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
864 Disable TLBIE instruction. Currently does not work
865 with KVM, with HASH MMU, or with coherent accelerators.
867 disable_cpu_apicid= [X86,APIC,SMP]
869 The number of initial APIC ID for the
870 corresponding CPU to be disabled at boot,
871 mostly used for the kdump 2nd kernel to
872 disable BSP to wake up multiple CPUs without
873 causing system reset or hang due to sending
876 perf_v4_pmi= [X86,INTEL]
878 Disable Intel PMU counter freezing feature.
879 The feature only exists starting from
880 Arch Perfmon v4 (Skylake and newer).
882 disable_ddw [PPC/PSERIES]
883 Disable Dynamic DMA Window support. Use this if
884 to workaround buggy firmware.
887 See Documentation/networking/ipv6.txt.
889 disable_mtrr_cleanup [X86]
890 The kernel tries to adjust MTRR layout from continuous
891 to discrete, to make X server driver able to add WB
892 entry later. This parameter disables that.
894 disable_mtrr_trim [X86, Intel and AMD only]
895 By default the kernel will trim any uncacheable
896 memory out of your available memory pool based on
897 MTRR settings. This parameter disables that behavior,
898 possibly causing your machine to run very slowly.
900 disable_timer_pin_1 [X86]
901 Disable PIN 1 of APIC timer
902 Can be useful to work around chipset bugs.
904 dis_ucode_ldr [X86] Disable the microcode loader.
906 dma_debug=off If the kernel is compiled with DMA_API_DEBUG support,
907 this option disables the debugging code at boot.
909 dma_debug_entries=<number>
910 This option allows to tune the number of preallocated
911 entries for DMA-API debugging code. One entry is
912 required per DMA-API allocation. Use this if the
913 DMA-API debugging code disables itself because the
914 architectural default is too low.
916 dma_debug_driver=<driver_name>
917 With this option the DMA-API debugging driver
918 filter feature can be enabled at boot time. Just
919 pass the driver to filter for as the parameter.
920 The filter can be disabled or changed to another
921 driver later using sysfs.
923 driver_async_probe= [KNL]
924 List of driver names to be probed asynchronously.
925 Format: <driver_name1>,<driver_name2>...
927 drm.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>]
928 Broken monitors, graphic adapters, KVMs and EDIDless
929 panels may send no or incorrect EDID data sets.
930 This parameter allows to specify an EDID data sets
931 in the /lib/firmware directory that are used instead.
932 Generic built-in EDID data sets are used, if one of
933 edid/1024x768.bin, edid/1280x1024.bin,
934 edid/1680x1050.bin, or edid/1920x1080.bin is given
935 and no file with the same name exists. Details and
936 instructions how to build your own EDID data are
937 available in Documentation/driver-api/edid.rst. An EDID
938 data set will only be used for a particular connector,
939 if its name and a colon are prepended to the EDID
940 name. Each connector may use a unique EDID data
941 set by separating the files with a comma. An EDID
942 data set with no connector name will be used for
943 any connectors not explicitly specified.
948 Format: {"off" | "known"}
949 Control how the dt_cpu_ftrs device-tree binding is
950 used for CPU feature discovery and setup (if it
952 off: Do not use it, fall back to legacy cpu table.
953 known: Do not pass through unknown features to guests
954 or userspace, only those that the kernel is aware of.
956 dump_apple_properties [X86]
957 Dump name and content of EFI device properties on
958 x86 Macs. Useful for driver authors to determine
959 what data is available or for reverse-engineering.
961 dyndbg[="val"] [KNL,DYNAMIC_DEBUG]
962 module.dyndbg[="val"]
963 Enable debug messages at boot time. See
964 Documentation/admin-guide/dynamic-debug-howto.rst
967 nompx [X86] Disables Intel Memory Protection Extensions.
968 See Documentation/x86/intel_mpx.rst for more
969 information about the feature.
971 nopku [X86] Disable Memory Protection Keys CPU feature found
974 module.async_probe [KNL]
975 Enable asynchronous probe on this module.
977 early_ioremap_debug [KNL]
978 Enable debug messages in early_ioremap support. This
979 is useful for tracking down temporary early mappings
980 which are not unmapped.
982 earlycon= [KNL] Output early console device and options.
984 [ARM64] The early console is determined by the
985 stdout-path property in device tree's chosen node,
986 or determined by the ACPI SPCR table.
988 [X86] When used with no options the early console is
989 determined by the ACPI SPCR table.
991 cdns,<addr>[,options]
992 Start an early, polled-mode console on a Cadence
993 (xuartps) serial port at the specified address. Only
994 supported option is baud rate. If baud rate is not
995 specified, the serial port must already be setup and
998 uart[8250],io,<addr>[,options]
999 uart[8250],mmio,<addr>[,options]
1000 uart[8250],mmio32,<addr>[,options]
1001 uart[8250],mmio32be,<addr>[,options]
1002 uart[8250],0x<addr>[,options]
1003 Start an early, polled-mode console on the 8250/16550
1004 UART at the specified I/O port or MMIO address.
1005 MMIO inter-register address stride is either 8-bit
1006 (mmio) or 32-bit (mmio32 or mmio32be).
1007 If none of [io|mmio|mmio32|mmio32be], <addr> is assumed
1008 to be equivalent to 'mmio'. 'options' are specified
1009 in the same format described for "console=ttyS<n>"; if
1010 unspecified, the h/w is not initialized.
1014 Start an early, polled-mode console on a pl011 serial
1015 port at the specified address. The pl011 serial port
1016 must already be setup and configured. Options are not
1017 yet supported. If 'mmio32' is specified, then only
1018 the driver will use only 32-bit accessors to read/write
1019 the device registers.
1022 Start an early, polled-mode console on a meson serial
1023 port at the specified address. The serial port must
1024 already be setup and configured. Options are not yet
1028 Start an early, polled-mode console on an msm serial
1029 port at the specified address. The serial port
1030 must already be setup and configured. Options are not
1033 msm_serial_dm,<addr>
1034 Start an early, polled-mode console on an msm serial
1035 dm port at the specified address. The serial port
1036 must already be setup and configured. Options are not
1040 Start an early, polled-mode console on a serial port
1041 of an Actions Semi SoC, such as S500 or S900, at the
1042 specified address. The serial port must already be
1043 setup and configured. Options are not yet supported.
1046 Start an early, polled-mode console on a serial port
1047 of an RDA Micro SoC, such as RDA8810PL, at the
1048 specified address. The serial port must already be
1049 setup and configured. Options are not yet supported.
1052 Use RISC-V SBI (Supervisor Binary Interface) for early
1055 smh Use ARM semihosting calls for early console.
1063 Use early console provided by serial driver available
1064 on Samsung SoCs, requires selecting proper type and
1065 a correct base address of the selected UART port. The
1066 serial port must already be setup and configured.
1067 Options are not yet supported.
1070 Start an early, polled-mode console on a lantiq serial
1071 (lqasc) port at the specified address. The serial port
1072 must already be setup and configured. Options are not
1077 Use early console provided by Freescale LP UART driver
1078 found on Freescale Vybrid and QorIQ LS1021A processors.
1079 A valid base address must be provided, and the serial
1080 port must already be setup and configured.
1083 Start an early, polled-mode console on the
1084 Armada 3700 serial port at the specified
1085 address. The serial port must already be setup
1086 and configured. Options are not yet supported.
1089 Start an early, polled-mode console on a Qualcomm
1090 Generic Interface (GENI) based serial port at the
1091 specified address. The serial port must already be
1092 setup and configured. Options are not yet supported.
1095 Start an early, unaccelerated console on the EFI
1096 memory mapped framebuffer (if available). On cache
1097 coherent non-x86 systems that use system memory for
1098 the framebuffer, pass the 'ram' option so that it is
1099 mapped with the correct attributes.
1102 Use early console provided by Freescale LinFlex UART
1103 serial driver for NXP S32V234 SoCs. A valid base
1104 address must be provided, and the serial port must
1105 already be setup and configured.
1107 earlyprintk= [X86,SH,ARM,M68k,S390]
1111 earlyprintk=serial[,ttySn[,baudrate]]
1112 earlyprintk=serial[,0x...[,baudrate]]
1113 earlyprintk=ttySn[,baudrate]
1114 earlyprintk=dbgp[debugController#]
1115 earlyprintk=pciserial[,force],bus:device.function[,baudrate]
1116 earlyprintk=xdbc[xhciController#]
1118 earlyprintk is useful when the kernel crashes before
1119 the normal console is initialized. It is not enabled by
1120 default because it has some cosmetic problems.
1122 Append ",keep" to not disable it when the real console
1125 Only one of vga, efi, serial, or usb debug port can
1128 Currently only ttyS0 and ttyS1 may be specified by
1129 name. Other I/O ports may be explicitly specified
1130 on some architectures (x86 and arm at least) by
1131 replacing ttySn with an I/O port address, like this:
1132 earlyprintk=serial,0x1008,115200
1133 You can find the port for a given device in
1134 /proc/tty/driver/serial:
1135 2: uart:ST16650V2 port:00001008 irq:18 ...
1137 Interaction with the standard serial driver is not
1140 The VGA and EFI output is eventually overwritten by
1143 The xen output can only be used by Xen PV guests.
1145 The sclp output can only be used on s390.
1147 The optional "force" to "pciserial" enables use of a
1148 PCI device even when its classcode is not of the
1151 edac_report= [HW,EDAC] Control how to report EDAC event
1152 Format: {"on" | "off" | "force"}
1153 on: enable EDAC to report H/W event. May be overridden
1154 by other higher priority error reporting module.
1155 off: disable H/W event reporting through EDAC.
1156 force: enforce the use of EDAC to report H/W event.
1159 ekgdboc= [X86,KGDB] Allow early kernel console debugging
1162 This is designed to be used in conjunction with
1163 the boot argument: earlyprintk=vga
1166 Format: {"off" | "on" | "skip[mbr]"}
1169 Format: { "old_map", "nochunk", "noruntime", "debug" }
1170 old_map [X86-64]: switch to the old ioremap-based EFI
1171 runtime services mapping. 32-bit still uses this one by
1173 nochunk: disable reading files in "chunks" in the EFI
1174 boot stub, as chunking can cause problems with some
1175 firmware implementations.
1176 noruntime : disable EFI runtime services support
1177 debug: enable misc debug output
1179 efi_no_storage_paranoia [EFI; X86]
1180 Using this parameter you can use more than 50% of
1181 your efi variable storage. Use this parameter only if
1182 you are really sure that your UEFI does sane gc and
1183 fulfills the spec otherwise your board may brick.
1185 efi_fake_mem= nn[KMG]@ss[KMG]:aa[,nn[KMG]@ss[KMG]:aa,..] [EFI; X86]
1186 Add arbitrary attribute to specific memory range by
1187 updating original EFI memory map.
1188 Region of memory which aa attribute is added to is
1190 If efi_fake_mem=2G@4G:0x10000,2G@0x10a0000000:0x10000
1191 is specified, EFI_MEMORY_MORE_RELIABLE(0x10000)
1192 attribute is added to range 0x100000000-0x180000000 and
1193 0x10a0000000-0x1120000000.
1195 Using this parameter you can do debugging of EFI memmap
1196 related feature. For example, you can do debugging of
1197 Address Range Mirroring feature even if your box
1200 efivar_ssdt= [EFI; X86] Name of an EFI variable that contains an SSDT
1201 that is to be dynamically loaded by Linux. If there are
1202 multiple variables with the same name but with different
1203 vendor GUIDs, all of them will be loaded. See
1204 Documentation/admin-guide/acpi/ssdt-overlays.rst for details.
1207 eisa_irq_edge= [PARISC,HW]
1208 See header of drivers/parisc/eisa.c.
1211 See comment before function elanfreq_setup() in
1212 arch/x86/kernel/cpu/cpufreq/elanfreq.c.
1214 elfcorehdr=[size[KMG]@]offset[KMG] [IA64,PPC,SH,X86,S390]
1215 Specifies physical address of start of kernel core
1216 image elf header and optionally the size. Generally
1217 kexec loader will pass this option to capture kernel.
1218 See Documentation/admin-guide/kdump/kdump.rst for details.
1220 enable_mtrr_cleanup [X86]
1221 The kernel tries to adjust MTRR layout from continuous
1222 to discrete, to make X server driver able to add WB
1223 entry later. This parameter enables that.
1225 enable_timer_pin_1 [X86]
1226 Enable PIN 1 of APIC timer
1227 Can be useful to work around chipset bugs
1228 (in particular on some ATI chipsets).
1229 The kernel tries to set a reasonable default.
1231 enforcing [SELINUX] Set initial enforcing status.
1233 See security/selinux/Kconfig help text.
1234 0 -- permissive (log only, no denials).
1235 1 -- enforcing (deny and log).
1237 Value can be changed at runtime via /selinux/enforce.
1240 Disable Error Record Serialization Table (ERST)
1243 ether= [HW,NET] Ethernet cards parameters
1244 This option is obsoleted by the "netdev=" option, which
1245 has equivalent usage. See its documentation for details.
1249 Permit 'security.evm' to be updated regardless of
1250 current integrity status.
1254 fail_make_request=[KNL]
1255 General fault injection mechanism.
1256 Format: <interval>,<probability>,<space>,<times>
1257 See also Documentation/fault-injection/.
1260 See Documentation/admin-guide/blockdev/floppy.rst.
1262 force_pal_cache_flush
1263 [IA-64] Avoid check_sal_cache_flush which may hang on
1264 buggy SAL_CACHE_FLUSH implementations. Using this
1265 parameter will force ia64_sal_cache_flush to call
1266 ia64_pal_cache_flush instead of SAL_CACHE_FLUSH.
1269 Forcefully enable Physical Address Extension (PAE).
1270 Many Pentium M systems disable PAE but may have a
1271 functionally usable PAE implementation.
1272 Warning: use of this parameter will taint the kernel
1273 and may cause unknown problems.
1276 [FTRACE] will set and start the specified tracer
1277 as early as possible in order to facilitate early
1280 ftrace_dump_on_oops[=orig_cpu]
1281 [FTRACE] will dump the trace buffers on oops.
1282 If no parameter is passed, ftrace will dump
1283 buffers of all CPUs, but if you pass orig_cpu, it will
1284 dump only the buffer of the CPU that triggered the
1287 ftrace_filter=[function-list]
1288 [FTRACE] Limit the functions traced by the function
1289 tracer at boot up. function-list is a comma separated
1290 list of functions. This list can be changed at run
1291 time by the set_ftrace_filter file in the debugfs
1294 ftrace_notrace=[function-list]
1295 [FTRACE] Do not trace the functions specified in
1296 function-list. This list can be changed at run time
1297 by the set_ftrace_notrace file in the debugfs
1300 ftrace_graph_filter=[function-list]
1301 [FTRACE] Limit the top level callers functions traced
1302 by the function graph tracer at boot up.
1303 function-list is a comma separated list of functions
1304 that can be changed at run time by the
1305 set_graph_function file in the debugfs tracing directory.
1307 ftrace_graph_notrace=[function-list]
1308 [FTRACE] Do not trace from the functions specified in
1309 function-list. This list is a comma separated list of
1310 functions that can be changed at run time by the
1311 set_graph_notrace file in the debugfs tracing directory.
1313 ftrace_graph_max_depth=<uint>
1314 [FTRACE] Used with the function graph tracer. This is
1315 the max depth it will trace into a function. This value
1316 can be changed at run time by the max_graph_depth file
1317 in the tracefs tracing directory. default: 0 (no limit)
1320 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad
1321 support via parallel port (up to 5 devices per port)
1322 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5>
1323 See also Documentation/input/devices/joystick-parport.rst
1327 gart_fix_e820= [X86_64] disable the fix e820 for K8 GART
1331 gcov_persist= [GCOV] When non-zero (default), profiling data for
1332 kernel modules is saved and remains accessible via
1333 debugfs, even when the module is unloaded/reloaded.
1334 When zero, profiling data is discarded and associated
1335 debugfs files are removed at module unload time.
1337 goldfish [X86] Enable the goldfish android emulator platform.
1338 Don't use this when you are not running on the
1341 gpt [EFI] Forces disk with valid GPT signature but
1342 invalid Protective MBR to be treated as GPT. If the
1343 primary GPT is corrupted, it enables the backup/alternate
1344 GPT to be used instead.
1346 grcan.enable0= [HW] Configuration of physical interface 0. Determines
1347 the "Enable 0" bit of the configuration register.
1350 grcan.enable1= [HW] Configuration of physical interface 1. Determines
1351 the "Enable 0" bit of the configuration register.
1354 grcan.select= [HW] Select which physical interface to use.
1357 grcan.txsize= [HW] Sets the size of the tx buffer.
1358 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0.
1360 grcan.rxsize= [HW] Sets the size of the rx buffer.
1361 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0.
1364 gpio-mockup.gpio_mockup_ranges
1365 [HW] Sets the ranges of gpiochip of for this device.
1366 Format: <start1>,<end1>,<start2>,<end2>...
1368 hardlockup_all_cpu_backtrace=
1369 [KNL] Should the hard-lockup detector generate
1370 backtraces on all cpus.
1373 hashdist= [KNL,NUMA] Large hashes allocated during boot
1374 are distributed across NUMA nodes. Defaults on
1375 for 64-bit NUMA, off otherwise.
1376 Format: 0 | 1 (for off | on)
1378 hcl= [IA-64] SGI's Hardware Graph compatibility layer
1380 hd= [EIDE] (E)IDE hard drive subsystem geometry
1381 Format: <cyl>,<head>,<sect>
1384 Disable Hardware Error Source Table (HEST) support;
1385 corresponding firmware-first mode error processing
1386 logic will be disabled.
1388 highmem=nn[KMG] [KNL,BOOT] forces the highmem zone to have an exact
1389 size of <nn>. This works even on boxes that have no
1390 highmem otherwise. This also works to reduce highmem
1391 size on bigger boxes.
1393 highres= [KNL] Enable/disable high resolution timer mode.
1394 Valid parameters: "on", "off"
1399 hpet= [X86-32,HPET] option to control HPET usage
1400 Format: { enable (default) | disable | force |
1402 disable: disable HPET and use PIT instead
1403 force: allow force enabled of undocumented chips (ICH4,
1405 verbose: show contents of HPET registers during setup
1407 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET
1408 registers. Default set by CONFIG_HPET_MMAP_DEFAULT.
1410 hugepages= [HW,X86-32,IA-64] HugeTLB pages to allocate at boot.
1411 hugepagesz= [HW,IA-64,PPC,X86-64] The size of the HugeTLB pages.
1412 On x86-64 and powerpc, this option can be specified
1413 multiple times interleaved with hugepages= to reserve
1414 huge pages of different sizes. Valid pages sizes on
1415 x86-64 are 2M (when the CPU supports "pse") and 1G
1416 (when the CPU supports the "pdpe1gb" cpuinfo flag).
1419 [KNL] Should the hung task detector generate panics.
1422 A nonzero value instructs the kernel to panic when a
1423 hung task is detected. The default value is controlled
1424 by the CONFIG_BOOTPARAM_HUNG_TASK_PANIC build-time
1425 option. The value selected by this boot parameter can
1426 be changed later by the kernel.hung_task_panic sysctl.
1428 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC)
1429 terminal devices. Valid values: 0..8
1430 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs.
1431 If specified, z/VM IUCV HVC accepts connections
1432 from listed z/VM user IDs only.
1434 hv_nopvspin [X86,HYPER_V] Disables the paravirt spinlock optimizations
1435 which allow the hypervisor to 'idle' the
1436 guest on lock contention.
1439 Do not unregister boot console at start. This is only
1440 useful for debugging when something happens in the window
1441 between unregistering the boot console and initializing
1444 i2c_bus= [HW] Override the default board specific I2C bus speed
1445 or register an additional I2C bus that is not
1446 registered from board initialization code.
1450 i8042.debug [HW] Toggle i8042 debug mode
1451 i8042.unmask_kbd_data
1452 [HW] Enable printing of interrupt data from the KBD port
1453 (disabled by default, and as a pre-condition
1454 requires that i8042.debug=1 be enabled)
1455 i8042.direct [HW] Put keyboard port into non-translated mode
1456 i8042.dumbkbd [HW] Pretend that controller can only read data from
1457 keyboard and cannot control its state
1458 (Don't attempt to blink the leds)
1459 i8042.noaux [HW] Don't check for auxiliary (== mouse) port
1460 i8042.nokbd [HW] Don't check/create keyboard port
1461 i8042.noloop [HW] Disable the AUX Loopback command while probing
1463 i8042.nomux [HW] Don't check presence of an active multiplexing
1465 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX
1467 i8042.notimeout [HW] Ignore timeout condition signalled by controller
1468 i8042.reset [HW] Reset the controller during init, cleanup and
1469 suspend-to-ram transitions, only during s2r
1470 transitions, or never reset
1471 Format: { 1 | Y | y | 0 | N | n }
1472 1, Y, y: always reset controller
1473 0, N, n: don't ever reset controller
1474 Default: only on s2r transitions on x86; most other
1475 architectures force reset to be always executed
1476 i8042.unlock [HW] Unlock (ignore) the keylock
1477 i8042.kbdreset [HW] Reset device connected to KBD port
1481 i8k.ignore_dmi [HW] Continue probing hardware even if DMI data
1482 indicates that the driver is running on unsupported
1484 i8k.force [HW] Activate i8k driver even if SMM BIOS signature
1485 does not match list of supported models.
1487 [HW] Report power status in /proc/i8k
1488 (disabled by default)
1489 i8k.restricted [HW] Allow controlling fans only if SYS_ADMIN
1492 i915.invert_brightness=
1493 [DRM] Invert the sense of the variable that is used to
1494 set the brightness of the panel backlight. Normally a
1495 brightness value of 0 indicates backlight switched off,
1496 and the maximum of the brightness value sets the backlight
1497 to maximum brightness. If this parameter is set to 0
1498 (default) and the machine requires it, or this parameter
1499 is set to 1, a brightness value of 0 sets the backlight
1500 to maximum brightness, and the maximum of the brightness
1501 value switches the backlight off.
1502 -1 -- never invert brightness
1503 0 -- machine default
1504 1 -- force brightness inversion
1507 Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]]
1509 ide-core.nodma= [HW] (E)IDE subsystem
1510 Format: =0.0 to prevent dma on hda, =0.1 hdb =1.0 hdc
1511 .vlb_clock .pci_clock .noflush .nohpa .noprobe .nowerr
1512 .cdrom .chs .ignore_cable are additional options
1513 See Documentation/ide/ide.rst.
1515 ide-generic.probe-mask= [HW] (E)IDE subsystem
1517 Probe mask for legacy ISA IDE ports. Depending on
1518 platform up to 6 ports are supported, enabled by
1519 setting corresponding bits in the mask to 1. The
1520 default value is 0x0, which has a special meaning.
1521 On systems that have PCI, it triggers scanning the
1522 PCI bus for the first and the second port, which
1523 are then probed. On systems without PCI the value
1524 of 0x0 enables probing the two first ports as if it
1527 ide-pci-generic.all-generic-ide [HW] (E)IDE subsystem
1528 Claim all unknown PCI IDE storage controllers.
1531 Format: idle=poll, idle=halt, idle=nomwait
1532 Poll forces a polling idle loop that can slightly
1533 improve the performance of waking up a idle CPU, but
1534 will use a lot of power and make the system run hot.
1536 idle=halt: Halt is forced to be used for CPU idle.
1537 In such case C2/C3 won't be used again.
1538 idle=nomwait: Disable mwait for CPU C-states
1540 ieee754= [MIPS] Select IEEE Std 754 conformance mode
1541 Format: { strict | legacy | 2008 | relaxed }
1544 Choose which programs will be accepted for execution
1545 based on the IEEE 754 NaN encoding(s) supported by
1546 the FPU and the NaN encoding requested with the value
1547 of an ELF file header flag individually set by each
1548 binary. Hardware implementations are permitted to
1549 support either or both of the legacy and the 2008 NaN
1552 Available settings are as follows:
1553 strict accept binaries that request a NaN encoding
1554 supported by the FPU
1555 legacy only accept legacy-NaN binaries, if supported
1557 2008 only accept 2008-NaN binaries, if supported
1559 relaxed accept any binaries regardless of whether
1560 supported by the FPU
1562 The FPU emulator is always able to support both NaN
1563 encodings, so if no FPU hardware is present or it has
1564 been disabled with 'nofpu', then the settings of
1565 'legacy' and '2008' strap the emulator accordingly,
1566 'relaxed' straps the emulator for both legacy-NaN and
1567 2008-NaN, whereas 'strict' enables legacy-NaN only on
1568 legacy processors and both NaN encodings on MIPS32 or
1571 The setting for ABS.fmt/NEG.fmt instruction execution
1572 mode generally follows that for the NaN encoding,
1573 except where unsupported by hardware.
1575 ignore_loglevel [KNL]
1576 Ignore loglevel setting - this will print /all/
1577 kernel messages to the console. Useful for debugging.
1578 We also add it as printk module parameter, so users
1579 could change it dynamically, usually by
1580 /sys/module/printk/parameters/ignore_loglevel.
1583 Ignore RLIMIT_DATA setting for data mappings,
1584 print warning at first misuse. Can be changed via
1585 /sys/module/kernel/parameters/ignore_rlimit_data.
1587 ihash_entries= [KNL]
1588 Set number of hash buckets for inode cache.
1590 ima_appraise= [IMA] appraise integrity measurements
1591 Format: { "off" | "enforce" | "fix" | "log" }
1594 ima_appraise_tcb [IMA] Deprecated. Use ima_policy= instead.
1595 The builtin appraise policy appraises all files
1598 ima_canonical_fmt [IMA]
1599 Use the canonical format for the binary runtime
1600 measurements, instead of host native format.
1603 Format: { md5 | sha1 | rmd160 | sha256 | sha384
1607 The list of supported hash algorithms is defined
1608 in crypto/hash_info.h.
1611 The builtin policies to load during IMA setup.
1612 Format: "tcb | appraise_tcb | secure_boot |
1615 The "tcb" policy measures all programs exec'd, files
1616 mmap'd for exec, and all files opened with the read
1617 mode bit set by either the effective uid (euid=0) or
1620 The "appraise_tcb" policy appraises the integrity of
1621 all files owned by root.
1623 The "secure_boot" policy appraises the integrity
1624 of files (eg. kexec kernel image, kernel modules,
1625 firmware, policy, etc) based on file signatures.
1627 The "fail_securely" policy forces file signature
1628 verification failure also on privileged mounted
1629 filesystems with the SB_I_UNVERIFIABLE_SIGNATURE
1632 ima_tcb [IMA] Deprecated. Use ima_policy= instead.
1633 Load a policy which meets the needs of the Trusted
1634 Computing Base. This means IMA will measure all
1635 programs exec'd, files mmap'd for exec, and all files
1636 opened for read by uid=0.
1639 Select one of defined IMA measurements template formats.
1640 Formats: { "ima" | "ima-ng" | "ima-sig" }
1644 [IMA] Define a custom template format.
1645 Format: { "field1|...|fieldN" }
1647 ima.ahash_minsize= [IMA] Minimum file size for asynchronous hash usage
1648 Format: <min_file_size>
1649 Set the minimal file size for using asynchronous hash.
1650 If left unspecified, ahash usage is disabled.
1652 ahash performance varies for different data sizes on
1653 different crypto accelerators. This option can be used
1654 to achieve the best performance for a particular HW.
1656 ima.ahash_bufsize= [IMA] Asynchronous hash buffer size
1658 Set hashing buffer size. Default: 4k.
1660 ahash performance varies for different chunk sizes on
1661 different crypto accelerators. This option can be used
1662 to achieve best performance for particular HW.
1666 Run specified binary instead of /sbin/init as init
1669 initcall_debug [KNL] Trace initcalls as they are executed. Useful
1670 for working out where the kernel is dying during
1673 initcall_blacklist= [KNL] Do not execute a comma-separated list of
1674 initcall functions. Useful for debugging built-in
1675 modules and initcalls.
1677 initrd= [BOOT] Specify the location of the initial ramdisk
1679 init_on_alloc= [MM] Fill newly allocated pages and heap objects with
1682 Default set by CONFIG_INIT_ON_ALLOC_DEFAULT_ON.
1684 init_on_free= [MM] Fill freed pages and heap objects with zeroes.
1686 Default set by CONFIG_INIT_ON_FREE_DEFAULT_ON.
1688 init_pkru= [x86] Specify the default memory protection keys rights
1689 register contents for all processes. 0x55555554 by
1690 default (disallow access to all but pkey 0). Can
1691 override in debugfs after boot.
1693 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver
1696 int_pln_enable [x86] Enable power limit notification interrupt
1698 integrity_audit=[IMA]
1699 Format: { "0" | "1" }
1700 0 -- basic integrity auditing messages. (Default)
1701 1 -- additional integrity auditing messages.
1703 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option
1705 Enable intel iommu driver.
1707 Disable intel iommu driver.
1708 igfx_off [Default Off]
1709 By default, gfx is mapped as normal device. If a gfx
1710 device has a dedicated DMAR unit, the DMAR unit is
1711 bypassed by not enabling DMAR with this option. In
1712 this case, gfx device will use physical address for
1715 With this option iommu will not optimize to look
1716 for io virtual address below 32-bit forcing dual
1717 address cycle on pci bus for cards supporting greater
1718 than 32-bit addressing. The default is to look
1719 for translation below 32-bit and if not available
1720 then look in the higher range.
1721 strict [Default Off]
1722 With this option on every unmap_single operation will
1723 result in a hardware IOTLB flush operation as opposed
1724 to batching them for performance.
1725 sp_off [Default Off]
1726 By default, super page will be supported if Intel IOMMU
1727 has the capability. With this option, super page will
1730 By default, scalable mode will be disabled even if the
1731 hardware advertises that it has support for the scalable
1732 mode translation. With this option set, scalable mode
1733 will be used on hardware which claims to support it.
1734 tboot_noforce [Default Off]
1735 Do not force the Intel IOMMU enabled under tboot.
1736 By default, tboot will force Intel IOMMU on, which
1737 could harm performance of some high-throughput
1738 devices like 40GBit network cards, even if identity
1740 Note that using this option lowers the security
1741 provided by tboot because it makes the system
1742 vulnerable to DMA attacks.
1743 nobounce [Default off]
1744 Disable bounce buffer for unstrusted devices such as
1745 the Thunderbolt devices. This will treat the untrusted
1746 devices as the trusted ones, hence might expose security
1747 risks of DMA attacks.
1749 intel_idle.max_cstate= [KNL,HW,ACPI,X86]
1750 0 disables intel_idle and fall back on acpi_idle.
1751 1 to 9 specify maximum depth of C-state.
1755 Do not enable intel_pstate as the default
1756 scaling driver for the supported processors
1758 Use intel_pstate as a scaling driver, but configure it
1759 to work with generic cpufreq governors (instead of
1760 enabling its internal governor). This mode cannot be
1761 used along with the hardware-managed P-states (HWP)
1764 Enable intel_pstate on systems that prohibit it by default
1765 in favor of acpi-cpufreq. Forcing the intel_pstate driver
1766 instead of acpi-cpufreq may disable platform features, such
1767 as thermal controls and power capping, that rely on ACPI
1768 P-States information being indicated to OSPM and therefore
1769 should be used with caution. This option does not work with
1770 processors that aren't supported by the intel_pstate driver
1771 or on platforms that use pcc-cpufreq instead of acpi-cpufreq.
1773 Do not enable hardware P state control (HWP)
1776 Only load intel_pstate on systems which support
1777 hardware P state control (HWP) if available.
1779 Enforce ACPI _PPC performance limits. If the Fixed ACPI
1780 Description Table, specifies preferred power management
1781 profile as "Enterprise Server" or "Performance Server",
1782 then this feature is turned on by default.
1784 Allow per-logical-CPU P-State performance control limits using
1785 cpufreq sysfs interface
1787 intremap= [X86-64, Intel-IOMMU]
1788 on enable Interrupt Remapping (default)
1789 off disable Interrupt Remapping
1790 nosid disable Source ID checking
1792 BIOS x2APIC opt-out request will be ignored
1793 nopost disable Interrupt Posting
1795 iomem= Disable strict checking of access to MMIO memory
1796 strict regions from userspace.
1811 nobypass [PPC/POWERNV]
1812 Disable IOMMU bypass, using IOMMU for PCI devices.
1814 iommu.strict= [ARM64] Configure TLB invalidation behaviour
1815 Format: { "0" | "1" }
1817 Request that DMA unmap operations use deferred
1818 invalidation of hardware TLBs, for increased
1819 throughput at the cost of reduced device isolation.
1820 Will fall back to strict mode if not supported by
1821 the relevant IOMMU driver.
1822 1 - Strict mode (default).
1823 DMA unmap operations invalidate IOMMU hardware TLBs
1827 [ARM64, X86] Configure DMA to bypass the IOMMU by default.
1828 Format: { "0" | "1" }
1829 0 - Use IOMMU translation for DMA.
1830 1 - Bypass the IOMMU for DMA.
1831 unset - Use value of CONFIG_IOMMU_DEFAULT_PASSTHROUGH.
1833 io7= [HW] IO7 for Marvel based alpha systems
1834 See comment before marvel_specify_io7 in
1835 arch/alpha/kernel/core_marvel.c.
1837 io_delay= [X86] I/O delay method
1839 Standard port 0x80 based delay
1841 Alternate port 0xed based delay (needed on some systems)
1843 Simple two microseconds delay
1848 See Documentation/filesystems/nfs/nfsroot.txt.
1850 ipcmni_extend [KNL] Extend the maximum number of unique System V
1851 IPC identifiers from 32,768 to 16,777,216.
1853 irqaffinity= [SMP] Set the default irq affinity mask
1854 The argument is a cpu list, as described above.
1856 irqchip.gicv2_force_probe=
1859 Force the kernel to look for the second 4kB page
1860 of a GICv2 controller even if the memory range
1861 exposed by the device tree is too small.
1863 irqchip.gicv3_nolpi=
1865 Force the kernel to ignore the availability of
1866 LPIs (and by consequence ITSs). Intended for system
1867 that use the kernel as a bootloader, and thus want
1868 to let secondary kernels in charge of setting up
1871 irqchip.gicv3_pseudo_nmi= [ARM64]
1872 Enables support for pseudo-NMIs in the kernel. This
1873 requires the kernel to be built with
1874 CONFIG_ARM64_PSEUDO_NMI.
1877 When an interrupt is not handled search all handlers
1878 for it. Intended to get systems with badly broken
1882 When an interrupt is not handled search all handlers
1883 for it. Also check all handlers each timer
1884 interrupt. Intended to get systems with badly broken
1888 Format: <RDP>,<reset>,<pci_scan>,<verbosity>
1890 isolcpus= [KNL,SMP,ISOL] Isolate a given set of CPUs from disturbance.
1891 [Deprecated - use cpusets instead]
1892 Format: [flag-list,]<cpu-list>
1894 Specify one or more CPUs to isolate from disturbances
1895 specified in the flag list (default: domain):
1898 Disable the tick when a single task runs.
1900 A residual 1Hz tick is offloaded to workqueues, which you
1901 need to affine to housekeeping through the global
1902 workqueue's affinity configured via the
1903 /sys/devices/virtual/workqueue/cpumask sysfs file, or
1904 by using the 'domain' flag described below.
1906 NOTE: by default the global workqueue runs on all CPUs,
1907 so to protect individual CPUs the 'cpumask' file has to
1908 be configured manually after bootup.
1911 Isolate from the general SMP balancing and scheduling
1912 algorithms. Note that performing domain isolation this way
1913 is irreversible: it's not possible to bring back a CPU to
1914 the domains once isolated through isolcpus. It's strongly
1915 advised to use cpusets instead to disable scheduler load
1916 balancing through the "cpuset.sched_load_balance" file.
1917 It offers a much more flexible interface where CPUs can
1918 move in and out of an isolated set anytime.
1920 You can move a process onto or off an "isolated" CPU via
1921 the CPU affinity syscalls or cpuset.
1922 <cpu number> begins at 0 and the maximum value is
1923 "number of CPUs in system - 1".
1925 The format of <cpu-list> is described above.
1931 ivrs_ioapic [HW,X86_64]
1932 Provide an override to the IOAPIC-ID<->DEVICE-ID
1933 mapping provided in the IVRS ACPI table. For
1934 example, to map IOAPIC-ID decimal 10 to
1935 PCI device 00:14.0 write the parameter as:
1936 ivrs_ioapic[10]=00:14.0
1938 ivrs_hpet [HW,X86_64]
1939 Provide an override to the HPET-ID<->DEVICE-ID
1940 mapping provided in the IVRS ACPI table. For
1941 example, to map HPET-ID decimal 0 to
1942 PCI device 00:14.0 write the parameter as:
1943 ivrs_hpet[0]=00:14.0
1945 ivrs_acpihid [HW,X86_64]
1946 Provide an override to the ACPI-HID:UID<->DEVICE-ID
1947 mapping provided in the IVRS ACPI table. For
1948 example, to map UART-HID:UID AMD0020:0 to
1949 PCI device 00:14.5 write the parameter as:
1950 ivrs_acpihid[00:14.5]=AMD0020:0
1952 js= [HW,JOY] Analog joystick
1953 See Documentation/input/joydev/joystick.rst.
1956 When CONFIG_RANDOMIZE_BASE is set, this disables
1957 kernel and module base offset ASLR (Address Space
1958 Layout Randomization).
1961 [KNL] Enforce KASAN (Kernel Address Sanitizer) to print
1962 report on every invalid memory access. Without this
1963 parameter KASAN will print report only for the first
1968 kernelcore= [KNL,X86,IA-64,PPC]
1969 Format: nn[KMGTPE] | nn% | "mirror"
1970 This parameter specifies the amount of memory usable by
1971 the kernel for non-movable allocations. The requested
1972 amount is spread evenly throughout all nodes in the
1973 system as ZONE_NORMAL. The remaining memory is used for
1974 movable memory in its own zone, ZONE_MOVABLE. In the
1975 event, a node is too small to have both ZONE_NORMAL and
1976 ZONE_MOVABLE, kernelcore memory will take priority and
1977 other nodes will have a larger ZONE_MOVABLE.
1979 ZONE_MOVABLE is used for the allocation of pages that
1980 may be reclaimed or moved by the page migration
1981 subsystem. Note that allocations like PTEs-from-HighMem
1982 still use the HighMem zone if it exists, and the Normal
1983 zone if it does not.
1985 It is possible to specify the exact amount of memory in
1986 the form of "nn[KMGTPE]", a percentage of total system
1987 memory in the form of "nn%", or "mirror". If "mirror"
1988 option is specified, mirrored (reliable) memory is used
1989 for non-movable allocations and remaining memory is used
1990 for Movable pages. "nn[KMGTPE]", "nn%", and "mirror"
1991 are exclusive, so you cannot specify multiple forms.
1993 kgdbdbgp= [KGDB,HW] kgdb over EHCI usb debug port.
1994 Format: <Controller#>[,poll interval]
1995 The controller # is the number of the ehci usb debug
1996 port as it is probed via PCI. The poll interval is
1997 optional and is the number seconds in between
1998 each poll cycle to the debug port in case you need
1999 the functionality for interrupting the kernel with
2000 gdb or control-c on the dbgp connection. When
2001 not using this parameter you use sysrq-g to break into
2002 the kernel debugger.
2004 kgdboc= [KGDB,HW] kgdb over consoles.
2005 Requires a tty driver that supports console polling,
2006 or a supported polling keyboard driver (non-usb).
2007 Serial only format: <serial_device>[,baud]
2008 keyboard only format: kbd
2009 keyboard and serial format: kbd,<serial_device>[,baud]
2010 Optional Kernel mode setting:
2011 kms, kbd format: kms,kbd
2012 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud]
2014 kgdbwait [KGDB] Stop kernel execution and enter the
2015 kernel debugger at the earliest opportunity.
2017 kmac= [MIPS] korina ethernet MAC address.
2018 Configure the RouterBoard 532 series on-chip
2019 Ethernet adapter MAC address.
2021 kmemleak= [KNL] Boot-time kmemleak enable/disable
2022 Valid arguments: on, off
2024 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y,
2027 kprobe_event=[probe-list]
2028 [FTRACE] Add kprobe events and enable at boot time.
2029 The probe-list is a semicolon delimited list of probe
2030 definitions. Each definition is same as kprobe_events
2031 interface, but the parameters are comma delimited.
2032 For example, to add a kprobe event on vfs_read with
2033 arg1 and arg2, add to the command line;
2035 kprobe_event=p,vfs_read,$arg1,$arg2
2037 See also Documentation/trace/kprobetrace.rst "Kernel
2038 Boot Parameter" section.
2040 kpti= [ARM64] Control page table isolation of user
2041 and kernel address spaces.
2042 Default: enabled on cores which need mitigation.
2046 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs.
2047 Default is 0 (don't ignore, but inject #GP)
2049 kvm.enable_vmware_backdoor=[KVM] Support VMware backdoor PV interface.
2050 Default is false (don't support).
2052 kvm.mmu_audit= [KVM] This is a R/W parameter which allows audit
2056 kvm-amd.nested= [KVM,AMD] Allow nested virtualization in KVM/SVM.
2057 Default is 1 (enabled)
2059 kvm-amd.npt= [KVM,AMD] Disable nested paging (virtualized MMU)
2061 Default is 1 (enabled) if in 64-bit or 32-bit PAE mode.
2063 kvm-arm.vgic_v3_group0_trap=
2064 [KVM,ARM] Trap guest accesses to GICv3 group-0
2067 kvm-arm.vgic_v3_group1_trap=
2068 [KVM,ARM] Trap guest accesses to GICv3 group-1
2071 kvm-arm.vgic_v3_common_trap=
2072 [KVM,ARM] Trap guest accesses to GICv3 common
2075 kvm-arm.vgic_v4_enable=
2076 [KVM,ARM] Allow use of GICv4 for direct injection of
2079 kvm-intel.ept= [KVM,Intel] Disable extended page tables
2080 (virtualized MMU) support on capable Intel chips.
2081 Default is 1 (enabled)
2083 kvm-intel.emulate_invalid_guest_state=
2084 [KVM,Intel] Enable emulation of invalid guest states
2085 Default is 0 (disabled)
2087 kvm-intel.flexpriority=
2088 [KVM,Intel] Disable FlexPriority feature (TPR shadow).
2089 Default is 1 (enabled)
2092 [KVM,Intel] Enable VMX nesting (nVMX).
2093 Default is 0 (disabled)
2095 kvm-intel.unrestricted_guest=
2096 [KVM,Intel] Disable unrestricted guest feature
2097 (virtualized real and unpaged mode) on capable
2098 Intel chips. Default is 1 (enabled)
2100 kvm-intel.vmentry_l1d_flush=[KVM,Intel] Mitigation for L1 Terminal Fault
2103 Valid arguments: never, cond, always
2105 always: L1D cache flush on every VMENTER.
2106 cond: Flush L1D on VMENTER only when the code between
2107 VMEXIT and VMENTER can leak host memory.
2108 never: Disables the mitigation
2110 Default is cond (do L1 cache flush in specific instances)
2112 kvm-intel.vpid= [KVM,Intel] Disable Virtual Processor Identification
2113 feature (tagged TLBs) on capable Intel chips.
2114 Default is 1 (enabled)
2116 l1tf= [X86] Control mitigation of the L1TF vulnerability on
2119 The kernel PTE inversion protection is unconditionally
2120 enabled and cannot be disabled.
2123 Provides all available mitigations for the
2124 L1TF vulnerability. Disables SMT and
2125 enables all mitigations in the
2126 hypervisors, i.e. unconditional L1D flush.
2128 SMT control and L1D flush control via the
2129 sysfs interface is still possible after
2130 boot. Hypervisors will issue a warning
2131 when the first VM is started in a
2132 potentially insecure configuration,
2133 i.e. SMT enabled or L1D flush disabled.
2136 Same as 'full', but disables SMT and L1D
2137 flush runtime control. Implies the
2138 'nosmt=force' command line option.
2139 (i.e. sysfs control of SMT is disabled.)
2142 Leaves SMT enabled and enables the default
2143 hypervisor mitigation, i.e. conditional
2146 SMT control and L1D flush control via the
2147 sysfs interface is still possible after
2148 boot. Hypervisors will issue a warning
2149 when the first VM is started in a
2150 potentially insecure configuration,
2151 i.e. SMT enabled or L1D flush disabled.
2155 Disables SMT and enables the default
2156 hypervisor mitigation.
2158 SMT control and L1D flush control via the
2159 sysfs interface is still possible after
2160 boot. Hypervisors will issue a warning
2161 when the first VM is started in a
2162 potentially insecure configuration,
2163 i.e. SMT enabled or L1D flush disabled.
2166 Same as 'flush', but hypervisors will not
2167 warn when a VM is started in a potentially
2168 insecure configuration.
2171 Disables hypervisor mitigations and doesn't
2173 It also drops the swap size and available
2174 RAM limit restriction on both hypervisor and
2179 For details see: Documentation/admin-guide/hw-vuln/l1tf.rst
2185 lapic [X86-32,APIC] Enable the local APIC even if BIOS
2188 lapic= [x86,APIC] "notscdeadline" Do not use TSC deadline
2189 value for LAPIC timer one-shot implementation. Default
2190 back to the programmable timer unit in the LAPIC.
2192 lapic_timer_c2_ok [X86,APIC] trust the local apic timer
2195 libata.dma= [LIBATA] DMA control
2196 libata.dma=0 Disable all PATA and SATA DMA
2197 libata.dma=1 PATA and SATA Disk DMA only
2198 libata.dma=2 ATAPI (CDROM) DMA only
2199 libata.dma=4 Compact Flash DMA only
2200 Combinations also work, so libata.dma=3 enables DMA
2201 for disks and CDROMs, but not CFs.
2203 libata.ignore_hpa= [LIBATA] Ignore HPA limit
2204 libata.ignore_hpa=0 keep BIOS limits (default)
2205 libata.ignore_hpa=1 ignore limits, using full disk
2207 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume
2211 libata.force= [LIBATA] Force configurations. The format is comma
2212 separated list of "[ID:]VAL" where ID is
2213 PORT[.DEVICE]. PORT and DEVICE are decimal numbers
2214 matching port, link or device. Basically, it matches
2215 the ATA ID string printed on console by libata. If
2216 the whole ID part is omitted, the last PORT and DEVICE
2217 values are used. If ID hasn't been specified yet, the
2218 configuration applies to all ports, links and devices.
2220 If only DEVICE is omitted, the parameter applies to
2221 the port and all links and devices behind it. DEVICE
2222 number of 0 either selects the first device or the
2223 first fan-out link behind PMP device. It does not
2224 select the host link. DEVICE number of 15 selects the
2225 host link and device attached to it.
2227 The VAL specifies the configuration to force. As long
2228 as there's no ambiguity shortcut notation is allowed.
2229 For example, both 1.5 and 1.5G would work for 1.5Gbps.
2230 The following configurations can be forced.
2232 * Cable type: 40c, 80c, short40c, unk, ign or sata.
2233 Any ID with matching PORT is used.
2235 * SATA link speed limit: 1.5Gbps or 3.0Gbps.
2237 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7].
2238 udma[/][16,25,33,44,66,100,133] notation is also
2241 * [no]ncq: Turn on or off NCQ.
2243 * [no]ncqtrim: Turn off queued DSM TRIM.
2245 * nohrst, nosrst, norst: suppress hard, soft
2248 * rstonce: only attempt one reset during
2249 hot-unplug link recovery
2251 * dump_id: dump IDENTIFY data.
2253 * atapi_dmadir: Enable ATAPI DMADIR bridge support
2255 * disable: Disable this device.
2257 If there are multiple matching configurations changing
2258 the same attribute, the last one is used.
2260 memblock=debug [KNL] Enable memblock debug messages.
2262 load_ramdisk= [RAM] List of ramdisks to load from floppy
2263 See Documentation/admin-guide/blockdev/ramdisk.rst.
2265 lockd.nlm_grace_period=P [NFS] Assign grace period.
2268 lockd.nlm_tcpport=N [NFS] Assign TCP port.
2271 lockd.nlm_timeout=T [NFS] Assign timeout value.
2274 lockd.nlm_udpport=M [NFS] Assign UDP port.
2277 locktorture.nreaders_stress= [KNL]
2278 Set the number of locking read-acquisition kthreads.
2279 Defaults to being automatically set based on the
2280 number of online CPUs.
2282 locktorture.nwriters_stress= [KNL]
2283 Set the number of locking write-acquisition kthreads.
2285 locktorture.onoff_holdoff= [KNL]
2286 Set time (s) after boot for CPU-hotplug testing.
2288 locktorture.onoff_interval= [KNL]
2289 Set time (s) between CPU-hotplug operations, or
2290 zero to disable CPU-hotplug testing.
2292 locktorture.shuffle_interval= [KNL]
2293 Set task-shuffle interval (jiffies). Shuffling
2294 tasks allows some CPUs to go into dyntick-idle
2295 mode during the locktorture test.
2297 locktorture.shutdown_secs= [KNL]
2298 Set time (s) after boot system shutdown. This
2299 is useful for hands-off automated testing.
2301 locktorture.stat_interval= [KNL]
2302 Time (s) between statistics printk()s.
2304 locktorture.stutter= [KNL]
2305 Time (s) to stutter testing, for example,
2306 specifying five seconds causes the test to run for
2307 five seconds, wait for five seconds, and so on.
2308 This tests the locking primitive's ability to
2309 transition abruptly to and from idle.
2311 locktorture.torture_type= [KNL]
2312 Specify the locking implementation to test.
2314 locktorture.verbose= [KNL]
2315 Enable additional printk() statements.
2317 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver
2320 loglevel= All Kernel Messages with a loglevel smaller than the
2321 console loglevel will be printed to the console. It can
2322 also be changed with klogd or other programs. The
2323 loglevels are defined as follows:
2325 0 (KERN_EMERG) system is unusable
2326 1 (KERN_ALERT) action must be taken immediately
2327 2 (KERN_CRIT) critical conditions
2328 3 (KERN_ERR) error conditions
2329 4 (KERN_WARNING) warning conditions
2330 5 (KERN_NOTICE) normal but significant condition
2331 6 (KERN_INFO) informational
2332 7 (KERN_DEBUG) debug-level messages
2334 log_buf_len=n[KMG] Sets the size of the printk ring buffer,
2335 in bytes. n must be a power of two and greater
2336 than the minimal size. The minimal size is defined
2337 by LOG_BUF_SHIFT kernel config parameter. There is
2338 also CONFIG_LOG_CPU_MAX_BUF_SHIFT config parameter
2339 that allows to increase the default size depending on
2340 the number of CPUs. See init/Kconfig for more details.
2342 logo.nologo [FB] Disables display of the built-in Linux logo.
2343 This may be used to provide more screen space for
2344 kernel log messages and is useful when debugging
2345 kernel boot problems.
2347 lp=0 [LP] Specify parallel ports to use, e.g,
2348 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses
2349 lp=reset first parallel port). 'lp=0' disables the
2350 lp=auto printer driver. 'lp=reset' (which can be
2351 specified in addition to the ports) causes
2352 attached printers to be reset. Using
2353 lp=port1,port2,... specifies the parallel ports
2354 to associate lp devices with, starting with
2355 lp0. A port specification may be 'none' to skip
2356 that lp device, or a parport name such as
2357 'parport0'. Specifying 'lp=auto' instead of a
2358 port specification list means that device IDs
2359 from each port should be examined, to see if
2360 an IEEE 1284-compliant printer is attached; if
2361 so, the driver will manage that printer.
2362 See also header of drivers/char/lp.c.
2365 Sets loops_per_jiffy to given constant, thus avoiding
2366 time-consuming boot-time autodetection (up to 250 ms per
2367 CPU). 0 enables autodetection (default). To determine
2368 the correct value for your kernel, boot with normal
2369 autodetection and see what value is printed. Note that
2370 on SMP systems the preset will be applied to all CPUs,
2371 which is likely to cause problems if your CPUs need
2372 significantly divergent settings. An incorrect value
2373 will cause delays in the kernel to be wrong, leading to
2374 unpredictable I/O errors and other breakage. Although
2375 unlikely, in the extreme case this might damage your
2379 Format: <io>,<irq>,<dma>
2381 lsm.debug [SECURITY] Enable LSM initialization debugging output.
2384 [SECURITY] Choose order of LSM initialization. This
2385 overrides CONFIG_LSM, and the "security=" parameter.
2387 machvec= [IA-64] Force the use of a particular machine-vector
2388 (machvec) in a generic kernel.
2389 Example: machvec=hpzx1
2391 machtype= [Loongson] Share the same kernel image file between different
2393 Example: machtype=lemote-yeeloong-2f-7inch
2395 max_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory greater
2396 than or equal to this physical address is ignored.
2398 maxcpus= [SMP] Maximum number of processors that an SMP kernel
2399 will bring up during bootup. maxcpus=n : n >= 0 limits
2400 the kernel to bring up 'n' processors. Surely after
2401 bootup you can bring up the other plugged cpu by executing
2402 "echo 1 > /sys/devices/system/cpu/cpuX/online". So maxcpus
2403 only takes effect during system bootup.
2404 While n=0 is a special case, it is equivalent to "nosmp",
2405 which also disables the IO APIC.
2407 max_loop= [LOOP] The number of loop block devices that get
2408 (loop.max_loop) unconditionally pre-created at init time. The default
2409 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead
2410 of statically allocating a predefined number, loop
2411 devices can be requested on-demand with the
2412 /dev/loop-control interface.
2414 mce [X86-32] Machine Check Exception
2416 mce=option [X86-64] See Documentation/x86/x86_64/boot-options.rst
2418 md= [HW] RAID subsystems devices and level
2419 See Documentation/admin-guide/md.rst.
2422 Format: <first>,<last>
2423 Specifies range of consoles to be captured by the MDA.
2426 Control mitigation for the Micro-architectural Data
2427 Sampling (MDS) vulnerability.
2429 Certain CPUs are vulnerable to an exploit against CPU
2430 internal buffers which can forward information to a
2431 disclosure gadget under certain conditions.
2433 In vulnerable processors, the speculatively
2434 forwarded data can be used in a cache side channel
2435 attack, to access data to which the attacker does
2436 not have direct access.
2438 This parameter controls the MDS mitigation. The
2441 full - Enable MDS mitigation on vulnerable CPUs
2442 full,nosmt - Enable MDS mitigation and disable
2443 SMT on vulnerable CPUs
2444 off - Unconditionally disable MDS mitigation
2446 Not specifying this option is equivalent to
2449 For details see: Documentation/admin-guide/hw-vuln/mds.rst
2451 mem=nn[KMG] [KNL,BOOT] Force usage of a specific amount of memory
2452 Amount of memory to be used when the kernel is not able
2453 to see the whole system memory or for test.
2454 [X86] Work as limiting max address. Use together
2455 with memmap= to avoid physical address space collisions.
2456 Without memmap= PCI devices could be placed at addresses
2457 belonging to unused RAM.
2459 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel
2463 [KNL,SH] Allow user to override the default size for
2464 per-device physically contiguous DMA buffers.
2466 memhp_default_state=online/offline
2467 [KNL] Set the initial state for the memory hotplug
2468 onlining policy. If not specified, the default value is
2469 set according to the
2470 CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE kernel config
2472 See Documentation/admin-guide/mm/memory-hotplug.rst.
2474 memmap=exactmap [KNL,X86] Enable setting of an exact
2475 E820 memory map, as specified by the user.
2476 Such memmap=exactmap lines can be constructed based on
2477 BIOS output or other requirements. See the memmap=nn@ss
2480 memmap=nn[KMG]@ss[KMG]
2481 [KNL] Force usage of a specific region of memory.
2482 Region of memory to be used is from ss to ss+nn.
2483 If @ss[KMG] is omitted, it is equivalent to mem=nn[KMG],
2484 which limits max address to nn[KMG].
2485 Multiple different regions can be specified,
2488 memmap=100M@2G,100M#3G,1G!1024G
2490 memmap=nn[KMG]#ss[KMG]
2491 [KNL,ACPI] Mark specific memory as ACPI data.
2492 Region of memory to be marked is from ss to ss+nn.
2494 memmap=nn[KMG]$ss[KMG]
2495 [KNL,ACPI] Mark specific memory as reserved.
2496 Region of memory to be reserved is from ss to ss+nn.
2497 Example: Exclude memory from 0x18690000-0x1869ffff
2498 memmap=64K$0x18690000
2500 memmap=0x10000$0x18690000
2501 Some bootloaders may need an escape character before '$',
2502 like Grub2, otherwise '$' and the following number
2505 memmap=nn[KMG]!ss[KMG]
2506 [KNL,X86] Mark specific memory as protected.
2507 Region of memory to be used, from ss to ss+nn.
2508 The memory region may be marked as e820 type 12 (0xc)
2509 and is NVDIMM or ADR memory.
2511 memmap=<size>%<offset>-<oldtype>+<newtype>
2512 [KNL,ACPI] Convert memory within the specified region
2513 from <oldtype> to <newtype>. If "-<oldtype>" is left
2514 out, the whole region will be marked as <newtype>,
2515 even if previously unavailable. If "+<newtype>" is left
2516 out, matching memory will be removed. Types are
2517 specified as e820 types, e.g., 1 = RAM, 2 = reserved,
2518 3 = ACPI, 12 = PRAM.
2520 memory_corruption_check=0/1 [X86]
2521 Some BIOSes seem to corrupt the first 64k of
2522 memory when doing things like suspend/resume.
2523 Setting this option will scan the memory
2524 looking for corruption. Enabling this will
2525 both detect corruption and prevent the kernel
2526 from using the memory being corrupted.
2527 However, its intended as a diagnostic tool; if
2528 repeatable BIOS-originated corruption always
2529 affects the same memory, you can use memmap=
2530 to prevent the kernel from using that memory.
2532 memory_corruption_check_size=size [X86]
2533 By default it checks for corruption in the low
2534 64k, making this memory unavailable for normal
2535 use. Use this parameter to scan for
2536 corruption in more or less memory.
2538 memory_corruption_check_period=seconds [X86]
2539 By default it checks for corruption every 60
2540 seconds. Use this parameter to check at some
2541 other rate. 0 disables periodic checking.
2543 memtest= [KNL,X86,ARM,PPC] Enable memtest
2545 default : 0 <disable>
2546 Specifies the number of memtest passes to be
2547 performed. Each pass selects another test
2548 pattern from a given set of patterns. Memtest
2549 fills the memory with this pattern, validates
2550 memory contents and reserves bad memory
2551 regions that are detected.
2553 mem_encrypt= [X86-64] AMD Secure Memory Encryption (SME) control
2554 Valid arguments: on, off
2555 Default (depends on kernel configuration option):
2556 on (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=y)
2557 off (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=n)
2558 mem_encrypt=on: Activate SME
2559 mem_encrypt=off: Do not activate SME
2561 Refer to Documentation/virt/kvm/amd-memory-encryption.rst
2562 for details on when memory encryption can be activated.
2564 mem_sleep_default= [SUSPEND] Default system suspend mode:
2565 s2idle - Suspend-To-Idle
2566 shallow - Power-On Suspend or equivalent (if supported)
2567 deep - Suspend-To-RAM or equivalent (if supported)
2568 See Documentation/admin-guide/pm/sleep-states.rst.
2570 meye.*= [HW] Set MotionEye Camera parameters
2571 See Documentation/media/v4l-drivers/meye.rst.
2573 mfgpt_irq= [IA-32] Specify the IRQ to use for the
2574 Multi-Function General Purpose Timers on AMD Geode
2577 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when
2578 the BIOS has incorrectly applied a workaround. TinyBIOS
2579 version 0.98 is known to be affected, 0.99 fixes the
2580 problem by letting the user disable the workaround.
2584 min_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory below this
2585 physical address is ignored.
2587 mini2440= [ARM,HW,KNL]
2588 Format:[0..2][b][c][t]
2590 MINI2440 configuration specification:
2591 0 - The attached screen is the 3.5" TFT
2592 1 - The attached screen is the 7" TFT
2593 2 - The VGA Shield is attached (1024x768)
2594 Leaving out the screen size parameter will not load
2595 the TFT driver, and the framebuffer will be left
2597 b - Enable backlight. The TFT backlight pin will be
2598 linked to the kernel VESA blanking code and a GPIO
2599 LED. This parameter is not necessary when using the
2601 c - Enable the s3c camera interface.
2602 t - Reserved for enabling touchscreen support. The
2603 touchscreen support is not enabled in the mainstream
2604 kernel as of 2.6.30, a preliminary port can be found
2605 in the "bleeding edge" mini2440 support kernel at
2606 http://repo.or.cz/w/linux-2.6/mini2440.git
2609 [X86,PPC,S390,ARM64] Control optional mitigations for
2610 CPU vulnerabilities. This is a set of curated,
2611 arch-independent options, each of which is an
2612 aggregation of existing arch-specific options.
2615 Disable all optional CPU mitigations. This
2616 improves system performance, but it may also
2617 expose users to several CPU vulnerabilities.
2618 Equivalent to: nopti [X86,PPC]
2620 nospectre_v1 [X86,PPC]
2622 nospectre_v2 [X86,PPC,S390,ARM64]
2623 spectre_v2_user=off [X86]
2624 spec_store_bypass_disable=off [X86,PPC]
2625 ssbd=force-off [ARM64]
2630 Mitigate all CPU vulnerabilities, but leave SMT
2631 enabled, even if it's vulnerable. This is for
2632 users who don't want to be surprised by SMT
2633 getting disabled across kernel upgrades, or who
2634 have other ways of avoiding SMT-based attacks.
2635 Equivalent to: (default behavior)
2638 Mitigate all CPU vulnerabilities, disabling SMT
2639 if needed. This is for users who always want to
2640 be fully mitigated, even if it means losing SMT.
2641 Equivalent to: l1tf=flush,nosmt [X86]
2642 mds=full,nosmt [X86]
2645 [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this
2646 parameter allows control of the logging verbosity for
2647 the additional memory initialisation checks. A value
2648 of 0 disables mminit logging and a level of 4 will
2649 log everything. Information is printed at KERN_DEBUG
2650 so loglevel=8 may also need to be specified.
2653 [KNL] When CONFIG_MODULE_SIG is set, this means that
2654 modules without (valid) signatures will fail to load.
2655 Note that if CONFIG_MODULE_SIG_FORCE is set, that
2656 is always true, so this option does nothing.
2658 module_blacklist= [KNL] Do not load a comma-separated list of
2659 modules. Useful for debugging problem modules.
2662 [MOUSE] Maximum time between finger touching and
2663 leaving touchpad surface for touch to be considered
2664 a tap and be reported as a left button click (for
2665 touchpads working in absolute mode only).
2667 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices
2668 reporting absolute coordinates, such as tablets
2669 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices
2670 reporting absolute coordinates, such as tablets
2672 movablecore= [KNL,X86,IA-64,PPC]
2673 Format: nn[KMGTPE] | nn%
2674 This parameter is the complement to kernelcore=, it
2675 specifies the amount of memory used for migratable
2676 allocations. If both kernelcore and movablecore is
2677 specified, then kernelcore will be at *least* the
2678 specified value but may be more. If movablecore on its
2679 own is specified, the administrator must be careful
2680 that the amount of memory usable for all allocations
2683 movable_node [KNL] Boot-time switch to make hotplugable memory
2684 NUMA nodes to be movable. This means that the memory
2685 of such nodes will be usable only for movable
2686 allocations which rules out almost all kernel
2687 allocations. Use with caution!
2689 MTD_Partition= [MTD]
2690 Format: <name>,<region-number>,<size>,<offset>
2692 MTD_Region= [MTD] Format:
2693 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>]
2696 See drivers/mtd/cmdlinepart.c.
2698 multitce=off [PPC] This parameter disables the use of the pSeries
2699 firmware feature for updating multiple TCE entries
2702 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration
2704 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock]
2706 boundary - index of last SLC block on Flex-OneNAND.
2707 The remaining blocks are configured as MLC blocks.
2708 lock - Configure if Flex-OneNAND boundary should be locked.
2709 Once locked, the boundary cannot be changed.
2710 1 indicates lock status, 0 indicates unlock status.
2713 ARM/S3C2412 JIVE boot control
2715 See arch/arm/mach-s3c2412/mach-jive.c
2717 mtouchusb.raw_coordinates=
2718 [HW] Make the MicroTouch USB driver use raw coordinates
2719 ('y', default) or cooked coordinates ('n')
2721 mtrr_chunk_size=nn[KMG] [X86]
2722 used for mtrr cleanup. It is largest continuous chunk
2723 that could hold holes aka. UC entries.
2725 mtrr_gran_size=nn[KMG] [X86]
2726 Used for mtrr cleanup. It is granularity of mtrr block.
2728 Large value could prevent small alignment from
2731 mtrr_spare_reg_nr=n [X86]
2733 Range: 0,7 : spare reg number
2735 Used for mtrr cleanup. It is spare mtrr entries number.
2736 Set to 2 or more if your graphical card needs more.
2738 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card
2740 netdev= [NET] Network devices parameters
2741 Format: <irq>,<io>,<mem_start>,<mem_end>,<name>
2742 Note that mem_start is often overloaded to mean
2743 something different and driver-specific.
2744 This usage is only documented in each driver source
2748 [NETFILTER] Enable connection tracking flow accounting
2749 0 to disable accounting
2750 1 to enable accounting
2753 nfsaddrs= [NFS] Deprecated. Use ip= instead.
2754 See Documentation/filesystems/nfs/nfsroot.txt.
2756 nfsroot= [NFS] nfs root filesystem for disk-less boxes.
2757 See Documentation/filesystems/nfs/nfsroot.txt.
2759 nfsrootdebug [NFS] enable nfsroot debugging messages.
2760 See Documentation/filesystems/nfs/nfsroot.txt.
2762 nfs.callback_nr_threads=
2763 [NFSv4] set the total number of threads that the
2764 NFS client will assign to service NFSv4 callback
2767 nfs.callback_tcpport=
2768 [NFS] set the TCP port on which the NFSv4 callback
2769 channel should listen.
2772 [NFS] sets the pathname to the program which is used
2773 to update the NFS client cache entries.
2775 nfs.cache_getent_timeout=
2776 [NFS] sets the timeout after which an attempt to
2777 update a cache entry is deemed to have failed.
2779 nfs.idmap_cache_timeout=
2780 [NFS] set the maximum lifetime for idmapper cache
2784 [NFS] enable 64-bit inode numbers.
2785 If zero, the NFS client will fake up a 32-bit inode
2786 number for the readdir() and stat() syscalls instead
2787 of returning the full 64-bit number.
2788 The default is to return 64-bit inode numbers.
2790 nfs.max_session_cb_slots=
2791 [NFSv4.1] Sets the maximum number of session
2792 slots the client will assign to the callback
2793 channel. This determines the maximum number of
2794 callbacks the client will process in parallel for
2795 a particular server.
2797 nfs.max_session_slots=
2798 [NFSv4.1] Sets the maximum number of session slots
2799 the client will attempt to negotiate with the server.
2800 This limits the number of simultaneous RPC requests
2801 that the client can send to the NFSv4.1 server.
2802 Note that there is little point in setting this
2803 value higher than the max_tcp_slot_table_limit.
2805 nfs.nfs4_disable_idmapping=
2806 [NFSv4] When set to the default of '1', this option
2807 ensures that both the RPC level authentication
2808 scheme and the NFS level operations agree to use
2809 numeric uids/gids if the mount is using the
2810 'sec=sys' security flavour. In effect it is
2811 disabling idmapping, which can make migration from
2812 legacy NFSv2/v3 systems to NFSv4 easier.
2813 Servers that do not support this mode of operation
2814 will be autodetected by the client, and it will fall
2815 back to using the idmapper.
2816 To turn off this behaviour, set the value to '0'.
2818 [NFS4] Specify an additional fixed unique ident-
2819 ification string that NFSv4 clients can insert into
2820 their nfs_client_id4 string. This is typically a
2821 UUID that is generated at system install time.
2823 nfs.send_implementation_id =
2824 [NFSv4.1] Send client implementation identification
2825 information in exchange_id requests.
2826 If zero, no implementation identification information
2828 The default is to send the implementation identification
2831 nfs.recover_lost_locks =
2832 [NFSv4] Attempt to recover locks that were lost due
2833 to a lease timeout on the server. Please note that
2834 doing this risks data corruption, since there are
2835 no guarantees that the file will remain unchanged
2836 after the locks are lost.
2837 If you want to enable the kernel legacy behaviour of
2838 attempting to recover these locks, then set this
2840 The default parameter value of '0' causes the kernel
2841 not to attempt recovery of lost locks.
2843 nfs4.layoutstats_timer =
2844 [NFSv4.2] Change the rate at which the kernel sends
2845 layoutstats to the pNFS metadata server.
2847 Setting this to value to 0 causes the kernel to use
2848 whatever value is the default set by the layout
2849 driver. A non-zero value sets the minimum interval
2850 in seconds between layoutstats transmissions.
2852 nfsd.nfs4_disable_idmapping=
2853 [NFSv4] When set to the default of '1', the NFSv4
2854 server will return only numeric uids and gids to
2855 clients using auth_sys, and will accept numeric uids
2856 and gids from such clients. This is intended to ease
2857 migration from NFSv2/v3.
2859 nmi_debug= [KNL,SH] Specify one or more actions to take
2860 when a NMI is triggered.
2861 Format: [state][,regs][,debounce][,die]
2863 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels
2864 Format: [panic,][nopanic,][num]
2866 0 - turn hardlockup detector in nmi_watchdog off
2867 1 - turn hardlockup detector in nmi_watchdog on
2868 When panic is specified, panic when an NMI watchdog
2869 timeout occurs (or 'nopanic' to not panic on an NMI
2870 watchdog, if CONFIG_BOOTPARAM_HARDLOCKUP_PANIC is set)
2871 To disable both hard and soft lockup detectors,
2872 please see 'nowatchdog'.
2873 This is useful when you use a panic=... timeout and
2874 need the box quickly up again.
2876 These settings can be accessed at runtime via
2877 the nmi_watchdog and hardlockup_panic sysctls.
2879 netpoll.carrier_timeout=
2880 [NET] Specifies amount of time (in seconds) that
2881 netpoll should wait for a carrier. By default netpoll
2884 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths
2885 emulation library even if a 387 maths coprocessor
2888 no5lvl [X86-64] Disable 5-level paging mode. Forces
2889 kernel to use 4-level paging instead.
2892 [HW] Never suspend the console
2893 Disable suspending of consoles during suspend and
2894 hibernate operations. Once disabled, debugging
2895 messages can reach various consoles while the rest
2896 of the system is being put to sleep (ie, while
2897 debugging driver suspend/resume hooks). This may
2898 not work reliably with all consoles, but is known
2899 to work with serial and VGA consoles.
2900 To facilitate more flexible debugging, we also add
2901 console_suspend, a printk module parameter to control
2902 it. Users could use console_suspend (usually
2903 /sys/module/printk/parameters/console_suspend) to
2904 turn on/off it dynamically.
2906 novmcoredd [KNL,KDUMP]
2907 Disable device dump. Device dump allows drivers to
2908 append dump data to vmcore so you can collect driver
2909 specified debug info. Drivers can append the data
2910 without any limit and this data is stored in memory,
2911 so this may cause significant memory stress. Disabling
2912 device dump can help save memory but the driver debug
2913 data will be no longer available. This parameter
2914 is only available when CONFIG_PROC_VMCORE_DEVICE_DUMP
2917 noaliencache [MM, NUMA, SLAB] Disables the allocation of alien
2918 caches in the slab allocator. Saves per-node memory,
2919 but will impact performance.
2923 noaltinstr [S390] Disables alternative instructions patching
2924 (CPU alternatives feature).
2926 noapic [SMP,APIC] Tells the kernel to not make use of any
2927 IOAPICs that may be present in the system.
2929 noautogroup Disable scheduler automatic task group creation.
2931 nobats [PPC] Do not use BATs for mapping kernel lowmem
2932 on "Classic" PPC cores.
2936 noclflush [BUGS=X86] Don't use the CLFLUSH instruction
2938 nodelayacct [KNL] Disable per-task delay accounting
2940 nodsp [SH] Disable hardware DSP at boot time.
2942 noefi Disable EFI runtime services support.
2947 On X86-32 available only on PAE configured kernels.
2948 noexec=on: enable non-executable mappings (default)
2949 noexec=off: disable non-executable mappings
2952 Disable SMAP (Supervisor Mode Access Prevention)
2953 even if it is supported by processor.
2956 Disable SMEP (Supervisor Mode Execution Prevention)
2957 even if it is supported by processor.
2960 This affects only 32-bit executables.
2961 noexec32=on: enable non-executable mappings (default)
2962 read doesn't imply executable mappings
2963 noexec32=off: disable non-executable mappings
2964 read implies executable mappings
2966 nofpu [MIPS,SH] Disable hardware FPU at boot time.
2968 nofxsr [BUGS=X86-32] Disables x86 floating point extended
2969 register save and restore. The kernel will only save
2970 legacy floating-point registers on task switch.
2972 nohugeiomap [KNL,x86,PPC] Disable kernel huge I/O mappings.
2974 nosmt [KNL,S390] Disable symmetric multithreading (SMT).
2975 Equivalent to smt=1.
2977 [KNL,x86] Disable symmetric multithreading (SMT).
2978 nosmt=force: Force disable SMT, cannot be undone
2979 via the sysfs control file.
2981 nospectre_v1 [X86,PPC] Disable mitigations for Spectre Variant 1
2982 (bounds check bypass). With this option data leaks are
2983 possible in the system.
2985 nospectre_v2 [X86,PPC_FSL_BOOK3E,ARM64] Disable all mitigations for
2986 the Spectre variant 2 (indirect branch prediction)
2987 vulnerability. System may allow data leaks with this
2990 nospec_store_bypass_disable
2991 [HW] Disable all mitigations for the Speculative Store Bypass vulnerability
2993 noxsave [BUGS=X86] Disables x86 extended register state save
2994 and restore using xsave. The kernel will fallback to
2995 enabling legacy floating-point and sse state.
2997 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended
2998 register states. The kernel will fall back to use
2999 xsave to save the states. By using this parameter,
3000 performance of saving the states is degraded because
3001 xsave doesn't support modified optimization while
3002 xsaveopt supports it on xsaveopt enabled systems.
3004 noxsaves [X86] Disables xsaves and xrstors used in saving and
3005 restoring x86 extended register state in compacted
3006 form of xsave area. The kernel will fall back to use
3007 xsaveopt and xrstor to save and restore the states
3008 in standard form of xsave area. By using this
3009 parameter, xsave area per process might occupy more
3010 memory on xsaves enabled systems.
3012 nohlt [BUGS=ARM,SH] Tells the kernel that the sleep(SH) or
3013 wfi(ARM) instruction doesn't work correctly and not to
3014 use it. This is also useful when using JTAG debugger.
3016 no_file_caps Tells the kernel not to honor file capabilities. The
3017 only way then for a file to be executed with privilege
3018 is to be setuid root or executed by root.
3020 nohalt [IA-64] Tells the kernel not to use the power saving
3021 function PAL_HALT_LIGHT when idle. This increases
3022 power-consumption. On the positive side, it reduces
3023 interrupt wake-up latency, which may improve performance
3024 in certain environments such as networked servers or
3027 nohibernate [HIBERNATION] Disable hibernation and resume.
3029 nohz= [KNL] Boottime enable/disable dynamic ticks
3030 Valid arguments: on, off
3033 nohz_full= [KNL,BOOT,SMP,ISOL]
3034 The argument is a cpu list, as described above.
3035 In kernels built with CONFIG_NO_HZ_FULL=y, set
3036 the specified list of CPUs whose tick will be stopped
3037 whenever possible. The boot CPU will be forced outside
3038 the range to maintain the timekeeping. Any CPUs
3039 in this list will have their RCU callbacks offloaded,
3040 just as if they had also been called out in the
3041 rcu_nocbs= boot parameter.
3043 noiotrap [SH] Disables trapped I/O port accesses.
3045 noirqdebug [X86-32] Disables the code which attempts to detect and
3046 disable unhandled interrupt sources.
3048 no_timer_check [X86,APIC] Disables the code which tests for
3049 broken timer IRQ sources.
3051 noisapnp [ISAPNP] Disables ISA PnP code.
3053 noinitrd [RAM] Tells the kernel not to load any configured
3056 nointremap [X86-64, Intel-IOMMU] Do not enable interrupt
3058 [Deprecated - use intremap=off]
3062 noinvpcid [X86] Disable the INVPCID cpu feature.
3064 nojitter [IA-64] Disables jitter checking for ITC timers.
3066 no-kvmclock [X86,KVM] Disable paravirtualized KVM clock driver
3068 no-kvmapf [X86,KVM] Disable paravirtualized asynchronous page
3072 [X86,PV_OPS] Disable paravirtualized VMware scheduler
3073 clock and use the default one.
3075 no-steal-acc [X86,KVM] Disable paravirtualized steal time accounting.
3076 steal time is computed, but won't influence scheduler
3079 nolapic [X86-32,APIC] Do not enable or use the local APIC.
3081 nolapic_timer [X86-32,APIC] Do not use the local APIC timer.
3083 noltlbs [PPC] Do not use large page/tlb entries for kernel
3084 lowmem mapping on PPC40x and PPC8xx
3086 nomca [IA-64] Disable machine check abort handling
3088 nomce [X86-32] Disable Machine Check Exception
3090 nomfgpt [X86-32] Disable Multi-Function General Purpose
3091 Timer usage (for AMD Geode machines).
3093 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to
3094 shutdown the other cpus. Instead use the REBOOT_VECTOR
3097 nomodule Disable module load
3099 nopat [X86] Disable PAT (page attribute table extension of
3100 pagetables) support.
3102 nopcid [X86-64] Disable the PCID cpu feature.
3104 norandmaps Don't use address space randomization. Equivalent to
3105 echo 0 > /proc/sys/kernel/randomize_va_space
3107 noreplace-smp [X86-32,SMP] Don't replace SMP instructions
3108 with UP alternatives
3110 nordrand [X86] Disable kernel use of the RDRAND and
3111 RDSEED instructions even if they are supported
3112 by the processor. RDRAND and RDSEED are still
3113 available to user space applications.
3115 noresume [SWSUSP] Disables resume and restores original swap
3118 no-scroll [VGA] Disables scrollback.
3119 This is required for the Braillex ib80-piezo Braille
3120 reader made by F.H. Papenmeier (Germany).
3124 nosep [BUGS=X86-32] Disables x86 SYSENTER/SYSEXIT support.
3126 nosmp [SMP] Tells an SMP kernel to act as a UP kernel,
3127 and disable the IO APIC. legacy for "maxcpus=0".
3129 nosoftlockup [KNL] Disable the soft-lockup detector.
3131 nosync [HW,M68K] Disables sync negotiation for all devices.
3133 nowatchdog [KNL] Disable both lockup detectors, i.e.
3134 soft-lockup and NMI watchdog (hard-lockup).
3138 nox2apic [X86-64,APIC] Do not enable x2APIC mode.
3140 cpu0_hotplug [X86] Turn on CPU0 hotplug feature when
3141 CONFIG_BOOTPARAM_HOTPLUG_CPU0 is off.
3142 Some features depend on CPU0. Known dependencies are:
3143 1. Resume from suspend/hibernate depends on CPU0.
3144 Suspend/hibernate will fail if CPU0 is offline and you
3145 need to online CPU0 before suspend/hibernate.
3146 2. PIC interrupts also depend on CPU0. CPU0 can't be
3147 removed if a PIC interrupt is detected.
3148 It's said poweroff/reboot may depend on CPU0 on some
3149 machines although I haven't seen such issues so far
3150 after CPU0 is offline on a few tested machines.
3151 If the dependencies are under your control, you can
3152 turn on cpu0_hotplug.
3154 nps_mtm_hs_ctr= [KNL,ARC]
3155 This parameter sets the maximum duration, in
3156 cycles, each HW thread of the CTOP can run
3157 without interruptions, before HW switches it.
3158 The actual maximum duration is 16 times this
3160 Format: integer between 1 and 255
3163 nptcg= [IA-64] Override max number of concurrent global TLB
3164 purges which is reported from either PAL_VM_SUMMARY or
3167 nr_cpus= [SMP] Maximum number of processors that an SMP kernel
3168 could support. nr_cpus=n : n >= 1 limits the kernel to
3169 support 'n' processors. It could be larger than the
3170 number of already plugged CPU during bootup, later in
3171 runtime you can physically add extra cpu until it reaches
3172 n. So during boot up some boot time memory for per-cpu
3173 variables need be pre-allocated for later physical cpu
3176 nr_uarts= [SERIAL] maximum number of UARTs to be registered.
3178 numa_balancing= [KNL,X86] Enable or disable automatic NUMA balancing.
3179 Allowed values are enable and disable
3181 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA.
3182 'node', 'default' can be specified
3183 This can be set from sysctl after boot.
3184 See Documentation/admin-guide/sysctl/vm.rst for details.
3186 ohci1394_dma=early [HW] enable debugging via the ohci1394 driver.
3187 See Documentation/debugging-via-ohci1394.txt for more
3190 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands
3191 Rather than timing out after 20 ms if an EC
3192 command is not properly ACKed, override the length
3193 of the timeout. We have interrupts disabled while
3194 waiting for the ACK, so if this is set too high
3195 interrupts *may* be lost!
3197 omap_mux= [OMAP] Override bootloader pin multiplexing.
3198 Format: <mux_mode0.mode_name=value>...
3199 For example, to override I2C bus2:
3200 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100
3202 oprofile.timer= [HW]
3203 Use timer interrupt instead of performance counters
3205 oprofile.cpu_type= Force an oprofile cpu type
3206 This might be useful if you have an older oprofile
3207 userland or if you want common events.
3208 Format: { arch_perfmon }
3209 arch_perfmon: [X86] Force use of architectural
3210 perfmon on Intel CPUs instead of the
3211 CPU specific event set.
3212 timer: [X86] Force use of architectural NMI
3213 timer mode (see also oprofile.timer
3214 for generic hr timer mode)
3216 oops=panic Always panic on oopses. Default is to just kill the
3217 process, but there is a small probability of
3218 deadlocking the machine.
3219 This will also cause panics on machine check exceptions.
3220 Useful together with panic=30 to trigger a reboot.
3223 [KNL] Boolean flag to control whether the page allocator
3224 should randomize its free lists. The randomization may
3225 be automatically enabled if the kernel detects it is
3226 running on a platform with a direct-mapped memory-side
3227 cache, and this parameter can be used to
3228 override/disable that behavior. The state of the flag
3229 can be read from sysfs at:
3230 /sys/module/page_alloc/parameters/shuffle.
3232 page_owner= [KNL] Boot-time page_owner enabling option.
3233 Storage of the information about who allocated
3234 each page is disabled in default. With this switch,
3236 on: enable the feature
3238 page_poison= [KNL] Boot-time parameter changing the state of
3239 poisoning on the buddy allocator, available with
3240 CONFIG_PAGE_POISONING=y.
3241 off: turn off poisoning (default)
3242 on: turn on poisoning
3244 panic= [KNL] Kernel behaviour on panic: delay <timeout>
3245 timeout > 0: seconds before rebooting
3246 timeout = 0: wait forever
3247 timeout < 0: reboot immediately
3250 panic_print= Bitmask for printing system info when panic happens.
3251 User can chose combination of the following bits:
3252 bit 0: print all tasks info
3253 bit 1: print system memory info
3254 bit 2: print timer info
3255 bit 3: print locks info if CONFIG_LOCKDEP is on
3256 bit 4: print ftrace buffer
3257 bit 5: print all printk messages in buffer
3259 panic_on_warn panic() instead of WARN(). Useful to cause kdump
3262 crash_kexec_post_notifiers
3263 Run kdump after running panic-notifiers and dumping
3264 kmsg. This only for the users who doubt kdump always
3265 succeeds in any situation.
3266 Note that this also increases risks of kdump failure,
3267 because some panic notifiers can make the crashed
3268 kernel more unstable.
3270 parkbd.port= [HW] Parallel port number the keyboard adapter is
3271 connected to, default is 0.
3273 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation,
3274 0 for XT, 1 for AT (default is AT).
3277 parport= [HW,PPT] Specify parallel ports. 0 disables.
3278 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] }
3279 Use 'auto' to force the driver to use any
3280 IRQ/DMA settings detected (the default is to
3281 ignore detected IRQ/DMA settings because of
3282 possible conflicts). You can specify the base
3283 address, IRQ, and DMA settings; IRQ and DMA
3284 should be numbers, or 'auto' (for using detected
3285 settings on that particular port), or 'nofifo'
3286 (to avoid using a FIFO even if it is detected).
3287 Parallel ports are assigned in the order they
3288 are specified on the command line, starting
3291 parport_init_mode= [HW,PPT]
3292 Configure VIA parallel port to operate in
3293 a specific mode. This is necessary on Pegasos
3294 computer where firmware has no options for setting
3295 up parallel port mode and sets it to spp.
3296 Currently this function knows 686a and 8231 chips.
3297 Format: [spp|ps2|epp|ecp|ecpepp]
3300 Halt all CPUs after the first oops has been printed for
3301 the specified number of seconds. This is to be used if
3302 your oopses keep scrolling off the screen.
3307 See header of drivers/block/paride/pcd.c.
3308 See also Documentation/admin-guide/blockdev/paride.rst.
3310 pci=option[,option...] [PCI] various PCI subsystem options.
3312 Some options herein operate on a specific device
3313 or a set of devices (<pci_dev>). These are
3314 specified in one of the following formats:
3316 [<domain>:]<bus>:<dev>.<func>[/<dev>.<func>]*
3317 pci:<vendor>:<device>[:<subvendor>:<subdevice>]
3319 Note: the first format specifies a PCI
3320 bus/device/function address which may change
3321 if new hardware is inserted, if motherboard
3322 firmware changes, or due to changes caused
3323 by other kernel parameters. If the
3324 domain is left unspecified, it is
3325 taken to be zero. Optionally, a path
3326 to a device through multiple device/function
3327 addresses can be specified after the base
3328 address (this is more robust against
3329 renumbering issues). The second format
3330 selects devices using IDs from the
3331 configuration space which may match multiple
3332 devices in the system.
3334 earlydump dump PCI config space before the kernel
3336 off [X86] don't probe for the PCI bus
3337 bios [X86-32] force use of PCI BIOS, don't access
3338 the hardware directly. Use this if your machine
3339 has a non-standard PCI host bridge.
3340 nobios [X86-32] disallow use of PCI BIOS, only direct
3341 hardware access methods are allowed. Use this
3342 if you experience crashes upon bootup and you
3343 suspect they are caused by the BIOS.
3344 conf1 [X86] Force use of PCI Configuration Access
3345 Mechanism 1 (config address in IO port 0xCF8,
3346 data in IO port 0xCFC, both 32-bit).
3347 conf2 [X86] Force use of PCI Configuration Access
3348 Mechanism 2 (IO port 0xCF8 is an 8-bit port for
3349 the function, IO port 0xCFA, also 8-bit, sets
3350 bus number. The config space is then accessed
3351 through ports 0xC000-0xCFFF).
3352 See http://wiki.osdev.org/PCI for more info
3353 on the configuration access mechanisms.
3354 noaer [PCIE] If the PCIEAER kernel config parameter is
3355 enabled, this kernel boot option can be used to
3356 disable the use of PCIE advanced error reporting.
3357 nodomains [PCI] Disable support for multiple PCI
3358 root domains (aka PCI segments, in ACPI-speak).
3359 nommconf [X86] Disable use of MMCONFIG for PCI
3361 check_enable_amd_mmconf [X86] check for and enable
3362 properly configured MMIO access to PCI
3363 config space on AMD family 10h CPU
3364 nomsi [MSI] If the PCI_MSI kernel config parameter is
3365 enabled, this kernel boot option can be used to
3366 disable the use of MSI interrupts system-wide.
3367 noioapicquirk [APIC] Disable all boot interrupt quirks.
3368 Safety option to keep boot IRQs enabled. This
3369 should never be necessary.
3370 ioapicreroute [APIC] Enable rerouting of boot IRQs to the
3371 primary IO-APIC for bridges that cannot disable
3372 boot IRQs. This fixes a source of spurious IRQs
3373 when the system masks IRQs.
3374 noioapicreroute [APIC] Disable workaround that uses the
3375 boot IRQ equivalent of an IRQ that connects to
3376 a chipset where boot IRQs cannot be disabled.
3377 The opposite of ioapicreroute.
3378 biosirq [X86-32] Use PCI BIOS calls to get the interrupt
3379 routing table. These calls are known to be buggy
3380 on several machines and they hang the machine
3381 when used, but on other computers it's the only
3382 way to get the interrupt routing table. Try
3383 this option if the kernel is unable to allocate
3384 IRQs or discover secondary PCI buses on your
3386 rom [X86] Assign address space to expansion ROMs.
3387 Use with caution as certain devices share
3388 address decoders between ROMs and other
3390 norom [X86] Do not assign address space to
3391 expansion ROMs that do not already have
3392 BIOS assigned address ranges.
3393 nobar [X86] Do not assign address space to the
3394 BARs that weren't assigned by the BIOS.
3395 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be
3396 assigned automatically to PCI devices. You can
3397 make the kernel exclude IRQs of your ISA cards
3399 pirqaddr=0xAAAAA [X86] Specify the physical address
3400 of the PIRQ table (normally generated
3401 by the BIOS) if it is outside the
3402 F0000h-100000h range.
3403 lastbus=N [X86] Scan all buses thru bus #N. Can be
3404 useful if the kernel is unable to find your
3405 secondary buses and you want to tell it
3406 explicitly which ones they are.
3407 assign-busses [X86] Always assign all PCI bus
3408 numbers ourselves, overriding
3409 whatever the firmware may have done.
3410 usepirqmask [X86] Honor the possible IRQ mask stored
3411 in the BIOS $PIR table. This is needed on
3412 some systems with broken BIOSes, notably
3413 some HP Pavilion N5400 and Omnibook XE3
3414 notebooks. This will have no effect if ACPI
3415 IRQ routing is enabled.
3416 noacpi [X86] Do not use ACPI for IRQ routing
3417 or for PCI scanning.
3418 use_crs [X86] Use PCI host bridge window information
3419 from ACPI. On BIOSes from 2008 or later, this
3420 is enabled by default. If you need to use this,
3421 please report a bug.
3422 nocrs [X86] Ignore PCI host bridge windows from ACPI.
3423 If you need to use this, please report a bug.
3424 routeirq Do IRQ routing for all PCI devices.
3425 This is normally done in pci_enable_device(),
3426 so this option is a temporary workaround
3427 for broken drivers that don't call it.
3428 skip_isa_align [X86] do not align io start addr, so can
3429 handle more pci cards
3430 noearly [X86] Don't do any early type 1 scanning.
3431 This might help on some broken boards which
3432 machine check when some devices' config space
3433 is read. But various workarounds are disabled
3434 and some IOMMU drivers will not work.
3435 bfsort Sort PCI devices into breadth-first order.
3436 This sorting is done to get a device
3437 order compatible with older (<= 2.4) kernels.
3438 nobfsort Don't sort PCI devices into breadth-first order.
3439 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size)
3440 tuning and use the BIOS-configured MPS defaults.
3441 pcie_bus_safe Set every device's MPS to the largest value
3442 supported by all devices below the root complex.
3443 pcie_bus_perf Set device MPS to the largest allowable MPS
3444 based on its parent bus. Also set MRRS (Max
3445 Read Request Size) to the largest supported
3446 value (no larger than the MPS that the device
3447 or bus can support) for best performance.
3448 pcie_bus_peer2peer Set every device's MPS to 128B, which
3449 every device is guaranteed to support. This
3450 configuration allows peer-to-peer DMA between
3451 any pair of devices, possibly at the cost of
3452 reduced performance. This also guarantees
3453 that hot-added devices will work.
3454 cbiosize=nn[KMG] The fixed amount of bus space which is
3455 reserved for the CardBus bridge's IO window.
3456 The default value is 256 bytes.
3457 cbmemsize=nn[KMG] The fixed amount of bus space which is
3458 reserved for the CardBus bridge's memory
3459 window. The default value is 64 megabytes.
3462 [<order of align>@]<pci_dev>[; ...]
3463 Specifies alignment and device to reassign
3464 aligned memory resources. How to
3465 specify the device is described above.
3466 If <order of align> is not specified,
3467 PAGE_SIZE is used as alignment.
3468 PCI-PCI bridge can be specified, if resource
3469 windows need to be expanded.
3470 To specify the alignment for several
3471 instances of a device, the PCI vendor,
3472 device, subvendor, and subdevice may be
3473 specified, e.g., 4096@pci:8086:9c22:103c:198f
3474 ecrc= Enable/disable PCIe ECRC (transaction layer
3475 end-to-end CRC checking).
3476 bios: Use BIOS/firmware settings. This is the
3480 hpiosize=nn[KMG] The fixed amount of bus space which is
3481 reserved for hotplug bridge's IO window.
3482 Default size is 256 bytes.
3483 hpmemsize=nn[KMG] The fixed amount of bus space which is
3484 reserved for hotplug bridge's memory window.
3485 Default size is 2 megabytes.
3486 hpbussize=nn The minimum amount of additional bus numbers
3487 reserved for buses below a hotplug bridge.
3489 realloc= Enable/disable reallocating PCI bridge resources
3490 if allocations done by BIOS are too small to
3491 accommodate resources required by all child
3493 off: Turn realloc off
3495 realloc same as realloc=on
3496 noari do not use PCIe ARI.
3497 noats [PCIE, Intel-IOMMU, AMD-IOMMU]
3498 do not use PCIe ATS (and IOMMU device IOTLB).
3499 pcie_scan_all Scan all possible PCIe devices. Otherwise we
3500 only look for one device below a PCIe downstream
3502 big_root_window Try to add a big 64bit memory window to the PCIe
3503 root complex on AMD CPUs. Some GFX hardware
3504 can resize a BAR to allow access to all VRAM.
3505 Adding the window is slightly risky (it may
3506 conflict with unreported devices), so this
3508 disable_acs_redir=<pci_dev>[; ...]
3509 Specify one or more PCI devices (in the format
3510 specified above) separated by semicolons.
3511 Each device specified will have the PCI ACS
3512 redirect capabilities forced off which will
3513 allow P2P traffic between devices through
3514 bridges without forcing it upstream. Note:
3515 this removes isolation between devices and
3516 may put more devices in an IOMMU group.
3517 force_floating [S390] Force usage of floating interrupts.
3518 nomio [S390] Do not use MIO instructions.
3520 pcie_aspm= [PCIE] Forcibly enable or disable PCIe Active State Power
3523 force Enable ASPM even on devices that claim not to support it.
3524 WARNING: Forcing ASPM on may cause system lockups.
3526 pcie_ports= [PCIE] PCIe port services handling:
3527 native Use native PCIe services (PME, AER, DPC, PCIe hotplug)
3528 even if the platform doesn't give the OS permission to
3529 use them. This may cause conflicts if the platform
3530 also tries to use these services.
3531 compat Disable native PCIe services (PME, AER, DPC, PCIe
3534 pcie_port_pm= [PCIE] PCIe port power management handling:
3535 off Disable power management of all PCIe ports
3536 force Forcibly enable power management of all PCIe ports
3538 pcie_pme= [PCIE,PM] Native PCIe PME signaling options:
3539 nomsi Do not use MSI for native PCIe PME signaling (this makes
3540 all PCIe root ports use INTx for all services).
3542 pcmv= [HW,PCMCIA] BadgePAD 4
3546 Keep all power-domains already enabled by bootloader on,
3547 even if no driver has claimed them. This is useful
3548 for debug and development, but should not be
3549 needed on a platform with proper driver support.
3552 See Documentation/admin-guide/blockdev/paride.rst.
3554 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at
3557 See arch/parisc/kernel/pdc_chassis.c
3559 percpu_alloc= Select which percpu first chunk allocator to use.
3560 Currently supported values are "embed" and "page".
3561 Archs may support subset or none of the selections.
3562 See comments in mm/percpu.c for details on each
3563 allocator. This parameter is primarily for debugging
3564 and performance comparison.
3567 See Documentation/admin-guide/blockdev/paride.rst.
3570 See Documentation/admin-guide/blockdev/paride.rst.
3572 pirq= [SMP,APIC] Manual mp-table setup
3573 See Documentation/x86/i386/IO-APIC.rst.
3575 plip= [PPT,NET] Parallel port network link
3576 Format: { parport<nr> | timid | 0 }
3577 See also Documentation/admin-guide/parport.rst.
3579 pmtmr= [X86] Manual setup of pmtmr I/O Port.
3580 Override pmtimer IOPort with a hex value.
3584 Enable PNP debug messages (depends on the
3585 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time
3586 via /sys/module/pnp/parameters/debug. We always show
3587 current resource usage; turning this on also shows
3588 possible settings and some assignment information.
3594 { on | off | curr | res | no-curr | no-res }
3597 [ISAPNP] Exclude IRQs for the autoconfiguration
3600 [ISAPNP] Exclude DMAs for the autoconfiguration
3602 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration
3603 Ranges are in pairs (I/O port base and size).
3606 [ISAPNP] Exclude memory regions for the
3608 Ranges are in pairs (memory base and size).
3610 ports= [IP_VS_FTP] IPVS ftp helper module
3612 Up to 8 (IP_VS_APP_MAX_PORTS) ports
3614 Format: <port>,<port>....
3616 powersave=off [PPC] This option disables power saving features.
3617 It specifically disables cpuidle and sets the
3618 platform machine description specific power_save
3619 function to NULL. On Idle the CPU just reduces
3622 ppc_strict_facility_enable
3623 [PPC] This option catches any kernel floating point,
3624 Altivec, VSX and SPE outside of regions specifically
3625 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()).
3626 There is some performance impact when enabling this.
3630 Disable Hardware Transactional Memory
3632 print-fatal-signals=
3633 [KNL] debug: print fatal signals
3635 If enabled, warn about various signal handling
3636 related application anomalies: too many signals,
3637 too many POSIX.1 timers, fatal signals causing a
3640 If you hit the warning due to signal overflow,
3641 you might want to try "ulimit -i unlimited".
3645 printk.always_kmsg_dump=
3646 Trigger kmsg_dump for cases other than kernel oops or
3648 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
3651 printk.devkmsg={on,off,ratelimit}
3652 Control writing to /dev/kmsg.
3653 on - unlimited logging to /dev/kmsg from userspace
3654 off - logging to /dev/kmsg disabled
3655 ratelimit - ratelimit the logging
3658 printk.time= Show timing data prefixed to each printk message line
3659 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
3661 processor.max_cstate= [HW,ACPI]
3662 Limit processor to maximum C-state
3663 max_cstate=9 overrides any DMI blacklist limit.
3665 processor.nocst [HW,ACPI]
3666 Ignore the _CST method to determine C-states,
3667 instead using the legacy FADT method
3669 profile= [KNL] Enable kernel profiling via /proc/profile
3670 Format: [<profiletype>,]<number>
3671 Param: <profiletype>: "schedule", "sleep", or "kvm"
3672 [defaults to kernel profiling]
3673 Param: "schedule" - profile schedule points.
3674 Param: "sleep" - profile D-state sleeping (millisecs).
3675 Requires CONFIG_SCHEDSTATS
3676 Param: "kvm" - profile VM exits.
3677 Param: <number> - step/bucket size as a power of 2 for
3678 statistical time based profiling.
3680 prompt_ramdisk= [RAM] List of RAM disks to prompt for floppy disk
3682 See Documentation/admin-guide/blockdev/ramdisk.rst.
3684 psi= [KNL] Enable or disable pressure stall information
3688 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to
3689 probe for; one of (bare|imps|exps|lifebook|any).
3690 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports
3692 psmouse.resetafter= [HW,MOUSE]
3693 Try to reset the device after so many bad packets
3696 [HW,MOUSE] Set desired mouse resolution, in dpi.
3697 psmouse.smartscroll=
3698 [HW,MOUSE] Controls Logitech smartscroll autorepeat.
3699 0 = disabled, 1 = enabled (default).
3701 pstore.backend= Specify the name of the pstore backend to use
3704 See Documentation/admin-guide/blockdev/paride.rst.
3706 pti= [X86_64] Control Page Table Isolation of user and
3707 kernel address spaces. Disabling this feature
3708 removes hardening, but improves performance of
3709 system calls and interrupts.
3711 on - unconditionally enable
3712 off - unconditionally disable
3713 auto - kernel detects whether your CPU model is
3714 vulnerable to issues that PTI mitigates
3716 Not specifying this option is equivalent to pti=auto.
3719 Equivalent to pti=off
3722 [KNL] Number of legacy pty's. Overwrites compiled-in
3725 quiet [KNL] Disable most log messages
3730 See Documentation/admin-guide/md.rst.
3732 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes
3733 See Documentation/admin-guide/blockdev/ramdisk.rst.
3735 random.trust_cpu={on,off}
3736 [KNL] Enable or disable trusting the use of the
3737 CPU's random number generator (if available) to
3738 fully seed the kernel's CRNG. Default is controlled
3739 by CONFIG_RANDOM_TRUST_CPU.
3741 ras=option[,option,...] [KNL] RAS-specific options
3744 Disable the Correctable Errors Collector,
3745 see CONFIG_RAS_CEC help text.
3748 The argument is a cpu list, as described above,
3749 except that the string "all" can be used to
3750 specify every CPU on the system.
3752 In kernels built with CONFIG_RCU_NOCB_CPU=y, set
3753 the specified list of CPUs to be no-callback CPUs.
3754 Invocation of these CPUs' RCU callbacks will be
3755 offloaded to "rcuox/N" kthreads created for that
3756 purpose, where "x" is "p" for RCU-preempt, and
3757 "s" for RCU-sched, and "N" is the CPU number.
3758 This reduces OS jitter on the offloaded CPUs,
3759 which can be useful for HPC and real-time
3760 workloads. It can also improve energy efficiency
3761 for asymmetric multiprocessors.
3764 Rather than requiring that offloaded CPUs
3765 (specified by rcu_nocbs= above) explicitly
3766 awaken the corresponding "rcuoN" kthreads,
3767 make these kthreads poll for callbacks.
3768 This improves the real-time response for the
3769 offloaded CPUs by relieving them of the need to
3770 wake up the corresponding kthread, but degrades
3771 energy efficiency by requiring that the kthreads
3772 periodically wake up to do the polling.
3774 rcutree.blimit= [KNL]
3775 Set maximum number of finished RCU callbacks to
3776 process in one batch.
3778 rcutree.dump_tree= [KNL]
3779 Dump the structure of the rcu_node combining tree
3780 out at early boot. This is used for diagnostic
3781 purposes, to verify correct tree setup.
3783 rcutree.gp_cleanup_delay= [KNL]
3784 Set the number of jiffies to delay each step of
3785 RCU grace-period cleanup.
3787 rcutree.gp_init_delay= [KNL]
3788 Set the number of jiffies to delay each step of
3789 RCU grace-period initialization.
3791 rcutree.gp_preinit_delay= [KNL]
3792 Set the number of jiffies to delay each step of
3793 RCU grace-period pre-initialization, that is,
3794 the propagation of recent CPU-hotplug changes up
3795 the rcu_node combining tree.
3797 rcutree.use_softirq= [KNL]
3798 If set to zero, move all RCU_SOFTIRQ processing to
3799 per-CPU rcuc kthreads. Defaults to a non-zero
3800 value, meaning that RCU_SOFTIRQ is used by default.
3801 Specify rcutree.use_softirq=0 to use rcuc kthreads.
3803 rcutree.rcu_fanout_exact= [KNL]
3804 Disable autobalancing of the rcu_node combining
3805 tree. This is used by rcutorture, and might
3806 possibly be useful for architectures having high
3807 cache-to-cache transfer latencies.
3809 rcutree.rcu_fanout_leaf= [KNL]
3810 Change the number of CPUs assigned to each
3811 leaf rcu_node structure. Useful for very
3812 large systems, which will choose the value 64,
3813 and for NUMA systems with large remote-access
3814 latencies, which will choose a value aligned
3815 with the appropriate hardware boundaries.
3817 rcutree.jiffies_till_first_fqs= [KNL]
3818 Set delay from grace-period initialization to
3819 first attempt to force quiescent states.
3820 Units are jiffies, minimum value is zero,
3821 and maximum value is HZ.
3823 rcutree.jiffies_till_next_fqs= [KNL]
3824 Set delay between subsequent attempts to force
3825 quiescent states. Units are jiffies, minimum
3826 value is one, and maximum value is HZ.
3828 rcutree.jiffies_till_sched_qs= [KNL]
3829 Set required age in jiffies for a
3830 given grace period before RCU starts
3831 soliciting quiescent-state help from
3832 rcu_note_context_switch() and cond_resched().
3833 If not specified, the kernel will calculate
3834 a value based on the most recent settings
3835 of rcutree.jiffies_till_first_fqs
3836 and rcutree.jiffies_till_next_fqs.
3837 This calculated value may be viewed in
3838 rcutree.jiffies_to_sched_qs. Any attempt to set
3839 rcutree.jiffies_to_sched_qs will be cheerfully
3842 rcutree.kthread_prio= [KNL,BOOT]
3843 Set the SCHED_FIFO priority of the RCU per-CPU
3844 kthreads (rcuc/N). This value is also used for
3845 the priority of the RCU boost threads (rcub/N)
3846 and for the RCU grace-period kthreads (rcu_bh,
3847 rcu_preempt, and rcu_sched). If RCU_BOOST is
3848 set, valid values are 1-99 and the default is 1
3849 (the least-favored priority). Otherwise, when
3850 RCU_BOOST is not set, valid values are 0-99 and
3851 the default is zero (non-realtime operation).
3853 rcutree.rcu_nocb_gp_stride= [KNL]
3854 Set the number of NOCB callback kthreads in
3855 each group, which defaults to the square root
3856 of the number of CPUs. Larger numbers reduce
3857 the wakeup overhead on the global grace-period
3858 kthread, but increases that same overhead on
3859 each group's NOCB grace-period kthread.
3861 rcutree.qhimark= [KNL]
3862 Set threshold of queued RCU callbacks beyond which
3863 batch limiting is disabled.
3865 rcutree.qlowmark= [KNL]
3866 Set threshold of queued RCU callbacks below which
3867 batch limiting is re-enabled.
3869 rcutree.rcu_idle_gp_delay= [KNL]
3870 Set wakeup interval for idle CPUs that have
3871 RCU callbacks (RCU_FAST_NO_HZ=y).
3873 rcutree.rcu_idle_lazy_gp_delay= [KNL]
3874 Set wakeup interval for idle CPUs that have
3875 only "lazy" RCU callbacks (RCU_FAST_NO_HZ=y).
3876 Lazy RCU callbacks are those which RCU can
3877 prove do nothing more than free memory.
3879 rcutree.rcu_kick_kthreads= [KNL]
3880 Cause the grace-period kthread to get an extra
3881 wake_up() if it sleeps three times longer than
3882 it should at force-quiescent-state time.
3883 This wake_up() will be accompanied by a
3884 WARN_ONCE() splat and an ftrace_dump().
3886 rcutree.sysrq_rcu= [KNL]
3887 Commandeer a sysrq key to dump out Tree RCU's
3888 rcu_node tree with an eye towards determining
3889 why a new grace period has not yet started.
3891 rcuperf.gp_async= [KNL]
3892 Measure performance of asynchronous
3893 grace-period primitives such as call_rcu().
3895 rcuperf.gp_async_max= [KNL]
3896 Specify the maximum number of outstanding
3897 callbacks per writer thread. When a writer
3898 thread exceeds this limit, it invokes the
3899 corresponding flavor of rcu_barrier() to allow
3900 previously posted callbacks to drain.
3902 rcuperf.gp_exp= [KNL]
3903 Measure performance of expedited synchronous
3904 grace-period primitives.
3906 rcuperf.holdoff= [KNL]
3907 Set test-start holdoff period. The purpose of
3908 this parameter is to delay the start of the
3909 test until boot completes in order to avoid
3912 rcuperf.nreaders= [KNL]
3913 Set number of RCU readers. The value -1 selects
3914 N, where N is the number of CPUs. A value
3915 "n" less than -1 selects N-n+1, where N is again
3916 the number of CPUs. For example, -2 selects N
3917 (the number of CPUs), -3 selects N+1, and so on.
3918 A value of "n" less than or equal to -N selects
3921 rcuperf.nwriters= [KNL]
3922 Set number of RCU writers. The values operate
3923 the same as for rcuperf.nreaders.
3924 N, where N is the number of CPUs
3926 rcuperf.perf_type= [KNL]
3927 Specify the RCU implementation to test.
3929 rcuperf.shutdown= [KNL]
3930 Shut the system down after performance tests
3931 complete. This is useful for hands-off automated
3934 rcuperf.verbose= [KNL]
3935 Enable additional printk() statements.
3937 rcuperf.writer_holdoff= [KNL]
3938 Write-side holdoff between grace periods,
3939 in microseconds. The default of zero says
3942 rcutorture.fqs_duration= [KNL]
3943 Set duration of force_quiescent_state bursts
3946 rcutorture.fqs_holdoff= [KNL]
3947 Set holdoff time within force_quiescent_state bursts
3950 rcutorture.fqs_stutter= [KNL]
3951 Set wait time between force_quiescent_state bursts
3954 rcutorture.fwd_progress= [KNL]
3955 Enable RCU grace-period forward-progress testing
3956 for the types of RCU supporting this notion.
3958 rcutorture.fwd_progress_div= [KNL]
3959 Specify the fraction of a CPU-stall-warning
3960 period to do tight-loop forward-progress testing.
3962 rcutorture.fwd_progress_holdoff= [KNL]
3963 Number of seconds to wait between successive
3964 forward-progress tests.
3966 rcutorture.fwd_progress_need_resched= [KNL]
3967 Enclose cond_resched() calls within checks for
3968 need_resched() during tight-loop forward-progress
3971 rcutorture.gp_cond= [KNL]
3972 Use conditional/asynchronous update-side
3973 primitives, if available.
3975 rcutorture.gp_exp= [KNL]
3976 Use expedited update-side primitives, if available.
3978 rcutorture.gp_normal= [KNL]
3979 Use normal (non-expedited) asynchronous
3980 update-side primitives, if available.
3982 rcutorture.gp_sync= [KNL]
3983 Use normal (non-expedited) synchronous
3984 update-side primitives, if available. If all
3985 of rcutorture.gp_cond=, rcutorture.gp_exp=,
3986 rcutorture.gp_normal=, and rcutorture.gp_sync=
3987 are zero, rcutorture acts as if is interpreted
3988 they are all non-zero.
3990 rcutorture.n_barrier_cbs= [KNL]
3991 Set callbacks/threads for rcu_barrier() testing.
3993 rcutorture.nfakewriters= [KNL]
3994 Set number of concurrent RCU writers. These just
3995 stress RCU, they don't participate in the actual
3996 test, hence the "fake".
3998 rcutorture.nreaders= [KNL]
3999 Set number of RCU readers. The value -1 selects
4000 N-1, where N is the number of CPUs. A value
4001 "n" less than -1 selects N-n-2, where N is again
4002 the number of CPUs. For example, -2 selects N
4003 (the number of CPUs), -3 selects N+1, and so on.
4005 rcutorture.object_debug= [KNL]
4006 Enable debug-object double-call_rcu() testing.
4008 rcutorture.onoff_holdoff= [KNL]
4009 Set time (s) after boot for CPU-hotplug testing.
4011 rcutorture.onoff_interval= [KNL]
4012 Set time (jiffies) between CPU-hotplug operations,
4013 or zero to disable CPU-hotplug testing.
4015 rcutorture.shuffle_interval= [KNL]
4016 Set task-shuffle interval (s). Shuffling tasks
4017 allows some CPUs to go into dyntick-idle mode
4018 during the rcutorture test.
4020 rcutorture.shutdown_secs= [KNL]
4021 Set time (s) after boot system shutdown. This
4022 is useful for hands-off automated testing.
4024 rcutorture.stall_cpu= [KNL]
4025 Duration of CPU stall (s) to test RCU CPU stall
4026 warnings, zero to disable.
4028 rcutorture.stall_cpu_holdoff= [KNL]
4029 Time to wait (s) after boot before inducing stall.
4031 rcutorture.stall_cpu_irqsoff= [KNL]
4032 Disable interrupts while stalling if set.
4034 rcutorture.stat_interval= [KNL]
4035 Time (s) between statistics printk()s.
4037 rcutorture.stutter= [KNL]
4038 Time (s) to stutter testing, for example, specifying
4039 five seconds causes the test to run for five seconds,
4040 wait for five seconds, and so on. This tests RCU's
4041 ability to transition abruptly to and from idle.
4043 rcutorture.test_boost= [KNL]
4044 Test RCU priority boosting? 0=no, 1=maybe, 2=yes.
4045 "Maybe" means test if the RCU implementation
4046 under test support RCU priority boosting.
4048 rcutorture.test_boost_duration= [KNL]
4049 Duration (s) of each individual boost test.
4051 rcutorture.test_boost_interval= [KNL]
4052 Interval (s) between each boost test.
4054 rcutorture.test_no_idle_hz= [KNL]
4055 Test RCU's dyntick-idle handling. See also the
4056 rcutorture.shuffle_interval parameter.
4058 rcutorture.torture_type= [KNL]
4059 Specify the RCU implementation to test.
4061 rcutorture.verbose= [KNL]
4062 Enable additional printk() statements.
4064 rcupdate.rcu_cpu_stall_ftrace_dump= [KNL]
4065 Dump ftrace buffer after reporting RCU CPU
4068 rcupdate.rcu_cpu_stall_suppress= [KNL]
4069 Suppress RCU CPU stall warning messages.
4071 rcupdate.rcu_cpu_stall_timeout= [KNL]
4072 Set timeout for RCU CPU stall warning messages.
4074 rcupdate.rcu_expedited= [KNL]
4075 Use expedited grace-period primitives, for
4076 example, synchronize_rcu_expedited() instead
4077 of synchronize_rcu(). This reduces latency,
4078 but can increase CPU utilization, degrade
4079 real-time latency, and degrade energy efficiency.
4080 No effect on CONFIG_TINY_RCU kernels.
4082 rcupdate.rcu_normal= [KNL]
4083 Use only normal grace-period primitives,
4084 for example, synchronize_rcu() instead of
4085 synchronize_rcu_expedited(). This improves
4086 real-time latency, CPU utilization, and
4087 energy efficiency, but can expose users to
4088 increased grace-period latency. This parameter
4089 overrides rcupdate.rcu_expedited. No effect on
4090 CONFIG_TINY_RCU kernels.
4092 rcupdate.rcu_normal_after_boot= [KNL]
4093 Once boot has completed (that is, after
4094 rcu_end_inkernel_boot() has been invoked), use
4095 only normal grace-period primitives. No effect
4096 on CONFIG_TINY_RCU kernels.
4098 rcupdate.rcu_task_stall_timeout= [KNL]
4099 Set timeout in jiffies for RCU task stall warning
4100 messages. Disable with a value less than or equal
4103 rcupdate.rcu_self_test= [KNL]
4104 Run the RCU early boot self tests
4108 Run specified binary instead of /init from the ramdisk,
4109 used for early userspace startup. See initrd.
4112 force - Override the decision by the kernel to hide the
4113 advertisement of RDRAND support (this affects
4114 certain AMD processors because of buggy BIOS
4115 support, specifically around the suspend/resume
4119 Turn on/off individual RDT features. List is:
4120 cmt, mbmtotal, mbmlocal, l3cat, l3cdp, l2cat, l2cdp,
4122 E.g. to turn on cmt and turn off mba use:
4126 Format (x86 or x86_64):
4127 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] \
4129 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \
4131 Where reboot_mode is one of warm (soft) or cold (hard) or gpio
4132 (prefix with 'panic_' to set mode for panic
4134 reboot_type is one of bios, acpi, kbd, triple, efi, or pci,
4135 reboot_force is either force or not specified,
4136 reboot_cpu is s[mp]#### with #### being the processor
4137 to be used for rebooting.
4140 [KNL, SMP] Set scheduler's default relax_domain_level.
4141 See Documentation/admin-guide/cgroup-v1/cpusets.rst.
4143 reserve= [KNL,BUGS] Force kernel to ignore I/O ports or memory
4144 Format: <base1>,<size1>[,<base2>,<size2>,...]
4145 Reserve I/O ports or memory so the kernel won't use
4146 them. If <base> is less than 0x10000, the region
4147 is assumed to be I/O ports; otherwise it is memory.
4149 reservetop= [X86-32]
4151 Reserves a hole at the top of the kernel virtual
4156 Set the amount of memory to reserve for BIOS at
4157 the bottom of the address space.
4159 reset_devices [KNL] Force drivers to reset the underlying device
4160 during initialization.
4163 Specify the partition device for software suspend
4165 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>}
4167 resume_offset= [SWSUSP]
4168 Specify the offset from the beginning of the partition
4169 given by "resume=" at which the swap header is located,
4170 in <PAGE_SIZE> units (needed only for swap files).
4171 See Documentation/power/swsusp-and-swap-files.rst
4173 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to
4174 read the resume files
4176 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up.
4177 Useful for devices that are detected asynchronously
4178 (e.g. USB and MMC devices).
4180 hibernate= [HIBERNATION]
4181 noresume Don't check if there's a hibernation image
4182 present during boot.
4183 nocompress Don't compress/decompress hibernation images.
4184 no Disable hibernation and resume.
4185 protect_image Turn on image protection during restoration
4186 (that will set all pages holding image data
4187 during restoration read-only).
4189 retain_initrd [RAM] Keep initrd memory after extraction
4191 rfkill.default_state=
4192 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm,
4193 etc. communication is blocked by default.
4196 rfkill.master_switch_mode=
4197 0 The "airplane mode" button does nothing.
4198 1 The "airplane mode" button toggles between everything
4199 blocked and the previous configuration.
4200 2 The "airplane mode" button toggles between everything
4201 blocked and everything unblocked.
4203 rhash_entries= [KNL,NET]
4204 Set number of hash buckets for route cache
4207 [KNL] Disable ring 3 MONITOR/MWAIT feature on supported
4210 ro [KNL] Mount root device read-only on boot
4213 on Mark read-only kernel memory as read-only (default).
4214 off Leave read-only kernel memory writable for debugging.
4217 Enable the uart passthrough on the designated usb port
4218 on Rockchip SoCs. When active, the signals of the
4219 debug-uart get routed to the D+ and D- pins of the usb
4220 port and the regular usb controller gets disabled.
4222 root= [KNL] Root filesystem
4223 See name_to_dev_t comment in init/do_mounts.c.
4225 rootdelay= [KNL] Delay (in seconds) to pause before attempting to
4226 mount the root filesystem
4228 rootflags= [KNL] Set root filesystem mount option string
4230 rootfstype= [KNL] Set root filesystem type
4232 rootwait [KNL] Wait (indefinitely) for root device to show up.
4233 Useful for devices that are detected asynchronously
4234 (e.g. USB and MMC devices).
4236 rproc_mem=nn[KMG][@address]
4237 [KNL,ARM,CMA] Remoteproc physical memory block.
4238 Memory area to be used by remote processor image,
4241 rw [KNL] Mount root device read-write on boot
4243 S [KNL] Run init in single mode
4245 s390_iommu= [HW,S390]
4246 Set s390 IOTLB flushing mode
4248 With strict flushing every unmap operation will result in
4249 an IOTLB flush. Default is lazy flushing before reuse,
4253 See drivers/net/irda/sa1100_ir.c.
4255 sbni= [NET] Granch SBNI12 leased line adapter
4257 sched_debug [KNL] Enables verbose scheduler debug messages.
4259 schedstats= [KNL,X86] Enable or disable scheduled statistics.
4260 Allowed values are enable and disable. This feature
4261 incurs a small amount of overhead in the scheduler
4262 but is useful for debugging and performance tuning.
4264 skew_tick= [KNL] Offset the periodic timer tick per cpu to mitigate
4265 xtime_lock contention on larger systems, and/or RCU lock
4266 contention on all systems with CONFIG_MAXSMP set.
4267 Format: { "0" | "1" }
4268 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1"
4270 Note: increases power consumption, thus should only be
4271 enabled if running jitter sensitive (HPC/RT) workloads.
4273 security= [SECURITY] Choose a legacy "major" security module to
4274 enable at boot. This has been deprecated by the
4277 selinux= [SELINUX] Disable or enable SELinux at boot time.
4278 Format: { "0" | "1" }
4279 See security/selinux/Kconfig help text.
4282 Default value is set via kernel config option.
4283 If enabled at boot time, /selinux/disable can be used
4284 later to disable prior to initial policy load.
4286 apparmor= [APPARMOR] Disable or enable AppArmor at boot time
4287 Format: { "0" | "1" }
4288 See security/apparmor/Kconfig help text
4291 Default value is set via kernel config option.
4293 serialnumber [BUGS=X86-32]
4296 Maximal number of shapers.
4304 Disable merging of slabs with similar size. May be
4305 necessary if there is some reason to distinguish
4306 allocs to different slabs, especially in hardened
4307 environments where the risk of heap overflows and
4308 layout control by attackers can usually be
4309 frustrated by disabling merging. This will reduce
4310 most of the exposure of a heap attack to a single
4311 cache (risks via metadata attacks are mostly
4312 unchanged). Debug options disable merging on their
4314 For more information see Documentation/vm/slub.rst.
4316 slab_max_order= [MM, SLAB]
4317 Determines the maximum allowed order for slabs.
4318 A high setting may cause OOMs due to memory
4319 fragmentation. Defaults to 1 for systems with
4320 more than 32MB of RAM, 0 otherwise.
4322 slub_debug[=options[,slabs]] [MM, SLUB]
4323 Enabling slub_debug allows one to determine the
4324 culprit if slab objects become corrupted. Enabling
4325 slub_debug can create guard zones around objects and
4326 may poison objects when not in use. Also tracks the
4327 last alloc / free. For more information see
4328 Documentation/vm/slub.rst.
4330 slub_memcg_sysfs= [MM, SLUB]
4331 Determines whether to enable sysfs directories for
4332 memory cgroup sub-caches. 1 to enable, 0 to disable.
4333 The default is determined by CONFIG_SLUB_MEMCG_SYSFS_ON.
4334 Enabling this can lead to a very high number of debug
4335 directories and files being created under
4338 slub_max_order= [MM, SLUB]
4339 Determines the maximum allowed order for slabs.
4340 A high setting may cause OOMs due to memory
4341 fragmentation. For more information see
4342 Documentation/vm/slub.rst.
4344 slub_min_objects= [MM, SLUB]
4345 The minimum number of objects per slab. SLUB will
4346 increase the slab order up to slub_max_order to
4347 generate a sufficiently large slab able to contain
4348 the number of objects indicated. The higher the number
4349 of objects the smaller the overhead of tracking slabs
4350 and the less frequently locks need to be acquired.
4351 For more information see Documentation/vm/slub.rst.
4353 slub_min_order= [MM, SLUB]
4354 Determines the minimum page order for slabs. Must be
4355 lower than slub_max_order.
4356 For more information see Documentation/vm/slub.rst.
4358 slub_nomerge [MM, SLUB]
4359 Same with slab_nomerge. This is supported for legacy.
4360 See slab_nomerge for more information.
4363 Format: <io1>[,<io2>[,...,<io8>]]
4365 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices
4366 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port
4367 smsc-ircc2.ircc_sir= [HW] SIR base I/O port
4368 smsc-ircc2.ircc_fir= [HW] FIR base I/O port
4369 smsc-ircc2.ircc_irq= [HW] IRQ line
4370 smsc-ircc2.ircc_dma= [HW] DMA channel
4371 smsc-ircc2.ircc_transceiver= [HW] Transceiver type:
4372 0: Toshiba Satellite 1800 (GP data pin select)
4373 1: Fast pin select (default)
4376 smt [KNL,S390] Set the maximum number of threads (logical
4377 CPUs) to use per physical CPU on systems capable of
4378 symmetric multithreading (SMT). Will be capped to the
4379 actual hardware limit.
4381 Default: -1 (no limit)
4384 [KNL] Should the soft-lockup detector generate panics.
4387 A nonzero value instructs the soft-lockup detector
4388 to panic the machine when a soft-lockup occurs. This
4389 is also controlled by CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC
4390 which is the respective build-time switch to that
4393 softlockup_all_cpu_backtrace=
4394 [KNL] Should the soft-lockup detector generate
4395 backtraces on all cpus.
4398 sonypi.*= [HW] Sony Programmable I/O Control Device driver
4399 See Documentation/admin-guide/laptops/sonypi.rst
4401 spectre_v2= [X86] Control mitigation of Spectre variant 2
4402 (indirect branch speculation) vulnerability.
4403 The default operation protects the kernel from
4406 on - unconditionally enable, implies
4408 off - unconditionally disable, implies
4410 auto - kernel detects whether your CPU model is
4413 Selecting 'on' will, and 'auto' may, choose a
4414 mitigation method at run time according to the
4415 CPU, the available microcode, the setting of the
4416 CONFIG_RETPOLINE configuration option, and the
4417 compiler with which the kernel was built.
4419 Selecting 'on' will also enable the mitigation
4420 against user space to user space task attacks.
4422 Selecting 'off' will disable both the kernel and
4423 the user space protections.
4425 Specific mitigations can also be selected manually:
4427 retpoline - replace indirect branches
4428 retpoline,generic - google's original retpoline
4429 retpoline,amd - AMD-specific minimal thunk
4431 Not specifying this option is equivalent to
4435 [X86] Control mitigation of Spectre variant 2
4436 (indirect branch speculation) vulnerability between
4439 on - Unconditionally enable mitigations. Is
4440 enforced by spectre_v2=on
4442 off - Unconditionally disable mitigations. Is
4443 enforced by spectre_v2=off
4445 prctl - Indirect branch speculation is enabled,
4446 but mitigation can be enabled via prctl
4447 per thread. The mitigation control state
4448 is inherited on fork.
4451 - Like "prctl" above, but only STIBP is
4452 controlled per thread. IBPB is issued
4453 always when switching between different user
4457 - Same as "prctl" above, but all seccomp
4458 threads will enable the mitigation unless
4459 they explicitly opt out.
4462 - Like "seccomp" above, but only STIBP is
4463 controlled per thread. IBPB is issued
4464 always when switching between different
4465 user space processes.
4467 auto - Kernel selects the mitigation depending on
4468 the available CPU features and vulnerability.
4471 If CONFIG_SECCOMP=y then "seccomp", otherwise "prctl"
4473 Not specifying this option is equivalent to
4474 spectre_v2_user=auto.
4476 spec_store_bypass_disable=
4477 [HW] Control Speculative Store Bypass (SSB) Disable mitigation
4478 (Speculative Store Bypass vulnerability)
4480 Certain CPUs are vulnerable to an exploit against a
4481 a common industry wide performance optimization known
4482 as "Speculative Store Bypass" in which recent stores
4483 to the same memory location may not be observed by
4484 later loads during speculative execution. The idea
4485 is that such stores are unlikely and that they can
4486 be detected prior to instruction retirement at the
4487 end of a particular speculation execution window.
4489 In vulnerable processors, the speculatively forwarded
4490 store can be used in a cache side channel attack, for
4491 example to read memory to which the attacker does not
4492 directly have access (e.g. inside sandboxed code).
4494 This parameter controls whether the Speculative Store
4495 Bypass optimization is used.
4497 On x86 the options are:
4499 on - Unconditionally disable Speculative Store Bypass
4500 off - Unconditionally enable Speculative Store Bypass
4501 auto - Kernel detects whether the CPU model contains an
4502 implementation of Speculative Store Bypass and
4503 picks the most appropriate mitigation. If the
4504 CPU is not vulnerable, "off" is selected. If the
4505 CPU is vulnerable the default mitigation is
4506 architecture and Kconfig dependent. See below.
4507 prctl - Control Speculative Store Bypass per thread
4508 via prctl. Speculative Store Bypass is enabled
4509 for a process by default. The state of the control
4510 is inherited on fork.
4511 seccomp - Same as "prctl" above, but all seccomp threads
4512 will disable SSB unless they explicitly opt out.
4514 Default mitigations:
4515 X86: If CONFIG_SECCOMP=y "seccomp", otherwise "prctl"
4517 On powerpc the options are:
4519 on,auto - On Power8 and Power9 insert a store-forwarding
4520 barrier on kernel entry and exit. On Power7
4521 perform a software flush on kernel entry and
4525 Not specifying this option is equivalent to
4526 spec_store_bypass_disable=auto.
4528 spia_io_base= [HW,MTD]
4533 srcutree.counter_wrap_check [KNL]
4534 Specifies how frequently to check for
4535 grace-period sequence counter wrap for the
4536 srcu_data structure's ->srcu_gp_seq_needed field.
4537 The greater the number of bits set in this kernel
4538 parameter, the less frequently counter wrap will
4539 be checked for. Note that the bottom two bits
4542 srcutree.exp_holdoff [KNL]
4543 Specifies how many nanoseconds must elapse
4544 since the end of the last SRCU grace period for
4545 a given srcu_struct until the next normal SRCU
4546 grace period will be considered for automatic
4547 expediting. Set to zero to disable automatic
4551 Speculative Store Bypass Disable control
4553 On CPUs that are vulnerable to the Speculative
4554 Store Bypass vulnerability and offer a
4555 firmware based mitigation, this parameter
4556 indicates how the mitigation should be used:
4558 force-on: Unconditionally enable mitigation for
4559 for both kernel and userspace
4560 force-off: Unconditionally disable mitigation for
4561 for both kernel and userspace
4562 kernel: Always enable mitigation in the
4563 kernel, and offer a prctl interface
4564 to allow userspace to register its
4565 interest in being mitigated too.
4567 stack_guard_gap= [MM]
4568 override the default stack gap protection. The value
4569 is in page units and it defines how many pages prior
4570 to (for stacks growing down) resp. after (for stacks
4571 growing up) the main stack are reserved for no other
4572 mapping. Default value is 256 pages.
4575 Enabled the stack tracer on boot up.
4577 stacktrace_filter=[function-list]
4578 [FTRACE] Limit the functions that the stack tracer
4579 will trace at boot up. function-list is a comma separated
4580 list of functions. This list can be changed at run
4581 time by the stack_trace_filter file in the debugfs
4582 tracing directory. Note, this enables stack tracing
4583 and the stacktrace above is not needed.
4587 Set the STI (builtin display/keyboard on the HP-PARISC
4588 machines) console (graphic card) which should be used
4589 as the initial boot-console.
4590 See also comment in drivers/video/console/sticore.c.
4593 See comment in drivers/video/console/sticore.c.
4596 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]]
4598 sunrpc.min_resvport=
4599 sunrpc.max_resvport=
4601 SunRPC servers often require that client requests
4602 originate from a privileged port (i.e. a port in the
4603 range 0 < portnr < 1024).
4604 An administrator who wishes to reserve some of these
4605 ports for other uses may adjust the range that the
4606 kernel's sunrpc client considers to be privileged
4607 using these two parameters to set the minimum and
4608 maximum port values.
4610 sunrpc.svc_rpc_per_connection_limit=
4612 Limit the number of requests that the server will
4613 process in parallel from a single connection.
4614 The default value is 0 (no limit).
4618 Control how the NFS server code allocates CPUs to
4619 service thread pools. Depending on how many NICs
4620 you have and where their interrupts are bound, this
4621 option will affect which CPUs will do NFS serving.
4622 Note: this parameter cannot be changed while the
4623 NFS server is running.
4625 auto the server chooses an appropriate mode
4626 automatically using heuristics
4627 global a single global pool contains all CPUs
4628 percpu one pool for each CPU
4629 pernode one pool for each NUMA node (equivalent
4630 to global on non-NUMA machines)
4632 sunrpc.tcp_slot_table_entries=
4633 sunrpc.udp_slot_table_entries=
4635 Sets the upper limit on the number of simultaneous
4636 RPC calls that can be sent from the client to a
4637 server. Increasing these values may allow you to
4638 improve throughput, but will also increase the
4639 amount of memory reserved for use by the client.
4641 suspend.pm_test_delay=
4643 Sets the number of seconds to remain in a suspend test
4644 mode before resuming the system (see
4645 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG
4646 is set. Default value is 5.
4649 Format: { on | off | y | n | 1 | 0 }
4650 This parameter controls use of the Protected
4651 Execution Facility on pSeries.
4654 [KNL] Enable accounting of swap in memory resource
4655 controller if no parameter or 1 is given or disable
4656 it if 0 is given (See Documentation/admin-guide/cgroup-v1/memory.rst)
4658 swiotlb= [ARM,IA-64,PPC,MIPS,X86]
4659 Format: { <int> | force | noforce }
4660 <int> -- Number of I/O TLB slabs
4661 force -- force using of bounce buffers even if they
4662 wouldn't be automatically used by the kernel
4663 noforce -- Never use bounce buffers (for debugging)
4667 sysfs.deprecated=0|1 [KNL]
4668 Enable/disable old style sysfs layout for old udev
4669 on older distributions. When this option is enabled
4670 very new udev will not work anymore. When this option
4671 is disabled (or CONFIG_SYSFS_DEPRECATED not compiled)
4672 in older udev will not work anymore.
4673 Default depends on CONFIG_SYSFS_DEPRECATED_V2 set in
4674 the kernel configuration.
4676 sysrq_always_enabled
4678 Ignore sysrq setting - this boot parameter will
4679 neutralize any effect of /proc/sys/kernel/sysrq.
4680 Useful for debugging.
4682 tcpmhash_entries= [KNL,NET]
4683 Set the number of tcp_metrics_hash slots.
4684 Default value is 8192 or 16384 depending on total
4685 ram pages. This is used to specify the TCP metrics
4686 cache size. See Documentation/networking/ip-sysctl.txt
4687 "tcp_no_metrics_save" section for more details.
4691 test_suspend= [SUSPEND][,N]
4692 Specify "mem" (for Suspend-to-RAM) or "standby" (for
4693 standby suspend) or "freeze" (for suspend type freeze)
4694 as the system sleep state during system startup with
4695 the optional capability to repeat N number of times.
4696 The system is woken from this state using a
4697 wakeup-capable RTC alarm.
4699 thash_entries= [KNL,NET]
4700 Set number of hash buckets for TCP connection
4702 thermal.act= [HW,ACPI]
4703 -1: disable all active trip points in all thermal zones
4704 <degrees C>: override all lowest active trip points
4706 thermal.crt= [HW,ACPI]
4707 -1: disable all critical trip points in all thermal zones
4708 <degrees C>: override all critical trip points
4710 thermal.nocrt= [HW,ACPI]
4711 Set to disable actions on ACPI thermal zone
4712 critical and hot trip points.
4714 thermal.off= [HW,ACPI]
4715 1: disable ACPI thermal control
4717 thermal.psv= [HW,ACPI]
4718 -1: disable all passive trip points
4719 <degrees C>: override all passive trip points to this
4722 thermal.tzp= [HW,ACPI]
4723 Specify global default ACPI thermal zone polling rate
4724 <deci-seconds>: poll all this frequency
4725 0: no polling (default)
4728 Force threading of all interrupt handlers except those
4729 marked explicitly IRQF_NO_THREAD.
4733 Specify if the kernel should make use of the cpu
4734 topology information if the hardware supports this.
4735 The scheduler will make use of this information and
4736 e.g. base its process migration decisions on it.
4739 topology_updates= [KNL, PPC, NUMA]
4741 Specify if the kernel should ignore (off)
4742 topology updates sent by the hypervisor to this
4747 tpm_suspend_pcr=[HW,TPM]
4748 Format: integer pcr id
4749 Specify that at suspend time, the tpm driver
4750 should extend the specified pcr with zeros,
4751 as a workaround for some chips which fail to
4752 flush the last written pcr on TPM_SaveState.
4753 This will guarantee that all the other pcrs
4756 trace_buf_size=nn[KMG]
4757 [FTRACE] will set tracing buffer size on each cpu.
4759 trace_event=[event-list]
4760 [FTRACE] Set and start specified trace events in order
4761 to facilitate early boot debugging. The event-list is a
4762 comma separated list of trace events to enable. See
4763 also Documentation/trace/events.rst
4765 trace_options=[option-list]
4766 [FTRACE] Enable or disable tracer options at boot.
4767 The option-list is a comma delimited list of options
4768 that can be enabled or disabled just as if you were
4769 to echo the option name into
4771 /sys/kernel/debug/tracing/trace_options
4773 For example, to enable stacktrace option (to dump the
4774 stack trace of each event), add to the command line:
4776 trace_options=stacktrace
4778 See also Documentation/trace/ftrace.rst "trace options"
4782 Have the tracepoints sent to printk as well as the
4783 tracing ring buffer. This is useful for early boot up
4784 where the system hangs or reboots and does not give the
4785 option for reading the tracing buffer or performing a
4786 ftrace_dump_on_oops.
4788 To turn off having tracepoints sent to printk,
4789 echo 0 > /proc/sys/kernel/tracepoint_printk
4790 Note, echoing 1 into this file without the
4791 tracepoint_printk kernel cmdline option has no effect.
4795 Having tracepoints sent to printk() and activating high
4796 frequency tracepoints such as irq or sched, can cause
4797 the system to live lock.
4800 [FTRACE] enable this option to disable tracing when a
4801 warning is hit. This turns off "tracing_on". Tracing can
4802 be enabled again by echoing '1' into the "tracing_on"
4803 file located in /sys/kernel/debug/tracing/
4805 This option is useful, as it disables the trace before
4806 the WARNING dump is called, which prevents the trace to
4807 be filled with content caused by the warning output.
4809 This option can also be set at run time via the sysctl
4810 option: kernel/traceoff_on_warning
4812 transparent_hugepage=
4814 Format: [always|madvise|never]
4815 Can be used to control the default behavior of the system
4816 with respect to transparent hugepages.
4817 See Documentation/admin-guide/mm/transhuge.rst
4820 tsc= Disable clocksource stability checks for TSC.
4822 [x86] reliable: mark tsc clocksource as reliable, this
4823 disables clocksource verification at runtime, as well
4824 as the stability checks done at bootup. Used to enable
4825 high-resolution timer mode on older hardware, and in
4826 virtualized environment.
4827 [x86] noirqtime: Do not use TSC to do irq accounting.
4828 Used to run time disable IRQ_TIME_ACCOUNTING on any
4829 platforms where RDTSC is slow and this accounting
4831 [x86] unstable: mark the TSC clocksource as unstable, this
4832 marks the TSC unconditionally unstable at bootup and
4833 avoids any further wobbles once the TSC watchdog notices.
4834 [x86] nowatchdog: disable clocksource watchdog. Used
4835 in situations with strict latency requirements (where
4836 interruptions from clocksource watchdog are not
4839 turbografx.map[2|3]= [HW,JOY]
4840 TurboGraFX parallel port interface
4842 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7>
4843 See also Documentation/input/devices/joystick-parport.rst
4845 udbg-immortal [PPC] When debugging early kernel crashes that
4846 happen after console_init() and before a proper
4847 console driver takes over, this boot options might
4848 help "seeing" what's going on.
4850 uhash_entries= [KNL,NET]
4851 Set number of hash buckets for UDP/UDP-Lite connections
4854 [USB] Ignore overcurrent events (default N).
4855 Some badly-designed motherboards generate lots of
4856 bogus events, for ports that aren't wired to
4857 anything. Set this parameter to avoid log spamming.
4858 Note that genuine overcurrent events won't be
4862 [X86] Cause panic on unknown NMI.
4864 usbcore.authorized_default=
4865 [USB] Default USB device authorization:
4866 (default -1 = authorized except for wireless USB,
4867 0 = not authorized, 1 = authorized, 2 = authorized
4868 if device connected to internal port)
4870 usbcore.autosuspend=
4871 [USB] The autosuspend time delay (in seconds) used
4872 for newly-detected USB devices (default 2). This
4873 is the time required before an idle device will be
4874 autosuspended. Devices for which the delay is set
4875 to a negative value won't be autosuspended at all.
4877 usbcore.usbfs_snoop=
4878 [USB] Set to log all usbfs traffic (default 0 = off).
4880 usbcore.usbfs_snoop_max=
4881 [USB] Maximum number of bytes to snoop in each URB
4884 usbcore.blinkenlights=
4885 [USB] Set to cycle leds on hubs (default 0 = off).
4887 usbcore.old_scheme_first=
4888 [USB] Start with the old device initialization
4889 scheme, applies only to low and full-speed devices
4892 usbcore.usbfs_memory_mb=
4893 [USB] Memory limit (in MB) for buffers allocated by
4894 usbfs (default = 16, 0 = max = 2047).
4896 usbcore.use_both_schemes=
4897 [USB] Try the other device initialization scheme
4898 if the first one fails (default 1 = enabled).
4900 usbcore.initial_descriptor_timeout=
4901 [USB] Specifies timeout for the initial 64-byte
4902 USB_REQ_GET_DESCRIPTOR request in milliseconds
4903 (default 5000 = 5.0 seconds).
4905 usbcore.nousb [USB] Disable the USB subsystem
4908 [USB] A list of quirk entries to augment the built-in
4909 usb core quirk list. List entries are separated by
4910 commas. Each entry has the form
4911 VendorID:ProductID:Flags. The IDs are 4-digit hex
4912 numbers and Flags is a set of letters. Each letter
4913 will change the built-in quirk; setting it if it is
4914 clear and clearing it if it is set. The letters have
4915 the following meanings:
4916 a = USB_QUIRK_STRING_FETCH_255 (string
4917 descriptors must not be fetched using
4919 b = USB_QUIRK_RESET_RESUME (device can't resume
4920 correctly so reset it instead);
4921 c = USB_QUIRK_NO_SET_INTF (device can't handle
4922 Set-Interface requests);
4923 d = USB_QUIRK_CONFIG_INTF_STRINGS (device can't
4924 handle its Configuration or Interface
4926 e = USB_QUIRK_RESET (device can't be reset
4927 (e.g morph devices), don't use reset);
4928 f = USB_QUIRK_HONOR_BNUMINTERFACES (device has
4929 more interface descriptions than the
4930 bNumInterfaces count, and can't handle
4931 talking to these interfaces);
4932 g = USB_QUIRK_DELAY_INIT (device needs a pause
4933 during initialization, after we read
4934 the device descriptor);
4935 h = USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL (For
4936 high speed and super speed interrupt
4937 endpoints, the USB 2.0 and USB 3.0 spec
4938 require the interval in microframes (1
4939 microframe = 125 microseconds) to be
4940 calculated as interval = 2 ^
4942 Devices with this quirk report their
4943 bInterval as the result of this
4944 calculation instead of the exponent
4945 variable used in the calculation);
4946 i = USB_QUIRK_DEVICE_QUALIFIER (device can't
4947 handle device_qualifier descriptor
4949 j = USB_QUIRK_IGNORE_REMOTE_WAKEUP (device
4950 generates spurious wakeup, ignore
4951 remote wakeup capability);
4952 k = USB_QUIRK_NO_LPM (device can't handle Link
4954 l = USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL
4955 (Device reports its bInterval as linear
4956 frames instead of the USB 2.0
4958 m = USB_QUIRK_DISCONNECT_SUSPEND (Device needs
4959 to be disconnected before suspend to
4960 prevent spurious wakeup);
4961 n = USB_QUIRK_DELAY_CTRL_MSG (Device needs a
4962 pause after every control message);
4963 o = USB_QUIRK_HUB_SLOW_RESET (Hub needs extra
4964 delay after resetting its port);
4965 Example: quirks=0781:5580:bk,0a5c:5834:gij
4968 [USBHID] The interval which mice are to be polled at.
4971 [USBHID] The interval which joysticks are to be polled at.
4974 [USBHID] The interval which keyboards are to be polled at.
4976 usb-storage.delay_use=
4977 [UMS] The delay in seconds before a new device is
4978 scanned for Logical Units (default 1).
4981 [UMS] A list of quirks entries to supplement or
4982 override the built-in unusual_devs list. List
4983 entries are separated by commas. Each entry has
4984 the form VID:PID:Flags where VID and PID are Vendor
4985 and Product ID values (4-digit hex numbers) and
4986 Flags is a set of characters, each corresponding
4987 to a common usb-storage quirk flag as follows:
4988 a = SANE_SENSE (collect more than 18 bytes
4990 b = BAD_SENSE (don't collect more than 18
4991 bytes of sense data);
4992 c = FIX_CAPACITY (decrease the reported
4993 device capacity by one sector);
4994 d = NO_READ_DISC_INFO (don't use
4995 READ_DISC_INFO command);
4996 e = NO_READ_CAPACITY_16 (don't use
4997 READ_CAPACITY_16 command);
4998 f = NO_REPORT_OPCODES (don't use report opcodes
5000 g = MAX_SECTORS_240 (don't transfer more than
5001 240 sectors at a time, uas only);
5002 h = CAPACITY_HEURISTICS (decrease the
5003 reported device capacity by one
5004 sector if the number is odd);
5005 i = IGNORE_DEVICE (don't bind to this
5007 j = NO_REPORT_LUNS (don't use report luns
5009 l = NOT_LOCKABLE (don't try to lock and
5010 unlock ejectable media);
5011 m = MAX_SECTORS_64 (don't transfer more
5012 than 64 sectors = 32 KB at a time);
5013 n = INITIAL_READ10 (force a retry of the
5014 initial READ(10) command);
5015 o = CAPACITY_OK (accept the capacity
5016 reported by the device);
5017 p = WRITE_CACHE (the device cache is ON
5019 r = IGNORE_RESIDUE (the device reports
5020 bogus residue values);
5021 s = SINGLE_LUN (the device has only one
5023 t = NO_ATA_1X (don't allow ATA(12) and ATA(16)
5024 commands, uas only);
5025 u = IGNORE_UAS (don't bind to the uas driver);
5026 w = NO_WP_DETECT (don't test whether the
5027 medium is write-protected).
5028 y = ALWAYS_SYNC (issue a SYNCHRONIZE_CACHE
5029 even if the device claims no cache)
5030 Example: quirks=0419:aaf5:rl,0421:0433:rc
5032 user_debug= [KNL,ARM]
5034 See arch/arm/Kconfig.debug help text.
5035 1 - undefined instruction events
5037 4 - invalid data aborts
5040 Example: user_debug=31
5043 [X86] Flags controlling user PTE allocations.
5045 nohigh = do not allocate PTE pages in
5046 HIGHMEM regardless of setting
5050 On X86_32, this is an alias for vdso32=. Otherwise:
5052 vdso=1: enable VDSO (the default)
5053 vdso=0: disable VDSO mapping
5055 vdso32= [X86] Control the 32-bit vDSO
5056 vdso32=1: enable 32-bit VDSO
5057 vdso32=0 or vdso32=2: disable 32-bit VDSO
5059 See the help text for CONFIG_COMPAT_VDSO for more
5060 details. If CONFIG_COMPAT_VDSO is set, the default is
5061 vdso32=0; otherwise, the default is vdso32=1.
5063 For compatibility with older kernels, vdso32=2 is an
5066 Try vdso32=0 if you encounter an error that says:
5067 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed!
5070 vector=percpu: enable percpu vector domain
5072 video= [FB] Frame buffer configuration
5073 See Documentation/fb/modedb.rst.
5075 video.brightness_switch_enabled= [0,1]
5076 If set to 1, on receiving an ACPI notify event
5077 generated by hotkey, video driver will adjust brightness
5078 level and then send out the event to user space through
5079 the allocated input device; If set to 0, video driver
5080 will only send out the event without touching backlight
5085 [VMMIO] Memory mapped virtio (platform) device.
5087 <size>@<baseaddr>:<irq>[:<id>]
5089 <size> := size (can use standard suffixes
5091 <baseaddr> := physical base address
5092 <irq> := interrupt number (as passed to
5094 <id> := (optional) platform device id
5096 virtio_mmio.device=1K@0x100b0000:48:7
5098 Can be used multiple times for multiple devices.
5100 vga= [BOOT,X86-32] Select a particular video mode
5101 See Documentation/x86/boot.rst and
5102 Documentation/admin-guide/svga.rst.
5103 Use vga=ask for menu.
5104 This is actually a boot loader parameter; the value is
5105 passed to the kernel using a special protocol.
5107 vm_debug[=options] [KNL] Available with CONFIG_DEBUG_VM=y.
5108 May slow down system boot speed, especially when
5109 enabled on systems with a large amount of memory.
5110 All options are enabled by default, and this
5111 interface is meant to allow for selectively
5112 enabling or disabling specific virtual memory
5115 Available options are:
5116 P Enable page structure init time poisoning
5117 - Disable all of the above options
5119 vmalloc=nn[KMG] [KNL,BOOT] Forces the vmalloc area to have an exact
5120 size of <nn>. This can be used to increase the
5121 minimum size (128MB on x86). It can also be used to
5122 decrease the size and leave more room for directly
5125 vmcp_cma=nn[MG] [KNL,S390]
5126 Sets the memory size reserved for contiguous memory
5127 allocations for the vmcp device driver.
5129 vmhalt= [KNL,S390] Perform z/VM CP command after system halt.
5132 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic.
5135 vmpoff= [KNL,S390] Perform z/VM CP command after power off.
5139 Controls the behavior of vsyscalls (i.e. calls to
5140 fixed addresses of 0xffffffffff600x00 from legacy
5141 code). Most statically-linked binaries and older
5142 versions of glibc use these calls. Because these
5143 functions are at fixed addresses, they make nice
5144 targets for exploits that can control RIP.
5146 emulate [default] Vsyscalls turn into traps and are
5147 emulated reasonably safely. The vsyscall
5150 xonly Vsyscalls turn into traps and are
5151 emulated reasonably safely. The vsyscall
5152 page is not readable.
5154 none Vsyscalls don't work at all. This makes
5155 them quite hard to use for exploits but
5156 might break your system.
5158 vt.color= [VT] Default text color.
5159 Format: 0xYX, X = foreground, Y = background.
5160 Default: 0x07 = light gray on black.
5162 vt.cur_default= [VT] Default cursor shape.
5163 Format: 0xCCBBAA, where AA, BB, and CC are the same as
5164 the parameters of the <Esc>[?A;B;Cc escape sequence;
5165 see VGA-softcursor.txt. Default: 2 = underline.
5167 vt.default_blu= [VT]
5168 Format: <blue0>,<blue1>,<blue2>,...,<blue15>
5169 Change the default blue palette of the console.
5170 This is a 16-member array composed of values
5173 vt.default_grn= [VT]
5174 Format: <green0>,<green1>,<green2>,...,<green15>
5175 Change the default green palette of the console.
5176 This is a 16-member array composed of values
5179 vt.default_red= [VT]
5180 Format: <red0>,<red1>,<red2>,...,<red15>
5181 Change the default red palette of the console.
5182 This is a 16-member array composed of values
5188 Set system-wide default UTF-8 mode for all tty's.
5189 Default is 1, i.e. UTF-8 mode is enabled for all
5190 newly opened terminals.
5192 vt.global_cursor_default=
5195 Set system-wide default for whether a cursor
5196 is shown on new VTs. Default is -1,
5197 i.e. cursors will be created by default unless
5198 overridden by individual drivers. 0 will hide
5199 cursors, 1 will display them.
5201 vt.italic= [VT] Default color for italic text; 0-15.
5204 vt.underline= [VT] Default color for underlined text; 0-15.
5207 watchdog timers [HW,WDT] For information on watchdog timers,
5208 see Documentation/watchdog/watchdog-parameters.rst
5209 or other driver-specific files in the
5210 Documentation/watchdog/ directory.
5214 Set the hard lockup detector stall duration
5215 threshold in seconds. The soft lockup detector
5216 threshold is set to twice the value. A value of 0
5217 disables both lockup detectors. Default is 10
5220 workqueue.watchdog_thresh=
5221 If CONFIG_WQ_WATCHDOG is configured, workqueue can
5222 warn stall conditions and dump internal state to
5223 help debugging. 0 disables workqueue stall
5224 detection; otherwise, it's the stall threshold
5225 duration in seconds. The default value is 30 and
5226 it can be updated at runtime by writing to the
5227 corresponding sysfs file.
5229 workqueue.disable_numa
5230 By default, all work items queued to unbound
5231 workqueues are affine to the NUMA nodes they're
5232 issued on, which results in better behavior in
5233 general. If NUMA affinity needs to be disabled for
5234 whatever reason, this option can be used. Note
5235 that this also can be controlled per-workqueue for
5236 workqueues visible under /sys/bus/workqueue/.
5238 workqueue.power_efficient
5239 Per-cpu workqueues are generally preferred because
5240 they show better performance thanks to cache
5241 locality; unfortunately, per-cpu workqueues tend to
5242 be more power hungry than unbound workqueues.
5244 Enabling this makes the per-cpu workqueues which
5245 were observed to contribute significantly to power
5246 consumption unbound, leading to measurably lower
5247 power usage at the cost of small performance
5250 The default value of this parameter is determined by
5251 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT.
5253 workqueue.debug_force_rr_cpu
5254 Workqueue used to implicitly guarantee that work
5255 items queued without explicit CPU specified are put
5256 on the local CPU. This guarantee is no longer true
5257 and while local CPU is still preferred work items
5258 may be put on foreign CPUs. This debug option
5259 forces round-robin CPU selection to flush out
5260 usages which depend on the now broken guarantee.
5261 When enabled, memory and cache locality will be
5264 x2apic_phys [X86-64,APIC] Use x2apic physical mode instead of
5265 default x2apic cluster mode on platforms
5268 x86_intel_mid_timer= [X86-32,APBT]
5269 Choose timer option for x86 Intel MID platform.
5270 Two valid options are apbt timer only and lapic timer
5271 plus one apbt timer for broadcast timer.
5272 x86_intel_mid_timer=apbt_only | lapic_and_apbt
5274 xen_512gb_limit [KNL,X86-64,XEN]
5275 Restricts the kernel running paravirtualized under Xen
5276 to use only up to 512 GB of RAM. The reason to do so is
5277 crash analysis tools and Xen tools for doing domain
5278 save/restore/migration must be enabled to handle larger
5281 xen_emul_unplug= [HW,X86,XEN]
5282 Unplug Xen emulated devices
5283 Format: [unplug0,][unplug1]
5284 ide-disks -- unplug primary master IDE devices
5285 aux-ide-disks -- unplug non-primary-master IDE devices
5286 nics -- unplug network devices
5287 all -- unplug all emulated devices (NICs and IDE disks)
5288 unnecessary -- unplugging emulated devices is
5289 unnecessary even if the host did not respond to
5291 never -- do not unplug even if version check succeeds
5293 xen_nopvspin [X86,XEN]
5294 Disables the ticketlock slowpath using Xen PV
5298 Disables the PV optimizations forcing the HVM guest to
5299 run as generic HVM guest with no PV drivers.
5300 This option is obsoleted by the "nopv" option, which
5301 has equivalent effect for XEN platform.
5303 xen_scrub_pages= [XEN]
5304 Boolean option to control scrubbing pages before giving them back
5305 to Xen, for use by other domains. Can be also changed at runtime
5306 with /sys/devices/system/xen_memory/xen_memory0/scrub_pages.
5307 Default value controlled with CONFIG_XEN_SCRUB_PAGES_DEFAULT.
5309 xen_timer_slop= [X86-64,XEN]
5310 Set the timer slop (in nanoseconds) for the virtual Xen
5311 timers (default is 100000). This adjusts the minimum
5312 delta of virtualized Xen timers, where lower values
5313 improve timer resolution at the expense of processing
5314 more timer interrupts.
5316 nopv= [X86,XEN,KVM,HYPER_V,VMWARE]
5317 Disables the PV optimizations forcing the guest to run
5318 as generic guest with no PV drivers. Currently support
5319 XEN HVM, KVM, HYPER_V and VMWARE guest.
5321 xirc2ps_cs= [NET,PCMCIA]
5323 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]]
5326 By default on POWER9 and above, the kernel will
5327 natively use the XIVE interrupt controller. This option
5328 allows the fallback firmware mode to be used:
5330 off Fallback to firmware control of XIVE interrupt
5331 controller on both pseries and powernv
5332 platforms. Only useful on POWER9 and above.
5334 xhci-hcd.quirks [USB,KNL]
5335 A hex value specifying bitmask with supplemental xhci
5336 host controller quirks. Meaning of each bit can be
5337 consulted in header drivers/usb/host/xhci.h.
5340 Format: { early | on | rw | ro | off }
5341 Controls if xmon debugger is enabled. Default is off.
5342 Passing only "xmon" is equivalent to "xmon=early".
5343 early Call xmon as early as possible on boot; xmon
5344 debugger is called from setup_arch().
5345 on xmon debugger hooks will be installed so xmon
5346 is only called on a kernel crash. Default mode,
5347 i.e. either "ro" or "rw" mode, is controlled
5348 with CONFIG_XMON_DEFAULT_RO_MODE.
5349 rw xmon debugger hooks will be installed so xmon
5350 is called only on a kernel crash, mode is write,
5351 meaning SPR registers, memory and, other data
5352 can be written using xmon commands.
5353 ro same as "rw" option above but SPR registers,
5354 memory, and other data can't be written using
5356 off xmon is disabled.