1 acpi= [HW,ACPI,X86,ARM64]
2 Advanced Configuration and Power Interface
3 Format: { force | on | off | strict | noirq | rsdt |
5 force -- enable ACPI if default was off
6 on -- enable ACPI but allow fallback to DT [arm64]
7 off -- disable ACPI if default was on
8 noirq -- do not use ACPI for IRQ routing
9 strict -- Be less tolerant of platforms that are not
10 strictly ACPI specification compliant.
11 rsdt -- prefer RSDT over (default) XSDT
12 copy_dsdt -- copy DSDT to memory
13 For ARM64, ONLY "acpi=off", "acpi=on" or "acpi=force"
16 See also Documentation/power/runtime_pm.rst, pci=noacpi
18 acpi_apic_instance= [ACPI, IOAPIC]
20 2: use 2nd APIC table, if available
21 1,0: use 1st APIC table
24 acpi_backlight= [HW,ACPI]
25 { vendor | video | native | none }
26 If set to vendor, prefer vendor-specific driver
27 (e.g. thinkpad_acpi, sony_acpi, etc.) instead
28 of the ACPI video.ko driver.
29 If set to video, use the ACPI video.ko driver.
30 If set to native, use the device's native backlight mode.
31 If set to none, disable the ACPI backlight interface.
33 acpi_force_32bit_fadt_addr
34 force FADT to use 32 bit addresses rather than the
35 64 bit X_* addresses. Some firmware have broken 64
36 bit addresses for force ACPI ignore these and use
37 the older legacy 32 bit addresses.
39 acpica_no_return_repair [HW, ACPI]
40 Disable AML predefined validation mechanism
41 This mechanism can repair the evaluation result to make
42 the return objects more ACPI specification compliant.
43 This option is useful for developers to identify the
44 root cause of an AML interpreter issue when the issue
45 has something to do with the repair mechanism.
47 acpi.debug_layer= [HW,ACPI,ACPI_DEBUG]
48 acpi.debug_level= [HW,ACPI,ACPI_DEBUG]
50 CONFIG_ACPI_DEBUG must be enabled to produce any ACPI
51 debug output. Bits in debug_layer correspond to a
52 _COMPONENT in an ACPI source file, e.g.,
53 #define _COMPONENT ACPI_EVENTS
54 Bits in debug_level correspond to a level in
55 ACPI_DEBUG_PRINT statements, e.g.,
56 ACPI_DEBUG_PRINT((ACPI_DB_INFO, ...
57 The debug_level mask defaults to "info". See
58 Documentation/firmware-guide/acpi/debug.rst for more information about
59 debug layers and levels.
61 Enable processor driver info messages:
62 acpi.debug_layer=0x20000000
63 Enable AML "Debug" output, i.e., stores to the Debug
64 object while interpreting AML:
65 acpi.debug_layer=0xffffffff acpi.debug_level=0x2
66 Enable all messages related to ACPI hardware:
67 acpi.debug_layer=0x2 acpi.debug_level=0xffffffff
69 Some values produce so much output that the system is
70 unusable. The "log_buf_len" parameter may be useful
71 if you need to capture more output.
73 acpi_enforce_resources= [ACPI]
75 Check for resource conflicts between native drivers
76 and ACPI OperationRegions (SystemIO and SystemMemory
77 only). IO ports and memory declared in ACPI might be
78 used by the ACPI subsystem in arbitrary AML code and
79 can interfere with legacy drivers.
80 strict (default): access to resources claimed by ACPI
81 is denied; legacy drivers trying to access reserved
82 resources will fail to bind to device using them.
83 lax: access to resources claimed by ACPI is allowed;
84 legacy drivers trying to access reserved resources
85 will bind successfully but a warning message is logged.
86 no: ACPI OperationRegions are not marked as reserved,
87 no further checks are performed.
89 acpi_force_table_verification [HW,ACPI]
90 Enable table checksum verification during early stage.
91 By default, this is disabled due to x86 early mapping
94 acpi_irq_balance [HW,ACPI]
95 ACPI will balance active IRQs
98 acpi_irq_nobalance [HW,ACPI]
99 ACPI will not move active IRQs (default)
102 acpi_irq_isa= [HW,ACPI] If irq_balance, mark listed IRQs used by ISA
103 Format: <irq>,<irq>...
105 acpi_irq_pci= [HW,ACPI] If irq_balance, clear listed IRQs for
107 Format: <irq>,<irq>...
109 acpi_mask_gpe= [HW,ACPI]
110 Due to the existence of _Lxx/_Exx, some GPEs triggered
111 by unsupported hardware/firmware features can result in
112 GPE floodings that cannot be automatically disabled by
114 This facility can be used to prevent such uncontrolled
116 Format: <byte> or <bitmap-list>
118 acpi_no_auto_serialize [HW,ACPI]
119 Disable auto-serialization of AML methods
120 AML control methods that contain the opcodes to create
121 named objects will be marked as "Serialized" by the
122 auto-serialization feature.
123 This feature is enabled by default.
124 This option allows to turn off the feature.
126 acpi_no_memhotplug [ACPI] Disable memory hotplug. Useful for kdump
129 acpi_no_static_ssdt [HW,ACPI]
130 Disable installation of static SSDTs at early boot time
131 By default, SSDTs contained in the RSDT/XSDT will be
132 installed automatically and they will appear under
133 /sys/firmware/acpi/tables.
134 This option turns off this feature.
135 Note that specifying this option does not affect
136 dynamic table installation which will install SSDT
137 tables to /sys/firmware/acpi/tables/dynamic.
139 acpi_no_watchdog [HW,ACPI,WDT]
140 Ignore the ACPI-based watchdog interface (WDAT) and let
141 a native driver control the watchdog device instead.
143 acpi_rsdp= [ACPI,EFI,KEXEC]
144 Pass the RSDP address to the kernel, mostly used
145 on machines running EFI runtime service to boot the
146 second kernel for kdump.
148 acpi_os_name= [HW,ACPI] Tell ACPI BIOS the name of the OS
149 Format: To spoof as Windows 98: ="Microsoft Windows"
151 acpi_rev_override [ACPI] Override the _REV object to return 5 (instead
152 of 2 which is mandated by ACPI 6) as the supported ACPI
153 specification revision (when using this switch, it may
154 be necessary to carry out a cold reboot _twice_ in a
155 row to make it take effect on the platform firmware).
157 acpi_osi= [HW,ACPI] Modify list of supported OS interface strings
158 acpi_osi="string1" # add string1
159 acpi_osi="!string2" # remove string2
160 acpi_osi=!* # remove all strings
161 acpi_osi=! # disable all built-in OS vendor
163 acpi_osi=!! # enable all built-in OS vendor
165 acpi_osi= # disable all strings
167 'acpi_osi=!' can be used in combination with single or
168 multiple 'acpi_osi="string1"' to support specific OS
169 vendor string(s). Note that such command can only
170 affect the default state of the OS vendor strings, thus
171 it cannot affect the default state of the feature group
172 strings and the current state of the OS vendor strings,
173 specifying it multiple times through kernel command line
174 is meaningless. This command is useful when one do not
175 care about the state of the feature group strings which
176 should be controlled by the OSPM.
178 1. 'acpi_osi=! acpi_osi="Windows 2000"' is equivalent
179 to 'acpi_osi="Windows 2000" acpi_osi=!', they all
180 can make '_OSI("Windows 2000")' TRUE.
182 'acpi_osi=' cannot be used in combination with other
183 'acpi_osi=' command lines, the _OSI method will not
184 exist in the ACPI namespace. NOTE that such command can
185 only affect the _OSI support state, thus specifying it
186 multiple times through kernel command line is also
189 1. 'acpi_osi=' can make 'CondRefOf(_OSI, Local1)'
192 'acpi_osi=!*' can be used in combination with single or
193 multiple 'acpi_osi="string1"' to support specific
194 string(s). Note that such command can affect the
195 current state of both the OS vendor strings and the
196 feature group strings, thus specifying it multiple times
197 through kernel command line is meaningful. But it may
198 still not able to affect the final state of a string if
199 there are quirks related to this string. This command
200 is useful when one want to control the state of the
201 feature group strings to debug BIOS issues related to
204 1. 'acpi_osi="Module Device" acpi_osi=!*' can make
205 '_OSI("Module Device")' FALSE.
206 2. 'acpi_osi=!* acpi_osi="Module Device"' can make
207 '_OSI("Module Device")' TRUE.
208 3. 'acpi_osi=! acpi_osi=!* acpi_osi="Windows 2000"' is
210 'acpi_osi=!* acpi_osi=! acpi_osi="Windows 2000"'
212 'acpi_osi=!* acpi_osi="Windows 2000" acpi_osi=!',
213 they all will make '_OSI("Windows 2000")' TRUE.
216 Override the pmtimer bug detection: force the kernel
217 to assume that this machine's pmtimer latches its value
218 and always returns good values.
220 acpi_sci= [HW,ACPI] ACPI System Control Interrupt trigger mode
221 Format: { level | edge | high | low }
223 acpi_skip_timer_override [HW,ACPI]
224 Recognize and ignore IRQ0/pin2 Interrupt Override.
225 For broken nForce2 BIOS resulting in XT-PIC timer.
227 acpi_sleep= [HW,ACPI] Sleep options
228 Format: { s3_bios, s3_mode, s3_beep, s4_hwsig,
229 s4_nohwsig, old_ordering, nonvs,
230 sci_force_enable, nobl }
231 See Documentation/power/video.rst for information on
233 s3_beep is for debugging; it makes the PC's speaker beep
234 as soon as the kernel's real-mode entry point is called.
235 s4_hwsig causes the kernel to check the ACPI hardware
236 signature during resume from hibernation, and gracefully
237 refuse to resume if it has changed. This complies with
238 the ACPI specification but not with reality, since
239 Windows does not do this and many laptops do change it
240 on docking. So the default behaviour is to allow resume
241 and simply warn when the signature changes, unless the
242 s4_hwsig option is enabled.
243 s4_nohwsig prevents ACPI hardware signature from being
244 used (or even warned about) during resume.
245 old_ordering causes the ACPI 1.0 ordering of the _PTS
246 control method, with respect to putting devices into
247 low power states, to be enforced (the ACPI 2.0 ordering
248 of _PTS is used by default).
249 nonvs prevents the kernel from saving/restoring the
250 ACPI NVS memory during suspend/hibernation and resume.
251 sci_force_enable causes the kernel to set SCI_EN directly
252 on resume from S1/S3 (which is against the ACPI spec,
253 but some broken systems don't work without it).
254 nobl causes the internal blacklist of systems known to
255 behave incorrectly in some ways with respect to system
256 suspend and resume to be ignored (use wisely).
258 acpi_use_timer_override [HW,ACPI]
259 Use timer override. For some broken Nvidia NF5 boards
260 that require a timer override, but don't have HPET
262 add_efi_memmap [EFI; X86] Include EFI memory map in
263 kernel's map of available physical RAM.
266 { off | try_unsupported }
267 off: disable AGP support
268 try_unsupported: try to drive unsupported chipsets
269 (may crash computer or cause data corruption)
272 See Documentation/sound/alsa-configuration.rst
275 Allow the default userspace alignment fault handler
276 behaviour to be specified. Bit 0 enables warnings,
277 bit 1 enables fixups, and bit 2 sends a segfault.
279 align_va_addr= [X86-64]
280 Align virtual addresses by clearing slice [14:12] when
281 allocating a VMA at process creation time. This option
282 gives you up to 3% performance improvement on AMD F15h
283 machines (where it is enabled by default) for a
284 CPU-intensive style benchmark, and it can vary highly in
285 a microbenchmark depending on workload and compiler.
287 32: only for 32-bit processes
288 64: only for 64-bit processes
289 on: enable for both 32- and 64-bit processes
290 off: disable for both 32- and 64-bit processes
292 alloc_snapshot [FTRACE]
293 Allocate the ftrace snapshot buffer on boot up when the
294 main buffer is allocated. This is handy if debugging
295 and you need to use tracing_snapshot() on boot up, and
296 do not want to use tracing_snapshot_alloc() as it needs
297 to be done where GFP_KERNEL allocations are allowed.
299 allow_mismatched_32bit_el0 [ARM64]
300 Allow execve() of 32-bit applications and setting of the
301 PER_LINUX32 personality on systems where only a strict
302 subset of the CPUs support 32-bit EL0. When this
303 parameter is present, the set of CPUs supporting 32-bit
304 EL0 is indicated by /sys/devices/system/cpu/aarch32_el0
305 and hot-unplug operations may be restricted.
307 See Documentation/arm64/asymmetric-32bit.rst for more
310 amd_iommu= [HW,X86-64]
311 Pass parameters to the AMD IOMMU driver in the system.
313 fullflush - Deprecated, equivalent to iommu.strict=1
314 off - do not initialize any AMD IOMMU found in
316 force_isolation - Force device isolation for all
317 devices. The IOMMU driver is not
318 allowed anymore to lift isolation
319 requirements as needed. This option
320 does not override iommu=pt
321 force_enable - Force enable the IOMMU on platforms known
322 to be buggy with IOMMU enabled. Use this
325 amd_iommu_dump= [HW,X86-64]
326 Enable AMD IOMMU driver option to dump the ACPI table
327 for AMD IOMMU. With this option enabled, AMD IOMMU
328 driver will print ACPI tables for AMD IOMMU during
329 IOMMU initialization.
331 amd_iommu_intr= [HW,X86-64]
332 Specifies one of the following AMD IOMMU interrupt
334 legacy - Use legacy interrupt remapping mode.
335 vapic - Use virtual APIC mode, which allows IOMMU
336 to inject interrupts directly into guest.
337 This mode requires kvm-amd.avic=1.
338 (Default when IOMMU HW support is present.)
340 amijoy.map= [HW,JOY] Amiga joystick support
341 Map of devices attached to JOY0DAT and JOY1DAT
343 See also Documentation/input/joydev/joystick.rst
345 analog.map= [HW,JOY] Analog joystick and gamepad support
346 Specifies type or capabilities of an analog joystick
347 connected to one of 16 gameports
348 Format: <type1>,<type2>,..<type16>
351 Power management functions (SPARCstation-4/5 + deriv.)
353 Disable APC CPU standby support. SPARCstation-Fox does
354 not play well with APC CPU idle - disable it if you have
355 APC and your system crashes randomly.
357 apic= [APIC,X86] Advanced Programmable Interrupt Controller
358 Change the output verbosity while booting
359 Format: { quiet (default) | verbose | debug }
360 Change the amount of debugging information output
361 when initialising the APIC and IO-APIC components.
362 For X86-32, this can also be used to specify an APIC
364 Format: apic=driver_name
365 Examples: apic=bigsmp
367 apic_extnmi= [APIC,X86] External NMI delivery setting
368 Format: { bsp (default) | all | none }
369 bsp: External NMI is delivered only to CPU 0
370 all: External NMIs are broadcast to all CPUs as a
372 none: External NMI is masked for all CPUs. This is
373 useful so that a dump capture kernel won't be
377 See Documentation/networking/ipv6.rst.
379 show_lapic= [APIC,X86] Advanced Programmable Interrupt Controller
380 Limit apic dumping. The parameter defines the maximal
381 number of local apics being dumped. Also it is possible
382 to set it to "all" by meaning -- no limit here.
383 Format: { 1 (default) | 2 | ... | all }.
384 The parameter valid if only apic=debug or
385 apic=verbose is specified.
386 Example: apic=debug show_lapic=all
388 apm= [APM] Advanced Power Management
389 See header of arch/x86/kernel/apm_32.c.
391 arcrimi= [HW,NET] ARCnet - "RIM I" (entirely mem-mapped) cards
392 Format: <io>,<irq>,<nodeID>
394 arm64.nobti [ARM64] Unconditionally disable Branch Target
395 Identification support
397 arm64.nopauth [ARM64] Unconditionally disable Pointer Authentication
400 arm64.nomte [ARM64] Unconditionally disable Memory Tagging Extension
403 arm64.nosve [ARM64] Unconditionally disable Scalable Vector
406 arm64.nosme [ARM64] Unconditionally disable Scalable Matrix
411 atarimouse= [HW,MOUSE] Atari Mouse
413 atkbd.extra= [HW] Enable extra LEDs and keys on IBM RapidAccess,
414 EzKey and similar keyboards
416 atkbd.reset= [HW] Reset keyboard during initialization
418 atkbd.set= [HW] Select keyboard code set
419 Format: <int> (2 = AT (default), 3 = PS/2)
421 atkbd.scroll= [HW] Enable scroll wheel on MS Office and similar
424 atkbd.softraw= [HW] Choose between synthetic and real raw mode
425 Format: <bool> (0 = real, 1 = synthetic (default))
427 atkbd.softrepeat= [HW]
428 Use software keyboard repeat
430 audit= [KNL] Enable the audit sub-system
431 Format: { "0" | "1" | "off" | "on" }
432 0 | off - kernel audit is disabled and can not be
433 enabled until the next reboot
434 unset - kernel audit is initialized but disabled and
435 will be fully enabled by the userspace auditd.
436 1 | on - kernel audit is initialized and partially
437 enabled, storing at most audit_backlog_limit
438 messages in RAM until it is fully enabled by the
442 audit_backlog_limit= [KNL] Set the audit queue size limit.
443 Format: <int> (must be >=0)
446 bau= [X86_UV] Enable the BAU on SGI UV. The default
447 behavior is to disable the BAU (i.e. bau=0).
448 Format: { "0" | "1" }
451 unset - Disable the BAU.
453 baycom_epp= [HW,AX25]
456 baycom_par= [HW,AX25] BayCom Parallel Port AX.25 Modem
458 See header of drivers/net/hamradio/baycom_par.c.
460 baycom_ser_fdx= [HW,AX25]
461 BayCom Serial Port AX.25 Modem (Full Duplex Mode)
462 Format: <io>,<irq>,<mode>[,<baud>]
463 See header of drivers/net/hamradio/baycom_ser_fdx.c.
465 baycom_ser_hdx= [HW,AX25]
466 BayCom Serial Port AX.25 Modem (Half Duplex Mode)
467 Format: <io>,<irq>,<mode>
468 See header of drivers/net/hamradio/baycom_ser_hdx.c.
471 Disable BERT OS support on buggy BIOSes.
473 bgrt_disable [ACPI][X86]
474 Disable BGRT to avoid flickering OEM logo.
476 blkdevparts= Manual partition parsing of block device(s) for
477 embedded devices based on command line input.
478 See Documentation/block/cmdline-partition.rst
480 boot_delay= Milliseconds to delay each printk during boot.
481 Values larger than 10 seconds (10000) are changed to
486 Extended command line options can be added to an initrd
487 and this will cause the kernel to look for it.
489 See Documentation/admin-guide/bootconfig.rst
491 bttv.card= [HW,V4L] bttv (bt848 + bt878 based grabber cards)
492 bttv.radio= Most important insmod options are available as
494 bttv.pll= See Documentation/admin-guide/media/bttv.rst
497 bulk_remove=off [PPC] This parameter disables the use of the pSeries
498 firmware feature for flushing multiple hpte entries
501 c101= [NET] Moxa C101 synchronous serial card
503 cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection.
504 Sometimes CPU hardware bugs make them report the cache
505 size incorrectly. The kernel will attempt work arounds
506 to fix known problems, but for some CPUs it is not
507 possible to determine what the correct size should be.
508 This option provides an override for these situations.
511 [NET] Specifies amount of time (in seconds) that
512 the kernel should wait for a network carrier. By default
513 it waits 120 seconds.
515 ca_keys= [KEYS] This parameter identifies a specific key(s) on
516 the system trusted keyring to be used for certificate
518 format: { id:<keyid> | builtin }
520 cca= [MIPS] Override the kernel pages' cache coherency
521 algorithm. Accepted values range from 0 to 7
522 inclusive. See arch/mips/include/asm/pgtable-bits.h
523 for platform specific values (SB1, Loongson3 and
526 ccw_timeout_log [S390]
527 See Documentation/s390/common_io.rst for details.
529 cgroup_disable= [KNL] Disable a particular controller or optional feature
530 Format: {name of the controller(s) or feature(s) to disable}
531 The effects of cgroup_disable=foo are:
532 - foo isn't auto-mounted if you mount all cgroups in
534 - foo isn't visible as an individually mountable
536 - if foo is an optional feature then the feature is
537 disabled and corresponding cgroup files are not
539 {Currently only "memory" controller deal with this and
540 cut the overhead, others just disable the usage. So
541 only cgroup_disable=memory is actually worthy}
542 Specifying "pressure" disables per-cgroup pressure
543 stall information accounting feature
545 cgroup_no_v1= [KNL] Disable cgroup controllers and named hierarchies in v1
546 Format: { { controller | "all" | "named" }
547 [,{ controller | "all" | "named" }...] }
548 Like cgroup_disable, but only applies to cgroup v1;
549 the blacklisted controllers remain available in cgroup2.
550 "all" blacklists all controllers and "named" disables
551 named mounts. Specifying both "all" and "named" disables
554 cgroup.memory= [KNL] Pass options to the cgroup memory controller.
556 nosocket -- Disable socket memory accounting.
557 nokmem -- Disable kernel memory accounting.
559 checkreqprot= [SELINUX] Set initial checkreqprot flag value.
560 Format: { "0" | "1" }
561 See security/selinux/Kconfig help text.
562 0 -- check protection applied by kernel (includes
563 any implied execute protection).
564 1 -- check protection requested by application.
565 Default value is set via a kernel config option.
566 Value can be changed at runtime via
567 /sys/fs/selinux/checkreqprot.
568 Setting checkreqprot to 1 is deprecated.
571 See Documentation/s390/common_io.rst for details.
573 clearcpuid=X[,X...] [X86]
574 Disable CPUID feature X for the kernel. See
575 arch/x86/include/asm/cpufeatures.h for the valid bit
576 numbers X. Note the Linux-specific bits are not necessarily
577 stable over kernel options, but the vendor-specific
579 X can also be a string as appearing in the flags: line
580 in /proc/cpuinfo which does not have the above
581 instability issue. However, not all features have names
583 Note that using this option will taint your kernel.
584 Also note that user programs calling CPUID directly
585 or using the feature without checking anything
586 will still see it. This just prevents it from
587 being used by the kernel or shown in /proc/cpuinfo.
588 Also note the kernel might malfunction if you disable
593 Prevents the clock framework from automatically gating
594 clocks that have not been explicitly enabled by a Linux
595 device driver but are enabled in hardware at reset or
596 by the bootloader/firmware. Note that this does not
597 force such clocks to be always-on nor does it reserve
598 those clocks in any way. This parameter is useful for
599 debug and development, but should not be needed on a
600 platform with proper driver support. For more
601 information, see Documentation/driver-api/clk.rst.
603 clock= [BUGS=X86-32, HW] gettimeofday clocksource override.
605 Forces specified clocksource (if available) to be used
606 when calculating gettimeofday(). If specified
607 clocksource is not available, it defaults to PIT.
608 Format: { pit | tsc | cyclone | pmtmr }
610 clocksource= Override the default clocksource
612 Override the default clocksource and use the clocksource
613 with the name specified.
614 Some clocksource names to choose from, depending on
616 [all] jiffies (this is the base, fallback clocksource)
618 [ARM] imx_timer1,OSTS,netx_timer,mpu_timer2,
619 pxa_timer,timer3,32k_counter,timer0_1
620 [X86-32] pit,hpet,tsc;
621 scx200_hrt on Geode; cyclone on IBM x440
629 clocksource.arm_arch_timer.evtstrm=
632 Enable/disable the eventstream feature of the ARM
633 architected timer so that code using WFE-based polling
634 loops can be debugged more effectively on production
637 clocksource.max_cswd_read_retries= [KNL]
638 Number of clocksource_watchdog() retries due to
639 external delays before the clock will be marked
640 unstable. Defaults to two retries, that is,
641 three attempts to read the clock under test.
643 clocksource.verify_n_cpus= [KNL]
644 Limit the number of CPUs checked for clocksources
645 marked with CLOCK_SOURCE_VERIFY_PERCPU that
646 are marked unstable due to excessive skew.
647 A negative value says to check all CPUs, while
648 zero says not to check any. Values larger than
649 nr_cpu_ids are silently truncated to nr_cpu_ids.
650 The actual CPUs are chosen randomly, with
651 no replacement if the same CPU is chosen twice.
653 clocksource-wdtest.holdoff= [KNL]
654 Set the time in seconds that the clocksource
655 watchdog test waits before commencing its tests.
656 Defaults to zero when built as a module and to
657 10 seconds when built into the kernel.
659 cma=nn[MG]@[start[MG][-end[MG]]]
661 Sets the size of kernel global memory area for
662 contiguous memory allocations and optionally the
663 placement constraint by the physical address range of
664 memory allocations. A value of 0 disables CMA
665 altogether. For more information, see
666 kernel/dma/contiguous.c
670 Sets the size of kernel per-numa memory area for
671 contiguous memory allocations. A value of 0 disables
672 per-numa CMA altogether. And If this option is not
673 specificed, the default value is 0.
674 With per-numa CMA enabled, DMA users on node nid will
675 first try to allocate buffer from the pernuma area
676 which is located in node nid, if the allocation fails,
677 they will fallback to the global default memory area.
679 cmo_free_hint= [PPC] Format: { yes | no }
680 Specify whether pages are marked as being inactive
681 when they are freed. This is used in CMO environments
682 to determine OS memory pressure for page stealing by
686 coherent_pool=nn[KMG] [ARM,KNL]
687 Sets the size of memory pool for coherent, atomic dma
688 allocations, by default set to 256K.
690 com20020= [HW,NET] ARCnet - COM20020 chipset
692 <io>[,<irq>[,<nodeID>[,<backplane>[,<ckp>[,<timeout>]]]]]
694 com90io= [HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers)
698 ARCnet - COM90xx chipset (memory-mapped buffers)
699 Format: <io>[,<irq>[,<memstart>]]
701 condev= [HW,S390] console device
704 console= [KNL] Output console device and options.
706 tty<n> Use the virtual console device <n>.
710 Use the specified serial port. The options are of
711 the form "bbbbpnf", where "bbbb" is the baud rate,
712 "p" is parity ("n", "o", or "e"), "n" is number of
713 bits, and "f" is flow control ("r" for RTS or
714 omit it). Default is "9600n8".
716 See Documentation/admin-guide/serial-console.rst for more
718 Documentation/networking/netconsole.rst for an
721 uart[8250],io,<addr>[,options]
722 uart[8250],mmio,<addr>[,options]
723 uart[8250],mmio16,<addr>[,options]
724 uart[8250],mmio32,<addr>[,options]
725 uart[8250],0x<addr>[,options]
726 Start an early, polled-mode console on the 8250/16550
727 UART at the specified I/O port or MMIO address,
728 switching to the matching ttyS device later.
729 MMIO inter-register address stride is either 8-bit
730 (mmio), 16-bit (mmio16), or 32-bit (mmio32).
731 If none of [io|mmio|mmio16|mmio32], <addr> is assumed
732 to be equivalent to 'mmio'. 'options' are specified in
733 the same format described for ttyS above; if unspecified,
734 the h/w is not re-initialized.
736 hvc<n> Use the hypervisor console device <n>. This is for
737 both Xen and PowerPC hypervisors.
740 Use to disable console output, i.e., to have kernel
741 console messages discarded.
742 This must be the only console= parameter used on the
745 If the device connected to the port is not a TTY but a braille
746 device, prepend "brl," before the device type, for instance
748 For now, only VisioBraille is supported.
751 [KNL] Change console messages format
753 By default we print messages on consoles in
754 "[time stamp] text\n" format (time stamp may not be
755 printed, depending on CONFIG_PRINTK_TIME or
756 `printk_time' param).
758 Switch to syslog format: "<%u>[time stamp] text\n"
759 IOW, each message will have a facility and loglevel
760 prefix. The format is similar to one used by syslog()
761 syscall, or to executing "dmesg -S --raw" or to reading
764 consoleblank= [KNL] The console blank (screen saver) timeout in
765 seconds. A value of 0 disables the blank timer.
769 [KNL] Change the default value for
770 /proc/<pid>/coredump_filter.
771 See also Documentation/filesystems/proc.rst.
773 coresight_cpu_debug.enable
776 Enable/disable the CPU sampling based debugging.
777 0: default value, disable debugging
778 1: enable debugging at boot time
780 cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver
782 <first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>]
784 cpu0_hotplug [X86] Turn on CPU0 hotplug feature when
785 CONFIG_BOOTPARAM_HOTPLUG_CPU0 is off.
786 Some features depend on CPU0. Known dependencies are:
787 1. Resume from suspend/hibernate depends on CPU0.
788 Suspend/hibernate will fail if CPU0 is offline and you
789 need to online CPU0 before suspend/hibernate.
790 2. PIC interrupts also depend on CPU0. CPU0 can't be
791 removed if a PIC interrupt is detected.
792 It's said poweroff/reboot may depend on CPU0 on some
793 machines although I haven't seen such issues so far
794 after CPU0 is offline on a few tested machines.
795 If the dependencies are under your control, you can
796 turn on cpu0_hotplug.
798 cpuidle.off=1 [CPU_IDLE]
799 disable the cpuidle sub-system
802 [CPU_IDLE] Name of the cpuidle governor to use.
804 cpufreq.off=1 [CPU_FREQ]
805 disable the cpufreq sub-system
807 cpufreq.default_governor=
808 [CPU_FREQ] Name of the default cpufreq governor or
809 policy to use. This governor must be registered in the
810 kernel before the cpufreq driver probes.
813 [X86] Delay for N microsec between assert and de-assert
814 of APIC INIT to start processors. This delay occurs
815 on every CPU online, such as boot, and resume from suspend.
818 crash_kexec_post_notifiers
819 Run kdump after running panic-notifiers and dumping
820 kmsg. This only for the users who doubt kdump always
821 succeeds in any situation.
822 Note that this also increases risks of kdump failure,
823 because some panic notifiers can make the crashed
824 kernel more unstable.
826 crashkernel=size[KMG][@offset[KMG]]
827 [KNL] Using kexec, Linux can switch to a 'crash kernel'
828 upon panic. This parameter reserves the physical
829 memory region [offset, offset + size] for that kernel
830 image. If '@offset' is omitted, then a suitable offset
831 is selected automatically.
832 [KNL, X86-64] Select a region under 4G first, and
833 fall back to reserve region above 4G when '@offset'
834 hasn't been specified.
835 See Documentation/admin-guide/kdump/kdump.rst for further details.
837 crashkernel=range1:size1[,range2:size2,...][@offset]
838 [KNL] Same as above, but depends on the memory
839 in the running system. The syntax of range is
840 start-[end] where start and end are both
841 a memory unit (amount[KMG]). See also
842 Documentation/admin-guide/kdump/kdump.rst for an example.
844 crashkernel=size[KMG],high
845 [KNL, X86-64, ARM64] range could be above 4G. Allow kernel
846 to allocate physical memory region from top, so could
847 be above 4G if system have more than 4G ram installed.
848 Otherwise memory region will be allocated below 4G, if
850 It will be ignored if crashkernel=X is specified.
851 crashkernel=size[KMG],low
852 [KNL, X86-64] range under 4G. When crashkernel=X,high
853 is passed, kernel could allocate physical memory region
854 above 4G, that cause second kernel crash on system
855 that require some amount of low memory, e.g. swiotlb
856 requires at least 64M+32K low memory, also enough extra
857 low memory is needed to make sure DMA buffers for 32-bit
858 devices won't run out. Kernel would try to allocate
859 at least 256M below 4G automatically.
860 This one lets the user specify own low range under 4G
861 for second kernel instead.
862 0: to disable low allocation.
863 It will be ignored when crashkernel=X,high is not used
864 or memory reserved is below 4G.
866 [KNL, ARM64] range in low memory.
867 This one lets the user specify a low range in the
868 DMA zone for the crash dump kernel.
869 It will be ignored when crashkernel=X,high is not used
870 or memory reserved is located in the DMA zones.
873 [KNL] Disable crypto self-tests
878 cs89x0_media= [HW,NET]
879 Format: { rj45 | aui | bnc }
881 csdlock_debug= [KNL] Enable debug add-ons of cross-CPU function call
882 handling. When switched on, additional debug data is
883 printed to the console in case a hanging CPU is
884 detected, and that CPU is pinged again in order to try
885 to resolve the hang situation.
886 0: disable csdlock debugging (default)
887 1: enable basic csdlock debugging (minor impact)
888 ext: enable extended csdlock debugging (more impact,
892 See header of drivers/s390/block/dasd_devmap.c.
894 db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port
895 (one device per port)
896 Format: <port#>,<type>
897 See also Documentation/input/devices/joystick-parport.rst
899 debug [KNL] Enable kernel debugging (events log level).
902 [KNL] Enable printing [hashed] pointers early in the
903 boot sequence. If enabled, we use a weak hash instead
904 of siphash to hash pointers. Use this option if you are
905 seeing instances of '(___ptrval___)') and need to see a
906 value (hashed pointer) instead. Cryptographically
907 insecure, please do not use on production kernels.
910 [KNL] verbose locking self-tests
912 Print debugging info while doing the locking API
914 Bitmask for the various LOCKTYPE_ tests. Defaults to 0
915 (no extra messages), setting it to -1 (all bits set)
916 will print _a_lot_ more information - normally only
917 useful to lockdep developers.
919 debug_objects [KNL] Enable object debugging
922 [KNL] Disable object debugging
924 debug_guardpage_minorder=
925 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this
926 parameter allows control of the order of pages that will
927 be intentionally kept free (and hence protected) by the
928 buddy allocator. Bigger value increase the probability
929 of catching random memory corruption, but reduce the
930 amount of memory for normal system use. The maximum
931 possible value is MAX_ORDER/2. Setting this parameter
932 to 1 or 2 should be enough to identify most random
933 memory corruption problems caused by bugs in kernel or
934 driver code when a CPU writes to (or reads from) a
935 random memory location. Note that there exists a class
936 of memory corruptions problems caused by buggy H/W or
937 F/W or by drivers badly programing DMA (basically when
938 memory is written at bus level and the CPU MMU is
939 bypassed) which are not detectable by
940 CONFIG_DEBUG_PAGEALLOC, hence this option will not help
941 tracking down these problems.
944 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this parameter
945 enables the feature at boot time. By default, it is
946 disabled and the system will work mostly the same as a
947 kernel built without CONFIG_DEBUG_PAGEALLOC.
948 Note: to get most of debug_pagealloc error reports, it's
949 useful to also enable the page_owner functionality.
950 on: enable the feature
952 debugfs= [KNL] This parameter enables what is exposed to userspace
953 and debugfs internal clients.
954 Format: { on, no-mount, off }
955 on: All functions are enabled.
957 Filesystem is not registered but kernel clients can
958 access APIs and a crashkernel can be used to read
959 its content. There is nothing to mount.
960 off: Filesystem is not registered and clients
961 get a -EPERM as result when trying to register files
962 or directories within debugfs.
963 This is equivalent of the runtime functionality if
964 debugfs was not enabled in the kernel at all.
965 Default value is set in build-time with a kernel configuration.
967 debugpat [X86] Enable PAT debugging
969 decnet.addr= [HW,NET]
970 Format: <area>[,<node>]
971 See also Documentation/networking/decnet.rst.
974 [HW] The size of the default HugeTLB page. This is
975 the size represented by the legacy /proc/ hugepages
976 APIs. In addition, this is the default hugetlb size
977 used for shmget(), mmap() and mounting hugetlbfs
978 filesystems. If not specified, defaults to the
979 architecture's default huge page size. Huge page
980 sizes are architecture dependent. See also
981 Documentation/admin-guide/mm/hugetlbpage.rst.
984 deferred_probe_timeout=
985 [KNL] Debugging option to set a timeout in seconds for
986 deferred probe to give up waiting on dependencies to
987 probe. Only specific dependencies (subsystems or
988 drivers) that have opted in will be ignored. A timeout
989 of 0 will timeout at the end of initcalls. If the time
990 out hasn't expired, it'll be restarted by each
991 successful driver registration. This option will also
992 dump out devices still on the deferred probe list after
995 delayacct [KNL] Enable per-task delay accounting
997 dell_smm_hwmon.ignore_dmi=
998 [HW] Continue probing hardware even if DMI data
999 indicates that the driver is running on unsupported
1002 dell_smm_hwmon.force=
1003 [HW] Activate driver even if SMM BIOS signature does
1004 not match list of supported models and enable otherwise
1005 blacklisted features.
1007 dell_smm_hwmon.power_status=
1008 [HW] Report power status in /proc/i8k
1009 (disabled by default).
1011 dell_smm_hwmon.restricted=
1012 [HW] Allow controlling fans only if SYS_ADMIN
1015 dell_smm_hwmon.fan_mult=
1016 [HW] Factor to multiply fan speed with.
1018 dell_smm_hwmon.fan_max=
1019 [HW] Maximum configurable fan speed.
1022 Format: { on | off | def_only | inf_only | always }
1023 on: s390 zlib hardware support for compression on
1024 level 1 and decompression (default)
1025 off: No s390 zlib hardware support
1026 def_only: s390 zlib hardware support for deflate
1027 only (compression on level 1)
1028 inf_only: s390 zlib hardware support for inflate
1029 only (decompression)
1030 always: Same as 'on' but ignores the selected compression
1031 level always using hardware support (used for debugging)
1033 dhash_entries= [KNL]
1034 Set number of hash buckets for dentry cache.
1036 disable_1tb_segments [PPC]
1037 Disables the use of 1TB hash page table segments. This
1038 causes the kernel to fall back to 256MB segments which
1039 can be useful when debugging issues that require an SLB
1043 Limits the number of kernel SLB entries, and flushes
1044 them frequently to increase the rate of SLB faults
1045 on kernel addresses.
1048 See Documentation/networking/ipv6.rst.
1051 Disable RADIX MMU mode on POWER9
1053 radix_hcall_invalidate=on [PPC/PSERIES]
1054 Disable RADIX GTSE feature and use hcall for TLB
1058 Disable TLBIE instruction. Currently does not work
1059 with KVM, with HASH MMU, or with coherent accelerators.
1061 disable_cpu_apicid= [X86,APIC,SMP]
1063 The number of initial APIC ID for the
1064 corresponding CPU to be disabled at boot,
1065 mostly used for the kdump 2nd kernel to
1066 disable BSP to wake up multiple CPUs without
1067 causing system reset or hang due to sending
1068 INIT from AP to BSP.
1070 disable_ddw [PPC/PSERIES]
1071 Disable Dynamic DMA Window support. Use this
1072 to workaround buggy firmware.
1074 disable_ipv6= [IPV6]
1075 See Documentation/networking/ipv6.rst.
1077 disable_mtrr_cleanup [X86]
1078 The kernel tries to adjust MTRR layout from continuous
1079 to discrete, to make X server driver able to add WB
1080 entry later. This parameter disables that.
1082 disable_mtrr_trim [X86, Intel and AMD only]
1083 By default the kernel will trim any uncacheable
1084 memory out of your available memory pool based on
1085 MTRR settings. This parameter disables that behavior,
1086 possibly causing your machine to run very slowly.
1088 disable_timer_pin_1 [X86]
1089 Disable PIN 1 of APIC timer
1090 Can be useful to work around chipset bugs.
1092 dis_ucode_ldr [X86] Disable the microcode loader.
1094 dma_debug=off If the kernel is compiled with DMA_API_DEBUG support,
1095 this option disables the debugging code at boot.
1097 dma_debug_entries=<number>
1098 This option allows to tune the number of preallocated
1099 entries for DMA-API debugging code. One entry is
1100 required per DMA-API allocation. Use this if the
1101 DMA-API debugging code disables itself because the
1102 architectural default is too low.
1104 dma_debug_driver=<driver_name>
1105 With this option the DMA-API debugging driver
1106 filter feature can be enabled at boot time. Just
1107 pass the driver to filter for as the parameter.
1108 The filter can be disabled or changed to another
1109 driver later using sysfs.
1111 driver_async_probe= [KNL]
1112 List of driver names to be probed asynchronously. *
1113 matches with all driver names. If * is specified, the
1114 rest of the listed driver names are those that will NOT
1116 Format: <driver_name1>,<driver_name2>...
1118 drm.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>]
1119 Broken monitors, graphic adapters, KVMs and EDIDless
1120 panels may send no or incorrect EDID data sets.
1121 This parameter allows to specify an EDID data sets
1122 in the /lib/firmware directory that are used instead.
1123 Generic built-in EDID data sets are used, if one of
1124 edid/1024x768.bin, edid/1280x1024.bin,
1125 edid/1680x1050.bin, or edid/1920x1080.bin is given
1126 and no file with the same name exists. Details and
1127 instructions how to build your own EDID data are
1128 available in Documentation/admin-guide/edid.rst. An EDID
1129 data set will only be used for a particular connector,
1130 if its name and a colon are prepended to the EDID
1131 name. Each connector may use a unique EDID data
1132 set by separating the files with a comma. An EDID
1133 data set with no connector name will be used for
1134 any connectors not explicitly specified.
1139 Format: {"off" | "known"}
1140 Control how the dt_cpu_ftrs device-tree binding is
1141 used for CPU feature discovery and setup (if it
1143 off: Do not use it, fall back to legacy cpu table.
1144 known: Do not pass through unknown features to guests
1145 or userspace, only those that the kernel is aware of.
1147 dump_apple_properties [X86]
1148 Dump name and content of EFI device properties on
1149 x86 Macs. Useful for driver authors to determine
1150 what data is available or for reverse-engineering.
1152 dyndbg[="val"] [KNL,DYNAMIC_DEBUG]
1153 <module>.dyndbg[="val"]
1154 Enable debug messages at boot time. See
1155 Documentation/admin-guide/dynamic-debug-howto.rst
1158 nopku [X86] Disable Memory Protection Keys CPU feature found
1161 <module>.async_probe [KNL]
1162 Enable asynchronous probe on this module.
1164 early_ioremap_debug [KNL]
1165 Enable debug messages in early_ioremap support. This
1166 is useful for tracking down temporary early mappings
1167 which are not unmapped.
1169 earlycon= [KNL] Output early console device and options.
1171 When used with no options, the early console is
1172 determined by stdout-path property in device tree's
1173 chosen node or the ACPI SPCR table if supported by
1176 cdns,<addr>[,options]
1177 Start an early, polled-mode console on a Cadence
1178 (xuartps) serial port at the specified address. Only
1179 supported option is baud rate. If baud rate is not
1180 specified, the serial port must already be setup and
1183 uart[8250],io,<addr>[,options]
1184 uart[8250],mmio,<addr>[,options]
1185 uart[8250],mmio32,<addr>[,options]
1186 uart[8250],mmio32be,<addr>[,options]
1187 uart[8250],0x<addr>[,options]
1188 Start an early, polled-mode console on the 8250/16550
1189 UART at the specified I/O port or MMIO address.
1190 MMIO inter-register address stride is either 8-bit
1191 (mmio) or 32-bit (mmio32 or mmio32be).
1192 If none of [io|mmio|mmio32|mmio32be], <addr> is assumed
1193 to be equivalent to 'mmio'. 'options' are specified
1194 in the same format described for "console=ttyS<n>"; if
1195 unspecified, the h/w is not initialized.
1199 Start an early, polled-mode console on a pl011 serial
1200 port at the specified address. The pl011 serial port
1201 must already be setup and configured. Options are not
1202 yet supported. If 'mmio32' is specified, then only
1203 the driver will use only 32-bit accessors to read/write
1204 the device registers.
1207 Start an early console on a litex serial port at the
1208 specified address. The serial port must already be
1209 setup and configured. Options are not yet supported.
1212 Start an early, polled-mode console on a meson serial
1213 port at the specified address. The serial port must
1214 already be setup and configured. Options are not yet
1218 Start an early, polled-mode console on an msm serial
1219 port at the specified address. The serial port
1220 must already be setup and configured. Options are not
1223 msm_serial_dm,<addr>
1224 Start an early, polled-mode console on an msm serial
1225 dm port at the specified address. The serial port
1226 must already be setup and configured. Options are not
1230 Start an early, polled-mode console on a serial port
1231 of an Actions Semi SoC, such as S500 or S900, at the
1232 specified address. The serial port must already be
1233 setup and configured. Options are not yet supported.
1236 Start an early, polled-mode console on a serial port
1237 of an RDA Micro SoC, such as RDA8810PL, at the
1238 specified address. The serial port must already be
1239 setup and configured. Options are not yet supported.
1242 Use RISC-V SBI (Supervisor Binary Interface) for early
1245 smh Use ARM semihosting calls for early console.
1253 Use early console provided by serial driver available
1254 on Samsung SoCs, requires selecting proper type and
1255 a correct base address of the selected UART port. The
1256 serial port must already be setup and configured.
1257 Options are not yet supported.
1260 Start an early, polled-mode console on a lantiq serial
1261 (lqasc) port at the specified address. The serial port
1262 must already be setup and configured. Options are not
1267 Use early console provided by Freescale LP UART driver
1268 found on Freescale Vybrid and QorIQ LS1021A processors.
1269 A valid base address must be provided, and the serial
1270 port must already be setup and configured.
1274 Start an early, polled-mode, output-only console on the
1275 Freescale i.MX UART at the specified address. The UART
1276 must already be setup and configured.
1279 Start an early, polled-mode console on the
1280 Armada 3700 serial port at the specified
1281 address. The serial port must already be setup
1282 and configured. Options are not yet supported.
1285 Start an early, polled-mode console on a Qualcomm
1286 Generic Interface (GENI) based serial port at the
1287 specified address. The serial port must already be
1288 setup and configured. Options are not yet supported.
1291 Start an early, unaccelerated console on the EFI
1292 memory mapped framebuffer (if available). On cache
1293 coherent non-x86 systems that use system memory for
1294 the framebuffer, pass the 'ram' option so that it is
1295 mapped with the correct attributes.
1298 Use early console provided by Freescale LINFlexD UART
1299 serial driver for NXP S32V234 SoCs. A valid base
1300 address must be provided, and the serial port must
1301 already be setup and configured.
1303 earlyprintk= [X86,SH,ARM,M68k,S390]
1307 earlyprintk=serial[,ttySn[,baudrate]]
1308 earlyprintk=serial[,0x...[,baudrate]]
1309 earlyprintk=ttySn[,baudrate]
1310 earlyprintk=dbgp[debugController#]
1311 earlyprintk=pciserial[,force],bus:device.function[,baudrate]
1312 earlyprintk=xdbc[xhciController#]
1314 earlyprintk is useful when the kernel crashes before
1315 the normal console is initialized. It is not enabled by
1316 default because it has some cosmetic problems.
1318 Append ",keep" to not disable it when the real console
1321 Only one of vga, serial, or usb debug port can
1324 Currently only ttyS0 and ttyS1 may be specified by
1325 name. Other I/O ports may be explicitly specified
1326 on some architectures (x86 and arm at least) by
1327 replacing ttySn with an I/O port address, like this:
1328 earlyprintk=serial,0x1008,115200
1329 You can find the port for a given device in
1330 /proc/tty/driver/serial:
1331 2: uart:ST16650V2 port:00001008 irq:18 ...
1333 Interaction with the standard serial driver is not
1336 The VGA output is eventually overwritten by
1339 The xen option can only be used in Xen domains.
1341 The sclp output can only be used on s390.
1343 The optional "force" to "pciserial" enables use of a
1344 PCI device even when its classcode is not of the
1347 edac_report= [HW,EDAC] Control how to report EDAC event
1348 Format: {"on" | "off" | "force"}
1349 on: enable EDAC to report H/W event. May be overridden
1350 by other higher priority error reporting module.
1351 off: disable H/W event reporting through EDAC.
1352 force: enforce the use of EDAC to report H/W event.
1356 Format: {"off" | "on" | "skip[mbr]"}
1359 Format: { "debug", "disable_early_pci_dma",
1360 "nochunk", "noruntime", "nosoftreserve",
1361 "novamap", "no_disable_early_pci_dma" }
1362 debug: enable misc debug output.
1363 disable_early_pci_dma: disable the busmaster bit on all
1364 PCI bridges while in the EFI boot stub.
1365 nochunk: disable reading files in "chunks" in the EFI
1366 boot stub, as chunking can cause problems with some
1367 firmware implementations.
1368 noruntime : disable EFI runtime services support
1369 nosoftreserve: The EFI_MEMORY_SP (Specific Purpose)
1370 attribute may cause the kernel to reserve the
1371 memory range for a memory mapping driver to
1372 claim. Specify efi=nosoftreserve to disable this
1373 reservation and treat the memory by its base type
1374 (i.e. EFI_CONVENTIONAL_MEMORY / "System RAM").
1375 novamap: do not call SetVirtualAddressMap().
1376 no_disable_early_pci_dma: Leave the busmaster bit set
1377 on all PCI bridges while in the EFI boot stub
1379 efi_no_storage_paranoia [EFI; X86]
1380 Using this parameter you can use more than 50% of
1381 your efi variable storage. Use this parameter only if
1382 you are really sure that your UEFI does sane gc and
1383 fulfills the spec otherwise your board may brick.
1385 efi_fake_mem= nn[KMG]@ss[KMG]:aa[,nn[KMG]@ss[KMG]:aa,..] [EFI; X86]
1386 Add arbitrary attribute to specific memory range by
1387 updating original EFI memory map.
1388 Region of memory which aa attribute is added to is
1391 If efi_fake_mem=2G@4G:0x10000,2G@0x10a0000000:0x10000
1392 is specified, EFI_MEMORY_MORE_RELIABLE(0x10000)
1393 attribute is added to range 0x100000000-0x180000000 and
1394 0x10a0000000-0x1120000000.
1396 If efi_fake_mem=8G@9G:0x40000 is specified, the
1397 EFI_MEMORY_SP(0x40000) attribute is added to
1398 range 0x240000000-0x43fffffff.
1400 Using this parameter you can do debugging of EFI memmap
1401 related features. For example, you can do debugging of
1402 Address Range Mirroring feature even if your box
1403 doesn't support it, or mark specific memory as
1406 efivar_ssdt= [EFI; X86] Name of an EFI variable that contains an SSDT
1407 that is to be dynamically loaded by Linux. If there are
1408 multiple variables with the same name but with different
1409 vendor GUIDs, all of them will be loaded. See
1410 Documentation/admin-guide/acpi/ssdt-overlays.rst for details.
1413 eisa_irq_edge= [PARISC,HW]
1414 See header of drivers/parisc/eisa.c.
1416 ekgdboc= [X86,KGDB] Allow early kernel console debugging
1419 This is designed to be used in conjunction with
1420 the boot argument: earlyprintk=vga
1422 This parameter works in place of the kgdboc parameter
1423 but can only be used if the backing tty is available
1424 very early in the boot process. For early debugging
1425 via a serial port see kgdboc_earlycon instead.
1428 See comment before function elanfreq_setup() in
1429 arch/x86/kernel/cpu/cpufreq/elanfreq.c.
1431 elfcorehdr=[size[KMG]@]offset[KMG] [IA64,PPC,SH,X86,S390]
1432 Specifies physical address of start of kernel core
1433 image elf header and optionally the size. Generally
1434 kexec loader will pass this option to capture kernel.
1435 See Documentation/admin-guide/kdump/kdump.rst for details.
1437 enable_mtrr_cleanup [X86]
1438 The kernel tries to adjust MTRR layout from continuous
1439 to discrete, to make X server driver able to add WB
1440 entry later. This parameter enables that.
1442 enable_timer_pin_1 [X86]
1443 Enable PIN 1 of APIC timer
1444 Can be useful to work around chipset bugs
1445 (in particular on some ATI chipsets).
1446 The kernel tries to set a reasonable default.
1448 enforcing= [SELINUX] Set initial enforcing status.
1450 See security/selinux/Kconfig help text.
1451 0 -- permissive (log only, no denials).
1452 1 -- enforcing (deny and log).
1454 Value can be changed at runtime via
1455 /sys/fs/selinux/enforce.
1458 Disable Error Record Serialization Table (ERST)
1461 ether= [HW,NET] Ethernet cards parameters
1462 This option is obsoleted by the "netdev=" option, which
1463 has equivalent usage. See its documentation for details.
1467 Permit 'security.evm' to be updated regardless of
1468 current integrity status.
1473 fail_make_request=[KNL]
1474 General fault injection mechanism.
1475 Format: <interval>,<probability>,<space>,<times>
1476 See also Documentation/fault-injection/.
1479 Format: { initns | none }
1480 See Documentation/admin-guide/sysctl/net.rst for
1481 fb_tunnels_only_for_init_ns
1484 See Documentation/admin-guide/blockdev/floppy.rst.
1486 force_pal_cache_flush
1487 [IA-64] Avoid check_sal_cache_flush which may hang on
1488 buggy SAL_CACHE_FLUSH implementations. Using this
1489 parameter will force ia64_sal_cache_flush to call
1490 ia64_pal_cache_flush instead of SAL_CACHE_FLUSH.
1493 Forcefully enable Physical Address Extension (PAE).
1494 Many Pentium M systems disable PAE but may have a
1495 functionally usable PAE implementation.
1496 Warning: use of this parameter will taint the kernel
1497 and may cause unknown problems.
1500 [FTRACE] will set and start the specified tracer
1501 as early as possible in order to facilitate early
1504 ftrace_boot_snapshot
1505 [FTRACE] On boot up, a snapshot will be taken of the
1506 ftrace ring buffer that can be read at:
1507 /sys/kernel/tracing/snapshot.
1508 This is useful if you need tracing information from kernel
1509 boot up that is likely to be overridden by user space
1510 start up functionality.
1512 ftrace_dump_on_oops[=orig_cpu]
1513 [FTRACE] will dump the trace buffers on oops.
1514 If no parameter is passed, ftrace will dump
1515 buffers of all CPUs, but if you pass orig_cpu, it will
1516 dump only the buffer of the CPU that triggered the
1519 ftrace_filter=[function-list]
1520 [FTRACE] Limit the functions traced by the function
1521 tracer at boot up. function-list is a comma-separated
1522 list of functions. This list can be changed at run
1523 time by the set_ftrace_filter file in the debugfs
1526 ftrace_notrace=[function-list]
1527 [FTRACE] Do not trace the functions specified in
1528 function-list. This list can be changed at run time
1529 by the set_ftrace_notrace file in the debugfs
1532 ftrace_graph_filter=[function-list]
1533 [FTRACE] Limit the top level callers functions traced
1534 by the function graph tracer at boot up.
1535 function-list is a comma-separated list of functions
1536 that can be changed at run time by the
1537 set_graph_function file in the debugfs tracing directory.
1539 ftrace_graph_notrace=[function-list]
1540 [FTRACE] Do not trace from the functions specified in
1541 function-list. This list is a comma-separated list of
1542 functions that can be changed at run time by the
1543 set_graph_notrace file in the debugfs tracing directory.
1545 ftrace_graph_max_depth=<uint>
1546 [FTRACE] Used with the function graph tracer. This is
1547 the max depth it will trace into a function. This value
1548 can be changed at run time by the max_graph_depth file
1549 in the tracefs tracing directory. default: 0 (no limit)
1551 fw_devlink= [KNL] Create device links between consumer and supplier
1552 devices by scanning the firmware to infer the
1553 consumer/supplier relationships. This feature is
1554 especially useful when drivers are loaded as modules as
1555 it ensures proper ordering of tasks like device probing
1556 (suppliers first, then consumers), supplier boot state
1557 clean up (only after all consumers have probed),
1558 suspend/resume & runtime PM (consumers first, then
1560 Format: { off | permissive | on | rpm }
1561 off -- Don't create device links from firmware info.
1562 permissive -- Create device links from firmware info
1563 but use it only for ordering boot state clean
1564 up (sync_state() calls).
1565 on -- Create device links from firmware info and use it
1566 to enforce probe and suspend/resume ordering.
1567 rpm -- Like "on", but also use to order runtime PM.
1569 fw_devlink.strict=<bool>
1570 [KNL] Treat all inferred dependencies as mandatory
1571 dependencies. This only applies for fw_devlink=on|rpm.
1575 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad
1576 support via parallel port (up to 5 devices per port)
1577 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5>
1578 See also Documentation/input/devices/joystick-parport.rst
1582 gart_fix_e820= [X86-64] disable the fix e820 for K8 GART
1586 gcov_persist= [GCOV] When non-zero (default), profiling data for
1587 kernel modules is saved and remains accessible via
1588 debugfs, even when the module is unloaded/reloaded.
1589 When zero, profiling data is discarded and associated
1590 debugfs files are removed at module unload time.
1592 goldfish [X86] Enable the goldfish android emulator platform.
1593 Don't use this when you are not running on the
1596 gpio-mockup.gpio_mockup_ranges
1597 [HW] Sets the ranges of gpiochip of for this device.
1598 Format: <start1>,<end1>,<start2>,<end2>...
1599 gpio-mockup.gpio_mockup_named_lines
1600 [HW] Let the driver know GPIO lines should be named.
1602 gpt [EFI] Forces disk with valid GPT signature but
1603 invalid Protective MBR to be treated as GPT. If the
1604 primary GPT is corrupted, it enables the backup/alternate
1605 GPT to be used instead.
1607 grcan.enable0= [HW] Configuration of physical interface 0. Determines
1608 the "Enable 0" bit of the configuration register.
1611 grcan.enable1= [HW] Configuration of physical interface 1. Determines
1612 the "Enable 0" bit of the configuration register.
1615 grcan.select= [HW] Select which physical interface to use.
1618 grcan.txsize= [HW] Sets the size of the tx buffer.
1619 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0.
1621 grcan.rxsize= [HW] Sets the size of the rx buffer.
1622 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0.
1626 [KNL] Under CONFIG_HARDENED_USERCOPY, whether
1627 hardening is enabled for this boot. Hardened
1628 usercopy checking is used to protect the kernel
1629 from reading or writing beyond known memory
1630 allocation boundaries as a proactive defense
1631 against bounds-checking flaws in the kernel's
1632 copy_to_user()/copy_from_user() interface.
1633 on Perform hardened usercopy checks (default).
1634 off Disable hardened usercopy checks.
1636 hardlockup_all_cpu_backtrace=
1637 [KNL] Should the hard-lockup detector generate
1638 backtraces on all cpus.
1641 hashdist= [KNL,NUMA] Large hashes allocated during boot
1642 are distributed across NUMA nodes. Defaults on
1643 for 64-bit NUMA, off otherwise.
1644 Format: 0 | 1 (for off | on)
1646 hcl= [IA-64] SGI's Hardware Graph compatibility layer
1648 hd= [EIDE] (E)IDE hard drive subsystem geometry
1649 Format: <cyl>,<head>,<sect>
1652 Disable Hardware Error Source Table (HEST) support;
1653 corresponding firmware-first mode error processing
1654 logic will be disabled.
1656 hibernate= [HIBERNATION]
1657 noresume Don't check if there's a hibernation image
1658 present during boot.
1659 nocompress Don't compress/decompress hibernation images.
1660 no Disable hibernation and resume.
1661 protect_image Turn on image protection during restoration
1662 (that will set all pages holding image data
1663 during restoration read-only).
1665 highmem=nn[KMG] [KNL,BOOT] forces the highmem zone to have an exact
1666 size of <nn>. This works even on boxes that have no
1667 highmem otherwise. This also works to reduce highmem
1668 size on bigger boxes.
1670 highres= [KNL] Enable/disable high resolution timer mode.
1671 Valid parameters: "on", "off"
1676 hostname= [KNL] Set the hostname (aka UTS nodename).
1678 This allows setting the system's hostname during early
1679 startup. This sets the name returned by gethostname.
1680 Using this parameter to set the hostname makes it
1681 possible to ensure the hostname is correctly set before
1682 any userspace processes run, avoiding the possibility
1683 that a process may call gethostname before the hostname
1684 has been explicitly set, resulting in the calling
1685 process getting an incorrect result. The string must
1686 not exceed the maximum allowed hostname length (usually
1687 64 characters) and will be truncated otherwise.
1689 hpet= [X86-32,HPET] option to control HPET usage
1690 Format: { enable (default) | disable | force |
1692 disable: disable HPET and use PIT instead
1693 force: allow force enabled of undocumented chips (ICH4,
1695 verbose: show contents of HPET registers during setup
1697 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET
1698 registers. Default set by CONFIG_HPET_MMAP_DEFAULT.
1700 hugepages= [HW] Number of HugeTLB pages to allocate at boot.
1701 If this follows hugepagesz (below), it specifies
1702 the number of pages of hugepagesz to be allocated.
1703 If this is the first HugeTLB parameter on the command
1704 line, it specifies the number of pages to allocate for
1705 the default huge page size. If using node format, the
1706 number of pages to allocate per-node can be specified.
1707 See also Documentation/admin-guide/mm/hugetlbpage.rst.
1708 Format: <integer> or (node format)
1709 <node>:<integer>[,<node>:<integer>]
1712 [HW] The size of the HugeTLB pages. This is used in
1713 conjunction with hugepages (above) to allocate huge
1714 pages of a specific size at boot. The pair
1715 hugepagesz=X hugepages=Y can be specified once for
1716 each supported huge page size. Huge page sizes are
1717 architecture dependent. See also
1718 Documentation/admin-guide/mm/hugetlbpage.rst.
1721 hugetlb_cma= [HW,CMA] The size of a CMA area used for allocation
1722 of gigantic hugepages. Or using node format, the size
1723 of a CMA area per node can be specified.
1724 Format: nn[KMGTPE] or (node format)
1725 <node>:nn[KMGTPE][,<node>:nn[KMGTPE]]
1727 Reserve a CMA area of given size and allocate gigantic
1728 hugepages using the CMA allocator. If enabled, the
1729 boot-time allocation of gigantic hugepages is skipped.
1731 hugetlb_free_vmemmap=
1732 [KNL] Reguires CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP
1734 Allows heavy hugetlb users to free up some more
1735 memory (7 * PAGE_SIZE for each 2MB hugetlb page).
1736 Format: { [oO][Nn]/Y/y/1 | [oO][Ff]/N/n/0 (default) }
1738 [oO][Nn]/Y/y/1: enable the feature
1739 [oO][Ff]/N/n/0: disable the feature
1741 Built with CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP_DEFAULT_ON=y,
1744 Note that the vmemmap pages may be allocated from the added
1745 memory block itself when memory_hotplug.memmap_on_memory is
1746 enabled, those vmemmap pages cannot be optimized even if this
1747 feature is enabled. Other vmemmap pages not allocated from
1748 the added memory block itself do not be affected.
1751 [KNL] Should the hung task detector generate panics.
1754 A value of 1 instructs the kernel to panic when a
1755 hung task is detected. The default value is controlled
1756 by the CONFIG_BOOTPARAM_HUNG_TASK_PANIC build-time
1757 option. The value selected by this boot parameter can
1758 be changed later by the kernel.hung_task_panic sysctl.
1760 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC)
1761 terminal devices. Valid values: 0..8
1762 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs.
1763 If specified, z/VM IUCV HVC accepts connections
1764 from listed z/VM user IDs only.
1766 hv_nopvspin [X86,HYPER_V] Disables the paravirt spinlock optimizations
1767 which allow the hypervisor to 'idle' the
1768 guest on lock contention.
1771 Do not unregister boot console at start. This is only
1772 useful for debugging when something happens in the window
1773 between unregistering the boot console and initializing
1776 i2c_bus= [HW] Override the default board specific I2C bus speed
1777 or register an additional I2C bus that is not
1778 registered from board initialization code.
1782 i8042.debug [HW] Toggle i8042 debug mode
1783 i8042.unmask_kbd_data
1784 [HW] Enable printing of interrupt data from the KBD port
1785 (disabled by default, and as a pre-condition
1786 requires that i8042.debug=1 be enabled)
1787 i8042.direct [HW] Put keyboard port into non-translated mode
1788 i8042.dumbkbd [HW] Pretend that controller can only read data from
1789 keyboard and cannot control its state
1790 (Don't attempt to blink the leds)
1791 i8042.noaux [HW] Don't check for auxiliary (== mouse) port
1792 i8042.nokbd [HW] Don't check/create keyboard port
1793 i8042.noloop [HW] Disable the AUX Loopback command while probing
1795 i8042.nomux [HW] Don't check presence of an active multiplexing
1797 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX
1799 i8042.notimeout [HW] Ignore timeout condition signalled by controller
1800 i8042.reset [HW] Reset the controller during init, cleanup and
1801 suspend-to-ram transitions, only during s2r
1802 transitions, or never reset
1803 Format: { 1 | Y | y | 0 | N | n }
1804 1, Y, y: always reset controller
1805 0, N, n: don't ever reset controller
1806 Default: only on s2r transitions on x86; most other
1807 architectures force reset to be always executed
1808 i8042.unlock [HW] Unlock (ignore) the keylock
1809 i8042.kbdreset [HW] Reset device connected to KBD port
1811 [HW] Allow deferred probing upon i8042 probe errors
1815 i915.invert_brightness=
1816 [DRM] Invert the sense of the variable that is used to
1817 set the brightness of the panel backlight. Normally a
1818 brightness value of 0 indicates backlight switched off,
1819 and the maximum of the brightness value sets the backlight
1820 to maximum brightness. If this parameter is set to 0
1821 (default) and the machine requires it, or this parameter
1822 is set to 1, a brightness value of 0 sets the backlight
1823 to maximum brightness, and the maximum of the brightness
1824 value switches the backlight off.
1825 -1 -- never invert brightness
1826 0 -- machine default
1827 1 -- force brightness inversion
1830 Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]]
1834 Format: idle=poll, idle=halt, idle=nomwait
1835 Poll forces a polling idle loop that can slightly
1836 improve the performance of waking up a idle CPU, but
1837 will use a lot of power and make the system run hot.
1839 idle=halt: Halt is forced to be used for CPU idle.
1840 In such case C2/C3 won't be used again.
1841 idle=nomwait: Disable mwait for CPU C-states
1845 Allow force disabling of Shared Virtual Memory (SVA)
1846 support for the idxd driver. By default it is set to
1849 idxd.tc_override= [HW]
1851 Allow override of default traffic class configuration
1852 for the device. By default it is set to false (0).
1854 ieee754= [MIPS] Select IEEE Std 754 conformance mode
1855 Format: { strict | legacy | 2008 | relaxed }
1858 Choose which programs will be accepted for execution
1859 based on the IEEE 754 NaN encoding(s) supported by
1860 the FPU and the NaN encoding requested with the value
1861 of an ELF file header flag individually set by each
1862 binary. Hardware implementations are permitted to
1863 support either or both of the legacy and the 2008 NaN
1866 Available settings are as follows:
1867 strict accept binaries that request a NaN encoding
1868 supported by the FPU
1869 legacy only accept legacy-NaN binaries, if supported
1871 2008 only accept 2008-NaN binaries, if supported
1873 relaxed accept any binaries regardless of whether
1874 supported by the FPU
1876 The FPU emulator is always able to support both NaN
1877 encodings, so if no FPU hardware is present or it has
1878 been disabled with 'nofpu', then the settings of
1879 'legacy' and '2008' strap the emulator accordingly,
1880 'relaxed' straps the emulator for both legacy-NaN and
1881 2008-NaN, whereas 'strict' enables legacy-NaN only on
1882 legacy processors and both NaN encodings on MIPS32 or
1885 The setting for ABS.fmt/NEG.fmt instruction execution
1886 mode generally follows that for the NaN encoding,
1887 except where unsupported by hardware.
1889 ignore_loglevel [KNL]
1890 Ignore loglevel setting - this will print /all/
1891 kernel messages to the console. Useful for debugging.
1892 We also add it as printk module parameter, so users
1893 could change it dynamically, usually by
1894 /sys/module/printk/parameters/ignore_loglevel.
1897 Ignore RLIMIT_DATA setting for data mappings,
1898 print warning at first misuse. Can be changed via
1899 /sys/module/kernel/parameters/ignore_rlimit_data.
1901 ihash_entries= [KNL]
1902 Set number of hash buckets for inode cache.
1904 ima_appraise= [IMA] appraise integrity measurements
1905 Format: { "off" | "enforce" | "fix" | "log" }
1908 ima_appraise_tcb [IMA] Deprecated. Use ima_policy= instead.
1909 The builtin appraise policy appraises all files
1912 ima_canonical_fmt [IMA]
1913 Use the canonical format for the binary runtime
1914 measurements, instead of host native format.
1917 Format: { md5 | sha1 | rmd160 | sha256 | sha384
1921 The list of supported hash algorithms is defined
1922 in crypto/hash_info.h.
1925 The builtin policies to load during IMA setup.
1926 Format: "tcb | appraise_tcb | secure_boot |
1927 fail_securely | critical_data"
1929 The "tcb" policy measures all programs exec'd, files
1930 mmap'd for exec, and all files opened with the read
1931 mode bit set by either the effective uid (euid=0) or
1934 The "appraise_tcb" policy appraises the integrity of
1935 all files owned by root.
1937 The "secure_boot" policy appraises the integrity
1938 of files (eg. kexec kernel image, kernel modules,
1939 firmware, policy, etc) based on file signatures.
1941 The "fail_securely" policy forces file signature
1942 verification failure also on privileged mounted
1943 filesystems with the SB_I_UNVERIFIABLE_SIGNATURE
1946 The "critical_data" policy measures kernel integrity
1949 ima_tcb [IMA] Deprecated. Use ima_policy= instead.
1950 Load a policy which meets the needs of the Trusted
1951 Computing Base. This means IMA will measure all
1952 programs exec'd, files mmap'd for exec, and all files
1953 opened for read by uid=0.
1956 Select one of defined IMA measurements template formats.
1957 Formats: { "ima" | "ima-ng" | "ima-ngv2" | "ima-sig" |
1962 [IMA] Define a custom template format.
1963 Format: { "field1|...|fieldN" }
1965 ima.ahash_minsize= [IMA] Minimum file size for asynchronous hash usage
1966 Format: <min_file_size>
1967 Set the minimal file size for using asynchronous hash.
1968 If left unspecified, ahash usage is disabled.
1970 ahash performance varies for different data sizes on
1971 different crypto accelerators. This option can be used
1972 to achieve the best performance for a particular HW.
1974 ima.ahash_bufsize= [IMA] Asynchronous hash buffer size
1976 Set hashing buffer size. Default: 4k.
1978 ahash performance varies for different chunk sizes on
1979 different crypto accelerators. This option can be used
1980 to achieve best performance for particular HW.
1984 Run specified binary instead of /sbin/init as init
1987 initcall_debug [KNL] Trace initcalls as they are executed. Useful
1988 for working out where the kernel is dying during
1991 initcall_blacklist= [KNL] Do not execute a comma-separated list of
1992 initcall functions. Useful for debugging built-in
1993 modules and initcalls.
1995 initramfs_async= [KNL]
1998 This parameter controls whether the initramfs
1999 image is unpacked asynchronously, concurrently
2000 with devices being probed and
2001 initialized. This should normally just work,
2002 but as a debugging aid, one can get the
2003 historical behaviour of the initramfs
2004 unpacking being completed before device_ and
2007 initrd= [BOOT] Specify the location of the initial ramdisk
2009 initrdmem= [KNL] Specify a physical address and size from which to
2010 load the initrd. If an initrd is compiled in or
2011 specified in the bootparams, it takes priority over this
2013 Format: ss[KMG],nn[KMG]
2016 init_on_alloc= [MM] Fill newly allocated pages and heap objects with
2019 Default set by CONFIG_INIT_ON_ALLOC_DEFAULT_ON.
2021 init_on_free= [MM] Fill freed pages and heap objects with zeroes.
2023 Default set by CONFIG_INIT_ON_FREE_DEFAULT_ON.
2025 init_pkru= [X86] Specify the default memory protection keys rights
2026 register contents for all processes. 0x55555554 by
2027 default (disallow access to all but pkey 0). Can
2028 override in debugfs after boot.
2030 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver
2033 int_pln_enable [X86] Enable power limit notification interrupt
2035 integrity_audit=[IMA]
2036 Format: { "0" | "1" }
2037 0 -- basic integrity auditing messages. (Default)
2038 1 -- additional integrity auditing messages.
2040 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option
2042 Enable intel iommu driver.
2044 Disable intel iommu driver.
2045 igfx_off [Default Off]
2046 By default, gfx is mapped as normal device. If a gfx
2047 device has a dedicated DMAR unit, the DMAR unit is
2048 bypassed by not enabling DMAR with this option. In
2049 this case, gfx device will use physical address for
2051 strict [Default Off]
2052 Deprecated, equivalent to iommu.strict=1.
2053 sp_off [Default Off]
2054 By default, super page will be supported if Intel IOMMU
2055 has the capability. With this option, super page will
2058 Enable the Intel IOMMU scalable mode if the hardware
2059 advertises that it has support for the scalable mode
2062 Disallow use of the Intel IOMMU scalable mode.
2063 tboot_noforce [Default Off]
2064 Do not force the Intel IOMMU enabled under tboot.
2065 By default, tboot will force Intel IOMMU on, which
2066 could harm performance of some high-throughput
2067 devices like 40GBit network cards, even if identity
2069 Note that using this option lowers the security
2070 provided by tboot because it makes the system
2071 vulnerable to DMA attacks.
2073 intel_idle.max_cstate= [KNL,HW,ACPI,X86]
2074 0 disables intel_idle and fall back on acpi_idle.
2075 1 to 9 specify maximum depth of C-state.
2079 Do not enable intel_pstate as the default
2080 scaling driver for the supported processors
2082 Use intel_pstate as a scaling driver, but configure it
2083 to work with generic cpufreq governors (instead of
2084 enabling its internal governor). This mode cannot be
2085 used along with the hardware-managed P-states (HWP)
2088 Enable intel_pstate on systems that prohibit it by default
2089 in favor of acpi-cpufreq. Forcing the intel_pstate driver
2090 instead of acpi-cpufreq may disable platform features, such
2091 as thermal controls and power capping, that rely on ACPI
2092 P-States information being indicated to OSPM and therefore
2093 should be used with caution. This option does not work with
2094 processors that aren't supported by the intel_pstate driver
2095 or on platforms that use pcc-cpufreq instead of acpi-cpufreq.
2097 Do not enable hardware P state control (HWP)
2100 Only load intel_pstate on systems which support
2101 hardware P state control (HWP) if available.
2103 Enforce ACPI _PPC performance limits. If the Fixed ACPI
2104 Description Table, specifies preferred power management
2105 profile as "Enterprise Server" or "Performance Server",
2106 then this feature is turned on by default.
2108 Allow per-logical-CPU P-State performance control limits using
2109 cpufreq sysfs interface
2111 intremap= [X86-64, Intel-IOMMU]
2112 on enable Interrupt Remapping (default)
2113 off disable Interrupt Remapping
2114 nosid disable Source ID checking
2116 BIOS x2APIC opt-out request will be ignored
2117 nopost disable Interrupt Posting
2119 iomem= Disable strict checking of access to MMIO memory
2120 strict regions from userspace.
2135 nobypass [PPC/POWERNV]
2136 Disable IOMMU bypass, using IOMMU for PCI devices.
2138 iommu.forcedac= [ARM64, X86] Control IOVA allocation for PCI devices.
2139 Format: { "0" | "1" }
2140 0 - Try to allocate a 32-bit DMA address first, before
2141 falling back to the full range if needed.
2142 1 - Allocate directly from the full usable range,
2143 forcing Dual Address Cycle for PCI cards supporting
2144 greater than 32-bit addressing.
2146 iommu.strict= [ARM64, X86] Configure TLB invalidation behaviour
2147 Format: { "0" | "1" }
2149 Request that DMA unmap operations use deferred
2150 invalidation of hardware TLBs, for increased
2151 throughput at the cost of reduced device isolation.
2152 Will fall back to strict mode if not supported by
2153 the relevant IOMMU driver.
2155 DMA unmap operations invalidate IOMMU hardware TLBs
2157 unset - Use value of CONFIG_IOMMU_DEFAULT_DMA_{LAZY,STRICT}.
2158 Note: on x86, strict mode specified via one of the
2159 legacy driver-specific options takes precedence.
2162 [ARM64, X86] Configure DMA to bypass the IOMMU by default.
2163 Format: { "0" | "1" }
2164 0 - Use IOMMU translation for DMA.
2165 1 - Bypass the IOMMU for DMA.
2166 unset - Use value of CONFIG_IOMMU_DEFAULT_PASSTHROUGH.
2168 io7= [HW] IO7 for Marvel-based Alpha systems
2169 See comment before marvel_specify_io7 in
2170 arch/alpha/kernel/core_marvel.c.
2172 io_delay= [X86] I/O delay method
2174 Standard port 0x80 based delay
2176 Alternate port 0xed based delay (needed on some systems)
2178 Simple two microseconds delay
2183 See Documentation/admin-guide/nfs/nfsroot.rst.
2185 ipcmni_extend [KNL] Extend the maximum number of unique System V
2186 IPC identifiers from 32,768 to 16,777,216.
2188 irqaffinity= [SMP] Set the default irq affinity mask
2189 The argument is a cpu list, as described above.
2191 irqchip.gicv2_force_probe=
2194 Force the kernel to look for the second 4kB page
2195 of a GICv2 controller even if the memory range
2196 exposed by the device tree is too small.
2198 irqchip.gicv3_nolpi=
2200 Force the kernel to ignore the availability of
2201 LPIs (and by consequence ITSs). Intended for system
2202 that use the kernel as a bootloader, and thus want
2203 to let secondary kernels in charge of setting up
2206 irqchip.gicv3_pseudo_nmi= [ARM64]
2207 Enables support for pseudo-NMIs in the kernel. This
2208 requires the kernel to be built with
2209 CONFIG_ARM64_PSEUDO_NMI.
2212 When an interrupt is not handled search all handlers
2213 for it. Intended to get systems with badly broken
2217 When an interrupt is not handled search all handlers
2218 for it. Also check all handlers each timer
2219 interrupt. Intended to get systems with badly broken
2223 Format: <RDP>,<reset>,<pci_scan>,<verbosity>
2225 isolcpus= [KNL,SMP,ISOL] Isolate a given set of CPUs from disturbance.
2226 [Deprecated - use cpusets instead]
2227 Format: [flag-list,]<cpu-list>
2229 Specify one or more CPUs to isolate from disturbances
2230 specified in the flag list (default: domain):
2233 Disable the tick when a single task runs.
2235 A residual 1Hz tick is offloaded to workqueues, which you
2236 need to affine to housekeeping through the global
2237 workqueue's affinity configured via the
2238 /sys/devices/virtual/workqueue/cpumask sysfs file, or
2239 by using the 'domain' flag described below.
2241 NOTE: by default the global workqueue runs on all CPUs,
2242 so to protect individual CPUs the 'cpumask' file has to
2243 be configured manually after bootup.
2246 Isolate from the general SMP balancing and scheduling
2247 algorithms. Note that performing domain isolation this way
2248 is irreversible: it's not possible to bring back a CPU to
2249 the domains once isolated through isolcpus. It's strongly
2250 advised to use cpusets instead to disable scheduler load
2251 balancing through the "cpuset.sched_load_balance" file.
2252 It offers a much more flexible interface where CPUs can
2253 move in and out of an isolated set anytime.
2255 You can move a process onto or off an "isolated" CPU via
2256 the CPU affinity syscalls or cpuset.
2257 <cpu number> begins at 0 and the maximum value is
2258 "number of CPUs in system - 1".
2262 Isolate from being targeted by managed interrupts
2263 which have an interrupt mask containing isolated
2264 CPUs. The affinity of managed interrupts is
2265 handled by the kernel and cannot be changed via
2266 the /proc/irq/* interfaces.
2268 This isolation is best effort and only effective
2269 if the automatically assigned interrupt mask of a
2270 device queue contains isolated and housekeeping
2271 CPUs. If housekeeping CPUs are online then such
2272 interrupts are directed to the housekeeping CPU
2273 so that IO submitted on the housekeeping CPU
2274 cannot disturb the isolated CPU.
2276 If a queue's affinity mask contains only isolated
2277 CPUs then this parameter has no effect on the
2278 interrupt routing decision, though interrupts are
2279 only delivered when tasks running on those
2280 isolated CPUs submit IO. IO submitted on
2281 housekeeping CPUs has no influence on those
2284 The format of <cpu-list> is described above.
2288 ivrs_ioapic [HW,X86-64]
2289 Provide an override to the IOAPIC-ID<->DEVICE-ID
2290 mapping provided in the IVRS ACPI table.
2291 By default, PCI segment is 0, and can be omitted.
2293 * To map IOAPIC-ID decimal 10 to PCI device 00:14.0
2294 write the parameter as:
2295 ivrs_ioapic[10]=00:14.0
2296 * To map IOAPIC-ID decimal 10 to PCI segment 0x1 and
2297 PCI device 00:14.0 write the parameter as:
2298 ivrs_ioapic[10]=0001:00:14.0
2300 ivrs_hpet [HW,X86-64]
2301 Provide an override to the HPET-ID<->DEVICE-ID
2302 mapping provided in the IVRS ACPI table.
2303 By default, PCI segment is 0, and can be omitted.
2305 * To map HPET-ID decimal 0 to PCI device 00:14.0
2306 write the parameter as:
2307 ivrs_hpet[0]=00:14.0
2308 * To map HPET-ID decimal 10 to PCI segment 0x1 and
2309 PCI device 00:14.0 write the parameter as:
2310 ivrs_ioapic[10]=0001:00:14.0
2312 ivrs_acpihid [HW,X86-64]
2313 Provide an override to the ACPI-HID:UID<->DEVICE-ID
2314 mapping provided in the IVRS ACPI table.
2316 For example, to map UART-HID:UID AMD0020:0 to
2317 PCI segment 0x1 and PCI device ID 00:14.5,
2318 write the parameter as:
2319 ivrs_acpihid[0001:00:14.5]=AMD0020:0
2321 By default, PCI segment is 0, and can be omitted.
2322 For example, PCI device 00:14.5 write the parameter as:
2323 ivrs_acpihid[00:14.5]=AMD0020:0
2325 js= [HW,JOY] Analog joystick
2326 See Documentation/input/joydev/joystick.rst.
2329 When CONFIG_RANDOMIZE_BASE is set, this disables
2330 kernel and module base offset ASLR (Address Space
2331 Layout Randomization).
2334 [KNL] Enforce KASAN (Kernel Address Sanitizer) to print
2335 report on every invalid memory access. Without this
2336 parameter KASAN will print report only for the first
2341 kernelcore= [KNL,X86,IA-64,PPC]
2342 Format: nn[KMGTPE] | nn% | "mirror"
2343 This parameter specifies the amount of memory usable by
2344 the kernel for non-movable allocations. The requested
2345 amount is spread evenly throughout all nodes in the
2346 system as ZONE_NORMAL. The remaining memory is used for
2347 movable memory in its own zone, ZONE_MOVABLE. In the
2348 event, a node is too small to have both ZONE_NORMAL and
2349 ZONE_MOVABLE, kernelcore memory will take priority and
2350 other nodes will have a larger ZONE_MOVABLE.
2352 ZONE_MOVABLE is used for the allocation of pages that
2353 may be reclaimed or moved by the page migration
2354 subsystem. Note that allocations like PTEs-from-HighMem
2355 still use the HighMem zone if it exists, and the Normal
2356 zone if it does not.
2358 It is possible to specify the exact amount of memory in
2359 the form of "nn[KMGTPE]", a percentage of total system
2360 memory in the form of "nn%", or "mirror". If "mirror"
2361 option is specified, mirrored (reliable) memory is used
2362 for non-movable allocations and remaining memory is used
2363 for Movable pages. "nn[KMGTPE]", "nn%", and "mirror"
2364 are exclusive, so you cannot specify multiple forms.
2366 kgdbdbgp= [KGDB,HW] kgdb over EHCI usb debug port.
2367 Format: <Controller#>[,poll interval]
2368 The controller # is the number of the ehci usb debug
2369 port as it is probed via PCI. The poll interval is
2370 optional and is the number seconds in between
2371 each poll cycle to the debug port in case you need
2372 the functionality for interrupting the kernel with
2373 gdb or control-c on the dbgp connection. When
2374 not using this parameter you use sysrq-g to break into
2375 the kernel debugger.
2377 kgdboc= [KGDB,HW] kgdb over consoles.
2378 Requires a tty driver that supports console polling,
2379 or a supported polling keyboard driver (non-usb).
2380 Serial only format: <serial_device>[,baud]
2381 keyboard only format: kbd
2382 keyboard and serial format: kbd,<serial_device>[,baud]
2383 Optional Kernel mode setting:
2384 kms, kbd format: kms,kbd
2385 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud]
2387 kgdboc_earlycon= [KGDB,HW]
2388 If the boot console provides the ability to read
2389 characters and can work in polling mode, you can use
2390 this parameter to tell kgdb to use it as a backend
2391 until the normal console is registered. Intended to
2392 be used together with the kgdboc parameter which
2393 specifies the normal console to transition to.
2395 The name of the early console should be specified
2396 as the value of this parameter. Note that the name of
2397 the early console might be different than the tty
2398 name passed to kgdboc. It's OK to leave the value
2399 blank and the first boot console that implements
2400 read() will be picked.
2402 kgdbwait [KGDB] Stop kernel execution and enter the
2403 kernel debugger at the earliest opportunity.
2405 kmac= [MIPS] Korina ethernet MAC address.
2406 Configure the RouterBoard 532 series on-chip
2407 Ethernet adapter MAC address.
2409 kmemleak= [KNL] Boot-time kmemleak enable/disable
2410 Valid arguments: on, off
2412 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y,
2415 kprobe_event=[probe-list]
2416 [FTRACE] Add kprobe events and enable at boot time.
2417 The probe-list is a semicolon delimited list of probe
2418 definitions. Each definition is same as kprobe_events
2419 interface, but the parameters are comma delimited.
2420 For example, to add a kprobe event on vfs_read with
2421 arg1 and arg2, add to the command line;
2423 kprobe_event=p,vfs_read,$arg1,$arg2
2425 See also Documentation/trace/kprobetrace.rst "Kernel
2426 Boot Parameter" section.
2428 kpti= [ARM64] Control page table isolation of user
2429 and kernel address spaces.
2430 Default: enabled on cores which need mitigation.
2434 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs.
2435 Default is 0 (don't ignore, but inject #GP)
2437 kvm.eager_page_split=
2438 [KVM,X86] Controls whether or not KVM will try to
2439 proactively split all huge pages during dirty logging.
2440 Eager page splitting reduces interruptions to vCPU
2441 execution by eliminating the write-protection faults
2442 and MMU lock contention that would otherwise be
2443 required to split huge pages lazily.
2445 VM workloads that rarely perform writes or that write
2446 only to a small region of VM memory may benefit from
2447 disabling eager page splitting to allow huge pages to
2448 still be used for reads.
2450 The behavior of eager page splitting depends on whether
2451 KVM_DIRTY_LOG_INITIALLY_SET is enabled or disabled. If
2452 disabled, all huge pages in a memslot will be eagerly
2453 split when dirty logging is enabled on that memslot. If
2454 enabled, eager page splitting will be performed during
2455 the KVM_CLEAR_DIRTY ioctl, and only for the pages being
2458 Eager page splitting is only supported when kvm.tdp_mmu=Y.
2462 kvm.enable_vmware_backdoor=[KVM] Support VMware backdoor PV interface.
2463 Default is false (don't support).
2466 [KVM] Controls the software workaround for the
2467 X86_BUG_ITLB_MULTIHIT bug.
2468 force : Always deploy workaround.
2469 off : Never deploy workaround.
2470 auto : Deploy workaround based on the presence of
2471 X86_BUG_ITLB_MULTIHIT.
2475 If the software workaround is enabled for the host,
2476 guests do need not to enable it for nested guests.
2478 kvm.nx_huge_pages_recovery_ratio=
2479 [KVM] Controls how many 4KiB pages are periodically zapped
2480 back to huge pages. 0 disables the recovery, otherwise if
2481 the value is N KVM will zap 1/Nth of the 4KiB pages every
2482 period (see below). The default is 60.
2484 kvm.nx_huge_pages_recovery_period_ms=
2485 [KVM] Controls the time period at which KVM zaps 4KiB pages
2486 back to huge pages. If the value is a non-zero N, KVM will
2487 zap a portion (see ratio above) of the pages every N msecs.
2488 If the value is 0 (the default), KVM will pick a period based
2489 on the ratio, such that a page is zapped after 1 hour on average.
2491 kvm-amd.nested= [KVM,AMD] Allow nested virtualization in KVM/SVM.
2492 Default is 1 (enabled)
2494 kvm-amd.npt= [KVM,AMD] Disable nested paging (virtualized MMU)
2496 Default is 1 (enabled) if in 64-bit or 32-bit PAE mode.
2499 [KVM,ARM] Select one of KVM/arm64's modes of operation.
2501 none: Forcefully disable KVM.
2503 nvhe: Standard nVHE-based mode, without support for
2506 protected: nVHE-based mode with support for guests whose
2507 state is kept private from the host.
2509 Defaults to VHE/nVHE based on hardware support. Setting
2510 mode to "protected" will disable kexec and hibernation
2513 kvm-arm.vgic_v3_group0_trap=
2514 [KVM,ARM] Trap guest accesses to GICv3 group-0
2517 kvm-arm.vgic_v3_group1_trap=
2518 [KVM,ARM] Trap guest accesses to GICv3 group-1
2521 kvm-arm.vgic_v3_common_trap=
2522 [KVM,ARM] Trap guest accesses to GICv3 common
2525 kvm-arm.vgic_v4_enable=
2526 [KVM,ARM] Allow use of GICv4 for direct injection of
2529 kvm_cma_resv_ratio=n [PPC]
2530 Reserves given percentage from system memory area for
2531 contiguous memory allocation for KVM hash pagetable
2533 By default it reserves 5% of total system memory.
2537 kvm-intel.ept= [KVM,Intel] Disable extended page tables
2538 (virtualized MMU) support on capable Intel chips.
2539 Default is 1 (enabled)
2541 kvm-intel.emulate_invalid_guest_state=
2542 [KVM,Intel] Disable emulation of invalid guest state.
2543 Ignored if kvm-intel.enable_unrestricted_guest=1, as
2544 guest state is never invalid for unrestricted guests.
2545 This param doesn't apply to nested guests (L2), as KVM
2546 never emulates invalid L2 guest state.
2547 Default is 1 (enabled)
2549 kvm-intel.flexpriority=
2550 [KVM,Intel] Disable FlexPriority feature (TPR shadow).
2551 Default is 1 (enabled)
2554 [KVM,Intel] Enable VMX nesting (nVMX).
2555 Default is 0 (disabled)
2557 kvm-intel.unrestricted_guest=
2558 [KVM,Intel] Disable unrestricted guest feature
2559 (virtualized real and unpaged mode) on capable
2560 Intel chips. Default is 1 (enabled)
2562 kvm-intel.vmentry_l1d_flush=[KVM,Intel] Mitigation for L1 Terminal Fault
2565 Valid arguments: never, cond, always
2567 always: L1D cache flush on every VMENTER.
2568 cond: Flush L1D on VMENTER only when the code between
2569 VMEXIT and VMENTER can leak host memory.
2570 never: Disables the mitigation
2572 Default is cond (do L1 cache flush in specific instances)
2574 kvm-intel.vpid= [KVM,Intel] Disable Virtual Processor Identification
2575 feature (tagged TLBs) on capable Intel chips.
2576 Default is 1 (enabled)
2578 l1d_flush= [X86,INTEL]
2579 Control mitigation for L1D based snooping vulnerability.
2581 Certain CPUs are vulnerable to an exploit against CPU
2582 internal buffers which can forward information to a
2583 disclosure gadget under certain conditions.
2585 In vulnerable processors, the speculatively
2586 forwarded data can be used in a cache side channel
2587 attack, to access data to which the attacker does
2588 not have direct access.
2590 This parameter controls the mitigation. The
2593 on - enable the interface for the mitigation
2595 l1tf= [X86] Control mitigation of the L1TF vulnerability on
2598 The kernel PTE inversion protection is unconditionally
2599 enabled and cannot be disabled.
2602 Provides all available mitigations for the
2603 L1TF vulnerability. Disables SMT and
2604 enables all mitigations in the
2605 hypervisors, i.e. unconditional L1D flush.
2607 SMT control and L1D flush control via the
2608 sysfs interface is still possible after
2609 boot. Hypervisors will issue a warning
2610 when the first VM is started in a
2611 potentially insecure configuration,
2612 i.e. SMT enabled or L1D flush disabled.
2615 Same as 'full', but disables SMT and L1D
2616 flush runtime control. Implies the
2617 'nosmt=force' command line option.
2618 (i.e. sysfs control of SMT is disabled.)
2621 Leaves SMT enabled and enables the default
2622 hypervisor mitigation, i.e. conditional
2625 SMT control and L1D flush control via the
2626 sysfs interface is still possible after
2627 boot. Hypervisors will issue a warning
2628 when the first VM is started in a
2629 potentially insecure configuration,
2630 i.e. SMT enabled or L1D flush disabled.
2634 Disables SMT and enables the default
2635 hypervisor mitigation.
2637 SMT control and L1D flush control via the
2638 sysfs interface is still possible after
2639 boot. Hypervisors will issue a warning
2640 when the first VM is started in a
2641 potentially insecure configuration,
2642 i.e. SMT enabled or L1D flush disabled.
2645 Same as 'flush', but hypervisors will not
2646 warn when a VM is started in a potentially
2647 insecure configuration.
2650 Disables hypervisor mitigations and doesn't
2652 It also drops the swap size and available
2653 RAM limit restriction on both hypervisor and
2658 For details see: Documentation/admin-guide/hw-vuln/l1tf.rst
2664 lapic [X86-32,APIC] Enable the local APIC even if BIOS
2667 lapic= [X86,APIC] Do not use TSC deadline
2668 value for LAPIC timer one-shot implementation. Default
2669 back to the programmable timer unit in the LAPIC.
2670 Format: notscdeadline
2672 lapic_timer_c2_ok [X86,APIC] trust the local apic timer
2675 libata.dma= [LIBATA] DMA control
2676 libata.dma=0 Disable all PATA and SATA DMA
2677 libata.dma=1 PATA and SATA Disk DMA only
2678 libata.dma=2 ATAPI (CDROM) DMA only
2679 libata.dma=4 Compact Flash DMA only
2680 Combinations also work, so libata.dma=3 enables DMA
2681 for disks and CDROMs, but not CFs.
2683 libata.ignore_hpa= [LIBATA] Ignore HPA limit
2684 libata.ignore_hpa=0 keep BIOS limits (default)
2685 libata.ignore_hpa=1 ignore limits, using full disk
2687 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume
2691 libata.force= [LIBATA] Force configurations. The format is a comma-
2692 separated list of "[ID:]VAL" where ID is PORT[.DEVICE].
2693 PORT and DEVICE are decimal numbers matching port, link
2694 or device. Basically, it matches the ATA ID string
2695 printed on console by libata. If the whole ID part is
2696 omitted, the last PORT and DEVICE values are used. If
2697 ID hasn't been specified yet, the configuration applies
2698 to all ports, links and devices.
2700 If only DEVICE is omitted, the parameter applies to
2701 the port and all links and devices behind it. DEVICE
2702 number of 0 either selects the first device or the
2703 first fan-out link behind PMP device. It does not
2704 select the host link. DEVICE number of 15 selects the
2705 host link and device attached to it.
2707 The VAL specifies the configuration to force. As long
2708 as there is no ambiguity, shortcut notation is allowed.
2709 For example, both 1.5 and 1.5G would work for 1.5Gbps.
2710 The following configurations can be forced.
2712 * Cable type: 40c, 80c, short40c, unk, ign or sata.
2713 Any ID with matching PORT is used.
2715 * SATA link speed limit: 1.5Gbps or 3.0Gbps.
2717 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7].
2718 udma[/][16,25,33,44,66,100,133] notation is also
2721 * nohrst, nosrst, norst: suppress hard, soft and both
2724 * rstonce: only attempt one reset during hot-unplug
2727 * [no]dbdelay: Enable or disable the extra 200ms delay
2728 before debouncing a link PHY and device presence
2731 * [no]ncq: Turn on or off NCQ.
2733 * [no]ncqtrim: Enable or disable queued DSM TRIM.
2735 * [no]ncqati: Enable or disable NCQ trim on ATI chipset.
2737 * [no]trim: Enable or disable (unqueued) TRIM.
2739 * trim_zero: Indicate that TRIM command zeroes data.
2741 * max_trim_128m: Set 128M maximum trim size limit.
2743 * [no]dma: Turn on or off DMA transfers.
2745 * atapi_dmadir: Enable ATAPI DMADIR bridge support.
2747 * atapi_mod16_dma: Enable the use of ATAPI DMA for
2748 commands that are not a multiple of 16 bytes.
2750 * [no]dmalog: Enable or disable the use of the
2751 READ LOG DMA EXT command to access logs.
2753 * [no]iddevlog: Enable or disable access to the
2754 identify device data log.
2756 * [no]logdir: Enable or disable access to the general
2757 purpose log directory.
2759 * max_sec_128: Set transfer size limit to 128 sectors.
2761 * max_sec_1024: Set or clear transfer size limit to
2764 * max_sec_lba48: Set or clear transfer size limit to
2767 * [no]lpm: Enable or disable link power management.
2769 * [no]setxfer: Indicate if transfer speed mode setting
2772 * dump_id: Dump IDENTIFY data.
2774 * disable: Disable this device.
2776 If there are multiple matching configurations changing
2777 the same attribute, the last one is used.
2779 load_ramdisk= [RAM] [Deprecated]
2781 lockd.nlm_grace_period=P [NFS] Assign grace period.
2784 lockd.nlm_tcpport=N [NFS] Assign TCP port.
2787 lockd.nlm_timeout=T [NFS] Assign timeout value.
2790 lockd.nlm_udpport=M [NFS] Assign UDP port.
2793 lockdown= [SECURITY]
2794 { integrity | confidentiality }
2795 Enable the kernel lockdown feature. If set to
2796 integrity, kernel features that allow userland to
2797 modify the running kernel are disabled. If set to
2798 confidentiality, kernel features that allow userland
2799 to extract confidential information from the kernel
2802 locktorture.nreaders_stress= [KNL]
2803 Set the number of locking read-acquisition kthreads.
2804 Defaults to being automatically set based on the
2805 number of online CPUs.
2807 locktorture.nwriters_stress= [KNL]
2808 Set the number of locking write-acquisition kthreads.
2810 locktorture.onoff_holdoff= [KNL]
2811 Set time (s) after boot for CPU-hotplug testing.
2813 locktorture.onoff_interval= [KNL]
2814 Set time (s) between CPU-hotplug operations, or
2815 zero to disable CPU-hotplug testing.
2817 locktorture.shuffle_interval= [KNL]
2818 Set task-shuffle interval (jiffies). Shuffling
2819 tasks allows some CPUs to go into dyntick-idle
2820 mode during the locktorture test.
2822 locktorture.shutdown_secs= [KNL]
2823 Set time (s) after boot system shutdown. This
2824 is useful for hands-off automated testing.
2826 locktorture.stat_interval= [KNL]
2827 Time (s) between statistics printk()s.
2829 locktorture.stutter= [KNL]
2830 Time (s) to stutter testing, for example,
2831 specifying five seconds causes the test to run for
2832 five seconds, wait for five seconds, and so on.
2833 This tests the locking primitive's ability to
2834 transition abruptly to and from idle.
2836 locktorture.torture_type= [KNL]
2837 Specify the locking implementation to test.
2839 locktorture.verbose= [KNL]
2840 Enable additional printk() statements.
2842 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver
2845 loglevel= All Kernel Messages with a loglevel smaller than the
2846 console loglevel will be printed to the console. It can
2847 also be changed with klogd or other programs. The
2848 loglevels are defined as follows:
2850 0 (KERN_EMERG) system is unusable
2851 1 (KERN_ALERT) action must be taken immediately
2852 2 (KERN_CRIT) critical conditions
2853 3 (KERN_ERR) error conditions
2854 4 (KERN_WARNING) warning conditions
2855 5 (KERN_NOTICE) normal but significant condition
2856 6 (KERN_INFO) informational
2857 7 (KERN_DEBUG) debug-level messages
2859 log_buf_len=n[KMG] Sets the size of the printk ring buffer,
2860 in bytes. n must be a power of two and greater
2861 than the minimal size. The minimal size is defined
2862 by LOG_BUF_SHIFT kernel config parameter. There is
2863 also CONFIG_LOG_CPU_MAX_BUF_SHIFT config parameter
2864 that allows to increase the default size depending on
2865 the number of CPUs. See init/Kconfig for more details.
2867 logo.nologo [FB] Disables display of the built-in Linux logo.
2868 This may be used to provide more screen space for
2869 kernel log messages and is useful when debugging
2870 kernel boot problems.
2872 lp=0 [LP] Specify parallel ports to use, e.g,
2873 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses
2874 lp=reset first parallel port). 'lp=0' disables the
2875 lp=auto printer driver. 'lp=reset' (which can be
2876 specified in addition to the ports) causes
2877 attached printers to be reset. Using
2878 lp=port1,port2,... specifies the parallel ports
2879 to associate lp devices with, starting with
2880 lp0. A port specification may be 'none' to skip
2881 that lp device, or a parport name such as
2882 'parport0'. Specifying 'lp=auto' instead of a
2883 port specification list means that device IDs
2884 from each port should be examined, to see if
2885 an IEEE 1284-compliant printer is attached; if
2886 so, the driver will manage that printer.
2887 See also header of drivers/char/lp.c.
2890 Sets loops_per_jiffy to given constant, thus avoiding
2891 time-consuming boot-time autodetection (up to 250 ms per
2892 CPU). 0 enables autodetection (default). To determine
2893 the correct value for your kernel, boot with normal
2894 autodetection and see what value is printed. Note that
2895 on SMP systems the preset will be applied to all CPUs,
2896 which is likely to cause problems if your CPUs need
2897 significantly divergent settings. An incorrect value
2898 will cause delays in the kernel to be wrong, leading to
2899 unpredictable I/O errors and other breakage. Although
2900 unlikely, in the extreme case this might damage your
2904 Format: <io>,<irq>,<dma>
2906 lsm.debug [SECURITY] Enable LSM initialization debugging output.
2909 [SECURITY] Choose order of LSM initialization. This
2910 overrides CONFIG_LSM, and the "security=" parameter.
2912 machvec= [IA-64] Force the use of a particular machine-vector
2913 (machvec) in a generic kernel.
2914 Example: machvec=hpzx1
2916 machtype= [Loongson] Share the same kernel image file between
2917 different yeeloong laptops.
2918 Example: machtype=lemote-yeeloong-2f-7inch
2920 max_addr=nn[KMG] [KNL,BOOT,IA-64] All physical memory greater
2921 than or equal to this physical address is ignored.
2923 maxcpus= [SMP] Maximum number of processors that an SMP kernel
2924 will bring up during bootup. maxcpus=n : n >= 0 limits
2925 the kernel to bring up 'n' processors. Surely after
2926 bootup you can bring up the other plugged cpu by executing
2927 "echo 1 > /sys/devices/system/cpu/cpuX/online". So maxcpus
2928 only takes effect during system bootup.
2929 While n=0 is a special case, it is equivalent to "nosmp",
2930 which also disables the IO APIC.
2932 max_loop= [LOOP] The number of loop block devices that get
2933 (loop.max_loop) unconditionally pre-created at init time. The default
2934 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead
2935 of statically allocating a predefined number, loop
2936 devices can be requested on-demand with the
2937 /dev/loop-control interface.
2939 mce [X86-32] Machine Check Exception
2941 mce=option [X86-64] See Documentation/x86/x86_64/boot-options.rst
2943 md= [HW] RAID subsystems devices and level
2944 See Documentation/admin-guide/md.rst.
2947 Format: <first>,<last>
2948 Specifies range of consoles to be captured by the MDA.
2951 Control mitigation for the Micro-architectural Data
2952 Sampling (MDS) vulnerability.
2954 Certain CPUs are vulnerable to an exploit against CPU
2955 internal buffers which can forward information to a
2956 disclosure gadget under certain conditions.
2958 In vulnerable processors, the speculatively
2959 forwarded data can be used in a cache side channel
2960 attack, to access data to which the attacker does
2961 not have direct access.
2963 This parameter controls the MDS mitigation. The
2966 full - Enable MDS mitigation on vulnerable CPUs
2967 full,nosmt - Enable MDS mitigation and disable
2968 SMT on vulnerable CPUs
2969 off - Unconditionally disable MDS mitigation
2971 On TAA-affected machines, mds=off can be prevented by
2972 an active TAA mitigation as both vulnerabilities are
2973 mitigated with the same mechanism so in order to disable
2974 this mitigation, you need to specify tsx_async_abort=off
2977 Not specifying this option is equivalent to
2980 For details see: Documentation/admin-guide/hw-vuln/mds.rst
2982 mem=nn[KMG] [HEXAGON] Set the memory size.
2983 Must be specified, otherwise memory size will be 0.
2985 mem=nn[KMG] [KNL,BOOT] Force usage of a specific amount of memory
2986 Amount of memory to be used in cases as follows:
2989 2 when the kernel is not able to see the whole system memory;
2990 3 memory that lies after 'mem=' boundary is excluded from
2991 the hypervisor, then assigned to KVM guests.
2992 4 to limit the memory available for kdump kernel.
2994 [ARC,MICROBLAZE] - the limit applies only to low memory,
2995 high memory is not affected.
2997 [ARM64] - only limits memory covered by the linear
2998 mapping. The NOMAP regions are not affected.
3000 [X86] Work as limiting max address. Use together
3001 with memmap= to avoid physical address space collisions.
3002 Without memmap= PCI devices could be placed at addresses
3003 belonging to unused RAM.
3005 Note that this only takes effects during boot time since
3006 in above case 3, memory may need be hot added after boot
3007 if system memory of hypervisor is not sufficient.
3010 [ARM,MIPS] - override the memory layout reported by
3012 Define a memory region of size nn[KMG] starting at
3014 Multiple different regions can be specified with
3015 multiple mem= parameters on the command line.
3017 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel
3020 memblock=debug [KNL] Enable memblock debug messages.
3023 [KNL,SH] Allow user to override the default size for
3024 per-device physically contiguous DMA buffers.
3026 memhp_default_state=online/offline
3027 [KNL] Set the initial state for the memory hotplug
3028 onlining policy. If not specified, the default value is
3029 set according to the
3030 CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE kernel config
3032 See Documentation/admin-guide/mm/memory-hotplug.rst.
3034 memmap=exactmap [KNL,X86] Enable setting of an exact
3035 E820 memory map, as specified by the user.
3036 Such memmap=exactmap lines can be constructed based on
3037 BIOS output or other requirements. See the memmap=nn@ss
3040 memmap=nn[KMG]@ss[KMG]
3041 [KNL, X86, MIPS, XTENSA] Force usage of a specific region of memory.
3042 Region of memory to be used is from ss to ss+nn.
3043 If @ss[KMG] is omitted, it is equivalent to mem=nn[KMG],
3044 which limits max address to nn[KMG].
3045 Multiple different regions can be specified,
3048 memmap=100M@2G,100M#3G,1G!1024G
3050 memmap=nn[KMG]#ss[KMG]
3051 [KNL,ACPI] Mark specific memory as ACPI data.
3052 Region of memory to be marked is from ss to ss+nn.
3054 memmap=nn[KMG]$ss[KMG]
3055 [KNL,ACPI] Mark specific memory as reserved.
3056 Region of memory to be reserved is from ss to ss+nn.
3057 Example: Exclude memory from 0x18690000-0x1869ffff
3058 memmap=64K$0x18690000
3060 memmap=0x10000$0x18690000
3061 Some bootloaders may need an escape character before '$',
3062 like Grub2, otherwise '$' and the following number
3065 memmap=nn[KMG]!ss[KMG]
3066 [KNL,X86] Mark specific memory as protected.
3067 Region of memory to be used, from ss to ss+nn.
3068 The memory region may be marked as e820 type 12 (0xc)
3069 and is NVDIMM or ADR memory.
3071 memmap=<size>%<offset>-<oldtype>+<newtype>
3072 [KNL,ACPI] Convert memory within the specified region
3073 from <oldtype> to <newtype>. If "-<oldtype>" is left
3074 out, the whole region will be marked as <newtype>,
3075 even if previously unavailable. If "+<newtype>" is left
3076 out, matching memory will be removed. Types are
3077 specified as e820 types, e.g., 1 = RAM, 2 = reserved,
3078 3 = ACPI, 12 = PRAM.
3080 memory_corruption_check=0/1 [X86]
3081 Some BIOSes seem to corrupt the first 64k of
3082 memory when doing things like suspend/resume.
3083 Setting this option will scan the memory
3084 looking for corruption. Enabling this will
3085 both detect corruption and prevent the kernel
3086 from using the memory being corrupted.
3087 However, its intended as a diagnostic tool; if
3088 repeatable BIOS-originated corruption always
3089 affects the same memory, you can use memmap=
3090 to prevent the kernel from using that memory.
3092 memory_corruption_check_size=size [X86]
3093 By default it checks for corruption in the low
3094 64k, making this memory unavailable for normal
3095 use. Use this parameter to scan for
3096 corruption in more or less memory.
3098 memory_corruption_check_period=seconds [X86]
3099 By default it checks for corruption every 60
3100 seconds. Use this parameter to check at some
3101 other rate. 0 disables periodic checking.
3103 memory_hotplug.memmap_on_memory
3104 [KNL,X86,ARM] Boolean flag to enable this feature.
3105 Format: {on | off (default)}
3106 When enabled, runtime hotplugged memory will
3107 allocate its internal metadata (struct pages,
3108 those vmemmap pages cannot be optimized even
3109 if hugetlb_free_vmemmap is enabled) from the
3110 hotadded memory which will allow to hotadd a
3111 lot of memory without requiring additional
3113 This feature is disabled by default because it
3114 has some implication on large (e.g. GB)
3115 allocations in some configurations (e.g. small
3117 The state of the flag can be read in
3118 /sys/module/memory_hotplug/parameters/memmap_on_memory.
3119 Note that even when enabled, there are a few cases where
3120 the feature is not effective.
3122 memtest= [KNL,X86,ARM,M68K,PPC,RISCV] Enable memtest
3124 default : 0 <disable>
3125 Specifies the number of memtest passes to be
3126 performed. Each pass selects another test
3127 pattern from a given set of patterns. Memtest
3128 fills the memory with this pattern, validates
3129 memory contents and reserves bad memory
3130 regions that are detected.
3132 mem_encrypt= [X86-64] AMD Secure Memory Encryption (SME) control
3133 Valid arguments: on, off
3134 Default (depends on kernel configuration option):
3135 on (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=y)
3136 off (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=n)
3137 mem_encrypt=on: Activate SME
3138 mem_encrypt=off: Do not activate SME
3140 Refer to Documentation/virt/kvm/x86/amd-memory-encryption.rst
3141 for details on when memory encryption can be activated.
3143 mem_sleep_default= [SUSPEND] Default system suspend mode:
3144 s2idle - Suspend-To-Idle
3145 shallow - Power-On Suspend or equivalent (if supported)
3146 deep - Suspend-To-RAM or equivalent (if supported)
3147 See Documentation/admin-guide/pm/sleep-states.rst.
3149 meye.*= [HW] Set MotionEye Camera parameters
3150 See Documentation/admin-guide/media/meye.rst.
3152 mfgpt_irq= [IA-32] Specify the IRQ to use for the
3153 Multi-Function General Purpose Timers on AMD Geode
3156 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when
3157 the BIOS has incorrectly applied a workaround. TinyBIOS
3158 version 0.98 is known to be affected, 0.99 fixes the
3159 problem by letting the user disable the workaround.
3163 min_addr=nn[KMG] [KNL,BOOT,IA-64] All physical memory below this
3164 physical address is ignored.
3166 mini2440= [ARM,HW,KNL]
3167 Format:[0..2][b][c][t]
3169 MINI2440 configuration specification:
3170 0 - The attached screen is the 3.5" TFT
3171 1 - The attached screen is the 7" TFT
3172 2 - The VGA Shield is attached (1024x768)
3173 Leaving out the screen size parameter will not load
3174 the TFT driver, and the framebuffer will be left
3176 b - Enable backlight. The TFT backlight pin will be
3177 linked to the kernel VESA blanking code and a GPIO
3178 LED. This parameter is not necessary when using the
3180 c - Enable the s3c camera interface.
3181 t - Reserved for enabling touchscreen support. The
3182 touchscreen support is not enabled in the mainstream
3183 kernel as of 2.6.30, a preliminary port can be found
3184 in the "bleeding edge" mini2440 support kernel at
3185 https://repo.or.cz/w/linux-2.6/mini2440.git
3188 [X86,PPC,S390,ARM64] Control optional mitigations for
3189 CPU vulnerabilities. This is a set of curated,
3190 arch-independent options, each of which is an
3191 aggregation of existing arch-specific options.
3194 Disable all optional CPU mitigations. This
3195 improves system performance, but it may also
3196 expose users to several CPU vulnerabilities.
3197 Equivalent to: nopti [X86,PPC]
3198 if nokaslr then kpti=0 [ARM64]
3199 nospectre_v1 [X86,PPC]
3201 nospectre_v2 [X86,PPC,S390,ARM64]
3202 spectre_v2_user=off [X86]
3203 spec_store_bypass_disable=off [X86,PPC]
3204 ssbd=force-off [ARM64]
3207 tsx_async_abort=off [X86]
3208 kvm.nx_huge_pages=off [X86]
3209 srbds=off [X86,INTEL]
3210 no_entry_flush [PPC]
3211 no_uaccess_flush [PPC]
3212 mmio_stale_data=off [X86]
3216 This does not have any effect on
3217 kvm.nx_huge_pages when
3218 kvm.nx_huge_pages=force.
3221 Mitigate all CPU vulnerabilities, but leave SMT
3222 enabled, even if it's vulnerable. This is for
3223 users who don't want to be surprised by SMT
3224 getting disabled across kernel upgrades, or who
3225 have other ways of avoiding SMT-based attacks.
3226 Equivalent to: (default behavior)
3229 Mitigate all CPU vulnerabilities, disabling SMT
3230 if needed. This is for users who always want to
3231 be fully mitigated, even if it means losing SMT.
3232 Equivalent to: l1tf=flush,nosmt [X86]
3233 mds=full,nosmt [X86]
3234 tsx_async_abort=full,nosmt [X86]
3235 mmio_stale_data=full,nosmt [X86]
3236 retbleed=auto,nosmt [X86]
3239 [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this
3240 parameter allows control of the logging verbosity for
3241 the additional memory initialisation checks. A value
3242 of 0 disables mminit logging and a level of 4 will
3243 log everything. Information is printed at KERN_DEBUG
3244 so loglevel=8 may also need to be specified.
3247 [X86,INTEL] Control mitigation for the Processor
3248 MMIO Stale Data vulnerabilities.
3250 Processor MMIO Stale Data is a class of
3251 vulnerabilities that may expose data after an MMIO
3252 operation. Exposed data could originate or end in
3253 the same CPU buffers as affected by MDS and TAA.
3254 Therefore, similar to MDS and TAA, the mitigation
3255 is to clear the affected CPU buffers.
3257 This parameter controls the mitigation. The
3260 full - Enable mitigation on vulnerable CPUs
3262 full,nosmt - Enable mitigation and disable SMT on
3265 off - Unconditionally disable mitigation
3267 On MDS or TAA affected machines,
3268 mmio_stale_data=off can be prevented by an active
3269 MDS or TAA mitigation as these vulnerabilities are
3270 mitigated with the same mechanism so in order to
3271 disable this mitigation, you need to specify
3272 mds=off and tsx_async_abort=off too.
3274 Not specifying this option is equivalent to
3275 mmio_stale_data=full.
3278 Documentation/admin-guide/hw-vuln/processor_mmio_stale_data.rst
3281 [KNL] When CONFIG_MODULE_SIG is set, this means that
3282 modules without (valid) signatures will fail to load.
3283 Note that if CONFIG_MODULE_SIG_FORCE is set, that
3284 is always true, so this option does nothing.
3286 module_blacklist= [KNL] Do not load a comma-separated list of
3287 modules. Useful for debugging problem modules.
3290 [MOUSE] Maximum time between finger touching and
3291 leaving touchpad surface for touch to be considered
3292 a tap and be reported as a left button click (for
3293 touchpads working in absolute mode only).
3295 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices
3296 reporting absolute coordinates, such as tablets
3297 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices
3298 reporting absolute coordinates, such as tablets
3300 movablecore= [KNL,X86,IA-64,PPC]
3301 Format: nn[KMGTPE] | nn%
3302 This parameter is the complement to kernelcore=, it
3303 specifies the amount of memory used for migratable
3304 allocations. If both kernelcore and movablecore is
3305 specified, then kernelcore will be at *least* the
3306 specified value but may be more. If movablecore on its
3307 own is specified, the administrator must be careful
3308 that the amount of memory usable for all allocations
3311 movable_node [KNL] Boot-time switch to make hotplugable memory
3312 NUMA nodes to be movable. This means that the memory
3313 of such nodes will be usable only for movable
3314 allocations which rules out almost all kernel
3315 allocations. Use with caution!
3317 MTD_Partition= [MTD]
3318 Format: <name>,<region-number>,<size>,<offset>
3320 MTD_Region= [MTD] Format:
3321 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>]
3324 See drivers/mtd/parsers/cmdlinepart.c
3327 ARM/S3C2412 JIVE boot control
3329 See arch/arm/mach-s3c/mach-jive.c
3331 mtouchusb.raw_coordinates=
3332 [HW] Make the MicroTouch USB driver use raw coordinates
3333 ('y', default) or cooked coordinates ('n')
3335 mtrr_chunk_size=nn[KMG] [X86]
3336 used for mtrr cleanup. It is largest continuous chunk
3337 that could hold holes aka. UC entries.
3339 mtrr_gran_size=nn[KMG] [X86]
3340 Used for mtrr cleanup. It is granularity of mtrr block.
3342 Large value could prevent small alignment from
3345 mtrr_spare_reg_nr=n [X86]
3347 Range: 0,7 : spare reg number
3349 Used for mtrr cleanup. It is spare mtrr entries number.
3350 Set to 2 or more if your graphical card needs more.
3352 multitce=off [PPC] This parameter disables the use of the pSeries
3353 firmware feature for updating multiple TCE entries
3356 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card
3358 netdev= [NET] Network devices parameters
3359 Format: <irq>,<io>,<mem_start>,<mem_end>,<name>
3360 Note that mem_start is often overloaded to mean
3361 something different and driver-specific.
3362 This usage is only documented in each driver source
3365 netpoll.carrier_timeout=
3366 [NET] Specifies amount of time (in seconds) that
3367 netpoll should wait for a carrier. By default netpoll
3371 [NETFILTER] Enable connection tracking flow accounting
3372 0 to disable accounting
3373 1 to enable accounting
3376 nfsaddrs= [NFS] Deprecated. Use ip= instead.
3377 See Documentation/admin-guide/nfs/nfsroot.rst.
3379 nfsroot= [NFS] nfs root filesystem for disk-less boxes.
3380 See Documentation/admin-guide/nfs/nfsroot.rst.
3382 nfsrootdebug [NFS] enable nfsroot debugging messages.
3383 See Documentation/admin-guide/nfs/nfsroot.rst.
3385 nfs.callback_nr_threads=
3386 [NFSv4] set the total number of threads that the
3387 NFS client will assign to service NFSv4 callback
3390 nfs.callback_tcpport=
3391 [NFS] set the TCP port on which the NFSv4 callback
3392 channel should listen.
3395 [NFS] sets the pathname to the program which is used
3396 to update the NFS client cache entries.
3398 nfs.cache_getent_timeout=
3399 [NFS] sets the timeout after which an attempt to
3400 update a cache entry is deemed to have failed.
3402 nfs.idmap_cache_timeout=
3403 [NFS] set the maximum lifetime for idmapper cache
3407 [NFS] enable 64-bit inode numbers.
3408 If zero, the NFS client will fake up a 32-bit inode
3409 number for the readdir() and stat() syscalls instead
3410 of returning the full 64-bit number.
3411 The default is to return 64-bit inode numbers.
3413 nfs.max_session_cb_slots=
3414 [NFSv4.1] Sets the maximum number of session
3415 slots the client will assign to the callback
3416 channel. This determines the maximum number of
3417 callbacks the client will process in parallel for
3418 a particular server.
3420 nfs.max_session_slots=
3421 [NFSv4.1] Sets the maximum number of session slots
3422 the client will attempt to negotiate with the server.
3423 This limits the number of simultaneous RPC requests
3424 that the client can send to the NFSv4.1 server.
3425 Note that there is little point in setting this
3426 value higher than the max_tcp_slot_table_limit.
3428 nfs.nfs4_disable_idmapping=
3429 [NFSv4] When set to the default of '1', this option
3430 ensures that both the RPC level authentication
3431 scheme and the NFS level operations agree to use
3432 numeric uids/gids if the mount is using the
3433 'sec=sys' security flavour. In effect it is
3434 disabling idmapping, which can make migration from
3435 legacy NFSv2/v3 systems to NFSv4 easier.
3436 Servers that do not support this mode of operation
3437 will be autodetected by the client, and it will fall
3438 back to using the idmapper.
3439 To turn off this behaviour, set the value to '0'.
3441 [NFS4] Specify an additional fixed unique ident-
3442 ification string that NFSv4 clients can insert into
3443 their nfs_client_id4 string. This is typically a
3444 UUID that is generated at system install time.
3446 nfs.send_implementation_id =
3447 [NFSv4.1] Send client implementation identification
3448 information in exchange_id requests.
3449 If zero, no implementation identification information
3451 The default is to send the implementation identification
3454 nfs.recover_lost_locks =
3455 [NFSv4] Attempt to recover locks that were lost due
3456 to a lease timeout on the server. Please note that
3457 doing this risks data corruption, since there are
3458 no guarantees that the file will remain unchanged
3459 after the locks are lost.
3460 If you want to enable the kernel legacy behaviour of
3461 attempting to recover these locks, then set this
3463 The default parameter value of '0' causes the kernel
3464 not to attempt recovery of lost locks.
3466 nfs4.layoutstats_timer =
3467 [NFSv4.2] Change the rate at which the kernel sends
3468 layoutstats to the pNFS metadata server.
3470 Setting this to value to 0 causes the kernel to use
3471 whatever value is the default set by the layout
3472 driver. A non-zero value sets the minimum interval
3473 in seconds between layoutstats transmissions.
3475 nfsd.inter_copy_offload_enable =
3476 [NFSv4.2] When set to 1, the server will support
3477 server-to-server copies for which this server is
3478 the destination of the copy.
3480 nfsd.nfsd4_ssc_umount_timeout =
3481 [NFSv4.2] When used as the destination of a
3482 server-to-server copy, knfsd temporarily mounts
3483 the source server. It caches the mount in case
3484 it will be needed again, and discards it if not
3485 used for the number of milliseconds specified by
3488 nfsd.nfs4_disable_idmapping=
3489 [NFSv4] When set to the default of '1', the NFSv4
3490 server will return only numeric uids and gids to
3491 clients using auth_sys, and will accept numeric uids
3492 and gids from such clients. This is intended to ease
3493 migration from NFSv2/v3.
3496 nmi_backtrace.backtrace_idle [KNL]
3497 Dump stacks even of idle CPUs in response to an
3498 NMI stack-backtrace request.
3500 nmi_debug= [KNL,SH] Specify one or more actions to take
3501 when a NMI is triggered.
3502 Format: [state][,regs][,debounce][,die]
3504 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels
3505 Format: [panic,][nopanic,][num]
3507 0 - turn hardlockup detector in nmi_watchdog off
3508 1 - turn hardlockup detector in nmi_watchdog on
3509 When panic is specified, panic when an NMI watchdog
3510 timeout occurs (or 'nopanic' to not panic on an NMI
3511 watchdog, if CONFIG_BOOTPARAM_HARDLOCKUP_PANIC is set)
3512 To disable both hard and soft lockup detectors,
3513 please see 'nowatchdog'.
3514 This is useful when you use a panic=... timeout and
3515 need the box quickly up again.
3517 These settings can be accessed at runtime via
3518 the nmi_watchdog and hardlockup_panic sysctls.
3520 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths
3521 emulation library even if a 387 maths coprocessor
3524 no5lvl [X86-64] Disable 5-level paging mode. Forces
3525 kernel to use 4-level paging instead.
3527 nofsgsbase [X86] Disables FSGSBASE instructions.
3530 [HW] Never suspend the console
3531 Disable suspending of consoles during suspend and
3532 hibernate operations. Once disabled, debugging
3533 messages can reach various consoles while the rest
3534 of the system is being put to sleep (ie, while
3535 debugging driver suspend/resume hooks). This may
3536 not work reliably with all consoles, but is known
3537 to work with serial and VGA consoles.
3538 To facilitate more flexible debugging, we also add
3539 console_suspend, a printk module parameter to control
3540 it. Users could use console_suspend (usually
3541 /sys/module/printk/parameters/console_suspend) to
3542 turn on/off it dynamically.
3544 novmcoredd [KNL,KDUMP]
3545 Disable device dump. Device dump allows drivers to
3546 append dump data to vmcore so you can collect driver
3547 specified debug info. Drivers can append the data
3548 without any limit and this data is stored in memory,
3549 so this may cause significant memory stress. Disabling
3550 device dump can help save memory but the driver debug
3551 data will be no longer available. This parameter
3552 is only available when CONFIG_PROC_VMCORE_DEVICE_DUMP
3555 noaliencache [MM, NUMA, SLAB] Disables the allocation of alien
3556 caches in the slab allocator. Saves per-node memory,
3557 but will impact performance.
3561 noaltinstr [S390] Disables alternative instructions patching
3562 (CPU alternatives feature).
3564 noapic [SMP,APIC] Tells the kernel to not make use of any
3565 IOAPICs that may be present in the system.
3567 noautogroup Disable scheduler automatic task group creation.
3571 nodsp [SH] Disable hardware DSP at boot time.
3573 noefi Disable EFI runtime services support.
3575 no_entry_flush [PPC] Don't flush the L1-D cache when entering the kernel.
3580 Disable SMAP (Supervisor Mode Access Prevention)
3581 even if it is supported by processor.
3584 Disable SMEP (Supervisor Mode Execution Prevention)
3585 even if it is supported by processor.
3588 This affects only 32-bit executables.
3589 noexec32=on: enable non-executable mappings (default)
3590 read doesn't imply executable mappings
3591 noexec32=off: disable non-executable mappings
3592 read implies executable mappings
3594 nofpu [MIPS,SH] Disable hardware FPU at boot time.
3596 nofxsr [BUGS=X86-32] Disables x86 floating point extended
3597 register save and restore. The kernel will only save
3598 legacy floating-point registers on task switch.
3600 nohugeiomap [KNL,X86,PPC,ARM64] Disable kernel huge I/O mappings.
3602 nohugevmalloc [PPC] Disable kernel huge vmalloc mappings.
3604 nosmt [KNL,S390] Disable symmetric multithreading (SMT).
3605 Equivalent to smt=1.
3607 [KNL,X86] Disable symmetric multithreading (SMT).
3608 nosmt=force: Force disable SMT, cannot be undone
3609 via the sysfs control file.
3611 nospectre_v1 [X86,PPC] Disable mitigations for Spectre Variant 1
3612 (bounds check bypass). With this option data leaks are
3613 possible in the system.
3615 nospectre_v2 [X86,PPC_FSL_BOOK3E,ARM64] Disable all mitigations for
3616 the Spectre variant 2 (indirect branch prediction)
3617 vulnerability. System may allow data leaks with this
3620 nospec_store_bypass_disable
3621 [HW] Disable all mitigations for the Speculative Store Bypass vulnerability
3624 [PPC] Don't flush the L1-D cache after accessing user data.
3626 noxsave [BUGS=X86] Disables x86 extended register state save
3627 and restore using xsave. The kernel will fallback to
3628 enabling legacy floating-point and sse state.
3630 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended
3631 register states. The kernel will fall back to use
3632 xsave to save the states. By using this parameter,
3633 performance of saving the states is degraded because
3634 xsave doesn't support modified optimization while
3635 xsaveopt supports it on xsaveopt enabled systems.
3637 noxsaves [X86] Disables xsaves and xrstors used in saving and
3638 restoring x86 extended register state in compacted
3639 form of xsave area. The kernel will fall back to use
3640 xsaveopt and xrstor to save and restore the states
3641 in standard form of xsave area. By using this
3642 parameter, xsave area per process might occupy more
3643 memory on xsaves enabled systems.
3645 nohlt [ARM,ARM64,MICROBLAZE,SH] Forces the kernel to busy wait
3646 in do_idle() and not use the arch_cpu_idle()
3647 implementation; requires CONFIG_GENERIC_IDLE_POLL_SETUP
3648 to be effective. This is useful on platforms where the
3649 sleep(SH) or wfi(ARM,ARM64) instructions do not work
3650 correctly or when doing power measurements to evalute
3651 the impact of the sleep instructions. This is also
3652 useful when using JTAG debugger.
3654 no_file_caps Tells the kernel not to honor file capabilities. The
3655 only way then for a file to be executed with privilege
3656 is to be setuid root or executed by root.
3658 nohalt [IA-64] Tells the kernel not to use the power saving
3659 function PAL_HALT_LIGHT when idle. This increases
3660 power-consumption. On the positive side, it reduces
3661 interrupt wake-up latency, which may improve performance
3662 in certain environments such as networked servers or
3666 Force pointers printed to the console or buffers to be
3667 unhashed. By default, when a pointer is printed via %p
3668 format string, that pointer is "hashed", i.e. obscured
3669 by hashing the pointer value. This is a security feature
3670 that hides actual kernel addresses from unprivileged
3671 users, but it also makes debugging the kernel more
3672 difficult since unequal pointers can no longer be
3673 compared. However, if this command-line option is
3674 specified, then all normal pointers will have their true
3675 value printed. This option should only be specified when
3676 debugging the kernel. Please do not use on production
3679 nohibernate [HIBERNATION] Disable hibernation and resume.
3681 nohz= [KNL] Boottime enable/disable dynamic ticks
3682 Valid arguments: on, off
3685 nohz_full= [KNL,BOOT,SMP,ISOL]
3686 The argument is a cpu list, as described above.
3687 In kernels built with CONFIG_NO_HZ_FULL=y, set
3688 the specified list of CPUs whose tick will be stopped
3689 whenever possible. The boot CPU will be forced outside
3690 the range to maintain the timekeeping. Any CPUs
3691 in this list will have their RCU callbacks offloaded,
3692 just as if they had also been called out in the
3693 rcu_nocbs= boot parameter.
3695 Note that this argument takes precedence over
3696 the CONFIG_RCU_NOCB_CPU_DEFAULT_ALL option.
3698 noiotrap [SH] Disables trapped I/O port accesses.
3700 noirqdebug [X86-32] Disables the code which attempts to detect and
3701 disable unhandled interrupt sources.
3703 no_timer_check [X86,APIC] Disables the code which tests for
3704 broken timer IRQ sources.
3706 noisapnp [ISAPNP] Disables ISA PnP code.
3708 noinitrd [RAM] Tells the kernel not to load any configured
3711 nointremap [X86-64, Intel-IOMMU] Do not enable interrupt
3713 [Deprecated - use intremap=off]
3717 noinvpcid [X86] Disable the INVPCID cpu feature.
3719 nojitter [IA-64] Disables jitter checking for ITC timers.
3721 no-kvmclock [X86,KVM] Disable paravirtualized KVM clock driver
3723 no-kvmapf [X86,KVM] Disable paravirtualized asynchronous page
3727 [X86,PV_OPS] Disable paravirtualized VMware scheduler
3728 clock and use the default one.
3730 no-steal-acc [X86,PV_OPS,ARM64] Disable paravirtualized steal time
3731 accounting. steal time is computed, but won't
3732 influence scheduler behaviour
3734 nolapic [X86-32,APIC] Do not enable or use the local APIC.
3736 nolapic_timer [X86-32,APIC] Do not use the local APIC timer.
3738 nomca [IA-64] Disable machine check abort handling
3740 nomce [X86-32] Disable Machine Check Exception
3742 nomfgpt [X86-32] Disable Multi-Function General Purpose
3743 Timer usage (for AMD Geode machines).
3745 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to
3746 shutdown the other cpus. Instead use the REBOOT_VECTOR
3749 nomodeset Disable kernel modesetting. DRM drivers will not perform
3750 display-mode changes or accelerated rendering. Only the
3751 system framebuffer will be available for use if this was
3752 set-up by the firmware or boot loader.
3754 Useful as fallback, or for testing and debugging.
3756 nomodule Disable module load
3758 nopat [X86] Disable PAT (page attribute table extension of
3759 pagetables) support.
3761 nopcid [X86-64] Disable the PCID cpu feature.
3763 norandmaps Don't use address space randomization. Equivalent to
3764 echo 0 > /proc/sys/kernel/randomize_va_space
3766 noreplace-smp [X86-32,SMP] Don't replace SMP instructions
3767 with UP alternatives
3769 noresume [SWSUSP] Disables resume and restores original swap
3772 no-scroll [VGA] Disables scrollback.
3773 This is required for the Braillex ib80-piezo Braille
3774 reader made by F.H. Papenmeier (Germany).
3778 nosgx [X86-64,SGX] Disables Intel SGX kernel support.
3780 nosmp [SMP] Tells an SMP kernel to act as a UP kernel,
3781 and disable the IO APIC. legacy for "maxcpus=0".
3783 nosoftlockup [KNL] Disable the soft-lockup detector.
3785 nosync [HW,M68K] Disables sync negotiation for all devices.
3787 nowatchdog [KNL] Disable both lockup detectors, i.e.
3788 soft-lockup and NMI watchdog (hard-lockup).
3792 nox2apic [X86-64,APIC] Do not enable x2APIC mode.
3794 nps_mtm_hs_ctr= [KNL,ARC]
3795 This parameter sets the maximum duration, in
3796 cycles, each HW thread of the CTOP can run
3797 without interruptions, before HW switches it.
3798 The actual maximum duration is 16 times this
3800 Format: integer between 1 and 255
3803 nptcg= [IA-64] Override max number of concurrent global TLB
3804 purges which is reported from either PAL_VM_SUMMARY or
3807 nr_cpus= [SMP] Maximum number of processors that an SMP kernel
3808 could support. nr_cpus=n : n >= 1 limits the kernel to
3809 support 'n' processors. It could be larger than the
3810 number of already plugged CPU during bootup, later in
3811 runtime you can physically add extra cpu until it reaches
3812 n. So during boot up some boot time memory for per-cpu
3813 variables need be pre-allocated for later physical cpu
3816 nr_uarts= [SERIAL] maximum number of UARTs to be registered.
3818 numa=off [KNL, ARM64, PPC, RISCV, SPARC, X86] Disable NUMA, Only
3819 set up a single NUMA node spanning all memory.
3821 numa_balancing= [KNL,ARM64,PPC,RISCV,S390,X86] Enable or disable automatic
3823 Allowed values are enable and disable
3825 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA.
3826 'node', 'default' can be specified
3827 This can be set from sysctl after boot.
3828 See Documentation/admin-guide/sysctl/vm.rst for details.
3830 ohci1394_dma=early [HW] enable debugging via the ohci1394 driver.
3831 See Documentation/core-api/debugging-via-ohci1394.rst for more
3834 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands
3835 Rather than timing out after 20 ms if an EC
3836 command is not properly ACKed, override the length
3837 of the timeout. We have interrupts disabled while
3838 waiting for the ACK, so if this is set too high
3839 interrupts *may* be lost!
3841 omap_mux= [OMAP] Override bootloader pin multiplexing.
3842 Format: <mux_mode0.mode_name=value>...
3843 For example, to override I2C bus2:
3844 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100
3846 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration
3848 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock]
3850 boundary - index of last SLC block on Flex-OneNAND.
3851 The remaining blocks are configured as MLC blocks.
3852 lock - Configure if Flex-OneNAND boundary should be locked.
3853 Once locked, the boundary cannot be changed.
3854 1 indicates lock status, 0 indicates unlock status.
3856 oops=panic Always panic on oopses. Default is to just kill the
3857 process, but there is a small probability of
3858 deadlocking the machine.
3859 This will also cause panics on machine check exceptions.
3860 Useful together with panic=30 to trigger a reboot.
3863 [KNL] Boolean flag to control whether the page allocator
3864 should randomize its free lists. The randomization may
3865 be automatically enabled if the kernel detects it is
3866 running on a platform with a direct-mapped memory-side
3867 cache, and this parameter can be used to
3868 override/disable that behavior. The state of the flag
3869 can be read from sysfs at:
3870 /sys/module/page_alloc/parameters/shuffle.
3872 page_owner= [KNL] Boot-time page_owner enabling option.
3873 Storage of the information about who allocated
3874 each page is disabled in default. With this switch,
3876 on: enable the feature
3878 page_poison= [KNL] Boot-time parameter changing the state of
3879 poisoning on the buddy allocator, available with
3880 CONFIG_PAGE_POISONING=y.
3881 off: turn off poisoning (default)
3882 on: turn on poisoning
3884 page_reporting.page_reporting_order=
3885 [KNL] Minimal page reporting order
3887 Adjust the minimal page reporting order. The page
3888 reporting is disabled when it exceeds (MAX_ORDER-1).
3890 panic= [KNL] Kernel behaviour on panic: delay <timeout>
3891 timeout > 0: seconds before rebooting
3892 timeout = 0: wait forever
3893 timeout < 0: reboot immediately
3896 panic_print= Bitmask for printing system info when panic happens.
3897 User can chose combination of the following bits:
3898 bit 0: print all tasks info
3899 bit 1: print system memory info
3900 bit 2: print timer info
3901 bit 3: print locks info if CONFIG_LOCKDEP is on
3902 bit 4: print ftrace buffer
3903 bit 5: print all printk messages in buffer
3904 bit 6: print all CPUs backtrace (if available in the arch)
3905 *Be aware* that this option may print a _lot_ of lines,
3906 so there are risks of losing older messages in the log.
3907 Use this option carefully, maybe worth to setup a
3908 bigger log buffer with "log_buf_len" along with this.
3910 panic_on_taint= Bitmask for conditionally calling panic() in add_taint()
3911 Format: <hex>[,nousertaint]
3912 Hexadecimal bitmask representing the set of TAINT flags
3913 that will cause the kernel to panic when add_taint() is
3914 called with any of the flags in this set.
3915 The optional switch "nousertaint" can be utilized to
3916 prevent userspace forced crashes by writing to sysctl
3917 /proc/sys/kernel/tainted any flagset matching with the
3918 bitmask set on panic_on_taint.
3919 See Documentation/admin-guide/tainted-kernels.rst for
3920 extra details on the taint flags that users can pick
3921 to compose the bitmask to assign to panic_on_taint.
3923 panic_on_warn panic() instead of WARN(). Useful to cause kdump
3926 parkbd.port= [HW] Parallel port number the keyboard adapter is
3927 connected to, default is 0.
3929 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation,
3930 0 for XT, 1 for AT (default is AT).
3933 parport= [HW,PPT] Specify parallel ports. 0 disables.
3934 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] }
3935 Use 'auto' to force the driver to use any
3936 IRQ/DMA settings detected (the default is to
3937 ignore detected IRQ/DMA settings because of
3938 possible conflicts). You can specify the base
3939 address, IRQ, and DMA settings; IRQ and DMA
3940 should be numbers, or 'auto' (for using detected
3941 settings on that particular port), or 'nofifo'
3942 (to avoid using a FIFO even if it is detected).
3943 Parallel ports are assigned in the order they
3944 are specified on the command line, starting
3947 parport_init_mode= [HW,PPT]
3948 Configure VIA parallel port to operate in
3949 a specific mode. This is necessary on Pegasos
3950 computer where firmware has no options for setting
3951 up parallel port mode and sets it to spp.
3952 Currently this function knows 686a and 8231 chips.
3953 Format: [spp|ps2|epp|ecp|ecpepp]
3955 pata_legacy.all= [HW,LIBATA]
3957 Set to non-zero to probe primary and secondary ISA
3958 port ranges on PCI systems where no PCI PATA device
3959 has been found at either range. Disabled by default.
3961 pata_legacy.autospeed= [HW,LIBATA]
3963 Set to non-zero if a chip is present that snoops speed
3964 changes. Disabled by default.
3966 pata_legacy.ht6560a= [HW,LIBATA]
3968 Set to 1, 2, or 3 for HT 6560A on the primary channel,
3969 the secondary channel, or both channels respectively.
3970 Disabled by default.
3972 pata_legacy.ht6560b= [HW,LIBATA]
3974 Set to 1, 2, or 3 for HT 6560B on the primary channel,
3975 the secondary channel, or both channels respectively.
3976 Disabled by default.
3978 pata_legacy.iordy_mask= [HW,LIBATA]
3980 IORDY enable mask. Set individual bits to allow IORDY
3981 for the respective channel. Bit 0 is for the first
3982 legacy channel handled by this driver, bit 1 is for
3983 the second channel, and so on. The sequence will often
3984 correspond to the primary legacy channel, the secondary
3985 legacy channel, and so on, but the handling of a PCI
3986 bus and the use of other driver options may interfere
3987 with the sequence. By default IORDY is allowed across
3990 pata_legacy.opti82c46x= [HW,LIBATA]
3992 Set to 1, 2, or 3 for Opti 82c611A on the primary
3993 channel, the secondary channel, or both channels
3994 respectively. Disabled by default.
3996 pata_legacy.opti82c611a= [HW,LIBATA]
3998 Set to 1, 2, or 3 for Opti 82c465MV on the primary
3999 channel, the secondary channel, or both channels
4000 respectively. Disabled by default.
4002 pata_legacy.pio_mask= [HW,LIBATA]
4004 PIO mode mask for autospeed devices. Set individual
4005 bits to allow the use of the respective PIO modes.
4006 Bit 0 is for mode 0, bit 1 is for mode 1, and so on.
4007 All modes allowed by default.
4009 pata_legacy.probe_all= [HW,LIBATA]
4011 Set to non-zero to probe tertiary and further ISA
4012 port ranges on PCI systems. Disabled by default.
4014 pata_legacy.probe_mask= [HW,LIBATA]
4016 Probe mask for legacy ISA PATA ports. Depending on
4017 platform configuration and the use of other driver
4018 options up to 6 legacy ports are supported: 0x1f0,
4019 0x170, 0x1e8, 0x168, 0x1e0, 0x160, however probing
4020 of individual ports can be disabled by setting the
4021 corresponding bits in the mask to 1. Bit 0 is for
4022 the first port in the list above (0x1f0), and so on.
4023 By default all supported ports are probed.
4025 pata_legacy.qdi= [HW,LIBATA]
4027 Set to non-zero to probe QDI controllers. By default
4028 set to 1 if CONFIG_PATA_QDI_MODULE, 0 otherwise.
4030 pata_legacy.winbond= [HW,LIBATA]
4032 Set to non-zero to probe Winbond controllers. Use
4033 the standard I/O port (0x130) if 1, otherwise the
4034 value given is the I/O port to use (typically 0x1b0).
4035 By default set to 1 if CONFIG_PATA_WINBOND_VLB_MODULE,
4038 pata_platform.pio_mask= [HW,LIBATA]
4040 Supported PIO mode mask. Set individual bits to allow
4041 the use of the respective PIO modes. Bit 0 is for
4042 mode 0, bit 1 is for mode 1, and so on. Mode 0 only
4046 Halt all CPUs after the first oops has been printed for
4047 the specified number of seconds. This is to be used if
4048 your oopses keep scrolling off the screen.
4053 See header of drivers/block/paride/pcd.c.
4054 See also Documentation/admin-guide/blockdev/paride.rst.
4056 pci=option[,option...] [PCI] various PCI subsystem options.
4058 Some options herein operate on a specific device
4059 or a set of devices (<pci_dev>). These are
4060 specified in one of the following formats:
4062 [<domain>:]<bus>:<dev>.<func>[/<dev>.<func>]*
4063 pci:<vendor>:<device>[:<subvendor>:<subdevice>]
4065 Note: the first format specifies a PCI
4066 bus/device/function address which may change
4067 if new hardware is inserted, if motherboard
4068 firmware changes, or due to changes caused
4069 by other kernel parameters. If the
4070 domain is left unspecified, it is
4071 taken to be zero. Optionally, a path
4072 to a device through multiple device/function
4073 addresses can be specified after the base
4074 address (this is more robust against
4075 renumbering issues). The second format
4076 selects devices using IDs from the
4077 configuration space which may match multiple
4078 devices in the system.
4080 earlydump dump PCI config space before the kernel
4082 off [X86] don't probe for the PCI bus
4083 bios [X86-32] force use of PCI BIOS, don't access
4084 the hardware directly. Use this if your machine
4085 has a non-standard PCI host bridge.
4086 nobios [X86-32] disallow use of PCI BIOS, only direct
4087 hardware access methods are allowed. Use this
4088 if you experience crashes upon bootup and you
4089 suspect they are caused by the BIOS.
4090 conf1 [X86] Force use of PCI Configuration Access
4091 Mechanism 1 (config address in IO port 0xCF8,
4092 data in IO port 0xCFC, both 32-bit).
4093 conf2 [X86] Force use of PCI Configuration Access
4094 Mechanism 2 (IO port 0xCF8 is an 8-bit port for
4095 the function, IO port 0xCFA, also 8-bit, sets
4096 bus number. The config space is then accessed
4097 through ports 0xC000-0xCFFF).
4098 See http://wiki.osdev.org/PCI for more info
4099 on the configuration access mechanisms.
4100 noaer [PCIE] If the PCIEAER kernel config parameter is
4101 enabled, this kernel boot option can be used to
4102 disable the use of PCIE advanced error reporting.
4103 nodomains [PCI] Disable support for multiple PCI
4104 root domains (aka PCI segments, in ACPI-speak).
4105 nommconf [X86] Disable use of MMCONFIG for PCI
4107 check_enable_amd_mmconf [X86] check for and enable
4108 properly configured MMIO access to PCI
4109 config space on AMD family 10h CPU
4110 nomsi [MSI] If the PCI_MSI kernel config parameter is
4111 enabled, this kernel boot option can be used to
4112 disable the use of MSI interrupts system-wide.
4113 noioapicquirk [APIC] Disable all boot interrupt quirks.
4114 Safety option to keep boot IRQs enabled. This
4115 should never be necessary.
4116 ioapicreroute [APIC] Enable rerouting of boot IRQs to the
4117 primary IO-APIC for bridges that cannot disable
4118 boot IRQs. This fixes a source of spurious IRQs
4119 when the system masks IRQs.
4120 noioapicreroute [APIC] Disable workaround that uses the
4121 boot IRQ equivalent of an IRQ that connects to
4122 a chipset where boot IRQs cannot be disabled.
4123 The opposite of ioapicreroute.
4124 biosirq [X86-32] Use PCI BIOS calls to get the interrupt
4125 routing table. These calls are known to be buggy
4126 on several machines and they hang the machine
4127 when used, but on other computers it's the only
4128 way to get the interrupt routing table. Try
4129 this option if the kernel is unable to allocate
4130 IRQs or discover secondary PCI buses on your
4132 rom [X86] Assign address space to expansion ROMs.
4133 Use with caution as certain devices share
4134 address decoders between ROMs and other
4136 norom [X86] Do not assign address space to
4137 expansion ROMs that do not already have
4138 BIOS assigned address ranges.
4139 nobar [X86] Do not assign address space to the
4140 BARs that weren't assigned by the BIOS.
4141 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be
4142 assigned automatically to PCI devices. You can
4143 make the kernel exclude IRQs of your ISA cards
4145 pirqaddr=0xAAAAA [X86] Specify the physical address
4146 of the PIRQ table (normally generated
4147 by the BIOS) if it is outside the
4148 F0000h-100000h range.
4149 lastbus=N [X86] Scan all buses thru bus #N. Can be
4150 useful if the kernel is unable to find your
4151 secondary buses and you want to tell it
4152 explicitly which ones they are.
4153 assign-busses [X86] Always assign all PCI bus
4154 numbers ourselves, overriding
4155 whatever the firmware may have done.
4156 usepirqmask [X86] Honor the possible IRQ mask stored
4157 in the BIOS $PIR table. This is needed on
4158 some systems with broken BIOSes, notably
4159 some HP Pavilion N5400 and Omnibook XE3
4160 notebooks. This will have no effect if ACPI
4161 IRQ routing is enabled.
4162 noacpi [X86] Do not use ACPI for IRQ routing
4163 or for PCI scanning.
4164 use_crs [X86] Use PCI host bridge window information
4165 from ACPI. On BIOSes from 2008 or later, this
4166 is enabled by default. If you need to use this,
4167 please report a bug.
4168 nocrs [X86] Ignore PCI host bridge windows from ACPI.
4169 If you need to use this, please report a bug.
4170 use_e820 [X86] Use E820 reservations to exclude parts of
4171 PCI host bridge windows. This is a workaround
4172 for BIOS defects in host bridge _CRS methods.
4173 If you need to use this, please report a bug to
4174 <linux-pci@vger.kernel.org>.
4175 no_e820 [X86] Ignore E820 reservations for PCI host
4176 bridge windows. This is the default on modern
4177 hardware. If you need to use this, please report
4178 a bug to <linux-pci@vger.kernel.org>.
4179 routeirq Do IRQ routing for all PCI devices.
4180 This is normally done in pci_enable_device(),
4181 so this option is a temporary workaround
4182 for broken drivers that don't call it.
4183 skip_isa_align [X86] do not align io start addr, so can
4184 handle more pci cards
4185 noearly [X86] Don't do any early type 1 scanning.
4186 This might help on some broken boards which
4187 machine check when some devices' config space
4188 is read. But various workarounds are disabled
4189 and some IOMMU drivers will not work.
4190 bfsort Sort PCI devices into breadth-first order.
4191 This sorting is done to get a device
4192 order compatible with older (<= 2.4) kernels.
4193 nobfsort Don't sort PCI devices into breadth-first order.
4194 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size)
4195 tuning and use the BIOS-configured MPS defaults.
4196 pcie_bus_safe Set every device's MPS to the largest value
4197 supported by all devices below the root complex.
4198 pcie_bus_perf Set device MPS to the largest allowable MPS
4199 based on its parent bus. Also set MRRS (Max
4200 Read Request Size) to the largest supported
4201 value (no larger than the MPS that the device
4202 or bus can support) for best performance.
4203 pcie_bus_peer2peer Set every device's MPS to 128B, which
4204 every device is guaranteed to support. This
4205 configuration allows peer-to-peer DMA between
4206 any pair of devices, possibly at the cost of
4207 reduced performance. This also guarantees
4208 that hot-added devices will work.
4209 cbiosize=nn[KMG] The fixed amount of bus space which is
4210 reserved for the CardBus bridge's IO window.
4211 The default value is 256 bytes.
4212 cbmemsize=nn[KMG] The fixed amount of bus space which is
4213 reserved for the CardBus bridge's memory
4214 window. The default value is 64 megabytes.
4217 [<order of align>@]<pci_dev>[; ...]
4218 Specifies alignment and device to reassign
4219 aligned memory resources. How to
4220 specify the device is described above.
4221 If <order of align> is not specified,
4222 PAGE_SIZE is used as alignment.
4223 A PCI-PCI bridge can be specified if resource
4224 windows need to be expanded.
4225 To specify the alignment for several
4226 instances of a device, the PCI vendor,
4227 device, subvendor, and subdevice may be
4228 specified, e.g., 12@pci:8086:9c22:103c:198f
4229 for 4096-byte alignment.
4230 ecrc= Enable/disable PCIe ECRC (transaction layer
4231 end-to-end CRC checking).
4232 bios: Use BIOS/firmware settings. This is the
4236 hpiosize=nn[KMG] The fixed amount of bus space which is
4237 reserved for hotplug bridge's IO window.
4238 Default size is 256 bytes.
4239 hpmmiosize=nn[KMG] The fixed amount of bus space which is
4240 reserved for hotplug bridge's MMIO window.
4241 Default size is 2 megabytes.
4242 hpmmioprefsize=nn[KMG] The fixed amount of bus space which is
4243 reserved for hotplug bridge's MMIO_PREF window.
4244 Default size is 2 megabytes.
4245 hpmemsize=nn[KMG] The fixed amount of bus space which is
4246 reserved for hotplug bridge's MMIO and
4248 Default size is 2 megabytes.
4249 hpbussize=nn The minimum amount of additional bus numbers
4250 reserved for buses below a hotplug bridge.
4252 realloc= Enable/disable reallocating PCI bridge resources
4253 if allocations done by BIOS are too small to
4254 accommodate resources required by all child
4256 off: Turn realloc off
4258 realloc same as realloc=on
4259 noari do not use PCIe ARI.
4260 noats [PCIE, Intel-IOMMU, AMD-IOMMU]
4261 do not use PCIe ATS (and IOMMU device IOTLB).
4262 pcie_scan_all Scan all possible PCIe devices. Otherwise we
4263 only look for one device below a PCIe downstream
4265 big_root_window Try to add a big 64bit memory window to the PCIe
4266 root complex on AMD CPUs. Some GFX hardware
4267 can resize a BAR to allow access to all VRAM.
4268 Adding the window is slightly risky (it may
4269 conflict with unreported devices), so this
4271 disable_acs_redir=<pci_dev>[; ...]
4272 Specify one or more PCI devices (in the format
4273 specified above) separated by semicolons.
4274 Each device specified will have the PCI ACS
4275 redirect capabilities forced off which will
4276 allow P2P traffic between devices through
4277 bridges without forcing it upstream. Note:
4278 this removes isolation between devices and
4279 may put more devices in an IOMMU group.
4280 force_floating [S390] Force usage of floating interrupts.
4281 nomio [S390] Do not use MIO instructions.
4282 norid [S390] ignore the RID field and force use of
4283 one PCI domain per PCI function
4285 pcie_aspm= [PCIE] Forcibly enable or disable PCIe Active State Power
4288 force Enable ASPM even on devices that claim not to support it.
4289 WARNING: Forcing ASPM on may cause system lockups.
4291 pcie_ports= [PCIE] PCIe port services handling:
4292 native Use native PCIe services (PME, AER, DPC, PCIe hotplug)
4293 even if the platform doesn't give the OS permission to
4294 use them. This may cause conflicts if the platform
4295 also tries to use these services.
4296 dpc-native Use native PCIe service for DPC only. May
4297 cause conflicts if firmware uses AER or DPC.
4298 compat Disable native PCIe services (PME, AER, DPC, PCIe
4301 pcie_port_pm= [PCIE] PCIe port power management handling:
4302 off Disable power management of all PCIe ports
4303 force Forcibly enable power management of all PCIe ports
4305 pcie_pme= [PCIE,PM] Native PCIe PME signaling options:
4306 nomsi Do not use MSI for native PCIe PME signaling (this makes
4307 all PCIe root ports use INTx for all services).
4309 pcmv= [HW,PCMCIA] BadgePAD 4
4313 Keep all power-domains already enabled by bootloader on,
4314 even if no driver has claimed them. This is useful
4315 for debug and development, but should not be
4316 needed on a platform with proper driver support.
4319 See Documentation/admin-guide/blockdev/paride.rst.
4321 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at
4324 See arch/parisc/kernel/pdc_chassis.c
4326 percpu_alloc= Select which percpu first chunk allocator to use.
4327 Currently supported values are "embed" and "page".
4328 Archs may support subset or none of the selections.
4329 See comments in mm/percpu.c for details on each
4330 allocator. This parameter is primarily for debugging
4331 and performance comparison.
4334 See Documentation/admin-guide/blockdev/paride.rst.
4337 See Documentation/admin-guide/blockdev/paride.rst.
4339 pirq= [SMP,APIC] Manual mp-table setup
4340 See Documentation/x86/i386/IO-APIC.rst.
4342 plip= [PPT,NET] Parallel port network link
4343 Format: { parport<nr> | timid | 0 }
4344 See also Documentation/admin-guide/parport.rst.
4346 pmtmr= [X86] Manual setup of pmtmr I/O Port.
4347 Override pmtimer IOPort with a hex value.
4350 pmu_override= [PPC] Override the PMU.
4351 This option takes over the PMU facility, so it is no
4352 longer usable by perf. Setting this option starts the
4353 PMU counters by setting MMCR0 to 0 (the FC bit is
4354 cleared). If a number is given, then MMCR1 is set to
4355 that number, otherwise (e.g., 'pmu_override=on'), MMCR1
4358 pm_debug_messages [SUSPEND,KNL]
4359 Enable suspend/resume debug messages during boot up.
4362 Enable PNP debug messages (depends on the
4363 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time
4364 via /sys/module/pnp/parameters/debug. We always show
4365 current resource usage; turning this on also shows
4366 possible settings and some assignment information.
4372 { on | off | curr | res | no-curr | no-res }
4375 [ISAPNP] Exclude IRQs for the autoconfiguration
4378 [ISAPNP] Exclude DMAs for the autoconfiguration
4380 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration
4381 Ranges are in pairs (I/O port base and size).
4384 [ISAPNP] Exclude memory regions for the
4386 Ranges are in pairs (memory base and size).
4388 ports= [IP_VS_FTP] IPVS ftp helper module
4390 Up to 8 (IP_VS_APP_MAX_PORTS) ports
4392 Format: <port>,<port>....
4394 powersave=off [PPC] This option disables power saving features.
4395 It specifically disables cpuidle and sets the
4396 platform machine description specific power_save
4397 function to NULL. On Idle the CPU just reduces
4400 ppc_strict_facility_enable
4401 [PPC] This option catches any kernel floating point,
4402 Altivec, VSX and SPE outside of regions specifically
4403 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()).
4404 There is some performance impact when enabling this.
4408 Disable Hardware Transactional Memory
4411 Select preemption mode if you have CONFIG_PREEMPT_DYNAMIC
4412 none - Limited to cond_resched() calls
4413 voluntary - Limited to cond_resched() and might_sleep() calls
4414 full - Any section that isn't explicitly preempt disabled
4415 can be preempted anytime.
4417 print-fatal-signals=
4418 [KNL] debug: print fatal signals
4420 If enabled, warn about various signal handling
4421 related application anomalies: too many signals,
4422 too many POSIX.1 timers, fatal signals causing a
4425 If you hit the warning due to signal overflow,
4426 you might want to try "ulimit -i unlimited".
4430 printk.always_kmsg_dump=
4431 Trigger kmsg_dump for cases other than kernel oops or
4433 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
4436 printk.console_no_auto_verbose=
4437 Disable console loglevel raise on oops, panic
4438 or lockdep-detected issues (only if lock debug is on).
4439 With an exception to setups with low baudrate on
4440 serial console, keeping this 0 is a good choice
4441 in order to provide more debug information.
4443 default: 0 (auto_verbose is enabled)
4445 printk.devkmsg={on,off,ratelimit}
4446 Control writing to /dev/kmsg.
4447 on - unlimited logging to /dev/kmsg from userspace
4448 off - logging to /dev/kmsg disabled
4449 ratelimit - ratelimit the logging
4452 printk.time= Show timing data prefixed to each printk message line
4453 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
4455 processor.max_cstate= [HW,ACPI]
4456 Limit processor to maximum C-state
4457 max_cstate=9 overrides any DMI blacklist limit.
4459 processor.nocst [HW,ACPI]
4460 Ignore the _CST method to determine C-states,
4461 instead using the legacy FADT method
4463 profile= [KNL] Enable kernel profiling via /proc/profile
4464 Format: [<profiletype>,]<number>
4465 Param: <profiletype>: "schedule", "sleep", or "kvm"
4466 [defaults to kernel profiling]
4467 Param: "schedule" - profile schedule points.
4468 Param: "sleep" - profile D-state sleeping (millisecs).
4469 Requires CONFIG_SCHEDSTATS
4470 Param: "kvm" - profile VM exits.
4471 Param: <number> - step/bucket size as a power of 2 for
4472 statistical time based profiling.
4474 prompt_ramdisk= [RAM] [Deprecated]
4476 prot_virt= [S390] enable hosting protected virtual machines
4477 isolated from the hypervisor (if hardware supports
4481 psi= [KNL] Enable or disable pressure stall information
4485 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to
4486 probe for; one of (bare|imps|exps|lifebook|any).
4487 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports
4489 psmouse.resetafter= [HW,MOUSE]
4490 Try to reset the device after so many bad packets
4493 [HW,MOUSE] Set desired mouse resolution, in dpi.
4494 psmouse.smartscroll=
4495 [HW,MOUSE] Controls Logitech smartscroll autorepeat.
4496 0 = disabled, 1 = enabled (default).
4498 pstore.backend= Specify the name of the pstore backend to use
4501 See Documentation/admin-guide/blockdev/paride.rst.
4503 pti= [X86-64] Control Page Table Isolation of user and
4504 kernel address spaces. Disabling this feature
4505 removes hardening, but improves performance of
4506 system calls and interrupts.
4508 on - unconditionally enable
4509 off - unconditionally disable
4510 auto - kernel detects whether your CPU model is
4511 vulnerable to issues that PTI mitigates
4513 Not specifying this option is equivalent to pti=auto.
4516 Equivalent to pti=off
4519 [KNL] Number of legacy pty's. Overwrites compiled-in
4522 quiet [KNL] Disable most log messages
4527 See Documentation/admin-guide/md.rst.
4529 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes
4530 See Documentation/admin-guide/blockdev/ramdisk.rst.
4532 ramdisk_start= [RAM] RAM disk image start address
4534 random.trust_cpu={on,off}
4535 [KNL] Enable or disable trusting the use of the
4536 CPU's random number generator (if available) to
4537 fully seed the kernel's CRNG. Default is controlled
4538 by CONFIG_RANDOM_TRUST_CPU.
4540 random.trust_bootloader={on,off}
4541 [KNL] Enable or disable trusting the use of a
4542 seed passed by the bootloader (if available) to
4543 fully seed the kernel's CRNG. Default is controlled
4544 by CONFIG_RANDOM_TRUST_BOOTLOADER.
4546 randomize_kstack_offset=
4547 [KNL] Enable or disable kernel stack offset
4548 randomization, which provides roughly 5 bits of
4549 entropy, frustrating memory corruption attacks
4550 that depend on stack address determinism or
4551 cross-syscall address exposures. This is only
4552 available on architectures that have defined
4553 CONFIG_HAVE_ARCH_RANDOMIZE_KSTACK_OFFSET.
4554 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
4555 Default is CONFIG_RANDOMIZE_KSTACK_OFFSET_DEFAULT.
4557 ras=option[,option,...] [KNL] RAS-specific options
4560 Disable the Correctable Errors Collector,
4561 see CONFIG_RAS_CEC help text.
4563 rcu_nocbs[=cpu-list]
4564 [KNL] The optional argument is a cpu list,
4567 In kernels built with CONFIG_RCU_NOCB_CPU=y,
4568 enable the no-callback CPU mode, which prevents
4569 such CPUs' callbacks from being invoked in
4570 softirq context. Invocation of such CPUs' RCU
4571 callbacks will instead be offloaded to "rcuox/N"
4572 kthreads created for that purpose, where "x" is
4573 "p" for RCU-preempt, "s" for RCU-sched, and "g"
4574 for the kthreads that mediate grace periods; and
4575 "N" is the CPU number. This reduces OS jitter on
4576 the offloaded CPUs, which can be useful for HPC
4577 and real-time workloads. It can also improve
4578 energy efficiency for asymmetric multiprocessors.
4580 If a cpulist is passed as an argument, the specified
4581 list of CPUs is set to no-callback mode from boot.
4583 Otherwise, if the '=' sign and the cpulist
4584 arguments are omitted, no CPU will be set to
4585 no-callback mode from boot but the mode may be
4586 toggled at runtime via cpusets.
4588 Note that this argument takes precedence over
4589 the CONFIG_RCU_NOCB_CPU_DEFAULT_ALL option.
4592 Rather than requiring that offloaded CPUs
4593 (specified by rcu_nocbs= above) explicitly
4594 awaken the corresponding "rcuoN" kthreads,
4595 make these kthreads poll for callbacks.
4596 This improves the real-time response for the
4597 offloaded CPUs by relieving them of the need to
4598 wake up the corresponding kthread, but degrades
4599 energy efficiency by requiring that the kthreads
4600 periodically wake up to do the polling.
4602 rcutree.blimit= [KNL]
4603 Set maximum number of finished RCU callbacks to
4604 process in one batch.
4606 rcutree.dump_tree= [KNL]
4607 Dump the structure of the rcu_node combining tree
4608 out at early boot. This is used for diagnostic
4609 purposes, to verify correct tree setup.
4611 rcutree.gp_cleanup_delay= [KNL]
4612 Set the number of jiffies to delay each step of
4613 RCU grace-period cleanup.
4615 rcutree.gp_init_delay= [KNL]
4616 Set the number of jiffies to delay each step of
4617 RCU grace-period initialization.
4619 rcutree.gp_preinit_delay= [KNL]
4620 Set the number of jiffies to delay each step of
4621 RCU grace-period pre-initialization, that is,
4622 the propagation of recent CPU-hotplug changes up
4623 the rcu_node combining tree.
4625 rcutree.use_softirq= [KNL]
4626 If set to zero, move all RCU_SOFTIRQ processing to
4627 per-CPU rcuc kthreads. Defaults to a non-zero
4628 value, meaning that RCU_SOFTIRQ is used by default.
4629 Specify rcutree.use_softirq=0 to use rcuc kthreads.
4631 But note that CONFIG_PREEMPT_RT=y kernels disable
4632 this kernel boot parameter, forcibly setting it
4635 rcutree.rcu_fanout_exact= [KNL]
4636 Disable autobalancing of the rcu_node combining
4637 tree. This is used by rcutorture, and might
4638 possibly be useful for architectures having high
4639 cache-to-cache transfer latencies.
4641 rcutree.rcu_fanout_leaf= [KNL]
4642 Change the number of CPUs assigned to each
4643 leaf rcu_node structure. Useful for very
4644 large systems, which will choose the value 64,
4645 and for NUMA systems with large remote-access
4646 latencies, which will choose a value aligned
4647 with the appropriate hardware boundaries.
4649 rcutree.rcu_min_cached_objs= [KNL]
4650 Minimum number of objects which are cached and
4651 maintained per one CPU. Object size is equal
4652 to PAGE_SIZE. The cache allows to reduce the
4653 pressure to page allocator, also it makes the
4654 whole algorithm to behave better in low memory
4657 rcutree.rcu_delay_page_cache_fill_msec= [KNL]
4658 Set the page-cache refill delay (in milliseconds)
4659 in response to low-memory conditions. The range
4660 of permitted values is in the range 0:100000.
4662 rcutree.jiffies_till_first_fqs= [KNL]
4663 Set delay from grace-period initialization to
4664 first attempt to force quiescent states.
4665 Units are jiffies, minimum value is zero,
4666 and maximum value is HZ.
4668 rcutree.jiffies_till_next_fqs= [KNL]
4669 Set delay between subsequent attempts to force
4670 quiescent states. Units are jiffies, minimum
4671 value is one, and maximum value is HZ.
4673 rcutree.jiffies_till_sched_qs= [KNL]
4674 Set required age in jiffies for a
4675 given grace period before RCU starts
4676 soliciting quiescent-state help from
4677 rcu_note_context_switch() and cond_resched().
4678 If not specified, the kernel will calculate
4679 a value based on the most recent settings
4680 of rcutree.jiffies_till_first_fqs
4681 and rcutree.jiffies_till_next_fqs.
4682 This calculated value may be viewed in
4683 rcutree.jiffies_to_sched_qs. Any attempt to set
4684 rcutree.jiffies_to_sched_qs will be cheerfully
4687 rcutree.kthread_prio= [KNL,BOOT]
4688 Set the SCHED_FIFO priority of the RCU per-CPU
4689 kthreads (rcuc/N). This value is also used for
4690 the priority of the RCU boost threads (rcub/N)
4691 and for the RCU grace-period kthreads (rcu_bh,
4692 rcu_preempt, and rcu_sched). If RCU_BOOST is
4693 set, valid values are 1-99 and the default is 1
4694 (the least-favored priority). Otherwise, when
4695 RCU_BOOST is not set, valid values are 0-99 and
4696 the default is zero (non-realtime operation).
4697 When RCU_NOCB_CPU is set, also adjust the
4698 priority of NOCB callback kthreads.
4700 rcutree.rcu_divisor= [KNL]
4701 Set the shift-right count to use to compute
4702 the callback-invocation batch limit bl from
4703 the number of callbacks queued on this CPU.
4704 The result will be bounded below by the value of
4705 the rcutree.blimit kernel parameter. Every bl
4706 callbacks, the softirq handler will exit in
4707 order to allow the CPU to do other work.
4709 Please note that this callback-invocation batch
4710 limit applies only to non-offloaded callback
4711 invocation. Offloaded callbacks are instead
4712 invoked in the context of an rcuoc kthread, which
4713 scheduler will preempt as it does any other task.
4715 rcutree.nocb_nobypass_lim_per_jiffy= [KNL]
4716 On callback-offloaded (rcu_nocbs) CPUs,
4717 RCU reduces the lock contention that would
4718 otherwise be caused by callback floods through
4719 use of the ->nocb_bypass list. However, in the
4720 common non-flooded case, RCU queues directly to
4721 the main ->cblist in order to avoid the extra
4722 overhead of the ->nocb_bypass list and its lock.
4723 But if there are too many callbacks queued during
4724 a single jiffy, RCU pre-queues the callbacks into
4725 the ->nocb_bypass queue. The definition of "too
4726 many" is supplied by this kernel boot parameter.
4728 rcutree.rcu_nocb_gp_stride= [KNL]
4729 Set the number of NOCB callback kthreads in
4730 each group, which defaults to the square root
4731 of the number of CPUs. Larger numbers reduce
4732 the wakeup overhead on the global grace-period
4733 kthread, but increases that same overhead on
4734 each group's NOCB grace-period kthread.
4736 rcutree.qhimark= [KNL]
4737 Set threshold of queued RCU callbacks beyond which
4738 batch limiting is disabled.
4740 rcutree.qlowmark= [KNL]
4741 Set threshold of queued RCU callbacks below which
4742 batch limiting is re-enabled.
4744 rcutree.qovld= [KNL]
4745 Set threshold of queued RCU callbacks beyond which
4746 RCU's force-quiescent-state scan will aggressively
4747 enlist help from cond_resched() and sched IPIs to
4748 help CPUs more quickly reach quiescent states.
4749 Set to less than zero to make this be set based
4750 on rcutree.qhimark at boot time and to zero to
4751 disable more aggressive help enlistment.
4753 rcutree.rcu_kick_kthreads= [KNL]
4754 Cause the grace-period kthread to get an extra
4755 wake_up() if it sleeps three times longer than
4756 it should at force-quiescent-state time.
4757 This wake_up() will be accompanied by a
4758 WARN_ONCE() splat and an ftrace_dump().
4760 rcutree.rcu_unlock_delay= [KNL]
4761 In CONFIG_RCU_STRICT_GRACE_PERIOD=y kernels,
4762 this specifies an rcu_read_unlock()-time delay
4763 in microseconds. This defaults to zero.
4764 Larger delays increase the probability of
4765 catching RCU pointer leaks, that is, buggy use
4766 of RCU-protected pointers after the relevant
4767 rcu_read_unlock() has completed.
4769 rcutree.sysrq_rcu= [KNL]
4770 Commandeer a sysrq key to dump out Tree RCU's
4771 rcu_node tree with an eye towards determining
4772 why a new grace period has not yet started.
4774 rcuscale.gp_async= [KNL]
4775 Measure performance of asynchronous
4776 grace-period primitives such as call_rcu().
4778 rcuscale.gp_async_max= [KNL]
4779 Specify the maximum number of outstanding
4780 callbacks per writer thread. When a writer
4781 thread exceeds this limit, it invokes the
4782 corresponding flavor of rcu_barrier() to allow
4783 previously posted callbacks to drain.
4785 rcuscale.gp_exp= [KNL]
4786 Measure performance of expedited synchronous
4787 grace-period primitives.
4789 rcuscale.holdoff= [KNL]
4790 Set test-start holdoff period. The purpose of
4791 this parameter is to delay the start of the
4792 test until boot completes in order to avoid
4795 rcuscale.kfree_rcu_test= [KNL]
4796 Set to measure performance of kfree_rcu() flooding.
4798 rcuscale.kfree_rcu_test_double= [KNL]
4799 Test the double-argument variant of kfree_rcu().
4800 If this parameter has the same value as
4801 rcuscale.kfree_rcu_test_single, both the single-
4802 and double-argument variants are tested.
4804 rcuscale.kfree_rcu_test_single= [KNL]
4805 Test the single-argument variant of kfree_rcu().
4806 If this parameter has the same value as
4807 rcuscale.kfree_rcu_test_double, both the single-
4808 and double-argument variants are tested.
4810 rcuscale.kfree_nthreads= [KNL]
4811 The number of threads running loops of kfree_rcu().
4813 rcuscale.kfree_alloc_num= [KNL]
4814 Number of allocations and frees done in an iteration.
4816 rcuscale.kfree_loops= [KNL]
4817 Number of loops doing rcuscale.kfree_alloc_num number
4818 of allocations and frees.
4820 rcuscale.nreaders= [KNL]
4821 Set number of RCU readers. The value -1 selects
4822 N, where N is the number of CPUs. A value
4823 "n" less than -1 selects N-n+1, where N is again
4824 the number of CPUs. For example, -2 selects N
4825 (the number of CPUs), -3 selects N+1, and so on.
4826 A value of "n" less than or equal to -N selects
4829 rcuscale.nwriters= [KNL]
4830 Set number of RCU writers. The values operate
4831 the same as for rcuscale.nreaders.
4832 N, where N is the number of CPUs
4834 rcuscale.perf_type= [KNL]
4835 Specify the RCU implementation to test.
4837 rcuscale.shutdown= [KNL]
4838 Shut the system down after performance tests
4839 complete. This is useful for hands-off automated
4842 rcuscale.verbose= [KNL]
4843 Enable additional printk() statements.
4845 rcuscale.writer_holdoff= [KNL]
4846 Write-side holdoff between grace periods,
4847 in microseconds. The default of zero says
4850 rcutorture.fqs_duration= [KNL]
4851 Set duration of force_quiescent_state bursts
4854 rcutorture.fqs_holdoff= [KNL]
4855 Set holdoff time within force_quiescent_state bursts
4858 rcutorture.fqs_stutter= [KNL]
4859 Set wait time between force_quiescent_state bursts
4862 rcutorture.fwd_progress= [KNL]
4863 Specifies the number of kthreads to be used
4864 for RCU grace-period forward-progress testing
4865 for the types of RCU supporting this notion.
4866 Defaults to 1 kthread, values less than zero or
4867 greater than the number of CPUs cause the number
4870 rcutorture.fwd_progress_div= [KNL]
4871 Specify the fraction of a CPU-stall-warning
4872 period to do tight-loop forward-progress testing.
4874 rcutorture.fwd_progress_holdoff= [KNL]
4875 Number of seconds to wait between successive
4876 forward-progress tests.
4878 rcutorture.fwd_progress_need_resched= [KNL]
4879 Enclose cond_resched() calls within checks for
4880 need_resched() during tight-loop forward-progress
4883 rcutorture.gp_cond= [KNL]
4884 Use conditional/asynchronous update-side
4885 primitives, if available.
4887 rcutorture.gp_exp= [KNL]
4888 Use expedited update-side primitives, if available.
4890 rcutorture.gp_normal= [KNL]
4891 Use normal (non-expedited) asynchronous
4892 update-side primitives, if available.
4894 rcutorture.gp_sync= [KNL]
4895 Use normal (non-expedited) synchronous
4896 update-side primitives, if available. If all
4897 of rcutorture.gp_cond=, rcutorture.gp_exp=,
4898 rcutorture.gp_normal=, and rcutorture.gp_sync=
4899 are zero, rcutorture acts as if is interpreted
4900 they are all non-zero.
4902 rcutorture.irqreader= [KNL]
4903 Run RCU readers from irq handlers, or, more
4904 accurately, from a timer handler. Not all RCU
4905 flavors take kindly to this sort of thing.
4907 rcutorture.leakpointer= [KNL]
4908 Leak an RCU-protected pointer out of the reader.
4909 This can of course result in splats, and is
4910 intended to test the ability of things like
4911 CONFIG_RCU_STRICT_GRACE_PERIOD=y to detect
4914 rcutorture.n_barrier_cbs= [KNL]
4915 Set callbacks/threads for rcu_barrier() testing.
4917 rcutorture.nfakewriters= [KNL]
4918 Set number of concurrent RCU writers. These just
4919 stress RCU, they don't participate in the actual
4920 test, hence the "fake".
4922 rcutorture.nocbs_nthreads= [KNL]
4923 Set number of RCU callback-offload togglers.
4924 Zero (the default) disables toggling.
4926 rcutorture.nocbs_toggle= [KNL]
4927 Set the delay in milliseconds between successive
4928 callback-offload toggling attempts.
4930 rcutorture.nreaders= [KNL]
4931 Set number of RCU readers. The value -1 selects
4932 N-1, where N is the number of CPUs. A value
4933 "n" less than -1 selects N-n-2, where N is again
4934 the number of CPUs. For example, -2 selects N
4935 (the number of CPUs), -3 selects N+1, and so on.
4937 rcutorture.object_debug= [KNL]
4938 Enable debug-object double-call_rcu() testing.
4940 rcutorture.onoff_holdoff= [KNL]
4941 Set time (s) after boot for CPU-hotplug testing.
4943 rcutorture.onoff_interval= [KNL]
4944 Set time (jiffies) between CPU-hotplug operations,
4945 or zero to disable CPU-hotplug testing.
4947 rcutorture.read_exit= [KNL]
4948 Set the number of read-then-exit kthreads used
4949 to test the interaction of RCU updaters and
4950 task-exit processing.
4952 rcutorture.read_exit_burst= [KNL]
4953 The number of times in a given read-then-exit
4954 episode that a set of read-then-exit kthreads
4957 rcutorture.read_exit_delay= [KNL]
4958 The delay, in seconds, between successive
4959 read-then-exit testing episodes.
4961 rcutorture.shuffle_interval= [KNL]
4962 Set task-shuffle interval (s). Shuffling tasks
4963 allows some CPUs to go into dyntick-idle mode
4964 during the rcutorture test.
4966 rcutorture.shutdown_secs= [KNL]
4967 Set time (s) after boot system shutdown. This
4968 is useful for hands-off automated testing.
4970 rcutorture.stall_cpu= [KNL]
4971 Duration of CPU stall (s) to test RCU CPU stall
4972 warnings, zero to disable.
4974 rcutorture.stall_cpu_block= [KNL]
4975 Sleep while stalling if set. This will result
4976 in warnings from preemptible RCU in addition
4977 to any other stall-related activity.
4979 rcutorture.stall_cpu_holdoff= [KNL]
4980 Time to wait (s) after boot before inducing stall.
4982 rcutorture.stall_cpu_irqsoff= [KNL]
4983 Disable interrupts while stalling if set.
4985 rcutorture.stall_gp_kthread= [KNL]
4986 Duration (s) of forced sleep within RCU
4987 grace-period kthread to test RCU CPU stall
4988 warnings, zero to disable. If both stall_cpu
4989 and stall_gp_kthread are specified, the
4990 kthread is starved first, then the CPU.
4992 rcutorture.stat_interval= [KNL]
4993 Time (s) between statistics printk()s.
4995 rcutorture.stutter= [KNL]
4996 Time (s) to stutter testing, for example, specifying
4997 five seconds causes the test to run for five seconds,
4998 wait for five seconds, and so on. This tests RCU's
4999 ability to transition abruptly to and from idle.
5001 rcutorture.test_boost= [KNL]
5002 Test RCU priority boosting? 0=no, 1=maybe, 2=yes.
5003 "Maybe" means test if the RCU implementation
5004 under test support RCU priority boosting.
5006 rcutorture.test_boost_duration= [KNL]
5007 Duration (s) of each individual boost test.
5009 rcutorture.test_boost_interval= [KNL]
5010 Interval (s) between each boost test.
5012 rcutorture.test_no_idle_hz= [KNL]
5013 Test RCU's dyntick-idle handling. See also the
5014 rcutorture.shuffle_interval parameter.
5016 rcutorture.torture_type= [KNL]
5017 Specify the RCU implementation to test.
5019 rcutorture.verbose= [KNL]
5020 Enable additional printk() statements.
5022 rcupdate.rcu_cpu_stall_ftrace_dump= [KNL]
5023 Dump ftrace buffer after reporting RCU CPU
5026 rcupdate.rcu_cpu_stall_suppress= [KNL]
5027 Suppress RCU CPU stall warning messages.
5029 rcupdate.rcu_cpu_stall_suppress_at_boot= [KNL]
5030 Suppress RCU CPU stall warning messages and
5031 rcutorture writer stall warnings that occur
5032 during early boot, that is, during the time
5033 before the init task is spawned.
5035 rcupdate.rcu_cpu_stall_timeout= [KNL]
5036 Set timeout for RCU CPU stall warning messages.
5037 The value is in seconds and the maximum allowed
5038 value is 300 seconds.
5040 rcupdate.rcu_exp_cpu_stall_timeout= [KNL]
5041 Set timeout for expedited RCU CPU stall warning
5042 messages. The value is in milliseconds
5043 and the maximum allowed value is 21000
5044 milliseconds. Please note that this value is
5045 adjusted to an arch timer tick resolution.
5046 Setting this to zero causes the value from
5047 rcupdate.rcu_cpu_stall_timeout to be used (after
5048 conversion from seconds to milliseconds).
5050 rcupdate.rcu_expedited= [KNL]
5051 Use expedited grace-period primitives, for
5052 example, synchronize_rcu_expedited() instead
5053 of synchronize_rcu(). This reduces latency,
5054 but can increase CPU utilization, degrade
5055 real-time latency, and degrade energy efficiency.
5056 No effect on CONFIG_TINY_RCU kernels.
5058 rcupdate.rcu_normal= [KNL]
5059 Use only normal grace-period primitives,
5060 for example, synchronize_rcu() instead of
5061 synchronize_rcu_expedited(). This improves
5062 real-time latency, CPU utilization, and
5063 energy efficiency, but can expose users to
5064 increased grace-period latency. This parameter
5065 overrides rcupdate.rcu_expedited. No effect on
5066 CONFIG_TINY_RCU kernels.
5068 rcupdate.rcu_normal_after_boot= [KNL]
5069 Once boot has completed (that is, after
5070 rcu_end_inkernel_boot() has been invoked), use
5071 only normal grace-period primitives. No effect
5072 on CONFIG_TINY_RCU kernels.
5074 But note that CONFIG_PREEMPT_RT=y kernels enables
5075 this kernel boot parameter, forcibly setting
5076 it to the value one, that is, converting any
5077 post-boot attempt at an expedited RCU grace
5078 period to instead use normal non-expedited
5079 grace-period processing.
5081 rcupdate.rcu_task_collapse_lim= [KNL]
5082 Set the maximum number of callbacks present
5083 at the beginning of a grace period that allows
5084 the RCU Tasks flavors to collapse back to using
5085 a single callback queue. This switching only
5086 occurs when rcupdate.rcu_task_enqueue_lim is
5087 set to the default value of -1.
5089 rcupdate.rcu_task_contend_lim= [KNL]
5090 Set the minimum number of callback-queuing-time
5091 lock-contention events per jiffy required to
5092 cause the RCU Tasks flavors to switch to per-CPU
5093 callback queuing. This switching only occurs
5094 when rcupdate.rcu_task_enqueue_lim is set to
5095 the default value of -1.
5097 rcupdate.rcu_task_enqueue_lim= [KNL]
5098 Set the number of callback queues to use for the
5099 RCU Tasks family of RCU flavors. The default
5100 of -1 allows this to be automatically (and
5101 dynamically) adjusted. This parameter is intended
5104 rcupdate.rcu_task_ipi_delay= [KNL]
5105 Set time in jiffies during which RCU tasks will
5106 avoid sending IPIs, starting with the beginning
5107 of a given grace period. Setting a large
5108 number avoids disturbing real-time workloads,
5109 but lengthens grace periods.
5111 rcupdate.rcu_task_stall_info= [KNL]
5112 Set initial timeout in jiffies for RCU task stall
5113 informational messages, which give some indication
5114 of the problem for those not patient enough to
5115 wait for ten minutes. Informational messages are
5116 only printed prior to the stall-warning message
5117 for a given grace period. Disable with a value
5118 less than or equal to zero. Defaults to ten
5119 seconds. A change in value does not take effect
5120 until the beginning of the next grace period.
5122 rcupdate.rcu_task_stall_info_mult= [KNL]
5123 Multiplier for time interval between successive
5124 RCU task stall informational messages for a given
5125 RCU tasks grace period. This value is clamped
5126 to one through ten, inclusive. It defaults to
5127 the value three, so that the first informational
5128 message is printed 10 seconds into the grace
5129 period, the second at 40 seconds, the third at
5130 160 seconds, and then the stall warning at 600
5131 seconds would prevent a fourth at 640 seconds.
5133 rcupdate.rcu_task_stall_timeout= [KNL]
5134 Set timeout in jiffies for RCU task stall
5135 warning messages. Disable with a value less
5136 than or equal to zero. Defaults to ten minutes.
5137 A change in value does not take effect until
5138 the beginning of the next grace period.
5140 rcupdate.rcu_self_test= [KNL]
5141 Run the RCU early boot self tests
5145 Run specified binary instead of /init from the ramdisk,
5146 used for early userspace startup. See initrd.
5149 force - Override the decision by the kernel to hide the
5150 advertisement of RDRAND support (this affects
5151 certain AMD processors because of buggy BIOS
5152 support, specifically around the suspend/resume
5156 Turn on/off individual RDT features. List is:
5157 cmt, mbmtotal, mbmlocal, l3cat, l3cdp, l2cat, l2cdp,
5159 E.g. to turn on cmt and turn off mba use:
5163 Format (x86 or x86_64):
5164 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] | d[efault] \
5166 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \
5168 Where reboot_mode is one of warm (soft) or cold (hard) or gpio
5169 (prefix with 'panic_' to set mode for panic
5171 reboot_type is one of bios, acpi, kbd, triple, efi, or pci,
5172 reboot_force is either force or not specified,
5173 reboot_cpu is s[mp]#### with #### being the processor
5174 to be used for rebooting.
5176 refscale.holdoff= [KNL]
5177 Set test-start holdoff period. The purpose of
5178 this parameter is to delay the start of the
5179 test until boot completes in order to avoid
5182 refscale.loops= [KNL]
5183 Set the number of loops over the synchronization
5184 primitive under test. Increasing this number
5185 reduces noise due to loop start/end overhead,
5186 but the default has already reduced the per-pass
5187 noise to a handful of picoseconds on ca. 2020
5190 refscale.nreaders= [KNL]
5191 Set number of readers. The default value of -1
5192 selects N, where N is roughly 75% of the number
5193 of CPUs. A value of zero is an interesting choice.
5195 refscale.nruns= [KNL]
5196 Set number of runs, each of which is dumped onto
5199 refscale.readdelay= [KNL]
5200 Set the read-side critical-section duration,
5201 measured in microseconds.
5203 refscale.scale_type= [KNL]
5204 Specify the read-protection implementation to test.
5206 refscale.shutdown= [KNL]
5207 Shut down the system at the end of the performance
5208 test. This defaults to 1 (shut it down) when
5209 refscale is built into the kernel and to 0 (leave
5210 it running) when refscale is built as a module.
5212 refscale.verbose= [KNL]
5213 Enable additional printk() statements.
5215 refscale.verbose_batched= [KNL]
5216 Batch the additional printk() statements. If zero
5217 (the default) or negative, print everything. Otherwise,
5218 print every Nth verbose statement, where N is the value
5222 [KNL, SMP] Set scheduler's default relax_domain_level.
5223 See Documentation/admin-guide/cgroup-v1/cpusets.rst.
5225 reserve= [KNL,BUGS] Force kernel to ignore I/O ports or memory
5226 Format: <base1>,<size1>[,<base2>,<size2>,...]
5227 Reserve I/O ports or memory so the kernel won't use
5228 them. If <base> is less than 0x10000, the region
5229 is assumed to be I/O ports; otherwise it is memory.
5231 reservetop= [X86-32]
5233 Reserves a hole at the top of the kernel virtual
5236 reset_devices [KNL] Force drivers to reset the underlying device
5237 during initialization.
5240 Specify the partition device for software suspend
5242 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>}
5244 resume_offset= [SWSUSP]
5245 Specify the offset from the beginning of the partition
5246 given by "resume=" at which the swap header is located,
5247 in <PAGE_SIZE> units (needed only for swap files).
5248 See Documentation/power/swsusp-and-swap-files.rst
5250 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to
5251 read the resume files
5253 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up.
5254 Useful for devices that are detected asynchronously
5255 (e.g. USB and MMC devices).
5257 retain_initrd [RAM] Keep initrd memory after extraction
5259 retbleed= [X86] Control mitigation of RETBleed (Arbitrary
5260 Speculative Code Execution with Return Instructions)
5264 auto - automatically select a migitation
5265 auto,nosmt - automatically select a mitigation,
5266 disabling SMT if necessary for
5267 the full mitigation (only on Zen1
5268 and older without STIBP).
5269 ibpb - mitigate short speculation windows on
5270 basic block boundaries too. Safe, highest
5272 unret - force enable untrained return thunks,
5273 only effective on AMD f15h-f17h
5275 unret,nosmt - like unret, will disable SMT when STIBP
5278 Selecting 'auto' will choose a mitigation method at run
5279 time according to the CPU.
5281 Not specifying this option is equivalent to retbleed=auto.
5283 rfkill.default_state=
5284 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm,
5285 etc. communication is blocked by default.
5288 rfkill.master_switch_mode=
5289 0 The "airplane mode" button does nothing.
5290 1 The "airplane mode" button toggles between everything
5291 blocked and the previous configuration.
5292 2 The "airplane mode" button toggles between everything
5293 blocked and everything unblocked.
5295 rhash_entries= [KNL,NET]
5296 Set number of hash buckets for route cache
5299 [KNL] Disable ring 3 MONITOR/MWAIT feature on supported
5302 ro [KNL] Mount root device read-only on boot
5305 on Mark read-only kernel memory as read-only (default).
5306 off Leave read-only kernel memory writable for debugging.
5309 Enable the uart passthrough on the designated usb port
5310 on Rockchip SoCs. When active, the signals of the
5311 debug-uart get routed to the D+ and D- pins of the usb
5312 port and the regular usb controller gets disabled.
5314 root= [KNL] Root filesystem
5315 See name_to_dev_t comment in init/do_mounts.c.
5317 rootdelay= [KNL] Delay (in seconds) to pause before attempting to
5318 mount the root filesystem
5320 rootflags= [KNL] Set root filesystem mount option string
5322 rootfstype= [KNL] Set root filesystem type
5324 rootwait [KNL] Wait (indefinitely) for root device to show up.
5325 Useful for devices that are detected asynchronously
5326 (e.g. USB and MMC devices).
5328 rproc_mem=nn[KMG][@address]
5329 [KNL,ARM,CMA] Remoteproc physical memory block.
5330 Memory area to be used by remote processor image,
5333 rw [KNL] Mount root device read-write on boot
5335 S [KNL] Run init in single mode
5337 s390_iommu= [HW,S390]
5338 Set s390 IOTLB flushing mode
5340 With strict flushing every unmap operation will result in
5341 an IOTLB flush. Default is lazy flushing before reuse,
5344 s390_iommu_aperture= [KNL,S390]
5345 Specifies the size of the per device DMA address space
5346 accessible through the DMA and IOMMU APIs as a decimal
5347 factor of the size of main memory.
5348 The default is 1 meaning that one can concurrently use
5349 as many DMA addresses as physical memory is installed,
5350 if supported by hardware, and thus map all of memory
5351 once. With a value of 2 one can map all of memory twice
5352 and so on. As a special case a factor of 0 imposes no
5353 restrictions other than those given by hardware at the
5354 cost of significant additional memory use for tables.
5357 See drivers/net/irda/sa1100_ir.c.
5359 sched_verbose [KNL] Enables verbose scheduler debug messages.
5361 schedstats= [KNL,X86] Enable or disable scheduled statistics.
5362 Allowed values are enable and disable. This feature
5363 incurs a small amount of overhead in the scheduler
5364 but is useful for debugging and performance tuning.
5366 sched_thermal_decay_shift=
5367 [KNL, SMP] Set a decay shift for scheduler thermal
5368 pressure signal. Thermal pressure signal follows the
5369 default decay period of other scheduler pelt
5370 signals(usually 32 ms but configurable). Setting
5371 sched_thermal_decay_shift will left shift the decay
5372 period for the thermal pressure signal by the shift
5374 i.e. with the default pelt decay period of 32 ms
5375 sched_thermal_decay_shift thermal pressure decay pr
5379 Format: integer between 0 and 10
5382 scftorture.holdoff= [KNL]
5383 Number of seconds to hold off before starting
5384 test. Defaults to zero for module insertion and
5385 to 10 seconds for built-in smp_call_function()
5388 scftorture.longwait= [KNL]
5389 Request ridiculously long waits randomly selected
5390 up to the chosen limit in seconds. Zero (the
5391 default) disables this feature. Please note
5392 that requesting even small non-zero numbers of
5393 seconds can result in RCU CPU stall warnings,
5394 softlockup complaints, and so on.
5396 scftorture.nthreads= [KNL]
5397 Number of kthreads to spawn to invoke the
5398 smp_call_function() family of functions.
5399 The default of -1 specifies a number of kthreads
5400 equal to the number of CPUs.
5402 scftorture.onoff_holdoff= [KNL]
5403 Number seconds to wait after the start of the
5404 test before initiating CPU-hotplug operations.
5406 scftorture.onoff_interval= [KNL]
5407 Number seconds to wait between successive
5408 CPU-hotplug operations. Specifying zero (which
5409 is the default) disables CPU-hotplug operations.
5411 scftorture.shutdown_secs= [KNL]
5412 The number of seconds following the start of the
5413 test after which to shut down the system. The
5414 default of zero avoids shutting down the system.
5415 Non-zero values are useful for automated tests.
5417 scftorture.stat_interval= [KNL]
5418 The number of seconds between outputting the
5419 current test statistics to the console. A value
5420 of zero disables statistics output.
5422 scftorture.stutter_cpus= [KNL]
5423 The number of jiffies to wait between each change
5424 to the set of CPUs under test.
5426 scftorture.use_cpus_read_lock= [KNL]
5427 Use use_cpus_read_lock() instead of the default
5428 preempt_disable() to disable CPU hotplug
5429 while invoking one of the smp_call_function*()
5432 scftorture.verbose= [KNL]
5433 Enable additional printk() statements.
5435 scftorture.weight_single= [KNL]
5436 The probability weighting to use for the
5437 smp_call_function_single() function with a zero
5438 "wait" parameter. A value of -1 selects the
5439 default if all other weights are -1. However,
5440 if at least one weight has some other value, a
5441 value of -1 will instead select a weight of zero.
5443 scftorture.weight_single_wait= [KNL]
5444 The probability weighting to use for the
5445 smp_call_function_single() function with a
5446 non-zero "wait" parameter. See weight_single.
5448 scftorture.weight_many= [KNL]
5449 The probability weighting to use for the
5450 smp_call_function_many() function with a zero
5451 "wait" parameter. See weight_single.
5452 Note well that setting a high probability for
5453 this weighting can place serious IPI load
5456 scftorture.weight_many_wait= [KNL]
5457 The probability weighting to use for the
5458 smp_call_function_many() function with a
5459 non-zero "wait" parameter. See weight_single
5462 scftorture.weight_all= [KNL]
5463 The probability weighting to use for the
5464 smp_call_function_all() function with a zero
5465 "wait" parameter. See weight_single and
5468 scftorture.weight_all_wait= [KNL]
5469 The probability weighting to use for the
5470 smp_call_function_all() function with a
5471 non-zero "wait" parameter. See weight_single
5474 skew_tick= [KNL] Offset the periodic timer tick per cpu to mitigate
5475 xtime_lock contention on larger systems, and/or RCU lock
5476 contention on all systems with CONFIG_MAXSMP set.
5477 Format: { "0" | "1" }
5478 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1"
5480 Note: increases power consumption, thus should only be
5481 enabled if running jitter sensitive (HPC/RT) workloads.
5483 security= [SECURITY] Choose a legacy "major" security module to
5484 enable at boot. This has been deprecated by the
5487 selinux= [SELINUX] Disable or enable SELinux at boot time.
5488 Format: { "0" | "1" }
5489 See security/selinux/Kconfig help text.
5494 apparmor= [APPARMOR] Disable or enable AppArmor at boot time
5495 Format: { "0" | "1" }
5496 See security/apparmor/Kconfig help text
5499 Default value is set via kernel config option.
5501 serialnumber [BUGS=X86-32]
5503 sev=option[,option...] [X86-64] See Documentation/x86/x86_64/boot-options.rst
5506 Maximal number of shapers.
5514 Enable merging of slabs with similar size when the
5515 kernel is built without CONFIG_SLAB_MERGE_DEFAULT.
5518 Disable merging of slabs with similar size. May be
5519 necessary if there is some reason to distinguish
5520 allocs to different slabs, especially in hardened
5521 environments where the risk of heap overflows and
5522 layout control by attackers can usually be
5523 frustrated by disabling merging. This will reduce
5524 most of the exposure of a heap attack to a single
5525 cache (risks via metadata attacks are mostly
5526 unchanged). Debug options disable merging on their
5528 For more information see Documentation/mm/slub.rst.
5530 slab_max_order= [MM, SLAB]
5531 Determines the maximum allowed order for slabs.
5532 A high setting may cause OOMs due to memory
5533 fragmentation. Defaults to 1 for systems with
5534 more than 32MB of RAM, 0 otherwise.
5536 slub_debug[=options[,slabs][;[options[,slabs]]...] [MM, SLUB]
5537 Enabling slub_debug allows one to determine the
5538 culprit if slab objects become corrupted. Enabling
5539 slub_debug can create guard zones around objects and
5540 may poison objects when not in use. Also tracks the
5541 last alloc / free. For more information see
5542 Documentation/mm/slub.rst.
5544 slub_max_order= [MM, SLUB]
5545 Determines the maximum allowed order for slabs.
5546 A high setting may cause OOMs due to memory
5547 fragmentation. For more information see
5548 Documentation/mm/slub.rst.
5550 slub_min_objects= [MM, SLUB]
5551 The minimum number of objects per slab. SLUB will
5552 increase the slab order up to slub_max_order to
5553 generate a sufficiently large slab able to contain
5554 the number of objects indicated. The higher the number
5555 of objects the smaller the overhead of tracking slabs
5556 and the less frequently locks need to be acquired.
5557 For more information see Documentation/mm/slub.rst.
5559 slub_min_order= [MM, SLUB]
5560 Determines the minimum page order for slabs. Must be
5561 lower than slub_max_order.
5562 For more information see Documentation/mm/slub.rst.
5564 slub_merge [MM, SLUB]
5565 Same with slab_merge.
5567 slub_nomerge [MM, SLUB]
5568 Same with slab_nomerge. This is supported for legacy.
5569 See slab_nomerge for more information.
5572 Format: <io1>[,<io2>[,...,<io8>]]
5574 smp.csd_lock_timeout= [KNL]
5575 Specify the period of time in milliseconds
5576 that smp_call_function() and friends will wait
5577 for a CPU to release the CSD lock. This is
5578 useful when diagnosing bugs involving CPUs
5579 disabling interrupts for extended periods
5580 of time. Defaults to 5,000 milliseconds, and
5581 setting a value of zero disables this feature.
5582 This feature may be more efficiently disabled
5583 using the csdlock_debug- kernel parameter.
5585 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices
5586 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port
5587 smsc-ircc2.ircc_sir= [HW] SIR base I/O port
5588 smsc-ircc2.ircc_fir= [HW] FIR base I/O port
5589 smsc-ircc2.ircc_irq= [HW] IRQ line
5590 smsc-ircc2.ircc_dma= [HW] DMA channel
5591 smsc-ircc2.ircc_transceiver= [HW] Transceiver type:
5592 0: Toshiba Satellite 1800 (GP data pin select)
5593 1: Fast pin select (default)
5596 smt= [KNL,S390] Set the maximum number of threads (logical
5597 CPUs) to use per physical CPU on systems capable of
5598 symmetric multithreading (SMT). Will be capped to the
5599 actual hardware limit.
5601 Default: -1 (no limit)
5604 [KNL] Should the soft-lockup detector generate panics.
5607 A value of 1 instructs the soft-lockup detector
5608 to panic the machine when a soft-lockup occurs. It is
5609 also controlled by the kernel.softlockup_panic sysctl
5610 and CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC, which is the
5611 respective build-time switch to that functionality.
5613 softlockup_all_cpu_backtrace=
5614 [KNL] Should the soft-lockup detector generate
5615 backtraces on all cpus.
5618 sonypi.*= [HW] Sony Programmable I/O Control Device driver
5619 See Documentation/admin-guide/laptops/sonypi.rst
5621 spectre_v2= [X86] Control mitigation of Spectre variant 2
5622 (indirect branch speculation) vulnerability.
5623 The default operation protects the kernel from
5626 on - unconditionally enable, implies
5628 off - unconditionally disable, implies
5630 auto - kernel detects whether your CPU model is
5633 Selecting 'on' will, and 'auto' may, choose a
5634 mitigation method at run time according to the
5635 CPU, the available microcode, the setting of the
5636 CONFIG_RETPOLINE configuration option, and the
5637 compiler with which the kernel was built.
5639 Selecting 'on' will also enable the mitigation
5640 against user space to user space task attacks.
5642 Selecting 'off' will disable both the kernel and
5643 the user space protections.
5645 Specific mitigations can also be selected manually:
5647 retpoline - replace indirect branches
5648 retpoline,generic - Retpolines
5649 retpoline,lfence - LFENCE; indirect branch
5650 retpoline,amd - alias for retpoline,lfence
5651 eibrs - enhanced IBRS
5652 eibrs,retpoline - enhanced IBRS + Retpolines
5653 eibrs,lfence - enhanced IBRS + LFENCE
5654 ibrs - use IBRS to protect kernel
5656 Not specifying this option is equivalent to
5660 [X86] Control mitigation of Spectre variant 2
5661 (indirect branch speculation) vulnerability between
5664 on - Unconditionally enable mitigations. Is
5665 enforced by spectre_v2=on
5667 off - Unconditionally disable mitigations. Is
5668 enforced by spectre_v2=off
5670 prctl - Indirect branch speculation is enabled,
5671 but mitigation can be enabled via prctl
5672 per thread. The mitigation control state
5673 is inherited on fork.
5676 - Like "prctl" above, but only STIBP is
5677 controlled per thread. IBPB is issued
5678 always when switching between different user
5682 - Same as "prctl" above, but all seccomp
5683 threads will enable the mitigation unless
5684 they explicitly opt out.
5687 - Like "seccomp" above, but only STIBP is
5688 controlled per thread. IBPB is issued
5689 always when switching between different
5690 user space processes.
5692 auto - Kernel selects the mitigation depending on
5693 the available CPU features and vulnerability.
5695 Default mitigation: "prctl"
5697 Not specifying this option is equivalent to
5698 spectre_v2_user=auto.
5700 spec_store_bypass_disable=
5701 [HW] Control Speculative Store Bypass (SSB) Disable mitigation
5702 (Speculative Store Bypass vulnerability)
5704 Certain CPUs are vulnerable to an exploit against a
5705 a common industry wide performance optimization known
5706 as "Speculative Store Bypass" in which recent stores
5707 to the same memory location may not be observed by
5708 later loads during speculative execution. The idea
5709 is that such stores are unlikely and that they can
5710 be detected prior to instruction retirement at the
5711 end of a particular speculation execution window.
5713 In vulnerable processors, the speculatively forwarded
5714 store can be used in a cache side channel attack, for
5715 example to read memory to which the attacker does not
5716 directly have access (e.g. inside sandboxed code).
5718 This parameter controls whether the Speculative Store
5719 Bypass optimization is used.
5721 On x86 the options are:
5723 on - Unconditionally disable Speculative Store Bypass
5724 off - Unconditionally enable Speculative Store Bypass
5725 auto - Kernel detects whether the CPU model contains an
5726 implementation of Speculative Store Bypass and
5727 picks the most appropriate mitigation. If the
5728 CPU is not vulnerable, "off" is selected. If the
5729 CPU is vulnerable the default mitigation is
5730 architecture and Kconfig dependent. See below.
5731 prctl - Control Speculative Store Bypass per thread
5732 via prctl. Speculative Store Bypass is enabled
5733 for a process by default. The state of the control
5734 is inherited on fork.
5735 seccomp - Same as "prctl" above, but all seccomp threads
5736 will disable SSB unless they explicitly opt out.
5738 Default mitigations:
5741 On powerpc the options are:
5743 on,auto - On Power8 and Power9 insert a store-forwarding
5744 barrier on kernel entry and exit. On Power7
5745 perform a software flush on kernel entry and
5749 Not specifying this option is equivalent to
5750 spec_store_bypass_disable=auto.
5752 spia_io_base= [HW,MTD]
5758 [X86] Enable split lock detection or bus lock detection
5760 When enabled (and if hardware support is present), atomic
5761 instructions that access data across cache line
5762 boundaries will result in an alignment check exception
5763 for split lock detection or a debug exception for
5768 warn - the kernel will emit rate-limited warnings
5769 about applications triggering the #AC
5770 exception or the #DB exception. This mode is
5771 the default on CPUs that support split lock
5772 detection or bus lock detection. Default
5773 behavior is by #AC if both features are
5774 enabled in hardware.
5776 fatal - the kernel will send SIGBUS to applications
5777 that trigger the #AC exception or the #DB
5778 exception. Default behavior is by #AC if
5779 both features are enabled in hardware.
5782 Set system wide rate limit to N bus locks
5783 per second for bus lock detection.
5786 N/A for split lock detection.
5789 If an #AC exception is hit in the kernel or in
5790 firmware (i.e. not while executing in user mode)
5791 the kernel will oops in either "warn" or "fatal"
5794 #DB exception for bus lock is triggered only when
5798 Control the Special Register Buffer Data Sampling
5801 Certain CPUs are vulnerable to an MDS-like
5802 exploit which can leak bits from the random
5805 By default, this issue is mitigated by
5806 microcode. However, the microcode fix can cause
5807 the RDRAND and RDSEED instructions to become
5808 much slower. Among other effects, this will
5809 result in reduced throughput from /dev/urandom.
5811 The microcode mitigation can be disabled with
5812 the following option:
5814 off: Disable mitigation and remove
5815 performance impact to RDRAND and RDSEED
5817 srcutree.big_cpu_lim [KNL]
5818 Specifies the number of CPUs constituting a
5819 large system, such that srcu_struct structures
5820 should immediately allocate an srcu_node array.
5821 This kernel-boot parameter defaults to 128,
5822 but takes effect only when the low-order four
5823 bits of srcutree.convert_to_big is equal to 3
5826 srcutree.convert_to_big [KNL]
5827 Specifies under what conditions an SRCU tree
5828 srcu_struct structure will be converted to big
5829 form, that is, with an rcu_node tree:
5832 1: At init_srcu_struct() time.
5833 2: When rcutorture decides to.
5834 3: Decide at boot time (default).
5835 0x1X: Above plus if high contention.
5837 Either way, the srcu_node tree will be sized based
5838 on the actual runtime number of CPUs (nr_cpu_ids)
5839 instead of the compile-time CONFIG_NR_CPUS.
5841 srcutree.counter_wrap_check [KNL]
5842 Specifies how frequently to check for
5843 grace-period sequence counter wrap for the
5844 srcu_data structure's ->srcu_gp_seq_needed field.
5845 The greater the number of bits set in this kernel
5846 parameter, the less frequently counter wrap will
5847 be checked for. Note that the bottom two bits
5850 srcutree.exp_holdoff [KNL]
5851 Specifies how many nanoseconds must elapse
5852 since the end of the last SRCU grace period for
5853 a given srcu_struct until the next normal SRCU
5854 grace period will be considered for automatic
5855 expediting. Set to zero to disable automatic
5858 srcutree.srcu_max_nodelay [KNL]
5859 Specifies the number of no-delay instances
5860 per jiffy for which the SRCU grace period
5861 worker thread will be rescheduled with zero
5862 delay. Beyond this limit, worker thread will
5863 be rescheduled with a sleep delay of one jiffy.
5865 srcutree.srcu_max_nodelay_phase [KNL]
5866 Specifies the per-grace-period phase, number of
5867 non-sleeping polls of readers. Beyond this limit,
5868 grace period worker thread will be rescheduled
5869 with a sleep delay of one jiffy, between each
5870 rescan of the readers, for a grace period phase.
5872 srcutree.srcu_retry_check_delay [KNL]
5873 Specifies number of microseconds of non-sleeping
5874 delay between each non-sleeping poll of readers.
5876 srcutree.small_contention_lim [KNL]
5877 Specifies the number of update-side contention
5878 events per jiffy will be tolerated before
5879 initiating a conversion of an srcu_struct
5880 structure to big form. Note that the value of
5881 srcutree.convert_to_big must have the 0x10 bit
5882 set for contention-based conversions to occur.
5885 Speculative Store Bypass Disable control
5887 On CPUs that are vulnerable to the Speculative
5888 Store Bypass vulnerability and offer a
5889 firmware based mitigation, this parameter
5890 indicates how the mitigation should be used:
5892 force-on: Unconditionally enable mitigation for
5893 for both kernel and userspace
5894 force-off: Unconditionally disable mitigation for
5895 for both kernel and userspace
5896 kernel: Always enable mitigation in the
5897 kernel, and offer a prctl interface
5898 to allow userspace to register its
5899 interest in being mitigated too.
5901 stack_guard_gap= [MM]
5902 override the default stack gap protection. The value
5903 is in page units and it defines how many pages prior
5904 to (for stacks growing down) resp. after (for stacks
5905 growing up) the main stack are reserved for no other
5906 mapping. Default value is 256 pages.
5908 stack_depot_disable= [KNL]
5909 Setting this to true through kernel command line will
5910 disable the stack depot thereby saving the static memory
5911 consumed by the stack hash table. By default this is set
5915 Enabled the stack tracer on boot up.
5917 stacktrace_filter=[function-list]
5918 [FTRACE] Limit the functions that the stack tracer
5919 will trace at boot up. function-list is a comma-separated
5920 list of functions. This list can be changed at run
5921 time by the stack_trace_filter file in the debugfs
5922 tracing directory. Note, this enables stack tracing
5923 and the stacktrace above is not needed.
5927 Set the STI (builtin display/keyboard on the HP-PARISC
5928 machines) console (graphic card) which should be used
5929 as the initial boot-console.
5930 See also comment in drivers/video/console/sticore.c.
5933 See comment in drivers/video/console/sticore.c.
5936 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]]
5941 Enable or disable strict sigaltstack size checks
5942 against the required signal frame size which
5943 depends on the supported FPU features. This can
5944 be used to filter out binaries which have
5945 not yet been made aware of AT_MINSIGSTKSZ.
5947 sunrpc.min_resvport=
5948 sunrpc.max_resvport=
5950 SunRPC servers often require that client requests
5951 originate from a privileged port (i.e. a port in the
5952 range 0 < portnr < 1024).
5953 An administrator who wishes to reserve some of these
5954 ports for other uses may adjust the range that the
5955 kernel's sunrpc client considers to be privileged
5956 using these two parameters to set the minimum and
5957 maximum port values.
5959 sunrpc.svc_rpc_per_connection_limit=
5961 Limit the number of requests that the server will
5962 process in parallel from a single connection.
5963 The default value is 0 (no limit).
5967 Control how the NFS server code allocates CPUs to
5968 service thread pools. Depending on how many NICs
5969 you have and where their interrupts are bound, this
5970 option will affect which CPUs will do NFS serving.
5971 Note: this parameter cannot be changed while the
5972 NFS server is running.
5974 auto the server chooses an appropriate mode
5975 automatically using heuristics
5976 global a single global pool contains all CPUs
5977 percpu one pool for each CPU
5978 pernode one pool for each NUMA node (equivalent
5979 to global on non-NUMA machines)
5981 sunrpc.tcp_slot_table_entries=
5982 sunrpc.udp_slot_table_entries=
5984 Sets the upper limit on the number of simultaneous
5985 RPC calls that can be sent from the client to a
5986 server. Increasing these values may allow you to
5987 improve throughput, but will also increase the
5988 amount of memory reserved for use by the client.
5990 suspend.pm_test_delay=
5992 Sets the number of seconds to remain in a suspend test
5993 mode before resuming the system (see
5994 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG
5995 is set. Default value is 5.
5998 Format: { on | off | y | n | 1 | 0 }
5999 This parameter controls use of the Protected
6000 Execution Facility on pSeries.
6004 Enable accounting of swap in memory resource
6005 controller if no parameter or 1 is given or disable
6006 it if 0 is given (See Documentation/admin-guide/cgroup-v1/memory.rst)
6008 swiotlb= [ARM,IA-64,PPC,MIPS,X86]
6009 Format: { <int> [,<int>] | force | noforce }
6010 <int> -- Number of I/O TLB slabs
6011 <int> -- Second integer after comma. Number of swiotlb
6012 areas with their own lock. Will be rounded up
6014 force -- force using of bounce buffers even if they
6015 wouldn't be automatically used by the kernel
6016 noforce -- Never use bounce buffers (for debugging)
6021 Set a sysctl parameter, right before loading the init
6022 process, as if the value was written to the respective
6023 /proc/sys/... file. Both '.' and '/' are recognized as
6024 separators. Unrecognized parameters and invalid values
6025 are reported in the kernel log. Sysctls registered
6026 later by a loaded module cannot be set this way.
6027 Example: sysctl.vm.swappiness=40
6029 sysfs.deprecated=0|1 [KNL]
6030 Enable/disable old style sysfs layout for old udev
6031 on older distributions. When this option is enabled
6032 very new udev will not work anymore. When this option
6033 is disabled (or CONFIG_SYSFS_DEPRECATED not compiled)
6034 in older udev will not work anymore.
6035 Default depends on CONFIG_SYSFS_DEPRECATED_V2 set in
6036 the kernel configuration.
6038 sysrq_always_enabled
6040 Ignore sysrq setting - this boot parameter will
6041 neutralize any effect of /proc/sys/kernel/sysrq.
6042 Useful for debugging.
6044 tcpmhash_entries= [KNL,NET]
6045 Set the number of tcp_metrics_hash slots.
6046 Default value is 8192 or 16384 depending on total
6047 ram pages. This is used to specify the TCP metrics
6048 cache size. See Documentation/networking/ip-sysctl.rst
6049 "tcp_no_metrics_save" section for more details.
6053 test_suspend= [SUSPEND]
6054 Format: { "mem" | "standby" | "freeze" }[,N]
6055 Specify "mem" (for Suspend-to-RAM) or "standby" (for
6056 standby suspend) or "freeze" (for suspend type freeze)
6057 as the system sleep state during system startup with
6058 the optional capability to repeat N number of times.
6059 The system is woken from this state using a
6060 wakeup-capable RTC alarm.
6062 thash_entries= [KNL,NET]
6063 Set number of hash buckets for TCP connection
6065 thermal.act= [HW,ACPI]
6066 -1: disable all active trip points in all thermal zones
6067 <degrees C>: override all lowest active trip points
6069 thermal.crt= [HW,ACPI]
6070 -1: disable all critical trip points in all thermal zones
6071 <degrees C>: override all critical trip points
6073 thermal.nocrt= [HW,ACPI]
6074 Set to disable actions on ACPI thermal zone
6075 critical and hot trip points.
6077 thermal.off= [HW,ACPI]
6078 1: disable ACPI thermal control
6080 thermal.psv= [HW,ACPI]
6081 -1: disable all passive trip points
6082 <degrees C>: override all passive trip points to this
6085 thermal.tzp= [HW,ACPI]
6086 Specify global default ACPI thermal zone polling rate
6087 <deci-seconds>: poll all this frequency
6088 0: no polling (default)
6091 Force threading of all interrupt handlers except those
6092 marked explicitly IRQF_NO_THREAD.
6096 Specify if the kernel should make use of the cpu
6097 topology information if the hardware supports this.
6098 The scheduler will make use of this information and
6099 e.g. base its process migration decisions on it.
6102 topology_updates= [KNL, PPC, NUMA]
6104 Specify if the kernel should ignore (off)
6105 topology updates sent by the hypervisor to this
6108 torture.disable_onoff_at_boot= [KNL]
6109 Prevent the CPU-hotplug component of torturing
6110 until after init has spawned.
6112 torture.ftrace_dump_at_shutdown= [KNL]
6113 Dump the ftrace buffer at torture-test shutdown,
6114 even if there were no errors. This can be a
6115 very costly operation when many torture tests
6116 are running concurrently, especially on systems
6117 with rotating-rust storage.
6119 torture.verbose_sleep_frequency= [KNL]
6120 Specifies how many verbose printk()s should be
6121 emitted between each sleep. The default of zero
6122 disables verbose-printk() sleeping.
6124 torture.verbose_sleep_duration= [KNL]
6125 Duration of each verbose-printk() sleep in jiffies.
6129 tpm_suspend_pcr=[HW,TPM]
6130 Format: integer pcr id
6131 Specify that at suspend time, the tpm driver
6132 should extend the specified pcr with zeros,
6133 as a workaround for some chips which fail to
6134 flush the last written pcr on TPM_SaveState.
6135 This will guarantee that all the other pcrs
6139 Have the tracepoints sent to printk as well as the
6140 tracing ring buffer. This is useful for early boot up
6141 where the system hangs or reboots and does not give the
6142 option for reading the tracing buffer or performing a
6143 ftrace_dump_on_oops.
6145 To turn off having tracepoints sent to printk,
6146 echo 0 > /proc/sys/kernel/tracepoint_printk
6147 Note, echoing 1 into this file without the
6148 tracepoint_printk kernel cmdline option has no effect.
6150 The tp_printk_stop_on_boot (see below) can also be used
6151 to stop the printing of events to console at
6156 Having tracepoints sent to printk() and activating high
6157 frequency tracepoints such as irq or sched, can cause
6158 the system to live lock.
6160 tp_printk_stop_on_boot [FTRACE]
6161 When tp_printk (above) is set, it can cause a lot of noise
6162 on the console. It may be useful to only include the
6163 printing of events during boot up, as user space may
6164 make the system inoperable.
6166 This command line option will stop the printing of events
6167 to console at the late_initcall_sync() time frame.
6169 trace_buf_size=nn[KMG]
6170 [FTRACE] will set tracing buffer size on each cpu.
6172 trace_clock= [FTRACE] Set the clock used for tracing events
6174 local - Use the per CPU time stamp counter
6175 (converted into nanoseconds). Fast, but
6176 depending on the architecture, may not be
6177 in sync between CPUs.
6178 global - Event time stamps are synchronize across
6179 CPUs. May be slower than the local clock,
6180 but better for some race conditions.
6181 counter - Simple counting of events (1, 2, ..)
6182 note, some counts may be skipped due to the
6183 infrastructure grabbing the clock more than
6185 uptime - Use jiffies as the time stamp.
6186 perf - Use the same clock that perf uses.
6187 mono - Use ktime_get_mono_fast_ns() for time stamps.
6188 mono_raw - Use ktime_get_raw_fast_ns() for time
6190 boot - Use ktime_get_boot_fast_ns() for time stamps.
6191 Architectures may add more clocks. See
6192 Documentation/trace/ftrace.rst for more details.
6194 trace_event=[event-list]
6195 [FTRACE] Set and start specified trace events in order
6196 to facilitate early boot debugging. The event-list is a
6197 comma-separated list of trace events to enable. See
6198 also Documentation/trace/events.rst
6200 trace_options=[option-list]
6201 [FTRACE] Enable or disable tracer options at boot.
6202 The option-list is a comma delimited list of options
6203 that can be enabled or disabled just as if you were
6204 to echo the option name into
6206 /sys/kernel/debug/tracing/trace_options
6208 For example, to enable stacktrace option (to dump the
6209 stack trace of each event), add to the command line:
6211 trace_options=stacktrace
6213 See also Documentation/trace/ftrace.rst "trace options"
6217 [FTRACE] enable this option to disable tracing when a
6218 warning is hit. This turns off "tracing_on". Tracing can
6219 be enabled again by echoing '1' into the "tracing_on"
6220 file located in /sys/kernel/debug/tracing/
6222 This option is useful, as it disables the trace before
6223 the WARNING dump is called, which prevents the trace to
6224 be filled with content caused by the warning output.
6226 This option can also be set at run time via the sysctl
6227 option: kernel/traceoff_on_warning
6229 transparent_hugepage=
6231 Format: [always|madvise|never]
6232 Can be used to control the default behavior of the system
6233 with respect to transparent hugepages.
6234 See Documentation/admin-guide/mm/transhuge.rst
6237 trusted.source= [KEYS]
6239 This parameter identifies the trust source as a backend
6240 for trusted keys implementation. Supported trust
6245 If not specified then it defaults to iterating through
6246 the trust source list starting with TPM and assigns the
6247 first trust source as a backend which is initialized
6248 successfully during iteration.
6252 The RNG used to generate key material for trusted keys.
6255 - the same value as trusted.source: "tpm" or "tee"
6257 If not specified, "default" is used. In this case,
6258 the RNG's choice is left to each individual trust source.
6260 tsc= Disable clocksource stability checks for TSC.
6262 [x86] reliable: mark tsc clocksource as reliable, this
6263 disables clocksource verification at runtime, as well
6264 as the stability checks done at bootup. Used to enable
6265 high-resolution timer mode on older hardware, and in
6266 virtualized environment.
6267 [x86] noirqtime: Do not use TSC to do irq accounting.
6268 Used to run time disable IRQ_TIME_ACCOUNTING on any
6269 platforms where RDTSC is slow and this accounting
6271 [x86] unstable: mark the TSC clocksource as unstable, this
6272 marks the TSC unconditionally unstable at bootup and
6273 avoids any further wobbles once the TSC watchdog notices.
6274 [x86] nowatchdog: disable clocksource watchdog. Used
6275 in situations with strict latency requirements (where
6276 interruptions from clocksource watchdog are not
6279 tsc_early_khz= [X86] Skip early TSC calibration and use the given
6280 value instead. Useful when the early TSC frequency discovery
6281 procedure is not reliable, such as on overclocked systems
6282 with CPUID.16h support and partial CPUID.15h support.
6283 Format: <unsigned int>
6285 tsx= [X86] Control Transactional Synchronization
6286 Extensions (TSX) feature in Intel processors that
6287 support TSX control.
6289 This parameter controls the TSX feature. The options are:
6291 on - Enable TSX on the system. Although there are
6292 mitigations for all known security vulnerabilities,
6293 TSX has been known to be an accelerator for
6294 several previous speculation-related CVEs, and
6295 so there may be unknown security risks associated
6296 with leaving it enabled.
6298 off - Disable TSX on the system. (Note that this
6299 option takes effect only on newer CPUs which are
6300 not vulnerable to MDS, i.e., have
6301 MSR_IA32_ARCH_CAPABILITIES.MDS_NO=1 and which get
6302 the new IA32_TSX_CTRL MSR through a microcode
6303 update. This new MSR allows for the reliable
6304 deactivation of the TSX functionality.)
6306 auto - Disable TSX if X86_BUG_TAA is present,
6307 otherwise enable TSX on the system.
6309 Not specifying this option is equivalent to tsx=off.
6311 See Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
6314 tsx_async_abort= [X86,INTEL] Control mitigation for the TSX Async
6315 Abort (TAA) vulnerability.
6317 Similar to Micro-architectural Data Sampling (MDS)
6318 certain CPUs that support Transactional
6319 Synchronization Extensions (TSX) are vulnerable to an
6320 exploit against CPU internal buffers which can forward
6321 information to a disclosure gadget under certain
6324 In vulnerable processors, the speculatively forwarded
6325 data can be used in a cache side channel attack, to
6326 access data to which the attacker does not have direct
6329 This parameter controls the TAA mitigation. The
6332 full - Enable TAA mitigation on vulnerable CPUs
6335 full,nosmt - Enable TAA mitigation and disable SMT on
6336 vulnerable CPUs. If TSX is disabled, SMT
6337 is not disabled because CPU is not
6338 vulnerable to cross-thread TAA attacks.
6339 off - Unconditionally disable TAA mitigation
6341 On MDS-affected machines, tsx_async_abort=off can be
6342 prevented by an active MDS mitigation as both vulnerabilities
6343 are mitigated with the same mechanism so in order to disable
6344 this mitigation, you need to specify mds=off too.
6346 Not specifying this option is equivalent to
6347 tsx_async_abort=full. On CPUs which are MDS affected
6348 and deploy MDS mitigation, TAA mitigation is not
6349 required and doesn't provide any additional
6353 Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
6355 turbografx.map[2|3]= [HW,JOY]
6356 TurboGraFX parallel port interface
6358 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7>
6359 See also Documentation/input/devices/joystick-parport.rst
6361 udbg-immortal [PPC] When debugging early kernel crashes that
6362 happen after console_init() and before a proper
6363 console driver takes over, this boot options might
6364 help "seeing" what's going on.
6366 uhash_entries= [KNL,NET]
6367 Set number of hash buckets for UDP/UDP-Lite connections
6370 [USB] Ignore overcurrent events (default N).
6371 Some badly-designed motherboards generate lots of
6372 bogus events, for ports that aren't wired to
6373 anything. Set this parameter to avoid log spamming.
6374 Note that genuine overcurrent events won't be
6378 [X86] Cause panic on unknown NMI.
6380 usbcore.authorized_default=
6381 [USB] Default USB device authorization:
6382 (default -1 = authorized except for wireless USB,
6383 0 = not authorized, 1 = authorized, 2 = authorized
6384 if device connected to internal port)
6386 usbcore.autosuspend=
6387 [USB] The autosuspend time delay (in seconds) used
6388 for newly-detected USB devices (default 2). This
6389 is the time required before an idle device will be
6390 autosuspended. Devices for which the delay is set
6391 to a negative value won't be autosuspended at all.
6393 usbcore.usbfs_snoop=
6394 [USB] Set to log all usbfs traffic (default 0 = off).
6396 usbcore.usbfs_snoop_max=
6397 [USB] Maximum number of bytes to snoop in each URB
6400 usbcore.blinkenlights=
6401 [USB] Set to cycle leds on hubs (default 0 = off).
6403 usbcore.old_scheme_first=
6404 [USB] Start with the old device initialization
6405 scheme (default 0 = off).
6407 usbcore.usbfs_memory_mb=
6408 [USB] Memory limit (in MB) for buffers allocated by
6409 usbfs (default = 16, 0 = max = 2047).
6411 usbcore.use_both_schemes=
6412 [USB] Try the other device initialization scheme
6413 if the first one fails (default 1 = enabled).
6415 usbcore.initial_descriptor_timeout=
6416 [USB] Specifies timeout for the initial 64-byte
6417 USB_REQ_GET_DESCRIPTOR request in milliseconds
6418 (default 5000 = 5.0 seconds).
6420 usbcore.nousb [USB] Disable the USB subsystem
6423 [USB] A list of quirk entries to augment the built-in
6424 usb core quirk list. List entries are separated by
6425 commas. Each entry has the form
6426 VendorID:ProductID:Flags. The IDs are 4-digit hex
6427 numbers and Flags is a set of letters. Each letter
6428 will change the built-in quirk; setting it if it is
6429 clear and clearing it if it is set. The letters have
6430 the following meanings:
6431 a = USB_QUIRK_STRING_FETCH_255 (string
6432 descriptors must not be fetched using
6434 b = USB_QUIRK_RESET_RESUME (device can't resume
6435 correctly so reset it instead);
6436 c = USB_QUIRK_NO_SET_INTF (device can't handle
6437 Set-Interface requests);
6438 d = USB_QUIRK_CONFIG_INTF_STRINGS (device can't
6439 handle its Configuration or Interface
6441 e = USB_QUIRK_RESET (device can't be reset
6442 (e.g morph devices), don't use reset);
6443 f = USB_QUIRK_HONOR_BNUMINTERFACES (device has
6444 more interface descriptions than the
6445 bNumInterfaces count, and can't handle
6446 talking to these interfaces);
6447 g = USB_QUIRK_DELAY_INIT (device needs a pause
6448 during initialization, after we read
6449 the device descriptor);
6450 h = USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL (For
6451 high speed and super speed interrupt
6452 endpoints, the USB 2.0 and USB 3.0 spec
6453 require the interval in microframes (1
6454 microframe = 125 microseconds) to be
6455 calculated as interval = 2 ^
6457 Devices with this quirk report their
6458 bInterval as the result of this
6459 calculation instead of the exponent
6460 variable used in the calculation);
6461 i = USB_QUIRK_DEVICE_QUALIFIER (device can't
6462 handle device_qualifier descriptor
6464 j = USB_QUIRK_IGNORE_REMOTE_WAKEUP (device
6465 generates spurious wakeup, ignore
6466 remote wakeup capability);
6467 k = USB_QUIRK_NO_LPM (device can't handle Link
6469 l = USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL
6470 (Device reports its bInterval as linear
6471 frames instead of the USB 2.0
6473 m = USB_QUIRK_DISCONNECT_SUSPEND (Device needs
6474 to be disconnected before suspend to
6475 prevent spurious wakeup);
6476 n = USB_QUIRK_DELAY_CTRL_MSG (Device needs a
6477 pause after every control message);
6478 o = USB_QUIRK_HUB_SLOW_RESET (Hub needs extra
6479 delay after resetting its port);
6480 Example: quirks=0781:5580:bk,0a5c:5834:gij
6483 [USBHID] The interval which mice are to be polled at.
6486 [USBHID] The interval which joysticks are to be polled at.
6489 [USBHID] The interval which keyboards are to be polled at.
6491 usb-storage.delay_use=
6492 [UMS] The delay in seconds before a new device is
6493 scanned for Logical Units (default 1).
6496 [UMS] A list of quirks entries to supplement or
6497 override the built-in unusual_devs list. List
6498 entries are separated by commas. Each entry has
6499 the form VID:PID:Flags where VID and PID are Vendor
6500 and Product ID values (4-digit hex numbers) and
6501 Flags is a set of characters, each corresponding
6502 to a common usb-storage quirk flag as follows:
6503 a = SANE_SENSE (collect more than 18 bytes
6504 of sense data, not on uas);
6505 b = BAD_SENSE (don't collect more than 18
6506 bytes of sense data, not on uas);
6507 c = FIX_CAPACITY (decrease the reported
6508 device capacity by one sector);
6509 d = NO_READ_DISC_INFO (don't use
6510 READ_DISC_INFO command, not on uas);
6511 e = NO_READ_CAPACITY_16 (don't use
6512 READ_CAPACITY_16 command);
6513 f = NO_REPORT_OPCODES (don't use report opcodes
6515 g = MAX_SECTORS_240 (don't transfer more than
6516 240 sectors at a time, uas only);
6517 h = CAPACITY_HEURISTICS (decrease the
6518 reported device capacity by one
6519 sector if the number is odd);
6520 i = IGNORE_DEVICE (don't bind to this
6522 j = NO_REPORT_LUNS (don't use report luns
6524 k = NO_SAME (do not use WRITE_SAME, uas only)
6525 l = NOT_LOCKABLE (don't try to lock and
6526 unlock ejectable media, not on uas);
6527 m = MAX_SECTORS_64 (don't transfer more
6528 than 64 sectors = 32 KB at a time,
6530 n = INITIAL_READ10 (force a retry of the
6531 initial READ(10) command, not on uas);
6532 o = CAPACITY_OK (accept the capacity
6533 reported by the device, not on uas);
6534 p = WRITE_CACHE (the device cache is ON
6535 by default, not on uas);
6536 r = IGNORE_RESIDUE (the device reports
6537 bogus residue values, not on uas);
6538 s = SINGLE_LUN (the device has only one
6540 t = NO_ATA_1X (don't allow ATA(12) and ATA(16)
6541 commands, uas only);
6542 u = IGNORE_UAS (don't bind to the uas driver);
6543 w = NO_WP_DETECT (don't test whether the
6544 medium is write-protected).
6545 y = ALWAYS_SYNC (issue a SYNCHRONIZE_CACHE
6546 even if the device claims no cache,
6548 Example: quirks=0419:aaf5:rl,0421:0433:rc
6550 user_debug= [KNL,ARM]
6552 See arch/arm/Kconfig.debug help text.
6553 1 - undefined instruction events
6555 4 - invalid data aborts
6558 Example: user_debug=31
6561 [X86] Flags controlling user PTE allocations.
6563 nohigh = do not allocate PTE pages in
6564 HIGHMEM regardless of setting
6567 vdso= [X86,SH,SPARC]
6568 On X86_32, this is an alias for vdso32=. Otherwise:
6570 vdso=1: enable VDSO (the default)
6571 vdso=0: disable VDSO mapping
6573 vdso32= [X86] Control the 32-bit vDSO
6574 vdso32=1: enable 32-bit VDSO
6575 vdso32=0 or vdso32=2: disable 32-bit VDSO
6577 See the help text for CONFIG_COMPAT_VDSO for more
6578 details. If CONFIG_COMPAT_VDSO is set, the default is
6579 vdso32=0; otherwise, the default is vdso32=1.
6581 For compatibility with older kernels, vdso32=2 is an
6584 Try vdso32=0 if you encounter an error that says:
6585 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed!
6588 vector=percpu: enable percpu vector domain
6590 video= [FB] Frame buffer configuration
6591 See Documentation/fb/modedb.rst.
6593 video.brightness_switch_enabled= [ACPI]
6595 If set to 1, on receiving an ACPI notify event
6596 generated by hotkey, video driver will adjust brightness
6597 level and then send out the event to user space through
6598 the allocated input device. If set to 0, video driver
6599 will only send out the event without touching backlight
6604 [VMMIO] Memory mapped virtio (platform) device.
6606 <size>@<baseaddr>:<irq>[:<id>]
6608 <size> := size (can use standard suffixes
6610 <baseaddr> := physical base address
6611 <irq> := interrupt number (as passed to
6613 <id> := (optional) platform device id
6615 virtio_mmio.device=1K@0x100b0000:48:7
6617 Can be used multiple times for multiple devices.
6619 vga= [BOOT,X86-32] Select a particular video mode
6620 See Documentation/x86/boot.rst and
6621 Documentation/admin-guide/svga.rst.
6622 Use vga=ask for menu.
6623 This is actually a boot loader parameter; the value is
6624 passed to the kernel using a special protocol.
6626 vm_debug[=options] [KNL] Available with CONFIG_DEBUG_VM=y.
6627 May slow down system boot speed, especially when
6628 enabled on systems with a large amount of memory.
6629 All options are enabled by default, and this
6630 interface is meant to allow for selectively
6631 enabling or disabling specific virtual memory
6634 Available options are:
6635 P Enable page structure init time poisoning
6636 - Disable all of the above options
6638 vmalloc=nn[KMG] [KNL,BOOT] Forces the vmalloc area to have an exact
6639 size of <nn>. This can be used to increase the
6640 minimum size (128MB on x86). It can also be used to
6641 decrease the size and leave more room for directly
6644 vmcp_cma=nn[MG] [KNL,S390]
6645 Sets the memory size reserved for contiguous memory
6646 allocations for the vmcp device driver.
6648 vmhalt= [KNL,S390] Perform z/VM CP command after system halt.
6651 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic.
6654 vmpoff= [KNL,S390] Perform z/VM CP command after power off.
6658 Controls the behavior of vsyscalls (i.e. calls to
6659 fixed addresses of 0xffffffffff600x00 from legacy
6660 code). Most statically-linked binaries and older
6661 versions of glibc use these calls. Because these
6662 functions are at fixed addresses, they make nice
6663 targets for exploits that can control RIP.
6665 emulate [default] Vsyscalls turn into traps and are
6666 emulated reasonably safely. The vsyscall
6669 xonly Vsyscalls turn into traps and are
6670 emulated reasonably safely. The vsyscall
6671 page is not readable.
6673 none Vsyscalls don't work at all. This makes
6674 them quite hard to use for exploits but
6675 might break your system.
6677 vt.color= [VT] Default text color.
6678 Format: 0xYX, X = foreground, Y = background.
6679 Default: 0x07 = light gray on black.
6681 vt.cur_default= [VT] Default cursor shape.
6682 Format: 0xCCBBAA, where AA, BB, and CC are the same as
6683 the parameters of the <Esc>[?A;B;Cc escape sequence;
6684 see VGA-softcursor.txt. Default: 2 = underline.
6686 vt.default_blu= [VT]
6687 Format: <blue0>,<blue1>,<blue2>,...,<blue15>
6688 Change the default blue palette of the console.
6689 This is a 16-member array composed of values
6692 vt.default_grn= [VT]
6693 Format: <green0>,<green1>,<green2>,...,<green15>
6694 Change the default green palette of the console.
6695 This is a 16-member array composed of values
6698 vt.default_red= [VT]
6699 Format: <red0>,<red1>,<red2>,...,<red15>
6700 Change the default red palette of the console.
6701 This is a 16-member array composed of values
6707 Set system-wide default UTF-8 mode for all tty's.
6708 Default is 1, i.e. UTF-8 mode is enabled for all
6709 newly opened terminals.
6711 vt.global_cursor_default=
6714 Set system-wide default for whether a cursor
6715 is shown on new VTs. Default is -1,
6716 i.e. cursors will be created by default unless
6717 overridden by individual drivers. 0 will hide
6718 cursors, 1 will display them.
6720 vt.italic= [VT] Default color for italic text; 0-15.
6723 vt.underline= [VT] Default color for underlined text; 0-15.
6726 watchdog timers [HW,WDT] For information on watchdog timers,
6727 see Documentation/watchdog/watchdog-parameters.rst
6728 or other driver-specific files in the
6729 Documentation/watchdog/ directory.
6733 Set the hard lockup detector stall duration
6734 threshold in seconds. The soft lockup detector
6735 threshold is set to twice the value. A value of 0
6736 disables both lockup detectors. Default is 10
6739 workqueue.watchdog_thresh=
6740 If CONFIG_WQ_WATCHDOG is configured, workqueue can
6741 warn stall conditions and dump internal state to
6742 help debugging. 0 disables workqueue stall
6743 detection; otherwise, it's the stall threshold
6744 duration in seconds. The default value is 30 and
6745 it can be updated at runtime by writing to the
6746 corresponding sysfs file.
6748 workqueue.disable_numa
6749 By default, all work items queued to unbound
6750 workqueues are affine to the NUMA nodes they're
6751 issued on, which results in better behavior in
6752 general. If NUMA affinity needs to be disabled for
6753 whatever reason, this option can be used. Note
6754 that this also can be controlled per-workqueue for
6755 workqueues visible under /sys/bus/workqueue/.
6757 workqueue.power_efficient
6758 Per-cpu workqueues are generally preferred because
6759 they show better performance thanks to cache
6760 locality; unfortunately, per-cpu workqueues tend to
6761 be more power hungry than unbound workqueues.
6763 Enabling this makes the per-cpu workqueues which
6764 were observed to contribute significantly to power
6765 consumption unbound, leading to measurably lower
6766 power usage at the cost of small performance
6769 The default value of this parameter is determined by
6770 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT.
6772 workqueue.debug_force_rr_cpu
6773 Workqueue used to implicitly guarantee that work
6774 items queued without explicit CPU specified are put
6775 on the local CPU. This guarantee is no longer true
6776 and while local CPU is still preferred work items
6777 may be put on foreign CPUs. This debug option
6778 forces round-robin CPU selection to flush out
6779 usages which depend on the now broken guarantee.
6780 When enabled, memory and cache locality will be
6783 x2apic_phys [X86-64,APIC] Use x2apic physical mode instead of
6784 default x2apic cluster mode on platforms
6787 xen_512gb_limit [KNL,X86-64,XEN]
6788 Restricts the kernel running paravirtualized under Xen
6789 to use only up to 512 GB of RAM. The reason to do so is
6790 crash analysis tools and Xen tools for doing domain
6791 save/restore/migration must be enabled to handle larger
6794 xen_emul_unplug= [HW,X86,XEN]
6795 Unplug Xen emulated devices
6796 Format: [unplug0,][unplug1]
6797 ide-disks -- unplug primary master IDE devices
6798 aux-ide-disks -- unplug non-primary-master IDE devices
6799 nics -- unplug network devices
6800 all -- unplug all emulated devices (NICs and IDE disks)
6801 unnecessary -- unplugging emulated devices is
6802 unnecessary even if the host did not respond to
6804 never -- do not unplug even if version check succeeds
6806 xen_legacy_crash [X86,XEN]
6807 Crash from Xen panic notifier, without executing late
6808 panic() code such as dumping handler.
6810 xen_nopvspin [X86,XEN]
6811 Disables the qspinlock slowpath using Xen PV optimizations.
6812 This parameter is obsoleted by "nopvspin" parameter, which
6813 has equivalent effect for XEN platform.
6816 Disables the PV optimizations forcing the HVM guest to
6817 run as generic HVM guest with no PV drivers.
6818 This option is obsoleted by the "nopv" option, which
6819 has equivalent effect for XEN platform.
6821 xen_no_vector_callback
6822 [KNL,X86,XEN] Disable the vector callback for Xen
6823 event channel interrupts.
6825 xen_scrub_pages= [XEN]
6826 Boolean option to control scrubbing pages before giving them back
6827 to Xen, for use by other domains. Can be also changed at runtime
6828 with /sys/devices/system/xen_memory/xen_memory0/scrub_pages.
6829 Default value controlled with CONFIG_XEN_SCRUB_PAGES_DEFAULT.
6831 xen_timer_slop= [X86-64,XEN]
6832 Set the timer slop (in nanoseconds) for the virtual Xen
6833 timers (default is 100000). This adjusts the minimum
6834 delta of virtualized Xen timers, where lower values
6835 improve timer resolution at the expense of processing
6836 more timer interrupts.
6838 xen.balloon_boot_timeout= [XEN]
6839 The time (in seconds) to wait before giving up to boot
6840 in case initial ballooning fails to free enough memory.
6841 Applies only when running as HVM or PVH guest and
6842 started with less memory configured than allowed at
6843 max. Default is 180.
6845 xen.event_eoi_delay= [XEN]
6846 How long to delay EOI handling in case of event
6847 storms (jiffies). Default is 10.
6849 xen.event_loop_timeout= [XEN]
6850 After which time (jiffies) the event handling loop
6851 should start to delay EOI handling. Default is 2.
6853 xen.fifo_events= [XEN]
6854 Boolean parameter to disable using fifo event handling
6855 even if available. Normally fifo event handling is
6856 preferred over the 2-level event handling, as it is
6857 fairer and the number of possible event channels is
6858 much higher. Default is on (use fifo events).
6860 nopv= [X86,XEN,KVM,HYPER_V,VMWARE]
6861 Disables the PV optimizations forcing the guest to run
6862 as generic guest with no PV drivers. Currently support
6863 XEN HVM, KVM, HYPER_V and VMWARE guest.
6865 nopvspin [X86,XEN,KVM]
6866 Disables the qspinlock slow path using PV optimizations
6867 which allow the hypervisor to 'idle' the guest on lock
6870 xirc2ps_cs= [NET,PCMCIA]
6872 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]]
6875 By default on POWER9 and above, the kernel will
6876 natively use the XIVE interrupt controller. This option
6877 allows the fallback firmware mode to be used:
6879 off Fallback to firmware control of XIVE interrupt
6880 controller on both pseries and powernv
6881 platforms. Only useful on POWER9 and above.
6883 xive.store-eoi=off [PPC]
6884 By default on POWER10 and above, the kernel will use
6885 stores for EOI handling when the XIVE interrupt mode
6886 is active. This option allows the XIVE driver to use
6887 loads instead, as on POWER9.
6889 xhci-hcd.quirks [USB,KNL]
6890 A hex value specifying bitmask with supplemental xhci
6891 host controller quirks. Meaning of each bit can be
6892 consulted in header drivers/usb/host/xhci.h.
6895 Format: { early | on | rw | ro | off }
6896 Controls if xmon debugger is enabled. Default is off.
6897 Passing only "xmon" is equivalent to "xmon=early".
6898 early Call xmon as early as possible on boot; xmon
6899 debugger is called from setup_arch().
6900 on xmon debugger hooks will be installed so xmon
6901 is only called on a kernel crash. Default mode,
6902 i.e. either "ro" or "rw" mode, is controlled
6903 with CONFIG_XMON_DEFAULT_RO_MODE.
6904 rw xmon debugger hooks will be installed so xmon
6905 is called only on a kernel crash, mode is write,
6906 meaning SPR registers, memory and, other data
6907 can be written using xmon commands.
6908 ro same as "rw" option above but SPR registers,
6909 memory, and other data can't be written using
6911 off xmon is disabled.