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]
27 If set to vendor, prefer vendor specific driver
28 (e.g. thinkpad_acpi, sony_acpi, etc.) instead
29 of the ACPI video.ko driver.
31 acpi_force_32bit_fadt_addr
32 force FADT to use 32 bit addresses rather than the
33 64 bit X_* addresses. Some firmware have broken 64
34 bit addresses for force ACPI ignore these and use
35 the older legacy 32 bit addresses.
37 acpica_no_return_repair [HW, ACPI]
38 Disable AML predefined validation mechanism
39 This mechanism can repair the evaluation result to make
40 the return objects more ACPI specification compliant.
41 This option is useful for developers to identify the
42 root cause of an AML interpreter issue when the issue
43 has something to do with the repair mechanism.
45 acpi.debug_layer= [HW,ACPI,ACPI_DEBUG]
46 acpi.debug_level= [HW,ACPI,ACPI_DEBUG]
48 CONFIG_ACPI_DEBUG must be enabled to produce any ACPI
49 debug output. Bits in debug_layer correspond to a
50 _COMPONENT in an ACPI source file, e.g.,
51 #define _COMPONENT ACPI_PCI_COMPONENT
52 Bits in debug_level correspond to a level in
53 ACPI_DEBUG_PRINT statements, e.g.,
54 ACPI_DEBUG_PRINT((ACPI_DB_INFO, ...
55 The debug_level mask defaults to "info". See
56 Documentation/firmware-guide/acpi/debug.rst for more information about
57 debug layers and levels.
59 Enable processor driver info messages:
60 acpi.debug_layer=0x20000000
61 Enable PCI/PCI interrupt routing info messages:
62 acpi.debug_layer=0x400000
63 Enable AML "Debug" output, i.e., stores to the Debug
64 object while interpreting AML:
65 acpi.debug_layer=0xffffffff acpi.debug_level=0x2
66 Enable all messages related to ACPI hardware:
67 acpi.debug_layer=0x2 acpi.debug_level=0xffffffff
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_rsdp= [ACPI,EFI,KEXEC]
140 Pass the RSDP address to the kernel, mostly used
141 on machines running EFI runtime service to boot the
142 second kernel for kdump.
144 acpi_os_name= [HW,ACPI] Tell ACPI BIOS the name of the OS
145 Format: To spoof as Windows 98: ="Microsoft Windows"
147 acpi_rev_override [ACPI] Override the _REV object to return 5 (instead
148 of 2 which is mandated by ACPI 6) as the supported ACPI
149 specification revision (when using this switch, it may
150 be necessary to carry out a cold reboot _twice_ in a
151 row to make it take effect on the platform firmware).
153 acpi_osi= [HW,ACPI] Modify list of supported OS interface strings
154 acpi_osi="string1" # add string1
155 acpi_osi="!string2" # remove string2
156 acpi_osi=!* # remove all strings
157 acpi_osi=! # disable all built-in OS vendor
159 acpi_osi=!! # enable all built-in OS vendor
161 acpi_osi= # disable all strings
163 'acpi_osi=!' can be used in combination with single or
164 multiple 'acpi_osi="string1"' to support specific OS
165 vendor string(s). Note that such command can only
166 affect the default state of the OS vendor strings, thus
167 it cannot affect the default state of the feature group
168 strings and the current state of the OS vendor strings,
169 specifying it multiple times through kernel command line
170 is meaningless. This command is useful when one do not
171 care about the state of the feature group strings which
172 should be controlled by the OSPM.
174 1. 'acpi_osi=! acpi_osi="Windows 2000"' is equivalent
175 to 'acpi_osi="Windows 2000" acpi_osi=!', they all
176 can make '_OSI("Windows 2000")' TRUE.
178 'acpi_osi=' cannot be used in combination with other
179 'acpi_osi=' command lines, the _OSI method will not
180 exist in the ACPI namespace. NOTE that such command can
181 only affect the _OSI support state, thus specifying it
182 multiple times through kernel command line is also
185 1. 'acpi_osi=' can make 'CondRefOf(_OSI, Local1)'
188 'acpi_osi=!*' can be used in combination with single or
189 multiple 'acpi_osi="string1"' to support specific
190 string(s). Note that such command can affect the
191 current state of both the OS vendor strings and the
192 feature group strings, thus specifying it multiple times
193 through kernel command line is meaningful. But it may
194 still not able to affect the final state of a string if
195 there are quirks related to this string. This command
196 is useful when one want to control the state of the
197 feature group strings to debug BIOS issues related to
200 1. 'acpi_osi="Module Device" acpi_osi=!*' can make
201 '_OSI("Module Device")' FALSE.
202 2. 'acpi_osi=!* acpi_osi="Module Device"' can make
203 '_OSI("Module Device")' TRUE.
204 3. 'acpi_osi=! acpi_osi=!* acpi_osi="Windows 2000"' is
206 'acpi_osi=!* acpi_osi=! acpi_osi="Windows 2000"'
208 'acpi_osi=!* acpi_osi="Windows 2000" acpi_osi=!',
209 they all will make '_OSI("Windows 2000")' TRUE.
212 Override the pmtimer bug detection: force the kernel
213 to assume that this machine's pmtimer latches its value
214 and always returns good values.
216 acpi_sci= [HW,ACPI] ACPI System Control Interrupt trigger mode
217 Format: { level | edge | high | low }
219 acpi_skip_timer_override [HW,ACPI]
220 Recognize and ignore IRQ0/pin2 Interrupt Override.
221 For broken nForce2 BIOS resulting in XT-PIC timer.
223 acpi_sleep= [HW,ACPI] Sleep options
224 Format: { s3_bios, s3_mode, s3_beep, s4_nohwsig,
225 old_ordering, nonvs, sci_force_enable, nobl }
226 See Documentation/power/video.rst for information on
228 s3_beep is for debugging; it makes the PC's speaker beep
229 as soon as the kernel's real-mode entry point is called.
230 s4_nohwsig prevents ACPI hardware signature from being
231 used during resume from hibernation.
232 old_ordering causes the ACPI 1.0 ordering of the _PTS
233 control method, with respect to putting devices into
234 low power states, to be enforced (the ACPI 2.0 ordering
235 of _PTS is used by default).
236 nonvs prevents the kernel from saving/restoring the
237 ACPI NVS memory during suspend/hibernation and resume.
238 sci_force_enable causes the kernel to set SCI_EN directly
239 on resume from S1/S3 (which is against the ACPI spec,
240 but some broken systems don't work without it).
241 nobl causes the internal blacklist of systems known to
242 behave incorrectly in some ways with respect to system
243 suspend and resume to be ignored (use wisely).
245 acpi_use_timer_override [HW,ACPI]
246 Use timer override. For some broken Nvidia NF5 boards
247 that require a timer override, but don't have HPET
249 add_efi_memmap [EFI; X86] Include EFI memory map in
250 kernel's map of available physical RAM.
253 { off | try_unsupported }
254 off: disable AGP support
255 try_unsupported: try to drive unsupported chipsets
256 (may crash computer or cause data corruption)
259 See Documentation/sound/alsa-configuration.rst
262 Allow the default userspace alignment fault handler
263 behaviour to be specified. Bit 0 enables warnings,
264 bit 1 enables fixups, and bit 2 sends a segfault.
266 align_va_addr= [X86-64]
267 Align virtual addresses by clearing slice [14:12] when
268 allocating a VMA at process creation time. This option
269 gives you up to 3% performance improvement on AMD F15h
270 machines (where it is enabled by default) for a
271 CPU-intensive style benchmark, and it can vary highly in
272 a microbenchmark depending on workload and compiler.
274 32: only for 32-bit processes
275 64: only for 64-bit processes
276 on: enable for both 32- and 64-bit processes
277 off: disable for both 32- and 64-bit processes
279 alloc_snapshot [FTRACE]
280 Allocate the ftrace snapshot buffer on boot up when the
281 main buffer is allocated. This is handy if debugging
282 and you need to use tracing_snapshot() on boot up, and
283 do not want to use tracing_snapshot_alloc() as it needs
284 to be done where GFP_KERNEL allocations are allowed.
286 amd_iommu= [HW,X86-64]
287 Pass parameters to the AMD IOMMU driver in the system.
289 fullflush - enable flushing of IO/TLB entries when
290 they are unmapped. Otherwise they are
291 flushed before they will be reused, which
293 off - do not initialize any AMD IOMMU found in
295 force_isolation - Force device isolation for all
296 devices. The IOMMU driver is not
297 allowed anymore to lift isolation
298 requirements as needed. This option
299 does not override iommu=pt
301 amd_iommu_dump= [HW,X86-64]
302 Enable AMD IOMMU driver option to dump the ACPI table
303 for AMD IOMMU. With this option enabled, AMD IOMMU
304 driver will print ACPI tables for AMD IOMMU during
305 IOMMU initialization.
307 amd_iommu_intr= [HW,X86-64]
308 Specifies one of the following AMD IOMMU interrupt
310 legacy - Use legacy interrupt remapping mode.
311 vapic - Use virtual APIC mode, which allows IOMMU
312 to inject interrupts directly into guest.
313 This mode requires kvm-amd.avic=1.
314 (Default when IOMMU HW support is present.)
316 amijoy.map= [HW,JOY] Amiga joystick support
317 Map of devices attached to JOY0DAT and JOY1DAT
319 See also Documentation/input/joydev/joystick.rst
321 analog.map= [HW,JOY] Analog joystick and gamepad support
322 Specifies type or capabilities of an analog joystick
323 connected to one of 16 gameports
324 Format: <type1>,<type2>,..<type16>
327 Power management functions (SPARCstation-4/5 + deriv.)
329 Disable APC CPU standby support. SPARCstation-Fox does
330 not play well with APC CPU idle - disable it if you have
331 APC and your system crashes randomly.
333 apic= [APIC,X86] Advanced Programmable Interrupt Controller
334 Change the output verbosity while booting
335 Format: { quiet (default) | verbose | debug }
336 Change the amount of debugging information output
337 when initialising the APIC and IO-APIC components.
338 For X86-32, this can also be used to specify an APIC
340 Format: apic=driver_name
341 Examples: apic=bigsmp
343 apic_extnmi= [APIC,X86] External NMI delivery setting
344 Format: { bsp (default) | all | none }
345 bsp: External NMI is delivered only to CPU 0
346 all: External NMIs are broadcast to all CPUs as a
348 none: External NMI is masked for all CPUs. This is
349 useful so that a dump capture kernel won't be
353 See Documentation/networking/ipv6.txt.
355 show_lapic= [APIC,X86] Advanced Programmable Interrupt Controller
356 Limit apic dumping. The parameter defines the maximal
357 number of local apics being dumped. Also it is possible
358 to set it to "all" by meaning -- no limit here.
359 Format: { 1 (default) | 2 | ... | all }.
360 The parameter valid if only apic=debug or
361 apic=verbose is specified.
362 Example: apic=debug show_lapic=all
364 apm= [APM] Advanced Power Management
365 See header of arch/x86/kernel/apm_32.c.
367 arcrimi= [HW,NET] ARCnet - "RIM I" (entirely mem-mapped) cards
368 Format: <io>,<irq>,<nodeID>
372 atarimouse= [HW,MOUSE] Atari Mouse
374 atkbd.extra= [HW] Enable extra LEDs and keys on IBM RapidAccess,
375 EzKey and similar keyboards
377 atkbd.reset= [HW] Reset keyboard during initialization
379 atkbd.set= [HW] Select keyboard code set
380 Format: <int> (2 = AT (default), 3 = PS/2)
382 atkbd.scroll= [HW] Enable scroll wheel on MS Office and similar
385 atkbd.softraw= [HW] Choose between synthetic and real raw mode
386 Format: <bool> (0 = real, 1 = synthetic (default))
388 atkbd.softrepeat= [HW]
389 Use software keyboard repeat
391 audit= [KNL] Enable the audit sub-system
392 Format: { "0" | "1" | "off" | "on" }
393 0 | off - kernel audit is disabled and can not be
394 enabled until the next reboot
395 unset - kernel audit is initialized but disabled and
396 will be fully enabled by the userspace auditd.
397 1 | on - kernel audit is initialized and partially
398 enabled, storing at most audit_backlog_limit
399 messages in RAM until it is fully enabled by the
403 audit_backlog_limit= [KNL] Set the audit queue size limit.
404 Format: <int> (must be >=0)
407 bau= [X86_UV] Enable the BAU on SGI UV. The default
408 behavior is to disable the BAU (i.e. bau=0).
409 Format: { "0" | "1" }
412 unset - Disable the BAU.
414 baycom_epp= [HW,AX25]
417 baycom_par= [HW,AX25] BayCom Parallel Port AX.25 Modem
419 See header of drivers/net/hamradio/baycom_par.c.
421 baycom_ser_fdx= [HW,AX25]
422 BayCom Serial Port AX.25 Modem (Full Duplex Mode)
423 Format: <io>,<irq>,<mode>[,<baud>]
424 See header of drivers/net/hamradio/baycom_ser_fdx.c.
426 baycom_ser_hdx= [HW,AX25]
427 BayCom Serial Port AX.25 Modem (Half Duplex Mode)
428 Format: <io>,<irq>,<mode>
429 See header of drivers/net/hamradio/baycom_ser_hdx.c.
431 blkdevparts= Manual partition parsing of block device(s) for
432 embedded devices based on command line input.
433 See Documentation/block/cmdline-partition.rst
435 boot_delay= Milliseconds to delay each printk during boot.
436 Values larger than 10 seconds (10000) are changed to
440 bootmem_debug [KNL] Enable bootmem allocator debug messages.
443 Disable BERT OS support on buggy BIOSes.
445 bttv.card= [HW,V4L] bttv (bt848 + bt878 based grabber cards)
446 bttv.radio= Most important insmod options are available as
448 bttv.pll= See Documentation/media/v4l-drivers/bttv.rst
451 bulk_remove=off [PPC] This parameter disables the use of the pSeries
452 firmware feature for flushing multiple hpte entries
455 c101= [NET] Moxa C101 synchronous serial card
457 cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection.
458 Sometimes CPU hardware bugs make them report the cache
459 size incorrectly. The kernel will attempt work arounds
460 to fix known problems, but for some CPUs it is not
461 possible to determine what the correct size should be.
462 This option provides an override for these situations.
465 [NET] Specifies amount of time (in seconds) that
466 the kernel should wait for a network carrier. By default
467 it waits 120 seconds.
469 ca_keys= [KEYS] This parameter identifies a specific key(s) on
470 the system trusted keyring to be used for certificate
472 format: { id:<keyid> | builtin }
474 cca= [MIPS] Override the kernel pages' cache coherency
475 algorithm. Accepted values range from 0 to 7
476 inclusive. See arch/mips/include/asm/pgtable-bits.h
477 for platform specific values (SB1, Loongson3 and
480 ccw_timeout_log [S390]
481 See Documentation/s390/common_io.rst for details.
483 cgroup_disable= [KNL] Disable a particular controller
484 Format: {name of the controller(s) to disable}
485 The effects of cgroup_disable=foo are:
486 - foo isn't auto-mounted if you mount all cgroups in
488 - foo isn't visible as an individually mountable
490 {Currently only "memory" controller deal with this and
491 cut the overhead, others just disable the usage. So
492 only cgroup_disable=memory is actually worthy}
494 cgroup_no_v1= [KNL] Disable cgroup controllers and named hierarchies in v1
495 Format: { { controller | "all" | "named" }
496 [,{ controller | "all" | "named" }...] }
497 Like cgroup_disable, but only applies to cgroup v1;
498 the blacklisted controllers remain available in cgroup2.
499 "all" blacklists all controllers and "named" disables
500 named mounts. Specifying both "all" and "named" disables
503 cgroup.memory= [KNL] Pass options to the cgroup memory controller.
505 nosocket -- Disable socket memory accounting.
506 nokmem -- Disable kernel memory accounting.
508 checkreqprot [SELINUX] Set initial checkreqprot flag value.
509 Format: { "0" | "1" }
510 See security/selinux/Kconfig help text.
511 0 -- check protection applied by kernel (includes
512 any implied execute protection).
513 1 -- check protection requested by application.
514 Default value is set via a kernel config option.
515 Value can be changed at runtime via
516 /selinux/checkreqprot.
519 See Documentation/s390/common_io.rst for details.
522 Prevents the clock framework from automatically gating
523 clocks that have not been explicitly enabled by a Linux
524 device driver but are enabled in hardware at reset or
525 by the bootloader/firmware. Note that this does not
526 force such clocks to be always-on nor does it reserve
527 those clocks in any way. This parameter is useful for
528 debug and development, but should not be needed on a
529 platform with proper driver support. For more
530 information, see Documentation/driver-api/clk.rst.
532 clock= [BUGS=X86-32, HW] gettimeofday clocksource override.
534 Forces specified clocksource (if available) to be used
535 when calculating gettimeofday(). If specified
536 clocksource is not available, it defaults to PIT.
537 Format: { pit | tsc | cyclone | pmtmr }
539 clocksource= Override the default clocksource
541 Override the default clocksource and use the clocksource
542 with the name specified.
543 Some clocksource names to choose from, depending on
545 [all] jiffies (this is the base, fallback clocksource)
547 [ARM] imx_timer1,OSTS,netx_timer,mpu_timer2,
548 pxa_timer,timer3,32k_counter,timer0_1
549 [X86-32] pit,hpet,tsc;
550 scx200_hrt on Geode; cyclone on IBM x440
558 clocksource.arm_arch_timer.evtstrm=
561 Enable/disable the eventstream feature of the ARM
562 architected timer so that code using WFE-based polling
563 loops can be debugged more effectively on production
566 clearcpuid=BITNUM [X86]
567 Disable CPUID feature X for the kernel. See
568 arch/x86/include/asm/cpufeatures.h for the valid bit
569 numbers. Note the Linux specific bits are not necessarily
570 stable over kernel options, but the vendor specific
572 Also note that user programs calling CPUID directly
573 or using the feature without checking anything
574 will still see it. This just prevents it from
575 being used by the kernel or shown in /proc/cpuinfo.
576 Also note the kernel might malfunction if you disable
579 cma=nn[MG]@[start[MG][-end[MG]]]
581 Sets the size of kernel global memory area for
582 contiguous memory allocations and optionally the
583 placement constraint by the physical address range of
584 memory allocations. A value of 0 disables CMA
585 altogether. For more information, see
586 include/linux/dma-contiguous.h
588 cmo_free_hint= [PPC] Format: { yes | no }
589 Specify whether pages are marked as being inactive
590 when they are freed. This is used in CMO environments
591 to determine OS memory pressure for page stealing by
595 coherent_pool=nn[KMG] [ARM,KNL]
596 Sets the size of memory pool for coherent, atomic dma
597 allocations, by default set to 256K.
599 com20020= [HW,NET] ARCnet - COM20020 chipset
601 <io>[,<irq>[,<nodeID>[,<backplane>[,<ckp>[,<timeout>]]]]]
603 com90io= [HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers)
607 ARCnet - COM90xx chipset (memory-mapped buffers)
608 Format: <io>[,<irq>[,<memstart>]]
610 condev= [HW,S390] console device
613 console= [KNL] Output console device and options.
615 tty<n> Use the virtual console device <n>.
619 Use the specified serial port. The options are of
620 the form "bbbbpnf", where "bbbb" is the baud rate,
621 "p" is parity ("n", "o", or "e"), "n" is number of
622 bits, and "f" is flow control ("r" for RTS or
623 omit it). Default is "9600n8".
625 See Documentation/admin-guide/serial-console.rst for more
627 Documentation/networking/netconsole.txt for an
630 uart[8250],io,<addr>[,options]
631 uart[8250],mmio,<addr>[,options]
632 uart[8250],mmio16,<addr>[,options]
633 uart[8250],mmio32,<addr>[,options]
634 uart[8250],0x<addr>[,options]
635 Start an early, polled-mode console on the 8250/16550
636 UART at the specified I/O port or MMIO address,
637 switching to the matching ttyS device later.
638 MMIO inter-register address stride is either 8-bit
639 (mmio), 16-bit (mmio16), or 32-bit (mmio32).
640 If none of [io|mmio|mmio16|mmio32], <addr> is assumed
641 to be equivalent to 'mmio'. 'options' are specified in
642 the same format described for ttyS above; if unspecified,
643 the h/w is not re-initialized.
645 hvc<n> Use the hypervisor console device <n>. This is for
646 both Xen and PowerPC hypervisors.
648 If the device connected to the port is not a TTY but a braille
649 device, prepend "brl," before the device type, for instance
651 For now, only VisioBraille is supported.
654 [KNL] Change console messages format
656 By default we print messages on consoles in
657 "[time stamp] text\n" format (time stamp may not be
658 printed, depending on CONFIG_PRINTK_TIME or
659 `printk_time' param).
661 Switch to syslog format: "<%u>[time stamp] text\n"
662 IOW, each message will have a facility and loglevel
663 prefix. The format is similar to one used by syslog()
664 syscall, or to executing "dmesg -S --raw" or to reading
667 consoleblank= [KNL] The console blank (screen saver) timeout in
668 seconds. A value of 0 disables the blank timer.
672 [KNL] Change the default value for
673 /proc/<pid>/coredump_filter.
674 See also Documentation/filesystems/proc.txt.
676 coresight_cpu_debug.enable
679 Enable/disable the CPU sampling based debugging.
680 0: default value, disable debugging
681 1: enable debugging at boot time
683 cpuidle.off=1 [CPU_IDLE]
684 disable the cpuidle sub-system
687 [CPU_IDLE] Name of the cpuidle governor to use.
689 cpufreq.off=1 [CPU_FREQ]
690 disable the cpufreq sub-system
693 [X86] Delay for N microsec between assert and de-assert
694 of APIC INIT to start processors. This delay occurs
695 on every CPU online, such as boot, and resume from suspend.
698 cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver
700 <first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>]
702 crashkernel=size[KMG][@offset[KMG]]
703 [KNL] Using kexec, Linux can switch to a 'crash kernel'
704 upon panic. This parameter reserves the physical
705 memory region [offset, offset + size] for that kernel
706 image. If '@offset' is omitted, then a suitable offset
707 is selected automatically.
708 [KNL, x86_64] select a region under 4G first, and
709 fall back to reserve region above 4G when '@offset'
710 hasn't been specified.
711 See Documentation/admin-guide/kdump/kdump.rst for further details.
713 crashkernel=range1:size1[,range2:size2,...][@offset]
714 [KNL] Same as above, but depends on the memory
715 in the running system. The syntax of range is
716 start-[end] where start and end are both
717 a memory unit (amount[KMG]). See also
718 Documentation/admin-guide/kdump/kdump.rst for an example.
720 crashkernel=size[KMG],high
721 [KNL, x86_64] range could be above 4G. Allow kernel
722 to allocate physical memory region from top, so could
723 be above 4G if system have more than 4G ram installed.
724 Otherwise memory region will be allocated below 4G, if
726 It will be ignored if crashkernel=X is specified.
727 crashkernel=size[KMG],low
728 [KNL, x86_64] range under 4G. When crashkernel=X,high
729 is passed, kernel could allocate physical memory region
730 above 4G, that cause second kernel crash on system
731 that require some amount of low memory, e.g. swiotlb
732 requires at least 64M+32K low memory, also enough extra
733 low memory is needed to make sure DMA buffers for 32-bit
734 devices won't run out. Kernel would try to allocate at
735 at least 256M below 4G automatically.
736 This one let user to specify own low range under 4G
737 for second kernel instead.
738 0: to disable low allocation.
739 It will be ignored when crashkernel=X,high is not used
740 or memory reserved is below 4G.
743 [KNL] Disable crypto self-tests
748 cs89x0_media= [HW,NET]
749 Format: { rj45 | aui | bnc }
752 See header of drivers/s390/block/dasd_devmap.c.
754 db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port
755 (one device per port)
756 Format: <port#>,<type>
757 See also Documentation/input/devices/joystick-parport.rst
759 ddebug_query= [KNL,DYNAMIC_DEBUG] Enable debug messages at early boot
761 Documentation/admin-guide/dynamic-debug-howto.rst for
762 details. Deprecated, see dyndbg.
764 debug [KNL] Enable kernel debugging (events log level).
767 [KNL] Enable printing [hashed] pointers early in the
768 boot sequence. If enabled, we use a weak hash instead
769 of siphash to hash pointers. Use this option if you are
770 seeing instances of '(___ptrval___)') and need to see a
771 value (hashed pointer) instead. Cryptographically
772 insecure, please do not use on production kernels.
775 [KNL] verbose self-tests
777 Print debugging info while doing the locking API
779 We default to 0 (no extra messages), setting it to
780 1 will print _a lot_ more information - normally
781 only useful to kernel developers.
783 debug_objects [KNL] Enable object debugging
786 [KNL] Disable object debugging
788 debug_guardpage_minorder=
789 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this
790 parameter allows control of the order of pages that will
791 be intentionally kept free (and hence protected) by the
792 buddy allocator. Bigger value increase the probability
793 of catching random memory corruption, but reduce the
794 amount of memory for normal system use. The maximum
795 possible value is MAX_ORDER/2. Setting this parameter
796 to 1 or 2 should be enough to identify most random
797 memory corruption problems caused by bugs in kernel or
798 driver code when a CPU writes to (or reads from) a
799 random memory location. Note that there exists a class
800 of memory corruptions problems caused by buggy H/W or
801 F/W or by drivers badly programing DMA (basically when
802 memory is written at bus level and the CPU MMU is
803 bypassed) which are not detectable by
804 CONFIG_DEBUG_PAGEALLOC, hence this option will not help
805 tracking down these problems.
808 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this parameter
809 enables the feature at boot time. By default, it is
810 disabled and the system will work mostly the same as a
811 kernel built without CONFIG_DEBUG_PAGEALLOC.
812 on: enable the feature
814 debugpat [X86] Enable PAT debugging
816 decnet.addr= [HW,NET]
817 Format: <area>[,<node>]
818 See also Documentation/networking/decnet.txt.
821 [same as hugepagesz=] The size of the default
822 HugeTLB page size. This is the size represented by
823 the legacy /proc/ hugepages APIs, used for SHM, and
824 default size when mounting hugetlbfs filesystems.
825 Defaults to the default architecture's huge page size
828 deferred_probe_timeout=
829 [KNL] Debugging option to set a timeout in seconds for
830 deferred probe to give up waiting on dependencies to
831 probe. Only specific dependencies (subsystems or
832 drivers) that have opted in will be ignored. A timeout of 0
833 will timeout at the end of initcalls. This option will also
834 dump out devices still on the deferred probe list after
838 Set number of hash buckets for dentry cache.
840 disable_1tb_segments [PPC]
841 Disables the use of 1TB hash page table segments. This
842 causes the kernel to fall back to 256MB segments which
843 can be useful when debugging issues that require an SLB
847 See Documentation/networking/ipv6.txt.
850 [KNL] Under CONFIG_HARDENED_USERCOPY, whether
851 hardening is enabled for this boot. Hardened
852 usercopy checking is used to protect the kernel
853 from reading or writing beyond known memory
854 allocation boundaries as a proactive defense
855 against bounds-checking flaws in the kernel's
856 copy_to_user()/copy_from_user() interface.
857 on Perform hardened usercopy checks (default).
858 off Disable hardened usercopy checks.
861 Disable RADIX MMU mode on POWER9
863 disable_cpu_apicid= [X86,APIC,SMP]
865 The number of initial APIC ID for the
866 corresponding CPU to be disabled at boot,
867 mostly used for the kdump 2nd kernel to
868 disable BSP to wake up multiple CPUs without
869 causing system reset or hang due to sending
872 perf_v4_pmi= [X86,INTEL]
874 Disable Intel PMU counter freezing feature.
875 The feature only exists starting from
876 Arch Perfmon v4 (Skylake and newer).
878 disable_ddw [PPC/PSERIES]
879 Disable Dynamic DMA Window support. Use this if
880 to workaround buggy firmware.
883 See Documentation/networking/ipv6.txt.
885 disable_mtrr_cleanup [X86]
886 The kernel tries to adjust MTRR layout from continuous
887 to discrete, to make X server driver able to add WB
888 entry later. This parameter disables that.
890 disable_mtrr_trim [X86, Intel and AMD only]
891 By default the kernel will trim any uncacheable
892 memory out of your available memory pool based on
893 MTRR settings. This parameter disables that behavior,
894 possibly causing your machine to run very slowly.
896 disable_timer_pin_1 [X86]
897 Disable PIN 1 of APIC timer
898 Can be useful to work around chipset bugs.
900 dis_ucode_ldr [X86] Disable the microcode loader.
902 dma_debug=off If the kernel is compiled with DMA_API_DEBUG support,
903 this option disables the debugging code at boot.
905 dma_debug_entries=<number>
906 This option allows to tune the number of preallocated
907 entries for DMA-API debugging code. One entry is
908 required per DMA-API allocation. Use this if the
909 DMA-API debugging code disables itself because the
910 architectural default is too low.
912 dma_debug_driver=<driver_name>
913 With this option the DMA-API debugging driver
914 filter feature can be enabled at boot time. Just
915 pass the driver to filter for as the parameter.
916 The filter can be disabled or changed to another
917 driver later using sysfs.
919 driver_async_probe= [KNL]
920 List of driver names to be probed asynchronously.
921 Format: <driver_name1>,<driver_name2>...
923 drm.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>]
924 Broken monitors, graphic adapters, KVMs and EDIDless
925 panels may send no or incorrect EDID data sets.
926 This parameter allows to specify an EDID data sets
927 in the /lib/firmware directory that are used instead.
928 Generic built-in EDID data sets are used, if one of
929 edid/1024x768.bin, edid/1280x1024.bin,
930 edid/1680x1050.bin, or edid/1920x1080.bin is given
931 and no file with the same name exists. Details and
932 instructions how to build your own EDID data are
933 available in Documentation/driver-api/edid.rst. An EDID
934 data set will only be used for a particular connector,
935 if its name and a colon are prepended to the EDID
936 name. Each connector may use a unique EDID data
937 set by separating the files with a comma. An EDID
938 data set with no connector name will be used for
939 any connectors not explicitly specified.
944 Format: {"off" | "known"}
945 Control how the dt_cpu_ftrs device-tree binding is
946 used for CPU feature discovery and setup (if it
948 off: Do not use it, fall back to legacy cpu table.
949 known: Do not pass through unknown features to guests
950 or userspace, only those that the kernel is aware of.
952 dump_apple_properties [X86]
953 Dump name and content of EFI device properties on
954 x86 Macs. Useful for driver authors to determine
955 what data is available or for reverse-engineering.
957 dyndbg[="val"] [KNL,DYNAMIC_DEBUG]
958 module.dyndbg[="val"]
959 Enable debug messages at boot time. See
960 Documentation/admin-guide/dynamic-debug-howto.rst
963 nompx [X86] Disables Intel Memory Protection Extensions.
964 See Documentation/x86/intel_mpx.rst for more
965 information about the feature.
967 nopku [X86] Disable Memory Protection Keys CPU feature found
970 module.async_probe [KNL]
971 Enable asynchronous probe on this module.
973 early_ioremap_debug [KNL]
974 Enable debug messages in early_ioremap support. This
975 is useful for tracking down temporary early mappings
976 which are not unmapped.
978 earlycon= [KNL] Output early console device and options.
980 [ARM64] The early console is determined by the
981 stdout-path property in device tree's chosen node,
982 or determined by the ACPI SPCR table.
984 [X86] When used with no options the early console is
985 determined by the ACPI SPCR table.
987 cdns,<addr>[,options]
988 Start an early, polled-mode console on a Cadence
989 (xuartps) serial port at the specified address. Only
990 supported option is baud rate. If baud rate is not
991 specified, the serial port must already be setup and
994 uart[8250],io,<addr>[,options]
995 uart[8250],mmio,<addr>[,options]
996 uart[8250],mmio32,<addr>[,options]
997 uart[8250],mmio32be,<addr>[,options]
998 uart[8250],0x<addr>[,options]
999 Start an early, polled-mode console on the 8250/16550
1000 UART at the specified I/O port or MMIO address.
1001 MMIO inter-register address stride is either 8-bit
1002 (mmio) or 32-bit (mmio32 or mmio32be).
1003 If none of [io|mmio|mmio32|mmio32be], <addr> is assumed
1004 to be equivalent to 'mmio'. 'options' are specified
1005 in the same format described for "console=ttyS<n>"; if
1006 unspecified, the h/w is not initialized.
1010 Start an early, polled-mode console on a pl011 serial
1011 port at the specified address. The pl011 serial port
1012 must already be setup and configured. Options are not
1013 yet supported. If 'mmio32' is specified, then only
1014 the driver will use only 32-bit accessors to read/write
1015 the device registers.
1018 Start an early, polled-mode console on a meson serial
1019 port at the specified address. The serial port must
1020 already be setup and configured. Options are not yet
1024 Start an early, polled-mode console on an msm serial
1025 port at the specified address. The serial port
1026 must already be setup and configured. Options are not
1029 msm_serial_dm,<addr>
1030 Start an early, polled-mode console on an msm serial
1031 dm port at the specified address. The serial port
1032 must already be setup and configured. Options are not
1036 Start an early, polled-mode console on a serial port
1037 of an Actions Semi SoC, such as S500 or S900, at the
1038 specified address. The serial port must already be
1039 setup and configured. Options are not yet supported.
1042 Start an early, polled-mode console on a serial port
1043 of an RDA Micro SoC, such as RDA8810PL, at the
1044 specified address. The serial port must already be
1045 setup and configured. Options are not yet supported.
1048 Use RISC-V SBI (Supervisor Binary Interface) for early
1051 smh Use ARM semihosting calls for early console.
1059 Use early console provided by serial driver available
1060 on Samsung SoCs, requires selecting proper type and
1061 a correct base address of the selected UART port. The
1062 serial port must already be setup and configured.
1063 Options are not yet supported.
1066 Start an early, polled-mode console on a lantiq serial
1067 (lqasc) port at the specified address. The serial port
1068 must already be setup and configured. Options are not
1073 Use early console provided by Freescale LP UART driver
1074 found on Freescale Vybrid and QorIQ LS1021A processors.
1075 A valid base address must be provided, and the serial
1076 port must already be setup and configured.
1079 Start an early, polled-mode console on the
1080 Armada 3700 serial port at the specified
1081 address. The serial port must already be setup
1082 and configured. Options are not yet supported.
1085 Start an early, polled-mode console on a Qualcomm
1086 Generic Interface (GENI) based serial port at the
1087 specified address. The serial port must already be
1088 setup and configured. Options are not yet supported.
1091 Start an early, unaccelerated console on the EFI
1092 memory mapped framebuffer (if available). On cache
1093 coherent non-x86 systems that use system memory for
1094 the framebuffer, pass the 'ram' option so that it is
1095 mapped with the correct attributes.
1097 earlyprintk= [X86,SH,ARM,M68k,S390]
1101 earlyprintk=serial[,ttySn[,baudrate]]
1102 earlyprintk=serial[,0x...[,baudrate]]
1103 earlyprintk=ttySn[,baudrate]
1104 earlyprintk=dbgp[debugController#]
1105 earlyprintk=pciserial[,force],bus:device.function[,baudrate]
1106 earlyprintk=xdbc[xhciController#]
1108 earlyprintk is useful when the kernel crashes before
1109 the normal console is initialized. It is not enabled by
1110 default because it has some cosmetic problems.
1112 Append ",keep" to not disable it when the real console
1115 Only one of vga, efi, serial, or usb debug port can
1118 Currently only ttyS0 and ttyS1 may be specified by
1119 name. Other I/O ports may be explicitly specified
1120 on some architectures (x86 and arm at least) by
1121 replacing ttySn with an I/O port address, like this:
1122 earlyprintk=serial,0x1008,115200
1123 You can find the port for a given device in
1124 /proc/tty/driver/serial:
1125 2: uart:ST16650V2 port:00001008 irq:18 ...
1127 Interaction with the standard serial driver is not
1130 The VGA and EFI output is eventually overwritten by
1133 The xen output can only be used by Xen PV guests.
1135 The sclp output can only be used on s390.
1137 The optional "force" to "pciserial" enables use of a
1138 PCI device even when its classcode is not of the
1141 edac_report= [HW,EDAC] Control how to report EDAC event
1142 Format: {"on" | "off" | "force"}
1143 on: enable EDAC to report H/W event. May be overridden
1144 by other higher priority error reporting module.
1145 off: disable H/W event reporting through EDAC.
1146 force: enforce the use of EDAC to report H/W event.
1149 ekgdboc= [X86,KGDB] Allow early kernel console debugging
1152 This is designed to be used in conjunction with
1153 the boot argument: earlyprintk=vga
1156 Format: {"off" | "on" | "skip[mbr]"}
1159 Format: { "old_map", "nochunk", "noruntime", "debug" }
1160 old_map [X86-64]: switch to the old ioremap-based EFI
1161 runtime services mapping. 32-bit still uses this one by
1163 nochunk: disable reading files in "chunks" in the EFI
1164 boot stub, as chunking can cause problems with some
1165 firmware implementations.
1166 noruntime : disable EFI runtime services support
1167 debug: enable misc debug output
1169 efi_no_storage_paranoia [EFI; X86]
1170 Using this parameter you can use more than 50% of
1171 your efi variable storage. Use this parameter only if
1172 you are really sure that your UEFI does sane gc and
1173 fulfills the spec otherwise your board may brick.
1175 efi_fake_mem= nn[KMG]@ss[KMG]:aa[,nn[KMG]@ss[KMG]:aa,..] [EFI; X86]
1176 Add arbitrary attribute to specific memory range by
1177 updating original EFI memory map.
1178 Region of memory which aa attribute is added to is
1180 If efi_fake_mem=2G@4G:0x10000,2G@0x10a0000000:0x10000
1181 is specified, EFI_MEMORY_MORE_RELIABLE(0x10000)
1182 attribute is added to range 0x100000000-0x180000000 and
1183 0x10a0000000-0x1120000000.
1185 Using this parameter you can do debugging of EFI memmap
1186 related feature. For example, you can do debugging of
1187 Address Range Mirroring feature even if your box
1190 efivar_ssdt= [EFI; X86] Name of an EFI variable that contains an SSDT
1191 that is to be dynamically loaded by Linux. If there are
1192 multiple variables with the same name but with different
1193 vendor GUIDs, all of them will be loaded. See
1194 Documentation/admin-guide/acpi/ssdt-overlays.rst for details.
1197 eisa_irq_edge= [PARISC,HW]
1198 See header of drivers/parisc/eisa.c.
1201 See comment before function elanfreq_setup() in
1202 arch/x86/kernel/cpu/cpufreq/elanfreq.c.
1205 Format: { "mq-deadline" | "kyber" | "bfq" }
1206 See Documentation/block/deadline-iosched.rst,
1207 Documentation/block/kyber-iosched.rst and
1208 Documentation/block/bfq-iosched.rst for details.
1210 elfcorehdr=[size[KMG]@]offset[KMG] [IA64,PPC,SH,X86,S390]
1211 Specifies physical address of start of kernel core
1212 image elf header and optionally the size. Generally
1213 kexec loader will pass this option to capture kernel.
1214 See Documentation/admin-guide/kdump/kdump.rst for details.
1216 enable_mtrr_cleanup [X86]
1217 The kernel tries to adjust MTRR layout from continuous
1218 to discrete, to make X server driver able to add WB
1219 entry later. This parameter enables that.
1221 enable_timer_pin_1 [X86]
1222 Enable PIN 1 of APIC timer
1223 Can be useful to work around chipset bugs
1224 (in particular on some ATI chipsets).
1225 The kernel tries to set a reasonable default.
1227 enforcing [SELINUX] Set initial enforcing status.
1229 See security/selinux/Kconfig help text.
1230 0 -- permissive (log only, no denials).
1231 1 -- enforcing (deny and log).
1233 Value can be changed at runtime via /selinux/enforce.
1236 Disable Error Record Serialization Table (ERST)
1239 ether= [HW,NET] Ethernet cards parameters
1240 This option is obsoleted by the "netdev=" option, which
1241 has equivalent usage. See its documentation for details.
1245 Permit 'security.evm' to be updated regardless of
1246 current integrity status.
1250 fail_make_request=[KNL]
1251 General fault injection mechanism.
1252 Format: <interval>,<probability>,<space>,<times>
1253 See also Documentation/fault-injection/.
1256 See Documentation/admin-guide/blockdev/floppy.rst.
1258 force_pal_cache_flush
1259 [IA-64] Avoid check_sal_cache_flush which may hang on
1260 buggy SAL_CACHE_FLUSH implementations. Using this
1261 parameter will force ia64_sal_cache_flush to call
1262 ia64_pal_cache_flush instead of SAL_CACHE_FLUSH.
1265 Forcefully enable Physical Address Extension (PAE).
1266 Many Pentium M systems disable PAE but may have a
1267 functionally usable PAE implementation.
1268 Warning: use of this parameter will taint the kernel
1269 and may cause unknown problems.
1272 [FTRACE] will set and start the specified tracer
1273 as early as possible in order to facilitate early
1276 ftrace_dump_on_oops[=orig_cpu]
1277 [FTRACE] will dump the trace buffers on oops.
1278 If no parameter is passed, ftrace will dump
1279 buffers of all CPUs, but if you pass orig_cpu, it will
1280 dump only the buffer of the CPU that triggered the
1283 ftrace_filter=[function-list]
1284 [FTRACE] Limit the functions traced by the function
1285 tracer at boot up. function-list is a comma separated
1286 list of functions. This list can be changed at run
1287 time by the set_ftrace_filter file in the debugfs
1290 ftrace_notrace=[function-list]
1291 [FTRACE] Do not trace the functions specified in
1292 function-list. This list can be changed at run time
1293 by the set_ftrace_notrace file in the debugfs
1296 ftrace_graph_filter=[function-list]
1297 [FTRACE] Limit the top level callers functions traced
1298 by the function graph tracer at boot up.
1299 function-list is a comma separated list of functions
1300 that can be changed at run time by the
1301 set_graph_function file in the debugfs tracing directory.
1303 ftrace_graph_notrace=[function-list]
1304 [FTRACE] Do not trace from the functions specified in
1305 function-list. This list is a comma separated list of
1306 functions that can be changed at run time by the
1307 set_graph_notrace file in the debugfs tracing directory.
1309 ftrace_graph_max_depth=<uint>
1310 [FTRACE] Used with the function graph tracer. This is
1311 the max depth it will trace into a function. This value
1312 can be changed at run time by the max_graph_depth file
1313 in the tracefs tracing directory. default: 0 (no limit)
1316 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad
1317 support via parallel port (up to 5 devices per port)
1318 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5>
1319 See also Documentation/input/devices/joystick-parport.rst
1323 gart_fix_e820= [X86_64] disable the fix e820 for K8 GART
1327 gcov_persist= [GCOV] When non-zero (default), profiling data for
1328 kernel modules is saved and remains accessible via
1329 debugfs, even when the module is unloaded/reloaded.
1330 When zero, profiling data is discarded and associated
1331 debugfs files are removed at module unload time.
1333 goldfish [X86] Enable the goldfish android emulator platform.
1334 Don't use this when you are not running on the
1337 gpt [EFI] Forces disk with valid GPT signature but
1338 invalid Protective MBR to be treated as GPT. If the
1339 primary GPT is corrupted, it enables the backup/alternate
1340 GPT to be used instead.
1342 grcan.enable0= [HW] Configuration of physical interface 0. Determines
1343 the "Enable 0" bit of the configuration register.
1346 grcan.enable1= [HW] Configuration of physical interface 1. Determines
1347 the "Enable 0" bit of the configuration register.
1350 grcan.select= [HW] Select which physical interface to use.
1353 grcan.txsize= [HW] Sets the size of the tx buffer.
1354 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0.
1356 grcan.rxsize= [HW] Sets the size of the rx buffer.
1357 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0.
1360 gpio-mockup.gpio_mockup_ranges
1361 [HW] Sets the ranges of gpiochip of for this device.
1362 Format: <start1>,<end1>,<start2>,<end2>...
1364 hardlockup_all_cpu_backtrace=
1365 [KNL] Should the hard-lockup detector generate
1366 backtraces on all cpus.
1369 hashdist= [KNL,NUMA] Large hashes allocated during boot
1370 are distributed across NUMA nodes. Defaults on
1371 for 64-bit NUMA, off otherwise.
1372 Format: 0 | 1 (for off | on)
1374 hcl= [IA-64] SGI's Hardware Graph compatibility layer
1376 hd= [EIDE] (E)IDE hard drive subsystem geometry
1377 Format: <cyl>,<head>,<sect>
1380 Disable Hardware Error Source Table (HEST) support;
1381 corresponding firmware-first mode error processing
1382 logic will be disabled.
1384 highmem=nn[KMG] [KNL,BOOT] forces the highmem zone to have an exact
1385 size of <nn>. This works even on boxes that have no
1386 highmem otherwise. This also works to reduce highmem
1387 size on bigger boxes.
1389 highres= [KNL] Enable/disable high resolution timer mode.
1390 Valid parameters: "on", "off"
1395 hpet= [X86-32,HPET] option to control HPET usage
1396 Format: { enable (default) | disable | force |
1398 disable: disable HPET and use PIT instead
1399 force: allow force enabled of undocumented chips (ICH4,
1401 verbose: show contents of HPET registers during setup
1403 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET
1404 registers. Default set by CONFIG_HPET_MMAP_DEFAULT.
1406 hugepages= [HW,X86-32,IA-64] HugeTLB pages to allocate at boot.
1407 hugepagesz= [HW,IA-64,PPC,X86-64] The size of the HugeTLB pages.
1408 On x86-64 and powerpc, this option can be specified
1409 multiple times interleaved with hugepages= to reserve
1410 huge pages of different sizes. Valid pages sizes on
1411 x86-64 are 2M (when the CPU supports "pse") and 1G
1412 (when the CPU supports the "pdpe1gb" cpuinfo flag).
1415 [KNL] Should the hung task detector generate panics.
1418 A nonzero value instructs the kernel to panic when a
1419 hung task is detected. The default value is controlled
1420 by the CONFIG_BOOTPARAM_HUNG_TASK_PANIC build-time
1421 option. The value selected by this boot parameter can
1422 be changed later by the kernel.hung_task_panic sysctl.
1424 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC)
1425 terminal devices. Valid values: 0..8
1426 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs.
1427 If specified, z/VM IUCV HVC accepts connections
1428 from listed z/VM user IDs only.
1430 hv_nopvspin [X86,HYPER_V] Disables the paravirt spinlock optimizations
1431 which allow the hypervisor to 'idle' the
1432 guest on lock contention.
1435 Do not unregister boot console at start. This is only
1436 useful for debugging when something happens in the window
1437 between unregistering the boot console and initializing
1440 i2c_bus= [HW] Override the default board specific I2C bus speed
1441 or register an additional I2C bus that is not
1442 registered from board initialization code.
1446 i8042.debug [HW] Toggle i8042 debug mode
1447 i8042.unmask_kbd_data
1448 [HW] Enable printing of interrupt data from the KBD port
1449 (disabled by default, and as a pre-condition
1450 requires that i8042.debug=1 be enabled)
1451 i8042.direct [HW] Put keyboard port into non-translated mode
1452 i8042.dumbkbd [HW] Pretend that controller can only read data from
1453 keyboard and cannot control its state
1454 (Don't attempt to blink the leds)
1455 i8042.noaux [HW] Don't check for auxiliary (== mouse) port
1456 i8042.nokbd [HW] Don't check/create keyboard port
1457 i8042.noloop [HW] Disable the AUX Loopback command while probing
1459 i8042.nomux [HW] Don't check presence of an active multiplexing
1461 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX
1463 i8042.notimeout [HW] Ignore timeout condition signalled by controller
1464 i8042.reset [HW] Reset the controller during init, cleanup and
1465 suspend-to-ram transitions, only during s2r
1466 transitions, or never reset
1467 Format: { 1 | Y | y | 0 | N | n }
1468 1, Y, y: always reset controller
1469 0, N, n: don't ever reset controller
1470 Default: only on s2r transitions on x86; most other
1471 architectures force reset to be always executed
1472 i8042.unlock [HW] Unlock (ignore) the keylock
1473 i8042.kbdreset [HW] Reset device connected to KBD port
1477 i8k.ignore_dmi [HW] Continue probing hardware even if DMI data
1478 indicates that the driver is running on unsupported
1480 i8k.force [HW] Activate i8k driver even if SMM BIOS signature
1481 does not match list of supported models.
1483 [HW] Report power status in /proc/i8k
1484 (disabled by default)
1485 i8k.restricted [HW] Allow controlling fans only if SYS_ADMIN
1488 i915.invert_brightness=
1489 [DRM] Invert the sense of the variable that is used to
1490 set the brightness of the panel backlight. Normally a
1491 brightness value of 0 indicates backlight switched off,
1492 and the maximum of the brightness value sets the backlight
1493 to maximum brightness. If this parameter is set to 0
1494 (default) and the machine requires it, or this parameter
1495 is set to 1, a brightness value of 0 sets the backlight
1496 to maximum brightness, and the maximum of the brightness
1497 value switches the backlight off.
1498 -1 -- never invert brightness
1499 0 -- machine default
1500 1 -- force brightness inversion
1503 Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]]
1505 ide-core.nodma= [HW] (E)IDE subsystem
1506 Format: =0.0 to prevent dma on hda, =0.1 hdb =1.0 hdc
1507 .vlb_clock .pci_clock .noflush .nohpa .noprobe .nowerr
1508 .cdrom .chs .ignore_cable are additional options
1509 See Documentation/ide/ide.rst.
1511 ide-generic.probe-mask= [HW] (E)IDE subsystem
1513 Probe mask for legacy ISA IDE ports. Depending on
1514 platform up to 6 ports are supported, enabled by
1515 setting corresponding bits in the mask to 1. The
1516 default value is 0x0, which has a special meaning.
1517 On systems that have PCI, it triggers scanning the
1518 PCI bus for the first and the second port, which
1519 are then probed. On systems without PCI the value
1520 of 0x0 enables probing the two first ports as if it
1523 ide-pci-generic.all-generic-ide [HW] (E)IDE subsystem
1524 Claim all unknown PCI IDE storage controllers.
1527 Format: idle=poll, idle=halt, idle=nomwait
1528 Poll forces a polling idle loop that can slightly
1529 improve the performance of waking up a idle CPU, but
1530 will use a lot of power and make the system run hot.
1532 idle=halt: Halt is forced to be used for CPU idle.
1533 In such case C2/C3 won't be used again.
1534 idle=nomwait: Disable mwait for CPU C-states
1536 ieee754= [MIPS] Select IEEE Std 754 conformance mode
1537 Format: { strict | legacy | 2008 | relaxed }
1540 Choose which programs will be accepted for execution
1541 based on the IEEE 754 NaN encoding(s) supported by
1542 the FPU and the NaN encoding requested with the value
1543 of an ELF file header flag individually set by each
1544 binary. Hardware implementations are permitted to
1545 support either or both of the legacy and the 2008 NaN
1548 Available settings are as follows:
1549 strict accept binaries that request a NaN encoding
1550 supported by the FPU
1551 legacy only accept legacy-NaN binaries, if supported
1553 2008 only accept 2008-NaN binaries, if supported
1555 relaxed accept any binaries regardless of whether
1556 supported by the FPU
1558 The FPU emulator is always able to support both NaN
1559 encodings, so if no FPU hardware is present or it has
1560 been disabled with 'nofpu', then the settings of
1561 'legacy' and '2008' strap the emulator accordingly,
1562 'relaxed' straps the emulator for both legacy-NaN and
1563 2008-NaN, whereas 'strict' enables legacy-NaN only on
1564 legacy processors and both NaN encodings on MIPS32 or
1567 The setting for ABS.fmt/NEG.fmt instruction execution
1568 mode generally follows that for the NaN encoding,
1569 except where unsupported by hardware.
1571 ignore_loglevel [KNL]
1572 Ignore loglevel setting - this will print /all/
1573 kernel messages to the console. Useful for debugging.
1574 We also add it as printk module parameter, so users
1575 could change it dynamically, usually by
1576 /sys/module/printk/parameters/ignore_loglevel.
1579 Ignore RLIMIT_DATA setting for data mappings,
1580 print warning at first misuse. Can be changed via
1581 /sys/module/kernel/parameters/ignore_rlimit_data.
1583 ihash_entries= [KNL]
1584 Set number of hash buckets for inode cache.
1586 ima_appraise= [IMA] appraise integrity measurements
1587 Format: { "off" | "enforce" | "fix" | "log" }
1590 ima_appraise_tcb [IMA] Deprecated. Use ima_policy= instead.
1591 The builtin appraise policy appraises all files
1594 ima_canonical_fmt [IMA]
1595 Use the canonical format for the binary runtime
1596 measurements, instead of host native format.
1599 Format: { md5 | sha1 | rmd160 | sha256 | sha384
1603 The list of supported hash algorithms is defined
1604 in crypto/hash_info.h.
1607 The builtin policies to load during IMA setup.
1608 Format: "tcb | appraise_tcb | secure_boot |
1611 The "tcb" policy measures all programs exec'd, files
1612 mmap'd for exec, and all files opened with the read
1613 mode bit set by either the effective uid (euid=0) or
1616 The "appraise_tcb" policy appraises the integrity of
1617 all files owned by root.
1619 The "secure_boot" policy appraises the integrity
1620 of files (eg. kexec kernel image, kernel modules,
1621 firmware, policy, etc) based on file signatures.
1623 The "fail_securely" policy forces file signature
1624 verification failure also on privileged mounted
1625 filesystems with the SB_I_UNVERIFIABLE_SIGNATURE
1628 ima_tcb [IMA] Deprecated. Use ima_policy= instead.
1629 Load a policy which meets the needs of the Trusted
1630 Computing Base. This means IMA will measure all
1631 programs exec'd, files mmap'd for exec, and all files
1632 opened for read by uid=0.
1635 Select one of defined IMA measurements template formats.
1636 Formats: { "ima" | "ima-ng" | "ima-sig" }
1640 [IMA] Define a custom template format.
1641 Format: { "field1|...|fieldN" }
1643 ima.ahash_minsize= [IMA] Minimum file size for asynchronous hash usage
1644 Format: <min_file_size>
1645 Set the minimal file size for using asynchronous hash.
1646 If left unspecified, ahash usage is disabled.
1648 ahash performance varies for different data sizes on
1649 different crypto accelerators. This option can be used
1650 to achieve the best performance for a particular HW.
1652 ima.ahash_bufsize= [IMA] Asynchronous hash buffer size
1654 Set hashing buffer size. Default: 4k.
1656 ahash performance varies for different chunk sizes on
1657 different crypto accelerators. This option can be used
1658 to achieve best performance for particular HW.
1662 Run specified binary instead of /sbin/init as init
1665 initcall_debug [KNL] Trace initcalls as they are executed. Useful
1666 for working out where the kernel is dying during
1669 initcall_blacklist= [KNL] Do not execute a comma-separated list of
1670 initcall functions. Useful for debugging built-in
1671 modules and initcalls.
1673 initrd= [BOOT] Specify the location of the initial ramdisk
1675 init_on_alloc= [MM] Fill newly allocated pages and heap objects with
1678 Default set by CONFIG_INIT_ON_ALLOC_DEFAULT_ON.
1680 init_on_free= [MM] Fill freed pages and heap objects with zeroes.
1682 Default set by CONFIG_INIT_ON_FREE_DEFAULT_ON.
1684 init_pkru= [x86] Specify the default memory protection keys rights
1685 register contents for all processes. 0x55555554 by
1686 default (disallow access to all but pkey 0). Can
1687 override in debugfs after boot.
1689 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver
1692 int_pln_enable [x86] Enable power limit notification interrupt
1694 integrity_audit=[IMA]
1695 Format: { "0" | "1" }
1696 0 -- basic integrity auditing messages. (Default)
1697 1 -- additional integrity auditing messages.
1699 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option
1701 Enable intel iommu driver.
1703 Disable intel iommu driver.
1704 igfx_off [Default Off]
1705 By default, gfx is mapped as normal device. If a gfx
1706 device has a dedicated DMAR unit, the DMAR unit is
1707 bypassed by not enabling DMAR with this option. In
1708 this case, gfx device will use physical address for
1711 With this option iommu will not optimize to look
1712 for io virtual address below 32-bit forcing dual
1713 address cycle on pci bus for cards supporting greater
1714 than 32-bit addressing. The default is to look
1715 for translation below 32-bit and if not available
1716 then look in the higher range.
1717 strict [Default Off]
1718 With this option on every unmap_single operation will
1719 result in a hardware IOTLB flush operation as opposed
1720 to batching them for performance.
1721 sp_off [Default Off]
1722 By default, super page will be supported if Intel IOMMU
1723 has the capability. With this option, super page will
1726 By default, scalable mode will be disabled even if the
1727 hardware advertises that it has support for the scalable
1728 mode translation. With this option set, scalable mode
1729 will be used on hardware which claims to support it.
1730 tboot_noforce [Default Off]
1731 Do not force the Intel IOMMU enabled under tboot.
1732 By default, tboot will force Intel IOMMU on, which
1733 could harm performance of some high-throughput
1734 devices like 40GBit network cards, even if identity
1736 Note that using this option lowers the security
1737 provided by tboot because it makes the system
1738 vulnerable to DMA attacks.
1739 nobounce [Default off]
1740 Disable bounce buffer for unstrusted devices such as
1741 the Thunderbolt devices. This will treat the untrusted
1742 devices as the trusted ones, hence might expose security
1743 risks of DMA attacks.
1745 intel_idle.max_cstate= [KNL,HW,ACPI,X86]
1746 0 disables intel_idle and fall back on acpi_idle.
1747 1 to 9 specify maximum depth of C-state.
1751 Do not enable intel_pstate as the default
1752 scaling driver for the supported processors
1754 Use intel_pstate as a scaling driver, but configure it
1755 to work with generic cpufreq governors (instead of
1756 enabling its internal governor). This mode cannot be
1757 used along with the hardware-managed P-states (HWP)
1760 Enable intel_pstate on systems that prohibit it by default
1761 in favor of acpi-cpufreq. Forcing the intel_pstate driver
1762 instead of acpi-cpufreq may disable platform features, such
1763 as thermal controls and power capping, that rely on ACPI
1764 P-States information being indicated to OSPM and therefore
1765 should be used with caution. This option does not work with
1766 processors that aren't supported by the intel_pstate driver
1767 or on platforms that use pcc-cpufreq instead of acpi-cpufreq.
1769 Do not enable hardware P state control (HWP)
1772 Only load intel_pstate on systems which support
1773 hardware P state control (HWP) if available.
1775 Enforce ACPI _PPC performance limits. If the Fixed ACPI
1776 Description Table, specifies preferred power management
1777 profile as "Enterprise Server" or "Performance Server",
1778 then this feature is turned on by default.
1780 Allow per-logical-CPU P-State performance control limits using
1781 cpufreq sysfs interface
1783 intremap= [X86-64, Intel-IOMMU]
1784 on enable Interrupt Remapping (default)
1785 off disable Interrupt Remapping
1786 nosid disable Source ID checking
1788 BIOS x2APIC opt-out request will be ignored
1789 nopost disable Interrupt Posting
1791 iomem= Disable strict checking of access to MMIO memory
1792 strict regions from userspace.
1807 nobypass [PPC/POWERNV]
1808 Disable IOMMU bypass, using IOMMU for PCI devices.
1810 iommu.strict= [ARM64] Configure TLB invalidation behaviour
1811 Format: { "0" | "1" }
1813 Request that DMA unmap operations use deferred
1814 invalidation of hardware TLBs, for increased
1815 throughput at the cost of reduced device isolation.
1816 Will fall back to strict mode if not supported by
1817 the relevant IOMMU driver.
1818 1 - Strict mode (default).
1819 DMA unmap operations invalidate IOMMU hardware TLBs
1823 [ARM64, X86] Configure DMA to bypass the IOMMU by default.
1824 Format: { "0" | "1" }
1825 0 - Use IOMMU translation for DMA.
1826 1 - Bypass the IOMMU for DMA.
1827 unset - Use value of CONFIG_IOMMU_DEFAULT_PASSTHROUGH.
1829 io7= [HW] IO7 for Marvel based alpha systems
1830 See comment before marvel_specify_io7 in
1831 arch/alpha/kernel/core_marvel.c.
1833 io_delay= [X86] I/O delay method
1835 Standard port 0x80 based delay
1837 Alternate port 0xed based delay (needed on some systems)
1839 Simple two microseconds delay
1844 See Documentation/filesystems/nfs/nfsroot.txt.
1846 ipcmni_extend [KNL] Extend the maximum number of unique System V
1847 IPC identifiers from 32,768 to 16,777,216.
1849 irqaffinity= [SMP] Set the default irq affinity mask
1850 The argument is a cpu list, as described above.
1852 irqchip.gicv2_force_probe=
1855 Force the kernel to look for the second 4kB page
1856 of a GICv2 controller even if the memory range
1857 exposed by the device tree is too small.
1859 irqchip.gicv3_nolpi=
1861 Force the kernel to ignore the availability of
1862 LPIs (and by consequence ITSs). Intended for system
1863 that use the kernel as a bootloader, and thus want
1864 to let secondary kernels in charge of setting up
1867 irqchip.gicv3_pseudo_nmi= [ARM64]
1868 Enables support for pseudo-NMIs in the kernel. This
1869 requires the kernel to be built with
1870 CONFIG_ARM64_PSEUDO_NMI.
1873 When an interrupt is not handled search all handlers
1874 for it. Intended to get systems with badly broken
1878 When an interrupt is not handled search all handlers
1879 for it. Also check all handlers each timer
1880 interrupt. Intended to get systems with badly broken
1884 Format: <RDP>,<reset>,<pci_scan>,<verbosity>
1886 isolcpus= [KNL,SMP,ISOL] Isolate a given set of CPUs from disturbance.
1887 [Deprecated - use cpusets instead]
1888 Format: [flag-list,]<cpu-list>
1890 Specify one or more CPUs to isolate from disturbances
1891 specified in the flag list (default: domain):
1894 Disable the tick when a single task runs.
1896 A residual 1Hz tick is offloaded to workqueues, which you
1897 need to affine to housekeeping through the global
1898 workqueue's affinity configured via the
1899 /sys/devices/virtual/workqueue/cpumask sysfs file, or
1900 by using the 'domain' flag described below.
1902 NOTE: by default the global workqueue runs on all CPUs,
1903 so to protect individual CPUs the 'cpumask' file has to
1904 be configured manually after bootup.
1907 Isolate from the general SMP balancing and scheduling
1908 algorithms. Note that performing domain isolation this way
1909 is irreversible: it's not possible to bring back a CPU to
1910 the domains once isolated through isolcpus. It's strongly
1911 advised to use cpusets instead to disable scheduler load
1912 balancing through the "cpuset.sched_load_balance" file.
1913 It offers a much more flexible interface where CPUs can
1914 move in and out of an isolated set anytime.
1916 You can move a process onto or off an "isolated" CPU via
1917 the CPU affinity syscalls or cpuset.
1918 <cpu number> begins at 0 and the maximum value is
1919 "number of CPUs in system - 1".
1921 The format of <cpu-list> is described above.
1927 ivrs_ioapic [HW,X86_64]
1928 Provide an override to the IOAPIC-ID<->DEVICE-ID
1929 mapping provided in the IVRS ACPI table. For
1930 example, to map IOAPIC-ID decimal 10 to
1931 PCI device 00:14.0 write the parameter as:
1932 ivrs_ioapic[10]=00:14.0
1934 ivrs_hpet [HW,X86_64]
1935 Provide an override to the HPET-ID<->DEVICE-ID
1936 mapping provided in the IVRS ACPI table. For
1937 example, to map HPET-ID decimal 0 to
1938 PCI device 00:14.0 write the parameter as:
1939 ivrs_hpet[0]=00:14.0
1941 ivrs_acpihid [HW,X86_64]
1942 Provide an override to the ACPI-HID:UID<->DEVICE-ID
1943 mapping provided in the IVRS ACPI table. For
1944 example, to map UART-HID:UID AMD0020:0 to
1945 PCI device 00:14.5 write the parameter as:
1946 ivrs_acpihid[00:14.5]=AMD0020:0
1948 js= [HW,JOY] Analog joystick
1949 See Documentation/input/joydev/joystick.rst.
1952 When CONFIG_RANDOMIZE_BASE is set, this disables
1953 kernel and module base offset ASLR (Address Space
1954 Layout Randomization).
1957 [KNL] Enforce KASAN (Kernel Address Sanitizer) to print
1958 report on every invalid memory access. Without this
1959 parameter KASAN will print report only for the first
1964 kernelcore= [KNL,X86,IA-64,PPC]
1965 Format: nn[KMGTPE] | nn% | "mirror"
1966 This parameter specifies the amount of memory usable by
1967 the kernel for non-movable allocations. The requested
1968 amount is spread evenly throughout all nodes in the
1969 system as ZONE_NORMAL. The remaining memory is used for
1970 movable memory in its own zone, ZONE_MOVABLE. In the
1971 event, a node is too small to have both ZONE_NORMAL and
1972 ZONE_MOVABLE, kernelcore memory will take priority and
1973 other nodes will have a larger ZONE_MOVABLE.
1975 ZONE_MOVABLE is used for the allocation of pages that
1976 may be reclaimed or moved by the page migration
1977 subsystem. Note that allocations like PTEs-from-HighMem
1978 still use the HighMem zone if it exists, and the Normal
1979 zone if it does not.
1981 It is possible to specify the exact amount of memory in
1982 the form of "nn[KMGTPE]", a percentage of total system
1983 memory in the form of "nn%", or "mirror". If "mirror"
1984 option is specified, mirrored (reliable) memory is used
1985 for non-movable allocations and remaining memory is used
1986 for Movable pages. "nn[KMGTPE]", "nn%", and "mirror"
1987 are exclusive, so you cannot specify multiple forms.
1989 kgdbdbgp= [KGDB,HW] kgdb over EHCI usb debug port.
1990 Format: <Controller#>[,poll interval]
1991 The controller # is the number of the ehci usb debug
1992 port as it is probed via PCI. The poll interval is
1993 optional and is the number seconds in between
1994 each poll cycle to the debug port in case you need
1995 the functionality for interrupting the kernel with
1996 gdb or control-c on the dbgp connection. When
1997 not using this parameter you use sysrq-g to break into
1998 the kernel debugger.
2000 kgdboc= [KGDB,HW] kgdb over consoles.
2001 Requires a tty driver that supports console polling,
2002 or a supported polling keyboard driver (non-usb).
2003 Serial only format: <serial_device>[,baud]
2004 keyboard only format: kbd
2005 keyboard and serial format: kbd,<serial_device>[,baud]
2006 Optional Kernel mode setting:
2007 kms, kbd format: kms,kbd
2008 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud]
2010 kgdbwait [KGDB] Stop kernel execution and enter the
2011 kernel debugger at the earliest opportunity.
2013 kmac= [MIPS] korina ethernet MAC address.
2014 Configure the RouterBoard 532 series on-chip
2015 Ethernet adapter MAC address.
2017 kmemleak= [KNL] Boot-time kmemleak enable/disable
2018 Valid arguments: on, off
2020 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y,
2023 kprobe_event=[probe-list]
2024 [FTRACE] Add kprobe events and enable at boot time.
2025 The probe-list is a semicolon delimited list of probe
2026 definitions. Each definition is same as kprobe_events
2027 interface, but the parameters are comma delimited.
2028 For example, to add a kprobe event on vfs_read with
2029 arg1 and arg2, add to the command line;
2031 kprobe_event=p,vfs_read,$arg1,$arg2
2033 See also Documentation/trace/kprobetrace.rst "Kernel
2034 Boot Parameter" section.
2036 kpti= [ARM64] Control page table isolation of user
2037 and kernel address spaces.
2038 Default: enabled on cores which need mitigation.
2042 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs.
2043 Default is 0 (don't ignore, but inject #GP)
2045 kvm.enable_vmware_backdoor=[KVM] Support VMware backdoor PV interface.
2046 Default is false (don't support).
2048 kvm.mmu_audit= [KVM] This is a R/W parameter which allows audit
2052 kvm-amd.nested= [KVM,AMD] Allow nested virtualization in KVM/SVM.
2053 Default is 1 (enabled)
2055 kvm-amd.npt= [KVM,AMD] Disable nested paging (virtualized MMU)
2057 Default is 1 (enabled) if in 64-bit or 32-bit PAE mode.
2059 kvm-arm.vgic_v3_group0_trap=
2060 [KVM,ARM] Trap guest accesses to GICv3 group-0
2063 kvm-arm.vgic_v3_group1_trap=
2064 [KVM,ARM] Trap guest accesses to GICv3 group-1
2067 kvm-arm.vgic_v3_common_trap=
2068 [KVM,ARM] Trap guest accesses to GICv3 common
2071 kvm-arm.vgic_v4_enable=
2072 [KVM,ARM] Allow use of GICv4 for direct injection of
2075 kvm-intel.ept= [KVM,Intel] Disable extended page tables
2076 (virtualized MMU) support on capable Intel chips.
2077 Default is 1 (enabled)
2079 kvm-intel.emulate_invalid_guest_state=
2080 [KVM,Intel] Enable emulation of invalid guest states
2081 Default is 0 (disabled)
2083 kvm-intel.flexpriority=
2084 [KVM,Intel] Disable FlexPriority feature (TPR shadow).
2085 Default is 1 (enabled)
2088 [KVM,Intel] Enable VMX nesting (nVMX).
2089 Default is 0 (disabled)
2091 kvm-intel.unrestricted_guest=
2092 [KVM,Intel] Disable unrestricted guest feature
2093 (virtualized real and unpaged mode) on capable
2094 Intel chips. Default is 1 (enabled)
2096 kvm-intel.vmentry_l1d_flush=[KVM,Intel] Mitigation for L1 Terminal Fault
2099 Valid arguments: never, cond, always
2101 always: L1D cache flush on every VMENTER.
2102 cond: Flush L1D on VMENTER only when the code between
2103 VMEXIT and VMENTER can leak host memory.
2104 never: Disables the mitigation
2106 Default is cond (do L1 cache flush in specific instances)
2108 kvm-intel.vpid= [KVM,Intel] Disable Virtual Processor Identification
2109 feature (tagged TLBs) on capable Intel chips.
2110 Default is 1 (enabled)
2112 l1tf= [X86] Control mitigation of the L1TF vulnerability on
2115 The kernel PTE inversion protection is unconditionally
2116 enabled and cannot be disabled.
2119 Provides all available mitigations for the
2120 L1TF vulnerability. Disables SMT and
2121 enables all mitigations in the
2122 hypervisors, i.e. unconditional L1D flush.
2124 SMT control and L1D flush control via the
2125 sysfs interface is still possible after
2126 boot. Hypervisors will issue a warning
2127 when the first VM is started in a
2128 potentially insecure configuration,
2129 i.e. SMT enabled or L1D flush disabled.
2132 Same as 'full', but disables SMT and L1D
2133 flush runtime control. Implies the
2134 'nosmt=force' command line option.
2135 (i.e. sysfs control of SMT is disabled.)
2138 Leaves SMT enabled and enables the default
2139 hypervisor mitigation, i.e. conditional
2142 SMT control and L1D flush control via the
2143 sysfs interface is still possible after
2144 boot. Hypervisors will issue a warning
2145 when the first VM is started in a
2146 potentially insecure configuration,
2147 i.e. SMT enabled or L1D flush disabled.
2151 Disables SMT and enables the default
2152 hypervisor mitigation.
2154 SMT control and L1D flush control via the
2155 sysfs interface is still possible after
2156 boot. Hypervisors will issue a warning
2157 when the first VM is started in a
2158 potentially insecure configuration,
2159 i.e. SMT enabled or L1D flush disabled.
2162 Same as 'flush', but hypervisors will not
2163 warn when a VM is started in a potentially
2164 insecure configuration.
2167 Disables hypervisor mitigations and doesn't
2169 It also drops the swap size and available
2170 RAM limit restriction on both hypervisor and
2175 For details see: Documentation/admin-guide/hw-vuln/l1tf.rst
2181 lapic [X86-32,APIC] Enable the local APIC even if BIOS
2184 lapic= [x86,APIC] "notscdeadline" Do not use TSC deadline
2185 value for LAPIC timer one-shot implementation. Default
2186 back to the programmable timer unit in the LAPIC.
2188 lapic_timer_c2_ok [X86,APIC] trust the local apic timer
2191 libata.dma= [LIBATA] DMA control
2192 libata.dma=0 Disable all PATA and SATA DMA
2193 libata.dma=1 PATA and SATA Disk DMA only
2194 libata.dma=2 ATAPI (CDROM) DMA only
2195 libata.dma=4 Compact Flash DMA only
2196 Combinations also work, so libata.dma=3 enables DMA
2197 for disks and CDROMs, but not CFs.
2199 libata.ignore_hpa= [LIBATA] Ignore HPA limit
2200 libata.ignore_hpa=0 keep BIOS limits (default)
2201 libata.ignore_hpa=1 ignore limits, using full disk
2203 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume
2207 libata.force= [LIBATA] Force configurations. The format is comma
2208 separated list of "[ID:]VAL" where ID is
2209 PORT[.DEVICE]. PORT and DEVICE are decimal numbers
2210 matching port, link or device. Basically, it matches
2211 the ATA ID string printed on console by libata. If
2212 the whole ID part is omitted, the last PORT and DEVICE
2213 values are used. If ID hasn't been specified yet, the
2214 configuration applies to all ports, links and devices.
2216 If only DEVICE is omitted, the parameter applies to
2217 the port and all links and devices behind it. DEVICE
2218 number of 0 either selects the first device or the
2219 first fan-out link behind PMP device. It does not
2220 select the host link. DEVICE number of 15 selects the
2221 host link and device attached to it.
2223 The VAL specifies the configuration to force. As long
2224 as there's no ambiguity shortcut notation is allowed.
2225 For example, both 1.5 and 1.5G would work for 1.5Gbps.
2226 The following configurations can be forced.
2228 * Cable type: 40c, 80c, short40c, unk, ign or sata.
2229 Any ID with matching PORT is used.
2231 * SATA link speed limit: 1.5Gbps or 3.0Gbps.
2233 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7].
2234 udma[/][16,25,33,44,66,100,133] notation is also
2237 * [no]ncq: Turn on or off NCQ.
2239 * [no]ncqtrim: Turn off queued DSM TRIM.
2241 * nohrst, nosrst, norst: suppress hard, soft
2244 * rstonce: only attempt one reset during
2245 hot-unplug link recovery
2247 * dump_id: dump IDENTIFY data.
2249 * atapi_dmadir: Enable ATAPI DMADIR bridge support
2251 * disable: Disable this device.
2253 If there are multiple matching configurations changing
2254 the same attribute, the last one is used.
2256 memblock=debug [KNL] Enable memblock debug messages.
2258 load_ramdisk= [RAM] List of ramdisks to load from floppy
2259 See Documentation/admin-guide/blockdev/ramdisk.rst.
2261 lockd.nlm_grace_period=P [NFS] Assign grace period.
2264 lockd.nlm_tcpport=N [NFS] Assign TCP port.
2267 lockd.nlm_timeout=T [NFS] Assign timeout value.
2270 lockd.nlm_udpport=M [NFS] Assign UDP port.
2273 locktorture.nreaders_stress= [KNL]
2274 Set the number of locking read-acquisition kthreads.
2275 Defaults to being automatically set based on the
2276 number of online CPUs.
2278 locktorture.nwriters_stress= [KNL]
2279 Set the number of locking write-acquisition kthreads.
2281 locktorture.onoff_holdoff= [KNL]
2282 Set time (s) after boot for CPU-hotplug testing.
2284 locktorture.onoff_interval= [KNL]
2285 Set time (s) between CPU-hotplug operations, or
2286 zero to disable CPU-hotplug testing.
2288 locktorture.shuffle_interval= [KNL]
2289 Set task-shuffle interval (jiffies). Shuffling
2290 tasks allows some CPUs to go into dyntick-idle
2291 mode during the locktorture test.
2293 locktorture.shutdown_secs= [KNL]
2294 Set time (s) after boot system shutdown. This
2295 is useful for hands-off automated testing.
2297 locktorture.stat_interval= [KNL]
2298 Time (s) between statistics printk()s.
2300 locktorture.stutter= [KNL]
2301 Time (s) to stutter testing, for example,
2302 specifying five seconds causes the test to run for
2303 five seconds, wait for five seconds, and so on.
2304 This tests the locking primitive's ability to
2305 transition abruptly to and from idle.
2307 locktorture.torture_type= [KNL]
2308 Specify the locking implementation to test.
2310 locktorture.verbose= [KNL]
2311 Enable additional printk() statements.
2313 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver
2316 loglevel= All Kernel Messages with a loglevel smaller than the
2317 console loglevel will be printed to the console. It can
2318 also be changed with klogd or other programs. The
2319 loglevels are defined as follows:
2321 0 (KERN_EMERG) system is unusable
2322 1 (KERN_ALERT) action must be taken immediately
2323 2 (KERN_CRIT) critical conditions
2324 3 (KERN_ERR) error conditions
2325 4 (KERN_WARNING) warning conditions
2326 5 (KERN_NOTICE) normal but significant condition
2327 6 (KERN_INFO) informational
2328 7 (KERN_DEBUG) debug-level messages
2330 log_buf_len=n[KMG] Sets the size of the printk ring buffer,
2331 in bytes. n must be a power of two and greater
2332 than the minimal size. The minimal size is defined
2333 by LOG_BUF_SHIFT kernel config parameter. There is
2334 also CONFIG_LOG_CPU_MAX_BUF_SHIFT config parameter
2335 that allows to increase the default size depending on
2336 the number of CPUs. See init/Kconfig for more details.
2338 logo.nologo [FB] Disables display of the built-in Linux logo.
2339 This may be used to provide more screen space for
2340 kernel log messages and is useful when debugging
2341 kernel boot problems.
2343 lp=0 [LP] Specify parallel ports to use, e.g,
2344 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses
2345 lp=reset first parallel port). 'lp=0' disables the
2346 lp=auto printer driver. 'lp=reset' (which can be
2347 specified in addition to the ports) causes
2348 attached printers to be reset. Using
2349 lp=port1,port2,... specifies the parallel ports
2350 to associate lp devices with, starting with
2351 lp0. A port specification may be 'none' to skip
2352 that lp device, or a parport name such as
2353 'parport0'. Specifying 'lp=auto' instead of a
2354 port specification list means that device IDs
2355 from each port should be examined, to see if
2356 an IEEE 1284-compliant printer is attached; if
2357 so, the driver will manage that printer.
2358 See also header of drivers/char/lp.c.
2361 Sets loops_per_jiffy to given constant, thus avoiding
2362 time-consuming boot-time autodetection (up to 250 ms per
2363 CPU). 0 enables autodetection (default). To determine
2364 the correct value for your kernel, boot with normal
2365 autodetection and see what value is printed. Note that
2366 on SMP systems the preset will be applied to all CPUs,
2367 which is likely to cause problems if your CPUs need
2368 significantly divergent settings. An incorrect value
2369 will cause delays in the kernel to be wrong, leading to
2370 unpredictable I/O errors and other breakage. Although
2371 unlikely, in the extreme case this might damage your
2375 Format: <io>,<irq>,<dma>
2377 lsm.debug [SECURITY] Enable LSM initialization debugging output.
2380 [SECURITY] Choose order of LSM initialization. This
2381 overrides CONFIG_LSM, and the "security=" parameter.
2383 machvec= [IA-64] Force the use of a particular machine-vector
2384 (machvec) in a generic kernel.
2385 Example: machvec=hpzx1
2387 machtype= [Loongson] Share the same kernel image file between different
2389 Example: machtype=lemote-yeeloong-2f-7inch
2391 max_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory greater
2392 than or equal to this physical address is ignored.
2394 maxcpus= [SMP] Maximum number of processors that an SMP kernel
2395 will bring up during bootup. maxcpus=n : n >= 0 limits
2396 the kernel to bring up 'n' processors. Surely after
2397 bootup you can bring up the other plugged cpu by executing
2398 "echo 1 > /sys/devices/system/cpu/cpuX/online". So maxcpus
2399 only takes effect during system bootup.
2400 While n=0 is a special case, it is equivalent to "nosmp",
2401 which also disables the IO APIC.
2403 max_loop= [LOOP] The number of loop block devices that get
2404 (loop.max_loop) unconditionally pre-created at init time. The default
2405 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead
2406 of statically allocating a predefined number, loop
2407 devices can be requested on-demand with the
2408 /dev/loop-control interface.
2410 mce [X86-32] Machine Check Exception
2412 mce=option [X86-64] See Documentation/x86/x86_64/boot-options.rst
2414 md= [HW] RAID subsystems devices and level
2415 See Documentation/admin-guide/md.rst.
2418 Format: <first>,<last>
2419 Specifies range of consoles to be captured by the MDA.
2422 Control mitigation for the Micro-architectural Data
2423 Sampling (MDS) vulnerability.
2425 Certain CPUs are vulnerable to an exploit against CPU
2426 internal buffers which can forward information to a
2427 disclosure gadget under certain conditions.
2429 In vulnerable processors, the speculatively
2430 forwarded data can be used in a cache side channel
2431 attack, to access data to which the attacker does
2432 not have direct access.
2434 This parameter controls the MDS mitigation. The
2437 full - Enable MDS mitigation on vulnerable CPUs
2438 full,nosmt - Enable MDS mitigation and disable
2439 SMT on vulnerable CPUs
2440 off - Unconditionally disable MDS mitigation
2442 Not specifying this option is equivalent to
2445 For details see: Documentation/admin-guide/hw-vuln/mds.rst
2447 mem=nn[KMG] [KNL,BOOT] Force usage of a specific amount of memory
2448 Amount of memory to be used when the kernel is not able
2449 to see the whole system memory or for test.
2450 [X86] Work as limiting max address. Use together
2451 with memmap= to avoid physical address space collisions.
2452 Without memmap= PCI devices could be placed at addresses
2453 belonging to unused RAM.
2455 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel
2459 [KNL,SH] Allow user to override the default size for
2460 per-device physically contiguous DMA buffers.
2462 memhp_default_state=online/offline
2463 [KNL] Set the initial state for the memory hotplug
2464 onlining policy. If not specified, the default value is
2465 set according to the
2466 CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE kernel config
2468 See Documentation/admin-guide/mm/memory-hotplug.rst.
2470 memmap=exactmap [KNL,X86] Enable setting of an exact
2471 E820 memory map, as specified by the user.
2472 Such memmap=exactmap lines can be constructed based on
2473 BIOS output or other requirements. See the memmap=nn@ss
2476 memmap=nn[KMG]@ss[KMG]
2477 [KNL] Force usage of a specific region of memory.
2478 Region of memory to be used is from ss to ss+nn.
2479 If @ss[KMG] is omitted, it is equivalent to mem=nn[KMG],
2480 which limits max address to nn[KMG].
2481 Multiple different regions can be specified,
2484 memmap=100M@2G,100M#3G,1G!1024G
2486 memmap=nn[KMG]#ss[KMG]
2487 [KNL,ACPI] Mark specific memory as ACPI data.
2488 Region of memory to be marked is from ss to ss+nn.
2490 memmap=nn[KMG]$ss[KMG]
2491 [KNL,ACPI] Mark specific memory as reserved.
2492 Region of memory to be reserved is from ss to ss+nn.
2493 Example: Exclude memory from 0x18690000-0x1869ffff
2494 memmap=64K$0x18690000
2496 memmap=0x10000$0x18690000
2497 Some bootloaders may need an escape character before '$',
2498 like Grub2, otherwise '$' and the following number
2501 memmap=nn[KMG]!ss[KMG]
2502 [KNL,X86] Mark specific memory as protected.
2503 Region of memory to be used, from ss to ss+nn.
2504 The memory region may be marked as e820 type 12 (0xc)
2505 and is NVDIMM or ADR memory.
2507 memmap=<size>%<offset>-<oldtype>+<newtype>
2508 [KNL,ACPI] Convert memory within the specified region
2509 from <oldtype> to <newtype>. If "-<oldtype>" is left
2510 out, the whole region will be marked as <newtype>,
2511 even if previously unavailable. If "+<newtype>" is left
2512 out, matching memory will be removed. Types are
2513 specified as e820 types, e.g., 1 = RAM, 2 = reserved,
2514 3 = ACPI, 12 = PRAM.
2516 memory_corruption_check=0/1 [X86]
2517 Some BIOSes seem to corrupt the first 64k of
2518 memory when doing things like suspend/resume.
2519 Setting this option will scan the memory
2520 looking for corruption. Enabling this will
2521 both detect corruption and prevent the kernel
2522 from using the memory being corrupted.
2523 However, its intended as a diagnostic tool; if
2524 repeatable BIOS-originated corruption always
2525 affects the same memory, you can use memmap=
2526 to prevent the kernel from using that memory.
2528 memory_corruption_check_size=size [X86]
2529 By default it checks for corruption in the low
2530 64k, making this memory unavailable for normal
2531 use. Use this parameter to scan for
2532 corruption in more or less memory.
2534 memory_corruption_check_period=seconds [X86]
2535 By default it checks for corruption every 60
2536 seconds. Use this parameter to check at some
2537 other rate. 0 disables periodic checking.
2539 memtest= [KNL,X86,ARM,PPC] Enable memtest
2541 default : 0 <disable>
2542 Specifies the number of memtest passes to be
2543 performed. Each pass selects another test
2544 pattern from a given set of patterns. Memtest
2545 fills the memory with this pattern, validates
2546 memory contents and reserves bad memory
2547 regions that are detected.
2549 mem_encrypt= [X86-64] AMD Secure Memory Encryption (SME) control
2550 Valid arguments: on, off
2551 Default (depends on kernel configuration option):
2552 on (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=y)
2553 off (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=n)
2554 mem_encrypt=on: Activate SME
2555 mem_encrypt=off: Do not activate SME
2557 Refer to Documentation/virt/kvm/amd-memory-encryption.rst
2558 for details on when memory encryption can be activated.
2560 mem_sleep_default= [SUSPEND] Default system suspend mode:
2561 s2idle - Suspend-To-Idle
2562 shallow - Power-On Suspend or equivalent (if supported)
2563 deep - Suspend-To-RAM or equivalent (if supported)
2564 See Documentation/admin-guide/pm/sleep-states.rst.
2566 meye.*= [HW] Set MotionEye Camera parameters
2567 See Documentation/media/v4l-drivers/meye.rst.
2569 mfgpt_irq= [IA-32] Specify the IRQ to use for the
2570 Multi-Function General Purpose Timers on AMD Geode
2573 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when
2574 the BIOS has incorrectly applied a workaround. TinyBIOS
2575 version 0.98 is known to be affected, 0.99 fixes the
2576 problem by letting the user disable the workaround.
2580 min_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory below this
2581 physical address is ignored.
2583 mini2440= [ARM,HW,KNL]
2584 Format:[0..2][b][c][t]
2586 MINI2440 configuration specification:
2587 0 - The attached screen is the 3.5" TFT
2588 1 - The attached screen is the 7" TFT
2589 2 - The VGA Shield is attached (1024x768)
2590 Leaving out the screen size parameter will not load
2591 the TFT driver, and the framebuffer will be left
2593 b - Enable backlight. The TFT backlight pin will be
2594 linked to the kernel VESA blanking code and a GPIO
2595 LED. This parameter is not necessary when using the
2597 c - Enable the s3c camera interface.
2598 t - Reserved for enabling touchscreen support. The
2599 touchscreen support is not enabled in the mainstream
2600 kernel as of 2.6.30, a preliminary port can be found
2601 in the "bleeding edge" mini2440 support kernel at
2602 http://repo.or.cz/w/linux-2.6/mini2440.git
2605 [X86,PPC,S390,ARM64] Control optional mitigations for
2606 CPU vulnerabilities. This is a set of curated,
2607 arch-independent options, each of which is an
2608 aggregation of existing arch-specific options.
2611 Disable all optional CPU mitigations. This
2612 improves system performance, but it may also
2613 expose users to several CPU vulnerabilities.
2614 Equivalent to: nopti [X86,PPC]
2616 nospectre_v1 [X86,PPC]
2618 nospectre_v2 [X86,PPC,S390,ARM64]
2619 spectre_v2_user=off [X86]
2620 spec_store_bypass_disable=off [X86,PPC]
2621 ssbd=force-off [ARM64]
2626 Mitigate all CPU vulnerabilities, but leave SMT
2627 enabled, even if it's vulnerable. This is for
2628 users who don't want to be surprised by SMT
2629 getting disabled across kernel upgrades, or who
2630 have other ways of avoiding SMT-based attacks.
2631 Equivalent to: (default behavior)
2634 Mitigate all CPU vulnerabilities, disabling SMT
2635 if needed. This is for users who always want to
2636 be fully mitigated, even if it means losing SMT.
2637 Equivalent to: l1tf=flush,nosmt [X86]
2638 mds=full,nosmt [X86]
2641 [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this
2642 parameter allows control of the logging verbosity for
2643 the additional memory initialisation checks. A value
2644 of 0 disables mminit logging and a level of 4 will
2645 log everything. Information is printed at KERN_DEBUG
2646 so loglevel=8 may also need to be specified.
2649 [KNL] When CONFIG_MODULE_SIG is set, this means that
2650 modules without (valid) signatures will fail to load.
2651 Note that if CONFIG_MODULE_SIG_FORCE is set, that
2652 is always true, so this option does nothing.
2654 module_blacklist= [KNL] Do not load a comma-separated list of
2655 modules. Useful for debugging problem modules.
2658 [MOUSE] Maximum time between finger touching and
2659 leaving touchpad surface for touch to be considered
2660 a tap and be reported as a left button click (for
2661 touchpads working in absolute mode only).
2663 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices
2664 reporting absolute coordinates, such as tablets
2665 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices
2666 reporting absolute coordinates, such as tablets
2668 movablecore= [KNL,X86,IA-64,PPC]
2669 Format: nn[KMGTPE] | nn%
2670 This parameter is the complement to kernelcore=, it
2671 specifies the amount of memory used for migratable
2672 allocations. If both kernelcore and movablecore is
2673 specified, then kernelcore will be at *least* the
2674 specified value but may be more. If movablecore on its
2675 own is specified, the administrator must be careful
2676 that the amount of memory usable for all allocations
2679 movable_node [KNL] Boot-time switch to make hotplugable memory
2680 NUMA nodes to be movable. This means that the memory
2681 of such nodes will be usable only for movable
2682 allocations which rules out almost all kernel
2683 allocations. Use with caution!
2685 MTD_Partition= [MTD]
2686 Format: <name>,<region-number>,<size>,<offset>
2688 MTD_Region= [MTD] Format:
2689 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>]
2692 See drivers/mtd/cmdlinepart.c.
2694 multitce=off [PPC] This parameter disables the use of the pSeries
2695 firmware feature for updating multiple TCE entries
2698 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration
2700 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock]
2702 boundary - index of last SLC block on Flex-OneNAND.
2703 The remaining blocks are configured as MLC blocks.
2704 lock - Configure if Flex-OneNAND boundary should be locked.
2705 Once locked, the boundary cannot be changed.
2706 1 indicates lock status, 0 indicates unlock status.
2709 ARM/S3C2412 JIVE boot control
2711 See arch/arm/mach-s3c2412/mach-jive.c
2713 mtouchusb.raw_coordinates=
2714 [HW] Make the MicroTouch USB driver use raw coordinates
2715 ('y', default) or cooked coordinates ('n')
2717 mtrr_chunk_size=nn[KMG] [X86]
2718 used for mtrr cleanup. It is largest continuous chunk
2719 that could hold holes aka. UC entries.
2721 mtrr_gran_size=nn[KMG] [X86]
2722 Used for mtrr cleanup. It is granularity of mtrr block.
2724 Large value could prevent small alignment from
2727 mtrr_spare_reg_nr=n [X86]
2729 Range: 0,7 : spare reg number
2731 Used for mtrr cleanup. It is spare mtrr entries number.
2732 Set to 2 or more if your graphical card needs more.
2734 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card
2736 netdev= [NET] Network devices parameters
2737 Format: <irq>,<io>,<mem_start>,<mem_end>,<name>
2738 Note that mem_start is often overloaded to mean
2739 something different and driver-specific.
2740 This usage is only documented in each driver source
2744 [NETFILTER] Enable connection tracking flow accounting
2745 0 to disable accounting
2746 1 to enable accounting
2749 nfsaddrs= [NFS] Deprecated. Use ip= instead.
2750 See Documentation/filesystems/nfs/nfsroot.txt.
2752 nfsroot= [NFS] nfs root filesystem for disk-less boxes.
2753 See Documentation/filesystems/nfs/nfsroot.txt.
2755 nfsrootdebug [NFS] enable nfsroot debugging messages.
2756 See Documentation/filesystems/nfs/nfsroot.txt.
2758 nfs.callback_nr_threads=
2759 [NFSv4] set the total number of threads that the
2760 NFS client will assign to service NFSv4 callback
2763 nfs.callback_tcpport=
2764 [NFS] set the TCP port on which the NFSv4 callback
2765 channel should listen.
2768 [NFS] sets the pathname to the program which is used
2769 to update the NFS client cache entries.
2771 nfs.cache_getent_timeout=
2772 [NFS] sets the timeout after which an attempt to
2773 update a cache entry is deemed to have failed.
2775 nfs.idmap_cache_timeout=
2776 [NFS] set the maximum lifetime for idmapper cache
2780 [NFS] enable 64-bit inode numbers.
2781 If zero, the NFS client will fake up a 32-bit inode
2782 number for the readdir() and stat() syscalls instead
2783 of returning the full 64-bit number.
2784 The default is to return 64-bit inode numbers.
2786 nfs.max_session_cb_slots=
2787 [NFSv4.1] Sets the maximum number of session
2788 slots the client will assign to the callback
2789 channel. This determines the maximum number of
2790 callbacks the client will process in parallel for
2791 a particular server.
2793 nfs.max_session_slots=
2794 [NFSv4.1] Sets the maximum number of session slots
2795 the client will attempt to negotiate with the server.
2796 This limits the number of simultaneous RPC requests
2797 that the client can send to the NFSv4.1 server.
2798 Note that there is little point in setting this
2799 value higher than the max_tcp_slot_table_limit.
2801 nfs.nfs4_disable_idmapping=
2802 [NFSv4] When set to the default of '1', this option
2803 ensures that both the RPC level authentication
2804 scheme and the NFS level operations agree to use
2805 numeric uids/gids if the mount is using the
2806 'sec=sys' security flavour. In effect it is
2807 disabling idmapping, which can make migration from
2808 legacy NFSv2/v3 systems to NFSv4 easier.
2809 Servers that do not support this mode of operation
2810 will be autodetected by the client, and it will fall
2811 back to using the idmapper.
2812 To turn off this behaviour, set the value to '0'.
2814 [NFS4] Specify an additional fixed unique ident-
2815 ification string that NFSv4 clients can insert into
2816 their nfs_client_id4 string. This is typically a
2817 UUID that is generated at system install time.
2819 nfs.send_implementation_id =
2820 [NFSv4.1] Send client implementation identification
2821 information in exchange_id requests.
2822 If zero, no implementation identification information
2824 The default is to send the implementation identification
2827 nfs.recover_lost_locks =
2828 [NFSv4] Attempt to recover locks that were lost due
2829 to a lease timeout on the server. Please note that
2830 doing this risks data corruption, since there are
2831 no guarantees that the file will remain unchanged
2832 after the locks are lost.
2833 If you want to enable the kernel legacy behaviour of
2834 attempting to recover these locks, then set this
2836 The default parameter value of '0' causes the kernel
2837 not to attempt recovery of lost locks.
2839 nfs4.layoutstats_timer =
2840 [NFSv4.2] Change the rate at which the kernel sends
2841 layoutstats to the pNFS metadata server.
2843 Setting this to value to 0 causes the kernel to use
2844 whatever value is the default set by the layout
2845 driver. A non-zero value sets the minimum interval
2846 in seconds between layoutstats transmissions.
2848 nfsd.nfs4_disable_idmapping=
2849 [NFSv4] When set to the default of '1', the NFSv4
2850 server will return only numeric uids and gids to
2851 clients using auth_sys, and will accept numeric uids
2852 and gids from such clients. This is intended to ease
2853 migration from NFSv2/v3.
2855 nmi_debug= [KNL,SH] Specify one or more actions to take
2856 when a NMI is triggered.
2857 Format: [state][,regs][,debounce][,die]
2859 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels
2860 Format: [panic,][nopanic,][num]
2862 0 - turn hardlockup detector in nmi_watchdog off
2863 1 - turn hardlockup detector in nmi_watchdog on
2864 When panic is specified, panic when an NMI watchdog
2865 timeout occurs (or 'nopanic' to not panic on an NMI
2866 watchdog, if CONFIG_BOOTPARAM_HARDLOCKUP_PANIC is set)
2867 To disable both hard and soft lockup detectors,
2868 please see 'nowatchdog'.
2869 This is useful when you use a panic=... timeout and
2870 need the box quickly up again.
2872 These settings can be accessed at runtime via
2873 the nmi_watchdog and hardlockup_panic sysctls.
2875 netpoll.carrier_timeout=
2876 [NET] Specifies amount of time (in seconds) that
2877 netpoll should wait for a carrier. By default netpoll
2880 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths
2881 emulation library even if a 387 maths coprocessor
2884 no5lvl [X86-64] Disable 5-level paging mode. Forces
2885 kernel to use 4-level paging instead.
2888 [HW] Never suspend the console
2889 Disable suspending of consoles during suspend and
2890 hibernate operations. Once disabled, debugging
2891 messages can reach various consoles while the rest
2892 of the system is being put to sleep (ie, while
2893 debugging driver suspend/resume hooks). This may
2894 not work reliably with all consoles, but is known
2895 to work with serial and VGA consoles.
2896 To facilitate more flexible debugging, we also add
2897 console_suspend, a printk module parameter to control
2898 it. Users could use console_suspend (usually
2899 /sys/module/printk/parameters/console_suspend) to
2900 turn on/off it dynamically.
2902 novmcoredd [KNL,KDUMP]
2903 Disable device dump. Device dump allows drivers to
2904 append dump data to vmcore so you can collect driver
2905 specified debug info. Drivers can append the data
2906 without any limit and this data is stored in memory,
2907 so this may cause significant memory stress. Disabling
2908 device dump can help save memory but the driver debug
2909 data will be no longer available. This parameter
2910 is only available when CONFIG_PROC_VMCORE_DEVICE_DUMP
2913 noaliencache [MM, NUMA, SLAB] Disables the allocation of alien
2914 caches in the slab allocator. Saves per-node memory,
2915 but will impact performance.
2919 noaltinstr [S390] Disables alternative instructions patching
2920 (CPU alternatives feature).
2922 noapic [SMP,APIC] Tells the kernel to not make use of any
2923 IOAPICs that may be present in the system.
2925 noautogroup Disable scheduler automatic task group creation.
2927 nobats [PPC] Do not use BATs for mapping kernel lowmem
2928 on "Classic" PPC cores.
2932 noclflush [BUGS=X86] Don't use the CLFLUSH instruction
2934 nodelayacct [KNL] Disable per-task delay accounting
2936 nodsp [SH] Disable hardware DSP at boot time.
2938 noefi Disable EFI runtime services support.
2943 On X86-32 available only on PAE configured kernels.
2944 noexec=on: enable non-executable mappings (default)
2945 noexec=off: disable non-executable mappings
2948 Disable SMAP (Supervisor Mode Access Prevention)
2949 even if it is supported by processor.
2952 Disable SMEP (Supervisor Mode Execution Prevention)
2953 even if it is supported by processor.
2956 This affects only 32-bit executables.
2957 noexec32=on: enable non-executable mappings (default)
2958 read doesn't imply executable mappings
2959 noexec32=off: disable non-executable mappings
2960 read implies executable mappings
2962 nofpu [MIPS,SH] Disable hardware FPU at boot time.
2964 nofxsr [BUGS=X86-32] Disables x86 floating point extended
2965 register save and restore. The kernel will only save
2966 legacy floating-point registers on task switch.
2968 nohugeiomap [KNL,x86,PPC] Disable kernel huge I/O mappings.
2970 nosmt [KNL,S390] Disable symmetric multithreading (SMT).
2971 Equivalent to smt=1.
2973 [KNL,x86] Disable symmetric multithreading (SMT).
2974 nosmt=force: Force disable SMT, cannot be undone
2975 via the sysfs control file.
2977 nospectre_v1 [X86,PPC] Disable mitigations for Spectre Variant 1
2978 (bounds check bypass). With this option data leaks are
2979 possible in the system.
2981 nospectre_v2 [X86,PPC_FSL_BOOK3E,ARM64] Disable all mitigations for
2982 the Spectre variant 2 (indirect branch prediction)
2983 vulnerability. System may allow data leaks with this
2986 nospec_store_bypass_disable
2987 [HW] Disable all mitigations for the Speculative Store Bypass vulnerability
2989 noxsave [BUGS=X86] Disables x86 extended register state save
2990 and restore using xsave. The kernel will fallback to
2991 enabling legacy floating-point and sse state.
2993 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended
2994 register states. The kernel will fall back to use
2995 xsave to save the states. By using this parameter,
2996 performance of saving the states is degraded because
2997 xsave doesn't support modified optimization while
2998 xsaveopt supports it on xsaveopt enabled systems.
3000 noxsaves [X86] Disables xsaves and xrstors used in saving and
3001 restoring x86 extended register state in compacted
3002 form of xsave area. The kernel will fall back to use
3003 xsaveopt and xrstor to save and restore the states
3004 in standard form of xsave area. By using this
3005 parameter, xsave area per process might occupy more
3006 memory on xsaves enabled systems.
3008 nohlt [BUGS=ARM,SH] Tells the kernel that the sleep(SH) or
3009 wfi(ARM) instruction doesn't work correctly and not to
3010 use it. This is also useful when using JTAG debugger.
3012 no_file_caps Tells the kernel not to honor file capabilities. The
3013 only way then for a file to be executed with privilege
3014 is to be setuid root or executed by root.
3016 nohalt [IA-64] Tells the kernel not to use the power saving
3017 function PAL_HALT_LIGHT when idle. This increases
3018 power-consumption. On the positive side, it reduces
3019 interrupt wake-up latency, which may improve performance
3020 in certain environments such as networked servers or
3023 nohibernate [HIBERNATION] Disable hibernation and resume.
3025 nohz= [KNL] Boottime enable/disable dynamic ticks
3026 Valid arguments: on, off
3029 nohz_full= [KNL,BOOT,SMP,ISOL]
3030 The argument is a cpu list, as described above.
3031 In kernels built with CONFIG_NO_HZ_FULL=y, set
3032 the specified list of CPUs whose tick will be stopped
3033 whenever possible. The boot CPU will be forced outside
3034 the range to maintain the timekeeping. Any CPUs
3035 in this list will have their RCU callbacks offloaded,
3036 just as if they had also been called out in the
3037 rcu_nocbs= boot parameter.
3039 noiotrap [SH] Disables trapped I/O port accesses.
3041 noirqdebug [X86-32] Disables the code which attempts to detect and
3042 disable unhandled interrupt sources.
3044 no_timer_check [X86,APIC] Disables the code which tests for
3045 broken timer IRQ sources.
3047 noisapnp [ISAPNP] Disables ISA PnP code.
3049 noinitrd [RAM] Tells the kernel not to load any configured
3052 nointremap [X86-64, Intel-IOMMU] Do not enable interrupt
3054 [Deprecated - use intremap=off]
3058 noinvpcid [X86] Disable the INVPCID cpu feature.
3060 nojitter [IA-64] Disables jitter checking for ITC timers.
3062 no-kvmclock [X86,KVM] Disable paravirtualized KVM clock driver
3064 no-kvmapf [X86,KVM] Disable paravirtualized asynchronous page
3068 [X86,PV_OPS] Disable paravirtualized VMware scheduler
3069 clock and use the default one.
3071 no-steal-acc [X86,KVM] Disable paravirtualized steal time accounting.
3072 steal time is computed, but won't influence scheduler
3075 nolapic [X86-32,APIC] Do not enable or use the local APIC.
3077 nolapic_timer [X86-32,APIC] Do not use the local APIC timer.
3079 noltlbs [PPC] Do not use large page/tlb entries for kernel
3080 lowmem mapping on PPC40x and PPC8xx
3082 nomca [IA-64] Disable machine check abort handling
3084 nomce [X86-32] Disable Machine Check Exception
3086 nomfgpt [X86-32] Disable Multi-Function General Purpose
3087 Timer usage (for AMD Geode machines).
3089 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to
3090 shutdown the other cpus. Instead use the REBOOT_VECTOR
3093 nomodule Disable module load
3095 nopat [X86] Disable PAT (page attribute table extension of
3096 pagetables) support.
3098 nopcid [X86-64] Disable the PCID cpu feature.
3100 norandmaps Don't use address space randomization. Equivalent to
3101 echo 0 > /proc/sys/kernel/randomize_va_space
3103 noreplace-smp [X86-32,SMP] Don't replace SMP instructions
3104 with UP alternatives
3106 nordrand [X86] Disable kernel use of the RDRAND and
3107 RDSEED instructions even if they are supported
3108 by the processor. RDRAND and RDSEED are still
3109 available to user space applications.
3111 noresume [SWSUSP] Disables resume and restores original swap
3114 no-scroll [VGA] Disables scrollback.
3115 This is required for the Braillex ib80-piezo Braille
3116 reader made by F.H. Papenmeier (Germany).
3120 nosep [BUGS=X86-32] Disables x86 SYSENTER/SYSEXIT support.
3122 nosmp [SMP] Tells an SMP kernel to act as a UP kernel,
3123 and disable the IO APIC. legacy for "maxcpus=0".
3125 nosoftlockup [KNL] Disable the soft-lockup detector.
3127 nosync [HW,M68K] Disables sync negotiation for all devices.
3129 nowatchdog [KNL] Disable both lockup detectors, i.e.
3130 soft-lockup and NMI watchdog (hard-lockup).
3134 nox2apic [X86-64,APIC] Do not enable x2APIC mode.
3136 cpu0_hotplug [X86] Turn on CPU0 hotplug feature when
3137 CONFIG_BOOTPARAM_HOTPLUG_CPU0 is off.
3138 Some features depend on CPU0. Known dependencies are:
3139 1. Resume from suspend/hibernate depends on CPU0.
3140 Suspend/hibernate will fail if CPU0 is offline and you
3141 need to online CPU0 before suspend/hibernate.
3142 2. PIC interrupts also depend on CPU0. CPU0 can't be
3143 removed if a PIC interrupt is detected.
3144 It's said poweroff/reboot may depend on CPU0 on some
3145 machines although I haven't seen such issues so far
3146 after CPU0 is offline on a few tested machines.
3147 If the dependencies are under your control, you can
3148 turn on cpu0_hotplug.
3150 nps_mtm_hs_ctr= [KNL,ARC]
3151 This parameter sets the maximum duration, in
3152 cycles, each HW thread of the CTOP can run
3153 without interruptions, before HW switches it.
3154 The actual maximum duration is 16 times this
3156 Format: integer between 1 and 255
3159 nptcg= [IA-64] Override max number of concurrent global TLB
3160 purges which is reported from either PAL_VM_SUMMARY or
3163 nr_cpus= [SMP] Maximum number of processors that an SMP kernel
3164 could support. nr_cpus=n : n >= 1 limits the kernel to
3165 support 'n' processors. It could be larger than the
3166 number of already plugged CPU during bootup, later in
3167 runtime you can physically add extra cpu until it reaches
3168 n. So during boot up some boot time memory for per-cpu
3169 variables need be pre-allocated for later physical cpu
3172 nr_uarts= [SERIAL] maximum number of UARTs to be registered.
3174 numa_balancing= [KNL,X86] Enable or disable automatic NUMA balancing.
3175 Allowed values are enable and disable
3177 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA.
3178 'node', 'default' can be specified
3179 This can be set from sysctl after boot.
3180 See Documentation/admin-guide/sysctl/vm.rst for details.
3182 ohci1394_dma=early [HW] enable debugging via the ohci1394 driver.
3183 See Documentation/debugging-via-ohci1394.txt for more
3186 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands
3187 Rather than timing out after 20 ms if an EC
3188 command is not properly ACKed, override the length
3189 of the timeout. We have interrupts disabled while
3190 waiting for the ACK, so if this is set too high
3191 interrupts *may* be lost!
3193 omap_mux= [OMAP] Override bootloader pin multiplexing.
3194 Format: <mux_mode0.mode_name=value>...
3195 For example, to override I2C bus2:
3196 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100
3198 oprofile.timer= [HW]
3199 Use timer interrupt instead of performance counters
3201 oprofile.cpu_type= Force an oprofile cpu type
3202 This might be useful if you have an older oprofile
3203 userland or if you want common events.
3204 Format: { arch_perfmon }
3205 arch_perfmon: [X86] Force use of architectural
3206 perfmon on Intel CPUs instead of the
3207 CPU specific event set.
3208 timer: [X86] Force use of architectural NMI
3209 timer mode (see also oprofile.timer
3210 for generic hr timer mode)
3212 oops=panic Always panic on oopses. Default is to just kill the
3213 process, but there is a small probability of
3214 deadlocking the machine.
3215 This will also cause panics on machine check exceptions.
3216 Useful together with panic=30 to trigger a reboot.
3219 [KNL] Boolean flag to control whether the page allocator
3220 should randomize its free lists. The randomization may
3221 be automatically enabled if the kernel detects it is
3222 running on a platform with a direct-mapped memory-side
3223 cache, and this parameter can be used to
3224 override/disable that behavior. The state of the flag
3225 can be read from sysfs at:
3226 /sys/module/page_alloc/parameters/shuffle.
3228 page_owner= [KNL] Boot-time page_owner enabling option.
3229 Storage of the information about who allocated
3230 each page is disabled in default. With this switch,
3232 on: enable the feature
3234 page_poison= [KNL] Boot-time parameter changing the state of
3235 poisoning on the buddy allocator, available with
3236 CONFIG_PAGE_POISONING=y.
3237 off: turn off poisoning (default)
3238 on: turn on poisoning
3240 panic= [KNL] Kernel behaviour on panic: delay <timeout>
3241 timeout > 0: seconds before rebooting
3242 timeout = 0: wait forever
3243 timeout < 0: reboot immediately
3246 panic_print= Bitmask for printing system info when panic happens.
3247 User can chose combination of the following bits:
3248 bit 0: print all tasks info
3249 bit 1: print system memory info
3250 bit 2: print timer info
3251 bit 3: print locks info if CONFIG_LOCKDEP is on
3252 bit 4: print ftrace buffer
3253 bit 5: print all printk messages in buffer
3255 panic_on_warn panic() instead of WARN(). Useful to cause kdump
3258 crash_kexec_post_notifiers
3259 Run kdump after running panic-notifiers and dumping
3260 kmsg. This only for the users who doubt kdump always
3261 succeeds in any situation.
3262 Note that this also increases risks of kdump failure,
3263 because some panic notifiers can make the crashed
3264 kernel more unstable.
3266 parkbd.port= [HW] Parallel port number the keyboard adapter is
3267 connected to, default is 0.
3269 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation,
3270 0 for XT, 1 for AT (default is AT).
3273 parport= [HW,PPT] Specify parallel ports. 0 disables.
3274 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] }
3275 Use 'auto' to force the driver to use any
3276 IRQ/DMA settings detected (the default is to
3277 ignore detected IRQ/DMA settings because of
3278 possible conflicts). You can specify the base
3279 address, IRQ, and DMA settings; IRQ and DMA
3280 should be numbers, or 'auto' (for using detected
3281 settings on that particular port), or 'nofifo'
3282 (to avoid using a FIFO even if it is detected).
3283 Parallel ports are assigned in the order they
3284 are specified on the command line, starting
3287 parport_init_mode= [HW,PPT]
3288 Configure VIA parallel port to operate in
3289 a specific mode. This is necessary on Pegasos
3290 computer where firmware has no options for setting
3291 up parallel port mode and sets it to spp.
3292 Currently this function knows 686a and 8231 chips.
3293 Format: [spp|ps2|epp|ecp|ecpepp]
3296 Halt all CPUs after the first oops has been printed for
3297 the specified number of seconds. This is to be used if
3298 your oopses keep scrolling off the screen.
3303 See header of drivers/block/paride/pcd.c.
3304 See also Documentation/admin-guide/blockdev/paride.rst.
3306 pci=option[,option...] [PCI] various PCI subsystem options.
3308 Some options herein operate on a specific device
3309 or a set of devices (<pci_dev>). These are
3310 specified in one of the following formats:
3312 [<domain>:]<bus>:<dev>.<func>[/<dev>.<func>]*
3313 pci:<vendor>:<device>[:<subvendor>:<subdevice>]
3315 Note: the first format specifies a PCI
3316 bus/device/function address which may change
3317 if new hardware is inserted, if motherboard
3318 firmware changes, or due to changes caused
3319 by other kernel parameters. If the
3320 domain is left unspecified, it is
3321 taken to be zero. Optionally, a path
3322 to a device through multiple device/function
3323 addresses can be specified after the base
3324 address (this is more robust against
3325 renumbering issues). The second format
3326 selects devices using IDs from the
3327 configuration space which may match multiple
3328 devices in the system.
3330 earlydump dump PCI config space before the kernel
3332 off [X86] don't probe for the PCI bus
3333 bios [X86-32] force use of PCI BIOS, don't access
3334 the hardware directly. Use this if your machine
3335 has a non-standard PCI host bridge.
3336 nobios [X86-32] disallow use of PCI BIOS, only direct
3337 hardware access methods are allowed. Use this
3338 if you experience crashes upon bootup and you
3339 suspect they are caused by the BIOS.
3340 conf1 [X86] Force use of PCI Configuration Access
3341 Mechanism 1 (config address in IO port 0xCF8,
3342 data in IO port 0xCFC, both 32-bit).
3343 conf2 [X86] Force use of PCI Configuration Access
3344 Mechanism 2 (IO port 0xCF8 is an 8-bit port for
3345 the function, IO port 0xCFA, also 8-bit, sets
3346 bus number. The config space is then accessed
3347 through ports 0xC000-0xCFFF).
3348 See http://wiki.osdev.org/PCI for more info
3349 on the configuration access mechanisms.
3350 noaer [PCIE] If the PCIEAER kernel config parameter is
3351 enabled, this kernel boot option can be used to
3352 disable the use of PCIE advanced error reporting.
3353 nodomains [PCI] Disable support for multiple PCI
3354 root domains (aka PCI segments, in ACPI-speak).
3355 nommconf [X86] Disable use of MMCONFIG for PCI
3357 check_enable_amd_mmconf [X86] check for and enable
3358 properly configured MMIO access to PCI
3359 config space on AMD family 10h CPU
3360 nomsi [MSI] If the PCI_MSI kernel config parameter is
3361 enabled, this kernel boot option can be used to
3362 disable the use of MSI interrupts system-wide.
3363 noioapicquirk [APIC] Disable all boot interrupt quirks.
3364 Safety option to keep boot IRQs enabled. This
3365 should never be necessary.
3366 ioapicreroute [APIC] Enable rerouting of boot IRQs to the
3367 primary IO-APIC for bridges that cannot disable
3368 boot IRQs. This fixes a source of spurious IRQs
3369 when the system masks IRQs.
3370 noioapicreroute [APIC] Disable workaround that uses the
3371 boot IRQ equivalent of an IRQ that connects to
3372 a chipset where boot IRQs cannot be disabled.
3373 The opposite of ioapicreroute.
3374 biosirq [X86-32] Use PCI BIOS calls to get the interrupt
3375 routing table. These calls are known to be buggy
3376 on several machines and they hang the machine
3377 when used, but on other computers it's the only
3378 way to get the interrupt routing table. Try
3379 this option if the kernel is unable to allocate
3380 IRQs or discover secondary PCI buses on your
3382 rom [X86] Assign address space to expansion ROMs.
3383 Use with caution as certain devices share
3384 address decoders between ROMs and other
3386 norom [X86] Do not assign address space to
3387 expansion ROMs that do not already have
3388 BIOS assigned address ranges.
3389 nobar [X86] Do not assign address space to the
3390 BARs that weren't assigned by the BIOS.
3391 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be
3392 assigned automatically to PCI devices. You can
3393 make the kernel exclude IRQs of your ISA cards
3395 pirqaddr=0xAAAAA [X86] Specify the physical address
3396 of the PIRQ table (normally generated
3397 by the BIOS) if it is outside the
3398 F0000h-100000h range.
3399 lastbus=N [X86] Scan all buses thru bus #N. Can be
3400 useful if the kernel is unable to find your
3401 secondary buses and you want to tell it
3402 explicitly which ones they are.
3403 assign-busses [X86] Always assign all PCI bus
3404 numbers ourselves, overriding
3405 whatever the firmware may have done.
3406 usepirqmask [X86] Honor the possible IRQ mask stored
3407 in the BIOS $PIR table. This is needed on
3408 some systems with broken BIOSes, notably
3409 some HP Pavilion N5400 and Omnibook XE3
3410 notebooks. This will have no effect if ACPI
3411 IRQ routing is enabled.
3412 noacpi [X86] Do not use ACPI for IRQ routing
3413 or for PCI scanning.
3414 use_crs [X86] Use PCI host bridge window information
3415 from ACPI. On BIOSes from 2008 or later, this
3416 is enabled by default. If you need to use this,
3417 please report a bug.
3418 nocrs [X86] Ignore PCI host bridge windows from ACPI.
3419 If you need to use this, please report a bug.
3420 routeirq Do IRQ routing for all PCI devices.
3421 This is normally done in pci_enable_device(),
3422 so this option is a temporary workaround
3423 for broken drivers that don't call it.
3424 skip_isa_align [X86] do not align io start addr, so can
3425 handle more pci cards
3426 noearly [X86] Don't do any early type 1 scanning.
3427 This might help on some broken boards which
3428 machine check when some devices' config space
3429 is read. But various workarounds are disabled
3430 and some IOMMU drivers will not work.
3431 bfsort Sort PCI devices into breadth-first order.
3432 This sorting is done to get a device
3433 order compatible with older (<= 2.4) kernels.
3434 nobfsort Don't sort PCI devices into breadth-first order.
3435 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size)
3436 tuning and use the BIOS-configured MPS defaults.
3437 pcie_bus_safe Set every device's MPS to the largest value
3438 supported by all devices below the root complex.
3439 pcie_bus_perf Set device MPS to the largest allowable MPS
3440 based on its parent bus. Also set MRRS (Max
3441 Read Request Size) to the largest supported
3442 value (no larger than the MPS that the device
3443 or bus can support) for best performance.
3444 pcie_bus_peer2peer Set every device's MPS to 128B, which
3445 every device is guaranteed to support. This
3446 configuration allows peer-to-peer DMA between
3447 any pair of devices, possibly at the cost of
3448 reduced performance. This also guarantees
3449 that hot-added devices will work.
3450 cbiosize=nn[KMG] The fixed amount of bus space which is
3451 reserved for the CardBus bridge's IO window.
3452 The default value is 256 bytes.
3453 cbmemsize=nn[KMG] The fixed amount of bus space which is
3454 reserved for the CardBus bridge's memory
3455 window. The default value is 64 megabytes.
3458 [<order of align>@]<pci_dev>[; ...]
3459 Specifies alignment and device to reassign
3460 aligned memory resources. How to
3461 specify the device is described above.
3462 If <order of align> is not specified,
3463 PAGE_SIZE is used as alignment.
3464 PCI-PCI bridge can be specified, if resource
3465 windows need to be expanded.
3466 To specify the alignment for several
3467 instances of a device, the PCI vendor,
3468 device, subvendor, and subdevice may be
3469 specified, e.g., 4096@pci:8086:9c22:103c:198f
3470 ecrc= Enable/disable PCIe ECRC (transaction layer
3471 end-to-end CRC checking).
3472 bios: Use BIOS/firmware settings. This is the
3476 hpiosize=nn[KMG] The fixed amount of bus space which is
3477 reserved for hotplug bridge's IO window.
3478 Default size is 256 bytes.
3479 hpmemsize=nn[KMG] The fixed amount of bus space which is
3480 reserved for hotplug bridge's memory window.
3481 Default size is 2 megabytes.
3482 hpbussize=nn The minimum amount of additional bus numbers
3483 reserved for buses below a hotplug bridge.
3485 realloc= Enable/disable reallocating PCI bridge resources
3486 if allocations done by BIOS are too small to
3487 accommodate resources required by all child
3489 off: Turn realloc off
3491 realloc same as realloc=on
3492 noari do not use PCIe ARI.
3493 noats [PCIE, Intel-IOMMU, AMD-IOMMU]
3494 do not use PCIe ATS (and IOMMU device IOTLB).
3495 pcie_scan_all Scan all possible PCIe devices. Otherwise we
3496 only look for one device below a PCIe downstream
3498 big_root_window Try to add a big 64bit memory window to the PCIe
3499 root complex on AMD CPUs. Some GFX hardware
3500 can resize a BAR to allow access to all VRAM.
3501 Adding the window is slightly risky (it may
3502 conflict with unreported devices), so this
3504 disable_acs_redir=<pci_dev>[; ...]
3505 Specify one or more PCI devices (in the format
3506 specified above) separated by semicolons.
3507 Each device specified will have the PCI ACS
3508 redirect capabilities forced off which will
3509 allow P2P traffic between devices through
3510 bridges without forcing it upstream. Note:
3511 this removes isolation between devices and
3512 may put more devices in an IOMMU group.
3513 force_floating [S390] Force usage of floating interrupts.
3514 nomio [S390] Do not use MIO instructions.
3516 pcie_aspm= [PCIE] Forcibly enable or disable PCIe Active State Power
3519 force Enable ASPM even on devices that claim not to support it.
3520 WARNING: Forcing ASPM on may cause system lockups.
3522 pcie_ports= [PCIE] PCIe port services handling:
3523 native Use native PCIe services (PME, AER, DPC, PCIe hotplug)
3524 even if the platform doesn't give the OS permission to
3525 use them. This may cause conflicts if the platform
3526 also tries to use these services.
3527 compat Disable native PCIe services (PME, AER, DPC, PCIe
3530 pcie_port_pm= [PCIE] PCIe port power management handling:
3531 off Disable power management of all PCIe ports
3532 force Forcibly enable power management of all PCIe ports
3534 pcie_pme= [PCIE,PM] Native PCIe PME signaling options:
3535 nomsi Do not use MSI for native PCIe PME signaling (this makes
3536 all PCIe root ports use INTx for all services).
3538 pcmv= [HW,PCMCIA] BadgePAD 4
3542 Keep all power-domains already enabled by bootloader on,
3543 even if no driver has claimed them. This is useful
3544 for debug and development, but should not be
3545 needed on a platform with proper driver support.
3548 See Documentation/admin-guide/blockdev/paride.rst.
3550 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at
3553 See arch/parisc/kernel/pdc_chassis.c
3555 percpu_alloc= Select which percpu first chunk allocator to use.
3556 Currently supported values are "embed" and "page".
3557 Archs may support subset or none of the selections.
3558 See comments in mm/percpu.c for details on each
3559 allocator. This parameter is primarily for debugging
3560 and performance comparison.
3563 See Documentation/admin-guide/blockdev/paride.rst.
3566 See Documentation/admin-guide/blockdev/paride.rst.
3568 pirq= [SMP,APIC] Manual mp-table setup
3569 See Documentation/x86/i386/IO-APIC.rst.
3571 plip= [PPT,NET] Parallel port network link
3572 Format: { parport<nr> | timid | 0 }
3573 See also Documentation/admin-guide/parport.rst.
3575 pmtmr= [X86] Manual setup of pmtmr I/O Port.
3576 Override pmtimer IOPort with a hex value.
3580 Enable PNP debug messages (depends on the
3581 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time
3582 via /sys/module/pnp/parameters/debug. We always show
3583 current resource usage; turning this on also shows
3584 possible settings and some assignment information.
3590 { on | off | curr | res | no-curr | no-res }
3593 [ISAPNP] Exclude IRQs for the autoconfiguration
3596 [ISAPNP] Exclude DMAs for the autoconfiguration
3598 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration
3599 Ranges are in pairs (I/O port base and size).
3602 [ISAPNP] Exclude memory regions for the
3604 Ranges are in pairs (memory base and size).
3606 ports= [IP_VS_FTP] IPVS ftp helper module
3608 Up to 8 (IP_VS_APP_MAX_PORTS) ports
3610 Format: <port>,<port>....
3612 powersave=off [PPC] This option disables power saving features.
3613 It specifically disables cpuidle and sets the
3614 platform machine description specific power_save
3615 function to NULL. On Idle the CPU just reduces
3618 ppc_strict_facility_enable
3619 [PPC] This option catches any kernel floating point,
3620 Altivec, VSX and SPE outside of regions specifically
3621 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()).
3622 There is some performance impact when enabling this.
3626 Disable Hardware Transactional Memory
3628 print-fatal-signals=
3629 [KNL] debug: print fatal signals
3631 If enabled, warn about various signal handling
3632 related application anomalies: too many signals,
3633 too many POSIX.1 timers, fatal signals causing a
3636 If you hit the warning due to signal overflow,
3637 you might want to try "ulimit -i unlimited".
3641 printk.always_kmsg_dump=
3642 Trigger kmsg_dump for cases other than kernel oops or
3644 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
3647 printk.devkmsg={on,off,ratelimit}
3648 Control writing to /dev/kmsg.
3649 on - unlimited logging to /dev/kmsg from userspace
3650 off - logging to /dev/kmsg disabled
3651 ratelimit - ratelimit the logging
3654 printk.time= Show timing data prefixed to each printk message line
3655 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
3657 processor.max_cstate= [HW,ACPI]
3658 Limit processor to maximum C-state
3659 max_cstate=9 overrides any DMI blacklist limit.
3661 processor.nocst [HW,ACPI]
3662 Ignore the _CST method to determine C-states,
3663 instead using the legacy FADT method
3665 profile= [KNL] Enable kernel profiling via /proc/profile
3666 Format: [<profiletype>,]<number>
3667 Param: <profiletype>: "schedule", "sleep", or "kvm"
3668 [defaults to kernel profiling]
3669 Param: "schedule" - profile schedule points.
3670 Param: "sleep" - profile D-state sleeping (millisecs).
3671 Requires CONFIG_SCHEDSTATS
3672 Param: "kvm" - profile VM exits.
3673 Param: <number> - step/bucket size as a power of 2 for
3674 statistical time based profiling.
3676 prompt_ramdisk= [RAM] List of RAM disks to prompt for floppy disk
3678 See Documentation/admin-guide/blockdev/ramdisk.rst.
3680 psi= [KNL] Enable or disable pressure stall information
3684 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to
3685 probe for; one of (bare|imps|exps|lifebook|any).
3686 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports
3688 psmouse.resetafter= [HW,MOUSE]
3689 Try to reset the device after so many bad packets
3692 [HW,MOUSE] Set desired mouse resolution, in dpi.
3693 psmouse.smartscroll=
3694 [HW,MOUSE] Controls Logitech smartscroll autorepeat.
3695 0 = disabled, 1 = enabled (default).
3697 pstore.backend= Specify the name of the pstore backend to use
3700 See Documentation/admin-guide/blockdev/paride.rst.
3702 pti= [X86_64] Control Page Table Isolation of user and
3703 kernel address spaces. Disabling this feature
3704 removes hardening, but improves performance of
3705 system calls and interrupts.
3707 on - unconditionally enable
3708 off - unconditionally disable
3709 auto - kernel detects whether your CPU model is
3710 vulnerable to issues that PTI mitigates
3712 Not specifying this option is equivalent to pti=auto.
3715 Equivalent to pti=off
3718 [KNL] Number of legacy pty's. Overwrites compiled-in
3721 quiet [KNL] Disable most log messages
3726 See Documentation/admin-guide/md.rst.
3728 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes
3729 See Documentation/admin-guide/blockdev/ramdisk.rst.
3731 random.trust_cpu={on,off}
3732 [KNL] Enable or disable trusting the use of the
3733 CPU's random number generator (if available) to
3734 fully seed the kernel's CRNG. Default is controlled
3735 by CONFIG_RANDOM_TRUST_CPU.
3737 ras=option[,option,...] [KNL] RAS-specific options
3740 Disable the Correctable Errors Collector,
3741 see CONFIG_RAS_CEC help text.
3744 The argument is a cpu list, as described above,
3745 except that the string "all" can be used to
3746 specify every CPU on the system.
3748 In kernels built with CONFIG_RCU_NOCB_CPU=y, set
3749 the specified list of CPUs to be no-callback CPUs.
3750 Invocation of these CPUs' RCU callbacks will be
3751 offloaded to "rcuox/N" kthreads created for that
3752 purpose, where "x" is "p" for RCU-preempt, and
3753 "s" for RCU-sched, and "N" is the CPU number.
3754 This reduces OS jitter on the offloaded CPUs,
3755 which can be useful for HPC and real-time
3756 workloads. It can also improve energy efficiency
3757 for asymmetric multiprocessors.
3760 Rather than requiring that offloaded CPUs
3761 (specified by rcu_nocbs= above) explicitly
3762 awaken the corresponding "rcuoN" kthreads,
3763 make these kthreads poll for callbacks.
3764 This improves the real-time response for the
3765 offloaded CPUs by relieving them of the need to
3766 wake up the corresponding kthread, but degrades
3767 energy efficiency by requiring that the kthreads
3768 periodically wake up to do the polling.
3770 rcutree.blimit= [KNL]
3771 Set maximum number of finished RCU callbacks to
3772 process in one batch.
3774 rcutree.dump_tree= [KNL]
3775 Dump the structure of the rcu_node combining tree
3776 out at early boot. This is used for diagnostic
3777 purposes, to verify correct tree setup.
3779 rcutree.gp_cleanup_delay= [KNL]
3780 Set the number of jiffies to delay each step of
3781 RCU grace-period cleanup.
3783 rcutree.gp_init_delay= [KNL]
3784 Set the number of jiffies to delay each step of
3785 RCU grace-period initialization.
3787 rcutree.gp_preinit_delay= [KNL]
3788 Set the number of jiffies to delay each step of
3789 RCU grace-period pre-initialization, that is,
3790 the propagation of recent CPU-hotplug changes up
3791 the rcu_node combining tree.
3793 rcutree.use_softirq= [KNL]
3794 If set to zero, move all RCU_SOFTIRQ processing to
3795 per-CPU rcuc kthreads. Defaults to a non-zero
3796 value, meaning that RCU_SOFTIRQ is used by default.
3797 Specify rcutree.use_softirq=0 to use rcuc kthreads.
3799 rcutree.rcu_fanout_exact= [KNL]
3800 Disable autobalancing of the rcu_node combining
3801 tree. This is used by rcutorture, and might
3802 possibly be useful for architectures having high
3803 cache-to-cache transfer latencies.
3805 rcutree.rcu_fanout_leaf= [KNL]
3806 Change the number of CPUs assigned to each
3807 leaf rcu_node structure. Useful for very
3808 large systems, which will choose the value 64,
3809 and for NUMA systems with large remote-access
3810 latencies, which will choose a value aligned
3811 with the appropriate hardware boundaries.
3813 rcutree.jiffies_till_first_fqs= [KNL]
3814 Set delay from grace-period initialization to
3815 first attempt to force quiescent states.
3816 Units are jiffies, minimum value is zero,
3817 and maximum value is HZ.
3819 rcutree.jiffies_till_next_fqs= [KNL]
3820 Set delay between subsequent attempts to force
3821 quiescent states. Units are jiffies, minimum
3822 value is one, and maximum value is HZ.
3824 rcutree.jiffies_till_sched_qs= [KNL]
3825 Set required age in jiffies for a
3826 given grace period before RCU starts
3827 soliciting quiescent-state help from
3828 rcu_note_context_switch() and cond_resched().
3829 If not specified, the kernel will calculate
3830 a value based on the most recent settings
3831 of rcutree.jiffies_till_first_fqs
3832 and rcutree.jiffies_till_next_fqs.
3833 This calculated value may be viewed in
3834 rcutree.jiffies_to_sched_qs. Any attempt to set
3835 rcutree.jiffies_to_sched_qs will be cheerfully
3838 rcutree.kthread_prio= [KNL,BOOT]
3839 Set the SCHED_FIFO priority of the RCU per-CPU
3840 kthreads (rcuc/N). This value is also used for
3841 the priority of the RCU boost threads (rcub/N)
3842 and for the RCU grace-period kthreads (rcu_bh,
3843 rcu_preempt, and rcu_sched). If RCU_BOOST is
3844 set, valid values are 1-99 and the default is 1
3845 (the least-favored priority). Otherwise, when
3846 RCU_BOOST is not set, valid values are 0-99 and
3847 the default is zero (non-realtime operation).
3849 rcutree.rcu_nocb_gp_stride= [KNL]
3850 Set the number of NOCB callback kthreads in
3851 each group, which defaults to the square root
3852 of the number of CPUs. Larger numbers reduce
3853 the wakeup overhead on the global grace-period
3854 kthread, but increases that same overhead on
3855 each group's NOCB grace-period kthread.
3857 rcutree.qhimark= [KNL]
3858 Set threshold of queued RCU callbacks beyond which
3859 batch limiting is disabled.
3861 rcutree.qlowmark= [KNL]
3862 Set threshold of queued RCU callbacks below which
3863 batch limiting is re-enabled.
3865 rcutree.rcu_idle_gp_delay= [KNL]
3866 Set wakeup interval for idle CPUs that have
3867 RCU callbacks (RCU_FAST_NO_HZ=y).
3869 rcutree.rcu_idle_lazy_gp_delay= [KNL]
3870 Set wakeup interval for idle CPUs that have
3871 only "lazy" RCU callbacks (RCU_FAST_NO_HZ=y).
3872 Lazy RCU callbacks are those which RCU can
3873 prove do nothing more than free memory.
3875 rcutree.rcu_kick_kthreads= [KNL]
3876 Cause the grace-period kthread to get an extra
3877 wake_up() if it sleeps three times longer than
3878 it should at force-quiescent-state time.
3879 This wake_up() will be accompanied by a
3880 WARN_ONCE() splat and an ftrace_dump().
3882 rcutree.sysrq_rcu= [KNL]
3883 Commandeer a sysrq key to dump out Tree RCU's
3884 rcu_node tree with an eye towards determining
3885 why a new grace period has not yet started.
3887 rcuperf.gp_async= [KNL]
3888 Measure performance of asynchronous
3889 grace-period primitives such as call_rcu().
3891 rcuperf.gp_async_max= [KNL]
3892 Specify the maximum number of outstanding
3893 callbacks per writer thread. When a writer
3894 thread exceeds this limit, it invokes the
3895 corresponding flavor of rcu_barrier() to allow
3896 previously posted callbacks to drain.
3898 rcuperf.gp_exp= [KNL]
3899 Measure performance of expedited synchronous
3900 grace-period primitives.
3902 rcuperf.holdoff= [KNL]
3903 Set test-start holdoff period. The purpose of
3904 this parameter is to delay the start of the
3905 test until boot completes in order to avoid
3908 rcuperf.nreaders= [KNL]
3909 Set number of RCU readers. The value -1 selects
3910 N, where N is the number of CPUs. A value
3911 "n" less than -1 selects N-n+1, where N is again
3912 the number of CPUs. For example, -2 selects N
3913 (the number of CPUs), -3 selects N+1, and so on.
3914 A value of "n" less than or equal to -N selects
3917 rcuperf.nwriters= [KNL]
3918 Set number of RCU writers. The values operate
3919 the same as for rcuperf.nreaders.
3920 N, where N is the number of CPUs
3922 rcuperf.perf_type= [KNL]
3923 Specify the RCU implementation to test.
3925 rcuperf.shutdown= [KNL]
3926 Shut the system down after performance tests
3927 complete. This is useful for hands-off automated
3930 rcuperf.verbose= [KNL]
3931 Enable additional printk() statements.
3933 rcuperf.writer_holdoff= [KNL]
3934 Write-side holdoff between grace periods,
3935 in microseconds. The default of zero says
3938 rcutorture.fqs_duration= [KNL]
3939 Set duration of force_quiescent_state bursts
3942 rcutorture.fqs_holdoff= [KNL]
3943 Set holdoff time within force_quiescent_state bursts
3946 rcutorture.fqs_stutter= [KNL]
3947 Set wait time between force_quiescent_state bursts
3950 rcutorture.fwd_progress= [KNL]
3951 Enable RCU grace-period forward-progress testing
3952 for the types of RCU supporting this notion.
3954 rcutorture.fwd_progress_div= [KNL]
3955 Specify the fraction of a CPU-stall-warning
3956 period to do tight-loop forward-progress testing.
3958 rcutorture.fwd_progress_holdoff= [KNL]
3959 Number of seconds to wait between successive
3960 forward-progress tests.
3962 rcutorture.fwd_progress_need_resched= [KNL]
3963 Enclose cond_resched() calls within checks for
3964 need_resched() during tight-loop forward-progress
3967 rcutorture.gp_cond= [KNL]
3968 Use conditional/asynchronous update-side
3969 primitives, if available.
3971 rcutorture.gp_exp= [KNL]
3972 Use expedited update-side primitives, if available.
3974 rcutorture.gp_normal= [KNL]
3975 Use normal (non-expedited) asynchronous
3976 update-side primitives, if available.
3978 rcutorture.gp_sync= [KNL]
3979 Use normal (non-expedited) synchronous
3980 update-side primitives, if available. If all
3981 of rcutorture.gp_cond=, rcutorture.gp_exp=,
3982 rcutorture.gp_normal=, and rcutorture.gp_sync=
3983 are zero, rcutorture acts as if is interpreted
3984 they are all non-zero.
3986 rcutorture.n_barrier_cbs= [KNL]
3987 Set callbacks/threads for rcu_barrier() testing.
3989 rcutorture.nfakewriters= [KNL]
3990 Set number of concurrent RCU writers. These just
3991 stress RCU, they don't participate in the actual
3992 test, hence the "fake".
3994 rcutorture.nreaders= [KNL]
3995 Set number of RCU readers. The value -1 selects
3996 N-1, where N is the number of CPUs. A value
3997 "n" less than -1 selects N-n-2, where N is again
3998 the number of CPUs. For example, -2 selects N
3999 (the number of CPUs), -3 selects N+1, and so on.
4001 rcutorture.object_debug= [KNL]
4002 Enable debug-object double-call_rcu() testing.
4004 rcutorture.onoff_holdoff= [KNL]
4005 Set time (s) after boot for CPU-hotplug testing.
4007 rcutorture.onoff_interval= [KNL]
4008 Set time (jiffies) between CPU-hotplug operations,
4009 or zero to disable CPU-hotplug testing.
4011 rcutorture.shuffle_interval= [KNL]
4012 Set task-shuffle interval (s). Shuffling tasks
4013 allows some CPUs to go into dyntick-idle mode
4014 during the rcutorture test.
4016 rcutorture.shutdown_secs= [KNL]
4017 Set time (s) after boot system shutdown. This
4018 is useful for hands-off automated testing.
4020 rcutorture.stall_cpu= [KNL]
4021 Duration of CPU stall (s) to test RCU CPU stall
4022 warnings, zero to disable.
4024 rcutorture.stall_cpu_holdoff= [KNL]
4025 Time to wait (s) after boot before inducing stall.
4027 rcutorture.stall_cpu_irqsoff= [KNL]
4028 Disable interrupts while stalling if set.
4030 rcutorture.stat_interval= [KNL]
4031 Time (s) between statistics printk()s.
4033 rcutorture.stutter= [KNL]
4034 Time (s) to stutter testing, for example, specifying
4035 five seconds causes the test to run for five seconds,
4036 wait for five seconds, and so on. This tests RCU's
4037 ability to transition abruptly to and from idle.
4039 rcutorture.test_boost= [KNL]
4040 Test RCU priority boosting? 0=no, 1=maybe, 2=yes.
4041 "Maybe" means test if the RCU implementation
4042 under test support RCU priority boosting.
4044 rcutorture.test_boost_duration= [KNL]
4045 Duration (s) of each individual boost test.
4047 rcutorture.test_boost_interval= [KNL]
4048 Interval (s) between each boost test.
4050 rcutorture.test_no_idle_hz= [KNL]
4051 Test RCU's dyntick-idle handling. See also the
4052 rcutorture.shuffle_interval parameter.
4054 rcutorture.torture_type= [KNL]
4055 Specify the RCU implementation to test.
4057 rcutorture.verbose= [KNL]
4058 Enable additional printk() statements.
4060 rcupdate.rcu_cpu_stall_ftrace_dump= [KNL]
4061 Dump ftrace buffer after reporting RCU CPU
4064 rcupdate.rcu_cpu_stall_suppress= [KNL]
4065 Suppress RCU CPU stall warning messages.
4067 rcupdate.rcu_cpu_stall_timeout= [KNL]
4068 Set timeout for RCU CPU stall warning messages.
4070 rcupdate.rcu_expedited= [KNL]
4071 Use expedited grace-period primitives, for
4072 example, synchronize_rcu_expedited() instead
4073 of synchronize_rcu(). This reduces latency,
4074 but can increase CPU utilization, degrade
4075 real-time latency, and degrade energy efficiency.
4076 No effect on CONFIG_TINY_RCU kernels.
4078 rcupdate.rcu_normal= [KNL]
4079 Use only normal grace-period primitives,
4080 for example, synchronize_rcu() instead of
4081 synchronize_rcu_expedited(). This improves
4082 real-time latency, CPU utilization, and
4083 energy efficiency, but can expose users to
4084 increased grace-period latency. This parameter
4085 overrides rcupdate.rcu_expedited. No effect on
4086 CONFIG_TINY_RCU kernels.
4088 rcupdate.rcu_normal_after_boot= [KNL]
4089 Once boot has completed (that is, after
4090 rcu_end_inkernel_boot() has been invoked), use
4091 only normal grace-period primitives. No effect
4092 on CONFIG_TINY_RCU kernels.
4094 rcupdate.rcu_task_stall_timeout= [KNL]
4095 Set timeout in jiffies for RCU task stall warning
4096 messages. Disable with a value less than or equal
4099 rcupdate.rcu_self_test= [KNL]
4100 Run the RCU early boot self tests
4104 Run specified binary instead of /init from the ramdisk,
4105 used for early userspace startup. See initrd.
4108 force - Override the decision by the kernel to hide the
4109 advertisement of RDRAND support (this affects
4110 certain AMD processors because of buggy BIOS
4111 support, specifically around the suspend/resume
4115 Turn on/off individual RDT features. List is:
4116 cmt, mbmtotal, mbmlocal, l3cat, l3cdp, l2cat, l2cdp,
4118 E.g. to turn on cmt and turn off mba use:
4122 Format (x86 or x86_64):
4123 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] \
4125 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \
4127 Where reboot_mode is one of warm (soft) or cold (hard) or gpio
4128 (prefix with 'panic_' to set mode for panic
4130 reboot_type is one of bios, acpi, kbd, triple, efi, or pci,
4131 reboot_force is either force or not specified,
4132 reboot_cpu is s[mp]#### with #### being the processor
4133 to be used for rebooting.
4136 [KNL, SMP] Set scheduler's default relax_domain_level.
4137 See Documentation/admin-guide/cgroup-v1/cpusets.rst.
4139 reserve= [KNL,BUGS] Force kernel to ignore I/O ports or memory
4140 Format: <base1>,<size1>[,<base2>,<size2>,...]
4141 Reserve I/O ports or memory so the kernel won't use
4142 them. If <base> is less than 0x10000, the region
4143 is assumed to be I/O ports; otherwise it is memory.
4145 reservetop= [X86-32]
4147 Reserves a hole at the top of the kernel virtual
4152 Set the amount of memory to reserve for BIOS at
4153 the bottom of the address space.
4155 reset_devices [KNL] Force drivers to reset the underlying device
4156 during initialization.
4159 Specify the partition device for software suspend
4161 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>}
4163 resume_offset= [SWSUSP]
4164 Specify the offset from the beginning of the partition
4165 given by "resume=" at which the swap header is located,
4166 in <PAGE_SIZE> units (needed only for swap files).
4167 See Documentation/power/swsusp-and-swap-files.rst
4169 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to
4170 read the resume files
4172 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up.
4173 Useful for devices that are detected asynchronously
4174 (e.g. USB and MMC devices).
4176 hibernate= [HIBERNATION]
4177 noresume Don't check if there's a hibernation image
4178 present during boot.
4179 nocompress Don't compress/decompress hibernation images.
4180 no Disable hibernation and resume.
4181 protect_image Turn on image protection during restoration
4182 (that will set all pages holding image data
4183 during restoration read-only).
4185 retain_initrd [RAM] Keep initrd memory after extraction
4187 rfkill.default_state=
4188 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm,
4189 etc. communication is blocked by default.
4192 rfkill.master_switch_mode=
4193 0 The "airplane mode" button does nothing.
4194 1 The "airplane mode" button toggles between everything
4195 blocked and the previous configuration.
4196 2 The "airplane mode" button toggles between everything
4197 blocked and everything unblocked.
4199 rhash_entries= [KNL,NET]
4200 Set number of hash buckets for route cache
4203 [KNL] Disable ring 3 MONITOR/MWAIT feature on supported
4206 ro [KNL] Mount root device read-only on boot
4209 on Mark read-only kernel memory as read-only (default).
4210 off Leave read-only kernel memory writable for debugging.
4213 Enable the uart passthrough on the designated usb port
4214 on Rockchip SoCs. When active, the signals of the
4215 debug-uart get routed to the D+ and D- pins of the usb
4216 port and the regular usb controller gets disabled.
4218 root= [KNL] Root filesystem
4219 See name_to_dev_t comment in init/do_mounts.c.
4221 rootdelay= [KNL] Delay (in seconds) to pause before attempting to
4222 mount the root filesystem
4224 rootflags= [KNL] Set root filesystem mount option string
4226 rootfstype= [KNL] Set root filesystem type
4228 rootwait [KNL] Wait (indefinitely) for root device to show up.
4229 Useful for devices that are detected asynchronously
4230 (e.g. USB and MMC devices).
4232 rproc_mem=nn[KMG][@address]
4233 [KNL,ARM,CMA] Remoteproc physical memory block.
4234 Memory area to be used by remote processor image,
4237 rw [KNL] Mount root device read-write on boot
4239 S [KNL] Run init in single mode
4241 s390_iommu= [HW,S390]
4242 Set s390 IOTLB flushing mode
4244 With strict flushing every unmap operation will result in
4245 an IOTLB flush. Default is lazy flushing before reuse,
4249 See drivers/net/irda/sa1100_ir.c.
4251 sbni= [NET] Granch SBNI12 leased line adapter
4253 sched_debug [KNL] Enables verbose scheduler debug messages.
4255 schedstats= [KNL,X86] Enable or disable scheduled statistics.
4256 Allowed values are enable and disable. This feature
4257 incurs a small amount of overhead in the scheduler
4258 but is useful for debugging and performance tuning.
4260 skew_tick= [KNL] Offset the periodic timer tick per cpu to mitigate
4261 xtime_lock contention on larger systems, and/or RCU lock
4262 contention on all systems with CONFIG_MAXSMP set.
4263 Format: { "0" | "1" }
4264 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1"
4266 Note: increases power consumption, thus should only be
4267 enabled if running jitter sensitive (HPC/RT) workloads.
4269 security= [SECURITY] Choose a legacy "major" security module to
4270 enable at boot. This has been deprecated by the
4273 selinux= [SELINUX] Disable or enable SELinux at boot time.
4274 Format: { "0" | "1" }
4275 See security/selinux/Kconfig help text.
4278 Default value is set via kernel config option.
4279 If enabled at boot time, /selinux/disable can be used
4280 later to disable prior to initial policy load.
4282 apparmor= [APPARMOR] Disable or enable AppArmor at boot time
4283 Format: { "0" | "1" }
4284 See security/apparmor/Kconfig help text
4287 Default value is set via kernel config option.
4289 serialnumber [BUGS=X86-32]
4292 Maximal number of shapers.
4300 Disable merging of slabs with similar size. May be
4301 necessary if there is some reason to distinguish
4302 allocs to different slabs, especially in hardened
4303 environments where the risk of heap overflows and
4304 layout control by attackers can usually be
4305 frustrated by disabling merging. This will reduce
4306 most of the exposure of a heap attack to a single
4307 cache (risks via metadata attacks are mostly
4308 unchanged). Debug options disable merging on their
4310 For more information see Documentation/vm/slub.rst.
4312 slab_max_order= [MM, SLAB]
4313 Determines the maximum allowed order for slabs.
4314 A high setting may cause OOMs due to memory
4315 fragmentation. Defaults to 1 for systems with
4316 more than 32MB of RAM, 0 otherwise.
4318 slub_debug[=options[,slabs]] [MM, SLUB]
4319 Enabling slub_debug allows one to determine the
4320 culprit if slab objects become corrupted. Enabling
4321 slub_debug can create guard zones around objects and
4322 may poison objects when not in use. Also tracks the
4323 last alloc / free. For more information see
4324 Documentation/vm/slub.rst.
4326 slub_memcg_sysfs= [MM, SLUB]
4327 Determines whether to enable sysfs directories for
4328 memory cgroup sub-caches. 1 to enable, 0 to disable.
4329 The default is determined by CONFIG_SLUB_MEMCG_SYSFS_ON.
4330 Enabling this can lead to a very high number of debug
4331 directories and files being created under
4334 slub_max_order= [MM, SLUB]
4335 Determines the maximum allowed order for slabs.
4336 A high setting may cause OOMs due to memory
4337 fragmentation. For more information see
4338 Documentation/vm/slub.rst.
4340 slub_min_objects= [MM, SLUB]
4341 The minimum number of objects per slab. SLUB will
4342 increase the slab order up to slub_max_order to
4343 generate a sufficiently large slab able to contain
4344 the number of objects indicated. The higher the number
4345 of objects the smaller the overhead of tracking slabs
4346 and the less frequently locks need to be acquired.
4347 For more information see Documentation/vm/slub.rst.
4349 slub_min_order= [MM, SLUB]
4350 Determines the minimum page order for slabs. Must be
4351 lower than slub_max_order.
4352 For more information see Documentation/vm/slub.rst.
4354 slub_nomerge [MM, SLUB]
4355 Same with slab_nomerge. This is supported for legacy.
4356 See slab_nomerge for more information.
4359 Format: <io1>[,<io2>[,...,<io8>]]
4361 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices
4362 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port
4363 smsc-ircc2.ircc_sir= [HW] SIR base I/O port
4364 smsc-ircc2.ircc_fir= [HW] FIR base I/O port
4365 smsc-ircc2.ircc_irq= [HW] IRQ line
4366 smsc-ircc2.ircc_dma= [HW] DMA channel
4367 smsc-ircc2.ircc_transceiver= [HW] Transceiver type:
4368 0: Toshiba Satellite 1800 (GP data pin select)
4369 1: Fast pin select (default)
4372 smt [KNL,S390] Set the maximum number of threads (logical
4373 CPUs) to use per physical CPU on systems capable of
4374 symmetric multithreading (SMT). Will be capped to the
4375 actual hardware limit.
4377 Default: -1 (no limit)
4380 [KNL] Should the soft-lockup detector generate panics.
4383 A nonzero value instructs the soft-lockup detector
4384 to panic the machine when a soft-lockup occurs. This
4385 is also controlled by CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC
4386 which is the respective build-time switch to that
4389 softlockup_all_cpu_backtrace=
4390 [KNL] Should the soft-lockup detector generate
4391 backtraces on all cpus.
4394 sonypi.*= [HW] Sony Programmable I/O Control Device driver
4395 See Documentation/admin-guide/laptops/sonypi.rst
4397 spectre_v2= [X86] Control mitigation of Spectre variant 2
4398 (indirect branch speculation) vulnerability.
4399 The default operation protects the kernel from
4402 on - unconditionally enable, implies
4404 off - unconditionally disable, implies
4406 auto - kernel detects whether your CPU model is
4409 Selecting 'on' will, and 'auto' may, choose a
4410 mitigation method at run time according to the
4411 CPU, the available microcode, the setting of the
4412 CONFIG_RETPOLINE configuration option, and the
4413 compiler with which the kernel was built.
4415 Selecting 'on' will also enable the mitigation
4416 against user space to user space task attacks.
4418 Selecting 'off' will disable both the kernel and
4419 the user space protections.
4421 Specific mitigations can also be selected manually:
4423 retpoline - replace indirect branches
4424 retpoline,generic - google's original retpoline
4425 retpoline,amd - AMD-specific minimal thunk
4427 Not specifying this option is equivalent to
4431 [X86] Control mitigation of Spectre variant 2
4432 (indirect branch speculation) vulnerability between
4435 on - Unconditionally enable mitigations. Is
4436 enforced by spectre_v2=on
4438 off - Unconditionally disable mitigations. Is
4439 enforced by spectre_v2=off
4441 prctl - Indirect branch speculation is enabled,
4442 but mitigation can be enabled via prctl
4443 per thread. The mitigation control state
4444 is inherited on fork.
4447 - Like "prctl" above, but only STIBP is
4448 controlled per thread. IBPB is issued
4449 always when switching between different user
4453 - Same as "prctl" above, but all seccomp
4454 threads will enable the mitigation unless
4455 they explicitly opt out.
4458 - Like "seccomp" above, but only STIBP is
4459 controlled per thread. IBPB is issued
4460 always when switching between different
4461 user space processes.
4463 auto - Kernel selects the mitigation depending on
4464 the available CPU features and vulnerability.
4467 If CONFIG_SECCOMP=y then "seccomp", otherwise "prctl"
4469 Not specifying this option is equivalent to
4470 spectre_v2_user=auto.
4472 spec_store_bypass_disable=
4473 [HW] Control Speculative Store Bypass (SSB) Disable mitigation
4474 (Speculative Store Bypass vulnerability)
4476 Certain CPUs are vulnerable to an exploit against a
4477 a common industry wide performance optimization known
4478 as "Speculative Store Bypass" in which recent stores
4479 to the same memory location may not be observed by
4480 later loads during speculative execution. The idea
4481 is that such stores are unlikely and that they can
4482 be detected prior to instruction retirement at the
4483 end of a particular speculation execution window.
4485 In vulnerable processors, the speculatively forwarded
4486 store can be used in a cache side channel attack, for
4487 example to read memory to which the attacker does not
4488 directly have access (e.g. inside sandboxed code).
4490 This parameter controls whether the Speculative Store
4491 Bypass optimization is used.
4493 On x86 the options are:
4495 on - Unconditionally disable Speculative Store Bypass
4496 off - Unconditionally enable Speculative Store Bypass
4497 auto - Kernel detects whether the CPU model contains an
4498 implementation of Speculative Store Bypass and
4499 picks the most appropriate mitigation. If the
4500 CPU is not vulnerable, "off" is selected. If the
4501 CPU is vulnerable the default mitigation is
4502 architecture and Kconfig dependent. See below.
4503 prctl - Control Speculative Store Bypass per thread
4504 via prctl. Speculative Store Bypass is enabled
4505 for a process by default. The state of the control
4506 is inherited on fork.
4507 seccomp - Same as "prctl" above, but all seccomp threads
4508 will disable SSB unless they explicitly opt out.
4510 Default mitigations:
4511 X86: If CONFIG_SECCOMP=y "seccomp", otherwise "prctl"
4513 On powerpc the options are:
4515 on,auto - On Power8 and Power9 insert a store-forwarding
4516 barrier on kernel entry and exit. On Power7
4517 perform a software flush on kernel entry and
4521 Not specifying this option is equivalent to
4522 spec_store_bypass_disable=auto.
4524 spia_io_base= [HW,MTD]
4529 srcutree.counter_wrap_check [KNL]
4530 Specifies how frequently to check for
4531 grace-period sequence counter wrap for the
4532 srcu_data structure's ->srcu_gp_seq_needed field.
4533 The greater the number of bits set in this kernel
4534 parameter, the less frequently counter wrap will
4535 be checked for. Note that the bottom two bits
4538 srcutree.exp_holdoff [KNL]
4539 Specifies how many nanoseconds must elapse
4540 since the end of the last SRCU grace period for
4541 a given srcu_struct until the next normal SRCU
4542 grace period will be considered for automatic
4543 expediting. Set to zero to disable automatic
4547 Speculative Store Bypass Disable control
4549 On CPUs that are vulnerable to the Speculative
4550 Store Bypass vulnerability and offer a
4551 firmware based mitigation, this parameter
4552 indicates how the mitigation should be used:
4554 force-on: Unconditionally enable mitigation for
4555 for both kernel and userspace
4556 force-off: Unconditionally disable mitigation for
4557 for both kernel and userspace
4558 kernel: Always enable mitigation in the
4559 kernel, and offer a prctl interface
4560 to allow userspace to register its
4561 interest in being mitigated too.
4563 stack_guard_gap= [MM]
4564 override the default stack gap protection. The value
4565 is in page units and it defines how many pages prior
4566 to (for stacks growing down) resp. after (for stacks
4567 growing up) the main stack are reserved for no other
4568 mapping. Default value is 256 pages.
4571 Enabled the stack tracer on boot up.
4573 stacktrace_filter=[function-list]
4574 [FTRACE] Limit the functions that the stack tracer
4575 will trace at boot up. function-list is a comma separated
4576 list of functions. This list can be changed at run
4577 time by the stack_trace_filter file in the debugfs
4578 tracing directory. Note, this enables stack tracing
4579 and the stacktrace above is not needed.
4583 Set the STI (builtin display/keyboard on the HP-PARISC
4584 machines) console (graphic card) which should be used
4585 as the initial boot-console.
4586 See also comment in drivers/video/console/sticore.c.
4589 See comment in drivers/video/console/sticore.c.
4592 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]]
4594 sunrpc.min_resvport=
4595 sunrpc.max_resvport=
4597 SunRPC servers often require that client requests
4598 originate from a privileged port (i.e. a port in the
4599 range 0 < portnr < 1024).
4600 An administrator who wishes to reserve some of these
4601 ports for other uses may adjust the range that the
4602 kernel's sunrpc client considers to be privileged
4603 using these two parameters to set the minimum and
4604 maximum port values.
4606 sunrpc.svc_rpc_per_connection_limit=
4608 Limit the number of requests that the server will
4609 process in parallel from a single connection.
4610 The default value is 0 (no limit).
4614 Control how the NFS server code allocates CPUs to
4615 service thread pools. Depending on how many NICs
4616 you have and where their interrupts are bound, this
4617 option will affect which CPUs will do NFS serving.
4618 Note: this parameter cannot be changed while the
4619 NFS server is running.
4621 auto the server chooses an appropriate mode
4622 automatically using heuristics
4623 global a single global pool contains all CPUs
4624 percpu one pool for each CPU
4625 pernode one pool for each NUMA node (equivalent
4626 to global on non-NUMA machines)
4628 sunrpc.tcp_slot_table_entries=
4629 sunrpc.udp_slot_table_entries=
4631 Sets the upper limit on the number of simultaneous
4632 RPC calls that can be sent from the client to a
4633 server. Increasing these values may allow you to
4634 improve throughput, but will also increase the
4635 amount of memory reserved for use by the client.
4637 suspend.pm_test_delay=
4639 Sets the number of seconds to remain in a suspend test
4640 mode before resuming the system (see
4641 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG
4642 is set. Default value is 5.
4645 [KNL] Enable accounting of swap in memory resource
4646 controller if no parameter or 1 is given or disable
4647 it if 0 is given (See Documentation/admin-guide/cgroup-v1/memory.rst)
4649 swiotlb= [ARM,IA-64,PPC,MIPS,X86]
4650 Format: { <int> | force | noforce }
4651 <int> -- Number of I/O TLB slabs
4652 force -- force using of bounce buffers even if they
4653 wouldn't be automatically used by the kernel
4654 noforce -- Never use bounce buffers (for debugging)
4658 sysfs.deprecated=0|1 [KNL]
4659 Enable/disable old style sysfs layout for old udev
4660 on older distributions. When this option is enabled
4661 very new udev will not work anymore. When this option
4662 is disabled (or CONFIG_SYSFS_DEPRECATED not compiled)
4663 in older udev will not work anymore.
4664 Default depends on CONFIG_SYSFS_DEPRECATED_V2 set in
4665 the kernel configuration.
4667 sysrq_always_enabled
4669 Ignore sysrq setting - this boot parameter will
4670 neutralize any effect of /proc/sys/kernel/sysrq.
4671 Useful for debugging.
4673 tcpmhash_entries= [KNL,NET]
4674 Set the number of tcp_metrics_hash slots.
4675 Default value is 8192 or 16384 depending on total
4676 ram pages. This is used to specify the TCP metrics
4677 cache size. See Documentation/networking/ip-sysctl.txt
4678 "tcp_no_metrics_save" section for more details.
4682 test_suspend= [SUSPEND][,N]
4683 Specify "mem" (for Suspend-to-RAM) or "standby" (for
4684 standby suspend) or "freeze" (for suspend type freeze)
4685 as the system sleep state during system startup with
4686 the optional capability to repeat N number of times.
4687 The system is woken from this state using a
4688 wakeup-capable RTC alarm.
4690 thash_entries= [KNL,NET]
4691 Set number of hash buckets for TCP connection
4693 thermal.act= [HW,ACPI]
4694 -1: disable all active trip points in all thermal zones
4695 <degrees C>: override all lowest active trip points
4697 thermal.crt= [HW,ACPI]
4698 -1: disable all critical trip points in all thermal zones
4699 <degrees C>: override all critical trip points
4701 thermal.nocrt= [HW,ACPI]
4702 Set to disable actions on ACPI thermal zone
4703 critical and hot trip points.
4705 thermal.off= [HW,ACPI]
4706 1: disable ACPI thermal control
4708 thermal.psv= [HW,ACPI]
4709 -1: disable all passive trip points
4710 <degrees C>: override all passive trip points to this
4713 thermal.tzp= [HW,ACPI]
4714 Specify global default ACPI thermal zone polling rate
4715 <deci-seconds>: poll all this frequency
4716 0: no polling (default)
4719 Force threading of all interrupt handlers except those
4720 marked explicitly IRQF_NO_THREAD.
4724 Specify if the kernel should make use of the cpu
4725 topology information if the hardware supports this.
4726 The scheduler will make use of this information and
4727 e.g. base its process migration decisions on it.
4730 topology_updates= [KNL, PPC, NUMA]
4732 Specify if the kernel should ignore (off)
4733 topology updates sent by the hypervisor to this
4738 tpm_suspend_pcr=[HW,TPM]
4739 Format: integer pcr id
4740 Specify that at suspend time, the tpm driver
4741 should extend the specified pcr with zeros,
4742 as a workaround for some chips which fail to
4743 flush the last written pcr on TPM_SaveState.
4744 This will guarantee that all the other pcrs
4747 trace_buf_size=nn[KMG]
4748 [FTRACE] will set tracing buffer size on each cpu.
4750 trace_event=[event-list]
4751 [FTRACE] Set and start specified trace events in order
4752 to facilitate early boot debugging. The event-list is a
4753 comma separated list of trace events to enable. See
4754 also Documentation/trace/events.rst
4756 trace_options=[option-list]
4757 [FTRACE] Enable or disable tracer options at boot.
4758 The option-list is a comma delimited list of options
4759 that can be enabled or disabled just as if you were
4760 to echo the option name into
4762 /sys/kernel/debug/tracing/trace_options
4764 For example, to enable stacktrace option (to dump the
4765 stack trace of each event), add to the command line:
4767 trace_options=stacktrace
4769 See also Documentation/trace/ftrace.rst "trace options"
4773 Have the tracepoints sent to printk as well as the
4774 tracing ring buffer. This is useful for early boot up
4775 where the system hangs or reboots and does not give the
4776 option for reading the tracing buffer or performing a
4777 ftrace_dump_on_oops.
4779 To turn off having tracepoints sent to printk,
4780 echo 0 > /proc/sys/kernel/tracepoint_printk
4781 Note, echoing 1 into this file without the
4782 tracepoint_printk kernel cmdline option has no effect.
4786 Having tracepoints sent to printk() and activating high
4787 frequency tracepoints such as irq or sched, can cause
4788 the system to live lock.
4791 [FTRACE] enable this option to disable tracing when a
4792 warning is hit. This turns off "tracing_on". Tracing can
4793 be enabled again by echoing '1' into the "tracing_on"
4794 file located in /sys/kernel/debug/tracing/
4796 This option is useful, as it disables the trace before
4797 the WARNING dump is called, which prevents the trace to
4798 be filled with content caused by the warning output.
4800 This option can also be set at run time via the sysctl
4801 option: kernel/traceoff_on_warning
4803 transparent_hugepage=
4805 Format: [always|madvise|never]
4806 Can be used to control the default behavior of the system
4807 with respect to transparent hugepages.
4808 See Documentation/admin-guide/mm/transhuge.rst
4811 tsc= Disable clocksource stability checks for TSC.
4813 [x86] reliable: mark tsc clocksource as reliable, this
4814 disables clocksource verification at runtime, as well
4815 as the stability checks done at bootup. Used to enable
4816 high-resolution timer mode on older hardware, and in
4817 virtualized environment.
4818 [x86] noirqtime: Do not use TSC to do irq accounting.
4819 Used to run time disable IRQ_TIME_ACCOUNTING on any
4820 platforms where RDTSC is slow and this accounting
4822 [x86] unstable: mark the TSC clocksource as unstable, this
4823 marks the TSC unconditionally unstable at bootup and
4824 avoids any further wobbles once the TSC watchdog notices.
4825 [x86] nowatchdog: disable clocksource watchdog. Used
4826 in situations with strict latency requirements (where
4827 interruptions from clocksource watchdog are not
4830 turbografx.map[2|3]= [HW,JOY]
4831 TurboGraFX parallel port interface
4833 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7>
4834 See also Documentation/input/devices/joystick-parport.rst
4836 udbg-immortal [PPC] When debugging early kernel crashes that
4837 happen after console_init() and before a proper
4838 console driver takes over, this boot options might
4839 help "seeing" what's going on.
4841 uhash_entries= [KNL,NET]
4842 Set number of hash buckets for UDP/UDP-Lite connections
4845 [USB] Ignore overcurrent events (default N).
4846 Some badly-designed motherboards generate lots of
4847 bogus events, for ports that aren't wired to
4848 anything. Set this parameter to avoid log spamming.
4849 Note that genuine overcurrent events won't be
4853 [X86] Cause panic on unknown NMI.
4855 usbcore.authorized_default=
4856 [USB] Default USB device authorization:
4857 (default -1 = authorized except for wireless USB,
4858 0 = not authorized, 1 = authorized, 2 = authorized
4859 if device connected to internal port)
4861 usbcore.autosuspend=
4862 [USB] The autosuspend time delay (in seconds) used
4863 for newly-detected USB devices (default 2). This
4864 is the time required before an idle device will be
4865 autosuspended. Devices for which the delay is set
4866 to a negative value won't be autosuspended at all.
4868 usbcore.usbfs_snoop=
4869 [USB] Set to log all usbfs traffic (default 0 = off).
4871 usbcore.usbfs_snoop_max=
4872 [USB] Maximum number of bytes to snoop in each URB
4875 usbcore.blinkenlights=
4876 [USB] Set to cycle leds on hubs (default 0 = off).
4878 usbcore.old_scheme_first=
4879 [USB] Start with the old device initialization
4880 scheme, applies only to low and full-speed devices
4883 usbcore.usbfs_memory_mb=
4884 [USB] Memory limit (in MB) for buffers allocated by
4885 usbfs (default = 16, 0 = max = 2047).
4887 usbcore.use_both_schemes=
4888 [USB] Try the other device initialization scheme
4889 if the first one fails (default 1 = enabled).
4891 usbcore.initial_descriptor_timeout=
4892 [USB] Specifies timeout for the initial 64-byte
4893 USB_REQ_GET_DESCRIPTOR request in milliseconds
4894 (default 5000 = 5.0 seconds).
4896 usbcore.nousb [USB] Disable the USB subsystem
4899 [USB] A list of quirk entries to augment the built-in
4900 usb core quirk list. List entries are separated by
4901 commas. Each entry has the form
4902 VendorID:ProductID:Flags. The IDs are 4-digit hex
4903 numbers and Flags is a set of letters. Each letter
4904 will change the built-in quirk; setting it if it is
4905 clear and clearing it if it is set. The letters have
4906 the following meanings:
4907 a = USB_QUIRK_STRING_FETCH_255 (string
4908 descriptors must not be fetched using
4910 b = USB_QUIRK_RESET_RESUME (device can't resume
4911 correctly so reset it instead);
4912 c = USB_QUIRK_NO_SET_INTF (device can't handle
4913 Set-Interface requests);
4914 d = USB_QUIRK_CONFIG_INTF_STRINGS (device can't
4915 handle its Configuration or Interface
4917 e = USB_QUIRK_RESET (device can't be reset
4918 (e.g morph devices), don't use reset);
4919 f = USB_QUIRK_HONOR_BNUMINTERFACES (device has
4920 more interface descriptions than the
4921 bNumInterfaces count, and can't handle
4922 talking to these interfaces);
4923 g = USB_QUIRK_DELAY_INIT (device needs a pause
4924 during initialization, after we read
4925 the device descriptor);
4926 h = USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL (For
4927 high speed and super speed interrupt
4928 endpoints, the USB 2.0 and USB 3.0 spec
4929 require the interval in microframes (1
4930 microframe = 125 microseconds) to be
4931 calculated as interval = 2 ^
4933 Devices with this quirk report their
4934 bInterval as the result of this
4935 calculation instead of the exponent
4936 variable used in the calculation);
4937 i = USB_QUIRK_DEVICE_QUALIFIER (device can't
4938 handle device_qualifier descriptor
4940 j = USB_QUIRK_IGNORE_REMOTE_WAKEUP (device
4941 generates spurious wakeup, ignore
4942 remote wakeup capability);
4943 k = USB_QUIRK_NO_LPM (device can't handle Link
4945 l = USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL
4946 (Device reports its bInterval as linear
4947 frames instead of the USB 2.0
4949 m = USB_QUIRK_DISCONNECT_SUSPEND (Device needs
4950 to be disconnected before suspend to
4951 prevent spurious wakeup);
4952 n = USB_QUIRK_DELAY_CTRL_MSG (Device needs a
4953 pause after every control message);
4954 o = USB_QUIRK_HUB_SLOW_RESET (Hub needs extra
4955 delay after resetting its port);
4956 Example: quirks=0781:5580:bk,0a5c:5834:gij
4959 [USBHID] The interval which mice are to be polled at.
4962 [USBHID] The interval which joysticks are to be polled at.
4965 [USBHID] The interval which keyboards are to be polled at.
4967 usb-storage.delay_use=
4968 [UMS] The delay in seconds before a new device is
4969 scanned for Logical Units (default 1).
4972 [UMS] A list of quirks entries to supplement or
4973 override the built-in unusual_devs list. List
4974 entries are separated by commas. Each entry has
4975 the form VID:PID:Flags where VID and PID are Vendor
4976 and Product ID values (4-digit hex numbers) and
4977 Flags is a set of characters, each corresponding
4978 to a common usb-storage quirk flag as follows:
4979 a = SANE_SENSE (collect more than 18 bytes
4981 b = BAD_SENSE (don't collect more than 18
4982 bytes of sense data);
4983 c = FIX_CAPACITY (decrease the reported
4984 device capacity by one sector);
4985 d = NO_READ_DISC_INFO (don't use
4986 READ_DISC_INFO command);
4987 e = NO_READ_CAPACITY_16 (don't use
4988 READ_CAPACITY_16 command);
4989 f = NO_REPORT_OPCODES (don't use report opcodes
4991 g = MAX_SECTORS_240 (don't transfer more than
4992 240 sectors at a time, uas only);
4993 h = CAPACITY_HEURISTICS (decrease the
4994 reported device capacity by one
4995 sector if the number is odd);
4996 i = IGNORE_DEVICE (don't bind to this
4998 j = NO_REPORT_LUNS (don't use report luns
5000 l = NOT_LOCKABLE (don't try to lock and
5001 unlock ejectable media);
5002 m = MAX_SECTORS_64 (don't transfer more
5003 than 64 sectors = 32 KB at a time);
5004 n = INITIAL_READ10 (force a retry of the
5005 initial READ(10) command);
5006 o = CAPACITY_OK (accept the capacity
5007 reported by the device);
5008 p = WRITE_CACHE (the device cache is ON
5010 r = IGNORE_RESIDUE (the device reports
5011 bogus residue values);
5012 s = SINGLE_LUN (the device has only one
5014 t = NO_ATA_1X (don't allow ATA(12) and ATA(16)
5015 commands, uas only);
5016 u = IGNORE_UAS (don't bind to the uas driver);
5017 w = NO_WP_DETECT (don't test whether the
5018 medium is write-protected).
5019 y = ALWAYS_SYNC (issue a SYNCHRONIZE_CACHE
5020 even if the device claims no cache)
5021 Example: quirks=0419:aaf5:rl,0421:0433:rc
5023 user_debug= [KNL,ARM]
5025 See arch/arm/Kconfig.debug help text.
5026 1 - undefined instruction events
5028 4 - invalid data aborts
5031 Example: user_debug=31
5034 [X86] Flags controlling user PTE allocations.
5036 nohigh = do not allocate PTE pages in
5037 HIGHMEM regardless of setting
5041 On X86_32, this is an alias for vdso32=. Otherwise:
5043 vdso=1: enable VDSO (the default)
5044 vdso=0: disable VDSO mapping
5046 vdso32= [X86] Control the 32-bit vDSO
5047 vdso32=1: enable 32-bit VDSO
5048 vdso32=0 or vdso32=2: disable 32-bit VDSO
5050 See the help text for CONFIG_COMPAT_VDSO for more
5051 details. If CONFIG_COMPAT_VDSO is set, the default is
5052 vdso32=0; otherwise, the default is vdso32=1.
5054 For compatibility with older kernels, vdso32=2 is an
5057 Try vdso32=0 if you encounter an error that says:
5058 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed!
5061 vector=percpu: enable percpu vector domain
5063 video= [FB] Frame buffer configuration
5064 See Documentation/fb/modedb.rst.
5066 video.brightness_switch_enabled= [0,1]
5067 If set to 1, on receiving an ACPI notify event
5068 generated by hotkey, video driver will adjust brightness
5069 level and then send out the event to user space through
5070 the allocated input device; If set to 0, video driver
5071 will only send out the event without touching backlight
5076 [VMMIO] Memory mapped virtio (platform) device.
5078 <size>@<baseaddr>:<irq>[:<id>]
5080 <size> := size (can use standard suffixes
5082 <baseaddr> := physical base address
5083 <irq> := interrupt number (as passed to
5085 <id> := (optional) platform device id
5087 virtio_mmio.device=1K@0x100b0000:48:7
5089 Can be used multiple times for multiple devices.
5091 vga= [BOOT,X86-32] Select a particular video mode
5092 See Documentation/x86/boot.rst and
5093 Documentation/admin-guide/svga.rst.
5094 Use vga=ask for menu.
5095 This is actually a boot loader parameter; the value is
5096 passed to the kernel using a special protocol.
5098 vm_debug[=options] [KNL] Available with CONFIG_DEBUG_VM=y.
5099 May slow down system boot speed, especially when
5100 enabled on systems with a large amount of memory.
5101 All options are enabled by default, and this
5102 interface is meant to allow for selectively
5103 enabling or disabling specific virtual memory
5106 Available options are:
5107 P Enable page structure init time poisoning
5108 - Disable all of the above options
5110 vmalloc=nn[KMG] [KNL,BOOT] Forces the vmalloc area to have an exact
5111 size of <nn>. This can be used to increase the
5112 minimum size (128MB on x86). It can also be used to
5113 decrease the size and leave more room for directly
5116 vmcp_cma=nn[MG] [KNL,S390]
5117 Sets the memory size reserved for contiguous memory
5118 allocations for the vmcp device driver.
5120 vmhalt= [KNL,S390] Perform z/VM CP command after system halt.
5123 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic.
5126 vmpoff= [KNL,S390] Perform z/VM CP command after power off.
5130 Controls the behavior of vsyscalls (i.e. calls to
5131 fixed addresses of 0xffffffffff600x00 from legacy
5132 code). Most statically-linked binaries and older
5133 versions of glibc use these calls. Because these
5134 functions are at fixed addresses, they make nice
5135 targets for exploits that can control RIP.
5137 emulate [default] Vsyscalls turn into traps and are
5138 emulated reasonably safely. The vsyscall
5141 xonly Vsyscalls turn into traps and are
5142 emulated reasonably safely. The vsyscall
5143 page is not readable.
5145 none Vsyscalls don't work at all. This makes
5146 them quite hard to use for exploits but
5147 might break your system.
5149 vt.color= [VT] Default text color.
5150 Format: 0xYX, X = foreground, Y = background.
5151 Default: 0x07 = light gray on black.
5153 vt.cur_default= [VT] Default cursor shape.
5154 Format: 0xCCBBAA, where AA, BB, and CC are the same as
5155 the parameters of the <Esc>[?A;B;Cc escape sequence;
5156 see VGA-softcursor.txt. Default: 2 = underline.
5158 vt.default_blu= [VT]
5159 Format: <blue0>,<blue1>,<blue2>,...,<blue15>
5160 Change the default blue palette of the console.
5161 This is a 16-member array composed of values
5164 vt.default_grn= [VT]
5165 Format: <green0>,<green1>,<green2>,...,<green15>
5166 Change the default green palette of the console.
5167 This is a 16-member array composed of values
5170 vt.default_red= [VT]
5171 Format: <red0>,<red1>,<red2>,...,<red15>
5172 Change the default red palette of the console.
5173 This is a 16-member array composed of values
5179 Set system-wide default UTF-8 mode for all tty's.
5180 Default is 1, i.e. UTF-8 mode is enabled for all
5181 newly opened terminals.
5183 vt.global_cursor_default=
5186 Set system-wide default for whether a cursor
5187 is shown on new VTs. Default is -1,
5188 i.e. cursors will be created by default unless
5189 overridden by individual drivers. 0 will hide
5190 cursors, 1 will display them.
5192 vt.italic= [VT] Default color for italic text; 0-15.
5195 vt.underline= [VT] Default color for underlined text; 0-15.
5198 watchdog timers [HW,WDT] For information on watchdog timers,
5199 see Documentation/watchdog/watchdog-parameters.rst
5200 or other driver-specific files in the
5201 Documentation/watchdog/ directory.
5205 Set the hard lockup detector stall duration
5206 threshold in seconds. The soft lockup detector
5207 threshold is set to twice the value. A value of 0
5208 disables both lockup detectors. Default is 10
5211 workqueue.watchdog_thresh=
5212 If CONFIG_WQ_WATCHDOG is configured, workqueue can
5213 warn stall conditions and dump internal state to
5214 help debugging. 0 disables workqueue stall
5215 detection; otherwise, it's the stall threshold
5216 duration in seconds. The default value is 30 and
5217 it can be updated at runtime by writing to the
5218 corresponding sysfs file.
5220 workqueue.disable_numa
5221 By default, all work items queued to unbound
5222 workqueues are affine to the NUMA nodes they're
5223 issued on, which results in better behavior in
5224 general. If NUMA affinity needs to be disabled for
5225 whatever reason, this option can be used. Note
5226 that this also can be controlled per-workqueue for
5227 workqueues visible under /sys/bus/workqueue/.
5229 workqueue.power_efficient
5230 Per-cpu workqueues are generally preferred because
5231 they show better performance thanks to cache
5232 locality; unfortunately, per-cpu workqueues tend to
5233 be more power hungry than unbound workqueues.
5235 Enabling this makes the per-cpu workqueues which
5236 were observed to contribute significantly to power
5237 consumption unbound, leading to measurably lower
5238 power usage at the cost of small performance
5241 The default value of this parameter is determined by
5242 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT.
5244 workqueue.debug_force_rr_cpu
5245 Workqueue used to implicitly guarantee that work
5246 items queued without explicit CPU specified are put
5247 on the local CPU. This guarantee is no longer true
5248 and while local CPU is still preferred work items
5249 may be put on foreign CPUs. This debug option
5250 forces round-robin CPU selection to flush out
5251 usages which depend on the now broken guarantee.
5252 When enabled, memory and cache locality will be
5255 x2apic_phys [X86-64,APIC] Use x2apic physical mode instead of
5256 default x2apic cluster mode on platforms
5259 x86_intel_mid_timer= [X86-32,APBT]
5260 Choose timer option for x86 Intel MID platform.
5261 Two valid options are apbt timer only and lapic timer
5262 plus one apbt timer for broadcast timer.
5263 x86_intel_mid_timer=apbt_only | lapic_and_apbt
5265 xen_512gb_limit [KNL,X86-64,XEN]
5266 Restricts the kernel running paravirtualized under Xen
5267 to use only up to 512 GB of RAM. The reason to do so is
5268 crash analysis tools and Xen tools for doing domain
5269 save/restore/migration must be enabled to handle larger
5272 xen_emul_unplug= [HW,X86,XEN]
5273 Unplug Xen emulated devices
5274 Format: [unplug0,][unplug1]
5275 ide-disks -- unplug primary master IDE devices
5276 aux-ide-disks -- unplug non-primary-master IDE devices
5277 nics -- unplug network devices
5278 all -- unplug all emulated devices (NICs and IDE disks)
5279 unnecessary -- unplugging emulated devices is
5280 unnecessary even if the host did not respond to
5282 never -- do not unplug even if version check succeeds
5284 xen_nopvspin [X86,XEN]
5285 Disables the ticketlock slowpath using Xen PV
5289 Disables the PV optimizations forcing the HVM guest to
5290 run as generic HVM guest with no PV drivers.
5291 This option is obsoleted by the "nopv" option, which
5292 has equivalent effect for XEN platform.
5294 xen_scrub_pages= [XEN]
5295 Boolean option to control scrubbing pages before giving them back
5296 to Xen, for use by other domains. Can be also changed at runtime
5297 with /sys/devices/system/xen_memory/xen_memory0/scrub_pages.
5298 Default value controlled with CONFIG_XEN_SCRUB_PAGES_DEFAULT.
5300 xen_timer_slop= [X86-64,XEN]
5301 Set the timer slop (in nanoseconds) for the virtual Xen
5302 timers (default is 100000). This adjusts the minimum
5303 delta of virtualized Xen timers, where lower values
5304 improve timer resolution at the expense of processing
5305 more timer interrupts.
5307 nopv= [X86,XEN,KVM,HYPER_V,VMWARE]
5308 Disables the PV optimizations forcing the guest to run
5309 as generic guest with no PV drivers. Currently support
5310 XEN HVM, KVM, HYPER_V and VMWARE guest.
5312 xirc2ps_cs= [NET,PCMCIA]
5314 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]]
5317 By default on POWER9 and above, the kernel will
5318 natively use the XIVE interrupt controller. This option
5319 allows the fallback firmware mode to be used:
5321 off Fallback to firmware control of XIVE interrupt
5322 controller on both pseries and powernv
5323 platforms. Only useful on POWER9 and above.
5325 xhci-hcd.quirks [USB,KNL]
5326 A hex value specifying bitmask with supplemental xhci
5327 host controller quirks. Meaning of each bit can be
5328 consulted in header drivers/usb/host/xhci.h.