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_hwsig,
229 s4_nohwsig, old_ordering, nonvs,
230 sci_force_enable, nobl }
231 See Documentation/power/video.rst for information on
233 s3_beep is for debugging; it makes the PC's speaker beep
234 as soon as the kernel's real-mode entry point is called.
235 s4_hwsig causes the kernel to check the ACPI hardware
236 signature during resume from hibernation, and gracefully
237 refuse to resume if it has changed. This complies with
238 the ACPI specification but not with reality, since
239 Windows does not do this and many laptops do change it
240 on docking. So the default behaviour is to allow resume
241 and simply warn when the signature changes, unless the
242 s4_hwsig option is enabled.
243 s4_nohwsig prevents ACPI hardware signature from being
244 used (or even warned about) during resume.
245 old_ordering causes the ACPI 1.0 ordering of the _PTS
246 control method, with respect to putting devices into
247 low power states, to be enforced (the ACPI 2.0 ordering
248 of _PTS is used by default).
249 nonvs prevents the kernel from saving/restoring the
250 ACPI NVS memory during suspend/hibernation and resume.
251 sci_force_enable causes the kernel to set SCI_EN directly
252 on resume from S1/S3 (which is against the ACPI spec,
253 but some broken systems don't work without it).
254 nobl causes the internal blacklist of systems known to
255 behave incorrectly in some ways with respect to system
256 suspend and resume to be ignored (use wisely).
258 acpi_use_timer_override [HW,ACPI]
259 Use timer override. For some broken Nvidia NF5 boards
260 that require a timer override, but don't have HPET
262 add_efi_memmap [EFI; X86] Include EFI memory map in
263 kernel's map of available physical RAM.
266 { off | try_unsupported }
267 off: disable AGP support
268 try_unsupported: try to drive unsupported chipsets
269 (may crash computer or cause data corruption)
272 See Documentation/sound/alsa-configuration.rst
275 Allow the default userspace alignment fault handler
276 behaviour to be specified. Bit 0 enables warnings,
277 bit 1 enables fixups, and bit 2 sends a segfault.
279 align_va_addr= [X86-64]
280 Align virtual addresses by clearing slice [14:12] when
281 allocating a VMA at process creation time. This option
282 gives you up to 3% performance improvement on AMD F15h
283 machines (where it is enabled by default) for a
284 CPU-intensive style benchmark, and it can vary highly in
285 a microbenchmark depending on workload and compiler.
287 32: only for 32-bit processes
288 64: only for 64-bit processes
289 on: enable for both 32- and 64-bit processes
290 off: disable for both 32- and 64-bit processes
292 alloc_snapshot [FTRACE]
293 Allocate the ftrace snapshot buffer on boot up when the
294 main buffer is allocated. This is handy if debugging
295 and you need to use tracing_snapshot() on boot up, and
296 do not want to use tracing_snapshot_alloc() as it needs
297 to be done where GFP_KERNEL allocations are allowed.
299 allow_mismatched_32bit_el0 [ARM64]
300 Allow execve() of 32-bit applications and setting of the
301 PER_LINUX32 personality on systems where only a strict
302 subset of the CPUs support 32-bit EL0. When this
303 parameter is present, the set of CPUs supporting 32-bit
304 EL0 is indicated by /sys/devices/system/cpu/aarch32_el0
305 and hot-unplug operations may be restricted.
307 See Documentation/arm64/asymmetric-32bit.rst for more
310 amd_iommu= [HW,X86-64]
311 Pass parameters to the AMD IOMMU driver in the system.
313 fullflush - Deprecated, equivalent to iommu.strict=1
314 off - do not initialize any AMD IOMMU found in
316 force_isolation - Force device isolation for all
317 devices. The IOMMU driver is not
318 allowed anymore to lift isolation
319 requirements as needed. This option
320 does not override iommu=pt
321 force_enable - Force enable the IOMMU on platforms known
322 to be buggy with IOMMU enabled. Use this
325 amd_iommu_dump= [HW,X86-64]
326 Enable AMD IOMMU driver option to dump the ACPI table
327 for AMD IOMMU. With this option enabled, AMD IOMMU
328 driver will print ACPI tables for AMD IOMMU during
329 IOMMU initialization.
331 amd_iommu_intr= [HW,X86-64]
332 Specifies one of the following AMD IOMMU interrupt
334 legacy - Use legacy interrupt remapping mode.
335 vapic - Use virtual APIC mode, which allows IOMMU
336 to inject interrupts directly into guest.
337 This mode requires kvm-amd.avic=1.
338 (Default when IOMMU HW support is present.)
340 amijoy.map= [HW,JOY] Amiga joystick support
341 Map of devices attached to JOY0DAT and JOY1DAT
343 See also Documentation/input/joydev/joystick.rst
345 analog.map= [HW,JOY] Analog joystick and gamepad support
346 Specifies type or capabilities of an analog joystick
347 connected to one of 16 gameports
348 Format: <type1>,<type2>,..<type16>
351 Power management functions (SPARCstation-4/5 + deriv.)
353 Disable APC CPU standby support. SPARCstation-Fox does
354 not play well with APC CPU idle - disable it if you have
355 APC and your system crashes randomly.
357 apic= [APIC,X86] Advanced Programmable Interrupt Controller
358 Change the output verbosity while booting
359 Format: { quiet (default) | verbose | debug }
360 Change the amount of debugging information output
361 when initialising the APIC and IO-APIC components.
362 For X86-32, this can also be used to specify an APIC
364 Format: apic=driver_name
365 Examples: apic=bigsmp
367 apic_extnmi= [APIC,X86] External NMI delivery setting
368 Format: { bsp (default) | all | none }
369 bsp: External NMI is delivered only to CPU 0
370 all: External NMIs are broadcast to all CPUs as a
372 none: External NMI is masked for all CPUs. This is
373 useful so that a dump capture kernel won't be
377 See Documentation/networking/ipv6.rst.
379 show_lapic= [APIC,X86] Advanced Programmable Interrupt Controller
380 Limit apic dumping. The parameter defines the maximal
381 number of local apics being dumped. Also it is possible
382 to set it to "all" by meaning -- no limit here.
383 Format: { 1 (default) | 2 | ... | all }.
384 The parameter valid if only apic=debug or
385 apic=verbose is specified.
386 Example: apic=debug show_lapic=all
388 apm= [APM] Advanced Power Management
389 See header of arch/x86/kernel/apm_32.c.
391 arcrimi= [HW,NET] ARCnet - "RIM I" (entirely mem-mapped) cards
392 Format: <io>,<irq>,<nodeID>
394 arm64.nobti [ARM64] Unconditionally disable Branch Target
395 Identification support
397 arm64.nopauth [ARM64] Unconditionally disable Pointer Authentication
400 arm64.nomte [ARM64] Unconditionally disable Memory Tagging Extension
405 atarimouse= [HW,MOUSE] Atari Mouse
407 atkbd.extra= [HW] Enable extra LEDs and keys on IBM RapidAccess,
408 EzKey and similar keyboards
410 atkbd.reset= [HW] Reset keyboard during initialization
412 atkbd.set= [HW] Select keyboard code set
413 Format: <int> (2 = AT (default), 3 = PS/2)
415 atkbd.scroll= [HW] Enable scroll wheel on MS Office and similar
418 atkbd.softraw= [HW] Choose between synthetic and real raw mode
419 Format: <bool> (0 = real, 1 = synthetic (default))
421 atkbd.softrepeat= [HW]
422 Use software keyboard repeat
424 audit= [KNL] Enable the audit sub-system
425 Format: { "0" | "1" | "off" | "on" }
426 0 | off - kernel audit is disabled and can not be
427 enabled until the next reboot
428 unset - kernel audit is initialized but disabled and
429 will be fully enabled by the userspace auditd.
430 1 | on - kernel audit is initialized and partially
431 enabled, storing at most audit_backlog_limit
432 messages in RAM until it is fully enabled by the
436 audit_backlog_limit= [KNL] Set the audit queue size limit.
437 Format: <int> (must be >=0)
440 bau= [X86_UV] Enable the BAU on SGI UV. The default
441 behavior is to disable the BAU (i.e. bau=0).
442 Format: { "0" | "1" }
445 unset - Disable the BAU.
447 baycom_epp= [HW,AX25]
450 baycom_par= [HW,AX25] BayCom Parallel Port AX.25 Modem
452 See header of drivers/net/hamradio/baycom_par.c.
454 baycom_ser_fdx= [HW,AX25]
455 BayCom Serial Port AX.25 Modem (Full Duplex Mode)
456 Format: <io>,<irq>,<mode>[,<baud>]
457 See header of drivers/net/hamradio/baycom_ser_fdx.c.
459 baycom_ser_hdx= [HW,AX25]
460 BayCom Serial Port AX.25 Modem (Half Duplex Mode)
461 Format: <io>,<irq>,<mode>
462 See header of drivers/net/hamradio/baycom_ser_hdx.c.
464 blkdevparts= Manual partition parsing of block device(s) for
465 embedded devices based on command line input.
466 See Documentation/block/cmdline-partition.rst
468 boot_delay= Milliseconds to delay each printk during boot.
469 Values larger than 10 seconds (10000) are changed to
474 Extended command line options can be added to an initrd
475 and this will cause the kernel to look for it.
477 See Documentation/admin-guide/bootconfig.rst
480 Disable BERT OS support on buggy BIOSes.
482 bgrt_disable [ACPI][X86]
483 Disable BGRT to avoid flickering OEM logo.
485 bttv.card= [HW,V4L] bttv (bt848 + bt878 based grabber cards)
486 bttv.radio= Most important insmod options are available as
488 bttv.pll= See Documentation/admin-guide/media/bttv.rst
491 bulk_remove=off [PPC] This parameter disables the use of the pSeries
492 firmware feature for flushing multiple hpte entries
495 c101= [NET] Moxa C101 synchronous serial card
497 cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection.
498 Sometimes CPU hardware bugs make them report the cache
499 size incorrectly. The kernel will attempt work arounds
500 to fix known problems, but for some CPUs it is not
501 possible to determine what the correct size should be.
502 This option provides an override for these situations.
505 [NET] Specifies amount of time (in seconds) that
506 the kernel should wait for a network carrier. By default
507 it waits 120 seconds.
509 ca_keys= [KEYS] This parameter identifies a specific key(s) on
510 the system trusted keyring to be used for certificate
512 format: { id:<keyid> | builtin }
514 cca= [MIPS] Override the kernel pages' cache coherency
515 algorithm. Accepted values range from 0 to 7
516 inclusive. See arch/mips/include/asm/pgtable-bits.h
517 for platform specific values (SB1, Loongson3 and
520 ccw_timeout_log [S390]
521 See Documentation/s390/common_io.rst for details.
523 cgroup_disable= [KNL] Disable a particular controller or optional feature
524 Format: {name of the controller(s) or feature(s) to disable}
525 The effects of cgroup_disable=foo are:
526 - foo isn't auto-mounted if you mount all cgroups in
528 - foo isn't visible as an individually mountable
530 - if foo is an optional feature then the feature is
531 disabled and corresponding cgroup files are not
533 {Currently only "memory" controller deal with this and
534 cut the overhead, others just disable the usage. So
535 only cgroup_disable=memory is actually worthy}
536 Specifying "pressure" disables per-cgroup pressure
537 stall information accounting feature
539 cgroup_no_v1= [KNL] Disable cgroup controllers and named hierarchies in v1
540 Format: { { controller | "all" | "named" }
541 [,{ controller | "all" | "named" }...] }
542 Like cgroup_disable, but only applies to cgroup v1;
543 the blacklisted controllers remain available in cgroup2.
544 "all" blacklists all controllers and "named" disables
545 named mounts. Specifying both "all" and "named" disables
548 cgroup.memory= [KNL] Pass options to the cgroup memory controller.
550 nosocket -- Disable socket memory accounting.
551 nokmem -- Disable kernel memory accounting.
553 checkreqprot [SELINUX] Set initial checkreqprot flag value.
554 Format: { "0" | "1" }
555 See security/selinux/Kconfig help text.
556 0 -- check protection applied by kernel (includes
557 any implied execute protection).
558 1 -- check protection requested by application.
559 Default value is set via a kernel config option.
560 Value can be changed at runtime via
561 /sys/fs/selinux/checkreqprot.
562 Setting checkreqprot to 1 is deprecated.
565 See Documentation/s390/common_io.rst for details.
568 Prevents the clock framework from automatically gating
569 clocks that have not been explicitly enabled by a Linux
570 device driver but are enabled in hardware at reset or
571 by the bootloader/firmware. Note that this does not
572 force such clocks to be always-on nor does it reserve
573 those clocks in any way. This parameter is useful for
574 debug and development, but should not be needed on a
575 platform with proper driver support. For more
576 information, see Documentation/driver-api/clk.rst.
578 clock= [BUGS=X86-32, HW] gettimeofday clocksource override.
580 Forces specified clocksource (if available) to be used
581 when calculating gettimeofday(). If specified
582 clocksource is not available, it defaults to PIT.
583 Format: { pit | tsc | cyclone | pmtmr }
585 clocksource= Override the default clocksource
587 Override the default clocksource and use the clocksource
588 with the name specified.
589 Some clocksource names to choose from, depending on
591 [all] jiffies (this is the base, fallback clocksource)
593 [ARM] imx_timer1,OSTS,netx_timer,mpu_timer2,
594 pxa_timer,timer3,32k_counter,timer0_1
595 [X86-32] pit,hpet,tsc;
596 scx200_hrt on Geode; cyclone on IBM x440
604 clocksource.arm_arch_timer.evtstrm=
607 Enable/disable the eventstream feature of the ARM
608 architected timer so that code using WFE-based polling
609 loops can be debugged more effectively on production
612 clocksource.max_cswd_read_retries= [KNL]
613 Number of clocksource_watchdog() retries due to
614 external delays before the clock will be marked
615 unstable. Defaults to three retries, that is,
616 four attempts to read the clock under test.
618 clocksource.verify_n_cpus= [KNL]
619 Limit the number of CPUs checked for clocksources
620 marked with CLOCK_SOURCE_VERIFY_PERCPU that
621 are marked unstable due to excessive skew.
622 A negative value says to check all CPUs, while
623 zero says not to check any. Values larger than
624 nr_cpu_ids are silently truncated to nr_cpu_ids.
625 The actual CPUs are chosen randomly, with
626 no replacement if the same CPU is chosen twice.
628 clocksource-wdtest.holdoff= [KNL]
629 Set the time in seconds that the clocksource
630 watchdog test waits before commencing its tests.
631 Defaults to zero when built as a module and to
632 10 seconds when built into the kernel.
634 clearcpuid=BITNUM[,BITNUM...] [X86]
635 Disable CPUID feature X for the kernel. See
636 arch/x86/include/asm/cpufeatures.h for the valid bit
637 numbers. Note the Linux specific bits are not necessarily
638 stable over kernel options, but the vendor specific
640 Also note that user programs calling CPUID directly
641 or using the feature without checking anything
642 will still see it. This just prevents it from
643 being used by the kernel or shown in /proc/cpuinfo.
644 Also note the kernel might malfunction if you disable
647 cma=nn[MG]@[start[MG][-end[MG]]]
649 Sets the size of kernel global memory area for
650 contiguous memory allocations and optionally the
651 placement constraint by the physical address range of
652 memory allocations. A value of 0 disables CMA
653 altogether. For more information, see
654 kernel/dma/contiguous.c
658 Sets the size of kernel per-numa memory area for
659 contiguous memory allocations. A value of 0 disables
660 per-numa CMA altogether. And If this option is not
661 specificed, the default value is 0.
662 With per-numa CMA enabled, DMA users on node nid will
663 first try to allocate buffer from the pernuma area
664 which is located in node nid, if the allocation fails,
665 they will fallback to the global default memory area.
667 cmo_free_hint= [PPC] Format: { yes | no }
668 Specify whether pages are marked as being inactive
669 when they are freed. This is used in CMO environments
670 to determine OS memory pressure for page stealing by
674 coherent_pool=nn[KMG] [ARM,KNL]
675 Sets the size of memory pool for coherent, atomic dma
676 allocations, by default set to 256K.
678 com20020= [HW,NET] ARCnet - COM20020 chipset
680 <io>[,<irq>[,<nodeID>[,<backplane>[,<ckp>[,<timeout>]]]]]
682 com90io= [HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers)
686 ARCnet - COM90xx chipset (memory-mapped buffers)
687 Format: <io>[,<irq>[,<memstart>]]
689 condev= [HW,S390] console device
692 console= [KNL] Output console device and options.
694 tty<n> Use the virtual console device <n>.
698 Use the specified serial port. The options are of
699 the form "bbbbpnf", where "bbbb" is the baud rate,
700 "p" is parity ("n", "o", or "e"), "n" is number of
701 bits, and "f" is flow control ("r" for RTS or
702 omit it). Default is "9600n8".
704 See Documentation/admin-guide/serial-console.rst for more
706 Documentation/networking/netconsole.rst for an
709 uart[8250],io,<addr>[,options]
710 uart[8250],mmio,<addr>[,options]
711 uart[8250],mmio16,<addr>[,options]
712 uart[8250],mmio32,<addr>[,options]
713 uart[8250],0x<addr>[,options]
714 Start an early, polled-mode console on the 8250/16550
715 UART at the specified I/O port or MMIO address,
716 switching to the matching ttyS device later.
717 MMIO inter-register address stride is either 8-bit
718 (mmio), 16-bit (mmio16), or 32-bit (mmio32).
719 If none of [io|mmio|mmio16|mmio32], <addr> is assumed
720 to be equivalent to 'mmio'. 'options' are specified in
721 the same format described for ttyS above; if unspecified,
722 the h/w is not re-initialized.
724 hvc<n> Use the hypervisor console device <n>. This is for
725 both Xen and PowerPC hypervisors.
727 If the device connected to the port is not a TTY but a braille
728 device, prepend "brl," before the device type, for instance
730 For now, only VisioBraille is supported.
733 [KNL] Change console messages format
735 By default we print messages on consoles in
736 "[time stamp] text\n" format (time stamp may not be
737 printed, depending on CONFIG_PRINTK_TIME or
738 `printk_time' param).
740 Switch to syslog format: "<%u>[time stamp] text\n"
741 IOW, each message will have a facility and loglevel
742 prefix. The format is similar to one used by syslog()
743 syscall, or to executing "dmesg -S --raw" or to reading
746 consoleblank= [KNL] The console blank (screen saver) timeout in
747 seconds. A value of 0 disables the blank timer.
751 [KNL] Change the default value for
752 /proc/<pid>/coredump_filter.
753 See also Documentation/filesystems/proc.rst.
755 coresight_cpu_debug.enable
758 Enable/disable the CPU sampling based debugging.
759 0: default value, disable debugging
760 1: enable debugging at boot time
762 cpuidle.off=1 [CPU_IDLE]
763 disable the cpuidle sub-system
766 [CPU_IDLE] Name of the cpuidle governor to use.
768 cpufreq.off=1 [CPU_FREQ]
769 disable the cpufreq sub-system
771 cpufreq.default_governor=
772 [CPU_FREQ] Name of the default cpufreq governor or
773 policy to use. This governor must be registered in the
774 kernel before the cpufreq driver probes.
777 [X86] Delay for N microsec between assert and de-assert
778 of APIC INIT to start processors. This delay occurs
779 on every CPU online, such as boot, and resume from suspend.
782 cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver
784 <first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>]
786 crashkernel=size[KMG][@offset[KMG]]
787 [KNL] Using kexec, Linux can switch to a 'crash kernel'
788 upon panic. This parameter reserves the physical
789 memory region [offset, offset + size] for that kernel
790 image. If '@offset' is omitted, then a suitable offset
791 is selected automatically.
792 [KNL, X86-64] Select a region under 4G first, and
793 fall back to reserve region above 4G when '@offset'
794 hasn't been specified.
795 See Documentation/admin-guide/kdump/kdump.rst for further details.
797 crashkernel=range1:size1[,range2:size2,...][@offset]
798 [KNL] Same as above, but depends on the memory
799 in the running system. The syntax of range is
800 start-[end] where start and end are both
801 a memory unit (amount[KMG]). See also
802 Documentation/admin-guide/kdump/kdump.rst for an example.
804 crashkernel=size[KMG],high
805 [KNL, X86-64] range could be above 4G. Allow kernel
806 to allocate physical memory region from top, so could
807 be above 4G if system have more than 4G ram installed.
808 Otherwise memory region will be allocated below 4G, if
810 It will be ignored if crashkernel=X is specified.
811 crashkernel=size[KMG],low
812 [KNL, X86-64] range under 4G. When crashkernel=X,high
813 is passed, kernel could allocate physical memory region
814 above 4G, that cause second kernel crash on system
815 that require some amount of low memory, e.g. swiotlb
816 requires at least 64M+32K low memory, also enough extra
817 low memory is needed to make sure DMA buffers for 32-bit
818 devices won't run out. Kernel would try to allocate at
819 at least 256M below 4G automatically.
820 This one let user to specify own low range under 4G
821 for second kernel instead.
822 0: to disable low allocation.
823 It will be ignored when crashkernel=X,high is not used
824 or memory reserved is below 4G.
827 [KNL] Disable crypto self-tests
832 cs89x0_media= [HW,NET]
833 Format: { rj45 | aui | bnc }
835 csdlock_debug= [KNL] Enable debug add-ons of cross-CPU function call
836 handling. When switched on, additional debug data is
837 printed to the console in case a hanging CPU is
838 detected, and that CPU is pinged again in order to try
839 to resolve the hang situation.
840 0: disable csdlock debugging (default)
841 1: enable basic csdlock debugging (minor impact)
842 ext: enable extended csdlock debugging (more impact,
846 See header of drivers/s390/block/dasd_devmap.c.
848 db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port
849 (one device per port)
850 Format: <port#>,<type>
851 See also Documentation/input/devices/joystick-parport.rst
853 debug [KNL] Enable kernel debugging (events log level).
856 [KNL] Enable printing [hashed] pointers early in the
857 boot sequence. If enabled, we use a weak hash instead
858 of siphash to hash pointers. Use this option if you are
859 seeing instances of '(___ptrval___)') and need to see a
860 value (hashed pointer) instead. Cryptographically
861 insecure, please do not use on production kernels.
864 [KNL] verbose locking self-tests
866 Print debugging info while doing the locking API
868 Bitmask for the various LOCKTYPE_ tests. Defaults to 0
869 (no extra messages), setting it to -1 (all bits set)
870 will print _a_lot_ more information - normally only
871 useful to lockdep developers.
873 debug_objects [KNL] Enable object debugging
876 [KNL] Disable object debugging
878 debug_guardpage_minorder=
879 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this
880 parameter allows control of the order of pages that will
881 be intentionally kept free (and hence protected) by the
882 buddy allocator. Bigger value increase the probability
883 of catching random memory corruption, but reduce the
884 amount of memory for normal system use. The maximum
885 possible value is MAX_ORDER/2. Setting this parameter
886 to 1 or 2 should be enough to identify most random
887 memory corruption problems caused by bugs in kernel or
888 driver code when a CPU writes to (or reads from) a
889 random memory location. Note that there exists a class
890 of memory corruptions problems caused by buggy H/W or
891 F/W or by drivers badly programing DMA (basically when
892 memory is written at bus level and the CPU MMU is
893 bypassed) which are not detectable by
894 CONFIG_DEBUG_PAGEALLOC, hence this option will not help
895 tracking down these problems.
898 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this parameter
899 enables the feature at boot time. By default, it is
900 disabled and the system will work mostly the same as a
901 kernel built without CONFIG_DEBUG_PAGEALLOC.
902 Note: to get most of debug_pagealloc error reports, it's
903 useful to also enable the page_owner functionality.
904 on: enable the feature
906 debugfs= [KNL] This parameter enables what is exposed to userspace
907 and debugfs internal clients.
908 Format: { on, no-mount, off }
909 on: All functions are enabled.
911 Filesystem is not registered but kernel clients can
912 access APIs and a crashkernel can be used to read
913 its content. There is nothing to mount.
914 off: Filesystem is not registered and clients
915 get a -EPERM as result when trying to register files
916 or directories within debugfs.
917 This is equivalent of the runtime functionality if
918 debugfs was not enabled in the kernel at all.
919 Default value is set in build-time with a kernel configuration.
921 debugpat [X86] Enable PAT debugging
923 decnet.addr= [HW,NET]
924 Format: <area>[,<node>]
925 See also Documentation/networking/decnet.rst.
928 [HW] The size of the default HugeTLB page. This is
929 the size represented by the legacy /proc/ hugepages
930 APIs. In addition, this is the default hugetlb size
931 used for shmget(), mmap() and mounting hugetlbfs
932 filesystems. If not specified, defaults to the
933 architecture's default huge page size. Huge page
934 sizes are architecture dependent. See also
935 Documentation/admin-guide/mm/hugetlbpage.rst.
938 deferred_probe_timeout=
939 [KNL] Debugging option to set a timeout in seconds for
940 deferred probe to give up waiting on dependencies to
941 probe. Only specific dependencies (subsystems or
942 drivers) that have opted in will be ignored. A timeout of 0
943 will timeout at the end of initcalls. This option will also
944 dump out devices still on the deferred probe list after
948 Format: { on | off | def_only | inf_only | always }
949 on: s390 zlib hardware support for compression on
950 level 1 and decompression (default)
951 off: No s390 zlib hardware support
952 def_only: s390 zlib hardware support for deflate
953 only (compression on level 1)
954 inf_only: s390 zlib hardware support for inflate
956 always: Same as 'on' but ignores the selected compression
957 level always using hardware support (used for debugging)
960 Set number of hash buckets for dentry cache.
962 disable_1tb_segments [PPC]
963 Disables the use of 1TB hash page table segments. This
964 causes the kernel to fall back to 256MB segments which
965 can be useful when debugging issues that require an SLB
969 Limits the number of kernel SLB entries, and flushes
970 them frequently to increase the rate of SLB faults
974 See Documentation/networking/ipv6.rst.
977 [KNL] Under CONFIG_HARDENED_USERCOPY, whether
978 hardening is enabled for this boot. Hardened
979 usercopy checking is used to protect the kernel
980 from reading or writing beyond known memory
981 allocation boundaries as a proactive defense
982 against bounds-checking flaws in the kernel's
983 copy_to_user()/copy_from_user() interface.
984 on Perform hardened usercopy checks (default).
985 off Disable hardened usercopy checks.
988 Disable RADIX MMU mode on POWER9
990 radix_hcall_invalidate=on [PPC/PSERIES]
991 Disable RADIX GTSE feature and use hcall for TLB
995 Disable TLBIE instruction. Currently does not work
996 with KVM, with HASH MMU, or with coherent accelerators.
998 disable_cpu_apicid= [X86,APIC,SMP]
1000 The number of initial APIC ID for the
1001 corresponding CPU to be disabled at boot,
1002 mostly used for the kdump 2nd kernel to
1003 disable BSP to wake up multiple CPUs without
1004 causing system reset or hang due to sending
1005 INIT from AP to BSP.
1007 disable_ddw [PPC/PSERIES]
1008 Disable Dynamic DMA Window support. Use this
1009 to workaround buggy firmware.
1011 disable_ipv6= [IPV6]
1012 See Documentation/networking/ipv6.rst.
1014 disable_mtrr_cleanup [X86]
1015 The kernel tries to adjust MTRR layout from continuous
1016 to discrete, to make X server driver able to add WB
1017 entry later. This parameter disables that.
1019 disable_mtrr_trim [X86, Intel and AMD only]
1020 By default the kernel will trim any uncacheable
1021 memory out of your available memory pool based on
1022 MTRR settings. This parameter disables that behavior,
1023 possibly causing your machine to run very slowly.
1025 disable_timer_pin_1 [X86]
1026 Disable PIN 1 of APIC timer
1027 Can be useful to work around chipset bugs.
1029 dis_ucode_ldr [X86] Disable the microcode loader.
1031 dma_debug=off If the kernel is compiled with DMA_API_DEBUG support,
1032 this option disables the debugging code at boot.
1034 dma_debug_entries=<number>
1035 This option allows to tune the number of preallocated
1036 entries for DMA-API debugging code. One entry is
1037 required per DMA-API allocation. Use this if the
1038 DMA-API debugging code disables itself because the
1039 architectural default is too low.
1041 dma_debug_driver=<driver_name>
1042 With this option the DMA-API debugging driver
1043 filter feature can be enabled at boot time. Just
1044 pass the driver to filter for as the parameter.
1045 The filter can be disabled or changed to another
1046 driver later using sysfs.
1048 driver_async_probe= [KNL]
1049 List of driver names to be probed asynchronously.
1050 Format: <driver_name1>,<driver_name2>...
1052 drm.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>]
1053 Broken monitors, graphic adapters, KVMs and EDIDless
1054 panels may send no or incorrect EDID data sets.
1055 This parameter allows to specify an EDID data sets
1056 in the /lib/firmware directory that are used instead.
1057 Generic built-in EDID data sets are used, if one of
1058 edid/1024x768.bin, edid/1280x1024.bin,
1059 edid/1680x1050.bin, or edid/1920x1080.bin is given
1060 and no file with the same name exists. Details and
1061 instructions how to build your own EDID data are
1062 available in Documentation/admin-guide/edid.rst. An EDID
1063 data set will only be used for a particular connector,
1064 if its name and a colon are prepended to the EDID
1065 name. Each connector may use a unique EDID data
1066 set by separating the files with a comma. An EDID
1067 data set with no connector name will be used for
1068 any connectors not explicitly specified.
1073 Format: {"off" | "known"}
1074 Control how the dt_cpu_ftrs device-tree binding is
1075 used for CPU feature discovery and setup (if it
1077 off: Do not use it, fall back to legacy cpu table.
1078 known: Do not pass through unknown features to guests
1079 or userspace, only those that the kernel is aware of.
1081 dump_apple_properties [X86]
1082 Dump name and content of EFI device properties on
1083 x86 Macs. Useful for driver authors to determine
1084 what data is available or for reverse-engineering.
1086 dyndbg[="val"] [KNL,DYNAMIC_DEBUG]
1087 <module>.dyndbg[="val"]
1088 Enable debug messages at boot time. See
1089 Documentation/admin-guide/dynamic-debug-howto.rst
1092 nopku [X86] Disable Memory Protection Keys CPU feature found
1095 <module>.async_probe [KNL]
1096 Enable asynchronous probe on this module.
1098 early_ioremap_debug [KNL]
1099 Enable debug messages in early_ioremap support. This
1100 is useful for tracking down temporary early mappings
1101 which are not unmapped.
1103 earlycon= [KNL] Output early console device and options.
1105 When used with no options, the early console is
1106 determined by stdout-path property in device tree's
1107 chosen node or the ACPI SPCR table if supported by
1110 cdns,<addr>[,options]
1111 Start an early, polled-mode console on a Cadence
1112 (xuartps) serial port at the specified address. Only
1113 supported option is baud rate. If baud rate is not
1114 specified, the serial port must already be setup and
1117 uart[8250],io,<addr>[,options]
1118 uart[8250],mmio,<addr>[,options]
1119 uart[8250],mmio32,<addr>[,options]
1120 uart[8250],mmio32be,<addr>[,options]
1121 uart[8250],0x<addr>[,options]
1122 Start an early, polled-mode console on the 8250/16550
1123 UART at the specified I/O port or MMIO address.
1124 MMIO inter-register address stride is either 8-bit
1125 (mmio) or 32-bit (mmio32 or mmio32be).
1126 If none of [io|mmio|mmio32|mmio32be], <addr> is assumed
1127 to be equivalent to 'mmio'. 'options' are specified
1128 in the same format described for "console=ttyS<n>"; if
1129 unspecified, the h/w is not initialized.
1133 Start an early, polled-mode console on a pl011 serial
1134 port at the specified address. The pl011 serial port
1135 must already be setup and configured. Options are not
1136 yet supported. If 'mmio32' is specified, then only
1137 the driver will use only 32-bit accessors to read/write
1138 the device registers.
1141 Start an early console on a litex serial port at the
1142 specified address. The serial port must already be
1143 setup and configured. Options are not yet supported.
1146 Start an early, polled-mode console on a meson serial
1147 port at the specified address. The serial port must
1148 already be setup and configured. Options are not yet
1152 Start an early, polled-mode console on an msm serial
1153 port at the specified address. The serial port
1154 must already be setup and configured. Options are not
1157 msm_serial_dm,<addr>
1158 Start an early, polled-mode console on an msm serial
1159 dm port at the specified address. The serial port
1160 must already be setup and configured. Options are not
1164 Start an early, polled-mode console on a serial port
1165 of an Actions Semi SoC, such as S500 or S900, at the
1166 specified address. The serial port must already be
1167 setup and configured. Options are not yet supported.
1170 Start an early, polled-mode console on a serial port
1171 of an RDA Micro SoC, such as RDA8810PL, at the
1172 specified address. The serial port must already be
1173 setup and configured. Options are not yet supported.
1176 Use RISC-V SBI (Supervisor Binary Interface) for early
1179 smh Use ARM semihosting calls for early console.
1187 Use early console provided by serial driver available
1188 on Samsung SoCs, requires selecting proper type and
1189 a correct base address of the selected UART port. The
1190 serial port must already be setup and configured.
1191 Options are not yet supported.
1194 Start an early, polled-mode console on a lantiq serial
1195 (lqasc) port at the specified address. The serial port
1196 must already be setup and configured. Options are not
1201 Use early console provided by Freescale LP UART driver
1202 found on Freescale Vybrid and QorIQ LS1021A processors.
1203 A valid base address must be provided, and the serial
1204 port must already be setup and configured.
1208 Start an early, polled-mode, output-only console on the
1209 Freescale i.MX UART at the specified address. The UART
1210 must already be setup and configured.
1213 Start an early, polled-mode console on the
1214 Armada 3700 serial port at the specified
1215 address. The serial port must already be setup
1216 and configured. Options are not yet supported.
1219 Start an early, polled-mode console on a Qualcomm
1220 Generic Interface (GENI) based serial port at the
1221 specified address. The serial port must already be
1222 setup and configured. Options are not yet supported.
1225 Start an early, unaccelerated console on the EFI
1226 memory mapped framebuffer (if available). On cache
1227 coherent non-x86 systems that use system memory for
1228 the framebuffer, pass the 'ram' option so that it is
1229 mapped with the correct attributes.
1232 Use early console provided by Freescale LINFlexD UART
1233 serial driver for NXP S32V234 SoCs. A valid base
1234 address must be provided, and the serial port must
1235 already be setup and configured.
1237 earlyprintk= [X86,SH,ARM,M68k,S390]
1241 earlyprintk=serial[,ttySn[,baudrate]]
1242 earlyprintk=serial[,0x...[,baudrate]]
1243 earlyprintk=ttySn[,baudrate]
1244 earlyprintk=dbgp[debugController#]
1245 earlyprintk=pciserial[,force],bus:device.function[,baudrate]
1246 earlyprintk=xdbc[xhciController#]
1248 earlyprintk is useful when the kernel crashes before
1249 the normal console is initialized. It is not enabled by
1250 default because it has some cosmetic problems.
1252 Append ",keep" to not disable it when the real console
1255 Only one of vga, efi, serial, or usb debug port can
1258 Currently only ttyS0 and ttyS1 may be specified by
1259 name. Other I/O ports may be explicitly specified
1260 on some architectures (x86 and arm at least) by
1261 replacing ttySn with an I/O port address, like this:
1262 earlyprintk=serial,0x1008,115200
1263 You can find the port for a given device in
1264 /proc/tty/driver/serial:
1265 2: uart:ST16650V2 port:00001008 irq:18 ...
1267 Interaction with the standard serial driver is not
1270 The VGA and EFI output is eventually overwritten by
1273 The xen option can only be used in Xen domains.
1275 The sclp output can only be used on s390.
1277 The optional "force" to "pciserial" enables use of a
1278 PCI device even when its classcode is not of the
1281 edac_report= [HW,EDAC] Control how to report EDAC event
1282 Format: {"on" | "off" | "force"}
1283 on: enable EDAC to report H/W event. May be overridden
1284 by other higher priority error reporting module.
1285 off: disable H/W event reporting through EDAC.
1286 force: enforce the use of EDAC to report H/W event.
1289 ekgdboc= [X86,KGDB] Allow early kernel console debugging
1292 This is designed to be used in conjunction with
1293 the boot argument: earlyprintk=vga
1295 This parameter works in place of the kgdboc parameter
1296 but can only be used if the backing tty is available
1297 very early in the boot process. For early debugging
1298 via a serial port see kgdboc_earlycon instead.
1301 Format: {"off" | "on" | "skip[mbr]"}
1304 Format: { "debug", "disable_early_pci_dma",
1305 "nochunk", "noruntime", "nosoftreserve",
1306 "novamap", "no_disable_early_pci_dma" }
1307 debug: enable misc debug output.
1308 disable_early_pci_dma: disable the busmaster bit on all
1309 PCI bridges while in the EFI boot stub.
1310 nochunk: disable reading files in "chunks" in the EFI
1311 boot stub, as chunking can cause problems with some
1312 firmware implementations.
1313 noruntime : disable EFI runtime services support
1314 nosoftreserve: The EFI_MEMORY_SP (Specific Purpose)
1315 attribute may cause the kernel to reserve the
1316 memory range for a memory mapping driver to
1317 claim. Specify efi=nosoftreserve to disable this
1318 reservation and treat the memory by its base type
1319 (i.e. EFI_CONVENTIONAL_MEMORY / "System RAM").
1320 novamap: do not call SetVirtualAddressMap().
1321 no_disable_early_pci_dma: Leave the busmaster bit set
1322 on all PCI bridges while in the EFI boot stub
1324 efi_no_storage_paranoia [EFI; X86]
1325 Using this parameter you can use more than 50% of
1326 your efi variable storage. Use this parameter only if
1327 you are really sure that your UEFI does sane gc and
1328 fulfills the spec otherwise your board may brick.
1330 efi_fake_mem= nn[KMG]@ss[KMG]:aa[,nn[KMG]@ss[KMG]:aa,..] [EFI; X86]
1331 Add arbitrary attribute to specific memory range by
1332 updating original EFI memory map.
1333 Region of memory which aa attribute is added to is
1336 If efi_fake_mem=2G@4G:0x10000,2G@0x10a0000000:0x10000
1337 is specified, EFI_MEMORY_MORE_RELIABLE(0x10000)
1338 attribute is added to range 0x100000000-0x180000000 and
1339 0x10a0000000-0x1120000000.
1341 If efi_fake_mem=8G@9G:0x40000 is specified, the
1342 EFI_MEMORY_SP(0x40000) attribute is added to
1343 range 0x240000000-0x43fffffff.
1345 Using this parameter you can do debugging of EFI memmap
1346 related features. For example, you can do debugging of
1347 Address Range Mirroring feature even if your box
1348 doesn't support it, or mark specific memory as
1351 efivar_ssdt= [EFI; X86] Name of an EFI variable that contains an SSDT
1352 that is to be dynamically loaded by Linux. If there are
1353 multiple variables with the same name but with different
1354 vendor GUIDs, all of them will be loaded. See
1355 Documentation/admin-guide/acpi/ssdt-overlays.rst for details.
1358 eisa_irq_edge= [PARISC,HW]
1359 See header of drivers/parisc/eisa.c.
1362 See comment before function elanfreq_setup() in
1363 arch/x86/kernel/cpu/cpufreq/elanfreq.c.
1365 elfcorehdr=[size[KMG]@]offset[KMG] [IA64,PPC,SH,X86,S390]
1366 Specifies physical address of start of kernel core
1367 image elf header and optionally the size. Generally
1368 kexec loader will pass this option to capture kernel.
1369 See Documentation/admin-guide/kdump/kdump.rst for details.
1371 enable_mtrr_cleanup [X86]
1372 The kernel tries to adjust MTRR layout from continuous
1373 to discrete, to make X server driver able to add WB
1374 entry later. This parameter enables that.
1376 enable_timer_pin_1 [X86]
1377 Enable PIN 1 of APIC timer
1378 Can be useful to work around chipset bugs
1379 (in particular on some ATI chipsets).
1380 The kernel tries to set a reasonable default.
1382 enforcing [SELINUX] Set initial enforcing status.
1384 See security/selinux/Kconfig help text.
1385 0 -- permissive (log only, no denials).
1386 1 -- enforcing (deny and log).
1388 Value can be changed at runtime via
1389 /sys/fs/selinux/enforce.
1392 Disable Error Record Serialization Table (ERST)
1395 ether= [HW,NET] Ethernet cards parameters
1396 This option is obsoleted by the "netdev=" option, which
1397 has equivalent usage. See its documentation for details.
1401 Permit 'security.evm' to be updated regardless of
1402 current integrity status.
1407 fail_make_request=[KNL]
1408 General fault injection mechanism.
1409 Format: <interval>,<probability>,<space>,<times>
1410 See also Documentation/fault-injection/.
1413 Format: { initns | none }
1414 See Documentation/admin-guide/sysctl/net.rst for
1415 fb_tunnels_only_for_init_ns
1418 See Documentation/admin-guide/blockdev/floppy.rst.
1420 force_pal_cache_flush
1421 [IA-64] Avoid check_sal_cache_flush which may hang on
1422 buggy SAL_CACHE_FLUSH implementations. Using this
1423 parameter will force ia64_sal_cache_flush to call
1424 ia64_pal_cache_flush instead of SAL_CACHE_FLUSH.
1427 Forcefully enable Physical Address Extension (PAE).
1428 Many Pentium M systems disable PAE but may have a
1429 functionally usable PAE implementation.
1430 Warning: use of this parameter will taint the kernel
1431 and may cause unknown problems.
1434 [FTRACE] will set and start the specified tracer
1435 as early as possible in order to facilitate early
1438 ftrace_dump_on_oops[=orig_cpu]
1439 [FTRACE] will dump the trace buffers on oops.
1440 If no parameter is passed, ftrace will dump
1441 buffers of all CPUs, but if you pass orig_cpu, it will
1442 dump only the buffer of the CPU that triggered the
1445 ftrace_filter=[function-list]
1446 [FTRACE] Limit the functions traced by the function
1447 tracer at boot up. function-list is a comma-separated
1448 list of functions. This list can be changed at run
1449 time by the set_ftrace_filter file in the debugfs
1452 ftrace_notrace=[function-list]
1453 [FTRACE] Do not trace the functions specified in
1454 function-list. This list can be changed at run time
1455 by the set_ftrace_notrace file in the debugfs
1458 ftrace_graph_filter=[function-list]
1459 [FTRACE] Limit the top level callers functions traced
1460 by the function graph tracer at boot up.
1461 function-list is a comma-separated list of functions
1462 that can be changed at run time by the
1463 set_graph_function file in the debugfs tracing directory.
1465 ftrace_graph_notrace=[function-list]
1466 [FTRACE] Do not trace from the functions specified in
1467 function-list. This list is a comma-separated list of
1468 functions that can be changed at run time by the
1469 set_graph_notrace file in the debugfs tracing directory.
1471 ftrace_graph_max_depth=<uint>
1472 [FTRACE] Used with the function graph tracer. This is
1473 the max depth it will trace into a function. This value
1474 can be changed at run time by the max_graph_depth file
1475 in the tracefs tracing directory. default: 0 (no limit)
1477 fw_devlink= [KNL] Create device links between consumer and supplier
1478 devices by scanning the firmware to infer the
1479 consumer/supplier relationships. This feature is
1480 especially useful when drivers are loaded as modules as
1481 it ensures proper ordering of tasks like device probing
1482 (suppliers first, then consumers), supplier boot state
1483 clean up (only after all consumers have probed),
1484 suspend/resume & runtime PM (consumers first, then
1486 Format: { off | permissive | on | rpm }
1487 off -- Don't create device links from firmware info.
1488 permissive -- Create device links from firmware info
1489 but use it only for ordering boot state clean
1490 up (sync_state() calls).
1491 on -- Create device links from firmware info and use it
1492 to enforce probe and suspend/resume ordering.
1493 rpm -- Like "on", but also use to order runtime PM.
1495 fw_devlink.strict=<bool>
1496 [KNL] Treat all inferred dependencies as mandatory
1497 dependencies. This only applies for fw_devlink=on|rpm.
1501 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad
1502 support via parallel port (up to 5 devices per port)
1503 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5>
1504 See also Documentation/input/devices/joystick-parport.rst
1508 gart_fix_e820= [X86-64] disable the fix e820 for K8 GART
1512 gcov_persist= [GCOV] When non-zero (default), profiling data for
1513 kernel modules is saved and remains accessible via
1514 debugfs, even when the module is unloaded/reloaded.
1515 When zero, profiling data is discarded and associated
1516 debugfs files are removed at module unload time.
1518 goldfish [X86] Enable the goldfish android emulator platform.
1519 Don't use this when you are not running on the
1522 gpio-mockup.gpio_mockup_ranges
1523 [HW] Sets the ranges of gpiochip of for this device.
1524 Format: <start1>,<end1>,<start2>,<end2>...
1525 gpio-mockup.gpio_mockup_named_lines
1526 [HW] Let the driver know GPIO lines should be named.
1528 gpt [EFI] Forces disk with valid GPT signature but
1529 invalid Protective MBR to be treated as GPT. If the
1530 primary GPT is corrupted, it enables the backup/alternate
1531 GPT to be used instead.
1533 grcan.enable0= [HW] Configuration of physical interface 0. Determines
1534 the "Enable 0" bit of the configuration register.
1537 grcan.enable1= [HW] Configuration of physical interface 1. Determines
1538 the "Enable 0" bit of the configuration register.
1541 grcan.select= [HW] Select which physical interface to use.
1544 grcan.txsize= [HW] Sets the size of the tx buffer.
1545 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0.
1547 grcan.rxsize= [HW] Sets the size of the rx buffer.
1548 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0.
1551 hardlockup_all_cpu_backtrace=
1552 [KNL] Should the hard-lockup detector generate
1553 backtraces on all cpus.
1556 hashdist= [KNL,NUMA] Large hashes allocated during boot
1557 are distributed across NUMA nodes. Defaults on
1558 for 64-bit NUMA, off otherwise.
1559 Format: 0 | 1 (for off | on)
1561 hcl= [IA-64] SGI's Hardware Graph compatibility layer
1563 hd= [EIDE] (E)IDE hard drive subsystem geometry
1564 Format: <cyl>,<head>,<sect>
1567 Disable Hardware Error Source Table (HEST) support;
1568 corresponding firmware-first mode error processing
1569 logic will be disabled.
1571 highmem=nn[KMG] [KNL,BOOT] forces the highmem zone to have an exact
1572 size of <nn>. This works even on boxes that have no
1573 highmem otherwise. This also works to reduce highmem
1574 size on bigger boxes.
1576 highres= [KNL] Enable/disable high resolution timer mode.
1577 Valid parameters: "on", "off"
1582 hpet= [X86-32,HPET] option to control HPET usage
1583 Format: { enable (default) | disable | force |
1585 disable: disable HPET and use PIT instead
1586 force: allow force enabled of undocumented chips (ICH4,
1588 verbose: show contents of HPET registers during setup
1590 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET
1591 registers. Default set by CONFIG_HPET_MMAP_DEFAULT.
1593 hugetlb_cma= [HW,CMA] The size of a CMA area used for allocation
1594 of gigantic hugepages. Or using node format, the size
1595 of a CMA area per node can be specified.
1596 Format: nn[KMGTPE] or (node format)
1597 <node>:nn[KMGTPE][,<node>:nn[KMGTPE]]
1599 Reserve a CMA area of given size and allocate gigantic
1600 hugepages using the CMA allocator. If enabled, the
1601 boot-time allocation of gigantic hugepages is skipped.
1603 hugepages= [HW] Number of HugeTLB pages to allocate at boot.
1604 If this follows hugepagesz (below), it specifies
1605 the number of pages of hugepagesz to be allocated.
1606 If this is the first HugeTLB parameter on the command
1607 line, it specifies the number of pages to allocate for
1608 the default huge page size. If using node format, the
1609 number of pages to allocate per-node can be specified.
1610 See also Documentation/admin-guide/mm/hugetlbpage.rst.
1611 Format: <integer> or (node format)
1612 <node>:<integer>[,<node>:<integer>]
1615 [HW] The size of the HugeTLB pages. This is used in
1616 conjunction with hugepages (above) to allocate huge
1617 pages of a specific size at boot. The pair
1618 hugepagesz=X hugepages=Y can be specified once for
1619 each supported huge page size. Huge page sizes are
1620 architecture dependent. See also
1621 Documentation/admin-guide/mm/hugetlbpage.rst.
1624 hugetlb_free_vmemmap=
1625 [KNL] Reguires CONFIG_HUGETLB_PAGE_FREE_VMEMMAP
1627 Allows heavy hugetlb users to free up some more
1628 memory (6 * PAGE_SIZE for each 2MB hugetlb page).
1629 Format: { on | off (default) }
1631 on: enable the feature
1632 off: disable the feature
1634 Built with CONFIG_HUGETLB_PAGE_FREE_VMEMMAP_DEFAULT_ON=y,
1637 This is not compatible with memory_hotplug.memmap_on_memory.
1638 If both parameters are enabled, hugetlb_free_vmemmap takes
1639 precedence over memory_hotplug.memmap_on_memory.
1642 [KNL] Should the hung task detector generate panics.
1645 A value of 1 instructs the kernel to panic when a
1646 hung task is detected. The default value is controlled
1647 by the CONFIG_BOOTPARAM_HUNG_TASK_PANIC build-time
1648 option. The value selected by this boot parameter can
1649 be changed later by the kernel.hung_task_panic sysctl.
1651 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC)
1652 terminal devices. Valid values: 0..8
1653 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs.
1654 If specified, z/VM IUCV HVC accepts connections
1655 from listed z/VM user IDs only.
1657 hv_nopvspin [X86,HYPER_V] Disables the paravirt spinlock optimizations
1658 which allow the hypervisor to 'idle' the
1659 guest on lock contention.
1662 Do not unregister boot console at start. This is only
1663 useful for debugging when something happens in the window
1664 between unregistering the boot console and initializing
1667 i2c_bus= [HW] Override the default board specific I2C bus speed
1668 or register an additional I2C bus that is not
1669 registered from board initialization code.
1673 i8042.debug [HW] Toggle i8042 debug mode
1674 i8042.unmask_kbd_data
1675 [HW] Enable printing of interrupt data from the KBD port
1676 (disabled by default, and as a pre-condition
1677 requires that i8042.debug=1 be enabled)
1678 i8042.direct [HW] Put keyboard port into non-translated mode
1679 i8042.dumbkbd [HW] Pretend that controller can only read data from
1680 keyboard and cannot control its state
1681 (Don't attempt to blink the leds)
1682 i8042.noaux [HW] Don't check for auxiliary (== mouse) port
1683 i8042.nokbd [HW] Don't check/create keyboard port
1684 i8042.noloop [HW] Disable the AUX Loopback command while probing
1686 i8042.nomux [HW] Don't check presence of an active multiplexing
1688 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX
1690 i8042.notimeout [HW] Ignore timeout condition signalled by controller
1691 i8042.reset [HW] Reset the controller during init, cleanup and
1692 suspend-to-ram transitions, only during s2r
1693 transitions, or never reset
1694 Format: { 1 | Y | y | 0 | N | n }
1695 1, Y, y: always reset controller
1696 0, N, n: don't ever reset controller
1697 Default: only on s2r transitions on x86; most other
1698 architectures force reset to be always executed
1699 i8042.unlock [HW] Unlock (ignore) the keylock
1700 i8042.kbdreset [HW] Reset device connected to KBD port
1704 i8k.ignore_dmi [HW] Continue probing hardware even if DMI data
1705 indicates that the driver is running on unsupported
1707 i8k.force [HW] Activate i8k driver even if SMM BIOS signature
1708 does not match list of supported models.
1710 [HW] Report power status in /proc/i8k
1711 (disabled by default)
1712 i8k.restricted [HW] Allow controlling fans only if SYS_ADMIN
1715 i915.invert_brightness=
1716 [DRM] Invert the sense of the variable that is used to
1717 set the brightness of the panel backlight. Normally a
1718 brightness value of 0 indicates backlight switched off,
1719 and the maximum of the brightness value sets the backlight
1720 to maximum brightness. If this parameter is set to 0
1721 (default) and the machine requires it, or this parameter
1722 is set to 1, a brightness value of 0 sets the backlight
1723 to maximum brightness, and the maximum of the brightness
1724 value switches the backlight off.
1725 -1 -- never invert brightness
1726 0 -- machine default
1727 1 -- force brightness inversion
1730 Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]]
1732 ide-core.nodma= [HW] (E)IDE subsystem
1733 Format: =0.0 to prevent dma on hda, =0.1 hdb =1.0 hdc
1734 .vlb_clock .pci_clock .noflush .nohpa .noprobe .nowerr
1735 .cdrom .chs .ignore_cable are additional options
1736 See Documentation/ide/ide.rst.
1738 ide-generic.probe-mask= [HW] (E)IDE subsystem
1740 Probe mask for legacy ISA IDE ports. Depending on
1741 platform up to 6 ports are supported, enabled by
1742 setting corresponding bits in the mask to 1. The
1743 default value is 0x0, which has a special meaning.
1744 On systems that have PCI, it triggers scanning the
1745 PCI bus for the first and the second port, which
1746 are then probed. On systems without PCI the value
1747 of 0x0 enables probing the two first ports as if it
1750 ide-pci-generic.all-generic-ide [HW] (E)IDE subsystem
1751 Claim all unknown PCI IDE storage controllers.
1754 Format: idle=poll, idle=halt, idle=nomwait
1755 Poll forces a polling idle loop that can slightly
1756 improve the performance of waking up a idle CPU, but
1757 will use a lot of power and make the system run hot.
1759 idle=halt: Halt is forced to be used for CPU idle.
1760 In such case C2/C3 won't be used again.
1761 idle=nomwait: Disable mwait for CPU C-states
1765 Allow force disabling of Shared Virtual Memory (SVA)
1766 support for the idxd driver. By default it is set to
1769 idxd.tc_override= [HW]
1771 Allow override of default traffic class configuration
1772 for the device. By default it is set to false (0).
1774 ieee754= [MIPS] Select IEEE Std 754 conformance mode
1775 Format: { strict | legacy | 2008 | relaxed }
1778 Choose which programs will be accepted for execution
1779 based on the IEEE 754 NaN encoding(s) supported by
1780 the FPU and the NaN encoding requested with the value
1781 of an ELF file header flag individually set by each
1782 binary. Hardware implementations are permitted to
1783 support either or both of the legacy and the 2008 NaN
1786 Available settings are as follows:
1787 strict accept binaries that request a NaN encoding
1788 supported by the FPU
1789 legacy only accept legacy-NaN binaries, if supported
1791 2008 only accept 2008-NaN binaries, if supported
1793 relaxed accept any binaries regardless of whether
1794 supported by the FPU
1796 The FPU emulator is always able to support both NaN
1797 encodings, so if no FPU hardware is present or it has
1798 been disabled with 'nofpu', then the settings of
1799 'legacy' and '2008' strap the emulator accordingly,
1800 'relaxed' straps the emulator for both legacy-NaN and
1801 2008-NaN, whereas 'strict' enables legacy-NaN only on
1802 legacy processors and both NaN encodings on MIPS32 or
1805 The setting for ABS.fmt/NEG.fmt instruction execution
1806 mode generally follows that for the NaN encoding,
1807 except where unsupported by hardware.
1809 ignore_loglevel [KNL]
1810 Ignore loglevel setting - this will print /all/
1811 kernel messages to the console. Useful for debugging.
1812 We also add it as printk module parameter, so users
1813 could change it dynamically, usually by
1814 /sys/module/printk/parameters/ignore_loglevel.
1817 Ignore RLIMIT_DATA setting for data mappings,
1818 print warning at first misuse. Can be changed via
1819 /sys/module/kernel/parameters/ignore_rlimit_data.
1821 ihash_entries= [KNL]
1822 Set number of hash buckets for inode cache.
1824 ima_appraise= [IMA] appraise integrity measurements
1825 Format: { "off" | "enforce" | "fix" | "log" }
1828 ima_appraise_tcb [IMA] Deprecated. Use ima_policy= instead.
1829 The builtin appraise policy appraises all files
1832 ima_canonical_fmt [IMA]
1833 Use the canonical format for the binary runtime
1834 measurements, instead of host native format.
1837 Format: { md5 | sha1 | rmd160 | sha256 | sha384
1841 The list of supported hash algorithms is defined
1842 in crypto/hash_info.h.
1845 The builtin policies to load during IMA setup.
1846 Format: "tcb | appraise_tcb | secure_boot |
1847 fail_securely | critical_data"
1849 The "tcb" policy measures all programs exec'd, files
1850 mmap'd for exec, and all files opened with the read
1851 mode bit set by either the effective uid (euid=0) or
1854 The "appraise_tcb" policy appraises the integrity of
1855 all files owned by root.
1857 The "secure_boot" policy appraises the integrity
1858 of files (eg. kexec kernel image, kernel modules,
1859 firmware, policy, etc) based on file signatures.
1861 The "fail_securely" policy forces file signature
1862 verification failure also on privileged mounted
1863 filesystems with the SB_I_UNVERIFIABLE_SIGNATURE
1866 The "critical_data" policy measures kernel integrity
1869 ima_tcb [IMA] Deprecated. Use ima_policy= instead.
1870 Load a policy which meets the needs of the Trusted
1871 Computing Base. This means IMA will measure all
1872 programs exec'd, files mmap'd for exec, and all files
1873 opened for read by uid=0.
1876 Select one of defined IMA measurements template formats.
1877 Formats: { "ima" | "ima-ng" | "ima-sig" }
1881 [IMA] Define a custom template format.
1882 Format: { "field1|...|fieldN" }
1884 ima.ahash_minsize= [IMA] Minimum file size for asynchronous hash usage
1885 Format: <min_file_size>
1886 Set the minimal file size for using asynchronous hash.
1887 If left unspecified, ahash usage is disabled.
1889 ahash performance varies for different data sizes on
1890 different crypto accelerators. This option can be used
1891 to achieve the best performance for a particular HW.
1893 ima.ahash_bufsize= [IMA] Asynchronous hash buffer size
1895 Set hashing buffer size. Default: 4k.
1897 ahash performance varies for different chunk sizes on
1898 different crypto accelerators. This option can be used
1899 to achieve best performance for particular HW.
1903 Run specified binary instead of /sbin/init as init
1906 initcall_debug [KNL] Trace initcalls as they are executed. Useful
1907 for working out where the kernel is dying during
1910 initcall_blacklist= [KNL] Do not execute a comma-separated list of
1911 initcall functions. Useful for debugging built-in
1912 modules and initcalls.
1914 initramfs_async= [KNL]
1917 This parameter controls whether the initramfs
1918 image is unpacked asynchronously, concurrently
1919 with devices being probed and
1920 initialized. This should normally just work,
1921 but as a debugging aid, one can get the
1922 historical behaviour of the initramfs
1923 unpacking being completed before device_ and
1926 initrd= [BOOT] Specify the location of the initial ramdisk
1928 initrdmem= [KNL] Specify a physical address and size from which to
1929 load the initrd. If an initrd is compiled in or
1930 specified in the bootparams, it takes priority over this
1932 Format: ss[KMG],nn[KMG]
1935 init_on_alloc= [MM] Fill newly allocated pages and heap objects with
1938 Default set by CONFIG_INIT_ON_ALLOC_DEFAULT_ON.
1940 init_on_free= [MM] Fill freed pages and heap objects with zeroes.
1942 Default set by CONFIG_INIT_ON_FREE_DEFAULT_ON.
1944 init_pkru= [X86] Specify the default memory protection keys rights
1945 register contents for all processes. 0x55555554 by
1946 default (disallow access to all but pkey 0). Can
1947 override in debugfs after boot.
1949 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver
1952 int_pln_enable [X86] Enable power limit notification interrupt
1954 integrity_audit=[IMA]
1955 Format: { "0" | "1" }
1956 0 -- basic integrity auditing messages. (Default)
1957 1 -- additional integrity auditing messages.
1959 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option
1961 Enable intel iommu driver.
1963 Disable intel iommu driver.
1964 igfx_off [Default Off]
1965 By default, gfx is mapped as normal device. If a gfx
1966 device has a dedicated DMAR unit, the DMAR unit is
1967 bypassed by not enabling DMAR with this option. In
1968 this case, gfx device will use physical address for
1970 strict [Default Off]
1971 Deprecated, equivalent to iommu.strict=1.
1972 sp_off [Default Off]
1973 By default, super page will be supported if Intel IOMMU
1974 has the capability. With this option, super page will
1977 Enable the Intel IOMMU scalable mode if the hardware
1978 advertises that it has support for the scalable mode
1981 Disallow use of the Intel IOMMU scalable mode.
1982 tboot_noforce [Default Off]
1983 Do not force the Intel IOMMU enabled under tboot.
1984 By default, tboot will force Intel IOMMU on, which
1985 could harm performance of some high-throughput
1986 devices like 40GBit network cards, even if identity
1988 Note that using this option lowers the security
1989 provided by tboot because it makes the system
1990 vulnerable to DMA attacks.
1992 intel_idle.max_cstate= [KNL,HW,ACPI,X86]
1993 0 disables intel_idle and fall back on acpi_idle.
1994 1 to 9 specify maximum depth of C-state.
1998 Do not enable intel_pstate as the default
1999 scaling driver for the supported processors
2001 Use intel_pstate as a scaling driver, but configure it
2002 to work with generic cpufreq governors (instead of
2003 enabling its internal governor). This mode cannot be
2004 used along with the hardware-managed P-states (HWP)
2007 Enable intel_pstate on systems that prohibit it by default
2008 in favor of acpi-cpufreq. Forcing the intel_pstate driver
2009 instead of acpi-cpufreq may disable platform features, such
2010 as thermal controls and power capping, that rely on ACPI
2011 P-States information being indicated to OSPM and therefore
2012 should be used with caution. This option does not work with
2013 processors that aren't supported by the intel_pstate driver
2014 or on platforms that use pcc-cpufreq instead of acpi-cpufreq.
2016 Do not enable hardware P state control (HWP)
2019 Only load intel_pstate on systems which support
2020 hardware P state control (HWP) if available.
2022 Enforce ACPI _PPC performance limits. If the Fixed ACPI
2023 Description Table, specifies preferred power management
2024 profile as "Enterprise Server" or "Performance Server",
2025 then this feature is turned on by default.
2027 Allow per-logical-CPU P-State performance control limits using
2028 cpufreq sysfs interface
2030 intremap= [X86-64, Intel-IOMMU]
2031 on enable Interrupt Remapping (default)
2032 off disable Interrupt Remapping
2033 nosid disable Source ID checking
2035 BIOS x2APIC opt-out request will be ignored
2036 nopost disable Interrupt Posting
2038 iomem= Disable strict checking of access to MMIO memory
2039 strict regions from userspace.
2054 nobypass [PPC/POWERNV]
2055 Disable IOMMU bypass, using IOMMU for PCI devices.
2057 iommu.forcedac= [ARM64, X86] Control IOVA allocation for PCI devices.
2058 Format: { "0" | "1" }
2059 0 - Try to allocate a 32-bit DMA address first, before
2060 falling back to the full range if needed.
2061 1 - Allocate directly from the full usable range,
2062 forcing Dual Address Cycle for PCI cards supporting
2063 greater than 32-bit addressing.
2065 iommu.strict= [ARM64, X86] Configure TLB invalidation behaviour
2066 Format: { "0" | "1" }
2068 Request that DMA unmap operations use deferred
2069 invalidation of hardware TLBs, for increased
2070 throughput at the cost of reduced device isolation.
2071 Will fall back to strict mode if not supported by
2072 the relevant IOMMU driver.
2074 DMA unmap operations invalidate IOMMU hardware TLBs
2076 unset - Use value of CONFIG_IOMMU_DEFAULT_DMA_{LAZY,STRICT}.
2077 Note: on x86, strict mode specified via one of the
2078 legacy driver-specific options takes precedence.
2081 [ARM64, X86] Configure DMA to bypass the IOMMU by default.
2082 Format: { "0" | "1" }
2083 0 - Use IOMMU translation for DMA.
2084 1 - Bypass the IOMMU for DMA.
2085 unset - Use value of CONFIG_IOMMU_DEFAULT_PASSTHROUGH.
2087 io7= [HW] IO7 for Marvel-based Alpha systems
2088 See comment before marvel_specify_io7 in
2089 arch/alpha/kernel/core_marvel.c.
2091 io_delay= [X86] I/O delay method
2093 Standard port 0x80 based delay
2095 Alternate port 0xed based delay (needed on some systems)
2097 Simple two microseconds delay
2102 See Documentation/admin-guide/nfs/nfsroot.rst.
2104 ipcmni_extend [KNL] Extend the maximum number of unique System V
2105 IPC identifiers from 32,768 to 16,777,216.
2107 irqaffinity= [SMP] Set the default irq affinity mask
2108 The argument is a cpu list, as described above.
2110 irqchip.gicv2_force_probe=
2113 Force the kernel to look for the second 4kB page
2114 of a GICv2 controller even if the memory range
2115 exposed by the device tree is too small.
2117 irqchip.gicv3_nolpi=
2119 Force the kernel to ignore the availability of
2120 LPIs (and by consequence ITSs). Intended for system
2121 that use the kernel as a bootloader, and thus want
2122 to let secondary kernels in charge of setting up
2125 irqchip.gicv3_pseudo_nmi= [ARM64]
2126 Enables support for pseudo-NMIs in the kernel. This
2127 requires the kernel to be built with
2128 CONFIG_ARM64_PSEUDO_NMI.
2131 When an interrupt is not handled search all handlers
2132 for it. Intended to get systems with badly broken
2136 When an interrupt is not handled search all handlers
2137 for it. Also check all handlers each timer
2138 interrupt. Intended to get systems with badly broken
2142 Format: <RDP>,<reset>,<pci_scan>,<verbosity>
2144 isolcpus= [KNL,SMP,ISOL] Isolate a given set of CPUs from disturbance.
2145 [Deprecated - use cpusets instead]
2146 Format: [flag-list,]<cpu-list>
2148 Specify one or more CPUs to isolate from disturbances
2149 specified in the flag list (default: domain):
2152 Disable the tick when a single task runs.
2154 A residual 1Hz tick is offloaded to workqueues, which you
2155 need to affine to housekeeping through the global
2156 workqueue's affinity configured via the
2157 /sys/devices/virtual/workqueue/cpumask sysfs file, or
2158 by using the 'domain' flag described below.
2160 NOTE: by default the global workqueue runs on all CPUs,
2161 so to protect individual CPUs the 'cpumask' file has to
2162 be configured manually after bootup.
2165 Isolate from the general SMP balancing and scheduling
2166 algorithms. Note that performing domain isolation this way
2167 is irreversible: it's not possible to bring back a CPU to
2168 the domains once isolated through isolcpus. It's strongly
2169 advised to use cpusets instead to disable scheduler load
2170 balancing through the "cpuset.sched_load_balance" file.
2171 It offers a much more flexible interface where CPUs can
2172 move in and out of an isolated set anytime.
2174 You can move a process onto or off an "isolated" CPU via
2175 the CPU affinity syscalls or cpuset.
2176 <cpu number> begins at 0 and the maximum value is
2177 "number of CPUs in system - 1".
2181 Isolate from being targeted by managed interrupts
2182 which have an interrupt mask containing isolated
2183 CPUs. The affinity of managed interrupts is
2184 handled by the kernel and cannot be changed via
2185 the /proc/irq/* interfaces.
2187 This isolation is best effort and only effective
2188 if the automatically assigned interrupt mask of a
2189 device queue contains isolated and housekeeping
2190 CPUs. If housekeeping CPUs are online then such
2191 interrupts are directed to the housekeeping CPU
2192 so that IO submitted on the housekeeping CPU
2193 cannot disturb the isolated CPU.
2195 If a queue's affinity mask contains only isolated
2196 CPUs then this parameter has no effect on the
2197 interrupt routing decision, though interrupts are
2198 only delivered when tasks running on those
2199 isolated CPUs submit IO. IO submitted on
2200 housekeeping CPUs has no influence on those
2203 The format of <cpu-list> is described above.
2207 ivrs_ioapic [HW,X86-64]
2208 Provide an override to the IOAPIC-ID<->DEVICE-ID
2209 mapping provided in the IVRS ACPI table. For
2210 example, to map IOAPIC-ID decimal 10 to
2211 PCI device 00:14.0 write the parameter as:
2212 ivrs_ioapic[10]=00:14.0
2214 ivrs_hpet [HW,X86-64]
2215 Provide an override to the HPET-ID<->DEVICE-ID
2216 mapping provided in the IVRS ACPI table. For
2217 example, to map HPET-ID decimal 0 to
2218 PCI device 00:14.0 write the parameter as:
2219 ivrs_hpet[0]=00:14.0
2221 ivrs_acpihid [HW,X86-64]
2222 Provide an override to the ACPI-HID:UID<->DEVICE-ID
2223 mapping provided in the IVRS ACPI table. For
2224 example, to map UART-HID:UID AMD0020:0 to
2225 PCI device 00:14.5 write the parameter as:
2226 ivrs_acpihid[00:14.5]=AMD0020:0
2228 js= [HW,JOY] Analog joystick
2229 See Documentation/input/joydev/joystick.rst.
2232 When CONFIG_RANDOMIZE_BASE is set, this disables
2233 kernel and module base offset ASLR (Address Space
2234 Layout Randomization).
2237 [KNL] Enforce KASAN (Kernel Address Sanitizer) to print
2238 report on every invalid memory access. Without this
2239 parameter KASAN will print report only for the first
2244 kernelcore= [KNL,X86,IA-64,PPC]
2245 Format: nn[KMGTPE] | nn% | "mirror"
2246 This parameter specifies the amount of memory usable by
2247 the kernel for non-movable allocations. The requested
2248 amount is spread evenly throughout all nodes in the
2249 system as ZONE_NORMAL. The remaining memory is used for
2250 movable memory in its own zone, ZONE_MOVABLE. In the
2251 event, a node is too small to have both ZONE_NORMAL and
2252 ZONE_MOVABLE, kernelcore memory will take priority and
2253 other nodes will have a larger ZONE_MOVABLE.
2255 ZONE_MOVABLE is used for the allocation of pages that
2256 may be reclaimed or moved by the page migration
2257 subsystem. Note that allocations like PTEs-from-HighMem
2258 still use the HighMem zone if it exists, and the Normal
2259 zone if it does not.
2261 It is possible to specify the exact amount of memory in
2262 the form of "nn[KMGTPE]", a percentage of total system
2263 memory in the form of "nn%", or "mirror". If "mirror"
2264 option is specified, mirrored (reliable) memory is used
2265 for non-movable allocations and remaining memory is used
2266 for Movable pages. "nn[KMGTPE]", "nn%", and "mirror"
2267 are exclusive, so you cannot specify multiple forms.
2269 kgdbdbgp= [KGDB,HW] kgdb over EHCI usb debug port.
2270 Format: <Controller#>[,poll interval]
2271 The controller # is the number of the ehci usb debug
2272 port as it is probed via PCI. The poll interval is
2273 optional and is the number seconds in between
2274 each poll cycle to the debug port in case you need
2275 the functionality for interrupting the kernel with
2276 gdb or control-c on the dbgp connection. When
2277 not using this parameter you use sysrq-g to break into
2278 the kernel debugger.
2280 kgdboc= [KGDB,HW] kgdb over consoles.
2281 Requires a tty driver that supports console polling,
2282 or a supported polling keyboard driver (non-usb).
2283 Serial only format: <serial_device>[,baud]
2284 keyboard only format: kbd
2285 keyboard and serial format: kbd,<serial_device>[,baud]
2286 Optional Kernel mode setting:
2287 kms, kbd format: kms,kbd
2288 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud]
2290 kgdboc_earlycon= [KGDB,HW]
2291 If the boot console provides the ability to read
2292 characters and can work in polling mode, you can use
2293 this parameter to tell kgdb to use it as a backend
2294 until the normal console is registered. Intended to
2295 be used together with the kgdboc parameter which
2296 specifies the normal console to transition to.
2298 The name of the early console should be specified
2299 as the value of this parameter. Note that the name of
2300 the early console might be different than the tty
2301 name passed to kgdboc. It's OK to leave the value
2302 blank and the first boot console that implements
2303 read() will be picked.
2305 kgdbwait [KGDB] Stop kernel execution and enter the
2306 kernel debugger at the earliest opportunity.
2308 kmac= [MIPS] Korina ethernet MAC address.
2309 Configure the RouterBoard 532 series on-chip
2310 Ethernet adapter MAC address.
2312 kmemleak= [KNL] Boot-time kmemleak enable/disable
2313 Valid arguments: on, off
2315 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y,
2318 kprobe_event=[probe-list]
2319 [FTRACE] Add kprobe events and enable at boot time.
2320 The probe-list is a semicolon delimited list of probe
2321 definitions. Each definition is same as kprobe_events
2322 interface, but the parameters are comma delimited.
2323 For example, to add a kprobe event on vfs_read with
2324 arg1 and arg2, add to the command line;
2326 kprobe_event=p,vfs_read,$arg1,$arg2
2328 See also Documentation/trace/kprobetrace.rst "Kernel
2329 Boot Parameter" section.
2331 kpti= [ARM64] Control page table isolation of user
2332 and kernel address spaces.
2333 Default: enabled on cores which need mitigation.
2337 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs.
2338 Default is 0 (don't ignore, but inject #GP)
2340 kvm.enable_vmware_backdoor=[KVM] Support VMware backdoor PV interface.
2341 Default is false (don't support).
2343 kvm.mmu_audit= [KVM] This is a R/W parameter which allows audit
2348 [KVM] Controls the software workaround for the
2349 X86_BUG_ITLB_MULTIHIT bug.
2350 force : Always deploy workaround.
2351 off : Never deploy workaround.
2352 auto : Deploy workaround based on the presence of
2353 X86_BUG_ITLB_MULTIHIT.
2357 If the software workaround is enabled for the host,
2358 guests do need not to enable it for nested guests.
2360 kvm.nx_huge_pages_recovery_ratio=
2361 [KVM] Controls how many 4KiB pages are periodically zapped
2362 back to huge pages. 0 disables the recovery, otherwise if
2363 the value is N KVM will zap 1/Nth of the 4KiB pages every
2364 period (see below). The default is 60.
2366 kvm.nx_huge_pages_recovery_period_ms=
2367 [KVM] Controls the time period at which KVM zaps 4KiB pages
2368 back to huge pages. If the value is a non-zero N, KVM will
2369 zap a portion (see ratio above) of the pages every N msecs.
2370 If the value is 0 (the default), KVM will pick a period based
2371 on the ratio, such that a page is zapped after 1 hour on average.
2373 kvm-amd.nested= [KVM,AMD] Allow nested virtualization in KVM/SVM.
2374 Default is 1 (enabled)
2376 kvm-amd.npt= [KVM,AMD] Disable nested paging (virtualized MMU)
2378 Default is 1 (enabled) if in 64-bit or 32-bit PAE mode.
2381 [KVM,ARM] Select one of KVM/arm64's modes of operation.
2383 none: Forcefully disable KVM.
2385 nvhe: Standard nVHE-based mode, without support for
2388 protected: nVHE-based mode with support for guests whose
2389 state is kept private from the host.
2390 Not valid if the kernel is running in EL2.
2392 Defaults to VHE/nVHE based on hardware support. Setting
2393 mode to "protected" will disable kexec and hibernation
2396 kvm-arm.vgic_v3_group0_trap=
2397 [KVM,ARM] Trap guest accesses to GICv3 group-0
2400 kvm-arm.vgic_v3_group1_trap=
2401 [KVM,ARM] Trap guest accesses to GICv3 group-1
2404 kvm-arm.vgic_v3_common_trap=
2405 [KVM,ARM] Trap guest accesses to GICv3 common
2408 kvm-arm.vgic_v4_enable=
2409 [KVM,ARM] Allow use of GICv4 for direct injection of
2412 kvm_cma_resv_ratio=n [PPC]
2413 Reserves given percentage from system memory area for
2414 contiguous memory allocation for KVM hash pagetable
2416 By default it reserves 5% of total system memory.
2420 kvm-intel.ept= [KVM,Intel] Disable extended page tables
2421 (virtualized MMU) support on capable Intel chips.
2422 Default is 1 (enabled)
2424 kvm-intel.emulate_invalid_guest_state=
2425 [KVM,Intel] Enable emulation of invalid guest states
2426 Default is 0 (disabled)
2428 kvm-intel.flexpriority=
2429 [KVM,Intel] Disable FlexPriority feature (TPR shadow).
2430 Default is 1 (enabled)
2433 [KVM,Intel] Enable VMX nesting (nVMX).
2434 Default is 0 (disabled)
2436 kvm-intel.unrestricted_guest=
2437 [KVM,Intel] Disable unrestricted guest feature
2438 (virtualized real and unpaged mode) on capable
2439 Intel chips. Default is 1 (enabled)
2441 kvm-intel.vmentry_l1d_flush=[KVM,Intel] Mitigation for L1 Terminal Fault
2444 Valid arguments: never, cond, always
2446 always: L1D cache flush on every VMENTER.
2447 cond: Flush L1D on VMENTER only when the code between
2448 VMEXIT and VMENTER can leak host memory.
2449 never: Disables the mitigation
2451 Default is cond (do L1 cache flush in specific instances)
2453 kvm-intel.vpid= [KVM,Intel] Disable Virtual Processor Identification
2454 feature (tagged TLBs) on capable Intel chips.
2455 Default is 1 (enabled)
2457 l1d_flush= [X86,INTEL]
2458 Control mitigation for L1D based snooping vulnerability.
2460 Certain CPUs are vulnerable to an exploit against CPU
2461 internal buffers which can forward information to a
2462 disclosure gadget under certain conditions.
2464 In vulnerable processors, the speculatively
2465 forwarded data can be used in a cache side channel
2466 attack, to access data to which the attacker does
2467 not have direct access.
2469 This parameter controls the mitigation. The
2472 on - enable the interface for the mitigation
2474 l1tf= [X86] Control mitigation of the L1TF vulnerability on
2477 The kernel PTE inversion protection is unconditionally
2478 enabled and cannot be disabled.
2481 Provides all available mitigations for the
2482 L1TF vulnerability. Disables SMT and
2483 enables all mitigations in the
2484 hypervisors, i.e. unconditional L1D flush.
2486 SMT control and L1D flush control via the
2487 sysfs interface is still possible after
2488 boot. Hypervisors will issue a warning
2489 when the first VM is started in a
2490 potentially insecure configuration,
2491 i.e. SMT enabled or L1D flush disabled.
2494 Same as 'full', but disables SMT and L1D
2495 flush runtime control. Implies the
2496 'nosmt=force' command line option.
2497 (i.e. sysfs control of SMT is disabled.)
2500 Leaves SMT enabled and enables the default
2501 hypervisor mitigation, i.e. conditional
2504 SMT control and L1D flush control via the
2505 sysfs interface is still possible after
2506 boot. Hypervisors will issue a warning
2507 when the first VM is started in a
2508 potentially insecure configuration,
2509 i.e. SMT enabled or L1D flush disabled.
2513 Disables SMT and enables the default
2514 hypervisor mitigation.
2516 SMT control and L1D flush control via the
2517 sysfs interface is still possible after
2518 boot. Hypervisors will issue a warning
2519 when the first VM is started in a
2520 potentially insecure configuration,
2521 i.e. SMT enabled or L1D flush disabled.
2524 Same as 'flush', but hypervisors will not
2525 warn when a VM is started in a potentially
2526 insecure configuration.
2529 Disables hypervisor mitigations and doesn't
2531 It also drops the swap size and available
2532 RAM limit restriction on both hypervisor and
2537 For details see: Documentation/admin-guide/hw-vuln/l1tf.rst
2543 lapic [X86-32,APIC] Enable the local APIC even if BIOS
2546 lapic= [X86,APIC] Do not use TSC deadline
2547 value for LAPIC timer one-shot implementation. Default
2548 back to the programmable timer unit in the LAPIC.
2549 Format: notscdeadline
2551 lapic_timer_c2_ok [X86,APIC] trust the local apic timer
2554 libata.dma= [LIBATA] DMA control
2555 libata.dma=0 Disable all PATA and SATA DMA
2556 libata.dma=1 PATA and SATA Disk DMA only
2557 libata.dma=2 ATAPI (CDROM) DMA only
2558 libata.dma=4 Compact Flash DMA only
2559 Combinations also work, so libata.dma=3 enables DMA
2560 for disks and CDROMs, but not CFs.
2562 libata.ignore_hpa= [LIBATA] Ignore HPA limit
2563 libata.ignore_hpa=0 keep BIOS limits (default)
2564 libata.ignore_hpa=1 ignore limits, using full disk
2566 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume
2570 libata.force= [LIBATA] Force configurations. The format is comma-
2571 separated list of "[ID:]VAL" where ID is
2572 PORT[.DEVICE]. PORT and DEVICE are decimal numbers
2573 matching port, link or device. Basically, it matches
2574 the ATA ID string printed on console by libata. If
2575 the whole ID part is omitted, the last PORT and DEVICE
2576 values are used. If ID hasn't been specified yet, the
2577 configuration applies to all ports, links and devices.
2579 If only DEVICE is omitted, the parameter applies to
2580 the port and all links and devices behind it. DEVICE
2581 number of 0 either selects the first device or the
2582 first fan-out link behind PMP device. It does not
2583 select the host link. DEVICE number of 15 selects the
2584 host link and device attached to it.
2586 The VAL specifies the configuration to force. As long
2587 as there's no ambiguity shortcut notation is allowed.
2588 For example, both 1.5 and 1.5G would work for 1.5Gbps.
2589 The following configurations can be forced.
2591 * Cable type: 40c, 80c, short40c, unk, ign or sata.
2592 Any ID with matching PORT is used.
2594 * SATA link speed limit: 1.5Gbps or 3.0Gbps.
2596 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7].
2597 udma[/][16,25,33,44,66,100,133] notation is also
2600 * [no]ncq: Turn on or off NCQ.
2602 * [no]ncqtrim: Turn off queued DSM TRIM.
2604 * nohrst, nosrst, norst: suppress hard, soft
2607 * rstonce: only attempt one reset during
2608 hot-unplug link recovery
2610 * dump_id: dump IDENTIFY data.
2612 * atapi_dmadir: Enable ATAPI DMADIR bridge support
2614 * disable: Disable this device.
2616 If there are multiple matching configurations changing
2617 the same attribute, the last one is used.
2619 memblock=debug [KNL] Enable memblock debug messages.
2621 load_ramdisk= [RAM] [Deprecated]
2623 lockd.nlm_grace_period=P [NFS] Assign grace period.
2626 lockd.nlm_tcpport=N [NFS] Assign TCP port.
2629 lockd.nlm_timeout=T [NFS] Assign timeout value.
2632 lockd.nlm_udpport=M [NFS] Assign UDP port.
2635 lockdown= [SECURITY]
2636 { integrity | confidentiality }
2637 Enable the kernel lockdown feature. If set to
2638 integrity, kernel features that allow userland to
2639 modify the running kernel are disabled. If set to
2640 confidentiality, kernel features that allow userland
2641 to extract confidential information from the kernel
2644 locktorture.nreaders_stress= [KNL]
2645 Set the number of locking read-acquisition kthreads.
2646 Defaults to being automatically set based on the
2647 number of online CPUs.
2649 locktorture.nwriters_stress= [KNL]
2650 Set the number of locking write-acquisition kthreads.
2652 locktorture.onoff_holdoff= [KNL]
2653 Set time (s) after boot for CPU-hotplug testing.
2655 locktorture.onoff_interval= [KNL]
2656 Set time (s) between CPU-hotplug operations, or
2657 zero to disable CPU-hotplug testing.
2659 locktorture.shuffle_interval= [KNL]
2660 Set task-shuffle interval (jiffies). Shuffling
2661 tasks allows some CPUs to go into dyntick-idle
2662 mode during the locktorture test.
2664 locktorture.shutdown_secs= [KNL]
2665 Set time (s) after boot system shutdown. This
2666 is useful for hands-off automated testing.
2668 locktorture.stat_interval= [KNL]
2669 Time (s) between statistics printk()s.
2671 locktorture.stutter= [KNL]
2672 Time (s) to stutter testing, for example,
2673 specifying five seconds causes the test to run for
2674 five seconds, wait for five seconds, and so on.
2675 This tests the locking primitive's ability to
2676 transition abruptly to and from idle.
2678 locktorture.torture_type= [KNL]
2679 Specify the locking implementation to test.
2681 locktorture.verbose= [KNL]
2682 Enable additional printk() statements.
2684 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver
2687 loglevel= All Kernel Messages with a loglevel smaller than the
2688 console loglevel will be printed to the console. It can
2689 also be changed with klogd or other programs. The
2690 loglevels are defined as follows:
2692 0 (KERN_EMERG) system is unusable
2693 1 (KERN_ALERT) action must be taken immediately
2694 2 (KERN_CRIT) critical conditions
2695 3 (KERN_ERR) error conditions
2696 4 (KERN_WARNING) warning conditions
2697 5 (KERN_NOTICE) normal but significant condition
2698 6 (KERN_INFO) informational
2699 7 (KERN_DEBUG) debug-level messages
2701 log_buf_len=n[KMG] Sets the size of the printk ring buffer,
2702 in bytes. n must be a power of two and greater
2703 than the minimal size. The minimal size is defined
2704 by LOG_BUF_SHIFT kernel config parameter. There is
2705 also CONFIG_LOG_CPU_MAX_BUF_SHIFT config parameter
2706 that allows to increase the default size depending on
2707 the number of CPUs. See init/Kconfig for more details.
2709 logo.nologo [FB] Disables display of the built-in Linux logo.
2710 This may be used to provide more screen space for
2711 kernel log messages and is useful when debugging
2712 kernel boot problems.
2714 lp=0 [LP] Specify parallel ports to use, e.g,
2715 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses
2716 lp=reset first parallel port). 'lp=0' disables the
2717 lp=auto printer driver. 'lp=reset' (which can be
2718 specified in addition to the ports) causes
2719 attached printers to be reset. Using
2720 lp=port1,port2,... specifies the parallel ports
2721 to associate lp devices with, starting with
2722 lp0. A port specification may be 'none' to skip
2723 that lp device, or a parport name such as
2724 'parport0'. Specifying 'lp=auto' instead of a
2725 port specification list means that device IDs
2726 from each port should be examined, to see if
2727 an IEEE 1284-compliant printer is attached; if
2728 so, the driver will manage that printer.
2729 See also header of drivers/char/lp.c.
2732 Sets loops_per_jiffy to given constant, thus avoiding
2733 time-consuming boot-time autodetection (up to 250 ms per
2734 CPU). 0 enables autodetection (default). To determine
2735 the correct value for your kernel, boot with normal
2736 autodetection and see what value is printed. Note that
2737 on SMP systems the preset will be applied to all CPUs,
2738 which is likely to cause problems if your CPUs need
2739 significantly divergent settings. An incorrect value
2740 will cause delays in the kernel to be wrong, leading to
2741 unpredictable I/O errors and other breakage. Although
2742 unlikely, in the extreme case this might damage your
2746 Format: <io>,<irq>,<dma>
2748 lsm.debug [SECURITY] Enable LSM initialization debugging output.
2751 [SECURITY] Choose order of LSM initialization. This
2752 overrides CONFIG_LSM, and the "security=" parameter.
2754 machvec= [IA-64] Force the use of a particular machine-vector
2755 (machvec) in a generic kernel.
2756 Example: machvec=hpzx1
2758 machtype= [Loongson] Share the same kernel image file between
2759 different yeeloong laptops.
2760 Example: machtype=lemote-yeeloong-2f-7inch
2762 max_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory greater
2763 than or equal to this physical address is ignored.
2765 maxcpus= [SMP] Maximum number of processors that an SMP kernel
2766 will bring up during bootup. maxcpus=n : n >= 0 limits
2767 the kernel to bring up 'n' processors. Surely after
2768 bootup you can bring up the other plugged cpu by executing
2769 "echo 1 > /sys/devices/system/cpu/cpuX/online". So maxcpus
2770 only takes effect during system bootup.
2771 While n=0 is a special case, it is equivalent to "nosmp",
2772 which also disables the IO APIC.
2774 max_loop= [LOOP] The number of loop block devices that get
2775 (loop.max_loop) unconditionally pre-created at init time. The default
2776 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead
2777 of statically allocating a predefined number, loop
2778 devices can be requested on-demand with the
2779 /dev/loop-control interface.
2781 mce [X86-32] Machine Check Exception
2783 mce=option [X86-64] See Documentation/x86/x86_64/boot-options.rst
2785 md= [HW] RAID subsystems devices and level
2786 See Documentation/admin-guide/md.rst.
2789 Format: <first>,<last>
2790 Specifies range of consoles to be captured by the MDA.
2793 Control mitigation for the Micro-architectural Data
2794 Sampling (MDS) vulnerability.
2796 Certain CPUs are vulnerable to an exploit against CPU
2797 internal buffers which can forward information to a
2798 disclosure gadget under certain conditions.
2800 In vulnerable processors, the speculatively
2801 forwarded data can be used in a cache side channel
2802 attack, to access data to which the attacker does
2803 not have direct access.
2805 This parameter controls the MDS mitigation. The
2808 full - Enable MDS mitigation on vulnerable CPUs
2809 full,nosmt - Enable MDS mitigation and disable
2810 SMT on vulnerable CPUs
2811 off - Unconditionally disable MDS mitigation
2813 On TAA-affected machines, mds=off can be prevented by
2814 an active TAA mitigation as both vulnerabilities are
2815 mitigated with the same mechanism so in order to disable
2816 this mitigation, you need to specify tsx_async_abort=off
2819 Not specifying this option is equivalent to
2822 For details see: Documentation/admin-guide/hw-vuln/mds.rst
2824 mem=nn[KMG] [KNL,BOOT] Force usage of a specific amount of memory
2825 Amount of memory to be used in cases as follows:
2828 2 when the kernel is not able to see the whole system memory;
2829 3 memory that lies after 'mem=' boundary is excluded from
2830 the hypervisor, then assigned to KVM guests.
2832 [X86] Work as limiting max address. Use together
2833 with memmap= to avoid physical address space collisions.
2834 Without memmap= PCI devices could be placed at addresses
2835 belonging to unused RAM.
2837 Note that this only takes effects during boot time since
2838 in above case 3, memory may need be hot added after boot
2839 if system memory of hypervisor is not sufficient.
2841 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel
2845 [KNL,SH] Allow user to override the default size for
2846 per-device physically contiguous DMA buffers.
2848 memhp_default_state=online/offline
2849 [KNL] Set the initial state for the memory hotplug
2850 onlining policy. If not specified, the default value is
2851 set according to the
2852 CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE kernel config
2854 See Documentation/admin-guide/mm/memory-hotplug.rst.
2856 memmap=exactmap [KNL,X86] Enable setting of an exact
2857 E820 memory map, as specified by the user.
2858 Such memmap=exactmap lines can be constructed based on
2859 BIOS output or other requirements. See the memmap=nn@ss
2862 memmap=nn[KMG]@ss[KMG]
2863 [KNL, X86, MIPS, XTENSA] Force usage of a specific region of memory.
2864 Region of memory to be used is from ss to ss+nn.
2865 If @ss[KMG] is omitted, it is equivalent to mem=nn[KMG],
2866 which limits max address to nn[KMG].
2867 Multiple different regions can be specified,
2870 memmap=100M@2G,100M#3G,1G!1024G
2872 memmap=nn[KMG]#ss[KMG]
2873 [KNL,ACPI] Mark specific memory as ACPI data.
2874 Region of memory to be marked is from ss to ss+nn.
2876 memmap=nn[KMG]$ss[KMG]
2877 [KNL,ACPI] Mark specific memory as reserved.
2878 Region of memory to be reserved is from ss to ss+nn.
2879 Example: Exclude memory from 0x18690000-0x1869ffff
2880 memmap=64K$0x18690000
2882 memmap=0x10000$0x18690000
2883 Some bootloaders may need an escape character before '$',
2884 like Grub2, otherwise '$' and the following number
2887 memmap=nn[KMG]!ss[KMG]
2888 [KNL,X86] Mark specific memory as protected.
2889 Region of memory to be used, from ss to ss+nn.
2890 The memory region may be marked as e820 type 12 (0xc)
2891 and is NVDIMM or ADR memory.
2893 memmap=<size>%<offset>-<oldtype>+<newtype>
2894 [KNL,ACPI] Convert memory within the specified region
2895 from <oldtype> to <newtype>. If "-<oldtype>" is left
2896 out, the whole region will be marked as <newtype>,
2897 even if previously unavailable. If "+<newtype>" is left
2898 out, matching memory will be removed. Types are
2899 specified as e820 types, e.g., 1 = RAM, 2 = reserved,
2900 3 = ACPI, 12 = PRAM.
2902 memory_corruption_check=0/1 [X86]
2903 Some BIOSes seem to corrupt the first 64k of
2904 memory when doing things like suspend/resume.
2905 Setting this option will scan the memory
2906 looking for corruption. Enabling this will
2907 both detect corruption and prevent the kernel
2908 from using the memory being corrupted.
2909 However, its intended as a diagnostic tool; if
2910 repeatable BIOS-originated corruption always
2911 affects the same memory, you can use memmap=
2912 to prevent the kernel from using that memory.
2914 memory_corruption_check_size=size [X86]
2915 By default it checks for corruption in the low
2916 64k, making this memory unavailable for normal
2917 use. Use this parameter to scan for
2918 corruption in more or less memory.
2920 memory_corruption_check_period=seconds [X86]
2921 By default it checks for corruption every 60
2922 seconds. Use this parameter to check at some
2923 other rate. 0 disables periodic checking.
2925 memory_hotplug.memmap_on_memory
2926 [KNL,X86,ARM] Boolean flag to enable this feature.
2927 Format: {on | off (default)}
2928 When enabled, runtime hotplugged memory will
2929 allocate its internal metadata (struct pages)
2930 from the hotadded memory which will allow to
2931 hotadd a lot of memory without requiring
2932 additional memory to do so.
2933 This feature is disabled by default because it
2934 has some implication on large (e.g. GB)
2935 allocations in some configurations (e.g. small
2937 The state of the flag can be read in
2938 /sys/module/memory_hotplug/parameters/memmap_on_memory.
2939 Note that even when enabled, there are a few cases where
2940 the feature is not effective.
2942 This is not compatible with hugetlb_free_vmemmap. If
2943 both parameters are enabled, hugetlb_free_vmemmap takes
2944 precedence over memory_hotplug.memmap_on_memory.
2946 memtest= [KNL,X86,ARM,PPC,RISCV] Enable memtest
2948 default : 0 <disable>
2949 Specifies the number of memtest passes to be
2950 performed. Each pass selects another test
2951 pattern from a given set of patterns. Memtest
2952 fills the memory with this pattern, validates
2953 memory contents and reserves bad memory
2954 regions that are detected.
2956 mem_encrypt= [X86-64] AMD Secure Memory Encryption (SME) control
2957 Valid arguments: on, off
2958 Default (depends on kernel configuration option):
2959 on (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=y)
2960 off (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=n)
2961 mem_encrypt=on: Activate SME
2962 mem_encrypt=off: Do not activate SME
2964 Refer to Documentation/virt/kvm/amd-memory-encryption.rst
2965 for details on when memory encryption can be activated.
2967 mem_sleep_default= [SUSPEND] Default system suspend mode:
2968 s2idle - Suspend-To-Idle
2969 shallow - Power-On Suspend or equivalent (if supported)
2970 deep - Suspend-To-RAM or equivalent (if supported)
2971 See Documentation/admin-guide/pm/sleep-states.rst.
2973 meye.*= [HW] Set MotionEye Camera parameters
2974 See Documentation/admin-guide/media/meye.rst.
2976 mfgpt_irq= [IA-32] Specify the IRQ to use for the
2977 Multi-Function General Purpose Timers on AMD Geode
2980 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when
2981 the BIOS has incorrectly applied a workaround. TinyBIOS
2982 version 0.98 is known to be affected, 0.99 fixes the
2983 problem by letting the user disable the workaround.
2987 min_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory below this
2988 physical address is ignored.
2990 mini2440= [ARM,HW,KNL]
2991 Format:[0..2][b][c][t]
2993 MINI2440 configuration specification:
2994 0 - The attached screen is the 3.5" TFT
2995 1 - The attached screen is the 7" TFT
2996 2 - The VGA Shield is attached (1024x768)
2997 Leaving out the screen size parameter will not load
2998 the TFT driver, and the framebuffer will be left
3000 b - Enable backlight. The TFT backlight pin will be
3001 linked to the kernel VESA blanking code and a GPIO
3002 LED. This parameter is not necessary when using the
3004 c - Enable the s3c camera interface.
3005 t - Reserved for enabling touchscreen support. The
3006 touchscreen support is not enabled in the mainstream
3007 kernel as of 2.6.30, a preliminary port can be found
3008 in the "bleeding edge" mini2440 support kernel at
3009 https://repo.or.cz/w/linux-2.6/mini2440.git
3012 [X86,PPC,S390,ARM64] Control optional mitigations for
3013 CPU vulnerabilities. This is a set of curated,
3014 arch-independent options, each of which is an
3015 aggregation of existing arch-specific options.
3018 Disable all optional CPU mitigations. This
3019 improves system performance, but it may also
3020 expose users to several CPU vulnerabilities.
3021 Equivalent to: nopti [X86,PPC]
3023 nospectre_v1 [X86,PPC]
3025 nospectre_v2 [X86,PPC,S390,ARM64]
3026 spectre_v2_user=off [X86]
3027 spec_store_bypass_disable=off [X86,PPC]
3028 ssbd=force-off [ARM64]
3031 tsx_async_abort=off [X86]
3032 kvm.nx_huge_pages=off [X86]
3033 no_entry_flush [PPC]
3034 no_uaccess_flush [PPC]
3037 This does not have any effect on
3038 kvm.nx_huge_pages when
3039 kvm.nx_huge_pages=force.
3042 Mitigate all CPU vulnerabilities, but leave SMT
3043 enabled, even if it's vulnerable. This is for
3044 users who don't want to be surprised by SMT
3045 getting disabled across kernel upgrades, or who
3046 have other ways of avoiding SMT-based attacks.
3047 Equivalent to: (default behavior)
3050 Mitigate all CPU vulnerabilities, disabling SMT
3051 if needed. This is for users who always want to
3052 be fully mitigated, even if it means losing SMT.
3053 Equivalent to: l1tf=flush,nosmt [X86]
3054 mds=full,nosmt [X86]
3055 tsx_async_abort=full,nosmt [X86]
3058 [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this
3059 parameter allows control of the logging verbosity for
3060 the additional memory initialisation checks. A value
3061 of 0 disables mminit logging and a level of 4 will
3062 log everything. Information is printed at KERN_DEBUG
3063 so loglevel=8 may also need to be specified.
3066 [KNL] When CONFIG_MODULE_SIG is set, this means that
3067 modules without (valid) signatures will fail to load.
3068 Note that if CONFIG_MODULE_SIG_FORCE is set, that
3069 is always true, so this option does nothing.
3071 module_blacklist= [KNL] Do not load a comma-separated list of
3072 modules. Useful for debugging problem modules.
3075 [MOUSE] Maximum time between finger touching and
3076 leaving touchpad surface for touch to be considered
3077 a tap and be reported as a left button click (for
3078 touchpads working in absolute mode only).
3080 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices
3081 reporting absolute coordinates, such as tablets
3082 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices
3083 reporting absolute coordinates, such as tablets
3085 movablecore= [KNL,X86,IA-64,PPC]
3086 Format: nn[KMGTPE] | nn%
3087 This parameter is the complement to kernelcore=, it
3088 specifies the amount of memory used for migratable
3089 allocations. If both kernelcore and movablecore is
3090 specified, then kernelcore will be at *least* the
3091 specified value but may be more. If movablecore on its
3092 own is specified, the administrator must be careful
3093 that the amount of memory usable for all allocations
3096 movable_node [KNL] Boot-time switch to make hotplugable memory
3097 NUMA nodes to be movable. This means that the memory
3098 of such nodes will be usable only for movable
3099 allocations which rules out almost all kernel
3100 allocations. Use with caution!
3102 MTD_Partition= [MTD]
3103 Format: <name>,<region-number>,<size>,<offset>
3105 MTD_Region= [MTD] Format:
3106 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>]
3109 See drivers/mtd/parsers/cmdlinepart.c
3111 multitce=off [PPC] This parameter disables the use of the pSeries
3112 firmware feature for updating multiple TCE entries
3115 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration
3117 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock]
3119 boundary - index of last SLC block on Flex-OneNAND.
3120 The remaining blocks are configured as MLC blocks.
3121 lock - Configure if Flex-OneNAND boundary should be locked.
3122 Once locked, the boundary cannot be changed.
3123 1 indicates lock status, 0 indicates unlock status.
3126 ARM/S3C2412 JIVE boot control
3128 See arch/arm/mach-s3c/mach-jive.c
3130 mtouchusb.raw_coordinates=
3131 [HW] Make the MicroTouch USB driver use raw coordinates
3132 ('y', default) or cooked coordinates ('n')
3134 mtrr_chunk_size=nn[KMG] [X86]
3135 used for mtrr cleanup. It is largest continuous chunk
3136 that could hold holes aka. UC entries.
3138 mtrr_gran_size=nn[KMG] [X86]
3139 Used for mtrr cleanup. It is granularity of mtrr block.
3141 Large value could prevent small alignment from
3144 mtrr_spare_reg_nr=n [X86]
3146 Range: 0,7 : spare reg number
3148 Used for mtrr cleanup. It is spare mtrr entries number.
3149 Set to 2 or more if your graphical card needs more.
3151 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card
3153 netdev= [NET] Network devices parameters
3154 Format: <irq>,<io>,<mem_start>,<mem_end>,<name>
3155 Note that mem_start is often overloaded to mean
3156 something different and driver-specific.
3157 This usage is only documented in each driver source
3161 [NETFILTER] Enable connection tracking flow accounting
3162 0 to disable accounting
3163 1 to enable accounting
3166 nfsaddrs= [NFS] Deprecated. Use ip= instead.
3167 See Documentation/admin-guide/nfs/nfsroot.rst.
3169 nfsroot= [NFS] nfs root filesystem for disk-less boxes.
3170 See Documentation/admin-guide/nfs/nfsroot.rst.
3172 nfsrootdebug [NFS] enable nfsroot debugging messages.
3173 See Documentation/admin-guide/nfs/nfsroot.rst.
3175 nfs.callback_nr_threads=
3176 [NFSv4] set the total number of threads that the
3177 NFS client will assign to service NFSv4 callback
3180 nfs.callback_tcpport=
3181 [NFS] set the TCP port on which the NFSv4 callback
3182 channel should listen.
3185 [NFS] sets the pathname to the program which is used
3186 to update the NFS client cache entries.
3188 nfs.cache_getent_timeout=
3189 [NFS] sets the timeout after which an attempt to
3190 update a cache entry is deemed to have failed.
3192 nfs.idmap_cache_timeout=
3193 [NFS] set the maximum lifetime for idmapper cache
3197 [NFS] enable 64-bit inode numbers.
3198 If zero, the NFS client will fake up a 32-bit inode
3199 number for the readdir() and stat() syscalls instead
3200 of returning the full 64-bit number.
3201 The default is to return 64-bit inode numbers.
3203 nfs.max_session_cb_slots=
3204 [NFSv4.1] Sets the maximum number of session
3205 slots the client will assign to the callback
3206 channel. This determines the maximum number of
3207 callbacks the client will process in parallel for
3208 a particular server.
3210 nfs.max_session_slots=
3211 [NFSv4.1] Sets the maximum number of session slots
3212 the client will attempt to negotiate with the server.
3213 This limits the number of simultaneous RPC requests
3214 that the client can send to the NFSv4.1 server.
3215 Note that there is little point in setting this
3216 value higher than the max_tcp_slot_table_limit.
3218 nfs.nfs4_disable_idmapping=
3219 [NFSv4] When set to the default of '1', this option
3220 ensures that both the RPC level authentication
3221 scheme and the NFS level operations agree to use
3222 numeric uids/gids if the mount is using the
3223 'sec=sys' security flavour. In effect it is
3224 disabling idmapping, which can make migration from
3225 legacy NFSv2/v3 systems to NFSv4 easier.
3226 Servers that do not support this mode of operation
3227 will be autodetected by the client, and it will fall
3228 back to using the idmapper.
3229 To turn off this behaviour, set the value to '0'.
3231 [NFS4] Specify an additional fixed unique ident-
3232 ification string that NFSv4 clients can insert into
3233 their nfs_client_id4 string. This is typically a
3234 UUID that is generated at system install time.
3236 nfs.send_implementation_id =
3237 [NFSv4.1] Send client implementation identification
3238 information in exchange_id requests.
3239 If zero, no implementation identification information
3241 The default is to send the implementation identification
3244 nfs.recover_lost_locks =
3245 [NFSv4] Attempt to recover locks that were lost due
3246 to a lease timeout on the server. Please note that
3247 doing this risks data corruption, since there are
3248 no guarantees that the file will remain unchanged
3249 after the locks are lost.
3250 If you want to enable the kernel legacy behaviour of
3251 attempting to recover these locks, then set this
3253 The default parameter value of '0' causes the kernel
3254 not to attempt recovery of lost locks.
3256 nfs4.layoutstats_timer =
3257 [NFSv4.2] Change the rate at which the kernel sends
3258 layoutstats to the pNFS metadata server.
3260 Setting this to value to 0 causes the kernel to use
3261 whatever value is the default set by the layout
3262 driver. A non-zero value sets the minimum interval
3263 in seconds between layoutstats transmissions.
3265 nfsd.inter_copy_offload_enable =
3266 [NFSv4.2] When set to 1, the server will support
3267 server-to-server copies for which this server is
3268 the destination of the copy.
3270 nfsd.nfsd4_ssc_umount_timeout =
3271 [NFSv4.2] When used as the destination of a
3272 server-to-server copy, knfsd temporarily mounts
3273 the source server. It caches the mount in case
3274 it will be needed again, and discards it if not
3275 used for the number of milliseconds specified by
3278 nfsd.nfs4_disable_idmapping=
3279 [NFSv4] When set to the default of '1', the NFSv4
3280 server will return only numeric uids and gids to
3281 clients using auth_sys, and will accept numeric uids
3282 and gids from such clients. This is intended to ease
3283 migration from NFSv2/v3.
3286 nmi_backtrace.backtrace_idle [KNL]
3287 Dump stacks even of idle CPUs in response to an
3288 NMI stack-backtrace request.
3290 nmi_debug= [KNL,SH] Specify one or more actions to take
3291 when a NMI is triggered.
3292 Format: [state][,regs][,debounce][,die]
3294 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels
3295 Format: [panic,][nopanic,][num]
3297 0 - turn hardlockup detector in nmi_watchdog off
3298 1 - turn hardlockup detector in nmi_watchdog on
3299 When panic is specified, panic when an NMI watchdog
3300 timeout occurs (or 'nopanic' to not panic on an NMI
3301 watchdog, if CONFIG_BOOTPARAM_HARDLOCKUP_PANIC is set)
3302 To disable both hard and soft lockup detectors,
3303 please see 'nowatchdog'.
3304 This is useful when you use a panic=... timeout and
3305 need the box quickly up again.
3307 These settings can be accessed at runtime via
3308 the nmi_watchdog and hardlockup_panic sysctls.
3310 netpoll.carrier_timeout=
3311 [NET] Specifies amount of time (in seconds) that
3312 netpoll should wait for a carrier. By default netpoll
3315 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths
3316 emulation library even if a 387 maths coprocessor
3319 no5lvl [X86-64] Disable 5-level paging mode. Forces
3320 kernel to use 4-level paging instead.
3322 nofsgsbase [X86] Disables FSGSBASE instructions.
3325 [HW] Never suspend the console
3326 Disable suspending of consoles during suspend and
3327 hibernate operations. Once disabled, debugging
3328 messages can reach various consoles while the rest
3329 of the system is being put to sleep (ie, while
3330 debugging driver suspend/resume hooks). This may
3331 not work reliably with all consoles, but is known
3332 to work with serial and VGA consoles.
3333 To facilitate more flexible debugging, we also add
3334 console_suspend, a printk module parameter to control
3335 it. Users could use console_suspend (usually
3336 /sys/module/printk/parameters/console_suspend) to
3337 turn on/off it dynamically.
3339 novmcoredd [KNL,KDUMP]
3340 Disable device dump. Device dump allows drivers to
3341 append dump data to vmcore so you can collect driver
3342 specified debug info. Drivers can append the data
3343 without any limit and this data is stored in memory,
3344 so this may cause significant memory stress. Disabling
3345 device dump can help save memory but the driver debug
3346 data will be no longer available. This parameter
3347 is only available when CONFIG_PROC_VMCORE_DEVICE_DUMP
3350 noaliencache [MM, NUMA, SLAB] Disables the allocation of alien
3351 caches in the slab allocator. Saves per-node memory,
3352 but will impact performance.
3356 noaltinstr [S390] Disables alternative instructions patching
3357 (CPU alternatives feature).
3359 noapic [SMP,APIC] Tells the kernel to not make use of any
3360 IOAPICs that may be present in the system.
3362 noautogroup Disable scheduler automatic task group creation.
3364 nobats [PPC] Do not use BATs for mapping kernel lowmem
3365 on "Classic" PPC cores.
3369 noclflush [BUGS=X86] Don't use the CLFLUSH instruction
3371 delayacct [KNL] Enable per-task delay accounting
3373 nodsp [SH] Disable hardware DSP at boot time.
3375 noefi Disable EFI runtime services support.
3377 no_entry_flush [PPC] Don't flush the L1-D cache when entering the kernel.
3382 On X86-32 available only on PAE configured kernels.
3383 noexec=on: enable non-executable mappings (default)
3384 noexec=off: disable non-executable mappings
3387 Disable SMAP (Supervisor Mode Access Prevention)
3388 even if it is supported by processor.
3391 Disable SMEP (Supervisor Mode Execution Prevention)
3392 even if it is supported by processor.
3395 This affects only 32-bit executables.
3396 noexec32=on: enable non-executable mappings (default)
3397 read doesn't imply executable mappings
3398 noexec32=off: disable non-executable mappings
3399 read implies executable mappings
3401 nofpu [MIPS,SH] Disable hardware FPU at boot time.
3403 nofxsr [BUGS=X86-32] Disables x86 floating point extended
3404 register save and restore. The kernel will only save
3405 legacy floating-point registers on task switch.
3407 nohugeiomap [KNL,X86,PPC,ARM64] Disable kernel huge I/O mappings.
3409 nohugevmalloc [PPC] Disable kernel huge vmalloc mappings.
3411 nosmt [KNL,S390] Disable symmetric multithreading (SMT).
3412 Equivalent to smt=1.
3414 [KNL,X86] Disable symmetric multithreading (SMT).
3415 nosmt=force: Force disable SMT, cannot be undone
3416 via the sysfs control file.
3418 nospectre_v1 [X86,PPC] Disable mitigations for Spectre Variant 1
3419 (bounds check bypass). With this option data leaks are
3420 possible in the system.
3422 nospectre_v2 [X86,PPC_FSL_BOOK3E,ARM64] Disable all mitigations for
3423 the Spectre variant 2 (indirect branch prediction)
3424 vulnerability. System may allow data leaks with this
3427 nospec_store_bypass_disable
3428 [HW] Disable all mitigations for the Speculative Store Bypass vulnerability
3431 [PPC] Don't flush the L1-D cache after accessing user data.
3433 noxsave [BUGS=X86] Disables x86 extended register state save
3434 and restore using xsave. The kernel will fallback to
3435 enabling legacy floating-point and sse state.
3437 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended
3438 register states. The kernel will fall back to use
3439 xsave to save the states. By using this parameter,
3440 performance of saving the states is degraded because
3441 xsave doesn't support modified optimization while
3442 xsaveopt supports it on xsaveopt enabled systems.
3444 noxsaves [X86] Disables xsaves and xrstors used in saving and
3445 restoring x86 extended register state in compacted
3446 form of xsave area. The kernel will fall back to use
3447 xsaveopt and xrstor to save and restore the states
3448 in standard form of xsave area. By using this
3449 parameter, xsave area per process might occupy more
3450 memory on xsaves enabled systems.
3452 nohlt [ARM,ARM64,MICROBLAZE,SH] Forces the kernel to busy wait
3453 in do_idle() and not use the arch_cpu_idle()
3454 implementation; requires CONFIG_GENERIC_IDLE_POLL_SETUP
3455 to be effective. This is useful on platforms where the
3456 sleep(SH) or wfi(ARM,ARM64) instructions do not work
3457 correctly or when doing power measurements to evalute
3458 the impact of the sleep instructions. This is also
3459 useful when using JTAG debugger.
3461 no_file_caps Tells the kernel not to honor file capabilities. The
3462 only way then for a file to be executed with privilege
3463 is to be setuid root or executed by root.
3465 nohalt [IA-64] Tells the kernel not to use the power saving
3466 function PAL_HALT_LIGHT when idle. This increases
3467 power-consumption. On the positive side, it reduces
3468 interrupt wake-up latency, which may improve performance
3469 in certain environments such as networked servers or
3473 Force pointers printed to the console or buffers to be
3474 unhashed. By default, when a pointer is printed via %p
3475 format string, that pointer is "hashed", i.e. obscured
3476 by hashing the pointer value. This is a security feature
3477 that hides actual kernel addresses from unprivileged
3478 users, but it also makes debugging the kernel more
3479 difficult since unequal pointers can no longer be
3480 compared. However, if this command-line option is
3481 specified, then all normal pointers will have their true
3482 value printed. Pointers printed via %pK may still be
3483 hashed. This option should only be specified when
3484 debugging the kernel. Please do not use on production
3487 nohibernate [HIBERNATION] Disable hibernation and resume.
3489 nohz= [KNL] Boottime enable/disable dynamic ticks
3490 Valid arguments: on, off
3493 nohz_full= [KNL,BOOT,SMP,ISOL]
3494 The argument is a cpu list, as described above.
3495 In kernels built with CONFIG_NO_HZ_FULL=y, set
3496 the specified list of CPUs whose tick will be stopped
3497 whenever possible. The boot CPU will be forced outside
3498 the range to maintain the timekeeping. Any CPUs
3499 in this list will have their RCU callbacks offloaded,
3500 just as if they had also been called out in the
3501 rcu_nocbs= boot parameter.
3503 noiotrap [SH] Disables trapped I/O port accesses.
3505 noirqdebug [X86-32] Disables the code which attempts to detect and
3506 disable unhandled interrupt sources.
3508 no_timer_check [X86,APIC] Disables the code which tests for
3509 broken timer IRQ sources.
3511 noisapnp [ISAPNP] Disables ISA PnP code.
3513 noinitrd [RAM] Tells the kernel not to load any configured
3516 nointremap [X86-64, Intel-IOMMU] Do not enable interrupt
3518 [Deprecated - use intremap=off]
3522 noinvpcid [X86] Disable the INVPCID cpu feature.
3524 nojitter [IA-64] Disables jitter checking for ITC timers.
3526 no-kvmclock [X86,KVM] Disable paravirtualized KVM clock driver
3528 no-kvmapf [X86,KVM] Disable paravirtualized asynchronous page
3532 [X86,PV_OPS] Disable paravirtualized VMware scheduler
3533 clock and use the default one.
3535 no-steal-acc [X86,PV_OPS,ARM64] Disable paravirtualized steal time
3536 accounting. steal time is computed, but won't
3537 influence scheduler behaviour
3539 nolapic [X86-32,APIC] Do not enable or use the local APIC.
3541 nolapic_timer [X86-32,APIC] Do not use the local APIC timer.
3543 noltlbs [PPC] Do not use large page/tlb entries for kernel
3544 lowmem mapping on PPC40x and PPC8xx
3546 nomca [IA-64] Disable machine check abort handling
3548 nomce [X86-32] Disable Machine Check Exception
3550 nomfgpt [X86-32] Disable Multi-Function General Purpose
3551 Timer usage (for AMD Geode machines).
3553 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to
3554 shutdown the other cpus. Instead use the REBOOT_VECTOR
3557 nomodule Disable module load
3559 nopat [X86] Disable PAT (page attribute table extension of
3560 pagetables) support.
3562 nopcid [X86-64] Disable the PCID cpu feature.
3564 norandmaps Don't use address space randomization. Equivalent to
3565 echo 0 > /proc/sys/kernel/randomize_va_space
3567 noreplace-smp [X86-32,SMP] Don't replace SMP instructions
3568 with UP alternatives
3570 nordrand [X86] Disable kernel use of the RDRAND and
3571 RDSEED instructions even if they are supported
3572 by the processor. RDRAND and RDSEED are still
3573 available to user space applications.
3575 noresume [SWSUSP] Disables resume and restores original swap
3578 no-scroll [VGA] Disables scrollback.
3579 This is required for the Braillex ib80-piezo Braille
3580 reader made by F.H. Papenmeier (Germany).
3584 nosep [BUGS=X86-32] Disables x86 SYSENTER/SYSEXIT support.
3586 nosgx [X86-64,SGX] Disables Intel SGX kernel support.
3588 nosmp [SMP] Tells an SMP kernel to act as a UP kernel,
3589 and disable the IO APIC. legacy for "maxcpus=0".
3591 nosoftlockup [KNL] Disable the soft-lockup detector.
3593 nosync [HW,M68K] Disables sync negotiation for all devices.
3595 nowatchdog [KNL] Disable both lockup detectors, i.e.
3596 soft-lockup and NMI watchdog (hard-lockup).
3600 nox2apic [X86-64,APIC] Do not enable x2APIC mode.
3602 cpu0_hotplug [X86] Turn on CPU0 hotplug feature when
3603 CONFIG_BOOTPARAM_HOTPLUG_CPU0 is off.
3604 Some features depend on CPU0. Known dependencies are:
3605 1. Resume from suspend/hibernate depends on CPU0.
3606 Suspend/hibernate will fail if CPU0 is offline and you
3607 need to online CPU0 before suspend/hibernate.
3608 2. PIC interrupts also depend on CPU0. CPU0 can't be
3609 removed if a PIC interrupt is detected.
3610 It's said poweroff/reboot may depend on CPU0 on some
3611 machines although I haven't seen such issues so far
3612 after CPU0 is offline on a few tested machines.
3613 If the dependencies are under your control, you can
3614 turn on cpu0_hotplug.
3616 nps_mtm_hs_ctr= [KNL,ARC]
3617 This parameter sets the maximum duration, in
3618 cycles, each HW thread of the CTOP can run
3619 without interruptions, before HW switches it.
3620 The actual maximum duration is 16 times this
3622 Format: integer between 1 and 255
3625 nptcg= [IA-64] Override max number of concurrent global TLB
3626 purges which is reported from either PAL_VM_SUMMARY or
3629 nr_cpus= [SMP] Maximum number of processors that an SMP kernel
3630 could support. nr_cpus=n : n >= 1 limits the kernel to
3631 support 'n' processors. It could be larger than the
3632 number of already plugged CPU during bootup, later in
3633 runtime you can physically add extra cpu until it reaches
3634 n. So during boot up some boot time memory for per-cpu
3635 variables need be pre-allocated for later physical cpu
3638 nr_uarts= [SERIAL] maximum number of UARTs to be registered.
3640 numa=off [KNL, ARM64, PPC, RISCV, SPARC, X86] Disable NUMA, Only
3641 set up a single NUMA node spanning all memory.
3643 numa_balancing= [KNL,ARM64,PPC,RISCV,S390,X86] Enable or disable automatic
3645 Allowed values are enable and disable
3647 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA.
3648 'node', 'default' can be specified
3649 This can be set from sysctl after boot.
3650 See Documentation/admin-guide/sysctl/vm.rst for details.
3652 ohci1394_dma=early [HW] enable debugging via the ohci1394 driver.
3653 See Documentation/core-api/debugging-via-ohci1394.rst for more
3656 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands
3657 Rather than timing out after 20 ms if an EC
3658 command is not properly ACKed, override the length
3659 of the timeout. We have interrupts disabled while
3660 waiting for the ACK, so if this is set too high
3661 interrupts *may* be lost!
3663 omap_mux= [OMAP] Override bootloader pin multiplexing.
3664 Format: <mux_mode0.mode_name=value>...
3665 For example, to override I2C bus2:
3666 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100
3668 oops=panic Always panic on oopses. Default is to just kill the
3669 process, but there is a small probability of
3670 deadlocking the machine.
3671 This will also cause panics on machine check exceptions.
3672 Useful together with panic=30 to trigger a reboot.
3675 [KNL] Boolean flag to control whether the page allocator
3676 should randomize its free lists. The randomization may
3677 be automatically enabled if the kernel detects it is
3678 running on a platform with a direct-mapped memory-side
3679 cache, and this parameter can be used to
3680 override/disable that behavior. The state of the flag
3681 can be read from sysfs at:
3682 /sys/module/page_alloc/parameters/shuffle.
3684 page_owner= [KNL] Boot-time page_owner enabling option.
3685 Storage of the information about who allocated
3686 each page is disabled in default. With this switch,
3688 on: enable the feature
3690 page_poison= [KNL] Boot-time parameter changing the state of
3691 poisoning on the buddy allocator, available with
3692 CONFIG_PAGE_POISONING=y.
3693 off: turn off poisoning (default)
3694 on: turn on poisoning
3696 page_reporting.page_reporting_order=
3697 [KNL] Minimal page reporting order
3699 Adjust the minimal page reporting order. The page
3700 reporting is disabled when it exceeds (MAX_ORDER-1).
3702 panic= [KNL] Kernel behaviour on panic: delay <timeout>
3703 timeout > 0: seconds before rebooting
3704 timeout = 0: wait forever
3705 timeout < 0: reboot immediately
3708 panic_print= Bitmask for printing system info when panic happens.
3709 User can chose combination of the following bits:
3710 bit 0: print all tasks info
3711 bit 1: print system memory info
3712 bit 2: print timer info
3713 bit 3: print locks info if CONFIG_LOCKDEP is on
3714 bit 4: print ftrace buffer
3715 bit 5: print all printk messages in buffer
3717 panic_on_taint= Bitmask for conditionally calling panic() in add_taint()
3718 Format: <hex>[,nousertaint]
3719 Hexadecimal bitmask representing the set of TAINT flags
3720 that will cause the kernel to panic when add_taint() is
3721 called with any of the flags in this set.
3722 The optional switch "nousertaint" can be utilized to
3723 prevent userspace forced crashes by writing to sysctl
3724 /proc/sys/kernel/tainted any flagset matching with the
3725 bitmask set on panic_on_taint.
3726 See Documentation/admin-guide/tainted-kernels.rst for
3727 extra details on the taint flags that users can pick
3728 to compose the bitmask to assign to panic_on_taint.
3730 panic_on_warn panic() instead of WARN(). Useful to cause kdump
3733 crash_kexec_post_notifiers
3734 Run kdump after running panic-notifiers and dumping
3735 kmsg. This only for the users who doubt kdump always
3736 succeeds in any situation.
3737 Note that this also increases risks of kdump failure,
3738 because some panic notifiers can make the crashed
3739 kernel more unstable.
3741 parkbd.port= [HW] Parallel port number the keyboard adapter is
3742 connected to, default is 0.
3744 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation,
3745 0 for XT, 1 for AT (default is AT).
3748 parport= [HW,PPT] Specify parallel ports. 0 disables.
3749 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] }
3750 Use 'auto' to force the driver to use any
3751 IRQ/DMA settings detected (the default is to
3752 ignore detected IRQ/DMA settings because of
3753 possible conflicts). You can specify the base
3754 address, IRQ, and DMA settings; IRQ and DMA
3755 should be numbers, or 'auto' (for using detected
3756 settings on that particular port), or 'nofifo'
3757 (to avoid using a FIFO even if it is detected).
3758 Parallel ports are assigned in the order they
3759 are specified on the command line, starting
3762 parport_init_mode= [HW,PPT]
3763 Configure VIA parallel port to operate in
3764 a specific mode. This is necessary on Pegasos
3765 computer where firmware has no options for setting
3766 up parallel port mode and sets it to spp.
3767 Currently this function knows 686a and 8231 chips.
3768 Format: [spp|ps2|epp|ecp|ecpepp]
3770 pata_legacy.all= [HW,LIBATA]
3772 Set to non-zero to probe primary and secondary ISA
3773 port ranges on PCI systems where no PCI PATA device
3774 has been found at either range. Disabled by default.
3776 pata_legacy.autospeed= [HW,LIBATA]
3778 Set to non-zero if a chip is present that snoops speed
3779 changes. Disabled by default.
3781 pata_legacy.ht6560a= [HW,LIBATA]
3783 Set to 1, 2, or 3 for HT 6560A on the primary channel,
3784 the secondary channel, or both channels respectively.
3785 Disabled by default.
3787 pata_legacy.ht6560b= [HW,LIBATA]
3789 Set to 1, 2, or 3 for HT 6560B on the primary channel,
3790 the secondary channel, or both channels respectively.
3791 Disabled by default.
3793 pata_legacy.iordy_mask= [HW,LIBATA]
3795 IORDY enable mask. Set individual bits to allow IORDY
3796 for the respective channel. Bit 0 is for the first
3797 legacy channel handled by this driver, bit 1 is for
3798 the second channel, and so on. The sequence will often
3799 correspond to the primary legacy channel, the secondary
3800 legacy channel, and so on, but the handling of a PCI
3801 bus and the use of other driver options may interfere
3802 with the sequence. By default IORDY is allowed across
3805 pata_legacy.opti82c46x= [HW,LIBATA]
3807 Set to 1, 2, or 3 for Opti 82c611A on the primary
3808 channel, the secondary channel, or both channels
3809 respectively. Disabled by default.
3811 pata_legacy.opti82c611a= [HW,LIBATA]
3813 Set to 1, 2, or 3 for Opti 82c465MV on the primary
3814 channel, the secondary channel, or both channels
3815 respectively. Disabled by default.
3817 pata_legacy.pio_mask= [HW,LIBATA]
3819 PIO mode mask for autospeed devices. Set individual
3820 bits to allow the use of the respective PIO modes.
3821 Bit 0 is for mode 0, bit 1 is for mode 1, and so on.
3822 All modes allowed by default.
3824 pata_legacy.probe_all= [HW,LIBATA]
3826 Set to non-zero to probe tertiary and further ISA
3827 port ranges on PCI systems. Disabled by default.
3829 pata_legacy.probe_mask= [HW,LIBATA]
3831 Probe mask for legacy ISA PATA ports. Depending on
3832 platform configuration and the use of other driver
3833 options up to 6 legacy ports are supported: 0x1f0,
3834 0x170, 0x1e8, 0x168, 0x1e0, 0x160, however probing
3835 of individual ports can be disabled by setting the
3836 corresponding bits in the mask to 1. Bit 0 is for
3837 the first port in the list above (0x1f0), and so on.
3838 By default all supported ports are probed.
3840 pata_legacy.qdi= [HW,LIBATA]
3842 Set to non-zero to probe QDI controllers. By default
3843 set to 1 if CONFIG_PATA_QDI_MODULE, 0 otherwise.
3845 pata_legacy.winbond= [HW,LIBATA]
3847 Set to non-zero to probe Winbond controllers. Use
3848 the standard I/O port (0x130) if 1, otherwise the
3849 value given is the I/O port to use (typically 0x1b0).
3850 By default set to 1 if CONFIG_PATA_WINBOND_VLB_MODULE,
3853 pata_platform.pio_mask= [HW,LIBATA]
3855 Supported PIO mode mask. Set individual bits to allow
3856 the use of the respective PIO modes. Bit 0 is for
3857 mode 0, bit 1 is for mode 1, and so on. Mode 0 only
3861 Halt all CPUs after the first oops has been printed for
3862 the specified number of seconds. This is to be used if
3863 your oopses keep scrolling off the screen.
3868 See header of drivers/block/paride/pcd.c.
3869 See also Documentation/admin-guide/blockdev/paride.rst.
3871 pci=option[,option...] [PCI] various PCI subsystem options.
3873 Some options herein operate on a specific device
3874 or a set of devices (<pci_dev>). These are
3875 specified in one of the following formats:
3877 [<domain>:]<bus>:<dev>.<func>[/<dev>.<func>]*
3878 pci:<vendor>:<device>[:<subvendor>:<subdevice>]
3880 Note: the first format specifies a PCI
3881 bus/device/function address which may change
3882 if new hardware is inserted, if motherboard
3883 firmware changes, or due to changes caused
3884 by other kernel parameters. If the
3885 domain is left unspecified, it is
3886 taken to be zero. Optionally, a path
3887 to a device through multiple device/function
3888 addresses can be specified after the base
3889 address (this is more robust against
3890 renumbering issues). The second format
3891 selects devices using IDs from the
3892 configuration space which may match multiple
3893 devices in the system.
3895 earlydump dump PCI config space before the kernel
3897 off [X86] don't probe for the PCI bus
3898 bios [X86-32] force use of PCI BIOS, don't access
3899 the hardware directly. Use this if your machine
3900 has a non-standard PCI host bridge.
3901 nobios [X86-32] disallow use of PCI BIOS, only direct
3902 hardware access methods are allowed. Use this
3903 if you experience crashes upon bootup and you
3904 suspect they are caused by the BIOS.
3905 conf1 [X86] Force use of PCI Configuration Access
3906 Mechanism 1 (config address in IO port 0xCF8,
3907 data in IO port 0xCFC, both 32-bit).
3908 conf2 [X86] Force use of PCI Configuration Access
3909 Mechanism 2 (IO port 0xCF8 is an 8-bit port for
3910 the function, IO port 0xCFA, also 8-bit, sets
3911 bus number. The config space is then accessed
3912 through ports 0xC000-0xCFFF).
3913 See http://wiki.osdev.org/PCI for more info
3914 on the configuration access mechanisms.
3915 noaer [PCIE] If the PCIEAER kernel config parameter is
3916 enabled, this kernel boot option can be used to
3917 disable the use of PCIE advanced error reporting.
3918 nodomains [PCI] Disable support for multiple PCI
3919 root domains (aka PCI segments, in ACPI-speak).
3920 nommconf [X86] Disable use of MMCONFIG for PCI
3922 check_enable_amd_mmconf [X86] check for and enable
3923 properly configured MMIO access to PCI
3924 config space on AMD family 10h CPU
3925 nomsi [MSI] If the PCI_MSI kernel config parameter is
3926 enabled, this kernel boot option can be used to
3927 disable the use of MSI interrupts system-wide.
3928 noioapicquirk [APIC] Disable all boot interrupt quirks.
3929 Safety option to keep boot IRQs enabled. This
3930 should never be necessary.
3931 ioapicreroute [APIC] Enable rerouting of boot IRQs to the
3932 primary IO-APIC for bridges that cannot disable
3933 boot IRQs. This fixes a source of spurious IRQs
3934 when the system masks IRQs.
3935 noioapicreroute [APIC] Disable workaround that uses the
3936 boot IRQ equivalent of an IRQ that connects to
3937 a chipset where boot IRQs cannot be disabled.
3938 The opposite of ioapicreroute.
3939 biosirq [X86-32] Use PCI BIOS calls to get the interrupt
3940 routing table. These calls are known to be buggy
3941 on several machines and they hang the machine
3942 when used, but on other computers it's the only
3943 way to get the interrupt routing table. Try
3944 this option if the kernel is unable to allocate
3945 IRQs or discover secondary PCI buses on your
3947 rom [X86] Assign address space to expansion ROMs.
3948 Use with caution as certain devices share
3949 address decoders between ROMs and other
3951 norom [X86] Do not assign address space to
3952 expansion ROMs that do not already have
3953 BIOS assigned address ranges.
3954 nobar [X86] Do not assign address space to the
3955 BARs that weren't assigned by the BIOS.
3956 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be
3957 assigned automatically to PCI devices. You can
3958 make the kernel exclude IRQs of your ISA cards
3960 pirqaddr=0xAAAAA [X86] Specify the physical address
3961 of the PIRQ table (normally generated
3962 by the BIOS) if it is outside the
3963 F0000h-100000h range.
3964 lastbus=N [X86] Scan all buses thru bus #N. Can be
3965 useful if the kernel is unable to find your
3966 secondary buses and you want to tell it
3967 explicitly which ones they are.
3968 assign-busses [X86] Always assign all PCI bus
3969 numbers ourselves, overriding
3970 whatever the firmware may have done.
3971 usepirqmask [X86] Honor the possible IRQ mask stored
3972 in the BIOS $PIR table. This is needed on
3973 some systems with broken BIOSes, notably
3974 some HP Pavilion N5400 and Omnibook XE3
3975 notebooks. This will have no effect if ACPI
3976 IRQ routing is enabled.
3977 noacpi [X86] Do not use ACPI for IRQ routing
3978 or for PCI scanning.
3979 use_crs [X86] Use PCI host bridge window information
3980 from ACPI. On BIOSes from 2008 or later, this
3981 is enabled by default. If you need to use this,
3982 please report a bug.
3983 nocrs [X86] Ignore PCI host bridge windows from ACPI.
3984 If you need to use this, please report a bug.
3985 routeirq Do IRQ routing for all PCI devices.
3986 This is normally done in pci_enable_device(),
3987 so this option is a temporary workaround
3988 for broken drivers that don't call it.
3989 skip_isa_align [X86] do not align io start addr, so can
3990 handle more pci cards
3991 noearly [X86] Don't do any early type 1 scanning.
3992 This might help on some broken boards which
3993 machine check when some devices' config space
3994 is read. But various workarounds are disabled
3995 and some IOMMU drivers will not work.
3996 bfsort Sort PCI devices into breadth-first order.
3997 This sorting is done to get a device
3998 order compatible with older (<= 2.4) kernels.
3999 nobfsort Don't sort PCI devices into breadth-first order.
4000 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size)
4001 tuning and use the BIOS-configured MPS defaults.
4002 pcie_bus_safe Set every device's MPS to the largest value
4003 supported by all devices below the root complex.
4004 pcie_bus_perf Set device MPS to the largest allowable MPS
4005 based on its parent bus. Also set MRRS (Max
4006 Read Request Size) to the largest supported
4007 value (no larger than the MPS that the device
4008 or bus can support) for best performance.
4009 pcie_bus_peer2peer Set every device's MPS to 128B, which
4010 every device is guaranteed to support. This
4011 configuration allows peer-to-peer DMA between
4012 any pair of devices, possibly at the cost of
4013 reduced performance. This also guarantees
4014 that hot-added devices will work.
4015 cbiosize=nn[KMG] The fixed amount of bus space which is
4016 reserved for the CardBus bridge's IO window.
4017 The default value is 256 bytes.
4018 cbmemsize=nn[KMG] The fixed amount of bus space which is
4019 reserved for the CardBus bridge's memory
4020 window. The default value is 64 megabytes.
4023 [<order of align>@]<pci_dev>[; ...]
4024 Specifies alignment and device to reassign
4025 aligned memory resources. How to
4026 specify the device is described above.
4027 If <order of align> is not specified,
4028 PAGE_SIZE is used as alignment.
4029 A PCI-PCI bridge can be specified if resource
4030 windows need to be expanded.
4031 To specify the alignment for several
4032 instances of a device, the PCI vendor,
4033 device, subvendor, and subdevice may be
4034 specified, e.g., 12@pci:8086:9c22:103c:198f
4035 for 4096-byte alignment.
4036 ecrc= Enable/disable PCIe ECRC (transaction layer
4037 end-to-end CRC checking).
4038 bios: Use BIOS/firmware settings. This is the
4042 hpiosize=nn[KMG] The fixed amount of bus space which is
4043 reserved for hotplug bridge's IO window.
4044 Default size is 256 bytes.
4045 hpmmiosize=nn[KMG] The fixed amount of bus space which is
4046 reserved for hotplug bridge's MMIO window.
4047 Default size is 2 megabytes.
4048 hpmmioprefsize=nn[KMG] The fixed amount of bus space which is
4049 reserved for hotplug bridge's MMIO_PREF window.
4050 Default size is 2 megabytes.
4051 hpmemsize=nn[KMG] The fixed amount of bus space which is
4052 reserved for hotplug bridge's MMIO and
4054 Default size is 2 megabytes.
4055 hpbussize=nn The minimum amount of additional bus numbers
4056 reserved for buses below a hotplug bridge.
4058 realloc= Enable/disable reallocating PCI bridge resources
4059 if allocations done by BIOS are too small to
4060 accommodate resources required by all child
4062 off: Turn realloc off
4064 realloc same as realloc=on
4065 noari do not use PCIe ARI.
4066 noats [PCIE, Intel-IOMMU, AMD-IOMMU]
4067 do not use PCIe ATS (and IOMMU device IOTLB).
4068 pcie_scan_all Scan all possible PCIe devices. Otherwise we
4069 only look for one device below a PCIe downstream
4071 big_root_window Try to add a big 64bit memory window to the PCIe
4072 root complex on AMD CPUs. Some GFX hardware
4073 can resize a BAR to allow access to all VRAM.
4074 Adding the window is slightly risky (it may
4075 conflict with unreported devices), so this
4077 disable_acs_redir=<pci_dev>[; ...]
4078 Specify one or more PCI devices (in the format
4079 specified above) separated by semicolons.
4080 Each device specified will have the PCI ACS
4081 redirect capabilities forced off which will
4082 allow P2P traffic between devices through
4083 bridges without forcing it upstream. Note:
4084 this removes isolation between devices and
4085 may put more devices in an IOMMU group.
4086 force_floating [S390] Force usage of floating interrupts.
4087 nomio [S390] Do not use MIO instructions.
4088 norid [S390] ignore the RID field and force use of
4089 one PCI domain per PCI function
4091 pcie_aspm= [PCIE] Forcibly enable or disable PCIe Active State Power
4094 force Enable ASPM even on devices that claim not to support it.
4095 WARNING: Forcing ASPM on may cause system lockups.
4097 pcie_ports= [PCIE] PCIe port services handling:
4098 native Use native PCIe services (PME, AER, DPC, PCIe hotplug)
4099 even if the platform doesn't give the OS permission to
4100 use them. This may cause conflicts if the platform
4101 also tries to use these services.
4102 dpc-native Use native PCIe service for DPC only. May
4103 cause conflicts if firmware uses AER or DPC.
4104 compat Disable native PCIe services (PME, AER, DPC, PCIe
4107 pcie_port_pm= [PCIE] PCIe port power management handling:
4108 off Disable power management of all PCIe ports
4109 force Forcibly enable power management of all PCIe ports
4111 pcie_pme= [PCIE,PM] Native PCIe PME signaling options:
4112 nomsi Do not use MSI for native PCIe PME signaling (this makes
4113 all PCIe root ports use INTx for all services).
4115 pcmv= [HW,PCMCIA] BadgePAD 4
4119 Keep all power-domains already enabled by bootloader on,
4120 even if no driver has claimed them. This is useful
4121 for debug and development, but should not be
4122 needed on a platform with proper driver support.
4125 See Documentation/admin-guide/blockdev/paride.rst.
4127 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at
4130 See arch/parisc/kernel/pdc_chassis.c
4132 percpu_alloc= Select which percpu first chunk allocator to use.
4133 Currently supported values are "embed" and "page".
4134 Archs may support subset or none of the selections.
4135 See comments in mm/percpu.c for details on each
4136 allocator. This parameter is primarily for debugging
4137 and performance comparison.
4140 See Documentation/admin-guide/blockdev/paride.rst.
4143 See Documentation/admin-guide/blockdev/paride.rst.
4145 pirq= [SMP,APIC] Manual mp-table setup
4146 See Documentation/x86/i386/IO-APIC.rst.
4148 plip= [PPT,NET] Parallel port network link
4149 Format: { parport<nr> | timid | 0 }
4150 See also Documentation/admin-guide/parport.rst.
4152 pmtmr= [X86] Manual setup of pmtmr I/O Port.
4153 Override pmtimer IOPort with a hex value.
4156 pm_debug_messages [SUSPEND,KNL]
4157 Enable suspend/resume debug messages during boot up.
4160 Enable PNP debug messages (depends on the
4161 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time
4162 via /sys/module/pnp/parameters/debug. We always show
4163 current resource usage; turning this on also shows
4164 possible settings and some assignment information.
4170 { on | off | curr | res | no-curr | no-res }
4173 [ISAPNP] Exclude IRQs for the autoconfiguration
4176 [ISAPNP] Exclude DMAs for the autoconfiguration
4178 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration
4179 Ranges are in pairs (I/O port base and size).
4182 [ISAPNP] Exclude memory regions for the
4184 Ranges are in pairs (memory base and size).
4186 ports= [IP_VS_FTP] IPVS ftp helper module
4188 Up to 8 (IP_VS_APP_MAX_PORTS) ports
4190 Format: <port>,<port>....
4192 powersave=off [PPC] This option disables power saving features.
4193 It specifically disables cpuidle and sets the
4194 platform machine description specific power_save
4195 function to NULL. On Idle the CPU just reduces
4198 ppc_strict_facility_enable
4199 [PPC] This option catches any kernel floating point,
4200 Altivec, VSX and SPE outside of regions specifically
4201 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()).
4202 There is some performance impact when enabling this.
4206 Disable Hardware Transactional Memory
4209 Select preemption mode if you have CONFIG_PREEMPT_DYNAMIC
4210 none - Limited to cond_resched() calls
4211 voluntary - Limited to cond_resched() and might_sleep() calls
4212 full - Any section that isn't explicitly preempt disabled
4213 can be preempted anytime.
4215 print-fatal-signals=
4216 [KNL] debug: print fatal signals
4218 If enabled, warn about various signal handling
4219 related application anomalies: too many signals,
4220 too many POSIX.1 timers, fatal signals causing a
4223 If you hit the warning due to signal overflow,
4224 you might want to try "ulimit -i unlimited".
4228 printk.always_kmsg_dump=
4229 Trigger kmsg_dump for cases other than kernel oops or
4231 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
4234 printk.console_no_auto_verbose=
4235 Disable console loglevel raise on oops, panic
4236 or lockdep-detected issues (only if lock debug is on).
4237 With an exception to setups with low baudrate on
4238 serial console, keeping this 0 is a good choice
4239 in order to provide more debug information.
4241 default: 0 (auto_verbose is enabled)
4243 printk.devkmsg={on,off,ratelimit}
4244 Control writing to /dev/kmsg.
4245 on - unlimited logging to /dev/kmsg from userspace
4246 off - logging to /dev/kmsg disabled
4247 ratelimit - ratelimit the logging
4250 printk.time= Show timing data prefixed to each printk message line
4251 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
4253 processor.max_cstate= [HW,ACPI]
4254 Limit processor to maximum C-state
4255 max_cstate=9 overrides any DMI blacklist limit.
4257 processor.nocst [HW,ACPI]
4258 Ignore the _CST method to determine C-states,
4259 instead using the legacy FADT method
4261 profile= [KNL] Enable kernel profiling via /proc/profile
4262 Format: [<profiletype>,]<number>
4263 Param: <profiletype>: "schedule", "sleep", or "kvm"
4264 [defaults to kernel profiling]
4265 Param: "schedule" - profile schedule points.
4266 Param: "sleep" - profile D-state sleeping (millisecs).
4267 Requires CONFIG_SCHEDSTATS
4268 Param: "kvm" - profile VM exits.
4269 Param: <number> - step/bucket size as a power of 2 for
4270 statistical time based profiling.
4272 prompt_ramdisk= [RAM] [Deprecated]
4274 prot_virt= [S390] enable hosting protected virtual machines
4275 isolated from the hypervisor (if hardware supports
4279 psi= [KNL] Enable or disable pressure stall information
4283 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to
4284 probe for; one of (bare|imps|exps|lifebook|any).
4285 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports
4287 psmouse.resetafter= [HW,MOUSE]
4288 Try to reset the device after so many bad packets
4291 [HW,MOUSE] Set desired mouse resolution, in dpi.
4292 psmouse.smartscroll=
4293 [HW,MOUSE] Controls Logitech smartscroll autorepeat.
4294 0 = disabled, 1 = enabled (default).
4296 pstore.backend= Specify the name of the pstore backend to use
4299 See Documentation/admin-guide/blockdev/paride.rst.
4301 pti= [X86-64] Control Page Table Isolation of user and
4302 kernel address spaces. Disabling this feature
4303 removes hardening, but improves performance of
4304 system calls and interrupts.
4306 on - unconditionally enable
4307 off - unconditionally disable
4308 auto - kernel detects whether your CPU model is
4309 vulnerable to issues that PTI mitigates
4311 Not specifying this option is equivalent to pti=auto.
4314 Equivalent to pti=off
4317 [KNL] Number of legacy pty's. Overwrites compiled-in
4320 quiet [KNL] Disable most log messages
4325 See Documentation/admin-guide/md.rst.
4327 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes
4328 See Documentation/admin-guide/blockdev/ramdisk.rst.
4330 ramdisk_start= [RAM] RAM disk image start address
4332 random.trust_cpu={on,off}
4333 [KNL] Enable or disable trusting the use of the
4334 CPU's random number generator (if available) to
4335 fully seed the kernel's CRNG. Default is controlled
4336 by CONFIG_RANDOM_TRUST_CPU.
4338 randomize_kstack_offset=
4339 [KNL] Enable or disable kernel stack offset
4340 randomization, which provides roughly 5 bits of
4341 entropy, frustrating memory corruption attacks
4342 that depend on stack address determinism or
4343 cross-syscall address exposures. This is only
4344 available on architectures that have defined
4345 CONFIG_HAVE_ARCH_RANDOMIZE_KSTACK_OFFSET.
4346 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
4347 Default is CONFIG_RANDOMIZE_KSTACK_OFFSET_DEFAULT.
4349 ras=option[,option,...] [KNL] RAS-specific options
4352 Disable the Correctable Errors Collector,
4353 see CONFIG_RAS_CEC help text.
4356 The argument is a cpu list, as described above.
4358 In kernels built with CONFIG_RCU_NOCB_CPU=y, set
4359 the specified list of CPUs to be no-callback CPUs.
4360 Invocation of these CPUs' RCU callbacks will be
4361 offloaded to "rcuox/N" kthreads created for that
4362 purpose, where "x" is "p" for RCU-preempt, and
4363 "s" for RCU-sched, and "N" is the CPU number.
4364 This reduces OS jitter on the offloaded CPUs,
4365 which can be useful for HPC and real-time
4366 workloads. It can also improve energy efficiency
4367 for asymmetric multiprocessors.
4370 Rather than requiring that offloaded CPUs
4371 (specified by rcu_nocbs= above) explicitly
4372 awaken the corresponding "rcuoN" kthreads,
4373 make these kthreads poll for callbacks.
4374 This improves the real-time response for the
4375 offloaded CPUs by relieving them of the need to
4376 wake up the corresponding kthread, but degrades
4377 energy efficiency by requiring that the kthreads
4378 periodically wake up to do the polling.
4380 rcutree.blimit= [KNL]
4381 Set maximum number of finished RCU callbacks to
4382 process in one batch.
4384 rcutree.dump_tree= [KNL]
4385 Dump the structure of the rcu_node combining tree
4386 out at early boot. This is used for diagnostic
4387 purposes, to verify correct tree setup.
4389 rcutree.gp_cleanup_delay= [KNL]
4390 Set the number of jiffies to delay each step of
4391 RCU grace-period cleanup.
4393 rcutree.gp_init_delay= [KNL]
4394 Set the number of jiffies to delay each step of
4395 RCU grace-period initialization.
4397 rcutree.gp_preinit_delay= [KNL]
4398 Set the number of jiffies to delay each step of
4399 RCU grace-period pre-initialization, that is,
4400 the propagation of recent CPU-hotplug changes up
4401 the rcu_node combining tree.
4403 rcutree.use_softirq= [KNL]
4404 If set to zero, move all RCU_SOFTIRQ processing to
4405 per-CPU rcuc kthreads. Defaults to a non-zero
4406 value, meaning that RCU_SOFTIRQ is used by default.
4407 Specify rcutree.use_softirq=0 to use rcuc kthreads.
4409 But note that CONFIG_PREEMPT_RT=y kernels disable
4410 this kernel boot parameter, forcibly setting it
4413 rcutree.rcu_fanout_exact= [KNL]
4414 Disable autobalancing of the rcu_node combining
4415 tree. This is used by rcutorture, and might
4416 possibly be useful for architectures having high
4417 cache-to-cache transfer latencies.
4419 rcutree.rcu_fanout_leaf= [KNL]
4420 Change the number of CPUs assigned to each
4421 leaf rcu_node structure. Useful for very
4422 large systems, which will choose the value 64,
4423 and for NUMA systems with large remote-access
4424 latencies, which will choose a value aligned
4425 with the appropriate hardware boundaries.
4427 rcutree.rcu_min_cached_objs= [KNL]
4428 Minimum number of objects which are cached and
4429 maintained per one CPU. Object size is equal
4430 to PAGE_SIZE. The cache allows to reduce the
4431 pressure to page allocator, also it makes the
4432 whole algorithm to behave better in low memory
4435 rcutree.rcu_delay_page_cache_fill_msec= [KNL]
4436 Set the page-cache refill delay (in milliseconds)
4437 in response to low-memory conditions. The range
4438 of permitted values is in the range 0:100000.
4440 rcutree.jiffies_till_first_fqs= [KNL]
4441 Set delay from grace-period initialization to
4442 first attempt to force quiescent states.
4443 Units are jiffies, minimum value is zero,
4444 and maximum value is HZ.
4446 rcutree.jiffies_till_next_fqs= [KNL]
4447 Set delay between subsequent attempts to force
4448 quiescent states. Units are jiffies, minimum
4449 value is one, and maximum value is HZ.
4451 rcutree.jiffies_till_sched_qs= [KNL]
4452 Set required age in jiffies for a
4453 given grace period before RCU starts
4454 soliciting quiescent-state help from
4455 rcu_note_context_switch() and cond_resched().
4456 If not specified, the kernel will calculate
4457 a value based on the most recent settings
4458 of rcutree.jiffies_till_first_fqs
4459 and rcutree.jiffies_till_next_fqs.
4460 This calculated value may be viewed in
4461 rcutree.jiffies_to_sched_qs. Any attempt to set
4462 rcutree.jiffies_to_sched_qs will be cheerfully
4465 rcutree.kthread_prio= [KNL,BOOT]
4466 Set the SCHED_FIFO priority of the RCU per-CPU
4467 kthreads (rcuc/N). This value is also used for
4468 the priority of the RCU boost threads (rcub/N)
4469 and for the RCU grace-period kthreads (rcu_bh,
4470 rcu_preempt, and rcu_sched). If RCU_BOOST is
4471 set, valid values are 1-99 and the default is 1
4472 (the least-favored priority). Otherwise, when
4473 RCU_BOOST is not set, valid values are 0-99 and
4474 the default is zero (non-realtime operation).
4476 rcutree.rcu_nocb_gp_stride= [KNL]
4477 Set the number of NOCB callback kthreads in
4478 each group, which defaults to the square root
4479 of the number of CPUs. Larger numbers reduce
4480 the wakeup overhead on the global grace-period
4481 kthread, but increases that same overhead on
4482 each group's NOCB grace-period kthread.
4484 rcutree.qhimark= [KNL]
4485 Set threshold of queued RCU callbacks beyond which
4486 batch limiting is disabled.
4488 rcutree.qlowmark= [KNL]
4489 Set threshold of queued RCU callbacks below which
4490 batch limiting is re-enabled.
4492 rcutree.qovld= [KNL]
4493 Set threshold of queued RCU callbacks beyond which
4494 RCU's force-quiescent-state scan will aggressively
4495 enlist help from cond_resched() and sched IPIs to
4496 help CPUs more quickly reach quiescent states.
4497 Set to less than zero to make this be set based
4498 on rcutree.qhimark at boot time and to zero to
4499 disable more aggressive help enlistment.
4501 rcutree.rcu_idle_gp_delay= [KNL]
4502 Set wakeup interval for idle CPUs that have
4503 RCU callbacks (RCU_FAST_NO_HZ=y).
4505 rcutree.rcu_kick_kthreads= [KNL]
4506 Cause the grace-period kthread to get an extra
4507 wake_up() if it sleeps three times longer than
4508 it should at force-quiescent-state time.
4509 This wake_up() will be accompanied by a
4510 WARN_ONCE() splat and an ftrace_dump().
4512 rcutree.rcu_unlock_delay= [KNL]
4513 In CONFIG_RCU_STRICT_GRACE_PERIOD=y kernels,
4514 this specifies an rcu_read_unlock()-time delay
4515 in microseconds. This defaults to zero.
4516 Larger delays increase the probability of
4517 catching RCU pointer leaks, that is, buggy use
4518 of RCU-protected pointers after the relevant
4519 rcu_read_unlock() has completed.
4521 rcutree.sysrq_rcu= [KNL]
4522 Commandeer a sysrq key to dump out Tree RCU's
4523 rcu_node tree with an eye towards determining
4524 why a new grace period has not yet started.
4526 rcuscale.gp_async= [KNL]
4527 Measure performance of asynchronous
4528 grace-period primitives such as call_rcu().
4530 rcuscale.gp_async_max= [KNL]
4531 Specify the maximum number of outstanding
4532 callbacks per writer thread. When a writer
4533 thread exceeds this limit, it invokes the
4534 corresponding flavor of rcu_barrier() to allow
4535 previously posted callbacks to drain.
4537 rcuscale.gp_exp= [KNL]
4538 Measure performance of expedited synchronous
4539 grace-period primitives.
4541 rcuscale.holdoff= [KNL]
4542 Set test-start holdoff period. The purpose of
4543 this parameter is to delay the start of the
4544 test until boot completes in order to avoid
4547 rcuscale.kfree_rcu_test= [KNL]
4548 Set to measure performance of kfree_rcu() flooding.
4550 rcuscale.kfree_rcu_test_double= [KNL]
4551 Test the double-argument variant of kfree_rcu().
4552 If this parameter has the same value as
4553 rcuscale.kfree_rcu_test_single, both the single-
4554 and double-argument variants are tested.
4556 rcuscale.kfree_rcu_test_single= [KNL]
4557 Test the single-argument variant of kfree_rcu().
4558 If this parameter has the same value as
4559 rcuscale.kfree_rcu_test_double, both the single-
4560 and double-argument variants are tested.
4562 rcuscale.kfree_nthreads= [KNL]
4563 The number of threads running loops of kfree_rcu().
4565 rcuscale.kfree_alloc_num= [KNL]
4566 Number of allocations and frees done in an iteration.
4568 rcuscale.kfree_loops= [KNL]
4569 Number of loops doing rcuscale.kfree_alloc_num number
4570 of allocations and frees.
4572 rcuscale.nreaders= [KNL]
4573 Set number of RCU readers. The value -1 selects
4574 N, where N is the number of CPUs. A value
4575 "n" less than -1 selects N-n+1, where N is again
4576 the number of CPUs. For example, -2 selects N
4577 (the number of CPUs), -3 selects N+1, and so on.
4578 A value of "n" less than or equal to -N selects
4581 rcuscale.nwriters= [KNL]
4582 Set number of RCU writers. The values operate
4583 the same as for rcuscale.nreaders.
4584 N, where N is the number of CPUs
4586 rcuscale.perf_type= [KNL]
4587 Specify the RCU implementation to test.
4589 rcuscale.shutdown= [KNL]
4590 Shut the system down after performance tests
4591 complete. This is useful for hands-off automated
4594 rcuscale.verbose= [KNL]
4595 Enable additional printk() statements.
4597 rcuscale.writer_holdoff= [KNL]
4598 Write-side holdoff between grace periods,
4599 in microseconds. The default of zero says
4602 rcutorture.fqs_duration= [KNL]
4603 Set duration of force_quiescent_state bursts
4606 rcutorture.fqs_holdoff= [KNL]
4607 Set holdoff time within force_quiescent_state bursts
4610 rcutorture.fqs_stutter= [KNL]
4611 Set wait time between force_quiescent_state bursts
4614 rcutorture.fwd_progress= [KNL]
4615 Enable RCU grace-period forward-progress testing
4616 for the types of RCU supporting this notion.
4618 rcutorture.fwd_progress_div= [KNL]
4619 Specify the fraction of a CPU-stall-warning
4620 period to do tight-loop forward-progress testing.
4622 rcutorture.fwd_progress_holdoff= [KNL]
4623 Number of seconds to wait between successive
4624 forward-progress tests.
4626 rcutorture.fwd_progress_need_resched= [KNL]
4627 Enclose cond_resched() calls within checks for
4628 need_resched() during tight-loop forward-progress
4631 rcutorture.gp_cond= [KNL]
4632 Use conditional/asynchronous update-side
4633 primitives, if available.
4635 rcutorture.gp_exp= [KNL]
4636 Use expedited update-side primitives, if available.
4638 rcutorture.gp_normal= [KNL]
4639 Use normal (non-expedited) asynchronous
4640 update-side primitives, if available.
4642 rcutorture.gp_sync= [KNL]
4643 Use normal (non-expedited) synchronous
4644 update-side primitives, if available. If all
4645 of rcutorture.gp_cond=, rcutorture.gp_exp=,
4646 rcutorture.gp_normal=, and rcutorture.gp_sync=
4647 are zero, rcutorture acts as if is interpreted
4648 they are all non-zero.
4650 rcutorture.irqreader= [KNL]
4651 Run RCU readers from irq handlers, or, more
4652 accurately, from a timer handler. Not all RCU
4653 flavors take kindly to this sort of thing.
4655 rcutorture.leakpointer= [KNL]
4656 Leak an RCU-protected pointer out of the reader.
4657 This can of course result in splats, and is
4658 intended to test the ability of things like
4659 CONFIG_RCU_STRICT_GRACE_PERIOD=y to detect
4662 rcutorture.n_barrier_cbs= [KNL]
4663 Set callbacks/threads for rcu_barrier() testing.
4665 rcutorture.nfakewriters= [KNL]
4666 Set number of concurrent RCU writers. These just
4667 stress RCU, they don't participate in the actual
4668 test, hence the "fake".
4670 rcutorture.nocbs_nthreads= [KNL]
4671 Set number of RCU callback-offload togglers.
4672 Zero (the default) disables toggling.
4674 rcutorture.nocbs_toggle= [KNL]
4675 Set the delay in milliseconds between successive
4676 callback-offload toggling attempts.
4678 rcutorture.nreaders= [KNL]
4679 Set number of RCU readers. The value -1 selects
4680 N-1, where N is the number of CPUs. A value
4681 "n" less than -1 selects N-n-2, where N is again
4682 the number of CPUs. For example, -2 selects N
4683 (the number of CPUs), -3 selects N+1, and so on.
4685 rcutorture.object_debug= [KNL]
4686 Enable debug-object double-call_rcu() testing.
4688 rcutorture.onoff_holdoff= [KNL]
4689 Set time (s) after boot for CPU-hotplug testing.
4691 rcutorture.onoff_interval= [KNL]
4692 Set time (jiffies) between CPU-hotplug operations,
4693 or zero to disable CPU-hotplug testing.
4695 rcutorture.read_exit= [KNL]
4696 Set the number of read-then-exit kthreads used
4697 to test the interaction of RCU updaters and
4698 task-exit processing.
4700 rcutorture.read_exit_burst= [KNL]
4701 The number of times in a given read-then-exit
4702 episode that a set of read-then-exit kthreads
4705 rcutorture.read_exit_delay= [KNL]
4706 The delay, in seconds, between successive
4707 read-then-exit testing episodes.
4709 rcutorture.shuffle_interval= [KNL]
4710 Set task-shuffle interval (s). Shuffling tasks
4711 allows some CPUs to go into dyntick-idle mode
4712 during the rcutorture test.
4714 rcutorture.shutdown_secs= [KNL]
4715 Set time (s) after boot system shutdown. This
4716 is useful for hands-off automated testing.
4718 rcutorture.stall_cpu= [KNL]
4719 Duration of CPU stall (s) to test RCU CPU stall
4720 warnings, zero to disable.
4722 rcutorture.stall_cpu_block= [KNL]
4723 Sleep while stalling if set. This will result
4724 in warnings from preemptible RCU in addition
4725 to any other stall-related activity.
4727 rcutorture.stall_cpu_holdoff= [KNL]
4728 Time to wait (s) after boot before inducing stall.
4730 rcutorture.stall_cpu_irqsoff= [KNL]
4731 Disable interrupts while stalling if set.
4733 rcutorture.stall_gp_kthread= [KNL]
4734 Duration (s) of forced sleep within RCU
4735 grace-period kthread to test RCU CPU stall
4736 warnings, zero to disable. If both stall_cpu
4737 and stall_gp_kthread are specified, the
4738 kthread is starved first, then the CPU.
4740 rcutorture.stat_interval= [KNL]
4741 Time (s) between statistics printk()s.
4743 rcutorture.stutter= [KNL]
4744 Time (s) to stutter testing, for example, specifying
4745 five seconds causes the test to run for five seconds,
4746 wait for five seconds, and so on. This tests RCU's
4747 ability to transition abruptly to and from idle.
4749 rcutorture.test_boost= [KNL]
4750 Test RCU priority boosting? 0=no, 1=maybe, 2=yes.
4751 "Maybe" means test if the RCU implementation
4752 under test support RCU priority boosting.
4754 rcutorture.test_boost_duration= [KNL]
4755 Duration (s) of each individual boost test.
4757 rcutorture.test_boost_interval= [KNL]
4758 Interval (s) between each boost test.
4760 rcutorture.test_no_idle_hz= [KNL]
4761 Test RCU's dyntick-idle handling. See also the
4762 rcutorture.shuffle_interval parameter.
4764 rcutorture.torture_type= [KNL]
4765 Specify the RCU implementation to test.
4767 rcutorture.verbose= [KNL]
4768 Enable additional printk() statements.
4770 rcupdate.rcu_cpu_stall_ftrace_dump= [KNL]
4771 Dump ftrace buffer after reporting RCU CPU
4774 rcupdate.rcu_cpu_stall_suppress= [KNL]
4775 Suppress RCU CPU stall warning messages.
4777 rcupdate.rcu_cpu_stall_suppress_at_boot= [KNL]
4778 Suppress RCU CPU stall warning messages and
4779 rcutorture writer stall warnings that occur
4780 during early boot, that is, during the time
4781 before the init task is spawned.
4783 rcupdate.rcu_cpu_stall_timeout= [KNL]
4784 Set timeout for RCU CPU stall warning messages.
4786 rcupdate.rcu_expedited= [KNL]
4787 Use expedited grace-period primitives, for
4788 example, synchronize_rcu_expedited() instead
4789 of synchronize_rcu(). This reduces latency,
4790 but can increase CPU utilization, degrade
4791 real-time latency, and degrade energy efficiency.
4792 No effect on CONFIG_TINY_RCU kernels.
4794 rcupdate.rcu_normal= [KNL]
4795 Use only normal grace-period primitives,
4796 for example, synchronize_rcu() instead of
4797 synchronize_rcu_expedited(). This improves
4798 real-time latency, CPU utilization, and
4799 energy efficiency, but can expose users to
4800 increased grace-period latency. This parameter
4801 overrides rcupdate.rcu_expedited. No effect on
4802 CONFIG_TINY_RCU kernels.
4804 rcupdate.rcu_normal_after_boot= [KNL]
4805 Once boot has completed (that is, after
4806 rcu_end_inkernel_boot() has been invoked), use
4807 only normal grace-period primitives. No effect
4808 on CONFIG_TINY_RCU kernels.
4810 But note that CONFIG_PREEMPT_RT=y kernels enables
4811 this kernel boot parameter, forcibly setting
4812 it to the value one, that is, converting any
4813 post-boot attempt at an expedited RCU grace
4814 period to instead use normal non-expedited
4815 grace-period processing.
4817 rcupdate.rcu_task_ipi_delay= [KNL]
4818 Set time in jiffies during which RCU tasks will
4819 avoid sending IPIs, starting with the beginning
4820 of a given grace period. Setting a large
4821 number avoids disturbing real-time workloads,
4822 but lengthens grace periods.
4824 rcupdate.rcu_task_stall_timeout= [KNL]
4825 Set timeout in jiffies for RCU task stall warning
4826 messages. Disable with a value less than or equal
4829 rcupdate.rcu_self_test= [KNL]
4830 Run the RCU early boot self tests
4834 Run specified binary instead of /init from the ramdisk,
4835 used for early userspace startup. See initrd.
4838 force - Override the decision by the kernel to hide the
4839 advertisement of RDRAND support (this affects
4840 certain AMD processors because of buggy BIOS
4841 support, specifically around the suspend/resume
4845 Turn on/off individual RDT features. List is:
4846 cmt, mbmtotal, mbmlocal, l3cat, l3cdp, l2cat, l2cdp,
4848 E.g. to turn on cmt and turn off mba use:
4852 Format (x86 or x86_64):
4853 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] | d[efault] \
4855 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \
4857 Where reboot_mode is one of warm (soft) or cold (hard) or gpio
4858 (prefix with 'panic_' to set mode for panic
4860 reboot_type is one of bios, acpi, kbd, triple, efi, or pci,
4861 reboot_force is either force or not specified,
4862 reboot_cpu is s[mp]#### with #### being the processor
4863 to be used for rebooting.
4865 refscale.holdoff= [KNL]
4866 Set test-start holdoff period. The purpose of
4867 this parameter is to delay the start of the
4868 test until boot completes in order to avoid
4871 refscale.loops= [KNL]
4872 Set the number of loops over the synchronization
4873 primitive under test. Increasing this number
4874 reduces noise due to loop start/end overhead,
4875 but the default has already reduced the per-pass
4876 noise to a handful of picoseconds on ca. 2020
4879 refscale.nreaders= [KNL]
4880 Set number of readers. The default value of -1
4881 selects N, where N is roughly 75% of the number
4882 of CPUs. A value of zero is an interesting choice.
4884 refscale.nruns= [KNL]
4885 Set number of runs, each of which is dumped onto
4888 refscale.readdelay= [KNL]
4889 Set the read-side critical-section duration,
4890 measured in microseconds.
4892 refscale.scale_type= [KNL]
4893 Specify the read-protection implementation to test.
4895 refscale.shutdown= [KNL]
4896 Shut down the system at the end of the performance
4897 test. This defaults to 1 (shut it down) when
4898 refscale is built into the kernel and to 0 (leave
4899 it running) when refscale is built as a module.
4901 refscale.verbose= [KNL]
4902 Enable additional printk() statements.
4904 refscale.verbose_batched= [KNL]
4905 Batch the additional printk() statements. If zero
4906 (the default) or negative, print everything. Otherwise,
4907 print every Nth verbose statement, where N is the value
4911 [KNL, SMP] Set scheduler's default relax_domain_level.
4912 See Documentation/admin-guide/cgroup-v1/cpusets.rst.
4914 reserve= [KNL,BUGS] Force kernel to ignore I/O ports or memory
4915 Format: <base1>,<size1>[,<base2>,<size2>,...]
4916 Reserve I/O ports or memory so the kernel won't use
4917 them. If <base> is less than 0x10000, the region
4918 is assumed to be I/O ports; otherwise it is memory.
4920 reservetop= [X86-32]
4922 Reserves a hole at the top of the kernel virtual
4925 reset_devices [KNL] Force drivers to reset the underlying device
4926 during initialization.
4929 Specify the partition device for software suspend
4931 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>}
4933 resume_offset= [SWSUSP]
4934 Specify the offset from the beginning of the partition
4935 given by "resume=" at which the swap header is located,
4936 in <PAGE_SIZE> units (needed only for swap files).
4937 See Documentation/power/swsusp-and-swap-files.rst
4939 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to
4940 read the resume files
4942 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up.
4943 Useful for devices that are detected asynchronously
4944 (e.g. USB and MMC devices).
4946 hibernate= [HIBERNATION]
4947 noresume Don't check if there's a hibernation image
4948 present during boot.
4949 nocompress Don't compress/decompress hibernation images.
4950 no Disable hibernation and resume.
4951 protect_image Turn on image protection during restoration
4952 (that will set all pages holding image data
4953 during restoration read-only).
4955 retain_initrd [RAM] Keep initrd memory after extraction
4957 rfkill.default_state=
4958 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm,
4959 etc. communication is blocked by default.
4962 rfkill.master_switch_mode=
4963 0 The "airplane mode" button does nothing.
4964 1 The "airplane mode" button toggles between everything
4965 blocked and the previous configuration.
4966 2 The "airplane mode" button toggles between everything
4967 blocked and everything unblocked.
4969 rhash_entries= [KNL,NET]
4970 Set number of hash buckets for route cache
4973 [KNL] Disable ring 3 MONITOR/MWAIT feature on supported
4976 ro [KNL] Mount root device read-only on boot
4979 on Mark read-only kernel memory as read-only (default).
4980 off Leave read-only kernel memory writable for debugging.
4983 Enable the uart passthrough on the designated usb port
4984 on Rockchip SoCs. When active, the signals of the
4985 debug-uart get routed to the D+ and D- pins of the usb
4986 port and the regular usb controller gets disabled.
4988 root= [KNL] Root filesystem
4989 See name_to_dev_t comment in init/do_mounts.c.
4991 rootdelay= [KNL] Delay (in seconds) to pause before attempting to
4992 mount the root filesystem
4994 rootflags= [KNL] Set root filesystem mount option string
4996 rootfstype= [KNL] Set root filesystem type
4998 rootwait [KNL] Wait (indefinitely) for root device to show up.
4999 Useful for devices that are detected asynchronously
5000 (e.g. USB and MMC devices).
5002 rproc_mem=nn[KMG][@address]
5003 [KNL,ARM,CMA] Remoteproc physical memory block.
5004 Memory area to be used by remote processor image,
5007 rw [KNL] Mount root device read-write on boot
5009 S [KNL] Run init in single mode
5011 s390_iommu= [HW,S390]
5012 Set s390 IOTLB flushing mode
5014 With strict flushing every unmap operation will result in
5015 an IOTLB flush. Default is lazy flushing before reuse,
5018 s390_iommu_aperture= [KNL,S390]
5019 Specifies the size of the per device DMA address space
5020 accessible through the DMA and IOMMU APIs as a decimal
5021 factor of the size of main memory.
5022 The default is 1 meaning that one can concurrently use
5023 as many DMA addresses as physical memory is installed,
5024 if supported by hardware, and thus map all of memory
5025 once. With a value of 2 one can map all of memory twice
5026 and so on. As a special case a factor of 0 imposes no
5027 restrictions other than those given by hardware at the
5028 cost of significant additional memory use for tables.
5031 See drivers/net/irda/sa1100_ir.c.
5033 sched_verbose [KNL] Enables verbose scheduler debug messages.
5035 schedstats= [KNL,X86] Enable or disable scheduled statistics.
5036 Allowed values are enable and disable. This feature
5037 incurs a small amount of overhead in the scheduler
5038 but is useful for debugging and performance tuning.
5040 sched_thermal_decay_shift=
5041 [KNL, SMP] Set a decay shift for scheduler thermal
5042 pressure signal. Thermal pressure signal follows the
5043 default decay period of other scheduler pelt
5044 signals(usually 32 ms but configurable). Setting
5045 sched_thermal_decay_shift will left shift the decay
5046 period for the thermal pressure signal by the shift
5048 i.e. with the default pelt decay period of 32 ms
5049 sched_thermal_decay_shift thermal pressure decay pr
5053 Format: integer between 0 and 10
5056 scftorture.holdoff= [KNL]
5057 Number of seconds to hold off before starting
5058 test. Defaults to zero for module insertion and
5059 to 10 seconds for built-in smp_call_function()
5062 scftorture.longwait= [KNL]
5063 Request ridiculously long waits randomly selected
5064 up to the chosen limit in seconds. Zero (the
5065 default) disables this feature. Please note
5066 that requesting even small non-zero numbers of
5067 seconds can result in RCU CPU stall warnings,
5068 softlockup complaints, and so on.
5070 scftorture.nthreads= [KNL]
5071 Number of kthreads to spawn to invoke the
5072 smp_call_function() family of functions.
5073 The default of -1 specifies a number of kthreads
5074 equal to the number of CPUs.
5076 scftorture.onoff_holdoff= [KNL]
5077 Number seconds to wait after the start of the
5078 test before initiating CPU-hotplug operations.
5080 scftorture.onoff_interval= [KNL]
5081 Number seconds to wait between successive
5082 CPU-hotplug operations. Specifying zero (which
5083 is the default) disables CPU-hotplug operations.
5085 scftorture.shutdown_secs= [KNL]
5086 The number of seconds following the start of the
5087 test after which to shut down the system. The
5088 default of zero avoids shutting down the system.
5089 Non-zero values are useful for automated tests.
5091 scftorture.stat_interval= [KNL]
5092 The number of seconds between outputting the
5093 current test statistics to the console. A value
5094 of zero disables statistics output.
5096 scftorture.stutter_cpus= [KNL]
5097 The number of jiffies to wait between each change
5098 to the set of CPUs under test.
5100 scftorture.use_cpus_read_lock= [KNL]
5101 Use use_cpus_read_lock() instead of the default
5102 preempt_disable() to disable CPU hotplug
5103 while invoking one of the smp_call_function*()
5106 scftorture.verbose= [KNL]
5107 Enable additional printk() statements.
5109 scftorture.weight_single= [KNL]
5110 The probability weighting to use for the
5111 smp_call_function_single() function with a zero
5112 "wait" parameter. A value of -1 selects the
5113 default if all other weights are -1. However,
5114 if at least one weight has some other value, a
5115 value of -1 will instead select a weight of zero.
5117 scftorture.weight_single_wait= [KNL]
5118 The probability weighting to use for the
5119 smp_call_function_single() function with a
5120 non-zero "wait" parameter. See weight_single.
5122 scftorture.weight_many= [KNL]
5123 The probability weighting to use for the
5124 smp_call_function_many() function with a zero
5125 "wait" parameter. See weight_single.
5126 Note well that setting a high probability for
5127 this weighting can place serious IPI load
5130 scftorture.weight_many_wait= [KNL]
5131 The probability weighting to use for the
5132 smp_call_function_many() function with a
5133 non-zero "wait" parameter. See weight_single
5136 scftorture.weight_all= [KNL]
5137 The probability weighting to use for the
5138 smp_call_function_all() function with a zero
5139 "wait" parameter. See weight_single and
5142 scftorture.weight_all_wait= [KNL]
5143 The probability weighting to use for the
5144 smp_call_function_all() function with a
5145 non-zero "wait" parameter. See weight_single
5148 skew_tick= [KNL] Offset the periodic timer tick per cpu to mitigate
5149 xtime_lock contention on larger systems, and/or RCU lock
5150 contention on all systems with CONFIG_MAXSMP set.
5151 Format: { "0" | "1" }
5152 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1"
5154 Note: increases power consumption, thus should only be
5155 enabled if running jitter sensitive (HPC/RT) workloads.
5157 security= [SECURITY] Choose a legacy "major" security module to
5158 enable at boot. This has been deprecated by the
5161 selinux= [SELINUX] Disable or enable SELinux at boot time.
5162 Format: { "0" | "1" }
5163 See security/selinux/Kconfig help text.
5168 apparmor= [APPARMOR] Disable or enable AppArmor at boot time
5169 Format: { "0" | "1" }
5170 See security/apparmor/Kconfig help text
5173 Default value is set via kernel config option.
5175 serialnumber [BUGS=X86-32]
5178 Maximal number of shapers.
5186 Enable merging of slabs with similar size when the
5187 kernel is built without CONFIG_SLAB_MERGE_DEFAULT.
5190 Disable merging of slabs with similar size. May be
5191 necessary if there is some reason to distinguish
5192 allocs to different slabs, especially in hardened
5193 environments where the risk of heap overflows and
5194 layout control by attackers can usually be
5195 frustrated by disabling merging. This will reduce
5196 most of the exposure of a heap attack to a single
5197 cache (risks via metadata attacks are mostly
5198 unchanged). Debug options disable merging on their
5200 For more information see Documentation/vm/slub.rst.
5202 slab_max_order= [MM, SLAB]
5203 Determines the maximum allowed order for slabs.
5204 A high setting may cause OOMs due to memory
5205 fragmentation. Defaults to 1 for systems with
5206 more than 32MB of RAM, 0 otherwise.
5208 slub_debug[=options[,slabs][;[options[,slabs]]...] [MM, SLUB]
5209 Enabling slub_debug allows one to determine the
5210 culprit if slab objects become corrupted. Enabling
5211 slub_debug can create guard zones around objects and
5212 may poison objects when not in use. Also tracks the
5213 last alloc / free. For more information see
5214 Documentation/vm/slub.rst.
5216 slub_max_order= [MM, SLUB]
5217 Determines the maximum allowed order for slabs.
5218 A high setting may cause OOMs due to memory
5219 fragmentation. For more information see
5220 Documentation/vm/slub.rst.
5222 slub_min_objects= [MM, SLUB]
5223 The minimum number of objects per slab. SLUB will
5224 increase the slab order up to slub_max_order to
5225 generate a sufficiently large slab able to contain
5226 the number of objects indicated. The higher the number
5227 of objects the smaller the overhead of tracking slabs
5228 and the less frequently locks need to be acquired.
5229 For more information see Documentation/vm/slub.rst.
5231 slub_min_order= [MM, SLUB]
5232 Determines the minimum page order for slabs. Must be
5233 lower than slub_max_order.
5234 For more information see Documentation/vm/slub.rst.
5236 slub_merge [MM, SLUB]
5237 Same with slab_merge.
5239 slub_nomerge [MM, SLUB]
5240 Same with slab_nomerge. This is supported for legacy.
5241 See slab_nomerge for more information.
5244 Format: <io1>[,<io2>[,...,<io8>]]
5246 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices
5247 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port
5248 smsc-ircc2.ircc_sir= [HW] SIR base I/O port
5249 smsc-ircc2.ircc_fir= [HW] FIR base I/O port
5250 smsc-ircc2.ircc_irq= [HW] IRQ line
5251 smsc-ircc2.ircc_dma= [HW] DMA channel
5252 smsc-ircc2.ircc_transceiver= [HW] Transceiver type:
5253 0: Toshiba Satellite 1800 (GP data pin select)
5254 1: Fast pin select (default)
5257 smt [KNL,S390] Set the maximum number of threads (logical
5258 CPUs) to use per physical CPU on systems capable of
5259 symmetric multithreading (SMT). Will be capped to the
5260 actual hardware limit.
5262 Default: -1 (no limit)
5265 [KNL] Should the soft-lockup detector generate panics.
5268 A value of 1 instructs the soft-lockup detector
5269 to panic the machine when a soft-lockup occurs. It is
5270 also controlled by the kernel.softlockup_panic sysctl
5271 and CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC, which is the
5272 respective build-time switch to that functionality.
5274 softlockup_all_cpu_backtrace=
5275 [KNL] Should the soft-lockup detector generate
5276 backtraces on all cpus.
5279 sonypi.*= [HW] Sony Programmable I/O Control Device driver
5280 See Documentation/admin-guide/laptops/sonypi.rst
5282 spectre_v2= [X86] Control mitigation of Spectre variant 2
5283 (indirect branch speculation) vulnerability.
5284 The default operation protects the kernel from
5287 on - unconditionally enable, implies
5289 off - unconditionally disable, implies
5291 auto - kernel detects whether your CPU model is
5294 Selecting 'on' will, and 'auto' may, choose a
5295 mitigation method at run time according to the
5296 CPU, the available microcode, the setting of the
5297 CONFIG_RETPOLINE configuration option, and the
5298 compiler with which the kernel was built.
5300 Selecting 'on' will also enable the mitigation
5301 against user space to user space task attacks.
5303 Selecting 'off' will disable both the kernel and
5304 the user space protections.
5306 Specific mitigations can also be selected manually:
5308 retpoline - replace indirect branches
5309 retpoline,generic - google's original retpoline
5310 retpoline,amd - AMD-specific minimal thunk
5312 Not specifying this option is equivalent to
5316 [X86] Control mitigation of Spectre variant 2
5317 (indirect branch speculation) vulnerability between
5320 on - Unconditionally enable mitigations. Is
5321 enforced by spectre_v2=on
5323 off - Unconditionally disable mitigations. Is
5324 enforced by spectre_v2=off
5326 prctl - Indirect branch speculation is enabled,
5327 but mitigation can be enabled via prctl
5328 per thread. The mitigation control state
5329 is inherited on fork.
5332 - Like "prctl" above, but only STIBP is
5333 controlled per thread. IBPB is issued
5334 always when switching between different user
5338 - Same as "prctl" above, but all seccomp
5339 threads will enable the mitigation unless
5340 they explicitly opt out.
5343 - Like "seccomp" above, but only STIBP is
5344 controlled per thread. IBPB is issued
5345 always when switching between different
5346 user space processes.
5348 auto - Kernel selects the mitigation depending on
5349 the available CPU features and vulnerability.
5351 Default mitigation: "prctl"
5353 Not specifying this option is equivalent to
5354 spectre_v2_user=auto.
5356 spec_store_bypass_disable=
5357 [HW] Control Speculative Store Bypass (SSB) Disable mitigation
5358 (Speculative Store Bypass vulnerability)
5360 Certain CPUs are vulnerable to an exploit against a
5361 a common industry wide performance optimization known
5362 as "Speculative Store Bypass" in which recent stores
5363 to the same memory location may not be observed by
5364 later loads during speculative execution. The idea
5365 is that such stores are unlikely and that they can
5366 be detected prior to instruction retirement at the
5367 end of a particular speculation execution window.
5369 In vulnerable processors, the speculatively forwarded
5370 store can be used in a cache side channel attack, for
5371 example to read memory to which the attacker does not
5372 directly have access (e.g. inside sandboxed code).
5374 This parameter controls whether the Speculative Store
5375 Bypass optimization is used.
5377 On x86 the options are:
5379 on - Unconditionally disable Speculative Store Bypass
5380 off - Unconditionally enable Speculative Store Bypass
5381 auto - Kernel detects whether the CPU model contains an
5382 implementation of Speculative Store Bypass and
5383 picks the most appropriate mitigation. If the
5384 CPU is not vulnerable, "off" is selected. If the
5385 CPU is vulnerable the default mitigation is
5386 architecture and Kconfig dependent. See below.
5387 prctl - Control Speculative Store Bypass per thread
5388 via prctl. Speculative Store Bypass is enabled
5389 for a process by default. The state of the control
5390 is inherited on fork.
5391 seccomp - Same as "prctl" above, but all seccomp threads
5392 will disable SSB unless they explicitly opt out.
5394 Default mitigations:
5397 On powerpc the options are:
5399 on,auto - On Power8 and Power9 insert a store-forwarding
5400 barrier on kernel entry and exit. On Power7
5401 perform a software flush on kernel entry and
5405 Not specifying this option is equivalent to
5406 spec_store_bypass_disable=auto.
5408 spia_io_base= [HW,MTD]
5414 [X86] Enable split lock detection or bus lock detection
5416 When enabled (and if hardware support is present), atomic
5417 instructions that access data across cache line
5418 boundaries will result in an alignment check exception
5419 for split lock detection or a debug exception for
5424 warn - the kernel will emit rate-limited warnings
5425 about applications triggering the #AC
5426 exception or the #DB exception. This mode is
5427 the default on CPUs that support split lock
5428 detection or bus lock detection. Default
5429 behavior is by #AC if both features are
5430 enabled in hardware.
5432 fatal - the kernel will send SIGBUS to applications
5433 that trigger the #AC exception or the #DB
5434 exception. Default behavior is by #AC if
5435 both features are enabled in hardware.
5438 Set system wide rate limit to N bus locks
5439 per second for bus lock detection.
5442 N/A for split lock detection.
5445 If an #AC exception is hit in the kernel or in
5446 firmware (i.e. not while executing in user mode)
5447 the kernel will oops in either "warn" or "fatal"
5450 #DB exception for bus lock is triggered only when
5454 Control the Special Register Buffer Data Sampling
5457 Certain CPUs are vulnerable to an MDS-like
5458 exploit which can leak bits from the random
5461 By default, this issue is mitigated by
5462 microcode. However, the microcode fix can cause
5463 the RDRAND and RDSEED instructions to become
5464 much slower. Among other effects, this will
5465 result in reduced throughput from /dev/urandom.
5467 The microcode mitigation can be disabled with
5468 the following option:
5470 off: Disable mitigation and remove
5471 performance impact to RDRAND and RDSEED
5473 srcutree.counter_wrap_check [KNL]
5474 Specifies how frequently to check for
5475 grace-period sequence counter wrap for the
5476 srcu_data structure's ->srcu_gp_seq_needed field.
5477 The greater the number of bits set in this kernel
5478 parameter, the less frequently counter wrap will
5479 be checked for. Note that the bottom two bits
5482 srcutree.exp_holdoff [KNL]
5483 Specifies how many nanoseconds must elapse
5484 since the end of the last SRCU grace period for
5485 a given srcu_struct until the next normal SRCU
5486 grace period will be considered for automatic
5487 expediting. Set to zero to disable automatic
5491 Speculative Store Bypass Disable control
5493 On CPUs that are vulnerable to the Speculative
5494 Store Bypass vulnerability and offer a
5495 firmware based mitigation, this parameter
5496 indicates how the mitigation should be used:
5498 force-on: Unconditionally enable mitigation for
5499 for both kernel and userspace
5500 force-off: Unconditionally disable mitigation for
5501 for both kernel and userspace
5502 kernel: Always enable mitigation in the
5503 kernel, and offer a prctl interface
5504 to allow userspace to register its
5505 interest in being mitigated too.
5507 stack_guard_gap= [MM]
5508 override the default stack gap protection. The value
5509 is in page units and it defines how many pages prior
5510 to (for stacks growing down) resp. after (for stacks
5511 growing up) the main stack are reserved for no other
5512 mapping. Default value is 256 pages.
5514 stack_depot_disable= [KNL]
5515 Setting this to true through kernel command line will
5516 disable the stack depot thereby saving the static memory
5517 consumed by the stack hash table. By default this is set
5521 Enabled the stack tracer on boot up.
5523 stacktrace_filter=[function-list]
5524 [FTRACE] Limit the functions that the stack tracer
5525 will trace at boot up. function-list is a comma-separated
5526 list of functions. This list can be changed at run
5527 time by the stack_trace_filter file in the debugfs
5528 tracing directory. Note, this enables stack tracing
5529 and the stacktrace above is not needed.
5533 Set the STI (builtin display/keyboard on the HP-PARISC
5534 machines) console (graphic card) which should be used
5535 as the initial boot-console.
5536 See also comment in drivers/video/console/sticore.c.
5539 See comment in drivers/video/console/sticore.c.
5542 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]]
5547 Enable or disable strict sigaltstack size checks
5548 against the required signal frame size which
5549 depends on the supported FPU features. This can
5550 be used to filter out binaries which have
5551 not yet been made aware of AT_MINSIGSTKSZ.
5553 sunrpc.min_resvport=
5554 sunrpc.max_resvport=
5556 SunRPC servers often require that client requests
5557 originate from a privileged port (i.e. a port in the
5558 range 0 < portnr < 1024).
5559 An administrator who wishes to reserve some of these
5560 ports for other uses may adjust the range that the
5561 kernel's sunrpc client considers to be privileged
5562 using these two parameters to set the minimum and
5563 maximum port values.
5565 sunrpc.svc_rpc_per_connection_limit=
5567 Limit the number of requests that the server will
5568 process in parallel from a single connection.
5569 The default value is 0 (no limit).
5573 Control how the NFS server code allocates CPUs to
5574 service thread pools. Depending on how many NICs
5575 you have and where their interrupts are bound, this
5576 option will affect which CPUs will do NFS serving.
5577 Note: this parameter cannot be changed while the
5578 NFS server is running.
5580 auto the server chooses an appropriate mode
5581 automatically using heuristics
5582 global a single global pool contains all CPUs
5583 percpu one pool for each CPU
5584 pernode one pool for each NUMA node (equivalent
5585 to global on non-NUMA machines)
5587 sunrpc.tcp_slot_table_entries=
5588 sunrpc.udp_slot_table_entries=
5590 Sets the upper limit on the number of simultaneous
5591 RPC calls that can be sent from the client to a
5592 server. Increasing these values may allow you to
5593 improve throughput, but will also increase the
5594 amount of memory reserved for use by the client.
5596 suspend.pm_test_delay=
5598 Sets the number of seconds to remain in a suspend test
5599 mode before resuming the system (see
5600 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG
5601 is set. Default value is 5.
5604 Format: { on | off | y | n | 1 | 0 }
5605 This parameter controls use of the Protected
5606 Execution Facility on pSeries.
5609 [KNL] Enable accounting of swap in memory resource
5610 controller if no parameter or 1 is given or disable
5611 it if 0 is given (See Documentation/admin-guide/cgroup-v1/memory.rst)
5613 swiotlb= [ARM,IA-64,PPC,MIPS,X86]
5614 Format: { <int> | force | noforce }
5615 <int> -- Number of I/O TLB slabs
5616 force -- force using of bounce buffers even if they
5617 wouldn't be automatically used by the kernel
5618 noforce -- Never use bounce buffers (for debugging)
5623 Set a sysctl parameter, right before loading the init
5624 process, as if the value was written to the respective
5625 /proc/sys/... file. Both '.' and '/' are recognized as
5626 separators. Unrecognized parameters and invalid values
5627 are reported in the kernel log. Sysctls registered
5628 later by a loaded module cannot be set this way.
5629 Example: sysctl.vm.swappiness=40
5631 sysfs.deprecated=0|1 [KNL]
5632 Enable/disable old style sysfs layout for old udev
5633 on older distributions. When this option is enabled
5634 very new udev will not work anymore. When this option
5635 is disabled (or CONFIG_SYSFS_DEPRECATED not compiled)
5636 in older udev will not work anymore.
5637 Default depends on CONFIG_SYSFS_DEPRECATED_V2 set in
5638 the kernel configuration.
5640 sysrq_always_enabled
5642 Ignore sysrq setting - this boot parameter will
5643 neutralize any effect of /proc/sys/kernel/sysrq.
5644 Useful for debugging.
5646 tcpmhash_entries= [KNL,NET]
5647 Set the number of tcp_metrics_hash slots.
5648 Default value is 8192 or 16384 depending on total
5649 ram pages. This is used to specify the TCP metrics
5650 cache size. See Documentation/networking/ip-sysctl.rst
5651 "tcp_no_metrics_save" section for more details.
5655 test_suspend= [SUSPEND][,N]
5656 Specify "mem" (for Suspend-to-RAM) or "standby" (for
5657 standby suspend) or "freeze" (for suspend type freeze)
5658 as the system sleep state during system startup with
5659 the optional capability to repeat N number of times.
5660 The system is woken from this state using a
5661 wakeup-capable RTC alarm.
5663 thash_entries= [KNL,NET]
5664 Set number of hash buckets for TCP connection
5666 thermal.act= [HW,ACPI]
5667 -1: disable all active trip points in all thermal zones
5668 <degrees C>: override all lowest active trip points
5670 thermal.crt= [HW,ACPI]
5671 -1: disable all critical trip points in all thermal zones
5672 <degrees C>: override all critical trip points
5674 thermal.nocrt= [HW,ACPI]
5675 Set to disable actions on ACPI thermal zone
5676 critical and hot trip points.
5678 thermal.off= [HW,ACPI]
5679 1: disable ACPI thermal control
5681 thermal.psv= [HW,ACPI]
5682 -1: disable all passive trip points
5683 <degrees C>: override all passive trip points to this
5686 thermal.tzp= [HW,ACPI]
5687 Specify global default ACPI thermal zone polling rate
5688 <deci-seconds>: poll all this frequency
5689 0: no polling (default)
5692 Force threading of all interrupt handlers except those
5693 marked explicitly IRQF_NO_THREAD.
5697 Specify if the kernel should make use of the cpu
5698 topology information if the hardware supports this.
5699 The scheduler will make use of this information and
5700 e.g. base its process migration decisions on it.
5703 topology_updates= [KNL, PPC, NUMA]
5705 Specify if the kernel should ignore (off)
5706 topology updates sent by the hypervisor to this
5709 torture.disable_onoff_at_boot= [KNL]
5710 Prevent the CPU-hotplug component of torturing
5711 until after init has spawned.
5713 torture.ftrace_dump_at_shutdown= [KNL]
5714 Dump the ftrace buffer at torture-test shutdown,
5715 even if there were no errors. This can be a
5716 very costly operation when many torture tests
5717 are running concurrently, especially on systems
5718 with rotating-rust storage.
5720 torture.verbose_sleep_frequency= [KNL]
5721 Specifies how many verbose printk()s should be
5722 emitted between each sleep. The default of zero
5723 disables verbose-printk() sleeping.
5725 torture.verbose_sleep_duration= [KNL]
5726 Duration of each verbose-printk() sleep in jiffies.
5730 tpm_suspend_pcr=[HW,TPM]
5731 Format: integer pcr id
5732 Specify that at suspend time, the tpm driver
5733 should extend the specified pcr with zeros,
5734 as a workaround for some chips which fail to
5735 flush the last written pcr on TPM_SaveState.
5736 This will guarantee that all the other pcrs
5739 trace_buf_size=nn[KMG]
5740 [FTRACE] will set tracing buffer size on each cpu.
5742 trace_event=[event-list]
5743 [FTRACE] Set and start specified trace events in order
5744 to facilitate early boot debugging. The event-list is a
5745 comma-separated list of trace events to enable. See
5746 also Documentation/trace/events.rst
5748 trace_options=[option-list]
5749 [FTRACE] Enable or disable tracer options at boot.
5750 The option-list is a comma delimited list of options
5751 that can be enabled or disabled just as if you were
5752 to echo the option name into
5754 /sys/kernel/debug/tracing/trace_options
5756 For example, to enable stacktrace option (to dump the
5757 stack trace of each event), add to the command line:
5759 trace_options=stacktrace
5761 See also Documentation/trace/ftrace.rst "trace options"
5765 Have the tracepoints sent to printk as well as the
5766 tracing ring buffer. This is useful for early boot up
5767 where the system hangs or reboots and does not give the
5768 option for reading the tracing buffer or performing a
5769 ftrace_dump_on_oops.
5771 To turn off having tracepoints sent to printk,
5772 echo 0 > /proc/sys/kernel/tracepoint_printk
5773 Note, echoing 1 into this file without the
5774 tracepoint_printk kernel cmdline option has no effect.
5776 The tp_printk_stop_on_boot (see below) can also be used
5777 to stop the printing of events to console at
5782 Having tracepoints sent to printk() and activating high
5783 frequency tracepoints such as irq or sched, can cause
5784 the system to live lock.
5786 tp_printk_stop_on_boot[FTRACE]
5787 When tp_printk (above) is set, it can cause a lot of noise
5788 on the console. It may be useful to only include the
5789 printing of events during boot up, as user space may
5790 make the system inoperable.
5792 This command line option will stop the printing of events
5793 to console at the late_initcall_sync() time frame.
5796 [FTRACE] enable this option to disable tracing when a
5797 warning is hit. This turns off "tracing_on". Tracing can
5798 be enabled again by echoing '1' into the "tracing_on"
5799 file located in /sys/kernel/debug/tracing/
5801 This option is useful, as it disables the trace before
5802 the WARNING dump is called, which prevents the trace to
5803 be filled with content caused by the warning output.
5805 This option can also be set at run time via the sysctl
5806 option: kernel/traceoff_on_warning
5808 transparent_hugepage=
5810 Format: [always|madvise|never]
5811 Can be used to control the default behavior of the system
5812 with respect to transparent hugepages.
5813 See Documentation/admin-guide/mm/transhuge.rst
5816 trusted.source= [KEYS]
5818 This parameter identifies the trust source as a backend
5819 for trusted keys implementation. Supported trust
5823 If not specified then it defaults to iterating through
5824 the trust source list starting with TPM and assigns the
5825 first trust source as a backend which is initialized
5826 successfully during iteration.
5828 tsc= Disable clocksource stability checks for TSC.
5830 [x86] reliable: mark tsc clocksource as reliable, this
5831 disables clocksource verification at runtime, as well
5832 as the stability checks done at bootup. Used to enable
5833 high-resolution timer mode on older hardware, and in
5834 virtualized environment.
5835 [x86] noirqtime: Do not use TSC to do irq accounting.
5836 Used to run time disable IRQ_TIME_ACCOUNTING on any
5837 platforms where RDTSC is slow and this accounting
5839 [x86] unstable: mark the TSC clocksource as unstable, this
5840 marks the TSC unconditionally unstable at bootup and
5841 avoids any further wobbles once the TSC watchdog notices.
5842 [x86] nowatchdog: disable clocksource watchdog. Used
5843 in situations with strict latency requirements (where
5844 interruptions from clocksource watchdog are not
5847 tsc_early_khz= [X86] Skip early TSC calibration and use the given
5848 value instead. Useful when the early TSC frequency discovery
5849 procedure is not reliable, such as on overclocked systems
5850 with CPUID.16h support and partial CPUID.15h support.
5851 Format: <unsigned int>
5853 tsx= [X86] Control Transactional Synchronization
5854 Extensions (TSX) feature in Intel processors that
5855 support TSX control.
5857 This parameter controls the TSX feature. The options are:
5859 on - Enable TSX on the system. Although there are
5860 mitigations for all known security vulnerabilities,
5861 TSX has been known to be an accelerator for
5862 several previous speculation-related CVEs, and
5863 so there may be unknown security risks associated
5864 with leaving it enabled.
5866 off - Disable TSX on the system. (Note that this
5867 option takes effect only on newer CPUs which are
5868 not vulnerable to MDS, i.e., have
5869 MSR_IA32_ARCH_CAPABILITIES.MDS_NO=1 and which get
5870 the new IA32_TSX_CTRL MSR through a microcode
5871 update. This new MSR allows for the reliable
5872 deactivation of the TSX functionality.)
5874 auto - Disable TSX if X86_BUG_TAA is present,
5875 otherwise enable TSX on the system.
5877 Not specifying this option is equivalent to tsx=off.
5879 See Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
5882 tsx_async_abort= [X86,INTEL] Control mitigation for the TSX Async
5883 Abort (TAA) vulnerability.
5885 Similar to Micro-architectural Data Sampling (MDS)
5886 certain CPUs that support Transactional
5887 Synchronization Extensions (TSX) are vulnerable to an
5888 exploit against CPU internal buffers which can forward
5889 information to a disclosure gadget under certain
5892 In vulnerable processors, the speculatively forwarded
5893 data can be used in a cache side channel attack, to
5894 access data to which the attacker does not have direct
5897 This parameter controls the TAA mitigation. The
5900 full - Enable TAA mitigation on vulnerable CPUs
5903 full,nosmt - Enable TAA mitigation and disable SMT on
5904 vulnerable CPUs. If TSX is disabled, SMT
5905 is not disabled because CPU is not
5906 vulnerable to cross-thread TAA attacks.
5907 off - Unconditionally disable TAA mitigation
5909 On MDS-affected machines, tsx_async_abort=off can be
5910 prevented by an active MDS mitigation as both vulnerabilities
5911 are mitigated with the same mechanism so in order to disable
5912 this mitigation, you need to specify mds=off too.
5914 Not specifying this option is equivalent to
5915 tsx_async_abort=full. On CPUs which are MDS affected
5916 and deploy MDS mitigation, TAA mitigation is not
5917 required and doesn't provide any additional
5921 Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
5923 turbografx.map[2|3]= [HW,JOY]
5924 TurboGraFX parallel port interface
5926 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7>
5927 See also Documentation/input/devices/joystick-parport.rst
5929 udbg-immortal [PPC] When debugging early kernel crashes that
5930 happen after console_init() and before a proper
5931 console driver takes over, this boot options might
5932 help "seeing" what's going on.
5934 uhash_entries= [KNL,NET]
5935 Set number of hash buckets for UDP/UDP-Lite connections
5938 [USB] Ignore overcurrent events (default N).
5939 Some badly-designed motherboards generate lots of
5940 bogus events, for ports that aren't wired to
5941 anything. Set this parameter to avoid log spamming.
5942 Note that genuine overcurrent events won't be
5946 [X86] Cause panic on unknown NMI.
5948 usbcore.authorized_default=
5949 [USB] Default USB device authorization:
5950 (default -1 = authorized except for wireless USB,
5951 0 = not authorized, 1 = authorized, 2 = authorized
5952 if device connected to internal port)
5954 usbcore.autosuspend=
5955 [USB] The autosuspend time delay (in seconds) used
5956 for newly-detected USB devices (default 2). This
5957 is the time required before an idle device will be
5958 autosuspended. Devices for which the delay is set
5959 to a negative value won't be autosuspended at all.
5961 usbcore.usbfs_snoop=
5962 [USB] Set to log all usbfs traffic (default 0 = off).
5964 usbcore.usbfs_snoop_max=
5965 [USB] Maximum number of bytes to snoop in each URB
5968 usbcore.blinkenlights=
5969 [USB] Set to cycle leds on hubs (default 0 = off).
5971 usbcore.old_scheme_first=
5972 [USB] Start with the old device initialization
5973 scheme (default 0 = off).
5975 usbcore.usbfs_memory_mb=
5976 [USB] Memory limit (in MB) for buffers allocated by
5977 usbfs (default = 16, 0 = max = 2047).
5979 usbcore.use_both_schemes=
5980 [USB] Try the other device initialization scheme
5981 if the first one fails (default 1 = enabled).
5983 usbcore.initial_descriptor_timeout=
5984 [USB] Specifies timeout for the initial 64-byte
5985 USB_REQ_GET_DESCRIPTOR request in milliseconds
5986 (default 5000 = 5.0 seconds).
5988 usbcore.nousb [USB] Disable the USB subsystem
5991 [USB] A list of quirk entries to augment the built-in
5992 usb core quirk list. List entries are separated by
5993 commas. Each entry has the form
5994 VendorID:ProductID:Flags. The IDs are 4-digit hex
5995 numbers and Flags is a set of letters. Each letter
5996 will change the built-in quirk; setting it if it is
5997 clear and clearing it if it is set. The letters have
5998 the following meanings:
5999 a = USB_QUIRK_STRING_FETCH_255 (string
6000 descriptors must not be fetched using
6002 b = USB_QUIRK_RESET_RESUME (device can't resume
6003 correctly so reset it instead);
6004 c = USB_QUIRK_NO_SET_INTF (device can't handle
6005 Set-Interface requests);
6006 d = USB_QUIRK_CONFIG_INTF_STRINGS (device can't
6007 handle its Configuration or Interface
6009 e = USB_QUIRK_RESET (device can't be reset
6010 (e.g morph devices), don't use reset);
6011 f = USB_QUIRK_HONOR_BNUMINTERFACES (device has
6012 more interface descriptions than the
6013 bNumInterfaces count, and can't handle
6014 talking to these interfaces);
6015 g = USB_QUIRK_DELAY_INIT (device needs a pause
6016 during initialization, after we read
6017 the device descriptor);
6018 h = USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL (For
6019 high speed and super speed interrupt
6020 endpoints, the USB 2.0 and USB 3.0 spec
6021 require the interval in microframes (1
6022 microframe = 125 microseconds) to be
6023 calculated as interval = 2 ^
6025 Devices with this quirk report their
6026 bInterval as the result of this
6027 calculation instead of the exponent
6028 variable used in the calculation);
6029 i = USB_QUIRK_DEVICE_QUALIFIER (device can't
6030 handle device_qualifier descriptor
6032 j = USB_QUIRK_IGNORE_REMOTE_WAKEUP (device
6033 generates spurious wakeup, ignore
6034 remote wakeup capability);
6035 k = USB_QUIRK_NO_LPM (device can't handle Link
6037 l = USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL
6038 (Device reports its bInterval as linear
6039 frames instead of the USB 2.0
6041 m = USB_QUIRK_DISCONNECT_SUSPEND (Device needs
6042 to be disconnected before suspend to
6043 prevent spurious wakeup);
6044 n = USB_QUIRK_DELAY_CTRL_MSG (Device needs a
6045 pause after every control message);
6046 o = USB_QUIRK_HUB_SLOW_RESET (Hub needs extra
6047 delay after resetting its port);
6048 Example: quirks=0781:5580:bk,0a5c:5834:gij
6051 [USBHID] The interval which mice are to be polled at.
6054 [USBHID] The interval which joysticks are to be polled at.
6057 [USBHID] The interval which keyboards are to be polled at.
6059 usb-storage.delay_use=
6060 [UMS] The delay in seconds before a new device is
6061 scanned for Logical Units (default 1).
6064 [UMS] A list of quirks entries to supplement or
6065 override the built-in unusual_devs list. List
6066 entries are separated by commas. Each entry has
6067 the form VID:PID:Flags where VID and PID are Vendor
6068 and Product ID values (4-digit hex numbers) and
6069 Flags is a set of characters, each corresponding
6070 to a common usb-storage quirk flag as follows:
6071 a = SANE_SENSE (collect more than 18 bytes
6072 of sense data, not on uas);
6073 b = BAD_SENSE (don't collect more than 18
6074 bytes of sense data, not on uas);
6075 c = FIX_CAPACITY (decrease the reported
6076 device capacity by one sector);
6077 d = NO_READ_DISC_INFO (don't use
6078 READ_DISC_INFO command, not on uas);
6079 e = NO_READ_CAPACITY_16 (don't use
6080 READ_CAPACITY_16 command);
6081 f = NO_REPORT_OPCODES (don't use report opcodes
6083 g = MAX_SECTORS_240 (don't transfer more than
6084 240 sectors at a time, uas only);
6085 h = CAPACITY_HEURISTICS (decrease the
6086 reported device capacity by one
6087 sector if the number is odd);
6088 i = IGNORE_DEVICE (don't bind to this
6090 j = NO_REPORT_LUNS (don't use report luns
6092 k = NO_SAME (do not use WRITE_SAME, uas only)
6093 l = NOT_LOCKABLE (don't try to lock and
6094 unlock ejectable media, not on uas);
6095 m = MAX_SECTORS_64 (don't transfer more
6096 than 64 sectors = 32 KB at a time,
6098 n = INITIAL_READ10 (force a retry of the
6099 initial READ(10) command, not on uas);
6100 o = CAPACITY_OK (accept the capacity
6101 reported by the device, not on uas);
6102 p = WRITE_CACHE (the device cache is ON
6103 by default, not on uas);
6104 r = IGNORE_RESIDUE (the device reports
6105 bogus residue values, not on uas);
6106 s = SINGLE_LUN (the device has only one
6108 t = NO_ATA_1X (don't allow ATA(12) and ATA(16)
6109 commands, uas only);
6110 u = IGNORE_UAS (don't bind to the uas driver);
6111 w = NO_WP_DETECT (don't test whether the
6112 medium is write-protected).
6113 y = ALWAYS_SYNC (issue a SYNCHRONIZE_CACHE
6114 even if the device claims no cache,
6116 Example: quirks=0419:aaf5:rl,0421:0433:rc
6118 user_debug= [KNL,ARM]
6120 See arch/arm/Kconfig.debug help text.
6121 1 - undefined instruction events
6123 4 - invalid data aborts
6126 Example: user_debug=31
6129 [X86] Flags controlling user PTE allocations.
6131 nohigh = do not allocate PTE pages in
6132 HIGHMEM regardless of setting
6136 On X86_32, this is an alias for vdso32=. Otherwise:
6138 vdso=1: enable VDSO (the default)
6139 vdso=0: disable VDSO mapping
6141 vdso32= [X86] Control the 32-bit vDSO
6142 vdso32=1: enable 32-bit VDSO
6143 vdso32=0 or vdso32=2: disable 32-bit VDSO
6145 See the help text for CONFIG_COMPAT_VDSO for more
6146 details. If CONFIG_COMPAT_VDSO is set, the default is
6147 vdso32=0; otherwise, the default is vdso32=1.
6149 For compatibility with older kernels, vdso32=2 is an
6152 Try vdso32=0 if you encounter an error that says:
6153 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed!
6156 vector=percpu: enable percpu vector domain
6158 video= [FB] Frame buffer configuration
6159 See Documentation/fb/modedb.rst.
6161 video.brightness_switch_enabled= [0,1]
6162 If set to 1, on receiving an ACPI notify event
6163 generated by hotkey, video driver will adjust brightness
6164 level and then send out the event to user space through
6165 the allocated input device; If set to 0, video driver
6166 will only send out the event without touching backlight
6171 [VMMIO] Memory mapped virtio (platform) device.
6173 <size>@<baseaddr>:<irq>[:<id>]
6175 <size> := size (can use standard suffixes
6177 <baseaddr> := physical base address
6178 <irq> := interrupt number (as passed to
6180 <id> := (optional) platform device id
6182 virtio_mmio.device=1K@0x100b0000:48:7
6184 Can be used multiple times for multiple devices.
6186 vga= [BOOT,X86-32] Select a particular video mode
6187 See Documentation/x86/boot.rst and
6188 Documentation/admin-guide/svga.rst.
6189 Use vga=ask for menu.
6190 This is actually a boot loader parameter; the value is
6191 passed to the kernel using a special protocol.
6193 vm_debug[=options] [KNL] Available with CONFIG_DEBUG_VM=y.
6194 May slow down system boot speed, especially when
6195 enabled on systems with a large amount of memory.
6196 All options are enabled by default, and this
6197 interface is meant to allow for selectively
6198 enabling or disabling specific virtual memory
6201 Available options are:
6202 P Enable page structure init time poisoning
6203 - Disable all of the above options
6205 vmalloc=nn[KMG] [KNL,BOOT] Forces the vmalloc area to have an exact
6206 size of <nn>. This can be used to increase the
6207 minimum size (128MB on x86). It can also be used to
6208 decrease the size and leave more room for directly
6211 vmcp_cma=nn[MG] [KNL,S390]
6212 Sets the memory size reserved for contiguous memory
6213 allocations for the vmcp device driver.
6215 vmhalt= [KNL,S390] Perform z/VM CP command after system halt.
6218 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic.
6221 vmpoff= [KNL,S390] Perform z/VM CP command after power off.
6225 Controls the behavior of vsyscalls (i.e. calls to
6226 fixed addresses of 0xffffffffff600x00 from legacy
6227 code). Most statically-linked binaries and older
6228 versions of glibc use these calls. Because these
6229 functions are at fixed addresses, they make nice
6230 targets for exploits that can control RIP.
6232 emulate [default] Vsyscalls turn into traps and are
6233 emulated reasonably safely. The vsyscall
6236 xonly Vsyscalls turn into traps and are
6237 emulated reasonably safely. The vsyscall
6238 page is not readable.
6240 none Vsyscalls don't work at all. This makes
6241 them quite hard to use for exploits but
6242 might break your system.
6244 vt.color= [VT] Default text color.
6245 Format: 0xYX, X = foreground, Y = background.
6246 Default: 0x07 = light gray on black.
6248 vt.cur_default= [VT] Default cursor shape.
6249 Format: 0xCCBBAA, where AA, BB, and CC are the same as
6250 the parameters of the <Esc>[?A;B;Cc escape sequence;
6251 see VGA-softcursor.txt. Default: 2 = underline.
6253 vt.default_blu= [VT]
6254 Format: <blue0>,<blue1>,<blue2>,...,<blue15>
6255 Change the default blue palette of the console.
6256 This is a 16-member array composed of values
6259 vt.default_grn= [VT]
6260 Format: <green0>,<green1>,<green2>,...,<green15>
6261 Change the default green palette of the console.
6262 This is a 16-member array composed of values
6265 vt.default_red= [VT]
6266 Format: <red0>,<red1>,<red2>,...,<red15>
6267 Change the default red palette of the console.
6268 This is a 16-member array composed of values
6274 Set system-wide default UTF-8 mode for all tty's.
6275 Default is 1, i.e. UTF-8 mode is enabled for all
6276 newly opened terminals.
6278 vt.global_cursor_default=
6281 Set system-wide default for whether a cursor
6282 is shown on new VTs. Default is -1,
6283 i.e. cursors will be created by default unless
6284 overridden by individual drivers. 0 will hide
6285 cursors, 1 will display them.
6287 vt.italic= [VT] Default color for italic text; 0-15.
6290 vt.underline= [VT] Default color for underlined text; 0-15.
6293 watchdog timers [HW,WDT] For information on watchdog timers,
6294 see Documentation/watchdog/watchdog-parameters.rst
6295 or other driver-specific files in the
6296 Documentation/watchdog/ directory.
6300 Set the hard lockup detector stall duration
6301 threshold in seconds. The soft lockup detector
6302 threshold is set to twice the value. A value of 0
6303 disables both lockup detectors. Default is 10
6306 workqueue.watchdog_thresh=
6307 If CONFIG_WQ_WATCHDOG is configured, workqueue can
6308 warn stall conditions and dump internal state to
6309 help debugging. 0 disables workqueue stall
6310 detection; otherwise, it's the stall threshold
6311 duration in seconds. The default value is 30 and
6312 it can be updated at runtime by writing to the
6313 corresponding sysfs file.
6315 workqueue.disable_numa
6316 By default, all work items queued to unbound
6317 workqueues are affine to the NUMA nodes they're
6318 issued on, which results in better behavior in
6319 general. If NUMA affinity needs to be disabled for
6320 whatever reason, this option can be used. Note
6321 that this also can be controlled per-workqueue for
6322 workqueues visible under /sys/bus/workqueue/.
6324 workqueue.power_efficient
6325 Per-cpu workqueues are generally preferred because
6326 they show better performance thanks to cache
6327 locality; unfortunately, per-cpu workqueues tend to
6328 be more power hungry than unbound workqueues.
6330 Enabling this makes the per-cpu workqueues which
6331 were observed to contribute significantly to power
6332 consumption unbound, leading to measurably lower
6333 power usage at the cost of small performance
6336 The default value of this parameter is determined by
6337 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT.
6339 workqueue.debug_force_rr_cpu
6340 Workqueue used to implicitly guarantee that work
6341 items queued without explicit CPU specified are put
6342 on the local CPU. This guarantee is no longer true
6343 and while local CPU is still preferred work items
6344 may be put on foreign CPUs. This debug option
6345 forces round-robin CPU selection to flush out
6346 usages which depend on the now broken guarantee.
6347 When enabled, memory and cache locality will be
6350 x2apic_phys [X86-64,APIC] Use x2apic physical mode instead of
6351 default x2apic cluster mode on platforms
6354 xen_512gb_limit [KNL,X86-64,XEN]
6355 Restricts the kernel running paravirtualized under Xen
6356 to use only up to 512 GB of RAM. The reason to do so is
6357 crash analysis tools and Xen tools for doing domain
6358 save/restore/migration must be enabled to handle larger
6361 xen_emul_unplug= [HW,X86,XEN]
6362 Unplug Xen emulated devices
6363 Format: [unplug0,][unplug1]
6364 ide-disks -- unplug primary master IDE devices
6365 aux-ide-disks -- unplug non-primary-master IDE devices
6366 nics -- unplug network devices
6367 all -- unplug all emulated devices (NICs and IDE disks)
6368 unnecessary -- unplugging emulated devices is
6369 unnecessary even if the host did not respond to
6371 never -- do not unplug even if version check succeeds
6373 xen_legacy_crash [X86,XEN]
6374 Crash from Xen panic notifier, without executing late
6375 panic() code such as dumping handler.
6377 xen_nopvspin [X86,XEN]
6378 Disables the qspinlock slowpath using Xen PV optimizations.
6379 This parameter is obsoleted by "nopvspin" parameter, which
6380 has equivalent effect for XEN platform.
6383 Disables the PV optimizations forcing the HVM guest to
6384 run as generic HVM guest with no PV drivers.
6385 This option is obsoleted by the "nopv" option, which
6386 has equivalent effect for XEN platform.
6388 xen_no_vector_callback
6389 [KNL,X86,XEN] Disable the vector callback for Xen
6390 event channel interrupts.
6392 xen_scrub_pages= [XEN]
6393 Boolean option to control scrubbing pages before giving them back
6394 to Xen, for use by other domains. Can be also changed at runtime
6395 with /sys/devices/system/xen_memory/xen_memory0/scrub_pages.
6396 Default value controlled with CONFIG_XEN_SCRUB_PAGES_DEFAULT.
6398 xen_timer_slop= [X86-64,XEN]
6399 Set the timer slop (in nanoseconds) for the virtual Xen
6400 timers (default is 100000). This adjusts the minimum
6401 delta of virtualized Xen timers, where lower values
6402 improve timer resolution at the expense of processing
6403 more timer interrupts.
6405 xen.balloon_boot_timeout= [XEN]
6406 The time (in seconds) to wait before giving up to boot
6407 in case initial ballooning fails to free enough memory.
6408 Applies only when running as HVM or PVH guest and
6409 started with less memory configured than allowed at
6410 max. Default is 180.
6412 xen.event_eoi_delay= [XEN]
6413 How long to delay EOI handling in case of event
6414 storms (jiffies). Default is 10.
6416 xen.event_loop_timeout= [XEN]
6417 After which time (jiffies) the event handling loop
6418 should start to delay EOI handling. Default is 2.
6420 xen.fifo_events= [XEN]
6421 Boolean parameter to disable using fifo event handling
6422 even if available. Normally fifo event handling is
6423 preferred over the 2-level event handling, as it is
6424 fairer and the number of possible event channels is
6425 much higher. Default is on (use fifo events).
6427 nopv= [X86,XEN,KVM,HYPER_V,VMWARE]
6428 Disables the PV optimizations forcing the guest to run
6429 as generic guest with no PV drivers. Currently support
6430 XEN HVM, KVM, HYPER_V and VMWARE guest.
6432 nopvspin [X86,XEN,KVM]
6433 Disables the qspinlock slow path using PV optimizations
6434 which allow the hypervisor to 'idle' the guest on lock
6437 xirc2ps_cs= [NET,PCMCIA]
6439 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]]
6442 By default on POWER9 and above, the kernel will
6443 natively use the XIVE interrupt controller. This option
6444 allows the fallback firmware mode to be used:
6446 off Fallback to firmware control of XIVE interrupt
6447 controller on both pseries and powernv
6448 platforms. Only useful on POWER9 and above.
6450 xhci-hcd.quirks [USB,KNL]
6451 A hex value specifying bitmask with supplemental xhci
6452 host controller quirks. Meaning of each bit can be
6453 consulted in header drivers/usb/host/xhci.h.
6456 Format: { early | on | rw | ro | off }
6457 Controls if xmon debugger is enabled. Default is off.
6458 Passing only "xmon" is equivalent to "xmon=early".
6459 early Call xmon as early as possible on boot; xmon
6460 debugger is called from setup_arch().
6461 on xmon debugger hooks will be installed so xmon
6462 is only called on a kernel crash. Default mode,
6463 i.e. either "ro" or "rw" mode, is controlled
6464 with CONFIG_XMON_DEFAULT_RO_MODE.
6465 rw xmon debugger hooks will be installed so xmon
6466 is called only on a kernel crash, mode is write,
6467 meaning SPR registers, memory and, other data
6468 can be written using xmon commands.
6469 ro same as "rw" option above but SPR registers,
6470 memory, and other data can't be written using
6472 off xmon is disabled.