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
324 pgtbl_v1 - Use v1 page table for DMA-API (Default).
325 pgtbl_v2 - Use v2 page table for DMA-API.
327 amd_iommu_dump= [HW,X86-64]
328 Enable AMD IOMMU driver option to dump the ACPI table
329 for AMD IOMMU. With this option enabled, AMD IOMMU
330 driver will print ACPI tables for AMD IOMMU during
331 IOMMU initialization.
333 amd_iommu_intr= [HW,X86-64]
334 Specifies one of the following AMD IOMMU interrupt
336 legacy - Use legacy interrupt remapping mode.
337 vapic - Use virtual APIC mode, which allows IOMMU
338 to inject interrupts directly into guest.
339 This mode requires kvm-amd.avic=1.
340 (Default when IOMMU HW support is present.)
344 Do not enable amd_pstate as the default
345 scaling driver for the supported processors
347 Use amd_pstate with passive mode as a scaling driver.
348 In this mode autonomous selection is disabled.
349 Driver requests a desired performance level and platform
350 tries to match the same performance level if it is
351 satisfied by guaranteed performance level.
353 Use amd_pstate_epp driver instance as the scaling driver,
354 driver provides a hint to the hardware if software wants
355 to bias toward performance (0x0) or energy efficiency (0xff)
356 to the CPPC firmware. then CPPC power algorithm will
357 calculate the runtime workload and adjust the realtime cores
360 Activate guided autonomous mode. Driver requests minimum and
361 maximum performance level and the platform autonomously
362 selects a performance level in this range and appropriate
363 to the current workload.
365 amijoy.map= [HW,JOY] Amiga joystick support
366 Map of devices attached to JOY0DAT and JOY1DAT
368 See also Documentation/input/joydev/joystick.rst
370 analog.map= [HW,JOY] Analog joystick and gamepad support
371 Specifies type or capabilities of an analog joystick
372 connected to one of 16 gameports
373 Format: <type1>,<type2>,..<type16>
376 Power management functions (SPARCstation-4/5 + deriv.)
378 Disable APC CPU standby support. SPARCstation-Fox does
379 not play well with APC CPU idle - disable it if you have
380 APC and your system crashes randomly.
382 apic= [APIC,X86] Advanced Programmable Interrupt Controller
383 Change the output verbosity while booting
384 Format: { quiet (default) | verbose | debug }
385 Change the amount of debugging information output
386 when initialising the APIC and IO-APIC components.
387 For X86-32, this can also be used to specify an APIC
389 Format: apic=driver_name
390 Examples: apic=bigsmp
392 apic_extnmi= [APIC,X86] External NMI delivery setting
393 Format: { bsp (default) | all | none }
394 bsp: External NMI is delivered only to CPU 0
395 all: External NMIs are broadcast to all CPUs as a
397 none: External NMI is masked for all CPUs. This is
398 useful so that a dump capture kernel won't be
402 See Documentation/networking/ipv6.rst.
404 apm= [APM] Advanced Power Management
405 See header of arch/x86/kernel/apm_32.c.
407 apparmor= [APPARMOR] Disable or enable AppArmor at boot time
408 Format: { "0" | "1" }
409 See security/apparmor/Kconfig help text
412 Default value is set via kernel config option.
414 arcrimi= [HW,NET] ARCnet - "RIM I" (entirely mem-mapped) cards
415 Format: <io>,<irq>,<nodeID>
417 arm64.nobti [ARM64] Unconditionally disable Branch Target
418 Identification support
420 arm64.nopauth [ARM64] Unconditionally disable Pointer Authentication
423 arm64.nomte [ARM64] Unconditionally disable Memory Tagging Extension
426 arm64.nosve [ARM64] Unconditionally disable Scalable Vector
429 arm64.nosme [ARM64] Unconditionally disable Scalable Matrix
432 arm64.nomops [ARM64] Unconditionally disable Memory Copy and Memory
433 Set instructions support
437 atarimouse= [HW,MOUSE] Atari Mouse
439 atkbd.extra= [HW] Enable extra LEDs and keys on IBM RapidAccess,
440 EzKey and similar keyboards
442 atkbd.reset= [HW] Reset keyboard during initialization
444 atkbd.set= [HW] Select keyboard code set
445 Format: <int> (2 = AT (default), 3 = PS/2)
447 atkbd.scroll= [HW] Enable scroll wheel on MS Office and similar
450 atkbd.softraw= [HW] Choose between synthetic and real raw mode
451 Format: <bool> (0 = real, 1 = synthetic (default))
453 atkbd.softrepeat= [HW]
454 Use software keyboard repeat
456 audit= [KNL] Enable the audit sub-system
457 Format: { "0" | "1" | "off" | "on" }
458 0 | off - kernel audit is disabled and can not be
459 enabled until the next reboot
460 unset - kernel audit is initialized but disabled and
461 will be fully enabled by the userspace auditd.
462 1 | on - kernel audit is initialized and partially
463 enabled, storing at most audit_backlog_limit
464 messages in RAM until it is fully enabled by the
468 audit_backlog_limit= [KNL] Set the audit queue size limit.
469 Format: <int> (must be >=0)
472 bau= [X86_UV] Enable the BAU on SGI UV. The default
473 behavior is to disable the BAU (i.e. bau=0).
474 Format: { "0" | "1" }
477 unset - Disable the BAU.
479 baycom_epp= [HW,AX25]
482 baycom_par= [HW,AX25] BayCom Parallel Port AX.25 Modem
484 See header of drivers/net/hamradio/baycom_par.c.
486 baycom_ser_fdx= [HW,AX25]
487 BayCom Serial Port AX.25 Modem (Full Duplex Mode)
488 Format: <io>,<irq>,<mode>[,<baud>]
489 See header of drivers/net/hamradio/baycom_ser_fdx.c.
491 baycom_ser_hdx= [HW,AX25]
492 BayCom Serial Port AX.25 Modem (Half Duplex Mode)
493 Format: <io>,<irq>,<mode>
494 See header of drivers/net/hamradio/baycom_ser_hdx.c.
497 Disable BERT OS support on buggy BIOSes.
499 bgrt_disable [ACPI][X86]
500 Disable BGRT to avoid flickering OEM logo.
502 blkdevparts= Manual partition parsing of block device(s) for
503 embedded devices based on command line input.
504 See Documentation/block/cmdline-partition.rst
506 boot_delay= Milliseconds to delay each printk during boot.
507 Only works if CONFIG_BOOT_PRINTK_DELAY is enabled,
508 and you may also have to specify "lpj=". Boot_delay
509 values larger than 10 seconds (10000) are assumed
510 erroneous and ignored.
514 Extended command line options can be added to an initrd
515 and this will cause the kernel to look for it.
517 See Documentation/admin-guide/bootconfig.rst
519 bttv.card= [HW,V4L] bttv (bt848 + bt878 based grabber cards)
520 bttv.radio= Most important insmod options are available as
522 bttv.pll= See Documentation/admin-guide/media/bttv.rst
525 bulk_remove=off [PPC] This parameter disables the use of the pSeries
526 firmware feature for flushing multiple hpte entries
529 c101= [NET] Moxa C101 synchronous serial card
531 cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection.
532 Sometimes CPU hardware bugs make them report the cache
533 size incorrectly. The kernel will attempt work arounds
534 to fix known problems, but for some CPUs it is not
535 possible to determine what the correct size should be.
536 This option provides an override for these situations.
539 [NET] Specifies amount of time (in seconds) that
540 the kernel should wait for a network carrier. By default
541 it waits 120 seconds.
543 ca_keys= [KEYS] This parameter identifies a specific key(s) on
544 the system trusted keyring to be used for certificate
546 format: { id:<keyid> | builtin }
548 cca= [MIPS] Override the kernel pages' cache coherency
549 algorithm. Accepted values range from 0 to 7
550 inclusive. See arch/mips/include/asm/pgtable-bits.h
551 for platform specific values (SB1, Loongson3 and
554 ccw_timeout_log [S390]
555 See Documentation/s390/common_io.rst for details.
557 cgroup_disable= [KNL] Disable a particular controller or optional feature
558 Format: {name of the controller(s) or feature(s) to disable}
559 The effects of cgroup_disable=foo are:
560 - foo isn't auto-mounted if you mount all cgroups in
562 - foo isn't visible as an individually mountable
564 - if foo is an optional feature then the feature is
565 disabled and corresponding cgroup files are not
567 {Currently only "memory" controller deal with this and
568 cut the overhead, others just disable the usage. So
569 only cgroup_disable=memory is actually worthy}
570 Specifying "pressure" disables per-cgroup pressure
571 stall information accounting feature
573 cgroup_no_v1= [KNL] Disable cgroup controllers and named hierarchies in v1
574 Format: { { controller | "all" | "named" }
575 [,{ controller | "all" | "named" }...] }
576 Like cgroup_disable, but only applies to cgroup v1;
577 the blacklisted controllers remain available in cgroup2.
578 "all" blacklists all controllers and "named" disables
579 named mounts. Specifying both "all" and "named" disables
582 cgroup.memory= [KNL] Pass options to the cgroup memory controller.
584 nosocket -- Disable socket memory accounting.
585 nokmem -- Disable kernel memory accounting.
586 nobpf -- Disable BPF memory accounting.
588 checkreqprot= [SELINUX] Set initial checkreqprot flag value.
589 Format: { "0" | "1" }
590 See security/selinux/Kconfig help text.
591 0 -- check protection applied by kernel (includes
592 any implied execute protection).
593 1 -- check protection requested by application.
594 Default value is set via a kernel config option.
595 Value can be changed at runtime via
596 /sys/fs/selinux/checkreqprot.
597 Setting checkreqprot to 1 is deprecated.
600 See Documentation/s390/common_io.rst for details.
602 clearcpuid=X[,X...] [X86]
603 Disable CPUID feature X for the kernel. See
604 arch/x86/include/asm/cpufeatures.h for the valid bit
605 numbers X. Note the Linux-specific bits are not necessarily
606 stable over kernel options, but the vendor-specific
608 X can also be a string as appearing in the flags: line
609 in /proc/cpuinfo which does not have the above
610 instability issue. However, not all features have names
612 Note that using this option will taint your kernel.
613 Also note that user programs calling CPUID directly
614 or using the feature without checking anything
615 will still see it. This just prevents it from
616 being used by the kernel or shown in /proc/cpuinfo.
617 Also note the kernel might malfunction if you disable
622 Prevents the clock framework from automatically gating
623 clocks that have not been explicitly enabled by a Linux
624 device driver but are enabled in hardware at reset or
625 by the bootloader/firmware. Note that this does not
626 force such clocks to be always-on nor does it reserve
627 those clocks in any way. This parameter is useful for
628 debug and development, but should not be needed on a
629 platform with proper driver support. For more
630 information, see Documentation/driver-api/clk.rst.
632 clock= [BUGS=X86-32, HW] gettimeofday clocksource override.
634 Forces specified clocksource (if available) to be used
635 when calculating gettimeofday(). If specified
636 clocksource is not available, it defaults to PIT.
637 Format: { pit | tsc | cyclone | pmtmr }
639 clocksource= Override the default clocksource
641 Override the default clocksource and use the clocksource
642 with the name specified.
643 Some clocksource names to choose from, depending on
645 [all] jiffies (this is the base, fallback clocksource)
647 [ARM] imx_timer1,OSTS,netx_timer,mpu_timer2,
648 pxa_timer,timer3,32k_counter,timer0_1
649 [X86-32] pit,hpet,tsc;
650 scx200_hrt on Geode; cyclone on IBM x440
658 clocksource.arm_arch_timer.evtstrm=
661 Enable/disable the eventstream feature of the ARM
662 architected timer so that code using WFE-based polling
663 loops can be debugged more effectively on production
666 clocksource.max_cswd_read_retries= [KNL]
667 Number of clocksource_watchdog() retries due to
668 external delays before the clock will be marked
669 unstable. Defaults to two retries, that is,
670 three attempts to read the clock under test.
672 clocksource.verify_n_cpus= [KNL]
673 Limit the number of CPUs checked for clocksources
674 marked with CLOCK_SOURCE_VERIFY_PERCPU that
675 are marked unstable due to excessive skew.
676 A negative value says to check all CPUs, while
677 zero says not to check any. Values larger than
678 nr_cpu_ids are silently truncated to nr_cpu_ids.
679 The actual CPUs are chosen randomly, with
680 no replacement if the same CPU is chosen twice.
682 clocksource-wdtest.holdoff= [KNL]
683 Set the time in seconds that the clocksource
684 watchdog test waits before commencing its tests.
685 Defaults to zero when built as a module and to
686 10 seconds when built into the kernel.
688 cma=nn[MG]@[start[MG][-end[MG]]]
690 Sets the size of kernel global memory area for
691 contiguous memory allocations and optionally the
692 placement constraint by the physical address range of
693 memory allocations. A value of 0 disables CMA
694 altogether. For more information, see
695 kernel/dma/contiguous.c
699 Sets the size of kernel per-numa memory area for
700 contiguous memory allocations. A value of 0 disables
701 per-numa CMA altogether. And If this option is not
702 specified, the default value is 0.
703 With per-numa CMA enabled, DMA users on node nid will
704 first try to allocate buffer from the pernuma area
705 which is located in node nid, if the allocation fails,
706 they will fallback to the global default memory area.
708 cmo_free_hint= [PPC] Format: { yes | no }
709 Specify whether pages are marked as being inactive
710 when they are freed. This is used in CMO environments
711 to determine OS memory pressure for page stealing by
715 coherent_pool=nn[KMG] [ARM,KNL]
716 Sets the size of memory pool for coherent, atomic dma
717 allocations, by default set to 256K.
719 com20020= [HW,NET] ARCnet - COM20020 chipset
721 <io>[,<irq>[,<nodeID>[,<backplane>[,<ckp>[,<timeout>]]]]]
723 com90io= [HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers)
727 ARCnet - COM90xx chipset (memory-mapped buffers)
728 Format: <io>[,<irq>[,<memstart>]]
730 condev= [HW,S390] console device
733 con3215_drop= [S390] 3215 console drop mode.
735 When set to true, drop data on the 3215 console when
736 the console buffer is full. In this case the
737 operator using a 3270 terminal emulator (for example
738 x3270) does not have to enter the clear key for the
739 console output to advance and the kernel to continue.
740 This leads to a much faster boot time when a 3270
741 terminal emulator is active. If no 3270 terminal
742 emulator is used, this parameter has no effect.
744 console= [KNL] Output console device and options.
746 tty<n> Use the virtual console device <n>.
750 Use the specified serial port. The options are of
751 the form "bbbbpnf", where "bbbb" is the baud rate,
752 "p" is parity ("n", "o", or "e"), "n" is number of
753 bits, and "f" is flow control ("r" for RTS or
754 omit it). Default is "9600n8".
756 See Documentation/admin-guide/serial-console.rst for more
758 Documentation/networking/netconsole.rst for an
761 uart[8250],io,<addr>[,options]
762 uart[8250],mmio,<addr>[,options]
763 uart[8250],mmio16,<addr>[,options]
764 uart[8250],mmio32,<addr>[,options]
765 uart[8250],0x<addr>[,options]
766 Start an early, polled-mode console on the 8250/16550
767 UART at the specified I/O port or MMIO address,
768 switching to the matching ttyS device later.
769 MMIO inter-register address stride is either 8-bit
770 (mmio), 16-bit (mmio16), or 32-bit (mmio32).
771 If none of [io|mmio|mmio16|mmio32], <addr> is assumed
772 to be equivalent to 'mmio'. 'options' are specified in
773 the same format described for ttyS above; if unspecified,
774 the h/w is not re-initialized.
776 hvc<n> Use the hypervisor console device <n>. This is for
777 both Xen and PowerPC hypervisors.
780 Use to disable console output, i.e., to have kernel
781 console messages discarded.
782 This must be the only console= parameter used on the
785 If the device connected to the port is not a TTY but a braille
786 device, prepend "brl," before the device type, for instance
788 For now, only VisioBraille is supported.
791 [KNL] Change console messages format
793 By default we print messages on consoles in
794 "[time stamp] text\n" format (time stamp may not be
795 printed, depending on CONFIG_PRINTK_TIME or
796 `printk_time' param).
798 Switch to syslog format: "<%u>[time stamp] text\n"
799 IOW, each message will have a facility and loglevel
800 prefix. The format is similar to one used by syslog()
801 syscall, or to executing "dmesg -S --raw" or to reading
804 consoleblank= [KNL] The console blank (screen saver) timeout in
805 seconds. A value of 0 disables the blank timer.
809 [KNL] Change the default value for
810 /proc/<pid>/coredump_filter.
811 See also Documentation/filesystems/proc.rst.
813 coresight_cpu_debug.enable
816 Enable/disable the CPU sampling based debugging.
817 0: default value, disable debugging
818 1: enable debugging at boot time
820 cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver
822 <first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>]
824 cpuidle.off=1 [CPU_IDLE]
825 disable the cpuidle sub-system
828 [CPU_IDLE] Name of the cpuidle governor to use.
830 cpufreq.off=1 [CPU_FREQ]
831 disable the cpufreq sub-system
833 cpufreq.default_governor=
834 [CPU_FREQ] Name of the default cpufreq governor or
835 policy to use. This governor must be registered in the
836 kernel before the cpufreq driver probes.
839 [X86] Delay for N microsec between assert and de-assert
840 of APIC INIT to start processors. This delay occurs
841 on every CPU online, such as boot, and resume from suspend.
845 [SMP] Enable/disable parallel bringup of secondary CPUs
847 Default is enabled if CONFIG_HOTPLUG_PARALLEL=y. Otherwise
848 the parameter has no effect.
850 crash_kexec_post_notifiers
851 Run kdump after running panic-notifiers and dumping
852 kmsg. This only for the users who doubt kdump always
853 succeeds in any situation.
854 Note that this also increases risks of kdump failure,
855 because some panic notifiers can make the crashed
856 kernel more unstable.
858 crashkernel=size[KMG][@offset[KMG]]
859 [KNL] Using kexec, Linux can switch to a 'crash kernel'
860 upon panic. This parameter reserves the physical
861 memory region [offset, offset + size] for that kernel
862 image. If '@offset' is omitted, then a suitable offset
863 is selected automatically.
864 [KNL, X86-64, ARM64] Select a region under 4G first, and
865 fall back to reserve region above 4G when '@offset'
866 hasn't been specified.
867 See Documentation/admin-guide/kdump/kdump.rst for further details.
869 crashkernel=range1:size1[,range2:size2,...][@offset]
870 [KNL] Same as above, but depends on the memory
871 in the running system. The syntax of range is
872 start-[end] where start and end are both
873 a memory unit (amount[KMG]). See also
874 Documentation/admin-guide/kdump/kdump.rst for an example.
876 crashkernel=size[KMG],high
877 [KNL, X86-64, ARM64] range could be above 4G. Allow kernel
878 to allocate physical memory region from top, so could
879 be above 4G if system have more than 4G ram installed.
880 Otherwise memory region will be allocated below 4G, if
882 It will be ignored if crashkernel=X is specified.
883 crashkernel=size[KMG],low
884 [KNL, X86-64, ARM64] range under 4G. When crashkernel=X,high
885 is passed, kernel could allocate physical memory region
886 above 4G, that cause second kernel crash on system
887 that require some amount of low memory, e.g. swiotlb
888 requires at least 64M+32K low memory, also enough extra
889 low memory is needed to make sure DMA buffers for 32-bit
890 devices won't run out. Kernel would try to allocate
891 default size of memory below 4G automatically. The default
892 size is platform dependent.
893 --> x86: max(swiotlb_size_or_default() + 8MiB, 256MiB)
895 This one lets the user specify own low range under 4G
896 for second kernel instead.
897 0: to disable low allocation.
898 It will be ignored when crashkernel=X,high is not used
899 or memory reserved is below 4G.
902 [KNL] Disable crypto self-tests
907 cs89x0_media= [HW,NET]
908 Format: { rj45 | aui | bnc }
910 csdlock_debug= [KNL] Enable or disable debug add-ons of cross-CPU
911 function call handling. When switched on,
912 additional debug data is printed to the console
913 in case a hanging CPU is detected, and that
914 CPU is pinged again in order to try to resolve
915 the hang situation. The default value of this
916 option depends on the CSD_LOCK_WAIT_DEBUG_DEFAULT
920 See header of drivers/s390/block/dasd_devmap.c.
922 db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port
923 (one device per port)
924 Format: <port#>,<type>
925 See also Documentation/input/devices/joystick-parport.rst
927 debug [KNL] Enable kernel debugging (events log level).
930 [KNL] Enable printing [hashed] pointers early in the
931 boot sequence. If enabled, we use a weak hash instead
932 of siphash to hash pointers. Use this option if you are
933 seeing instances of '(___ptrval___)') and need to see a
934 value (hashed pointer) instead. Cryptographically
935 insecure, please do not use on production kernels.
938 [KNL] verbose locking self-tests
940 Print debugging info while doing the locking API
942 Bitmask for the various LOCKTYPE_ tests. Defaults to 0
943 (no extra messages), setting it to -1 (all bits set)
944 will print _a_lot_ more information - normally only
945 useful to lockdep developers.
947 debug_objects [KNL] Enable object debugging
949 debug_guardpage_minorder=
950 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this
951 parameter allows control of the order of pages that will
952 be intentionally kept free (and hence protected) by the
953 buddy allocator. Bigger value increase the probability
954 of catching random memory corruption, but reduce the
955 amount of memory for normal system use. The maximum
956 possible value is MAX_ORDER/2. Setting this parameter
957 to 1 or 2 should be enough to identify most random
958 memory corruption problems caused by bugs in kernel or
959 driver code when a CPU writes to (or reads from) a
960 random memory location. Note that there exists a class
961 of memory corruptions problems caused by buggy H/W or
962 F/W or by drivers badly programming DMA (basically when
963 memory is written at bus level and the CPU MMU is
964 bypassed) which are not detectable by
965 CONFIG_DEBUG_PAGEALLOC, hence this option will not help
966 tracking down these problems.
969 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this parameter
970 enables the feature at boot time. By default, it is
971 disabled and the system will work mostly the same as a
972 kernel built without CONFIG_DEBUG_PAGEALLOC.
973 Note: to get most of debug_pagealloc error reports, it's
974 useful to also enable the page_owner functionality.
975 on: enable the feature
977 debugfs= [KNL] This parameter enables what is exposed to userspace
978 and debugfs internal clients.
979 Format: { on, no-mount, off }
980 on: All functions are enabled.
982 Filesystem is not registered but kernel clients can
983 access APIs and a crashkernel can be used to read
984 its content. There is nothing to mount.
985 off: Filesystem is not registered and clients
986 get a -EPERM as result when trying to register files
987 or directories within debugfs.
988 This is equivalent of the runtime functionality if
989 debugfs was not enabled in the kernel at all.
990 Default value is set in build-time with a kernel configuration.
992 debugpat [X86] Enable PAT debugging
995 [HW] The size of the default HugeTLB page. This is
996 the size represented by the legacy /proc/ hugepages
997 APIs. In addition, this is the default hugetlb size
998 used for shmget(), mmap() and mounting hugetlbfs
999 filesystems. If not specified, defaults to the
1000 architecture's default huge page size. Huge page
1001 sizes are architecture dependent. See also
1002 Documentation/admin-guide/mm/hugetlbpage.rst.
1005 deferred_probe_timeout=
1006 [KNL] Debugging option to set a timeout in seconds for
1007 deferred probe to give up waiting on dependencies to
1008 probe. Only specific dependencies (subsystems or
1009 drivers) that have opted in will be ignored. A timeout
1010 of 0 will timeout at the end of initcalls. If the time
1011 out hasn't expired, it'll be restarted by each
1012 successful driver registration. This option will also
1013 dump out devices still on the deferred probe list after
1016 delayacct [KNL] Enable per-task delay accounting
1018 dell_smm_hwmon.ignore_dmi=
1019 [HW] Continue probing hardware even if DMI data
1020 indicates that the driver is running on unsupported
1023 dell_smm_hwmon.force=
1024 [HW] Activate driver even if SMM BIOS signature does
1025 not match list of supported models and enable otherwise
1026 blacklisted features.
1028 dell_smm_hwmon.power_status=
1029 [HW] Report power status in /proc/i8k
1030 (disabled by default).
1032 dell_smm_hwmon.restricted=
1033 [HW] Allow controlling fans only if SYS_ADMIN
1036 dell_smm_hwmon.fan_mult=
1037 [HW] Factor to multiply fan speed with.
1039 dell_smm_hwmon.fan_max=
1040 [HW] Maximum configurable fan speed.
1043 Format: { on | off | def_only | inf_only | always }
1044 on: s390 zlib hardware support for compression on
1045 level 1 and decompression (default)
1046 off: No s390 zlib hardware support
1047 def_only: s390 zlib hardware support for deflate
1048 only (compression on level 1)
1049 inf_only: s390 zlib hardware support for inflate
1050 only (decompression)
1051 always: Same as 'on' but ignores the selected compression
1052 level always using hardware support (used for debugging)
1054 dhash_entries= [KNL]
1055 Set number of hash buckets for dentry cache.
1057 disable_1tb_segments [PPC]
1058 Disables the use of 1TB hash page table segments. This
1059 causes the kernel to fall back to 256MB segments which
1060 can be useful when debugging issues that require an SLB
1064 See Documentation/networking/ipv6.rst.
1067 Disable RADIX MMU mode on POWER9
1070 Disable TLBIE instruction. Currently does not work
1071 with KVM, with HASH MMU, or with coherent accelerators.
1073 disable_cpu_apicid= [X86,APIC,SMP]
1075 The number of initial APIC ID for the
1076 corresponding CPU to be disabled at boot,
1077 mostly used for the kdump 2nd kernel to
1078 disable BSP to wake up multiple CPUs without
1079 causing system reset or hang due to sending
1080 INIT from AP to BSP.
1082 disable_ddw [PPC/PSERIES]
1083 Disable Dynamic DMA Window support. Use this
1084 to workaround buggy firmware.
1086 disable_ipv6= [IPV6]
1087 See Documentation/networking/ipv6.rst.
1089 disable_mtrr_cleanup [X86]
1090 The kernel tries to adjust MTRR layout from continuous
1091 to discrete, to make X server driver able to add WB
1092 entry later. This parameter disables that.
1094 disable_mtrr_trim [X86, Intel and AMD only]
1095 By default the kernel will trim any uncacheable
1096 memory out of your available memory pool based on
1097 MTRR settings. This parameter disables that behavior,
1098 possibly causing your machine to run very slowly.
1100 disable_timer_pin_1 [X86]
1101 Disable PIN 1 of APIC timer
1102 Can be useful to work around chipset bugs.
1104 dis_ucode_ldr [X86] Disable the microcode loader.
1106 dma_debug=off If the kernel is compiled with DMA_API_DEBUG support,
1107 this option disables the debugging code at boot.
1109 dma_debug_entries=<number>
1110 This option allows to tune the number of preallocated
1111 entries for DMA-API debugging code. One entry is
1112 required per DMA-API allocation. Use this if the
1113 DMA-API debugging code disables itself because the
1114 architectural default is too low.
1116 dma_debug_driver=<driver_name>
1117 With this option the DMA-API debugging driver
1118 filter feature can be enabled at boot time. Just
1119 pass the driver to filter for as the parameter.
1120 The filter can be disabled or changed to another
1121 driver later using sysfs.
1123 driver_async_probe= [KNL]
1124 List of driver names to be probed asynchronously. *
1125 matches with all driver names. If * is specified, the
1126 rest of the listed driver names are those that will NOT
1128 Format: <driver_name1>,<driver_name2>...
1130 drm.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>]
1131 Broken monitors, graphic adapters, KVMs and EDIDless
1132 panels may send no or incorrect EDID data sets.
1133 This parameter allows to specify an EDID data sets
1134 in the /lib/firmware directory that are used instead.
1135 Generic built-in EDID data sets are used, if one of
1136 edid/1024x768.bin, edid/1280x1024.bin,
1137 edid/1680x1050.bin, or edid/1920x1080.bin is given
1138 and no file with the same name exists. Details and
1139 instructions how to build your own EDID data are
1140 available in Documentation/admin-guide/edid.rst. An EDID
1141 data set will only be used for a particular connector,
1142 if its name and a colon are prepended to the EDID
1143 name. Each connector may use a unique EDID data
1144 set by separating the files with a comma. An EDID
1145 data set with no connector name will be used for
1146 any connectors not explicitly specified.
1151 Format: {"off" | "known"}
1152 Control how the dt_cpu_ftrs device-tree binding is
1153 used for CPU feature discovery and setup (if it
1155 off: Do not use it, fall back to legacy cpu table.
1156 known: Do not pass through unknown features to guests
1157 or userspace, only those that the kernel is aware of.
1159 dump_apple_properties [X86]
1160 Dump name and content of EFI device properties on
1161 x86 Macs. Useful for driver authors to determine
1162 what data is available or for reverse-engineering.
1164 dyndbg[="val"] [KNL,DYNAMIC_DEBUG]
1165 <module>.dyndbg[="val"]
1166 Enable debug messages at boot time. See
1167 Documentation/admin-guide/dynamic-debug-howto.rst
1170 early_ioremap_debug [KNL]
1171 Enable debug messages in early_ioremap support. This
1172 is useful for tracking down temporary early mappings
1173 which are not unmapped.
1175 earlycon= [KNL] Output early console device and options.
1177 When used with no options, the early console is
1178 determined by stdout-path property in device tree's
1179 chosen node or the ACPI SPCR table if supported by
1182 cdns,<addr>[,options]
1183 Start an early, polled-mode console on a Cadence
1184 (xuartps) serial port at the specified address. Only
1185 supported option is baud rate. If baud rate is not
1186 specified, the serial port must already be setup and
1189 uart[8250],io,<addr>[,options[,uartclk]]
1190 uart[8250],mmio,<addr>[,options[,uartclk]]
1191 uart[8250],mmio32,<addr>[,options[,uartclk]]
1192 uart[8250],mmio32be,<addr>[,options[,uartclk]]
1193 uart[8250],0x<addr>[,options]
1194 Start an early, polled-mode console on the 8250/16550
1195 UART at the specified I/O port or MMIO address.
1196 MMIO inter-register address stride is either 8-bit
1197 (mmio) or 32-bit (mmio32 or mmio32be).
1198 If none of [io|mmio|mmio32|mmio32be], <addr> is assumed
1199 to be equivalent to 'mmio'. 'options' are specified
1200 in the same format described for "console=ttyS<n>"; if
1201 unspecified, the h/w is not initialized. 'uartclk' is
1202 the uart clock frequency; if unspecified, it is set
1203 to 'BASE_BAUD' * 16.
1207 Start an early, polled-mode console on a pl011 serial
1208 port at the specified address. The pl011 serial port
1209 must already be setup and configured. Options are not
1210 yet supported. If 'mmio32' is specified, then only
1211 the driver will use only 32-bit accessors to read/write
1212 the device registers.
1215 Start an early console on a litex serial port at the
1216 specified address. The serial port must already be
1217 setup and configured. Options are not yet supported.
1220 Start an early, polled-mode console on a meson serial
1221 port at the specified address. The serial port must
1222 already be setup and configured. Options are not yet
1226 Start an early, polled-mode console on an msm serial
1227 port at the specified address. The serial port
1228 must already be setup and configured. Options are not
1231 msm_serial_dm,<addr>
1232 Start an early, polled-mode console on an msm serial
1233 dm port at the specified address. The serial port
1234 must already be setup and configured. Options are not
1238 Start an early, polled-mode console on a serial port
1239 of an Actions Semi SoC, such as S500 or S900, at the
1240 specified address. The serial port must already be
1241 setup and configured. Options are not yet supported.
1244 Start an early, polled-mode console on a serial port
1245 of an RDA Micro SoC, such as RDA8810PL, at the
1246 specified address. The serial port must already be
1247 setup and configured. Options are not yet supported.
1250 Use RISC-V SBI (Supervisor Binary Interface) for early
1253 smh Use ARM semihosting calls for early console.
1261 Use early console provided by serial driver available
1262 on Samsung SoCs, requires selecting proper type and
1263 a correct base address of the selected UART port. The
1264 serial port must already be setup and configured.
1265 Options are not yet supported.
1268 Start an early, polled-mode console on a lantiq serial
1269 (lqasc) port at the specified address. The serial port
1270 must already be setup and configured. Options are not
1275 Use early console provided by Freescale LP UART driver
1276 found on Freescale Vybrid and QorIQ LS1021A processors.
1277 A valid base address must be provided, and the serial
1278 port must already be setup and configured.
1282 Start an early, polled-mode, output-only console on the
1283 Freescale i.MX UART at the specified address. The UART
1284 must already be setup and configured.
1287 Start an early, polled-mode console on the
1288 Armada 3700 serial port at the specified
1289 address. The serial port must already be setup
1290 and configured. Options are not yet supported.
1293 Start an early, polled-mode console on a Qualcomm
1294 Generic Interface (GENI) based serial port at the
1295 specified address. The serial port must already be
1296 setup and configured. Options are not yet supported.
1299 Start an early, unaccelerated console on the EFI
1300 memory mapped framebuffer (if available). On cache
1301 coherent non-x86 systems that use system memory for
1302 the framebuffer, pass the 'ram' option so that it is
1303 mapped with the correct attributes.
1306 Use early console provided by Freescale LINFlexD UART
1307 serial driver for NXP S32V234 SoCs. A valid base
1308 address must be provided, and the serial port must
1309 already be setup and configured.
1311 earlyprintk= [X86,SH,ARM,M68k,S390]
1315 earlyprintk=serial[,ttySn[,baudrate]]
1316 earlyprintk=serial[,0x...[,baudrate]]
1317 earlyprintk=ttySn[,baudrate]
1318 earlyprintk=dbgp[debugController#]
1319 earlyprintk=pciserial[,force],bus:device.function[,baudrate]
1320 earlyprintk=xdbc[xhciController#]
1322 earlyprintk is useful when the kernel crashes before
1323 the normal console is initialized. It is not enabled by
1324 default because it has some cosmetic problems.
1326 Append ",keep" to not disable it when the real console
1329 Only one of vga, serial, or usb debug port can
1332 Currently only ttyS0 and ttyS1 may be specified by
1333 name. Other I/O ports may be explicitly specified
1334 on some architectures (x86 and arm at least) by
1335 replacing ttySn with an I/O port address, like this:
1336 earlyprintk=serial,0x1008,115200
1337 You can find the port for a given device in
1338 /proc/tty/driver/serial:
1339 2: uart:ST16650V2 port:00001008 irq:18 ...
1341 Interaction with the standard serial driver is not
1344 The VGA output is eventually overwritten by
1347 The xen option can only be used in Xen domains.
1349 The sclp output can only be used on s390.
1351 The optional "force" to "pciserial" enables use of a
1352 PCI device even when its classcode is not of the
1355 edac_report= [HW,EDAC] Control how to report EDAC event
1356 Format: {"on" | "off" | "force"}
1357 on: enable EDAC to report H/W event. May be overridden
1358 by other higher priority error reporting module.
1359 off: disable H/W event reporting through EDAC.
1360 force: enforce the use of EDAC to report H/W event.
1364 Format: {"off" | "on" | "skip[mbr]"}
1367 Format: { "debug", "disable_early_pci_dma",
1368 "nochunk", "noruntime", "nosoftreserve",
1369 "novamap", "no_disable_early_pci_dma" }
1370 debug: enable misc debug output.
1371 disable_early_pci_dma: disable the busmaster bit on all
1372 PCI bridges while in the EFI boot stub.
1373 nochunk: disable reading files in "chunks" in the EFI
1374 boot stub, as chunking can cause problems with some
1375 firmware implementations.
1376 noruntime : disable EFI runtime services support
1377 nosoftreserve: The EFI_MEMORY_SP (Specific Purpose)
1378 attribute may cause the kernel to reserve the
1379 memory range for a memory mapping driver to
1380 claim. Specify efi=nosoftreserve to disable this
1381 reservation and treat the memory by its base type
1382 (i.e. EFI_CONVENTIONAL_MEMORY / "System RAM").
1383 novamap: do not call SetVirtualAddressMap().
1384 no_disable_early_pci_dma: Leave the busmaster bit set
1385 on all PCI bridges while in the EFI boot stub
1387 efi_no_storage_paranoia [EFI; X86]
1388 Using this parameter you can use more than 50% of
1389 your efi variable storage. Use this parameter only if
1390 you are really sure that your UEFI does sane gc and
1391 fulfills the spec otherwise your board may brick.
1393 efi_fake_mem= nn[KMG]@ss[KMG]:aa[,nn[KMG]@ss[KMG]:aa,..] [EFI; X86]
1394 Add arbitrary attribute to specific memory range by
1395 updating original EFI memory map.
1396 Region of memory which aa attribute is added to is
1399 If efi_fake_mem=2G@4G:0x10000,2G@0x10a0000000:0x10000
1400 is specified, EFI_MEMORY_MORE_RELIABLE(0x10000)
1401 attribute is added to range 0x100000000-0x180000000 and
1402 0x10a0000000-0x1120000000.
1404 If efi_fake_mem=8G@9G:0x40000 is specified, the
1405 EFI_MEMORY_SP(0x40000) attribute is added to
1406 range 0x240000000-0x43fffffff.
1408 Using this parameter you can do debugging of EFI memmap
1409 related features. For example, you can do debugging of
1410 Address Range Mirroring feature even if your box
1411 doesn't support it, or mark specific memory as
1414 efivar_ssdt= [EFI; X86] Name of an EFI variable that contains an SSDT
1415 that is to be dynamically loaded by Linux. If there are
1416 multiple variables with the same name but with different
1417 vendor GUIDs, all of them will be loaded. See
1418 Documentation/admin-guide/acpi/ssdt-overlays.rst for details.
1421 eisa_irq_edge= [PARISC,HW]
1422 See header of drivers/parisc/eisa.c.
1424 ekgdboc= [X86,KGDB] Allow early kernel console debugging
1427 This is designed to be used in conjunction with
1428 the boot argument: earlyprintk=vga
1430 This parameter works in place of the kgdboc parameter
1431 but can only be used if the backing tty is available
1432 very early in the boot process. For early debugging
1433 via a serial port see kgdboc_earlycon instead.
1436 See comment before function elanfreq_setup() in
1437 arch/x86/kernel/cpu/cpufreq/elanfreq.c.
1439 elfcorehdr=[size[KMG]@]offset[KMG] [IA64,PPC,SH,X86,S390]
1440 Specifies physical address of start of kernel core
1441 image elf header and optionally the size. Generally
1442 kexec loader will pass this option to capture kernel.
1443 See Documentation/admin-guide/kdump/kdump.rst for details.
1445 enable_mtrr_cleanup [X86]
1446 The kernel tries to adjust MTRR layout from continuous
1447 to discrete, to make X server driver able to add WB
1448 entry later. This parameter enables that.
1450 enable_timer_pin_1 [X86]
1451 Enable PIN 1 of APIC timer
1452 Can be useful to work around chipset bugs
1453 (in particular on some ATI chipsets).
1454 The kernel tries to set a reasonable default.
1456 enforcing= [SELINUX] Set initial enforcing status.
1458 See security/selinux/Kconfig help text.
1459 0 -- permissive (log only, no denials).
1460 1 -- enforcing (deny and log).
1462 Value can be changed at runtime via
1463 /sys/fs/selinux/enforce.
1466 Disable Error Record Serialization Table (ERST)
1469 ether= [HW,NET] Ethernet cards parameters
1470 This option is obsoleted by the "netdev=" option, which
1471 has equivalent usage. See its documentation for details.
1475 Permit 'security.evm' to be updated regardless of
1476 current integrity status.
1478 early_page_ext [KNL] Enforces page_ext initialization to earlier
1479 stages so cover more early boot allocations.
1480 Please note that as side effect some optimizations
1481 might be disabled to achieve that (e.g. parallelized
1482 memory initialization is disabled) so the boot process
1483 might take longer, especially on systems with a lot of
1484 memory. Available with CONFIG_PAGE_EXTENSION=y.
1489 fail_make_request=[KNL]
1490 General fault injection mechanism.
1491 Format: <interval>,<probability>,<space>,<times>
1492 See also Documentation/fault-injection/.
1495 Format: { initns | none }
1496 See Documentation/admin-guide/sysctl/net.rst for
1497 fb_tunnels_only_for_init_ns
1500 See Documentation/admin-guide/blockdev/floppy.rst.
1502 force_pal_cache_flush
1503 [IA-64] Avoid check_sal_cache_flush which may hang on
1504 buggy SAL_CACHE_FLUSH implementations. Using this
1505 parameter will force ia64_sal_cache_flush to call
1506 ia64_pal_cache_flush instead of SAL_CACHE_FLUSH.
1509 Forcefully enable Physical Address Extension (PAE).
1510 Many Pentium M systems disable PAE but may have a
1511 functionally usable PAE implementation.
1512 Warning: use of this parameter will taint the kernel
1513 and may cause unknown problems.
1516 [FTRACE] will set and start the specified tracer
1517 as early as possible in order to facilitate early
1520 ftrace_boot_snapshot
1521 [FTRACE] On boot up, a snapshot will be taken of the
1522 ftrace ring buffer that can be read at:
1523 /sys/kernel/tracing/snapshot.
1524 This is useful if you need tracing information from kernel
1525 boot up that is likely to be overridden by user space
1526 start up functionality.
1528 Optionally, the snapshot can also be defined for a tracing
1529 instance that was created by the trace_instance= command
1532 trace_instance=foo,sched_switch ftrace_boot_snapshot=foo
1534 The above will cause the "foo" tracing instance to trigger
1535 a snapshot at the end of boot up.
1537 ftrace_dump_on_oops[=orig_cpu]
1538 [FTRACE] will dump the trace buffers on oops.
1539 If no parameter is passed, ftrace will dump
1540 buffers of all CPUs, but if you pass orig_cpu, it will
1541 dump only the buffer of the CPU that triggered the
1544 ftrace_filter=[function-list]
1545 [FTRACE] Limit the functions traced by the function
1546 tracer at boot up. function-list is a comma-separated
1547 list of functions. This list can be changed at run
1548 time by the set_ftrace_filter file in the debugfs
1551 ftrace_notrace=[function-list]
1552 [FTRACE] Do not trace the functions specified in
1553 function-list. This list can be changed at run time
1554 by the set_ftrace_notrace file in the debugfs
1557 ftrace_graph_filter=[function-list]
1558 [FTRACE] Limit the top level callers functions traced
1559 by the function graph tracer at boot up.
1560 function-list is a comma-separated list of functions
1561 that can be changed at run time by the
1562 set_graph_function file in the debugfs tracing directory.
1564 ftrace_graph_notrace=[function-list]
1565 [FTRACE] Do not trace from the functions specified in
1566 function-list. This list is a comma-separated list of
1567 functions that can be changed at run time by the
1568 set_graph_notrace file in the debugfs tracing directory.
1570 ftrace_graph_max_depth=<uint>
1571 [FTRACE] Used with the function graph tracer. This is
1572 the max depth it will trace into a function. This value
1573 can be changed at run time by the max_graph_depth file
1574 in the tracefs tracing directory. default: 0 (no limit)
1576 fw_devlink= [KNL] Create device links between consumer and supplier
1577 devices by scanning the firmware to infer the
1578 consumer/supplier relationships. This feature is
1579 especially useful when drivers are loaded as modules as
1580 it ensures proper ordering of tasks like device probing
1581 (suppliers first, then consumers), supplier boot state
1582 clean up (only after all consumers have probed),
1583 suspend/resume & runtime PM (consumers first, then
1585 Format: { off | permissive | on | rpm }
1586 off -- Don't create device links from firmware info.
1587 permissive -- Create device links from firmware info
1588 but use it only for ordering boot state clean
1589 up (sync_state() calls).
1590 on -- Create device links from firmware info and use it
1591 to enforce probe and suspend/resume ordering.
1592 rpm -- Like "on", but also use to order runtime PM.
1594 fw_devlink.strict=<bool>
1595 [KNL] Treat all inferred dependencies as mandatory
1596 dependencies. This only applies for fw_devlink=on|rpm.
1599 fw_devlink.sync_state =
1600 [KNL] When all devices that could probe have finished
1601 probing, this parameter controls what to do with
1602 devices that haven't yet received their sync_state()
1604 Format: { strict | timeout }
1605 strict -- Default. Continue waiting on consumers to
1607 timeout -- Give up waiting on consumers and call
1608 sync_state() on any devices that haven't yet
1609 received their sync_state() calls after
1610 deferred_probe_timeout has expired or by
1611 late_initcall() if !CONFIG_MODULES.
1614 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad
1615 support via parallel port (up to 5 devices per port)
1616 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5>
1617 See also Documentation/input/devices/joystick-parport.rst
1621 gart_fix_e820= [X86-64] disable the fix e820 for K8 GART
1625 gcov_persist= [GCOV] When non-zero (default), profiling data for
1626 kernel modules is saved and remains accessible via
1627 debugfs, even when the module is unloaded/reloaded.
1628 When zero, profiling data is discarded and associated
1629 debugfs files are removed at module unload time.
1631 goldfish [X86] Enable the goldfish android emulator platform.
1632 Don't use this when you are not running on the
1635 gpio-mockup.gpio_mockup_ranges
1636 [HW] Sets the ranges of gpiochip of for this device.
1637 Format: <start1>,<end1>,<start2>,<end2>...
1638 gpio-mockup.gpio_mockup_named_lines
1639 [HW] Let the driver know GPIO lines should be named.
1641 gpt [EFI] Forces disk with valid GPT signature but
1642 invalid Protective MBR to be treated as GPT. If the
1643 primary GPT is corrupted, it enables the backup/alternate
1644 GPT to be used instead.
1646 grcan.enable0= [HW] Configuration of physical interface 0. Determines
1647 the "Enable 0" bit of the configuration register.
1650 grcan.enable1= [HW] Configuration of physical interface 1. Determines
1651 the "Enable 0" bit of the configuration register.
1654 grcan.select= [HW] Select which physical interface to use.
1657 grcan.txsize= [HW] Sets the size of the tx buffer.
1658 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0.
1660 grcan.rxsize= [HW] Sets the size of the rx buffer.
1661 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0.
1665 [KNL] Under CONFIG_HARDENED_USERCOPY, whether
1666 hardening is enabled for this boot. Hardened
1667 usercopy checking is used to protect the kernel
1668 from reading or writing beyond known memory
1669 allocation boundaries as a proactive defense
1670 against bounds-checking flaws in the kernel's
1671 copy_to_user()/copy_from_user() interface.
1672 on Perform hardened usercopy checks (default).
1673 off Disable hardened usercopy checks.
1675 hardlockup_all_cpu_backtrace=
1676 [KNL] Should the hard-lockup detector generate
1677 backtraces on all cpus.
1680 hashdist= [KNL,NUMA] Large hashes allocated during boot
1681 are distributed across NUMA nodes. Defaults on
1682 for 64-bit NUMA, off otherwise.
1683 Format: 0 | 1 (for off | on)
1685 hcl= [IA-64] SGI's Hardware Graph compatibility layer
1687 hd= [EIDE] (E)IDE hard drive subsystem geometry
1688 Format: <cyl>,<head>,<sect>
1691 Disable Hardware Error Source Table (HEST) support;
1692 corresponding firmware-first mode error processing
1693 logic will be disabled.
1695 hibernate= [HIBERNATION]
1696 noresume Don't check if there's a hibernation image
1697 present during boot.
1698 nocompress Don't compress/decompress hibernation images.
1699 no Disable hibernation and resume.
1700 protect_image Turn on image protection during restoration
1701 (that will set all pages holding image data
1702 during restoration read-only).
1704 highmem=nn[KMG] [KNL,BOOT] forces the highmem zone to have an exact
1705 size of <nn>. This works even on boxes that have no
1706 highmem otherwise. This also works to reduce highmem
1707 size on bigger boxes.
1709 highres= [KNL] Enable/disable high resolution timer mode.
1710 Valid parameters: "on", "off"
1715 hostname= [KNL] Set the hostname (aka UTS nodename).
1717 This allows setting the system's hostname during early
1718 startup. This sets the name returned by gethostname.
1719 Using this parameter to set the hostname makes it
1720 possible to ensure the hostname is correctly set before
1721 any userspace processes run, avoiding the possibility
1722 that a process may call gethostname before the hostname
1723 has been explicitly set, resulting in the calling
1724 process getting an incorrect result. The string must
1725 not exceed the maximum allowed hostname length (usually
1726 64 characters) and will be truncated otherwise.
1728 hpet= [X86-32,HPET] option to control HPET usage
1729 Format: { enable (default) | disable | force |
1731 disable: disable HPET and use PIT instead
1732 force: allow force enabled of undocumented chips (ICH4,
1734 verbose: show contents of HPET registers during setup
1736 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET
1737 registers. Default set by CONFIG_HPET_MMAP_DEFAULT.
1739 hugepages= [HW] Number of HugeTLB pages to allocate at boot.
1740 If this follows hugepagesz (below), it specifies
1741 the number of pages of hugepagesz to be allocated.
1742 If this is the first HugeTLB parameter on the command
1743 line, it specifies the number of pages to allocate for
1744 the default huge page size. If using node format, the
1745 number of pages to allocate per-node can be specified.
1746 See also Documentation/admin-guide/mm/hugetlbpage.rst.
1747 Format: <integer> or (node format)
1748 <node>:<integer>[,<node>:<integer>]
1751 [HW] The size of the HugeTLB pages. This is used in
1752 conjunction with hugepages (above) to allocate huge
1753 pages of a specific size at boot. The pair
1754 hugepagesz=X hugepages=Y can be specified once for
1755 each supported huge page size. Huge page sizes are
1756 architecture dependent. See also
1757 Documentation/admin-guide/mm/hugetlbpage.rst.
1760 hugetlb_cma= [HW,CMA] The size of a CMA area used for allocation
1761 of gigantic hugepages. Or using node format, the size
1762 of a CMA area per node can be specified.
1763 Format: nn[KMGTPE] or (node format)
1764 <node>:nn[KMGTPE][,<node>:nn[KMGTPE]]
1766 Reserve a CMA area of given size and allocate gigantic
1767 hugepages using the CMA allocator. If enabled, the
1768 boot-time allocation of gigantic hugepages is skipped.
1770 hugetlb_free_vmemmap=
1771 [KNL] Requires CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP
1773 Control if HugeTLB Vmemmap Optimization (HVO) is enabled.
1774 Allows heavy hugetlb users to free up some more
1775 memory (7 * PAGE_SIZE for each 2MB hugetlb page).
1776 Format: { on | off (default) }
1781 Built with CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP_DEFAULT_ON=y,
1784 Note that the vmemmap pages may be allocated from the added
1785 memory block itself when memory_hotplug.memmap_on_memory is
1786 enabled, those vmemmap pages cannot be optimized even if this
1787 feature is enabled. Other vmemmap pages not allocated from
1788 the added memory block itself do not be affected.
1791 [KNL] Should the hung task detector generate panics.
1794 A value of 1 instructs the kernel to panic when a
1795 hung task is detected. The default value is controlled
1796 by the CONFIG_BOOTPARAM_HUNG_TASK_PANIC build-time
1797 option. The value selected by this boot parameter can
1798 be changed later by the kernel.hung_task_panic sysctl.
1800 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC)
1801 terminal devices. Valid values: 0..8
1802 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs.
1803 If specified, z/VM IUCV HVC accepts connections
1804 from listed z/VM user IDs only.
1806 hv_nopvspin [X86,HYPER_V] Disables the paravirt spinlock optimizations
1807 which allow the hypervisor to 'idle' the
1808 guest on lock contention.
1810 i2c_bus= [HW] Override the default board specific I2C bus speed
1811 or register an additional I2C bus that is not
1812 registered from board initialization code.
1816 i8042.debug [HW] Toggle i8042 debug mode
1817 i8042.unmask_kbd_data
1818 [HW] Enable printing of interrupt data from the KBD port
1819 (disabled by default, and as a pre-condition
1820 requires that i8042.debug=1 be enabled)
1821 i8042.direct [HW] Put keyboard port into non-translated mode
1822 i8042.dumbkbd [HW] Pretend that controller can only read data from
1823 keyboard and cannot control its state
1824 (Don't attempt to blink the leds)
1825 i8042.noaux [HW] Don't check for auxiliary (== mouse) port
1826 i8042.nokbd [HW] Don't check/create keyboard port
1827 i8042.noloop [HW] Disable the AUX Loopback command while probing
1829 i8042.nomux [HW] Don't check presence of an active multiplexing
1831 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX
1833 i8042.notimeout [HW] Ignore timeout condition signalled by controller
1834 i8042.reset [HW] Reset the controller during init, cleanup and
1835 suspend-to-ram transitions, only during s2r
1836 transitions, or never reset
1837 Format: { 1 | Y | y | 0 | N | n }
1838 1, Y, y: always reset controller
1839 0, N, n: don't ever reset controller
1840 Default: only on s2r transitions on x86; most other
1841 architectures force reset to be always executed
1842 i8042.unlock [HW] Unlock (ignore) the keylock
1843 i8042.kbdreset [HW] Reset device connected to KBD port
1845 [HW] Allow deferred probing upon i8042 probe errors
1849 i915.invert_brightness=
1850 [DRM] Invert the sense of the variable that is used to
1851 set the brightness of the panel backlight. Normally a
1852 brightness value of 0 indicates backlight switched off,
1853 and the maximum of the brightness value sets the backlight
1854 to maximum brightness. If this parameter is set to 0
1855 (default) and the machine requires it, or this parameter
1856 is set to 1, a brightness value of 0 sets the backlight
1857 to maximum brightness, and the maximum of the brightness
1858 value switches the backlight off.
1859 -1 -- never invert brightness
1860 0 -- machine default
1861 1 -- force brightness inversion
1864 Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]]
1868 Format: idle=poll, idle=halt, idle=nomwait
1869 Poll forces a polling idle loop that can slightly
1870 improve the performance of waking up a idle CPU, but
1871 will use a lot of power and make the system run hot.
1873 idle=halt: Halt is forced to be used for CPU idle.
1874 In such case C2/C3 won't be used again.
1875 idle=nomwait: Disable mwait for CPU C-states
1879 Allow force disabling of Shared Virtual Memory (SVA)
1880 support for the idxd driver. By default it is set to
1883 idxd.tc_override= [HW]
1885 Allow override of default traffic class configuration
1886 for the device. By default it is set to false (0).
1888 ieee754= [MIPS] Select IEEE Std 754 conformance mode
1889 Format: { strict | legacy | 2008 | relaxed }
1892 Choose which programs will be accepted for execution
1893 based on the IEEE 754 NaN encoding(s) supported by
1894 the FPU and the NaN encoding requested with the value
1895 of an ELF file header flag individually set by each
1896 binary. Hardware implementations are permitted to
1897 support either or both of the legacy and the 2008 NaN
1900 Available settings are as follows:
1901 strict accept binaries that request a NaN encoding
1902 supported by the FPU
1903 legacy only accept legacy-NaN binaries, if supported
1905 2008 only accept 2008-NaN binaries, if supported
1907 relaxed accept any binaries regardless of whether
1908 supported by the FPU
1910 The FPU emulator is always able to support both NaN
1911 encodings, so if no FPU hardware is present or it has
1912 been disabled with 'nofpu', then the settings of
1913 'legacy' and '2008' strap the emulator accordingly,
1914 'relaxed' straps the emulator for both legacy-NaN and
1915 2008-NaN, whereas 'strict' enables legacy-NaN only on
1916 legacy processors and both NaN encodings on MIPS32 or
1919 The setting for ABS.fmt/NEG.fmt instruction execution
1920 mode generally follows that for the NaN encoding,
1921 except where unsupported by hardware.
1923 ignore_loglevel [KNL]
1924 Ignore loglevel setting - this will print /all/
1925 kernel messages to the console. Useful for debugging.
1926 We also add it as printk module parameter, so users
1927 could change it dynamically, usually by
1928 /sys/module/printk/parameters/ignore_loglevel.
1931 Ignore RLIMIT_DATA setting for data mappings,
1932 print warning at first misuse. Can be changed via
1933 /sys/module/kernel/parameters/ignore_rlimit_data.
1935 ihash_entries= [KNL]
1936 Set number of hash buckets for inode cache.
1938 ima_appraise= [IMA] appraise integrity measurements
1939 Format: { "off" | "enforce" | "fix" | "log" }
1942 ima_appraise_tcb [IMA] Deprecated. Use ima_policy= instead.
1943 The builtin appraise policy appraises all files
1946 ima_canonical_fmt [IMA]
1947 Use the canonical format for the binary runtime
1948 measurements, instead of host native format.
1951 Format: { md5 | sha1 | rmd160 | sha256 | sha384
1955 The list of supported hash algorithms is defined
1956 in crypto/hash_info.h.
1959 The builtin policies to load during IMA setup.
1960 Format: "tcb | appraise_tcb | secure_boot |
1961 fail_securely | critical_data"
1963 The "tcb" policy measures all programs exec'd, files
1964 mmap'd for exec, and all files opened with the read
1965 mode bit set by either the effective uid (euid=0) or
1968 The "appraise_tcb" policy appraises the integrity of
1969 all files owned by root.
1971 The "secure_boot" policy appraises the integrity
1972 of files (eg. kexec kernel image, kernel modules,
1973 firmware, policy, etc) based on file signatures.
1975 The "fail_securely" policy forces file signature
1976 verification failure also on privileged mounted
1977 filesystems with the SB_I_UNVERIFIABLE_SIGNATURE
1980 The "critical_data" policy measures kernel integrity
1983 ima_tcb [IMA] Deprecated. Use ima_policy= instead.
1984 Load a policy which meets the needs of the Trusted
1985 Computing Base. This means IMA will measure all
1986 programs exec'd, files mmap'd for exec, and all files
1987 opened for read by uid=0.
1990 Select one of defined IMA measurements template formats.
1991 Formats: { "ima" | "ima-ng" | "ima-ngv2" | "ima-sig" |
1996 [IMA] Define a custom template format.
1997 Format: { "field1|...|fieldN" }
1999 ima.ahash_minsize= [IMA] Minimum file size for asynchronous hash usage
2000 Format: <min_file_size>
2001 Set the minimal file size for using asynchronous hash.
2002 If left unspecified, ahash usage is disabled.
2004 ahash performance varies for different data sizes on
2005 different crypto accelerators. This option can be used
2006 to achieve the best performance for a particular HW.
2008 ima.ahash_bufsize= [IMA] Asynchronous hash buffer size
2010 Set hashing buffer size. Default: 4k.
2012 ahash performance varies for different chunk sizes on
2013 different crypto accelerators. This option can be used
2014 to achieve best performance for particular HW.
2018 Run specified binary instead of /sbin/init as init
2021 initcall_debug [KNL] Trace initcalls as they are executed. Useful
2022 for working out where the kernel is dying during
2025 initcall_blacklist= [KNL] Do not execute a comma-separated list of
2026 initcall functions. Useful for debugging built-in
2027 modules and initcalls.
2029 initramfs_async= [KNL]
2032 This parameter controls whether the initramfs
2033 image is unpacked asynchronously, concurrently
2034 with devices being probed and
2035 initialized. This should normally just work,
2036 but as a debugging aid, one can get the
2037 historical behaviour of the initramfs
2038 unpacking being completed before device_ and
2041 initrd= [BOOT] Specify the location of the initial ramdisk
2043 initrdmem= [KNL] Specify a physical address and size from which to
2044 load the initrd. If an initrd is compiled in or
2045 specified in the bootparams, it takes priority over this
2047 Format: ss[KMG],nn[KMG]
2050 init_on_alloc= [MM] Fill newly allocated pages and heap objects with
2053 Default set by CONFIG_INIT_ON_ALLOC_DEFAULT_ON.
2055 init_on_free= [MM] Fill freed pages and heap objects with zeroes.
2057 Default set by CONFIG_INIT_ON_FREE_DEFAULT_ON.
2059 init_pkru= [X86] Specify the default memory protection keys rights
2060 register contents for all processes. 0x55555554 by
2061 default (disallow access to all but pkey 0). Can
2062 override in debugfs after boot.
2064 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver
2067 int_pln_enable [X86] Enable power limit notification interrupt
2069 integrity_audit=[IMA]
2070 Format: { "0" | "1" }
2071 0 -- basic integrity auditing messages. (Default)
2072 1 -- additional integrity auditing messages.
2074 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option
2076 Enable intel iommu driver.
2078 Disable intel iommu driver.
2079 igfx_off [Default Off]
2080 By default, gfx is mapped as normal device. If a gfx
2081 device has a dedicated DMAR unit, the DMAR unit is
2082 bypassed by not enabling DMAR with this option. In
2083 this case, gfx device will use physical address for
2085 strict [Default Off]
2086 Deprecated, equivalent to iommu.strict=1.
2087 sp_off [Default Off]
2088 By default, super page will be supported if Intel IOMMU
2089 has the capability. With this option, super page will
2092 Enable the Intel IOMMU scalable mode if the hardware
2093 advertises that it has support for the scalable mode
2096 Disallow use of the Intel IOMMU scalable mode.
2097 tboot_noforce [Default Off]
2098 Do not force the Intel IOMMU enabled under tboot.
2099 By default, tboot will force Intel IOMMU on, which
2100 could harm performance of some high-throughput
2101 devices like 40GBit network cards, even if identity
2103 Note that using this option lowers the security
2104 provided by tboot because it makes the system
2105 vulnerable to DMA attacks.
2107 intel_idle.max_cstate= [KNL,HW,ACPI,X86]
2108 0 disables intel_idle and fall back on acpi_idle.
2109 1 to 9 specify maximum depth of C-state.
2113 Do not enable intel_pstate as the default
2114 scaling driver for the supported processors
2116 Use intel_pstate driver to bypass the scaling
2117 governors layer of cpufreq and provides it own
2118 algorithms for p-state selection. There are two
2119 P-state selection algorithms provided by
2120 intel_pstate in the active mode: powersave and
2121 performance. The way they both operate depends
2122 on whether or not the hardware managed P-states
2123 (HWP) feature has been enabled in the processor
2124 and possibly on the processor model.
2126 Use intel_pstate as a scaling driver, but configure it
2127 to work with generic cpufreq governors (instead of
2128 enabling its internal governor). This mode cannot be
2129 used along with the hardware-managed P-states (HWP)
2132 Enable intel_pstate on systems that prohibit it by default
2133 in favor of acpi-cpufreq. Forcing the intel_pstate driver
2134 instead of acpi-cpufreq may disable platform features, such
2135 as thermal controls and power capping, that rely on ACPI
2136 P-States information being indicated to OSPM and therefore
2137 should be used with caution. This option does not work with
2138 processors that aren't supported by the intel_pstate driver
2139 or on platforms that use pcc-cpufreq instead of acpi-cpufreq.
2141 Do not enable hardware P state control (HWP)
2144 Only load intel_pstate on systems which support
2145 hardware P state control (HWP) if available.
2147 Enforce ACPI _PPC performance limits. If the Fixed ACPI
2148 Description Table, specifies preferred power management
2149 profile as "Enterprise Server" or "Performance Server",
2150 then this feature is turned on by default.
2152 Allow per-logical-CPU P-State performance control limits using
2153 cpufreq sysfs interface
2155 intremap= [X86-64, Intel-IOMMU]
2156 on enable Interrupt Remapping (default)
2157 off disable Interrupt Remapping
2158 nosid disable Source ID checking
2160 BIOS x2APIC opt-out request will be ignored
2161 nopost disable Interrupt Posting
2163 iomem= Disable strict checking of access to MMIO memory
2164 strict regions from userspace.
2179 nobypass [PPC/POWERNV]
2180 Disable IOMMU bypass, using IOMMU for PCI devices.
2182 iommu.forcedac= [ARM64, X86] Control IOVA allocation for PCI devices.
2183 Format: { "0" | "1" }
2184 0 - Try to allocate a 32-bit DMA address first, before
2185 falling back to the full range if needed.
2186 1 - Allocate directly from the full usable range,
2187 forcing Dual Address Cycle for PCI cards supporting
2188 greater than 32-bit addressing.
2190 iommu.strict= [ARM64, X86] Configure TLB invalidation behaviour
2191 Format: { "0" | "1" }
2193 Request that DMA unmap operations use deferred
2194 invalidation of hardware TLBs, for increased
2195 throughput at the cost of reduced device isolation.
2196 Will fall back to strict mode if not supported by
2197 the relevant IOMMU driver.
2199 DMA unmap operations invalidate IOMMU hardware TLBs
2201 unset - Use value of CONFIG_IOMMU_DEFAULT_DMA_{LAZY,STRICT}.
2202 Note: on x86, strict mode specified via one of the
2203 legacy driver-specific options takes precedence.
2206 [ARM64, X86] Configure DMA to bypass the IOMMU by default.
2207 Format: { "0" | "1" }
2208 0 - Use IOMMU translation for DMA.
2209 1 - Bypass the IOMMU for DMA.
2210 unset - Use value of CONFIG_IOMMU_DEFAULT_PASSTHROUGH.
2212 io7= [HW] IO7 for Marvel-based Alpha systems
2213 See comment before marvel_specify_io7 in
2214 arch/alpha/kernel/core_marvel.c.
2216 io_delay= [X86] I/O delay method
2218 Standard port 0x80 based delay
2220 Alternate port 0xed based delay (needed on some systems)
2222 Simple two microseconds delay
2227 See Documentation/admin-guide/nfs/nfsroot.rst.
2229 ipcmni_extend [KNL] Extend the maximum number of unique System V
2230 IPC identifiers from 32,768 to 16,777,216.
2232 irqaffinity= [SMP] Set the default irq affinity mask
2233 The argument is a cpu list, as described above.
2235 irqchip.gicv2_force_probe=
2238 Force the kernel to look for the second 4kB page
2239 of a GICv2 controller even if the memory range
2240 exposed by the device tree is too small.
2242 irqchip.gicv3_nolpi=
2244 Force the kernel to ignore the availability of
2245 LPIs (and by consequence ITSs). Intended for system
2246 that use the kernel as a bootloader, and thus want
2247 to let secondary kernels in charge of setting up
2250 irqchip.gicv3_pseudo_nmi= [ARM64]
2251 Enables support for pseudo-NMIs in the kernel. This
2252 requires the kernel to be built with
2253 CONFIG_ARM64_PSEUDO_NMI.
2256 When an interrupt is not handled search all handlers
2257 for it. Intended to get systems with badly broken
2261 When an interrupt is not handled search all handlers
2262 for it. Also check all handlers each timer
2263 interrupt. Intended to get systems with badly broken
2267 Format: <RDP>,<reset>,<pci_scan>,<verbosity>
2269 isolcpus= [KNL,SMP,ISOL] Isolate a given set of CPUs from disturbance.
2270 [Deprecated - use cpusets instead]
2271 Format: [flag-list,]<cpu-list>
2273 Specify one or more CPUs to isolate from disturbances
2274 specified in the flag list (default: domain):
2277 Disable the tick when a single task runs.
2279 A residual 1Hz tick is offloaded to workqueues, which you
2280 need to affine to housekeeping through the global
2281 workqueue's affinity configured via the
2282 /sys/devices/virtual/workqueue/cpumask sysfs file, or
2283 by using the 'domain' flag described below.
2285 NOTE: by default the global workqueue runs on all CPUs,
2286 so to protect individual CPUs the 'cpumask' file has to
2287 be configured manually after bootup.
2290 Isolate from the general SMP balancing and scheduling
2291 algorithms. Note that performing domain isolation this way
2292 is irreversible: it's not possible to bring back a CPU to
2293 the domains once isolated through isolcpus. It's strongly
2294 advised to use cpusets instead to disable scheduler load
2295 balancing through the "cpuset.sched_load_balance" file.
2296 It offers a much more flexible interface where CPUs can
2297 move in and out of an isolated set anytime.
2299 You can move a process onto or off an "isolated" CPU via
2300 the CPU affinity syscalls or cpuset.
2301 <cpu number> begins at 0 and the maximum value is
2302 "number of CPUs in system - 1".
2306 Isolate from being targeted by managed interrupts
2307 which have an interrupt mask containing isolated
2308 CPUs. The affinity of managed interrupts is
2309 handled by the kernel and cannot be changed via
2310 the /proc/irq/* interfaces.
2312 This isolation is best effort and only effective
2313 if the automatically assigned interrupt mask of a
2314 device queue contains isolated and housekeeping
2315 CPUs. If housekeeping CPUs are online then such
2316 interrupts are directed to the housekeeping CPU
2317 so that IO submitted on the housekeeping CPU
2318 cannot disturb the isolated CPU.
2320 If a queue's affinity mask contains only isolated
2321 CPUs then this parameter has no effect on the
2322 interrupt routing decision, though interrupts are
2323 only delivered when tasks running on those
2324 isolated CPUs submit IO. IO submitted on
2325 housekeeping CPUs has no influence on those
2328 The format of <cpu-list> is described above.
2332 ivrs_ioapic [HW,X86-64]
2333 Provide an override to the IOAPIC-ID<->DEVICE-ID
2334 mapping provided in the IVRS ACPI table.
2335 By default, PCI segment is 0, and can be omitted.
2337 For example, to map IOAPIC-ID decimal 10 to
2338 PCI segment 0x1 and PCI device 00:14.0,
2339 write the parameter as:
2340 ivrs_ioapic=10@0001:00:14.0
2343 * To map IOAPIC-ID decimal 10 to PCI device 00:14.0
2344 write the parameter as:
2345 ivrs_ioapic[10]=00:14.0
2346 * To map IOAPIC-ID decimal 10 to PCI segment 0x1 and
2347 PCI device 00:14.0 write the parameter as:
2348 ivrs_ioapic[10]=0001:00:14.0
2350 ivrs_hpet [HW,X86-64]
2351 Provide an override to the HPET-ID<->DEVICE-ID
2352 mapping provided in the IVRS ACPI table.
2353 By default, PCI segment is 0, and can be omitted.
2355 For example, to map HPET-ID decimal 10 to
2356 PCI segment 0x1 and PCI device 00:14.0,
2357 write the parameter as:
2358 ivrs_hpet=10@0001:00:14.0
2361 * To map HPET-ID decimal 0 to PCI device 00:14.0
2362 write the parameter as:
2363 ivrs_hpet[0]=00:14.0
2364 * To map HPET-ID decimal 10 to PCI segment 0x1 and
2365 PCI device 00:14.0 write the parameter as:
2366 ivrs_ioapic[10]=0001:00:14.0
2368 ivrs_acpihid [HW,X86-64]
2369 Provide an override to the ACPI-HID:UID<->DEVICE-ID
2370 mapping provided in the IVRS ACPI table.
2371 By default, PCI segment is 0, and can be omitted.
2373 For example, to map UART-HID:UID AMD0020:0 to
2374 PCI segment 0x1 and PCI device ID 00:14.5,
2375 write the parameter as:
2376 ivrs_acpihid=AMD0020:0@0001:00:14.5
2379 * To map UART-HID:UID AMD0020:0 to PCI segment is 0,
2380 PCI device ID 00:14.5, write the parameter as:
2381 ivrs_acpihid[00:14.5]=AMD0020:0
2382 * To map UART-HID:UID AMD0020:0 to PCI segment 0x1 and
2383 PCI device ID 00:14.5, write the parameter as:
2384 ivrs_acpihid[0001:00:14.5]=AMD0020:0
2386 js= [HW,JOY] Analog joystick
2387 See Documentation/input/joydev/joystick.rst.
2390 [KNL] Enforce KASAN (Kernel Address Sanitizer) to print
2391 report on every invalid memory access. Without this
2392 parameter KASAN will print report only for the first
2396 Do not unregister boot console at start. This is only
2397 useful for debugging when something happens in the window
2398 between unregistering the boot console and initializing
2403 kernelcore= [KNL,X86,IA-64,PPC]
2404 Format: nn[KMGTPE] | nn% | "mirror"
2405 This parameter specifies the amount of memory usable by
2406 the kernel for non-movable allocations. The requested
2407 amount is spread evenly throughout all nodes in the
2408 system as ZONE_NORMAL. The remaining memory is used for
2409 movable memory in its own zone, ZONE_MOVABLE. In the
2410 event, a node is too small to have both ZONE_NORMAL and
2411 ZONE_MOVABLE, kernelcore memory will take priority and
2412 other nodes will have a larger ZONE_MOVABLE.
2414 ZONE_MOVABLE is used for the allocation of pages that
2415 may be reclaimed or moved by the page migration
2416 subsystem. Note that allocations like PTEs-from-HighMem
2417 still use the HighMem zone if it exists, and the Normal
2418 zone if it does not.
2420 It is possible to specify the exact amount of memory in
2421 the form of "nn[KMGTPE]", a percentage of total system
2422 memory in the form of "nn%", or "mirror". If "mirror"
2423 option is specified, mirrored (reliable) memory is used
2424 for non-movable allocations and remaining memory is used
2425 for Movable pages. "nn[KMGTPE]", "nn%", and "mirror"
2426 are exclusive, so you cannot specify multiple forms.
2428 kgdbdbgp= [KGDB,HW] kgdb over EHCI usb debug port.
2429 Format: <Controller#>[,poll interval]
2430 The controller # is the number of the ehci usb debug
2431 port as it is probed via PCI. The poll interval is
2432 optional and is the number seconds in between
2433 each poll cycle to the debug port in case you need
2434 the functionality for interrupting the kernel with
2435 gdb or control-c on the dbgp connection. When
2436 not using this parameter you use sysrq-g to break into
2437 the kernel debugger.
2439 kgdboc= [KGDB,HW] kgdb over consoles.
2440 Requires a tty driver that supports console polling,
2441 or a supported polling keyboard driver (non-usb).
2442 Serial only format: <serial_device>[,baud]
2443 keyboard only format: kbd
2444 keyboard and serial format: kbd,<serial_device>[,baud]
2445 Optional Kernel mode setting:
2446 kms, kbd format: kms,kbd
2447 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud]
2449 kgdboc_earlycon= [KGDB,HW]
2450 If the boot console provides the ability to read
2451 characters and can work in polling mode, you can use
2452 this parameter to tell kgdb to use it as a backend
2453 until the normal console is registered. Intended to
2454 be used together with the kgdboc parameter which
2455 specifies the normal console to transition to.
2457 The name of the early console should be specified
2458 as the value of this parameter. Note that the name of
2459 the early console might be different than the tty
2460 name passed to kgdboc. It's OK to leave the value
2461 blank and the first boot console that implements
2462 read() will be picked.
2464 kgdbwait [KGDB] Stop kernel execution and enter the
2465 kernel debugger at the earliest opportunity.
2467 kmac= [MIPS] Korina ethernet MAC address.
2468 Configure the RouterBoard 532 series on-chip
2469 Ethernet adapter MAC address.
2471 kmemleak= [KNL] Boot-time kmemleak enable/disable
2472 Valid arguments: on, off
2474 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y,
2477 kprobe_event=[probe-list]
2478 [FTRACE] Add kprobe events and enable at boot time.
2479 The probe-list is a semicolon delimited list of probe
2480 definitions. Each definition is same as kprobe_events
2481 interface, but the parameters are comma delimited.
2482 For example, to add a kprobe event on vfs_read with
2483 arg1 and arg2, add to the command line;
2485 kprobe_event=p,vfs_read,$arg1,$arg2
2487 See also Documentation/trace/kprobetrace.rst "Kernel
2488 Boot Parameter" section.
2490 kpti= [ARM64] Control page table isolation of user
2491 and kernel address spaces.
2492 Default: enabled on cores which need mitigation.
2496 kunit.enable= [KUNIT] Enable executing KUnit tests. Requires
2497 CONFIG_KUNIT to be set to be fully enabled. The
2498 default value can be overridden via
2499 KUNIT_DEFAULT_ENABLED.
2500 Default is 1 (enabled)
2502 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs.
2503 Default is 0 (don't ignore, but inject #GP)
2505 kvm.eager_page_split=
2506 [KVM,X86] Controls whether or not KVM will try to
2507 proactively split all huge pages during dirty logging.
2508 Eager page splitting reduces interruptions to vCPU
2509 execution by eliminating the write-protection faults
2510 and MMU lock contention that would otherwise be
2511 required to split huge pages lazily.
2513 VM workloads that rarely perform writes or that write
2514 only to a small region of VM memory may benefit from
2515 disabling eager page splitting to allow huge pages to
2516 still be used for reads.
2518 The behavior of eager page splitting depends on whether
2519 KVM_DIRTY_LOG_INITIALLY_SET is enabled or disabled. If
2520 disabled, all huge pages in a memslot will be eagerly
2521 split when dirty logging is enabled on that memslot. If
2522 enabled, eager page splitting will be performed during
2523 the KVM_CLEAR_DIRTY ioctl, and only for the pages being
2526 Eager page splitting is only supported when kvm.tdp_mmu=Y.
2530 kvm.enable_vmware_backdoor=[KVM] Support VMware backdoor PV interface.
2531 Default is false (don't support).
2534 [KVM] Controls the software workaround for the
2535 X86_BUG_ITLB_MULTIHIT bug.
2536 force : Always deploy workaround.
2537 off : Never deploy workaround.
2538 auto : Deploy workaround based on the presence of
2539 X86_BUG_ITLB_MULTIHIT.
2543 If the software workaround is enabled for the host,
2544 guests do need not to enable it for nested guests.
2546 kvm.nx_huge_pages_recovery_ratio=
2547 [KVM] Controls how many 4KiB pages are periodically zapped
2548 back to huge pages. 0 disables the recovery, otherwise if
2549 the value is N KVM will zap 1/Nth of the 4KiB pages every
2550 period (see below). The default is 60.
2552 kvm.nx_huge_pages_recovery_period_ms=
2553 [KVM] Controls the time period at which KVM zaps 4KiB pages
2554 back to huge pages. If the value is a non-zero N, KVM will
2555 zap a portion (see ratio above) of the pages every N msecs.
2556 If the value is 0 (the default), KVM will pick a period based
2557 on the ratio, such that a page is zapped after 1 hour on average.
2559 kvm-amd.nested= [KVM,AMD] Control nested virtualization feature in
2560 KVM/SVM. Default is 1 (enabled).
2562 kvm-amd.npt= [KVM,AMD] Control KVM's use of Nested Page Tables,
2563 a.k.a. Two-Dimensional Page Tables. Default is 1
2564 (enabled). Disable by KVM if hardware lacks support
2568 [KVM,ARM] Select one of KVM/arm64's modes of operation.
2570 none: Forcefully disable KVM.
2572 nvhe: Standard nVHE-based mode, without support for
2575 protected: nVHE-based mode with support for guests whose
2576 state is kept private from the host.
2578 nested: VHE-based mode with support for nested
2579 virtualization. Requires at least ARMv8.3
2582 Defaults to VHE/nVHE based on hardware support. Setting
2583 mode to "protected" will disable kexec and hibernation
2584 for the host. "nested" is experimental and should be
2585 used with extreme caution.
2587 kvm-arm.vgic_v3_group0_trap=
2588 [KVM,ARM] Trap guest accesses to GICv3 group-0
2591 kvm-arm.vgic_v3_group1_trap=
2592 [KVM,ARM] Trap guest accesses to GICv3 group-1
2595 kvm-arm.vgic_v3_common_trap=
2596 [KVM,ARM] Trap guest accesses to GICv3 common
2599 kvm-arm.vgic_v4_enable=
2600 [KVM,ARM] Allow use of GICv4 for direct injection of
2603 kvm_cma_resv_ratio=n [PPC]
2604 Reserves given percentage from system memory area for
2605 contiguous memory allocation for KVM hash pagetable
2607 By default it reserves 5% of total system memory.
2611 kvm-intel.ept= [KVM,Intel] Control KVM's use of Extended Page Tables,
2612 a.k.a. Two-Dimensional Page Tables. Default is 1
2613 (enabled). Disable by KVM if hardware lacks support
2616 kvm-intel.emulate_invalid_guest_state=
2617 [KVM,Intel] Control whether to emulate invalid guest
2618 state. Ignored if kvm-intel.enable_unrestricted_guest=1,
2619 as guest state is never invalid for unrestricted
2620 guests. This param doesn't apply to nested guests (L2),
2621 as KVM never emulates invalid L2 guest state.
2622 Default is 1 (enabled).
2624 kvm-intel.flexpriority=
2625 [KVM,Intel] Control KVM's use of FlexPriority feature
2626 (TPR shadow). Default is 1 (enabled). Disalbe by KVM if
2627 hardware lacks support for it.
2630 [KVM,Intel] Control nested virtualization feature in
2631 KVM/VMX. Default is 1 (enabled).
2633 kvm-intel.unrestricted_guest=
2634 [KVM,Intel] Control KVM's use of unrestricted guest
2635 feature (virtualized real and unpaged mode). Default
2636 is 1 (enabled). Disable by KVM if EPT is disabled or
2637 hardware lacks support for it.
2639 kvm-intel.vmentry_l1d_flush=[KVM,Intel] Mitigation for L1 Terminal Fault
2642 Valid arguments: never, cond, always
2644 always: L1D cache flush on every VMENTER.
2645 cond: Flush L1D on VMENTER only when the code between
2646 VMEXIT and VMENTER can leak host memory.
2647 never: Disables the mitigation
2649 Default is cond (do L1 cache flush in specific instances)
2651 kvm-intel.vpid= [KVM,Intel] Control KVM's use of Virtual Processor
2652 Identification feature (tagged TLBs). Default is 1
2653 (enabled). Disable by KVM if hardware lacks support
2656 l1d_flush= [X86,INTEL]
2657 Control mitigation for L1D based snooping vulnerability.
2659 Certain CPUs are vulnerable to an exploit against CPU
2660 internal buffers which can forward information to a
2661 disclosure gadget under certain conditions.
2663 In vulnerable processors, the speculatively
2664 forwarded data can be used in a cache side channel
2665 attack, to access data to which the attacker does
2666 not have direct access.
2668 This parameter controls the mitigation. The
2671 on - enable the interface for the mitigation
2673 l1tf= [X86] Control mitigation of the L1TF vulnerability on
2676 The kernel PTE inversion protection is unconditionally
2677 enabled and cannot be disabled.
2680 Provides all available mitigations for the
2681 L1TF vulnerability. Disables SMT and
2682 enables all mitigations in the
2683 hypervisors, i.e. unconditional L1D flush.
2685 SMT control and L1D flush control via the
2686 sysfs interface is still possible after
2687 boot. Hypervisors will issue a warning
2688 when the first VM is started in a
2689 potentially insecure configuration,
2690 i.e. SMT enabled or L1D flush disabled.
2693 Same as 'full', but disables SMT and L1D
2694 flush runtime control. Implies the
2695 'nosmt=force' command line option.
2696 (i.e. sysfs control of SMT is disabled.)
2699 Leaves SMT enabled and enables the default
2700 hypervisor mitigation, i.e. conditional
2703 SMT control and L1D flush control via the
2704 sysfs interface is still possible after
2705 boot. Hypervisors will issue a warning
2706 when the first VM is started in a
2707 potentially insecure configuration,
2708 i.e. SMT enabled or L1D flush disabled.
2712 Disables SMT and enables the default
2713 hypervisor mitigation.
2715 SMT control and L1D flush control via the
2716 sysfs interface is still possible after
2717 boot. Hypervisors will issue a warning
2718 when the first VM is started in a
2719 potentially insecure configuration,
2720 i.e. SMT enabled or L1D flush disabled.
2723 Same as 'flush', but hypervisors will not
2724 warn when a VM is started in a potentially
2725 insecure configuration.
2728 Disables hypervisor mitigations and doesn't
2730 It also drops the swap size and available
2731 RAM limit restriction on both hypervisor and
2736 For details see: Documentation/admin-guide/hw-vuln/l1tf.rst
2742 lapic [X86-32,APIC] Enable the local APIC even if BIOS
2745 lapic= [X86,APIC] Do not use TSC deadline
2746 value for LAPIC timer one-shot implementation. Default
2747 back to the programmable timer unit in the LAPIC.
2748 Format: notscdeadline
2750 lapic_timer_c2_ok [X86,APIC] trust the local apic timer
2753 libata.dma= [LIBATA] DMA control
2754 libata.dma=0 Disable all PATA and SATA DMA
2755 libata.dma=1 PATA and SATA Disk DMA only
2756 libata.dma=2 ATAPI (CDROM) DMA only
2757 libata.dma=4 Compact Flash DMA only
2758 Combinations also work, so libata.dma=3 enables DMA
2759 for disks and CDROMs, but not CFs.
2761 libata.ignore_hpa= [LIBATA] Ignore HPA limit
2762 libata.ignore_hpa=0 keep BIOS limits (default)
2763 libata.ignore_hpa=1 ignore limits, using full disk
2765 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume
2769 libata.force= [LIBATA] Force configurations. The format is a comma-
2770 separated list of "[ID:]VAL" where ID is PORT[.DEVICE].
2771 PORT and DEVICE are decimal numbers matching port, link
2772 or device. Basically, it matches the ATA ID string
2773 printed on console by libata. If the whole ID part is
2774 omitted, the last PORT and DEVICE values are used. If
2775 ID hasn't been specified yet, the configuration applies
2776 to all ports, links and devices.
2778 If only DEVICE is omitted, the parameter applies to
2779 the port and all links and devices behind it. DEVICE
2780 number of 0 either selects the first device or the
2781 first fan-out link behind PMP device. It does not
2782 select the host link. DEVICE number of 15 selects the
2783 host link and device attached to it.
2785 The VAL specifies the configuration to force. As long
2786 as there is no ambiguity, shortcut notation is allowed.
2787 For example, both 1.5 and 1.5G would work for 1.5Gbps.
2788 The following configurations can be forced.
2790 * Cable type: 40c, 80c, short40c, unk, ign or sata.
2791 Any ID with matching PORT is used.
2793 * SATA link speed limit: 1.5Gbps or 3.0Gbps.
2795 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7].
2796 udma[/][16,25,33,44,66,100,133] notation is also
2799 * nohrst, nosrst, norst: suppress hard, soft and both
2802 * rstonce: only attempt one reset during hot-unplug
2805 * [no]dbdelay: Enable or disable the extra 200ms delay
2806 before debouncing a link PHY and device presence
2809 * [no]ncq: Turn on or off NCQ.
2811 * [no]ncqtrim: Enable or disable queued DSM TRIM.
2813 * [no]ncqati: Enable or disable NCQ trim on ATI chipset.
2815 * [no]trim: Enable or disable (unqueued) TRIM.
2817 * trim_zero: Indicate that TRIM command zeroes data.
2819 * max_trim_128m: Set 128M maximum trim size limit.
2821 * [no]dma: Turn on or off DMA transfers.
2823 * atapi_dmadir: Enable ATAPI DMADIR bridge support.
2825 * atapi_mod16_dma: Enable the use of ATAPI DMA for
2826 commands that are not a multiple of 16 bytes.
2828 * [no]dmalog: Enable or disable the use of the
2829 READ LOG DMA EXT command to access logs.
2831 * [no]iddevlog: Enable or disable access to the
2832 identify device data log.
2834 * [no]logdir: Enable or disable access to the general
2835 purpose log directory.
2837 * max_sec_128: Set transfer size limit to 128 sectors.
2839 * max_sec_1024: Set or clear transfer size limit to
2842 * max_sec_lba48: Set or clear transfer size limit to
2845 * [no]lpm: Enable or disable link power management.
2847 * [no]setxfer: Indicate if transfer speed mode setting
2850 * [no]fua: Disable or enable FUA (Force Unit Access)
2851 support for devices supporting this feature.
2853 * dump_id: Dump IDENTIFY data.
2855 * disable: Disable this device.
2857 If there are multiple matching configurations changing
2858 the same attribute, the last one is used.
2860 load_ramdisk= [RAM] [Deprecated]
2862 lockd.nlm_grace_period=P [NFS] Assign grace period.
2865 lockd.nlm_tcpport=N [NFS] Assign TCP port.
2868 lockd.nlm_timeout=T [NFS] Assign timeout value.
2871 lockd.nlm_udpport=M [NFS] Assign UDP port.
2874 lockdown= [SECURITY]
2875 { integrity | confidentiality }
2876 Enable the kernel lockdown feature. If set to
2877 integrity, kernel features that allow userland to
2878 modify the running kernel are disabled. If set to
2879 confidentiality, kernel features that allow userland
2880 to extract confidential information from the kernel
2883 locktorture.nreaders_stress= [KNL]
2884 Set the number of locking read-acquisition kthreads.
2885 Defaults to being automatically set based on the
2886 number of online CPUs.
2888 locktorture.nwriters_stress= [KNL]
2889 Set the number of locking write-acquisition kthreads.
2891 locktorture.onoff_holdoff= [KNL]
2892 Set time (s) after boot for CPU-hotplug testing.
2894 locktorture.onoff_interval= [KNL]
2895 Set time (s) between CPU-hotplug operations, or
2896 zero to disable CPU-hotplug testing.
2898 locktorture.shuffle_interval= [KNL]
2899 Set task-shuffle interval (jiffies). Shuffling
2900 tasks allows some CPUs to go into dyntick-idle
2901 mode during the locktorture test.
2903 locktorture.shutdown_secs= [KNL]
2904 Set time (s) after boot system shutdown. This
2905 is useful for hands-off automated testing.
2907 locktorture.stat_interval= [KNL]
2908 Time (s) between statistics printk()s.
2910 locktorture.stutter= [KNL]
2911 Time (s) to stutter testing, for example,
2912 specifying five seconds causes the test to run for
2913 five seconds, wait for five seconds, and so on.
2914 This tests the locking primitive's ability to
2915 transition abruptly to and from idle.
2917 locktorture.torture_type= [KNL]
2918 Specify the locking implementation to test.
2920 locktorture.verbose= [KNL]
2921 Enable additional printk() statements.
2923 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver
2926 loglevel= All Kernel Messages with a loglevel smaller than the
2927 console loglevel will be printed to the console. It can
2928 also be changed with klogd or other programs. The
2929 loglevels are defined as follows:
2931 0 (KERN_EMERG) system is unusable
2932 1 (KERN_ALERT) action must be taken immediately
2933 2 (KERN_CRIT) critical conditions
2934 3 (KERN_ERR) error conditions
2935 4 (KERN_WARNING) warning conditions
2936 5 (KERN_NOTICE) normal but significant condition
2937 6 (KERN_INFO) informational
2938 7 (KERN_DEBUG) debug-level messages
2940 log_buf_len=n[KMG] Sets the size of the printk ring buffer,
2941 in bytes. n must be a power of two and greater
2942 than the minimal size. The minimal size is defined
2943 by LOG_BUF_SHIFT kernel config parameter. There is
2944 also CONFIG_LOG_CPU_MAX_BUF_SHIFT config parameter
2945 that allows to increase the default size depending on
2946 the number of CPUs. See init/Kconfig for more details.
2948 logo.nologo [FB] Disables display of the built-in Linux logo.
2949 This may be used to provide more screen space for
2950 kernel log messages and is useful when debugging
2951 kernel boot problems.
2953 lp=0 [LP] Specify parallel ports to use, e.g,
2954 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses
2955 lp=reset first parallel port). 'lp=0' disables the
2956 lp=auto printer driver. 'lp=reset' (which can be
2957 specified in addition to the ports) causes
2958 attached printers to be reset. Using
2959 lp=port1,port2,... specifies the parallel ports
2960 to associate lp devices with, starting with
2961 lp0. A port specification may be 'none' to skip
2962 that lp device, or a parport name such as
2963 'parport0'. Specifying 'lp=auto' instead of a
2964 port specification list means that device IDs
2965 from each port should be examined, to see if
2966 an IEEE 1284-compliant printer is attached; if
2967 so, the driver will manage that printer.
2968 See also header of drivers/char/lp.c.
2971 Sets loops_per_jiffy to given constant, thus avoiding
2972 time-consuming boot-time autodetection (up to 250 ms per
2973 CPU). 0 enables autodetection (default). To determine
2974 the correct value for your kernel, boot with normal
2975 autodetection and see what value is printed. Note that
2976 on SMP systems the preset will be applied to all CPUs,
2977 which is likely to cause problems if your CPUs need
2978 significantly divergent settings. An incorrect value
2979 will cause delays in the kernel to be wrong, leading to
2980 unpredictable I/O errors and other breakage. Although
2981 unlikely, in the extreme case this might damage your
2985 Format: <io>,<irq>,<dma>
2987 lsm.debug [SECURITY] Enable LSM initialization debugging output.
2990 [SECURITY] Choose order of LSM initialization. This
2991 overrides CONFIG_LSM, and the "security=" parameter.
2993 machvec= [IA-64] Force the use of a particular machine-vector
2994 (machvec) in a generic kernel.
2995 Example: machvec=hpzx1
2997 machtype= [Loongson] Share the same kernel image file between
2998 different yeeloong laptops.
2999 Example: machtype=lemote-yeeloong-2f-7inch
3001 max_addr=nn[KMG] [KNL,BOOT,IA-64] All physical memory greater
3002 than or equal to this physical address is ignored.
3004 maxcpus= [SMP] Maximum number of processors that an SMP kernel
3005 will bring up during bootup. maxcpus=n : n >= 0 limits
3006 the kernel to bring up 'n' processors. Surely after
3007 bootup you can bring up the other plugged cpu by executing
3008 "echo 1 > /sys/devices/system/cpu/cpuX/online". So maxcpus
3009 only takes effect during system bootup.
3010 While n=0 is a special case, it is equivalent to "nosmp",
3011 which also disables the IO APIC.
3013 max_loop= [LOOP] The number of loop block devices that get
3014 (loop.max_loop) unconditionally pre-created at init time. The default
3015 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead
3016 of statically allocating a predefined number, loop
3017 devices can be requested on-demand with the
3018 /dev/loop-control interface.
3020 mce [X86-32] Machine Check Exception
3022 mce=option [X86-64] See Documentation/arch/x86/x86_64/boot-options.rst
3024 md= [HW] RAID subsystems devices and level
3025 See Documentation/admin-guide/md.rst.
3028 Format: <first>,<last>
3029 Specifies range of consoles to be captured by the MDA.
3032 Control mitigation for the Micro-architectural Data
3033 Sampling (MDS) vulnerability.
3035 Certain CPUs are vulnerable to an exploit against CPU
3036 internal buffers which can forward information to a
3037 disclosure gadget under certain conditions.
3039 In vulnerable processors, the speculatively
3040 forwarded data can be used in a cache side channel
3041 attack, to access data to which the attacker does
3042 not have direct access.
3044 This parameter controls the MDS mitigation. The
3047 full - Enable MDS mitigation on vulnerable CPUs
3048 full,nosmt - Enable MDS mitigation and disable
3049 SMT on vulnerable CPUs
3050 off - Unconditionally disable MDS mitigation
3052 On TAA-affected machines, mds=off can be prevented by
3053 an active TAA mitigation as both vulnerabilities are
3054 mitigated with the same mechanism so in order to disable
3055 this mitigation, you need to specify tsx_async_abort=off
3058 Not specifying this option is equivalent to
3061 For details see: Documentation/admin-guide/hw-vuln/mds.rst
3063 mem=nn[KMG] [HEXAGON] Set the memory size.
3064 Must be specified, otherwise memory size will be 0.
3066 mem=nn[KMG] [KNL,BOOT] Force usage of a specific amount of memory
3067 Amount of memory to be used in cases as follows:
3070 2 when the kernel is not able to see the whole system memory;
3071 3 memory that lies after 'mem=' boundary is excluded from
3072 the hypervisor, then assigned to KVM guests.
3073 4 to limit the memory available for kdump kernel.
3075 [ARC,MICROBLAZE] - the limit applies only to low memory,
3076 high memory is not affected.
3078 [ARM64] - only limits memory covered by the linear
3079 mapping. The NOMAP regions are not affected.
3081 [X86] Work as limiting max address. Use together
3082 with memmap= to avoid physical address space collisions.
3083 Without memmap= PCI devices could be placed at addresses
3084 belonging to unused RAM.
3086 Note that this only takes effects during boot time since
3087 in above case 3, memory may need be hot added after boot
3088 if system memory of hypervisor is not sufficient.
3091 [ARM,MIPS] - override the memory layout reported by
3093 Define a memory region of size nn[KMG] starting at
3095 Multiple different regions can be specified with
3096 multiple mem= parameters on the command line.
3098 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel
3101 memblock=debug [KNL] Enable memblock debug messages.
3104 [KNL,SH] Allow user to override the default size for
3105 per-device physically contiguous DMA buffers.
3107 memhp_default_state=online/offline
3108 [KNL] Set the initial state for the memory hotplug
3109 onlining policy. If not specified, the default value is
3110 set according to the
3111 CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE kernel config
3113 See Documentation/admin-guide/mm/memory-hotplug.rst.
3115 memmap=exactmap [KNL,X86] Enable setting of an exact
3116 E820 memory map, as specified by the user.
3117 Such memmap=exactmap lines can be constructed based on
3118 BIOS output or other requirements. See the memmap=nn@ss
3121 memmap=nn[KMG]@ss[KMG]
3122 [KNL, X86, MIPS, XTENSA] Force usage of a specific region of memory.
3123 Region of memory to be used is from ss to ss+nn.
3124 If @ss[KMG] is omitted, it is equivalent to mem=nn[KMG],
3125 which limits max address to nn[KMG].
3126 Multiple different regions can be specified,
3129 memmap=100M@2G,100M#3G,1G!1024G
3131 memmap=nn[KMG]#ss[KMG]
3132 [KNL,ACPI] Mark specific memory as ACPI data.
3133 Region of memory to be marked is from ss to ss+nn.
3135 memmap=nn[KMG]$ss[KMG]
3136 [KNL,ACPI] Mark specific memory as reserved.
3137 Region of memory to be reserved is from ss to ss+nn.
3138 Example: Exclude memory from 0x18690000-0x1869ffff
3139 memmap=64K$0x18690000
3141 memmap=0x10000$0x18690000
3142 Some bootloaders may need an escape character before '$',
3143 like Grub2, otherwise '$' and the following number
3146 memmap=nn[KMG]!ss[KMG]
3147 [KNL,X86] Mark specific memory as protected.
3148 Region of memory to be used, from ss to ss+nn.
3149 The memory region may be marked as e820 type 12 (0xc)
3150 and is NVDIMM or ADR memory.
3152 memmap=<size>%<offset>-<oldtype>+<newtype>
3153 [KNL,ACPI] Convert memory within the specified region
3154 from <oldtype> to <newtype>. If "-<oldtype>" is left
3155 out, the whole region will be marked as <newtype>,
3156 even if previously unavailable. If "+<newtype>" is left
3157 out, matching memory will be removed. Types are
3158 specified as e820 types, e.g., 1 = RAM, 2 = reserved,
3159 3 = ACPI, 12 = PRAM.
3161 memory_corruption_check=0/1 [X86]
3162 Some BIOSes seem to corrupt the first 64k of
3163 memory when doing things like suspend/resume.
3164 Setting this option will scan the memory
3165 looking for corruption. Enabling this will
3166 both detect corruption and prevent the kernel
3167 from using the memory being corrupted.
3168 However, its intended as a diagnostic tool; if
3169 repeatable BIOS-originated corruption always
3170 affects the same memory, you can use memmap=
3171 to prevent the kernel from using that memory.
3173 memory_corruption_check_size=size [X86]
3174 By default it checks for corruption in the low
3175 64k, making this memory unavailable for normal
3176 use. Use this parameter to scan for
3177 corruption in more or less memory.
3179 memory_corruption_check_period=seconds [X86]
3180 By default it checks for corruption every 60
3181 seconds. Use this parameter to check at some
3182 other rate. 0 disables periodic checking.
3184 memory_hotplug.memmap_on_memory
3185 [KNL,X86,ARM] Boolean flag to enable this feature.
3186 Format: {on | off (default)}
3187 When enabled, runtime hotplugged memory will
3188 allocate its internal metadata (struct pages,
3189 those vmemmap pages cannot be optimized even
3190 if hugetlb_free_vmemmap is enabled) from the
3191 hotadded memory which will allow to hotadd a
3192 lot of memory without requiring additional
3194 This feature is disabled by default because it
3195 has some implication on large (e.g. GB)
3196 allocations in some configurations (e.g. small
3198 The state of the flag can be read in
3199 /sys/module/memory_hotplug/parameters/memmap_on_memory.
3200 Note that even when enabled, there are a few cases where
3201 the feature is not effective.
3203 memtest= [KNL,X86,ARM,M68K,PPC,RISCV] Enable memtest
3205 default : 0 <disable>
3206 Specifies the number of memtest passes to be
3207 performed. Each pass selects another test
3208 pattern from a given set of patterns. Memtest
3209 fills the memory with this pattern, validates
3210 memory contents and reserves bad memory
3211 regions that are detected.
3213 mem_encrypt= [X86-64] AMD Secure Memory Encryption (SME) control
3214 Valid arguments: on, off
3215 Default (depends on kernel configuration option):
3216 on (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=y)
3217 off (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=n)
3218 mem_encrypt=on: Activate SME
3219 mem_encrypt=off: Do not activate SME
3221 Refer to Documentation/virt/kvm/x86/amd-memory-encryption.rst
3222 for details on when memory encryption can be activated.
3224 mem_sleep_default= [SUSPEND] Default system suspend mode:
3225 s2idle - Suspend-To-Idle
3226 shallow - Power-On Suspend or equivalent (if supported)
3227 deep - Suspend-To-RAM or equivalent (if supported)
3228 See Documentation/admin-guide/pm/sleep-states.rst.
3230 mfgpt_irq= [IA-32] Specify the IRQ to use for the
3231 Multi-Function General Purpose Timers on AMD Geode
3234 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when
3235 the BIOS has incorrectly applied a workaround. TinyBIOS
3236 version 0.98 is known to be affected, 0.99 fixes the
3237 problem by letting the user disable the workaround.
3241 min_addr=nn[KMG] [KNL,BOOT,IA-64] All physical memory below this
3242 physical address is ignored.
3244 mini2440= [ARM,HW,KNL]
3245 Format:[0..2][b][c][t]
3247 MINI2440 configuration specification:
3248 0 - The attached screen is the 3.5" TFT
3249 1 - The attached screen is the 7" TFT
3250 2 - The VGA Shield is attached (1024x768)
3251 Leaving out the screen size parameter will not load
3252 the TFT driver, and the framebuffer will be left
3254 b - Enable backlight. The TFT backlight pin will be
3255 linked to the kernel VESA blanking code and a GPIO
3256 LED. This parameter is not necessary when using the
3258 c - Enable the s3c camera interface.
3259 t - Reserved for enabling touchscreen support. The
3260 touchscreen support is not enabled in the mainstream
3261 kernel as of 2.6.30, a preliminary port can be found
3262 in the "bleeding edge" mini2440 support kernel at
3263 https://repo.or.cz/w/linux-2.6/mini2440.git
3266 [X86,PPC,S390,ARM64] Control optional mitigations for
3267 CPU vulnerabilities. This is a set of curated,
3268 arch-independent options, each of which is an
3269 aggregation of existing arch-specific options.
3272 Disable all optional CPU mitigations. This
3273 improves system performance, but it may also
3274 expose users to several CPU vulnerabilities.
3275 Equivalent to: nopti [X86,PPC]
3276 if nokaslr then kpti=0 [ARM64]
3277 nospectre_v1 [X86,PPC]
3279 nospectre_v2 [X86,PPC,S390,ARM64]
3280 spectre_v2_user=off [X86]
3281 spec_store_bypass_disable=off [X86,PPC]
3282 ssbd=force-off [ARM64]
3283 nospectre_bhb [ARM64]
3286 tsx_async_abort=off [X86]
3287 kvm.nx_huge_pages=off [X86]
3288 srbds=off [X86,INTEL]
3289 no_entry_flush [PPC]
3290 no_uaccess_flush [PPC]
3291 mmio_stale_data=off [X86]
3295 This does not have any effect on
3296 kvm.nx_huge_pages when
3297 kvm.nx_huge_pages=force.
3300 Mitigate all CPU vulnerabilities, but leave SMT
3301 enabled, even if it's vulnerable. This is for
3302 users who don't want to be surprised by SMT
3303 getting disabled across kernel upgrades, or who
3304 have other ways of avoiding SMT-based attacks.
3305 Equivalent to: (default behavior)
3308 Mitigate all CPU vulnerabilities, disabling SMT
3309 if needed. This is for users who always want to
3310 be fully mitigated, even if it means losing SMT.
3311 Equivalent to: l1tf=flush,nosmt [X86]
3312 mds=full,nosmt [X86]
3313 tsx_async_abort=full,nosmt [X86]
3314 mmio_stale_data=full,nosmt [X86]
3315 retbleed=auto,nosmt [X86]
3318 [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this
3319 parameter allows control of the logging verbosity for
3320 the additional memory initialisation checks. A value
3321 of 0 disables mminit logging and a level of 4 will
3322 log everything. Information is printed at KERN_DEBUG
3323 so loglevel=8 may also need to be specified.
3326 [X86,INTEL] Control mitigation for the Processor
3327 MMIO Stale Data vulnerabilities.
3329 Processor MMIO Stale Data is a class of
3330 vulnerabilities that may expose data after an MMIO
3331 operation. Exposed data could originate or end in
3332 the same CPU buffers as affected by MDS and TAA.
3333 Therefore, similar to MDS and TAA, the mitigation
3334 is to clear the affected CPU buffers.
3336 This parameter controls the mitigation. The
3339 full - Enable mitigation on vulnerable CPUs
3341 full,nosmt - Enable mitigation and disable SMT on
3344 off - Unconditionally disable mitigation
3346 On MDS or TAA affected machines,
3347 mmio_stale_data=off can be prevented by an active
3348 MDS or TAA mitigation as these vulnerabilities are
3349 mitigated with the same mechanism so in order to
3350 disable this mitigation, you need to specify
3351 mds=off and tsx_async_abort=off too.
3353 Not specifying this option is equivalent to
3354 mmio_stale_data=full.
3357 Documentation/admin-guide/hw-vuln/processor_mmio_stale_data.rst
3359 <module>.async_probe[=<bool>] [KNL]
3360 If no <bool> value is specified or if the value
3361 specified is not a valid <bool>, enable asynchronous
3362 probe on this module. Otherwise, enable/disable
3363 asynchronous probe on this module as indicated by the
3364 <bool> value. See also: module.async_probe
3366 module.async_probe=<bool>
3367 [KNL] When set to true, modules will use async probing
3368 by default. To enable/disable async probing for a
3369 specific module, use the module specific control that
3370 is documented under <module>.async_probe. When both
3371 module.async_probe and <module>.async_probe are
3372 specified, <module>.async_probe takes precedence for
3373 the specific module.
3375 module.enable_dups_trace
3376 [KNL] When CONFIG_MODULE_DEBUG_AUTOLOAD_DUPS is set,
3377 this means that duplicate request_module() calls will
3378 trigger a WARN_ON() instead of a pr_warn(). Note that
3379 if MODULE_DEBUG_AUTOLOAD_DUPS_TRACE is set, WARN_ON()s
3380 will always be issued and this option does nothing.
3382 [KNL] When CONFIG_MODULE_SIG is set, this means that
3383 modules without (valid) signatures will fail to load.
3384 Note that if CONFIG_MODULE_SIG_FORCE is set, that
3385 is always true, so this option does nothing.
3387 module_blacklist= [KNL] Do not load a comma-separated list of
3388 modules. Useful for debugging problem modules.
3391 [MOUSE] Maximum time between finger touching and
3392 leaving touchpad surface for touch to be considered
3393 a tap and be reported as a left button click (for
3394 touchpads working in absolute mode only).
3396 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices
3397 reporting absolute coordinates, such as tablets
3398 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices
3399 reporting absolute coordinates, such as tablets
3401 movablecore= [KNL,X86,IA-64,PPC]
3402 Format: nn[KMGTPE] | nn%
3403 This parameter is the complement to kernelcore=, it
3404 specifies the amount of memory used for migratable
3405 allocations. If both kernelcore and movablecore is
3406 specified, then kernelcore will be at *least* the
3407 specified value but may be more. If movablecore on its
3408 own is specified, the administrator must be careful
3409 that the amount of memory usable for all allocations
3412 movable_node [KNL] Boot-time switch to make hotplugable memory
3413 NUMA nodes to be movable. This means that the memory
3414 of such nodes will be usable only for movable
3415 allocations which rules out almost all kernel
3416 allocations. Use with caution!
3418 MTD_Partition= [MTD]
3419 Format: <name>,<region-number>,<size>,<offset>
3421 MTD_Region= [MTD] Format:
3422 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>]
3425 See drivers/mtd/parsers/cmdlinepart.c
3428 ARM/S3C2412 JIVE boot control
3430 See arch/arm/mach-s3c/mach-jive.c
3432 mtouchusb.raw_coordinates=
3433 [HW] Make the MicroTouch USB driver use raw coordinates
3434 ('y', default) or cooked coordinates ('n')
3436 mtrr_chunk_size=nn[KMG] [X86]
3437 used for mtrr cleanup. It is largest continuous chunk
3438 that could hold holes aka. UC entries.
3440 mtrr_gran_size=nn[KMG] [X86]
3441 Used for mtrr cleanup. It is granularity of mtrr block.
3443 Large value could prevent small alignment from
3446 mtrr_spare_reg_nr=n [X86]
3448 Range: 0,7 : spare reg number
3450 Used for mtrr cleanup. It is spare mtrr entries number.
3451 Set to 2 or more if your graphical card needs more.
3453 multitce=off [PPC] This parameter disables the use of the pSeries
3454 firmware feature for updating multiple TCE entries
3457 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card
3459 netdev= [NET] Network devices parameters
3460 Format: <irq>,<io>,<mem_start>,<mem_end>,<name>
3461 Note that mem_start is often overloaded to mean
3462 something different and driver-specific.
3463 This usage is only documented in each driver source
3466 netpoll.carrier_timeout=
3467 [NET] Specifies amount of time (in seconds) that
3468 netpoll should wait for a carrier. By default netpoll
3472 [NETFILTER] Enable connection tracking flow accounting
3473 0 to disable accounting
3474 1 to enable accounting
3478 [NFS] sets the pathname to the program which is used
3479 to update the NFS client cache entries.
3481 nfs.cache_getent_timeout=
3482 [NFS] sets the timeout after which an attempt to
3483 update a cache entry is deemed to have failed.
3485 nfs.callback_nr_threads=
3486 [NFSv4] set the total number of threads that the
3487 NFS client will assign to service NFSv4 callback
3490 nfs.callback_tcpport=
3491 [NFS] set the TCP port on which the NFSv4 callback
3492 channel should listen.
3495 [NFS] enable 64-bit inode numbers.
3496 If zero, the NFS client will fake up a 32-bit inode
3497 number for the readdir() and stat() syscalls instead
3498 of returning the full 64-bit number.
3499 The default is to return 64-bit inode numbers.
3501 nfs.idmap_cache_timeout=
3502 [NFS] set the maximum lifetime for idmapper cache
3505 nfs.max_session_cb_slots=
3506 [NFSv4.1] Sets the maximum number of session
3507 slots the client will assign to the callback
3508 channel. This determines the maximum number of
3509 callbacks the client will process in parallel for
3510 a particular server.
3512 nfs.max_session_slots=
3513 [NFSv4.1] Sets the maximum number of session slots
3514 the client will attempt to negotiate with the server.
3515 This limits the number of simultaneous RPC requests
3516 that the client can send to the NFSv4.1 server.
3517 Note that there is little point in setting this
3518 value higher than the max_tcp_slot_table_limit.
3520 nfs.nfs4_disable_idmapping=
3521 [NFSv4] When set to the default of '1', this option
3522 ensures that both the RPC level authentication
3523 scheme and the NFS level operations agree to use
3524 numeric uids/gids if the mount is using the
3525 'sec=sys' security flavour. In effect it is
3526 disabling idmapping, which can make migration from
3527 legacy NFSv2/v3 systems to NFSv4 easier.
3528 Servers that do not support this mode of operation
3529 will be autodetected by the client, and it will fall
3530 back to using the idmapper.
3531 To turn off this behaviour, set the value to '0'.
3534 [NFS4] Specify an additional fixed unique ident-
3535 ification string that NFSv4 clients can insert into
3536 their nfs_client_id4 string. This is typically a
3537 UUID that is generated at system install time.
3539 nfs.recover_lost_locks=
3540 [NFSv4] Attempt to recover locks that were lost due
3541 to a lease timeout on the server. Please note that
3542 doing this risks data corruption, since there are
3543 no guarantees that the file will remain unchanged
3544 after the locks are lost.
3545 If you want to enable the kernel legacy behaviour of
3546 attempting to recover these locks, then set this
3548 The default parameter value of '0' causes the kernel
3549 not to attempt recovery of lost locks.
3551 nfs.send_implementation_id=
3552 [NFSv4.1] Send client implementation identification
3553 information in exchange_id requests.
3554 If zero, no implementation identification information
3556 The default is to send the implementation identification
3559 nfs4.layoutstats_timer=
3560 [NFSv4.2] Change the rate at which the kernel sends
3561 layoutstats to the pNFS metadata server.
3563 Setting this to value to 0 causes the kernel to use
3564 whatever value is the default set by the layout
3565 driver. A non-zero value sets the minimum interval
3566 in seconds between layoutstats transmissions.
3568 nfsd.inter_copy_offload_enable=
3569 [NFSv4.2] When set to 1, the server will support
3570 server-to-server copies for which this server is
3571 the destination of the copy.
3573 nfsd.nfs4_disable_idmapping=
3574 [NFSv4] When set to the default of '1', the NFSv4
3575 server will return only numeric uids and gids to
3576 clients using auth_sys, and will accept numeric uids
3577 and gids from such clients. This is intended to ease
3578 migration from NFSv2/v3.
3580 nfsd.nfsd4_ssc_umount_timeout=
3581 [NFSv4.2] When used as the destination of a
3582 server-to-server copy, knfsd temporarily mounts
3583 the source server. It caches the mount in case
3584 it will be needed again, and discards it if not
3585 used for the number of milliseconds specified by
3588 nfsaddrs= [NFS] Deprecated. Use ip= instead.
3589 See Documentation/admin-guide/nfs/nfsroot.rst.
3591 nfsroot= [NFS] nfs root filesystem for disk-less boxes.
3592 See Documentation/admin-guide/nfs/nfsroot.rst.
3594 nfsrootdebug [NFS] enable nfsroot debugging messages.
3595 See Documentation/admin-guide/nfs/nfsroot.rst.
3597 nmi_backtrace.backtrace_idle [KNL]
3598 Dump stacks even of idle CPUs in response to an
3599 NMI stack-backtrace request.
3601 nmi_debug= [KNL,SH] Specify one or more actions to take
3602 when a NMI is triggered.
3603 Format: [state][,regs][,debounce][,die]
3605 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels
3606 Format: [panic,][nopanic,][num]
3608 0 - turn hardlockup detector in nmi_watchdog off
3609 1 - turn hardlockup detector in nmi_watchdog on
3610 When panic is specified, panic when an NMI watchdog
3611 timeout occurs (or 'nopanic' to not panic on an NMI
3612 watchdog, if CONFIG_BOOTPARAM_HARDLOCKUP_PANIC is set)
3613 To disable both hard and soft lockup detectors,
3614 please see 'nowatchdog'.
3615 This is useful when you use a panic=... timeout and
3616 need the box quickly up again.
3618 These settings can be accessed at runtime via
3619 the nmi_watchdog and hardlockup_panic sysctls.
3621 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths
3622 emulation library even if a 387 maths coprocessor
3625 no4lvl [RISCV] Disable 4-level and 5-level paging modes. Forces
3626 kernel to use 3-level paging instead.
3628 no5lvl [X86-64,RISCV] Disable 5-level paging mode. Forces
3629 kernel to use 4-level paging instead.
3631 noaliencache [MM, NUMA, SLAB] Disables the allocation of alien
3632 caches in the slab allocator. Saves per-node memory,
3633 but will impact performance.
3637 noaltinstr [S390] Disables alternative instructions patching
3638 (CPU alternatives feature).
3640 noapic [SMP,APIC] Tells the kernel to not make use of any
3641 IOAPICs that may be present in the system.
3643 noautogroup Disable scheduler automatic task group creation.
3648 [HW] Never suspend the console
3649 Disable suspending of consoles during suspend and
3650 hibernate operations. Once disabled, debugging
3651 messages can reach various consoles while the rest
3652 of the system is being put to sleep (ie, while
3653 debugging driver suspend/resume hooks). This may
3654 not work reliably with all consoles, but is known
3655 to work with serial and VGA consoles.
3656 To facilitate more flexible debugging, we also add
3657 console_suspend, a printk module parameter to control
3658 it. Users could use console_suspend (usually
3659 /sys/module/printk/parameters/console_suspend) to
3660 turn on/off it dynamically.
3663 [KNL] Disable object debugging
3665 nodsp [SH] Disable hardware DSP at boot time.
3667 noefi Disable EFI runtime services support.
3669 no_entry_flush [PPC] Don't flush the L1-D cache when entering the kernel.
3674 This affects only 32-bit executables.
3675 noexec32=on: enable non-executable mappings (default)
3676 read doesn't imply executable mappings
3677 noexec32=off: disable non-executable mappings
3678 read implies executable mappings
3680 no_file_caps Tells the kernel not to honor file capabilities. The
3681 only way then for a file to be executed with privilege
3682 is to be setuid root or executed by root.
3684 nofpu [MIPS,SH] Disable hardware FPU at boot time.
3686 nofsgsbase [X86] Disables FSGSBASE instructions.
3688 nofxsr [BUGS=X86-32] Disables x86 floating point extended
3689 register save and restore. The kernel will only save
3690 legacy floating-point registers on task switch.
3692 nohalt [IA-64] Tells the kernel not to use the power saving
3693 function PAL_HALT_LIGHT when idle. This increases
3694 power-consumption. On the positive side, it reduces
3695 interrupt wake-up latency, which may improve performance
3696 in certain environments such as networked servers or
3700 Force pointers printed to the console or buffers to be
3701 unhashed. By default, when a pointer is printed via %p
3702 format string, that pointer is "hashed", i.e. obscured
3703 by hashing the pointer value. This is a security feature
3704 that hides actual kernel addresses from unprivileged
3705 users, but it also makes debugging the kernel more
3706 difficult since unequal pointers can no longer be
3707 compared. However, if this command-line option is
3708 specified, then all normal pointers will have their true
3709 value printed. This option should only be specified when
3710 debugging the kernel. Please do not use on production
3713 nohibernate [HIBERNATION] Disable hibernation and resume.
3715 nohlt [ARM,ARM64,MICROBLAZE,SH] Forces the kernel to busy wait
3716 in do_idle() and not use the arch_cpu_idle()
3717 implementation; requires CONFIG_GENERIC_IDLE_POLL_SETUP
3718 to be effective. This is useful on platforms where the
3719 sleep(SH) or wfi(ARM,ARM64) instructions do not work
3720 correctly or when doing power measurements to evaluate
3721 the impact of the sleep instructions. This is also
3722 useful when using JTAG debugger.
3724 nohugeiomap [KNL,X86,PPC,ARM64] Disable kernel huge I/O mappings.
3726 nohugevmalloc [KNL,X86,PPC,ARM64] Disable kernel huge vmalloc mappings.
3728 nohz= [KNL] Boottime enable/disable dynamic ticks
3729 Valid arguments: on, off
3732 nohz_full= [KNL,BOOT,SMP,ISOL]
3733 The argument is a cpu list, as described above.
3734 In kernels built with CONFIG_NO_HZ_FULL=y, set
3735 the specified list of CPUs whose tick will be stopped
3736 whenever possible. The boot CPU will be forced outside
3737 the range to maintain the timekeeping. Any CPUs
3738 in this list will have their RCU callbacks offloaded,
3739 just as if they had also been called out in the
3740 rcu_nocbs= boot parameter.
3742 Note that this argument takes precedence over
3743 the CONFIG_RCU_NOCB_CPU_DEFAULT_ALL option.
3745 noinitrd [RAM] Tells the kernel not to load any configured
3748 nointremap [X86-64, Intel-IOMMU] Do not enable interrupt
3750 [Deprecated - use intremap=off]
3754 noinvpcid [X86] Disable the INVPCID cpu feature.
3756 noiotrap [SH] Disables trapped I/O port accesses.
3758 noirqdebug [X86-32] Disables the code which attempts to detect and
3759 disable unhandled interrupt sources.
3761 noisapnp [ISAPNP] Disables ISA PnP code.
3763 nojitter [IA-64] Disables jitter checking for ITC timers.
3766 When CONFIG_RANDOMIZE_BASE is set, this disables
3767 kernel and module base offset ASLR (Address Space
3768 Layout Randomization).
3770 no-kvmapf [X86,KVM] Disable paravirtualized asynchronous page
3773 no-kvmclock [X86,KVM] Disable paravirtualized KVM clock driver
3775 nolapic [X86-32,APIC] Do not enable or use the local APIC.
3777 nolapic_timer [X86-32,APIC] Do not use the local APIC timer.
3779 nomca [IA-64] Disable machine check abort handling
3781 nomce [X86-32] Disable Machine Check Exception
3783 nomfgpt [X86-32] Disable Multi-Function General Purpose
3784 Timer usage (for AMD Geode machines).
3786 nomodeset Disable kernel modesetting. Most systems' firmware
3787 sets up a display mode and provides framebuffer memory
3788 for output. With nomodeset, DRM and fbdev drivers will
3789 not load if they could possibly displace the pre-
3790 initialized output. Only the system framebuffer will
3791 be available for use. The respective drivers will not
3792 perform display-mode changes or accelerated rendering.
3794 Useful as error fallback, or for testing and debugging.
3796 nomodule Disable module load
3798 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to
3799 shutdown the other cpus. Instead use the REBOOT_VECTOR
3802 nopat [X86] Disable PAT (page attribute table extension of
3803 pagetables) support.
3805 nopcid [X86-64] Disable the PCID cpu feature.
3807 nopku [X86] Disable Memory Protection Keys CPU feature found
3811 Equivalent to pti=off
3813 nopv= [X86,XEN,KVM,HYPER_V,VMWARE]
3814 Disables the PV optimizations forcing the guest to run
3815 as generic guest with no PV drivers. Currently support
3816 XEN HVM, KVM, HYPER_V and VMWARE guest.
3818 nopvspin [X86,XEN,KVM]
3819 Disables the qspinlock slow path using PV optimizations
3820 which allow the hypervisor to 'idle' the guest on lock
3823 norandmaps Don't use address space randomization. Equivalent to
3824 echo 0 > /proc/sys/kernel/randomize_va_space
3826 noreplace-smp [X86-32,SMP] Don't replace SMP instructions
3827 with UP alternatives
3829 noresume [SWSUSP] Disables resume and restores original swap
3834 no-scroll [VGA] Disables scrollback.
3835 This is required for the Braillex ib80-piezo Braille
3836 reader made by F.H. Papenmeier (Germany).
3838 nosgx [X86-64,SGX] Disables Intel SGX kernel support.
3841 Disable SMAP (Supervisor Mode Access Prevention)
3842 even if it is supported by processor.
3845 Disable SMEP (Supervisor Mode Execution Prevention)
3846 even if it is supported by processor.
3848 nosmp [SMP] Tells an SMP kernel to act as a UP kernel,
3849 and disable the IO APIC. legacy for "maxcpus=0".
3851 nosmt [KNL,S390] Disable symmetric multithreading (SMT).
3852 Equivalent to smt=1.
3854 [KNL,X86] Disable symmetric multithreading (SMT).
3855 nosmt=force: Force disable SMT, cannot be undone
3856 via the sysfs control file.
3858 nosoftlockup [KNL] Disable the soft-lockup detector.
3860 nospec_store_bypass_disable
3861 [HW] Disable all mitigations for the Speculative Store Bypass vulnerability
3863 nospectre_bhb [ARM64] Disable all mitigations for Spectre-BHB (branch
3864 history injection) vulnerability. System may allow data leaks
3867 nospectre_v1 [X86,PPC] Disable mitigations for Spectre Variant 1
3868 (bounds check bypass). With this option data leaks are
3869 possible in the system.
3871 nospectre_v2 [X86,PPC_E500,ARM64] Disable all mitigations for
3872 the Spectre variant 2 (indirect branch prediction)
3873 vulnerability. System may allow data leaks with this
3876 no-steal-acc [X86,PV_OPS,ARM64,PPC/PSERIES] Disable paravirtualized
3877 steal time accounting. steal time is computed, but
3878 won't influence scheduler behaviour
3880 nosync [HW,M68K] Disables sync negotiation for all devices.
3882 no_timer_check [X86,APIC] Disables the code which tests for
3883 broken timer IRQ sources.
3886 [PPC] Don't flush the L1-D cache after accessing user data.
3888 novmcoredd [KNL,KDUMP]
3889 Disable device dump. Device dump allows drivers to
3890 append dump data to vmcore so you can collect driver
3891 specified debug info. Drivers can append the data
3892 without any limit and this data is stored in memory,
3893 so this may cause significant memory stress. Disabling
3894 device dump can help save memory but the driver debug
3895 data will be no longer available. This parameter
3896 is only available when CONFIG_PROC_VMCORE_DEVICE_DUMP
3900 [X86,PV_OPS] Disable paravirtualized VMware scheduler
3901 clock and use the default one.
3903 nowatchdog [KNL] Disable both lockup detectors, i.e.
3904 soft-lockup and NMI watchdog (hard-lockup).
3908 nox2apic [X86-64,APIC] Do not enable x2APIC mode.
3910 NOTE: this parameter will be ignored on systems with the
3911 LEGACY_XAPIC_DISABLED bit set in the
3912 IA32_XAPIC_DISABLE_STATUS MSR.
3914 noxsave [BUGS=X86] Disables x86 extended register state save
3915 and restore using xsave. The kernel will fallback to
3916 enabling legacy floating-point and sse state.
3918 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended
3919 register states. The kernel will fall back to use
3920 xsave to save the states. By using this parameter,
3921 performance of saving the states is degraded because
3922 xsave doesn't support modified optimization while
3923 xsaveopt supports it on xsaveopt enabled systems.
3925 noxsaves [X86] Disables xsaves and xrstors used in saving and
3926 restoring x86 extended register state in compacted
3927 form of xsave area. The kernel will fall back to use
3928 xsaveopt and xrstor to save and restore the states
3929 in standard form of xsave area. By using this
3930 parameter, xsave area per process might occupy more
3931 memory on xsaves enabled systems.
3933 nps_mtm_hs_ctr= [KNL,ARC]
3934 This parameter sets the maximum duration, in
3935 cycles, each HW thread of the CTOP can run
3936 without interruptions, before HW switches it.
3937 The actual maximum duration is 16 times this
3939 Format: integer between 1 and 255
3942 nptcg= [IA-64] Override max number of concurrent global TLB
3943 purges which is reported from either PAL_VM_SUMMARY or
3946 nr_cpus= [SMP] Maximum number of processors that an SMP kernel
3947 could support. nr_cpus=n : n >= 1 limits the kernel to
3948 support 'n' processors. It could be larger than the
3949 number of already plugged CPU during bootup, later in
3950 runtime you can physically add extra cpu until it reaches
3951 n. So during boot up some boot time memory for per-cpu
3952 variables need be pre-allocated for later physical cpu
3955 nr_uarts= [SERIAL] maximum number of UARTs to be registered.
3957 numa=off [KNL, ARM64, PPC, RISCV, SPARC, X86] Disable NUMA, Only
3958 set up a single NUMA node spanning all memory.
3960 numa_balancing= [KNL,ARM64,PPC,RISCV,S390,X86] Enable or disable automatic
3962 Allowed values are enable and disable
3964 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA.
3965 'node', 'default' can be specified
3966 This can be set from sysctl after boot.
3967 See Documentation/admin-guide/sysctl/vm.rst for details.
3969 ohci1394_dma=early [HW] enable debugging via the ohci1394 driver.
3970 See Documentation/core-api/debugging-via-ohci1394.rst for more
3973 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands
3974 Rather than timing out after 20 ms if an EC
3975 command is not properly ACKed, override the length
3976 of the timeout. We have interrupts disabled while
3977 waiting for the ACK, so if this is set too high
3978 interrupts *may* be lost!
3980 omap_mux= [OMAP] Override bootloader pin multiplexing.
3981 Format: <mux_mode0.mode_name=value>...
3982 For example, to override I2C bus2:
3983 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100
3985 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration
3987 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock]
3989 boundary - index of last SLC block on Flex-OneNAND.
3990 The remaining blocks are configured as MLC blocks.
3991 lock - Configure if Flex-OneNAND boundary should be locked.
3992 Once locked, the boundary cannot be changed.
3993 1 indicates lock status, 0 indicates unlock status.
3995 oops=panic Always panic on oopses. Default is to just kill the
3996 process, but there is a small probability of
3997 deadlocking the machine.
3998 This will also cause panics on machine check exceptions.
3999 Useful together with panic=30 to trigger a reboot.
4002 [KNL] Boolean flag to control whether the page allocator
4003 should randomize its free lists. The randomization may
4004 be automatically enabled if the kernel detects it is
4005 running on a platform with a direct-mapped memory-side
4006 cache, and this parameter can be used to
4007 override/disable that behavior. The state of the flag
4008 can be read from sysfs at:
4009 /sys/module/page_alloc/parameters/shuffle.
4011 page_owner= [KNL] Boot-time page_owner enabling option.
4012 Storage of the information about who allocated
4013 each page is disabled in default. With this switch,
4015 on: enable the feature
4017 page_poison= [KNL] Boot-time parameter changing the state of
4018 poisoning on the buddy allocator, available with
4019 CONFIG_PAGE_POISONING=y.
4020 off: turn off poisoning (default)
4021 on: turn on poisoning
4023 page_reporting.page_reporting_order=
4024 [KNL] Minimal page reporting order
4026 Adjust the minimal page reporting order. The page
4027 reporting is disabled when it exceeds MAX_ORDER.
4029 panic= [KNL] Kernel behaviour on panic: delay <timeout>
4030 timeout > 0: seconds before rebooting
4031 timeout = 0: wait forever
4032 timeout < 0: reboot immediately
4035 panic_print= Bitmask for printing system info when panic happens.
4036 User can chose combination of the following bits:
4037 bit 0: print all tasks info
4038 bit 1: print system memory info
4039 bit 2: print timer info
4040 bit 3: print locks info if CONFIG_LOCKDEP is on
4041 bit 4: print ftrace buffer
4042 bit 5: print all printk messages in buffer
4043 bit 6: print all CPUs backtrace (if available in the arch)
4044 *Be aware* that this option may print a _lot_ of lines,
4045 so there are risks of losing older messages in the log.
4046 Use this option carefully, maybe worth to setup a
4047 bigger log buffer with "log_buf_len" along with this.
4049 panic_on_taint= Bitmask for conditionally calling panic() in add_taint()
4050 Format: <hex>[,nousertaint]
4051 Hexadecimal bitmask representing the set of TAINT flags
4052 that will cause the kernel to panic when add_taint() is
4053 called with any of the flags in this set.
4054 The optional switch "nousertaint" can be utilized to
4055 prevent userspace forced crashes by writing to sysctl
4056 /proc/sys/kernel/tainted any flagset matching with the
4057 bitmask set on panic_on_taint.
4058 See Documentation/admin-guide/tainted-kernels.rst for
4059 extra details on the taint flags that users can pick
4060 to compose the bitmask to assign to panic_on_taint.
4062 panic_on_warn panic() instead of WARN(). Useful to cause kdump
4065 parkbd.port= [HW] Parallel port number the keyboard adapter is
4066 connected to, default is 0.
4068 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation,
4069 0 for XT, 1 for AT (default is AT).
4072 parport= [HW,PPT] Specify parallel ports. 0 disables.
4073 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] }
4074 Use 'auto' to force the driver to use any
4075 IRQ/DMA settings detected (the default is to
4076 ignore detected IRQ/DMA settings because of
4077 possible conflicts). You can specify the base
4078 address, IRQ, and DMA settings; IRQ and DMA
4079 should be numbers, or 'auto' (for using detected
4080 settings on that particular port), or 'nofifo'
4081 (to avoid using a FIFO even if it is detected).
4082 Parallel ports are assigned in the order they
4083 are specified on the command line, starting
4086 parport_init_mode= [HW,PPT]
4087 Configure VIA parallel port to operate in
4088 a specific mode. This is necessary on Pegasos
4089 computer where firmware has no options for setting
4090 up parallel port mode and sets it to spp.
4091 Currently this function knows 686a and 8231 chips.
4092 Format: [spp|ps2|epp|ecp|ecpepp]
4094 pata_legacy.all= [HW,LIBATA]
4096 Set to non-zero to probe primary and secondary ISA
4097 port ranges on PCI systems where no PCI PATA device
4098 has been found at either range. Disabled by default.
4100 pata_legacy.autospeed= [HW,LIBATA]
4102 Set to non-zero if a chip is present that snoops speed
4103 changes. Disabled by default.
4105 pata_legacy.ht6560a= [HW,LIBATA]
4107 Set to 1, 2, or 3 for HT 6560A on the primary channel,
4108 the secondary channel, or both channels respectively.
4109 Disabled by default.
4111 pata_legacy.ht6560b= [HW,LIBATA]
4113 Set to 1, 2, or 3 for HT 6560B on the primary channel,
4114 the secondary channel, or both channels respectively.
4115 Disabled by default.
4117 pata_legacy.iordy_mask= [HW,LIBATA]
4119 IORDY enable mask. Set individual bits to allow IORDY
4120 for the respective channel. Bit 0 is for the first
4121 legacy channel handled by this driver, bit 1 is for
4122 the second channel, and so on. The sequence will often
4123 correspond to the primary legacy channel, the secondary
4124 legacy channel, and so on, but the handling of a PCI
4125 bus and the use of other driver options may interfere
4126 with the sequence. By default IORDY is allowed across
4129 pata_legacy.opti82c46x= [HW,LIBATA]
4131 Set to 1, 2, or 3 for Opti 82c611A on the primary
4132 channel, the secondary channel, or both channels
4133 respectively. Disabled by default.
4135 pata_legacy.opti82c611a= [HW,LIBATA]
4137 Set to 1, 2, or 3 for Opti 82c465MV on the primary
4138 channel, the secondary channel, or both channels
4139 respectively. Disabled by default.
4141 pata_legacy.pio_mask= [HW,LIBATA]
4143 PIO mode mask for autospeed devices. Set individual
4144 bits to allow the use of the respective PIO modes.
4145 Bit 0 is for mode 0, bit 1 is for mode 1, and so on.
4146 All modes allowed by default.
4148 pata_legacy.probe_all= [HW,LIBATA]
4150 Set to non-zero to probe tertiary and further ISA
4151 port ranges on PCI systems. Disabled by default.
4153 pata_legacy.probe_mask= [HW,LIBATA]
4155 Probe mask for legacy ISA PATA ports. Depending on
4156 platform configuration and the use of other driver
4157 options up to 6 legacy ports are supported: 0x1f0,
4158 0x170, 0x1e8, 0x168, 0x1e0, 0x160, however probing
4159 of individual ports can be disabled by setting the
4160 corresponding bits in the mask to 1. Bit 0 is for
4161 the first port in the list above (0x1f0), and so on.
4162 By default all supported ports are probed.
4164 pata_legacy.qdi= [HW,LIBATA]
4166 Set to non-zero to probe QDI controllers. By default
4167 set to 1 if CONFIG_PATA_QDI_MODULE, 0 otherwise.
4169 pata_legacy.winbond= [HW,LIBATA]
4171 Set to non-zero to probe Winbond controllers. Use
4172 the standard I/O port (0x130) if 1, otherwise the
4173 value given is the I/O port to use (typically 0x1b0).
4174 By default set to 1 if CONFIG_PATA_WINBOND_VLB_MODULE,
4177 pata_platform.pio_mask= [HW,LIBATA]
4179 Supported PIO mode mask. Set individual bits to allow
4180 the use of the respective PIO modes. Bit 0 is for
4181 mode 0, bit 1 is for mode 1, and so on. Mode 0 only
4185 Halt all CPUs after the first oops has been printed for
4186 the specified number of seconds. This is to be used if
4187 your oopses keep scrolling off the screen.
4191 pci=option[,option...] [PCI] various PCI subsystem options.
4193 Some options herein operate on a specific device
4194 or a set of devices (<pci_dev>). These are
4195 specified in one of the following formats:
4197 [<domain>:]<bus>:<dev>.<func>[/<dev>.<func>]*
4198 pci:<vendor>:<device>[:<subvendor>:<subdevice>]
4200 Note: the first format specifies a PCI
4201 bus/device/function address which may change
4202 if new hardware is inserted, if motherboard
4203 firmware changes, or due to changes caused
4204 by other kernel parameters. If the
4205 domain is left unspecified, it is
4206 taken to be zero. Optionally, a path
4207 to a device through multiple device/function
4208 addresses can be specified after the base
4209 address (this is more robust against
4210 renumbering issues). The second format
4211 selects devices using IDs from the
4212 configuration space which may match multiple
4213 devices in the system.
4215 earlydump dump PCI config space before the kernel
4217 off [X86] don't probe for the PCI bus
4218 bios [X86-32] force use of PCI BIOS, don't access
4219 the hardware directly. Use this if your machine
4220 has a non-standard PCI host bridge.
4221 nobios [X86-32] disallow use of PCI BIOS, only direct
4222 hardware access methods are allowed. Use this
4223 if you experience crashes upon bootup and you
4224 suspect they are caused by the BIOS.
4225 conf1 [X86] Force use of PCI Configuration Access
4226 Mechanism 1 (config address in IO port 0xCF8,
4227 data in IO port 0xCFC, both 32-bit).
4228 conf2 [X86] Force use of PCI Configuration Access
4229 Mechanism 2 (IO port 0xCF8 is an 8-bit port for
4230 the function, IO port 0xCFA, also 8-bit, sets
4231 bus number. The config space is then accessed
4232 through ports 0xC000-0xCFFF).
4233 See http://wiki.osdev.org/PCI for more info
4234 on the configuration access mechanisms.
4235 noaer [PCIE] If the PCIEAER kernel config parameter is
4236 enabled, this kernel boot option can be used to
4237 disable the use of PCIE advanced error reporting.
4238 nodomains [PCI] Disable support for multiple PCI
4239 root domains (aka PCI segments, in ACPI-speak).
4240 nommconf [X86] Disable use of MMCONFIG for PCI
4242 check_enable_amd_mmconf [X86] check for and enable
4243 properly configured MMIO access to PCI
4244 config space on AMD family 10h CPU
4245 nomsi [MSI] If the PCI_MSI kernel config parameter is
4246 enabled, this kernel boot option can be used to
4247 disable the use of MSI interrupts system-wide.
4248 noioapicquirk [APIC] Disable all boot interrupt quirks.
4249 Safety option to keep boot IRQs enabled. This
4250 should never be necessary.
4251 ioapicreroute [APIC] Enable rerouting of boot IRQs to the
4252 primary IO-APIC for bridges that cannot disable
4253 boot IRQs. This fixes a source of spurious IRQs
4254 when the system masks IRQs.
4255 noioapicreroute [APIC] Disable workaround that uses the
4256 boot IRQ equivalent of an IRQ that connects to
4257 a chipset where boot IRQs cannot be disabled.
4258 The opposite of ioapicreroute.
4259 biosirq [X86-32] Use PCI BIOS calls to get the interrupt
4260 routing table. These calls are known to be buggy
4261 on several machines and they hang the machine
4262 when used, but on other computers it's the only
4263 way to get the interrupt routing table. Try
4264 this option if the kernel is unable to allocate
4265 IRQs or discover secondary PCI buses on your
4267 rom [X86] Assign address space to expansion ROMs.
4268 Use with caution as certain devices share
4269 address decoders between ROMs and other
4271 norom [X86] Do not assign address space to
4272 expansion ROMs that do not already have
4273 BIOS assigned address ranges.
4274 nobar [X86] Do not assign address space to the
4275 BARs that weren't assigned by the BIOS.
4276 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be
4277 assigned automatically to PCI devices. You can
4278 make the kernel exclude IRQs of your ISA cards
4280 pirqaddr=0xAAAAA [X86] Specify the physical address
4281 of the PIRQ table (normally generated
4282 by the BIOS) if it is outside the
4283 F0000h-100000h range.
4284 lastbus=N [X86] Scan all buses thru bus #N. Can be
4285 useful if the kernel is unable to find your
4286 secondary buses and you want to tell it
4287 explicitly which ones they are.
4288 assign-busses [X86] Always assign all PCI bus
4289 numbers ourselves, overriding
4290 whatever the firmware may have done.
4291 usepirqmask [X86] Honor the possible IRQ mask stored
4292 in the BIOS $PIR table. This is needed on
4293 some systems with broken BIOSes, notably
4294 some HP Pavilion N5400 and Omnibook XE3
4295 notebooks. This will have no effect if ACPI
4296 IRQ routing is enabled.
4297 noacpi [X86] Do not use ACPI for IRQ routing
4298 or for PCI scanning.
4299 use_crs [X86] Use PCI host bridge window information
4300 from ACPI. On BIOSes from 2008 or later, this
4301 is enabled by default. If you need to use this,
4302 please report a bug.
4303 nocrs [X86] Ignore PCI host bridge windows from ACPI.
4304 If you need to use this, please report a bug.
4305 use_e820 [X86] Use E820 reservations to exclude parts of
4306 PCI host bridge windows. This is a workaround
4307 for BIOS defects in host bridge _CRS methods.
4308 If you need to use this, please report a bug to
4309 <linux-pci@vger.kernel.org>.
4310 no_e820 [X86] Ignore E820 reservations for PCI host
4311 bridge windows. This is the default on modern
4312 hardware. If you need to use this, please report
4313 a bug to <linux-pci@vger.kernel.org>.
4314 routeirq Do IRQ routing for all PCI devices.
4315 This is normally done in pci_enable_device(),
4316 so this option is a temporary workaround
4317 for broken drivers that don't call it.
4318 skip_isa_align [X86] do not align io start addr, so can
4319 handle more pci cards
4320 noearly [X86] Don't do any early type 1 scanning.
4321 This might help on some broken boards which
4322 machine check when some devices' config space
4323 is read. But various workarounds are disabled
4324 and some IOMMU drivers will not work.
4325 bfsort Sort PCI devices into breadth-first order.
4326 This sorting is done to get a device
4327 order compatible with older (<= 2.4) kernels.
4328 nobfsort Don't sort PCI devices into breadth-first order.
4329 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size)
4330 tuning and use the BIOS-configured MPS defaults.
4331 pcie_bus_safe Set every device's MPS to the largest value
4332 supported by all devices below the root complex.
4333 pcie_bus_perf Set device MPS to the largest allowable MPS
4334 based on its parent bus. Also set MRRS (Max
4335 Read Request Size) to the largest supported
4336 value (no larger than the MPS that the device
4337 or bus can support) for best performance.
4338 pcie_bus_peer2peer Set every device's MPS to 128B, which
4339 every device is guaranteed to support. This
4340 configuration allows peer-to-peer DMA between
4341 any pair of devices, possibly at the cost of
4342 reduced performance. This also guarantees
4343 that hot-added devices will work.
4344 cbiosize=nn[KMG] The fixed amount of bus space which is
4345 reserved for the CardBus bridge's IO window.
4346 The default value is 256 bytes.
4347 cbmemsize=nn[KMG] The fixed amount of bus space which is
4348 reserved for the CardBus bridge's memory
4349 window. The default value is 64 megabytes.
4352 [<order of align>@]<pci_dev>[; ...]
4353 Specifies alignment and device to reassign
4354 aligned memory resources. How to
4355 specify the device is described above.
4356 If <order of align> is not specified,
4357 PAGE_SIZE is used as alignment.
4358 A PCI-PCI bridge can be specified if resource
4359 windows need to be expanded.
4360 To specify the alignment for several
4361 instances of a device, the PCI vendor,
4362 device, subvendor, and subdevice may be
4363 specified, e.g., 12@pci:8086:9c22:103c:198f
4364 for 4096-byte alignment.
4365 ecrc= Enable/disable PCIe ECRC (transaction layer
4366 end-to-end CRC checking). Only effective if
4367 OS has native AER control (either granted by
4368 ACPI _OSC or forced via "pcie_ports=native")
4369 bios: Use BIOS/firmware settings. This is the
4373 hpiosize=nn[KMG] The fixed amount of bus space which is
4374 reserved for hotplug bridge's IO window.
4375 Default size is 256 bytes.
4376 hpmmiosize=nn[KMG] The fixed amount of bus space which is
4377 reserved for hotplug bridge's MMIO window.
4378 Default size is 2 megabytes.
4379 hpmmioprefsize=nn[KMG] The fixed amount of bus space which is
4380 reserved for hotplug bridge's MMIO_PREF window.
4381 Default size is 2 megabytes.
4382 hpmemsize=nn[KMG] The fixed amount of bus space which is
4383 reserved for hotplug bridge's MMIO and
4385 Default size is 2 megabytes.
4386 hpbussize=nn The minimum amount of additional bus numbers
4387 reserved for buses below a hotplug bridge.
4389 realloc= Enable/disable reallocating PCI bridge resources
4390 if allocations done by BIOS are too small to
4391 accommodate resources required by all child
4393 off: Turn realloc off
4395 realloc same as realloc=on
4396 noari do not use PCIe ARI.
4397 noats [PCIE, Intel-IOMMU, AMD-IOMMU]
4398 do not use PCIe ATS (and IOMMU device IOTLB).
4399 pcie_scan_all Scan all possible PCIe devices. Otherwise we
4400 only look for one device below a PCIe downstream
4402 big_root_window Try to add a big 64bit memory window to the PCIe
4403 root complex on AMD CPUs. Some GFX hardware
4404 can resize a BAR to allow access to all VRAM.
4405 Adding the window is slightly risky (it may
4406 conflict with unreported devices), so this
4408 disable_acs_redir=<pci_dev>[; ...]
4409 Specify one or more PCI devices (in the format
4410 specified above) separated by semicolons.
4411 Each device specified will have the PCI ACS
4412 redirect capabilities forced off which will
4413 allow P2P traffic between devices through
4414 bridges without forcing it upstream. Note:
4415 this removes isolation between devices and
4416 may put more devices in an IOMMU group.
4417 force_floating [S390] Force usage of floating interrupts.
4418 nomio [S390] Do not use MIO instructions.
4419 norid [S390] ignore the RID field and force use of
4420 one PCI domain per PCI function
4422 pcie_aspm= [PCIE] Forcibly enable or disable PCIe Active State Power
4425 force Enable ASPM even on devices that claim not to support it.
4426 WARNING: Forcing ASPM on may cause system lockups.
4428 pcie_ports= [PCIE] PCIe port services handling:
4429 native Use native PCIe services (PME, AER, DPC, PCIe hotplug)
4430 even if the platform doesn't give the OS permission to
4431 use them. This may cause conflicts if the platform
4432 also tries to use these services.
4433 dpc-native Use native PCIe service for DPC only. May
4434 cause conflicts if firmware uses AER or DPC.
4435 compat Disable native PCIe services (PME, AER, DPC, PCIe
4438 pcie_port_pm= [PCIE] PCIe port power management handling:
4439 off Disable power management of all PCIe ports
4440 force Forcibly enable power management of all PCIe ports
4442 pcie_pme= [PCIE,PM] Native PCIe PME signaling options:
4443 nomsi Do not use MSI for native PCIe PME signaling (this makes
4444 all PCIe root ports use INTx for all services).
4446 pcmv= [HW,PCMCIA] BadgePAD 4
4450 Keep all power-domains already enabled by bootloader on,
4451 even if no driver has claimed them. This is useful
4452 for debug and development, but should not be
4453 needed on a platform with proper driver support.
4455 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at
4458 See arch/parisc/kernel/pdc_chassis.c
4460 percpu_alloc= Select which percpu first chunk allocator to use.
4461 Currently supported values are "embed" and "page".
4462 Archs may support subset or none of the selections.
4463 See comments in mm/percpu.c for details on each
4464 allocator. This parameter is primarily for debugging
4465 and performance comparison.
4467 pirq= [SMP,APIC] Manual mp-table setup
4468 See Documentation/arch/x86/i386/IO-APIC.rst.
4470 plip= [PPT,NET] Parallel port network link
4471 Format: { parport<nr> | timid | 0 }
4472 See also Documentation/admin-guide/parport.rst.
4474 pmtmr= [X86] Manual setup of pmtmr I/O Port.
4475 Override pmtimer IOPort with a hex value.
4478 pmu_override= [PPC] Override the PMU.
4479 This option takes over the PMU facility, so it is no
4480 longer usable by perf. Setting this option starts the
4481 PMU counters by setting MMCR0 to 0 (the FC bit is
4482 cleared). If a number is given, then MMCR1 is set to
4483 that number, otherwise (e.g., 'pmu_override=on'), MMCR1
4486 pm_debug_messages [SUSPEND,KNL]
4487 Enable suspend/resume debug messages during boot up.
4490 Enable PNP debug messages (depends on the
4491 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time
4492 via /sys/module/pnp/parameters/debug. We always show
4493 current resource usage; turning this on also shows
4494 possible settings and some assignment information.
4500 { on | off | curr | res | no-curr | no-res }
4503 [ISAPNP] Exclude IRQs for the autoconfiguration
4506 [ISAPNP] Exclude DMAs for the autoconfiguration
4508 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration
4509 Ranges are in pairs (I/O port base and size).
4512 [ISAPNP] Exclude memory regions for the
4514 Ranges are in pairs (memory base and size).
4516 ports= [IP_VS_FTP] IPVS ftp helper module
4518 Up to 8 (IP_VS_APP_MAX_PORTS) ports
4520 Format: <port>,<port>....
4522 powersave=off [PPC] This option disables power saving features.
4523 It specifically disables cpuidle and sets the
4524 platform machine description specific power_save
4525 function to NULL. On Idle the CPU just reduces
4528 ppc_strict_facility_enable
4529 [PPC] This option catches any kernel floating point,
4530 Altivec, VSX and SPE outside of regions specifically
4531 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()).
4532 There is some performance impact when enabling this.
4536 Disable Hardware Transactional Memory
4539 Select preemption mode if you have CONFIG_PREEMPT_DYNAMIC
4540 none - Limited to cond_resched() calls
4541 voluntary - Limited to cond_resched() and might_sleep() calls
4542 full - Any section that isn't explicitly preempt disabled
4543 can be preempted anytime.
4545 print-fatal-signals=
4546 [KNL] debug: print fatal signals
4548 If enabled, warn about various signal handling
4549 related application anomalies: too many signals,
4550 too many POSIX.1 timers, fatal signals causing a
4553 If you hit the warning due to signal overflow,
4554 you might want to try "ulimit -i unlimited".
4558 printk.always_kmsg_dump=
4559 Trigger kmsg_dump for cases other than kernel oops or
4561 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
4564 printk.console_no_auto_verbose=
4565 Disable console loglevel raise on oops, panic
4566 or lockdep-detected issues (only if lock debug is on).
4567 With an exception to setups with low baudrate on
4568 serial console, keeping this 0 is a good choice
4569 in order to provide more debug information.
4571 default: 0 (auto_verbose is enabled)
4573 printk.devkmsg={on,off,ratelimit}
4574 Control writing to /dev/kmsg.
4575 on - unlimited logging to /dev/kmsg from userspace
4576 off - logging to /dev/kmsg disabled
4577 ratelimit - ratelimit the logging
4580 printk.time= Show timing data prefixed to each printk message line
4581 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
4583 processor.max_cstate= [HW,ACPI]
4584 Limit processor to maximum C-state
4585 max_cstate=9 overrides any DMI blacklist limit.
4587 processor.nocst [HW,ACPI]
4588 Ignore the _CST method to determine C-states,
4589 instead using the legacy FADT method
4591 profile= [KNL] Enable kernel profiling via /proc/profile
4592 Format: [<profiletype>,]<number>
4593 Param: <profiletype>: "schedule", "sleep", or "kvm"
4594 [defaults to kernel profiling]
4595 Param: "schedule" - profile schedule points.
4596 Param: "sleep" - profile D-state sleeping (millisecs).
4597 Requires CONFIG_SCHEDSTATS
4598 Param: "kvm" - profile VM exits.
4599 Param: <number> - step/bucket size as a power of 2 for
4600 statistical time based profiling.
4602 prompt_ramdisk= [RAM] [Deprecated]
4604 prot_virt= [S390] enable hosting protected virtual machines
4605 isolated from the hypervisor (if hardware supports
4609 psi= [KNL] Enable or disable pressure stall information
4613 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to
4614 probe for; one of (bare|imps|exps|lifebook|any).
4615 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports
4617 psmouse.resetafter= [HW,MOUSE]
4618 Try to reset the device after so many bad packets
4621 [HW,MOUSE] Set desired mouse resolution, in dpi.
4622 psmouse.smartscroll=
4623 [HW,MOUSE] Controls Logitech smartscroll autorepeat.
4624 0 = disabled, 1 = enabled (default).
4626 pstore.backend= Specify the name of the pstore backend to use
4628 pti= [X86-64] Control Page Table Isolation of user and
4629 kernel address spaces. Disabling this feature
4630 removes hardening, but improves performance of
4631 system calls and interrupts.
4633 on - unconditionally enable
4634 off - unconditionally disable
4635 auto - kernel detects whether your CPU model is
4636 vulnerable to issues that PTI mitigates
4638 Not specifying this option is equivalent to pti=auto.
4641 [KNL] Number of legacy pty's. Overwrites compiled-in
4644 quiet [KNL] Disable most log messages
4648 radix_hcall_invalidate=on [PPC/PSERIES]
4649 Disable RADIX GTSE feature and use hcall for TLB
4653 See Documentation/admin-guide/md.rst.
4655 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes
4656 See Documentation/admin-guide/blockdev/ramdisk.rst.
4658 ramdisk_start= [RAM] RAM disk image start address
4660 random.trust_cpu=off
4661 [KNL] Disable trusting the use of the CPU's
4662 random number generator (if available) to
4663 initialize the kernel's RNG.
4665 random.trust_bootloader=off
4666 [KNL] Disable trusting the use of the a seed
4667 passed by the bootloader (if available) to
4668 initialize the kernel's RNG.
4670 randomize_kstack_offset=
4671 [KNL] Enable or disable kernel stack offset
4672 randomization, which provides roughly 5 bits of
4673 entropy, frustrating memory corruption attacks
4674 that depend on stack address determinism or
4675 cross-syscall address exposures. This is only
4676 available on architectures that have defined
4677 CONFIG_HAVE_ARCH_RANDOMIZE_KSTACK_OFFSET.
4678 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
4679 Default is CONFIG_RANDOMIZE_KSTACK_OFFSET_DEFAULT.
4681 ras=option[,option,...] [KNL] RAS-specific options
4684 Disable the Correctable Errors Collector,
4685 see CONFIG_RAS_CEC help text.
4687 rcu_nocbs[=cpu-list]
4688 [KNL] The optional argument is a cpu list,
4691 In kernels built with CONFIG_RCU_NOCB_CPU=y,
4692 enable the no-callback CPU mode, which prevents
4693 such CPUs' callbacks from being invoked in
4694 softirq context. Invocation of such CPUs' RCU
4695 callbacks will instead be offloaded to "rcuox/N"
4696 kthreads created for that purpose, where "x" is
4697 "p" for RCU-preempt, "s" for RCU-sched, and "g"
4698 for the kthreads that mediate grace periods; and
4699 "N" is the CPU number. This reduces OS jitter on
4700 the offloaded CPUs, which can be useful for HPC
4701 and real-time workloads. It can also improve
4702 energy efficiency for asymmetric multiprocessors.
4704 If a cpulist is passed as an argument, the specified
4705 list of CPUs is set to no-callback mode from boot.
4707 Otherwise, if the '=' sign and the cpulist
4708 arguments are omitted, no CPU will be set to
4709 no-callback mode from boot but the mode may be
4710 toggled at runtime via cpusets.
4712 Note that this argument takes precedence over
4713 the CONFIG_RCU_NOCB_CPU_DEFAULT_ALL option.
4716 Rather than requiring that offloaded CPUs
4717 (specified by rcu_nocbs= above) explicitly
4718 awaken the corresponding "rcuoN" kthreads,
4719 make these kthreads poll for callbacks.
4720 This improves the real-time response for the
4721 offloaded CPUs by relieving them of the need to
4722 wake up the corresponding kthread, but degrades
4723 energy efficiency by requiring that the kthreads
4724 periodically wake up to do the polling.
4726 rcutree.blimit= [KNL]
4727 Set maximum number of finished RCU callbacks to
4728 process in one batch.
4730 rcutree.dump_tree= [KNL]
4731 Dump the structure of the rcu_node combining tree
4732 out at early boot. This is used for diagnostic
4733 purposes, to verify correct tree setup.
4735 rcutree.gp_cleanup_delay= [KNL]
4736 Set the number of jiffies to delay each step of
4737 RCU grace-period cleanup.
4739 rcutree.gp_init_delay= [KNL]
4740 Set the number of jiffies to delay each step of
4741 RCU grace-period initialization.
4743 rcutree.gp_preinit_delay= [KNL]
4744 Set the number of jiffies to delay each step of
4745 RCU grace-period pre-initialization, that is,
4746 the propagation of recent CPU-hotplug changes up
4747 the rcu_node combining tree.
4749 rcutree.jiffies_till_first_fqs= [KNL]
4750 Set delay from grace-period initialization to
4751 first attempt to force quiescent states.
4752 Units are jiffies, minimum value is zero,
4753 and maximum value is HZ.
4755 rcutree.jiffies_till_next_fqs= [KNL]
4756 Set delay between subsequent attempts to force
4757 quiescent states. Units are jiffies, minimum
4758 value is one, and maximum value is HZ.
4760 rcutree.jiffies_till_sched_qs= [KNL]
4761 Set required age in jiffies for a
4762 given grace period before RCU starts
4763 soliciting quiescent-state help from
4764 rcu_note_context_switch() and cond_resched().
4765 If not specified, the kernel will calculate
4766 a value based on the most recent settings
4767 of rcutree.jiffies_till_first_fqs
4768 and rcutree.jiffies_till_next_fqs.
4769 This calculated value may be viewed in
4770 rcutree.jiffies_to_sched_qs. Any attempt to set
4771 rcutree.jiffies_to_sched_qs will be cheerfully
4774 rcutree.kthread_prio= [KNL,BOOT]
4775 Set the SCHED_FIFO priority of the RCU per-CPU
4776 kthreads (rcuc/N). This value is also used for
4777 the priority of the RCU boost threads (rcub/N)
4778 and for the RCU grace-period kthreads (rcu_bh,
4779 rcu_preempt, and rcu_sched). If RCU_BOOST is
4780 set, valid values are 1-99 and the default is 1
4781 (the least-favored priority). Otherwise, when
4782 RCU_BOOST is not set, valid values are 0-99 and
4783 the default is zero (non-realtime operation).
4784 When RCU_NOCB_CPU is set, also adjust the
4785 priority of NOCB callback kthreads.
4787 rcutree.nocb_nobypass_lim_per_jiffy= [KNL]
4788 On callback-offloaded (rcu_nocbs) CPUs,
4789 RCU reduces the lock contention that would
4790 otherwise be caused by callback floods through
4791 use of the ->nocb_bypass list. However, in the
4792 common non-flooded case, RCU queues directly to
4793 the main ->cblist in order to avoid the extra
4794 overhead of the ->nocb_bypass list and its lock.
4795 But if there are too many callbacks queued during
4796 a single jiffy, RCU pre-queues the callbacks into
4797 the ->nocb_bypass queue. The definition of "too
4798 many" is supplied by this kernel boot parameter.
4800 rcutree.qhimark= [KNL]
4801 Set threshold of queued RCU callbacks beyond which
4802 batch limiting is disabled.
4804 rcutree.qlowmark= [KNL]
4805 Set threshold of queued RCU callbacks below which
4806 batch limiting is re-enabled.
4808 rcutree.qovld= [KNL]
4809 Set threshold of queued RCU callbacks beyond which
4810 RCU's force-quiescent-state scan will aggressively
4811 enlist help from cond_resched() and sched IPIs to
4812 help CPUs more quickly reach quiescent states.
4813 Set to less than zero to make this be set based
4814 on rcutree.qhimark at boot time and to zero to
4815 disable more aggressive help enlistment.
4817 rcutree.rcu_delay_page_cache_fill_msec= [KNL]
4818 Set the page-cache refill delay (in milliseconds)
4819 in response to low-memory conditions. The range
4820 of permitted values is in the range 0:100000.
4822 rcutree.rcu_divisor= [KNL]
4823 Set the shift-right count to use to compute
4824 the callback-invocation batch limit bl from
4825 the number of callbacks queued on this CPU.
4826 The result will be bounded below by the value of
4827 the rcutree.blimit kernel parameter. Every bl
4828 callbacks, the softirq handler will exit in
4829 order to allow the CPU to do other work.
4831 Please note that this callback-invocation batch
4832 limit applies only to non-offloaded callback
4833 invocation. Offloaded callbacks are instead
4834 invoked in the context of an rcuoc kthread, which
4835 scheduler will preempt as it does any other task.
4837 rcutree.rcu_fanout_exact= [KNL]
4838 Disable autobalancing of the rcu_node combining
4839 tree. This is used by rcutorture, and might
4840 possibly be useful for architectures having high
4841 cache-to-cache transfer latencies.
4843 rcutree.rcu_fanout_leaf= [KNL]
4844 Change the number of CPUs assigned to each
4845 leaf rcu_node structure. Useful for very
4846 large systems, which will choose the value 64,
4847 and for NUMA systems with large remote-access
4848 latencies, which will choose a value aligned
4849 with the appropriate hardware boundaries.
4851 rcutree.rcu_min_cached_objs= [KNL]
4852 Minimum number of objects which are cached and
4853 maintained per one CPU. Object size is equal
4854 to PAGE_SIZE. The cache allows to reduce the
4855 pressure to page allocator, also it makes the
4856 whole algorithm to behave better in low memory
4859 rcutree.rcu_nocb_gp_stride= [KNL]
4860 Set the number of NOCB callback kthreads in
4861 each group, which defaults to the square root
4862 of the number of CPUs. Larger numbers reduce
4863 the wakeup overhead on the global grace-period
4864 kthread, but increases that same overhead on
4865 each group's NOCB grace-period kthread.
4867 rcutree.rcu_kick_kthreads= [KNL]
4868 Cause the grace-period kthread to get an extra
4869 wake_up() if it sleeps three times longer than
4870 it should at force-quiescent-state time.
4871 This wake_up() will be accompanied by a
4872 WARN_ONCE() splat and an ftrace_dump().
4874 rcutree.rcu_resched_ns= [KNL]
4875 Limit the time spend invoking a batch of RCU
4876 callbacks to the specified number of nanoseconds.
4877 By default, this limit is checked only once
4878 every 32 callbacks in order to limit the pain
4879 inflicted by local_clock() overhead.
4881 rcutree.rcu_unlock_delay= [KNL]
4882 In CONFIG_RCU_STRICT_GRACE_PERIOD=y kernels,
4883 this specifies an rcu_read_unlock()-time delay
4884 in microseconds. This defaults to zero.
4885 Larger delays increase the probability of
4886 catching RCU pointer leaks, that is, buggy use
4887 of RCU-protected pointers after the relevant
4888 rcu_read_unlock() has completed.
4890 rcutree.sysrq_rcu= [KNL]
4891 Commandeer a sysrq key to dump out Tree RCU's
4892 rcu_node tree with an eye towards determining
4893 why a new grace period has not yet started.
4895 rcutree.use_softirq= [KNL]
4896 If set to zero, move all RCU_SOFTIRQ processing to
4897 per-CPU rcuc kthreads. Defaults to a non-zero
4898 value, meaning that RCU_SOFTIRQ is used by default.
4899 Specify rcutree.use_softirq=0 to use rcuc kthreads.
4901 But note that CONFIG_PREEMPT_RT=y kernels disable
4902 this kernel boot parameter, forcibly setting it
4905 rcuscale.gp_async= [KNL]
4906 Measure performance of asynchronous
4907 grace-period primitives such as call_rcu().
4909 rcuscale.gp_async_max= [KNL]
4910 Specify the maximum number of outstanding
4911 callbacks per writer thread. When a writer
4912 thread exceeds this limit, it invokes the
4913 corresponding flavor of rcu_barrier() to allow
4914 previously posted callbacks to drain.
4916 rcuscale.gp_exp= [KNL]
4917 Measure performance of expedited synchronous
4918 grace-period primitives.
4920 rcuscale.holdoff= [KNL]
4921 Set test-start holdoff period. The purpose of
4922 this parameter is to delay the start of the
4923 test until boot completes in order to avoid
4926 rcuscale.kfree_rcu_test= [KNL]
4927 Set to measure performance of kfree_rcu() flooding.
4929 rcuscale.kfree_rcu_test_double= [KNL]
4930 Test the double-argument variant of kfree_rcu().
4931 If this parameter has the same value as
4932 rcuscale.kfree_rcu_test_single, both the single-
4933 and double-argument variants are tested.
4935 rcuscale.kfree_rcu_test_single= [KNL]
4936 Test the single-argument variant of kfree_rcu().
4937 If this parameter has the same value as
4938 rcuscale.kfree_rcu_test_double, both the single-
4939 and double-argument variants are tested.
4941 rcuscale.kfree_nthreads= [KNL]
4942 The number of threads running loops of kfree_rcu().
4944 rcuscale.kfree_alloc_num= [KNL]
4945 Number of allocations and frees done in an iteration.
4947 rcuscale.kfree_loops= [KNL]
4948 Number of loops doing rcuscale.kfree_alloc_num number
4949 of allocations and frees.
4951 rcuscale.nreaders= [KNL]
4952 Set number of RCU readers. The value -1 selects
4953 N, where N is the number of CPUs. A value
4954 "n" less than -1 selects N-n+1, where N is again
4955 the number of CPUs. For example, -2 selects N
4956 (the number of CPUs), -3 selects N+1, and so on.
4957 A value of "n" less than or equal to -N selects
4960 rcuscale.nwriters= [KNL]
4961 Set number of RCU writers. The values operate
4962 the same as for rcuscale.nreaders.
4963 N, where N is the number of CPUs
4965 rcuscale.perf_type= [KNL]
4966 Specify the RCU implementation to test.
4968 rcuscale.shutdown= [KNL]
4969 Shut the system down after performance tests
4970 complete. This is useful for hands-off automated
4973 rcuscale.verbose= [KNL]
4974 Enable additional printk() statements.
4976 rcuscale.writer_holdoff= [KNL]
4977 Write-side holdoff between grace periods,
4978 in microseconds. The default of zero says
4981 rcutorture.fqs_duration= [KNL]
4982 Set duration of force_quiescent_state bursts
4985 rcutorture.fqs_holdoff= [KNL]
4986 Set holdoff time within force_quiescent_state bursts
4989 rcutorture.fqs_stutter= [KNL]
4990 Set wait time between force_quiescent_state bursts
4993 rcutorture.fwd_progress= [KNL]
4994 Specifies the number of kthreads to be used
4995 for RCU grace-period forward-progress testing
4996 for the types of RCU supporting this notion.
4997 Defaults to 1 kthread, values less than zero or
4998 greater than the number of CPUs cause the number
5001 rcutorture.fwd_progress_div= [KNL]
5002 Specify the fraction of a CPU-stall-warning
5003 period to do tight-loop forward-progress testing.
5005 rcutorture.fwd_progress_holdoff= [KNL]
5006 Number of seconds to wait between successive
5007 forward-progress tests.
5009 rcutorture.fwd_progress_need_resched= [KNL]
5010 Enclose cond_resched() calls within checks for
5011 need_resched() during tight-loop forward-progress
5014 rcutorture.gp_cond= [KNL]
5015 Use conditional/asynchronous update-side
5016 primitives, if available.
5018 rcutorture.gp_exp= [KNL]
5019 Use expedited update-side primitives, if available.
5021 rcutorture.gp_normal= [KNL]
5022 Use normal (non-expedited) asynchronous
5023 update-side primitives, if available.
5025 rcutorture.gp_sync= [KNL]
5026 Use normal (non-expedited) synchronous
5027 update-side primitives, if available. If all
5028 of rcutorture.gp_cond=, rcutorture.gp_exp=,
5029 rcutorture.gp_normal=, and rcutorture.gp_sync=
5030 are zero, rcutorture acts as if is interpreted
5031 they are all non-zero.
5033 rcutorture.irqreader= [KNL]
5034 Run RCU readers from irq handlers, or, more
5035 accurately, from a timer handler. Not all RCU
5036 flavors take kindly to this sort of thing.
5038 rcutorture.leakpointer= [KNL]
5039 Leak an RCU-protected pointer out of the reader.
5040 This can of course result in splats, and is
5041 intended to test the ability of things like
5042 CONFIG_RCU_STRICT_GRACE_PERIOD=y to detect
5045 rcutorture.n_barrier_cbs= [KNL]
5046 Set callbacks/threads for rcu_barrier() testing.
5048 rcutorture.nfakewriters= [KNL]
5049 Set number of concurrent RCU writers. These just
5050 stress RCU, they don't participate in the actual
5051 test, hence the "fake".
5053 rcutorture.nocbs_nthreads= [KNL]
5054 Set number of RCU callback-offload togglers.
5055 Zero (the default) disables toggling.
5057 rcutorture.nocbs_toggle= [KNL]
5058 Set the delay in milliseconds between successive
5059 callback-offload toggling attempts.
5061 rcutorture.nreaders= [KNL]
5062 Set number of RCU readers. The value -1 selects
5063 N-1, where N is the number of CPUs. A value
5064 "n" less than -1 selects N-n-2, where N is again
5065 the number of CPUs. For example, -2 selects N
5066 (the number of CPUs), -3 selects N+1, and so on.
5068 rcutorture.object_debug= [KNL]
5069 Enable debug-object double-call_rcu() testing.
5071 rcutorture.onoff_holdoff= [KNL]
5072 Set time (s) after boot for CPU-hotplug testing.
5074 rcutorture.onoff_interval= [KNL]
5075 Set time (jiffies) between CPU-hotplug operations,
5076 or zero to disable CPU-hotplug testing.
5078 rcutorture.read_exit= [KNL]
5079 Set the number of read-then-exit kthreads used
5080 to test the interaction of RCU updaters and
5081 task-exit processing.
5083 rcutorture.read_exit_burst= [KNL]
5084 The number of times in a given read-then-exit
5085 episode that a set of read-then-exit kthreads
5088 rcutorture.read_exit_delay= [KNL]
5089 The delay, in seconds, between successive
5090 read-then-exit testing episodes.
5092 rcutorture.shuffle_interval= [KNL]
5093 Set task-shuffle interval (s). Shuffling tasks
5094 allows some CPUs to go into dyntick-idle mode
5095 during the rcutorture test.
5097 rcutorture.shutdown_secs= [KNL]
5098 Set time (s) after boot system shutdown. This
5099 is useful for hands-off automated testing.
5101 rcutorture.stall_cpu= [KNL]
5102 Duration of CPU stall (s) to test RCU CPU stall
5103 warnings, zero to disable.
5105 rcutorture.stall_cpu_block= [KNL]
5106 Sleep while stalling if set. This will result
5107 in warnings from preemptible RCU in addition to
5108 any other stall-related activity. Note that
5109 in kernels built with CONFIG_PREEMPTION=n and
5110 CONFIG_PREEMPT_COUNT=y, this parameter will
5111 cause the CPU to pass through a quiescent state.
5112 Given CONFIG_PREEMPTION=n, this will suppress
5113 RCU CPU stall warnings, but will instead result
5114 in scheduling-while-atomic splats.
5116 Use of this module parameter results in splats.
5119 rcutorture.stall_cpu_holdoff= [KNL]
5120 Time to wait (s) after boot before inducing stall.
5122 rcutorture.stall_cpu_irqsoff= [KNL]
5123 Disable interrupts while stalling if set.
5125 rcutorture.stall_gp_kthread= [KNL]
5126 Duration (s) of forced sleep within RCU
5127 grace-period kthread to test RCU CPU stall
5128 warnings, zero to disable. If both stall_cpu
5129 and stall_gp_kthread are specified, the
5130 kthread is starved first, then the CPU.
5132 rcutorture.stat_interval= [KNL]
5133 Time (s) between statistics printk()s.
5135 rcutorture.stutter= [KNL]
5136 Time (s) to stutter testing, for example, specifying
5137 five seconds causes the test to run for five seconds,
5138 wait for five seconds, and so on. This tests RCU's
5139 ability to transition abruptly to and from idle.
5141 rcutorture.test_boost= [KNL]
5142 Test RCU priority boosting? 0=no, 1=maybe, 2=yes.
5143 "Maybe" means test if the RCU implementation
5144 under test support RCU priority boosting.
5146 rcutorture.test_boost_duration= [KNL]
5147 Duration (s) of each individual boost test.
5149 rcutorture.test_boost_interval= [KNL]
5150 Interval (s) between each boost test.
5152 rcutorture.test_no_idle_hz= [KNL]
5153 Test RCU's dyntick-idle handling. See also the
5154 rcutorture.shuffle_interval parameter.
5156 rcutorture.torture_type= [KNL]
5157 Specify the RCU implementation to test.
5159 rcutorture.verbose= [KNL]
5160 Enable additional printk() statements.
5162 rcupdate.rcu_cpu_stall_ftrace_dump= [KNL]
5163 Dump ftrace buffer after reporting RCU CPU
5166 rcupdate.rcu_cpu_stall_suppress= [KNL]
5167 Suppress RCU CPU stall warning messages.
5169 rcupdate.rcu_cpu_stall_suppress_at_boot= [KNL]
5170 Suppress RCU CPU stall warning messages and
5171 rcutorture writer stall warnings that occur
5172 during early boot, that is, during the time
5173 before the init task is spawned.
5175 rcupdate.rcu_cpu_stall_timeout= [KNL]
5176 Set timeout for RCU CPU stall warning messages.
5177 The value is in seconds and the maximum allowed
5178 value is 300 seconds.
5180 rcupdate.rcu_exp_cpu_stall_timeout= [KNL]
5181 Set timeout for expedited RCU CPU stall warning
5182 messages. The value is in milliseconds
5183 and the maximum allowed value is 21000
5184 milliseconds. Please note that this value is
5185 adjusted to an arch timer tick resolution.
5186 Setting this to zero causes the value from
5187 rcupdate.rcu_cpu_stall_timeout to be used (after
5188 conversion from seconds to milliseconds).
5190 rcupdate.rcu_cpu_stall_cputime= [KNL]
5191 Provide statistics on the cputime and count of
5192 interrupts and tasks during the sampling period. For
5193 multiple continuous RCU stalls, all sampling periods
5194 begin at half of the first RCU stall timeout.
5196 rcupdate.rcu_exp_stall_task_details= [KNL]
5197 Print stack dumps of any tasks blocking the
5198 current expedited RCU grace period during an
5199 expedited RCU CPU stall warning.
5201 rcupdate.rcu_expedited= [KNL]
5202 Use expedited grace-period primitives, for
5203 example, synchronize_rcu_expedited() instead
5204 of synchronize_rcu(). This reduces latency,
5205 but can increase CPU utilization, degrade
5206 real-time latency, and degrade energy efficiency.
5207 No effect on CONFIG_TINY_RCU kernels.
5209 rcupdate.rcu_normal= [KNL]
5210 Use only normal grace-period primitives,
5211 for example, synchronize_rcu() instead of
5212 synchronize_rcu_expedited(). This improves
5213 real-time latency, CPU utilization, and
5214 energy efficiency, but can expose users to
5215 increased grace-period latency. This parameter
5216 overrides rcupdate.rcu_expedited. No effect on
5217 CONFIG_TINY_RCU kernels.
5219 rcupdate.rcu_normal_after_boot= [KNL]
5220 Once boot has completed (that is, after
5221 rcu_end_inkernel_boot() has been invoked), use
5222 only normal grace-period primitives. No effect
5223 on CONFIG_TINY_RCU kernels.
5225 But note that CONFIG_PREEMPT_RT=y kernels enables
5226 this kernel boot parameter, forcibly setting
5227 it to the value one, that is, converting any
5228 post-boot attempt at an expedited RCU grace
5229 period to instead use normal non-expedited
5230 grace-period processing.
5232 rcupdate.rcu_task_collapse_lim= [KNL]
5233 Set the maximum number of callbacks present
5234 at the beginning of a grace period that allows
5235 the RCU Tasks flavors to collapse back to using
5236 a single callback queue. This switching only
5237 occurs when rcupdate.rcu_task_enqueue_lim is
5238 set to the default value of -1.
5240 rcupdate.rcu_task_contend_lim= [KNL]
5241 Set the minimum number of callback-queuing-time
5242 lock-contention events per jiffy required to
5243 cause the RCU Tasks flavors to switch to per-CPU
5244 callback queuing. This switching only occurs
5245 when rcupdate.rcu_task_enqueue_lim is set to
5246 the default value of -1.
5248 rcupdate.rcu_task_enqueue_lim= [KNL]
5249 Set the number of callback queues to use for the
5250 RCU Tasks family of RCU flavors. The default
5251 of -1 allows this to be automatically (and
5252 dynamically) adjusted. This parameter is intended
5255 rcupdate.rcu_task_ipi_delay= [KNL]
5256 Set time in jiffies during which RCU tasks will
5257 avoid sending IPIs, starting with the beginning
5258 of a given grace period. Setting a large
5259 number avoids disturbing real-time workloads,
5260 but lengthens grace periods.
5262 rcupdate.rcu_task_stall_info= [KNL]
5263 Set initial timeout in jiffies for RCU task stall
5264 informational messages, which give some indication
5265 of the problem for those not patient enough to
5266 wait for ten minutes. Informational messages are
5267 only printed prior to the stall-warning message
5268 for a given grace period. Disable with a value
5269 less than or equal to zero. Defaults to ten
5270 seconds. A change in value does not take effect
5271 until the beginning of the next grace period.
5273 rcupdate.rcu_task_stall_info_mult= [KNL]
5274 Multiplier for time interval between successive
5275 RCU task stall informational messages for a given
5276 RCU tasks grace period. This value is clamped
5277 to one through ten, inclusive. It defaults to
5278 the value three, so that the first informational
5279 message is printed 10 seconds into the grace
5280 period, the second at 40 seconds, the third at
5281 160 seconds, and then the stall warning at 600
5282 seconds would prevent a fourth at 640 seconds.
5284 rcupdate.rcu_task_stall_timeout= [KNL]
5285 Set timeout in jiffies for RCU task stall
5286 warning messages. Disable with a value less
5287 than or equal to zero. Defaults to ten minutes.
5288 A change in value does not take effect until
5289 the beginning of the next grace period.
5291 rcupdate.rcu_self_test= [KNL]
5292 Run the RCU early boot self tests
5296 Run specified binary instead of /init from the ramdisk,
5297 used for early userspace startup. See initrd.
5300 force - Override the decision by the kernel to hide the
5301 advertisement of RDRAND support (this affects
5302 certain AMD processors because of buggy BIOS
5303 support, specifically around the suspend/resume
5307 Turn on/off individual RDT features. List is:
5308 cmt, mbmtotal, mbmlocal, l3cat, l3cdp, l2cat, l2cdp,
5310 E.g. to turn on cmt and turn off mba use:
5314 Format (x86 or x86_64):
5315 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] | d[efault] \
5317 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \
5319 Where reboot_mode is one of warm (soft) or cold (hard) or gpio
5320 (prefix with 'panic_' to set mode for panic
5322 reboot_type is one of bios, acpi, kbd, triple, efi, or pci,
5323 reboot_force is either force or not specified,
5324 reboot_cpu is s[mp]#### with #### being the processor
5325 to be used for rebooting.
5327 refscale.holdoff= [KNL]
5328 Set test-start holdoff period. The purpose of
5329 this parameter is to delay the start of the
5330 test until boot completes in order to avoid
5333 refscale.loops= [KNL]
5334 Set the number of loops over the synchronization
5335 primitive under test. Increasing this number
5336 reduces noise due to loop start/end overhead,
5337 but the default has already reduced the per-pass
5338 noise to a handful of picoseconds on ca. 2020
5341 refscale.nreaders= [KNL]
5342 Set number of readers. The default value of -1
5343 selects N, where N is roughly 75% of the number
5344 of CPUs. A value of zero is an interesting choice.
5346 refscale.nruns= [KNL]
5347 Set number of runs, each of which is dumped onto
5350 refscale.readdelay= [KNL]
5351 Set the read-side critical-section duration,
5352 measured in microseconds.
5354 refscale.scale_type= [KNL]
5355 Specify the read-protection implementation to test.
5357 refscale.shutdown= [KNL]
5358 Shut down the system at the end of the performance
5359 test. This defaults to 1 (shut it down) when
5360 refscale is built into the kernel and to 0 (leave
5361 it running) when refscale is built as a module.
5363 refscale.verbose= [KNL]
5364 Enable additional printk() statements.
5366 refscale.verbose_batched= [KNL]
5367 Batch the additional printk() statements. If zero
5368 (the default) or negative, print everything. Otherwise,
5369 print every Nth verbose statement, where N is the value
5373 [KNL, SMP] Set scheduler's default relax_domain_level.
5374 See Documentation/admin-guide/cgroup-v1/cpusets.rst.
5376 reserve= [KNL,BUGS] Force kernel to ignore I/O ports or memory
5377 Format: <base1>,<size1>[,<base2>,<size2>,...]
5378 Reserve I/O ports or memory so the kernel won't use
5379 them. If <base> is less than 0x10000, the region
5380 is assumed to be I/O ports; otherwise it is memory.
5382 reservetop= [X86-32]
5384 Reserves a hole at the top of the kernel virtual
5387 reset_devices [KNL] Force drivers to reset the underlying device
5388 during initialization.
5391 Specify the partition device for software suspend
5393 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>}
5395 resume_offset= [SWSUSP]
5396 Specify the offset from the beginning of the partition
5397 given by "resume=" at which the swap header is located,
5398 in <PAGE_SIZE> units (needed only for swap files).
5399 See Documentation/power/swsusp-and-swap-files.rst
5401 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to
5402 read the resume files
5404 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up.
5405 Useful for devices that are detected asynchronously
5406 (e.g. USB and MMC devices).
5408 retain_initrd [RAM] Keep initrd memory after extraction
5410 retbleed= [X86] Control mitigation of RETBleed (Arbitrary
5411 Speculative Code Execution with Return Instructions)
5414 AMD-based UNRET and IBPB mitigations alone do not stop
5415 sibling threads from influencing the predictions of other
5416 sibling threads. For that reason, STIBP is used on pro-
5417 cessors that support it, and mitigate SMT on processors
5421 auto - automatically select a migitation
5422 auto,nosmt - automatically select a mitigation,
5423 disabling SMT if necessary for
5424 the full mitigation (only on Zen1
5425 and older without STIBP).
5426 ibpb - On AMD, mitigate short speculation
5427 windows on basic block boundaries too.
5428 Safe, highest perf impact. It also
5429 enables STIBP if present. Not suitable
5431 ibpb,nosmt - Like "ibpb" above but will disable SMT
5432 when STIBP is not available. This is
5433 the alternative for systems which do not
5435 unret - Force enable untrained return thunks,
5436 only effective on AMD f15h-f17h based
5438 unret,nosmt - Like unret, but will disable SMT when STIBP
5439 is not available. This is the alternative for
5440 systems which do not have STIBP.
5442 Selecting 'auto' will choose a mitigation method at run
5443 time according to the CPU.
5445 Not specifying this option is equivalent to retbleed=auto.
5447 rfkill.default_state=
5448 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm,
5449 etc. communication is blocked by default.
5452 rfkill.master_switch_mode=
5453 0 The "airplane mode" button does nothing.
5454 1 The "airplane mode" button toggles between everything
5455 blocked and the previous configuration.
5456 2 The "airplane mode" button toggles between everything
5457 blocked and everything unblocked.
5459 rhash_entries= [KNL,NET]
5460 Set number of hash buckets for route cache
5463 [KNL] Disable ring 3 MONITOR/MWAIT feature on supported
5466 ro [KNL] Mount root device read-only on boot
5469 on Mark read-only kernel memory as read-only (default).
5470 off Leave read-only kernel memory writable for debugging.
5471 full Mark read-only kernel memory and aliases as read-only
5475 Enable the uart passthrough on the designated usb port
5476 on Rockchip SoCs. When active, the signals of the
5477 debug-uart get routed to the D+ and D- pins of the usb
5478 port and the regular usb controller gets disabled.
5480 root= [KNL] Root filesystem
5481 Usually this a a block device specifier of some kind,
5482 see the early_lookup_bdev comment in
5483 block/early-lookup.c for details.
5484 Alternatively this can be "ram" for the legacy initial
5485 ramdisk, "nfs" and "cifs" for root on a network file
5486 system, or "mtd" and "ubi" for mounting from raw flash.
5488 rootdelay= [KNL] Delay (in seconds) to pause before attempting to
5489 mount the root filesystem
5491 rootflags= [KNL] Set root filesystem mount option string
5493 rootfstype= [KNL] Set root filesystem type
5495 rootwait [KNL] Wait (indefinitely) for root device to show up.
5496 Useful for devices that are detected asynchronously
5497 (e.g. USB and MMC devices).
5499 rproc_mem=nn[KMG][@address]
5500 [KNL,ARM,CMA] Remoteproc physical memory block.
5501 Memory area to be used by remote processor image,
5504 rw [KNL] Mount root device read-write on boot
5506 S [KNL] Run init in single mode
5508 s390_iommu= [HW,S390]
5509 Set s390 IOTLB flushing mode
5511 With strict flushing every unmap operation will result in
5512 an IOTLB flush. Default is lazy flushing before reuse,
5515 s390_iommu_aperture= [KNL,S390]
5516 Specifies the size of the per device DMA address space
5517 accessible through the DMA and IOMMU APIs as a decimal
5518 factor of the size of main memory.
5519 The default is 1 meaning that one can concurrently use
5520 as many DMA addresses as physical memory is installed,
5521 if supported by hardware, and thus map all of memory
5522 once. With a value of 2 one can map all of memory twice
5523 and so on. As a special case a factor of 0 imposes no
5524 restrictions other than those given by hardware at the
5525 cost of significant additional memory use for tables.
5528 See drivers/net/irda/sa1100_ir.c.
5530 sched_verbose [KNL] Enables verbose scheduler debug messages.
5532 schedstats= [KNL,X86] Enable or disable scheduled statistics.
5533 Allowed values are enable and disable. This feature
5534 incurs a small amount of overhead in the scheduler
5535 but is useful for debugging and performance tuning.
5537 sched_thermal_decay_shift=
5538 [KNL, SMP] Set a decay shift for scheduler thermal
5539 pressure signal. Thermal pressure signal follows the
5540 default decay period of other scheduler pelt
5541 signals(usually 32 ms but configurable). Setting
5542 sched_thermal_decay_shift will left shift the decay
5543 period for the thermal pressure signal by the shift
5545 i.e. with the default pelt decay period of 32 ms
5546 sched_thermal_decay_shift thermal pressure decay pr
5550 Format: integer between 0 and 10
5553 scftorture.holdoff= [KNL]
5554 Number of seconds to hold off before starting
5555 test. Defaults to zero for module insertion and
5556 to 10 seconds for built-in smp_call_function()
5559 scftorture.longwait= [KNL]
5560 Request ridiculously long waits randomly selected
5561 up to the chosen limit in seconds. Zero (the
5562 default) disables this feature. Please note
5563 that requesting even small non-zero numbers of
5564 seconds can result in RCU CPU stall warnings,
5565 softlockup complaints, and so on.
5567 scftorture.nthreads= [KNL]
5568 Number of kthreads to spawn to invoke the
5569 smp_call_function() family of functions.
5570 The default of -1 specifies a number of kthreads
5571 equal to the number of CPUs.
5573 scftorture.onoff_holdoff= [KNL]
5574 Number seconds to wait after the start of the
5575 test before initiating CPU-hotplug operations.
5577 scftorture.onoff_interval= [KNL]
5578 Number seconds to wait between successive
5579 CPU-hotplug operations. Specifying zero (which
5580 is the default) disables CPU-hotplug operations.
5582 scftorture.shutdown_secs= [KNL]
5583 The number of seconds following the start of the
5584 test after which to shut down the system. The
5585 default of zero avoids shutting down the system.
5586 Non-zero values are useful for automated tests.
5588 scftorture.stat_interval= [KNL]
5589 The number of seconds between outputting the
5590 current test statistics to the console. A value
5591 of zero disables statistics output.
5593 scftorture.stutter_cpus= [KNL]
5594 The number of jiffies to wait between each change
5595 to the set of CPUs under test.
5597 scftorture.use_cpus_read_lock= [KNL]
5598 Use use_cpus_read_lock() instead of the default
5599 preempt_disable() to disable CPU hotplug
5600 while invoking one of the smp_call_function*()
5603 scftorture.verbose= [KNL]
5604 Enable additional printk() statements.
5606 scftorture.weight_single= [KNL]
5607 The probability weighting to use for the
5608 smp_call_function_single() function with a zero
5609 "wait" parameter. A value of -1 selects the
5610 default if all other weights are -1. However,
5611 if at least one weight has some other value, a
5612 value of -1 will instead select a weight of zero.
5614 scftorture.weight_single_wait= [KNL]
5615 The probability weighting to use for the
5616 smp_call_function_single() function with a
5617 non-zero "wait" parameter. See weight_single.
5619 scftorture.weight_many= [KNL]
5620 The probability weighting to use for the
5621 smp_call_function_many() function with a zero
5622 "wait" parameter. See weight_single.
5623 Note well that setting a high probability for
5624 this weighting can place serious IPI load
5627 scftorture.weight_many_wait= [KNL]
5628 The probability weighting to use for the
5629 smp_call_function_many() function with a
5630 non-zero "wait" parameter. See weight_single
5633 scftorture.weight_all= [KNL]
5634 The probability weighting to use for the
5635 smp_call_function_all() function with a zero
5636 "wait" parameter. See weight_single and
5639 scftorture.weight_all_wait= [KNL]
5640 The probability weighting to use for the
5641 smp_call_function_all() function with a
5642 non-zero "wait" parameter. See weight_single
5645 skew_tick= [KNL] Offset the periodic timer tick per cpu to mitigate
5646 xtime_lock contention on larger systems, and/or RCU lock
5647 contention on all systems with CONFIG_MAXSMP set.
5648 Format: { "0" | "1" }
5649 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1"
5651 Note: increases power consumption, thus should only be
5652 enabled if running jitter sensitive (HPC/RT) workloads.
5654 security= [SECURITY] Choose a legacy "major" security module to
5655 enable at boot. This has been deprecated by the
5658 selinux= [SELINUX] Disable or enable SELinux at boot time.
5659 Format: { "0" | "1" }
5660 See security/selinux/Kconfig help text.
5665 serialnumber [BUGS=X86-32]
5667 sev=option[,option...] [X86-64] See Documentation/arch/x86/x86_64/boot-options.rst
5670 Maximal number of shapers.
5672 show_lapic= [APIC,X86] Advanced Programmable Interrupt Controller
5673 Limit apic dumping. The parameter defines the maximal
5674 number of local apics being dumped. Also it is possible
5675 to set it to "all" by meaning -- no limit here.
5676 Format: { 1 (default) | 2 | ... | all }.
5677 The parameter valid if only apic=debug or
5678 apic=verbose is specified.
5679 Example: apic=debug show_lapic=all
5687 Enable merging of slabs with similar size when the
5688 kernel is built without CONFIG_SLAB_MERGE_DEFAULT.
5691 Disable merging of slabs with similar size. May be
5692 necessary if there is some reason to distinguish
5693 allocs to different slabs, especially in hardened
5694 environments where the risk of heap overflows and
5695 layout control by attackers can usually be
5696 frustrated by disabling merging. This will reduce
5697 most of the exposure of a heap attack to a single
5698 cache (risks via metadata attacks are mostly
5699 unchanged). Debug options disable merging on their
5701 For more information see Documentation/mm/slub.rst.
5703 slab_max_order= [MM, SLAB]
5704 Determines the maximum allowed order for slabs.
5705 A high setting may cause OOMs due to memory
5706 fragmentation. Defaults to 1 for systems with
5707 more than 32MB of RAM, 0 otherwise.
5709 slub_debug[=options[,slabs][;[options[,slabs]]...] [MM, SLUB]
5710 Enabling slub_debug allows one to determine the
5711 culprit if slab objects become corrupted. Enabling
5712 slub_debug can create guard zones around objects and
5713 may poison objects when not in use. Also tracks the
5714 last alloc / free. For more information see
5715 Documentation/mm/slub.rst.
5717 slub_max_order= [MM, SLUB]
5718 Determines the maximum allowed order for slabs.
5719 A high setting may cause OOMs due to memory
5720 fragmentation. For more information see
5721 Documentation/mm/slub.rst.
5723 slub_min_objects= [MM, SLUB]
5724 The minimum number of objects per slab. SLUB will
5725 increase the slab order up to slub_max_order to
5726 generate a sufficiently large slab able to contain
5727 the number of objects indicated. The higher the number
5728 of objects the smaller the overhead of tracking slabs
5729 and the less frequently locks need to be acquired.
5730 For more information see Documentation/mm/slub.rst.
5732 slub_min_order= [MM, SLUB]
5733 Determines the minimum page order for slabs. Must be
5734 lower than slub_max_order.
5735 For more information see Documentation/mm/slub.rst.
5737 slub_merge [MM, SLUB]
5738 Same with slab_merge.
5740 slub_nomerge [MM, SLUB]
5741 Same with slab_nomerge. This is supported for legacy.
5742 See slab_nomerge for more information.
5745 Format: <io1>[,<io2>[,...,<io8>]]
5747 smp.csd_lock_timeout= [KNL]
5748 Specify the period of time in milliseconds
5749 that smp_call_function() and friends will wait
5750 for a CPU to release the CSD lock. This is
5751 useful when diagnosing bugs involving CPUs
5752 disabling interrupts for extended periods
5753 of time. Defaults to 5,000 milliseconds, and
5754 setting a value of zero disables this feature.
5755 This feature may be more efficiently disabled
5756 using the csdlock_debug- kernel parameter.
5758 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices
5759 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port
5760 smsc-ircc2.ircc_sir= [HW] SIR base I/O port
5761 smsc-ircc2.ircc_fir= [HW] FIR base I/O port
5762 smsc-ircc2.ircc_irq= [HW] IRQ line
5763 smsc-ircc2.ircc_dma= [HW] DMA channel
5764 smsc-ircc2.ircc_transceiver= [HW] Transceiver type:
5765 0: Toshiba Satellite 1800 (GP data pin select)
5766 1: Fast pin select (default)
5769 smt= [KNL,S390] Set the maximum number of threads (logical
5770 CPUs) to use per physical CPU on systems capable of
5771 symmetric multithreading (SMT). Will be capped to the
5772 actual hardware limit.
5774 Default: -1 (no limit)
5777 [KNL] Should the soft-lockup detector generate panics.
5780 A value of 1 instructs the soft-lockup detector
5781 to panic the machine when a soft-lockup occurs. It is
5782 also controlled by the kernel.softlockup_panic sysctl
5783 and CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC, which is the
5784 respective build-time switch to that functionality.
5786 softlockup_all_cpu_backtrace=
5787 [KNL] Should the soft-lockup detector generate
5788 backtraces on all cpus.
5791 sonypi.*= [HW] Sony Programmable I/O Control Device driver
5792 See Documentation/admin-guide/laptops/sonypi.rst
5794 spectre_v2= [X86] Control mitigation of Spectre variant 2
5795 (indirect branch speculation) vulnerability.
5796 The default operation protects the kernel from
5799 on - unconditionally enable, implies
5801 off - unconditionally disable, implies
5803 auto - kernel detects whether your CPU model is
5806 Selecting 'on' will, and 'auto' may, choose a
5807 mitigation method at run time according to the
5808 CPU, the available microcode, the setting of the
5809 CONFIG_RETPOLINE configuration option, and the
5810 compiler with which the kernel was built.
5812 Selecting 'on' will also enable the mitigation
5813 against user space to user space task attacks.
5815 Selecting 'off' will disable both the kernel and
5816 the user space protections.
5818 Specific mitigations can also be selected manually:
5820 retpoline - replace indirect branches
5821 retpoline,generic - Retpolines
5822 retpoline,lfence - LFENCE; indirect branch
5823 retpoline,amd - alias for retpoline,lfence
5824 eibrs - Enhanced/Auto IBRS
5825 eibrs,retpoline - Enhanced/Auto IBRS + Retpolines
5826 eibrs,lfence - Enhanced/Auto IBRS + LFENCE
5827 ibrs - use IBRS to protect kernel
5829 Not specifying this option is equivalent to
5833 [X86] Control mitigation of Spectre variant 2
5834 (indirect branch speculation) vulnerability between
5837 on - Unconditionally enable mitigations. Is
5838 enforced by spectre_v2=on
5840 off - Unconditionally disable mitigations. Is
5841 enforced by spectre_v2=off
5843 prctl - Indirect branch speculation is enabled,
5844 but mitigation can be enabled via prctl
5845 per thread. The mitigation control state
5846 is inherited on fork.
5849 - Like "prctl" above, but only STIBP is
5850 controlled per thread. IBPB is issued
5851 always when switching between different user
5855 - Same as "prctl" above, but all seccomp
5856 threads will enable the mitigation unless
5857 they explicitly opt out.
5860 - Like "seccomp" above, but only STIBP is
5861 controlled per thread. IBPB is issued
5862 always when switching between different
5863 user space processes.
5865 auto - Kernel selects the mitigation depending on
5866 the available CPU features and vulnerability.
5868 Default mitigation: "prctl"
5870 Not specifying this option is equivalent to
5871 spectre_v2_user=auto.
5873 spec_store_bypass_disable=
5874 [HW] Control Speculative Store Bypass (SSB) Disable mitigation
5875 (Speculative Store Bypass vulnerability)
5877 Certain CPUs are vulnerable to an exploit against a
5878 a common industry wide performance optimization known
5879 as "Speculative Store Bypass" in which recent stores
5880 to the same memory location may not be observed by
5881 later loads during speculative execution. The idea
5882 is that such stores are unlikely and that they can
5883 be detected prior to instruction retirement at the
5884 end of a particular speculation execution window.
5886 In vulnerable processors, the speculatively forwarded
5887 store can be used in a cache side channel attack, for
5888 example to read memory to which the attacker does not
5889 directly have access (e.g. inside sandboxed code).
5891 This parameter controls whether the Speculative Store
5892 Bypass optimization is used.
5894 On x86 the options are:
5896 on - Unconditionally disable Speculative Store Bypass
5897 off - Unconditionally enable Speculative Store Bypass
5898 auto - Kernel detects whether the CPU model contains an
5899 implementation of Speculative Store Bypass and
5900 picks the most appropriate mitigation. If the
5901 CPU is not vulnerable, "off" is selected. If the
5902 CPU is vulnerable the default mitigation is
5903 architecture and Kconfig dependent. See below.
5904 prctl - Control Speculative Store Bypass per thread
5905 via prctl. Speculative Store Bypass is enabled
5906 for a process by default. The state of the control
5907 is inherited on fork.
5908 seccomp - Same as "prctl" above, but all seccomp threads
5909 will disable SSB unless they explicitly opt out.
5911 Default mitigations:
5914 On powerpc the options are:
5916 on,auto - On Power8 and Power9 insert a store-forwarding
5917 barrier on kernel entry and exit. On Power7
5918 perform a software flush on kernel entry and
5922 Not specifying this option is equivalent to
5923 spec_store_bypass_disable=auto.
5925 spia_io_base= [HW,MTD]
5931 [X86] Enable split lock detection or bus lock detection
5933 When enabled (and if hardware support is present), atomic
5934 instructions that access data across cache line
5935 boundaries will result in an alignment check exception
5936 for split lock detection or a debug exception for
5941 warn - the kernel will emit rate-limited warnings
5942 about applications triggering the #AC
5943 exception or the #DB exception. This mode is
5944 the default on CPUs that support split lock
5945 detection or bus lock detection. Default
5946 behavior is by #AC if both features are
5947 enabled in hardware.
5949 fatal - the kernel will send SIGBUS to applications
5950 that trigger the #AC exception or the #DB
5951 exception. Default behavior is by #AC if
5952 both features are enabled in hardware.
5955 Set system wide rate limit to N bus locks
5956 per second for bus lock detection.
5959 N/A for split lock detection.
5962 If an #AC exception is hit in the kernel or in
5963 firmware (i.e. not while executing in user mode)
5964 the kernel will oops in either "warn" or "fatal"
5967 #DB exception for bus lock is triggered only when
5971 Control the Special Register Buffer Data Sampling
5974 Certain CPUs are vulnerable to an MDS-like
5975 exploit which can leak bits from the random
5978 By default, this issue is mitigated by
5979 microcode. However, the microcode fix can cause
5980 the RDRAND and RDSEED instructions to become
5981 much slower. Among other effects, this will
5982 result in reduced throughput from /dev/urandom.
5984 The microcode mitigation can be disabled with
5985 the following option:
5987 off: Disable mitigation and remove
5988 performance impact to RDRAND and RDSEED
5990 srcutree.big_cpu_lim [KNL]
5991 Specifies the number of CPUs constituting a
5992 large system, such that srcu_struct structures
5993 should immediately allocate an srcu_node array.
5994 This kernel-boot parameter defaults to 128,
5995 but takes effect only when the low-order four
5996 bits of srcutree.convert_to_big is equal to 3
5999 srcutree.convert_to_big [KNL]
6000 Specifies under what conditions an SRCU tree
6001 srcu_struct structure will be converted to big
6002 form, that is, with an rcu_node tree:
6005 1: At init_srcu_struct() time.
6006 2: When rcutorture decides to.
6007 3: Decide at boot time (default).
6008 0x1X: Above plus if high contention.
6010 Either way, the srcu_node tree will be sized based
6011 on the actual runtime number of CPUs (nr_cpu_ids)
6012 instead of the compile-time CONFIG_NR_CPUS.
6014 srcutree.counter_wrap_check [KNL]
6015 Specifies how frequently to check for
6016 grace-period sequence counter wrap for the
6017 srcu_data structure's ->srcu_gp_seq_needed field.
6018 The greater the number of bits set in this kernel
6019 parameter, the less frequently counter wrap will
6020 be checked for. Note that the bottom two bits
6023 srcutree.exp_holdoff [KNL]
6024 Specifies how many nanoseconds must elapse
6025 since the end of the last SRCU grace period for
6026 a given srcu_struct until the next normal SRCU
6027 grace period will be considered for automatic
6028 expediting. Set to zero to disable automatic
6031 srcutree.srcu_max_nodelay [KNL]
6032 Specifies the number of no-delay instances
6033 per jiffy for which the SRCU grace period
6034 worker thread will be rescheduled with zero
6035 delay. Beyond this limit, worker thread will
6036 be rescheduled with a sleep delay of one jiffy.
6038 srcutree.srcu_max_nodelay_phase [KNL]
6039 Specifies the per-grace-period phase, number of
6040 non-sleeping polls of readers. Beyond this limit,
6041 grace period worker thread will be rescheduled
6042 with a sleep delay of one jiffy, between each
6043 rescan of the readers, for a grace period phase.
6045 srcutree.srcu_retry_check_delay [KNL]
6046 Specifies number of microseconds of non-sleeping
6047 delay between each non-sleeping poll of readers.
6049 srcutree.small_contention_lim [KNL]
6050 Specifies the number of update-side contention
6051 events per jiffy will be tolerated before
6052 initiating a conversion of an srcu_struct
6053 structure to big form. Note that the value of
6054 srcutree.convert_to_big must have the 0x10 bit
6055 set for contention-based conversions to occur.
6058 Speculative Store Bypass Disable control
6060 On CPUs that are vulnerable to the Speculative
6061 Store Bypass vulnerability and offer a
6062 firmware based mitigation, this parameter
6063 indicates how the mitigation should be used:
6065 force-on: Unconditionally enable mitigation for
6066 for both kernel and userspace
6067 force-off: Unconditionally disable mitigation for
6068 for both kernel and userspace
6069 kernel: Always enable mitigation in the
6070 kernel, and offer a prctl interface
6071 to allow userspace to register its
6072 interest in being mitigated too.
6074 stack_guard_gap= [MM]
6075 override the default stack gap protection. The value
6076 is in page units and it defines how many pages prior
6077 to (for stacks growing down) resp. after (for stacks
6078 growing up) the main stack are reserved for no other
6079 mapping. Default value is 256 pages.
6081 stack_depot_disable= [KNL]
6082 Setting this to true through kernel command line will
6083 disable the stack depot thereby saving the static memory
6084 consumed by the stack hash table. By default this is set
6088 Enabled the stack tracer on boot up.
6090 stacktrace_filter=[function-list]
6091 [FTRACE] Limit the functions that the stack tracer
6092 will trace at boot up. function-list is a comma-separated
6093 list of functions. This list can be changed at run
6094 time by the stack_trace_filter file in the debugfs
6095 tracing directory. Note, this enables stack tracing
6096 and the stacktrace above is not needed.
6100 Set the STI (builtin display/keyboard on the HP-PARISC
6101 machines) console (graphic card) which should be used
6102 as the initial boot-console.
6103 See also comment in drivers/video/console/sticore.c.
6106 See comment in drivers/video/console/sticore.c.
6109 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]]
6114 Enable or disable strict sigaltstack size checks
6115 against the required signal frame size which
6116 depends on the supported FPU features. This can
6117 be used to filter out binaries which have
6118 not yet been made aware of AT_MINSIGSTKSZ.
6121 Limits the number of kernel HPT entries in the hash
6122 page table to increase the rate of hash page table
6123 faults on kernel addresses.
6126 Limits the number of kernel SLB entries, and flushes
6127 them frequently to increase the rate of SLB faults
6128 on kernel addresses.
6130 sunrpc.min_resvport=
6131 sunrpc.max_resvport=
6133 SunRPC servers often require that client requests
6134 originate from a privileged port (i.e. a port in the
6135 range 0 < portnr < 1024).
6136 An administrator who wishes to reserve some of these
6137 ports for other uses may adjust the range that the
6138 kernel's sunrpc client considers to be privileged
6139 using these two parameters to set the minimum and
6140 maximum port values.
6142 sunrpc.svc_rpc_per_connection_limit=
6144 Limit the number of requests that the server will
6145 process in parallel from a single connection.
6146 The default value is 0 (no limit).
6150 Control how the NFS server code allocates CPUs to
6151 service thread pools. Depending on how many NICs
6152 you have and where their interrupts are bound, this
6153 option will affect which CPUs will do NFS serving.
6154 Note: this parameter cannot be changed while the
6155 NFS server is running.
6157 auto the server chooses an appropriate mode
6158 automatically using heuristics
6159 global a single global pool contains all CPUs
6160 percpu one pool for each CPU
6161 pernode one pool for each NUMA node (equivalent
6162 to global on non-NUMA machines)
6164 sunrpc.tcp_slot_table_entries=
6165 sunrpc.udp_slot_table_entries=
6167 Sets the upper limit on the number of simultaneous
6168 RPC calls that can be sent from the client to a
6169 server. Increasing these values may allow you to
6170 improve throughput, but will also increase the
6171 amount of memory reserved for use by the client.
6173 suspend.pm_test_delay=
6175 Sets the number of seconds to remain in a suspend test
6176 mode before resuming the system (see
6177 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG
6178 is set. Default value is 5.
6181 Format: { on | off | y | n | 1 | 0 }
6182 This parameter controls use of the Protected
6183 Execution Facility on pSeries.
6185 swiotlb= [ARM,IA-64,PPC,MIPS,X86]
6186 Format: { <int> [,<int>] | force | noforce }
6187 <int> -- Number of I/O TLB slabs
6188 <int> -- Second integer after comma. Number of swiotlb
6189 areas with their own lock. Will be rounded up
6191 force -- force using of bounce buffers even if they
6192 wouldn't be automatically used by the kernel
6193 noforce -- Never use bounce buffers (for debugging)
6198 Set a sysctl parameter, right before loading the init
6199 process, as if the value was written to the respective
6200 /proc/sys/... file. Both '.' and '/' are recognized as
6201 separators. Unrecognized parameters and invalid values
6202 are reported in the kernel log. Sysctls registered
6203 later by a loaded module cannot be set this way.
6204 Example: sysctl.vm.swappiness=40
6206 sysrq_always_enabled
6208 Ignore sysrq setting - this boot parameter will
6209 neutralize any effect of /proc/sys/kernel/sysrq.
6210 Useful for debugging.
6212 tcpmhash_entries= [KNL,NET]
6213 Set the number of tcp_metrics_hash slots.
6214 Default value is 8192 or 16384 depending on total
6215 ram pages. This is used to specify the TCP metrics
6216 cache size. See Documentation/networking/ip-sysctl.rst
6217 "tcp_no_metrics_save" section for more details.
6221 test_suspend= [SUSPEND]
6222 Format: { "mem" | "standby" | "freeze" }[,N]
6223 Specify "mem" (for Suspend-to-RAM) or "standby" (for
6224 standby suspend) or "freeze" (for suspend type freeze)
6225 as the system sleep state during system startup with
6226 the optional capability to repeat N number of times.
6227 The system is woken from this state using a
6228 wakeup-capable RTC alarm.
6230 thash_entries= [KNL,NET]
6231 Set number of hash buckets for TCP connection
6233 thermal.act= [HW,ACPI]
6234 -1: disable all active trip points in all thermal zones
6235 <degrees C>: override all lowest active trip points
6237 thermal.crt= [HW,ACPI]
6238 -1: disable all critical trip points in all thermal zones
6239 <degrees C>: override all critical trip points
6241 thermal.nocrt= [HW,ACPI]
6242 Set to disable actions on ACPI thermal zone
6243 critical and hot trip points.
6245 thermal.off= [HW,ACPI]
6246 1: disable ACPI thermal control
6248 thermal.psv= [HW,ACPI]
6249 -1: disable all passive trip points
6250 <degrees C>: override all passive trip points to this
6253 thermal.tzp= [HW,ACPI]
6254 Specify global default ACPI thermal zone polling rate
6255 <deci-seconds>: poll all this frequency
6256 0: no polling (default)
6259 Force threading of all interrupt handlers except those
6260 marked explicitly IRQF_NO_THREAD.
6264 Specify if the kernel should make use of the cpu
6265 topology information if the hardware supports this.
6266 The scheduler will make use of this information and
6267 e.g. base its process migration decisions on it.
6270 topology_updates= [KNL, PPC, NUMA]
6272 Specify if the kernel should ignore (off)
6273 topology updates sent by the hypervisor to this
6276 torture.disable_onoff_at_boot= [KNL]
6277 Prevent the CPU-hotplug component of torturing
6278 until after init has spawned.
6280 torture.ftrace_dump_at_shutdown= [KNL]
6281 Dump the ftrace buffer at torture-test shutdown,
6282 even if there were no errors. This can be a
6283 very costly operation when many torture tests
6284 are running concurrently, especially on systems
6285 with rotating-rust storage.
6287 torture.verbose_sleep_frequency= [KNL]
6288 Specifies how many verbose printk()s should be
6289 emitted between each sleep. The default of zero
6290 disables verbose-printk() sleeping.
6292 torture.verbose_sleep_duration= [KNL]
6293 Duration of each verbose-printk() sleep in jiffies.
6297 tpm_suspend_pcr=[HW,TPM]
6298 Format: integer pcr id
6299 Specify that at suspend time, the tpm driver
6300 should extend the specified pcr with zeros,
6301 as a workaround for some chips which fail to
6302 flush the last written pcr on TPM_SaveState.
6303 This will guarantee that all the other pcrs
6307 Have the tracepoints sent to printk as well as the
6308 tracing ring buffer. This is useful for early boot up
6309 where the system hangs or reboots and does not give the
6310 option for reading the tracing buffer or performing a
6311 ftrace_dump_on_oops.
6313 To turn off having tracepoints sent to printk,
6314 echo 0 > /proc/sys/kernel/tracepoint_printk
6315 Note, echoing 1 into this file without the
6316 tracepoint_printk kernel cmdline option has no effect.
6318 The tp_printk_stop_on_boot (see below) can also be used
6319 to stop the printing of events to console at
6324 Having tracepoints sent to printk() and activating high
6325 frequency tracepoints such as irq or sched, can cause
6326 the system to live lock.
6328 tp_printk_stop_on_boot [FTRACE]
6329 When tp_printk (above) is set, it can cause a lot of noise
6330 on the console. It may be useful to only include the
6331 printing of events during boot up, as user space may
6332 make the system inoperable.
6334 This command line option will stop the printing of events
6335 to console at the late_initcall_sync() time frame.
6337 trace_buf_size=nn[KMG]
6338 [FTRACE] will set tracing buffer size on each cpu.
6340 trace_clock= [FTRACE] Set the clock used for tracing events
6342 local - Use the per CPU time stamp counter
6343 (converted into nanoseconds). Fast, but
6344 depending on the architecture, may not be
6345 in sync between CPUs.
6346 global - Event time stamps are synchronize across
6347 CPUs. May be slower than the local clock,
6348 but better for some race conditions.
6349 counter - Simple counting of events (1, 2, ..)
6350 note, some counts may be skipped due to the
6351 infrastructure grabbing the clock more than
6353 uptime - Use jiffies as the time stamp.
6354 perf - Use the same clock that perf uses.
6355 mono - Use ktime_get_mono_fast_ns() for time stamps.
6356 mono_raw - Use ktime_get_raw_fast_ns() for time
6358 boot - Use ktime_get_boot_fast_ns() for time stamps.
6359 Architectures may add more clocks. See
6360 Documentation/trace/ftrace.rst for more details.
6362 trace_event=[event-list]
6363 [FTRACE] Set and start specified trace events in order
6364 to facilitate early boot debugging. The event-list is a
6365 comma-separated list of trace events to enable. See
6366 also Documentation/trace/events.rst
6368 trace_instance=[instance-info]
6369 [FTRACE] Create a ring buffer instance early in boot up.
6370 This will be listed in:
6372 /sys/kernel/tracing/instances
6374 Events can be enabled at the time the instance is created
6377 trace_instance=<name>,<system1>:<event1>,<system2>:<event2>
6379 Note, the "<system*>:" portion is optional if the event is
6382 trace_instance=foo,sched:sched_switch,irq_handler_entry,initcall
6384 will enable the "sched_switch" event (note, the "sched:" is optional, and
6385 the same thing would happen if it was left off). The irq_handler_entry
6386 event, and all events under the "initcall" system.
6388 trace_options=[option-list]
6389 [FTRACE] Enable or disable tracer options at boot.
6390 The option-list is a comma delimited list of options
6391 that can be enabled or disabled just as if you were
6392 to echo the option name into
6394 /sys/kernel/tracing/trace_options
6396 For example, to enable stacktrace option (to dump the
6397 stack trace of each event), add to the command line:
6399 trace_options=stacktrace
6401 See also Documentation/trace/ftrace.rst "trace options"
6404 trace_trigger=[trigger-list]
6405 [FTRACE] Add a event trigger on specific events.
6406 Set a trigger on top of a specific event, with an optional
6409 The format is is "trace_trigger=<event>.<trigger>[ if <filter>],..."
6410 Where more than one trigger may be specified that are comma deliminated.
6414 trace_trigger="sched_switch.stacktrace if prev_state == 2"
6416 The above will enable the "stacktrace" trigger on the "sched_switch"
6417 event but only trigger it if the "prev_state" of the "sched_switch"
6418 event is "2" (TASK_UNINTERUPTIBLE).
6420 See also "Event triggers" in Documentation/trace/events.rst
6424 [FTRACE] enable this option to disable tracing when a
6425 warning is hit. This turns off "tracing_on". Tracing can
6426 be enabled again by echoing '1' into the "tracing_on"
6427 file located in /sys/kernel/tracing/
6429 This option is useful, as it disables the trace before
6430 the WARNING dump is called, which prevents the trace to
6431 be filled with content caused by the warning output.
6433 This option can also be set at run time via the sysctl
6434 option: kernel/traceoff_on_warning
6436 transparent_hugepage=
6438 Format: [always|madvise|never]
6439 Can be used to control the default behavior of the system
6440 with respect to transparent hugepages.
6441 See Documentation/admin-guide/mm/transhuge.rst
6444 trusted.source= [KEYS]
6446 This parameter identifies the trust source as a backend
6447 for trusted keys implementation. Supported trust
6452 If not specified then it defaults to iterating through
6453 the trust source list starting with TPM and assigns the
6454 first trust source as a backend which is initialized
6455 successfully during iteration.
6459 The RNG used to generate key material for trusted keys.
6462 - the same value as trusted.source: "tpm" or "tee"
6464 If not specified, "default" is used. In this case,
6465 the RNG's choice is left to each individual trust source.
6467 tsc= Disable clocksource stability checks for TSC.
6469 [x86] reliable: mark tsc clocksource as reliable, this
6470 disables clocksource verification at runtime, as well
6471 as the stability checks done at bootup. Used to enable
6472 high-resolution timer mode on older hardware, and in
6473 virtualized environment.
6474 [x86] noirqtime: Do not use TSC to do irq accounting.
6475 Used to run time disable IRQ_TIME_ACCOUNTING on any
6476 platforms where RDTSC is slow and this accounting
6478 [x86] unstable: mark the TSC clocksource as unstable, this
6479 marks the TSC unconditionally unstable at bootup and
6480 avoids any further wobbles once the TSC watchdog notices.
6481 [x86] nowatchdog: disable clocksource watchdog. Used
6482 in situations with strict latency requirements (where
6483 interruptions from clocksource watchdog are not
6485 [x86] recalibrate: force recalibration against a HW timer
6486 (HPET or PM timer) on systems whose TSC frequency was
6487 obtained from HW or FW using either an MSR or CPUID(0x15).
6488 Warn if the difference is more than 500 ppm.
6489 [x86] watchdog: Use TSC as the watchdog clocksource with
6490 which to check other HW timers (HPET or PM timer), but
6491 only on systems where TSC has been deemed trustworthy.
6492 This will be suppressed by an earlier tsc=nowatchdog and
6493 can be overridden by a later tsc=nowatchdog. A console
6494 message will flag any such suppression or overriding.
6496 tsc_early_khz= [X86] Skip early TSC calibration and use the given
6497 value instead. Useful when the early TSC frequency discovery
6498 procedure is not reliable, such as on overclocked systems
6499 with CPUID.16h support and partial CPUID.15h support.
6500 Format: <unsigned int>
6502 tsx= [X86] Control Transactional Synchronization
6503 Extensions (TSX) feature in Intel processors that
6504 support TSX control.
6506 This parameter controls the TSX feature. The options are:
6508 on - Enable TSX on the system. Although there are
6509 mitigations for all known security vulnerabilities,
6510 TSX has been known to be an accelerator for
6511 several previous speculation-related CVEs, and
6512 so there may be unknown security risks associated
6513 with leaving it enabled.
6515 off - Disable TSX on the system. (Note that this
6516 option takes effect only on newer CPUs which are
6517 not vulnerable to MDS, i.e., have
6518 MSR_IA32_ARCH_CAPABILITIES.MDS_NO=1 and which get
6519 the new IA32_TSX_CTRL MSR through a microcode
6520 update. This new MSR allows for the reliable
6521 deactivation of the TSX functionality.)
6523 auto - Disable TSX if X86_BUG_TAA is present,
6524 otherwise enable TSX on the system.
6526 Not specifying this option is equivalent to tsx=off.
6528 See Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
6531 tsx_async_abort= [X86,INTEL] Control mitigation for the TSX Async
6532 Abort (TAA) vulnerability.
6534 Similar to Micro-architectural Data Sampling (MDS)
6535 certain CPUs that support Transactional
6536 Synchronization Extensions (TSX) are vulnerable to an
6537 exploit against CPU internal buffers which can forward
6538 information to a disclosure gadget under certain
6541 In vulnerable processors, the speculatively forwarded
6542 data can be used in a cache side channel attack, to
6543 access data to which the attacker does not have direct
6546 This parameter controls the TAA mitigation. The
6549 full - Enable TAA mitigation on vulnerable CPUs
6552 full,nosmt - Enable TAA mitigation and disable SMT on
6553 vulnerable CPUs. If TSX is disabled, SMT
6554 is not disabled because CPU is not
6555 vulnerable to cross-thread TAA attacks.
6556 off - Unconditionally disable TAA mitigation
6558 On MDS-affected machines, tsx_async_abort=off can be
6559 prevented by an active MDS mitigation as both vulnerabilities
6560 are mitigated with the same mechanism so in order to disable
6561 this mitigation, you need to specify mds=off too.
6563 Not specifying this option is equivalent to
6564 tsx_async_abort=full. On CPUs which are MDS affected
6565 and deploy MDS mitigation, TAA mitigation is not
6566 required and doesn't provide any additional
6570 Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
6572 turbografx.map[2|3]= [HW,JOY]
6573 TurboGraFX parallel port interface
6575 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7>
6576 See also Documentation/input/devices/joystick-parport.rst
6578 udbg-immortal [PPC] When debugging early kernel crashes that
6579 happen after console_init() and before a proper
6580 console driver takes over, this boot options might
6581 help "seeing" what's going on.
6583 uhash_entries= [KNL,NET]
6584 Set number of hash buckets for UDP/UDP-Lite connections
6587 [USB] Ignore overcurrent events (default N).
6588 Some badly-designed motherboards generate lots of
6589 bogus events, for ports that aren't wired to
6590 anything. Set this parameter to avoid log spamming.
6591 Note that genuine overcurrent events won't be
6595 [X86] Cause panic on unknown NMI.
6597 usbcore.authorized_default=
6598 [USB] Default USB device authorization:
6599 (default -1 = authorized except for wireless USB,
6600 0 = not authorized, 1 = authorized, 2 = authorized
6601 if device connected to internal port)
6603 usbcore.autosuspend=
6604 [USB] The autosuspend time delay (in seconds) used
6605 for newly-detected USB devices (default 2). This
6606 is the time required before an idle device will be
6607 autosuspended. Devices for which the delay is set
6608 to a negative value won't be autosuspended at all.
6610 usbcore.usbfs_snoop=
6611 [USB] Set to log all usbfs traffic (default 0 = off).
6613 usbcore.usbfs_snoop_max=
6614 [USB] Maximum number of bytes to snoop in each URB
6617 usbcore.blinkenlights=
6618 [USB] Set to cycle leds on hubs (default 0 = off).
6620 usbcore.old_scheme_first=
6621 [USB] Start with the old device initialization
6622 scheme (default 0 = off).
6624 usbcore.usbfs_memory_mb=
6625 [USB] Memory limit (in MB) for buffers allocated by
6626 usbfs (default = 16, 0 = max = 2047).
6628 usbcore.use_both_schemes=
6629 [USB] Try the other device initialization scheme
6630 if the first one fails (default 1 = enabled).
6632 usbcore.initial_descriptor_timeout=
6633 [USB] Specifies timeout for the initial 64-byte
6634 USB_REQ_GET_DESCRIPTOR request in milliseconds
6635 (default 5000 = 5.0 seconds).
6637 usbcore.nousb [USB] Disable the USB subsystem
6640 [USB] A list of quirk entries to augment the built-in
6641 usb core quirk list. List entries are separated by
6642 commas. Each entry has the form
6643 VendorID:ProductID:Flags. The IDs are 4-digit hex
6644 numbers and Flags is a set of letters. Each letter
6645 will change the built-in quirk; setting it if it is
6646 clear and clearing it if it is set. The letters have
6647 the following meanings:
6648 a = USB_QUIRK_STRING_FETCH_255 (string
6649 descriptors must not be fetched using
6651 b = USB_QUIRK_RESET_RESUME (device can't resume
6652 correctly so reset it instead);
6653 c = USB_QUIRK_NO_SET_INTF (device can't handle
6654 Set-Interface requests);
6655 d = USB_QUIRK_CONFIG_INTF_STRINGS (device can't
6656 handle its Configuration or Interface
6658 e = USB_QUIRK_RESET (device can't be reset
6659 (e.g morph devices), don't use reset);
6660 f = USB_QUIRK_HONOR_BNUMINTERFACES (device has
6661 more interface descriptions than the
6662 bNumInterfaces count, and can't handle
6663 talking to these interfaces);
6664 g = USB_QUIRK_DELAY_INIT (device needs a pause
6665 during initialization, after we read
6666 the device descriptor);
6667 h = USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL (For
6668 high speed and super speed interrupt
6669 endpoints, the USB 2.0 and USB 3.0 spec
6670 require the interval in microframes (1
6671 microframe = 125 microseconds) to be
6672 calculated as interval = 2 ^
6674 Devices with this quirk report their
6675 bInterval as the result of this
6676 calculation instead of the exponent
6677 variable used in the calculation);
6678 i = USB_QUIRK_DEVICE_QUALIFIER (device can't
6679 handle device_qualifier descriptor
6681 j = USB_QUIRK_IGNORE_REMOTE_WAKEUP (device
6682 generates spurious wakeup, ignore
6683 remote wakeup capability);
6684 k = USB_QUIRK_NO_LPM (device can't handle Link
6686 l = USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL
6687 (Device reports its bInterval as linear
6688 frames instead of the USB 2.0
6690 m = USB_QUIRK_DISCONNECT_SUSPEND (Device needs
6691 to be disconnected before suspend to
6692 prevent spurious wakeup);
6693 n = USB_QUIRK_DELAY_CTRL_MSG (Device needs a
6694 pause after every control message);
6695 o = USB_QUIRK_HUB_SLOW_RESET (Hub needs extra
6696 delay after resetting its port);
6697 Example: quirks=0781:5580:bk,0a5c:5834:gij
6700 [USBHID] The interval which mice are to be polled at.
6703 [USBHID] The interval which joysticks are to be polled at.
6706 [USBHID] The interval which keyboards are to be polled at.
6708 usb-storage.delay_use=
6709 [UMS] The delay in seconds before a new device is
6710 scanned for Logical Units (default 1).
6713 [UMS] A list of quirks entries to supplement or
6714 override the built-in unusual_devs list. List
6715 entries are separated by commas. Each entry has
6716 the form VID:PID:Flags where VID and PID are Vendor
6717 and Product ID values (4-digit hex numbers) and
6718 Flags is a set of characters, each corresponding
6719 to a common usb-storage quirk flag as follows:
6720 a = SANE_SENSE (collect more than 18 bytes
6721 of sense data, not on uas);
6722 b = BAD_SENSE (don't collect more than 18
6723 bytes of sense data, not on uas);
6724 c = FIX_CAPACITY (decrease the reported
6725 device capacity by one sector);
6726 d = NO_READ_DISC_INFO (don't use
6727 READ_DISC_INFO command, not on uas);
6728 e = NO_READ_CAPACITY_16 (don't use
6729 READ_CAPACITY_16 command);
6730 f = NO_REPORT_OPCODES (don't use report opcodes
6732 g = MAX_SECTORS_240 (don't transfer more than
6733 240 sectors at a time, uas only);
6734 h = CAPACITY_HEURISTICS (decrease the
6735 reported device capacity by one
6736 sector if the number is odd);
6737 i = IGNORE_DEVICE (don't bind to this
6739 j = NO_REPORT_LUNS (don't use report luns
6741 k = NO_SAME (do not use WRITE_SAME, uas only)
6742 l = NOT_LOCKABLE (don't try to lock and
6743 unlock ejectable media, not on uas);
6744 m = MAX_SECTORS_64 (don't transfer more
6745 than 64 sectors = 32 KB at a time,
6747 n = INITIAL_READ10 (force a retry of the
6748 initial READ(10) command, not on uas);
6749 o = CAPACITY_OK (accept the capacity
6750 reported by the device, not on uas);
6751 p = WRITE_CACHE (the device cache is ON
6752 by default, not on uas);
6753 r = IGNORE_RESIDUE (the device reports
6754 bogus residue values, not on uas);
6755 s = SINGLE_LUN (the device has only one
6757 t = NO_ATA_1X (don't allow ATA(12) and ATA(16)
6758 commands, uas only);
6759 u = IGNORE_UAS (don't bind to the uas driver);
6760 w = NO_WP_DETECT (don't test whether the
6761 medium is write-protected).
6762 y = ALWAYS_SYNC (issue a SYNCHRONIZE_CACHE
6763 even if the device claims no cache,
6765 Example: quirks=0419:aaf5:rl,0421:0433:rc
6767 user_debug= [KNL,ARM]
6769 See arch/arm/Kconfig.debug help text.
6770 1 - undefined instruction events
6772 4 - invalid data aborts
6775 Example: user_debug=31
6778 [X86] Flags controlling user PTE allocations.
6780 nohigh = do not allocate PTE pages in
6781 HIGHMEM regardless of setting
6784 vdso= [X86,SH,SPARC]
6785 On X86_32, this is an alias for vdso32=. Otherwise:
6787 vdso=1: enable VDSO (the default)
6788 vdso=0: disable VDSO mapping
6790 vdso32= [X86] Control the 32-bit vDSO
6791 vdso32=1: enable 32-bit VDSO
6792 vdso32=0 or vdso32=2: disable 32-bit VDSO
6794 See the help text for CONFIG_COMPAT_VDSO for more
6795 details. If CONFIG_COMPAT_VDSO is set, the default is
6796 vdso32=0; otherwise, the default is vdso32=1.
6798 For compatibility with older kernels, vdso32=2 is an
6801 Try vdso32=0 if you encounter an error that says:
6802 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed!
6805 vector=percpu: enable percpu vector domain
6807 video= [FB] Frame buffer configuration
6808 See Documentation/fb/modedb.rst.
6810 video.brightness_switch_enabled= [ACPI]
6812 If set to 1, on receiving an ACPI notify event
6813 generated by hotkey, video driver will adjust brightness
6814 level and then send out the event to user space through
6815 the allocated input device. If set to 0, video driver
6816 will only send out the event without touching backlight
6821 [VMMIO] Memory mapped virtio (platform) device.
6823 <size>@<baseaddr>:<irq>[:<id>]
6825 <size> := size (can use standard suffixes
6827 <baseaddr> := physical base address
6828 <irq> := interrupt number (as passed to
6830 <id> := (optional) platform device id
6832 virtio_mmio.device=1K@0x100b0000:48:7
6834 Can be used multiple times for multiple devices.
6836 vga= [BOOT,X86-32] Select a particular video mode
6837 See Documentation/arch/x86/boot.rst and
6838 Documentation/admin-guide/svga.rst.
6839 Use vga=ask for menu.
6840 This is actually a boot loader parameter; the value is
6841 passed to the kernel using a special protocol.
6843 vm_debug[=options] [KNL] Available with CONFIG_DEBUG_VM=y.
6844 May slow down system boot speed, especially when
6845 enabled on systems with a large amount of memory.
6846 All options are enabled by default, and this
6847 interface is meant to allow for selectively
6848 enabling or disabling specific virtual memory
6851 Available options are:
6852 P Enable page structure init time poisoning
6853 - Disable all of the above options
6855 vmalloc=nn[KMG] [KNL,BOOT] Forces the vmalloc area to have an exact
6856 size of <nn>. This can be used to increase the
6857 minimum size (128MB on x86). It can also be used to
6858 decrease the size and leave more room for directly
6861 vmcp_cma=nn[MG] [KNL,S390]
6862 Sets the memory size reserved for contiguous memory
6863 allocations for the vmcp device driver.
6865 vmhalt= [KNL,S390] Perform z/VM CP command after system halt.
6868 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic.
6871 vmpoff= [KNL,S390] Perform z/VM CP command after power off.
6875 Controls the behavior of vsyscalls (i.e. calls to
6876 fixed addresses of 0xffffffffff600x00 from legacy
6877 code). Most statically-linked binaries and older
6878 versions of glibc use these calls. Because these
6879 functions are at fixed addresses, they make nice
6880 targets for exploits that can control RIP.
6882 emulate Vsyscalls turn into traps and are emulated
6883 reasonably safely. The vsyscall page is
6886 xonly [default] Vsyscalls turn into traps and are
6887 emulated reasonably safely. The vsyscall
6888 page is not readable.
6890 none Vsyscalls don't work at all. This makes
6891 them quite hard to use for exploits but
6892 might break your system.
6894 vt.color= [VT] Default text color.
6895 Format: 0xYX, X = foreground, Y = background.
6896 Default: 0x07 = light gray on black.
6898 vt.cur_default= [VT] Default cursor shape.
6899 Format: 0xCCBBAA, where AA, BB, and CC are the same as
6900 the parameters of the <Esc>[?A;B;Cc escape sequence;
6901 see VGA-softcursor.txt. Default: 2 = underline.
6903 vt.default_blu= [VT]
6904 Format: <blue0>,<blue1>,<blue2>,...,<blue15>
6905 Change the default blue palette of the console.
6906 This is a 16-member array composed of values
6909 vt.default_grn= [VT]
6910 Format: <green0>,<green1>,<green2>,...,<green15>
6911 Change the default green palette of the console.
6912 This is a 16-member array composed of values
6915 vt.default_red= [VT]
6916 Format: <red0>,<red1>,<red2>,...,<red15>
6917 Change the default red palette of the console.
6918 This is a 16-member array composed of values
6924 Set system-wide default UTF-8 mode for all tty's.
6925 Default is 1, i.e. UTF-8 mode is enabled for all
6926 newly opened terminals.
6928 vt.global_cursor_default=
6931 Set system-wide default for whether a cursor
6932 is shown on new VTs. Default is -1,
6933 i.e. cursors will be created by default unless
6934 overridden by individual drivers. 0 will hide
6935 cursors, 1 will display them.
6937 vt.italic= [VT] Default color for italic text; 0-15.
6940 vt.underline= [VT] Default color for underlined text; 0-15.
6943 watchdog timers [HW,WDT] For information on watchdog timers,
6944 see Documentation/watchdog/watchdog-parameters.rst
6945 or other driver-specific files in the
6946 Documentation/watchdog/ directory.
6950 Set the hard lockup detector stall duration
6951 threshold in seconds. The soft lockup detector
6952 threshold is set to twice the value. A value of 0
6953 disables both lockup detectors. Default is 10
6956 workqueue.watchdog_thresh=
6957 If CONFIG_WQ_WATCHDOG is configured, workqueue can
6958 warn stall conditions and dump internal state to
6959 help debugging. 0 disables workqueue stall
6960 detection; otherwise, it's the stall threshold
6961 duration in seconds. The default value is 30 and
6962 it can be updated at runtime by writing to the
6963 corresponding sysfs file.
6965 workqueue.disable_numa
6966 By default, all work items queued to unbound
6967 workqueues are affine to the NUMA nodes they're
6968 issued on, which results in better behavior in
6969 general. If NUMA affinity needs to be disabled for
6970 whatever reason, this option can be used. Note
6971 that this also can be controlled per-workqueue for
6972 workqueues visible under /sys/bus/workqueue/.
6974 workqueue.power_efficient
6975 Per-cpu workqueues are generally preferred because
6976 they show better performance thanks to cache
6977 locality; unfortunately, per-cpu workqueues tend to
6978 be more power hungry than unbound workqueues.
6980 Enabling this makes the per-cpu workqueues which
6981 were observed to contribute significantly to power
6982 consumption unbound, leading to measurably lower
6983 power usage at the cost of small performance
6986 The default value of this parameter is determined by
6987 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT.
6989 workqueue.debug_force_rr_cpu
6990 Workqueue used to implicitly guarantee that work
6991 items queued without explicit CPU specified are put
6992 on the local CPU. This guarantee is no longer true
6993 and while local CPU is still preferred work items
6994 may be put on foreign CPUs. This debug option
6995 forces round-robin CPU selection to flush out
6996 usages which depend on the now broken guarantee.
6997 When enabled, memory and cache locality will be
7000 writecombine= [LOONGARCH] Control the MAT (Memory Access Type) of
7003 on - Enable writecombine, use WUC for ioremap_wc()
7004 off - Disable writecombine, use SUC for ioremap_wc()
7006 x2apic_phys [X86-64,APIC] Use x2apic physical mode instead of
7007 default x2apic cluster mode on platforms
7010 xen_512gb_limit [KNL,X86-64,XEN]
7011 Restricts the kernel running paravirtualized under Xen
7012 to use only up to 512 GB of RAM. The reason to do so is
7013 crash analysis tools and Xen tools for doing domain
7014 save/restore/migration must be enabled to handle larger
7017 xen_emul_unplug= [HW,X86,XEN]
7018 Unplug Xen emulated devices
7019 Format: [unplug0,][unplug1]
7020 ide-disks -- unplug primary master IDE devices
7021 aux-ide-disks -- unplug non-primary-master IDE devices
7022 nics -- unplug network devices
7023 all -- unplug all emulated devices (NICs and IDE disks)
7024 unnecessary -- unplugging emulated devices is
7025 unnecessary even if the host did not respond to
7027 never -- do not unplug even if version check succeeds
7029 xen_legacy_crash [X86,XEN]
7030 Crash from Xen panic notifier, without executing late
7031 panic() code such as dumping handler.
7033 xen_msr_safe= [X86,XEN]
7035 Select whether to always use non-faulting (safe) MSR
7036 access functions when running as Xen PV guest. The
7037 default value is controlled by CONFIG_XEN_PV_MSR_SAFE.
7039 xen_nopvspin [X86,XEN]
7040 Disables the qspinlock slowpath using Xen PV optimizations.
7041 This parameter is obsoleted by "nopvspin" parameter, which
7042 has equivalent effect for XEN platform.
7045 Disables the PV optimizations forcing the HVM guest to
7046 run as generic HVM guest with no PV drivers.
7047 This option is obsoleted by the "nopv" option, which
7048 has equivalent effect for XEN platform.
7050 xen_no_vector_callback
7051 [KNL,X86,XEN] Disable the vector callback for Xen
7052 event channel interrupts.
7054 xen_scrub_pages= [XEN]
7055 Boolean option to control scrubbing pages before giving them back
7056 to Xen, for use by other domains. Can be also changed at runtime
7057 with /sys/devices/system/xen_memory/xen_memory0/scrub_pages.
7058 Default value controlled with CONFIG_XEN_SCRUB_PAGES_DEFAULT.
7060 xen_timer_slop= [X86-64,XEN]
7061 Set the timer slop (in nanoseconds) for the virtual Xen
7062 timers (default is 100000). This adjusts the minimum
7063 delta of virtualized Xen timers, where lower values
7064 improve timer resolution at the expense of processing
7065 more timer interrupts.
7067 xen.balloon_boot_timeout= [XEN]
7068 The time (in seconds) to wait before giving up to boot
7069 in case initial ballooning fails to free enough memory.
7070 Applies only when running as HVM or PVH guest and
7071 started with less memory configured than allowed at
7072 max. Default is 180.
7074 xen.event_eoi_delay= [XEN]
7075 How long to delay EOI handling in case of event
7076 storms (jiffies). Default is 10.
7078 xen.event_loop_timeout= [XEN]
7079 After which time (jiffies) the event handling loop
7080 should start to delay EOI handling. Default is 2.
7082 xen.fifo_events= [XEN]
7083 Boolean parameter to disable using fifo event handling
7084 even if available. Normally fifo event handling is
7085 preferred over the 2-level event handling, as it is
7086 fairer and the number of possible event channels is
7087 much higher. Default is on (use fifo events).
7089 xirc2ps_cs= [NET,PCMCIA]
7091 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]]
7094 By default on POWER9 and above, the kernel will
7095 natively use the XIVE interrupt controller. This option
7096 allows the fallback firmware mode to be used:
7098 off Fallback to firmware control of XIVE interrupt
7099 controller on both pseries and powernv
7100 platforms. Only useful on POWER9 and above.
7102 xive.store-eoi=off [PPC]
7103 By default on POWER10 and above, the kernel will use
7104 stores for EOI handling when the XIVE interrupt mode
7105 is active. This option allows the XIVE driver to use
7106 loads instead, as on POWER9.
7108 xhci-hcd.quirks [USB,KNL]
7109 A hex value specifying bitmask with supplemental xhci
7110 host controller quirks. Meaning of each bit can be
7111 consulted in header drivers/usb/host/xhci.h.
7114 Format: { early | on | rw | ro | off }
7115 Controls if xmon debugger is enabled. Default is off.
7116 Passing only "xmon" is equivalent to "xmon=early".
7117 early Call xmon as early as possible on boot; xmon
7118 debugger is called from setup_arch().
7119 on xmon debugger hooks will be installed so xmon
7120 is only called on a kernel crash. Default mode,
7121 i.e. either "ro" or "rw" mode, is controlled
7122 with CONFIG_XMON_DEFAULT_RO_MODE.
7123 rw xmon debugger hooks will be installed so xmon
7124 is called only on a kernel crash, mode is write,
7125 meaning SPR registers, memory and, other data
7126 can be written using xmon commands.
7127 ro same as "rw" option above but SPR registers,
7128 memory, and other data can't be written using
7130 off xmon is disabled.