1 acpi= [HW,ACPI,X86,ARM64]
2 Advanced Configuration and Power Interface
3 Format: { force | on | off | strict | noirq | rsdt |
5 force -- enable ACPI if default was off
6 on -- enable ACPI but allow fallback to DT [arm64]
7 off -- disable ACPI if default was on
8 noirq -- do not use ACPI for IRQ routing
9 strict -- Be less tolerant of platforms that are not
10 strictly ACPI specification compliant.
11 rsdt -- prefer RSDT over (default) XSDT
12 copy_dsdt -- copy DSDT to memory
13 For ARM64, ONLY "acpi=off", "acpi=on" or "acpi=force"
16 See also Documentation/power/runtime_pm.rst, pci=noacpi
18 acpi_apic_instance= [ACPI, IOAPIC]
20 2: use 2nd APIC table, if available
21 1,0: use 1st APIC table
24 acpi_backlight= [HW,ACPI]
25 { vendor | video | native | none }
26 If set to vendor, prefer vendor-specific driver
27 (e.g. thinkpad_acpi, sony_acpi, etc.) instead
28 of the ACPI video.ko driver.
29 If set to video, use the ACPI video.ko driver.
30 If set to native, use the device's native backlight mode.
31 If set to none, disable the ACPI backlight interface.
33 acpi_force_32bit_fadt_addr
34 force FADT to use 32 bit addresses rather than the
35 64 bit X_* addresses. Some firmware have broken 64
36 bit addresses for force ACPI ignore these and use
37 the older legacy 32 bit addresses.
39 acpica_no_return_repair [HW, ACPI]
40 Disable AML predefined validation mechanism
41 This mechanism can repair the evaluation result to make
42 the return objects more ACPI specification compliant.
43 This option is useful for developers to identify the
44 root cause of an AML interpreter issue when the issue
45 has something to do with the repair mechanism.
47 acpi.debug_layer= [HW,ACPI,ACPI_DEBUG]
48 acpi.debug_level= [HW,ACPI,ACPI_DEBUG]
50 CONFIG_ACPI_DEBUG must be enabled to produce any ACPI
51 debug output. Bits in debug_layer correspond to a
52 _COMPONENT in an ACPI source file, e.g.,
53 #define _COMPONENT ACPI_PCI_COMPONENT
54 Bits in debug_level correspond to a level in
55 ACPI_DEBUG_PRINT statements, e.g.,
56 ACPI_DEBUG_PRINT((ACPI_DB_INFO, ...
57 The debug_level mask defaults to "info". See
58 Documentation/firmware-guide/acpi/debug.rst for more information about
59 debug layers and levels.
61 Enable processor driver info messages:
62 acpi.debug_layer=0x20000000
63 Enable PCI/PCI interrupt routing info messages:
64 acpi.debug_layer=0x400000
65 Enable AML "Debug" output, i.e., stores to the Debug
66 object while interpreting AML:
67 acpi.debug_layer=0xffffffff acpi.debug_level=0x2
68 Enable all messages related to ACPI hardware:
69 acpi.debug_layer=0x2 acpi.debug_level=0xffffffff
71 Some values produce so much output that the system is
72 unusable. The "log_buf_len" parameter may be useful
73 if you need to capture more output.
75 acpi_enforce_resources= [ACPI]
77 Check for resource conflicts between native drivers
78 and ACPI OperationRegions (SystemIO and SystemMemory
79 only). IO ports and memory declared in ACPI might be
80 used by the ACPI subsystem in arbitrary AML code and
81 can interfere with legacy drivers.
82 strict (default): access to resources claimed by ACPI
83 is denied; legacy drivers trying to access reserved
84 resources will fail to bind to device using them.
85 lax: access to resources claimed by ACPI is allowed;
86 legacy drivers trying to access reserved resources
87 will bind successfully but a warning message is logged.
88 no: ACPI OperationRegions are not marked as reserved,
89 no further checks are performed.
91 acpi_force_table_verification [HW,ACPI]
92 Enable table checksum verification during early stage.
93 By default, this is disabled due to x86 early mapping
96 acpi_irq_balance [HW,ACPI]
97 ACPI will balance active IRQs
100 acpi_irq_nobalance [HW,ACPI]
101 ACPI will not move active IRQs (default)
104 acpi_irq_isa= [HW,ACPI] If irq_balance, mark listed IRQs used by ISA
105 Format: <irq>,<irq>...
107 acpi_irq_pci= [HW,ACPI] If irq_balance, clear listed IRQs for
109 Format: <irq>,<irq>...
111 acpi_mask_gpe= [HW,ACPI]
112 Due to the existence of _Lxx/_Exx, some GPEs triggered
113 by unsupported hardware/firmware features can result in
114 GPE floodings that cannot be automatically disabled by
116 This facility can be used to prevent such uncontrolled
120 acpi_no_auto_serialize [HW,ACPI]
121 Disable auto-serialization of AML methods
122 AML control methods that contain the opcodes to create
123 named objects will be marked as "Serialized" by the
124 auto-serialization feature.
125 This feature is enabled by default.
126 This option allows to turn off the feature.
128 acpi_no_memhotplug [ACPI] Disable memory hotplug. Useful for kdump
131 acpi_no_static_ssdt [HW,ACPI]
132 Disable installation of static SSDTs at early boot time
133 By default, SSDTs contained in the RSDT/XSDT will be
134 installed automatically and they will appear under
135 /sys/firmware/acpi/tables.
136 This option turns off this feature.
137 Note that specifying this option does not affect
138 dynamic table installation which will install SSDT
139 tables to /sys/firmware/acpi/tables/dynamic.
141 acpi_no_watchdog [HW,ACPI,WDT]
142 Ignore the ACPI-based watchdog interface (WDAT) and let
143 a native driver control the watchdog device instead.
145 acpi_rsdp= [ACPI,EFI,KEXEC]
146 Pass the RSDP address to the kernel, mostly used
147 on machines running EFI runtime service to boot the
148 second kernel for kdump.
150 acpi_os_name= [HW,ACPI] Tell ACPI BIOS the name of the OS
151 Format: To spoof as Windows 98: ="Microsoft Windows"
153 acpi_rev_override [ACPI] Override the _REV object to return 5 (instead
154 of 2 which is mandated by ACPI 6) as the supported ACPI
155 specification revision (when using this switch, it may
156 be necessary to carry out a cold reboot _twice_ in a
157 row to make it take effect on the platform firmware).
159 acpi_osi= [HW,ACPI] Modify list of supported OS interface strings
160 acpi_osi="string1" # add string1
161 acpi_osi="!string2" # remove string2
162 acpi_osi=!* # remove all strings
163 acpi_osi=! # disable all built-in OS vendor
165 acpi_osi=!! # enable all built-in OS vendor
167 acpi_osi= # disable all strings
169 'acpi_osi=!' can be used in combination with single or
170 multiple 'acpi_osi="string1"' to support specific OS
171 vendor string(s). Note that such command can only
172 affect the default state of the OS vendor strings, thus
173 it cannot affect the default state of the feature group
174 strings and the current state of the OS vendor strings,
175 specifying it multiple times through kernel command line
176 is meaningless. This command is useful when one do not
177 care about the state of the feature group strings which
178 should be controlled by the OSPM.
180 1. 'acpi_osi=! acpi_osi="Windows 2000"' is equivalent
181 to 'acpi_osi="Windows 2000" acpi_osi=!', they all
182 can make '_OSI("Windows 2000")' TRUE.
184 'acpi_osi=' cannot be used in combination with other
185 'acpi_osi=' command lines, the _OSI method will not
186 exist in the ACPI namespace. NOTE that such command can
187 only affect the _OSI support state, thus specifying it
188 multiple times through kernel command line is also
191 1. 'acpi_osi=' can make 'CondRefOf(_OSI, Local1)'
194 'acpi_osi=!*' can be used in combination with single or
195 multiple 'acpi_osi="string1"' to support specific
196 string(s). Note that such command can affect the
197 current state of both the OS vendor strings and the
198 feature group strings, thus specifying it multiple times
199 through kernel command line is meaningful. But it may
200 still not able to affect the final state of a string if
201 there are quirks related to this string. This command
202 is useful when one want to control the state of the
203 feature group strings to debug BIOS issues related to
206 1. 'acpi_osi="Module Device" acpi_osi=!*' can make
207 '_OSI("Module Device")' FALSE.
208 2. 'acpi_osi=!* acpi_osi="Module Device"' can make
209 '_OSI("Module Device")' TRUE.
210 3. 'acpi_osi=! acpi_osi=!* acpi_osi="Windows 2000"' is
212 'acpi_osi=!* acpi_osi=! acpi_osi="Windows 2000"'
214 'acpi_osi=!* acpi_osi="Windows 2000" acpi_osi=!',
215 they all will make '_OSI("Windows 2000")' TRUE.
218 Override the pmtimer bug detection: force the kernel
219 to assume that this machine's pmtimer latches its value
220 and always returns good values.
222 acpi_sci= [HW,ACPI] ACPI System Control Interrupt trigger mode
223 Format: { level | edge | high | low }
225 acpi_skip_timer_override [HW,ACPI]
226 Recognize and ignore IRQ0/pin2 Interrupt Override.
227 For broken nForce2 BIOS resulting in XT-PIC timer.
229 acpi_sleep= [HW,ACPI] Sleep options
230 Format: { s3_bios, s3_mode, s3_beep, s4_nohwsig,
231 old_ordering, nonvs, sci_force_enable, nobl }
232 See Documentation/power/video.rst for information on
234 s3_beep is for debugging; it makes the PC's speaker beep
235 as soon as the kernel's real-mode entry point is called.
236 s4_nohwsig prevents ACPI hardware signature from being
237 used during resume from hibernation.
238 old_ordering causes the ACPI 1.0 ordering of the _PTS
239 control method, with respect to putting devices into
240 low power states, to be enforced (the ACPI 2.0 ordering
241 of _PTS is used by default).
242 nonvs prevents the kernel from saving/restoring the
243 ACPI NVS memory during suspend/hibernation and resume.
244 sci_force_enable causes the kernel to set SCI_EN directly
245 on resume from S1/S3 (which is against the ACPI spec,
246 but some broken systems don't work without it).
247 nobl causes the internal blacklist of systems known to
248 behave incorrectly in some ways with respect to system
249 suspend and resume to be ignored (use wisely).
251 acpi_use_timer_override [HW,ACPI]
252 Use timer override. For some broken Nvidia NF5 boards
253 that require a timer override, but don't have HPET
255 add_efi_memmap [EFI; X86] Include EFI memory map in
256 kernel's map of available physical RAM.
259 { off | try_unsupported }
260 off: disable AGP support
261 try_unsupported: try to drive unsupported chipsets
262 (may crash computer or cause data corruption)
265 See Documentation/sound/alsa-configuration.rst
268 Allow the default userspace alignment fault handler
269 behaviour to be specified. Bit 0 enables warnings,
270 bit 1 enables fixups, and bit 2 sends a segfault.
272 align_va_addr= [X86-64]
273 Align virtual addresses by clearing slice [14:12] when
274 allocating a VMA at process creation time. This option
275 gives you up to 3% performance improvement on AMD F15h
276 machines (where it is enabled by default) for a
277 CPU-intensive style benchmark, and it can vary highly in
278 a microbenchmark depending on workload and compiler.
280 32: only for 32-bit processes
281 64: only for 64-bit processes
282 on: enable for both 32- and 64-bit processes
283 off: disable for both 32- and 64-bit processes
285 alloc_snapshot [FTRACE]
286 Allocate the ftrace snapshot buffer on boot up when the
287 main buffer is allocated. This is handy if debugging
288 and you need to use tracing_snapshot() on boot up, and
289 do not want to use tracing_snapshot_alloc() as it needs
290 to be done where GFP_KERNEL allocations are allowed.
292 amd_iommu= [HW,X86-64]
293 Pass parameters to the AMD IOMMU driver in the system.
295 fullflush - enable flushing of IO/TLB entries when
296 they are unmapped. Otherwise they are
297 flushed before they will be reused, which
299 off - do not initialize any AMD IOMMU found in
301 force_isolation - Force device isolation for all
302 devices. The IOMMU driver is not
303 allowed anymore to lift isolation
304 requirements as needed. This option
305 does not override iommu=pt
307 amd_iommu_dump= [HW,X86-64]
308 Enable AMD IOMMU driver option to dump the ACPI table
309 for AMD IOMMU. With this option enabled, AMD IOMMU
310 driver will print ACPI tables for AMD IOMMU during
311 IOMMU initialization.
313 amd_iommu_intr= [HW,X86-64]
314 Specifies one of the following AMD IOMMU interrupt
316 legacy - Use legacy interrupt remapping mode.
317 vapic - Use virtual APIC mode, which allows IOMMU
318 to inject interrupts directly into guest.
319 This mode requires kvm-amd.avic=1.
320 (Default when IOMMU HW support is present.)
322 amijoy.map= [HW,JOY] Amiga joystick support
323 Map of devices attached to JOY0DAT and JOY1DAT
325 See also Documentation/input/joydev/joystick.rst
327 analog.map= [HW,JOY] Analog joystick and gamepad support
328 Specifies type or capabilities of an analog joystick
329 connected to one of 16 gameports
330 Format: <type1>,<type2>,..<type16>
333 Power management functions (SPARCstation-4/5 + deriv.)
335 Disable APC CPU standby support. SPARCstation-Fox does
336 not play well with APC CPU idle - disable it if you have
337 APC and your system crashes randomly.
339 apic= [APIC,X86] Advanced Programmable Interrupt Controller
340 Change the output verbosity while booting
341 Format: { quiet (default) | verbose | debug }
342 Change the amount of debugging information output
343 when initialising the APIC and IO-APIC components.
344 For X86-32, this can also be used to specify an APIC
346 Format: apic=driver_name
347 Examples: apic=bigsmp
349 apic_extnmi= [APIC,X86] External NMI delivery setting
350 Format: { bsp (default) | all | none }
351 bsp: External NMI is delivered only to CPU 0
352 all: External NMIs are broadcast to all CPUs as a
354 none: External NMI is masked for all CPUs. This is
355 useful so that a dump capture kernel won't be
359 See Documentation/networking/ipv6.rst.
361 show_lapic= [APIC,X86] Advanced Programmable Interrupt Controller
362 Limit apic dumping. The parameter defines the maximal
363 number of local apics being dumped. Also it is possible
364 to set it to "all" by meaning -- no limit here.
365 Format: { 1 (default) | 2 | ... | all }.
366 The parameter valid if only apic=debug or
367 apic=verbose is specified.
368 Example: apic=debug show_lapic=all
370 apm= [APM] Advanced Power Management
371 See header of arch/x86/kernel/apm_32.c.
373 arcrimi= [HW,NET] ARCnet - "RIM I" (entirely mem-mapped) cards
374 Format: <io>,<irq>,<nodeID>
378 atarimouse= [HW,MOUSE] Atari Mouse
380 atkbd.extra= [HW] Enable extra LEDs and keys on IBM RapidAccess,
381 EzKey and similar keyboards
383 atkbd.reset= [HW] Reset keyboard during initialization
385 atkbd.set= [HW] Select keyboard code set
386 Format: <int> (2 = AT (default), 3 = PS/2)
388 atkbd.scroll= [HW] Enable scroll wheel on MS Office and similar
391 atkbd.softraw= [HW] Choose between synthetic and real raw mode
392 Format: <bool> (0 = real, 1 = synthetic (default))
394 atkbd.softrepeat= [HW]
395 Use software keyboard repeat
397 audit= [KNL] Enable the audit sub-system
398 Format: { "0" | "1" | "off" | "on" }
399 0 | off - kernel audit is disabled and can not be
400 enabled until the next reboot
401 unset - kernel audit is initialized but disabled and
402 will be fully enabled by the userspace auditd.
403 1 | on - kernel audit is initialized and partially
404 enabled, storing at most audit_backlog_limit
405 messages in RAM until it is fully enabled by the
409 audit_backlog_limit= [KNL] Set the audit queue size limit.
410 Format: <int> (must be >=0)
413 bau= [X86_UV] Enable the BAU on SGI UV. The default
414 behavior is to disable the BAU (i.e. bau=0).
415 Format: { "0" | "1" }
418 unset - Disable the BAU.
420 baycom_epp= [HW,AX25]
423 baycom_par= [HW,AX25] BayCom Parallel Port AX.25 Modem
425 See header of drivers/net/hamradio/baycom_par.c.
427 baycom_ser_fdx= [HW,AX25]
428 BayCom Serial Port AX.25 Modem (Full Duplex Mode)
429 Format: <io>,<irq>,<mode>[,<baud>]
430 See header of drivers/net/hamradio/baycom_ser_fdx.c.
432 baycom_ser_hdx= [HW,AX25]
433 BayCom Serial Port AX.25 Modem (Half Duplex Mode)
434 Format: <io>,<irq>,<mode>
435 See header of drivers/net/hamradio/baycom_ser_hdx.c.
437 blkdevparts= Manual partition parsing of block device(s) for
438 embedded devices based on command line input.
439 See Documentation/block/cmdline-partition.rst
441 boot_delay= Milliseconds to delay each printk during boot.
442 Values larger than 10 seconds (10000) are changed to
447 Extended command line options can be added to an initrd
448 and this will cause the kernel to look for it.
450 See Documentation/admin-guide/bootconfig.rst
453 Disable BERT OS support on buggy BIOSes.
455 bgrt_disable [ACPI][X86]
456 Disable BGRT to avoid flickering OEM logo.
458 bttv.card= [HW,V4L] bttv (bt848 + bt878 based grabber cards)
459 bttv.radio= Most important insmod options are available as
461 bttv.pll= See Documentation/admin-guide/media/bttv.rst
464 bulk_remove=off [PPC] This parameter disables the use of the pSeries
465 firmware feature for flushing multiple hpte entries
468 c101= [NET] Moxa C101 synchronous serial card
470 cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection.
471 Sometimes CPU hardware bugs make them report the cache
472 size incorrectly. The kernel will attempt work arounds
473 to fix known problems, but for some CPUs it is not
474 possible to determine what the correct size should be.
475 This option provides an override for these situations.
478 [NET] Specifies amount of time (in seconds) that
479 the kernel should wait for a network carrier. By default
480 it waits 120 seconds.
482 ca_keys= [KEYS] This parameter identifies a specific key(s) on
483 the system trusted keyring to be used for certificate
485 format: { id:<keyid> | builtin }
487 cca= [MIPS] Override the kernel pages' cache coherency
488 algorithm. Accepted values range from 0 to 7
489 inclusive. See arch/mips/include/asm/pgtable-bits.h
490 for platform specific values (SB1, Loongson3 and
493 ccw_timeout_log [S390]
494 See Documentation/s390/common_io.rst for details.
496 cgroup_disable= [KNL] Disable a particular controller
497 Format: {name of the controller(s) to disable}
498 The effects of cgroup_disable=foo are:
499 - foo isn't auto-mounted if you mount all cgroups in
501 - foo isn't visible as an individually mountable
503 {Currently only "memory" controller deal with this and
504 cut the overhead, others just disable the usage. So
505 only cgroup_disable=memory is actually worthy}
507 cgroup_no_v1= [KNL] Disable cgroup controllers and named hierarchies in v1
508 Format: { { controller | "all" | "named" }
509 [,{ controller | "all" | "named" }...] }
510 Like cgroup_disable, but only applies to cgroup v1;
511 the blacklisted controllers remain available in cgroup2.
512 "all" blacklists all controllers and "named" disables
513 named mounts. Specifying both "all" and "named" disables
516 cgroup.memory= [KNL] Pass options to the cgroup memory controller.
518 nosocket -- Disable socket memory accounting.
519 nokmem -- Disable kernel memory accounting.
521 checkreqprot [SELINUX] Set initial checkreqprot flag value.
522 Format: { "0" | "1" }
523 See security/selinux/Kconfig help text.
524 0 -- check protection applied by kernel (includes
525 any implied execute protection).
526 1 -- check protection requested by application.
527 Default value is set via a kernel config option.
528 Value can be changed at runtime via
529 /sys/fs/selinux/checkreqprot.
530 Setting checkreqprot to 1 is deprecated.
533 See Documentation/s390/common_io.rst for details.
536 Prevents the clock framework from automatically gating
537 clocks that have not been explicitly enabled by a Linux
538 device driver but are enabled in hardware at reset or
539 by the bootloader/firmware. Note that this does not
540 force such clocks to be always-on nor does it reserve
541 those clocks in any way. This parameter is useful for
542 debug and development, but should not be needed on a
543 platform with proper driver support. For more
544 information, see Documentation/driver-api/clk.rst.
546 clock= [BUGS=X86-32, HW] gettimeofday clocksource override.
548 Forces specified clocksource (if available) to be used
549 when calculating gettimeofday(). If specified
550 clocksource is not available, it defaults to PIT.
551 Format: { pit | tsc | cyclone | pmtmr }
553 clocksource= Override the default clocksource
555 Override the default clocksource and use the clocksource
556 with the name specified.
557 Some clocksource names to choose from, depending on
559 [all] jiffies (this is the base, fallback clocksource)
561 [ARM] imx_timer1,OSTS,netx_timer,mpu_timer2,
562 pxa_timer,timer3,32k_counter,timer0_1
563 [X86-32] pit,hpet,tsc;
564 scx200_hrt on Geode; cyclone on IBM x440
572 clocksource.arm_arch_timer.evtstrm=
575 Enable/disable the eventstream feature of the ARM
576 architected timer so that code using WFE-based polling
577 loops can be debugged more effectively on production
580 clearcpuid=BITNUM [X86]
581 Disable CPUID feature X for the kernel. See
582 arch/x86/include/asm/cpufeatures.h for the valid bit
583 numbers. Note the Linux specific bits are not necessarily
584 stable over kernel options, but the vendor specific
586 Also note that user programs calling CPUID directly
587 or using the feature without checking anything
588 will still see it. This just prevents it from
589 being used by the kernel or shown in /proc/cpuinfo.
590 Also note the kernel might malfunction if you disable
593 cma=nn[MG]@[start[MG][-end[MG]]]
595 Sets the size of kernel global memory area for
596 contiguous memory allocations and optionally the
597 placement constraint by the physical address range of
598 memory allocations. A value of 0 disables CMA
599 altogether. For more information, see
600 include/linux/dma-contiguous.h
602 cmo_free_hint= [PPC] Format: { yes | no }
603 Specify whether pages are marked as being inactive
604 when they are freed. This is used in CMO environments
605 to determine OS memory pressure for page stealing by
609 coherent_pool=nn[KMG] [ARM,KNL]
610 Sets the size of memory pool for coherent, atomic dma
611 allocations, by default set to 256K.
613 com20020= [HW,NET] ARCnet - COM20020 chipset
615 <io>[,<irq>[,<nodeID>[,<backplane>[,<ckp>[,<timeout>]]]]]
617 com90io= [HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers)
621 ARCnet - COM90xx chipset (memory-mapped buffers)
622 Format: <io>[,<irq>[,<memstart>]]
624 condev= [HW,S390] console device
627 console= [KNL] Output console device and options.
629 tty<n> Use the virtual console device <n>.
633 Use the specified serial port. The options are of
634 the form "bbbbpnf", where "bbbb" is the baud rate,
635 "p" is parity ("n", "o", or "e"), "n" is number of
636 bits, and "f" is flow control ("r" for RTS or
637 omit it). Default is "9600n8".
639 See Documentation/admin-guide/serial-console.rst for more
641 Documentation/networking/netconsole.rst for an
644 uart[8250],io,<addr>[,options]
645 uart[8250],mmio,<addr>[,options]
646 uart[8250],mmio16,<addr>[,options]
647 uart[8250],mmio32,<addr>[,options]
648 uart[8250],0x<addr>[,options]
649 Start an early, polled-mode console on the 8250/16550
650 UART at the specified I/O port or MMIO address,
651 switching to the matching ttyS device later.
652 MMIO inter-register address stride is either 8-bit
653 (mmio), 16-bit (mmio16), or 32-bit (mmio32).
654 If none of [io|mmio|mmio16|mmio32], <addr> is assumed
655 to be equivalent to 'mmio'. 'options' are specified in
656 the same format described for ttyS above; if unspecified,
657 the h/w is not re-initialized.
659 hvc<n> Use the hypervisor console device <n>. This is for
660 both Xen and PowerPC hypervisors.
662 If the device connected to the port is not a TTY but a braille
663 device, prepend "brl," before the device type, for instance
665 For now, only VisioBraille is supported.
668 [KNL] Change console messages format
670 By default we print messages on consoles in
671 "[time stamp] text\n" format (time stamp may not be
672 printed, depending on CONFIG_PRINTK_TIME or
673 `printk_time' param).
675 Switch to syslog format: "<%u>[time stamp] text\n"
676 IOW, each message will have a facility and loglevel
677 prefix. The format is similar to one used by syslog()
678 syscall, or to executing "dmesg -S --raw" or to reading
681 consoleblank= [KNL] The console blank (screen saver) timeout in
682 seconds. A value of 0 disables the blank timer.
686 [KNL] Change the default value for
687 /proc/<pid>/coredump_filter.
688 See also Documentation/filesystems/proc.rst.
690 coresight_cpu_debug.enable
693 Enable/disable the CPU sampling based debugging.
694 0: default value, disable debugging
695 1: enable debugging at boot time
697 cpuidle.off=1 [CPU_IDLE]
698 disable the cpuidle sub-system
701 [CPU_IDLE] Name of the cpuidle governor to use.
703 cpufreq.off=1 [CPU_FREQ]
704 disable the cpufreq sub-system
706 cpufreq.default_governor=
707 [CPU_FREQ] Name of the default cpufreq governor or
708 policy to use. This governor must be registered in the
709 kernel before the cpufreq driver probes.
712 [X86] Delay for N microsec between assert and de-assert
713 of APIC INIT to start processors. This delay occurs
714 on every CPU online, such as boot, and resume from suspend.
717 cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver
719 <first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>]
721 crashkernel=size[KMG][@offset[KMG]]
722 [KNL] Using kexec, Linux can switch to a 'crash kernel'
723 upon panic. This parameter reserves the physical
724 memory region [offset, offset + size] for that kernel
725 image. If '@offset' is omitted, then a suitable offset
726 is selected automatically.
727 [KNL, x86_64] select a region under 4G first, and
728 fall back to reserve region above 4G when '@offset'
729 hasn't been specified.
730 See Documentation/admin-guide/kdump/kdump.rst for further details.
732 crashkernel=range1:size1[,range2:size2,...][@offset]
733 [KNL] Same as above, but depends on the memory
734 in the running system. The syntax of range is
735 start-[end] where start and end are both
736 a memory unit (amount[KMG]). See also
737 Documentation/admin-guide/kdump/kdump.rst for an example.
739 crashkernel=size[KMG],high
740 [KNL, x86_64] range could be above 4G. Allow kernel
741 to allocate physical memory region from top, so could
742 be above 4G if system have more than 4G ram installed.
743 Otherwise memory region will be allocated below 4G, if
745 It will be ignored if crashkernel=X is specified.
746 crashkernel=size[KMG],low
747 [KNL, x86_64] range under 4G. When crashkernel=X,high
748 is passed, kernel could allocate physical memory region
749 above 4G, that cause second kernel crash on system
750 that require some amount of low memory, e.g. swiotlb
751 requires at least 64M+32K low memory, also enough extra
752 low memory is needed to make sure DMA buffers for 32-bit
753 devices won't run out. Kernel would try to allocate at
754 at least 256M below 4G automatically.
755 This one let user to specify own low range under 4G
756 for second kernel instead.
757 0: to disable low allocation.
758 It will be ignored when crashkernel=X,high is not used
759 or memory reserved is below 4G.
762 [KNL] Disable crypto self-tests
767 cs89x0_media= [HW,NET]
768 Format: { rj45 | aui | bnc }
771 See header of drivers/s390/block/dasd_devmap.c.
773 db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port
774 (one device per port)
775 Format: <port#>,<type>
776 See also Documentation/input/devices/joystick-parport.rst
778 ddebug_query= [KNL,DYNAMIC_DEBUG] Enable debug messages at early boot
780 Documentation/admin-guide/dynamic-debug-howto.rst for
781 details. Deprecated, see dyndbg.
783 debug [KNL] Enable kernel debugging (events log level).
786 [KNL] Enable printing [hashed] pointers early in the
787 boot sequence. If enabled, we use a weak hash instead
788 of siphash to hash pointers. Use this option if you are
789 seeing instances of '(___ptrval___)') and need to see a
790 value (hashed pointer) instead. Cryptographically
791 insecure, please do not use on production kernels.
794 [KNL] verbose self-tests
796 Print debugging info while doing the locking API
798 We default to 0 (no extra messages), setting it to
799 1 will print _a lot_ more information - normally
800 only useful to kernel developers.
802 debug_objects [KNL] Enable object debugging
805 [KNL] Disable object debugging
807 debug_guardpage_minorder=
808 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this
809 parameter allows control of the order of pages that will
810 be intentionally kept free (and hence protected) by the
811 buddy allocator. Bigger value increase the probability
812 of catching random memory corruption, but reduce the
813 amount of memory for normal system use. The maximum
814 possible value is MAX_ORDER/2. Setting this parameter
815 to 1 or 2 should be enough to identify most random
816 memory corruption problems caused by bugs in kernel or
817 driver code when a CPU writes to (or reads from) a
818 random memory location. Note that there exists a class
819 of memory corruptions problems caused by buggy H/W or
820 F/W or by drivers badly programing DMA (basically when
821 memory is written at bus level and the CPU MMU is
822 bypassed) which are not detectable by
823 CONFIG_DEBUG_PAGEALLOC, hence this option will not help
824 tracking down these problems.
827 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this parameter
828 enables the feature at boot time. By default, it is
829 disabled and the system will work mostly the same as a
830 kernel built without CONFIG_DEBUG_PAGEALLOC.
831 Note: to get most of debug_pagealloc error reports, it's
832 useful to also enable the page_owner functionality.
833 on: enable the feature
835 debugfs= [KNL] This parameter enables what is exposed to userspace
836 and debugfs internal clients.
837 Format: { on, no-mount, off }
838 on: All functions are enabled.
840 Filesystem is not registered but kernel clients can
841 access APIs and a crashkernel can be used to read
842 its content. There is nothing to mount.
843 off: Filesystem is not registered and clients
844 get a -EPERM as result when trying to register files
845 or directories within debugfs.
846 This is equivalent of the runtime functionality if
847 debugfs was not enabled in the kernel at all.
848 Default value is set in build-time with a kernel configuration.
850 debugpat [X86] Enable PAT debugging
852 decnet.addr= [HW,NET]
853 Format: <area>[,<node>]
854 See also Documentation/networking/decnet.rst.
857 [HW] The size of the default HugeTLB page. This is
858 the size represented by the legacy /proc/ hugepages
859 APIs. In addition, this is the default hugetlb size
860 used for shmget(), mmap() and mounting hugetlbfs
861 filesystems. If not specified, defaults to the
862 architecture's default huge page size. Huge page
863 sizes are architecture dependent. See also
864 Documentation/admin-guide/mm/hugetlbpage.rst.
867 deferred_probe_timeout=
868 [KNL] Debugging option to set a timeout in seconds for
869 deferred probe to give up waiting on dependencies to
870 probe. Only specific dependencies (subsystems or
871 drivers) that have opted in will be ignored. A timeout of 0
872 will timeout at the end of initcalls. This option will also
873 dump out devices still on the deferred probe list after
877 Format: { on | off | def_only | inf_only | always }
878 on: s390 zlib hardware support for compression on
879 level 1 and decompression (default)
880 off: No s390 zlib hardware support
881 def_only: s390 zlib hardware support for deflate
882 only (compression on level 1)
883 inf_only: s390 zlib hardware support for inflate
885 always: Same as 'on' but ignores the selected compression
886 level always using hardware support (used for debugging)
889 Set number of hash buckets for dentry cache.
891 disable_1tb_segments [PPC]
892 Disables the use of 1TB hash page table segments. This
893 causes the kernel to fall back to 256MB segments which
894 can be useful when debugging issues that require an SLB
898 Limits the number of kernel SLB entries, and flushes
899 them frequently to increase the rate of SLB faults
903 See Documentation/networking/ipv6.rst.
906 [KNL] Under CONFIG_HARDENED_USERCOPY, whether
907 hardening is enabled for this boot. Hardened
908 usercopy checking is used to protect the kernel
909 from reading or writing beyond known memory
910 allocation boundaries as a proactive defense
911 against bounds-checking flaws in the kernel's
912 copy_to_user()/copy_from_user() interface.
913 on Perform hardened usercopy checks (default).
914 off Disable hardened usercopy checks.
917 Disable RADIX MMU mode on POWER9
919 radix_hcall_invalidate=on [PPC/PSERIES]
920 Disable RADIX GTSE feature and use hcall for TLB
924 Disable TLBIE instruction. Currently does not work
925 with KVM, with HASH MMU, or with coherent accelerators.
927 disable_cpu_apicid= [X86,APIC,SMP]
929 The number of initial APIC ID for the
930 corresponding CPU to be disabled at boot,
931 mostly used for the kdump 2nd kernel to
932 disable BSP to wake up multiple CPUs without
933 causing system reset or hang due to sending
936 perf_v4_pmi= [X86,INTEL]
938 Disable Intel PMU counter freezing feature.
939 The feature only exists starting from
940 Arch Perfmon v4 (Skylake and newer).
942 disable_ddw [PPC/PSERIES]
943 Disable Dynamic DMA Window support. Use this if
944 to workaround buggy firmware.
947 See Documentation/networking/ipv6.rst.
949 disable_mtrr_cleanup [X86]
950 The kernel tries to adjust MTRR layout from continuous
951 to discrete, to make X server driver able to add WB
952 entry later. This parameter disables that.
954 disable_mtrr_trim [X86, Intel and AMD only]
955 By default the kernel will trim any uncacheable
956 memory out of your available memory pool based on
957 MTRR settings. This parameter disables that behavior,
958 possibly causing your machine to run very slowly.
960 disable_timer_pin_1 [X86]
961 Disable PIN 1 of APIC timer
962 Can be useful to work around chipset bugs.
964 dis_ucode_ldr [X86] Disable the microcode loader.
966 dma_debug=off If the kernel is compiled with DMA_API_DEBUG support,
967 this option disables the debugging code at boot.
969 dma_debug_entries=<number>
970 This option allows to tune the number of preallocated
971 entries for DMA-API debugging code. One entry is
972 required per DMA-API allocation. Use this if the
973 DMA-API debugging code disables itself because the
974 architectural default is too low.
976 dma_debug_driver=<driver_name>
977 With this option the DMA-API debugging driver
978 filter feature can be enabled at boot time. Just
979 pass the driver to filter for as the parameter.
980 The filter can be disabled or changed to another
981 driver later using sysfs.
983 driver_async_probe= [KNL]
984 List of driver names to be probed asynchronously.
985 Format: <driver_name1>,<driver_name2>...
987 drm.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>]
988 Broken monitors, graphic adapters, KVMs and EDIDless
989 panels may send no or incorrect EDID data sets.
990 This parameter allows to specify an EDID data sets
991 in the /lib/firmware directory that are used instead.
992 Generic built-in EDID data sets are used, if one of
993 edid/1024x768.bin, edid/1280x1024.bin,
994 edid/1680x1050.bin, or edid/1920x1080.bin is given
995 and no file with the same name exists. Details and
996 instructions how to build your own EDID data are
997 available in Documentation/admin-guide/edid.rst. An EDID
998 data set will only be used for a particular connector,
999 if its name and a colon are prepended to the EDID
1000 name. Each connector may use a unique EDID data
1001 set by separating the files with a comma. An EDID
1002 data set with no connector name will be used for
1003 any connectors not explicitly specified.
1008 Format: {"off" | "known"}
1009 Control how the dt_cpu_ftrs device-tree binding is
1010 used for CPU feature discovery and setup (if it
1012 off: Do not use it, fall back to legacy cpu table.
1013 known: Do not pass through unknown features to guests
1014 or userspace, only those that the kernel is aware of.
1016 dump_apple_properties [X86]
1017 Dump name and content of EFI device properties on
1018 x86 Macs. Useful for driver authors to determine
1019 what data is available or for reverse-engineering.
1021 dyndbg[="val"] [KNL,DYNAMIC_DEBUG]
1022 module.dyndbg[="val"]
1023 Enable debug messages at boot time. See
1024 Documentation/admin-guide/dynamic-debug-howto.rst
1027 nopku [X86] Disable Memory Protection Keys CPU feature found
1030 module.async_probe [KNL]
1031 Enable asynchronous probe on this module.
1033 early_ioremap_debug [KNL]
1034 Enable debug messages in early_ioremap support. This
1035 is useful for tracking down temporary early mappings
1036 which are not unmapped.
1038 earlycon= [KNL] Output early console device and options.
1040 When used with no options, the early console is
1041 determined by stdout-path property in device tree's
1042 chosen node or the ACPI SPCR table if supported by
1045 cdns,<addr>[,options]
1046 Start an early, polled-mode console on a Cadence
1047 (xuartps) serial port at the specified address. Only
1048 supported option is baud rate. If baud rate is not
1049 specified, the serial port must already be setup and
1052 uart[8250],io,<addr>[,options]
1053 uart[8250],mmio,<addr>[,options]
1054 uart[8250],mmio32,<addr>[,options]
1055 uart[8250],mmio32be,<addr>[,options]
1056 uart[8250],0x<addr>[,options]
1057 Start an early, polled-mode console on the 8250/16550
1058 UART at the specified I/O port or MMIO address.
1059 MMIO inter-register address stride is either 8-bit
1060 (mmio) or 32-bit (mmio32 or mmio32be).
1061 If none of [io|mmio|mmio32|mmio32be], <addr> is assumed
1062 to be equivalent to 'mmio'. 'options' are specified
1063 in the same format described for "console=ttyS<n>"; if
1064 unspecified, the h/w is not initialized.
1068 Start an early, polled-mode console on a pl011 serial
1069 port at the specified address. The pl011 serial port
1070 must already be setup and configured. Options are not
1071 yet supported. If 'mmio32' is specified, then only
1072 the driver will use only 32-bit accessors to read/write
1073 the device registers.
1076 Start an early, polled-mode console on a meson serial
1077 port at the specified address. The serial port must
1078 already be setup and configured. Options are not yet
1082 Start an early, polled-mode console on an msm serial
1083 port at the specified address. The serial port
1084 must already be setup and configured. Options are not
1087 msm_serial_dm,<addr>
1088 Start an early, polled-mode console on an msm serial
1089 dm port at the specified address. The serial port
1090 must already be setup and configured. Options are not
1094 Start an early, polled-mode console on a serial port
1095 of an Actions Semi SoC, such as S500 or S900, at the
1096 specified address. The serial port must already be
1097 setup and configured. Options are not yet supported.
1100 Start an early, polled-mode console on a serial port
1101 of an RDA Micro SoC, such as RDA8810PL, at the
1102 specified address. The serial port must already be
1103 setup and configured. Options are not yet supported.
1106 Use RISC-V SBI (Supervisor Binary Interface) for early
1109 smh Use ARM semihosting calls for early console.
1117 Use early console provided by serial driver available
1118 on Samsung SoCs, requires selecting proper type and
1119 a correct base address of the selected UART port. The
1120 serial port must already be setup and configured.
1121 Options are not yet supported.
1124 Start an early, polled-mode console on a lantiq serial
1125 (lqasc) port at the specified address. The serial port
1126 must already be setup and configured. Options are not
1131 Use early console provided by Freescale LP UART driver
1132 found on Freescale Vybrid and QorIQ LS1021A processors.
1133 A valid base address must be provided, and the serial
1134 port must already be setup and configured.
1138 Start an early, polled-mode, output-only console on the
1139 Freescale i.MX UART at the specified address. The UART
1140 must already be setup and configured.
1143 Start an early, polled-mode console on the
1144 Armada 3700 serial port at the specified
1145 address. The serial port must already be setup
1146 and configured. Options are not yet supported.
1149 Start an early, polled-mode console on a Qualcomm
1150 Generic Interface (GENI) based serial port at the
1151 specified address. The serial port must already be
1152 setup and configured. Options are not yet supported.
1155 Start an early, unaccelerated console on the EFI
1156 memory mapped framebuffer (if available). On cache
1157 coherent non-x86 systems that use system memory for
1158 the framebuffer, pass the 'ram' option so that it is
1159 mapped with the correct attributes.
1162 Use early console provided by Freescale LINFlexD UART
1163 serial driver for NXP S32V234 SoCs. A valid base
1164 address must be provided, and the serial port must
1165 already be setup and configured.
1167 earlyprintk= [X86,SH,ARM,M68k,S390]
1171 earlyprintk=serial[,ttySn[,baudrate]]
1172 earlyprintk=serial[,0x...[,baudrate]]
1173 earlyprintk=ttySn[,baudrate]
1174 earlyprintk=dbgp[debugController#]
1175 earlyprintk=pciserial[,force],bus:device.function[,baudrate]
1176 earlyprintk=xdbc[xhciController#]
1178 earlyprintk is useful when the kernel crashes before
1179 the normal console is initialized. It is not enabled by
1180 default because it has some cosmetic problems.
1182 Append ",keep" to not disable it when the real console
1185 Only one of vga, efi, serial, or usb debug port can
1188 Currently only ttyS0 and ttyS1 may be specified by
1189 name. Other I/O ports may be explicitly specified
1190 on some architectures (x86 and arm at least) by
1191 replacing ttySn with an I/O port address, like this:
1192 earlyprintk=serial,0x1008,115200
1193 You can find the port for a given device in
1194 /proc/tty/driver/serial:
1195 2: uart:ST16650V2 port:00001008 irq:18 ...
1197 Interaction with the standard serial driver is not
1200 The VGA and EFI output is eventually overwritten by
1203 The xen output can only be used by Xen PV guests.
1205 The sclp output can only be used on s390.
1207 The optional "force" to "pciserial" enables use of a
1208 PCI device even when its classcode is not of the
1211 edac_report= [HW,EDAC] Control how to report EDAC event
1212 Format: {"on" | "off" | "force"}
1213 on: enable EDAC to report H/W event. May be overridden
1214 by other higher priority error reporting module.
1215 off: disable H/W event reporting through EDAC.
1216 force: enforce the use of EDAC to report H/W event.
1219 ekgdboc= [X86,KGDB] Allow early kernel console debugging
1222 This is designed to be used in conjunction with
1223 the boot argument: earlyprintk=vga
1225 This parameter works in place of the kgdboc parameter
1226 but can only be used if the backing tty is available
1227 very early in the boot process. For early debugging
1228 via a serial port see kgdboc_earlycon instead.
1231 Format: {"off" | "on" | "skip[mbr]"}
1234 Format: { "debug", "disable_early_pci_dma",
1235 "nochunk", "noruntime", "nosoftreserve",
1236 "novamap", "no_disable_early_pci_dma",
1238 debug: enable misc debug output.
1239 disable_early_pci_dma: disable the busmaster bit on all
1240 PCI bridges while in the EFI boot stub.
1241 nochunk: disable reading files in "chunks" in the EFI
1242 boot stub, as chunking can cause problems with some
1243 firmware implementations.
1244 noruntime : disable EFI runtime services support
1245 nosoftreserve: The EFI_MEMORY_SP (Specific Purpose)
1246 attribute may cause the kernel to reserve the
1247 memory range for a memory mapping driver to
1248 claim. Specify efi=nosoftreserve to disable this
1249 reservation and treat the memory by its base type
1250 (i.e. EFI_CONVENTIONAL_MEMORY / "System RAM").
1251 novamap: do not call SetVirtualAddressMap().
1252 no_disable_early_pci_dma: Leave the busmaster bit set
1253 on all PCI bridges while in the EFI boot stub
1254 old_map [X86-64]: switch to the old ioremap-based EFI
1255 runtime services mapping. [Needs CONFIG_X86_UV=y]
1257 efi_no_storage_paranoia [EFI; X86]
1258 Using this parameter you can use more than 50% of
1259 your efi variable storage. Use this parameter only if
1260 you are really sure that your UEFI does sane gc and
1261 fulfills the spec otherwise your board may brick.
1263 efi_fake_mem= nn[KMG]@ss[KMG]:aa[,nn[KMG]@ss[KMG]:aa,..] [EFI; X86]
1264 Add arbitrary attribute to specific memory range by
1265 updating original EFI memory map.
1266 Region of memory which aa attribute is added to is
1269 If efi_fake_mem=2G@4G:0x10000,2G@0x10a0000000:0x10000
1270 is specified, EFI_MEMORY_MORE_RELIABLE(0x10000)
1271 attribute is added to range 0x100000000-0x180000000 and
1272 0x10a0000000-0x1120000000.
1274 If efi_fake_mem=8G@9G:0x40000 is specified, the
1275 EFI_MEMORY_SP(0x40000) attribute is added to
1276 range 0x240000000-0x43fffffff.
1278 Using this parameter you can do debugging of EFI memmap
1279 related features. For example, you can do debugging of
1280 Address Range Mirroring feature even if your box
1281 doesn't support it, or mark specific memory as
1284 efivar_ssdt= [EFI; X86] Name of an EFI variable that contains an SSDT
1285 that is to be dynamically loaded by Linux. If there are
1286 multiple variables with the same name but with different
1287 vendor GUIDs, all of them will be loaded. See
1288 Documentation/admin-guide/acpi/ssdt-overlays.rst for details.
1291 eisa_irq_edge= [PARISC,HW]
1292 See header of drivers/parisc/eisa.c.
1295 See comment before function elanfreq_setup() in
1296 arch/x86/kernel/cpu/cpufreq/elanfreq.c.
1298 elfcorehdr=[size[KMG]@]offset[KMG] [IA64,PPC,SH,X86,S390]
1299 Specifies physical address of start of kernel core
1300 image elf header and optionally the size. Generally
1301 kexec loader will pass this option to capture kernel.
1302 See Documentation/admin-guide/kdump/kdump.rst for details.
1304 enable_mtrr_cleanup [X86]
1305 The kernel tries to adjust MTRR layout from continuous
1306 to discrete, to make X server driver able to add WB
1307 entry later. This parameter enables that.
1309 enable_timer_pin_1 [X86]
1310 Enable PIN 1 of APIC timer
1311 Can be useful to work around chipset bugs
1312 (in particular on some ATI chipsets).
1313 The kernel tries to set a reasonable default.
1315 enforcing [SELINUX] Set initial enforcing status.
1317 See security/selinux/Kconfig help text.
1318 0 -- permissive (log only, no denials).
1319 1 -- enforcing (deny and log).
1321 Value can be changed at runtime via
1322 /sys/fs/selinux/enforce.
1325 Disable Error Record Serialization Table (ERST)
1328 ether= [HW,NET] Ethernet cards parameters
1329 This option is obsoleted by the "netdev=" option, which
1330 has equivalent usage. See its documentation for details.
1334 Permit 'security.evm' to be updated regardless of
1335 current integrity status.
1339 fail_make_request=[KNL]
1340 General fault injection mechanism.
1341 Format: <interval>,<probability>,<space>,<times>
1342 See also Documentation/fault-injection/.
1345 See Documentation/admin-guide/blockdev/floppy.rst.
1347 force_pal_cache_flush
1348 [IA-64] Avoid check_sal_cache_flush which may hang on
1349 buggy SAL_CACHE_FLUSH implementations. Using this
1350 parameter will force ia64_sal_cache_flush to call
1351 ia64_pal_cache_flush instead of SAL_CACHE_FLUSH.
1354 Forcefully enable Physical Address Extension (PAE).
1355 Many Pentium M systems disable PAE but may have a
1356 functionally usable PAE implementation.
1357 Warning: use of this parameter will taint the kernel
1358 and may cause unknown problems.
1361 [FTRACE] will set and start the specified tracer
1362 as early as possible in order to facilitate early
1365 ftrace_dump_on_oops[=orig_cpu]
1366 [FTRACE] will dump the trace buffers on oops.
1367 If no parameter is passed, ftrace will dump
1368 buffers of all CPUs, but if you pass orig_cpu, it will
1369 dump only the buffer of the CPU that triggered the
1372 ftrace_filter=[function-list]
1373 [FTRACE] Limit the functions traced by the function
1374 tracer at boot up. function-list is a comma separated
1375 list of functions. This list can be changed at run
1376 time by the set_ftrace_filter file in the debugfs
1379 ftrace_notrace=[function-list]
1380 [FTRACE] Do not trace the functions specified in
1381 function-list. This list can be changed at run time
1382 by the set_ftrace_notrace file in the debugfs
1385 ftrace_graph_filter=[function-list]
1386 [FTRACE] Limit the top level callers functions traced
1387 by the function graph tracer at boot up.
1388 function-list is a comma separated list of functions
1389 that can be changed at run time by the
1390 set_graph_function file in the debugfs tracing directory.
1392 ftrace_graph_notrace=[function-list]
1393 [FTRACE] Do not trace from the functions specified in
1394 function-list. This list is a comma separated list of
1395 functions that can be changed at run time by the
1396 set_graph_notrace file in the debugfs tracing directory.
1398 ftrace_graph_max_depth=<uint>
1399 [FTRACE] Used with the function graph tracer. This is
1400 the max depth it will trace into a function. This value
1401 can be changed at run time by the max_graph_depth file
1402 in the tracefs tracing directory. default: 0 (no limit)
1404 fw_devlink= [KNL] Create device links between consumer and supplier
1405 devices by scanning the firmware to infer the
1406 consumer/supplier relationships. This feature is
1407 especially useful when drivers are loaded as modules as
1408 it ensures proper ordering of tasks like device probing
1409 (suppliers first, then consumers), supplier boot state
1410 clean up (only after all consumers have probed),
1411 suspend/resume & runtime PM (consumers first, then
1413 Format: { off | permissive | on | rpm }
1414 off -- Don't create device links from firmware info.
1415 permissive -- Create device links from firmware info
1416 but use it only for ordering boot state clean
1417 up (sync_state() calls).
1418 on -- Create device links from firmware info and use it
1419 to enforce probe and suspend/resume ordering.
1420 rpm -- Like "on", but also use to order runtime PM.
1423 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad
1424 support via parallel port (up to 5 devices per port)
1425 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5>
1426 See also Documentation/input/devices/joystick-parport.rst
1430 gart_fix_e820= [X86_64] disable the fix e820 for K8 GART
1434 gcov_persist= [GCOV] When non-zero (default), profiling data for
1435 kernel modules is saved and remains accessible via
1436 debugfs, even when the module is unloaded/reloaded.
1437 When zero, profiling data is discarded and associated
1438 debugfs files are removed at module unload time.
1440 goldfish [X86] Enable the goldfish android emulator platform.
1441 Don't use this when you are not running on the
1444 gpt [EFI] Forces disk with valid GPT signature but
1445 invalid Protective MBR to be treated as GPT. If the
1446 primary GPT is corrupted, it enables the backup/alternate
1447 GPT to be used instead.
1449 grcan.enable0= [HW] Configuration of physical interface 0. Determines
1450 the "Enable 0" bit of the configuration register.
1453 grcan.enable1= [HW] Configuration of physical interface 1. Determines
1454 the "Enable 0" bit of the configuration register.
1457 grcan.select= [HW] Select which physical interface to use.
1460 grcan.txsize= [HW] Sets the size of the tx buffer.
1461 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0.
1463 grcan.rxsize= [HW] Sets the size of the rx buffer.
1464 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0.
1467 gpio-mockup.gpio_mockup_ranges
1468 [HW] Sets the ranges of gpiochip of for this device.
1469 Format: <start1>,<end1>,<start2>,<end2>...
1471 hardlockup_all_cpu_backtrace=
1472 [KNL] Should the hard-lockup detector generate
1473 backtraces on all cpus.
1476 hashdist= [KNL,NUMA] Large hashes allocated during boot
1477 are distributed across NUMA nodes. Defaults on
1478 for 64-bit NUMA, off otherwise.
1479 Format: 0 | 1 (for off | on)
1481 hcl= [IA-64] SGI's Hardware Graph compatibility layer
1483 hd= [EIDE] (E)IDE hard drive subsystem geometry
1484 Format: <cyl>,<head>,<sect>
1487 Disable Hardware Error Source Table (HEST) support;
1488 corresponding firmware-first mode error processing
1489 logic will be disabled.
1491 highmem=nn[KMG] [KNL,BOOT] forces the highmem zone to have an exact
1492 size of <nn>. This works even on boxes that have no
1493 highmem otherwise. This also works to reduce highmem
1494 size on bigger boxes.
1496 highres= [KNL] Enable/disable high resolution timer mode.
1497 Valid parameters: "on", "off"
1502 hpet= [X86-32,HPET] option to control HPET usage
1503 Format: { enable (default) | disable | force |
1505 disable: disable HPET and use PIT instead
1506 force: allow force enabled of undocumented chips (ICH4,
1508 verbose: show contents of HPET registers during setup
1510 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET
1511 registers. Default set by CONFIG_HPET_MMAP_DEFAULT.
1513 hugetlb_cma= [HW] The size of a cma area used for allocation
1514 of gigantic hugepages.
1517 Reserve a cma area of given size and allocate gigantic
1518 hugepages using the cma allocator. If enabled, the
1519 boot-time allocation of gigantic hugepages is skipped.
1521 hugepages= [HW] Number of HugeTLB pages to allocate at boot.
1522 If this follows hugepagesz (below), it specifies
1523 the number of pages of hugepagesz to be allocated.
1524 If this is the first HugeTLB parameter on the command
1525 line, it specifies the number of pages to allocate for
1526 the default huge page size. See also
1527 Documentation/admin-guide/mm/hugetlbpage.rst.
1531 [HW] The size of the HugeTLB pages. This is used in
1532 conjunction with hugepages (above) to allocate huge
1533 pages of a specific size at boot. The pair
1534 hugepagesz=X hugepages=Y can be specified once for
1535 each supported huge page size. Huge page sizes are
1536 architecture dependent. See also
1537 Documentation/admin-guide/mm/hugetlbpage.rst.
1541 [KNL] Should the hung task detector generate panics.
1544 A value of 1 instructs the kernel to panic when a
1545 hung task is detected. The default value is controlled
1546 by the CONFIG_BOOTPARAM_HUNG_TASK_PANIC build-time
1547 option. The value selected by this boot parameter can
1548 be changed later by the kernel.hung_task_panic sysctl.
1550 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC)
1551 terminal devices. Valid values: 0..8
1552 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs.
1553 If specified, z/VM IUCV HVC accepts connections
1554 from listed z/VM user IDs only.
1556 hv_nopvspin [X86,HYPER_V] Disables the paravirt spinlock optimizations
1557 which allow the hypervisor to 'idle' the
1558 guest on lock contention.
1561 Do not unregister boot console at start. This is only
1562 useful for debugging when something happens in the window
1563 between unregistering the boot console and initializing
1566 i2c_bus= [HW] Override the default board specific I2C bus speed
1567 or register an additional I2C bus that is not
1568 registered from board initialization code.
1572 i8042.debug [HW] Toggle i8042 debug mode
1573 i8042.unmask_kbd_data
1574 [HW] Enable printing of interrupt data from the KBD port
1575 (disabled by default, and as a pre-condition
1576 requires that i8042.debug=1 be enabled)
1577 i8042.direct [HW] Put keyboard port into non-translated mode
1578 i8042.dumbkbd [HW] Pretend that controller can only read data from
1579 keyboard and cannot control its state
1580 (Don't attempt to blink the leds)
1581 i8042.noaux [HW] Don't check for auxiliary (== mouse) port
1582 i8042.nokbd [HW] Don't check/create keyboard port
1583 i8042.noloop [HW] Disable the AUX Loopback command while probing
1585 i8042.nomux [HW] Don't check presence of an active multiplexing
1587 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX
1589 i8042.notimeout [HW] Ignore timeout condition signalled by controller
1590 i8042.reset [HW] Reset the controller during init, cleanup and
1591 suspend-to-ram transitions, only during s2r
1592 transitions, or never reset
1593 Format: { 1 | Y | y | 0 | N | n }
1594 1, Y, y: always reset controller
1595 0, N, n: don't ever reset controller
1596 Default: only on s2r transitions on x86; most other
1597 architectures force reset to be always executed
1598 i8042.unlock [HW] Unlock (ignore) the keylock
1599 i8042.kbdreset [HW] Reset device connected to KBD port
1603 i8k.ignore_dmi [HW] Continue probing hardware even if DMI data
1604 indicates that the driver is running on unsupported
1606 i8k.force [HW] Activate i8k driver even if SMM BIOS signature
1607 does not match list of supported models.
1609 [HW] Report power status in /proc/i8k
1610 (disabled by default)
1611 i8k.restricted [HW] Allow controlling fans only if SYS_ADMIN
1614 i915.invert_brightness=
1615 [DRM] Invert the sense of the variable that is used to
1616 set the brightness of the panel backlight. Normally a
1617 brightness value of 0 indicates backlight switched off,
1618 and the maximum of the brightness value sets the backlight
1619 to maximum brightness. If this parameter is set to 0
1620 (default) and the machine requires it, or this parameter
1621 is set to 1, a brightness value of 0 sets the backlight
1622 to maximum brightness, and the maximum of the brightness
1623 value switches the backlight off.
1624 -1 -- never invert brightness
1625 0 -- machine default
1626 1 -- force brightness inversion
1629 Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]]
1631 ide-core.nodma= [HW] (E)IDE subsystem
1632 Format: =0.0 to prevent dma on hda, =0.1 hdb =1.0 hdc
1633 .vlb_clock .pci_clock .noflush .nohpa .noprobe .nowerr
1634 .cdrom .chs .ignore_cable are additional options
1635 See Documentation/ide/ide.rst.
1637 ide-generic.probe-mask= [HW] (E)IDE subsystem
1639 Probe mask for legacy ISA IDE ports. Depending on
1640 platform up to 6 ports are supported, enabled by
1641 setting corresponding bits in the mask to 1. The
1642 default value is 0x0, which has a special meaning.
1643 On systems that have PCI, it triggers scanning the
1644 PCI bus for the first and the second port, which
1645 are then probed. On systems without PCI the value
1646 of 0x0 enables probing the two first ports as if it
1649 ide-pci-generic.all-generic-ide [HW] (E)IDE subsystem
1650 Claim all unknown PCI IDE storage controllers.
1653 Format: idle=poll, idle=halt, idle=nomwait
1654 Poll forces a polling idle loop that can slightly
1655 improve the performance of waking up a idle CPU, but
1656 will use a lot of power and make the system run hot.
1658 idle=halt: Halt is forced to be used for CPU idle.
1659 In such case C2/C3 won't be used again.
1660 idle=nomwait: Disable mwait for CPU C-states
1662 ieee754= [MIPS] Select IEEE Std 754 conformance mode
1663 Format: { strict | legacy | 2008 | relaxed }
1666 Choose which programs will be accepted for execution
1667 based on the IEEE 754 NaN encoding(s) supported by
1668 the FPU and the NaN encoding requested with the value
1669 of an ELF file header flag individually set by each
1670 binary. Hardware implementations are permitted to
1671 support either or both of the legacy and the 2008 NaN
1674 Available settings are as follows:
1675 strict accept binaries that request a NaN encoding
1676 supported by the FPU
1677 legacy only accept legacy-NaN binaries, if supported
1679 2008 only accept 2008-NaN binaries, if supported
1681 relaxed accept any binaries regardless of whether
1682 supported by the FPU
1684 The FPU emulator is always able to support both NaN
1685 encodings, so if no FPU hardware is present or it has
1686 been disabled with 'nofpu', then the settings of
1687 'legacy' and '2008' strap the emulator accordingly,
1688 'relaxed' straps the emulator for both legacy-NaN and
1689 2008-NaN, whereas 'strict' enables legacy-NaN only on
1690 legacy processors and both NaN encodings on MIPS32 or
1693 The setting for ABS.fmt/NEG.fmt instruction execution
1694 mode generally follows that for the NaN encoding,
1695 except where unsupported by hardware.
1697 ignore_loglevel [KNL]
1698 Ignore loglevel setting - this will print /all/
1699 kernel messages to the console. Useful for debugging.
1700 We also add it as printk module parameter, so users
1701 could change it dynamically, usually by
1702 /sys/module/printk/parameters/ignore_loglevel.
1705 Ignore RLIMIT_DATA setting for data mappings,
1706 print warning at first misuse. Can be changed via
1707 /sys/module/kernel/parameters/ignore_rlimit_data.
1709 ihash_entries= [KNL]
1710 Set number of hash buckets for inode cache.
1712 ima_appraise= [IMA] appraise integrity measurements
1713 Format: { "off" | "enforce" | "fix" | "log" }
1716 ima_appraise_tcb [IMA] Deprecated. Use ima_policy= instead.
1717 The builtin appraise policy appraises all files
1720 ima_canonical_fmt [IMA]
1721 Use the canonical format for the binary runtime
1722 measurements, instead of host native format.
1725 Format: { md5 | sha1 | rmd160 | sha256 | sha384
1729 The list of supported hash algorithms is defined
1730 in crypto/hash_info.h.
1733 The builtin policies to load during IMA setup.
1734 Format: "tcb | appraise_tcb | secure_boot |
1737 The "tcb" policy measures all programs exec'd, files
1738 mmap'd for exec, and all files opened with the read
1739 mode bit set by either the effective uid (euid=0) or
1742 The "appraise_tcb" policy appraises the integrity of
1743 all files owned by root.
1745 The "secure_boot" policy appraises the integrity
1746 of files (eg. kexec kernel image, kernel modules,
1747 firmware, policy, etc) based on file signatures.
1749 The "fail_securely" policy forces file signature
1750 verification failure also on privileged mounted
1751 filesystems with the SB_I_UNVERIFIABLE_SIGNATURE
1754 ima_tcb [IMA] Deprecated. Use ima_policy= instead.
1755 Load a policy which meets the needs of the Trusted
1756 Computing Base. This means IMA will measure all
1757 programs exec'd, files mmap'd for exec, and all files
1758 opened for read by uid=0.
1761 Select one of defined IMA measurements template formats.
1762 Formats: { "ima" | "ima-ng" | "ima-sig" }
1766 [IMA] Define a custom template format.
1767 Format: { "field1|...|fieldN" }
1769 ima.ahash_minsize= [IMA] Minimum file size for asynchronous hash usage
1770 Format: <min_file_size>
1771 Set the minimal file size for using asynchronous hash.
1772 If left unspecified, ahash usage is disabled.
1774 ahash performance varies for different data sizes on
1775 different crypto accelerators. This option can be used
1776 to achieve the best performance for a particular HW.
1778 ima.ahash_bufsize= [IMA] Asynchronous hash buffer size
1780 Set hashing buffer size. Default: 4k.
1782 ahash performance varies for different chunk sizes on
1783 different crypto accelerators. This option can be used
1784 to achieve best performance for particular HW.
1788 Run specified binary instead of /sbin/init as init
1791 initcall_debug [KNL] Trace initcalls as they are executed. Useful
1792 for working out where the kernel is dying during
1795 initcall_blacklist= [KNL] Do not execute a comma-separated list of
1796 initcall functions. Useful for debugging built-in
1797 modules and initcalls.
1799 initrd= [BOOT] Specify the location of the initial ramdisk
1801 initrdmem= [KNL] Specify a physical address and size from which to
1802 load the initrd. If an initrd is compiled in or
1803 specified in the bootparams, it takes priority over this
1805 Format: ss[KMG],nn[KMG]
1808 init_on_alloc= [MM] Fill newly allocated pages and heap objects with
1811 Default set by CONFIG_INIT_ON_ALLOC_DEFAULT_ON.
1813 init_on_free= [MM] Fill freed pages and heap objects with zeroes.
1815 Default set by CONFIG_INIT_ON_FREE_DEFAULT_ON.
1817 init_pkru= [x86] Specify the default memory protection keys rights
1818 register contents for all processes. 0x55555554 by
1819 default (disallow access to all but pkey 0). Can
1820 override in debugfs after boot.
1822 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver
1825 int_pln_enable [x86] Enable power limit notification interrupt
1827 integrity_audit=[IMA]
1828 Format: { "0" | "1" }
1829 0 -- basic integrity auditing messages. (Default)
1830 1 -- additional integrity auditing messages.
1832 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option
1834 Enable intel iommu driver.
1836 Disable intel iommu driver.
1837 igfx_off [Default Off]
1838 By default, gfx is mapped as normal device. If a gfx
1839 device has a dedicated DMAR unit, the DMAR unit is
1840 bypassed by not enabling DMAR with this option. In
1841 this case, gfx device will use physical address for
1844 With this option iommu will not optimize to look
1845 for io virtual address below 32-bit forcing dual
1846 address cycle on pci bus for cards supporting greater
1847 than 32-bit addressing. The default is to look
1848 for translation below 32-bit and if not available
1849 then look in the higher range.
1850 strict [Default Off]
1851 With this option on every unmap_single operation will
1852 result in a hardware IOTLB flush operation as opposed
1853 to batching them for performance.
1854 sp_off [Default Off]
1855 By default, super page will be supported if Intel IOMMU
1856 has the capability. With this option, super page will
1859 By default, scalable mode will be disabled even if the
1860 hardware advertises that it has support for the scalable
1861 mode translation. With this option set, scalable mode
1862 will be used on hardware which claims to support it.
1863 tboot_noforce [Default Off]
1864 Do not force the Intel IOMMU enabled under tboot.
1865 By default, tboot will force Intel IOMMU on, which
1866 could harm performance of some high-throughput
1867 devices like 40GBit network cards, even if identity
1869 Note that using this option lowers the security
1870 provided by tboot because it makes the system
1871 vulnerable to DMA attacks.
1872 nobounce [Default off]
1873 Disable bounce buffer for untrusted devices such as
1874 the Thunderbolt devices. This will treat the untrusted
1875 devices as the trusted ones, hence might expose security
1876 risks of DMA attacks.
1878 intel_idle.max_cstate= [KNL,HW,ACPI,X86]
1879 0 disables intel_idle and fall back on acpi_idle.
1880 1 to 9 specify maximum depth of C-state.
1884 Do not enable intel_pstate as the default
1885 scaling driver for the supported processors
1887 Use intel_pstate as a scaling driver, but configure it
1888 to work with generic cpufreq governors (instead of
1889 enabling its internal governor). This mode cannot be
1890 used along with the hardware-managed P-states (HWP)
1893 Enable intel_pstate on systems that prohibit it by default
1894 in favor of acpi-cpufreq. Forcing the intel_pstate driver
1895 instead of acpi-cpufreq may disable platform features, such
1896 as thermal controls and power capping, that rely on ACPI
1897 P-States information being indicated to OSPM and therefore
1898 should be used with caution. This option does not work with
1899 processors that aren't supported by the intel_pstate driver
1900 or on platforms that use pcc-cpufreq instead of acpi-cpufreq.
1902 Do not enable hardware P state control (HWP)
1905 Only load intel_pstate on systems which support
1906 hardware P state control (HWP) if available.
1908 Enforce ACPI _PPC performance limits. If the Fixed ACPI
1909 Description Table, specifies preferred power management
1910 profile as "Enterprise Server" or "Performance Server",
1911 then this feature is turned on by default.
1913 Allow per-logical-CPU P-State performance control limits using
1914 cpufreq sysfs interface
1916 intremap= [X86-64, Intel-IOMMU]
1917 on enable Interrupt Remapping (default)
1918 off disable Interrupt Remapping
1919 nosid disable Source ID checking
1921 BIOS x2APIC opt-out request will be ignored
1922 nopost disable Interrupt Posting
1924 iomem= Disable strict checking of access to MMIO memory
1925 strict regions from userspace.
1940 nobypass [PPC/POWERNV]
1941 Disable IOMMU bypass, using IOMMU for PCI devices.
1943 iommu.strict= [ARM64] Configure TLB invalidation behaviour
1944 Format: { "0" | "1" }
1946 Request that DMA unmap operations use deferred
1947 invalidation of hardware TLBs, for increased
1948 throughput at the cost of reduced device isolation.
1949 Will fall back to strict mode if not supported by
1950 the relevant IOMMU driver.
1951 1 - Strict mode (default).
1952 DMA unmap operations invalidate IOMMU hardware TLBs
1956 [ARM64, X86] Configure DMA to bypass the IOMMU by default.
1957 Format: { "0" | "1" }
1958 0 - Use IOMMU translation for DMA.
1959 1 - Bypass the IOMMU for DMA.
1960 unset - Use value of CONFIG_IOMMU_DEFAULT_PASSTHROUGH.
1962 io7= [HW] IO7 for Marvel based alpha systems
1963 See comment before marvel_specify_io7 in
1964 arch/alpha/kernel/core_marvel.c.
1966 io_delay= [X86] I/O delay method
1968 Standard port 0x80 based delay
1970 Alternate port 0xed based delay (needed on some systems)
1972 Simple two microseconds delay
1977 See Documentation/admin-guide/nfs/nfsroot.rst.
1979 ipcmni_extend [KNL] Extend the maximum number of unique System V
1980 IPC identifiers from 32,768 to 16,777,216.
1982 irqaffinity= [SMP] Set the default irq affinity mask
1983 The argument is a cpu list, as described above.
1985 irqchip.gicv2_force_probe=
1988 Force the kernel to look for the second 4kB page
1989 of a GICv2 controller even if the memory range
1990 exposed by the device tree is too small.
1992 irqchip.gicv3_nolpi=
1994 Force the kernel to ignore the availability of
1995 LPIs (and by consequence ITSs). Intended for system
1996 that use the kernel as a bootloader, and thus want
1997 to let secondary kernels in charge of setting up
2000 irqchip.gicv3_pseudo_nmi= [ARM64]
2001 Enables support for pseudo-NMIs in the kernel. This
2002 requires the kernel to be built with
2003 CONFIG_ARM64_PSEUDO_NMI.
2006 When an interrupt is not handled search all handlers
2007 for it. Intended to get systems with badly broken
2011 When an interrupt is not handled search all handlers
2012 for it. Also check all handlers each timer
2013 interrupt. Intended to get systems with badly broken
2017 Format: <RDP>,<reset>,<pci_scan>,<verbosity>
2019 isolcpus= [KNL,SMP,ISOL] Isolate a given set of CPUs from disturbance.
2020 [Deprecated - use cpusets instead]
2021 Format: [flag-list,]<cpu-list>
2023 Specify one or more CPUs to isolate from disturbances
2024 specified in the flag list (default: domain):
2027 Disable the tick when a single task runs.
2029 A residual 1Hz tick is offloaded to workqueues, which you
2030 need to affine to housekeeping through the global
2031 workqueue's affinity configured via the
2032 /sys/devices/virtual/workqueue/cpumask sysfs file, or
2033 by using the 'domain' flag described below.
2035 NOTE: by default the global workqueue runs on all CPUs,
2036 so to protect individual CPUs the 'cpumask' file has to
2037 be configured manually after bootup.
2040 Isolate from the general SMP balancing and scheduling
2041 algorithms. Note that performing domain isolation this way
2042 is irreversible: it's not possible to bring back a CPU to
2043 the domains once isolated through isolcpus. It's strongly
2044 advised to use cpusets instead to disable scheduler load
2045 balancing through the "cpuset.sched_load_balance" file.
2046 It offers a much more flexible interface where CPUs can
2047 move in and out of an isolated set anytime.
2049 You can move a process onto or off an "isolated" CPU via
2050 the CPU affinity syscalls or cpuset.
2051 <cpu number> begins at 0 and the maximum value is
2052 "number of CPUs in system - 1".
2056 Isolate from being targeted by managed interrupts
2057 which have an interrupt mask containing isolated
2058 CPUs. The affinity of managed interrupts is
2059 handled by the kernel and cannot be changed via
2060 the /proc/irq/* interfaces.
2062 This isolation is best effort and only effective
2063 if the automatically assigned interrupt mask of a
2064 device queue contains isolated and housekeeping
2065 CPUs. If housekeeping CPUs are online then such
2066 interrupts are directed to the housekeeping CPU
2067 so that IO submitted on the housekeeping CPU
2068 cannot disturb the isolated CPU.
2070 If a queue's affinity mask contains only isolated
2071 CPUs then this parameter has no effect on the
2072 interrupt routing decision, though interrupts are
2073 only delivered when tasks running on those
2074 isolated CPUs submit IO. IO submitted on
2075 housekeeping CPUs has no influence on those
2078 The format of <cpu-list> is described above.
2082 ivrs_ioapic [HW,X86_64]
2083 Provide an override to the IOAPIC-ID<->DEVICE-ID
2084 mapping provided in the IVRS ACPI table. For
2085 example, to map IOAPIC-ID decimal 10 to
2086 PCI device 00:14.0 write the parameter as:
2087 ivrs_ioapic[10]=00:14.0
2089 ivrs_hpet [HW,X86_64]
2090 Provide an override to the HPET-ID<->DEVICE-ID
2091 mapping provided in the IVRS ACPI table. For
2092 example, to map HPET-ID decimal 0 to
2093 PCI device 00:14.0 write the parameter as:
2094 ivrs_hpet[0]=00:14.0
2096 ivrs_acpihid [HW,X86_64]
2097 Provide an override to the ACPI-HID:UID<->DEVICE-ID
2098 mapping provided in the IVRS ACPI table. For
2099 example, to map UART-HID:UID AMD0020:0 to
2100 PCI device 00:14.5 write the parameter as:
2101 ivrs_acpihid[00:14.5]=AMD0020:0
2103 js= [HW,JOY] Analog joystick
2104 See Documentation/input/joydev/joystick.rst.
2107 When CONFIG_RANDOMIZE_BASE is set, this disables
2108 kernel and module base offset ASLR (Address Space
2109 Layout Randomization).
2112 [KNL] Enforce KASAN (Kernel Address Sanitizer) to print
2113 report on every invalid memory access. Without this
2114 parameter KASAN will print report only for the first
2119 kernelcore= [KNL,X86,IA-64,PPC]
2120 Format: nn[KMGTPE] | nn% | "mirror"
2121 This parameter specifies the amount of memory usable by
2122 the kernel for non-movable allocations. The requested
2123 amount is spread evenly throughout all nodes in the
2124 system as ZONE_NORMAL. The remaining memory is used for
2125 movable memory in its own zone, ZONE_MOVABLE. In the
2126 event, a node is too small to have both ZONE_NORMAL and
2127 ZONE_MOVABLE, kernelcore memory will take priority and
2128 other nodes will have a larger ZONE_MOVABLE.
2130 ZONE_MOVABLE is used for the allocation of pages that
2131 may be reclaimed or moved by the page migration
2132 subsystem. Note that allocations like PTEs-from-HighMem
2133 still use the HighMem zone if it exists, and the Normal
2134 zone if it does not.
2136 It is possible to specify the exact amount of memory in
2137 the form of "nn[KMGTPE]", a percentage of total system
2138 memory in the form of "nn%", or "mirror". If "mirror"
2139 option is specified, mirrored (reliable) memory is used
2140 for non-movable allocations and remaining memory is used
2141 for Movable pages. "nn[KMGTPE]", "nn%", and "mirror"
2142 are exclusive, so you cannot specify multiple forms.
2144 kgdbdbgp= [KGDB,HW] kgdb over EHCI usb debug port.
2145 Format: <Controller#>[,poll interval]
2146 The controller # is the number of the ehci usb debug
2147 port as it is probed via PCI. The poll interval is
2148 optional and is the number seconds in between
2149 each poll cycle to the debug port in case you need
2150 the functionality for interrupting the kernel with
2151 gdb or control-c on the dbgp connection. When
2152 not using this parameter you use sysrq-g to break into
2153 the kernel debugger.
2155 kgdboc= [KGDB,HW] kgdb over consoles.
2156 Requires a tty driver that supports console polling,
2157 or a supported polling keyboard driver (non-usb).
2158 Serial only format: <serial_device>[,baud]
2159 keyboard only format: kbd
2160 keyboard and serial format: kbd,<serial_device>[,baud]
2161 Optional Kernel mode setting:
2162 kms, kbd format: kms,kbd
2163 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud]
2165 kgdboc_earlycon= [KGDB,HW]
2166 If the boot console provides the ability to read
2167 characters and can work in polling mode, you can use
2168 this parameter to tell kgdb to use it as a backend
2169 until the normal console is registered. Intended to
2170 be used together with the kgdboc parameter which
2171 specifies the normal console to transition to.
2173 The name of the early console should be specified
2174 as the value of this parameter. Note that the name of
2175 the early console might be different than the tty
2176 name passed to kgdboc. It's OK to leave the value
2177 blank and the first boot console that implements
2178 read() will be picked.
2180 kgdbwait [KGDB] Stop kernel execution and enter the
2181 kernel debugger at the earliest opportunity.
2183 kmac= [MIPS] korina ethernet MAC address.
2184 Configure the RouterBoard 532 series on-chip
2185 Ethernet adapter MAC address.
2187 kmemleak= [KNL] Boot-time kmemleak enable/disable
2188 Valid arguments: on, off
2190 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y,
2193 kprobe_event=[probe-list]
2194 [FTRACE] Add kprobe events and enable at boot time.
2195 The probe-list is a semicolon delimited list of probe
2196 definitions. Each definition is same as kprobe_events
2197 interface, but the parameters are comma delimited.
2198 For example, to add a kprobe event on vfs_read with
2199 arg1 and arg2, add to the command line;
2201 kprobe_event=p,vfs_read,$arg1,$arg2
2203 See also Documentation/trace/kprobetrace.rst "Kernel
2204 Boot Parameter" section.
2206 kpti= [ARM64] Control page table isolation of user
2207 and kernel address spaces.
2208 Default: enabled on cores which need mitigation.
2212 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs.
2213 Default is 0 (don't ignore, but inject #GP)
2215 kvm.enable_vmware_backdoor=[KVM] Support VMware backdoor PV interface.
2216 Default is false (don't support).
2218 kvm.mmu_audit= [KVM] This is a R/W parameter which allows audit
2223 [KVM] Controls the software workaround for the
2224 X86_BUG_ITLB_MULTIHIT bug.
2225 force : Always deploy workaround.
2226 off : Never deploy workaround.
2227 auto : Deploy workaround based on the presence of
2228 X86_BUG_ITLB_MULTIHIT.
2232 If the software workaround is enabled for the host,
2233 guests do need not to enable it for nested guests.
2235 kvm.nx_huge_pages_recovery_ratio=
2236 [KVM] Controls how many 4KiB pages are periodically zapped
2237 back to huge pages. 0 disables the recovery, otherwise if
2238 the value is N KVM will zap 1/Nth of the 4KiB pages every
2239 minute. The default is 60.
2241 kvm-amd.nested= [KVM,AMD] Allow nested virtualization in KVM/SVM.
2242 Default is 1 (enabled)
2244 kvm-amd.npt= [KVM,AMD] Disable nested paging (virtualized MMU)
2246 Default is 1 (enabled) if in 64-bit or 32-bit PAE mode.
2248 kvm-arm.vgic_v3_group0_trap=
2249 [KVM,ARM] Trap guest accesses to GICv3 group-0
2252 kvm-arm.vgic_v3_group1_trap=
2253 [KVM,ARM] Trap guest accesses to GICv3 group-1
2256 kvm-arm.vgic_v3_common_trap=
2257 [KVM,ARM] Trap guest accesses to GICv3 common
2260 kvm-arm.vgic_v4_enable=
2261 [KVM,ARM] Allow use of GICv4 for direct injection of
2264 kvm-intel.ept= [KVM,Intel] Disable extended page tables
2265 (virtualized MMU) support on capable Intel chips.
2266 Default is 1 (enabled)
2268 kvm-intel.emulate_invalid_guest_state=
2269 [KVM,Intel] Enable emulation of invalid guest states
2270 Default is 0 (disabled)
2272 kvm-intel.flexpriority=
2273 [KVM,Intel] Disable FlexPriority feature (TPR shadow).
2274 Default is 1 (enabled)
2277 [KVM,Intel] Enable VMX nesting (nVMX).
2278 Default is 0 (disabled)
2280 kvm-intel.unrestricted_guest=
2281 [KVM,Intel] Disable unrestricted guest feature
2282 (virtualized real and unpaged mode) on capable
2283 Intel chips. Default is 1 (enabled)
2285 kvm-intel.vmentry_l1d_flush=[KVM,Intel] Mitigation for L1 Terminal Fault
2288 Valid arguments: never, cond, always
2290 always: L1D cache flush on every VMENTER.
2291 cond: Flush L1D on VMENTER only when the code between
2292 VMEXIT and VMENTER can leak host memory.
2293 never: Disables the mitigation
2295 Default is cond (do L1 cache flush in specific instances)
2297 kvm-intel.vpid= [KVM,Intel] Disable Virtual Processor Identification
2298 feature (tagged TLBs) on capable Intel chips.
2299 Default is 1 (enabled)
2301 l1tf= [X86] Control mitigation of the L1TF vulnerability on
2304 The kernel PTE inversion protection is unconditionally
2305 enabled and cannot be disabled.
2308 Provides all available mitigations for the
2309 L1TF vulnerability. Disables SMT and
2310 enables all mitigations in the
2311 hypervisors, i.e. unconditional L1D flush.
2313 SMT control and L1D flush control via the
2314 sysfs interface is still possible after
2315 boot. Hypervisors will issue a warning
2316 when the first VM is started in a
2317 potentially insecure configuration,
2318 i.e. SMT enabled or L1D flush disabled.
2321 Same as 'full', but disables SMT and L1D
2322 flush runtime control. Implies the
2323 'nosmt=force' command line option.
2324 (i.e. sysfs control of SMT is disabled.)
2327 Leaves SMT enabled and enables the default
2328 hypervisor mitigation, i.e. conditional
2331 SMT control and L1D flush control via the
2332 sysfs interface is still possible after
2333 boot. Hypervisors will issue a warning
2334 when the first VM is started in a
2335 potentially insecure configuration,
2336 i.e. SMT enabled or L1D flush disabled.
2340 Disables SMT and enables the default
2341 hypervisor mitigation.
2343 SMT control and L1D flush control via the
2344 sysfs interface is still possible after
2345 boot. Hypervisors will issue a warning
2346 when the first VM is started in a
2347 potentially insecure configuration,
2348 i.e. SMT enabled or L1D flush disabled.
2351 Same as 'flush', but hypervisors will not
2352 warn when a VM is started in a potentially
2353 insecure configuration.
2356 Disables hypervisor mitigations and doesn't
2358 It also drops the swap size and available
2359 RAM limit restriction on both hypervisor and
2364 For details see: Documentation/admin-guide/hw-vuln/l1tf.rst
2370 lapic [X86-32,APIC] Enable the local APIC even if BIOS
2373 lapic= [x86,APIC] "notscdeadline" Do not use TSC deadline
2374 value for LAPIC timer one-shot implementation. Default
2375 back to the programmable timer unit in the LAPIC.
2377 lapic_timer_c2_ok [X86,APIC] trust the local apic timer
2380 libata.dma= [LIBATA] DMA control
2381 libata.dma=0 Disable all PATA and SATA DMA
2382 libata.dma=1 PATA and SATA Disk DMA only
2383 libata.dma=2 ATAPI (CDROM) DMA only
2384 libata.dma=4 Compact Flash DMA only
2385 Combinations also work, so libata.dma=3 enables DMA
2386 for disks and CDROMs, but not CFs.
2388 libata.ignore_hpa= [LIBATA] Ignore HPA limit
2389 libata.ignore_hpa=0 keep BIOS limits (default)
2390 libata.ignore_hpa=1 ignore limits, using full disk
2392 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume
2396 libata.force= [LIBATA] Force configurations. The format is comma
2397 separated list of "[ID:]VAL" where ID is
2398 PORT[.DEVICE]. PORT and DEVICE are decimal numbers
2399 matching port, link or device. Basically, it matches
2400 the ATA ID string printed on console by libata. If
2401 the whole ID part is omitted, the last PORT and DEVICE
2402 values are used. If ID hasn't been specified yet, the
2403 configuration applies to all ports, links and devices.
2405 If only DEVICE is omitted, the parameter applies to
2406 the port and all links and devices behind it. DEVICE
2407 number of 0 either selects the first device or the
2408 first fan-out link behind PMP device. It does not
2409 select the host link. DEVICE number of 15 selects the
2410 host link and device attached to it.
2412 The VAL specifies the configuration to force. As long
2413 as there's no ambiguity shortcut notation is allowed.
2414 For example, both 1.5 and 1.5G would work for 1.5Gbps.
2415 The following configurations can be forced.
2417 * Cable type: 40c, 80c, short40c, unk, ign or sata.
2418 Any ID with matching PORT is used.
2420 * SATA link speed limit: 1.5Gbps or 3.0Gbps.
2422 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7].
2423 udma[/][16,25,33,44,66,100,133] notation is also
2426 * [no]ncq: Turn on or off NCQ.
2428 * [no]ncqtrim: Turn off queued DSM TRIM.
2430 * nohrst, nosrst, norst: suppress hard, soft
2433 * rstonce: only attempt one reset during
2434 hot-unplug link recovery
2436 * dump_id: dump IDENTIFY data.
2438 * atapi_dmadir: Enable ATAPI DMADIR bridge support
2440 * disable: Disable this device.
2442 If there are multiple matching configurations changing
2443 the same attribute, the last one is used.
2445 memblock=debug [KNL] Enable memblock debug messages.
2447 load_ramdisk= [RAM] List of ramdisks to load from floppy
2448 See Documentation/admin-guide/blockdev/ramdisk.rst.
2450 lockd.nlm_grace_period=P [NFS] Assign grace period.
2453 lockd.nlm_tcpport=N [NFS] Assign TCP port.
2456 lockd.nlm_timeout=T [NFS] Assign timeout value.
2459 lockd.nlm_udpport=M [NFS] Assign UDP port.
2462 lockdown= [SECURITY]
2463 { integrity | confidentiality }
2464 Enable the kernel lockdown feature. If set to
2465 integrity, kernel features that allow userland to
2466 modify the running kernel are disabled. If set to
2467 confidentiality, kernel features that allow userland
2468 to extract confidential information from the kernel
2471 locktorture.nreaders_stress= [KNL]
2472 Set the number of locking read-acquisition kthreads.
2473 Defaults to being automatically set based on the
2474 number of online CPUs.
2476 locktorture.nwriters_stress= [KNL]
2477 Set the number of locking write-acquisition kthreads.
2479 locktorture.onoff_holdoff= [KNL]
2480 Set time (s) after boot for CPU-hotplug testing.
2482 locktorture.onoff_interval= [KNL]
2483 Set time (s) between CPU-hotplug operations, or
2484 zero to disable CPU-hotplug testing.
2486 locktorture.shuffle_interval= [KNL]
2487 Set task-shuffle interval (jiffies). Shuffling
2488 tasks allows some CPUs to go into dyntick-idle
2489 mode during the locktorture test.
2491 locktorture.shutdown_secs= [KNL]
2492 Set time (s) after boot system shutdown. This
2493 is useful for hands-off automated testing.
2495 locktorture.stat_interval= [KNL]
2496 Time (s) between statistics printk()s.
2498 locktorture.stutter= [KNL]
2499 Time (s) to stutter testing, for example,
2500 specifying five seconds causes the test to run for
2501 five seconds, wait for five seconds, and so on.
2502 This tests the locking primitive's ability to
2503 transition abruptly to and from idle.
2505 locktorture.torture_type= [KNL]
2506 Specify the locking implementation to test.
2508 locktorture.verbose= [KNL]
2509 Enable additional printk() statements.
2511 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver
2514 loglevel= All Kernel Messages with a loglevel smaller than the
2515 console loglevel will be printed to the console. It can
2516 also be changed with klogd or other programs. The
2517 loglevels are defined as follows:
2519 0 (KERN_EMERG) system is unusable
2520 1 (KERN_ALERT) action must be taken immediately
2521 2 (KERN_CRIT) critical conditions
2522 3 (KERN_ERR) error conditions
2523 4 (KERN_WARNING) warning conditions
2524 5 (KERN_NOTICE) normal but significant condition
2525 6 (KERN_INFO) informational
2526 7 (KERN_DEBUG) debug-level messages
2528 log_buf_len=n[KMG] Sets the size of the printk ring buffer,
2529 in bytes. n must be a power of two and greater
2530 than the minimal size. The minimal size is defined
2531 by LOG_BUF_SHIFT kernel config parameter. There is
2532 also CONFIG_LOG_CPU_MAX_BUF_SHIFT config parameter
2533 that allows to increase the default size depending on
2534 the number of CPUs. See init/Kconfig for more details.
2536 logo.nologo [FB] Disables display of the built-in Linux logo.
2537 This may be used to provide more screen space for
2538 kernel log messages and is useful when debugging
2539 kernel boot problems.
2541 lp=0 [LP] Specify parallel ports to use, e.g,
2542 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses
2543 lp=reset first parallel port). 'lp=0' disables the
2544 lp=auto printer driver. 'lp=reset' (which can be
2545 specified in addition to the ports) causes
2546 attached printers to be reset. Using
2547 lp=port1,port2,... specifies the parallel ports
2548 to associate lp devices with, starting with
2549 lp0. A port specification may be 'none' to skip
2550 that lp device, or a parport name such as
2551 'parport0'. Specifying 'lp=auto' instead of a
2552 port specification list means that device IDs
2553 from each port should be examined, to see if
2554 an IEEE 1284-compliant printer is attached; if
2555 so, the driver will manage that printer.
2556 See also header of drivers/char/lp.c.
2559 Sets loops_per_jiffy to given constant, thus avoiding
2560 time-consuming boot-time autodetection (up to 250 ms per
2561 CPU). 0 enables autodetection (default). To determine
2562 the correct value for your kernel, boot with normal
2563 autodetection and see what value is printed. Note that
2564 on SMP systems the preset will be applied to all CPUs,
2565 which is likely to cause problems if your CPUs need
2566 significantly divergent settings. An incorrect value
2567 will cause delays in the kernel to be wrong, leading to
2568 unpredictable I/O errors and other breakage. Although
2569 unlikely, in the extreme case this might damage your
2573 Format: <io>,<irq>,<dma>
2575 lsm.debug [SECURITY] Enable LSM initialization debugging output.
2578 [SECURITY] Choose order of LSM initialization. This
2579 overrides CONFIG_LSM, and the "security=" parameter.
2581 machvec= [IA-64] Force the use of a particular machine-vector
2582 (machvec) in a generic kernel.
2583 Example: machvec=hpzx1
2585 machtype= [Loongson] Share the same kernel image file between different
2587 Example: machtype=lemote-yeeloong-2f-7inch
2589 max_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory greater
2590 than or equal to this physical address is ignored.
2592 maxcpus= [SMP] Maximum number of processors that an SMP kernel
2593 will bring up during bootup. maxcpus=n : n >= 0 limits
2594 the kernel to bring up 'n' processors. Surely after
2595 bootup you can bring up the other plugged cpu by executing
2596 "echo 1 > /sys/devices/system/cpu/cpuX/online". So maxcpus
2597 only takes effect during system bootup.
2598 While n=0 is a special case, it is equivalent to "nosmp",
2599 which also disables the IO APIC.
2601 max_loop= [LOOP] The number of loop block devices that get
2602 (loop.max_loop) unconditionally pre-created at init time. The default
2603 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead
2604 of statically allocating a predefined number, loop
2605 devices can be requested on-demand with the
2606 /dev/loop-control interface.
2608 mce [X86-32] Machine Check Exception
2610 mce=option [X86-64] See Documentation/x86/x86_64/boot-options.rst
2612 md= [HW] RAID subsystems devices and level
2613 See Documentation/admin-guide/md.rst.
2616 Format: <first>,<last>
2617 Specifies range of consoles to be captured by the MDA.
2620 Control mitigation for the Micro-architectural Data
2621 Sampling (MDS) vulnerability.
2623 Certain CPUs are vulnerable to an exploit against CPU
2624 internal buffers which can forward information to a
2625 disclosure gadget under certain conditions.
2627 In vulnerable processors, the speculatively
2628 forwarded data can be used in a cache side channel
2629 attack, to access data to which the attacker does
2630 not have direct access.
2632 This parameter controls the MDS mitigation. The
2635 full - Enable MDS mitigation on vulnerable CPUs
2636 full,nosmt - Enable MDS mitigation and disable
2637 SMT on vulnerable CPUs
2638 off - Unconditionally disable MDS mitigation
2640 On TAA-affected machines, mds=off can be prevented by
2641 an active TAA mitigation as both vulnerabilities are
2642 mitigated with the same mechanism so in order to disable
2643 this mitigation, you need to specify tsx_async_abort=off
2646 Not specifying this option is equivalent to
2649 For details see: Documentation/admin-guide/hw-vuln/mds.rst
2651 mem=nn[KMG] [KNL,BOOT] Force usage of a specific amount of memory
2652 Amount of memory to be used in cases as follows:
2655 2 when the kernel is not able to see the whole system memory;
2656 3 memory that lies after 'mem=' boundary is excluded from
2657 the hypervisor, then assigned to KVM guests.
2659 [X86] Work as limiting max address. Use together
2660 with memmap= to avoid physical address space collisions.
2661 Without memmap= PCI devices could be placed at addresses
2662 belonging to unused RAM.
2664 Note that this only takes effects during boot time since
2665 in above case 3, memory may need be hot added after boot
2666 if system memory of hypervisor is not sufficient.
2668 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel
2672 [KNL,SH] Allow user to override the default size for
2673 per-device physically contiguous DMA buffers.
2675 memhp_default_state=online/offline
2676 [KNL] Set the initial state for the memory hotplug
2677 onlining policy. If not specified, the default value is
2678 set according to the
2679 CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE kernel config
2681 See Documentation/admin-guide/mm/memory-hotplug.rst.
2683 memmap=exactmap [KNL,X86] Enable setting of an exact
2684 E820 memory map, as specified by the user.
2685 Such memmap=exactmap lines can be constructed based on
2686 BIOS output or other requirements. See the memmap=nn@ss
2689 memmap=nn[KMG]@ss[KMG]
2690 [KNL] Force usage of a specific region of memory.
2691 Region of memory to be used is from ss to ss+nn.
2692 If @ss[KMG] is omitted, it is equivalent to mem=nn[KMG],
2693 which limits max address to nn[KMG].
2694 Multiple different regions can be specified,
2697 memmap=100M@2G,100M#3G,1G!1024G
2699 memmap=nn[KMG]#ss[KMG]
2700 [KNL,ACPI] Mark specific memory as ACPI data.
2701 Region of memory to be marked is from ss to ss+nn.
2703 memmap=nn[KMG]$ss[KMG]
2704 [KNL,ACPI] Mark specific memory as reserved.
2705 Region of memory to be reserved is from ss to ss+nn.
2706 Example: Exclude memory from 0x18690000-0x1869ffff
2707 memmap=64K$0x18690000
2709 memmap=0x10000$0x18690000
2710 Some bootloaders may need an escape character before '$',
2711 like Grub2, otherwise '$' and the following number
2714 memmap=nn[KMG]!ss[KMG]
2715 [KNL,X86] Mark specific memory as protected.
2716 Region of memory to be used, from ss to ss+nn.
2717 The memory region may be marked as e820 type 12 (0xc)
2718 and is NVDIMM or ADR memory.
2720 memmap=<size>%<offset>-<oldtype>+<newtype>
2721 [KNL,ACPI] Convert memory within the specified region
2722 from <oldtype> to <newtype>. If "-<oldtype>" is left
2723 out, the whole region will be marked as <newtype>,
2724 even if previously unavailable. If "+<newtype>" is left
2725 out, matching memory will be removed. Types are
2726 specified as e820 types, e.g., 1 = RAM, 2 = reserved,
2727 3 = ACPI, 12 = PRAM.
2729 memory_corruption_check=0/1 [X86]
2730 Some BIOSes seem to corrupt the first 64k of
2731 memory when doing things like suspend/resume.
2732 Setting this option will scan the memory
2733 looking for corruption. Enabling this will
2734 both detect corruption and prevent the kernel
2735 from using the memory being corrupted.
2736 However, its intended as a diagnostic tool; if
2737 repeatable BIOS-originated corruption always
2738 affects the same memory, you can use memmap=
2739 to prevent the kernel from using that memory.
2741 memory_corruption_check_size=size [X86]
2742 By default it checks for corruption in the low
2743 64k, making this memory unavailable for normal
2744 use. Use this parameter to scan for
2745 corruption in more or less memory.
2747 memory_corruption_check_period=seconds [X86]
2748 By default it checks for corruption every 60
2749 seconds. Use this parameter to check at some
2750 other rate. 0 disables periodic checking.
2752 memtest= [KNL,X86,ARM,PPC] Enable memtest
2754 default : 0 <disable>
2755 Specifies the number of memtest passes to be
2756 performed. Each pass selects another test
2757 pattern from a given set of patterns. Memtest
2758 fills the memory with this pattern, validates
2759 memory contents and reserves bad memory
2760 regions that are detected.
2762 mem_encrypt= [X86-64] AMD Secure Memory Encryption (SME) control
2763 Valid arguments: on, off
2764 Default (depends on kernel configuration option):
2765 on (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=y)
2766 off (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=n)
2767 mem_encrypt=on: Activate SME
2768 mem_encrypt=off: Do not activate SME
2770 Refer to Documentation/virt/kvm/amd-memory-encryption.rst
2771 for details on when memory encryption can be activated.
2773 mem_sleep_default= [SUSPEND] Default system suspend mode:
2774 s2idle - Suspend-To-Idle
2775 shallow - Power-On Suspend or equivalent (if supported)
2776 deep - Suspend-To-RAM or equivalent (if supported)
2777 See Documentation/admin-guide/pm/sleep-states.rst.
2779 meye.*= [HW] Set MotionEye Camera parameters
2780 See Documentation/admin-guide/media/meye.rst.
2782 mfgpt_irq= [IA-32] Specify the IRQ to use for the
2783 Multi-Function General Purpose Timers on AMD Geode
2786 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when
2787 the BIOS has incorrectly applied a workaround. TinyBIOS
2788 version 0.98 is known to be affected, 0.99 fixes the
2789 problem by letting the user disable the workaround.
2793 min_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory below this
2794 physical address is ignored.
2796 mini2440= [ARM,HW,KNL]
2797 Format:[0..2][b][c][t]
2799 MINI2440 configuration specification:
2800 0 - The attached screen is the 3.5" TFT
2801 1 - The attached screen is the 7" TFT
2802 2 - The VGA Shield is attached (1024x768)
2803 Leaving out the screen size parameter will not load
2804 the TFT driver, and the framebuffer will be left
2806 b - Enable backlight. The TFT backlight pin will be
2807 linked to the kernel VESA blanking code and a GPIO
2808 LED. This parameter is not necessary when using the
2810 c - Enable the s3c camera interface.
2811 t - Reserved for enabling touchscreen support. The
2812 touchscreen support is not enabled in the mainstream
2813 kernel as of 2.6.30, a preliminary port can be found
2814 in the "bleeding edge" mini2440 support kernel at
2815 https://repo.or.cz/w/linux-2.6/mini2440.git
2818 [X86,PPC,S390,ARM64] Control optional mitigations for
2819 CPU vulnerabilities. This is a set of curated,
2820 arch-independent options, each of which is an
2821 aggregation of existing arch-specific options.
2824 Disable all optional CPU mitigations. This
2825 improves system performance, but it may also
2826 expose users to several CPU vulnerabilities.
2827 Equivalent to: nopti [X86,PPC]
2829 nospectre_v1 [X86,PPC]
2831 nospectre_v2 [X86,PPC,S390,ARM64]
2832 spectre_v2_user=off [X86]
2833 spec_store_bypass_disable=off [X86,PPC]
2834 ssbd=force-off [ARM64]
2837 tsx_async_abort=off [X86]
2838 kvm.nx_huge_pages=off [X86]
2841 This does not have any effect on
2842 kvm.nx_huge_pages when
2843 kvm.nx_huge_pages=force.
2846 Mitigate all CPU vulnerabilities, but leave SMT
2847 enabled, even if it's vulnerable. This is for
2848 users who don't want to be surprised by SMT
2849 getting disabled across kernel upgrades, or who
2850 have other ways of avoiding SMT-based attacks.
2851 Equivalent to: (default behavior)
2854 Mitigate all CPU vulnerabilities, disabling SMT
2855 if needed. This is for users who always want to
2856 be fully mitigated, even if it means losing SMT.
2857 Equivalent to: l1tf=flush,nosmt [X86]
2858 mds=full,nosmt [X86]
2859 tsx_async_abort=full,nosmt [X86]
2862 [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this
2863 parameter allows control of the logging verbosity for
2864 the additional memory initialisation checks. A value
2865 of 0 disables mminit logging and a level of 4 will
2866 log everything. Information is printed at KERN_DEBUG
2867 so loglevel=8 may also need to be specified.
2870 [KNL] When CONFIG_MODULE_SIG is set, this means that
2871 modules without (valid) signatures will fail to load.
2872 Note that if CONFIG_MODULE_SIG_FORCE is set, that
2873 is always true, so this option does nothing.
2875 module_blacklist= [KNL] Do not load a comma-separated list of
2876 modules. Useful for debugging problem modules.
2879 [MOUSE] Maximum time between finger touching and
2880 leaving touchpad surface for touch to be considered
2881 a tap and be reported as a left button click (for
2882 touchpads working in absolute mode only).
2884 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices
2885 reporting absolute coordinates, such as tablets
2886 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices
2887 reporting absolute coordinates, such as tablets
2889 movablecore= [KNL,X86,IA-64,PPC]
2890 Format: nn[KMGTPE] | nn%
2891 This parameter is the complement to kernelcore=, it
2892 specifies the amount of memory used for migratable
2893 allocations. If both kernelcore and movablecore is
2894 specified, then kernelcore will be at *least* the
2895 specified value but may be more. If movablecore on its
2896 own is specified, the administrator must be careful
2897 that the amount of memory usable for all allocations
2900 movable_node [KNL] Boot-time switch to make hotplugable memory
2901 NUMA nodes to be movable. This means that the memory
2902 of such nodes will be usable only for movable
2903 allocations which rules out almost all kernel
2904 allocations. Use with caution!
2906 MTD_Partition= [MTD]
2907 Format: <name>,<region-number>,<size>,<offset>
2909 MTD_Region= [MTD] Format:
2910 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>]
2913 See drivers/mtd/parsers/cmdlinepart.c
2915 multitce=off [PPC] This parameter disables the use of the pSeries
2916 firmware feature for updating multiple TCE entries
2919 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration
2921 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock]
2923 boundary - index of last SLC block on Flex-OneNAND.
2924 The remaining blocks are configured as MLC blocks.
2925 lock - Configure if Flex-OneNAND boundary should be locked.
2926 Once locked, the boundary cannot be changed.
2927 1 indicates lock status, 0 indicates unlock status.
2930 ARM/S3C2412 JIVE boot control
2932 See arch/arm/mach-s3c2412/mach-jive.c
2934 mtouchusb.raw_coordinates=
2935 [HW] Make the MicroTouch USB driver use raw coordinates
2936 ('y', default) or cooked coordinates ('n')
2938 mtrr_chunk_size=nn[KMG] [X86]
2939 used for mtrr cleanup. It is largest continuous chunk
2940 that could hold holes aka. UC entries.
2942 mtrr_gran_size=nn[KMG] [X86]
2943 Used for mtrr cleanup. It is granularity of mtrr block.
2945 Large value could prevent small alignment from
2948 mtrr_spare_reg_nr=n [X86]
2950 Range: 0,7 : spare reg number
2952 Used for mtrr cleanup. It is spare mtrr entries number.
2953 Set to 2 or more if your graphical card needs more.
2955 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card
2957 netdev= [NET] Network devices parameters
2958 Format: <irq>,<io>,<mem_start>,<mem_end>,<name>
2959 Note that mem_start is often overloaded to mean
2960 something different and driver-specific.
2961 This usage is only documented in each driver source
2965 [NETFILTER] Enable connection tracking flow accounting
2966 0 to disable accounting
2967 1 to enable accounting
2970 nfsaddrs= [NFS] Deprecated. Use ip= instead.
2971 See Documentation/admin-guide/nfs/nfsroot.rst.
2973 nfsroot= [NFS] nfs root filesystem for disk-less boxes.
2974 See Documentation/admin-guide/nfs/nfsroot.rst.
2976 nfsrootdebug [NFS] enable nfsroot debugging messages.
2977 See Documentation/admin-guide/nfs/nfsroot.rst.
2979 nfs.callback_nr_threads=
2980 [NFSv4] set the total number of threads that the
2981 NFS client will assign to service NFSv4 callback
2984 nfs.callback_tcpport=
2985 [NFS] set the TCP port on which the NFSv4 callback
2986 channel should listen.
2989 [NFS] sets the pathname to the program which is used
2990 to update the NFS client cache entries.
2992 nfs.cache_getent_timeout=
2993 [NFS] sets the timeout after which an attempt to
2994 update a cache entry is deemed to have failed.
2996 nfs.idmap_cache_timeout=
2997 [NFS] set the maximum lifetime for idmapper cache
3001 [NFS] enable 64-bit inode numbers.
3002 If zero, the NFS client will fake up a 32-bit inode
3003 number for the readdir() and stat() syscalls instead
3004 of returning the full 64-bit number.
3005 The default is to return 64-bit inode numbers.
3007 nfs.max_session_cb_slots=
3008 [NFSv4.1] Sets the maximum number of session
3009 slots the client will assign to the callback
3010 channel. This determines the maximum number of
3011 callbacks the client will process in parallel for
3012 a particular server.
3014 nfs.max_session_slots=
3015 [NFSv4.1] Sets the maximum number of session slots
3016 the client will attempt to negotiate with the server.
3017 This limits the number of simultaneous RPC requests
3018 that the client can send to the NFSv4.1 server.
3019 Note that there is little point in setting this
3020 value higher than the max_tcp_slot_table_limit.
3022 nfs.nfs4_disable_idmapping=
3023 [NFSv4] When set to the default of '1', this option
3024 ensures that both the RPC level authentication
3025 scheme and the NFS level operations agree to use
3026 numeric uids/gids if the mount is using the
3027 'sec=sys' security flavour. In effect it is
3028 disabling idmapping, which can make migration from
3029 legacy NFSv2/v3 systems to NFSv4 easier.
3030 Servers that do not support this mode of operation
3031 will be autodetected by the client, and it will fall
3032 back to using the idmapper.
3033 To turn off this behaviour, set the value to '0'.
3035 [NFS4] Specify an additional fixed unique ident-
3036 ification string that NFSv4 clients can insert into
3037 their nfs_client_id4 string. This is typically a
3038 UUID that is generated at system install time.
3040 nfs.send_implementation_id =
3041 [NFSv4.1] Send client implementation identification
3042 information in exchange_id requests.
3043 If zero, no implementation identification information
3045 The default is to send the implementation identification
3048 nfs.recover_lost_locks =
3049 [NFSv4] Attempt to recover locks that were lost due
3050 to a lease timeout on the server. Please note that
3051 doing this risks data corruption, since there are
3052 no guarantees that the file will remain unchanged
3053 after the locks are lost.
3054 If you want to enable the kernel legacy behaviour of
3055 attempting to recover these locks, then set this
3057 The default parameter value of '0' causes the kernel
3058 not to attempt recovery of lost locks.
3060 nfs4.layoutstats_timer =
3061 [NFSv4.2] Change the rate at which the kernel sends
3062 layoutstats to the pNFS metadata server.
3064 Setting this to value to 0 causes the kernel to use
3065 whatever value is the default set by the layout
3066 driver. A non-zero value sets the minimum interval
3067 in seconds between layoutstats transmissions.
3069 nfsd.nfs4_disable_idmapping=
3070 [NFSv4] When set to the default of '1', the NFSv4
3071 server will return only numeric uids and gids to
3072 clients using auth_sys, and will accept numeric uids
3073 and gids from such clients. This is intended to ease
3074 migration from NFSv2/v3.
3076 nmi_debug= [KNL,SH] Specify one or more actions to take
3077 when a NMI is triggered.
3078 Format: [state][,regs][,debounce][,die]
3080 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels
3081 Format: [panic,][nopanic,][num]
3083 0 - turn hardlockup detector in nmi_watchdog off
3084 1 - turn hardlockup detector in nmi_watchdog on
3085 When panic is specified, panic when an NMI watchdog
3086 timeout occurs (or 'nopanic' to not panic on an NMI
3087 watchdog, if CONFIG_BOOTPARAM_HARDLOCKUP_PANIC is set)
3088 To disable both hard and soft lockup detectors,
3089 please see 'nowatchdog'.
3090 This is useful when you use a panic=... timeout and
3091 need the box quickly up again.
3093 These settings can be accessed at runtime via
3094 the nmi_watchdog and hardlockup_panic sysctls.
3096 netpoll.carrier_timeout=
3097 [NET] Specifies amount of time (in seconds) that
3098 netpoll should wait for a carrier. By default netpoll
3101 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths
3102 emulation library even if a 387 maths coprocessor
3105 no5lvl [X86-64] Disable 5-level paging mode. Forces
3106 kernel to use 4-level paging instead.
3108 nofsgsbase [X86] Disables FSGSBASE instructions.
3111 [HW] Never suspend the console
3112 Disable suspending of consoles during suspend and
3113 hibernate operations. Once disabled, debugging
3114 messages can reach various consoles while the rest
3115 of the system is being put to sleep (ie, while
3116 debugging driver suspend/resume hooks). This may
3117 not work reliably with all consoles, but is known
3118 to work with serial and VGA consoles.
3119 To facilitate more flexible debugging, we also add
3120 console_suspend, a printk module parameter to control
3121 it. Users could use console_suspend (usually
3122 /sys/module/printk/parameters/console_suspend) to
3123 turn on/off it dynamically.
3125 novmcoredd [KNL,KDUMP]
3126 Disable device dump. Device dump allows drivers to
3127 append dump data to vmcore so you can collect driver
3128 specified debug info. Drivers can append the data
3129 without any limit and this data is stored in memory,
3130 so this may cause significant memory stress. Disabling
3131 device dump can help save memory but the driver debug
3132 data will be no longer available. This parameter
3133 is only available when CONFIG_PROC_VMCORE_DEVICE_DUMP
3136 noaliencache [MM, NUMA, SLAB] Disables the allocation of alien
3137 caches in the slab allocator. Saves per-node memory,
3138 but will impact performance.
3142 noaltinstr [S390] Disables alternative instructions patching
3143 (CPU alternatives feature).
3145 noapic [SMP,APIC] Tells the kernel to not make use of any
3146 IOAPICs that may be present in the system.
3148 noautogroup Disable scheduler automatic task group creation.
3150 nobats [PPC] Do not use BATs for mapping kernel lowmem
3151 on "Classic" PPC cores.
3155 noclflush [BUGS=X86] Don't use the CLFLUSH instruction
3157 nodelayacct [KNL] Disable per-task delay accounting
3159 nodsp [SH] Disable hardware DSP at boot time.
3161 noefi Disable EFI runtime services support.
3166 On X86-32 available only on PAE configured kernels.
3167 noexec=on: enable non-executable mappings (default)
3168 noexec=off: disable non-executable mappings
3171 Disable SMAP (Supervisor Mode Access Prevention)
3172 even if it is supported by processor.
3175 Disable SMEP (Supervisor Mode Execution Prevention)
3176 even if it is supported by processor.
3179 This affects only 32-bit executables.
3180 noexec32=on: enable non-executable mappings (default)
3181 read doesn't imply executable mappings
3182 noexec32=off: disable non-executable mappings
3183 read implies executable mappings
3185 nofpu [MIPS,SH] Disable hardware FPU at boot time.
3187 nofxsr [BUGS=X86-32] Disables x86 floating point extended
3188 register save and restore. The kernel will only save
3189 legacy floating-point registers on task switch.
3191 nohugeiomap [KNL,x86,PPC] Disable kernel huge I/O mappings.
3193 nosmt [KNL,S390] Disable symmetric multithreading (SMT).
3194 Equivalent to smt=1.
3196 [KNL,x86] Disable symmetric multithreading (SMT).
3197 nosmt=force: Force disable SMT, cannot be undone
3198 via the sysfs control file.
3200 nospectre_v1 [X86,PPC] Disable mitigations for Spectre Variant 1
3201 (bounds check bypass). With this option data leaks are
3202 possible in the system.
3204 nospectre_v2 [X86,PPC_FSL_BOOK3E,ARM64] Disable all mitigations for
3205 the Spectre variant 2 (indirect branch prediction)
3206 vulnerability. System may allow data leaks with this
3209 nospec_store_bypass_disable
3210 [HW] Disable all mitigations for the Speculative Store Bypass vulnerability
3212 noxsave [BUGS=X86] Disables x86 extended register state save
3213 and restore using xsave. The kernel will fallback to
3214 enabling legacy floating-point and sse state.
3216 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended
3217 register states. The kernel will fall back to use
3218 xsave to save the states. By using this parameter,
3219 performance of saving the states is degraded because
3220 xsave doesn't support modified optimization while
3221 xsaveopt supports it on xsaveopt enabled systems.
3223 noxsaves [X86] Disables xsaves and xrstors used in saving and
3224 restoring x86 extended register state in compacted
3225 form of xsave area. The kernel will fall back to use
3226 xsaveopt and xrstor to save and restore the states
3227 in standard form of xsave area. By using this
3228 parameter, xsave area per process might occupy more
3229 memory on xsaves enabled systems.
3231 nohlt [BUGS=ARM,SH] Tells the kernel that the sleep(SH) or
3232 wfi(ARM) instruction doesn't work correctly and not to
3233 use it. This is also useful when using JTAG debugger.
3235 no_file_caps Tells the kernel not to honor file capabilities. The
3236 only way then for a file to be executed with privilege
3237 is to be setuid root or executed by root.
3239 nohalt [IA-64] Tells the kernel not to use the power saving
3240 function PAL_HALT_LIGHT when idle. This increases
3241 power-consumption. On the positive side, it reduces
3242 interrupt wake-up latency, which may improve performance
3243 in certain environments such as networked servers or
3246 nohibernate [HIBERNATION] Disable hibernation and resume.
3248 nohz= [KNL] Boottime enable/disable dynamic ticks
3249 Valid arguments: on, off
3252 nohz_full= [KNL,BOOT,SMP,ISOL]
3253 The argument is a cpu list, as described above.
3254 In kernels built with CONFIG_NO_HZ_FULL=y, set
3255 the specified list of CPUs whose tick will be stopped
3256 whenever possible. The boot CPU will be forced outside
3257 the range to maintain the timekeeping. Any CPUs
3258 in this list will have their RCU callbacks offloaded,
3259 just as if they had also been called out in the
3260 rcu_nocbs= boot parameter.
3262 noiotrap [SH] Disables trapped I/O port accesses.
3264 noirqdebug [X86-32] Disables the code which attempts to detect and
3265 disable unhandled interrupt sources.
3267 no_timer_check [X86,APIC] Disables the code which tests for
3268 broken timer IRQ sources.
3270 noisapnp [ISAPNP] Disables ISA PnP code.
3272 noinitrd [RAM] Tells the kernel not to load any configured
3275 nointremap [X86-64, Intel-IOMMU] Do not enable interrupt
3277 [Deprecated - use intremap=off]
3281 noinvpcid [X86] Disable the INVPCID cpu feature.
3283 nojitter [IA-64] Disables jitter checking for ITC timers.
3285 no-kvmclock [X86,KVM] Disable paravirtualized KVM clock driver
3287 no-kvmapf [X86,KVM] Disable paravirtualized asynchronous page
3291 [X86,PV_OPS] Disable paravirtualized VMware scheduler
3292 clock and use the default one.
3294 no-steal-acc [X86,PV_OPS,ARM64] Disable paravirtualized steal time
3295 accounting. steal time is computed, but won't
3296 influence scheduler behaviour
3298 nolapic [X86-32,APIC] Do not enable or use the local APIC.
3300 nolapic_timer [X86-32,APIC] Do not use the local APIC timer.
3302 noltlbs [PPC] Do not use large page/tlb entries for kernel
3303 lowmem mapping on PPC40x and PPC8xx
3305 nomca [IA-64] Disable machine check abort handling
3307 nomce [X86-32] Disable Machine Check Exception
3309 nomfgpt [X86-32] Disable Multi-Function General Purpose
3310 Timer usage (for AMD Geode machines).
3312 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to
3313 shutdown the other cpus. Instead use the REBOOT_VECTOR
3316 nomodule Disable module load
3318 nopat [X86] Disable PAT (page attribute table extension of
3319 pagetables) support.
3321 nopcid [X86-64] Disable the PCID cpu feature.
3323 norandmaps Don't use address space randomization. Equivalent to
3324 echo 0 > /proc/sys/kernel/randomize_va_space
3326 noreplace-smp [X86-32,SMP] Don't replace SMP instructions
3327 with UP alternatives
3329 nordrand [X86] Disable kernel use of the RDRAND and
3330 RDSEED instructions even if they are supported
3331 by the processor. RDRAND and RDSEED are still
3332 available to user space applications.
3334 noresume [SWSUSP] Disables resume and restores original swap
3337 no-scroll [VGA] Disables scrollback.
3338 This is required for the Braillex ib80-piezo Braille
3339 reader made by F.H. Papenmeier (Germany).
3343 nosep [BUGS=X86-32] Disables x86 SYSENTER/SYSEXIT support.
3345 nosmp [SMP] Tells an SMP kernel to act as a UP kernel,
3346 and disable the IO APIC. legacy for "maxcpus=0".
3348 nosoftlockup [KNL] Disable the soft-lockup detector.
3350 nosync [HW,M68K] Disables sync negotiation for all devices.
3352 nowatchdog [KNL] Disable both lockup detectors, i.e.
3353 soft-lockup and NMI watchdog (hard-lockup).
3357 nox2apic [X86-64,APIC] Do not enable x2APIC mode.
3359 cpu0_hotplug [X86] Turn on CPU0 hotplug feature when
3360 CONFIG_BOOTPARAM_HOTPLUG_CPU0 is off.
3361 Some features depend on CPU0. Known dependencies are:
3362 1. Resume from suspend/hibernate depends on CPU0.
3363 Suspend/hibernate will fail if CPU0 is offline and you
3364 need to online CPU0 before suspend/hibernate.
3365 2. PIC interrupts also depend on CPU0. CPU0 can't be
3366 removed if a PIC interrupt is detected.
3367 It's said poweroff/reboot may depend on CPU0 on some
3368 machines although I haven't seen such issues so far
3369 after CPU0 is offline on a few tested machines.
3370 If the dependencies are under your control, you can
3371 turn on cpu0_hotplug.
3373 nps_mtm_hs_ctr= [KNL,ARC]
3374 This parameter sets the maximum duration, in
3375 cycles, each HW thread of the CTOP can run
3376 without interruptions, before HW switches it.
3377 The actual maximum duration is 16 times this
3379 Format: integer between 1 and 255
3382 nptcg= [IA-64] Override max number of concurrent global TLB
3383 purges which is reported from either PAL_VM_SUMMARY or
3386 nr_cpus= [SMP] Maximum number of processors that an SMP kernel
3387 could support. nr_cpus=n : n >= 1 limits the kernel to
3388 support 'n' processors. It could be larger than the
3389 number of already plugged CPU during bootup, later in
3390 runtime you can physically add extra cpu until it reaches
3391 n. So during boot up some boot time memory for per-cpu
3392 variables need be pre-allocated for later physical cpu
3395 nr_uarts= [SERIAL] maximum number of UARTs to be registered.
3397 numa_balancing= [KNL,X86] Enable or disable automatic NUMA balancing.
3398 Allowed values are enable and disable
3400 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA.
3401 'node', 'default' can be specified
3402 This can be set from sysctl after boot.
3403 See Documentation/admin-guide/sysctl/vm.rst for details.
3405 ohci1394_dma=early [HW] enable debugging via the ohci1394 driver.
3406 See Documentation/core-api/debugging-via-ohci1394.rst for more
3409 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands
3410 Rather than timing out after 20 ms if an EC
3411 command is not properly ACKed, override the length
3412 of the timeout. We have interrupts disabled while
3413 waiting for the ACK, so if this is set too high
3414 interrupts *may* be lost!
3416 omap_mux= [OMAP] Override bootloader pin multiplexing.
3417 Format: <mux_mode0.mode_name=value>...
3418 For example, to override I2C bus2:
3419 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100
3421 oprofile.timer= [HW]
3422 Use timer interrupt instead of performance counters
3424 oprofile.cpu_type= Force an oprofile cpu type
3425 This might be useful if you have an older oprofile
3426 userland or if you want common events.
3427 Format: { arch_perfmon }
3428 arch_perfmon: [X86] Force use of architectural
3429 perfmon on Intel CPUs instead of the
3430 CPU specific event set.
3431 timer: [X86] Force use of architectural NMI
3432 timer mode (see also oprofile.timer
3433 for generic hr timer mode)
3435 oops=panic Always panic on oopses. Default is to just kill the
3436 process, but there is a small probability of
3437 deadlocking the machine.
3438 This will also cause panics on machine check exceptions.
3439 Useful together with panic=30 to trigger a reboot.
3442 [KNL] Boolean flag to control whether the page allocator
3443 should randomize its free lists. The randomization may
3444 be automatically enabled if the kernel detects it is
3445 running on a platform with a direct-mapped memory-side
3446 cache, and this parameter can be used to
3447 override/disable that behavior. The state of the flag
3448 can be read from sysfs at:
3449 /sys/module/page_alloc/parameters/shuffle.
3451 page_owner= [KNL] Boot-time page_owner enabling option.
3452 Storage of the information about who allocated
3453 each page is disabled in default. With this switch,
3455 on: enable the feature
3457 page_poison= [KNL] Boot-time parameter changing the state of
3458 poisoning on the buddy allocator, available with
3459 CONFIG_PAGE_POISONING=y.
3460 off: turn off poisoning (default)
3461 on: turn on poisoning
3463 panic= [KNL] Kernel behaviour on panic: delay <timeout>
3464 timeout > 0: seconds before rebooting
3465 timeout = 0: wait forever
3466 timeout < 0: reboot immediately
3469 panic_print= Bitmask for printing system info when panic happens.
3470 User can chose combination of the following bits:
3471 bit 0: print all tasks info
3472 bit 1: print system memory info
3473 bit 2: print timer info
3474 bit 3: print locks info if CONFIG_LOCKDEP is on
3475 bit 4: print ftrace buffer
3476 bit 5: print all printk messages in buffer
3478 panic_on_taint= Bitmask for conditionally calling panic() in add_taint()
3479 Format: <hex>[,nousertaint]
3480 Hexadecimal bitmask representing the set of TAINT flags
3481 that will cause the kernel to panic when add_taint() is
3482 called with any of the flags in this set.
3483 The optional switch "nousertaint" can be utilized to
3484 prevent userspace forced crashes by writing to sysctl
3485 /proc/sys/kernel/tainted any flagset matching with the
3486 bitmask set on panic_on_taint.
3487 See Documentation/admin-guide/tainted-kernels.rst for
3488 extra details on the taint flags that users can pick
3489 to compose the bitmask to assign to panic_on_taint.
3491 panic_on_warn panic() instead of WARN(). Useful to cause kdump
3494 crash_kexec_post_notifiers
3495 Run kdump after running panic-notifiers and dumping
3496 kmsg. This only for the users who doubt kdump always
3497 succeeds in any situation.
3498 Note that this also increases risks of kdump failure,
3499 because some panic notifiers can make the crashed
3500 kernel more unstable.
3502 parkbd.port= [HW] Parallel port number the keyboard adapter is
3503 connected to, default is 0.
3505 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation,
3506 0 for XT, 1 for AT (default is AT).
3509 parport= [HW,PPT] Specify parallel ports. 0 disables.
3510 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] }
3511 Use 'auto' to force the driver to use any
3512 IRQ/DMA settings detected (the default is to
3513 ignore detected IRQ/DMA settings because of
3514 possible conflicts). You can specify the base
3515 address, IRQ, and DMA settings; IRQ and DMA
3516 should be numbers, or 'auto' (for using detected
3517 settings on that particular port), or 'nofifo'
3518 (to avoid using a FIFO even if it is detected).
3519 Parallel ports are assigned in the order they
3520 are specified on the command line, starting
3523 parport_init_mode= [HW,PPT]
3524 Configure VIA parallel port to operate in
3525 a specific mode. This is necessary on Pegasos
3526 computer where firmware has no options for setting
3527 up parallel port mode and sets it to spp.
3528 Currently this function knows 686a and 8231 chips.
3529 Format: [spp|ps2|epp|ecp|ecpepp]
3532 Halt all CPUs after the first oops has been printed for
3533 the specified number of seconds. This is to be used if
3534 your oopses keep scrolling off the screen.
3539 See header of drivers/block/paride/pcd.c.
3540 See also Documentation/admin-guide/blockdev/paride.rst.
3542 pci=option[,option...] [PCI] various PCI subsystem options.
3544 Some options herein operate on a specific device
3545 or a set of devices (<pci_dev>). These are
3546 specified in one of the following formats:
3548 [<domain>:]<bus>:<dev>.<func>[/<dev>.<func>]*
3549 pci:<vendor>:<device>[:<subvendor>:<subdevice>]
3551 Note: the first format specifies a PCI
3552 bus/device/function address which may change
3553 if new hardware is inserted, if motherboard
3554 firmware changes, or due to changes caused
3555 by other kernel parameters. If the
3556 domain is left unspecified, it is
3557 taken to be zero. Optionally, a path
3558 to a device through multiple device/function
3559 addresses can be specified after the base
3560 address (this is more robust against
3561 renumbering issues). The second format
3562 selects devices using IDs from the
3563 configuration space which may match multiple
3564 devices in the system.
3566 earlydump dump PCI config space before the kernel
3568 off [X86] don't probe for the PCI bus
3569 bios [X86-32] force use of PCI BIOS, don't access
3570 the hardware directly. Use this if your machine
3571 has a non-standard PCI host bridge.
3572 nobios [X86-32] disallow use of PCI BIOS, only direct
3573 hardware access methods are allowed. Use this
3574 if you experience crashes upon bootup and you
3575 suspect they are caused by the BIOS.
3576 conf1 [X86] Force use of PCI Configuration Access
3577 Mechanism 1 (config address in IO port 0xCF8,
3578 data in IO port 0xCFC, both 32-bit).
3579 conf2 [X86] Force use of PCI Configuration Access
3580 Mechanism 2 (IO port 0xCF8 is an 8-bit port for
3581 the function, IO port 0xCFA, also 8-bit, sets
3582 bus number. The config space is then accessed
3583 through ports 0xC000-0xCFFF).
3584 See http://wiki.osdev.org/PCI for more info
3585 on the configuration access mechanisms.
3586 noaer [PCIE] If the PCIEAER kernel config parameter is
3587 enabled, this kernel boot option can be used to
3588 disable the use of PCIE advanced error reporting.
3589 nodomains [PCI] Disable support for multiple PCI
3590 root domains (aka PCI segments, in ACPI-speak).
3591 nommconf [X86] Disable use of MMCONFIG for PCI
3593 check_enable_amd_mmconf [X86] check for and enable
3594 properly configured MMIO access to PCI
3595 config space on AMD family 10h CPU
3596 nomsi [MSI] If the PCI_MSI kernel config parameter is
3597 enabled, this kernel boot option can be used to
3598 disable the use of MSI interrupts system-wide.
3599 noioapicquirk [APIC] Disable all boot interrupt quirks.
3600 Safety option to keep boot IRQs enabled. This
3601 should never be necessary.
3602 ioapicreroute [APIC] Enable rerouting of boot IRQs to the
3603 primary IO-APIC for bridges that cannot disable
3604 boot IRQs. This fixes a source of spurious IRQs
3605 when the system masks IRQs.
3606 noioapicreroute [APIC] Disable workaround that uses the
3607 boot IRQ equivalent of an IRQ that connects to
3608 a chipset where boot IRQs cannot be disabled.
3609 The opposite of ioapicreroute.
3610 biosirq [X86-32] Use PCI BIOS calls to get the interrupt
3611 routing table. These calls are known to be buggy
3612 on several machines and they hang the machine
3613 when used, but on other computers it's the only
3614 way to get the interrupt routing table. Try
3615 this option if the kernel is unable to allocate
3616 IRQs or discover secondary PCI buses on your
3618 rom [X86] Assign address space to expansion ROMs.
3619 Use with caution as certain devices share
3620 address decoders between ROMs and other
3622 norom [X86] Do not assign address space to
3623 expansion ROMs that do not already have
3624 BIOS assigned address ranges.
3625 nobar [X86] Do not assign address space to the
3626 BARs that weren't assigned by the BIOS.
3627 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be
3628 assigned automatically to PCI devices. You can
3629 make the kernel exclude IRQs of your ISA cards
3631 pirqaddr=0xAAAAA [X86] Specify the physical address
3632 of the PIRQ table (normally generated
3633 by the BIOS) if it is outside the
3634 F0000h-100000h range.
3635 lastbus=N [X86] Scan all buses thru bus #N. Can be
3636 useful if the kernel is unable to find your
3637 secondary buses and you want to tell it
3638 explicitly which ones they are.
3639 assign-busses [X86] Always assign all PCI bus
3640 numbers ourselves, overriding
3641 whatever the firmware may have done.
3642 usepirqmask [X86] Honor the possible IRQ mask stored
3643 in the BIOS $PIR table. This is needed on
3644 some systems with broken BIOSes, notably
3645 some HP Pavilion N5400 and Omnibook XE3
3646 notebooks. This will have no effect if ACPI
3647 IRQ routing is enabled.
3648 noacpi [X86] Do not use ACPI for IRQ routing
3649 or for PCI scanning.
3650 use_crs [X86] Use PCI host bridge window information
3651 from ACPI. On BIOSes from 2008 or later, this
3652 is enabled by default. If you need to use this,
3653 please report a bug.
3654 nocrs [X86] Ignore PCI host bridge windows from ACPI.
3655 If you need to use this, please report a bug.
3656 routeirq Do IRQ routing for all PCI devices.
3657 This is normally done in pci_enable_device(),
3658 so this option is a temporary workaround
3659 for broken drivers that don't call it.
3660 skip_isa_align [X86] do not align io start addr, so can
3661 handle more pci cards
3662 noearly [X86] Don't do any early type 1 scanning.
3663 This might help on some broken boards which
3664 machine check when some devices' config space
3665 is read. But various workarounds are disabled
3666 and some IOMMU drivers will not work.
3667 bfsort Sort PCI devices into breadth-first order.
3668 This sorting is done to get a device
3669 order compatible with older (<= 2.4) kernels.
3670 nobfsort Don't sort PCI devices into breadth-first order.
3671 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size)
3672 tuning and use the BIOS-configured MPS defaults.
3673 pcie_bus_safe Set every device's MPS to the largest value
3674 supported by all devices below the root complex.
3675 pcie_bus_perf Set device MPS to the largest allowable MPS
3676 based on its parent bus. Also set MRRS (Max
3677 Read Request Size) to the largest supported
3678 value (no larger than the MPS that the device
3679 or bus can support) for best performance.
3680 pcie_bus_peer2peer Set every device's MPS to 128B, which
3681 every device is guaranteed to support. This
3682 configuration allows peer-to-peer DMA between
3683 any pair of devices, possibly at the cost of
3684 reduced performance. This also guarantees
3685 that hot-added devices will work.
3686 cbiosize=nn[KMG] The fixed amount of bus space which is
3687 reserved for the CardBus bridge's IO window.
3688 The default value is 256 bytes.
3689 cbmemsize=nn[KMG] The fixed amount of bus space which is
3690 reserved for the CardBus bridge's memory
3691 window. The default value is 64 megabytes.
3694 [<order of align>@]<pci_dev>[; ...]
3695 Specifies alignment and device to reassign
3696 aligned memory resources. How to
3697 specify the device is described above.
3698 If <order of align> is not specified,
3699 PAGE_SIZE is used as alignment.
3700 A PCI-PCI bridge can be specified if resource
3701 windows need to be expanded.
3702 To specify the alignment for several
3703 instances of a device, the PCI vendor,
3704 device, subvendor, and subdevice may be
3705 specified, e.g., 12@pci:8086:9c22:103c:198f
3706 for 4096-byte alignment.
3707 ecrc= Enable/disable PCIe ECRC (transaction layer
3708 end-to-end CRC checking).
3709 bios: Use BIOS/firmware settings. This is the
3713 hpiosize=nn[KMG] The fixed amount of bus space which is
3714 reserved for hotplug bridge's IO window.
3715 Default size is 256 bytes.
3716 hpmmiosize=nn[KMG] The fixed amount of bus space which is
3717 reserved for hotplug bridge's MMIO window.
3718 Default size is 2 megabytes.
3719 hpmmioprefsize=nn[KMG] The fixed amount of bus space which is
3720 reserved for hotplug bridge's MMIO_PREF window.
3721 Default size is 2 megabytes.
3722 hpmemsize=nn[KMG] The fixed amount of bus space which is
3723 reserved for hotplug bridge's MMIO and
3725 Default size is 2 megabytes.
3726 hpbussize=nn The minimum amount of additional bus numbers
3727 reserved for buses below a hotplug bridge.
3729 realloc= Enable/disable reallocating PCI bridge resources
3730 if allocations done by BIOS are too small to
3731 accommodate resources required by all child
3733 off: Turn realloc off
3735 realloc same as realloc=on
3736 noari do not use PCIe ARI.
3737 noats [PCIE, Intel-IOMMU, AMD-IOMMU]
3738 do not use PCIe ATS (and IOMMU device IOTLB).
3739 pcie_scan_all Scan all possible PCIe devices. Otherwise we
3740 only look for one device below a PCIe downstream
3742 big_root_window Try to add a big 64bit memory window to the PCIe
3743 root complex on AMD CPUs. Some GFX hardware
3744 can resize a BAR to allow access to all VRAM.
3745 Adding the window is slightly risky (it may
3746 conflict with unreported devices), so this
3748 disable_acs_redir=<pci_dev>[; ...]
3749 Specify one or more PCI devices (in the format
3750 specified above) separated by semicolons.
3751 Each device specified will have the PCI ACS
3752 redirect capabilities forced off which will
3753 allow P2P traffic between devices through
3754 bridges without forcing it upstream. Note:
3755 this removes isolation between devices and
3756 may put more devices in an IOMMU group.
3757 force_floating [S390] Force usage of floating interrupts.
3758 nomio [S390] Do not use MIO instructions.
3759 norid [S390] ignore the RID field and force use of
3760 one PCI domain per PCI function
3762 pcie_aspm= [PCIE] Forcibly enable or disable PCIe Active State Power
3765 force Enable ASPM even on devices that claim not to support it.
3766 WARNING: Forcing ASPM on may cause system lockups.
3768 pcie_ports= [PCIE] PCIe port services handling:
3769 native Use native PCIe services (PME, AER, DPC, PCIe hotplug)
3770 even if the platform doesn't give the OS permission to
3771 use them. This may cause conflicts if the platform
3772 also tries to use these services.
3773 dpc-native Use native PCIe service for DPC only. May
3774 cause conflicts if firmware uses AER or DPC.
3775 compat Disable native PCIe services (PME, AER, DPC, PCIe
3778 pcie_port_pm= [PCIE] PCIe port power management handling:
3779 off Disable power management of all PCIe ports
3780 force Forcibly enable power management of all PCIe ports
3782 pcie_pme= [PCIE,PM] Native PCIe PME signaling options:
3783 nomsi Do not use MSI for native PCIe PME signaling (this makes
3784 all PCIe root ports use INTx for all services).
3786 pcmv= [HW,PCMCIA] BadgePAD 4
3790 Keep all power-domains already enabled by bootloader on,
3791 even if no driver has claimed them. This is useful
3792 for debug and development, but should not be
3793 needed on a platform with proper driver support.
3796 See Documentation/admin-guide/blockdev/paride.rst.
3798 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at
3801 See arch/parisc/kernel/pdc_chassis.c
3803 percpu_alloc= Select which percpu first chunk allocator to use.
3804 Currently supported values are "embed" and "page".
3805 Archs may support subset or none of the selections.
3806 See comments in mm/percpu.c for details on each
3807 allocator. This parameter is primarily for debugging
3808 and performance comparison.
3811 See Documentation/admin-guide/blockdev/paride.rst.
3814 See Documentation/admin-guide/blockdev/paride.rst.
3816 pirq= [SMP,APIC] Manual mp-table setup
3817 See Documentation/x86/i386/IO-APIC.rst.
3819 plip= [PPT,NET] Parallel port network link
3820 Format: { parport<nr> | timid | 0 }
3821 See also Documentation/admin-guide/parport.rst.
3823 pmtmr= [X86] Manual setup of pmtmr I/O Port.
3824 Override pmtimer IOPort with a hex value.
3827 pm_debug_messages [SUSPEND,KNL]
3828 Enable suspend/resume debug messages during boot up.
3831 Enable PNP debug messages (depends on the
3832 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time
3833 via /sys/module/pnp/parameters/debug. We always show
3834 current resource usage; turning this on also shows
3835 possible settings and some assignment information.
3841 { on | off | curr | res | no-curr | no-res }
3844 [ISAPNP] Exclude IRQs for the autoconfiguration
3847 [ISAPNP] Exclude DMAs for the autoconfiguration
3849 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration
3850 Ranges are in pairs (I/O port base and size).
3853 [ISAPNP] Exclude memory regions for the
3855 Ranges are in pairs (memory base and size).
3857 ports= [IP_VS_FTP] IPVS ftp helper module
3859 Up to 8 (IP_VS_APP_MAX_PORTS) ports
3861 Format: <port>,<port>....
3863 powersave=off [PPC] This option disables power saving features.
3864 It specifically disables cpuidle and sets the
3865 platform machine description specific power_save
3866 function to NULL. On Idle the CPU just reduces
3869 ppc_strict_facility_enable
3870 [PPC] This option catches any kernel floating point,
3871 Altivec, VSX and SPE outside of regions specifically
3872 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()).
3873 There is some performance impact when enabling this.
3877 Disable Hardware Transactional Memory
3879 print-fatal-signals=
3880 [KNL] debug: print fatal signals
3882 If enabled, warn about various signal handling
3883 related application anomalies: too many signals,
3884 too many POSIX.1 timers, fatal signals causing a
3887 If you hit the warning due to signal overflow,
3888 you might want to try "ulimit -i unlimited".
3892 printk.always_kmsg_dump=
3893 Trigger kmsg_dump for cases other than kernel oops or
3895 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
3898 printk.devkmsg={on,off,ratelimit}
3899 Control writing to /dev/kmsg.
3900 on - unlimited logging to /dev/kmsg from userspace
3901 off - logging to /dev/kmsg disabled
3902 ratelimit - ratelimit the logging
3905 printk.time= Show timing data prefixed to each printk message line
3906 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
3908 processor.max_cstate= [HW,ACPI]
3909 Limit processor to maximum C-state
3910 max_cstate=9 overrides any DMI blacklist limit.
3912 processor.nocst [HW,ACPI]
3913 Ignore the _CST method to determine C-states,
3914 instead using the legacy FADT method
3916 profile= [KNL] Enable kernel profiling via /proc/profile
3917 Format: [<profiletype>,]<number>
3918 Param: <profiletype>: "schedule", "sleep", or "kvm"
3919 [defaults to kernel profiling]
3920 Param: "schedule" - profile schedule points.
3921 Param: "sleep" - profile D-state sleeping (millisecs).
3922 Requires CONFIG_SCHEDSTATS
3923 Param: "kvm" - profile VM exits.
3924 Param: <number> - step/bucket size as a power of 2 for
3925 statistical time based profiling.
3927 prompt_ramdisk= [RAM] List of RAM disks to prompt for floppy disk
3929 See Documentation/admin-guide/blockdev/ramdisk.rst.
3931 prot_virt= [S390] enable hosting protected virtual machines
3932 isolated from the hypervisor (if hardware supports
3936 psi= [KNL] Enable or disable pressure stall information
3940 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to
3941 probe for; one of (bare|imps|exps|lifebook|any).
3942 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports
3944 psmouse.resetafter= [HW,MOUSE]
3945 Try to reset the device after so many bad packets
3948 [HW,MOUSE] Set desired mouse resolution, in dpi.
3949 psmouse.smartscroll=
3950 [HW,MOUSE] Controls Logitech smartscroll autorepeat.
3951 0 = disabled, 1 = enabled (default).
3953 pstore.backend= Specify the name of the pstore backend to use
3956 See Documentation/admin-guide/blockdev/paride.rst.
3958 pti= [X86_64] Control Page Table Isolation of user and
3959 kernel address spaces. Disabling this feature
3960 removes hardening, but improves performance of
3961 system calls and interrupts.
3963 on - unconditionally enable
3964 off - unconditionally disable
3965 auto - kernel detects whether your CPU model is
3966 vulnerable to issues that PTI mitigates
3968 Not specifying this option is equivalent to pti=auto.
3971 Equivalent to pti=off
3974 [KNL] Number of legacy pty's. Overwrites compiled-in
3977 quiet [KNL] Disable most log messages
3982 See Documentation/admin-guide/md.rst.
3984 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes
3985 See Documentation/admin-guide/blockdev/ramdisk.rst.
3987 random.trust_cpu={on,off}
3988 [KNL] Enable or disable trusting the use of the
3989 CPU's random number generator (if available) to
3990 fully seed the kernel's CRNG. Default is controlled
3991 by CONFIG_RANDOM_TRUST_CPU.
3993 ras=option[,option,...] [KNL] RAS-specific options
3996 Disable the Correctable Errors Collector,
3997 see CONFIG_RAS_CEC help text.
4000 The argument is a cpu list, as described above,
4001 except that the string "all" can be used to
4002 specify every CPU on the system.
4004 In kernels built with CONFIG_RCU_NOCB_CPU=y, set
4005 the specified list of CPUs to be no-callback CPUs.
4006 Invocation of these CPUs' RCU callbacks will be
4007 offloaded to "rcuox/N" kthreads created for that
4008 purpose, where "x" is "p" for RCU-preempt, and
4009 "s" for RCU-sched, and "N" is the CPU number.
4010 This reduces OS jitter on the offloaded CPUs,
4011 which can be useful for HPC and real-time
4012 workloads. It can also improve energy efficiency
4013 for asymmetric multiprocessors.
4016 Rather than requiring that offloaded CPUs
4017 (specified by rcu_nocbs= above) explicitly
4018 awaken the corresponding "rcuoN" kthreads,
4019 make these kthreads poll for callbacks.
4020 This improves the real-time response for the
4021 offloaded CPUs by relieving them of the need to
4022 wake up the corresponding kthread, but degrades
4023 energy efficiency by requiring that the kthreads
4024 periodically wake up to do the polling.
4026 rcutree.blimit= [KNL]
4027 Set maximum number of finished RCU callbacks to
4028 process in one batch.
4030 rcutree.dump_tree= [KNL]
4031 Dump the structure of the rcu_node combining tree
4032 out at early boot. This is used for diagnostic
4033 purposes, to verify correct tree setup.
4035 rcutree.gp_cleanup_delay= [KNL]
4036 Set the number of jiffies to delay each step of
4037 RCU grace-period cleanup.
4039 rcutree.gp_init_delay= [KNL]
4040 Set the number of jiffies to delay each step of
4041 RCU grace-period initialization.
4043 rcutree.gp_preinit_delay= [KNL]
4044 Set the number of jiffies to delay each step of
4045 RCU grace-period pre-initialization, that is,
4046 the propagation of recent CPU-hotplug changes up
4047 the rcu_node combining tree.
4049 rcutree.use_softirq= [KNL]
4050 If set to zero, move all RCU_SOFTIRQ processing to
4051 per-CPU rcuc kthreads. Defaults to a non-zero
4052 value, meaning that RCU_SOFTIRQ is used by default.
4053 Specify rcutree.use_softirq=0 to use rcuc kthreads.
4055 rcutree.rcu_fanout_exact= [KNL]
4056 Disable autobalancing of the rcu_node combining
4057 tree. This is used by rcutorture, and might
4058 possibly be useful for architectures having high
4059 cache-to-cache transfer latencies.
4061 rcutree.rcu_fanout_leaf= [KNL]
4062 Change the number of CPUs assigned to each
4063 leaf rcu_node structure. Useful for very
4064 large systems, which will choose the value 64,
4065 and for NUMA systems with large remote-access
4066 latencies, which will choose a value aligned
4067 with the appropriate hardware boundaries.
4069 rcutree.rcu_min_cached_objs= [KNL]
4070 Minimum number of objects which are cached and
4071 maintained per one CPU. Object size is equal
4072 to PAGE_SIZE. The cache allows to reduce the
4073 pressure to page allocator, also it makes the
4074 whole algorithm to behave better in low memory
4077 rcutree.jiffies_till_first_fqs= [KNL]
4078 Set delay from grace-period initialization to
4079 first attempt to force quiescent states.
4080 Units are jiffies, minimum value is zero,
4081 and maximum value is HZ.
4083 rcutree.jiffies_till_next_fqs= [KNL]
4084 Set delay between subsequent attempts to force
4085 quiescent states. Units are jiffies, minimum
4086 value is one, and maximum value is HZ.
4088 rcutree.jiffies_till_sched_qs= [KNL]
4089 Set required age in jiffies for a
4090 given grace period before RCU starts
4091 soliciting quiescent-state help from
4092 rcu_note_context_switch() and cond_resched().
4093 If not specified, the kernel will calculate
4094 a value based on the most recent settings
4095 of rcutree.jiffies_till_first_fqs
4096 and rcutree.jiffies_till_next_fqs.
4097 This calculated value may be viewed in
4098 rcutree.jiffies_to_sched_qs. Any attempt to set
4099 rcutree.jiffies_to_sched_qs will be cheerfully
4102 rcutree.kthread_prio= [KNL,BOOT]
4103 Set the SCHED_FIFO priority of the RCU per-CPU
4104 kthreads (rcuc/N). This value is also used for
4105 the priority of the RCU boost threads (rcub/N)
4106 and for the RCU grace-period kthreads (rcu_bh,
4107 rcu_preempt, and rcu_sched). If RCU_BOOST is
4108 set, valid values are 1-99 and the default is 1
4109 (the least-favored priority). Otherwise, when
4110 RCU_BOOST is not set, valid values are 0-99 and
4111 the default is zero (non-realtime operation).
4113 rcutree.rcu_nocb_gp_stride= [KNL]
4114 Set the number of NOCB callback kthreads in
4115 each group, which defaults to the square root
4116 of the number of CPUs. Larger numbers reduce
4117 the wakeup overhead on the global grace-period
4118 kthread, but increases that same overhead on
4119 each group's NOCB grace-period kthread.
4121 rcutree.qhimark= [KNL]
4122 Set threshold of queued RCU callbacks beyond which
4123 batch limiting is disabled.
4125 rcutree.qlowmark= [KNL]
4126 Set threshold of queued RCU callbacks below which
4127 batch limiting is re-enabled.
4129 rcutree.qovld= [KNL]
4130 Set threshold of queued RCU callbacks beyond which
4131 RCU's force-quiescent-state scan will aggressively
4132 enlist help from cond_resched() and sched IPIs to
4133 help CPUs more quickly reach quiescent states.
4134 Set to less than zero to make this be set based
4135 on rcutree.qhimark at boot time and to zero to
4136 disable more aggressive help enlistment.
4138 rcutree.rcu_idle_gp_delay= [KNL]
4139 Set wakeup interval for idle CPUs that have
4140 RCU callbacks (RCU_FAST_NO_HZ=y).
4142 rcutree.rcu_idle_lazy_gp_delay= [KNL]
4143 Set wakeup interval for idle CPUs that have
4144 only "lazy" RCU callbacks (RCU_FAST_NO_HZ=y).
4145 Lazy RCU callbacks are those which RCU can
4146 prove do nothing more than free memory.
4148 rcutree.rcu_kick_kthreads= [KNL]
4149 Cause the grace-period kthread to get an extra
4150 wake_up() if it sleeps three times longer than
4151 it should at force-quiescent-state time.
4152 This wake_up() will be accompanied by a
4153 WARN_ONCE() splat and an ftrace_dump().
4155 rcutree.sysrq_rcu= [KNL]
4156 Commandeer a sysrq key to dump out Tree RCU's
4157 rcu_node tree with an eye towards determining
4158 why a new grace period has not yet started.
4160 rcuperf.gp_async= [KNL]
4161 Measure performance of asynchronous
4162 grace-period primitives such as call_rcu().
4164 rcuperf.gp_async_max= [KNL]
4165 Specify the maximum number of outstanding
4166 callbacks per writer thread. When a writer
4167 thread exceeds this limit, it invokes the
4168 corresponding flavor of rcu_barrier() to allow
4169 previously posted callbacks to drain.
4171 rcuperf.gp_exp= [KNL]
4172 Measure performance of expedited synchronous
4173 grace-period primitives.
4175 rcuperf.holdoff= [KNL]
4176 Set test-start holdoff period. The purpose of
4177 this parameter is to delay the start of the
4178 test until boot completes in order to avoid
4181 rcuperf.kfree_rcu_test= [KNL]
4182 Set to measure performance of kfree_rcu() flooding.
4184 rcuperf.kfree_nthreads= [KNL]
4185 The number of threads running loops of kfree_rcu().
4187 rcuperf.kfree_alloc_num= [KNL]
4188 Number of allocations and frees done in an iteration.
4190 rcuperf.kfree_loops= [KNL]
4191 Number of loops doing rcuperf.kfree_alloc_num number
4192 of allocations and frees.
4194 rcuperf.nreaders= [KNL]
4195 Set number of RCU readers. The value -1 selects
4196 N, where N is the number of CPUs. A value
4197 "n" less than -1 selects N-n+1, where N is again
4198 the number of CPUs. For example, -2 selects N
4199 (the number of CPUs), -3 selects N+1, and so on.
4200 A value of "n" less than or equal to -N selects
4203 rcuperf.nwriters= [KNL]
4204 Set number of RCU writers. The values operate
4205 the same as for rcuperf.nreaders.
4206 N, where N is the number of CPUs
4208 rcuperf.perf_type= [KNL]
4209 Specify the RCU implementation to test.
4211 rcuperf.shutdown= [KNL]
4212 Shut the system down after performance tests
4213 complete. This is useful for hands-off automated
4216 rcuperf.verbose= [KNL]
4217 Enable additional printk() statements.
4219 rcuperf.writer_holdoff= [KNL]
4220 Write-side holdoff between grace periods,
4221 in microseconds. The default of zero says
4224 rcutorture.fqs_duration= [KNL]
4225 Set duration of force_quiescent_state bursts
4228 rcutorture.fqs_holdoff= [KNL]
4229 Set holdoff time within force_quiescent_state bursts
4232 rcutorture.fqs_stutter= [KNL]
4233 Set wait time between force_quiescent_state bursts
4236 rcutorture.fwd_progress= [KNL]
4237 Enable RCU grace-period forward-progress testing
4238 for the types of RCU supporting this notion.
4240 rcutorture.fwd_progress_div= [KNL]
4241 Specify the fraction of a CPU-stall-warning
4242 period to do tight-loop forward-progress testing.
4244 rcutorture.fwd_progress_holdoff= [KNL]
4245 Number of seconds to wait between successive
4246 forward-progress tests.
4248 rcutorture.fwd_progress_need_resched= [KNL]
4249 Enclose cond_resched() calls within checks for
4250 need_resched() during tight-loop forward-progress
4253 rcutorture.gp_cond= [KNL]
4254 Use conditional/asynchronous update-side
4255 primitives, if available.
4257 rcutorture.gp_exp= [KNL]
4258 Use expedited update-side primitives, if available.
4260 rcutorture.gp_normal= [KNL]
4261 Use normal (non-expedited) asynchronous
4262 update-side primitives, if available.
4264 rcutorture.gp_sync= [KNL]
4265 Use normal (non-expedited) synchronous
4266 update-side primitives, if available. If all
4267 of rcutorture.gp_cond=, rcutorture.gp_exp=,
4268 rcutorture.gp_normal=, and rcutorture.gp_sync=
4269 are zero, rcutorture acts as if is interpreted
4270 they are all non-zero.
4272 rcutorture.n_barrier_cbs= [KNL]
4273 Set callbacks/threads for rcu_barrier() testing.
4275 rcutorture.nfakewriters= [KNL]
4276 Set number of concurrent RCU writers. These just
4277 stress RCU, they don't participate in the actual
4278 test, hence the "fake".
4280 rcutorture.nreaders= [KNL]
4281 Set number of RCU readers. The value -1 selects
4282 N-1, where N is the number of CPUs. A value
4283 "n" less than -1 selects N-n-2, where N is again
4284 the number of CPUs. For example, -2 selects N
4285 (the number of CPUs), -3 selects N+1, and so on.
4287 rcutorture.object_debug= [KNL]
4288 Enable debug-object double-call_rcu() testing.
4290 rcutorture.onoff_holdoff= [KNL]
4291 Set time (s) after boot for CPU-hotplug testing.
4293 rcutorture.onoff_interval= [KNL]
4294 Set time (jiffies) between CPU-hotplug operations,
4295 or zero to disable CPU-hotplug testing.
4297 rcutorture.read_exit= [KNL]
4298 Set the number of read-then-exit kthreads used
4299 to test the interaction of RCU updaters and
4300 task-exit processing.
4302 rcutorture.read_exit_burst= [KNL]
4303 The number of times in a given read-then-exit
4304 episode that a set of read-then-exit kthreads
4307 rcutorture.read_exit_delay= [KNL]
4308 The delay, in seconds, between successive
4309 read-then-exit testing episodes.
4311 rcutorture.shuffle_interval= [KNL]
4312 Set task-shuffle interval (s). Shuffling tasks
4313 allows some CPUs to go into dyntick-idle mode
4314 during the rcutorture test.
4316 rcutorture.shutdown_secs= [KNL]
4317 Set time (s) after boot system shutdown. This
4318 is useful for hands-off automated testing.
4320 rcutorture.stall_cpu= [KNL]
4321 Duration of CPU stall (s) to test RCU CPU stall
4322 warnings, zero to disable.
4324 rcutorture.stall_cpu_block= [KNL]
4325 Sleep while stalling if set. This will result
4326 in warnings from preemptible RCU in addition
4327 to any other stall-related activity.
4329 rcutorture.stall_cpu_holdoff= [KNL]
4330 Time to wait (s) after boot before inducing stall.
4332 rcutorture.stall_cpu_irqsoff= [KNL]
4333 Disable interrupts while stalling if set.
4335 rcutorture.stall_gp_kthread= [KNL]
4336 Duration (s) of forced sleep within RCU
4337 grace-period kthread to test RCU CPU stall
4338 warnings, zero to disable. If both stall_cpu
4339 and stall_gp_kthread are specified, the
4340 kthread is starved first, then the CPU.
4342 rcutorture.stat_interval= [KNL]
4343 Time (s) between statistics printk()s.
4345 rcutorture.stutter= [KNL]
4346 Time (s) to stutter testing, for example, specifying
4347 five seconds causes the test to run for five seconds,
4348 wait for five seconds, and so on. This tests RCU's
4349 ability to transition abruptly to and from idle.
4351 rcutorture.test_boost= [KNL]
4352 Test RCU priority boosting? 0=no, 1=maybe, 2=yes.
4353 "Maybe" means test if the RCU implementation
4354 under test support RCU priority boosting.
4356 rcutorture.test_boost_duration= [KNL]
4357 Duration (s) of each individual boost test.
4359 rcutorture.test_boost_interval= [KNL]
4360 Interval (s) between each boost test.
4362 rcutorture.test_no_idle_hz= [KNL]
4363 Test RCU's dyntick-idle handling. See also the
4364 rcutorture.shuffle_interval parameter.
4366 rcutorture.torture_type= [KNL]
4367 Specify the RCU implementation to test.
4369 rcutorture.verbose= [KNL]
4370 Enable additional printk() statements.
4372 rcupdate.rcu_cpu_stall_ftrace_dump= [KNL]
4373 Dump ftrace buffer after reporting RCU CPU
4376 rcupdate.rcu_cpu_stall_suppress= [KNL]
4377 Suppress RCU CPU stall warning messages.
4379 rcupdate.rcu_cpu_stall_suppress_at_boot= [KNL]
4380 Suppress RCU CPU stall warning messages and
4381 rcutorture writer stall warnings that occur
4382 during early boot, that is, during the time
4383 before the init task is spawned.
4385 rcupdate.rcu_cpu_stall_timeout= [KNL]
4386 Set timeout for RCU CPU stall warning messages.
4388 rcupdate.rcu_expedited= [KNL]
4389 Use expedited grace-period primitives, for
4390 example, synchronize_rcu_expedited() instead
4391 of synchronize_rcu(). This reduces latency,
4392 but can increase CPU utilization, degrade
4393 real-time latency, and degrade energy efficiency.
4394 No effect on CONFIG_TINY_RCU kernels.
4396 rcupdate.rcu_normal= [KNL]
4397 Use only normal grace-period primitives,
4398 for example, synchronize_rcu() instead of
4399 synchronize_rcu_expedited(). This improves
4400 real-time latency, CPU utilization, and
4401 energy efficiency, but can expose users to
4402 increased grace-period latency. This parameter
4403 overrides rcupdate.rcu_expedited. No effect on
4404 CONFIG_TINY_RCU kernels.
4406 rcupdate.rcu_normal_after_boot= [KNL]
4407 Once boot has completed (that is, after
4408 rcu_end_inkernel_boot() has been invoked), use
4409 only normal grace-period primitives. No effect
4410 on CONFIG_TINY_RCU kernels.
4412 rcupdate.rcu_task_ipi_delay= [KNL]
4413 Set time in jiffies during which RCU tasks will
4414 avoid sending IPIs, starting with the beginning
4415 of a given grace period. Setting a large
4416 number avoids disturbing real-time workloads,
4417 but lengthens grace periods.
4419 rcupdate.rcu_task_stall_timeout= [KNL]
4420 Set timeout in jiffies for RCU task stall warning
4421 messages. Disable with a value less than or equal
4424 rcupdate.rcu_self_test= [KNL]
4425 Run the RCU early boot self tests
4429 Run specified binary instead of /init from the ramdisk,
4430 used for early userspace startup. See initrd.
4433 force - Override the decision by the kernel to hide the
4434 advertisement of RDRAND support (this affects
4435 certain AMD processors because of buggy BIOS
4436 support, specifically around the suspend/resume
4440 Turn on/off individual RDT features. List is:
4441 cmt, mbmtotal, mbmlocal, l3cat, l3cdp, l2cat, l2cdp,
4443 E.g. to turn on cmt and turn off mba use:
4447 Format (x86 or x86_64):
4448 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] \
4450 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \
4452 Where reboot_mode is one of warm (soft) or cold (hard) or gpio
4453 (prefix with 'panic_' to set mode for panic
4455 reboot_type is one of bios, acpi, kbd, triple, efi, or pci,
4456 reboot_force is either force or not specified,
4457 reboot_cpu is s[mp]#### with #### being the processor
4458 to be used for rebooting.
4460 refscale.holdoff= [KNL]
4461 Set test-start holdoff period. The purpose of
4462 this parameter is to delay the start of the
4463 test until boot completes in order to avoid
4466 refscale.loops= [KNL]
4467 Set the number of loops over the synchronization
4468 primitive under test. Increasing this number
4469 reduces noise due to loop start/end overhead,
4470 but the default has already reduced the per-pass
4471 noise to a handful of picoseconds on ca. 2020
4474 refscale.nreaders= [KNL]
4475 Set number of readers. The default value of -1
4476 selects N, where N is roughly 75% of the number
4477 of CPUs. A value of zero is an interesting choice.
4479 refscale.nruns= [KNL]
4480 Set number of runs, each of which is dumped onto
4483 refscale.readdelay= [KNL]
4484 Set the read-side critical-section duration,
4485 measured in microseconds.
4487 refscale.scale_type= [KNL]
4488 Specify the read-protection implementation to test.
4490 refscale.shutdown= [KNL]
4491 Shut down the system at the end of the performance
4492 test. This defaults to 1 (shut it down) when
4493 rcuperf is built into the kernel and to 0 (leave
4494 it running) when rcuperf is built as a module.
4496 refscale.verbose= [KNL]
4497 Enable additional printk() statements.
4500 [KNL, SMP] Set scheduler's default relax_domain_level.
4501 See Documentation/admin-guide/cgroup-v1/cpusets.rst.
4503 reserve= [KNL,BUGS] Force kernel to ignore I/O ports or memory
4504 Format: <base1>,<size1>[,<base2>,<size2>,...]
4505 Reserve I/O ports or memory so the kernel won't use
4506 them. If <base> is less than 0x10000, the region
4507 is assumed to be I/O ports; otherwise it is memory.
4509 reservetop= [X86-32]
4511 Reserves a hole at the top of the kernel virtual
4516 Set the amount of memory to reserve for BIOS at
4517 the bottom of the address space.
4519 reset_devices [KNL] Force drivers to reset the underlying device
4520 during initialization.
4523 Specify the partition device for software suspend
4525 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>}
4527 resume_offset= [SWSUSP]
4528 Specify the offset from the beginning of the partition
4529 given by "resume=" at which the swap header is located,
4530 in <PAGE_SIZE> units (needed only for swap files).
4531 See Documentation/power/swsusp-and-swap-files.rst
4533 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to
4534 read the resume files
4536 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up.
4537 Useful for devices that are detected asynchronously
4538 (e.g. USB and MMC devices).
4540 hibernate= [HIBERNATION]
4541 noresume Don't check if there's a hibernation image
4542 present during boot.
4543 nocompress Don't compress/decompress hibernation images.
4544 no Disable hibernation and resume.
4545 protect_image Turn on image protection during restoration
4546 (that will set all pages holding image data
4547 during restoration read-only).
4549 retain_initrd [RAM] Keep initrd memory after extraction
4551 rfkill.default_state=
4552 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm,
4553 etc. communication is blocked by default.
4556 rfkill.master_switch_mode=
4557 0 The "airplane mode" button does nothing.
4558 1 The "airplane mode" button toggles between everything
4559 blocked and the previous configuration.
4560 2 The "airplane mode" button toggles between everything
4561 blocked and everything unblocked.
4563 rhash_entries= [KNL,NET]
4564 Set number of hash buckets for route cache
4567 [KNL] Disable ring 3 MONITOR/MWAIT feature on supported
4570 ro [KNL] Mount root device read-only on boot
4573 on Mark read-only kernel memory as read-only (default).
4574 off Leave read-only kernel memory writable for debugging.
4577 Enable the uart passthrough on the designated usb port
4578 on Rockchip SoCs. When active, the signals of the
4579 debug-uart get routed to the D+ and D- pins of the usb
4580 port and the regular usb controller gets disabled.
4582 root= [KNL] Root filesystem
4583 See name_to_dev_t comment in init/do_mounts.c.
4585 rootdelay= [KNL] Delay (in seconds) to pause before attempting to
4586 mount the root filesystem
4588 rootflags= [KNL] Set root filesystem mount option string
4590 rootfstype= [KNL] Set root filesystem type
4592 rootwait [KNL] Wait (indefinitely) for root device to show up.
4593 Useful for devices that are detected asynchronously
4594 (e.g. USB and MMC devices).
4596 rproc_mem=nn[KMG][@address]
4597 [KNL,ARM,CMA] Remoteproc physical memory block.
4598 Memory area to be used by remote processor image,
4601 rw [KNL] Mount root device read-write on boot
4603 S [KNL] Run init in single mode
4605 s390_iommu= [HW,S390]
4606 Set s390 IOTLB flushing mode
4608 With strict flushing every unmap operation will result in
4609 an IOTLB flush. Default is lazy flushing before reuse,
4613 See drivers/net/irda/sa1100_ir.c.
4615 sbni= [NET] Granch SBNI12 leased line adapter
4617 sched_debug [KNL] Enables verbose scheduler debug messages.
4619 schedstats= [KNL,X86] Enable or disable scheduled statistics.
4620 Allowed values are enable and disable. This feature
4621 incurs a small amount of overhead in the scheduler
4622 but is useful for debugging and performance tuning.
4624 sched_thermal_decay_shift=
4625 [KNL, SMP] Set a decay shift for scheduler thermal
4626 pressure signal. Thermal pressure signal follows the
4627 default decay period of other scheduler pelt
4628 signals(usually 32 ms but configurable). Setting
4629 sched_thermal_decay_shift will left shift the decay
4630 period for the thermal pressure signal by the shift
4632 i.e. with the default pelt decay period of 32 ms
4633 sched_thermal_decay_shift thermal pressure decay pr
4637 Format: integer between 0 and 10
4640 skew_tick= [KNL] Offset the periodic timer tick per cpu to mitigate
4641 xtime_lock contention on larger systems, and/or RCU lock
4642 contention on all systems with CONFIG_MAXSMP set.
4643 Format: { "0" | "1" }
4644 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1"
4646 Note: increases power consumption, thus should only be
4647 enabled if running jitter sensitive (HPC/RT) workloads.
4649 security= [SECURITY] Choose a legacy "major" security module to
4650 enable at boot. This has been deprecated by the
4653 selinux= [SELINUX] Disable or enable SELinux at boot time.
4654 Format: { "0" | "1" }
4655 See security/selinux/Kconfig help text.
4660 apparmor= [APPARMOR] Disable or enable AppArmor at boot time
4661 Format: { "0" | "1" }
4662 See security/apparmor/Kconfig help text
4665 Default value is set via kernel config option.
4667 serialnumber [BUGS=X86-32]
4670 Maximal number of shapers.
4678 Disable merging of slabs with similar size. May be
4679 necessary if there is some reason to distinguish
4680 allocs to different slabs, especially in hardened
4681 environments where the risk of heap overflows and
4682 layout control by attackers can usually be
4683 frustrated by disabling merging. This will reduce
4684 most of the exposure of a heap attack to a single
4685 cache (risks via metadata attacks are mostly
4686 unchanged). Debug options disable merging on their
4688 For more information see Documentation/vm/slub.rst.
4690 slab_max_order= [MM, SLAB]
4691 Determines the maximum allowed order for slabs.
4692 A high setting may cause OOMs due to memory
4693 fragmentation. Defaults to 1 for systems with
4694 more than 32MB of RAM, 0 otherwise.
4696 slub_debug[=options[,slabs][;[options[,slabs]]...] [MM, SLUB]
4697 Enabling slub_debug allows one to determine the
4698 culprit if slab objects become corrupted. Enabling
4699 slub_debug can create guard zones around objects and
4700 may poison objects when not in use. Also tracks the
4701 last alloc / free. For more information see
4702 Documentation/vm/slub.rst.
4704 slub_memcg_sysfs= [MM, SLUB]
4705 Determines whether to enable sysfs directories for
4706 memory cgroup sub-caches. 1 to enable, 0 to disable.
4707 The default is determined by CONFIG_SLUB_MEMCG_SYSFS_ON.
4708 Enabling this can lead to a very high number of debug
4709 directories and files being created under
4712 slub_max_order= [MM, SLUB]
4713 Determines the maximum allowed order for slabs.
4714 A high setting may cause OOMs due to memory
4715 fragmentation. For more information see
4716 Documentation/vm/slub.rst.
4718 slub_min_objects= [MM, SLUB]
4719 The minimum number of objects per slab. SLUB will
4720 increase the slab order up to slub_max_order to
4721 generate a sufficiently large slab able to contain
4722 the number of objects indicated. The higher the number
4723 of objects the smaller the overhead of tracking slabs
4724 and the less frequently locks need to be acquired.
4725 For more information see Documentation/vm/slub.rst.
4727 slub_min_order= [MM, SLUB]
4728 Determines the minimum page order for slabs. Must be
4729 lower than slub_max_order.
4730 For more information see Documentation/vm/slub.rst.
4732 slub_nomerge [MM, SLUB]
4733 Same with slab_nomerge. This is supported for legacy.
4734 See slab_nomerge for more information.
4737 Format: <io1>[,<io2>[,...,<io8>]]
4739 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices
4740 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port
4741 smsc-ircc2.ircc_sir= [HW] SIR base I/O port
4742 smsc-ircc2.ircc_fir= [HW] FIR base I/O port
4743 smsc-ircc2.ircc_irq= [HW] IRQ line
4744 smsc-ircc2.ircc_dma= [HW] DMA channel
4745 smsc-ircc2.ircc_transceiver= [HW] Transceiver type:
4746 0: Toshiba Satellite 1800 (GP data pin select)
4747 1: Fast pin select (default)
4750 smt [KNL,S390] Set the maximum number of threads (logical
4751 CPUs) to use per physical CPU on systems capable of
4752 symmetric multithreading (SMT). Will be capped to the
4753 actual hardware limit.
4755 Default: -1 (no limit)
4758 [KNL] Should the soft-lockup detector generate panics.
4761 A value of 1 instructs the soft-lockup detector
4762 to panic the machine when a soft-lockup occurs. It is
4763 also controlled by the kernel.softlockup_panic sysctl
4764 and CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC, which is the
4765 respective build-time switch to that functionality.
4767 softlockup_all_cpu_backtrace=
4768 [KNL] Should the soft-lockup detector generate
4769 backtraces on all cpus.
4772 sonypi.*= [HW] Sony Programmable I/O Control Device driver
4773 See Documentation/admin-guide/laptops/sonypi.rst
4775 spectre_v2= [X86] Control mitigation of Spectre variant 2
4776 (indirect branch speculation) vulnerability.
4777 The default operation protects the kernel from
4780 on - unconditionally enable, implies
4782 off - unconditionally disable, implies
4784 auto - kernel detects whether your CPU model is
4787 Selecting 'on' will, and 'auto' may, choose a
4788 mitigation method at run time according to the
4789 CPU, the available microcode, the setting of the
4790 CONFIG_RETPOLINE configuration option, and the
4791 compiler with which the kernel was built.
4793 Selecting 'on' will also enable the mitigation
4794 against user space to user space task attacks.
4796 Selecting 'off' will disable both the kernel and
4797 the user space protections.
4799 Specific mitigations can also be selected manually:
4801 retpoline - replace indirect branches
4802 retpoline,generic - google's original retpoline
4803 retpoline,amd - AMD-specific minimal thunk
4805 Not specifying this option is equivalent to
4809 [X86] Control mitigation of Spectre variant 2
4810 (indirect branch speculation) vulnerability between
4813 on - Unconditionally enable mitigations. Is
4814 enforced by spectre_v2=on
4816 off - Unconditionally disable mitigations. Is
4817 enforced by spectre_v2=off
4819 prctl - Indirect branch speculation is enabled,
4820 but mitigation can be enabled via prctl
4821 per thread. The mitigation control state
4822 is inherited on fork.
4825 - Like "prctl" above, but only STIBP is
4826 controlled per thread. IBPB is issued
4827 always when switching between different user
4831 - Same as "prctl" above, but all seccomp
4832 threads will enable the mitigation unless
4833 they explicitly opt out.
4836 - Like "seccomp" above, but only STIBP is
4837 controlled per thread. IBPB is issued
4838 always when switching between different
4839 user space processes.
4841 auto - Kernel selects the mitigation depending on
4842 the available CPU features and vulnerability.
4845 If CONFIG_SECCOMP=y then "seccomp", otherwise "prctl"
4847 Not specifying this option is equivalent to
4848 spectre_v2_user=auto.
4850 spec_store_bypass_disable=
4851 [HW] Control Speculative Store Bypass (SSB) Disable mitigation
4852 (Speculative Store Bypass vulnerability)
4854 Certain CPUs are vulnerable to an exploit against a
4855 a common industry wide performance optimization known
4856 as "Speculative Store Bypass" in which recent stores
4857 to the same memory location may not be observed by
4858 later loads during speculative execution. The idea
4859 is that such stores are unlikely and that they can
4860 be detected prior to instruction retirement at the
4861 end of a particular speculation execution window.
4863 In vulnerable processors, the speculatively forwarded
4864 store can be used in a cache side channel attack, for
4865 example to read memory to which the attacker does not
4866 directly have access (e.g. inside sandboxed code).
4868 This parameter controls whether the Speculative Store
4869 Bypass optimization is used.
4871 On x86 the options are:
4873 on - Unconditionally disable Speculative Store Bypass
4874 off - Unconditionally enable Speculative Store Bypass
4875 auto - Kernel detects whether the CPU model contains an
4876 implementation of Speculative Store Bypass and
4877 picks the most appropriate mitigation. If the
4878 CPU is not vulnerable, "off" is selected. If the
4879 CPU is vulnerable the default mitigation is
4880 architecture and Kconfig dependent. See below.
4881 prctl - Control Speculative Store Bypass per thread
4882 via prctl. Speculative Store Bypass is enabled
4883 for a process by default. The state of the control
4884 is inherited on fork.
4885 seccomp - Same as "prctl" above, but all seccomp threads
4886 will disable SSB unless they explicitly opt out.
4888 Default mitigations:
4889 X86: If CONFIG_SECCOMP=y "seccomp", otherwise "prctl"
4891 On powerpc the options are:
4893 on,auto - On Power8 and Power9 insert a store-forwarding
4894 barrier on kernel entry and exit. On Power7
4895 perform a software flush on kernel entry and
4899 Not specifying this option is equivalent to
4900 spec_store_bypass_disable=auto.
4902 spia_io_base= [HW,MTD]
4908 [X86] Enable split lock detection
4910 When enabled (and if hardware support is present), atomic
4911 instructions that access data across cache line
4912 boundaries will result in an alignment check exception.
4916 warn - the kernel will emit rate limited warnings
4917 about applications triggering the #AC
4918 exception. This mode is the default on CPUs
4919 that supports split lock detection.
4921 fatal - the kernel will send SIGBUS to applications
4922 that trigger the #AC exception.
4924 If an #AC exception is hit in the kernel or in
4925 firmware (i.e. not while executing in user mode)
4926 the kernel will oops in either "warn" or "fatal"
4930 Control the Special Register Buffer Data Sampling
4933 Certain CPUs are vulnerable to an MDS-like
4934 exploit which can leak bits from the random
4937 By default, this issue is mitigated by
4938 microcode. However, the microcode fix can cause
4939 the RDRAND and RDSEED instructions to become
4940 much slower. Among other effects, this will
4941 result in reduced throughput from /dev/urandom.
4943 The microcode mitigation can be disabled with
4944 the following option:
4946 off: Disable mitigation and remove
4947 performance impact to RDRAND and RDSEED
4949 srcutree.counter_wrap_check [KNL]
4950 Specifies how frequently to check for
4951 grace-period sequence counter wrap for the
4952 srcu_data structure's ->srcu_gp_seq_needed field.
4953 The greater the number of bits set in this kernel
4954 parameter, the less frequently counter wrap will
4955 be checked for. Note that the bottom two bits
4958 srcutree.exp_holdoff [KNL]
4959 Specifies how many nanoseconds must elapse
4960 since the end of the last SRCU grace period for
4961 a given srcu_struct until the next normal SRCU
4962 grace period will be considered for automatic
4963 expediting. Set to zero to disable automatic
4967 Speculative Store Bypass Disable control
4969 On CPUs that are vulnerable to the Speculative
4970 Store Bypass vulnerability and offer a
4971 firmware based mitigation, this parameter
4972 indicates how the mitigation should be used:
4974 force-on: Unconditionally enable mitigation for
4975 for both kernel and userspace
4976 force-off: Unconditionally disable mitigation for
4977 for both kernel and userspace
4978 kernel: Always enable mitigation in the
4979 kernel, and offer a prctl interface
4980 to allow userspace to register its
4981 interest in being mitigated too.
4983 stack_guard_gap= [MM]
4984 override the default stack gap protection. The value
4985 is in page units and it defines how many pages prior
4986 to (for stacks growing down) resp. after (for stacks
4987 growing up) the main stack are reserved for no other
4988 mapping. Default value is 256 pages.
4991 Enabled the stack tracer on boot up.
4993 stacktrace_filter=[function-list]
4994 [FTRACE] Limit the functions that the stack tracer
4995 will trace at boot up. function-list is a comma separated
4996 list of functions. This list can be changed at run
4997 time by the stack_trace_filter file in the debugfs
4998 tracing directory. Note, this enables stack tracing
4999 and the stacktrace above is not needed.
5003 Set the STI (builtin display/keyboard on the HP-PARISC
5004 machines) console (graphic card) which should be used
5005 as the initial boot-console.
5006 See also comment in drivers/video/console/sticore.c.
5009 See comment in drivers/video/console/sticore.c.
5012 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]]
5014 sunrpc.min_resvport=
5015 sunrpc.max_resvport=
5017 SunRPC servers often require that client requests
5018 originate from a privileged port (i.e. a port in the
5019 range 0 < portnr < 1024).
5020 An administrator who wishes to reserve some of these
5021 ports for other uses may adjust the range that the
5022 kernel's sunrpc client considers to be privileged
5023 using these two parameters to set the minimum and
5024 maximum port values.
5026 sunrpc.svc_rpc_per_connection_limit=
5028 Limit the number of requests that the server will
5029 process in parallel from a single connection.
5030 The default value is 0 (no limit).
5034 Control how the NFS server code allocates CPUs to
5035 service thread pools. Depending on how many NICs
5036 you have and where their interrupts are bound, this
5037 option will affect which CPUs will do NFS serving.
5038 Note: this parameter cannot be changed while the
5039 NFS server is running.
5041 auto the server chooses an appropriate mode
5042 automatically using heuristics
5043 global a single global pool contains all CPUs
5044 percpu one pool for each CPU
5045 pernode one pool for each NUMA node (equivalent
5046 to global on non-NUMA machines)
5048 sunrpc.tcp_slot_table_entries=
5049 sunrpc.udp_slot_table_entries=
5051 Sets the upper limit on the number of simultaneous
5052 RPC calls that can be sent from the client to a
5053 server. Increasing these values may allow you to
5054 improve throughput, but will also increase the
5055 amount of memory reserved for use by the client.
5057 suspend.pm_test_delay=
5059 Sets the number of seconds to remain in a suspend test
5060 mode before resuming the system (see
5061 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG
5062 is set. Default value is 5.
5065 Format: { on | off | y | n | 1 | 0 }
5066 This parameter controls use of the Protected
5067 Execution Facility on pSeries.
5070 [KNL] Enable accounting of swap in memory resource
5071 controller if no parameter or 1 is given or disable
5072 it if 0 is given (See Documentation/admin-guide/cgroup-v1/memory.rst)
5074 swiotlb= [ARM,IA-64,PPC,MIPS,X86]
5075 Format: { <int> | force | noforce }
5076 <int> -- Number of I/O TLB slabs
5077 force -- force using of bounce buffers even if they
5078 wouldn't be automatically used by the kernel
5079 noforce -- Never use bounce buffers (for debugging)
5084 Set a sysctl parameter, right before loading the init
5085 process, as if the value was written to the respective
5086 /proc/sys/... file. Both '.' and '/' are recognized as
5087 separators. Unrecognized parameters and invalid values
5088 are reported in the kernel log. Sysctls registered
5089 later by a loaded module cannot be set this way.
5090 Example: sysctl.vm.swappiness=40
5092 sysfs.deprecated=0|1 [KNL]
5093 Enable/disable old style sysfs layout for old udev
5094 on older distributions. When this option is enabled
5095 very new udev will not work anymore. When this option
5096 is disabled (or CONFIG_SYSFS_DEPRECATED not compiled)
5097 in older udev will not work anymore.
5098 Default depends on CONFIG_SYSFS_DEPRECATED_V2 set in
5099 the kernel configuration.
5101 sysrq_always_enabled
5103 Ignore sysrq setting - this boot parameter will
5104 neutralize any effect of /proc/sys/kernel/sysrq.
5105 Useful for debugging.
5107 tcpmhash_entries= [KNL,NET]
5108 Set the number of tcp_metrics_hash slots.
5109 Default value is 8192 or 16384 depending on total
5110 ram pages. This is used to specify the TCP metrics
5111 cache size. See Documentation/networking/ip-sysctl.rst
5112 "tcp_no_metrics_save" section for more details.
5116 test_suspend= [SUSPEND][,N]
5117 Specify "mem" (for Suspend-to-RAM) or "standby" (for
5118 standby suspend) or "freeze" (for suspend type freeze)
5119 as the system sleep state during system startup with
5120 the optional capability to repeat N number of times.
5121 The system is woken from this state using a
5122 wakeup-capable RTC alarm.
5124 thash_entries= [KNL,NET]
5125 Set number of hash buckets for TCP connection
5127 thermal.act= [HW,ACPI]
5128 -1: disable all active trip points in all thermal zones
5129 <degrees C>: override all lowest active trip points
5131 thermal.crt= [HW,ACPI]
5132 -1: disable all critical trip points in all thermal zones
5133 <degrees C>: override all critical trip points
5135 thermal.nocrt= [HW,ACPI]
5136 Set to disable actions on ACPI thermal zone
5137 critical and hot trip points.
5139 thermal.off= [HW,ACPI]
5140 1: disable ACPI thermal control
5142 thermal.psv= [HW,ACPI]
5143 -1: disable all passive trip points
5144 <degrees C>: override all passive trip points to this
5147 thermal.tzp= [HW,ACPI]
5148 Specify global default ACPI thermal zone polling rate
5149 <deci-seconds>: poll all this frequency
5150 0: no polling (default)
5153 Force threading of all interrupt handlers except those
5154 marked explicitly IRQF_NO_THREAD.
5158 Specify if the kernel should make use of the cpu
5159 topology information if the hardware supports this.
5160 The scheduler will make use of this information and
5161 e.g. base its process migration decisions on it.
5164 topology_updates= [KNL, PPC, NUMA]
5166 Specify if the kernel should ignore (off)
5167 topology updates sent by the hypervisor to this
5170 torture.disable_onoff_at_boot= [KNL]
5171 Prevent the CPU-hotplug component of torturing
5172 until after init has spawned.
5174 torture.ftrace_dump_at_shutdown= [KNL]
5175 Dump the ftrace buffer at torture-test shutdown,
5176 even if there were no errors. This can be a
5177 very costly operation when many torture tests
5178 are running concurrently, especially on systems
5179 with rotating-rust storage.
5183 tpm_suspend_pcr=[HW,TPM]
5184 Format: integer pcr id
5185 Specify that at suspend time, the tpm driver
5186 should extend the specified pcr with zeros,
5187 as a workaround for some chips which fail to
5188 flush the last written pcr on TPM_SaveState.
5189 This will guarantee that all the other pcrs
5192 trace_buf_size=nn[KMG]
5193 [FTRACE] will set tracing buffer size on each cpu.
5195 trace_event=[event-list]
5196 [FTRACE] Set and start specified trace events in order
5197 to facilitate early boot debugging. The event-list is a
5198 comma separated list of trace events to enable. See
5199 also Documentation/trace/events.rst
5201 trace_options=[option-list]
5202 [FTRACE] Enable or disable tracer options at boot.
5203 The option-list is a comma delimited list of options
5204 that can be enabled or disabled just as if you were
5205 to echo the option name into
5207 /sys/kernel/debug/tracing/trace_options
5209 For example, to enable stacktrace option (to dump the
5210 stack trace of each event), add to the command line:
5212 trace_options=stacktrace
5214 See also Documentation/trace/ftrace.rst "trace options"
5218 Have the tracepoints sent to printk as well as the
5219 tracing ring buffer. This is useful for early boot up
5220 where the system hangs or reboots and does not give the
5221 option for reading the tracing buffer or performing a
5222 ftrace_dump_on_oops.
5224 To turn off having tracepoints sent to printk,
5225 echo 0 > /proc/sys/kernel/tracepoint_printk
5226 Note, echoing 1 into this file without the
5227 tracepoint_printk kernel cmdline option has no effect.
5231 Having tracepoints sent to printk() and activating high
5232 frequency tracepoints such as irq or sched, can cause
5233 the system to live lock.
5236 [FTRACE] enable this option to disable tracing when a
5237 warning is hit. This turns off "tracing_on". Tracing can
5238 be enabled again by echoing '1' into the "tracing_on"
5239 file located in /sys/kernel/debug/tracing/
5241 This option is useful, as it disables the trace before
5242 the WARNING dump is called, which prevents the trace to
5243 be filled with content caused by the warning output.
5245 This option can also be set at run time via the sysctl
5246 option: kernel/traceoff_on_warning
5248 transparent_hugepage=
5250 Format: [always|madvise|never]
5251 Can be used to control the default behavior of the system
5252 with respect to transparent hugepages.
5253 See Documentation/admin-guide/mm/transhuge.rst
5256 tsc= Disable clocksource stability checks for TSC.
5258 [x86] reliable: mark tsc clocksource as reliable, this
5259 disables clocksource verification at runtime, as well
5260 as the stability checks done at bootup. Used to enable
5261 high-resolution timer mode on older hardware, and in
5262 virtualized environment.
5263 [x86] noirqtime: Do not use TSC to do irq accounting.
5264 Used to run time disable IRQ_TIME_ACCOUNTING on any
5265 platforms where RDTSC is slow and this accounting
5267 [x86] unstable: mark the TSC clocksource as unstable, this
5268 marks the TSC unconditionally unstable at bootup and
5269 avoids any further wobbles once the TSC watchdog notices.
5270 [x86] nowatchdog: disable clocksource watchdog. Used
5271 in situations with strict latency requirements (where
5272 interruptions from clocksource watchdog are not
5275 tsc_early_khz= [X86] Skip early TSC calibration and use the given
5276 value instead. Useful when the early TSC frequency discovery
5277 procedure is not reliable, such as on overclocked systems
5278 with CPUID.16h support and partial CPUID.15h support.
5279 Format: <unsigned int>
5281 tsx= [X86] Control Transactional Synchronization
5282 Extensions (TSX) feature in Intel processors that
5283 support TSX control.
5285 This parameter controls the TSX feature. The options are:
5287 on - Enable TSX on the system. Although there are
5288 mitigations for all known security vulnerabilities,
5289 TSX has been known to be an accelerator for
5290 several previous speculation-related CVEs, and
5291 so there may be unknown security risks associated
5292 with leaving it enabled.
5294 off - Disable TSX on the system. (Note that this
5295 option takes effect only on newer CPUs which are
5296 not vulnerable to MDS, i.e., have
5297 MSR_IA32_ARCH_CAPABILITIES.MDS_NO=1 and which get
5298 the new IA32_TSX_CTRL MSR through a microcode
5299 update. This new MSR allows for the reliable
5300 deactivation of the TSX functionality.)
5302 auto - Disable TSX if X86_BUG_TAA is present,
5303 otherwise enable TSX on the system.
5305 Not specifying this option is equivalent to tsx=off.
5307 See Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
5310 tsx_async_abort= [X86,INTEL] Control mitigation for the TSX Async
5311 Abort (TAA) vulnerability.
5313 Similar to Micro-architectural Data Sampling (MDS)
5314 certain CPUs that support Transactional
5315 Synchronization Extensions (TSX) are vulnerable to an
5316 exploit against CPU internal buffers which can forward
5317 information to a disclosure gadget under certain
5320 In vulnerable processors, the speculatively forwarded
5321 data can be used in a cache side channel attack, to
5322 access data to which the attacker does not have direct
5325 This parameter controls the TAA mitigation. The
5328 full - Enable TAA mitigation on vulnerable CPUs
5331 full,nosmt - Enable TAA mitigation and disable SMT on
5332 vulnerable CPUs. If TSX is disabled, SMT
5333 is not disabled because CPU is not
5334 vulnerable to cross-thread TAA attacks.
5335 off - Unconditionally disable TAA mitigation
5337 On MDS-affected machines, tsx_async_abort=off can be
5338 prevented by an active MDS mitigation as both vulnerabilities
5339 are mitigated with the same mechanism so in order to disable
5340 this mitigation, you need to specify mds=off too.
5342 Not specifying this option is equivalent to
5343 tsx_async_abort=full. On CPUs which are MDS affected
5344 and deploy MDS mitigation, TAA mitigation is not
5345 required and doesn't provide any additional
5349 Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
5351 turbografx.map[2|3]= [HW,JOY]
5352 TurboGraFX parallel port interface
5354 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7>
5355 See also Documentation/input/devices/joystick-parport.rst
5357 udbg-immortal [PPC] When debugging early kernel crashes that
5358 happen after console_init() and before a proper
5359 console driver takes over, this boot options might
5360 help "seeing" what's going on.
5362 uhash_entries= [KNL,NET]
5363 Set number of hash buckets for UDP/UDP-Lite connections
5366 [USB] Ignore overcurrent events (default N).
5367 Some badly-designed motherboards generate lots of
5368 bogus events, for ports that aren't wired to
5369 anything. Set this parameter to avoid log spamming.
5370 Note that genuine overcurrent events won't be
5374 [X86] Cause panic on unknown NMI.
5376 usbcore.authorized_default=
5377 [USB] Default USB device authorization:
5378 (default -1 = authorized except for wireless USB,
5379 0 = not authorized, 1 = authorized, 2 = authorized
5380 if device connected to internal port)
5382 usbcore.autosuspend=
5383 [USB] The autosuspend time delay (in seconds) used
5384 for newly-detected USB devices (default 2). This
5385 is the time required before an idle device will be
5386 autosuspended. Devices for which the delay is set
5387 to a negative value won't be autosuspended at all.
5389 usbcore.usbfs_snoop=
5390 [USB] Set to log all usbfs traffic (default 0 = off).
5392 usbcore.usbfs_snoop_max=
5393 [USB] Maximum number of bytes to snoop in each URB
5396 usbcore.blinkenlights=
5397 [USB] Set to cycle leds on hubs (default 0 = off).
5399 usbcore.old_scheme_first=
5400 [USB] Start with the old device initialization
5401 scheme (default 0 = off).
5403 usbcore.usbfs_memory_mb=
5404 [USB] Memory limit (in MB) for buffers allocated by
5405 usbfs (default = 16, 0 = max = 2047).
5407 usbcore.use_both_schemes=
5408 [USB] Try the other device initialization scheme
5409 if the first one fails (default 1 = enabled).
5411 usbcore.initial_descriptor_timeout=
5412 [USB] Specifies timeout for the initial 64-byte
5413 USB_REQ_GET_DESCRIPTOR request in milliseconds
5414 (default 5000 = 5.0 seconds).
5416 usbcore.nousb [USB] Disable the USB subsystem
5419 [USB] A list of quirk entries to augment the built-in
5420 usb core quirk list. List entries are separated by
5421 commas. Each entry has the form
5422 VendorID:ProductID:Flags. The IDs are 4-digit hex
5423 numbers and Flags is a set of letters. Each letter
5424 will change the built-in quirk; setting it if it is
5425 clear and clearing it if it is set. The letters have
5426 the following meanings:
5427 a = USB_QUIRK_STRING_FETCH_255 (string
5428 descriptors must not be fetched using
5430 b = USB_QUIRK_RESET_RESUME (device can't resume
5431 correctly so reset it instead);
5432 c = USB_QUIRK_NO_SET_INTF (device can't handle
5433 Set-Interface requests);
5434 d = USB_QUIRK_CONFIG_INTF_STRINGS (device can't
5435 handle its Configuration or Interface
5437 e = USB_QUIRK_RESET (device can't be reset
5438 (e.g morph devices), don't use reset);
5439 f = USB_QUIRK_HONOR_BNUMINTERFACES (device has
5440 more interface descriptions than the
5441 bNumInterfaces count, and can't handle
5442 talking to these interfaces);
5443 g = USB_QUIRK_DELAY_INIT (device needs a pause
5444 during initialization, after we read
5445 the device descriptor);
5446 h = USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL (For
5447 high speed and super speed interrupt
5448 endpoints, the USB 2.0 and USB 3.0 spec
5449 require the interval in microframes (1
5450 microframe = 125 microseconds) to be
5451 calculated as interval = 2 ^
5453 Devices with this quirk report their
5454 bInterval as the result of this
5455 calculation instead of the exponent
5456 variable used in the calculation);
5457 i = USB_QUIRK_DEVICE_QUALIFIER (device can't
5458 handle device_qualifier descriptor
5460 j = USB_QUIRK_IGNORE_REMOTE_WAKEUP (device
5461 generates spurious wakeup, ignore
5462 remote wakeup capability);
5463 k = USB_QUIRK_NO_LPM (device can't handle Link
5465 l = USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL
5466 (Device reports its bInterval as linear
5467 frames instead of the USB 2.0
5469 m = USB_QUIRK_DISCONNECT_SUSPEND (Device needs
5470 to be disconnected before suspend to
5471 prevent spurious wakeup);
5472 n = USB_QUIRK_DELAY_CTRL_MSG (Device needs a
5473 pause after every control message);
5474 o = USB_QUIRK_HUB_SLOW_RESET (Hub needs extra
5475 delay after resetting its port);
5476 Example: quirks=0781:5580:bk,0a5c:5834:gij
5479 [USBHID] The interval which mice are to be polled at.
5482 [USBHID] The interval which joysticks are to be polled at.
5485 [USBHID] The interval which keyboards are to be polled at.
5487 usb-storage.delay_use=
5488 [UMS] The delay in seconds before a new device is
5489 scanned for Logical Units (default 1).
5492 [UMS] A list of quirks entries to supplement or
5493 override the built-in unusual_devs list. List
5494 entries are separated by commas. Each entry has
5495 the form VID:PID:Flags where VID and PID are Vendor
5496 and Product ID values (4-digit hex numbers) and
5497 Flags is a set of characters, each corresponding
5498 to a common usb-storage quirk flag as follows:
5499 a = SANE_SENSE (collect more than 18 bytes
5500 of sense data, not on uas);
5501 b = BAD_SENSE (don't collect more than 18
5502 bytes of sense data, not on uas);
5503 c = FIX_CAPACITY (decrease the reported
5504 device capacity by one sector);
5505 d = NO_READ_DISC_INFO (don't use
5506 READ_DISC_INFO command, not on uas);
5507 e = NO_READ_CAPACITY_16 (don't use
5508 READ_CAPACITY_16 command);
5509 f = NO_REPORT_OPCODES (don't use report opcodes
5511 g = MAX_SECTORS_240 (don't transfer more than
5512 240 sectors at a time, uas only);
5513 h = CAPACITY_HEURISTICS (decrease the
5514 reported device capacity by one
5515 sector if the number is odd);
5516 i = IGNORE_DEVICE (don't bind to this
5518 j = NO_REPORT_LUNS (don't use report luns
5520 l = NOT_LOCKABLE (don't try to lock and
5521 unlock ejectable media, not on uas);
5522 m = MAX_SECTORS_64 (don't transfer more
5523 than 64 sectors = 32 KB at a time,
5525 n = INITIAL_READ10 (force a retry of the
5526 initial READ(10) command, not on uas);
5527 o = CAPACITY_OK (accept the capacity
5528 reported by the device, not on uas);
5529 p = WRITE_CACHE (the device cache is ON
5530 by default, not on uas);
5531 r = IGNORE_RESIDUE (the device reports
5532 bogus residue values, not on uas);
5533 s = SINGLE_LUN (the device has only one
5535 t = NO_ATA_1X (don't allow ATA(12) and ATA(16)
5536 commands, uas only);
5537 u = IGNORE_UAS (don't bind to the uas driver);
5538 w = NO_WP_DETECT (don't test whether the
5539 medium is write-protected).
5540 y = ALWAYS_SYNC (issue a SYNCHRONIZE_CACHE
5541 even if the device claims no cache,
5543 Example: quirks=0419:aaf5:rl,0421:0433:rc
5545 user_debug= [KNL,ARM]
5547 See arch/arm/Kconfig.debug help text.
5548 1 - undefined instruction events
5550 4 - invalid data aborts
5553 Example: user_debug=31
5556 [X86] Flags controlling user PTE allocations.
5558 nohigh = do not allocate PTE pages in
5559 HIGHMEM regardless of setting
5563 On X86_32, this is an alias for vdso32=. Otherwise:
5565 vdso=1: enable VDSO (the default)
5566 vdso=0: disable VDSO mapping
5568 vdso32= [X86] Control the 32-bit vDSO
5569 vdso32=1: enable 32-bit VDSO
5570 vdso32=0 or vdso32=2: disable 32-bit VDSO
5572 See the help text for CONFIG_COMPAT_VDSO for more
5573 details. If CONFIG_COMPAT_VDSO is set, the default is
5574 vdso32=0; otherwise, the default is vdso32=1.
5576 For compatibility with older kernels, vdso32=2 is an
5579 Try vdso32=0 if you encounter an error that says:
5580 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed!
5583 vector=percpu: enable percpu vector domain
5585 video= [FB] Frame buffer configuration
5586 See Documentation/fb/modedb.rst.
5588 video.brightness_switch_enabled= [0,1]
5589 If set to 1, on receiving an ACPI notify event
5590 generated by hotkey, video driver will adjust brightness
5591 level and then send out the event to user space through
5592 the allocated input device; If set to 0, video driver
5593 will only send out the event without touching backlight
5598 [VMMIO] Memory mapped virtio (platform) device.
5600 <size>@<baseaddr>:<irq>[:<id>]
5602 <size> := size (can use standard suffixes
5604 <baseaddr> := physical base address
5605 <irq> := interrupt number (as passed to
5607 <id> := (optional) platform device id
5609 virtio_mmio.device=1K@0x100b0000:48:7
5611 Can be used multiple times for multiple devices.
5613 vga= [BOOT,X86-32] Select a particular video mode
5614 See Documentation/x86/boot.rst and
5615 Documentation/admin-guide/svga.rst.
5616 Use vga=ask for menu.
5617 This is actually a boot loader parameter; the value is
5618 passed to the kernel using a special protocol.
5620 vm_debug[=options] [KNL] Available with CONFIG_DEBUG_VM=y.
5621 May slow down system boot speed, especially when
5622 enabled on systems with a large amount of memory.
5623 All options are enabled by default, and this
5624 interface is meant to allow for selectively
5625 enabling or disabling specific virtual memory
5628 Available options are:
5629 P Enable page structure init time poisoning
5630 - Disable all of the above options
5632 vmalloc=nn[KMG] [KNL,BOOT] Forces the vmalloc area to have an exact
5633 size of <nn>. This can be used to increase the
5634 minimum size (128MB on x86). It can also be used to
5635 decrease the size and leave more room for directly
5638 vmcp_cma=nn[MG] [KNL,S390]
5639 Sets the memory size reserved for contiguous memory
5640 allocations for the vmcp device driver.
5642 vmhalt= [KNL,S390] Perform z/VM CP command after system halt.
5645 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic.
5648 vmpoff= [KNL,S390] Perform z/VM CP command after power off.
5652 Controls the behavior of vsyscalls (i.e. calls to
5653 fixed addresses of 0xffffffffff600x00 from legacy
5654 code). Most statically-linked binaries and older
5655 versions of glibc use these calls. Because these
5656 functions are at fixed addresses, they make nice
5657 targets for exploits that can control RIP.
5659 emulate [default] Vsyscalls turn into traps and are
5660 emulated reasonably safely. The vsyscall
5663 xonly Vsyscalls turn into traps and are
5664 emulated reasonably safely. The vsyscall
5665 page is not readable.
5667 none Vsyscalls don't work at all. This makes
5668 them quite hard to use for exploits but
5669 might break your system.
5671 vt.color= [VT] Default text color.
5672 Format: 0xYX, X = foreground, Y = background.
5673 Default: 0x07 = light gray on black.
5675 vt.cur_default= [VT] Default cursor shape.
5676 Format: 0xCCBBAA, where AA, BB, and CC are the same as
5677 the parameters of the <Esc>[?A;B;Cc escape sequence;
5678 see VGA-softcursor.txt. Default: 2 = underline.
5680 vt.default_blu= [VT]
5681 Format: <blue0>,<blue1>,<blue2>,...,<blue15>
5682 Change the default blue palette of the console.
5683 This is a 16-member array composed of values
5686 vt.default_grn= [VT]
5687 Format: <green0>,<green1>,<green2>,...,<green15>
5688 Change the default green palette of the console.
5689 This is a 16-member array composed of values
5692 vt.default_red= [VT]
5693 Format: <red0>,<red1>,<red2>,...,<red15>
5694 Change the default red palette of the console.
5695 This is a 16-member array composed of values
5701 Set system-wide default UTF-8 mode for all tty's.
5702 Default is 1, i.e. UTF-8 mode is enabled for all
5703 newly opened terminals.
5705 vt.global_cursor_default=
5708 Set system-wide default for whether a cursor
5709 is shown on new VTs. Default is -1,
5710 i.e. cursors will be created by default unless
5711 overridden by individual drivers. 0 will hide
5712 cursors, 1 will display them.
5714 vt.italic= [VT] Default color for italic text; 0-15.
5717 vt.underline= [VT] Default color for underlined text; 0-15.
5720 watchdog timers [HW,WDT] For information on watchdog timers,
5721 see Documentation/watchdog/watchdog-parameters.rst
5722 or other driver-specific files in the
5723 Documentation/watchdog/ directory.
5727 Set the hard lockup detector stall duration
5728 threshold in seconds. The soft lockup detector
5729 threshold is set to twice the value. A value of 0
5730 disables both lockup detectors. Default is 10
5733 workqueue.watchdog_thresh=
5734 If CONFIG_WQ_WATCHDOG is configured, workqueue can
5735 warn stall conditions and dump internal state to
5736 help debugging. 0 disables workqueue stall
5737 detection; otherwise, it's the stall threshold
5738 duration in seconds. The default value is 30 and
5739 it can be updated at runtime by writing to the
5740 corresponding sysfs file.
5742 workqueue.disable_numa
5743 By default, all work items queued to unbound
5744 workqueues are affine to the NUMA nodes they're
5745 issued on, which results in better behavior in
5746 general. If NUMA affinity needs to be disabled for
5747 whatever reason, this option can be used. Note
5748 that this also can be controlled per-workqueue for
5749 workqueues visible under /sys/bus/workqueue/.
5751 workqueue.power_efficient
5752 Per-cpu workqueues are generally preferred because
5753 they show better performance thanks to cache
5754 locality; unfortunately, per-cpu workqueues tend to
5755 be more power hungry than unbound workqueues.
5757 Enabling this makes the per-cpu workqueues which
5758 were observed to contribute significantly to power
5759 consumption unbound, leading to measurably lower
5760 power usage at the cost of small performance
5763 The default value of this parameter is determined by
5764 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT.
5766 workqueue.debug_force_rr_cpu
5767 Workqueue used to implicitly guarantee that work
5768 items queued without explicit CPU specified are put
5769 on the local CPU. This guarantee is no longer true
5770 and while local CPU is still preferred work items
5771 may be put on foreign CPUs. This debug option
5772 forces round-robin CPU selection to flush out
5773 usages which depend on the now broken guarantee.
5774 When enabled, memory and cache locality will be
5777 x2apic_phys [X86-64,APIC] Use x2apic physical mode instead of
5778 default x2apic cluster mode on platforms
5781 x86_intel_mid_timer= [X86-32,APBT]
5782 Choose timer option for x86 Intel MID platform.
5783 Two valid options are apbt timer only and lapic timer
5784 plus one apbt timer for broadcast timer.
5785 x86_intel_mid_timer=apbt_only | lapic_and_apbt
5787 xen_512gb_limit [KNL,X86-64,XEN]
5788 Restricts the kernel running paravirtualized under Xen
5789 to use only up to 512 GB of RAM. The reason to do so is
5790 crash analysis tools and Xen tools for doing domain
5791 save/restore/migration must be enabled to handle larger
5794 xen_emul_unplug= [HW,X86,XEN]
5795 Unplug Xen emulated devices
5796 Format: [unplug0,][unplug1]
5797 ide-disks -- unplug primary master IDE devices
5798 aux-ide-disks -- unplug non-primary-master IDE devices
5799 nics -- unplug network devices
5800 all -- unplug all emulated devices (NICs and IDE disks)
5801 unnecessary -- unplugging emulated devices is
5802 unnecessary even if the host did not respond to
5804 never -- do not unplug even if version check succeeds
5806 xen_legacy_crash [X86,XEN]
5807 Crash from Xen panic notifier, without executing late
5808 panic() code such as dumping handler.
5810 xen_nopvspin [X86,XEN]
5811 Disables the qspinlock slowpath using Xen PV optimizations.
5812 This parameter is obsoleted by "nopvspin" parameter, which
5813 has equivalent effect for XEN platform.
5816 Disables the PV optimizations forcing the HVM guest to
5817 run as generic HVM guest with no PV drivers.
5818 This option is obsoleted by the "nopv" option, which
5819 has equivalent effect for XEN platform.
5821 xen_scrub_pages= [XEN]
5822 Boolean option to control scrubbing pages before giving them back
5823 to Xen, for use by other domains. Can be also changed at runtime
5824 with /sys/devices/system/xen_memory/xen_memory0/scrub_pages.
5825 Default value controlled with CONFIG_XEN_SCRUB_PAGES_DEFAULT.
5827 xen_timer_slop= [X86-64,XEN]
5828 Set the timer slop (in nanoseconds) for the virtual Xen
5829 timers (default is 100000). This adjusts the minimum
5830 delta of virtualized Xen timers, where lower values
5831 improve timer resolution at the expense of processing
5832 more timer interrupts.
5834 nopv= [X86,XEN,KVM,HYPER_V,VMWARE]
5835 Disables the PV optimizations forcing the guest to run
5836 as generic guest with no PV drivers. Currently support
5837 XEN HVM, KVM, HYPER_V and VMWARE guest.
5839 nopvspin [X86,XEN,KVM]
5840 Disables the qspinlock slow path using PV optimizations
5841 which allow the hypervisor to 'idle' the guest on lock
5844 xirc2ps_cs= [NET,PCMCIA]
5846 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]]
5849 By default on POWER9 and above, the kernel will
5850 natively use the XIVE interrupt controller. This option
5851 allows the fallback firmware mode to be used:
5853 off Fallback to firmware control of XIVE interrupt
5854 controller on both pseries and powernv
5855 platforms. Only useful on POWER9 and above.
5857 xhci-hcd.quirks [USB,KNL]
5858 A hex value specifying bitmask with supplemental xhci
5859 host controller quirks. Meaning of each bit can be
5860 consulted in header drivers/usb/host/xhci.h.
5863 Format: { early | on | rw | ro | off }
5864 Controls if xmon debugger is enabled. Default is off.
5865 Passing only "xmon" is equivalent to "xmon=early".
5866 early Call xmon as early as possible on boot; xmon
5867 debugger is called from setup_arch().
5868 on xmon debugger hooks will be installed so xmon
5869 is only called on a kernel crash. Default mode,
5870 i.e. either "ro" or "rw" mode, is controlled
5871 with CONFIG_XMON_DEFAULT_RO_MODE.
5872 rw xmon debugger hooks will be installed so xmon
5873 is called only on a kernel crash, mode is write,
5874 meaning SPR registers, memory and, other data
5875 can be written using xmon commands.
5876 ro same as "rw" option above but SPR registers,
5877 memory, and other data can't be written using
5879 off xmon is disabled.