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 Control if HugeTLB Vmemmap Optimization (HVO) is enabled.
1735 Allows heavy hugetlb users to free up some more
1736 memory (7 * PAGE_SIZE for each 2MB hugetlb page).
1737 Format: { on | off (default) }
1742 Built with CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP_DEFAULT_ON=y,
1745 Note that the vmemmap pages may be allocated from the added
1746 memory block itself when memory_hotplug.memmap_on_memory is
1747 enabled, those vmemmap pages cannot be optimized even if this
1748 feature is enabled. Other vmemmap pages not allocated from
1749 the added memory block itself do not be affected.
1752 [KNL] Should the hung task detector generate panics.
1755 A value of 1 instructs the kernel to panic when a
1756 hung task is detected. The default value is controlled
1757 by the CONFIG_BOOTPARAM_HUNG_TASK_PANIC build-time
1758 option. The value selected by this boot parameter can
1759 be changed later by the kernel.hung_task_panic sysctl.
1761 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC)
1762 terminal devices. Valid values: 0..8
1763 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs.
1764 If specified, z/VM IUCV HVC accepts connections
1765 from listed z/VM user IDs only.
1767 hv_nopvspin [X86,HYPER_V] Disables the paravirt spinlock optimizations
1768 which allow the hypervisor to 'idle' the
1769 guest on lock contention.
1772 Do not unregister boot console at start. This is only
1773 useful for debugging when something happens in the window
1774 between unregistering the boot console and initializing
1777 i2c_bus= [HW] Override the default board specific I2C bus speed
1778 or register an additional I2C bus that is not
1779 registered from board initialization code.
1783 i8042.debug [HW] Toggle i8042 debug mode
1784 i8042.unmask_kbd_data
1785 [HW] Enable printing of interrupt data from the KBD port
1786 (disabled by default, and as a pre-condition
1787 requires that i8042.debug=1 be enabled)
1788 i8042.direct [HW] Put keyboard port into non-translated mode
1789 i8042.dumbkbd [HW] Pretend that controller can only read data from
1790 keyboard and cannot control its state
1791 (Don't attempt to blink the leds)
1792 i8042.noaux [HW] Don't check for auxiliary (== mouse) port
1793 i8042.nokbd [HW] Don't check/create keyboard port
1794 i8042.noloop [HW] Disable the AUX Loopback command while probing
1796 i8042.nomux [HW] Don't check presence of an active multiplexing
1798 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX
1800 i8042.notimeout [HW] Ignore timeout condition signalled by controller
1801 i8042.reset [HW] Reset the controller during init, cleanup and
1802 suspend-to-ram transitions, only during s2r
1803 transitions, or never reset
1804 Format: { 1 | Y | y | 0 | N | n }
1805 1, Y, y: always reset controller
1806 0, N, n: don't ever reset controller
1807 Default: only on s2r transitions on x86; most other
1808 architectures force reset to be always executed
1809 i8042.unlock [HW] Unlock (ignore) the keylock
1810 i8042.kbdreset [HW] Reset device connected to KBD port
1812 [HW] Allow deferred probing upon i8042 probe errors
1816 i915.invert_brightness=
1817 [DRM] Invert the sense of the variable that is used to
1818 set the brightness of the panel backlight. Normally a
1819 brightness value of 0 indicates backlight switched off,
1820 and the maximum of the brightness value sets the backlight
1821 to maximum brightness. If this parameter is set to 0
1822 (default) and the machine requires it, or this parameter
1823 is set to 1, a brightness value of 0 sets the backlight
1824 to maximum brightness, and the maximum of the brightness
1825 value switches the backlight off.
1826 -1 -- never invert brightness
1827 0 -- machine default
1828 1 -- force brightness inversion
1831 Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]]
1835 Format: idle=poll, idle=halt, idle=nomwait
1836 Poll forces a polling idle loop that can slightly
1837 improve the performance of waking up a idle CPU, but
1838 will use a lot of power and make the system run hot.
1840 idle=halt: Halt is forced to be used for CPU idle.
1841 In such case C2/C3 won't be used again.
1842 idle=nomwait: Disable mwait for CPU C-states
1846 Allow force disabling of Shared Virtual Memory (SVA)
1847 support for the idxd driver. By default it is set to
1850 idxd.tc_override= [HW]
1852 Allow override of default traffic class configuration
1853 for the device. By default it is set to false (0).
1855 ieee754= [MIPS] Select IEEE Std 754 conformance mode
1856 Format: { strict | legacy | 2008 | relaxed }
1859 Choose which programs will be accepted for execution
1860 based on the IEEE 754 NaN encoding(s) supported by
1861 the FPU and the NaN encoding requested with the value
1862 of an ELF file header flag individually set by each
1863 binary. Hardware implementations are permitted to
1864 support either or both of the legacy and the 2008 NaN
1867 Available settings are as follows:
1868 strict accept binaries that request a NaN encoding
1869 supported by the FPU
1870 legacy only accept legacy-NaN binaries, if supported
1872 2008 only accept 2008-NaN binaries, if supported
1874 relaxed accept any binaries regardless of whether
1875 supported by the FPU
1877 The FPU emulator is always able to support both NaN
1878 encodings, so if no FPU hardware is present or it has
1879 been disabled with 'nofpu', then the settings of
1880 'legacy' and '2008' strap the emulator accordingly,
1881 'relaxed' straps the emulator for both legacy-NaN and
1882 2008-NaN, whereas 'strict' enables legacy-NaN only on
1883 legacy processors and both NaN encodings on MIPS32 or
1886 The setting for ABS.fmt/NEG.fmt instruction execution
1887 mode generally follows that for the NaN encoding,
1888 except where unsupported by hardware.
1890 ignore_loglevel [KNL]
1891 Ignore loglevel setting - this will print /all/
1892 kernel messages to the console. Useful for debugging.
1893 We also add it as printk module parameter, so users
1894 could change it dynamically, usually by
1895 /sys/module/printk/parameters/ignore_loglevel.
1898 Ignore RLIMIT_DATA setting for data mappings,
1899 print warning at first misuse. Can be changed via
1900 /sys/module/kernel/parameters/ignore_rlimit_data.
1902 ihash_entries= [KNL]
1903 Set number of hash buckets for inode cache.
1905 ima_appraise= [IMA] appraise integrity measurements
1906 Format: { "off" | "enforce" | "fix" | "log" }
1909 ima_appraise_tcb [IMA] Deprecated. Use ima_policy= instead.
1910 The builtin appraise policy appraises all files
1913 ima_canonical_fmt [IMA]
1914 Use the canonical format for the binary runtime
1915 measurements, instead of host native format.
1918 Format: { md5 | sha1 | rmd160 | sha256 | sha384
1922 The list of supported hash algorithms is defined
1923 in crypto/hash_info.h.
1926 The builtin policies to load during IMA setup.
1927 Format: "tcb | appraise_tcb | secure_boot |
1928 fail_securely | critical_data"
1930 The "tcb" policy measures all programs exec'd, files
1931 mmap'd for exec, and all files opened with the read
1932 mode bit set by either the effective uid (euid=0) or
1935 The "appraise_tcb" policy appraises the integrity of
1936 all files owned by root.
1938 The "secure_boot" policy appraises the integrity
1939 of files (eg. kexec kernel image, kernel modules,
1940 firmware, policy, etc) based on file signatures.
1942 The "fail_securely" policy forces file signature
1943 verification failure also on privileged mounted
1944 filesystems with the SB_I_UNVERIFIABLE_SIGNATURE
1947 The "critical_data" policy measures kernel integrity
1950 ima_tcb [IMA] Deprecated. Use ima_policy= instead.
1951 Load a policy which meets the needs of the Trusted
1952 Computing Base. This means IMA will measure all
1953 programs exec'd, files mmap'd for exec, and all files
1954 opened for read by uid=0.
1957 Select one of defined IMA measurements template formats.
1958 Formats: { "ima" | "ima-ng" | "ima-ngv2" | "ima-sig" |
1963 [IMA] Define a custom template format.
1964 Format: { "field1|...|fieldN" }
1966 ima.ahash_minsize= [IMA] Minimum file size for asynchronous hash usage
1967 Format: <min_file_size>
1968 Set the minimal file size for using asynchronous hash.
1969 If left unspecified, ahash usage is disabled.
1971 ahash performance varies for different data sizes on
1972 different crypto accelerators. This option can be used
1973 to achieve the best performance for a particular HW.
1975 ima.ahash_bufsize= [IMA] Asynchronous hash buffer size
1977 Set hashing buffer size. Default: 4k.
1979 ahash performance varies for different chunk sizes on
1980 different crypto accelerators. This option can be used
1981 to achieve best performance for particular HW.
1985 Run specified binary instead of /sbin/init as init
1988 initcall_debug [KNL] Trace initcalls as they are executed. Useful
1989 for working out where the kernel is dying during
1992 initcall_blacklist= [KNL] Do not execute a comma-separated list of
1993 initcall functions. Useful for debugging built-in
1994 modules and initcalls.
1996 initramfs_async= [KNL]
1999 This parameter controls whether the initramfs
2000 image is unpacked asynchronously, concurrently
2001 with devices being probed and
2002 initialized. This should normally just work,
2003 but as a debugging aid, one can get the
2004 historical behaviour of the initramfs
2005 unpacking being completed before device_ and
2008 initrd= [BOOT] Specify the location of the initial ramdisk
2010 initrdmem= [KNL] Specify a physical address and size from which to
2011 load the initrd. If an initrd is compiled in or
2012 specified in the bootparams, it takes priority over this
2014 Format: ss[KMG],nn[KMG]
2017 init_on_alloc= [MM] Fill newly allocated pages and heap objects with
2020 Default set by CONFIG_INIT_ON_ALLOC_DEFAULT_ON.
2022 init_on_free= [MM] Fill freed pages and heap objects with zeroes.
2024 Default set by CONFIG_INIT_ON_FREE_DEFAULT_ON.
2026 init_pkru= [X86] Specify the default memory protection keys rights
2027 register contents for all processes. 0x55555554 by
2028 default (disallow access to all but pkey 0). Can
2029 override in debugfs after boot.
2031 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver
2034 int_pln_enable [X86] Enable power limit notification interrupt
2036 integrity_audit=[IMA]
2037 Format: { "0" | "1" }
2038 0 -- basic integrity auditing messages. (Default)
2039 1 -- additional integrity auditing messages.
2041 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option
2043 Enable intel iommu driver.
2045 Disable intel iommu driver.
2046 igfx_off [Default Off]
2047 By default, gfx is mapped as normal device. If a gfx
2048 device has a dedicated DMAR unit, the DMAR unit is
2049 bypassed by not enabling DMAR with this option. In
2050 this case, gfx device will use physical address for
2052 strict [Default Off]
2053 Deprecated, equivalent to iommu.strict=1.
2054 sp_off [Default Off]
2055 By default, super page will be supported if Intel IOMMU
2056 has the capability. With this option, super page will
2059 Enable the Intel IOMMU scalable mode if the hardware
2060 advertises that it has support for the scalable mode
2063 Disallow use of the Intel IOMMU scalable mode.
2064 tboot_noforce [Default Off]
2065 Do not force the Intel IOMMU enabled under tboot.
2066 By default, tboot will force Intel IOMMU on, which
2067 could harm performance of some high-throughput
2068 devices like 40GBit network cards, even if identity
2070 Note that using this option lowers the security
2071 provided by tboot because it makes the system
2072 vulnerable to DMA attacks.
2074 intel_idle.max_cstate= [KNL,HW,ACPI,X86]
2075 0 disables intel_idle and fall back on acpi_idle.
2076 1 to 9 specify maximum depth of C-state.
2080 Do not enable intel_pstate as the default
2081 scaling driver for the supported processors
2083 Use intel_pstate as a scaling driver, but configure it
2084 to work with generic cpufreq governors (instead of
2085 enabling its internal governor). This mode cannot be
2086 used along with the hardware-managed P-states (HWP)
2089 Enable intel_pstate on systems that prohibit it by default
2090 in favor of acpi-cpufreq. Forcing the intel_pstate driver
2091 instead of acpi-cpufreq may disable platform features, such
2092 as thermal controls and power capping, that rely on ACPI
2093 P-States information being indicated to OSPM and therefore
2094 should be used with caution. This option does not work with
2095 processors that aren't supported by the intel_pstate driver
2096 or on platforms that use pcc-cpufreq instead of acpi-cpufreq.
2098 Do not enable hardware P state control (HWP)
2101 Only load intel_pstate on systems which support
2102 hardware P state control (HWP) if available.
2104 Enforce ACPI _PPC performance limits. If the Fixed ACPI
2105 Description Table, specifies preferred power management
2106 profile as "Enterprise Server" or "Performance Server",
2107 then this feature is turned on by default.
2109 Allow per-logical-CPU P-State performance control limits using
2110 cpufreq sysfs interface
2112 intremap= [X86-64, Intel-IOMMU]
2113 on enable Interrupt Remapping (default)
2114 off disable Interrupt Remapping
2115 nosid disable Source ID checking
2117 BIOS x2APIC opt-out request will be ignored
2118 nopost disable Interrupt Posting
2120 iomem= Disable strict checking of access to MMIO memory
2121 strict regions from userspace.
2136 nobypass [PPC/POWERNV]
2137 Disable IOMMU bypass, using IOMMU for PCI devices.
2139 iommu.forcedac= [ARM64, X86] Control IOVA allocation for PCI devices.
2140 Format: { "0" | "1" }
2141 0 - Try to allocate a 32-bit DMA address first, before
2142 falling back to the full range if needed.
2143 1 - Allocate directly from the full usable range,
2144 forcing Dual Address Cycle for PCI cards supporting
2145 greater than 32-bit addressing.
2147 iommu.strict= [ARM64, X86] Configure TLB invalidation behaviour
2148 Format: { "0" | "1" }
2150 Request that DMA unmap operations use deferred
2151 invalidation of hardware TLBs, for increased
2152 throughput at the cost of reduced device isolation.
2153 Will fall back to strict mode if not supported by
2154 the relevant IOMMU driver.
2156 DMA unmap operations invalidate IOMMU hardware TLBs
2158 unset - Use value of CONFIG_IOMMU_DEFAULT_DMA_{LAZY,STRICT}.
2159 Note: on x86, strict mode specified via one of the
2160 legacy driver-specific options takes precedence.
2163 [ARM64, X86] Configure DMA to bypass the IOMMU by default.
2164 Format: { "0" | "1" }
2165 0 - Use IOMMU translation for DMA.
2166 1 - Bypass the IOMMU for DMA.
2167 unset - Use value of CONFIG_IOMMU_DEFAULT_PASSTHROUGH.
2169 io7= [HW] IO7 for Marvel-based Alpha systems
2170 See comment before marvel_specify_io7 in
2171 arch/alpha/kernel/core_marvel.c.
2173 io_delay= [X86] I/O delay method
2175 Standard port 0x80 based delay
2177 Alternate port 0xed based delay (needed on some systems)
2179 Simple two microseconds delay
2184 See Documentation/admin-guide/nfs/nfsroot.rst.
2186 ipcmni_extend [KNL] Extend the maximum number of unique System V
2187 IPC identifiers from 32,768 to 16,777,216.
2189 irqaffinity= [SMP] Set the default irq affinity mask
2190 The argument is a cpu list, as described above.
2192 irqchip.gicv2_force_probe=
2195 Force the kernel to look for the second 4kB page
2196 of a GICv2 controller even if the memory range
2197 exposed by the device tree is too small.
2199 irqchip.gicv3_nolpi=
2201 Force the kernel to ignore the availability of
2202 LPIs (and by consequence ITSs). Intended for system
2203 that use the kernel as a bootloader, and thus want
2204 to let secondary kernels in charge of setting up
2207 irqchip.gicv3_pseudo_nmi= [ARM64]
2208 Enables support for pseudo-NMIs in the kernel. This
2209 requires the kernel to be built with
2210 CONFIG_ARM64_PSEUDO_NMI.
2213 When an interrupt is not handled search all handlers
2214 for it. Intended to get systems with badly broken
2218 When an interrupt is not handled search all handlers
2219 for it. Also check all handlers each timer
2220 interrupt. Intended to get systems with badly broken
2224 Format: <RDP>,<reset>,<pci_scan>,<verbosity>
2226 isolcpus= [KNL,SMP,ISOL] Isolate a given set of CPUs from disturbance.
2227 [Deprecated - use cpusets instead]
2228 Format: [flag-list,]<cpu-list>
2230 Specify one or more CPUs to isolate from disturbances
2231 specified in the flag list (default: domain):
2234 Disable the tick when a single task runs.
2236 A residual 1Hz tick is offloaded to workqueues, which you
2237 need to affine to housekeeping through the global
2238 workqueue's affinity configured via the
2239 /sys/devices/virtual/workqueue/cpumask sysfs file, or
2240 by using the 'domain' flag described below.
2242 NOTE: by default the global workqueue runs on all CPUs,
2243 so to protect individual CPUs the 'cpumask' file has to
2244 be configured manually after bootup.
2247 Isolate from the general SMP balancing and scheduling
2248 algorithms. Note that performing domain isolation this way
2249 is irreversible: it's not possible to bring back a CPU to
2250 the domains once isolated through isolcpus. It's strongly
2251 advised to use cpusets instead to disable scheduler load
2252 balancing through the "cpuset.sched_load_balance" file.
2253 It offers a much more flexible interface where CPUs can
2254 move in and out of an isolated set anytime.
2256 You can move a process onto or off an "isolated" CPU via
2257 the CPU affinity syscalls or cpuset.
2258 <cpu number> begins at 0 and the maximum value is
2259 "number of CPUs in system - 1".
2263 Isolate from being targeted by managed interrupts
2264 which have an interrupt mask containing isolated
2265 CPUs. The affinity of managed interrupts is
2266 handled by the kernel and cannot be changed via
2267 the /proc/irq/* interfaces.
2269 This isolation is best effort and only effective
2270 if the automatically assigned interrupt mask of a
2271 device queue contains isolated and housekeeping
2272 CPUs. If housekeeping CPUs are online then such
2273 interrupts are directed to the housekeeping CPU
2274 so that IO submitted on the housekeeping CPU
2275 cannot disturb the isolated CPU.
2277 If a queue's affinity mask contains only isolated
2278 CPUs then this parameter has no effect on the
2279 interrupt routing decision, though interrupts are
2280 only delivered when tasks running on those
2281 isolated CPUs submit IO. IO submitted on
2282 housekeeping CPUs has no influence on those
2285 The format of <cpu-list> is described above.
2289 ivrs_ioapic [HW,X86-64]
2290 Provide an override to the IOAPIC-ID<->DEVICE-ID
2291 mapping provided in the IVRS ACPI table.
2292 By default, PCI segment is 0, and can be omitted.
2294 * To map IOAPIC-ID decimal 10 to PCI device 00:14.0
2295 write the parameter as:
2296 ivrs_ioapic[10]=00:14.0
2297 * To map IOAPIC-ID decimal 10 to PCI segment 0x1 and
2298 PCI device 00:14.0 write the parameter as:
2299 ivrs_ioapic[10]=0001:00:14.0
2301 ivrs_hpet [HW,X86-64]
2302 Provide an override to the HPET-ID<->DEVICE-ID
2303 mapping provided in the IVRS ACPI table.
2304 By default, PCI segment is 0, and can be omitted.
2306 * To map HPET-ID decimal 0 to PCI device 00:14.0
2307 write the parameter as:
2308 ivrs_hpet[0]=00:14.0
2309 * To map HPET-ID decimal 10 to PCI segment 0x1 and
2310 PCI device 00:14.0 write the parameter as:
2311 ivrs_ioapic[10]=0001:00:14.0
2313 ivrs_acpihid [HW,X86-64]
2314 Provide an override to the ACPI-HID:UID<->DEVICE-ID
2315 mapping provided in the IVRS ACPI table.
2317 For example, to map UART-HID:UID AMD0020:0 to
2318 PCI segment 0x1 and PCI device ID 00:14.5,
2319 write the parameter as:
2320 ivrs_acpihid[0001:00:14.5]=AMD0020:0
2322 By default, PCI segment is 0, and can be omitted.
2323 For example, PCI device 00:14.5 write the parameter as:
2324 ivrs_acpihid[00:14.5]=AMD0020:0
2326 js= [HW,JOY] Analog joystick
2327 See Documentation/input/joydev/joystick.rst.
2330 When CONFIG_RANDOMIZE_BASE is set, this disables
2331 kernel and module base offset ASLR (Address Space
2332 Layout Randomization).
2335 [KNL] Enforce KASAN (Kernel Address Sanitizer) to print
2336 report on every invalid memory access. Without this
2337 parameter KASAN will print report only for the first
2342 kernelcore= [KNL,X86,IA-64,PPC]
2343 Format: nn[KMGTPE] | nn% | "mirror"
2344 This parameter specifies the amount of memory usable by
2345 the kernel for non-movable allocations. The requested
2346 amount is spread evenly throughout all nodes in the
2347 system as ZONE_NORMAL. The remaining memory is used for
2348 movable memory in its own zone, ZONE_MOVABLE. In the
2349 event, a node is too small to have both ZONE_NORMAL and
2350 ZONE_MOVABLE, kernelcore memory will take priority and
2351 other nodes will have a larger ZONE_MOVABLE.
2353 ZONE_MOVABLE is used for the allocation of pages that
2354 may be reclaimed or moved by the page migration
2355 subsystem. Note that allocations like PTEs-from-HighMem
2356 still use the HighMem zone if it exists, and the Normal
2357 zone if it does not.
2359 It is possible to specify the exact amount of memory in
2360 the form of "nn[KMGTPE]", a percentage of total system
2361 memory in the form of "nn%", or "mirror". If "mirror"
2362 option is specified, mirrored (reliable) memory is used
2363 for non-movable allocations and remaining memory is used
2364 for Movable pages. "nn[KMGTPE]", "nn%", and "mirror"
2365 are exclusive, so you cannot specify multiple forms.
2367 kgdbdbgp= [KGDB,HW] kgdb over EHCI usb debug port.
2368 Format: <Controller#>[,poll interval]
2369 The controller # is the number of the ehci usb debug
2370 port as it is probed via PCI. The poll interval is
2371 optional and is the number seconds in between
2372 each poll cycle to the debug port in case you need
2373 the functionality for interrupting the kernel with
2374 gdb or control-c on the dbgp connection. When
2375 not using this parameter you use sysrq-g to break into
2376 the kernel debugger.
2378 kgdboc= [KGDB,HW] kgdb over consoles.
2379 Requires a tty driver that supports console polling,
2380 or a supported polling keyboard driver (non-usb).
2381 Serial only format: <serial_device>[,baud]
2382 keyboard only format: kbd
2383 keyboard and serial format: kbd,<serial_device>[,baud]
2384 Optional Kernel mode setting:
2385 kms, kbd format: kms,kbd
2386 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud]
2388 kgdboc_earlycon= [KGDB,HW]
2389 If the boot console provides the ability to read
2390 characters and can work in polling mode, you can use
2391 this parameter to tell kgdb to use it as a backend
2392 until the normal console is registered. Intended to
2393 be used together with the kgdboc parameter which
2394 specifies the normal console to transition to.
2396 The name of the early console should be specified
2397 as the value of this parameter. Note that the name of
2398 the early console might be different than the tty
2399 name passed to kgdboc. It's OK to leave the value
2400 blank and the first boot console that implements
2401 read() will be picked.
2403 kgdbwait [KGDB] Stop kernel execution and enter the
2404 kernel debugger at the earliest opportunity.
2406 kmac= [MIPS] Korina ethernet MAC address.
2407 Configure the RouterBoard 532 series on-chip
2408 Ethernet adapter MAC address.
2410 kmemleak= [KNL] Boot-time kmemleak enable/disable
2411 Valid arguments: on, off
2413 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y,
2416 kprobe_event=[probe-list]
2417 [FTRACE] Add kprobe events and enable at boot time.
2418 The probe-list is a semicolon delimited list of probe
2419 definitions. Each definition is same as kprobe_events
2420 interface, but the parameters are comma delimited.
2421 For example, to add a kprobe event on vfs_read with
2422 arg1 and arg2, add to the command line;
2424 kprobe_event=p,vfs_read,$arg1,$arg2
2426 See also Documentation/trace/kprobetrace.rst "Kernel
2427 Boot Parameter" section.
2429 kpti= [ARM64] Control page table isolation of user
2430 and kernel address spaces.
2431 Default: enabled on cores which need mitigation.
2435 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs.
2436 Default is 0 (don't ignore, but inject #GP)
2438 kvm.eager_page_split=
2439 [KVM,X86] Controls whether or not KVM will try to
2440 proactively split all huge pages during dirty logging.
2441 Eager page splitting reduces interruptions to vCPU
2442 execution by eliminating the write-protection faults
2443 and MMU lock contention that would otherwise be
2444 required to split huge pages lazily.
2446 VM workloads that rarely perform writes or that write
2447 only to a small region of VM memory may benefit from
2448 disabling eager page splitting to allow huge pages to
2449 still be used for reads.
2451 The behavior of eager page splitting depends on whether
2452 KVM_DIRTY_LOG_INITIALLY_SET is enabled or disabled. If
2453 disabled, all huge pages in a memslot will be eagerly
2454 split when dirty logging is enabled on that memslot. If
2455 enabled, eager page splitting will be performed during
2456 the KVM_CLEAR_DIRTY ioctl, and only for the pages being
2459 Eager page splitting is only supported when kvm.tdp_mmu=Y.
2463 kvm.enable_vmware_backdoor=[KVM] Support VMware backdoor PV interface.
2464 Default is false (don't support).
2467 [KVM] Controls the software workaround for the
2468 X86_BUG_ITLB_MULTIHIT bug.
2469 force : Always deploy workaround.
2470 off : Never deploy workaround.
2471 auto : Deploy workaround based on the presence of
2472 X86_BUG_ITLB_MULTIHIT.
2476 If the software workaround is enabled for the host,
2477 guests do need not to enable it for nested guests.
2479 kvm.nx_huge_pages_recovery_ratio=
2480 [KVM] Controls how many 4KiB pages are periodically zapped
2481 back to huge pages. 0 disables the recovery, otherwise if
2482 the value is N KVM will zap 1/Nth of the 4KiB pages every
2483 period (see below). The default is 60.
2485 kvm.nx_huge_pages_recovery_period_ms=
2486 [KVM] Controls the time period at which KVM zaps 4KiB pages
2487 back to huge pages. If the value is a non-zero N, KVM will
2488 zap a portion (see ratio above) of the pages every N msecs.
2489 If the value is 0 (the default), KVM will pick a period based
2490 on the ratio, such that a page is zapped after 1 hour on average.
2492 kvm-amd.nested= [KVM,AMD] Allow nested virtualization in KVM/SVM.
2493 Default is 1 (enabled)
2495 kvm-amd.npt= [KVM,AMD] Disable nested paging (virtualized MMU)
2497 Default is 1 (enabled) if in 64-bit or 32-bit PAE mode.
2500 [KVM,ARM] Select one of KVM/arm64's modes of operation.
2502 none: Forcefully disable KVM.
2504 nvhe: Standard nVHE-based mode, without support for
2507 protected: nVHE-based mode with support for guests whose
2508 state is kept private from the host.
2510 Defaults to VHE/nVHE based on hardware support. Setting
2511 mode to "protected" will disable kexec and hibernation
2514 kvm-arm.vgic_v3_group0_trap=
2515 [KVM,ARM] Trap guest accesses to GICv3 group-0
2518 kvm-arm.vgic_v3_group1_trap=
2519 [KVM,ARM] Trap guest accesses to GICv3 group-1
2522 kvm-arm.vgic_v3_common_trap=
2523 [KVM,ARM] Trap guest accesses to GICv3 common
2526 kvm-arm.vgic_v4_enable=
2527 [KVM,ARM] Allow use of GICv4 for direct injection of
2530 kvm_cma_resv_ratio=n [PPC]
2531 Reserves given percentage from system memory area for
2532 contiguous memory allocation for KVM hash pagetable
2534 By default it reserves 5% of total system memory.
2538 kvm-intel.ept= [KVM,Intel] Disable extended page tables
2539 (virtualized MMU) support on capable Intel chips.
2540 Default is 1 (enabled)
2542 kvm-intel.emulate_invalid_guest_state=
2543 [KVM,Intel] Disable emulation of invalid guest state.
2544 Ignored if kvm-intel.enable_unrestricted_guest=1, as
2545 guest state is never invalid for unrestricted guests.
2546 This param doesn't apply to nested guests (L2), as KVM
2547 never emulates invalid L2 guest state.
2548 Default is 1 (enabled)
2550 kvm-intel.flexpriority=
2551 [KVM,Intel] Disable FlexPriority feature (TPR shadow).
2552 Default is 1 (enabled)
2555 [KVM,Intel] Enable VMX nesting (nVMX).
2556 Default is 0 (disabled)
2558 kvm-intel.unrestricted_guest=
2559 [KVM,Intel] Disable unrestricted guest feature
2560 (virtualized real and unpaged mode) on capable
2561 Intel chips. Default is 1 (enabled)
2563 kvm-intel.vmentry_l1d_flush=[KVM,Intel] Mitigation for L1 Terminal Fault
2566 Valid arguments: never, cond, always
2568 always: L1D cache flush on every VMENTER.
2569 cond: Flush L1D on VMENTER only when the code between
2570 VMEXIT and VMENTER can leak host memory.
2571 never: Disables the mitigation
2573 Default is cond (do L1 cache flush in specific instances)
2575 kvm-intel.vpid= [KVM,Intel] Disable Virtual Processor Identification
2576 feature (tagged TLBs) on capable Intel chips.
2577 Default is 1 (enabled)
2579 l1d_flush= [X86,INTEL]
2580 Control mitigation for L1D based snooping vulnerability.
2582 Certain CPUs are vulnerable to an exploit against CPU
2583 internal buffers which can forward information to a
2584 disclosure gadget under certain conditions.
2586 In vulnerable processors, the speculatively
2587 forwarded data can be used in a cache side channel
2588 attack, to access data to which the attacker does
2589 not have direct access.
2591 This parameter controls the mitigation. The
2594 on - enable the interface for the mitigation
2596 l1tf= [X86] Control mitigation of the L1TF vulnerability on
2599 The kernel PTE inversion protection is unconditionally
2600 enabled and cannot be disabled.
2603 Provides all available mitigations for the
2604 L1TF vulnerability. Disables SMT and
2605 enables all mitigations in the
2606 hypervisors, i.e. unconditional L1D flush.
2608 SMT control and L1D flush control via the
2609 sysfs interface is still possible after
2610 boot. Hypervisors will issue a warning
2611 when the first VM is started in a
2612 potentially insecure configuration,
2613 i.e. SMT enabled or L1D flush disabled.
2616 Same as 'full', but disables SMT and L1D
2617 flush runtime control. Implies the
2618 'nosmt=force' command line option.
2619 (i.e. sysfs control of SMT is disabled.)
2622 Leaves SMT enabled and enables the default
2623 hypervisor mitigation, i.e. conditional
2626 SMT control and L1D flush control via the
2627 sysfs interface is still possible after
2628 boot. Hypervisors will issue a warning
2629 when the first VM is started in a
2630 potentially insecure configuration,
2631 i.e. SMT enabled or L1D flush disabled.
2635 Disables SMT and enables the default
2636 hypervisor mitigation.
2638 SMT control and L1D flush control via the
2639 sysfs interface is still possible after
2640 boot. Hypervisors will issue a warning
2641 when the first VM is started in a
2642 potentially insecure configuration,
2643 i.e. SMT enabled or L1D flush disabled.
2646 Same as 'flush', but hypervisors will not
2647 warn when a VM is started in a potentially
2648 insecure configuration.
2651 Disables hypervisor mitigations and doesn't
2653 It also drops the swap size and available
2654 RAM limit restriction on both hypervisor and
2659 For details see: Documentation/admin-guide/hw-vuln/l1tf.rst
2665 lapic [X86-32,APIC] Enable the local APIC even if BIOS
2668 lapic= [X86,APIC] Do not use TSC deadline
2669 value for LAPIC timer one-shot implementation. Default
2670 back to the programmable timer unit in the LAPIC.
2671 Format: notscdeadline
2673 lapic_timer_c2_ok [X86,APIC] trust the local apic timer
2676 libata.dma= [LIBATA] DMA control
2677 libata.dma=0 Disable all PATA and SATA DMA
2678 libata.dma=1 PATA and SATA Disk DMA only
2679 libata.dma=2 ATAPI (CDROM) DMA only
2680 libata.dma=4 Compact Flash DMA only
2681 Combinations also work, so libata.dma=3 enables DMA
2682 for disks and CDROMs, but not CFs.
2684 libata.ignore_hpa= [LIBATA] Ignore HPA limit
2685 libata.ignore_hpa=0 keep BIOS limits (default)
2686 libata.ignore_hpa=1 ignore limits, using full disk
2688 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume
2692 libata.force= [LIBATA] Force configurations. The format is a comma-
2693 separated list of "[ID:]VAL" where ID is PORT[.DEVICE].
2694 PORT and DEVICE are decimal numbers matching port, link
2695 or device. Basically, it matches the ATA ID string
2696 printed on console by libata. If the whole ID part is
2697 omitted, the last PORT and DEVICE values are used. If
2698 ID hasn't been specified yet, the configuration applies
2699 to all ports, links and devices.
2701 If only DEVICE is omitted, the parameter applies to
2702 the port and all links and devices behind it. DEVICE
2703 number of 0 either selects the first device or the
2704 first fan-out link behind PMP device. It does not
2705 select the host link. DEVICE number of 15 selects the
2706 host link and device attached to it.
2708 The VAL specifies the configuration to force. As long
2709 as there is no ambiguity, shortcut notation is allowed.
2710 For example, both 1.5 and 1.5G would work for 1.5Gbps.
2711 The following configurations can be forced.
2713 * Cable type: 40c, 80c, short40c, unk, ign or sata.
2714 Any ID with matching PORT is used.
2716 * SATA link speed limit: 1.5Gbps or 3.0Gbps.
2718 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7].
2719 udma[/][16,25,33,44,66,100,133] notation is also
2722 * nohrst, nosrst, norst: suppress hard, soft and both
2725 * rstonce: only attempt one reset during hot-unplug
2728 * [no]dbdelay: Enable or disable the extra 200ms delay
2729 before debouncing a link PHY and device presence
2732 * [no]ncq: Turn on or off NCQ.
2734 * [no]ncqtrim: Enable or disable queued DSM TRIM.
2736 * [no]ncqati: Enable or disable NCQ trim on ATI chipset.
2738 * [no]trim: Enable or disable (unqueued) TRIM.
2740 * trim_zero: Indicate that TRIM command zeroes data.
2742 * max_trim_128m: Set 128M maximum trim size limit.
2744 * [no]dma: Turn on or off DMA transfers.
2746 * atapi_dmadir: Enable ATAPI DMADIR bridge support.
2748 * atapi_mod16_dma: Enable the use of ATAPI DMA for
2749 commands that are not a multiple of 16 bytes.
2751 * [no]dmalog: Enable or disable the use of the
2752 READ LOG DMA EXT command to access logs.
2754 * [no]iddevlog: Enable or disable access to the
2755 identify device data log.
2757 * [no]logdir: Enable or disable access to the general
2758 purpose log directory.
2760 * max_sec_128: Set transfer size limit to 128 sectors.
2762 * max_sec_1024: Set or clear transfer size limit to
2765 * max_sec_lba48: Set or clear transfer size limit to
2768 * [no]lpm: Enable or disable link power management.
2770 * [no]setxfer: Indicate if transfer speed mode setting
2773 * dump_id: Dump IDENTIFY data.
2775 * disable: Disable this device.
2777 If there are multiple matching configurations changing
2778 the same attribute, the last one is used.
2780 load_ramdisk= [RAM] [Deprecated]
2782 lockd.nlm_grace_period=P [NFS] Assign grace period.
2785 lockd.nlm_tcpport=N [NFS] Assign TCP port.
2788 lockd.nlm_timeout=T [NFS] Assign timeout value.
2791 lockd.nlm_udpport=M [NFS] Assign UDP port.
2794 lockdown= [SECURITY]
2795 { integrity | confidentiality }
2796 Enable the kernel lockdown feature. If set to
2797 integrity, kernel features that allow userland to
2798 modify the running kernel are disabled. If set to
2799 confidentiality, kernel features that allow userland
2800 to extract confidential information from the kernel
2803 locktorture.nreaders_stress= [KNL]
2804 Set the number of locking read-acquisition kthreads.
2805 Defaults to being automatically set based on the
2806 number of online CPUs.
2808 locktorture.nwriters_stress= [KNL]
2809 Set the number of locking write-acquisition kthreads.
2811 locktorture.onoff_holdoff= [KNL]
2812 Set time (s) after boot for CPU-hotplug testing.
2814 locktorture.onoff_interval= [KNL]
2815 Set time (s) between CPU-hotplug operations, or
2816 zero to disable CPU-hotplug testing.
2818 locktorture.shuffle_interval= [KNL]
2819 Set task-shuffle interval (jiffies). Shuffling
2820 tasks allows some CPUs to go into dyntick-idle
2821 mode during the locktorture test.
2823 locktorture.shutdown_secs= [KNL]
2824 Set time (s) after boot system shutdown. This
2825 is useful for hands-off automated testing.
2827 locktorture.stat_interval= [KNL]
2828 Time (s) between statistics printk()s.
2830 locktorture.stutter= [KNL]
2831 Time (s) to stutter testing, for example,
2832 specifying five seconds causes the test to run for
2833 five seconds, wait for five seconds, and so on.
2834 This tests the locking primitive's ability to
2835 transition abruptly to and from idle.
2837 locktorture.torture_type= [KNL]
2838 Specify the locking implementation to test.
2840 locktorture.verbose= [KNL]
2841 Enable additional printk() statements.
2843 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver
2846 loglevel= All Kernel Messages with a loglevel smaller than the
2847 console loglevel will be printed to the console. It can
2848 also be changed with klogd or other programs. The
2849 loglevels are defined as follows:
2851 0 (KERN_EMERG) system is unusable
2852 1 (KERN_ALERT) action must be taken immediately
2853 2 (KERN_CRIT) critical conditions
2854 3 (KERN_ERR) error conditions
2855 4 (KERN_WARNING) warning conditions
2856 5 (KERN_NOTICE) normal but significant condition
2857 6 (KERN_INFO) informational
2858 7 (KERN_DEBUG) debug-level messages
2860 log_buf_len=n[KMG] Sets the size of the printk ring buffer,
2861 in bytes. n must be a power of two and greater
2862 than the minimal size. The minimal size is defined
2863 by LOG_BUF_SHIFT kernel config parameter. There is
2864 also CONFIG_LOG_CPU_MAX_BUF_SHIFT config parameter
2865 that allows to increase the default size depending on
2866 the number of CPUs. See init/Kconfig for more details.
2868 logo.nologo [FB] Disables display of the built-in Linux logo.
2869 This may be used to provide more screen space for
2870 kernel log messages and is useful when debugging
2871 kernel boot problems.
2873 lp=0 [LP] Specify parallel ports to use, e.g,
2874 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses
2875 lp=reset first parallel port). 'lp=0' disables the
2876 lp=auto printer driver. 'lp=reset' (which can be
2877 specified in addition to the ports) causes
2878 attached printers to be reset. Using
2879 lp=port1,port2,... specifies the parallel ports
2880 to associate lp devices with, starting with
2881 lp0. A port specification may be 'none' to skip
2882 that lp device, or a parport name such as
2883 'parport0'. Specifying 'lp=auto' instead of a
2884 port specification list means that device IDs
2885 from each port should be examined, to see if
2886 an IEEE 1284-compliant printer is attached; if
2887 so, the driver will manage that printer.
2888 See also header of drivers/char/lp.c.
2891 Sets loops_per_jiffy to given constant, thus avoiding
2892 time-consuming boot-time autodetection (up to 250 ms per
2893 CPU). 0 enables autodetection (default). To determine
2894 the correct value for your kernel, boot with normal
2895 autodetection and see what value is printed. Note that
2896 on SMP systems the preset will be applied to all CPUs,
2897 which is likely to cause problems if your CPUs need
2898 significantly divergent settings. An incorrect value
2899 will cause delays in the kernel to be wrong, leading to
2900 unpredictable I/O errors and other breakage. Although
2901 unlikely, in the extreme case this might damage your
2905 Format: <io>,<irq>,<dma>
2907 lsm.debug [SECURITY] Enable LSM initialization debugging output.
2910 [SECURITY] Choose order of LSM initialization. This
2911 overrides CONFIG_LSM, and the "security=" parameter.
2913 machvec= [IA-64] Force the use of a particular machine-vector
2914 (machvec) in a generic kernel.
2915 Example: machvec=hpzx1
2917 machtype= [Loongson] Share the same kernel image file between
2918 different yeeloong laptops.
2919 Example: machtype=lemote-yeeloong-2f-7inch
2921 max_addr=nn[KMG] [KNL,BOOT,IA-64] All physical memory greater
2922 than or equal to this physical address is ignored.
2924 maxcpus= [SMP] Maximum number of processors that an SMP kernel
2925 will bring up during bootup. maxcpus=n : n >= 0 limits
2926 the kernel to bring up 'n' processors. Surely after
2927 bootup you can bring up the other plugged cpu by executing
2928 "echo 1 > /sys/devices/system/cpu/cpuX/online". So maxcpus
2929 only takes effect during system bootup.
2930 While n=0 is a special case, it is equivalent to "nosmp",
2931 which also disables the IO APIC.
2933 max_loop= [LOOP] The number of loop block devices that get
2934 (loop.max_loop) unconditionally pre-created at init time. The default
2935 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead
2936 of statically allocating a predefined number, loop
2937 devices can be requested on-demand with the
2938 /dev/loop-control interface.
2940 mce [X86-32] Machine Check Exception
2942 mce=option [X86-64] See Documentation/x86/x86_64/boot-options.rst
2944 md= [HW] RAID subsystems devices and level
2945 See Documentation/admin-guide/md.rst.
2948 Format: <first>,<last>
2949 Specifies range of consoles to be captured by the MDA.
2952 Control mitigation for the Micro-architectural Data
2953 Sampling (MDS) vulnerability.
2955 Certain CPUs are vulnerable to an exploit against CPU
2956 internal buffers which can forward information to a
2957 disclosure gadget under certain conditions.
2959 In vulnerable processors, the speculatively
2960 forwarded data can be used in a cache side channel
2961 attack, to access data to which the attacker does
2962 not have direct access.
2964 This parameter controls the MDS mitigation. The
2967 full - Enable MDS mitigation on vulnerable CPUs
2968 full,nosmt - Enable MDS mitigation and disable
2969 SMT on vulnerable CPUs
2970 off - Unconditionally disable MDS mitigation
2972 On TAA-affected machines, mds=off can be prevented by
2973 an active TAA mitigation as both vulnerabilities are
2974 mitigated with the same mechanism so in order to disable
2975 this mitigation, you need to specify tsx_async_abort=off
2978 Not specifying this option is equivalent to
2981 For details see: Documentation/admin-guide/hw-vuln/mds.rst
2983 mem=nn[KMG] [HEXAGON] Set the memory size.
2984 Must be specified, otherwise memory size will be 0.
2986 mem=nn[KMG] [KNL,BOOT] Force usage of a specific amount of memory
2987 Amount of memory to be used in cases as follows:
2990 2 when the kernel is not able to see the whole system memory;
2991 3 memory that lies after 'mem=' boundary is excluded from
2992 the hypervisor, then assigned to KVM guests.
2993 4 to limit the memory available for kdump kernel.
2995 [ARC,MICROBLAZE] - the limit applies only to low memory,
2996 high memory is not affected.
2998 [ARM64] - only limits memory covered by the linear
2999 mapping. The NOMAP regions are not affected.
3001 [X86] Work as limiting max address. Use together
3002 with memmap= to avoid physical address space collisions.
3003 Without memmap= PCI devices could be placed at addresses
3004 belonging to unused RAM.
3006 Note that this only takes effects during boot time since
3007 in above case 3, memory may need be hot added after boot
3008 if system memory of hypervisor is not sufficient.
3011 [ARM,MIPS] - override the memory layout reported by
3013 Define a memory region of size nn[KMG] starting at
3015 Multiple different regions can be specified with
3016 multiple mem= parameters on the command line.
3018 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel
3021 memblock=debug [KNL] Enable memblock debug messages.
3024 [KNL,SH] Allow user to override the default size for
3025 per-device physically contiguous DMA buffers.
3027 memhp_default_state=online/offline
3028 [KNL] Set the initial state for the memory hotplug
3029 onlining policy. If not specified, the default value is
3030 set according to the
3031 CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE kernel config
3033 See Documentation/admin-guide/mm/memory-hotplug.rst.
3035 memmap=exactmap [KNL,X86] Enable setting of an exact
3036 E820 memory map, as specified by the user.
3037 Such memmap=exactmap lines can be constructed based on
3038 BIOS output or other requirements. See the memmap=nn@ss
3041 memmap=nn[KMG]@ss[KMG]
3042 [KNL, X86, MIPS, XTENSA] Force usage of a specific region of memory.
3043 Region of memory to be used is from ss to ss+nn.
3044 If @ss[KMG] is omitted, it is equivalent to mem=nn[KMG],
3045 which limits max address to nn[KMG].
3046 Multiple different regions can be specified,
3049 memmap=100M@2G,100M#3G,1G!1024G
3051 memmap=nn[KMG]#ss[KMG]
3052 [KNL,ACPI] Mark specific memory as ACPI data.
3053 Region of memory to be marked is from ss to ss+nn.
3055 memmap=nn[KMG]$ss[KMG]
3056 [KNL,ACPI] Mark specific memory as reserved.
3057 Region of memory to be reserved is from ss to ss+nn.
3058 Example: Exclude memory from 0x18690000-0x1869ffff
3059 memmap=64K$0x18690000
3061 memmap=0x10000$0x18690000
3062 Some bootloaders may need an escape character before '$',
3063 like Grub2, otherwise '$' and the following number
3066 memmap=nn[KMG]!ss[KMG]
3067 [KNL,X86] Mark specific memory as protected.
3068 Region of memory to be used, from ss to ss+nn.
3069 The memory region may be marked as e820 type 12 (0xc)
3070 and is NVDIMM or ADR memory.
3072 memmap=<size>%<offset>-<oldtype>+<newtype>
3073 [KNL,ACPI] Convert memory within the specified region
3074 from <oldtype> to <newtype>. If "-<oldtype>" is left
3075 out, the whole region will be marked as <newtype>,
3076 even if previously unavailable. If "+<newtype>" is left
3077 out, matching memory will be removed. Types are
3078 specified as e820 types, e.g., 1 = RAM, 2 = reserved,
3079 3 = ACPI, 12 = PRAM.
3081 memory_corruption_check=0/1 [X86]
3082 Some BIOSes seem to corrupt the first 64k of
3083 memory when doing things like suspend/resume.
3084 Setting this option will scan the memory
3085 looking for corruption. Enabling this will
3086 both detect corruption and prevent the kernel
3087 from using the memory being corrupted.
3088 However, its intended as a diagnostic tool; if
3089 repeatable BIOS-originated corruption always
3090 affects the same memory, you can use memmap=
3091 to prevent the kernel from using that memory.
3093 memory_corruption_check_size=size [X86]
3094 By default it checks for corruption in the low
3095 64k, making this memory unavailable for normal
3096 use. Use this parameter to scan for
3097 corruption in more or less memory.
3099 memory_corruption_check_period=seconds [X86]
3100 By default it checks for corruption every 60
3101 seconds. Use this parameter to check at some
3102 other rate. 0 disables periodic checking.
3104 memory_hotplug.memmap_on_memory
3105 [KNL,X86,ARM] Boolean flag to enable this feature.
3106 Format: {on | off (default)}
3107 When enabled, runtime hotplugged memory will
3108 allocate its internal metadata (struct pages,
3109 those vmemmap pages cannot be optimized even
3110 if hugetlb_free_vmemmap is enabled) from the
3111 hotadded memory which will allow to hotadd a
3112 lot of memory without requiring additional
3114 This feature is disabled by default because it
3115 has some implication on large (e.g. GB)
3116 allocations in some configurations (e.g. small
3118 The state of the flag can be read in
3119 /sys/module/memory_hotplug/parameters/memmap_on_memory.
3120 Note that even when enabled, there are a few cases where
3121 the feature is not effective.
3123 memtest= [KNL,X86,ARM,M68K,PPC,RISCV] Enable memtest
3125 default : 0 <disable>
3126 Specifies the number of memtest passes to be
3127 performed. Each pass selects another test
3128 pattern from a given set of patterns. Memtest
3129 fills the memory with this pattern, validates
3130 memory contents and reserves bad memory
3131 regions that are detected.
3133 mem_encrypt= [X86-64] AMD Secure Memory Encryption (SME) control
3134 Valid arguments: on, off
3135 Default (depends on kernel configuration option):
3136 on (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=y)
3137 off (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=n)
3138 mem_encrypt=on: Activate SME
3139 mem_encrypt=off: Do not activate SME
3141 Refer to Documentation/virt/kvm/x86/amd-memory-encryption.rst
3142 for details on when memory encryption can be activated.
3144 mem_sleep_default= [SUSPEND] Default system suspend mode:
3145 s2idle - Suspend-To-Idle
3146 shallow - Power-On Suspend or equivalent (if supported)
3147 deep - Suspend-To-RAM or equivalent (if supported)
3148 See Documentation/admin-guide/pm/sleep-states.rst.
3150 meye.*= [HW] Set MotionEye Camera parameters
3151 See Documentation/admin-guide/media/meye.rst.
3153 mfgpt_irq= [IA-32] Specify the IRQ to use for the
3154 Multi-Function General Purpose Timers on AMD Geode
3157 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when
3158 the BIOS has incorrectly applied a workaround. TinyBIOS
3159 version 0.98 is known to be affected, 0.99 fixes the
3160 problem by letting the user disable the workaround.
3164 min_addr=nn[KMG] [KNL,BOOT,IA-64] All physical memory below this
3165 physical address is ignored.
3167 mini2440= [ARM,HW,KNL]
3168 Format:[0..2][b][c][t]
3170 MINI2440 configuration specification:
3171 0 - The attached screen is the 3.5" TFT
3172 1 - The attached screen is the 7" TFT
3173 2 - The VGA Shield is attached (1024x768)
3174 Leaving out the screen size parameter will not load
3175 the TFT driver, and the framebuffer will be left
3177 b - Enable backlight. The TFT backlight pin will be
3178 linked to the kernel VESA blanking code and a GPIO
3179 LED. This parameter is not necessary when using the
3181 c - Enable the s3c camera interface.
3182 t - Reserved for enabling touchscreen support. The
3183 touchscreen support is not enabled in the mainstream
3184 kernel as of 2.6.30, a preliminary port can be found
3185 in the "bleeding edge" mini2440 support kernel at
3186 https://repo.or.cz/w/linux-2.6/mini2440.git
3189 [X86,PPC,S390,ARM64] Control optional mitigations for
3190 CPU vulnerabilities. This is a set of curated,
3191 arch-independent options, each of which is an
3192 aggregation of existing arch-specific options.
3195 Disable all optional CPU mitigations. This
3196 improves system performance, but it may also
3197 expose users to several CPU vulnerabilities.
3198 Equivalent to: nopti [X86,PPC]
3199 if nokaslr then kpti=0 [ARM64]
3200 nospectre_v1 [X86,PPC]
3202 nospectre_v2 [X86,PPC,S390,ARM64]
3203 spectre_v2_user=off [X86]
3204 spec_store_bypass_disable=off [X86,PPC]
3205 ssbd=force-off [ARM64]
3208 tsx_async_abort=off [X86]
3209 kvm.nx_huge_pages=off [X86]
3210 srbds=off [X86,INTEL]
3211 no_entry_flush [PPC]
3212 no_uaccess_flush [PPC]
3213 mmio_stale_data=off [X86]
3217 This does not have any effect on
3218 kvm.nx_huge_pages when
3219 kvm.nx_huge_pages=force.
3222 Mitigate all CPU vulnerabilities, but leave SMT
3223 enabled, even if it's vulnerable. This is for
3224 users who don't want to be surprised by SMT
3225 getting disabled across kernel upgrades, or who
3226 have other ways of avoiding SMT-based attacks.
3227 Equivalent to: (default behavior)
3230 Mitigate all CPU vulnerabilities, disabling SMT
3231 if needed. This is for users who always want to
3232 be fully mitigated, even if it means losing SMT.
3233 Equivalent to: l1tf=flush,nosmt [X86]
3234 mds=full,nosmt [X86]
3235 tsx_async_abort=full,nosmt [X86]
3236 mmio_stale_data=full,nosmt [X86]
3237 retbleed=auto,nosmt [X86]
3240 [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this
3241 parameter allows control of the logging verbosity for
3242 the additional memory initialisation checks. A value
3243 of 0 disables mminit logging and a level of 4 will
3244 log everything. Information is printed at KERN_DEBUG
3245 so loglevel=8 may also need to be specified.
3248 [X86,INTEL] Control mitigation for the Processor
3249 MMIO Stale Data vulnerabilities.
3251 Processor MMIO Stale Data is a class of
3252 vulnerabilities that may expose data after an MMIO
3253 operation. Exposed data could originate or end in
3254 the same CPU buffers as affected by MDS and TAA.
3255 Therefore, similar to MDS and TAA, the mitigation
3256 is to clear the affected CPU buffers.
3258 This parameter controls the mitigation. The
3261 full - Enable mitigation on vulnerable CPUs
3263 full,nosmt - Enable mitigation and disable SMT on
3266 off - Unconditionally disable mitigation
3268 On MDS or TAA affected machines,
3269 mmio_stale_data=off can be prevented by an active
3270 MDS or TAA mitigation as these vulnerabilities are
3271 mitigated with the same mechanism so in order to
3272 disable this mitigation, you need to specify
3273 mds=off and tsx_async_abort=off too.
3275 Not specifying this option is equivalent to
3276 mmio_stale_data=full.
3279 Documentation/admin-guide/hw-vuln/processor_mmio_stale_data.rst
3282 [KNL] When CONFIG_MODULE_SIG is set, this means that
3283 modules without (valid) signatures will fail to load.
3284 Note that if CONFIG_MODULE_SIG_FORCE is set, that
3285 is always true, so this option does nothing.
3287 module_blacklist= [KNL] Do not load a comma-separated list of
3288 modules. Useful for debugging problem modules.
3291 [MOUSE] Maximum time between finger touching and
3292 leaving touchpad surface for touch to be considered
3293 a tap and be reported as a left button click (for
3294 touchpads working in absolute mode only).
3296 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices
3297 reporting absolute coordinates, such as tablets
3298 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices
3299 reporting absolute coordinates, such as tablets
3301 movablecore= [KNL,X86,IA-64,PPC]
3302 Format: nn[KMGTPE] | nn%
3303 This parameter is the complement to kernelcore=, it
3304 specifies the amount of memory used for migratable
3305 allocations. If both kernelcore and movablecore is
3306 specified, then kernelcore will be at *least* the
3307 specified value but may be more. If movablecore on its
3308 own is specified, the administrator must be careful
3309 that the amount of memory usable for all allocations
3312 movable_node [KNL] Boot-time switch to make hotplugable memory
3313 NUMA nodes to be movable. This means that the memory
3314 of such nodes will be usable only for movable
3315 allocations which rules out almost all kernel
3316 allocations. Use with caution!
3318 MTD_Partition= [MTD]
3319 Format: <name>,<region-number>,<size>,<offset>
3321 MTD_Region= [MTD] Format:
3322 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>]
3325 See drivers/mtd/parsers/cmdlinepart.c
3328 ARM/S3C2412 JIVE boot control
3330 See arch/arm/mach-s3c/mach-jive.c
3332 mtouchusb.raw_coordinates=
3333 [HW] Make the MicroTouch USB driver use raw coordinates
3334 ('y', default) or cooked coordinates ('n')
3336 mtrr_chunk_size=nn[KMG] [X86]
3337 used for mtrr cleanup. It is largest continuous chunk
3338 that could hold holes aka. UC entries.
3340 mtrr_gran_size=nn[KMG] [X86]
3341 Used for mtrr cleanup. It is granularity of mtrr block.
3343 Large value could prevent small alignment from
3346 mtrr_spare_reg_nr=n [X86]
3348 Range: 0,7 : spare reg number
3350 Used for mtrr cleanup. It is spare mtrr entries number.
3351 Set to 2 or more if your graphical card needs more.
3353 multitce=off [PPC] This parameter disables the use of the pSeries
3354 firmware feature for updating multiple TCE entries
3357 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card
3359 netdev= [NET] Network devices parameters
3360 Format: <irq>,<io>,<mem_start>,<mem_end>,<name>
3361 Note that mem_start is often overloaded to mean
3362 something different and driver-specific.
3363 This usage is only documented in each driver source
3366 netpoll.carrier_timeout=
3367 [NET] Specifies amount of time (in seconds) that
3368 netpoll should wait for a carrier. By default netpoll
3372 [NETFILTER] Enable connection tracking flow accounting
3373 0 to disable accounting
3374 1 to enable accounting
3377 nfsaddrs= [NFS] Deprecated. Use ip= instead.
3378 See Documentation/admin-guide/nfs/nfsroot.rst.
3380 nfsroot= [NFS] nfs root filesystem for disk-less boxes.
3381 See Documentation/admin-guide/nfs/nfsroot.rst.
3383 nfsrootdebug [NFS] enable nfsroot debugging messages.
3384 See Documentation/admin-guide/nfs/nfsroot.rst.
3386 nfs.callback_nr_threads=
3387 [NFSv4] set the total number of threads that the
3388 NFS client will assign to service NFSv4 callback
3391 nfs.callback_tcpport=
3392 [NFS] set the TCP port on which the NFSv4 callback
3393 channel should listen.
3396 [NFS] sets the pathname to the program which is used
3397 to update the NFS client cache entries.
3399 nfs.cache_getent_timeout=
3400 [NFS] sets the timeout after which an attempt to
3401 update a cache entry is deemed to have failed.
3403 nfs.idmap_cache_timeout=
3404 [NFS] set the maximum lifetime for idmapper cache
3408 [NFS] enable 64-bit inode numbers.
3409 If zero, the NFS client will fake up a 32-bit inode
3410 number for the readdir() and stat() syscalls instead
3411 of returning the full 64-bit number.
3412 The default is to return 64-bit inode numbers.
3414 nfs.max_session_cb_slots=
3415 [NFSv4.1] Sets the maximum number of session
3416 slots the client will assign to the callback
3417 channel. This determines the maximum number of
3418 callbacks the client will process in parallel for
3419 a particular server.
3421 nfs.max_session_slots=
3422 [NFSv4.1] Sets the maximum number of session slots
3423 the client will attempt to negotiate with the server.
3424 This limits the number of simultaneous RPC requests
3425 that the client can send to the NFSv4.1 server.
3426 Note that there is little point in setting this
3427 value higher than the max_tcp_slot_table_limit.
3429 nfs.nfs4_disable_idmapping=
3430 [NFSv4] When set to the default of '1', this option
3431 ensures that both the RPC level authentication
3432 scheme and the NFS level operations agree to use
3433 numeric uids/gids if the mount is using the
3434 'sec=sys' security flavour. In effect it is
3435 disabling idmapping, which can make migration from
3436 legacy NFSv2/v3 systems to NFSv4 easier.
3437 Servers that do not support this mode of operation
3438 will be autodetected by the client, and it will fall
3439 back to using the idmapper.
3440 To turn off this behaviour, set the value to '0'.
3442 [NFS4] Specify an additional fixed unique ident-
3443 ification string that NFSv4 clients can insert into
3444 their nfs_client_id4 string. This is typically a
3445 UUID that is generated at system install time.
3447 nfs.send_implementation_id =
3448 [NFSv4.1] Send client implementation identification
3449 information in exchange_id requests.
3450 If zero, no implementation identification information
3452 The default is to send the implementation identification
3455 nfs.recover_lost_locks =
3456 [NFSv4] Attempt to recover locks that were lost due
3457 to a lease timeout on the server. Please note that
3458 doing this risks data corruption, since there are
3459 no guarantees that the file will remain unchanged
3460 after the locks are lost.
3461 If you want to enable the kernel legacy behaviour of
3462 attempting to recover these locks, then set this
3464 The default parameter value of '0' causes the kernel
3465 not to attempt recovery of lost locks.
3467 nfs4.layoutstats_timer =
3468 [NFSv4.2] Change the rate at which the kernel sends
3469 layoutstats to the pNFS metadata server.
3471 Setting this to value to 0 causes the kernel to use
3472 whatever value is the default set by the layout
3473 driver. A non-zero value sets the minimum interval
3474 in seconds between layoutstats transmissions.
3476 nfsd.inter_copy_offload_enable =
3477 [NFSv4.2] When set to 1, the server will support
3478 server-to-server copies for which this server is
3479 the destination of the copy.
3481 nfsd.nfsd4_ssc_umount_timeout =
3482 [NFSv4.2] When used as the destination of a
3483 server-to-server copy, knfsd temporarily mounts
3484 the source server. It caches the mount in case
3485 it will be needed again, and discards it if not
3486 used for the number of milliseconds specified by
3489 nfsd.nfs4_disable_idmapping=
3490 [NFSv4] When set to the default of '1', the NFSv4
3491 server will return only numeric uids and gids to
3492 clients using auth_sys, and will accept numeric uids
3493 and gids from such clients. This is intended to ease
3494 migration from NFSv2/v3.
3497 nmi_backtrace.backtrace_idle [KNL]
3498 Dump stacks even of idle CPUs in response to an
3499 NMI stack-backtrace request.
3501 nmi_debug= [KNL,SH] Specify one or more actions to take
3502 when a NMI is triggered.
3503 Format: [state][,regs][,debounce][,die]
3505 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels
3506 Format: [panic,][nopanic,][num]
3508 0 - turn hardlockup detector in nmi_watchdog off
3509 1 - turn hardlockup detector in nmi_watchdog on
3510 When panic is specified, panic when an NMI watchdog
3511 timeout occurs (or 'nopanic' to not panic on an NMI
3512 watchdog, if CONFIG_BOOTPARAM_HARDLOCKUP_PANIC is set)
3513 To disable both hard and soft lockup detectors,
3514 please see 'nowatchdog'.
3515 This is useful when you use a panic=... timeout and
3516 need the box quickly up again.
3518 These settings can be accessed at runtime via
3519 the nmi_watchdog and hardlockup_panic sysctls.
3521 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths
3522 emulation library even if a 387 maths coprocessor
3525 no5lvl [X86-64] Disable 5-level paging mode. Forces
3526 kernel to use 4-level paging instead.
3528 nofsgsbase [X86] Disables FSGSBASE instructions.
3531 [HW] Never suspend the console
3532 Disable suspending of consoles during suspend and
3533 hibernate operations. Once disabled, debugging
3534 messages can reach various consoles while the rest
3535 of the system is being put to sleep (ie, while
3536 debugging driver suspend/resume hooks). This may
3537 not work reliably with all consoles, but is known
3538 to work with serial and VGA consoles.
3539 To facilitate more flexible debugging, we also add
3540 console_suspend, a printk module parameter to control
3541 it. Users could use console_suspend (usually
3542 /sys/module/printk/parameters/console_suspend) to
3543 turn on/off it dynamically.
3545 novmcoredd [KNL,KDUMP]
3546 Disable device dump. Device dump allows drivers to
3547 append dump data to vmcore so you can collect driver
3548 specified debug info. Drivers can append the data
3549 without any limit and this data is stored in memory,
3550 so this may cause significant memory stress. Disabling
3551 device dump can help save memory but the driver debug
3552 data will be no longer available. This parameter
3553 is only available when CONFIG_PROC_VMCORE_DEVICE_DUMP
3556 noaliencache [MM, NUMA, SLAB] Disables the allocation of alien
3557 caches in the slab allocator. Saves per-node memory,
3558 but will impact performance.
3562 noaltinstr [S390] Disables alternative instructions patching
3563 (CPU alternatives feature).
3565 noapic [SMP,APIC] Tells the kernel to not make use of any
3566 IOAPICs that may be present in the system.
3568 noautogroup Disable scheduler automatic task group creation.
3572 nodsp [SH] Disable hardware DSP at boot time.
3574 noefi Disable EFI runtime services support.
3576 no_entry_flush [PPC] Don't flush the L1-D cache when entering the kernel.
3581 Disable SMAP (Supervisor Mode Access Prevention)
3582 even if it is supported by processor.
3585 Disable SMEP (Supervisor Mode Execution Prevention)
3586 even if it is supported by processor.
3589 This affects only 32-bit executables.
3590 noexec32=on: enable non-executable mappings (default)
3591 read doesn't imply executable mappings
3592 noexec32=off: disable non-executable mappings
3593 read implies executable mappings
3595 nofpu [MIPS,SH] Disable hardware FPU at boot time.
3597 nofxsr [BUGS=X86-32] Disables x86 floating point extended
3598 register save and restore. The kernel will only save
3599 legacy floating-point registers on task switch.
3601 nohugeiomap [KNL,X86,PPC,ARM64] Disable kernel huge I/O mappings.
3603 nohugevmalloc [PPC] Disable kernel huge vmalloc mappings.
3605 nosmt [KNL,S390] Disable symmetric multithreading (SMT).
3606 Equivalent to smt=1.
3608 [KNL,X86] Disable symmetric multithreading (SMT).
3609 nosmt=force: Force disable SMT, cannot be undone
3610 via the sysfs control file.
3612 nospectre_v1 [X86,PPC] Disable mitigations for Spectre Variant 1
3613 (bounds check bypass). With this option data leaks are
3614 possible in the system.
3616 nospectre_v2 [X86,PPC_FSL_BOOK3E,ARM64] Disable all mitigations for
3617 the Spectre variant 2 (indirect branch prediction)
3618 vulnerability. System may allow data leaks with this
3621 nospec_store_bypass_disable
3622 [HW] Disable all mitigations for the Speculative Store Bypass vulnerability
3625 [PPC] Don't flush the L1-D cache after accessing user data.
3627 noxsave [BUGS=X86] Disables x86 extended register state save
3628 and restore using xsave. The kernel will fallback to
3629 enabling legacy floating-point and sse state.
3631 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended
3632 register states. The kernel will fall back to use
3633 xsave to save the states. By using this parameter,
3634 performance of saving the states is degraded because
3635 xsave doesn't support modified optimization while
3636 xsaveopt supports it on xsaveopt enabled systems.
3638 noxsaves [X86] Disables xsaves and xrstors used in saving and
3639 restoring x86 extended register state in compacted
3640 form of xsave area. The kernel will fall back to use
3641 xsaveopt and xrstor to save and restore the states
3642 in standard form of xsave area. By using this
3643 parameter, xsave area per process might occupy more
3644 memory on xsaves enabled systems.
3646 nohlt [ARM,ARM64,MICROBLAZE,SH] Forces the kernel to busy wait
3647 in do_idle() and not use the arch_cpu_idle()
3648 implementation; requires CONFIG_GENERIC_IDLE_POLL_SETUP
3649 to be effective. This is useful on platforms where the
3650 sleep(SH) or wfi(ARM,ARM64) instructions do not work
3651 correctly or when doing power measurements to evalute
3652 the impact of the sleep instructions. This is also
3653 useful when using JTAG debugger.
3655 no_file_caps Tells the kernel not to honor file capabilities. The
3656 only way then for a file to be executed with privilege
3657 is to be setuid root or executed by root.
3659 nohalt [IA-64] Tells the kernel not to use the power saving
3660 function PAL_HALT_LIGHT when idle. This increases
3661 power-consumption. On the positive side, it reduces
3662 interrupt wake-up latency, which may improve performance
3663 in certain environments such as networked servers or
3667 Force pointers printed to the console or buffers to be
3668 unhashed. By default, when a pointer is printed via %p
3669 format string, that pointer is "hashed", i.e. obscured
3670 by hashing the pointer value. This is a security feature
3671 that hides actual kernel addresses from unprivileged
3672 users, but it also makes debugging the kernel more
3673 difficult since unequal pointers can no longer be
3674 compared. However, if this command-line option is
3675 specified, then all normal pointers will have their true
3676 value printed. This option should only be specified when
3677 debugging the kernel. Please do not use on production
3680 nohibernate [HIBERNATION] Disable hibernation and resume.
3682 nohz= [KNL] Boottime enable/disable dynamic ticks
3683 Valid arguments: on, off
3686 nohz_full= [KNL,BOOT,SMP,ISOL]
3687 The argument is a cpu list, as described above.
3688 In kernels built with CONFIG_NO_HZ_FULL=y, set
3689 the specified list of CPUs whose tick will be stopped
3690 whenever possible. The boot CPU will be forced outside
3691 the range to maintain the timekeeping. Any CPUs
3692 in this list will have their RCU callbacks offloaded,
3693 just as if they had also been called out in the
3694 rcu_nocbs= boot parameter.
3696 Note that this argument takes precedence over
3697 the CONFIG_RCU_NOCB_CPU_DEFAULT_ALL option.
3699 noiotrap [SH] Disables trapped I/O port accesses.
3701 noirqdebug [X86-32] Disables the code which attempts to detect and
3702 disable unhandled interrupt sources.
3704 no_timer_check [X86,APIC] Disables the code which tests for
3705 broken timer IRQ sources.
3707 noisapnp [ISAPNP] Disables ISA PnP code.
3709 noinitrd [RAM] Tells the kernel not to load any configured
3712 nointremap [X86-64, Intel-IOMMU] Do not enable interrupt
3714 [Deprecated - use intremap=off]
3718 noinvpcid [X86] Disable the INVPCID cpu feature.
3720 nojitter [IA-64] Disables jitter checking for ITC timers.
3722 no-kvmclock [X86,KVM] Disable paravirtualized KVM clock driver
3724 no-kvmapf [X86,KVM] Disable paravirtualized asynchronous page
3728 [X86,PV_OPS] Disable paravirtualized VMware scheduler
3729 clock and use the default one.
3731 no-steal-acc [X86,PV_OPS,ARM64] Disable paravirtualized steal time
3732 accounting. steal time is computed, but won't
3733 influence scheduler behaviour
3735 nolapic [X86-32,APIC] Do not enable or use the local APIC.
3737 nolapic_timer [X86-32,APIC] Do not use the local APIC timer.
3739 nomca [IA-64] Disable machine check abort handling
3741 nomce [X86-32] Disable Machine Check Exception
3743 nomfgpt [X86-32] Disable Multi-Function General Purpose
3744 Timer usage (for AMD Geode machines).
3746 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to
3747 shutdown the other cpus. Instead use the REBOOT_VECTOR
3750 nomodeset Disable kernel modesetting. DRM drivers will not perform
3751 display-mode changes or accelerated rendering. Only the
3752 system framebuffer will be available for use if this was
3753 set-up by the firmware or boot loader.
3755 Useful as fallback, or for testing and debugging.
3757 nomodule Disable module load
3759 nopat [X86] Disable PAT (page attribute table extension of
3760 pagetables) support.
3762 nopcid [X86-64] Disable the PCID cpu feature.
3764 norandmaps Don't use address space randomization. Equivalent to
3765 echo 0 > /proc/sys/kernel/randomize_va_space
3767 noreplace-smp [X86-32,SMP] Don't replace SMP instructions
3768 with UP alternatives
3770 noresume [SWSUSP] Disables resume and restores original swap
3773 no-scroll [VGA] Disables scrollback.
3774 This is required for the Braillex ib80-piezo Braille
3775 reader made by F.H. Papenmeier (Germany).
3779 nosgx [X86-64,SGX] Disables Intel SGX kernel support.
3781 nosmp [SMP] Tells an SMP kernel to act as a UP kernel,
3782 and disable the IO APIC. legacy for "maxcpus=0".
3784 nosoftlockup [KNL] Disable the soft-lockup detector.
3786 nosync [HW,M68K] Disables sync negotiation for all devices.
3788 nowatchdog [KNL] Disable both lockup detectors, i.e.
3789 soft-lockup and NMI watchdog (hard-lockup).
3793 nox2apic [X86-64,APIC] Do not enable x2APIC mode.
3795 nps_mtm_hs_ctr= [KNL,ARC]
3796 This parameter sets the maximum duration, in
3797 cycles, each HW thread of the CTOP can run
3798 without interruptions, before HW switches it.
3799 The actual maximum duration is 16 times this
3801 Format: integer between 1 and 255
3804 nptcg= [IA-64] Override max number of concurrent global TLB
3805 purges which is reported from either PAL_VM_SUMMARY or
3808 nr_cpus= [SMP] Maximum number of processors that an SMP kernel
3809 could support. nr_cpus=n : n >= 1 limits the kernel to
3810 support 'n' processors. It could be larger than the
3811 number of already plugged CPU during bootup, later in
3812 runtime you can physically add extra cpu until it reaches
3813 n. So during boot up some boot time memory for per-cpu
3814 variables need be pre-allocated for later physical cpu
3817 nr_uarts= [SERIAL] maximum number of UARTs to be registered.
3819 numa=off [KNL, ARM64, PPC, RISCV, SPARC, X86] Disable NUMA, Only
3820 set up a single NUMA node spanning all memory.
3822 numa_balancing= [KNL,ARM64,PPC,RISCV,S390,X86] Enable or disable automatic
3824 Allowed values are enable and disable
3826 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA.
3827 'node', 'default' can be specified
3828 This can be set from sysctl after boot.
3829 See Documentation/admin-guide/sysctl/vm.rst for details.
3831 ohci1394_dma=early [HW] enable debugging via the ohci1394 driver.
3832 See Documentation/core-api/debugging-via-ohci1394.rst for more
3835 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands
3836 Rather than timing out after 20 ms if an EC
3837 command is not properly ACKed, override the length
3838 of the timeout. We have interrupts disabled while
3839 waiting for the ACK, so if this is set too high
3840 interrupts *may* be lost!
3842 omap_mux= [OMAP] Override bootloader pin multiplexing.
3843 Format: <mux_mode0.mode_name=value>...
3844 For example, to override I2C bus2:
3845 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100
3847 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration
3849 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock]
3851 boundary - index of last SLC block on Flex-OneNAND.
3852 The remaining blocks are configured as MLC blocks.
3853 lock - Configure if Flex-OneNAND boundary should be locked.
3854 Once locked, the boundary cannot be changed.
3855 1 indicates lock status, 0 indicates unlock status.
3857 oops=panic Always panic on oopses. Default is to just kill the
3858 process, but there is a small probability of
3859 deadlocking the machine.
3860 This will also cause panics on machine check exceptions.
3861 Useful together with panic=30 to trigger a reboot.
3864 [KNL] Boolean flag to control whether the page allocator
3865 should randomize its free lists. The randomization may
3866 be automatically enabled if the kernel detects it is
3867 running on a platform with a direct-mapped memory-side
3868 cache, and this parameter can be used to
3869 override/disable that behavior. The state of the flag
3870 can be read from sysfs at:
3871 /sys/module/page_alloc/parameters/shuffle.
3873 page_owner= [KNL] Boot-time page_owner enabling option.
3874 Storage of the information about who allocated
3875 each page is disabled in default. With this switch,
3877 on: enable the feature
3879 page_poison= [KNL] Boot-time parameter changing the state of
3880 poisoning on the buddy allocator, available with
3881 CONFIG_PAGE_POISONING=y.
3882 off: turn off poisoning (default)
3883 on: turn on poisoning
3885 page_reporting.page_reporting_order=
3886 [KNL] Minimal page reporting order
3888 Adjust the minimal page reporting order. The page
3889 reporting is disabled when it exceeds (MAX_ORDER-1).
3891 panic= [KNL] Kernel behaviour on panic: delay <timeout>
3892 timeout > 0: seconds before rebooting
3893 timeout = 0: wait forever
3894 timeout < 0: reboot immediately
3897 panic_print= Bitmask for printing system info when panic happens.
3898 User can chose combination of the following bits:
3899 bit 0: print all tasks info
3900 bit 1: print system memory info
3901 bit 2: print timer info
3902 bit 3: print locks info if CONFIG_LOCKDEP is on
3903 bit 4: print ftrace buffer
3904 bit 5: print all printk messages in buffer
3905 bit 6: print all CPUs backtrace (if available in the arch)
3906 *Be aware* that this option may print a _lot_ of lines,
3907 so there are risks of losing older messages in the log.
3908 Use this option carefully, maybe worth to setup a
3909 bigger log buffer with "log_buf_len" along with this.
3911 panic_on_taint= Bitmask for conditionally calling panic() in add_taint()
3912 Format: <hex>[,nousertaint]
3913 Hexadecimal bitmask representing the set of TAINT flags
3914 that will cause the kernel to panic when add_taint() is
3915 called with any of the flags in this set.
3916 The optional switch "nousertaint" can be utilized to
3917 prevent userspace forced crashes by writing to sysctl
3918 /proc/sys/kernel/tainted any flagset matching with the
3919 bitmask set on panic_on_taint.
3920 See Documentation/admin-guide/tainted-kernels.rst for
3921 extra details on the taint flags that users can pick
3922 to compose the bitmask to assign to panic_on_taint.
3924 panic_on_warn panic() instead of WARN(). Useful to cause kdump
3927 parkbd.port= [HW] Parallel port number the keyboard adapter is
3928 connected to, default is 0.
3930 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation,
3931 0 for XT, 1 for AT (default is AT).
3934 parport= [HW,PPT] Specify parallel ports. 0 disables.
3935 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] }
3936 Use 'auto' to force the driver to use any
3937 IRQ/DMA settings detected (the default is to
3938 ignore detected IRQ/DMA settings because of
3939 possible conflicts). You can specify the base
3940 address, IRQ, and DMA settings; IRQ and DMA
3941 should be numbers, or 'auto' (for using detected
3942 settings on that particular port), or 'nofifo'
3943 (to avoid using a FIFO even if it is detected).
3944 Parallel ports are assigned in the order they
3945 are specified on the command line, starting
3948 parport_init_mode= [HW,PPT]
3949 Configure VIA parallel port to operate in
3950 a specific mode. This is necessary on Pegasos
3951 computer where firmware has no options for setting
3952 up parallel port mode and sets it to spp.
3953 Currently this function knows 686a and 8231 chips.
3954 Format: [spp|ps2|epp|ecp|ecpepp]
3956 pata_legacy.all= [HW,LIBATA]
3958 Set to non-zero to probe primary and secondary ISA
3959 port ranges on PCI systems where no PCI PATA device
3960 has been found at either range. Disabled by default.
3962 pata_legacy.autospeed= [HW,LIBATA]
3964 Set to non-zero if a chip is present that snoops speed
3965 changes. Disabled by default.
3967 pata_legacy.ht6560a= [HW,LIBATA]
3969 Set to 1, 2, or 3 for HT 6560A on the primary channel,
3970 the secondary channel, or both channels respectively.
3971 Disabled by default.
3973 pata_legacy.ht6560b= [HW,LIBATA]
3975 Set to 1, 2, or 3 for HT 6560B on the primary channel,
3976 the secondary channel, or both channels respectively.
3977 Disabled by default.
3979 pata_legacy.iordy_mask= [HW,LIBATA]
3981 IORDY enable mask. Set individual bits to allow IORDY
3982 for the respective channel. Bit 0 is for the first
3983 legacy channel handled by this driver, bit 1 is for
3984 the second channel, and so on. The sequence will often
3985 correspond to the primary legacy channel, the secondary
3986 legacy channel, and so on, but the handling of a PCI
3987 bus and the use of other driver options may interfere
3988 with the sequence. By default IORDY is allowed across
3991 pata_legacy.opti82c46x= [HW,LIBATA]
3993 Set to 1, 2, or 3 for Opti 82c611A on the primary
3994 channel, the secondary channel, or both channels
3995 respectively. Disabled by default.
3997 pata_legacy.opti82c611a= [HW,LIBATA]
3999 Set to 1, 2, or 3 for Opti 82c465MV on the primary
4000 channel, the secondary channel, or both channels
4001 respectively. Disabled by default.
4003 pata_legacy.pio_mask= [HW,LIBATA]
4005 PIO mode mask for autospeed devices. Set individual
4006 bits to allow the use of the respective PIO modes.
4007 Bit 0 is for mode 0, bit 1 is for mode 1, and so on.
4008 All modes allowed by default.
4010 pata_legacy.probe_all= [HW,LIBATA]
4012 Set to non-zero to probe tertiary and further ISA
4013 port ranges on PCI systems. Disabled by default.
4015 pata_legacy.probe_mask= [HW,LIBATA]
4017 Probe mask for legacy ISA PATA ports. Depending on
4018 platform configuration and the use of other driver
4019 options up to 6 legacy ports are supported: 0x1f0,
4020 0x170, 0x1e8, 0x168, 0x1e0, 0x160, however probing
4021 of individual ports can be disabled by setting the
4022 corresponding bits in the mask to 1. Bit 0 is for
4023 the first port in the list above (0x1f0), and so on.
4024 By default all supported ports are probed.
4026 pata_legacy.qdi= [HW,LIBATA]
4028 Set to non-zero to probe QDI controllers. By default
4029 set to 1 if CONFIG_PATA_QDI_MODULE, 0 otherwise.
4031 pata_legacy.winbond= [HW,LIBATA]
4033 Set to non-zero to probe Winbond controllers. Use
4034 the standard I/O port (0x130) if 1, otherwise the
4035 value given is the I/O port to use (typically 0x1b0).
4036 By default set to 1 if CONFIG_PATA_WINBOND_VLB_MODULE,
4039 pata_platform.pio_mask= [HW,LIBATA]
4041 Supported PIO mode mask. Set individual bits to allow
4042 the use of the respective PIO modes. Bit 0 is for
4043 mode 0, bit 1 is for mode 1, and so on. Mode 0 only
4047 Halt all CPUs after the first oops has been printed for
4048 the specified number of seconds. This is to be used if
4049 your oopses keep scrolling off the screen.
4054 See header of drivers/block/paride/pcd.c.
4055 See also Documentation/admin-guide/blockdev/paride.rst.
4057 pci=option[,option...] [PCI] various PCI subsystem options.
4059 Some options herein operate on a specific device
4060 or a set of devices (<pci_dev>). These are
4061 specified in one of the following formats:
4063 [<domain>:]<bus>:<dev>.<func>[/<dev>.<func>]*
4064 pci:<vendor>:<device>[:<subvendor>:<subdevice>]
4066 Note: the first format specifies a PCI
4067 bus/device/function address which may change
4068 if new hardware is inserted, if motherboard
4069 firmware changes, or due to changes caused
4070 by other kernel parameters. If the
4071 domain is left unspecified, it is
4072 taken to be zero. Optionally, a path
4073 to a device through multiple device/function
4074 addresses can be specified after the base
4075 address (this is more robust against
4076 renumbering issues). The second format
4077 selects devices using IDs from the
4078 configuration space which may match multiple
4079 devices in the system.
4081 earlydump dump PCI config space before the kernel
4083 off [X86] don't probe for the PCI bus
4084 bios [X86-32] force use of PCI BIOS, don't access
4085 the hardware directly. Use this if your machine
4086 has a non-standard PCI host bridge.
4087 nobios [X86-32] disallow use of PCI BIOS, only direct
4088 hardware access methods are allowed. Use this
4089 if you experience crashes upon bootup and you
4090 suspect they are caused by the BIOS.
4091 conf1 [X86] Force use of PCI Configuration Access
4092 Mechanism 1 (config address in IO port 0xCF8,
4093 data in IO port 0xCFC, both 32-bit).
4094 conf2 [X86] Force use of PCI Configuration Access
4095 Mechanism 2 (IO port 0xCF8 is an 8-bit port for
4096 the function, IO port 0xCFA, also 8-bit, sets
4097 bus number. The config space is then accessed
4098 through ports 0xC000-0xCFFF).
4099 See http://wiki.osdev.org/PCI for more info
4100 on the configuration access mechanisms.
4101 noaer [PCIE] If the PCIEAER kernel config parameter is
4102 enabled, this kernel boot option can be used to
4103 disable the use of PCIE advanced error reporting.
4104 nodomains [PCI] Disable support for multiple PCI
4105 root domains (aka PCI segments, in ACPI-speak).
4106 nommconf [X86] Disable use of MMCONFIG for PCI
4108 check_enable_amd_mmconf [X86] check for and enable
4109 properly configured MMIO access to PCI
4110 config space on AMD family 10h CPU
4111 nomsi [MSI] If the PCI_MSI kernel config parameter is
4112 enabled, this kernel boot option can be used to
4113 disable the use of MSI interrupts system-wide.
4114 noioapicquirk [APIC] Disable all boot interrupt quirks.
4115 Safety option to keep boot IRQs enabled. This
4116 should never be necessary.
4117 ioapicreroute [APIC] Enable rerouting of boot IRQs to the
4118 primary IO-APIC for bridges that cannot disable
4119 boot IRQs. This fixes a source of spurious IRQs
4120 when the system masks IRQs.
4121 noioapicreroute [APIC] Disable workaround that uses the
4122 boot IRQ equivalent of an IRQ that connects to
4123 a chipset where boot IRQs cannot be disabled.
4124 The opposite of ioapicreroute.
4125 biosirq [X86-32] Use PCI BIOS calls to get the interrupt
4126 routing table. These calls are known to be buggy
4127 on several machines and they hang the machine
4128 when used, but on other computers it's the only
4129 way to get the interrupt routing table. Try
4130 this option if the kernel is unable to allocate
4131 IRQs or discover secondary PCI buses on your
4133 rom [X86] Assign address space to expansion ROMs.
4134 Use with caution as certain devices share
4135 address decoders between ROMs and other
4137 norom [X86] Do not assign address space to
4138 expansion ROMs that do not already have
4139 BIOS assigned address ranges.
4140 nobar [X86] Do not assign address space to the
4141 BARs that weren't assigned by the BIOS.
4142 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be
4143 assigned automatically to PCI devices. You can
4144 make the kernel exclude IRQs of your ISA cards
4146 pirqaddr=0xAAAAA [X86] Specify the physical address
4147 of the PIRQ table (normally generated
4148 by the BIOS) if it is outside the
4149 F0000h-100000h range.
4150 lastbus=N [X86] Scan all buses thru bus #N. Can be
4151 useful if the kernel is unable to find your
4152 secondary buses and you want to tell it
4153 explicitly which ones they are.
4154 assign-busses [X86] Always assign all PCI bus
4155 numbers ourselves, overriding
4156 whatever the firmware may have done.
4157 usepirqmask [X86] Honor the possible IRQ mask stored
4158 in the BIOS $PIR table. This is needed on
4159 some systems with broken BIOSes, notably
4160 some HP Pavilion N5400 and Omnibook XE3
4161 notebooks. This will have no effect if ACPI
4162 IRQ routing is enabled.
4163 noacpi [X86] Do not use ACPI for IRQ routing
4164 or for PCI scanning.
4165 use_crs [X86] Use PCI host bridge window information
4166 from ACPI. On BIOSes from 2008 or later, this
4167 is enabled by default. If you need to use this,
4168 please report a bug.
4169 nocrs [X86] Ignore PCI host bridge windows from ACPI.
4170 If you need to use this, please report a bug.
4171 use_e820 [X86] Use E820 reservations to exclude parts of
4172 PCI host bridge windows. This is a workaround
4173 for BIOS defects in host bridge _CRS methods.
4174 If you need to use this, please report a bug to
4175 <linux-pci@vger.kernel.org>.
4176 no_e820 [X86] Ignore E820 reservations for PCI host
4177 bridge windows. This is the default on modern
4178 hardware. If you need to use this, please report
4179 a bug to <linux-pci@vger.kernel.org>.
4180 routeirq Do IRQ routing for all PCI devices.
4181 This is normally done in pci_enable_device(),
4182 so this option is a temporary workaround
4183 for broken drivers that don't call it.
4184 skip_isa_align [X86] do not align io start addr, so can
4185 handle more pci cards
4186 noearly [X86] Don't do any early type 1 scanning.
4187 This might help on some broken boards which
4188 machine check when some devices' config space
4189 is read. But various workarounds are disabled
4190 and some IOMMU drivers will not work.
4191 bfsort Sort PCI devices into breadth-first order.
4192 This sorting is done to get a device
4193 order compatible with older (<= 2.4) kernels.
4194 nobfsort Don't sort PCI devices into breadth-first order.
4195 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size)
4196 tuning and use the BIOS-configured MPS defaults.
4197 pcie_bus_safe Set every device's MPS to the largest value
4198 supported by all devices below the root complex.
4199 pcie_bus_perf Set device MPS to the largest allowable MPS
4200 based on its parent bus. Also set MRRS (Max
4201 Read Request Size) to the largest supported
4202 value (no larger than the MPS that the device
4203 or bus can support) for best performance.
4204 pcie_bus_peer2peer Set every device's MPS to 128B, which
4205 every device is guaranteed to support. This
4206 configuration allows peer-to-peer DMA between
4207 any pair of devices, possibly at the cost of
4208 reduced performance. This also guarantees
4209 that hot-added devices will work.
4210 cbiosize=nn[KMG] The fixed amount of bus space which is
4211 reserved for the CardBus bridge's IO window.
4212 The default value is 256 bytes.
4213 cbmemsize=nn[KMG] The fixed amount of bus space which is
4214 reserved for the CardBus bridge's memory
4215 window. The default value is 64 megabytes.
4218 [<order of align>@]<pci_dev>[; ...]
4219 Specifies alignment and device to reassign
4220 aligned memory resources. How to
4221 specify the device is described above.
4222 If <order of align> is not specified,
4223 PAGE_SIZE is used as alignment.
4224 A PCI-PCI bridge can be specified if resource
4225 windows need to be expanded.
4226 To specify the alignment for several
4227 instances of a device, the PCI vendor,
4228 device, subvendor, and subdevice may be
4229 specified, e.g., 12@pci:8086:9c22:103c:198f
4230 for 4096-byte alignment.
4231 ecrc= Enable/disable PCIe ECRC (transaction layer
4232 end-to-end CRC checking).
4233 bios: Use BIOS/firmware settings. This is the
4237 hpiosize=nn[KMG] The fixed amount of bus space which is
4238 reserved for hotplug bridge's IO window.
4239 Default size is 256 bytes.
4240 hpmmiosize=nn[KMG] The fixed amount of bus space which is
4241 reserved for hotplug bridge's MMIO window.
4242 Default size is 2 megabytes.
4243 hpmmioprefsize=nn[KMG] The fixed amount of bus space which is
4244 reserved for hotplug bridge's MMIO_PREF window.
4245 Default size is 2 megabytes.
4246 hpmemsize=nn[KMG] The fixed amount of bus space which is
4247 reserved for hotplug bridge's MMIO and
4249 Default size is 2 megabytes.
4250 hpbussize=nn The minimum amount of additional bus numbers
4251 reserved for buses below a hotplug bridge.
4253 realloc= Enable/disable reallocating PCI bridge resources
4254 if allocations done by BIOS are too small to
4255 accommodate resources required by all child
4257 off: Turn realloc off
4259 realloc same as realloc=on
4260 noari do not use PCIe ARI.
4261 noats [PCIE, Intel-IOMMU, AMD-IOMMU]
4262 do not use PCIe ATS (and IOMMU device IOTLB).
4263 pcie_scan_all Scan all possible PCIe devices. Otherwise we
4264 only look for one device below a PCIe downstream
4266 big_root_window Try to add a big 64bit memory window to the PCIe
4267 root complex on AMD CPUs. Some GFX hardware
4268 can resize a BAR to allow access to all VRAM.
4269 Adding the window is slightly risky (it may
4270 conflict with unreported devices), so this
4272 disable_acs_redir=<pci_dev>[; ...]
4273 Specify one or more PCI devices (in the format
4274 specified above) separated by semicolons.
4275 Each device specified will have the PCI ACS
4276 redirect capabilities forced off which will
4277 allow P2P traffic between devices through
4278 bridges without forcing it upstream. Note:
4279 this removes isolation between devices and
4280 may put more devices in an IOMMU group.
4281 force_floating [S390] Force usage of floating interrupts.
4282 nomio [S390] Do not use MIO instructions.
4283 norid [S390] ignore the RID field and force use of
4284 one PCI domain per PCI function
4286 pcie_aspm= [PCIE] Forcibly enable or disable PCIe Active State Power
4289 force Enable ASPM even on devices that claim not to support it.
4290 WARNING: Forcing ASPM on may cause system lockups.
4292 pcie_ports= [PCIE] PCIe port services handling:
4293 native Use native PCIe services (PME, AER, DPC, PCIe hotplug)
4294 even if the platform doesn't give the OS permission to
4295 use them. This may cause conflicts if the platform
4296 also tries to use these services.
4297 dpc-native Use native PCIe service for DPC only. May
4298 cause conflicts if firmware uses AER or DPC.
4299 compat Disable native PCIe services (PME, AER, DPC, PCIe
4302 pcie_port_pm= [PCIE] PCIe port power management handling:
4303 off Disable power management of all PCIe ports
4304 force Forcibly enable power management of all PCIe ports
4306 pcie_pme= [PCIE,PM] Native PCIe PME signaling options:
4307 nomsi Do not use MSI for native PCIe PME signaling (this makes
4308 all PCIe root ports use INTx for all services).
4310 pcmv= [HW,PCMCIA] BadgePAD 4
4314 Keep all power-domains already enabled by bootloader on,
4315 even if no driver has claimed them. This is useful
4316 for debug and development, but should not be
4317 needed on a platform with proper driver support.
4320 See Documentation/admin-guide/blockdev/paride.rst.
4322 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at
4325 See arch/parisc/kernel/pdc_chassis.c
4327 percpu_alloc= Select which percpu first chunk allocator to use.
4328 Currently supported values are "embed" and "page".
4329 Archs may support subset or none of the selections.
4330 See comments in mm/percpu.c for details on each
4331 allocator. This parameter is primarily for debugging
4332 and performance comparison.
4335 See Documentation/admin-guide/blockdev/paride.rst.
4338 See Documentation/admin-guide/blockdev/paride.rst.
4340 pirq= [SMP,APIC] Manual mp-table setup
4341 See Documentation/x86/i386/IO-APIC.rst.
4343 plip= [PPT,NET] Parallel port network link
4344 Format: { parport<nr> | timid | 0 }
4345 See also Documentation/admin-guide/parport.rst.
4347 pmtmr= [X86] Manual setup of pmtmr I/O Port.
4348 Override pmtimer IOPort with a hex value.
4351 pmu_override= [PPC] Override the PMU.
4352 This option takes over the PMU facility, so it is no
4353 longer usable by perf. Setting this option starts the
4354 PMU counters by setting MMCR0 to 0 (the FC bit is
4355 cleared). If a number is given, then MMCR1 is set to
4356 that number, otherwise (e.g., 'pmu_override=on'), MMCR1
4359 pm_debug_messages [SUSPEND,KNL]
4360 Enable suspend/resume debug messages during boot up.
4363 Enable PNP debug messages (depends on the
4364 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time
4365 via /sys/module/pnp/parameters/debug. We always show
4366 current resource usage; turning this on also shows
4367 possible settings and some assignment information.
4373 { on | off | curr | res | no-curr | no-res }
4376 [ISAPNP] Exclude IRQs for the autoconfiguration
4379 [ISAPNP] Exclude DMAs for the autoconfiguration
4381 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration
4382 Ranges are in pairs (I/O port base and size).
4385 [ISAPNP] Exclude memory regions for the
4387 Ranges are in pairs (memory base and size).
4389 ports= [IP_VS_FTP] IPVS ftp helper module
4391 Up to 8 (IP_VS_APP_MAX_PORTS) ports
4393 Format: <port>,<port>....
4395 powersave=off [PPC] This option disables power saving features.
4396 It specifically disables cpuidle and sets the
4397 platform machine description specific power_save
4398 function to NULL. On Idle the CPU just reduces
4401 ppc_strict_facility_enable
4402 [PPC] This option catches any kernel floating point,
4403 Altivec, VSX and SPE outside of regions specifically
4404 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()).
4405 There is some performance impact when enabling this.
4409 Disable Hardware Transactional Memory
4412 Select preemption mode if you have CONFIG_PREEMPT_DYNAMIC
4413 none - Limited to cond_resched() calls
4414 voluntary - Limited to cond_resched() and might_sleep() calls
4415 full - Any section that isn't explicitly preempt disabled
4416 can be preempted anytime.
4418 print-fatal-signals=
4419 [KNL] debug: print fatal signals
4421 If enabled, warn about various signal handling
4422 related application anomalies: too many signals,
4423 too many POSIX.1 timers, fatal signals causing a
4426 If you hit the warning due to signal overflow,
4427 you might want to try "ulimit -i unlimited".
4431 printk.always_kmsg_dump=
4432 Trigger kmsg_dump for cases other than kernel oops or
4434 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
4437 printk.console_no_auto_verbose=
4438 Disable console loglevel raise on oops, panic
4439 or lockdep-detected issues (only if lock debug is on).
4440 With an exception to setups with low baudrate on
4441 serial console, keeping this 0 is a good choice
4442 in order to provide more debug information.
4444 default: 0 (auto_verbose is enabled)
4446 printk.devkmsg={on,off,ratelimit}
4447 Control writing to /dev/kmsg.
4448 on - unlimited logging to /dev/kmsg from userspace
4449 off - logging to /dev/kmsg disabled
4450 ratelimit - ratelimit the logging
4453 printk.time= Show timing data prefixed to each printk message line
4454 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
4456 processor.max_cstate= [HW,ACPI]
4457 Limit processor to maximum C-state
4458 max_cstate=9 overrides any DMI blacklist limit.
4460 processor.nocst [HW,ACPI]
4461 Ignore the _CST method to determine C-states,
4462 instead using the legacy FADT method
4464 profile= [KNL] Enable kernel profiling via /proc/profile
4465 Format: [<profiletype>,]<number>
4466 Param: <profiletype>: "schedule", "sleep", or "kvm"
4467 [defaults to kernel profiling]
4468 Param: "schedule" - profile schedule points.
4469 Param: "sleep" - profile D-state sleeping (millisecs).
4470 Requires CONFIG_SCHEDSTATS
4471 Param: "kvm" - profile VM exits.
4472 Param: <number> - step/bucket size as a power of 2 for
4473 statistical time based profiling.
4475 prompt_ramdisk= [RAM] [Deprecated]
4477 prot_virt= [S390] enable hosting protected virtual machines
4478 isolated from the hypervisor (if hardware supports
4482 psi= [KNL] Enable or disable pressure stall information
4486 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to
4487 probe for; one of (bare|imps|exps|lifebook|any).
4488 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports
4490 psmouse.resetafter= [HW,MOUSE]
4491 Try to reset the device after so many bad packets
4494 [HW,MOUSE] Set desired mouse resolution, in dpi.
4495 psmouse.smartscroll=
4496 [HW,MOUSE] Controls Logitech smartscroll autorepeat.
4497 0 = disabled, 1 = enabled (default).
4499 pstore.backend= Specify the name of the pstore backend to use
4502 See Documentation/admin-guide/blockdev/paride.rst.
4504 pti= [X86-64] Control Page Table Isolation of user and
4505 kernel address spaces. Disabling this feature
4506 removes hardening, but improves performance of
4507 system calls and interrupts.
4509 on - unconditionally enable
4510 off - unconditionally disable
4511 auto - kernel detects whether your CPU model is
4512 vulnerable to issues that PTI mitigates
4514 Not specifying this option is equivalent to pti=auto.
4517 Equivalent to pti=off
4520 [KNL] Number of legacy pty's. Overwrites compiled-in
4523 quiet [KNL] Disable most log messages
4528 See Documentation/admin-guide/md.rst.
4530 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes
4531 See Documentation/admin-guide/blockdev/ramdisk.rst.
4533 ramdisk_start= [RAM] RAM disk image start address
4535 random.trust_cpu={on,off}
4536 [KNL] Enable or disable trusting the use of the
4537 CPU's random number generator (if available) to
4538 fully seed the kernel's CRNG. Default is controlled
4539 by CONFIG_RANDOM_TRUST_CPU.
4541 random.trust_bootloader={on,off}
4542 [KNL] Enable or disable trusting the use of a
4543 seed passed by the bootloader (if available) to
4544 fully seed the kernel's CRNG. Default is controlled
4545 by CONFIG_RANDOM_TRUST_BOOTLOADER.
4547 randomize_kstack_offset=
4548 [KNL] Enable or disable kernel stack offset
4549 randomization, which provides roughly 5 bits of
4550 entropy, frustrating memory corruption attacks
4551 that depend on stack address determinism or
4552 cross-syscall address exposures. This is only
4553 available on architectures that have defined
4554 CONFIG_HAVE_ARCH_RANDOMIZE_KSTACK_OFFSET.
4555 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
4556 Default is CONFIG_RANDOMIZE_KSTACK_OFFSET_DEFAULT.
4558 ras=option[,option,...] [KNL] RAS-specific options
4561 Disable the Correctable Errors Collector,
4562 see CONFIG_RAS_CEC help text.
4564 rcu_nocbs[=cpu-list]
4565 [KNL] The optional argument is a cpu list,
4568 In kernels built with CONFIG_RCU_NOCB_CPU=y,
4569 enable the no-callback CPU mode, which prevents
4570 such CPUs' callbacks from being invoked in
4571 softirq context. Invocation of such CPUs' RCU
4572 callbacks will instead be offloaded to "rcuox/N"
4573 kthreads created for that purpose, where "x" is
4574 "p" for RCU-preempt, "s" for RCU-sched, and "g"
4575 for the kthreads that mediate grace periods; and
4576 "N" is the CPU number. This reduces OS jitter on
4577 the offloaded CPUs, which can be useful for HPC
4578 and real-time workloads. It can also improve
4579 energy efficiency for asymmetric multiprocessors.
4581 If a cpulist is passed as an argument, the specified
4582 list of CPUs is set to no-callback mode from boot.
4584 Otherwise, if the '=' sign and the cpulist
4585 arguments are omitted, no CPU will be set to
4586 no-callback mode from boot but the mode may be
4587 toggled at runtime via cpusets.
4589 Note that this argument takes precedence over
4590 the CONFIG_RCU_NOCB_CPU_DEFAULT_ALL option.
4593 Rather than requiring that offloaded CPUs
4594 (specified by rcu_nocbs= above) explicitly
4595 awaken the corresponding "rcuoN" kthreads,
4596 make these kthreads poll for callbacks.
4597 This improves the real-time response for the
4598 offloaded CPUs by relieving them of the need to
4599 wake up the corresponding kthread, but degrades
4600 energy efficiency by requiring that the kthreads
4601 periodically wake up to do the polling.
4603 rcutree.blimit= [KNL]
4604 Set maximum number of finished RCU callbacks to
4605 process in one batch.
4607 rcutree.dump_tree= [KNL]
4608 Dump the structure of the rcu_node combining tree
4609 out at early boot. This is used for diagnostic
4610 purposes, to verify correct tree setup.
4612 rcutree.gp_cleanup_delay= [KNL]
4613 Set the number of jiffies to delay each step of
4614 RCU grace-period cleanup.
4616 rcutree.gp_init_delay= [KNL]
4617 Set the number of jiffies to delay each step of
4618 RCU grace-period initialization.
4620 rcutree.gp_preinit_delay= [KNL]
4621 Set the number of jiffies to delay each step of
4622 RCU grace-period pre-initialization, that is,
4623 the propagation of recent CPU-hotplug changes up
4624 the rcu_node combining tree.
4626 rcutree.use_softirq= [KNL]
4627 If set to zero, move all RCU_SOFTIRQ processing to
4628 per-CPU rcuc kthreads. Defaults to a non-zero
4629 value, meaning that RCU_SOFTIRQ is used by default.
4630 Specify rcutree.use_softirq=0 to use rcuc kthreads.
4632 But note that CONFIG_PREEMPT_RT=y kernels disable
4633 this kernel boot parameter, forcibly setting it
4636 rcutree.rcu_fanout_exact= [KNL]
4637 Disable autobalancing of the rcu_node combining
4638 tree. This is used by rcutorture, and might
4639 possibly be useful for architectures having high
4640 cache-to-cache transfer latencies.
4642 rcutree.rcu_fanout_leaf= [KNL]
4643 Change the number of CPUs assigned to each
4644 leaf rcu_node structure. Useful for very
4645 large systems, which will choose the value 64,
4646 and for NUMA systems with large remote-access
4647 latencies, which will choose a value aligned
4648 with the appropriate hardware boundaries.
4650 rcutree.rcu_min_cached_objs= [KNL]
4651 Minimum number of objects which are cached and
4652 maintained per one CPU. Object size is equal
4653 to PAGE_SIZE. The cache allows to reduce the
4654 pressure to page allocator, also it makes the
4655 whole algorithm to behave better in low memory
4658 rcutree.rcu_delay_page_cache_fill_msec= [KNL]
4659 Set the page-cache refill delay (in milliseconds)
4660 in response to low-memory conditions. The range
4661 of permitted values is in the range 0:100000.
4663 rcutree.jiffies_till_first_fqs= [KNL]
4664 Set delay from grace-period initialization to
4665 first attempt to force quiescent states.
4666 Units are jiffies, minimum value is zero,
4667 and maximum value is HZ.
4669 rcutree.jiffies_till_next_fqs= [KNL]
4670 Set delay between subsequent attempts to force
4671 quiescent states. Units are jiffies, minimum
4672 value is one, and maximum value is HZ.
4674 rcutree.jiffies_till_sched_qs= [KNL]
4675 Set required age in jiffies for a
4676 given grace period before RCU starts
4677 soliciting quiescent-state help from
4678 rcu_note_context_switch() and cond_resched().
4679 If not specified, the kernel will calculate
4680 a value based on the most recent settings
4681 of rcutree.jiffies_till_first_fqs
4682 and rcutree.jiffies_till_next_fqs.
4683 This calculated value may be viewed in
4684 rcutree.jiffies_to_sched_qs. Any attempt to set
4685 rcutree.jiffies_to_sched_qs will be cheerfully
4688 rcutree.kthread_prio= [KNL,BOOT]
4689 Set the SCHED_FIFO priority of the RCU per-CPU
4690 kthreads (rcuc/N). This value is also used for
4691 the priority of the RCU boost threads (rcub/N)
4692 and for the RCU grace-period kthreads (rcu_bh,
4693 rcu_preempt, and rcu_sched). If RCU_BOOST is
4694 set, valid values are 1-99 and the default is 1
4695 (the least-favored priority). Otherwise, when
4696 RCU_BOOST is not set, valid values are 0-99 and
4697 the default is zero (non-realtime operation).
4698 When RCU_NOCB_CPU is set, also adjust the
4699 priority of NOCB callback kthreads.
4701 rcutree.rcu_divisor= [KNL]
4702 Set the shift-right count to use to compute
4703 the callback-invocation batch limit bl from
4704 the number of callbacks queued on this CPU.
4705 The result will be bounded below by the value of
4706 the rcutree.blimit kernel parameter. Every bl
4707 callbacks, the softirq handler will exit in
4708 order to allow the CPU to do other work.
4710 Please note that this callback-invocation batch
4711 limit applies only to non-offloaded callback
4712 invocation. Offloaded callbacks are instead
4713 invoked in the context of an rcuoc kthread, which
4714 scheduler will preempt as it does any other task.
4716 rcutree.nocb_nobypass_lim_per_jiffy= [KNL]
4717 On callback-offloaded (rcu_nocbs) CPUs,
4718 RCU reduces the lock contention that would
4719 otherwise be caused by callback floods through
4720 use of the ->nocb_bypass list. However, in the
4721 common non-flooded case, RCU queues directly to
4722 the main ->cblist in order to avoid the extra
4723 overhead of the ->nocb_bypass list and its lock.
4724 But if there are too many callbacks queued during
4725 a single jiffy, RCU pre-queues the callbacks into
4726 the ->nocb_bypass queue. The definition of "too
4727 many" is supplied by this kernel boot parameter.
4729 rcutree.rcu_nocb_gp_stride= [KNL]
4730 Set the number of NOCB callback kthreads in
4731 each group, which defaults to the square root
4732 of the number of CPUs. Larger numbers reduce
4733 the wakeup overhead on the global grace-period
4734 kthread, but increases that same overhead on
4735 each group's NOCB grace-period kthread.
4737 rcutree.qhimark= [KNL]
4738 Set threshold of queued RCU callbacks beyond which
4739 batch limiting is disabled.
4741 rcutree.qlowmark= [KNL]
4742 Set threshold of queued RCU callbacks below which
4743 batch limiting is re-enabled.
4745 rcutree.qovld= [KNL]
4746 Set threshold of queued RCU callbacks beyond which
4747 RCU's force-quiescent-state scan will aggressively
4748 enlist help from cond_resched() and sched IPIs to
4749 help CPUs more quickly reach quiescent states.
4750 Set to less than zero to make this be set based
4751 on rcutree.qhimark at boot time and to zero to
4752 disable more aggressive help enlistment.
4754 rcutree.rcu_kick_kthreads= [KNL]
4755 Cause the grace-period kthread to get an extra
4756 wake_up() if it sleeps three times longer than
4757 it should at force-quiescent-state time.
4758 This wake_up() will be accompanied by a
4759 WARN_ONCE() splat and an ftrace_dump().
4761 rcutree.rcu_unlock_delay= [KNL]
4762 In CONFIG_RCU_STRICT_GRACE_PERIOD=y kernels,
4763 this specifies an rcu_read_unlock()-time delay
4764 in microseconds. This defaults to zero.
4765 Larger delays increase the probability of
4766 catching RCU pointer leaks, that is, buggy use
4767 of RCU-protected pointers after the relevant
4768 rcu_read_unlock() has completed.
4770 rcutree.sysrq_rcu= [KNL]
4771 Commandeer a sysrq key to dump out Tree RCU's
4772 rcu_node tree with an eye towards determining
4773 why a new grace period has not yet started.
4775 rcuscale.gp_async= [KNL]
4776 Measure performance of asynchronous
4777 grace-period primitives such as call_rcu().
4779 rcuscale.gp_async_max= [KNL]
4780 Specify the maximum number of outstanding
4781 callbacks per writer thread. When a writer
4782 thread exceeds this limit, it invokes the
4783 corresponding flavor of rcu_barrier() to allow
4784 previously posted callbacks to drain.
4786 rcuscale.gp_exp= [KNL]
4787 Measure performance of expedited synchronous
4788 grace-period primitives.
4790 rcuscale.holdoff= [KNL]
4791 Set test-start holdoff period. The purpose of
4792 this parameter is to delay the start of the
4793 test until boot completes in order to avoid
4796 rcuscale.kfree_rcu_test= [KNL]
4797 Set to measure performance of kfree_rcu() flooding.
4799 rcuscale.kfree_rcu_test_double= [KNL]
4800 Test the double-argument variant of kfree_rcu().
4801 If this parameter has the same value as
4802 rcuscale.kfree_rcu_test_single, both the single-
4803 and double-argument variants are tested.
4805 rcuscale.kfree_rcu_test_single= [KNL]
4806 Test the single-argument variant of kfree_rcu().
4807 If this parameter has the same value as
4808 rcuscale.kfree_rcu_test_double, both the single-
4809 and double-argument variants are tested.
4811 rcuscale.kfree_nthreads= [KNL]
4812 The number of threads running loops of kfree_rcu().
4814 rcuscale.kfree_alloc_num= [KNL]
4815 Number of allocations and frees done in an iteration.
4817 rcuscale.kfree_loops= [KNL]
4818 Number of loops doing rcuscale.kfree_alloc_num number
4819 of allocations and frees.
4821 rcuscale.nreaders= [KNL]
4822 Set number of RCU readers. The value -1 selects
4823 N, where N is the number of CPUs. A value
4824 "n" less than -1 selects N-n+1, where N is again
4825 the number of CPUs. For example, -2 selects N
4826 (the number of CPUs), -3 selects N+1, and so on.
4827 A value of "n" less than or equal to -N selects
4830 rcuscale.nwriters= [KNL]
4831 Set number of RCU writers. The values operate
4832 the same as for rcuscale.nreaders.
4833 N, where N is the number of CPUs
4835 rcuscale.perf_type= [KNL]
4836 Specify the RCU implementation to test.
4838 rcuscale.shutdown= [KNL]
4839 Shut the system down after performance tests
4840 complete. This is useful for hands-off automated
4843 rcuscale.verbose= [KNL]
4844 Enable additional printk() statements.
4846 rcuscale.writer_holdoff= [KNL]
4847 Write-side holdoff between grace periods,
4848 in microseconds. The default of zero says
4851 rcutorture.fqs_duration= [KNL]
4852 Set duration of force_quiescent_state bursts
4855 rcutorture.fqs_holdoff= [KNL]
4856 Set holdoff time within force_quiescent_state bursts
4859 rcutorture.fqs_stutter= [KNL]
4860 Set wait time between force_quiescent_state bursts
4863 rcutorture.fwd_progress= [KNL]
4864 Specifies the number of kthreads to be used
4865 for RCU grace-period forward-progress testing
4866 for the types of RCU supporting this notion.
4867 Defaults to 1 kthread, values less than zero or
4868 greater than the number of CPUs cause the number
4871 rcutorture.fwd_progress_div= [KNL]
4872 Specify the fraction of a CPU-stall-warning
4873 period to do tight-loop forward-progress testing.
4875 rcutorture.fwd_progress_holdoff= [KNL]
4876 Number of seconds to wait between successive
4877 forward-progress tests.
4879 rcutorture.fwd_progress_need_resched= [KNL]
4880 Enclose cond_resched() calls within checks for
4881 need_resched() during tight-loop forward-progress
4884 rcutorture.gp_cond= [KNL]
4885 Use conditional/asynchronous update-side
4886 primitives, if available.
4888 rcutorture.gp_exp= [KNL]
4889 Use expedited update-side primitives, if available.
4891 rcutorture.gp_normal= [KNL]
4892 Use normal (non-expedited) asynchronous
4893 update-side primitives, if available.
4895 rcutorture.gp_sync= [KNL]
4896 Use normal (non-expedited) synchronous
4897 update-side primitives, if available. If all
4898 of rcutorture.gp_cond=, rcutorture.gp_exp=,
4899 rcutorture.gp_normal=, and rcutorture.gp_sync=
4900 are zero, rcutorture acts as if is interpreted
4901 they are all non-zero.
4903 rcutorture.irqreader= [KNL]
4904 Run RCU readers from irq handlers, or, more
4905 accurately, from a timer handler. Not all RCU
4906 flavors take kindly to this sort of thing.
4908 rcutorture.leakpointer= [KNL]
4909 Leak an RCU-protected pointer out of the reader.
4910 This can of course result in splats, and is
4911 intended to test the ability of things like
4912 CONFIG_RCU_STRICT_GRACE_PERIOD=y to detect
4915 rcutorture.n_barrier_cbs= [KNL]
4916 Set callbacks/threads for rcu_barrier() testing.
4918 rcutorture.nfakewriters= [KNL]
4919 Set number of concurrent RCU writers. These just
4920 stress RCU, they don't participate in the actual
4921 test, hence the "fake".
4923 rcutorture.nocbs_nthreads= [KNL]
4924 Set number of RCU callback-offload togglers.
4925 Zero (the default) disables toggling.
4927 rcutorture.nocbs_toggle= [KNL]
4928 Set the delay in milliseconds between successive
4929 callback-offload toggling attempts.
4931 rcutorture.nreaders= [KNL]
4932 Set number of RCU readers. The value -1 selects
4933 N-1, where N is the number of CPUs. A value
4934 "n" less than -1 selects N-n-2, where N is again
4935 the number of CPUs. For example, -2 selects N
4936 (the number of CPUs), -3 selects N+1, and so on.
4938 rcutorture.object_debug= [KNL]
4939 Enable debug-object double-call_rcu() testing.
4941 rcutorture.onoff_holdoff= [KNL]
4942 Set time (s) after boot for CPU-hotplug testing.
4944 rcutorture.onoff_interval= [KNL]
4945 Set time (jiffies) between CPU-hotplug operations,
4946 or zero to disable CPU-hotplug testing.
4948 rcutorture.read_exit= [KNL]
4949 Set the number of read-then-exit kthreads used
4950 to test the interaction of RCU updaters and
4951 task-exit processing.
4953 rcutorture.read_exit_burst= [KNL]
4954 The number of times in a given read-then-exit
4955 episode that a set of read-then-exit kthreads
4958 rcutorture.read_exit_delay= [KNL]
4959 The delay, in seconds, between successive
4960 read-then-exit testing episodes.
4962 rcutorture.shuffle_interval= [KNL]
4963 Set task-shuffle interval (s). Shuffling tasks
4964 allows some CPUs to go into dyntick-idle mode
4965 during the rcutorture test.
4967 rcutorture.shutdown_secs= [KNL]
4968 Set time (s) after boot system shutdown. This
4969 is useful for hands-off automated testing.
4971 rcutorture.stall_cpu= [KNL]
4972 Duration of CPU stall (s) to test RCU CPU stall
4973 warnings, zero to disable.
4975 rcutorture.stall_cpu_block= [KNL]
4976 Sleep while stalling if set. This will result
4977 in warnings from preemptible RCU in addition
4978 to any other stall-related activity.
4980 rcutorture.stall_cpu_holdoff= [KNL]
4981 Time to wait (s) after boot before inducing stall.
4983 rcutorture.stall_cpu_irqsoff= [KNL]
4984 Disable interrupts while stalling if set.
4986 rcutorture.stall_gp_kthread= [KNL]
4987 Duration (s) of forced sleep within RCU
4988 grace-period kthread to test RCU CPU stall
4989 warnings, zero to disable. If both stall_cpu
4990 and stall_gp_kthread are specified, the
4991 kthread is starved first, then the CPU.
4993 rcutorture.stat_interval= [KNL]
4994 Time (s) between statistics printk()s.
4996 rcutorture.stutter= [KNL]
4997 Time (s) to stutter testing, for example, specifying
4998 five seconds causes the test to run for five seconds,
4999 wait for five seconds, and so on. This tests RCU's
5000 ability to transition abruptly to and from idle.
5002 rcutorture.test_boost= [KNL]
5003 Test RCU priority boosting? 0=no, 1=maybe, 2=yes.
5004 "Maybe" means test if the RCU implementation
5005 under test support RCU priority boosting.
5007 rcutorture.test_boost_duration= [KNL]
5008 Duration (s) of each individual boost test.
5010 rcutorture.test_boost_interval= [KNL]
5011 Interval (s) between each boost test.
5013 rcutorture.test_no_idle_hz= [KNL]
5014 Test RCU's dyntick-idle handling. See also the
5015 rcutorture.shuffle_interval parameter.
5017 rcutorture.torture_type= [KNL]
5018 Specify the RCU implementation to test.
5020 rcutorture.verbose= [KNL]
5021 Enable additional printk() statements.
5023 rcupdate.rcu_cpu_stall_ftrace_dump= [KNL]
5024 Dump ftrace buffer after reporting RCU CPU
5027 rcupdate.rcu_cpu_stall_suppress= [KNL]
5028 Suppress RCU CPU stall warning messages.
5030 rcupdate.rcu_cpu_stall_suppress_at_boot= [KNL]
5031 Suppress RCU CPU stall warning messages and
5032 rcutorture writer stall warnings that occur
5033 during early boot, that is, during the time
5034 before the init task is spawned.
5036 rcupdate.rcu_cpu_stall_timeout= [KNL]
5037 Set timeout for RCU CPU stall warning messages.
5038 The value is in seconds and the maximum allowed
5039 value is 300 seconds.
5041 rcupdate.rcu_exp_cpu_stall_timeout= [KNL]
5042 Set timeout for expedited RCU CPU stall warning
5043 messages. The value is in milliseconds
5044 and the maximum allowed value is 21000
5045 milliseconds. Please note that this value is
5046 adjusted to an arch timer tick resolution.
5047 Setting this to zero causes the value from
5048 rcupdate.rcu_cpu_stall_timeout to be used (after
5049 conversion from seconds to milliseconds).
5051 rcupdate.rcu_expedited= [KNL]
5052 Use expedited grace-period primitives, for
5053 example, synchronize_rcu_expedited() instead
5054 of synchronize_rcu(). This reduces latency,
5055 but can increase CPU utilization, degrade
5056 real-time latency, and degrade energy efficiency.
5057 No effect on CONFIG_TINY_RCU kernels.
5059 rcupdate.rcu_normal= [KNL]
5060 Use only normal grace-period primitives,
5061 for example, synchronize_rcu() instead of
5062 synchronize_rcu_expedited(). This improves
5063 real-time latency, CPU utilization, and
5064 energy efficiency, but can expose users to
5065 increased grace-period latency. This parameter
5066 overrides rcupdate.rcu_expedited. No effect on
5067 CONFIG_TINY_RCU kernels.
5069 rcupdate.rcu_normal_after_boot= [KNL]
5070 Once boot has completed (that is, after
5071 rcu_end_inkernel_boot() has been invoked), use
5072 only normal grace-period primitives. No effect
5073 on CONFIG_TINY_RCU kernels.
5075 But note that CONFIG_PREEMPT_RT=y kernels enables
5076 this kernel boot parameter, forcibly setting
5077 it to the value one, that is, converting any
5078 post-boot attempt at an expedited RCU grace
5079 period to instead use normal non-expedited
5080 grace-period processing.
5082 rcupdate.rcu_task_collapse_lim= [KNL]
5083 Set the maximum number of callbacks present
5084 at the beginning of a grace period that allows
5085 the RCU Tasks flavors to collapse back to using
5086 a single callback queue. This switching only
5087 occurs when rcupdate.rcu_task_enqueue_lim is
5088 set to the default value of -1.
5090 rcupdate.rcu_task_contend_lim= [KNL]
5091 Set the minimum number of callback-queuing-time
5092 lock-contention events per jiffy required to
5093 cause the RCU Tasks flavors to switch to per-CPU
5094 callback queuing. This switching only occurs
5095 when rcupdate.rcu_task_enqueue_lim is set to
5096 the default value of -1.
5098 rcupdate.rcu_task_enqueue_lim= [KNL]
5099 Set the number of callback queues to use for the
5100 RCU Tasks family of RCU flavors. The default
5101 of -1 allows this to be automatically (and
5102 dynamically) adjusted. This parameter is intended
5105 rcupdate.rcu_task_ipi_delay= [KNL]
5106 Set time in jiffies during which RCU tasks will
5107 avoid sending IPIs, starting with the beginning
5108 of a given grace period. Setting a large
5109 number avoids disturbing real-time workloads,
5110 but lengthens grace periods.
5112 rcupdate.rcu_task_stall_info= [KNL]
5113 Set initial timeout in jiffies for RCU task stall
5114 informational messages, which give some indication
5115 of the problem for those not patient enough to
5116 wait for ten minutes. Informational messages are
5117 only printed prior to the stall-warning message
5118 for a given grace period. Disable with a value
5119 less than or equal to zero. Defaults to ten
5120 seconds. A change in value does not take effect
5121 until the beginning of the next grace period.
5123 rcupdate.rcu_task_stall_info_mult= [KNL]
5124 Multiplier for time interval between successive
5125 RCU task stall informational messages for a given
5126 RCU tasks grace period. This value is clamped
5127 to one through ten, inclusive. It defaults to
5128 the value three, so that the first informational
5129 message is printed 10 seconds into the grace
5130 period, the second at 40 seconds, the third at
5131 160 seconds, and then the stall warning at 600
5132 seconds would prevent a fourth at 640 seconds.
5134 rcupdate.rcu_task_stall_timeout= [KNL]
5135 Set timeout in jiffies for RCU task stall
5136 warning messages. Disable with a value less
5137 than or equal to zero. Defaults to ten minutes.
5138 A change in value does not take effect until
5139 the beginning of the next grace period.
5141 rcupdate.rcu_self_test= [KNL]
5142 Run the RCU early boot self tests
5146 Run specified binary instead of /init from the ramdisk,
5147 used for early userspace startup. See initrd.
5150 force - Override the decision by the kernel to hide the
5151 advertisement of RDRAND support (this affects
5152 certain AMD processors because of buggy BIOS
5153 support, specifically around the suspend/resume
5157 Turn on/off individual RDT features. List is:
5158 cmt, mbmtotal, mbmlocal, l3cat, l3cdp, l2cat, l2cdp,
5160 E.g. to turn on cmt and turn off mba use:
5164 Format (x86 or x86_64):
5165 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] | d[efault] \
5167 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \
5169 Where reboot_mode is one of warm (soft) or cold (hard) or gpio
5170 (prefix with 'panic_' to set mode for panic
5172 reboot_type is one of bios, acpi, kbd, triple, efi, or pci,
5173 reboot_force is either force or not specified,
5174 reboot_cpu is s[mp]#### with #### being the processor
5175 to be used for rebooting.
5177 refscale.holdoff= [KNL]
5178 Set test-start holdoff period. The purpose of
5179 this parameter is to delay the start of the
5180 test until boot completes in order to avoid
5183 refscale.loops= [KNL]
5184 Set the number of loops over the synchronization
5185 primitive under test. Increasing this number
5186 reduces noise due to loop start/end overhead,
5187 but the default has already reduced the per-pass
5188 noise to a handful of picoseconds on ca. 2020
5191 refscale.nreaders= [KNL]
5192 Set number of readers. The default value of -1
5193 selects N, where N is roughly 75% of the number
5194 of CPUs. A value of zero is an interesting choice.
5196 refscale.nruns= [KNL]
5197 Set number of runs, each of which is dumped onto
5200 refscale.readdelay= [KNL]
5201 Set the read-side critical-section duration,
5202 measured in microseconds.
5204 refscale.scale_type= [KNL]
5205 Specify the read-protection implementation to test.
5207 refscale.shutdown= [KNL]
5208 Shut down the system at the end of the performance
5209 test. This defaults to 1 (shut it down) when
5210 refscale is built into the kernel and to 0 (leave
5211 it running) when refscale is built as a module.
5213 refscale.verbose= [KNL]
5214 Enable additional printk() statements.
5216 refscale.verbose_batched= [KNL]
5217 Batch the additional printk() statements. If zero
5218 (the default) or negative, print everything. Otherwise,
5219 print every Nth verbose statement, where N is the value
5223 [KNL, SMP] Set scheduler's default relax_domain_level.
5224 See Documentation/admin-guide/cgroup-v1/cpusets.rst.
5226 reserve= [KNL,BUGS] Force kernel to ignore I/O ports or memory
5227 Format: <base1>,<size1>[,<base2>,<size2>,...]
5228 Reserve I/O ports or memory so the kernel won't use
5229 them. If <base> is less than 0x10000, the region
5230 is assumed to be I/O ports; otherwise it is memory.
5232 reservetop= [X86-32]
5234 Reserves a hole at the top of the kernel virtual
5237 reset_devices [KNL] Force drivers to reset the underlying device
5238 during initialization.
5241 Specify the partition device for software suspend
5243 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>}
5245 resume_offset= [SWSUSP]
5246 Specify the offset from the beginning of the partition
5247 given by "resume=" at which the swap header is located,
5248 in <PAGE_SIZE> units (needed only for swap files).
5249 See Documentation/power/swsusp-and-swap-files.rst
5251 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to
5252 read the resume files
5254 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up.
5255 Useful for devices that are detected asynchronously
5256 (e.g. USB and MMC devices).
5258 retain_initrd [RAM] Keep initrd memory after extraction
5260 retbleed= [X86] Control mitigation of RETBleed (Arbitrary
5261 Speculative Code Execution with Return Instructions)
5265 auto - automatically select a migitation
5266 auto,nosmt - automatically select a mitigation,
5267 disabling SMT if necessary for
5268 the full mitigation (only on Zen1
5269 and older without STIBP).
5270 ibpb - mitigate short speculation windows on
5271 basic block boundaries too. Safe, highest
5273 unret - force enable untrained return thunks,
5274 only effective on AMD f15h-f17h
5276 unret,nosmt - like unret, will disable SMT when STIBP
5279 Selecting 'auto' will choose a mitigation method at run
5280 time according to the CPU.
5282 Not specifying this option is equivalent to retbleed=auto.
5284 rfkill.default_state=
5285 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm,
5286 etc. communication is blocked by default.
5289 rfkill.master_switch_mode=
5290 0 The "airplane mode" button does nothing.
5291 1 The "airplane mode" button toggles between everything
5292 blocked and the previous configuration.
5293 2 The "airplane mode" button toggles between everything
5294 blocked and everything unblocked.
5296 rhash_entries= [KNL,NET]
5297 Set number of hash buckets for route cache
5300 [KNL] Disable ring 3 MONITOR/MWAIT feature on supported
5303 ro [KNL] Mount root device read-only on boot
5306 on Mark read-only kernel memory as read-only (default).
5307 off Leave read-only kernel memory writable for debugging.
5310 Enable the uart passthrough on the designated usb port
5311 on Rockchip SoCs. When active, the signals of the
5312 debug-uart get routed to the D+ and D- pins of the usb
5313 port and the regular usb controller gets disabled.
5315 root= [KNL] Root filesystem
5316 See name_to_dev_t comment in init/do_mounts.c.
5318 rootdelay= [KNL] Delay (in seconds) to pause before attempting to
5319 mount the root filesystem
5321 rootflags= [KNL] Set root filesystem mount option string
5323 rootfstype= [KNL] Set root filesystem type
5325 rootwait [KNL] Wait (indefinitely) for root device to show up.
5326 Useful for devices that are detected asynchronously
5327 (e.g. USB and MMC devices).
5329 rproc_mem=nn[KMG][@address]
5330 [KNL,ARM,CMA] Remoteproc physical memory block.
5331 Memory area to be used by remote processor image,
5334 rw [KNL] Mount root device read-write on boot
5336 S [KNL] Run init in single mode
5338 s390_iommu= [HW,S390]
5339 Set s390 IOTLB flushing mode
5341 With strict flushing every unmap operation will result in
5342 an IOTLB flush. Default is lazy flushing before reuse,
5345 s390_iommu_aperture= [KNL,S390]
5346 Specifies the size of the per device DMA address space
5347 accessible through the DMA and IOMMU APIs as a decimal
5348 factor of the size of main memory.
5349 The default is 1 meaning that one can concurrently use
5350 as many DMA addresses as physical memory is installed,
5351 if supported by hardware, and thus map all of memory
5352 once. With a value of 2 one can map all of memory twice
5353 and so on. As a special case a factor of 0 imposes no
5354 restrictions other than those given by hardware at the
5355 cost of significant additional memory use for tables.
5358 See drivers/net/irda/sa1100_ir.c.
5360 sched_verbose [KNL] Enables verbose scheduler debug messages.
5362 schedstats= [KNL,X86] Enable or disable scheduled statistics.
5363 Allowed values are enable and disable. This feature
5364 incurs a small amount of overhead in the scheduler
5365 but is useful for debugging and performance tuning.
5367 sched_thermal_decay_shift=
5368 [KNL, SMP] Set a decay shift for scheduler thermal
5369 pressure signal. Thermal pressure signal follows the
5370 default decay period of other scheduler pelt
5371 signals(usually 32 ms but configurable). Setting
5372 sched_thermal_decay_shift will left shift the decay
5373 period for the thermal pressure signal by the shift
5375 i.e. with the default pelt decay period of 32 ms
5376 sched_thermal_decay_shift thermal pressure decay pr
5380 Format: integer between 0 and 10
5383 scftorture.holdoff= [KNL]
5384 Number of seconds to hold off before starting
5385 test. Defaults to zero for module insertion and
5386 to 10 seconds for built-in smp_call_function()
5389 scftorture.longwait= [KNL]
5390 Request ridiculously long waits randomly selected
5391 up to the chosen limit in seconds. Zero (the
5392 default) disables this feature. Please note
5393 that requesting even small non-zero numbers of
5394 seconds can result in RCU CPU stall warnings,
5395 softlockup complaints, and so on.
5397 scftorture.nthreads= [KNL]
5398 Number of kthreads to spawn to invoke the
5399 smp_call_function() family of functions.
5400 The default of -1 specifies a number of kthreads
5401 equal to the number of CPUs.
5403 scftorture.onoff_holdoff= [KNL]
5404 Number seconds to wait after the start of the
5405 test before initiating CPU-hotplug operations.
5407 scftorture.onoff_interval= [KNL]
5408 Number seconds to wait between successive
5409 CPU-hotplug operations. Specifying zero (which
5410 is the default) disables CPU-hotplug operations.
5412 scftorture.shutdown_secs= [KNL]
5413 The number of seconds following the start of the
5414 test after which to shut down the system. The
5415 default of zero avoids shutting down the system.
5416 Non-zero values are useful for automated tests.
5418 scftorture.stat_interval= [KNL]
5419 The number of seconds between outputting the
5420 current test statistics to the console. A value
5421 of zero disables statistics output.
5423 scftorture.stutter_cpus= [KNL]
5424 The number of jiffies to wait between each change
5425 to the set of CPUs under test.
5427 scftorture.use_cpus_read_lock= [KNL]
5428 Use use_cpus_read_lock() instead of the default
5429 preempt_disable() to disable CPU hotplug
5430 while invoking one of the smp_call_function*()
5433 scftorture.verbose= [KNL]
5434 Enable additional printk() statements.
5436 scftorture.weight_single= [KNL]
5437 The probability weighting to use for the
5438 smp_call_function_single() function with a zero
5439 "wait" parameter. A value of -1 selects the
5440 default if all other weights are -1. However,
5441 if at least one weight has some other value, a
5442 value of -1 will instead select a weight of zero.
5444 scftorture.weight_single_wait= [KNL]
5445 The probability weighting to use for the
5446 smp_call_function_single() function with a
5447 non-zero "wait" parameter. See weight_single.
5449 scftorture.weight_many= [KNL]
5450 The probability weighting to use for the
5451 smp_call_function_many() function with a zero
5452 "wait" parameter. See weight_single.
5453 Note well that setting a high probability for
5454 this weighting can place serious IPI load
5457 scftorture.weight_many_wait= [KNL]
5458 The probability weighting to use for the
5459 smp_call_function_many() function with a
5460 non-zero "wait" parameter. See weight_single
5463 scftorture.weight_all= [KNL]
5464 The probability weighting to use for the
5465 smp_call_function_all() function with a zero
5466 "wait" parameter. See weight_single and
5469 scftorture.weight_all_wait= [KNL]
5470 The probability weighting to use for the
5471 smp_call_function_all() function with a
5472 non-zero "wait" parameter. See weight_single
5475 skew_tick= [KNL] Offset the periodic timer tick per cpu to mitigate
5476 xtime_lock contention on larger systems, and/or RCU lock
5477 contention on all systems with CONFIG_MAXSMP set.
5478 Format: { "0" | "1" }
5479 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1"
5481 Note: increases power consumption, thus should only be
5482 enabled if running jitter sensitive (HPC/RT) workloads.
5484 security= [SECURITY] Choose a legacy "major" security module to
5485 enable at boot. This has been deprecated by the
5488 selinux= [SELINUX] Disable or enable SELinux at boot time.
5489 Format: { "0" | "1" }
5490 See security/selinux/Kconfig help text.
5495 apparmor= [APPARMOR] Disable or enable AppArmor at boot time
5496 Format: { "0" | "1" }
5497 See security/apparmor/Kconfig help text
5500 Default value is set via kernel config option.
5502 serialnumber [BUGS=X86-32]
5504 sev=option[,option...] [X86-64] See Documentation/x86/x86_64/boot-options.rst
5507 Maximal number of shapers.
5515 Enable merging of slabs with similar size when the
5516 kernel is built without CONFIG_SLAB_MERGE_DEFAULT.
5519 Disable merging of slabs with similar size. May be
5520 necessary if there is some reason to distinguish
5521 allocs to different slabs, especially in hardened
5522 environments where the risk of heap overflows and
5523 layout control by attackers can usually be
5524 frustrated by disabling merging. This will reduce
5525 most of the exposure of a heap attack to a single
5526 cache (risks via metadata attacks are mostly
5527 unchanged). Debug options disable merging on their
5529 For more information see Documentation/mm/slub.rst.
5531 slab_max_order= [MM, SLAB]
5532 Determines the maximum allowed order for slabs.
5533 A high setting may cause OOMs due to memory
5534 fragmentation. Defaults to 1 for systems with
5535 more than 32MB of RAM, 0 otherwise.
5537 slub_debug[=options[,slabs][;[options[,slabs]]...] [MM, SLUB]
5538 Enabling slub_debug allows one to determine the
5539 culprit if slab objects become corrupted. Enabling
5540 slub_debug can create guard zones around objects and
5541 may poison objects when not in use. Also tracks the
5542 last alloc / free. For more information see
5543 Documentation/mm/slub.rst.
5545 slub_max_order= [MM, SLUB]
5546 Determines the maximum allowed order for slabs.
5547 A high setting may cause OOMs due to memory
5548 fragmentation. For more information see
5549 Documentation/mm/slub.rst.
5551 slub_min_objects= [MM, SLUB]
5552 The minimum number of objects per slab. SLUB will
5553 increase the slab order up to slub_max_order to
5554 generate a sufficiently large slab able to contain
5555 the number of objects indicated. The higher the number
5556 of objects the smaller the overhead of tracking slabs
5557 and the less frequently locks need to be acquired.
5558 For more information see Documentation/mm/slub.rst.
5560 slub_min_order= [MM, SLUB]
5561 Determines the minimum page order for slabs. Must be
5562 lower than slub_max_order.
5563 For more information see Documentation/mm/slub.rst.
5565 slub_merge [MM, SLUB]
5566 Same with slab_merge.
5568 slub_nomerge [MM, SLUB]
5569 Same with slab_nomerge. This is supported for legacy.
5570 See slab_nomerge for more information.
5573 Format: <io1>[,<io2>[,...,<io8>]]
5575 smp.csd_lock_timeout= [KNL]
5576 Specify the period of time in milliseconds
5577 that smp_call_function() and friends will wait
5578 for a CPU to release the CSD lock. This is
5579 useful when diagnosing bugs involving CPUs
5580 disabling interrupts for extended periods
5581 of time. Defaults to 5,000 milliseconds, and
5582 setting a value of zero disables this feature.
5583 This feature may be more efficiently disabled
5584 using the csdlock_debug- kernel parameter.
5586 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices
5587 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port
5588 smsc-ircc2.ircc_sir= [HW] SIR base I/O port
5589 smsc-ircc2.ircc_fir= [HW] FIR base I/O port
5590 smsc-ircc2.ircc_irq= [HW] IRQ line
5591 smsc-ircc2.ircc_dma= [HW] DMA channel
5592 smsc-ircc2.ircc_transceiver= [HW] Transceiver type:
5593 0: Toshiba Satellite 1800 (GP data pin select)
5594 1: Fast pin select (default)
5597 smt= [KNL,S390] Set the maximum number of threads (logical
5598 CPUs) to use per physical CPU on systems capable of
5599 symmetric multithreading (SMT). Will be capped to the
5600 actual hardware limit.
5602 Default: -1 (no limit)
5605 [KNL] Should the soft-lockup detector generate panics.
5608 A value of 1 instructs the soft-lockup detector
5609 to panic the machine when a soft-lockup occurs. It is
5610 also controlled by the kernel.softlockup_panic sysctl
5611 and CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC, which is the
5612 respective build-time switch to that functionality.
5614 softlockup_all_cpu_backtrace=
5615 [KNL] Should the soft-lockup detector generate
5616 backtraces on all cpus.
5619 sonypi.*= [HW] Sony Programmable I/O Control Device driver
5620 See Documentation/admin-guide/laptops/sonypi.rst
5622 spectre_v2= [X86] Control mitigation of Spectre variant 2
5623 (indirect branch speculation) vulnerability.
5624 The default operation protects the kernel from
5627 on - unconditionally enable, implies
5629 off - unconditionally disable, implies
5631 auto - kernel detects whether your CPU model is
5634 Selecting 'on' will, and 'auto' may, choose a
5635 mitigation method at run time according to the
5636 CPU, the available microcode, the setting of the
5637 CONFIG_RETPOLINE configuration option, and the
5638 compiler with which the kernel was built.
5640 Selecting 'on' will also enable the mitigation
5641 against user space to user space task attacks.
5643 Selecting 'off' will disable both the kernel and
5644 the user space protections.
5646 Specific mitigations can also be selected manually:
5648 retpoline - replace indirect branches
5649 retpoline,generic - Retpolines
5650 retpoline,lfence - LFENCE; indirect branch
5651 retpoline,amd - alias for retpoline,lfence
5652 eibrs - enhanced IBRS
5653 eibrs,retpoline - enhanced IBRS + Retpolines
5654 eibrs,lfence - enhanced IBRS + LFENCE
5655 ibrs - use IBRS to protect kernel
5657 Not specifying this option is equivalent to
5661 [X86] Control mitigation of Spectre variant 2
5662 (indirect branch speculation) vulnerability between
5665 on - Unconditionally enable mitigations. Is
5666 enforced by spectre_v2=on
5668 off - Unconditionally disable mitigations. Is
5669 enforced by spectre_v2=off
5671 prctl - Indirect branch speculation is enabled,
5672 but mitigation can be enabled via prctl
5673 per thread. The mitigation control state
5674 is inherited on fork.
5677 - Like "prctl" above, but only STIBP is
5678 controlled per thread. IBPB is issued
5679 always when switching between different user
5683 - Same as "prctl" above, but all seccomp
5684 threads will enable the mitigation unless
5685 they explicitly opt out.
5688 - Like "seccomp" above, but only STIBP is
5689 controlled per thread. IBPB is issued
5690 always when switching between different
5691 user space processes.
5693 auto - Kernel selects the mitigation depending on
5694 the available CPU features and vulnerability.
5696 Default mitigation: "prctl"
5698 Not specifying this option is equivalent to
5699 spectre_v2_user=auto.
5701 spec_store_bypass_disable=
5702 [HW] Control Speculative Store Bypass (SSB) Disable mitigation
5703 (Speculative Store Bypass vulnerability)
5705 Certain CPUs are vulnerable to an exploit against a
5706 a common industry wide performance optimization known
5707 as "Speculative Store Bypass" in which recent stores
5708 to the same memory location may not be observed by
5709 later loads during speculative execution. The idea
5710 is that such stores are unlikely and that they can
5711 be detected prior to instruction retirement at the
5712 end of a particular speculation execution window.
5714 In vulnerable processors, the speculatively forwarded
5715 store can be used in a cache side channel attack, for
5716 example to read memory to which the attacker does not
5717 directly have access (e.g. inside sandboxed code).
5719 This parameter controls whether the Speculative Store
5720 Bypass optimization is used.
5722 On x86 the options are:
5724 on - Unconditionally disable Speculative Store Bypass
5725 off - Unconditionally enable Speculative Store Bypass
5726 auto - Kernel detects whether the CPU model contains an
5727 implementation of Speculative Store Bypass and
5728 picks the most appropriate mitigation. If the
5729 CPU is not vulnerable, "off" is selected. If the
5730 CPU is vulnerable the default mitigation is
5731 architecture and Kconfig dependent. See below.
5732 prctl - Control Speculative Store Bypass per thread
5733 via prctl. Speculative Store Bypass is enabled
5734 for a process by default. The state of the control
5735 is inherited on fork.
5736 seccomp - Same as "prctl" above, but all seccomp threads
5737 will disable SSB unless they explicitly opt out.
5739 Default mitigations:
5742 On powerpc the options are:
5744 on,auto - On Power8 and Power9 insert a store-forwarding
5745 barrier on kernel entry and exit. On Power7
5746 perform a software flush on kernel entry and
5750 Not specifying this option is equivalent to
5751 spec_store_bypass_disable=auto.
5753 spia_io_base= [HW,MTD]
5759 [X86] Enable split lock detection or bus lock detection
5761 When enabled (and if hardware support is present), atomic
5762 instructions that access data across cache line
5763 boundaries will result in an alignment check exception
5764 for split lock detection or a debug exception for
5769 warn - the kernel will emit rate-limited warnings
5770 about applications triggering the #AC
5771 exception or the #DB exception. This mode is
5772 the default on CPUs that support split lock
5773 detection or bus lock detection. Default
5774 behavior is by #AC if both features are
5775 enabled in hardware.
5777 fatal - the kernel will send SIGBUS to applications
5778 that trigger the #AC exception or the #DB
5779 exception. Default behavior is by #AC if
5780 both features are enabled in hardware.
5783 Set system wide rate limit to N bus locks
5784 per second for bus lock detection.
5787 N/A for split lock detection.
5790 If an #AC exception is hit in the kernel or in
5791 firmware (i.e. not while executing in user mode)
5792 the kernel will oops in either "warn" or "fatal"
5795 #DB exception for bus lock is triggered only when
5799 Control the Special Register Buffer Data Sampling
5802 Certain CPUs are vulnerable to an MDS-like
5803 exploit which can leak bits from the random
5806 By default, this issue is mitigated by
5807 microcode. However, the microcode fix can cause
5808 the RDRAND and RDSEED instructions to become
5809 much slower. Among other effects, this will
5810 result in reduced throughput from /dev/urandom.
5812 The microcode mitigation can be disabled with
5813 the following option:
5815 off: Disable mitigation and remove
5816 performance impact to RDRAND and RDSEED
5818 srcutree.big_cpu_lim [KNL]
5819 Specifies the number of CPUs constituting a
5820 large system, such that srcu_struct structures
5821 should immediately allocate an srcu_node array.
5822 This kernel-boot parameter defaults to 128,
5823 but takes effect only when the low-order four
5824 bits of srcutree.convert_to_big is equal to 3
5827 srcutree.convert_to_big [KNL]
5828 Specifies under what conditions an SRCU tree
5829 srcu_struct structure will be converted to big
5830 form, that is, with an rcu_node tree:
5833 1: At init_srcu_struct() time.
5834 2: When rcutorture decides to.
5835 3: Decide at boot time (default).
5836 0x1X: Above plus if high contention.
5838 Either way, the srcu_node tree will be sized based
5839 on the actual runtime number of CPUs (nr_cpu_ids)
5840 instead of the compile-time CONFIG_NR_CPUS.
5842 srcutree.counter_wrap_check [KNL]
5843 Specifies how frequently to check for
5844 grace-period sequence counter wrap for the
5845 srcu_data structure's ->srcu_gp_seq_needed field.
5846 The greater the number of bits set in this kernel
5847 parameter, the less frequently counter wrap will
5848 be checked for. Note that the bottom two bits
5851 srcutree.exp_holdoff [KNL]
5852 Specifies how many nanoseconds must elapse
5853 since the end of the last SRCU grace period for
5854 a given srcu_struct until the next normal SRCU
5855 grace period will be considered for automatic
5856 expediting. Set to zero to disable automatic
5859 srcutree.srcu_max_nodelay [KNL]
5860 Specifies the number of no-delay instances
5861 per jiffy for which the SRCU grace period
5862 worker thread will be rescheduled with zero
5863 delay. Beyond this limit, worker thread will
5864 be rescheduled with a sleep delay of one jiffy.
5866 srcutree.srcu_max_nodelay_phase [KNL]
5867 Specifies the per-grace-period phase, number of
5868 non-sleeping polls of readers. Beyond this limit,
5869 grace period worker thread will be rescheduled
5870 with a sleep delay of one jiffy, between each
5871 rescan of the readers, for a grace period phase.
5873 srcutree.srcu_retry_check_delay [KNL]
5874 Specifies number of microseconds of non-sleeping
5875 delay between each non-sleeping poll of readers.
5877 srcutree.small_contention_lim [KNL]
5878 Specifies the number of update-side contention
5879 events per jiffy will be tolerated before
5880 initiating a conversion of an srcu_struct
5881 structure to big form. Note that the value of
5882 srcutree.convert_to_big must have the 0x10 bit
5883 set for contention-based conversions to occur.
5886 Speculative Store Bypass Disable control
5888 On CPUs that are vulnerable to the Speculative
5889 Store Bypass vulnerability and offer a
5890 firmware based mitigation, this parameter
5891 indicates how the mitigation should be used:
5893 force-on: Unconditionally enable mitigation for
5894 for both kernel and userspace
5895 force-off: Unconditionally disable mitigation for
5896 for both kernel and userspace
5897 kernel: Always enable mitigation in the
5898 kernel, and offer a prctl interface
5899 to allow userspace to register its
5900 interest in being mitigated too.
5902 stack_guard_gap= [MM]
5903 override the default stack gap protection. The value
5904 is in page units and it defines how many pages prior
5905 to (for stacks growing down) resp. after (for stacks
5906 growing up) the main stack are reserved for no other
5907 mapping. Default value is 256 pages.
5909 stack_depot_disable= [KNL]
5910 Setting this to true through kernel command line will
5911 disable the stack depot thereby saving the static memory
5912 consumed by the stack hash table. By default this is set
5916 Enabled the stack tracer on boot up.
5918 stacktrace_filter=[function-list]
5919 [FTRACE] Limit the functions that the stack tracer
5920 will trace at boot up. function-list is a comma-separated
5921 list of functions. This list can be changed at run
5922 time by the stack_trace_filter file in the debugfs
5923 tracing directory. Note, this enables stack tracing
5924 and the stacktrace above is not needed.
5928 Set the STI (builtin display/keyboard on the HP-PARISC
5929 machines) console (graphic card) which should be used
5930 as the initial boot-console.
5931 See also comment in drivers/video/console/sticore.c.
5934 See comment in drivers/video/console/sticore.c.
5937 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]]
5942 Enable or disable strict sigaltstack size checks
5943 against the required signal frame size which
5944 depends on the supported FPU features. This can
5945 be used to filter out binaries which have
5946 not yet been made aware of AT_MINSIGSTKSZ.
5948 sunrpc.min_resvport=
5949 sunrpc.max_resvport=
5951 SunRPC servers often require that client requests
5952 originate from a privileged port (i.e. a port in the
5953 range 0 < portnr < 1024).
5954 An administrator who wishes to reserve some of these
5955 ports for other uses may adjust the range that the
5956 kernel's sunrpc client considers to be privileged
5957 using these two parameters to set the minimum and
5958 maximum port values.
5960 sunrpc.svc_rpc_per_connection_limit=
5962 Limit the number of requests that the server will
5963 process in parallel from a single connection.
5964 The default value is 0 (no limit).
5968 Control how the NFS server code allocates CPUs to
5969 service thread pools. Depending on how many NICs
5970 you have and where their interrupts are bound, this
5971 option will affect which CPUs will do NFS serving.
5972 Note: this parameter cannot be changed while the
5973 NFS server is running.
5975 auto the server chooses an appropriate mode
5976 automatically using heuristics
5977 global a single global pool contains all CPUs
5978 percpu one pool for each CPU
5979 pernode one pool for each NUMA node (equivalent
5980 to global on non-NUMA machines)
5982 sunrpc.tcp_slot_table_entries=
5983 sunrpc.udp_slot_table_entries=
5985 Sets the upper limit on the number of simultaneous
5986 RPC calls that can be sent from the client to a
5987 server. Increasing these values may allow you to
5988 improve throughput, but will also increase the
5989 amount of memory reserved for use by the client.
5991 suspend.pm_test_delay=
5993 Sets the number of seconds to remain in a suspend test
5994 mode before resuming the system (see
5995 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG
5996 is set. Default value is 5.
5999 Format: { on | off | y | n | 1 | 0 }
6000 This parameter controls use of the Protected
6001 Execution Facility on pSeries.
6005 Enable accounting of swap in memory resource
6006 controller if no parameter or 1 is given or disable
6007 it if 0 is given (See Documentation/admin-guide/cgroup-v1/memory.rst)
6009 swiotlb= [ARM,IA-64,PPC,MIPS,X86]
6010 Format: { <int> [,<int>] | force | noforce }
6011 <int> -- Number of I/O TLB slabs
6012 <int> -- Second integer after comma. Number of swiotlb
6013 areas with their own lock. Will be rounded up
6015 force -- force using of bounce buffers even if they
6016 wouldn't be automatically used by the kernel
6017 noforce -- Never use bounce buffers (for debugging)
6022 Set a sysctl parameter, right before loading the init
6023 process, as if the value was written to the respective
6024 /proc/sys/... file. Both '.' and '/' are recognized as
6025 separators. Unrecognized parameters and invalid values
6026 are reported in the kernel log. Sysctls registered
6027 later by a loaded module cannot be set this way.
6028 Example: sysctl.vm.swappiness=40
6030 sysfs.deprecated=0|1 [KNL]
6031 Enable/disable old style sysfs layout for old udev
6032 on older distributions. When this option is enabled
6033 very new udev will not work anymore. When this option
6034 is disabled (or CONFIG_SYSFS_DEPRECATED not compiled)
6035 in older udev will not work anymore.
6036 Default depends on CONFIG_SYSFS_DEPRECATED_V2 set in
6037 the kernel configuration.
6039 sysrq_always_enabled
6041 Ignore sysrq setting - this boot parameter will
6042 neutralize any effect of /proc/sys/kernel/sysrq.
6043 Useful for debugging.
6045 tcpmhash_entries= [KNL,NET]
6046 Set the number of tcp_metrics_hash slots.
6047 Default value is 8192 or 16384 depending on total
6048 ram pages. This is used to specify the TCP metrics
6049 cache size. See Documentation/networking/ip-sysctl.rst
6050 "tcp_no_metrics_save" section for more details.
6054 test_suspend= [SUSPEND]
6055 Format: { "mem" | "standby" | "freeze" }[,N]
6056 Specify "mem" (for Suspend-to-RAM) or "standby" (for
6057 standby suspend) or "freeze" (for suspend type freeze)
6058 as the system sleep state during system startup with
6059 the optional capability to repeat N number of times.
6060 The system is woken from this state using a
6061 wakeup-capable RTC alarm.
6063 thash_entries= [KNL,NET]
6064 Set number of hash buckets for TCP connection
6066 thermal.act= [HW,ACPI]
6067 -1: disable all active trip points in all thermal zones
6068 <degrees C>: override all lowest active trip points
6070 thermal.crt= [HW,ACPI]
6071 -1: disable all critical trip points in all thermal zones
6072 <degrees C>: override all critical trip points
6074 thermal.nocrt= [HW,ACPI]
6075 Set to disable actions on ACPI thermal zone
6076 critical and hot trip points.
6078 thermal.off= [HW,ACPI]
6079 1: disable ACPI thermal control
6081 thermal.psv= [HW,ACPI]
6082 -1: disable all passive trip points
6083 <degrees C>: override all passive trip points to this
6086 thermal.tzp= [HW,ACPI]
6087 Specify global default ACPI thermal zone polling rate
6088 <deci-seconds>: poll all this frequency
6089 0: no polling (default)
6092 Force threading of all interrupt handlers except those
6093 marked explicitly IRQF_NO_THREAD.
6097 Specify if the kernel should make use of the cpu
6098 topology information if the hardware supports this.
6099 The scheduler will make use of this information and
6100 e.g. base its process migration decisions on it.
6103 topology_updates= [KNL, PPC, NUMA]
6105 Specify if the kernel should ignore (off)
6106 topology updates sent by the hypervisor to this
6109 torture.disable_onoff_at_boot= [KNL]
6110 Prevent the CPU-hotplug component of torturing
6111 until after init has spawned.
6113 torture.ftrace_dump_at_shutdown= [KNL]
6114 Dump the ftrace buffer at torture-test shutdown,
6115 even if there were no errors. This can be a
6116 very costly operation when many torture tests
6117 are running concurrently, especially on systems
6118 with rotating-rust storage.
6120 torture.verbose_sleep_frequency= [KNL]
6121 Specifies how many verbose printk()s should be
6122 emitted between each sleep. The default of zero
6123 disables verbose-printk() sleeping.
6125 torture.verbose_sleep_duration= [KNL]
6126 Duration of each verbose-printk() sleep in jiffies.
6130 tpm_suspend_pcr=[HW,TPM]
6131 Format: integer pcr id
6132 Specify that at suspend time, the tpm driver
6133 should extend the specified pcr with zeros,
6134 as a workaround for some chips which fail to
6135 flush the last written pcr on TPM_SaveState.
6136 This will guarantee that all the other pcrs
6140 Have the tracepoints sent to printk as well as the
6141 tracing ring buffer. This is useful for early boot up
6142 where the system hangs or reboots and does not give the
6143 option for reading the tracing buffer or performing a
6144 ftrace_dump_on_oops.
6146 To turn off having tracepoints sent to printk,
6147 echo 0 > /proc/sys/kernel/tracepoint_printk
6148 Note, echoing 1 into this file without the
6149 tracepoint_printk kernel cmdline option has no effect.
6151 The tp_printk_stop_on_boot (see below) can also be used
6152 to stop the printing of events to console at
6157 Having tracepoints sent to printk() and activating high
6158 frequency tracepoints such as irq or sched, can cause
6159 the system to live lock.
6161 tp_printk_stop_on_boot [FTRACE]
6162 When tp_printk (above) is set, it can cause a lot of noise
6163 on the console. It may be useful to only include the
6164 printing of events during boot up, as user space may
6165 make the system inoperable.
6167 This command line option will stop the printing of events
6168 to console at the late_initcall_sync() time frame.
6170 trace_buf_size=nn[KMG]
6171 [FTRACE] will set tracing buffer size on each cpu.
6173 trace_clock= [FTRACE] Set the clock used for tracing events
6175 local - Use the per CPU time stamp counter
6176 (converted into nanoseconds). Fast, but
6177 depending on the architecture, may not be
6178 in sync between CPUs.
6179 global - Event time stamps are synchronize across
6180 CPUs. May be slower than the local clock,
6181 but better for some race conditions.
6182 counter - Simple counting of events (1, 2, ..)
6183 note, some counts may be skipped due to the
6184 infrastructure grabbing the clock more than
6186 uptime - Use jiffies as the time stamp.
6187 perf - Use the same clock that perf uses.
6188 mono - Use ktime_get_mono_fast_ns() for time stamps.
6189 mono_raw - Use ktime_get_raw_fast_ns() for time
6191 boot - Use ktime_get_boot_fast_ns() for time stamps.
6192 Architectures may add more clocks. See
6193 Documentation/trace/ftrace.rst for more details.
6195 trace_event=[event-list]
6196 [FTRACE] Set and start specified trace events in order
6197 to facilitate early boot debugging. The event-list is a
6198 comma-separated list of trace events to enable. See
6199 also Documentation/trace/events.rst
6201 trace_options=[option-list]
6202 [FTRACE] Enable or disable tracer options at boot.
6203 The option-list is a comma delimited list of options
6204 that can be enabled or disabled just as if you were
6205 to echo the option name into
6207 /sys/kernel/debug/tracing/trace_options
6209 For example, to enable stacktrace option (to dump the
6210 stack trace of each event), add to the command line:
6212 trace_options=stacktrace
6214 See also Documentation/trace/ftrace.rst "trace options"
6218 [FTRACE] enable this option to disable tracing when a
6219 warning is hit. This turns off "tracing_on". Tracing can
6220 be enabled again by echoing '1' into the "tracing_on"
6221 file located in /sys/kernel/debug/tracing/
6223 This option is useful, as it disables the trace before
6224 the WARNING dump is called, which prevents the trace to
6225 be filled with content caused by the warning output.
6227 This option can also be set at run time via the sysctl
6228 option: kernel/traceoff_on_warning
6230 transparent_hugepage=
6232 Format: [always|madvise|never]
6233 Can be used to control the default behavior of the system
6234 with respect to transparent hugepages.
6235 See Documentation/admin-guide/mm/transhuge.rst
6238 trusted.source= [KEYS]
6240 This parameter identifies the trust source as a backend
6241 for trusted keys implementation. Supported trust
6246 If not specified then it defaults to iterating through
6247 the trust source list starting with TPM and assigns the
6248 first trust source as a backend which is initialized
6249 successfully during iteration.
6253 The RNG used to generate key material for trusted keys.
6256 - the same value as trusted.source: "tpm" or "tee"
6258 If not specified, "default" is used. In this case,
6259 the RNG's choice is left to each individual trust source.
6261 tsc= Disable clocksource stability checks for TSC.
6263 [x86] reliable: mark tsc clocksource as reliable, this
6264 disables clocksource verification at runtime, as well
6265 as the stability checks done at bootup. Used to enable
6266 high-resolution timer mode on older hardware, and in
6267 virtualized environment.
6268 [x86] noirqtime: Do not use TSC to do irq accounting.
6269 Used to run time disable IRQ_TIME_ACCOUNTING on any
6270 platforms where RDTSC is slow and this accounting
6272 [x86] unstable: mark the TSC clocksource as unstable, this
6273 marks the TSC unconditionally unstable at bootup and
6274 avoids any further wobbles once the TSC watchdog notices.
6275 [x86] nowatchdog: disable clocksource watchdog. Used
6276 in situations with strict latency requirements (where
6277 interruptions from clocksource watchdog are not
6280 tsc_early_khz= [X86] Skip early TSC calibration and use the given
6281 value instead. Useful when the early TSC frequency discovery
6282 procedure is not reliable, such as on overclocked systems
6283 with CPUID.16h support and partial CPUID.15h support.
6284 Format: <unsigned int>
6286 tsx= [X86] Control Transactional Synchronization
6287 Extensions (TSX) feature in Intel processors that
6288 support TSX control.
6290 This parameter controls the TSX feature. The options are:
6292 on - Enable TSX on the system. Although there are
6293 mitigations for all known security vulnerabilities,
6294 TSX has been known to be an accelerator for
6295 several previous speculation-related CVEs, and
6296 so there may be unknown security risks associated
6297 with leaving it enabled.
6299 off - Disable TSX on the system. (Note that this
6300 option takes effect only on newer CPUs which are
6301 not vulnerable to MDS, i.e., have
6302 MSR_IA32_ARCH_CAPABILITIES.MDS_NO=1 and which get
6303 the new IA32_TSX_CTRL MSR through a microcode
6304 update. This new MSR allows for the reliable
6305 deactivation of the TSX functionality.)
6307 auto - Disable TSX if X86_BUG_TAA is present,
6308 otherwise enable TSX on the system.
6310 Not specifying this option is equivalent to tsx=off.
6312 See Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
6315 tsx_async_abort= [X86,INTEL] Control mitigation for the TSX Async
6316 Abort (TAA) vulnerability.
6318 Similar to Micro-architectural Data Sampling (MDS)
6319 certain CPUs that support Transactional
6320 Synchronization Extensions (TSX) are vulnerable to an
6321 exploit against CPU internal buffers which can forward
6322 information to a disclosure gadget under certain
6325 In vulnerable processors, the speculatively forwarded
6326 data can be used in a cache side channel attack, to
6327 access data to which the attacker does not have direct
6330 This parameter controls the TAA mitigation. The
6333 full - Enable TAA mitigation on vulnerable CPUs
6336 full,nosmt - Enable TAA mitigation and disable SMT on
6337 vulnerable CPUs. If TSX is disabled, SMT
6338 is not disabled because CPU is not
6339 vulnerable to cross-thread TAA attacks.
6340 off - Unconditionally disable TAA mitigation
6342 On MDS-affected machines, tsx_async_abort=off can be
6343 prevented by an active MDS mitigation as both vulnerabilities
6344 are mitigated with the same mechanism so in order to disable
6345 this mitigation, you need to specify mds=off too.
6347 Not specifying this option is equivalent to
6348 tsx_async_abort=full. On CPUs which are MDS affected
6349 and deploy MDS mitigation, TAA mitigation is not
6350 required and doesn't provide any additional
6354 Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
6356 turbografx.map[2|3]= [HW,JOY]
6357 TurboGraFX parallel port interface
6359 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7>
6360 See also Documentation/input/devices/joystick-parport.rst
6362 udbg-immortal [PPC] When debugging early kernel crashes that
6363 happen after console_init() and before a proper
6364 console driver takes over, this boot options might
6365 help "seeing" what's going on.
6367 uhash_entries= [KNL,NET]
6368 Set number of hash buckets for UDP/UDP-Lite connections
6371 [USB] Ignore overcurrent events (default N).
6372 Some badly-designed motherboards generate lots of
6373 bogus events, for ports that aren't wired to
6374 anything. Set this parameter to avoid log spamming.
6375 Note that genuine overcurrent events won't be
6379 [X86] Cause panic on unknown NMI.
6381 usbcore.authorized_default=
6382 [USB] Default USB device authorization:
6383 (default -1 = authorized except for wireless USB,
6384 0 = not authorized, 1 = authorized, 2 = authorized
6385 if device connected to internal port)
6387 usbcore.autosuspend=
6388 [USB] The autosuspend time delay (in seconds) used
6389 for newly-detected USB devices (default 2). This
6390 is the time required before an idle device will be
6391 autosuspended. Devices for which the delay is set
6392 to a negative value won't be autosuspended at all.
6394 usbcore.usbfs_snoop=
6395 [USB] Set to log all usbfs traffic (default 0 = off).
6397 usbcore.usbfs_snoop_max=
6398 [USB] Maximum number of bytes to snoop in each URB
6401 usbcore.blinkenlights=
6402 [USB] Set to cycle leds on hubs (default 0 = off).
6404 usbcore.old_scheme_first=
6405 [USB] Start with the old device initialization
6406 scheme (default 0 = off).
6408 usbcore.usbfs_memory_mb=
6409 [USB] Memory limit (in MB) for buffers allocated by
6410 usbfs (default = 16, 0 = max = 2047).
6412 usbcore.use_both_schemes=
6413 [USB] Try the other device initialization scheme
6414 if the first one fails (default 1 = enabled).
6416 usbcore.initial_descriptor_timeout=
6417 [USB] Specifies timeout for the initial 64-byte
6418 USB_REQ_GET_DESCRIPTOR request in milliseconds
6419 (default 5000 = 5.0 seconds).
6421 usbcore.nousb [USB] Disable the USB subsystem
6424 [USB] A list of quirk entries to augment the built-in
6425 usb core quirk list. List entries are separated by
6426 commas. Each entry has the form
6427 VendorID:ProductID:Flags. The IDs are 4-digit hex
6428 numbers and Flags is a set of letters. Each letter
6429 will change the built-in quirk; setting it if it is
6430 clear and clearing it if it is set. The letters have
6431 the following meanings:
6432 a = USB_QUIRK_STRING_FETCH_255 (string
6433 descriptors must not be fetched using
6435 b = USB_QUIRK_RESET_RESUME (device can't resume
6436 correctly so reset it instead);
6437 c = USB_QUIRK_NO_SET_INTF (device can't handle
6438 Set-Interface requests);
6439 d = USB_QUIRK_CONFIG_INTF_STRINGS (device can't
6440 handle its Configuration or Interface
6442 e = USB_QUIRK_RESET (device can't be reset
6443 (e.g morph devices), don't use reset);
6444 f = USB_QUIRK_HONOR_BNUMINTERFACES (device has
6445 more interface descriptions than the
6446 bNumInterfaces count, and can't handle
6447 talking to these interfaces);
6448 g = USB_QUIRK_DELAY_INIT (device needs a pause
6449 during initialization, after we read
6450 the device descriptor);
6451 h = USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL (For
6452 high speed and super speed interrupt
6453 endpoints, the USB 2.0 and USB 3.0 spec
6454 require the interval in microframes (1
6455 microframe = 125 microseconds) to be
6456 calculated as interval = 2 ^
6458 Devices with this quirk report their
6459 bInterval as the result of this
6460 calculation instead of the exponent
6461 variable used in the calculation);
6462 i = USB_QUIRK_DEVICE_QUALIFIER (device can't
6463 handle device_qualifier descriptor
6465 j = USB_QUIRK_IGNORE_REMOTE_WAKEUP (device
6466 generates spurious wakeup, ignore
6467 remote wakeup capability);
6468 k = USB_QUIRK_NO_LPM (device can't handle Link
6470 l = USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL
6471 (Device reports its bInterval as linear
6472 frames instead of the USB 2.0
6474 m = USB_QUIRK_DISCONNECT_SUSPEND (Device needs
6475 to be disconnected before suspend to
6476 prevent spurious wakeup);
6477 n = USB_QUIRK_DELAY_CTRL_MSG (Device needs a
6478 pause after every control message);
6479 o = USB_QUIRK_HUB_SLOW_RESET (Hub needs extra
6480 delay after resetting its port);
6481 Example: quirks=0781:5580:bk,0a5c:5834:gij
6484 [USBHID] The interval which mice are to be polled at.
6487 [USBHID] The interval which joysticks are to be polled at.
6490 [USBHID] The interval which keyboards are to be polled at.
6492 usb-storage.delay_use=
6493 [UMS] The delay in seconds before a new device is
6494 scanned for Logical Units (default 1).
6497 [UMS] A list of quirks entries to supplement or
6498 override the built-in unusual_devs list. List
6499 entries are separated by commas. Each entry has
6500 the form VID:PID:Flags where VID and PID are Vendor
6501 and Product ID values (4-digit hex numbers) and
6502 Flags is a set of characters, each corresponding
6503 to a common usb-storage quirk flag as follows:
6504 a = SANE_SENSE (collect more than 18 bytes
6505 of sense data, not on uas);
6506 b = BAD_SENSE (don't collect more than 18
6507 bytes of sense data, not on uas);
6508 c = FIX_CAPACITY (decrease the reported
6509 device capacity by one sector);
6510 d = NO_READ_DISC_INFO (don't use
6511 READ_DISC_INFO command, not on uas);
6512 e = NO_READ_CAPACITY_16 (don't use
6513 READ_CAPACITY_16 command);
6514 f = NO_REPORT_OPCODES (don't use report opcodes
6516 g = MAX_SECTORS_240 (don't transfer more than
6517 240 sectors at a time, uas only);
6518 h = CAPACITY_HEURISTICS (decrease the
6519 reported device capacity by one
6520 sector if the number is odd);
6521 i = IGNORE_DEVICE (don't bind to this
6523 j = NO_REPORT_LUNS (don't use report luns
6525 k = NO_SAME (do not use WRITE_SAME, uas only)
6526 l = NOT_LOCKABLE (don't try to lock and
6527 unlock ejectable media, not on uas);
6528 m = MAX_SECTORS_64 (don't transfer more
6529 than 64 sectors = 32 KB at a time,
6531 n = INITIAL_READ10 (force a retry of the
6532 initial READ(10) command, not on uas);
6533 o = CAPACITY_OK (accept the capacity
6534 reported by the device, not on uas);
6535 p = WRITE_CACHE (the device cache is ON
6536 by default, not on uas);
6537 r = IGNORE_RESIDUE (the device reports
6538 bogus residue values, not on uas);
6539 s = SINGLE_LUN (the device has only one
6541 t = NO_ATA_1X (don't allow ATA(12) and ATA(16)
6542 commands, uas only);
6543 u = IGNORE_UAS (don't bind to the uas driver);
6544 w = NO_WP_DETECT (don't test whether the
6545 medium is write-protected).
6546 y = ALWAYS_SYNC (issue a SYNCHRONIZE_CACHE
6547 even if the device claims no cache,
6549 Example: quirks=0419:aaf5:rl,0421:0433:rc
6551 user_debug= [KNL,ARM]
6553 See arch/arm/Kconfig.debug help text.
6554 1 - undefined instruction events
6556 4 - invalid data aborts
6559 Example: user_debug=31
6562 [X86] Flags controlling user PTE allocations.
6564 nohigh = do not allocate PTE pages in
6565 HIGHMEM regardless of setting
6568 vdso= [X86,SH,SPARC]
6569 On X86_32, this is an alias for vdso32=. Otherwise:
6571 vdso=1: enable VDSO (the default)
6572 vdso=0: disable VDSO mapping
6574 vdso32= [X86] Control the 32-bit vDSO
6575 vdso32=1: enable 32-bit VDSO
6576 vdso32=0 or vdso32=2: disable 32-bit VDSO
6578 See the help text for CONFIG_COMPAT_VDSO for more
6579 details. If CONFIG_COMPAT_VDSO is set, the default is
6580 vdso32=0; otherwise, the default is vdso32=1.
6582 For compatibility with older kernels, vdso32=2 is an
6585 Try vdso32=0 if you encounter an error that says:
6586 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed!
6589 vector=percpu: enable percpu vector domain
6591 video= [FB] Frame buffer configuration
6592 See Documentation/fb/modedb.rst.
6594 video.brightness_switch_enabled= [ACPI]
6596 If set to 1, on receiving an ACPI notify event
6597 generated by hotkey, video driver will adjust brightness
6598 level and then send out the event to user space through
6599 the allocated input device. If set to 0, video driver
6600 will only send out the event without touching backlight
6605 [VMMIO] Memory mapped virtio (platform) device.
6607 <size>@<baseaddr>:<irq>[:<id>]
6609 <size> := size (can use standard suffixes
6611 <baseaddr> := physical base address
6612 <irq> := interrupt number (as passed to
6614 <id> := (optional) platform device id
6616 virtio_mmio.device=1K@0x100b0000:48:7
6618 Can be used multiple times for multiple devices.
6620 vga= [BOOT,X86-32] Select a particular video mode
6621 See Documentation/x86/boot.rst and
6622 Documentation/admin-guide/svga.rst.
6623 Use vga=ask for menu.
6624 This is actually a boot loader parameter; the value is
6625 passed to the kernel using a special protocol.
6627 vm_debug[=options] [KNL] Available with CONFIG_DEBUG_VM=y.
6628 May slow down system boot speed, especially when
6629 enabled on systems with a large amount of memory.
6630 All options are enabled by default, and this
6631 interface is meant to allow for selectively
6632 enabling or disabling specific virtual memory
6635 Available options are:
6636 P Enable page structure init time poisoning
6637 - Disable all of the above options
6639 vmalloc=nn[KMG] [KNL,BOOT] Forces the vmalloc area to have an exact
6640 size of <nn>. This can be used to increase the
6641 minimum size (128MB on x86). It can also be used to
6642 decrease the size and leave more room for directly
6645 vmcp_cma=nn[MG] [KNL,S390]
6646 Sets the memory size reserved for contiguous memory
6647 allocations for the vmcp device driver.
6649 vmhalt= [KNL,S390] Perform z/VM CP command after system halt.
6652 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic.
6655 vmpoff= [KNL,S390] Perform z/VM CP command after power off.
6659 Controls the behavior of vsyscalls (i.e. calls to
6660 fixed addresses of 0xffffffffff600x00 from legacy
6661 code). Most statically-linked binaries and older
6662 versions of glibc use these calls. Because these
6663 functions are at fixed addresses, they make nice
6664 targets for exploits that can control RIP.
6666 emulate [default] Vsyscalls turn into traps and are
6667 emulated reasonably safely. The vsyscall
6670 xonly Vsyscalls turn into traps and are
6671 emulated reasonably safely. The vsyscall
6672 page is not readable.
6674 none Vsyscalls don't work at all. This makes
6675 them quite hard to use for exploits but
6676 might break your system.
6678 vt.color= [VT] Default text color.
6679 Format: 0xYX, X = foreground, Y = background.
6680 Default: 0x07 = light gray on black.
6682 vt.cur_default= [VT] Default cursor shape.
6683 Format: 0xCCBBAA, where AA, BB, and CC are the same as
6684 the parameters of the <Esc>[?A;B;Cc escape sequence;
6685 see VGA-softcursor.txt. Default: 2 = underline.
6687 vt.default_blu= [VT]
6688 Format: <blue0>,<blue1>,<blue2>,...,<blue15>
6689 Change the default blue palette of the console.
6690 This is a 16-member array composed of values
6693 vt.default_grn= [VT]
6694 Format: <green0>,<green1>,<green2>,...,<green15>
6695 Change the default green palette of the console.
6696 This is a 16-member array composed of values
6699 vt.default_red= [VT]
6700 Format: <red0>,<red1>,<red2>,...,<red15>
6701 Change the default red palette of the console.
6702 This is a 16-member array composed of values
6708 Set system-wide default UTF-8 mode for all tty's.
6709 Default is 1, i.e. UTF-8 mode is enabled for all
6710 newly opened terminals.
6712 vt.global_cursor_default=
6715 Set system-wide default for whether a cursor
6716 is shown on new VTs. Default is -1,
6717 i.e. cursors will be created by default unless
6718 overridden by individual drivers. 0 will hide
6719 cursors, 1 will display them.
6721 vt.italic= [VT] Default color for italic text; 0-15.
6724 vt.underline= [VT] Default color for underlined text; 0-15.
6727 watchdog timers [HW,WDT] For information on watchdog timers,
6728 see Documentation/watchdog/watchdog-parameters.rst
6729 or other driver-specific files in the
6730 Documentation/watchdog/ directory.
6734 Set the hard lockup detector stall duration
6735 threshold in seconds. The soft lockup detector
6736 threshold is set to twice the value. A value of 0
6737 disables both lockup detectors. Default is 10
6740 workqueue.watchdog_thresh=
6741 If CONFIG_WQ_WATCHDOG is configured, workqueue can
6742 warn stall conditions and dump internal state to
6743 help debugging. 0 disables workqueue stall
6744 detection; otherwise, it's the stall threshold
6745 duration in seconds. The default value is 30 and
6746 it can be updated at runtime by writing to the
6747 corresponding sysfs file.
6749 workqueue.disable_numa
6750 By default, all work items queued to unbound
6751 workqueues are affine to the NUMA nodes they're
6752 issued on, which results in better behavior in
6753 general. If NUMA affinity needs to be disabled for
6754 whatever reason, this option can be used. Note
6755 that this also can be controlled per-workqueue for
6756 workqueues visible under /sys/bus/workqueue/.
6758 workqueue.power_efficient
6759 Per-cpu workqueues are generally preferred because
6760 they show better performance thanks to cache
6761 locality; unfortunately, per-cpu workqueues tend to
6762 be more power hungry than unbound workqueues.
6764 Enabling this makes the per-cpu workqueues which
6765 were observed to contribute significantly to power
6766 consumption unbound, leading to measurably lower
6767 power usage at the cost of small performance
6770 The default value of this parameter is determined by
6771 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT.
6773 workqueue.debug_force_rr_cpu
6774 Workqueue used to implicitly guarantee that work
6775 items queued without explicit CPU specified are put
6776 on the local CPU. This guarantee is no longer true
6777 and while local CPU is still preferred work items
6778 may be put on foreign CPUs. This debug option
6779 forces round-robin CPU selection to flush out
6780 usages which depend on the now broken guarantee.
6781 When enabled, memory and cache locality will be
6784 x2apic_phys [X86-64,APIC] Use x2apic physical mode instead of
6785 default x2apic cluster mode on platforms
6788 xen_512gb_limit [KNL,X86-64,XEN]
6789 Restricts the kernel running paravirtualized under Xen
6790 to use only up to 512 GB of RAM. The reason to do so is
6791 crash analysis tools and Xen tools for doing domain
6792 save/restore/migration must be enabled to handle larger
6795 xen_emul_unplug= [HW,X86,XEN]
6796 Unplug Xen emulated devices
6797 Format: [unplug0,][unplug1]
6798 ide-disks -- unplug primary master IDE devices
6799 aux-ide-disks -- unplug non-primary-master IDE devices
6800 nics -- unplug network devices
6801 all -- unplug all emulated devices (NICs and IDE disks)
6802 unnecessary -- unplugging emulated devices is
6803 unnecessary even if the host did not respond to
6805 never -- do not unplug even if version check succeeds
6807 xen_legacy_crash [X86,XEN]
6808 Crash from Xen panic notifier, without executing late
6809 panic() code such as dumping handler.
6811 xen_nopvspin [X86,XEN]
6812 Disables the qspinlock slowpath using Xen PV optimizations.
6813 This parameter is obsoleted by "nopvspin" parameter, which
6814 has equivalent effect for XEN platform.
6817 Disables the PV optimizations forcing the HVM guest to
6818 run as generic HVM guest with no PV drivers.
6819 This option is obsoleted by the "nopv" option, which
6820 has equivalent effect for XEN platform.
6822 xen_no_vector_callback
6823 [KNL,X86,XEN] Disable the vector callback for Xen
6824 event channel interrupts.
6826 xen_scrub_pages= [XEN]
6827 Boolean option to control scrubbing pages before giving them back
6828 to Xen, for use by other domains. Can be also changed at runtime
6829 with /sys/devices/system/xen_memory/xen_memory0/scrub_pages.
6830 Default value controlled with CONFIG_XEN_SCRUB_PAGES_DEFAULT.
6832 xen_timer_slop= [X86-64,XEN]
6833 Set the timer slop (in nanoseconds) for the virtual Xen
6834 timers (default is 100000). This adjusts the minimum
6835 delta of virtualized Xen timers, where lower values
6836 improve timer resolution at the expense of processing
6837 more timer interrupts.
6839 xen.balloon_boot_timeout= [XEN]
6840 The time (in seconds) to wait before giving up to boot
6841 in case initial ballooning fails to free enough memory.
6842 Applies only when running as HVM or PVH guest and
6843 started with less memory configured than allowed at
6844 max. Default is 180.
6846 xen.event_eoi_delay= [XEN]
6847 How long to delay EOI handling in case of event
6848 storms (jiffies). Default is 10.
6850 xen.event_loop_timeout= [XEN]
6851 After which time (jiffies) the event handling loop
6852 should start to delay EOI handling. Default is 2.
6854 xen.fifo_events= [XEN]
6855 Boolean parameter to disable using fifo event handling
6856 even if available. Normally fifo event handling is
6857 preferred over the 2-level event handling, as it is
6858 fairer and the number of possible event channels is
6859 much higher. Default is on (use fifo events).
6861 nopv= [X86,XEN,KVM,HYPER_V,VMWARE]
6862 Disables the PV optimizations forcing the guest to run
6863 as generic guest with no PV drivers. Currently support
6864 XEN HVM, KVM, HYPER_V and VMWARE guest.
6866 nopvspin [X86,XEN,KVM]
6867 Disables the qspinlock slow path using PV optimizations
6868 which allow the hypervisor to 'idle' the guest on lock
6871 xirc2ps_cs= [NET,PCMCIA]
6873 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]]
6876 By default on POWER9 and above, the kernel will
6877 natively use the XIVE interrupt controller. This option
6878 allows the fallback firmware mode to be used:
6880 off Fallback to firmware control of XIVE interrupt
6881 controller on both pseries and powernv
6882 platforms. Only useful on POWER9 and above.
6884 xive.store-eoi=off [PPC]
6885 By default on POWER10 and above, the kernel will use
6886 stores for EOI handling when the XIVE interrupt mode
6887 is active. This option allows the XIVE driver to use
6888 loads instead, as on POWER9.
6890 xhci-hcd.quirks [USB,KNL]
6891 A hex value specifying bitmask with supplemental xhci
6892 host controller quirks. Meaning of each bit can be
6893 consulted in header drivers/usb/host/xhci.h.
6896 Format: { early | on | rw | ro | off }
6897 Controls if xmon debugger is enabled. Default is off.
6898 Passing only "xmon" is equivalent to "xmon=early".
6899 early Call xmon as early as possible on boot; xmon
6900 debugger is called from setup_arch().
6901 on xmon debugger hooks will be installed so xmon
6902 is only called on a kernel crash. Default mode,
6903 i.e. either "ro" or "rw" mode, is controlled
6904 with CONFIG_XMON_DEFAULT_RO_MODE.
6905 rw xmon debugger hooks will be installed so xmon
6906 is called only on a kernel crash, mode is write,
6907 meaning SPR registers, memory and, other data
6908 can be written using xmon commands.
6909 ro same as "rw" option above but SPR registers,
6910 memory, and other data can't be written using
6912 off xmon is disabled.