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[=<bool>] [KNL]
1162 If no <bool> value is specified or if the value
1163 specified is not a valid <bool>, enable asynchronous
1164 probe on this module. Otherwise, enable/disable
1165 asynchronous probe on this module as indicated by the
1166 <bool> value. See also: module.async_probe
1168 early_ioremap_debug [KNL]
1169 Enable debug messages in early_ioremap support. This
1170 is useful for tracking down temporary early mappings
1171 which are not unmapped.
1173 earlycon= [KNL] Output early console device and options.
1175 When used with no options, the early console is
1176 determined by stdout-path property in device tree's
1177 chosen node or the ACPI SPCR table if supported by
1180 cdns,<addr>[,options]
1181 Start an early, polled-mode console on a Cadence
1182 (xuartps) serial port at the specified address. Only
1183 supported option is baud rate. If baud rate is not
1184 specified, the serial port must already be setup and
1187 uart[8250],io,<addr>[,options]
1188 uart[8250],mmio,<addr>[,options]
1189 uart[8250],mmio32,<addr>[,options]
1190 uart[8250],mmio32be,<addr>[,options]
1191 uart[8250],0x<addr>[,options]
1192 Start an early, polled-mode console on the 8250/16550
1193 UART at the specified I/O port or MMIO address.
1194 MMIO inter-register address stride is either 8-bit
1195 (mmio) or 32-bit (mmio32 or mmio32be).
1196 If none of [io|mmio|mmio32|mmio32be], <addr> is assumed
1197 to be equivalent to 'mmio'. 'options' are specified
1198 in the same format described for "console=ttyS<n>"; if
1199 unspecified, the h/w is not initialized.
1203 Start an early, polled-mode console on a pl011 serial
1204 port at the specified address. The pl011 serial port
1205 must already be setup and configured. Options are not
1206 yet supported. If 'mmio32' is specified, then only
1207 the driver will use only 32-bit accessors to read/write
1208 the device registers.
1211 Start an early console on a litex serial port at the
1212 specified address. The serial port must already be
1213 setup and configured. Options are not yet supported.
1216 Start an early, polled-mode console on a meson serial
1217 port at the specified address. The serial port must
1218 already be setup and configured. Options are not yet
1222 Start an early, polled-mode console on an msm serial
1223 port at the specified address. The serial port
1224 must already be setup and configured. Options are not
1227 msm_serial_dm,<addr>
1228 Start an early, polled-mode console on an msm serial
1229 dm port at the specified address. The serial port
1230 must already be setup and configured. Options are not
1234 Start an early, polled-mode console on a serial port
1235 of an Actions Semi SoC, such as S500 or S900, at the
1236 specified address. The serial port must already be
1237 setup and configured. Options are not yet supported.
1240 Start an early, polled-mode console on a serial port
1241 of an RDA Micro SoC, such as RDA8810PL, at the
1242 specified address. The serial port must already be
1243 setup and configured. Options are not yet supported.
1246 Use RISC-V SBI (Supervisor Binary Interface) for early
1249 smh Use ARM semihosting calls for early console.
1257 Use early console provided by serial driver available
1258 on Samsung SoCs, requires selecting proper type and
1259 a correct base address of the selected UART port. The
1260 serial port must already be setup and configured.
1261 Options are not yet supported.
1264 Start an early, polled-mode console on a lantiq serial
1265 (lqasc) port at the specified address. The serial port
1266 must already be setup and configured. Options are not
1271 Use early console provided by Freescale LP UART driver
1272 found on Freescale Vybrid and QorIQ LS1021A processors.
1273 A valid base address must be provided, and the serial
1274 port must already be setup and configured.
1278 Start an early, polled-mode, output-only console on the
1279 Freescale i.MX UART at the specified address. The UART
1280 must already be setup and configured.
1283 Start an early, polled-mode console on the
1284 Armada 3700 serial port at the specified
1285 address. The serial port must already be setup
1286 and configured. Options are not yet supported.
1289 Start an early, polled-mode console on a Qualcomm
1290 Generic Interface (GENI) based serial port at the
1291 specified address. The serial port must already be
1292 setup and configured. Options are not yet supported.
1295 Start an early, unaccelerated console on the EFI
1296 memory mapped framebuffer (if available). On cache
1297 coherent non-x86 systems that use system memory for
1298 the framebuffer, pass the 'ram' option so that it is
1299 mapped with the correct attributes.
1302 Use early console provided by Freescale LINFlexD UART
1303 serial driver for NXP S32V234 SoCs. A valid base
1304 address must be provided, and the serial port must
1305 already be setup and configured.
1307 earlyprintk= [X86,SH,ARM,M68k,S390]
1311 earlyprintk=serial[,ttySn[,baudrate]]
1312 earlyprintk=serial[,0x...[,baudrate]]
1313 earlyprintk=ttySn[,baudrate]
1314 earlyprintk=dbgp[debugController#]
1315 earlyprintk=pciserial[,force],bus:device.function[,baudrate]
1316 earlyprintk=xdbc[xhciController#]
1318 earlyprintk is useful when the kernel crashes before
1319 the normal console is initialized. It is not enabled by
1320 default because it has some cosmetic problems.
1322 Append ",keep" to not disable it when the real console
1325 Only one of vga, serial, or usb debug port can
1328 Currently only ttyS0 and ttyS1 may be specified by
1329 name. Other I/O ports may be explicitly specified
1330 on some architectures (x86 and arm at least) by
1331 replacing ttySn with an I/O port address, like this:
1332 earlyprintk=serial,0x1008,115200
1333 You can find the port for a given device in
1334 /proc/tty/driver/serial:
1335 2: uart:ST16650V2 port:00001008 irq:18 ...
1337 Interaction with the standard serial driver is not
1340 The VGA output is eventually overwritten by
1343 The xen option can only be used in Xen domains.
1345 The sclp output can only be used on s390.
1347 The optional "force" to "pciserial" enables use of a
1348 PCI device even when its classcode is not of the
1351 edac_report= [HW,EDAC] Control how to report EDAC event
1352 Format: {"on" | "off" | "force"}
1353 on: enable EDAC to report H/W event. May be overridden
1354 by other higher priority error reporting module.
1355 off: disable H/W event reporting through EDAC.
1356 force: enforce the use of EDAC to report H/W event.
1360 Format: {"off" | "on" | "skip[mbr]"}
1363 Format: { "debug", "disable_early_pci_dma",
1364 "nochunk", "noruntime", "nosoftreserve",
1365 "novamap", "no_disable_early_pci_dma" }
1366 debug: enable misc debug output.
1367 disable_early_pci_dma: disable the busmaster bit on all
1368 PCI bridges while in the EFI boot stub.
1369 nochunk: disable reading files in "chunks" in the EFI
1370 boot stub, as chunking can cause problems with some
1371 firmware implementations.
1372 noruntime : disable EFI runtime services support
1373 nosoftreserve: The EFI_MEMORY_SP (Specific Purpose)
1374 attribute may cause the kernel to reserve the
1375 memory range for a memory mapping driver to
1376 claim. Specify efi=nosoftreserve to disable this
1377 reservation and treat the memory by its base type
1378 (i.e. EFI_CONVENTIONAL_MEMORY / "System RAM").
1379 novamap: do not call SetVirtualAddressMap().
1380 no_disable_early_pci_dma: Leave the busmaster bit set
1381 on all PCI bridges while in the EFI boot stub
1383 efi_no_storage_paranoia [EFI; X86]
1384 Using this parameter you can use more than 50% of
1385 your efi variable storage. Use this parameter only if
1386 you are really sure that your UEFI does sane gc and
1387 fulfills the spec otherwise your board may brick.
1389 efi_fake_mem= nn[KMG]@ss[KMG]:aa[,nn[KMG]@ss[KMG]:aa,..] [EFI; X86]
1390 Add arbitrary attribute to specific memory range by
1391 updating original EFI memory map.
1392 Region of memory which aa attribute is added to is
1395 If efi_fake_mem=2G@4G:0x10000,2G@0x10a0000000:0x10000
1396 is specified, EFI_MEMORY_MORE_RELIABLE(0x10000)
1397 attribute is added to range 0x100000000-0x180000000 and
1398 0x10a0000000-0x1120000000.
1400 If efi_fake_mem=8G@9G:0x40000 is specified, the
1401 EFI_MEMORY_SP(0x40000) attribute is added to
1402 range 0x240000000-0x43fffffff.
1404 Using this parameter you can do debugging of EFI memmap
1405 related features. For example, you can do debugging of
1406 Address Range Mirroring feature even if your box
1407 doesn't support it, or mark specific memory as
1410 efivar_ssdt= [EFI; X86] Name of an EFI variable that contains an SSDT
1411 that is to be dynamically loaded by Linux. If there are
1412 multiple variables with the same name but with different
1413 vendor GUIDs, all of them will be loaded. See
1414 Documentation/admin-guide/acpi/ssdt-overlays.rst for details.
1417 eisa_irq_edge= [PARISC,HW]
1418 See header of drivers/parisc/eisa.c.
1420 ekgdboc= [X86,KGDB] Allow early kernel console debugging
1423 This is designed to be used in conjunction with
1424 the boot argument: earlyprintk=vga
1426 This parameter works in place of the kgdboc parameter
1427 but can only be used if the backing tty is available
1428 very early in the boot process. For early debugging
1429 via a serial port see kgdboc_earlycon instead.
1432 See comment before function elanfreq_setup() in
1433 arch/x86/kernel/cpu/cpufreq/elanfreq.c.
1435 elfcorehdr=[size[KMG]@]offset[KMG] [IA64,PPC,SH,X86,S390]
1436 Specifies physical address of start of kernel core
1437 image elf header and optionally the size. Generally
1438 kexec loader will pass this option to capture kernel.
1439 See Documentation/admin-guide/kdump/kdump.rst for details.
1441 enable_mtrr_cleanup [X86]
1442 The kernel tries to adjust MTRR layout from continuous
1443 to discrete, to make X server driver able to add WB
1444 entry later. This parameter enables that.
1446 enable_timer_pin_1 [X86]
1447 Enable PIN 1 of APIC timer
1448 Can be useful to work around chipset bugs
1449 (in particular on some ATI chipsets).
1450 The kernel tries to set a reasonable default.
1452 enforcing= [SELINUX] Set initial enforcing status.
1454 See security/selinux/Kconfig help text.
1455 0 -- permissive (log only, no denials).
1456 1 -- enforcing (deny and log).
1458 Value can be changed at runtime via
1459 /sys/fs/selinux/enforce.
1462 Disable Error Record Serialization Table (ERST)
1465 ether= [HW,NET] Ethernet cards parameters
1466 This option is obsoleted by the "netdev=" option, which
1467 has equivalent usage. See its documentation for details.
1471 Permit 'security.evm' to be updated regardless of
1472 current integrity status.
1477 fail_make_request=[KNL]
1478 General fault injection mechanism.
1479 Format: <interval>,<probability>,<space>,<times>
1480 See also Documentation/fault-injection/.
1483 Format: { initns | none }
1484 See Documentation/admin-guide/sysctl/net.rst for
1485 fb_tunnels_only_for_init_ns
1488 See Documentation/admin-guide/blockdev/floppy.rst.
1490 force_pal_cache_flush
1491 [IA-64] Avoid check_sal_cache_flush which may hang on
1492 buggy SAL_CACHE_FLUSH implementations. Using this
1493 parameter will force ia64_sal_cache_flush to call
1494 ia64_pal_cache_flush instead of SAL_CACHE_FLUSH.
1497 Forcefully enable Physical Address Extension (PAE).
1498 Many Pentium M systems disable PAE but may have a
1499 functionally usable PAE implementation.
1500 Warning: use of this parameter will taint the kernel
1501 and may cause unknown problems.
1504 [FTRACE] will set and start the specified tracer
1505 as early as possible in order to facilitate early
1508 ftrace_boot_snapshot
1509 [FTRACE] On boot up, a snapshot will be taken of the
1510 ftrace ring buffer that can be read at:
1511 /sys/kernel/tracing/snapshot.
1512 This is useful if you need tracing information from kernel
1513 boot up that is likely to be overridden by user space
1514 start up functionality.
1516 ftrace_dump_on_oops[=orig_cpu]
1517 [FTRACE] will dump the trace buffers on oops.
1518 If no parameter is passed, ftrace will dump
1519 buffers of all CPUs, but if you pass orig_cpu, it will
1520 dump only the buffer of the CPU that triggered the
1523 ftrace_filter=[function-list]
1524 [FTRACE] Limit the functions traced by the function
1525 tracer at boot up. function-list is a comma-separated
1526 list of functions. This list can be changed at run
1527 time by the set_ftrace_filter file in the debugfs
1530 ftrace_notrace=[function-list]
1531 [FTRACE] Do not trace the functions specified in
1532 function-list. This list can be changed at run time
1533 by the set_ftrace_notrace file in the debugfs
1536 ftrace_graph_filter=[function-list]
1537 [FTRACE] Limit the top level callers functions traced
1538 by the function graph tracer at boot up.
1539 function-list is a comma-separated list of functions
1540 that can be changed at run time by the
1541 set_graph_function file in the debugfs tracing directory.
1543 ftrace_graph_notrace=[function-list]
1544 [FTRACE] Do not trace from the functions specified in
1545 function-list. This list is a comma-separated list of
1546 functions that can be changed at run time by the
1547 set_graph_notrace file in the debugfs tracing directory.
1549 ftrace_graph_max_depth=<uint>
1550 [FTRACE] Used with the function graph tracer. This is
1551 the max depth it will trace into a function. This value
1552 can be changed at run time by the max_graph_depth file
1553 in the tracefs tracing directory. default: 0 (no limit)
1555 fw_devlink= [KNL] Create device links between consumer and supplier
1556 devices by scanning the firmware to infer the
1557 consumer/supplier relationships. This feature is
1558 especially useful when drivers are loaded as modules as
1559 it ensures proper ordering of tasks like device probing
1560 (suppliers first, then consumers), supplier boot state
1561 clean up (only after all consumers have probed),
1562 suspend/resume & runtime PM (consumers first, then
1564 Format: { off | permissive | on | rpm }
1565 off -- Don't create device links from firmware info.
1566 permissive -- Create device links from firmware info
1567 but use it only for ordering boot state clean
1568 up (sync_state() calls).
1569 on -- Create device links from firmware info and use it
1570 to enforce probe and suspend/resume ordering.
1571 rpm -- Like "on", but also use to order runtime PM.
1573 fw_devlink.strict=<bool>
1574 [KNL] Treat all inferred dependencies as mandatory
1575 dependencies. This only applies for fw_devlink=on|rpm.
1579 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad
1580 support via parallel port (up to 5 devices per port)
1581 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5>
1582 See also Documentation/input/devices/joystick-parport.rst
1586 gart_fix_e820= [X86-64] disable the fix e820 for K8 GART
1590 gcov_persist= [GCOV] When non-zero (default), profiling data for
1591 kernel modules is saved and remains accessible via
1592 debugfs, even when the module is unloaded/reloaded.
1593 When zero, profiling data is discarded and associated
1594 debugfs files are removed at module unload time.
1596 goldfish [X86] Enable the goldfish android emulator platform.
1597 Don't use this when you are not running on the
1600 gpio-mockup.gpio_mockup_ranges
1601 [HW] Sets the ranges of gpiochip of for this device.
1602 Format: <start1>,<end1>,<start2>,<end2>...
1603 gpio-mockup.gpio_mockup_named_lines
1604 [HW] Let the driver know GPIO lines should be named.
1606 gpt [EFI] Forces disk with valid GPT signature but
1607 invalid Protective MBR to be treated as GPT. If the
1608 primary GPT is corrupted, it enables the backup/alternate
1609 GPT to be used instead.
1611 grcan.enable0= [HW] Configuration of physical interface 0. Determines
1612 the "Enable 0" bit of the configuration register.
1615 grcan.enable1= [HW] Configuration of physical interface 1. Determines
1616 the "Enable 0" bit of the configuration register.
1619 grcan.select= [HW] Select which physical interface to use.
1622 grcan.txsize= [HW] Sets the size of the tx buffer.
1623 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0.
1625 grcan.rxsize= [HW] Sets the size of the rx buffer.
1626 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0.
1630 [KNL] Under CONFIG_HARDENED_USERCOPY, whether
1631 hardening is enabled for this boot. Hardened
1632 usercopy checking is used to protect the kernel
1633 from reading or writing beyond known memory
1634 allocation boundaries as a proactive defense
1635 against bounds-checking flaws in the kernel's
1636 copy_to_user()/copy_from_user() interface.
1637 on Perform hardened usercopy checks (default).
1638 off Disable hardened usercopy checks.
1640 hardlockup_all_cpu_backtrace=
1641 [KNL] Should the hard-lockup detector generate
1642 backtraces on all cpus.
1645 hashdist= [KNL,NUMA] Large hashes allocated during boot
1646 are distributed across NUMA nodes. Defaults on
1647 for 64-bit NUMA, off otherwise.
1648 Format: 0 | 1 (for off | on)
1650 hcl= [IA-64] SGI's Hardware Graph compatibility layer
1652 hd= [EIDE] (E)IDE hard drive subsystem geometry
1653 Format: <cyl>,<head>,<sect>
1656 Disable Hardware Error Source Table (HEST) support;
1657 corresponding firmware-first mode error processing
1658 logic will be disabled.
1660 hibernate= [HIBERNATION]
1661 noresume Don't check if there's a hibernation image
1662 present during boot.
1663 nocompress Don't compress/decompress hibernation images.
1664 no Disable hibernation and resume.
1665 protect_image Turn on image protection during restoration
1666 (that will set all pages holding image data
1667 during restoration read-only).
1669 highmem=nn[KMG] [KNL,BOOT] forces the highmem zone to have an exact
1670 size of <nn>. This works even on boxes that have no
1671 highmem otherwise. This also works to reduce highmem
1672 size on bigger boxes.
1674 highres= [KNL] Enable/disable high resolution timer mode.
1675 Valid parameters: "on", "off"
1680 hostname= [KNL] Set the hostname (aka UTS nodename).
1682 This allows setting the system's hostname during early
1683 startup. This sets the name returned by gethostname.
1684 Using this parameter to set the hostname makes it
1685 possible to ensure the hostname is correctly set before
1686 any userspace processes run, avoiding the possibility
1687 that a process may call gethostname before the hostname
1688 has been explicitly set, resulting in the calling
1689 process getting an incorrect result. The string must
1690 not exceed the maximum allowed hostname length (usually
1691 64 characters) and will be truncated otherwise.
1693 hpet= [X86-32,HPET] option to control HPET usage
1694 Format: { enable (default) | disable | force |
1696 disable: disable HPET and use PIT instead
1697 force: allow force enabled of undocumented chips (ICH4,
1699 verbose: show contents of HPET registers during setup
1701 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET
1702 registers. Default set by CONFIG_HPET_MMAP_DEFAULT.
1704 hugepages= [HW] Number of HugeTLB pages to allocate at boot.
1705 If this follows hugepagesz (below), it specifies
1706 the number of pages of hugepagesz to be allocated.
1707 If this is the first HugeTLB parameter on the command
1708 line, it specifies the number of pages to allocate for
1709 the default huge page size. If using node format, the
1710 number of pages to allocate per-node can be specified.
1711 See also Documentation/admin-guide/mm/hugetlbpage.rst.
1712 Format: <integer> or (node format)
1713 <node>:<integer>[,<node>:<integer>]
1716 [HW] The size of the HugeTLB pages. This is used in
1717 conjunction with hugepages (above) to allocate huge
1718 pages of a specific size at boot. The pair
1719 hugepagesz=X hugepages=Y can be specified once for
1720 each supported huge page size. Huge page sizes are
1721 architecture dependent. See also
1722 Documentation/admin-guide/mm/hugetlbpage.rst.
1725 hugetlb_cma= [HW,CMA] The size of a CMA area used for allocation
1726 of gigantic hugepages. Or using node format, the size
1727 of a CMA area per node can be specified.
1728 Format: nn[KMGTPE] or (node format)
1729 <node>:nn[KMGTPE][,<node>:nn[KMGTPE]]
1731 Reserve a CMA area of given size and allocate gigantic
1732 hugepages using the CMA allocator. If enabled, the
1733 boot-time allocation of gigantic hugepages is skipped.
1735 hugetlb_free_vmemmap=
1736 [KNL] Reguires CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP
1738 Allows heavy hugetlb users to free up some more
1739 memory (7 * PAGE_SIZE for each 2MB hugetlb page).
1740 Format: { [oO][Nn]/Y/y/1 | [oO][Ff]/N/n/0 (default) }
1742 [oO][Nn]/Y/y/1: enable the feature
1743 [oO][Ff]/N/n/0: disable the feature
1745 Built with CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP_DEFAULT_ON=y,
1748 Note that the vmemmap pages may be allocated from the added
1749 memory block itself when memory_hotplug.memmap_on_memory is
1750 enabled, those vmemmap pages cannot be optimized even if this
1751 feature is enabled. Other vmemmap pages not allocated from
1752 the added memory block itself do not be affected.
1755 [KNL] Should the hung task detector generate panics.
1758 A value of 1 instructs the kernel to panic when a
1759 hung task is detected. The default value is controlled
1760 by the CONFIG_BOOTPARAM_HUNG_TASK_PANIC build-time
1761 option. The value selected by this boot parameter can
1762 be changed later by the kernel.hung_task_panic sysctl.
1764 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC)
1765 terminal devices. Valid values: 0..8
1766 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs.
1767 If specified, z/VM IUCV HVC accepts connections
1768 from listed z/VM user IDs only.
1770 hv_nopvspin [X86,HYPER_V] Disables the paravirt spinlock optimizations
1771 which allow the hypervisor to 'idle' the
1772 guest on lock contention.
1775 Do not unregister boot console at start. This is only
1776 useful for debugging when something happens in the window
1777 between unregistering the boot console and initializing
1780 i2c_bus= [HW] Override the default board specific I2C bus speed
1781 or register an additional I2C bus that is not
1782 registered from board initialization code.
1786 i8042.debug [HW] Toggle i8042 debug mode
1787 i8042.unmask_kbd_data
1788 [HW] Enable printing of interrupt data from the KBD port
1789 (disabled by default, and as a pre-condition
1790 requires that i8042.debug=1 be enabled)
1791 i8042.direct [HW] Put keyboard port into non-translated mode
1792 i8042.dumbkbd [HW] Pretend that controller can only read data from
1793 keyboard and cannot control its state
1794 (Don't attempt to blink the leds)
1795 i8042.noaux [HW] Don't check for auxiliary (== mouse) port
1796 i8042.nokbd [HW] Don't check/create keyboard port
1797 i8042.noloop [HW] Disable the AUX Loopback command while probing
1799 i8042.nomux [HW] Don't check presence of an active multiplexing
1801 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX
1803 i8042.notimeout [HW] Ignore timeout condition signalled by controller
1804 i8042.reset [HW] Reset the controller during init, cleanup and
1805 suspend-to-ram transitions, only during s2r
1806 transitions, or never reset
1807 Format: { 1 | Y | y | 0 | N | n }
1808 1, Y, y: always reset controller
1809 0, N, n: don't ever reset controller
1810 Default: only on s2r transitions on x86; most other
1811 architectures force reset to be always executed
1812 i8042.unlock [HW] Unlock (ignore) the keylock
1813 i8042.kbdreset [HW] Reset device connected to KBD port
1815 [HW] Allow deferred probing upon i8042 probe errors
1819 i915.invert_brightness=
1820 [DRM] Invert the sense of the variable that is used to
1821 set the brightness of the panel backlight. Normally a
1822 brightness value of 0 indicates backlight switched off,
1823 and the maximum of the brightness value sets the backlight
1824 to maximum brightness. If this parameter is set to 0
1825 (default) and the machine requires it, or this parameter
1826 is set to 1, a brightness value of 0 sets the backlight
1827 to maximum brightness, and the maximum of the brightness
1828 value switches the backlight off.
1829 -1 -- never invert brightness
1830 0 -- machine default
1831 1 -- force brightness inversion
1834 Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]]
1838 Format: idle=poll, idle=halt, idle=nomwait
1839 Poll forces a polling idle loop that can slightly
1840 improve the performance of waking up a idle CPU, but
1841 will use a lot of power and make the system run hot.
1843 idle=halt: Halt is forced to be used for CPU idle.
1844 In such case C2/C3 won't be used again.
1845 idle=nomwait: Disable mwait for CPU C-states
1849 Allow force disabling of Shared Virtual Memory (SVA)
1850 support for the idxd driver. By default it is set to
1853 idxd.tc_override= [HW]
1855 Allow override of default traffic class configuration
1856 for the device. By default it is set to false (0).
1858 ieee754= [MIPS] Select IEEE Std 754 conformance mode
1859 Format: { strict | legacy | 2008 | relaxed }
1862 Choose which programs will be accepted for execution
1863 based on the IEEE 754 NaN encoding(s) supported by
1864 the FPU and the NaN encoding requested with the value
1865 of an ELF file header flag individually set by each
1866 binary. Hardware implementations are permitted to
1867 support either or both of the legacy and the 2008 NaN
1870 Available settings are as follows:
1871 strict accept binaries that request a NaN encoding
1872 supported by the FPU
1873 legacy only accept legacy-NaN binaries, if supported
1875 2008 only accept 2008-NaN binaries, if supported
1877 relaxed accept any binaries regardless of whether
1878 supported by the FPU
1880 The FPU emulator is always able to support both NaN
1881 encodings, so if no FPU hardware is present or it has
1882 been disabled with 'nofpu', then the settings of
1883 'legacy' and '2008' strap the emulator accordingly,
1884 'relaxed' straps the emulator for both legacy-NaN and
1885 2008-NaN, whereas 'strict' enables legacy-NaN only on
1886 legacy processors and both NaN encodings on MIPS32 or
1889 The setting for ABS.fmt/NEG.fmt instruction execution
1890 mode generally follows that for the NaN encoding,
1891 except where unsupported by hardware.
1893 ignore_loglevel [KNL]
1894 Ignore loglevel setting - this will print /all/
1895 kernel messages to the console. Useful for debugging.
1896 We also add it as printk module parameter, so users
1897 could change it dynamically, usually by
1898 /sys/module/printk/parameters/ignore_loglevel.
1901 Ignore RLIMIT_DATA setting for data mappings,
1902 print warning at first misuse. Can be changed via
1903 /sys/module/kernel/parameters/ignore_rlimit_data.
1905 ihash_entries= [KNL]
1906 Set number of hash buckets for inode cache.
1908 ima_appraise= [IMA] appraise integrity measurements
1909 Format: { "off" | "enforce" | "fix" | "log" }
1912 ima_appraise_tcb [IMA] Deprecated. Use ima_policy= instead.
1913 The builtin appraise policy appraises all files
1916 ima_canonical_fmt [IMA]
1917 Use the canonical format for the binary runtime
1918 measurements, instead of host native format.
1921 Format: { md5 | sha1 | rmd160 | sha256 | sha384
1925 The list of supported hash algorithms is defined
1926 in crypto/hash_info.h.
1929 The builtin policies to load during IMA setup.
1930 Format: "tcb | appraise_tcb | secure_boot |
1931 fail_securely | critical_data"
1933 The "tcb" policy measures all programs exec'd, files
1934 mmap'd for exec, and all files opened with the read
1935 mode bit set by either the effective uid (euid=0) or
1938 The "appraise_tcb" policy appraises the integrity of
1939 all files owned by root.
1941 The "secure_boot" policy appraises the integrity
1942 of files (eg. kexec kernel image, kernel modules,
1943 firmware, policy, etc) based on file signatures.
1945 The "fail_securely" policy forces file signature
1946 verification failure also on privileged mounted
1947 filesystems with the SB_I_UNVERIFIABLE_SIGNATURE
1950 The "critical_data" policy measures kernel integrity
1953 ima_tcb [IMA] Deprecated. Use ima_policy= instead.
1954 Load a policy which meets the needs of the Trusted
1955 Computing Base. This means IMA will measure all
1956 programs exec'd, files mmap'd for exec, and all files
1957 opened for read by uid=0.
1960 Select one of defined IMA measurements template formats.
1961 Formats: { "ima" | "ima-ng" | "ima-ngv2" | "ima-sig" |
1966 [IMA] Define a custom template format.
1967 Format: { "field1|...|fieldN" }
1969 ima.ahash_minsize= [IMA] Minimum file size for asynchronous hash usage
1970 Format: <min_file_size>
1971 Set the minimal file size for using asynchronous hash.
1972 If left unspecified, ahash usage is disabled.
1974 ahash performance varies for different data sizes on
1975 different crypto accelerators. This option can be used
1976 to achieve the best performance for a particular HW.
1978 ima.ahash_bufsize= [IMA] Asynchronous hash buffer size
1980 Set hashing buffer size. Default: 4k.
1982 ahash performance varies for different chunk sizes on
1983 different crypto accelerators. This option can be used
1984 to achieve best performance for particular HW.
1988 Run specified binary instead of /sbin/init as init
1991 initcall_debug [KNL] Trace initcalls as they are executed. Useful
1992 for working out where the kernel is dying during
1995 initcall_blacklist= [KNL] Do not execute a comma-separated list of
1996 initcall functions. Useful for debugging built-in
1997 modules and initcalls.
1999 initramfs_async= [KNL]
2002 This parameter controls whether the initramfs
2003 image is unpacked asynchronously, concurrently
2004 with devices being probed and
2005 initialized. This should normally just work,
2006 but as a debugging aid, one can get the
2007 historical behaviour of the initramfs
2008 unpacking being completed before device_ and
2011 initrd= [BOOT] Specify the location of the initial ramdisk
2013 initrdmem= [KNL] Specify a physical address and size from which to
2014 load the initrd. If an initrd is compiled in or
2015 specified in the bootparams, it takes priority over this
2017 Format: ss[KMG],nn[KMG]
2020 init_on_alloc= [MM] Fill newly allocated pages and heap objects with
2023 Default set by CONFIG_INIT_ON_ALLOC_DEFAULT_ON.
2025 init_on_free= [MM] Fill freed pages and heap objects with zeroes.
2027 Default set by CONFIG_INIT_ON_FREE_DEFAULT_ON.
2029 init_pkru= [X86] Specify the default memory protection keys rights
2030 register contents for all processes. 0x55555554 by
2031 default (disallow access to all but pkey 0). Can
2032 override in debugfs after boot.
2034 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver
2037 int_pln_enable [X86] Enable power limit notification interrupt
2039 integrity_audit=[IMA]
2040 Format: { "0" | "1" }
2041 0 -- basic integrity auditing messages. (Default)
2042 1 -- additional integrity auditing messages.
2044 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option
2046 Enable intel iommu driver.
2048 Disable intel iommu driver.
2049 igfx_off [Default Off]
2050 By default, gfx is mapped as normal device. If a gfx
2051 device has a dedicated DMAR unit, the DMAR unit is
2052 bypassed by not enabling DMAR with this option. In
2053 this case, gfx device will use physical address for
2055 strict [Default Off]
2056 Deprecated, equivalent to iommu.strict=1.
2057 sp_off [Default Off]
2058 By default, super page will be supported if Intel IOMMU
2059 has the capability. With this option, super page will
2062 Enable the Intel IOMMU scalable mode if the hardware
2063 advertises that it has support for the scalable mode
2066 Disallow use of the Intel IOMMU scalable mode.
2067 tboot_noforce [Default Off]
2068 Do not force the Intel IOMMU enabled under tboot.
2069 By default, tboot will force Intel IOMMU on, which
2070 could harm performance of some high-throughput
2071 devices like 40GBit network cards, even if identity
2073 Note that using this option lowers the security
2074 provided by tboot because it makes the system
2075 vulnerable to DMA attacks.
2077 intel_idle.max_cstate= [KNL,HW,ACPI,X86]
2078 0 disables intel_idle and fall back on acpi_idle.
2079 1 to 9 specify maximum depth of C-state.
2083 Do not enable intel_pstate as the default
2084 scaling driver for the supported processors
2086 Use intel_pstate as a scaling driver, but configure it
2087 to work with generic cpufreq governors (instead of
2088 enabling its internal governor). This mode cannot be
2089 used along with the hardware-managed P-states (HWP)
2092 Enable intel_pstate on systems that prohibit it by default
2093 in favor of acpi-cpufreq. Forcing the intel_pstate driver
2094 instead of acpi-cpufreq may disable platform features, such
2095 as thermal controls and power capping, that rely on ACPI
2096 P-States information being indicated to OSPM and therefore
2097 should be used with caution. This option does not work with
2098 processors that aren't supported by the intel_pstate driver
2099 or on platforms that use pcc-cpufreq instead of acpi-cpufreq.
2101 Do not enable hardware P state control (HWP)
2104 Only load intel_pstate on systems which support
2105 hardware P state control (HWP) if available.
2107 Enforce ACPI _PPC performance limits. If the Fixed ACPI
2108 Description Table, specifies preferred power management
2109 profile as "Enterprise Server" or "Performance Server",
2110 then this feature is turned on by default.
2112 Allow per-logical-CPU P-State performance control limits using
2113 cpufreq sysfs interface
2115 intremap= [X86-64, Intel-IOMMU]
2116 on enable Interrupt Remapping (default)
2117 off disable Interrupt Remapping
2118 nosid disable Source ID checking
2120 BIOS x2APIC opt-out request will be ignored
2121 nopost disable Interrupt Posting
2123 iomem= Disable strict checking of access to MMIO memory
2124 strict regions from userspace.
2139 nobypass [PPC/POWERNV]
2140 Disable IOMMU bypass, using IOMMU for PCI devices.
2142 iommu.forcedac= [ARM64, X86] Control IOVA allocation for PCI devices.
2143 Format: { "0" | "1" }
2144 0 - Try to allocate a 32-bit DMA address first, before
2145 falling back to the full range if needed.
2146 1 - Allocate directly from the full usable range,
2147 forcing Dual Address Cycle for PCI cards supporting
2148 greater than 32-bit addressing.
2150 iommu.strict= [ARM64, X86] Configure TLB invalidation behaviour
2151 Format: { "0" | "1" }
2153 Request that DMA unmap operations use deferred
2154 invalidation of hardware TLBs, for increased
2155 throughput at the cost of reduced device isolation.
2156 Will fall back to strict mode if not supported by
2157 the relevant IOMMU driver.
2159 DMA unmap operations invalidate IOMMU hardware TLBs
2161 unset - Use value of CONFIG_IOMMU_DEFAULT_DMA_{LAZY,STRICT}.
2162 Note: on x86, strict mode specified via one of the
2163 legacy driver-specific options takes precedence.
2166 [ARM64, X86] Configure DMA to bypass the IOMMU by default.
2167 Format: { "0" | "1" }
2168 0 - Use IOMMU translation for DMA.
2169 1 - Bypass the IOMMU for DMA.
2170 unset - Use value of CONFIG_IOMMU_DEFAULT_PASSTHROUGH.
2172 io7= [HW] IO7 for Marvel-based Alpha systems
2173 See comment before marvel_specify_io7 in
2174 arch/alpha/kernel/core_marvel.c.
2176 io_delay= [X86] I/O delay method
2178 Standard port 0x80 based delay
2180 Alternate port 0xed based delay (needed on some systems)
2182 Simple two microseconds delay
2187 See Documentation/admin-guide/nfs/nfsroot.rst.
2189 ipcmni_extend [KNL] Extend the maximum number of unique System V
2190 IPC identifiers from 32,768 to 16,777,216.
2192 irqaffinity= [SMP] Set the default irq affinity mask
2193 The argument is a cpu list, as described above.
2195 irqchip.gicv2_force_probe=
2198 Force the kernel to look for the second 4kB page
2199 of a GICv2 controller even if the memory range
2200 exposed by the device tree is too small.
2202 irqchip.gicv3_nolpi=
2204 Force the kernel to ignore the availability of
2205 LPIs (and by consequence ITSs). Intended for system
2206 that use the kernel as a bootloader, and thus want
2207 to let secondary kernels in charge of setting up
2210 irqchip.gicv3_pseudo_nmi= [ARM64]
2211 Enables support for pseudo-NMIs in the kernel. This
2212 requires the kernel to be built with
2213 CONFIG_ARM64_PSEUDO_NMI.
2216 When an interrupt is not handled search all handlers
2217 for it. Intended to get systems with badly broken
2221 When an interrupt is not handled search all handlers
2222 for it. Also check all handlers each timer
2223 interrupt. Intended to get systems with badly broken
2227 Format: <RDP>,<reset>,<pci_scan>,<verbosity>
2229 isolcpus= [KNL,SMP,ISOL] Isolate a given set of CPUs from disturbance.
2230 [Deprecated - use cpusets instead]
2231 Format: [flag-list,]<cpu-list>
2233 Specify one or more CPUs to isolate from disturbances
2234 specified in the flag list (default: domain):
2237 Disable the tick when a single task runs.
2239 A residual 1Hz tick is offloaded to workqueues, which you
2240 need to affine to housekeeping through the global
2241 workqueue's affinity configured via the
2242 /sys/devices/virtual/workqueue/cpumask sysfs file, or
2243 by using the 'domain' flag described below.
2245 NOTE: by default the global workqueue runs on all CPUs,
2246 so to protect individual CPUs the 'cpumask' file has to
2247 be configured manually after bootup.
2250 Isolate from the general SMP balancing and scheduling
2251 algorithms. Note that performing domain isolation this way
2252 is irreversible: it's not possible to bring back a CPU to
2253 the domains once isolated through isolcpus. It's strongly
2254 advised to use cpusets instead to disable scheduler load
2255 balancing through the "cpuset.sched_load_balance" file.
2256 It offers a much more flexible interface where CPUs can
2257 move in and out of an isolated set anytime.
2259 You can move a process onto or off an "isolated" CPU via
2260 the CPU affinity syscalls or cpuset.
2261 <cpu number> begins at 0 and the maximum value is
2262 "number of CPUs in system - 1".
2266 Isolate from being targeted by managed interrupts
2267 which have an interrupt mask containing isolated
2268 CPUs. The affinity of managed interrupts is
2269 handled by the kernel and cannot be changed via
2270 the /proc/irq/* interfaces.
2272 This isolation is best effort and only effective
2273 if the automatically assigned interrupt mask of a
2274 device queue contains isolated and housekeeping
2275 CPUs. If housekeeping CPUs are online then such
2276 interrupts are directed to the housekeeping CPU
2277 so that IO submitted on the housekeeping CPU
2278 cannot disturb the isolated CPU.
2280 If a queue's affinity mask contains only isolated
2281 CPUs then this parameter has no effect on the
2282 interrupt routing decision, though interrupts are
2283 only delivered when tasks running on those
2284 isolated CPUs submit IO. IO submitted on
2285 housekeeping CPUs has no influence on those
2288 The format of <cpu-list> is described above.
2292 ivrs_ioapic [HW,X86-64]
2293 Provide an override to the IOAPIC-ID<->DEVICE-ID
2294 mapping provided in the IVRS ACPI table.
2295 By default, PCI segment is 0, and can be omitted.
2297 * To map IOAPIC-ID decimal 10 to PCI device 00:14.0
2298 write the parameter as:
2299 ivrs_ioapic[10]=00:14.0
2300 * To map IOAPIC-ID decimal 10 to PCI segment 0x1 and
2301 PCI device 00:14.0 write the parameter as:
2302 ivrs_ioapic[10]=0001:00:14.0
2304 ivrs_hpet [HW,X86-64]
2305 Provide an override to the HPET-ID<->DEVICE-ID
2306 mapping provided in the IVRS ACPI table.
2307 By default, PCI segment is 0, and can be omitted.
2309 * To map HPET-ID decimal 0 to PCI device 00:14.0
2310 write the parameter as:
2311 ivrs_hpet[0]=00:14.0
2312 * To map HPET-ID decimal 10 to PCI segment 0x1 and
2313 PCI device 00:14.0 write the parameter as:
2314 ivrs_ioapic[10]=0001:00:14.0
2316 ivrs_acpihid [HW,X86-64]
2317 Provide an override to the ACPI-HID:UID<->DEVICE-ID
2318 mapping provided in the IVRS ACPI table.
2320 For example, to map UART-HID:UID AMD0020:0 to
2321 PCI segment 0x1 and PCI device ID 00:14.5,
2322 write the parameter as:
2323 ivrs_acpihid[0001:00:14.5]=AMD0020:0
2325 By default, PCI segment is 0, and can be omitted.
2326 For example, PCI device 00:14.5 write the parameter as:
2327 ivrs_acpihid[00:14.5]=AMD0020:0
2329 js= [HW,JOY] Analog joystick
2330 See Documentation/input/joydev/joystick.rst.
2333 When CONFIG_RANDOMIZE_BASE is set, this disables
2334 kernel and module base offset ASLR (Address Space
2335 Layout Randomization).
2338 [KNL] Enforce KASAN (Kernel Address Sanitizer) to print
2339 report on every invalid memory access. Without this
2340 parameter KASAN will print report only for the first
2345 kernelcore= [KNL,X86,IA-64,PPC]
2346 Format: nn[KMGTPE] | nn% | "mirror"
2347 This parameter specifies the amount of memory usable by
2348 the kernel for non-movable allocations. The requested
2349 amount is spread evenly throughout all nodes in the
2350 system as ZONE_NORMAL. The remaining memory is used for
2351 movable memory in its own zone, ZONE_MOVABLE. In the
2352 event, a node is too small to have both ZONE_NORMAL and
2353 ZONE_MOVABLE, kernelcore memory will take priority and
2354 other nodes will have a larger ZONE_MOVABLE.
2356 ZONE_MOVABLE is used for the allocation of pages that
2357 may be reclaimed or moved by the page migration
2358 subsystem. Note that allocations like PTEs-from-HighMem
2359 still use the HighMem zone if it exists, and the Normal
2360 zone if it does not.
2362 It is possible to specify the exact amount of memory in
2363 the form of "nn[KMGTPE]", a percentage of total system
2364 memory in the form of "nn%", or "mirror". If "mirror"
2365 option is specified, mirrored (reliable) memory is used
2366 for non-movable allocations and remaining memory is used
2367 for Movable pages. "nn[KMGTPE]", "nn%", and "mirror"
2368 are exclusive, so you cannot specify multiple forms.
2370 kgdbdbgp= [KGDB,HW] kgdb over EHCI usb debug port.
2371 Format: <Controller#>[,poll interval]
2372 The controller # is the number of the ehci usb debug
2373 port as it is probed via PCI. The poll interval is
2374 optional and is the number seconds in between
2375 each poll cycle to the debug port in case you need
2376 the functionality for interrupting the kernel with
2377 gdb or control-c on the dbgp connection. When
2378 not using this parameter you use sysrq-g to break into
2379 the kernel debugger.
2381 kgdboc= [KGDB,HW] kgdb over consoles.
2382 Requires a tty driver that supports console polling,
2383 or a supported polling keyboard driver (non-usb).
2384 Serial only format: <serial_device>[,baud]
2385 keyboard only format: kbd
2386 keyboard and serial format: kbd,<serial_device>[,baud]
2387 Optional Kernel mode setting:
2388 kms, kbd format: kms,kbd
2389 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud]
2391 kgdboc_earlycon= [KGDB,HW]
2392 If the boot console provides the ability to read
2393 characters and can work in polling mode, you can use
2394 this parameter to tell kgdb to use it as a backend
2395 until the normal console is registered. Intended to
2396 be used together with the kgdboc parameter which
2397 specifies the normal console to transition to.
2399 The name of the early console should be specified
2400 as the value of this parameter. Note that the name of
2401 the early console might be different than the tty
2402 name passed to kgdboc. It's OK to leave the value
2403 blank and the first boot console that implements
2404 read() will be picked.
2406 kgdbwait [KGDB] Stop kernel execution and enter the
2407 kernel debugger at the earliest opportunity.
2409 kmac= [MIPS] Korina ethernet MAC address.
2410 Configure the RouterBoard 532 series on-chip
2411 Ethernet adapter MAC address.
2413 kmemleak= [KNL] Boot-time kmemleak enable/disable
2414 Valid arguments: on, off
2416 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y,
2419 kprobe_event=[probe-list]
2420 [FTRACE] Add kprobe events and enable at boot time.
2421 The probe-list is a semicolon delimited list of probe
2422 definitions. Each definition is same as kprobe_events
2423 interface, but the parameters are comma delimited.
2424 For example, to add a kprobe event on vfs_read with
2425 arg1 and arg2, add to the command line;
2427 kprobe_event=p,vfs_read,$arg1,$arg2
2429 See also Documentation/trace/kprobetrace.rst "Kernel
2430 Boot Parameter" section.
2432 kpti= [ARM64] Control page table isolation of user
2433 and kernel address spaces.
2434 Default: enabled on cores which need mitigation.
2438 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs.
2439 Default is 0 (don't ignore, but inject #GP)
2441 kvm.eager_page_split=
2442 [KVM,X86] Controls whether or not KVM will try to
2443 proactively split all huge pages during dirty logging.
2444 Eager page splitting reduces interruptions to vCPU
2445 execution by eliminating the write-protection faults
2446 and MMU lock contention that would otherwise be
2447 required to split huge pages lazily.
2449 VM workloads that rarely perform writes or that write
2450 only to a small region of VM memory may benefit from
2451 disabling eager page splitting to allow huge pages to
2452 still be used for reads.
2454 The behavior of eager page splitting depends on whether
2455 KVM_DIRTY_LOG_INITIALLY_SET is enabled or disabled. If
2456 disabled, all huge pages in a memslot will be eagerly
2457 split when dirty logging is enabled on that memslot. If
2458 enabled, eager page splitting will be performed during
2459 the KVM_CLEAR_DIRTY ioctl, and only for the pages being
2462 Eager page splitting is only supported when kvm.tdp_mmu=Y.
2466 kvm.enable_vmware_backdoor=[KVM] Support VMware backdoor PV interface.
2467 Default is false (don't support).
2470 [KVM] Controls the software workaround for the
2471 X86_BUG_ITLB_MULTIHIT bug.
2472 force : Always deploy workaround.
2473 off : Never deploy workaround.
2474 auto : Deploy workaround based on the presence of
2475 X86_BUG_ITLB_MULTIHIT.
2479 If the software workaround is enabled for the host,
2480 guests do need not to enable it for nested guests.
2482 kvm.nx_huge_pages_recovery_ratio=
2483 [KVM] Controls how many 4KiB pages are periodically zapped
2484 back to huge pages. 0 disables the recovery, otherwise if
2485 the value is N KVM will zap 1/Nth of the 4KiB pages every
2486 period (see below). The default is 60.
2488 kvm.nx_huge_pages_recovery_period_ms=
2489 [KVM] Controls the time period at which KVM zaps 4KiB pages
2490 back to huge pages. If the value is a non-zero N, KVM will
2491 zap a portion (see ratio above) of the pages every N msecs.
2492 If the value is 0 (the default), KVM will pick a period based
2493 on the ratio, such that a page is zapped after 1 hour on average.
2495 kvm-amd.nested= [KVM,AMD] Allow nested virtualization in KVM/SVM.
2496 Default is 1 (enabled)
2498 kvm-amd.npt= [KVM,AMD] Disable nested paging (virtualized MMU)
2500 Default is 1 (enabled) if in 64-bit or 32-bit PAE mode.
2503 [KVM,ARM] Select one of KVM/arm64's modes of operation.
2505 none: Forcefully disable KVM.
2507 nvhe: Standard nVHE-based mode, without support for
2510 protected: nVHE-based mode with support for guests whose
2511 state is kept private from the host.
2513 Defaults to VHE/nVHE based on hardware support. Setting
2514 mode to "protected" will disable kexec and hibernation
2517 kvm-arm.vgic_v3_group0_trap=
2518 [KVM,ARM] Trap guest accesses to GICv3 group-0
2521 kvm-arm.vgic_v3_group1_trap=
2522 [KVM,ARM] Trap guest accesses to GICv3 group-1
2525 kvm-arm.vgic_v3_common_trap=
2526 [KVM,ARM] Trap guest accesses to GICv3 common
2529 kvm-arm.vgic_v4_enable=
2530 [KVM,ARM] Allow use of GICv4 for direct injection of
2533 kvm_cma_resv_ratio=n [PPC]
2534 Reserves given percentage from system memory area for
2535 contiguous memory allocation for KVM hash pagetable
2537 By default it reserves 5% of total system memory.
2541 kvm-intel.ept= [KVM,Intel] Disable extended page tables
2542 (virtualized MMU) support on capable Intel chips.
2543 Default is 1 (enabled)
2545 kvm-intel.emulate_invalid_guest_state=
2546 [KVM,Intel] Disable emulation of invalid guest state.
2547 Ignored if kvm-intel.enable_unrestricted_guest=1, as
2548 guest state is never invalid for unrestricted guests.
2549 This param doesn't apply to nested guests (L2), as KVM
2550 never emulates invalid L2 guest state.
2551 Default is 1 (enabled)
2553 kvm-intel.flexpriority=
2554 [KVM,Intel] Disable FlexPriority feature (TPR shadow).
2555 Default is 1 (enabled)
2558 [KVM,Intel] Enable VMX nesting (nVMX).
2559 Default is 0 (disabled)
2561 kvm-intel.unrestricted_guest=
2562 [KVM,Intel] Disable unrestricted guest feature
2563 (virtualized real and unpaged mode) on capable
2564 Intel chips. Default is 1 (enabled)
2566 kvm-intel.vmentry_l1d_flush=[KVM,Intel] Mitigation for L1 Terminal Fault
2569 Valid arguments: never, cond, always
2571 always: L1D cache flush on every VMENTER.
2572 cond: Flush L1D on VMENTER only when the code between
2573 VMEXIT and VMENTER can leak host memory.
2574 never: Disables the mitigation
2576 Default is cond (do L1 cache flush in specific instances)
2578 kvm-intel.vpid= [KVM,Intel] Disable Virtual Processor Identification
2579 feature (tagged TLBs) on capable Intel chips.
2580 Default is 1 (enabled)
2582 l1d_flush= [X86,INTEL]
2583 Control mitigation for L1D based snooping vulnerability.
2585 Certain CPUs are vulnerable to an exploit against CPU
2586 internal buffers which can forward information to a
2587 disclosure gadget under certain conditions.
2589 In vulnerable processors, the speculatively
2590 forwarded data can be used in a cache side channel
2591 attack, to access data to which the attacker does
2592 not have direct access.
2594 This parameter controls the mitigation. The
2597 on - enable the interface for the mitigation
2599 l1tf= [X86] Control mitigation of the L1TF vulnerability on
2602 The kernel PTE inversion protection is unconditionally
2603 enabled and cannot be disabled.
2606 Provides all available mitigations for the
2607 L1TF vulnerability. Disables SMT and
2608 enables all mitigations in the
2609 hypervisors, i.e. unconditional L1D flush.
2611 SMT control and L1D flush control via the
2612 sysfs interface is still possible after
2613 boot. Hypervisors will issue a warning
2614 when the first VM is started in a
2615 potentially insecure configuration,
2616 i.e. SMT enabled or L1D flush disabled.
2619 Same as 'full', but disables SMT and L1D
2620 flush runtime control. Implies the
2621 'nosmt=force' command line option.
2622 (i.e. sysfs control of SMT is disabled.)
2625 Leaves SMT enabled and enables the default
2626 hypervisor mitigation, i.e. conditional
2629 SMT control and L1D flush control via the
2630 sysfs interface is still possible after
2631 boot. Hypervisors will issue a warning
2632 when the first VM is started in a
2633 potentially insecure configuration,
2634 i.e. SMT enabled or L1D flush disabled.
2638 Disables SMT and enables the default
2639 hypervisor mitigation.
2641 SMT control and L1D flush control via the
2642 sysfs interface is still possible after
2643 boot. Hypervisors will issue a warning
2644 when the first VM is started in a
2645 potentially insecure configuration,
2646 i.e. SMT enabled or L1D flush disabled.
2649 Same as 'flush', but hypervisors will not
2650 warn when a VM is started in a potentially
2651 insecure configuration.
2654 Disables hypervisor mitigations and doesn't
2656 It also drops the swap size and available
2657 RAM limit restriction on both hypervisor and
2662 For details see: Documentation/admin-guide/hw-vuln/l1tf.rst
2668 lapic [X86-32,APIC] Enable the local APIC even if BIOS
2671 lapic= [X86,APIC] Do not use TSC deadline
2672 value for LAPIC timer one-shot implementation. Default
2673 back to the programmable timer unit in the LAPIC.
2674 Format: notscdeadline
2676 lapic_timer_c2_ok [X86,APIC] trust the local apic timer
2679 libata.dma= [LIBATA] DMA control
2680 libata.dma=0 Disable all PATA and SATA DMA
2681 libata.dma=1 PATA and SATA Disk DMA only
2682 libata.dma=2 ATAPI (CDROM) DMA only
2683 libata.dma=4 Compact Flash DMA only
2684 Combinations also work, so libata.dma=3 enables DMA
2685 for disks and CDROMs, but not CFs.
2687 libata.ignore_hpa= [LIBATA] Ignore HPA limit
2688 libata.ignore_hpa=0 keep BIOS limits (default)
2689 libata.ignore_hpa=1 ignore limits, using full disk
2691 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume
2695 libata.force= [LIBATA] Force configurations. The format is a comma-
2696 separated list of "[ID:]VAL" where ID is PORT[.DEVICE].
2697 PORT and DEVICE are decimal numbers matching port, link
2698 or device. Basically, it matches the ATA ID string
2699 printed on console by libata. If the whole ID part is
2700 omitted, the last PORT and DEVICE values are used. If
2701 ID hasn't been specified yet, the configuration applies
2702 to all ports, links and devices.
2704 If only DEVICE is omitted, the parameter applies to
2705 the port and all links and devices behind it. DEVICE
2706 number of 0 either selects the first device or the
2707 first fan-out link behind PMP device. It does not
2708 select the host link. DEVICE number of 15 selects the
2709 host link and device attached to it.
2711 The VAL specifies the configuration to force. As long
2712 as there is no ambiguity, shortcut notation is allowed.
2713 For example, both 1.5 and 1.5G would work for 1.5Gbps.
2714 The following configurations can be forced.
2716 * Cable type: 40c, 80c, short40c, unk, ign or sata.
2717 Any ID with matching PORT is used.
2719 * SATA link speed limit: 1.5Gbps or 3.0Gbps.
2721 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7].
2722 udma[/][16,25,33,44,66,100,133] notation is also
2725 * nohrst, nosrst, norst: suppress hard, soft and both
2728 * rstonce: only attempt one reset during hot-unplug
2731 * [no]dbdelay: Enable or disable the extra 200ms delay
2732 before debouncing a link PHY and device presence
2735 * [no]ncq: Turn on or off NCQ.
2737 * [no]ncqtrim: Enable or disable queued DSM TRIM.
2739 * [no]ncqati: Enable or disable NCQ trim on ATI chipset.
2741 * [no]trim: Enable or disable (unqueued) TRIM.
2743 * trim_zero: Indicate that TRIM command zeroes data.
2745 * max_trim_128m: Set 128M maximum trim size limit.
2747 * [no]dma: Turn on or off DMA transfers.
2749 * atapi_dmadir: Enable ATAPI DMADIR bridge support.
2751 * atapi_mod16_dma: Enable the use of ATAPI DMA for
2752 commands that are not a multiple of 16 bytes.
2754 * [no]dmalog: Enable or disable the use of the
2755 READ LOG DMA EXT command to access logs.
2757 * [no]iddevlog: Enable or disable access to the
2758 identify device data log.
2760 * [no]logdir: Enable or disable access to the general
2761 purpose log directory.
2763 * max_sec_128: Set transfer size limit to 128 sectors.
2765 * max_sec_1024: Set or clear transfer size limit to
2768 * max_sec_lba48: Set or clear transfer size limit to
2771 * [no]lpm: Enable or disable link power management.
2773 * [no]setxfer: Indicate if transfer speed mode setting
2776 * dump_id: Dump IDENTIFY data.
2778 * disable: Disable this device.
2780 If there are multiple matching configurations changing
2781 the same attribute, the last one is used.
2783 load_ramdisk= [RAM] [Deprecated]
2785 lockd.nlm_grace_period=P [NFS] Assign grace period.
2788 lockd.nlm_tcpport=N [NFS] Assign TCP port.
2791 lockd.nlm_timeout=T [NFS] Assign timeout value.
2794 lockd.nlm_udpport=M [NFS] Assign UDP port.
2797 lockdown= [SECURITY]
2798 { integrity | confidentiality }
2799 Enable the kernel lockdown feature. If set to
2800 integrity, kernel features that allow userland to
2801 modify the running kernel are disabled. If set to
2802 confidentiality, kernel features that allow userland
2803 to extract confidential information from the kernel
2806 locktorture.nreaders_stress= [KNL]
2807 Set the number of locking read-acquisition kthreads.
2808 Defaults to being automatically set based on the
2809 number of online CPUs.
2811 locktorture.nwriters_stress= [KNL]
2812 Set the number of locking write-acquisition kthreads.
2814 locktorture.onoff_holdoff= [KNL]
2815 Set time (s) after boot for CPU-hotplug testing.
2817 locktorture.onoff_interval= [KNL]
2818 Set time (s) between CPU-hotplug operations, or
2819 zero to disable CPU-hotplug testing.
2821 locktorture.shuffle_interval= [KNL]
2822 Set task-shuffle interval (jiffies). Shuffling
2823 tasks allows some CPUs to go into dyntick-idle
2824 mode during the locktorture test.
2826 locktorture.shutdown_secs= [KNL]
2827 Set time (s) after boot system shutdown. This
2828 is useful for hands-off automated testing.
2830 locktorture.stat_interval= [KNL]
2831 Time (s) between statistics printk()s.
2833 locktorture.stutter= [KNL]
2834 Time (s) to stutter testing, for example,
2835 specifying five seconds causes the test to run for
2836 five seconds, wait for five seconds, and so on.
2837 This tests the locking primitive's ability to
2838 transition abruptly to and from idle.
2840 locktorture.torture_type= [KNL]
2841 Specify the locking implementation to test.
2843 locktorture.verbose= [KNL]
2844 Enable additional printk() statements.
2846 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver
2849 loglevel= All Kernel Messages with a loglevel smaller than the
2850 console loglevel will be printed to the console. It can
2851 also be changed with klogd or other programs. The
2852 loglevels are defined as follows:
2854 0 (KERN_EMERG) system is unusable
2855 1 (KERN_ALERT) action must be taken immediately
2856 2 (KERN_CRIT) critical conditions
2857 3 (KERN_ERR) error conditions
2858 4 (KERN_WARNING) warning conditions
2859 5 (KERN_NOTICE) normal but significant condition
2860 6 (KERN_INFO) informational
2861 7 (KERN_DEBUG) debug-level messages
2863 log_buf_len=n[KMG] Sets the size of the printk ring buffer,
2864 in bytes. n must be a power of two and greater
2865 than the minimal size. The minimal size is defined
2866 by LOG_BUF_SHIFT kernel config parameter. There is
2867 also CONFIG_LOG_CPU_MAX_BUF_SHIFT config parameter
2868 that allows to increase the default size depending on
2869 the number of CPUs. See init/Kconfig for more details.
2871 logo.nologo [FB] Disables display of the built-in Linux logo.
2872 This may be used to provide more screen space for
2873 kernel log messages and is useful when debugging
2874 kernel boot problems.
2876 lp=0 [LP] Specify parallel ports to use, e.g,
2877 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses
2878 lp=reset first parallel port). 'lp=0' disables the
2879 lp=auto printer driver. 'lp=reset' (which can be
2880 specified in addition to the ports) causes
2881 attached printers to be reset. Using
2882 lp=port1,port2,... specifies the parallel ports
2883 to associate lp devices with, starting with
2884 lp0. A port specification may be 'none' to skip
2885 that lp device, or a parport name such as
2886 'parport0'. Specifying 'lp=auto' instead of a
2887 port specification list means that device IDs
2888 from each port should be examined, to see if
2889 an IEEE 1284-compliant printer is attached; if
2890 so, the driver will manage that printer.
2891 See also header of drivers/char/lp.c.
2894 Sets loops_per_jiffy to given constant, thus avoiding
2895 time-consuming boot-time autodetection (up to 250 ms per
2896 CPU). 0 enables autodetection (default). To determine
2897 the correct value for your kernel, boot with normal
2898 autodetection and see what value is printed. Note that
2899 on SMP systems the preset will be applied to all CPUs,
2900 which is likely to cause problems if your CPUs need
2901 significantly divergent settings. An incorrect value
2902 will cause delays in the kernel to be wrong, leading to
2903 unpredictable I/O errors and other breakage. Although
2904 unlikely, in the extreme case this might damage your
2908 Format: <io>,<irq>,<dma>
2910 lsm.debug [SECURITY] Enable LSM initialization debugging output.
2913 [SECURITY] Choose order of LSM initialization. This
2914 overrides CONFIG_LSM, and the "security=" parameter.
2916 machvec= [IA-64] Force the use of a particular machine-vector
2917 (machvec) in a generic kernel.
2918 Example: machvec=hpzx1
2920 machtype= [Loongson] Share the same kernel image file between
2921 different yeeloong laptops.
2922 Example: machtype=lemote-yeeloong-2f-7inch
2924 max_addr=nn[KMG] [KNL,BOOT,IA-64] All physical memory greater
2925 than or equal to this physical address is ignored.
2927 maxcpus= [SMP] Maximum number of processors that an SMP kernel
2928 will bring up during bootup. maxcpus=n : n >= 0 limits
2929 the kernel to bring up 'n' processors. Surely after
2930 bootup you can bring up the other plugged cpu by executing
2931 "echo 1 > /sys/devices/system/cpu/cpuX/online". So maxcpus
2932 only takes effect during system bootup.
2933 While n=0 is a special case, it is equivalent to "nosmp",
2934 which also disables the IO APIC.
2936 max_loop= [LOOP] The number of loop block devices that get
2937 (loop.max_loop) unconditionally pre-created at init time. The default
2938 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead
2939 of statically allocating a predefined number, loop
2940 devices can be requested on-demand with the
2941 /dev/loop-control interface.
2943 mce [X86-32] Machine Check Exception
2945 mce=option [X86-64] See Documentation/x86/x86_64/boot-options.rst
2947 md= [HW] RAID subsystems devices and level
2948 See Documentation/admin-guide/md.rst.
2951 Format: <first>,<last>
2952 Specifies range of consoles to be captured by the MDA.
2955 Control mitigation for the Micro-architectural Data
2956 Sampling (MDS) vulnerability.
2958 Certain CPUs are vulnerable to an exploit against CPU
2959 internal buffers which can forward information to a
2960 disclosure gadget under certain conditions.
2962 In vulnerable processors, the speculatively
2963 forwarded data can be used in a cache side channel
2964 attack, to access data to which the attacker does
2965 not have direct access.
2967 This parameter controls the MDS mitigation. The
2970 full - Enable MDS mitigation on vulnerable CPUs
2971 full,nosmt - Enable MDS mitigation and disable
2972 SMT on vulnerable CPUs
2973 off - Unconditionally disable MDS mitigation
2975 On TAA-affected machines, mds=off can be prevented by
2976 an active TAA mitigation as both vulnerabilities are
2977 mitigated with the same mechanism so in order to disable
2978 this mitigation, you need to specify tsx_async_abort=off
2981 Not specifying this option is equivalent to
2984 For details see: Documentation/admin-guide/hw-vuln/mds.rst
2986 mem=nn[KMG] [HEXAGON] Set the memory size.
2987 Must be specified, otherwise memory size will be 0.
2989 mem=nn[KMG] [KNL,BOOT] Force usage of a specific amount of memory
2990 Amount of memory to be used in cases as follows:
2993 2 when the kernel is not able to see the whole system memory;
2994 3 memory that lies after 'mem=' boundary is excluded from
2995 the hypervisor, then assigned to KVM guests.
2996 4 to limit the memory available for kdump kernel.
2998 [ARC,MICROBLAZE] - the limit applies only to low memory,
2999 high memory is not affected.
3001 [ARM64] - only limits memory covered by the linear
3002 mapping. The NOMAP regions are not affected.
3004 [X86] Work as limiting max address. Use together
3005 with memmap= to avoid physical address space collisions.
3006 Without memmap= PCI devices could be placed at addresses
3007 belonging to unused RAM.
3009 Note that this only takes effects during boot time since
3010 in above case 3, memory may need be hot added after boot
3011 if system memory of hypervisor is not sufficient.
3014 [ARM,MIPS] - override the memory layout reported by
3016 Define a memory region of size nn[KMG] starting at
3018 Multiple different regions can be specified with
3019 multiple mem= parameters on the command line.
3021 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel
3024 memblock=debug [KNL] Enable memblock debug messages.
3027 [KNL,SH] Allow user to override the default size for
3028 per-device physically contiguous DMA buffers.
3030 memhp_default_state=online/offline
3031 [KNL] Set the initial state for the memory hotplug
3032 onlining policy. If not specified, the default value is
3033 set according to the
3034 CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE kernel config
3036 See Documentation/admin-guide/mm/memory-hotplug.rst.
3038 memmap=exactmap [KNL,X86] Enable setting of an exact
3039 E820 memory map, as specified by the user.
3040 Such memmap=exactmap lines can be constructed based on
3041 BIOS output or other requirements. See the memmap=nn@ss
3044 memmap=nn[KMG]@ss[KMG]
3045 [KNL, X86, MIPS, XTENSA] Force usage of a specific region of memory.
3046 Region of memory to be used is from ss to ss+nn.
3047 If @ss[KMG] is omitted, it is equivalent to mem=nn[KMG],
3048 which limits max address to nn[KMG].
3049 Multiple different regions can be specified,
3052 memmap=100M@2G,100M#3G,1G!1024G
3054 memmap=nn[KMG]#ss[KMG]
3055 [KNL,ACPI] Mark specific memory as ACPI data.
3056 Region of memory to be marked is from ss to ss+nn.
3058 memmap=nn[KMG]$ss[KMG]
3059 [KNL,ACPI] Mark specific memory as reserved.
3060 Region of memory to be reserved is from ss to ss+nn.
3061 Example: Exclude memory from 0x18690000-0x1869ffff
3062 memmap=64K$0x18690000
3064 memmap=0x10000$0x18690000
3065 Some bootloaders may need an escape character before '$',
3066 like Grub2, otherwise '$' and the following number
3069 memmap=nn[KMG]!ss[KMG]
3070 [KNL,X86] Mark specific memory as protected.
3071 Region of memory to be used, from ss to ss+nn.
3072 The memory region may be marked as e820 type 12 (0xc)
3073 and is NVDIMM or ADR memory.
3075 memmap=<size>%<offset>-<oldtype>+<newtype>
3076 [KNL,ACPI] Convert memory within the specified region
3077 from <oldtype> to <newtype>. If "-<oldtype>" is left
3078 out, the whole region will be marked as <newtype>,
3079 even if previously unavailable. If "+<newtype>" is left
3080 out, matching memory will be removed. Types are
3081 specified as e820 types, e.g., 1 = RAM, 2 = reserved,
3082 3 = ACPI, 12 = PRAM.
3084 memory_corruption_check=0/1 [X86]
3085 Some BIOSes seem to corrupt the first 64k of
3086 memory when doing things like suspend/resume.
3087 Setting this option will scan the memory
3088 looking for corruption. Enabling this will
3089 both detect corruption and prevent the kernel
3090 from using the memory being corrupted.
3091 However, its intended as a diagnostic tool; if
3092 repeatable BIOS-originated corruption always
3093 affects the same memory, you can use memmap=
3094 to prevent the kernel from using that memory.
3096 memory_corruption_check_size=size [X86]
3097 By default it checks for corruption in the low
3098 64k, making this memory unavailable for normal
3099 use. Use this parameter to scan for
3100 corruption in more or less memory.
3102 memory_corruption_check_period=seconds [X86]
3103 By default it checks for corruption every 60
3104 seconds. Use this parameter to check at some
3105 other rate. 0 disables periodic checking.
3107 memory_hotplug.memmap_on_memory
3108 [KNL,X86,ARM] Boolean flag to enable this feature.
3109 Format: {on | off (default)}
3110 When enabled, runtime hotplugged memory will
3111 allocate its internal metadata (struct pages,
3112 those vmemmap pages cannot be optimized even
3113 if hugetlb_free_vmemmap is enabled) from the
3114 hotadded memory which will allow to hotadd a
3115 lot of memory without requiring additional
3117 This feature is disabled by default because it
3118 has some implication on large (e.g. GB)
3119 allocations in some configurations (e.g. small
3121 The state of the flag can be read in
3122 /sys/module/memory_hotplug/parameters/memmap_on_memory.
3123 Note that even when enabled, there are a few cases where
3124 the feature is not effective.
3126 memtest= [KNL,X86,ARM,M68K,PPC,RISCV] Enable memtest
3128 default : 0 <disable>
3129 Specifies the number of memtest passes to be
3130 performed. Each pass selects another test
3131 pattern from a given set of patterns. Memtest
3132 fills the memory with this pattern, validates
3133 memory contents and reserves bad memory
3134 regions that are detected.
3136 mem_encrypt= [X86-64] AMD Secure Memory Encryption (SME) control
3137 Valid arguments: on, off
3138 Default (depends on kernel configuration option):
3139 on (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=y)
3140 off (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=n)
3141 mem_encrypt=on: Activate SME
3142 mem_encrypt=off: Do not activate SME
3144 Refer to Documentation/virt/kvm/x86/amd-memory-encryption.rst
3145 for details on when memory encryption can be activated.
3147 mem_sleep_default= [SUSPEND] Default system suspend mode:
3148 s2idle - Suspend-To-Idle
3149 shallow - Power-On Suspend or equivalent (if supported)
3150 deep - Suspend-To-RAM or equivalent (if supported)
3151 See Documentation/admin-guide/pm/sleep-states.rst.
3153 meye.*= [HW] Set MotionEye Camera parameters
3154 See Documentation/admin-guide/media/meye.rst.
3156 mfgpt_irq= [IA-32] Specify the IRQ to use for the
3157 Multi-Function General Purpose Timers on AMD Geode
3160 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when
3161 the BIOS has incorrectly applied a workaround. TinyBIOS
3162 version 0.98 is known to be affected, 0.99 fixes the
3163 problem by letting the user disable the workaround.
3167 min_addr=nn[KMG] [KNL,BOOT,IA-64] All physical memory below this
3168 physical address is ignored.
3170 mini2440= [ARM,HW,KNL]
3171 Format:[0..2][b][c][t]
3173 MINI2440 configuration specification:
3174 0 - The attached screen is the 3.5" TFT
3175 1 - The attached screen is the 7" TFT
3176 2 - The VGA Shield is attached (1024x768)
3177 Leaving out the screen size parameter will not load
3178 the TFT driver, and the framebuffer will be left
3180 b - Enable backlight. The TFT backlight pin will be
3181 linked to the kernel VESA blanking code and a GPIO
3182 LED. This parameter is not necessary when using the
3184 c - Enable the s3c camera interface.
3185 t - Reserved for enabling touchscreen support. The
3186 touchscreen support is not enabled in the mainstream
3187 kernel as of 2.6.30, a preliminary port can be found
3188 in the "bleeding edge" mini2440 support kernel at
3189 https://repo.or.cz/w/linux-2.6/mini2440.git
3192 [X86,PPC,S390,ARM64] Control optional mitigations for
3193 CPU vulnerabilities. This is a set of curated,
3194 arch-independent options, each of which is an
3195 aggregation of existing arch-specific options.
3198 Disable all optional CPU mitigations. This
3199 improves system performance, but it may also
3200 expose users to several CPU vulnerabilities.
3201 Equivalent to: nopti [X86,PPC]
3202 if nokaslr then kpti=0 [ARM64]
3203 nospectre_v1 [X86,PPC]
3205 nospectre_v2 [X86,PPC,S390,ARM64]
3206 spectre_v2_user=off [X86]
3207 spec_store_bypass_disable=off [X86,PPC]
3208 ssbd=force-off [ARM64]
3211 tsx_async_abort=off [X86]
3212 kvm.nx_huge_pages=off [X86]
3213 srbds=off [X86,INTEL]
3214 no_entry_flush [PPC]
3215 no_uaccess_flush [PPC]
3216 mmio_stale_data=off [X86]
3220 This does not have any effect on
3221 kvm.nx_huge_pages when
3222 kvm.nx_huge_pages=force.
3225 Mitigate all CPU vulnerabilities, but leave SMT
3226 enabled, even if it's vulnerable. This is for
3227 users who don't want to be surprised by SMT
3228 getting disabled across kernel upgrades, or who
3229 have other ways of avoiding SMT-based attacks.
3230 Equivalent to: (default behavior)
3233 Mitigate all CPU vulnerabilities, disabling SMT
3234 if needed. This is for users who always want to
3235 be fully mitigated, even if it means losing SMT.
3236 Equivalent to: l1tf=flush,nosmt [X86]
3237 mds=full,nosmt [X86]
3238 tsx_async_abort=full,nosmt [X86]
3239 mmio_stale_data=full,nosmt [X86]
3240 retbleed=auto,nosmt [X86]
3243 [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this
3244 parameter allows control of the logging verbosity for
3245 the additional memory initialisation checks. A value
3246 of 0 disables mminit logging and a level of 4 will
3247 log everything. Information is printed at KERN_DEBUG
3248 so loglevel=8 may also need to be specified.
3251 [X86,INTEL] Control mitigation for the Processor
3252 MMIO Stale Data vulnerabilities.
3254 Processor MMIO Stale Data is a class of
3255 vulnerabilities that may expose data after an MMIO
3256 operation. Exposed data could originate or end in
3257 the same CPU buffers as affected by MDS and TAA.
3258 Therefore, similar to MDS and TAA, the mitigation
3259 is to clear the affected CPU buffers.
3261 This parameter controls the mitigation. The
3264 full - Enable mitigation on vulnerable CPUs
3266 full,nosmt - Enable mitigation and disable SMT on
3269 off - Unconditionally disable mitigation
3271 On MDS or TAA affected machines,
3272 mmio_stale_data=off can be prevented by an active
3273 MDS or TAA mitigation as these vulnerabilities are
3274 mitigated with the same mechanism so in order to
3275 disable this mitigation, you need to specify
3276 mds=off and tsx_async_abort=off too.
3278 Not specifying this option is equivalent to
3279 mmio_stale_data=full.
3282 Documentation/admin-guide/hw-vuln/processor_mmio_stale_data.rst
3284 module.async_probe=<bool>
3285 [KNL] When set to true, modules will use async probing
3286 by default. To enable/disable async probing for a
3287 specific module, use the module specific control that
3288 is documented under <module>.async_probe. When both
3289 module.async_probe and <module>.async_probe are
3290 specified, <module>.async_probe takes precedence for
3291 the specific module.
3294 [KNL] When CONFIG_MODULE_SIG is set, this means that
3295 modules without (valid) signatures will fail to load.
3296 Note that if CONFIG_MODULE_SIG_FORCE is set, that
3297 is always true, so this option does nothing.
3299 module_blacklist= [KNL] Do not load a comma-separated list of
3300 modules. Useful for debugging problem modules.
3303 [MOUSE] Maximum time between finger touching and
3304 leaving touchpad surface for touch to be considered
3305 a tap and be reported as a left button click (for
3306 touchpads working in absolute mode only).
3308 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices
3309 reporting absolute coordinates, such as tablets
3310 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices
3311 reporting absolute coordinates, such as tablets
3313 movablecore= [KNL,X86,IA-64,PPC]
3314 Format: nn[KMGTPE] | nn%
3315 This parameter is the complement to kernelcore=, it
3316 specifies the amount of memory used for migratable
3317 allocations. If both kernelcore and movablecore is
3318 specified, then kernelcore will be at *least* the
3319 specified value but may be more. If movablecore on its
3320 own is specified, the administrator must be careful
3321 that the amount of memory usable for all allocations
3324 movable_node [KNL] Boot-time switch to make hotplugable memory
3325 NUMA nodes to be movable. This means that the memory
3326 of such nodes will be usable only for movable
3327 allocations which rules out almost all kernel
3328 allocations. Use with caution!
3330 MTD_Partition= [MTD]
3331 Format: <name>,<region-number>,<size>,<offset>
3333 MTD_Region= [MTD] Format:
3334 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>]
3337 See drivers/mtd/parsers/cmdlinepart.c
3340 ARM/S3C2412 JIVE boot control
3342 See arch/arm/mach-s3c/mach-jive.c
3344 mtouchusb.raw_coordinates=
3345 [HW] Make the MicroTouch USB driver use raw coordinates
3346 ('y', default) or cooked coordinates ('n')
3348 mtrr_chunk_size=nn[KMG] [X86]
3349 used for mtrr cleanup. It is largest continuous chunk
3350 that could hold holes aka. UC entries.
3352 mtrr_gran_size=nn[KMG] [X86]
3353 Used for mtrr cleanup. It is granularity of mtrr block.
3355 Large value could prevent small alignment from
3358 mtrr_spare_reg_nr=n [X86]
3360 Range: 0,7 : spare reg number
3362 Used for mtrr cleanup. It is spare mtrr entries number.
3363 Set to 2 or more if your graphical card needs more.
3365 multitce=off [PPC] This parameter disables the use of the pSeries
3366 firmware feature for updating multiple TCE entries
3369 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card
3371 netdev= [NET] Network devices parameters
3372 Format: <irq>,<io>,<mem_start>,<mem_end>,<name>
3373 Note that mem_start is often overloaded to mean
3374 something different and driver-specific.
3375 This usage is only documented in each driver source
3378 netpoll.carrier_timeout=
3379 [NET] Specifies amount of time (in seconds) that
3380 netpoll should wait for a carrier. By default netpoll
3384 [NETFILTER] Enable connection tracking flow accounting
3385 0 to disable accounting
3386 1 to enable accounting
3389 nfsaddrs= [NFS] Deprecated. Use ip= instead.
3390 See Documentation/admin-guide/nfs/nfsroot.rst.
3392 nfsroot= [NFS] nfs root filesystem for disk-less boxes.
3393 See Documentation/admin-guide/nfs/nfsroot.rst.
3395 nfsrootdebug [NFS] enable nfsroot debugging messages.
3396 See Documentation/admin-guide/nfs/nfsroot.rst.
3398 nfs.callback_nr_threads=
3399 [NFSv4] set the total number of threads that the
3400 NFS client will assign to service NFSv4 callback
3403 nfs.callback_tcpport=
3404 [NFS] set the TCP port on which the NFSv4 callback
3405 channel should listen.
3408 [NFS] sets the pathname to the program which is used
3409 to update the NFS client cache entries.
3411 nfs.cache_getent_timeout=
3412 [NFS] sets the timeout after which an attempt to
3413 update a cache entry is deemed to have failed.
3415 nfs.idmap_cache_timeout=
3416 [NFS] set the maximum lifetime for idmapper cache
3420 [NFS] enable 64-bit inode numbers.
3421 If zero, the NFS client will fake up a 32-bit inode
3422 number for the readdir() and stat() syscalls instead
3423 of returning the full 64-bit number.
3424 The default is to return 64-bit inode numbers.
3426 nfs.max_session_cb_slots=
3427 [NFSv4.1] Sets the maximum number of session
3428 slots the client will assign to the callback
3429 channel. This determines the maximum number of
3430 callbacks the client will process in parallel for
3431 a particular server.
3433 nfs.max_session_slots=
3434 [NFSv4.1] Sets the maximum number of session slots
3435 the client will attempt to negotiate with the server.
3436 This limits the number of simultaneous RPC requests
3437 that the client can send to the NFSv4.1 server.
3438 Note that there is little point in setting this
3439 value higher than the max_tcp_slot_table_limit.
3441 nfs.nfs4_disable_idmapping=
3442 [NFSv4] When set to the default of '1', this option
3443 ensures that both the RPC level authentication
3444 scheme and the NFS level operations agree to use
3445 numeric uids/gids if the mount is using the
3446 'sec=sys' security flavour. In effect it is
3447 disabling idmapping, which can make migration from
3448 legacy NFSv2/v3 systems to NFSv4 easier.
3449 Servers that do not support this mode of operation
3450 will be autodetected by the client, and it will fall
3451 back to using the idmapper.
3452 To turn off this behaviour, set the value to '0'.
3454 [NFS4] Specify an additional fixed unique ident-
3455 ification string that NFSv4 clients can insert into
3456 their nfs_client_id4 string. This is typically a
3457 UUID that is generated at system install time.
3459 nfs.send_implementation_id =
3460 [NFSv4.1] Send client implementation identification
3461 information in exchange_id requests.
3462 If zero, no implementation identification information
3464 The default is to send the implementation identification
3467 nfs.recover_lost_locks =
3468 [NFSv4] Attempt to recover locks that were lost due
3469 to a lease timeout on the server. Please note that
3470 doing this risks data corruption, since there are
3471 no guarantees that the file will remain unchanged
3472 after the locks are lost.
3473 If you want to enable the kernel legacy behaviour of
3474 attempting to recover these locks, then set this
3476 The default parameter value of '0' causes the kernel
3477 not to attempt recovery of lost locks.
3479 nfs4.layoutstats_timer =
3480 [NFSv4.2] Change the rate at which the kernel sends
3481 layoutstats to the pNFS metadata server.
3483 Setting this to value to 0 causes the kernel to use
3484 whatever value is the default set by the layout
3485 driver. A non-zero value sets the minimum interval
3486 in seconds between layoutstats transmissions.
3488 nfsd.inter_copy_offload_enable =
3489 [NFSv4.2] When set to 1, the server will support
3490 server-to-server copies for which this server is
3491 the destination of the copy.
3493 nfsd.nfsd4_ssc_umount_timeout =
3494 [NFSv4.2] When used as the destination of a
3495 server-to-server copy, knfsd temporarily mounts
3496 the source server. It caches the mount in case
3497 it will be needed again, and discards it if not
3498 used for the number of milliseconds specified by
3501 nfsd.nfs4_disable_idmapping=
3502 [NFSv4] When set to the default of '1', the NFSv4
3503 server will return only numeric uids and gids to
3504 clients using auth_sys, and will accept numeric uids
3505 and gids from such clients. This is intended to ease
3506 migration from NFSv2/v3.
3509 nmi_backtrace.backtrace_idle [KNL]
3510 Dump stacks even of idle CPUs in response to an
3511 NMI stack-backtrace request.
3513 nmi_debug= [KNL,SH] Specify one or more actions to take
3514 when a NMI is triggered.
3515 Format: [state][,regs][,debounce][,die]
3517 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels
3518 Format: [panic,][nopanic,][num]
3520 0 - turn hardlockup detector in nmi_watchdog off
3521 1 - turn hardlockup detector in nmi_watchdog on
3522 When panic is specified, panic when an NMI watchdog
3523 timeout occurs (or 'nopanic' to not panic on an NMI
3524 watchdog, if CONFIG_BOOTPARAM_HARDLOCKUP_PANIC is set)
3525 To disable both hard and soft lockup detectors,
3526 please see 'nowatchdog'.
3527 This is useful when you use a panic=... timeout and
3528 need the box quickly up again.
3530 These settings can be accessed at runtime via
3531 the nmi_watchdog and hardlockup_panic sysctls.
3533 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths
3534 emulation library even if a 387 maths coprocessor
3537 no5lvl [X86-64] Disable 5-level paging mode. Forces
3538 kernel to use 4-level paging instead.
3540 nofsgsbase [X86] Disables FSGSBASE instructions.
3543 [HW] Never suspend the console
3544 Disable suspending of consoles during suspend and
3545 hibernate operations. Once disabled, debugging
3546 messages can reach various consoles while the rest
3547 of the system is being put to sleep (ie, while
3548 debugging driver suspend/resume hooks). This may
3549 not work reliably with all consoles, but is known
3550 to work with serial and VGA consoles.
3551 To facilitate more flexible debugging, we also add
3552 console_suspend, a printk module parameter to control
3553 it. Users could use console_suspend (usually
3554 /sys/module/printk/parameters/console_suspend) to
3555 turn on/off it dynamically.
3557 novmcoredd [KNL,KDUMP]
3558 Disable device dump. Device dump allows drivers to
3559 append dump data to vmcore so you can collect driver
3560 specified debug info. Drivers can append the data
3561 without any limit and this data is stored in memory,
3562 so this may cause significant memory stress. Disabling
3563 device dump can help save memory but the driver debug
3564 data will be no longer available. This parameter
3565 is only available when CONFIG_PROC_VMCORE_DEVICE_DUMP
3568 noaliencache [MM, NUMA, SLAB] Disables the allocation of alien
3569 caches in the slab allocator. Saves per-node memory,
3570 but will impact performance.
3574 noaltinstr [S390] Disables alternative instructions patching
3575 (CPU alternatives feature).
3577 noapic [SMP,APIC] Tells the kernel to not make use of any
3578 IOAPICs that may be present in the system.
3580 noautogroup Disable scheduler automatic task group creation.
3584 nodsp [SH] Disable hardware DSP at boot time.
3586 noefi Disable EFI runtime services support.
3588 no_entry_flush [PPC] Don't flush the L1-D cache when entering the kernel.
3593 Disable SMAP (Supervisor Mode Access Prevention)
3594 even if it is supported by processor.
3597 Disable SMEP (Supervisor Mode Execution Prevention)
3598 even if it is supported by processor.
3601 This affects only 32-bit executables.
3602 noexec32=on: enable non-executable mappings (default)
3603 read doesn't imply executable mappings
3604 noexec32=off: disable non-executable mappings
3605 read implies executable mappings
3607 nofpu [MIPS,SH] Disable hardware FPU at boot time.
3609 nofxsr [BUGS=X86-32] Disables x86 floating point extended
3610 register save and restore. The kernel will only save
3611 legacy floating-point registers on task switch.
3613 nohugeiomap [KNL,X86,PPC,ARM64] Disable kernel huge I/O mappings.
3615 nohugevmalloc [PPC] Disable kernel huge vmalloc mappings.
3617 nosmt [KNL,S390] Disable symmetric multithreading (SMT).
3618 Equivalent to smt=1.
3620 [KNL,X86] Disable symmetric multithreading (SMT).
3621 nosmt=force: Force disable SMT, cannot be undone
3622 via the sysfs control file.
3624 nospectre_v1 [X86,PPC] Disable mitigations for Spectre Variant 1
3625 (bounds check bypass). With this option data leaks are
3626 possible in the system.
3628 nospectre_v2 [X86,PPC_FSL_BOOK3E,ARM64] Disable all mitigations for
3629 the Spectre variant 2 (indirect branch prediction)
3630 vulnerability. System may allow data leaks with this
3633 nospec_store_bypass_disable
3634 [HW] Disable all mitigations for the Speculative Store Bypass vulnerability
3637 [PPC] Don't flush the L1-D cache after accessing user data.
3639 noxsave [BUGS=X86] Disables x86 extended register state save
3640 and restore using xsave. The kernel will fallback to
3641 enabling legacy floating-point and sse state.
3643 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended
3644 register states. The kernel will fall back to use
3645 xsave to save the states. By using this parameter,
3646 performance of saving the states is degraded because
3647 xsave doesn't support modified optimization while
3648 xsaveopt supports it on xsaveopt enabled systems.
3650 noxsaves [X86] Disables xsaves and xrstors used in saving and
3651 restoring x86 extended register state in compacted
3652 form of xsave area. The kernel will fall back to use
3653 xsaveopt and xrstor to save and restore the states
3654 in standard form of xsave area. By using this
3655 parameter, xsave area per process might occupy more
3656 memory on xsaves enabled systems.
3658 nohlt [ARM,ARM64,MICROBLAZE,SH] Forces the kernel to busy wait
3659 in do_idle() and not use the arch_cpu_idle()
3660 implementation; requires CONFIG_GENERIC_IDLE_POLL_SETUP
3661 to be effective. This is useful on platforms where the
3662 sleep(SH) or wfi(ARM,ARM64) instructions do not work
3663 correctly or when doing power measurements to evalute
3664 the impact of the sleep instructions. This is also
3665 useful when using JTAG debugger.
3667 no_file_caps Tells the kernel not to honor file capabilities. The
3668 only way then for a file to be executed with privilege
3669 is to be setuid root or executed by root.
3671 nohalt [IA-64] Tells the kernel not to use the power saving
3672 function PAL_HALT_LIGHT when idle. This increases
3673 power-consumption. On the positive side, it reduces
3674 interrupt wake-up latency, which may improve performance
3675 in certain environments such as networked servers or
3679 Force pointers printed to the console or buffers to be
3680 unhashed. By default, when a pointer is printed via %p
3681 format string, that pointer is "hashed", i.e. obscured
3682 by hashing the pointer value. This is a security feature
3683 that hides actual kernel addresses from unprivileged
3684 users, but it also makes debugging the kernel more
3685 difficult since unequal pointers can no longer be
3686 compared. However, if this command-line option is
3687 specified, then all normal pointers will have their true
3688 value printed. This option should only be specified when
3689 debugging the kernel. Please do not use on production
3692 nohibernate [HIBERNATION] Disable hibernation and resume.
3694 nohz= [KNL] Boottime enable/disable dynamic ticks
3695 Valid arguments: on, off
3698 nohz_full= [KNL,BOOT,SMP,ISOL]
3699 The argument is a cpu list, as described above.
3700 In kernels built with CONFIG_NO_HZ_FULL=y, set
3701 the specified list of CPUs whose tick will be stopped
3702 whenever possible. The boot CPU will be forced outside
3703 the range to maintain the timekeeping. Any CPUs
3704 in this list will have their RCU callbacks offloaded,
3705 just as if they had also been called out in the
3706 rcu_nocbs= boot parameter.
3708 Note that this argument takes precedence over
3709 the CONFIG_RCU_NOCB_CPU_DEFAULT_ALL option.
3711 noiotrap [SH] Disables trapped I/O port accesses.
3713 noirqdebug [X86-32] Disables the code which attempts to detect and
3714 disable unhandled interrupt sources.
3716 no_timer_check [X86,APIC] Disables the code which tests for
3717 broken timer IRQ sources.
3719 noisapnp [ISAPNP] Disables ISA PnP code.
3721 noinitrd [RAM] Tells the kernel not to load any configured
3724 nointremap [X86-64, Intel-IOMMU] Do not enable interrupt
3726 [Deprecated - use intremap=off]
3730 noinvpcid [X86] Disable the INVPCID cpu feature.
3732 nojitter [IA-64] Disables jitter checking for ITC timers.
3734 no-kvmclock [X86,KVM] Disable paravirtualized KVM clock driver
3736 no-kvmapf [X86,KVM] Disable paravirtualized asynchronous page
3740 [X86,PV_OPS] Disable paravirtualized VMware scheduler
3741 clock and use the default one.
3743 no-steal-acc [X86,PV_OPS,ARM64] Disable paravirtualized steal time
3744 accounting. steal time is computed, but won't
3745 influence scheduler behaviour
3747 nolapic [X86-32,APIC] Do not enable or use the local APIC.
3749 nolapic_timer [X86-32,APIC] Do not use the local APIC timer.
3751 nomca [IA-64] Disable machine check abort handling
3753 nomce [X86-32] Disable Machine Check Exception
3755 nomfgpt [X86-32] Disable Multi-Function General Purpose
3756 Timer usage (for AMD Geode machines).
3758 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to
3759 shutdown the other cpus. Instead use the REBOOT_VECTOR
3762 nomodeset Disable kernel modesetting. DRM drivers will not perform
3763 display-mode changes or accelerated rendering. Only the
3764 system framebuffer will be available for use if this was
3765 set-up by the firmware or boot loader.
3767 Useful as fallback, or for testing and debugging.
3769 nomodule Disable module load
3771 nopat [X86] Disable PAT (page attribute table extension of
3772 pagetables) support.
3774 nopcid [X86-64] Disable the PCID cpu feature.
3776 norandmaps Don't use address space randomization. Equivalent to
3777 echo 0 > /proc/sys/kernel/randomize_va_space
3779 noreplace-smp [X86-32,SMP] Don't replace SMP instructions
3780 with UP alternatives
3782 noresume [SWSUSP] Disables resume and restores original swap
3785 no-scroll [VGA] Disables scrollback.
3786 This is required for the Braillex ib80-piezo Braille
3787 reader made by F.H. Papenmeier (Germany).
3791 nosgx [X86-64,SGX] Disables Intel SGX kernel support.
3793 nosmp [SMP] Tells an SMP kernel to act as a UP kernel,
3794 and disable the IO APIC. legacy for "maxcpus=0".
3796 nosoftlockup [KNL] Disable the soft-lockup detector.
3798 nosync [HW,M68K] Disables sync negotiation for all devices.
3800 nowatchdog [KNL] Disable both lockup detectors, i.e.
3801 soft-lockup and NMI watchdog (hard-lockup).
3805 nox2apic [X86-64,APIC] Do not enable x2APIC mode.
3807 nps_mtm_hs_ctr= [KNL,ARC]
3808 This parameter sets the maximum duration, in
3809 cycles, each HW thread of the CTOP can run
3810 without interruptions, before HW switches it.
3811 The actual maximum duration is 16 times this
3813 Format: integer between 1 and 255
3816 nptcg= [IA-64] Override max number of concurrent global TLB
3817 purges which is reported from either PAL_VM_SUMMARY or
3820 nr_cpus= [SMP] Maximum number of processors that an SMP kernel
3821 could support. nr_cpus=n : n >= 1 limits the kernel to
3822 support 'n' processors. It could be larger than the
3823 number of already plugged CPU during bootup, later in
3824 runtime you can physically add extra cpu until it reaches
3825 n. So during boot up some boot time memory for per-cpu
3826 variables need be pre-allocated for later physical cpu
3829 nr_uarts= [SERIAL] maximum number of UARTs to be registered.
3831 numa=off [KNL, ARM64, PPC, RISCV, SPARC, X86] Disable NUMA, Only
3832 set up a single NUMA node spanning all memory.
3834 numa_balancing= [KNL,ARM64,PPC,RISCV,S390,X86] Enable or disable automatic
3836 Allowed values are enable and disable
3838 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA.
3839 'node', 'default' can be specified
3840 This can be set from sysctl after boot.
3841 See Documentation/admin-guide/sysctl/vm.rst for details.
3843 ohci1394_dma=early [HW] enable debugging via the ohci1394 driver.
3844 See Documentation/core-api/debugging-via-ohci1394.rst for more
3847 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands
3848 Rather than timing out after 20 ms if an EC
3849 command is not properly ACKed, override the length
3850 of the timeout. We have interrupts disabled while
3851 waiting for the ACK, so if this is set too high
3852 interrupts *may* be lost!
3854 omap_mux= [OMAP] Override bootloader pin multiplexing.
3855 Format: <mux_mode0.mode_name=value>...
3856 For example, to override I2C bus2:
3857 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100
3859 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration
3861 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock]
3863 boundary - index of last SLC block on Flex-OneNAND.
3864 The remaining blocks are configured as MLC blocks.
3865 lock - Configure if Flex-OneNAND boundary should be locked.
3866 Once locked, the boundary cannot be changed.
3867 1 indicates lock status, 0 indicates unlock status.
3869 oops=panic Always panic on oopses. Default is to just kill the
3870 process, but there is a small probability of
3871 deadlocking the machine.
3872 This will also cause panics on machine check exceptions.
3873 Useful together with panic=30 to trigger a reboot.
3876 [KNL] Boolean flag to control whether the page allocator
3877 should randomize its free lists. The randomization may
3878 be automatically enabled if the kernel detects it is
3879 running on a platform with a direct-mapped memory-side
3880 cache, and this parameter can be used to
3881 override/disable that behavior. The state of the flag
3882 can be read from sysfs at:
3883 /sys/module/page_alloc/parameters/shuffle.
3885 page_owner= [KNL] Boot-time page_owner enabling option.
3886 Storage of the information about who allocated
3887 each page is disabled in default. With this switch,
3889 on: enable the feature
3891 page_poison= [KNL] Boot-time parameter changing the state of
3892 poisoning on the buddy allocator, available with
3893 CONFIG_PAGE_POISONING=y.
3894 off: turn off poisoning (default)
3895 on: turn on poisoning
3897 page_reporting.page_reporting_order=
3898 [KNL] Minimal page reporting order
3900 Adjust the minimal page reporting order. The page
3901 reporting is disabled when it exceeds (MAX_ORDER-1).
3903 panic= [KNL] Kernel behaviour on panic: delay <timeout>
3904 timeout > 0: seconds before rebooting
3905 timeout = 0: wait forever
3906 timeout < 0: reboot immediately
3909 panic_print= Bitmask for printing system info when panic happens.
3910 User can chose combination of the following bits:
3911 bit 0: print all tasks info
3912 bit 1: print system memory info
3913 bit 2: print timer info
3914 bit 3: print locks info if CONFIG_LOCKDEP is on
3915 bit 4: print ftrace buffer
3916 bit 5: print all printk messages in buffer
3917 bit 6: print all CPUs backtrace (if available in the arch)
3918 *Be aware* that this option may print a _lot_ of lines,
3919 so there are risks of losing older messages in the log.
3920 Use this option carefully, maybe worth to setup a
3921 bigger log buffer with "log_buf_len" along with this.
3923 panic_on_taint= Bitmask for conditionally calling panic() in add_taint()
3924 Format: <hex>[,nousertaint]
3925 Hexadecimal bitmask representing the set of TAINT flags
3926 that will cause the kernel to panic when add_taint() is
3927 called with any of the flags in this set.
3928 The optional switch "nousertaint" can be utilized to
3929 prevent userspace forced crashes by writing to sysctl
3930 /proc/sys/kernel/tainted any flagset matching with the
3931 bitmask set on panic_on_taint.
3932 See Documentation/admin-guide/tainted-kernels.rst for
3933 extra details on the taint flags that users can pick
3934 to compose the bitmask to assign to panic_on_taint.
3936 panic_on_warn panic() instead of WARN(). Useful to cause kdump
3939 parkbd.port= [HW] Parallel port number the keyboard adapter is
3940 connected to, default is 0.
3942 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation,
3943 0 for XT, 1 for AT (default is AT).
3946 parport= [HW,PPT] Specify parallel ports. 0 disables.
3947 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] }
3948 Use 'auto' to force the driver to use any
3949 IRQ/DMA settings detected (the default is to
3950 ignore detected IRQ/DMA settings because of
3951 possible conflicts). You can specify the base
3952 address, IRQ, and DMA settings; IRQ and DMA
3953 should be numbers, or 'auto' (for using detected
3954 settings on that particular port), or 'nofifo'
3955 (to avoid using a FIFO even if it is detected).
3956 Parallel ports are assigned in the order they
3957 are specified on the command line, starting
3960 parport_init_mode= [HW,PPT]
3961 Configure VIA parallel port to operate in
3962 a specific mode. This is necessary on Pegasos
3963 computer where firmware has no options for setting
3964 up parallel port mode and sets it to spp.
3965 Currently this function knows 686a and 8231 chips.
3966 Format: [spp|ps2|epp|ecp|ecpepp]
3968 pata_legacy.all= [HW,LIBATA]
3970 Set to non-zero to probe primary and secondary ISA
3971 port ranges on PCI systems where no PCI PATA device
3972 has been found at either range. Disabled by default.
3974 pata_legacy.autospeed= [HW,LIBATA]
3976 Set to non-zero if a chip is present that snoops speed
3977 changes. Disabled by default.
3979 pata_legacy.ht6560a= [HW,LIBATA]
3981 Set to 1, 2, or 3 for HT 6560A on the primary channel,
3982 the secondary channel, or both channels respectively.
3983 Disabled by default.
3985 pata_legacy.ht6560b= [HW,LIBATA]
3987 Set to 1, 2, or 3 for HT 6560B on the primary channel,
3988 the secondary channel, or both channels respectively.
3989 Disabled by default.
3991 pata_legacy.iordy_mask= [HW,LIBATA]
3993 IORDY enable mask. Set individual bits to allow IORDY
3994 for the respective channel. Bit 0 is for the first
3995 legacy channel handled by this driver, bit 1 is for
3996 the second channel, and so on. The sequence will often
3997 correspond to the primary legacy channel, the secondary
3998 legacy channel, and so on, but the handling of a PCI
3999 bus and the use of other driver options may interfere
4000 with the sequence. By default IORDY is allowed across
4003 pata_legacy.opti82c46x= [HW,LIBATA]
4005 Set to 1, 2, or 3 for Opti 82c611A on the primary
4006 channel, the secondary channel, or both channels
4007 respectively. Disabled by default.
4009 pata_legacy.opti82c611a= [HW,LIBATA]
4011 Set to 1, 2, or 3 for Opti 82c465MV on the primary
4012 channel, the secondary channel, or both channels
4013 respectively. Disabled by default.
4015 pata_legacy.pio_mask= [HW,LIBATA]
4017 PIO mode mask for autospeed devices. Set individual
4018 bits to allow the use of the respective PIO modes.
4019 Bit 0 is for mode 0, bit 1 is for mode 1, and so on.
4020 All modes allowed by default.
4022 pata_legacy.probe_all= [HW,LIBATA]
4024 Set to non-zero to probe tertiary and further ISA
4025 port ranges on PCI systems. Disabled by default.
4027 pata_legacy.probe_mask= [HW,LIBATA]
4029 Probe mask for legacy ISA PATA ports. Depending on
4030 platform configuration and the use of other driver
4031 options up to 6 legacy ports are supported: 0x1f0,
4032 0x170, 0x1e8, 0x168, 0x1e0, 0x160, however probing
4033 of individual ports can be disabled by setting the
4034 corresponding bits in the mask to 1. Bit 0 is for
4035 the first port in the list above (0x1f0), and so on.
4036 By default all supported ports are probed.
4038 pata_legacy.qdi= [HW,LIBATA]
4040 Set to non-zero to probe QDI controllers. By default
4041 set to 1 if CONFIG_PATA_QDI_MODULE, 0 otherwise.
4043 pata_legacy.winbond= [HW,LIBATA]
4045 Set to non-zero to probe Winbond controllers. Use
4046 the standard I/O port (0x130) if 1, otherwise the
4047 value given is the I/O port to use (typically 0x1b0).
4048 By default set to 1 if CONFIG_PATA_WINBOND_VLB_MODULE,
4051 pata_platform.pio_mask= [HW,LIBATA]
4053 Supported PIO mode mask. Set individual bits to allow
4054 the use of the respective PIO modes. Bit 0 is for
4055 mode 0, bit 1 is for mode 1, and so on. Mode 0 only
4059 Halt all CPUs after the first oops has been printed for
4060 the specified number of seconds. This is to be used if
4061 your oopses keep scrolling off the screen.
4066 See header of drivers/block/paride/pcd.c.
4067 See also Documentation/admin-guide/blockdev/paride.rst.
4069 pci=option[,option...] [PCI] various PCI subsystem options.
4071 Some options herein operate on a specific device
4072 or a set of devices (<pci_dev>). These are
4073 specified in one of the following formats:
4075 [<domain>:]<bus>:<dev>.<func>[/<dev>.<func>]*
4076 pci:<vendor>:<device>[:<subvendor>:<subdevice>]
4078 Note: the first format specifies a PCI
4079 bus/device/function address which may change
4080 if new hardware is inserted, if motherboard
4081 firmware changes, or due to changes caused
4082 by other kernel parameters. If the
4083 domain is left unspecified, it is
4084 taken to be zero. Optionally, a path
4085 to a device through multiple device/function
4086 addresses can be specified after the base
4087 address (this is more robust against
4088 renumbering issues). The second format
4089 selects devices using IDs from the
4090 configuration space which may match multiple
4091 devices in the system.
4093 earlydump dump PCI config space before the kernel
4095 off [X86] don't probe for the PCI bus
4096 bios [X86-32] force use of PCI BIOS, don't access
4097 the hardware directly. Use this if your machine
4098 has a non-standard PCI host bridge.
4099 nobios [X86-32] disallow use of PCI BIOS, only direct
4100 hardware access methods are allowed. Use this
4101 if you experience crashes upon bootup and you
4102 suspect they are caused by the BIOS.
4103 conf1 [X86] Force use of PCI Configuration Access
4104 Mechanism 1 (config address in IO port 0xCF8,
4105 data in IO port 0xCFC, both 32-bit).
4106 conf2 [X86] Force use of PCI Configuration Access
4107 Mechanism 2 (IO port 0xCF8 is an 8-bit port for
4108 the function, IO port 0xCFA, also 8-bit, sets
4109 bus number. The config space is then accessed
4110 through ports 0xC000-0xCFFF).
4111 See http://wiki.osdev.org/PCI for more info
4112 on the configuration access mechanisms.
4113 noaer [PCIE] If the PCIEAER kernel config parameter is
4114 enabled, this kernel boot option can be used to
4115 disable the use of PCIE advanced error reporting.
4116 nodomains [PCI] Disable support for multiple PCI
4117 root domains (aka PCI segments, in ACPI-speak).
4118 nommconf [X86] Disable use of MMCONFIG for PCI
4120 check_enable_amd_mmconf [X86] check for and enable
4121 properly configured MMIO access to PCI
4122 config space on AMD family 10h CPU
4123 nomsi [MSI] If the PCI_MSI kernel config parameter is
4124 enabled, this kernel boot option can be used to
4125 disable the use of MSI interrupts system-wide.
4126 noioapicquirk [APIC] Disable all boot interrupt quirks.
4127 Safety option to keep boot IRQs enabled. This
4128 should never be necessary.
4129 ioapicreroute [APIC] Enable rerouting of boot IRQs to the
4130 primary IO-APIC for bridges that cannot disable
4131 boot IRQs. This fixes a source of spurious IRQs
4132 when the system masks IRQs.
4133 noioapicreroute [APIC] Disable workaround that uses the
4134 boot IRQ equivalent of an IRQ that connects to
4135 a chipset where boot IRQs cannot be disabled.
4136 The opposite of ioapicreroute.
4137 biosirq [X86-32] Use PCI BIOS calls to get the interrupt
4138 routing table. These calls are known to be buggy
4139 on several machines and they hang the machine
4140 when used, but on other computers it's the only
4141 way to get the interrupt routing table. Try
4142 this option if the kernel is unable to allocate
4143 IRQs or discover secondary PCI buses on your
4145 rom [X86] Assign address space to expansion ROMs.
4146 Use with caution as certain devices share
4147 address decoders between ROMs and other
4149 norom [X86] Do not assign address space to
4150 expansion ROMs that do not already have
4151 BIOS assigned address ranges.
4152 nobar [X86] Do not assign address space to the
4153 BARs that weren't assigned by the BIOS.
4154 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be
4155 assigned automatically to PCI devices. You can
4156 make the kernel exclude IRQs of your ISA cards
4158 pirqaddr=0xAAAAA [X86] Specify the physical address
4159 of the PIRQ table (normally generated
4160 by the BIOS) if it is outside the
4161 F0000h-100000h range.
4162 lastbus=N [X86] Scan all buses thru bus #N. Can be
4163 useful if the kernel is unable to find your
4164 secondary buses and you want to tell it
4165 explicitly which ones they are.
4166 assign-busses [X86] Always assign all PCI bus
4167 numbers ourselves, overriding
4168 whatever the firmware may have done.
4169 usepirqmask [X86] Honor the possible IRQ mask stored
4170 in the BIOS $PIR table. This is needed on
4171 some systems with broken BIOSes, notably
4172 some HP Pavilion N5400 and Omnibook XE3
4173 notebooks. This will have no effect if ACPI
4174 IRQ routing is enabled.
4175 noacpi [X86] Do not use ACPI for IRQ routing
4176 or for PCI scanning.
4177 use_crs [X86] Use PCI host bridge window information
4178 from ACPI. On BIOSes from 2008 or later, this
4179 is enabled by default. If you need to use this,
4180 please report a bug.
4181 nocrs [X86] Ignore PCI host bridge windows from ACPI.
4182 If you need to use this, please report a bug.
4183 use_e820 [X86] Use E820 reservations to exclude parts of
4184 PCI host bridge windows. This is a workaround
4185 for BIOS defects in host bridge _CRS methods.
4186 If you need to use this, please report a bug to
4187 <linux-pci@vger.kernel.org>.
4188 no_e820 [X86] Ignore E820 reservations for PCI host
4189 bridge windows. This is the default on modern
4190 hardware. If you need to use this, please report
4191 a bug to <linux-pci@vger.kernel.org>.
4192 routeirq Do IRQ routing for all PCI devices.
4193 This is normally done in pci_enable_device(),
4194 so this option is a temporary workaround
4195 for broken drivers that don't call it.
4196 skip_isa_align [X86] do not align io start addr, so can
4197 handle more pci cards
4198 noearly [X86] Don't do any early type 1 scanning.
4199 This might help on some broken boards which
4200 machine check when some devices' config space
4201 is read. But various workarounds are disabled
4202 and some IOMMU drivers will not work.
4203 bfsort Sort PCI devices into breadth-first order.
4204 This sorting is done to get a device
4205 order compatible with older (<= 2.4) kernels.
4206 nobfsort Don't sort PCI devices into breadth-first order.
4207 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size)
4208 tuning and use the BIOS-configured MPS defaults.
4209 pcie_bus_safe Set every device's MPS to the largest value
4210 supported by all devices below the root complex.
4211 pcie_bus_perf Set device MPS to the largest allowable MPS
4212 based on its parent bus. Also set MRRS (Max
4213 Read Request Size) to the largest supported
4214 value (no larger than the MPS that the device
4215 or bus can support) for best performance.
4216 pcie_bus_peer2peer Set every device's MPS to 128B, which
4217 every device is guaranteed to support. This
4218 configuration allows peer-to-peer DMA between
4219 any pair of devices, possibly at the cost of
4220 reduced performance. This also guarantees
4221 that hot-added devices will work.
4222 cbiosize=nn[KMG] The fixed amount of bus space which is
4223 reserved for the CardBus bridge's IO window.
4224 The default value is 256 bytes.
4225 cbmemsize=nn[KMG] The fixed amount of bus space which is
4226 reserved for the CardBus bridge's memory
4227 window. The default value is 64 megabytes.
4230 [<order of align>@]<pci_dev>[; ...]
4231 Specifies alignment and device to reassign
4232 aligned memory resources. How to
4233 specify the device is described above.
4234 If <order of align> is not specified,
4235 PAGE_SIZE is used as alignment.
4236 A PCI-PCI bridge can be specified if resource
4237 windows need to be expanded.
4238 To specify the alignment for several
4239 instances of a device, the PCI vendor,
4240 device, subvendor, and subdevice may be
4241 specified, e.g., 12@pci:8086:9c22:103c:198f
4242 for 4096-byte alignment.
4243 ecrc= Enable/disable PCIe ECRC (transaction layer
4244 end-to-end CRC checking).
4245 bios: Use BIOS/firmware settings. This is the
4249 hpiosize=nn[KMG] The fixed amount of bus space which is
4250 reserved for hotplug bridge's IO window.
4251 Default size is 256 bytes.
4252 hpmmiosize=nn[KMG] The fixed amount of bus space which is
4253 reserved for hotplug bridge's MMIO window.
4254 Default size is 2 megabytes.
4255 hpmmioprefsize=nn[KMG] The fixed amount of bus space which is
4256 reserved for hotplug bridge's MMIO_PREF window.
4257 Default size is 2 megabytes.
4258 hpmemsize=nn[KMG] The fixed amount of bus space which is
4259 reserved for hotplug bridge's MMIO and
4261 Default size is 2 megabytes.
4262 hpbussize=nn The minimum amount of additional bus numbers
4263 reserved for buses below a hotplug bridge.
4265 realloc= Enable/disable reallocating PCI bridge resources
4266 if allocations done by BIOS are too small to
4267 accommodate resources required by all child
4269 off: Turn realloc off
4271 realloc same as realloc=on
4272 noari do not use PCIe ARI.
4273 noats [PCIE, Intel-IOMMU, AMD-IOMMU]
4274 do not use PCIe ATS (and IOMMU device IOTLB).
4275 pcie_scan_all Scan all possible PCIe devices. Otherwise we
4276 only look for one device below a PCIe downstream
4278 big_root_window Try to add a big 64bit memory window to the PCIe
4279 root complex on AMD CPUs. Some GFX hardware
4280 can resize a BAR to allow access to all VRAM.
4281 Adding the window is slightly risky (it may
4282 conflict with unreported devices), so this
4284 disable_acs_redir=<pci_dev>[; ...]
4285 Specify one or more PCI devices (in the format
4286 specified above) separated by semicolons.
4287 Each device specified will have the PCI ACS
4288 redirect capabilities forced off which will
4289 allow P2P traffic between devices through
4290 bridges without forcing it upstream. Note:
4291 this removes isolation between devices and
4292 may put more devices in an IOMMU group.
4293 force_floating [S390] Force usage of floating interrupts.
4294 nomio [S390] Do not use MIO instructions.
4295 norid [S390] ignore the RID field and force use of
4296 one PCI domain per PCI function
4298 pcie_aspm= [PCIE] Forcibly enable or disable PCIe Active State Power
4301 force Enable ASPM even on devices that claim not to support it.
4302 WARNING: Forcing ASPM on may cause system lockups.
4304 pcie_ports= [PCIE] PCIe port services handling:
4305 native Use native PCIe services (PME, AER, DPC, PCIe hotplug)
4306 even if the platform doesn't give the OS permission to
4307 use them. This may cause conflicts if the platform
4308 also tries to use these services.
4309 dpc-native Use native PCIe service for DPC only. May
4310 cause conflicts if firmware uses AER or DPC.
4311 compat Disable native PCIe services (PME, AER, DPC, PCIe
4314 pcie_port_pm= [PCIE] PCIe port power management handling:
4315 off Disable power management of all PCIe ports
4316 force Forcibly enable power management of all PCIe ports
4318 pcie_pme= [PCIE,PM] Native PCIe PME signaling options:
4319 nomsi Do not use MSI for native PCIe PME signaling (this makes
4320 all PCIe root ports use INTx for all services).
4322 pcmv= [HW,PCMCIA] BadgePAD 4
4326 Keep all power-domains already enabled by bootloader on,
4327 even if no driver has claimed them. This is useful
4328 for debug and development, but should not be
4329 needed on a platform with proper driver support.
4332 See Documentation/admin-guide/blockdev/paride.rst.
4334 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at
4337 See arch/parisc/kernel/pdc_chassis.c
4339 percpu_alloc= Select which percpu first chunk allocator to use.
4340 Currently supported values are "embed" and "page".
4341 Archs may support subset or none of the selections.
4342 See comments in mm/percpu.c for details on each
4343 allocator. This parameter is primarily for debugging
4344 and performance comparison.
4347 See Documentation/admin-guide/blockdev/paride.rst.
4350 See Documentation/admin-guide/blockdev/paride.rst.
4352 pirq= [SMP,APIC] Manual mp-table setup
4353 See Documentation/x86/i386/IO-APIC.rst.
4355 plip= [PPT,NET] Parallel port network link
4356 Format: { parport<nr> | timid | 0 }
4357 See also Documentation/admin-guide/parport.rst.
4359 pmtmr= [X86] Manual setup of pmtmr I/O Port.
4360 Override pmtimer IOPort with a hex value.
4363 pmu_override= [PPC] Override the PMU.
4364 This option takes over the PMU facility, so it is no
4365 longer usable by perf. Setting this option starts the
4366 PMU counters by setting MMCR0 to 0 (the FC bit is
4367 cleared). If a number is given, then MMCR1 is set to
4368 that number, otherwise (e.g., 'pmu_override=on'), MMCR1
4371 pm_debug_messages [SUSPEND,KNL]
4372 Enable suspend/resume debug messages during boot up.
4375 Enable PNP debug messages (depends on the
4376 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time
4377 via /sys/module/pnp/parameters/debug. We always show
4378 current resource usage; turning this on also shows
4379 possible settings and some assignment information.
4385 { on | off | curr | res | no-curr | no-res }
4388 [ISAPNP] Exclude IRQs for the autoconfiguration
4391 [ISAPNP] Exclude DMAs for the autoconfiguration
4393 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration
4394 Ranges are in pairs (I/O port base and size).
4397 [ISAPNP] Exclude memory regions for the
4399 Ranges are in pairs (memory base and size).
4401 ports= [IP_VS_FTP] IPVS ftp helper module
4403 Up to 8 (IP_VS_APP_MAX_PORTS) ports
4405 Format: <port>,<port>....
4407 powersave=off [PPC] This option disables power saving features.
4408 It specifically disables cpuidle and sets the
4409 platform machine description specific power_save
4410 function to NULL. On Idle the CPU just reduces
4413 ppc_strict_facility_enable
4414 [PPC] This option catches any kernel floating point,
4415 Altivec, VSX and SPE outside of regions specifically
4416 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()).
4417 There is some performance impact when enabling this.
4421 Disable Hardware Transactional Memory
4424 Select preemption mode if you have CONFIG_PREEMPT_DYNAMIC
4425 none - Limited to cond_resched() calls
4426 voluntary - Limited to cond_resched() and might_sleep() calls
4427 full - Any section that isn't explicitly preempt disabled
4428 can be preempted anytime.
4430 print-fatal-signals=
4431 [KNL] debug: print fatal signals
4433 If enabled, warn about various signal handling
4434 related application anomalies: too many signals,
4435 too many POSIX.1 timers, fatal signals causing a
4438 If you hit the warning due to signal overflow,
4439 you might want to try "ulimit -i unlimited".
4443 printk.always_kmsg_dump=
4444 Trigger kmsg_dump for cases other than kernel oops or
4446 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
4449 printk.console_no_auto_verbose=
4450 Disable console loglevel raise on oops, panic
4451 or lockdep-detected issues (only if lock debug is on).
4452 With an exception to setups with low baudrate on
4453 serial console, keeping this 0 is a good choice
4454 in order to provide more debug information.
4456 default: 0 (auto_verbose is enabled)
4458 printk.devkmsg={on,off,ratelimit}
4459 Control writing to /dev/kmsg.
4460 on - unlimited logging to /dev/kmsg from userspace
4461 off - logging to /dev/kmsg disabled
4462 ratelimit - ratelimit the logging
4465 printk.time= Show timing data prefixed to each printk message line
4466 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
4468 processor.max_cstate= [HW,ACPI]
4469 Limit processor to maximum C-state
4470 max_cstate=9 overrides any DMI blacklist limit.
4472 processor.nocst [HW,ACPI]
4473 Ignore the _CST method to determine C-states,
4474 instead using the legacy FADT method
4476 profile= [KNL] Enable kernel profiling via /proc/profile
4477 Format: [<profiletype>,]<number>
4478 Param: <profiletype>: "schedule", "sleep", or "kvm"
4479 [defaults to kernel profiling]
4480 Param: "schedule" - profile schedule points.
4481 Param: "sleep" - profile D-state sleeping (millisecs).
4482 Requires CONFIG_SCHEDSTATS
4483 Param: "kvm" - profile VM exits.
4484 Param: <number> - step/bucket size as a power of 2 for
4485 statistical time based profiling.
4487 prompt_ramdisk= [RAM] [Deprecated]
4489 prot_virt= [S390] enable hosting protected virtual machines
4490 isolated from the hypervisor (if hardware supports
4494 psi= [KNL] Enable or disable pressure stall information
4498 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to
4499 probe for; one of (bare|imps|exps|lifebook|any).
4500 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports
4502 psmouse.resetafter= [HW,MOUSE]
4503 Try to reset the device after so many bad packets
4506 [HW,MOUSE] Set desired mouse resolution, in dpi.
4507 psmouse.smartscroll=
4508 [HW,MOUSE] Controls Logitech smartscroll autorepeat.
4509 0 = disabled, 1 = enabled (default).
4511 pstore.backend= Specify the name of the pstore backend to use
4514 See Documentation/admin-guide/blockdev/paride.rst.
4516 pti= [X86-64] Control Page Table Isolation of user and
4517 kernel address spaces. Disabling this feature
4518 removes hardening, but improves performance of
4519 system calls and interrupts.
4521 on - unconditionally enable
4522 off - unconditionally disable
4523 auto - kernel detects whether your CPU model is
4524 vulnerable to issues that PTI mitigates
4526 Not specifying this option is equivalent to pti=auto.
4529 Equivalent to pti=off
4532 [KNL] Number of legacy pty's. Overwrites compiled-in
4535 quiet [KNL] Disable most log messages
4540 See Documentation/admin-guide/md.rst.
4542 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes
4543 See Documentation/admin-guide/blockdev/ramdisk.rst.
4545 ramdisk_start= [RAM] RAM disk image start address
4547 random.trust_cpu={on,off}
4548 [KNL] Enable or disable trusting the use of the
4549 CPU's random number generator (if available) to
4550 fully seed the kernel's CRNG. Default is controlled
4551 by CONFIG_RANDOM_TRUST_CPU.
4553 random.trust_bootloader={on,off}
4554 [KNL] Enable or disable trusting the use of a
4555 seed passed by the bootloader (if available) to
4556 fully seed the kernel's CRNG. Default is controlled
4557 by CONFIG_RANDOM_TRUST_BOOTLOADER.
4559 randomize_kstack_offset=
4560 [KNL] Enable or disable kernel stack offset
4561 randomization, which provides roughly 5 bits of
4562 entropy, frustrating memory corruption attacks
4563 that depend on stack address determinism or
4564 cross-syscall address exposures. This is only
4565 available on architectures that have defined
4566 CONFIG_HAVE_ARCH_RANDOMIZE_KSTACK_OFFSET.
4567 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
4568 Default is CONFIG_RANDOMIZE_KSTACK_OFFSET_DEFAULT.
4570 ras=option[,option,...] [KNL] RAS-specific options
4573 Disable the Correctable Errors Collector,
4574 see CONFIG_RAS_CEC help text.
4576 rcu_nocbs[=cpu-list]
4577 [KNL] The optional argument is a cpu list,
4580 In kernels built with CONFIG_RCU_NOCB_CPU=y,
4581 enable the no-callback CPU mode, which prevents
4582 such CPUs' callbacks from being invoked in
4583 softirq context. Invocation of such CPUs' RCU
4584 callbacks will instead be offloaded to "rcuox/N"
4585 kthreads created for that purpose, where "x" is
4586 "p" for RCU-preempt, "s" for RCU-sched, and "g"
4587 for the kthreads that mediate grace periods; and
4588 "N" is the CPU number. This reduces OS jitter on
4589 the offloaded CPUs, which can be useful for HPC
4590 and real-time workloads. It can also improve
4591 energy efficiency for asymmetric multiprocessors.
4593 If a cpulist is passed as an argument, the specified
4594 list of CPUs is set to no-callback mode from boot.
4596 Otherwise, if the '=' sign and the cpulist
4597 arguments are omitted, no CPU will be set to
4598 no-callback mode from boot but the mode may be
4599 toggled at runtime via cpusets.
4601 Note that this argument takes precedence over
4602 the CONFIG_RCU_NOCB_CPU_DEFAULT_ALL option.
4605 Rather than requiring that offloaded CPUs
4606 (specified by rcu_nocbs= above) explicitly
4607 awaken the corresponding "rcuoN" kthreads,
4608 make these kthreads poll for callbacks.
4609 This improves the real-time response for the
4610 offloaded CPUs by relieving them of the need to
4611 wake up the corresponding kthread, but degrades
4612 energy efficiency by requiring that the kthreads
4613 periodically wake up to do the polling.
4615 rcutree.blimit= [KNL]
4616 Set maximum number of finished RCU callbacks to
4617 process in one batch.
4619 rcutree.dump_tree= [KNL]
4620 Dump the structure of the rcu_node combining tree
4621 out at early boot. This is used for diagnostic
4622 purposes, to verify correct tree setup.
4624 rcutree.gp_cleanup_delay= [KNL]
4625 Set the number of jiffies to delay each step of
4626 RCU grace-period cleanup.
4628 rcutree.gp_init_delay= [KNL]
4629 Set the number of jiffies to delay each step of
4630 RCU grace-period initialization.
4632 rcutree.gp_preinit_delay= [KNL]
4633 Set the number of jiffies to delay each step of
4634 RCU grace-period pre-initialization, that is,
4635 the propagation of recent CPU-hotplug changes up
4636 the rcu_node combining tree.
4638 rcutree.use_softirq= [KNL]
4639 If set to zero, move all RCU_SOFTIRQ processing to
4640 per-CPU rcuc kthreads. Defaults to a non-zero
4641 value, meaning that RCU_SOFTIRQ is used by default.
4642 Specify rcutree.use_softirq=0 to use rcuc kthreads.
4644 But note that CONFIG_PREEMPT_RT=y kernels disable
4645 this kernel boot parameter, forcibly setting it
4648 rcutree.rcu_fanout_exact= [KNL]
4649 Disable autobalancing of the rcu_node combining
4650 tree. This is used by rcutorture, and might
4651 possibly be useful for architectures having high
4652 cache-to-cache transfer latencies.
4654 rcutree.rcu_fanout_leaf= [KNL]
4655 Change the number of CPUs assigned to each
4656 leaf rcu_node structure. Useful for very
4657 large systems, which will choose the value 64,
4658 and for NUMA systems with large remote-access
4659 latencies, which will choose a value aligned
4660 with the appropriate hardware boundaries.
4662 rcutree.rcu_min_cached_objs= [KNL]
4663 Minimum number of objects which are cached and
4664 maintained per one CPU. Object size is equal
4665 to PAGE_SIZE. The cache allows to reduce the
4666 pressure to page allocator, also it makes the
4667 whole algorithm to behave better in low memory
4670 rcutree.rcu_delay_page_cache_fill_msec= [KNL]
4671 Set the page-cache refill delay (in milliseconds)
4672 in response to low-memory conditions. The range
4673 of permitted values is in the range 0:100000.
4675 rcutree.jiffies_till_first_fqs= [KNL]
4676 Set delay from grace-period initialization to
4677 first attempt to force quiescent states.
4678 Units are jiffies, minimum value is zero,
4679 and maximum value is HZ.
4681 rcutree.jiffies_till_next_fqs= [KNL]
4682 Set delay between subsequent attempts to force
4683 quiescent states. Units are jiffies, minimum
4684 value is one, and maximum value is HZ.
4686 rcutree.jiffies_till_sched_qs= [KNL]
4687 Set required age in jiffies for a
4688 given grace period before RCU starts
4689 soliciting quiescent-state help from
4690 rcu_note_context_switch() and cond_resched().
4691 If not specified, the kernel will calculate
4692 a value based on the most recent settings
4693 of rcutree.jiffies_till_first_fqs
4694 and rcutree.jiffies_till_next_fqs.
4695 This calculated value may be viewed in
4696 rcutree.jiffies_to_sched_qs. Any attempt to set
4697 rcutree.jiffies_to_sched_qs will be cheerfully
4700 rcutree.kthread_prio= [KNL,BOOT]
4701 Set the SCHED_FIFO priority of the RCU per-CPU
4702 kthreads (rcuc/N). This value is also used for
4703 the priority of the RCU boost threads (rcub/N)
4704 and for the RCU grace-period kthreads (rcu_bh,
4705 rcu_preempt, and rcu_sched). If RCU_BOOST is
4706 set, valid values are 1-99 and the default is 1
4707 (the least-favored priority). Otherwise, when
4708 RCU_BOOST is not set, valid values are 0-99 and
4709 the default is zero (non-realtime operation).
4710 When RCU_NOCB_CPU is set, also adjust the
4711 priority of NOCB callback kthreads.
4713 rcutree.rcu_divisor= [KNL]
4714 Set the shift-right count to use to compute
4715 the callback-invocation batch limit bl from
4716 the number of callbacks queued on this CPU.
4717 The result will be bounded below by the value of
4718 the rcutree.blimit kernel parameter. Every bl
4719 callbacks, the softirq handler will exit in
4720 order to allow the CPU to do other work.
4722 Please note that this callback-invocation batch
4723 limit applies only to non-offloaded callback
4724 invocation. Offloaded callbacks are instead
4725 invoked in the context of an rcuoc kthread, which
4726 scheduler will preempt as it does any other task.
4728 rcutree.nocb_nobypass_lim_per_jiffy= [KNL]
4729 On callback-offloaded (rcu_nocbs) CPUs,
4730 RCU reduces the lock contention that would
4731 otherwise be caused by callback floods through
4732 use of the ->nocb_bypass list. However, in the
4733 common non-flooded case, RCU queues directly to
4734 the main ->cblist in order to avoid the extra
4735 overhead of the ->nocb_bypass list and its lock.
4736 But if there are too many callbacks queued during
4737 a single jiffy, RCU pre-queues the callbacks into
4738 the ->nocb_bypass queue. The definition of "too
4739 many" is supplied by this kernel boot parameter.
4741 rcutree.rcu_nocb_gp_stride= [KNL]
4742 Set the number of NOCB callback kthreads in
4743 each group, which defaults to the square root
4744 of the number of CPUs. Larger numbers reduce
4745 the wakeup overhead on the global grace-period
4746 kthread, but increases that same overhead on
4747 each group's NOCB grace-period kthread.
4749 rcutree.qhimark= [KNL]
4750 Set threshold of queued RCU callbacks beyond which
4751 batch limiting is disabled.
4753 rcutree.qlowmark= [KNL]
4754 Set threshold of queued RCU callbacks below which
4755 batch limiting is re-enabled.
4757 rcutree.qovld= [KNL]
4758 Set threshold of queued RCU callbacks beyond which
4759 RCU's force-quiescent-state scan will aggressively
4760 enlist help from cond_resched() and sched IPIs to
4761 help CPUs more quickly reach quiescent states.
4762 Set to less than zero to make this be set based
4763 on rcutree.qhimark at boot time and to zero to
4764 disable more aggressive help enlistment.
4766 rcutree.rcu_kick_kthreads= [KNL]
4767 Cause the grace-period kthread to get an extra
4768 wake_up() if it sleeps three times longer than
4769 it should at force-quiescent-state time.
4770 This wake_up() will be accompanied by a
4771 WARN_ONCE() splat and an ftrace_dump().
4773 rcutree.rcu_unlock_delay= [KNL]
4774 In CONFIG_RCU_STRICT_GRACE_PERIOD=y kernels,
4775 this specifies an rcu_read_unlock()-time delay
4776 in microseconds. This defaults to zero.
4777 Larger delays increase the probability of
4778 catching RCU pointer leaks, that is, buggy use
4779 of RCU-protected pointers after the relevant
4780 rcu_read_unlock() has completed.
4782 rcutree.sysrq_rcu= [KNL]
4783 Commandeer a sysrq key to dump out Tree RCU's
4784 rcu_node tree with an eye towards determining
4785 why a new grace period has not yet started.
4787 rcuscale.gp_async= [KNL]
4788 Measure performance of asynchronous
4789 grace-period primitives such as call_rcu().
4791 rcuscale.gp_async_max= [KNL]
4792 Specify the maximum number of outstanding
4793 callbacks per writer thread. When a writer
4794 thread exceeds this limit, it invokes the
4795 corresponding flavor of rcu_barrier() to allow
4796 previously posted callbacks to drain.
4798 rcuscale.gp_exp= [KNL]
4799 Measure performance of expedited synchronous
4800 grace-period primitives.
4802 rcuscale.holdoff= [KNL]
4803 Set test-start holdoff period. The purpose of
4804 this parameter is to delay the start of the
4805 test until boot completes in order to avoid
4808 rcuscale.kfree_rcu_test= [KNL]
4809 Set to measure performance of kfree_rcu() flooding.
4811 rcuscale.kfree_rcu_test_double= [KNL]
4812 Test the double-argument variant of kfree_rcu().
4813 If this parameter has the same value as
4814 rcuscale.kfree_rcu_test_single, both the single-
4815 and double-argument variants are tested.
4817 rcuscale.kfree_rcu_test_single= [KNL]
4818 Test the single-argument variant of kfree_rcu().
4819 If this parameter has the same value as
4820 rcuscale.kfree_rcu_test_double, both the single-
4821 and double-argument variants are tested.
4823 rcuscale.kfree_nthreads= [KNL]
4824 The number of threads running loops of kfree_rcu().
4826 rcuscale.kfree_alloc_num= [KNL]
4827 Number of allocations and frees done in an iteration.
4829 rcuscale.kfree_loops= [KNL]
4830 Number of loops doing rcuscale.kfree_alloc_num number
4831 of allocations and frees.
4833 rcuscale.nreaders= [KNL]
4834 Set number of RCU readers. The value -1 selects
4835 N, where N is the number of CPUs. A value
4836 "n" less than -1 selects N-n+1, where N is again
4837 the number of CPUs. For example, -2 selects N
4838 (the number of CPUs), -3 selects N+1, and so on.
4839 A value of "n" less than or equal to -N selects
4842 rcuscale.nwriters= [KNL]
4843 Set number of RCU writers. The values operate
4844 the same as for rcuscale.nreaders.
4845 N, where N is the number of CPUs
4847 rcuscale.perf_type= [KNL]
4848 Specify the RCU implementation to test.
4850 rcuscale.shutdown= [KNL]
4851 Shut the system down after performance tests
4852 complete. This is useful for hands-off automated
4855 rcuscale.verbose= [KNL]
4856 Enable additional printk() statements.
4858 rcuscale.writer_holdoff= [KNL]
4859 Write-side holdoff between grace periods,
4860 in microseconds. The default of zero says
4863 rcutorture.fqs_duration= [KNL]
4864 Set duration of force_quiescent_state bursts
4867 rcutorture.fqs_holdoff= [KNL]
4868 Set holdoff time within force_quiescent_state bursts
4871 rcutorture.fqs_stutter= [KNL]
4872 Set wait time between force_quiescent_state bursts
4875 rcutorture.fwd_progress= [KNL]
4876 Specifies the number of kthreads to be used
4877 for RCU grace-period forward-progress testing
4878 for the types of RCU supporting this notion.
4879 Defaults to 1 kthread, values less than zero or
4880 greater than the number of CPUs cause the number
4883 rcutorture.fwd_progress_div= [KNL]
4884 Specify the fraction of a CPU-stall-warning
4885 period to do tight-loop forward-progress testing.
4887 rcutorture.fwd_progress_holdoff= [KNL]
4888 Number of seconds to wait between successive
4889 forward-progress tests.
4891 rcutorture.fwd_progress_need_resched= [KNL]
4892 Enclose cond_resched() calls within checks for
4893 need_resched() during tight-loop forward-progress
4896 rcutorture.gp_cond= [KNL]
4897 Use conditional/asynchronous update-side
4898 primitives, if available.
4900 rcutorture.gp_exp= [KNL]
4901 Use expedited update-side primitives, if available.
4903 rcutorture.gp_normal= [KNL]
4904 Use normal (non-expedited) asynchronous
4905 update-side primitives, if available.
4907 rcutorture.gp_sync= [KNL]
4908 Use normal (non-expedited) synchronous
4909 update-side primitives, if available. If all
4910 of rcutorture.gp_cond=, rcutorture.gp_exp=,
4911 rcutorture.gp_normal=, and rcutorture.gp_sync=
4912 are zero, rcutorture acts as if is interpreted
4913 they are all non-zero.
4915 rcutorture.irqreader= [KNL]
4916 Run RCU readers from irq handlers, or, more
4917 accurately, from a timer handler. Not all RCU
4918 flavors take kindly to this sort of thing.
4920 rcutorture.leakpointer= [KNL]
4921 Leak an RCU-protected pointer out of the reader.
4922 This can of course result in splats, and is
4923 intended to test the ability of things like
4924 CONFIG_RCU_STRICT_GRACE_PERIOD=y to detect
4927 rcutorture.n_barrier_cbs= [KNL]
4928 Set callbacks/threads for rcu_barrier() testing.
4930 rcutorture.nfakewriters= [KNL]
4931 Set number of concurrent RCU writers. These just
4932 stress RCU, they don't participate in the actual
4933 test, hence the "fake".
4935 rcutorture.nocbs_nthreads= [KNL]
4936 Set number of RCU callback-offload togglers.
4937 Zero (the default) disables toggling.
4939 rcutorture.nocbs_toggle= [KNL]
4940 Set the delay in milliseconds between successive
4941 callback-offload toggling attempts.
4943 rcutorture.nreaders= [KNL]
4944 Set number of RCU readers. The value -1 selects
4945 N-1, where N is the number of CPUs. A value
4946 "n" less than -1 selects N-n-2, where N is again
4947 the number of CPUs. For example, -2 selects N
4948 (the number of CPUs), -3 selects N+1, and so on.
4950 rcutorture.object_debug= [KNL]
4951 Enable debug-object double-call_rcu() testing.
4953 rcutorture.onoff_holdoff= [KNL]
4954 Set time (s) after boot for CPU-hotplug testing.
4956 rcutorture.onoff_interval= [KNL]
4957 Set time (jiffies) between CPU-hotplug operations,
4958 or zero to disable CPU-hotplug testing.
4960 rcutorture.read_exit= [KNL]
4961 Set the number of read-then-exit kthreads used
4962 to test the interaction of RCU updaters and
4963 task-exit processing.
4965 rcutorture.read_exit_burst= [KNL]
4966 The number of times in a given read-then-exit
4967 episode that a set of read-then-exit kthreads
4970 rcutorture.read_exit_delay= [KNL]
4971 The delay, in seconds, between successive
4972 read-then-exit testing episodes.
4974 rcutorture.shuffle_interval= [KNL]
4975 Set task-shuffle interval (s). Shuffling tasks
4976 allows some CPUs to go into dyntick-idle mode
4977 during the rcutorture test.
4979 rcutorture.shutdown_secs= [KNL]
4980 Set time (s) after boot system shutdown. This
4981 is useful for hands-off automated testing.
4983 rcutorture.stall_cpu= [KNL]
4984 Duration of CPU stall (s) to test RCU CPU stall
4985 warnings, zero to disable.
4987 rcutorture.stall_cpu_block= [KNL]
4988 Sleep while stalling if set. This will result
4989 in warnings from preemptible RCU in addition
4990 to any other stall-related activity.
4992 rcutorture.stall_cpu_holdoff= [KNL]
4993 Time to wait (s) after boot before inducing stall.
4995 rcutorture.stall_cpu_irqsoff= [KNL]
4996 Disable interrupts while stalling if set.
4998 rcutorture.stall_gp_kthread= [KNL]
4999 Duration (s) of forced sleep within RCU
5000 grace-period kthread to test RCU CPU stall
5001 warnings, zero to disable. If both stall_cpu
5002 and stall_gp_kthread are specified, the
5003 kthread is starved first, then the CPU.
5005 rcutorture.stat_interval= [KNL]
5006 Time (s) between statistics printk()s.
5008 rcutorture.stutter= [KNL]
5009 Time (s) to stutter testing, for example, specifying
5010 five seconds causes the test to run for five seconds,
5011 wait for five seconds, and so on. This tests RCU's
5012 ability to transition abruptly to and from idle.
5014 rcutorture.test_boost= [KNL]
5015 Test RCU priority boosting? 0=no, 1=maybe, 2=yes.
5016 "Maybe" means test if the RCU implementation
5017 under test support RCU priority boosting.
5019 rcutorture.test_boost_duration= [KNL]
5020 Duration (s) of each individual boost test.
5022 rcutorture.test_boost_interval= [KNL]
5023 Interval (s) between each boost test.
5025 rcutorture.test_no_idle_hz= [KNL]
5026 Test RCU's dyntick-idle handling. See also the
5027 rcutorture.shuffle_interval parameter.
5029 rcutorture.torture_type= [KNL]
5030 Specify the RCU implementation to test.
5032 rcutorture.verbose= [KNL]
5033 Enable additional printk() statements.
5035 rcupdate.rcu_cpu_stall_ftrace_dump= [KNL]
5036 Dump ftrace buffer after reporting RCU CPU
5039 rcupdate.rcu_cpu_stall_suppress= [KNL]
5040 Suppress RCU CPU stall warning messages.
5042 rcupdate.rcu_cpu_stall_suppress_at_boot= [KNL]
5043 Suppress RCU CPU stall warning messages and
5044 rcutorture writer stall warnings that occur
5045 during early boot, that is, during the time
5046 before the init task is spawned.
5048 rcupdate.rcu_cpu_stall_timeout= [KNL]
5049 Set timeout for RCU CPU stall warning messages.
5050 The value is in seconds and the maximum allowed
5051 value is 300 seconds.
5053 rcupdate.rcu_exp_cpu_stall_timeout= [KNL]
5054 Set timeout for expedited RCU CPU stall warning
5055 messages. The value is in milliseconds
5056 and the maximum allowed value is 21000
5057 milliseconds. Please note that this value is
5058 adjusted to an arch timer tick resolution.
5059 Setting this to zero causes the value from
5060 rcupdate.rcu_cpu_stall_timeout to be used (after
5061 conversion from seconds to milliseconds).
5063 rcupdate.rcu_expedited= [KNL]
5064 Use expedited grace-period primitives, for
5065 example, synchronize_rcu_expedited() instead
5066 of synchronize_rcu(). This reduces latency,
5067 but can increase CPU utilization, degrade
5068 real-time latency, and degrade energy efficiency.
5069 No effect on CONFIG_TINY_RCU kernels.
5071 rcupdate.rcu_normal= [KNL]
5072 Use only normal grace-period primitives,
5073 for example, synchronize_rcu() instead of
5074 synchronize_rcu_expedited(). This improves
5075 real-time latency, CPU utilization, and
5076 energy efficiency, but can expose users to
5077 increased grace-period latency. This parameter
5078 overrides rcupdate.rcu_expedited. No effect on
5079 CONFIG_TINY_RCU kernels.
5081 rcupdate.rcu_normal_after_boot= [KNL]
5082 Once boot has completed (that is, after
5083 rcu_end_inkernel_boot() has been invoked), use
5084 only normal grace-period primitives. No effect
5085 on CONFIG_TINY_RCU kernels.
5087 But note that CONFIG_PREEMPT_RT=y kernels enables
5088 this kernel boot parameter, forcibly setting
5089 it to the value one, that is, converting any
5090 post-boot attempt at an expedited RCU grace
5091 period to instead use normal non-expedited
5092 grace-period processing.
5094 rcupdate.rcu_task_collapse_lim= [KNL]
5095 Set the maximum number of callbacks present
5096 at the beginning of a grace period that allows
5097 the RCU Tasks flavors to collapse back to using
5098 a single callback queue. This switching only
5099 occurs when rcupdate.rcu_task_enqueue_lim is
5100 set to the default value of -1.
5102 rcupdate.rcu_task_contend_lim= [KNL]
5103 Set the minimum number of callback-queuing-time
5104 lock-contention events per jiffy required to
5105 cause the RCU Tasks flavors to switch to per-CPU
5106 callback queuing. This switching only occurs
5107 when rcupdate.rcu_task_enqueue_lim is set to
5108 the default value of -1.
5110 rcupdate.rcu_task_enqueue_lim= [KNL]
5111 Set the number of callback queues to use for the
5112 RCU Tasks family of RCU flavors. The default
5113 of -1 allows this to be automatically (and
5114 dynamically) adjusted. This parameter is intended
5117 rcupdate.rcu_task_ipi_delay= [KNL]
5118 Set time in jiffies during which RCU tasks will
5119 avoid sending IPIs, starting with the beginning
5120 of a given grace period. Setting a large
5121 number avoids disturbing real-time workloads,
5122 but lengthens grace periods.
5124 rcupdate.rcu_task_stall_info= [KNL]
5125 Set initial timeout in jiffies for RCU task stall
5126 informational messages, which give some indication
5127 of the problem for those not patient enough to
5128 wait for ten minutes. Informational messages are
5129 only printed prior to the stall-warning message
5130 for a given grace period. Disable with a value
5131 less than or equal to zero. Defaults to ten
5132 seconds. A change in value does not take effect
5133 until the beginning of the next grace period.
5135 rcupdate.rcu_task_stall_info_mult= [KNL]
5136 Multiplier for time interval between successive
5137 RCU task stall informational messages for a given
5138 RCU tasks grace period. This value is clamped
5139 to one through ten, inclusive. It defaults to
5140 the value three, so that the first informational
5141 message is printed 10 seconds into the grace
5142 period, the second at 40 seconds, the third at
5143 160 seconds, and then the stall warning at 600
5144 seconds would prevent a fourth at 640 seconds.
5146 rcupdate.rcu_task_stall_timeout= [KNL]
5147 Set timeout in jiffies for RCU task stall
5148 warning messages. Disable with a value less
5149 than or equal to zero. Defaults to ten minutes.
5150 A change in value does not take effect until
5151 the beginning of the next grace period.
5153 rcupdate.rcu_self_test= [KNL]
5154 Run the RCU early boot self tests
5158 Run specified binary instead of /init from the ramdisk,
5159 used for early userspace startup. See initrd.
5162 force - Override the decision by the kernel to hide the
5163 advertisement of RDRAND support (this affects
5164 certain AMD processors because of buggy BIOS
5165 support, specifically around the suspend/resume
5169 Turn on/off individual RDT features. List is:
5170 cmt, mbmtotal, mbmlocal, l3cat, l3cdp, l2cat, l2cdp,
5172 E.g. to turn on cmt and turn off mba use:
5176 Format (x86 or x86_64):
5177 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] | d[efault] \
5179 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \
5181 Where reboot_mode is one of warm (soft) or cold (hard) or gpio
5182 (prefix with 'panic_' to set mode for panic
5184 reboot_type is one of bios, acpi, kbd, triple, efi, or pci,
5185 reboot_force is either force or not specified,
5186 reboot_cpu is s[mp]#### with #### being the processor
5187 to be used for rebooting.
5189 refscale.holdoff= [KNL]
5190 Set test-start holdoff period. The purpose of
5191 this parameter is to delay the start of the
5192 test until boot completes in order to avoid
5195 refscale.loops= [KNL]
5196 Set the number of loops over the synchronization
5197 primitive under test. Increasing this number
5198 reduces noise due to loop start/end overhead,
5199 but the default has already reduced the per-pass
5200 noise to a handful of picoseconds on ca. 2020
5203 refscale.nreaders= [KNL]
5204 Set number of readers. The default value of -1
5205 selects N, where N is roughly 75% of the number
5206 of CPUs. A value of zero is an interesting choice.
5208 refscale.nruns= [KNL]
5209 Set number of runs, each of which is dumped onto
5212 refscale.readdelay= [KNL]
5213 Set the read-side critical-section duration,
5214 measured in microseconds.
5216 refscale.scale_type= [KNL]
5217 Specify the read-protection implementation to test.
5219 refscale.shutdown= [KNL]
5220 Shut down the system at the end of the performance
5221 test. This defaults to 1 (shut it down) when
5222 refscale is built into the kernel and to 0 (leave
5223 it running) when refscale is built as a module.
5225 refscale.verbose= [KNL]
5226 Enable additional printk() statements.
5228 refscale.verbose_batched= [KNL]
5229 Batch the additional printk() statements. If zero
5230 (the default) or negative, print everything. Otherwise,
5231 print every Nth verbose statement, where N is the value
5235 [KNL, SMP] Set scheduler's default relax_domain_level.
5236 See Documentation/admin-guide/cgroup-v1/cpusets.rst.
5238 reserve= [KNL,BUGS] Force kernel to ignore I/O ports or memory
5239 Format: <base1>,<size1>[,<base2>,<size2>,...]
5240 Reserve I/O ports or memory so the kernel won't use
5241 them. If <base> is less than 0x10000, the region
5242 is assumed to be I/O ports; otherwise it is memory.
5244 reservetop= [X86-32]
5246 Reserves a hole at the top of the kernel virtual
5249 reset_devices [KNL] Force drivers to reset the underlying device
5250 during initialization.
5253 Specify the partition device for software suspend
5255 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>}
5257 resume_offset= [SWSUSP]
5258 Specify the offset from the beginning of the partition
5259 given by "resume=" at which the swap header is located,
5260 in <PAGE_SIZE> units (needed only for swap files).
5261 See Documentation/power/swsusp-and-swap-files.rst
5263 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to
5264 read the resume files
5266 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up.
5267 Useful for devices that are detected asynchronously
5268 (e.g. USB and MMC devices).
5270 retain_initrd [RAM] Keep initrd memory after extraction
5272 retbleed= [X86] Control mitigation of RETBleed (Arbitrary
5273 Speculative Code Execution with Return Instructions)
5277 auto - automatically select a migitation
5278 auto,nosmt - automatically select a mitigation,
5279 disabling SMT if necessary for
5280 the full mitigation (only on Zen1
5281 and older without STIBP).
5282 ibpb - mitigate short speculation windows on
5283 basic block boundaries too. Safe, highest
5285 unret - force enable untrained return thunks,
5286 only effective on AMD f15h-f17h
5288 unret,nosmt - like unret, will disable SMT when STIBP
5291 Selecting 'auto' will choose a mitigation method at run
5292 time according to the CPU.
5294 Not specifying this option is equivalent to retbleed=auto.
5296 rfkill.default_state=
5297 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm,
5298 etc. communication is blocked by default.
5301 rfkill.master_switch_mode=
5302 0 The "airplane mode" button does nothing.
5303 1 The "airplane mode" button toggles between everything
5304 blocked and the previous configuration.
5305 2 The "airplane mode" button toggles between everything
5306 blocked and everything unblocked.
5308 rhash_entries= [KNL,NET]
5309 Set number of hash buckets for route cache
5312 [KNL] Disable ring 3 MONITOR/MWAIT feature on supported
5315 ro [KNL] Mount root device read-only on boot
5318 on Mark read-only kernel memory as read-only (default).
5319 off Leave read-only kernel memory writable for debugging.
5322 Enable the uart passthrough on the designated usb port
5323 on Rockchip SoCs. When active, the signals of the
5324 debug-uart get routed to the D+ and D- pins of the usb
5325 port and the regular usb controller gets disabled.
5327 root= [KNL] Root filesystem
5328 See name_to_dev_t comment in init/do_mounts.c.
5330 rootdelay= [KNL] Delay (in seconds) to pause before attempting to
5331 mount the root filesystem
5333 rootflags= [KNL] Set root filesystem mount option string
5335 rootfstype= [KNL] Set root filesystem type
5337 rootwait [KNL] Wait (indefinitely) for root device to show up.
5338 Useful for devices that are detected asynchronously
5339 (e.g. USB and MMC devices).
5341 rproc_mem=nn[KMG][@address]
5342 [KNL,ARM,CMA] Remoteproc physical memory block.
5343 Memory area to be used by remote processor image,
5346 rw [KNL] Mount root device read-write on boot
5348 S [KNL] Run init in single mode
5350 s390_iommu= [HW,S390]
5351 Set s390 IOTLB flushing mode
5353 With strict flushing every unmap operation will result in
5354 an IOTLB flush. Default is lazy flushing before reuse,
5357 s390_iommu_aperture= [KNL,S390]
5358 Specifies the size of the per device DMA address space
5359 accessible through the DMA and IOMMU APIs as a decimal
5360 factor of the size of main memory.
5361 The default is 1 meaning that one can concurrently use
5362 as many DMA addresses as physical memory is installed,
5363 if supported by hardware, and thus map all of memory
5364 once. With a value of 2 one can map all of memory twice
5365 and so on. As a special case a factor of 0 imposes no
5366 restrictions other than those given by hardware at the
5367 cost of significant additional memory use for tables.
5370 See drivers/net/irda/sa1100_ir.c.
5372 sched_verbose [KNL] Enables verbose scheduler debug messages.
5374 schedstats= [KNL,X86] Enable or disable scheduled statistics.
5375 Allowed values are enable and disable. This feature
5376 incurs a small amount of overhead in the scheduler
5377 but is useful for debugging and performance tuning.
5379 sched_thermal_decay_shift=
5380 [KNL, SMP] Set a decay shift for scheduler thermal
5381 pressure signal. Thermal pressure signal follows the
5382 default decay period of other scheduler pelt
5383 signals(usually 32 ms but configurable). Setting
5384 sched_thermal_decay_shift will left shift the decay
5385 period for the thermal pressure signal by the shift
5387 i.e. with the default pelt decay period of 32 ms
5388 sched_thermal_decay_shift thermal pressure decay pr
5392 Format: integer between 0 and 10
5395 scftorture.holdoff= [KNL]
5396 Number of seconds to hold off before starting
5397 test. Defaults to zero for module insertion and
5398 to 10 seconds for built-in smp_call_function()
5401 scftorture.longwait= [KNL]
5402 Request ridiculously long waits randomly selected
5403 up to the chosen limit in seconds. Zero (the
5404 default) disables this feature. Please note
5405 that requesting even small non-zero numbers of
5406 seconds can result in RCU CPU stall warnings,
5407 softlockup complaints, and so on.
5409 scftorture.nthreads= [KNL]
5410 Number of kthreads to spawn to invoke the
5411 smp_call_function() family of functions.
5412 The default of -1 specifies a number of kthreads
5413 equal to the number of CPUs.
5415 scftorture.onoff_holdoff= [KNL]
5416 Number seconds to wait after the start of the
5417 test before initiating CPU-hotplug operations.
5419 scftorture.onoff_interval= [KNL]
5420 Number seconds to wait between successive
5421 CPU-hotplug operations. Specifying zero (which
5422 is the default) disables CPU-hotplug operations.
5424 scftorture.shutdown_secs= [KNL]
5425 The number of seconds following the start of the
5426 test after which to shut down the system. The
5427 default of zero avoids shutting down the system.
5428 Non-zero values are useful for automated tests.
5430 scftorture.stat_interval= [KNL]
5431 The number of seconds between outputting the
5432 current test statistics to the console. A value
5433 of zero disables statistics output.
5435 scftorture.stutter_cpus= [KNL]
5436 The number of jiffies to wait between each change
5437 to the set of CPUs under test.
5439 scftorture.use_cpus_read_lock= [KNL]
5440 Use use_cpus_read_lock() instead of the default
5441 preempt_disable() to disable CPU hotplug
5442 while invoking one of the smp_call_function*()
5445 scftorture.verbose= [KNL]
5446 Enable additional printk() statements.
5448 scftorture.weight_single= [KNL]
5449 The probability weighting to use for the
5450 smp_call_function_single() function with a zero
5451 "wait" parameter. A value of -1 selects the
5452 default if all other weights are -1. However,
5453 if at least one weight has some other value, a
5454 value of -1 will instead select a weight of zero.
5456 scftorture.weight_single_wait= [KNL]
5457 The probability weighting to use for the
5458 smp_call_function_single() function with a
5459 non-zero "wait" parameter. See weight_single.
5461 scftorture.weight_many= [KNL]
5462 The probability weighting to use for the
5463 smp_call_function_many() function with a zero
5464 "wait" parameter. See weight_single.
5465 Note well that setting a high probability for
5466 this weighting can place serious IPI load
5469 scftorture.weight_many_wait= [KNL]
5470 The probability weighting to use for the
5471 smp_call_function_many() function with a
5472 non-zero "wait" parameter. See weight_single
5475 scftorture.weight_all= [KNL]
5476 The probability weighting to use for the
5477 smp_call_function_all() function with a zero
5478 "wait" parameter. See weight_single and
5481 scftorture.weight_all_wait= [KNL]
5482 The probability weighting to use for the
5483 smp_call_function_all() function with a
5484 non-zero "wait" parameter. See weight_single
5487 skew_tick= [KNL] Offset the periodic timer tick per cpu to mitigate
5488 xtime_lock contention on larger systems, and/or RCU lock
5489 contention on all systems with CONFIG_MAXSMP set.
5490 Format: { "0" | "1" }
5491 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1"
5493 Note: increases power consumption, thus should only be
5494 enabled if running jitter sensitive (HPC/RT) workloads.
5496 security= [SECURITY] Choose a legacy "major" security module to
5497 enable at boot. This has been deprecated by the
5500 selinux= [SELINUX] Disable or enable SELinux at boot time.
5501 Format: { "0" | "1" }
5502 See security/selinux/Kconfig help text.
5507 apparmor= [APPARMOR] Disable or enable AppArmor at boot time
5508 Format: { "0" | "1" }
5509 See security/apparmor/Kconfig help text
5512 Default value is set via kernel config option.
5514 serialnumber [BUGS=X86-32]
5516 sev=option[,option...] [X86-64] See Documentation/x86/x86_64/boot-options.rst
5519 Maximal number of shapers.
5527 Enable merging of slabs with similar size when the
5528 kernel is built without CONFIG_SLAB_MERGE_DEFAULT.
5531 Disable merging of slabs with similar size. May be
5532 necessary if there is some reason to distinguish
5533 allocs to different slabs, especially in hardened
5534 environments where the risk of heap overflows and
5535 layout control by attackers can usually be
5536 frustrated by disabling merging. This will reduce
5537 most of the exposure of a heap attack to a single
5538 cache (risks via metadata attacks are mostly
5539 unchanged). Debug options disable merging on their
5541 For more information see Documentation/mm/slub.rst.
5543 slab_max_order= [MM, SLAB]
5544 Determines the maximum allowed order for slabs.
5545 A high setting may cause OOMs due to memory
5546 fragmentation. Defaults to 1 for systems with
5547 more than 32MB of RAM, 0 otherwise.
5549 slub_debug[=options[,slabs][;[options[,slabs]]...] [MM, SLUB]
5550 Enabling slub_debug allows one to determine the
5551 culprit if slab objects become corrupted. Enabling
5552 slub_debug can create guard zones around objects and
5553 may poison objects when not in use. Also tracks the
5554 last alloc / free. For more information see
5555 Documentation/mm/slub.rst.
5557 slub_max_order= [MM, SLUB]
5558 Determines the maximum allowed order for slabs.
5559 A high setting may cause OOMs due to memory
5560 fragmentation. For more information see
5561 Documentation/mm/slub.rst.
5563 slub_min_objects= [MM, SLUB]
5564 The minimum number of objects per slab. SLUB will
5565 increase the slab order up to slub_max_order to
5566 generate a sufficiently large slab able to contain
5567 the number of objects indicated. The higher the number
5568 of objects the smaller the overhead of tracking slabs
5569 and the less frequently locks need to be acquired.
5570 For more information see Documentation/mm/slub.rst.
5572 slub_min_order= [MM, SLUB]
5573 Determines the minimum page order for slabs. Must be
5574 lower than slub_max_order.
5575 For more information see Documentation/mm/slub.rst.
5577 slub_merge [MM, SLUB]
5578 Same with slab_merge.
5580 slub_nomerge [MM, SLUB]
5581 Same with slab_nomerge. This is supported for legacy.
5582 See slab_nomerge for more information.
5585 Format: <io1>[,<io2>[,...,<io8>]]
5587 smp.csd_lock_timeout= [KNL]
5588 Specify the period of time in milliseconds
5589 that smp_call_function() and friends will wait
5590 for a CPU to release the CSD lock. This is
5591 useful when diagnosing bugs involving CPUs
5592 disabling interrupts for extended periods
5593 of time. Defaults to 5,000 milliseconds, and
5594 setting a value of zero disables this feature.
5595 This feature may be more efficiently disabled
5596 using the csdlock_debug- kernel parameter.
5598 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices
5599 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port
5600 smsc-ircc2.ircc_sir= [HW] SIR base I/O port
5601 smsc-ircc2.ircc_fir= [HW] FIR base I/O port
5602 smsc-ircc2.ircc_irq= [HW] IRQ line
5603 smsc-ircc2.ircc_dma= [HW] DMA channel
5604 smsc-ircc2.ircc_transceiver= [HW] Transceiver type:
5605 0: Toshiba Satellite 1800 (GP data pin select)
5606 1: Fast pin select (default)
5609 smt= [KNL,S390] Set the maximum number of threads (logical
5610 CPUs) to use per physical CPU on systems capable of
5611 symmetric multithreading (SMT). Will be capped to the
5612 actual hardware limit.
5614 Default: -1 (no limit)
5617 [KNL] Should the soft-lockup detector generate panics.
5620 A value of 1 instructs the soft-lockup detector
5621 to panic the machine when a soft-lockup occurs. It is
5622 also controlled by the kernel.softlockup_panic sysctl
5623 and CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC, which is the
5624 respective build-time switch to that functionality.
5626 softlockup_all_cpu_backtrace=
5627 [KNL] Should the soft-lockup detector generate
5628 backtraces on all cpus.
5631 sonypi.*= [HW] Sony Programmable I/O Control Device driver
5632 See Documentation/admin-guide/laptops/sonypi.rst
5634 spectre_v2= [X86] Control mitigation of Spectre variant 2
5635 (indirect branch speculation) vulnerability.
5636 The default operation protects the kernel from
5639 on - unconditionally enable, implies
5641 off - unconditionally disable, implies
5643 auto - kernel detects whether your CPU model is
5646 Selecting 'on' will, and 'auto' may, choose a
5647 mitigation method at run time according to the
5648 CPU, the available microcode, the setting of the
5649 CONFIG_RETPOLINE configuration option, and the
5650 compiler with which the kernel was built.
5652 Selecting 'on' will also enable the mitigation
5653 against user space to user space task attacks.
5655 Selecting 'off' will disable both the kernel and
5656 the user space protections.
5658 Specific mitigations can also be selected manually:
5660 retpoline - replace indirect branches
5661 retpoline,generic - Retpolines
5662 retpoline,lfence - LFENCE; indirect branch
5663 retpoline,amd - alias for retpoline,lfence
5664 eibrs - enhanced IBRS
5665 eibrs,retpoline - enhanced IBRS + Retpolines
5666 eibrs,lfence - enhanced IBRS + LFENCE
5667 ibrs - use IBRS to protect kernel
5669 Not specifying this option is equivalent to
5673 [X86] Control mitigation of Spectre variant 2
5674 (indirect branch speculation) vulnerability between
5677 on - Unconditionally enable mitigations. Is
5678 enforced by spectre_v2=on
5680 off - Unconditionally disable mitigations. Is
5681 enforced by spectre_v2=off
5683 prctl - Indirect branch speculation is enabled,
5684 but mitigation can be enabled via prctl
5685 per thread. The mitigation control state
5686 is inherited on fork.
5689 - Like "prctl" above, but only STIBP is
5690 controlled per thread. IBPB is issued
5691 always when switching between different user
5695 - Same as "prctl" above, but all seccomp
5696 threads will enable the mitigation unless
5697 they explicitly opt out.
5700 - Like "seccomp" above, but only STIBP is
5701 controlled per thread. IBPB is issued
5702 always when switching between different
5703 user space processes.
5705 auto - Kernel selects the mitigation depending on
5706 the available CPU features and vulnerability.
5708 Default mitigation: "prctl"
5710 Not specifying this option is equivalent to
5711 spectre_v2_user=auto.
5713 spec_store_bypass_disable=
5714 [HW] Control Speculative Store Bypass (SSB) Disable mitigation
5715 (Speculative Store Bypass vulnerability)
5717 Certain CPUs are vulnerable to an exploit against a
5718 a common industry wide performance optimization known
5719 as "Speculative Store Bypass" in which recent stores
5720 to the same memory location may not be observed by
5721 later loads during speculative execution. The idea
5722 is that such stores are unlikely and that they can
5723 be detected prior to instruction retirement at the
5724 end of a particular speculation execution window.
5726 In vulnerable processors, the speculatively forwarded
5727 store can be used in a cache side channel attack, for
5728 example to read memory to which the attacker does not
5729 directly have access (e.g. inside sandboxed code).
5731 This parameter controls whether the Speculative Store
5732 Bypass optimization is used.
5734 On x86 the options are:
5736 on - Unconditionally disable Speculative Store Bypass
5737 off - Unconditionally enable Speculative Store Bypass
5738 auto - Kernel detects whether the CPU model contains an
5739 implementation of Speculative Store Bypass and
5740 picks the most appropriate mitigation. If the
5741 CPU is not vulnerable, "off" is selected. If the
5742 CPU is vulnerable the default mitigation is
5743 architecture and Kconfig dependent. See below.
5744 prctl - Control Speculative Store Bypass per thread
5745 via prctl. Speculative Store Bypass is enabled
5746 for a process by default. The state of the control
5747 is inherited on fork.
5748 seccomp - Same as "prctl" above, but all seccomp threads
5749 will disable SSB unless they explicitly opt out.
5751 Default mitigations:
5754 On powerpc the options are:
5756 on,auto - On Power8 and Power9 insert a store-forwarding
5757 barrier on kernel entry and exit. On Power7
5758 perform a software flush on kernel entry and
5762 Not specifying this option is equivalent to
5763 spec_store_bypass_disable=auto.
5765 spia_io_base= [HW,MTD]
5771 [X86] Enable split lock detection or bus lock detection
5773 When enabled (and if hardware support is present), atomic
5774 instructions that access data across cache line
5775 boundaries will result in an alignment check exception
5776 for split lock detection or a debug exception for
5781 warn - the kernel will emit rate-limited warnings
5782 about applications triggering the #AC
5783 exception or the #DB exception. This mode is
5784 the default on CPUs that support split lock
5785 detection or bus lock detection. Default
5786 behavior is by #AC if both features are
5787 enabled in hardware.
5789 fatal - the kernel will send SIGBUS to applications
5790 that trigger the #AC exception or the #DB
5791 exception. Default behavior is by #AC if
5792 both features are enabled in hardware.
5795 Set system wide rate limit to N bus locks
5796 per second for bus lock detection.
5799 N/A for split lock detection.
5802 If an #AC exception is hit in the kernel or in
5803 firmware (i.e. not while executing in user mode)
5804 the kernel will oops in either "warn" or "fatal"
5807 #DB exception for bus lock is triggered only when
5811 Control the Special Register Buffer Data Sampling
5814 Certain CPUs are vulnerable to an MDS-like
5815 exploit which can leak bits from the random
5818 By default, this issue is mitigated by
5819 microcode. However, the microcode fix can cause
5820 the RDRAND and RDSEED instructions to become
5821 much slower. Among other effects, this will
5822 result in reduced throughput from /dev/urandom.
5824 The microcode mitigation can be disabled with
5825 the following option:
5827 off: Disable mitigation and remove
5828 performance impact to RDRAND and RDSEED
5830 srcutree.big_cpu_lim [KNL]
5831 Specifies the number of CPUs constituting a
5832 large system, such that srcu_struct structures
5833 should immediately allocate an srcu_node array.
5834 This kernel-boot parameter defaults to 128,
5835 but takes effect only when the low-order four
5836 bits of srcutree.convert_to_big is equal to 3
5839 srcutree.convert_to_big [KNL]
5840 Specifies under what conditions an SRCU tree
5841 srcu_struct structure will be converted to big
5842 form, that is, with an rcu_node tree:
5845 1: At init_srcu_struct() time.
5846 2: When rcutorture decides to.
5847 3: Decide at boot time (default).
5848 0x1X: Above plus if high contention.
5850 Either way, the srcu_node tree will be sized based
5851 on the actual runtime number of CPUs (nr_cpu_ids)
5852 instead of the compile-time CONFIG_NR_CPUS.
5854 srcutree.counter_wrap_check [KNL]
5855 Specifies how frequently to check for
5856 grace-period sequence counter wrap for the
5857 srcu_data structure's ->srcu_gp_seq_needed field.
5858 The greater the number of bits set in this kernel
5859 parameter, the less frequently counter wrap will
5860 be checked for. Note that the bottom two bits
5863 srcutree.exp_holdoff [KNL]
5864 Specifies how many nanoseconds must elapse
5865 since the end of the last SRCU grace period for
5866 a given srcu_struct until the next normal SRCU
5867 grace period will be considered for automatic
5868 expediting. Set to zero to disable automatic
5871 srcutree.srcu_max_nodelay [KNL]
5872 Specifies the number of no-delay instances
5873 per jiffy for which the SRCU grace period
5874 worker thread will be rescheduled with zero
5875 delay. Beyond this limit, worker thread will
5876 be rescheduled with a sleep delay of one jiffy.
5878 srcutree.srcu_max_nodelay_phase [KNL]
5879 Specifies the per-grace-period phase, number of
5880 non-sleeping polls of readers. Beyond this limit,
5881 grace period worker thread will be rescheduled
5882 with a sleep delay of one jiffy, between each
5883 rescan of the readers, for a grace period phase.
5885 srcutree.srcu_retry_check_delay [KNL]
5886 Specifies number of microseconds of non-sleeping
5887 delay between each non-sleeping poll of readers.
5889 srcutree.small_contention_lim [KNL]
5890 Specifies the number of update-side contention
5891 events per jiffy will be tolerated before
5892 initiating a conversion of an srcu_struct
5893 structure to big form. Note that the value of
5894 srcutree.convert_to_big must have the 0x10 bit
5895 set for contention-based conversions to occur.
5898 Speculative Store Bypass Disable control
5900 On CPUs that are vulnerable to the Speculative
5901 Store Bypass vulnerability and offer a
5902 firmware based mitigation, this parameter
5903 indicates how the mitigation should be used:
5905 force-on: Unconditionally enable mitigation for
5906 for both kernel and userspace
5907 force-off: Unconditionally disable mitigation for
5908 for both kernel and userspace
5909 kernel: Always enable mitigation in the
5910 kernel, and offer a prctl interface
5911 to allow userspace to register its
5912 interest in being mitigated too.
5914 stack_guard_gap= [MM]
5915 override the default stack gap protection. The value
5916 is in page units and it defines how many pages prior
5917 to (for stacks growing down) resp. after (for stacks
5918 growing up) the main stack are reserved for no other
5919 mapping. Default value is 256 pages.
5921 stack_depot_disable= [KNL]
5922 Setting this to true through kernel command line will
5923 disable the stack depot thereby saving the static memory
5924 consumed by the stack hash table. By default this is set
5928 Enabled the stack tracer on boot up.
5930 stacktrace_filter=[function-list]
5931 [FTRACE] Limit the functions that the stack tracer
5932 will trace at boot up. function-list is a comma-separated
5933 list of functions. This list can be changed at run
5934 time by the stack_trace_filter file in the debugfs
5935 tracing directory. Note, this enables stack tracing
5936 and the stacktrace above is not needed.
5940 Set the STI (builtin display/keyboard on the HP-PARISC
5941 machines) console (graphic card) which should be used
5942 as the initial boot-console.
5943 See also comment in drivers/video/console/sticore.c.
5946 See comment in drivers/video/console/sticore.c.
5949 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]]
5954 Enable or disable strict sigaltstack size checks
5955 against the required signal frame size which
5956 depends on the supported FPU features. This can
5957 be used to filter out binaries which have
5958 not yet been made aware of AT_MINSIGSTKSZ.
5960 sunrpc.min_resvport=
5961 sunrpc.max_resvport=
5963 SunRPC servers often require that client requests
5964 originate from a privileged port (i.e. a port in the
5965 range 0 < portnr < 1024).
5966 An administrator who wishes to reserve some of these
5967 ports for other uses may adjust the range that the
5968 kernel's sunrpc client considers to be privileged
5969 using these two parameters to set the minimum and
5970 maximum port values.
5972 sunrpc.svc_rpc_per_connection_limit=
5974 Limit the number of requests that the server will
5975 process in parallel from a single connection.
5976 The default value is 0 (no limit).
5980 Control how the NFS server code allocates CPUs to
5981 service thread pools. Depending on how many NICs
5982 you have and where their interrupts are bound, this
5983 option will affect which CPUs will do NFS serving.
5984 Note: this parameter cannot be changed while the
5985 NFS server is running.
5987 auto the server chooses an appropriate mode
5988 automatically using heuristics
5989 global a single global pool contains all CPUs
5990 percpu one pool for each CPU
5991 pernode one pool for each NUMA node (equivalent
5992 to global on non-NUMA machines)
5994 sunrpc.tcp_slot_table_entries=
5995 sunrpc.udp_slot_table_entries=
5997 Sets the upper limit on the number of simultaneous
5998 RPC calls that can be sent from the client to a
5999 server. Increasing these values may allow you to
6000 improve throughput, but will also increase the
6001 amount of memory reserved for use by the client.
6003 suspend.pm_test_delay=
6005 Sets the number of seconds to remain in a suspend test
6006 mode before resuming the system (see
6007 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG
6008 is set. Default value is 5.
6011 Format: { on | off | y | n | 1 | 0 }
6012 This parameter controls use of the Protected
6013 Execution Facility on pSeries.
6017 Enable accounting of swap in memory resource
6018 controller if no parameter or 1 is given or disable
6019 it if 0 is given (See Documentation/admin-guide/cgroup-v1/memory.rst)
6021 swiotlb= [ARM,IA-64,PPC,MIPS,X86]
6022 Format: { <int> [,<int>] | force | noforce }
6023 <int> -- Number of I/O TLB slabs
6024 <int> -- Second integer after comma. Number of swiotlb
6025 areas with their own lock. Will be rounded up
6027 force -- force using of bounce buffers even if they
6028 wouldn't be automatically used by the kernel
6029 noforce -- Never use bounce buffers (for debugging)
6034 Set a sysctl parameter, right before loading the init
6035 process, as if the value was written to the respective
6036 /proc/sys/... file. Both '.' and '/' are recognized as
6037 separators. Unrecognized parameters and invalid values
6038 are reported in the kernel log. Sysctls registered
6039 later by a loaded module cannot be set this way.
6040 Example: sysctl.vm.swappiness=40
6042 sysfs.deprecated=0|1 [KNL]
6043 Enable/disable old style sysfs layout for old udev
6044 on older distributions. When this option is enabled
6045 very new udev will not work anymore. When this option
6046 is disabled (or CONFIG_SYSFS_DEPRECATED not compiled)
6047 in older udev will not work anymore.
6048 Default depends on CONFIG_SYSFS_DEPRECATED_V2 set in
6049 the kernel configuration.
6051 sysrq_always_enabled
6053 Ignore sysrq setting - this boot parameter will
6054 neutralize any effect of /proc/sys/kernel/sysrq.
6055 Useful for debugging.
6057 tcpmhash_entries= [KNL,NET]
6058 Set the number of tcp_metrics_hash slots.
6059 Default value is 8192 or 16384 depending on total
6060 ram pages. This is used to specify the TCP metrics
6061 cache size. See Documentation/networking/ip-sysctl.rst
6062 "tcp_no_metrics_save" section for more details.
6066 test_suspend= [SUSPEND]
6067 Format: { "mem" | "standby" | "freeze" }[,N]
6068 Specify "mem" (for Suspend-to-RAM) or "standby" (for
6069 standby suspend) or "freeze" (for suspend type freeze)
6070 as the system sleep state during system startup with
6071 the optional capability to repeat N number of times.
6072 The system is woken from this state using a
6073 wakeup-capable RTC alarm.
6075 thash_entries= [KNL,NET]
6076 Set number of hash buckets for TCP connection
6078 thermal.act= [HW,ACPI]
6079 -1: disable all active trip points in all thermal zones
6080 <degrees C>: override all lowest active trip points
6082 thermal.crt= [HW,ACPI]
6083 -1: disable all critical trip points in all thermal zones
6084 <degrees C>: override all critical trip points
6086 thermal.nocrt= [HW,ACPI]
6087 Set to disable actions on ACPI thermal zone
6088 critical and hot trip points.
6090 thermal.off= [HW,ACPI]
6091 1: disable ACPI thermal control
6093 thermal.psv= [HW,ACPI]
6094 -1: disable all passive trip points
6095 <degrees C>: override all passive trip points to this
6098 thermal.tzp= [HW,ACPI]
6099 Specify global default ACPI thermal zone polling rate
6100 <deci-seconds>: poll all this frequency
6101 0: no polling (default)
6104 Force threading of all interrupt handlers except those
6105 marked explicitly IRQF_NO_THREAD.
6109 Specify if the kernel should make use of the cpu
6110 topology information if the hardware supports this.
6111 The scheduler will make use of this information and
6112 e.g. base its process migration decisions on it.
6115 topology_updates= [KNL, PPC, NUMA]
6117 Specify if the kernel should ignore (off)
6118 topology updates sent by the hypervisor to this
6121 torture.disable_onoff_at_boot= [KNL]
6122 Prevent the CPU-hotplug component of torturing
6123 until after init has spawned.
6125 torture.ftrace_dump_at_shutdown= [KNL]
6126 Dump the ftrace buffer at torture-test shutdown,
6127 even if there were no errors. This can be a
6128 very costly operation when many torture tests
6129 are running concurrently, especially on systems
6130 with rotating-rust storage.
6132 torture.verbose_sleep_frequency= [KNL]
6133 Specifies how many verbose printk()s should be
6134 emitted between each sleep. The default of zero
6135 disables verbose-printk() sleeping.
6137 torture.verbose_sleep_duration= [KNL]
6138 Duration of each verbose-printk() sleep in jiffies.
6142 tpm_suspend_pcr=[HW,TPM]
6143 Format: integer pcr id
6144 Specify that at suspend time, the tpm driver
6145 should extend the specified pcr with zeros,
6146 as a workaround for some chips which fail to
6147 flush the last written pcr on TPM_SaveState.
6148 This will guarantee that all the other pcrs
6152 Have the tracepoints sent to printk as well as the
6153 tracing ring buffer. This is useful for early boot up
6154 where the system hangs or reboots and does not give the
6155 option for reading the tracing buffer or performing a
6156 ftrace_dump_on_oops.
6158 To turn off having tracepoints sent to printk,
6159 echo 0 > /proc/sys/kernel/tracepoint_printk
6160 Note, echoing 1 into this file without the
6161 tracepoint_printk kernel cmdline option has no effect.
6163 The tp_printk_stop_on_boot (see below) can also be used
6164 to stop the printing of events to console at
6169 Having tracepoints sent to printk() and activating high
6170 frequency tracepoints such as irq or sched, can cause
6171 the system to live lock.
6173 tp_printk_stop_on_boot [FTRACE]
6174 When tp_printk (above) is set, it can cause a lot of noise
6175 on the console. It may be useful to only include the
6176 printing of events during boot up, as user space may
6177 make the system inoperable.
6179 This command line option will stop the printing of events
6180 to console at the late_initcall_sync() time frame.
6182 trace_buf_size=nn[KMG]
6183 [FTRACE] will set tracing buffer size on each cpu.
6185 trace_clock= [FTRACE] Set the clock used for tracing events
6187 local - Use the per CPU time stamp counter
6188 (converted into nanoseconds). Fast, but
6189 depending on the architecture, may not be
6190 in sync between CPUs.
6191 global - Event time stamps are synchronize across
6192 CPUs. May be slower than the local clock,
6193 but better for some race conditions.
6194 counter - Simple counting of events (1, 2, ..)
6195 note, some counts may be skipped due to the
6196 infrastructure grabbing the clock more than
6198 uptime - Use jiffies as the time stamp.
6199 perf - Use the same clock that perf uses.
6200 mono - Use ktime_get_mono_fast_ns() for time stamps.
6201 mono_raw - Use ktime_get_raw_fast_ns() for time
6203 boot - Use ktime_get_boot_fast_ns() for time stamps.
6204 Architectures may add more clocks. See
6205 Documentation/trace/ftrace.rst for more details.
6207 trace_event=[event-list]
6208 [FTRACE] Set and start specified trace events in order
6209 to facilitate early boot debugging. The event-list is a
6210 comma-separated list of trace events to enable. See
6211 also Documentation/trace/events.rst
6213 trace_options=[option-list]
6214 [FTRACE] Enable or disable tracer options at boot.
6215 The option-list is a comma delimited list of options
6216 that can be enabled or disabled just as if you were
6217 to echo the option name into
6219 /sys/kernel/debug/tracing/trace_options
6221 For example, to enable stacktrace option (to dump the
6222 stack trace of each event), add to the command line:
6224 trace_options=stacktrace
6226 See also Documentation/trace/ftrace.rst "trace options"
6230 [FTRACE] enable this option to disable tracing when a
6231 warning is hit. This turns off "tracing_on". Tracing can
6232 be enabled again by echoing '1' into the "tracing_on"
6233 file located in /sys/kernel/debug/tracing/
6235 This option is useful, as it disables the trace before
6236 the WARNING dump is called, which prevents the trace to
6237 be filled with content caused by the warning output.
6239 This option can also be set at run time via the sysctl
6240 option: kernel/traceoff_on_warning
6242 transparent_hugepage=
6244 Format: [always|madvise|never]
6245 Can be used to control the default behavior of the system
6246 with respect to transparent hugepages.
6247 See Documentation/admin-guide/mm/transhuge.rst
6250 trusted.source= [KEYS]
6252 This parameter identifies the trust source as a backend
6253 for trusted keys implementation. Supported trust
6258 If not specified then it defaults to iterating through
6259 the trust source list starting with TPM and assigns the
6260 first trust source as a backend which is initialized
6261 successfully during iteration.
6265 The RNG used to generate key material for trusted keys.
6268 - the same value as trusted.source: "tpm" or "tee"
6270 If not specified, "default" is used. In this case,
6271 the RNG's choice is left to each individual trust source.
6273 tsc= Disable clocksource stability checks for TSC.
6275 [x86] reliable: mark tsc clocksource as reliable, this
6276 disables clocksource verification at runtime, as well
6277 as the stability checks done at bootup. Used to enable
6278 high-resolution timer mode on older hardware, and in
6279 virtualized environment.
6280 [x86] noirqtime: Do not use TSC to do irq accounting.
6281 Used to run time disable IRQ_TIME_ACCOUNTING on any
6282 platforms where RDTSC is slow and this accounting
6284 [x86] unstable: mark the TSC clocksource as unstable, this
6285 marks the TSC unconditionally unstable at bootup and
6286 avoids any further wobbles once the TSC watchdog notices.
6287 [x86] nowatchdog: disable clocksource watchdog. Used
6288 in situations with strict latency requirements (where
6289 interruptions from clocksource watchdog are not
6292 tsc_early_khz= [X86] Skip early TSC calibration and use the given
6293 value instead. Useful when the early TSC frequency discovery
6294 procedure is not reliable, such as on overclocked systems
6295 with CPUID.16h support and partial CPUID.15h support.
6296 Format: <unsigned int>
6298 tsx= [X86] Control Transactional Synchronization
6299 Extensions (TSX) feature in Intel processors that
6300 support TSX control.
6302 This parameter controls the TSX feature. The options are:
6304 on - Enable TSX on the system. Although there are
6305 mitigations for all known security vulnerabilities,
6306 TSX has been known to be an accelerator for
6307 several previous speculation-related CVEs, and
6308 so there may be unknown security risks associated
6309 with leaving it enabled.
6311 off - Disable TSX on the system. (Note that this
6312 option takes effect only on newer CPUs which are
6313 not vulnerable to MDS, i.e., have
6314 MSR_IA32_ARCH_CAPABILITIES.MDS_NO=1 and which get
6315 the new IA32_TSX_CTRL MSR through a microcode
6316 update. This new MSR allows for the reliable
6317 deactivation of the TSX functionality.)
6319 auto - Disable TSX if X86_BUG_TAA is present,
6320 otherwise enable TSX on the system.
6322 Not specifying this option is equivalent to tsx=off.
6324 See Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
6327 tsx_async_abort= [X86,INTEL] Control mitigation for the TSX Async
6328 Abort (TAA) vulnerability.
6330 Similar to Micro-architectural Data Sampling (MDS)
6331 certain CPUs that support Transactional
6332 Synchronization Extensions (TSX) are vulnerable to an
6333 exploit against CPU internal buffers which can forward
6334 information to a disclosure gadget under certain
6337 In vulnerable processors, the speculatively forwarded
6338 data can be used in a cache side channel attack, to
6339 access data to which the attacker does not have direct
6342 This parameter controls the TAA mitigation. The
6345 full - Enable TAA mitigation on vulnerable CPUs
6348 full,nosmt - Enable TAA mitigation and disable SMT on
6349 vulnerable CPUs. If TSX is disabled, SMT
6350 is not disabled because CPU is not
6351 vulnerable to cross-thread TAA attacks.
6352 off - Unconditionally disable TAA mitigation
6354 On MDS-affected machines, tsx_async_abort=off can be
6355 prevented by an active MDS mitigation as both vulnerabilities
6356 are mitigated with the same mechanism so in order to disable
6357 this mitigation, you need to specify mds=off too.
6359 Not specifying this option is equivalent to
6360 tsx_async_abort=full. On CPUs which are MDS affected
6361 and deploy MDS mitigation, TAA mitigation is not
6362 required and doesn't provide any additional
6366 Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
6368 turbografx.map[2|3]= [HW,JOY]
6369 TurboGraFX parallel port interface
6371 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7>
6372 See also Documentation/input/devices/joystick-parport.rst
6374 udbg-immortal [PPC] When debugging early kernel crashes that
6375 happen after console_init() and before a proper
6376 console driver takes over, this boot options might
6377 help "seeing" what's going on.
6379 uhash_entries= [KNL,NET]
6380 Set number of hash buckets for UDP/UDP-Lite connections
6383 [USB] Ignore overcurrent events (default N).
6384 Some badly-designed motherboards generate lots of
6385 bogus events, for ports that aren't wired to
6386 anything. Set this parameter to avoid log spamming.
6387 Note that genuine overcurrent events won't be
6391 [X86] Cause panic on unknown NMI.
6393 usbcore.authorized_default=
6394 [USB] Default USB device authorization:
6395 (default -1 = authorized except for wireless USB,
6396 0 = not authorized, 1 = authorized, 2 = authorized
6397 if device connected to internal port)
6399 usbcore.autosuspend=
6400 [USB] The autosuspend time delay (in seconds) used
6401 for newly-detected USB devices (default 2). This
6402 is the time required before an idle device will be
6403 autosuspended. Devices for which the delay is set
6404 to a negative value won't be autosuspended at all.
6406 usbcore.usbfs_snoop=
6407 [USB] Set to log all usbfs traffic (default 0 = off).
6409 usbcore.usbfs_snoop_max=
6410 [USB] Maximum number of bytes to snoop in each URB
6413 usbcore.blinkenlights=
6414 [USB] Set to cycle leds on hubs (default 0 = off).
6416 usbcore.old_scheme_first=
6417 [USB] Start with the old device initialization
6418 scheme (default 0 = off).
6420 usbcore.usbfs_memory_mb=
6421 [USB] Memory limit (in MB) for buffers allocated by
6422 usbfs (default = 16, 0 = max = 2047).
6424 usbcore.use_both_schemes=
6425 [USB] Try the other device initialization scheme
6426 if the first one fails (default 1 = enabled).
6428 usbcore.initial_descriptor_timeout=
6429 [USB] Specifies timeout for the initial 64-byte
6430 USB_REQ_GET_DESCRIPTOR request in milliseconds
6431 (default 5000 = 5.0 seconds).
6433 usbcore.nousb [USB] Disable the USB subsystem
6436 [USB] A list of quirk entries to augment the built-in
6437 usb core quirk list. List entries are separated by
6438 commas. Each entry has the form
6439 VendorID:ProductID:Flags. The IDs are 4-digit hex
6440 numbers and Flags is a set of letters. Each letter
6441 will change the built-in quirk; setting it if it is
6442 clear and clearing it if it is set. The letters have
6443 the following meanings:
6444 a = USB_QUIRK_STRING_FETCH_255 (string
6445 descriptors must not be fetched using
6447 b = USB_QUIRK_RESET_RESUME (device can't resume
6448 correctly so reset it instead);
6449 c = USB_QUIRK_NO_SET_INTF (device can't handle
6450 Set-Interface requests);
6451 d = USB_QUIRK_CONFIG_INTF_STRINGS (device can't
6452 handle its Configuration or Interface
6454 e = USB_QUIRK_RESET (device can't be reset
6455 (e.g morph devices), don't use reset);
6456 f = USB_QUIRK_HONOR_BNUMINTERFACES (device has
6457 more interface descriptions than the
6458 bNumInterfaces count, and can't handle
6459 talking to these interfaces);
6460 g = USB_QUIRK_DELAY_INIT (device needs a pause
6461 during initialization, after we read
6462 the device descriptor);
6463 h = USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL (For
6464 high speed and super speed interrupt
6465 endpoints, the USB 2.0 and USB 3.0 spec
6466 require the interval in microframes (1
6467 microframe = 125 microseconds) to be
6468 calculated as interval = 2 ^
6470 Devices with this quirk report their
6471 bInterval as the result of this
6472 calculation instead of the exponent
6473 variable used in the calculation);
6474 i = USB_QUIRK_DEVICE_QUALIFIER (device can't
6475 handle device_qualifier descriptor
6477 j = USB_QUIRK_IGNORE_REMOTE_WAKEUP (device
6478 generates spurious wakeup, ignore
6479 remote wakeup capability);
6480 k = USB_QUIRK_NO_LPM (device can't handle Link
6482 l = USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL
6483 (Device reports its bInterval as linear
6484 frames instead of the USB 2.0
6486 m = USB_QUIRK_DISCONNECT_SUSPEND (Device needs
6487 to be disconnected before suspend to
6488 prevent spurious wakeup);
6489 n = USB_QUIRK_DELAY_CTRL_MSG (Device needs a
6490 pause after every control message);
6491 o = USB_QUIRK_HUB_SLOW_RESET (Hub needs extra
6492 delay after resetting its port);
6493 Example: quirks=0781:5580:bk,0a5c:5834:gij
6496 [USBHID] The interval which mice are to be polled at.
6499 [USBHID] The interval which joysticks are to be polled at.
6502 [USBHID] The interval which keyboards are to be polled at.
6504 usb-storage.delay_use=
6505 [UMS] The delay in seconds before a new device is
6506 scanned for Logical Units (default 1).
6509 [UMS] A list of quirks entries to supplement or
6510 override the built-in unusual_devs list. List
6511 entries are separated by commas. Each entry has
6512 the form VID:PID:Flags where VID and PID are Vendor
6513 and Product ID values (4-digit hex numbers) and
6514 Flags is a set of characters, each corresponding
6515 to a common usb-storage quirk flag as follows:
6516 a = SANE_SENSE (collect more than 18 bytes
6517 of sense data, not on uas);
6518 b = BAD_SENSE (don't collect more than 18
6519 bytes of sense data, not on uas);
6520 c = FIX_CAPACITY (decrease the reported
6521 device capacity by one sector);
6522 d = NO_READ_DISC_INFO (don't use
6523 READ_DISC_INFO command, not on uas);
6524 e = NO_READ_CAPACITY_16 (don't use
6525 READ_CAPACITY_16 command);
6526 f = NO_REPORT_OPCODES (don't use report opcodes
6528 g = MAX_SECTORS_240 (don't transfer more than
6529 240 sectors at a time, uas only);
6530 h = CAPACITY_HEURISTICS (decrease the
6531 reported device capacity by one
6532 sector if the number is odd);
6533 i = IGNORE_DEVICE (don't bind to this
6535 j = NO_REPORT_LUNS (don't use report luns
6537 k = NO_SAME (do not use WRITE_SAME, uas only)
6538 l = NOT_LOCKABLE (don't try to lock and
6539 unlock ejectable media, not on uas);
6540 m = MAX_SECTORS_64 (don't transfer more
6541 than 64 sectors = 32 KB at a time,
6543 n = INITIAL_READ10 (force a retry of the
6544 initial READ(10) command, not on uas);
6545 o = CAPACITY_OK (accept the capacity
6546 reported by the device, not on uas);
6547 p = WRITE_CACHE (the device cache is ON
6548 by default, not on uas);
6549 r = IGNORE_RESIDUE (the device reports
6550 bogus residue values, not on uas);
6551 s = SINGLE_LUN (the device has only one
6553 t = NO_ATA_1X (don't allow ATA(12) and ATA(16)
6554 commands, uas only);
6555 u = IGNORE_UAS (don't bind to the uas driver);
6556 w = NO_WP_DETECT (don't test whether the
6557 medium is write-protected).
6558 y = ALWAYS_SYNC (issue a SYNCHRONIZE_CACHE
6559 even if the device claims no cache,
6561 Example: quirks=0419:aaf5:rl,0421:0433:rc
6563 user_debug= [KNL,ARM]
6565 See arch/arm/Kconfig.debug help text.
6566 1 - undefined instruction events
6568 4 - invalid data aborts
6571 Example: user_debug=31
6574 [X86] Flags controlling user PTE allocations.
6576 nohigh = do not allocate PTE pages in
6577 HIGHMEM regardless of setting
6580 vdso= [X86,SH,SPARC]
6581 On X86_32, this is an alias for vdso32=. Otherwise:
6583 vdso=1: enable VDSO (the default)
6584 vdso=0: disable VDSO mapping
6586 vdso32= [X86] Control the 32-bit vDSO
6587 vdso32=1: enable 32-bit VDSO
6588 vdso32=0 or vdso32=2: disable 32-bit VDSO
6590 See the help text for CONFIG_COMPAT_VDSO for more
6591 details. If CONFIG_COMPAT_VDSO is set, the default is
6592 vdso32=0; otherwise, the default is vdso32=1.
6594 For compatibility with older kernels, vdso32=2 is an
6597 Try vdso32=0 if you encounter an error that says:
6598 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed!
6601 vector=percpu: enable percpu vector domain
6603 video= [FB] Frame buffer configuration
6604 See Documentation/fb/modedb.rst.
6606 video.brightness_switch_enabled= [ACPI]
6608 If set to 1, on receiving an ACPI notify event
6609 generated by hotkey, video driver will adjust brightness
6610 level and then send out the event to user space through
6611 the allocated input device. If set to 0, video driver
6612 will only send out the event without touching backlight
6617 [VMMIO] Memory mapped virtio (platform) device.
6619 <size>@<baseaddr>:<irq>[:<id>]
6621 <size> := size (can use standard suffixes
6623 <baseaddr> := physical base address
6624 <irq> := interrupt number (as passed to
6626 <id> := (optional) platform device id
6628 virtio_mmio.device=1K@0x100b0000:48:7
6630 Can be used multiple times for multiple devices.
6632 vga= [BOOT,X86-32] Select a particular video mode
6633 See Documentation/x86/boot.rst and
6634 Documentation/admin-guide/svga.rst.
6635 Use vga=ask for menu.
6636 This is actually a boot loader parameter; the value is
6637 passed to the kernel using a special protocol.
6639 vm_debug[=options] [KNL] Available with CONFIG_DEBUG_VM=y.
6640 May slow down system boot speed, especially when
6641 enabled on systems with a large amount of memory.
6642 All options are enabled by default, and this
6643 interface is meant to allow for selectively
6644 enabling or disabling specific virtual memory
6647 Available options are:
6648 P Enable page structure init time poisoning
6649 - Disable all of the above options
6651 vmalloc=nn[KMG] [KNL,BOOT] Forces the vmalloc area to have an exact
6652 size of <nn>. This can be used to increase the
6653 minimum size (128MB on x86). It can also be used to
6654 decrease the size and leave more room for directly
6657 vmcp_cma=nn[MG] [KNL,S390]
6658 Sets the memory size reserved for contiguous memory
6659 allocations for the vmcp device driver.
6661 vmhalt= [KNL,S390] Perform z/VM CP command after system halt.
6664 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic.
6667 vmpoff= [KNL,S390] Perform z/VM CP command after power off.
6671 Controls the behavior of vsyscalls (i.e. calls to
6672 fixed addresses of 0xffffffffff600x00 from legacy
6673 code). Most statically-linked binaries and older
6674 versions of glibc use these calls. Because these
6675 functions are at fixed addresses, they make nice
6676 targets for exploits that can control RIP.
6678 emulate [default] Vsyscalls turn into traps and are
6679 emulated reasonably safely. The vsyscall
6682 xonly Vsyscalls turn into traps and are
6683 emulated reasonably safely. The vsyscall
6684 page is not readable.
6686 none Vsyscalls don't work at all. This makes
6687 them quite hard to use for exploits but
6688 might break your system.
6690 vt.color= [VT] Default text color.
6691 Format: 0xYX, X = foreground, Y = background.
6692 Default: 0x07 = light gray on black.
6694 vt.cur_default= [VT] Default cursor shape.
6695 Format: 0xCCBBAA, where AA, BB, and CC are the same as
6696 the parameters of the <Esc>[?A;B;Cc escape sequence;
6697 see VGA-softcursor.txt. Default: 2 = underline.
6699 vt.default_blu= [VT]
6700 Format: <blue0>,<blue1>,<blue2>,...,<blue15>
6701 Change the default blue palette of the console.
6702 This is a 16-member array composed of values
6705 vt.default_grn= [VT]
6706 Format: <green0>,<green1>,<green2>,...,<green15>
6707 Change the default green palette of the console.
6708 This is a 16-member array composed of values
6711 vt.default_red= [VT]
6712 Format: <red0>,<red1>,<red2>,...,<red15>
6713 Change the default red palette of the console.
6714 This is a 16-member array composed of values
6720 Set system-wide default UTF-8 mode for all tty's.
6721 Default is 1, i.e. UTF-8 mode is enabled for all
6722 newly opened terminals.
6724 vt.global_cursor_default=
6727 Set system-wide default for whether a cursor
6728 is shown on new VTs. Default is -1,
6729 i.e. cursors will be created by default unless
6730 overridden by individual drivers. 0 will hide
6731 cursors, 1 will display them.
6733 vt.italic= [VT] Default color for italic text; 0-15.
6736 vt.underline= [VT] Default color for underlined text; 0-15.
6739 watchdog timers [HW,WDT] For information on watchdog timers,
6740 see Documentation/watchdog/watchdog-parameters.rst
6741 or other driver-specific files in the
6742 Documentation/watchdog/ directory.
6746 Set the hard lockup detector stall duration
6747 threshold in seconds. The soft lockup detector
6748 threshold is set to twice the value. A value of 0
6749 disables both lockup detectors. Default is 10
6752 workqueue.watchdog_thresh=
6753 If CONFIG_WQ_WATCHDOG is configured, workqueue can
6754 warn stall conditions and dump internal state to
6755 help debugging. 0 disables workqueue stall
6756 detection; otherwise, it's the stall threshold
6757 duration in seconds. The default value is 30 and
6758 it can be updated at runtime by writing to the
6759 corresponding sysfs file.
6761 workqueue.disable_numa
6762 By default, all work items queued to unbound
6763 workqueues are affine to the NUMA nodes they're
6764 issued on, which results in better behavior in
6765 general. If NUMA affinity needs to be disabled for
6766 whatever reason, this option can be used. Note
6767 that this also can be controlled per-workqueue for
6768 workqueues visible under /sys/bus/workqueue/.
6770 workqueue.power_efficient
6771 Per-cpu workqueues are generally preferred because
6772 they show better performance thanks to cache
6773 locality; unfortunately, per-cpu workqueues tend to
6774 be more power hungry than unbound workqueues.
6776 Enabling this makes the per-cpu workqueues which
6777 were observed to contribute significantly to power
6778 consumption unbound, leading to measurably lower
6779 power usage at the cost of small performance
6782 The default value of this parameter is determined by
6783 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT.
6785 workqueue.debug_force_rr_cpu
6786 Workqueue used to implicitly guarantee that work
6787 items queued without explicit CPU specified are put
6788 on the local CPU. This guarantee is no longer true
6789 and while local CPU is still preferred work items
6790 may be put on foreign CPUs. This debug option
6791 forces round-robin CPU selection to flush out
6792 usages which depend on the now broken guarantee.
6793 When enabled, memory and cache locality will be
6796 x2apic_phys [X86-64,APIC] Use x2apic physical mode instead of
6797 default x2apic cluster mode on platforms
6800 xen_512gb_limit [KNL,X86-64,XEN]
6801 Restricts the kernel running paravirtualized under Xen
6802 to use only up to 512 GB of RAM. The reason to do so is
6803 crash analysis tools and Xen tools for doing domain
6804 save/restore/migration must be enabled to handle larger
6807 xen_emul_unplug= [HW,X86,XEN]
6808 Unplug Xen emulated devices
6809 Format: [unplug0,][unplug1]
6810 ide-disks -- unplug primary master IDE devices
6811 aux-ide-disks -- unplug non-primary-master IDE devices
6812 nics -- unplug network devices
6813 all -- unplug all emulated devices (NICs and IDE disks)
6814 unnecessary -- unplugging emulated devices is
6815 unnecessary even if the host did not respond to
6817 never -- do not unplug even if version check succeeds
6819 xen_legacy_crash [X86,XEN]
6820 Crash from Xen panic notifier, without executing late
6821 panic() code such as dumping handler.
6823 xen_nopvspin [X86,XEN]
6824 Disables the qspinlock slowpath using Xen PV optimizations.
6825 This parameter is obsoleted by "nopvspin" parameter, which
6826 has equivalent effect for XEN platform.
6829 Disables the PV optimizations forcing the HVM guest to
6830 run as generic HVM guest with no PV drivers.
6831 This option is obsoleted by the "nopv" option, which
6832 has equivalent effect for XEN platform.
6834 xen_no_vector_callback
6835 [KNL,X86,XEN] Disable the vector callback for Xen
6836 event channel interrupts.
6838 xen_scrub_pages= [XEN]
6839 Boolean option to control scrubbing pages before giving them back
6840 to Xen, for use by other domains. Can be also changed at runtime
6841 with /sys/devices/system/xen_memory/xen_memory0/scrub_pages.
6842 Default value controlled with CONFIG_XEN_SCRUB_PAGES_DEFAULT.
6844 xen_timer_slop= [X86-64,XEN]
6845 Set the timer slop (in nanoseconds) for the virtual Xen
6846 timers (default is 100000). This adjusts the minimum
6847 delta of virtualized Xen timers, where lower values
6848 improve timer resolution at the expense of processing
6849 more timer interrupts.
6851 xen.balloon_boot_timeout= [XEN]
6852 The time (in seconds) to wait before giving up to boot
6853 in case initial ballooning fails to free enough memory.
6854 Applies only when running as HVM or PVH guest and
6855 started with less memory configured than allowed at
6856 max. Default is 180.
6858 xen.event_eoi_delay= [XEN]
6859 How long to delay EOI handling in case of event
6860 storms (jiffies). Default is 10.
6862 xen.event_loop_timeout= [XEN]
6863 After which time (jiffies) the event handling loop
6864 should start to delay EOI handling. Default is 2.
6866 xen.fifo_events= [XEN]
6867 Boolean parameter to disable using fifo event handling
6868 even if available. Normally fifo event handling is
6869 preferred over the 2-level event handling, as it is
6870 fairer and the number of possible event channels is
6871 much higher. Default is on (use fifo events).
6873 nopv= [X86,XEN,KVM,HYPER_V,VMWARE]
6874 Disables the PV optimizations forcing the guest to run
6875 as generic guest with no PV drivers. Currently support
6876 XEN HVM, KVM, HYPER_V and VMWARE guest.
6878 nopvspin [X86,XEN,KVM]
6879 Disables the qspinlock slow path using PV optimizations
6880 which allow the hypervisor to 'idle' the guest on lock
6883 xirc2ps_cs= [NET,PCMCIA]
6885 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]]
6888 By default on POWER9 and above, the kernel will
6889 natively use the XIVE interrupt controller. This option
6890 allows the fallback firmware mode to be used:
6892 off Fallback to firmware control of XIVE interrupt
6893 controller on both pseries and powernv
6894 platforms. Only useful on POWER9 and above.
6896 xive.store-eoi=off [PPC]
6897 By default on POWER10 and above, the kernel will use
6898 stores for EOI handling when the XIVE interrupt mode
6899 is active. This option allows the XIVE driver to use
6900 loads instead, as on POWER9.
6902 xhci-hcd.quirks [USB,KNL]
6903 A hex value specifying bitmask with supplemental xhci
6904 host controller quirks. Meaning of each bit can be
6905 consulted in header drivers/usb/host/xhci.h.
6908 Format: { early | on | rw | ro | off }
6909 Controls if xmon debugger is enabled. Default is off.
6910 Passing only "xmon" is equivalent to "xmon=early".
6911 early Call xmon as early as possible on boot; xmon
6912 debugger is called from setup_arch().
6913 on xmon debugger hooks will be installed so xmon
6914 is only called on a kernel crash. Default mode,
6915 i.e. either "ro" or "rw" mode, is controlled
6916 with CONFIG_XMON_DEFAULT_RO_MODE.
6917 rw xmon debugger hooks will be installed so xmon
6918 is called only on a kernel crash, mode is write,
6919 meaning SPR registers, memory and, other data
6920 can be written using xmon commands.
6921 ro same as "rw" option above but SPR registers,
6922 memory, and other data can't be written using
6924 off xmon is disabled.