disables verbose-printk() sleeping.
5524 torture.verbose_sleep_duration= [KNL]
5525 Duration of each verbose-printk() sleep in jiffies.
5529 tpm_suspend_pcr=[HW,TPM]
5530 Format: integer pcr id
5531 Specify that at suspend time, the tpm driver
5532 should extend the specified pcr with zeros,
5533 as a workaround for some chips which fail to
5534 flush the last written pcr on TPM_SaveState.
5535 This will guarantee that all the other pcrs
5538 trace_buf_size=nn[KMG]
5539 [FTRACE] will set tracing buffer size on each cpu.
5541 trace_event=[event-list]
5542 [FTRACE] Set and start specified trace events in order
5543 to facilitate early boot debugging. The event-list is a
5544 comma-separated list of trace events to enable. See
5545 also Documentation/trace/events.rst
5547 trace_options=[option-list]
5548 [FTRACE] Enable or disable tracer options at boot.
5549 The option-list is a comma delimited list of options
5550 that can be enabled or disabled just as if you were
5551 to echo the option name into
5553 /sys/kernel/debug/tracing/trace_options
5555 For example, to enable stacktrace option (to dump the
5556 stack trace of each event), add to the command line:
5558 trace_options=stacktrace
5560 See also Documentation/trace/ftrace.rst "trace options"
5564 Have the tracepoints sent to printk as well as the
5565 tracing ring buffer. This is useful for early boot up
5566 where the system hangs or reboots and does not give the
5567 option for reading the tracing buffer or performing a
5568 ftrace_dump_on_oops.
5570 To turn off having tracepoints sent to printk,
5571 echo 0 > /proc/sys/kernel/tracepoint_printk
5572 Note, echoing 1 into this file without the
5573 tracepoint_printk kernel cmdline option has no effect.
5577 Having tracepoints sent to printk() and activating high
5578 frequency tracepoints such as irq or sched, can cause
5579 the system to live lock.
5582 [FTRACE] enable this option to disable tracing when a
5583 warning is hit. This turns off "tracing_on". Tracing can
5584 be enabled again by echoing '1' into the "tracing_on"
5585 file located in /sys/kernel/debug/tracing/
5587 This option is useful, as it disables the trace before
5588 the WARNING dump is called, which prevents the trace to
5589 be filled with content caused by the warning output.
5591 This option can also be set at run time via the sysctl
5592 option: kernel/traceoff_on_warning
5594 transparent_hugepage=
5596 Format: [always|madvise|never]
5597 Can be used to control the default behavior of the system
5598 with respect to transparent hugepages.
5599 See Documentation/admin-guide/mm/transhuge.rst
5602 trusted.source= [KEYS]
5604 This parameter identifies the trust source as a backend
5605 for trusted keys implementation. Supported trust
5609 If not specified then it defaults to iterating through
5610 the trust source list starting with TPM and assigns the
5611 first trust source as a backend which is initialized
5612 successfully during iteration.
5614 tsc= Disable clocksource stability checks for TSC.
5616 [x86] reliable: mark tsc clocksource as reliable, this
5617 disables clocksource verification at runtime, as well
5618 as the stability checks done at bootup. Used to enable
5619 high-resolution timer mode on older hardware, and in
5620 virtualized environment.
5621 [x86] noirqtime: Do not use TSC to do irq accounting.
5622 Used to run time disable IRQ_TIME_ACCOUNTING on any
5623 platforms where RDTSC is slow and this accounting
5625 [x86] unstable: mark the TSC clocksource as unstable, this
5626 marks the TSC unconditionally unstable at bootup and
5627 avoids any further wobbles once the TSC watchdog notices.
5628 [x86] nowatchdog: disable clocksource watchdog. Used
5629 in situations with strict latency requirements (where
5630 interruptions from clocksource watchdog are not
5633 tsc_early_khz= [X86] Skip early TSC calibration and use the given
5634 value instead. Useful when the early TSC frequency discovery
5635 procedure is not reliable, such as on overclocked systems
5636 with CPUID.16h support and partial CPUID.15h support.
5637 Format: <unsigned int>
5639 tsx= [X86] Control Transactional Synchronization
5640 Extensions (TSX) feature in Intel processors that
5641 support TSX control.
5643 This parameter controls the TSX feature. The options are:
5645 on - Enable TSX on the system. Although there are
5646 mitigations for all known security vulnerabilities,
5647 TSX has been known to be an accelerator for
5648 several previous speculation-related CVEs, and
5649 so there may be unknown security risks associated
5650 with leaving it enabled.
5652 off - Disable TSX on the system. (Note that this
5653 option takes effect only on newer CPUs which are
5654 not vulnerable to MDS, i.e., have
5655 MSR_IA32_ARCH_CAPABILITIES.MDS_NO=1 and which get
5656 the new IA32_TSX_CTRL MSR through a microcode
5657 update. This new MSR allows for the reliable
5658 deactivation of the TSX functionality.)
5660 auto - Disable TSX if X86_BUG_TAA is present,
5661 otherwise enable TSX on the system.
5663 Not specifying this option is equivalent to tsx=off.
5665 See Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
5668 tsx_async_abort= [X86,INTEL] Control mitigation for the TSX Async
5669 Abort (TAA) vulnerability.
5671 Similar to Micro-architectural Data Sampling (MDS)
5672 certain CPUs that support Transactional
5673 Synchronization Extensions (TSX) are vulnerable to an
5674 exploit against CPU internal buffers which can forward
5675 information to a disclosure gadget under certain
5678 In vulnerable processors, the speculatively forwarded
5679 data can be used in a cache side channel attack, to
5680 access data to which the attacker does not have direct
5683 This parameter controls the TAA mitigation. The
5686 full - Enable TAA mitigation on vulnerable CPUs
5689 full,nosmt - Enable TAA mitigation and disable SMT on
5690 vulnerable CPUs. If TSX is disabled, SMT
5691 is not disabled because CPU is not
5692 vulnerable to cross-thread TAA attacks.
5693 off - Unconditionally disable TAA mitigation
5695 On MDS-affected machines, tsx_async_abort=off can be
5696 prevented by an active MDS mitigation as both vulnerabilities
5697 are mitigated with the same mechanism so in order to disable
5698 this mitigation, you need to specify mds=off too.
5700 Not specifying this option is equivalent to
5701 tsx_async_abort=full. On CPUs which are MDS affected
5702 and deploy MDS mitigation, TAA mitigation is not
5703 required and doesn't provide any additional
5707 Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
5709 turbografx.map[2|3]= [HW,JOY]
5710 TurboGraFX parallel port interface
5712 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7>
5713 See also Documentation/input/devices/joystick-parport.rst
5715 udbg-immortal [PPC] When debugging early kernel crashes that
5716 happen after console_init() and before a proper
5717 console driver takes over, this boot options might
5718 help "seeing" what's going on.
5720 uhash_entries= [KNL,NET]
5721 Set number of hash buckets for UDP/UDP-Lite connections
5724 [USB] Ignore overcurrent events (default N).
5725 Some badly-designed motherboards generate lots of
5726 bogus events, for ports that aren't wired to
5727 anything. Set this parameter to avoid log spamming.
5728 Note that genuine overcurrent events won't be
5732 [X86] Cause panic on unknown NMI.
5734 usbcore.authorized_default=
5735 [USB] Default USB device authorization:
5736 (default -1 = authorized except for wireless USB,
5737 0 = not authorized, 1 = authorized, 2 = authorized
5738 if device connected to internal port)
5740 usbcore.autosuspend=
5741 [USB] The autosuspend time delay (in seconds) used
5742 for newly-detected USB devices (default 2). This
5743 is the time required before an idle device will be
5744 autosuspended. Devices for which the delay is set
5745 to a negative value won't be autosuspended at all.
5747 usbcore.usbfs_snoop=
5748 [USB] Set to log all usbfs traffic (default 0 = off).
5750 usbcore.usbfs_snoop_max=
5751 [USB] Maximum number of bytes to snoop in each URB
5754 usbcore.blinkenlights=
5755 [USB] Set to cycle leds on hubs (default 0 = off).
5757 usbcore.old_scheme_first=
5758 [USB] Start with the old device initialization
5759 scheme (default 0 = off).
5761 usbcore.usbfs_memory_mb=
5762 [USB] Memory limit (in MB) for buffers allocated by
5763 usbfs (default = 16, 0 = max = 2047).
5765 usbcore.use_both_schemes=
5766 [USB] Try the other device initialization scheme
5767 if the first one fails (default 1 = enabled).
5769 usbcore.initial_descriptor_timeout=
5770 [USB] Specifies timeout for the initial 64-byte
5771 USB_REQ_GET_DESCRIPTOR request in milliseconds
5772 (default 5000 = 5.0 seconds).
5774 usbcore.nousb [USB] Disable the USB subsystem
5777 [USB] A list of quirk entries to augment the built-in
5778 usb core quirk list. List entries are separated by
5779 commas. Each entry has the form
5780 VendorID:ProductID:Flags. The IDs are 4-digit hex
5781 numbers and Flags is a set of letters. Each letter
5782 will change the built-in quirk; setting it if it is
5783 clear and clearing it if it is set. The letters have
5784 the following meanings:
5785 a = USB_QUIRK_STRING_FETCH_255 (string
5786 descriptors must not be fetched using
5788 b = USB_QUIRK_RESET_RESUME (device can't resume
5789 correctly so reset it instead);
5790 c = USB_QUIRK_NO_SET_INTF (device can't handle
5791 Set-Interface requests);
5792 d = USB_QUIRK_CONFIG_INTF_STRINGS (device can't
5793 handle its Configuration or Interface
5795 e = USB_QUIRK_RESET (device can't be reset
5796 (e.g morph devices), don't use reset);
5797 f = USB_QUIRK_HONOR_BNUMINTERFACES (device has
5798 more interface descriptions than the
5799 bNumInterfaces count, and can't handle
5800 talking to these interfaces);
5801 g = USB_QUIRK_DELAY_INIT (device needs a pause
5802 during initialization, after we read
5803 the device descriptor);
5804 h = USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL (For
5805 high speed and super speed interrupt
5806 endpoints, the USB 2.0 and USB 3.0 spec
5807 require the interval in microframes (1
5808 microframe = 125 microseconds) to be
5809 calculated as interval = 2 ^
5811 Devices with this quirk report their
5812 bInterval as the result of this
5813 calculation instead of the exponent
5814 variable used in the calculation);
5815 i = USB_QUIRK_DEVICE_QUALIFIER (device can't
5816 handle device_qualifier descriptor
5818 j = USB_QUIRK_IGNORE_REMOTE_WAKEUP (device
5819 generates spurious wakeup, ignore
5820 remote wakeup capability);
5821 k = USB_QUIRK_NO_LPM (device can't handle Link
5823 l = USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL
5824 (Device reports its bInterval as linear
5825 frames instead of the USB 2.0
5827 m = USB_QUIRK_DISCONNECT_SUSPEND (Device needs
5828 to be disconnected before suspend to
5829 prevent spurious wakeup);
5830 n = USB_QUIRK_DELAY_CTRL_MSG (Device needs a
5831 pause after every control message);
5832 o = USB_QUIRK_HUB_SLOW_RESET (Hub needs extra
5833 delay after resetting its port);
5834 Example: quirks=0781:5580:bk,0a5c:5834:gij
5837 [USBHID] The interval which mice are to be polled at.
5840 [USBHID] The interval which joysticks are to be polled at.
5843 [USBHID] The interval which keyboards are to be polled at.
5845 usb-storage.delay_use=
5846 [UMS] The delay in seconds before a new device is
5847 scanned for Logical Units (default 1).
5850 [UMS] A list of quirks entries to supplement or
5851 override the built-in unusual_devs list. List
5852 entries are separated by commas. Each entry has
5853 the form VID:PID:Flags where VID and PID are Vendor
5854 and Product ID values (4-digit hex numbers) and
5855 Flags is a set of characters, each corresponding
5856 to a common usb-storage quirk flag as follows:
5857 a = SANE_SENSE (collect more than 18 bytes
5858 of sense data, not on uas);
5859 b = BAD_SENSE (don't collect more than 18
5860 bytes of sense data, not on uas);
5861 c = FIX_CAPACITY (decrease the reported
5862 device capacity by one sector);
5863 d = NO_READ_DISC_INFO (don't use
5864 READ_DISC_INFO command, not on uas);
5865 e = NO_READ_CAPACITY_16 (don't use
5866 READ_CAPACITY_16 command);
5867 f = NO_REPORT_OPCODES (don't use report opcodes
5869 g = MAX_SECTORS_240 (don't transfer more than
5870 240 sectors at a time, uas only);
5871 h = CAPACITY_HEURISTICS (decrease the
5872 reported device capacity by one
5873 sector if the number is odd);
5874 i = IGNORE_DEVICE (don't bind to this
5876 j = NO_REPORT_LUNS (don't use report luns
5878 k = NO_SAME (do not use WRITE_SAME, uas only)
5879 l = NOT_LOCKABLE (don't try to lock and
5880 unlock ejectable media, not on uas);
5881 m = MAX_SECTORS_64 (don't transfer more
5882 than 64 sectors = 32 KB at a time,
5884 n = INITIAL_READ10 (force a retry of the
5885 initial READ(10) command, not on uas);
5886 o = CAPACITY_OK (accept the capacity
5887 reported by the device, not on uas);
5888 p = WRITE_CACHE (the device cache is ON
5889 by default, not on uas);
5890 r = IGNORE_RESIDUE (the device reports
5891 bogus residue values, not on uas);
5892 s = SINGLE_LUN (the device has only one
5894 t = NO_ATA_1X (don't allow ATA(12) and ATA(16)
5895 commands, uas only);
5896 u = IGNORE_UAS (don't bind to the uas driver);
5897 w = NO_WP_DETECT (don't test whether the
5898 medium is write-protected).
5899 y = ALWAYS_SYNC (issue a SYNCHRONIZE_CACHE
5900 even if the device claims no cache,
5902 Example: quirks=0419:aaf5:rl,0421:0433:rc
5904 user_debug= [KNL,ARM]
5906 See arch/arm/Kconfig.debug help text.
5907 1 - undefined instruction events
5909 4 - invalid data aborts
5912 Example: user_debug=31
5915 [X86] Flags controlling user PTE allocations.
5917 nohigh = do not allocate PTE pages in
5918 HIGHMEM regardless of setting
5922 On X86_32, this is an alias for vdso32=. Otherwise:
5924 vdso=1: enable VDSO (the default)
5925 vdso=0: disable VDSO mapping
5927 vdso32= [X86] Control the 32-bit vDSO
5928 vdso32=1: enable 32-bit VDSO
5929 vdso32=0 or vdso32=2: disable 32-bit VDSO
5931 See the help text for CONFIG_COMPAT_VDSO for more
5932 details. If CONFIG_COMPAT_VDSO is set, the default is
5933 vdso32=0; otherwise, the default is vdso32=1.
5935 For compatibility with older kernels, vdso32=2 is an
5938 Try vdso32=0 if you encounter an error that says:
5939 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed!
5942 vector=percpu: enable percpu vector domain
5944 video= [FB] Frame buffer configuration
5945 See Documentation/fb/modedb.rst.
5947 video.brightness_switch_enabled= [0,1]
5948 If set to 1, on receiving an ACPI notify event
5949 generated by hotkey, video driver will adjust brightness
5950 level and then send out the event to user space through
5951 the allocated input device; If set to 0, video driver
5952 will only send out the event without touching backlight
5957 [VMMIO] Memory mapped virtio (platform) device.
5959 <size>@<baseaddr>:<irq>[:<id>]
5961 <size> := size (can use standard suffixes
5963 <baseaddr> := physical base address
5964 <irq> := interrupt number (as passed to
5966 <id> := (optional) platform device id
5968 virtio_mmio.device=1K@0x100b0000:48:7
5970 Can be used multiple times for multiple devices.
5972 vga= [BOOT,X86-32] Select a particular video mode
5973 See Documentation/x86/boot.rst and
5974 Documentation/admin-guide/svga.rst.
5975 Use vga=ask for menu.
5976 This is actually a boot loader parameter; the value is
5977 passed to the kernel using a special protocol.
5979 vm_debug[=options] [KNL] Available with CONFIG_DEBUG_VM=y.
5980 May slow down system boot speed, especially when
5981 enabled on systems with a large amount of memory.
5982 All options are enabled by default, and this
5983 interface is meant to allow for selectively
5984 enabling or disabling specific virtual memory
5987 Available options are:
5988 P Enable page structure init time poisoning
5989 - Disable all of the above options
5991 vmalloc=nn[KMG] [KNL,BOOT] Forces the vmalloc area to have an exact
5992 size of <nn>. This can be used to increase the
5993 minimum size (128MB on x86). It can also be used to
5994 decrease the size and leave more room for directly
5997 vmcp_cma=nn[MG] [KNL,S390]
5998 Sets the memory size reserved for contiguous memory
5999 allocations for the vmcp device driver.
6001 vmhalt= [KNL,S390] Perform z/VM CP command after system halt.
6004 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic.
6007 vmpoff= [KNL,S390] Perform z/VM CP command after power off.
6011 Controls the behavior of vsyscalls (i.e. calls to
6012 fixed addresses of 0xffffffffff600x00 from legacy
6013 code). Most statically-linked binaries and older
6014 versions of glibc use these calls. Because these
6015 functions are at fixed addresses, they make nice
6016 targets for exploits that can control RIP.
6018 emulate [default] Vsyscalls turn into traps and are
6019 emulated reasonably safely. The vsyscall
6022 xonly Vsyscalls turn into traps and are
6023 emulated reasonably safely. The vsyscall
6024 page is not readable.
6026 none Vsyscalls don't work at all. This makes
6027 them quite hard to use for exploits but
6028 might break your system.
6030 vt.color= [VT] Default text color.
6031 Format: 0xYX, X = foreground, Y = background.
6032 Default: 0x07 = light gray on black.
6034 vt.cur_default= [VT] Default cursor shape.
6035 Format: 0xCCBBAA, where AA, BB, and CC are the same as
6036 the parameters of the <Esc>[?A;B;Cc escape sequence;
6037 see VGA-softcursor.txt. Default: 2 = underline.
6039 vt.default_blu= [VT]
6040 Format: <blue0>,<blue1>,<blue2>,...,<blue15>
6041 Change the default blue palette of the console.
6042 This is a 16-member array composed of values
6045 vt.default_grn= [VT]
6046 Format: <green0>,<green1>,<green2>,...,<green15>
6047 Change the default green palette of the console.
6048 This is a 16-member array composed of values
6051 vt.default_red= [VT]
6052 Format: <red0>,<red1>,<red2>,...,<red15>
6053 Change the default red palette of the console.
6054 This is a 16-member array composed of values
6060 Set system-wide default UTF-8 mode for all tty's.
6061 Default is 1, i.e. UTF-8 mode is enabled for all
6062 newly opened terminals.
6064 vt.global_cursor_default=
6067 Set system-wide default for whether a cursor
6068 is shown on new VTs. Default is -1,
6069 i.e. cursors will be created by default unless
6070 overridden by individual drivers. 0 will hide
6071 cursors, 1 will display them.
6073 vt.italic= [VT] Default color for italic text; 0-15.
6076 vt.underline= [VT] Default color for underlined text; 0-15.
6079 watchdog timers [HW,WDT] For information on watchdog timers,
6080 see Documentation/watchdog/watchdog-parameters.rst
6081 or other driver-specific files in the
6082 Documentation/watchdog/ directory.
6086 Set the hard lockup detector stall duration
6087 threshold in seconds. The soft lockup detector
6088 threshold is set to twice the value. A value of 0
6089 disables both lockup detectors. Default is 10
6092 workqueue.watchdog_thresh=
6093 If CONFIG_WQ_WATCHDOG is configured, workqueue can
6094 warn stall conditions and dump internal state to
6095 help debugging. 0 disables workqueue stall
6096 detection; otherwise, it's the stall threshold
6097 duration in seconds. The default value is 30 and
6098 it can be updated at runtime by writing to the
6099 corresponding sysfs file.
6101 workqueue.disable_numa
6102 By default, all work items queued to unbound
6103 workqueues are affine to the NUMA nodes they're
6104 issued on, which results in better behavior in
6105 general. If NUMA affinity needs to be disabled for
6106 whatever reason, this option can be used. Note
6107 that this also can be controlled per-workqueue for
6108 workqueues visible under /sys/bus/workqueue/.
6110 workqueue.power_efficient
6111 Per-cpu workqueues are generally preferred because
6112 they show better performance thanks to cache
6113 locality; unfortunately, per-cpu workqueues tend to
6114 be more power hungry than unbound workqueues.
6116 Enabling this makes the per-cpu workqueues which
6117 were observed to contribute significantly to power
6118 consumption unbound, leading to measurably lower
6119 power usage at the cost of small performance
6122 The default value of this parameter is determined by
6123 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT.
6125 workqueue.debug_force_rr_cpu
6126 Workqueue used to implicitly guarantee that work
6127 items queued without explicit CPU specified are put
6128 on the local CPU. This guarantee is no longer true
6129 and while local CPU is still preferred work items
6130 may be put on foreign CPUs. This debug option
6131 forces round-robin CPU selection to flush out
6132 usages which depend on the now broken guarantee.
6133 When enabled, memory and cache locality will be
6136 x2apic_phys [X86-64,APIC] Use x2apic physical mode instead of
6137 default x2apic cluster mode on platforms
6140 xen_512gb_limit [KNL,X86-64,XEN]
6141 Restricts the kernel running paravirtualized under Xen
6142 to use only up to 512 GB of RAM. The reason to do so is
6143 crash analysis tools and Xen tools for doing domain
6144 save/restore/migration must be enabled to handle larger
6147 xen_emul_unplug= [HW,X86,XEN]
6148 Unplug Xen emulated devices
6149 Format: [unplug0,][unplug1]
6150 ide-disks -- unplug primary master IDE devices
6151 aux-ide-disks -- unplug non-primary-master IDE devices
6152 nics -- unplug network devices
6153 all -- unplug all emulated devices (NICs and IDE disks)
6154 unnecessary -- unplugging emulated devices is
6155 unnecessary even if the host did not respond to
6157 never -- do not unplug even if version check succeeds
6159 xen_legacy_crash [X86,XEN]
6160 Crash from Xen panic notifier, without executing late
6161 panic() code such as dumping handler.
6163 xen_nopvspin [X86,XEN]
6164 Disables the qspinlock slowpath using Xen PV optimizations.
6165 This parameter is obsoleted by "nopvspin" parameter, which
6166 has equivalent effect for XEN platform.
6169 Disables the PV optimizations forcing the HVM guest to
6170 run as generic HVM guest with no PV drivers.
6171 This option is obsoleted by the "nopv" option, which
6172 has equivalent effect for XEN platform.
6174 xen_no_vector_callback
6175 [KNL,X86,XEN] Disable the vector callback for Xen
6176 event channel interrupts.
6178 xen_scrub_pages= [XEN]
6179 Boolean option to control scrubbing pages before giving them back
6180 to Xen, for use by other domains. Can be also changed at runtime
6181 with /sys/devices/system/xen_memory/xen_memory0/scrub_pages.
6182 Default value controlled with CONFIG_XEN_SCRUB_PAGES_DEFAULT.
6184 xen_timer_slop= [X86-64,XEN]
6185 Set the timer slop (in nanoseconds) for the virtual Xen
6186 timers (default is 100000). This adjusts the minimum
6187 delta of virtualized Xen timers, where lower values
6188 improve timer resolution at the expense of processing
6189 more timer interrupts.
6191 xen.event_eoi_delay= [XEN]
6192 How long to delay EOI handling in case of event
6193 storms (jiffies). Default is 10.
6195 xen.event_loop_timeout= [XEN]
6196 After which time (jiffies) the event handling loop
6197 should start to delay EOI handling. Default is 2.
6199 xen.fifo_events= [XEN]
6200 Boolean parameter to disable using fifo event handling
6201 even if available. Normally fifo event handling is
6202 preferred over the 2-level event handling, as it is
6203 fairer and the number of possible event channels is
6204 much higher. Default is on (use fifo events).
6206 nopv= [X86,XEN,KVM,HYPER_V,VMWARE]
6207 Disables the PV optimizations forcing the guest to run
6208 as generic guest with no PV drivers. Currently support
6209 XEN HVM, KVM, HYPER_V and VMWARE guest.
6211 nopvspin [X86,XEN,KVM]
6212 Disables the qspinlock slow path using PV optimizations
6213 which allow the hypervisor to 'idle' the guest on lock
6216 xirc2ps_cs= [NET,PCMCIA]
6218 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]]
6221 By default on POWER9 and above, the kernel will
6222 natively use the XIVE interrupt controller. This option
6223 allows the fallback firmware mode to be used:
6225 off Fallback to firmware control of XIVE interrupt
6226 controller on both pseries and powernv
6227 platforms. Only useful on POWER9 and above.
6229 xhci-hcd.quirks [USB,KNL]
6230 A hex value specifying bitmask with supplemental xhci
6231 host controller quirks. Meaning of each bit can be
6232 consulted in header drivers/usb/host/xhci.h.
6235 Format: { early | on | rw | ro | off }
6236 Controls if xmon debugger is enabled. Default is off.
6237 Passing only "xmon" is equivalent to "xmon=early".
6238 early Call xmon as early as possible on boot; xmon
6239 debugger is called from setup_arch().
6240 on xmon debugger hooks will be installed so xmon
6241 is only called on a kernel crash. Default mode,
6242 i.e. either "ro" or "rw" mode, is controlled
6243 with CONFIG_XMON_DEFAULT_RO_MODE.
6244 rw xmon debugger hooks will be installed so xmon
6245 is called only on a kernel crash, mode is write,
6246 meaning SPR registers, memory and, other data
6247 can be written using xmon commands.
6248 ro same as "rw" option above but SPR registers,
6249 memory, and other data can't be written using
6251 off xmon is disabled.