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
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
305 amd_iommu_dump= [HW,X86-64]
306 Enable AMD IOMMU driver option to dump the ACPI table
307 for AMD IOMMU. With this option enabled, AMD IOMMU
308 driver will print ACPI tables for AMD IOMMU during
309 IOMMU initialization.
311 amd_iommu_intr= [HW,X86-64]
312 Specifies one of the following AMD IOMMU interrupt
314 legacy - Use legacy interrupt remapping mode.
315 vapic - Use virtual APIC mode, which allows IOMMU
316 to inject interrupts directly into guest.
317 This mode requires kvm-amd.avic=1.
318 (Default when IOMMU HW support is present.)
320 amijoy.map= [HW,JOY] Amiga joystick support
321 Map of devices attached to JOY0DAT and JOY1DAT
323 See also Documentation/input/joydev/joystick.rst
325 analog.map= [HW,JOY] Analog joystick and gamepad support
326 Specifies type or capabilities of an analog joystick
327 connected to one of 16 gameports
328 Format: <type1>,<type2>,..<type16>
331 Power management functions (SPARCstation-4/5 + deriv.)
333 Disable APC CPU standby support. SPARCstation-Fox does
334 not play well with APC CPU idle - disable it if you have
335 APC and your system crashes randomly.
337 apic= [APIC,X86] Advanced Programmable Interrupt Controller
338 Change the output verbosity while booting
339 Format: { quiet (default) | verbose | debug }
340 Change the amount of debugging information output
341 when initialising the APIC and IO-APIC components.
342 For X86-32, this can also be used to specify an APIC
344 Format: apic=driver_name
345 Examples: apic=bigsmp
347 apic_extnmi= [APIC,X86] External NMI delivery setting
348 Format: { bsp (default) | all | none }
349 bsp: External NMI is delivered only to CPU 0
350 all: External NMIs are broadcast to all CPUs as a
352 none: External NMI is masked for all CPUs. This is
353 useful so that a dump capture kernel won't be
357 See Documentation/networking/ipv6.rst.
359 show_lapic= [APIC,X86] Advanced Programmable Interrupt Controller
360 Limit apic dumping. The parameter defines the maximal
361 number of local apics being dumped. Also it is possible
362 to set it to "all" by meaning -- no limit here.
363 Format: { 1 (default) | 2 | ... | all }.
364 The parameter valid if only apic=debug or
365 apic=verbose is specified.
366 Example: apic=debug show_lapic=all
368 apm= [APM] Advanced Power Management
369 See header of arch/x86/kernel/apm_32.c.
371 arcrimi= [HW,NET] ARCnet - "RIM I" (entirely mem-mapped) cards
372 Format: <io>,<irq>,<nodeID>
374 arm64.nobti [ARM64] Unconditionally disable Branch Target
375 Identification support
377 arm64.nopauth [ARM64] Unconditionally disable Pointer Authentication
382 atarimouse= [HW,MOUSE] Atari Mouse
384 atkbd.extra= [HW] Enable extra LEDs and keys on IBM RapidAccess,
385 EzKey and similar keyboards
387 atkbd.reset= [HW] Reset keyboard during initialization
389 atkbd.set= [HW] Select keyboard code set
390 Format: <int> (2 = AT (default), 3 = PS/2)
392 atkbd.scroll= [HW] Enable scroll wheel on MS Office and similar
395 atkbd.softraw= [HW] Choose between synthetic and real raw mode
396 Format: <bool> (0 = real, 1 = synthetic (default))
398 atkbd.softrepeat= [HW]
399 Use software keyboard repeat
401 audit= [KNL] Enable the audit sub-system
402 Format: { "0" | "1" | "off" | "on" }
403 0 | off - kernel audit is disabled and can not be
404 enabled until the next reboot
405 unset - kernel audit is initialized but disabled and
406 will be fully enabled by the userspace auditd.
407 1 | on - kernel audit is initialized and partially
408 enabled, storing at most audit_backlog_limit
409 messages in RAM until it is fully enabled by the
413 audit_backlog_limit= [KNL] Set the audit queue size limit.
414 Format: <int> (must be >=0)
417 bau= [X86_UV] Enable the BAU on SGI UV. The default
418 behavior is to disable the BAU (i.e. bau=0).
419 Format: { "0" | "1" }
422 unset - Disable the BAU.
424 baycom_epp= [HW,AX25]
427 baycom_par= [HW,AX25] BayCom Parallel Port AX.25 Modem
429 See header of drivers/net/hamradio/baycom_par.c.
431 baycom_ser_fdx= [HW,AX25]
432 BayCom Serial Port AX.25 Modem (Full Duplex Mode)
433 Format: <io>,<irq>,<mode>[,<baud>]
434 See header of drivers/net/hamradio/baycom_ser_fdx.c.
436 baycom_ser_hdx= [HW,AX25]
437 BayCom Serial Port AX.25 Modem (Half Duplex Mode)
438 Format: <io>,<irq>,<mode>
439 See header of drivers/net/hamradio/baycom_ser_hdx.c.
441 blkdevparts= Manual partition parsing of block device(s) for
442 embedded devices based on command line input.
443 See Documentation/block/cmdline-partition.rst
445 boot_delay= Milliseconds to delay each printk during boot.
446 Values larger than 10 seconds (10000) are changed to
451 Extended command line options can be added to an initrd
452 and this will cause the kernel to look for it.
454 See Documentation/admin-guide/bootconfig.rst
457 Disable BERT OS support on buggy BIOSes.
459 bgrt_disable [ACPI][X86]
460 Disable BGRT to avoid flickering OEM logo.
462 bttv.card= [HW,V4L] bttv (bt848 + bt878 based grabber cards)
463 bttv.radio= Most important insmod options are available as
465 bttv.pll= See Documentation/admin-guide/media/bttv.rst
468 bulk_remove=off [PPC] This parameter disables the use of the pSeries
469 firmware feature for flushing multiple hpte entries
472 c101= [NET] Moxa C101 synchronous serial card
474 cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection.
475 Sometimes CPU hardware bugs make them report the cache
476 size incorrectly. The kernel will attempt work arounds
477 to fix known problems, but for some CPUs it is not
478 possible to determine what the correct size should be.
479 This option provides an override for these situations.
482 [NET] Specifies amount of time (in seconds) that
483 the kernel should wait for a network carrier. By default
484 it waits 120 seconds.
486 ca_keys= [KEYS] This parameter identifies a specific key(s) on
487 the system trusted keyring to be used for certificate
489 format: { id:<keyid> | builtin }
491 cca= [MIPS] Override the kernel pages' cache coherency
492 algorithm. Accepted values range from 0 to 7
493 inclusive. See arch/mips/include/asm/pgtable-bits.h
494 for platform specific values (SB1, Loongson3 and
497 ccw_timeout_log [S390]
498 See Documentation/s390/common_io.rst for details.
500 cgroup_disable= [KNL] Disable a particular controller
501 Format: {name of the controller(s) to disable}
502 The effects of cgroup_disable=foo are:
503 - foo isn't auto-mounted if you mount all cgroups in
505 - foo isn't visible as an individually mountable
507 {Currently only "memory" controller deal with this and
508 cut the overhead, others just disable the usage. So
509 only cgroup_disable=memory is actually worthy}
511 cgroup_no_v1= [KNL] Disable cgroup controllers and named hierarchies in v1
512 Format: { { controller | "all" | "named" }
513 [,{ controller | "all" | "named" }...] }
514 Like cgroup_disable, but only applies to cgroup v1;
515 the blacklisted controllers remain available in cgroup2.
516 "all" blacklists all controllers and "named" disables
517 named mounts. Specifying both "all" and "named" disables
520 cgroup.memory= [KNL] Pass options to the cgroup memory controller.
522 nosocket -- Disable socket memory accounting.
523 nokmem -- Disable kernel memory accounting.
525 checkreqprot [SELINUX] Set initial checkreqprot flag value.
526 Format: { "0" | "1" }
527 See security/selinux/Kconfig help text.
528 0 -- check protection applied by kernel (includes
529 any implied execute protection).
530 1 -- check protection requested by application.
531 Default value is set via a kernel config option.
532 Value can be changed at runtime via
533 /sys/fs/selinux/checkreqprot.
534 Setting checkreqprot to 1 is deprecated.
537 See Documentation/s390/common_io.rst for details.
540 Prevents the clock framework from automatically gating
541 clocks that have not been explicitly enabled by a Linux
542 device driver but are enabled in hardware at reset or
543 by the bootloader/firmware. Note that this does not
544 force such clocks to be always-on nor does it reserve
545 those clocks in any way. This parameter is useful for
546 debug and development, but should not be needed on a
547 platform with proper driver support. For more
548 information, see Documentation/driver-api/clk.rst.
550 clock= [BUGS=X86-32, HW] gettimeofday clocksource override.
552 Forces specified clocksource (if available) to be used
553 when calculating gettimeofday(). If specified
554 clocksource is not available, it defaults to PIT.
555 Format: { pit | tsc | cyclone | pmtmr }
557 clocksource= Override the default clocksource
559 Override the default clocksource and use the clocksource
560 with the name specified.
561 Some clocksource names to choose from, depending on
563 [all] jiffies (this is the base, fallback clocksource)
565 [ARM] imx_timer1,OSTS,netx_timer,mpu_timer2,
566 pxa_timer,timer3,32k_counter,timer0_1
567 [X86-32] pit,hpet,tsc;
568 scx200_hrt on Geode; cyclone on IBM x440
576 clocksource.arm_arch_timer.evtstrm=
579 Enable/disable the eventstream feature of the ARM
580 architected timer so that code using WFE-based polling
581 loops can be debugged more effectively on production
584 clearcpuid=BITNUM[,BITNUM...] [X86]
585 Disable CPUID feature X for the kernel. See
586 arch/x86/include/asm/cpufeatures.h for the valid bit
587 numbers. Note the Linux specific bits are not necessarily
588 stable over kernel options, but the vendor specific
590 Also note that user programs calling CPUID directly
591 or using the feature without checking anything
592 will still see it. This just prevents it from
593 being used by the kernel or shown in /proc/cpuinfo.
594 Also note the kernel might malfunction if you disable
597 cma=nn[MG]@[start[MG][-end[MG]]]
599 Sets the size of kernel global memory area for
600 contiguous memory allocations and optionally the
601 placement constraint by the physical address range of
602 memory allocations. A value of 0 disables CMA
603 altogether. For more information, see
604 kernel/dma/contiguous.c
608 Sets the size of kernel per-numa memory area for
609 contiguous memory allocations. A value of 0 disables
610 per-numa CMA altogether. And If this option is not
611 specificed, the default value is 0.
612 With per-numa CMA enabled, DMA users on node nid will
613 first try to allocate buffer from the pernuma area
614 which is located in node nid, if the allocation fails,
615 they will fallback to the global default memory area.
617 cmo_free_hint= [PPC] Format: { yes | no }
618 Specify whether pages are marked as being inactive
619 when they are freed. This is used in CMO environments
620 to determine OS memory pressure for page stealing by
624 coherent_pool=nn[KMG] [ARM,KNL]
625 Sets the size of memory pool for coherent, atomic dma
626 allocations, by default set to 256K.
628 com20020= [HW,NET] ARCnet - COM20020 chipset
630 <io>[,<irq>[,<nodeID>[,<backplane>[,<ckp>[,<timeout>]]]]]
632 com90io= [HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers)
636 ARCnet - COM90xx chipset (memory-mapped buffers)
637 Format: <io>[,<irq>[,<memstart>]]
639 condev= [HW,S390] console device
642 console= [KNL] Output console device and options.
644 tty<n> Use the virtual console device <n>.
648 Use the specified serial port. The options are of
649 the form "bbbbpnf", where "bbbb" is the baud rate,
650 "p" is parity ("n", "o", or "e"), "n" is number of
651 bits, and "f" is flow control ("r" for RTS or
652 omit it). Default is "9600n8".
654 See Documentation/admin-guide/serial-console.rst for more
656 Documentation/networking/netconsole.rst for an
659 uart[8250],io,<addr>[,options]
660 uart[8250],mmio,<addr>[,options]
661 uart[8250],mmio16,<addr>[,options]
662 uart[8250],mmio32,<addr>[,options]
663 uart[8250],0x<addr>[,options]
664 Start an early, polled-mode console on the 8250/16550
665 UART at the specified I/O port or MMIO address,
666 switching to the matching ttyS device later.
667 MMIO inter-register address stride is either 8-bit
668 (mmio), 16-bit (mmio16), or 32-bit (mmio32).
669 If none of [io|mmio|mmio16|mmio32], <addr> is assumed
670 to be equivalent to 'mmio'. 'options' are specified in
671 the same format described for ttyS above; if unspecified,
672 the h/w is not re-initialized.
674 hvc<n> Use the hypervisor console device <n>. This is for
675 both Xen and PowerPC hypervisors.
677 If the device connected to the port is not a TTY but a braille
678 device, prepend "brl," before the device type, for instance
680 For now, only VisioBraille is supported.
683 [KNL] Change console messages format
685 By default we print messages on consoles in
686 "[time stamp] text\n" format (time stamp may not be
687 printed, depending on CONFIG_PRINTK_TIME or
688 `printk_time' param).
690 Switch to syslog format: "<%u>[time stamp] text\n"
691 IOW, each message will have a facility and loglevel
692 prefix. The format is similar to one used by syslog()
693 syscall, or to executing "dmesg -S --raw" or to reading
696 consoleblank= [KNL] The console blank (screen saver) timeout in
697 seconds. A value of 0 disables the blank timer.
701 [KNL] Change the default value for
702 /proc/<pid>/coredump_filter.
703 See also Documentation/filesystems/proc.rst.
705 coresight_cpu_debug.enable
708 Enable/disable the CPU sampling based debugging.
709 0: default value, disable debugging
710 1: enable debugging at boot time
712 cpuidle.off=1 [CPU_IDLE]
713 disable the cpuidle sub-system
716 [CPU_IDLE] Name of the cpuidle governor to use.
718 cpufreq.off=1 [CPU_FREQ]
719 disable the cpufreq sub-system
721 cpufreq.default_governor=
722 [CPU_FREQ] Name of the default cpufreq governor or
723 policy to use. This governor must be registered in the
724 kernel before the cpufreq driver probes.
727 [X86] Delay for N microsec between assert and de-assert
728 of APIC INIT to start processors. This delay occurs
729 on every CPU online, such as boot, and resume from suspend.
732 cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver
734 <first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>]
736 crashkernel=size[KMG][@offset[KMG]]
737 [KNL] Using kexec, Linux can switch to a 'crash kernel'
738 upon panic. This parameter reserves the physical
739 memory region [offset, offset + size] for that kernel
740 image. If '@offset' is omitted, then a suitable offset
741 is selected automatically.
742 [KNL, X86-64] Select a region under 4G first, and
743 fall back to reserve region above 4G when '@offset'
744 hasn't been specified.
745 See Documentation/admin-guide/kdump/kdump.rst for further details.
747 crashkernel=range1:size1[,range2:size2,...][@offset]
748 [KNL] Same as above, but depends on the memory
749 in the running system. The syntax of range is
750 start-[end] where start and end are both
751 a memory unit (amount[KMG]). See also
752 Documentation/admin-guide/kdump/kdump.rst for an example.
754 crashkernel=size[KMG],high
755 [KNL, X86-64] range could be above 4G. Allow kernel
756 to allocate physical memory region from top, so could
757 be above 4G if system have more than 4G ram installed.
758 Otherwise memory region will be allocated below 4G, if
760 It will be ignored if crashkernel=X is specified.
761 crashkernel=size[KMG],low
762 [KNL, X86-64] range under 4G. When crashkernel=X,high
763 is passed, kernel could allocate physical memory region
764 above 4G, that cause second kernel crash on system
765 that require some amount of low memory, e.g. swiotlb
766 requires at least 64M+32K low memory, also enough extra
767 low memory is needed to make sure DMA buffers for 32-bit
768 devices won't run out. Kernel would try to allocate at
769 at least 256M below 4G automatically.
770 This one let user to specify own low range under 4G
771 for second kernel instead.
772 0: to disable low allocation.
773 It will be ignored when crashkernel=X,high is not used
774 or memory reserved is below 4G.
777 [KNL] Disable crypto self-tests
782 cs89x0_media= [HW,NET]
783 Format: { rj45 | aui | bnc }
785 csdlock_debug= [KNL] Enable debug add-ons of cross-CPU function call
786 handling. When switched on, additional debug data is
787 printed to the console in case a hanging CPU is
788 detected, and that CPU is pinged again in order to try
789 to resolve the hang situation.
790 0: disable csdlock debugging (default)
791 1: enable basic csdlock debugging (minor impact)
792 ext: enable extended csdlock debugging (more impact,
796 See header of drivers/s390/block/dasd_devmap.c.
798 db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port
799 (one device per port)
800 Format: <port#>,<type>
801 See also Documentation/input/devices/joystick-parport.rst
803 ddebug_query= [KNL,DYNAMIC_DEBUG] Enable debug messages at early boot
805 Documentation/admin-guide/dynamic-debug-howto.rst for
806 details. Deprecated, see dyndbg.
808 debug [KNL] Enable kernel debugging (events log level).
811 [KNL] Enable printing [hashed] pointers early in the
812 boot sequence. If enabled, we use a weak hash instead
813 of siphash to hash pointers. Use this option if you are
814 seeing instances of '(___ptrval___)') and need to see a
815 value (hashed pointer) instead. Cryptographically
816 insecure, please do not use on production kernels.
819 [KNL] verbose locking self-tests
821 Print debugging info while doing the locking API
823 Bitmask for the various LOCKTYPE_ tests. Defaults to 0
824 (no extra messages), setting it to -1 (all bits set)
825 will print _a_lot_ more information - normally only
826 useful to lockdep developers.
828 debug_objects [KNL] Enable object debugging
831 [KNL] Disable object debugging
833 debug_guardpage_minorder=
834 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this
835 parameter allows control of the order of pages that will
836 be intentionally kept free (and hence protected) by the
837 buddy allocator. Bigger value increase the probability
838 of catching random memory corruption, but reduce the
839 amount of memory for normal system use. The maximum
840 possible value is MAX_ORDER/2. Setting this parameter
841 to 1 or 2 should be enough to identify most random
842 memory corruption problems caused by bugs in kernel or
843 driver code when a CPU writes to (or reads from) a
844 random memory location. Note that there exists a class
845 of memory corruptions problems caused by buggy H/W or
846 F/W or by drivers badly programing DMA (basically when
847 memory is written at bus level and the CPU MMU is
848 bypassed) which are not detectable by
849 CONFIG_DEBUG_PAGEALLOC, hence this option will not help
850 tracking down these problems.
853 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this parameter
854 enables the feature at boot time. By default, it is
855 disabled and the system will work mostly the same as a
856 kernel built without CONFIG_DEBUG_PAGEALLOC.
857 Note: to get most of debug_pagealloc error reports, it's
858 useful to also enable the page_owner functionality.
859 on: enable the feature
861 debugfs= [KNL] This parameter enables what is exposed to userspace
862 and debugfs internal clients.
863 Format: { on, no-mount, off }
864 on: All functions are enabled.
866 Filesystem is not registered but kernel clients can
867 access APIs and a crashkernel can be used to read
868 its content. There is nothing to mount.
869 off: Filesystem is not registered and clients
870 get a -EPERM as result when trying to register files
871 or directories within debugfs.
872 This is equivalent of the runtime functionality if
873 debugfs was not enabled in the kernel at all.
874 Default value is set in build-time with a kernel configuration.
876 debugpat [X86] Enable PAT debugging
878 decnet.addr= [HW,NET]
879 Format: <area>[,<node>]
880 See also Documentation/networking/decnet.rst.
883 [HW] The size of the default HugeTLB page. This is
884 the size represented by the legacy /proc/ hugepages
885 APIs. In addition, this is the default hugetlb size
886 used for shmget(), mmap() and mounting hugetlbfs
887 filesystems. If not specified, defaults to the
888 architecture's default huge page size. Huge page
889 sizes are architecture dependent. See also
890 Documentation/admin-guide/mm/hugetlbpage.rst.
893 deferred_probe_timeout=
894 [KNL] Debugging option to set a timeout in seconds for
895 deferred probe to give up waiting on dependencies to
896 probe. Only specific dependencies (subsystems or
897 drivers) that have opted in will be ignored. A timeout of 0
898 will timeout at the end of initcalls. This option will also
899 dump out devices still on the deferred probe list after
903 Format: { on | off | def_only | inf_only | always }
904 on: s390 zlib hardware support for compression on
905 level 1 and decompression (default)
906 off: No s390 zlib hardware support
907 def_only: s390 zlib hardware support for deflate
908 only (compression on level 1)
909 inf_only: s390 zlib hardware support for inflate
911 always: Same as 'on' but ignores the selected compression
912 level always using hardware support (used for debugging)
915 Set number of hash buckets for dentry cache.
917 disable_1tb_segments [PPC]
918 Disables the use of 1TB hash page table segments. This
919 causes the kernel to fall back to 256MB segments which
920 can be useful when debugging issues that require an SLB
924 Limits the number of kernel SLB entries, and flushes
925 them frequently to increase the rate of SLB faults
929 See Documentation/networking/ipv6.rst.
932 [KNL] Under CONFIG_HARDENED_USERCOPY, whether
933 hardening is enabled for this boot. Hardened
934 usercopy checking is used to protect the kernel
935 from reading or writing beyond known memory
936 allocation boundaries as a proactive defense
937 against bounds-checking flaws in the kernel's
938 copy_to_user()/copy_from_user() interface.
939 on Perform hardened usercopy checks (default).
940 off Disable hardened usercopy checks.
943 Disable RADIX MMU mode on POWER9
945 radix_hcall_invalidate=on [PPC/PSERIES]
946 Disable RADIX GTSE feature and use hcall for TLB
950 Disable TLBIE instruction. Currently does not work
951 with KVM, with HASH MMU, or with coherent accelerators.
953 disable_cpu_apicid= [X86,APIC,SMP]
955 The number of initial APIC ID for the
956 corresponding CPU to be disabled at boot,
957 mostly used for the kdump 2nd kernel to
958 disable BSP to wake up multiple CPUs without
959 causing system reset or hang due to sending
962 disable_ddw [PPC/PSERIES]
963 Disable Dynamic DMA Window support. Use this
964 to workaround buggy firmware.
967 See Documentation/networking/ipv6.rst.
969 disable_mtrr_cleanup [X86]
970 The kernel tries to adjust MTRR layout from continuous
971 to discrete, to make X server driver able to add WB
972 entry later. This parameter disables that.
974 disable_mtrr_trim [X86, Intel and AMD only]
975 By default the kernel will trim any uncacheable
976 memory out of your available memory pool based on
977 MTRR settings. This parameter disables that behavior,
978 possibly causing your machine to run very slowly.
980 disable_timer_pin_1 [X86]
981 Disable PIN 1 of APIC timer
982 Can be useful to work around chipset bugs.
984 dis_ucode_ldr [X86] Disable the microcode loader.
986 dma_debug=off If the kernel is compiled with DMA_API_DEBUG support,
987 this option disables the debugging code at boot.
989 dma_debug_entries=<number>
990 This option allows to tune the number of preallocated
991 entries for DMA-API debugging code. One entry is
992 required per DMA-API allocation. Use this if the
993 DMA-API debugging code disables itself because the
994 architectural default is too low.
996 dma_debug_driver=<driver_name>
997 With this option the DMA-API debugging driver
998 filter feature can be enabled at boot time. Just
999 pass the driver to filter for as the parameter.
1000 The filter can be disabled or changed to another
1001 driver later using sysfs.
1003 driver_async_probe= [KNL]
1004 List of driver names to be probed asynchronously.
1005 Format: <driver_name1>,<driver_name2>...
1007 drm.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>]
1008 Broken monitors, graphic adapters, KVMs and EDIDless
1009 panels may send no or incorrect EDID data sets.
1010 This parameter allows to specify an EDID data sets
1011 in the /lib/firmware directory that are used instead.
1012 Generic built-in EDID data sets are used, if one of
1013 edid/1024x768.bin, edid/1280x1024.bin,
1014 edid/1680x1050.bin, or edid/1920x1080.bin is given
1015 and no file with the same name exists. Details and
1016 instructions how to build your own EDID data are
1017 available in Documentation/admin-guide/edid.rst. An EDID
1018 data set will only be used for a particular connector,
1019 if its name and a colon are prepended to the EDID
1020 name. Each connector may use a unique EDID data
1021 set by separating the files with a comma. An EDID
1022 data set with no connector name will be used for
1023 any connectors not explicitly specified.
1028 Format: {"off" | "known"}
1029 Control how the dt_cpu_ftrs device-tree binding is
1030 used for CPU feature discovery and setup (if it
1032 off: Do not use it, fall back to legacy cpu table.
1033 known: Do not pass through unknown features to guests
1034 or userspace, only those that the kernel is aware of.
1036 dump_apple_properties [X86]
1037 Dump name and content of EFI device properties on
1038 x86 Macs. Useful for driver authors to determine
1039 what data is available or for reverse-engineering.
1041 dyndbg[="val"] [KNL,DYNAMIC_DEBUG]
1042 <module>.dyndbg[="val"]
1043 Enable debug messages at boot time. See
1044 Documentation/admin-guide/dynamic-debug-howto.rst
1047 nopku [X86] Disable Memory Protection Keys CPU feature found
1050 <module>.async_probe [KNL]
1051 Enable asynchronous probe on this module.
1053 early_ioremap_debug [KNL]
1054 Enable debug messages in early_ioremap support. This
1055 is useful for tracking down temporary early mappings
1056 which are not unmapped.
1058 earlycon= [KNL] Output early console device and options.
1060 When used with no options, the early console is
1061 determined by stdout-path property in device tree's
1062 chosen node or the ACPI SPCR table if supported by
1065 cdns,<addr>[,options]
1066 Start an early, polled-mode console on a Cadence
1067 (xuartps) serial port at the specified address. Only
1068 supported option is baud rate. If baud rate is not
1069 specified, the serial port must already be setup and
1072 uart[8250],io,<addr>[,options]
1073 uart[8250],mmio,<addr>[,options]
1074 uart[8250],mmio32,<addr>[,options]
1075 uart[8250],mmio32be,<addr>[,options]
1076 uart[8250],0x<addr>[,options]
1077 Start an early, polled-mode console on the 8250/16550
1078 UART at the specified I/O port or MMIO address.
1079 MMIO inter-register address stride is either 8-bit
1080 (mmio) or 32-bit (mmio32 or mmio32be).
1081 If none of [io|mmio|mmio32|mmio32be], <addr> is assumed
1082 to be equivalent to 'mmio'. 'options' are specified
1083 in the same format described for "console=ttyS<n>"; if
1084 unspecified, the h/w is not initialized.
1088 Start an early, polled-mode console on a pl011 serial
1089 port at the specified address. The pl011 serial port
1090 must already be setup and configured. Options are not
1091 yet supported. If 'mmio32' is specified, then only
1092 the driver will use only 32-bit accessors to read/write
1093 the device registers.
1096 Start an early, polled-mode console on a meson serial
1097 port at the specified address. The serial port must
1098 already be setup and configured. Options are not yet
1102 Start an early, polled-mode console on an msm serial
1103 port at the specified address. The serial port
1104 must already be setup and configured. Options are not
1107 msm_serial_dm,<addr>
1108 Start an early, polled-mode console on an msm serial
1109 dm port at the specified address. The serial port
1110 must already be setup and configured. Options are not
1114 Start an early, polled-mode console on a serial port
1115 of an Actions Semi SoC, such as S500 or S900, at the
1116 specified address. The serial port must already be
1117 setup and configured. Options are not yet supported.
1120 Start an early, polled-mode console on a serial port
1121 of an RDA Micro SoC, such as RDA8810PL, at the
1122 specified address. The serial port must already be
1123 setup and configured. Options are not yet supported.
1126 Use RISC-V SBI (Supervisor Binary Interface) for early
1129 smh Use ARM semihosting calls for early console.
1137 Use early console provided by serial driver available
1138 on Samsung SoCs, requires selecting proper type and
1139 a correct base address of the selected UART port. The
1140 serial port must already be setup and configured.
1141 Options are not yet supported.
1144 Start an early, polled-mode console on a lantiq serial
1145 (lqasc) port at the specified address. The serial port
1146 must already be setup and configured. Options are not
1151 Use early console provided by Freescale LP UART driver
1152 found on Freescale Vybrid and QorIQ LS1021A processors.
1153 A valid base address must be provided, and the serial
1154 port must already be setup and configured.
1158 Start an early, polled-mode, output-only console on the
1159 Freescale i.MX UART at the specified address. The UART
1160 must already be setup and configured.
1163 Start an early, polled-mode console on the
1164 Armada 3700 serial port at the specified
1165 address. The serial port must already be setup
1166 and configured. Options are not yet supported.
1169 Start an early, polled-mode console on a Qualcomm
1170 Generic Interface (GENI) based serial port at the
1171 specified address. The serial port must already be
1172 setup and configured. Options are not yet supported.
1175 Start an early, unaccelerated console on the EFI
1176 memory mapped framebuffer (if available). On cache
1177 coherent non-x86 systems that use system memory for
1178 the framebuffer, pass the 'ram' option so that it is
1179 mapped with the correct attributes.
1182 Use early console provided by Freescale LINFlexD UART
1183 serial driver for NXP S32V234 SoCs. A valid base
1184 address must be provided, and the serial port must
1185 already be setup and configured.
1187 earlyprintk= [X86,SH,ARM,M68k,S390]
1191 earlyprintk=serial[,ttySn[,baudrate]]
1192 earlyprintk=serial[,0x...[,baudrate]]
1193 earlyprintk=ttySn[,baudrate]
1194 earlyprintk=dbgp[debugController#]
1195 earlyprintk=pciserial[,force],bus:device.function[,baudrate]
1196 earlyprintk=xdbc[xhciController#]
1198 earlyprintk is useful when the kernel crashes before
1199 the normal console is initialized. It is not enabled by
1200 default because it has some cosmetic problems.
1202 Append ",keep" to not disable it when the real console
1205 Only one of vga, efi, serial, or usb debug port can
1208 Currently only ttyS0 and ttyS1 may be specified by
1209 name. Other I/O ports may be explicitly specified
1210 on some architectures (x86 and arm at least) by
1211 replacing ttySn with an I/O port address, like this:
1212 earlyprintk=serial,0x1008,115200
1213 You can find the port for a given device in
1214 /proc/tty/driver/serial:
1215 2: uart:ST16650V2 port:00001008 irq:18 ...
1217 Interaction with the standard serial driver is not
1220 The VGA and EFI output is eventually overwritten by
1223 The xen output can only be used by Xen PV guests.
1225 The sclp output can only be used on s390.
1227 The optional "force" to "pciserial" enables use of a
1228 PCI device even when its classcode is not of the
1231 edac_report= [HW,EDAC] Control how to report EDAC event
1232 Format: {"on" | "off" | "force"}
1233 on: enable EDAC to report H/W event. May be overridden
1234 by other higher priority error reporting module.
1235 off: disable H/W event reporting through EDAC.
1236 force: enforce the use of EDAC to report H/W event.
1239 ekgdboc= [X86,KGDB] Allow early kernel console debugging
1242 This is designed to be used in conjunction with
1243 the boot argument: earlyprintk=vga
1245 This parameter works in place of the kgdboc parameter
1246 but can only be used if the backing tty is available
1247 very early in the boot process. For early debugging
1248 via a serial port see kgdboc_earlycon instead.
1251 Format: {"off" | "on" | "skip[mbr]"}
1254 Format: { "debug", "disable_early_pci_dma",
1255 "nochunk", "noruntime", "nosoftreserve",
1256 "novamap", "no_disable_early_pci_dma" }
1257 debug: enable misc debug output.
1258 disable_early_pci_dma: disable the busmaster bit on all
1259 PCI bridges while in the EFI boot stub.
1260 nochunk: disable reading files in "chunks" in the EFI
1261 boot stub, as chunking can cause problems with some
1262 firmware implementations.
1263 noruntime : disable EFI runtime services support
1264 nosoftreserve: The EFI_MEMORY_SP (Specific Purpose)
1265 attribute may cause the kernel to reserve the
1266 memory range for a memory mapping driver to
1267 claim. Specify efi=nosoftreserve to disable this
1268 reservation and treat the memory by its base type
1269 (i.e. EFI_CONVENTIONAL_MEMORY / "System RAM").
1270 novamap: do not call SetVirtualAddressMap().
1271 no_disable_early_pci_dma: Leave the busmaster bit set
1272 on all PCI bridges while in the EFI boot stub
1274 efi_no_storage_paranoia [EFI; X86]
1275 Using this parameter you can use more than 50% of
1276 your efi variable storage. Use this parameter only if
1277 you are really sure that your UEFI does sane gc and
1278 fulfills the spec otherwise your board may brick.
1280 efi_fake_mem= nn[KMG]@ss[KMG]:aa[,nn[KMG]@ss[KMG]:aa,..] [EFI; X86]
1281 Add arbitrary attribute to specific memory range by
1282 updating original EFI memory map.
1283 Region of memory which aa attribute is added to is
1286 If efi_fake_mem=2G@4G:0x10000,2G@0x10a0000000:0x10000
1287 is specified, EFI_MEMORY_MORE_RELIABLE(0x10000)
1288 attribute is added to range 0x100000000-0x180000000 and
1289 0x10a0000000-0x1120000000.
1291 If efi_fake_mem=8G@9G:0x40000 is specified, the
1292 EFI_MEMORY_SP(0x40000) attribute is added to
1293 range 0x240000000-0x43fffffff.
1295 Using this parameter you can do debugging of EFI memmap
1296 related features. For example, you can do debugging of
1297 Address Range Mirroring feature even if your box
1298 doesn't support it, or mark specific memory as
1301 efivar_ssdt= [EFI; X86] Name of an EFI variable that contains an SSDT
1302 that is to be dynamically loaded by Linux. If there are
1303 multiple variables with the same name but with different
1304 vendor GUIDs, all of them will be loaded. See
1305 Documentation/admin-guide/acpi/ssdt-overlays.rst for details.
1308 eisa_irq_edge= [PARISC,HW]
1309 See header of drivers/parisc/eisa.c.
1312 See comment before function elanfreq_setup() in
1313 arch/x86/kernel/cpu/cpufreq/elanfreq.c.
1315 elfcorehdr=[size[KMG]@]offset[KMG] [IA64,PPC,SH,X86,S390]
1316 Specifies physical address of start of kernel core
1317 image elf header and optionally the size. Generally
1318 kexec loader will pass this option to capture kernel.
1319 See Documentation/admin-guide/kdump/kdump.rst for details.
1321 enable_mtrr_cleanup [X86]
1322 The kernel tries to adjust MTRR layout from continuous
1323 to discrete, to make X server driver able to add WB
1324 entry later. This parameter enables that.
1326 enable_timer_pin_1 [X86]
1327 Enable PIN 1 of APIC timer
1328 Can be useful to work around chipset bugs
1329 (in particular on some ATI chipsets).
1330 The kernel tries to set a reasonable default.
1332 enforcing [SELINUX] Set initial enforcing status.
1334 See security/selinux/Kconfig help text.
1335 0 -- permissive (log only, no denials).
1336 1 -- enforcing (deny and log).
1338 Value can be changed at runtime via
1339 /sys/fs/selinux/enforce.
1342 Disable Error Record Serialization Table (ERST)
1345 ether= [HW,NET] Ethernet cards parameters
1346 This option is obsoleted by the "netdev=" option, which
1347 has equivalent usage. See its documentation for details.
1351 Permit 'security.evm' to be updated regardless of
1352 current integrity status.
1357 fail_make_request=[KNL]
1358 General fault injection mechanism.
1359 Format: <interval>,<probability>,<space>,<times>
1360 See also Documentation/fault-injection/.
1363 Format: { initns | none }
1364 See Documentation/admin-guide/sysctl/net.rst for
1365 fb_tunnels_only_for_init_ns
1368 See Documentation/admin-guide/blockdev/floppy.rst.
1370 force_pal_cache_flush
1371 [IA-64] Avoid check_sal_cache_flush which may hang on
1372 buggy SAL_CACHE_FLUSH implementations. Using this
1373 parameter will force ia64_sal_cache_flush to call
1374 ia64_pal_cache_flush instead of SAL_CACHE_FLUSH.
1377 Forcefully enable Physical Address Extension (PAE).
1378 Many Pentium M systems disable PAE but may have a
1379 functionally usable PAE implementation.
1380 Warning: use of this parameter will taint the kernel
1381 and may cause unknown problems.
1384 [FTRACE] will set and start the specified tracer
1385 as early as possible in order to facilitate early
1388 ftrace_dump_on_oops[=orig_cpu]
1389 [FTRACE] will dump the trace buffers on oops.
1390 If no parameter is passed, ftrace will dump
1391 buffers of all CPUs, but if you pass orig_cpu, it will
1392 dump only the buffer of the CPU that triggered the
1395 ftrace_filter=[function-list]
1396 [FTRACE] Limit the functions traced by the function
1397 tracer at boot up. function-list is a comma-separated
1398 list of functions. This list can be changed at run
1399 time by the set_ftrace_filter file in the debugfs
1402 ftrace_notrace=[function-list]
1403 [FTRACE] Do not trace the functions specified in
1404 function-list. This list can be changed at run time
1405 by the set_ftrace_notrace file in the debugfs
1408 ftrace_graph_filter=[function-list]
1409 [FTRACE] Limit the top level callers functions traced
1410 by the function graph tracer at boot up.
1411 function-list is a comma-separated list of functions
1412 that can be changed at run time by the
1413 set_graph_function file in the debugfs tracing directory.
1415 ftrace_graph_notrace=[function-list]
1416 [FTRACE] Do not trace from the functions specified in
1417 function-list. This list is a comma-separated list of
1418 functions that can be changed at run time by the
1419 set_graph_notrace file in the debugfs tracing directory.
1421 ftrace_graph_max_depth=<uint>
1422 [FTRACE] Used with the function graph tracer. This is
1423 the max depth it will trace into a function. This value
1424 can be changed at run time by the max_graph_depth file
1425 in the tracefs tracing directory. default: 0 (no limit)
1427 fw_devlink= [KNL] Create device links between consumer and supplier
1428 devices by scanning the firmware to infer the
1429 consumer/supplier relationships. This feature is
1430 especially useful when drivers are loaded as modules as
1431 it ensures proper ordering of tasks like device probing
1432 (suppliers first, then consumers), supplier boot state
1433 clean up (only after all consumers have probed),
1434 suspend/resume & runtime PM (consumers first, then
1436 Format: { off | permissive | on | rpm }
1437 off -- Don't create device links from firmware info.
1438 permissive -- Create device links from firmware info
1439 but use it only for ordering boot state clean
1440 up (sync_state() calls).
1441 on -- Create device links from firmware info and use it
1442 to enforce probe and suspend/resume ordering.
1443 rpm -- Like "on", but also use to order runtime PM.
1445 fw_devlink.strict=<bool>
1446 [KNL] Treat all inferred dependencies as mandatory
1447 dependencies. This only applies for fw_devlink=on|rpm.
1451 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad
1452 support via parallel port (up to 5 devices per port)
1453 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5>
1454 See also Documentation/input/devices/joystick-parport.rst
1458 gart_fix_e820= [X86-64] disable the fix e820 for K8 GART
1462 gcov_persist= [GCOV] When non-zero (default), profiling data for
1463 kernel modules is saved and remains accessible via
1464 debugfs, even when the module is unloaded/reloaded.
1465 When zero, profiling data is discarded and associated
1466 debugfs files are removed at module unload time.
1468 goldfish [X86] Enable the goldfish android emulator platform.
1469 Don't use this when you are not running on the
1472 gpio-mockup.gpio_mockup_ranges
1473 [HW] Sets the ranges of gpiochip of for this device.
1474 Format: <start1>,<end1>,<start2>,<end2>...
1475 gpio-mockup.gpio_mockup_named_lines
1476 [HW] Let the driver know GPIO lines should be named.
1478 gpt [EFI] Forces disk with valid GPT signature but
1479 invalid Protective MBR to be treated as GPT. If the
1480 primary GPT is corrupted, it enables the backup/alternate
1481 GPT to be used instead.
1483 grcan.enable0= [HW] Configuration of physical interface 0. Determines
1484 the "Enable 0" bit of the configuration register.
1487 grcan.enable1= [HW] Configuration of physical interface 1. Determines
1488 the "Enable 0" bit of the configuration register.
1491 grcan.select= [HW] Select which physical interface to use.
1494 grcan.txsize= [HW] Sets the size of the tx buffer.
1495 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0.
1497 grcan.rxsize= [HW] Sets the size of the rx buffer.
1498 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0.
1501 hardlockup_all_cpu_backtrace=
1502 [KNL] Should the hard-lockup detector generate
1503 backtraces on all cpus.
1506 hashdist= [KNL,NUMA] Large hashes allocated during boot
1507 are distributed across NUMA nodes. Defaults on
1508 for 64-bit NUMA, off otherwise.
1509 Format: 0 | 1 (for off | on)
1511 hcl= [IA-64] SGI's Hardware Graph compatibility layer
1513 hd= [EIDE] (E)IDE hard drive subsystem geometry
1514 Format: <cyl>,<head>,<sect>
1517 Disable Hardware Error Source Table (HEST) support;
1518 corresponding firmware-first mode error processing
1519 logic will be disabled.
1521 highmem=nn[KMG] [KNL,BOOT] forces the highmem zone to have an exact
1522 size of <nn>. This works even on boxes that have no
1523 highmem otherwise. This also works to reduce highmem
1524 size on bigger boxes.
1526 highres= [KNL] Enable/disable high resolution timer mode.
1527 Valid parameters: "on", "off"
1532 hpet= [X86-32,HPET] option to control HPET usage
1533 Format: { enable (default) | disable | force |
1535 disable: disable HPET and use PIT instead
1536 force: allow force enabled of undocumented chips (ICH4,
1538 verbose: show contents of HPET registers during setup
1540 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET
1541 registers. Default set by CONFIG_HPET_MMAP_DEFAULT.
1543 hugetlb_cma= [HW,CMA] The size of a CMA area used for allocation
1544 of gigantic hugepages.
1547 Reserve a CMA area of given size and allocate gigantic
1548 hugepages using the CMA allocator. If enabled, the
1549 boot-time allocation of gigantic hugepages is skipped.
1551 hugepages= [HW] Number of HugeTLB pages to allocate at boot.
1552 If this follows hugepagesz (below), it specifies
1553 the number of pages of hugepagesz to be allocated.
1554 If this is the first HugeTLB parameter on the command
1555 line, it specifies the number of pages to allocate for
1556 the default huge page size. See also
1557 Documentation/admin-guide/mm/hugetlbpage.rst.
1561 [HW] The size of the HugeTLB pages. This is used in
1562 conjunction with hugepages (above) to allocate huge
1563 pages of a specific size at boot. The pair
1564 hugepagesz=X hugepages=Y can be specified once for
1565 each supported huge page size. Huge page sizes are
1566 architecture dependent. See also
1567 Documentation/admin-guide/mm/hugetlbpage.rst.
1571 [KNL] Should the hung task detector generate panics.
1574 A value of 1 instructs the kernel to panic when a
1575 hung task is detected. The default value is controlled
1576 by the CONFIG_BOOTPARAM_HUNG_TASK_PANIC build-time
1577 option. The value selected by this boot parameter can
1578 be changed later by the kernel.hung_task_panic sysctl.
1580 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC)
1581 terminal devices. Valid values: 0..8
1582 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs.
1583 If specified, z/VM IUCV HVC accepts connections
1584 from listed z/VM user IDs only.
1586 hv_nopvspin [X86,HYPER_V] Disables the paravirt spinlock optimizations
1587 which allow the hypervisor to 'idle' the
1588 guest on lock contention.
1591 Do not unregister boot console at start. This is only
1592 useful for debugging when something happens in the window
1593 between unregistering the boot console and initializing
1596 i2c_bus= [HW] Override the default board specific I2C bus speed
1597 or register an additional I2C bus that is not
1598 registered from board initialization code.
1602 i8042.debug [HW] Toggle i8042 debug mode
1603 i8042.unmask_kbd_data
1604 [HW] Enable printing of interrupt data from the KBD port
1605 (disabled by default, and as a pre-condition
1606 requires that i8042.debug=1 be enabled)
1607 i8042.direct [HW] Put keyboard port into non-translated mode
1608 i8042.dumbkbd [HW] Pretend that controller can only read data from
1609 keyboard and cannot control its state
1610 (Don't attempt to blink the leds)
1611 i8042.noaux [HW] Don't check for auxiliary (== mouse) port
1612 i8042.nokbd [HW] Don't check/create keyboard port
1613 i8042.noloop [HW] Disable the AUX Loopback command while probing
1615 i8042.nomux [HW] Don't check presence of an active multiplexing
1617 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX
1619 i8042.notimeout [HW] Ignore timeout condition signalled by controller
1620 i8042.reset [HW] Reset the controller during init, cleanup and
1621 suspend-to-ram transitions, only during s2r
1622 transitions, or never reset
1623 Format: { 1 | Y | y | 0 | N | n }
1624 1, Y, y: always reset controller
1625 0, N, n: don't ever reset controller
1626 Default: only on s2r transitions on x86; most other
1627 architectures force reset to be always executed
1628 i8042.unlock [HW] Unlock (ignore) the keylock
1629 i8042.kbdreset [HW] Reset device connected to KBD port
1633 i8k.ignore_dmi [HW] Continue probing hardware even if DMI data
1634 indicates that the driver is running on unsupported
1636 i8k.force [HW] Activate i8k driver even if SMM BIOS signature
1637 does not match list of supported models.
1639 [HW] Report power status in /proc/i8k
1640 (disabled by default)
1641 i8k.restricted [HW] Allow controlling fans only if SYS_ADMIN
1644 i915.invert_brightness=
1645 [DRM] Invert the sense of the variable that is used to
1646 set the brightness of the panel backlight. Normally a
1647 brightness value of 0 indicates backlight switched off,
1648 and the maximum of the brightness value sets the backlight
1649 to maximum brightness. If this parameter is set to 0
1650 (default) and the machine requires it, or this parameter
1651 is set to 1, a brightness value of 0 sets the backlight
1652 to maximum brightness, and the maximum of the brightness
1653 value switches the backlight off.
1654 -1 -- never invert brightness
1655 0 -- machine default
1656 1 -- force brightness inversion
1659 Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]]
1661 ide-core.nodma= [HW] (E)IDE subsystem
1662 Format: =0.0 to prevent dma on hda, =0.1 hdb =1.0 hdc
1663 .vlb_clock .pci_clock .noflush .nohpa .noprobe .nowerr
1664 .cdrom .chs .ignore_cable are additional options
1665 See Documentation/ide/ide.rst.
1667 ide-generic.probe-mask= [HW] (E)IDE subsystem
1669 Probe mask for legacy ISA IDE ports. Depending on
1670 platform up to 6 ports are supported, enabled by
1671 setting corresponding bits in the mask to 1. The
1672 default value is 0x0, which has a special meaning.
1673 On systems that have PCI, it triggers scanning the
1674 PCI bus for the first and the second port, which
1675 are then probed. On systems without PCI the value
1676 of 0x0 enables probing the two first ports as if it
1679 ide-pci-generic.all-generic-ide [HW] (E)IDE subsystem
1680 Claim all unknown PCI IDE storage controllers.
1683 Format: idle=poll, idle=halt, idle=nomwait
1684 Poll forces a polling idle loop that can slightly
1685 improve the performance of waking up a idle CPU, but
1686 will use a lot of power and make the system run hot.
1688 idle=halt: Halt is forced to be used for CPU idle.
1689 In such case C2/C3 won't be used again.
1690 idle=nomwait: Disable mwait for CPU C-states
1694 Allow force disabling of Shared Virtual Memory (SVA)
1695 support for the idxd driver. By default it is set to
1698 ieee754= [MIPS] Select IEEE Std 754 conformance mode
1699 Format: { strict | legacy | 2008 | relaxed }
1702 Choose which programs will be accepted for execution
1703 based on the IEEE 754 NaN encoding(s) supported by
1704 the FPU and the NaN encoding requested with the value
1705 of an ELF file header flag individually set by each
1706 binary. Hardware implementations are permitted to
1707 support either or both of the legacy and the 2008 NaN
1710 Available settings are as follows:
1711 strict accept binaries that request a NaN encoding
1712 supported by the FPU
1713 legacy only accept legacy-NaN binaries, if supported
1715 2008 only accept 2008-NaN binaries, if supported
1717 relaxed accept any binaries regardless of whether
1718 supported by the FPU
1720 The FPU emulator is always able to support both NaN
1721 encodings, so if no FPU hardware is present or it has
1722 been disabled with 'nofpu', then the settings of
1723 'legacy' and '2008' strap the emulator accordingly,
1724 'relaxed' straps the emulator for both legacy-NaN and
1725 2008-NaN, whereas 'strict' enables legacy-NaN only on
1726 legacy processors and both NaN encodings on MIPS32 or
1729 The setting for ABS.fmt/NEG.fmt instruction execution
1730 mode generally follows that for the NaN encoding,
1731 except where unsupported by hardware.
1733 ignore_loglevel [KNL]
1734 Ignore loglevel setting - this will print /all/
1735 kernel messages to the console. Useful for debugging.
1736 We also add it as printk module parameter, so users
1737 could change it dynamically, usually by
1738 /sys/module/printk/parameters/ignore_loglevel.
1741 Ignore RLIMIT_DATA setting for data mappings,
1742 print warning at first misuse. Can be changed via
1743 /sys/module/kernel/parameters/ignore_rlimit_data.
1745 ihash_entries= [KNL]
1746 Set number of hash buckets for inode cache.
1748 ima_appraise= [IMA] appraise integrity measurements
1749 Format: { "off" | "enforce" | "fix" | "log" }
1752 ima_appraise_tcb [IMA] Deprecated. Use ima_policy= instead.
1753 The builtin appraise policy appraises all files
1756 ima_canonical_fmt [IMA]
1757 Use the canonical format for the binary runtime
1758 measurements, instead of host native format.
1761 Format: { md5 | sha1 | rmd160 | sha256 | sha384
1765 The list of supported hash algorithms is defined
1766 in crypto/hash_info.h.
1769 The builtin policies to load during IMA setup.
1770 Format: "tcb | appraise_tcb | secure_boot |
1771 fail_securely | critical_data"
1773 The "tcb" policy measures all programs exec'd, files
1774 mmap'd for exec, and all files opened with the read
1775 mode bit set by either the effective uid (euid=0) or
1778 The "appraise_tcb" policy appraises the integrity of
1779 all files owned by root.
1781 The "secure_boot" policy appraises the integrity
1782 of files (eg. kexec kernel image, kernel modules,
1783 firmware, policy, etc) based on file signatures.
1785 The "fail_securely" policy forces file signature
1786 verification failure also on privileged mounted
1787 filesystems with the SB_I_UNVERIFIABLE_SIGNATURE
1790 The "critical_data" policy measures kernel integrity
1793 ima_tcb [IMA] Deprecated. Use ima_policy= instead.
1794 Load a policy which meets the needs of the Trusted
1795 Computing Base. This means IMA will measure all
1796 programs exec'd, files mmap'd for exec, and all files
1797 opened for read by uid=0.
1800 Select one of defined IMA measurements template formats.
1801 Formats: { "ima" | "ima-ng" | "ima-sig" }
1805 [IMA] Define a custom template format.
1806 Format: { "field1|...|fieldN" }
1808 ima.ahash_minsize= [IMA] Minimum file size for asynchronous hash usage
1809 Format: <min_file_size>
1810 Set the minimal file size for using asynchronous hash.
1811 If left unspecified, ahash usage is disabled.
1813 ahash performance varies for different data sizes on
1814 different crypto accelerators. This option can be used
1815 to achieve the best performance for a particular HW.
1817 ima.ahash_bufsize= [IMA] Asynchronous hash buffer size
1819 Set hashing buffer size. Default: 4k.
1821 ahash performance varies for different chunk sizes on
1822 different crypto accelerators. This option can be used
1823 to achieve best performance for particular HW.
1827 Run specified binary instead of /sbin/init as init
1830 initcall_debug [KNL] Trace initcalls as they are executed. Useful
1831 for working out where the kernel is dying during
1834 initcall_blacklist= [KNL] Do not execute a comma-separated list of
1835 initcall functions. Useful for debugging built-in
1836 modules and initcalls.
1838 initrd= [BOOT] Specify the location of the initial ramdisk
1840 initrdmem= [KNL] Specify a physical address and size from which to
1841 load the initrd. If an initrd is compiled in or
1842 specified in the bootparams, it takes priority over this
1844 Format: ss[KMG],nn[KMG]
1847 init_on_alloc= [MM] Fill newly allocated pages and heap objects with
1850 Default set by CONFIG_INIT_ON_ALLOC_DEFAULT_ON.
1852 init_on_free= [MM] Fill freed pages and heap objects with zeroes.
1854 Default set by CONFIG_INIT_ON_FREE_DEFAULT_ON.
1856 init_pkru= [X86] Specify the default memory protection keys rights
1857 register contents for all processes. 0x55555554 by
1858 default (disallow access to all but pkey 0). Can
1859 override in debugfs after boot.
1861 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver
1864 int_pln_enable [X86] Enable power limit notification interrupt
1866 integrity_audit=[IMA]
1867 Format: { "0" | "1" }
1868 0 -- basic integrity auditing messages. (Default)
1869 1 -- additional integrity auditing messages.
1871 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option
1873 Enable intel iommu driver.
1875 Disable intel iommu driver.
1876 igfx_off [Default Off]
1877 By default, gfx is mapped as normal device. If a gfx
1878 device has a dedicated DMAR unit, the DMAR unit is
1879 bypassed by not enabling DMAR with this option. In
1880 this case, gfx device will use physical address for
1882 strict [Default Off]
1883 With this option on every unmap_single operation will
1884 result in a hardware IOTLB flush operation as opposed
1885 to batching them for performance.
1886 sp_off [Default Off]
1887 By default, super page will be supported if Intel IOMMU
1888 has the capability. With this option, super page will
1891 By default, scalable mode will be disabled even if the
1892 hardware advertises that it has support for the scalable
1893 mode translation. With this option set, scalable mode
1894 will be used on hardware which claims to support it.
1895 tboot_noforce [Default Off]
1896 Do not force the Intel IOMMU enabled under tboot.
1897 By default, tboot will force Intel IOMMU on, which
1898 could harm performance of some high-throughput
1899 devices like 40GBit network cards, even if identity
1901 Note that using this option lowers the security
1902 provided by tboot because it makes the system
1903 vulnerable to DMA attacks.
1905 intel_idle.max_cstate= [KNL,HW,ACPI,X86]
1906 0 disables intel_idle and fall back on acpi_idle.
1907 1 to 9 specify maximum depth of C-state.
1911 Do not enable intel_pstate as the default
1912 scaling driver for the supported processors
1914 Use intel_pstate as a scaling driver, but configure it
1915 to work with generic cpufreq governors (instead of
1916 enabling its internal governor). This mode cannot be
1917 used along with the hardware-managed P-states (HWP)
1920 Enable intel_pstate on systems that prohibit it by default
1921 in favor of acpi-cpufreq. Forcing the intel_pstate driver
1922 instead of acpi-cpufreq may disable platform features, such
1923 as thermal controls and power capping, that rely on ACPI
1924 P-States information being indicated to OSPM and therefore
1925 should be used with caution. This option does not work with
1926 processors that aren't supported by the intel_pstate driver
1927 or on platforms that use pcc-cpufreq instead of acpi-cpufreq.
1929 Do not enable hardware P state control (HWP)
1932 Only load intel_pstate on systems which support
1933 hardware P state control (HWP) if available.
1935 Enforce ACPI _PPC performance limits. If the Fixed ACPI
1936 Description Table, specifies preferred power management
1937 profile as "Enterprise Server" or "Performance Server",
1938 then this feature is turned on by default.
1940 Allow per-logical-CPU P-State performance control limits using
1941 cpufreq sysfs interface
1943 intremap= [X86-64, Intel-IOMMU]
1944 on enable Interrupt Remapping (default)
1945 off disable Interrupt Remapping
1946 nosid disable Source ID checking
1948 BIOS x2APIC opt-out request will be ignored
1949 nopost disable Interrupt Posting
1951 iomem= Disable strict checking of access to MMIO memory
1952 strict regions from userspace.
1967 nobypass [PPC/POWERNV]
1968 Disable IOMMU bypass, using IOMMU for PCI devices.
1970 iommu.forcedac= [ARM64, X86] Control IOVA allocation for PCI devices.
1971 Format: { "0" | "1" }
1972 0 - Try to allocate a 32-bit DMA address first, before
1973 falling back to the full range if needed.
1974 1 - Allocate directly from the full usable range,
1975 forcing Dual Address Cycle for PCI cards supporting
1976 greater than 32-bit addressing.
1978 iommu.strict= [ARM64] Configure TLB invalidation behaviour
1979 Format: { "0" | "1" }
1981 Request that DMA unmap operations use deferred
1982 invalidation of hardware TLBs, for increased
1983 throughput at the cost of reduced device isolation.
1984 Will fall back to strict mode if not supported by
1985 the relevant IOMMU driver.
1986 1 - Strict mode (default).
1987 DMA unmap operations invalidate IOMMU hardware TLBs
1991 [ARM64, X86] Configure DMA to bypass the IOMMU by default.
1992 Format: { "0" | "1" }
1993 0 - Use IOMMU translation for DMA.
1994 1 - Bypass the IOMMU for DMA.
1995 unset - Use value of CONFIG_IOMMU_DEFAULT_PASSTHROUGH.
1997 io7= [HW] IO7 for Marvel-based Alpha systems
1998 See comment before marvel_specify_io7 in
1999 arch/alpha/kernel/core_marvel.c.
2001 io_delay= [X86] I/O delay method
2003 Standard port 0x80 based delay
2005 Alternate port 0xed based delay (needed on some systems)
2007 Simple two microseconds delay
2012 See Documentation/admin-guide/nfs/nfsroot.rst.
2014 ipcmni_extend [KNL] Extend the maximum number of unique System V
2015 IPC identifiers from 32,768 to 16,777,216.
2017 irqaffinity= [SMP] Set the default irq affinity mask
2018 The argument is a cpu list, as described above.
2020 irqchip.gicv2_force_probe=
2023 Force the kernel to look for the second 4kB page
2024 of a GICv2 controller even if the memory range
2025 exposed by the device tree is too small.
2027 irqchip.gicv3_nolpi=
2029 Force the kernel to ignore the availability of
2030 LPIs (and by consequence ITSs). Intended for system
2031 that use the kernel as a bootloader, and thus want
2032 to let secondary kernels in charge of setting up
2035 irqchip.gicv3_pseudo_nmi= [ARM64]
2036 Enables support for pseudo-NMIs in the kernel. This
2037 requires the kernel to be built with
2038 CONFIG_ARM64_PSEUDO_NMI.
2041 When an interrupt is not handled search all handlers
2042 for it. Intended to get systems with badly broken
2046 When an interrupt is not handled search all handlers
2047 for it. Also check all handlers each timer
2048 interrupt. Intended to get systems with badly broken
2052 Format: <RDP>,<reset>,<pci_scan>,<verbosity>
2054 isolcpus= [KNL,SMP,ISOL] Isolate a given set of CPUs from disturbance.
2055 [Deprecated - use cpusets instead]
2056 Format: [flag-list,]<cpu-list>
2058 Specify one or more CPUs to isolate from disturbances
2059 specified in the flag list (default: domain):
2062 Disable the tick when a single task runs.
2064 A residual 1Hz tick is offloaded to workqueues, which you
2065 need to affine to housekeeping through the global
2066 workqueue's affinity configured via the
2067 /sys/devices/virtual/workqueue/cpumask sysfs file, or
2068 by using the 'domain' flag described below.
2070 NOTE: by default the global workqueue runs on all CPUs,
2071 so to protect individual CPUs the 'cpumask' file has to
2072 be configured manually after bootup.
2075 Isolate from the general SMP balancing and scheduling
2076 algorithms. Note that performing domain isolation this way
2077 is irreversible: it's not possible to bring back a CPU to
2078 the domains once isolated through isolcpus. It's strongly
2079 advised to use cpusets instead to disable scheduler load
2080 balancing through the "cpuset.sched_load_balance" file.
2081 It offers a much more flexible interface where CPUs can
2082 move in and out of an isolated set anytime.
2084 You can move a process onto or off an "isolated" CPU via
2085 the CPU affinity syscalls or cpuset.
2086 <cpu number> begins at 0 and the maximum value is
2087 "number of CPUs in system - 1".
2091 Isolate from being targeted by managed interrupts
2092 which have an interrupt mask containing isolated
2093 CPUs. The affinity of managed interrupts is
2094 handled by the kernel and cannot be changed via
2095 the /proc/irq/* interfaces.
2097 This isolation is best effort and only effective
2098 if the automatically assigned interrupt mask of a
2099 device queue contains isolated and housekeeping
2100 CPUs. If housekeeping CPUs are online then such
2101 interrupts are directed to the housekeeping CPU
2102 so that IO submitted on the housekeeping CPU
2103 cannot disturb the isolated CPU.
2105 If a queue's affinity mask contains only isolated
2106 CPUs then this parameter has no effect on the
2107 interrupt routing decision, though interrupts are
2108 only delivered when tasks running on those
2109 isolated CPUs submit IO. IO submitted on
2110 housekeeping CPUs has no influence on those
2113 The format of <cpu-list> is described above.
2117 ivrs_ioapic [HW,X86-64]
2118 Provide an override to the IOAPIC-ID<->DEVICE-ID
2119 mapping provided in the IVRS ACPI table. For
2120 example, to map IOAPIC-ID decimal 10 to
2121 PCI device 00:14.0 write the parameter as:
2122 ivrs_ioapic[10]=00:14.0
2124 ivrs_hpet [HW,X86-64]
2125 Provide an override to the HPET-ID<->DEVICE-ID
2126 mapping provided in the IVRS ACPI table. For
2127 example, to map HPET-ID decimal 0 to
2128 PCI device 00:14.0 write the parameter as:
2129 ivrs_hpet[0]=00:14.0
2131 ivrs_acpihid [HW,X86-64]
2132 Provide an override to the ACPI-HID:UID<->DEVICE-ID
2133 mapping provided in the IVRS ACPI table. For
2134 example, to map UART-HID:UID AMD0020:0 to
2135 PCI device 00:14.5 write the parameter as:
2136 ivrs_acpihid[00:14.5]=AMD0020:0
2138 js= [HW,JOY] Analog joystick
2139 See Documentation/input/joydev/joystick.rst.
2142 When CONFIG_RANDOMIZE_BASE is set, this disables
2143 kernel and module base offset ASLR (Address Space
2144 Layout Randomization).
2147 [KNL] Enforce KASAN (Kernel Address Sanitizer) to print
2148 report on every invalid memory access. Without this
2149 parameter KASAN will print report only for the first
2154 kernelcore= [KNL,X86,IA-64,PPC]
2155 Format: nn[KMGTPE] | nn% | "mirror"
2156 This parameter specifies the amount of memory usable by
2157 the kernel for non-movable allocations. The requested
2158 amount is spread evenly throughout all nodes in the
2159 system as ZONE_NORMAL. The remaining memory is used for
2160 movable memory in its own zone, ZONE_MOVABLE. In the
2161 event, a node is too small to have both ZONE_NORMAL and
2162 ZONE_MOVABLE, kernelcore memory will take priority and
2163 other nodes will have a larger ZONE_MOVABLE.
2165 ZONE_MOVABLE is used for the allocation of pages that
2166 may be reclaimed or moved by the page migration
2167 subsystem. Note that allocations like PTEs-from-HighMem
2168 still use the HighMem zone if it exists, and the Normal
2169 zone if it does not.
2171 It is possible to specify the exact amount of memory in
2172 the form of "nn[KMGTPE]", a percentage of total system
2173 memory in the form of "nn%", or "mirror". If "mirror"
2174 option is specified, mirrored (reliable) memory is used
2175 for non-movable allocations and remaining memory is used
2176 for Movable pages. "nn[KMGTPE]", "nn%", and "mirror"
2177 are exclusive, so you cannot specify multiple forms.
2179 kgdbdbgp= [KGDB,HW] kgdb over EHCI usb debug port.
2180 Format: <Controller#>[,poll interval]
2181 The controller # is the number of the ehci usb debug
2182 port as it is probed via PCI. The poll interval is
2183 optional and is the number seconds in between
2184 each poll cycle to the debug port in case you need
2185 the functionality for interrupting the kernel with
2186 gdb or control-c on the dbgp connection. When
2187 not using this parameter you use sysrq-g to break into
2188 the kernel debugger.
2190 kgdboc= [KGDB,HW] kgdb over consoles.
2191 Requires a tty driver that supports console polling,
2192 or a supported polling keyboard driver (non-usb).
2193 Serial only format: <serial_device>[,baud]
2194 keyboard only format: kbd
2195 keyboard and serial format: kbd,<serial_device>[,baud]
2196 Optional Kernel mode setting:
2197 kms, kbd format: kms,kbd
2198 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud]
2200 kgdboc_earlycon= [KGDB,HW]
2201 If the boot console provides the ability to read
2202 characters and can work in polling mode, you can use
2203 this parameter to tell kgdb to use it as a backend
2204 until the normal console is registered. Intended to
2205 be used together with the kgdboc parameter which
2206 specifies the normal console to transition to.
2208 The name of the early console should be specified
2209 as the value of this parameter. Note that the name of
2210 the early console might be different than the tty
2211 name passed to kgdboc. It's OK to leave the value
2212 blank and the first boot console that implements
2213 read() will be picked.
2215 kgdbwait [KGDB] Stop kernel execution and enter the
2216 kernel debugger at the earliest opportunity.
2218 kmac= [MIPS] Korina ethernet MAC address.
2219 Configure the RouterBoard 532 series on-chip
2220 Ethernet adapter MAC address.
2222 kmemleak= [KNL] Boot-time kmemleak enable/disable
2223 Valid arguments: on, off
2225 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y,
2228 kprobe_event=[probe-list]
2229 [FTRACE] Add kprobe events and enable at boot time.
2230 The probe-list is a semicolon delimited list of probe
2231 definitions. Each definition is same as kprobe_events
2232 interface, but the parameters are comma delimited.
2233 For example, to add a kprobe event on vfs_read with
2234 arg1 and arg2, add to the command line;
2236 kprobe_event=p,vfs_read,$arg1,$arg2
2238 See also Documentation/trace/kprobetrace.rst "Kernel
2239 Boot Parameter" section.
2241 kpti= [ARM64] Control page table isolation of user
2242 and kernel address spaces.
2243 Default: enabled on cores which need mitigation.
2247 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs.
2248 Default is 0 (don't ignore, but inject #GP)
2250 kvm.enable_vmware_backdoor=[KVM] Support VMware backdoor PV interface.
2251 Default is false (don't support).
2253 kvm.mmu_audit= [KVM] This is a R/W parameter which allows audit
2258 [KVM] Controls the software workaround for the
2259 X86_BUG_ITLB_MULTIHIT bug.
2260 force : Always deploy workaround.
2261 off : Never deploy workaround.
2262 auto : Deploy workaround based on the presence of
2263 X86_BUG_ITLB_MULTIHIT.
2267 If the software workaround is enabled for the host,
2268 guests do need not to enable it for nested guests.
2270 kvm.nx_huge_pages_recovery_ratio=
2271 [KVM] Controls how many 4KiB pages are periodically zapped
2272 back to huge pages. 0 disables the recovery, otherwise if
2273 the value is N KVM will zap 1/Nth of the 4KiB pages every
2274 minute. The default is 60.
2276 kvm-amd.nested= [KVM,AMD] Allow nested virtualization in KVM/SVM.
2277 Default is 1 (enabled)
2279 kvm-amd.npt= [KVM,AMD] Disable nested paging (virtualized MMU)
2281 Default is 1 (enabled) if in 64-bit or 32-bit PAE mode.
2284 [KVM,ARM] Select one of KVM/arm64's modes of operation.
2286 nvhe: Standard nVHE-based mode, without support for
2289 protected: nVHE-based mode with support for guests whose
2290 state is kept private from the host.
2291 Not valid if the kernel is running in EL2.
2293 Defaults to VHE/nVHE based on hardware support.
2295 kvm-arm.vgic_v3_group0_trap=
2296 [KVM,ARM] Trap guest accesses to GICv3 group-0
2299 kvm-arm.vgic_v3_group1_trap=
2300 [KVM,ARM] Trap guest accesses to GICv3 group-1
2303 kvm-arm.vgic_v3_common_trap=
2304 [KVM,ARM] Trap guest accesses to GICv3 common
2307 kvm-arm.vgic_v4_enable=
2308 [KVM,ARM] Allow use of GICv4 for direct injection of
2311 kvm_cma_resv_ratio=n [PPC]
2312 Reserves given percentage from system memory area for
2313 contiguous memory allocation for KVM hash pagetable
2315 By default it reserves 5% of total system memory.
2319 kvm-intel.ept= [KVM,Intel] Disable extended page tables
2320 (virtualized MMU) support on capable Intel chips.
2321 Default is 1 (enabled)
2323 kvm-intel.emulate_invalid_guest_state=
2324 [KVM,Intel] Enable emulation of invalid guest states
2325 Default is 0 (disabled)
2327 kvm-intel.flexpriority=
2328 [KVM,Intel] Disable FlexPriority feature (TPR shadow).
2329 Default is 1 (enabled)
2332 [KVM,Intel] Enable VMX nesting (nVMX).
2333 Default is 0 (disabled)
2335 kvm-intel.unrestricted_guest=
2336 [KVM,Intel] Disable unrestricted guest feature
2337 (virtualized real and unpaged mode) on capable
2338 Intel chips. Default is 1 (enabled)
2340 kvm-intel.vmentry_l1d_flush=[KVM,Intel] Mitigation for L1 Terminal Fault
2343 Valid arguments: never, cond, always
2345 always: L1D cache flush on every VMENTER.
2346 cond: Flush L1D on VMENTER only when the code between
2347 VMEXIT and VMENTER can leak host memory.
2348 never: Disables the mitigation
2350 Default is cond (do L1 cache flush in specific instances)
2352 kvm-intel.vpid= [KVM,Intel] Disable Virtual Processor Identification
2353 feature (tagged TLBs) on capable Intel chips.
2354 Default is 1 (enabled)
2356 l1tf= [X86] Control mitigation of the L1TF vulnerability on
2359 The kernel PTE inversion protection is unconditionally
2360 enabled and cannot be disabled.
2363 Provides all available mitigations for the
2364 L1TF vulnerability. Disables SMT and
2365 enables all mitigations in the
2366 hypervisors, i.e. unconditional L1D flush.
2368 SMT control and L1D flush control via the
2369 sysfs interface is still possible after
2370 boot. Hypervisors will issue a warning
2371 when the first VM is started in a
2372 potentially insecure configuration,
2373 i.e. SMT enabled or L1D flush disabled.
2376 Same as 'full', but disables SMT and L1D
2377 flush runtime control. Implies the
2378 'nosmt=force' command line option.
2379 (i.e. sysfs control of SMT is disabled.)
2382 Leaves SMT enabled and enables the default
2383 hypervisor mitigation, i.e. conditional
2386 SMT control and L1D flush control via the
2387 sysfs interface is still possible after
2388 boot. Hypervisors will issue a warning
2389 when the first VM is started in a
2390 potentially insecure configuration,
2391 i.e. SMT enabled or L1D flush disabled.
2395 Disables SMT and enables the default
2396 hypervisor mitigation.
2398 SMT control and L1D flush control via the
2399 sysfs interface is still possible after
2400 boot. Hypervisors will issue a warning
2401 when the first VM is started in a
2402 potentially insecure configuration,
2403 i.e. SMT enabled or L1D flush disabled.
2406 Same as 'flush', but hypervisors will not
2407 warn when a VM is started in a potentially
2408 insecure configuration.
2411 Disables hypervisor mitigations and doesn't
2413 It also drops the swap size and available
2414 RAM limit restriction on both hypervisor and
2419 For details see: Documentation/admin-guide/hw-vuln/l1tf.rst
2425 lapic [X86-32,APIC] Enable the local APIC even if BIOS
2428 lapic= [X86,APIC] Do not use TSC deadline
2429 value for LAPIC timer one-shot implementation. Default
2430 back to the programmable timer unit in the LAPIC.
2431 Format: notscdeadline
2433 lapic_timer_c2_ok [X86,APIC] trust the local apic timer
2436 libata.dma= [LIBATA] DMA control
2437 libata.dma=0 Disable all PATA and SATA DMA
2438 libata.dma=1 PATA and SATA Disk DMA only
2439 libata.dma=2 ATAPI (CDROM) DMA only
2440 libata.dma=4 Compact Flash DMA only
2441 Combinations also work, so libata.dma=3 enables DMA
2442 for disks and CDROMs, but not CFs.
2444 libata.ignore_hpa= [LIBATA] Ignore HPA limit
2445 libata.ignore_hpa=0 keep BIOS limits (default)
2446 libata.ignore_hpa=1 ignore limits, using full disk
2448 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume
2452 libata.force= [LIBATA] Force configurations. The format is comma-
2453 separated list of "[ID:]VAL" where ID is
2454 PORT[.DEVICE]. PORT and DEVICE are decimal numbers
2455 matching port, link or device. Basically, it matches
2456 the ATA ID string printed on console by libata. If
2457 the whole ID part is omitted, the last PORT and DEVICE
2458 values are used. If ID hasn't been specified yet, the
2459 configuration applies to all ports, links and devices.
2461 If only DEVICE is omitted, the parameter applies to
2462 the port and all links and devices behind it. DEVICE
2463 number of 0 either selects the first device or the
2464 first fan-out link behind PMP device. It does not
2465 select the host link. DEVICE number of 15 selects the
2466 host link and device attached to it.
2468 The VAL specifies the configuration to force. As long
2469 as there's no ambiguity shortcut notation is allowed.
2470 For example, both 1.5 and 1.5G would work for 1.5Gbps.
2471 The following configurations can be forced.
2473 * Cable type: 40c, 80c, short40c, unk, ign or sata.
2474 Any ID with matching PORT is used.
2476 * SATA link speed limit: 1.5Gbps or 3.0Gbps.
2478 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7].
2479 udma[/][16,25,33,44,66,100,133] notation is also
2482 * [no]ncq: Turn on or off NCQ.
2484 * [no]ncqtrim: Turn off queued DSM TRIM.
2486 * nohrst, nosrst, norst: suppress hard, soft
2489 * rstonce: only attempt one reset during
2490 hot-unplug link recovery
2492 * dump_id: dump IDENTIFY data.
2494 * atapi_dmadir: Enable ATAPI DMADIR bridge support
2496 * disable: Disable this device.
2498 If there are multiple matching configurations changing
2499 the same attribute, the last one is used.
2501 memblock=debug [KNL] Enable memblock debug messages.
2503 load_ramdisk= [RAM] [Deprecated]
2505 lockd.nlm_grace_period=P [NFS] Assign grace period.
2508 lockd.nlm_tcpport=N [NFS] Assign TCP port.
2511 lockd.nlm_timeout=T [NFS] Assign timeout value.
2514 lockd.nlm_udpport=M [NFS] Assign UDP port.
2517 lockdown= [SECURITY]
2518 { integrity | confidentiality }
2519 Enable the kernel lockdown feature. If set to
2520 integrity, kernel features that allow userland to
2521 modify the running kernel are disabled. If set to
2522 confidentiality, kernel features that allow userland
2523 to extract confidential information from the kernel
2526 locktorture.nreaders_stress= [KNL]
2527 Set the number of locking read-acquisition kthreads.
2528 Defaults to being automatically set based on the
2529 number of online CPUs.
2531 locktorture.nwriters_stress= [KNL]
2532 Set the number of locking write-acquisition kthreads.
2534 locktorture.onoff_holdoff= [KNL]
2535 Set time (s) after boot for CPU-hotplug testing.
2537 locktorture.onoff_interval= [KNL]
2538 Set time (s) between CPU-hotplug operations, or
2539 zero to disable CPU-hotplug testing.
2541 locktorture.shuffle_interval= [KNL]
2542 Set task-shuffle interval (jiffies). Shuffling
2543 tasks allows some CPUs to go into dyntick-idle
2544 mode during the locktorture test.
2546 locktorture.shutdown_secs= [KNL]
2547 Set time (s) after boot system shutdown. This
2548 is useful for hands-off automated testing.
2550 locktorture.stat_interval= [KNL]
2551 Time (s) between statistics printk()s.
2553 locktorture.stutter= [KNL]
2554 Time (s) to stutter testing, for example,
2555 specifying five seconds causes the test to run for
2556 five seconds, wait for five seconds, and so on.
2557 This tests the locking primitive's ability to
2558 transition abruptly to and from idle.
2560 locktorture.torture_type= [KNL]
2561 Specify the locking implementation to test.
2563 locktorture.verbose= [KNL]
2564 Enable additional printk() statements.
2566 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver
2569 loglevel= All Kernel Messages with a loglevel smaller than the
2570 console loglevel will be printed to the console. It can
2571 also be changed with klogd or other programs. The
2572 loglevels are defined as follows:
2574 0 (KERN_EMERG) system is unusable
2575 1 (KERN_ALERT) action must be taken immediately
2576 2 (KERN_CRIT) critical conditions
2577 3 (KERN_ERR) error conditions
2578 4 (KERN_WARNING) warning conditions
2579 5 (KERN_NOTICE) normal but significant condition
2580 6 (KERN_INFO) informational
2581 7 (KERN_DEBUG) debug-level messages
2583 log_buf_len=n[KMG] Sets the size of the printk ring buffer,
2584 in bytes. n must be a power of two and greater
2585 than the minimal size. The minimal size is defined
2586 by LOG_BUF_SHIFT kernel config parameter. There is
2587 also CONFIG_LOG_CPU_MAX_BUF_SHIFT config parameter
2588 that allows to increase the default size depending on
2589 the number of CPUs. See init/Kconfig for more details.
2591 logo.nologo [FB] Disables display of the built-in Linux logo.
2592 This may be used to provide more screen space for
2593 kernel log messages and is useful when debugging
2594 kernel boot problems.
2596 lp=0 [LP] Specify parallel ports to use, e.g,
2597 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses
2598 lp=reset first parallel port). 'lp=0' disables the
2599 lp=auto printer driver. 'lp=reset' (which can be
2600 specified in addition to the ports) causes
2601 attached printers to be reset. Using
2602 lp=port1,port2,... specifies the parallel ports
2603 to associate lp devices with, starting with
2604 lp0. A port specification may be 'none' to skip
2605 that lp device, or a parport name such as
2606 'parport0'. Specifying 'lp=auto' instead of a
2607 port specification list means that device IDs
2608 from each port should be examined, to see if
2609 an IEEE 1284-compliant printer is attached; if
2610 so, the driver will manage that printer.
2611 See also header of drivers/char/lp.c.
2614 Sets loops_per_jiffy to given constant, thus avoiding
2615 time-consuming boot-time autodetection (up to 250 ms per
2616 CPU). 0 enables autodetection (default). To determine
2617 the correct value for your kernel, boot with normal
2618 autodetection and see what value is printed. Note that
2619 on SMP systems the preset will be applied to all CPUs,
2620 which is likely to cause problems if your CPUs need
2621 significantly divergent settings. An incorrect value
2622 will cause delays in the kernel to be wrong, leading to
2623 unpredictable I/O errors and other breakage. Although
2624 unlikely, in the extreme case this might damage your
2628 Format: <io>,<irq>,<dma>
2630 lsm.debug [SECURITY] Enable LSM initialization debugging output.
2633 [SECURITY] Choose order of LSM initialization. This
2634 overrides CONFIG_LSM, and the "security=" parameter.
2636 machvec= [IA-64] Force the use of a particular machine-vector
2637 (machvec) in a generic kernel.
2638 Example: machvec=hpzx1
2640 machtype= [Loongson] Share the same kernel image file between
2641 different yeeloong laptops.
2642 Example: machtype=lemote-yeeloong-2f-7inch
2644 max_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory greater
2645 than or equal to this physical address is ignored.
2647 maxcpus= [SMP] Maximum number of processors that an SMP kernel
2648 will bring up during bootup. maxcpus=n : n >= 0 limits
2649 the kernel to bring up 'n' processors. Surely after
2650 bootup you can bring up the other plugged cpu by executing
2651 "echo 1 > /sys/devices/system/cpu/cpuX/online". So maxcpus
2652 only takes effect during system bootup.
2653 While n=0 is a special case, it is equivalent to "nosmp",
2654 which also disables the IO APIC.
2656 max_loop= [LOOP] The number of loop block devices that get
2657 (loop.max_loop) unconditionally pre-created at init time. The default
2658 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead
2659 of statically allocating a predefined number, loop
2660 devices can be requested on-demand with the
2661 /dev/loop-control interface.
2663 mce [X86-32] Machine Check Exception
2665 mce=option [X86-64] See Documentation/x86/x86_64/boot-options.rst
2667 md= [HW] RAID subsystems devices and level
2668 See Documentation/admin-guide/md.rst.
2671 Format: <first>,<last>
2672 Specifies range of consoles to be captured by the MDA.
2675 Control mitigation for the Micro-architectural Data
2676 Sampling (MDS) vulnerability.
2678 Certain CPUs are vulnerable to an exploit against CPU
2679 internal buffers which can forward information to a
2680 disclosure gadget under certain conditions.
2682 In vulnerable processors, the speculatively
2683 forwarded data can be used in a cache side channel
2684 attack, to access data to which the attacker does
2685 not have direct access.
2687 This parameter controls the MDS mitigation. The
2690 full - Enable MDS mitigation on vulnerable CPUs
2691 full,nosmt - Enable MDS mitigation and disable
2692 SMT on vulnerable CPUs
2693 off - Unconditionally disable MDS mitigation
2695 On TAA-affected machines, mds=off can be prevented by
2696 an active TAA mitigation as both vulnerabilities are
2697 mitigated with the same mechanism so in order to disable
2698 this mitigation, you need to specify tsx_async_abort=off
2701 Not specifying this option is equivalent to
2704 For details see: Documentation/admin-guide/hw-vuln/mds.rst
2706 mem=nn[KMG] [KNL,BOOT] Force usage of a specific amount of memory
2707 Amount of memory to be used in cases as follows:
2710 2 when the kernel is not able to see the whole system memory;
2711 3 memory that lies after 'mem=' boundary is excluded from
2712 the hypervisor, then assigned to KVM guests.
2714 [X86] Work as limiting max address. Use together
2715 with memmap= to avoid physical address space collisions.
2716 Without memmap= PCI devices could be placed at addresses
2717 belonging to unused RAM.
2719 Note that this only takes effects during boot time since
2720 in above case 3, memory may need be hot added after boot
2721 if system memory of hypervisor is not sufficient.
2723 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel
2727 [KNL,SH] Allow user to override the default size for
2728 per-device physically contiguous DMA buffers.
2730 memhp_default_state=online/offline
2731 [KNL] Set the initial state for the memory hotplug
2732 onlining policy. If not specified, the default value is
2733 set according to the
2734 CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE kernel config
2736 See Documentation/admin-guide/mm/memory-hotplug.rst.
2738 memmap=exactmap [KNL,X86] Enable setting of an exact
2739 E820 memory map, as specified by the user.
2740 Such memmap=exactmap lines can be constructed based on
2741 BIOS output or other requirements. See the memmap=nn@ss
2744 memmap=nn[KMG]@ss[KMG]
2745 [KNL, X86, MIPS, XTENSA] Force usage of a specific region of memory.
2746 Region of memory to be used is from ss to ss+nn.
2747 If @ss[KMG] is omitted, it is equivalent to mem=nn[KMG],
2748 which limits max address to nn[KMG].
2749 Multiple different regions can be specified,
2752 memmap=100M@2G,100M#3G,1G!1024G
2754 memmap=nn[KMG]#ss[KMG]
2755 [KNL,ACPI] Mark specific memory as ACPI data.
2756 Region of memory to be marked is from ss to ss+nn.
2758 memmap=nn[KMG]$ss[KMG]
2759 [KNL,ACPI] Mark specific memory as reserved.
2760 Region of memory to be reserved is from ss to ss+nn.
2761 Example: Exclude memory from 0x18690000-0x1869ffff
2762 memmap=64K$0x18690000
2764 memmap=0x10000$0x18690000
2765 Some bootloaders may need an escape character before '$',
2766 like Grub2, otherwise '$' and the following number
2769 memmap=nn[KMG]!ss[KMG]
2770 [KNL,X86] Mark specific memory as protected.
2771 Region of memory to be used, from ss to ss+nn.
2772 The memory region may be marked as e820 type 12 (0xc)
2773 and is NVDIMM or ADR memory.
2775 memmap=<size>%<offset>-<oldtype>+<newtype>
2776 [KNL,ACPI] Convert memory within the specified region
2777 from <oldtype> to <newtype>. If "-<oldtype>" is left
2778 out, the whole region will be marked as <newtype>,
2779 even if previously unavailable. If "+<newtype>" is left
2780 out, matching memory will be removed. Types are
2781 specified as e820 types, e.g., 1 = RAM, 2 = reserved,
2782 3 = ACPI, 12 = PRAM.
2784 memory_corruption_check=0/1 [X86]
2785 Some BIOSes seem to corrupt the first 64k of
2786 memory when doing things like suspend/resume.
2787 Setting this option will scan the memory
2788 looking for corruption. Enabling this will
2789 both detect corruption and prevent the kernel
2790 from using the memory being corrupted.
2791 However, its intended as a diagnostic tool; if
2792 repeatable BIOS-originated corruption always
2793 affects the same memory, you can use memmap=
2794 to prevent the kernel from using that memory.
2796 memory_corruption_check_size=size [X86]
2797 By default it checks for corruption in the low
2798 64k, making this memory unavailable for normal
2799 use. Use this parameter to scan for
2800 corruption in more or less memory.
2802 memory_corruption_check_period=seconds [X86]
2803 By default it checks for corruption every 60
2804 seconds. Use this parameter to check at some
2805 other rate. 0 disables periodic checking.
2807 memory_hotplug.memmap_on_memory
2808 [KNL,X86,ARM] Boolean flag to enable this feature.
2809 Format: {on | off (default)}
2810 When enabled, runtime hotplugged memory will
2811 allocate its internal metadata (struct pages)
2812 from the hotadded memory which will allow to
2813 hotadd a lot of memory without requiring
2814 additional memory to do so.
2815 This feature is disabled by default because it
2816 has some implication on large (e.g. GB)
2817 allocations in some configurations (e.g. small
2819 The state of the flag can be read in
2820 /sys/module/memory_hotplug/parameters/memmap_on_memory.
2821 Note that even when enabled, there are a few cases where
2822 the feature is not effective.
2824 memtest= [KNL,X86,ARM,PPC,RISCV] Enable memtest
2826 default : 0 <disable>
2827 Specifies the number of memtest passes to be
2828 performed. Each pass selects another test
2829 pattern from a given set of patterns. Memtest
2830 fills the memory with this pattern, validates
2831 memory contents and reserves bad memory
2832 regions that are detected.
2834 mem_encrypt= [X86-64] AMD Secure Memory Encryption (SME) control
2835 Valid arguments: on, off
2836 Default (depends on kernel configuration option):
2837 on (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=y)
2838 off (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=n)
2839 mem_encrypt=on: Activate SME
2840 mem_encrypt=off: Do not activate SME
2842 Refer to Documentation/virt/kvm/amd-memory-encryption.rst
2843 for details on when memory encryption can be activated.
2845 mem_sleep_default= [SUSPEND] Default system suspend mode:
2846 s2idle - Suspend-To-Idle
2847 shallow - Power-On Suspend or equivalent (if supported)
2848 deep - Suspend-To-RAM or equivalent (if supported)
2849 See Documentation/admin-guide/pm/sleep-states.rst.
2851 meye.*= [HW] Set MotionEye Camera parameters
2852 See Documentation/admin-guide/media/meye.rst.
2854 mfgpt_irq= [IA-32] Specify the IRQ to use for the
2855 Multi-Function General Purpose Timers on AMD Geode
2858 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when
2859 the BIOS has incorrectly applied a workaround. TinyBIOS
2860 version 0.98 is known to be affected, 0.99 fixes the
2861 problem by letting the user disable the workaround.
2865 min_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory below this
2866 physical address is ignored.
2868 mini2440= [ARM,HW,KNL]
2869 Format:[0..2][b][c][t]
2871 MINI2440 configuration specification:
2872 0 - The attached screen is the 3.5" TFT
2873 1 - The attached screen is the 7" TFT
2874 2 - The VGA Shield is attached (1024x768)
2875 Leaving out the screen size parameter will not load
2876 the TFT driver, and the framebuffer will be left
2878 b - Enable backlight. The TFT backlight pin will be
2879 linked to the kernel VESA blanking code and a GPIO
2880 LED. This parameter is not necessary when using the
2882 c - Enable the s3c camera interface.
2883 t - Reserved for enabling touchscreen support. The
2884 touchscreen support is not enabled in the mainstream
2885 kernel as of 2.6.30, a preliminary port can be found
2886 in the "bleeding edge" mini2440 support kernel at
2887 https://repo.or.cz/w/linux-2.6/mini2440.git
2890 [X86,PPC,S390,ARM64] Control optional mitigations for
2891 CPU vulnerabilities. This is a set of curated,
2892 arch-independent options, each of which is an
2893 aggregation of existing arch-specific options.
2896 Disable all optional CPU mitigations. This
2897 improves system performance, but it may also
2898 expose users to several CPU vulnerabilities.
2899 Equivalent to: nopti [X86,PPC]
2901 nospectre_v1 [X86,PPC]
2903 nospectre_v2 [X86,PPC,S390,ARM64]
2904 spectre_v2_user=off [X86]
2905 spec_store_bypass_disable=off [X86,PPC]
2906 ssbd=force-off [ARM64]
2909 tsx_async_abort=off [X86]
2910 kvm.nx_huge_pages=off [X86]
2911 no_entry_flush [PPC]
2912 no_uaccess_flush [PPC]
2915 This does not have any effect on
2916 kvm.nx_huge_pages when
2917 kvm.nx_huge_pages=force.
2920 Mitigate all CPU vulnerabilities, but leave SMT
2921 enabled, even if it's vulnerable. This is for
2922 users who don't want to be surprised by SMT
2923 getting disabled across kernel upgrades, or who
2924 have other ways of avoiding SMT-based attacks.
2925 Equivalent to: (default behavior)
2928 Mitigate all CPU vulnerabilities, disabling SMT
2929 if needed. This is for users who always want to
2930 be fully mitigated, even if it means losing SMT.
2931 Equivalent to: l1tf=flush,nosmt [X86]
2932 mds=full,nosmt [X86]
2933 tsx_async_abort=full,nosmt [X86]
2936 [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this
2937 parameter allows control of the logging verbosity for
2938 the additional memory initialisation checks. A value
2939 of 0 disables mminit logging and a level of 4 will
2940 log everything. Information is printed at KERN_DEBUG
2941 so loglevel=8 may also need to be specified.
2944 [KNL] When CONFIG_MODULE_SIG is set, this means that
2945 modules without (valid) signatures will fail to load.
2946 Note that if CONFIG_MODULE_SIG_FORCE is set, that
2947 is always true, so this option does nothing.
2949 module_blacklist= [KNL] Do not load a comma-separated list of
2950 modules. Useful for debugging problem modules.
2953 [MOUSE] Maximum time between finger touching and
2954 leaving touchpad surface for touch to be considered
2955 a tap and be reported as a left button click (for
2956 touchpads working in absolute mode only).
2958 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices
2959 reporting absolute coordinates, such as tablets
2960 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices
2961 reporting absolute coordinates, such as tablets
2963 movablecore= [KNL,X86,IA-64,PPC]
2964 Format: nn[KMGTPE] | nn%
2965 This parameter is the complement to kernelcore=, it
2966 specifies the amount of memory used for migratable
2967 allocations. If both kernelcore and movablecore is
2968 specified, then kernelcore will be at *least* the
2969 specified value but may be more. If movablecore on its
2970 own is specified, the administrator must be careful
2971 that the amount of memory usable for all allocations
2974 movable_node [KNL] Boot-time switch to make hotplugable memory
2975 NUMA nodes to be movable. This means that the memory
2976 of such nodes will be usable only for movable
2977 allocations which rules out almost all kernel
2978 allocations. Use with caution!
2980 MTD_Partition= [MTD]
2981 Format: <name>,<region-number>,<size>,<offset>
2983 MTD_Region= [MTD] Format:
2984 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>]
2987 See drivers/mtd/parsers/cmdlinepart.c
2989 multitce=off [PPC] This parameter disables the use of the pSeries
2990 firmware feature for updating multiple TCE entries
2993 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration
2995 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock]
2997 boundary - index of last SLC block on Flex-OneNAND.
2998 The remaining blocks are configured as MLC blocks.
2999 lock - Configure if Flex-OneNAND boundary should be locked.
3000 Once locked, the boundary cannot be changed.
3001 1 indicates lock status, 0 indicates unlock status.
3004 ARM/S3C2412 JIVE boot control
3006 See arch/arm/mach-s3c/mach-jive.c
3008 mtouchusb.raw_coordinates=
3009 [HW] Make the MicroTouch USB driver use raw coordinates
3010 ('y', default) or cooked coordinates ('n')
3012 mtrr_chunk_size=nn[KMG] [X86]
3013 used for mtrr cleanup. It is largest continuous chunk
3014 that could hold holes aka. UC entries.
3016 mtrr_gran_size=nn[KMG] [X86]
3017 Used for mtrr cleanup. It is granularity of mtrr block.
3019 Large value could prevent small alignment from
3022 mtrr_spare_reg_nr=n [X86]
3024 Range: 0,7 : spare reg number
3026 Used for mtrr cleanup. It is spare mtrr entries number.
3027 Set to 2 or more if your graphical card needs more.
3029 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card
3031 netdev= [NET] Network devices parameters
3032 Format: <irq>,<io>,<mem_start>,<mem_end>,<name>
3033 Note that mem_start is often overloaded to mean
3034 something different and driver-specific.
3035 This usage is only documented in each driver source
3039 [NETFILTER] Enable connection tracking flow accounting
3040 0 to disable accounting
3041 1 to enable accounting
3044 nfsaddrs= [NFS] Deprecated. Use ip= instead.
3045 See Documentation/admin-guide/nfs/nfsroot.rst.
3047 nfsroot= [NFS] nfs root filesystem for disk-less boxes.
3048 See Documentation/admin-guide/nfs/nfsroot.rst.
3050 nfsrootdebug [NFS] enable nfsroot debugging messages.
3051 See Documentation/admin-guide/nfs/nfsroot.rst.
3053 nfs.callback_nr_threads=
3054 [NFSv4] set the total number of threads that the
3055 NFS client will assign to service NFSv4 callback
3058 nfs.callback_tcpport=
3059 [NFS] set the TCP port on which the NFSv4 callback
3060 channel should listen.
3063 [NFS] sets the pathname to the program which is used
3064 to update the NFS client cache entries.
3066 nfs.cache_getent_timeout=
3067 [NFS] sets the timeout after which an attempt to
3068 update a cache entry is deemed to have failed.
3070 nfs.idmap_cache_timeout=
3071 [NFS] set the maximum lifetime for idmapper cache
3075 [NFS] enable 64-bit inode numbers.
3076 If zero, the NFS client will fake up a 32-bit inode
3077 number for the readdir() and stat() syscalls instead
3078 of returning the full 64-bit number.
3079 The default is to return 64-bit inode numbers.
3081 nfs.max_session_cb_slots=
3082 [NFSv4.1] Sets the maximum number of session
3083 slots the client will assign to the callback
3084 channel. This determines the maximum number of
3085 callbacks the client will process in parallel for
3086 a particular server.
3088 nfs.max_session_slots=
3089 [NFSv4.1] Sets the maximum number of session slots
3090 the client will attempt to negotiate with the server.
3091 This limits the number of simultaneous RPC requests
3092 that the client can send to the NFSv4.1 server.
3093 Note that there is little point in setting this
3094 value higher than the max_tcp_slot_table_limit.
3096 nfs.nfs4_disable_idmapping=
3097 [NFSv4] When set to the default of '1', this option
3098 ensures that both the RPC level authentication
3099 scheme and the NFS level operations agree to use
3100 numeric uids/gids if the mount is using the
3101 'sec=sys' security flavour. In effect it is
3102 disabling idmapping, which can make migration from
3103 legacy NFSv2/v3 systems to NFSv4 easier.
3104 Servers that do not support this mode of operation
3105 will be autodetected by the client, and it will fall
3106 back to using the idmapper.
3107 To turn off this behaviour, set the value to '0'.
3109 [NFS4] Specify an additional fixed unique ident-
3110 ification string that NFSv4 clients can insert into
3111 their nfs_client_id4 string. This is typically a
3112 UUID that is generated at system install time.
3114 nfs.send_implementation_id =
3115 [NFSv4.1] Send client implementation identification
3116 information in exchange_id requests.
3117 If zero, no implementation identification information
3119 The default is to send the implementation identification
3122 nfs.recover_lost_locks =
3123 [NFSv4] Attempt to recover locks that were lost due
3124 to a lease timeout on the server. Please note that
3125 doing this risks data corruption, since there are
3126 no guarantees that the file will remain unchanged
3127 after the locks are lost.
3128 If you want to enable the kernel legacy behaviour of
3129 attempting to recover these locks, then set this
3131 The default parameter value of '0' causes the kernel
3132 not to attempt recovery of lost locks.
3134 nfs4.layoutstats_timer =
3135 [NFSv4.2] Change the rate at which the kernel sends
3136 layoutstats to the pNFS metadata server.
3138 Setting this to value to 0 causes the kernel to use
3139 whatever value is the default set by the layout
3140 driver. A non-zero value sets the minimum interval
3141 in seconds between layoutstats transmissions.
3143 nfsd.nfs4_disable_idmapping=
3144 [NFSv4] When set to the default of '1', the NFSv4
3145 server will return only numeric uids and gids to
3146 clients using auth_sys, and will accept numeric uids
3147 and gids from such clients. This is intended to ease
3148 migration from NFSv2/v3.
3150 nmi_backtrace.backtrace_idle [KNL]
3151 Dump stacks even of idle CPUs in response to an
3152 NMI stack-backtrace request.
3154 nmi_debug= [KNL,SH] Specify one or more actions to take
3155 when a NMI is triggered.
3156 Format: [state][,regs][,debounce][,die]
3158 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels
3159 Format: [panic,][nopanic,][num]
3161 0 - turn hardlockup detector in nmi_watchdog off
3162 1 - turn hardlockup detector in nmi_watchdog on
3163 When panic is specified, panic when an NMI watchdog
3164 timeout occurs (or 'nopanic' to not panic on an NMI
3165 watchdog, if CONFIG_BOOTPARAM_HARDLOCKUP_PANIC is set)
3166 To disable both hard and soft lockup detectors,
3167 please see 'nowatchdog'.
3168 This is useful when you use a panic=... timeout and
3169 need the box quickly up again.
3171 These settings can be accessed at runtime via
3172 the nmi_watchdog and hardlockup_panic sysctls.
3174 netpoll.carrier_timeout=
3175 [NET] Specifies amount of time (in seconds) that
3176 netpoll should wait for a carrier. By default netpoll
3179 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths
3180 emulation library even if a 387 maths coprocessor
3183 no5lvl [X86-64] Disable 5-level paging mode. Forces
3184 kernel to use 4-level paging instead.
3186 nofsgsbase [X86] Disables FSGSBASE instructions.
3189 [HW] Never suspend the console
3190 Disable suspending of consoles during suspend and
3191 hibernate operations. Once disabled, debugging
3192 messages can reach various consoles while the rest
3193 of the system is being put to sleep (ie, while
3194 debugging driver suspend/resume hooks). This may
3195 not work reliably with all consoles, but is known
3196 to work with serial and VGA consoles.
3197 To facilitate more flexible debugging, we also add
3198 console_suspend, a printk module parameter to control
3199 it. Users could use console_suspend (usually
3200 /sys/module/printk/parameters/console_suspend) to
3201 turn on/off it dynamically.
3203 novmcoredd [KNL,KDUMP]
3204 Disable device dump. Device dump allows drivers to
3205 append dump data to vmcore so you can collect driver
3206 specified debug info. Drivers can append the data
3207 without any limit and this data is stored in memory,
3208 so this may cause significant memory stress. Disabling
3209 device dump can help save memory but the driver debug
3210 data will be no longer available. This parameter
3211 is only available when CONFIG_PROC_VMCORE_DEVICE_DUMP
3214 noaliencache [MM, NUMA, SLAB] Disables the allocation of alien
3215 caches in the slab allocator. Saves per-node memory,
3216 but will impact performance.
3220 noaltinstr [S390] Disables alternative instructions patching
3221 (CPU alternatives feature).
3223 noapic [SMP,APIC] Tells the kernel to not make use of any
3224 IOAPICs that may be present in the system.
3226 noautogroup Disable scheduler automatic task group creation.
3228 nobats [PPC] Do not use BATs for mapping kernel lowmem
3229 on "Classic" PPC cores.
3233 noclflush [BUGS=X86] Don't use the CLFLUSH instruction
3235 nodelayacct [KNL] Disable per-task delay accounting
3237 nodsp [SH] Disable hardware DSP at boot time.
3239 noefi Disable EFI runtime services support.
3241 no_entry_flush [PPC] Don't flush the L1-D cache when entering the kernel.
3246 On X86-32 available only on PAE configured kernels.
3247 noexec=on: enable non-executable mappings (default)
3248 noexec=off: disable non-executable mappings
3251 Disable SMAP (Supervisor Mode Access Prevention)
3252 even if it is supported by processor.
3255 Disable SMEP (Supervisor Mode Execution Prevention)
3256 even if it is supported by processor.
3259 This affects only 32-bit executables.
3260 noexec32=on: enable non-executable mappings (default)
3261 read doesn't imply executable mappings
3262 noexec32=off: disable non-executable mappings
3263 read implies executable mappings
3265 nofpu [MIPS,SH] Disable hardware FPU at boot time.
3267 nofxsr [BUGS=X86-32] Disables x86 floating point extended
3268 register save and restore. The kernel will only save
3269 legacy floating-point registers on task switch.
3271 nohugeiomap [KNL,X86,PPC,ARM64] Disable kernel huge I/O mappings.
3273 nosmt [KNL,S390] Disable symmetric multithreading (SMT).
3274 Equivalent to smt=1.
3276 [KNL,X86] Disable symmetric multithreading (SMT).
3277 nosmt=force: Force disable SMT, cannot be undone
3278 via the sysfs control file.
3280 nospectre_v1 [X86,PPC] Disable mitigations for Spectre Variant 1
3281 (bounds check bypass). With this option data leaks are
3282 possible in the system.
3284 nospectre_v2 [X86,PPC_FSL_BOOK3E,ARM64] Disable all mitigations for
3285 the Spectre variant 2 (indirect branch prediction)
3286 vulnerability. System may allow data leaks with this
3289 nospec_store_bypass_disable
3290 [HW] Disable all mitigations for the Speculative Store Bypass vulnerability
3293 [PPC] Don't flush the L1-D cache after accessing user data.
3295 noxsave [BUGS=X86] Disables x86 extended register state save
3296 and restore using xsave. The kernel will fallback to
3297 enabling legacy floating-point and sse state.
3299 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended
3300 register states. The kernel will fall back to use
3301 xsave to save the states. By using this parameter,
3302 performance of saving the states is degraded because
3303 xsave doesn't support modified optimization while
3304 xsaveopt supports it on xsaveopt enabled systems.
3306 noxsaves [X86] Disables xsaves and xrstors used in saving and
3307 restoring x86 extended register state in compacted
3308 form of xsave area. The kernel will fall back to use
3309 xsaveopt and xrstor to save and restore the states
3310 in standard form of xsave area. By using this
3311 parameter, xsave area per process might occupy more
3312 memory on xsaves enabled systems.
3314 nohlt [ARM,ARM64,MICROBLAZE,SH] Forces the kernel to busy wait
3315 in do_idle() and not use the arch_cpu_idle()
3316 implementation; requires CONFIG_GENERIC_IDLE_POLL_SETUP
3317 to be effective. This is useful on platforms where the
3318 sleep(SH) or wfi(ARM,ARM64) instructions do not work
3319 correctly or when doing power measurements to evalute
3320 the impact of the sleep instructions. This is also
3321 useful when using JTAG debugger.
3323 no_file_caps Tells the kernel not to honor file capabilities. The
3324 only way then for a file to be executed with privilege
3325 is to be setuid root or executed by root.
3327 nohalt [IA-64] Tells the kernel not to use the power saving
3328 function PAL_HALT_LIGHT when idle. This increases
3329 power-consumption. On the positive side, it reduces
3330 interrupt wake-up latency, which may improve performance
3331 in certain environments such as networked servers or
3335 Force pointers printed to the console or buffers to be
3336 unhashed. By default, when a pointer is printed via %p
3337 format string, that pointer is "hashed", i.e. obscured
3338 by hashing the pointer value. This is a security feature
3339 that hides actual kernel addresses from unprivileged
3340 users, but it also makes debugging the kernel more
3341 difficult since unequal pointers can no longer be
3342 compared. However, if this command-line option is
3343 specified, then all normal pointers will have their true
3344 value printed. Pointers printed via %pK may still be
3345 hashed. This option should only be specified when
3346 debugging the kernel. Please do not use on production
3349 nohibernate [HIBERNATION] Disable hibernation and resume.
3351 nohz= [KNL] Boottime enable/disable dynamic ticks
3352 Valid arguments: on, off
3355 nohz_full= [KNL,BOOT,SMP,ISOL]
3356 The argument is a cpu list, as described above.
3357 In kernels built with CONFIG_NO_HZ_FULL=y, set
3358 the specified list of CPUs whose tick will be stopped
3359 whenever possible. The boot CPU will be forced outside
3360 the range to maintain the timekeeping. Any CPUs
3361 in this list will have their RCU callbacks offloaded,
3362 just as if they had also been called out in the
3363 rcu_nocbs= boot parameter.
3365 noiotrap [SH] Disables trapped I/O port accesses.
3367 noirqdebug [X86-32] Disables the code which attempts to detect and
3368 disable unhandled interrupt sources.
3370 no_timer_check [X86,APIC] Disables the code which tests for
3371 broken timer IRQ sources.
3373 noisapnp [ISAPNP] Disables ISA PnP code.
3375 noinitrd [RAM] Tells the kernel not to load any configured
3378 nointremap [X86-64, Intel-IOMMU] Do not enable interrupt
3380 [Deprecated - use intremap=off]
3384 noinvpcid [X86] Disable the INVPCID cpu feature.
3386 nojitter [IA-64] Disables jitter checking for ITC timers.
3388 no-kvmclock [X86,KVM] Disable paravirtualized KVM clock driver
3390 no-kvmapf [X86,KVM] Disable paravirtualized asynchronous page
3394 [X86,PV_OPS] Disable paravirtualized VMware scheduler
3395 clock and use the default one.
3397 no-steal-acc [X86,PV_OPS,ARM64] Disable paravirtualized steal time
3398 accounting. steal time is computed, but won't
3399 influence scheduler behaviour
3401 nolapic [X86-32,APIC] Do not enable or use the local APIC.
3403 nolapic_timer [X86-32,APIC] Do not use the local APIC timer.
3405 noltlbs [PPC] Do not use large page/tlb entries for kernel
3406 lowmem mapping on PPC40x and PPC8xx
3408 nomca [IA-64] Disable machine check abort handling
3410 nomce [X86-32] Disable Machine Check Exception
3412 nomfgpt [X86-32] Disable Multi-Function General Purpose
3413 Timer usage (for AMD Geode machines).
3415 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to
3416 shutdown the other cpus. Instead use the REBOOT_VECTOR
3419 nomodule Disable module load
3421 nopat [X86] Disable PAT (page attribute table extension of
3422 pagetables) support.
3424 nopcid [X86-64] Disable the PCID cpu feature.
3426 norandmaps Don't use address space randomization. Equivalent to
3427 echo 0 > /proc/sys/kernel/randomize_va_space
3429 noreplace-smp [X86-32,SMP] Don't replace SMP instructions
3430 with UP alternatives
3432 nordrand [X86] Disable kernel use of the RDRAND and
3433 RDSEED instructions even if they are supported
3434 by the processor. RDRAND and RDSEED are still
3435 available to user space applications.
3437 noresume [SWSUSP] Disables resume and restores original swap
3440 no-scroll [VGA] Disables scrollback.
3441 This is required for the Braillex ib80-piezo Braille
3442 reader made by F.H. Papenmeier (Germany).
3446 nosep [BUGS=X86-32] Disables x86 SYSENTER/SYSEXIT support.
3448 nosgx [X86-64,SGX] Disables Intel SGX kernel support.
3450 nosmp [SMP] Tells an SMP kernel to act as a UP kernel,
3451 and disable the IO APIC. legacy for "maxcpus=0".
3453 nosoftlockup [KNL] Disable the soft-lockup detector.
3455 nosync [HW,M68K] Disables sync negotiation for all devices.
3457 nowatchdog [KNL] Disable both lockup detectors, i.e.
3458 soft-lockup and NMI watchdog (hard-lockup).
3462 nox2apic [X86-64,APIC] Do not enable x2APIC mode.
3464 cpu0_hotplug [X86] Turn on CPU0 hotplug feature when
3465 CONFIG_BOOTPARAM_HOTPLUG_CPU0 is off.
3466 Some features depend on CPU0. Known dependencies are:
3467 1. Resume from suspend/hibernate depends on CPU0.
3468 Suspend/hibernate will fail if CPU0 is offline and you
3469 need to online CPU0 before suspend/hibernate.
3470 2. PIC interrupts also depend on CPU0. CPU0 can't be
3471 removed if a PIC interrupt is detected.
3472 It's said poweroff/reboot may depend on CPU0 on some
3473 machines although I haven't seen such issues so far
3474 after CPU0 is offline on a few tested machines.
3475 If the dependencies are under your control, you can
3476 turn on cpu0_hotplug.
3478 nps_mtm_hs_ctr= [KNL,ARC]
3479 This parameter sets the maximum duration, in
3480 cycles, each HW thread of the CTOP can run
3481 without interruptions, before HW switches it.
3482 The actual maximum duration is 16 times this
3484 Format: integer between 1 and 255
3487 nptcg= [IA-64] Override max number of concurrent global TLB
3488 purges which is reported from either PAL_VM_SUMMARY or
3491 nr_cpus= [SMP] Maximum number of processors that an SMP kernel
3492 could support. nr_cpus=n : n >= 1 limits the kernel to
3493 support 'n' processors. It could be larger than the
3494 number of already plugged CPU during bootup, later in
3495 runtime you can physically add extra cpu until it reaches
3496 n. So during boot up some boot time memory for per-cpu
3497 variables need be pre-allocated for later physical cpu
3500 nr_uarts= [SERIAL] maximum number of UARTs to be registered.
3502 numa_balancing= [KNL,ARM64,PPC,RISCV,S390,X86] Enable or disable automatic
3504 Allowed values are enable and disable
3506 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA.
3507 'node', 'default' can be specified
3508 This can be set from sysctl after boot.
3509 See Documentation/admin-guide/sysctl/vm.rst for details.
3511 ohci1394_dma=early [HW] enable debugging via the ohci1394 driver.
3512 See Documentation/core-api/debugging-via-ohci1394.rst for more
3515 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands
3516 Rather than timing out after 20 ms if an EC
3517 command is not properly ACKed, override the length
3518 of the timeout. We have interrupts disabled while
3519 waiting for the ACK, so if this is set too high
3520 interrupts *may* be lost!
3522 omap_mux= [OMAP] Override bootloader pin multiplexing.
3523 Format: <mux_mode0.mode_name=value>...
3524 For example, to override I2C bus2:
3525 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100
3527 oops=panic Always panic on oopses. Default is to just kill the
3528 process, but there is a small probability of
3529 deadlocking the machine.
3530 This will also cause panics on machine check exceptions.
3531 Useful together with panic=30 to trigger a reboot.
3534 [KNL] Boolean flag to control whether the page allocator
3535 should randomize its free lists. The randomization may
3536 be automatically enabled if the kernel detects it is
3537 running on a platform with a direct-mapped memory-side
3538 cache, and this parameter can be used to
3539 override/disable that behavior. The state of the flag
3540 can be read from sysfs at:
3541 /sys/module/page_alloc/parameters/shuffle.
3543 page_owner= [KNL] Boot-time page_owner enabling option.
3544 Storage of the information about who allocated
3545 each page is disabled in default. With this switch,
3547 on: enable the feature
3549 page_poison= [KNL] Boot-time parameter changing the state of
3550 poisoning on the buddy allocator, available with
3551 CONFIG_PAGE_POISONING=y.
3552 off: turn off poisoning (default)
3553 on: turn on poisoning
3555 panic= [KNL] Kernel behaviour on panic: delay <timeout>
3556 timeout > 0: seconds before rebooting
3557 timeout = 0: wait forever
3558 timeout < 0: reboot immediately
3561 panic_print= Bitmask for printing system info when panic happens.
3562 User can chose combination of the following bits:
3563 bit 0: print all tasks info
3564 bit 1: print system memory info
3565 bit 2: print timer info
3566 bit 3: print locks info if CONFIG_LOCKDEP is on
3567 bit 4: print ftrace buffer
3568 bit 5: print all printk messages in buffer
3570 panic_on_taint= Bitmask for conditionally calling panic() in add_taint()
3571 Format: <hex>[,nousertaint]
3572 Hexadecimal bitmask representing the set of TAINT flags
3573 that will cause the kernel to panic when add_taint() is
3574 called with any of the flags in this set.
3575 The optional switch "nousertaint" can be utilized to
3576 prevent userspace forced crashes by writing to sysctl
3577 /proc/sys/kernel/tainted any flagset matching with the
3578 bitmask set on panic_on_taint.
3579 See Documentation/admin-guide/tainted-kernels.rst for
3580 extra details on the taint flags that users can pick
3581 to compose the bitmask to assign to panic_on_taint.
3583 panic_on_warn panic() instead of WARN(). Useful to cause kdump
3586 crash_kexec_post_notifiers
3587 Run kdump after running panic-notifiers and dumping
3588 kmsg. This only for the users who doubt kdump always
3589 succeeds in any situation.
3590 Note that this also increases risks of kdump failure,
3591 because some panic notifiers can make the crashed
3592 kernel more unstable.
3594 parkbd.port= [HW] Parallel port number the keyboard adapter is
3595 connected to, default is 0.
3597 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation,
3598 0 for XT, 1 for AT (default is AT).
3601 parport= [HW,PPT] Specify parallel ports. 0 disables.
3602 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] }
3603 Use 'auto' to force the driver to use any
3604 IRQ/DMA settings detected (the default is to
3605 ignore detected IRQ/DMA settings because of
3606 possible conflicts). You can specify the base
3607 address, IRQ, and DMA settings; IRQ and DMA
3608 should be numbers, or 'auto' (for using detected
3609 settings on that particular port), or 'nofifo'
3610 (to avoid using a FIFO even if it is detected).
3611 Parallel ports are assigned in the order they
3612 are specified on the command line, starting
3615 parport_init_mode= [HW,PPT]
3616 Configure VIA parallel port to operate in
3617 a specific mode. This is necessary on Pegasos
3618 computer where firmware has no options for setting
3619 up parallel port mode and sets it to spp.
3620 Currently this function knows 686a and 8231 chips.
3621 Format: [spp|ps2|epp|ecp|ecpepp]
3623 pata_legacy.all= [HW,LIBATA]
3625 Set to non-zero to probe primary and secondary ISA
3626 port ranges on PCI systems where no PCI PATA device
3627 has been found at either range. Disabled by default.
3629 pata_legacy.autospeed= [HW,LIBATA]
3631 Set to non-zero if a chip is present that snoops speed
3632 changes. Disabled by default.
3634 pata_legacy.ht6560a= [HW,LIBATA]
3636 Set to 1, 2, or 3 for HT 6560A on the primary channel,
3637 the secondary channel, or both channels respectively.
3638 Disabled by default.
3640 pata_legacy.ht6560b= [HW,LIBATA]
3642 Set to 1, 2, or 3 for HT 6560B on the primary channel,
3643 the secondary channel, or both channels respectively.
3644 Disabled by default.
3646 pata_legacy.iordy_mask= [HW,LIBATA]
3648 IORDY enable mask. Set individual bits to allow IORDY
3649 for the respective channel. Bit 0 is for the first
3650 legacy channel handled by this driver, bit 1 is for
3651 the second channel, and so on. The sequence will often
3652 correspond to the primary legacy channel, the secondary
3653 legacy channel, and so on, but the handling of a PCI
3654 bus and the use of other driver options may interfere
3655 with the sequence. By default IORDY is allowed across
3658 pata_legacy.opti82c46x= [HW,LIBATA]
3660 Set to 1, 2, or 3 for Opti 82c611A on the primary
3661 channel, the secondary channel, or both channels
3662 respectively. Disabled by default.
3664 pata_legacy.opti82c611a= [HW,LIBATA]
3666 Set to 1, 2, or 3 for Opti 82c465MV on the primary
3667 channel, the secondary channel, or both channels
3668 respectively. Disabled by default.
3670 pata_legacy.pio_mask= [HW,LIBATA]
3672 PIO mode mask for autospeed devices. Set individual
3673 bits to allow the use of the respective PIO modes.
3674 Bit 0 is for mode 0, bit 1 is for mode 1, and so on.
3675 All modes allowed by default.
3677 pata_legacy.probe_all= [HW,LIBATA]
3679 Set to non-zero to probe tertiary and further ISA
3680 port ranges on PCI systems. Disabled by default.
3682 pata_legacy.probe_mask= [HW,LIBATA]
3684 Probe mask for legacy ISA PATA ports. Depending on
3685 platform configuration and the use of other driver
3686 options up to 6 legacy ports are supported: 0x1f0,
3687 0x170, 0x1e8, 0x168, 0x1e0, 0x160, however probing
3688 of individual ports can be disabled by setting the
3689 corresponding bits in the mask to 1. Bit 0 is for
3690 the first port in the list above (0x1f0), and so on.
3691 By default all supported ports are probed.
3693 pata_legacy.qdi= [HW,LIBATA]
3695 Set to non-zero to probe QDI controllers. By default
3696 set to 1 if CONFIG_PATA_QDI_MODULE, 0 otherwise.
3698 pata_legacy.winbond= [HW,LIBATA]
3700 Set to non-zero to probe Winbond controllers. Use
3701 the standard I/O port (0x130) if 1, otherwise the
3702 value given is the I/O port to use (typically 0x1b0).
3703 By default set to 1 if CONFIG_PATA_WINBOND_VLB_MODULE,
3706 pata_platform.pio_mask= [HW,LIBATA]
3708 Supported PIO mode mask. Set individual bits to allow
3709 the use of the respective PIO modes. Bit 0 is for
3710 mode 0, bit 1 is for mode 1, and so on. Mode 0 only
3714 Halt all CPUs after the first oops has been printed for
3715 the specified number of seconds. This is to be used if
3716 your oopses keep scrolling off the screen.
3721 See header of drivers/block/paride/pcd.c.
3722 See also Documentation/admin-guide/blockdev/paride.rst.
3724 pci=option[,option...] [PCI] various PCI subsystem options.
3726 Some options herein operate on a specific device
3727 or a set of devices (<pci_dev>). These are
3728 specified in one of the following formats:
3730 [<domain>:]<bus>:<dev>.<func>[/<dev>.<func>]*
3731 pci:<vendor>:<device>[:<subvendor>:<subdevice>]
3733 Note: the first format specifies a PCI
3734 bus/device/function address which may change
3735 if new hardware is inserted, if motherboard
3736 firmware changes, or due to changes caused
3737 by other kernel parameters. If the
3738 domain is left unspecified, it is
3739 taken to be zero. Optionally, a path
3740 to a device through multiple device/function
3741 addresses can be specified after the base
3742 address (this is more robust against
3743 renumbering issues). The second format
3744 selects devices using IDs from the
3745 configuration space which may match multiple
3746 devices in the system.
3748 earlydump dump PCI config space before the kernel
3750 off [X86] don't probe for the PCI bus
3751 bios [X86-32] force use of PCI BIOS, don't access
3752 the hardware directly. Use this if your machine
3753 has a non-standard PCI host bridge.
3754 nobios [X86-32] disallow use of PCI BIOS, only direct
3755 hardware access methods are allowed. Use this
3756 if you experience crashes upon bootup and you
3757 suspect they are caused by the BIOS.
3758 conf1 [X86] Force use of PCI Configuration Access
3759 Mechanism 1 (config address in IO port 0xCF8,
3760 data in IO port 0xCFC, both 32-bit).
3761 conf2 [X86] Force use of PCI Configuration Access
3762 Mechanism 2 (IO port 0xCF8 is an 8-bit port for
3763 the function, IO port 0xCFA, also 8-bit, sets
3764 bus number. The config space is then accessed
3765 through ports 0xC000-0xCFFF).
3766 See http://wiki.osdev.org/PCI for more info
3767 on the configuration access mechanisms.
3768 noaer [PCIE] If the PCIEAER kernel config parameter is
3769 enabled, this kernel boot option can be used to
3770 disable the use of PCIE advanced error reporting.
3771 nodomains [PCI] Disable support for multiple PCI
3772 root domains (aka PCI segments, in ACPI-speak).
3773 nommconf [X86] Disable use of MMCONFIG for PCI
3775 check_enable_amd_mmconf [X86] check for and enable
3776 properly configured MMIO access to PCI
3777 config space on AMD family 10h CPU
3778 nomsi [MSI] If the PCI_MSI kernel config parameter is
3779 enabled, this kernel boot option can be used to
3780 disable the use of MSI interrupts system-wide.
3781 noioapicquirk [APIC] Disable all boot interrupt quirks.
3782 Safety option to keep boot IRQs enabled. This
3783 should never be necessary.
3784 ioapicreroute [APIC] Enable rerouting of boot IRQs to the
3785 primary IO-APIC for bridges that cannot disable
3786 boot IRQs. This fixes a source of spurious IRQs
3787 when the system masks IRQs.
3788 noioapicreroute [APIC] Disable workaround that uses the
3789 boot IRQ equivalent of an IRQ that connects to
3790 a chipset where boot IRQs cannot be disabled.
3791 The opposite of ioapicreroute.
3792 biosirq [X86-32] Use PCI BIOS calls to get the interrupt
3793 routing table. These calls are known to be buggy
3794 on several machines and they hang the machine
3795 when used, but on other computers it's the only
3796 way to get the interrupt routing table. Try
3797 this option if the kernel is unable to allocate
3798 IRQs or discover secondary PCI buses on your
3800 rom [X86] Assign address space to expansion ROMs.
3801 Use with caution as certain devices share
3802 address decoders between ROMs and other
3804 norom [X86] Do not assign address space to
3805 expansion ROMs that do not already have
3806 BIOS assigned address ranges.
3807 nobar [X86] Do not assign address space to the
3808 BARs that weren't assigned by the BIOS.
3809 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be
3810 assigned automatically to PCI devices. You can
3811 make the kernel exclude IRQs of your ISA cards
3813 pirqaddr=0xAAAAA [X86] Specify the physical address
3814 of the PIRQ table (normally generated
3815 by the BIOS) if it is outside the
3816 F0000h-100000h range.
3817 lastbus=N [X86] Scan all buses thru bus #N. Can be
3818 useful if the kernel is unable to find your
3819 secondary buses and you want to tell it
3820 explicitly which ones they are.
3821 assign-busses [X86] Always assign all PCI bus
3822 numbers ourselves, overriding
3823 whatever the firmware may have done.
3824 usepirqmask [X86] Honor the possible IRQ mask stored
3825 in the BIOS $PIR table. This is needed on
3826 some systems with broken BIOSes, notably
3827 some HP Pavilion N5400 and Omnibook XE3
3828 notebooks. This will have no effect if ACPI
3829 IRQ routing is enabled.
3830 noacpi [X86] Do not use ACPI for IRQ routing
3831 or for PCI scanning.
3832 use_crs [X86] Use PCI host bridge window information
3833 from ACPI. On BIOSes from 2008 or later, this
3834 is enabled by default. If you need to use this,
3835 please report a bug.
3836 nocrs [X86] Ignore PCI host bridge windows from ACPI.
3837 If you need to use this, please report a bug.
3838 routeirq Do IRQ routing for all PCI devices.
3839 This is normally done in pci_enable_device(),
3840 so this option is a temporary workaround
3841 for broken drivers that don't call it.
3842 skip_isa_align [X86] do not align io start addr, so can
3843 handle more pci cards
3844 noearly [X86] Don't do any early type 1 scanning.
3845 This might help on some broken boards which
3846 machine check when some devices' config space
3847 is read. But various workarounds are disabled
3848 and some IOMMU drivers will not work.
3849 bfsort Sort PCI devices into breadth-first order.
3850 This sorting is done to get a device
3851 order compatible with older (<= 2.4) kernels.
3852 nobfsort Don't sort PCI devices into breadth-first order.
3853 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size)
3854 tuning and use the BIOS-configured MPS defaults.
3855 pcie_bus_safe Set every device's MPS to the largest value
3856 supported by all devices below the root complex.
3857 pcie_bus_perf Set device MPS to the largest allowable MPS
3858 based on its parent bus. Also set MRRS (Max
3859 Read Request Size) to the largest supported
3860 value (no larger than the MPS that the device
3861 or bus can support) for best performance.
3862 pcie_bus_peer2peer Set every device's MPS to 128B, which
3863 every device is guaranteed to support. This
3864 configuration allows peer-to-peer DMA between
3865 any pair of devices, possibly at the cost of
3866 reduced performance. This also guarantees
3867 that hot-added devices will work.
3868 cbiosize=nn[KMG] The fixed amount of bus space which is
3869 reserved for the CardBus bridge's IO window.
3870 The default value is 256 bytes.
3871 cbmemsize=nn[KMG] The fixed amount of bus space which is
3872 reserved for the CardBus bridge's memory
3873 window. The default value is 64 megabytes.
3876 [<order of align>@]<pci_dev>[; ...]
3877 Specifies alignment and device to reassign
3878 aligned memory resources. How to
3879 specify the device is described above.
3880 If <order of align> is not specified,
3881 PAGE_SIZE is used as alignment.
3882 A PCI-PCI bridge can be specified if resource
3883 windows need to be expanded.
3884 To specify the alignment for several
3885 instances of a device, the PCI vendor,
3886 device, subvendor, and subdevice may be
3887 specified, e.g., 12@pci:8086:9c22:103c:198f
3888 for 4096-byte alignment.
3889 ecrc= Enable/disable PCIe ECRC (transaction layer
3890 end-to-end CRC checking).
3891 bios: Use BIOS/firmware settings. This is the
3895 hpiosize=nn[KMG] The fixed amount of bus space which is
3896 reserved for hotplug bridge's IO window.
3897 Default size is 256 bytes.
3898 hpmmiosize=nn[KMG] The fixed amount of bus space which is
3899 reserved for hotplug bridge's MMIO window.
3900 Default size is 2 megabytes.
3901 hpmmioprefsize=nn[KMG] The fixed amount of bus space which is
3902 reserved for hotplug bridge's MMIO_PREF window.
3903 Default size is 2 megabytes.
3904 hpmemsize=nn[KMG] The fixed amount of bus space which is
3905 reserved for hotplug bridge's MMIO and
3907 Default size is 2 megabytes.
3908 hpbussize=nn The minimum amount of additional bus numbers
3909 reserved for buses below a hotplug bridge.
3911 realloc= Enable/disable reallocating PCI bridge resources
3912 if allocations done by BIOS are too small to
3913 accommodate resources required by all child
3915 off: Turn realloc off
3917 realloc same as realloc=on
3918 noari do not use PCIe ARI.
3919 noats [PCIE, Intel-IOMMU, AMD-IOMMU]
3920 do not use PCIe ATS (and IOMMU device IOTLB).
3921 pcie_scan_all Scan all possible PCIe devices. Otherwise we
3922 only look for one device below a PCIe downstream
3924 big_root_window Try to add a big 64bit memory window to the PCIe
3925 root complex on AMD CPUs. Some GFX hardware
3926 can resize a BAR to allow access to all VRAM.
3927 Adding the window is slightly risky (it may
3928 conflict with unreported devices), so this
3930 disable_acs_redir=<pci_dev>[; ...]
3931 Specify one or more PCI devices (in the format
3932 specified above) separated by semicolons.
3933 Each device specified will have the PCI ACS
3934 redirect capabilities forced off which will
3935 allow P2P traffic between devices through
3936 bridges without forcing it upstream. Note:
3937 this removes isolation between devices and
3938 may put more devices in an IOMMU group.
3939 force_floating [S390] Force usage of floating interrupts.
3940 nomio [S390] Do not use MIO instructions.
3941 norid [S390] ignore the RID field and force use of
3942 one PCI domain per PCI function
3944 pcie_aspm= [PCIE] Forcibly enable or disable PCIe Active State Power
3947 force Enable ASPM even on devices that claim not to support it.
3948 WARNING: Forcing ASPM on may cause system lockups.
3950 pcie_ports= [PCIE] PCIe port services handling:
3951 native Use native PCIe services (PME, AER, DPC, PCIe hotplug)
3952 even if the platform doesn't give the OS permission to
3953 use them. This may cause conflicts if the platform
3954 also tries to use these services.
3955 dpc-native Use native PCIe service for DPC only. May
3956 cause conflicts if firmware uses AER or DPC.
3957 compat Disable native PCIe services (PME, AER, DPC, PCIe
3960 pcie_port_pm= [PCIE] PCIe port power management handling:
3961 off Disable power management of all PCIe ports
3962 force Forcibly enable power management of all PCIe ports
3964 pcie_pme= [PCIE,PM] Native PCIe PME signaling options:
3965 nomsi Do not use MSI for native PCIe PME signaling (this makes
3966 all PCIe root ports use INTx for all services).
3968 pcmv= [HW,PCMCIA] BadgePAD 4
3972 Keep all power-domains already enabled by bootloader on,
3973 even if no driver has claimed them. This is useful
3974 for debug and development, but should not be
3975 needed on a platform with proper driver support.
3978 See Documentation/admin-guide/blockdev/paride.rst.
3980 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at
3983 See arch/parisc/kernel/pdc_chassis.c
3985 percpu_alloc= Select which percpu first chunk allocator to use.
3986 Currently supported values are "embed" and "page".
3987 Archs may support subset or none of the selections.
3988 See comments in mm/percpu.c for details on each
3989 allocator. This parameter is primarily for debugging
3990 and performance comparison.
3993 See Documentation/admin-guide/blockdev/paride.rst.
3996 See Documentation/admin-guide/blockdev/paride.rst.
3998 pirq= [SMP,APIC] Manual mp-table setup
3999 See Documentation/x86/i386/IO-APIC.rst.
4001 plip= [PPT,NET] Parallel port network link
4002 Format: { parport<nr> | timid | 0 }
4003 See also Documentation/admin-guide/parport.rst.
4005 pmtmr= [X86] Manual setup of pmtmr I/O Port.
4006 Override pmtimer IOPort with a hex value.
4009 pm_debug_messages [SUSPEND,KNL]
4010 Enable suspend/resume debug messages during boot up.
4013 Enable PNP debug messages (depends on the
4014 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time
4015 via /sys/module/pnp/parameters/debug. We always show
4016 current resource usage; turning this on also shows
4017 possible settings and some assignment information.
4023 { on | off | curr | res | no-curr | no-res }
4026 [ISAPNP] Exclude IRQs for the autoconfiguration
4029 [ISAPNP] Exclude DMAs for the autoconfiguration
4031 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration
4032 Ranges are in pairs (I/O port base and size).
4035 [ISAPNP] Exclude memory regions for the
4037 Ranges are in pairs (memory base and size).
4039 ports= [IP_VS_FTP] IPVS ftp helper module
4041 Up to 8 (IP_VS_APP_MAX_PORTS) ports
4043 Format: <port>,<port>....
4045 powersave=off [PPC] This option disables power saving features.
4046 It specifically disables cpuidle and sets the
4047 platform machine description specific power_save
4048 function to NULL. On Idle the CPU just reduces
4051 ppc_strict_facility_enable
4052 [PPC] This option catches any kernel floating point,
4053 Altivec, VSX and SPE outside of regions specifically
4054 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()).
4055 There is some performance impact when enabling this.
4059 Disable Hardware Transactional Memory
4062 Select preemption mode if you have CONFIG_PREEMPT_DYNAMIC
4063 none - Limited to cond_resched() calls
4064 voluntary - Limited to cond_resched() and might_sleep() calls
4065 full - Any section that isn't explicitly preempt disabled
4066 can be preempted anytime.
4068 print-fatal-signals=
4069 [KNL] debug: print fatal signals
4071 If enabled, warn about various signal handling
4072 related application anomalies: too many signals,
4073 too many POSIX.1 timers, fatal signals causing a
4076 If you hit the warning due to signal overflow,
4077 you might want to try "ulimit -i unlimited".
4081 printk.always_kmsg_dump=
4082 Trigger kmsg_dump for cases other than kernel oops or
4084 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
4087 printk.devkmsg={on,off,ratelimit}
4088 Control writing to /dev/kmsg.
4089 on - unlimited logging to /dev/kmsg from userspace
4090 off - logging to /dev/kmsg disabled
4091 ratelimit - ratelimit the logging
4094 printk.time= Show timing data prefixed to each printk message line
4095 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
4097 processor.max_cstate= [HW,ACPI]
4098 Limit processor to maximum C-state
4099 max_cstate=9 overrides any DMI blacklist limit.
4101 processor.nocst [HW,ACPI]
4102 Ignore the _CST method to determine C-states,
4103 instead using the legacy FADT method
4105 profile= [KNL] Enable kernel profiling via /proc/profile
4106 Format: [<profiletype>,]<number>
4107 Param: <profiletype>: "schedule", "sleep", or "kvm"
4108 [defaults to kernel profiling]
4109 Param: "schedule" - profile schedule points.
4110 Param: "sleep" - profile D-state sleeping (millisecs).
4111 Requires CONFIG_SCHEDSTATS
4112 Param: "kvm" - profile VM exits.
4113 Param: <number> - step/bucket size as a power of 2 for
4114 statistical time based profiling.
4116 prompt_ramdisk= [RAM] [Deprecated]
4118 prot_virt= [S390] enable hosting protected virtual machines
4119 isolated from the hypervisor (if hardware supports
4123 psi= [KNL] Enable or disable pressure stall information
4127 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to
4128 probe for; one of (bare|imps|exps|lifebook|any).
4129 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports
4131 psmouse.resetafter= [HW,MOUSE]
4132 Try to reset the device after so many bad packets
4135 [HW,MOUSE] Set desired mouse resolution, in dpi.
4136 psmouse.smartscroll=
4137 [HW,MOUSE] Controls Logitech smartscroll autorepeat.
4138 0 = disabled, 1 = enabled (default).
4140 pstore.backend= Specify the name of the pstore backend to use
4143 See Documentation/admin-guide/blockdev/paride.rst.
4145 pti= [X86-64] Control Page Table Isolation of user and
4146 kernel address spaces. Disabling this feature
4147 removes hardening, but improves performance of
4148 system calls and interrupts.
4150 on - unconditionally enable
4151 off - unconditionally disable
4152 auto - kernel detects whether your CPU model is
4153 vulnerable to issues that PTI mitigates
4155 Not specifying this option is equivalent to pti=auto.
4158 Equivalent to pti=off
4161 [KNL] Number of legacy pty's. Overwrites compiled-in
4164 quiet [KNL] Disable most log messages
4169 See Documentation/admin-guide/md.rst.
4171 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes
4172 See Documentation/admin-guide/blockdev/ramdisk.rst.
4174 ramdisk_start= [RAM] RAM disk image start address
4176 random.trust_cpu={on,off}
4177 [KNL] Enable or disable trusting the use of the
4178 CPU's random number generator (if available) to
4179 fully seed the kernel's CRNG. Default is controlled
4180 by CONFIG_RANDOM_TRUST_CPU.
4182 randomize_kstack_offset=
4183 [KNL] Enable or disable kernel stack offset
4184 randomization, which provides roughly 5 bits of
4185 entropy, frustrating memory corruption attacks
4186 that depend on stack address determinism or
4187 cross-syscall address exposures. This is only
4188 available on architectures that have defined
4189 CONFIG_HAVE_ARCH_RANDOMIZE_KSTACK_OFFSET.
4190 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
4191 Default is CONFIG_RANDOMIZE_KSTACK_OFFSET_DEFAULT.
4193 ras=option[,option,...] [KNL] RAS-specific options
4196 Disable the Correctable Errors Collector,
4197 see CONFIG_RAS_CEC help text.
4200 The argument is a cpu list, as described above.
4202 In kernels built with CONFIG_RCU_NOCB_CPU=y, set
4203 the specified list of CPUs to be no-callback CPUs.
4204 Invocation of these CPUs' RCU callbacks will be
4205 offloaded to "rcuox/N" kthreads created for that
4206 purpose, where "x" is "p" for RCU-preempt, and
4207 "s" for RCU-sched, and "N" is the CPU number.
4208 This reduces OS jitter on the offloaded CPUs,
4209 which can be useful for HPC and real-time
4210 workloads. It can also improve energy efficiency
4211 for asymmetric multiprocessors.
4214 Rather than requiring that offloaded CPUs
4215 (specified by rcu_nocbs= above) explicitly
4216 awaken the corresponding "rcuoN" kthreads,
4217 make these kthreads poll for callbacks.
4218 This improves the real-time response for the
4219 offloaded CPUs by relieving them of the need to
4220 wake up the corresponding kthread, but degrades
4221 energy efficiency by requiring that the kthreads
4222 periodically wake up to do the polling.
4224 rcutree.blimit= [KNL]
4225 Set maximum number of finished RCU callbacks to
4226 process in one batch.
4228 rcutree.dump_tree= [KNL]
4229 Dump the structure of the rcu_node combining tree
4230 out at early boot. This is used for diagnostic
4231 purposes, to verify correct tree setup.
4233 rcutree.gp_cleanup_delay= [KNL]
4234 Set the number of jiffies to delay each step of
4235 RCU grace-period cleanup.
4237 rcutree.gp_init_delay= [KNL]
4238 Set the number of jiffies to delay each step of
4239 RCU grace-period initialization.
4241 rcutree.gp_preinit_delay= [KNL]
4242 Set the number of jiffies to delay each step of
4243 RCU grace-period pre-initialization, that is,
4244 the propagation of recent CPU-hotplug changes up
4245 the rcu_node combining tree.
4247 rcutree.use_softirq= [KNL]
4248 If set to zero, move all RCU_SOFTIRQ processing to
4249 per-CPU rcuc kthreads. Defaults to a non-zero
4250 value, meaning that RCU_SOFTIRQ is used by default.
4251 Specify rcutree.use_softirq=0 to use rcuc kthreads.
4253 But note that CONFIG_PREEMPT_RT=y kernels disable
4254 this kernel boot parameter, forcibly setting it
4257 rcutree.rcu_fanout_exact= [KNL]
4258 Disable autobalancing of the rcu_node combining
4259 tree. This is used by rcutorture, and might
4260 possibly be useful for architectures having high
4261 cache-to-cache transfer latencies.
4263 rcutree.rcu_fanout_leaf= [KNL]
4264 Change the number of CPUs assigned to each
4265 leaf rcu_node structure. Useful for very
4266 large systems, which will choose the value 64,
4267 and for NUMA systems with large remote-access
4268 latencies, which will choose a value aligned
4269 with the appropriate hardware boundaries.
4271 rcutree.rcu_min_cached_objs= [KNL]
4272 Minimum number of objects which are cached and
4273 maintained per one CPU. Object size is equal
4274 to PAGE_SIZE. The cache allows to reduce the
4275 pressure to page allocator, also it makes the
4276 whole algorithm to behave better in low memory
4279 rcutree.jiffies_till_first_fqs= [KNL]
4280 Set delay from grace-period initialization to
4281 first attempt to force quiescent states.
4282 Units are jiffies, minimum value is zero,
4283 and maximum value is HZ.
4285 rcutree.jiffies_till_next_fqs= [KNL]
4286 Set delay between subsequent attempts to force
4287 quiescent states. Units are jiffies, minimum
4288 value is one, and maximum value is HZ.
4290 rcutree.jiffies_till_sched_qs= [KNL]
4291 Set required age in jiffies for a
4292 given grace period before RCU starts
4293 soliciting quiescent-state help from
4294 rcu_note_context_switch() and cond_resched().
4295 If not specified, the kernel will calculate
4296 a value based on the most recent settings
4297 of rcutree.jiffies_till_first_fqs
4298 and rcutree.jiffies_till_next_fqs.
4299 This calculated value may be viewed in
4300 rcutree.jiffies_to_sched_qs. Any attempt to set
4301 rcutree.jiffies_to_sched_qs will be cheerfully
4304 rcutree.kthread_prio= [KNL,BOOT]
4305 Set the SCHED_FIFO priority of the RCU per-CPU
4306 kthreads (rcuc/N). This value is also used for
4307 the priority of the RCU boost threads (rcub/N)
4308 and for the RCU grace-period kthreads (rcu_bh,
4309 rcu_preempt, and rcu_sched). If RCU_BOOST is
4310 set, valid values are 1-99 and the default is 1
4311 (the least-favored priority). Otherwise, when
4312 RCU_BOOST is not set, valid values are 0-99 and
4313 the default is zero (non-realtime operation).
4315 rcutree.rcu_nocb_gp_stride= [KNL]
4316 Set the number of NOCB callback kthreads in
4317 each group, which defaults to the square root
4318 of the number of CPUs. Larger numbers reduce
4319 the wakeup overhead on the global grace-period
4320 kthread, but increases that same overhead on
4321 each group's NOCB grace-period kthread.
4323 rcutree.qhimark= [KNL]
4324 Set threshold of queued RCU callbacks beyond which
4325 batch limiting is disabled.
4327 rcutree.qlowmark= [KNL]
4328 Set threshold of queued RCU callbacks below which
4329 batch limiting is re-enabled.
4331 rcutree.qovld= [KNL]
4332 Set threshold of queued RCU callbacks beyond which
4333 RCU's force-quiescent-state scan will aggressively
4334 enlist help from cond_resched() and sched IPIs to
4335 help CPUs more quickly reach quiescent states.
4336 Set to less than zero to make this be set based
4337 on rcutree.qhimark at boot time and to zero to
4338 disable more aggressive help enlistment.
4340 rcutree.rcu_idle_gp_delay= [KNL]
4341 Set wakeup interval for idle CPUs that have
4342 RCU callbacks (RCU_FAST_NO_HZ=y).
4344 rcutree.rcu_kick_kthreads= [KNL]
4345 Cause the grace-period kthread to get an extra
4346 wake_up() if it sleeps three times longer than
4347 it should at force-quiescent-state time.
4348 This wake_up() will be accompanied by a
4349 WARN_ONCE() splat and an ftrace_dump().
4351 rcutree.rcu_unlock_delay= [KNL]
4352 In CONFIG_RCU_STRICT_GRACE_PERIOD=y kernels,
4353 this specifies an rcu_read_unlock()-time delay
4354 in microseconds. This defaults to zero.
4355 Larger delays increase the probability of
4356 catching RCU pointer leaks, that is, buggy use
4357 of RCU-protected pointers after the relevant
4358 rcu_read_unlock() has completed.
4360 rcutree.sysrq_rcu= [KNL]
4361 Commandeer a sysrq key to dump out Tree RCU's
4362 rcu_node tree with an eye towards determining
4363 why a new grace period has not yet started.
4365 rcuscale.gp_async= [KNL]
4366 Measure performance of asynchronous
4367 grace-period primitives such as call_rcu().
4369 rcuscale.gp_async_max= [KNL]
4370 Specify the maximum number of outstanding
4371 callbacks per writer thread. When a writer
4372 thread exceeds this limit, it invokes the
4373 corresponding flavor of rcu_barrier() to allow
4374 previously posted callbacks to drain.
4376 rcuscale.gp_exp= [KNL]
4377 Measure performance of expedited synchronous
4378 grace-period primitives.
4380 rcuscale.holdoff= [KNL]
4381 Set test-start holdoff period. The purpose of
4382 this parameter is to delay the start of the
4383 test until boot completes in order to avoid
4386 rcuscale.kfree_rcu_test= [KNL]
4387 Set to measure performance of kfree_rcu() flooding.
4389 rcuscale.kfree_rcu_test_double= [KNL]
4390 Test the double-argument variant of kfree_rcu().
4391 If this parameter has the same value as
4392 rcuscale.kfree_rcu_test_single, both the single-
4393 and double-argument variants are tested.
4395 rcuscale.kfree_rcu_test_single= [KNL]
4396 Test the single-argument variant of kfree_rcu().
4397 If this parameter has the same value as
4398 rcuscale.kfree_rcu_test_double, both the single-
4399 and double-argument variants are tested.
4401 rcuscale.kfree_nthreads= [KNL]
4402 The number of threads running loops of kfree_rcu().
4404 rcuscale.kfree_alloc_num= [KNL]
4405 Number of allocations and frees done in an iteration.
4407 rcuscale.kfree_loops= [KNL]
4408 Number of loops doing rcuscale.kfree_alloc_num number
4409 of allocations and frees.
4411 rcuscale.nreaders= [KNL]
4412 Set number of RCU readers. The value -1 selects
4413 N, where N is the number of CPUs. A value
4414 "n" less than -1 selects N-n+1, where N is again
4415 the number of CPUs. For example, -2 selects N
4416 (the number of CPUs), -3 selects N+1, and so on.
4417 A value of "n" less than or equal to -N selects
4420 rcuscale.nwriters= [KNL]
4421 Set number of RCU writers. The values operate
4422 the same as for rcuscale.nreaders.
4423 N, where N is the number of CPUs
4425 rcuscale.perf_type= [KNL]
4426 Specify the RCU implementation to test.
4428 rcuscale.shutdown= [KNL]
4429 Shut the system down after performance tests
4430 complete. This is useful for hands-off automated
4433 rcuscale.verbose= [KNL]
4434 Enable additional printk() statements.
4436 rcuscale.writer_holdoff= [KNL]
4437 Write-side holdoff between grace periods,
4438 in microseconds. The default of zero says
4441 rcutorture.fqs_duration= [KNL]
4442 Set duration of force_quiescent_state bursts
4445 rcutorture.fqs_holdoff= [KNL]
4446 Set holdoff time within force_quiescent_state bursts
4449 rcutorture.fqs_stutter= [KNL]
4450 Set wait time between force_quiescent_state bursts
4453 rcutorture.fwd_progress= [KNL]
4454 Enable RCU grace-period forward-progress testing
4455 for the types of RCU supporting this notion.
4457 rcutorture.fwd_progress_div= [KNL]
4458 Specify the fraction of a CPU-stall-warning
4459 period to do tight-loop forward-progress testing.
4461 rcutorture.fwd_progress_holdoff= [KNL]
4462 Number of seconds to wait between successive
4463 forward-progress tests.
4465 rcutorture.fwd_progress_need_resched= [KNL]
4466 Enclose cond_resched() calls within checks for
4467 need_resched() during tight-loop forward-progress
4470 rcutorture.gp_cond= [KNL]
4471 Use conditional/asynchronous update-side
4472 primitives, if available.
4474 rcutorture.gp_exp= [KNL]
4475 Use expedited update-side primitives, if available.
4477 rcutorture.gp_normal= [KNL]
4478 Use normal (non-expedited) asynchronous
4479 update-side primitives, if available.
4481 rcutorture.gp_sync= [KNL]
4482 Use normal (non-expedited) synchronous
4483 update-side primitives, if available. If all
4484 of rcutorture.gp_cond=, rcutorture.gp_exp=,
4485 rcutorture.gp_normal=, and rcutorture.gp_sync=
4486 are zero, rcutorture acts as if is interpreted
4487 they are all non-zero.
4489 rcutorture.irqreader= [KNL]
4490 Run RCU readers from irq handlers, or, more
4491 accurately, from a timer handler. Not all RCU
4492 flavors take kindly to this sort of thing.
4494 rcutorture.leakpointer= [KNL]
4495 Leak an RCU-protected pointer out of the reader.
4496 This can of course result in splats, and is
4497 intended to test the ability of things like
4498 CONFIG_RCU_STRICT_GRACE_PERIOD=y to detect
4501 rcutorture.n_barrier_cbs= [KNL]
4502 Set callbacks/threads for rcu_barrier() testing.
4504 rcutorture.nfakewriters= [KNL]
4505 Set number of concurrent RCU writers. These just
4506 stress RCU, they don't participate in the actual
4507 test, hence the "fake".
4509 rcutorture.nocbs_nthreads= [KNL]
4510 Set number of RCU callback-offload togglers.
4511 Zero (the default) disables toggling.
4513 rcutorture.nocbs_toggle= [KNL]
4514 Set the delay in milliseconds between successive
4515 callback-offload toggling attempts.
4517 rcutorture.nreaders= [KNL]
4518 Set number of RCU readers. The value -1 selects
4519 N-1, where N is the number of CPUs. A value
4520 "n" less than -1 selects N-n-2, where N is again
4521 the number of CPUs. For example, -2 selects N
4522 (the number of CPUs), -3 selects N+1, and so on.
4524 rcutorture.object_debug= [KNL]
4525 Enable debug-object double-call_rcu() testing.
4527 rcutorture.onoff_holdoff= [KNL]
4528 Set time (s) after boot for CPU-hotplug testing.
4530 rcutorture.onoff_interval= [KNL]
4531 Set time (jiffies) between CPU-hotplug operations,
4532 or zero to disable CPU-hotplug testing.
4534 rcutorture.read_exit= [KNL]
4535 Set the number of read-then-exit kthreads used
4536 to test the interaction of RCU updaters and
4537 task-exit processing.
4539 rcutorture.read_exit_burst= [KNL]
4540 The number of times in a given read-then-exit
4541 episode that a set of read-then-exit kthreads
4544 rcutorture.read_exit_delay= [KNL]
4545 The delay, in seconds, between successive
4546 read-then-exit testing episodes.
4548 rcutorture.shuffle_interval= [KNL]
4549 Set task-shuffle interval (s). Shuffling tasks
4550 allows some CPUs to go into dyntick-idle mode
4551 during the rcutorture test.
4553 rcutorture.shutdown_secs= [KNL]
4554 Set time (s) after boot system shutdown. This
4555 is useful for hands-off automated testing.
4557 rcutorture.stall_cpu= [KNL]
4558 Duration of CPU stall (s) to test RCU CPU stall
4559 warnings, zero to disable.
4561 rcutorture.stall_cpu_block= [KNL]
4562 Sleep while stalling if set. This will result
4563 in warnings from preemptible RCU in addition
4564 to any other stall-related activity.
4566 rcutorture.stall_cpu_holdoff= [KNL]
4567 Time to wait (s) after boot before inducing stall.
4569 rcutorture.stall_cpu_irqsoff= [KNL]
4570 Disable interrupts while stalling if set.
4572 rcutorture.stall_gp_kthread= [KNL]
4573 Duration (s) of forced sleep within RCU
4574 grace-period kthread to test RCU CPU stall
4575 warnings, zero to disable. If both stall_cpu
4576 and stall_gp_kthread are specified, the
4577 kthread is starved first, then the CPU.
4579 rcutorture.stat_interval= [KNL]
4580 Time (s) between statistics printk()s.
4582 rcutorture.stutter= [KNL]
4583 Time (s) to stutter testing, for example, specifying
4584 five seconds causes the test to run for five seconds,
4585 wait for five seconds, and so on. This tests RCU's
4586 ability to transition abruptly to and from idle.
4588 rcutorture.test_boost= [KNL]
4589 Test RCU priority boosting? 0=no, 1=maybe, 2=yes.
4590 "Maybe" means test if the RCU implementation
4591 under test support RCU priority boosting.
4593 rcutorture.test_boost_duration= [KNL]
4594 Duration (s) of each individual boost test.
4596 rcutorture.test_boost_interval= [KNL]
4597 Interval (s) between each boost test.
4599 rcutorture.test_no_idle_hz= [KNL]
4600 Test RCU's dyntick-idle handling. See also the
4601 rcutorture.shuffle_interval parameter.
4603 rcutorture.torture_type= [KNL]
4604 Specify the RCU implementation to test.
4606 rcutorture.verbose= [KNL]
4607 Enable additional printk() statements.
4609 rcupdate.rcu_cpu_stall_ftrace_dump= [KNL]
4610 Dump ftrace buffer after reporting RCU CPU
4613 rcupdate.rcu_cpu_stall_suppress= [KNL]
4614 Suppress RCU CPU stall warning messages.
4616 rcupdate.rcu_cpu_stall_suppress_at_boot= [KNL]
4617 Suppress RCU CPU stall warning messages and
4618 rcutorture writer stall warnings that occur
4619 during early boot, that is, during the time
4620 before the init task is spawned.
4622 rcupdate.rcu_cpu_stall_timeout= [KNL]
4623 Set timeout for RCU CPU stall warning messages.
4625 rcupdate.rcu_expedited= [KNL]
4626 Use expedited grace-period primitives, for
4627 example, synchronize_rcu_expedited() instead
4628 of synchronize_rcu(). This reduces latency,
4629 but can increase CPU utilization, degrade
4630 real-time latency, and degrade energy efficiency.
4631 No effect on CONFIG_TINY_RCU kernels.
4633 rcupdate.rcu_normal= [KNL]
4634 Use only normal grace-period primitives,
4635 for example, synchronize_rcu() instead of
4636 synchronize_rcu_expedited(). This improves
4637 real-time latency, CPU utilization, and
4638 energy efficiency, but can expose users to
4639 increased grace-period latency. This parameter
4640 overrides rcupdate.rcu_expedited. No effect on
4641 CONFIG_TINY_RCU kernels.
4643 rcupdate.rcu_normal_after_boot= [KNL]
4644 Once boot has completed (that is, after
4645 rcu_end_inkernel_boot() has been invoked), use
4646 only normal grace-period primitives. No effect
4647 on CONFIG_TINY_RCU kernels.
4649 But note that CONFIG_PREEMPT_RT=y kernels enables
4650 this kernel boot parameter, forcibly setting
4651 it to the value one, that is, converting any
4652 post-boot attempt at an expedited RCU grace
4653 period to instead use normal non-expedited
4654 grace-period processing.
4656 rcupdate.rcu_task_ipi_delay= [KNL]
4657 Set time in jiffies during which RCU tasks will
4658 avoid sending IPIs, starting with the beginning
4659 of a given grace period. Setting a large
4660 number avoids disturbing real-time workloads,
4661 but lengthens grace periods.
4663 rcupdate.rcu_task_stall_timeout= [KNL]
4664 Set timeout in jiffies for RCU task stall warning
4665 messages. Disable with a value less than or equal
4668 rcupdate.rcu_self_test= [KNL]
4669 Run the RCU early boot self tests
4673 Run specified binary instead of /init from the ramdisk,
4674 used for early userspace startup. See initrd.
4677 force - Override the decision by the kernel to hide the
4678 advertisement of RDRAND support (this affects
4679 certain AMD processors because of buggy BIOS
4680 support, specifically around the suspend/resume
4684 Turn on/off individual RDT features. List is:
4685 cmt, mbmtotal, mbmlocal, l3cat, l3cdp, l2cat, l2cdp,
4687 E.g. to turn on cmt and turn off mba use:
4691 Format (x86 or x86_64):
4692 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] \
4694 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \
4696 Where reboot_mode is one of warm (soft) or cold (hard) or gpio
4697 (prefix with 'panic_' to set mode for panic
4699 reboot_type is one of bios, acpi, kbd, triple, efi, or pci,
4700 reboot_force is either force or not specified,
4701 reboot_cpu is s[mp]#### with #### being the processor
4702 to be used for rebooting.
4704 refscale.holdoff= [KNL]
4705 Set test-start holdoff period. The purpose of
4706 this parameter is to delay the start of the
4707 test until boot completes in order to avoid
4710 refscale.loops= [KNL]
4711 Set the number of loops over the synchronization
4712 primitive under test. Increasing this number
4713 reduces noise due to loop start/end overhead,
4714 but the default has already reduced the per-pass
4715 noise to a handful of picoseconds on ca. 2020
4718 refscale.nreaders= [KNL]
4719 Set number of readers. The default value of -1
4720 selects N, where N is roughly 75% of the number
4721 of CPUs. A value of zero is an interesting choice.
4723 refscale.nruns= [KNL]
4724 Set number of runs, each of which is dumped onto
4727 refscale.readdelay= [KNL]
4728 Set the read-side critical-section duration,
4729 measured in microseconds.
4731 refscale.scale_type= [KNL]
4732 Specify the read-protection implementation to test.
4734 refscale.shutdown= [KNL]
4735 Shut down the system at the end of the performance
4736 test. This defaults to 1 (shut it down) when
4737 refscale is built into the kernel and to 0 (leave
4738 it running) when refscale is built as a module.
4740 refscale.verbose= [KNL]
4741 Enable additional printk() statements.
4743 refscale.verbose_batched= [KNL]
4744 Batch the additional printk() statements. If zero
4745 (the default) or negative, print everything. Otherwise,
4746 print every Nth verbose statement, where N is the value
4750 [KNL, SMP] Set scheduler's default relax_domain_level.
4751 See Documentation/admin-guide/cgroup-v1/cpusets.rst.
4753 reserve= [KNL,BUGS] Force kernel to ignore I/O ports or memory
4754 Format: <base1>,<size1>[,<base2>,<size2>,...]
4755 Reserve I/O ports or memory so the kernel won't use
4756 them. If <base> is less than 0x10000, the region
4757 is assumed to be I/O ports; otherwise it is memory.
4759 reservetop= [X86-32]
4761 Reserves a hole at the top of the kernel virtual
4766 Set the amount of memory to reserve for BIOS at
4767 the bottom of the address space.
4769 reset_devices [KNL] Force drivers to reset the underlying device
4770 during initialization.
4773 Specify the partition device for software suspend
4775 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>}
4777 resume_offset= [SWSUSP]
4778 Specify the offset from the beginning of the partition
4779 given by "resume=" at which the swap header is located,
4780 in <PAGE_SIZE> units (needed only for swap files).
4781 See Documentation/power/swsusp-and-swap-files.rst
4783 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to
4784 read the resume files
4786 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up.
4787 Useful for devices that are detected asynchronously
4788 (e.g. USB and MMC devices).
4790 hibernate= [HIBERNATION]
4791 noresume Don't check if there's a hibernation image
4792 present during boot.
4793 nocompress Don't compress/decompress hibernation images.
4794 no Disable hibernation and resume.
4795 protect_image Turn on image protection during restoration
4796 (that will set all pages holding image data
4797 during restoration read-only).
4799 retain_initrd [RAM] Keep initrd memory after extraction
4801 rfkill.default_state=
4802 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm,
4803 etc. communication is blocked by default.
4806 rfkill.master_switch_mode=
4807 0 The "airplane mode" button does nothing.
4808 1 The "airplane mode" button toggles between everything
4809 blocked and the previous configuration.
4810 2 The "airplane mode" button toggles between everything
4811 blocked and everything unblocked.
4813 rhash_entries= [KNL,NET]
4814 Set number of hash buckets for route cache
4817 [KNL] Disable ring 3 MONITOR/MWAIT feature on supported
4820 ro [KNL] Mount root device read-only on boot
4823 on Mark read-only kernel memory as read-only (default).
4824 off Leave read-only kernel memory writable for debugging.
4827 Enable the uart passthrough on the designated usb port
4828 on Rockchip SoCs. When active, the signals of the
4829 debug-uart get routed to the D+ and D- pins of the usb
4830 port and the regular usb controller gets disabled.
4832 root= [KNL] Root filesystem
4833 See name_to_dev_t comment in init/do_mounts.c.
4835 rootdelay= [KNL] Delay (in seconds) to pause before attempting to
4836 mount the root filesystem
4838 rootflags= [KNL] Set root filesystem mount option string
4840 rootfstype= [KNL] Set root filesystem type
4842 rootwait [KNL] Wait (indefinitely) for root device to show up.
4843 Useful for devices that are detected asynchronously
4844 (e.g. USB and MMC devices).
4846 rproc_mem=nn[KMG][@address]
4847 [KNL,ARM,CMA] Remoteproc physical memory block.
4848 Memory area to be used by remote processor image,
4851 rw [KNL] Mount root device read-write on boot
4853 S [KNL] Run init in single mode
4855 s390_iommu= [HW,S390]
4856 Set s390 IOTLB flushing mode
4858 With strict flushing every unmap operation will result in
4859 an IOTLB flush. Default is lazy flushing before reuse,
4863 See drivers/net/irda/sa1100_ir.c.
4865 sbni= [NET] Granch SBNI12 leased line adapter
4867 sched_verbose [KNL] Enables verbose scheduler debug messages.
4869 schedstats= [KNL,X86] Enable or disable scheduled statistics.
4870 Allowed values are enable and disable. This feature
4871 incurs a small amount of overhead in the scheduler
4872 but is useful for debugging and performance tuning.
4874 sched_thermal_decay_shift=
4875 [KNL, SMP] Set a decay shift for scheduler thermal
4876 pressure signal. Thermal pressure signal follows the
4877 default decay period of other scheduler pelt
4878 signals(usually 32 ms but configurable). Setting
4879 sched_thermal_decay_shift will left shift the decay
4880 period for the thermal pressure signal by the shift
4882 i.e. with the default pelt decay period of 32 ms
4883 sched_thermal_decay_shift thermal pressure decay pr
4887 Format: integer between 0 and 10
4890 scftorture.holdoff= [KNL]
4891 Number of seconds to hold off before starting
4892 test. Defaults to zero for module insertion and
4893 to 10 seconds for built-in smp_call_function()
4896 scftorture.longwait= [KNL]
4897 Request ridiculously long waits randomly selected
4898 up to the chosen limit in seconds. Zero (the
4899 default) disables this feature. Please note
4900 that requesting even small non-zero numbers of
4901 seconds can result in RCU CPU stall warnings,
4902 softlockup complaints, and so on.
4904 scftorture.nthreads= [KNL]
4905 Number of kthreads to spawn to invoke the
4906 smp_call_function() family of functions.
4907 The default of -1 specifies a number of kthreads
4908 equal to the number of CPUs.
4910 scftorture.onoff_holdoff= [KNL]
4911 Number seconds to wait after the start of the
4912 test before initiating CPU-hotplug operations.
4914 scftorture.onoff_interval= [KNL]
4915 Number seconds to wait between successive
4916 CPU-hotplug operations. Specifying zero (which
4917 is the default) disables CPU-hotplug operations.
4919 scftorture.shutdown_secs= [KNL]
4920 The number of seconds following the start of the
4921 test after which to shut down the system. The
4922 default of zero avoids shutting down the system.
4923 Non-zero values are useful for automated tests.
4925 scftorture.stat_interval= [KNL]
4926 The number of seconds between outputting the
4927 current test statistics to the console. A value
4928 of zero disables statistics output.
4930 scftorture.stutter_cpus= [KNL]
4931 The number of jiffies to wait between each change
4932 to the set of CPUs under test.
4934 scftorture.use_cpus_read_lock= [KNL]
4935 Use use_cpus_read_lock() instead of the default
4936 preempt_disable() to disable CPU hotplug
4937 while invoking one of the smp_call_function*()
4940 scftorture.verbose= [KNL]
4941 Enable additional printk() statements.
4943 scftorture.weight_single= [KNL]
4944 The probability weighting to use for the
4945 smp_call_function_single() function with a zero
4946 "wait" parameter. A value of -1 selects the
4947 default if all other weights are -1. However,
4948 if at least one weight has some other value, a
4949 value of -1 will instead select a weight of zero.
4951 scftorture.weight_single_wait= [KNL]
4952 The probability weighting to use for the
4953 smp_call_function_single() function with a
4954 non-zero "wait" parameter. See weight_single.
4956 scftorture.weight_many= [KNL]
4957 The probability weighting to use for the
4958 smp_call_function_many() function with a zero
4959 "wait" parameter. See weight_single.
4960 Note well that setting a high probability for
4961 this weighting can place serious IPI load
4964 scftorture.weight_many_wait= [KNL]
4965 The probability weighting to use for the
4966 smp_call_function_many() function with a
4967 non-zero "wait" parameter. See weight_single
4970 scftorture.weight_all= [KNL]
4971 The probability weighting to use for the
4972 smp_call_function_all() function with a zero
4973 "wait" parameter. See weight_single and
4976 scftorture.weight_all_wait= [KNL]
4977 The probability weighting to use for the
4978 smp_call_function_all() function with a
4979 non-zero "wait" parameter. See weight_single
4982 skew_tick= [KNL] Offset the periodic timer tick per cpu to mitigate
4983 xtime_lock contention on larger systems, and/or RCU lock
4984 contention on all systems with CONFIG_MAXSMP set.
4985 Format: { "0" | "1" }
4986 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1"
4988 Note: increases power consumption, thus should only be
4989 enabled if running jitter sensitive (HPC/RT) workloads.
4991 security= [SECURITY] Choose a legacy "major" security module to
4992 enable at boot. This has been deprecated by the
4995 selinux= [SELINUX] Disable or enable SELinux at boot time.
4996 Format: { "0" | "1" }
4997 See security/selinux/Kconfig help text.
5002 apparmor= [APPARMOR] Disable or enable AppArmor at boot time
5003 Format: { "0" | "1" }
5004 See security/apparmor/Kconfig help text
5007 Default value is set via kernel config option.
5009 serialnumber [BUGS=X86-32]
5012 Maximal number of shapers.
5020 Enable merging of slabs with similar size when the
5021 kernel is built without CONFIG_SLAB_MERGE_DEFAULT.
5024 Disable merging of slabs with similar size. May be
5025 necessary if there is some reason to distinguish
5026 allocs to different slabs, especially in hardened
5027 environments where the risk of heap overflows and
5028 layout control by attackers can usually be
5029 frustrated by disabling merging. This will reduce
5030 most of the exposure of a heap attack to a single
5031 cache (risks via metadata attacks are mostly
5032 unchanged). Debug options disable merging on their
5034 For more information see Documentation/vm/slub.rst.
5036 slab_max_order= [MM, SLAB]
5037 Determines the maximum allowed order for slabs.
5038 A high setting may cause OOMs due to memory
5039 fragmentation. Defaults to 1 for systems with
5040 more than 32MB of RAM, 0 otherwise.
5042 slub_debug[=options[,slabs][;[options[,slabs]]...] [MM, SLUB]
5043 Enabling slub_debug allows one to determine the
5044 culprit if slab objects become corrupted. Enabling
5045 slub_debug can create guard zones around objects and
5046 may poison objects when not in use. Also tracks the
5047 last alloc / free. For more information see
5048 Documentation/vm/slub.rst.
5050 slub_max_order= [MM, SLUB]
5051 Determines the maximum allowed order for slabs.
5052 A high setting may cause OOMs due to memory
5053 fragmentation. For more information see
5054 Documentation/vm/slub.rst.
5056 slub_min_objects= [MM, SLUB]
5057 The minimum number of objects per slab. SLUB will
5058 increase the slab order up to slub_max_order to
5059 generate a sufficiently large slab able to contain
5060 the number of objects indicated. The higher the number
5061 of objects the smaller the overhead of tracking slabs
5062 and the less frequently locks need to be acquired.
5063 For more information see Documentation/vm/slub.rst.
5065 slub_min_order= [MM, SLUB]
5066 Determines the minimum page order for slabs. Must be
5067 lower than slub_max_order.
5068 For more information see Documentation/vm/slub.rst.
5070 slub_merge [MM, SLUB]
5071 Same with slab_merge.
5073 slub_nomerge [MM, SLUB]
5074 Same with slab_nomerge. This is supported for legacy.
5075 See slab_nomerge for more information.
5078 Format: <io1>[,<io2>[,...,<io8>]]
5080 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices
5081 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port
5082 smsc-ircc2.ircc_sir= [HW] SIR base I/O port
5083 smsc-ircc2.ircc_fir= [HW] FIR base I/O port
5084 smsc-ircc2.ircc_irq= [HW] IRQ line
5085 smsc-ircc2.ircc_dma= [HW] DMA channel
5086 smsc-ircc2.ircc_transceiver= [HW] Transceiver type:
5087 0: Toshiba Satellite 1800 (GP data pin select)
5088 1: Fast pin select (default)
5091 smt [KNL,S390] Set the maximum number of threads (logical
5092 CPUs) to use per physical CPU on systems capable of
5093 symmetric multithreading (SMT). Will be capped to the
5094 actual hardware limit.
5096 Default: -1 (no limit)
5099 [KNL] Should the soft-lockup detector generate panics.
5102 A value of 1 instructs the soft-lockup detector
5103 to panic the machine when a soft-lockup occurs. It is
5104 also controlled by the kernel.softlockup_panic sysctl
5105 and CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC, which is the
5106 respective build-time switch to that functionality.
5108 softlockup_all_cpu_backtrace=
5109 [KNL] Should the soft-lockup detector generate
5110 backtraces on all cpus.
5113 sonypi.*= [HW] Sony Programmable I/O Control Device driver
5114 See Documentation/admin-guide/laptops/sonypi.rst
5116 spectre_v2= [X86] Control mitigation of Spectre variant 2
5117 (indirect branch speculation) vulnerability.
5118 The default operation protects the kernel from
5121 on - unconditionally enable, implies
5123 off - unconditionally disable, implies
5125 auto - kernel detects whether your CPU model is
5128 Selecting 'on' will, and 'auto' may, choose a
5129 mitigation method at run time according to the
5130 CPU, the available microcode, the setting of the
5131 CONFIG_RETPOLINE configuration option, and the
5132 compiler with which the kernel was built.
5134 Selecting 'on' will also enable the mitigation
5135 against user space to user space task attacks.
5137 Selecting 'off' will disable both the kernel and
5138 the user space protections.
5140 Specific mitigations can also be selected manually:
5142 retpoline - replace indirect branches
5143 retpoline,generic - google's original retpoline
5144 retpoline,amd - AMD-specific minimal thunk
5146 Not specifying this option is equivalent to
5150 [X86] Control mitigation of Spectre variant 2
5151 (indirect branch speculation) vulnerability between
5154 on - Unconditionally enable mitigations. Is
5155 enforced by spectre_v2=on
5157 off - Unconditionally disable mitigations. Is
5158 enforced by spectre_v2=off
5160 prctl - Indirect branch speculation is enabled,
5161 but mitigation can be enabled via prctl
5162 per thread. The mitigation control state
5163 is inherited on fork.
5166 - Like "prctl" above, but only STIBP is
5167 controlled per thread. IBPB is issued
5168 always when switching between different user
5172 - Same as "prctl" above, but all seccomp
5173 threads will enable the mitigation unless
5174 they explicitly opt out.
5177 - Like "seccomp" above, but only STIBP is
5178 controlled per thread. IBPB is issued
5179 always when switching between different
5180 user space processes.
5182 auto - Kernel selects the mitigation depending on
5183 the available CPU features and vulnerability.
5186 If CONFIG_SECCOMP=y then "seccomp", otherwise "prctl"
5188 Not specifying this option is equivalent to
5189 spectre_v2_user=auto.
5191 spec_store_bypass_disable=
5192 [HW] Control Speculative Store Bypass (SSB) Disable mitigation
5193 (Speculative Store Bypass vulnerability)
5195 Certain CPUs are vulnerable to an exploit against a
5196 a common industry wide performance optimization known
5197 as "Speculative Store Bypass" in which recent stores
5198 to the same memory location may not be observed by
5199 later loads during speculative execution. The idea
5200 is that such stores are unlikely and that they can
5201 be detected prior to instruction retirement at the
5202 end of a particular speculation execution window.
5204 In vulnerable processors, the speculatively forwarded
5205 store can be used in a cache side channel attack, for
5206 example to read memory to which the attacker does not
5207 directly have access (e.g. inside sandboxed code).
5209 This parameter controls whether the Speculative Store
5210 Bypass optimization is used.
5212 On x86 the options are:
5214 on - Unconditionally disable Speculative Store Bypass
5215 off - Unconditionally enable Speculative Store Bypass
5216 auto - Kernel detects whether the CPU model contains an
5217 implementation of Speculative Store Bypass and
5218 picks the most appropriate mitigation. If the
5219 CPU is not vulnerable, "off" is selected. If the
5220 CPU is vulnerable the default mitigation is
5221 architecture and Kconfig dependent. See below.
5222 prctl - Control Speculative Store Bypass per thread
5223 via prctl. Speculative Store Bypass is enabled
5224 for a process by default. The state of the control
5225 is inherited on fork.
5226 seccomp - Same as "prctl" above, but all seccomp threads
5227 will disable SSB unless they explicitly opt out.
5229 Default mitigations:
5230 X86: If CONFIG_SECCOMP=y "seccomp", otherwise "prctl"
5232 On powerpc the options are:
5234 on,auto - On Power8 and Power9 insert a store-forwarding
5235 barrier on kernel entry and exit. On Power7
5236 perform a software flush on kernel entry and
5240 Not specifying this option is equivalent to
5241 spec_store_bypass_disable=auto.
5243 spia_io_base= [HW,MTD]
5249 [X86] Enable split lock detection or bus lock detection
5251 When enabled (and if hardware support is present), atomic
5252 instructions that access data across cache line
5253 boundaries will result in an alignment check exception
5254 for split lock detection or a debug exception for
5259 warn - the kernel will emit rate-limited warnings
5260 about applications triggering the #AC
5261 exception or the #DB exception. This mode is
5262 the default on CPUs that support split lock
5263 detection or bus lock detection. Default
5264 behavior is by #AC if both features are
5265 enabled in hardware.
5267 fatal - the kernel will send SIGBUS to applications
5268 that trigger the #AC exception or the #DB
5269 exception. Default behavior is by #AC if
5270 both features are enabled in hardware.
5272 If an #AC exception is hit in the kernel or in
5273 firmware (i.e. not while executing in user mode)
5274 the kernel will oops in either "warn" or "fatal"
5277 #DB exception for bus lock is triggered only when
5281 Control the Special Register Buffer Data Sampling
5284 Certain CPUs are vulnerable to an MDS-like
5285 exploit which can leak bits from the random
5288 By default, this issue is mitigated by
5289 microcode. However, the microcode fix can cause
5290 the RDRAND and RDSEED instructions to become
5291 much slower. Among other effects, this will
5292 result in reduced throughput from /dev/urandom.
5294 The microcode mitigation can be disabled with
5295 the following option:
5297 off: Disable mitigation and remove
5298 performance impact to RDRAND and RDSEED
5300 srcutree.counter_wrap_check [KNL]
5301 Specifies how frequently to check for
5302 grace-period sequence counter wrap for the
5303 srcu_data structure's ->srcu_gp_seq_needed field.
5304 The greater the number of bits set in this kernel
5305 parameter, the less frequently counter wrap will
5306 be checked for. Note that the bottom two bits
5309 srcutree.exp_holdoff [KNL]
5310 Specifies how many nanoseconds must elapse
5311 since the end of the last SRCU grace period for
5312 a given srcu_struct until the next normal SRCU
5313 grace period will be considered for automatic
5314 expediting. Set to zero to disable automatic
5318 Speculative Store Bypass Disable control
5320 On CPUs that are vulnerable to the Speculative
5321 Store Bypass vulnerability and offer a
5322 firmware based mitigation, this parameter
5323 indicates how the mitigation should be used:
5325 force-on: Unconditionally enable mitigation for
5326 for both kernel and userspace
5327 force-off: Unconditionally disable mitigation for
5328 for both kernel and userspace
5329 kernel: Always enable mitigation in the
5330 kernel, and offer a prctl interface
5331 to allow userspace to register its
5332 interest in being mitigated too.
5334 stack_guard_gap= [MM]
5335 override the default stack gap protection. The value
5336 is in page units and it defines how many pages prior
5337 to (for stacks growing down) resp. after (for stacks
5338 growing up) the main stack are reserved for no other
5339 mapping. Default value is 256 pages.
5341 stack_depot_disable= [KNL]
5342 Setting this to true through kernel command line will
5343 disable the stack depot thereby saving the static memory
5344 consumed by the stack hash table. By default this is set
5348 Enabled the stack tracer on boot up.
5350 stacktrace_filter=[function-list]
5351 [FTRACE] Limit the functions that the stack tracer
5352 will trace at boot up. function-list is a comma-separated
5353 list of functions. This list can be changed at run
5354 time by the stack_trace_filter file in the debugfs
5355 tracing directory. Note, this enables stack tracing
5356 and the stacktrace above is not needed.
5360 Set the STI (builtin display/keyboard on the HP-PARISC
5361 machines) console (graphic card) which should be used
5362 as the initial boot-console.
5363 See also comment in drivers/video/console/sticore.c.
5366 See comment in drivers/video/console/sticore.c.
5369 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]]
5371 sunrpc.min_resvport=
5372 sunrpc.max_resvport=
5374 SunRPC servers often require that client requests
5375 originate from a privileged port (i.e. a port in the
5376 range 0 < portnr < 1024).
5377 An administrator who wishes to reserve some of these
5378 ports for other uses may adjust the range that the
5379 kernel's sunrpc client considers to be privileged
5380 using these two parameters to set the minimum and
5381 maximum port values.
5383 sunrpc.svc_rpc_per_connection_limit=
5385 Limit the number of requests that the server will
5386 process in parallel from a single connection.
5387 The default value is 0 (no limit).
5391 Control how the NFS server code allocates CPUs to
5392 service thread pools. Depending on how many NICs
5393 you have and where their interrupts are bound, this
5394 option will affect which CPUs will do NFS serving.
5395 Note: this parameter cannot be changed while the
5396 NFS server is running.
5398 auto the server chooses an appropriate mode
5399 automatically using heuristics
5400 global a single global pool contains all CPUs
5401 percpu one pool for each CPU
5402 pernode one pool for each NUMA node (equivalent
5403 to global on non-NUMA machines)
5405 sunrpc.tcp_slot_table_entries=
5406 sunrpc.udp_slot_table_entries=
5408 Sets the upper limit on the number of simultaneous
5409 RPC calls that can be sent from the client to a
5410 server. Increasing these values may allow you to
5411 improve throughput, but will also increase the
5412 amount of memory reserved for use by the client.
5414 suspend.pm_test_delay=
5416 Sets the number of seconds to remain in a suspend test
5417 mode before resuming the system (see
5418 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG
5419 is set. Default value is 5.
5422 Format: { on | off | y | n | 1 | 0 }
5423 This parameter controls use of the Protected
5424 Execution Facility on pSeries.
5427 [KNL] Enable accounting of swap in memory resource
5428 controller if no parameter or 1 is given or disable
5429 it if 0 is given (See Documentation/admin-guide/cgroup-v1/memory.rst)
5431 swiotlb= [ARM,IA-64,PPC,MIPS,X86]
5432 Format: { <int> | force | noforce }
5433 <int> -- Number of I/O TLB slabs
5434 force -- force using of bounce buffers even if they
5435 wouldn't be automatically used by the kernel
5436 noforce -- Never use bounce buffers (for debugging)
5441 Set a sysctl parameter, right before loading the init
5442 process, as if the value was written to the respective
5443 /proc/sys/... file. Both '.' and '/' are recognized as
5444 separators. Unrecognized parameters and invalid values
5445 are reported in the kernel log. Sysctls registered
5446 later by a loaded module cannot be set this way.
5447 Example: sysctl.vm.swappiness=40
5449 sysfs.deprecated=0|1 [KNL]
5450 Enable/disable old style sysfs layout for old udev
5451 on older distributions. When this option is enabled
5452 very new udev will not work anymore. When this option
5453 is disabled (or CONFIG_SYSFS_DEPRECATED not compiled)
5454 in older udev will not work anymore.
5455 Default depends on CONFIG_SYSFS_DEPRECATED_V2 set in
5456 the kernel configuration.
5458 sysrq_always_enabled
5460 Ignore sysrq setting - this boot parameter will
5461 neutralize any effect of /proc/sys/kernel/sysrq.
5462 Useful for debugging.
5464 tcpmhash_entries= [KNL,NET]
5465 Set the number of tcp_metrics_hash slots.
5466 Default value is 8192 or 16384 depending on total
5467 ram pages. This is used to specify the TCP metrics
5468 cache size. See Documentation/networking/ip-sysctl.rst
5469 "tcp_no_metrics_save" section for more details.
5473 test_suspend= [SUSPEND][,N]
5474 Specify "mem" (for Suspend-to-RAM) or "standby" (for
5475 standby suspend) or "freeze" (for suspend type freeze)
5476 as the system sleep state during system startup with
5477 the optional capability to repeat N number of times.
5478 The system is woken from this state using a
5479 wakeup-capable RTC alarm.
5481 thash_entries= [KNL,NET]
5482 Set number of hash buckets for TCP connection
5484 thermal.act= [HW,ACPI]
5485 -1: disable all active trip points in all thermal zones
5486 <degrees C>: override all lowest active trip points
5488 thermal.crt= [HW,ACPI]
5489 -1: disable all critical trip points in all thermal zones
5490 <degrees C>: override all critical trip points
5492 thermal.nocrt= [HW,ACPI]
5493 Set to disable actions on ACPI thermal zone
5494 critical and hot trip points.
5496 thermal.off= [HW,ACPI]
5497 1: disable ACPI thermal control
5499 thermal.psv= [HW,ACPI]
5500 -1: disable all passive trip points
5501 <degrees C>: override all passive trip points to this
5504 thermal.tzp= [HW,ACPI]
5505 Specify global default ACPI thermal zone polling rate
5506 <deci-seconds>: poll all this frequency
5507 0: no polling (default)
5510 Force threading of all interrupt handlers except those
5511 marked explicitly IRQF_NO_THREAD.
5515 Specify if the kernel should make use of the cpu
5516 topology information if the hardware supports this.
5517 The scheduler will make use of this information and
5518 e.g. base its process migration decisions on it.
5521 topology_updates= [KNL, PPC, NUMA]
5523 Specify if the kernel should ignore (off)
5524 topology updates sent by the hypervisor to this
5527 torture.disable_onoff_at_boot= [KNL]
5528 Prevent the CPU-hotplug component of torturing
5529 until after init has spawned.
5531 torture.ftrace_dump_at_shutdown= [KNL]
5532 Dump the ftrace buffer at torture-test shutdown,
5533 even if there were no errors. This can be a
5534 very costly operation when many torture tests
5535 are running concurrently, especially on systems
5536 with rotating-rust storage.
5538 torture.verbose_sleep_frequency= [KNL]
5539 Specifies how many verbose printk()s should be
5540 emitted between each sleep. The default of zero
5541 disables verbose-printk() sleeping.
5543 torture.verbose_sleep_duration= [KNL]
5544 Duration of each verbose-printk() sleep in jiffies.
5548 tpm_suspend_pcr=[HW,TPM]
5549 Format: integer pcr id
5550 Specify that at suspend time, the tpm driver
5551 should extend the specified pcr with zeros,
5552 as a workaround for some chips which fail to
5553 flush the last written pcr on TPM_SaveState.
5554 This will guarantee that all the other pcrs
5557 trace_buf_size=nn[KMG]
5558 [FTRACE] will set tracing buffer size on each cpu.
5560 trace_event=[event-list]
5561 [FTRACE] Set and start specified trace events in order
5562 to facilitate early boot debugging. The event-list is a
5563 comma-separated list of trace events to enable. See
5564 also Documentation/trace/events.rst
5566 trace_options=[option-list]
5567 [FTRACE] Enable or disable tracer options at boot.
5568 The option-list is a comma delimited list of options
5569 that can be enabled or disabled just as if you were
5570 to echo the option name into
5572 /sys/kernel/debug/tracing/trace_options
5574 For example, to enable stacktrace option (to dump the
5575 stack trace of each event), add to the command line:
5577 trace_options=stacktrace
5579 See also Documentation/trace/ftrace.rst "trace options"
5583 Have the tracepoints sent to printk as well as the
5584 tracing ring buffer. This is useful for early boot up
5585 where the system hangs or reboots and does not give the
5586 option for reading the tracing buffer or performing a
5587 ftrace_dump_on_oops.
5589 To turn off having tracepoints sent to printk,
5590 echo 0 > /proc/sys/kernel/tracepoint_printk
5591 Note, echoing 1 into this file without the
5592 tracepoint_printk kernel cmdline option has no effect.
5596 Having tracepoints sent to printk() and activating high
5597 frequency tracepoints such as irq or sched, can cause
5598 the system to live lock.
5601 [FTRACE] enable this option to disable tracing when a
5602 warning is hit. This turns off "tracing_on". Tracing can
5603 be enabled again by echoing '1' into the "tracing_on"
5604 file located in /sys/kernel/debug/tracing/
5606 This option is useful, as it disables the trace before
5607 the WARNING dump is called, which prevents the trace to
5608 be filled with content caused by the warning output.
5610 This option can also be set at run time via the sysctl
5611 option: kernel/traceoff_on_warning
5613 transparent_hugepage=
5615 Format: [always|madvise|never]
5616 Can be used to control the default behavior of the system
5617 with respect to transparent hugepages.
5618 See Documentation/admin-guide/mm/transhuge.rst
5621 trusted.source= [KEYS]
5623 This parameter identifies the trust source as a backend
5624 for trusted keys implementation. Supported trust
5628 If not specified then it defaults to iterating through
5629 the trust source list starting with TPM and assigns the
5630 first trust source as a backend which is initialized
5631 successfully during iteration.
5633 tsc= Disable clocksource stability checks for TSC.
5635 [x86] reliable: mark tsc clocksource as reliable, this
5636 disables clocksource verification at runtime, as well
5637 as the stability checks done at bootup. Used to enable
5638 high-resolution timer mode on older hardware, and in
5639 virtualized environment.
5640 [x86] noirqtime: Do not use TSC to do irq accounting.
5641 Used to run time disable IRQ_TIME_ACCOUNTING on any
5642 platforms where RDTSC is slow and this accounting
5644 [x86] unstable: mark the TSC clocksource as unstable, this
5645 marks the TSC unconditionally unstable at bootup and
5646 avoids any further wobbles once the TSC watchdog notices.
5647 [x86] nowatchdog: disable clocksource watchdog. Used
5648 in situations with strict latency requirements (where
5649 interruptions from clocksource watchdog are not
5652 tsc_early_khz= [X86] Skip early TSC calibration and use the given
5653 value instead. Useful when the early TSC frequency discovery
5654 procedure is not reliable, such as on overclocked systems
5655 with CPUID.16h support and partial CPUID.15h support.
5656 Format: <unsigned int>
5658 tsx= [X86] Control Transactional Synchronization
5659 Extensions (TSX) feature in Intel processors that
5660 support TSX control.
5662 This parameter controls the TSX feature. The options are:
5664 on - Enable TSX on the system. Although there are
5665 mitigations for all known security vulnerabilities,
5666 TSX has been known to be an accelerator for
5667 several previous speculation-related CVEs, and
5668 so there may be unknown security risks associated
5669 with leaving it enabled.
5671 off - Disable TSX on the system. (Note that this
5672 option takes effect only on newer CPUs which are
5673 not vulnerable to MDS, i.e., have
5674 MSR_IA32_ARCH_CAPABILITIES.MDS_NO=1 and which get
5675 the new IA32_TSX_CTRL MSR through a microcode
5676 update. This new MSR allows for the reliable
5677 deactivation of the TSX functionality.)
5679 auto - Disable TSX if X86_BUG_TAA is present,
5680 otherwise enable TSX on the system.
5682 Not specifying this option is equivalent to tsx=off.
5684 See Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
5687 tsx_async_abort= [X86,INTEL] Control mitigation for the TSX Async
5688 Abort (TAA) vulnerability.
5690 Similar to Micro-architectural Data Sampling (MDS)
5691 certain CPUs that support Transactional
5692 Synchronization Extensions (TSX) are vulnerable to an
5693 exploit against CPU internal buffers which can forward
5694 information to a disclosure gadget under certain
5697 In vulnerable processors, the speculatively forwarded
5698 data can be used in a cache side channel attack, to
5699 access data to which the attacker does not have direct
5702 This parameter controls the TAA mitigation. The
5705 full - Enable TAA mitigation on vulnerable CPUs
5708 full,nosmt - Enable TAA mitigation and disable SMT on
5709 vulnerable CPUs. If TSX is disabled, SMT
5710 is not disabled because CPU is not
5711 vulnerable to cross-thread TAA attacks.
5712 off - Unconditionally disable TAA mitigation
5714 On MDS-affected machines, tsx_async_abort=off can be
5715 prevented by an active MDS mitigation as both vulnerabilities
5716 are mitigated with the same mechanism so in order to disable
5717 this mitigation, you need to specify mds=off too.
5719 Not specifying this option is equivalent to
5720 tsx_async_abort=full. On CPUs which are MDS affected
5721 and deploy MDS mitigation, TAA mitigation is not
5722 required and doesn't provide any additional
5726 Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
5728 turbografx.map[2|3]= [HW,JOY]
5729 TurboGraFX parallel port interface
5731 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7>
5732 See also Documentation/input/devices/joystick-parport.rst
5734 udbg-immortal [PPC] When debugging early kernel crashes that
5735 happen after console_init() and before a proper
5736 console driver takes over, this boot options might
5737 help "seeing" what's going on.
5739 uhash_entries= [KNL,NET]
5740 Set number of hash buckets for UDP/UDP-Lite connections
5743 [USB] Ignore overcurrent events (default N).
5744 Some badly-designed motherboards generate lots of
5745 bogus events, for ports that aren't wired to
5746 anything. Set this parameter to avoid log spamming.
5747 Note that genuine overcurrent events won't be
5751 [X86] Cause panic on unknown NMI.
5753 usbcore.authorized_default=
5754 [USB] Default USB device authorization:
5755 (default -1 = authorized except for wireless USB,
5756 0 = not authorized, 1 = authorized, 2 = authorized
5757 if device connected to internal port)
5759 usbcore.autosuspend=
5760 [USB] The autosuspend time delay (in seconds) used
5761 for newly-detected USB devices (default 2). This
5762 is the time required before an idle device will be
5763 autosuspended. Devices for which the delay is set
5764 to a negative value won't be autosuspended at all.
5766 usbcore.usbfs_snoop=
5767 [USB] Set to log all usbfs traffic (default 0 = off).
5769 usbcore.usbfs_snoop_max=
5770 [USB] Maximum number of bytes to snoop in each URB
5773 usbcore.blinkenlights=
5774 [USB] Set to cycle leds on hubs (default 0 = off).
5776 usbcore.old_scheme_first=
5777 [USB] Start with the old device initialization
5778 scheme (default 0 = off).
5780 usbcore.usbfs_memory_mb=
5781 [USB] Memory limit (in MB) for buffers allocated by
5782 usbfs (default = 16, 0 = max = 2047).
5784 usbcore.use_both_schemes=
5785 [USB] Try the other device initialization scheme
5786 if the first one fails (default 1 = enabled).
5788 usbcore.initial_descriptor_timeout=
5789 [USB] Specifies timeout for the initial 64-byte
5790 USB_REQ_GET_DESCRIPTOR request in milliseconds
5791 (default 5000 = 5.0 seconds).
5793 usbcore.nousb [USB] Disable the USB subsystem
5796 [USB] A list of quirk entries to augment the built-in
5797 usb core quirk list. List entries are separated by
5798 commas. Each entry has the form
5799 VendorID:ProductID:Flags. The IDs are 4-digit hex
5800 numbers and Flags is a set of letters. Each letter
5801 will change the built-in quirk; setting it if it is
5802 clear and clearing it if it is set. The letters have
5803 the following meanings:
5804 a = USB_QUIRK_STRING_FETCH_255 (string
5805 descriptors must not be fetched using
5807 b = USB_QUIRK_RESET_RESUME (device can't resume
5808 correctly so reset it instead);
5809 c = USB_QUIRK_NO_SET_INTF (device can't handle
5810 Set-Interface requests);
5811 d = USB_QUIRK_CONFIG_INTF_STRINGS (device can't
5812 handle its Configuration or Interface
5814 e = USB_QUIRK_RESET (device can't be reset
5815 (e.g morph devices), don't use reset);
5816 f = USB_QUIRK_HONOR_BNUMINTERFACES (device has
5817 more interface descriptions than the
5818 bNumInterfaces count, and can't handle
5819 talking to these interfaces);
5820 g = USB_QUIRK_DELAY_INIT (device needs a pause
5821 during initialization, after we read
5822 the device descriptor);
5823 h = USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL (For
5824 high speed and super speed interrupt
5825 endpoints, the USB 2.0 and USB 3.0 spec
5826 require the interval in microframes (1
5827 microframe = 125 microseconds) to be
5828 calculated as interval = 2 ^
5830 Devices with this quirk report their
5831 bInterval as the result of this
5832 calculation instead of the exponent
5833 variable used in the calculation);
5834 i = USB_QUIRK_DEVICE_QUALIFIER (device can't
5835 handle device_qualifier descriptor
5837 j = USB_QUIRK_IGNORE_REMOTE_WAKEUP (device
5838 generates spurious wakeup, ignore
5839 remote wakeup capability);
5840 k = USB_QUIRK_NO_LPM (device can't handle Link
5842 l = USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL
5843 (Device reports its bInterval as linear
5844 frames instead of the USB 2.0
5846 m = USB_QUIRK_DISCONNECT_SUSPEND (Device needs
5847 to be disconnected before suspend to
5848 prevent spurious wakeup);
5849 n = USB_QUIRK_DELAY_CTRL_MSG (Device needs a
5850 pause after every control message);
5851 o = USB_QUIRK_HUB_SLOW_RESET (Hub needs extra
5852 delay after resetting its port);
5853 Example: quirks=0781:5580:bk,0a5c:5834:gij
5856 [USBHID] The interval which mice are to be polled at.
5859 [USBHID] The interval which joysticks are to be polled at.
5862 [USBHID] The interval which keyboards are to be polled at.
5864 usb-storage.delay_use=
5865 [UMS] The delay in seconds before a new device is
5866 scanned for Logical Units (default 1).
5869 [UMS] A list of quirks entries to supplement or
5870 override the built-in unusual_devs list. List
5871 entries are separated by commas. Each entry has
5872 the form VID:PID:Flags where VID and PID are Vendor
5873 and Product ID values (4-digit hex numbers) and
5874 Flags is a set of characters, each corresponding
5875 to a common usb-storage quirk flag as follows:
5876 a = SANE_SENSE (collect more than 18 bytes
5877 of sense data, not on uas);
5878 b = BAD_SENSE (don't collect more than 18
5879 bytes of sense data, not on uas);
5880 c = FIX_CAPACITY (decrease the reported
5881 device capacity by one sector);
5882 d = NO_READ_DISC_INFO (don't use
5883 READ_DISC_INFO command, not on uas);
5884 e = NO_READ_CAPACITY_16 (don't use
5885 READ_CAPACITY_16 command);
5886 f = NO_REPORT_OPCODES (don't use report opcodes
5888 g = MAX_SECTORS_240 (don't transfer more than
5889 240 sectors at a time, uas only);
5890 h = CAPACITY_HEURISTICS (decrease the
5891 reported device capacity by one
5892 sector if the number is odd);
5893 i = IGNORE_DEVICE (don't bind to this
5895 j = NO_REPORT_LUNS (don't use report luns
5897 k = NO_SAME (do not use WRITE_SAME, uas only)
5898 l = NOT_LOCKABLE (don't try to lock and
5899 unlock ejectable media, not on uas);
5900 m = MAX_SECTORS_64 (don't transfer more
5901 than 64 sectors = 32 KB at a time,
5903 n = INITIAL_READ10 (force a retry of the
5904 initial READ(10) command, not on uas);
5905 o = CAPACITY_OK (accept the capacity
5906 reported by the device, not on uas);
5907 p = WRITE_CACHE (the device cache is ON
5908 by default, not on uas);
5909 r = IGNORE_RESIDUE (the device reports
5910 bogus residue values, not on uas);
5911 s = SINGLE_LUN (the device has only one
5913 t = NO_ATA_1X (don't allow ATA(12) and ATA(16)
5914 commands, uas only);
5915 u = IGNORE_UAS (don't bind to the uas driver);
5916 w = NO_WP_DETECT (don't test whether the
5917 medium is write-protected).
5918 y = ALWAYS_SYNC (issue a SYNCHRONIZE_CACHE
5919 even if the device claims no cache,
5921 Example: quirks=0419:aaf5:rl,0421:0433:rc
5923 user_debug= [KNL,ARM]
5925 See arch/arm/Kconfig.debug help text.
5926 1 - undefined instruction events
5928 4 - invalid data aborts
5931 Example: user_debug=31
5934 [X86] Flags controlling user PTE allocations.
5936 nohigh = do not allocate PTE pages in
5937 HIGHMEM regardless of setting
5941 On X86_32, this is an alias for vdso32=. Otherwise:
5943 vdso=1: enable VDSO (the default)
5944 vdso=0: disable VDSO mapping
5946 vdso32= [X86] Control the 32-bit vDSO
5947 vdso32=1: enable 32-bit VDSO
5948 vdso32=0 or vdso32=2: disable 32-bit VDSO
5950 See the help text for CONFIG_COMPAT_VDSO for more
5951 details. If CONFIG_COMPAT_VDSO is set, the default is
5952 vdso32=0; otherwise, the default is vdso32=1.
5954 For compatibility with older kernels, vdso32=2 is an
5957 Try vdso32=0 if you encounter an error that says:
5958 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed!
5961 vector=percpu: enable percpu vector domain
5963 video= [FB] Frame buffer configuration
5964 See Documentation/fb/modedb.rst.
5966 video.brightness_switch_enabled= [0,1]
5967 If set to 1, on receiving an ACPI notify event
5968 generated by hotkey, video driver will adjust brightness
5969 level and then send out the event to user space through
5970 the allocated input device; If set to 0, video driver
5971 will only send out the event without touching backlight
5976 [VMMIO] Memory mapped virtio (platform) device.
5978 <size>@<baseaddr>:<irq>[:<id>]
5980 <size> := size (can use standard suffixes
5982 <baseaddr> := physical base address
5983 <irq> := interrupt number (as passed to
5985 <id> := (optional) platform device id
5987 virtio_mmio.device=1K@0x100b0000:48:7
5989 Can be used multiple times for multiple devices.
5991 vga= [BOOT,X86-32] Select a particular video mode
5992 See Documentation/x86/boot.rst and
5993 Documentation/admin-guide/svga.rst.
5994 Use vga=ask for menu.
5995 This is actually a boot loader parameter; the value is
5996 passed to the kernel using a special protocol.
5998 vm_debug[=options] [KNL] Available with CONFIG_DEBUG_VM=y.
5999 May slow down system boot speed, especially when
6000 enabled on systems with a large amount of memory.
6001 All options are enabled by default, and this
6002 interface is meant to allow for selectively
6003 enabling or disabling specific virtual memory
6006 Available options are:
6007 P Enable page structure init time poisoning
6008 - Disable all of the above options
6010 vmalloc=nn[KMG] [KNL,BOOT] Forces the vmalloc area to have an exact
6011 size of <nn>. This can be used to increase the
6012 minimum size (128MB on x86). It can also be used to
6013 decrease the size and leave more room for directly
6016 vmcp_cma=nn[MG] [KNL,S390]
6017 Sets the memory size reserved for contiguous memory
6018 allocations for the vmcp device driver.
6020 vmhalt= [KNL,S390] Perform z/VM CP command after system halt.
6023 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic.
6026 vmpoff= [KNL,S390] Perform z/VM CP command after power off.
6030 Controls the behavior of vsyscalls (i.e. calls to
6031 fixed addresses of 0xffffffffff600x00 from legacy
6032 code). Most statically-linked binaries and older
6033 versions of glibc use these calls. Because these
6034 functions are at fixed addresses, they make nice
6035 targets for exploits that can control RIP.
6037 emulate [default] Vsyscalls turn into traps and are
6038 emulated reasonably safely. The vsyscall
6041 xonly Vsyscalls turn into traps and are
6042 emulated reasonably safely. The vsyscall
6043 page is not readable.
6045 none Vsyscalls don't work at all. This makes
6046 them quite hard to use for exploits but
6047 might break your system.
6049 vt.color= [VT] Default text color.
6050 Format: 0xYX, X = foreground, Y = background.
6051 Default: 0x07 = light gray on black.
6053 vt.cur_default= [VT] Default cursor shape.
6054 Format: 0xCCBBAA, where AA, BB, and CC are the same as
6055 the parameters of the <Esc>[?A;B;Cc escape sequence;
6056 see VGA-softcursor.txt. Default: 2 = underline.
6058 vt.default_blu= [VT]
6059 Format: <blue0>,<blue1>,<blue2>,...,<blue15>
6060 Change the default blue palette of the console.
6061 This is a 16-member array composed of values
6064 vt.default_grn= [VT]
6065 Format: <green0>,<green1>,<green2>,...,<green15>
6066 Change the default green palette of the console.
6067 This is a 16-member array composed of values
6070 vt.default_red= [VT]
6071 Format: <red0>,<red1>,<red2>,...,<red15>
6072 Change the default red palette of the console.
6073 This is a 16-member array composed of values
6079 Set system-wide default UTF-8 mode for all tty's.
6080 Default is 1, i.e. UTF-8 mode is enabled for all
6081 newly opened terminals.
6083 vt.global_cursor_default=
6086 Set system-wide default for whether a cursor
6087 is shown on new VTs. Default is -1,
6088 i.e. cursors will be created by default unless
6089 overridden by individual drivers. 0 will hide
6090 cursors, 1 will display them.
6092 vt.italic= [VT] Default color for italic text; 0-15.
6095 vt.underline= [VT] Default color for underlined text; 0-15.
6098 watchdog timers [HW,WDT] For information on watchdog timers,
6099 see Documentation/watchdog/watchdog-parameters.rst
6100 or other driver-specific files in the
6101 Documentation/watchdog/ directory.
6105 Set the hard lockup detector stall duration
6106 threshold in seconds. The soft lockup detector
6107 threshold is set to twice the value. A value of 0
6108 disables both lockup detectors. Default is 10
6111 workqueue.watchdog_thresh=
6112 If CONFIG_WQ_WATCHDOG is configured, workqueue can
6113 warn stall conditions and dump internal state to
6114 help debugging. 0 disables workqueue stall
6115 detection; otherwise, it's the stall threshold
6116 duration in seconds. The default value is 30 and
6117 it can be updated at runtime by writing to the
6118 corresponding sysfs file.
6120 workqueue.disable_numa
6121 By default, all work items queued to unbound
6122 workqueues are affine to the NUMA nodes they're
6123 issued on, which results in better behavior in
6124 general. If NUMA affinity needs to be disabled for
6125 whatever reason, this option can be used. Note
6126 that this also can be controlled per-workqueue for
6127 workqueues visible under /sys/bus/workqueue/.
6129 workqueue.power_efficient
6130 Per-cpu workqueues are generally preferred because
6131 they show better performance thanks to cache
6132 locality; unfortunately, per-cpu workqueues tend to
6133 be more power hungry than unbound workqueues.
6135 Enabling this makes the per-cpu workqueues which
6136 were observed to contribute significantly to power
6137 consumption unbound, leading to measurably lower
6138 power usage at the cost of small performance
6141 The default value of this parameter is determined by
6142 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT.
6144 workqueue.debug_force_rr_cpu
6145 Workqueue used to implicitly guarantee that work
6146 items queued without explicit CPU specified are put
6147 on the local CPU. This guarantee is no longer true
6148 and while local CPU is still preferred work items
6149 may be put on foreign CPUs. This debug option
6150 forces round-robin CPU selection to flush out
6151 usages which depend on the now broken guarantee.
6152 When enabled, memory and cache locality will be
6155 x2apic_phys [X86-64,APIC] Use x2apic physical mode instead of
6156 default x2apic cluster mode on platforms
6159 xen_512gb_limit [KNL,X86-64,XEN]
6160 Restricts the kernel running paravirtualized under Xen
6161 to use only up to 512 GB of RAM. The reason to do so is
6162 crash analysis tools and Xen tools for doing domain
6163 save/restore/migration must be enabled to handle larger
6166 xen_emul_unplug= [HW,X86,XEN]
6167 Unplug Xen emulated devices
6168 Format: [unplug0,][unplug1]
6169 ide-disks -- unplug primary master IDE devices
6170 aux-ide-disks -- unplug non-primary-master IDE devices
6171 nics -- unplug network devices
6172 all -- unplug all emulated devices (NICs and IDE disks)
6173 unnecessary -- unplugging emulated devices is
6174 unnecessary even if the host did not respond to
6176 never -- do not unplug even if version check succeeds
6178 xen_legacy_crash [X86,XEN]
6179 Crash from Xen panic notifier, without executing late
6180 panic() code such as dumping handler.
6182 xen_nopvspin [X86,XEN]
6183 Disables the qspinlock slowpath using Xen PV optimizations.
6184 This parameter is obsoleted by "nopvspin" parameter, which
6185 has equivalent effect for XEN platform.
6188 Disables the PV optimizations forcing the HVM guest to
6189 run as generic HVM guest with no PV drivers.
6190 This option is obsoleted by the "nopv" option, which
6191 has equivalent effect for XEN platform.
6193 xen_no_vector_callback
6194 [KNL,X86,XEN] Disable the vector callback for Xen
6195 event channel interrupts.
6197 xen_scrub_pages= [XEN]
6198 Boolean option to control scrubbing pages before giving them back
6199 to Xen, for use by other domains. Can be also changed at runtime
6200 with /sys/devices/system/xen_memory/xen_memory0/scrub_pages.
6201 Default value controlled with CONFIG_XEN_SCRUB_PAGES_DEFAULT.
6203 xen_timer_slop= [X86-64,XEN]
6204 Set the timer slop (in nanoseconds) for the virtual Xen
6205 timers (default is 100000). This adjusts the minimum
6206 delta of virtualized Xen timers, where lower values
6207 improve timer resolution at the expense of processing
6208 more timer interrupts.
6210 xen.event_eoi_delay= [XEN]
6211 How long to delay EOI handling in case of event
6212 storms (jiffies). Default is 10.
6214 xen.event_loop_timeout= [XEN]
6215 After which time (jiffies) the event handling loop
6216 should start to delay EOI handling. Default is 2.
6218 xen.fifo_events= [XEN]
6219 Boolean parameter to disable using fifo event handling
6220 even if available. Normally fifo event handling is
6221 preferred over the 2-level event handling, as it is
6222 fairer and the number of possible event channels is
6223 much higher. Default is on (use fifo events).
6225 nopv= [X86,XEN,KVM,HYPER_V,VMWARE]
6226 Disables the PV optimizations forcing the guest to run
6227 as generic guest with no PV drivers. Currently support
6228 XEN HVM, KVM, HYPER_V and VMWARE guest.
6230 nopvspin [X86,XEN,KVM]
6231 Disables the qspinlock slow path using PV optimizations
6232 which allow the hypervisor to 'idle' the guest on lock
6235 xirc2ps_cs= [NET,PCMCIA]
6237 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]]
6240 By default on POWER9 and above, the kernel will
6241 natively use the XIVE interrupt controller. This option
6242 allows the fallback firmware mode to be used:
6244 off Fallback to firmware control of XIVE interrupt
6245 controller on both pseries and powernv
6246 platforms. Only useful on POWER9 and above.
6248 xhci-hcd.quirks [USB,KNL]
6249 A hex value specifying bitmask with supplemental xhci
6250 host controller quirks. Meaning of each bit can be
6251 consulted in header drivers/usb/host/xhci.h.
6254 Format: { early | on | rw | ro | off }
6255 Controls if xmon debugger is enabled. Default is off.
6256 Passing only "xmon" is equivalent to "xmon=early".
6257 early Call xmon as early as possible on boot; xmon
6258 debugger is called from setup_arch().
6259 on xmon debugger hooks will be installed so xmon
6260 is only called on a kernel crash. Default mode,
6261 i.e. either "ro" or "rw" mode, is controlled
6262 with CONFIG_XMON_DEFAULT_RO_MODE.
6263 rw xmon debugger hooks will be installed so xmon
6264 is called only on a kernel crash, mode is write,
6265 meaning SPR registers, memory and, other data
6266 can be written using xmon commands.
6267 ro same as "rw" option above but SPR registers,
6268 memory, and other data can't be written using
6270 off xmon is disabled.
projects / linux-2.6-microblaze.git / blob