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_no_watchdog [HW,ACPI,WDT]
140 Ignore the ACPI-based watchdog interface (WDAT) and let
141 a native driver control the watchdog device instead.
143 acpi_rsdp= [ACPI,EFI,KEXEC]
144 Pass the RSDP address to the kernel, mostly used
145 on machines running EFI runtime service to boot the
146 second kernel for kdump.
148 acpi_os_name= [HW,ACPI] Tell ACPI BIOS the name of the OS
149 Format: To spoof as Windows 98: ="Microsoft Windows"
151 acpi_rev_override [ACPI] Override the _REV object to return 5 (instead
152 of 2 which is mandated by ACPI 6) as the supported ACPI
153 specification revision (when using this switch, it may
154 be necessary to carry out a cold reboot _twice_ in a
155 row to make it take effect on the platform firmware).
157 acpi_osi= [HW,ACPI] Modify list of supported OS interface strings
158 acpi_osi="string1" # add string1
159 acpi_osi="!string2" # remove string2
160 acpi_osi=!* # remove all strings
161 acpi_osi=! # disable all built-in OS vendor
163 acpi_osi=!! # enable all built-in OS vendor
165 acpi_osi= # disable all strings
167 'acpi_osi=!' can be used in combination with single or
168 multiple 'acpi_osi="string1"' to support specific OS
169 vendor string(s). Note that such command can only
170 affect the default state of the OS vendor strings, thus
171 it cannot affect the default state of the feature group
172 strings and the current state of the OS vendor strings,
173 specifying it multiple times through kernel command line
174 is meaningless. This command is useful when one do not
175 care about the state of the feature group strings which
176 should be controlled by the OSPM.
178 1. 'acpi_osi=! acpi_osi="Windows 2000"' is equivalent
179 to 'acpi_osi="Windows 2000" acpi_osi=!', they all
180 can make '_OSI("Windows 2000")' TRUE.
182 'acpi_osi=' cannot be used in combination with other
183 'acpi_osi=' command lines, the _OSI method will not
184 exist in the ACPI namespace. NOTE that such command can
185 only affect the _OSI support state, thus specifying it
186 multiple times through kernel command line is also
189 1. 'acpi_osi=' can make 'CondRefOf(_OSI, Local1)'
192 'acpi_osi=!*' can be used in combination with single or
193 multiple 'acpi_osi="string1"' to support specific
194 string(s). Note that such command can affect the
195 current state of both the OS vendor strings and the
196 feature group strings, thus specifying it multiple times
197 through kernel command line is meaningful. But it may
198 still not able to affect the final state of a string if
199 there are quirks related to this string. This command
200 is useful when one want to control the state of the
201 feature group strings to debug BIOS issues related to
204 1. 'acpi_osi="Module Device" acpi_osi=!*' can make
205 '_OSI("Module Device")' FALSE.
206 2. 'acpi_osi=!* acpi_osi="Module Device"' can make
207 '_OSI("Module Device")' TRUE.
208 3. 'acpi_osi=! acpi_osi=!* acpi_osi="Windows 2000"' is
210 'acpi_osi=!* acpi_osi=! acpi_osi="Windows 2000"'
212 'acpi_osi=!* acpi_osi="Windows 2000" acpi_osi=!',
213 they all will make '_OSI("Windows 2000")' TRUE.
216 Override the pmtimer bug detection: force the kernel
217 to assume that this machine's pmtimer latches its value
218 and always returns good values.
220 acpi_sci= [HW,ACPI] ACPI System Control Interrupt trigger mode
221 Format: { level | edge | high | low }
223 acpi_skip_timer_override [HW,ACPI]
224 Recognize and ignore IRQ0/pin2 Interrupt Override.
225 For broken nForce2 BIOS resulting in XT-PIC timer.
227 acpi_sleep= [HW,ACPI] Sleep options
228 Format: { s3_bios, s3_mode, s3_beep, s4_nohwsig,
229 old_ordering, nonvs, sci_force_enable, nobl }
230 See Documentation/power/video.rst for information on
232 s3_beep is for debugging; it makes the PC's speaker beep
233 as soon as the kernel's real-mode entry point is called.
234 s4_nohwsig prevents ACPI hardware signature from being
235 used during resume from hibernation.
236 old_ordering causes the ACPI 1.0 ordering of the _PTS
237 control method, with respect to putting devices into
238 low power states, to be enforced (the ACPI 2.0 ordering
239 of _PTS is used by default).
240 nonvs prevents the kernel from saving/restoring the
241 ACPI NVS memory during suspend/hibernation and resume.
242 sci_force_enable causes the kernel to set SCI_EN directly
243 on resume from S1/S3 (which is against the ACPI spec,
244 but some broken systems don't work without it).
245 nobl causes the internal blacklist of systems known to
246 behave incorrectly in some ways with respect to system
247 suspend and resume to be ignored (use wisely).
249 acpi_use_timer_override [HW,ACPI]
250 Use timer override. For some broken Nvidia NF5 boards
251 that require a timer override, but don't have HPET
253 add_efi_memmap [EFI; X86] Include EFI memory map in
254 kernel's map of available physical RAM.
257 { off | try_unsupported }
258 off: disable AGP support
259 try_unsupported: try to drive unsupported chipsets
260 (may crash computer or cause data corruption)
263 See Documentation/sound/alsa-configuration.rst
266 Allow the default userspace alignment fault handler
267 behaviour to be specified. Bit 0 enables warnings,
268 bit 1 enables fixups, and bit 2 sends a segfault.
270 align_va_addr= [X86-64]
271 Align virtual addresses by clearing slice [14:12] when
272 allocating a VMA at process creation time. This option
273 gives you up to 3% performance improvement on AMD F15h
274 machines (where it is enabled by default) for a
275 CPU-intensive style benchmark, and it can vary highly in
276 a microbenchmark depending on workload and compiler.
278 32: only for 32-bit processes
279 64: only for 64-bit processes
280 on: enable for both 32- and 64-bit processes
281 off: disable for both 32- and 64-bit processes
283 alloc_snapshot [FTRACE]
284 Allocate the ftrace snapshot buffer on boot up when the
285 main buffer is allocated. This is handy if debugging
286 and you need to use tracing_snapshot() on boot up, and
287 do not want to use tracing_snapshot_alloc() as it needs
288 to be done where GFP_KERNEL allocations are allowed.
290 amd_iommu= [HW,X86-64]
291 Pass parameters to the AMD IOMMU driver in the system.
293 fullflush - enable flushing of IO/TLB entries when
294 they are unmapped. Otherwise they are
295 flushed before they will be reused, which
297 off - do not initialize any AMD IOMMU found in
299 force_isolation - Force device isolation for all
300 devices. The IOMMU driver is not
301 allowed anymore to lift isolation
302 requirements as needed. This option
303 does not override iommu=pt
305 amd_iommu_dump= [HW,X86-64]
306 Enable AMD IOMMU driver option to dump the ACPI table
307 for AMD IOMMU. With this option enabled, AMD IOMMU
308 driver will print ACPI tables for AMD IOMMU during
309 IOMMU initialization.
311 amd_iommu_intr= [HW,X86-64]
312 Specifies one of the following AMD IOMMU interrupt
314 legacy - Use legacy interrupt remapping mode.
315 vapic - Use virtual APIC mode, which allows IOMMU
316 to inject interrupts directly into guest.
317 This mode requires kvm-amd.avic=1.
318 (Default when IOMMU HW support is present.)
320 amijoy.map= [HW,JOY] Amiga joystick support
321 Map of devices attached to JOY0DAT and JOY1DAT
323 See also Documentation/input/joydev/joystick.rst
325 analog.map= [HW,JOY] Analog joystick and gamepad support
326 Specifies type or capabilities of an analog joystick
327 connected to one of 16 gameports
328 Format: <type1>,<type2>,..<type16>
331 Power management functions (SPARCstation-4/5 + deriv.)
333 Disable APC CPU standby support. SPARCstation-Fox does
334 not play well with APC CPU idle - disable it if you have
335 APC and your system crashes randomly.
337 apic= [APIC,X86] Advanced Programmable Interrupt Controller
338 Change the output verbosity while booting
339 Format: { quiet (default) | verbose | debug }
340 Change the amount of debugging information output
341 when initialising the APIC and IO-APIC components.
342 For X86-32, this can also be used to specify an APIC
344 Format: apic=driver_name
345 Examples: apic=bigsmp
347 apic_extnmi= [APIC,X86] External NMI delivery setting
348 Format: { bsp (default) | all | none }
349 bsp: External NMI is delivered only to CPU 0
350 all: External NMIs are broadcast to all CPUs as a
352 none: External NMI is masked for all CPUs. This is
353 useful so that a dump capture kernel won't be
357 See Documentation/networking/ipv6.txt.
359 show_lapic= [APIC,X86] Advanced Programmable Interrupt Controller
360 Limit apic dumping. The parameter defines the maximal
361 number of local apics being dumped. Also it is possible
362 to set it to "all" by meaning -- no limit here.
363 Format: { 1 (default) | 2 | ... | all }.
364 The parameter valid if only apic=debug or
365 apic=verbose is specified.
366 Example: apic=debug show_lapic=all
368 apm= [APM] Advanced Power Management
369 See header of arch/x86/kernel/apm_32.c.
371 arcrimi= [HW,NET] ARCnet - "RIM I" (entirely mem-mapped) cards
372 Format: <io>,<irq>,<nodeID>
376 atarimouse= [HW,MOUSE] Atari Mouse
378 atkbd.extra= [HW] Enable extra LEDs and keys on IBM RapidAccess,
379 EzKey and similar keyboards
381 atkbd.reset= [HW] Reset keyboard during initialization
383 atkbd.set= [HW] Select keyboard code set
384 Format: <int> (2 = AT (default), 3 = PS/2)
386 atkbd.scroll= [HW] Enable scroll wheel on MS Office and similar
389 atkbd.softraw= [HW] Choose between synthetic and real raw mode
390 Format: <bool> (0 = real, 1 = synthetic (default))
392 atkbd.softrepeat= [HW]
393 Use software keyboard repeat
395 audit= [KNL] Enable the audit sub-system
396 Format: { "0" | "1" | "off" | "on" }
397 0 | off - kernel audit is disabled and can not be
398 enabled until the next reboot
399 unset - kernel audit is initialized but disabled and
400 will be fully enabled by the userspace auditd.
401 1 | on - kernel audit is initialized and partially
402 enabled, storing at most audit_backlog_limit
403 messages in RAM until it is fully enabled by the
407 audit_backlog_limit= [KNL] Set the audit queue size limit.
408 Format: <int> (must be >=0)
411 bau= [X86_UV] Enable the BAU on SGI UV. The default
412 behavior is to disable the BAU (i.e. bau=0).
413 Format: { "0" | "1" }
416 unset - Disable the BAU.
418 baycom_epp= [HW,AX25]
421 baycom_par= [HW,AX25] BayCom Parallel Port AX.25 Modem
423 See header of drivers/net/hamradio/baycom_par.c.
425 baycom_ser_fdx= [HW,AX25]
426 BayCom Serial Port AX.25 Modem (Full Duplex Mode)
427 Format: <io>,<irq>,<mode>[,<baud>]
428 See header of drivers/net/hamradio/baycom_ser_fdx.c.
430 baycom_ser_hdx= [HW,AX25]
431 BayCom Serial Port AX.25 Modem (Half Duplex Mode)
432 Format: <io>,<irq>,<mode>
433 See header of drivers/net/hamradio/baycom_ser_hdx.c.
435 blkdevparts= Manual partition parsing of block device(s) for
436 embedded devices based on command line input.
437 See Documentation/block/cmdline-partition.rst
439 boot_delay= Milliseconds to delay each printk during boot.
440 Values larger than 10 seconds (10000) are changed to
445 Extended command line options can be added to an initrd
446 and this will cause the kernel to look for it.
448 See Documentation/admin-guide/bootconfig.rst
451 Disable BERT OS support on buggy BIOSes.
453 bttv.card= [HW,V4L] bttv (bt848 + bt878 based grabber cards)
454 bttv.radio= Most important insmod options are available as
456 bttv.pll= See Documentation/media/v4l-drivers/bttv.rst
459 bulk_remove=off [PPC] This parameter disables the use of the pSeries
460 firmware feature for flushing multiple hpte entries
463 c101= [NET] Moxa C101 synchronous serial card
465 cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection.
466 Sometimes CPU hardware bugs make them report the cache
467 size incorrectly. The kernel will attempt work arounds
468 to fix known problems, but for some CPUs it is not
469 possible to determine what the correct size should be.
470 This option provides an override for these situations.
473 [NET] Specifies amount of time (in seconds) that
474 the kernel should wait for a network carrier. By default
475 it waits 120 seconds.
477 ca_keys= [KEYS] This parameter identifies a specific key(s) on
478 the system trusted keyring to be used for certificate
480 format: { id:<keyid> | builtin }
482 cca= [MIPS] Override the kernel pages' cache coherency
483 algorithm. Accepted values range from 0 to 7
484 inclusive. See arch/mips/include/asm/pgtable-bits.h
485 for platform specific values (SB1, Loongson3 and
488 ccw_timeout_log [S390]
489 See Documentation/s390/common_io.rst for details.
491 cgroup_disable= [KNL] Disable a particular controller
492 Format: {name of the controller(s) to disable}
493 The effects of cgroup_disable=foo are:
494 - foo isn't auto-mounted if you mount all cgroups in
496 - foo isn't visible as an individually mountable
498 {Currently only "memory" controller deal with this and
499 cut the overhead, others just disable the usage. So
500 only cgroup_disable=memory is actually worthy}
502 cgroup_no_v1= [KNL] Disable cgroup controllers and named hierarchies in v1
503 Format: { { controller | "all" | "named" }
504 [,{ controller | "all" | "named" }...] }
505 Like cgroup_disable, but only applies to cgroup v1;
506 the blacklisted controllers remain available in cgroup2.
507 "all" blacklists all controllers and "named" disables
508 named mounts. Specifying both "all" and "named" disables
511 cgroup.memory= [KNL] Pass options to the cgroup memory controller.
513 nosocket -- Disable socket memory accounting.
514 nokmem -- Disable kernel memory accounting.
516 checkreqprot [SELINUX] Set initial checkreqprot flag value.
517 Format: { "0" | "1" }
518 See security/selinux/Kconfig help text.
519 0 -- check protection applied by kernel (includes
520 any implied execute protection).
521 1 -- check protection requested by application.
522 Default value is set via a kernel config option.
523 Value can be changed at runtime via
524 /sys/fs/selinux/checkreqprot.
527 See Documentation/s390/common_io.rst for details.
530 Prevents the clock framework from automatically gating
531 clocks that have not been explicitly enabled by a Linux
532 device driver but are enabled in hardware at reset or
533 by the bootloader/firmware. Note that this does not
534 force such clocks to be always-on nor does it reserve
535 those clocks in any way. This parameter is useful for
536 debug and development, but should not be needed on a
537 platform with proper driver support. For more
538 information, see Documentation/driver-api/clk.rst.
540 clock= [BUGS=X86-32, HW] gettimeofday clocksource override.
542 Forces specified clocksource (if available) to be used
543 when calculating gettimeofday(). If specified
544 clocksource is not available, it defaults to PIT.
545 Format: { pit | tsc | cyclone | pmtmr }
547 clocksource= Override the default clocksource
549 Override the default clocksource and use the clocksource
550 with the name specified.
551 Some clocksource names to choose from, depending on
553 [all] jiffies (this is the base, fallback clocksource)
555 [ARM] imx_timer1,OSTS,netx_timer,mpu_timer2,
556 pxa_timer,timer3,32k_counter,timer0_1
557 [X86-32] pit,hpet,tsc;
558 scx200_hrt on Geode; cyclone on IBM x440
566 clocksource.arm_arch_timer.evtstrm=
569 Enable/disable the eventstream feature of the ARM
570 architected timer so that code using WFE-based polling
571 loops can be debugged more effectively on production
574 clearcpuid=BITNUM [X86]
575 Disable CPUID feature X for the kernel. See
576 arch/x86/include/asm/cpufeatures.h for the valid bit
577 numbers. Note the Linux specific bits are not necessarily
578 stable over kernel options, but the vendor specific
580 Also note that user programs calling CPUID directly
581 or using the feature without checking anything
582 will still see it. This just prevents it from
583 being used by the kernel or shown in /proc/cpuinfo.
584 Also note the kernel might malfunction if you disable
587 cma=nn[MG]@[start[MG][-end[MG]]]
589 Sets the size of kernel global memory area for
590 contiguous memory allocations and optionally the
591 placement constraint by the physical address range of
592 memory allocations. A value of 0 disables CMA
593 altogether. For more information, see
594 include/linux/dma-contiguous.h
596 cmo_free_hint= [PPC] Format: { yes | no }
597 Specify whether pages are marked as being inactive
598 when they are freed. This is used in CMO environments
599 to determine OS memory pressure for page stealing by
603 coherent_pool=nn[KMG] [ARM,KNL]
604 Sets the size of memory pool for coherent, atomic dma
605 allocations, by default set to 256K.
607 com20020= [HW,NET] ARCnet - COM20020 chipset
609 <io>[,<irq>[,<nodeID>[,<backplane>[,<ckp>[,<timeout>]]]]]
611 com90io= [HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers)
615 ARCnet - COM90xx chipset (memory-mapped buffers)
616 Format: <io>[,<irq>[,<memstart>]]
618 condev= [HW,S390] console device
621 console= [KNL] Output console device and options.
623 tty<n> Use the virtual console device <n>.
627 Use the specified serial port. The options are of
628 the form "bbbbpnf", where "bbbb" is the baud rate,
629 "p" is parity ("n", "o", or "e"), "n" is number of
630 bits, and "f" is flow control ("r" for RTS or
631 omit it). Default is "9600n8".
633 See Documentation/admin-guide/serial-console.rst for more
635 Documentation/networking/netconsole.txt for an
638 uart[8250],io,<addr>[,options]
639 uart[8250],mmio,<addr>[,options]
640 uart[8250],mmio16,<addr>[,options]
641 uart[8250],mmio32,<addr>[,options]
642 uart[8250],0x<addr>[,options]
643 Start an early, polled-mode console on the 8250/16550
644 UART at the specified I/O port or MMIO address,
645 switching to the matching ttyS device later.
646 MMIO inter-register address stride is either 8-bit
647 (mmio), 16-bit (mmio16), or 32-bit (mmio32).
648 If none of [io|mmio|mmio16|mmio32], <addr> is assumed
649 to be equivalent to 'mmio'. 'options' are specified in
650 the same format described for ttyS above; if unspecified,
651 the h/w is not re-initialized.
653 hvc<n> Use the hypervisor console device <n>. This is for
654 both Xen and PowerPC hypervisors.
656 If the device connected to the port is not a TTY but a braille
657 device, prepend "brl," before the device type, for instance
659 For now, only VisioBraille is supported.
662 [KNL] Change console messages format
664 By default we print messages on consoles in
665 "[time stamp] text\n" format (time stamp may not be
666 printed, depending on CONFIG_PRINTK_TIME or
667 `printk_time' param).
669 Switch to syslog format: "<%u>[time stamp] text\n"
670 IOW, each message will have a facility and loglevel
671 prefix. The format is similar to one used by syslog()
672 syscall, or to executing "dmesg -S --raw" or to reading
675 consoleblank= [KNL] The console blank (screen saver) timeout in
676 seconds. A value of 0 disables the blank timer.
680 [KNL] Change the default value for
681 /proc/<pid>/coredump_filter.
682 See also Documentation/filesystems/proc.txt.
684 coresight_cpu_debug.enable
687 Enable/disable the CPU sampling based debugging.
688 0: default value, disable debugging
689 1: enable debugging at boot time
691 cpuidle.off=1 [CPU_IDLE]
692 disable the cpuidle sub-system
695 [CPU_IDLE] Name of the cpuidle governor to use.
697 cpufreq.off=1 [CPU_FREQ]
698 disable the cpufreq sub-system
701 [X86] Delay for N microsec between assert and de-assert
702 of APIC INIT to start processors. This delay occurs
703 on every CPU online, such as boot, and resume from suspend.
706 cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver
708 <first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>]
710 crashkernel=size[KMG][@offset[KMG]]
711 [KNL] Using kexec, Linux can switch to a 'crash kernel'
712 upon panic. This parameter reserves the physical
713 memory region [offset, offset + size] for that kernel
714 image. If '@offset' is omitted, then a suitable offset
715 is selected automatically.
716 [KNL, x86_64] select a region under 4G first, and
717 fall back to reserve region above 4G when '@offset'
718 hasn't been specified.
719 See Documentation/admin-guide/kdump/kdump.rst for further details.
721 crashkernel=range1:size1[,range2:size2,...][@offset]
722 [KNL] Same as above, but depends on the memory
723 in the running system. The syntax of range is
724 start-[end] where start and end are both
725 a memory unit (amount[KMG]). See also
726 Documentation/admin-guide/kdump/kdump.rst for an example.
728 crashkernel=size[KMG],high
729 [KNL, x86_64] range could be above 4G. Allow kernel
730 to allocate physical memory region from top, so could
731 be above 4G if system have more than 4G ram installed.
732 Otherwise memory region will be allocated below 4G, if
734 It will be ignored if crashkernel=X is specified.
735 crashkernel=size[KMG],low
736 [KNL, x86_64] range under 4G. When crashkernel=X,high
737 is passed, kernel could allocate physical memory region
738 above 4G, that cause second kernel crash on system
739 that require some amount of low memory, e.g. swiotlb
740 requires at least 64M+32K low memory, also enough extra
741 low memory is needed to make sure DMA buffers for 32-bit
742 devices won't run out. Kernel would try to allocate at
743 at least 256M below 4G automatically.
744 This one let user to specify own low range under 4G
745 for second kernel instead.
746 0: to disable low allocation.
747 It will be ignored when crashkernel=X,high is not used
748 or memory reserved is below 4G.
751 [KNL] Disable crypto self-tests
756 cs89x0_media= [HW,NET]
757 Format: { rj45 | aui | bnc }
760 See header of drivers/s390/block/dasd_devmap.c.
762 db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port
763 (one device per port)
764 Format: <port#>,<type>
765 See also Documentation/input/devices/joystick-parport.rst
767 ddebug_query= [KNL,DYNAMIC_DEBUG] Enable debug messages at early boot
769 Documentation/admin-guide/dynamic-debug-howto.rst for
770 details. Deprecated, see dyndbg.
772 debug [KNL] Enable kernel debugging (events log level).
775 [KNL] Enable printing [hashed] pointers early in the
776 boot sequence. If enabled, we use a weak hash instead
777 of siphash to hash pointers. Use this option if you are
778 seeing instances of '(___ptrval___)') and need to see a
779 value (hashed pointer) instead. Cryptographically
780 insecure, please do not use on production kernels.
783 [KNL] verbose self-tests
785 Print debugging info while doing the locking API
787 We default to 0 (no extra messages), setting it to
788 1 will print _a lot_ more information - normally
789 only useful to kernel developers.
791 debug_objects [KNL] Enable object debugging
794 [KNL] Disable object debugging
796 debug_guardpage_minorder=
797 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this
798 parameter allows control of the order of pages that will
799 be intentionally kept free (and hence protected) by the
800 buddy allocator. Bigger value increase the probability
801 of catching random memory corruption, but reduce the
802 amount of memory for normal system use. The maximum
803 possible value is MAX_ORDER/2. Setting this parameter
804 to 1 or 2 should be enough to identify most random
805 memory corruption problems caused by bugs in kernel or
806 driver code when a CPU writes to (or reads from) a
807 random memory location. Note that there exists a class
808 of memory corruptions problems caused by buggy H/W or
809 F/W or by drivers badly programing DMA (basically when
810 memory is written at bus level and the CPU MMU is
811 bypassed) which are not detectable by
812 CONFIG_DEBUG_PAGEALLOC, hence this option will not help
813 tracking down these problems.
816 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this parameter
817 enables the feature at boot time. By default, it is
818 disabled and the system will work mostly the same as a
819 kernel built without CONFIG_DEBUG_PAGEALLOC.
820 Note: to get most of debug_pagealloc error reports, it's
821 useful to also enable the page_owner functionality.
822 on: enable the feature
824 debugpat [X86] Enable PAT debugging
826 decnet.addr= [HW,NET]
827 Format: <area>[,<node>]
828 See also Documentation/networking/decnet.txt.
831 [same as hugepagesz=] The size of the default
832 HugeTLB page size. This is the size represented by
833 the legacy /proc/ hugepages APIs, used for SHM, and
834 default size when mounting hugetlbfs filesystems.
835 Defaults to the default architecture's huge page size
838 deferred_probe_timeout=
839 [KNL] Debugging option to set a timeout in seconds for
840 deferred probe to give up waiting on dependencies to
841 probe. Only specific dependencies (subsystems or
842 drivers) that have opted in will be ignored. A timeout of 0
843 will timeout at the end of initcalls. This option will also
844 dump out devices still on the deferred probe list after
848 Format: { on | off | def_only | inf_only | always }
849 on: s390 zlib hardware support for compression on
850 level 1 and decompression (default)
851 off: No s390 zlib hardware support
852 def_only: s390 zlib hardware support for deflate
853 only (compression on level 1)
854 inf_only: s390 zlib hardware support for inflate
856 always: Same as 'on' but ignores the selected compression
857 level always using hardware support (used for debugging)
860 Set number of hash buckets for dentry cache.
862 disable_1tb_segments [PPC]
863 Disables the use of 1TB hash page table segments. This
864 causes the kernel to fall back to 256MB segments which
865 can be useful when debugging issues that require an SLB
869 See Documentation/networking/ipv6.txt.
872 [KNL] Under CONFIG_HARDENED_USERCOPY, whether
873 hardening is enabled for this boot. Hardened
874 usercopy checking is used to protect the kernel
875 from reading or writing beyond known memory
876 allocation boundaries as a proactive defense
877 against bounds-checking flaws in the kernel's
878 copy_to_user()/copy_from_user() interface.
879 on Perform hardened usercopy checks (default).
880 off Disable hardened usercopy checks.
883 Disable RADIX MMU mode on POWER9
886 Disable TLBIE instruction. Currently does not work
887 with KVM, with HASH MMU, or with coherent accelerators.
889 disable_cpu_apicid= [X86,APIC,SMP]
891 The number of initial APIC ID for the
892 corresponding CPU to be disabled at boot,
893 mostly used for the kdump 2nd kernel to
894 disable BSP to wake up multiple CPUs without
895 causing system reset or hang due to sending
898 perf_v4_pmi= [X86,INTEL]
900 Disable Intel PMU counter freezing feature.
901 The feature only exists starting from
902 Arch Perfmon v4 (Skylake and newer).
904 disable_ddw [PPC/PSERIES]
905 Disable Dynamic DMA Window support. Use this if
906 to workaround buggy firmware.
909 See Documentation/networking/ipv6.txt.
911 disable_mtrr_cleanup [X86]
912 The kernel tries to adjust MTRR layout from continuous
913 to discrete, to make X server driver able to add WB
914 entry later. This parameter disables that.
916 disable_mtrr_trim [X86, Intel and AMD only]
917 By default the kernel will trim any uncacheable
918 memory out of your available memory pool based on
919 MTRR settings. This parameter disables that behavior,
920 possibly causing your machine to run very slowly.
922 disable_timer_pin_1 [X86]
923 Disable PIN 1 of APIC timer
924 Can be useful to work around chipset bugs.
926 dis_ucode_ldr [X86] Disable the microcode loader.
928 dma_debug=off If the kernel is compiled with DMA_API_DEBUG support,
929 this option disables the debugging code at boot.
931 dma_debug_entries=<number>
932 This option allows to tune the number of preallocated
933 entries for DMA-API debugging code. One entry is
934 required per DMA-API allocation. Use this if the
935 DMA-API debugging code disables itself because the
936 architectural default is too low.
938 dma_debug_driver=<driver_name>
939 With this option the DMA-API debugging driver
940 filter feature can be enabled at boot time. Just
941 pass the driver to filter for as the parameter.
942 The filter can be disabled or changed to another
943 driver later using sysfs.
945 driver_async_probe= [KNL]
946 List of driver names to be probed asynchronously.
947 Format: <driver_name1>,<driver_name2>...
949 drm.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>]
950 Broken monitors, graphic adapters, KVMs and EDIDless
951 panels may send no or incorrect EDID data sets.
952 This parameter allows to specify an EDID data sets
953 in the /lib/firmware directory that are used instead.
954 Generic built-in EDID data sets are used, if one of
955 edid/1024x768.bin, edid/1280x1024.bin,
956 edid/1680x1050.bin, or edid/1920x1080.bin is given
957 and no file with the same name exists. Details and
958 instructions how to build your own EDID data are
959 available in Documentation/driver-api/edid.rst. An EDID
960 data set will only be used for a particular connector,
961 if its name and a colon are prepended to the EDID
962 name. Each connector may use a unique EDID data
963 set by separating the files with a comma. An EDID
964 data set with no connector name will be used for
965 any connectors not explicitly specified.
970 Format: {"off" | "known"}
971 Control how the dt_cpu_ftrs device-tree binding is
972 used for CPU feature discovery and setup (if it
974 off: Do not use it, fall back to legacy cpu table.
975 known: Do not pass through unknown features to guests
976 or userspace, only those that the kernel is aware of.
978 dump_apple_properties [X86]
979 Dump name and content of EFI device properties on
980 x86 Macs. Useful for driver authors to determine
981 what data is available or for reverse-engineering.
983 dyndbg[="val"] [KNL,DYNAMIC_DEBUG]
984 module.dyndbg[="val"]
985 Enable debug messages at boot time. See
986 Documentation/admin-guide/dynamic-debug-howto.rst
989 nompx [X86] Disables Intel Memory Protection Extensions.
990 See Documentation/x86/intel_mpx.rst for more
991 information about the feature.
993 nopku [X86] Disable Memory Protection Keys CPU feature found
996 module.async_probe [KNL]
997 Enable asynchronous probe on this module.
999 early_ioremap_debug [KNL]
1000 Enable debug messages in early_ioremap support. This
1001 is useful for tracking down temporary early mappings
1002 which are not unmapped.
1004 earlycon= [KNL] Output early console device and options.
1006 When used with no options, the early console is
1007 determined by stdout-path property in device tree's
1008 chosen node or the ACPI SPCR table if supported by
1011 cdns,<addr>[,options]
1012 Start an early, polled-mode console on a Cadence
1013 (xuartps) serial port at the specified address. Only
1014 supported option is baud rate. If baud rate is not
1015 specified, the serial port must already be setup and
1018 uart[8250],io,<addr>[,options]
1019 uart[8250],mmio,<addr>[,options]
1020 uart[8250],mmio32,<addr>[,options]
1021 uart[8250],mmio32be,<addr>[,options]
1022 uart[8250],0x<addr>[,options]
1023 Start an early, polled-mode console on the 8250/16550
1024 UART at the specified I/O port or MMIO address.
1025 MMIO inter-register address stride is either 8-bit
1026 (mmio) or 32-bit (mmio32 or mmio32be).
1027 If none of [io|mmio|mmio32|mmio32be], <addr> is assumed
1028 to be equivalent to 'mmio'. 'options' are specified
1029 in the same format described for "console=ttyS<n>"; if
1030 unspecified, the h/w is not initialized.
1034 Start an early, polled-mode console on a pl011 serial
1035 port at the specified address. The pl011 serial port
1036 must already be setup and configured. Options are not
1037 yet supported. If 'mmio32' is specified, then only
1038 the driver will use only 32-bit accessors to read/write
1039 the device registers.
1042 Start an early, polled-mode console on a meson serial
1043 port at the specified address. The serial port must
1044 already be setup and configured. Options are not yet
1048 Start an early, polled-mode console on an msm serial
1049 port at the specified address. The serial port
1050 must already be setup and configured. Options are not
1053 msm_serial_dm,<addr>
1054 Start an early, polled-mode console on an msm serial
1055 dm port at the specified address. The serial port
1056 must already be setup and configured. Options are not
1060 Start an early, polled-mode console on a serial port
1061 of an Actions Semi SoC, such as S500 or S900, at the
1062 specified address. The serial port must already be
1063 setup and configured. Options are not yet supported.
1066 Start an early, polled-mode console on a serial port
1067 of an RDA Micro SoC, such as RDA8810PL, at the
1068 specified address. The serial port must already be
1069 setup and configured. Options are not yet supported.
1072 Use RISC-V SBI (Supervisor Binary Interface) for early
1075 smh Use ARM semihosting calls for early console.
1083 Use early console provided by serial driver available
1084 on Samsung SoCs, requires selecting proper type and
1085 a correct base address of the selected UART port. The
1086 serial port must already be setup and configured.
1087 Options are not yet supported.
1090 Start an early, polled-mode console on a lantiq serial
1091 (lqasc) port at the specified address. The serial port
1092 must already be setup and configured. Options are not
1097 Use early console provided by Freescale LP UART driver
1098 found on Freescale Vybrid and QorIQ LS1021A processors.
1099 A valid base address must be provided, and the serial
1100 port must already be setup and configured.
1103 Start an early, polled-mode console on the
1104 Armada 3700 serial port at the specified
1105 address. The serial port must already be setup
1106 and configured. Options are not yet supported.
1109 Start an early, polled-mode console on a Qualcomm
1110 Generic Interface (GENI) based serial port at the
1111 specified address. The serial port must already be
1112 setup and configured. Options are not yet supported.
1115 Start an early, unaccelerated console on the EFI
1116 memory mapped framebuffer (if available). On cache
1117 coherent non-x86 systems that use system memory for
1118 the framebuffer, pass the 'ram' option so that it is
1119 mapped with the correct attributes.
1122 Use early console provided by Freescale LINFlexD UART
1123 serial driver for NXP S32V234 SoCs. A valid base
1124 address must be provided, and the serial port must
1125 already be setup and configured.
1127 earlyprintk= [X86,SH,ARM,M68k,S390]
1131 earlyprintk=serial[,ttySn[,baudrate]]
1132 earlyprintk=serial[,0x...[,baudrate]]
1133 earlyprintk=ttySn[,baudrate]
1134 earlyprintk=dbgp[debugController#]
1135 earlyprintk=pciserial[,force],bus:device.function[,baudrate]
1136 earlyprintk=xdbc[xhciController#]
1138 earlyprintk is useful when the kernel crashes before
1139 the normal console is initialized. It is not enabled by
1140 default because it has some cosmetic problems.
1142 Append ",keep" to not disable it when the real console
1145 Only one of vga, efi, serial, or usb debug port can
1148 Currently only ttyS0 and ttyS1 may be specified by
1149 name. Other I/O ports may be explicitly specified
1150 on some architectures (x86 and arm at least) by
1151 replacing ttySn with an I/O port address, like this:
1152 earlyprintk=serial,0x1008,115200
1153 You can find the port for a given device in
1154 /proc/tty/driver/serial:
1155 2: uart:ST16650V2 port:00001008 irq:18 ...
1157 Interaction with the standard serial driver is not
1160 The VGA and EFI output is eventually overwritten by
1163 The xen output can only be used by Xen PV guests.
1165 The sclp output can only be used on s390.
1167 The optional "force" to "pciserial" enables use of a
1168 PCI device even when its classcode is not of the
1171 edac_report= [HW,EDAC] Control how to report EDAC event
1172 Format: {"on" | "off" | "force"}
1173 on: enable EDAC to report H/W event. May be overridden
1174 by other higher priority error reporting module.
1175 off: disable H/W event reporting through EDAC.
1176 force: enforce the use of EDAC to report H/W event.
1179 ekgdboc= [X86,KGDB] Allow early kernel console debugging
1182 This is designed to be used in conjunction with
1183 the boot argument: earlyprintk=vga
1186 Format: {"off" | "on" | "skip[mbr]"}
1189 Format: { "old_map", "nochunk", "noruntime", "debug",
1190 "nosoftreserve", "disable_early_pci_dma",
1191 "no_disable_early_pci_dma" }
1192 old_map [X86-64]: switch to the old ioremap-based EFI
1193 runtime services mapping. [Needs CONFIG_X86_UV=y]
1194 nochunk: disable reading files in "chunks" in the EFI
1195 boot stub, as chunking can cause problems with some
1196 firmware implementations.
1197 noruntime : disable EFI runtime services support
1198 debug: enable misc debug output
1199 nosoftreserve: The EFI_MEMORY_SP (Specific Purpose)
1200 attribute may cause the kernel to reserve the
1201 memory range for a memory mapping driver to
1202 claim. Specify efi=nosoftreserve to disable this
1203 reservation and treat the memory by its base type
1204 (i.e. EFI_CONVENTIONAL_MEMORY / "System RAM").
1205 disable_early_pci_dma: Disable the busmaster bit on all
1206 PCI bridges while in the EFI boot stub
1207 no_disable_early_pci_dma: Leave the busmaster bit set
1208 on all PCI bridges while in the EFI boot stub
1210 efi_no_storage_paranoia [EFI; X86]
1211 Using this parameter you can use more than 50% of
1212 your efi variable storage. Use this parameter only if
1213 you are really sure that your UEFI does sane gc and
1214 fulfills the spec otherwise your board may brick.
1216 efi_fake_mem= nn[KMG]@ss[KMG]:aa[,nn[KMG]@ss[KMG]:aa,..] [EFI; X86]
1217 Add arbitrary attribute to specific memory range by
1218 updating original EFI memory map.
1219 Region of memory which aa attribute is added to is
1222 If efi_fake_mem=2G@4G:0x10000,2G@0x10a0000000:0x10000
1223 is specified, EFI_MEMORY_MORE_RELIABLE(0x10000)
1224 attribute is added to range 0x100000000-0x180000000 and
1225 0x10a0000000-0x1120000000.
1227 If efi_fake_mem=8G@9G:0x40000 is specified, the
1228 EFI_MEMORY_SP(0x40000) attribute is added to
1229 range 0x240000000-0x43fffffff.
1231 Using this parameter you can do debugging of EFI memmap
1232 related features. For example, you can do debugging of
1233 Address Range Mirroring feature even if your box
1234 doesn't support it, or mark specific memory as
1237 efivar_ssdt= [EFI; X86] Name of an EFI variable that contains an SSDT
1238 that is to be dynamically loaded by Linux. If there are
1239 multiple variables with the same name but with different
1240 vendor GUIDs, all of them will be loaded. See
1241 Documentation/admin-guide/acpi/ssdt-overlays.rst for details.
1244 eisa_irq_edge= [PARISC,HW]
1245 See header of drivers/parisc/eisa.c.
1248 See comment before function elanfreq_setup() in
1249 arch/x86/kernel/cpu/cpufreq/elanfreq.c.
1251 elfcorehdr=[size[KMG]@]offset[KMG] [IA64,PPC,SH,X86,S390]
1252 Specifies physical address of start of kernel core
1253 image elf header and optionally the size. Generally
1254 kexec loader will pass this option to capture kernel.
1255 See Documentation/admin-guide/kdump/kdump.rst for details.
1257 enable_mtrr_cleanup [X86]
1258 The kernel tries to adjust MTRR layout from continuous
1259 to discrete, to make X server driver able to add WB
1260 entry later. This parameter enables that.
1262 enable_timer_pin_1 [X86]
1263 Enable PIN 1 of APIC timer
1264 Can be useful to work around chipset bugs
1265 (in particular on some ATI chipsets).
1266 The kernel tries to set a reasonable default.
1268 enforcing [SELINUX] Set initial enforcing status.
1270 See security/selinux/Kconfig help text.
1271 0 -- permissive (log only, no denials).
1272 1 -- enforcing (deny and log).
1274 Value can be changed at runtime via
1275 /sys/fs/selinux/enforce.
1278 Disable Error Record Serialization Table (ERST)
1281 ether= [HW,NET] Ethernet cards parameters
1282 This option is obsoleted by the "netdev=" option, which
1283 has equivalent usage. See its documentation for details.
1287 Permit 'security.evm' to be updated regardless of
1288 current integrity status.
1292 fail_make_request=[KNL]
1293 General fault injection mechanism.
1294 Format: <interval>,<probability>,<space>,<times>
1295 See also Documentation/fault-injection/.
1298 See Documentation/admin-guide/blockdev/floppy.rst.
1300 force_pal_cache_flush
1301 [IA-64] Avoid check_sal_cache_flush which may hang on
1302 buggy SAL_CACHE_FLUSH implementations. Using this
1303 parameter will force ia64_sal_cache_flush to call
1304 ia64_pal_cache_flush instead of SAL_CACHE_FLUSH.
1307 Forcefully enable Physical Address Extension (PAE).
1308 Many Pentium M systems disable PAE but may have a
1309 functionally usable PAE implementation.
1310 Warning: use of this parameter will taint the kernel
1311 and may cause unknown problems.
1314 [FTRACE] will set and start the specified tracer
1315 as early as possible in order to facilitate early
1318 ftrace_dump_on_oops[=orig_cpu]
1319 [FTRACE] will dump the trace buffers on oops.
1320 If no parameter is passed, ftrace will dump
1321 buffers of all CPUs, but if you pass orig_cpu, it will
1322 dump only the buffer of the CPU that triggered the
1325 ftrace_filter=[function-list]
1326 [FTRACE] Limit the functions traced by the function
1327 tracer at boot up. function-list is a comma separated
1328 list of functions. This list can be changed at run
1329 time by the set_ftrace_filter file in the debugfs
1332 ftrace_notrace=[function-list]
1333 [FTRACE] Do not trace the functions specified in
1334 function-list. This list can be changed at run time
1335 by the set_ftrace_notrace file in the debugfs
1338 ftrace_graph_filter=[function-list]
1339 [FTRACE] Limit the top level callers functions traced
1340 by the function graph tracer at boot up.
1341 function-list is a comma separated list of functions
1342 that can be changed at run time by the
1343 set_graph_function file in the debugfs tracing directory.
1345 ftrace_graph_notrace=[function-list]
1346 [FTRACE] Do not trace from the functions specified in
1347 function-list. This list is a comma separated list of
1348 functions that can be changed at run time by the
1349 set_graph_notrace file in the debugfs tracing directory.
1351 ftrace_graph_max_depth=<uint>
1352 [FTRACE] Used with the function graph tracer. This is
1353 the max depth it will trace into a function. This value
1354 can be changed at run time by the max_graph_depth file
1355 in the tracefs tracing directory. default: 0 (no limit)
1358 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad
1359 support via parallel port (up to 5 devices per port)
1360 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5>
1361 See also Documentation/input/devices/joystick-parport.rst
1365 gart_fix_e820= [X86_64] disable the fix e820 for K8 GART
1369 gcov_persist= [GCOV] When non-zero (default), profiling data for
1370 kernel modules is saved and remains accessible via
1371 debugfs, even when the module is unloaded/reloaded.
1372 When zero, profiling data is discarded and associated
1373 debugfs files are removed at module unload time.
1375 goldfish [X86] Enable the goldfish android emulator platform.
1376 Don't use this when you are not running on the
1379 gpt [EFI] Forces disk with valid GPT signature but
1380 invalid Protective MBR to be treated as GPT. If the
1381 primary GPT is corrupted, it enables the backup/alternate
1382 GPT to be used instead.
1384 grcan.enable0= [HW] Configuration of physical interface 0. Determines
1385 the "Enable 0" bit of the configuration register.
1388 grcan.enable1= [HW] Configuration of physical interface 1. Determines
1389 the "Enable 0" bit of the configuration register.
1392 grcan.select= [HW] Select which physical interface to use.
1395 grcan.txsize= [HW] Sets the size of the tx buffer.
1396 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0.
1398 grcan.rxsize= [HW] Sets the size of the rx buffer.
1399 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0.
1402 gpio-mockup.gpio_mockup_ranges
1403 [HW] Sets the ranges of gpiochip of for this device.
1404 Format: <start1>,<end1>,<start2>,<end2>...
1406 hardlockup_all_cpu_backtrace=
1407 [KNL] Should the hard-lockup detector generate
1408 backtraces on all cpus.
1411 hashdist= [KNL,NUMA] Large hashes allocated during boot
1412 are distributed across NUMA nodes. Defaults on
1413 for 64-bit NUMA, off otherwise.
1414 Format: 0 | 1 (for off | on)
1416 hcl= [IA-64] SGI's Hardware Graph compatibility layer
1418 hd= [EIDE] (E)IDE hard drive subsystem geometry
1419 Format: <cyl>,<head>,<sect>
1422 Disable Hardware Error Source Table (HEST) support;
1423 corresponding firmware-first mode error processing
1424 logic will be disabled.
1426 highmem=nn[KMG] [KNL,BOOT] forces the highmem zone to have an exact
1427 size of <nn>. This works even on boxes that have no
1428 highmem otherwise. This also works to reduce highmem
1429 size on bigger boxes.
1431 highres= [KNL] Enable/disable high resolution timer mode.
1432 Valid parameters: "on", "off"
1437 hpet= [X86-32,HPET] option to control HPET usage
1438 Format: { enable (default) | disable | force |
1440 disable: disable HPET and use PIT instead
1441 force: allow force enabled of undocumented chips (ICH4,
1443 verbose: show contents of HPET registers during setup
1445 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET
1446 registers. Default set by CONFIG_HPET_MMAP_DEFAULT.
1448 hugepages= [HW,X86-32,IA-64] HugeTLB pages to allocate at boot.
1449 hugepagesz= [HW,IA-64,PPC,X86-64] The size of the HugeTLB pages.
1450 On x86-64 and powerpc, this option can be specified
1451 multiple times interleaved with hugepages= to reserve
1452 huge pages of different sizes. Valid pages sizes on
1453 x86-64 are 2M (when the CPU supports "pse") and 1G
1454 (when the CPU supports the "pdpe1gb" cpuinfo flag).
1457 [KNL] Should the hung task detector generate panics.
1460 A nonzero value instructs the kernel to panic when a
1461 hung task is detected. The default value is controlled
1462 by the CONFIG_BOOTPARAM_HUNG_TASK_PANIC build-time
1463 option. The value selected by this boot parameter can
1464 be changed later by the kernel.hung_task_panic sysctl.
1466 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC)
1467 terminal devices. Valid values: 0..8
1468 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs.
1469 If specified, z/VM IUCV HVC accepts connections
1470 from listed z/VM user IDs only.
1472 hv_nopvspin [X86,HYPER_V] Disables the paravirt spinlock optimizations
1473 which allow the hypervisor to 'idle' the
1474 guest on lock contention.
1477 Do not unregister boot console at start. This is only
1478 useful for debugging when something happens in the window
1479 between unregistering the boot console and initializing
1482 i2c_bus= [HW] Override the default board specific I2C bus speed
1483 or register an additional I2C bus that is not
1484 registered from board initialization code.
1488 i8042.debug [HW] Toggle i8042 debug mode
1489 i8042.unmask_kbd_data
1490 [HW] Enable printing of interrupt data from the KBD port
1491 (disabled by default, and as a pre-condition
1492 requires that i8042.debug=1 be enabled)
1493 i8042.direct [HW] Put keyboard port into non-translated mode
1494 i8042.dumbkbd [HW] Pretend that controller can only read data from
1495 keyboard and cannot control its state
1496 (Don't attempt to blink the leds)
1497 i8042.noaux [HW] Don't check for auxiliary (== mouse) port
1498 i8042.nokbd [HW] Don't check/create keyboard port
1499 i8042.noloop [HW] Disable the AUX Loopback command while probing
1501 i8042.nomux [HW] Don't check presence of an active multiplexing
1503 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX
1505 i8042.notimeout [HW] Ignore timeout condition signalled by controller
1506 i8042.reset [HW] Reset the controller during init, cleanup and
1507 suspend-to-ram transitions, only during s2r
1508 transitions, or never reset
1509 Format: { 1 | Y | y | 0 | N | n }
1510 1, Y, y: always reset controller
1511 0, N, n: don't ever reset controller
1512 Default: only on s2r transitions on x86; most other
1513 architectures force reset to be always executed
1514 i8042.unlock [HW] Unlock (ignore) the keylock
1515 i8042.kbdreset [HW] Reset device connected to KBD port
1519 i8k.ignore_dmi [HW] Continue probing hardware even if DMI data
1520 indicates that the driver is running on unsupported
1522 i8k.force [HW] Activate i8k driver even if SMM BIOS signature
1523 does not match list of supported models.
1525 [HW] Report power status in /proc/i8k
1526 (disabled by default)
1527 i8k.restricted [HW] Allow controlling fans only if SYS_ADMIN
1530 i915.invert_brightness=
1531 [DRM] Invert the sense of the variable that is used to
1532 set the brightness of the panel backlight. Normally a
1533 brightness value of 0 indicates backlight switched off,
1534 and the maximum of the brightness value sets the backlight
1535 to maximum brightness. If this parameter is set to 0
1536 (default) and the machine requires it, or this parameter
1537 is set to 1, a brightness value of 0 sets the backlight
1538 to maximum brightness, and the maximum of the brightness
1539 value switches the backlight off.
1540 -1 -- never invert brightness
1541 0 -- machine default
1542 1 -- force brightness inversion
1545 Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]]
1547 ide-core.nodma= [HW] (E)IDE subsystem
1548 Format: =0.0 to prevent dma on hda, =0.1 hdb =1.0 hdc
1549 .vlb_clock .pci_clock .noflush .nohpa .noprobe .nowerr
1550 .cdrom .chs .ignore_cable are additional options
1551 See Documentation/ide/ide.rst.
1553 ide-generic.probe-mask= [HW] (E)IDE subsystem
1555 Probe mask for legacy ISA IDE ports. Depending on
1556 platform up to 6 ports are supported, enabled by
1557 setting corresponding bits in the mask to 1. The
1558 default value is 0x0, which has a special meaning.
1559 On systems that have PCI, it triggers scanning the
1560 PCI bus for the first and the second port, which
1561 are then probed. On systems without PCI the value
1562 of 0x0 enables probing the two first ports as if it
1565 ide-pci-generic.all-generic-ide [HW] (E)IDE subsystem
1566 Claim all unknown PCI IDE storage controllers.
1569 Format: idle=poll, idle=halt, idle=nomwait
1570 Poll forces a polling idle loop that can slightly
1571 improve the performance of waking up a idle CPU, but
1572 will use a lot of power and make the system run hot.
1574 idle=halt: Halt is forced to be used for CPU idle.
1575 In such case C2/C3 won't be used again.
1576 idle=nomwait: Disable mwait for CPU C-states
1578 ieee754= [MIPS] Select IEEE Std 754 conformance mode
1579 Format: { strict | legacy | 2008 | relaxed }
1582 Choose which programs will be accepted for execution
1583 based on the IEEE 754 NaN encoding(s) supported by
1584 the FPU and the NaN encoding requested with the value
1585 of an ELF file header flag individually set by each
1586 binary. Hardware implementations are permitted to
1587 support either or both of the legacy and the 2008 NaN
1590 Available settings are as follows:
1591 strict accept binaries that request a NaN encoding
1592 supported by the FPU
1593 legacy only accept legacy-NaN binaries, if supported
1595 2008 only accept 2008-NaN binaries, if supported
1597 relaxed accept any binaries regardless of whether
1598 supported by the FPU
1600 The FPU emulator is always able to support both NaN
1601 encodings, so if no FPU hardware is present or it has
1602 been disabled with 'nofpu', then the settings of
1603 'legacy' and '2008' strap the emulator accordingly,
1604 'relaxed' straps the emulator for both legacy-NaN and
1605 2008-NaN, whereas 'strict' enables legacy-NaN only on
1606 legacy processors and both NaN encodings on MIPS32 or
1609 The setting for ABS.fmt/NEG.fmt instruction execution
1610 mode generally follows that for the NaN encoding,
1611 except where unsupported by hardware.
1613 ignore_loglevel [KNL]
1614 Ignore loglevel setting - this will print /all/
1615 kernel messages to the console. Useful for debugging.
1616 We also add it as printk module parameter, so users
1617 could change it dynamically, usually by
1618 /sys/module/printk/parameters/ignore_loglevel.
1621 Ignore RLIMIT_DATA setting for data mappings,
1622 print warning at first misuse. Can be changed via
1623 /sys/module/kernel/parameters/ignore_rlimit_data.
1625 ihash_entries= [KNL]
1626 Set number of hash buckets for inode cache.
1628 ima_appraise= [IMA] appraise integrity measurements
1629 Format: { "off" | "enforce" | "fix" | "log" }
1632 ima_appraise_tcb [IMA] Deprecated. Use ima_policy= instead.
1633 The builtin appraise policy appraises all files
1636 ima_canonical_fmt [IMA]
1637 Use the canonical format for the binary runtime
1638 measurements, instead of host native format.
1641 Format: { md5 | sha1 | rmd160 | sha256 | sha384
1645 The list of supported hash algorithms is defined
1646 in crypto/hash_info.h.
1649 The builtin policies to load during IMA setup.
1650 Format: "tcb | appraise_tcb | secure_boot |
1653 The "tcb" policy measures all programs exec'd, files
1654 mmap'd for exec, and all files opened with the read
1655 mode bit set by either the effective uid (euid=0) or
1658 The "appraise_tcb" policy appraises the integrity of
1659 all files owned by root.
1661 The "secure_boot" policy appraises the integrity
1662 of files (eg. kexec kernel image, kernel modules,
1663 firmware, policy, etc) based on file signatures.
1665 The "fail_securely" policy forces file signature
1666 verification failure also on privileged mounted
1667 filesystems with the SB_I_UNVERIFIABLE_SIGNATURE
1670 ima_tcb [IMA] Deprecated. Use ima_policy= instead.
1671 Load a policy which meets the needs of the Trusted
1672 Computing Base. This means IMA will measure all
1673 programs exec'd, files mmap'd for exec, and all files
1674 opened for read by uid=0.
1677 Select one of defined IMA measurements template formats.
1678 Formats: { "ima" | "ima-ng" | "ima-sig" }
1682 [IMA] Define a custom template format.
1683 Format: { "field1|...|fieldN" }
1685 ima.ahash_minsize= [IMA] Minimum file size for asynchronous hash usage
1686 Format: <min_file_size>
1687 Set the minimal file size for using asynchronous hash.
1688 If left unspecified, ahash usage is disabled.
1690 ahash performance varies for different data sizes on
1691 different crypto accelerators. This option can be used
1692 to achieve the best performance for a particular HW.
1694 ima.ahash_bufsize= [IMA] Asynchronous hash buffer size
1696 Set hashing buffer size. Default: 4k.
1698 ahash performance varies for different chunk sizes on
1699 different crypto accelerators. This option can be used
1700 to achieve best performance for particular HW.
1704 Run specified binary instead of /sbin/init as init
1707 initcall_debug [KNL] Trace initcalls as they are executed. Useful
1708 for working out where the kernel is dying during
1711 initcall_blacklist= [KNL] Do not execute a comma-separated list of
1712 initcall functions. Useful for debugging built-in
1713 modules and initcalls.
1715 initrd= [BOOT] Specify the location of the initial ramdisk
1717 init_on_alloc= [MM] Fill newly allocated pages and heap objects with
1720 Default set by CONFIG_INIT_ON_ALLOC_DEFAULT_ON.
1722 init_on_free= [MM] Fill freed pages and heap objects with zeroes.
1724 Default set by CONFIG_INIT_ON_FREE_DEFAULT_ON.
1726 init_pkru= [x86] Specify the default memory protection keys rights
1727 register contents for all processes. 0x55555554 by
1728 default (disallow access to all but pkey 0). Can
1729 override in debugfs after boot.
1731 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver
1734 int_pln_enable [x86] Enable power limit notification interrupt
1736 integrity_audit=[IMA]
1737 Format: { "0" | "1" }
1738 0 -- basic integrity auditing messages. (Default)
1739 1 -- additional integrity auditing messages.
1741 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option
1743 Enable intel iommu driver.
1745 Disable intel iommu driver.
1746 igfx_off [Default Off]
1747 By default, gfx is mapped as normal device. If a gfx
1748 device has a dedicated DMAR unit, the DMAR unit is
1749 bypassed by not enabling DMAR with this option. In
1750 this case, gfx device will use physical address for
1753 With this option iommu will not optimize to look
1754 for io virtual address below 32-bit forcing dual
1755 address cycle on pci bus for cards supporting greater
1756 than 32-bit addressing. The default is to look
1757 for translation below 32-bit and if not available
1758 then look in the higher range.
1759 strict [Default Off]
1760 With this option on every unmap_single operation will
1761 result in a hardware IOTLB flush operation as opposed
1762 to batching them for performance.
1763 sp_off [Default Off]
1764 By default, super page will be supported if Intel IOMMU
1765 has the capability. With this option, super page will
1768 By default, scalable mode will be disabled even if the
1769 hardware advertises that it has support for the scalable
1770 mode translation. With this option set, scalable mode
1771 will be used on hardware which claims to support it.
1772 tboot_noforce [Default Off]
1773 Do not force the Intel IOMMU enabled under tboot.
1774 By default, tboot will force Intel IOMMU on, which
1775 could harm performance of some high-throughput
1776 devices like 40GBit network cards, even if identity
1778 Note that using this option lowers the security
1779 provided by tboot because it makes the system
1780 vulnerable to DMA attacks.
1781 nobounce [Default off]
1782 Disable bounce buffer for unstrusted devices such as
1783 the Thunderbolt devices. This will treat the untrusted
1784 devices as the trusted ones, hence might expose security
1785 risks of DMA attacks.
1787 intel_idle.max_cstate= [KNL,HW,ACPI,X86]
1788 0 disables intel_idle and fall back on acpi_idle.
1789 1 to 9 specify maximum depth of C-state.
1793 Do not enable intel_pstate as the default
1794 scaling driver for the supported processors
1796 Use intel_pstate as a scaling driver, but configure it
1797 to work with generic cpufreq governors (instead of
1798 enabling its internal governor). This mode cannot be
1799 used along with the hardware-managed P-states (HWP)
1802 Enable intel_pstate on systems that prohibit it by default
1803 in favor of acpi-cpufreq. Forcing the intel_pstate driver
1804 instead of acpi-cpufreq may disable platform features, such
1805 as thermal controls and power capping, that rely on ACPI
1806 P-States information being indicated to OSPM and therefore
1807 should be used with caution. This option does not work with
1808 processors that aren't supported by the intel_pstate driver
1809 or on platforms that use pcc-cpufreq instead of acpi-cpufreq.
1811 Do not enable hardware P state control (HWP)
1814 Only load intel_pstate on systems which support
1815 hardware P state control (HWP) if available.
1817 Enforce ACPI _PPC performance limits. If the Fixed ACPI
1818 Description Table, specifies preferred power management
1819 profile as "Enterprise Server" or "Performance Server",
1820 then this feature is turned on by default.
1822 Allow per-logical-CPU P-State performance control limits using
1823 cpufreq sysfs interface
1825 intremap= [X86-64, Intel-IOMMU]
1826 on enable Interrupt Remapping (default)
1827 off disable Interrupt Remapping
1828 nosid disable Source ID checking
1830 BIOS x2APIC opt-out request will be ignored
1831 nopost disable Interrupt Posting
1833 iomem= Disable strict checking of access to MMIO memory
1834 strict regions from userspace.
1849 nobypass [PPC/POWERNV]
1850 Disable IOMMU bypass, using IOMMU for PCI devices.
1852 iommu.strict= [ARM64] Configure TLB invalidation behaviour
1853 Format: { "0" | "1" }
1855 Request that DMA unmap operations use deferred
1856 invalidation of hardware TLBs, for increased
1857 throughput at the cost of reduced device isolation.
1858 Will fall back to strict mode if not supported by
1859 the relevant IOMMU driver.
1860 1 - Strict mode (default).
1861 DMA unmap operations invalidate IOMMU hardware TLBs
1865 [ARM64, X86] Configure DMA to bypass the IOMMU by default.
1866 Format: { "0" | "1" }
1867 0 - Use IOMMU translation for DMA.
1868 1 - Bypass the IOMMU for DMA.
1869 unset - Use value of CONFIG_IOMMU_DEFAULT_PASSTHROUGH.
1871 io7= [HW] IO7 for Marvel based alpha systems
1872 See comment before marvel_specify_io7 in
1873 arch/alpha/kernel/core_marvel.c.
1875 io_delay= [X86] I/O delay method
1877 Standard port 0x80 based delay
1879 Alternate port 0xed based delay (needed on some systems)
1881 Simple two microseconds delay
1886 See Documentation/filesystems/nfs/nfsroot.txt.
1888 ipcmni_extend [KNL] Extend the maximum number of unique System V
1889 IPC identifiers from 32,768 to 16,777,216.
1891 irqaffinity= [SMP] Set the default irq affinity mask
1892 The argument is a cpu list, as described above.
1894 irqchip.gicv2_force_probe=
1897 Force the kernel to look for the second 4kB page
1898 of a GICv2 controller even if the memory range
1899 exposed by the device tree is too small.
1901 irqchip.gicv3_nolpi=
1903 Force the kernel to ignore the availability of
1904 LPIs (and by consequence ITSs). Intended for system
1905 that use the kernel as a bootloader, and thus want
1906 to let secondary kernels in charge of setting up
1909 irqchip.gicv3_pseudo_nmi= [ARM64]
1910 Enables support for pseudo-NMIs in the kernel. This
1911 requires the kernel to be built with
1912 CONFIG_ARM64_PSEUDO_NMI.
1915 When an interrupt is not handled search all handlers
1916 for it. Intended to get systems with badly broken
1920 When an interrupt is not handled search all handlers
1921 for it. Also check all handlers each timer
1922 interrupt. Intended to get systems with badly broken
1926 Format: <RDP>,<reset>,<pci_scan>,<verbosity>
1928 isolcpus= [KNL,SMP,ISOL] Isolate a given set of CPUs from disturbance.
1929 [Deprecated - use cpusets instead]
1930 Format: [flag-list,]<cpu-list>
1932 Specify one or more CPUs to isolate from disturbances
1933 specified in the flag list (default: domain):
1936 Disable the tick when a single task runs.
1938 A residual 1Hz tick is offloaded to workqueues, which you
1939 need to affine to housekeeping through the global
1940 workqueue's affinity configured via the
1941 /sys/devices/virtual/workqueue/cpumask sysfs file, or
1942 by using the 'domain' flag described below.
1944 NOTE: by default the global workqueue runs on all CPUs,
1945 so to protect individual CPUs the 'cpumask' file has to
1946 be configured manually after bootup.
1949 Isolate from the general SMP balancing and scheduling
1950 algorithms. Note that performing domain isolation this way
1951 is irreversible: it's not possible to bring back a CPU to
1952 the domains once isolated through isolcpus. It's strongly
1953 advised to use cpusets instead to disable scheduler load
1954 balancing through the "cpuset.sched_load_balance" file.
1955 It offers a much more flexible interface where CPUs can
1956 move in and out of an isolated set anytime.
1958 You can move a process onto or off an "isolated" CPU via
1959 the CPU affinity syscalls or cpuset.
1960 <cpu number> begins at 0 and the maximum value is
1961 "number of CPUs in system - 1".
1965 Isolate from being targeted by managed interrupts
1966 which have an interrupt mask containing isolated
1967 CPUs. The affinity of managed interrupts is
1968 handled by the kernel and cannot be changed via
1969 the /proc/irq/* interfaces.
1971 This isolation is best effort and only effective
1972 if the automatically assigned interrupt mask of a
1973 device queue contains isolated and housekeeping
1974 CPUs. If housekeeping CPUs are online then such
1975 interrupts are directed to the housekeeping CPU
1976 so that IO submitted on the housekeeping CPU
1977 cannot disturb the isolated CPU.
1979 If a queue's affinity mask contains only isolated
1980 CPUs then this parameter has no effect on the
1981 interrupt routing decision, though interrupts are
1982 only delivered when tasks running on those
1983 isolated CPUs submit IO. IO submitted on
1984 housekeeping CPUs has no influence on those
1987 The format of <cpu-list> is described above.
1991 ivrs_ioapic [HW,X86_64]
1992 Provide an override to the IOAPIC-ID<->DEVICE-ID
1993 mapping provided in the IVRS ACPI table. For
1994 example, to map IOAPIC-ID decimal 10 to
1995 PCI device 00:14.0 write the parameter as:
1996 ivrs_ioapic[10]=00:14.0
1998 ivrs_hpet [HW,X86_64]
1999 Provide an override to the HPET-ID<->DEVICE-ID
2000 mapping provided in the IVRS ACPI table. For
2001 example, to map HPET-ID decimal 0 to
2002 PCI device 00:14.0 write the parameter as:
2003 ivrs_hpet[0]=00:14.0
2005 ivrs_acpihid [HW,X86_64]
2006 Provide an override to the ACPI-HID:UID<->DEVICE-ID
2007 mapping provided in the IVRS ACPI table. For
2008 example, to map UART-HID:UID AMD0020:0 to
2009 PCI device 00:14.5 write the parameter as:
2010 ivrs_acpihid[00:14.5]=AMD0020:0
2012 js= [HW,JOY] Analog joystick
2013 See Documentation/input/joydev/joystick.rst.
2016 When CONFIG_RANDOMIZE_BASE is set, this disables
2017 kernel and module base offset ASLR (Address Space
2018 Layout Randomization).
2021 [KNL] Enforce KASAN (Kernel Address Sanitizer) to print
2022 report on every invalid memory access. Without this
2023 parameter KASAN will print report only for the first
2028 kernelcore= [KNL,X86,IA-64,PPC]
2029 Format: nn[KMGTPE] | nn% | "mirror"
2030 This parameter specifies the amount of memory usable by
2031 the kernel for non-movable allocations. The requested
2032 amount is spread evenly throughout all nodes in the
2033 system as ZONE_NORMAL. The remaining memory is used for
2034 movable memory in its own zone, ZONE_MOVABLE. In the
2035 event, a node is too small to have both ZONE_NORMAL and
2036 ZONE_MOVABLE, kernelcore memory will take priority and
2037 other nodes will have a larger ZONE_MOVABLE.
2039 ZONE_MOVABLE is used for the allocation of pages that
2040 may be reclaimed or moved by the page migration
2041 subsystem. Note that allocations like PTEs-from-HighMem
2042 still use the HighMem zone if it exists, and the Normal
2043 zone if it does not.
2045 It is possible to specify the exact amount of memory in
2046 the form of "nn[KMGTPE]", a percentage of total system
2047 memory in the form of "nn%", or "mirror". If "mirror"
2048 option is specified, mirrored (reliable) memory is used
2049 for non-movable allocations and remaining memory is used
2050 for Movable pages. "nn[KMGTPE]", "nn%", and "mirror"
2051 are exclusive, so you cannot specify multiple forms.
2053 kgdbdbgp= [KGDB,HW] kgdb over EHCI usb debug port.
2054 Format: <Controller#>[,poll interval]
2055 The controller # is the number of the ehci usb debug
2056 port as it is probed via PCI. The poll interval is
2057 optional and is the number seconds in between
2058 each poll cycle to the debug port in case you need
2059 the functionality for interrupting the kernel with
2060 gdb or control-c on the dbgp connection. When
2061 not using this parameter you use sysrq-g to break into
2062 the kernel debugger.
2064 kgdboc= [KGDB,HW] kgdb over consoles.
2065 Requires a tty driver that supports console polling,
2066 or a supported polling keyboard driver (non-usb).
2067 Serial only format: <serial_device>[,baud]
2068 keyboard only format: kbd
2069 keyboard and serial format: kbd,<serial_device>[,baud]
2070 Optional Kernel mode setting:
2071 kms, kbd format: kms,kbd
2072 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud]
2074 kgdbwait [KGDB] Stop kernel execution and enter the
2075 kernel debugger at the earliest opportunity.
2077 kmac= [MIPS] korina ethernet MAC address.
2078 Configure the RouterBoard 532 series on-chip
2079 Ethernet adapter MAC address.
2081 kmemleak= [KNL] Boot-time kmemleak enable/disable
2082 Valid arguments: on, off
2084 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y,
2087 kprobe_event=[probe-list]
2088 [FTRACE] Add kprobe events and enable at boot time.
2089 The probe-list is a semicolon delimited list of probe
2090 definitions. Each definition is same as kprobe_events
2091 interface, but the parameters are comma delimited.
2092 For example, to add a kprobe event on vfs_read with
2093 arg1 and arg2, add to the command line;
2095 kprobe_event=p,vfs_read,$arg1,$arg2
2097 See also Documentation/trace/kprobetrace.rst "Kernel
2098 Boot Parameter" section.
2100 kpti= [ARM64] Control page table isolation of user
2101 and kernel address spaces.
2102 Default: enabled on cores which need mitigation.
2106 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs.
2107 Default is 0 (don't ignore, but inject #GP)
2109 kvm.enable_vmware_backdoor=[KVM] Support VMware backdoor PV interface.
2110 Default is false (don't support).
2112 kvm.mmu_audit= [KVM] This is a R/W parameter which allows audit
2117 [KVM] Controls the software workaround for the
2118 X86_BUG_ITLB_MULTIHIT bug.
2119 force : Always deploy workaround.
2120 off : Never deploy workaround.
2121 auto : Deploy workaround based on the presence of
2122 X86_BUG_ITLB_MULTIHIT.
2126 If the software workaround is enabled for the host,
2127 guests do need not to enable it for nested guests.
2129 kvm.nx_huge_pages_recovery_ratio=
2130 [KVM] Controls how many 4KiB pages are periodically zapped
2131 back to huge pages. 0 disables the recovery, otherwise if
2132 the value is N KVM will zap 1/Nth of the 4KiB pages every
2133 minute. The default is 60.
2135 kvm-amd.nested= [KVM,AMD] Allow nested virtualization in KVM/SVM.
2136 Default is 1 (enabled)
2138 kvm-amd.npt= [KVM,AMD] Disable nested paging (virtualized MMU)
2140 Default is 1 (enabled) if in 64-bit or 32-bit PAE mode.
2142 kvm-arm.vgic_v3_group0_trap=
2143 [KVM,ARM] Trap guest accesses to GICv3 group-0
2146 kvm-arm.vgic_v3_group1_trap=
2147 [KVM,ARM] Trap guest accesses to GICv3 group-1
2150 kvm-arm.vgic_v3_common_trap=
2151 [KVM,ARM] Trap guest accesses to GICv3 common
2154 kvm-arm.vgic_v4_enable=
2155 [KVM,ARM] Allow use of GICv4 for direct injection of
2158 kvm-intel.ept= [KVM,Intel] Disable extended page tables
2159 (virtualized MMU) support on capable Intel chips.
2160 Default is 1 (enabled)
2162 kvm-intel.emulate_invalid_guest_state=
2163 [KVM,Intel] Enable emulation of invalid guest states
2164 Default is 0 (disabled)
2166 kvm-intel.flexpriority=
2167 [KVM,Intel] Disable FlexPriority feature (TPR shadow).
2168 Default is 1 (enabled)
2171 [KVM,Intel] Enable VMX nesting (nVMX).
2172 Default is 0 (disabled)
2174 kvm-intel.unrestricted_guest=
2175 [KVM,Intel] Disable unrestricted guest feature
2176 (virtualized real and unpaged mode) on capable
2177 Intel chips. Default is 1 (enabled)
2179 kvm-intel.vmentry_l1d_flush=[KVM,Intel] Mitigation for L1 Terminal Fault
2182 Valid arguments: never, cond, always
2184 always: L1D cache flush on every VMENTER.
2185 cond: Flush L1D on VMENTER only when the code between
2186 VMEXIT and VMENTER can leak host memory.
2187 never: Disables the mitigation
2189 Default is cond (do L1 cache flush in specific instances)
2191 kvm-intel.vpid= [KVM,Intel] Disable Virtual Processor Identification
2192 feature (tagged TLBs) on capable Intel chips.
2193 Default is 1 (enabled)
2195 l1tf= [X86] Control mitigation of the L1TF vulnerability on
2198 The kernel PTE inversion protection is unconditionally
2199 enabled and cannot be disabled.
2202 Provides all available mitigations for the
2203 L1TF vulnerability. Disables SMT and
2204 enables all mitigations in the
2205 hypervisors, i.e. unconditional L1D flush.
2207 SMT control and L1D flush control via the
2208 sysfs interface is still possible after
2209 boot. Hypervisors will issue a warning
2210 when the first VM is started in a
2211 potentially insecure configuration,
2212 i.e. SMT enabled or L1D flush disabled.
2215 Same as 'full', but disables SMT and L1D
2216 flush runtime control. Implies the
2217 'nosmt=force' command line option.
2218 (i.e. sysfs control of SMT is disabled.)
2221 Leaves SMT enabled and enables the default
2222 hypervisor mitigation, i.e. conditional
2225 SMT control and L1D flush control via the
2226 sysfs interface is still possible after
2227 boot. Hypervisors will issue a warning
2228 when the first VM is started in a
2229 potentially insecure configuration,
2230 i.e. SMT enabled or L1D flush disabled.
2234 Disables SMT and enables the default
2235 hypervisor mitigation.
2237 SMT control and L1D flush control via the
2238 sysfs interface is still possible after
2239 boot. Hypervisors will issue a warning
2240 when the first VM is started in a
2241 potentially insecure configuration,
2242 i.e. SMT enabled or L1D flush disabled.
2245 Same as 'flush', but hypervisors will not
2246 warn when a VM is started in a potentially
2247 insecure configuration.
2250 Disables hypervisor mitigations and doesn't
2252 It also drops the swap size and available
2253 RAM limit restriction on both hypervisor and
2258 For details see: Documentation/admin-guide/hw-vuln/l1tf.rst
2264 lapic [X86-32,APIC] Enable the local APIC even if BIOS
2267 lapic= [x86,APIC] "notscdeadline" Do not use TSC deadline
2268 value for LAPIC timer one-shot implementation. Default
2269 back to the programmable timer unit in the LAPIC.
2271 lapic_timer_c2_ok [X86,APIC] trust the local apic timer
2274 libata.dma= [LIBATA] DMA control
2275 libata.dma=0 Disable all PATA and SATA DMA
2276 libata.dma=1 PATA and SATA Disk DMA only
2277 libata.dma=2 ATAPI (CDROM) DMA only
2278 libata.dma=4 Compact Flash DMA only
2279 Combinations also work, so libata.dma=3 enables DMA
2280 for disks and CDROMs, but not CFs.
2282 libata.ignore_hpa= [LIBATA] Ignore HPA limit
2283 libata.ignore_hpa=0 keep BIOS limits (default)
2284 libata.ignore_hpa=1 ignore limits, using full disk
2286 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume
2290 libata.force= [LIBATA] Force configurations. The format is comma
2291 separated list of "[ID:]VAL" where ID is
2292 PORT[.DEVICE]. PORT and DEVICE are decimal numbers
2293 matching port, link or device. Basically, it matches
2294 the ATA ID string printed on console by libata. If
2295 the whole ID part is omitted, the last PORT and DEVICE
2296 values are used. If ID hasn't been specified yet, the
2297 configuration applies to all ports, links and devices.
2299 If only DEVICE is omitted, the parameter applies to
2300 the port and all links and devices behind it. DEVICE
2301 number of 0 either selects the first device or the
2302 first fan-out link behind PMP device. It does not
2303 select the host link. DEVICE number of 15 selects the
2304 host link and device attached to it.
2306 The VAL specifies the configuration to force. As long
2307 as there's no ambiguity shortcut notation is allowed.
2308 For example, both 1.5 and 1.5G would work for 1.5Gbps.
2309 The following configurations can be forced.
2311 * Cable type: 40c, 80c, short40c, unk, ign or sata.
2312 Any ID with matching PORT is used.
2314 * SATA link speed limit: 1.5Gbps or 3.0Gbps.
2316 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7].
2317 udma[/][16,25,33,44,66,100,133] notation is also
2320 * [no]ncq: Turn on or off NCQ.
2322 * [no]ncqtrim: Turn off queued DSM TRIM.
2324 * nohrst, nosrst, norst: suppress hard, soft
2327 * rstonce: only attempt one reset during
2328 hot-unplug link recovery
2330 * dump_id: dump IDENTIFY data.
2332 * atapi_dmadir: Enable ATAPI DMADIR bridge support
2334 * disable: Disable this device.
2336 If there are multiple matching configurations changing
2337 the same attribute, the last one is used.
2339 memblock=debug [KNL] Enable memblock debug messages.
2341 load_ramdisk= [RAM] List of ramdisks to load from floppy
2342 See Documentation/admin-guide/blockdev/ramdisk.rst.
2344 lockd.nlm_grace_period=P [NFS] Assign grace period.
2347 lockd.nlm_tcpport=N [NFS] Assign TCP port.
2350 lockd.nlm_timeout=T [NFS] Assign timeout value.
2353 lockd.nlm_udpport=M [NFS] Assign UDP port.
2356 lockdown= [SECURITY]
2357 { integrity | confidentiality }
2358 Enable the kernel lockdown feature. If set to
2359 integrity, kernel features that allow userland to
2360 modify the running kernel are disabled. If set to
2361 confidentiality, kernel features that allow userland
2362 to extract confidential information from the kernel
2365 locktorture.nreaders_stress= [KNL]
2366 Set the number of locking read-acquisition kthreads.
2367 Defaults to being automatically set based on the
2368 number of online CPUs.
2370 locktorture.nwriters_stress= [KNL]
2371 Set the number of locking write-acquisition kthreads.
2373 locktorture.onoff_holdoff= [KNL]
2374 Set time (s) after boot for CPU-hotplug testing.
2376 locktorture.onoff_interval= [KNL]
2377 Set time (s) between CPU-hotplug operations, or
2378 zero to disable CPU-hotplug testing.
2380 locktorture.shuffle_interval= [KNL]
2381 Set task-shuffle interval (jiffies). Shuffling
2382 tasks allows some CPUs to go into dyntick-idle
2383 mode during the locktorture test.
2385 locktorture.shutdown_secs= [KNL]
2386 Set time (s) after boot system shutdown. This
2387 is useful for hands-off automated testing.
2389 locktorture.stat_interval= [KNL]
2390 Time (s) between statistics printk()s.
2392 locktorture.stutter= [KNL]
2393 Time (s) to stutter testing, for example,
2394 specifying five seconds causes the test to run for
2395 five seconds, wait for five seconds, and so on.
2396 This tests the locking primitive's ability to
2397 transition abruptly to and from idle.
2399 locktorture.torture_type= [KNL]
2400 Specify the locking implementation to test.
2402 locktorture.verbose= [KNL]
2403 Enable additional printk() statements.
2405 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver
2408 loglevel= All Kernel Messages with a loglevel smaller than the
2409 console loglevel will be printed to the console. It can
2410 also be changed with klogd or other programs. The
2411 loglevels are defined as follows:
2413 0 (KERN_EMERG) system is unusable
2414 1 (KERN_ALERT) action must be taken immediately
2415 2 (KERN_CRIT) critical conditions
2416 3 (KERN_ERR) error conditions
2417 4 (KERN_WARNING) warning conditions
2418 5 (KERN_NOTICE) normal but significant condition
2419 6 (KERN_INFO) informational
2420 7 (KERN_DEBUG) debug-level messages
2422 log_buf_len=n[KMG] Sets the size of the printk ring buffer,
2423 in bytes. n must be a power of two and greater
2424 than the minimal size. The minimal size is defined
2425 by LOG_BUF_SHIFT kernel config parameter. There is
2426 also CONFIG_LOG_CPU_MAX_BUF_SHIFT config parameter
2427 that allows to increase the default size depending on
2428 the number of CPUs. See init/Kconfig for more details.
2430 logo.nologo [FB] Disables display of the built-in Linux logo.
2431 This may be used to provide more screen space for
2432 kernel log messages and is useful when debugging
2433 kernel boot problems.
2435 lp=0 [LP] Specify parallel ports to use, e.g,
2436 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses
2437 lp=reset first parallel port). 'lp=0' disables the
2438 lp=auto printer driver. 'lp=reset' (which can be
2439 specified in addition to the ports) causes
2440 attached printers to be reset. Using
2441 lp=port1,port2,... specifies the parallel ports
2442 to associate lp devices with, starting with
2443 lp0. A port specification may be 'none' to skip
2444 that lp device, or a parport name such as
2445 'parport0'. Specifying 'lp=auto' instead of a
2446 port specification list means that device IDs
2447 from each port should be examined, to see if
2448 an IEEE 1284-compliant printer is attached; if
2449 so, the driver will manage that printer.
2450 See also header of drivers/char/lp.c.
2453 Sets loops_per_jiffy to given constant, thus avoiding
2454 time-consuming boot-time autodetection (up to 250 ms per
2455 CPU). 0 enables autodetection (default). To determine
2456 the correct value for your kernel, boot with normal
2457 autodetection and see what value is printed. Note that
2458 on SMP systems the preset will be applied to all CPUs,
2459 which is likely to cause problems if your CPUs need
2460 significantly divergent settings. An incorrect value
2461 will cause delays in the kernel to be wrong, leading to
2462 unpredictable I/O errors and other breakage. Although
2463 unlikely, in the extreme case this might damage your
2467 Format: <io>,<irq>,<dma>
2469 lsm.debug [SECURITY] Enable LSM initialization debugging output.
2472 [SECURITY] Choose order of LSM initialization. This
2473 overrides CONFIG_LSM, and the "security=" parameter.
2475 machvec= [IA-64] Force the use of a particular machine-vector
2476 (machvec) in a generic kernel.
2477 Example: machvec=hpzx1
2479 machtype= [Loongson] Share the same kernel image file between different
2481 Example: machtype=lemote-yeeloong-2f-7inch
2483 max_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory greater
2484 than or equal to this physical address is ignored.
2486 maxcpus= [SMP] Maximum number of processors that an SMP kernel
2487 will bring up during bootup. maxcpus=n : n >= 0 limits
2488 the kernel to bring up 'n' processors. Surely after
2489 bootup you can bring up the other plugged cpu by executing
2490 "echo 1 > /sys/devices/system/cpu/cpuX/online". So maxcpus
2491 only takes effect during system bootup.
2492 While n=0 is a special case, it is equivalent to "nosmp",
2493 which also disables the IO APIC.
2495 max_loop= [LOOP] The number of loop block devices that get
2496 (loop.max_loop) unconditionally pre-created at init time. The default
2497 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead
2498 of statically allocating a predefined number, loop
2499 devices can be requested on-demand with the
2500 /dev/loop-control interface.
2502 mce [X86-32] Machine Check Exception
2504 mce=option [X86-64] See Documentation/x86/x86_64/boot-options.rst
2506 md= [HW] RAID subsystems devices and level
2507 See Documentation/admin-guide/md.rst.
2510 Format: <first>,<last>
2511 Specifies range of consoles to be captured by the MDA.
2514 Control mitigation for the Micro-architectural Data
2515 Sampling (MDS) vulnerability.
2517 Certain CPUs are vulnerable to an exploit against CPU
2518 internal buffers which can forward information to a
2519 disclosure gadget under certain conditions.
2521 In vulnerable processors, the speculatively
2522 forwarded data can be used in a cache side channel
2523 attack, to access data to which the attacker does
2524 not have direct access.
2526 This parameter controls the MDS mitigation. The
2529 full - Enable MDS mitigation on vulnerable CPUs
2530 full,nosmt - Enable MDS mitigation and disable
2531 SMT on vulnerable CPUs
2532 off - Unconditionally disable MDS mitigation
2534 On TAA-affected machines, mds=off can be prevented by
2535 an active TAA mitigation as both vulnerabilities are
2536 mitigated with the same mechanism so in order to disable
2537 this mitigation, you need to specify tsx_async_abort=off
2540 Not specifying this option is equivalent to
2543 For details see: Documentation/admin-guide/hw-vuln/mds.rst
2545 mem=nn[KMG] [KNL,BOOT] Force usage of a specific amount of memory
2546 Amount of memory to be used when the kernel is not able
2547 to see the whole system memory or for test.
2548 [X86] Work as limiting max address. Use together
2549 with memmap= to avoid physical address space collisions.
2550 Without memmap= PCI devices could be placed at addresses
2551 belonging to unused RAM.
2553 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel
2557 [KNL,SH] Allow user to override the default size for
2558 per-device physically contiguous DMA buffers.
2560 memhp_default_state=online/offline
2561 [KNL] Set the initial state for the memory hotplug
2562 onlining policy. If not specified, the default value is
2563 set according to the
2564 CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE kernel config
2566 See Documentation/admin-guide/mm/memory-hotplug.rst.
2568 memmap=exactmap [KNL,X86] Enable setting of an exact
2569 E820 memory map, as specified by the user.
2570 Such memmap=exactmap lines can be constructed based on
2571 BIOS output or other requirements. See the memmap=nn@ss
2574 memmap=nn[KMG]@ss[KMG]
2575 [KNL] Force usage of a specific region of memory.
2576 Region of memory to be used is from ss to ss+nn.
2577 If @ss[KMG] is omitted, it is equivalent to mem=nn[KMG],
2578 which limits max address to nn[KMG].
2579 Multiple different regions can be specified,
2582 memmap=100M@2G,100M#3G,1G!1024G
2584 memmap=nn[KMG]#ss[KMG]
2585 [KNL,ACPI] Mark specific memory as ACPI data.
2586 Region of memory to be marked is from ss to ss+nn.
2588 memmap=nn[KMG]$ss[KMG]
2589 [KNL,ACPI] Mark specific memory as reserved.
2590 Region of memory to be reserved is from ss to ss+nn.
2591 Example: Exclude memory from 0x18690000-0x1869ffff
2592 memmap=64K$0x18690000
2594 memmap=0x10000$0x18690000
2595 Some bootloaders may need an escape character before '$',
2596 like Grub2, otherwise '$' and the following number
2599 memmap=nn[KMG]!ss[KMG]
2600 [KNL,X86] Mark specific memory as protected.
2601 Region of memory to be used, from ss to ss+nn.
2602 The memory region may be marked as e820 type 12 (0xc)
2603 and is NVDIMM or ADR memory.
2605 memmap=<size>%<offset>-<oldtype>+<newtype>
2606 [KNL,ACPI] Convert memory within the specified region
2607 from <oldtype> to <newtype>. If "-<oldtype>" is left
2608 out, the whole region will be marked as <newtype>,
2609 even if previously unavailable. If "+<newtype>" is left
2610 out, matching memory will be removed. Types are
2611 specified as e820 types, e.g., 1 = RAM, 2 = reserved,
2612 3 = ACPI, 12 = PRAM.
2614 memory_corruption_check=0/1 [X86]
2615 Some BIOSes seem to corrupt the first 64k of
2616 memory when doing things like suspend/resume.
2617 Setting this option will scan the memory
2618 looking for corruption. Enabling this will
2619 both detect corruption and prevent the kernel
2620 from using the memory being corrupted.
2621 However, its intended as a diagnostic tool; if
2622 repeatable BIOS-originated corruption always
2623 affects the same memory, you can use memmap=
2624 to prevent the kernel from using that memory.
2626 memory_corruption_check_size=size [X86]
2627 By default it checks for corruption in the low
2628 64k, making this memory unavailable for normal
2629 use. Use this parameter to scan for
2630 corruption in more or less memory.
2632 memory_corruption_check_period=seconds [X86]
2633 By default it checks for corruption every 60
2634 seconds. Use this parameter to check at some
2635 other rate. 0 disables periodic checking.
2637 memtest= [KNL,X86,ARM,PPC] Enable memtest
2639 default : 0 <disable>
2640 Specifies the number of memtest passes to be
2641 performed. Each pass selects another test
2642 pattern from a given set of patterns. Memtest
2643 fills the memory with this pattern, validates
2644 memory contents and reserves bad memory
2645 regions that are detected.
2647 mem_encrypt= [X86-64] AMD Secure Memory Encryption (SME) control
2648 Valid arguments: on, off
2649 Default (depends on kernel configuration option):
2650 on (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=y)
2651 off (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=n)
2652 mem_encrypt=on: Activate SME
2653 mem_encrypt=off: Do not activate SME
2655 Refer to Documentation/virt/kvm/amd-memory-encryption.rst
2656 for details on when memory encryption can be activated.
2658 mem_sleep_default= [SUSPEND] Default system suspend mode:
2659 s2idle - Suspend-To-Idle
2660 shallow - Power-On Suspend or equivalent (if supported)
2661 deep - Suspend-To-RAM or equivalent (if supported)
2662 See Documentation/admin-guide/pm/sleep-states.rst.
2664 meye.*= [HW] Set MotionEye Camera parameters
2665 See Documentation/media/v4l-drivers/meye.rst.
2667 mfgpt_irq= [IA-32] Specify the IRQ to use for the
2668 Multi-Function General Purpose Timers on AMD Geode
2671 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when
2672 the BIOS has incorrectly applied a workaround. TinyBIOS
2673 version 0.98 is known to be affected, 0.99 fixes the
2674 problem by letting the user disable the workaround.
2678 min_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory below this
2679 physical address is ignored.
2681 mini2440= [ARM,HW,KNL]
2682 Format:[0..2][b][c][t]
2684 MINI2440 configuration specification:
2685 0 - The attached screen is the 3.5" TFT
2686 1 - The attached screen is the 7" TFT
2687 2 - The VGA Shield is attached (1024x768)
2688 Leaving out the screen size parameter will not load
2689 the TFT driver, and the framebuffer will be left
2691 b - Enable backlight. The TFT backlight pin will be
2692 linked to the kernel VESA blanking code and a GPIO
2693 LED. This parameter is not necessary when using the
2695 c - Enable the s3c camera interface.
2696 t - Reserved for enabling touchscreen support. The
2697 touchscreen support is not enabled in the mainstream
2698 kernel as of 2.6.30, a preliminary port can be found
2699 in the "bleeding edge" mini2440 support kernel at
2700 http://repo.or.cz/w/linux-2.6/mini2440.git
2703 [X86,PPC,S390,ARM64] Control optional mitigations for
2704 CPU vulnerabilities. This is a set of curated,
2705 arch-independent options, each of which is an
2706 aggregation of existing arch-specific options.
2709 Disable all optional CPU mitigations. This
2710 improves system performance, but it may also
2711 expose users to several CPU vulnerabilities.
2712 Equivalent to: nopti [X86,PPC]
2714 nospectre_v1 [X86,PPC]
2716 nospectre_v2 [X86,PPC,S390,ARM64]
2717 spectre_v2_user=off [X86]
2718 spec_store_bypass_disable=off [X86,PPC]
2719 ssbd=force-off [ARM64]
2722 tsx_async_abort=off [X86]
2723 kvm.nx_huge_pages=off [X86]
2726 This does not have any effect on
2727 kvm.nx_huge_pages when
2728 kvm.nx_huge_pages=force.
2731 Mitigate all CPU vulnerabilities, but leave SMT
2732 enabled, even if it's vulnerable. This is for
2733 users who don't want to be surprised by SMT
2734 getting disabled across kernel upgrades, or who
2735 have other ways of avoiding SMT-based attacks.
2736 Equivalent to: (default behavior)
2739 Mitigate all CPU vulnerabilities, disabling SMT
2740 if needed. This is for users who always want to
2741 be fully mitigated, even if it means losing SMT.
2742 Equivalent to: l1tf=flush,nosmt [X86]
2743 mds=full,nosmt [X86]
2744 tsx_async_abort=full,nosmt [X86]
2747 [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this
2748 parameter allows control of the logging verbosity for
2749 the additional memory initialisation checks. A value
2750 of 0 disables mminit logging and a level of 4 will
2751 log everything. Information is printed at KERN_DEBUG
2752 so loglevel=8 may also need to be specified.
2755 [KNL] When CONFIG_MODULE_SIG is set, this means that
2756 modules without (valid) signatures will fail to load.
2757 Note that if CONFIG_MODULE_SIG_FORCE is set, that
2758 is always true, so this option does nothing.
2760 module_blacklist= [KNL] Do not load a comma-separated list of
2761 modules. Useful for debugging problem modules.
2764 [MOUSE] Maximum time between finger touching and
2765 leaving touchpad surface for touch to be considered
2766 a tap and be reported as a left button click (for
2767 touchpads working in absolute mode only).
2769 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices
2770 reporting absolute coordinates, such as tablets
2771 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices
2772 reporting absolute coordinates, such as tablets
2774 movablecore= [KNL,X86,IA-64,PPC]
2775 Format: nn[KMGTPE] | nn%
2776 This parameter is the complement to kernelcore=, it
2777 specifies the amount of memory used for migratable
2778 allocations. If both kernelcore and movablecore is
2779 specified, then kernelcore will be at *least* the
2780 specified value but may be more. If movablecore on its
2781 own is specified, the administrator must be careful
2782 that the amount of memory usable for all allocations
2785 movable_node [KNL] Boot-time switch to make hotplugable memory
2786 NUMA nodes to be movable. This means that the memory
2787 of such nodes will be usable only for movable
2788 allocations which rules out almost all kernel
2789 allocations. Use with caution!
2791 MTD_Partition= [MTD]
2792 Format: <name>,<region-number>,<size>,<offset>
2794 MTD_Region= [MTD] Format:
2795 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>]
2798 See drivers/mtd/cmdlinepart.c.
2800 multitce=off [PPC] This parameter disables the use of the pSeries
2801 firmware feature for updating multiple TCE entries
2804 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration
2806 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock]
2808 boundary - index of last SLC block on Flex-OneNAND.
2809 The remaining blocks are configured as MLC blocks.
2810 lock - Configure if Flex-OneNAND boundary should be locked.
2811 Once locked, the boundary cannot be changed.
2812 1 indicates lock status, 0 indicates unlock status.
2815 ARM/S3C2412 JIVE boot control
2817 See arch/arm/mach-s3c2412/mach-jive.c
2819 mtouchusb.raw_coordinates=
2820 [HW] Make the MicroTouch USB driver use raw coordinates
2821 ('y', default) or cooked coordinates ('n')
2823 mtrr_chunk_size=nn[KMG] [X86]
2824 used for mtrr cleanup. It is largest continuous chunk
2825 that could hold holes aka. UC entries.
2827 mtrr_gran_size=nn[KMG] [X86]
2828 Used for mtrr cleanup. It is granularity of mtrr block.
2830 Large value could prevent small alignment from
2833 mtrr_spare_reg_nr=n [X86]
2835 Range: 0,7 : spare reg number
2837 Used for mtrr cleanup. It is spare mtrr entries number.
2838 Set to 2 or more if your graphical card needs more.
2840 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card
2842 netdev= [NET] Network devices parameters
2843 Format: <irq>,<io>,<mem_start>,<mem_end>,<name>
2844 Note that mem_start is often overloaded to mean
2845 something different and driver-specific.
2846 This usage is only documented in each driver source
2850 [NETFILTER] Enable connection tracking flow accounting
2851 0 to disable accounting
2852 1 to enable accounting
2855 nfsaddrs= [NFS] Deprecated. Use ip= instead.
2856 See Documentation/filesystems/nfs/nfsroot.txt.
2858 nfsroot= [NFS] nfs root filesystem for disk-less boxes.
2859 See Documentation/filesystems/nfs/nfsroot.txt.
2861 nfsrootdebug [NFS] enable nfsroot debugging messages.
2862 See Documentation/filesystems/nfs/nfsroot.txt.
2864 nfs.callback_nr_threads=
2865 [NFSv4] set the total number of threads that the
2866 NFS client will assign to service NFSv4 callback
2869 nfs.callback_tcpport=
2870 [NFS] set the TCP port on which the NFSv4 callback
2871 channel should listen.
2874 [NFS] sets the pathname to the program which is used
2875 to update the NFS client cache entries.
2877 nfs.cache_getent_timeout=
2878 [NFS] sets the timeout after which an attempt to
2879 update a cache entry is deemed to have failed.
2881 nfs.idmap_cache_timeout=
2882 [NFS] set the maximum lifetime for idmapper cache
2886 [NFS] enable 64-bit inode numbers.
2887 If zero, the NFS client will fake up a 32-bit inode
2888 number for the readdir() and stat() syscalls instead
2889 of returning the full 64-bit number.
2890 The default is to return 64-bit inode numbers.
2892 nfs.max_session_cb_slots=
2893 [NFSv4.1] Sets the maximum number of session
2894 slots the client will assign to the callback
2895 channel. This determines the maximum number of
2896 callbacks the client will process in parallel for
2897 a particular server.
2899 nfs.max_session_slots=
2900 [NFSv4.1] Sets the maximum number of session slots
2901 the client will attempt to negotiate with the server.
2902 This limits the number of simultaneous RPC requests
2903 that the client can send to the NFSv4.1 server.
2904 Note that there is little point in setting this
2905 value higher than the max_tcp_slot_table_limit.
2907 nfs.nfs4_disable_idmapping=
2908 [NFSv4] When set to the default of '1', this option
2909 ensures that both the RPC level authentication
2910 scheme and the NFS level operations agree to use
2911 numeric uids/gids if the mount is using the
2912 'sec=sys' security flavour. In effect it is
2913 disabling idmapping, which can make migration from
2914 legacy NFSv2/v3 systems to NFSv4 easier.
2915 Servers that do not support this mode of operation
2916 will be autodetected by the client, and it will fall
2917 back to using the idmapper.
2918 To turn off this behaviour, set the value to '0'.
2920 [NFS4] Specify an additional fixed unique ident-
2921 ification string that NFSv4 clients can insert into
2922 their nfs_client_id4 string. This is typically a
2923 UUID that is generated at system install time.
2925 nfs.send_implementation_id =
2926 [NFSv4.1] Send client implementation identification
2927 information in exchange_id requests.
2928 If zero, no implementation identification information
2930 The default is to send the implementation identification
2933 nfs.recover_lost_locks =
2934 [NFSv4] Attempt to recover locks that were lost due
2935 to a lease timeout on the server. Please note that
2936 doing this risks data corruption, since there are
2937 no guarantees that the file will remain unchanged
2938 after the locks are lost.
2939 If you want to enable the kernel legacy behaviour of
2940 attempting to recover these locks, then set this
2942 The default parameter value of '0' causes the kernel
2943 not to attempt recovery of lost locks.
2945 nfs4.layoutstats_timer =
2946 [NFSv4.2] Change the rate at which the kernel sends
2947 layoutstats to the pNFS metadata server.
2949 Setting this to value to 0 causes the kernel to use
2950 whatever value is the default set by the layout
2951 driver. A non-zero value sets the minimum interval
2952 in seconds between layoutstats transmissions.
2954 nfsd.nfs4_disable_idmapping=
2955 [NFSv4] When set to the default of '1', the NFSv4
2956 server will return only numeric uids and gids to
2957 clients using auth_sys, and will accept numeric uids
2958 and gids from such clients. This is intended to ease
2959 migration from NFSv2/v3.
2961 nmi_debug= [KNL,SH] Specify one or more actions to take
2962 when a NMI is triggered.
2963 Format: [state][,regs][,debounce][,die]
2965 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels
2966 Format: [panic,][nopanic,][num]
2968 0 - turn hardlockup detector in nmi_watchdog off
2969 1 - turn hardlockup detector in nmi_watchdog on
2970 When panic is specified, panic when an NMI watchdog
2971 timeout occurs (or 'nopanic' to not panic on an NMI
2972 watchdog, if CONFIG_BOOTPARAM_HARDLOCKUP_PANIC is set)
2973 To disable both hard and soft lockup detectors,
2974 please see 'nowatchdog'.
2975 This is useful when you use a panic=... timeout and
2976 need the box quickly up again.
2978 These settings can be accessed at runtime via
2979 the nmi_watchdog and hardlockup_panic sysctls.
2981 netpoll.carrier_timeout=
2982 [NET] Specifies amount of time (in seconds) that
2983 netpoll should wait for a carrier. By default netpoll
2986 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths
2987 emulation library even if a 387 maths coprocessor
2990 no5lvl [X86-64] Disable 5-level paging mode. Forces
2991 kernel to use 4-level paging instead.
2994 [HW] Never suspend the console
2995 Disable suspending of consoles during suspend and
2996 hibernate operations. Once disabled, debugging
2997 messages can reach various consoles while the rest
2998 of the system is being put to sleep (ie, while
2999 debugging driver suspend/resume hooks). This may
3000 not work reliably with all consoles, but is known
3001 to work with serial and VGA consoles.
3002 To facilitate more flexible debugging, we also add
3003 console_suspend, a printk module parameter to control
3004 it. Users could use console_suspend (usually
3005 /sys/module/printk/parameters/console_suspend) to
3006 turn on/off it dynamically.
3008 novmcoredd [KNL,KDUMP]
3009 Disable device dump. Device dump allows drivers to
3010 append dump data to vmcore so you can collect driver
3011 specified debug info. Drivers can append the data
3012 without any limit and this data is stored in memory,
3013 so this may cause significant memory stress. Disabling
3014 device dump can help save memory but the driver debug
3015 data will be no longer available. This parameter
3016 is only available when CONFIG_PROC_VMCORE_DEVICE_DUMP
3019 noaliencache [MM, NUMA, SLAB] Disables the allocation of alien
3020 caches in the slab allocator. Saves per-node memory,
3021 but will impact performance.
3025 noaltinstr [S390] Disables alternative instructions patching
3026 (CPU alternatives feature).
3028 noapic [SMP,APIC] Tells the kernel to not make use of any
3029 IOAPICs that may be present in the system.
3031 noautogroup Disable scheduler automatic task group creation.
3033 nobats [PPC] Do not use BATs for mapping kernel lowmem
3034 on "Classic" PPC cores.
3038 noclflush [BUGS=X86] Don't use the CLFLUSH instruction
3040 nodelayacct [KNL] Disable per-task delay accounting
3042 nodsp [SH] Disable hardware DSP at boot time.
3044 noefi Disable EFI runtime services support.
3049 On X86-32 available only on PAE configured kernels.
3050 noexec=on: enable non-executable mappings (default)
3051 noexec=off: disable non-executable mappings
3054 Disable SMAP (Supervisor Mode Access Prevention)
3055 even if it is supported by processor.
3058 Disable SMEP (Supervisor Mode Execution Prevention)
3059 even if it is supported by processor.
3062 This affects only 32-bit executables.
3063 noexec32=on: enable non-executable mappings (default)
3064 read doesn't imply executable mappings
3065 noexec32=off: disable non-executable mappings
3066 read implies executable mappings
3068 nofpu [MIPS,SH] Disable hardware FPU at boot time.
3070 nofxsr [BUGS=X86-32] Disables x86 floating point extended
3071 register save and restore. The kernel will only save
3072 legacy floating-point registers on task switch.
3074 nohugeiomap [KNL,x86,PPC] Disable kernel huge I/O mappings.
3076 nosmt [KNL,S390] Disable symmetric multithreading (SMT).
3077 Equivalent to smt=1.
3079 [KNL,x86] Disable symmetric multithreading (SMT).
3080 nosmt=force: Force disable SMT, cannot be undone
3081 via the sysfs control file.
3083 nospectre_v1 [X86,PPC] Disable mitigations for Spectre Variant 1
3084 (bounds check bypass). With this option data leaks are
3085 possible in the system.
3087 nospectre_v2 [X86,PPC_FSL_BOOK3E,ARM64] Disable all mitigations for
3088 the Spectre variant 2 (indirect branch prediction)
3089 vulnerability. System may allow data leaks with this
3092 nospec_store_bypass_disable
3093 [HW] Disable all mitigations for the Speculative Store Bypass vulnerability
3095 noxsave [BUGS=X86] Disables x86 extended register state save
3096 and restore using xsave. The kernel will fallback to
3097 enabling legacy floating-point and sse state.
3099 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended
3100 register states. The kernel will fall back to use
3101 xsave to save the states. By using this parameter,
3102 performance of saving the states is degraded because
3103 xsave doesn't support modified optimization while
3104 xsaveopt supports it on xsaveopt enabled systems.
3106 noxsaves [X86] Disables xsaves and xrstors used in saving and
3107 restoring x86 extended register state in compacted
3108 form of xsave area. The kernel will fall back to use
3109 xsaveopt and xrstor to save and restore the states
3110 in standard form of xsave area. By using this
3111 parameter, xsave area per process might occupy more
3112 memory on xsaves enabled systems.
3114 nohlt [BUGS=ARM,SH] Tells the kernel that the sleep(SH) or
3115 wfi(ARM) instruction doesn't work correctly and not to
3116 use it. This is also useful when using JTAG debugger.
3118 no_file_caps Tells the kernel not to honor file capabilities. The
3119 only way then for a file to be executed with privilege
3120 is to be setuid root or executed by root.
3122 nohalt [IA-64] Tells the kernel not to use the power saving
3123 function PAL_HALT_LIGHT when idle. This increases
3124 power-consumption. On the positive side, it reduces
3125 interrupt wake-up latency, which may improve performance
3126 in certain environments such as networked servers or
3129 nohibernate [HIBERNATION] Disable hibernation and resume.
3131 nohz= [KNL] Boottime enable/disable dynamic ticks
3132 Valid arguments: on, off
3135 nohz_full= [KNL,BOOT,SMP,ISOL]
3136 The argument is a cpu list, as described above.
3137 In kernels built with CONFIG_NO_HZ_FULL=y, set
3138 the specified list of CPUs whose tick will be stopped
3139 whenever possible. The boot CPU will be forced outside
3140 the range to maintain the timekeeping. Any CPUs
3141 in this list will have their RCU callbacks offloaded,
3142 just as if they had also been called out in the
3143 rcu_nocbs= boot parameter.
3145 noiotrap [SH] Disables trapped I/O port accesses.
3147 noirqdebug [X86-32] Disables the code which attempts to detect and
3148 disable unhandled interrupt sources.
3150 no_timer_check [X86,APIC] Disables the code which tests for
3151 broken timer IRQ sources.
3153 noisapnp [ISAPNP] Disables ISA PnP code.
3155 noinitrd [RAM] Tells the kernel not to load any configured
3158 nointremap [X86-64, Intel-IOMMU] Do not enable interrupt
3160 [Deprecated - use intremap=off]
3164 noinvpcid [X86] Disable the INVPCID cpu feature.
3166 nojitter [IA-64] Disables jitter checking for ITC timers.
3168 no-kvmclock [X86,KVM] Disable paravirtualized KVM clock driver
3170 no-kvmapf [X86,KVM] Disable paravirtualized asynchronous page
3174 [X86,PV_OPS] Disable paravirtualized VMware scheduler
3175 clock and use the default one.
3177 no-steal-acc [X86,KVM,ARM64] Disable paravirtualized steal time
3178 accounting. steal time is computed, but won't
3179 influence scheduler behaviour
3181 nolapic [X86-32,APIC] Do not enable or use the local APIC.
3183 nolapic_timer [X86-32,APIC] Do not use the local APIC timer.
3185 noltlbs [PPC] Do not use large page/tlb entries for kernel
3186 lowmem mapping on PPC40x and PPC8xx
3188 nomca [IA-64] Disable machine check abort handling
3190 nomce [X86-32] Disable Machine Check Exception
3192 nomfgpt [X86-32] Disable Multi-Function General Purpose
3193 Timer usage (for AMD Geode machines).
3195 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to
3196 shutdown the other cpus. Instead use the REBOOT_VECTOR
3199 nomodule Disable module load
3201 nopat [X86] Disable PAT (page attribute table extension of
3202 pagetables) support.
3204 nopcid [X86-64] Disable the PCID cpu feature.
3206 norandmaps Don't use address space randomization. Equivalent to
3207 echo 0 > /proc/sys/kernel/randomize_va_space
3209 noreplace-smp [X86-32,SMP] Don't replace SMP instructions
3210 with UP alternatives
3212 nordrand [X86] Disable kernel use of the RDRAND and
3213 RDSEED instructions even if they are supported
3214 by the processor. RDRAND and RDSEED are still
3215 available to user space applications.
3217 noresume [SWSUSP] Disables resume and restores original swap
3220 no-scroll [VGA] Disables scrollback.
3221 This is required for the Braillex ib80-piezo Braille
3222 reader made by F.H. Papenmeier (Germany).
3226 nosep [BUGS=X86-32] Disables x86 SYSENTER/SYSEXIT support.
3228 nosmp [SMP] Tells an SMP kernel to act as a UP kernel,
3229 and disable the IO APIC. legacy for "maxcpus=0".
3231 nosoftlockup [KNL] Disable the soft-lockup detector.
3233 nosync [HW,M68K] Disables sync negotiation for all devices.
3235 nowatchdog [KNL] Disable both lockup detectors, i.e.
3236 soft-lockup and NMI watchdog (hard-lockup).
3240 nox2apic [X86-64,APIC] Do not enable x2APIC mode.
3242 cpu0_hotplug [X86] Turn on CPU0 hotplug feature when
3243 CONFIG_BOOTPARAM_HOTPLUG_CPU0 is off.
3244 Some features depend on CPU0. Known dependencies are:
3245 1. Resume from suspend/hibernate depends on CPU0.
3246 Suspend/hibernate will fail if CPU0 is offline and you
3247 need to online CPU0 before suspend/hibernate.
3248 2. PIC interrupts also depend on CPU0. CPU0 can't be
3249 removed if a PIC interrupt is detected.
3250 It's said poweroff/reboot may depend on CPU0 on some
3251 machines although I haven't seen such issues so far
3252 after CPU0 is offline on a few tested machines.
3253 If the dependencies are under your control, you can
3254 turn on cpu0_hotplug.
3256 nps_mtm_hs_ctr= [KNL,ARC]
3257 This parameter sets the maximum duration, in
3258 cycles, each HW thread of the CTOP can run
3259 without interruptions, before HW switches it.
3260 The actual maximum duration is 16 times this
3262 Format: integer between 1 and 255
3265 nptcg= [IA-64] Override max number of concurrent global TLB
3266 purges which is reported from either PAL_VM_SUMMARY or
3269 nr_cpus= [SMP] Maximum number of processors that an SMP kernel
3270 could support. nr_cpus=n : n >= 1 limits the kernel to
3271 support 'n' processors. It could be larger than the
3272 number of already plugged CPU during bootup, later in
3273 runtime you can physically add extra cpu until it reaches
3274 n. So during boot up some boot time memory for per-cpu
3275 variables need be pre-allocated for later physical cpu
3278 nr_uarts= [SERIAL] maximum number of UARTs to be registered.
3280 numa_balancing= [KNL,X86] Enable or disable automatic NUMA balancing.
3281 Allowed values are enable and disable
3283 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA.
3284 'node', 'default' can be specified
3285 This can be set from sysctl after boot.
3286 See Documentation/admin-guide/sysctl/vm.rst for details.
3288 of_devlink [OF, KNL] Create device links between consumer and
3289 supplier devices by scanning the devictree to infer the
3290 consumer/supplier relationships. A consumer device
3291 will not be probed until all the supplier devices have
3292 probed successfully.
3294 ohci1394_dma=early [HW] enable debugging via the ohci1394 driver.
3295 See Documentation/debugging-via-ohci1394.txt for more
3298 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands
3299 Rather than timing out after 20 ms if an EC
3300 command is not properly ACKed, override the length
3301 of the timeout. We have interrupts disabled while
3302 waiting for the ACK, so if this is set too high
3303 interrupts *may* be lost!
3305 omap_mux= [OMAP] Override bootloader pin multiplexing.
3306 Format: <mux_mode0.mode_name=value>...
3307 For example, to override I2C bus2:
3308 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100
3310 oprofile.timer= [HW]
3311 Use timer interrupt instead of performance counters
3313 oprofile.cpu_type= Force an oprofile cpu type
3314 This might be useful if you have an older oprofile
3315 userland or if you want common events.
3316 Format: { arch_perfmon }
3317 arch_perfmon: [X86] Force use of architectural
3318 perfmon on Intel CPUs instead of the
3319 CPU specific event set.
3320 timer: [X86] Force use of architectural NMI
3321 timer mode (see also oprofile.timer
3322 for generic hr timer mode)
3324 oops=panic Always panic on oopses. Default is to just kill the
3325 process, but there is a small probability of
3326 deadlocking the machine.
3327 This will also cause panics on machine check exceptions.
3328 Useful together with panic=30 to trigger a reboot.
3331 [KNL] Boolean flag to control whether the page allocator
3332 should randomize its free lists. The randomization may
3333 be automatically enabled if the kernel detects it is
3334 running on a platform with a direct-mapped memory-side
3335 cache, and this parameter can be used to
3336 override/disable that behavior. The state of the flag
3337 can be read from sysfs at:
3338 /sys/module/page_alloc/parameters/shuffle.
3340 page_owner= [KNL] Boot-time page_owner enabling option.
3341 Storage of the information about who allocated
3342 each page is disabled in default. With this switch,
3344 on: enable the feature
3346 page_poison= [KNL] Boot-time parameter changing the state of
3347 poisoning on the buddy allocator, available with
3348 CONFIG_PAGE_POISONING=y.
3349 off: turn off poisoning (default)
3350 on: turn on poisoning
3352 panic= [KNL] Kernel behaviour on panic: delay <timeout>
3353 timeout > 0: seconds before rebooting
3354 timeout = 0: wait forever
3355 timeout < 0: reboot immediately
3358 panic_print= Bitmask for printing system info when panic happens.
3359 User can chose combination of the following bits:
3360 bit 0: print all tasks info
3361 bit 1: print system memory info
3362 bit 2: print timer info
3363 bit 3: print locks info if CONFIG_LOCKDEP is on
3364 bit 4: print ftrace buffer
3365 bit 5: print all printk messages in buffer
3367 panic_on_warn panic() instead of WARN(). Useful to cause kdump
3370 crash_kexec_post_notifiers
3371 Run kdump after running panic-notifiers and dumping
3372 kmsg. This only for the users who doubt kdump always
3373 succeeds in any situation.
3374 Note that this also increases risks of kdump failure,
3375 because some panic notifiers can make the crashed
3376 kernel more unstable.
3378 parkbd.port= [HW] Parallel port number the keyboard adapter is
3379 connected to, default is 0.
3381 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation,
3382 0 for XT, 1 for AT (default is AT).
3385 parport= [HW,PPT] Specify parallel ports. 0 disables.
3386 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] }
3387 Use 'auto' to force the driver to use any
3388 IRQ/DMA settings detected (the default is to
3389 ignore detected IRQ/DMA settings because of
3390 possible conflicts). You can specify the base
3391 address, IRQ, and DMA settings; IRQ and DMA
3392 should be numbers, or 'auto' (for using detected
3393 settings on that particular port), or 'nofifo'
3394 (to avoid using a FIFO even if it is detected).
3395 Parallel ports are assigned in the order they
3396 are specified on the command line, starting
3399 parport_init_mode= [HW,PPT]
3400 Configure VIA parallel port to operate in
3401 a specific mode. This is necessary on Pegasos
3402 computer where firmware has no options for setting
3403 up parallel port mode and sets it to spp.
3404 Currently this function knows 686a and 8231 chips.
3405 Format: [spp|ps2|epp|ecp|ecpepp]
3408 Halt all CPUs after the first oops has been printed for
3409 the specified number of seconds. This is to be used if
3410 your oopses keep scrolling off the screen.
3415 See header of drivers/block/paride/pcd.c.
3416 See also Documentation/admin-guide/blockdev/paride.rst.
3418 pci=option[,option...] [PCI] various PCI subsystem options.
3420 Some options herein operate on a specific device
3421 or a set of devices (<pci_dev>). These are
3422 specified in one of the following formats:
3424 [<domain>:]<bus>:<dev>.<func>[/<dev>.<func>]*
3425 pci:<vendor>:<device>[:<subvendor>:<subdevice>]
3427 Note: the first format specifies a PCI
3428 bus/device/function address which may change
3429 if new hardware is inserted, if motherboard
3430 firmware changes, or due to changes caused
3431 by other kernel parameters. If the
3432 domain is left unspecified, it is
3433 taken to be zero. Optionally, a path
3434 to a device through multiple device/function
3435 addresses can be specified after the base
3436 address (this is more robust against
3437 renumbering issues). The second format
3438 selects devices using IDs from the
3439 configuration space which may match multiple
3440 devices in the system.
3442 earlydump dump PCI config space before the kernel
3444 off [X86] don't probe for the PCI bus
3445 bios [X86-32] force use of PCI BIOS, don't access
3446 the hardware directly. Use this if your machine
3447 has a non-standard PCI host bridge.
3448 nobios [X86-32] disallow use of PCI BIOS, only direct
3449 hardware access methods are allowed. Use this
3450 if you experience crashes upon bootup and you
3451 suspect they are caused by the BIOS.
3452 conf1 [X86] Force use of PCI Configuration Access
3453 Mechanism 1 (config address in IO port 0xCF8,
3454 data in IO port 0xCFC, both 32-bit).
3455 conf2 [X86] Force use of PCI Configuration Access
3456 Mechanism 2 (IO port 0xCF8 is an 8-bit port for
3457 the function, IO port 0xCFA, also 8-bit, sets
3458 bus number. The config space is then accessed
3459 through ports 0xC000-0xCFFF).
3460 See http://wiki.osdev.org/PCI for more info
3461 on the configuration access mechanisms.
3462 noaer [PCIE] If the PCIEAER kernel config parameter is
3463 enabled, this kernel boot option can be used to
3464 disable the use of PCIE advanced error reporting.
3465 nodomains [PCI] Disable support for multiple PCI
3466 root domains (aka PCI segments, in ACPI-speak).
3467 nommconf [X86] Disable use of MMCONFIG for PCI
3469 check_enable_amd_mmconf [X86] check for and enable
3470 properly configured MMIO access to PCI
3471 config space on AMD family 10h CPU
3472 nomsi [MSI] If the PCI_MSI kernel config parameter is
3473 enabled, this kernel boot option can be used to
3474 disable the use of MSI interrupts system-wide.
3475 noioapicquirk [APIC] Disable all boot interrupt quirks.
3476 Safety option to keep boot IRQs enabled. This
3477 should never be necessary.
3478 ioapicreroute [APIC] Enable rerouting of boot IRQs to the
3479 primary IO-APIC for bridges that cannot disable
3480 boot IRQs. This fixes a source of spurious IRQs
3481 when the system masks IRQs.
3482 noioapicreroute [APIC] Disable workaround that uses the
3483 boot IRQ equivalent of an IRQ that connects to
3484 a chipset where boot IRQs cannot be disabled.
3485 The opposite of ioapicreroute.
3486 biosirq [X86-32] Use PCI BIOS calls to get the interrupt
3487 routing table. These calls are known to be buggy
3488 on several machines and they hang the machine
3489 when used, but on other computers it's the only
3490 way to get the interrupt routing table. Try
3491 this option if the kernel is unable to allocate
3492 IRQs or discover secondary PCI buses on your
3494 rom [X86] Assign address space to expansion ROMs.
3495 Use with caution as certain devices share
3496 address decoders between ROMs and other
3498 norom [X86] Do not assign address space to
3499 expansion ROMs that do not already have
3500 BIOS assigned address ranges.
3501 nobar [X86] Do not assign address space to the
3502 BARs that weren't assigned by the BIOS.
3503 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be
3504 assigned automatically to PCI devices. You can
3505 make the kernel exclude IRQs of your ISA cards
3507 pirqaddr=0xAAAAA [X86] Specify the physical address
3508 of the PIRQ table (normally generated
3509 by the BIOS) if it is outside the
3510 F0000h-100000h range.
3511 lastbus=N [X86] Scan all buses thru bus #N. Can be
3512 useful if the kernel is unable to find your
3513 secondary buses and you want to tell it
3514 explicitly which ones they are.
3515 assign-busses [X86] Always assign all PCI bus
3516 numbers ourselves, overriding
3517 whatever the firmware may have done.
3518 usepirqmask [X86] Honor the possible IRQ mask stored
3519 in the BIOS $PIR table. This is needed on
3520 some systems with broken BIOSes, notably
3521 some HP Pavilion N5400 and Omnibook XE3
3522 notebooks. This will have no effect if ACPI
3523 IRQ routing is enabled.
3524 noacpi [X86] Do not use ACPI for IRQ routing
3525 or for PCI scanning.
3526 use_crs [X86] Use PCI host bridge window information
3527 from ACPI. On BIOSes from 2008 or later, this
3528 is enabled by default. If you need to use this,
3529 please report a bug.
3530 nocrs [X86] Ignore PCI host bridge windows from ACPI.
3531 If you need to use this, please report a bug.
3532 routeirq Do IRQ routing for all PCI devices.
3533 This is normally done in pci_enable_device(),
3534 so this option is a temporary workaround
3535 for broken drivers that don't call it.
3536 skip_isa_align [X86] do not align io start addr, so can
3537 handle more pci cards
3538 noearly [X86] Don't do any early type 1 scanning.
3539 This might help on some broken boards which
3540 machine check when some devices' config space
3541 is read. But various workarounds are disabled
3542 and some IOMMU drivers will not work.
3543 bfsort Sort PCI devices into breadth-first order.
3544 This sorting is done to get a device
3545 order compatible with older (<= 2.4) kernels.
3546 nobfsort Don't sort PCI devices into breadth-first order.
3547 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size)
3548 tuning and use the BIOS-configured MPS defaults.
3549 pcie_bus_safe Set every device's MPS to the largest value
3550 supported by all devices below the root complex.
3551 pcie_bus_perf Set device MPS to the largest allowable MPS
3552 based on its parent bus. Also set MRRS (Max
3553 Read Request Size) to the largest supported
3554 value (no larger than the MPS that the device
3555 or bus can support) for best performance.
3556 pcie_bus_peer2peer Set every device's MPS to 128B, which
3557 every device is guaranteed to support. This
3558 configuration allows peer-to-peer DMA between
3559 any pair of devices, possibly at the cost of
3560 reduced performance. This also guarantees
3561 that hot-added devices will work.
3562 cbiosize=nn[KMG] The fixed amount of bus space which is
3563 reserved for the CardBus bridge's IO window.
3564 The default value is 256 bytes.
3565 cbmemsize=nn[KMG] The fixed amount of bus space which is
3566 reserved for the CardBus bridge's memory
3567 window. The default value is 64 megabytes.
3570 [<order of align>@]<pci_dev>[; ...]
3571 Specifies alignment and device to reassign
3572 aligned memory resources. How to
3573 specify the device is described above.
3574 If <order of align> is not specified,
3575 PAGE_SIZE is used as alignment.
3576 A PCI-PCI bridge can be specified if resource
3577 windows need to be expanded.
3578 To specify the alignment for several
3579 instances of a device, the PCI vendor,
3580 device, subvendor, and subdevice may be
3581 specified, e.g., 12@pci:8086:9c22:103c:198f
3582 for 4096-byte alignment.
3583 ecrc= Enable/disable PCIe ECRC (transaction layer
3584 end-to-end CRC checking).
3585 bios: Use BIOS/firmware settings. This is the
3589 hpiosize=nn[KMG] The fixed amount of bus space which is
3590 reserved for hotplug bridge's IO window.
3591 Default size is 256 bytes.
3592 hpmmiosize=nn[KMG] The fixed amount of bus space which is
3593 reserved for hotplug bridge's MMIO window.
3594 Default size is 2 megabytes.
3595 hpmmioprefsize=nn[KMG] The fixed amount of bus space which is
3596 reserved for hotplug bridge's MMIO_PREF window.
3597 Default size is 2 megabytes.
3598 hpmemsize=nn[KMG] The fixed amount of bus space which is
3599 reserved for hotplug bridge's MMIO and
3601 Default size is 2 megabytes.
3602 hpbussize=nn The minimum amount of additional bus numbers
3603 reserved for buses below a hotplug bridge.
3605 realloc= Enable/disable reallocating PCI bridge resources
3606 if allocations done by BIOS are too small to
3607 accommodate resources required by all child
3609 off: Turn realloc off
3611 realloc same as realloc=on
3612 noari do not use PCIe ARI.
3613 noats [PCIE, Intel-IOMMU, AMD-IOMMU]
3614 do not use PCIe ATS (and IOMMU device IOTLB).
3615 pcie_scan_all Scan all possible PCIe devices. Otherwise we
3616 only look for one device below a PCIe downstream
3618 big_root_window Try to add a big 64bit memory window to the PCIe
3619 root complex on AMD CPUs. Some GFX hardware
3620 can resize a BAR to allow access to all VRAM.
3621 Adding the window is slightly risky (it may
3622 conflict with unreported devices), so this
3624 disable_acs_redir=<pci_dev>[; ...]
3625 Specify one or more PCI devices (in the format
3626 specified above) separated by semicolons.
3627 Each device specified will have the PCI ACS
3628 redirect capabilities forced off which will
3629 allow P2P traffic between devices through
3630 bridges without forcing it upstream. Note:
3631 this removes isolation between devices and
3632 may put more devices in an IOMMU group.
3633 force_floating [S390] Force usage of floating interrupts.
3634 nomio [S390] Do not use MIO instructions.
3636 pcie_aspm= [PCIE] Forcibly enable or disable PCIe Active State Power
3639 force Enable ASPM even on devices that claim not to support it.
3640 WARNING: Forcing ASPM on may cause system lockups.
3642 pcie_ports= [PCIE] PCIe port services handling:
3643 native Use native PCIe services (PME, AER, DPC, PCIe hotplug)
3644 even if the platform doesn't give the OS permission to
3645 use them. This may cause conflicts if the platform
3646 also tries to use these services.
3647 dpc-native Use native PCIe service for DPC only. May
3648 cause conflicts if firmware uses AER or DPC.
3649 compat Disable native PCIe services (PME, AER, DPC, PCIe
3652 pcie_port_pm= [PCIE] PCIe port power management handling:
3653 off Disable power management of all PCIe ports
3654 force Forcibly enable power management of all PCIe ports
3656 pcie_pme= [PCIE,PM] Native PCIe PME signaling options:
3657 nomsi Do not use MSI for native PCIe PME signaling (this makes
3658 all PCIe root ports use INTx for all services).
3660 pcmv= [HW,PCMCIA] BadgePAD 4
3664 Keep all power-domains already enabled by bootloader on,
3665 even if no driver has claimed them. This is useful
3666 for debug and development, but should not be
3667 needed on a platform with proper driver support.
3670 See Documentation/admin-guide/blockdev/paride.rst.
3672 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at
3675 See arch/parisc/kernel/pdc_chassis.c
3677 percpu_alloc= Select which percpu first chunk allocator to use.
3678 Currently supported values are "embed" and "page".
3679 Archs may support subset or none of the selections.
3680 See comments in mm/percpu.c for details on each
3681 allocator. This parameter is primarily for debugging
3682 and performance comparison.
3685 See Documentation/admin-guide/blockdev/paride.rst.
3688 See Documentation/admin-guide/blockdev/paride.rst.
3690 pirq= [SMP,APIC] Manual mp-table setup
3691 See Documentation/x86/i386/IO-APIC.rst.
3693 plip= [PPT,NET] Parallel port network link
3694 Format: { parport<nr> | timid | 0 }
3695 See also Documentation/admin-guide/parport.rst.
3697 pmtmr= [X86] Manual setup of pmtmr I/O Port.
3698 Override pmtimer IOPort with a hex value.
3702 Enable PNP debug messages (depends on the
3703 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time
3704 via /sys/module/pnp/parameters/debug. We always show
3705 current resource usage; turning this on also shows
3706 possible settings and some assignment information.
3712 { on | off | curr | res | no-curr | no-res }
3715 [ISAPNP] Exclude IRQs for the autoconfiguration
3718 [ISAPNP] Exclude DMAs for the autoconfiguration
3720 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration
3721 Ranges are in pairs (I/O port base and size).
3724 [ISAPNP] Exclude memory regions for the
3726 Ranges are in pairs (memory base and size).
3728 ports= [IP_VS_FTP] IPVS ftp helper module
3730 Up to 8 (IP_VS_APP_MAX_PORTS) ports
3732 Format: <port>,<port>....
3734 powersave=off [PPC] This option disables power saving features.
3735 It specifically disables cpuidle and sets the
3736 platform machine description specific power_save
3737 function to NULL. On Idle the CPU just reduces
3740 ppc_strict_facility_enable
3741 [PPC] This option catches any kernel floating point,
3742 Altivec, VSX and SPE outside of regions specifically
3743 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()).
3744 There is some performance impact when enabling this.
3748 Disable Hardware Transactional Memory
3750 print-fatal-signals=
3751 [KNL] debug: print fatal signals
3753 If enabled, warn about various signal handling
3754 related application anomalies: too many signals,
3755 too many POSIX.1 timers, fatal signals causing a
3758 If you hit the warning due to signal overflow,
3759 you might want to try "ulimit -i unlimited".
3763 printk.always_kmsg_dump=
3764 Trigger kmsg_dump for cases other than kernel oops or
3766 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
3769 printk.devkmsg={on,off,ratelimit}
3770 Control writing to /dev/kmsg.
3771 on - unlimited logging to /dev/kmsg from userspace
3772 off - logging to /dev/kmsg disabled
3773 ratelimit - ratelimit the logging
3776 printk.time= Show timing data prefixed to each printk message line
3777 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
3779 processor.max_cstate= [HW,ACPI]
3780 Limit processor to maximum C-state
3781 max_cstate=9 overrides any DMI blacklist limit.
3783 processor.nocst [HW,ACPI]
3784 Ignore the _CST method to determine C-states,
3785 instead using the legacy FADT method
3787 profile= [KNL] Enable kernel profiling via /proc/profile
3788 Format: [<profiletype>,]<number>
3789 Param: <profiletype>: "schedule", "sleep", or "kvm"
3790 [defaults to kernel profiling]
3791 Param: "schedule" - profile schedule points.
3792 Param: "sleep" - profile D-state sleeping (millisecs).
3793 Requires CONFIG_SCHEDSTATS
3794 Param: "kvm" - profile VM exits.
3795 Param: <number> - step/bucket size as a power of 2 for
3796 statistical time based profiling.
3798 prompt_ramdisk= [RAM] List of RAM disks to prompt for floppy disk
3800 See Documentation/admin-guide/blockdev/ramdisk.rst.
3802 psi= [KNL] Enable or disable pressure stall information
3806 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to
3807 probe for; one of (bare|imps|exps|lifebook|any).
3808 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports
3810 psmouse.resetafter= [HW,MOUSE]
3811 Try to reset the device after so many bad packets
3814 [HW,MOUSE] Set desired mouse resolution, in dpi.
3815 psmouse.smartscroll=
3816 [HW,MOUSE] Controls Logitech smartscroll autorepeat.
3817 0 = disabled, 1 = enabled (default).
3819 pstore.backend= Specify the name of the pstore backend to use
3822 See Documentation/admin-guide/blockdev/paride.rst.
3824 pti= [X86_64] Control Page Table Isolation of user and
3825 kernel address spaces. Disabling this feature
3826 removes hardening, but improves performance of
3827 system calls and interrupts.
3829 on - unconditionally enable
3830 off - unconditionally disable
3831 auto - kernel detects whether your CPU model is
3832 vulnerable to issues that PTI mitigates
3834 Not specifying this option is equivalent to pti=auto.
3837 Equivalent to pti=off
3840 [KNL] Number of legacy pty's. Overwrites compiled-in
3843 quiet [KNL] Disable most log messages
3848 See Documentation/admin-guide/md.rst.
3850 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes
3851 See Documentation/admin-guide/blockdev/ramdisk.rst.
3853 random.trust_cpu={on,off}
3854 [KNL] Enable or disable trusting the use of the
3855 CPU's random number generator (if available) to
3856 fully seed the kernel's CRNG. Default is controlled
3857 by CONFIG_RANDOM_TRUST_CPU.
3859 ras=option[,option,...] [KNL] RAS-specific options
3862 Disable the Correctable Errors Collector,
3863 see CONFIG_RAS_CEC help text.
3866 The argument is a cpu list, as described above,
3867 except that the string "all" can be used to
3868 specify every CPU on the system.
3870 In kernels built with CONFIG_RCU_NOCB_CPU=y, set
3871 the specified list of CPUs to be no-callback CPUs.
3872 Invocation of these CPUs' RCU callbacks will be
3873 offloaded to "rcuox/N" kthreads created for that
3874 purpose, where "x" is "p" for RCU-preempt, and
3875 "s" for RCU-sched, and "N" is the CPU number.
3876 This reduces OS jitter on the offloaded CPUs,
3877 which can be useful for HPC and real-time
3878 workloads. It can also improve energy efficiency
3879 for asymmetric multiprocessors.
3882 Rather than requiring that offloaded CPUs
3883 (specified by rcu_nocbs= above) explicitly
3884 awaken the corresponding "rcuoN" kthreads,
3885 make these kthreads poll for callbacks.
3886 This improves the real-time response for the
3887 offloaded CPUs by relieving them of the need to
3888 wake up the corresponding kthread, but degrades
3889 energy efficiency by requiring that the kthreads
3890 periodically wake up to do the polling.
3892 rcutree.blimit= [KNL]
3893 Set maximum number of finished RCU callbacks to
3894 process in one batch.
3896 rcutree.dump_tree= [KNL]
3897 Dump the structure of the rcu_node combining tree
3898 out at early boot. This is used for diagnostic
3899 purposes, to verify correct tree setup.
3901 rcutree.gp_cleanup_delay= [KNL]
3902 Set the number of jiffies to delay each step of
3903 RCU grace-period cleanup.
3905 rcutree.gp_init_delay= [KNL]
3906 Set the number of jiffies to delay each step of
3907 RCU grace-period initialization.
3909 rcutree.gp_preinit_delay= [KNL]
3910 Set the number of jiffies to delay each step of
3911 RCU grace-period pre-initialization, that is,
3912 the propagation of recent CPU-hotplug changes up
3913 the rcu_node combining tree.
3915 rcutree.use_softirq= [KNL]
3916 If set to zero, move all RCU_SOFTIRQ processing to
3917 per-CPU rcuc kthreads. Defaults to a non-zero
3918 value, meaning that RCU_SOFTIRQ is used by default.
3919 Specify rcutree.use_softirq=0 to use rcuc kthreads.
3921 rcutree.rcu_fanout_exact= [KNL]
3922 Disable autobalancing of the rcu_node combining
3923 tree. This is used by rcutorture, and might
3924 possibly be useful for architectures having high
3925 cache-to-cache transfer latencies.
3927 rcutree.rcu_fanout_leaf= [KNL]
3928 Change the number of CPUs assigned to each
3929 leaf rcu_node structure. Useful for very
3930 large systems, which will choose the value 64,
3931 and for NUMA systems with large remote-access
3932 latencies, which will choose a value aligned
3933 with the appropriate hardware boundaries.
3935 rcutree.jiffies_till_first_fqs= [KNL]
3936 Set delay from grace-period initialization to
3937 first attempt to force quiescent states.
3938 Units are jiffies, minimum value is zero,
3939 and maximum value is HZ.
3941 rcutree.jiffies_till_next_fqs= [KNL]
3942 Set delay between subsequent attempts to force
3943 quiescent states. Units are jiffies, minimum
3944 value is one, and maximum value is HZ.
3946 rcutree.jiffies_till_sched_qs= [KNL]
3947 Set required age in jiffies for a
3948 given grace period before RCU starts
3949 soliciting quiescent-state help from
3950 rcu_note_context_switch() and cond_resched().
3951 If not specified, the kernel will calculate
3952 a value based on the most recent settings
3953 of rcutree.jiffies_till_first_fqs
3954 and rcutree.jiffies_till_next_fqs.
3955 This calculated value may be viewed in
3956 rcutree.jiffies_to_sched_qs. Any attempt to set
3957 rcutree.jiffies_to_sched_qs will be cheerfully
3960 rcutree.kthread_prio= [KNL,BOOT]
3961 Set the SCHED_FIFO priority of the RCU per-CPU
3962 kthreads (rcuc/N). This value is also used for
3963 the priority of the RCU boost threads (rcub/N)
3964 and for the RCU grace-period kthreads (rcu_bh,
3965 rcu_preempt, and rcu_sched). If RCU_BOOST is
3966 set, valid values are 1-99 and the default is 1
3967 (the least-favored priority). Otherwise, when
3968 RCU_BOOST is not set, valid values are 0-99 and
3969 the default is zero (non-realtime operation).
3971 rcutree.rcu_nocb_gp_stride= [KNL]
3972 Set the number of NOCB callback kthreads in
3973 each group, which defaults to the square root
3974 of the number of CPUs. Larger numbers reduce
3975 the wakeup overhead on the global grace-period
3976 kthread, but increases that same overhead on
3977 each group's NOCB grace-period kthread.
3979 rcutree.qhimark= [KNL]
3980 Set threshold of queued RCU callbacks beyond which
3981 batch limiting is disabled.
3983 rcutree.qlowmark= [KNL]
3984 Set threshold of queued RCU callbacks below which
3985 batch limiting is re-enabled.
3987 rcutree.rcu_idle_gp_delay= [KNL]
3988 Set wakeup interval for idle CPUs that have
3989 RCU callbacks (RCU_FAST_NO_HZ=y).
3991 rcutree.rcu_idle_lazy_gp_delay= [KNL]
3992 Set wakeup interval for idle CPUs that have
3993 only "lazy" RCU callbacks (RCU_FAST_NO_HZ=y).
3994 Lazy RCU callbacks are those which RCU can
3995 prove do nothing more than free memory.
3997 rcutree.rcu_kick_kthreads= [KNL]
3998 Cause the grace-period kthread to get an extra
3999 wake_up() if it sleeps three times longer than
4000 it should at force-quiescent-state time.
4001 This wake_up() will be accompanied by a
4002 WARN_ONCE() splat and an ftrace_dump().
4004 rcutree.sysrq_rcu= [KNL]
4005 Commandeer a sysrq key to dump out Tree RCU's
4006 rcu_node tree with an eye towards determining
4007 why a new grace period has not yet started.
4009 rcuperf.gp_async= [KNL]
4010 Measure performance of asynchronous
4011 grace-period primitives such as call_rcu().
4013 rcuperf.gp_async_max= [KNL]
4014 Specify the maximum number of outstanding
4015 callbacks per writer thread. When a writer
4016 thread exceeds this limit, it invokes the
4017 corresponding flavor of rcu_barrier() to allow
4018 previously posted callbacks to drain.
4020 rcuperf.gp_exp= [KNL]
4021 Measure performance of expedited synchronous
4022 grace-period primitives.
4024 rcuperf.holdoff= [KNL]
4025 Set test-start holdoff period. The purpose of
4026 this parameter is to delay the start of the
4027 test until boot completes in order to avoid
4030 rcuperf.kfree_rcu_test= [KNL]
4031 Set to measure performance of kfree_rcu() flooding.
4033 rcuperf.kfree_nthreads= [KNL]
4034 The number of threads running loops of kfree_rcu().
4036 rcuperf.kfree_alloc_num= [KNL]
4037 Number of allocations and frees done in an iteration.
4039 rcuperf.kfree_loops= [KNL]
4040 Number of loops doing rcuperf.kfree_alloc_num number
4041 of allocations and frees.
4043 rcuperf.nreaders= [KNL]
4044 Set number of RCU readers. The value -1 selects
4045 N, where N is the number of CPUs. A value
4046 "n" less than -1 selects N-n+1, where N is again
4047 the number of CPUs. For example, -2 selects N
4048 (the number of CPUs), -3 selects N+1, and so on.
4049 A value of "n" less than or equal to -N selects
4052 rcuperf.nwriters= [KNL]
4053 Set number of RCU writers. The values operate
4054 the same as for rcuperf.nreaders.
4055 N, where N is the number of CPUs
4057 rcuperf.perf_type= [KNL]
4058 Specify the RCU implementation to test.
4060 rcuperf.shutdown= [KNL]
4061 Shut the system down after performance tests
4062 complete. This is useful for hands-off automated
4065 rcuperf.verbose= [KNL]
4066 Enable additional printk() statements.
4068 rcuperf.writer_holdoff= [KNL]
4069 Write-side holdoff between grace periods,
4070 in microseconds. The default of zero says
4073 rcutorture.fqs_duration= [KNL]
4074 Set duration of force_quiescent_state bursts
4077 rcutorture.fqs_holdoff= [KNL]
4078 Set holdoff time within force_quiescent_state bursts
4081 rcutorture.fqs_stutter= [KNL]
4082 Set wait time between force_quiescent_state bursts
4085 rcutorture.fwd_progress= [KNL]
4086 Enable RCU grace-period forward-progress testing
4087 for the types of RCU supporting this notion.
4089 rcutorture.fwd_progress_div= [KNL]
4090 Specify the fraction of a CPU-stall-warning
4091 period to do tight-loop forward-progress testing.
4093 rcutorture.fwd_progress_holdoff= [KNL]
4094 Number of seconds to wait between successive
4095 forward-progress tests.
4097 rcutorture.fwd_progress_need_resched= [KNL]
4098 Enclose cond_resched() calls within checks for
4099 need_resched() during tight-loop forward-progress
4102 rcutorture.gp_cond= [KNL]
4103 Use conditional/asynchronous update-side
4104 primitives, if available.
4106 rcutorture.gp_exp= [KNL]
4107 Use expedited update-side primitives, if available.
4109 rcutorture.gp_normal= [KNL]
4110 Use normal (non-expedited) asynchronous
4111 update-side primitives, if available.
4113 rcutorture.gp_sync= [KNL]
4114 Use normal (non-expedited) synchronous
4115 update-side primitives, if available. If all
4116 of rcutorture.gp_cond=, rcutorture.gp_exp=,
4117 rcutorture.gp_normal=, and rcutorture.gp_sync=
4118 are zero, rcutorture acts as if is interpreted
4119 they are all non-zero.
4121 rcutorture.n_barrier_cbs= [KNL]
4122 Set callbacks/threads for rcu_barrier() testing.
4124 rcutorture.nfakewriters= [KNL]
4125 Set number of concurrent RCU writers. These just
4126 stress RCU, they don't participate in the actual
4127 test, hence the "fake".
4129 rcutorture.nreaders= [KNL]
4130 Set number of RCU readers. The value -1 selects
4131 N-1, where N is the number of CPUs. A value
4132 "n" less than -1 selects N-n-2, where N is again
4133 the number of CPUs. For example, -2 selects N
4134 (the number of CPUs), -3 selects N+1, and so on.
4136 rcutorture.object_debug= [KNL]
4137 Enable debug-object double-call_rcu() testing.
4139 rcutorture.onoff_holdoff= [KNL]
4140 Set time (s) after boot for CPU-hotplug testing.
4142 rcutorture.onoff_interval= [KNL]
4143 Set time (jiffies) between CPU-hotplug operations,
4144 or zero to disable CPU-hotplug testing.
4146 rcutorture.shuffle_interval= [KNL]
4147 Set task-shuffle interval (s). Shuffling tasks
4148 allows some CPUs to go into dyntick-idle mode
4149 during the rcutorture test.
4151 rcutorture.shutdown_secs= [KNL]
4152 Set time (s) after boot system shutdown. This
4153 is useful for hands-off automated testing.
4155 rcutorture.stall_cpu= [KNL]
4156 Duration of CPU stall (s) to test RCU CPU stall
4157 warnings, zero to disable.
4159 rcutorture.stall_cpu_holdoff= [KNL]
4160 Time to wait (s) after boot before inducing stall.
4162 rcutorture.stall_cpu_irqsoff= [KNL]
4163 Disable interrupts while stalling if set.
4165 rcutorture.stat_interval= [KNL]
4166 Time (s) between statistics printk()s.
4168 rcutorture.stutter= [KNL]
4169 Time (s) to stutter testing, for example, specifying
4170 five seconds causes the test to run for five seconds,
4171 wait for five seconds, and so on. This tests RCU's
4172 ability to transition abruptly to and from idle.
4174 rcutorture.test_boost= [KNL]
4175 Test RCU priority boosting? 0=no, 1=maybe, 2=yes.
4176 "Maybe" means test if the RCU implementation
4177 under test support RCU priority boosting.
4179 rcutorture.test_boost_duration= [KNL]
4180 Duration (s) of each individual boost test.
4182 rcutorture.test_boost_interval= [KNL]
4183 Interval (s) between each boost test.
4185 rcutorture.test_no_idle_hz= [KNL]
4186 Test RCU's dyntick-idle handling. See also the
4187 rcutorture.shuffle_interval parameter.
4189 rcutorture.torture_type= [KNL]
4190 Specify the RCU implementation to test.
4192 rcutorture.verbose= [KNL]
4193 Enable additional printk() statements.
4195 rcupdate.rcu_cpu_stall_ftrace_dump= [KNL]
4196 Dump ftrace buffer after reporting RCU CPU
4199 rcupdate.rcu_cpu_stall_suppress= [KNL]
4200 Suppress RCU CPU stall warning messages.
4202 rcupdate.rcu_cpu_stall_timeout= [KNL]
4203 Set timeout for RCU CPU stall warning messages.
4205 rcupdate.rcu_expedited= [KNL]
4206 Use expedited grace-period primitives, for
4207 example, synchronize_rcu_expedited() instead
4208 of synchronize_rcu(). This reduces latency,
4209 but can increase CPU utilization, degrade
4210 real-time latency, and degrade energy efficiency.
4211 No effect on CONFIG_TINY_RCU kernels.
4213 rcupdate.rcu_normal= [KNL]
4214 Use only normal grace-period primitives,
4215 for example, synchronize_rcu() instead of
4216 synchronize_rcu_expedited(). This improves
4217 real-time latency, CPU utilization, and
4218 energy efficiency, but can expose users to
4219 increased grace-period latency. This parameter
4220 overrides rcupdate.rcu_expedited. No effect on
4221 CONFIG_TINY_RCU kernels.
4223 rcupdate.rcu_normal_after_boot= [KNL]
4224 Once boot has completed (that is, after
4225 rcu_end_inkernel_boot() has been invoked), use
4226 only normal grace-period primitives. No effect
4227 on CONFIG_TINY_RCU kernels.
4229 rcupdate.rcu_task_stall_timeout= [KNL]
4230 Set timeout in jiffies for RCU task stall warning
4231 messages. Disable with a value less than or equal
4234 rcupdate.rcu_self_test= [KNL]
4235 Run the RCU early boot self tests
4239 Run specified binary instead of /init from the ramdisk,
4240 used for early userspace startup. See initrd.
4243 force - Override the decision by the kernel to hide the
4244 advertisement of RDRAND support (this affects
4245 certain AMD processors because of buggy BIOS
4246 support, specifically around the suspend/resume
4250 Turn on/off individual RDT features. List is:
4251 cmt, mbmtotal, mbmlocal, l3cat, l3cdp, l2cat, l2cdp,
4253 E.g. to turn on cmt and turn off mba use:
4257 Format (x86 or x86_64):
4258 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] \
4260 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \
4262 Where reboot_mode is one of warm (soft) or cold (hard) or gpio
4263 (prefix with 'panic_' to set mode for panic
4265 reboot_type is one of bios, acpi, kbd, triple, efi, or pci,
4266 reboot_force is either force or not specified,
4267 reboot_cpu is s[mp]#### with #### being the processor
4268 to be used for rebooting.
4271 [KNL, SMP] Set scheduler's default relax_domain_level.
4272 See Documentation/admin-guide/cgroup-v1/cpusets.rst.
4274 reserve= [KNL,BUGS] Force kernel to ignore I/O ports or memory
4275 Format: <base1>,<size1>[,<base2>,<size2>,...]
4276 Reserve I/O ports or memory so the kernel won't use
4277 them. If <base> is less than 0x10000, the region
4278 is assumed to be I/O ports; otherwise it is memory.
4280 reservetop= [X86-32]
4282 Reserves a hole at the top of the kernel virtual
4287 Set the amount of memory to reserve for BIOS at
4288 the bottom of the address space.
4290 reset_devices [KNL] Force drivers to reset the underlying device
4291 during initialization.
4294 Specify the partition device for software suspend
4296 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>}
4298 resume_offset= [SWSUSP]
4299 Specify the offset from the beginning of the partition
4300 given by "resume=" at which the swap header is located,
4301 in <PAGE_SIZE> units (needed only for swap files).
4302 See Documentation/power/swsusp-and-swap-files.rst
4304 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to
4305 read the resume files
4307 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up.
4308 Useful for devices that are detected asynchronously
4309 (e.g. USB and MMC devices).
4311 hibernate= [HIBERNATION]
4312 noresume Don't check if there's a hibernation image
4313 present during boot.
4314 nocompress Don't compress/decompress hibernation images.
4315 no Disable hibernation and resume.
4316 protect_image Turn on image protection during restoration
4317 (that will set all pages holding image data
4318 during restoration read-only).
4320 retain_initrd [RAM] Keep initrd memory after extraction
4322 rfkill.default_state=
4323 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm,
4324 etc. communication is blocked by default.
4327 rfkill.master_switch_mode=
4328 0 The "airplane mode" button does nothing.
4329 1 The "airplane mode" button toggles between everything
4330 blocked and the previous configuration.
4331 2 The "airplane mode" button toggles between everything
4332 blocked and everything unblocked.
4334 rhash_entries= [KNL,NET]
4335 Set number of hash buckets for route cache
4338 [KNL] Disable ring 3 MONITOR/MWAIT feature on supported
4341 ro [KNL] Mount root device read-only on boot
4344 on Mark read-only kernel memory as read-only (default).
4345 off Leave read-only kernel memory writable for debugging.
4348 Enable the uart passthrough on the designated usb port
4349 on Rockchip SoCs. When active, the signals of the
4350 debug-uart get routed to the D+ and D- pins of the usb
4351 port and the regular usb controller gets disabled.
4353 root= [KNL] Root filesystem
4354 See name_to_dev_t comment in init/do_mounts.c.
4356 rootdelay= [KNL] Delay (in seconds) to pause before attempting to
4357 mount the root filesystem
4359 rootflags= [KNL] Set root filesystem mount option string
4361 rootfstype= [KNL] Set root filesystem type
4363 rootwait [KNL] Wait (indefinitely) for root device to show up.
4364 Useful for devices that are detected asynchronously
4365 (e.g. USB and MMC devices).
4367 rproc_mem=nn[KMG][@address]
4368 [KNL,ARM,CMA] Remoteproc physical memory block.
4369 Memory area to be used by remote processor image,
4372 rw [KNL] Mount root device read-write on boot
4374 S [KNL] Run init in single mode
4376 s390_iommu= [HW,S390]
4377 Set s390 IOTLB flushing mode
4379 With strict flushing every unmap operation will result in
4380 an IOTLB flush. Default is lazy flushing before reuse,
4384 See drivers/net/irda/sa1100_ir.c.
4386 sbni= [NET] Granch SBNI12 leased line adapter
4388 sched_debug [KNL] Enables verbose scheduler debug messages.
4390 schedstats= [KNL,X86] Enable or disable scheduled statistics.
4391 Allowed values are enable and disable. This feature
4392 incurs a small amount of overhead in the scheduler
4393 but is useful for debugging and performance tuning.
4395 skew_tick= [KNL] Offset the periodic timer tick per cpu to mitigate
4396 xtime_lock contention on larger systems, and/or RCU lock
4397 contention on all systems with CONFIG_MAXSMP set.
4398 Format: { "0" | "1" }
4399 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1"
4401 Note: increases power consumption, thus should only be
4402 enabled if running jitter sensitive (HPC/RT) workloads.
4404 security= [SECURITY] Choose a legacy "major" security module to
4405 enable at boot. This has been deprecated by the
4408 selinux= [SELINUX] Disable or enable SELinux at boot time.
4409 Format: { "0" | "1" }
4410 See security/selinux/Kconfig help text.
4415 apparmor= [APPARMOR] Disable or enable AppArmor at boot time
4416 Format: { "0" | "1" }
4417 See security/apparmor/Kconfig help text
4420 Default value is set via kernel config option.
4422 serialnumber [BUGS=X86-32]
4425 Maximal number of shapers.
4433 Disable merging of slabs with similar size. May be
4434 necessary if there is some reason to distinguish
4435 allocs to different slabs, especially in hardened
4436 environments where the risk of heap overflows and
4437 layout control by attackers can usually be
4438 frustrated by disabling merging. This will reduce
4439 most of the exposure of a heap attack to a single
4440 cache (risks via metadata attacks are mostly
4441 unchanged). Debug options disable merging on their
4443 For more information see Documentation/vm/slub.rst.
4445 slab_max_order= [MM, SLAB]
4446 Determines the maximum allowed order for slabs.
4447 A high setting may cause OOMs due to memory
4448 fragmentation. Defaults to 1 for systems with
4449 more than 32MB of RAM, 0 otherwise.
4451 slub_debug[=options[,slabs]] [MM, SLUB]
4452 Enabling slub_debug allows one to determine the
4453 culprit if slab objects become corrupted. Enabling
4454 slub_debug can create guard zones around objects and
4455 may poison objects when not in use. Also tracks the
4456 last alloc / free. For more information see
4457 Documentation/vm/slub.rst.
4459 slub_memcg_sysfs= [MM, SLUB]
4460 Determines whether to enable sysfs directories for
4461 memory cgroup sub-caches. 1 to enable, 0 to disable.
4462 The default is determined by CONFIG_SLUB_MEMCG_SYSFS_ON.
4463 Enabling this can lead to a very high number of debug
4464 directories and files being created under
4467 slub_max_order= [MM, SLUB]
4468 Determines the maximum allowed order for slabs.
4469 A high setting may cause OOMs due to memory
4470 fragmentation. For more information see
4471 Documentation/vm/slub.rst.
4473 slub_min_objects= [MM, SLUB]
4474 The minimum number of objects per slab. SLUB will
4475 increase the slab order up to slub_max_order to
4476 generate a sufficiently large slab able to contain
4477 the number of objects indicated. The higher the number
4478 of objects the smaller the overhead of tracking slabs
4479 and the less frequently locks need to be acquired.
4480 For more information see Documentation/vm/slub.rst.
4482 slub_min_order= [MM, SLUB]
4483 Determines the minimum page order for slabs. Must be
4484 lower than slub_max_order.
4485 For more information see Documentation/vm/slub.rst.
4487 slub_nomerge [MM, SLUB]
4488 Same with slab_nomerge. This is supported for legacy.
4489 See slab_nomerge for more information.
4492 Format: <io1>[,<io2>[,...,<io8>]]
4494 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices
4495 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port
4496 smsc-ircc2.ircc_sir= [HW] SIR base I/O port
4497 smsc-ircc2.ircc_fir= [HW] FIR base I/O port
4498 smsc-ircc2.ircc_irq= [HW] IRQ line
4499 smsc-ircc2.ircc_dma= [HW] DMA channel
4500 smsc-ircc2.ircc_transceiver= [HW] Transceiver type:
4501 0: Toshiba Satellite 1800 (GP data pin select)
4502 1: Fast pin select (default)
4505 smt [KNL,S390] Set the maximum number of threads (logical
4506 CPUs) to use per physical CPU on systems capable of
4507 symmetric multithreading (SMT). Will be capped to the
4508 actual hardware limit.
4510 Default: -1 (no limit)
4513 [KNL] Should the soft-lockup detector generate panics.
4516 A nonzero value instructs the soft-lockup detector
4517 to panic the machine when a soft-lockup occurs. This
4518 is also controlled by CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC
4519 which is the respective build-time switch to that
4522 softlockup_all_cpu_backtrace=
4523 [KNL] Should the soft-lockup detector generate
4524 backtraces on all cpus.
4527 sonypi.*= [HW] Sony Programmable I/O Control Device driver
4528 See Documentation/admin-guide/laptops/sonypi.rst
4530 spectre_v2= [X86] Control mitigation of Spectre variant 2
4531 (indirect branch speculation) vulnerability.
4532 The default operation protects the kernel from
4535 on - unconditionally enable, implies
4537 off - unconditionally disable, implies
4539 auto - kernel detects whether your CPU model is
4542 Selecting 'on' will, and 'auto' may, choose a
4543 mitigation method at run time according to the
4544 CPU, the available microcode, the setting of the
4545 CONFIG_RETPOLINE configuration option, and the
4546 compiler with which the kernel was built.
4548 Selecting 'on' will also enable the mitigation
4549 against user space to user space task attacks.
4551 Selecting 'off' will disable both the kernel and
4552 the user space protections.
4554 Specific mitigations can also be selected manually:
4556 retpoline - replace indirect branches
4557 retpoline,generic - google's original retpoline
4558 retpoline,amd - AMD-specific minimal thunk
4560 Not specifying this option is equivalent to
4564 [X86] Control mitigation of Spectre variant 2
4565 (indirect branch speculation) vulnerability between
4568 on - Unconditionally enable mitigations. Is
4569 enforced by spectre_v2=on
4571 off - Unconditionally disable mitigations. Is
4572 enforced by spectre_v2=off
4574 prctl - Indirect branch speculation is enabled,
4575 but mitigation can be enabled via prctl
4576 per thread. The mitigation control state
4577 is inherited on fork.
4580 - Like "prctl" above, but only STIBP is
4581 controlled per thread. IBPB is issued
4582 always when switching between different user
4586 - Same as "prctl" above, but all seccomp
4587 threads will enable the mitigation unless
4588 they explicitly opt out.
4591 - Like "seccomp" above, but only STIBP is
4592 controlled per thread. IBPB is issued
4593 always when switching between different
4594 user space processes.
4596 auto - Kernel selects the mitigation depending on
4597 the available CPU features and vulnerability.
4600 If CONFIG_SECCOMP=y then "seccomp", otherwise "prctl"
4602 Not specifying this option is equivalent to
4603 spectre_v2_user=auto.
4605 spec_store_bypass_disable=
4606 [HW] Control Speculative Store Bypass (SSB) Disable mitigation
4607 (Speculative Store Bypass vulnerability)
4609 Certain CPUs are vulnerable to an exploit against a
4610 a common industry wide performance optimization known
4611 as "Speculative Store Bypass" in which recent stores
4612 to the same memory location may not be observed by
4613 later loads during speculative execution. The idea
4614 is that such stores are unlikely and that they can
4615 be detected prior to instruction retirement at the
4616 end of a particular speculation execution window.
4618 In vulnerable processors, the speculatively forwarded
4619 store can be used in a cache side channel attack, for
4620 example to read memory to which the attacker does not
4621 directly have access (e.g. inside sandboxed code).
4623 This parameter controls whether the Speculative Store
4624 Bypass optimization is used.
4626 On x86 the options are:
4628 on - Unconditionally disable Speculative Store Bypass
4629 off - Unconditionally enable Speculative Store Bypass
4630 auto - Kernel detects whether the CPU model contains an
4631 implementation of Speculative Store Bypass and
4632 picks the most appropriate mitigation. If the
4633 CPU is not vulnerable, "off" is selected. If the
4634 CPU is vulnerable the default mitigation is
4635 architecture and Kconfig dependent. See below.
4636 prctl - Control Speculative Store Bypass per thread
4637 via prctl. Speculative Store Bypass is enabled
4638 for a process by default. The state of the control
4639 is inherited on fork.
4640 seccomp - Same as "prctl" above, but all seccomp threads
4641 will disable SSB unless they explicitly opt out.
4643 Default mitigations:
4644 X86: If CONFIG_SECCOMP=y "seccomp", otherwise "prctl"
4646 On powerpc the options are:
4648 on,auto - On Power8 and Power9 insert a store-forwarding
4649 barrier on kernel entry and exit. On Power7
4650 perform a software flush on kernel entry and
4654 Not specifying this option is equivalent to
4655 spec_store_bypass_disable=auto.
4657 spia_io_base= [HW,MTD]
4662 srcutree.counter_wrap_check [KNL]
4663 Specifies how frequently to check for
4664 grace-period sequence counter wrap for the
4665 srcu_data structure's ->srcu_gp_seq_needed field.
4666 The greater the number of bits set in this kernel
4667 parameter, the less frequently counter wrap will
4668 be checked for. Note that the bottom two bits
4671 srcutree.exp_holdoff [KNL]
4672 Specifies how many nanoseconds must elapse
4673 since the end of the last SRCU grace period for
4674 a given srcu_struct until the next normal SRCU
4675 grace period will be considered for automatic
4676 expediting. Set to zero to disable automatic
4680 Speculative Store Bypass Disable control
4682 On CPUs that are vulnerable to the Speculative
4683 Store Bypass vulnerability and offer a
4684 firmware based mitigation, this parameter
4685 indicates how the mitigation should be used:
4687 force-on: Unconditionally enable mitigation for
4688 for both kernel and userspace
4689 force-off: Unconditionally disable mitigation for
4690 for both kernel and userspace
4691 kernel: Always enable mitigation in the
4692 kernel, and offer a prctl interface
4693 to allow userspace to register its
4694 interest in being mitigated too.
4696 stack_guard_gap= [MM]
4697 override the default stack gap protection. The value
4698 is in page units and it defines how many pages prior
4699 to (for stacks growing down) resp. after (for stacks
4700 growing up) the main stack are reserved for no other
4701 mapping. Default value is 256 pages.
4704 Enabled the stack tracer on boot up.
4706 stacktrace_filter=[function-list]
4707 [FTRACE] Limit the functions that the stack tracer
4708 will trace at boot up. function-list is a comma separated
4709 list of functions. This list can be changed at run
4710 time by the stack_trace_filter file in the debugfs
4711 tracing directory. Note, this enables stack tracing
4712 and the stacktrace above is not needed.
4716 Set the STI (builtin display/keyboard on the HP-PARISC
4717 machines) console (graphic card) which should be used
4718 as the initial boot-console.
4719 See also comment in drivers/video/console/sticore.c.
4722 See comment in drivers/video/console/sticore.c.
4725 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]]
4727 sunrpc.min_resvport=
4728 sunrpc.max_resvport=
4730 SunRPC servers often require that client requests
4731 originate from a privileged port (i.e. a port in the
4732 range 0 < portnr < 1024).
4733 An administrator who wishes to reserve some of these
4734 ports for other uses may adjust the range that the
4735 kernel's sunrpc client considers to be privileged
4736 using these two parameters to set the minimum and
4737 maximum port values.
4739 sunrpc.svc_rpc_per_connection_limit=
4741 Limit the number of requests that the server will
4742 process in parallel from a single connection.
4743 The default value is 0 (no limit).
4747 Control how the NFS server code allocates CPUs to
4748 service thread pools. Depending on how many NICs
4749 you have and where their interrupts are bound, this
4750 option will affect which CPUs will do NFS serving.
4751 Note: this parameter cannot be changed while the
4752 NFS server is running.
4754 auto the server chooses an appropriate mode
4755 automatically using heuristics
4756 global a single global pool contains all CPUs
4757 percpu one pool for each CPU
4758 pernode one pool for each NUMA node (equivalent
4759 to global on non-NUMA machines)
4761 sunrpc.tcp_slot_table_entries=
4762 sunrpc.udp_slot_table_entries=
4764 Sets the upper limit on the number of simultaneous
4765 RPC calls that can be sent from the client to a
4766 server. Increasing these values may allow you to
4767 improve throughput, but will also increase the
4768 amount of memory reserved for use by the client.
4770 suspend.pm_test_delay=
4772 Sets the number of seconds to remain in a suspend test
4773 mode before resuming the system (see
4774 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG
4775 is set. Default value is 5.
4778 Format: { on | off | y | n | 1 | 0 }
4779 This parameter controls use of the Protected
4780 Execution Facility on pSeries.
4783 [KNL] Enable accounting of swap in memory resource
4784 controller if no parameter or 1 is given or disable
4785 it if 0 is given (See Documentation/admin-guide/cgroup-v1/memory.rst)
4787 swiotlb= [ARM,IA-64,PPC,MIPS,X86]
4788 Format: { <int> | force | noforce }
4789 <int> -- Number of I/O TLB slabs
4790 force -- force using of bounce buffers even if they
4791 wouldn't be automatically used by the kernel
4792 noforce -- Never use bounce buffers (for debugging)
4796 sysfs.deprecated=0|1 [KNL]
4797 Enable/disable old style sysfs layout for old udev
4798 on older distributions. When this option is enabled
4799 very new udev will not work anymore. When this option
4800 is disabled (or CONFIG_SYSFS_DEPRECATED not compiled)
4801 in older udev will not work anymore.
4802 Default depends on CONFIG_SYSFS_DEPRECATED_V2 set in
4803 the kernel configuration.
4805 sysrq_always_enabled
4807 Ignore sysrq setting - this boot parameter will
4808 neutralize any effect of /proc/sys/kernel/sysrq.
4809 Useful for debugging.
4811 tcpmhash_entries= [KNL,NET]
4812 Set the number of tcp_metrics_hash slots.
4813 Default value is 8192 or 16384 depending on total
4814 ram pages. This is used to specify the TCP metrics
4815 cache size. See Documentation/networking/ip-sysctl.txt
4816 "tcp_no_metrics_save" section for more details.
4820 test_suspend= [SUSPEND][,N]
4821 Specify "mem" (for Suspend-to-RAM) or "standby" (for
4822 standby suspend) or "freeze" (for suspend type freeze)
4823 as the system sleep state during system startup with
4824 the optional capability to repeat N number of times.
4825 The system is woken from this state using a
4826 wakeup-capable RTC alarm.
4828 thash_entries= [KNL,NET]
4829 Set number of hash buckets for TCP connection
4831 thermal.act= [HW,ACPI]
4832 -1: disable all active trip points in all thermal zones
4833 <degrees C>: override all lowest active trip points
4835 thermal.crt= [HW,ACPI]
4836 -1: disable all critical trip points in all thermal zones
4837 <degrees C>: override all critical trip points
4839 thermal.nocrt= [HW,ACPI]
4840 Set to disable actions on ACPI thermal zone
4841 critical and hot trip points.
4843 thermal.off= [HW,ACPI]
4844 1: disable ACPI thermal control
4846 thermal.psv= [HW,ACPI]
4847 -1: disable all passive trip points
4848 <degrees C>: override all passive trip points to this
4851 thermal.tzp= [HW,ACPI]
4852 Specify global default ACPI thermal zone polling rate
4853 <deci-seconds>: poll all this frequency
4854 0: no polling (default)
4857 Force threading of all interrupt handlers except those
4858 marked explicitly IRQF_NO_THREAD.
4862 Specify if the kernel should make use of the cpu
4863 topology information if the hardware supports this.
4864 The scheduler will make use of this information and
4865 e.g. base its process migration decisions on it.
4868 topology_updates= [KNL, PPC, NUMA]
4870 Specify if the kernel should ignore (off)
4871 topology updates sent by the hypervisor to this
4876 tpm_suspend_pcr=[HW,TPM]
4877 Format: integer pcr id
4878 Specify that at suspend time, the tpm driver
4879 should extend the specified pcr with zeros,
4880 as a workaround for some chips which fail to
4881 flush the last written pcr on TPM_SaveState.
4882 This will guarantee that all the other pcrs
4885 trace_buf_size=nn[KMG]
4886 [FTRACE] will set tracing buffer size on each cpu.
4888 trace_event=[event-list]
4889 [FTRACE] Set and start specified trace events in order
4890 to facilitate early boot debugging. The event-list is a
4891 comma separated list of trace events to enable. See
4892 also Documentation/trace/events.rst
4894 trace_options=[option-list]
4895 [FTRACE] Enable or disable tracer options at boot.
4896 The option-list is a comma delimited list of options
4897 that can be enabled or disabled just as if you were
4898 to echo the option name into
4900 /sys/kernel/debug/tracing/trace_options
4902 For example, to enable stacktrace option (to dump the
4903 stack trace of each event), add to the command line:
4905 trace_options=stacktrace
4907 See also Documentation/trace/ftrace.rst "trace options"
4911 Have the tracepoints sent to printk as well as the
4912 tracing ring buffer. This is useful for early boot up
4913 where the system hangs or reboots and does not give the
4914 option for reading the tracing buffer or performing a
4915 ftrace_dump_on_oops.
4917 To turn off having tracepoints sent to printk,
4918 echo 0 > /proc/sys/kernel/tracepoint_printk
4919 Note, echoing 1 into this file without the
4920 tracepoint_printk kernel cmdline option has no effect.
4924 Having tracepoints sent to printk() and activating high
4925 frequency tracepoints such as irq or sched, can cause
4926 the system to live lock.
4929 [FTRACE] enable this option to disable tracing when a
4930 warning is hit. This turns off "tracing_on". Tracing can
4931 be enabled again by echoing '1' into the "tracing_on"
4932 file located in /sys/kernel/debug/tracing/
4934 This option is useful, as it disables the trace before
4935 the WARNING dump is called, which prevents the trace to
4936 be filled with content caused by the warning output.
4938 This option can also be set at run time via the sysctl
4939 option: kernel/traceoff_on_warning
4941 transparent_hugepage=
4943 Format: [always|madvise|never]
4944 Can be used to control the default behavior of the system
4945 with respect to transparent hugepages.
4946 See Documentation/admin-guide/mm/transhuge.rst
4949 tsc= Disable clocksource stability checks for TSC.
4951 [x86] reliable: mark tsc clocksource as reliable, this
4952 disables clocksource verification at runtime, as well
4953 as the stability checks done at bootup. Used to enable
4954 high-resolution timer mode on older hardware, and in
4955 virtualized environment.
4956 [x86] noirqtime: Do not use TSC to do irq accounting.
4957 Used to run time disable IRQ_TIME_ACCOUNTING on any
4958 platforms where RDTSC is slow and this accounting
4960 [x86] unstable: mark the TSC clocksource as unstable, this
4961 marks the TSC unconditionally unstable at bootup and
4962 avoids any further wobbles once the TSC watchdog notices.
4963 [x86] nowatchdog: disable clocksource watchdog. Used
4964 in situations with strict latency requirements (where
4965 interruptions from clocksource watchdog are not
4968 tsx= [X86] Control Transactional Synchronization
4969 Extensions (TSX) feature in Intel processors that
4970 support TSX control.
4972 This parameter controls the TSX feature. The options are:
4974 on - Enable TSX on the system. Although there are
4975 mitigations for all known security vulnerabilities,
4976 TSX has been known to be an accelerator for
4977 several previous speculation-related CVEs, and
4978 so there may be unknown security risks associated
4979 with leaving it enabled.
4981 off - Disable TSX on the system. (Note that this
4982 option takes effect only on newer CPUs which are
4983 not vulnerable to MDS, i.e., have
4984 MSR_IA32_ARCH_CAPABILITIES.MDS_NO=1 and which get
4985 the new IA32_TSX_CTRL MSR through a microcode
4986 update. This new MSR allows for the reliable
4987 deactivation of the TSX functionality.)
4989 auto - Disable TSX if X86_BUG_TAA is present,
4990 otherwise enable TSX on the system.
4992 Not specifying this option is equivalent to tsx=off.
4994 See Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
4997 tsx_async_abort= [X86,INTEL] Control mitigation for the TSX Async
4998 Abort (TAA) vulnerability.
5000 Similar to Micro-architectural Data Sampling (MDS)
5001 certain CPUs that support Transactional
5002 Synchronization Extensions (TSX) are vulnerable to an
5003 exploit against CPU internal buffers which can forward
5004 information to a disclosure gadget under certain
5007 In vulnerable processors, the speculatively forwarded
5008 data can be used in a cache side channel attack, to
5009 access data to which the attacker does not have direct
5012 This parameter controls the TAA mitigation. The
5015 full - Enable TAA mitigation on vulnerable CPUs
5018 full,nosmt - Enable TAA mitigation and disable SMT on
5019 vulnerable CPUs. If TSX is disabled, SMT
5020 is not disabled because CPU is not
5021 vulnerable to cross-thread TAA attacks.
5022 off - Unconditionally disable TAA mitigation
5024 On MDS-affected machines, tsx_async_abort=off can be
5025 prevented by an active MDS mitigation as both vulnerabilities
5026 are mitigated with the same mechanism so in order to disable
5027 this mitigation, you need to specify mds=off too.
5029 Not specifying this option is equivalent to
5030 tsx_async_abort=full. On CPUs which are MDS affected
5031 and deploy MDS mitigation, TAA mitigation is not
5032 required and doesn't provide any additional
5036 Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
5038 turbografx.map[2|3]= [HW,JOY]
5039 TurboGraFX parallel port interface
5041 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7>
5042 See also Documentation/input/devices/joystick-parport.rst
5044 udbg-immortal [PPC] When debugging early kernel crashes that
5045 happen after console_init() and before a proper
5046 console driver takes over, this boot options might
5047 help "seeing" what's going on.
5049 uhash_entries= [KNL,NET]
5050 Set number of hash buckets for UDP/UDP-Lite connections
5053 [USB] Ignore overcurrent events (default N).
5054 Some badly-designed motherboards generate lots of
5055 bogus events, for ports that aren't wired to
5056 anything. Set this parameter to avoid log spamming.
5057 Note that genuine overcurrent events won't be
5061 [X86] Cause panic on unknown NMI.
5063 usbcore.authorized_default=
5064 [USB] Default USB device authorization:
5065 (default -1 = authorized except for wireless USB,
5066 0 = not authorized, 1 = authorized, 2 = authorized
5067 if device connected to internal port)
5069 usbcore.autosuspend=
5070 [USB] The autosuspend time delay (in seconds) used
5071 for newly-detected USB devices (default 2). This
5072 is the time required before an idle device will be
5073 autosuspended. Devices for which the delay is set
5074 to a negative value won't be autosuspended at all.
5076 usbcore.usbfs_snoop=
5077 [USB] Set to log all usbfs traffic (default 0 = off).
5079 usbcore.usbfs_snoop_max=
5080 [USB] Maximum number of bytes to snoop in each URB
5083 usbcore.blinkenlights=
5084 [USB] Set to cycle leds on hubs (default 0 = off).
5086 usbcore.old_scheme_first=
5087 [USB] Start with the old device initialization
5088 scheme, applies only to low and full-speed devices
5091 usbcore.usbfs_memory_mb=
5092 [USB] Memory limit (in MB) for buffers allocated by
5093 usbfs (default = 16, 0 = max = 2047).
5095 usbcore.use_both_schemes=
5096 [USB] Try the other device initialization scheme
5097 if the first one fails (default 1 = enabled).
5099 usbcore.initial_descriptor_timeout=
5100 [USB] Specifies timeout for the initial 64-byte
5101 USB_REQ_GET_DESCRIPTOR request in milliseconds
5102 (default 5000 = 5.0 seconds).
5104 usbcore.nousb [USB] Disable the USB subsystem
5107 [USB] A list of quirk entries to augment the built-in
5108 usb core quirk list. List entries are separated by
5109 commas. Each entry has the form
5110 VendorID:ProductID:Flags. The IDs are 4-digit hex
5111 numbers and Flags is a set of letters. Each letter
5112 will change the built-in quirk; setting it if it is
5113 clear and clearing it if it is set. The letters have
5114 the following meanings:
5115 a = USB_QUIRK_STRING_FETCH_255 (string
5116 descriptors must not be fetched using
5118 b = USB_QUIRK_RESET_RESUME (device can't resume
5119 correctly so reset it instead);
5120 c = USB_QUIRK_NO_SET_INTF (device can't handle
5121 Set-Interface requests);
5122 d = USB_QUIRK_CONFIG_INTF_STRINGS (device can't
5123 handle its Configuration or Interface
5125 e = USB_QUIRK_RESET (device can't be reset
5126 (e.g morph devices), don't use reset);
5127 f = USB_QUIRK_HONOR_BNUMINTERFACES (device has
5128 more interface descriptions than the
5129 bNumInterfaces count, and can't handle
5130 talking to these interfaces);
5131 g = USB_QUIRK_DELAY_INIT (device needs a pause
5132 during initialization, after we read
5133 the device descriptor);
5134 h = USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL (For
5135 high speed and super speed interrupt
5136 endpoints, the USB 2.0 and USB 3.0 spec
5137 require the interval in microframes (1
5138 microframe = 125 microseconds) to be
5139 calculated as interval = 2 ^
5141 Devices with this quirk report their
5142 bInterval as the result of this
5143 calculation instead of the exponent
5144 variable used in the calculation);
5145 i = USB_QUIRK_DEVICE_QUALIFIER (device can't
5146 handle device_qualifier descriptor
5148 j = USB_QUIRK_IGNORE_REMOTE_WAKEUP (device
5149 generates spurious wakeup, ignore
5150 remote wakeup capability);
5151 k = USB_QUIRK_NO_LPM (device can't handle Link
5153 l = USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL
5154 (Device reports its bInterval as linear
5155 frames instead of the USB 2.0
5157 m = USB_QUIRK_DISCONNECT_SUSPEND (Device needs
5158 to be disconnected before suspend to
5159 prevent spurious wakeup);
5160 n = USB_QUIRK_DELAY_CTRL_MSG (Device needs a
5161 pause after every control message);
5162 o = USB_QUIRK_HUB_SLOW_RESET (Hub needs extra
5163 delay after resetting its port);
5164 Example: quirks=0781:5580:bk,0a5c:5834:gij
5167 [USBHID] The interval which mice are to be polled at.
5170 [USBHID] The interval which joysticks are to be polled at.
5173 [USBHID] The interval which keyboards are to be polled at.
5175 usb-storage.delay_use=
5176 [UMS] The delay in seconds before a new device is
5177 scanned for Logical Units (default 1).
5180 [UMS] A list of quirks entries to supplement or
5181 override the built-in unusual_devs list. List
5182 entries are separated by commas. Each entry has
5183 the form VID:PID:Flags where VID and PID are Vendor
5184 and Product ID values (4-digit hex numbers) and
5185 Flags is a set of characters, each corresponding
5186 to a common usb-storage quirk flag as follows:
5187 a = SANE_SENSE (collect more than 18 bytes
5188 of sense data, not on uas);
5189 b = BAD_SENSE (don't collect more than 18
5190 bytes of sense data, not on uas);
5191 c = FIX_CAPACITY (decrease the reported
5192 device capacity by one sector);
5193 d = NO_READ_DISC_INFO (don't use
5194 READ_DISC_INFO command, not on uas);
5195 e = NO_READ_CAPACITY_16 (don't use
5196 READ_CAPACITY_16 command);
5197 f = NO_REPORT_OPCODES (don't use report opcodes
5199 g = MAX_SECTORS_240 (don't transfer more than
5200 240 sectors at a time, uas only);
5201 h = CAPACITY_HEURISTICS (decrease the
5202 reported device capacity by one
5203 sector if the number is odd);
5204 i = IGNORE_DEVICE (don't bind to this
5206 j = NO_REPORT_LUNS (don't use report luns
5208 l = NOT_LOCKABLE (don't try to lock and
5209 unlock ejectable media, not on uas);
5210 m = MAX_SECTORS_64 (don't transfer more
5211 than 64 sectors = 32 KB at a time,
5213 n = INITIAL_READ10 (force a retry of the
5214 initial READ(10) command, not on uas);
5215 o = CAPACITY_OK (accept the capacity
5216 reported by the device, not on uas);
5217 p = WRITE_CACHE (the device cache is ON
5218 by default, not on uas);
5219 r = IGNORE_RESIDUE (the device reports
5220 bogus residue values, not on uas);
5221 s = SINGLE_LUN (the device has only one
5223 t = NO_ATA_1X (don't allow ATA(12) and ATA(16)
5224 commands, uas only);
5225 u = IGNORE_UAS (don't bind to the uas driver);
5226 w = NO_WP_DETECT (don't test whether the
5227 medium is write-protected).
5228 y = ALWAYS_SYNC (issue a SYNCHRONIZE_CACHE
5229 even if the device claims no cache,
5231 Example: quirks=0419:aaf5:rl,0421:0433:rc
5233 user_debug= [KNL,ARM]
5235 See arch/arm/Kconfig.debug help text.
5236 1 - undefined instruction events
5238 4 - invalid data aborts
5241 Example: user_debug=31
5244 [X86] Flags controlling user PTE allocations.
5246 nohigh = do not allocate PTE pages in
5247 HIGHMEM regardless of setting
5251 On X86_32, this is an alias for vdso32=. Otherwise:
5253 vdso=1: enable VDSO (the default)
5254 vdso=0: disable VDSO mapping
5256 vdso32= [X86] Control the 32-bit vDSO
5257 vdso32=1: enable 32-bit VDSO
5258 vdso32=0 or vdso32=2: disable 32-bit VDSO
5260 See the help text for CONFIG_COMPAT_VDSO for more
5261 details. If CONFIG_COMPAT_VDSO is set, the default is
5262 vdso32=0; otherwise, the default is vdso32=1.
5264 For compatibility with older kernels, vdso32=2 is an
5267 Try vdso32=0 if you encounter an error that says:
5268 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed!
5271 vector=percpu: enable percpu vector domain
5273 video= [FB] Frame buffer configuration
5274 See Documentation/fb/modedb.rst.
5276 video.brightness_switch_enabled= [0,1]
5277 If set to 1, on receiving an ACPI notify event
5278 generated by hotkey, video driver will adjust brightness
5279 level and then send out the event to user space through
5280 the allocated input device; If set to 0, video driver
5281 will only send out the event without touching backlight
5286 [VMMIO] Memory mapped virtio (platform) device.
5288 <size>@<baseaddr>:<irq>[:<id>]
5290 <size> := size (can use standard suffixes
5292 <baseaddr> := physical base address
5293 <irq> := interrupt number (as passed to
5295 <id> := (optional) platform device id
5297 virtio_mmio.device=1K@0x100b0000:48:7
5299 Can be used multiple times for multiple devices.
5301 vga= [BOOT,X86-32] Select a particular video mode
5302 See Documentation/x86/boot.rst and
5303 Documentation/admin-guide/svga.rst.
5304 Use vga=ask for menu.
5305 This is actually a boot loader parameter; the value is
5306 passed to the kernel using a special protocol.
5308 vm_debug[=options] [KNL] Available with CONFIG_DEBUG_VM=y.
5309 May slow down system boot speed, especially when
5310 enabled on systems with a large amount of memory.
5311 All options are enabled by default, and this
5312 interface is meant to allow for selectively
5313 enabling or disabling specific virtual memory
5316 Available options are:
5317 P Enable page structure init time poisoning
5318 - Disable all of the above options
5320 vmalloc=nn[KMG] [KNL,BOOT] Forces the vmalloc area to have an exact
5321 size of <nn>. This can be used to increase the
5322 minimum size (128MB on x86). It can also be used to
5323 decrease the size and leave more room for directly
5326 vmcp_cma=nn[MG] [KNL,S390]
5327 Sets the memory size reserved for contiguous memory
5328 allocations for the vmcp device driver.
5330 vmhalt= [KNL,S390] Perform z/VM CP command after system halt.
5333 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic.
5336 vmpoff= [KNL,S390] Perform z/VM CP command after power off.
5340 Controls the behavior of vsyscalls (i.e. calls to
5341 fixed addresses of 0xffffffffff600x00 from legacy
5342 code). Most statically-linked binaries and older
5343 versions of glibc use these calls. Because these
5344 functions are at fixed addresses, they make nice
5345 targets for exploits that can control RIP.
5347 emulate [default] Vsyscalls turn into traps and are
5348 emulated reasonably safely. The vsyscall
5351 xonly Vsyscalls turn into traps and are
5352 emulated reasonably safely. The vsyscall
5353 page is not readable.
5355 none Vsyscalls don't work at all. This makes
5356 them quite hard to use for exploits but
5357 might break your system.
5359 vt.color= [VT] Default text color.
5360 Format: 0xYX, X = foreground, Y = background.
5361 Default: 0x07 = light gray on black.
5363 vt.cur_default= [VT] Default cursor shape.
5364 Format: 0xCCBBAA, where AA, BB, and CC are the same as
5365 the parameters of the <Esc>[?A;B;Cc escape sequence;
5366 see VGA-softcursor.txt. Default: 2 = underline.
5368 vt.default_blu= [VT]
5369 Format: <blue0>,<blue1>,<blue2>,...,<blue15>
5370 Change the default blue palette of the console.
5371 This is a 16-member array composed of values
5374 vt.default_grn= [VT]
5375 Format: <green0>,<green1>,<green2>,...,<green15>
5376 Change the default green palette of the console.
5377 This is a 16-member array composed of values
5380 vt.default_red= [VT]
5381 Format: <red0>,<red1>,<red2>,...,<red15>
5382 Change the default red palette of the console.
5383 This is a 16-member array composed of values
5389 Set system-wide default UTF-8 mode for all tty's.
5390 Default is 1, i.e. UTF-8 mode is enabled for all
5391 newly opened terminals.
5393 vt.global_cursor_default=
5396 Set system-wide default for whether a cursor
5397 is shown on new VTs. Default is -1,
5398 i.e. cursors will be created by default unless
5399 overridden by individual drivers. 0 will hide
5400 cursors, 1 will display them.
5402 vt.italic= [VT] Default color for italic text; 0-15.
5405 vt.underline= [VT] Default color for underlined text; 0-15.
5408 watchdog timers [HW,WDT] For information on watchdog timers,
5409 see Documentation/watchdog/watchdog-parameters.rst
5410 or other driver-specific files in the
5411 Documentation/watchdog/ directory.
5415 Set the hard lockup detector stall duration
5416 threshold in seconds. The soft lockup detector
5417 threshold is set to twice the value. A value of 0
5418 disables both lockup detectors. Default is 10
5421 workqueue.watchdog_thresh=
5422 If CONFIG_WQ_WATCHDOG is configured, workqueue can
5423 warn stall conditions and dump internal state to
5424 help debugging. 0 disables workqueue stall
5425 detection; otherwise, it's the stall threshold
5426 duration in seconds. The default value is 30 and
5427 it can be updated at runtime by writing to the
5428 corresponding sysfs file.
5430 workqueue.disable_numa
5431 By default, all work items queued to unbound
5432 workqueues are affine to the NUMA nodes they're
5433 issued on, which results in better behavior in
5434 general. If NUMA affinity needs to be disabled for
5435 whatever reason, this option can be used. Note
5436 that this also can be controlled per-workqueue for
5437 workqueues visible under /sys/bus/workqueue/.
5439 workqueue.power_efficient
5440 Per-cpu workqueues are generally preferred because
5441 they show better performance thanks to cache
5442 locality; unfortunately, per-cpu workqueues tend to
5443 be more power hungry than unbound workqueues.
5445 Enabling this makes the per-cpu workqueues which
5446 were observed to contribute significantly to power
5447 consumption unbound, leading to measurably lower
5448 power usage at the cost of small performance
5451 The default value of this parameter is determined by
5452 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT.
5454 workqueue.debug_force_rr_cpu
5455 Workqueue used to implicitly guarantee that work
5456 items queued without explicit CPU specified are put
5457 on the local CPU. This guarantee is no longer true
5458 and while local CPU is still preferred work items
5459 may be put on foreign CPUs. This debug option
5460 forces round-robin CPU selection to flush out
5461 usages which depend on the now broken guarantee.
5462 When enabled, memory and cache locality will be
5465 x2apic_phys [X86-64,APIC] Use x2apic physical mode instead of
5466 default x2apic cluster mode on platforms
5469 x86_intel_mid_timer= [X86-32,APBT]
5470 Choose timer option for x86 Intel MID platform.
5471 Two valid options are apbt timer only and lapic timer
5472 plus one apbt timer for broadcast timer.
5473 x86_intel_mid_timer=apbt_only | lapic_and_apbt
5475 xen_512gb_limit [KNL,X86-64,XEN]
5476 Restricts the kernel running paravirtualized under Xen
5477 to use only up to 512 GB of RAM. The reason to do so is
5478 crash analysis tools and Xen tools for doing domain
5479 save/restore/migration must be enabled to handle larger
5482 xen_emul_unplug= [HW,X86,XEN]
5483 Unplug Xen emulated devices
5484 Format: [unplug0,][unplug1]
5485 ide-disks -- unplug primary master IDE devices
5486 aux-ide-disks -- unplug non-primary-master IDE devices
5487 nics -- unplug network devices
5488 all -- unplug all emulated devices (NICs and IDE disks)
5489 unnecessary -- unplugging emulated devices is
5490 unnecessary even if the host did not respond to
5492 never -- do not unplug even if version check succeeds
5494 xen_legacy_crash [X86,XEN]
5495 Crash from Xen panic notifier, without executing late
5496 panic() code such as dumping handler.
5498 xen_nopvspin [X86,XEN]
5499 Disables the ticketlock slowpath using Xen PV
5503 Disables the PV optimizations forcing the HVM guest to
5504 run as generic HVM guest with no PV drivers.
5505 This option is obsoleted by the "nopv" option, which
5506 has equivalent effect for XEN platform.
5508 xen_scrub_pages= [XEN]
5509 Boolean option to control scrubbing pages before giving them back
5510 to Xen, for use by other domains. Can be also changed at runtime
5511 with /sys/devices/system/xen_memory/xen_memory0/scrub_pages.
5512 Default value controlled with CONFIG_XEN_SCRUB_PAGES_DEFAULT.
5514 xen_timer_slop= [X86-64,XEN]
5515 Set the timer slop (in nanoseconds) for the virtual Xen
5516 timers (default is 100000). This adjusts the minimum
5517 delta of virtualized Xen timers, where lower values
5518 improve timer resolution at the expense of processing
5519 more timer interrupts.
5521 nopv= [X86,XEN,KVM,HYPER_V,VMWARE]
5522 Disables the PV optimizations forcing the guest to run
5523 as generic guest with no PV drivers. Currently support
5524 XEN HVM, KVM, HYPER_V and VMWARE guest.
5526 xirc2ps_cs= [NET,PCMCIA]
5528 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]]
5531 By default on POWER9 and above, the kernel will
5532 natively use the XIVE interrupt controller. This option
5533 allows the fallback firmware mode to be used:
5535 off Fallback to firmware control of XIVE interrupt
5536 controller on both pseries and powernv
5537 platforms. Only useful on POWER9 and above.
5539 xhci-hcd.quirks [USB,KNL]
5540 A hex value specifying bitmask with supplemental xhci
5541 host controller quirks. Meaning of each bit can be
5542 consulted in header drivers/usb/host/xhci.h.
5545 Format: { early | on | rw | ro | off }
5546 Controls if xmon debugger is enabled. Default is off.
5547 Passing only "xmon" is equivalent to "xmon=early".
5548 early Call xmon as early as possible on boot; xmon
5549 debugger is called from setup_arch().
5550 on xmon debugger hooks will be installed so xmon
5551 is only called on a kernel crash. Default mode,
5552 i.e. either "ro" or "rw" mode, is controlled
5553 with CONFIG_XMON_DEFAULT_RO_MODE.
5554 rw xmon debugger hooks will be installed so xmon
5555 is called only on a kernel crash, mode is write,
5556 meaning SPR registers, memory and, other data
5557 can be written using xmon commands.
5558 ro same as "rw" option above but SPR registers,
5559 memory, and other data can't be written using
5561 off xmon is disabled.