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 gpt [EFI] Forces disk with valid GPT signature but
1473 invalid Protective MBR to be treated as GPT. If the
1474 primary GPT is corrupted, it enables the backup/alternate
1475 GPT to be used instead.
1477 grcan.enable0= [HW] Configuration of physical interface 0. Determines
1478 the "Enable 0" bit of the configuration register.
1481 grcan.enable1= [HW] Configuration of physical interface 1. Determines
1482 the "Enable 0" bit of the configuration register.
1485 grcan.select= [HW] Select which physical interface to use.
1488 grcan.txsize= [HW] Sets the size of the tx buffer.
1489 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0.
1491 grcan.rxsize= [HW] Sets the size of the rx buffer.
1492 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0.
1495 gpio-mockup.gpio_mockup_ranges
1496 [HW] Sets the ranges of gpiochip of for this device.
1497 Format: <start1>,<end1>,<start2>,<end2>...
1499 hardlockup_all_cpu_backtrace=
1500 [KNL] Should the hard-lockup detector generate
1501 backtraces on all cpus.
1504 hashdist= [KNL,NUMA] Large hashes allocated during boot
1505 are distributed across NUMA nodes. Defaults on
1506 for 64-bit NUMA, off otherwise.
1507 Format: 0 | 1 (for off | on)
1509 hcl= [IA-64] SGI's Hardware Graph compatibility layer
1511 hd= [EIDE] (E)IDE hard drive subsystem geometry
1512 Format: <cyl>,<head>,<sect>
1515 Disable Hardware Error Source Table (HEST) support;
1516 corresponding firmware-first mode error processing
1517 logic will be disabled.
1519 highmem=nn[KMG] [KNL,BOOT] forces the highmem zone to have an exact
1520 size of <nn>. This works even on boxes that have no
1521 highmem otherwise. This also works to reduce highmem
1522 size on bigger boxes.
1524 highres= [KNL] Enable/disable high resolution timer mode.
1525 Valid parameters: "on", "off"
1530 hpet= [X86-32,HPET] option to control HPET usage
1531 Format: { enable (default) | disable | force |
1533 disable: disable HPET and use PIT instead
1534 force: allow force enabled of undocumented chips (ICH4,
1536 verbose: show contents of HPET registers during setup
1538 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET
1539 registers. Default set by CONFIG_HPET_MMAP_DEFAULT.
1541 hugetlb_cma= [HW,CMA] The size of a CMA area used for allocation
1542 of gigantic hugepages.
1545 Reserve a CMA area of given size and allocate gigantic
1546 hugepages using the CMA allocator. If enabled, the
1547 boot-time allocation of gigantic hugepages is skipped.
1549 hugepages= [HW] Number of HugeTLB pages to allocate at boot.
1550 If this follows hugepagesz (below), it specifies
1551 the number of pages of hugepagesz to be allocated.
1552 If this is the first HugeTLB parameter on the command
1553 line, it specifies the number of pages to allocate for
1554 the default huge page size. See also
1555 Documentation/admin-guide/mm/hugetlbpage.rst.
1559 [HW] The size of the HugeTLB pages. This is used in
1560 conjunction with hugepages (above) to allocate huge
1561 pages of a specific size at boot. The pair
1562 hugepagesz=X hugepages=Y can be specified once for
1563 each supported huge page size. Huge page sizes are
1564 architecture dependent. See also
1565 Documentation/admin-guide/mm/hugetlbpage.rst.
1569 [KNL] Should the hung task detector generate panics.
1572 A value of 1 instructs the kernel to panic when a
1573 hung task is detected. The default value is controlled
1574 by the CONFIG_BOOTPARAM_HUNG_TASK_PANIC build-time
1575 option. The value selected by this boot parameter can
1576 be changed later by the kernel.hung_task_panic sysctl.
1578 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC)
1579 terminal devices. Valid values: 0..8
1580 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs.
1581 If specified, z/VM IUCV HVC accepts connections
1582 from listed z/VM user IDs only.
1584 hv_nopvspin [X86,HYPER_V] Disables the paravirt spinlock optimizations
1585 which allow the hypervisor to 'idle' the
1586 guest on lock contention.
1589 Do not unregister boot console at start. This is only
1590 useful for debugging when something happens in the window
1591 between unregistering the boot console and initializing
1594 i2c_bus= [HW] Override the default board specific I2C bus speed
1595 or register an additional I2C bus that is not
1596 registered from board initialization code.
1600 i8042.debug [HW] Toggle i8042 debug mode
1601 i8042.unmask_kbd_data
1602 [HW] Enable printing of interrupt data from the KBD port
1603 (disabled by default, and as a pre-condition
1604 requires that i8042.debug=1 be enabled)
1605 i8042.direct [HW] Put keyboard port into non-translated mode
1606 i8042.dumbkbd [HW] Pretend that controller can only read data from
1607 keyboard and cannot control its state
1608 (Don't attempt to blink the leds)
1609 i8042.noaux [HW] Don't check for auxiliary (== mouse) port
1610 i8042.nokbd [HW] Don't check/create keyboard port
1611 i8042.noloop [HW] Disable the AUX Loopback command while probing
1613 i8042.nomux [HW] Don't check presence of an active multiplexing
1615 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX
1617 i8042.notimeout [HW] Ignore timeout condition signalled by controller
1618 i8042.reset [HW] Reset the controller during init, cleanup and
1619 suspend-to-ram transitions, only during s2r
1620 transitions, or never reset
1621 Format: { 1 | Y | y | 0 | N | n }
1622 1, Y, y: always reset controller
1623 0, N, n: don't ever reset controller
1624 Default: only on s2r transitions on x86; most other
1625 architectures force reset to be always executed
1626 i8042.unlock [HW] Unlock (ignore) the keylock
1627 i8042.kbdreset [HW] Reset device connected to KBD port
1631 i8k.ignore_dmi [HW] Continue probing hardware even if DMI data
1632 indicates that the driver is running on unsupported
1634 i8k.force [HW] Activate i8k driver even if SMM BIOS signature
1635 does not match list of supported models.
1637 [HW] Report power status in /proc/i8k
1638 (disabled by default)
1639 i8k.restricted [HW] Allow controlling fans only if SYS_ADMIN
1642 i915.invert_brightness=
1643 [DRM] Invert the sense of the variable that is used to
1644 set the brightness of the panel backlight. Normally a
1645 brightness value of 0 indicates backlight switched off,
1646 and the maximum of the brightness value sets the backlight
1647 to maximum brightness. If this parameter is set to 0
1648 (default) and the machine requires it, or this parameter
1649 is set to 1, a brightness value of 0 sets the backlight
1650 to maximum brightness, and the maximum of the brightness
1651 value switches the backlight off.
1652 -1 -- never invert brightness
1653 0 -- machine default
1654 1 -- force brightness inversion
1657 Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]]
1659 ide-core.nodma= [HW] (E)IDE subsystem
1660 Format: =0.0 to prevent dma on hda, =0.1 hdb =1.0 hdc
1661 .vlb_clock .pci_clock .noflush .nohpa .noprobe .nowerr
1662 .cdrom .chs .ignore_cable are additional options
1663 See Documentation/ide/ide.rst.
1665 ide-generic.probe-mask= [HW] (E)IDE subsystem
1667 Probe mask for legacy ISA IDE ports. Depending on
1668 platform up to 6 ports are supported, enabled by
1669 setting corresponding bits in the mask to 1. The
1670 default value is 0x0, which has a special meaning.
1671 On systems that have PCI, it triggers scanning the
1672 PCI bus for the first and the second port, which
1673 are then probed. On systems without PCI the value
1674 of 0x0 enables probing the two first ports as if it
1677 ide-pci-generic.all-generic-ide [HW] (E)IDE subsystem
1678 Claim all unknown PCI IDE storage controllers.
1681 Format: idle=poll, idle=halt, idle=nomwait
1682 Poll forces a polling idle loop that can slightly
1683 improve the performance of waking up a idle CPU, but
1684 will use a lot of power and make the system run hot.
1686 idle=halt: Halt is forced to be used for CPU idle.
1687 In such case C2/C3 won't be used again.
1688 idle=nomwait: Disable mwait for CPU C-states
1692 Allow force disabling of Shared Virtual Memory (SVA)
1693 support for the idxd driver. By default it is set to
1696 ieee754= [MIPS] Select IEEE Std 754 conformance mode
1697 Format: { strict | legacy | 2008 | relaxed }
1700 Choose which programs will be accepted for execution
1701 based on the IEEE 754 NaN encoding(s) supported by
1702 the FPU and the NaN encoding requested with the value
1703 of an ELF file header flag individually set by each
1704 binary. Hardware implementations are permitted to
1705 support either or both of the legacy and the 2008 NaN
1708 Available settings are as follows:
1709 strict accept binaries that request a NaN encoding
1710 supported by the FPU
1711 legacy only accept legacy-NaN binaries, if supported
1713 2008 only accept 2008-NaN binaries, if supported
1715 relaxed accept any binaries regardless of whether
1716 supported by the FPU
1718 The FPU emulator is always able to support both NaN
1719 encodings, so if no FPU hardware is present or it has
1720 been disabled with 'nofpu', then the settings of
1721 'legacy' and '2008' strap the emulator accordingly,
1722 'relaxed' straps the emulator for both legacy-NaN and
1723 2008-NaN, whereas 'strict' enables legacy-NaN only on
1724 legacy processors and both NaN encodings on MIPS32 or
1727 The setting for ABS.fmt/NEG.fmt instruction execution
1728 mode generally follows that for the NaN encoding,
1729 except where unsupported by hardware.
1731 ignore_loglevel [KNL]
1732 Ignore loglevel setting - this will print /all/
1733 kernel messages to the console. Useful for debugging.
1734 We also add it as printk module parameter, so users
1735 could change it dynamically, usually by
1736 /sys/module/printk/parameters/ignore_loglevel.
1739 Ignore RLIMIT_DATA setting for data mappings,
1740 print warning at first misuse. Can be changed via
1741 /sys/module/kernel/parameters/ignore_rlimit_data.
1743 ihash_entries= [KNL]
1744 Set number of hash buckets for inode cache.
1746 ima_appraise= [IMA] appraise integrity measurements
1747 Format: { "off" | "enforce" | "fix" | "log" }
1750 ima_appraise_tcb [IMA] Deprecated. Use ima_policy= instead.
1751 The builtin appraise policy appraises all files
1754 ima_canonical_fmt [IMA]
1755 Use the canonical format for the binary runtime
1756 measurements, instead of host native format.
1759 Format: { md5 | sha1 | rmd160 | sha256 | sha384
1763 The list of supported hash algorithms is defined
1764 in crypto/hash_info.h.
1767 The builtin policies to load during IMA setup.
1768 Format: "tcb | appraise_tcb | secure_boot |
1769 fail_securely | critical_data"
1771 The "tcb" policy measures all programs exec'd, files
1772 mmap'd for exec, and all files opened with the read
1773 mode bit set by either the effective uid (euid=0) or
1776 The "appraise_tcb" policy appraises the integrity of
1777 all files owned by root.
1779 The "secure_boot" policy appraises the integrity
1780 of files (eg. kexec kernel image, kernel modules,
1781 firmware, policy, etc) based on file signatures.
1783 The "fail_securely" policy forces file signature
1784 verification failure also on privileged mounted
1785 filesystems with the SB_I_UNVERIFIABLE_SIGNATURE
1788 The "critical_data" policy measures kernel integrity
1791 ima_tcb [IMA] Deprecated. Use ima_policy= instead.
1792 Load a policy which meets the needs of the Trusted
1793 Computing Base. This means IMA will measure all
1794 programs exec'd, files mmap'd for exec, and all files
1795 opened for read by uid=0.
1798 Select one of defined IMA measurements template formats.
1799 Formats: { "ima" | "ima-ng" | "ima-sig" }
1803 [IMA] Define a custom template format.
1804 Format: { "field1|...|fieldN" }
1806 ima.ahash_minsize= [IMA] Minimum file size for asynchronous hash usage
1807 Format: <min_file_size>
1808 Set the minimal file size for using asynchronous hash.
1809 If left unspecified, ahash usage is disabled.
1811 ahash performance varies for different data sizes on
1812 different crypto accelerators. This option can be used
1813 to achieve the best performance for a particular HW.
1815 ima.ahash_bufsize= [IMA] Asynchronous hash buffer size
1817 Set hashing buffer size. Default: 4k.
1819 ahash performance varies for different chunk sizes on
1820 different crypto accelerators. This option can be used
1821 to achieve best performance for particular HW.
1825 Run specified binary instead of /sbin/init as init
1828 initcall_debug [KNL] Trace initcalls as they are executed. Useful
1829 for working out where the kernel is dying during
1832 initcall_blacklist= [KNL] Do not execute a comma-separated list of
1833 initcall functions. Useful for debugging built-in
1834 modules and initcalls.
1836 initrd= [BOOT] Specify the location of the initial ramdisk
1838 initrdmem= [KNL] Specify a physical address and size from which to
1839 load the initrd. If an initrd is compiled in or
1840 specified in the bootparams, it takes priority over this
1842 Format: ss[KMG],nn[KMG]
1845 init_on_alloc= [MM] Fill newly allocated pages and heap objects with
1848 Default set by CONFIG_INIT_ON_ALLOC_DEFAULT_ON.
1850 init_on_free= [MM] Fill freed pages and heap objects with zeroes.
1852 Default set by CONFIG_INIT_ON_FREE_DEFAULT_ON.
1854 init_pkru= [X86] Specify the default memory protection keys rights
1855 register contents for all processes. 0x55555554 by
1856 default (disallow access to all but pkey 0). Can
1857 override in debugfs after boot.
1859 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver
1862 int_pln_enable [X86] Enable power limit notification interrupt
1864 integrity_audit=[IMA]
1865 Format: { "0" | "1" }
1866 0 -- basic integrity auditing messages. (Default)
1867 1 -- additional integrity auditing messages.
1869 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option
1871 Enable intel iommu driver.
1873 Disable intel iommu driver.
1874 igfx_off [Default Off]
1875 By default, gfx is mapped as normal device. If a gfx
1876 device has a dedicated DMAR unit, the DMAR unit is
1877 bypassed by not enabling DMAR with this option. In
1878 this case, gfx device will use physical address for
1881 With this option iommu will not optimize to look
1882 for io virtual address below 32-bit forcing dual
1883 address cycle on pci bus for cards supporting greater
1884 than 32-bit addressing. The default is to look
1885 for translation below 32-bit and if not available
1886 then look in the higher range.
1887 strict [Default Off]
1888 With this option on every unmap_single operation will
1889 result in a hardware IOTLB flush operation as opposed
1890 to batching them for performance.
1891 sp_off [Default Off]
1892 By default, super page will be supported if Intel IOMMU
1893 has the capability. With this option, super page will
1896 By default, scalable mode will be disabled even if the
1897 hardware advertises that it has support for the scalable
1898 mode translation. With this option set, scalable mode
1899 will be used on hardware which claims to support it.
1900 tboot_noforce [Default Off]
1901 Do not force the Intel IOMMU enabled under tboot.
1902 By default, tboot will force Intel IOMMU on, which
1903 could harm performance of some high-throughput
1904 devices like 40GBit network cards, even if identity
1906 Note that using this option lowers the security
1907 provided by tboot because it makes the system
1908 vulnerable to DMA attacks.
1910 intel_idle.max_cstate= [KNL,HW,ACPI,X86]
1911 0 disables intel_idle and fall back on acpi_idle.
1912 1 to 9 specify maximum depth of C-state.
1916 Do not enable intel_pstate as the default
1917 scaling driver for the supported processors
1919 Use intel_pstate as a scaling driver, but configure it
1920 to work with generic cpufreq governors (instead of
1921 enabling its internal governor). This mode cannot be
1922 used along with the hardware-managed P-states (HWP)
1925 Enable intel_pstate on systems that prohibit it by default
1926 in favor of acpi-cpufreq. Forcing the intel_pstate driver
1927 instead of acpi-cpufreq may disable platform features, such
1928 as thermal controls and power capping, that rely on ACPI
1929 P-States information being indicated to OSPM and therefore
1930 should be used with caution. This option does not work with
1931 processors that aren't supported by the intel_pstate driver
1932 or on platforms that use pcc-cpufreq instead of acpi-cpufreq.
1934 Do not enable hardware P state control (HWP)
1937 Only load intel_pstate on systems which support
1938 hardware P state control (HWP) if available.
1940 Enforce ACPI _PPC performance limits. If the Fixed ACPI
1941 Description Table, specifies preferred power management
1942 profile as "Enterprise Server" or "Performance Server",
1943 then this feature is turned on by default.
1945 Allow per-logical-CPU P-State performance control limits using
1946 cpufreq sysfs interface
1948 intremap= [X86-64, Intel-IOMMU]
1949 on enable Interrupt Remapping (default)
1950 off disable Interrupt Remapping
1951 nosid disable Source ID checking
1953 BIOS x2APIC opt-out request will be ignored
1954 nopost disable Interrupt Posting
1956 iomem= Disable strict checking of access to MMIO memory
1957 strict regions from userspace.
1972 nobypass [PPC/POWERNV]
1973 Disable IOMMU bypass, using IOMMU for PCI devices.
1975 iommu.strict= [ARM64] Configure TLB invalidation behaviour
1976 Format: { "0" | "1" }
1978 Request that DMA unmap operations use deferred
1979 invalidation of hardware TLBs, for increased
1980 throughput at the cost of reduced device isolation.
1981 Will fall back to strict mode if not supported by
1982 the relevant IOMMU driver.
1983 1 - Strict mode (default).
1984 DMA unmap operations invalidate IOMMU hardware TLBs
1988 [ARM64, X86] Configure DMA to bypass the IOMMU by default.
1989 Format: { "0" | "1" }
1990 0 - Use IOMMU translation for DMA.
1991 1 - Bypass the IOMMU for DMA.
1992 unset - Use value of CONFIG_IOMMU_DEFAULT_PASSTHROUGH.
1994 io7= [HW] IO7 for Marvel-based Alpha systems
1995 See comment before marvel_specify_io7 in
1996 arch/alpha/kernel/core_marvel.c.
1998 io_delay= [X86] I/O delay method
2000 Standard port 0x80 based delay
2002 Alternate port 0xed based delay (needed on some systems)
2004 Simple two microseconds delay
2009 See Documentation/admin-guide/nfs/nfsroot.rst.
2011 ipcmni_extend [KNL] Extend the maximum number of unique System V
2012 IPC identifiers from 32,768 to 16,777,216.
2014 irqaffinity= [SMP] Set the default irq affinity mask
2015 The argument is a cpu list, as described above.
2017 irqchip.gicv2_force_probe=
2020 Force the kernel to look for the second 4kB page
2021 of a GICv2 controller even if the memory range
2022 exposed by the device tree is too small.
2024 irqchip.gicv3_nolpi=
2026 Force the kernel to ignore the availability of
2027 LPIs (and by consequence ITSs). Intended for system
2028 that use the kernel as a bootloader, and thus want
2029 to let secondary kernels in charge of setting up
2032 irqchip.gicv3_pseudo_nmi= [ARM64]
2033 Enables support for pseudo-NMIs in the kernel. This
2034 requires the kernel to be built with
2035 CONFIG_ARM64_PSEUDO_NMI.
2038 When an interrupt is not handled search all handlers
2039 for it. Intended to get systems with badly broken
2043 When an interrupt is not handled search all handlers
2044 for it. Also check all handlers each timer
2045 interrupt. Intended to get systems with badly broken
2049 Format: <RDP>,<reset>,<pci_scan>,<verbosity>
2051 isolcpus= [KNL,SMP,ISOL] Isolate a given set of CPUs from disturbance.
2052 [Deprecated - use cpusets instead]
2053 Format: [flag-list,]<cpu-list>
2055 Specify one or more CPUs to isolate from disturbances
2056 specified in the flag list (default: domain):
2059 Disable the tick when a single task runs.
2061 A residual 1Hz tick is offloaded to workqueues, which you
2062 need to affine to housekeeping through the global
2063 workqueue's affinity configured via the
2064 /sys/devices/virtual/workqueue/cpumask sysfs file, or
2065 by using the 'domain' flag described below.
2067 NOTE: by default the global workqueue runs on all CPUs,
2068 so to protect individual CPUs the 'cpumask' file has to
2069 be configured manually after bootup.
2072 Isolate from the general SMP balancing and scheduling
2073 algorithms. Note that performing domain isolation this way
2074 is irreversible: it's not possible to bring back a CPU to
2075 the domains once isolated through isolcpus. It's strongly
2076 advised to use cpusets instead to disable scheduler load
2077 balancing through the "cpuset.sched_load_balance" file.
2078 It offers a much more flexible interface where CPUs can
2079 move in and out of an isolated set anytime.
2081 You can move a process onto or off an "isolated" CPU via
2082 the CPU affinity syscalls or cpuset.
2083 <cpu number> begins at 0 and the maximum value is
2084 "number of CPUs in system - 1".
2088 Isolate from being targeted by managed interrupts
2089 which have an interrupt mask containing isolated
2090 CPUs. The affinity of managed interrupts is
2091 handled by the kernel and cannot be changed via
2092 the /proc/irq/* interfaces.
2094 This isolation is best effort and only effective
2095 if the automatically assigned interrupt mask of a
2096 device queue contains isolated and housekeeping
2097 CPUs. If housekeeping CPUs are online then such
2098 interrupts are directed to the housekeeping CPU
2099 so that IO submitted on the housekeeping CPU
2100 cannot disturb the isolated CPU.
2102 If a queue's affinity mask contains only isolated
2103 CPUs then this parameter has no effect on the
2104 interrupt routing decision, though interrupts are
2105 only delivered when tasks running on those
2106 isolated CPUs submit IO. IO submitted on
2107 housekeeping CPUs has no influence on those
2110 The format of <cpu-list> is described above.
2114 ivrs_ioapic [HW,X86-64]
2115 Provide an override to the IOAPIC-ID<->DEVICE-ID
2116 mapping provided in the IVRS ACPI table. For
2117 example, to map IOAPIC-ID decimal 10 to
2118 PCI device 00:14.0 write the parameter as:
2119 ivrs_ioapic[10]=00:14.0
2121 ivrs_hpet [HW,X86-64]
2122 Provide an override to the HPET-ID<->DEVICE-ID
2123 mapping provided in the IVRS ACPI table. For
2124 example, to map HPET-ID decimal 0 to
2125 PCI device 00:14.0 write the parameter as:
2126 ivrs_hpet[0]=00:14.0
2128 ivrs_acpihid [HW,X86-64]
2129 Provide an override to the ACPI-HID:UID<->DEVICE-ID
2130 mapping provided in the IVRS ACPI table. For
2131 example, to map UART-HID:UID AMD0020:0 to
2132 PCI device 00:14.5 write the parameter as:
2133 ivrs_acpihid[00:14.5]=AMD0020:0
2135 js= [HW,JOY] Analog joystick
2136 See Documentation/input/joydev/joystick.rst.
2139 When CONFIG_RANDOMIZE_BASE is set, this disables
2140 kernel and module base offset ASLR (Address Space
2141 Layout Randomization).
2144 [KNL] Enforce KASAN (Kernel Address Sanitizer) to print
2145 report on every invalid memory access. Without this
2146 parameter KASAN will print report only for the first
2151 kernelcore= [KNL,X86,IA-64,PPC]
2152 Format: nn[KMGTPE] | nn% | "mirror"
2153 This parameter specifies the amount of memory usable by
2154 the kernel for non-movable allocations. The requested
2155 amount is spread evenly throughout all nodes in the
2156 system as ZONE_NORMAL. The remaining memory is used for
2157 movable memory in its own zone, ZONE_MOVABLE. In the
2158 event, a node is too small to have both ZONE_NORMAL and
2159 ZONE_MOVABLE, kernelcore memory will take priority and
2160 other nodes will have a larger ZONE_MOVABLE.
2162 ZONE_MOVABLE is used for the allocation of pages that
2163 may be reclaimed or moved by the page migration
2164 subsystem. Note that allocations like PTEs-from-HighMem
2165 still use the HighMem zone if it exists, and the Normal
2166 zone if it does not.
2168 It is possible to specify the exact amount of memory in
2169 the form of "nn[KMGTPE]", a percentage of total system
2170 memory in the form of "nn%", or "mirror". If "mirror"
2171 option is specified, mirrored (reliable) memory is used
2172 for non-movable allocations and remaining memory is used
2173 for Movable pages. "nn[KMGTPE]", "nn%", and "mirror"
2174 are exclusive, so you cannot specify multiple forms.
2176 kgdbdbgp= [KGDB,HW] kgdb over EHCI usb debug port.
2177 Format: <Controller#>[,poll interval]
2178 The controller # is the number of the ehci usb debug
2179 port as it is probed via PCI. The poll interval is
2180 optional and is the number seconds in between
2181 each poll cycle to the debug port in case you need
2182 the functionality for interrupting the kernel with
2183 gdb or control-c on the dbgp connection. When
2184 not using this parameter you use sysrq-g to break into
2185 the kernel debugger.
2187 kgdboc= [KGDB,HW] kgdb over consoles.
2188 Requires a tty driver that supports console polling,
2189 or a supported polling keyboard driver (non-usb).
2190 Serial only format: <serial_device>[,baud]
2191 keyboard only format: kbd
2192 keyboard and serial format: kbd,<serial_device>[,baud]
2193 Optional Kernel mode setting:
2194 kms, kbd format: kms,kbd
2195 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud]
2197 kgdboc_earlycon= [KGDB,HW]
2198 If the boot console provides the ability to read
2199 characters and can work in polling mode, you can use
2200 this parameter to tell kgdb to use it as a backend
2201 until the normal console is registered. Intended to
2202 be used together with the kgdboc parameter which
2203 specifies the normal console to transition to.
2205 The name of the early console should be specified
2206 as the value of this parameter. Note that the name of
2207 the early console might be different than the tty
2208 name passed to kgdboc. It's OK to leave the value
2209 blank and the first boot console that implements
2210 read() will be picked.
2212 kgdbwait [KGDB] Stop kernel execution and enter the
2213 kernel debugger at the earliest opportunity.
2215 kmac= [MIPS] Korina ethernet MAC address.
2216 Configure the RouterBoard 532 series on-chip
2217 Ethernet adapter MAC address.
2219 kmemleak= [KNL] Boot-time kmemleak enable/disable
2220 Valid arguments: on, off
2222 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y,
2225 kprobe_event=[probe-list]
2226 [FTRACE] Add kprobe events and enable at boot time.
2227 The probe-list is a semicolon delimited list of probe
2228 definitions. Each definition is same as kprobe_events
2229 interface, but the parameters are comma delimited.
2230 For example, to add a kprobe event on vfs_read with
2231 arg1 and arg2, add to the command line;
2233 kprobe_event=p,vfs_read,$arg1,$arg2
2235 See also Documentation/trace/kprobetrace.rst "Kernel
2236 Boot Parameter" section.
2238 kpti= [ARM64] Control page table isolation of user
2239 and kernel address spaces.
2240 Default: enabled on cores which need mitigation.
2244 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs.
2245 Default is 0 (don't ignore, but inject #GP)
2247 kvm.enable_vmware_backdoor=[KVM] Support VMware backdoor PV interface.
2248 Default is false (don't support).
2250 kvm.mmu_audit= [KVM] This is a R/W parameter which allows audit
2255 [KVM] Controls the software workaround for the
2256 X86_BUG_ITLB_MULTIHIT bug.
2257 force : Always deploy workaround.
2258 off : Never deploy workaround.
2259 auto : Deploy workaround based on the presence of
2260 X86_BUG_ITLB_MULTIHIT.
2264 If the software workaround is enabled for the host,
2265 guests do need not to enable it for nested guests.
2267 kvm.nx_huge_pages_recovery_ratio=
2268 [KVM] Controls how many 4KiB pages are periodically zapped
2269 back to huge pages. 0 disables the recovery, otherwise if
2270 the value is N KVM will zap 1/Nth of the 4KiB pages every
2271 minute. The default is 60.
2273 kvm-amd.nested= [KVM,AMD] Allow nested virtualization in KVM/SVM.
2274 Default is 1 (enabled)
2276 kvm-amd.npt= [KVM,AMD] Disable nested paging (virtualized MMU)
2278 Default is 1 (enabled) if in 64-bit or 32-bit PAE mode.
2281 [KVM,ARM] Select one of KVM/arm64's modes of operation.
2283 nvhe: Standard nVHE-based mode, without support for
2286 protected: nVHE-based mode with support for guests whose
2287 state is kept private from the host.
2288 Not valid if the kernel is running in EL2.
2290 Defaults to VHE/nVHE based on hardware support.
2292 kvm-arm.vgic_v3_group0_trap=
2293 [KVM,ARM] Trap guest accesses to GICv3 group-0
2296 kvm-arm.vgic_v3_group1_trap=
2297 [KVM,ARM] Trap guest accesses to GICv3 group-1
2300 kvm-arm.vgic_v3_common_trap=
2301 [KVM,ARM] Trap guest accesses to GICv3 common
2304 kvm-arm.vgic_v4_enable=
2305 [KVM,ARM] Allow use of GICv4 for direct injection of
2308 kvm_cma_resv_ratio=n [PPC]
2309 Reserves given percentage from system memory area for
2310 contiguous memory allocation for KVM hash pagetable
2312 By default it reserves 5% of total system memory.
2316 kvm-intel.ept= [KVM,Intel] Disable extended page tables
2317 (virtualized MMU) support on capable Intel chips.
2318 Default is 1 (enabled)
2320 kvm-intel.emulate_invalid_guest_state=
2321 [KVM,Intel] Enable emulation of invalid guest states
2322 Default is 0 (disabled)
2324 kvm-intel.flexpriority=
2325 [KVM,Intel] Disable FlexPriority feature (TPR shadow).
2326 Default is 1 (enabled)
2329 [KVM,Intel] Enable VMX nesting (nVMX).
2330 Default is 0 (disabled)
2332 kvm-intel.unrestricted_guest=
2333 [KVM,Intel] Disable unrestricted guest feature
2334 (virtualized real and unpaged mode) on capable
2335 Intel chips. Default is 1 (enabled)
2337 kvm-intel.vmentry_l1d_flush=[KVM,Intel] Mitigation for L1 Terminal Fault
2340 Valid arguments: never, cond, always
2342 always: L1D cache flush on every VMENTER.
2343 cond: Flush L1D on VMENTER only when the code between
2344 VMEXIT and VMENTER can leak host memory.
2345 never: Disables the mitigation
2347 Default is cond (do L1 cache flush in specific instances)
2349 kvm-intel.vpid= [KVM,Intel] Disable Virtual Processor Identification
2350 feature (tagged TLBs) on capable Intel chips.
2351 Default is 1 (enabled)
2353 l1tf= [X86] Control mitigation of the L1TF vulnerability on
2356 The kernel PTE inversion protection is unconditionally
2357 enabled and cannot be disabled.
2360 Provides all available mitigations for the
2361 L1TF vulnerability. Disables SMT and
2362 enables all mitigations in the
2363 hypervisors, i.e. unconditional L1D flush.
2365 SMT control and L1D flush control via the
2366 sysfs interface is still possible after
2367 boot. Hypervisors will issue a warning
2368 when the first VM is started in a
2369 potentially insecure configuration,
2370 i.e. SMT enabled or L1D flush disabled.
2373 Same as 'full', but disables SMT and L1D
2374 flush runtime control. Implies the
2375 'nosmt=force' command line option.
2376 (i.e. sysfs control of SMT is disabled.)
2379 Leaves SMT enabled and enables the default
2380 hypervisor mitigation, i.e. conditional
2383 SMT control and L1D flush control via the
2384 sysfs interface is still possible after
2385 boot. Hypervisors will issue a warning
2386 when the first VM is started in a
2387 potentially insecure configuration,
2388 i.e. SMT enabled or L1D flush disabled.
2392 Disables SMT and enables the default
2393 hypervisor mitigation.
2395 SMT control and L1D flush control via the
2396 sysfs interface is still possible after
2397 boot. Hypervisors will issue a warning
2398 when the first VM is started in a
2399 potentially insecure configuration,
2400 i.e. SMT enabled or L1D flush disabled.
2403 Same as 'flush', but hypervisors will not
2404 warn when a VM is started in a potentially
2405 insecure configuration.
2408 Disables hypervisor mitigations and doesn't
2410 It also drops the swap size and available
2411 RAM limit restriction on both hypervisor and
2416 For details see: Documentation/admin-guide/hw-vuln/l1tf.rst
2422 lapic [X86-32,APIC] Enable the local APIC even if BIOS
2425 lapic= [X86,APIC] Do not use TSC deadline
2426 value for LAPIC timer one-shot implementation. Default
2427 back to the programmable timer unit in the LAPIC.
2428 Format: notscdeadline
2430 lapic_timer_c2_ok [X86,APIC] trust the local apic timer
2433 libata.dma= [LIBATA] DMA control
2434 libata.dma=0 Disable all PATA and SATA DMA
2435 libata.dma=1 PATA and SATA Disk DMA only
2436 libata.dma=2 ATAPI (CDROM) DMA only
2437 libata.dma=4 Compact Flash DMA only
2438 Combinations also work, so libata.dma=3 enables DMA
2439 for disks and CDROMs, but not CFs.
2441 libata.ignore_hpa= [LIBATA] Ignore HPA limit
2442 libata.ignore_hpa=0 keep BIOS limits (default)
2443 libata.ignore_hpa=1 ignore limits, using full disk
2445 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume
2449 libata.force= [LIBATA] Force configurations. The format is comma-
2450 separated list of "[ID:]VAL" where ID is
2451 PORT[.DEVICE]. PORT and DEVICE are decimal numbers
2452 matching port, link or device. Basically, it matches
2453 the ATA ID string printed on console by libata. If
2454 the whole ID part is omitted, the last PORT and DEVICE
2455 values are used. If ID hasn't been specified yet, the
2456 configuration applies to all ports, links and devices.
2458 If only DEVICE is omitted, the parameter applies to
2459 the port and all links and devices behind it. DEVICE
2460 number of 0 either selects the first device or the
2461 first fan-out link behind PMP device. It does not
2462 select the host link. DEVICE number of 15 selects the
2463 host link and device attached to it.
2465 The VAL specifies the configuration to force. As long
2466 as there's no ambiguity shortcut notation is allowed.
2467 For example, both 1.5 and 1.5G would work for 1.5Gbps.
2468 The following configurations can be forced.
2470 * Cable type: 40c, 80c, short40c, unk, ign or sata.
2471 Any ID with matching PORT is used.
2473 * SATA link speed limit: 1.5Gbps or 3.0Gbps.
2475 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7].
2476 udma[/][16,25,33,44,66,100,133] notation is also
2479 * [no]ncq: Turn on or off NCQ.
2481 * [no]ncqtrim: Turn off queued DSM TRIM.
2483 * nohrst, nosrst, norst: suppress hard, soft
2486 * rstonce: only attempt one reset during
2487 hot-unplug link recovery
2489 * dump_id: dump IDENTIFY data.
2491 * atapi_dmadir: Enable ATAPI DMADIR bridge support
2493 * disable: Disable this device.
2495 If there are multiple matching configurations changing
2496 the same attribute, the last one is used.
2498 memblock=debug [KNL] Enable memblock debug messages.
2500 load_ramdisk= [RAM] [Deprecated]
2502 lockd.nlm_grace_period=P [NFS] Assign grace period.
2505 lockd.nlm_tcpport=N [NFS] Assign TCP port.
2508 lockd.nlm_timeout=T [NFS] Assign timeout value.
2511 lockd.nlm_udpport=M [NFS] Assign UDP port.
2514 lockdown= [SECURITY]
2515 { integrity | confidentiality }
2516 Enable the kernel lockdown feature. If set to
2517 integrity, kernel features that allow userland to
2518 modify the running kernel are disabled. If set to
2519 confidentiality, kernel features that allow userland
2520 to extract confidential information from the kernel
2523 locktorture.nreaders_stress= [KNL]
2524 Set the number of locking read-acquisition kthreads.
2525 Defaults to being automatically set based on the
2526 number of online CPUs.
2528 locktorture.nwriters_stress= [KNL]
2529 Set the number of locking write-acquisition kthreads.
2531 locktorture.onoff_holdoff= [KNL]
2532 Set time (s) after boot for CPU-hotplug testing.
2534 locktorture.onoff_interval= [KNL]
2535 Set time (s) between CPU-hotplug operations, or
2536 zero to disable CPU-hotplug testing.
2538 locktorture.shuffle_interval= [KNL]
2539 Set task-shuffle interval (jiffies). Shuffling
2540 tasks allows some CPUs to go into dyntick-idle
2541 mode during the locktorture test.
2543 locktorture.shutdown_secs= [KNL]
2544 Set time (s) after boot system shutdown. This
2545 is useful for hands-off automated testing.
2547 locktorture.stat_interval= [KNL]
2548 Time (s) between statistics printk()s.
2550 locktorture.stutter= [KNL]
2551 Time (s) to stutter testing, for example,
2552 specifying five seconds causes the test to run for
2553 five seconds, wait for five seconds, and so on.
2554 This tests the locking primitive's ability to
2555 transition abruptly to and from idle.
2557 locktorture.torture_type= [KNL]
2558 Specify the locking implementation to test.
2560 locktorture.verbose= [KNL]
2561 Enable additional printk() statements.
2563 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver
2566 loglevel= All Kernel Messages with a loglevel smaller than the
2567 console loglevel will be printed to the console. It can
2568 also be changed with klogd or other programs. The
2569 loglevels are defined as follows:
2571 0 (KERN_EMERG) system is unusable
2572 1 (KERN_ALERT) action must be taken immediately
2573 2 (KERN_CRIT) critical conditions
2574 3 (KERN_ERR) error conditions
2575 4 (KERN_WARNING) warning conditions
2576 5 (KERN_NOTICE) normal but significant condition
2577 6 (KERN_INFO) informational
2578 7 (KERN_DEBUG) debug-level messages
2580 log_buf_len=n[KMG] Sets the size of the printk ring buffer,
2581 in bytes. n must be a power of two and greater
2582 than the minimal size. The minimal size is defined
2583 by LOG_BUF_SHIFT kernel config parameter. There is
2584 also CONFIG_LOG_CPU_MAX_BUF_SHIFT config parameter
2585 that allows to increase the default size depending on
2586 the number of CPUs. See init/Kconfig for more details.
2588 logo.nologo [FB] Disables display of the built-in Linux logo.
2589 This may be used to provide more screen space for
2590 kernel log messages and is useful when debugging
2591 kernel boot problems.
2593 lp=0 [LP] Specify parallel ports to use, e.g,
2594 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses
2595 lp=reset first parallel port). 'lp=0' disables the
2596 lp=auto printer driver. 'lp=reset' (which can be
2597 specified in addition to the ports) causes
2598 attached printers to be reset. Using
2599 lp=port1,port2,... specifies the parallel ports
2600 to associate lp devices with, starting with
2601 lp0. A port specification may be 'none' to skip
2602 that lp device, or a parport name such as
2603 'parport0'. Specifying 'lp=auto' instead of a
2604 port specification list means that device IDs
2605 from each port should be examined, to see if
2606 an IEEE 1284-compliant printer is attached; if
2607 so, the driver will manage that printer.
2608 See also header of drivers/char/lp.c.
2611 Sets loops_per_jiffy to given constant, thus avoiding
2612 time-consuming boot-time autodetection (up to 250 ms per
2613 CPU). 0 enables autodetection (default). To determine
2614 the correct value for your kernel, boot with normal
2615 autodetection and see what value is printed. Note that
2616 on SMP systems the preset will be applied to all CPUs,
2617 which is likely to cause problems if your CPUs need
2618 significantly divergent settings. An incorrect value
2619 will cause delays in the kernel to be wrong, leading to
2620 unpredictable I/O errors and other breakage. Although
2621 unlikely, in the extreme case this might damage your
2625 Format: <io>,<irq>,<dma>
2627 lsm.debug [SECURITY] Enable LSM initialization debugging output.
2630 [SECURITY] Choose order of LSM initialization. This
2631 overrides CONFIG_LSM, and the "security=" parameter.
2633 machvec= [IA-64] Force the use of a particular machine-vector
2634 (machvec) in a generic kernel.
2635 Example: machvec=hpzx1
2637 machtype= [Loongson] Share the same kernel image file between
2638 different yeeloong laptops.
2639 Example: machtype=lemote-yeeloong-2f-7inch
2641 max_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory greater
2642 than or equal to this physical address is ignored.
2644 maxcpus= [SMP] Maximum number of processors that an SMP kernel
2645 will bring up during bootup. maxcpus=n : n >= 0 limits
2646 the kernel to bring up 'n' processors. Surely after
2647 bootup you can bring up the other plugged cpu by executing
2648 "echo 1 > /sys/devices/system/cpu/cpuX/online". So maxcpus
2649 only takes effect during system bootup.
2650 While n=0 is a special case, it is equivalent to "nosmp",
2651 which also disables the IO APIC.
2653 max_loop= [LOOP] The number of loop block devices that get
2654 (loop.max_loop) unconditionally pre-created at init time. The default
2655 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead
2656 of statically allocating a predefined number, loop
2657 devices can be requested on-demand with the
2658 /dev/loop-control interface.
2660 mce [X86-32] Machine Check Exception
2662 mce=option [X86-64] See Documentation/x86/x86_64/boot-options.rst
2664 md= [HW] RAID subsystems devices and level
2665 See Documentation/admin-guide/md.rst.
2668 Format: <first>,<last>
2669 Specifies range of consoles to be captured by the MDA.
2672 Control mitigation for the Micro-architectural Data
2673 Sampling (MDS) vulnerability.
2675 Certain CPUs are vulnerable to an exploit against CPU
2676 internal buffers which can forward information to a
2677 disclosure gadget under certain conditions.
2679 In vulnerable processors, the speculatively
2680 forwarded data can be used in a cache side channel
2681 attack, to access data to which the attacker does
2682 not have direct access.
2684 This parameter controls the MDS mitigation. The
2687 full - Enable MDS mitigation on vulnerable CPUs
2688 full,nosmt - Enable MDS mitigation and disable
2689 SMT on vulnerable CPUs
2690 off - Unconditionally disable MDS mitigation
2692 On TAA-affected machines, mds=off can be prevented by
2693 an active TAA mitigation as both vulnerabilities are
2694 mitigated with the same mechanism so in order to disable
2695 this mitigation, you need to specify tsx_async_abort=off
2698 Not specifying this option is equivalent to
2701 For details see: Documentation/admin-guide/hw-vuln/mds.rst
2703 mem=nn[KMG] [KNL,BOOT] Force usage of a specific amount of memory
2704 Amount of memory to be used in cases as follows:
2707 2 when the kernel is not able to see the whole system memory;
2708 3 memory that lies after 'mem=' boundary is excluded from
2709 the hypervisor, then assigned to KVM guests.
2711 [X86] Work as limiting max address. Use together
2712 with memmap= to avoid physical address space collisions.
2713 Without memmap= PCI devices could be placed at addresses
2714 belonging to unused RAM.
2716 Note that this only takes effects during boot time since
2717 in above case 3, memory may need be hot added after boot
2718 if system memory of hypervisor is not sufficient.
2720 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel
2724 [KNL,SH] Allow user to override the default size for
2725 per-device physically contiguous DMA buffers.
2727 memhp_default_state=online/offline
2728 [KNL] Set the initial state for the memory hotplug
2729 onlining policy. If not specified, the default value is
2730 set according to the
2731 CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE kernel config
2733 See Documentation/admin-guide/mm/memory-hotplug.rst.
2735 memmap=exactmap [KNL,X86] Enable setting of an exact
2736 E820 memory map, as specified by the user.
2737 Such memmap=exactmap lines can be constructed based on
2738 BIOS output or other requirements. See the memmap=nn@ss
2741 memmap=nn[KMG]@ss[KMG]
2742 [KNL, X86, MIPS, XTENSA] Force usage of a specific region of memory.
2743 Region of memory to be used is from ss to ss+nn.
2744 If @ss[KMG] is omitted, it is equivalent to mem=nn[KMG],
2745 which limits max address to nn[KMG].
2746 Multiple different regions can be specified,
2749 memmap=100M@2G,100M#3G,1G!1024G
2751 memmap=nn[KMG]#ss[KMG]
2752 [KNL,ACPI] Mark specific memory as ACPI data.
2753 Region of memory to be marked is from ss to ss+nn.
2755 memmap=nn[KMG]$ss[KMG]
2756 [KNL,ACPI] Mark specific memory as reserved.
2757 Region of memory to be reserved is from ss to ss+nn.
2758 Example: Exclude memory from 0x18690000-0x1869ffff
2759 memmap=64K$0x18690000
2761 memmap=0x10000$0x18690000
2762 Some bootloaders may need an escape character before '$',
2763 like Grub2, otherwise '$' and the following number
2766 memmap=nn[KMG]!ss[KMG]
2767 [KNL,X86] Mark specific memory as protected.
2768 Region of memory to be used, from ss to ss+nn.
2769 The memory region may be marked as e820 type 12 (0xc)
2770 and is NVDIMM or ADR memory.
2772 memmap=<size>%<offset>-<oldtype>+<newtype>
2773 [KNL,ACPI] Convert memory within the specified region
2774 from <oldtype> to <newtype>. If "-<oldtype>" is left
2775 out, the whole region will be marked as <newtype>,
2776 even if previously unavailable. If "+<newtype>" is left
2777 out, matching memory will be removed. Types are
2778 specified as e820 types, e.g., 1 = RAM, 2 = reserved,
2779 3 = ACPI, 12 = PRAM.
2781 memory_corruption_check=0/1 [X86]
2782 Some BIOSes seem to corrupt the first 64k of
2783 memory when doing things like suspend/resume.
2784 Setting this option will scan the memory
2785 looking for corruption. Enabling this will
2786 both detect corruption and prevent the kernel
2787 from using the memory being corrupted.
2788 However, its intended as a diagnostic tool; if
2789 repeatable BIOS-originated corruption always
2790 affects the same memory, you can use memmap=
2791 to prevent the kernel from using that memory.
2793 memory_corruption_check_size=size [X86]
2794 By default it checks for corruption in the low
2795 64k, making this memory unavailable for normal
2796 use. Use this parameter to scan for
2797 corruption in more or less memory.
2799 memory_corruption_check_period=seconds [X86]
2800 By default it checks for corruption every 60
2801 seconds. Use this parameter to check at some
2802 other rate. 0 disables periodic checking.
2804 memtest= [KNL,X86,ARM,PPC] Enable memtest
2806 default : 0 <disable>
2807 Specifies the number of memtest passes to be
2808 performed. Each pass selects another test
2809 pattern from a given set of patterns. Memtest
2810 fills the memory with this pattern, validates
2811 memory contents and reserves bad memory
2812 regions that are detected.
2814 mem_encrypt= [X86-64] AMD Secure Memory Encryption (SME) control
2815 Valid arguments: on, off
2816 Default (depends on kernel configuration option):
2817 on (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=y)
2818 off (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=n)
2819 mem_encrypt=on: Activate SME
2820 mem_encrypt=off: Do not activate SME
2822 Refer to Documentation/virt/kvm/amd-memory-encryption.rst
2823 for details on when memory encryption can be activated.
2825 mem_sleep_default= [SUSPEND] Default system suspend mode:
2826 s2idle - Suspend-To-Idle
2827 shallow - Power-On Suspend or equivalent (if supported)
2828 deep - Suspend-To-RAM or equivalent (if supported)
2829 See Documentation/admin-guide/pm/sleep-states.rst.
2831 meye.*= [HW] Set MotionEye Camera parameters
2832 See Documentation/admin-guide/media/meye.rst.
2834 mfgpt_irq= [IA-32] Specify the IRQ to use for the
2835 Multi-Function General Purpose Timers on AMD Geode
2838 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when
2839 the BIOS has incorrectly applied a workaround. TinyBIOS
2840 version 0.98 is known to be affected, 0.99 fixes the
2841 problem by letting the user disable the workaround.
2845 min_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory below this
2846 physical address is ignored.
2848 mini2440= [ARM,HW,KNL]
2849 Format:[0..2][b][c][t]
2851 MINI2440 configuration specification:
2852 0 - The attached screen is the 3.5" TFT
2853 1 - The attached screen is the 7" TFT
2854 2 - The VGA Shield is attached (1024x768)
2855 Leaving out the screen size parameter will not load
2856 the TFT driver, and the framebuffer will be left
2858 b - Enable backlight. The TFT backlight pin will be
2859 linked to the kernel VESA blanking code and a GPIO
2860 LED. This parameter is not necessary when using the
2862 c - Enable the s3c camera interface.
2863 t - Reserved for enabling touchscreen support. The
2864 touchscreen support is not enabled in the mainstream
2865 kernel as of 2.6.30, a preliminary port can be found
2866 in the "bleeding edge" mini2440 support kernel at
2867 https://repo.or.cz/w/linux-2.6/mini2440.git
2870 [X86,PPC,S390,ARM64] Control optional mitigations for
2871 CPU vulnerabilities. This is a set of curated,
2872 arch-independent options, each of which is an
2873 aggregation of existing arch-specific options.
2876 Disable all optional CPU mitigations. This
2877 improves system performance, but it may also
2878 expose users to several CPU vulnerabilities.
2879 Equivalent to: nopti [X86,PPC]
2881 nospectre_v1 [X86,PPC]
2883 nospectre_v2 [X86,PPC,S390,ARM64]
2884 spectre_v2_user=off [X86]
2885 spec_store_bypass_disable=off [X86,PPC]
2886 ssbd=force-off [ARM64]
2889 tsx_async_abort=off [X86]
2890 kvm.nx_huge_pages=off [X86]
2891 no_entry_flush [PPC]
2892 no_uaccess_flush [PPC]
2895 This does not have any effect on
2896 kvm.nx_huge_pages when
2897 kvm.nx_huge_pages=force.
2900 Mitigate all CPU vulnerabilities, but leave SMT
2901 enabled, even if it's vulnerable. This is for
2902 users who don't want to be surprised by SMT
2903 getting disabled across kernel upgrades, or who
2904 have other ways of avoiding SMT-based attacks.
2905 Equivalent to: (default behavior)
2908 Mitigate all CPU vulnerabilities, disabling SMT
2909 if needed. This is for users who always want to
2910 be fully mitigated, even if it means losing SMT.
2911 Equivalent to: l1tf=flush,nosmt [X86]
2912 mds=full,nosmt [X86]
2913 tsx_async_abort=full,nosmt [X86]
2916 [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this
2917 parameter allows control of the logging verbosity for
2918 the additional memory initialisation checks. A value
2919 of 0 disables mminit logging and a level of 4 will
2920 log everything. Information is printed at KERN_DEBUG
2921 so loglevel=8 may also need to be specified.
2924 [KNL] When CONFIG_MODULE_SIG is set, this means that
2925 modules without (valid) signatures will fail to load.
2926 Note that if CONFIG_MODULE_SIG_FORCE is set, that
2927 is always true, so this option does nothing.
2929 module_blacklist= [KNL] Do not load a comma-separated list of
2930 modules. Useful for debugging problem modules.
2933 [MOUSE] Maximum time between finger touching and
2934 leaving touchpad surface for touch to be considered
2935 a tap and be reported as a left button click (for
2936 touchpads working in absolute mode only).
2938 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices
2939 reporting absolute coordinates, such as tablets
2940 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices
2941 reporting absolute coordinates, such as tablets
2943 movablecore= [KNL,X86,IA-64,PPC]
2944 Format: nn[KMGTPE] | nn%
2945 This parameter is the complement to kernelcore=, it
2946 specifies the amount of memory used for migratable
2947 allocations. If both kernelcore and movablecore is
2948 specified, then kernelcore will be at *least* the
2949 specified value but may be more. If movablecore on its
2950 own is specified, the administrator must be careful
2951 that the amount of memory usable for all allocations
2954 movable_node [KNL] Boot-time switch to make hotplugable memory
2955 NUMA nodes to be movable. This means that the memory
2956 of such nodes will be usable only for movable
2957 allocations which rules out almost all kernel
2958 allocations. Use with caution!
2960 MTD_Partition= [MTD]
2961 Format: <name>,<region-number>,<size>,<offset>
2963 MTD_Region= [MTD] Format:
2964 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>]
2967 See drivers/mtd/parsers/cmdlinepart.c
2969 multitce=off [PPC] This parameter disables the use of the pSeries
2970 firmware feature for updating multiple TCE entries
2973 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration
2975 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock]
2977 boundary - index of last SLC block on Flex-OneNAND.
2978 The remaining blocks are configured as MLC blocks.
2979 lock - Configure if Flex-OneNAND boundary should be locked.
2980 Once locked, the boundary cannot be changed.
2981 1 indicates lock status, 0 indicates unlock status.
2984 ARM/S3C2412 JIVE boot control
2986 See arch/arm/mach-s3c/mach-jive.c
2988 mtouchusb.raw_coordinates=
2989 [HW] Make the MicroTouch USB driver use raw coordinates
2990 ('y', default) or cooked coordinates ('n')
2992 mtrr_chunk_size=nn[KMG] [X86]
2993 used for mtrr cleanup. It is largest continuous chunk
2994 that could hold holes aka. UC entries.
2996 mtrr_gran_size=nn[KMG] [X86]
2997 Used for mtrr cleanup. It is granularity of mtrr block.
2999 Large value could prevent small alignment from
3002 mtrr_spare_reg_nr=n [X86]
3004 Range: 0,7 : spare reg number
3006 Used for mtrr cleanup. It is spare mtrr entries number.
3007 Set to 2 or more if your graphical card needs more.
3009 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card
3011 netdev= [NET] Network devices parameters
3012 Format: <irq>,<io>,<mem_start>,<mem_end>,<name>
3013 Note that mem_start is often overloaded to mean
3014 something different and driver-specific.
3015 This usage is only documented in each driver source
3019 [NETFILTER] Enable connection tracking flow accounting
3020 0 to disable accounting
3021 1 to enable accounting
3024 nfsaddrs= [NFS] Deprecated. Use ip= instead.
3025 See Documentation/admin-guide/nfs/nfsroot.rst.
3027 nfsroot= [NFS] nfs root filesystem for disk-less boxes.
3028 See Documentation/admin-guide/nfs/nfsroot.rst.
3030 nfsrootdebug [NFS] enable nfsroot debugging messages.
3031 See Documentation/admin-guide/nfs/nfsroot.rst.
3033 nfs.callback_nr_threads=
3034 [NFSv4] set the total number of threads that the
3035 NFS client will assign to service NFSv4 callback
3038 nfs.callback_tcpport=
3039 [NFS] set the TCP port on which the NFSv4 callback
3040 channel should listen.
3043 [NFS] sets the pathname to the program which is used
3044 to update the NFS client cache entries.
3046 nfs.cache_getent_timeout=
3047 [NFS] sets the timeout after which an attempt to
3048 update a cache entry is deemed to have failed.
3050 nfs.idmap_cache_timeout=
3051 [NFS] set the maximum lifetime for idmapper cache
3055 [NFS] enable 64-bit inode numbers.
3056 If zero, the NFS client will fake up a 32-bit inode
3057 number for the readdir() and stat() syscalls instead
3058 of returning the full 64-bit number.
3059 The default is to return 64-bit inode numbers.
3061 nfs.max_session_cb_slots=
3062 [NFSv4.1] Sets the maximum number of session
3063 slots the client will assign to the callback
3064 channel. This determines the maximum number of
3065 callbacks the client will process in parallel for
3066 a particular server.
3068 nfs.max_session_slots=
3069 [NFSv4.1] Sets the maximum number of session slots
3070 the client will attempt to negotiate with the server.
3071 This limits the number of simultaneous RPC requests
3072 that the client can send to the NFSv4.1 server.
3073 Note that there is little point in setting this
3074 value higher than the max_tcp_slot_table_limit.
3076 nfs.nfs4_disable_idmapping=
3077 [NFSv4] When set to the default of '1', this option
3078 ensures that both the RPC level authentication
3079 scheme and the NFS level operations agree to use
3080 numeric uids/gids if the mount is using the
3081 'sec=sys' security flavour. In effect it is
3082 disabling idmapping, which can make migration from
3083 legacy NFSv2/v3 systems to NFSv4 easier.
3084 Servers that do not support this mode of operation
3085 will be autodetected by the client, and it will fall
3086 back to using the idmapper.
3087 To turn off this behaviour, set the value to '0'.
3089 [NFS4] Specify an additional fixed unique ident-
3090 ification string that NFSv4 clients can insert into
3091 their nfs_client_id4 string. This is typically a
3092 UUID that is generated at system install time.
3094 nfs.send_implementation_id =
3095 [NFSv4.1] Send client implementation identification
3096 information in exchange_id requests.
3097 If zero, no implementation identification information
3099 The default is to send the implementation identification
3102 nfs.recover_lost_locks =
3103 [NFSv4] Attempt to recover locks that were lost due
3104 to a lease timeout on the server. Please note that
3105 doing this risks data corruption, since there are
3106 no guarantees that the file will remain unchanged
3107 after the locks are lost.
3108 If you want to enable the kernel legacy behaviour of
3109 attempting to recover these locks, then set this
3111 The default parameter value of '0' causes the kernel
3112 not to attempt recovery of lost locks.
3114 nfs4.layoutstats_timer =
3115 [NFSv4.2] Change the rate at which the kernel sends
3116 layoutstats to the pNFS metadata server.
3118 Setting this to value to 0 causes the kernel to use
3119 whatever value is the default set by the layout
3120 driver. A non-zero value sets the minimum interval
3121 in seconds between layoutstats transmissions.
3123 nfsd.nfs4_disable_idmapping=
3124 [NFSv4] When set to the default of '1', the NFSv4
3125 server will return only numeric uids and gids to
3126 clients using auth_sys, and will accept numeric uids
3127 and gids from such clients. This is intended to ease
3128 migration from NFSv2/v3.
3130 nmi_backtrace.backtrace_idle [KNL]
3131 Dump stacks even of idle CPUs in response to an
3132 NMI stack-backtrace request.
3134 nmi_debug= [KNL,SH] Specify one or more actions to take
3135 when a NMI is triggered.
3136 Format: [state][,regs][,debounce][,die]
3138 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels
3139 Format: [panic,][nopanic,][num]
3141 0 - turn hardlockup detector in nmi_watchdog off
3142 1 - turn hardlockup detector in nmi_watchdog on
3143 When panic is specified, panic when an NMI watchdog
3144 timeout occurs (or 'nopanic' to not panic on an NMI
3145 watchdog, if CONFIG_BOOTPARAM_HARDLOCKUP_PANIC is set)
3146 To disable both hard and soft lockup detectors,
3147 please see 'nowatchdog'.
3148 This is useful when you use a panic=... timeout and
3149 need the box quickly up again.
3151 These settings can be accessed at runtime via
3152 the nmi_watchdog and hardlockup_panic sysctls.
3154 netpoll.carrier_timeout=
3155 [NET] Specifies amount of time (in seconds) that
3156 netpoll should wait for a carrier. By default netpoll
3159 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths
3160 emulation library even if a 387 maths coprocessor
3163 no5lvl [X86-64] Disable 5-level paging mode. Forces
3164 kernel to use 4-level paging instead.
3166 nofsgsbase [X86] Disables FSGSBASE instructions.
3169 [HW] Never suspend the console
3170 Disable suspending of consoles during suspend and
3171 hibernate operations. Once disabled, debugging
3172 messages can reach various consoles while the rest
3173 of the system is being put to sleep (ie, while
3174 debugging driver suspend/resume hooks). This may
3175 not work reliably with all consoles, but is known
3176 to work with serial and VGA consoles.
3177 To facilitate more flexible debugging, we also add
3178 console_suspend, a printk module parameter to control
3179 it. Users could use console_suspend (usually
3180 /sys/module/printk/parameters/console_suspend) to
3181 turn on/off it dynamically.
3183 novmcoredd [KNL,KDUMP]
3184 Disable device dump. Device dump allows drivers to
3185 append dump data to vmcore so you can collect driver
3186 specified debug info. Drivers can append the data
3187 without any limit and this data is stored in memory,
3188 so this may cause significant memory stress. Disabling
3189 device dump can help save memory but the driver debug
3190 data will be no longer available. This parameter
3191 is only available when CONFIG_PROC_VMCORE_DEVICE_DUMP
3194 noaliencache [MM, NUMA, SLAB] Disables the allocation of alien
3195 caches in the slab allocator. Saves per-node memory,
3196 but will impact performance.
3200 noaltinstr [S390] Disables alternative instructions patching
3201 (CPU alternatives feature).
3203 noapic [SMP,APIC] Tells the kernel to not make use of any
3204 IOAPICs that may be present in the system.
3206 noautogroup Disable scheduler automatic task group creation.
3208 nobats [PPC] Do not use BATs for mapping kernel lowmem
3209 on "Classic" PPC cores.
3213 noclflush [BUGS=X86] Don't use the CLFLUSH instruction
3215 nodelayacct [KNL] Disable per-task delay accounting
3217 nodsp [SH] Disable hardware DSP at boot time.
3219 noefi Disable EFI runtime services support.
3221 no_entry_flush [PPC] Don't flush the L1-D cache when entering the kernel.
3226 On X86-32 available only on PAE configured kernels.
3227 noexec=on: enable non-executable mappings (default)
3228 noexec=off: disable non-executable mappings
3231 Disable SMAP (Supervisor Mode Access Prevention)
3232 even if it is supported by processor.
3235 Disable SMEP (Supervisor Mode Execution Prevention)
3236 even if it is supported by processor.
3239 This affects only 32-bit executables.
3240 noexec32=on: enable non-executable mappings (default)
3241 read doesn't imply executable mappings
3242 noexec32=off: disable non-executable mappings
3243 read implies executable mappings
3245 nofpu [MIPS,SH] Disable hardware FPU at boot time.
3247 nofxsr [BUGS=X86-32] Disables x86 floating point extended
3248 register save and restore. The kernel will only save
3249 legacy floating-point registers on task switch.
3251 nohugeiomap [KNL,X86,PPC,ARM64] Disable kernel huge I/O mappings.
3253 nosmt [KNL,S390] Disable symmetric multithreading (SMT).
3254 Equivalent to smt=1.
3256 [KNL,X86] Disable symmetric multithreading (SMT).
3257 nosmt=force: Force disable SMT, cannot be undone
3258 via the sysfs control file.
3260 nospectre_v1 [X86,PPC] Disable mitigations for Spectre Variant 1
3261 (bounds check bypass). With this option data leaks are
3262 possible in the system.
3264 nospectre_v2 [X86,PPC_FSL_BOOK3E,ARM64] Disable all mitigations for
3265 the Spectre variant 2 (indirect branch prediction)
3266 vulnerability. System may allow data leaks with this
3269 nospec_store_bypass_disable
3270 [HW] Disable all mitigations for the Speculative Store Bypass vulnerability
3273 [PPC] Don't flush the L1-D cache after accessing user data.
3275 noxsave [BUGS=X86] Disables x86 extended register state save
3276 and restore using xsave. The kernel will fallback to
3277 enabling legacy floating-point and sse state.
3279 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended
3280 register states. The kernel will fall back to use
3281 xsave to save the states. By using this parameter,
3282 performance of saving the states is degraded because
3283 xsave doesn't support modified optimization while
3284 xsaveopt supports it on xsaveopt enabled systems.
3286 noxsaves [X86] Disables xsaves and xrstors used in saving and
3287 restoring x86 extended register state in compacted
3288 form of xsave area. The kernel will fall back to use
3289 xsaveopt and xrstor to save and restore the states
3290 in standard form of xsave area. By using this
3291 parameter, xsave area per process might occupy more
3292 memory on xsaves enabled systems.
3294 nohlt [ARM,ARM64,MICROBLAZE,SH] Forces the kernel to busy wait
3295 in do_idle() and not use the arch_cpu_idle()
3296 implementation; requires CONFIG_GENERIC_IDLE_POLL_SETUP
3297 to be effective. This is useful on platforms where the
3298 sleep(SH) or wfi(ARM,ARM64) instructions do not work
3299 correctly or when doing power measurements to evalute
3300 the impact of the sleep instructions. This is also
3301 useful when using JTAG debugger.
3303 no_file_caps Tells the kernel not to honor file capabilities. The
3304 only way then for a file to be executed with privilege
3305 is to be setuid root or executed by root.
3307 nohalt [IA-64] Tells the kernel not to use the power saving
3308 function PAL_HALT_LIGHT when idle. This increases
3309 power-consumption. On the positive side, it reduces
3310 interrupt wake-up latency, which may improve performance
3311 in certain environments such as networked servers or
3315 Force pointers printed to the console or buffers to be
3316 unhashed. By default, when a pointer is printed via %p
3317 format string, that pointer is "hashed", i.e. obscured
3318 by hashing the pointer value. This is a security feature
3319 that hides actual kernel addresses from unprivileged
3320 users, but it also makes debugging the kernel more
3321 difficult since unequal pointers can no longer be
3322 compared. However, if this command-line option is
3323 specified, then all normal pointers will have their true
3324 value printed. Pointers printed via %pK may still be
3325 hashed. This option should only be specified when
3326 debugging the kernel. Please do not use on production
3329 nohibernate [HIBERNATION] Disable hibernation and resume.
3331 nohz= [KNL] Boottime enable/disable dynamic ticks
3332 Valid arguments: on, off
3335 nohz_full= [KNL,BOOT,SMP,ISOL]
3336 The argument is a cpu list, as described above.
3337 In kernels built with CONFIG_NO_HZ_FULL=y, set
3338 the specified list of CPUs whose tick will be stopped
3339 whenever possible. The boot CPU will be forced outside
3340 the range to maintain the timekeeping. Any CPUs
3341 in this list will have their RCU callbacks offloaded,
3342 just as if they had also been called out in the
3343 rcu_nocbs= boot parameter.
3345 noiotrap [SH] Disables trapped I/O port accesses.
3347 noirqdebug [X86-32] Disables the code which attempts to detect and
3348 disable unhandled interrupt sources.
3350 no_timer_check [X86,APIC] Disables the code which tests for
3351 broken timer IRQ sources.
3353 noisapnp [ISAPNP] Disables ISA PnP code.
3355 noinitrd [RAM] Tells the kernel not to load any configured
3358 nointremap [X86-64, Intel-IOMMU] Do not enable interrupt
3360 [Deprecated - use intremap=off]
3364 noinvpcid [X86] Disable the INVPCID cpu feature.
3366 nojitter [IA-64] Disables jitter checking for ITC timers.
3368 no-kvmclock [X86,KVM] Disable paravirtualized KVM clock driver
3370 no-kvmapf [X86,KVM] Disable paravirtualized asynchronous page
3374 [X86,PV_OPS] Disable paravirtualized VMware scheduler
3375 clock and use the default one.
3377 no-steal-acc [X86,PV_OPS,ARM64] Disable paravirtualized steal time
3378 accounting. steal time is computed, but won't
3379 influence scheduler behaviour
3381 nolapic [X86-32,APIC] Do not enable or use the local APIC.
3383 nolapic_timer [X86-32,APIC] Do not use the local APIC timer.
3385 noltlbs [PPC] Do not use large page/tlb entries for kernel
3386 lowmem mapping on PPC40x and PPC8xx
3388 nomca [IA-64] Disable machine check abort handling
3390 nomce [X86-32] Disable Machine Check Exception
3392 nomfgpt [X86-32] Disable Multi-Function General Purpose
3393 Timer usage (for AMD Geode machines).
3395 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to
3396 shutdown the other cpus. Instead use the REBOOT_VECTOR
3399 nomodule Disable module load
3401 nopat [X86] Disable PAT (page attribute table extension of
3402 pagetables) support.
3404 nopcid [X86-64] Disable the PCID cpu feature.
3406 norandmaps Don't use address space randomization. Equivalent to
3407 echo 0 > /proc/sys/kernel/randomize_va_space
3409 noreplace-smp [X86-32,SMP] Don't replace SMP instructions
3410 with UP alternatives
3412 nordrand [X86] Disable kernel use of the RDRAND and
3413 RDSEED instructions even if they are supported
3414 by the processor. RDRAND and RDSEED are still
3415 available to user space applications.
3417 noresume [SWSUSP] Disables resume and restores original swap
3420 no-scroll [VGA] Disables scrollback.
3421 This is required for the Braillex ib80-piezo Braille
3422 reader made by F.H. Papenmeier (Germany).
3426 nosep [BUGS=X86-32] Disables x86 SYSENTER/SYSEXIT support.
3428 nosgx [X86-64,SGX] Disables Intel SGX kernel support.
3430 nosmp [SMP] Tells an SMP kernel to act as a UP kernel,
3431 and disable the IO APIC. legacy for "maxcpus=0".
3433 nosoftlockup [KNL] Disable the soft-lockup detector.
3435 nosync [HW,M68K] Disables sync negotiation for all devices.
3437 nowatchdog [KNL] Disable both lockup detectors, i.e.
3438 soft-lockup and NMI watchdog (hard-lockup).
3442 nox2apic [X86-64,APIC] Do not enable x2APIC mode.
3444 cpu0_hotplug [X86] Turn on CPU0 hotplug feature when
3445 CONFIG_BOOTPARAM_HOTPLUG_CPU0 is off.
3446 Some features depend on CPU0. Known dependencies are:
3447 1. Resume from suspend/hibernate depends on CPU0.
3448 Suspend/hibernate will fail if CPU0 is offline and you
3449 need to online CPU0 before suspend/hibernate.
3450 2. PIC interrupts also depend on CPU0. CPU0 can't be
3451 removed if a PIC interrupt is detected.
3452 It's said poweroff/reboot may depend on CPU0 on some
3453 machines although I haven't seen such issues so far
3454 after CPU0 is offline on a few tested machines.
3455 If the dependencies are under your control, you can
3456 turn on cpu0_hotplug.
3458 nps_mtm_hs_ctr= [KNL,ARC]
3459 This parameter sets the maximum duration, in
3460 cycles, each HW thread of the CTOP can run
3461 without interruptions, before HW switches it.
3462 The actual maximum duration is 16 times this
3464 Format: integer between 1 and 255
3467 nptcg= [IA-64] Override max number of concurrent global TLB
3468 purges which is reported from either PAL_VM_SUMMARY or
3471 nr_cpus= [SMP] Maximum number of processors that an SMP kernel
3472 could support. nr_cpus=n : n >= 1 limits the kernel to
3473 support 'n' processors. It could be larger than the
3474 number of already plugged CPU during bootup, later in
3475 runtime you can physically add extra cpu until it reaches
3476 n. So during boot up some boot time memory for per-cpu
3477 variables need be pre-allocated for later physical cpu
3480 nr_uarts= [SERIAL] maximum number of UARTs to be registered.
3482 numa_balancing= [KNL,ARM64,PPC,RISCV,S390,X86] Enable or disable automatic
3484 Allowed values are enable and disable
3486 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA.
3487 'node', 'default' can be specified
3488 This can be set from sysctl after boot.
3489 See Documentation/admin-guide/sysctl/vm.rst for details.
3491 ohci1394_dma=early [HW] enable debugging via the ohci1394 driver.
3492 See Documentation/core-api/debugging-via-ohci1394.rst for more
3495 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands
3496 Rather than timing out after 20 ms if an EC
3497 command is not properly ACKed, override the length
3498 of the timeout. We have interrupts disabled while
3499 waiting for the ACK, so if this is set too high
3500 interrupts *may* be lost!
3502 omap_mux= [OMAP] Override bootloader pin multiplexing.
3503 Format: <mux_mode0.mode_name=value>...
3504 For example, to override I2C bus2:
3505 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100
3507 oops=panic Always panic on oopses. Default is to just kill the
3508 process, but there is a small probability of
3509 deadlocking the machine.
3510 This will also cause panics on machine check exceptions.
3511 Useful together with panic=30 to trigger a reboot.
3514 [KNL] Boolean flag to control whether the page allocator
3515 should randomize its free lists. The randomization may
3516 be automatically enabled if the kernel detects it is
3517 running on a platform with a direct-mapped memory-side
3518 cache, and this parameter can be used to
3519 override/disable that behavior. The state of the flag
3520 can be read from sysfs at:
3521 /sys/module/page_alloc/parameters/shuffle.
3523 page_owner= [KNL] Boot-time page_owner enabling option.
3524 Storage of the information about who allocated
3525 each page is disabled in default. With this switch,
3527 on: enable the feature
3529 page_poison= [KNL] Boot-time parameter changing the state of
3530 poisoning on the buddy allocator, available with
3531 CONFIG_PAGE_POISONING=y.
3532 off: turn off poisoning (default)
3533 on: turn on poisoning
3535 panic= [KNL] Kernel behaviour on panic: delay <timeout>
3536 timeout > 0: seconds before rebooting
3537 timeout = 0: wait forever
3538 timeout < 0: reboot immediately
3541 panic_print= Bitmask for printing system info when panic happens.
3542 User can chose combination of the following bits:
3543 bit 0: print all tasks info
3544 bit 1: print system memory info
3545 bit 2: print timer info
3546 bit 3: print locks info if CONFIG_LOCKDEP is on
3547 bit 4: print ftrace buffer
3548 bit 5: print all printk messages in buffer
3550 panic_on_taint= Bitmask for conditionally calling panic() in add_taint()
3551 Format: <hex>[,nousertaint]
3552 Hexadecimal bitmask representing the set of TAINT flags
3553 that will cause the kernel to panic when add_taint() is
3554 called with any of the flags in this set.
3555 The optional switch "nousertaint" can be utilized to
3556 prevent userspace forced crashes by writing to sysctl
3557 /proc/sys/kernel/tainted any flagset matching with the
3558 bitmask set on panic_on_taint.
3559 See Documentation/admin-guide/tainted-kernels.rst for
3560 extra details on the taint flags that users can pick
3561 to compose the bitmask to assign to panic_on_taint.
3563 panic_on_warn panic() instead of WARN(). Useful to cause kdump
3566 crash_kexec_post_notifiers
3567 Run kdump after running panic-notifiers and dumping
3568 kmsg. This only for the users who doubt kdump always
3569 succeeds in any situation.
3570 Note that this also increases risks of kdump failure,
3571 because some panic notifiers can make the crashed
3572 kernel more unstable.
3574 parkbd.port= [HW] Parallel port number the keyboard adapter is
3575 connected to, default is 0.
3577 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation,
3578 0 for XT, 1 for AT (default is AT).
3581 parport= [HW,PPT] Specify parallel ports. 0 disables.
3582 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] }
3583 Use 'auto' to force the driver to use any
3584 IRQ/DMA settings detected (the default is to
3585 ignore detected IRQ/DMA settings because of
3586 possible conflicts). You can specify the base
3587 address, IRQ, and DMA settings; IRQ and DMA
3588 should be numbers, or 'auto' (for using detected
3589 settings on that particular port), or 'nofifo'
3590 (to avoid using a FIFO even if it is detected).
3591 Parallel ports are assigned in the order they
3592 are specified on the command line, starting
3595 parport_init_mode= [HW,PPT]
3596 Configure VIA parallel port to operate in
3597 a specific mode. This is necessary on Pegasos
3598 computer where firmware has no options for setting
3599 up parallel port mode and sets it to spp.
3600 Currently this function knows 686a and 8231 chips.
3601 Format: [spp|ps2|epp|ecp|ecpepp]
3604 Halt all CPUs after the first oops has been printed for
3605 the specified number of seconds. This is to be used if
3606 your oopses keep scrolling off the screen.
3611 See header of drivers/block/paride/pcd.c.
3612 See also Documentation/admin-guide/blockdev/paride.rst.
3614 pci=option[,option...] [PCI] various PCI subsystem options.
3616 Some options herein operate on a specific device
3617 or a set of devices (<pci_dev>). These are
3618 specified in one of the following formats:
3620 [<domain>:]<bus>:<dev>.<func>[/<dev>.<func>]*
3621 pci:<vendor>:<device>[:<subvendor>:<subdevice>]
3623 Note: the first format specifies a PCI
3624 bus/device/function address which may change
3625 if new hardware is inserted, if motherboard
3626 firmware changes, or due to changes caused
3627 by other kernel parameters. If the
3628 domain is left unspecified, it is
3629 taken to be zero. Optionally, a path
3630 to a device through multiple device/function
3631 addresses can be specified after the base
3632 address (this is more robust against
3633 renumbering issues). The second format
3634 selects devices using IDs from the
3635 configuration space which may match multiple
3636 devices in the system.
3638 earlydump dump PCI config space before the kernel
3640 off [X86] don't probe for the PCI bus
3641 bios [X86-32] force use of PCI BIOS, don't access
3642 the hardware directly. Use this if your machine
3643 has a non-standard PCI host bridge.
3644 nobios [X86-32] disallow use of PCI BIOS, only direct
3645 hardware access methods are allowed. Use this
3646 if you experience crashes upon bootup and you
3647 suspect they are caused by the BIOS.
3648 conf1 [X86] Force use of PCI Configuration Access
3649 Mechanism 1 (config address in IO port 0xCF8,
3650 data in IO port 0xCFC, both 32-bit).
3651 conf2 [X86] Force use of PCI Configuration Access
3652 Mechanism 2 (IO port 0xCF8 is an 8-bit port for
3653 the function, IO port 0xCFA, also 8-bit, sets
3654 bus number. The config space is then accessed
3655 through ports 0xC000-0xCFFF).
3656 See http://wiki.osdev.org/PCI for more info
3657 on the configuration access mechanisms.
3658 noaer [PCIE] If the PCIEAER kernel config parameter is
3659 enabled, this kernel boot option can be used to
3660 disable the use of PCIE advanced error reporting.
3661 nodomains [PCI] Disable support for multiple PCI
3662 root domains (aka PCI segments, in ACPI-speak).
3663 nommconf [X86] Disable use of MMCONFIG for PCI
3665 check_enable_amd_mmconf [X86] check for and enable
3666 properly configured MMIO access to PCI
3667 config space on AMD family 10h CPU
3668 nomsi [MSI] If the PCI_MSI kernel config parameter is
3669 enabled, this kernel boot option can be used to
3670 disable the use of MSI interrupts system-wide.
3671 noioapicquirk [APIC] Disable all boot interrupt quirks.
3672 Safety option to keep boot IRQs enabled. This
3673 should never be necessary.
3674 ioapicreroute [APIC] Enable rerouting of boot IRQs to the
3675 primary IO-APIC for bridges that cannot disable
3676 boot IRQs. This fixes a source of spurious IRQs
3677 when the system masks IRQs.
3678 noioapicreroute [APIC] Disable workaround that uses the
3679 boot IRQ equivalent of an IRQ that connects to
3680 a chipset where boot IRQs cannot be disabled.
3681 The opposite of ioapicreroute.
3682 biosirq [X86-32] Use PCI BIOS calls to get the interrupt
3683 routing table. These calls are known to be buggy
3684 on several machines and they hang the machine
3685 when used, but on other computers it's the only
3686 way to get the interrupt routing table. Try
3687 this option if the kernel is unable to allocate
3688 IRQs or discover secondary PCI buses on your
3690 rom [X86] Assign address space to expansion ROMs.
3691 Use with caution as certain devices share
3692 address decoders between ROMs and other
3694 norom [X86] Do not assign address space to
3695 expansion ROMs that do not already have
3696 BIOS assigned address ranges.
3697 nobar [X86] Do not assign address space to the
3698 BARs that weren't assigned by the BIOS.
3699 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be
3700 assigned automatically to PCI devices. You can
3701 make the kernel exclude IRQs of your ISA cards
3703 pirqaddr=0xAAAAA [X86] Specify the physical address
3704 of the PIRQ table (normally generated
3705 by the BIOS) if it is outside the
3706 F0000h-100000h range.
3707 lastbus=N [X86] Scan all buses thru bus #N. Can be
3708 useful if the kernel is unable to find your
3709 secondary buses and you want to tell it
3710 explicitly which ones they are.
3711 assign-busses [X86] Always assign all PCI bus
3712 numbers ourselves, overriding
3713 whatever the firmware may have done.
3714 usepirqmask [X86] Honor the possible IRQ mask stored
3715 in the BIOS $PIR table. This is needed on
3716 some systems with broken BIOSes, notably
3717 some HP Pavilion N5400 and Omnibook XE3
3718 notebooks. This will have no effect if ACPI
3719 IRQ routing is enabled.
3720 noacpi [X86] Do not use ACPI for IRQ routing
3721 or for PCI scanning.
3722 use_crs [X86] Use PCI host bridge window information
3723 from ACPI. On BIOSes from 2008 or later, this
3724 is enabled by default. If you need to use this,
3725 please report a bug.
3726 nocrs [X86] Ignore PCI host bridge windows from ACPI.
3727 If you need to use this, please report a bug.
3728 routeirq Do IRQ routing for all PCI devices.
3729 This is normally done in pci_enable_device(),
3730 so this option is a temporary workaround
3731 for broken drivers that don't call it.
3732 skip_isa_align [X86] do not align io start addr, so can
3733 handle more pci cards
3734 noearly [X86] Don't do any early type 1 scanning.
3735 This might help on some broken boards which
3736 machine check when some devices' config space
3737 is read. But various workarounds are disabled
3738 and some IOMMU drivers will not work.
3739 bfsort Sort PCI devices into breadth-first order.
3740 This sorting is done to get a device
3741 order compatible with older (<= 2.4) kernels.
3742 nobfsort Don't sort PCI devices into breadth-first order.
3743 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size)
3744 tuning and use the BIOS-configured MPS defaults.
3745 pcie_bus_safe Set every device's MPS to the largest value
3746 supported by all devices below the root complex.
3747 pcie_bus_perf Set device MPS to the largest allowable MPS
3748 based on its parent bus. Also set MRRS (Max
3749 Read Request Size) to the largest supported
3750 value (no larger than the MPS that the device
3751 or bus can support) for best performance.
3752 pcie_bus_peer2peer Set every device's MPS to 128B, which
3753 every device is guaranteed to support. This
3754 configuration allows peer-to-peer DMA between
3755 any pair of devices, possibly at the cost of
3756 reduced performance. This also guarantees
3757 that hot-added devices will work.
3758 cbiosize=nn[KMG] The fixed amount of bus space which is
3759 reserved for the CardBus bridge's IO window.
3760 The default value is 256 bytes.
3761 cbmemsize=nn[KMG] The fixed amount of bus space which is
3762 reserved for the CardBus bridge's memory
3763 window. The default value is 64 megabytes.
3766 [<order of align>@]<pci_dev>[; ...]
3767 Specifies alignment and device to reassign
3768 aligned memory resources. How to
3769 specify the device is described above.
3770 If <order of align> is not specified,
3771 PAGE_SIZE is used as alignment.
3772 A PCI-PCI bridge can be specified if resource
3773 windows need to be expanded.
3774 To specify the alignment for several
3775 instances of a device, the PCI vendor,
3776 device, subvendor, and subdevice may be
3777 specified, e.g., 12@pci:8086:9c22:103c:198f
3778 for 4096-byte alignment.
3779 ecrc= Enable/disable PCIe ECRC (transaction layer
3780 end-to-end CRC checking).
3781 bios: Use BIOS/firmware settings. This is the
3785 hpiosize=nn[KMG] The fixed amount of bus space which is
3786 reserved for hotplug bridge's IO window.
3787 Default size is 256 bytes.
3788 hpmmiosize=nn[KMG] The fixed amount of bus space which is
3789 reserved for hotplug bridge's MMIO window.
3790 Default size is 2 megabytes.
3791 hpmmioprefsize=nn[KMG] The fixed amount of bus space which is
3792 reserved for hotplug bridge's MMIO_PREF window.
3793 Default size is 2 megabytes.
3794 hpmemsize=nn[KMG] The fixed amount of bus space which is
3795 reserved for hotplug bridge's MMIO and
3797 Default size is 2 megabytes.
3798 hpbussize=nn The minimum amount of additional bus numbers
3799 reserved for buses below a hotplug bridge.
3801 realloc= Enable/disable reallocating PCI bridge resources
3802 if allocations done by BIOS are too small to
3803 accommodate resources required by all child
3805 off: Turn realloc off
3807 realloc same as realloc=on
3808 noari do not use PCIe ARI.
3809 noats [PCIE, Intel-IOMMU, AMD-IOMMU]
3810 do not use PCIe ATS (and IOMMU device IOTLB).
3811 pcie_scan_all Scan all possible PCIe devices. Otherwise we
3812 only look for one device below a PCIe downstream
3814 big_root_window Try to add a big 64bit memory window to the PCIe
3815 root complex on AMD CPUs. Some GFX hardware
3816 can resize a BAR to allow access to all VRAM.
3817 Adding the window is slightly risky (it may
3818 conflict with unreported devices), so this
3820 disable_acs_redir=<pci_dev>[; ...]
3821 Specify one or more PCI devices (in the format
3822 specified above) separated by semicolons.
3823 Each device specified will have the PCI ACS
3824 redirect capabilities forced off which will
3825 allow P2P traffic between devices through
3826 bridges without forcing it upstream. Note:
3827 this removes isolation between devices and
3828 may put more devices in an IOMMU group.
3829 force_floating [S390] Force usage of floating interrupts.
3830 nomio [S390] Do not use MIO instructions.
3831 norid [S390] ignore the RID field and force use of
3832 one PCI domain per PCI function
3834 pcie_aspm= [PCIE] Forcibly enable or disable PCIe Active State Power
3837 force Enable ASPM even on devices that claim not to support it.
3838 WARNING: Forcing ASPM on may cause system lockups.
3840 pcie_ports= [PCIE] PCIe port services handling:
3841 native Use native PCIe services (PME, AER, DPC, PCIe hotplug)
3842 even if the platform doesn't give the OS permission to
3843 use them. This may cause conflicts if the platform
3844 also tries to use these services.
3845 dpc-native Use native PCIe service for DPC only. May
3846 cause conflicts if firmware uses AER or DPC.
3847 compat Disable native PCIe services (PME, AER, DPC, PCIe
3850 pcie_port_pm= [PCIE] PCIe port power management handling:
3851 off Disable power management of all PCIe ports
3852 force Forcibly enable power management of all PCIe ports
3854 pcie_pme= [PCIE,PM] Native PCIe PME signaling options:
3855 nomsi Do not use MSI for native PCIe PME signaling (this makes
3856 all PCIe root ports use INTx for all services).
3858 pcmv= [HW,PCMCIA] BadgePAD 4
3862 Keep all power-domains already enabled by bootloader on,
3863 even if no driver has claimed them. This is useful
3864 for debug and development, but should not be
3865 needed on a platform with proper driver support.
3868 See Documentation/admin-guide/blockdev/paride.rst.
3870 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at
3873 See arch/parisc/kernel/pdc_chassis.c
3875 percpu_alloc= Select which percpu first chunk allocator to use.
3876 Currently supported values are "embed" and "page".
3877 Archs may support subset or none of the selections.
3878 See comments in mm/percpu.c for details on each
3879 allocator. This parameter is primarily for debugging
3880 and performance comparison.
3883 See Documentation/admin-guide/blockdev/paride.rst.
3886 See Documentation/admin-guide/blockdev/paride.rst.
3888 pirq= [SMP,APIC] Manual mp-table setup
3889 See Documentation/x86/i386/IO-APIC.rst.
3891 plip= [PPT,NET] Parallel port network link
3892 Format: { parport<nr> | timid | 0 }
3893 See also Documentation/admin-guide/parport.rst.
3895 pmtmr= [X86] Manual setup of pmtmr I/O Port.
3896 Override pmtimer IOPort with a hex value.
3899 pm_debug_messages [SUSPEND,KNL]
3900 Enable suspend/resume debug messages during boot up.
3903 Enable PNP debug messages (depends on the
3904 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time
3905 via /sys/module/pnp/parameters/debug. We always show
3906 current resource usage; turning this on also shows
3907 possible settings and some assignment information.
3913 { on | off | curr | res | no-curr | no-res }
3916 [ISAPNP] Exclude IRQs for the autoconfiguration
3919 [ISAPNP] Exclude DMAs for the autoconfiguration
3921 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration
3922 Ranges are in pairs (I/O port base and size).
3925 [ISAPNP] Exclude memory regions for the
3927 Ranges are in pairs (memory base and size).
3929 ports= [IP_VS_FTP] IPVS ftp helper module
3931 Up to 8 (IP_VS_APP_MAX_PORTS) ports
3933 Format: <port>,<port>....
3935 powersave=off [PPC] This option disables power saving features.
3936 It specifically disables cpuidle and sets the
3937 platform machine description specific power_save
3938 function to NULL. On Idle the CPU just reduces
3941 ppc_strict_facility_enable
3942 [PPC] This option catches any kernel floating point,
3943 Altivec, VSX and SPE outside of regions specifically
3944 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()).
3945 There is some performance impact when enabling this.
3949 Disable Hardware Transactional Memory
3952 Select preemption mode if you have CONFIG_PREEMPT_DYNAMIC
3953 none - Limited to cond_resched() calls
3954 voluntary - Limited to cond_resched() and might_sleep() calls
3955 full - Any section that isn't explicitly preempt disabled
3956 can be preempted anytime.
3958 print-fatal-signals=
3959 [KNL] debug: print fatal signals
3961 If enabled, warn about various signal handling
3962 related application anomalies: too many signals,
3963 too many POSIX.1 timers, fatal signals causing a
3966 If you hit the warning due to signal overflow,
3967 you might want to try "ulimit -i unlimited".
3971 printk.always_kmsg_dump=
3972 Trigger kmsg_dump for cases other than kernel oops or
3974 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
3977 printk.devkmsg={on,off,ratelimit}
3978 Control writing to /dev/kmsg.
3979 on - unlimited logging to /dev/kmsg from userspace
3980 off - logging to /dev/kmsg disabled
3981 ratelimit - ratelimit the logging
3984 printk.time= Show timing data prefixed to each printk message line
3985 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
3987 processor.max_cstate= [HW,ACPI]
3988 Limit processor to maximum C-state
3989 max_cstate=9 overrides any DMI blacklist limit.
3991 processor.nocst [HW,ACPI]
3992 Ignore the _CST method to determine C-states,
3993 instead using the legacy FADT method
3995 profile= [KNL] Enable kernel profiling via /proc/profile
3996 Format: [<profiletype>,]<number>
3997 Param: <profiletype>: "schedule", "sleep", or "kvm"
3998 [defaults to kernel profiling]
3999 Param: "schedule" - profile schedule points.
4000 Param: "sleep" - profile D-state sleeping (millisecs).
4001 Requires CONFIG_SCHEDSTATS
4002 Param: "kvm" - profile VM exits.
4003 Param: <number> - step/bucket size as a power of 2 for
4004 statistical time based profiling.
4006 prompt_ramdisk= [RAM] [Deprecated]
4008 prot_virt= [S390] enable hosting protected virtual machines
4009 isolated from the hypervisor (if hardware supports
4013 psi= [KNL] Enable or disable pressure stall information
4017 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to
4018 probe for; one of (bare|imps|exps|lifebook|any).
4019 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports
4021 psmouse.resetafter= [HW,MOUSE]
4022 Try to reset the device after so many bad packets
4025 [HW,MOUSE] Set desired mouse resolution, in dpi.
4026 psmouse.smartscroll=
4027 [HW,MOUSE] Controls Logitech smartscroll autorepeat.
4028 0 = disabled, 1 = enabled (default).
4030 pstore.backend= Specify the name of the pstore backend to use
4033 See Documentation/admin-guide/blockdev/paride.rst.
4035 pti= [X86-64] Control Page Table Isolation of user and
4036 kernel address spaces. Disabling this feature
4037 removes hardening, but improves performance of
4038 system calls and interrupts.
4040 on - unconditionally enable
4041 off - unconditionally disable
4042 auto - kernel detects whether your CPU model is
4043 vulnerable to issues that PTI mitigates
4045 Not specifying this option is equivalent to pti=auto.
4048 Equivalent to pti=off
4051 [KNL] Number of legacy pty's. Overwrites compiled-in
4054 quiet [KNL] Disable most log messages
4059 See Documentation/admin-guide/md.rst.
4061 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes
4062 See Documentation/admin-guide/blockdev/ramdisk.rst.
4064 ramdisk_start= [RAM] RAM disk image start address
4066 random.trust_cpu={on,off}
4067 [KNL] Enable or disable trusting the use of the
4068 CPU's random number generator (if available) to
4069 fully seed the kernel's CRNG. Default is controlled
4070 by CONFIG_RANDOM_TRUST_CPU.
4072 randomize_kstack_offset=
4073 [KNL] Enable or disable kernel stack offset
4074 randomization, which provides roughly 5 bits of
4075 entropy, frustrating memory corruption attacks
4076 that depend on stack address determinism or
4077 cross-syscall address exposures. This is only
4078 available on architectures that have defined
4079 CONFIG_HAVE_ARCH_RANDOMIZE_KSTACK_OFFSET.
4080 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
4081 Default is CONFIG_RANDOMIZE_KSTACK_OFFSET_DEFAULT.
4083 ras=option[,option,...] [KNL] RAS-specific options
4086 Disable the Correctable Errors Collector,
4087 see CONFIG_RAS_CEC help text.
4090 The argument is a cpu list, as described above.
4092 In kernels built with CONFIG_RCU_NOCB_CPU=y, set
4093 the specified list of CPUs to be no-callback CPUs.
4094 Invocation of these CPUs' RCU callbacks will be
4095 offloaded to "rcuox/N" kthreads created for that
4096 purpose, where "x" is "p" for RCU-preempt, and
4097 "s" for RCU-sched, and "N" is the CPU number.
4098 This reduces OS jitter on the offloaded CPUs,
4099 which can be useful for HPC and real-time
4100 workloads. It can also improve energy efficiency
4101 for asymmetric multiprocessors.
4104 Rather than requiring that offloaded CPUs
4105 (specified by rcu_nocbs= above) explicitly
4106 awaken the corresponding "rcuoN" kthreads,
4107 make these kthreads poll for callbacks.
4108 This improves the real-time response for the
4109 offloaded CPUs by relieving them of the need to
4110 wake up the corresponding kthread, but degrades
4111 energy efficiency by requiring that the kthreads
4112 periodically wake up to do the polling.
4114 rcutree.blimit= [KNL]
4115 Set maximum number of finished RCU callbacks to
4116 process in one batch.
4118 rcutree.dump_tree= [KNL]
4119 Dump the structure of the rcu_node combining tree
4120 out at early boot. This is used for diagnostic
4121 purposes, to verify correct tree setup.
4123 rcutree.gp_cleanup_delay= [KNL]
4124 Set the number of jiffies to delay each step of
4125 RCU grace-period cleanup.
4127 rcutree.gp_init_delay= [KNL]
4128 Set the number of jiffies to delay each step of
4129 RCU grace-period initialization.
4131 rcutree.gp_preinit_delay= [KNL]
4132 Set the number of jiffies to delay each step of
4133 RCU grace-period pre-initialization, that is,
4134 the propagation of recent CPU-hotplug changes up
4135 the rcu_node combining tree.
4137 rcutree.use_softirq= [KNL]
4138 If set to zero, move all RCU_SOFTIRQ processing to
4139 per-CPU rcuc kthreads. Defaults to a non-zero
4140 value, meaning that RCU_SOFTIRQ is used by default.
4141 Specify rcutree.use_softirq=0 to use rcuc kthreads.
4143 But note that CONFIG_PREEMPT_RT=y kernels disable
4144 this kernel boot parameter, forcibly setting it
4147 rcutree.rcu_fanout_exact= [KNL]
4148 Disable autobalancing of the rcu_node combining
4149 tree. This is used by rcutorture, and might
4150 possibly be useful for architectures having high
4151 cache-to-cache transfer latencies.
4153 rcutree.rcu_fanout_leaf= [KNL]
4154 Change the number of CPUs assigned to each
4155 leaf rcu_node structure. Useful for very
4156 large systems, which will choose the value 64,
4157 and for NUMA systems with large remote-access
4158 latencies, which will choose a value aligned
4159 with the appropriate hardware boundaries.
4161 rcutree.rcu_min_cached_objs= [KNL]
4162 Minimum number of objects which are cached and
4163 maintained per one CPU. Object size is equal
4164 to PAGE_SIZE. The cache allows to reduce the
4165 pressure to page allocator, also it makes the
4166 whole algorithm to behave better in low memory
4169 rcutree.jiffies_till_first_fqs= [KNL]
4170 Set delay from grace-period initialization to
4171 first attempt to force quiescent states.
4172 Units are jiffies, minimum value is zero,
4173 and maximum value is HZ.
4175 rcutree.jiffies_till_next_fqs= [KNL]
4176 Set delay between subsequent attempts to force
4177 quiescent states. Units are jiffies, minimum
4178 value is one, and maximum value is HZ.
4180 rcutree.jiffies_till_sched_qs= [KNL]
4181 Set required age in jiffies for a
4182 given grace period before RCU starts
4183 soliciting quiescent-state help from
4184 rcu_note_context_switch() and cond_resched().
4185 If not specified, the kernel will calculate
4186 a value based on the most recent settings
4187 of rcutree.jiffies_till_first_fqs
4188 and rcutree.jiffies_till_next_fqs.
4189 This calculated value may be viewed in
4190 rcutree.jiffies_to_sched_qs. Any attempt to set
4191 rcutree.jiffies_to_sched_qs will be cheerfully
4194 rcutree.kthread_prio= [KNL,BOOT]
4195 Set the SCHED_FIFO priority of the RCU per-CPU
4196 kthreads (rcuc/N). This value is also used for
4197 the priority of the RCU boost threads (rcub/N)
4198 and for the RCU grace-period kthreads (rcu_bh,
4199 rcu_preempt, and rcu_sched). If RCU_BOOST is
4200 set, valid values are 1-99 and the default is 1
4201 (the least-favored priority). Otherwise, when
4202 RCU_BOOST is not set, valid values are 0-99 and
4203 the default is zero (non-realtime operation).
4205 rcutree.rcu_nocb_gp_stride= [KNL]
4206 Set the number of NOCB callback kthreads in
4207 each group, which defaults to the square root
4208 of the number of CPUs. Larger numbers reduce
4209 the wakeup overhead on the global grace-period
4210 kthread, but increases that same overhead on
4211 each group's NOCB grace-period kthread.
4213 rcutree.qhimark= [KNL]
4214 Set threshold of queued RCU callbacks beyond which
4215 batch limiting is disabled.
4217 rcutree.qlowmark= [KNL]
4218 Set threshold of queued RCU callbacks below which
4219 batch limiting is re-enabled.
4221 rcutree.qovld= [KNL]
4222 Set threshold of queued RCU callbacks beyond which
4223 RCU's force-quiescent-state scan will aggressively
4224 enlist help from cond_resched() and sched IPIs to
4225 help CPUs more quickly reach quiescent states.
4226 Set to less than zero to make this be set based
4227 on rcutree.qhimark at boot time and to zero to
4228 disable more aggressive help enlistment.
4230 rcutree.rcu_idle_gp_delay= [KNL]
4231 Set wakeup interval for idle CPUs that have
4232 RCU callbacks (RCU_FAST_NO_HZ=y).
4234 rcutree.rcu_kick_kthreads= [KNL]
4235 Cause the grace-period kthread to get an extra
4236 wake_up() if it sleeps three times longer than
4237 it should at force-quiescent-state time.
4238 This wake_up() will be accompanied by a
4239 WARN_ONCE() splat and an ftrace_dump().
4241 rcutree.rcu_unlock_delay= [KNL]
4242 In CONFIG_RCU_STRICT_GRACE_PERIOD=y kernels,
4243 this specifies an rcu_read_unlock()-time delay
4244 in microseconds. This defaults to zero.
4245 Larger delays increase the probability of
4246 catching RCU pointer leaks, that is, buggy use
4247 of RCU-protected pointers after the relevant
4248 rcu_read_unlock() has completed.
4250 rcutree.sysrq_rcu= [KNL]
4251 Commandeer a sysrq key to dump out Tree RCU's
4252 rcu_node tree with an eye towards determining
4253 why a new grace period has not yet started.
4255 rcuscale.gp_async= [KNL]
4256 Measure performance of asynchronous
4257 grace-period primitives such as call_rcu().
4259 rcuscale.gp_async_max= [KNL]
4260 Specify the maximum number of outstanding
4261 callbacks per writer thread. When a writer
4262 thread exceeds this limit, it invokes the
4263 corresponding flavor of rcu_barrier() to allow
4264 previously posted callbacks to drain.
4266 rcuscale.gp_exp= [KNL]
4267 Measure performance of expedited synchronous
4268 grace-period primitives.
4270 rcuscale.holdoff= [KNL]
4271 Set test-start holdoff period. The purpose of
4272 this parameter is to delay the start of the
4273 test until boot completes in order to avoid
4276 rcuscale.kfree_rcu_test= [KNL]
4277 Set to measure performance of kfree_rcu() flooding.
4279 rcuscale.kfree_rcu_test_double= [KNL]
4280 Test the double-argument variant of kfree_rcu().
4281 If this parameter has the same value as
4282 rcuscale.kfree_rcu_test_single, both the single-
4283 and double-argument variants are tested.
4285 rcuscale.kfree_rcu_test_single= [KNL]
4286 Test the single-argument variant of kfree_rcu().
4287 If this parameter has the same value as
4288 rcuscale.kfree_rcu_test_double, both the single-
4289 and double-argument variants are tested.
4291 rcuscale.kfree_nthreads= [KNL]
4292 The number of threads running loops of kfree_rcu().
4294 rcuscale.kfree_alloc_num= [KNL]
4295 Number of allocations and frees done in an iteration.
4297 rcuscale.kfree_loops= [KNL]
4298 Number of loops doing rcuscale.kfree_alloc_num number
4299 of allocations and frees.
4301 rcuscale.nreaders= [KNL]
4302 Set number of RCU readers. The value -1 selects
4303 N, where N is the number of CPUs. A value
4304 "n" less than -1 selects N-n+1, where N is again
4305 the number of CPUs. For example, -2 selects N
4306 (the number of CPUs), -3 selects N+1, and so on.
4307 A value of "n" less than or equal to -N selects
4310 rcuscale.nwriters= [KNL]
4311 Set number of RCU writers. The values operate
4312 the same as for rcuscale.nreaders.
4313 N, where N is the number of CPUs
4315 rcuscale.perf_type= [KNL]
4316 Specify the RCU implementation to test.
4318 rcuscale.shutdown= [KNL]
4319 Shut the system down after performance tests
4320 complete. This is useful for hands-off automated
4323 rcuscale.verbose= [KNL]
4324 Enable additional printk() statements.
4326 rcuscale.writer_holdoff= [KNL]
4327 Write-side holdoff between grace periods,
4328 in microseconds. The default of zero says
4331 rcutorture.fqs_duration= [KNL]
4332 Set duration of force_quiescent_state bursts
4335 rcutorture.fqs_holdoff= [KNL]
4336 Set holdoff time within force_quiescent_state bursts
4339 rcutorture.fqs_stutter= [KNL]
4340 Set wait time between force_quiescent_state bursts
4343 rcutorture.fwd_progress= [KNL]
4344 Enable RCU grace-period forward-progress testing
4345 for the types of RCU supporting this notion.
4347 rcutorture.fwd_progress_div= [KNL]
4348 Specify the fraction of a CPU-stall-warning
4349 period to do tight-loop forward-progress testing.
4351 rcutorture.fwd_progress_holdoff= [KNL]
4352 Number of seconds to wait between successive
4353 forward-progress tests.
4355 rcutorture.fwd_progress_need_resched= [KNL]
4356 Enclose cond_resched() calls within checks for
4357 need_resched() during tight-loop forward-progress
4360 rcutorture.gp_cond= [KNL]
4361 Use conditional/asynchronous update-side
4362 primitives, if available.
4364 rcutorture.gp_exp= [KNL]
4365 Use expedited update-side primitives, if available.
4367 rcutorture.gp_normal= [KNL]
4368 Use normal (non-expedited) asynchronous
4369 update-side primitives, if available.
4371 rcutorture.gp_sync= [KNL]
4372 Use normal (non-expedited) synchronous
4373 update-side primitives, if available. If all
4374 of rcutorture.gp_cond=, rcutorture.gp_exp=,
4375 rcutorture.gp_normal=, and rcutorture.gp_sync=
4376 are zero, rcutorture acts as if is interpreted
4377 they are all non-zero.
4379 rcutorture.irqreader= [KNL]
4380 Run RCU readers from irq handlers, or, more
4381 accurately, from a timer handler. Not all RCU
4382 flavors take kindly to this sort of thing.
4384 rcutorture.leakpointer= [KNL]
4385 Leak an RCU-protected pointer out of the reader.
4386 This can of course result in splats, and is
4387 intended to test the ability of things like
4388 CONFIG_RCU_STRICT_GRACE_PERIOD=y to detect
4391 rcutorture.n_barrier_cbs= [KNL]
4392 Set callbacks/threads for rcu_barrier() testing.
4394 rcutorture.nfakewriters= [KNL]
4395 Set number of concurrent RCU writers. These just
4396 stress RCU, they don't participate in the actual
4397 test, hence the "fake".
4399 rcutorture.nocbs_nthreads= [KNL]
4400 Set number of RCU callback-offload togglers.
4401 Zero (the default) disables toggling.
4403 rcutorture.nocbs_toggle= [KNL]
4404 Set the delay in milliseconds between successive
4405 callback-offload toggling attempts.
4407 rcutorture.nreaders= [KNL]
4408 Set number of RCU readers. The value -1 selects
4409 N-1, where N is the number of CPUs. A value
4410 "n" less than -1 selects N-n-2, where N is again
4411 the number of CPUs. For example, -2 selects N
4412 (the number of CPUs), -3 selects N+1, and so on.
4414 rcutorture.object_debug= [KNL]
4415 Enable debug-object double-call_rcu() testing.
4417 rcutorture.onoff_holdoff= [KNL]
4418 Set time (s) after boot for CPU-hotplug testing.
4420 rcutorture.onoff_interval= [KNL]
4421 Set time (jiffies) between CPU-hotplug operations,
4422 or zero to disable CPU-hotplug testing.
4424 rcutorture.read_exit= [KNL]
4425 Set the number of read-then-exit kthreads used
4426 to test the interaction of RCU updaters and
4427 task-exit processing.
4429 rcutorture.read_exit_burst= [KNL]
4430 The number of times in a given read-then-exit
4431 episode that a set of read-then-exit kthreads
4434 rcutorture.read_exit_delay= [KNL]
4435 The delay, in seconds, between successive
4436 read-then-exit testing episodes.
4438 rcutorture.shuffle_interval= [KNL]
4439 Set task-shuffle interval (s). Shuffling tasks
4440 allows some CPUs to go into dyntick-idle mode
4441 during the rcutorture test.
4443 rcutorture.shutdown_secs= [KNL]
4444 Set time (s) after boot system shutdown. This
4445 is useful for hands-off automated testing.
4447 rcutorture.stall_cpu= [KNL]
4448 Duration of CPU stall (s) to test RCU CPU stall
4449 warnings, zero to disable.
4451 rcutorture.stall_cpu_block= [KNL]
4452 Sleep while stalling if set. This will result
4453 in warnings from preemptible RCU in addition
4454 to any other stall-related activity.
4456 rcutorture.stall_cpu_holdoff= [KNL]
4457 Time to wait (s) after boot before inducing stall.
4459 rcutorture.stall_cpu_irqsoff= [KNL]
4460 Disable interrupts while stalling if set.
4462 rcutorture.stall_gp_kthread= [KNL]
4463 Duration (s) of forced sleep within RCU
4464 grace-period kthread to test RCU CPU stall
4465 warnings, zero to disable. If both stall_cpu
4466 and stall_gp_kthread are specified, the
4467 kthread is starved first, then the CPU.
4469 rcutorture.stat_interval= [KNL]
4470 Time (s) between statistics printk()s.
4472 rcutorture.stutter= [KNL]
4473 Time (s) to stutter testing, for example, specifying
4474 five seconds causes the test to run for five seconds,
4475 wait for five seconds, and so on. This tests RCU's
4476 ability to transition abruptly to and from idle.
4478 rcutorture.test_boost= [KNL]
4479 Test RCU priority boosting? 0=no, 1=maybe, 2=yes.
4480 "Maybe" means test if the RCU implementation
4481 under test support RCU priority boosting.
4483 rcutorture.test_boost_duration= [KNL]
4484 Duration (s) of each individual boost test.
4486 rcutorture.test_boost_interval= [KNL]
4487 Interval (s) between each boost test.
4489 rcutorture.test_no_idle_hz= [KNL]
4490 Test RCU's dyntick-idle handling. See also the
4491 rcutorture.shuffle_interval parameter.
4493 rcutorture.torture_type= [KNL]
4494 Specify the RCU implementation to test.
4496 rcutorture.verbose= [KNL]
4497 Enable additional printk() statements.
4499 rcupdate.rcu_cpu_stall_ftrace_dump= [KNL]
4500 Dump ftrace buffer after reporting RCU CPU
4503 rcupdate.rcu_cpu_stall_suppress= [KNL]
4504 Suppress RCU CPU stall warning messages.
4506 rcupdate.rcu_cpu_stall_suppress_at_boot= [KNL]
4507 Suppress RCU CPU stall warning messages and
4508 rcutorture writer stall warnings that occur
4509 during early boot, that is, during the time
4510 before the init task is spawned.
4512 rcupdate.rcu_cpu_stall_timeout= [KNL]
4513 Set timeout for RCU CPU stall warning messages.
4515 rcupdate.rcu_expedited= [KNL]
4516 Use expedited grace-period primitives, for
4517 example, synchronize_rcu_expedited() instead
4518 of synchronize_rcu(). This reduces latency,
4519 but can increase CPU utilization, degrade
4520 real-time latency, and degrade energy efficiency.
4521 No effect on CONFIG_TINY_RCU kernels.
4523 rcupdate.rcu_normal= [KNL]
4524 Use only normal grace-period primitives,
4525 for example, synchronize_rcu() instead of
4526 synchronize_rcu_expedited(). This improves
4527 real-time latency, CPU utilization, and
4528 energy efficiency, but can expose users to
4529 increased grace-period latency. This parameter
4530 overrides rcupdate.rcu_expedited. No effect on
4531 CONFIG_TINY_RCU kernels.
4533 rcupdate.rcu_normal_after_boot= [KNL]
4534 Once boot has completed (that is, after
4535 rcu_end_inkernel_boot() has been invoked), use
4536 only normal grace-period primitives. No effect
4537 on CONFIG_TINY_RCU kernels.
4539 But note that CONFIG_PREEMPT_RT=y kernels enables
4540 this kernel boot parameter, forcibly setting
4541 it to the value one, that is, converting any
4542 post-boot attempt at an expedited RCU grace
4543 period to instead use normal non-expedited
4544 grace-period processing.
4546 rcupdate.rcu_task_ipi_delay= [KNL]
4547 Set time in jiffies during which RCU tasks will
4548 avoid sending IPIs, starting with the beginning
4549 of a given grace period. Setting a large
4550 number avoids disturbing real-time workloads,
4551 but lengthens grace periods.
4553 rcupdate.rcu_task_stall_timeout= [KNL]
4554 Set timeout in jiffies for RCU task stall warning
4555 messages. Disable with a value less than or equal
4558 rcupdate.rcu_self_test= [KNL]
4559 Run the RCU early boot self tests
4563 Run specified binary instead of /init from the ramdisk,
4564 used for early userspace startup. See initrd.
4567 force - Override the decision by the kernel to hide the
4568 advertisement of RDRAND support (this affects
4569 certain AMD processors because of buggy BIOS
4570 support, specifically around the suspend/resume
4574 Turn on/off individual RDT features. List is:
4575 cmt, mbmtotal, mbmlocal, l3cat, l3cdp, l2cat, l2cdp,
4577 E.g. to turn on cmt and turn off mba use:
4581 Format (x86 or x86_64):
4582 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] \
4584 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \
4586 Where reboot_mode is one of warm (soft) or cold (hard) or gpio
4587 (prefix with 'panic_' to set mode for panic
4589 reboot_type is one of bios, acpi, kbd, triple, efi, or pci,
4590 reboot_force is either force or not specified,
4591 reboot_cpu is s[mp]#### with #### being the processor
4592 to be used for rebooting.
4594 refscale.holdoff= [KNL]
4595 Set test-start holdoff period. The purpose of
4596 this parameter is to delay the start of the
4597 test until boot completes in order to avoid
4600 refscale.loops= [KNL]
4601 Set the number of loops over the synchronization
4602 primitive under test. Increasing this number
4603 reduces noise due to loop start/end overhead,
4604 but the default has already reduced the per-pass
4605 noise to a handful of picoseconds on ca. 2020
4608 refscale.nreaders= [KNL]
4609 Set number of readers. The default value of -1
4610 selects N, where N is roughly 75% of the number
4611 of CPUs. A value of zero is an interesting choice.
4613 refscale.nruns= [KNL]
4614 Set number of runs, each of which is dumped onto
4617 refscale.readdelay= [KNL]
4618 Set the read-side critical-section duration,
4619 measured in microseconds.
4621 refscale.scale_type= [KNL]
4622 Specify the read-protection implementation to test.
4624 refscale.shutdown= [KNL]
4625 Shut down the system at the end of the performance
4626 test. This defaults to 1 (shut it down) when
4627 refscale is built into the kernel and to 0 (leave
4628 it running) when refscale is built as a module.
4630 refscale.verbose= [KNL]
4631 Enable additional printk() statements.
4633 refscale.verbose_batched= [KNL]
4634 Batch the additional printk() statements. If zero
4635 (the default) or negative, print everything. Otherwise,
4636 print every Nth verbose statement, where N is the value
4640 [KNL, SMP] Set scheduler's default relax_domain_level.
4641 See Documentation/admin-guide/cgroup-v1/cpusets.rst.
4643 reserve= [KNL,BUGS] Force kernel to ignore I/O ports or memory
4644 Format: <base1>,<size1>[,<base2>,<size2>,...]
4645 Reserve I/O ports or memory so the kernel won't use
4646 them. If <base> is less than 0x10000, the region
4647 is assumed to be I/O ports; otherwise it is memory.
4649 reservetop= [X86-32]
4651 Reserves a hole at the top of the kernel virtual
4656 Set the amount of memory to reserve for BIOS at
4657 the bottom of the address space.
4659 reset_devices [KNL] Force drivers to reset the underlying device
4660 during initialization.
4663 Specify the partition device for software suspend
4665 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>}
4667 resume_offset= [SWSUSP]
4668 Specify the offset from the beginning of the partition
4669 given by "resume=" at which the swap header is located,
4670 in <PAGE_SIZE> units (needed only for swap files).
4671 See Documentation/power/swsusp-and-swap-files.rst
4673 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to
4674 read the resume files
4676 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up.
4677 Useful for devices that are detected asynchronously
4678 (e.g. USB and MMC devices).
4680 hibernate= [HIBERNATION]
4681 noresume Don't check if there's a hibernation image
4682 present during boot.
4683 nocompress Don't compress/decompress hibernation images.
4684 no Disable hibernation and resume.
4685 protect_image Turn on image protection during restoration
4686 (that will set all pages holding image data
4687 during restoration read-only).
4689 retain_initrd [RAM] Keep initrd memory after extraction
4691 rfkill.default_state=
4692 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm,
4693 etc. communication is blocked by default.
4696 rfkill.master_switch_mode=
4697 0 The "airplane mode" button does nothing.
4698 1 The "airplane mode" button toggles between everything
4699 blocked and the previous configuration.
4700 2 The "airplane mode" button toggles between everything
4701 blocked and everything unblocked.
4703 rhash_entries= [KNL,NET]
4704 Set number of hash buckets for route cache
4707 [KNL] Disable ring 3 MONITOR/MWAIT feature on supported
4710 ro [KNL] Mount root device read-only on boot
4713 on Mark read-only kernel memory as read-only (default).
4714 off Leave read-only kernel memory writable for debugging.
4717 Enable the uart passthrough on the designated usb port
4718 on Rockchip SoCs. When active, the signals of the
4719 debug-uart get routed to the D+ and D- pins of the usb
4720 port and the regular usb controller gets disabled.
4722 root= [KNL] Root filesystem
4723 See name_to_dev_t comment in init/do_mounts.c.
4725 rootdelay= [KNL] Delay (in seconds) to pause before attempting to
4726 mount the root filesystem
4728 rootflags= [KNL] Set root filesystem mount option string
4730 rootfstype= [KNL] Set root filesystem type
4732 rootwait [KNL] Wait (indefinitely) for root device to show up.
4733 Useful for devices that are detected asynchronously
4734 (e.g. USB and MMC devices).
4736 rproc_mem=nn[KMG][@address]
4737 [KNL,ARM,CMA] Remoteproc physical memory block.
4738 Memory area to be used by remote processor image,
4741 rw [KNL] Mount root device read-write on boot
4743 S [KNL] Run init in single mode
4745 s390_iommu= [HW,S390]
4746 Set s390 IOTLB flushing mode
4748 With strict flushing every unmap operation will result in
4749 an IOTLB flush. Default is lazy flushing before reuse,
4753 See drivers/net/irda/sa1100_ir.c.
4755 sbni= [NET] Granch SBNI12 leased line adapter
4757 sched_debug [KNL] Enables verbose scheduler debug messages.
4759 schedstats= [KNL,X86] Enable or disable scheduled statistics.
4760 Allowed values are enable and disable. This feature
4761 incurs a small amount of overhead in the scheduler
4762 but is useful for debugging and performance tuning.
4764 sched_thermal_decay_shift=
4765 [KNL, SMP] Set a decay shift for scheduler thermal
4766 pressure signal. Thermal pressure signal follows the
4767 default decay period of other scheduler pelt
4768 signals(usually 32 ms but configurable). Setting
4769 sched_thermal_decay_shift will left shift the decay
4770 period for the thermal pressure signal by the shift
4772 i.e. with the default pelt decay period of 32 ms
4773 sched_thermal_decay_shift thermal pressure decay pr
4777 Format: integer between 0 and 10
4780 scftorture.holdoff= [KNL]
4781 Number of seconds to hold off before starting
4782 test. Defaults to zero for module insertion and
4783 to 10 seconds for built-in smp_call_function()
4786 scftorture.longwait= [KNL]
4787 Request ridiculously long waits randomly selected
4788 up to the chosen limit in seconds. Zero (the
4789 default) disables this feature. Please note
4790 that requesting even small non-zero numbers of
4791 seconds can result in RCU CPU stall warnings,
4792 softlockup complaints, and so on.
4794 scftorture.nthreads= [KNL]
4795 Number of kthreads to spawn to invoke the
4796 smp_call_function() family of functions.
4797 The default of -1 specifies a number of kthreads
4798 equal to the number of CPUs.
4800 scftorture.onoff_holdoff= [KNL]
4801 Number seconds to wait after the start of the
4802 test before initiating CPU-hotplug operations.
4804 scftorture.onoff_interval= [KNL]
4805 Number seconds to wait between successive
4806 CPU-hotplug operations. Specifying zero (which
4807 is the default) disables CPU-hotplug operations.
4809 scftorture.shutdown_secs= [KNL]
4810 The number of seconds following the start of the
4811 test after which to shut down the system. The
4812 default of zero avoids shutting down the system.
4813 Non-zero values are useful for automated tests.
4815 scftorture.stat_interval= [KNL]
4816 The number of seconds between outputting the
4817 current test statistics to the console. A value
4818 of zero disables statistics output.
4820 scftorture.stutter_cpus= [KNL]
4821 The number of jiffies to wait between each change
4822 to the set of CPUs under test.
4824 scftorture.use_cpus_read_lock= [KNL]
4825 Use use_cpus_read_lock() instead of the default
4826 preempt_disable() to disable CPU hotplug
4827 while invoking one of the smp_call_function*()
4830 scftorture.verbose= [KNL]
4831 Enable additional printk() statements.
4833 scftorture.weight_single= [KNL]
4834 The probability weighting to use for the
4835 smp_call_function_single() function with a zero
4836 "wait" parameter. A value of -1 selects the
4837 default if all other weights are -1. However,
4838 if at least one weight has some other value, a
4839 value of -1 will instead select a weight of zero.
4841 scftorture.weight_single_wait= [KNL]
4842 The probability weighting to use for the
4843 smp_call_function_single() function with a
4844 non-zero "wait" parameter. See weight_single.
4846 scftorture.weight_many= [KNL]
4847 The probability weighting to use for the
4848 smp_call_function_many() function with a zero
4849 "wait" parameter. See weight_single.
4850 Note well that setting a high probability for
4851 this weighting can place serious IPI load
4854 scftorture.weight_many_wait= [KNL]
4855 The probability weighting to use for the
4856 smp_call_function_many() function with a
4857 non-zero "wait" parameter. See weight_single
4860 scftorture.weight_all= [KNL]
4861 The probability weighting to use for the
4862 smp_call_function_all() function with a zero
4863 "wait" parameter. See weight_single and
4866 scftorture.weight_all_wait= [KNL]
4867 The probability weighting to use for the
4868 smp_call_function_all() function with a
4869 non-zero "wait" parameter. See weight_single
4872 skew_tick= [KNL] Offset the periodic timer tick per cpu to mitigate
4873 xtime_lock contention on larger systems, and/or RCU lock
4874 contention on all systems with CONFIG_MAXSMP set.
4875 Format: { "0" | "1" }
4876 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1"
4878 Note: increases power consumption, thus should only be
4879 enabled if running jitter sensitive (HPC/RT) workloads.
4881 security= [SECURITY] Choose a legacy "major" security module to
4882 enable at boot. This has been deprecated by the
4885 selinux= [SELINUX] Disable or enable SELinux at boot time.
4886 Format: { "0" | "1" }
4887 See security/selinux/Kconfig help text.
4892 apparmor= [APPARMOR] Disable or enable AppArmor at boot time
4893 Format: { "0" | "1" }
4894 See security/apparmor/Kconfig help text
4897 Default value is set via kernel config option.
4899 serialnumber [BUGS=X86-32]
4902 Maximal number of shapers.
4910 Disable merging of slabs with similar size. May be
4911 necessary if there is some reason to distinguish
4912 allocs to different slabs, especially in hardened
4913 environments where the risk of heap overflows and
4914 layout control by attackers can usually be
4915 frustrated by disabling merging. This will reduce
4916 most of the exposure of a heap attack to a single
4917 cache (risks via metadata attacks are mostly
4918 unchanged). Debug options disable merging on their
4920 For more information see Documentation/vm/slub.rst.
4922 slab_max_order= [MM, SLAB]
4923 Determines the maximum allowed order for slabs.
4924 A high setting may cause OOMs due to memory
4925 fragmentation. Defaults to 1 for systems with
4926 more than 32MB of RAM, 0 otherwise.
4928 slub_debug[=options[,slabs][;[options[,slabs]]...] [MM, SLUB]
4929 Enabling slub_debug allows one to determine the
4930 culprit if slab objects become corrupted. Enabling
4931 slub_debug can create guard zones around objects and
4932 may poison objects when not in use. Also tracks the
4933 last alloc / free. For more information see
4934 Documentation/vm/slub.rst.
4936 slub_max_order= [MM, SLUB]
4937 Determines the maximum allowed order for slabs.
4938 A high setting may cause OOMs due to memory
4939 fragmentation. For more information see
4940 Documentation/vm/slub.rst.
4942 slub_min_objects= [MM, SLUB]
4943 The minimum number of objects per slab. SLUB will
4944 increase the slab order up to slub_max_order to
4945 generate a sufficiently large slab able to contain
4946 the number of objects indicated. The higher the number
4947 of objects the smaller the overhead of tracking slabs
4948 and the less frequently locks need to be acquired.
4949 For more information see Documentation/vm/slub.rst.
4951 slub_min_order= [MM, SLUB]
4952 Determines the minimum page order for slabs. Must be
4953 lower than slub_max_order.
4954 For more information see Documentation/vm/slub.rst.
4956 slub_nomerge [MM, SLUB]
4957 Same with slab_nomerge. This is supported for legacy.
4958 See slab_nomerge for more information.
4961 Format: <io1>[,<io2>[,...,<io8>]]
4963 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices
4964 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port
4965 smsc-ircc2.ircc_sir= [HW] SIR base I/O port
4966 smsc-ircc2.ircc_fir= [HW] FIR base I/O port
4967 smsc-ircc2.ircc_irq= [HW] IRQ line
4968 smsc-ircc2.ircc_dma= [HW] DMA channel
4969 smsc-ircc2.ircc_transceiver= [HW] Transceiver type:
4970 0: Toshiba Satellite 1800 (GP data pin select)
4971 1: Fast pin select (default)
4974 smt [KNL,S390] Set the maximum number of threads (logical
4975 CPUs) to use per physical CPU on systems capable of
4976 symmetric multithreading (SMT). Will be capped to the
4977 actual hardware limit.
4979 Default: -1 (no limit)
4982 [KNL] Should the soft-lockup detector generate panics.
4985 A value of 1 instructs the soft-lockup detector
4986 to panic the machine when a soft-lockup occurs. It is
4987 also controlled by the kernel.softlockup_panic sysctl
4988 and CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC, which is the
4989 respective build-time switch to that functionality.
4991 softlockup_all_cpu_backtrace=
4992 [KNL] Should the soft-lockup detector generate
4993 backtraces on all cpus.
4996 sonypi.*= [HW] Sony Programmable I/O Control Device driver
4997 See Documentation/admin-guide/laptops/sonypi.rst
4999 spectre_v2= [X86] Control mitigation of Spectre variant 2
5000 (indirect branch speculation) vulnerability.
5001 The default operation protects the kernel from
5004 on - unconditionally enable, implies
5006 off - unconditionally disable, implies
5008 auto - kernel detects whether your CPU model is
5011 Selecting 'on' will, and 'auto' may, choose a
5012 mitigation method at run time according to the
5013 CPU, the available microcode, the setting of the
5014 CONFIG_RETPOLINE configuration option, and the
5015 compiler with which the kernel was built.
5017 Selecting 'on' will also enable the mitigation
5018 against user space to user space task attacks.
5020 Selecting 'off' will disable both the kernel and
5021 the user space protections.
5023 Specific mitigations can also be selected manually:
5025 retpoline - replace indirect branches
5026 retpoline,generic - google's original retpoline
5027 retpoline,amd - AMD-specific minimal thunk
5029 Not specifying this option is equivalent to
5033 [X86] Control mitigation of Spectre variant 2
5034 (indirect branch speculation) vulnerability between
5037 on - Unconditionally enable mitigations. Is
5038 enforced by spectre_v2=on
5040 off - Unconditionally disable mitigations. Is
5041 enforced by spectre_v2=off
5043 prctl - Indirect branch speculation is enabled,
5044 but mitigation can be enabled via prctl
5045 per thread. The mitigation control state
5046 is inherited on fork.
5049 - Like "prctl" above, but only STIBP is
5050 controlled per thread. IBPB is issued
5051 always when switching between different user
5055 - Same as "prctl" above, but all seccomp
5056 threads will enable the mitigation unless
5057 they explicitly opt out.
5060 - Like "seccomp" above, but only STIBP is
5061 controlled per thread. IBPB is issued
5062 always when switching between different
5063 user space processes.
5065 auto - Kernel selects the mitigation depending on
5066 the available CPU features and vulnerability.
5069 If CONFIG_SECCOMP=y then "seccomp", otherwise "prctl"
5071 Not specifying this option is equivalent to
5072 spectre_v2_user=auto.
5074 spec_store_bypass_disable=
5075 [HW] Control Speculative Store Bypass (SSB) Disable mitigation
5076 (Speculative Store Bypass vulnerability)
5078 Certain CPUs are vulnerable to an exploit against a
5079 a common industry wide performance optimization known
5080 as "Speculative Store Bypass" in which recent stores
5081 to the same memory location may not be observed by
5082 later loads during speculative execution. The idea
5083 is that such stores are unlikely and that they can
5084 be detected prior to instruction retirement at the
5085 end of a particular speculation execution window.
5087 In vulnerable processors, the speculatively forwarded
5088 store can be used in a cache side channel attack, for
5089 example to read memory to which the attacker does not
5090 directly have access (e.g. inside sandboxed code).
5092 This parameter controls whether the Speculative Store
5093 Bypass optimization is used.
5095 On x86 the options are:
5097 on - Unconditionally disable Speculative Store Bypass
5098 off - Unconditionally enable Speculative Store Bypass
5099 auto - Kernel detects whether the CPU model contains an
5100 implementation of Speculative Store Bypass and
5101 picks the most appropriate mitigation. If the
5102 CPU is not vulnerable, "off" is selected. If the
5103 CPU is vulnerable the default mitigation is
5104 architecture and Kconfig dependent. See below.
5105 prctl - Control Speculative Store Bypass per thread
5106 via prctl. Speculative Store Bypass is enabled
5107 for a process by default. The state of the control
5108 is inherited on fork.
5109 seccomp - Same as "prctl" above, but all seccomp threads
5110 will disable SSB unless they explicitly opt out.
5112 Default mitigations:
5113 X86: If CONFIG_SECCOMP=y "seccomp", otherwise "prctl"
5115 On powerpc the options are:
5117 on,auto - On Power8 and Power9 insert a store-forwarding
5118 barrier on kernel entry and exit. On Power7
5119 perform a software flush on kernel entry and
5123 Not specifying this option is equivalent to
5124 spec_store_bypass_disable=auto.
5126 spia_io_base= [HW,MTD]
5132 [X86] Enable split lock detection or bus lock detection
5134 When enabled (and if hardware support is present), atomic
5135 instructions that access data across cache line
5136 boundaries will result in an alignment check exception
5137 for split lock detection or a debug exception for
5142 warn - the kernel will emit rate-limited warnings
5143 about applications triggering the #AC
5144 exception or the #DB exception. This mode is
5145 the default on CPUs that support split lock
5146 detection or bus lock detection. Default
5147 behavior is by #AC if both features are
5148 enabled in hardware.
5150 fatal - the kernel will send SIGBUS to applications
5151 that trigger the #AC exception or the #DB
5152 exception. Default behavior is by #AC if
5153 both features are enabled in hardware.
5155 If an #AC exception is hit in the kernel or in
5156 firmware (i.e. not while executing in user mode)
5157 the kernel will oops in either "warn" or "fatal"
5160 #DB exception for bus lock is triggered only when
5164 Control the Special Register Buffer Data Sampling
5167 Certain CPUs are vulnerable to an MDS-like
5168 exploit which can leak bits from the random
5171 By default, this issue is mitigated by
5172 microcode. However, the microcode fix can cause
5173 the RDRAND and RDSEED instructions to become
5174 much slower. Among other effects, this will
5175 result in reduced throughput from /dev/urandom.
5177 The microcode mitigation can be disabled with
5178 the following option:
5180 off: Disable mitigation and remove
5181 performance impact to RDRAND and RDSEED
5183 srcutree.counter_wrap_check [KNL]
5184 Specifies how frequently to check for
5185 grace-period sequence counter wrap for the
5186 srcu_data structure's ->srcu_gp_seq_needed field.
5187 The greater the number of bits set in this kernel
5188 parameter, the less frequently counter wrap will
5189 be checked for. Note that the bottom two bits
5192 srcutree.exp_holdoff [KNL]
5193 Specifies how many nanoseconds must elapse
5194 since the end of the last SRCU grace period for
5195 a given srcu_struct until the next normal SRCU
5196 grace period will be considered for automatic
5197 expediting. Set to zero to disable automatic
5201 Speculative Store Bypass Disable control
5203 On CPUs that are vulnerable to the Speculative
5204 Store Bypass vulnerability and offer a
5205 firmware based mitigation, this parameter
5206 indicates how the mitigation should be used:
5208 force-on: Unconditionally enable mitigation for
5209 for both kernel and userspace
5210 force-off: Unconditionally disable mitigation for
5211 for both kernel and userspace
5212 kernel: Always enable mitigation in the
5213 kernel, and offer a prctl interface
5214 to allow userspace to register its
5215 interest in being mitigated too.
5217 stack_guard_gap= [MM]
5218 override the default stack gap protection. The value
5219 is in page units and it defines how many pages prior
5220 to (for stacks growing down) resp. after (for stacks
5221 growing up) the main stack are reserved for no other
5222 mapping. Default value is 256 pages.
5224 stack_depot_disable= [KNL]
5225 Setting this to true through kernel command line will
5226 disable the stack depot thereby saving the static memory
5227 consumed by the stack hash table. By default this is set
5231 Enabled the stack tracer on boot up.
5233 stacktrace_filter=[function-list]
5234 [FTRACE] Limit the functions that the stack tracer
5235 will trace at boot up. function-list is a comma-separated
5236 list of functions. This list can be changed at run
5237 time by the stack_trace_filter file in the debugfs
5238 tracing directory. Note, this enables stack tracing
5239 and the stacktrace above is not needed.
5243 Set the STI (builtin display/keyboard on the HP-PARISC
5244 machines) console (graphic card) which should be used
5245 as the initial boot-console.
5246 See also comment in drivers/video/console/sticore.c.
5249 See comment in drivers/video/console/sticore.c.
5252 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]]
5254 sunrpc.min_resvport=
5255 sunrpc.max_resvport=
5257 SunRPC servers often require that client requests
5258 originate from a privileged port (i.e. a port in the
5259 range 0 < portnr < 1024).
5260 An administrator who wishes to reserve some of these
5261 ports for other uses may adjust the range that the
5262 kernel's sunrpc client considers to be privileged
5263 using these two parameters to set the minimum and
5264 maximum port values.
5266 sunrpc.svc_rpc_per_connection_limit=
5268 Limit the number of requests that the server will
5269 process in parallel from a single connection.
5270 The default value is 0 (no limit).
5274 Control how the NFS server code allocates CPUs to
5275 service thread pools. Depending on how many NICs
5276 you have and where their interrupts are bound, this
5277 option will affect which CPUs will do NFS serving.
5278 Note: this parameter cannot be changed while the
5279 NFS server is running.
5281 auto the server chooses an appropriate mode
5282 automatically using heuristics
5283 global a single global pool contains all CPUs
5284 percpu one pool for each CPU
5285 pernode one pool for each NUMA node (equivalent
5286 to global on non-NUMA machines)
5288 sunrpc.tcp_slot_table_entries=
5289 sunrpc.udp_slot_table_entries=
5291 Sets the upper limit on the number of simultaneous
5292 RPC calls that can be sent from the client to a
5293 server. Increasing these values may allow you to
5294 improve throughput, but will also increase the
5295 amount of memory reserved for use by the client.
5297 suspend.pm_test_delay=
5299 Sets the number of seconds to remain in a suspend test
5300 mode before resuming the system (see
5301 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG
5302 is set. Default value is 5.
5305 Format: { on | off | y | n | 1 | 0 }
5306 This parameter controls use of the Protected
5307 Execution Facility on pSeries.
5310 [KNL] Enable accounting of swap in memory resource
5311 controller if no parameter or 1 is given or disable
5312 it if 0 is given (See Documentation/admin-guide/cgroup-v1/memory.rst)
5314 swiotlb= [ARM,IA-64,PPC,MIPS,X86]
5315 Format: { <int> | force | noforce }
5316 <int> -- Number of I/O TLB slabs
5317 force -- force using of bounce buffers even if they
5318 wouldn't be automatically used by the kernel
5319 noforce -- Never use bounce buffers (for debugging)
5324 Set a sysctl parameter, right before loading the init
5325 process, as if the value was written to the respective
5326 /proc/sys/... file. Both '.' and '/' are recognized as
5327 separators. Unrecognized parameters and invalid values
5328 are reported in the kernel log. Sysctls registered
5329 later by a loaded module cannot be set this way.
5330 Example: sysctl.vm.swappiness=40
5332 sysfs.deprecated=0|1 [KNL]
5333 Enable/disable old style sysfs layout for old udev
5334 on older distributions. When this option is enabled
5335 very new udev will not work anymore. When this option
5336 is disabled (or CONFIG_SYSFS_DEPRECATED not compiled)
5337 in older udev will not work anymore.
5338 Default depends on CONFIG_SYSFS_DEPRECATED_V2 set in
5339 the kernel configuration.
5341 sysrq_always_enabled
5343 Ignore sysrq setting - this boot parameter will
5344 neutralize any effect of /proc/sys/kernel/sysrq.
5345 Useful for debugging.
5347 tcpmhash_entries= [KNL,NET]
5348 Set the number of tcp_metrics_hash slots.
5349 Default value is 8192 or 16384 depending on total
5350 ram pages. This is used to specify the TCP metrics
5351 cache size. See Documentation/networking/ip-sysctl.rst
5352 "tcp_no_metrics_save" section for more details.
5356 test_suspend= [SUSPEND][,N]
5357 Specify "mem" (for Suspend-to-RAM) or "standby" (for
5358 standby suspend) or "freeze" (for suspend type freeze)
5359 as the system sleep state during system startup with
5360 the optional capability to repeat N number of times.
5361 The system is woken from this state using a
5362 wakeup-capable RTC alarm.
5364 thash_entries= [KNL,NET]
5365 Set number of hash buckets for TCP connection
5367 thermal.act= [HW,ACPI]
5368 -1: disable all active trip points in all thermal zones
5369 <degrees C>: override all lowest active trip points
5371 thermal.crt= [HW,ACPI]
5372 -1: disable all critical trip points in all thermal zones
5373 <degrees C>: override all critical trip points
5375 thermal.nocrt= [HW,ACPI]
5376 Set to disable actions on ACPI thermal zone
5377 critical and hot trip points.
5379 thermal.off= [HW,ACPI]
5380 1: disable ACPI thermal control
5382 thermal.psv= [HW,ACPI]
5383 -1: disable all passive trip points
5384 <degrees C>: override all passive trip points to this
5387 thermal.tzp= [HW,ACPI]
5388 Specify global default ACPI thermal zone polling rate
5389 <deci-seconds>: poll all this frequency
5390 0: no polling (default)
5393 Force threading of all interrupt handlers except those
5394 marked explicitly IRQF_NO_THREAD.
5398 Specify if the kernel should make use of the cpu
5399 topology information if the hardware supports this.
5400 The scheduler will make use of this information and
5401 e.g. base its process migration decisions on it.
5404 topology_updates= [KNL, PPC, NUMA]
5406 Specify if the kernel should ignore (off)
5407 topology updates sent by the hypervisor to this
5410 torture.disable_onoff_at_boot= [KNL]
5411 Prevent the CPU-hotplug component of torturing
5412 until after init has spawned.
5414 torture.ftrace_dump_at_shutdown= [KNL]
5415 Dump the ftrace buffer at torture-test shutdown,
5416 even if there were no errors. This can be a
5417 very costly operation when many torture tests
5418 are running concurrently, especially on systems
5419 with rotating-rust storage.
5421 torture.verbose_sleep_frequency= [KNL]
5422 Specifies how many verbose printk()s should be
5423 emitted between each sleep. The default of zero
5424 disables verbose-printk() sleeping.
5426 torture.verbose_sleep_duration= [KNL]
5427 Duration of each verbose-printk() sleep in jiffies.
5431 tpm_suspend_pcr=[HW,TPM]
5432 Format: integer pcr id
5433 Specify that at suspend time, the tpm driver
5434 should extend the specified pcr with zeros,
5435 as a workaround for some chips which fail to
5436 flush the last written pcr on TPM_SaveState.
5437 This will guarantee that all the other pcrs
5440 trace_buf_size=nn[KMG]
5441 [FTRACE] will set tracing buffer size on each cpu.
5443 trace_event=[event-list]
5444 [FTRACE] Set and start specified trace events in order
5445 to facilitate early boot debugging. The event-list is a
5446 comma-separated list of trace events to enable. See
5447 also Documentation/trace/events.rst
5449 trace_options=[option-list]
5450 [FTRACE] Enable or disable tracer options at boot.
5451 The option-list is a comma delimited list of options
5452 that can be enabled or disabled just as if you were
5453 to echo the option name into
5455 /sys/kernel/debug/tracing/trace_options
5457 For example, to enable stacktrace option (to dump the
5458 stack trace of each event), add to the command line:
5460 trace_options=stacktrace
5462 See also Documentation/trace/ftrace.rst "trace options"
5466 Have the tracepoints sent to printk as well as the
5467 tracing ring buffer. This is useful for early boot up
5468 where the system hangs or reboots and does not give the
5469 option for reading the tracing buffer or performing a
5470 ftrace_dump_on_oops.
5472 To turn off having tracepoints sent to printk,
5473 echo 0 > /proc/sys/kernel/tracepoint_printk
5474 Note, echoing 1 into this file without the
5475 tracepoint_printk kernel cmdline option has no effect.
5479 Having tracepoints sent to printk() and activating high
5480 frequency tracepoints such as irq or sched, can cause
5481 the system to live lock.
5484 [FTRACE] enable this option to disable tracing when a
5485 warning is hit. This turns off "tracing_on". Tracing can
5486 be enabled again by echoing '1' into the "tracing_on"
5487 file located in /sys/kernel/debug/tracing/
5489 This option is useful, as it disables the trace before
5490 the WARNING dump is called, which prevents the trace to
5491 be filled with content caused by the warning output.
5493 This option can also be set at run time via the sysctl
5494 option: kernel/traceoff_on_warning
5496 transparent_hugepage=
5498 Format: [always|madvise|never]
5499 Can be used to control the default behavior of the system
5500 with respect to transparent hugepages.
5501 See Documentation/admin-guide/mm/transhuge.rst
5504 trusted.source= [KEYS]
5506 This parameter identifies the trust source as a backend
5507 for trusted keys implementation. Supported trust
5511 If not specified then it defaults to iterating through
5512 the trust source list starting with TPM and assigns the
5513 first trust source as a backend which is initialized
5514 successfully during iteration.
5516 tsc= Disable clocksource stability checks for TSC.
5518 [x86] reliable: mark tsc clocksource as reliable, this
5519 disables clocksource verification at runtime, as well
5520 as the stability checks done at bootup. Used to enable
5521 high-resolution timer mode on older hardware, and in
5522 virtualized environment.
5523 [x86] noirqtime: Do not use TSC to do irq accounting.
5524 Used to run time disable IRQ_TIME_ACCOUNTING on any
5525 platforms where RDTSC is slow and this accounting
5527 [x86] unstable: mark the TSC clocksource as unstable, this
5528 marks the TSC unconditionally unstable at bootup and
5529 avoids any further wobbles once the TSC watchdog notices.
5530 [x86] nowatchdog: disable clocksource watchdog. Used
5531 in situations with strict latency requirements (where
5532 interruptions from clocksource watchdog are not
5535 tsc_early_khz= [X86] Skip early TSC calibration and use the given
5536 value instead. Useful when the early TSC frequency discovery
5537 procedure is not reliable, such as on overclocked systems
5538 with CPUID.16h support and partial CPUID.15h support.
5539 Format: <unsigned int>
5541 tsx= [X86] Control Transactional Synchronization
5542 Extensions (TSX) feature in Intel processors that
5543 support TSX control.
5545 This parameter controls the TSX feature. The options are:
5547 on - Enable TSX on the system. Although there are
5548 mitigations for all known security vulnerabilities,
5549 TSX has been known to be an accelerator for
5550 several previous speculation-related CVEs, and
5551 so there may be unknown security risks associated
5552 with leaving it enabled.
5554 off - Disable TSX on the system. (Note that this
5555 option takes effect only on newer CPUs which are
5556 not vulnerable to MDS, i.e., have
5557 MSR_IA32_ARCH_CAPABILITIES.MDS_NO=1 and which get
5558 the new IA32_TSX_CTRL MSR through a microcode
5559 update. This new MSR allows for the reliable
5560 deactivation of the TSX functionality.)
5562 auto - Disable TSX if X86_BUG_TAA is present,
5563 otherwise enable TSX on the system.
5565 Not specifying this option is equivalent to tsx=off.
5567 See Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
5570 tsx_async_abort= [X86,INTEL] Control mitigation for the TSX Async
5571 Abort (TAA) vulnerability.
5573 Similar to Micro-architectural Data Sampling (MDS)
5574 certain CPUs that support Transactional
5575 Synchronization Extensions (TSX) are vulnerable to an
5576 exploit against CPU internal buffers which can forward
5577 information to a disclosure gadget under certain
5580 In vulnerable processors, the speculatively forwarded
5581 data can be used in a cache side channel attack, to
5582 access data to which the attacker does not have direct
5585 This parameter controls the TAA mitigation. The
5588 full - Enable TAA mitigation on vulnerable CPUs
5591 full,nosmt - Enable TAA mitigation and disable SMT on
5592 vulnerable CPUs. If TSX is disabled, SMT
5593 is not disabled because CPU is not
5594 vulnerable to cross-thread TAA attacks.
5595 off - Unconditionally disable TAA mitigation
5597 On MDS-affected machines, tsx_async_abort=off can be
5598 prevented by an active MDS mitigation as both vulnerabilities
5599 are mitigated with the same mechanism so in order to disable
5600 this mitigation, you need to specify mds=off too.
5602 Not specifying this option is equivalent to
5603 tsx_async_abort=full. On CPUs which are MDS affected
5604 and deploy MDS mitigation, TAA mitigation is not
5605 required and doesn't provide any additional
5609 Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
5611 turbografx.map[2|3]= [HW,JOY]
5612 TurboGraFX parallel port interface
5614 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7>
5615 See also Documentation/input/devices/joystick-parport.rst
5617 udbg-immortal [PPC] When debugging early kernel crashes that
5618 happen after console_init() and before a proper
5619 console driver takes over, this boot options might
5620 help "seeing" what's going on.
5622 uhash_entries= [KNL,NET]
5623 Set number of hash buckets for UDP/UDP-Lite connections
5626 [USB] Ignore overcurrent events (default N).
5627 Some badly-designed motherboards generate lots of
5628 bogus events, for ports that aren't wired to
5629 anything. Set this parameter to avoid log spamming.
5630 Note that genuine overcurrent events won't be
5634 [X86] Cause panic on unknown NMI.
5636 usbcore.authorized_default=
5637 [USB] Default USB device authorization:
5638 (default -1 = authorized except for wireless USB,
5639 0 = not authorized, 1 = authorized, 2 = authorized
5640 if device connected to internal port)
5642 usbcore.autosuspend=
5643 [USB] The autosuspend time delay (in seconds) used
5644 for newly-detected USB devices (default 2). This
5645 is the time required before an idle device will be
5646 autosuspended. Devices for which the delay is set
5647 to a negative value won't be autosuspended at all.
5649 usbcore.usbfs_snoop=
5650 [USB] Set to log all usbfs traffic (default 0 = off).
5652 usbcore.usbfs_snoop_max=
5653 [USB] Maximum number of bytes to snoop in each URB
5656 usbcore.blinkenlights=
5657 [USB] Set to cycle leds on hubs (default 0 = off).
5659 usbcore.old_scheme_first=
5660 [USB] Start with the old device initialization
5661 scheme (default 0 = off).
5663 usbcore.usbfs_memory_mb=
5664 [USB] Memory limit (in MB) for buffers allocated by
5665 usbfs (default = 16, 0 = max = 2047).
5667 usbcore.use_both_schemes=
5668 [USB] Try the other device initialization scheme
5669 if the first one fails (default 1 = enabled).
5671 usbcore.initial_descriptor_timeout=
5672 [USB] Specifies timeout for the initial 64-byte
5673 USB_REQ_GET_DESCRIPTOR request in milliseconds
5674 (default 5000 = 5.0 seconds).
5676 usbcore.nousb [USB] Disable the USB subsystem
5679 [USB] A list of quirk entries to augment the built-in
5680 usb core quirk list. List entries are separated by
5681 commas. Each entry has the form
5682 VendorID:ProductID:Flags. The IDs are 4-digit hex
5683 numbers and Flags is a set of letters. Each letter
5684 will change the built-in quirk; setting it if it is
5685 clear and clearing it if it is set. The letters have
5686 the following meanings:
5687 a = USB_QUIRK_STRING_FETCH_255 (string
5688 descriptors must not be fetched using
5690 b = USB_QUIRK_RESET_RESUME (device can't resume
5691 correctly so reset it instead);
5692 c = USB_QUIRK_NO_SET_INTF (device can't handle
5693 Set-Interface requests);
5694 d = USB_QUIRK_CONFIG_INTF_STRINGS (device can't
5695 handle its Configuration or Interface
5697 e = USB_QUIRK_RESET (device can't be reset
5698 (e.g morph devices), don't use reset);
5699 f = USB_QUIRK_HONOR_BNUMINTERFACES (device has
5700 more interface descriptions than the
5701 bNumInterfaces count, and can't handle
5702 talking to these interfaces);
5703 g = USB_QUIRK_DELAY_INIT (device needs a pause
5704 during initialization, after we read
5705 the device descriptor);
5706 h = USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL (For
5707 high speed and super speed interrupt
5708 endpoints, the USB 2.0 and USB 3.0 spec
5709 require the interval in microframes (1
5710 microframe = 125 microseconds) to be
5711 calculated as interval = 2 ^
5713 Devices with this quirk report their
5714 bInterval as the result of this
5715 calculation instead of the exponent
5716 variable used in the calculation);
5717 i = USB_QUIRK_DEVICE_QUALIFIER (device can't
5718 handle device_qualifier descriptor
5720 j = USB_QUIRK_IGNORE_REMOTE_WAKEUP (device
5721 generates spurious wakeup, ignore
5722 remote wakeup capability);
5723 k = USB_QUIRK_NO_LPM (device can't handle Link
5725 l = USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL
5726 (Device reports its bInterval as linear
5727 frames instead of the USB 2.0
5729 m = USB_QUIRK_DISCONNECT_SUSPEND (Device needs
5730 to be disconnected before suspend to
5731 prevent spurious wakeup);
5732 n = USB_QUIRK_DELAY_CTRL_MSG (Device needs a
5733 pause after every control message);
5734 o = USB_QUIRK_HUB_SLOW_RESET (Hub needs extra
5735 delay after resetting its port);
5736 Example: quirks=0781:5580:bk,0a5c:5834:gij
5739 [USBHID] The interval which mice are to be polled at.
5742 [USBHID] The interval which joysticks are to be polled at.
5745 [USBHID] The interval which keyboards are to be polled at.
5747 usb-storage.delay_use=
5748 [UMS] The delay in seconds before a new device is
5749 scanned for Logical Units (default 1).
5752 [UMS] A list of quirks entries to supplement or
5753 override the built-in unusual_devs list. List
5754 entries are separated by commas. Each entry has
5755 the form VID:PID:Flags where VID and PID are Vendor
5756 and Product ID values (4-digit hex numbers) and
5757 Flags is a set of characters, each corresponding
5758 to a common usb-storage quirk flag as follows:
5759 a = SANE_SENSE (collect more than 18 bytes
5760 of sense data, not on uas);
5761 b = BAD_SENSE (don't collect more than 18
5762 bytes of sense data, not on uas);
5763 c = FIX_CAPACITY (decrease the reported
5764 device capacity by one sector);
5765 d = NO_READ_DISC_INFO (don't use
5766 READ_DISC_INFO command, not on uas);
5767 e = NO_READ_CAPACITY_16 (don't use
5768 READ_CAPACITY_16 command);
5769 f = NO_REPORT_OPCODES (don't use report opcodes
5771 g = MAX_SECTORS_240 (don't transfer more than
5772 240 sectors at a time, uas only);
5773 h = CAPACITY_HEURISTICS (decrease the
5774 reported device capacity by one
5775 sector if the number is odd);
5776 i = IGNORE_DEVICE (don't bind to this
5778 j = NO_REPORT_LUNS (don't use report luns
5780 k = NO_SAME (do not use WRITE_SAME, uas only)
5781 l = NOT_LOCKABLE (don't try to lock and
5782 unlock ejectable media, not on uas);
5783 m = MAX_SECTORS_64 (don't transfer more
5784 than 64 sectors = 32 KB at a time,
5786 n = INITIAL_READ10 (force a retry of the
5787 initial READ(10) command, not on uas);
5788 o = CAPACITY_OK (accept the capacity
5789 reported by the device, not on uas);
5790 p = WRITE_CACHE (the device cache is ON
5791 by default, not on uas);
5792 r = IGNORE_RESIDUE (the device reports
5793 bogus residue values, not on uas);
5794 s = SINGLE_LUN (the device has only one
5796 t = NO_ATA_1X (don't allow ATA(12) and ATA(16)
5797 commands, uas only);
5798 u = IGNORE_UAS (don't bind to the uas driver);
5799 w = NO_WP_DETECT (don't test whether the
5800 medium is write-protected).
5801 y = ALWAYS_SYNC (issue a SYNCHRONIZE_CACHE
5802 even if the device claims no cache,
5804 Example: quirks=0419:aaf5:rl,0421:0433:rc
5806 user_debug= [KNL,ARM]
5808 See arch/arm/Kconfig.debug help text.
5809 1 - undefined instruction events
5811 4 - invalid data aborts
5814 Example: user_debug=31
5817 [X86] Flags controlling user PTE allocations.
5819 nohigh = do not allocate PTE pages in
5820 HIGHMEM regardless of setting
5824 On X86_32, this is an alias for vdso32=. Otherwise:
5826 vdso=1: enable VDSO (the default)
5827 vdso=0: disable VDSO mapping
5829 vdso32= [X86] Control the 32-bit vDSO
5830 vdso32=1: enable 32-bit VDSO
5831 vdso32=0 or vdso32=2: disable 32-bit VDSO
5833 See the help text for CONFIG_COMPAT_VDSO for more
5834 details. If CONFIG_COMPAT_VDSO is set, the default is
5835 vdso32=0; otherwise, the default is vdso32=1.
5837 For compatibility with older kernels, vdso32=2 is an
5840 Try vdso32=0 if you encounter an error that says:
5841 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed!
5844 vector=percpu: enable percpu vector domain
5846 video= [FB] Frame buffer configuration
5847 See Documentation/fb/modedb.rst.
5849 video.brightness_switch_enabled= [0,1]
5850 If set to 1, on receiving an ACPI notify event
5851 generated by hotkey, video driver will adjust brightness
5852 level and then send out the event to user space through
5853 the allocated input device; If set to 0, video driver
5854 will only send out the event without touching backlight
5859 [VMMIO] Memory mapped virtio (platform) device.
5861 <size>@<baseaddr>:<irq>[:<id>]
5863 <size> := size (can use standard suffixes
5865 <baseaddr> := physical base address
5866 <irq> := interrupt number (as passed to
5868 <id> := (optional) platform device id
5870 virtio_mmio.device=1K@0x100b0000:48:7
5872 Can be used multiple times for multiple devices.
5874 vga= [BOOT,X86-32] Select a particular video mode
5875 See Documentation/x86/boot.rst and
5876 Documentation/admin-guide/svga.rst.
5877 Use vga=ask for menu.
5878 This is actually a boot loader parameter; the value is
5879 passed to the kernel using a special protocol.
5881 vm_debug[=options] [KNL] Available with CONFIG_DEBUG_VM=y.
5882 May slow down system boot speed, especially when
5883 enabled on systems with a large amount of memory.
5884 All options are enabled by default, and this
5885 interface is meant to allow for selectively
5886 enabling or disabling specific virtual memory
5889 Available options are:
5890 P Enable page structure init time poisoning
5891 - Disable all of the above options
5893 vmalloc=nn[KMG] [KNL,BOOT] Forces the vmalloc area to have an exact
5894 size of <nn>. This can be used to increase the
5895 minimum size (128MB on x86). It can also be used to
5896 decrease the size and leave more room for directly
5899 vmcp_cma=nn[MG] [KNL,S390]
5900 Sets the memory size reserved for contiguous memory
5901 allocations for the vmcp device driver.
5903 vmhalt= [KNL,S390] Perform z/VM CP command after system halt.
5906 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic.
5909 vmpoff= [KNL,S390] Perform z/VM CP command after power off.
5913 Controls the behavior of vsyscalls (i.e. calls to
5914 fixed addresses of 0xffffffffff600x00 from legacy
5915 code). Most statically-linked binaries and older
5916 versions of glibc use these calls. Because these
5917 functions are at fixed addresses, they make nice
5918 targets for exploits that can control RIP.
5920 emulate [default] Vsyscalls turn into traps and are
5921 emulated reasonably safely. The vsyscall
5924 xonly Vsyscalls turn into traps and are
5925 emulated reasonably safely. The vsyscall
5926 page is not readable.
5928 none Vsyscalls don't work at all. This makes
5929 them quite hard to use for exploits but
5930 might break your system.
5932 vt.color= [VT] Default text color.
5933 Format: 0xYX, X = foreground, Y = background.
5934 Default: 0x07 = light gray on black.
5936 vt.cur_default= [VT] Default cursor shape.
5937 Format: 0xCCBBAA, where AA, BB, and CC are the same as
5938 the parameters of the <Esc>[?A;B;Cc escape sequence;
5939 see VGA-softcursor.txt. Default: 2 = underline.
5941 vt.default_blu= [VT]
5942 Format: <blue0>,<blue1>,<blue2>,...,<blue15>
5943 Change the default blue palette of the console.
5944 This is a 16-member array composed of values
5947 vt.default_grn= [VT]
5948 Format: <green0>,<green1>,<green2>,...,<green15>
5949 Change the default green palette of the console.
5950 This is a 16-member array composed of values
5953 vt.default_red= [VT]
5954 Format: <red0>,<red1>,<red2>,...,<red15>
5955 Change the default red palette of the console.
5956 This is a 16-member array composed of values
5962 Set system-wide default UTF-8 mode for all tty's.
5963 Default is 1, i.e. UTF-8 mode is enabled for all
5964 newly opened terminals.
5966 vt.global_cursor_default=
5969 Set system-wide default for whether a cursor
5970 is shown on new VTs. Default is -1,
5971 i.e. cursors will be created by default unless
5972 overridden by individual drivers. 0 will hide
5973 cursors, 1 will display them.
5975 vt.italic= [VT] Default color for italic text; 0-15.
5978 vt.underline= [VT] Default color for underlined text; 0-15.
5981 watchdog timers [HW,WDT] For information on watchdog timers,
5982 see Documentation/watchdog/watchdog-parameters.rst
5983 or other driver-specific files in the
5984 Documentation/watchdog/ directory.
5988 Set the hard lockup detector stall duration
5989 threshold in seconds. The soft lockup detector
5990 threshold is set to twice the value. A value of 0
5991 disables both lockup detectors. Default is 10
5994 workqueue.watchdog_thresh=
5995 If CONFIG_WQ_WATCHDOG is configured, workqueue can
5996 warn stall conditions and dump internal state to
5997 help debugging. 0 disables workqueue stall
5998 detection; otherwise, it's the stall threshold
5999 duration in seconds. The default value is 30 and
6000 it can be updated at runtime by writing to the
6001 corresponding sysfs file.
6003 workqueue.disable_numa
6004 By default, all work items queued to unbound
6005 workqueues are affine to the NUMA nodes they're
6006 issued on, which results in better behavior in
6007 general. If NUMA affinity needs to be disabled for
6008 whatever reason, this option can be used. Note
6009 that this also can be controlled per-workqueue for
6010 workqueues visible under /sys/bus/workqueue/.
6012 workqueue.power_efficient
6013 Per-cpu workqueues are generally preferred because
6014 they show better performance thanks to cache
6015 locality; unfortunately, per-cpu workqueues tend to
6016 be more power hungry than unbound workqueues.
6018 Enabling this makes the per-cpu workqueues which
6019 were observed to contribute significantly to power
6020 consumption unbound, leading to measurably lower
6021 power usage at the cost of small performance
6024 The default value of this parameter is determined by
6025 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT.
6027 workqueue.debug_force_rr_cpu
6028 Workqueue used to implicitly guarantee that work
6029 items queued without explicit CPU specified are put
6030 on the local CPU. This guarantee is no longer true
6031 and while local CPU is still preferred work items
6032 may be put on foreign CPUs. This debug option
6033 forces round-robin CPU selection to flush out
6034 usages which depend on the now broken guarantee.
6035 When enabled, memory and cache locality will be
6038 x2apic_phys [X86-64,APIC] Use x2apic physical mode instead of
6039 default x2apic cluster mode on platforms
6042 xen_512gb_limit [KNL,X86-64,XEN]
6043 Restricts the kernel running paravirtualized under Xen
6044 to use only up to 512 GB of RAM. The reason to do so is
6045 crash analysis tools and Xen tools for doing domain
6046 save/restore/migration must be enabled to handle larger
6049 xen_emul_unplug= [HW,X86,XEN]
6050 Unplug Xen emulated devices
6051 Format: [unplug0,][unplug1]
6052 ide-disks -- unplug primary master IDE devices
6053 aux-ide-disks -- unplug non-primary-master IDE devices
6054 nics -- unplug network devices
6055 all -- unplug all emulated devices (NICs and IDE disks)
6056 unnecessary -- unplugging emulated devices is
6057 unnecessary even if the host did not respond to
6059 never -- do not unplug even if version check succeeds
6061 xen_legacy_crash [X86,XEN]
6062 Crash from Xen panic notifier, without executing late
6063 panic() code such as dumping handler.
6065 xen_nopvspin [X86,XEN]
6066 Disables the qspinlock slowpath using Xen PV optimizations.
6067 This parameter is obsoleted by "nopvspin" parameter, which
6068 has equivalent effect for XEN platform.
6071 Disables the PV optimizations forcing the HVM guest to
6072 run as generic HVM guest with no PV drivers.
6073 This option is obsoleted by the "nopv" option, which
6074 has equivalent effect for XEN platform.
6076 xen_no_vector_callback
6077 [KNL,X86,XEN] Disable the vector callback for Xen
6078 event channel interrupts.
6080 xen_scrub_pages= [XEN]
6081 Boolean option to control scrubbing pages before giving them back
6082 to Xen, for use by other domains. Can be also changed at runtime
6083 with /sys/devices/system/xen_memory/xen_memory0/scrub_pages.
6084 Default value controlled with CONFIG_XEN_SCRUB_PAGES_DEFAULT.
6086 xen_timer_slop= [X86-64,XEN]
6087 Set the timer slop (in nanoseconds) for the virtual Xen
6088 timers (default is 100000). This adjusts the minimum
6089 delta of virtualized Xen timers, where lower values
6090 improve timer resolution at the expense of processing
6091 more timer interrupts.
6093 xen.event_eoi_delay= [XEN]
6094 How long to delay EOI handling in case of event
6095 storms (jiffies). Default is 10.
6097 xen.event_loop_timeout= [XEN]
6098 After which time (jiffies) the event handling loop
6099 should start to delay EOI handling. Default is 2.
6101 xen.fifo_events= [XEN]
6102 Boolean parameter to disable using fifo event handling
6103 even if available. Normally fifo event handling is
6104 preferred over the 2-level event handling, as it is
6105 fairer and the number of possible event channels is
6106 much higher. Default is on (use fifo events).
6108 nopv= [X86,XEN,KVM,HYPER_V,VMWARE]
6109 Disables the PV optimizations forcing the guest to run
6110 as generic guest with no PV drivers. Currently support
6111 XEN HVM, KVM, HYPER_V and VMWARE guest.
6113 nopvspin [X86,XEN,KVM]
6114 Disables the qspinlock slow path using PV optimizations
6115 which allow the hypervisor to 'idle' the guest on lock
6118 xirc2ps_cs= [NET,PCMCIA]
6120 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]]
6123 By default on POWER9 and above, the kernel will
6124 natively use the XIVE interrupt controller. This option
6125 allows the fallback firmware mode to be used:
6127 off Fallback to firmware control of XIVE interrupt
6128 controller on both pseries and powernv
6129 platforms. Only useful on POWER9 and above.
6131 xhci-hcd.quirks [USB,KNL]
6132 A hex value specifying bitmask with supplemental xhci
6133 host controller quirks. Meaning of each bit can be
6134 consulted in header drivers/usb/host/xhci.h.
6137 Format: { early | on | rw | ro | off }
6138 Controls if xmon debugger is enabled. Default is off.
6139 Passing only "xmon" is equivalent to "xmon=early".
6140 early Call xmon as early as possible on boot; xmon
6141 debugger is called from setup_arch().
6142 on xmon debugger hooks will be installed so xmon
6143 is only called on a kernel crash. Default mode,
6144 i.e. either "ro" or "rw" mode, is controlled
6145 with CONFIG_XMON_DEFAULT_RO_MODE.
6146 rw xmon debugger hooks will be installed so xmon
6147 is called only on a kernel crash, mode is write,
6148 meaning SPR registers, memory and, other data
6149 can be written using xmon commands.
6150 ro same as "rw" option above but SPR registers,
6151 memory, and other data can't be written using
6153 off xmon is disabled.
projects / linux-2.6-microblaze.git / blob