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_nohwsig,
229 old_ordering, nonvs, sci_force_enable, nobl }
230 See Documentation/power/video.rst for information on
232 s3_beep is for debugging; it makes the PC's speaker beep
233 as soon as the kernel's real-mode entry point is called.
234 s4_nohwsig prevents ACPI hardware signature from being
235 used during resume from hibernation.
236 old_ordering causes the ACPI 1.0 ordering of the _PTS
237 control method, with respect to putting devices into
238 low power states, to be enforced (the ACPI 2.0 ordering
239 of _PTS is used by default).
240 nonvs prevents the kernel from saving/restoring the
241 ACPI NVS memory during suspend/hibernation and resume.
242 sci_force_enable causes the kernel to set SCI_EN directly
243 on resume from S1/S3 (which is against the ACPI spec,
244 but some broken systems don't work without it).
245 nobl causes the internal blacklist of systems known to
246 behave incorrectly in some ways with respect to system
247 suspend and resume to be ignored (use wisely).
249 acpi_use_timer_override [HW,ACPI]
250 Use timer override. For some broken Nvidia NF5 boards
251 that require a timer override, but don't have HPET
253 add_efi_memmap [EFI; X86] Include EFI memory map in
254 kernel's map of available physical RAM.
257 { off | try_unsupported }
258 off: disable AGP support
259 try_unsupported: try to drive unsupported chipsets
260 (may crash computer or cause data corruption)
263 See Documentation/sound/alsa-configuration.rst
266 Allow the default userspace alignment fault handler
267 behaviour to be specified. Bit 0 enables warnings,
268 bit 1 enables fixups, and bit 2 sends a segfault.
270 align_va_addr= [X86-64]
271 Align virtual addresses by clearing slice [14:12] when
272 allocating a VMA at process creation time. This option
273 gives you up to 3% performance improvement on AMD F15h
274 machines (where it is enabled by default) for a
275 CPU-intensive style benchmark, and it can vary highly in
276 a microbenchmark depending on workload and compiler.
278 32: only for 32-bit processes
279 64: only for 64-bit processes
280 on: enable for both 32- and 64-bit processes
281 off: disable for both 32- and 64-bit processes
283 alloc_snapshot [FTRACE]
284 Allocate the ftrace snapshot buffer on boot up when the
285 main buffer is allocated. This is handy if debugging
286 and you need to use tracing_snapshot() on boot up, and
287 do not want to use tracing_snapshot_alloc() as it needs
288 to be done where GFP_KERNEL allocations are allowed.
290 amd_iommu= [HW,X86-64]
291 Pass parameters to the AMD IOMMU driver in the system.
293 fullflush - enable flushing of IO/TLB entries when
294 they are unmapped. Otherwise they are
295 flushed before they will be reused, which
297 off - do not initialize any AMD IOMMU found in
299 force_isolation - Force device isolation for all
300 devices. The IOMMU driver is not
301 allowed anymore to lift isolation
302 requirements as needed. This option
303 does not override iommu=pt
304 force_enable - Force enable the IOMMU on platforms known
305 to be buggy with IOMMU enabled. Use this
308 amd_iommu_dump= [HW,X86-64]
309 Enable AMD IOMMU driver option to dump the ACPI table
310 for AMD IOMMU. With this option enabled, AMD IOMMU
311 driver will print ACPI tables for AMD IOMMU during
312 IOMMU initialization.
314 amd_iommu_intr= [HW,X86-64]
315 Specifies one of the following AMD IOMMU interrupt
317 legacy - Use legacy interrupt remapping mode.
318 vapic - Use virtual APIC mode, which allows IOMMU
319 to inject interrupts directly into guest.
320 This mode requires kvm-amd.avic=1.
321 (Default when IOMMU HW support is present.)
323 amijoy.map= [HW,JOY] Amiga joystick support
324 Map of devices attached to JOY0DAT and JOY1DAT
326 See also Documentation/input/joydev/joystick.rst
328 analog.map= [HW,JOY] Analog joystick and gamepad support
329 Specifies type or capabilities of an analog joystick
330 connected to one of 16 gameports
331 Format: <type1>,<type2>,..<type16>
334 Power management functions (SPARCstation-4/5 + deriv.)
336 Disable APC CPU standby support. SPARCstation-Fox does
337 not play well with APC CPU idle - disable it if you have
338 APC and your system crashes randomly.
340 apic= [APIC,X86] Advanced Programmable Interrupt Controller
341 Change the output verbosity while booting
342 Format: { quiet (default) | verbose | debug }
343 Change the amount of debugging information output
344 when initialising the APIC and IO-APIC components.
345 For X86-32, this can also be used to specify an APIC
347 Format: apic=driver_name
348 Examples: apic=bigsmp
350 apic_extnmi= [APIC,X86] External NMI delivery setting
351 Format: { bsp (default) | all | none }
352 bsp: External NMI is delivered only to CPU 0
353 all: External NMIs are broadcast to all CPUs as a
355 none: External NMI is masked for all CPUs. This is
356 useful so that a dump capture kernel won't be
360 See Documentation/networking/ipv6.rst.
362 show_lapic= [APIC,X86] Advanced Programmable Interrupt Controller
363 Limit apic dumping. The parameter defines the maximal
364 number of local apics being dumped. Also it is possible
365 to set it to "all" by meaning -- no limit here.
366 Format: { 1 (default) | 2 | ... | all }.
367 The parameter valid if only apic=debug or
368 apic=verbose is specified.
369 Example: apic=debug show_lapic=all
371 apm= [APM] Advanced Power Management
372 See header of arch/x86/kernel/apm_32.c.
374 arcrimi= [HW,NET] ARCnet - "RIM I" (entirely mem-mapped) cards
375 Format: <io>,<irq>,<nodeID>
377 arm64.nobti [ARM64] Unconditionally disable Branch Target
378 Identification support
380 arm64.nopauth [ARM64] Unconditionally disable Pointer Authentication
385 atarimouse= [HW,MOUSE] Atari Mouse
387 atkbd.extra= [HW] Enable extra LEDs and keys on IBM RapidAccess,
388 EzKey and similar keyboards
390 atkbd.reset= [HW] Reset keyboard during initialization
392 atkbd.set= [HW] Select keyboard code set
393 Format: <int> (2 = AT (default), 3 = PS/2)
395 atkbd.scroll= [HW] Enable scroll wheel on MS Office and similar
398 atkbd.softraw= [HW] Choose between synthetic and real raw mode
399 Format: <bool> (0 = real, 1 = synthetic (default))
401 atkbd.softrepeat= [HW]
402 Use software keyboard repeat
404 audit= [KNL] Enable the audit sub-system
405 Format: { "0" | "1" | "off" | "on" }
406 0 | off - kernel audit is disabled and can not be
407 enabled until the next reboot
408 unset - kernel audit is initialized but disabled and
409 will be fully enabled by the userspace auditd.
410 1 | on - kernel audit is initialized and partially
411 enabled, storing at most audit_backlog_limit
412 messages in RAM until it is fully enabled by the
416 audit_backlog_limit= [KNL] Set the audit queue size limit.
417 Format: <int> (must be >=0)
420 bau= [X86_UV] Enable the BAU on SGI UV. The default
421 behavior is to disable the BAU (i.e. bau=0).
422 Format: { "0" | "1" }
425 unset - Disable the BAU.
427 baycom_epp= [HW,AX25]
430 baycom_par= [HW,AX25] BayCom Parallel Port AX.25 Modem
432 See header of drivers/net/hamradio/baycom_par.c.
434 baycom_ser_fdx= [HW,AX25]
435 BayCom Serial Port AX.25 Modem (Full Duplex Mode)
436 Format: <io>,<irq>,<mode>[,<baud>]
437 See header of drivers/net/hamradio/baycom_ser_fdx.c.
439 baycom_ser_hdx= [HW,AX25]
440 BayCom Serial Port AX.25 Modem (Half Duplex Mode)
441 Format: <io>,<irq>,<mode>
442 See header of drivers/net/hamradio/baycom_ser_hdx.c.
444 blkdevparts= Manual partition parsing of block device(s) for
445 embedded devices based on command line input.
446 See Documentation/block/cmdline-partition.rst
448 boot_delay= Milliseconds to delay each printk during boot.
449 Values larger than 10 seconds (10000) are changed to
454 Extended command line options can be added to an initrd
455 and this will cause the kernel to look for it.
457 See Documentation/admin-guide/bootconfig.rst
460 Disable BERT OS support on buggy BIOSes.
462 bgrt_disable [ACPI][X86]
463 Disable BGRT to avoid flickering OEM logo.
465 bttv.card= [HW,V4L] bttv (bt848 + bt878 based grabber cards)
466 bttv.radio= Most important insmod options are available as
468 bttv.pll= See Documentation/admin-guide/media/bttv.rst
471 bulk_remove=off [PPC] This parameter disables the use of the pSeries
472 firmware feature for flushing multiple hpte entries
475 c101= [NET] Moxa C101 synchronous serial card
477 cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection.
478 Sometimes CPU hardware bugs make them report the cache
479 size incorrectly. The kernel will attempt work arounds
480 to fix known problems, but for some CPUs it is not
481 possible to determine what the correct size should be.
482 This option provides an override for these situations.
485 [NET] Specifies amount of time (in seconds) that
486 the kernel should wait for a network carrier. By default
487 it waits 120 seconds.
489 ca_keys= [KEYS] This parameter identifies a specific key(s) on
490 the system trusted keyring to be used for certificate
492 format: { id:<keyid> | builtin }
494 cca= [MIPS] Override the kernel pages' cache coherency
495 algorithm. Accepted values range from 0 to 7
496 inclusive. See arch/mips/include/asm/pgtable-bits.h
497 for platform specific values (SB1, Loongson3 and
500 ccw_timeout_log [S390]
501 See Documentation/s390/common_io.rst for details.
503 cgroup_disable= [KNL] Disable a particular controller or optional feature
504 Format: {name of the controller(s) or feature(s) to disable}
505 The effects of cgroup_disable=foo are:
506 - foo isn't auto-mounted if you mount all cgroups in
508 - foo isn't visible as an individually mountable
510 - if foo is an optional feature then the feature is
511 disabled and corresponding cgroup files are not
513 {Currently only "memory" controller deal with this and
514 cut the overhead, others just disable the usage. So
515 only cgroup_disable=memory is actually worthy}
516 Specifying "pressure" disables per-cgroup pressure
517 stall information accounting feature
519 cgroup_no_v1= [KNL] Disable cgroup controllers and named hierarchies in v1
520 Format: { { controller | "all" | "named" }
521 [,{ controller | "all" | "named" }...] }
522 Like cgroup_disable, but only applies to cgroup v1;
523 the blacklisted controllers remain available in cgroup2.
524 "all" blacklists all controllers and "named" disables
525 named mounts. Specifying both "all" and "named" disables
528 cgroup.memory= [KNL] Pass options to the cgroup memory controller.
530 nosocket -- Disable socket memory accounting.
531 nokmem -- Disable kernel memory accounting.
533 checkreqprot [SELINUX] Set initial checkreqprot flag value.
534 Format: { "0" | "1" }
535 See security/selinux/Kconfig help text.
536 0 -- check protection applied by kernel (includes
537 any implied execute protection).
538 1 -- check protection requested by application.
539 Default value is set via a kernel config option.
540 Value can be changed at runtime via
541 /sys/fs/selinux/checkreqprot.
542 Setting checkreqprot to 1 is deprecated.
545 See Documentation/s390/common_io.rst for details.
548 Prevents the clock framework from automatically gating
549 clocks that have not been explicitly enabled by a Linux
550 device driver but are enabled in hardware at reset or
551 by the bootloader/firmware. Note that this does not
552 force such clocks to be always-on nor does it reserve
553 those clocks in any way. This parameter is useful for
554 debug and development, but should not be needed on a
555 platform with proper driver support. For more
556 information, see Documentation/driver-api/clk.rst.
558 clock= [BUGS=X86-32, HW] gettimeofday clocksource override.
560 Forces specified clocksource (if available) to be used
561 when calculating gettimeofday(). If specified
562 clocksource is not available, it defaults to PIT.
563 Format: { pit | tsc | cyclone | pmtmr }
565 clocksource= Override the default clocksource
567 Override the default clocksource and use the clocksource
568 with the name specified.
569 Some clocksource names to choose from, depending on
571 [all] jiffies (this is the base, fallback clocksource)
573 [ARM] imx_timer1,OSTS,netx_timer,mpu_timer2,
574 pxa_timer,timer3,32k_counter,timer0_1
575 [X86-32] pit,hpet,tsc;
576 scx200_hrt on Geode; cyclone on IBM x440
584 clocksource.arm_arch_timer.evtstrm=
587 Enable/disable the eventstream feature of the ARM
588 architected timer so that code using WFE-based polling
589 loops can be debugged more effectively on production
592 clocksource.max_cswd_read_retries= [KNL]
593 Number of clocksource_watchdog() retries due to
594 external delays before the clock will be marked
595 unstable. Defaults to three retries, that is,
596 four attempts to read the clock under test.
598 clocksource.verify_n_cpus= [KNL]
599 Limit the number of CPUs checked for clocksources
600 marked with CLOCK_SOURCE_VERIFY_PERCPU that
601 are marked unstable due to excessive skew.
602 A negative value says to check all CPUs, while
603 zero says not to check any. Values larger than
604 nr_cpu_ids are silently truncated to nr_cpu_ids.
605 The actual CPUs are chosen randomly, with
606 no replacement if the same CPU is chosen twice.
608 clocksource-wdtest.holdoff= [KNL]
609 Set the time in seconds that the clocksource
610 watchdog test waits before commencing its tests.
611 Defaults to zero when built as a module and to
612 10 seconds when built into the kernel.
614 clearcpuid=BITNUM[,BITNUM...] [X86]
615 Disable CPUID feature X for the kernel. See
616 arch/x86/include/asm/cpufeatures.h for the valid bit
617 numbers. Note the Linux specific bits are not necessarily
618 stable over kernel options, but the vendor specific
620 Also note that user programs calling CPUID directly
621 or using the feature without checking anything
622 will still see it. This just prevents it from
623 being used by the kernel or shown in /proc/cpuinfo.
624 Also note the kernel might malfunction if you disable
627 cma=nn[MG]@[start[MG][-end[MG]]]
629 Sets the size of kernel global memory area for
630 contiguous memory allocations and optionally the
631 placement constraint by the physical address range of
632 memory allocations. A value of 0 disables CMA
633 altogether. For more information, see
634 kernel/dma/contiguous.c
638 Sets the size of kernel per-numa memory area for
639 contiguous memory allocations. A value of 0 disables
640 per-numa CMA altogether. And If this option is not
641 specificed, the default value is 0.
642 With per-numa CMA enabled, DMA users on node nid will
643 first try to allocate buffer from the pernuma area
644 which is located in node nid, if the allocation fails,
645 they will fallback to the global default memory area.
647 cmo_free_hint= [PPC] Format: { yes | no }
648 Specify whether pages are marked as being inactive
649 when they are freed. This is used in CMO environments
650 to determine OS memory pressure for page stealing by
654 coherent_pool=nn[KMG] [ARM,KNL]
655 Sets the size of memory pool for coherent, atomic dma
656 allocations, by default set to 256K.
658 com20020= [HW,NET] ARCnet - COM20020 chipset
660 <io>[,<irq>[,<nodeID>[,<backplane>[,<ckp>[,<timeout>]]]]]
662 com90io= [HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers)
666 ARCnet - COM90xx chipset (memory-mapped buffers)
667 Format: <io>[,<irq>[,<memstart>]]
669 condev= [HW,S390] console device
672 console= [KNL] Output console device and options.
674 tty<n> Use the virtual console device <n>.
678 Use the specified serial port. The options are of
679 the form "bbbbpnf", where "bbbb" is the baud rate,
680 "p" is parity ("n", "o", or "e"), "n" is number of
681 bits, and "f" is flow control ("r" for RTS or
682 omit it). Default is "9600n8".
684 See Documentation/admin-guide/serial-console.rst for more
686 Documentation/networking/netconsole.rst for an
689 uart[8250],io,<addr>[,options]
690 uart[8250],mmio,<addr>[,options]
691 uart[8250],mmio16,<addr>[,options]
692 uart[8250],mmio32,<addr>[,options]
693 uart[8250],0x<addr>[,options]
694 Start an early, polled-mode console on the 8250/16550
695 UART at the specified I/O port or MMIO address,
696 switching to the matching ttyS device later.
697 MMIO inter-register address stride is either 8-bit
698 (mmio), 16-bit (mmio16), or 32-bit (mmio32).
699 If none of [io|mmio|mmio16|mmio32], <addr> is assumed
700 to be equivalent to 'mmio'. 'options' are specified in
701 the same format described for ttyS above; if unspecified,
702 the h/w is not re-initialized.
704 hvc<n> Use the hypervisor console device <n>. This is for
705 both Xen and PowerPC hypervisors.
707 If the device connected to the port is not a TTY but a braille
708 device, prepend "brl," before the device type, for instance
710 For now, only VisioBraille is supported.
713 [KNL] Change console messages format
715 By default we print messages on consoles in
716 "[time stamp] text\n" format (time stamp may not be
717 printed, depending on CONFIG_PRINTK_TIME or
718 `printk_time' param).
720 Switch to syslog format: "<%u>[time stamp] text\n"
721 IOW, each message will have a facility and loglevel
722 prefix. The format is similar to one used by syslog()
723 syscall, or to executing "dmesg -S --raw" or to reading
726 consoleblank= [KNL] The console blank (screen saver) timeout in
727 seconds. A value of 0 disables the blank timer.
731 [KNL] Change the default value for
732 /proc/<pid>/coredump_filter.
733 See also Documentation/filesystems/proc.rst.
735 coresight_cpu_debug.enable
738 Enable/disable the CPU sampling based debugging.
739 0: default value, disable debugging
740 1: enable debugging at boot time
742 cpuidle.off=1 [CPU_IDLE]
743 disable the cpuidle sub-system
746 [CPU_IDLE] Name of the cpuidle governor to use.
748 cpufreq.off=1 [CPU_FREQ]
749 disable the cpufreq sub-system
751 cpufreq.default_governor=
752 [CPU_FREQ] Name of the default cpufreq governor or
753 policy to use. This governor must be registered in the
754 kernel before the cpufreq driver probes.
757 [X86] Delay for N microsec between assert and de-assert
758 of APIC INIT to start processors. This delay occurs
759 on every CPU online, such as boot, and resume from suspend.
762 cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver
764 <first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>]
766 crashkernel=size[KMG][@offset[KMG]]
767 [KNL] Using kexec, Linux can switch to a 'crash kernel'
768 upon panic. This parameter reserves the physical
769 memory region [offset, offset + size] for that kernel
770 image. If '@offset' is omitted, then a suitable offset
771 is selected automatically.
772 [KNL, X86-64] Select a region under 4G first, and
773 fall back to reserve region above 4G when '@offset'
774 hasn't been specified.
775 See Documentation/admin-guide/kdump/kdump.rst for further details.
777 crashkernel=range1:size1[,range2:size2,...][@offset]
778 [KNL] Same as above, but depends on the memory
779 in the running system. The syntax of range is
780 start-[end] where start and end are both
781 a memory unit (amount[KMG]). See also
782 Documentation/admin-guide/kdump/kdump.rst for an example.
784 crashkernel=size[KMG],high
785 [KNL, X86-64] range could be above 4G. Allow kernel
786 to allocate physical memory region from top, so could
787 be above 4G if system have more than 4G ram installed.
788 Otherwise memory region will be allocated below 4G, if
790 It will be ignored if crashkernel=X is specified.
791 crashkernel=size[KMG],low
792 [KNL, X86-64] range under 4G. When crashkernel=X,high
793 is passed, kernel could allocate physical memory region
794 above 4G, that cause second kernel crash on system
795 that require some amount of low memory, e.g. swiotlb
796 requires at least 64M+32K low memory, also enough extra
797 low memory is needed to make sure DMA buffers for 32-bit
798 devices won't run out. Kernel would try to allocate at
799 at least 256M below 4G automatically.
800 This one let user to specify own low range under 4G
801 for second kernel instead.
802 0: to disable low allocation.
803 It will be ignored when crashkernel=X,high is not used
804 or memory reserved is below 4G.
807 [KNL] Disable crypto self-tests
812 cs89x0_media= [HW,NET]
813 Format: { rj45 | aui | bnc }
815 csdlock_debug= [KNL] Enable debug add-ons of cross-CPU function call
816 handling. When switched on, additional debug data is
817 printed to the console in case a hanging CPU is
818 detected, and that CPU is pinged again in order to try
819 to resolve the hang situation.
820 0: disable csdlock debugging (default)
821 1: enable basic csdlock debugging (minor impact)
822 ext: enable extended csdlock debugging (more impact,
826 See header of drivers/s390/block/dasd_devmap.c.
828 db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port
829 (one device per port)
830 Format: <port#>,<type>
831 See also Documentation/input/devices/joystick-parport.rst
833 ddebug_query= [KNL,DYNAMIC_DEBUG] Enable debug messages at early boot
835 Documentation/admin-guide/dynamic-debug-howto.rst for
836 details. Deprecated, see dyndbg.
838 debug [KNL] Enable kernel debugging (events log level).
841 [KNL] Enable printing [hashed] pointers early in the
842 boot sequence. If enabled, we use a weak hash instead
843 of siphash to hash pointers. Use this option if you are
844 seeing instances of '(___ptrval___)') and need to see a
845 value (hashed pointer) instead. Cryptographically
846 insecure, please do not use on production kernels.
849 [KNL] verbose locking self-tests
851 Print debugging info while doing the locking API
853 Bitmask for the various LOCKTYPE_ tests. Defaults to 0
854 (no extra messages), setting it to -1 (all bits set)
855 will print _a_lot_ more information - normally only
856 useful to lockdep developers.
858 debug_objects [KNL] Enable object debugging
861 [KNL] Disable object debugging
863 debug_guardpage_minorder=
864 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this
865 parameter allows control of the order of pages that will
866 be intentionally kept free (and hence protected) by the
867 buddy allocator. Bigger value increase the probability
868 of catching random memory corruption, but reduce the
869 amount of memory for normal system use. The maximum
870 possible value is MAX_ORDER/2. Setting this parameter
871 to 1 or 2 should be enough to identify most random
872 memory corruption problems caused by bugs in kernel or
873 driver code when a CPU writes to (or reads from) a
874 random memory location. Note that there exists a class
875 of memory corruptions problems caused by buggy H/W or
876 F/W or by drivers badly programing DMA (basically when
877 memory is written at bus level and the CPU MMU is
878 bypassed) which are not detectable by
879 CONFIG_DEBUG_PAGEALLOC, hence this option will not help
880 tracking down these problems.
883 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this parameter
884 enables the feature at boot time. By default, it is
885 disabled and the system will work mostly the same as a
886 kernel built without CONFIG_DEBUG_PAGEALLOC.
887 Note: to get most of debug_pagealloc error reports, it's
888 useful to also enable the page_owner functionality.
889 on: enable the feature
891 debugfs= [KNL] This parameter enables what is exposed to userspace
892 and debugfs internal clients.
893 Format: { on, no-mount, off }
894 on: All functions are enabled.
896 Filesystem is not registered but kernel clients can
897 access APIs and a crashkernel can be used to read
898 its content. There is nothing to mount.
899 off: Filesystem is not registered and clients
900 get a -EPERM as result when trying to register files
901 or directories within debugfs.
902 This is equivalent of the runtime functionality if
903 debugfs was not enabled in the kernel at all.
904 Default value is set in build-time with a kernel configuration.
906 debugpat [X86] Enable PAT debugging
908 decnet.addr= [HW,NET]
909 Format: <area>[,<node>]
910 See also Documentation/networking/decnet.rst.
913 [HW] The size of the default HugeTLB page. This is
914 the size represented by the legacy /proc/ hugepages
915 APIs. In addition, this is the default hugetlb size
916 used for shmget(), mmap() and mounting hugetlbfs
917 filesystems. If not specified, defaults to the
918 architecture's default huge page size. Huge page
919 sizes are architecture dependent. See also
920 Documentation/admin-guide/mm/hugetlbpage.rst.
923 deferred_probe_timeout=
924 [KNL] Debugging option to set a timeout in seconds for
925 deferred probe to give up waiting on dependencies to
926 probe. Only specific dependencies (subsystems or
927 drivers) that have opted in will be ignored. A timeout of 0
928 will timeout at the end of initcalls. This option will also
929 dump out devices still on the deferred probe list after
933 Format: { on | off | def_only | inf_only | always }
934 on: s390 zlib hardware support for compression on
935 level 1 and decompression (default)
936 off: No s390 zlib hardware support
937 def_only: s390 zlib hardware support for deflate
938 only (compression on level 1)
939 inf_only: s390 zlib hardware support for inflate
941 always: Same as 'on' but ignores the selected compression
942 level always using hardware support (used for debugging)
945 Set number of hash buckets for dentry cache.
947 disable_1tb_segments [PPC]
948 Disables the use of 1TB hash page table segments. This
949 causes the kernel to fall back to 256MB segments which
950 can be useful when debugging issues that require an SLB
954 Limits the number of kernel SLB entries, and flushes
955 them frequently to increase the rate of SLB faults
959 See Documentation/networking/ipv6.rst.
962 [KNL] Under CONFIG_HARDENED_USERCOPY, whether
963 hardening is enabled for this boot. Hardened
964 usercopy checking is used to protect the kernel
965 from reading or writing beyond known memory
966 allocation boundaries as a proactive defense
967 against bounds-checking flaws in the kernel's
968 copy_to_user()/copy_from_user() interface.
969 on Perform hardened usercopy checks (default).
970 off Disable hardened usercopy checks.
973 Disable RADIX MMU mode on POWER9
975 radix_hcall_invalidate=on [PPC/PSERIES]
976 Disable RADIX GTSE feature and use hcall for TLB
980 Disable TLBIE instruction. Currently does not work
981 with KVM, with HASH MMU, or with coherent accelerators.
983 disable_cpu_apicid= [X86,APIC,SMP]
985 The number of initial APIC ID for the
986 corresponding CPU to be disabled at boot,
987 mostly used for the kdump 2nd kernel to
988 disable BSP to wake up multiple CPUs without
989 causing system reset or hang due to sending
992 disable_ddw [PPC/PSERIES]
993 Disable Dynamic DMA Window support. Use this
994 to workaround buggy firmware.
997 See Documentation/networking/ipv6.rst.
999 disable_mtrr_cleanup [X86]
1000 The kernel tries to adjust MTRR layout from continuous
1001 to discrete, to make X server driver able to add WB
1002 entry later. This parameter disables that.
1004 disable_mtrr_trim [X86, Intel and AMD only]
1005 By default the kernel will trim any uncacheable
1006 memory out of your available memory pool based on
1007 MTRR settings. This parameter disables that behavior,
1008 possibly causing your machine to run very slowly.
1010 disable_timer_pin_1 [X86]
1011 Disable PIN 1 of APIC timer
1012 Can be useful to work around chipset bugs.
1014 dis_ucode_ldr [X86] Disable the microcode loader.
1016 dma_debug=off If the kernel is compiled with DMA_API_DEBUG support,
1017 this option disables the debugging code at boot.
1019 dma_debug_entries=<number>
1020 This option allows to tune the number of preallocated
1021 entries for DMA-API debugging code. One entry is
1022 required per DMA-API allocation. Use this if the
1023 DMA-API debugging code disables itself because the
1024 architectural default is too low.
1026 dma_debug_driver=<driver_name>
1027 With this option the DMA-API debugging driver
1028 filter feature can be enabled at boot time. Just
1029 pass the driver to filter for as the parameter.
1030 The filter can be disabled or changed to another
1031 driver later using sysfs.
1033 driver_async_probe= [KNL]
1034 List of driver names to be probed asynchronously.
1035 Format: <driver_name1>,<driver_name2>...
1037 drm.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>]
1038 Broken monitors, graphic adapters, KVMs and EDIDless
1039 panels may send no or incorrect EDID data sets.
1040 This parameter allows to specify an EDID data sets
1041 in the /lib/firmware directory that are used instead.
1042 Generic built-in EDID data sets are used, if one of
1043 edid/1024x768.bin, edid/1280x1024.bin,
1044 edid/1680x1050.bin, or edid/1920x1080.bin is given
1045 and no file with the same name exists. Details and
1046 instructions how to build your own EDID data are
1047 available in Documentation/admin-guide/edid.rst. An EDID
1048 data set will only be used for a particular connector,
1049 if its name and a colon are prepended to the EDID
1050 name. Each connector may use a unique EDID data
1051 set by separating the files with a comma. An EDID
1052 data set with no connector name will be used for
1053 any connectors not explicitly specified.
1058 Format: {"off" | "known"}
1059 Control how the dt_cpu_ftrs device-tree binding is
1060 used for CPU feature discovery and setup (if it
1062 off: Do not use it, fall back to legacy cpu table.
1063 known: Do not pass through unknown features to guests
1064 or userspace, only those that the kernel is aware of.
1066 dump_apple_properties [X86]
1067 Dump name and content of EFI device properties on
1068 x86 Macs. Useful for driver authors to determine
1069 what data is available or for reverse-engineering.
1071 dyndbg[="val"] [KNL,DYNAMIC_DEBUG]
1072 <module>.dyndbg[="val"]
1073 Enable debug messages at boot time. See
1074 Documentation/admin-guide/dynamic-debug-howto.rst
1077 nopku [X86] Disable Memory Protection Keys CPU feature found
1080 <module>.async_probe [KNL]
1081 Enable asynchronous probe on this module.
1083 early_ioremap_debug [KNL]
1084 Enable debug messages in early_ioremap support. This
1085 is useful for tracking down temporary early mappings
1086 which are not unmapped.
1088 earlycon= [KNL] Output early console device and options.
1090 When used with no options, the early console is
1091 determined by stdout-path property in device tree's
1092 chosen node or the ACPI SPCR table if supported by
1095 cdns,<addr>[,options]
1096 Start an early, polled-mode console on a Cadence
1097 (xuartps) serial port at the specified address. Only
1098 supported option is baud rate. If baud rate is not
1099 specified, the serial port must already be setup and
1102 uart[8250],io,<addr>[,options]
1103 uart[8250],mmio,<addr>[,options]
1104 uart[8250],mmio32,<addr>[,options]
1105 uart[8250],mmio32be,<addr>[,options]
1106 uart[8250],0x<addr>[,options]
1107 Start an early, polled-mode console on the 8250/16550
1108 UART at the specified I/O port or MMIO address.
1109 MMIO inter-register address stride is either 8-bit
1110 (mmio) or 32-bit (mmio32 or mmio32be).
1111 If none of [io|mmio|mmio32|mmio32be], <addr> is assumed
1112 to be equivalent to 'mmio'. 'options' are specified
1113 in the same format described for "console=ttyS<n>"; if
1114 unspecified, the h/w is not initialized.
1118 Start an early, polled-mode console on a pl011 serial
1119 port at the specified address. The pl011 serial port
1120 must already be setup and configured. Options are not
1121 yet supported. If 'mmio32' is specified, then only
1122 the driver will use only 32-bit accessors to read/write
1123 the device registers.
1126 Start an early console on a litex serial port at the
1127 specified address. The serial port must already be
1128 setup and configured. Options are not yet supported.
1131 Start an early, polled-mode console on a meson serial
1132 port at the specified address. The serial port must
1133 already be setup and configured. Options are not yet
1137 Start an early, polled-mode console on an msm serial
1138 port at the specified address. The serial port
1139 must already be setup and configured. Options are not
1142 msm_serial_dm,<addr>
1143 Start an early, polled-mode console on an msm serial
1144 dm port at the specified address. The serial port
1145 must already be setup and configured. Options are not
1149 Start an early, polled-mode console on a serial port
1150 of an Actions Semi SoC, such as S500 or S900, at the
1151 specified address. The serial port must already be
1152 setup and configured. Options are not yet supported.
1155 Start an early, polled-mode console on a serial port
1156 of an RDA Micro SoC, such as RDA8810PL, at the
1157 specified address. The serial port must already be
1158 setup and configured. Options are not yet supported.
1161 Use RISC-V SBI (Supervisor Binary Interface) for early
1164 smh Use ARM semihosting calls for early console.
1172 Use early console provided by serial driver available
1173 on Samsung SoCs, requires selecting proper type and
1174 a correct base address of the selected UART port. The
1175 serial port must already be setup and configured.
1176 Options are not yet supported.
1179 Start an early, polled-mode console on a lantiq serial
1180 (lqasc) port at the specified address. The serial port
1181 must already be setup and configured. Options are not
1186 Use early console provided by Freescale LP UART driver
1187 found on Freescale Vybrid and QorIQ LS1021A processors.
1188 A valid base address must be provided, and the serial
1189 port must already be setup and configured.
1193 Start an early, polled-mode, output-only console on the
1194 Freescale i.MX UART at the specified address. The UART
1195 must already be setup and configured.
1198 Start an early, polled-mode console on the
1199 Armada 3700 serial port at the specified
1200 address. The serial port must already be setup
1201 and configured. Options are not yet supported.
1204 Start an early, polled-mode console on a Qualcomm
1205 Generic Interface (GENI) based serial port at the
1206 specified address. The serial port must already be
1207 setup and configured. Options are not yet supported.
1210 Start an early, unaccelerated console on the EFI
1211 memory mapped framebuffer (if available). On cache
1212 coherent non-x86 systems that use system memory for
1213 the framebuffer, pass the 'ram' option so that it is
1214 mapped with the correct attributes.
1217 Use early console provided by Freescale LINFlexD UART
1218 serial driver for NXP S32V234 SoCs. A valid base
1219 address must be provided, and the serial port must
1220 already be setup and configured.
1222 earlyprintk= [X86,SH,ARM,M68k,S390]
1226 earlyprintk=serial[,ttySn[,baudrate]]
1227 earlyprintk=serial[,0x...[,baudrate]]
1228 earlyprintk=ttySn[,baudrate]
1229 earlyprintk=dbgp[debugController#]
1230 earlyprintk=pciserial[,force],bus:device.function[,baudrate]
1231 earlyprintk=xdbc[xhciController#]
1233 earlyprintk is useful when the kernel crashes before
1234 the normal console is initialized. It is not enabled by
1235 default because it has some cosmetic problems.
1237 Append ",keep" to not disable it when the real console
1240 Only one of vga, efi, serial, or usb debug port can
1243 Currently only ttyS0 and ttyS1 may be specified by
1244 name. Other I/O ports may be explicitly specified
1245 on some architectures (x86 and arm at least) by
1246 replacing ttySn with an I/O port address, like this:
1247 earlyprintk=serial,0x1008,115200
1248 You can find the port for a given device in
1249 /proc/tty/driver/serial:
1250 2: uart:ST16650V2 port:00001008 irq:18 ...
1252 Interaction with the standard serial driver is not
1255 The VGA and EFI output is eventually overwritten by
1258 The xen output can only be used by Xen PV guests.
1260 The sclp output can only be used on s390.
1262 The optional "force" to "pciserial" enables use of a
1263 PCI device even when its classcode is not of the
1266 edac_report= [HW,EDAC] Control how to report EDAC event
1267 Format: {"on" | "off" | "force"}
1268 on: enable EDAC to report H/W event. May be overridden
1269 by other higher priority error reporting module.
1270 off: disable H/W event reporting through EDAC.
1271 force: enforce the use of EDAC to report H/W event.
1274 ekgdboc= [X86,KGDB] Allow early kernel console debugging
1277 This is designed to be used in conjunction with
1278 the boot argument: earlyprintk=vga
1280 This parameter works in place of the kgdboc parameter
1281 but can only be used if the backing tty is available
1282 very early in the boot process. For early debugging
1283 via a serial port see kgdboc_earlycon instead.
1286 Format: {"off" | "on" | "skip[mbr]"}
1289 Format: { "debug", "disable_early_pci_dma",
1290 "nochunk", "noruntime", "nosoftreserve",
1291 "novamap", "no_disable_early_pci_dma" }
1292 debug: enable misc debug output.
1293 disable_early_pci_dma: disable the busmaster bit on all
1294 PCI bridges while in the EFI boot stub.
1295 nochunk: disable reading files in "chunks" in the EFI
1296 boot stub, as chunking can cause problems with some
1297 firmware implementations.
1298 noruntime : disable EFI runtime services support
1299 nosoftreserve: The EFI_MEMORY_SP (Specific Purpose)
1300 attribute may cause the kernel to reserve the
1301 memory range for a memory mapping driver to
1302 claim. Specify efi=nosoftreserve to disable this
1303 reservation and treat the memory by its base type
1304 (i.e. EFI_CONVENTIONAL_MEMORY / "System RAM").
1305 novamap: do not call SetVirtualAddressMap().
1306 no_disable_early_pci_dma: Leave the busmaster bit set
1307 on all PCI bridges while in the EFI boot stub
1309 efi_no_storage_paranoia [EFI; X86]
1310 Using this parameter you can use more than 50% of
1311 your efi variable storage. Use this parameter only if
1312 you are really sure that your UEFI does sane gc and
1313 fulfills the spec otherwise your board may brick.
1315 efi_fake_mem= nn[KMG]@ss[KMG]:aa[,nn[KMG]@ss[KMG]:aa,..] [EFI; X86]
1316 Add arbitrary attribute to specific memory range by
1317 updating original EFI memory map.
1318 Region of memory which aa attribute is added to is
1321 If efi_fake_mem=2G@4G:0x10000,2G@0x10a0000000:0x10000
1322 is specified, EFI_MEMORY_MORE_RELIABLE(0x10000)
1323 attribute is added to range 0x100000000-0x180000000 and
1324 0x10a0000000-0x1120000000.
1326 If efi_fake_mem=8G@9G:0x40000 is specified, the
1327 EFI_MEMORY_SP(0x40000) attribute is added to
1328 range 0x240000000-0x43fffffff.
1330 Using this parameter you can do debugging of EFI memmap
1331 related features. For example, you can do debugging of
1332 Address Range Mirroring feature even if your box
1333 doesn't support it, or mark specific memory as
1336 efivar_ssdt= [EFI; X86] Name of an EFI variable that contains an SSDT
1337 that is to be dynamically loaded by Linux. If there are
1338 multiple variables with the same name but with different
1339 vendor GUIDs, all of them will be loaded. See
1340 Documentation/admin-guide/acpi/ssdt-overlays.rst for details.
1343 eisa_irq_edge= [PARISC,HW]
1344 See header of drivers/parisc/eisa.c.
1347 See comment before function elanfreq_setup() in
1348 arch/x86/kernel/cpu/cpufreq/elanfreq.c.
1350 elfcorehdr=[size[KMG]@]offset[KMG] [IA64,PPC,SH,X86,S390]
1351 Specifies physical address of start of kernel core
1352 image elf header and optionally the size. Generally
1353 kexec loader will pass this option to capture kernel.
1354 See Documentation/admin-guide/kdump/kdump.rst for details.
1356 enable_mtrr_cleanup [X86]
1357 The kernel tries to adjust MTRR layout from continuous
1358 to discrete, to make X server driver able to add WB
1359 entry later. This parameter enables that.
1361 enable_timer_pin_1 [X86]
1362 Enable PIN 1 of APIC timer
1363 Can be useful to work around chipset bugs
1364 (in particular on some ATI chipsets).
1365 The kernel tries to set a reasonable default.
1367 enforcing [SELINUX] Set initial enforcing status.
1369 See security/selinux/Kconfig help text.
1370 0 -- permissive (log only, no denials).
1371 1 -- enforcing (deny and log).
1373 Value can be changed at runtime via
1374 /sys/fs/selinux/enforce.
1377 Disable Error Record Serialization Table (ERST)
1380 ether= [HW,NET] Ethernet cards parameters
1381 This option is obsoleted by the "netdev=" option, which
1382 has equivalent usage. See its documentation for details.
1386 Permit 'security.evm' to be updated regardless of
1387 current integrity status.
1392 fail_make_request=[KNL]
1393 General fault injection mechanism.
1394 Format: <interval>,<probability>,<space>,<times>
1395 See also Documentation/fault-injection/.
1398 Format: { initns | none }
1399 See Documentation/admin-guide/sysctl/net.rst for
1400 fb_tunnels_only_for_init_ns
1403 See Documentation/admin-guide/blockdev/floppy.rst.
1405 force_pal_cache_flush
1406 [IA-64] Avoid check_sal_cache_flush which may hang on
1407 buggy SAL_CACHE_FLUSH implementations. Using this
1408 parameter will force ia64_sal_cache_flush to call
1409 ia64_pal_cache_flush instead of SAL_CACHE_FLUSH.
1412 Forcefully enable Physical Address Extension (PAE).
1413 Many Pentium M systems disable PAE but may have a
1414 functionally usable PAE implementation.
1415 Warning: use of this parameter will taint the kernel
1416 and may cause unknown problems.
1419 [FTRACE] will set and start the specified tracer
1420 as early as possible in order to facilitate early
1423 ftrace_dump_on_oops[=orig_cpu]
1424 [FTRACE] will dump the trace buffers on oops.
1425 If no parameter is passed, ftrace will dump
1426 buffers of all CPUs, but if you pass orig_cpu, it will
1427 dump only the buffer of the CPU that triggered the
1430 ftrace_filter=[function-list]
1431 [FTRACE] Limit the functions traced by the function
1432 tracer at boot up. function-list is a comma-separated
1433 list of functions. This list can be changed at run
1434 time by the set_ftrace_filter file in the debugfs
1437 ftrace_notrace=[function-list]
1438 [FTRACE] Do not trace the functions specified in
1439 function-list. This list can be changed at run time
1440 by the set_ftrace_notrace file in the debugfs
1443 ftrace_graph_filter=[function-list]
1444 [FTRACE] Limit the top level callers functions traced
1445 by the function graph tracer at boot up.
1446 function-list is a comma-separated list of functions
1447 that can be changed at run time by the
1448 set_graph_function file in the debugfs tracing directory.
1450 ftrace_graph_notrace=[function-list]
1451 [FTRACE] Do not trace from the functions specified in
1452 function-list. This list is a comma-separated list of
1453 functions that can be changed at run time by the
1454 set_graph_notrace file in the debugfs tracing directory.
1456 ftrace_graph_max_depth=<uint>
1457 [FTRACE] Used with the function graph tracer. This is
1458 the max depth it will trace into a function. This value
1459 can be changed at run time by the max_graph_depth file
1460 in the tracefs tracing directory. default: 0 (no limit)
1462 fw_devlink= [KNL] Create device links between consumer and supplier
1463 devices by scanning the firmware to infer the
1464 consumer/supplier relationships. This feature is
1465 especially useful when drivers are loaded as modules as
1466 it ensures proper ordering of tasks like device probing
1467 (suppliers first, then consumers), supplier boot state
1468 clean up (only after all consumers have probed),
1469 suspend/resume & runtime PM (consumers first, then
1471 Format: { off | permissive | on | rpm }
1472 off -- Don't create device links from firmware info.
1473 permissive -- Create device links from firmware info
1474 but use it only for ordering boot state clean
1475 up (sync_state() calls).
1476 on -- Create device links from firmware info and use it
1477 to enforce probe and suspend/resume ordering.
1478 rpm -- Like "on", but also use to order runtime PM.
1480 fw_devlink.strict=<bool>
1481 [KNL] Treat all inferred dependencies as mandatory
1482 dependencies. This only applies for fw_devlink=on|rpm.
1486 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad
1487 support via parallel port (up to 5 devices per port)
1488 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5>
1489 See also Documentation/input/devices/joystick-parport.rst
1493 gart_fix_e820= [X86-64] disable the fix e820 for K8 GART
1497 gcov_persist= [GCOV] When non-zero (default), profiling data for
1498 kernel modules is saved and remains accessible via
1499 debugfs, even when the module is unloaded/reloaded.
1500 When zero, profiling data is discarded and associated
1501 debugfs files are removed at module unload time.
1503 goldfish [X86] Enable the goldfish android emulator platform.
1504 Don't use this when you are not running on the
1507 gpio-mockup.gpio_mockup_ranges
1508 [HW] Sets the ranges of gpiochip of for this device.
1509 Format: <start1>,<end1>,<start2>,<end2>...
1510 gpio-mockup.gpio_mockup_named_lines
1511 [HW] Let the driver know GPIO lines should be named.
1513 gpt [EFI] Forces disk with valid GPT signature but
1514 invalid Protective MBR to be treated as GPT. If the
1515 primary GPT is corrupted, it enables the backup/alternate
1516 GPT to be used instead.
1518 grcan.enable0= [HW] Configuration of physical interface 0. Determines
1519 the "Enable 0" bit of the configuration register.
1522 grcan.enable1= [HW] Configuration of physical interface 1. Determines
1523 the "Enable 0" bit of the configuration register.
1526 grcan.select= [HW] Select which physical interface to use.
1529 grcan.txsize= [HW] Sets the size of the tx buffer.
1530 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0.
1532 grcan.rxsize= [HW] Sets the size of the rx buffer.
1533 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0.
1536 hardlockup_all_cpu_backtrace=
1537 [KNL] Should the hard-lockup detector generate
1538 backtraces on all cpus.
1541 hashdist= [KNL,NUMA] Large hashes allocated during boot
1542 are distributed across NUMA nodes. Defaults on
1543 for 64-bit NUMA, off otherwise.
1544 Format: 0 | 1 (for off | on)
1546 hcl= [IA-64] SGI's Hardware Graph compatibility layer
1548 hd= [EIDE] (E)IDE hard drive subsystem geometry
1549 Format: <cyl>,<head>,<sect>
1552 Disable Hardware Error Source Table (HEST) support;
1553 corresponding firmware-first mode error processing
1554 logic will be disabled.
1556 highmem=nn[KMG] [KNL,BOOT] forces the highmem zone to have an exact
1557 size of <nn>. This works even on boxes that have no
1558 highmem otherwise. This also works to reduce highmem
1559 size on bigger boxes.
1561 highres= [KNL] Enable/disable high resolution timer mode.
1562 Valid parameters: "on", "off"
1567 hpet= [X86-32,HPET] option to control HPET usage
1568 Format: { enable (default) | disable | force |
1570 disable: disable HPET and use PIT instead
1571 force: allow force enabled of undocumented chips (ICH4,
1573 verbose: show contents of HPET registers during setup
1575 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET
1576 registers. Default set by CONFIG_HPET_MMAP_DEFAULT.
1578 hugetlb_cma= [HW,CMA] The size of a CMA area used for allocation
1579 of gigantic hugepages.
1582 Reserve a CMA area of given size and allocate gigantic
1583 hugepages using the CMA allocator. If enabled, the
1584 boot-time allocation of gigantic hugepages is skipped.
1586 hugepages= [HW] Number of HugeTLB pages to allocate at boot.
1587 If this follows hugepagesz (below), it specifies
1588 the number of pages of hugepagesz to be allocated.
1589 If this is the first HugeTLB parameter on the command
1590 line, it specifies the number of pages to allocate for
1591 the default huge page size. See also
1592 Documentation/admin-guide/mm/hugetlbpage.rst.
1596 [HW] The size of the HugeTLB pages. This is used in
1597 conjunction with hugepages (above) to allocate huge
1598 pages of a specific size at boot. The pair
1599 hugepagesz=X hugepages=Y can be specified once for
1600 each supported huge page size. Huge page sizes are
1601 architecture dependent. See also
1602 Documentation/admin-guide/mm/hugetlbpage.rst.
1605 hugetlb_free_vmemmap=
1606 [KNL] Reguires CONFIG_HUGETLB_PAGE_FREE_VMEMMAP
1608 Allows heavy hugetlb users to free up some more
1609 memory (6 * PAGE_SIZE for each 2MB hugetlb page).
1610 Format: { on | off (default) }
1612 on: enable the feature
1613 off: disable the feature
1615 Built with CONFIG_HUGETLB_PAGE_FREE_VMEMMAP_DEFAULT_ON=y,
1618 This is not compatible with memory_hotplug.memmap_on_memory.
1619 If both parameters are enabled, hugetlb_free_vmemmap takes
1620 precedence over memory_hotplug.memmap_on_memory.
1623 [KNL] Should the hung task detector generate panics.
1626 A value of 1 instructs the kernel to panic when a
1627 hung task is detected. The default value is controlled
1628 by the CONFIG_BOOTPARAM_HUNG_TASK_PANIC build-time
1629 option. The value selected by this boot parameter can
1630 be changed later by the kernel.hung_task_panic sysctl.
1632 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC)
1633 terminal devices. Valid values: 0..8
1634 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs.
1635 If specified, z/VM IUCV HVC accepts connections
1636 from listed z/VM user IDs only.
1638 hv_nopvspin [X86,HYPER_V] Disables the paravirt spinlock optimizations
1639 which allow the hypervisor to 'idle' the
1640 guest on lock contention.
1643 Do not unregister boot console at start. This is only
1644 useful for debugging when something happens in the window
1645 between unregistering the boot console and initializing
1648 i2c_bus= [HW] Override the default board specific I2C bus speed
1649 or register an additional I2C bus that is not
1650 registered from board initialization code.
1654 i8042.debug [HW] Toggle i8042 debug mode
1655 i8042.unmask_kbd_data
1656 [HW] Enable printing of interrupt data from the KBD port
1657 (disabled by default, and as a pre-condition
1658 requires that i8042.debug=1 be enabled)
1659 i8042.direct [HW] Put keyboard port into non-translated mode
1660 i8042.dumbkbd [HW] Pretend that controller can only read data from
1661 keyboard and cannot control its state
1662 (Don't attempt to blink the leds)
1663 i8042.noaux [HW] Don't check for auxiliary (== mouse) port
1664 i8042.nokbd [HW] Don't check/create keyboard port
1665 i8042.noloop [HW] Disable the AUX Loopback command while probing
1667 i8042.nomux [HW] Don't check presence of an active multiplexing
1669 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX
1671 i8042.notimeout [HW] Ignore timeout condition signalled by controller
1672 i8042.reset [HW] Reset the controller during init, cleanup and
1673 suspend-to-ram transitions, only during s2r
1674 transitions, or never reset
1675 Format: { 1 | Y | y | 0 | N | n }
1676 1, Y, y: always reset controller
1677 0, N, n: don't ever reset controller
1678 Default: only on s2r transitions on x86; most other
1679 architectures force reset to be always executed
1680 i8042.unlock [HW] Unlock (ignore) the keylock
1681 i8042.kbdreset [HW] Reset device connected to KBD port
1685 i8k.ignore_dmi [HW] Continue probing hardware even if DMI data
1686 indicates that the driver is running on unsupported
1688 i8k.force [HW] Activate i8k driver even if SMM BIOS signature
1689 does not match list of supported models.
1691 [HW] Report power status in /proc/i8k
1692 (disabled by default)
1693 i8k.restricted [HW] Allow controlling fans only if SYS_ADMIN
1696 i915.invert_brightness=
1697 [DRM] Invert the sense of the variable that is used to
1698 set the brightness of the panel backlight. Normally a
1699 brightness value of 0 indicates backlight switched off,
1700 and the maximum of the brightness value sets the backlight
1701 to maximum brightness. If this parameter is set to 0
1702 (default) and the machine requires it, or this parameter
1703 is set to 1, a brightness value of 0 sets the backlight
1704 to maximum brightness, and the maximum of the brightness
1705 value switches the backlight off.
1706 -1 -- never invert brightness
1707 0 -- machine default
1708 1 -- force brightness inversion
1711 Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]]
1713 ide-core.nodma= [HW] (E)IDE subsystem
1714 Format: =0.0 to prevent dma on hda, =0.1 hdb =1.0 hdc
1715 .vlb_clock .pci_clock .noflush .nohpa .noprobe .nowerr
1716 .cdrom .chs .ignore_cable are additional options
1717 See Documentation/ide/ide.rst.
1719 ide-generic.probe-mask= [HW] (E)IDE subsystem
1721 Probe mask for legacy ISA IDE ports. Depending on
1722 platform up to 6 ports are supported, enabled by
1723 setting corresponding bits in the mask to 1. The
1724 default value is 0x0, which has a special meaning.
1725 On systems that have PCI, it triggers scanning the
1726 PCI bus for the first and the second port, which
1727 are then probed. On systems without PCI the value
1728 of 0x0 enables probing the two first ports as if it
1731 ide-pci-generic.all-generic-ide [HW] (E)IDE subsystem
1732 Claim all unknown PCI IDE storage controllers.
1735 Format: idle=poll, idle=halt, idle=nomwait
1736 Poll forces a polling idle loop that can slightly
1737 improve the performance of waking up a idle CPU, but
1738 will use a lot of power and make the system run hot.
1740 idle=halt: Halt is forced to be used for CPU idle.
1741 In such case C2/C3 won't be used again.
1742 idle=nomwait: Disable mwait for CPU C-states
1746 Allow force disabling of Shared Virtual Memory (SVA)
1747 support for the idxd driver. By default it is set to
1750 idxd.tc_override= [HW]
1752 Allow override of default traffic class configuration
1753 for the device. By default it is set to false (0).
1755 ieee754= [MIPS] Select IEEE Std 754 conformance mode
1756 Format: { strict | legacy | 2008 | relaxed }
1759 Choose which programs will be accepted for execution
1760 based on the IEEE 754 NaN encoding(s) supported by
1761 the FPU and the NaN encoding requested with the value
1762 of an ELF file header flag individually set by each
1763 binary. Hardware implementations are permitted to
1764 support either or both of the legacy and the 2008 NaN
1767 Available settings are as follows:
1768 strict accept binaries that request a NaN encoding
1769 supported by the FPU
1770 legacy only accept legacy-NaN binaries, if supported
1772 2008 only accept 2008-NaN binaries, if supported
1774 relaxed accept any binaries regardless of whether
1775 supported by the FPU
1777 The FPU emulator is always able to support both NaN
1778 encodings, so if no FPU hardware is present or it has
1779 been disabled with 'nofpu', then the settings of
1780 'legacy' and '2008' strap the emulator accordingly,
1781 'relaxed' straps the emulator for both legacy-NaN and
1782 2008-NaN, whereas 'strict' enables legacy-NaN only on
1783 legacy processors and both NaN encodings on MIPS32 or
1786 The setting for ABS.fmt/NEG.fmt instruction execution
1787 mode generally follows that for the NaN encoding,
1788 except where unsupported by hardware.
1790 ignore_loglevel [KNL]
1791 Ignore loglevel setting - this will print /all/
1792 kernel messages to the console. Useful for debugging.
1793 We also add it as printk module parameter, so users
1794 could change it dynamically, usually by
1795 /sys/module/printk/parameters/ignore_loglevel.
1798 Ignore RLIMIT_DATA setting for data mappings,
1799 print warning at first misuse. Can be changed via
1800 /sys/module/kernel/parameters/ignore_rlimit_data.
1802 ihash_entries= [KNL]
1803 Set number of hash buckets for inode cache.
1805 ima_appraise= [IMA] appraise integrity measurements
1806 Format: { "off" | "enforce" | "fix" | "log" }
1809 ima_appraise_tcb [IMA] Deprecated. Use ima_policy= instead.
1810 The builtin appraise policy appraises all files
1813 ima_canonical_fmt [IMA]
1814 Use the canonical format for the binary runtime
1815 measurements, instead of host native format.
1818 Format: { md5 | sha1 | rmd160 | sha256 | sha384
1822 The list of supported hash algorithms is defined
1823 in crypto/hash_info.h.
1826 The builtin policies to load during IMA setup.
1827 Format: "tcb | appraise_tcb | secure_boot |
1828 fail_securely | critical_data"
1830 The "tcb" policy measures all programs exec'd, files
1831 mmap'd for exec, and all files opened with the read
1832 mode bit set by either the effective uid (euid=0) or
1835 The "appraise_tcb" policy appraises the integrity of
1836 all files owned by root.
1838 The "secure_boot" policy appraises the integrity
1839 of files (eg. kexec kernel image, kernel modules,
1840 firmware, policy, etc) based on file signatures.
1842 The "fail_securely" policy forces file signature
1843 verification failure also on privileged mounted
1844 filesystems with the SB_I_UNVERIFIABLE_SIGNATURE
1847 The "critical_data" policy measures kernel integrity
1850 ima_tcb [IMA] Deprecated. Use ima_policy= instead.
1851 Load a policy which meets the needs of the Trusted
1852 Computing Base. This means IMA will measure all
1853 programs exec'd, files mmap'd for exec, and all files
1854 opened for read by uid=0.
1857 Select one of defined IMA measurements template formats.
1858 Formats: { "ima" | "ima-ng" | "ima-sig" }
1862 [IMA] Define a custom template format.
1863 Format: { "field1|...|fieldN" }
1865 ima.ahash_minsize= [IMA] Minimum file size for asynchronous hash usage
1866 Format: <min_file_size>
1867 Set the minimal file size for using asynchronous hash.
1868 If left unspecified, ahash usage is disabled.
1870 ahash performance varies for different data sizes on
1871 different crypto accelerators. This option can be used
1872 to achieve the best performance for a particular HW.
1874 ima.ahash_bufsize= [IMA] Asynchronous hash buffer size
1876 Set hashing buffer size. Default: 4k.
1878 ahash performance varies for different chunk sizes on
1879 different crypto accelerators. This option can be used
1880 to achieve best performance for particular HW.
1884 Run specified binary instead of /sbin/init as init
1887 initcall_debug [KNL] Trace initcalls as they are executed. Useful
1888 for working out where the kernel is dying during
1891 initcall_blacklist= [KNL] Do not execute a comma-separated list of
1892 initcall functions. Useful for debugging built-in
1893 modules and initcalls.
1895 initramfs_async= [KNL]
1898 This parameter controls whether the initramfs
1899 image is unpacked asynchronously, concurrently
1900 with devices being probed and
1901 initialized. This should normally just work,
1902 but as a debugging aid, one can get the
1903 historical behaviour of the initramfs
1904 unpacking being completed before device_ and
1907 initrd= [BOOT] Specify the location of the initial ramdisk
1909 initrdmem= [KNL] Specify a physical address and size from which to
1910 load the initrd. If an initrd is compiled in or
1911 specified in the bootparams, it takes priority over this
1913 Format: ss[KMG],nn[KMG]
1916 init_on_alloc= [MM] Fill newly allocated pages and heap objects with
1919 Default set by CONFIG_INIT_ON_ALLOC_DEFAULT_ON.
1921 init_on_free= [MM] Fill freed pages and heap objects with zeroes.
1923 Default set by CONFIG_INIT_ON_FREE_DEFAULT_ON.
1925 init_pkru= [X86] Specify the default memory protection keys rights
1926 register contents for all processes. 0x55555554 by
1927 default (disallow access to all but pkey 0). Can
1928 override in debugfs after boot.
1930 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver
1933 int_pln_enable [X86] Enable power limit notification interrupt
1935 integrity_audit=[IMA]
1936 Format: { "0" | "1" }
1937 0 -- basic integrity auditing messages. (Default)
1938 1 -- additional integrity auditing messages.
1940 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option
1942 Enable intel iommu driver.
1944 Disable intel iommu driver.
1945 igfx_off [Default Off]
1946 By default, gfx is mapped as normal device. If a gfx
1947 device has a dedicated DMAR unit, the DMAR unit is
1948 bypassed by not enabling DMAR with this option. In
1949 this case, gfx device will use physical address for
1951 strict [Default Off]
1952 With this option on every unmap_single operation will
1953 result in a hardware IOTLB flush operation as opposed
1954 to batching them for performance.
1955 sp_off [Default Off]
1956 By default, super page will be supported if Intel IOMMU
1957 has the capability. With this option, super page will
1960 By default, scalable mode will be disabled even if the
1961 hardware advertises that it has support for the scalable
1962 mode translation. With this option set, scalable mode
1963 will be used on hardware which claims to support it.
1964 tboot_noforce [Default Off]
1965 Do not force the Intel IOMMU enabled under tboot.
1966 By default, tboot will force Intel IOMMU on, which
1967 could harm performance of some high-throughput
1968 devices like 40GBit network cards, even if identity
1970 Note that using this option lowers the security
1971 provided by tboot because it makes the system
1972 vulnerable to DMA attacks.
1974 intel_idle.max_cstate= [KNL,HW,ACPI,X86]
1975 0 disables intel_idle and fall back on acpi_idle.
1976 1 to 9 specify maximum depth of C-state.
1980 Do not enable intel_pstate as the default
1981 scaling driver for the supported processors
1983 Use intel_pstate as a scaling driver, but configure it
1984 to work with generic cpufreq governors (instead of
1985 enabling its internal governor). This mode cannot be
1986 used along with the hardware-managed P-states (HWP)
1989 Enable intel_pstate on systems that prohibit it by default
1990 in favor of acpi-cpufreq. Forcing the intel_pstate driver
1991 instead of acpi-cpufreq may disable platform features, such
1992 as thermal controls and power capping, that rely on ACPI
1993 P-States information being indicated to OSPM and therefore
1994 should be used with caution. This option does not work with
1995 processors that aren't supported by the intel_pstate driver
1996 or on platforms that use pcc-cpufreq instead of acpi-cpufreq.
1998 Do not enable hardware P state control (HWP)
2001 Only load intel_pstate on systems which support
2002 hardware P state control (HWP) if available.
2004 Enforce ACPI _PPC performance limits. If the Fixed ACPI
2005 Description Table, specifies preferred power management
2006 profile as "Enterprise Server" or "Performance Server",
2007 then this feature is turned on by default.
2009 Allow per-logical-CPU P-State performance control limits using
2010 cpufreq sysfs interface
2012 intremap= [X86-64, Intel-IOMMU]
2013 on enable Interrupt Remapping (default)
2014 off disable Interrupt Remapping
2015 nosid disable Source ID checking
2017 BIOS x2APIC opt-out request will be ignored
2018 nopost disable Interrupt Posting
2020 iomem= Disable strict checking of access to MMIO memory
2021 strict regions from userspace.
2036 nobypass [PPC/POWERNV]
2037 Disable IOMMU bypass, using IOMMU for PCI devices.
2039 iommu.forcedac= [ARM64, X86] Control IOVA allocation for PCI devices.
2040 Format: { "0" | "1" }
2041 0 - Try to allocate a 32-bit DMA address first, before
2042 falling back to the full range if needed.
2043 1 - Allocate directly from the full usable range,
2044 forcing Dual Address Cycle for PCI cards supporting
2045 greater than 32-bit addressing.
2047 iommu.strict= [ARM64, X86] Configure TLB invalidation behaviour
2048 Format: { "0" | "1" }
2050 Request that DMA unmap operations use deferred
2051 invalidation of hardware TLBs, for increased
2052 throughput at the cost of reduced device isolation.
2053 Will fall back to strict mode if not supported by
2054 the relevant IOMMU driver.
2055 1 - Strict mode (default).
2056 DMA unmap operations invalidate IOMMU hardware TLBs
2058 Note: on x86, the default behaviour depends on the
2059 equivalent driver-specific parameters, but a strict
2060 mode explicitly specified by either method takes
2064 [ARM64, X86] Configure DMA to bypass the IOMMU by default.
2065 Format: { "0" | "1" }
2066 0 - Use IOMMU translation for DMA.
2067 1 - Bypass the IOMMU for DMA.
2068 unset - Use value of CONFIG_IOMMU_DEFAULT_PASSTHROUGH.
2070 io7= [HW] IO7 for Marvel-based Alpha systems
2071 See comment before marvel_specify_io7 in
2072 arch/alpha/kernel/core_marvel.c.
2074 io_delay= [X86] I/O delay method
2076 Standard port 0x80 based delay
2078 Alternate port 0xed based delay (needed on some systems)
2080 Simple two microseconds delay
2085 See Documentation/admin-guide/nfs/nfsroot.rst.
2087 ipcmni_extend [KNL] Extend the maximum number of unique System V
2088 IPC identifiers from 32,768 to 16,777,216.
2090 irqaffinity= [SMP] Set the default irq affinity mask
2091 The argument is a cpu list, as described above.
2093 irqchip.gicv2_force_probe=
2096 Force the kernel to look for the second 4kB page
2097 of a GICv2 controller even if the memory range
2098 exposed by the device tree is too small.
2100 irqchip.gicv3_nolpi=
2102 Force the kernel to ignore the availability of
2103 LPIs (and by consequence ITSs). Intended for system
2104 that use the kernel as a bootloader, and thus want
2105 to let secondary kernels in charge of setting up
2108 irqchip.gicv3_pseudo_nmi= [ARM64]
2109 Enables support for pseudo-NMIs in the kernel. This
2110 requires the kernel to be built with
2111 CONFIG_ARM64_PSEUDO_NMI.
2114 When an interrupt is not handled search all handlers
2115 for it. Intended to get systems with badly broken
2119 When an interrupt is not handled search all handlers
2120 for it. Also check all handlers each timer
2121 interrupt. Intended to get systems with badly broken
2125 Format: <RDP>,<reset>,<pci_scan>,<verbosity>
2127 isolcpus= [KNL,SMP,ISOL] Isolate a given set of CPUs from disturbance.
2128 [Deprecated - use cpusets instead]
2129 Format: [flag-list,]<cpu-list>
2131 Specify one or more CPUs to isolate from disturbances
2132 specified in the flag list (default: domain):
2135 Disable the tick when a single task runs.
2137 A residual 1Hz tick is offloaded to workqueues, which you
2138 need to affine to housekeeping through the global
2139 workqueue's affinity configured via the
2140 /sys/devices/virtual/workqueue/cpumask sysfs file, or
2141 by using the 'domain' flag described below.
2143 NOTE: by default the global workqueue runs on all CPUs,
2144 so to protect individual CPUs the 'cpumask' file has to
2145 be configured manually after bootup.
2148 Isolate from the general SMP balancing and scheduling
2149 algorithms. Note that performing domain isolation this way
2150 is irreversible: it's not possible to bring back a CPU to
2151 the domains once isolated through isolcpus. It's strongly
2152 advised to use cpusets instead to disable scheduler load
2153 balancing through the "cpuset.sched_load_balance" file.
2154 It offers a much more flexible interface where CPUs can
2155 move in and out of an isolated set anytime.
2157 You can move a process onto or off an "isolated" CPU via
2158 the CPU affinity syscalls or cpuset.
2159 <cpu number> begins at 0 and the maximum value is
2160 "number of CPUs in system - 1".
2164 Isolate from being targeted by managed interrupts
2165 which have an interrupt mask containing isolated
2166 CPUs. The affinity of managed interrupts is
2167 handled by the kernel and cannot be changed via
2168 the /proc/irq/* interfaces.
2170 This isolation is best effort and only effective
2171 if the automatically assigned interrupt mask of a
2172 device queue contains isolated and housekeeping
2173 CPUs. If housekeeping CPUs are online then such
2174 interrupts are directed to the housekeeping CPU
2175 so that IO submitted on the housekeeping CPU
2176 cannot disturb the isolated CPU.
2178 If a queue's affinity mask contains only isolated
2179 CPUs then this parameter has no effect on the
2180 interrupt routing decision, though interrupts are
2181 only delivered when tasks running on those
2182 isolated CPUs submit IO. IO submitted on
2183 housekeeping CPUs has no influence on those
2186 The format of <cpu-list> is described above.
2190 ivrs_ioapic [HW,X86-64]
2191 Provide an override to the IOAPIC-ID<->DEVICE-ID
2192 mapping provided in the IVRS ACPI table. For
2193 example, to map IOAPIC-ID decimal 10 to
2194 PCI device 00:14.0 write the parameter as:
2195 ivrs_ioapic[10]=00:14.0
2197 ivrs_hpet [HW,X86-64]
2198 Provide an override to the HPET-ID<->DEVICE-ID
2199 mapping provided in the IVRS ACPI table. For
2200 example, to map HPET-ID decimal 0 to
2201 PCI device 00:14.0 write the parameter as:
2202 ivrs_hpet[0]=00:14.0
2204 ivrs_acpihid [HW,X86-64]
2205 Provide an override to the ACPI-HID:UID<->DEVICE-ID
2206 mapping provided in the IVRS ACPI table. For
2207 example, to map UART-HID:UID AMD0020:0 to
2208 PCI device 00:14.5 write the parameter as:
2209 ivrs_acpihid[00:14.5]=AMD0020:0
2211 js= [HW,JOY] Analog joystick
2212 See Documentation/input/joydev/joystick.rst.
2215 When CONFIG_RANDOMIZE_BASE is set, this disables
2216 kernel and module base offset ASLR (Address Space
2217 Layout Randomization).
2220 [KNL] Enforce KASAN (Kernel Address Sanitizer) to print
2221 report on every invalid memory access. Without this
2222 parameter KASAN will print report only for the first
2227 kernelcore= [KNL,X86,IA-64,PPC]
2228 Format: nn[KMGTPE] | nn% | "mirror"
2229 This parameter specifies the amount of memory usable by
2230 the kernel for non-movable allocations. The requested
2231 amount is spread evenly throughout all nodes in the
2232 system as ZONE_NORMAL. The remaining memory is used for
2233 movable memory in its own zone, ZONE_MOVABLE. In the
2234 event, a node is too small to have both ZONE_NORMAL and
2235 ZONE_MOVABLE, kernelcore memory will take priority and
2236 other nodes will have a larger ZONE_MOVABLE.
2238 ZONE_MOVABLE is used for the allocation of pages that
2239 may be reclaimed or moved by the page migration
2240 subsystem. Note that allocations like PTEs-from-HighMem
2241 still use the HighMem zone if it exists, and the Normal
2242 zone if it does not.
2244 It is possible to specify the exact amount of memory in
2245 the form of "nn[KMGTPE]", a percentage of total system
2246 memory in the form of "nn%", or "mirror". If "mirror"
2247 option is specified, mirrored (reliable) memory is used
2248 for non-movable allocations and remaining memory is used
2249 for Movable pages. "nn[KMGTPE]", "nn%", and "mirror"
2250 are exclusive, so you cannot specify multiple forms.
2252 kgdbdbgp= [KGDB,HW] kgdb over EHCI usb debug port.
2253 Format: <Controller#>[,poll interval]
2254 The controller # is the number of the ehci usb debug
2255 port as it is probed via PCI. The poll interval is
2256 optional and is the number seconds in between
2257 each poll cycle to the debug port in case you need
2258 the functionality for interrupting the kernel with
2259 gdb or control-c on the dbgp connection. When
2260 not using this parameter you use sysrq-g to break into
2261 the kernel debugger.
2263 kgdboc= [KGDB,HW] kgdb over consoles.
2264 Requires a tty driver that supports console polling,
2265 or a supported polling keyboard driver (non-usb).
2266 Serial only format: <serial_device>[,baud]
2267 keyboard only format: kbd
2268 keyboard and serial format: kbd,<serial_device>[,baud]
2269 Optional Kernel mode setting:
2270 kms, kbd format: kms,kbd
2271 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud]
2273 kgdboc_earlycon= [KGDB,HW]
2274 If the boot console provides the ability to read
2275 characters and can work in polling mode, you can use
2276 this parameter to tell kgdb to use it as a backend
2277 until the normal console is registered. Intended to
2278 be used together with the kgdboc parameter which
2279 specifies the normal console to transition to.
2281 The name of the early console should be specified
2282 as the value of this parameter. Note that the name of
2283 the early console might be different than the tty
2284 name passed to kgdboc. It's OK to leave the value
2285 blank and the first boot console that implements
2286 read() will be picked.
2288 kgdbwait [KGDB] Stop kernel execution and enter the
2289 kernel debugger at the earliest opportunity.
2291 kmac= [MIPS] Korina ethernet MAC address.
2292 Configure the RouterBoard 532 series on-chip
2293 Ethernet adapter MAC address.
2295 kmemleak= [KNL] Boot-time kmemleak enable/disable
2296 Valid arguments: on, off
2298 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y,
2301 kprobe_event=[probe-list]
2302 [FTRACE] Add kprobe events and enable at boot time.
2303 The probe-list is a semicolon delimited list of probe
2304 definitions. Each definition is same as kprobe_events
2305 interface, but the parameters are comma delimited.
2306 For example, to add a kprobe event on vfs_read with
2307 arg1 and arg2, add to the command line;
2309 kprobe_event=p,vfs_read,$arg1,$arg2
2311 See also Documentation/trace/kprobetrace.rst "Kernel
2312 Boot Parameter" section.
2314 kpti= [ARM64] Control page table isolation of user
2315 and kernel address spaces.
2316 Default: enabled on cores which need mitigation.
2320 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs.
2321 Default is 0 (don't ignore, but inject #GP)
2323 kvm.enable_vmware_backdoor=[KVM] Support VMware backdoor PV interface.
2324 Default is false (don't support).
2326 kvm.mmu_audit= [KVM] This is a R/W parameter which allows audit
2331 [KVM] Controls the software workaround for the
2332 X86_BUG_ITLB_MULTIHIT bug.
2333 force : Always deploy workaround.
2334 off : Never deploy workaround.
2335 auto : Deploy workaround based on the presence of
2336 X86_BUG_ITLB_MULTIHIT.
2340 If the software workaround is enabled for the host,
2341 guests do need not to enable it for nested guests.
2343 kvm.nx_huge_pages_recovery_ratio=
2344 [KVM] Controls how many 4KiB pages are periodically zapped
2345 back to huge pages. 0 disables the recovery, otherwise if
2346 the value is N KVM will zap 1/Nth of the 4KiB pages every
2347 minute. The default is 60.
2349 kvm-amd.nested= [KVM,AMD] Allow nested virtualization in KVM/SVM.
2350 Default is 1 (enabled)
2352 kvm-amd.npt= [KVM,AMD] Disable nested paging (virtualized MMU)
2354 Default is 1 (enabled) if in 64-bit or 32-bit PAE mode.
2357 [KVM,ARM] Select one of KVM/arm64's modes of operation.
2359 nvhe: Standard nVHE-based mode, without support for
2362 protected: nVHE-based mode with support for guests whose
2363 state is kept private from the host.
2364 Not valid if the kernel is running in EL2.
2366 Defaults to VHE/nVHE based on hardware support.
2368 kvm-arm.vgic_v3_group0_trap=
2369 [KVM,ARM] Trap guest accesses to GICv3 group-0
2372 kvm-arm.vgic_v3_group1_trap=
2373 [KVM,ARM] Trap guest accesses to GICv3 group-1
2376 kvm-arm.vgic_v3_common_trap=
2377 [KVM,ARM] Trap guest accesses to GICv3 common
2380 kvm-arm.vgic_v4_enable=
2381 [KVM,ARM] Allow use of GICv4 for direct injection of
2384 kvm_cma_resv_ratio=n [PPC]
2385 Reserves given percentage from system memory area for
2386 contiguous memory allocation for KVM hash pagetable
2388 By default it reserves 5% of total system memory.
2392 kvm-intel.ept= [KVM,Intel] Disable extended page tables
2393 (virtualized MMU) support on capable Intel chips.
2394 Default is 1 (enabled)
2396 kvm-intel.emulate_invalid_guest_state=
2397 [KVM,Intel] Enable emulation of invalid guest states
2398 Default is 0 (disabled)
2400 kvm-intel.flexpriority=
2401 [KVM,Intel] Disable FlexPriority feature (TPR shadow).
2402 Default is 1 (enabled)
2405 [KVM,Intel] Enable VMX nesting (nVMX).
2406 Default is 0 (disabled)
2408 kvm-intel.unrestricted_guest=
2409 [KVM,Intel] Disable unrestricted guest feature
2410 (virtualized real and unpaged mode) on capable
2411 Intel chips. Default is 1 (enabled)
2413 kvm-intel.vmentry_l1d_flush=[KVM,Intel] Mitigation for L1 Terminal Fault
2416 Valid arguments: never, cond, always
2418 always: L1D cache flush on every VMENTER.
2419 cond: Flush L1D on VMENTER only when the code between
2420 VMEXIT and VMENTER can leak host memory.
2421 never: Disables the mitigation
2423 Default is cond (do L1 cache flush in specific instances)
2425 kvm-intel.vpid= [KVM,Intel] Disable Virtual Processor Identification
2426 feature (tagged TLBs) on capable Intel chips.
2427 Default is 1 (enabled)
2429 l1tf= [X86] Control mitigation of the L1TF vulnerability on
2432 The kernel PTE inversion protection is unconditionally
2433 enabled and cannot be disabled.
2436 Provides all available mitigations for the
2437 L1TF vulnerability. Disables SMT and
2438 enables all mitigations in the
2439 hypervisors, i.e. unconditional L1D flush.
2441 SMT control and L1D flush control via the
2442 sysfs interface is still possible after
2443 boot. Hypervisors will issue a warning
2444 when the first VM is started in a
2445 potentially insecure configuration,
2446 i.e. SMT enabled or L1D flush disabled.
2449 Same as 'full', but disables SMT and L1D
2450 flush runtime control. Implies the
2451 'nosmt=force' command line option.
2452 (i.e. sysfs control of SMT is disabled.)
2455 Leaves SMT enabled and enables the default
2456 hypervisor mitigation, i.e. conditional
2459 SMT control and L1D flush control via the
2460 sysfs interface is still possible after
2461 boot. Hypervisors will issue a warning
2462 when the first VM is started in a
2463 potentially insecure configuration,
2464 i.e. SMT enabled or L1D flush disabled.
2468 Disables SMT and enables the default
2469 hypervisor mitigation.
2471 SMT control and L1D flush control via the
2472 sysfs interface is still possible after
2473 boot. Hypervisors will issue a warning
2474 when the first VM is started in a
2475 potentially insecure configuration,
2476 i.e. SMT enabled or L1D flush disabled.
2479 Same as 'flush', but hypervisors will not
2480 warn when a VM is started in a potentially
2481 insecure configuration.
2484 Disables hypervisor mitigations and doesn't
2486 It also drops the swap size and available
2487 RAM limit restriction on both hypervisor and
2492 For details see: Documentation/admin-guide/hw-vuln/l1tf.rst
2498 lapic [X86-32,APIC] Enable the local APIC even if BIOS
2501 lapic= [X86,APIC] Do not use TSC deadline
2502 value for LAPIC timer one-shot implementation. Default
2503 back to the programmable timer unit in the LAPIC.
2504 Format: notscdeadline
2506 lapic_timer_c2_ok [X86,APIC] trust the local apic timer
2509 libata.dma= [LIBATA] DMA control
2510 libata.dma=0 Disable all PATA and SATA DMA
2511 libata.dma=1 PATA and SATA Disk DMA only
2512 libata.dma=2 ATAPI (CDROM) DMA only
2513 libata.dma=4 Compact Flash DMA only
2514 Combinations also work, so libata.dma=3 enables DMA
2515 for disks and CDROMs, but not CFs.
2517 libata.ignore_hpa= [LIBATA] Ignore HPA limit
2518 libata.ignore_hpa=0 keep BIOS limits (default)
2519 libata.ignore_hpa=1 ignore limits, using full disk
2521 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume
2525 libata.force= [LIBATA] Force configurations. The format is comma-
2526 separated list of "[ID:]VAL" where ID is
2527 PORT[.DEVICE]. PORT and DEVICE are decimal numbers
2528 matching port, link or device. Basically, it matches
2529 the ATA ID string printed on console by libata. If
2530 the whole ID part is omitted, the last PORT and DEVICE
2531 values are used. If ID hasn't been specified yet, the
2532 configuration applies to all ports, links and devices.
2534 If only DEVICE is omitted, the parameter applies to
2535 the port and all links and devices behind it. DEVICE
2536 number of 0 either selects the first device or the
2537 first fan-out link behind PMP device. It does not
2538 select the host link. DEVICE number of 15 selects the
2539 host link and device attached to it.
2541 The VAL specifies the configuration to force. As long
2542 as there's no ambiguity shortcut notation is allowed.
2543 For example, both 1.5 and 1.5G would work for 1.5Gbps.
2544 The following configurations can be forced.
2546 * Cable type: 40c, 80c, short40c, unk, ign or sata.
2547 Any ID with matching PORT is used.
2549 * SATA link speed limit: 1.5Gbps or 3.0Gbps.
2551 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7].
2552 udma[/][16,25,33,44,66,100,133] notation is also
2555 * [no]ncq: Turn on or off NCQ.
2557 * [no]ncqtrim: Turn off queued DSM TRIM.
2559 * nohrst, nosrst, norst: suppress hard, soft
2562 * rstonce: only attempt one reset during
2563 hot-unplug link recovery
2565 * dump_id: dump IDENTIFY data.
2567 * atapi_dmadir: Enable ATAPI DMADIR bridge support
2569 * disable: Disable this device.
2571 If there are multiple matching configurations changing
2572 the same attribute, the last one is used.
2574 memblock=debug [KNL] Enable memblock debug messages.
2576 load_ramdisk= [RAM] [Deprecated]
2578 lockd.nlm_grace_period=P [NFS] Assign grace period.
2581 lockd.nlm_tcpport=N [NFS] Assign TCP port.
2584 lockd.nlm_timeout=T [NFS] Assign timeout value.
2587 lockd.nlm_udpport=M [NFS] Assign UDP port.
2590 lockdown= [SECURITY]
2591 { integrity | confidentiality }
2592 Enable the kernel lockdown feature. If set to
2593 integrity, kernel features that allow userland to
2594 modify the running kernel are disabled. If set to
2595 confidentiality, kernel features that allow userland
2596 to extract confidential information from the kernel
2599 locktorture.nreaders_stress= [KNL]
2600 Set the number of locking read-acquisition kthreads.
2601 Defaults to being automatically set based on the
2602 number of online CPUs.
2604 locktorture.nwriters_stress= [KNL]
2605 Set the number of locking write-acquisition kthreads.
2607 locktorture.onoff_holdoff= [KNL]
2608 Set time (s) after boot for CPU-hotplug testing.
2610 locktorture.onoff_interval= [KNL]
2611 Set time (s) between CPU-hotplug operations, or
2612 zero to disable CPU-hotplug testing.
2614 locktorture.shuffle_interval= [KNL]
2615 Set task-shuffle interval (jiffies). Shuffling
2616 tasks allows some CPUs to go into dyntick-idle
2617 mode during the locktorture test.
2619 locktorture.shutdown_secs= [KNL]
2620 Set time (s) after boot system shutdown. This
2621 is useful for hands-off automated testing.
2623 locktorture.stat_interval= [KNL]
2624 Time (s) between statistics printk()s.
2626 locktorture.stutter= [KNL]
2627 Time (s) to stutter testing, for example,
2628 specifying five seconds causes the test to run for
2629 five seconds, wait for five seconds, and so on.
2630 This tests the locking primitive's ability to
2631 transition abruptly to and from idle.
2633 locktorture.torture_type= [KNL]
2634 Specify the locking implementation to test.
2636 locktorture.verbose= [KNL]
2637 Enable additional printk() statements.
2639 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver
2642 loglevel= All Kernel Messages with a loglevel smaller than the
2643 console loglevel will be printed to the console. It can
2644 also be changed with klogd or other programs. The
2645 loglevels are defined as follows:
2647 0 (KERN_EMERG) system is unusable
2648 1 (KERN_ALERT) action must be taken immediately
2649 2 (KERN_CRIT) critical conditions
2650 3 (KERN_ERR) error conditions
2651 4 (KERN_WARNING) warning conditions
2652 5 (KERN_NOTICE) normal but significant condition
2653 6 (KERN_INFO) informational
2654 7 (KERN_DEBUG) debug-level messages
2656 log_buf_len=n[KMG] Sets the size of the printk ring buffer,
2657 in bytes. n must be a power of two and greater
2658 than the minimal size. The minimal size is defined
2659 by LOG_BUF_SHIFT kernel config parameter. There is
2660 also CONFIG_LOG_CPU_MAX_BUF_SHIFT config parameter
2661 that allows to increase the default size depending on
2662 the number of CPUs. See init/Kconfig for more details.
2664 logo.nologo [FB] Disables display of the built-in Linux logo.
2665 This may be used to provide more screen space for
2666 kernel log messages and is useful when debugging
2667 kernel boot problems.
2669 lp=0 [LP] Specify parallel ports to use, e.g,
2670 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses
2671 lp=reset first parallel port). 'lp=0' disables the
2672 lp=auto printer driver. 'lp=reset' (which can be
2673 specified in addition to the ports) causes
2674 attached printers to be reset. Using
2675 lp=port1,port2,... specifies the parallel ports
2676 to associate lp devices with, starting with
2677 lp0. A port specification may be 'none' to skip
2678 that lp device, or a parport name such as
2679 'parport0'. Specifying 'lp=auto' instead of a
2680 port specification list means that device IDs
2681 from each port should be examined, to see if
2682 an IEEE 1284-compliant printer is attached; if
2683 so, the driver will manage that printer.
2684 See also header of drivers/char/lp.c.
2687 Sets loops_per_jiffy to given constant, thus avoiding
2688 time-consuming boot-time autodetection (up to 250 ms per
2689 CPU). 0 enables autodetection (default). To determine
2690 the correct value for your kernel, boot with normal
2691 autodetection and see what value is printed. Note that
2692 on SMP systems the preset will be applied to all CPUs,
2693 which is likely to cause problems if your CPUs need
2694 significantly divergent settings. An incorrect value
2695 will cause delays in the kernel to be wrong, leading to
2696 unpredictable I/O errors and other breakage. Although
2697 unlikely, in the extreme case this might damage your
2701 Format: <io>,<irq>,<dma>
2703 lsm.debug [SECURITY] Enable LSM initialization debugging output.
2706 [SECURITY] Choose order of LSM initialization. This
2707 overrides CONFIG_LSM, and the "security=" parameter.
2709 machvec= [IA-64] Force the use of a particular machine-vector
2710 (machvec) in a generic kernel.
2711 Example: machvec=hpzx1
2713 machtype= [Loongson] Share the same kernel image file between
2714 different yeeloong laptops.
2715 Example: machtype=lemote-yeeloong-2f-7inch
2717 max_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory greater
2718 than or equal to this physical address is ignored.
2720 maxcpus= [SMP] Maximum number of processors that an SMP kernel
2721 will bring up during bootup. maxcpus=n : n >= 0 limits
2722 the kernel to bring up 'n' processors. Surely after
2723 bootup you can bring up the other plugged cpu by executing
2724 "echo 1 > /sys/devices/system/cpu/cpuX/online". So maxcpus
2725 only takes effect during system bootup.
2726 While n=0 is a special case, it is equivalent to "nosmp",
2727 which also disables the IO APIC.
2729 max_loop= [LOOP] The number of loop block devices that get
2730 (loop.max_loop) unconditionally pre-created at init time. The default
2731 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead
2732 of statically allocating a predefined number, loop
2733 devices can be requested on-demand with the
2734 /dev/loop-control interface.
2736 mce [X86-32] Machine Check Exception
2738 mce=option [X86-64] See Documentation/x86/x86_64/boot-options.rst
2740 md= [HW] RAID subsystems devices and level
2741 See Documentation/admin-guide/md.rst.
2744 Format: <first>,<last>
2745 Specifies range of consoles to be captured by the MDA.
2748 Control mitigation for the Micro-architectural Data
2749 Sampling (MDS) vulnerability.
2751 Certain CPUs are vulnerable to an exploit against CPU
2752 internal buffers which can forward information to a
2753 disclosure gadget under certain conditions.
2755 In vulnerable processors, the speculatively
2756 forwarded data can be used in a cache side channel
2757 attack, to access data to which the attacker does
2758 not have direct access.
2760 This parameter controls the MDS mitigation. The
2763 full - Enable MDS mitigation on vulnerable CPUs
2764 full,nosmt - Enable MDS mitigation and disable
2765 SMT on vulnerable CPUs
2766 off - Unconditionally disable MDS mitigation
2768 On TAA-affected machines, mds=off can be prevented by
2769 an active TAA mitigation as both vulnerabilities are
2770 mitigated with the same mechanism so in order to disable
2771 this mitigation, you need to specify tsx_async_abort=off
2774 Not specifying this option is equivalent to
2777 For details see: Documentation/admin-guide/hw-vuln/mds.rst
2779 mem=nn[KMG] [KNL,BOOT] Force usage of a specific amount of memory
2780 Amount of memory to be used in cases as follows:
2783 2 when the kernel is not able to see the whole system memory;
2784 3 memory that lies after 'mem=' boundary is excluded from
2785 the hypervisor, then assigned to KVM guests.
2787 [X86] Work as limiting max address. Use together
2788 with memmap= to avoid physical address space collisions.
2789 Without memmap= PCI devices could be placed at addresses
2790 belonging to unused RAM.
2792 Note that this only takes effects during boot time since
2793 in above case 3, memory may need be hot added after boot
2794 if system memory of hypervisor is not sufficient.
2796 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel
2800 [KNL,SH] Allow user to override the default size for
2801 per-device physically contiguous DMA buffers.
2803 memhp_default_state=online/offline
2804 [KNL] Set the initial state for the memory hotplug
2805 onlining policy. If not specified, the default value is
2806 set according to the
2807 CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE kernel config
2809 See Documentation/admin-guide/mm/memory-hotplug.rst.
2811 memmap=exactmap [KNL,X86] Enable setting of an exact
2812 E820 memory map, as specified by the user.
2813 Such memmap=exactmap lines can be constructed based on
2814 BIOS output or other requirements. See the memmap=nn@ss
2817 memmap=nn[KMG]@ss[KMG]
2818 [KNL, X86, MIPS, XTENSA] Force usage of a specific region of memory.
2819 Region of memory to be used is from ss to ss+nn.
2820 If @ss[KMG] is omitted, it is equivalent to mem=nn[KMG],
2821 which limits max address to nn[KMG].
2822 Multiple different regions can be specified,
2825 memmap=100M@2G,100M#3G,1G!1024G
2827 memmap=nn[KMG]#ss[KMG]
2828 [KNL,ACPI] Mark specific memory as ACPI data.
2829 Region of memory to be marked is from ss to ss+nn.
2831 memmap=nn[KMG]$ss[KMG]
2832 [KNL,ACPI] Mark specific memory as reserved.
2833 Region of memory to be reserved is from ss to ss+nn.
2834 Example: Exclude memory from 0x18690000-0x1869ffff
2835 memmap=64K$0x18690000
2837 memmap=0x10000$0x18690000
2838 Some bootloaders may need an escape character before '$',
2839 like Grub2, otherwise '$' and the following number
2842 memmap=nn[KMG]!ss[KMG]
2843 [KNL,X86] Mark specific memory as protected.
2844 Region of memory to be used, from ss to ss+nn.
2845 The memory region may be marked as e820 type 12 (0xc)
2846 and is NVDIMM or ADR memory.
2848 memmap=<size>%<offset>-<oldtype>+<newtype>
2849 [KNL,ACPI] Convert memory within the specified region
2850 from <oldtype> to <newtype>. If "-<oldtype>" is left
2851 out, the whole region will be marked as <newtype>,
2852 even if previously unavailable. If "+<newtype>" is left
2853 out, matching memory will be removed. Types are
2854 specified as e820 types, e.g., 1 = RAM, 2 = reserved,
2855 3 = ACPI, 12 = PRAM.
2857 memory_corruption_check=0/1 [X86]
2858 Some BIOSes seem to corrupt the first 64k of
2859 memory when doing things like suspend/resume.
2860 Setting this option will scan the memory
2861 looking for corruption. Enabling this will
2862 both detect corruption and prevent the kernel
2863 from using the memory being corrupted.
2864 However, its intended as a diagnostic tool; if
2865 repeatable BIOS-originated corruption always
2866 affects the same memory, you can use memmap=
2867 to prevent the kernel from using that memory.
2869 memory_corruption_check_size=size [X86]
2870 By default it checks for corruption in the low
2871 64k, making this memory unavailable for normal
2872 use. Use this parameter to scan for
2873 corruption in more or less memory.
2875 memory_corruption_check_period=seconds [X86]
2876 By default it checks for corruption every 60
2877 seconds. Use this parameter to check at some
2878 other rate. 0 disables periodic checking.
2880 memory_hotplug.memmap_on_memory
2881 [KNL,X86,ARM] Boolean flag to enable this feature.
2882 Format: {on | off (default)}
2883 When enabled, runtime hotplugged memory will
2884 allocate its internal metadata (struct pages)
2885 from the hotadded memory which will allow to
2886 hotadd a lot of memory without requiring
2887 additional memory to do so.
2888 This feature is disabled by default because it
2889 has some implication on large (e.g. GB)
2890 allocations in some configurations (e.g. small
2892 The state of the flag can be read in
2893 /sys/module/memory_hotplug/parameters/memmap_on_memory.
2894 Note that even when enabled, there are a few cases where
2895 the feature is not effective.
2897 This is not compatible with hugetlb_free_vmemmap. If
2898 both parameters are enabled, hugetlb_free_vmemmap takes
2899 precedence over memory_hotplug.memmap_on_memory.
2901 memtest= [KNL,X86,ARM,PPC,RISCV] Enable memtest
2903 default : 0 <disable>
2904 Specifies the number of memtest passes to be
2905 performed. Each pass selects another test
2906 pattern from a given set of patterns. Memtest
2907 fills the memory with this pattern, validates
2908 memory contents and reserves bad memory
2909 regions that are detected.
2911 mem_encrypt= [X86-64] AMD Secure Memory Encryption (SME) control
2912 Valid arguments: on, off
2913 Default (depends on kernel configuration option):
2914 on (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=y)
2915 off (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=n)
2916 mem_encrypt=on: Activate SME
2917 mem_encrypt=off: Do not activate SME
2919 Refer to Documentation/virt/kvm/amd-memory-encryption.rst
2920 for details on when memory encryption can be activated.
2922 mem_sleep_default= [SUSPEND] Default system suspend mode:
2923 s2idle - Suspend-To-Idle
2924 shallow - Power-On Suspend or equivalent (if supported)
2925 deep - Suspend-To-RAM or equivalent (if supported)
2926 See Documentation/admin-guide/pm/sleep-states.rst.
2928 meye.*= [HW] Set MotionEye Camera parameters
2929 See Documentation/admin-guide/media/meye.rst.
2931 mfgpt_irq= [IA-32] Specify the IRQ to use for the
2932 Multi-Function General Purpose Timers on AMD Geode
2935 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when
2936 the BIOS has incorrectly applied a workaround. TinyBIOS
2937 version 0.98 is known to be affected, 0.99 fixes the
2938 problem by letting the user disable the workaround.
2942 min_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory below this
2943 physical address is ignored.
2945 mini2440= [ARM,HW,KNL]
2946 Format:[0..2][b][c][t]
2948 MINI2440 configuration specification:
2949 0 - The attached screen is the 3.5" TFT
2950 1 - The attached screen is the 7" TFT
2951 2 - The VGA Shield is attached (1024x768)
2952 Leaving out the screen size parameter will not load
2953 the TFT driver, and the framebuffer will be left
2955 b - Enable backlight. The TFT backlight pin will be
2956 linked to the kernel VESA blanking code and a GPIO
2957 LED. This parameter is not necessary when using the
2959 c - Enable the s3c camera interface.
2960 t - Reserved for enabling touchscreen support. The
2961 touchscreen support is not enabled in the mainstream
2962 kernel as of 2.6.30, a preliminary port can be found
2963 in the "bleeding edge" mini2440 support kernel at
2964 https://repo.or.cz/w/linux-2.6/mini2440.git
2967 [X86,PPC,S390,ARM64] Control optional mitigations for
2968 CPU vulnerabilities. This is a set of curated,
2969 arch-independent options, each of which is an
2970 aggregation of existing arch-specific options.
2973 Disable all optional CPU mitigations. This
2974 improves system performance, but it may also
2975 expose users to several CPU vulnerabilities.
2976 Equivalent to: nopti [X86,PPC]
2978 nospectre_v1 [X86,PPC]
2980 nospectre_v2 [X86,PPC,S390,ARM64]
2981 spectre_v2_user=off [X86]
2982 spec_store_bypass_disable=off [X86,PPC]
2983 ssbd=force-off [ARM64]
2986 tsx_async_abort=off [X86]
2987 kvm.nx_huge_pages=off [X86]
2988 no_entry_flush [PPC]
2989 no_uaccess_flush [PPC]
2992 This does not have any effect on
2993 kvm.nx_huge_pages when
2994 kvm.nx_huge_pages=force.
2997 Mitigate all CPU vulnerabilities, but leave SMT
2998 enabled, even if it's vulnerable. This is for
2999 users who don't want to be surprised by SMT
3000 getting disabled across kernel upgrades, or who
3001 have other ways of avoiding SMT-based attacks.
3002 Equivalent to: (default behavior)
3005 Mitigate all CPU vulnerabilities, disabling SMT
3006 if needed. This is for users who always want to
3007 be fully mitigated, even if it means losing SMT.
3008 Equivalent to: l1tf=flush,nosmt [X86]
3009 mds=full,nosmt [X86]
3010 tsx_async_abort=full,nosmt [X86]
3013 [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this
3014 parameter allows control of the logging verbosity for
3015 the additional memory initialisation checks. A value
3016 of 0 disables mminit logging and a level of 4 will
3017 log everything. Information is printed at KERN_DEBUG
3018 so loglevel=8 may also need to be specified.
3021 [KNL] When CONFIG_MODULE_SIG is set, this means that
3022 modules without (valid) signatures will fail to load.
3023 Note that if CONFIG_MODULE_SIG_FORCE is set, that
3024 is always true, so this option does nothing.
3026 module_blacklist= [KNL] Do not load a comma-separated list of
3027 modules. Useful for debugging problem modules.
3030 [MOUSE] Maximum time between finger touching and
3031 leaving touchpad surface for touch to be considered
3032 a tap and be reported as a left button click (for
3033 touchpads working in absolute mode only).
3035 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices
3036 reporting absolute coordinates, such as tablets
3037 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices
3038 reporting absolute coordinates, such as tablets
3040 movablecore= [KNL,X86,IA-64,PPC]
3041 Format: nn[KMGTPE] | nn%
3042 This parameter is the complement to kernelcore=, it
3043 specifies the amount of memory used for migratable
3044 allocations. If both kernelcore and movablecore is
3045 specified, then kernelcore will be at *least* the
3046 specified value but may be more. If movablecore on its
3047 own is specified, the administrator must be careful
3048 that the amount of memory usable for all allocations
3051 movable_node [KNL] Boot-time switch to make hotplugable memory
3052 NUMA nodes to be movable. This means that the memory
3053 of such nodes will be usable only for movable
3054 allocations which rules out almost all kernel
3055 allocations. Use with caution!
3057 MTD_Partition= [MTD]
3058 Format: <name>,<region-number>,<size>,<offset>
3060 MTD_Region= [MTD] Format:
3061 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>]
3064 See drivers/mtd/parsers/cmdlinepart.c
3066 multitce=off [PPC] This parameter disables the use of the pSeries
3067 firmware feature for updating multiple TCE entries
3070 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration
3072 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock]
3074 boundary - index of last SLC block on Flex-OneNAND.
3075 The remaining blocks are configured as MLC blocks.
3076 lock - Configure if Flex-OneNAND boundary should be locked.
3077 Once locked, the boundary cannot be changed.
3078 1 indicates lock status, 0 indicates unlock status.
3081 ARM/S3C2412 JIVE boot control
3083 See arch/arm/mach-s3c/mach-jive.c
3085 mtouchusb.raw_coordinates=
3086 [HW] Make the MicroTouch USB driver use raw coordinates
3087 ('y', default) or cooked coordinates ('n')
3089 mtrr_chunk_size=nn[KMG] [X86]
3090 used for mtrr cleanup. It is largest continuous chunk
3091 that could hold holes aka. UC entries.
3093 mtrr_gran_size=nn[KMG] [X86]
3094 Used for mtrr cleanup. It is granularity of mtrr block.
3096 Large value could prevent small alignment from
3099 mtrr_spare_reg_nr=n [X86]
3101 Range: 0,7 : spare reg number
3103 Used for mtrr cleanup. It is spare mtrr entries number.
3104 Set to 2 or more if your graphical card needs more.
3106 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card
3108 netdev= [NET] Network devices parameters
3109 Format: <irq>,<io>,<mem_start>,<mem_end>,<name>
3110 Note that mem_start is often overloaded to mean
3111 something different and driver-specific.
3112 This usage is only documented in each driver source
3116 [NETFILTER] Enable connection tracking flow accounting
3117 0 to disable accounting
3118 1 to enable accounting
3121 nfsaddrs= [NFS] Deprecated. Use ip= instead.
3122 See Documentation/admin-guide/nfs/nfsroot.rst.
3124 nfsroot= [NFS] nfs root filesystem for disk-less boxes.
3125 See Documentation/admin-guide/nfs/nfsroot.rst.
3127 nfsrootdebug [NFS] enable nfsroot debugging messages.
3128 See Documentation/admin-guide/nfs/nfsroot.rst.
3130 nfs.callback_nr_threads=
3131 [NFSv4] set the total number of threads that the
3132 NFS client will assign to service NFSv4 callback
3135 nfs.callback_tcpport=
3136 [NFS] set the TCP port on which the NFSv4 callback
3137 channel should listen.
3140 [NFS] sets the pathname to the program which is used
3141 to update the NFS client cache entries.
3143 nfs.cache_getent_timeout=
3144 [NFS] sets the timeout after which an attempt to
3145 update a cache entry is deemed to have failed.
3147 nfs.idmap_cache_timeout=
3148 [NFS] set the maximum lifetime for idmapper cache
3152 [NFS] enable 64-bit inode numbers.
3153 If zero, the NFS client will fake up a 32-bit inode
3154 number for the readdir() and stat() syscalls instead
3155 of returning the full 64-bit number.
3156 The default is to return 64-bit inode numbers.
3158 nfs.max_session_cb_slots=
3159 [NFSv4.1] Sets the maximum number of session
3160 slots the client will assign to the callback
3161 channel. This determines the maximum number of
3162 callbacks the client will process in parallel for
3163 a particular server.
3165 nfs.max_session_slots=
3166 [NFSv4.1] Sets the maximum number of session slots
3167 the client will attempt to negotiate with the server.
3168 This limits the number of simultaneous RPC requests
3169 that the client can send to the NFSv4.1 server.
3170 Note that there is little point in setting this
3171 value higher than the max_tcp_slot_table_limit.
3173 nfs.nfs4_disable_idmapping=
3174 [NFSv4] When set to the default of '1', this option
3175 ensures that both the RPC level authentication
3176 scheme and the NFS level operations agree to use
3177 numeric uids/gids if the mount is using the
3178 'sec=sys' security flavour. In effect it is
3179 disabling idmapping, which can make migration from
3180 legacy NFSv2/v3 systems to NFSv4 easier.
3181 Servers that do not support this mode of operation
3182 will be autodetected by the client, and it will fall
3183 back to using the idmapper.
3184 To turn off this behaviour, set the value to '0'.
3186 [NFS4] Specify an additional fixed unique ident-
3187 ification string that NFSv4 clients can insert into
3188 their nfs_client_id4 string. This is typically a
3189 UUID that is generated at system install time.
3191 nfs.send_implementation_id =
3192 [NFSv4.1] Send client implementation identification
3193 information in exchange_id requests.
3194 If zero, no implementation identification information
3196 The default is to send the implementation identification
3199 nfs.recover_lost_locks =
3200 [NFSv4] Attempt to recover locks that were lost due
3201 to a lease timeout on the server. Please note that
3202 doing this risks data corruption, since there are
3203 no guarantees that the file will remain unchanged
3204 after the locks are lost.
3205 If you want to enable the kernel legacy behaviour of
3206 attempting to recover these locks, then set this
3208 The default parameter value of '0' causes the kernel
3209 not to attempt recovery of lost locks.
3211 nfs4.layoutstats_timer =
3212 [NFSv4.2] Change the rate at which the kernel sends
3213 layoutstats to the pNFS metadata server.
3215 Setting this to value to 0 causes the kernel to use
3216 whatever value is the default set by the layout
3217 driver. A non-zero value sets the minimum interval
3218 in seconds between layoutstats transmissions.
3220 nfsd.nfs4_disable_idmapping=
3221 [NFSv4] When set to the default of '1', the NFSv4
3222 server will return only numeric uids and gids to
3223 clients using auth_sys, and will accept numeric uids
3224 and gids from such clients. This is intended to ease
3225 migration from NFSv2/v3.
3227 nmi_backtrace.backtrace_idle [KNL]
3228 Dump stacks even of idle CPUs in response to an
3229 NMI stack-backtrace request.
3231 nmi_debug= [KNL,SH] Specify one or more actions to take
3232 when a NMI is triggered.
3233 Format: [state][,regs][,debounce][,die]
3235 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels
3236 Format: [panic,][nopanic,][num]
3238 0 - turn hardlockup detector in nmi_watchdog off
3239 1 - turn hardlockup detector in nmi_watchdog on
3240 When panic is specified, panic when an NMI watchdog
3241 timeout occurs (or 'nopanic' to not panic on an NMI
3242 watchdog, if CONFIG_BOOTPARAM_HARDLOCKUP_PANIC is set)
3243 To disable both hard and soft lockup detectors,
3244 please see 'nowatchdog'.
3245 This is useful when you use a panic=... timeout and
3246 need the box quickly up again.
3248 These settings can be accessed at runtime via
3249 the nmi_watchdog and hardlockup_panic sysctls.
3251 netpoll.carrier_timeout=
3252 [NET] Specifies amount of time (in seconds) that
3253 netpoll should wait for a carrier. By default netpoll
3256 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths
3257 emulation library even if a 387 maths coprocessor
3260 no5lvl [X86-64] Disable 5-level paging mode. Forces
3261 kernel to use 4-level paging instead.
3263 nofsgsbase [X86] Disables FSGSBASE instructions.
3266 [HW] Never suspend the console
3267 Disable suspending of consoles during suspend and
3268 hibernate operations. Once disabled, debugging
3269 messages can reach various consoles while the rest
3270 of the system is being put to sleep (ie, while
3271 debugging driver suspend/resume hooks). This may
3272 not work reliably with all consoles, but is known
3273 to work with serial and VGA consoles.
3274 To facilitate more flexible debugging, we also add
3275 console_suspend, a printk module parameter to control
3276 it. Users could use console_suspend (usually
3277 /sys/module/printk/parameters/console_suspend) to
3278 turn on/off it dynamically.
3280 novmcoredd [KNL,KDUMP]
3281 Disable device dump. Device dump allows drivers to
3282 append dump data to vmcore so you can collect driver
3283 specified debug info. Drivers can append the data
3284 without any limit and this data is stored in memory,
3285 so this may cause significant memory stress. Disabling
3286 device dump can help save memory but the driver debug
3287 data will be no longer available. This parameter
3288 is only available when CONFIG_PROC_VMCORE_DEVICE_DUMP
3291 noaliencache [MM, NUMA, SLAB] Disables the allocation of alien
3292 caches in the slab allocator. Saves per-node memory,
3293 but will impact performance.
3297 noaltinstr [S390] Disables alternative instructions patching
3298 (CPU alternatives feature).
3300 noapic [SMP,APIC] Tells the kernel to not make use of any
3301 IOAPICs that may be present in the system.
3303 noautogroup Disable scheduler automatic task group creation.
3305 nobats [PPC] Do not use BATs for mapping kernel lowmem
3306 on "Classic" PPC cores.
3310 noclflush [BUGS=X86] Don't use the CLFLUSH instruction
3312 delayacct [KNL] Enable per-task delay accounting
3314 nodsp [SH] Disable hardware DSP at boot time.
3316 noefi Disable EFI runtime services support.
3318 no_entry_flush [PPC] Don't flush the L1-D cache when entering the kernel.
3323 On X86-32 available only on PAE configured kernels.
3324 noexec=on: enable non-executable mappings (default)
3325 noexec=off: disable non-executable mappings
3328 Disable SMAP (Supervisor Mode Access Prevention)
3329 even if it is supported by processor.
3332 Disable SMEP (Supervisor Mode Execution Prevention)
3333 even if it is supported by processor.
3336 This affects only 32-bit executables.
3337 noexec32=on: enable non-executable mappings (default)
3338 read doesn't imply executable mappings
3339 noexec32=off: disable non-executable mappings
3340 read implies executable mappings
3342 nofpu [MIPS,SH] Disable hardware FPU at boot time.
3344 nofxsr [BUGS=X86-32] Disables x86 floating point extended
3345 register save and restore. The kernel will only save
3346 legacy floating-point registers on task switch.
3348 nohugeiomap [KNL,X86,PPC,ARM64] Disable kernel huge I/O mappings.
3350 nohugevmalloc [PPC] Disable kernel huge vmalloc mappings.
3352 nosmt [KNL,S390] Disable symmetric multithreading (SMT).
3353 Equivalent to smt=1.
3355 [KNL,X86] Disable symmetric multithreading (SMT).
3356 nosmt=force: Force disable SMT, cannot be undone
3357 via the sysfs control file.
3359 nospectre_v1 [X86,PPC] Disable mitigations for Spectre Variant 1
3360 (bounds check bypass). With this option data leaks are
3361 possible in the system.
3363 nospectre_v2 [X86,PPC_FSL_BOOK3E,ARM64] Disable all mitigations for
3364 the Spectre variant 2 (indirect branch prediction)
3365 vulnerability. System may allow data leaks with this
3368 nospec_store_bypass_disable
3369 [HW] Disable all mitigations for the Speculative Store Bypass vulnerability
3372 [PPC] Don't flush the L1-D cache after accessing user data.
3374 noxsave [BUGS=X86] Disables x86 extended register state save
3375 and restore using xsave. The kernel will fallback to
3376 enabling legacy floating-point and sse state.
3378 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended
3379 register states. The kernel will fall back to use
3380 xsave to save the states. By using this parameter,
3381 performance of saving the states is degraded because
3382 xsave doesn't support modified optimization while
3383 xsaveopt supports it on xsaveopt enabled systems.
3385 noxsaves [X86] Disables xsaves and xrstors used in saving and
3386 restoring x86 extended register state in compacted
3387 form of xsave area. The kernel will fall back to use
3388 xsaveopt and xrstor to save and restore the states
3389 in standard form of xsave area. By using this
3390 parameter, xsave area per process might occupy more
3391 memory on xsaves enabled systems.
3393 nohlt [ARM,ARM64,MICROBLAZE,SH] Forces the kernel to busy wait
3394 in do_idle() and not use the arch_cpu_idle()
3395 implementation; requires CONFIG_GENERIC_IDLE_POLL_SETUP
3396 to be effective. This is useful on platforms where the
3397 sleep(SH) or wfi(ARM,ARM64) instructions do not work
3398 correctly or when doing power measurements to evalute
3399 the impact of the sleep instructions. This is also
3400 useful when using JTAG debugger.
3402 no_file_caps Tells the kernel not to honor file capabilities. The
3403 only way then for a file to be executed with privilege
3404 is to be setuid root or executed by root.
3406 nohalt [IA-64] Tells the kernel not to use the power saving
3407 function PAL_HALT_LIGHT when idle. This increases
3408 power-consumption. On the positive side, it reduces
3409 interrupt wake-up latency, which may improve performance
3410 in certain environments such as networked servers or
3414 Force pointers printed to the console or buffers to be
3415 unhashed. By default, when a pointer is printed via %p
3416 format string, that pointer is "hashed", i.e. obscured
3417 by hashing the pointer value. This is a security feature
3418 that hides actual kernel addresses from unprivileged
3419 users, but it also makes debugging the kernel more
3420 difficult since unequal pointers can no longer be
3421 compared. However, if this command-line option is
3422 specified, then all normal pointers will have their true
3423 value printed. Pointers printed via %pK may still be
3424 hashed. This option should only be specified when
3425 debugging the kernel. Please do not use on production
3428 nohibernate [HIBERNATION] Disable hibernation and resume.
3430 nohz= [KNL] Boottime enable/disable dynamic ticks
3431 Valid arguments: on, off
3434 nohz_full= [KNL,BOOT,SMP,ISOL]
3435 The argument is a cpu list, as described above.
3436 In kernels built with CONFIG_NO_HZ_FULL=y, set
3437 the specified list of CPUs whose tick will be stopped
3438 whenever possible. The boot CPU will be forced outside
3439 the range to maintain the timekeeping. Any CPUs
3440 in this list will have their RCU callbacks offloaded,
3441 just as if they had also been called out in the
3442 rcu_nocbs= boot parameter.
3444 noiotrap [SH] Disables trapped I/O port accesses.
3446 noirqdebug [X86-32] Disables the code which attempts to detect and
3447 disable unhandled interrupt sources.
3449 no_timer_check [X86,APIC] Disables the code which tests for
3450 broken timer IRQ sources.
3452 noisapnp [ISAPNP] Disables ISA PnP code.
3454 noinitrd [RAM] Tells the kernel not to load any configured
3457 nointremap [X86-64, Intel-IOMMU] Do not enable interrupt
3459 [Deprecated - use intremap=off]
3463 noinvpcid [X86] Disable the INVPCID cpu feature.
3465 nojitter [IA-64] Disables jitter checking for ITC timers.
3467 no-kvmclock [X86,KVM] Disable paravirtualized KVM clock driver
3469 no-kvmapf [X86,KVM] Disable paravirtualized asynchronous page
3473 [X86,PV_OPS] Disable paravirtualized VMware scheduler
3474 clock and use the default one.
3476 no-steal-acc [X86,PV_OPS,ARM64] Disable paravirtualized steal time
3477 accounting. steal time is computed, but won't
3478 influence scheduler behaviour
3480 nolapic [X86-32,APIC] Do not enable or use the local APIC.
3482 nolapic_timer [X86-32,APIC] Do not use the local APIC timer.
3484 noltlbs [PPC] Do not use large page/tlb entries for kernel
3485 lowmem mapping on PPC40x and PPC8xx
3487 nomca [IA-64] Disable machine check abort handling
3489 nomce [X86-32] Disable Machine Check Exception
3491 nomfgpt [X86-32] Disable Multi-Function General Purpose
3492 Timer usage (for AMD Geode machines).
3494 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to
3495 shutdown the other cpus. Instead use the REBOOT_VECTOR
3498 nomodule Disable module load
3500 nopat [X86] Disable PAT (page attribute table extension of
3501 pagetables) support.
3503 nopcid [X86-64] Disable the PCID cpu feature.
3505 norandmaps Don't use address space randomization. Equivalent to
3506 echo 0 > /proc/sys/kernel/randomize_va_space
3508 noreplace-smp [X86-32,SMP] Don't replace SMP instructions
3509 with UP alternatives
3511 nordrand [X86] Disable kernel use of the RDRAND and
3512 RDSEED instructions even if they are supported
3513 by the processor. RDRAND and RDSEED are still
3514 available to user space applications.
3516 noresume [SWSUSP] Disables resume and restores original swap
3519 no-scroll [VGA] Disables scrollback.
3520 This is required for the Braillex ib80-piezo Braille
3521 reader made by F.H. Papenmeier (Germany).
3525 nosep [BUGS=X86-32] Disables x86 SYSENTER/SYSEXIT support.
3527 nosgx [X86-64,SGX] Disables Intel SGX kernel support.
3529 nosmp [SMP] Tells an SMP kernel to act as a UP kernel,
3530 and disable the IO APIC. legacy for "maxcpus=0".
3532 nosoftlockup [KNL] Disable the soft-lockup detector.
3534 nosync [HW,M68K] Disables sync negotiation for all devices.
3536 nowatchdog [KNL] Disable both lockup detectors, i.e.
3537 soft-lockup and NMI watchdog (hard-lockup).
3541 nox2apic [X86-64,APIC] Do not enable x2APIC mode.
3543 cpu0_hotplug [X86] Turn on CPU0 hotplug feature when
3544 CONFIG_BOOTPARAM_HOTPLUG_CPU0 is off.
3545 Some features depend on CPU0. Known dependencies are:
3546 1. Resume from suspend/hibernate depends on CPU0.
3547 Suspend/hibernate will fail if CPU0 is offline and you
3548 need to online CPU0 before suspend/hibernate.
3549 2. PIC interrupts also depend on CPU0. CPU0 can't be
3550 removed if a PIC interrupt is detected.
3551 It's said poweroff/reboot may depend on CPU0 on some
3552 machines although I haven't seen such issues so far
3553 after CPU0 is offline on a few tested machines.
3554 If the dependencies are under your control, you can
3555 turn on cpu0_hotplug.
3557 nps_mtm_hs_ctr= [KNL,ARC]
3558 This parameter sets the maximum duration, in
3559 cycles, each HW thread of the CTOP can run
3560 without interruptions, before HW switches it.
3561 The actual maximum duration is 16 times this
3563 Format: integer between 1 and 255
3566 nptcg= [IA-64] Override max number of concurrent global TLB
3567 purges which is reported from either PAL_VM_SUMMARY or
3570 nr_cpus= [SMP] Maximum number of processors that an SMP kernel
3571 could support. nr_cpus=n : n >= 1 limits the kernel to
3572 support 'n' processors. It could be larger than the
3573 number of already plugged CPU during bootup, later in
3574 runtime you can physically add extra cpu until it reaches
3575 n. So during boot up some boot time memory for per-cpu
3576 variables need be pre-allocated for later physical cpu
3579 nr_uarts= [SERIAL] maximum number of UARTs to be registered.
3581 numa=off [KNL, ARM64, PPC, RISCV, SPARC, X86] Disable NUMA, Only
3582 set up a single NUMA node spanning all memory.
3584 numa_balancing= [KNL,ARM64,PPC,RISCV,S390,X86] Enable or disable automatic
3586 Allowed values are enable and disable
3588 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA.
3589 'node', 'default' can be specified
3590 This can be set from sysctl after boot.
3591 See Documentation/admin-guide/sysctl/vm.rst for details.
3593 ohci1394_dma=early [HW] enable debugging via the ohci1394 driver.
3594 See Documentation/core-api/debugging-via-ohci1394.rst for more
3597 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands
3598 Rather than timing out after 20 ms if an EC
3599 command is not properly ACKed, override the length
3600 of the timeout. We have interrupts disabled while
3601 waiting for the ACK, so if this is set too high
3602 interrupts *may* be lost!
3604 omap_mux= [OMAP] Override bootloader pin multiplexing.
3605 Format: <mux_mode0.mode_name=value>...
3606 For example, to override I2C bus2:
3607 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100
3609 oops=panic Always panic on oopses. Default is to just kill the
3610 process, but there is a small probability of
3611 deadlocking the machine.
3612 This will also cause panics on machine check exceptions.
3613 Useful together with panic=30 to trigger a reboot.
3616 [KNL] Boolean flag to control whether the page allocator
3617 should randomize its free lists. The randomization may
3618 be automatically enabled if the kernel detects it is
3619 running on a platform with a direct-mapped memory-side
3620 cache, and this parameter can be used to
3621 override/disable that behavior. The state of the flag
3622 can be read from sysfs at:
3623 /sys/module/page_alloc/parameters/shuffle.
3625 page_owner= [KNL] Boot-time page_owner enabling option.
3626 Storage of the information about who allocated
3627 each page is disabled in default. With this switch,
3629 on: enable the feature
3631 page_poison= [KNL] Boot-time parameter changing the state of
3632 poisoning on the buddy allocator, available with
3633 CONFIG_PAGE_POISONING=y.
3634 off: turn off poisoning (default)
3635 on: turn on poisoning
3637 page_reporting.page_reporting_order=
3638 [KNL] Minimal page reporting order
3640 Adjust the minimal page reporting order. The page
3641 reporting is disabled when it exceeds (MAX_ORDER-1).
3643 panic= [KNL] Kernel behaviour on panic: delay <timeout>
3644 timeout > 0: seconds before rebooting
3645 timeout = 0: wait forever
3646 timeout < 0: reboot immediately
3649 panic_print= Bitmask for printing system info when panic happens.
3650 User can chose combination of the following bits:
3651 bit 0: print all tasks info
3652 bit 1: print system memory info
3653 bit 2: print timer info
3654 bit 3: print locks info if CONFIG_LOCKDEP is on
3655 bit 4: print ftrace buffer
3656 bit 5: print all printk messages in buffer
3658 panic_on_taint= Bitmask for conditionally calling panic() in add_taint()
3659 Format: <hex>[,nousertaint]
3660 Hexadecimal bitmask representing the set of TAINT flags
3661 that will cause the kernel to panic when add_taint() is
3662 called with any of the flags in this set.
3663 The optional switch "nousertaint" can be utilized to
3664 prevent userspace forced crashes by writing to sysctl
3665 /proc/sys/kernel/tainted any flagset matching with the
3666 bitmask set on panic_on_taint.
3667 See Documentation/admin-guide/tainted-kernels.rst for
3668 extra details on the taint flags that users can pick
3669 to compose the bitmask to assign to panic_on_taint.
3671 panic_on_warn panic() instead of WARN(). Useful to cause kdump
3674 crash_kexec_post_notifiers
3675 Run kdump after running panic-notifiers and dumping
3676 kmsg. This only for the users who doubt kdump always
3677 succeeds in any situation.
3678 Note that this also increases risks of kdump failure,
3679 because some panic notifiers can make the crashed
3680 kernel more unstable.
3682 parkbd.port= [HW] Parallel port number the keyboard adapter is
3683 connected to, default is 0.
3685 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation,
3686 0 for XT, 1 for AT (default is AT).
3689 parport= [HW,PPT] Specify parallel ports. 0 disables.
3690 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] }
3691 Use 'auto' to force the driver to use any
3692 IRQ/DMA settings detected (the default is to
3693 ignore detected IRQ/DMA settings because of
3694 possible conflicts). You can specify the base
3695 address, IRQ, and DMA settings; IRQ and DMA
3696 should be numbers, or 'auto' (for using detected
3697 settings on that particular port), or 'nofifo'
3698 (to avoid using a FIFO even if it is detected).
3699 Parallel ports are assigned in the order they
3700 are specified on the command line, starting
3703 parport_init_mode= [HW,PPT]
3704 Configure VIA parallel port to operate in
3705 a specific mode. This is necessary on Pegasos
3706 computer where firmware has no options for setting
3707 up parallel port mode and sets it to spp.
3708 Currently this function knows 686a and 8231 chips.
3709 Format: [spp|ps2|epp|ecp|ecpepp]
3711 pata_legacy.all= [HW,LIBATA]
3713 Set to non-zero to probe primary and secondary ISA
3714 port ranges on PCI systems where no PCI PATA device
3715 has been found at either range. Disabled by default.
3717 pata_legacy.autospeed= [HW,LIBATA]
3719 Set to non-zero if a chip is present that snoops speed
3720 changes. Disabled by default.
3722 pata_legacy.ht6560a= [HW,LIBATA]
3724 Set to 1, 2, or 3 for HT 6560A on the primary channel,
3725 the secondary channel, or both channels respectively.
3726 Disabled by default.
3728 pata_legacy.ht6560b= [HW,LIBATA]
3730 Set to 1, 2, or 3 for HT 6560B on the primary channel,
3731 the secondary channel, or both channels respectively.
3732 Disabled by default.
3734 pata_legacy.iordy_mask= [HW,LIBATA]
3736 IORDY enable mask. Set individual bits to allow IORDY
3737 for the respective channel. Bit 0 is for the first
3738 legacy channel handled by this driver, bit 1 is for
3739 the second channel, and so on. The sequence will often
3740 correspond to the primary legacy channel, the secondary
3741 legacy channel, and so on, but the handling of a PCI
3742 bus and the use of other driver options may interfere
3743 with the sequence. By default IORDY is allowed across
3746 pata_legacy.opti82c46x= [HW,LIBATA]
3748 Set to 1, 2, or 3 for Opti 82c611A on the primary
3749 channel, the secondary channel, or both channels
3750 respectively. Disabled by default.
3752 pata_legacy.opti82c611a= [HW,LIBATA]
3754 Set to 1, 2, or 3 for Opti 82c465MV on the primary
3755 channel, the secondary channel, or both channels
3756 respectively. Disabled by default.
3758 pata_legacy.pio_mask= [HW,LIBATA]
3760 PIO mode mask for autospeed devices. Set individual
3761 bits to allow the use of the respective PIO modes.
3762 Bit 0 is for mode 0, bit 1 is for mode 1, and so on.
3763 All modes allowed by default.
3765 pata_legacy.probe_all= [HW,LIBATA]
3767 Set to non-zero to probe tertiary and further ISA
3768 port ranges on PCI systems. Disabled by default.
3770 pata_legacy.probe_mask= [HW,LIBATA]
3772 Probe mask for legacy ISA PATA ports. Depending on
3773 platform configuration and the use of other driver
3774 options up to 6 legacy ports are supported: 0x1f0,
3775 0x170, 0x1e8, 0x168, 0x1e0, 0x160, however probing
3776 of individual ports can be disabled by setting the
3777 corresponding bits in the mask to 1. Bit 0 is for
3778 the first port in the list above (0x1f0), and so on.
3779 By default all supported ports are probed.
3781 pata_legacy.qdi= [HW,LIBATA]
3783 Set to non-zero to probe QDI controllers. By default
3784 set to 1 if CONFIG_PATA_QDI_MODULE, 0 otherwise.
3786 pata_legacy.winbond= [HW,LIBATA]
3788 Set to non-zero to probe Winbond controllers. Use
3789 the standard I/O port (0x130) if 1, otherwise the
3790 value given is the I/O port to use (typically 0x1b0).
3791 By default set to 1 if CONFIG_PATA_WINBOND_VLB_MODULE,
3794 pata_platform.pio_mask= [HW,LIBATA]
3796 Supported PIO mode mask. Set individual bits to allow
3797 the use of the respective PIO modes. Bit 0 is for
3798 mode 0, bit 1 is for mode 1, and so on. Mode 0 only
3802 Halt all CPUs after the first oops has been printed for
3803 the specified number of seconds. This is to be used if
3804 your oopses keep scrolling off the screen.
3809 See header of drivers/block/paride/pcd.c.
3810 See also Documentation/admin-guide/blockdev/paride.rst.
3812 pci=option[,option...] [PCI] various PCI subsystem options.
3814 Some options herein operate on a specific device
3815 or a set of devices (<pci_dev>). These are
3816 specified in one of the following formats:
3818 [<domain>:]<bus>:<dev>.<func>[/<dev>.<func>]*
3819 pci:<vendor>:<device>[:<subvendor>:<subdevice>]
3821 Note: the first format specifies a PCI
3822 bus/device/function address which may change
3823 if new hardware is inserted, if motherboard
3824 firmware changes, or due to changes caused
3825 by other kernel parameters. If the
3826 domain is left unspecified, it is
3827 taken to be zero. Optionally, a path
3828 to a device through multiple device/function
3829 addresses can be specified after the base
3830 address (this is more robust against
3831 renumbering issues). The second format
3832 selects devices using IDs from the
3833 configuration space which may match multiple
3834 devices in the system.
3836 earlydump dump PCI config space before the kernel
3838 off [X86] don't probe for the PCI bus
3839 bios [X86-32] force use of PCI BIOS, don't access
3840 the hardware directly. Use this if your machine
3841 has a non-standard PCI host bridge.
3842 nobios [X86-32] disallow use of PCI BIOS, only direct
3843 hardware access methods are allowed. Use this
3844 if you experience crashes upon bootup and you
3845 suspect they are caused by the BIOS.
3846 conf1 [X86] Force use of PCI Configuration Access
3847 Mechanism 1 (config address in IO port 0xCF8,
3848 data in IO port 0xCFC, both 32-bit).
3849 conf2 [X86] Force use of PCI Configuration Access
3850 Mechanism 2 (IO port 0xCF8 is an 8-bit port for
3851 the function, IO port 0xCFA, also 8-bit, sets
3852 bus number. The config space is then accessed
3853 through ports 0xC000-0xCFFF).
3854 See http://wiki.osdev.org/PCI for more info
3855 on the configuration access mechanisms.
3856 noaer [PCIE] If the PCIEAER kernel config parameter is
3857 enabled, this kernel boot option can be used to
3858 disable the use of PCIE advanced error reporting.
3859 nodomains [PCI] Disable support for multiple PCI
3860 root domains (aka PCI segments, in ACPI-speak).
3861 nommconf [X86] Disable use of MMCONFIG for PCI
3863 check_enable_amd_mmconf [X86] check for and enable
3864 properly configured MMIO access to PCI
3865 config space on AMD family 10h CPU
3866 nomsi [MSI] If the PCI_MSI kernel config parameter is
3867 enabled, this kernel boot option can be used to
3868 disable the use of MSI interrupts system-wide.
3869 noioapicquirk [APIC] Disable all boot interrupt quirks.
3870 Safety option to keep boot IRQs enabled. This
3871 should never be necessary.
3872 ioapicreroute [APIC] Enable rerouting of boot IRQs to the
3873 primary IO-APIC for bridges that cannot disable
3874 boot IRQs. This fixes a source of spurious IRQs
3875 when the system masks IRQs.
3876 noioapicreroute [APIC] Disable workaround that uses the
3877 boot IRQ equivalent of an IRQ that connects to
3878 a chipset where boot IRQs cannot be disabled.
3879 The opposite of ioapicreroute.
3880 biosirq [X86-32] Use PCI BIOS calls to get the interrupt
3881 routing table. These calls are known to be buggy
3882 on several machines and they hang the machine
3883 when used, but on other computers it's the only
3884 way to get the interrupt routing table. Try
3885 this option if the kernel is unable to allocate
3886 IRQs or discover secondary PCI buses on your
3888 rom [X86] Assign address space to expansion ROMs.
3889 Use with caution as certain devices share
3890 address decoders between ROMs and other
3892 norom [X86] Do not assign address space to
3893 expansion ROMs that do not already have
3894 BIOS assigned address ranges.
3895 nobar [X86] Do not assign address space to the
3896 BARs that weren't assigned by the BIOS.
3897 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be
3898 assigned automatically to PCI devices. You can
3899 make the kernel exclude IRQs of your ISA cards
3901 pirqaddr=0xAAAAA [X86] Specify the physical address
3902 of the PIRQ table (normally generated
3903 by the BIOS) if it is outside the
3904 F0000h-100000h range.
3905 lastbus=N [X86] Scan all buses thru bus #N. Can be
3906 useful if the kernel is unable to find your
3907 secondary buses and you want to tell it
3908 explicitly which ones they are.
3909 assign-busses [X86] Always assign all PCI bus
3910 numbers ourselves, overriding
3911 whatever the firmware may have done.
3912 usepirqmask [X86] Honor the possible IRQ mask stored
3913 in the BIOS $PIR table. This is needed on
3914 some systems with broken BIOSes, notably
3915 some HP Pavilion N5400 and Omnibook XE3
3916 notebooks. This will have no effect if ACPI
3917 IRQ routing is enabled.
3918 noacpi [X86] Do not use ACPI for IRQ routing
3919 or for PCI scanning.
3920 use_crs [X86] Use PCI host bridge window information
3921 from ACPI. On BIOSes from 2008 or later, this
3922 is enabled by default. If you need to use this,
3923 please report a bug.
3924 nocrs [X86] Ignore PCI host bridge windows from ACPI.
3925 If you need to use this, please report a bug.
3926 routeirq Do IRQ routing for all PCI devices.
3927 This is normally done in pci_enable_device(),
3928 so this option is a temporary workaround
3929 for broken drivers that don't call it.
3930 skip_isa_align [X86] do not align io start addr, so can
3931 handle more pci cards
3932 noearly [X86] Don't do any early type 1 scanning.
3933 This might help on some broken boards which
3934 machine check when some devices' config space
3935 is read. But various workarounds are disabled
3936 and some IOMMU drivers will not work.
3937 bfsort Sort PCI devices into breadth-first order.
3938 This sorting is done to get a device
3939 order compatible with older (<= 2.4) kernels.
3940 nobfsort Don't sort PCI devices into breadth-first order.
3941 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size)
3942 tuning and use the BIOS-configured MPS defaults.
3943 pcie_bus_safe Set every device's MPS to the largest value
3944 supported by all devices below the root complex.
3945 pcie_bus_perf Set device MPS to the largest allowable MPS
3946 based on its parent bus. Also set MRRS (Max
3947 Read Request Size) to the largest supported
3948 value (no larger than the MPS that the device
3949 or bus can support) for best performance.
3950 pcie_bus_peer2peer Set every device's MPS to 128B, which
3951 every device is guaranteed to support. This
3952 configuration allows peer-to-peer DMA between
3953 any pair of devices, possibly at the cost of
3954 reduced performance. This also guarantees
3955 that hot-added devices will work.
3956 cbiosize=nn[KMG] The fixed amount of bus space which is
3957 reserved for the CardBus bridge's IO window.
3958 The default value is 256 bytes.
3959 cbmemsize=nn[KMG] The fixed amount of bus space which is
3960 reserved for the CardBus bridge's memory
3961 window. The default value is 64 megabytes.
3964 [<order of align>@]<pci_dev>[; ...]
3965 Specifies alignment and device to reassign
3966 aligned memory resources. How to
3967 specify the device is described above.
3968 If <order of align> is not specified,
3969 PAGE_SIZE is used as alignment.
3970 A PCI-PCI bridge can be specified if resource
3971 windows need to be expanded.
3972 To specify the alignment for several
3973 instances of a device, the PCI vendor,
3974 device, subvendor, and subdevice may be
3975 specified, e.g., 12@pci:8086:9c22:103c:198f
3976 for 4096-byte alignment.
3977 ecrc= Enable/disable PCIe ECRC (transaction layer
3978 end-to-end CRC checking).
3979 bios: Use BIOS/firmware settings. This is the
3983 hpiosize=nn[KMG] The fixed amount of bus space which is
3984 reserved for hotplug bridge's IO window.
3985 Default size is 256 bytes.
3986 hpmmiosize=nn[KMG] The fixed amount of bus space which is
3987 reserved for hotplug bridge's MMIO window.
3988 Default size is 2 megabytes.
3989 hpmmioprefsize=nn[KMG] The fixed amount of bus space which is
3990 reserved for hotplug bridge's MMIO_PREF window.
3991 Default size is 2 megabytes.
3992 hpmemsize=nn[KMG] The fixed amount of bus space which is
3993 reserved for hotplug bridge's MMIO and
3995 Default size is 2 megabytes.
3996 hpbussize=nn The minimum amount of additional bus numbers
3997 reserved for buses below a hotplug bridge.
3999 realloc= Enable/disable reallocating PCI bridge resources
4000 if allocations done by BIOS are too small to
4001 accommodate resources required by all child
4003 off: Turn realloc off
4005 realloc same as realloc=on
4006 noari do not use PCIe ARI.
4007 noats [PCIE, Intel-IOMMU, AMD-IOMMU]
4008 do not use PCIe ATS (and IOMMU device IOTLB).
4009 pcie_scan_all Scan all possible PCIe devices. Otherwise we
4010 only look for one device below a PCIe downstream
4012 big_root_window Try to add a big 64bit memory window to the PCIe
4013 root complex on AMD CPUs. Some GFX hardware
4014 can resize a BAR to allow access to all VRAM.
4015 Adding the window is slightly risky (it may
4016 conflict with unreported devices), so this
4018 disable_acs_redir=<pci_dev>[; ...]
4019 Specify one or more PCI devices (in the format
4020 specified above) separated by semicolons.
4021 Each device specified will have the PCI ACS
4022 redirect capabilities forced off which will
4023 allow P2P traffic between devices through
4024 bridges without forcing it upstream. Note:
4025 this removes isolation between devices and
4026 may put more devices in an IOMMU group.
4027 force_floating [S390] Force usage of floating interrupts.
4028 nomio [S390] Do not use MIO instructions.
4029 norid [S390] ignore the RID field and force use of
4030 one PCI domain per PCI function
4032 pcie_aspm= [PCIE] Forcibly enable or disable PCIe Active State Power
4035 force Enable ASPM even on devices that claim not to support it.
4036 WARNING: Forcing ASPM on may cause system lockups.
4038 pcie_ports= [PCIE] PCIe port services handling:
4039 native Use native PCIe services (PME, AER, DPC, PCIe hotplug)
4040 even if the platform doesn't give the OS permission to
4041 use them. This may cause conflicts if the platform
4042 also tries to use these services.
4043 dpc-native Use native PCIe service for DPC only. May
4044 cause conflicts if firmware uses AER or DPC.
4045 compat Disable native PCIe services (PME, AER, DPC, PCIe
4048 pcie_port_pm= [PCIE] PCIe port power management handling:
4049 off Disable power management of all PCIe ports
4050 force Forcibly enable power management of all PCIe ports
4052 pcie_pme= [PCIE,PM] Native PCIe PME signaling options:
4053 nomsi Do not use MSI for native PCIe PME signaling (this makes
4054 all PCIe root ports use INTx for all services).
4056 pcmv= [HW,PCMCIA] BadgePAD 4
4060 Keep all power-domains already enabled by bootloader on,
4061 even if no driver has claimed them. This is useful
4062 for debug and development, but should not be
4063 needed on a platform with proper driver support.
4066 See Documentation/admin-guide/blockdev/paride.rst.
4068 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at
4071 See arch/parisc/kernel/pdc_chassis.c
4073 percpu_alloc= Select which percpu first chunk allocator to use.
4074 Currently supported values are "embed" and "page".
4075 Archs may support subset or none of the selections.
4076 See comments in mm/percpu.c for details on each
4077 allocator. This parameter is primarily for debugging
4078 and performance comparison.
4081 See Documentation/admin-guide/blockdev/paride.rst.
4084 See Documentation/admin-guide/blockdev/paride.rst.
4086 pirq= [SMP,APIC] Manual mp-table setup
4087 See Documentation/x86/i386/IO-APIC.rst.
4089 plip= [PPT,NET] Parallel port network link
4090 Format: { parport<nr> | timid | 0 }
4091 See also Documentation/admin-guide/parport.rst.
4093 pmtmr= [X86] Manual setup of pmtmr I/O Port.
4094 Override pmtimer IOPort with a hex value.
4097 pm_debug_messages [SUSPEND,KNL]
4098 Enable suspend/resume debug messages during boot up.
4101 Enable PNP debug messages (depends on the
4102 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time
4103 via /sys/module/pnp/parameters/debug. We always show
4104 current resource usage; turning this on also shows
4105 possible settings and some assignment information.
4111 { on | off | curr | res | no-curr | no-res }
4114 [ISAPNP] Exclude IRQs for the autoconfiguration
4117 [ISAPNP] Exclude DMAs for the autoconfiguration
4119 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration
4120 Ranges are in pairs (I/O port base and size).
4123 [ISAPNP] Exclude memory regions for the
4125 Ranges are in pairs (memory base and size).
4127 ports= [IP_VS_FTP] IPVS ftp helper module
4129 Up to 8 (IP_VS_APP_MAX_PORTS) ports
4131 Format: <port>,<port>....
4133 powersave=off [PPC] This option disables power saving features.
4134 It specifically disables cpuidle and sets the
4135 platform machine description specific power_save
4136 function to NULL. On Idle the CPU just reduces
4139 ppc_strict_facility_enable
4140 [PPC] This option catches any kernel floating point,
4141 Altivec, VSX and SPE outside of regions specifically
4142 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()).
4143 There is some performance impact when enabling this.
4147 Disable Hardware Transactional Memory
4150 Select preemption mode if you have CONFIG_PREEMPT_DYNAMIC
4151 none - Limited to cond_resched() calls
4152 voluntary - Limited to cond_resched() and might_sleep() calls
4153 full - Any section that isn't explicitly preempt disabled
4154 can be preempted anytime.
4156 print-fatal-signals=
4157 [KNL] debug: print fatal signals
4159 If enabled, warn about various signal handling
4160 related application anomalies: too many signals,
4161 too many POSIX.1 timers, fatal signals causing a
4164 If you hit the warning due to signal overflow,
4165 you might want to try "ulimit -i unlimited".
4169 printk.always_kmsg_dump=
4170 Trigger kmsg_dump for cases other than kernel oops or
4172 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
4175 printk.devkmsg={on,off,ratelimit}
4176 Control writing to /dev/kmsg.
4177 on - unlimited logging to /dev/kmsg from userspace
4178 off - logging to /dev/kmsg disabled
4179 ratelimit - ratelimit the logging
4182 printk.time= Show timing data prefixed to each printk message line
4183 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
4185 processor.max_cstate= [HW,ACPI]
4186 Limit processor to maximum C-state
4187 max_cstate=9 overrides any DMI blacklist limit.
4189 processor.nocst [HW,ACPI]
4190 Ignore the _CST method to determine C-states,
4191 instead using the legacy FADT method
4193 profile= [KNL] Enable kernel profiling via /proc/profile
4194 Format: [<profiletype>,]<number>
4195 Param: <profiletype>: "schedule", "sleep", or "kvm"
4196 [defaults to kernel profiling]
4197 Param: "schedule" - profile schedule points.
4198 Param: "sleep" - profile D-state sleeping (millisecs).
4199 Requires CONFIG_SCHEDSTATS
4200 Param: "kvm" - profile VM exits.
4201 Param: <number> - step/bucket size as a power of 2 for
4202 statistical time based profiling.
4204 prompt_ramdisk= [RAM] [Deprecated]
4206 prot_virt= [S390] enable hosting protected virtual machines
4207 isolated from the hypervisor (if hardware supports
4211 psi= [KNL] Enable or disable pressure stall information
4215 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to
4216 probe for; one of (bare|imps|exps|lifebook|any).
4217 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports
4219 psmouse.resetafter= [HW,MOUSE]
4220 Try to reset the device after so many bad packets
4223 [HW,MOUSE] Set desired mouse resolution, in dpi.
4224 psmouse.smartscroll=
4225 [HW,MOUSE] Controls Logitech smartscroll autorepeat.
4226 0 = disabled, 1 = enabled (default).
4228 pstore.backend= Specify the name of the pstore backend to use
4231 See Documentation/admin-guide/blockdev/paride.rst.
4233 pti= [X86-64] Control Page Table Isolation of user and
4234 kernel address spaces. Disabling this feature
4235 removes hardening, but improves performance of
4236 system calls and interrupts.
4238 on - unconditionally enable
4239 off - unconditionally disable
4240 auto - kernel detects whether your CPU model is
4241 vulnerable to issues that PTI mitigates
4243 Not specifying this option is equivalent to pti=auto.
4246 Equivalent to pti=off
4249 [KNL] Number of legacy pty's. Overwrites compiled-in
4252 quiet [KNL] Disable most log messages
4257 See Documentation/admin-guide/md.rst.
4259 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes
4260 See Documentation/admin-guide/blockdev/ramdisk.rst.
4262 ramdisk_start= [RAM] RAM disk image start address
4264 random.trust_cpu={on,off}
4265 [KNL] Enable or disable trusting the use of the
4266 CPU's random number generator (if available) to
4267 fully seed the kernel's CRNG. Default is controlled
4268 by CONFIG_RANDOM_TRUST_CPU.
4270 randomize_kstack_offset=
4271 [KNL] Enable or disable kernel stack offset
4272 randomization, which provides roughly 5 bits of
4273 entropy, frustrating memory corruption attacks
4274 that depend on stack address determinism or
4275 cross-syscall address exposures. This is only
4276 available on architectures that have defined
4277 CONFIG_HAVE_ARCH_RANDOMIZE_KSTACK_OFFSET.
4278 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
4279 Default is CONFIG_RANDOMIZE_KSTACK_OFFSET_DEFAULT.
4281 ras=option[,option,...] [KNL] RAS-specific options
4284 Disable the Correctable Errors Collector,
4285 see CONFIG_RAS_CEC help text.
4288 The argument is a cpu list, as described above.
4290 In kernels built with CONFIG_RCU_NOCB_CPU=y, set
4291 the specified list of CPUs to be no-callback CPUs.
4292 Invocation of these CPUs' RCU callbacks will be
4293 offloaded to "rcuox/N" kthreads created for that
4294 purpose, where "x" is "p" for RCU-preempt, and
4295 "s" for RCU-sched, and "N" is the CPU number.
4296 This reduces OS jitter on the offloaded CPUs,
4297 which can be useful for HPC and real-time
4298 workloads. It can also improve energy efficiency
4299 for asymmetric multiprocessors.
4302 Rather than requiring that offloaded CPUs
4303 (specified by rcu_nocbs= above) explicitly
4304 awaken the corresponding "rcuoN" kthreads,
4305 make these kthreads poll for callbacks.
4306 This improves the real-time response for the
4307 offloaded CPUs by relieving them of the need to
4308 wake up the corresponding kthread, but degrades
4309 energy efficiency by requiring that the kthreads
4310 periodically wake up to do the polling.
4312 rcutree.blimit= [KNL]
4313 Set maximum number of finished RCU callbacks to
4314 process in one batch.
4316 rcutree.dump_tree= [KNL]
4317 Dump the structure of the rcu_node combining tree
4318 out at early boot. This is used for diagnostic
4319 purposes, to verify correct tree setup.
4321 rcutree.gp_cleanup_delay= [KNL]
4322 Set the number of jiffies to delay each step of
4323 RCU grace-period cleanup.
4325 rcutree.gp_init_delay= [KNL]
4326 Set the number of jiffies to delay each step of
4327 RCU grace-period initialization.
4329 rcutree.gp_preinit_delay= [KNL]
4330 Set the number of jiffies to delay each step of
4331 RCU grace-period pre-initialization, that is,
4332 the propagation of recent CPU-hotplug changes up
4333 the rcu_node combining tree.
4335 rcutree.use_softirq= [KNL]
4336 If set to zero, move all RCU_SOFTIRQ processing to
4337 per-CPU rcuc kthreads. Defaults to a non-zero
4338 value, meaning that RCU_SOFTIRQ is used by default.
4339 Specify rcutree.use_softirq=0 to use rcuc kthreads.
4341 But note that CONFIG_PREEMPT_RT=y kernels disable
4342 this kernel boot parameter, forcibly setting it
4345 rcutree.rcu_fanout_exact= [KNL]
4346 Disable autobalancing of the rcu_node combining
4347 tree. This is used by rcutorture, and might
4348 possibly be useful for architectures having high
4349 cache-to-cache transfer latencies.
4351 rcutree.rcu_fanout_leaf= [KNL]
4352 Change the number of CPUs assigned to each
4353 leaf rcu_node structure. Useful for very
4354 large systems, which will choose the value 64,
4355 and for NUMA systems with large remote-access
4356 latencies, which will choose a value aligned
4357 with the appropriate hardware boundaries.
4359 rcutree.rcu_min_cached_objs= [KNL]
4360 Minimum number of objects which are cached and
4361 maintained per one CPU. Object size is equal
4362 to PAGE_SIZE. The cache allows to reduce the
4363 pressure to page allocator, also it makes the
4364 whole algorithm to behave better in low memory
4367 rcutree.rcu_delay_page_cache_fill_msec= [KNL]
4368 Set the page-cache refill delay (in milliseconds)
4369 in response to low-memory conditions. The range
4370 of permitted values is in the range 0:100000.
4372 rcutree.jiffies_till_first_fqs= [KNL]
4373 Set delay from grace-period initialization to
4374 first attempt to force quiescent states.
4375 Units are jiffies, minimum value is zero,
4376 and maximum value is HZ.
4378 rcutree.jiffies_till_next_fqs= [KNL]
4379 Set delay between subsequent attempts to force
4380 quiescent states. Units are jiffies, minimum
4381 value is one, and maximum value is HZ.
4383 rcutree.jiffies_till_sched_qs= [KNL]
4384 Set required age in jiffies for a
4385 given grace period before RCU starts
4386 soliciting quiescent-state help from
4387 rcu_note_context_switch() and cond_resched().
4388 If not specified, the kernel will calculate
4389 a value based on the most recent settings
4390 of rcutree.jiffies_till_first_fqs
4391 and rcutree.jiffies_till_next_fqs.
4392 This calculated value may be viewed in
4393 rcutree.jiffies_to_sched_qs. Any attempt to set
4394 rcutree.jiffies_to_sched_qs will be cheerfully
4397 rcutree.kthread_prio= [KNL,BOOT]
4398 Set the SCHED_FIFO priority of the RCU per-CPU
4399 kthreads (rcuc/N). This value is also used for
4400 the priority of the RCU boost threads (rcub/N)
4401 and for the RCU grace-period kthreads (rcu_bh,
4402 rcu_preempt, and rcu_sched). If RCU_BOOST is
4403 set, valid values are 1-99 and the default is 1
4404 (the least-favored priority). Otherwise, when
4405 RCU_BOOST is not set, valid values are 0-99 and
4406 the default is zero (non-realtime operation).
4408 rcutree.rcu_nocb_gp_stride= [KNL]
4409 Set the number of NOCB callback kthreads in
4410 each group, which defaults to the square root
4411 of the number of CPUs. Larger numbers reduce
4412 the wakeup overhead on the global grace-period
4413 kthread, but increases that same overhead on
4414 each group's NOCB grace-period kthread.
4416 rcutree.qhimark= [KNL]
4417 Set threshold of queued RCU callbacks beyond which
4418 batch limiting is disabled.
4420 rcutree.qlowmark= [KNL]
4421 Set threshold of queued RCU callbacks below which
4422 batch limiting is re-enabled.
4424 rcutree.qovld= [KNL]
4425 Set threshold of queued RCU callbacks beyond which
4426 RCU's force-quiescent-state scan will aggressively
4427 enlist help from cond_resched() and sched IPIs to
4428 help CPUs more quickly reach quiescent states.
4429 Set to less than zero to make this be set based
4430 on rcutree.qhimark at boot time and to zero to
4431 disable more aggressive help enlistment.
4433 rcutree.rcu_idle_gp_delay= [KNL]
4434 Set wakeup interval for idle CPUs that have
4435 RCU callbacks (RCU_FAST_NO_HZ=y).
4437 rcutree.rcu_kick_kthreads= [KNL]
4438 Cause the grace-period kthread to get an extra
4439 wake_up() if it sleeps three times longer than
4440 it should at force-quiescent-state time.
4441 This wake_up() will be accompanied by a
4442 WARN_ONCE() splat and an ftrace_dump().
4444 rcutree.rcu_unlock_delay= [KNL]
4445 In CONFIG_RCU_STRICT_GRACE_PERIOD=y kernels,
4446 this specifies an rcu_read_unlock()-time delay
4447 in microseconds. This defaults to zero.
4448 Larger delays increase the probability of
4449 catching RCU pointer leaks, that is, buggy use
4450 of RCU-protected pointers after the relevant
4451 rcu_read_unlock() has completed.
4453 rcutree.sysrq_rcu= [KNL]
4454 Commandeer a sysrq key to dump out Tree RCU's
4455 rcu_node tree with an eye towards determining
4456 why a new grace period has not yet started.
4458 rcuscale.gp_async= [KNL]
4459 Measure performance of asynchronous
4460 grace-period primitives such as call_rcu().
4462 rcuscale.gp_async_max= [KNL]
4463 Specify the maximum number of outstanding
4464 callbacks per writer thread. When a writer
4465 thread exceeds this limit, it invokes the
4466 corresponding flavor of rcu_barrier() to allow
4467 previously posted callbacks to drain.
4469 rcuscale.gp_exp= [KNL]
4470 Measure performance of expedited synchronous
4471 grace-period primitives.
4473 rcuscale.holdoff= [KNL]
4474 Set test-start holdoff period. The purpose of
4475 this parameter is to delay the start of the
4476 test until boot completes in order to avoid
4479 rcuscale.kfree_rcu_test= [KNL]
4480 Set to measure performance of kfree_rcu() flooding.
4482 rcuscale.kfree_rcu_test_double= [KNL]
4483 Test the double-argument variant of kfree_rcu().
4484 If this parameter has the same value as
4485 rcuscale.kfree_rcu_test_single, both the single-
4486 and double-argument variants are tested.
4488 rcuscale.kfree_rcu_test_single= [KNL]
4489 Test the single-argument variant of kfree_rcu().
4490 If this parameter has the same value as
4491 rcuscale.kfree_rcu_test_double, both the single-
4492 and double-argument variants are tested.
4494 rcuscale.kfree_nthreads= [KNL]
4495 The number of threads running loops of kfree_rcu().
4497 rcuscale.kfree_alloc_num= [KNL]
4498 Number of allocations and frees done in an iteration.
4500 rcuscale.kfree_loops= [KNL]
4501 Number of loops doing rcuscale.kfree_alloc_num number
4502 of allocations and frees.
4504 rcuscale.nreaders= [KNL]
4505 Set number of RCU readers. The value -1 selects
4506 N, where N is the number of CPUs. A value
4507 "n" less than -1 selects N-n+1, where N is again
4508 the number of CPUs. For example, -2 selects N
4509 (the number of CPUs), -3 selects N+1, and so on.
4510 A value of "n" less than or equal to -N selects
4513 rcuscale.nwriters= [KNL]
4514 Set number of RCU writers. The values operate
4515 the same as for rcuscale.nreaders.
4516 N, where N is the number of CPUs
4518 rcuscale.perf_type= [KNL]
4519 Specify the RCU implementation to test.
4521 rcuscale.shutdown= [KNL]
4522 Shut the system down after performance tests
4523 complete. This is useful for hands-off automated
4526 rcuscale.verbose= [KNL]
4527 Enable additional printk() statements.
4529 rcuscale.writer_holdoff= [KNL]
4530 Write-side holdoff between grace periods,
4531 in microseconds. The default of zero says
4534 rcutorture.fqs_duration= [KNL]
4535 Set duration of force_quiescent_state bursts
4538 rcutorture.fqs_holdoff= [KNL]
4539 Set holdoff time within force_quiescent_state bursts
4542 rcutorture.fqs_stutter= [KNL]
4543 Set wait time between force_quiescent_state bursts
4546 rcutorture.fwd_progress= [KNL]
4547 Enable RCU grace-period forward-progress testing
4548 for the types of RCU supporting this notion.
4550 rcutorture.fwd_progress_div= [KNL]
4551 Specify the fraction of a CPU-stall-warning
4552 period to do tight-loop forward-progress testing.
4554 rcutorture.fwd_progress_holdoff= [KNL]
4555 Number of seconds to wait between successive
4556 forward-progress tests.
4558 rcutorture.fwd_progress_need_resched= [KNL]
4559 Enclose cond_resched() calls within checks for
4560 need_resched() during tight-loop forward-progress
4563 rcutorture.gp_cond= [KNL]
4564 Use conditional/asynchronous update-side
4565 primitives, if available.
4567 rcutorture.gp_exp= [KNL]
4568 Use expedited update-side primitives, if available.
4570 rcutorture.gp_normal= [KNL]
4571 Use normal (non-expedited) asynchronous
4572 update-side primitives, if available.
4574 rcutorture.gp_sync= [KNL]
4575 Use normal (non-expedited) synchronous
4576 update-side primitives, if available. If all
4577 of rcutorture.gp_cond=, rcutorture.gp_exp=,
4578 rcutorture.gp_normal=, and rcutorture.gp_sync=
4579 are zero, rcutorture acts as if is interpreted
4580 they are all non-zero.
4582 rcutorture.irqreader= [KNL]
4583 Run RCU readers from irq handlers, or, more
4584 accurately, from a timer handler. Not all RCU
4585 flavors take kindly to this sort of thing.
4587 rcutorture.leakpointer= [KNL]
4588 Leak an RCU-protected pointer out of the reader.
4589 This can of course result in splats, and is
4590 intended to test the ability of things like
4591 CONFIG_RCU_STRICT_GRACE_PERIOD=y to detect
4594 rcutorture.n_barrier_cbs= [KNL]
4595 Set callbacks/threads for rcu_barrier() testing.
4597 rcutorture.nfakewriters= [KNL]
4598 Set number of concurrent RCU writers. These just
4599 stress RCU, they don't participate in the actual
4600 test, hence the "fake".
4602 rcutorture.nocbs_nthreads= [KNL]
4603 Set number of RCU callback-offload togglers.
4604 Zero (the default) disables toggling.
4606 rcutorture.nocbs_toggle= [KNL]
4607 Set the delay in milliseconds between successive
4608 callback-offload toggling attempts.
4610 rcutorture.nreaders= [KNL]
4611 Set number of RCU readers. The value -1 selects
4612 N-1, where N is the number of CPUs. A value
4613 "n" less than -1 selects N-n-2, where N is again
4614 the number of CPUs. For example, -2 selects N
4615 (the number of CPUs), -3 selects N+1, and so on.
4617 rcutorture.object_debug= [KNL]
4618 Enable debug-object double-call_rcu() testing.
4620 rcutorture.onoff_holdoff= [KNL]
4621 Set time (s) after boot for CPU-hotplug testing.
4623 rcutorture.onoff_interval= [KNL]
4624 Set time (jiffies) between CPU-hotplug operations,
4625 or zero to disable CPU-hotplug testing.
4627 rcutorture.read_exit= [KNL]
4628 Set the number of read-then-exit kthreads used
4629 to test the interaction of RCU updaters and
4630 task-exit processing.
4632 rcutorture.read_exit_burst= [KNL]
4633 The number of times in a given read-then-exit
4634 episode that a set of read-then-exit kthreads
4637 rcutorture.read_exit_delay= [KNL]
4638 The delay, in seconds, between successive
4639 read-then-exit testing episodes.
4641 rcutorture.shuffle_interval= [KNL]
4642 Set task-shuffle interval (s). Shuffling tasks
4643 allows some CPUs to go into dyntick-idle mode
4644 during the rcutorture test.
4646 rcutorture.shutdown_secs= [KNL]
4647 Set time (s) after boot system shutdown. This
4648 is useful for hands-off automated testing.
4650 rcutorture.stall_cpu= [KNL]
4651 Duration of CPU stall (s) to test RCU CPU stall
4652 warnings, zero to disable.
4654 rcutorture.stall_cpu_block= [KNL]
4655 Sleep while stalling if set. This will result
4656 in warnings from preemptible RCU in addition
4657 to any other stall-related activity.
4659 rcutorture.stall_cpu_holdoff= [KNL]
4660 Time to wait (s) after boot before inducing stall.
4662 rcutorture.stall_cpu_irqsoff= [KNL]
4663 Disable interrupts while stalling if set.
4665 rcutorture.stall_gp_kthread= [KNL]
4666 Duration (s) of forced sleep within RCU
4667 grace-period kthread to test RCU CPU stall
4668 warnings, zero to disable. If both stall_cpu
4669 and stall_gp_kthread are specified, the
4670 kthread is starved first, then the CPU.
4672 rcutorture.stat_interval= [KNL]
4673 Time (s) between statistics printk()s.
4675 rcutorture.stutter= [KNL]
4676 Time (s) to stutter testing, for example, specifying
4677 five seconds causes the test to run for five seconds,
4678 wait for five seconds, and so on. This tests RCU's
4679 ability to transition abruptly to and from idle.
4681 rcutorture.test_boost= [KNL]
4682 Test RCU priority boosting? 0=no, 1=maybe, 2=yes.
4683 "Maybe" means test if the RCU implementation
4684 under test support RCU priority boosting.
4686 rcutorture.test_boost_duration= [KNL]
4687 Duration (s) of each individual boost test.
4689 rcutorture.test_boost_interval= [KNL]
4690 Interval (s) between each boost test.
4692 rcutorture.test_no_idle_hz= [KNL]
4693 Test RCU's dyntick-idle handling. See also the
4694 rcutorture.shuffle_interval parameter.
4696 rcutorture.torture_type= [KNL]
4697 Specify the RCU implementation to test.
4699 rcutorture.verbose= [KNL]
4700 Enable additional printk() statements.
4702 rcupdate.rcu_cpu_stall_ftrace_dump= [KNL]
4703 Dump ftrace buffer after reporting RCU CPU
4706 rcupdate.rcu_cpu_stall_suppress= [KNL]
4707 Suppress RCU CPU stall warning messages.
4709 rcupdate.rcu_cpu_stall_suppress_at_boot= [KNL]
4710 Suppress RCU CPU stall warning messages and
4711 rcutorture writer stall warnings that occur
4712 during early boot, that is, during the time
4713 before the init task is spawned.
4715 rcupdate.rcu_cpu_stall_timeout= [KNL]
4716 Set timeout for RCU CPU stall warning messages.
4718 rcupdate.rcu_expedited= [KNL]
4719 Use expedited grace-period primitives, for
4720 example, synchronize_rcu_expedited() instead
4721 of synchronize_rcu(). This reduces latency,
4722 but can increase CPU utilization, degrade
4723 real-time latency, and degrade energy efficiency.
4724 No effect on CONFIG_TINY_RCU kernels.
4726 rcupdate.rcu_normal= [KNL]
4727 Use only normal grace-period primitives,
4728 for example, synchronize_rcu() instead of
4729 synchronize_rcu_expedited(). This improves
4730 real-time latency, CPU utilization, and
4731 energy efficiency, but can expose users to
4732 increased grace-period latency. This parameter
4733 overrides rcupdate.rcu_expedited. No effect on
4734 CONFIG_TINY_RCU kernels.
4736 rcupdate.rcu_normal_after_boot= [KNL]
4737 Once boot has completed (that is, after
4738 rcu_end_inkernel_boot() has been invoked), use
4739 only normal grace-period primitives. No effect
4740 on CONFIG_TINY_RCU kernels.
4742 But note that CONFIG_PREEMPT_RT=y kernels enables
4743 this kernel boot parameter, forcibly setting
4744 it to the value one, that is, converting any
4745 post-boot attempt at an expedited RCU grace
4746 period to instead use normal non-expedited
4747 grace-period processing.
4749 rcupdate.rcu_task_ipi_delay= [KNL]
4750 Set time in jiffies during which RCU tasks will
4751 avoid sending IPIs, starting with the beginning
4752 of a given grace period. Setting a large
4753 number avoids disturbing real-time workloads,
4754 but lengthens grace periods.
4756 rcupdate.rcu_task_stall_timeout= [KNL]
4757 Set timeout in jiffies for RCU task stall warning
4758 messages. Disable with a value less than or equal
4761 rcupdate.rcu_self_test= [KNL]
4762 Run the RCU early boot self tests
4766 Run specified binary instead of /init from the ramdisk,
4767 used for early userspace startup. See initrd.
4770 force - Override the decision by the kernel to hide the
4771 advertisement of RDRAND support (this affects
4772 certain AMD processors because of buggy BIOS
4773 support, specifically around the suspend/resume
4777 Turn on/off individual RDT features. List is:
4778 cmt, mbmtotal, mbmlocal, l3cat, l3cdp, l2cat, l2cdp,
4780 E.g. to turn on cmt and turn off mba use:
4784 Format (x86 or x86_64):
4785 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] \
4787 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \
4789 Where reboot_mode is one of warm (soft) or cold (hard) or gpio
4790 (prefix with 'panic_' to set mode for panic
4792 reboot_type is one of bios, acpi, kbd, triple, efi, or pci,
4793 reboot_force is either force or not specified,
4794 reboot_cpu is s[mp]#### with #### being the processor
4795 to be used for rebooting.
4797 refscale.holdoff= [KNL]
4798 Set test-start holdoff period. The purpose of
4799 this parameter is to delay the start of the
4800 test until boot completes in order to avoid
4803 refscale.loops= [KNL]
4804 Set the number of loops over the synchronization
4805 primitive under test. Increasing this number
4806 reduces noise due to loop start/end overhead,
4807 but the default has already reduced the per-pass
4808 noise to a handful of picoseconds on ca. 2020
4811 refscale.nreaders= [KNL]
4812 Set number of readers. The default value of -1
4813 selects N, where N is roughly 75% of the number
4814 of CPUs. A value of zero is an interesting choice.
4816 refscale.nruns= [KNL]
4817 Set number of runs, each of which is dumped onto
4820 refscale.readdelay= [KNL]
4821 Set the read-side critical-section duration,
4822 measured in microseconds.
4824 refscale.scale_type= [KNL]
4825 Specify the read-protection implementation to test.
4827 refscale.shutdown= [KNL]
4828 Shut down the system at the end of the performance
4829 test. This defaults to 1 (shut it down) when
4830 refscale is built into the kernel and to 0 (leave
4831 it running) when refscale is built as a module.
4833 refscale.verbose= [KNL]
4834 Enable additional printk() statements.
4836 refscale.verbose_batched= [KNL]
4837 Batch the additional printk() statements. If zero
4838 (the default) or negative, print everything. Otherwise,
4839 print every Nth verbose statement, where N is the value
4843 [KNL, SMP] Set scheduler's default relax_domain_level.
4844 See Documentation/admin-guide/cgroup-v1/cpusets.rst.
4846 reserve= [KNL,BUGS] Force kernel to ignore I/O ports or memory
4847 Format: <base1>,<size1>[,<base2>,<size2>,...]
4848 Reserve I/O ports or memory so the kernel won't use
4849 them. If <base> is less than 0x10000, the region
4850 is assumed to be I/O ports; otherwise it is memory.
4852 reservetop= [X86-32]
4854 Reserves a hole at the top of the kernel virtual
4857 reset_devices [KNL] Force drivers to reset the underlying device
4858 during initialization.
4861 Specify the partition device for software suspend
4863 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>}
4865 resume_offset= [SWSUSP]
4866 Specify the offset from the beginning of the partition
4867 given by "resume=" at which the swap header is located,
4868 in <PAGE_SIZE> units (needed only for swap files).
4869 See Documentation/power/swsusp-and-swap-files.rst
4871 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to
4872 read the resume files
4874 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up.
4875 Useful for devices that are detected asynchronously
4876 (e.g. USB and MMC devices).
4878 hibernate= [HIBERNATION]
4879 noresume Don't check if there's a hibernation image
4880 present during boot.
4881 nocompress Don't compress/decompress hibernation images.
4882 no Disable hibernation and resume.
4883 protect_image Turn on image protection during restoration
4884 (that will set all pages holding image data
4885 during restoration read-only).
4887 retain_initrd [RAM] Keep initrd memory after extraction
4889 rfkill.default_state=
4890 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm,
4891 etc. communication is blocked by default.
4894 rfkill.master_switch_mode=
4895 0 The "airplane mode" button does nothing.
4896 1 The "airplane mode" button toggles between everything
4897 blocked and the previous configuration.
4898 2 The "airplane mode" button toggles between everything
4899 blocked and everything unblocked.
4901 rhash_entries= [KNL,NET]
4902 Set number of hash buckets for route cache
4905 [KNL] Disable ring 3 MONITOR/MWAIT feature on supported
4908 ro [KNL] Mount root device read-only on boot
4911 on Mark read-only kernel memory as read-only (default).
4912 off Leave read-only kernel memory writable for debugging.
4915 Enable the uart passthrough on the designated usb port
4916 on Rockchip SoCs. When active, the signals of the
4917 debug-uart get routed to the D+ and D- pins of the usb
4918 port and the regular usb controller gets disabled.
4920 root= [KNL] Root filesystem
4921 See name_to_dev_t comment in init/do_mounts.c.
4923 rootdelay= [KNL] Delay (in seconds) to pause before attempting to
4924 mount the root filesystem
4926 rootflags= [KNL] Set root filesystem mount option string
4928 rootfstype= [KNL] Set root filesystem type
4930 rootwait [KNL] Wait (indefinitely) for root device to show up.
4931 Useful for devices that are detected asynchronously
4932 (e.g. USB and MMC devices).
4934 rproc_mem=nn[KMG][@address]
4935 [KNL,ARM,CMA] Remoteproc physical memory block.
4936 Memory area to be used by remote processor image,
4939 rw [KNL] Mount root device read-write on boot
4941 S [KNL] Run init in single mode
4943 s390_iommu= [HW,S390]
4944 Set s390 IOTLB flushing mode
4946 With strict flushing every unmap operation will result in
4947 an IOTLB flush. Default is lazy flushing before reuse,
4951 See drivers/net/irda/sa1100_ir.c.
4953 sbni= [NET] Granch SBNI12 leased line adapter
4955 sched_verbose [KNL] Enables verbose scheduler debug messages.
4957 schedstats= [KNL,X86] Enable or disable scheduled statistics.
4958 Allowed values are enable and disable. This feature
4959 incurs a small amount of overhead in the scheduler
4960 but is useful for debugging and performance tuning.
4962 sched_thermal_decay_shift=
4963 [KNL, SMP] Set a decay shift for scheduler thermal
4964 pressure signal. Thermal pressure signal follows the
4965 default decay period of other scheduler pelt
4966 signals(usually 32 ms but configurable). Setting
4967 sched_thermal_decay_shift will left shift the decay
4968 period for the thermal pressure signal by the shift
4970 i.e. with the default pelt decay period of 32 ms
4971 sched_thermal_decay_shift thermal pressure decay pr
4975 Format: integer between 0 and 10
4978 scftorture.holdoff= [KNL]
4979 Number of seconds to hold off before starting
4980 test. Defaults to zero for module insertion and
4981 to 10 seconds for built-in smp_call_function()
4984 scftorture.longwait= [KNL]
4985 Request ridiculously long waits randomly selected
4986 up to the chosen limit in seconds. Zero (the
4987 default) disables this feature. Please note
4988 that requesting even small non-zero numbers of
4989 seconds can result in RCU CPU stall warnings,
4990 softlockup complaints, and so on.
4992 scftorture.nthreads= [KNL]
4993 Number of kthreads to spawn to invoke the
4994 smp_call_function() family of functions.
4995 The default of -1 specifies a number of kthreads
4996 equal to the number of CPUs.
4998 scftorture.onoff_holdoff= [KNL]
4999 Number seconds to wait after the start of the
5000 test before initiating CPU-hotplug operations.
5002 scftorture.onoff_interval= [KNL]
5003 Number seconds to wait between successive
5004 CPU-hotplug operations. Specifying zero (which
5005 is the default) disables CPU-hotplug operations.
5007 scftorture.shutdown_secs= [KNL]
5008 The number of seconds following the start of the
5009 test after which to shut down the system. The
5010 default of zero avoids shutting down the system.
5011 Non-zero values are useful for automated tests.
5013 scftorture.stat_interval= [KNL]
5014 The number of seconds between outputting the
5015 current test statistics to the console. A value
5016 of zero disables statistics output.
5018 scftorture.stutter_cpus= [KNL]
5019 The number of jiffies to wait between each change
5020 to the set of CPUs under test.
5022 scftorture.use_cpus_read_lock= [KNL]
5023 Use use_cpus_read_lock() instead of the default
5024 preempt_disable() to disable CPU hotplug
5025 while invoking one of the smp_call_function*()
5028 scftorture.verbose= [KNL]
5029 Enable additional printk() statements.
5031 scftorture.weight_single= [KNL]
5032 The probability weighting to use for the
5033 smp_call_function_single() function with a zero
5034 "wait" parameter. A value of -1 selects the
5035 default if all other weights are -1. However,
5036 if at least one weight has some other value, a
5037 value of -1 will instead select a weight of zero.
5039 scftorture.weight_single_wait= [KNL]
5040 The probability weighting to use for the
5041 smp_call_function_single() function with a
5042 non-zero "wait" parameter. See weight_single.
5044 scftorture.weight_many= [KNL]
5045 The probability weighting to use for the
5046 smp_call_function_many() function with a zero
5047 "wait" parameter. See weight_single.
5048 Note well that setting a high probability for
5049 this weighting can place serious IPI load
5052 scftorture.weight_many_wait= [KNL]
5053 The probability weighting to use for the
5054 smp_call_function_many() function with a
5055 non-zero "wait" parameter. See weight_single
5058 scftorture.weight_all= [KNL]
5059 The probability weighting to use for the
5060 smp_call_function_all() function with a zero
5061 "wait" parameter. See weight_single and
5064 scftorture.weight_all_wait= [KNL]
5065 The probability weighting to use for the
5066 smp_call_function_all() function with a
5067 non-zero "wait" parameter. See weight_single
5070 skew_tick= [KNL] Offset the periodic timer tick per cpu to mitigate
5071 xtime_lock contention on larger systems, and/or RCU lock
5072 contention on all systems with CONFIG_MAXSMP set.
5073 Format: { "0" | "1" }
5074 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1"
5076 Note: increases power consumption, thus should only be
5077 enabled if running jitter sensitive (HPC/RT) workloads.
5079 security= [SECURITY] Choose a legacy "major" security module to
5080 enable at boot. This has been deprecated by the
5083 selinux= [SELINUX] Disable or enable SELinux at boot time.
5084 Format: { "0" | "1" }
5085 See security/selinux/Kconfig help text.
5090 apparmor= [APPARMOR] Disable or enable AppArmor at boot time
5091 Format: { "0" | "1" }
5092 See security/apparmor/Kconfig help text
5095 Default value is set via kernel config option.
5097 serialnumber [BUGS=X86-32]
5100 Maximal number of shapers.
5108 Enable merging of slabs with similar size when the
5109 kernel is built without CONFIG_SLAB_MERGE_DEFAULT.
5112 Disable merging of slabs with similar size. May be
5113 necessary if there is some reason to distinguish
5114 allocs to different slabs, especially in hardened
5115 environments where the risk of heap overflows and
5116 layout control by attackers can usually be
5117 frustrated by disabling merging. This will reduce
5118 most of the exposure of a heap attack to a single
5119 cache (risks via metadata attacks are mostly
5120 unchanged). Debug options disable merging on their
5122 For more information see Documentation/vm/slub.rst.
5124 slab_max_order= [MM, SLAB]
5125 Determines the maximum allowed order for slabs.
5126 A high setting may cause OOMs due to memory
5127 fragmentation. Defaults to 1 for systems with
5128 more than 32MB of RAM, 0 otherwise.
5130 slub_debug[=options[,slabs][;[options[,slabs]]...] [MM, SLUB]
5131 Enabling slub_debug allows one to determine the
5132 culprit if slab objects become corrupted. Enabling
5133 slub_debug can create guard zones around objects and
5134 may poison objects when not in use. Also tracks the
5135 last alloc / free. For more information see
5136 Documentation/vm/slub.rst.
5138 slub_max_order= [MM, SLUB]
5139 Determines the maximum allowed order for slabs.
5140 A high setting may cause OOMs due to memory
5141 fragmentation. For more information see
5142 Documentation/vm/slub.rst.
5144 slub_min_objects= [MM, SLUB]
5145 The minimum number of objects per slab. SLUB will
5146 increase the slab order up to slub_max_order to
5147 generate a sufficiently large slab able to contain
5148 the number of objects indicated. The higher the number
5149 of objects the smaller the overhead of tracking slabs
5150 and the less frequently locks need to be acquired.
5151 For more information see Documentation/vm/slub.rst.
5153 slub_min_order= [MM, SLUB]
5154 Determines the minimum page order for slabs. Must be
5155 lower than slub_max_order.
5156 For more information see Documentation/vm/slub.rst.
5158 slub_merge [MM, SLUB]
5159 Same with slab_merge.
5161 slub_nomerge [MM, SLUB]
5162 Same with slab_nomerge. This is supported for legacy.
5163 See slab_nomerge for more information.
5166 Format: <io1>[,<io2>[,...,<io8>]]
5168 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices
5169 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port
5170 smsc-ircc2.ircc_sir= [HW] SIR base I/O port
5171 smsc-ircc2.ircc_fir= [HW] FIR base I/O port
5172 smsc-ircc2.ircc_irq= [HW] IRQ line
5173 smsc-ircc2.ircc_dma= [HW] DMA channel
5174 smsc-ircc2.ircc_transceiver= [HW] Transceiver type:
5175 0: Toshiba Satellite 1800 (GP data pin select)
5176 1: Fast pin select (default)
5179 smt [KNL,S390] Set the maximum number of threads (logical
5180 CPUs) to use per physical CPU on systems capable of
5181 symmetric multithreading (SMT). Will be capped to the
5182 actual hardware limit.
5184 Default: -1 (no limit)
5187 [KNL] Should the soft-lockup detector generate panics.
5190 A value of 1 instructs the soft-lockup detector
5191 to panic the machine when a soft-lockup occurs. It is
5192 also controlled by the kernel.softlockup_panic sysctl
5193 and CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC, which is the
5194 respective build-time switch to that functionality.
5196 softlockup_all_cpu_backtrace=
5197 [KNL] Should the soft-lockup detector generate
5198 backtraces on all cpus.
5201 sonypi.*= [HW] Sony Programmable I/O Control Device driver
5202 See Documentation/admin-guide/laptops/sonypi.rst
5204 spectre_v2= [X86] Control mitigation of Spectre variant 2
5205 (indirect branch speculation) vulnerability.
5206 The default operation protects the kernel from
5209 on - unconditionally enable, implies
5211 off - unconditionally disable, implies
5213 auto - kernel detects whether your CPU model is
5216 Selecting 'on' will, and 'auto' may, choose a
5217 mitigation method at run time according to the
5218 CPU, the available microcode, the setting of the
5219 CONFIG_RETPOLINE configuration option, and the
5220 compiler with which the kernel was built.
5222 Selecting 'on' will also enable the mitigation
5223 against user space to user space task attacks.
5225 Selecting 'off' will disable both the kernel and
5226 the user space protections.
5228 Specific mitigations can also be selected manually:
5230 retpoline - replace indirect branches
5231 retpoline,generic - google's original retpoline
5232 retpoline,amd - AMD-specific minimal thunk
5234 Not specifying this option is equivalent to
5238 [X86] Control mitigation of Spectre variant 2
5239 (indirect branch speculation) vulnerability between
5242 on - Unconditionally enable mitigations. Is
5243 enforced by spectre_v2=on
5245 off - Unconditionally disable mitigations. Is
5246 enforced by spectre_v2=off
5248 prctl - Indirect branch speculation is enabled,
5249 but mitigation can be enabled via prctl
5250 per thread. The mitigation control state
5251 is inherited on fork.
5254 - Like "prctl" above, but only STIBP is
5255 controlled per thread. IBPB is issued
5256 always when switching between different user
5260 - Same as "prctl" above, but all seccomp
5261 threads will enable the mitigation unless
5262 they explicitly opt out.
5265 - Like "seccomp" above, but only STIBP is
5266 controlled per thread. IBPB is issued
5267 always when switching between different
5268 user space processes.
5270 auto - Kernel selects the mitigation depending on
5271 the available CPU features and vulnerability.
5274 If CONFIG_SECCOMP=y then "seccomp", otherwise "prctl"
5276 Not specifying this option is equivalent to
5277 spectre_v2_user=auto.
5279 spec_store_bypass_disable=
5280 [HW] Control Speculative Store Bypass (SSB) Disable mitigation
5281 (Speculative Store Bypass vulnerability)
5283 Certain CPUs are vulnerable to an exploit against a
5284 a common industry wide performance optimization known
5285 as "Speculative Store Bypass" in which recent stores
5286 to the same memory location may not be observed by
5287 later loads during speculative execution. The idea
5288 is that such stores are unlikely and that they can
5289 be detected prior to instruction retirement at the
5290 end of a particular speculation execution window.
5292 In vulnerable processors, the speculatively forwarded
5293 store can be used in a cache side channel attack, for
5294 example to read memory to which the attacker does not
5295 directly have access (e.g. inside sandboxed code).
5297 This parameter controls whether the Speculative Store
5298 Bypass optimization is used.
5300 On x86 the options are:
5302 on - Unconditionally disable Speculative Store Bypass
5303 off - Unconditionally enable Speculative Store Bypass
5304 auto - Kernel detects whether the CPU model contains an
5305 implementation of Speculative Store Bypass and
5306 picks the most appropriate mitigation. If the
5307 CPU is not vulnerable, "off" is selected. If the
5308 CPU is vulnerable the default mitigation is
5309 architecture and Kconfig dependent. See below.
5310 prctl - Control Speculative Store Bypass per thread
5311 via prctl. Speculative Store Bypass is enabled
5312 for a process by default. The state of the control
5313 is inherited on fork.
5314 seccomp - Same as "prctl" above, but all seccomp threads
5315 will disable SSB unless they explicitly opt out.
5317 Default mitigations:
5318 X86: If CONFIG_SECCOMP=y "seccomp", otherwise "prctl"
5320 On powerpc the options are:
5322 on,auto - On Power8 and Power9 insert a store-forwarding
5323 barrier on kernel entry and exit. On Power7
5324 perform a software flush on kernel entry and
5328 Not specifying this option is equivalent to
5329 spec_store_bypass_disable=auto.
5331 spia_io_base= [HW,MTD]
5337 [X86] Enable split lock detection or bus lock detection
5339 When enabled (and if hardware support is present), atomic
5340 instructions that access data across cache line
5341 boundaries will result in an alignment check exception
5342 for split lock detection or a debug exception for
5347 warn - the kernel will emit rate-limited warnings
5348 about applications triggering the #AC
5349 exception or the #DB exception. This mode is
5350 the default on CPUs that support split lock
5351 detection or bus lock detection. Default
5352 behavior is by #AC if both features are
5353 enabled in hardware.
5355 fatal - the kernel will send SIGBUS to applications
5356 that trigger the #AC exception or the #DB
5357 exception. Default behavior is by #AC if
5358 both features are enabled in hardware.
5361 Set system wide rate limit to N bus locks
5362 per second for bus lock detection.
5365 N/A for split lock detection.
5368 If an #AC exception is hit in the kernel or in
5369 firmware (i.e. not while executing in user mode)
5370 the kernel will oops in either "warn" or "fatal"
5373 #DB exception for bus lock is triggered only when
5377 Control the Special Register Buffer Data Sampling
5380 Certain CPUs are vulnerable to an MDS-like
5381 exploit which can leak bits from the random
5384 By default, this issue is mitigated by
5385 microcode. However, the microcode fix can cause
5386 the RDRAND and RDSEED instructions to become
5387 much slower. Among other effects, this will
5388 result in reduced throughput from /dev/urandom.
5390 The microcode mitigation can be disabled with
5391 the following option:
5393 off: Disable mitigation and remove
5394 performance impact to RDRAND and RDSEED
5396 srcutree.counter_wrap_check [KNL]
5397 Specifies how frequently to check for
5398 grace-period sequence counter wrap for the
5399 srcu_data structure's ->srcu_gp_seq_needed field.
5400 The greater the number of bits set in this kernel
5401 parameter, the less frequently counter wrap will
5402 be checked for. Note that the bottom two bits
5405 srcutree.exp_holdoff [KNL]
5406 Specifies how many nanoseconds must elapse
5407 since the end of the last SRCU grace period for
5408 a given srcu_struct until the next normal SRCU
5409 grace period will be considered for automatic
5410 expediting. Set to zero to disable automatic
5414 Speculative Store Bypass Disable control
5416 On CPUs that are vulnerable to the Speculative
5417 Store Bypass vulnerability and offer a
5418 firmware based mitigation, this parameter
5419 indicates how the mitigation should be used:
5421 force-on: Unconditionally enable mitigation for
5422 for both kernel and userspace
5423 force-off: Unconditionally disable mitigation for
5424 for both kernel and userspace
5425 kernel: Always enable mitigation in the
5426 kernel, and offer a prctl interface
5427 to allow userspace to register its
5428 interest in being mitigated too.
5430 stack_guard_gap= [MM]
5431 override the default stack gap protection. The value
5432 is in page units and it defines how many pages prior
5433 to (for stacks growing down) resp. after (for stacks
5434 growing up) the main stack are reserved for no other
5435 mapping. Default value is 256 pages.
5437 stack_depot_disable= [KNL]
5438 Setting this to true through kernel command line will
5439 disable the stack depot thereby saving the static memory
5440 consumed by the stack hash table. By default this is set
5444 Enabled the stack tracer on boot up.
5446 stacktrace_filter=[function-list]
5447 [FTRACE] Limit the functions that the stack tracer
5448 will trace at boot up. function-list is a comma-separated
5449 list of functions. This list can be changed at run
5450 time by the stack_trace_filter file in the debugfs
5451 tracing directory. Note, this enables stack tracing
5452 and the stacktrace above is not needed.
5456 Set the STI (builtin display/keyboard on the HP-PARISC
5457 machines) console (graphic card) which should be used
5458 as the initial boot-console.
5459 See also comment in drivers/video/console/sticore.c.
5462 See comment in drivers/video/console/sticore.c.
5465 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]]
5467 sunrpc.min_resvport=
5468 sunrpc.max_resvport=
5470 SunRPC servers often require that client requests
5471 originate from a privileged port (i.e. a port in the
5472 range 0 < portnr < 1024).
5473 An administrator who wishes to reserve some of these
5474 ports for other uses may adjust the range that the
5475 kernel's sunrpc client considers to be privileged
5476 using these two parameters to set the minimum and
5477 maximum port values.
5479 sunrpc.svc_rpc_per_connection_limit=
5481 Limit the number of requests that the server will
5482 process in parallel from a single connection.
5483 The default value is 0 (no limit).
5487 Control how the NFS server code allocates CPUs to
5488 service thread pools. Depending on how many NICs
5489 you have and where their interrupts are bound, this
5490 option will affect which CPUs will do NFS serving.
5491 Note: this parameter cannot be changed while the
5492 NFS server is running.
5494 auto the server chooses an appropriate mode
5495 automatically using heuristics
5496 global a single global pool contains all CPUs
5497 percpu one pool for each CPU
5498 pernode one pool for each NUMA node (equivalent
5499 to global on non-NUMA machines)
5501 sunrpc.tcp_slot_table_entries=
5502 sunrpc.udp_slot_table_entries=
5504 Sets the upper limit on the number of simultaneous
5505 RPC calls that can be sent from the client to a
5506 server. Increasing these values may allow you to
5507 improve throughput, but will also increase the
5508 amount of memory reserved for use by the client.
5510 suspend.pm_test_delay=
5512 Sets the number of seconds to remain in a suspend test
5513 mode before resuming the system (see
5514 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG
5515 is set. Default value is 5.
5518 Format: { on | off | y | n | 1 | 0 }
5519 This parameter controls use of the Protected
5520 Execution Facility on pSeries.
5523 [KNL] Enable accounting of swap in memory resource
5524 controller if no parameter or 1 is given or disable
5525 it if 0 is given (See Documentation/admin-guide/cgroup-v1/memory.rst)
5527 swiotlb= [ARM,IA-64,PPC,MIPS,X86]
5528 Format: { <int> | force | noforce }
5529 <int> -- Number of I/O TLB slabs
5530 force -- force using of bounce buffers even if they
5531 wouldn't be automatically used by the kernel
5532 noforce -- Never use bounce buffers (for debugging)
5537 Set a sysctl parameter, right before loading the init
5538 process, as if the value was written to the respective
5539 /proc/sys/... file. Both '.' and '/' are recognized as
5540 separators. Unrecognized parameters and invalid values
5541 are reported in the kernel log. Sysctls registered
5542 later by a loaded module cannot be set this way.
5543 Example: sysctl.vm.swappiness=40
5545 sysfs.deprecated=0|1 [KNL]
5546 Enable/disable old style sysfs layout for old udev
5547 on older distributions. When this option is enabled
5548 very new udev will not work anymore. When this option
5549 is disabled (or CONFIG_SYSFS_DEPRECATED not compiled)
5550 in older udev will not work anymore.
5551 Default depends on CONFIG_SYSFS_DEPRECATED_V2 set in
5552 the kernel configuration.
5554 sysrq_always_enabled
5556 Ignore sysrq setting - this boot parameter will
5557 neutralize any effect of /proc/sys/kernel/sysrq.
5558 Useful for debugging.
5560 tcpmhash_entries= [KNL,NET]
5561 Set the number of tcp_metrics_hash slots.
5562 Default value is 8192 or 16384 depending on total
5563 ram pages. This is used to specify the TCP metrics
5564 cache size. See Documentation/networking/ip-sysctl.rst
5565 "tcp_no_metrics_save" section for more details.
5569 test_suspend= [SUSPEND][,N]
5570 Specify "mem" (for Suspend-to-RAM) or "standby" (for
5571 standby suspend) or "freeze" (for suspend type freeze)
5572 as the system sleep state during system startup with
5573 the optional capability to repeat N number of times.
5574 The system is woken from this state using a
5575 wakeup-capable RTC alarm.
5577 thash_entries= [KNL,NET]
5578 Set number of hash buckets for TCP connection
5580 thermal.act= [HW,ACPI]
5581 -1: disable all active trip points in all thermal zones
5582 <degrees C>: override all lowest active trip points
5584 thermal.crt= [HW,ACPI]
5585 -1: disable all critical trip points in all thermal zones
5586 <degrees C>: override all critical trip points
5588 thermal.nocrt= [HW,ACPI]
5589 Set to disable actions on ACPI thermal zone
5590 critical and hot trip points.
5592 thermal.off= [HW,ACPI]
5593 1: disable ACPI thermal control
5595 thermal.psv= [HW,ACPI]
5596 -1: disable all passive trip points
5597 <degrees C>: override all passive trip points to this
5600 thermal.tzp= [HW,ACPI]
5601 Specify global default ACPI thermal zone polling rate
5602 <deci-seconds>: poll all this frequency
5603 0: no polling (default)
5606 Force threading of all interrupt handlers except those
5607 marked explicitly IRQF_NO_THREAD.
5611 Specify if the kernel should make use of the cpu
5612 topology information if the hardware supports this.
5613 The scheduler will make use of this information and
5614 e.g. base its process migration decisions on it.
5617 topology_updates= [KNL, PPC, NUMA]
5619 Specify if the kernel should ignore (off)
5620 topology updates sent by the hypervisor to this
5623 torture.disable_onoff_at_boot= [KNL]
5624 Prevent the CPU-hotplug component of torturing
5625 until after init has spawned.
5627 torture.ftrace_dump_at_shutdown= [KNL]
5628 Dump the ftrace buffer at torture-test shutdown,
5629 even if there were no errors. This can be a
5630 very costly operation when many torture tests
5631 are running concurrently, especially on systems
5632 with rotating-rust storage.
5634 torture.verbose_sleep_frequency= [KNL]
5635 Specifies how many verbose printk()s should be
5636 emitted between each sleep. The default of zero
5637 disables verbose-printk() sleeping.
5639 torture.verbose_sleep_duration= [KNL]
5640 Duration of each verbose-printk() sleep in jiffies.
5644 tpm_suspend_pcr=[HW,TPM]
5645 Format: integer pcr id
5646 Specify that at suspend time, the tpm driver
5647 should extend the specified pcr with zeros,
5648 as a workaround for some chips which fail to
5649 flush the last written pcr on TPM_SaveState.
5650 This will guarantee that all the other pcrs
5653 trace_buf_size=nn[KMG]
5654 [FTRACE] will set tracing buffer size on each cpu.
5656 trace_event=[event-list]
5657 [FTRACE] Set and start specified trace events in order
5658 to facilitate early boot debugging. The event-list is a
5659 comma-separated list of trace events to enable. See
5660 also Documentation/trace/events.rst
5662 trace_options=[option-list]
5663 [FTRACE] Enable or disable tracer options at boot.
5664 The option-list is a comma delimited list of options
5665 that can be enabled or disabled just as if you were
5666 to echo the option name into
5668 /sys/kernel/debug/tracing/trace_options
5670 For example, to enable stacktrace option (to dump the
5671 stack trace of each event), add to the command line:
5673 trace_options=stacktrace
5675 See also Documentation/trace/ftrace.rst "trace options"
5679 Have the tracepoints sent to printk as well as the
5680 tracing ring buffer. This is useful for early boot up
5681 where the system hangs or reboots and does not give the
5682 option for reading the tracing buffer or performing a
5683 ftrace_dump_on_oops.
5685 To turn off having tracepoints sent to printk,
5686 echo 0 > /proc/sys/kernel/tracepoint_printk
5687 Note, echoing 1 into this file without the
5688 tracepoint_printk kernel cmdline option has no effect.
5690 The tp_printk_stop_on_boot (see below) can also be used
5691 to stop the printing of events to console at
5696 Having tracepoints sent to printk() and activating high
5697 frequency tracepoints such as irq or sched, can cause
5698 the system to live lock.
5700 tp_printk_stop_on_boot[FTRACE]
5701 When tp_printk (above) is set, it can cause a lot of noise
5702 on the console. It may be useful to only include the
5703 printing of events during boot up, as user space may
5704 make the system inoperable.
5706 This command line option will stop the printing of events
5707 to console at the late_initcall_sync() time frame.
5710 [FTRACE] enable this option to disable tracing when a
5711 warning is hit. This turns off "tracing_on". Tracing can
5712 be enabled again by echoing '1' into the "tracing_on"
5713 file located in /sys/kernel/debug/tracing/
5715 This option is useful, as it disables the trace before
5716 the WARNING dump is called, which prevents the trace to
5717 be filled with content caused by the warning output.
5719 This option can also be set at run time via the sysctl
5720 option: kernel/traceoff_on_warning
5722 transparent_hugepage=
5724 Format: [always|madvise|never]
5725 Can be used to control the default behavior of the system
5726 with respect to transparent hugepages.
5727 See Documentation/admin-guide/mm/transhuge.rst
5730 trusted.source= [KEYS]
5732 This parameter identifies the trust source as a backend
5733 for trusted keys implementation. Supported trust
5737 If not specified then it defaults to iterating through
5738 the trust source list starting with TPM and assigns the
5739 first trust source as a backend which is initialized
5740 successfully during iteration.
5742 tsc= Disable clocksource stability checks for TSC.
5744 [x86] reliable: mark tsc clocksource as reliable, this
5745 disables clocksource verification at runtime, as well
5746 as the stability checks done at bootup. Used to enable
5747 high-resolution timer mode on older hardware, and in
5748 virtualized environment.
5749 [x86] noirqtime: Do not use TSC to do irq accounting.
5750 Used to run time disable IRQ_TIME_ACCOUNTING on any
5751 platforms where RDTSC is slow and this accounting
5753 [x86] unstable: mark the TSC clocksource as unstable, this
5754 marks the TSC unconditionally unstable at bootup and
5755 avoids any further wobbles once the TSC watchdog notices.
5756 [x86] nowatchdog: disable clocksource watchdog. Used
5757 in situations with strict latency requirements (where
5758 interruptions from clocksource watchdog are not
5761 tsc_early_khz= [X86] Skip early TSC calibration and use the given
5762 value instead. Useful when the early TSC frequency discovery
5763 procedure is not reliable, such as on overclocked systems
5764 with CPUID.16h support and partial CPUID.15h support.
5765 Format: <unsigned int>
5767 tsx= [X86] Control Transactional Synchronization
5768 Extensions (TSX) feature in Intel processors that
5769 support TSX control.
5771 This parameter controls the TSX feature. The options are:
5773 on - Enable TSX on the system. Although there are
5774 mitigations for all known security vulnerabilities,
5775 TSX has been known to be an accelerator for
5776 several previous speculation-related CVEs, and
5777 so there may be unknown security risks associated
5778 with leaving it enabled.
5780 off - Disable TSX on the system. (Note that this
5781 option takes effect only on newer CPUs which are
5782 not vulnerable to MDS, i.e., have
5783 MSR_IA32_ARCH_CAPABILITIES.MDS_NO=1 and which get
5784 the new IA32_TSX_CTRL MSR through a microcode
5785 update. This new MSR allows for the reliable
5786 deactivation of the TSX functionality.)
5788 auto - Disable TSX if X86_BUG_TAA is present,
5789 otherwise enable TSX on the system.
5791 Not specifying this option is equivalent to tsx=off.
5793 See Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
5796 tsx_async_abort= [X86,INTEL] Control mitigation for the TSX Async
5797 Abort (TAA) vulnerability.
5799 Similar to Micro-architectural Data Sampling (MDS)
5800 certain CPUs that support Transactional
5801 Synchronization Extensions (TSX) are vulnerable to an
5802 exploit against CPU internal buffers which can forward
5803 information to a disclosure gadget under certain
5806 In vulnerable processors, the speculatively forwarded
5807 data can be used in a cache side channel attack, to
5808 access data to which the attacker does not have direct
5811 This parameter controls the TAA mitigation. The
5814 full - Enable TAA mitigation on vulnerable CPUs
5817 full,nosmt - Enable TAA mitigation and disable SMT on
5818 vulnerable CPUs. If TSX is disabled, SMT
5819 is not disabled because CPU is not
5820 vulnerable to cross-thread TAA attacks.
5821 off - Unconditionally disable TAA mitigation
5823 On MDS-affected machines, tsx_async_abort=off can be
5824 prevented by an active MDS mitigation as both vulnerabilities
5825 are mitigated with the same mechanism so in order to disable
5826 this mitigation, you need to specify mds=off too.
5828 Not specifying this option is equivalent to
5829 tsx_async_abort=full. On CPUs which are MDS affected
5830 and deploy MDS mitigation, TAA mitigation is not
5831 required and doesn't provide any additional
5835 Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
5837 turbografx.map[2|3]= [HW,JOY]
5838 TurboGraFX parallel port interface
5840 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7>
5841 See also Documentation/input/devices/joystick-parport.rst
5843 udbg-immortal [PPC] When debugging early kernel crashes that
5844 happen after console_init() and before a proper
5845 console driver takes over, this boot options might
5846 help "seeing" what's going on.
5848 uhash_entries= [KNL,NET]
5849 Set number of hash buckets for UDP/UDP-Lite connections
5852 [USB] Ignore overcurrent events (default N).
5853 Some badly-designed motherboards generate lots of
5854 bogus events, for ports that aren't wired to
5855 anything. Set this parameter to avoid log spamming.
5856 Note that genuine overcurrent events won't be
5860 [X86] Cause panic on unknown NMI.
5862 usbcore.authorized_default=
5863 [USB] Default USB device authorization:
5864 (default -1 = authorized except for wireless USB,
5865 0 = not authorized, 1 = authorized, 2 = authorized
5866 if device connected to internal port)
5868 usbcore.autosuspend=
5869 [USB] The autosuspend time delay (in seconds) used
5870 for newly-detected USB devices (default 2). This
5871 is the time required before an idle device will be
5872 autosuspended. Devices for which the delay is set
5873 to a negative value won't be autosuspended at all.
5875 usbcore.usbfs_snoop=
5876 [USB] Set to log all usbfs traffic (default 0 = off).
5878 usbcore.usbfs_snoop_max=
5879 [USB] Maximum number of bytes to snoop in each URB
5882 usbcore.blinkenlights=
5883 [USB] Set to cycle leds on hubs (default 0 = off).
5885 usbcore.old_scheme_first=
5886 [USB] Start with the old device initialization
5887 scheme (default 0 = off).
5889 usbcore.usbfs_memory_mb=
5890 [USB] Memory limit (in MB) for buffers allocated by
5891 usbfs (default = 16, 0 = max = 2047).
5893 usbcore.use_both_schemes=
5894 [USB] Try the other device initialization scheme
5895 if the first one fails (default 1 = enabled).
5897 usbcore.initial_descriptor_timeout=
5898 [USB] Specifies timeout for the initial 64-byte
5899 USB_REQ_GET_DESCRIPTOR request in milliseconds
5900 (default 5000 = 5.0 seconds).
5902 usbcore.nousb [USB] Disable the USB subsystem
5905 [USB] A list of quirk entries to augment the built-in
5906 usb core quirk list. List entries are separated by
5907 commas. Each entry has the form
5908 VendorID:ProductID:Flags. The IDs are 4-digit hex
5909 numbers and Flags is a set of letters. Each letter
5910 will change the built-in quirk; setting it if it is
5911 clear and clearing it if it is set. The letters have
5912 the following meanings:
5913 a = USB_QUIRK_STRING_FETCH_255 (string
5914 descriptors must not be fetched using
5916 b = USB_QUIRK_RESET_RESUME (device can't resume
5917 correctly so reset it instead);
5918 c = USB_QUIRK_NO_SET_INTF (device can't handle
5919 Set-Interface requests);
5920 d = USB_QUIRK_CONFIG_INTF_STRINGS (device can't
5921 handle its Configuration or Interface
5923 e = USB_QUIRK_RESET (device can't be reset
5924 (e.g morph devices), don't use reset);
5925 f = USB_QUIRK_HONOR_BNUMINTERFACES (device has
5926 more interface descriptions than the
5927 bNumInterfaces count, and can't handle
5928 talking to these interfaces);
5929 g = USB_QUIRK_DELAY_INIT (device needs a pause
5930 during initialization, after we read
5931 the device descriptor);
5932 h = USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL (For
5933 high speed and super speed interrupt
5934 endpoints, the USB 2.0 and USB 3.0 spec
5935 require the interval in microframes (1
5936 microframe = 125 microseconds) to be
5937 calculated as interval = 2 ^
5939 Devices with this quirk report their
5940 bInterval as the result of this
5941 calculation instead of the exponent
5942 variable used in the calculation);
5943 i = USB_QUIRK_DEVICE_QUALIFIER (device can't
5944 handle device_qualifier descriptor
5946 j = USB_QUIRK_IGNORE_REMOTE_WAKEUP (device
5947 generates spurious wakeup, ignore
5948 remote wakeup capability);
5949 k = USB_QUIRK_NO_LPM (device can't handle Link
5951 l = USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL
5952 (Device reports its bInterval as linear
5953 frames instead of the USB 2.0
5955 m = USB_QUIRK_DISCONNECT_SUSPEND (Device needs
5956 to be disconnected before suspend to
5957 prevent spurious wakeup);
5958 n = USB_QUIRK_DELAY_CTRL_MSG (Device needs a
5959 pause after every control message);
5960 o = USB_QUIRK_HUB_SLOW_RESET (Hub needs extra
5961 delay after resetting its port);
5962 Example: quirks=0781:5580:bk,0a5c:5834:gij
5965 [USBHID] The interval which mice are to be polled at.
5968 [USBHID] The interval which joysticks are to be polled at.
5971 [USBHID] The interval which keyboards are to be polled at.
5973 usb-storage.delay_use=
5974 [UMS] The delay in seconds before a new device is
5975 scanned for Logical Units (default 1).
5978 [UMS] A list of quirks entries to supplement or
5979 override the built-in unusual_devs list. List
5980 entries are separated by commas. Each entry has
5981 the form VID:PID:Flags where VID and PID are Vendor
5982 and Product ID values (4-digit hex numbers) and
5983 Flags is a set of characters, each corresponding
5984 to a common usb-storage quirk flag as follows:
5985 a = SANE_SENSE (collect more than 18 bytes
5986 of sense data, not on uas);
5987 b = BAD_SENSE (don't collect more than 18
5988 bytes of sense data, not on uas);
5989 c = FIX_CAPACITY (decrease the reported
5990 device capacity by one sector);
5991 d = NO_READ_DISC_INFO (don't use
5992 READ_DISC_INFO command, not on uas);
5993 e = NO_READ_CAPACITY_16 (don't use
5994 READ_CAPACITY_16 command);
5995 f = NO_REPORT_OPCODES (don't use report opcodes
5997 g = MAX_SECTORS_240 (don't transfer more than
5998 240 sectors at a time, uas only);
5999 h = CAPACITY_HEURISTICS (decrease the
6000 reported device capacity by one
6001 sector if the number is odd);
6002 i = IGNORE_DEVICE (don't bind to this
6004 j = NO_REPORT_LUNS (don't use report luns
6006 k = NO_SAME (do not use WRITE_SAME, uas only)
6007 l = NOT_LOCKABLE (don't try to lock and
6008 unlock ejectable media, not on uas);
6009 m = MAX_SECTORS_64 (don't transfer more
6010 than 64 sectors = 32 KB at a time,
6012 n = INITIAL_READ10 (force a retry of the
6013 initial READ(10) command, not on uas);
6014 o = CAPACITY_OK (accept the capacity
6015 reported by the device, not on uas);
6016 p = WRITE_CACHE (the device cache is ON
6017 by default, not on uas);
6018 r = IGNORE_RESIDUE (the device reports
6019 bogus residue values, not on uas);
6020 s = SINGLE_LUN (the device has only one
6022 t = NO_ATA_1X (don't allow ATA(12) and ATA(16)
6023 commands, uas only);
6024 u = IGNORE_UAS (don't bind to the uas driver);
6025 w = NO_WP_DETECT (don't test whether the
6026 medium is write-protected).
6027 y = ALWAYS_SYNC (issue a SYNCHRONIZE_CACHE
6028 even if the device claims no cache,
6030 Example: quirks=0419:aaf5:rl,0421:0433:rc
6032 user_debug= [KNL,ARM]
6034 See arch/arm/Kconfig.debug help text.
6035 1 - undefined instruction events
6037 4 - invalid data aborts
6040 Example: user_debug=31
6043 [X86] Flags controlling user PTE allocations.
6045 nohigh = do not allocate PTE pages in
6046 HIGHMEM regardless of setting
6050 On X86_32, this is an alias for vdso32=. Otherwise:
6052 vdso=1: enable VDSO (the default)
6053 vdso=0: disable VDSO mapping
6055 vdso32= [X86] Control the 32-bit vDSO
6056 vdso32=1: enable 32-bit VDSO
6057 vdso32=0 or vdso32=2: disable 32-bit VDSO
6059 See the help text for CONFIG_COMPAT_VDSO for more
6060 details. If CONFIG_COMPAT_VDSO is set, the default is
6061 vdso32=0; otherwise, the default is vdso32=1.
6063 For compatibility with older kernels, vdso32=2 is an
6066 Try vdso32=0 if you encounter an error that says:
6067 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed!
6070 vector=percpu: enable percpu vector domain
6072 video= [FB] Frame buffer configuration
6073 See Documentation/fb/modedb.rst.
6075 video.brightness_switch_enabled= [0,1]
6076 If set to 1, on receiving an ACPI notify event
6077 generated by hotkey, video driver will adjust brightness
6078 level and then send out the event to user space through
6079 the allocated input device; If set to 0, video driver
6080 will only send out the event without touching backlight
6085 [VMMIO] Memory mapped virtio (platform) device.
6087 <size>@<baseaddr>:<irq>[:<id>]
6089 <size> := size (can use standard suffixes
6091 <baseaddr> := physical base address
6092 <irq> := interrupt number (as passed to
6094 <id> := (optional) platform device id
6096 virtio_mmio.device=1K@0x100b0000:48:7
6098 Can be used multiple times for multiple devices.
6100 vga= [BOOT,X86-32] Select a particular video mode
6101 See Documentation/x86/boot.rst and
6102 Documentation/admin-guide/svga.rst.
6103 Use vga=ask for menu.
6104 This is actually a boot loader parameter; the value is
6105 passed to the kernel using a special protocol.
6107 vm_debug[=options] [KNL] Available with CONFIG_DEBUG_VM=y.
6108 May slow down system boot speed, especially when
6109 enabled on systems with a large amount of memory.
6110 All options are enabled by default, and this
6111 interface is meant to allow for selectively
6112 enabling or disabling specific virtual memory
6115 Available options are:
6116 P Enable page structure init time poisoning
6117 - Disable all of the above options
6119 vmalloc=nn[KMG] [KNL,BOOT] Forces the vmalloc area to have an exact
6120 size of <nn>. This can be used to increase the
6121 minimum size (128MB on x86). It can also be used to
6122 decrease the size and leave more room for directly
6125 vmcp_cma=nn[MG] [KNL,S390]
6126 Sets the memory size reserved for contiguous memory
6127 allocations for the vmcp device driver.
6129 vmhalt= [KNL,S390] Perform z/VM CP command after system halt.
6132 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic.
6135 vmpoff= [KNL,S390] Perform z/VM CP command after power off.
6139 Controls the behavior of vsyscalls (i.e. calls to
6140 fixed addresses of 0xffffffffff600x00 from legacy
6141 code). Most statically-linked binaries and older
6142 versions of glibc use these calls. Because these
6143 functions are at fixed addresses, they make nice
6144 targets for exploits that can control RIP.
6146 emulate [default] Vsyscalls turn into traps and are
6147 emulated reasonably safely. The vsyscall
6150 xonly Vsyscalls turn into traps and are
6151 emulated reasonably safely. The vsyscall
6152 page is not readable.
6154 none Vsyscalls don't work at all. This makes
6155 them quite hard to use for exploits but
6156 might break your system.
6158 vt.color= [VT] Default text color.
6159 Format: 0xYX, X = foreground, Y = background.
6160 Default: 0x07 = light gray on black.
6162 vt.cur_default= [VT] Default cursor shape.
6163 Format: 0xCCBBAA, where AA, BB, and CC are the same as
6164 the parameters of the <Esc>[?A;B;Cc escape sequence;
6165 see VGA-softcursor.txt. Default: 2 = underline.
6167 vt.default_blu= [VT]
6168 Format: <blue0>,<blue1>,<blue2>,...,<blue15>
6169 Change the default blue palette of the console.
6170 This is a 16-member array composed of values
6173 vt.default_grn= [VT]
6174 Format: <green0>,<green1>,<green2>,...,<green15>
6175 Change the default green palette of the console.
6176 This is a 16-member array composed of values
6179 vt.default_red= [VT]
6180 Format: <red0>,<red1>,<red2>,...,<red15>
6181 Change the default red palette of the console.
6182 This is a 16-member array composed of values
6188 Set system-wide default UTF-8 mode for all tty's.
6189 Default is 1, i.e. UTF-8 mode is enabled for all
6190 newly opened terminals.
6192 vt.global_cursor_default=
6195 Set system-wide default for whether a cursor
6196 is shown on new VTs. Default is -1,
6197 i.e. cursors will be created by default unless
6198 overridden by individual drivers. 0 will hide
6199 cursors, 1 will display them.
6201 vt.italic= [VT] Default color for italic text; 0-15.
6204 vt.underline= [VT] Default color for underlined text; 0-15.
6207 watchdog timers [HW,WDT] For information on watchdog timers,
6208 see Documentation/watchdog/watchdog-parameters.rst
6209 or other driver-specific files in the
6210 Documentation/watchdog/ directory.
6214 Set the hard lockup detector stall duration
6215 threshold in seconds. The soft lockup detector
6216 threshold is set to twice the value. A value of 0
6217 disables both lockup detectors. Default is 10
6220 workqueue.watchdog_thresh=
6221 If CONFIG_WQ_WATCHDOG is configured, workqueue can
6222 warn stall conditions and dump internal state to
6223 help debugging. 0 disables workqueue stall
6224 detection; otherwise, it's the stall threshold
6225 duration in seconds. The default value is 30 and
6226 it can be updated at runtime by writing to the
6227 corresponding sysfs file.
6229 workqueue.disable_numa
6230 By default, all work items queued to unbound
6231 workqueues are affine to the NUMA nodes they're
6232 issued on, which results in better behavior in
6233 general. If NUMA affinity needs to be disabled for
6234 whatever reason, this option can be used. Note
6235 that this also can be controlled per-workqueue for
6236 workqueues visible under /sys/bus/workqueue/.
6238 workqueue.power_efficient
6239 Per-cpu workqueues are generally preferred because
6240 they show better performance thanks to cache
6241 locality; unfortunately, per-cpu workqueues tend to
6242 be more power hungry than unbound workqueues.
6244 Enabling this makes the per-cpu workqueues which
6245 were observed to contribute significantly to power
6246 consumption unbound, leading to measurably lower
6247 power usage at the cost of small performance
6250 The default value of this parameter is determined by
6251 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT.
6253 workqueue.debug_force_rr_cpu
6254 Workqueue used to implicitly guarantee that work
6255 items queued without explicit CPU specified are put
6256 on the local CPU. This guarantee is no longer true
6257 and while local CPU is still preferred work items
6258 may be put on foreign CPUs. This debug option
6259 forces round-robin CPU selection to flush out
6260 usages which depend on the now broken guarantee.
6261 When enabled, memory and cache locality will be
6264 x2apic_phys [X86-64,APIC] Use x2apic physical mode instead of
6265 default x2apic cluster mode on platforms
6268 xen_512gb_limit [KNL,X86-64,XEN]
6269 Restricts the kernel running paravirtualized under Xen
6270 to use only up to 512 GB of RAM. The reason to do so is
6271 crash analysis tools and Xen tools for doing domain
6272 save/restore/migration must be enabled to handle larger
6275 xen_emul_unplug= [HW,X86,XEN]
6276 Unplug Xen emulated devices
6277 Format: [unplug0,][unplug1]
6278 ide-disks -- unplug primary master IDE devices
6279 aux-ide-disks -- unplug non-primary-master IDE devices
6280 nics -- unplug network devices
6281 all -- unplug all emulated devices (NICs and IDE disks)
6282 unnecessary -- unplugging emulated devices is
6283 unnecessary even if the host did not respond to
6285 never -- do not unplug even if version check succeeds
6287 xen_legacy_crash [X86,XEN]
6288 Crash from Xen panic notifier, without executing late
6289 panic() code such as dumping handler.
6291 xen_nopvspin [X86,XEN]
6292 Disables the qspinlock slowpath using Xen PV optimizations.
6293 This parameter is obsoleted by "nopvspin" parameter, which
6294 has equivalent effect for XEN platform.
6297 Disables the PV optimizations forcing the HVM guest to
6298 run as generic HVM guest with no PV drivers.
6299 This option is obsoleted by the "nopv" option, which
6300 has equivalent effect for XEN platform.
6302 xen_no_vector_callback
6303 [KNL,X86,XEN] Disable the vector callback for Xen
6304 event channel interrupts.
6306 xen_scrub_pages= [XEN]
6307 Boolean option to control scrubbing pages before giving them back
6308 to Xen, for use by other domains. Can be also changed at runtime
6309 with /sys/devices/system/xen_memory/xen_memory0/scrub_pages.
6310 Default value controlled with CONFIG_XEN_SCRUB_PAGES_DEFAULT.
6312 xen_timer_slop= [X86-64,XEN]
6313 Set the timer slop (in nanoseconds) for the virtual Xen
6314 timers (default is 100000). This adjusts the minimum
6315 delta of virtualized Xen timers, where lower values
6316 improve timer resolution at the expense of processing
6317 more timer interrupts.
6319 xen.event_eoi_delay= [XEN]
6320 How long to delay EOI handling in case of event
6321 storms (jiffies). Default is 10.
6323 xen.event_loop_timeout= [XEN]
6324 After which time (jiffies) the event handling loop
6325 should start to delay EOI handling. Default is 2.
6327 xen.fifo_events= [XEN]
6328 Boolean parameter to disable using fifo event handling
6329 even if available. Normally fifo event handling is
6330 preferred over the 2-level event handling, as it is
6331 fairer and the number of possible event channels is
6332 much higher. Default is on (use fifo events).
6334 nopv= [X86,XEN,KVM,HYPER_V,VMWARE]
6335 Disables the PV optimizations forcing the guest to run
6336 as generic guest with no PV drivers. Currently support
6337 XEN HVM, KVM, HYPER_V and VMWARE guest.
6339 nopvspin [X86,XEN,KVM]
6340 Disables the qspinlock slow path using PV optimizations
6341 which allow the hypervisor to 'idle' the guest on lock
6344 xirc2ps_cs= [NET,PCMCIA]
6346 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]]
6349 By default on POWER9 and above, the kernel will
6350 natively use the XIVE interrupt controller. This option
6351 allows the fallback firmware mode to be used:
6353 off Fallback to firmware control of XIVE interrupt
6354 controller on both pseries and powernv
6355 platforms. Only useful on POWER9 and above.
6357 xhci-hcd.quirks [USB,KNL]
6358 A hex value specifying bitmask with supplemental xhci
6359 host controller quirks. Meaning of each bit can be
6360 consulted in header drivers/usb/host/xhci.h.
6363 Format: { early | on | rw | ro | off }
6364 Controls if xmon debugger is enabled. Default is off.
6365 Passing only "xmon" is equivalent to "xmon=early".
6366 early Call xmon as early as possible on boot; xmon
6367 debugger is called from setup_arch().
6368 on xmon debugger hooks will be installed so xmon
6369 is only called on a kernel crash. Default mode,
6370 i.e. either "ro" or "rw" mode, is controlled
6371 with CONFIG_XMON_DEFAULT_RO_MODE.
6372 rw xmon debugger hooks will be installed so xmon
6373 is called only on a kernel crash, mode is write,
6374 meaning SPR registers, memory and, other data
6375 can be written using xmon commands.
6376 ro same as "rw" option above but SPR registers,
6377 memory, and other data can't be written using
6379 off xmon is disabled.