1 acpi= [HW,ACPI,X86,ARM64]
2 Advanced Configuration and Power Interface
3 Format: { force | on | off | strict | noirq | rsdt |
5 force -- enable ACPI if default was off
6 on -- enable ACPI but allow fallback to DT [arm64]
7 off -- disable ACPI if default was on
8 noirq -- do not use ACPI for IRQ routing
9 strict -- Be less tolerant of platforms that are not
10 strictly ACPI specification compliant.
11 rsdt -- prefer RSDT over (default) XSDT
12 copy_dsdt -- copy DSDT to memory
13 For ARM64, ONLY "acpi=off", "acpi=on" or "acpi=force"
16 See also Documentation/power/runtime_pm.rst, pci=noacpi
18 acpi_apic_instance= [ACPI, IOAPIC]
20 2: use 2nd APIC table, if available
21 1,0: use 1st APIC table
24 acpi_backlight= [HW,ACPI]
25 { vendor | video | native | none }
26 If set to vendor, prefer vendor-specific driver
27 (e.g. thinkpad_acpi, sony_acpi, etc.) instead
28 of the ACPI video.ko driver.
29 If set to video, use the ACPI video.ko driver.
30 If set to native, use the device's native backlight mode.
31 If set to none, disable the ACPI backlight interface.
33 acpi_force_32bit_fadt_addr
34 force FADT to use 32 bit addresses rather than the
35 64 bit X_* addresses. Some firmware have broken 64
36 bit addresses for force ACPI ignore these and use
37 the older legacy 32 bit addresses.
39 acpica_no_return_repair [HW, ACPI]
40 Disable AML predefined validation mechanism
41 This mechanism can repair the evaluation result to make
42 the return objects more ACPI specification compliant.
43 This option is useful for developers to identify the
44 root cause of an AML interpreter issue when the issue
45 has something to do with the repair mechanism.
47 acpi.debug_layer= [HW,ACPI,ACPI_DEBUG]
48 acpi.debug_level= [HW,ACPI,ACPI_DEBUG]
50 CONFIG_ACPI_DEBUG must be enabled to produce any ACPI
51 debug output. Bits in debug_layer correspond to a
52 _COMPONENT in an ACPI source file, e.g.,
53 #define _COMPONENT ACPI_EVENTS
54 Bits in debug_level correspond to a level in
55 ACPI_DEBUG_PRINT statements, e.g.,
56 ACPI_DEBUG_PRINT((ACPI_DB_INFO, ...
57 The debug_level mask defaults to "info". See
58 Documentation/firmware-guide/acpi/debug.rst for more information about
59 debug layers and levels.
61 Enable processor driver info messages:
62 acpi.debug_layer=0x20000000
63 Enable AML "Debug" output, i.e., stores to the Debug
64 object while interpreting AML:
65 acpi.debug_layer=0xffffffff acpi.debug_level=0x2
66 Enable all messages related to ACPI hardware:
67 acpi.debug_layer=0x2 acpi.debug_level=0xffffffff
69 Some values produce so much output that the system is
70 unusable. The "log_buf_len" parameter may be useful
71 if you need to capture more output.
73 acpi_enforce_resources= [ACPI]
75 Check for resource conflicts between native drivers
76 and ACPI OperationRegions (SystemIO and SystemMemory
77 only). IO ports and memory declared in ACPI might be
78 used by the ACPI subsystem in arbitrary AML code and
79 can interfere with legacy drivers.
80 strict (default): access to resources claimed by ACPI
81 is denied; legacy drivers trying to access reserved
82 resources will fail to bind to device using them.
83 lax: access to resources claimed by ACPI is allowed;
84 legacy drivers trying to access reserved resources
85 will bind successfully but a warning message is logged.
86 no: ACPI OperationRegions are not marked as reserved,
87 no further checks are performed.
89 acpi_force_table_verification [HW,ACPI]
90 Enable table checksum verification during early stage.
91 By default, this is disabled due to x86 early mapping
94 acpi_irq_balance [HW,ACPI]
95 ACPI will balance active IRQs
98 acpi_irq_nobalance [HW,ACPI]
99 ACPI will not move active IRQs (default)
102 acpi_irq_isa= [HW,ACPI] If irq_balance, mark listed IRQs used by ISA
103 Format: <irq>,<irq>...
105 acpi_irq_pci= [HW,ACPI] If irq_balance, clear listed IRQs for
107 Format: <irq>,<irq>...
109 acpi_mask_gpe= [HW,ACPI]
110 Due to the existence of _Lxx/_Exx, some GPEs triggered
111 by unsupported hardware/firmware features can result in
112 GPE floodings that cannot be automatically disabled by
114 This facility can be used to prevent such uncontrolled
116 Format: <byte> or <bitmap-list>
118 acpi_no_auto_serialize [HW,ACPI]
119 Disable auto-serialization of AML methods
120 AML control methods that contain the opcodes to create
121 named objects will be marked as "Serialized" by the
122 auto-serialization feature.
123 This feature is enabled by default.
124 This option allows to turn off the feature.
126 acpi_no_memhotplug [ACPI] Disable memory hotplug. Useful for kdump
129 acpi_no_static_ssdt [HW,ACPI]
130 Disable installation of static SSDTs at early boot time
131 By default, SSDTs contained in the RSDT/XSDT will be
132 installed automatically and they will appear under
133 /sys/firmware/acpi/tables.
134 This option turns off this feature.
135 Note that specifying this option does not affect
136 dynamic table installation which will install SSDT
137 tables to /sys/firmware/acpi/tables/dynamic.
139 acpi_no_watchdog [HW,ACPI,WDT]
140 Ignore the ACPI-based watchdog interface (WDAT) and let
141 a native driver control the watchdog device instead.
143 acpi_rsdp= [ACPI,EFI,KEXEC]
144 Pass the RSDP address to the kernel, mostly used
145 on machines running EFI runtime service to boot the
146 second kernel for kdump.
148 acpi_os_name= [HW,ACPI] Tell ACPI BIOS the name of the OS
149 Format: To spoof as Windows 98: ="Microsoft Windows"
151 acpi_rev_override [ACPI] Override the _REV object to return 5 (instead
152 of 2 which is mandated by ACPI 6) as the supported ACPI
153 specification revision (when using this switch, it may
154 be necessary to carry out a cold reboot _twice_ in a
155 row to make it take effect on the platform firmware).
157 acpi_osi= [HW,ACPI] Modify list of supported OS interface strings
158 acpi_osi="string1" # add string1
159 acpi_osi="!string2" # remove string2
160 acpi_osi=!* # remove all strings
161 acpi_osi=! # disable all built-in OS vendor
163 acpi_osi=!! # enable all built-in OS vendor
165 acpi_osi= # disable all strings
167 'acpi_osi=!' can be used in combination with single or
168 multiple 'acpi_osi="string1"' to support specific OS
169 vendor string(s). Note that such command can only
170 affect the default state of the OS vendor strings, thus
171 it cannot affect the default state of the feature group
172 strings and the current state of the OS vendor strings,
173 specifying it multiple times through kernel command line
174 is meaningless. This command is useful when one do not
175 care about the state of the feature group strings which
176 should be controlled by the OSPM.
178 1. 'acpi_osi=! acpi_osi="Windows 2000"' is equivalent
179 to 'acpi_osi="Windows 2000" acpi_osi=!', they all
180 can make '_OSI("Windows 2000")' TRUE.
182 'acpi_osi=' cannot be used in combination with other
183 'acpi_osi=' command lines, the _OSI method will not
184 exist in the ACPI namespace. NOTE that such command can
185 only affect the _OSI support state, thus specifying it
186 multiple times through kernel command line is also
189 1. 'acpi_osi=' can make 'CondRefOf(_OSI, Local1)'
192 'acpi_osi=!*' can be used in combination with single or
193 multiple 'acpi_osi="string1"' to support specific
194 string(s). Note that such command can affect the
195 current state of both the OS vendor strings and the
196 feature group strings, thus specifying it multiple times
197 through kernel command line is meaningful. But it may
198 still not able to affect the final state of a string if
199 there are quirks related to this string. This command
200 is useful when one want to control the state of the
201 feature group strings to debug BIOS issues related to
204 1. 'acpi_osi="Module Device" acpi_osi=!*' can make
205 '_OSI("Module Device")' FALSE.
206 2. 'acpi_osi=!* acpi_osi="Module Device"' can make
207 '_OSI("Module Device")' TRUE.
208 3. 'acpi_osi=! acpi_osi=!* acpi_osi="Windows 2000"' is
210 'acpi_osi=!* acpi_osi=! acpi_osi="Windows 2000"'
212 'acpi_osi=!* acpi_osi="Windows 2000" acpi_osi=!',
213 they all will make '_OSI("Windows 2000")' TRUE.
216 Override the pmtimer bug detection: force the kernel
217 to assume that this machine's pmtimer latches its value
218 and always returns good values.
220 acpi_sci= [HW,ACPI] ACPI System Control Interrupt trigger mode
221 Format: { level | edge | high | low }
223 acpi_skip_timer_override [HW,ACPI]
224 Recognize and ignore IRQ0/pin2 Interrupt Override.
225 For broken nForce2 BIOS resulting in XT-PIC timer.
227 acpi_sleep= [HW,ACPI] Sleep options
228 Format: { s3_bios, s3_mode, s3_beep, s4_hwsig,
229 s4_nohwsig, old_ordering, nonvs,
230 sci_force_enable, nobl }
231 See Documentation/power/video.rst for information on
233 s3_beep is for debugging; it makes the PC's speaker beep
234 as soon as the kernel's real-mode entry point is called.
235 s4_hwsig causes the kernel to check the ACPI hardware
236 signature during resume from hibernation, and gracefully
237 refuse to resume if it has changed. This complies with
238 the ACPI specification but not with reality, since
239 Windows does not do this and many laptops do change it
240 on docking. So the default behaviour is to allow resume
241 and simply warn when the signature changes, unless the
242 s4_hwsig option is enabled.
243 s4_nohwsig prevents ACPI hardware signature from being
244 used (or even warned about) during resume.
245 old_ordering causes the ACPI 1.0 ordering of the _PTS
246 control method, with respect to putting devices into
247 low power states, to be enforced (the ACPI 2.0 ordering
248 of _PTS is used by default).
249 nonvs prevents the kernel from saving/restoring the
250 ACPI NVS memory during suspend/hibernation and resume.
251 sci_force_enable causes the kernel to set SCI_EN directly
252 on resume from S1/S3 (which is against the ACPI spec,
253 but some broken systems don't work without it).
254 nobl causes the internal blacklist of systems known to
255 behave incorrectly in some ways with respect to system
256 suspend and resume to be ignored (use wisely).
258 acpi_use_timer_override [HW,ACPI]
259 Use timer override. For some broken Nvidia NF5 boards
260 that require a timer override, but don't have HPET
262 add_efi_memmap [EFI; X86] Include EFI memory map in
263 kernel's map of available physical RAM.
266 { off | try_unsupported }
267 off: disable AGP support
268 try_unsupported: try to drive unsupported chipsets
269 (may crash computer or cause data corruption)
272 See Documentation/sound/alsa-configuration.rst
275 Allow the default userspace alignment fault handler
276 behaviour to be specified. Bit 0 enables warnings,
277 bit 1 enables fixups, and bit 2 sends a segfault.
279 align_va_addr= [X86-64]
280 Align virtual addresses by clearing slice [14:12] when
281 allocating a VMA at process creation time. This option
282 gives you up to 3% performance improvement on AMD F15h
283 machines (where it is enabled by default) for a
284 CPU-intensive style benchmark, and it can vary highly in
285 a microbenchmark depending on workload and compiler.
287 32: only for 32-bit processes
288 64: only for 64-bit processes
289 on: enable for both 32- and 64-bit processes
290 off: disable for both 32- and 64-bit processes
292 alloc_snapshot [FTRACE]
293 Allocate the ftrace snapshot buffer on boot up when the
294 main buffer is allocated. This is handy if debugging
295 and you need to use tracing_snapshot() on boot up, and
296 do not want to use tracing_snapshot_alloc() as it needs
297 to be done where GFP_KERNEL allocations are allowed.
299 allow_mismatched_32bit_el0 [ARM64]
300 Allow execve() of 32-bit applications and setting of the
301 PER_LINUX32 personality on systems where only a strict
302 subset of the CPUs support 32-bit EL0. When this
303 parameter is present, the set of CPUs supporting 32-bit
304 EL0 is indicated by /sys/devices/system/cpu/aarch32_el0
305 and hot-unplug operations may be restricted.
307 See Documentation/arm64/asymmetric-32bit.rst for more
310 amd_iommu= [HW,X86-64]
311 Pass parameters to the AMD IOMMU driver in the system.
313 fullflush - Deprecated, equivalent to iommu.strict=1
314 off - do not initialize any AMD IOMMU found in
316 force_isolation - Force device isolation for all
317 devices. The IOMMU driver is not
318 allowed anymore to lift isolation
319 requirements as needed. This option
320 does not override iommu=pt
321 force_enable - Force enable the IOMMU on platforms known
322 to be buggy with IOMMU enabled. Use this
325 amd_iommu_dump= [HW,X86-64]
326 Enable AMD IOMMU driver option to dump the ACPI table
327 for AMD IOMMU. With this option enabled, AMD IOMMU
328 driver will print ACPI tables for AMD IOMMU during
329 IOMMU initialization.
331 amd_iommu_intr= [HW,X86-64]
332 Specifies one of the following AMD IOMMU interrupt
334 legacy - Use legacy interrupt remapping mode.
335 vapic - Use virtual APIC mode, which allows IOMMU
336 to inject interrupts directly into guest.
337 This mode requires kvm-amd.avic=1.
338 (Default when IOMMU HW support is present.)
340 amijoy.map= [HW,JOY] Amiga joystick support
341 Map of devices attached to JOY0DAT and JOY1DAT
343 See also Documentation/input/joydev/joystick.rst
345 analog.map= [HW,JOY] Analog joystick and gamepad support
346 Specifies type or capabilities of an analog joystick
347 connected to one of 16 gameports
348 Format: <type1>,<type2>,..<type16>
351 Power management functions (SPARCstation-4/5 + deriv.)
353 Disable APC CPU standby support. SPARCstation-Fox does
354 not play well with APC CPU idle - disable it if you have
355 APC and your system crashes randomly.
357 apic= [APIC,X86] Advanced Programmable Interrupt Controller
358 Change the output verbosity while booting
359 Format: { quiet (default) | verbose | debug }
360 Change the amount of debugging information output
361 when initialising the APIC and IO-APIC components.
362 For X86-32, this can also be used to specify an APIC
364 Format: apic=driver_name
365 Examples: apic=bigsmp
367 apic_extnmi= [APIC,X86] External NMI delivery setting
368 Format: { bsp (default) | all | none }
369 bsp: External NMI is delivered only to CPU 0
370 all: External NMIs are broadcast to all CPUs as a
372 none: External NMI is masked for all CPUs. This is
373 useful so that a dump capture kernel won't be
377 See Documentation/networking/ipv6.rst.
379 show_lapic= [APIC,X86] Advanced Programmable Interrupt Controller
380 Limit apic dumping. The parameter defines the maximal
381 number of local apics being dumped. Also it is possible
382 to set it to "all" by meaning -- no limit here.
383 Format: { 1 (default) | 2 | ... | all }.
384 The parameter valid if only apic=debug or
385 apic=verbose is specified.
386 Example: apic=debug show_lapic=all
388 apm= [APM] Advanced Power Management
389 See header of arch/x86/kernel/apm_32.c.
391 arcrimi= [HW,NET] ARCnet - "RIM I" (entirely mem-mapped) cards
392 Format: <io>,<irq>,<nodeID>
394 arm64.nobti [ARM64] Unconditionally disable Branch Target
395 Identification support
397 arm64.nopauth [ARM64] Unconditionally disable Pointer Authentication
400 arm64.nomte [ARM64] Unconditionally disable Memory Tagging Extension
403 arm64.nosve [ARM64] Unconditionally disable Scalable Vector
406 arm64.nosme [ARM64] Unconditionally disable Scalable Matrix
411 atarimouse= [HW,MOUSE] Atari Mouse
413 atkbd.extra= [HW] Enable extra LEDs and keys on IBM RapidAccess,
414 EzKey and similar keyboards
416 atkbd.reset= [HW] Reset keyboard during initialization
418 atkbd.set= [HW] Select keyboard code set
419 Format: <int> (2 = AT (default), 3 = PS/2)
421 atkbd.scroll= [HW] Enable scroll wheel on MS Office and similar
424 atkbd.softraw= [HW] Choose between synthetic and real raw mode
425 Format: <bool> (0 = real, 1 = synthetic (default))
427 atkbd.softrepeat= [HW]
428 Use software keyboard repeat
430 audit= [KNL] Enable the audit sub-system
431 Format: { "0" | "1" | "off" | "on" }
432 0 | off - kernel audit is disabled and can not be
433 enabled until the next reboot
434 unset - kernel audit is initialized but disabled and
435 will be fully enabled by the userspace auditd.
436 1 | on - kernel audit is initialized and partially
437 enabled, storing at most audit_backlog_limit
438 messages in RAM until it is fully enabled by the
442 audit_backlog_limit= [KNL] Set the audit queue size limit.
443 Format: <int> (must be >=0)
446 bau= [X86_UV] Enable the BAU on SGI UV. The default
447 behavior is to disable the BAU (i.e. bau=0).
448 Format: { "0" | "1" }
451 unset - Disable the BAU.
453 baycom_epp= [HW,AX25]
456 baycom_par= [HW,AX25] BayCom Parallel Port AX.25 Modem
458 See header of drivers/net/hamradio/baycom_par.c.
460 baycom_ser_fdx= [HW,AX25]
461 BayCom Serial Port AX.25 Modem (Full Duplex Mode)
462 Format: <io>,<irq>,<mode>[,<baud>]
463 See header of drivers/net/hamradio/baycom_ser_fdx.c.
465 baycom_ser_hdx= [HW,AX25]
466 BayCom Serial Port AX.25 Modem (Half Duplex Mode)
467 Format: <io>,<irq>,<mode>
468 See header of drivers/net/hamradio/baycom_ser_hdx.c.
471 Disable BERT OS support on buggy BIOSes.
473 bgrt_disable [ACPI][X86]
474 Disable BGRT to avoid flickering OEM logo.
476 blkdevparts= Manual partition parsing of block device(s) for
477 embedded devices based on command line input.
478 See Documentation/block/cmdline-partition.rst
480 boot_delay= Milliseconds to delay each printk during boot.
481 Values larger than 10 seconds (10000) are changed to
486 Extended command line options can be added to an initrd
487 and this will cause the kernel to look for it.
489 See Documentation/admin-guide/bootconfig.rst
491 bttv.card= [HW,V4L] bttv (bt848 + bt878 based grabber cards)
492 bttv.radio= Most important insmod options are available as
494 bttv.pll= See Documentation/admin-guide/media/bttv.rst
497 bulk_remove=off [PPC] This parameter disables the use of the pSeries
498 firmware feature for flushing multiple hpte entries
501 c101= [NET] Moxa C101 synchronous serial card
503 cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection.
504 Sometimes CPU hardware bugs make them report the cache
505 size incorrectly. The kernel will attempt work arounds
506 to fix known problems, but for some CPUs it is not
507 possible to determine what the correct size should be.
508 This option provides an override for these situations.
511 [NET] Specifies amount of time (in seconds) that
512 the kernel should wait for a network carrier. By default
513 it waits 120 seconds.
515 ca_keys= [KEYS] This parameter identifies a specific key(s) on
516 the system trusted keyring to be used for certificate
518 format: { id:<keyid> | builtin }
520 cca= [MIPS] Override the kernel pages' cache coherency
521 algorithm. Accepted values range from 0 to 7
522 inclusive. See arch/mips/include/asm/pgtable-bits.h
523 for platform specific values (SB1, Loongson3 and
526 ccw_timeout_log [S390]
527 See Documentation/s390/common_io.rst for details.
529 cgroup_disable= [KNL] Disable a particular controller or optional feature
530 Format: {name of the controller(s) or feature(s) to disable}
531 The effects of cgroup_disable=foo are:
532 - foo isn't auto-mounted if you mount all cgroups in
534 - foo isn't visible as an individually mountable
536 - if foo is an optional feature then the feature is
537 disabled and corresponding cgroup files are not
539 {Currently only "memory" controller deal with this and
540 cut the overhead, others just disable the usage. So
541 only cgroup_disable=memory is actually worthy}
542 Specifying "pressure" disables per-cgroup pressure
543 stall information accounting feature
545 cgroup_no_v1= [KNL] Disable cgroup controllers and named hierarchies in v1
546 Format: { { controller | "all" | "named" }
547 [,{ controller | "all" | "named" }...] }
548 Like cgroup_disable, but only applies to cgroup v1;
549 the blacklisted controllers remain available in cgroup2.
550 "all" blacklists all controllers and "named" disables
551 named mounts. Specifying both "all" and "named" disables
554 cgroup.memory= [KNL] Pass options to the cgroup memory controller.
556 nosocket -- Disable socket memory accounting.
557 nokmem -- Disable kernel memory accounting.
559 checkreqprot= [SELINUX] Set initial checkreqprot flag value.
560 Format: { "0" | "1" }
561 See security/selinux/Kconfig help text.
562 0 -- check protection applied by kernel (includes
563 any implied execute protection).
564 1 -- check protection requested by application.
565 Default value is set via a kernel config option.
566 Value can be changed at runtime via
567 /sys/fs/selinux/checkreqprot.
568 Setting checkreqprot to 1 is deprecated.
571 See Documentation/s390/common_io.rst for details.
573 clearcpuid=X[,X...] [X86]
574 Disable CPUID feature X for the kernel. See
575 arch/x86/include/asm/cpufeatures.h for the valid bit
576 numbers X. Note the Linux-specific bits are not necessarily
577 stable over kernel options, but the vendor-specific
579 X can also be a string as appearing in the flags: line
580 in /proc/cpuinfo which does not have the above
581 instability issue. However, not all features have names
583 Note that using this option will taint your kernel.
584 Also note that user programs calling CPUID directly
585 or using the feature without checking anything
586 will still see it. This just prevents it from
587 being used by the kernel or shown in /proc/cpuinfo.
588 Also note the kernel might malfunction if you disable
593 Prevents the clock framework from automatically gating
594 clocks that have not been explicitly enabled by a Linux
595 device driver but are enabled in hardware at reset or
596 by the bootloader/firmware. Note that this does not
597 force such clocks to be always-on nor does it reserve
598 those clocks in any way. This parameter is useful for
599 debug and development, but should not be needed on a
600 platform with proper driver support. For more
601 information, see Documentation/driver-api/clk.rst.
603 clock= [BUGS=X86-32, HW] gettimeofday clocksource override.
605 Forces specified clocksource (if available) to be used
606 when calculating gettimeofday(). If specified
607 clocksource is not available, it defaults to PIT.
608 Format: { pit | tsc | cyclone | pmtmr }
610 clocksource= Override the default clocksource
612 Override the default clocksource and use the clocksource
613 with the name specified.
614 Some clocksource names to choose from, depending on
616 [all] jiffies (this is the base, fallback clocksource)
618 [ARM] imx_timer1,OSTS,netx_timer,mpu_timer2,
619 pxa_timer,timer3,32k_counter,timer0_1
620 [X86-32] pit,hpet,tsc;
621 scx200_hrt on Geode; cyclone on IBM x440
629 clocksource.arm_arch_timer.evtstrm=
632 Enable/disable the eventstream feature of the ARM
633 architected timer so that code using WFE-based polling
634 loops can be debugged more effectively on production
637 clocksource.max_cswd_read_retries= [KNL]
638 Number of clocksource_watchdog() retries due to
639 external delays before the clock will be marked
640 unstable. Defaults to two retries, that is,
641 three attempts to read the clock under test.
643 clocksource.verify_n_cpus= [KNL]
644 Limit the number of CPUs checked for clocksources
645 marked with CLOCK_SOURCE_VERIFY_PERCPU that
646 are marked unstable due to excessive skew.
647 A negative value says to check all CPUs, while
648 zero says not to check any. Values larger than
649 nr_cpu_ids are silently truncated to nr_cpu_ids.
650 The actual CPUs are chosen randomly, with
651 no replacement if the same CPU is chosen twice.
653 clocksource-wdtest.holdoff= [KNL]
654 Set the time in seconds that the clocksource
655 watchdog test waits before commencing its tests.
656 Defaults to zero when built as a module and to
657 10 seconds when built into the kernel.
659 cma=nn[MG]@[start[MG][-end[MG]]]
661 Sets the size of kernel global memory area for
662 contiguous memory allocations and optionally the
663 placement constraint by the physical address range of
664 memory allocations. A value of 0 disables CMA
665 altogether. For more information, see
666 kernel/dma/contiguous.c
670 Sets the size of kernel per-numa memory area for
671 contiguous memory allocations. A value of 0 disables
672 per-numa CMA altogether. And If this option is not
673 specificed, the default value is 0.
674 With per-numa CMA enabled, DMA users on node nid will
675 first try to allocate buffer from the pernuma area
676 which is located in node nid, if the allocation fails,
677 they will fallback to the global default memory area.
679 cmo_free_hint= [PPC] Format: { yes | no }
680 Specify whether pages are marked as being inactive
681 when they are freed. This is used in CMO environments
682 to determine OS memory pressure for page stealing by
686 coherent_pool=nn[KMG] [ARM,KNL]
687 Sets the size of memory pool for coherent, atomic dma
688 allocations, by default set to 256K.
690 com20020= [HW,NET] ARCnet - COM20020 chipset
692 <io>[,<irq>[,<nodeID>[,<backplane>[,<ckp>[,<timeout>]]]]]
694 com90io= [HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers)
698 ARCnet - COM90xx chipset (memory-mapped buffers)
699 Format: <io>[,<irq>[,<memstart>]]
701 condev= [HW,S390] console device
704 console= [KNL] Output console device and options.
706 tty<n> Use the virtual console device <n>.
710 Use the specified serial port. The options are of
711 the form "bbbbpnf", where "bbbb" is the baud rate,
712 "p" is parity ("n", "o", or "e"), "n" is number of
713 bits, and "f" is flow control ("r" for RTS or
714 omit it). Default is "9600n8".
716 See Documentation/admin-guide/serial-console.rst for more
718 Documentation/networking/netconsole.rst for an
721 uart[8250],io,<addr>[,options]
722 uart[8250],mmio,<addr>[,options]
723 uart[8250],mmio16,<addr>[,options]
724 uart[8250],mmio32,<addr>[,options]
725 uart[8250],0x<addr>[,options]
726 Start an early, polled-mode console on the 8250/16550
727 UART at the specified I/O port or MMIO address,
728 switching to the matching ttyS device later.
729 MMIO inter-register address stride is either 8-bit
730 (mmio), 16-bit (mmio16), or 32-bit (mmio32).
731 If none of [io|mmio|mmio16|mmio32], <addr> is assumed
732 to be equivalent to 'mmio'. 'options' are specified in
733 the same format described for ttyS above; if unspecified,
734 the h/w is not re-initialized.
736 hvc<n> Use the hypervisor console device <n>. This is for
737 both Xen and PowerPC hypervisors.
740 Use to disable console output, i.e., to have kernel
741 console messages discarded.
742 This must be the only console= parameter used on the
745 If the device connected to the port is not a TTY but a braille
746 device, prepend "brl," before the device type, for instance
748 For now, only VisioBraille is supported.
751 [KNL] Change console messages format
753 By default we print messages on consoles in
754 "[time stamp] text\n" format (time stamp may not be
755 printed, depending on CONFIG_PRINTK_TIME or
756 `printk_time' param).
758 Switch to syslog format: "<%u>[time stamp] text\n"
759 IOW, each message will have a facility and loglevel
760 prefix. The format is similar to one used by syslog()
761 syscall, or to executing "dmesg -S --raw" or to reading
764 consoleblank= [KNL] The console blank (screen saver) timeout in
765 seconds. A value of 0 disables the blank timer.
769 [KNL] Change the default value for
770 /proc/<pid>/coredump_filter.
771 See also Documentation/filesystems/proc.rst.
773 coresight_cpu_debug.enable
776 Enable/disable the CPU sampling based debugging.
777 0: default value, disable debugging
778 1: enable debugging at boot time
780 cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver
782 <first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>]
784 cpu0_hotplug [X86] Turn on CPU0 hotplug feature when
785 CONFIG_BOOTPARAM_HOTPLUG_CPU0 is off.
786 Some features depend on CPU0. Known dependencies are:
787 1. Resume from suspend/hibernate depends on CPU0.
788 Suspend/hibernate will fail if CPU0 is offline and you
789 need to online CPU0 before suspend/hibernate.
790 2. PIC interrupts also depend on CPU0. CPU0 can't be
791 removed if a PIC interrupt is detected.
792 It's said poweroff/reboot may depend on CPU0 on some
793 machines although I haven't seen such issues so far
794 after CPU0 is offline on a few tested machines.
795 If the dependencies are under your control, you can
796 turn on cpu0_hotplug.
798 cpuidle.off=1 [CPU_IDLE]
799 disable the cpuidle sub-system
802 [CPU_IDLE] Name of the cpuidle governor to use.
804 cpufreq.off=1 [CPU_FREQ]
805 disable the cpufreq sub-system
807 cpufreq.default_governor=
808 [CPU_FREQ] Name of the default cpufreq governor or
809 policy to use. This governor must be registered in the
810 kernel before the cpufreq driver probes.
813 [X86] Delay for N microsec between assert and de-assert
814 of APIC INIT to start processors. This delay occurs
815 on every CPU online, such as boot, and resume from suspend.
818 crash_kexec_post_notifiers
819 Run kdump after running panic-notifiers and dumping
820 kmsg. This only for the users who doubt kdump always
821 succeeds in any situation.
822 Note that this also increases risks of kdump failure,
823 because some panic notifiers can make the crashed
824 kernel more unstable.
826 crashkernel=size[KMG][@offset[KMG]]
827 [KNL] Using kexec, Linux can switch to a 'crash kernel'
828 upon panic. This parameter reserves the physical
829 memory region [offset, offset + size] for that kernel
830 image. If '@offset' is omitted, then a suitable offset
831 is selected automatically.
832 [KNL, X86-64] Select a region under 4G first, and
833 fall back to reserve region above 4G when '@offset'
834 hasn't been specified.
835 See Documentation/admin-guide/kdump/kdump.rst for further details.
837 crashkernel=range1:size1[,range2:size2,...][@offset]
838 [KNL] Same as above, but depends on the memory
839 in the running system. The syntax of range is
840 start-[end] where start and end are both
841 a memory unit (amount[KMG]). See also
842 Documentation/admin-guide/kdump/kdump.rst for an example.
844 crashkernel=size[KMG],high
845 [KNL, X86-64, ARM64] range could be above 4G. Allow kernel
846 to allocate physical memory region from top, so could
847 be above 4G if system have more than 4G ram installed.
848 Otherwise memory region will be allocated below 4G, if
850 It will be ignored if crashkernel=X is specified.
851 crashkernel=size[KMG],low
852 [KNL, X86-64] range under 4G. When crashkernel=X,high
853 is passed, kernel could allocate physical memory region
854 above 4G, that cause second kernel crash on system
855 that require some amount of low memory, e.g. swiotlb
856 requires at least 64M+32K low memory, also enough extra
857 low memory is needed to make sure DMA buffers for 32-bit
858 devices won't run out. Kernel would try to allocate
859 at least 256M below 4G automatically.
860 This one lets the user specify own low range under 4G
861 for second kernel instead.
862 0: to disable low allocation.
863 It will be ignored when crashkernel=X,high is not used
864 or memory reserved is below 4G.
866 [KNL, ARM64] range in low memory.
867 This one lets the user specify a low range in the
868 DMA zone for the crash dump kernel.
869 It will be ignored when crashkernel=X,high is not used
870 or memory reserved is located in the DMA zones.
873 [KNL] Disable crypto self-tests
878 cs89x0_media= [HW,NET]
879 Format: { rj45 | aui | bnc }
881 csdlock_debug= [KNL] Enable debug add-ons of cross-CPU function call
882 handling. When switched on, additional debug data is
883 printed to the console in case a hanging CPU is
884 detected, and that CPU is pinged again in order to try
885 to resolve the hang situation.
886 0: disable csdlock debugging (default)
887 1: enable basic csdlock debugging (minor impact)
888 ext: enable extended csdlock debugging (more impact,
892 See header of drivers/s390/block/dasd_devmap.c.
894 db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port
895 (one device per port)
896 Format: <port#>,<type>
897 See also Documentation/input/devices/joystick-parport.rst
899 debug [KNL] Enable kernel debugging (events log level).
902 [KNL] Enable printing [hashed] pointers early in the
903 boot sequence. If enabled, we use a weak hash instead
904 of siphash to hash pointers. Use this option if you are
905 seeing instances of '(___ptrval___)') and need to see a
906 value (hashed pointer) instead. Cryptographically
907 insecure, please do not use on production kernels.
910 [KNL] verbose locking self-tests
912 Print debugging info while doing the locking API
914 Bitmask for the various LOCKTYPE_ tests. Defaults to 0
915 (no extra messages), setting it to -1 (all bits set)
916 will print _a_lot_ more information - normally only
917 useful to lockdep developers.
919 debug_objects [KNL] Enable object debugging
922 [KNL] Disable object debugging
924 debug_guardpage_minorder=
925 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this
926 parameter allows control of the order of pages that will
927 be intentionally kept free (and hence protected) by the
928 buddy allocator. Bigger value increase the probability
929 of catching random memory corruption, but reduce the
930 amount of memory for normal system use. The maximum
931 possible value is MAX_ORDER/2. Setting this parameter
932 to 1 or 2 should be enough to identify most random
933 memory corruption problems caused by bugs in kernel or
934 driver code when a CPU writes to (or reads from) a
935 random memory location. Note that there exists a class
936 of memory corruptions problems caused by buggy H/W or
937 F/W or by drivers badly programing DMA (basically when
938 memory is written at bus level and the CPU MMU is
939 bypassed) which are not detectable by
940 CONFIG_DEBUG_PAGEALLOC, hence this option will not help
941 tracking down these problems.
944 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this parameter
945 enables the feature at boot time. By default, it is
946 disabled and the system will work mostly the same as a
947 kernel built without CONFIG_DEBUG_PAGEALLOC.
948 Note: to get most of debug_pagealloc error reports, it's
949 useful to also enable the page_owner functionality.
950 on: enable the feature
952 debugfs= [KNL] This parameter enables what is exposed to userspace
953 and debugfs internal clients.
954 Format: { on, no-mount, off }
955 on: All functions are enabled.
957 Filesystem is not registered but kernel clients can
958 access APIs and a crashkernel can be used to read
959 its content. There is nothing to mount.
960 off: Filesystem is not registered and clients
961 get a -EPERM as result when trying to register files
962 or directories within debugfs.
963 This is equivalent of the runtime functionality if
964 debugfs was not enabled in the kernel at all.
965 Default value is set in build-time with a kernel configuration.
967 debugpat [X86] Enable PAT debugging
969 decnet.addr= [HW,NET]
970 Format: <area>[,<node>]
971 See also Documentation/networking/decnet.rst.
974 [HW] The size of the default HugeTLB page. This is
975 the size represented by the legacy /proc/ hugepages
976 APIs. In addition, this is the default hugetlb size
977 used for shmget(), mmap() and mounting hugetlbfs
978 filesystems. If not specified, defaults to the
979 architecture's default huge page size. Huge page
980 sizes are architecture dependent. See also
981 Documentation/admin-guide/mm/hugetlbpage.rst.
984 deferred_probe_timeout=
985 [KNL] Debugging option to set a timeout in seconds for
986 deferred probe to give up waiting on dependencies to
987 probe. Only specific dependencies (subsystems or
988 drivers) that have opted in will be ignored. A timeout
989 of 0 will timeout at the end of initcalls. If the time
990 out hasn't expired, it'll be restarted by each
991 successful driver registration. This option will also
992 dump out devices still on the deferred probe list after
995 delayacct [KNL] Enable per-task delay accounting
997 dell_smm_hwmon.ignore_dmi=
998 [HW] Continue probing hardware even if DMI data
999 indicates that the driver is running on unsupported
1002 dell_smm_hwmon.force=
1003 [HW] Activate driver even if SMM BIOS signature does
1004 not match list of supported models and enable otherwise
1005 blacklisted features.
1007 dell_smm_hwmon.power_status=
1008 [HW] Report power status in /proc/i8k
1009 (disabled by default).
1011 dell_smm_hwmon.restricted=
1012 [HW] Allow controlling fans only if SYS_ADMIN
1015 dell_smm_hwmon.fan_mult=
1016 [HW] Factor to multiply fan speed with.
1018 dell_smm_hwmon.fan_max=
1019 [HW] Maximum configurable fan speed.
1022 Format: { on | off | def_only | inf_only | always }
1023 on: s390 zlib hardware support for compression on
1024 level 1 and decompression (default)
1025 off: No s390 zlib hardware support
1026 def_only: s390 zlib hardware support for deflate
1027 only (compression on level 1)
1028 inf_only: s390 zlib hardware support for inflate
1029 only (decompression)
1030 always: Same as 'on' but ignores the selected compression
1031 level always using hardware support (used for debugging)
1033 dhash_entries= [KNL]
1034 Set number of hash buckets for dentry cache.
1036 disable_1tb_segments [PPC]
1037 Disables the use of 1TB hash page table segments. This
1038 causes the kernel to fall back to 256MB segments which
1039 can be useful when debugging issues that require an SLB
1043 Limits the number of kernel SLB entries, and flushes
1044 them frequently to increase the rate of SLB faults
1045 on kernel addresses.
1048 See Documentation/networking/ipv6.rst.
1051 Disable RADIX MMU mode on POWER9
1053 radix_hcall_invalidate=on [PPC/PSERIES]
1054 Disable RADIX GTSE feature and use hcall for TLB
1058 Disable TLBIE instruction. Currently does not work
1059 with KVM, with HASH MMU, or with coherent accelerators.
1061 disable_cpu_apicid= [X86,APIC,SMP]
1063 The number of initial APIC ID for the
1064 corresponding CPU to be disabled at boot,
1065 mostly used for the kdump 2nd kernel to
1066 disable BSP to wake up multiple CPUs without
1067 causing system reset or hang due to sending
1068 INIT from AP to BSP.
1070 disable_ddw [PPC/PSERIES]
1071 Disable Dynamic DMA Window support. Use this
1072 to workaround buggy firmware.
1074 disable_ipv6= [IPV6]
1075 See Documentation/networking/ipv6.rst.
1077 disable_mtrr_cleanup [X86]
1078 The kernel tries to adjust MTRR layout from continuous
1079 to discrete, to make X server driver able to add WB
1080 entry later. This parameter disables that.
1082 disable_mtrr_trim [X86, Intel and AMD only]
1083 By default the kernel will trim any uncacheable
1084 memory out of your available memory pool based on
1085 MTRR settings. This parameter disables that behavior,
1086 possibly causing your machine to run very slowly.
1088 disable_timer_pin_1 [X86]
1089 Disable PIN 1 of APIC timer
1090 Can be useful to work around chipset bugs.
1092 dis_ucode_ldr [X86] Disable the microcode loader.
1094 dma_debug=off If the kernel is compiled with DMA_API_DEBUG support,
1095 this option disables the debugging code at boot.
1097 dma_debug_entries=<number>
1098 This option allows to tune the number of preallocated
1099 entries for DMA-API debugging code. One entry is
1100 required per DMA-API allocation. Use this if the
1101 DMA-API debugging code disables itself because the
1102 architectural default is too low.
1104 dma_debug_driver=<driver_name>
1105 With this option the DMA-API debugging driver
1106 filter feature can be enabled at boot time. Just
1107 pass the driver to filter for as the parameter.
1108 The filter can be disabled or changed to another
1109 driver later using sysfs.
1111 driver_async_probe= [KNL]
1112 List of driver names to be probed asynchronously. *
1113 matches with all driver names. If * is specified, the
1114 rest of the listed driver names are those that will NOT
1116 Format: <driver_name1>,<driver_name2>...
1118 drm.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>]
1119 Broken monitors, graphic adapters, KVMs and EDIDless
1120 panels may send no or incorrect EDID data sets.
1121 This parameter allows to specify an EDID data sets
1122 in the /lib/firmware directory that are used instead.
1123 Generic built-in EDID data sets are used, if one of
1124 edid/1024x768.bin, edid/1280x1024.bin,
1125 edid/1680x1050.bin, or edid/1920x1080.bin is given
1126 and no file with the same name exists. Details and
1127 instructions how to build your own EDID data are
1128 available in Documentation/admin-guide/edid.rst. An EDID
1129 data set will only be used for a particular connector,
1130 if its name and a colon are prepended to the EDID
1131 name. Each connector may use a unique EDID data
1132 set by separating the files with a comma. An EDID
1133 data set with no connector name will be used for
1134 any connectors not explicitly specified.
1139 Format: {"off" | "known"}
1140 Control how the dt_cpu_ftrs device-tree binding is
1141 used for CPU feature discovery and setup (if it
1143 off: Do not use it, fall back to legacy cpu table.
1144 known: Do not pass through unknown features to guests
1145 or userspace, only those that the kernel is aware of.
1147 dump_apple_properties [X86]
1148 Dump name and content of EFI device properties on
1149 x86 Macs. Useful for driver authors to determine
1150 what data is available or for reverse-engineering.
1152 dyndbg[="val"] [KNL,DYNAMIC_DEBUG]
1153 <module>.dyndbg[="val"]
1154 Enable debug messages at boot time. See
1155 Documentation/admin-guide/dynamic-debug-howto.rst
1158 nopku [X86] Disable Memory Protection Keys CPU feature found
1161 <module>.async_probe [KNL]
1162 Enable asynchronous probe on this module.
1164 early_ioremap_debug [KNL]
1165 Enable debug messages in early_ioremap support. This
1166 is useful for tracking down temporary early mappings
1167 which are not unmapped.
1169 earlycon= [KNL] Output early console device and options.
1171 When used with no options, the early console is
1172 determined by stdout-path property in device tree's
1173 chosen node or the ACPI SPCR table if supported by
1176 cdns,<addr>[,options]
1177 Start an early, polled-mode console on a Cadence
1178 (xuartps) serial port at the specified address. Only
1179 supported option is baud rate. If baud rate is not
1180 specified, the serial port must already be setup and
1183 uart[8250],io,<addr>[,options]
1184 uart[8250],mmio,<addr>[,options]
1185 uart[8250],mmio32,<addr>[,options]
1186 uart[8250],mmio32be,<addr>[,options]
1187 uart[8250],0x<addr>[,options]
1188 Start an early, polled-mode console on the 8250/16550
1189 UART at the specified I/O port or MMIO address.
1190 MMIO inter-register address stride is either 8-bit
1191 (mmio) or 32-bit (mmio32 or mmio32be).
1192 If none of [io|mmio|mmio32|mmio32be], <addr> is assumed
1193 to be equivalent to 'mmio'. 'options' are specified
1194 in the same format described for "console=ttyS<n>"; if
1195 unspecified, the h/w is not initialized.
1199 Start an early, polled-mode console on a pl011 serial
1200 port at the specified address. The pl011 serial port
1201 must already be setup and configured. Options are not
1202 yet supported. If 'mmio32' is specified, then only
1203 the driver will use only 32-bit accessors to read/write
1204 the device registers.
1207 Start an early console on a litex serial port at the
1208 specified address. The serial port must already be
1209 setup and configured. Options are not yet supported.
1212 Start an early, polled-mode console on a meson serial
1213 port at the specified address. The serial port must
1214 already be setup and configured. Options are not yet
1218 Start an early, polled-mode console on an msm serial
1219 port at the specified address. The serial port
1220 must already be setup and configured. Options are not
1223 msm_serial_dm,<addr>
1224 Start an early, polled-mode console on an msm serial
1225 dm port at the specified address. The serial port
1226 must already be setup and configured. Options are not
1230 Start an early, polled-mode console on a serial port
1231 of an Actions Semi SoC, such as S500 or S900, at the
1232 specified address. The serial port must already be
1233 setup and configured. Options are not yet supported.
1236 Start an early, polled-mode console on a serial port
1237 of an RDA Micro SoC, such as RDA8810PL, at the
1238 specified address. The serial port must already be
1239 setup and configured. Options are not yet supported.
1242 Use RISC-V SBI (Supervisor Binary Interface) for early
1245 smh Use ARM semihosting calls for early console.
1253 Use early console provided by serial driver available
1254 on Samsung SoCs, requires selecting proper type and
1255 a correct base address of the selected UART port. The
1256 serial port must already be setup and configured.
1257 Options are not yet supported.
1260 Start an early, polled-mode console on a lantiq serial
1261 (lqasc) port at the specified address. The serial port
1262 must already be setup and configured. Options are not
1267 Use early console provided by Freescale LP UART driver
1268 found on Freescale Vybrid and QorIQ LS1021A processors.
1269 A valid base address must be provided, and the serial
1270 port must already be setup and configured.
1274 Start an early, polled-mode, output-only console on the
1275 Freescale i.MX UART at the specified address. The UART
1276 must already be setup and configured.
1279 Start an early, polled-mode console on the
1280 Armada 3700 serial port at the specified
1281 address. The serial port must already be setup
1282 and configured. Options are not yet supported.
1285 Start an early, polled-mode console on a Qualcomm
1286 Generic Interface (GENI) based serial port at the
1287 specified address. The serial port must already be
1288 setup and configured. Options are not yet supported.
1291 Start an early, unaccelerated console on the EFI
1292 memory mapped framebuffer (if available). On cache
1293 coherent non-x86 systems that use system memory for
1294 the framebuffer, pass the 'ram' option so that it is
1295 mapped with the correct attributes.
1298 Use early console provided by Freescale LINFlexD UART
1299 serial driver for NXP S32V234 SoCs. A valid base
1300 address must be provided, and the serial port must
1301 already be setup and configured.
1303 earlyprintk= [X86,SH,ARM,M68k,S390]
1307 earlyprintk=serial[,ttySn[,baudrate]]
1308 earlyprintk=serial[,0x...[,baudrate]]
1309 earlyprintk=ttySn[,baudrate]
1310 earlyprintk=dbgp[debugController#]
1311 earlyprintk=pciserial[,force],bus:device.function[,baudrate]
1312 earlyprintk=xdbc[xhciController#]
1314 earlyprintk is useful when the kernel crashes before
1315 the normal console is initialized. It is not enabled by
1316 default because it has some cosmetic problems.
1318 Append ",keep" to not disable it when the real console
1321 Only one of vga, serial, or usb debug port can
1324 Currently only ttyS0 and ttyS1 may be specified by
1325 name. Other I/O ports may be explicitly specified
1326 on some architectures (x86 and arm at least) by
1327 replacing ttySn with an I/O port address, like this:
1328 earlyprintk=serial,0x1008,115200
1329 You can find the port for a given device in
1330 /proc/tty/driver/serial:
1331 2: uart:ST16650V2 port:00001008 irq:18 ...
1333 Interaction with the standard serial driver is not
1336 The VGA output is eventually overwritten by
1339 The xen option can only be used in Xen domains.
1341 The sclp output can only be used on s390.
1343 The optional "force" to "pciserial" enables use of a
1344 PCI device even when its classcode is not of the
1347 edac_report= [HW,EDAC] Control how to report EDAC event
1348 Format: {"on" | "off" | "force"}
1349 on: enable EDAC to report H/W event. May be overridden
1350 by other higher priority error reporting module.
1351 off: disable H/W event reporting through EDAC.
1352 force: enforce the use of EDAC to report H/W event.
1356 Format: {"off" | "on" | "skip[mbr]"}
1359 Format: { "debug", "disable_early_pci_dma",
1360 "nochunk", "noruntime", "nosoftreserve",
1361 "novamap", "no_disable_early_pci_dma" }
1362 debug: enable misc debug output.
1363 disable_early_pci_dma: disable the busmaster bit on all
1364 PCI bridges while in the EFI boot stub.
1365 nochunk: disable reading files in "chunks" in the EFI
1366 boot stub, as chunking can cause problems with some
1367 firmware implementations.
1368 noruntime : disable EFI runtime services support
1369 nosoftreserve: The EFI_MEMORY_SP (Specific Purpose)
1370 attribute may cause the kernel to reserve the
1371 memory range for a memory mapping driver to
1372 claim. Specify efi=nosoftreserve to disable this
1373 reservation and treat the memory by its base type
1374 (i.e. EFI_CONVENTIONAL_MEMORY / "System RAM").
1375 novamap: do not call SetVirtualAddressMap().
1376 no_disable_early_pci_dma: Leave the busmaster bit set
1377 on all PCI bridges while in the EFI boot stub
1379 efi_no_storage_paranoia [EFI; X86]
1380 Using this parameter you can use more than 50% of
1381 your efi variable storage. Use this parameter only if
1382 you are really sure that your UEFI does sane gc and
1383 fulfills the spec otherwise your board may brick.
1385 efi_fake_mem= nn[KMG]@ss[KMG]:aa[,nn[KMG]@ss[KMG]:aa,..] [EFI; X86]
1386 Add arbitrary attribute to specific memory range by
1387 updating original EFI memory map.
1388 Region of memory which aa attribute is added to is
1391 If efi_fake_mem=2G@4G:0x10000,2G@0x10a0000000:0x10000
1392 is specified, EFI_MEMORY_MORE_RELIABLE(0x10000)
1393 attribute is added to range 0x100000000-0x180000000 and
1394 0x10a0000000-0x1120000000.
1396 If efi_fake_mem=8G@9G:0x40000 is specified, the
1397 EFI_MEMORY_SP(0x40000) attribute is added to
1398 range 0x240000000-0x43fffffff.
1400 Using this parameter you can do debugging of EFI memmap
1401 related features. For example, you can do debugging of
1402 Address Range Mirroring feature even if your box
1403 doesn't support it, or mark specific memory as
1406 efivar_ssdt= [EFI; X86] Name of an EFI variable that contains an SSDT
1407 that is to be dynamically loaded by Linux. If there are
1408 multiple variables with the same name but with different
1409 vendor GUIDs, all of them will be loaded. See
1410 Documentation/admin-guide/acpi/ssdt-overlays.rst for details.
1413 eisa_irq_edge= [PARISC,HW]
1414 See header of drivers/parisc/eisa.c.
1416 ekgdboc= [X86,KGDB] Allow early kernel console debugging
1419 This is designed to be used in conjunction with
1420 the boot argument: earlyprintk=vga
1422 This parameter works in place of the kgdboc parameter
1423 but can only be used if the backing tty is available
1424 very early in the boot process. For early debugging
1425 via a serial port see kgdboc_earlycon instead.
1428 See comment before function elanfreq_setup() in
1429 arch/x86/kernel/cpu/cpufreq/elanfreq.c.
1431 elfcorehdr=[size[KMG]@]offset[KMG] [IA64,PPC,SH,X86,S390]
1432 Specifies physical address of start of kernel core
1433 image elf header and optionally the size. Generally
1434 kexec loader will pass this option to capture kernel.
1435 See Documentation/admin-guide/kdump/kdump.rst for details.
1437 enable_mtrr_cleanup [X86]
1438 The kernel tries to adjust MTRR layout from continuous
1439 to discrete, to make X server driver able to add WB
1440 entry later. This parameter enables that.
1442 enable_timer_pin_1 [X86]
1443 Enable PIN 1 of APIC timer
1444 Can be useful to work around chipset bugs
1445 (in particular on some ATI chipsets).
1446 The kernel tries to set a reasonable default.
1448 enforcing= [SELINUX] Set initial enforcing status.
1450 See security/selinux/Kconfig help text.
1451 0 -- permissive (log only, no denials).
1452 1 -- enforcing (deny and log).
1454 Value can be changed at runtime via
1455 /sys/fs/selinux/enforce.
1458 Disable Error Record Serialization Table (ERST)
1461 ether= [HW,NET] Ethernet cards parameters
1462 This option is obsoleted by the "netdev=" option, which
1463 has equivalent usage. See its documentation for details.
1467 Permit 'security.evm' to be updated regardless of
1468 current integrity status.
1473 fail_make_request=[KNL]
1474 General fault injection mechanism.
1475 Format: <interval>,<probability>,<space>,<times>
1476 See also Documentation/fault-injection/.
1479 Format: { initns | none }
1480 See Documentation/admin-guide/sysctl/net.rst for
1481 fb_tunnels_only_for_init_ns
1484 See Documentation/admin-guide/blockdev/floppy.rst.
1486 force_pal_cache_flush
1487 [IA-64] Avoid check_sal_cache_flush which may hang on
1488 buggy SAL_CACHE_FLUSH implementations. Using this
1489 parameter will force ia64_sal_cache_flush to call
1490 ia64_pal_cache_flush instead of SAL_CACHE_FLUSH.
1493 Forcefully enable Physical Address Extension (PAE).
1494 Many Pentium M systems disable PAE but may have a
1495 functionally usable PAE implementation.
1496 Warning: use of this parameter will taint the kernel
1497 and may cause unknown problems.
1500 [FTRACE] will set and start the specified tracer
1501 as early as possible in order to facilitate early
1504 ftrace_boot_snapshot
1505 [FTRACE] On boot up, a snapshot will be taken of the
1506 ftrace ring buffer that can be read at:
1507 /sys/kernel/tracing/snapshot.
1508 This is useful if you need tracing information from kernel
1509 boot up that is likely to be overridden by user space
1510 start up functionality.
1512 ftrace_dump_on_oops[=orig_cpu]
1513 [FTRACE] will dump the trace buffers on oops.
1514 If no parameter is passed, ftrace will dump
1515 buffers of all CPUs, but if you pass orig_cpu, it will
1516 dump only the buffer of the CPU that triggered the
1519 ftrace_filter=[function-list]
1520 [FTRACE] Limit the functions traced by the function
1521 tracer at boot up. function-list is a comma-separated
1522 list of functions. This list can be changed at run
1523 time by the set_ftrace_filter file in the debugfs
1526 ftrace_notrace=[function-list]
1527 [FTRACE] Do not trace the functions specified in
1528 function-list. This list can be changed at run time
1529 by the set_ftrace_notrace file in the debugfs
1532 ftrace_graph_filter=[function-list]
1533 [FTRACE] Limit the top level callers functions traced
1534 by the function graph tracer at boot up.
1535 function-list is a comma-separated list of functions
1536 that can be changed at run time by the
1537 set_graph_function file in the debugfs tracing directory.
1539 ftrace_graph_notrace=[function-list]
1540 [FTRACE] Do not trace from the functions specified in
1541 function-list. This list is a comma-separated list of
1542 functions that can be changed at run time by the
1543 set_graph_notrace file in the debugfs tracing directory.
1545 ftrace_graph_max_depth=<uint>
1546 [FTRACE] Used with the function graph tracer. This is
1547 the max depth it will trace into a function. This value
1548 can be changed at run time by the max_graph_depth file
1549 in the tracefs tracing directory. default: 0 (no limit)
1551 fw_devlink= [KNL] Create device links between consumer and supplier
1552 devices by scanning the firmware to infer the
1553 consumer/supplier relationships. This feature is
1554 especially useful when drivers are loaded as modules as
1555 it ensures proper ordering of tasks like device probing
1556 (suppliers first, then consumers), supplier boot state
1557 clean up (only after all consumers have probed),
1558 suspend/resume & runtime PM (consumers first, then
1560 Format: { off | permissive | on | rpm }
1561 off -- Don't create device links from firmware info.
1562 permissive -- Create device links from firmware info
1563 but use it only for ordering boot state clean
1564 up (sync_state() calls).
1565 on -- Create device links from firmware info and use it
1566 to enforce probe and suspend/resume ordering.
1567 rpm -- Like "on", but also use to order runtime PM.
1569 fw_devlink.strict=<bool>
1570 [KNL] Treat all inferred dependencies as mandatory
1571 dependencies. This only applies for fw_devlink=on|rpm.
1575 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad
1576 support via parallel port (up to 5 devices per port)
1577 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5>
1578 See also Documentation/input/devices/joystick-parport.rst
1582 gart_fix_e820= [X86-64] disable the fix e820 for K8 GART
1586 gcov_persist= [GCOV] When non-zero (default), profiling data for
1587 kernel modules is saved and remains accessible via
1588 debugfs, even when the module is unloaded/reloaded.
1589 When zero, profiling data is discarded and associated
1590 debugfs files are removed at module unload time.
1592 goldfish [X86] Enable the goldfish android emulator platform.
1593 Don't use this when you are not running on the
1596 gpio-mockup.gpio_mockup_ranges
1597 [HW] Sets the ranges of gpiochip of for this device.
1598 Format: <start1>,<end1>,<start2>,<end2>...
1599 gpio-mockup.gpio_mockup_named_lines
1600 [HW] Let the driver know GPIO lines should be named.
1602 gpt [EFI] Forces disk with valid GPT signature but
1603 invalid Protective MBR to be treated as GPT. If the
1604 primary GPT is corrupted, it enables the backup/alternate
1605 GPT to be used instead.
1607 grcan.enable0= [HW] Configuration of physical interface 0. Determines
1608 the "Enable 0" bit of the configuration register.
1611 grcan.enable1= [HW] Configuration of physical interface 1. Determines
1612 the "Enable 0" bit of the configuration register.
1615 grcan.select= [HW] Select which physical interface to use.
1618 grcan.txsize= [HW] Sets the size of the tx buffer.
1619 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0.
1621 grcan.rxsize= [HW] Sets the size of the rx buffer.
1622 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0.
1626 [KNL] Under CONFIG_HARDENED_USERCOPY, whether
1627 hardening is enabled for this boot. Hardened
1628 usercopy checking is used to protect the kernel
1629 from reading or writing beyond known memory
1630 allocation boundaries as a proactive defense
1631 against bounds-checking flaws in the kernel's
1632 copy_to_user()/copy_from_user() interface.
1633 on Perform hardened usercopy checks (default).
1634 off Disable hardened usercopy checks.
1636 hardlockup_all_cpu_backtrace=
1637 [KNL] Should the hard-lockup detector generate
1638 backtraces on all cpus.
1641 hashdist= [KNL,NUMA] Large hashes allocated during boot
1642 are distributed across NUMA nodes. Defaults on
1643 for 64-bit NUMA, off otherwise.
1644 Format: 0 | 1 (for off | on)
1646 hcl= [IA-64] SGI's Hardware Graph compatibility layer
1648 hd= [EIDE] (E)IDE hard drive subsystem geometry
1649 Format: <cyl>,<head>,<sect>
1652 Disable Hardware Error Source Table (HEST) support;
1653 corresponding firmware-first mode error processing
1654 logic will be disabled.
1656 hibernate= [HIBERNATION]
1657 noresume Don't check if there's a hibernation image
1658 present during boot.
1659 nocompress Don't compress/decompress hibernation images.
1660 no Disable hibernation and resume.
1661 protect_image Turn on image protection during restoration
1662 (that will set all pages holding image data
1663 during restoration read-only).
1665 highmem=nn[KMG] [KNL,BOOT] forces the highmem zone to have an exact
1666 size of <nn>. This works even on boxes that have no
1667 highmem otherwise. This also works to reduce highmem
1668 size on bigger boxes.
1670 highres= [KNL] Enable/disable high resolution timer mode.
1671 Valid parameters: "on", "off"
1676 hpet= [X86-32,HPET] option to control HPET usage
1677 Format: { enable (default) | disable | force |
1679 disable: disable HPET and use PIT instead
1680 force: allow force enabled of undocumented chips (ICH4,
1682 verbose: show contents of HPET registers during setup
1684 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET
1685 registers. Default set by CONFIG_HPET_MMAP_DEFAULT.
1687 hugepages= [HW] Number of HugeTLB pages to allocate at boot.
1688 If this follows hugepagesz (below), it specifies
1689 the number of pages of hugepagesz to be allocated.
1690 If this is the first HugeTLB parameter on the command
1691 line, it specifies the number of pages to allocate for
1692 the default huge page size. If using node format, the
1693 number of pages to allocate per-node can be specified.
1694 See also Documentation/admin-guide/mm/hugetlbpage.rst.
1695 Format: <integer> or (node format)
1696 <node>:<integer>[,<node>:<integer>]
1699 [HW] The size of the HugeTLB pages. This is used in
1700 conjunction with hugepages (above) to allocate huge
1701 pages of a specific size at boot. The pair
1702 hugepagesz=X hugepages=Y can be specified once for
1703 each supported huge page size. Huge page sizes are
1704 architecture dependent. See also
1705 Documentation/admin-guide/mm/hugetlbpage.rst.
1708 hugetlb_cma= [HW,CMA] The size of a CMA area used for allocation
1709 of gigantic hugepages. Or using node format, the size
1710 of a CMA area per node can be specified.
1711 Format: nn[KMGTPE] or (node format)
1712 <node>:nn[KMGTPE][,<node>:nn[KMGTPE]]
1714 Reserve a CMA area of given size and allocate gigantic
1715 hugepages using the CMA allocator. If enabled, the
1716 boot-time allocation of gigantic hugepages is skipped.
1718 hugetlb_free_vmemmap=
1719 [KNL] Reguires CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP
1721 Allows heavy hugetlb users to free up some more
1722 memory (7 * PAGE_SIZE for each 2MB hugetlb page).
1723 Format: { [oO][Nn]/Y/y/1 | [oO][Ff]/N/n/0 (default) }
1725 [oO][Nn]/Y/y/1: enable the feature
1726 [oO][Ff]/N/n/0: disable the feature
1728 Built with CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP_DEFAULT_ON=y,
1731 Note that the vmemmap pages may be allocated from the added
1732 memory block itself when memory_hotplug.memmap_on_memory is
1733 enabled, those vmemmap pages cannot be optimized even if this
1734 feature is enabled. Other vmemmap pages not allocated from
1735 the added memory block itself do not be affected.
1738 [KNL] Should the hung task detector generate panics.
1741 A value of 1 instructs the kernel to panic when a
1742 hung task is detected. The default value is controlled
1743 by the CONFIG_BOOTPARAM_HUNG_TASK_PANIC build-time
1744 option. The value selected by this boot parameter can
1745 be changed later by the kernel.hung_task_panic sysctl.
1747 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC)
1748 terminal devices. Valid values: 0..8
1749 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs.
1750 If specified, z/VM IUCV HVC accepts connections
1751 from listed z/VM user IDs only.
1753 hv_nopvspin [X86,HYPER_V] Disables the paravirt spinlock optimizations
1754 which allow the hypervisor to 'idle' the
1755 guest on lock contention.
1758 Do not unregister boot console at start. This is only
1759 useful for debugging when something happens in the window
1760 between unregistering the boot console and initializing
1763 i2c_bus= [HW] Override the default board specific I2C bus speed
1764 or register an additional I2C bus that is not
1765 registered from board initialization code.
1769 i8042.debug [HW] Toggle i8042 debug mode
1770 i8042.unmask_kbd_data
1771 [HW] Enable printing of interrupt data from the KBD port
1772 (disabled by default, and as a pre-condition
1773 requires that i8042.debug=1 be enabled)
1774 i8042.direct [HW] Put keyboard port into non-translated mode
1775 i8042.dumbkbd [HW] Pretend that controller can only read data from
1776 keyboard and cannot control its state
1777 (Don't attempt to blink the leds)
1778 i8042.noaux [HW] Don't check for auxiliary (== mouse) port
1779 i8042.nokbd [HW] Don't check/create keyboard port
1780 i8042.noloop [HW] Disable the AUX Loopback command while probing
1782 i8042.nomux [HW] Don't check presence of an active multiplexing
1784 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX
1786 i8042.notimeout [HW] Ignore timeout condition signalled by controller
1787 i8042.reset [HW] Reset the controller during init, cleanup and
1788 suspend-to-ram transitions, only during s2r
1789 transitions, or never reset
1790 Format: { 1 | Y | y | 0 | N | n }
1791 1, Y, y: always reset controller
1792 0, N, n: don't ever reset controller
1793 Default: only on s2r transitions on x86; most other
1794 architectures force reset to be always executed
1795 i8042.unlock [HW] Unlock (ignore) the keylock
1796 i8042.kbdreset [HW] Reset device connected to KBD port
1798 [HW] Allow deferred probing upon i8042 probe errors
1802 i915.invert_brightness=
1803 [DRM] Invert the sense of the variable that is used to
1804 set the brightness of the panel backlight. Normally a
1805 brightness value of 0 indicates backlight switched off,
1806 and the maximum of the brightness value sets the backlight
1807 to maximum brightness. If this parameter is set to 0
1808 (default) and the machine requires it, or this parameter
1809 is set to 1, a brightness value of 0 sets the backlight
1810 to maximum brightness, and the maximum of the brightness
1811 value switches the backlight off.
1812 -1 -- never invert brightness
1813 0 -- machine default
1814 1 -- force brightness inversion
1817 Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]]
1821 Format: idle=poll, idle=halt, idle=nomwait
1822 Poll forces a polling idle loop that can slightly
1823 improve the performance of waking up a idle CPU, but
1824 will use a lot of power and make the system run hot.
1826 idle=halt: Halt is forced to be used for CPU idle.
1827 In such case C2/C3 won't be used again.
1828 idle=nomwait: Disable mwait for CPU C-states
1832 Allow force disabling of Shared Virtual Memory (SVA)
1833 support for the idxd driver. By default it is set to
1836 idxd.tc_override= [HW]
1838 Allow override of default traffic class configuration
1839 for the device. By default it is set to false (0).
1841 ieee754= [MIPS] Select IEEE Std 754 conformance mode
1842 Format: { strict | legacy | 2008 | relaxed }
1845 Choose which programs will be accepted for execution
1846 based on the IEEE 754 NaN encoding(s) supported by
1847 the FPU and the NaN encoding requested with the value
1848 of an ELF file header flag individually set by each
1849 binary. Hardware implementations are permitted to
1850 support either or both of the legacy and the 2008 NaN
1853 Available settings are as follows:
1854 strict accept binaries that request a NaN encoding
1855 supported by the FPU
1856 legacy only accept legacy-NaN binaries, if supported
1858 2008 only accept 2008-NaN binaries, if supported
1860 relaxed accept any binaries regardless of whether
1861 supported by the FPU
1863 The FPU emulator is always able to support both NaN
1864 encodings, so if no FPU hardware is present or it has
1865 been disabled with 'nofpu', then the settings of
1866 'legacy' and '2008' strap the emulator accordingly,
1867 'relaxed' straps the emulator for both legacy-NaN and
1868 2008-NaN, whereas 'strict' enables legacy-NaN only on
1869 legacy processors and both NaN encodings on MIPS32 or
1872 The setting for ABS.fmt/NEG.fmt instruction execution
1873 mode generally follows that for the NaN encoding,
1874 except where unsupported by hardware.
1876 ignore_loglevel [KNL]
1877 Ignore loglevel setting - this will print /all/
1878 kernel messages to the console. Useful for debugging.
1879 We also add it as printk module parameter, so users
1880 could change it dynamically, usually by
1881 /sys/module/printk/parameters/ignore_loglevel.
1884 Ignore RLIMIT_DATA setting for data mappings,
1885 print warning at first misuse. Can be changed via
1886 /sys/module/kernel/parameters/ignore_rlimit_data.
1888 ihash_entries= [KNL]
1889 Set number of hash buckets for inode cache.
1891 ima_appraise= [IMA] appraise integrity measurements
1892 Format: { "off" | "enforce" | "fix" | "log" }
1895 ima_appraise_tcb [IMA] Deprecated. Use ima_policy= instead.
1896 The builtin appraise policy appraises all files
1899 ima_canonical_fmt [IMA]
1900 Use the canonical format for the binary runtime
1901 measurements, instead of host native format.
1904 Format: { md5 | sha1 | rmd160 | sha256 | sha384
1908 The list of supported hash algorithms is defined
1909 in crypto/hash_info.h.
1912 The builtin policies to load during IMA setup.
1913 Format: "tcb | appraise_tcb | secure_boot |
1914 fail_securely | critical_data"
1916 The "tcb" policy measures all programs exec'd, files
1917 mmap'd for exec, and all files opened with the read
1918 mode bit set by either the effective uid (euid=0) or
1921 The "appraise_tcb" policy appraises the integrity of
1922 all files owned by root.
1924 The "secure_boot" policy appraises the integrity
1925 of files (eg. kexec kernel image, kernel modules,
1926 firmware, policy, etc) based on file signatures.
1928 The "fail_securely" policy forces file signature
1929 verification failure also on privileged mounted
1930 filesystems with the SB_I_UNVERIFIABLE_SIGNATURE
1933 The "critical_data" policy measures kernel integrity
1936 ima_tcb [IMA] Deprecated. Use ima_policy= instead.
1937 Load a policy which meets the needs of the Trusted
1938 Computing Base. This means IMA will measure all
1939 programs exec'd, files mmap'd for exec, and all files
1940 opened for read by uid=0.
1943 Select one of defined IMA measurements template formats.
1944 Formats: { "ima" | "ima-ng" | "ima-ngv2" | "ima-sig" |
1949 [IMA] Define a custom template format.
1950 Format: { "field1|...|fieldN" }
1952 ima.ahash_minsize= [IMA] Minimum file size for asynchronous hash usage
1953 Format: <min_file_size>
1954 Set the minimal file size for using asynchronous hash.
1955 If left unspecified, ahash usage is disabled.
1957 ahash performance varies for different data sizes on
1958 different crypto accelerators. This option can be used
1959 to achieve the best performance for a particular HW.
1961 ima.ahash_bufsize= [IMA] Asynchronous hash buffer size
1963 Set hashing buffer size. Default: 4k.
1965 ahash performance varies for different chunk sizes on
1966 different crypto accelerators. This option can be used
1967 to achieve best performance for particular HW.
1971 Run specified binary instead of /sbin/init as init
1974 initcall_debug [KNL] Trace initcalls as they are executed. Useful
1975 for working out where the kernel is dying during
1978 initcall_blacklist= [KNL] Do not execute a comma-separated list of
1979 initcall functions. Useful for debugging built-in
1980 modules and initcalls.
1982 initramfs_async= [KNL]
1985 This parameter controls whether the initramfs
1986 image is unpacked asynchronously, concurrently
1987 with devices being probed and
1988 initialized. This should normally just work,
1989 but as a debugging aid, one can get the
1990 historical behaviour of the initramfs
1991 unpacking being completed before device_ and
1994 initrd= [BOOT] Specify the location of the initial ramdisk
1996 initrdmem= [KNL] Specify a physical address and size from which to
1997 load the initrd. If an initrd is compiled in or
1998 specified in the bootparams, it takes priority over this
2000 Format: ss[KMG],nn[KMG]
2003 init_on_alloc= [MM] Fill newly allocated pages and heap objects with
2006 Default set by CONFIG_INIT_ON_ALLOC_DEFAULT_ON.
2008 init_on_free= [MM] Fill freed pages and heap objects with zeroes.
2010 Default set by CONFIG_INIT_ON_FREE_DEFAULT_ON.
2012 init_pkru= [X86] Specify the default memory protection keys rights
2013 register contents for all processes. 0x55555554 by
2014 default (disallow access to all but pkey 0). Can
2015 override in debugfs after boot.
2017 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver
2020 int_pln_enable [X86] Enable power limit notification interrupt
2022 integrity_audit=[IMA]
2023 Format: { "0" | "1" }
2024 0 -- basic integrity auditing messages. (Default)
2025 1 -- additional integrity auditing messages.
2027 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option
2029 Enable intel iommu driver.
2031 Disable intel iommu driver.
2032 igfx_off [Default Off]
2033 By default, gfx is mapped as normal device. If a gfx
2034 device has a dedicated DMAR unit, the DMAR unit is
2035 bypassed by not enabling DMAR with this option. In
2036 this case, gfx device will use physical address for
2038 strict [Default Off]
2039 Deprecated, equivalent to iommu.strict=1.
2040 sp_off [Default Off]
2041 By default, super page will be supported if Intel IOMMU
2042 has the capability. With this option, super page will
2045 Enable the Intel IOMMU scalable mode if the hardware
2046 advertises that it has support for the scalable mode
2049 Disallow use of the Intel IOMMU scalable mode.
2050 tboot_noforce [Default Off]
2051 Do not force the Intel IOMMU enabled under tboot.
2052 By default, tboot will force Intel IOMMU on, which
2053 could harm performance of some high-throughput
2054 devices like 40GBit network cards, even if identity
2056 Note that using this option lowers the security
2057 provided by tboot because it makes the system
2058 vulnerable to DMA attacks.
2060 intel_idle.max_cstate= [KNL,HW,ACPI,X86]
2061 0 disables intel_idle and fall back on acpi_idle.
2062 1 to 9 specify maximum depth of C-state.
2066 Do not enable intel_pstate as the default
2067 scaling driver for the supported processors
2069 Use intel_pstate as a scaling driver, but configure it
2070 to work with generic cpufreq governors (instead of
2071 enabling its internal governor). This mode cannot be
2072 used along with the hardware-managed P-states (HWP)
2075 Enable intel_pstate on systems that prohibit it by default
2076 in favor of acpi-cpufreq. Forcing the intel_pstate driver
2077 instead of acpi-cpufreq may disable platform features, such
2078 as thermal controls and power capping, that rely on ACPI
2079 P-States information being indicated to OSPM and therefore
2080 should be used with caution. This option does not work with
2081 processors that aren't supported by the intel_pstate driver
2082 or on platforms that use pcc-cpufreq instead of acpi-cpufreq.
2084 Do not enable hardware P state control (HWP)
2087 Only load intel_pstate on systems which support
2088 hardware P state control (HWP) if available.
2090 Enforce ACPI _PPC performance limits. If the Fixed ACPI
2091 Description Table, specifies preferred power management
2092 profile as "Enterprise Server" or "Performance Server",
2093 then this feature is turned on by default.
2095 Allow per-logical-CPU P-State performance control limits using
2096 cpufreq sysfs interface
2098 intremap= [X86-64, Intel-IOMMU]
2099 on enable Interrupt Remapping (default)
2100 off disable Interrupt Remapping
2101 nosid disable Source ID checking
2103 BIOS x2APIC opt-out request will be ignored
2104 nopost disable Interrupt Posting
2106 iomem= Disable strict checking of access to MMIO memory
2107 strict regions from userspace.
2122 nobypass [PPC/POWERNV]
2123 Disable IOMMU bypass, using IOMMU for PCI devices.
2125 iommu.forcedac= [ARM64, X86] Control IOVA allocation for PCI devices.
2126 Format: { "0" | "1" }
2127 0 - Try to allocate a 32-bit DMA address first, before
2128 falling back to the full range if needed.
2129 1 - Allocate directly from the full usable range,
2130 forcing Dual Address Cycle for PCI cards supporting
2131 greater than 32-bit addressing.
2133 iommu.strict= [ARM64, X86] Configure TLB invalidation behaviour
2134 Format: { "0" | "1" }
2136 Request that DMA unmap operations use deferred
2137 invalidation of hardware TLBs, for increased
2138 throughput at the cost of reduced device isolation.
2139 Will fall back to strict mode if not supported by
2140 the relevant IOMMU driver.
2142 DMA unmap operations invalidate IOMMU hardware TLBs
2144 unset - Use value of CONFIG_IOMMU_DEFAULT_DMA_{LAZY,STRICT}.
2145 Note: on x86, strict mode specified via one of the
2146 legacy driver-specific options takes precedence.
2149 [ARM64, X86] Configure DMA to bypass the IOMMU by default.
2150 Format: { "0" | "1" }
2151 0 - Use IOMMU translation for DMA.
2152 1 - Bypass the IOMMU for DMA.
2153 unset - Use value of CONFIG_IOMMU_DEFAULT_PASSTHROUGH.
2155 io7= [HW] IO7 for Marvel-based Alpha systems
2156 See comment before marvel_specify_io7 in
2157 arch/alpha/kernel/core_marvel.c.
2159 io_delay= [X86] I/O delay method
2161 Standard port 0x80 based delay
2163 Alternate port 0xed based delay (needed on some systems)
2165 Simple two microseconds delay
2170 See Documentation/admin-guide/nfs/nfsroot.rst.
2172 ipcmni_extend [KNL] Extend the maximum number of unique System V
2173 IPC identifiers from 32,768 to 16,777,216.
2175 irqaffinity= [SMP] Set the default irq affinity mask
2176 The argument is a cpu list, as described above.
2178 irqchip.gicv2_force_probe=
2181 Force the kernel to look for the second 4kB page
2182 of a GICv2 controller even if the memory range
2183 exposed by the device tree is too small.
2185 irqchip.gicv3_nolpi=
2187 Force the kernel to ignore the availability of
2188 LPIs (and by consequence ITSs). Intended for system
2189 that use the kernel as a bootloader, and thus want
2190 to let secondary kernels in charge of setting up
2193 irqchip.gicv3_pseudo_nmi= [ARM64]
2194 Enables support for pseudo-NMIs in the kernel. This
2195 requires the kernel to be built with
2196 CONFIG_ARM64_PSEUDO_NMI.
2199 When an interrupt is not handled search all handlers
2200 for it. Intended to get systems with badly broken
2204 When an interrupt is not handled search all handlers
2205 for it. Also check all handlers each timer
2206 interrupt. Intended to get systems with badly broken
2210 Format: <RDP>,<reset>,<pci_scan>,<verbosity>
2212 isolcpus= [KNL,SMP,ISOL] Isolate a given set of CPUs from disturbance.
2213 [Deprecated - use cpusets instead]
2214 Format: [flag-list,]<cpu-list>
2216 Specify one or more CPUs to isolate from disturbances
2217 specified in the flag list (default: domain):
2220 Disable the tick when a single task runs.
2222 A residual 1Hz tick is offloaded to workqueues, which you
2223 need to affine to housekeeping through the global
2224 workqueue's affinity configured via the
2225 /sys/devices/virtual/workqueue/cpumask sysfs file, or
2226 by using the 'domain' flag described below.
2228 NOTE: by default the global workqueue runs on all CPUs,
2229 so to protect individual CPUs the 'cpumask' file has to
2230 be configured manually after bootup.
2233 Isolate from the general SMP balancing and scheduling
2234 algorithms. Note that performing domain isolation this way
2235 is irreversible: it's not possible to bring back a CPU to
2236 the domains once isolated through isolcpus. It's strongly
2237 advised to use cpusets instead to disable scheduler load
2238 balancing through the "cpuset.sched_load_balance" file.
2239 It offers a much more flexible interface where CPUs can
2240 move in and out of an isolated set anytime.
2242 You can move a process onto or off an "isolated" CPU via
2243 the CPU affinity syscalls or cpuset.
2244 <cpu number> begins at 0 and the maximum value is
2245 "number of CPUs in system - 1".
2249 Isolate from being targeted by managed interrupts
2250 which have an interrupt mask containing isolated
2251 CPUs. The affinity of managed interrupts is
2252 handled by the kernel and cannot be changed via
2253 the /proc/irq/* interfaces.
2255 This isolation is best effort and only effective
2256 if the automatically assigned interrupt mask of a
2257 device queue contains isolated and housekeeping
2258 CPUs. If housekeeping CPUs are online then such
2259 interrupts are directed to the housekeeping CPU
2260 so that IO submitted on the housekeeping CPU
2261 cannot disturb the isolated CPU.
2263 If a queue's affinity mask contains only isolated
2264 CPUs then this parameter has no effect on the
2265 interrupt routing decision, though interrupts are
2266 only delivered when tasks running on those
2267 isolated CPUs submit IO. IO submitted on
2268 housekeeping CPUs has no influence on those
2271 The format of <cpu-list> is described above.
2275 ivrs_ioapic [HW,X86-64]
2276 Provide an override to the IOAPIC-ID<->DEVICE-ID
2277 mapping provided in the IVRS ACPI table.
2278 By default, PCI segment is 0, and can be omitted.
2280 * To map IOAPIC-ID decimal 10 to PCI device 00:14.0
2281 write the parameter as:
2282 ivrs_ioapic[10]=00:14.0
2283 * To map IOAPIC-ID decimal 10 to PCI segment 0x1 and
2284 PCI device 00:14.0 write the parameter as:
2285 ivrs_ioapic[10]=0001:00:14.0
2287 ivrs_hpet [HW,X86-64]
2288 Provide an override to the HPET-ID<->DEVICE-ID
2289 mapping provided in the IVRS ACPI table.
2290 By default, PCI segment is 0, and can be omitted.
2292 * To map HPET-ID decimal 0 to PCI device 00:14.0
2293 write the parameter as:
2294 ivrs_hpet[0]=00:14.0
2295 * To map HPET-ID decimal 10 to PCI segment 0x1 and
2296 PCI device 00:14.0 write the parameter as:
2297 ivrs_ioapic[10]=0001:00:14.0
2299 ivrs_acpihid [HW,X86-64]
2300 Provide an override to the ACPI-HID:UID<->DEVICE-ID
2301 mapping provided in the IVRS ACPI table.
2303 For example, to map UART-HID:UID AMD0020:0 to
2304 PCI segment 0x1 and PCI device ID 00:14.5,
2305 write the parameter as:
2306 ivrs_acpihid[0001:00:14.5]=AMD0020:0
2308 By default, PCI segment is 0, and can be omitted.
2309 For example, PCI device 00:14.5 write the parameter as:
2310 ivrs_acpihid[00:14.5]=AMD0020:0
2312 js= [HW,JOY] Analog joystick
2313 See Documentation/input/joydev/joystick.rst.
2316 When CONFIG_RANDOMIZE_BASE is set, this disables
2317 kernel and module base offset ASLR (Address Space
2318 Layout Randomization).
2321 [KNL] Enforce KASAN (Kernel Address Sanitizer) to print
2322 report on every invalid memory access. Without this
2323 parameter KASAN will print report only for the first
2328 kernelcore= [KNL,X86,IA-64,PPC]
2329 Format: nn[KMGTPE] | nn% | "mirror"
2330 This parameter specifies the amount of memory usable by
2331 the kernel for non-movable allocations. The requested
2332 amount is spread evenly throughout all nodes in the
2333 system as ZONE_NORMAL. The remaining memory is used for
2334 movable memory in its own zone, ZONE_MOVABLE. In the
2335 event, a node is too small to have both ZONE_NORMAL and
2336 ZONE_MOVABLE, kernelcore memory will take priority and
2337 other nodes will have a larger ZONE_MOVABLE.
2339 ZONE_MOVABLE is used for the allocation of pages that
2340 may be reclaimed or moved by the page migration
2341 subsystem. Note that allocations like PTEs-from-HighMem
2342 still use the HighMem zone if it exists, and the Normal
2343 zone if it does not.
2345 It is possible to specify the exact amount of memory in
2346 the form of "nn[KMGTPE]", a percentage of total system
2347 memory in the form of "nn%", or "mirror". If "mirror"
2348 option is specified, mirrored (reliable) memory is used
2349 for non-movable allocations and remaining memory is used
2350 for Movable pages. "nn[KMGTPE]", "nn%", and "mirror"
2351 are exclusive, so you cannot specify multiple forms.
2353 kgdbdbgp= [KGDB,HW] kgdb over EHCI usb debug port.
2354 Format: <Controller#>[,poll interval]
2355 The controller # is the number of the ehci usb debug
2356 port as it is probed via PCI. The poll interval is
2357 optional and is the number seconds in between
2358 each poll cycle to the debug port in case you need
2359 the functionality for interrupting the kernel with
2360 gdb or control-c on the dbgp connection. When
2361 not using this parameter you use sysrq-g to break into
2362 the kernel debugger.
2364 kgdboc= [KGDB,HW] kgdb over consoles.
2365 Requires a tty driver that supports console polling,
2366 or a supported polling keyboard driver (non-usb).
2367 Serial only format: <serial_device>[,baud]
2368 keyboard only format: kbd
2369 keyboard and serial format: kbd,<serial_device>[,baud]
2370 Optional Kernel mode setting:
2371 kms, kbd format: kms,kbd
2372 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud]
2374 kgdboc_earlycon= [KGDB,HW]
2375 If the boot console provides the ability to read
2376 characters and can work in polling mode, you can use
2377 this parameter to tell kgdb to use it as a backend
2378 until the normal console is registered. Intended to
2379 be used together with the kgdboc parameter which
2380 specifies the normal console to transition to.
2382 The name of the early console should be specified
2383 as the value of this parameter. Note that the name of
2384 the early console might be different than the tty
2385 name passed to kgdboc. It's OK to leave the value
2386 blank and the first boot console that implements
2387 read() will be picked.
2389 kgdbwait [KGDB] Stop kernel execution and enter the
2390 kernel debugger at the earliest opportunity.
2392 kmac= [MIPS] Korina ethernet MAC address.
2393 Configure the RouterBoard 532 series on-chip
2394 Ethernet adapter MAC address.
2396 kmemleak= [KNL] Boot-time kmemleak enable/disable
2397 Valid arguments: on, off
2399 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y,
2402 kprobe_event=[probe-list]
2403 [FTRACE] Add kprobe events and enable at boot time.
2404 The probe-list is a semicolon delimited list of probe
2405 definitions. Each definition is same as kprobe_events
2406 interface, but the parameters are comma delimited.
2407 For example, to add a kprobe event on vfs_read with
2408 arg1 and arg2, add to the command line;
2410 kprobe_event=p,vfs_read,$arg1,$arg2
2412 See also Documentation/trace/kprobetrace.rst "Kernel
2413 Boot Parameter" section.
2415 kpti= [ARM64] Control page table isolation of user
2416 and kernel address spaces.
2417 Default: enabled on cores which need mitigation.
2421 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs.
2422 Default is 0 (don't ignore, but inject #GP)
2424 kvm.eager_page_split=
2425 [KVM,X86] Controls whether or not KVM will try to
2426 proactively split all huge pages during dirty logging.
2427 Eager page splitting reduces interruptions to vCPU
2428 execution by eliminating the write-protection faults
2429 and MMU lock contention that would otherwise be
2430 required to split huge pages lazily.
2432 VM workloads that rarely perform writes or that write
2433 only to a small region of VM memory may benefit from
2434 disabling eager page splitting to allow huge pages to
2435 still be used for reads.
2437 The behavior of eager page splitting depends on whether
2438 KVM_DIRTY_LOG_INITIALLY_SET is enabled or disabled. If
2439 disabled, all huge pages in a memslot will be eagerly
2440 split when dirty logging is enabled on that memslot. If
2441 enabled, eager page splitting will be performed during
2442 the KVM_CLEAR_DIRTY ioctl, and only for the pages being
2445 Eager page splitting is only supported when kvm.tdp_mmu=Y.
2449 kvm.enable_vmware_backdoor=[KVM] Support VMware backdoor PV interface.
2450 Default is false (don't support).
2453 [KVM] Controls the software workaround for the
2454 X86_BUG_ITLB_MULTIHIT bug.
2455 force : Always deploy workaround.
2456 off : Never deploy workaround.
2457 auto : Deploy workaround based on the presence of
2458 X86_BUG_ITLB_MULTIHIT.
2462 If the software workaround is enabled for the host,
2463 guests do need not to enable it for nested guests.
2465 kvm.nx_huge_pages_recovery_ratio=
2466 [KVM] Controls how many 4KiB pages are periodically zapped
2467 back to huge pages. 0 disables the recovery, otherwise if
2468 the value is N KVM will zap 1/Nth of the 4KiB pages every
2469 period (see below). The default is 60.
2471 kvm.nx_huge_pages_recovery_period_ms=
2472 [KVM] Controls the time period at which KVM zaps 4KiB pages
2473 back to huge pages. If the value is a non-zero N, KVM will
2474 zap a portion (see ratio above) of the pages every N msecs.
2475 If the value is 0 (the default), KVM will pick a period based
2476 on the ratio, such that a page is zapped after 1 hour on average.
2478 kvm-amd.nested= [KVM,AMD] Allow nested virtualization in KVM/SVM.
2479 Default is 1 (enabled)
2481 kvm-amd.npt= [KVM,AMD] Disable nested paging (virtualized MMU)
2483 Default is 1 (enabled) if in 64-bit or 32-bit PAE mode.
2486 [KVM,ARM] Select one of KVM/arm64's modes of operation.
2488 none: Forcefully disable KVM.
2490 nvhe: Standard nVHE-based mode, without support for
2493 protected: nVHE-based mode with support for guests whose
2494 state is kept private from the host.
2496 Defaults to VHE/nVHE based on hardware support. Setting
2497 mode to "protected" will disable kexec and hibernation
2500 kvm-arm.vgic_v3_group0_trap=
2501 [KVM,ARM] Trap guest accesses to GICv3 group-0
2504 kvm-arm.vgic_v3_group1_trap=
2505 [KVM,ARM] Trap guest accesses to GICv3 group-1
2508 kvm-arm.vgic_v3_common_trap=
2509 [KVM,ARM] Trap guest accesses to GICv3 common
2512 kvm-arm.vgic_v4_enable=
2513 [KVM,ARM] Allow use of GICv4 for direct injection of
2516 kvm_cma_resv_ratio=n [PPC]
2517 Reserves given percentage from system memory area for
2518 contiguous memory allocation for KVM hash pagetable
2520 By default it reserves 5% of total system memory.
2524 kvm-intel.ept= [KVM,Intel] Disable extended page tables
2525 (virtualized MMU) support on capable Intel chips.
2526 Default is 1 (enabled)
2528 kvm-intel.emulate_invalid_guest_state=
2529 [KVM,Intel] Disable emulation of invalid guest state.
2530 Ignored if kvm-intel.enable_unrestricted_guest=1, as
2531 guest state is never invalid for unrestricted guests.
2532 This param doesn't apply to nested guests (L2), as KVM
2533 never emulates invalid L2 guest state.
2534 Default is 1 (enabled)
2536 kvm-intel.flexpriority=
2537 [KVM,Intel] Disable FlexPriority feature (TPR shadow).
2538 Default is 1 (enabled)
2541 [KVM,Intel] Enable VMX nesting (nVMX).
2542 Default is 0 (disabled)
2544 kvm-intel.unrestricted_guest=
2545 [KVM,Intel] Disable unrestricted guest feature
2546 (virtualized real and unpaged mode) on capable
2547 Intel chips. Default is 1 (enabled)
2549 kvm-intel.vmentry_l1d_flush=[KVM,Intel] Mitigation for L1 Terminal Fault
2552 Valid arguments: never, cond, always
2554 always: L1D cache flush on every VMENTER.
2555 cond: Flush L1D on VMENTER only when the code between
2556 VMEXIT and VMENTER can leak host memory.
2557 never: Disables the mitigation
2559 Default is cond (do L1 cache flush in specific instances)
2561 kvm-intel.vpid= [KVM,Intel] Disable Virtual Processor Identification
2562 feature (tagged TLBs) on capable Intel chips.
2563 Default is 1 (enabled)
2565 l1d_flush= [X86,INTEL]
2566 Control mitigation for L1D based snooping vulnerability.
2568 Certain CPUs are vulnerable to an exploit against CPU
2569 internal buffers which can forward information to a
2570 disclosure gadget under certain conditions.
2572 In vulnerable processors, the speculatively
2573 forwarded data can be used in a cache side channel
2574 attack, to access data to which the attacker does
2575 not have direct access.
2577 This parameter controls the mitigation. The
2580 on - enable the interface for the mitigation
2582 l1tf= [X86] Control mitigation of the L1TF vulnerability on
2585 The kernel PTE inversion protection is unconditionally
2586 enabled and cannot be disabled.
2589 Provides all available mitigations for the
2590 L1TF vulnerability. Disables SMT and
2591 enables all mitigations in the
2592 hypervisors, i.e. unconditional L1D flush.
2594 SMT control and L1D flush control via the
2595 sysfs interface is still possible after
2596 boot. Hypervisors will issue a warning
2597 when the first VM is started in a
2598 potentially insecure configuration,
2599 i.e. SMT enabled or L1D flush disabled.
2602 Same as 'full', but disables SMT and L1D
2603 flush runtime control. Implies the
2604 'nosmt=force' command line option.
2605 (i.e. sysfs control of SMT is disabled.)
2608 Leaves SMT enabled and enables the default
2609 hypervisor mitigation, i.e. conditional
2612 SMT control and L1D flush control via the
2613 sysfs interface is still possible after
2614 boot. Hypervisors will issue a warning
2615 when the first VM is started in a
2616 potentially insecure configuration,
2617 i.e. SMT enabled or L1D flush disabled.
2621 Disables SMT and enables the default
2622 hypervisor mitigation.
2624 SMT control and L1D flush control via the
2625 sysfs interface is still possible after
2626 boot. Hypervisors will issue a warning
2627 when the first VM is started in a
2628 potentially insecure configuration,
2629 i.e. SMT enabled or L1D flush disabled.
2632 Same as 'flush', but hypervisors will not
2633 warn when a VM is started in a potentially
2634 insecure configuration.
2637 Disables hypervisor mitigations and doesn't
2639 It also drops the swap size and available
2640 RAM limit restriction on both hypervisor and
2645 For details see: Documentation/admin-guide/hw-vuln/l1tf.rst
2651 lapic [X86-32,APIC] Enable the local APIC even if BIOS
2654 lapic= [X86,APIC] Do not use TSC deadline
2655 value for LAPIC timer one-shot implementation. Default
2656 back to the programmable timer unit in the LAPIC.
2657 Format: notscdeadline
2659 lapic_timer_c2_ok [X86,APIC] trust the local apic timer
2662 libata.dma= [LIBATA] DMA control
2663 libata.dma=0 Disable all PATA and SATA DMA
2664 libata.dma=1 PATA and SATA Disk DMA only
2665 libata.dma=2 ATAPI (CDROM) DMA only
2666 libata.dma=4 Compact Flash DMA only
2667 Combinations also work, so libata.dma=3 enables DMA
2668 for disks and CDROMs, but not CFs.
2670 libata.ignore_hpa= [LIBATA] Ignore HPA limit
2671 libata.ignore_hpa=0 keep BIOS limits (default)
2672 libata.ignore_hpa=1 ignore limits, using full disk
2674 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume
2678 libata.force= [LIBATA] Force configurations. The format is a comma-
2679 separated list of "[ID:]VAL" where ID is PORT[.DEVICE].
2680 PORT and DEVICE are decimal numbers matching port, link
2681 or device. Basically, it matches the ATA ID string
2682 printed on console by libata. If the whole ID part is
2683 omitted, the last PORT and DEVICE values are used. If
2684 ID hasn't been specified yet, the configuration applies
2685 to all ports, links and devices.
2687 If only DEVICE is omitted, the parameter applies to
2688 the port and all links and devices behind it. DEVICE
2689 number of 0 either selects the first device or the
2690 first fan-out link behind PMP device. It does not
2691 select the host link. DEVICE number of 15 selects the
2692 host link and device attached to it.
2694 The VAL specifies the configuration to force. As long
2695 as there is no ambiguity, shortcut notation is allowed.
2696 For example, both 1.5 and 1.5G would work for 1.5Gbps.
2697 The following configurations can be forced.
2699 * Cable type: 40c, 80c, short40c, unk, ign or sata.
2700 Any ID with matching PORT is used.
2702 * SATA link speed limit: 1.5Gbps or 3.0Gbps.
2704 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7].
2705 udma[/][16,25,33,44,66,100,133] notation is also
2708 * nohrst, nosrst, norst: suppress hard, soft and both
2711 * rstonce: only attempt one reset during hot-unplug
2714 * [no]dbdelay: Enable or disable the extra 200ms delay
2715 before debouncing a link PHY and device presence
2718 * [no]ncq: Turn on or off NCQ.
2720 * [no]ncqtrim: Enable or disable queued DSM TRIM.
2722 * [no]ncqati: Enable or disable NCQ trim on ATI chipset.
2724 * [no]trim: Enable or disable (unqueued) TRIM.
2726 * trim_zero: Indicate that TRIM command zeroes data.
2728 * max_trim_128m: Set 128M maximum trim size limit.
2730 * [no]dma: Turn on or off DMA transfers.
2732 * atapi_dmadir: Enable ATAPI DMADIR bridge support.
2734 * atapi_mod16_dma: Enable the use of ATAPI DMA for
2735 commands that are not a multiple of 16 bytes.
2737 * [no]dmalog: Enable or disable the use of the
2738 READ LOG DMA EXT command to access logs.
2740 * [no]iddevlog: Enable or disable access to the
2741 identify device data log.
2743 * [no]logdir: Enable or disable access to the general
2744 purpose log directory.
2746 * max_sec_128: Set transfer size limit to 128 sectors.
2748 * max_sec_1024: Set or clear transfer size limit to
2751 * max_sec_lba48: Set or clear transfer size limit to
2754 * [no]lpm: Enable or disable link power management.
2756 * [no]setxfer: Indicate if transfer speed mode setting
2759 * dump_id: Dump IDENTIFY data.
2761 * disable: Disable this device.
2763 If there are multiple matching configurations changing
2764 the same attribute, the last one is used.
2766 load_ramdisk= [RAM] [Deprecated]
2768 lockd.nlm_grace_period=P [NFS] Assign grace period.
2771 lockd.nlm_tcpport=N [NFS] Assign TCP port.
2774 lockd.nlm_timeout=T [NFS] Assign timeout value.
2777 lockd.nlm_udpport=M [NFS] Assign UDP port.
2780 lockdown= [SECURITY]
2781 { integrity | confidentiality }
2782 Enable the kernel lockdown feature. If set to
2783 integrity, kernel features that allow userland to
2784 modify the running kernel are disabled. If set to
2785 confidentiality, kernel features that allow userland
2786 to extract confidential information from the kernel
2789 locktorture.nreaders_stress= [KNL]
2790 Set the number of locking read-acquisition kthreads.
2791 Defaults to being automatically set based on the
2792 number of online CPUs.
2794 locktorture.nwriters_stress= [KNL]
2795 Set the number of locking write-acquisition kthreads.
2797 locktorture.onoff_holdoff= [KNL]
2798 Set time (s) after boot for CPU-hotplug testing.
2800 locktorture.onoff_interval= [KNL]
2801 Set time (s) between CPU-hotplug operations, or
2802 zero to disable CPU-hotplug testing.
2804 locktorture.shuffle_interval= [KNL]
2805 Set task-shuffle interval (jiffies). Shuffling
2806 tasks allows some CPUs to go into dyntick-idle
2807 mode during the locktorture test.
2809 locktorture.shutdown_secs= [KNL]
2810 Set time (s) after boot system shutdown. This
2811 is useful for hands-off automated testing.
2813 locktorture.stat_interval= [KNL]
2814 Time (s) between statistics printk()s.
2816 locktorture.stutter= [KNL]
2817 Time (s) to stutter testing, for example,
2818 specifying five seconds causes the test to run for
2819 five seconds, wait for five seconds, and so on.
2820 This tests the locking primitive's ability to
2821 transition abruptly to and from idle.
2823 locktorture.torture_type= [KNL]
2824 Specify the locking implementation to test.
2826 locktorture.verbose= [KNL]
2827 Enable additional printk() statements.
2829 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver
2832 loglevel= All Kernel Messages with a loglevel smaller than the
2833 console loglevel will be printed to the console. It can
2834 also be changed with klogd or other programs. The
2835 loglevels are defined as follows:
2837 0 (KERN_EMERG) system is unusable
2838 1 (KERN_ALERT) action must be taken immediately
2839 2 (KERN_CRIT) critical conditions
2840 3 (KERN_ERR) error conditions
2841 4 (KERN_WARNING) warning conditions
2842 5 (KERN_NOTICE) normal but significant condition
2843 6 (KERN_INFO) informational
2844 7 (KERN_DEBUG) debug-level messages
2846 log_buf_len=n[KMG] Sets the size of the printk ring buffer,
2847 in bytes. n must be a power of two and greater
2848 than the minimal size. The minimal size is defined
2849 by LOG_BUF_SHIFT kernel config parameter. There is
2850 also CONFIG_LOG_CPU_MAX_BUF_SHIFT config parameter
2851 that allows to increase the default size depending on
2852 the number of CPUs. See init/Kconfig for more details.
2854 logo.nologo [FB] Disables display of the built-in Linux logo.
2855 This may be used to provide more screen space for
2856 kernel log messages and is useful when debugging
2857 kernel boot problems.
2859 lp=0 [LP] Specify parallel ports to use, e.g,
2860 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses
2861 lp=reset first parallel port). 'lp=0' disables the
2862 lp=auto printer driver. 'lp=reset' (which can be
2863 specified in addition to the ports) causes
2864 attached printers to be reset. Using
2865 lp=port1,port2,... specifies the parallel ports
2866 to associate lp devices with, starting with
2867 lp0. A port specification may be 'none' to skip
2868 that lp device, or a parport name such as
2869 'parport0'. Specifying 'lp=auto' instead of a
2870 port specification list means that device IDs
2871 from each port should be examined, to see if
2872 an IEEE 1284-compliant printer is attached; if
2873 so, the driver will manage that printer.
2874 See also header of drivers/char/lp.c.
2877 Sets loops_per_jiffy to given constant, thus avoiding
2878 time-consuming boot-time autodetection (up to 250 ms per
2879 CPU). 0 enables autodetection (default). To determine
2880 the correct value for your kernel, boot with normal
2881 autodetection and see what value is printed. Note that
2882 on SMP systems the preset will be applied to all CPUs,
2883 which is likely to cause problems if your CPUs need
2884 significantly divergent settings. An incorrect value
2885 will cause delays in the kernel to be wrong, leading to
2886 unpredictable I/O errors and other breakage. Although
2887 unlikely, in the extreme case this might damage your
2891 Format: <io>,<irq>,<dma>
2893 lsm.debug [SECURITY] Enable LSM initialization debugging output.
2896 [SECURITY] Choose order of LSM initialization. This
2897 overrides CONFIG_LSM, and the "security=" parameter.
2899 machvec= [IA-64] Force the use of a particular machine-vector
2900 (machvec) in a generic kernel.
2901 Example: machvec=hpzx1
2903 machtype= [Loongson] Share the same kernel image file between
2904 different yeeloong laptops.
2905 Example: machtype=lemote-yeeloong-2f-7inch
2907 max_addr=nn[KMG] [KNL,BOOT,IA-64] All physical memory greater
2908 than or equal to this physical address is ignored.
2910 maxcpus= [SMP] Maximum number of processors that an SMP kernel
2911 will bring up during bootup. maxcpus=n : n >= 0 limits
2912 the kernel to bring up 'n' processors. Surely after
2913 bootup you can bring up the other plugged cpu by executing
2914 "echo 1 > /sys/devices/system/cpu/cpuX/online". So maxcpus
2915 only takes effect during system bootup.
2916 While n=0 is a special case, it is equivalent to "nosmp",
2917 which also disables the IO APIC.
2919 max_loop= [LOOP] The number of loop block devices that get
2920 (loop.max_loop) unconditionally pre-created at init time. The default
2921 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead
2922 of statically allocating a predefined number, loop
2923 devices can be requested on-demand with the
2924 /dev/loop-control interface.
2926 mce [X86-32] Machine Check Exception
2928 mce=option [X86-64] See Documentation/x86/x86_64/boot-options.rst
2930 md= [HW] RAID subsystems devices and level
2931 See Documentation/admin-guide/md.rst.
2934 Format: <first>,<last>
2935 Specifies range of consoles to be captured by the MDA.
2938 Control mitigation for the Micro-architectural Data
2939 Sampling (MDS) vulnerability.
2941 Certain CPUs are vulnerable to an exploit against CPU
2942 internal buffers which can forward information to a
2943 disclosure gadget under certain conditions.
2945 In vulnerable processors, the speculatively
2946 forwarded data can be used in a cache side channel
2947 attack, to access data to which the attacker does
2948 not have direct access.
2950 This parameter controls the MDS mitigation. The
2953 full - Enable MDS mitigation on vulnerable CPUs
2954 full,nosmt - Enable MDS mitigation and disable
2955 SMT on vulnerable CPUs
2956 off - Unconditionally disable MDS mitigation
2958 On TAA-affected machines, mds=off can be prevented by
2959 an active TAA mitigation as both vulnerabilities are
2960 mitigated with the same mechanism so in order to disable
2961 this mitigation, you need to specify tsx_async_abort=off
2964 Not specifying this option is equivalent to
2967 For details see: Documentation/admin-guide/hw-vuln/mds.rst
2969 mem=nn[KMG] [HEXAGON] Set the memory size.
2970 Must be specified, otherwise memory size will be 0.
2972 mem=nn[KMG] [KNL,BOOT] Force usage of a specific amount of memory
2973 Amount of memory to be used in cases as follows:
2976 2 when the kernel is not able to see the whole system memory;
2977 3 memory that lies after 'mem=' boundary is excluded from
2978 the hypervisor, then assigned to KVM guests.
2979 4 to limit the memory available for kdump kernel.
2981 [ARC,MICROBLAZE] - the limit applies only to low memory,
2982 high memory is not affected.
2984 [ARM64] - only limits memory covered by the linear
2985 mapping. The NOMAP regions are not affected.
2987 [X86] Work as limiting max address. Use together
2988 with memmap= to avoid physical address space collisions.
2989 Without memmap= PCI devices could be placed at addresses
2990 belonging to unused RAM.
2992 Note that this only takes effects during boot time since
2993 in above case 3, memory may need be hot added after boot
2994 if system memory of hypervisor is not sufficient.
2997 [ARM,MIPS] - override the memory layout reported by
2999 Define a memory region of size nn[KMG] starting at
3001 Multiple different regions can be specified with
3002 multiple mem= parameters on the command line.
3004 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel
3007 memblock=debug [KNL] Enable memblock debug messages.
3010 [KNL,SH] Allow user to override the default size for
3011 per-device physically contiguous DMA buffers.
3013 memhp_default_state=online/offline
3014 [KNL] Set the initial state for the memory hotplug
3015 onlining policy. If not specified, the default value is
3016 set according to the
3017 CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE kernel config
3019 See Documentation/admin-guide/mm/memory-hotplug.rst.
3021 memmap=exactmap [KNL,X86] Enable setting of an exact
3022 E820 memory map, as specified by the user.
3023 Such memmap=exactmap lines can be constructed based on
3024 BIOS output or other requirements. See the memmap=nn@ss
3027 memmap=nn[KMG]@ss[KMG]
3028 [KNL, X86, MIPS, XTENSA] Force usage of a specific region of memory.
3029 Region of memory to be used is from ss to ss+nn.
3030 If @ss[KMG] is omitted, it is equivalent to mem=nn[KMG],
3031 which limits max address to nn[KMG].
3032 Multiple different regions can be specified,
3035 memmap=100M@2G,100M#3G,1G!1024G
3037 memmap=nn[KMG]#ss[KMG]
3038 [KNL,ACPI] Mark specific memory as ACPI data.
3039 Region of memory to be marked is from ss to ss+nn.
3041 memmap=nn[KMG]$ss[KMG]
3042 [KNL,ACPI] Mark specific memory as reserved.
3043 Region of memory to be reserved is from ss to ss+nn.
3044 Example: Exclude memory from 0x18690000-0x1869ffff
3045 memmap=64K$0x18690000
3047 memmap=0x10000$0x18690000
3048 Some bootloaders may need an escape character before '$',
3049 like Grub2, otherwise '$' and the following number
3052 memmap=nn[KMG]!ss[KMG]
3053 [KNL,X86] Mark specific memory as protected.
3054 Region of memory to be used, from ss to ss+nn.
3055 The memory region may be marked as e820 type 12 (0xc)
3056 and is NVDIMM or ADR memory.
3058 memmap=<size>%<offset>-<oldtype>+<newtype>
3059 [KNL,ACPI] Convert memory within the specified region
3060 from <oldtype> to <newtype>. If "-<oldtype>" is left
3061 out, the whole region will be marked as <newtype>,
3062 even if previously unavailable. If "+<newtype>" is left
3063 out, matching memory will be removed. Types are
3064 specified as e820 types, e.g., 1 = RAM, 2 = reserved,
3065 3 = ACPI, 12 = PRAM.
3067 memory_corruption_check=0/1 [X86]
3068 Some BIOSes seem to corrupt the first 64k of
3069 memory when doing things like suspend/resume.
3070 Setting this option will scan the memory
3071 looking for corruption. Enabling this will
3072 both detect corruption and prevent the kernel
3073 from using the memory being corrupted.
3074 However, its intended as a diagnostic tool; if
3075 repeatable BIOS-originated corruption always
3076 affects the same memory, you can use memmap=
3077 to prevent the kernel from using that memory.
3079 memory_corruption_check_size=size [X86]
3080 By default it checks for corruption in the low
3081 64k, making this memory unavailable for normal
3082 use. Use this parameter to scan for
3083 corruption in more or less memory.
3085 memory_corruption_check_period=seconds [X86]
3086 By default it checks for corruption every 60
3087 seconds. Use this parameter to check at some
3088 other rate. 0 disables periodic checking.
3090 memory_hotplug.memmap_on_memory
3091 [KNL,X86,ARM] Boolean flag to enable this feature.
3092 Format: {on | off (default)}
3093 When enabled, runtime hotplugged memory will
3094 allocate its internal metadata (struct pages,
3095 those vmemmap pages cannot be optimized even
3096 if hugetlb_free_vmemmap is enabled) from the
3097 hotadded memory which will allow to hotadd a
3098 lot of memory without requiring additional
3100 This feature is disabled by default because it
3101 has some implication on large (e.g. GB)
3102 allocations in some configurations (e.g. small
3104 The state of the flag can be read in
3105 /sys/module/memory_hotplug/parameters/memmap_on_memory.
3106 Note that even when enabled, there are a few cases where
3107 the feature is not effective.
3109 memtest= [KNL,X86,ARM,M68K,PPC,RISCV] Enable memtest
3111 default : 0 <disable>
3112 Specifies the number of memtest passes to be
3113 performed. Each pass selects another test
3114 pattern from a given set of patterns. Memtest
3115 fills the memory with this pattern, validates
3116 memory contents and reserves bad memory
3117 regions that are detected.
3119 mem_encrypt= [X86-64] AMD Secure Memory Encryption (SME) control
3120 Valid arguments: on, off
3121 Default (depends on kernel configuration option):
3122 on (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=y)
3123 off (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=n)
3124 mem_encrypt=on: Activate SME
3125 mem_encrypt=off: Do not activate SME
3127 Refer to Documentation/virt/kvm/x86/amd-memory-encryption.rst
3128 for details on when memory encryption can be activated.
3130 mem_sleep_default= [SUSPEND] Default system suspend mode:
3131 s2idle - Suspend-To-Idle
3132 shallow - Power-On Suspend or equivalent (if supported)
3133 deep - Suspend-To-RAM or equivalent (if supported)
3134 See Documentation/admin-guide/pm/sleep-states.rst.
3136 meye.*= [HW] Set MotionEye Camera parameters
3137 See Documentation/admin-guide/media/meye.rst.
3139 mfgpt_irq= [IA-32] Specify the IRQ to use for the
3140 Multi-Function General Purpose Timers on AMD Geode
3143 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when
3144 the BIOS has incorrectly applied a workaround. TinyBIOS
3145 version 0.98 is known to be affected, 0.99 fixes the
3146 problem by letting the user disable the workaround.
3150 min_addr=nn[KMG] [KNL,BOOT,IA-64] All physical memory below this
3151 physical address is ignored.
3153 mini2440= [ARM,HW,KNL]
3154 Format:[0..2][b][c][t]
3156 MINI2440 configuration specification:
3157 0 - The attached screen is the 3.5" TFT
3158 1 - The attached screen is the 7" TFT
3159 2 - The VGA Shield is attached (1024x768)
3160 Leaving out the screen size parameter will not load
3161 the TFT driver, and the framebuffer will be left
3163 b - Enable backlight. The TFT backlight pin will be
3164 linked to the kernel VESA blanking code and a GPIO
3165 LED. This parameter is not necessary when using the
3167 c - Enable the s3c camera interface.
3168 t - Reserved for enabling touchscreen support. The
3169 touchscreen support is not enabled in the mainstream
3170 kernel as of 2.6.30, a preliminary port can be found
3171 in the "bleeding edge" mini2440 support kernel at
3172 https://repo.or.cz/w/linux-2.6/mini2440.git
3175 [X86,PPC,S390,ARM64] Control optional mitigations for
3176 CPU vulnerabilities. This is a set of curated,
3177 arch-independent options, each of which is an
3178 aggregation of existing arch-specific options.
3181 Disable all optional CPU mitigations. This
3182 improves system performance, but it may also
3183 expose users to several CPU vulnerabilities.
3184 Equivalent to: nopti [X86,PPC]
3185 if nokaslr then kpti=0 [ARM64]
3186 nospectre_v1 [X86,PPC]
3188 nospectre_v2 [X86,PPC,S390,ARM64]
3189 spectre_v2_user=off [X86]
3190 spec_store_bypass_disable=off [X86,PPC]
3191 ssbd=force-off [ARM64]
3194 tsx_async_abort=off [X86]
3195 kvm.nx_huge_pages=off [X86]
3196 srbds=off [X86,INTEL]
3197 no_entry_flush [PPC]
3198 no_uaccess_flush [PPC]
3199 mmio_stale_data=off [X86]
3203 This does not have any effect on
3204 kvm.nx_huge_pages when
3205 kvm.nx_huge_pages=force.
3208 Mitigate all CPU vulnerabilities, but leave SMT
3209 enabled, even if it's vulnerable. This is for
3210 users who don't want to be surprised by SMT
3211 getting disabled across kernel upgrades, or who
3212 have other ways of avoiding SMT-based attacks.
3213 Equivalent to: (default behavior)
3216 Mitigate all CPU vulnerabilities, disabling SMT
3217 if needed. This is for users who always want to
3218 be fully mitigated, even if it means losing SMT.
3219 Equivalent to: l1tf=flush,nosmt [X86]
3220 mds=full,nosmt [X86]
3221 tsx_async_abort=full,nosmt [X86]
3222 mmio_stale_data=full,nosmt [X86]
3223 retbleed=auto,nosmt [X86]
3226 [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this
3227 parameter allows control of the logging verbosity for
3228 the additional memory initialisation checks. A value
3229 of 0 disables mminit logging and a level of 4 will
3230 log everything. Information is printed at KERN_DEBUG
3231 so loglevel=8 may also need to be specified.
3234 [X86,INTEL] Control mitigation for the Processor
3235 MMIO Stale Data vulnerabilities.
3237 Processor MMIO Stale Data is a class of
3238 vulnerabilities that may expose data after an MMIO
3239 operation. Exposed data could originate or end in
3240 the same CPU buffers as affected by MDS and TAA.
3241 Therefore, similar to MDS and TAA, the mitigation
3242 is to clear the affected CPU buffers.
3244 This parameter controls the mitigation. The
3247 full - Enable mitigation on vulnerable CPUs
3249 full,nosmt - Enable mitigation and disable SMT on
3252 off - Unconditionally disable mitigation
3254 On MDS or TAA affected machines,
3255 mmio_stale_data=off can be prevented by an active
3256 MDS or TAA mitigation as these vulnerabilities are
3257 mitigated with the same mechanism so in order to
3258 disable this mitigation, you need to specify
3259 mds=off and tsx_async_abort=off too.
3261 Not specifying this option is equivalent to
3262 mmio_stale_data=full.
3265 Documentation/admin-guide/hw-vuln/processor_mmio_stale_data.rst
3268 [KNL] When CONFIG_MODULE_SIG is set, this means that
3269 modules without (valid) signatures will fail to load.
3270 Note that if CONFIG_MODULE_SIG_FORCE is set, that
3271 is always true, so this option does nothing.
3273 module_blacklist= [KNL] Do not load a comma-separated list of
3274 modules. Useful for debugging problem modules.
3277 [MOUSE] Maximum time between finger touching and
3278 leaving touchpad surface for touch to be considered
3279 a tap and be reported as a left button click (for
3280 touchpads working in absolute mode only).
3282 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices
3283 reporting absolute coordinates, such as tablets
3284 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices
3285 reporting absolute coordinates, such as tablets
3287 movablecore= [KNL,X86,IA-64,PPC]
3288 Format: nn[KMGTPE] | nn%
3289 This parameter is the complement to kernelcore=, it
3290 specifies the amount of memory used for migratable
3291 allocations. If both kernelcore and movablecore is
3292 specified, then kernelcore will be at *least* the
3293 specified value but may be more. If movablecore on its
3294 own is specified, the administrator must be careful
3295 that the amount of memory usable for all allocations
3298 movable_node [KNL] Boot-time switch to make hotplugable memory
3299 NUMA nodes to be movable. This means that the memory
3300 of such nodes will be usable only for movable
3301 allocations which rules out almost all kernel
3302 allocations. Use with caution!
3304 MTD_Partition= [MTD]
3305 Format: <name>,<region-number>,<size>,<offset>
3307 MTD_Region= [MTD] Format:
3308 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>]
3311 See drivers/mtd/parsers/cmdlinepart.c
3314 ARM/S3C2412 JIVE boot control
3316 See arch/arm/mach-s3c/mach-jive.c
3318 mtouchusb.raw_coordinates=
3319 [HW] Make the MicroTouch USB driver use raw coordinates
3320 ('y', default) or cooked coordinates ('n')
3322 mtrr_chunk_size=nn[KMG] [X86]
3323 used for mtrr cleanup. It is largest continuous chunk
3324 that could hold holes aka. UC entries.
3326 mtrr_gran_size=nn[KMG] [X86]
3327 Used for mtrr cleanup. It is granularity of mtrr block.
3329 Large value could prevent small alignment from
3332 mtrr_spare_reg_nr=n [X86]
3334 Range: 0,7 : spare reg number
3336 Used for mtrr cleanup. It is spare mtrr entries number.
3337 Set to 2 or more if your graphical card needs more.
3339 multitce=off [PPC] This parameter disables the use of the pSeries
3340 firmware feature for updating multiple TCE entries
3343 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card
3345 netdev= [NET] Network devices parameters
3346 Format: <irq>,<io>,<mem_start>,<mem_end>,<name>
3347 Note that mem_start is often overloaded to mean
3348 something different and driver-specific.
3349 This usage is only documented in each driver source
3352 netpoll.carrier_timeout=
3353 [NET] Specifies amount of time (in seconds) that
3354 netpoll should wait for a carrier. By default netpoll
3358 [NETFILTER] Enable connection tracking flow accounting
3359 0 to disable accounting
3360 1 to enable accounting
3363 nfsaddrs= [NFS] Deprecated. Use ip= instead.
3364 See Documentation/admin-guide/nfs/nfsroot.rst.
3366 nfsroot= [NFS] nfs root filesystem for disk-less boxes.
3367 See Documentation/admin-guide/nfs/nfsroot.rst.
3369 nfsrootdebug [NFS] enable nfsroot debugging messages.
3370 See Documentation/admin-guide/nfs/nfsroot.rst.
3372 nfs.callback_nr_threads=
3373 [NFSv4] set the total number of threads that the
3374 NFS client will assign to service NFSv4 callback
3377 nfs.callback_tcpport=
3378 [NFS] set the TCP port on which the NFSv4 callback
3379 channel should listen.
3382 [NFS] sets the pathname to the program which is used
3383 to update the NFS client cache entries.
3385 nfs.cache_getent_timeout=
3386 [NFS] sets the timeout after which an attempt to
3387 update a cache entry is deemed to have failed.
3389 nfs.idmap_cache_timeout=
3390 [NFS] set the maximum lifetime for idmapper cache
3394 [NFS] enable 64-bit inode numbers.
3395 If zero, the NFS client will fake up a 32-bit inode
3396 number for the readdir() and stat() syscalls instead
3397 of returning the full 64-bit number.
3398 The default is to return 64-bit inode numbers.
3400 nfs.max_session_cb_slots=
3401 [NFSv4.1] Sets the maximum number of session
3402 slots the client will assign to the callback
3403 channel. This determines the maximum number of
3404 callbacks the client will process in parallel for
3405 a particular server.
3407 nfs.max_session_slots=
3408 [NFSv4.1] Sets the maximum number of session slots
3409 the client will attempt to negotiate with the server.
3410 This limits the number of simultaneous RPC requests
3411 that the client can send to the NFSv4.1 server.
3412 Note that there is little point in setting this
3413 value higher than the max_tcp_slot_table_limit.
3415 nfs.nfs4_disable_idmapping=
3416 [NFSv4] When set to the default of '1', this option
3417 ensures that both the RPC level authentication
3418 scheme and the NFS level operations agree to use
3419 numeric uids/gids if the mount is using the
3420 'sec=sys' security flavour. In effect it is
3421 disabling idmapping, which can make migration from
3422 legacy NFSv2/v3 systems to NFSv4 easier.
3423 Servers that do not support this mode of operation
3424 will be autodetected by the client, and it will fall
3425 back to using the idmapper.
3426 To turn off this behaviour, set the value to '0'.
3428 [NFS4] Specify an additional fixed unique ident-
3429 ification string that NFSv4 clients can insert into
3430 their nfs_client_id4 string. This is typically a
3431 UUID that is generated at system install time.
3433 nfs.send_implementation_id =
3434 [NFSv4.1] Send client implementation identification
3435 information in exchange_id requests.
3436 If zero, no implementation identification information
3438 The default is to send the implementation identification
3441 nfs.recover_lost_locks =
3442 [NFSv4] Attempt to recover locks that were lost due
3443 to a lease timeout on the server. Please note that
3444 doing this risks data corruption, since there are
3445 no guarantees that the file will remain unchanged
3446 after the locks are lost.
3447 If you want to enable the kernel legacy behaviour of
3448 attempting to recover these locks, then set this
3450 The default parameter value of '0' causes the kernel
3451 not to attempt recovery of lost locks.
3453 nfs4.layoutstats_timer =
3454 [NFSv4.2] Change the rate at which the kernel sends
3455 layoutstats to the pNFS metadata server.
3457 Setting this to value to 0 causes the kernel to use
3458 whatever value is the default set by the layout
3459 driver. A non-zero value sets the minimum interval
3460 in seconds between layoutstats transmissions.
3462 nfsd.inter_copy_offload_enable =
3463 [NFSv4.2] When set to 1, the server will support
3464 server-to-server copies for which this server is
3465 the destination of the copy.
3467 nfsd.nfsd4_ssc_umount_timeout =
3468 [NFSv4.2] When used as the destination of a
3469 server-to-server copy, knfsd temporarily mounts
3470 the source server. It caches the mount in case
3471 it will be needed again, and discards it if not
3472 used for the number of milliseconds specified by
3475 nfsd.nfs4_disable_idmapping=
3476 [NFSv4] When set to the default of '1', the NFSv4
3477 server will return only numeric uids and gids to
3478 clients using auth_sys, and will accept numeric uids
3479 and gids from such clients. This is intended to ease
3480 migration from NFSv2/v3.
3483 nmi_backtrace.backtrace_idle [KNL]
3484 Dump stacks even of idle CPUs in response to an
3485 NMI stack-backtrace request.
3487 nmi_debug= [KNL,SH] Specify one or more actions to take
3488 when a NMI is triggered.
3489 Format: [state][,regs][,debounce][,die]
3491 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels
3492 Format: [panic,][nopanic,][num]
3494 0 - turn hardlockup detector in nmi_watchdog off
3495 1 - turn hardlockup detector in nmi_watchdog on
3496 When panic is specified, panic when an NMI watchdog
3497 timeout occurs (or 'nopanic' to not panic on an NMI
3498 watchdog, if CONFIG_BOOTPARAM_HARDLOCKUP_PANIC is set)
3499 To disable both hard and soft lockup detectors,
3500 please see 'nowatchdog'.
3501 This is useful when you use a panic=... timeout and
3502 need the box quickly up again.
3504 These settings can be accessed at runtime via
3505 the nmi_watchdog and hardlockup_panic sysctls.
3507 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths
3508 emulation library even if a 387 maths coprocessor
3511 no5lvl [X86-64] Disable 5-level paging mode. Forces
3512 kernel to use 4-level paging instead.
3514 nofsgsbase [X86] Disables FSGSBASE instructions.
3517 [HW] Never suspend the console
3518 Disable suspending of consoles during suspend and
3519 hibernate operations. Once disabled, debugging
3520 messages can reach various consoles while the rest
3521 of the system is being put to sleep (ie, while
3522 debugging driver suspend/resume hooks). This may
3523 not work reliably with all consoles, but is known
3524 to work with serial and VGA consoles.
3525 To facilitate more flexible debugging, we also add
3526 console_suspend, a printk module parameter to control
3527 it. Users could use console_suspend (usually
3528 /sys/module/printk/parameters/console_suspend) to
3529 turn on/off it dynamically.
3531 novmcoredd [KNL,KDUMP]
3532 Disable device dump. Device dump allows drivers to
3533 append dump data to vmcore so you can collect driver
3534 specified debug info. Drivers can append the data
3535 without any limit and this data is stored in memory,
3536 so this may cause significant memory stress. Disabling
3537 device dump can help save memory but the driver debug
3538 data will be no longer available. This parameter
3539 is only available when CONFIG_PROC_VMCORE_DEVICE_DUMP
3542 noaliencache [MM, NUMA, SLAB] Disables the allocation of alien
3543 caches in the slab allocator. Saves per-node memory,
3544 but will impact performance.
3548 noaltinstr [S390] Disables alternative instructions patching
3549 (CPU alternatives feature).
3551 noapic [SMP,APIC] Tells the kernel to not make use of any
3552 IOAPICs that may be present in the system.
3554 noautogroup Disable scheduler automatic task group creation.
3556 nobats [PPC] Do not use BATs for mapping kernel lowmem
3557 on "Classic" PPC cores.
3561 nodsp [SH] Disable hardware DSP at boot time.
3563 noefi Disable EFI runtime services support.
3565 no_entry_flush [PPC] Don't flush the L1-D cache when entering the kernel.
3570 Disable SMAP (Supervisor Mode Access Prevention)
3571 even if it is supported by processor.
3574 Disable SMEP (Supervisor Mode Execution Prevention)
3575 even if it is supported by processor.
3578 This affects only 32-bit executables.
3579 noexec32=on: enable non-executable mappings (default)
3580 read doesn't imply executable mappings
3581 noexec32=off: disable non-executable mappings
3582 read implies executable mappings
3584 nofpu [MIPS,SH] Disable hardware FPU at boot time.
3586 nofxsr [BUGS=X86-32] Disables x86 floating point extended
3587 register save and restore. The kernel will only save
3588 legacy floating-point registers on task switch.
3590 nohugeiomap [KNL,X86,PPC,ARM64] Disable kernel huge I/O mappings.
3592 nohugevmalloc [PPC] Disable kernel huge vmalloc mappings.
3594 nosmt [KNL,S390] Disable symmetric multithreading (SMT).
3595 Equivalent to smt=1.
3597 [KNL,X86] Disable symmetric multithreading (SMT).
3598 nosmt=force: Force disable SMT, cannot be undone
3599 via the sysfs control file.
3601 nospectre_v1 [X86,PPC] Disable mitigations for Spectre Variant 1
3602 (bounds check bypass). With this option data leaks are
3603 possible in the system.
3605 nospectre_v2 [X86,PPC_FSL_BOOK3E,ARM64] Disable all mitigations for
3606 the Spectre variant 2 (indirect branch prediction)
3607 vulnerability. System may allow data leaks with this
3610 nospec_store_bypass_disable
3611 [HW] Disable all mitigations for the Speculative Store Bypass vulnerability
3614 [PPC] Don't flush the L1-D cache after accessing user data.
3616 noxsave [BUGS=X86] Disables x86 extended register state save
3617 and restore using xsave. The kernel will fallback to
3618 enabling legacy floating-point and sse state.
3620 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended
3621 register states. The kernel will fall back to use
3622 xsave to save the states. By using this parameter,
3623 performance of saving the states is degraded because
3624 xsave doesn't support modified optimization while
3625 xsaveopt supports it on xsaveopt enabled systems.
3627 noxsaves [X86] Disables xsaves and xrstors used in saving and
3628 restoring x86 extended register state in compacted
3629 form of xsave area. The kernel will fall back to use
3630 xsaveopt and xrstor to save and restore the states
3631 in standard form of xsave area. By using this
3632 parameter, xsave area per process might occupy more
3633 memory on xsaves enabled systems.
3635 nohlt [ARM,ARM64,MICROBLAZE,SH] Forces the kernel to busy wait
3636 in do_idle() and not use the arch_cpu_idle()
3637 implementation; requires CONFIG_GENERIC_IDLE_POLL_SETUP
3638 to be effective. This is useful on platforms where the
3639 sleep(SH) or wfi(ARM,ARM64) instructions do not work
3640 correctly or when doing power measurements to evalute
3641 the impact of the sleep instructions. This is also
3642 useful when using JTAG debugger.
3644 no_file_caps Tells the kernel not to honor file capabilities. The
3645 only way then for a file to be executed with privilege
3646 is to be setuid root or executed by root.
3648 nohalt [IA-64] Tells the kernel not to use the power saving
3649 function PAL_HALT_LIGHT when idle. This increases
3650 power-consumption. On the positive side, it reduces
3651 interrupt wake-up latency, which may improve performance
3652 in certain environments such as networked servers or
3656 Force pointers printed to the console or buffers to be
3657 unhashed. By default, when a pointer is printed via %p
3658 format string, that pointer is "hashed", i.e. obscured
3659 by hashing the pointer value. This is a security feature
3660 that hides actual kernel addresses from unprivileged
3661 users, but it also makes debugging the kernel more
3662 difficult since unequal pointers can no longer be
3663 compared. However, if this command-line option is
3664 specified, then all normal pointers will have their true
3665 value printed. This option should only be specified when
3666 debugging the kernel. Please do not use on production
3669 nohibernate [HIBERNATION] Disable hibernation and resume.
3671 nohz= [KNL] Boottime enable/disable dynamic ticks
3672 Valid arguments: on, off
3675 nohz_full= [KNL,BOOT,SMP,ISOL]
3676 The argument is a cpu list, as described above.
3677 In kernels built with CONFIG_NO_HZ_FULL=y, set
3678 the specified list of CPUs whose tick will be stopped
3679 whenever possible. The boot CPU will be forced outside
3680 the range to maintain the timekeeping. Any CPUs
3681 in this list will have their RCU callbacks offloaded,
3682 just as if they had also been called out in the
3683 rcu_nocbs= boot parameter.
3685 Note that this argument takes precedence over
3686 the CONFIG_RCU_NOCB_CPU_DEFAULT_ALL option.
3688 noiotrap [SH] Disables trapped I/O port accesses.
3690 noirqdebug [X86-32] Disables the code which attempts to detect and
3691 disable unhandled interrupt sources.
3693 no_timer_check [X86,APIC] Disables the code which tests for
3694 broken timer IRQ sources.
3696 noisapnp [ISAPNP] Disables ISA PnP code.
3698 noinitrd [RAM] Tells the kernel not to load any configured
3701 nointremap [X86-64, Intel-IOMMU] Do not enable interrupt
3703 [Deprecated - use intremap=off]
3707 noinvpcid [X86] Disable the INVPCID cpu feature.
3709 nojitter [IA-64] Disables jitter checking for ITC timers.
3711 no-kvmclock [X86,KVM] Disable paravirtualized KVM clock driver
3713 no-kvmapf [X86,KVM] Disable paravirtualized asynchronous page
3717 [X86,PV_OPS] Disable paravirtualized VMware scheduler
3718 clock and use the default one.
3720 no-steal-acc [X86,PV_OPS,ARM64] Disable paravirtualized steal time
3721 accounting. steal time is computed, but won't
3722 influence scheduler behaviour
3724 nolapic [X86-32,APIC] Do not enable or use the local APIC.
3726 nolapic_timer [X86-32,APIC] Do not use the local APIC timer.
3728 noltlbs [PPC] Do not use large page/tlb entries for kernel
3729 lowmem mapping on PPC40x and PPC8xx
3731 nomca [IA-64] Disable machine check abort handling
3733 nomce [X86-32] Disable Machine Check Exception
3735 nomfgpt [X86-32] Disable Multi-Function General Purpose
3736 Timer usage (for AMD Geode machines).
3738 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to
3739 shutdown the other cpus. Instead use the REBOOT_VECTOR
3742 nomodeset Disable kernel modesetting. DRM drivers will not perform
3743 display-mode changes or accelerated rendering. Only the
3744 system framebuffer will be available for use if this was
3745 set-up by the firmware or boot loader.
3747 Useful as fallback, or for testing and debugging.
3749 nomodule Disable module load
3751 nopat [X86] Disable PAT (page attribute table extension of
3752 pagetables) support.
3754 nopcid [X86-64] Disable the PCID cpu feature.
3756 norandmaps Don't use address space randomization. Equivalent to
3757 echo 0 > /proc/sys/kernel/randomize_va_space
3759 noreplace-smp [X86-32,SMP] Don't replace SMP instructions
3760 with UP alternatives
3762 noresume [SWSUSP] Disables resume and restores original swap
3765 no-scroll [VGA] Disables scrollback.
3766 This is required for the Braillex ib80-piezo Braille
3767 reader made by F.H. Papenmeier (Germany).
3771 nosgx [X86-64,SGX] Disables Intel SGX kernel support.
3773 nosmp [SMP] Tells an SMP kernel to act as a UP kernel,
3774 and disable the IO APIC. legacy for "maxcpus=0".
3776 nosoftlockup [KNL] Disable the soft-lockup detector.
3778 nosync [HW,M68K] Disables sync negotiation for all devices.
3780 nowatchdog [KNL] Disable both lockup detectors, i.e.
3781 soft-lockup and NMI watchdog (hard-lockup).
3785 nox2apic [X86-64,APIC] Do not enable x2APIC mode.
3787 nps_mtm_hs_ctr= [KNL,ARC]
3788 This parameter sets the maximum duration, in
3789 cycles, each HW thread of the CTOP can run
3790 without interruptions, before HW switches it.
3791 The actual maximum duration is 16 times this
3793 Format: integer between 1 and 255
3796 nptcg= [IA-64] Override max number of concurrent global TLB
3797 purges which is reported from either PAL_VM_SUMMARY or
3800 nr_cpus= [SMP] Maximum number of processors that an SMP kernel
3801 could support. nr_cpus=n : n >= 1 limits the kernel to
3802 support 'n' processors. It could be larger than the
3803 number of already plugged CPU during bootup, later in
3804 runtime you can physically add extra cpu until it reaches
3805 n. So during boot up some boot time memory for per-cpu
3806 variables need be pre-allocated for later physical cpu
3809 nr_uarts= [SERIAL] maximum number of UARTs to be registered.
3811 numa=off [KNL, ARM64, PPC, RISCV, SPARC, X86] Disable NUMA, Only
3812 set up a single NUMA node spanning all memory.
3814 numa_balancing= [KNL,ARM64,PPC,RISCV,S390,X86] Enable or disable automatic
3816 Allowed values are enable and disable
3818 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA.
3819 'node', 'default' can be specified
3820 This can be set from sysctl after boot.
3821 See Documentation/admin-guide/sysctl/vm.rst for details.
3823 ohci1394_dma=early [HW] enable debugging via the ohci1394 driver.
3824 See Documentation/core-api/debugging-via-ohci1394.rst for more
3827 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands
3828 Rather than timing out after 20 ms if an EC
3829 command is not properly ACKed, override the length
3830 of the timeout. We have interrupts disabled while
3831 waiting for the ACK, so if this is set too high
3832 interrupts *may* be lost!
3834 omap_mux= [OMAP] Override bootloader pin multiplexing.
3835 Format: <mux_mode0.mode_name=value>...
3836 For example, to override I2C bus2:
3837 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100
3839 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration
3841 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock]
3843 boundary - index of last SLC block on Flex-OneNAND.
3844 The remaining blocks are configured as MLC blocks.
3845 lock - Configure if Flex-OneNAND boundary should be locked.
3846 Once locked, the boundary cannot be changed.
3847 1 indicates lock status, 0 indicates unlock status.
3849 oops=panic Always panic on oopses. Default is to just kill the
3850 process, but there is a small probability of
3851 deadlocking the machine.
3852 This will also cause panics on machine check exceptions.
3853 Useful together with panic=30 to trigger a reboot.
3856 [KNL] Boolean flag to control whether the page allocator
3857 should randomize its free lists. The randomization may
3858 be automatically enabled if the kernel detects it is
3859 running on a platform with a direct-mapped memory-side
3860 cache, and this parameter can be used to
3861 override/disable that behavior. The state of the flag
3862 can be read from sysfs at:
3863 /sys/module/page_alloc/parameters/shuffle.
3865 page_owner= [KNL] Boot-time page_owner enabling option.
3866 Storage of the information about who allocated
3867 each page is disabled in default. With this switch,
3869 on: enable the feature
3871 page_poison= [KNL] Boot-time parameter changing the state of
3872 poisoning on the buddy allocator, available with
3873 CONFIG_PAGE_POISONING=y.
3874 off: turn off poisoning (default)
3875 on: turn on poisoning
3877 page_reporting.page_reporting_order=
3878 [KNL] Minimal page reporting order
3880 Adjust the minimal page reporting order. The page
3881 reporting is disabled when it exceeds (MAX_ORDER-1).
3883 panic= [KNL] Kernel behaviour on panic: delay <timeout>
3884 timeout > 0: seconds before rebooting
3885 timeout = 0: wait forever
3886 timeout < 0: reboot immediately
3889 panic_print= Bitmask for printing system info when panic happens.
3890 User can chose combination of the following bits:
3891 bit 0: print all tasks info
3892 bit 1: print system memory info
3893 bit 2: print timer info
3894 bit 3: print locks info if CONFIG_LOCKDEP is on
3895 bit 4: print ftrace buffer
3896 bit 5: print all printk messages in buffer
3897 bit 6: print all CPUs backtrace (if available in the arch)
3898 *Be aware* that this option may print a _lot_ of lines,
3899 so there are risks of losing older messages in the log.
3900 Use this option carefully, maybe worth to setup a
3901 bigger log buffer with "log_buf_len" along with this.
3903 panic_on_taint= Bitmask for conditionally calling panic() in add_taint()
3904 Format: <hex>[,nousertaint]
3905 Hexadecimal bitmask representing the set of TAINT flags
3906 that will cause the kernel to panic when add_taint() is
3907 called with any of the flags in this set.
3908 The optional switch "nousertaint" can be utilized to
3909 prevent userspace forced crashes by writing to sysctl
3910 /proc/sys/kernel/tainted any flagset matching with the
3911 bitmask set on panic_on_taint.
3912 See Documentation/admin-guide/tainted-kernels.rst for
3913 extra details on the taint flags that users can pick
3914 to compose the bitmask to assign to panic_on_taint.
3916 panic_on_warn panic() instead of WARN(). Useful to cause kdump
3919 parkbd.port= [HW] Parallel port number the keyboard adapter is
3920 connected to, default is 0.
3922 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation,
3923 0 for XT, 1 for AT (default is AT).
3926 parport= [HW,PPT] Specify parallel ports. 0 disables.
3927 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] }
3928 Use 'auto' to force the driver to use any
3929 IRQ/DMA settings detected (the default is to
3930 ignore detected IRQ/DMA settings because of
3931 possible conflicts). You can specify the base
3932 address, IRQ, and DMA settings; IRQ and DMA
3933 should be numbers, or 'auto' (for using detected
3934 settings on that particular port), or 'nofifo'
3935 (to avoid using a FIFO even if it is detected).
3936 Parallel ports are assigned in the order they
3937 are specified on the command line, starting
3940 parport_init_mode= [HW,PPT]
3941 Configure VIA parallel port to operate in
3942 a specific mode. This is necessary on Pegasos
3943 computer where firmware has no options for setting
3944 up parallel port mode and sets it to spp.
3945 Currently this function knows 686a and 8231 chips.
3946 Format: [spp|ps2|epp|ecp|ecpepp]
3948 pata_legacy.all= [HW,LIBATA]
3950 Set to non-zero to probe primary and secondary ISA
3951 port ranges on PCI systems where no PCI PATA device
3952 has been found at either range. Disabled by default.
3954 pata_legacy.autospeed= [HW,LIBATA]
3956 Set to non-zero if a chip is present that snoops speed
3957 changes. Disabled by default.
3959 pata_legacy.ht6560a= [HW,LIBATA]
3961 Set to 1, 2, or 3 for HT 6560A on the primary channel,
3962 the secondary channel, or both channels respectively.
3963 Disabled by default.
3965 pata_legacy.ht6560b= [HW,LIBATA]
3967 Set to 1, 2, or 3 for HT 6560B on the primary channel,
3968 the secondary channel, or both channels respectively.
3969 Disabled by default.
3971 pata_legacy.iordy_mask= [HW,LIBATA]
3973 IORDY enable mask. Set individual bits to allow IORDY
3974 for the respective channel. Bit 0 is for the first
3975 legacy channel handled by this driver, bit 1 is for
3976 the second channel, and so on. The sequence will often
3977 correspond to the primary legacy channel, the secondary
3978 legacy channel, and so on, but the handling of a PCI
3979 bus and the use of other driver options may interfere
3980 with the sequence. By default IORDY is allowed across
3983 pata_legacy.opti82c46x= [HW,LIBATA]
3985 Set to 1, 2, or 3 for Opti 82c611A on the primary
3986 channel, the secondary channel, or both channels
3987 respectively. Disabled by default.
3989 pata_legacy.opti82c611a= [HW,LIBATA]
3991 Set to 1, 2, or 3 for Opti 82c465MV on the primary
3992 channel, the secondary channel, or both channels
3993 respectively. Disabled by default.
3995 pata_legacy.pio_mask= [HW,LIBATA]
3997 PIO mode mask for autospeed devices. Set individual
3998 bits to allow the use of the respective PIO modes.
3999 Bit 0 is for mode 0, bit 1 is for mode 1, and so on.
4000 All modes allowed by default.
4002 pata_legacy.probe_all= [HW,LIBATA]
4004 Set to non-zero to probe tertiary and further ISA
4005 port ranges on PCI systems. Disabled by default.
4007 pata_legacy.probe_mask= [HW,LIBATA]
4009 Probe mask for legacy ISA PATA ports. Depending on
4010 platform configuration and the use of other driver
4011 options up to 6 legacy ports are supported: 0x1f0,
4012 0x170, 0x1e8, 0x168, 0x1e0, 0x160, however probing
4013 of individual ports can be disabled by setting the
4014 corresponding bits in the mask to 1. Bit 0 is for
4015 the first port in the list above (0x1f0), and so on.
4016 By default all supported ports are probed.
4018 pata_legacy.qdi= [HW,LIBATA]
4020 Set to non-zero to probe QDI controllers. By default
4021 set to 1 if CONFIG_PATA_QDI_MODULE, 0 otherwise.
4023 pata_legacy.winbond= [HW,LIBATA]
4025 Set to non-zero to probe Winbond controllers. Use
4026 the standard I/O port (0x130) if 1, otherwise the
4027 value given is the I/O port to use (typically 0x1b0).
4028 By default set to 1 if CONFIG_PATA_WINBOND_VLB_MODULE,
4031 pata_platform.pio_mask= [HW,LIBATA]
4033 Supported PIO mode mask. Set individual bits to allow
4034 the use of the respective PIO modes. Bit 0 is for
4035 mode 0, bit 1 is for mode 1, and so on. Mode 0 only
4039 Halt all CPUs after the first oops has been printed for
4040 the specified number of seconds. This is to be used if
4041 your oopses keep scrolling off the screen.
4046 See header of drivers/block/paride/pcd.c.
4047 See also Documentation/admin-guide/blockdev/paride.rst.
4049 pci=option[,option...] [PCI] various PCI subsystem options.
4051 Some options herein operate on a specific device
4052 or a set of devices (<pci_dev>). These are
4053 specified in one of the following formats:
4055 [<domain>:]<bus>:<dev>.<func>[/<dev>.<func>]*
4056 pci:<vendor>:<device>[:<subvendor>:<subdevice>]
4058 Note: the first format specifies a PCI
4059 bus/device/function address which may change
4060 if new hardware is inserted, if motherboard
4061 firmware changes, or due to changes caused
4062 by other kernel parameters. If the
4063 domain is left unspecified, it is
4064 taken to be zero. Optionally, a path
4065 to a device through multiple device/function
4066 addresses can be specified after the base
4067 address (this is more robust against
4068 renumbering issues). The second format
4069 selects devices using IDs from the
4070 configuration space which may match multiple
4071 devices in the system.
4073 earlydump dump PCI config space before the kernel
4075 off [X86] don't probe for the PCI bus
4076 bios [X86-32] force use of PCI BIOS, don't access
4077 the hardware directly. Use this if your machine
4078 has a non-standard PCI host bridge.
4079 nobios [X86-32] disallow use of PCI BIOS, only direct
4080 hardware access methods are allowed. Use this
4081 if you experience crashes upon bootup and you
4082 suspect they are caused by the BIOS.
4083 conf1 [X86] Force use of PCI Configuration Access
4084 Mechanism 1 (config address in IO port 0xCF8,
4085 data in IO port 0xCFC, both 32-bit).
4086 conf2 [X86] Force use of PCI Configuration Access
4087 Mechanism 2 (IO port 0xCF8 is an 8-bit port for
4088 the function, IO port 0xCFA, also 8-bit, sets
4089 bus number. The config space is then accessed
4090 through ports 0xC000-0xCFFF).
4091 See http://wiki.osdev.org/PCI for more info
4092 on the configuration access mechanisms.
4093 noaer [PCIE] If the PCIEAER kernel config parameter is
4094 enabled, this kernel boot option can be used to
4095 disable the use of PCIE advanced error reporting.
4096 nodomains [PCI] Disable support for multiple PCI
4097 root domains (aka PCI segments, in ACPI-speak).
4098 nommconf [X86] Disable use of MMCONFIG for PCI
4100 check_enable_amd_mmconf [X86] check for and enable
4101 properly configured MMIO access to PCI
4102 config space on AMD family 10h CPU
4103 nomsi [MSI] If the PCI_MSI kernel config parameter is
4104 enabled, this kernel boot option can be used to
4105 disable the use of MSI interrupts system-wide.
4106 noioapicquirk [APIC] Disable all boot interrupt quirks.
4107 Safety option to keep boot IRQs enabled. This
4108 should never be necessary.
4109 ioapicreroute [APIC] Enable rerouting of boot IRQs to the
4110 primary IO-APIC for bridges that cannot disable
4111 boot IRQs. This fixes a source of spurious IRQs
4112 when the system masks IRQs.
4113 noioapicreroute [APIC] Disable workaround that uses the
4114 boot IRQ equivalent of an IRQ that connects to
4115 a chipset where boot IRQs cannot be disabled.
4116 The opposite of ioapicreroute.
4117 biosirq [X86-32] Use PCI BIOS calls to get the interrupt
4118 routing table. These calls are known to be buggy
4119 on several machines and they hang the machine
4120 when used, but on other computers it's the only
4121 way to get the interrupt routing table. Try
4122 this option if the kernel is unable to allocate
4123 IRQs or discover secondary PCI buses on your
4125 rom [X86] Assign address space to expansion ROMs.
4126 Use with caution as certain devices share
4127 address decoders between ROMs and other
4129 norom [X86] Do not assign address space to
4130 expansion ROMs that do not already have
4131 BIOS assigned address ranges.
4132 nobar [X86] Do not assign address space to the
4133 BARs that weren't assigned by the BIOS.
4134 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be
4135 assigned automatically to PCI devices. You can
4136 make the kernel exclude IRQs of your ISA cards
4138 pirqaddr=0xAAAAA [X86] Specify the physical address
4139 of the PIRQ table (normally generated
4140 by the BIOS) if it is outside the
4141 F0000h-100000h range.
4142 lastbus=N [X86] Scan all buses thru bus #N. Can be
4143 useful if the kernel is unable to find your
4144 secondary buses and you want to tell it
4145 explicitly which ones they are.
4146 assign-busses [X86] Always assign all PCI bus
4147 numbers ourselves, overriding
4148 whatever the firmware may have done.
4149 usepirqmask [X86] Honor the possible IRQ mask stored
4150 in the BIOS $PIR table. This is needed on
4151 some systems with broken BIOSes, notably
4152 some HP Pavilion N5400 and Omnibook XE3
4153 notebooks. This will have no effect if ACPI
4154 IRQ routing is enabled.
4155 noacpi [X86] Do not use ACPI for IRQ routing
4156 or for PCI scanning.
4157 use_crs [X86] Use PCI host bridge window information
4158 from ACPI. On BIOSes from 2008 or later, this
4159 is enabled by default. If you need to use this,
4160 please report a bug.
4161 nocrs [X86] Ignore PCI host bridge windows from ACPI.
4162 If you need to use this, please report a bug.
4163 use_e820 [X86] Use E820 reservations to exclude parts of
4164 PCI host bridge windows. This is a workaround
4165 for BIOS defects in host bridge _CRS methods.
4166 If you need to use this, please report a bug to
4167 <linux-pci@vger.kernel.org>.
4168 no_e820 [X86] Ignore E820 reservations for PCI host
4169 bridge windows. This is the default on modern
4170 hardware. If you need to use this, please report
4171 a bug to <linux-pci@vger.kernel.org>.
4172 routeirq Do IRQ routing for all PCI devices.
4173 This is normally done in pci_enable_device(),
4174 so this option is a temporary workaround
4175 for broken drivers that don't call it.
4176 skip_isa_align [X86] do not align io start addr, so can
4177 handle more pci cards
4178 noearly [X86] Don't do any early type 1 scanning.
4179 This might help on some broken boards which
4180 machine check when some devices' config space
4181 is read. But various workarounds are disabled
4182 and some IOMMU drivers will not work.
4183 bfsort Sort PCI devices into breadth-first order.
4184 This sorting is done to get a device
4185 order compatible with older (<= 2.4) kernels.
4186 nobfsort Don't sort PCI devices into breadth-first order.
4187 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size)
4188 tuning and use the BIOS-configured MPS defaults.
4189 pcie_bus_safe Set every device's MPS to the largest value
4190 supported by all devices below the root complex.
4191 pcie_bus_perf Set device MPS to the largest allowable MPS
4192 based on its parent bus. Also set MRRS (Max
4193 Read Request Size) to the largest supported
4194 value (no larger than the MPS that the device
4195 or bus can support) for best performance.
4196 pcie_bus_peer2peer Set every device's MPS to 128B, which
4197 every device is guaranteed to support. This
4198 configuration allows peer-to-peer DMA between
4199 any pair of devices, possibly at the cost of
4200 reduced performance. This also guarantees
4201 that hot-added devices will work.
4202 cbiosize=nn[KMG] The fixed amount of bus space which is
4203 reserved for the CardBus bridge's IO window.
4204 The default value is 256 bytes.
4205 cbmemsize=nn[KMG] The fixed amount of bus space which is
4206 reserved for the CardBus bridge's memory
4207 window. The default value is 64 megabytes.
4210 [<order of align>@]<pci_dev>[; ...]
4211 Specifies alignment and device to reassign
4212 aligned memory resources. How to
4213 specify the device is described above.
4214 If <order of align> is not specified,
4215 PAGE_SIZE is used as alignment.
4216 A PCI-PCI bridge can be specified if resource
4217 windows need to be expanded.
4218 To specify the alignment for several
4219 instances of a device, the PCI vendor,
4220 device, subvendor, and subdevice may be
4221 specified, e.g., 12@pci:8086:9c22:103c:198f
4222 for 4096-byte alignment.
4223 ecrc= Enable/disable PCIe ECRC (transaction layer
4224 end-to-end CRC checking).
4225 bios: Use BIOS/firmware settings. This is the
4229 hpiosize=nn[KMG] The fixed amount of bus space which is
4230 reserved for hotplug bridge's IO window.
4231 Default size is 256 bytes.
4232 hpmmiosize=nn[KMG] The fixed amount of bus space which is
4233 reserved for hotplug bridge's MMIO window.
4234 Default size is 2 megabytes.
4235 hpmmioprefsize=nn[KMG] The fixed amount of bus space which is
4236 reserved for hotplug bridge's MMIO_PREF window.
4237 Default size is 2 megabytes.
4238 hpmemsize=nn[KMG] The fixed amount of bus space which is
4239 reserved for hotplug bridge's MMIO and
4241 Default size is 2 megabytes.
4242 hpbussize=nn The minimum amount of additional bus numbers
4243 reserved for buses below a hotplug bridge.
4245 realloc= Enable/disable reallocating PCI bridge resources
4246 if allocations done by BIOS are too small to
4247 accommodate resources required by all child
4249 off: Turn realloc off
4251 realloc same as realloc=on
4252 noari do not use PCIe ARI.
4253 noats [PCIE, Intel-IOMMU, AMD-IOMMU]
4254 do not use PCIe ATS (and IOMMU device IOTLB).
4255 pcie_scan_all Scan all possible PCIe devices. Otherwise we
4256 only look for one device below a PCIe downstream
4258 big_root_window Try to add a big 64bit memory window to the PCIe
4259 root complex on AMD CPUs. Some GFX hardware
4260 can resize a BAR to allow access to all VRAM.
4261 Adding the window is slightly risky (it may
4262 conflict with unreported devices), so this
4264 disable_acs_redir=<pci_dev>[; ...]
4265 Specify one or more PCI devices (in the format
4266 specified above) separated by semicolons.
4267 Each device specified will have the PCI ACS
4268 redirect capabilities forced off which will
4269 allow P2P traffic between devices through
4270 bridges without forcing it upstream. Note:
4271 this removes isolation between devices and
4272 may put more devices in an IOMMU group.
4273 force_floating [S390] Force usage of floating interrupts.
4274 nomio [S390] Do not use MIO instructions.
4275 norid [S390] ignore the RID field and force use of
4276 one PCI domain per PCI function
4278 pcie_aspm= [PCIE] Forcibly enable or disable PCIe Active State Power
4281 force Enable ASPM even on devices that claim not to support it.
4282 WARNING: Forcing ASPM on may cause system lockups.
4284 pcie_ports= [PCIE] PCIe port services handling:
4285 native Use native PCIe services (PME, AER, DPC, PCIe hotplug)
4286 even if the platform doesn't give the OS permission to
4287 use them. This may cause conflicts if the platform
4288 also tries to use these services.
4289 dpc-native Use native PCIe service for DPC only. May
4290 cause conflicts if firmware uses AER or DPC.
4291 compat Disable native PCIe services (PME, AER, DPC, PCIe
4294 pcie_port_pm= [PCIE] PCIe port power management handling:
4295 off Disable power management of all PCIe ports
4296 force Forcibly enable power management of all PCIe ports
4298 pcie_pme= [PCIE,PM] Native PCIe PME signaling options:
4299 nomsi Do not use MSI for native PCIe PME signaling (this makes
4300 all PCIe root ports use INTx for all services).
4302 pcmv= [HW,PCMCIA] BadgePAD 4
4306 Keep all power-domains already enabled by bootloader on,
4307 even if no driver has claimed them. This is useful
4308 for debug and development, but should not be
4309 needed on a platform with proper driver support.
4312 See Documentation/admin-guide/blockdev/paride.rst.
4314 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at
4317 See arch/parisc/kernel/pdc_chassis.c
4319 percpu_alloc= Select which percpu first chunk allocator to use.
4320 Currently supported values are "embed" and "page".
4321 Archs may support subset or none of the selections.
4322 See comments in mm/percpu.c for details on each
4323 allocator. This parameter is primarily for debugging
4324 and performance comparison.
4327 See Documentation/admin-guide/blockdev/paride.rst.
4330 See Documentation/admin-guide/blockdev/paride.rst.
4332 pirq= [SMP,APIC] Manual mp-table setup
4333 See Documentation/x86/i386/IO-APIC.rst.
4335 plip= [PPT,NET] Parallel port network link
4336 Format: { parport<nr> | timid | 0 }
4337 See also Documentation/admin-guide/parport.rst.
4339 pmtmr= [X86] Manual setup of pmtmr I/O Port.
4340 Override pmtimer IOPort with a hex value.
4343 pmu_override= [PPC] Override the PMU.
4344 This option takes over the PMU facility, so it is no
4345 longer usable by perf. Setting this option starts the
4346 PMU counters by setting MMCR0 to 0 (the FC bit is
4347 cleared). If a number is given, then MMCR1 is set to
4348 that number, otherwise (e.g., 'pmu_override=on'), MMCR1
4351 pm_debug_messages [SUSPEND,KNL]
4352 Enable suspend/resume debug messages during boot up.
4355 Enable PNP debug messages (depends on the
4356 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time
4357 via /sys/module/pnp/parameters/debug. We always show
4358 current resource usage; turning this on also shows
4359 possible settings and some assignment information.
4365 { on | off | curr | res | no-curr | no-res }
4368 [ISAPNP] Exclude IRQs for the autoconfiguration
4371 [ISAPNP] Exclude DMAs for the autoconfiguration
4373 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration
4374 Ranges are in pairs (I/O port base and size).
4377 [ISAPNP] Exclude memory regions for the
4379 Ranges are in pairs (memory base and size).
4381 ports= [IP_VS_FTP] IPVS ftp helper module
4383 Up to 8 (IP_VS_APP_MAX_PORTS) ports
4385 Format: <port>,<port>....
4387 powersave=off [PPC] This option disables power saving features.
4388 It specifically disables cpuidle and sets the
4389 platform machine description specific power_save
4390 function to NULL. On Idle the CPU just reduces
4393 ppc_strict_facility_enable
4394 [PPC] This option catches any kernel floating point,
4395 Altivec, VSX and SPE outside of regions specifically
4396 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()).
4397 There is some performance impact when enabling this.
4401 Disable Hardware Transactional Memory
4404 Select preemption mode if you have CONFIG_PREEMPT_DYNAMIC
4405 none - Limited to cond_resched() calls
4406 voluntary - Limited to cond_resched() and might_sleep() calls
4407 full - Any section that isn't explicitly preempt disabled
4408 can be preempted anytime.
4410 print-fatal-signals=
4411 [KNL] debug: print fatal signals
4413 If enabled, warn about various signal handling
4414 related application anomalies: too many signals,
4415 too many POSIX.1 timers, fatal signals causing a
4418 If you hit the warning due to signal overflow,
4419 you might want to try "ulimit -i unlimited".
4423 printk.always_kmsg_dump=
4424 Trigger kmsg_dump for cases other than kernel oops or
4426 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
4429 printk.console_no_auto_verbose=
4430 Disable console loglevel raise on oops, panic
4431 or lockdep-detected issues (only if lock debug is on).
4432 With an exception to setups with low baudrate on
4433 serial console, keeping this 0 is a good choice
4434 in order to provide more debug information.
4436 default: 0 (auto_verbose is enabled)
4438 printk.devkmsg={on,off,ratelimit}
4439 Control writing to /dev/kmsg.
4440 on - unlimited logging to /dev/kmsg from userspace
4441 off - logging to /dev/kmsg disabled
4442 ratelimit - ratelimit the logging
4445 printk.time= Show timing data prefixed to each printk message line
4446 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
4448 processor.max_cstate= [HW,ACPI]
4449 Limit processor to maximum C-state
4450 max_cstate=9 overrides any DMI blacklist limit.
4452 processor.nocst [HW,ACPI]
4453 Ignore the _CST method to determine C-states,
4454 instead using the legacy FADT method
4456 profile= [KNL] Enable kernel profiling via /proc/profile
4457 Format: [<profiletype>,]<number>
4458 Param: <profiletype>: "schedule", "sleep", or "kvm"
4459 [defaults to kernel profiling]
4460 Param: "schedule" - profile schedule points.
4461 Param: "sleep" - profile D-state sleeping (millisecs).
4462 Requires CONFIG_SCHEDSTATS
4463 Param: "kvm" - profile VM exits.
4464 Param: <number> - step/bucket size as a power of 2 for
4465 statistical time based profiling.
4467 prompt_ramdisk= [RAM] [Deprecated]
4469 prot_virt= [S390] enable hosting protected virtual machines
4470 isolated from the hypervisor (if hardware supports
4474 psi= [KNL] Enable or disable pressure stall information
4478 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to
4479 probe for; one of (bare|imps|exps|lifebook|any).
4480 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports
4482 psmouse.resetafter= [HW,MOUSE]
4483 Try to reset the device after so many bad packets
4486 [HW,MOUSE] Set desired mouse resolution, in dpi.
4487 psmouse.smartscroll=
4488 [HW,MOUSE] Controls Logitech smartscroll autorepeat.
4489 0 = disabled, 1 = enabled (default).
4491 pstore.backend= Specify the name of the pstore backend to use
4494 See Documentation/admin-guide/blockdev/paride.rst.
4496 pti= [X86-64] Control Page Table Isolation of user and
4497 kernel address spaces. Disabling this feature
4498 removes hardening, but improves performance of
4499 system calls and interrupts.
4501 on - unconditionally enable
4502 off - unconditionally disable
4503 auto - kernel detects whether your CPU model is
4504 vulnerable to issues that PTI mitigates
4506 Not specifying this option is equivalent to pti=auto.
4509 Equivalent to pti=off
4512 [KNL] Number of legacy pty's. Overwrites compiled-in
4515 quiet [KNL] Disable most log messages
4520 See Documentation/admin-guide/md.rst.
4522 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes
4523 See Documentation/admin-guide/blockdev/ramdisk.rst.
4525 ramdisk_start= [RAM] RAM disk image start address
4527 random.trust_cpu={on,off}
4528 [KNL] Enable or disable trusting the use of the
4529 CPU's random number generator (if available) to
4530 fully seed the kernel's CRNG. Default is controlled
4531 by CONFIG_RANDOM_TRUST_CPU.
4533 random.trust_bootloader={on,off}
4534 [KNL] Enable or disable trusting the use of a
4535 seed passed by the bootloader (if available) to
4536 fully seed the kernel's CRNG. Default is controlled
4537 by CONFIG_RANDOM_TRUST_BOOTLOADER.
4539 randomize_kstack_offset=
4540 [KNL] Enable or disable kernel stack offset
4541 randomization, which provides roughly 5 bits of
4542 entropy, frustrating memory corruption attacks
4543 that depend on stack address determinism or
4544 cross-syscall address exposures. This is only
4545 available on architectures that have defined
4546 CONFIG_HAVE_ARCH_RANDOMIZE_KSTACK_OFFSET.
4547 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
4548 Default is CONFIG_RANDOMIZE_KSTACK_OFFSET_DEFAULT.
4550 ras=option[,option,...] [KNL] RAS-specific options
4553 Disable the Correctable Errors Collector,
4554 see CONFIG_RAS_CEC help text.
4556 rcu_nocbs[=cpu-list]
4557 [KNL] The optional argument is a cpu list,
4560 In kernels built with CONFIG_RCU_NOCB_CPU=y,
4561 enable the no-callback CPU mode, which prevents
4562 such CPUs' callbacks from being invoked in
4563 softirq context. Invocation of such CPUs' RCU
4564 callbacks will instead be offloaded to "rcuox/N"
4565 kthreads created for that purpose, where "x" is
4566 "p" for RCU-preempt, "s" for RCU-sched, and "g"
4567 for the kthreads that mediate grace periods; and
4568 "N" is the CPU number. This reduces OS jitter on
4569 the offloaded CPUs, which can be useful for HPC
4570 and real-time workloads. It can also improve
4571 energy efficiency for asymmetric multiprocessors.
4573 If a cpulist is passed as an argument, the specified
4574 list of CPUs is set to no-callback mode from boot.
4576 Otherwise, if the '=' sign and the cpulist
4577 arguments are omitted, no CPU will be set to
4578 no-callback mode from boot but the mode may be
4579 toggled at runtime via cpusets.
4581 Note that this argument takes precedence over
4582 the CONFIG_RCU_NOCB_CPU_DEFAULT_ALL option.
4585 Rather than requiring that offloaded CPUs
4586 (specified by rcu_nocbs= above) explicitly
4587 awaken the corresponding "rcuoN" kthreads,
4588 make these kthreads poll for callbacks.
4589 This improves the real-time response for the
4590 offloaded CPUs by relieving them of the need to
4591 wake up the corresponding kthread, but degrades
4592 energy efficiency by requiring that the kthreads
4593 periodically wake up to do the polling.
4595 rcutree.blimit= [KNL]
4596 Set maximum number of finished RCU callbacks to
4597 process in one batch.
4599 rcutree.dump_tree= [KNL]
4600 Dump the structure of the rcu_node combining tree
4601 out at early boot. This is used for diagnostic
4602 purposes, to verify correct tree setup.
4604 rcutree.gp_cleanup_delay= [KNL]
4605 Set the number of jiffies to delay each step of
4606 RCU grace-period cleanup.
4608 rcutree.gp_init_delay= [KNL]
4609 Set the number of jiffies to delay each step of
4610 RCU grace-period initialization.
4612 rcutree.gp_preinit_delay= [KNL]
4613 Set the number of jiffies to delay each step of
4614 RCU grace-period pre-initialization, that is,
4615 the propagation of recent CPU-hotplug changes up
4616 the rcu_node combining tree.
4618 rcutree.use_softirq= [KNL]
4619 If set to zero, move all RCU_SOFTIRQ processing to
4620 per-CPU rcuc kthreads. Defaults to a non-zero
4621 value, meaning that RCU_SOFTIRQ is used by default.
4622 Specify rcutree.use_softirq=0 to use rcuc kthreads.
4624 But note that CONFIG_PREEMPT_RT=y kernels disable
4625 this kernel boot parameter, forcibly setting it
4628 rcutree.rcu_fanout_exact= [KNL]
4629 Disable autobalancing of the rcu_node combining
4630 tree. This is used by rcutorture, and might
4631 possibly be useful for architectures having high
4632 cache-to-cache transfer latencies.
4634 rcutree.rcu_fanout_leaf= [KNL]
4635 Change the number of CPUs assigned to each
4636 leaf rcu_node structure. Useful for very
4637 large systems, which will choose the value 64,
4638 and for NUMA systems with large remote-access
4639 latencies, which will choose a value aligned
4640 with the appropriate hardware boundaries.
4642 rcutree.rcu_min_cached_objs= [KNL]
4643 Minimum number of objects which are cached and
4644 maintained per one CPU. Object size is equal
4645 to PAGE_SIZE. The cache allows to reduce the
4646 pressure to page allocator, also it makes the
4647 whole algorithm to behave better in low memory
4650 rcutree.rcu_delay_page_cache_fill_msec= [KNL]
4651 Set the page-cache refill delay (in milliseconds)
4652 in response to low-memory conditions. The range
4653 of permitted values is in the range 0:100000.
4655 rcutree.jiffies_till_first_fqs= [KNL]
4656 Set delay from grace-period initialization to
4657 first attempt to force quiescent states.
4658 Units are jiffies, minimum value is zero,
4659 and maximum value is HZ.
4661 rcutree.jiffies_till_next_fqs= [KNL]
4662 Set delay between subsequent attempts to force
4663 quiescent states. Units are jiffies, minimum
4664 value is one, and maximum value is HZ.
4666 rcutree.jiffies_till_sched_qs= [KNL]
4667 Set required age in jiffies for a
4668 given grace period before RCU starts
4669 soliciting quiescent-state help from
4670 rcu_note_context_switch() and cond_resched().
4671 If not specified, the kernel will calculate
4672 a value based on the most recent settings
4673 of rcutree.jiffies_till_first_fqs
4674 and rcutree.jiffies_till_next_fqs.
4675 This calculated value may be viewed in
4676 rcutree.jiffies_to_sched_qs. Any attempt to set
4677 rcutree.jiffies_to_sched_qs will be cheerfully
4680 rcutree.kthread_prio= [KNL,BOOT]
4681 Set the SCHED_FIFO priority of the RCU per-CPU
4682 kthreads (rcuc/N). This value is also used for
4683 the priority of the RCU boost threads (rcub/N)
4684 and for the RCU grace-period kthreads (rcu_bh,
4685 rcu_preempt, and rcu_sched). If RCU_BOOST is
4686 set, valid values are 1-99 and the default is 1
4687 (the least-favored priority). Otherwise, when
4688 RCU_BOOST is not set, valid values are 0-99 and
4689 the default is zero (non-realtime operation).
4690 When RCU_NOCB_CPU is set, also adjust the
4691 priority of NOCB callback kthreads.
4693 rcutree.rcu_divisor= [KNL]
4694 Set the shift-right count to use to compute
4695 the callback-invocation batch limit bl from
4696 the number of callbacks queued on this CPU.
4697 The result will be bounded below by the value of
4698 the rcutree.blimit kernel parameter. Every bl
4699 callbacks, the softirq handler will exit in
4700 order to allow the CPU to do other work.
4702 Please note that this callback-invocation batch
4703 limit applies only to non-offloaded callback
4704 invocation. Offloaded callbacks are instead
4705 invoked in the context of an rcuoc kthread, which
4706 scheduler will preempt as it does any other task.
4708 rcutree.nocb_nobypass_lim_per_jiffy= [KNL]
4709 On callback-offloaded (rcu_nocbs) CPUs,
4710 RCU reduces the lock contention that would
4711 otherwise be caused by callback floods through
4712 use of the ->nocb_bypass list. However, in the
4713 common non-flooded case, RCU queues directly to
4714 the main ->cblist in order to avoid the extra
4715 overhead of the ->nocb_bypass list and its lock.
4716 But if there are too many callbacks queued during
4717 a single jiffy, RCU pre-queues the callbacks into
4718 the ->nocb_bypass queue. The definition of "too
4719 many" is supplied by this kernel boot parameter.
4721 rcutree.rcu_nocb_gp_stride= [KNL]
4722 Set the number of NOCB callback kthreads in
4723 each group, which defaults to the square root
4724 of the number of CPUs. Larger numbers reduce
4725 the wakeup overhead on the global grace-period
4726 kthread, but increases that same overhead on
4727 each group's NOCB grace-period kthread.
4729 rcutree.qhimark= [KNL]
4730 Set threshold of queued RCU callbacks beyond which
4731 batch limiting is disabled.
4733 rcutree.qlowmark= [KNL]
4734 Set threshold of queued RCU callbacks below which
4735 batch limiting is re-enabled.
4737 rcutree.qovld= [KNL]
4738 Set threshold of queued RCU callbacks beyond which
4739 RCU's force-quiescent-state scan will aggressively
4740 enlist help from cond_resched() and sched IPIs to
4741 help CPUs more quickly reach quiescent states.
4742 Set to less than zero to make this be set based
4743 on rcutree.qhimark at boot time and to zero to
4744 disable more aggressive help enlistment.
4746 rcutree.rcu_kick_kthreads= [KNL]
4747 Cause the grace-period kthread to get an extra
4748 wake_up() if it sleeps three times longer than
4749 it should at force-quiescent-state time.
4750 This wake_up() will be accompanied by a
4751 WARN_ONCE() splat and an ftrace_dump().
4753 rcutree.rcu_unlock_delay= [KNL]
4754 In CONFIG_RCU_STRICT_GRACE_PERIOD=y kernels,
4755 this specifies an rcu_read_unlock()-time delay
4756 in microseconds. This defaults to zero.
4757 Larger delays increase the probability of
4758 catching RCU pointer leaks, that is, buggy use
4759 of RCU-protected pointers after the relevant
4760 rcu_read_unlock() has completed.
4762 rcutree.sysrq_rcu= [KNL]
4763 Commandeer a sysrq key to dump out Tree RCU's
4764 rcu_node tree with an eye towards determining
4765 why a new grace period has not yet started.
4767 rcuscale.gp_async= [KNL]
4768 Measure performance of asynchronous
4769 grace-period primitives such as call_rcu().
4771 rcuscale.gp_async_max= [KNL]
4772 Specify the maximum number of outstanding
4773 callbacks per writer thread. When a writer
4774 thread exceeds this limit, it invokes the
4775 corresponding flavor of rcu_barrier() to allow
4776 previously posted callbacks to drain.
4778 rcuscale.gp_exp= [KNL]
4779 Measure performance of expedited synchronous
4780 grace-period primitives.
4782 rcuscale.holdoff= [KNL]
4783 Set test-start holdoff period. The purpose of
4784 this parameter is to delay the start of the
4785 test until boot completes in order to avoid
4788 rcuscale.kfree_rcu_test= [KNL]
4789 Set to measure performance of kfree_rcu() flooding.
4791 rcuscale.kfree_rcu_test_double= [KNL]
4792 Test the double-argument variant of kfree_rcu().
4793 If this parameter has the same value as
4794 rcuscale.kfree_rcu_test_single, both the single-
4795 and double-argument variants are tested.
4797 rcuscale.kfree_rcu_test_single= [KNL]
4798 Test the single-argument variant of kfree_rcu().
4799 If this parameter has the same value as
4800 rcuscale.kfree_rcu_test_double, both the single-
4801 and double-argument variants are tested.
4803 rcuscale.kfree_nthreads= [KNL]
4804 The number of threads running loops of kfree_rcu().
4806 rcuscale.kfree_alloc_num= [KNL]
4807 Number of allocations and frees done in an iteration.
4809 rcuscale.kfree_loops= [KNL]
4810 Number of loops doing rcuscale.kfree_alloc_num number
4811 of allocations and frees.
4813 rcuscale.nreaders= [KNL]
4814 Set number of RCU readers. The value -1 selects
4815 N, where N is the number of CPUs. A value
4816 "n" less than -1 selects N-n+1, where N is again
4817 the number of CPUs. For example, -2 selects N
4818 (the number of CPUs), -3 selects N+1, and so on.
4819 A value of "n" less than or equal to -N selects
4822 rcuscale.nwriters= [KNL]
4823 Set number of RCU writers. The values operate
4824 the same as for rcuscale.nreaders.
4825 N, where N is the number of CPUs
4827 rcuscale.perf_type= [KNL]
4828 Specify the RCU implementation to test.
4830 rcuscale.shutdown= [KNL]
4831 Shut the system down after performance tests
4832 complete. This is useful for hands-off automated
4835 rcuscale.verbose= [KNL]
4836 Enable additional printk() statements.
4838 rcuscale.writer_holdoff= [KNL]
4839 Write-side holdoff between grace periods,
4840 in microseconds. The default of zero says
4843 rcutorture.fqs_duration= [KNL]
4844 Set duration of force_quiescent_state bursts
4847 rcutorture.fqs_holdoff= [KNL]
4848 Set holdoff time within force_quiescent_state bursts
4851 rcutorture.fqs_stutter= [KNL]
4852 Set wait time between force_quiescent_state bursts
4855 rcutorture.fwd_progress= [KNL]
4856 Specifies the number of kthreads to be used
4857 for RCU grace-period forward-progress testing
4858 for the types of RCU supporting this notion.
4859 Defaults to 1 kthread, values less than zero or
4860 greater than the number of CPUs cause the number
4863 rcutorture.fwd_progress_div= [KNL]
4864 Specify the fraction of a CPU-stall-warning
4865 period to do tight-loop forward-progress testing.
4867 rcutorture.fwd_progress_holdoff= [KNL]
4868 Number of seconds to wait between successive
4869 forward-progress tests.
4871 rcutorture.fwd_progress_need_resched= [KNL]
4872 Enclose cond_resched() calls within checks for
4873 need_resched() during tight-loop forward-progress
4876 rcutorture.gp_cond= [KNL]
4877 Use conditional/asynchronous update-side
4878 primitives, if available.
4880 rcutorture.gp_exp= [KNL]
4881 Use expedited update-side primitives, if available.
4883 rcutorture.gp_normal= [KNL]
4884 Use normal (non-expedited) asynchronous
4885 update-side primitives, if available.
4887 rcutorture.gp_sync= [KNL]
4888 Use normal (non-expedited) synchronous
4889 update-side primitives, if available. If all
4890 of rcutorture.gp_cond=, rcutorture.gp_exp=,
4891 rcutorture.gp_normal=, and rcutorture.gp_sync=
4892 are zero, rcutorture acts as if is interpreted
4893 they are all non-zero.
4895 rcutorture.irqreader= [KNL]
4896 Run RCU readers from irq handlers, or, more
4897 accurately, from a timer handler. Not all RCU
4898 flavors take kindly to this sort of thing.
4900 rcutorture.leakpointer= [KNL]
4901 Leak an RCU-protected pointer out of the reader.
4902 This can of course result in splats, and is
4903 intended to test the ability of things like
4904 CONFIG_RCU_STRICT_GRACE_PERIOD=y to detect
4907 rcutorture.n_barrier_cbs= [KNL]
4908 Set callbacks/threads for rcu_barrier() testing.
4910 rcutorture.nfakewriters= [KNL]
4911 Set number of concurrent RCU writers. These just
4912 stress RCU, they don't participate in the actual
4913 test, hence the "fake".
4915 rcutorture.nocbs_nthreads= [KNL]
4916 Set number of RCU callback-offload togglers.
4917 Zero (the default) disables toggling.
4919 rcutorture.nocbs_toggle= [KNL]
4920 Set the delay in milliseconds between successive
4921 callback-offload toggling attempts.
4923 rcutorture.nreaders= [KNL]
4924 Set number of RCU readers. The value -1 selects
4925 N-1, where N is the number of CPUs. A value
4926 "n" less than -1 selects N-n-2, where N is again
4927 the number of CPUs. For example, -2 selects N
4928 (the number of CPUs), -3 selects N+1, and so on.
4930 rcutorture.object_debug= [KNL]
4931 Enable debug-object double-call_rcu() testing.
4933 rcutorture.onoff_holdoff= [KNL]
4934 Set time (s) after boot for CPU-hotplug testing.
4936 rcutorture.onoff_interval= [KNL]
4937 Set time (jiffies) between CPU-hotplug operations,
4938 or zero to disable CPU-hotplug testing.
4940 rcutorture.read_exit= [KNL]
4941 Set the number of read-then-exit kthreads used
4942 to test the interaction of RCU updaters and
4943 task-exit processing.
4945 rcutorture.read_exit_burst= [KNL]
4946 The number of times in a given read-then-exit
4947 episode that a set of read-then-exit kthreads
4950 rcutorture.read_exit_delay= [KNL]
4951 The delay, in seconds, between successive
4952 read-then-exit testing episodes.
4954 rcutorture.shuffle_interval= [KNL]
4955 Set task-shuffle interval (s). Shuffling tasks
4956 allows some CPUs to go into dyntick-idle mode
4957 during the rcutorture test.
4959 rcutorture.shutdown_secs= [KNL]
4960 Set time (s) after boot system shutdown. This
4961 is useful for hands-off automated testing.
4963 rcutorture.stall_cpu= [KNL]
4964 Duration of CPU stall (s) to test RCU CPU stall
4965 warnings, zero to disable.
4967 rcutorture.stall_cpu_block= [KNL]
4968 Sleep while stalling if set. This will result
4969 in warnings from preemptible RCU in addition
4970 to any other stall-related activity.
4972 rcutorture.stall_cpu_holdoff= [KNL]
4973 Time to wait (s) after boot before inducing stall.
4975 rcutorture.stall_cpu_irqsoff= [KNL]
4976 Disable interrupts while stalling if set.
4978 rcutorture.stall_gp_kthread= [KNL]
4979 Duration (s) of forced sleep within RCU
4980 grace-period kthread to test RCU CPU stall
4981 warnings, zero to disable. If both stall_cpu
4982 and stall_gp_kthread are specified, the
4983 kthread is starved first, then the CPU.
4985 rcutorture.stat_interval= [KNL]
4986 Time (s) between statistics printk()s.
4988 rcutorture.stutter= [KNL]
4989 Time (s) to stutter testing, for example, specifying
4990 five seconds causes the test to run for five seconds,
4991 wait for five seconds, and so on. This tests RCU's
4992 ability to transition abruptly to and from idle.
4994 rcutorture.test_boost= [KNL]
4995 Test RCU priority boosting? 0=no, 1=maybe, 2=yes.
4996 "Maybe" means test if the RCU implementation
4997 under test support RCU priority boosting.
4999 rcutorture.test_boost_duration= [KNL]
5000 Duration (s) of each individual boost test.
5002 rcutorture.test_boost_interval= [KNL]
5003 Interval (s) between each boost test.
5005 rcutorture.test_no_idle_hz= [KNL]
5006 Test RCU's dyntick-idle handling. See also the
5007 rcutorture.shuffle_interval parameter.
5009 rcutorture.torture_type= [KNL]
5010 Specify the RCU implementation to test.
5012 rcutorture.verbose= [KNL]
5013 Enable additional printk() statements.
5015 rcupdate.rcu_cpu_stall_ftrace_dump= [KNL]
5016 Dump ftrace buffer after reporting RCU CPU
5019 rcupdate.rcu_cpu_stall_suppress= [KNL]
5020 Suppress RCU CPU stall warning messages.
5022 rcupdate.rcu_cpu_stall_suppress_at_boot= [KNL]
5023 Suppress RCU CPU stall warning messages and
5024 rcutorture writer stall warnings that occur
5025 during early boot, that is, during the time
5026 before the init task is spawned.
5028 rcupdate.rcu_cpu_stall_timeout= [KNL]
5029 Set timeout for RCU CPU stall warning messages.
5030 The value is in seconds and the maximum allowed
5031 value is 300 seconds.
5033 rcupdate.rcu_exp_cpu_stall_timeout= [KNL]
5034 Set timeout for expedited RCU CPU stall warning
5035 messages. The value is in milliseconds
5036 and the maximum allowed value is 21000
5037 milliseconds. Please note that this value is
5038 adjusted to an arch timer tick resolution.
5039 Setting this to zero causes the value from
5040 rcupdate.rcu_cpu_stall_timeout to be used (after
5041 conversion from seconds to milliseconds).
5043 rcupdate.rcu_expedited= [KNL]
5044 Use expedited grace-period primitives, for
5045 example, synchronize_rcu_expedited() instead
5046 of synchronize_rcu(). This reduces latency,
5047 but can increase CPU utilization, degrade
5048 real-time latency, and degrade energy efficiency.
5049 No effect on CONFIG_TINY_RCU kernels.
5051 rcupdate.rcu_normal= [KNL]
5052 Use only normal grace-period primitives,
5053 for example, synchronize_rcu() instead of
5054 synchronize_rcu_expedited(). This improves
5055 real-time latency, CPU utilization, and
5056 energy efficiency, but can expose users to
5057 increased grace-period latency. This parameter
5058 overrides rcupdate.rcu_expedited. No effect on
5059 CONFIG_TINY_RCU kernels.
5061 rcupdate.rcu_normal_after_boot= [KNL]
5062 Once boot has completed (that is, after
5063 rcu_end_inkernel_boot() has been invoked), use
5064 only normal grace-period primitives. No effect
5065 on CONFIG_TINY_RCU kernels.
5067 But note that CONFIG_PREEMPT_RT=y kernels enables
5068 this kernel boot parameter, forcibly setting
5069 it to the value one, that is, converting any
5070 post-boot attempt at an expedited RCU grace
5071 period to instead use normal non-expedited
5072 grace-period processing.
5074 rcupdate.rcu_task_collapse_lim= [KNL]
5075 Set the maximum number of callbacks present
5076 at the beginning of a grace period that allows
5077 the RCU Tasks flavors to collapse back to using
5078 a single callback queue. This switching only
5079 occurs when rcupdate.rcu_task_enqueue_lim is
5080 set to the default value of -1.
5082 rcupdate.rcu_task_contend_lim= [KNL]
5083 Set the minimum number of callback-queuing-time
5084 lock-contention events per jiffy required to
5085 cause the RCU Tasks flavors to switch to per-CPU
5086 callback queuing. This switching only occurs
5087 when rcupdate.rcu_task_enqueue_lim is set to
5088 the default value of -1.
5090 rcupdate.rcu_task_enqueue_lim= [KNL]
5091 Set the number of callback queues to use for the
5092 RCU Tasks family of RCU flavors. The default
5093 of -1 allows this to be automatically (and
5094 dynamically) adjusted. This parameter is intended
5097 rcupdate.rcu_task_ipi_delay= [KNL]
5098 Set time in jiffies during which RCU tasks will
5099 avoid sending IPIs, starting with the beginning
5100 of a given grace period. Setting a large
5101 number avoids disturbing real-time workloads,
5102 but lengthens grace periods.
5104 rcupdate.rcu_task_stall_info= [KNL]
5105 Set initial timeout in jiffies for RCU task stall
5106 informational messages, which give some indication
5107 of the problem for those not patient enough to
5108 wait for ten minutes. Informational messages are
5109 only printed prior to the stall-warning message
5110 for a given grace period. Disable with a value
5111 less than or equal to zero. Defaults to ten
5112 seconds. A change in value does not take effect
5113 until the beginning of the next grace period.
5115 rcupdate.rcu_task_stall_info_mult= [KNL]
5116 Multiplier for time interval between successive
5117 RCU task stall informational messages for a given
5118 RCU tasks grace period. This value is clamped
5119 to one through ten, inclusive. It defaults to
5120 the value three, so that the first informational
5121 message is printed 10 seconds into the grace
5122 period, the second at 40 seconds, the third at
5123 160 seconds, and then the stall warning at 600
5124 seconds would prevent a fourth at 640 seconds.
5126 rcupdate.rcu_task_stall_timeout= [KNL]
5127 Set timeout in jiffies for RCU task stall
5128 warning messages. Disable with a value less
5129 than or equal to zero. Defaults to ten minutes.
5130 A change in value does not take effect until
5131 the beginning of the next grace period.
5133 rcupdate.rcu_self_test= [KNL]
5134 Run the RCU early boot self tests
5138 Run specified binary instead of /init from the ramdisk,
5139 used for early userspace startup. See initrd.
5142 force - Override the decision by the kernel to hide the
5143 advertisement of RDRAND support (this affects
5144 certain AMD processors because of buggy BIOS
5145 support, specifically around the suspend/resume
5149 Turn on/off individual RDT features. List is:
5150 cmt, mbmtotal, mbmlocal, l3cat, l3cdp, l2cat, l2cdp,
5152 E.g. to turn on cmt and turn off mba use:
5156 Format (x86 or x86_64):
5157 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] | d[efault] \
5159 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \
5161 Where reboot_mode is one of warm (soft) or cold (hard) or gpio
5162 (prefix with 'panic_' to set mode for panic
5164 reboot_type is one of bios, acpi, kbd, triple, efi, or pci,
5165 reboot_force is either force or not specified,
5166 reboot_cpu is s[mp]#### with #### being the processor
5167 to be used for rebooting.
5169 refscale.holdoff= [KNL]
5170 Set test-start holdoff period. The purpose of
5171 this parameter is to delay the start of the
5172 test until boot completes in order to avoid
5175 refscale.loops= [KNL]
5176 Set the number of loops over the synchronization
5177 primitive under test. Increasing this number
5178 reduces noise due to loop start/end overhead,
5179 but the default has already reduced the per-pass
5180 noise to a handful of picoseconds on ca. 2020
5183 refscale.nreaders= [KNL]
5184 Set number of readers. The default value of -1
5185 selects N, where N is roughly 75% of the number
5186 of CPUs. A value of zero is an interesting choice.
5188 refscale.nruns= [KNL]
5189 Set number of runs, each of which is dumped onto
5192 refscale.readdelay= [KNL]
5193 Set the read-side critical-section duration,
5194 measured in microseconds.
5196 refscale.scale_type= [KNL]
5197 Specify the read-protection implementation to test.
5199 refscale.shutdown= [KNL]
5200 Shut down the system at the end of the performance
5201 test. This defaults to 1 (shut it down) when
5202 refscale is built into the kernel and to 0 (leave
5203 it running) when refscale is built as a module.
5205 refscale.verbose= [KNL]
5206 Enable additional printk() statements.
5208 refscale.verbose_batched= [KNL]
5209 Batch the additional printk() statements. If zero
5210 (the default) or negative, print everything. Otherwise,
5211 print every Nth verbose statement, where N is the value
5215 [KNL, SMP] Set scheduler's default relax_domain_level.
5216 See Documentation/admin-guide/cgroup-v1/cpusets.rst.
5218 reserve= [KNL,BUGS] Force kernel to ignore I/O ports or memory
5219 Format: <base1>,<size1>[,<base2>,<size2>,...]
5220 Reserve I/O ports or memory so the kernel won't use
5221 them. If <base> is less than 0x10000, the region
5222 is assumed to be I/O ports; otherwise it is memory.
5224 reservetop= [X86-32]
5226 Reserves a hole at the top of the kernel virtual
5229 reset_devices [KNL] Force drivers to reset the underlying device
5230 during initialization.
5233 Specify the partition device for software suspend
5235 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>}
5237 resume_offset= [SWSUSP]
5238 Specify the offset from the beginning of the partition
5239 given by "resume=" at which the swap header is located,
5240 in <PAGE_SIZE> units (needed only for swap files).
5241 See Documentation/power/swsusp-and-swap-files.rst
5243 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to
5244 read the resume files
5246 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up.
5247 Useful for devices that are detected asynchronously
5248 (e.g. USB and MMC devices).
5250 retain_initrd [RAM] Keep initrd memory after extraction
5252 retbleed= [X86] Control mitigation of RETBleed (Arbitrary
5253 Speculative Code Execution with Return Instructions)
5257 auto - automatically select a migitation
5258 auto,nosmt - automatically select a mitigation,
5259 disabling SMT if necessary for
5260 the full mitigation (only on Zen1
5261 and older without STIBP).
5262 ibpb - mitigate short speculation windows on
5263 basic block boundaries too. Safe, highest
5265 unret - force enable untrained return thunks,
5266 only effective on AMD f15h-f17h
5268 unret,nosmt - like unret, will disable SMT when STIBP
5271 Selecting 'auto' will choose a mitigation method at run
5272 time according to the CPU.
5274 Not specifying this option is equivalent to retbleed=auto.
5276 rfkill.default_state=
5277 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm,
5278 etc. communication is blocked by default.
5281 rfkill.master_switch_mode=
5282 0 The "airplane mode" button does nothing.
5283 1 The "airplane mode" button toggles between everything
5284 blocked and the previous configuration.
5285 2 The "airplane mode" button toggles between everything
5286 blocked and everything unblocked.
5288 rhash_entries= [KNL,NET]
5289 Set number of hash buckets for route cache
5292 [KNL] Disable ring 3 MONITOR/MWAIT feature on supported
5295 ro [KNL] Mount root device read-only on boot
5298 on Mark read-only kernel memory as read-only (default).
5299 off Leave read-only kernel memory writable for debugging.
5302 Enable the uart passthrough on the designated usb port
5303 on Rockchip SoCs. When active, the signals of the
5304 debug-uart get routed to the D+ and D- pins of the usb
5305 port and the regular usb controller gets disabled.
5307 root= [KNL] Root filesystem
5308 See name_to_dev_t comment in init/do_mounts.c.
5310 rootdelay= [KNL] Delay (in seconds) to pause before attempting to
5311 mount the root filesystem
5313 rootflags= [KNL] Set root filesystem mount option string
5315 rootfstype= [KNL] Set root filesystem type
5317 rootwait [KNL] Wait (indefinitely) for root device to show up.
5318 Useful for devices that are detected asynchronously
5319 (e.g. USB and MMC devices).
5321 rproc_mem=nn[KMG][@address]
5322 [KNL,ARM,CMA] Remoteproc physical memory block.
5323 Memory area to be used by remote processor image,
5326 rw [KNL] Mount root device read-write on boot
5328 S [KNL] Run init in single mode
5330 s390_iommu= [HW,S390]
5331 Set s390 IOTLB flushing mode
5333 With strict flushing every unmap operation will result in
5334 an IOTLB flush. Default is lazy flushing before reuse,
5337 s390_iommu_aperture= [KNL,S390]
5338 Specifies the size of the per device DMA address space
5339 accessible through the DMA and IOMMU APIs as a decimal
5340 factor of the size of main memory.
5341 The default is 1 meaning that one can concurrently use
5342 as many DMA addresses as physical memory is installed,
5343 if supported by hardware, and thus map all of memory
5344 once. With a value of 2 one can map all of memory twice
5345 and so on. As a special case a factor of 0 imposes no
5346 restrictions other than those given by hardware at the
5347 cost of significant additional memory use for tables.
5350 See drivers/net/irda/sa1100_ir.c.
5352 sched_verbose [KNL] Enables verbose scheduler debug messages.
5354 schedstats= [KNL,X86] Enable or disable scheduled statistics.
5355 Allowed values are enable and disable. This feature
5356 incurs a small amount of overhead in the scheduler
5357 but is useful for debugging and performance tuning.
5359 sched_thermal_decay_shift=
5360 [KNL, SMP] Set a decay shift for scheduler thermal
5361 pressure signal. Thermal pressure signal follows the
5362 default decay period of other scheduler pelt
5363 signals(usually 32 ms but configurable). Setting
5364 sched_thermal_decay_shift will left shift the decay
5365 period for the thermal pressure signal by the shift
5367 i.e. with the default pelt decay period of 32 ms
5368 sched_thermal_decay_shift thermal pressure decay pr
5372 Format: integer between 0 and 10
5375 scftorture.holdoff= [KNL]
5376 Number of seconds to hold off before starting
5377 test. Defaults to zero for module insertion and
5378 to 10 seconds for built-in smp_call_function()
5381 scftorture.longwait= [KNL]
5382 Request ridiculously long waits randomly selected
5383 up to the chosen limit in seconds. Zero (the
5384 default) disables this feature. Please note
5385 that requesting even small non-zero numbers of
5386 seconds can result in RCU CPU stall warnings,
5387 softlockup complaints, and so on.
5389 scftorture.nthreads= [KNL]
5390 Number of kthreads to spawn to invoke the
5391 smp_call_function() family of functions.
5392 The default of -1 specifies a number of kthreads
5393 equal to the number of CPUs.
5395 scftorture.onoff_holdoff= [KNL]
5396 Number seconds to wait after the start of the
5397 test before initiating CPU-hotplug operations.
5399 scftorture.onoff_interval= [KNL]
5400 Number seconds to wait between successive
5401 CPU-hotplug operations. Specifying zero (which
5402 is the default) disables CPU-hotplug operations.
5404 scftorture.shutdown_secs= [KNL]
5405 The number of seconds following the start of the
5406 test after which to shut down the system. The
5407 default of zero avoids shutting down the system.
5408 Non-zero values are useful for automated tests.
5410 scftorture.stat_interval= [KNL]
5411 The number of seconds between outputting the
5412 current test statistics to the console. A value
5413 of zero disables statistics output.
5415 scftorture.stutter_cpus= [KNL]
5416 The number of jiffies to wait between each change
5417 to the set of CPUs under test.
5419 scftorture.use_cpus_read_lock= [KNL]
5420 Use use_cpus_read_lock() instead of the default
5421 preempt_disable() to disable CPU hotplug
5422 while invoking one of the smp_call_function*()
5425 scftorture.verbose= [KNL]
5426 Enable additional printk() statements.
5428 scftorture.weight_single= [KNL]
5429 The probability weighting to use for the
5430 smp_call_function_single() function with a zero
5431 "wait" parameter. A value of -1 selects the
5432 default if all other weights are -1. However,
5433 if at least one weight has some other value, a
5434 value of -1 will instead select a weight of zero.
5436 scftorture.weight_single_wait= [KNL]
5437 The probability weighting to use for the
5438 smp_call_function_single() function with a
5439 non-zero "wait" parameter. See weight_single.
5441 scftorture.weight_many= [KNL]
5442 The probability weighting to use for the
5443 smp_call_function_many() function with a zero
5444 "wait" parameter. See weight_single.
5445 Note well that setting a high probability for
5446 this weighting can place serious IPI load
5449 scftorture.weight_many_wait= [KNL]
5450 The probability weighting to use for the
5451 smp_call_function_many() function with a
5452 non-zero "wait" parameter. See weight_single
5455 scftorture.weight_all= [KNL]
5456 The probability weighting to use for the
5457 smp_call_function_all() function with a zero
5458 "wait" parameter. See weight_single and
5461 scftorture.weight_all_wait= [KNL]
5462 The probability weighting to use for the
5463 smp_call_function_all() function with a
5464 non-zero "wait" parameter. See weight_single
5467 skew_tick= [KNL] Offset the periodic timer tick per cpu to mitigate
5468 xtime_lock contention on larger systems, and/or RCU lock
5469 contention on all systems with CONFIG_MAXSMP set.
5470 Format: { "0" | "1" }
5471 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1"
5473 Note: increases power consumption, thus should only be
5474 enabled if running jitter sensitive (HPC/RT) workloads.
5476 security= [SECURITY] Choose a legacy "major" security module to
5477 enable at boot. This has been deprecated by the
5480 selinux= [SELINUX] Disable or enable SELinux at boot time.
5481 Format: { "0" | "1" }
5482 See security/selinux/Kconfig help text.
5487 apparmor= [APPARMOR] Disable or enable AppArmor at boot time
5488 Format: { "0" | "1" }
5489 See security/apparmor/Kconfig help text
5492 Default value is set via kernel config option.
5494 serialnumber [BUGS=X86-32]
5496 sev=option[,option...] [X86-64] See Documentation/x86/x86_64/boot-options.rst
5499 Maximal number of shapers.
5507 Enable merging of slabs with similar size when the
5508 kernel is built without CONFIG_SLAB_MERGE_DEFAULT.
5511 Disable merging of slabs with similar size. May be
5512 necessary if there is some reason to distinguish
5513 allocs to different slabs, especially in hardened
5514 environments where the risk of heap overflows and
5515 layout control by attackers can usually be
5516 frustrated by disabling merging. This will reduce
5517 most of the exposure of a heap attack to a single
5518 cache (risks via metadata attacks are mostly
5519 unchanged). Debug options disable merging on their
5521 For more information see Documentation/mm/slub.rst.
5523 slab_max_order= [MM, SLAB]
5524 Determines the maximum allowed order for slabs.
5525 A high setting may cause OOMs due to memory
5526 fragmentation. Defaults to 1 for systems with
5527 more than 32MB of RAM, 0 otherwise.
5529 slub_debug[=options[,slabs][;[options[,slabs]]...] [MM, SLUB]
5530 Enabling slub_debug allows one to determine the
5531 culprit if slab objects become corrupted. Enabling
5532 slub_debug can create guard zones around objects and
5533 may poison objects when not in use. Also tracks the
5534 last alloc / free. For more information see
5535 Documentation/mm/slub.rst.
5537 slub_max_order= [MM, SLUB]
5538 Determines the maximum allowed order for slabs.
5539 A high setting may cause OOMs due to memory
5540 fragmentation. For more information see
5541 Documentation/mm/slub.rst.
5543 slub_min_objects= [MM, SLUB]
5544 The minimum number of objects per slab. SLUB will
5545 increase the slab order up to slub_max_order to
5546 generate a sufficiently large slab able to contain
5547 the number of objects indicated. The higher the number
5548 of objects the smaller the overhead of tracking slabs
5549 and the less frequently locks need to be acquired.
5550 For more information see Documentation/mm/slub.rst.
5552 slub_min_order= [MM, SLUB]
5553 Determines the minimum page order for slabs. Must be
5554 lower than slub_max_order.
5555 For more information see Documentation/mm/slub.rst.
5557 slub_merge [MM, SLUB]
5558 Same with slab_merge.
5560 slub_nomerge [MM, SLUB]
5561 Same with slab_nomerge. This is supported for legacy.
5562 See slab_nomerge for more information.
5565 Format: <io1>[,<io2>[,...,<io8>]]
5567 smp.csd_lock_timeout= [KNL]
5568 Specify the period of time in milliseconds
5569 that smp_call_function() and friends will wait
5570 for a CPU to release the CSD lock. This is
5571 useful when diagnosing bugs involving CPUs
5572 disabling interrupts for extended periods
5573 of time. Defaults to 5,000 milliseconds, and
5574 setting a value of zero disables this feature.
5575 This feature may be more efficiently disabled
5576 using the csdlock_debug- kernel parameter.
5578 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices
5579 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port
5580 smsc-ircc2.ircc_sir= [HW] SIR base I/O port
5581 smsc-ircc2.ircc_fir= [HW] FIR base I/O port
5582 smsc-ircc2.ircc_irq= [HW] IRQ line
5583 smsc-ircc2.ircc_dma= [HW] DMA channel
5584 smsc-ircc2.ircc_transceiver= [HW] Transceiver type:
5585 0: Toshiba Satellite 1800 (GP data pin select)
5586 1: Fast pin select (default)
5589 smt= [KNL,S390] Set the maximum number of threads (logical
5590 CPUs) to use per physical CPU on systems capable of
5591 symmetric multithreading (SMT). Will be capped to the
5592 actual hardware limit.
5594 Default: -1 (no limit)
5597 [KNL] Should the soft-lockup detector generate panics.
5600 A value of 1 instructs the soft-lockup detector
5601 to panic the machine when a soft-lockup occurs. It is
5602 also controlled by the kernel.softlockup_panic sysctl
5603 and CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC, which is the
5604 respective build-time switch to that functionality.
5606 softlockup_all_cpu_backtrace=
5607 [KNL] Should the soft-lockup detector generate
5608 backtraces on all cpus.
5611 sonypi.*= [HW] Sony Programmable I/O Control Device driver
5612 See Documentation/admin-guide/laptops/sonypi.rst
5614 spectre_v2= [X86] Control mitigation of Spectre variant 2
5615 (indirect branch speculation) vulnerability.
5616 The default operation protects the kernel from
5619 on - unconditionally enable, implies
5621 off - unconditionally disable, implies
5623 auto - kernel detects whether your CPU model is
5626 Selecting 'on' will, and 'auto' may, choose a
5627 mitigation method at run time according to the
5628 CPU, the available microcode, the setting of the
5629 CONFIG_RETPOLINE configuration option, and the
5630 compiler with which the kernel was built.
5632 Selecting 'on' will also enable the mitigation
5633 against user space to user space task attacks.
5635 Selecting 'off' will disable both the kernel and
5636 the user space protections.
5638 Specific mitigations can also be selected manually:
5640 retpoline - replace indirect branches
5641 retpoline,generic - Retpolines
5642 retpoline,lfence - LFENCE; indirect branch
5643 retpoline,amd - alias for retpoline,lfence
5644 eibrs - enhanced IBRS
5645 eibrs,retpoline - enhanced IBRS + Retpolines
5646 eibrs,lfence - enhanced IBRS + LFENCE
5647 ibrs - use IBRS to protect kernel
5649 Not specifying this option is equivalent to
5653 [X86] Control mitigation of Spectre variant 2
5654 (indirect branch speculation) vulnerability between
5657 on - Unconditionally enable mitigations. Is
5658 enforced by spectre_v2=on
5660 off - Unconditionally disable mitigations. Is
5661 enforced by spectre_v2=off
5663 prctl - Indirect branch speculation is enabled,
5664 but mitigation can be enabled via prctl
5665 per thread. The mitigation control state
5666 is inherited on fork.
5669 - Like "prctl" above, but only STIBP is
5670 controlled per thread. IBPB is issued
5671 always when switching between different user
5675 - Same as "prctl" above, but all seccomp
5676 threads will enable the mitigation unless
5677 they explicitly opt out.
5680 - Like "seccomp" above, but only STIBP is
5681 controlled per thread. IBPB is issued
5682 always when switching between different
5683 user space processes.
5685 auto - Kernel selects the mitigation depending on
5686 the available CPU features and vulnerability.
5688 Default mitigation: "prctl"
5690 Not specifying this option is equivalent to
5691 spectre_v2_user=auto.
5693 spec_store_bypass_disable=
5694 [HW] Control Speculative Store Bypass (SSB) Disable mitigation
5695 (Speculative Store Bypass vulnerability)
5697 Certain CPUs are vulnerable to an exploit against a
5698 a common industry wide performance optimization known
5699 as "Speculative Store Bypass" in which recent stores
5700 to the same memory location may not be observed by
5701 later loads during speculative execution. The idea
5702 is that such stores are unlikely and that they can
5703 be detected prior to instruction retirement at the
5704 end of a particular speculation execution window.
5706 In vulnerable processors, the speculatively forwarded
5707 store can be used in a cache side channel attack, for
5708 example to read memory to which the attacker does not
5709 directly have access (e.g. inside sandboxed code).
5711 This parameter controls whether the Speculative Store
5712 Bypass optimization is used.
5714 On x86 the options are:
5716 on - Unconditionally disable Speculative Store Bypass
5717 off - Unconditionally enable Speculative Store Bypass
5718 auto - Kernel detects whether the CPU model contains an
5719 implementation of Speculative Store Bypass and
5720 picks the most appropriate mitigation. If the
5721 CPU is not vulnerable, "off" is selected. If the
5722 CPU is vulnerable the default mitigation is
5723 architecture and Kconfig dependent. See below.
5724 prctl - Control Speculative Store Bypass per thread
5725 via prctl. Speculative Store Bypass is enabled
5726 for a process by default. The state of the control
5727 is inherited on fork.
5728 seccomp - Same as "prctl" above, but all seccomp threads
5729 will disable SSB unless they explicitly opt out.
5731 Default mitigations:
5734 On powerpc the options are:
5736 on,auto - On Power8 and Power9 insert a store-forwarding
5737 barrier on kernel entry and exit. On Power7
5738 perform a software flush on kernel entry and
5742 Not specifying this option is equivalent to
5743 spec_store_bypass_disable=auto.
5745 spia_io_base= [HW,MTD]
5751 [X86] Enable split lock detection or bus lock detection
5753 When enabled (and if hardware support is present), atomic
5754 instructions that access data across cache line
5755 boundaries will result in an alignment check exception
5756 for split lock detection or a debug exception for
5761 warn - the kernel will emit rate-limited warnings
5762 about applications triggering the #AC
5763 exception or the #DB exception. This mode is
5764 the default on CPUs that support split lock
5765 detection or bus lock detection. Default
5766 behavior is by #AC if both features are
5767 enabled in hardware.
5769 fatal - the kernel will send SIGBUS to applications
5770 that trigger the #AC exception or the #DB
5771 exception. Default behavior is by #AC if
5772 both features are enabled in hardware.
5775 Set system wide rate limit to N bus locks
5776 per second for bus lock detection.
5779 N/A for split lock detection.
5782 If an #AC exception is hit in the kernel or in
5783 firmware (i.e. not while executing in user mode)
5784 the kernel will oops in either "warn" or "fatal"
5787 #DB exception for bus lock is triggered only when
5791 Control the Special Register Buffer Data Sampling
5794 Certain CPUs are vulnerable to an MDS-like
5795 exploit which can leak bits from the random
5798 By default, this issue is mitigated by
5799 microcode. However, the microcode fix can cause
5800 the RDRAND and RDSEED instructions to become
5801 much slower. Among other effects, this will
5802 result in reduced throughput from /dev/urandom.
5804 The microcode mitigation can be disabled with
5805 the following option:
5807 off: Disable mitigation and remove
5808 performance impact to RDRAND and RDSEED
5810 srcutree.big_cpu_lim [KNL]
5811 Specifies the number of CPUs constituting a
5812 large system, such that srcu_struct structures
5813 should immediately allocate an srcu_node array.
5814 This kernel-boot parameter defaults to 128,
5815 but takes effect only when the low-order four
5816 bits of srcutree.convert_to_big is equal to 3
5819 srcutree.convert_to_big [KNL]
5820 Specifies under what conditions an SRCU tree
5821 srcu_struct structure will be converted to big
5822 form, that is, with an rcu_node tree:
5825 1: At init_srcu_struct() time.
5826 2: When rcutorture decides to.
5827 3: Decide at boot time (default).
5828 0x1X: Above plus if high contention.
5830 Either way, the srcu_node tree will be sized based
5831 on the actual runtime number of CPUs (nr_cpu_ids)
5832 instead of the compile-time CONFIG_NR_CPUS.
5834 srcutree.counter_wrap_check [KNL]
5835 Specifies how frequently to check for
5836 grace-period sequence counter wrap for the
5837 srcu_data structure's ->srcu_gp_seq_needed field.
5838 The greater the number of bits set in this kernel
5839 parameter, the less frequently counter wrap will
5840 be checked for. Note that the bottom two bits
5843 srcutree.exp_holdoff [KNL]
5844 Specifies how many nanoseconds must elapse
5845 since the end of the last SRCU grace period for
5846 a given srcu_struct until the next normal SRCU
5847 grace period will be considered for automatic
5848 expediting. Set to zero to disable automatic
5851 srcutree.srcu_max_nodelay [KNL]
5852 Specifies the number of no-delay instances
5853 per jiffy for which the SRCU grace period
5854 worker thread will be rescheduled with zero
5855 delay. Beyond this limit, worker thread will
5856 be rescheduled with a sleep delay of one jiffy.
5858 srcutree.srcu_max_nodelay_phase [KNL]
5859 Specifies the per-grace-period phase, number of
5860 non-sleeping polls of readers. Beyond this limit,
5861 grace period worker thread will be rescheduled
5862 with a sleep delay of one jiffy, between each
5863 rescan of the readers, for a grace period phase.
5865 srcutree.srcu_retry_check_delay [KNL]
5866 Specifies number of microseconds of non-sleeping
5867 delay between each non-sleeping poll of readers.
5869 srcutree.small_contention_lim [KNL]
5870 Specifies the number of update-side contention
5871 events per jiffy will be tolerated before
5872 initiating a conversion of an srcu_struct
5873 structure to big form. Note that the value of
5874 srcutree.convert_to_big must have the 0x10 bit
5875 set for contention-based conversions to occur.
5878 Speculative Store Bypass Disable control
5880 On CPUs that are vulnerable to the Speculative
5881 Store Bypass vulnerability and offer a
5882 firmware based mitigation, this parameter
5883 indicates how the mitigation should be used:
5885 force-on: Unconditionally enable mitigation for
5886 for both kernel and userspace
5887 force-off: Unconditionally disable mitigation for
5888 for both kernel and userspace
5889 kernel: Always enable mitigation in the
5890 kernel, and offer a prctl interface
5891 to allow userspace to register its
5892 interest in being mitigated too.
5894 stack_guard_gap= [MM]
5895 override the default stack gap protection. The value
5896 is in page units and it defines how many pages prior
5897 to (for stacks growing down) resp. after (for stacks
5898 growing up) the main stack are reserved for no other
5899 mapping. Default value is 256 pages.
5901 stack_depot_disable= [KNL]
5902 Setting this to true through kernel command line will
5903 disable the stack depot thereby saving the static memory
5904 consumed by the stack hash table. By default this is set
5908 Enabled the stack tracer on boot up.
5910 stacktrace_filter=[function-list]
5911 [FTRACE] Limit the functions that the stack tracer
5912 will trace at boot up. function-list is a comma-separated
5913 list of functions. This list can be changed at run
5914 time by the stack_trace_filter file in the debugfs
5915 tracing directory. Note, this enables stack tracing
5916 and the stacktrace above is not needed.
5920 Set the STI (builtin display/keyboard on the HP-PARISC
5921 machines) console (graphic card) which should be used
5922 as the initial boot-console.
5923 See also comment in drivers/video/console/sticore.c.
5926 See comment in drivers/video/console/sticore.c.
5929 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]]
5934 Enable or disable strict sigaltstack size checks
5935 against the required signal frame size which
5936 depends on the supported FPU features. This can
5937 be used to filter out binaries which have
5938 not yet been made aware of AT_MINSIGSTKSZ.
5940 sunrpc.min_resvport=
5941 sunrpc.max_resvport=
5943 SunRPC servers often require that client requests
5944 originate from a privileged port (i.e. a port in the
5945 range 0 < portnr < 1024).
5946 An administrator who wishes to reserve some of these
5947 ports for other uses may adjust the range that the
5948 kernel's sunrpc client considers to be privileged
5949 using these two parameters to set the minimum and
5950 maximum port values.
5952 sunrpc.svc_rpc_per_connection_limit=
5954 Limit the number of requests that the server will
5955 process in parallel from a single connection.
5956 The default value is 0 (no limit).
5960 Control how the NFS server code allocates CPUs to
5961 service thread pools. Depending on how many NICs
5962 you have and where their interrupts are bound, this
5963 option will affect which CPUs will do NFS serving.
5964 Note: this parameter cannot be changed while the
5965 NFS server is running.
5967 auto the server chooses an appropriate mode
5968 automatically using heuristics
5969 global a single global pool contains all CPUs
5970 percpu one pool for each CPU
5971 pernode one pool for each NUMA node (equivalent
5972 to global on non-NUMA machines)
5974 sunrpc.tcp_slot_table_entries=
5975 sunrpc.udp_slot_table_entries=
5977 Sets the upper limit on the number of simultaneous
5978 RPC calls that can be sent from the client to a
5979 server. Increasing these values may allow you to
5980 improve throughput, but will also increase the
5981 amount of memory reserved for use by the client.
5983 suspend.pm_test_delay=
5985 Sets the number of seconds to remain in a suspend test
5986 mode before resuming the system (see
5987 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG
5988 is set. Default value is 5.
5991 Format: { on | off | y | n | 1 | 0 }
5992 This parameter controls use of the Protected
5993 Execution Facility on pSeries.
5997 Enable accounting of swap in memory resource
5998 controller if no parameter or 1 is given or disable
5999 it if 0 is given (See Documentation/admin-guide/cgroup-v1/memory.rst)
6001 swiotlb= [ARM,IA-64,PPC,MIPS,X86]
6002 Format: { <int> [,<int>] | force | noforce }
6003 <int> -- Number of I/O TLB slabs
6004 <int> -- Second integer after comma. Number of swiotlb
6005 areas with their own lock. Will be rounded up
6007 force -- force using of bounce buffers even if they
6008 wouldn't be automatically used by the kernel
6009 noforce -- Never use bounce buffers (for debugging)
6014 Set a sysctl parameter, right before loading the init
6015 process, as if the value was written to the respective
6016 /proc/sys/... file. Both '.' and '/' are recognized as
6017 separators. Unrecognized parameters and invalid values
6018 are reported in the kernel log. Sysctls registered
6019 later by a loaded module cannot be set this way.
6020 Example: sysctl.vm.swappiness=40
6022 sysfs.deprecated=0|1 [KNL]
6023 Enable/disable old style sysfs layout for old udev
6024 on older distributions. When this option is enabled
6025 very new udev will not work anymore. When this option
6026 is disabled (or CONFIG_SYSFS_DEPRECATED not compiled)
6027 in older udev will not work anymore.
6028 Default depends on CONFIG_SYSFS_DEPRECATED_V2 set in
6029 the kernel configuration.
6031 sysrq_always_enabled
6033 Ignore sysrq setting - this boot parameter will
6034 neutralize any effect of /proc/sys/kernel/sysrq.
6035 Useful for debugging.
6037 tcpmhash_entries= [KNL,NET]
6038 Set the number of tcp_metrics_hash slots.
6039 Default value is 8192 or 16384 depending on total
6040 ram pages. This is used to specify the TCP metrics
6041 cache size. See Documentation/networking/ip-sysctl.rst
6042 "tcp_no_metrics_save" section for more details.
6046 test_suspend= [SUSPEND]
6047 Format: { "mem" | "standby" | "freeze" }[,N]
6048 Specify "mem" (for Suspend-to-RAM) or "standby" (for
6049 standby suspend) or "freeze" (for suspend type freeze)
6050 as the system sleep state during system startup with
6051 the optional capability to repeat N number of times.
6052 The system is woken from this state using a
6053 wakeup-capable RTC alarm.
6055 thash_entries= [KNL,NET]
6056 Set number of hash buckets for TCP connection
6058 thermal.act= [HW,ACPI]
6059 -1: disable all active trip points in all thermal zones
6060 <degrees C>: override all lowest active trip points
6062 thermal.crt= [HW,ACPI]
6063 -1: disable all critical trip points in all thermal zones
6064 <degrees C>: override all critical trip points
6066 thermal.nocrt= [HW,ACPI]
6067 Set to disable actions on ACPI thermal zone
6068 critical and hot trip points.
6070 thermal.off= [HW,ACPI]
6071 1: disable ACPI thermal control
6073 thermal.psv= [HW,ACPI]
6074 -1: disable all passive trip points
6075 <degrees C>: override all passive trip points to this
6078 thermal.tzp= [HW,ACPI]
6079 Specify global default ACPI thermal zone polling rate
6080 <deci-seconds>: poll all this frequency
6081 0: no polling (default)
6084 Force threading of all interrupt handlers except those
6085 marked explicitly IRQF_NO_THREAD.
6089 Specify if the kernel should make use of the cpu
6090 topology information if the hardware supports this.
6091 The scheduler will make use of this information and
6092 e.g. base its process migration decisions on it.
6095 topology_updates= [KNL, PPC, NUMA]
6097 Specify if the kernel should ignore (off)
6098 topology updates sent by the hypervisor to this
6101 torture.disable_onoff_at_boot= [KNL]
6102 Prevent the CPU-hotplug component of torturing
6103 until after init has spawned.
6105 torture.ftrace_dump_at_shutdown= [KNL]
6106 Dump the ftrace buffer at torture-test shutdown,
6107 even if there were no errors. This can be a
6108 very costly operation when many torture tests
6109 are running concurrently, especially on systems
6110 with rotating-rust storage.
6112 torture.verbose_sleep_frequency= [KNL]
6113 Specifies how many verbose printk()s should be
6114 emitted between each sleep. The default of zero
6115 disables verbose-printk() sleeping.
6117 torture.verbose_sleep_duration= [KNL]
6118 Duration of each verbose-printk() sleep in jiffies.
6122 tpm_suspend_pcr=[HW,TPM]
6123 Format: integer pcr id
6124 Specify that at suspend time, the tpm driver
6125 should extend the specified pcr with zeros,
6126 as a workaround for some chips which fail to
6127 flush the last written pcr on TPM_SaveState.
6128 This will guarantee that all the other pcrs
6132 Have the tracepoints sent to printk as well as the
6133 tracing ring buffer. This is useful for early boot up
6134 where the system hangs or reboots and does not give the
6135 option for reading the tracing buffer or performing a
6136 ftrace_dump_on_oops.
6138 To turn off having tracepoints sent to printk,
6139 echo 0 > /proc/sys/kernel/tracepoint_printk
6140 Note, echoing 1 into this file without the
6141 tracepoint_printk kernel cmdline option has no effect.
6143 The tp_printk_stop_on_boot (see below) can also be used
6144 to stop the printing of events to console at
6149 Having tracepoints sent to printk() and activating high
6150 frequency tracepoints such as irq or sched, can cause
6151 the system to live lock.
6153 tp_printk_stop_on_boot [FTRACE]
6154 When tp_printk (above) is set, it can cause a lot of noise
6155 on the console. It may be useful to only include the
6156 printing of events during boot up, as user space may
6157 make the system inoperable.
6159 This command line option will stop the printing of events
6160 to console at the late_initcall_sync() time frame.
6162 trace_buf_size=nn[KMG]
6163 [FTRACE] will set tracing buffer size on each cpu.
6165 trace_clock= [FTRACE] Set the clock used for tracing events
6167 local - Use the per CPU time stamp counter
6168 (converted into nanoseconds). Fast, but
6169 depending on the architecture, may not be
6170 in sync between CPUs.
6171 global - Event time stamps are synchronize across
6172 CPUs. May be slower than the local clock,
6173 but better for some race conditions.
6174 counter - Simple counting of events (1, 2, ..)
6175 note, some counts may be skipped due to the
6176 infrastructure grabbing the clock more than
6178 uptime - Use jiffies as the time stamp.
6179 perf - Use the same clock that perf uses.
6180 mono - Use ktime_get_mono_fast_ns() for time stamps.
6181 mono_raw - Use ktime_get_raw_fast_ns() for time
6183 boot - Use ktime_get_boot_fast_ns() for time stamps.
6184 Architectures may add more clocks. See
6185 Documentation/trace/ftrace.rst for more details.
6187 trace_event=[event-list]
6188 [FTRACE] Set and start specified trace events in order
6189 to facilitate early boot debugging. The event-list is a
6190 comma-separated list of trace events to enable. See
6191 also Documentation/trace/events.rst
6193 trace_options=[option-list]
6194 [FTRACE] Enable or disable tracer options at boot.
6195 The option-list is a comma delimited list of options
6196 that can be enabled or disabled just as if you were
6197 to echo the option name into
6199 /sys/kernel/debug/tracing/trace_options
6201 For example, to enable stacktrace option (to dump the
6202 stack trace of each event), add to the command line:
6204 trace_options=stacktrace
6206 See also Documentation/trace/ftrace.rst "trace options"
6210 [FTRACE] enable this option to disable tracing when a
6211 warning is hit. This turns off "tracing_on". Tracing can
6212 be enabled again by echoing '1' into the "tracing_on"
6213 file located in /sys/kernel/debug/tracing/
6215 This option is useful, as it disables the trace before
6216 the WARNING dump is called, which prevents the trace to
6217 be filled with content caused by the warning output.
6219 This option can also be set at run time via the sysctl
6220 option: kernel/traceoff_on_warning
6222 transparent_hugepage=
6224 Format: [always|madvise|never]
6225 Can be used to control the default behavior of the system
6226 with respect to transparent hugepages.
6227 See Documentation/admin-guide/mm/transhuge.rst
6230 trusted.source= [KEYS]
6232 This parameter identifies the trust source as a backend
6233 for trusted keys implementation. Supported trust
6238 If not specified then it defaults to iterating through
6239 the trust source list starting with TPM and assigns the
6240 first trust source as a backend which is initialized
6241 successfully during iteration.
6245 The RNG used to generate key material for trusted keys.
6248 - the same value as trusted.source: "tpm" or "tee"
6250 If not specified, "default" is used. In this case,
6251 the RNG's choice is left to each individual trust source.
6253 tsc= Disable clocksource stability checks for TSC.
6255 [x86] reliable: mark tsc clocksource as reliable, this
6256 disables clocksource verification at runtime, as well
6257 as the stability checks done at bootup. Used to enable
6258 high-resolution timer mode on older hardware, and in
6259 virtualized environment.
6260 [x86] noirqtime: Do not use TSC to do irq accounting.
6261 Used to run time disable IRQ_TIME_ACCOUNTING on any
6262 platforms where RDTSC is slow and this accounting
6264 [x86] unstable: mark the TSC clocksource as unstable, this
6265 marks the TSC unconditionally unstable at bootup and
6266 avoids any further wobbles once the TSC watchdog notices.
6267 [x86] nowatchdog: disable clocksource watchdog. Used
6268 in situations with strict latency requirements (where
6269 interruptions from clocksource watchdog are not
6272 tsc_early_khz= [X86] Skip early TSC calibration and use the given
6273 value instead. Useful when the early TSC frequency discovery
6274 procedure is not reliable, such as on overclocked systems
6275 with CPUID.16h support and partial CPUID.15h support.
6276 Format: <unsigned int>
6278 tsx= [X86] Control Transactional Synchronization
6279 Extensions (TSX) feature in Intel processors that
6280 support TSX control.
6282 This parameter controls the TSX feature. The options are:
6284 on - Enable TSX on the system. Although there are
6285 mitigations for all known security vulnerabilities,
6286 TSX has been known to be an accelerator for
6287 several previous speculation-related CVEs, and
6288 so there may be unknown security risks associated
6289 with leaving it enabled.
6291 off - Disable TSX on the system. (Note that this
6292 option takes effect only on newer CPUs which are
6293 not vulnerable to MDS, i.e., have
6294 MSR_IA32_ARCH_CAPABILITIES.MDS_NO=1 and which get
6295 the new IA32_TSX_CTRL MSR through a microcode
6296 update. This new MSR allows for the reliable
6297 deactivation of the TSX functionality.)
6299 auto - Disable TSX if X86_BUG_TAA is present,
6300 otherwise enable TSX on the system.
6302 Not specifying this option is equivalent to tsx=off.
6304 See Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
6307 tsx_async_abort= [X86,INTEL] Control mitigation for the TSX Async
6308 Abort (TAA) vulnerability.
6310 Similar to Micro-architectural Data Sampling (MDS)
6311 certain CPUs that support Transactional
6312 Synchronization Extensions (TSX) are vulnerable to an
6313 exploit against CPU internal buffers which can forward
6314 information to a disclosure gadget under certain
6317 In vulnerable processors, the speculatively forwarded
6318 data can be used in a cache side channel attack, to
6319 access data to which the attacker does not have direct
6322 This parameter controls the TAA mitigation. The
6325 full - Enable TAA mitigation on vulnerable CPUs
6328 full,nosmt - Enable TAA mitigation and disable SMT on
6329 vulnerable CPUs. If TSX is disabled, SMT
6330 is not disabled because CPU is not
6331 vulnerable to cross-thread TAA attacks.
6332 off - Unconditionally disable TAA mitigation
6334 On MDS-affected machines, tsx_async_abort=off can be
6335 prevented by an active MDS mitigation as both vulnerabilities
6336 are mitigated with the same mechanism so in order to disable
6337 this mitigation, you need to specify mds=off too.
6339 Not specifying this option is equivalent to
6340 tsx_async_abort=full. On CPUs which are MDS affected
6341 and deploy MDS mitigation, TAA mitigation is not
6342 required and doesn't provide any additional
6346 Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
6348 turbografx.map[2|3]= [HW,JOY]
6349 TurboGraFX parallel port interface
6351 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7>
6352 See also Documentation/input/devices/joystick-parport.rst
6354 udbg-immortal [PPC] When debugging early kernel crashes that
6355 happen after console_init() and before a proper
6356 console driver takes over, this boot options might
6357 help "seeing" what's going on.
6359 uhash_entries= [KNL,NET]
6360 Set number of hash buckets for UDP/UDP-Lite connections
6363 [USB] Ignore overcurrent events (default N).
6364 Some badly-designed motherboards generate lots of
6365 bogus events, for ports that aren't wired to
6366 anything. Set this parameter to avoid log spamming.
6367 Note that genuine overcurrent events won't be
6371 [X86] Cause panic on unknown NMI.
6373 usbcore.authorized_default=
6374 [USB] Default USB device authorization:
6375 (default -1 = authorized except for wireless USB,
6376 0 = not authorized, 1 = authorized, 2 = authorized
6377 if device connected to internal port)
6379 usbcore.autosuspend=
6380 [USB] The autosuspend time delay (in seconds) used
6381 for newly-detected USB devices (default 2). This
6382 is the time required before an idle device will be
6383 autosuspended. Devices for which the delay is set
6384 to a negative value won't be autosuspended at all.
6386 usbcore.usbfs_snoop=
6387 [USB] Set to log all usbfs traffic (default 0 = off).
6389 usbcore.usbfs_snoop_max=
6390 [USB] Maximum number of bytes to snoop in each URB
6393 usbcore.blinkenlights=
6394 [USB] Set to cycle leds on hubs (default 0 = off).
6396 usbcore.old_scheme_first=
6397 [USB] Start with the old device initialization
6398 scheme (default 0 = off).
6400 usbcore.usbfs_memory_mb=
6401 [USB] Memory limit (in MB) for buffers allocated by
6402 usbfs (default = 16, 0 = max = 2047).
6404 usbcore.use_both_schemes=
6405 [USB] Try the other device initialization scheme
6406 if the first one fails (default 1 = enabled).
6408 usbcore.initial_descriptor_timeout=
6409 [USB] Specifies timeout for the initial 64-byte
6410 USB_REQ_GET_DESCRIPTOR request in milliseconds
6411 (default 5000 = 5.0 seconds).
6413 usbcore.nousb [USB] Disable the USB subsystem
6416 [USB] A list of quirk entries to augment the built-in
6417 usb core quirk list. List entries are separated by
6418 commas. Each entry has the form
6419 VendorID:ProductID:Flags. The IDs are 4-digit hex
6420 numbers and Flags is a set of letters. Each letter
6421 will change the built-in quirk; setting it if it is
6422 clear and clearing it if it is set. The letters have
6423 the following meanings:
6424 a = USB_QUIRK_STRING_FETCH_255 (string
6425 descriptors must not be fetched using
6427 b = USB_QUIRK_RESET_RESUME (device can't resume
6428 correctly so reset it instead);
6429 c = USB_QUIRK_NO_SET_INTF (device can't handle
6430 Set-Interface requests);
6431 d = USB_QUIRK_CONFIG_INTF_STRINGS (device can't
6432 handle its Configuration or Interface
6434 e = USB_QUIRK_RESET (device can't be reset
6435 (e.g morph devices), don't use reset);
6436 f = USB_QUIRK_HONOR_BNUMINTERFACES (device has
6437 more interface descriptions than the
6438 bNumInterfaces count, and can't handle
6439 talking to these interfaces);
6440 g = USB_QUIRK_DELAY_INIT (device needs a pause
6441 during initialization, after we read
6442 the device descriptor);
6443 h = USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL (For
6444 high speed and super speed interrupt
6445 endpoints, the USB 2.0 and USB 3.0 spec
6446 require the interval in microframes (1
6447 microframe = 125 microseconds) to be
6448 calculated as interval = 2 ^
6450 Devices with this quirk report their
6451 bInterval as the result of this
6452 calculation instead of the exponent
6453 variable used in the calculation);
6454 i = USB_QUIRK_DEVICE_QUALIFIER (device can't
6455 handle device_qualifier descriptor
6457 j = USB_QUIRK_IGNORE_REMOTE_WAKEUP (device
6458 generates spurious wakeup, ignore
6459 remote wakeup capability);
6460 k = USB_QUIRK_NO_LPM (device can't handle Link
6462 l = USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL
6463 (Device reports its bInterval as linear
6464 frames instead of the USB 2.0
6466 m = USB_QUIRK_DISCONNECT_SUSPEND (Device needs
6467 to be disconnected before suspend to
6468 prevent spurious wakeup);
6469 n = USB_QUIRK_DELAY_CTRL_MSG (Device needs a
6470 pause after every control message);
6471 o = USB_QUIRK_HUB_SLOW_RESET (Hub needs extra
6472 delay after resetting its port);
6473 Example: quirks=0781:5580:bk,0a5c:5834:gij
6476 [USBHID] The interval which mice are to be polled at.
6479 [USBHID] The interval which joysticks are to be polled at.
6482 [USBHID] The interval which keyboards are to be polled at.
6484 usb-storage.delay_use=
6485 [UMS] The delay in seconds before a new device is
6486 scanned for Logical Units (default 1).
6489 [UMS] A list of quirks entries to supplement or
6490 override the built-in unusual_devs list. List
6491 entries are separated by commas. Each entry has
6492 the form VID:PID:Flags where VID and PID are Vendor
6493 and Product ID values (4-digit hex numbers) and
6494 Flags is a set of characters, each corresponding
6495 to a common usb-storage quirk flag as follows:
6496 a = SANE_SENSE (collect more than 18 bytes
6497 of sense data, not on uas);
6498 b = BAD_SENSE (don't collect more than 18
6499 bytes of sense data, not on uas);
6500 c = FIX_CAPACITY (decrease the reported
6501 device capacity by one sector);
6502 d = NO_READ_DISC_INFO (don't use
6503 READ_DISC_INFO command, not on uas);
6504 e = NO_READ_CAPACITY_16 (don't use
6505 READ_CAPACITY_16 command);
6506 f = NO_REPORT_OPCODES (don't use report opcodes
6508 g = MAX_SECTORS_240 (don't transfer more than
6509 240 sectors at a time, uas only);
6510 h = CAPACITY_HEURISTICS (decrease the
6511 reported device capacity by one
6512 sector if the number is odd);
6513 i = IGNORE_DEVICE (don't bind to this
6515 j = NO_REPORT_LUNS (don't use report luns
6517 k = NO_SAME (do not use WRITE_SAME, uas only)
6518 l = NOT_LOCKABLE (don't try to lock and
6519 unlock ejectable media, not on uas);
6520 m = MAX_SECTORS_64 (don't transfer more
6521 than 64 sectors = 32 KB at a time,
6523 n = INITIAL_READ10 (force a retry of the
6524 initial READ(10) command, not on uas);
6525 o = CAPACITY_OK (accept the capacity
6526 reported by the device, not on uas);
6527 p = WRITE_CACHE (the device cache is ON
6528 by default, not on uas);
6529 r = IGNORE_RESIDUE (the device reports
6530 bogus residue values, not on uas);
6531 s = SINGLE_LUN (the device has only one
6533 t = NO_ATA_1X (don't allow ATA(12) and ATA(16)
6534 commands, uas only);
6535 u = IGNORE_UAS (don't bind to the uas driver);
6536 w = NO_WP_DETECT (don't test whether the
6537 medium is write-protected).
6538 y = ALWAYS_SYNC (issue a SYNCHRONIZE_CACHE
6539 even if the device claims no cache,
6541 Example: quirks=0419:aaf5:rl,0421:0433:rc
6543 user_debug= [KNL,ARM]
6545 See arch/arm/Kconfig.debug help text.
6546 1 - undefined instruction events
6548 4 - invalid data aborts
6551 Example: user_debug=31
6554 [X86] Flags controlling user PTE allocations.
6556 nohigh = do not allocate PTE pages in
6557 HIGHMEM regardless of setting
6560 vdso= [X86,SH,SPARC]
6561 On X86_32, this is an alias for vdso32=. Otherwise:
6563 vdso=1: enable VDSO (the default)
6564 vdso=0: disable VDSO mapping
6566 vdso32= [X86] Control the 32-bit vDSO
6567 vdso32=1: enable 32-bit VDSO
6568 vdso32=0 or vdso32=2: disable 32-bit VDSO
6570 See the help text for CONFIG_COMPAT_VDSO for more
6571 details. If CONFIG_COMPAT_VDSO is set, the default is
6572 vdso32=0; otherwise, the default is vdso32=1.
6574 For compatibility with older kernels, vdso32=2 is an
6577 Try vdso32=0 if you encounter an error that says:
6578 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed!
6581 vector=percpu: enable percpu vector domain
6583 video= [FB] Frame buffer configuration
6584 See Documentation/fb/modedb.rst.
6586 video.brightness_switch_enabled= [ACPI]
6588 If set to 1, on receiving an ACPI notify event
6589 generated by hotkey, video driver will adjust brightness
6590 level and then send out the event to user space through
6591 the allocated input device. If set to 0, video driver
6592 will only send out the event without touching backlight
6597 [VMMIO] Memory mapped virtio (platform) device.
6599 <size>@<baseaddr>:<irq>[:<id>]
6601 <size> := size (can use standard suffixes
6603 <baseaddr> := physical base address
6604 <irq> := interrupt number (as passed to
6606 <id> := (optional) platform device id
6608 virtio_mmio.device=1K@0x100b0000:48:7
6610 Can be used multiple times for multiple devices.
6612 vga= [BOOT,X86-32] Select a particular video mode
6613 See Documentation/x86/boot.rst and
6614 Documentation/admin-guide/svga.rst.
6615 Use vga=ask for menu.
6616 This is actually a boot loader parameter; the value is
6617 passed to the kernel using a special protocol.
6619 vm_debug[=options] [KNL] Available with CONFIG_DEBUG_VM=y.
6620 May slow down system boot speed, especially when
6621 enabled on systems with a large amount of memory.
6622 All options are enabled by default, and this
6623 interface is meant to allow for selectively
6624 enabling or disabling specific virtual memory
6627 Available options are:
6628 P Enable page structure init time poisoning
6629 - Disable all of the above options
6631 vmalloc=nn[KMG] [KNL,BOOT] Forces the vmalloc area to have an exact
6632 size of <nn>. This can be used to increase the
6633 minimum size (128MB on x86). It can also be used to
6634 decrease the size and leave more room for directly
6637 vmcp_cma=nn[MG] [KNL,S390]
6638 Sets the memory size reserved for contiguous memory
6639 allocations for the vmcp device driver.
6641 vmhalt= [KNL,S390] Perform z/VM CP command after system halt.
6644 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic.
6647 vmpoff= [KNL,S390] Perform z/VM CP command after power off.
6651 Controls the behavior of vsyscalls (i.e. calls to
6652 fixed addresses of 0xffffffffff600x00 from legacy
6653 code). Most statically-linked binaries and older
6654 versions of glibc use these calls. Because these
6655 functions are at fixed addresses, they make nice
6656 targets for exploits that can control RIP.
6658 emulate [default] Vsyscalls turn into traps and are
6659 emulated reasonably safely. The vsyscall
6662 xonly Vsyscalls turn into traps and are
6663 emulated reasonably safely. The vsyscall
6664 page is not readable.
6666 none Vsyscalls don't work at all. This makes
6667 them quite hard to use for exploits but
6668 might break your system.
6670 vt.color= [VT] Default text color.
6671 Format: 0xYX, X = foreground, Y = background.
6672 Default: 0x07 = light gray on black.
6674 vt.cur_default= [VT] Default cursor shape.
6675 Format: 0xCCBBAA, where AA, BB, and CC are the same as
6676 the parameters of the <Esc>[?A;B;Cc escape sequence;
6677 see VGA-softcursor.txt. Default: 2 = underline.
6679 vt.default_blu= [VT]
6680 Format: <blue0>,<blue1>,<blue2>,...,<blue15>
6681 Change the default blue palette of the console.
6682 This is a 16-member array composed of values
6685 vt.default_grn= [VT]
6686 Format: <green0>,<green1>,<green2>,...,<green15>
6687 Change the default green palette of the console.
6688 This is a 16-member array composed of values
6691 vt.default_red= [VT]
6692 Format: <red0>,<red1>,<red2>,...,<red15>
6693 Change the default red palette of the console.
6694 This is a 16-member array composed of values
6700 Set system-wide default UTF-8 mode for all tty's.
6701 Default is 1, i.e. UTF-8 mode is enabled for all
6702 newly opened terminals.
6704 vt.global_cursor_default=
6707 Set system-wide default for whether a cursor
6708 is shown on new VTs. Default is -1,
6709 i.e. cursors will be created by default unless
6710 overridden by individual drivers. 0 will hide
6711 cursors, 1 will display them.
6713 vt.italic= [VT] Default color for italic text; 0-15.
6716 vt.underline= [VT] Default color for underlined text; 0-15.
6719 watchdog timers [HW,WDT] For information on watchdog timers,
6720 see Documentation/watchdog/watchdog-parameters.rst
6721 or other driver-specific files in the
6722 Documentation/watchdog/ directory.
6726 Set the hard lockup detector stall duration
6727 threshold in seconds. The soft lockup detector
6728 threshold is set to twice the value. A value of 0
6729 disables both lockup detectors. Default is 10
6732 workqueue.watchdog_thresh=
6733 If CONFIG_WQ_WATCHDOG is configured, workqueue can
6734 warn stall conditions and dump internal state to
6735 help debugging. 0 disables workqueue stall
6736 detection; otherwise, it's the stall threshold
6737 duration in seconds. The default value is 30 and
6738 it can be updated at runtime by writing to the
6739 corresponding sysfs file.
6741 workqueue.disable_numa
6742 By default, all work items queued to unbound
6743 workqueues are affine to the NUMA nodes they're
6744 issued on, which results in better behavior in
6745 general. If NUMA affinity needs to be disabled for
6746 whatever reason, this option can be used. Note
6747 that this also can be controlled per-workqueue for
6748 workqueues visible under /sys/bus/workqueue/.
6750 workqueue.power_efficient
6751 Per-cpu workqueues are generally preferred because
6752 they show better performance thanks to cache
6753 locality; unfortunately, per-cpu workqueues tend to
6754 be more power hungry than unbound workqueues.
6756 Enabling this makes the per-cpu workqueues which
6757 were observed to contribute significantly to power
6758 consumption unbound, leading to measurably lower
6759 power usage at the cost of small performance
6762 The default value of this parameter is determined by
6763 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT.
6765 workqueue.debug_force_rr_cpu
6766 Workqueue used to implicitly guarantee that work
6767 items queued without explicit CPU specified are put
6768 on the local CPU. This guarantee is no longer true
6769 and while local CPU is still preferred work items
6770 may be put on foreign CPUs. This debug option
6771 forces round-robin CPU selection to flush out
6772 usages which depend on the now broken guarantee.
6773 When enabled, memory and cache locality will be
6776 x2apic_phys [X86-64,APIC] Use x2apic physical mode instead of
6777 default x2apic cluster mode on platforms
6780 xen_512gb_limit [KNL,X86-64,XEN]
6781 Restricts the kernel running paravirtualized under Xen
6782 to use only up to 512 GB of RAM. The reason to do so is
6783 crash analysis tools and Xen tools for doing domain
6784 save/restore/migration must be enabled to handle larger
6787 xen_emul_unplug= [HW,X86,XEN]
6788 Unplug Xen emulated devices
6789 Format: [unplug0,][unplug1]
6790 ide-disks -- unplug primary master IDE devices
6791 aux-ide-disks -- unplug non-primary-master IDE devices
6792 nics -- unplug network devices
6793 all -- unplug all emulated devices (NICs and IDE disks)
6794 unnecessary -- unplugging emulated devices is
6795 unnecessary even if the host did not respond to
6797 never -- do not unplug even if version check succeeds
6799 xen_legacy_crash [X86,XEN]
6800 Crash from Xen panic notifier, without executing late
6801 panic() code such as dumping handler.
6803 xen_nopvspin [X86,XEN]
6804 Disables the qspinlock slowpath using Xen PV optimizations.
6805 This parameter is obsoleted by "nopvspin" parameter, which
6806 has equivalent effect for XEN platform.
6809 Disables the PV optimizations forcing the HVM guest to
6810 run as generic HVM guest with no PV drivers.
6811 This option is obsoleted by the "nopv" option, which
6812 has equivalent effect for XEN platform.
6814 xen_no_vector_callback
6815 [KNL,X86,XEN] Disable the vector callback for Xen
6816 event channel interrupts.
6818 xen_scrub_pages= [XEN]
6819 Boolean option to control scrubbing pages before giving them back
6820 to Xen, for use by other domains. Can be also changed at runtime
6821 with /sys/devices/system/xen_memory/xen_memory0/scrub_pages.
6822 Default value controlled with CONFIG_XEN_SCRUB_PAGES_DEFAULT.
6824 xen_timer_slop= [X86-64,XEN]
6825 Set the timer slop (in nanoseconds) for the virtual Xen
6826 timers (default is 100000). This adjusts the minimum
6827 delta of virtualized Xen timers, where lower values
6828 improve timer resolution at the expense of processing
6829 more timer interrupts.
6831 xen.balloon_boot_timeout= [XEN]
6832 The time (in seconds) to wait before giving up to boot
6833 in case initial ballooning fails to free enough memory.
6834 Applies only when running as HVM or PVH guest and
6835 started with less memory configured than allowed at
6836 max. Default is 180.
6838 xen.event_eoi_delay= [XEN]
6839 How long to delay EOI handling in case of event
6840 storms (jiffies). Default is 10.
6842 xen.event_loop_timeout= [XEN]
6843 After which time (jiffies) the event handling loop
6844 should start to delay EOI handling. Default is 2.
6846 xen.fifo_events= [XEN]
6847 Boolean parameter to disable using fifo event handling
6848 even if available. Normally fifo event handling is
6849 preferred over the 2-level event handling, as it is
6850 fairer and the number of possible event channels is
6851 much higher. Default is on (use fifo events).
6853 nopv= [X86,XEN,KVM,HYPER_V,VMWARE]
6854 Disables the PV optimizations forcing the guest to run
6855 as generic guest with no PV drivers. Currently support
6856 XEN HVM, KVM, HYPER_V and VMWARE guest.
6858 nopvspin [X86,XEN,KVM]
6859 Disables the qspinlock slow path using PV optimizations
6860 which allow the hypervisor to 'idle' the guest on lock
6863 xirc2ps_cs= [NET,PCMCIA]
6865 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]]
6868 By default on POWER9 and above, the kernel will
6869 natively use the XIVE interrupt controller. This option
6870 allows the fallback firmware mode to be used:
6872 off Fallback to firmware control of XIVE interrupt
6873 controller on both pseries and powernv
6874 platforms. Only useful on POWER9 and above.
6876 xive.store-eoi=off [PPC]
6877 By default on POWER10 and above, the kernel will use
6878 stores for EOI handling when the XIVE interrupt mode
6879 is active. This option allows the XIVE driver to use
6880 loads instead, as on POWER9.
6882 xhci-hcd.quirks [USB,KNL]
6883 A hex value specifying bitmask with supplemental xhci
6884 host controller quirks. Meaning of each bit can be
6885 consulted in header drivers/usb/host/xhci.h.
6888 Format: { early | on | rw | ro | off }
6889 Controls if xmon debugger is enabled. Default is off.
6890 Passing only "xmon" is equivalent to "xmon=early".
6891 early Call xmon as early as possible on boot; xmon
6892 debugger is called from setup_arch().
6893 on xmon debugger hooks will be installed so xmon
6894 is only called on a kernel crash. Default mode,
6895 i.e. either "ro" or "rw" mode, is controlled
6896 with CONFIG_XMON_DEFAULT_RO_MODE.
6897 rw xmon debugger hooks will be installed so xmon
6898 is called only on a kernel crash, mode is write,
6899 meaning SPR registers, memory and, other data
6900 can be written using xmon commands.
6901 ro same as "rw" option above but SPR registers,
6902 memory, and other data can't be written using
6904 off xmon is disabled.
git.samba.org / sfrench / cifs-2.6.git / blob