1 acpi= [HW,ACPI,X86,ARM64]
2 Advanced Configuration and Power Interface
3 Format: { force | on | off | strict | noirq | rsdt |
5 force -- enable ACPI if default was off
6 on -- enable ACPI but allow fallback to DT [arm64]
7 off -- disable ACPI if default was on
8 noirq -- do not use ACPI for IRQ routing
9 strict -- Be less tolerant of platforms that are not
10 strictly ACPI specification compliant.
11 rsdt -- prefer RSDT over (default) XSDT
12 copy_dsdt -- copy DSDT to memory
13 For ARM64, ONLY "acpi=off", "acpi=on" or "acpi=force"
16 See also Documentation/power/runtime_pm.rst, pci=noacpi
18 acpi_apic_instance= [ACPI, IOAPIC]
20 2: use 2nd APIC table, if available
21 1,0: use 1st APIC table
24 acpi_backlight= [HW,ACPI]
25 { vendor | video | native | none }
26 If set to vendor, prefer vendor-specific driver
27 (e.g. thinkpad_acpi, sony_acpi, etc.) instead
28 of the ACPI video.ko driver.
29 If set to video, use the ACPI video.ko driver.
30 If set to native, use the device's native backlight mode.
31 If set to none, disable the ACPI backlight interface.
33 acpi_force_32bit_fadt_addr
34 force FADT to use 32 bit addresses rather than the
35 64 bit X_* addresses. Some firmware have broken 64
36 bit addresses for force ACPI ignore these and use
37 the older legacy 32 bit addresses.
39 acpica_no_return_repair [HW, ACPI]
40 Disable AML predefined validation mechanism
41 This mechanism can repair the evaluation result to make
42 the return objects more ACPI specification compliant.
43 This option is useful for developers to identify the
44 root cause of an AML interpreter issue when the issue
45 has something to do with the repair mechanism.
47 acpi.debug_layer= [HW,ACPI,ACPI_DEBUG]
48 acpi.debug_level= [HW,ACPI,ACPI_DEBUG]
50 CONFIG_ACPI_DEBUG must be enabled to produce any ACPI
51 debug output. Bits in debug_layer correspond to a
52 _COMPONENT in an ACPI source file, e.g.,
53 #define _COMPONENT ACPI_EVENTS
54 Bits in debug_level correspond to a level in
55 ACPI_DEBUG_PRINT statements, e.g.,
56 ACPI_DEBUG_PRINT((ACPI_DB_INFO, ...
57 The debug_level mask defaults to "info". See
58 Documentation/firmware-guide/acpi/debug.rst for more information about
59 debug layers and levels.
61 Enable processor driver info messages:
62 acpi.debug_layer=0x20000000
63 Enable AML "Debug" output, i.e., stores to the Debug
64 object while interpreting AML:
65 acpi.debug_layer=0xffffffff acpi.debug_level=0x2
66 Enable all messages related to ACPI hardware:
67 acpi.debug_layer=0x2 acpi.debug_level=0xffffffff
69 Some values produce so much output that the system is
70 unusable. The "log_buf_len" parameter may be useful
71 if you need to capture more output.
73 acpi_enforce_resources= [ACPI]
75 Check for resource conflicts between native drivers
76 and ACPI OperationRegions (SystemIO and SystemMemory
77 only). IO ports and memory declared in ACPI might be
78 used by the ACPI subsystem in arbitrary AML code and
79 can interfere with legacy drivers.
80 strict (default): access to resources claimed by ACPI
81 is denied; legacy drivers trying to access reserved
82 resources will fail to bind to device using them.
83 lax: access to resources claimed by ACPI is allowed;
84 legacy drivers trying to access reserved resources
85 will bind successfully but a warning message is logged.
86 no: ACPI OperationRegions are not marked as reserved,
87 no further checks are performed.
89 acpi_force_table_verification [HW,ACPI]
90 Enable table checksum verification during early stage.
91 By default, this is disabled due to x86 early mapping
94 acpi_irq_balance [HW,ACPI]
95 ACPI will balance active IRQs
98 acpi_irq_nobalance [HW,ACPI]
99 ACPI will not move active IRQs (default)
102 acpi_irq_isa= [HW,ACPI] If irq_balance, mark listed IRQs used by ISA
103 Format: <irq>,<irq>...
105 acpi_irq_pci= [HW,ACPI] If irq_balance, clear listed IRQs for
107 Format: <irq>,<irq>...
109 acpi_mask_gpe= [HW,ACPI]
110 Due to the existence of _Lxx/_Exx, some GPEs triggered
111 by unsupported hardware/firmware features can result in
112 GPE floodings that cannot be automatically disabled by
114 This facility can be used to prevent such uncontrolled
116 Format: <byte> or <bitmap-list>
118 acpi_no_auto_serialize [HW,ACPI]
119 Disable auto-serialization of AML methods
120 AML control methods that contain the opcodes to create
121 named objects will be marked as "Serialized" by the
122 auto-serialization feature.
123 This feature is enabled by default.
124 This option allows to turn off the feature.
126 acpi_no_memhotplug [ACPI] Disable memory hotplug. Useful for kdump
129 acpi_no_static_ssdt [HW,ACPI]
130 Disable installation of static SSDTs at early boot time
131 By default, SSDTs contained in the RSDT/XSDT will be
132 installed automatically and they will appear under
133 /sys/firmware/acpi/tables.
134 This option turns off this feature.
135 Note that specifying this option does not affect
136 dynamic table installation which will install SSDT
137 tables to /sys/firmware/acpi/tables/dynamic.
139 acpi_no_watchdog [HW,ACPI,WDT]
140 Ignore the ACPI-based watchdog interface (WDAT) and let
141 a native driver control the watchdog device instead.
143 acpi_rsdp= [ACPI,EFI,KEXEC]
144 Pass the RSDP address to the kernel, mostly used
145 on machines running EFI runtime service to boot the
146 second kernel for kdump.
148 acpi_os_name= [HW,ACPI] Tell ACPI BIOS the name of the OS
149 Format: To spoof as Windows 98: ="Microsoft Windows"
151 acpi_rev_override [ACPI] Override the _REV object to return 5 (instead
152 of 2 which is mandated by ACPI 6) as the supported ACPI
153 specification revision (when using this switch, it may
154 be necessary to carry out a cold reboot _twice_ in a
155 row to make it take effect on the platform firmware).
157 acpi_osi= [HW,ACPI] Modify list of supported OS interface strings
158 acpi_osi="string1" # add string1
159 acpi_osi="!string2" # remove string2
160 acpi_osi=!* # remove all strings
161 acpi_osi=! # disable all built-in OS vendor
163 acpi_osi=!! # enable all built-in OS vendor
165 acpi_osi= # disable all strings
167 'acpi_osi=!' can be used in combination with single or
168 multiple 'acpi_osi="string1"' to support specific OS
169 vendor string(s). Note that such command can only
170 affect the default state of the OS vendor strings, thus
171 it cannot affect the default state of the feature group
172 strings and the current state of the OS vendor strings,
173 specifying it multiple times through kernel command line
174 is meaningless. This command is useful when one do not
175 care about the state of the feature group strings which
176 should be controlled by the OSPM.
178 1. 'acpi_osi=! acpi_osi="Windows 2000"' is equivalent
179 to 'acpi_osi="Windows 2000" acpi_osi=!', they all
180 can make '_OSI("Windows 2000")' TRUE.
182 'acpi_osi=' cannot be used in combination with other
183 'acpi_osi=' command lines, the _OSI method will not
184 exist in the ACPI namespace. NOTE that such command can
185 only affect the _OSI support state, thus specifying it
186 multiple times through kernel command line is also
189 1. 'acpi_osi=' can make 'CondRefOf(_OSI, Local1)'
192 'acpi_osi=!*' can be used in combination with single or
193 multiple 'acpi_osi="string1"' to support specific
194 string(s). Note that such command can affect the
195 current state of both the OS vendor strings and the
196 feature group strings, thus specifying it multiple times
197 through kernel command line is meaningful. But it may
198 still not able to affect the final state of a string if
199 there are quirks related to this string. This command
200 is useful when one want to control the state of the
201 feature group strings to debug BIOS issues related to
204 1. 'acpi_osi="Module Device" acpi_osi=!*' can make
205 '_OSI("Module Device")' FALSE.
206 2. 'acpi_osi=!* acpi_osi="Module Device"' can make
207 '_OSI("Module Device")' TRUE.
208 3. 'acpi_osi=! acpi_osi=!* acpi_osi="Windows 2000"' is
210 'acpi_osi=!* acpi_osi=! acpi_osi="Windows 2000"'
212 'acpi_osi=!* acpi_osi="Windows 2000" acpi_osi=!',
213 they all will make '_OSI("Windows 2000")' TRUE.
216 Override the pmtimer bug detection: force the kernel
217 to assume that this machine's pmtimer latches its value
218 and always returns good values.
220 acpi_sci= [HW,ACPI] ACPI System Control Interrupt trigger mode
221 Format: { level | edge | high | low }
223 acpi_skip_timer_override [HW,ACPI]
224 Recognize and ignore IRQ0/pin2 Interrupt Override.
225 For broken nForce2 BIOS resulting in XT-PIC timer.
227 acpi_sleep= [HW,ACPI] Sleep options
228 Format: { s3_bios, s3_mode, s3_beep, s4_nohwsig,
229 old_ordering, nonvs, sci_force_enable, nobl }
230 See Documentation/power/video.rst for information on
232 s3_beep is for debugging; it makes the PC's speaker beep
233 as soon as the kernel's real-mode entry point is called.
234 s4_nohwsig prevents ACPI hardware signature from being
235 used during resume from hibernation.
236 old_ordering causes the ACPI 1.0 ordering of the _PTS
237 control method, with respect to putting devices into
238 low power states, to be enforced (the ACPI 2.0 ordering
239 of _PTS is used by default).
240 nonvs prevents the kernel from saving/restoring the
241 ACPI NVS memory during suspend/hibernation and resume.
242 sci_force_enable causes the kernel to set SCI_EN directly
243 on resume from S1/S3 (which is against the ACPI spec,
244 but some broken systems don't work without it).
245 nobl causes the internal blacklist of systems known to
246 behave incorrectly in some ways with respect to system
247 suspend and resume to be ignored (use wisely).
249 acpi_use_timer_override [HW,ACPI]
250 Use timer override. For some broken Nvidia NF5 boards
251 that require a timer override, but don't have HPET
253 add_efi_memmap [EFI; X86] Include EFI memory map in
254 kernel's map of available physical RAM.
257 { off | try_unsupported }
258 off: disable AGP support
259 try_unsupported: try to drive unsupported chipsets
260 (may crash computer or cause data corruption)
263 See Documentation/sound/alsa-configuration.rst
266 Allow the default userspace alignment fault handler
267 behaviour to be specified. Bit 0 enables warnings,
268 bit 1 enables fixups, and bit 2 sends a segfault.
270 align_va_addr= [X86-64]
271 Align virtual addresses by clearing slice [14:12] when
272 allocating a VMA at process creation time. This option
273 gives you up to 3% performance improvement on AMD F15h
274 machines (where it is enabled by default) for a
275 CPU-intensive style benchmark, and it can vary highly in
276 a microbenchmark depending on workload and compiler.
278 32: only for 32-bit processes
279 64: only for 64-bit processes
280 on: enable for both 32- and 64-bit processes
281 off: disable for both 32- and 64-bit processes
283 alloc_snapshot [FTRACE]
284 Allocate the ftrace snapshot buffer on boot up when the
285 main buffer is allocated. This is handy if debugging
286 and you need to use tracing_snapshot() on boot up, and
287 do not want to use tracing_snapshot_alloc() as it needs
288 to be done where GFP_KERNEL allocations are allowed.
290 allow_mismatched_32bit_el0 [ARM64]
291 Allow execve() of 32-bit applications and setting of the
292 PER_LINUX32 personality on systems where only a strict
293 subset of the CPUs support 32-bit EL0. When this
294 parameter is present, the set of CPUs supporting 32-bit
295 EL0 is indicated by /sys/devices/system/cpu/aarch32_el0
296 and hot-unplug operations may be restricted.
298 See Documentation/arm64/asymmetric-32bit.rst for more
301 amd_iommu= [HW,X86-64]
302 Pass parameters to the AMD IOMMU driver in the system.
304 fullflush - Deprecated, equivalent to iommu.strict=1
305 off - do not initialize any AMD IOMMU found in
307 force_isolation - Force device isolation for all
308 devices. The IOMMU driver is not
309 allowed anymore to lift isolation
310 requirements as needed. This option
311 does not override iommu=pt
312 force_enable - Force enable the IOMMU on platforms known
313 to be buggy with IOMMU enabled. Use this
316 amd_iommu_dump= [HW,X86-64]
317 Enable AMD IOMMU driver option to dump the ACPI table
318 for AMD IOMMU. With this option enabled, AMD IOMMU
319 driver will print ACPI tables for AMD IOMMU during
320 IOMMU initialization.
322 amd_iommu_intr= [HW,X86-64]
323 Specifies one of the following AMD IOMMU interrupt
325 legacy - Use legacy interrupt remapping mode.
326 vapic - Use virtual APIC mode, which allows IOMMU
327 to inject interrupts directly into guest.
328 This mode requires kvm-amd.avic=1.
329 (Default when IOMMU HW support is present.)
331 amijoy.map= [HW,JOY] Amiga joystick support
332 Map of devices attached to JOY0DAT and JOY1DAT
334 See also Documentation/input/joydev/joystick.rst
336 analog.map= [HW,JOY] Analog joystick and gamepad support
337 Specifies type or capabilities of an analog joystick
338 connected to one of 16 gameports
339 Format: <type1>,<type2>,..<type16>
342 Power management functions (SPARCstation-4/5 + deriv.)
344 Disable APC CPU standby support. SPARCstation-Fox does
345 not play well with APC CPU idle - disable it if you have
346 APC and your system crashes randomly.
348 apic= [APIC,X86] Advanced Programmable Interrupt Controller
349 Change the output verbosity while booting
350 Format: { quiet (default) | verbose | debug }
351 Change the amount of debugging information output
352 when initialising the APIC and IO-APIC components.
353 For X86-32, this can also be used to specify an APIC
355 Format: apic=driver_name
356 Examples: apic=bigsmp
358 apic_extnmi= [APIC,X86] External NMI delivery setting
359 Format: { bsp (default) | all | none }
360 bsp: External NMI is delivered only to CPU 0
361 all: External NMIs are broadcast to all CPUs as a
363 none: External NMI is masked for all CPUs. This is
364 useful so that a dump capture kernel won't be
368 See Documentation/networking/ipv6.rst.
370 show_lapic= [APIC,X86] Advanced Programmable Interrupt Controller
371 Limit apic dumping. The parameter defines the maximal
372 number of local apics being dumped. Also it is possible
373 to set it to "all" by meaning -- no limit here.
374 Format: { 1 (default) | 2 | ... | all }.
375 The parameter valid if only apic=debug or
376 apic=verbose is specified.
377 Example: apic=debug show_lapic=all
379 apm= [APM] Advanced Power Management
380 See header of arch/x86/kernel/apm_32.c.
382 arcrimi= [HW,NET] ARCnet - "RIM I" (entirely mem-mapped) cards
383 Format: <io>,<irq>,<nodeID>
385 arm64.nobti [ARM64] Unconditionally disable Branch Target
386 Identification support
388 arm64.nopauth [ARM64] Unconditionally disable Pointer Authentication
391 arm64.nomte [ARM64] Unconditionally disable Memory Tagging Extension
396 atarimouse= [HW,MOUSE] Atari Mouse
398 atkbd.extra= [HW] Enable extra LEDs and keys on IBM RapidAccess,
399 EzKey and similar keyboards
401 atkbd.reset= [HW] Reset keyboard during initialization
403 atkbd.set= [HW] Select keyboard code set
404 Format: <int> (2 = AT (default), 3 = PS/2)
406 atkbd.scroll= [HW] Enable scroll wheel on MS Office and similar
409 atkbd.softraw= [HW] Choose between synthetic and real raw mode
410 Format: <bool> (0 = real, 1 = synthetic (default))
412 atkbd.softrepeat= [HW]
413 Use software keyboard repeat
415 audit= [KNL] Enable the audit sub-system
416 Format: { "0" | "1" | "off" | "on" }
417 0 | off - kernel audit is disabled and can not be
418 enabled until the next reboot
419 unset - kernel audit is initialized but disabled and
420 will be fully enabled by the userspace auditd.
421 1 | on - kernel audit is initialized and partially
422 enabled, storing at most audit_backlog_limit
423 messages in RAM until it is fully enabled by the
427 audit_backlog_limit= [KNL] Set the audit queue size limit.
428 Format: <int> (must be >=0)
431 bau= [X86_UV] Enable the BAU on SGI UV. The default
432 behavior is to disable the BAU (i.e. bau=0).
433 Format: { "0" | "1" }
436 unset - Disable the BAU.
438 baycom_epp= [HW,AX25]
441 baycom_par= [HW,AX25] BayCom Parallel Port AX.25 Modem
443 See header of drivers/net/hamradio/baycom_par.c.
445 baycom_ser_fdx= [HW,AX25]
446 BayCom Serial Port AX.25 Modem (Full Duplex Mode)
447 Format: <io>,<irq>,<mode>[,<baud>]
448 See header of drivers/net/hamradio/baycom_ser_fdx.c.
450 baycom_ser_hdx= [HW,AX25]
451 BayCom Serial Port AX.25 Modem (Half Duplex Mode)
452 Format: <io>,<irq>,<mode>
453 See header of drivers/net/hamradio/baycom_ser_hdx.c.
455 blkdevparts= Manual partition parsing of block device(s) for
456 embedded devices based on command line input.
457 See Documentation/block/cmdline-partition.rst
459 boot_delay= Milliseconds to delay each printk during boot.
460 Values larger than 10 seconds (10000) are changed to
465 Extended command line options can be added to an initrd
466 and this will cause the kernel to look for it.
468 See Documentation/admin-guide/bootconfig.rst
471 Disable BERT OS support on buggy BIOSes.
473 bgrt_disable [ACPI][X86]
474 Disable BGRT to avoid flickering OEM logo.
476 bttv.card= [HW,V4L] bttv (bt848 + bt878 based grabber cards)
477 bttv.radio= Most important insmod options are available as
479 bttv.pll= See Documentation/admin-guide/media/bttv.rst
482 bulk_remove=off [PPC] This parameter disables the use of the pSeries
483 firmware feature for flushing multiple hpte entries
486 c101= [NET] Moxa C101 synchronous serial card
488 cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection.
489 Sometimes CPU hardware bugs make them report the cache
490 size incorrectly. The kernel will attempt work arounds
491 to fix known problems, but for some CPUs it is not
492 possible to determine what the correct size should be.
493 This option provides an override for these situations.
496 [NET] Specifies amount of time (in seconds) that
497 the kernel should wait for a network carrier. By default
498 it waits 120 seconds.
500 ca_keys= [KEYS] This parameter identifies a specific key(s) on
501 the system trusted keyring to be used for certificate
503 format: { id:<keyid> | builtin }
505 cca= [MIPS] Override the kernel pages' cache coherency
506 algorithm. Accepted values range from 0 to 7
507 inclusive. See arch/mips/include/asm/pgtable-bits.h
508 for platform specific values (SB1, Loongson3 and
511 ccw_timeout_log [S390]
512 See Documentation/s390/common_io.rst for details.
514 cgroup_disable= [KNL] Disable a particular controller or optional feature
515 Format: {name of the controller(s) or feature(s) to disable}
516 The effects of cgroup_disable=foo are:
517 - foo isn't auto-mounted if you mount all cgroups in
519 - foo isn't visible as an individually mountable
521 - if foo is an optional feature then the feature is
522 disabled and corresponding cgroup files are not
524 {Currently only "memory" controller deal with this and
525 cut the overhead, others just disable the usage. So
526 only cgroup_disable=memory is actually worthy}
527 Specifying "pressure" disables per-cgroup pressure
528 stall information accounting feature
530 cgroup_no_v1= [KNL] Disable cgroup controllers and named hierarchies in v1
531 Format: { { controller | "all" | "named" }
532 [,{ controller | "all" | "named" }...] }
533 Like cgroup_disable, but only applies to cgroup v1;
534 the blacklisted controllers remain available in cgroup2.
535 "all" blacklists all controllers and "named" disables
536 named mounts. Specifying both "all" and "named" disables
539 cgroup.memory= [KNL] Pass options to the cgroup memory controller.
541 nosocket -- Disable socket memory accounting.
542 nokmem -- Disable kernel memory accounting.
544 checkreqprot [SELINUX] Set initial checkreqprot flag value.
545 Format: { "0" | "1" }
546 See security/selinux/Kconfig help text.
547 0 -- check protection applied by kernel (includes
548 any implied execute protection).
549 1 -- check protection requested by application.
550 Default value is set via a kernel config option.
551 Value can be changed at runtime via
552 /sys/fs/selinux/checkreqprot.
553 Setting checkreqprot to 1 is deprecated.
556 See Documentation/s390/common_io.rst for details.
559 Prevents the clock framework from automatically gating
560 clocks that have not been explicitly enabled by a Linux
561 device driver but are enabled in hardware at reset or
562 by the bootloader/firmware. Note that this does not
563 force such clocks to be always-on nor does it reserve
564 those clocks in any way. This parameter is useful for
565 debug and development, but should not be needed on a
566 platform with proper driver support. For more
567 information, see Documentation/driver-api/clk.rst.
569 clock= [BUGS=X86-32, HW] gettimeofday clocksource override.
571 Forces specified clocksource (if available) to be used
572 when calculating gettimeofday(). If specified
573 clocksource is not available, it defaults to PIT.
574 Format: { pit | tsc | cyclone | pmtmr }
576 clocksource= Override the default clocksource
578 Override the default clocksource and use the clocksource
579 with the name specified.
580 Some clocksource names to choose from, depending on
582 [all] jiffies (this is the base, fallback clocksource)
584 [ARM] imx_timer1,OSTS,netx_timer,mpu_timer2,
585 pxa_timer,timer3,32k_counter,timer0_1
586 [X86-32] pit,hpet,tsc;
587 scx200_hrt on Geode; cyclone on IBM x440
595 clocksource.arm_arch_timer.evtstrm=
598 Enable/disable the eventstream feature of the ARM
599 architected timer so that code using WFE-based polling
600 loops can be debugged more effectively on production
603 clocksource.max_cswd_read_retries= [KNL]
604 Number of clocksource_watchdog() retries due to
605 external delays before the clock will be marked
606 unstable. Defaults to three retries, that is,
607 four attempts to read the clock under test.
609 clocksource.verify_n_cpus= [KNL]
610 Limit the number of CPUs checked for clocksources
611 marked with CLOCK_SOURCE_VERIFY_PERCPU that
612 are marked unstable due to excessive skew.
613 A negative value says to check all CPUs, while
614 zero says not to check any. Values larger than
615 nr_cpu_ids are silently truncated to nr_cpu_ids.
616 The actual CPUs are chosen randomly, with
617 no replacement if the same CPU is chosen twice.
619 clocksource-wdtest.holdoff= [KNL]
620 Set the time in seconds that the clocksource
621 watchdog test waits before commencing its tests.
622 Defaults to zero when built as a module and to
623 10 seconds when built into the kernel.
625 clearcpuid=BITNUM[,BITNUM...] [X86]
626 Disable CPUID feature X for the kernel. See
627 arch/x86/include/asm/cpufeatures.h for the valid bit
628 numbers. Note the Linux specific bits are not necessarily
629 stable over kernel options, but the vendor specific
631 Also note that user programs calling CPUID directly
632 or using the feature without checking anything
633 will still see it. This just prevents it from
634 being used by the kernel or shown in /proc/cpuinfo.
635 Also note the kernel might malfunction if you disable
638 cma=nn[MG]@[start[MG][-end[MG]]]
640 Sets the size of kernel global memory area for
641 contiguous memory allocations and optionally the
642 placement constraint by the physical address range of
643 memory allocations. A value of 0 disables CMA
644 altogether. For more information, see
645 kernel/dma/contiguous.c
649 Sets the size of kernel per-numa memory area for
650 contiguous memory allocations. A value of 0 disables
651 per-numa CMA altogether. And If this option is not
652 specificed, the default value is 0.
653 With per-numa CMA enabled, DMA users on node nid will
654 first try to allocate buffer from the pernuma area
655 which is located in node nid, if the allocation fails,
656 they will fallback to the global default memory area.
658 cmo_free_hint= [PPC] Format: { yes | no }
659 Specify whether pages are marked as being inactive
660 when they are freed. This is used in CMO environments
661 to determine OS memory pressure for page stealing by
665 coherent_pool=nn[KMG] [ARM,KNL]
666 Sets the size of memory pool for coherent, atomic dma
667 allocations, by default set to 256K.
669 com20020= [HW,NET] ARCnet - COM20020 chipset
671 <io>[,<irq>[,<nodeID>[,<backplane>[,<ckp>[,<timeout>]]]]]
673 com90io= [HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers)
677 ARCnet - COM90xx chipset (memory-mapped buffers)
678 Format: <io>[,<irq>[,<memstart>]]
680 condev= [HW,S390] console device
683 console= [KNL] Output console device and options.
685 tty<n> Use the virtual console device <n>.
689 Use the specified serial port. The options are of
690 the form "bbbbpnf", where "bbbb" is the baud rate,
691 "p" is parity ("n", "o", or "e"), "n" is number of
692 bits, and "f" is flow control ("r" for RTS or
693 omit it). Default is "9600n8".
695 See Documentation/admin-guide/serial-console.rst for more
697 Documentation/networking/netconsole.rst for an
700 uart[8250],io,<addr>[,options]
701 uart[8250],mmio,<addr>[,options]
702 uart[8250],mmio16,<addr>[,options]
703 uart[8250],mmio32,<addr>[,options]
704 uart[8250],0x<addr>[,options]
705 Start an early, polled-mode console on the 8250/16550
706 UART at the specified I/O port or MMIO address,
707 switching to the matching ttyS device later.
708 MMIO inter-register address stride is either 8-bit
709 (mmio), 16-bit (mmio16), or 32-bit (mmio32).
710 If none of [io|mmio|mmio16|mmio32], <addr> is assumed
711 to be equivalent to 'mmio'. 'options' are specified in
712 the same format described for ttyS above; if unspecified,
713 the h/w is not re-initialized.
715 hvc<n> Use the hypervisor console device <n>. This is for
716 both Xen and PowerPC hypervisors.
718 If the device connected to the port is not a TTY but a braille
719 device, prepend "brl," before the device type, for instance
721 For now, only VisioBraille is supported.
724 [KNL] Change console messages format
726 By default we print messages on consoles in
727 "[time stamp] text\n" format (time stamp may not be
728 printed, depending on CONFIG_PRINTK_TIME or
729 `printk_time' param).
731 Switch to syslog format: "<%u>[time stamp] text\n"
732 IOW, each message will have a facility and loglevel
733 prefix. The format is similar to one used by syslog()
734 syscall, or to executing "dmesg -S --raw" or to reading
737 consoleblank= [KNL] The console blank (screen saver) timeout in
738 seconds. A value of 0 disables the blank timer.
742 [KNL] Change the default value for
743 /proc/<pid>/coredump_filter.
744 See also Documentation/filesystems/proc.rst.
746 coresight_cpu_debug.enable
749 Enable/disable the CPU sampling based debugging.
750 0: default value, disable debugging
751 1: enable debugging at boot time
753 cpuidle.off=1 [CPU_IDLE]
754 disable the cpuidle sub-system
757 [CPU_IDLE] Name of the cpuidle governor to use.
759 cpufreq.off=1 [CPU_FREQ]
760 disable the cpufreq sub-system
762 cpufreq.default_governor=
763 [CPU_FREQ] Name of the default cpufreq governor or
764 policy to use. This governor must be registered in the
765 kernel before the cpufreq driver probes.
768 [X86] Delay for N microsec between assert and de-assert
769 of APIC INIT to start processors. This delay occurs
770 on every CPU online, such as boot, and resume from suspend.
773 cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver
775 <first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>]
777 crashkernel=size[KMG][@offset[KMG]]
778 [KNL] Using kexec, Linux can switch to a 'crash kernel'
779 upon panic. This parameter reserves the physical
780 memory region [offset, offset + size] for that kernel
781 image. If '@offset' is omitted, then a suitable offset
782 is selected automatically.
783 [KNL, X86-64] Select a region under 4G first, and
784 fall back to reserve region above 4G when '@offset'
785 hasn't been specified.
786 See Documentation/admin-guide/kdump/kdump.rst for further details.
788 crashkernel=range1:size1[,range2:size2,...][@offset]
789 [KNL] Same as above, but depends on the memory
790 in the running system. The syntax of range is
791 start-[end] where start and end are both
792 a memory unit (amount[KMG]). See also
793 Documentation/admin-guide/kdump/kdump.rst for an example.
795 crashkernel=size[KMG],high
796 [KNL, X86-64] range could be above 4G. Allow kernel
797 to allocate physical memory region from top, so could
798 be above 4G if system have more than 4G ram installed.
799 Otherwise memory region will be allocated below 4G, if
801 It will be ignored if crashkernel=X is specified.
802 crashkernel=size[KMG],low
803 [KNL, X86-64] range under 4G. When crashkernel=X,high
804 is passed, kernel could allocate physical memory region
805 above 4G, that cause second kernel crash on system
806 that require some amount of low memory, e.g. swiotlb
807 requires at least 64M+32K low memory, also enough extra
808 low memory is needed to make sure DMA buffers for 32-bit
809 devices won't run out. Kernel would try to allocate at
810 at least 256M below 4G automatically.
811 This one let user to specify own low range under 4G
812 for second kernel instead.
813 0: to disable low allocation.
814 It will be ignored when crashkernel=X,high is not used
815 or memory reserved is below 4G.
818 [KNL] Disable crypto self-tests
823 cs89x0_media= [HW,NET]
824 Format: { rj45 | aui | bnc }
826 csdlock_debug= [KNL] Enable debug add-ons of cross-CPU function call
827 handling. When switched on, additional debug data is
828 printed to the console in case a hanging CPU is
829 detected, and that CPU is pinged again in order to try
830 to resolve the hang situation.
831 0: disable csdlock debugging (default)
832 1: enable basic csdlock debugging (minor impact)
833 ext: enable extended csdlock debugging (more impact,
837 See header of drivers/s390/block/dasd_devmap.c.
839 db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port
840 (one device per port)
841 Format: <port#>,<type>
842 See also Documentation/input/devices/joystick-parport.rst
844 debug [KNL] Enable kernel debugging (events log level).
847 [KNL] Enable printing [hashed] pointers early in the
848 boot sequence. If enabled, we use a weak hash instead
849 of siphash to hash pointers. Use this option if you are
850 seeing instances of '(___ptrval___)') and need to see a
851 value (hashed pointer) instead. Cryptographically
852 insecure, please do not use on production kernels.
855 [KNL] verbose locking self-tests
857 Print debugging info while doing the locking API
859 Bitmask for the various LOCKTYPE_ tests. Defaults to 0
860 (no extra messages), setting it to -1 (all bits set)
861 will print _a_lot_ more information - normally only
862 useful to lockdep developers.
864 debug_objects [KNL] Enable object debugging
867 [KNL] Disable object debugging
869 debug_guardpage_minorder=
870 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this
871 parameter allows control of the order of pages that will
872 be intentionally kept free (and hence protected) by the
873 buddy allocator. Bigger value increase the probability
874 of catching random memory corruption, but reduce the
875 amount of memory for normal system use. The maximum
876 possible value is MAX_ORDER/2. Setting this parameter
877 to 1 or 2 should be enough to identify most random
878 memory corruption problems caused by bugs in kernel or
879 driver code when a CPU writes to (or reads from) a
880 random memory location. Note that there exists a class
881 of memory corruptions problems caused by buggy H/W or
882 F/W or by drivers badly programing DMA (basically when
883 memory is written at bus level and the CPU MMU is
884 bypassed) which are not detectable by
885 CONFIG_DEBUG_PAGEALLOC, hence this option will not help
886 tracking down these problems.
889 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this parameter
890 enables the feature at boot time. By default, it is
891 disabled and the system will work mostly the same as a
892 kernel built without CONFIG_DEBUG_PAGEALLOC.
893 Note: to get most of debug_pagealloc error reports, it's
894 useful to also enable the page_owner functionality.
895 on: enable the feature
897 debugfs= [KNL] This parameter enables what is exposed to userspace
898 and debugfs internal clients.
899 Format: { on, no-mount, off }
900 on: All functions are enabled.
902 Filesystem is not registered but kernel clients can
903 access APIs and a crashkernel can be used to read
904 its content. There is nothing to mount.
905 off: Filesystem is not registered and clients
906 get a -EPERM as result when trying to register files
907 or directories within debugfs.
908 This is equivalent of the runtime functionality if
909 debugfs was not enabled in the kernel at all.
910 Default value is set in build-time with a kernel configuration.
912 debugpat [X86] Enable PAT debugging
914 decnet.addr= [HW,NET]
915 Format: <area>[,<node>]
916 See also Documentation/networking/decnet.rst.
919 [HW] The size of the default HugeTLB page. This is
920 the size represented by the legacy /proc/ hugepages
921 APIs. In addition, this is the default hugetlb size
922 used for shmget(), mmap() and mounting hugetlbfs
923 filesystems. If not specified, defaults to the
924 architecture's default huge page size. Huge page
925 sizes are architecture dependent. See also
926 Documentation/admin-guide/mm/hugetlbpage.rst.
929 deferred_probe_timeout=
930 [KNL] Debugging option to set a timeout in seconds for
931 deferred probe to give up waiting on dependencies to
932 probe. Only specific dependencies (subsystems or
933 drivers) that have opted in will be ignored. A timeout of 0
934 will timeout at the end of initcalls. This option will also
935 dump out devices still on the deferred probe list after
939 Format: { on | off | def_only | inf_only | always }
940 on: s390 zlib hardware support for compression on
941 level 1 and decompression (default)
942 off: No s390 zlib hardware support
943 def_only: s390 zlib hardware support for deflate
944 only (compression on level 1)
945 inf_only: s390 zlib hardware support for inflate
947 always: Same as 'on' but ignores the selected compression
948 level always using hardware support (used for debugging)
951 Set number of hash buckets for dentry cache.
953 disable_1tb_segments [PPC]
954 Disables the use of 1TB hash page table segments. This
955 causes the kernel to fall back to 256MB segments which
956 can be useful when debugging issues that require an SLB
960 Limits the number of kernel SLB entries, and flushes
961 them frequently to increase the rate of SLB faults
965 See Documentation/networking/ipv6.rst.
968 [KNL] Under CONFIG_HARDENED_USERCOPY, whether
969 hardening is enabled for this boot. Hardened
970 usercopy checking is used to protect the kernel
971 from reading or writing beyond known memory
972 allocation boundaries as a proactive defense
973 against bounds-checking flaws in the kernel's
974 copy_to_user()/copy_from_user() interface.
975 on Perform hardened usercopy checks (default).
976 off Disable hardened usercopy checks.
979 Disable RADIX MMU mode on POWER9
981 radix_hcall_invalidate=on [PPC/PSERIES]
982 Disable RADIX GTSE feature and use hcall for TLB
986 Disable TLBIE instruction. Currently does not work
987 with KVM, with HASH MMU, or with coherent accelerators.
989 disable_cpu_apicid= [X86,APIC,SMP]
991 The number of initial APIC ID for the
992 corresponding CPU to be disabled at boot,
993 mostly used for the kdump 2nd kernel to
994 disable BSP to wake up multiple CPUs without
995 causing system reset or hang due to sending
998 disable_ddw [PPC/PSERIES]
999 Disable Dynamic DMA Window support. Use this
1000 to workaround buggy firmware.
1002 disable_ipv6= [IPV6]
1003 See Documentation/networking/ipv6.rst.
1005 disable_mtrr_cleanup [X86]
1006 The kernel tries to adjust MTRR layout from continuous
1007 to discrete, to make X server driver able to add WB
1008 entry later. This parameter disables that.
1010 disable_mtrr_trim [X86, Intel and AMD only]
1011 By default the kernel will trim any uncacheable
1012 memory out of your available memory pool based on
1013 MTRR settings. This parameter disables that behavior,
1014 possibly causing your machine to run very slowly.
1016 disable_timer_pin_1 [X86]
1017 Disable PIN 1 of APIC timer
1018 Can be useful to work around chipset bugs.
1020 dis_ucode_ldr [X86] Disable the microcode loader.
1022 dma_debug=off If the kernel is compiled with DMA_API_DEBUG support,
1023 this option disables the debugging code at boot.
1025 dma_debug_entries=<number>
1026 This option allows to tune the number of preallocated
1027 entries for DMA-API debugging code. One entry is
1028 required per DMA-API allocation. Use this if the
1029 DMA-API debugging code disables itself because the
1030 architectural default is too low.
1032 dma_debug_driver=<driver_name>
1033 With this option the DMA-API debugging driver
1034 filter feature can be enabled at boot time. Just
1035 pass the driver to filter for as the parameter.
1036 The filter can be disabled or changed to another
1037 driver later using sysfs.
1039 driver_async_probe= [KNL]
1040 List of driver names to be probed asynchronously.
1041 Format: <driver_name1>,<driver_name2>...
1043 drm.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>]
1044 Broken monitors, graphic adapters, KVMs and EDIDless
1045 panels may send no or incorrect EDID data sets.
1046 This parameter allows to specify an EDID data sets
1047 in the /lib/firmware directory that are used instead.
1048 Generic built-in EDID data sets are used, if one of
1049 edid/1024x768.bin, edid/1280x1024.bin,
1050 edid/1680x1050.bin, or edid/1920x1080.bin is given
1051 and no file with the same name exists. Details and
1052 instructions how to build your own EDID data are
1053 available in Documentation/admin-guide/edid.rst. An EDID
1054 data set will only be used for a particular connector,
1055 if its name and a colon are prepended to the EDID
1056 name. Each connector may use a unique EDID data
1057 set by separating the files with a comma. An EDID
1058 data set with no connector name will be used for
1059 any connectors not explicitly specified.
1064 Format: {"off" | "known"}
1065 Control how the dt_cpu_ftrs device-tree binding is
1066 used for CPU feature discovery and setup (if it
1068 off: Do not use it, fall back to legacy cpu table.
1069 known: Do not pass through unknown features to guests
1070 or userspace, only those that the kernel is aware of.
1072 dump_apple_properties [X86]
1073 Dump name and content of EFI device properties on
1074 x86 Macs. Useful for driver authors to determine
1075 what data is available or for reverse-engineering.
1077 dyndbg[="val"] [KNL,DYNAMIC_DEBUG]
1078 <module>.dyndbg[="val"]
1079 Enable debug messages at boot time. See
1080 Documentation/admin-guide/dynamic-debug-howto.rst
1083 nopku [X86] Disable Memory Protection Keys CPU feature found
1086 <module>.async_probe [KNL]
1087 Enable asynchronous probe on this module.
1089 early_ioremap_debug [KNL]
1090 Enable debug messages in early_ioremap support. This
1091 is useful for tracking down temporary early mappings
1092 which are not unmapped.
1094 earlycon= [KNL] Output early console device and options.
1096 When used with no options, the early console is
1097 determined by stdout-path property in device tree's
1098 chosen node or the ACPI SPCR table if supported by
1101 cdns,<addr>[,options]
1102 Start an early, polled-mode console on a Cadence
1103 (xuartps) serial port at the specified address. Only
1104 supported option is baud rate. If baud rate is not
1105 specified, the serial port must already be setup and
1108 uart[8250],io,<addr>[,options]
1109 uart[8250],mmio,<addr>[,options]
1110 uart[8250],mmio32,<addr>[,options]
1111 uart[8250],mmio32be,<addr>[,options]
1112 uart[8250],0x<addr>[,options]
1113 Start an early, polled-mode console on the 8250/16550
1114 UART at the specified I/O port or MMIO address.
1115 MMIO inter-register address stride is either 8-bit
1116 (mmio) or 32-bit (mmio32 or mmio32be).
1117 If none of [io|mmio|mmio32|mmio32be], <addr> is assumed
1118 to be equivalent to 'mmio'. 'options' are specified
1119 in the same format described for "console=ttyS<n>"; if
1120 unspecified, the h/w is not initialized.
1124 Start an early, polled-mode console on a pl011 serial
1125 port at the specified address. The pl011 serial port
1126 must already be setup and configured. Options are not
1127 yet supported. If 'mmio32' is specified, then only
1128 the driver will use only 32-bit accessors to read/write
1129 the device registers.
1132 Start an early console on a litex serial port at the
1133 specified address. The serial port must already be
1134 setup and configured. Options are not yet supported.
1137 Start an early, polled-mode console on a meson serial
1138 port at the specified address. The serial port must
1139 already be setup and configured. Options are not yet
1143 Start an early, polled-mode console on an msm serial
1144 port at the specified address. The serial port
1145 must already be setup and configured. Options are not
1148 msm_serial_dm,<addr>
1149 Start an early, polled-mode console on an msm serial
1150 dm port at the specified address. The serial port
1151 must already be setup and configured. Options are not
1155 Start an early, polled-mode console on a serial port
1156 of an Actions Semi SoC, such as S500 or S900, at the
1157 specified address. The serial port must already be
1158 setup and configured. Options are not yet supported.
1161 Start an early, polled-mode console on a serial port
1162 of an RDA Micro SoC, such as RDA8810PL, at the
1163 specified address. The serial port must already be
1164 setup and configured. Options are not yet supported.
1167 Use RISC-V SBI (Supervisor Binary Interface) for early
1170 smh Use ARM semihosting calls for early console.
1178 Use early console provided by serial driver available
1179 on Samsung SoCs, requires selecting proper type and
1180 a correct base address of the selected UART port. The
1181 serial port must already be setup and configured.
1182 Options are not yet supported.
1185 Start an early, polled-mode console on a lantiq serial
1186 (lqasc) port at the specified address. The serial port
1187 must already be setup and configured. Options are not
1192 Use early console provided by Freescale LP UART driver
1193 found on Freescale Vybrid and QorIQ LS1021A processors.
1194 A valid base address must be provided, and the serial
1195 port must already be setup and configured.
1199 Start an early, polled-mode, output-only console on the
1200 Freescale i.MX UART at the specified address. The UART
1201 must already be setup and configured.
1204 Start an early, polled-mode console on the
1205 Armada 3700 serial port at the specified
1206 address. The serial port must already be setup
1207 and configured. Options are not yet supported.
1210 Start an early, polled-mode console on a Qualcomm
1211 Generic Interface (GENI) based serial port at the
1212 specified address. The serial port must already be
1213 setup and configured. Options are not yet supported.
1216 Start an early, unaccelerated console on the EFI
1217 memory mapped framebuffer (if available). On cache
1218 coherent non-x86 systems that use system memory for
1219 the framebuffer, pass the 'ram' option so that it is
1220 mapped with the correct attributes.
1223 Use early console provided by Freescale LINFlexD UART
1224 serial driver for NXP S32V234 SoCs. A valid base
1225 address must be provided, and the serial port must
1226 already be setup and configured.
1228 earlyprintk= [X86,SH,ARM,M68k,S390]
1232 earlyprintk=serial[,ttySn[,baudrate]]
1233 earlyprintk=serial[,0x...[,baudrate]]
1234 earlyprintk=ttySn[,baudrate]
1235 earlyprintk=dbgp[debugController#]
1236 earlyprintk=pciserial[,force],bus:device.function[,baudrate]
1237 earlyprintk=xdbc[xhciController#]
1239 earlyprintk is useful when the kernel crashes before
1240 the normal console is initialized. It is not enabled by
1241 default because it has some cosmetic problems.
1243 Append ",keep" to not disable it when the real console
1246 Only one of vga, efi, serial, or usb debug port can
1249 Currently only ttyS0 and ttyS1 may be specified by
1250 name. Other I/O ports may be explicitly specified
1251 on some architectures (x86 and arm at least) by
1252 replacing ttySn with an I/O port address, like this:
1253 earlyprintk=serial,0x1008,115200
1254 You can find the port for a given device in
1255 /proc/tty/driver/serial:
1256 2: uart:ST16650V2 port:00001008 irq:18 ...
1258 Interaction with the standard serial driver is not
1261 The VGA and EFI output is eventually overwritten by
1264 The xen option can only be used in Xen domains.
1266 The sclp output can only be used on s390.
1268 The optional "force" to "pciserial" enables use of a
1269 PCI device even when its classcode is not of the
1272 edac_report= [HW,EDAC] Control how to report EDAC event
1273 Format: {"on" | "off" | "force"}
1274 on: enable EDAC to report H/W event. May be overridden
1275 by other higher priority error reporting module.
1276 off: disable H/W event reporting through EDAC.
1277 force: enforce the use of EDAC to report H/W event.
1280 ekgdboc= [X86,KGDB] Allow early kernel console debugging
1283 This is designed to be used in conjunction with
1284 the boot argument: earlyprintk=vga
1286 This parameter works in place of the kgdboc parameter
1287 but can only be used if the backing tty is available
1288 very early in the boot process. For early debugging
1289 via a serial port see kgdboc_earlycon instead.
1292 Format: {"off" | "on" | "skip[mbr]"}
1295 Format: { "debug", "disable_early_pci_dma",
1296 "nochunk", "noruntime", "nosoftreserve",
1297 "novamap", "no_disable_early_pci_dma" }
1298 debug: enable misc debug output.
1299 disable_early_pci_dma: disable the busmaster bit on all
1300 PCI bridges while in the EFI boot stub.
1301 nochunk: disable reading files in "chunks" in the EFI
1302 boot stub, as chunking can cause problems with some
1303 firmware implementations.
1304 noruntime : disable EFI runtime services support
1305 nosoftreserve: The EFI_MEMORY_SP (Specific Purpose)
1306 attribute may cause the kernel to reserve the
1307 memory range for a memory mapping driver to
1308 claim. Specify efi=nosoftreserve to disable this
1309 reservation and treat the memory by its base type
1310 (i.e. EFI_CONVENTIONAL_MEMORY / "System RAM").
1311 novamap: do not call SetVirtualAddressMap().
1312 no_disable_early_pci_dma: Leave the busmaster bit set
1313 on all PCI bridges while in the EFI boot stub
1315 efi_no_storage_paranoia [EFI; X86]
1316 Using this parameter you can use more than 50% of
1317 your efi variable storage. Use this parameter only if
1318 you are really sure that your UEFI does sane gc and
1319 fulfills the spec otherwise your board may brick.
1321 efi_fake_mem= nn[KMG]@ss[KMG]:aa[,nn[KMG]@ss[KMG]:aa,..] [EFI; X86]
1322 Add arbitrary attribute to specific memory range by
1323 updating original EFI memory map.
1324 Region of memory which aa attribute is added to is
1327 If efi_fake_mem=2G@4G:0x10000,2G@0x10a0000000:0x10000
1328 is specified, EFI_MEMORY_MORE_RELIABLE(0x10000)
1329 attribute is added to range 0x100000000-0x180000000 and
1330 0x10a0000000-0x1120000000.
1332 If efi_fake_mem=8G@9G:0x40000 is specified, the
1333 EFI_MEMORY_SP(0x40000) attribute is added to
1334 range 0x240000000-0x43fffffff.
1336 Using this parameter you can do debugging of EFI memmap
1337 related features. For example, you can do debugging of
1338 Address Range Mirroring feature even if your box
1339 doesn't support it, or mark specific memory as
1342 efivar_ssdt= [EFI; X86] Name of an EFI variable that contains an SSDT
1343 that is to be dynamically loaded by Linux. If there are
1344 multiple variables with the same name but with different
1345 vendor GUIDs, all of them will be loaded. See
1346 Documentation/admin-guide/acpi/ssdt-overlays.rst for details.
1349 eisa_irq_edge= [PARISC,HW]
1350 See header of drivers/parisc/eisa.c.
1353 See comment before function elanfreq_setup() in
1354 arch/x86/kernel/cpu/cpufreq/elanfreq.c.
1356 elfcorehdr=[size[KMG]@]offset[KMG] [IA64,PPC,SH,X86,S390]
1357 Specifies physical address of start of kernel core
1358 image elf header and optionally the size. Generally
1359 kexec loader will pass this option to capture kernel.
1360 See Documentation/admin-guide/kdump/kdump.rst for details.
1362 enable_mtrr_cleanup [X86]
1363 The kernel tries to adjust MTRR layout from continuous
1364 to discrete, to make X server driver able to add WB
1365 entry later. This parameter enables that.
1367 enable_timer_pin_1 [X86]
1368 Enable PIN 1 of APIC timer
1369 Can be useful to work around chipset bugs
1370 (in particular on some ATI chipsets).
1371 The kernel tries to set a reasonable default.
1373 enforcing [SELINUX] Set initial enforcing status.
1375 See security/selinux/Kconfig help text.
1376 0 -- permissive (log only, no denials).
1377 1 -- enforcing (deny and log).
1379 Value can be changed at runtime via
1380 /sys/fs/selinux/enforce.
1383 Disable Error Record Serialization Table (ERST)
1386 ether= [HW,NET] Ethernet cards parameters
1387 This option is obsoleted by the "netdev=" option, which
1388 has equivalent usage. See its documentation for details.
1392 Permit 'security.evm' to be updated regardless of
1393 current integrity status.
1398 fail_make_request=[KNL]
1399 General fault injection mechanism.
1400 Format: <interval>,<probability>,<space>,<times>
1401 See also Documentation/fault-injection/.
1404 Format: { initns | none }
1405 See Documentation/admin-guide/sysctl/net.rst for
1406 fb_tunnels_only_for_init_ns
1409 See Documentation/admin-guide/blockdev/floppy.rst.
1411 force_pal_cache_flush
1412 [IA-64] Avoid check_sal_cache_flush which may hang on
1413 buggy SAL_CACHE_FLUSH implementations. Using this
1414 parameter will force ia64_sal_cache_flush to call
1415 ia64_pal_cache_flush instead of SAL_CACHE_FLUSH.
1418 Forcefully enable Physical Address Extension (PAE).
1419 Many Pentium M systems disable PAE but may have a
1420 functionally usable PAE implementation.
1421 Warning: use of this parameter will taint the kernel
1422 and may cause unknown problems.
1425 [FTRACE] will set and start the specified tracer
1426 as early as possible in order to facilitate early
1429 ftrace_dump_on_oops[=orig_cpu]
1430 [FTRACE] will dump the trace buffers on oops.
1431 If no parameter is passed, ftrace will dump
1432 buffers of all CPUs, but if you pass orig_cpu, it will
1433 dump only the buffer of the CPU that triggered the
1436 ftrace_filter=[function-list]
1437 [FTRACE] Limit the functions traced by the function
1438 tracer at boot up. function-list is a comma-separated
1439 list of functions. This list can be changed at run
1440 time by the set_ftrace_filter file in the debugfs
1443 ftrace_notrace=[function-list]
1444 [FTRACE] Do not trace the functions specified in
1445 function-list. This list can be changed at run time
1446 by the set_ftrace_notrace file in the debugfs
1449 ftrace_graph_filter=[function-list]
1450 [FTRACE] Limit the top level callers functions traced
1451 by the function graph tracer at boot up.
1452 function-list is a comma-separated list of functions
1453 that can be changed at run time by the
1454 set_graph_function file in the debugfs tracing directory.
1456 ftrace_graph_notrace=[function-list]
1457 [FTRACE] Do not trace from the functions specified in
1458 function-list. This list is a comma-separated list of
1459 functions that can be changed at run time by the
1460 set_graph_notrace file in the debugfs tracing directory.
1462 ftrace_graph_max_depth=<uint>
1463 [FTRACE] Used with the function graph tracer. This is
1464 the max depth it will trace into a function. This value
1465 can be changed at run time by the max_graph_depth file
1466 in the tracefs tracing directory. default: 0 (no limit)
1468 fw_devlink= [KNL] Create device links between consumer and supplier
1469 devices by scanning the firmware to infer the
1470 consumer/supplier relationships. This feature is
1471 especially useful when drivers are loaded as modules as
1472 it ensures proper ordering of tasks like device probing
1473 (suppliers first, then consumers), supplier boot state
1474 clean up (only after all consumers have probed),
1475 suspend/resume & runtime PM (consumers first, then
1477 Format: { off | permissive | on | rpm }
1478 off -- Don't create device links from firmware info.
1479 permissive -- Create device links from firmware info
1480 but use it only for ordering boot state clean
1481 up (sync_state() calls).
1482 on -- Create device links from firmware info and use it
1483 to enforce probe and suspend/resume ordering.
1484 rpm -- Like "on", but also use to order runtime PM.
1486 fw_devlink.strict=<bool>
1487 [KNL] Treat all inferred dependencies as mandatory
1488 dependencies. This only applies for fw_devlink=on|rpm.
1492 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad
1493 support via parallel port (up to 5 devices per port)
1494 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5>
1495 See also Documentation/input/devices/joystick-parport.rst
1499 gart_fix_e820= [X86-64] disable the fix e820 for K8 GART
1503 gcov_persist= [GCOV] When non-zero (default), profiling data for
1504 kernel modules is saved and remains accessible via
1505 debugfs, even when the module is unloaded/reloaded.
1506 When zero, profiling data is discarded and associated
1507 debugfs files are removed at module unload time.
1509 goldfish [X86] Enable the goldfish android emulator platform.
1510 Don't use this when you are not running on the
1513 gpio-mockup.gpio_mockup_ranges
1514 [HW] Sets the ranges of gpiochip of for this device.
1515 Format: <start1>,<end1>,<start2>,<end2>...
1516 gpio-mockup.gpio_mockup_named_lines
1517 [HW] Let the driver know GPIO lines should be named.
1519 gpt [EFI] Forces disk with valid GPT signature but
1520 invalid Protective MBR to be treated as GPT. If the
1521 primary GPT is corrupted, it enables the backup/alternate
1522 GPT to be used instead.
1524 grcan.enable0= [HW] Configuration of physical interface 0. Determines
1525 the "Enable 0" bit of the configuration register.
1528 grcan.enable1= [HW] Configuration of physical interface 1. Determines
1529 the "Enable 0" bit of the configuration register.
1532 grcan.select= [HW] Select which physical interface to use.
1535 grcan.txsize= [HW] Sets the size of the tx buffer.
1536 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0.
1538 grcan.rxsize= [HW] Sets the size of the rx buffer.
1539 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0.
1542 hardlockup_all_cpu_backtrace=
1543 [KNL] Should the hard-lockup detector generate
1544 backtraces on all cpus.
1547 hashdist= [KNL,NUMA] Large hashes allocated during boot
1548 are distributed across NUMA nodes. Defaults on
1549 for 64-bit NUMA, off otherwise.
1550 Format: 0 | 1 (for off | on)
1552 hcl= [IA-64] SGI's Hardware Graph compatibility layer
1554 hd= [EIDE] (E)IDE hard drive subsystem geometry
1555 Format: <cyl>,<head>,<sect>
1558 Disable Hardware Error Source Table (HEST) support;
1559 corresponding firmware-first mode error processing
1560 logic will be disabled.
1562 highmem=nn[KMG] [KNL,BOOT] forces the highmem zone to have an exact
1563 size of <nn>. This works even on boxes that have no
1564 highmem otherwise. This also works to reduce highmem
1565 size on bigger boxes.
1567 highres= [KNL] Enable/disable high resolution timer mode.
1568 Valid parameters: "on", "off"
1573 hpet= [X86-32,HPET] option to control HPET usage
1574 Format: { enable (default) | disable | force |
1576 disable: disable HPET and use PIT instead
1577 force: allow force enabled of undocumented chips (ICH4,
1579 verbose: show contents of HPET registers during setup
1581 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET
1582 registers. Default set by CONFIG_HPET_MMAP_DEFAULT.
1584 hugetlb_cma= [HW,CMA] The size of a CMA area used for allocation
1585 of gigantic hugepages. Or using node format, the size
1586 of a CMA area per node can be specified.
1587 Format: nn[KMGTPE] or (node format)
1588 <node>:nn[KMGTPE][,<node>:nn[KMGTPE]]
1590 Reserve a CMA area of given size and allocate gigantic
1591 hugepages using the CMA allocator. If enabled, the
1592 boot-time allocation of gigantic hugepages is skipped.
1594 hugepages= [HW] Number of HugeTLB pages to allocate at boot.
1595 If this follows hugepagesz (below), it specifies
1596 the number of pages of hugepagesz to be allocated.
1597 If this is the first HugeTLB parameter on the command
1598 line, it specifies the number of pages to allocate for
1599 the default huge page size. If using node format, the
1600 number of pages to allocate per-node can be specified.
1601 See also Documentation/admin-guide/mm/hugetlbpage.rst.
1602 Format: <integer> or (node format)
1603 <node>:<integer>[,<node>:<integer>]
1606 [HW] The size of the HugeTLB pages. This is used in
1607 conjunction with hugepages (above) to allocate huge
1608 pages of a specific size at boot. The pair
1609 hugepagesz=X hugepages=Y can be specified once for
1610 each supported huge page size. Huge page sizes are
1611 architecture dependent. See also
1612 Documentation/admin-guide/mm/hugetlbpage.rst.
1615 hugetlb_free_vmemmap=
1616 [KNL] Reguires CONFIG_HUGETLB_PAGE_FREE_VMEMMAP
1618 Allows heavy hugetlb users to free up some more
1619 memory (6 * PAGE_SIZE for each 2MB hugetlb page).
1620 Format: { on | off (default) }
1622 on: enable the feature
1623 off: disable the feature
1625 Built with CONFIG_HUGETLB_PAGE_FREE_VMEMMAP_DEFAULT_ON=y,
1628 This is not compatible with memory_hotplug.memmap_on_memory.
1629 If both parameters are enabled, hugetlb_free_vmemmap takes
1630 precedence over memory_hotplug.memmap_on_memory.
1633 [KNL] Should the hung task detector generate panics.
1636 A value of 1 instructs the kernel to panic when a
1637 hung task is detected. The default value is controlled
1638 by the CONFIG_BOOTPARAM_HUNG_TASK_PANIC build-time
1639 option. The value selected by this boot parameter can
1640 be changed later by the kernel.hung_task_panic sysctl.
1642 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC)
1643 terminal devices. Valid values: 0..8
1644 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs.
1645 If specified, z/VM IUCV HVC accepts connections
1646 from listed z/VM user IDs only.
1648 hv_nopvspin [X86,HYPER_V] Disables the paravirt spinlock optimizations
1649 which allow the hypervisor to 'idle' the
1650 guest on lock contention.
1653 Do not unregister boot console at start. This is only
1654 useful for debugging when something happens in the window
1655 between unregistering the boot console and initializing
1658 i2c_bus= [HW] Override the default board specific I2C bus speed
1659 or register an additional I2C bus that is not
1660 registered from board initialization code.
1664 i8042.debug [HW] Toggle i8042 debug mode
1665 i8042.unmask_kbd_data
1666 [HW] Enable printing of interrupt data from the KBD port
1667 (disabled by default, and as a pre-condition
1668 requires that i8042.debug=1 be enabled)
1669 i8042.direct [HW] Put keyboard port into non-translated mode
1670 i8042.dumbkbd [HW] Pretend that controller can only read data from
1671 keyboard and cannot control its state
1672 (Don't attempt to blink the leds)
1673 i8042.noaux [HW] Don't check for auxiliary (== mouse) port
1674 i8042.nokbd [HW] Don't check/create keyboard port
1675 i8042.noloop [HW] Disable the AUX Loopback command while probing
1677 i8042.nomux [HW] Don't check presence of an active multiplexing
1679 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX
1681 i8042.notimeout [HW] Ignore timeout condition signalled by controller
1682 i8042.reset [HW] Reset the controller during init, cleanup and
1683 suspend-to-ram transitions, only during s2r
1684 transitions, or never reset
1685 Format: { 1 | Y | y | 0 | N | n }
1686 1, Y, y: always reset controller
1687 0, N, n: don't ever reset controller
1688 Default: only on s2r transitions on x86; most other
1689 architectures force reset to be always executed
1690 i8042.unlock [HW] Unlock (ignore) the keylock
1691 i8042.kbdreset [HW] Reset device connected to KBD port
1693 [HW] Allow deferred probing upon i8042 probe errors
1697 i8k.ignore_dmi [HW] Continue probing hardware even if DMI data
1698 indicates that the driver is running on unsupported
1700 i8k.force [HW] Activate i8k driver even if SMM BIOS signature
1701 does not match list of supported models.
1703 [HW] Report power status in /proc/i8k
1704 (disabled by default)
1705 i8k.restricted [HW] Allow controlling fans only if SYS_ADMIN
1708 i915.invert_brightness=
1709 [DRM] Invert the sense of the variable that is used to
1710 set the brightness of the panel backlight. Normally a
1711 brightness value of 0 indicates backlight switched off,
1712 and the maximum of the brightness value sets the backlight
1713 to maximum brightness. If this parameter is set to 0
1714 (default) and the machine requires it, or this parameter
1715 is set to 1, a brightness value of 0 sets the backlight
1716 to maximum brightness, and the maximum of the brightness
1717 value switches the backlight off.
1718 -1 -- never invert brightness
1719 0 -- machine default
1720 1 -- force brightness inversion
1723 Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]]
1725 ide-core.nodma= [HW] (E)IDE subsystem
1726 Format: =0.0 to prevent dma on hda, =0.1 hdb =1.0 hdc
1727 .vlb_clock .pci_clock .noflush .nohpa .noprobe .nowerr
1728 .cdrom .chs .ignore_cable are additional options
1729 See Documentation/ide/ide.rst.
1731 ide-generic.probe-mask= [HW] (E)IDE subsystem
1733 Probe mask for legacy ISA IDE ports. Depending on
1734 platform up to 6 ports are supported, enabled by
1735 setting corresponding bits in the mask to 1. The
1736 default value is 0x0, which has a special meaning.
1737 On systems that have PCI, it triggers scanning the
1738 PCI bus for the first and the second port, which
1739 are then probed. On systems without PCI the value
1740 of 0x0 enables probing the two first ports as if it
1743 ide-pci-generic.all-generic-ide [HW] (E)IDE subsystem
1744 Claim all unknown PCI IDE storage controllers.
1747 Format: idle=poll, idle=halt, idle=nomwait
1748 Poll forces a polling idle loop that can slightly
1749 improve the performance of waking up a idle CPU, but
1750 will use a lot of power and make the system run hot.
1752 idle=halt: Halt is forced to be used for CPU idle.
1753 In such case C2/C3 won't be used again.
1754 idle=nomwait: Disable mwait for CPU C-states
1758 Allow force disabling of Shared Virtual Memory (SVA)
1759 support for the idxd driver. By default it is set to
1762 idxd.tc_override= [HW]
1764 Allow override of default traffic class configuration
1765 for the device. By default it is set to false (0).
1767 ieee754= [MIPS] Select IEEE Std 754 conformance mode
1768 Format: { strict | legacy | 2008 | relaxed }
1771 Choose which programs will be accepted for execution
1772 based on the IEEE 754 NaN encoding(s) supported by
1773 the FPU and the NaN encoding requested with the value
1774 of an ELF file header flag individually set by each
1775 binary. Hardware implementations are permitted to
1776 support either or both of the legacy and the 2008 NaN
1779 Available settings are as follows:
1780 strict accept binaries that request a NaN encoding
1781 supported by the FPU
1782 legacy only accept legacy-NaN binaries, if supported
1784 2008 only accept 2008-NaN binaries, if supported
1786 relaxed accept any binaries regardless of whether
1787 supported by the FPU
1789 The FPU emulator is always able to support both NaN
1790 encodings, so if no FPU hardware is present or it has
1791 been disabled with 'nofpu', then the settings of
1792 'legacy' and '2008' strap the emulator accordingly,
1793 'relaxed' straps the emulator for both legacy-NaN and
1794 2008-NaN, whereas 'strict' enables legacy-NaN only on
1795 legacy processors and both NaN encodings on MIPS32 or
1798 The setting for ABS.fmt/NEG.fmt instruction execution
1799 mode generally follows that for the NaN encoding,
1800 except where unsupported by hardware.
1802 ignore_loglevel [KNL]
1803 Ignore loglevel setting - this will print /all/
1804 kernel messages to the console. Useful for debugging.
1805 We also add it as printk module parameter, so users
1806 could change it dynamically, usually by
1807 /sys/module/printk/parameters/ignore_loglevel.
1810 Ignore RLIMIT_DATA setting for data mappings,
1811 print warning at first misuse. Can be changed via
1812 /sys/module/kernel/parameters/ignore_rlimit_data.
1814 ihash_entries= [KNL]
1815 Set number of hash buckets for inode cache.
1817 ima_appraise= [IMA] appraise integrity measurements
1818 Format: { "off" | "enforce" | "fix" | "log" }
1821 ima_appraise_tcb [IMA] Deprecated. Use ima_policy= instead.
1822 The builtin appraise policy appraises all files
1825 ima_canonical_fmt [IMA]
1826 Use the canonical format for the binary runtime
1827 measurements, instead of host native format.
1830 Format: { md5 | sha1 | rmd160 | sha256 | sha384
1834 The list of supported hash algorithms is defined
1835 in crypto/hash_info.h.
1838 The builtin policies to load during IMA setup.
1839 Format: "tcb | appraise_tcb | secure_boot |
1840 fail_securely | critical_data"
1842 The "tcb" policy measures all programs exec'd, files
1843 mmap'd for exec, and all files opened with the read
1844 mode bit set by either the effective uid (euid=0) or
1847 The "appraise_tcb" policy appraises the integrity of
1848 all files owned by root.
1850 The "secure_boot" policy appraises the integrity
1851 of files (eg. kexec kernel image, kernel modules,
1852 firmware, policy, etc) based on file signatures.
1854 The "fail_securely" policy forces file signature
1855 verification failure also on privileged mounted
1856 filesystems with the SB_I_UNVERIFIABLE_SIGNATURE
1859 The "critical_data" policy measures kernel integrity
1862 ima_tcb [IMA] Deprecated. Use ima_policy= instead.
1863 Load a policy which meets the needs of the Trusted
1864 Computing Base. This means IMA will measure all
1865 programs exec'd, files mmap'd for exec, and all files
1866 opened for read by uid=0.
1869 Select one of defined IMA measurements template formats.
1870 Formats: { "ima" | "ima-ng" | "ima-sig" }
1874 [IMA] Define a custom template format.
1875 Format: { "field1|...|fieldN" }
1877 ima.ahash_minsize= [IMA] Minimum file size for asynchronous hash usage
1878 Format: <min_file_size>
1879 Set the minimal file size for using asynchronous hash.
1880 If left unspecified, ahash usage is disabled.
1882 ahash performance varies for different data sizes on
1883 different crypto accelerators. This option can be used
1884 to achieve the best performance for a particular HW.
1886 ima.ahash_bufsize= [IMA] Asynchronous hash buffer size
1888 Set hashing buffer size. Default: 4k.
1890 ahash performance varies for different chunk sizes on
1891 different crypto accelerators. This option can be used
1892 to achieve best performance for particular HW.
1896 Run specified binary instead of /sbin/init as init
1899 initcall_debug [KNL] Trace initcalls as they are executed. Useful
1900 for working out where the kernel is dying during
1903 initcall_blacklist= [KNL] Do not execute a comma-separated list of
1904 initcall functions. Useful for debugging built-in
1905 modules and initcalls.
1907 initramfs_async= [KNL]
1910 This parameter controls whether the initramfs
1911 image is unpacked asynchronously, concurrently
1912 with devices being probed and
1913 initialized. This should normally just work,
1914 but as a debugging aid, one can get the
1915 historical behaviour of the initramfs
1916 unpacking being completed before device_ and
1919 initrd= [BOOT] Specify the location of the initial ramdisk
1921 initrdmem= [KNL] Specify a physical address and size from which to
1922 load the initrd. If an initrd is compiled in or
1923 specified in the bootparams, it takes priority over this
1925 Format: ss[KMG],nn[KMG]
1928 init_on_alloc= [MM] Fill newly allocated pages and heap objects with
1931 Default set by CONFIG_INIT_ON_ALLOC_DEFAULT_ON.
1933 init_on_free= [MM] Fill freed pages and heap objects with zeroes.
1935 Default set by CONFIG_INIT_ON_FREE_DEFAULT_ON.
1937 init_pkru= [X86] Specify the default memory protection keys rights
1938 register contents for all processes. 0x55555554 by
1939 default (disallow access to all but pkey 0). Can
1940 override in debugfs after boot.
1942 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver
1945 int_pln_enable [X86] Enable power limit notification interrupt
1947 integrity_audit=[IMA]
1948 Format: { "0" | "1" }
1949 0 -- basic integrity auditing messages. (Default)
1950 1 -- additional integrity auditing messages.
1952 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option
1954 Enable intel iommu driver.
1956 Disable intel iommu driver.
1957 igfx_off [Default Off]
1958 By default, gfx is mapped as normal device. If a gfx
1959 device has a dedicated DMAR unit, the DMAR unit is
1960 bypassed by not enabling DMAR with this option. In
1961 this case, gfx device will use physical address for
1963 strict [Default Off]
1964 Deprecated, equivalent to iommu.strict=1.
1965 sp_off [Default Off]
1966 By default, super page will be supported if Intel IOMMU
1967 has the capability. With this option, super page will
1970 Enable the Intel IOMMU scalable mode if the hardware
1971 advertises that it has support for the scalable mode
1974 Disallow use of the Intel IOMMU scalable mode.
1975 tboot_noforce [Default Off]
1976 Do not force the Intel IOMMU enabled under tboot.
1977 By default, tboot will force Intel IOMMU on, which
1978 could harm performance of some high-throughput
1979 devices like 40GBit network cards, even if identity
1981 Note that using this option lowers the security
1982 provided by tboot because it makes the system
1983 vulnerable to DMA attacks.
1985 intel_idle.max_cstate= [KNL,HW,ACPI,X86]
1986 0 disables intel_idle and fall back on acpi_idle.
1987 1 to 9 specify maximum depth of C-state.
1991 Do not enable intel_pstate as the default
1992 scaling driver for the supported processors
1994 Use intel_pstate as a scaling driver, but configure it
1995 to work with generic cpufreq governors (instead of
1996 enabling its internal governor). This mode cannot be
1997 used along with the hardware-managed P-states (HWP)
2000 Enable intel_pstate on systems that prohibit it by default
2001 in favor of acpi-cpufreq. Forcing the intel_pstate driver
2002 instead of acpi-cpufreq may disable platform features, such
2003 as thermal controls and power capping, that rely on ACPI
2004 P-States information being indicated to OSPM and therefore
2005 should be used with caution. This option does not work with
2006 processors that aren't supported by the intel_pstate driver
2007 or on platforms that use pcc-cpufreq instead of acpi-cpufreq.
2009 Do not enable hardware P state control (HWP)
2012 Only load intel_pstate on systems which support
2013 hardware P state control (HWP) if available.
2015 Enforce ACPI _PPC performance limits. If the Fixed ACPI
2016 Description Table, specifies preferred power management
2017 profile as "Enterprise Server" or "Performance Server",
2018 then this feature is turned on by default.
2020 Allow per-logical-CPU P-State performance control limits using
2021 cpufreq sysfs interface
2023 intremap= [X86-64, Intel-IOMMU]
2024 on enable Interrupt Remapping (default)
2025 off disable Interrupt Remapping
2026 nosid disable Source ID checking
2028 BIOS x2APIC opt-out request will be ignored
2029 nopost disable Interrupt Posting
2031 iomem= Disable strict checking of access to MMIO memory
2032 strict regions from userspace.
2047 nobypass [PPC/POWERNV]
2048 Disable IOMMU bypass, using IOMMU for PCI devices.
2050 iommu.forcedac= [ARM64, X86] Control IOVA allocation for PCI devices.
2051 Format: { "0" | "1" }
2052 0 - Try to allocate a 32-bit DMA address first, before
2053 falling back to the full range if needed.
2054 1 - Allocate directly from the full usable range,
2055 forcing Dual Address Cycle for PCI cards supporting
2056 greater than 32-bit addressing.
2058 iommu.strict= [ARM64, X86] Configure TLB invalidation behaviour
2059 Format: { "0" | "1" }
2061 Request that DMA unmap operations use deferred
2062 invalidation of hardware TLBs, for increased
2063 throughput at the cost of reduced device isolation.
2064 Will fall back to strict mode if not supported by
2065 the relevant IOMMU driver.
2067 DMA unmap operations invalidate IOMMU hardware TLBs
2069 unset - Use value of CONFIG_IOMMU_DEFAULT_DMA_{LAZY,STRICT}.
2070 Note: on x86, strict mode specified via one of the
2071 legacy driver-specific options takes precedence.
2074 [ARM64, X86] Configure DMA to bypass the IOMMU by default.
2075 Format: { "0" | "1" }
2076 0 - Use IOMMU translation for DMA.
2077 1 - Bypass the IOMMU for DMA.
2078 unset - Use value of CONFIG_IOMMU_DEFAULT_PASSTHROUGH.
2080 io7= [HW] IO7 for Marvel-based Alpha systems
2081 See comment before marvel_specify_io7 in
2082 arch/alpha/kernel/core_marvel.c.
2084 io_delay= [X86] I/O delay method
2086 Standard port 0x80 based delay
2088 Alternate port 0xed based delay (needed on some systems)
2090 Simple two microseconds delay
2095 See Documentation/admin-guide/nfs/nfsroot.rst.
2097 ipcmni_extend [KNL] Extend the maximum number of unique System V
2098 IPC identifiers from 32,768 to 16,777,216.
2100 irqaffinity= [SMP] Set the default irq affinity mask
2101 The argument is a cpu list, as described above.
2103 irqchip.gicv2_force_probe=
2106 Force the kernel to look for the second 4kB page
2107 of a GICv2 controller even if the memory range
2108 exposed by the device tree is too small.
2110 irqchip.gicv3_nolpi=
2112 Force the kernel to ignore the availability of
2113 LPIs (and by consequence ITSs). Intended for system
2114 that use the kernel as a bootloader, and thus want
2115 to let secondary kernels in charge of setting up
2118 irqchip.gicv3_pseudo_nmi= [ARM64]
2119 Enables support for pseudo-NMIs in the kernel. This
2120 requires the kernel to be built with
2121 CONFIG_ARM64_PSEUDO_NMI.
2124 When an interrupt is not handled search all handlers
2125 for it. Intended to get systems with badly broken
2129 When an interrupt is not handled search all handlers
2130 for it. Also check all handlers each timer
2131 interrupt. Intended to get systems with badly broken
2135 Format: <RDP>,<reset>,<pci_scan>,<verbosity>
2137 isolcpus= [KNL,SMP,ISOL] Isolate a given set of CPUs from disturbance.
2138 [Deprecated - use cpusets instead]
2139 Format: [flag-list,]<cpu-list>
2141 Specify one or more CPUs to isolate from disturbances
2142 specified in the flag list (default: domain):
2145 Disable the tick when a single task runs.
2147 A residual 1Hz tick is offloaded to workqueues, which you
2148 need to affine to housekeeping through the global
2149 workqueue's affinity configured via the
2150 /sys/devices/virtual/workqueue/cpumask sysfs file, or
2151 by using the 'domain' flag described below.
2153 NOTE: by default the global workqueue runs on all CPUs,
2154 so to protect individual CPUs the 'cpumask' file has to
2155 be configured manually after bootup.
2158 Isolate from the general SMP balancing and scheduling
2159 algorithms. Note that performing domain isolation this way
2160 is irreversible: it's not possible to bring back a CPU to
2161 the domains once isolated through isolcpus. It's strongly
2162 advised to use cpusets instead to disable scheduler load
2163 balancing through the "cpuset.sched_load_balance" file.
2164 It offers a much more flexible interface where CPUs can
2165 move in and out of an isolated set anytime.
2167 You can move a process onto or off an "isolated" CPU via
2168 the CPU affinity syscalls or cpuset.
2169 <cpu number> begins at 0 and the maximum value is
2170 "number of CPUs in system - 1".
2174 Isolate from being targeted by managed interrupts
2175 which have an interrupt mask containing isolated
2176 CPUs. The affinity of managed interrupts is
2177 handled by the kernel and cannot be changed via
2178 the /proc/irq/* interfaces.
2180 This isolation is best effort and only effective
2181 if the automatically assigned interrupt mask of a
2182 device queue contains isolated and housekeeping
2183 CPUs. If housekeeping CPUs are online then such
2184 interrupts are directed to the housekeeping CPU
2185 so that IO submitted on the housekeeping CPU
2186 cannot disturb the isolated CPU.
2188 If a queue's affinity mask contains only isolated
2189 CPUs then this parameter has no effect on the
2190 interrupt routing decision, though interrupts are
2191 only delivered when tasks running on those
2192 isolated CPUs submit IO. IO submitted on
2193 housekeeping CPUs has no influence on those
2196 The format of <cpu-list> is described above.
2200 ivrs_ioapic [HW,X86-64]
2201 Provide an override to the IOAPIC-ID<->DEVICE-ID
2202 mapping provided in the IVRS ACPI table. For
2203 example, to map IOAPIC-ID decimal 10 to
2204 PCI device 00:14.0 write the parameter as:
2205 ivrs_ioapic[10]=00:14.0
2207 ivrs_hpet [HW,X86-64]
2208 Provide an override to the HPET-ID<->DEVICE-ID
2209 mapping provided in the IVRS ACPI table. For
2210 example, to map HPET-ID decimal 0 to
2211 PCI device 00:14.0 write the parameter as:
2212 ivrs_hpet[0]=00:14.0
2214 ivrs_acpihid [HW,X86-64]
2215 Provide an override to the ACPI-HID:UID<->DEVICE-ID
2216 mapping provided in the IVRS ACPI table. For
2217 example, to map UART-HID:UID AMD0020:0 to
2218 PCI device 00:14.5 write the parameter as:
2219 ivrs_acpihid[00:14.5]=AMD0020:0
2221 js= [HW,JOY] Analog joystick
2222 See Documentation/input/joydev/joystick.rst.
2225 When CONFIG_RANDOMIZE_BASE is set, this disables
2226 kernel and module base offset ASLR (Address Space
2227 Layout Randomization).
2230 [KNL] Enforce KASAN (Kernel Address Sanitizer) to print
2231 report on every invalid memory access. Without this
2232 parameter KASAN will print report only for the first
2237 kernelcore= [KNL,X86,IA-64,PPC]
2238 Format: nn[KMGTPE] | nn% | "mirror"
2239 This parameter specifies the amount of memory usable by
2240 the kernel for non-movable allocations. The requested
2241 amount is spread evenly throughout all nodes in the
2242 system as ZONE_NORMAL. The remaining memory is used for
2243 movable memory in its own zone, ZONE_MOVABLE. In the
2244 event, a node is too small to have both ZONE_NORMAL and
2245 ZONE_MOVABLE, kernelcore memory will take priority and
2246 other nodes will have a larger ZONE_MOVABLE.
2248 ZONE_MOVABLE is used for the allocation of pages that
2249 may be reclaimed or moved by the page migration
2250 subsystem. Note that allocations like PTEs-from-HighMem
2251 still use the HighMem zone if it exists, and the Normal
2252 zone if it does not.
2254 It is possible to specify the exact amount of memory in
2255 the form of "nn[KMGTPE]", a percentage of total system
2256 memory in the form of "nn%", or "mirror". If "mirror"
2257 option is specified, mirrored (reliable) memory is used
2258 for non-movable allocations and remaining memory is used
2259 for Movable pages. "nn[KMGTPE]", "nn%", and "mirror"
2260 are exclusive, so you cannot specify multiple forms.
2262 kgdbdbgp= [KGDB,HW] kgdb over EHCI usb debug port.
2263 Format: <Controller#>[,poll interval]
2264 The controller # is the number of the ehci usb debug
2265 port as it is probed via PCI. The poll interval is
2266 optional and is the number seconds in between
2267 each poll cycle to the debug port in case you need
2268 the functionality for interrupting the kernel with
2269 gdb or control-c on the dbgp connection. When
2270 not using this parameter you use sysrq-g to break into
2271 the kernel debugger.
2273 kgdboc= [KGDB,HW] kgdb over consoles.
2274 Requires a tty driver that supports console polling,
2275 or a supported polling keyboard driver (non-usb).
2276 Serial only format: <serial_device>[,baud]
2277 keyboard only format: kbd
2278 keyboard and serial format: kbd,<serial_device>[,baud]
2279 Optional Kernel mode setting:
2280 kms, kbd format: kms,kbd
2281 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud]
2283 kgdboc_earlycon= [KGDB,HW]
2284 If the boot console provides the ability to read
2285 characters and can work in polling mode, you can use
2286 this parameter to tell kgdb to use it as a backend
2287 until the normal console is registered. Intended to
2288 be used together with the kgdboc parameter which
2289 specifies the normal console to transition to.
2291 The name of the early console should be specified
2292 as the value of this parameter. Note that the name of
2293 the early console might be different than the tty
2294 name passed to kgdboc. It's OK to leave the value
2295 blank and the first boot console that implements
2296 read() will be picked.
2298 kgdbwait [KGDB] Stop kernel execution and enter the
2299 kernel debugger at the earliest opportunity.
2301 kmac= [MIPS] Korina ethernet MAC address.
2302 Configure the RouterBoard 532 series on-chip
2303 Ethernet adapter MAC address.
2305 kmemleak= [KNL] Boot-time kmemleak enable/disable
2306 Valid arguments: on, off
2308 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y,
2311 kprobe_event=[probe-list]
2312 [FTRACE] Add kprobe events and enable at boot time.
2313 The probe-list is a semicolon delimited list of probe
2314 definitions. Each definition is same as kprobe_events
2315 interface, but the parameters are comma delimited.
2316 For example, to add a kprobe event on vfs_read with
2317 arg1 and arg2, add to the command line;
2319 kprobe_event=p,vfs_read,$arg1,$arg2
2321 See also Documentation/trace/kprobetrace.rst "Kernel
2322 Boot Parameter" section.
2324 kpti= [ARM64] Control page table isolation of user
2325 and kernel address spaces.
2326 Default: enabled on cores which need mitigation.
2330 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs.
2331 Default is 0 (don't ignore, but inject #GP)
2333 kvm.enable_vmware_backdoor=[KVM] Support VMware backdoor PV interface.
2334 Default is false (don't support).
2336 kvm.mmu_audit= [KVM] This is a R/W parameter which allows audit
2341 [KVM] Controls the software workaround for the
2342 X86_BUG_ITLB_MULTIHIT bug.
2343 force : Always deploy workaround.
2344 off : Never deploy workaround.
2345 auto : Deploy workaround based on the presence of
2346 X86_BUG_ITLB_MULTIHIT.
2350 If the software workaround is enabled for the host,
2351 guests do need not to enable it for nested guests.
2353 kvm.nx_huge_pages_recovery_ratio=
2354 [KVM] Controls how many 4KiB pages are periodically zapped
2355 back to huge pages. 0 disables the recovery, otherwise if
2356 the value is N KVM will zap 1/Nth of the 4KiB pages every
2357 period (see below). The default is 60.
2359 kvm.nx_huge_pages_recovery_period_ms=
2360 [KVM] Controls the time period at which KVM zaps 4KiB pages
2361 back to huge pages. If the value is a non-zero N, KVM will
2362 zap a portion (see ratio above) of the pages every N msecs.
2363 If the value is 0 (the default), KVM will pick a period based
2364 on the ratio, such that a page is zapped after 1 hour on average.
2366 kvm-amd.nested= [KVM,AMD] Allow nested virtualization in KVM/SVM.
2367 Default is 1 (enabled)
2369 kvm-amd.npt= [KVM,AMD] Disable nested paging (virtualized MMU)
2371 Default is 1 (enabled) if in 64-bit or 32-bit PAE mode.
2374 [KVM,ARM] Select one of KVM/arm64's modes of operation.
2376 none: Forcefully disable KVM.
2378 nvhe: Standard nVHE-based mode, without support for
2381 protected: nVHE-based mode with support for guests whose
2382 state is kept private from the host.
2383 Not valid if the kernel is running in EL2.
2385 Defaults to VHE/nVHE based on hardware support. Setting
2386 mode to "protected" will disable kexec and hibernation
2389 kvm-arm.vgic_v3_group0_trap=
2390 [KVM,ARM] Trap guest accesses to GICv3 group-0
2393 kvm-arm.vgic_v3_group1_trap=
2394 [KVM,ARM] Trap guest accesses to GICv3 group-1
2397 kvm-arm.vgic_v3_common_trap=
2398 [KVM,ARM] Trap guest accesses to GICv3 common
2401 kvm-arm.vgic_v4_enable=
2402 [KVM,ARM] Allow use of GICv4 for direct injection of
2405 kvm_cma_resv_ratio=n [PPC]
2406 Reserves given percentage from system memory area for
2407 contiguous memory allocation for KVM hash pagetable
2409 By default it reserves 5% of total system memory.
2413 kvm-intel.ept= [KVM,Intel] Disable extended page tables
2414 (virtualized MMU) support on capable Intel chips.
2415 Default is 1 (enabled)
2417 kvm-intel.emulate_invalid_guest_state=
2418 [KVM,Intel] Disable emulation of invalid guest state.
2419 Ignored if kvm-intel.enable_unrestricted_guest=1, as
2420 guest state is never invalid for unrestricted guests.
2421 This param doesn't apply to nested guests (L2), as KVM
2422 never emulates invalid L2 guest state.
2423 Default is 1 (enabled)
2425 kvm-intel.flexpriority=
2426 [KVM,Intel] Disable FlexPriority feature (TPR shadow).
2427 Default is 1 (enabled)
2430 [KVM,Intel] Enable VMX nesting (nVMX).
2431 Default is 0 (disabled)
2433 kvm-intel.unrestricted_guest=
2434 [KVM,Intel] Disable unrestricted guest feature
2435 (virtualized real and unpaged mode) on capable
2436 Intel chips. Default is 1 (enabled)
2438 kvm-intel.vmentry_l1d_flush=[KVM,Intel] Mitigation for L1 Terminal Fault
2441 Valid arguments: never, cond, always
2443 always: L1D cache flush on every VMENTER.
2444 cond: Flush L1D on VMENTER only when the code between
2445 VMEXIT and VMENTER can leak host memory.
2446 never: Disables the mitigation
2448 Default is cond (do L1 cache flush in specific instances)
2450 kvm-intel.vpid= [KVM,Intel] Disable Virtual Processor Identification
2451 feature (tagged TLBs) on capable Intel chips.
2452 Default is 1 (enabled)
2454 l1d_flush= [X86,INTEL]
2455 Control mitigation for L1D based snooping vulnerability.
2457 Certain CPUs are vulnerable to an exploit against CPU
2458 internal buffers which can forward information to a
2459 disclosure gadget under certain conditions.
2461 In vulnerable processors, the speculatively
2462 forwarded data can be used in a cache side channel
2463 attack, to access data to which the attacker does
2464 not have direct access.
2466 This parameter controls the mitigation. The
2469 on - enable the interface for the mitigation
2471 l1tf= [X86] Control mitigation of the L1TF vulnerability on
2474 The kernel PTE inversion protection is unconditionally
2475 enabled and cannot be disabled.
2478 Provides all available mitigations for the
2479 L1TF vulnerability. Disables SMT and
2480 enables all mitigations in the
2481 hypervisors, i.e. unconditional L1D flush.
2483 SMT control and L1D flush control via the
2484 sysfs interface is still possible after
2485 boot. Hypervisors will issue a warning
2486 when the first VM is started in a
2487 potentially insecure configuration,
2488 i.e. SMT enabled or L1D flush disabled.
2491 Same as 'full', but disables SMT and L1D
2492 flush runtime control. Implies the
2493 'nosmt=force' command line option.
2494 (i.e. sysfs control of SMT is disabled.)
2497 Leaves SMT enabled and enables the default
2498 hypervisor mitigation, i.e. conditional
2501 SMT control and L1D flush control via the
2502 sysfs interface is still possible after
2503 boot. Hypervisors will issue a warning
2504 when the first VM is started in a
2505 potentially insecure configuration,
2506 i.e. SMT enabled or L1D flush disabled.
2510 Disables SMT and enables the default
2511 hypervisor mitigation.
2513 SMT control and L1D flush control via the
2514 sysfs interface is still possible after
2515 boot. Hypervisors will issue a warning
2516 when the first VM is started in a
2517 potentially insecure configuration,
2518 i.e. SMT enabled or L1D flush disabled.
2521 Same as 'flush', but hypervisors will not
2522 warn when a VM is started in a potentially
2523 insecure configuration.
2526 Disables hypervisor mitigations and doesn't
2528 It also drops the swap size and available
2529 RAM limit restriction on both hypervisor and
2534 For details see: Documentation/admin-guide/hw-vuln/l1tf.rst
2540 lapic [X86-32,APIC] Enable the local APIC even if BIOS
2543 lapic= [X86,APIC] Do not use TSC deadline
2544 value for LAPIC timer one-shot implementation. Default
2545 back to the programmable timer unit in the LAPIC.
2546 Format: notscdeadline
2548 lapic_timer_c2_ok [X86,APIC] trust the local apic timer
2551 libata.dma= [LIBATA] DMA control
2552 libata.dma=0 Disable all PATA and SATA DMA
2553 libata.dma=1 PATA and SATA Disk DMA only
2554 libata.dma=2 ATAPI (CDROM) DMA only
2555 libata.dma=4 Compact Flash DMA only
2556 Combinations also work, so libata.dma=3 enables DMA
2557 for disks and CDROMs, but not CFs.
2559 libata.ignore_hpa= [LIBATA] Ignore HPA limit
2560 libata.ignore_hpa=0 keep BIOS limits (default)
2561 libata.ignore_hpa=1 ignore limits, using full disk
2563 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume
2567 libata.force= [LIBATA] Force configurations. The format is comma-
2568 separated list of "[ID:]VAL" where ID is
2569 PORT[.DEVICE]. PORT and DEVICE are decimal numbers
2570 matching port, link or device. Basically, it matches
2571 the ATA ID string printed on console by libata. If
2572 the whole ID part is omitted, the last PORT and DEVICE
2573 values are used. If ID hasn't been specified yet, the
2574 configuration applies to all ports, links and devices.
2576 If only DEVICE is omitted, the parameter applies to
2577 the port and all links and devices behind it. DEVICE
2578 number of 0 either selects the first device or the
2579 first fan-out link behind PMP device. It does not
2580 select the host link. DEVICE number of 15 selects the
2581 host link and device attached to it.
2583 The VAL specifies the configuration to force. As long
2584 as there's no ambiguity shortcut notation is allowed.
2585 For example, both 1.5 and 1.5G would work for 1.5Gbps.
2586 The following configurations can be forced.
2588 * Cable type: 40c, 80c, short40c, unk, ign or sata.
2589 Any ID with matching PORT is used.
2591 * SATA link speed limit: 1.5Gbps or 3.0Gbps.
2593 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7].
2594 udma[/][16,25,33,44,66,100,133] notation is also
2597 * [no]ncq: Turn on or off NCQ.
2599 * [no]ncqtrim: Turn off queued DSM TRIM.
2601 * nohrst, nosrst, norst: suppress hard, soft
2604 * rstonce: only attempt one reset during
2605 hot-unplug link recovery
2607 * dump_id: dump IDENTIFY data.
2609 * atapi_dmadir: Enable ATAPI DMADIR bridge support
2611 * disable: Disable this device.
2613 If there are multiple matching configurations changing
2614 the same attribute, the last one is used.
2616 memblock=debug [KNL] Enable memblock debug messages.
2618 load_ramdisk= [RAM] [Deprecated]
2620 lockd.nlm_grace_period=P [NFS] Assign grace period.
2623 lockd.nlm_tcpport=N [NFS] Assign TCP port.
2626 lockd.nlm_timeout=T [NFS] Assign timeout value.
2629 lockd.nlm_udpport=M [NFS] Assign UDP port.
2632 lockdown= [SECURITY]
2633 { integrity | confidentiality }
2634 Enable the kernel lockdown feature. If set to
2635 integrity, kernel features that allow userland to
2636 modify the running kernel are disabled. If set to
2637 confidentiality, kernel features that allow userland
2638 to extract confidential information from the kernel
2641 locktorture.nreaders_stress= [KNL]
2642 Set the number of locking read-acquisition kthreads.
2643 Defaults to being automatically set based on the
2644 number of online CPUs.
2646 locktorture.nwriters_stress= [KNL]
2647 Set the number of locking write-acquisition kthreads.
2649 locktorture.onoff_holdoff= [KNL]
2650 Set time (s) after boot for CPU-hotplug testing.
2652 locktorture.onoff_interval= [KNL]
2653 Set time (s) between CPU-hotplug operations, or
2654 zero to disable CPU-hotplug testing.
2656 locktorture.shuffle_interval= [KNL]
2657 Set task-shuffle interval (jiffies). Shuffling
2658 tasks allows some CPUs to go into dyntick-idle
2659 mode during the locktorture test.
2661 locktorture.shutdown_secs= [KNL]
2662 Set time (s) after boot system shutdown. This
2663 is useful for hands-off automated testing.
2665 locktorture.stat_interval= [KNL]
2666 Time (s) between statistics printk()s.
2668 locktorture.stutter= [KNL]
2669 Time (s) to stutter testing, for example,
2670 specifying five seconds causes the test to run for
2671 five seconds, wait for five seconds, and so on.
2672 This tests the locking primitive's ability to
2673 transition abruptly to and from idle.
2675 locktorture.torture_type= [KNL]
2676 Specify the locking implementation to test.
2678 locktorture.verbose= [KNL]
2679 Enable additional printk() statements.
2681 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver
2684 loglevel= All Kernel Messages with a loglevel smaller than the
2685 console loglevel will be printed to the console. It can
2686 also be changed with klogd or other programs. The
2687 loglevels are defined as follows:
2689 0 (KERN_EMERG) system is unusable
2690 1 (KERN_ALERT) action must be taken immediately
2691 2 (KERN_CRIT) critical conditions
2692 3 (KERN_ERR) error conditions
2693 4 (KERN_WARNING) warning conditions
2694 5 (KERN_NOTICE) normal but significant condition
2695 6 (KERN_INFO) informational
2696 7 (KERN_DEBUG) debug-level messages
2698 log_buf_len=n[KMG] Sets the size of the printk ring buffer,
2699 in bytes. n must be a power of two and greater
2700 than the minimal size. The minimal size is defined
2701 by LOG_BUF_SHIFT kernel config parameter. There is
2702 also CONFIG_LOG_CPU_MAX_BUF_SHIFT config parameter
2703 that allows to increase the default size depending on
2704 the number of CPUs. See init/Kconfig for more details.
2706 logo.nologo [FB] Disables display of the built-in Linux logo.
2707 This may be used to provide more screen space for
2708 kernel log messages and is useful when debugging
2709 kernel boot problems.
2711 lp=0 [LP] Specify parallel ports to use, e.g,
2712 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses
2713 lp=reset first parallel port). 'lp=0' disables the
2714 lp=auto printer driver. 'lp=reset' (which can be
2715 specified in addition to the ports) causes
2716 attached printers to be reset. Using
2717 lp=port1,port2,... specifies the parallel ports
2718 to associate lp devices with, starting with
2719 lp0. A port specification may be 'none' to skip
2720 that lp device, or a parport name such as
2721 'parport0'. Specifying 'lp=auto' instead of a
2722 port specification list means that device IDs
2723 from each port should be examined, to see if
2724 an IEEE 1284-compliant printer is attached; if
2725 so, the driver will manage that printer.
2726 See also header of drivers/char/lp.c.
2729 Sets loops_per_jiffy to given constant, thus avoiding
2730 time-consuming boot-time autodetection (up to 250 ms per
2731 CPU). 0 enables autodetection (default). To determine
2732 the correct value for your kernel, boot with normal
2733 autodetection and see what value is printed. Note that
2734 on SMP systems the preset will be applied to all CPUs,
2735 which is likely to cause problems if your CPUs need
2736 significantly divergent settings. An incorrect value
2737 will cause delays in the kernel to be wrong, leading to
2738 unpredictable I/O errors and other breakage. Although
2739 unlikely, in the extreme case this might damage your
2743 Format: <io>,<irq>,<dma>
2745 lsm.debug [SECURITY] Enable LSM initialization debugging output.
2748 [SECURITY] Choose order of LSM initialization. This
2749 overrides CONFIG_LSM, and the "security=" parameter.
2751 machvec= [IA-64] Force the use of a particular machine-vector
2752 (machvec) in a generic kernel.
2753 Example: machvec=hpzx1
2755 machtype= [Loongson] Share the same kernel image file between
2756 different yeeloong laptops.
2757 Example: machtype=lemote-yeeloong-2f-7inch
2759 max_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory greater
2760 than or equal to this physical address is ignored.
2762 maxcpus= [SMP] Maximum number of processors that an SMP kernel
2763 will bring up during bootup. maxcpus=n : n >= 0 limits
2764 the kernel to bring up 'n' processors. Surely after
2765 bootup you can bring up the other plugged cpu by executing
2766 "echo 1 > /sys/devices/system/cpu/cpuX/online". So maxcpus
2767 only takes effect during system bootup.
2768 While n=0 is a special case, it is equivalent to "nosmp",
2769 which also disables the IO APIC.
2771 max_loop= [LOOP] The number of loop block devices that get
2772 (loop.max_loop) unconditionally pre-created at init time. The default
2773 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead
2774 of statically allocating a predefined number, loop
2775 devices can be requested on-demand with the
2776 /dev/loop-control interface.
2778 mce [X86-32] Machine Check Exception
2780 mce=option [X86-64] See Documentation/x86/x86_64/boot-options.rst
2782 md= [HW] RAID subsystems devices and level
2783 See Documentation/admin-guide/md.rst.
2786 Format: <first>,<last>
2787 Specifies range of consoles to be captured by the MDA.
2790 Control mitigation for the Micro-architectural Data
2791 Sampling (MDS) vulnerability.
2793 Certain CPUs are vulnerable to an exploit against CPU
2794 internal buffers which can forward information to a
2795 disclosure gadget under certain conditions.
2797 In vulnerable processors, the speculatively
2798 forwarded data can be used in a cache side channel
2799 attack, to access data to which the attacker does
2800 not have direct access.
2802 This parameter controls the MDS mitigation. The
2805 full - Enable MDS mitigation on vulnerable CPUs
2806 full,nosmt - Enable MDS mitigation and disable
2807 SMT on vulnerable CPUs
2808 off - Unconditionally disable MDS mitigation
2810 On TAA-affected machines, mds=off can be prevented by
2811 an active TAA mitigation as both vulnerabilities are
2812 mitigated with the same mechanism so in order to disable
2813 this mitigation, you need to specify tsx_async_abort=off
2816 Not specifying this option is equivalent to
2819 For details see: Documentation/admin-guide/hw-vuln/mds.rst
2821 mem=nn[KMG] [KNL,BOOT] Force usage of a specific amount of memory
2822 Amount of memory to be used in cases as follows:
2825 2 when the kernel is not able to see the whole system memory;
2826 3 memory that lies after 'mem=' boundary is excluded from
2827 the hypervisor, then assigned to KVM guests.
2829 [X86] Work as limiting max address. Use together
2830 with memmap= to avoid physical address space collisions.
2831 Without memmap= PCI devices could be placed at addresses
2832 belonging to unused RAM.
2834 Note that this only takes effects during boot time since
2835 in above case 3, memory may need be hot added after boot
2836 if system memory of hypervisor is not sufficient.
2838 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel
2842 [KNL,SH] Allow user to override the default size for
2843 per-device physically contiguous DMA buffers.
2845 memhp_default_state=online/offline
2846 [KNL] Set the initial state for the memory hotplug
2847 onlining policy. If not specified, the default value is
2848 set according to the
2849 CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE kernel config
2851 See Documentation/admin-guide/mm/memory-hotplug.rst.
2853 memmap=exactmap [KNL,X86] Enable setting of an exact
2854 E820 memory map, as specified by the user.
2855 Such memmap=exactmap lines can be constructed based on
2856 BIOS output or other requirements. See the memmap=nn@ss
2859 memmap=nn[KMG]@ss[KMG]
2860 [KNL, X86, MIPS, XTENSA] Force usage of a specific region of memory.
2861 Region of memory to be used is from ss to ss+nn.
2862 If @ss[KMG] is omitted, it is equivalent to mem=nn[KMG],
2863 which limits max address to nn[KMG].
2864 Multiple different regions can be specified,
2867 memmap=100M@2G,100M#3G,1G!1024G
2869 memmap=nn[KMG]#ss[KMG]
2870 [KNL,ACPI] Mark specific memory as ACPI data.
2871 Region of memory to be marked is from ss to ss+nn.
2873 memmap=nn[KMG]$ss[KMG]
2874 [KNL,ACPI] Mark specific memory as reserved.
2875 Region of memory to be reserved is from ss to ss+nn.
2876 Example: Exclude memory from 0x18690000-0x1869ffff
2877 memmap=64K$0x18690000
2879 memmap=0x10000$0x18690000
2880 Some bootloaders may need an escape character before '$',
2881 like Grub2, otherwise '$' and the following number
2884 memmap=nn[KMG]!ss[KMG]
2885 [KNL,X86] Mark specific memory as protected.
2886 Region of memory to be used, from ss to ss+nn.
2887 The memory region may be marked as e820 type 12 (0xc)
2888 and is NVDIMM or ADR memory.
2890 memmap=<size>%<offset>-<oldtype>+<newtype>
2891 [KNL,ACPI] Convert memory within the specified region
2892 from <oldtype> to <newtype>. If "-<oldtype>" is left
2893 out, the whole region will be marked as <newtype>,
2894 even if previously unavailable. If "+<newtype>" is left
2895 out, matching memory will be removed. Types are
2896 specified as e820 types, e.g., 1 = RAM, 2 = reserved,
2897 3 = ACPI, 12 = PRAM.
2899 memory_corruption_check=0/1 [X86]
2900 Some BIOSes seem to corrupt the first 64k of
2901 memory when doing things like suspend/resume.
2902 Setting this option will scan the memory
2903 looking for corruption. Enabling this will
2904 both detect corruption and prevent the kernel
2905 from using the memory being corrupted.
2906 However, its intended as a diagnostic tool; if
2907 repeatable BIOS-originated corruption always
2908 affects the same memory, you can use memmap=
2909 to prevent the kernel from using that memory.
2911 memory_corruption_check_size=size [X86]
2912 By default it checks for corruption in the low
2913 64k, making this memory unavailable for normal
2914 use. Use this parameter to scan for
2915 corruption in more or less memory.
2917 memory_corruption_check_period=seconds [X86]
2918 By default it checks for corruption every 60
2919 seconds. Use this parameter to check at some
2920 other rate. 0 disables periodic checking.
2922 memory_hotplug.memmap_on_memory
2923 [KNL,X86,ARM] Boolean flag to enable this feature.
2924 Format: {on | off (default)}
2925 When enabled, runtime hotplugged memory will
2926 allocate its internal metadata (struct pages)
2927 from the hotadded memory which will allow to
2928 hotadd a lot of memory without requiring
2929 additional memory to do so.
2930 This feature is disabled by default because it
2931 has some implication on large (e.g. GB)
2932 allocations in some configurations (e.g. small
2934 The state of the flag can be read in
2935 /sys/module/memory_hotplug/parameters/memmap_on_memory.
2936 Note that even when enabled, there are a few cases where
2937 the feature is not effective.
2939 This is not compatible with hugetlb_free_vmemmap. If
2940 both parameters are enabled, hugetlb_free_vmemmap takes
2941 precedence over memory_hotplug.memmap_on_memory.
2943 memtest= [KNL,X86,ARM,PPC,RISCV] Enable memtest
2945 default : 0 <disable>
2946 Specifies the number of memtest passes to be
2947 performed. Each pass selects another test
2948 pattern from a given set of patterns. Memtest
2949 fills the memory with this pattern, validates
2950 memory contents and reserves bad memory
2951 regions that are detected.
2953 mem_encrypt= [X86-64] AMD Secure Memory Encryption (SME) control
2954 Valid arguments: on, off
2955 Default (depends on kernel configuration option):
2956 on (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=y)
2957 off (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=n)
2958 mem_encrypt=on: Activate SME
2959 mem_encrypt=off: Do not activate SME
2961 Refer to Documentation/virt/kvm/amd-memory-encryption.rst
2962 for details on when memory encryption can be activated.
2964 mem_sleep_default= [SUSPEND] Default system suspend mode:
2965 s2idle - Suspend-To-Idle
2966 shallow - Power-On Suspend or equivalent (if supported)
2967 deep - Suspend-To-RAM or equivalent (if supported)
2968 See Documentation/admin-guide/pm/sleep-states.rst.
2970 meye.*= [HW] Set MotionEye Camera parameters
2971 See Documentation/admin-guide/media/meye.rst.
2973 mfgpt_irq= [IA-32] Specify the IRQ to use for the
2974 Multi-Function General Purpose Timers on AMD Geode
2977 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when
2978 the BIOS has incorrectly applied a workaround. TinyBIOS
2979 version 0.98 is known to be affected, 0.99 fixes the
2980 problem by letting the user disable the workaround.
2984 min_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory below this
2985 physical address is ignored.
2987 mini2440= [ARM,HW,KNL]
2988 Format:[0..2][b][c][t]
2990 MINI2440 configuration specification:
2991 0 - The attached screen is the 3.5" TFT
2992 1 - The attached screen is the 7" TFT
2993 2 - The VGA Shield is attached (1024x768)
2994 Leaving out the screen size parameter will not load
2995 the TFT driver, and the framebuffer will be left
2997 b - Enable backlight. The TFT backlight pin will be
2998 linked to the kernel VESA blanking code and a GPIO
2999 LED. This parameter is not necessary when using the
3001 c - Enable the s3c camera interface.
3002 t - Reserved for enabling touchscreen support. The
3003 touchscreen support is not enabled in the mainstream
3004 kernel as of 2.6.30, a preliminary port can be found
3005 in the "bleeding edge" mini2440 support kernel at
3006 https://repo.or.cz/w/linux-2.6/mini2440.git
3009 [X86,PPC,S390,ARM64] Control optional mitigations for
3010 CPU vulnerabilities. This is a set of curated,
3011 arch-independent options, each of which is an
3012 aggregation of existing arch-specific options.
3015 Disable all optional CPU mitigations. This
3016 improves system performance, but it may also
3017 expose users to several CPU vulnerabilities.
3018 Equivalent to: nopti [X86,PPC]
3020 nospectre_v1 [X86,PPC]
3022 nospectre_v2 [X86,PPC,S390,ARM64]
3023 spectre_v2_user=off [X86]
3024 spec_store_bypass_disable=off [X86,PPC]
3025 ssbd=force-off [ARM64]
3028 tsx_async_abort=off [X86]
3029 kvm.nx_huge_pages=off [X86]
3030 no_entry_flush [PPC]
3031 no_uaccess_flush [PPC]
3034 This does not have any effect on
3035 kvm.nx_huge_pages when
3036 kvm.nx_huge_pages=force.
3039 Mitigate all CPU vulnerabilities, but leave SMT
3040 enabled, even if it's vulnerable. This is for
3041 users who don't want to be surprised by SMT
3042 getting disabled across kernel upgrades, or who
3043 have other ways of avoiding SMT-based attacks.
3044 Equivalent to: (default behavior)
3047 Mitigate all CPU vulnerabilities, disabling SMT
3048 if needed. This is for users who always want to
3049 be fully mitigated, even if it means losing SMT.
3050 Equivalent to: l1tf=flush,nosmt [X86]
3051 mds=full,nosmt [X86]
3052 tsx_async_abort=full,nosmt [X86]
3055 [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this
3056 parameter allows control of the logging verbosity for
3057 the additional memory initialisation checks. A value
3058 of 0 disables mminit logging and a level of 4 will
3059 log everything. Information is printed at KERN_DEBUG
3060 so loglevel=8 may also need to be specified.
3063 [KNL] When CONFIG_MODULE_SIG is set, this means that
3064 modules without (valid) signatures will fail to load.
3065 Note that if CONFIG_MODULE_SIG_FORCE is set, that
3066 is always true, so this option does nothing.
3068 module_blacklist= [KNL] Do not load a comma-separated list of
3069 modules. Useful for debugging problem modules.
3072 [MOUSE] Maximum time between finger touching and
3073 leaving touchpad surface for touch to be considered
3074 a tap and be reported as a left button click (for
3075 touchpads working in absolute mode only).
3077 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices
3078 reporting absolute coordinates, such as tablets
3079 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices
3080 reporting absolute coordinates, such as tablets
3082 movablecore= [KNL,X86,IA-64,PPC]
3083 Format: nn[KMGTPE] | nn%
3084 This parameter is the complement to kernelcore=, it
3085 specifies the amount of memory used for migratable
3086 allocations. If both kernelcore and movablecore is
3087 specified, then kernelcore will be at *least* the
3088 specified value but may be more. If movablecore on its
3089 own is specified, the administrator must be careful
3090 that the amount of memory usable for all allocations
3093 movable_node [KNL] Boot-time switch to make hotplugable memory
3094 NUMA nodes to be movable. This means that the memory
3095 of such nodes will be usable only for movable
3096 allocations which rules out almost all kernel
3097 allocations. Use with caution!
3099 MTD_Partition= [MTD]
3100 Format: <name>,<region-number>,<size>,<offset>
3102 MTD_Region= [MTD] Format:
3103 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>]
3106 See drivers/mtd/parsers/cmdlinepart.c
3108 multitce=off [PPC] This parameter disables the use of the pSeries
3109 firmware feature for updating multiple TCE entries
3112 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration
3114 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock]
3116 boundary - index of last SLC block on Flex-OneNAND.
3117 The remaining blocks are configured as MLC blocks.
3118 lock - Configure if Flex-OneNAND boundary should be locked.
3119 Once locked, the boundary cannot be changed.
3120 1 indicates lock status, 0 indicates unlock status.
3123 ARM/S3C2412 JIVE boot control
3125 See arch/arm/mach-s3c/mach-jive.c
3127 mtouchusb.raw_coordinates=
3128 [HW] Make the MicroTouch USB driver use raw coordinates
3129 ('y', default) or cooked coordinates ('n')
3131 mtrr_chunk_size=nn[KMG] [X86]
3132 used for mtrr cleanup. It is largest continuous chunk
3133 that could hold holes aka. UC entries.
3135 mtrr_gran_size=nn[KMG] [X86]
3136 Used for mtrr cleanup. It is granularity of mtrr block.
3138 Large value could prevent small alignment from
3141 mtrr_spare_reg_nr=n [X86]
3143 Range: 0,7 : spare reg number
3145 Used for mtrr cleanup. It is spare mtrr entries number.
3146 Set to 2 or more if your graphical card needs more.
3148 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card
3150 netdev= [NET] Network devices parameters
3151 Format: <irq>,<io>,<mem_start>,<mem_end>,<name>
3152 Note that mem_start is often overloaded to mean
3153 something different and driver-specific.
3154 This usage is only documented in each driver source
3158 [NETFILTER] Enable connection tracking flow accounting
3159 0 to disable accounting
3160 1 to enable accounting
3163 nfsaddrs= [NFS] Deprecated. Use ip= instead.
3164 See Documentation/admin-guide/nfs/nfsroot.rst.
3166 nfsroot= [NFS] nfs root filesystem for disk-less boxes.
3167 See Documentation/admin-guide/nfs/nfsroot.rst.
3169 nfsrootdebug [NFS] enable nfsroot debugging messages.
3170 See Documentation/admin-guide/nfs/nfsroot.rst.
3172 nfs.callback_nr_threads=
3173 [NFSv4] set the total number of threads that the
3174 NFS client will assign to service NFSv4 callback
3177 nfs.callback_tcpport=
3178 [NFS] set the TCP port on which the NFSv4 callback
3179 channel should listen.
3182 [NFS] sets the pathname to the program which is used
3183 to update the NFS client cache entries.
3185 nfs.cache_getent_timeout=
3186 [NFS] sets the timeout after which an attempt to
3187 update a cache entry is deemed to have failed.
3189 nfs.idmap_cache_timeout=
3190 [NFS] set the maximum lifetime for idmapper cache
3194 [NFS] enable 64-bit inode numbers.
3195 If zero, the NFS client will fake up a 32-bit inode
3196 number for the readdir() and stat() syscalls instead
3197 of returning the full 64-bit number.
3198 The default is to return 64-bit inode numbers.
3200 nfs.max_session_cb_slots=
3201 [NFSv4.1] Sets the maximum number of session
3202 slots the client will assign to the callback
3203 channel. This determines the maximum number of
3204 callbacks the client will process in parallel for
3205 a particular server.
3207 nfs.max_session_slots=
3208 [NFSv4.1] Sets the maximum number of session slots
3209 the client will attempt to negotiate with the server.
3210 This limits the number of simultaneous RPC requests
3211 that the client can send to the NFSv4.1 server.
3212 Note that there is little point in setting this
3213 value higher than the max_tcp_slot_table_limit.
3215 nfs.nfs4_disable_idmapping=
3216 [NFSv4] When set to the default of '1', this option
3217 ensures that both the RPC level authentication
3218 scheme and the NFS level operations agree to use
3219 numeric uids/gids if the mount is using the
3220 'sec=sys' security flavour. In effect it is
3221 disabling idmapping, which can make migration from
3222 legacy NFSv2/v3 systems to NFSv4 easier.
3223 Servers that do not support this mode of operation
3224 will be autodetected by the client, and it will fall
3225 back to using the idmapper.
3226 To turn off this behaviour, set the value to '0'.
3228 [NFS4] Specify an additional fixed unique ident-
3229 ification string that NFSv4 clients can insert into
3230 their nfs_client_id4 string. This is typically a
3231 UUID that is generated at system install time.
3233 nfs.send_implementation_id =
3234 [NFSv4.1] Send client implementation identification
3235 information in exchange_id requests.
3236 If zero, no implementation identification information
3238 The default is to send the implementation identification
3241 nfs.recover_lost_locks =
3242 [NFSv4] Attempt to recover locks that were lost due
3243 to a lease timeout on the server. Please note that
3244 doing this risks data corruption, since there are
3245 no guarantees that the file will remain unchanged
3246 after the locks are lost.
3247 If you want to enable the kernel legacy behaviour of
3248 attempting to recover these locks, then set this
3250 The default parameter value of '0' causes the kernel
3251 not to attempt recovery of lost locks.
3253 nfs4.layoutstats_timer =
3254 [NFSv4.2] Change the rate at which the kernel sends
3255 layoutstats to the pNFS metadata server.
3257 Setting this to value to 0 causes the kernel to use
3258 whatever value is the default set by the layout
3259 driver. A non-zero value sets the minimum interval
3260 in seconds between layoutstats transmissions.
3262 nfsd.inter_copy_offload_enable =
3263 [NFSv4.2] When set to 1, the server will support
3264 server-to-server copies for which this server is
3265 the destination of the copy.
3267 nfsd.nfsd4_ssc_umount_timeout =
3268 [NFSv4.2] When used as the destination of a
3269 server-to-server copy, knfsd temporarily mounts
3270 the source server. It caches the mount in case
3271 it will be needed again, and discards it if not
3272 used for the number of milliseconds specified by
3275 nfsd.nfs4_disable_idmapping=
3276 [NFSv4] When set to the default of '1', the NFSv4
3277 server will return only numeric uids and gids to
3278 clients using auth_sys, and will accept numeric uids
3279 and gids from such clients. This is intended to ease
3280 migration from NFSv2/v3.
3283 nmi_backtrace.backtrace_idle [KNL]
3284 Dump stacks even of idle CPUs in response to an
3285 NMI stack-backtrace request.
3287 nmi_debug= [KNL,SH] Specify one or more actions to take
3288 when a NMI is triggered.
3289 Format: [state][,regs][,debounce][,die]
3291 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels
3292 Format: [panic,][nopanic,][num]
3294 0 - turn hardlockup detector in nmi_watchdog off
3295 1 - turn hardlockup detector in nmi_watchdog on
3296 When panic is specified, panic when an NMI watchdog
3297 timeout occurs (or 'nopanic' to not panic on an NMI
3298 watchdog, if CONFIG_BOOTPARAM_HARDLOCKUP_PANIC is set)
3299 To disable both hard and soft lockup detectors,
3300 please see 'nowatchdog'.
3301 This is useful when you use a panic=... timeout and
3302 need the box quickly up again.
3304 These settings can be accessed at runtime via
3305 the nmi_watchdog and hardlockup_panic sysctls.
3307 netpoll.carrier_timeout=
3308 [NET] Specifies amount of time (in seconds) that
3309 netpoll should wait for a carrier. By default netpoll
3312 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths
3313 emulation library even if a 387 maths coprocessor
3316 no5lvl [X86-64] Disable 5-level paging mode. Forces
3317 kernel to use 4-level paging instead.
3319 nofsgsbase [X86] Disables FSGSBASE instructions.
3322 [HW] Never suspend the console
3323 Disable suspending of consoles during suspend and
3324 hibernate operations. Once disabled, debugging
3325 messages can reach various consoles while the rest
3326 of the system is being put to sleep (ie, while
3327 debugging driver suspend/resume hooks). This may
3328 not work reliably with all consoles, but is known
3329 to work with serial and VGA consoles.
3330 To facilitate more flexible debugging, we also add
3331 console_suspend, a printk module parameter to control
3332 it. Users could use console_suspend (usually
3333 /sys/module/printk/parameters/console_suspend) to
3334 turn on/off it dynamically.
3336 novmcoredd [KNL,KDUMP]
3337 Disable device dump. Device dump allows drivers to
3338 append dump data to vmcore so you can collect driver
3339 specified debug info. Drivers can append the data
3340 without any limit and this data is stored in memory,
3341 so this may cause significant memory stress. Disabling
3342 device dump can help save memory but the driver debug
3343 data will be no longer available. This parameter
3344 is only available when CONFIG_PROC_VMCORE_DEVICE_DUMP
3347 noaliencache [MM, NUMA, SLAB] Disables the allocation of alien
3348 caches in the slab allocator. Saves per-node memory,
3349 but will impact performance.
3353 noaltinstr [S390] Disables alternative instructions patching
3354 (CPU alternatives feature).
3356 noapic [SMP,APIC] Tells the kernel to not make use of any
3357 IOAPICs that may be present in the system.
3359 noautogroup Disable scheduler automatic task group creation.
3361 nobats [PPC] Do not use BATs for mapping kernel lowmem
3362 on "Classic" PPC cores.
3366 noclflush [BUGS=X86] Don't use the CLFLUSH instruction
3368 delayacct [KNL] Enable per-task delay accounting
3370 nodsp [SH] Disable hardware DSP at boot time.
3372 noefi Disable EFI runtime services support.
3374 no_entry_flush [PPC] Don't flush the L1-D cache when entering the kernel.
3379 On X86-32 available only on PAE configured kernels.
3380 noexec=on: enable non-executable mappings (default)
3381 noexec=off: disable non-executable mappings
3384 Disable SMAP (Supervisor Mode Access Prevention)
3385 even if it is supported by processor.
3388 Disable SMEP (Supervisor Mode Execution Prevention)
3389 even if it is supported by processor.
3392 This affects only 32-bit executables.
3393 noexec32=on: enable non-executable mappings (default)
3394 read doesn't imply executable mappings
3395 noexec32=off: disable non-executable mappings
3396 read implies executable mappings
3398 nofpu [MIPS,SH] Disable hardware FPU at boot time.
3400 nofxsr [BUGS=X86-32] Disables x86 floating point extended
3401 register save and restore. The kernel will only save
3402 legacy floating-point registers on task switch.
3404 nohugeiomap [KNL,X86,PPC,ARM64] Disable kernel huge I/O mappings.
3406 nohugevmalloc [PPC] Disable kernel huge vmalloc mappings.
3408 nosmt [KNL,S390] Disable symmetric multithreading (SMT).
3409 Equivalent to smt=1.
3411 [KNL,X86] Disable symmetric multithreading (SMT).
3412 nosmt=force: Force disable SMT, cannot be undone
3413 via the sysfs control file.
3415 nospectre_v1 [X86,PPC] Disable mitigations for Spectre Variant 1
3416 (bounds check bypass). With this option data leaks are
3417 possible in the system.
3419 nospectre_v2 [X86,PPC_FSL_BOOK3E,ARM64] Disable all mitigations for
3420 the Spectre variant 2 (indirect branch prediction)
3421 vulnerability. System may allow data leaks with this
3424 nospec_store_bypass_disable
3425 [HW] Disable all mitigations for the Speculative Store Bypass vulnerability
3428 [PPC] Don't flush the L1-D cache after accessing user data.
3430 noxsave [BUGS=X86] Disables x86 extended register state save
3431 and restore using xsave. The kernel will fallback to
3432 enabling legacy floating-point and sse state.
3434 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended
3435 register states. The kernel will fall back to use
3436 xsave to save the states. By using this parameter,
3437 performance of saving the states is degraded because
3438 xsave doesn't support modified optimization while
3439 xsaveopt supports it on xsaveopt enabled systems.
3441 noxsaves [X86] Disables xsaves and xrstors used in saving and
3442 restoring x86 extended register state in compacted
3443 form of xsave area. The kernel will fall back to use
3444 xsaveopt and xrstor to save and restore the states
3445 in standard form of xsave area. By using this
3446 parameter, xsave area per process might occupy more
3447 memory on xsaves enabled systems.
3449 nohlt [ARM,ARM64,MICROBLAZE,SH] Forces the kernel to busy wait
3450 in do_idle() and not use the arch_cpu_idle()
3451 implementation; requires CONFIG_GENERIC_IDLE_POLL_SETUP
3452 to be effective. This is useful on platforms where the
3453 sleep(SH) or wfi(ARM,ARM64) instructions do not work
3454 correctly or when doing power measurements to evalute
3455 the impact of the sleep instructions. This is also
3456 useful when using JTAG debugger.
3458 no_file_caps Tells the kernel not to honor file capabilities. The
3459 only way then for a file to be executed with privilege
3460 is to be setuid root or executed by root.
3462 nohalt [IA-64] Tells the kernel not to use the power saving
3463 function PAL_HALT_LIGHT when idle. This increases
3464 power-consumption. On the positive side, it reduces
3465 interrupt wake-up latency, which may improve performance
3466 in certain environments such as networked servers or
3470 Force pointers printed to the console or buffers to be
3471 unhashed. By default, when a pointer is printed via %p
3472 format string, that pointer is "hashed", i.e. obscured
3473 by hashing the pointer value. This is a security feature
3474 that hides actual kernel addresses from unprivileged
3475 users, but it also makes debugging the kernel more
3476 difficult since unequal pointers can no longer be
3477 compared. However, if this command-line option is
3478 specified, then all normal pointers will have their true
3479 value printed. Pointers printed via %pK may still be
3480 hashed. This option should only be specified when
3481 debugging the kernel. Please do not use on production
3484 nohibernate [HIBERNATION] Disable hibernation and resume.
3486 nohz= [KNL] Boottime enable/disable dynamic ticks
3487 Valid arguments: on, off
3490 nohz_full= [KNL,BOOT,SMP,ISOL]
3491 The argument is a cpu list, as described above.
3492 In kernels built with CONFIG_NO_HZ_FULL=y, set
3493 the specified list of CPUs whose tick will be stopped
3494 whenever possible. The boot CPU will be forced outside
3495 the range to maintain the timekeeping. Any CPUs
3496 in this list will have their RCU callbacks offloaded,
3497 just as if they had also been called out in the
3498 rcu_nocbs= boot parameter.
3500 noiotrap [SH] Disables trapped I/O port accesses.
3502 noirqdebug [X86-32] Disables the code which attempts to detect and
3503 disable unhandled interrupt sources.
3505 no_timer_check [X86,APIC] Disables the code which tests for
3506 broken timer IRQ sources.
3508 noisapnp [ISAPNP] Disables ISA PnP code.
3510 noinitrd [RAM] Tells the kernel not to load any configured
3513 nointremap [X86-64, Intel-IOMMU] Do not enable interrupt
3515 [Deprecated - use intremap=off]
3519 noinvpcid [X86] Disable the INVPCID cpu feature.
3521 nojitter [IA-64] Disables jitter checking for ITC timers.
3523 no-kvmclock [X86,KVM] Disable paravirtualized KVM clock driver
3525 no-kvmapf [X86,KVM] Disable paravirtualized asynchronous page
3529 [X86,PV_OPS] Disable paravirtualized VMware scheduler
3530 clock and use the default one.
3532 no-steal-acc [X86,PV_OPS,ARM64] Disable paravirtualized steal time
3533 accounting. steal time is computed, but won't
3534 influence scheduler behaviour
3536 nolapic [X86-32,APIC] Do not enable or use the local APIC.
3538 nolapic_timer [X86-32,APIC] Do not use the local APIC timer.
3540 noltlbs [PPC] Do not use large page/tlb entries for kernel
3541 lowmem mapping on PPC40x and PPC8xx
3543 nomca [IA-64] Disable machine check abort handling
3545 nomce [X86-32] Disable Machine Check Exception
3547 nomfgpt [X86-32] Disable Multi-Function General Purpose
3548 Timer usage (for AMD Geode machines).
3550 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to
3551 shutdown the other cpus. Instead use the REBOOT_VECTOR
3554 nomodule Disable module load
3556 nopat [X86] Disable PAT (page attribute table extension of
3557 pagetables) support.
3559 nopcid [X86-64] Disable the PCID cpu feature.
3561 norandmaps Don't use address space randomization. Equivalent to
3562 echo 0 > /proc/sys/kernel/randomize_va_space
3564 noreplace-smp [X86-32,SMP] Don't replace SMP instructions
3565 with UP alternatives
3567 nordrand [X86] Disable kernel use of the RDRAND and
3568 RDSEED instructions even if they are supported
3569 by the processor. RDRAND and RDSEED are still
3570 available to user space applications.
3572 noresume [SWSUSP] Disables resume and restores original swap
3575 no-scroll [VGA] Disables scrollback.
3576 This is required for the Braillex ib80-piezo Braille
3577 reader made by F.H. Papenmeier (Germany).
3581 nosep [BUGS=X86-32] Disables x86 SYSENTER/SYSEXIT support.
3583 nosgx [X86-64,SGX] Disables Intel SGX kernel support.
3585 nosmp [SMP] Tells an SMP kernel to act as a UP kernel,
3586 and disable the IO APIC. legacy for "maxcpus=0".
3588 nosoftlockup [KNL] Disable the soft-lockup detector.
3590 nosync [HW,M68K] Disables sync negotiation for all devices.
3592 nowatchdog [KNL] Disable both lockup detectors, i.e.
3593 soft-lockup and NMI watchdog (hard-lockup).
3597 nox2apic [X86-64,APIC] Do not enable x2APIC mode.
3599 cpu0_hotplug [X86] Turn on CPU0 hotplug feature when
3600 CONFIG_BOOTPARAM_HOTPLUG_CPU0 is off.
3601 Some features depend on CPU0. Known dependencies are:
3602 1. Resume from suspend/hibernate depends on CPU0.
3603 Suspend/hibernate will fail if CPU0 is offline and you
3604 need to online CPU0 before suspend/hibernate.
3605 2. PIC interrupts also depend on CPU0. CPU0 can't be
3606 removed if a PIC interrupt is detected.
3607 It's said poweroff/reboot may depend on CPU0 on some
3608 machines although I haven't seen such issues so far
3609 after CPU0 is offline on a few tested machines.
3610 If the dependencies are under your control, you can
3611 turn on cpu0_hotplug.
3613 nps_mtm_hs_ctr= [KNL,ARC]
3614 This parameter sets the maximum duration, in
3615 cycles, each HW thread of the CTOP can run
3616 without interruptions, before HW switches it.
3617 The actual maximum duration is 16 times this
3619 Format: integer between 1 and 255
3622 nptcg= [IA-64] Override max number of concurrent global TLB
3623 purges which is reported from either PAL_VM_SUMMARY or
3626 nr_cpus= [SMP] Maximum number of processors that an SMP kernel
3627 could support. nr_cpus=n : n >= 1 limits the kernel to
3628 support 'n' processors. It could be larger than the
3629 number of already plugged CPU during bootup, later in
3630 runtime you can physically add extra cpu until it reaches
3631 n. So during boot up some boot time memory for per-cpu
3632 variables need be pre-allocated for later physical cpu
3635 nr_uarts= [SERIAL] maximum number of UARTs to be registered.
3637 numa=off [KNL, ARM64, PPC, RISCV, SPARC, X86] Disable NUMA, Only
3638 set up a single NUMA node spanning all memory.
3640 numa_balancing= [KNL,ARM64,PPC,RISCV,S390,X86] Enable or disable automatic
3642 Allowed values are enable and disable
3644 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA.
3645 'node', 'default' can be specified
3646 This can be set from sysctl after boot.
3647 See Documentation/admin-guide/sysctl/vm.rst for details.
3649 ohci1394_dma=early [HW] enable debugging via the ohci1394 driver.
3650 See Documentation/core-api/debugging-via-ohci1394.rst for more
3653 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands
3654 Rather than timing out after 20 ms if an EC
3655 command is not properly ACKed, override the length
3656 of the timeout. We have interrupts disabled while
3657 waiting for the ACK, so if this is set too high
3658 interrupts *may* be lost!
3660 omap_mux= [OMAP] Override bootloader pin multiplexing.
3661 Format: <mux_mode0.mode_name=value>...
3662 For example, to override I2C bus2:
3663 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100
3665 oops=panic Always panic on oopses. Default is to just kill the
3666 process, but there is a small probability of
3667 deadlocking the machine.
3668 This will also cause panics on machine check exceptions.
3669 Useful together with panic=30 to trigger a reboot.
3672 [KNL] Boolean flag to control whether the page allocator
3673 should randomize its free lists. The randomization may
3674 be automatically enabled if the kernel detects it is
3675 running on a platform with a direct-mapped memory-side
3676 cache, and this parameter can be used to
3677 override/disable that behavior. The state of the flag
3678 can be read from sysfs at:
3679 /sys/module/page_alloc/parameters/shuffle.
3681 page_owner= [KNL] Boot-time page_owner enabling option.
3682 Storage of the information about who allocated
3683 each page is disabled in default. With this switch,
3685 on: enable the feature
3687 page_poison= [KNL] Boot-time parameter changing the state of
3688 poisoning on the buddy allocator, available with
3689 CONFIG_PAGE_POISONING=y.
3690 off: turn off poisoning (default)
3691 on: turn on poisoning
3693 page_reporting.page_reporting_order=
3694 [KNL] Minimal page reporting order
3696 Adjust the minimal page reporting order. The page
3697 reporting is disabled when it exceeds (MAX_ORDER-1).
3699 panic= [KNL] Kernel behaviour on panic: delay <timeout>
3700 timeout > 0: seconds before rebooting
3701 timeout = 0: wait forever
3702 timeout < 0: reboot immediately
3705 panic_print= Bitmask for printing system info when panic happens.
3706 User can chose combination of the following bits:
3707 bit 0: print all tasks info
3708 bit 1: print system memory info
3709 bit 2: print timer info
3710 bit 3: print locks info if CONFIG_LOCKDEP is on
3711 bit 4: print ftrace buffer
3712 bit 5: print all printk messages in buffer
3714 panic_on_taint= Bitmask for conditionally calling panic() in add_taint()
3715 Format: <hex>[,nousertaint]
3716 Hexadecimal bitmask representing the set of TAINT flags
3717 that will cause the kernel to panic when add_taint() is
3718 called with any of the flags in this set.
3719 The optional switch "nousertaint" can be utilized to
3720 prevent userspace forced crashes by writing to sysctl
3721 /proc/sys/kernel/tainted any flagset matching with the
3722 bitmask set on panic_on_taint.
3723 See Documentation/admin-guide/tainted-kernels.rst for
3724 extra details on the taint flags that users can pick
3725 to compose the bitmask to assign to panic_on_taint.
3727 panic_on_warn panic() instead of WARN(). Useful to cause kdump
3730 crash_kexec_post_notifiers
3731 Run kdump after running panic-notifiers and dumping
3732 kmsg. This only for the users who doubt kdump always
3733 succeeds in any situation.
3734 Note that this also increases risks of kdump failure,
3735 because some panic notifiers can make the crashed
3736 kernel more unstable.
3738 parkbd.port= [HW] Parallel port number the keyboard adapter is
3739 connected to, default is 0.
3741 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation,
3742 0 for XT, 1 for AT (default is AT).
3745 parport= [HW,PPT] Specify parallel ports. 0 disables.
3746 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] }
3747 Use 'auto' to force the driver to use any
3748 IRQ/DMA settings detected (the default is to
3749 ignore detected IRQ/DMA settings because of
3750 possible conflicts). You can specify the base
3751 address, IRQ, and DMA settings; IRQ and DMA
3752 should be numbers, or 'auto' (for using detected
3753 settings on that particular port), or 'nofifo'
3754 (to avoid using a FIFO even if it is detected).
3755 Parallel ports are assigned in the order they
3756 are specified on the command line, starting
3759 parport_init_mode= [HW,PPT]
3760 Configure VIA parallel port to operate in
3761 a specific mode. This is necessary on Pegasos
3762 computer where firmware has no options for setting
3763 up parallel port mode and sets it to spp.
3764 Currently this function knows 686a and 8231 chips.
3765 Format: [spp|ps2|epp|ecp|ecpepp]
3767 pata_legacy.all= [HW,LIBATA]
3769 Set to non-zero to probe primary and secondary ISA
3770 port ranges on PCI systems where no PCI PATA device
3771 has been found at either range. Disabled by default.
3773 pata_legacy.autospeed= [HW,LIBATA]
3775 Set to non-zero if a chip is present that snoops speed
3776 changes. Disabled by default.
3778 pata_legacy.ht6560a= [HW,LIBATA]
3780 Set to 1, 2, or 3 for HT 6560A on the primary channel,
3781 the secondary channel, or both channels respectively.
3782 Disabled by default.
3784 pata_legacy.ht6560b= [HW,LIBATA]
3786 Set to 1, 2, or 3 for HT 6560B on the primary channel,
3787 the secondary channel, or both channels respectively.
3788 Disabled by default.
3790 pata_legacy.iordy_mask= [HW,LIBATA]
3792 IORDY enable mask. Set individual bits to allow IORDY
3793 for the respective channel. Bit 0 is for the first
3794 legacy channel handled by this driver, bit 1 is for
3795 the second channel, and so on. The sequence will often
3796 correspond to the primary legacy channel, the secondary
3797 legacy channel, and so on, but the handling of a PCI
3798 bus and the use of other driver options may interfere
3799 with the sequence. By default IORDY is allowed across
3802 pata_legacy.opti82c46x= [HW,LIBATA]
3804 Set to 1, 2, or 3 for Opti 82c611A on the primary
3805 channel, the secondary channel, or both channels
3806 respectively. Disabled by default.
3808 pata_legacy.opti82c611a= [HW,LIBATA]
3810 Set to 1, 2, or 3 for Opti 82c465MV on the primary
3811 channel, the secondary channel, or both channels
3812 respectively. Disabled by default.
3814 pata_legacy.pio_mask= [HW,LIBATA]
3816 PIO mode mask for autospeed devices. Set individual
3817 bits to allow the use of the respective PIO modes.
3818 Bit 0 is for mode 0, bit 1 is for mode 1, and so on.
3819 All modes allowed by default.
3821 pata_legacy.probe_all= [HW,LIBATA]
3823 Set to non-zero to probe tertiary and further ISA
3824 port ranges on PCI systems. Disabled by default.
3826 pata_legacy.probe_mask= [HW,LIBATA]
3828 Probe mask for legacy ISA PATA ports. Depending on
3829 platform configuration and the use of other driver
3830 options up to 6 legacy ports are supported: 0x1f0,
3831 0x170, 0x1e8, 0x168, 0x1e0, 0x160, however probing
3832 of individual ports can be disabled by setting the
3833 corresponding bits in the mask to 1. Bit 0 is for
3834 the first port in the list above (0x1f0), and so on.
3835 By default all supported ports are probed.
3837 pata_legacy.qdi= [HW,LIBATA]
3839 Set to non-zero to probe QDI controllers. By default
3840 set to 1 if CONFIG_PATA_QDI_MODULE, 0 otherwise.
3842 pata_legacy.winbond= [HW,LIBATA]
3844 Set to non-zero to probe Winbond controllers. Use
3845 the standard I/O port (0x130) if 1, otherwise the
3846 value given is the I/O port to use (typically 0x1b0).
3847 By default set to 1 if CONFIG_PATA_WINBOND_VLB_MODULE,
3850 pata_platform.pio_mask= [HW,LIBATA]
3852 Supported PIO mode mask. Set individual bits to allow
3853 the use of the respective PIO modes. Bit 0 is for
3854 mode 0, bit 1 is for mode 1, and so on. Mode 0 only
3858 Halt all CPUs after the first oops has been printed for
3859 the specified number of seconds. This is to be used if
3860 your oopses keep scrolling off the screen.
3865 See header of drivers/block/paride/pcd.c.
3866 See also Documentation/admin-guide/blockdev/paride.rst.
3868 pci=option[,option...] [PCI] various PCI subsystem options.
3870 Some options herein operate on a specific device
3871 or a set of devices (<pci_dev>). These are
3872 specified in one of the following formats:
3874 [<domain>:]<bus>:<dev>.<func>[/<dev>.<func>]*
3875 pci:<vendor>:<device>[:<subvendor>:<subdevice>]
3877 Note: the first format specifies a PCI
3878 bus/device/function address which may change
3879 if new hardware is inserted, if motherboard
3880 firmware changes, or due to changes caused
3881 by other kernel parameters. If the
3882 domain is left unspecified, it is
3883 taken to be zero. Optionally, a path
3884 to a device through multiple device/function
3885 addresses can be specified after the base
3886 address (this is more robust against
3887 renumbering issues). The second format
3888 selects devices using IDs from the
3889 configuration space which may match multiple
3890 devices in the system.
3892 earlydump dump PCI config space before the kernel
3894 off [X86] don't probe for the PCI bus
3895 bios [X86-32] force use of PCI BIOS, don't access
3896 the hardware directly. Use this if your machine
3897 has a non-standard PCI host bridge.
3898 nobios [X86-32] disallow use of PCI BIOS, only direct
3899 hardware access methods are allowed. Use this
3900 if you experience crashes upon bootup and you
3901 suspect they are caused by the BIOS.
3902 conf1 [X86] Force use of PCI Configuration Access
3903 Mechanism 1 (config address in IO port 0xCF8,
3904 data in IO port 0xCFC, both 32-bit).
3905 conf2 [X86] Force use of PCI Configuration Access
3906 Mechanism 2 (IO port 0xCF8 is an 8-bit port for
3907 the function, IO port 0xCFA, also 8-bit, sets
3908 bus number. The config space is then accessed
3909 through ports 0xC000-0xCFFF).
3910 See http://wiki.osdev.org/PCI for more info
3911 on the configuration access mechanisms.
3912 noaer [PCIE] If the PCIEAER kernel config parameter is
3913 enabled, this kernel boot option can be used to
3914 disable the use of PCIE advanced error reporting.
3915 nodomains [PCI] Disable support for multiple PCI
3916 root domains (aka PCI segments, in ACPI-speak).
3917 nommconf [X86] Disable use of MMCONFIG for PCI
3919 check_enable_amd_mmconf [X86] check for and enable
3920 properly configured MMIO access to PCI
3921 config space on AMD family 10h CPU
3922 nomsi [MSI] If the PCI_MSI kernel config parameter is
3923 enabled, this kernel boot option can be used to
3924 disable the use of MSI interrupts system-wide.
3925 noioapicquirk [APIC] Disable all boot interrupt quirks.
3926 Safety option to keep boot IRQs enabled. This
3927 should never be necessary.
3928 ioapicreroute [APIC] Enable rerouting of boot IRQs to the
3929 primary IO-APIC for bridges that cannot disable
3930 boot IRQs. This fixes a source of spurious IRQs
3931 when the system masks IRQs.
3932 noioapicreroute [APIC] Disable workaround that uses the
3933 boot IRQ equivalent of an IRQ that connects to
3934 a chipset where boot IRQs cannot be disabled.
3935 The opposite of ioapicreroute.
3936 biosirq [X86-32] Use PCI BIOS calls to get the interrupt
3937 routing table. These calls are known to be buggy
3938 on several machines and they hang the machine
3939 when used, but on other computers it's the only
3940 way to get the interrupt routing table. Try
3941 this option if the kernel is unable to allocate
3942 IRQs or discover secondary PCI buses on your
3944 rom [X86] Assign address space to expansion ROMs.
3945 Use with caution as certain devices share
3946 address decoders between ROMs and other
3948 norom [X86] Do not assign address space to
3949 expansion ROMs that do not already have
3950 BIOS assigned address ranges.
3951 nobar [X86] Do not assign address space to the
3952 BARs that weren't assigned by the BIOS.
3953 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be
3954 assigned automatically to PCI devices. You can
3955 make the kernel exclude IRQs of your ISA cards
3957 pirqaddr=0xAAAAA [X86] Specify the physical address
3958 of the PIRQ table (normally generated
3959 by the BIOS) if it is outside the
3960 F0000h-100000h range.
3961 lastbus=N [X86] Scan all buses thru bus #N. Can be
3962 useful if the kernel is unable to find your
3963 secondary buses and you want to tell it
3964 explicitly which ones they are.
3965 assign-busses [X86] Always assign all PCI bus
3966 numbers ourselves, overriding
3967 whatever the firmware may have done.
3968 usepirqmask [X86] Honor the possible IRQ mask stored
3969 in the BIOS $PIR table. This is needed on
3970 some systems with broken BIOSes, notably
3971 some HP Pavilion N5400 and Omnibook XE3
3972 notebooks. This will have no effect if ACPI
3973 IRQ routing is enabled.
3974 noacpi [X86] Do not use ACPI for IRQ routing
3975 or for PCI scanning.
3976 use_crs [X86] Use PCI host bridge window information
3977 from ACPI. On BIOSes from 2008 or later, this
3978 is enabled by default. If you need to use this,
3979 please report a bug.
3980 nocrs [X86] Ignore PCI host bridge windows from ACPI.
3981 If you need to use this, please report a bug.
3982 routeirq Do IRQ routing for all PCI devices.
3983 This is normally done in pci_enable_device(),
3984 so this option is a temporary workaround
3985 for broken drivers that don't call it.
3986 skip_isa_align [X86] do not align io start addr, so can
3987 handle more pci cards
3988 noearly [X86] Don't do any early type 1 scanning.
3989 This might help on some broken boards which
3990 machine check when some devices' config space
3991 is read. But various workarounds are disabled
3992 and some IOMMU drivers will not work.
3993 bfsort Sort PCI devices into breadth-first order.
3994 This sorting is done to get a device
3995 order compatible with older (<= 2.4) kernels.
3996 nobfsort Don't sort PCI devices into breadth-first order.
3997 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size)
3998 tuning and use the BIOS-configured MPS defaults.
3999 pcie_bus_safe Set every device's MPS to the largest value
4000 supported by all devices below the root complex.
4001 pcie_bus_perf Set device MPS to the largest allowable MPS
4002 based on its parent bus. Also set MRRS (Max
4003 Read Request Size) to the largest supported
4004 value (no larger than the MPS that the device
4005 or bus can support) for best performance.
4006 pcie_bus_peer2peer Set every device's MPS to 128B, which
4007 every device is guaranteed to support. This
4008 configuration allows peer-to-peer DMA between
4009 any pair of devices, possibly at the cost of
4010 reduced performance. This also guarantees
4011 that hot-added devices will work.
4012 cbiosize=nn[KMG] The fixed amount of bus space which is
4013 reserved for the CardBus bridge's IO window.
4014 The default value is 256 bytes.
4015 cbmemsize=nn[KMG] The fixed amount of bus space which is
4016 reserved for the CardBus bridge's memory
4017 window. The default value is 64 megabytes.
4020 [<order of align>@]<pci_dev>[; ...]
4021 Specifies alignment and device to reassign
4022 aligned memory resources. How to
4023 specify the device is described above.
4024 If <order of align> is not specified,
4025 PAGE_SIZE is used as alignment.
4026 A PCI-PCI bridge can be specified if resource
4027 windows need to be expanded.
4028 To specify the alignment for several
4029 instances of a device, the PCI vendor,
4030 device, subvendor, and subdevice may be
4031 specified, e.g., 12@pci:8086:9c22:103c:198f
4032 for 4096-byte alignment.
4033 ecrc= Enable/disable PCIe ECRC (transaction layer
4034 end-to-end CRC checking).
4035 bios: Use BIOS/firmware settings. This is the
4039 hpiosize=nn[KMG] The fixed amount of bus space which is
4040 reserved for hotplug bridge's IO window.
4041 Default size is 256 bytes.
4042 hpmmiosize=nn[KMG] The fixed amount of bus space which is
4043 reserved for hotplug bridge's MMIO window.
4044 Default size is 2 megabytes.
4045 hpmmioprefsize=nn[KMG] The fixed amount of bus space which is
4046 reserved for hotplug bridge's MMIO_PREF window.
4047 Default size is 2 megabytes.
4048 hpmemsize=nn[KMG] The fixed amount of bus space which is
4049 reserved for hotplug bridge's MMIO and
4051 Default size is 2 megabytes.
4052 hpbussize=nn The minimum amount of additional bus numbers
4053 reserved for buses below a hotplug bridge.
4055 realloc= Enable/disable reallocating PCI bridge resources
4056 if allocations done by BIOS are too small to
4057 accommodate resources required by all child
4059 off: Turn realloc off
4061 realloc same as realloc=on
4062 noari do not use PCIe ARI.
4063 noats [PCIE, Intel-IOMMU, AMD-IOMMU]
4064 do not use PCIe ATS (and IOMMU device IOTLB).
4065 pcie_scan_all Scan all possible PCIe devices. Otherwise we
4066 only look for one device below a PCIe downstream
4068 big_root_window Try to add a big 64bit memory window to the PCIe
4069 root complex on AMD CPUs. Some GFX hardware
4070 can resize a BAR to allow access to all VRAM.
4071 Adding the window is slightly risky (it may
4072 conflict with unreported devices), so this
4074 disable_acs_redir=<pci_dev>[; ...]
4075 Specify one or more PCI devices (in the format
4076 specified above) separated by semicolons.
4077 Each device specified will have the PCI ACS
4078 redirect capabilities forced off which will
4079 allow P2P traffic between devices through
4080 bridges without forcing it upstream. Note:
4081 this removes isolation between devices and
4082 may put more devices in an IOMMU group.
4083 force_floating [S390] Force usage of floating interrupts.
4084 nomio [S390] Do not use MIO instructions.
4085 norid [S390] ignore the RID field and force use of
4086 one PCI domain per PCI function
4088 pcie_aspm= [PCIE] Forcibly enable or disable PCIe Active State Power
4091 force Enable ASPM even on devices that claim not to support it.
4092 WARNING: Forcing ASPM on may cause system lockups.
4094 pcie_ports= [PCIE] PCIe port services handling:
4095 native Use native PCIe services (PME, AER, DPC, PCIe hotplug)
4096 even if the platform doesn't give the OS permission to
4097 use them. This may cause conflicts if the platform
4098 also tries to use these services.
4099 dpc-native Use native PCIe service for DPC only. May
4100 cause conflicts if firmware uses AER or DPC.
4101 compat Disable native PCIe services (PME, AER, DPC, PCIe
4104 pcie_port_pm= [PCIE] PCIe port power management handling:
4105 off Disable power management of all PCIe ports
4106 force Forcibly enable power management of all PCIe ports
4108 pcie_pme= [PCIE,PM] Native PCIe PME signaling options:
4109 nomsi Do not use MSI for native PCIe PME signaling (this makes
4110 all PCIe root ports use INTx for all services).
4112 pcmv= [HW,PCMCIA] BadgePAD 4
4116 Keep all power-domains already enabled by bootloader on,
4117 even if no driver has claimed them. This is useful
4118 for debug and development, but should not be
4119 needed on a platform with proper driver support.
4122 See Documentation/admin-guide/blockdev/paride.rst.
4124 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at
4127 See arch/parisc/kernel/pdc_chassis.c
4129 percpu_alloc= Select which percpu first chunk allocator to use.
4130 Currently supported values are "embed" and "page".
4131 Archs may support subset or none of the selections.
4132 See comments in mm/percpu.c for details on each
4133 allocator. This parameter is primarily for debugging
4134 and performance comparison.
4137 See Documentation/admin-guide/blockdev/paride.rst.
4140 See Documentation/admin-guide/blockdev/paride.rst.
4142 pirq= [SMP,APIC] Manual mp-table setup
4143 See Documentation/x86/i386/IO-APIC.rst.
4145 plip= [PPT,NET] Parallel port network link
4146 Format: { parport<nr> | timid | 0 }
4147 See also Documentation/admin-guide/parport.rst.
4149 pmtmr= [X86] Manual setup of pmtmr I/O Port.
4150 Override pmtimer IOPort with a hex value.
4153 pm_debug_messages [SUSPEND,KNL]
4154 Enable suspend/resume debug messages during boot up.
4157 Enable PNP debug messages (depends on the
4158 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time
4159 via /sys/module/pnp/parameters/debug. We always show
4160 current resource usage; turning this on also shows
4161 possible settings and some assignment information.
4167 { on | off | curr | res | no-curr | no-res }
4170 [ISAPNP] Exclude IRQs for the autoconfiguration
4173 [ISAPNP] Exclude DMAs for the autoconfiguration
4175 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration
4176 Ranges are in pairs (I/O port base and size).
4179 [ISAPNP] Exclude memory regions for the
4181 Ranges are in pairs (memory base and size).
4183 ports= [IP_VS_FTP] IPVS ftp helper module
4185 Up to 8 (IP_VS_APP_MAX_PORTS) ports
4187 Format: <port>,<port>....
4189 powersave=off [PPC] This option disables power saving features.
4190 It specifically disables cpuidle and sets the
4191 platform machine description specific power_save
4192 function to NULL. On Idle the CPU just reduces
4195 ppc_strict_facility_enable
4196 [PPC] This option catches any kernel floating point,
4197 Altivec, VSX and SPE outside of regions specifically
4198 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()).
4199 There is some performance impact when enabling this.
4203 Disable Hardware Transactional Memory
4206 Select preemption mode if you have CONFIG_PREEMPT_DYNAMIC
4207 none - Limited to cond_resched() calls
4208 voluntary - Limited to cond_resched() and might_sleep() calls
4209 full - Any section that isn't explicitly preempt disabled
4210 can be preempted anytime.
4212 print-fatal-signals=
4213 [KNL] debug: print fatal signals
4215 If enabled, warn about various signal handling
4216 related application anomalies: too many signals,
4217 too many POSIX.1 timers, fatal signals causing a
4220 If you hit the warning due to signal overflow,
4221 you might want to try "ulimit -i unlimited".
4225 printk.always_kmsg_dump=
4226 Trigger kmsg_dump for cases other than kernel oops or
4228 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
4231 printk.console_no_auto_verbose=
4232 Disable console loglevel raise on oops, panic
4233 or lockdep-detected issues (only if lock debug is on).
4234 With an exception to setups with low baudrate on
4235 serial console, keeping this 0 is a good choice
4236 in order to provide more debug information.
4238 default: 0 (auto_verbose is enabled)
4240 printk.devkmsg={on,off,ratelimit}
4241 Control writing to /dev/kmsg.
4242 on - unlimited logging to /dev/kmsg from userspace
4243 off - logging to /dev/kmsg disabled
4244 ratelimit - ratelimit the logging
4247 printk.time= Show timing data prefixed to each printk message line
4248 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
4250 processor.max_cstate= [HW,ACPI]
4251 Limit processor to maximum C-state
4252 max_cstate=9 overrides any DMI blacklist limit.
4254 processor.nocst [HW,ACPI]
4255 Ignore the _CST method to determine C-states,
4256 instead using the legacy FADT method
4258 profile= [KNL] Enable kernel profiling via /proc/profile
4259 Format: [<profiletype>,]<number>
4260 Param: <profiletype>: "schedule", "sleep", or "kvm"
4261 [defaults to kernel profiling]
4262 Param: "schedule" - profile schedule points.
4263 Param: "sleep" - profile D-state sleeping (millisecs).
4264 Requires CONFIG_SCHEDSTATS
4265 Param: "kvm" - profile VM exits.
4266 Param: <number> - step/bucket size as a power of 2 for
4267 statistical time based profiling.
4269 prompt_ramdisk= [RAM] [Deprecated]
4271 prot_virt= [S390] enable hosting protected virtual machines
4272 isolated from the hypervisor (if hardware supports
4276 psi= [KNL] Enable or disable pressure stall information
4280 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to
4281 probe for; one of (bare|imps|exps|lifebook|any).
4282 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports
4284 psmouse.resetafter= [HW,MOUSE]
4285 Try to reset the device after so many bad packets
4288 [HW,MOUSE] Set desired mouse resolution, in dpi.
4289 psmouse.smartscroll=
4290 [HW,MOUSE] Controls Logitech smartscroll autorepeat.
4291 0 = disabled, 1 = enabled (default).
4293 pstore.backend= Specify the name of the pstore backend to use
4296 See Documentation/admin-guide/blockdev/paride.rst.
4298 pti= [X86-64] Control Page Table Isolation of user and
4299 kernel address spaces. Disabling this feature
4300 removes hardening, but improves performance of
4301 system calls and interrupts.
4303 on - unconditionally enable
4304 off - unconditionally disable
4305 auto - kernel detects whether your CPU model is
4306 vulnerable to issues that PTI mitigates
4308 Not specifying this option is equivalent to pti=auto.
4311 Equivalent to pti=off
4314 [KNL] Number of legacy pty's. Overwrites compiled-in
4317 quiet [KNL] Disable most log messages
4322 See Documentation/admin-guide/md.rst.
4324 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes
4325 See Documentation/admin-guide/blockdev/ramdisk.rst.
4327 ramdisk_start= [RAM] RAM disk image start address
4329 random.trust_cpu={on,off}
4330 [KNL] Enable or disable trusting the use of the
4331 CPU's random number generator (if available) to
4332 fully seed the kernel's CRNG. Default is controlled
4333 by CONFIG_RANDOM_TRUST_CPU.
4335 randomize_kstack_offset=
4336 [KNL] Enable or disable kernel stack offset
4337 randomization, which provides roughly 5 bits of
4338 entropy, frustrating memory corruption attacks
4339 that depend on stack address determinism or
4340 cross-syscall address exposures. This is only
4341 available on architectures that have defined
4342 CONFIG_HAVE_ARCH_RANDOMIZE_KSTACK_OFFSET.
4343 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
4344 Default is CONFIG_RANDOMIZE_KSTACK_OFFSET_DEFAULT.
4346 ras=option[,option,...] [KNL] RAS-specific options
4349 Disable the Correctable Errors Collector,
4350 see CONFIG_RAS_CEC help text.
4353 The argument is a cpu list, as described above.
4355 In kernels built with CONFIG_RCU_NOCB_CPU=y, set
4356 the specified list of CPUs to be no-callback CPUs.
4357 Invocation of these CPUs' RCU callbacks will be
4358 offloaded to "rcuox/N" kthreads created for that
4359 purpose, where "x" is "p" for RCU-preempt, and
4360 "s" for RCU-sched, and "N" is the CPU number.
4361 This reduces OS jitter on the offloaded CPUs,
4362 which can be useful for HPC and real-time
4363 workloads. It can also improve energy efficiency
4364 for asymmetric multiprocessors.
4367 Rather than requiring that offloaded CPUs
4368 (specified by rcu_nocbs= above) explicitly
4369 awaken the corresponding "rcuoN" kthreads,
4370 make these kthreads poll for callbacks.
4371 This improves the real-time response for the
4372 offloaded CPUs by relieving them of the need to
4373 wake up the corresponding kthread, but degrades
4374 energy efficiency by requiring that the kthreads
4375 periodically wake up to do the polling.
4377 rcutree.blimit= [KNL]
4378 Set maximum number of finished RCU callbacks to
4379 process in one batch.
4381 rcutree.dump_tree= [KNL]
4382 Dump the structure of the rcu_node combining tree
4383 out at early boot. This is used for diagnostic
4384 purposes, to verify correct tree setup.
4386 rcutree.gp_cleanup_delay= [KNL]
4387 Set the number of jiffies to delay each step of
4388 RCU grace-period cleanup.
4390 rcutree.gp_init_delay= [KNL]
4391 Set the number of jiffies to delay each step of
4392 RCU grace-period initialization.
4394 rcutree.gp_preinit_delay= [KNL]
4395 Set the number of jiffies to delay each step of
4396 RCU grace-period pre-initialization, that is,
4397 the propagation of recent CPU-hotplug changes up
4398 the rcu_node combining tree.
4400 rcutree.use_softirq= [KNL]
4401 If set to zero, move all RCU_SOFTIRQ processing to
4402 per-CPU rcuc kthreads. Defaults to a non-zero
4403 value, meaning that RCU_SOFTIRQ is used by default.
4404 Specify rcutree.use_softirq=0 to use rcuc kthreads.
4406 But note that CONFIG_PREEMPT_RT=y kernels disable
4407 this kernel boot parameter, forcibly setting it
4410 rcutree.rcu_fanout_exact= [KNL]
4411 Disable autobalancing of the rcu_node combining
4412 tree. This is used by rcutorture, and might
4413 possibly be useful for architectures having high
4414 cache-to-cache transfer latencies.
4416 rcutree.rcu_fanout_leaf= [KNL]
4417 Change the number of CPUs assigned to each
4418 leaf rcu_node structure. Useful for very
4419 large systems, which will choose the value 64,
4420 and for NUMA systems with large remote-access
4421 latencies, which will choose a value aligned
4422 with the appropriate hardware boundaries.
4424 rcutree.rcu_min_cached_objs= [KNL]
4425 Minimum number of objects which are cached and
4426 maintained per one CPU. Object size is equal
4427 to PAGE_SIZE. The cache allows to reduce the
4428 pressure to page allocator, also it makes the
4429 whole algorithm to behave better in low memory
4432 rcutree.rcu_delay_page_cache_fill_msec= [KNL]
4433 Set the page-cache refill delay (in milliseconds)
4434 in response to low-memory conditions. The range
4435 of permitted values is in the range 0:100000.
4437 rcutree.jiffies_till_first_fqs= [KNL]
4438 Set delay from grace-period initialization to
4439 first attempt to force quiescent states.
4440 Units are jiffies, minimum value is zero,
4441 and maximum value is HZ.
4443 rcutree.jiffies_till_next_fqs= [KNL]
4444 Set delay between subsequent attempts to force
4445 quiescent states. Units are jiffies, minimum
4446 value is one, and maximum value is HZ.
4448 rcutree.jiffies_till_sched_qs= [KNL]
4449 Set required age in jiffies for a
4450 given grace period before RCU starts
4451 soliciting quiescent-state help from
4452 rcu_note_context_switch() and cond_resched().
4453 If not specified, the kernel will calculate
4454 a value based on the most recent settings
4455 of rcutree.jiffies_till_first_fqs
4456 and rcutree.jiffies_till_next_fqs.
4457 This calculated value may be viewed in
4458 rcutree.jiffies_to_sched_qs. Any attempt to set
4459 rcutree.jiffies_to_sched_qs will be cheerfully
4462 rcutree.kthread_prio= [KNL,BOOT]
4463 Set the SCHED_FIFO priority of the RCU per-CPU
4464 kthreads (rcuc/N). This value is also used for
4465 the priority of the RCU boost threads (rcub/N)
4466 and for the RCU grace-period kthreads (rcu_bh,
4467 rcu_preempt, and rcu_sched). If RCU_BOOST is
4468 set, valid values are 1-99 and the default is 1
4469 (the least-favored priority). Otherwise, when
4470 RCU_BOOST is not set, valid values are 0-99 and
4471 the default is zero (non-realtime operation).
4473 rcutree.rcu_nocb_gp_stride= [KNL]
4474 Set the number of NOCB callback kthreads in
4475 each group, which defaults to the square root
4476 of the number of CPUs. Larger numbers reduce
4477 the wakeup overhead on the global grace-period
4478 kthread, but increases that same overhead on
4479 each group's NOCB grace-period kthread.
4481 rcutree.qhimark= [KNL]
4482 Set threshold of queued RCU callbacks beyond which
4483 batch limiting is disabled.
4485 rcutree.qlowmark= [KNL]
4486 Set threshold of queued RCU callbacks below which
4487 batch limiting is re-enabled.
4489 rcutree.qovld= [KNL]
4490 Set threshold of queued RCU callbacks beyond which
4491 RCU's force-quiescent-state scan will aggressively
4492 enlist help from cond_resched() and sched IPIs to
4493 help CPUs more quickly reach quiescent states.
4494 Set to less than zero to make this be set based
4495 on rcutree.qhimark at boot time and to zero to
4496 disable more aggressive help enlistment.
4498 rcutree.rcu_idle_gp_delay= [KNL]
4499 Set wakeup interval for idle CPUs that have
4500 RCU callbacks (RCU_FAST_NO_HZ=y).
4502 rcutree.rcu_kick_kthreads= [KNL]
4503 Cause the grace-period kthread to get an extra
4504 wake_up() if it sleeps three times longer than
4505 it should at force-quiescent-state time.
4506 This wake_up() will be accompanied by a
4507 WARN_ONCE() splat and an ftrace_dump().
4509 rcutree.rcu_unlock_delay= [KNL]
4510 In CONFIG_RCU_STRICT_GRACE_PERIOD=y kernels,
4511 this specifies an rcu_read_unlock()-time delay
4512 in microseconds. This defaults to zero.
4513 Larger delays increase the probability of
4514 catching RCU pointer leaks, that is, buggy use
4515 of RCU-protected pointers after the relevant
4516 rcu_read_unlock() has completed.
4518 rcutree.sysrq_rcu= [KNL]
4519 Commandeer a sysrq key to dump out Tree RCU's
4520 rcu_node tree with an eye towards determining
4521 why a new grace period has not yet started.
4523 rcuscale.gp_async= [KNL]
4524 Measure performance of asynchronous
4525 grace-period primitives such as call_rcu().
4527 rcuscale.gp_async_max= [KNL]
4528 Specify the maximum number of outstanding
4529 callbacks per writer thread. When a writer
4530 thread exceeds this limit, it invokes the
4531 corresponding flavor of rcu_barrier() to allow
4532 previously posted callbacks to drain.
4534 rcuscale.gp_exp= [KNL]
4535 Measure performance of expedited synchronous
4536 grace-period primitives.
4538 rcuscale.holdoff= [KNL]
4539 Set test-start holdoff period. The purpose of
4540 this parameter is to delay the start of the
4541 test until boot completes in order to avoid
4544 rcuscale.kfree_rcu_test= [KNL]
4545 Set to measure performance of kfree_rcu() flooding.
4547 rcuscale.kfree_rcu_test_double= [KNL]
4548 Test the double-argument variant of kfree_rcu().
4549 If this parameter has the same value as
4550 rcuscale.kfree_rcu_test_single, both the single-
4551 and double-argument variants are tested.
4553 rcuscale.kfree_rcu_test_single= [KNL]
4554 Test the single-argument variant of kfree_rcu().
4555 If this parameter has the same value as
4556 rcuscale.kfree_rcu_test_double, both the single-
4557 and double-argument variants are tested.
4559 rcuscale.kfree_nthreads= [KNL]
4560 The number of threads running loops of kfree_rcu().
4562 rcuscale.kfree_alloc_num= [KNL]
4563 Number of allocations and frees done in an iteration.
4565 rcuscale.kfree_loops= [KNL]
4566 Number of loops doing rcuscale.kfree_alloc_num number
4567 of allocations and frees.
4569 rcuscale.nreaders= [KNL]
4570 Set number of RCU readers. The value -1 selects
4571 N, where N is the number of CPUs. A value
4572 "n" less than -1 selects N-n+1, where N is again
4573 the number of CPUs. For example, -2 selects N
4574 (the number of CPUs), -3 selects N+1, and so on.
4575 A value of "n" less than or equal to -N selects
4578 rcuscale.nwriters= [KNL]
4579 Set number of RCU writers. The values operate
4580 the same as for rcuscale.nreaders.
4581 N, where N is the number of CPUs
4583 rcuscale.perf_type= [KNL]
4584 Specify the RCU implementation to test.
4586 rcuscale.shutdown= [KNL]
4587 Shut the system down after performance tests
4588 complete. This is useful for hands-off automated
4591 rcuscale.verbose= [KNL]
4592 Enable additional printk() statements.
4594 rcuscale.writer_holdoff= [KNL]
4595 Write-side holdoff between grace periods,
4596 in microseconds. The default of zero says
4599 rcutorture.fqs_duration= [KNL]
4600 Set duration of force_quiescent_state bursts
4603 rcutorture.fqs_holdoff= [KNL]
4604 Set holdoff time within force_quiescent_state bursts
4607 rcutorture.fqs_stutter= [KNL]
4608 Set wait time between force_quiescent_state bursts
4611 rcutorture.fwd_progress= [KNL]
4612 Enable RCU grace-period forward-progress testing
4613 for the types of RCU supporting this notion.
4615 rcutorture.fwd_progress_div= [KNL]
4616 Specify the fraction of a CPU-stall-warning
4617 period to do tight-loop forward-progress testing.
4619 rcutorture.fwd_progress_holdoff= [KNL]
4620 Number of seconds to wait between successive
4621 forward-progress tests.
4623 rcutorture.fwd_progress_need_resched= [KNL]
4624 Enclose cond_resched() calls within checks for
4625 need_resched() during tight-loop forward-progress
4628 rcutorture.gp_cond= [KNL]
4629 Use conditional/asynchronous update-side
4630 primitives, if available.
4632 rcutorture.gp_exp= [KNL]
4633 Use expedited update-side primitives, if available.
4635 rcutorture.gp_normal= [KNL]
4636 Use normal (non-expedited) asynchronous
4637 update-side primitives, if available.
4639 rcutorture.gp_sync= [KNL]
4640 Use normal (non-expedited) synchronous
4641 update-side primitives, if available. If all
4642 of rcutorture.gp_cond=, rcutorture.gp_exp=,
4643 rcutorture.gp_normal=, and rcutorture.gp_sync=
4644 are zero, rcutorture acts as if is interpreted
4645 they are all non-zero.
4647 rcutorture.irqreader= [KNL]
4648 Run RCU readers from irq handlers, or, more
4649 accurately, from a timer handler. Not all RCU
4650 flavors take kindly to this sort of thing.
4652 rcutorture.leakpointer= [KNL]
4653 Leak an RCU-protected pointer out of the reader.
4654 This can of course result in splats, and is
4655 intended to test the ability of things like
4656 CONFIG_RCU_STRICT_GRACE_PERIOD=y to detect
4659 rcutorture.n_barrier_cbs= [KNL]
4660 Set callbacks/threads for rcu_barrier() testing.
4662 rcutorture.nfakewriters= [KNL]
4663 Set number of concurrent RCU writers. These just
4664 stress RCU, they don't participate in the actual
4665 test, hence the "fake".
4667 rcutorture.nocbs_nthreads= [KNL]
4668 Set number of RCU callback-offload togglers.
4669 Zero (the default) disables toggling.
4671 rcutorture.nocbs_toggle= [KNL]
4672 Set the delay in milliseconds between successive
4673 callback-offload toggling attempts.
4675 rcutorture.nreaders= [KNL]
4676 Set number of RCU readers. The value -1 selects
4677 N-1, where N is the number of CPUs. A value
4678 "n" less than -1 selects N-n-2, where N is again
4679 the number of CPUs. For example, -2 selects N
4680 (the number of CPUs), -3 selects N+1, and so on.
4682 rcutorture.object_debug= [KNL]
4683 Enable debug-object double-call_rcu() testing.
4685 rcutorture.onoff_holdoff= [KNL]
4686 Set time (s) after boot for CPU-hotplug testing.
4688 rcutorture.onoff_interval= [KNL]
4689 Set time (jiffies) between CPU-hotplug operations,
4690 or zero to disable CPU-hotplug testing.
4692 rcutorture.read_exit= [KNL]
4693 Set the number of read-then-exit kthreads used
4694 to test the interaction of RCU updaters and
4695 task-exit processing.
4697 rcutorture.read_exit_burst= [KNL]
4698 The number of times in a given read-then-exit
4699 episode that a set of read-then-exit kthreads
4702 rcutorture.read_exit_delay= [KNL]
4703 The delay, in seconds, between successive
4704 read-then-exit testing episodes.
4706 rcutorture.shuffle_interval= [KNL]
4707 Set task-shuffle interval (s). Shuffling tasks
4708 allows some CPUs to go into dyntick-idle mode
4709 during the rcutorture test.
4711 rcutorture.shutdown_secs= [KNL]
4712 Set time (s) after boot system shutdown. This
4713 is useful for hands-off automated testing.
4715 rcutorture.stall_cpu= [KNL]
4716 Duration of CPU stall (s) to test RCU CPU stall
4717 warnings, zero to disable.
4719 rcutorture.stall_cpu_block= [KNL]
4720 Sleep while stalling if set. This will result
4721 in warnings from preemptible RCU in addition
4722 to any other stall-related activity.
4724 rcutorture.stall_cpu_holdoff= [KNL]
4725 Time to wait (s) after boot before inducing stall.
4727 rcutorture.stall_cpu_irqsoff= [KNL]
4728 Disable interrupts while stalling if set.
4730 rcutorture.stall_gp_kthread= [KNL]
4731 Duration (s) of forced sleep within RCU
4732 grace-period kthread to test RCU CPU stall
4733 warnings, zero to disable. If both stall_cpu
4734 and stall_gp_kthread are specified, the
4735 kthread is starved first, then the CPU.
4737 rcutorture.stat_interval= [KNL]
4738 Time (s) between statistics printk()s.
4740 rcutorture.stutter= [KNL]
4741 Time (s) to stutter testing, for example, specifying
4742 five seconds causes the test to run for five seconds,
4743 wait for five seconds, and so on. This tests RCU's
4744 ability to transition abruptly to and from idle.
4746 rcutorture.test_boost= [KNL]
4747 Test RCU priority boosting? 0=no, 1=maybe, 2=yes.
4748 "Maybe" means test if the RCU implementation
4749 under test support RCU priority boosting.
4751 rcutorture.test_boost_duration= [KNL]
4752 Duration (s) of each individual boost test.
4754 rcutorture.test_boost_interval= [KNL]
4755 Interval (s) between each boost test.
4757 rcutorture.test_no_idle_hz= [KNL]
4758 Test RCU's dyntick-idle handling. See also the
4759 rcutorture.shuffle_interval parameter.
4761 rcutorture.torture_type= [KNL]
4762 Specify the RCU implementation to test.
4764 rcutorture.verbose= [KNL]
4765 Enable additional printk() statements.
4767 rcupdate.rcu_cpu_stall_ftrace_dump= [KNL]
4768 Dump ftrace buffer after reporting RCU CPU
4771 rcupdate.rcu_cpu_stall_suppress= [KNL]
4772 Suppress RCU CPU stall warning messages.
4774 rcupdate.rcu_cpu_stall_suppress_at_boot= [KNL]
4775 Suppress RCU CPU stall warning messages and
4776 rcutorture writer stall warnings that occur
4777 during early boot, that is, during the time
4778 before the init task is spawned.
4780 rcupdate.rcu_cpu_stall_timeout= [KNL]
4781 Set timeout for RCU CPU stall warning messages.
4783 rcupdate.rcu_expedited= [KNL]
4784 Use expedited grace-period primitives, for
4785 example, synchronize_rcu_expedited() instead
4786 of synchronize_rcu(). This reduces latency,
4787 but can increase CPU utilization, degrade
4788 real-time latency, and degrade energy efficiency.
4789 No effect on CONFIG_TINY_RCU kernels.
4791 rcupdate.rcu_normal= [KNL]
4792 Use only normal grace-period primitives,
4793 for example, synchronize_rcu() instead of
4794 synchronize_rcu_expedited(). This improves
4795 real-time latency, CPU utilization, and
4796 energy efficiency, but can expose users to
4797 increased grace-period latency. This parameter
4798 overrides rcupdate.rcu_expedited. No effect on
4799 CONFIG_TINY_RCU kernels.
4801 rcupdate.rcu_normal_after_boot= [KNL]
4802 Once boot has completed (that is, after
4803 rcu_end_inkernel_boot() has been invoked), use
4804 only normal grace-period primitives. No effect
4805 on CONFIG_TINY_RCU kernels.
4807 But note that CONFIG_PREEMPT_RT=y kernels enables
4808 this kernel boot parameter, forcibly setting
4809 it to the value one, that is, converting any
4810 post-boot attempt at an expedited RCU grace
4811 period to instead use normal non-expedited
4812 grace-period processing.
4814 rcupdate.rcu_task_ipi_delay= [KNL]
4815 Set time in jiffies during which RCU tasks will
4816 avoid sending IPIs, starting with the beginning
4817 of a given grace period. Setting a large
4818 number avoids disturbing real-time workloads,
4819 but lengthens grace periods.
4821 rcupdate.rcu_task_stall_timeout= [KNL]
4822 Set timeout in jiffies for RCU task stall warning
4823 messages. Disable with a value less than or equal
4826 rcupdate.rcu_self_test= [KNL]
4827 Run the RCU early boot self tests
4831 Run specified binary instead of /init from the ramdisk,
4832 used for early userspace startup. See initrd.
4835 force - Override the decision by the kernel to hide the
4836 advertisement of RDRAND support (this affects
4837 certain AMD processors because of buggy BIOS
4838 support, specifically around the suspend/resume
4842 Turn on/off individual RDT features. List is:
4843 cmt, mbmtotal, mbmlocal, l3cat, l3cdp, l2cat, l2cdp,
4845 E.g. to turn on cmt and turn off mba use:
4849 Format (x86 or x86_64):
4850 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] | d[efault] \
4852 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \
4854 Where reboot_mode is one of warm (soft) or cold (hard) or gpio
4855 (prefix with 'panic_' to set mode for panic
4857 reboot_type is one of bios, acpi, kbd, triple, efi, or pci,
4858 reboot_force is either force or not specified,
4859 reboot_cpu is s[mp]#### with #### being the processor
4860 to be used for rebooting.
4862 refscale.holdoff= [KNL]
4863 Set test-start holdoff period. The purpose of
4864 this parameter is to delay the start of the
4865 test until boot completes in order to avoid
4868 refscale.loops= [KNL]
4869 Set the number of loops over the synchronization
4870 primitive under test. Increasing this number
4871 reduces noise due to loop start/end overhead,
4872 but the default has already reduced the per-pass
4873 noise to a handful of picoseconds on ca. 2020
4876 refscale.nreaders= [KNL]
4877 Set number of readers. The default value of -1
4878 selects N, where N is roughly 75% of the number
4879 of CPUs. A value of zero is an interesting choice.
4881 refscale.nruns= [KNL]
4882 Set number of runs, each of which is dumped onto
4885 refscale.readdelay= [KNL]
4886 Set the read-side critical-section duration,
4887 measured in microseconds.
4889 refscale.scale_type= [KNL]
4890 Specify the read-protection implementation to test.
4892 refscale.shutdown= [KNL]
4893 Shut down the system at the end of the performance
4894 test. This defaults to 1 (shut it down) when
4895 refscale is built into the kernel and to 0 (leave
4896 it running) when refscale is built as a module.
4898 refscale.verbose= [KNL]
4899 Enable additional printk() statements.
4901 refscale.verbose_batched= [KNL]
4902 Batch the additional printk() statements. If zero
4903 (the default) or negative, print everything. Otherwise,
4904 print every Nth verbose statement, where N is the value
4908 [KNL, SMP] Set scheduler's default relax_domain_level.
4909 See Documentation/admin-guide/cgroup-v1/cpusets.rst.
4911 reserve= [KNL,BUGS] Force kernel to ignore I/O ports or memory
4912 Format: <base1>,<size1>[,<base2>,<size2>,...]
4913 Reserve I/O ports or memory so the kernel won't use
4914 them. If <base> is less than 0x10000, the region
4915 is assumed to be I/O ports; otherwise it is memory.
4917 reservetop= [X86-32]
4919 Reserves a hole at the top of the kernel virtual
4922 reset_devices [KNL] Force drivers to reset the underlying device
4923 during initialization.
4926 Specify the partition device for software suspend
4928 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>}
4930 resume_offset= [SWSUSP]
4931 Specify the offset from the beginning of the partition
4932 given by "resume=" at which the swap header is located,
4933 in <PAGE_SIZE> units (needed only for swap files).
4934 See Documentation/power/swsusp-and-swap-files.rst
4936 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to
4937 read the resume files
4939 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up.
4940 Useful for devices that are detected asynchronously
4941 (e.g. USB and MMC devices).
4943 hibernate= [HIBERNATION]
4944 noresume Don't check if there's a hibernation image
4945 present during boot.
4946 nocompress Don't compress/decompress hibernation images.
4947 no Disable hibernation and resume.
4948 protect_image Turn on image protection during restoration
4949 (that will set all pages holding image data
4950 during restoration read-only).
4952 retain_initrd [RAM] Keep initrd memory after extraction
4954 rfkill.default_state=
4955 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm,
4956 etc. communication is blocked by default.
4959 rfkill.master_switch_mode=
4960 0 The "airplane mode" button does nothing.
4961 1 The "airplane mode" button toggles between everything
4962 blocked and the previous configuration.
4963 2 The "airplane mode" button toggles between everything
4964 blocked and everything unblocked.
4966 rhash_entries= [KNL,NET]
4967 Set number of hash buckets for route cache
4970 [KNL] Disable ring 3 MONITOR/MWAIT feature on supported
4973 ro [KNL] Mount root device read-only on boot
4976 on Mark read-only kernel memory as read-only (default).
4977 off Leave read-only kernel memory writable for debugging.
4980 Enable the uart passthrough on the designated usb port
4981 on Rockchip SoCs. When active, the signals of the
4982 debug-uart get routed to the D+ and D- pins of the usb
4983 port and the regular usb controller gets disabled.
4985 root= [KNL] Root filesystem
4986 See name_to_dev_t comment in init/do_mounts.c.
4988 rootdelay= [KNL] Delay (in seconds) to pause before attempting to
4989 mount the root filesystem
4991 rootflags= [KNL] Set root filesystem mount option string
4993 rootfstype= [KNL] Set root filesystem type
4995 rootwait [KNL] Wait (indefinitely) for root device to show up.
4996 Useful for devices that are detected asynchronously
4997 (e.g. USB and MMC devices).
4999 rproc_mem=nn[KMG][@address]
5000 [KNL,ARM,CMA] Remoteproc physical memory block.
5001 Memory area to be used by remote processor image,
5004 rw [KNL] Mount root device read-write on boot
5006 S [KNL] Run init in single mode
5008 s390_iommu= [HW,S390]
5009 Set s390 IOTLB flushing mode
5011 With strict flushing every unmap operation will result in
5012 an IOTLB flush. Default is lazy flushing before reuse,
5015 s390_iommu_aperture= [KNL,S390]
5016 Specifies the size of the per device DMA address space
5017 accessible through the DMA and IOMMU APIs as a decimal
5018 factor of the size of main memory.
5019 The default is 1 meaning that one can concurrently use
5020 as many DMA addresses as physical memory is installed,
5021 if supported by hardware, and thus map all of memory
5022 once. With a value of 2 one can map all of memory twice
5023 and so on. As a special case a factor of 0 imposes no
5024 restrictions other than those given by hardware at the
5025 cost of significant additional memory use for tables.
5028 See drivers/net/irda/sa1100_ir.c.
5030 sched_verbose [KNL] Enables verbose scheduler debug messages.
5032 schedstats= [KNL,X86] Enable or disable scheduled statistics.
5033 Allowed values are enable and disable. This feature
5034 incurs a small amount of overhead in the scheduler
5035 but is useful for debugging and performance tuning.
5037 sched_thermal_decay_shift=
5038 [KNL, SMP] Set a decay shift for scheduler thermal
5039 pressure signal. Thermal pressure signal follows the
5040 default decay period of other scheduler pelt
5041 signals(usually 32 ms but configurable). Setting
5042 sched_thermal_decay_shift will left shift the decay
5043 period for the thermal pressure signal by the shift
5045 i.e. with the default pelt decay period of 32 ms
5046 sched_thermal_decay_shift thermal pressure decay pr
5050 Format: integer between 0 and 10
5053 scftorture.holdoff= [KNL]
5054 Number of seconds to hold off before starting
5055 test. Defaults to zero for module insertion and
5056 to 10 seconds for built-in smp_call_function()
5059 scftorture.longwait= [KNL]
5060 Request ridiculously long waits randomly selected
5061 up to the chosen limit in seconds. Zero (the
5062 default) disables this feature. Please note
5063 that requesting even small non-zero numbers of
5064 seconds can result in RCU CPU stall warnings,
5065 softlockup complaints, and so on.
5067 scftorture.nthreads= [KNL]
5068 Number of kthreads to spawn to invoke the
5069 smp_call_function() family of functions.
5070 The default of -1 specifies a number of kthreads
5071 equal to the number of CPUs.
5073 scftorture.onoff_holdoff= [KNL]
5074 Number seconds to wait after the start of the
5075 test before initiating CPU-hotplug operations.
5077 scftorture.onoff_interval= [KNL]
5078 Number seconds to wait between successive
5079 CPU-hotplug operations. Specifying zero (which
5080 is the default) disables CPU-hotplug operations.
5082 scftorture.shutdown_secs= [KNL]
5083 The number of seconds following the start of the
5084 test after which to shut down the system. The
5085 default of zero avoids shutting down the system.
5086 Non-zero values are useful for automated tests.
5088 scftorture.stat_interval= [KNL]
5089 The number of seconds between outputting the
5090 current test statistics to the console. A value
5091 of zero disables statistics output.
5093 scftorture.stutter_cpus= [KNL]
5094 The number of jiffies to wait between each change
5095 to the set of CPUs under test.
5097 scftorture.use_cpus_read_lock= [KNL]
5098 Use use_cpus_read_lock() instead of the default
5099 preempt_disable() to disable CPU hotplug
5100 while invoking one of the smp_call_function*()
5103 scftorture.verbose= [KNL]
5104 Enable additional printk() statements.
5106 scftorture.weight_single= [KNL]
5107 The probability weighting to use for the
5108 smp_call_function_single() function with a zero
5109 "wait" parameter. A value of -1 selects the
5110 default if all other weights are -1. However,
5111 if at least one weight has some other value, a
5112 value of -1 will instead select a weight of zero.
5114 scftorture.weight_single_wait= [KNL]
5115 The probability weighting to use for the
5116 smp_call_function_single() function with a
5117 non-zero "wait" parameter. See weight_single.
5119 scftorture.weight_many= [KNL]
5120 The probability weighting to use for the
5121 smp_call_function_many() function with a zero
5122 "wait" parameter. See weight_single.
5123 Note well that setting a high probability for
5124 this weighting can place serious IPI load
5127 scftorture.weight_many_wait= [KNL]
5128 The probability weighting to use for the
5129 smp_call_function_many() function with a
5130 non-zero "wait" parameter. See weight_single
5133 scftorture.weight_all= [KNL]
5134 The probability weighting to use for the
5135 smp_call_function_all() function with a zero
5136 "wait" parameter. See weight_single and
5139 scftorture.weight_all_wait= [KNL]
5140 The probability weighting to use for the
5141 smp_call_function_all() function with a
5142 non-zero "wait" parameter. See weight_single
5145 skew_tick= [KNL] Offset the periodic timer tick per cpu to mitigate
5146 xtime_lock contention on larger systems, and/or RCU lock
5147 contention on all systems with CONFIG_MAXSMP set.
5148 Format: { "0" | "1" }
5149 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1"
5151 Note: increases power consumption, thus should only be
5152 enabled if running jitter sensitive (HPC/RT) workloads.
5154 security= [SECURITY] Choose a legacy "major" security module to
5155 enable at boot. This has been deprecated by the
5158 selinux= [SELINUX] Disable or enable SELinux at boot time.
5159 Format: { "0" | "1" }
5160 See security/selinux/Kconfig help text.
5165 apparmor= [APPARMOR] Disable or enable AppArmor at boot time
5166 Format: { "0" | "1" }
5167 See security/apparmor/Kconfig help text
5170 Default value is set via kernel config option.
5172 serialnumber [BUGS=X86-32]
5175 Maximal number of shapers.
5183 Enable merging of slabs with similar size when the
5184 kernel is built without CONFIG_SLAB_MERGE_DEFAULT.
5187 Disable merging of slabs with similar size. May be
5188 necessary if there is some reason to distinguish
5189 allocs to different slabs, especially in hardened
5190 environments where the risk of heap overflows and
5191 layout control by attackers can usually be
5192 frustrated by disabling merging. This will reduce
5193 most of the exposure of a heap attack to a single
5194 cache (risks via metadata attacks are mostly
5195 unchanged). Debug options disable merging on their
5197 For more information see Documentation/vm/slub.rst.
5199 slab_max_order= [MM, SLAB]
5200 Determines the maximum allowed order for slabs.
5201 A high setting may cause OOMs due to memory
5202 fragmentation. Defaults to 1 for systems with
5203 more than 32MB of RAM, 0 otherwise.
5205 slub_debug[=options[,slabs][;[options[,slabs]]...] [MM, SLUB]
5206 Enabling slub_debug allows one to determine the
5207 culprit if slab objects become corrupted. Enabling
5208 slub_debug can create guard zones around objects and
5209 may poison objects when not in use. Also tracks the
5210 last alloc / free. For more information see
5211 Documentation/vm/slub.rst.
5213 slub_max_order= [MM, SLUB]
5214 Determines the maximum allowed order for slabs.
5215 A high setting may cause OOMs due to memory
5216 fragmentation. For more information see
5217 Documentation/vm/slub.rst.
5219 slub_min_objects= [MM, SLUB]
5220 The minimum number of objects per slab. SLUB will
5221 increase the slab order up to slub_max_order to
5222 generate a sufficiently large slab able to contain
5223 the number of objects indicated. The higher the number
5224 of objects the smaller the overhead of tracking slabs
5225 and the less frequently locks need to be acquired.
5226 For more information see Documentation/vm/slub.rst.
5228 slub_min_order= [MM, SLUB]
5229 Determines the minimum page order for slabs. Must be
5230 lower than slub_max_order.
5231 For more information see Documentation/vm/slub.rst.
5233 slub_merge [MM, SLUB]
5234 Same with slab_merge.
5236 slub_nomerge [MM, SLUB]
5237 Same with slab_nomerge. This is supported for legacy.
5238 See slab_nomerge for more information.
5241 Format: <io1>[,<io2>[,...,<io8>]]
5243 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices
5244 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port
5245 smsc-ircc2.ircc_sir= [HW] SIR base I/O port
5246 smsc-ircc2.ircc_fir= [HW] FIR base I/O port
5247 smsc-ircc2.ircc_irq= [HW] IRQ line
5248 smsc-ircc2.ircc_dma= [HW] DMA channel
5249 smsc-ircc2.ircc_transceiver= [HW] Transceiver type:
5250 0: Toshiba Satellite 1800 (GP data pin select)
5251 1: Fast pin select (default)
5254 smt [KNL,S390] Set the maximum number of threads (logical
5255 CPUs) to use per physical CPU on systems capable of
5256 symmetric multithreading (SMT). Will be capped to the
5257 actual hardware limit.
5259 Default: -1 (no limit)
5262 [KNL] Should the soft-lockup detector generate panics.
5265 A value of 1 instructs the soft-lockup detector
5266 to panic the machine when a soft-lockup occurs. It is
5267 also controlled by the kernel.softlockup_panic sysctl
5268 and CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC, which is the
5269 respective build-time switch to that functionality.
5271 softlockup_all_cpu_backtrace=
5272 [KNL] Should the soft-lockup detector generate
5273 backtraces on all cpus.
5276 sonypi.*= [HW] Sony Programmable I/O Control Device driver
5277 See Documentation/admin-guide/laptops/sonypi.rst
5279 spectre_v2= [X86] Control mitigation of Spectre variant 2
5280 (indirect branch speculation) vulnerability.
5281 The default operation protects the kernel from
5284 on - unconditionally enable, implies
5286 off - unconditionally disable, implies
5288 auto - kernel detects whether your CPU model is
5291 Selecting 'on' will, and 'auto' may, choose a
5292 mitigation method at run time according to the
5293 CPU, the available microcode, the setting of the
5294 CONFIG_RETPOLINE configuration option, and the
5295 compiler with which the kernel was built.
5297 Selecting 'on' will also enable the mitigation
5298 against user space to user space task attacks.
5300 Selecting 'off' will disable both the kernel and
5301 the user space protections.
5303 Specific mitigations can also be selected manually:
5305 retpoline - replace indirect branches
5306 retpoline,generic - google's original retpoline
5307 retpoline,amd - AMD-specific minimal thunk
5309 Not specifying this option is equivalent to
5313 [X86] Control mitigation of Spectre variant 2
5314 (indirect branch speculation) vulnerability between
5317 on - Unconditionally enable mitigations. Is
5318 enforced by spectre_v2=on
5320 off - Unconditionally disable mitigations. Is
5321 enforced by spectre_v2=off
5323 prctl - Indirect branch speculation is enabled,
5324 but mitigation can be enabled via prctl
5325 per thread. The mitigation control state
5326 is inherited on fork.
5329 - Like "prctl" above, but only STIBP is
5330 controlled per thread. IBPB is issued
5331 always when switching between different user
5335 - Same as "prctl" above, but all seccomp
5336 threads will enable the mitigation unless
5337 they explicitly opt out.
5340 - Like "seccomp" above, but only STIBP is
5341 controlled per thread. IBPB is issued
5342 always when switching between different
5343 user space processes.
5345 auto - Kernel selects the mitigation depending on
5346 the available CPU features and vulnerability.
5348 Default mitigation: "prctl"
5350 Not specifying this option is equivalent to
5351 spectre_v2_user=auto.
5353 spec_store_bypass_disable=
5354 [HW] Control Speculative Store Bypass (SSB) Disable mitigation
5355 (Speculative Store Bypass vulnerability)
5357 Certain CPUs are vulnerable to an exploit against a
5358 a common industry wide performance optimization known
5359 as "Speculative Store Bypass" in which recent stores
5360 to the same memory location may not be observed by
5361 later loads during speculative execution. The idea
5362 is that such stores are unlikely and that they can
5363 be detected prior to instruction retirement at the
5364 end of a particular speculation execution window.
5366 In vulnerable processors, the speculatively forwarded
5367 store can be used in a cache side channel attack, for
5368 example to read memory to which the attacker does not
5369 directly have access (e.g. inside sandboxed code).
5371 This parameter controls whether the Speculative Store
5372 Bypass optimization is used.
5374 On x86 the options are:
5376 on - Unconditionally disable Speculative Store Bypass
5377 off - Unconditionally enable Speculative Store Bypass
5378 auto - Kernel detects whether the CPU model contains an
5379 implementation of Speculative Store Bypass and
5380 picks the most appropriate mitigation. If the
5381 CPU is not vulnerable, "off" is selected. If the
5382 CPU is vulnerable the default mitigation is
5383 architecture and Kconfig dependent. See below.
5384 prctl - Control Speculative Store Bypass per thread
5385 via prctl. Speculative Store Bypass is enabled
5386 for a process by default. The state of the control
5387 is inherited on fork.
5388 seccomp - Same as "prctl" above, but all seccomp threads
5389 will disable SSB unless they explicitly opt out.
5391 Default mitigations:
5394 On powerpc the options are:
5396 on,auto - On Power8 and Power9 insert a store-forwarding
5397 barrier on kernel entry and exit. On Power7
5398 perform a software flush on kernel entry and
5402 Not specifying this option is equivalent to
5403 spec_store_bypass_disable=auto.
5405 spia_io_base= [HW,MTD]
5411 [X86] Enable split lock detection or bus lock detection
5413 When enabled (and if hardware support is present), atomic
5414 instructions that access data across cache line
5415 boundaries will result in an alignment check exception
5416 for split lock detection or a debug exception for
5421 warn - the kernel will emit rate-limited warnings
5422 about applications triggering the #AC
5423 exception or the #DB exception. This mode is
5424 the default on CPUs that support split lock
5425 detection or bus lock detection. Default
5426 behavior is by #AC if both features are
5427 enabled in hardware.
5429 fatal - the kernel will send SIGBUS to applications
5430 that trigger the #AC exception or the #DB
5431 exception. Default behavior is by #AC if
5432 both features are enabled in hardware.
5435 Set system wide rate limit to N bus locks
5436 per second for bus lock detection.
5439 N/A for split lock detection.
5442 If an #AC exception is hit in the kernel or in
5443 firmware (i.e. not while executing in user mode)
5444 the kernel will oops in either "warn" or "fatal"
5447 #DB exception for bus lock is triggered only when
5451 Control the Special Register Buffer Data Sampling
5454 Certain CPUs are vulnerable to an MDS-like
5455 exploit which can leak bits from the random
5458 By default, this issue is mitigated by
5459 microcode. However, the microcode fix can cause
5460 the RDRAND and RDSEED instructions to become
5461 much slower. Among other effects, this will
5462 result in reduced throughput from /dev/urandom.
5464 The microcode mitigation can be disabled with
5465 the following option:
5467 off: Disable mitigation and remove
5468 performance impact to RDRAND and RDSEED
5470 srcutree.counter_wrap_check [KNL]
5471 Specifies how frequently to check for
5472 grace-period sequence counter wrap for the
5473 srcu_data structure's ->srcu_gp_seq_needed field.
5474 The greater the number of bits set in this kernel
5475 parameter, the less frequently counter wrap will
5476 be checked for. Note that the bottom two bits
5479 srcutree.exp_holdoff [KNL]
5480 Specifies how many nanoseconds must elapse
5481 since the end of the last SRCU grace period for
5482 a given srcu_struct until the next normal SRCU
5483 grace period will be considered for automatic
5484 expediting. Set to zero to disable automatic
5488 Speculative Store Bypass Disable control
5490 On CPUs that are vulnerable to the Speculative
5491 Store Bypass vulnerability and offer a
5492 firmware based mitigation, this parameter
5493 indicates how the mitigation should be used:
5495 force-on: Unconditionally enable mitigation for
5496 for both kernel and userspace
5497 force-off: Unconditionally disable mitigation for
5498 for both kernel and userspace
5499 kernel: Always enable mitigation in the
5500 kernel, and offer a prctl interface
5501 to allow userspace to register its
5502 interest in being mitigated too.
5504 stack_guard_gap= [MM]
5505 override the default stack gap protection. The value
5506 is in page units and it defines how many pages prior
5507 to (for stacks growing down) resp. after (for stacks
5508 growing up) the main stack are reserved for no other
5509 mapping. Default value is 256 pages.
5511 stack_depot_disable= [KNL]
5512 Setting this to true through kernel command line will
5513 disable the stack depot thereby saving the static memory
5514 consumed by the stack hash table. By default this is set
5518 Enabled the stack tracer on boot up.
5520 stacktrace_filter=[function-list]
5521 [FTRACE] Limit the functions that the stack tracer
5522 will trace at boot up. function-list is a comma-separated
5523 list of functions. This list can be changed at run
5524 time by the stack_trace_filter file in the debugfs
5525 tracing directory. Note, this enables stack tracing
5526 and the stacktrace above is not needed.
5530 Set the STI (builtin display/keyboard on the HP-PARISC
5531 machines) console (graphic card) which should be used
5532 as the initial boot-console.
5533 See also comment in drivers/video/console/sticore.c.
5536 See comment in drivers/video/console/sticore.c.
5539 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]]
5544 Enable or disable strict sigaltstack size checks
5545 against the required signal frame size which
5546 depends on the supported FPU features. This can
5547 be used to filter out binaries which have
5548 not yet been made aware of AT_MINSIGSTKSZ.
5550 sunrpc.min_resvport=
5551 sunrpc.max_resvport=
5553 SunRPC servers often require that client requests
5554 originate from a privileged port (i.e. a port in the
5555 range 0 < portnr < 1024).
5556 An administrator who wishes to reserve some of these
5557 ports for other uses may adjust the range that the
5558 kernel's sunrpc client considers to be privileged
5559 using these two parameters to set the minimum and
5560 maximum port values.
5562 sunrpc.svc_rpc_per_connection_limit=
5564 Limit the number of requests that the server will
5565 process in parallel from a single connection.
5566 The default value is 0 (no limit).
5570 Control how the NFS server code allocates CPUs to
5571 service thread pools. Depending on how many NICs
5572 you have and where their interrupts are bound, this
5573 option will affect which CPUs will do NFS serving.
5574 Note: this parameter cannot be changed while the
5575 NFS server is running.
5577 auto the server chooses an appropriate mode
5578 automatically using heuristics
5579 global a single global pool contains all CPUs
5580 percpu one pool for each CPU
5581 pernode one pool for each NUMA node (equivalent
5582 to global on non-NUMA machines)
5584 sunrpc.tcp_slot_table_entries=
5585 sunrpc.udp_slot_table_entries=
5587 Sets the upper limit on the number of simultaneous
5588 RPC calls that can be sent from the client to a
5589 server. Increasing these values may allow you to
5590 improve throughput, but will also increase the
5591 amount of memory reserved for use by the client.
5593 suspend.pm_test_delay=
5595 Sets the number of seconds to remain in a suspend test
5596 mode before resuming the system (see
5597 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG
5598 is set. Default value is 5.
5601 Format: { on | off | y | n | 1 | 0 }
5602 This parameter controls use of the Protected
5603 Execution Facility on pSeries.
5606 [KNL] Enable accounting of swap in memory resource
5607 controller if no parameter or 1 is given or disable
5608 it if 0 is given (See Documentation/admin-guide/cgroup-v1/memory.rst)
5610 swiotlb= [ARM,IA-64,PPC,MIPS,X86]
5611 Format: { <int> | force | noforce }
5612 <int> -- Number of I/O TLB slabs
5613 force -- force using of bounce buffers even if they
5614 wouldn't be automatically used by the kernel
5615 noforce -- Never use bounce buffers (for debugging)
5620 Set a sysctl parameter, right before loading the init
5621 process, as if the value was written to the respective
5622 /proc/sys/... file. Both '.' and '/' are recognized as
5623 separators. Unrecognized parameters and invalid values
5624 are reported in the kernel log. Sysctls registered
5625 later by a loaded module cannot be set this way.
5626 Example: sysctl.vm.swappiness=40
5628 sysfs.deprecated=0|1 [KNL]
5629 Enable/disable old style sysfs layout for old udev
5630 on older distributions. When this option is enabled
5631 very new udev will not work anymore. When this option
5632 is disabled (or CONFIG_SYSFS_DEPRECATED not compiled)
5633 in older udev will not work anymore.
5634 Default depends on CONFIG_SYSFS_DEPRECATED_V2 set in
5635 the kernel configuration.
5637 sysrq_always_enabled
5639 Ignore sysrq setting - this boot parameter will
5640 neutralize any effect of /proc/sys/kernel/sysrq.
5641 Useful for debugging.
5643 tcpmhash_entries= [KNL,NET]
5644 Set the number of tcp_metrics_hash slots.
5645 Default value is 8192 or 16384 depending on total
5646 ram pages. This is used to specify the TCP metrics
5647 cache size. See Documentation/networking/ip-sysctl.rst
5648 "tcp_no_metrics_save" section for more details.
5652 test_suspend= [SUSPEND][,N]
5653 Specify "mem" (for Suspend-to-RAM) or "standby" (for
5654 standby suspend) or "freeze" (for suspend type freeze)
5655 as the system sleep state during system startup with
5656 the optional capability to repeat N number of times.
5657 The system is woken from this state using a
5658 wakeup-capable RTC alarm.
5660 thash_entries= [KNL,NET]
5661 Set number of hash buckets for TCP connection
5663 thermal.act= [HW,ACPI]
5664 -1: disable all active trip points in all thermal zones
5665 <degrees C>: override all lowest active trip points
5667 thermal.crt= [HW,ACPI]
5668 -1: disable all critical trip points in all thermal zones
5669 <degrees C>: override all critical trip points
5671 thermal.nocrt= [HW,ACPI]
5672 Set to disable actions on ACPI thermal zone
5673 critical and hot trip points.
5675 thermal.off= [HW,ACPI]
5676 1: disable ACPI thermal control
5678 thermal.psv= [HW,ACPI]
5679 -1: disable all passive trip points
5680 <degrees C>: override all passive trip points to this
5683 thermal.tzp= [HW,ACPI]
5684 Specify global default ACPI thermal zone polling rate
5685 <deci-seconds>: poll all this frequency
5686 0: no polling (default)
5689 Force threading of all interrupt handlers except those
5690 marked explicitly IRQF_NO_THREAD.
5694 Specify if the kernel should make use of the cpu
5695 topology information if the hardware supports this.
5696 The scheduler will make use of this information and
5697 e.g. base its process migration decisions on it.
5700 topology_updates= [KNL, PPC, NUMA]
5702 Specify if the kernel should ignore (off)
5703 topology updates sent by the hypervisor to this
5706 torture.disable_onoff_at_boot= [KNL]
5707 Prevent the CPU-hotplug component of torturing
5708 until after init has spawned.
5710 torture.ftrace_dump_at_shutdown= [KNL]
5711 Dump the ftrace buffer at torture-test shutdown,
5712 even if there were no errors. This can be a
5713 very costly operation when many torture tests
5714 are running concurrently, especially on systems
5715 with rotating-rust storage.
5717 torture.verbose_sleep_frequency= [KNL]
5718 Specifies how many verbose printk()s should be
5719 emitted between each sleep. The default of zero
5720 disables verbose-printk() sleeping.
5722 torture.verbose_sleep_duration= [KNL]
5723 Duration of each verbose-printk() sleep in jiffies.
5727 tpm_suspend_pcr=[HW,TPM]
5728 Format: integer pcr id
5729 Specify that at suspend time, the tpm driver
5730 should extend the specified pcr with zeros,
5731 as a workaround for some chips which fail to
5732 flush the last written pcr on TPM_SaveState.
5733 This will guarantee that all the other pcrs
5736 trace_buf_size=nn[KMG]
5737 [FTRACE] will set tracing buffer size on each cpu.
5739 trace_event=[event-list]
5740 [FTRACE] Set and start specified trace events in order
5741 to facilitate early boot debugging. The event-list is a
5742 comma-separated list of trace events to enable. See
5743 also Documentation/trace/events.rst
5745 trace_options=[option-list]
5746 [FTRACE] Enable or disable tracer options at boot.
5747 The option-list is a comma delimited list of options
5748 that can be enabled or disabled just as if you were
5749 to echo the option name into
5751 /sys/kernel/debug/tracing/trace_options
5753 For example, to enable stacktrace option (to dump the
5754 stack trace of each event), add to the command line:
5756 trace_options=stacktrace
5758 See also Documentation/trace/ftrace.rst "trace options"
5762 Have the tracepoints sent to printk as well as the
5763 tracing ring buffer. This is useful for early boot up
5764 where the system hangs or reboots and does not give the
5765 option for reading the tracing buffer or performing a
5766 ftrace_dump_on_oops.
5768 To turn off having tracepoints sent to printk,
5769 echo 0 > /proc/sys/kernel/tracepoint_printk
5770 Note, echoing 1 into this file without the
5771 tracepoint_printk kernel cmdline option has no effect.
5773 The tp_printk_stop_on_boot (see below) can also be used
5774 to stop the printing of events to console at
5779 Having tracepoints sent to printk() and activating high
5780 frequency tracepoints such as irq or sched, can cause
5781 the system to live lock.
5783 tp_printk_stop_on_boot[FTRACE]
5784 When tp_printk (above) is set, it can cause a lot of noise
5785 on the console. It may be useful to only include the
5786 printing of events during boot up, as user space may
5787 make the system inoperable.
5789 This command line option will stop the printing of events
5790 to console at the late_initcall_sync() time frame.
5793 [FTRACE] enable this option to disable tracing when a
5794 warning is hit. This turns off "tracing_on". Tracing can
5795 be enabled again by echoing '1' into the "tracing_on"
5796 file located in /sys/kernel/debug/tracing/
5798 This option is useful, as it disables the trace before
5799 the WARNING dump is called, which prevents the trace to
5800 be filled with content caused by the warning output.
5802 This option can also be set at run time via the sysctl
5803 option: kernel/traceoff_on_warning
5805 transparent_hugepage=
5807 Format: [always|madvise|never]
5808 Can be used to control the default behavior of the system
5809 with respect to transparent hugepages.
5810 See Documentation/admin-guide/mm/transhuge.rst
5813 trusted.source= [KEYS]
5815 This parameter identifies the trust source as a backend
5816 for trusted keys implementation. Supported trust
5820 If not specified then it defaults to iterating through
5821 the trust source list starting with TPM and assigns the
5822 first trust source as a backend which is initialized
5823 successfully during iteration.
5825 tsc= Disable clocksource stability checks for TSC.
5827 [x86] reliable: mark tsc clocksource as reliable, this
5828 disables clocksource verification at runtime, as well
5829 as the stability checks done at bootup. Used to enable
5830 high-resolution timer mode on older hardware, and in
5831 virtualized environment.
5832 [x86] noirqtime: Do not use TSC to do irq accounting.
5833 Used to run time disable IRQ_TIME_ACCOUNTING on any
5834 platforms where RDTSC is slow and this accounting
5836 [x86] unstable: mark the TSC clocksource as unstable, this
5837 marks the TSC unconditionally unstable at bootup and
5838 avoids any further wobbles once the TSC watchdog notices.
5839 [x86] nowatchdog: disable clocksource watchdog. Used
5840 in situations with strict latency requirements (where
5841 interruptions from clocksource watchdog are not
5844 tsc_early_khz= [X86] Skip early TSC calibration and use the given
5845 value instead. Useful when the early TSC frequency discovery
5846 procedure is not reliable, such as on overclocked systems
5847 with CPUID.16h support and partial CPUID.15h support.
5848 Format: <unsigned int>
5850 tsx= [X86] Control Transactional Synchronization
5851 Extensions (TSX) feature in Intel processors that
5852 support TSX control.
5854 This parameter controls the TSX feature. The options are:
5856 on - Enable TSX on the system. Although there are
5857 mitigations for all known security vulnerabilities,
5858 TSX has been known to be an accelerator for
5859 several previous speculation-related CVEs, and
5860 so there may be unknown security risks associated
5861 with leaving it enabled.
5863 off - Disable TSX on the system. (Note that this
5864 option takes effect only on newer CPUs which are
5865 not vulnerable to MDS, i.e., have
5866 MSR_IA32_ARCH_CAPABILITIES.MDS_NO=1 and which get
5867 the new IA32_TSX_CTRL MSR through a microcode
5868 update. This new MSR allows for the reliable
5869 deactivation of the TSX functionality.)
5871 auto - Disable TSX if X86_BUG_TAA is present,
5872 otherwise enable TSX on the system.
5874 Not specifying this option is equivalent to tsx=off.
5876 See Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
5879 tsx_async_abort= [X86,INTEL] Control mitigation for the TSX Async
5880 Abort (TAA) vulnerability.
5882 Similar to Micro-architectural Data Sampling (MDS)
5883 certain CPUs that support Transactional
5884 Synchronization Extensions (TSX) are vulnerable to an
5885 exploit against CPU internal buffers which can forward
5886 information to a disclosure gadget under certain
5889 In vulnerable processors, the speculatively forwarded
5890 data can be used in a cache side channel attack, to
5891 access data to which the attacker does not have direct
5894 This parameter controls the TAA mitigation. The
5897 full - Enable TAA mitigation on vulnerable CPUs
5900 full,nosmt - Enable TAA mitigation and disable SMT on
5901 vulnerable CPUs. If TSX is disabled, SMT
5902 is not disabled because CPU is not
5903 vulnerable to cross-thread TAA attacks.
5904 off - Unconditionally disable TAA mitigation
5906 On MDS-affected machines, tsx_async_abort=off can be
5907 prevented by an active MDS mitigation as both vulnerabilities
5908 are mitigated with the same mechanism so in order to disable
5909 this mitigation, you need to specify mds=off too.
5911 Not specifying this option is equivalent to
5912 tsx_async_abort=full. On CPUs which are MDS affected
5913 and deploy MDS mitigation, TAA mitigation is not
5914 required and doesn't provide any additional
5918 Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
5920 turbografx.map[2|3]= [HW,JOY]
5921 TurboGraFX parallel port interface
5923 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7>
5924 See also Documentation/input/devices/joystick-parport.rst
5926 udbg-immortal [PPC] When debugging early kernel crashes that
5927 happen after console_init() and before a proper
5928 console driver takes over, this boot options might
5929 help "seeing" what's going on.
5931 uhash_entries= [KNL,NET]
5932 Set number of hash buckets for UDP/UDP-Lite connections
5935 [USB] Ignore overcurrent events (default N).
5936 Some badly-designed motherboards generate lots of
5937 bogus events, for ports that aren't wired to
5938 anything. Set this parameter to avoid log spamming.
5939 Note that genuine overcurrent events won't be
5943 [X86] Cause panic on unknown NMI.
5945 usbcore.authorized_default=
5946 [USB] Default USB device authorization:
5947 (default -1 = authorized except for wireless USB,
5948 0 = not authorized, 1 = authorized, 2 = authorized
5949 if device connected to internal port)
5951 usbcore.autosuspend=
5952 [USB] The autosuspend time delay (in seconds) used
5953 for newly-detected USB devices (default 2). This
5954 is the time required before an idle device will be
5955 autosuspended. Devices for which the delay is set
5956 to a negative value won't be autosuspended at all.
5958 usbcore.usbfs_snoop=
5959 [USB] Set to log all usbfs traffic (default 0 = off).
5961 usbcore.usbfs_snoop_max=
5962 [USB] Maximum number of bytes to snoop in each URB
5965 usbcore.blinkenlights=
5966 [USB] Set to cycle leds on hubs (default 0 = off).
5968 usbcore.old_scheme_first=
5969 [USB] Start with the old device initialization
5970 scheme (default 0 = off).
5972 usbcore.usbfs_memory_mb=
5973 [USB] Memory limit (in MB) for buffers allocated by
5974 usbfs (default = 16, 0 = max = 2047).
5976 usbcore.use_both_schemes=
5977 [USB] Try the other device initialization scheme
5978 if the first one fails (default 1 = enabled).
5980 usbcore.initial_descriptor_timeout=
5981 [USB] Specifies timeout for the initial 64-byte
5982 USB_REQ_GET_DESCRIPTOR request in milliseconds
5983 (default 5000 = 5.0 seconds).
5985 usbcore.nousb [USB] Disable the USB subsystem
5988 [USB] A list of quirk entries to augment the built-in
5989 usb core quirk list. List entries are separated by
5990 commas. Each entry has the form
5991 VendorID:ProductID:Flags. The IDs are 4-digit hex
5992 numbers and Flags is a set of letters. Each letter
5993 will change the built-in quirk; setting it if it is
5994 clear and clearing it if it is set. The letters have
5995 the following meanings:
5996 a = USB_QUIRK_STRING_FETCH_255 (string
5997 descriptors must not be fetched using
5999 b = USB_QUIRK_RESET_RESUME (device can't resume
6000 correctly so reset it instead);
6001 c = USB_QUIRK_NO_SET_INTF (device can't handle
6002 Set-Interface requests);
6003 d = USB_QUIRK_CONFIG_INTF_STRINGS (device can't
6004 handle its Configuration or Interface
6006 e = USB_QUIRK_RESET (device can't be reset
6007 (e.g morph devices), don't use reset);
6008 f = USB_QUIRK_HONOR_BNUMINTERFACES (device has
6009 more interface descriptions than the
6010 bNumInterfaces count, and can't handle
6011 talking to these interfaces);
6012 g = USB_QUIRK_DELAY_INIT (device needs a pause
6013 during initialization, after we read
6014 the device descriptor);
6015 h = USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL (For
6016 high speed and super speed interrupt
6017 endpoints, the USB 2.0 and USB 3.0 spec
6018 require the interval in microframes (1
6019 microframe = 125 microseconds) to be
6020 calculated as interval = 2 ^
6022 Devices with this quirk report their
6023 bInterval as the result of this
6024 calculation instead of the exponent
6025 variable used in the calculation);
6026 i = USB_QUIRK_DEVICE_QUALIFIER (device can't
6027 handle device_qualifier descriptor
6029 j = USB_QUIRK_IGNORE_REMOTE_WAKEUP (device
6030 generates spurious wakeup, ignore
6031 remote wakeup capability);
6032 k = USB_QUIRK_NO_LPM (device can't handle Link
6034 l = USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL
6035 (Device reports its bInterval as linear
6036 frames instead of the USB 2.0
6038 m = USB_QUIRK_DISCONNECT_SUSPEND (Device needs
6039 to be disconnected before suspend to
6040 prevent spurious wakeup);
6041 n = USB_QUIRK_DELAY_CTRL_MSG (Device needs a
6042 pause after every control message);
6043 o = USB_QUIRK_HUB_SLOW_RESET (Hub needs extra
6044 delay after resetting its port);
6045 Example: quirks=0781:5580:bk,0a5c:5834:gij
6048 [USBHID] The interval which mice are to be polled at.
6051 [USBHID] The interval which joysticks are to be polled at.
6054 [USBHID] The interval which keyboards are to be polled at.
6056 usb-storage.delay_use=
6057 [UMS] The delay in seconds before a new device is
6058 scanned for Logical Units (default 1).
6061 [UMS] A list of quirks entries to supplement or
6062 override the built-in unusual_devs list. List
6063 entries are separated by commas. Each entry has
6064 the form VID:PID:Flags where VID and PID are Vendor
6065 and Product ID values (4-digit hex numbers) and
6066 Flags is a set of characters, each corresponding
6067 to a common usb-storage quirk flag as follows:
6068 a = SANE_SENSE (collect more than 18 bytes
6069 of sense data, not on uas);
6070 b = BAD_SENSE (don't collect more than 18
6071 bytes of sense data, not on uas);
6072 c = FIX_CAPACITY (decrease the reported
6073 device capacity by one sector);
6074 d = NO_READ_DISC_INFO (don't use
6075 READ_DISC_INFO command, not on uas);
6076 e = NO_READ_CAPACITY_16 (don't use
6077 READ_CAPACITY_16 command);
6078 f = NO_REPORT_OPCODES (don't use report opcodes
6080 g = MAX_SECTORS_240 (don't transfer more than
6081 240 sectors at a time, uas only);
6082 h = CAPACITY_HEURISTICS (decrease the
6083 reported device capacity by one
6084 sector if the number is odd);
6085 i = IGNORE_DEVICE (don't bind to this
6087 j = NO_REPORT_LUNS (don't use report luns
6089 k = NO_SAME (do not use WRITE_SAME, uas only)
6090 l = NOT_LOCKABLE (don't try to lock and
6091 unlock ejectable media, not on uas);
6092 m = MAX_SECTORS_64 (don't transfer more
6093 than 64 sectors = 32 KB at a time,
6095 n = INITIAL_READ10 (force a retry of the
6096 initial READ(10) command, not on uas);
6097 o = CAPACITY_OK (accept the capacity
6098 reported by the device, not on uas);
6099 p = WRITE_CACHE (the device cache is ON
6100 by default, not on uas);
6101 r = IGNORE_RESIDUE (the device reports
6102 bogus residue values, not on uas);
6103 s = SINGLE_LUN (the device has only one
6105 t = NO_ATA_1X (don't allow ATA(12) and ATA(16)
6106 commands, uas only);
6107 u = IGNORE_UAS (don't bind to the uas driver);
6108 w = NO_WP_DETECT (don't test whether the
6109 medium is write-protected).
6110 y = ALWAYS_SYNC (issue a SYNCHRONIZE_CACHE
6111 even if the device claims no cache,
6113 Example: quirks=0419:aaf5:rl,0421:0433:rc
6115 user_debug= [KNL,ARM]
6117 See arch/arm/Kconfig.debug help text.
6118 1 - undefined instruction events
6120 4 - invalid data aborts
6123 Example: user_debug=31
6126 [X86] Flags controlling user PTE allocations.
6128 nohigh = do not allocate PTE pages in
6129 HIGHMEM regardless of setting
6133 On X86_32, this is an alias for vdso32=. Otherwise:
6135 vdso=1: enable VDSO (the default)
6136 vdso=0: disable VDSO mapping
6138 vdso32= [X86] Control the 32-bit vDSO
6139 vdso32=1: enable 32-bit VDSO
6140 vdso32=0 or vdso32=2: disable 32-bit VDSO
6142 See the help text for CONFIG_COMPAT_VDSO for more
6143 details. If CONFIG_COMPAT_VDSO is set, the default is
6144 vdso32=0; otherwise, the default is vdso32=1.
6146 For compatibility with older kernels, vdso32=2 is an
6149 Try vdso32=0 if you encounter an error that says:
6150 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed!
6153 vector=percpu: enable percpu vector domain
6155 video= [FB] Frame buffer configuration
6156 See Documentation/fb/modedb.rst.
6158 video.brightness_switch_enabled= [0,1]
6159 If set to 1, on receiving an ACPI notify event
6160 generated by hotkey, video driver will adjust brightness
6161 level and then send out the event to user space through
6162 the allocated input device; If set to 0, video driver
6163 will only send out the event without touching backlight
6168 [VMMIO] Memory mapped virtio (platform) device.
6170 <size>@<baseaddr>:<irq>[:<id>]
6172 <size> := size (can use standard suffixes
6174 <baseaddr> := physical base address
6175 <irq> := interrupt number (as passed to
6177 <id> := (optional) platform device id
6179 virtio_mmio.device=1K@0x100b0000:48:7
6181 Can be used multiple times for multiple devices.
6183 vga= [BOOT,X86-32] Select a particular video mode
6184 See Documentation/x86/boot.rst and
6185 Documentation/admin-guide/svga.rst.
6186 Use vga=ask for menu.
6187 This is actually a boot loader parameter; the value is
6188 passed to the kernel using a special protocol.
6190 vm_debug[=options] [KNL] Available with CONFIG_DEBUG_VM=y.
6191 May slow down system boot speed, especially when
6192 enabled on systems with a large amount of memory.
6193 All options are enabled by default, and this
6194 interface is meant to allow for selectively
6195 enabling or disabling specific virtual memory
6198 Available options are:
6199 P Enable page structure init time poisoning
6200 - Disable all of the above options
6202 vmalloc=nn[KMG] [KNL,BOOT] Forces the vmalloc area to have an exact
6203 size of <nn>. This can be used to increase the
6204 minimum size (128MB on x86). It can also be used to
6205 decrease the size and leave more room for directly
6208 vmcp_cma=nn[MG] [KNL,S390]
6209 Sets the memory size reserved for contiguous memory
6210 allocations for the vmcp device driver.
6212 vmhalt= [KNL,S390] Perform z/VM CP command after system halt.
6215 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic.
6218 vmpoff= [KNL,S390] Perform z/VM CP command after power off.
6222 Controls the behavior of vsyscalls (i.e. calls to
6223 fixed addresses of 0xffffffffff600x00 from legacy
6224 code). Most statically-linked binaries and older
6225 versions of glibc use these calls. Because these
6226 functions are at fixed addresses, they make nice
6227 targets for exploits that can control RIP.
6229 emulate [default] Vsyscalls turn into traps and are
6230 emulated reasonably safely. The vsyscall
6233 xonly Vsyscalls turn into traps and are
6234 emulated reasonably safely. The vsyscall
6235 page is not readable.
6237 none Vsyscalls don't work at all. This makes
6238 them quite hard to use for exploits but
6239 might break your system.
6241 vt.color= [VT] Default text color.
6242 Format: 0xYX, X = foreground, Y = background.
6243 Default: 0x07 = light gray on black.
6245 vt.cur_default= [VT] Default cursor shape.
6246 Format: 0xCCBBAA, where AA, BB, and CC are the same as
6247 the parameters of the <Esc>[?A;B;Cc escape sequence;
6248 see VGA-softcursor.txt. Default: 2 = underline.
6250 vt.default_blu= [VT]
6251 Format: <blue0>,<blue1>,<blue2>,...,<blue15>
6252 Change the default blue palette of the console.
6253 This is a 16-member array composed of values
6256 vt.default_grn= [VT]
6257 Format: <green0>,<green1>,<green2>,...,<green15>
6258 Change the default green palette of the console.
6259 This is a 16-member array composed of values
6262 vt.default_red= [VT]
6263 Format: <red0>,<red1>,<red2>,...,<red15>
6264 Change the default red palette of the console.
6265 This is a 16-member array composed of values
6271 Set system-wide default UTF-8 mode for all tty's.
6272 Default is 1, i.e. UTF-8 mode is enabled for all
6273 newly opened terminals.
6275 vt.global_cursor_default=
6278 Set system-wide default for whether a cursor
6279 is shown on new VTs. Default is -1,
6280 i.e. cursors will be created by default unless
6281 overridden by individual drivers. 0 will hide
6282 cursors, 1 will display them.
6284 vt.italic= [VT] Default color for italic text; 0-15.
6287 vt.underline= [VT] Default color for underlined text; 0-15.
6290 watchdog timers [HW,WDT] For information on watchdog timers,
6291 see Documentation/watchdog/watchdog-parameters.rst
6292 or other driver-specific files in the
6293 Documentation/watchdog/ directory.
6297 Set the hard lockup detector stall duration
6298 threshold in seconds. The soft lockup detector
6299 threshold is set to twice the value. A value of 0
6300 disables both lockup detectors. Default is 10
6303 workqueue.watchdog_thresh=
6304 If CONFIG_WQ_WATCHDOG is configured, workqueue can
6305 warn stall conditions and dump internal state to
6306 help debugging. 0 disables workqueue stall
6307 detection; otherwise, it's the stall threshold
6308 duration in seconds. The default value is 30 and
6309 it can be updated at runtime by writing to the
6310 corresponding sysfs file.
6312 workqueue.disable_numa
6313 By default, all work items queued to unbound
6314 workqueues are affine to the NUMA nodes they're
6315 issued on, which results in better behavior in
6316 general. If NUMA affinity needs to be disabled for
6317 whatever reason, this option can be used. Note
6318 that this also can be controlled per-workqueue for
6319 workqueues visible under /sys/bus/workqueue/.
6321 workqueue.power_efficient
6322 Per-cpu workqueues are generally preferred because
6323 they show better performance thanks to cache
6324 locality; unfortunately, per-cpu workqueues tend to
6325 be more power hungry than unbound workqueues.
6327 Enabling this makes the per-cpu workqueues which
6328 were observed to contribute significantly to power
6329 consumption unbound, leading to measurably lower
6330 power usage at the cost of small performance
6333 The default value of this parameter is determined by
6334 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT.
6336 workqueue.debug_force_rr_cpu
6337 Workqueue used to implicitly guarantee that work
6338 items queued without explicit CPU specified are put
6339 on the local CPU. This guarantee is no longer true
6340 and while local CPU is still preferred work items
6341 may be put on foreign CPUs. This debug option
6342 forces round-robin CPU selection to flush out
6343 usages which depend on the now broken guarantee.
6344 When enabled, memory and cache locality will be
6347 x2apic_phys [X86-64,APIC] Use x2apic physical mode instead of
6348 default x2apic cluster mode on platforms
6351 xen_512gb_limit [KNL,X86-64,XEN]
6352 Restricts the kernel running paravirtualized under Xen
6353 to use only up to 512 GB of RAM. The reason to do so is
6354 crash analysis tools and Xen tools for doing domain
6355 save/restore/migration must be enabled to handle larger
6358 xen_emul_unplug= [HW,X86,XEN]
6359 Unplug Xen emulated devices
6360 Format: [unplug0,][unplug1]
6361 ide-disks -- unplug primary master IDE devices
6362 aux-ide-disks -- unplug non-primary-master IDE devices
6363 nics -- unplug network devices
6364 all -- unplug all emulated devices (NICs and IDE disks)
6365 unnecessary -- unplugging emulated devices is
6366 unnecessary even if the host did not respond to
6368 never -- do not unplug even if version check succeeds
6370 xen_legacy_crash [X86,XEN]
6371 Crash from Xen panic notifier, without executing late
6372 panic() code such as dumping handler.
6374 xen_nopvspin [X86,XEN]
6375 Disables the qspinlock slowpath using Xen PV optimizations.
6376 This parameter is obsoleted by "nopvspin" parameter, which
6377 has equivalent effect for XEN platform.
6380 Disables the PV optimizations forcing the HVM guest to
6381 run as generic HVM guest with no PV drivers.
6382 This option is obsoleted by the "nopv" option, which
6383 has equivalent effect for XEN platform.
6385 xen_no_vector_callback
6386 [KNL,X86,XEN] Disable the vector callback for Xen
6387 event channel interrupts.
6389 xen_scrub_pages= [XEN]
6390 Boolean option to control scrubbing pages before giving them back
6391 to Xen, for use by other domains. Can be also changed at runtime
6392 with /sys/devices/system/xen_memory/xen_memory0/scrub_pages.
6393 Default value controlled with CONFIG_XEN_SCRUB_PAGES_DEFAULT.
6395 xen_timer_slop= [X86-64,XEN]
6396 Set the timer slop (in nanoseconds) for the virtual Xen
6397 timers (default is 100000). This adjusts the minimum
6398 delta of virtualized Xen timers, where lower values
6399 improve timer resolution at the expense of processing
6400 more timer interrupts.
6402 xen.balloon_boot_timeout= [XEN]
6403 The time (in seconds) to wait before giving up to boot
6404 in case initial ballooning fails to free enough memory.
6405 Applies only when running as HVM or PVH guest and
6406 started with less memory configured than allowed at
6407 max. Default is 180.
6409 xen.event_eoi_delay= [XEN]
6410 How long to delay EOI handling in case of event
6411 storms (jiffies). Default is 10.
6413 xen.event_loop_timeout= [XEN]
6414 After which time (jiffies) the event handling loop
6415 should start to delay EOI handling. Default is 2.
6417 xen.fifo_events= [XEN]
6418 Boolean parameter to disable using fifo event handling
6419 even if available. Normally fifo event handling is
6420 preferred over the 2-level event handling, as it is
6421 fairer and the number of possible event channels is
6422 much higher. Default is on (use fifo events).
6424 nopv= [X86,XEN,KVM,HYPER_V,VMWARE]
6425 Disables the PV optimizations forcing the guest to run
6426 as generic guest with no PV drivers. Currently support
6427 XEN HVM, KVM, HYPER_V and VMWARE guest.
6429 nopvspin [X86,XEN,KVM]
6430 Disables the qspinlock slow path using PV optimizations
6431 which allow the hypervisor to 'idle' the guest on lock
6434 xirc2ps_cs= [NET,PCMCIA]
6436 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]]
6439 By default on POWER9 and above, the kernel will
6440 natively use the XIVE interrupt controller. This option
6441 allows the fallback firmware mode to be used:
6443 off Fallback to firmware control of XIVE interrupt
6444 controller on both pseries and powernv
6445 platforms. Only useful on POWER9 and above.
6447 xhci-hcd.quirks [USB,KNL]
6448 A hex value specifying bitmask with supplemental xhci
6449 host controller quirks. Meaning of each bit can be
6450 consulted in header drivers/usb/host/xhci.h.
6453 Format: { early | on | rw | ro | off }
6454 Controls if xmon debugger is enabled. Default is off.
6455 Passing only "xmon" is equivalent to "xmon=early".
6456 early Call xmon as early as possible on boot; xmon
6457 debugger is called from setup_arch().
6458 on xmon debugger hooks will be installed so xmon
6459 is only called on a kernel crash. Default mode,
6460 i.e. either "ro" or "rw" mode, is controlled
6461 with CONFIG_XMON_DEFAULT_RO_MODE.
6462 rw xmon debugger hooks will be installed so xmon
6463 is called only on a kernel crash, mode is write,
6464 meaning SPR registers, memory and, other data
6465 can be written using xmon commands.
6466 ro same as "rw" option above but SPR registers,
6467 memory, and other data can't be written using
6469 off xmon is disabled.
git.samba.org / sfrench / cifs-2.6.git / blob