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_PCI_COMPONENT
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 PCI/PCI interrupt routing info messages:
64 acpi.debug_layer=0x400000
65 Enable AML "Debug" output, i.e., stores to the Debug
66 object while interpreting AML:
67 acpi.debug_layer=0xffffffff acpi.debug_level=0x2
68 Enable all messages related to ACPI hardware:
69 acpi.debug_layer=0x2 acpi.debug_level=0xffffffff
71 Some values produce so much output that the system is
72 unusable. The "log_buf_len" parameter may be useful
73 if you need to capture more output.
75 acpi_enforce_resources= [ACPI]
77 Check for resource conflicts between native drivers
78 and ACPI OperationRegions (SystemIO and SystemMemory
79 only). IO ports and memory declared in ACPI might be
80 used by the ACPI subsystem in arbitrary AML code and
81 can interfere with legacy drivers.
82 strict (default): access to resources claimed by ACPI
83 is denied; legacy drivers trying to access reserved
84 resources will fail to bind to device using them.
85 lax: access to resources claimed by ACPI is allowed;
86 legacy drivers trying to access reserved resources
87 will bind successfully but a warning message is logged.
88 no: ACPI OperationRegions are not marked as reserved,
89 no further checks are performed.
91 acpi_force_table_verification [HW,ACPI]
92 Enable table checksum verification during early stage.
93 By default, this is disabled due to x86 early mapping
96 acpi_irq_balance [HW,ACPI]
97 ACPI will balance active IRQs
100 acpi_irq_nobalance [HW,ACPI]
101 ACPI will not move active IRQs (default)
104 acpi_irq_isa= [HW,ACPI] If irq_balance, mark listed IRQs used by ISA
105 Format: <irq>,<irq>...
107 acpi_irq_pci= [HW,ACPI] If irq_balance, clear listed IRQs for
109 Format: <irq>,<irq>...
111 acpi_mask_gpe= [HW,ACPI]
112 Due to the existence of _Lxx/_Exx, some GPEs triggered
113 by unsupported hardware/firmware features can result in
114 GPE floodings that cannot be automatically disabled by
116 This facility can be used to prevent such uncontrolled
120 acpi_no_auto_serialize [HW,ACPI]
121 Disable auto-serialization of AML methods
122 AML control methods that contain the opcodes to create
123 named objects will be marked as "Serialized" by the
124 auto-serialization feature.
125 This feature is enabled by default.
126 This option allows to turn off the feature.
128 acpi_no_memhotplug [ACPI] Disable memory hotplug. Useful for kdump
131 acpi_no_static_ssdt [HW,ACPI]
132 Disable installation of static SSDTs at early boot time
133 By default, SSDTs contained in the RSDT/XSDT will be
134 installed automatically and they will appear under
135 /sys/firmware/acpi/tables.
136 This option turns off this feature.
137 Note that specifying this option does not affect
138 dynamic table installation which will install SSDT
139 tables to /sys/firmware/acpi/tables/dynamic.
141 acpi_no_watchdog [HW,ACPI,WDT]
142 Ignore the ACPI-based watchdog interface (WDAT) and let
143 a native driver control the watchdog device instead.
145 acpi_rsdp= [ACPI,EFI,KEXEC]
146 Pass the RSDP address to the kernel, mostly used
147 on machines running EFI runtime service to boot the
148 second kernel for kdump.
150 acpi_os_name= [HW,ACPI] Tell ACPI BIOS the name of the OS
151 Format: To spoof as Windows 98: ="Microsoft Windows"
153 acpi_rev_override [ACPI] Override the _REV object to return 5 (instead
154 of 2 which is mandated by ACPI 6) as the supported ACPI
155 specification revision (when using this switch, it may
156 be necessary to carry out a cold reboot _twice_ in a
157 row to make it take effect on the platform firmware).
159 acpi_osi= [HW,ACPI] Modify list of supported OS interface strings
160 acpi_osi="string1" # add string1
161 acpi_osi="!string2" # remove string2
162 acpi_osi=!* # remove all strings
163 acpi_osi=! # disable all built-in OS vendor
165 acpi_osi=!! # enable all built-in OS vendor
167 acpi_osi= # disable all strings
169 'acpi_osi=!' can be used in combination with single or
170 multiple 'acpi_osi="string1"' to support specific OS
171 vendor string(s). Note that such command can only
172 affect the default state of the OS vendor strings, thus
173 it cannot affect the default state of the feature group
174 strings and the current state of the OS vendor strings,
175 specifying it multiple times through kernel command line
176 is meaningless. This command is useful when one do not
177 care about the state of the feature group strings which
178 should be controlled by the OSPM.
180 1. 'acpi_osi=! acpi_osi="Windows 2000"' is equivalent
181 to 'acpi_osi="Windows 2000" acpi_osi=!', they all
182 can make '_OSI("Windows 2000")' TRUE.
184 'acpi_osi=' cannot be used in combination with other
185 'acpi_osi=' command lines, the _OSI method will not
186 exist in the ACPI namespace. NOTE that such command can
187 only affect the _OSI support state, thus specifying it
188 multiple times through kernel command line is also
191 1. 'acpi_osi=' can make 'CondRefOf(_OSI, Local1)'
194 'acpi_osi=!*' can be used in combination with single or
195 multiple 'acpi_osi="string1"' to support specific
196 string(s). Note that such command can affect the
197 current state of both the OS vendor strings and the
198 feature group strings, thus specifying it multiple times
199 through kernel command line is meaningful. But it may
200 still not able to affect the final state of a string if
201 there are quirks related to this string. This command
202 is useful when one want to control the state of the
203 feature group strings to debug BIOS issues related to
206 1. 'acpi_osi="Module Device" acpi_osi=!*' can make
207 '_OSI("Module Device")' FALSE.
208 2. 'acpi_osi=!* acpi_osi="Module Device"' can make
209 '_OSI("Module Device")' TRUE.
210 3. 'acpi_osi=! acpi_osi=!* acpi_osi="Windows 2000"' is
212 'acpi_osi=!* acpi_osi=! acpi_osi="Windows 2000"'
214 'acpi_osi=!* acpi_osi="Windows 2000" acpi_osi=!',
215 they all will make '_OSI("Windows 2000")' TRUE.
218 Override the pmtimer bug detection: force the kernel
219 to assume that this machine's pmtimer latches its value
220 and always returns good values.
222 acpi_sci= [HW,ACPI] ACPI System Control Interrupt trigger mode
223 Format: { level | edge | high | low }
225 acpi_skip_timer_override [HW,ACPI]
226 Recognize and ignore IRQ0/pin2 Interrupt Override.
227 For broken nForce2 BIOS resulting in XT-PIC timer.
229 acpi_sleep= [HW,ACPI] Sleep options
230 Format: { s3_bios, s3_mode, s3_beep, s4_nohwsig,
231 old_ordering, nonvs, sci_force_enable, nobl }
232 See Documentation/power/video.rst for information on
234 s3_beep is for debugging; it makes the PC's speaker beep
235 as soon as the kernel's real-mode entry point is called.
236 s4_nohwsig prevents ACPI hardware signature from being
237 used during resume from hibernation.
238 old_ordering causes the ACPI 1.0 ordering of the _PTS
239 control method, with respect to putting devices into
240 low power states, to be enforced (the ACPI 2.0 ordering
241 of _PTS is used by default).
242 nonvs prevents the kernel from saving/restoring the
243 ACPI NVS memory during suspend/hibernation and resume.
244 sci_force_enable causes the kernel to set SCI_EN directly
245 on resume from S1/S3 (which is against the ACPI spec,
246 but some broken systems don't work without it).
247 nobl causes the internal blacklist of systems known to
248 behave incorrectly in some ways with respect to system
249 suspend and resume to be ignored (use wisely).
251 acpi_use_timer_override [HW,ACPI]
252 Use timer override. For some broken Nvidia NF5 boards
253 that require a timer override, but don't have HPET
255 add_efi_memmap [EFI; X86] Include EFI memory map in
256 kernel's map of available physical RAM.
259 { off | try_unsupported }
260 off: disable AGP support
261 try_unsupported: try to drive unsupported chipsets
262 (may crash computer or cause data corruption)
265 See Documentation/sound/alsa-configuration.rst
268 Allow the default userspace alignment fault handler
269 behaviour to be specified. Bit 0 enables warnings,
270 bit 1 enables fixups, and bit 2 sends a segfault.
272 align_va_addr= [X86-64]
273 Align virtual addresses by clearing slice [14:12] when
274 allocating a VMA at process creation time. This option
275 gives you up to 3% performance improvement on AMD F15h
276 machines (where it is enabled by default) for a
277 CPU-intensive style benchmark, and it can vary highly in
278 a microbenchmark depending on workload and compiler.
280 32: only for 32-bit processes
281 64: only for 64-bit processes
282 on: enable for both 32- and 64-bit processes
283 off: disable for both 32- and 64-bit processes
285 alloc_snapshot [FTRACE]
286 Allocate the ftrace snapshot buffer on boot up when the
287 main buffer is allocated. This is handy if debugging
288 and you need to use tracing_snapshot() on boot up, and
289 do not want to use tracing_snapshot_alloc() as it needs
290 to be done where GFP_KERNEL allocations are allowed.
292 amd_iommu= [HW,X86-64]
293 Pass parameters to the AMD IOMMU driver in the system.
295 fullflush - enable flushing of IO/TLB entries when
296 they are unmapped. Otherwise they are
297 flushed before they will be reused, which
299 off - do not initialize any AMD IOMMU found in
301 force_isolation - Force device isolation for all
302 devices. The IOMMU driver is not
303 allowed anymore to lift isolation
304 requirements as needed. This option
305 does not override iommu=pt
307 amd_iommu_dump= [HW,X86-64]
308 Enable AMD IOMMU driver option to dump the ACPI table
309 for AMD IOMMU. With this option enabled, AMD IOMMU
310 driver will print ACPI tables for AMD IOMMU during
311 IOMMU initialization.
313 amd_iommu_intr= [HW,X86-64]
314 Specifies one of the following AMD IOMMU interrupt
316 legacy - Use legacy interrupt remapping mode.
317 vapic - Use virtual APIC mode, which allows IOMMU
318 to inject interrupts directly into guest.
319 This mode requires kvm-amd.avic=1.
320 (Default when IOMMU HW support is present.)
322 amijoy.map= [HW,JOY] Amiga joystick support
323 Map of devices attached to JOY0DAT and JOY1DAT
325 See also Documentation/input/joydev/joystick.rst
327 analog.map= [HW,JOY] Analog joystick and gamepad support
328 Specifies type or capabilities of an analog joystick
329 connected to one of 16 gameports
330 Format: <type1>,<type2>,..<type16>
333 Power management functions (SPARCstation-4/5 + deriv.)
335 Disable APC CPU standby support. SPARCstation-Fox does
336 not play well with APC CPU idle - disable it if you have
337 APC and your system crashes randomly.
339 apic= [APIC,X86] Advanced Programmable Interrupt Controller
340 Change the output verbosity while booting
341 Format: { quiet (default) | verbose | debug }
342 Change the amount of debugging information output
343 when initialising the APIC and IO-APIC components.
344 For X86-32, this can also be used to specify an APIC
346 Format: apic=driver_name
347 Examples: apic=bigsmp
349 apic_extnmi= [APIC,X86] External NMI delivery setting
350 Format: { bsp (default) | all | none }
351 bsp: External NMI is delivered only to CPU 0
352 all: External NMIs are broadcast to all CPUs as a
354 none: External NMI is masked for all CPUs. This is
355 useful so that a dump capture kernel won't be
359 See Documentation/networking/ipv6.rst.
361 show_lapic= [APIC,X86] Advanced Programmable Interrupt Controller
362 Limit apic dumping. The parameter defines the maximal
363 number of local apics being dumped. Also it is possible
364 to set it to "all" by meaning -- no limit here.
365 Format: { 1 (default) | 2 | ... | all }.
366 The parameter valid if only apic=debug or
367 apic=verbose is specified.
368 Example: apic=debug show_lapic=all
370 apm= [APM] Advanced Power Management
371 See header of arch/x86/kernel/apm_32.c.
373 arcrimi= [HW,NET] ARCnet - "RIM I" (entirely mem-mapped) cards
374 Format: <io>,<irq>,<nodeID>
376 arm64.nobti [ARM64] Unconditionally disable Branch Target
377 Identification support
379 arm64.nopauth [ARM64] Unconditionally disable Pointer Authentication
384 atarimouse= [HW,MOUSE] Atari Mouse
386 atkbd.extra= [HW] Enable extra LEDs and keys on IBM RapidAccess,
387 EzKey and similar keyboards
389 atkbd.reset= [HW] Reset keyboard during initialization
391 atkbd.set= [HW] Select keyboard code set
392 Format: <int> (2 = AT (default), 3 = PS/2)
394 atkbd.scroll= [HW] Enable scroll wheel on MS Office and similar
397 atkbd.softraw= [HW] Choose between synthetic and real raw mode
398 Format: <bool> (0 = real, 1 = synthetic (default))
400 atkbd.softrepeat= [HW]
401 Use software keyboard repeat
403 audit= [KNL] Enable the audit sub-system
404 Format: { "0" | "1" | "off" | "on" }
405 0 | off - kernel audit is disabled and can not be
406 enabled until the next reboot
407 unset - kernel audit is initialized but disabled and
408 will be fully enabled by the userspace auditd.
409 1 | on - kernel audit is initialized and partially
410 enabled, storing at most audit_backlog_limit
411 messages in RAM until it is fully enabled by the
415 audit_backlog_limit= [KNL] Set the audit queue size limit.
416 Format: <int> (must be >=0)
419 bau= [X86_UV] Enable the BAU on SGI UV. The default
420 behavior is to disable the BAU (i.e. bau=0).
421 Format: { "0" | "1" }
424 unset - Disable the BAU.
426 baycom_epp= [HW,AX25]
429 baycom_par= [HW,AX25] BayCom Parallel Port AX.25 Modem
431 See header of drivers/net/hamradio/baycom_par.c.
433 baycom_ser_fdx= [HW,AX25]
434 BayCom Serial Port AX.25 Modem (Full Duplex Mode)
435 Format: <io>,<irq>,<mode>[,<baud>]
436 See header of drivers/net/hamradio/baycom_ser_fdx.c.
438 baycom_ser_hdx= [HW,AX25]
439 BayCom Serial Port AX.25 Modem (Half Duplex Mode)
440 Format: <io>,<irq>,<mode>
441 See header of drivers/net/hamradio/baycom_ser_hdx.c.
443 blkdevparts= Manual partition parsing of block device(s) for
444 embedded devices based on command line input.
445 See Documentation/block/cmdline-partition.rst
447 boot_delay= Milliseconds to delay each printk during boot.
448 Values larger than 10 seconds (10000) are changed to
453 Extended command line options can be added to an initrd
454 and this will cause the kernel to look for it.
456 See Documentation/admin-guide/bootconfig.rst
459 Disable BERT OS support on buggy BIOSes.
461 bgrt_disable [ACPI][X86]
462 Disable BGRT to avoid flickering OEM logo.
464 bttv.card= [HW,V4L] bttv (bt848 + bt878 based grabber cards)
465 bttv.radio= Most important insmod options are available as
467 bttv.pll= See Documentation/admin-guide/media/bttv.rst
470 bulk_remove=off [PPC] This parameter disables the use of the pSeries
471 firmware feature for flushing multiple hpte entries
474 c101= [NET] Moxa C101 synchronous serial card
476 cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection.
477 Sometimes CPU hardware bugs make them report the cache
478 size incorrectly. The kernel will attempt work arounds
479 to fix known problems, but for some CPUs it is not
480 possible to determine what the correct size should be.
481 This option provides an override for these situations.
484 [NET] Specifies amount of time (in seconds) that
485 the kernel should wait for a network carrier. By default
486 it waits 120 seconds.
488 ca_keys= [KEYS] This parameter identifies a specific key(s) on
489 the system trusted keyring to be used for certificate
491 format: { id:<keyid> | builtin }
493 cca= [MIPS] Override the kernel pages' cache coherency
494 algorithm. Accepted values range from 0 to 7
495 inclusive. See arch/mips/include/asm/pgtable-bits.h
496 for platform specific values (SB1, Loongson3 and
499 ccw_timeout_log [S390]
500 See Documentation/s390/common_io.rst for details.
502 cgroup_disable= [KNL] Disable a particular controller
503 Format: {name of the controller(s) to disable}
504 The effects of cgroup_disable=foo are:
505 - foo isn't auto-mounted if you mount all cgroups in
507 - foo isn't visible as an individually mountable
509 {Currently only "memory" controller deal with this and
510 cut the overhead, others just disable the usage. So
511 only cgroup_disable=memory is actually worthy}
513 cgroup_no_v1= [KNL] Disable cgroup controllers and named hierarchies in v1
514 Format: { { controller | "all" | "named" }
515 [,{ controller | "all" | "named" }...] }
516 Like cgroup_disable, but only applies to cgroup v1;
517 the blacklisted controllers remain available in cgroup2.
518 "all" blacklists all controllers and "named" disables
519 named mounts. Specifying both "all" and "named" disables
522 cgroup.memory= [KNL] Pass options to the cgroup memory controller.
524 nosocket -- Disable socket memory accounting.
525 nokmem -- Disable kernel memory accounting.
527 checkreqprot [SELINUX] Set initial checkreqprot flag value.
528 Format: { "0" | "1" }
529 See security/selinux/Kconfig help text.
530 0 -- check protection applied by kernel (includes
531 any implied execute protection).
532 1 -- check protection requested by application.
533 Default value is set via a kernel config option.
534 Value can be changed at runtime via
535 /sys/fs/selinux/checkreqprot.
536 Setting checkreqprot to 1 is deprecated.
539 See Documentation/s390/common_io.rst for details.
542 Prevents the clock framework from automatically gating
543 clocks that have not been explicitly enabled by a Linux
544 device driver but are enabled in hardware at reset or
545 by the bootloader/firmware. Note that this does not
546 force such clocks to be always-on nor does it reserve
547 those clocks in any way. This parameter is useful for
548 debug and development, but should not be needed on a
549 platform with proper driver support. For more
550 information, see Documentation/driver-api/clk.rst.
552 clock= [BUGS=X86-32, HW] gettimeofday clocksource override.
554 Forces specified clocksource (if available) to be used
555 when calculating gettimeofday(). If specified
556 clocksource is not available, it defaults to PIT.
557 Format: { pit | tsc | cyclone | pmtmr }
559 clocksource= Override the default clocksource
561 Override the default clocksource and use the clocksource
562 with the name specified.
563 Some clocksource names to choose from, depending on
565 [all] jiffies (this is the base, fallback clocksource)
567 [ARM] imx_timer1,OSTS,netx_timer,mpu_timer2,
568 pxa_timer,timer3,32k_counter,timer0_1
569 [X86-32] pit,hpet,tsc;
570 scx200_hrt on Geode; cyclone on IBM x440
578 clocksource.arm_arch_timer.evtstrm=
581 Enable/disable the eventstream feature of the ARM
582 architected timer so that code using WFE-based polling
583 loops can be debugged more effectively on production
586 clearcpuid=BITNUM[,BITNUM...] [X86]
587 Disable CPUID feature X for the kernel. See
588 arch/x86/include/asm/cpufeatures.h for the valid bit
589 numbers. Note the Linux specific bits are not necessarily
590 stable over kernel options, but the vendor specific
592 Also note that user programs calling CPUID directly
593 or using the feature without checking anything
594 will still see it. This just prevents it from
595 being used by the kernel or shown in /proc/cpuinfo.
596 Also note the kernel might malfunction if you disable
599 cma=nn[MG]@[start[MG][-end[MG]]]
601 Sets the size of kernel global memory area for
602 contiguous memory allocations and optionally the
603 placement constraint by the physical address range of
604 memory allocations. A value of 0 disables CMA
605 altogether. For more information, see
606 kernel/dma/contiguous.c
610 Sets the size of kernel per-numa memory area for
611 contiguous memory allocations. A value of 0 disables
612 per-numa CMA altogether. And If this option is not
613 specificed, the default value is 0.
614 With per-numa CMA enabled, DMA users on node nid will
615 first try to allocate buffer from the pernuma area
616 which is located in node nid, if the allocation fails,
617 they will fallback to the global default memory area.
619 cmo_free_hint= [PPC] Format: { yes | no }
620 Specify whether pages are marked as being inactive
621 when they are freed. This is used in CMO environments
622 to determine OS memory pressure for page stealing by
626 coherent_pool=nn[KMG] [ARM,KNL]
627 Sets the size of memory pool for coherent, atomic dma
628 allocations, by default set to 256K.
630 com20020= [HW,NET] ARCnet - COM20020 chipset
632 <io>[,<irq>[,<nodeID>[,<backplane>[,<ckp>[,<timeout>]]]]]
634 com90io= [HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers)
638 ARCnet - COM90xx chipset (memory-mapped buffers)
639 Format: <io>[,<irq>[,<memstart>]]
641 condev= [HW,S390] console device
644 console= [KNL] Output console device and options.
646 tty<n> Use the virtual console device <n>.
650 Use the specified serial port. The options are of
651 the form "bbbbpnf", where "bbbb" is the baud rate,
652 "p" is parity ("n", "o", or "e"), "n" is number of
653 bits, and "f" is flow control ("r" for RTS or
654 omit it). Default is "9600n8".
656 See Documentation/admin-guide/serial-console.rst for more
658 Documentation/networking/netconsole.rst for an
661 uart[8250],io,<addr>[,options]
662 uart[8250],mmio,<addr>[,options]
663 uart[8250],mmio16,<addr>[,options]
664 uart[8250],mmio32,<addr>[,options]
665 uart[8250],0x<addr>[,options]
666 Start an early, polled-mode console on the 8250/16550
667 UART at the specified I/O port or MMIO address,
668 switching to the matching ttyS device later.
669 MMIO inter-register address stride is either 8-bit
670 (mmio), 16-bit (mmio16), or 32-bit (mmio32).
671 If none of [io|mmio|mmio16|mmio32], <addr> is assumed
672 to be equivalent to 'mmio'. 'options' are specified in
673 the same format described for ttyS above; if unspecified,
674 the h/w is not re-initialized.
676 hvc<n> Use the hypervisor console device <n>. This is for
677 both Xen and PowerPC hypervisors.
679 If the device connected to the port is not a TTY but a braille
680 device, prepend "brl," before the device type, for instance
682 For now, only VisioBraille is supported.
685 [KNL] Change console messages format
687 By default we print messages on consoles in
688 "[time stamp] text\n" format (time stamp may not be
689 printed, depending on CONFIG_PRINTK_TIME or
690 `printk_time' param).
692 Switch to syslog format: "<%u>[time stamp] text\n"
693 IOW, each message will have a facility and loglevel
694 prefix. The format is similar to one used by syslog()
695 syscall, or to executing "dmesg -S --raw" or to reading
698 consoleblank= [KNL] The console blank (screen saver) timeout in
699 seconds. A value of 0 disables the blank timer.
703 [KNL] Change the default value for
704 /proc/<pid>/coredump_filter.
705 See also Documentation/filesystems/proc.rst.
707 coresight_cpu_debug.enable
710 Enable/disable the CPU sampling based debugging.
711 0: default value, disable debugging
712 1: enable debugging at boot time
714 cpuidle.off=1 [CPU_IDLE]
715 disable the cpuidle sub-system
718 [CPU_IDLE] Name of the cpuidle governor to use.
720 cpufreq.off=1 [CPU_FREQ]
721 disable the cpufreq sub-system
723 cpufreq.default_governor=
724 [CPU_FREQ] Name of the default cpufreq governor or
725 policy to use. This governor must be registered in the
726 kernel before the cpufreq driver probes.
729 [X86] Delay for N microsec between assert and de-assert
730 of APIC INIT to start processors. This delay occurs
731 on every CPU online, such as boot, and resume from suspend.
734 cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver
736 <first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>]
738 crashkernel=size[KMG][@offset[KMG]]
739 [KNL] Using kexec, Linux can switch to a 'crash kernel'
740 upon panic. This parameter reserves the physical
741 memory region [offset, offset + size] for that kernel
742 image. If '@offset' is omitted, then a suitable offset
743 is selected automatically.
744 [KNL, X86-64] Select a region under 4G first, and
745 fall back to reserve region above 4G when '@offset'
746 hasn't been specified.
747 See Documentation/admin-guide/kdump/kdump.rst for further details.
749 crashkernel=range1:size1[,range2:size2,...][@offset]
750 [KNL] Same as above, but depends on the memory
751 in the running system. The syntax of range is
752 start-[end] where start and end are both
753 a memory unit (amount[KMG]). See also
754 Documentation/admin-guide/kdump/kdump.rst for an example.
756 crashkernel=size[KMG],high
757 [KNL, X86-64] range could be above 4G. Allow kernel
758 to allocate physical memory region from top, so could
759 be above 4G if system have more than 4G ram installed.
760 Otherwise memory region will be allocated below 4G, if
762 It will be ignored if crashkernel=X is specified.
763 crashkernel=size[KMG],low
764 [KNL, X86-64] range under 4G. When crashkernel=X,high
765 is passed, kernel could allocate physical memory region
766 above 4G, that cause second kernel crash on system
767 that require some amount of low memory, e.g. swiotlb
768 requires at least 64M+32K low memory, also enough extra
769 low memory is needed to make sure DMA buffers for 32-bit
770 devices won't run out. Kernel would try to allocate at
771 at least 256M below 4G automatically.
772 This one let user to specify own low range under 4G
773 for second kernel instead.
774 0: to disable low allocation.
775 It will be ignored when crashkernel=X,high is not used
776 or memory reserved is below 4G.
779 [KNL] Disable crypto self-tests
784 cs89x0_media= [HW,NET]
785 Format: { rj45 | aui | bnc }
788 See header of drivers/s390/block/dasd_devmap.c.
790 db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port
791 (one device per port)
792 Format: <port#>,<type>
793 See also Documentation/input/devices/joystick-parport.rst
795 ddebug_query= [KNL,DYNAMIC_DEBUG] Enable debug messages at early boot
797 Documentation/admin-guide/dynamic-debug-howto.rst for
798 details. Deprecated, see dyndbg.
800 debug [KNL] Enable kernel debugging (events log level).
803 [KNL] Enable printing [hashed] pointers early in the
804 boot sequence. If enabled, we use a weak hash instead
805 of siphash to hash pointers. Use this option if you are
806 seeing instances of '(___ptrval___)') and need to see a
807 value (hashed pointer) instead. Cryptographically
808 insecure, please do not use on production kernels.
811 [KNL] verbose locking self-tests
813 Print debugging info while doing the locking API
815 Bitmask for the various LOCKTYPE_ tests. Defaults to 0
816 (no extra messages), setting it to -1 (all bits set)
817 will print _a_lot_ more information - normally only
818 useful to lockdep developers.
820 debug_objects [KNL] Enable object debugging
823 [KNL] Disable object debugging
825 debug_guardpage_minorder=
826 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this
827 parameter allows control of the order of pages that will
828 be intentionally kept free (and hence protected) by the
829 buddy allocator. Bigger value increase the probability
830 of catching random memory corruption, but reduce the
831 amount of memory for normal system use. The maximum
832 possible value is MAX_ORDER/2. Setting this parameter
833 to 1 or 2 should be enough to identify most random
834 memory corruption problems caused by bugs in kernel or
835 driver code when a CPU writes to (or reads from) a
836 random memory location. Note that there exists a class
837 of memory corruptions problems caused by buggy H/W or
838 F/W or by drivers badly programing DMA (basically when
839 memory is written at bus level and the CPU MMU is
840 bypassed) which are not detectable by
841 CONFIG_DEBUG_PAGEALLOC, hence this option will not help
842 tracking down these problems.
845 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this parameter
846 enables the feature at boot time. By default, it is
847 disabled and the system will work mostly the same as a
848 kernel built without CONFIG_DEBUG_PAGEALLOC.
849 Note: to get most of debug_pagealloc error reports, it's
850 useful to also enable the page_owner functionality.
851 on: enable the feature
853 debugfs= [KNL] This parameter enables what is exposed to userspace
854 and debugfs internal clients.
855 Format: { on, no-mount, off }
856 on: All functions are enabled.
858 Filesystem is not registered but kernel clients can
859 access APIs and a crashkernel can be used to read
860 its content. There is nothing to mount.
861 off: Filesystem is not registered and clients
862 get a -EPERM as result when trying to register files
863 or directories within debugfs.
864 This is equivalent of the runtime functionality if
865 debugfs was not enabled in the kernel at all.
866 Default value is set in build-time with a kernel configuration.
868 debugpat [X86] Enable PAT debugging
870 decnet.addr= [HW,NET]
871 Format: <area>[,<node>]
872 See also Documentation/networking/decnet.rst.
875 [HW] The size of the default HugeTLB page. This is
876 the size represented by the legacy /proc/ hugepages
877 APIs. In addition, this is the default hugetlb size
878 used for shmget(), mmap() and mounting hugetlbfs
879 filesystems. If not specified, defaults to the
880 architecture's default huge page size. Huge page
881 sizes are architecture dependent. See also
882 Documentation/admin-guide/mm/hugetlbpage.rst.
885 deferred_probe_timeout=
886 [KNL] Debugging option to set a timeout in seconds for
887 deferred probe to give up waiting on dependencies to
888 probe. Only specific dependencies (subsystems or
889 drivers) that have opted in will be ignored. A timeout of 0
890 will timeout at the end of initcalls. This option will also
891 dump out devices still on the deferred probe list after
895 Format: { on | off | def_only | inf_only | always }
896 on: s390 zlib hardware support for compression on
897 level 1 and decompression (default)
898 off: No s390 zlib hardware support
899 def_only: s390 zlib hardware support for deflate
900 only (compression on level 1)
901 inf_only: s390 zlib hardware support for inflate
903 always: Same as 'on' but ignores the selected compression
904 level always using hardware support (used for debugging)
907 Set number of hash buckets for dentry cache.
909 disable_1tb_segments [PPC]
910 Disables the use of 1TB hash page table segments. This
911 causes the kernel to fall back to 256MB segments which
912 can be useful when debugging issues that require an SLB
916 Limits the number of kernel SLB entries, and flushes
917 them frequently to increase the rate of SLB faults
921 See Documentation/networking/ipv6.rst.
924 [KNL] Under CONFIG_HARDENED_USERCOPY, whether
925 hardening is enabled for this boot. Hardened
926 usercopy checking is used to protect the kernel
927 from reading or writing beyond known memory
928 allocation boundaries as a proactive defense
929 against bounds-checking flaws in the kernel's
930 copy_to_user()/copy_from_user() interface.
931 on Perform hardened usercopy checks (default).
932 off Disable hardened usercopy checks.
935 Disable RADIX MMU mode on POWER9
937 radix_hcall_invalidate=on [PPC/PSERIES]
938 Disable RADIX GTSE feature and use hcall for TLB
942 Disable TLBIE instruction. Currently does not work
943 with KVM, with HASH MMU, or with coherent accelerators.
945 disable_cpu_apicid= [X86,APIC,SMP]
947 The number of initial APIC ID for the
948 corresponding CPU to be disabled at boot,
949 mostly used for the kdump 2nd kernel to
950 disable BSP to wake up multiple CPUs without
951 causing system reset or hang due to sending
954 disable_ddw [PPC/PSERIES]
955 Disable Dynamic DMA Window support. Use this
956 to workaround buggy firmware.
959 See Documentation/networking/ipv6.rst.
961 disable_mtrr_cleanup [X86]
962 The kernel tries to adjust MTRR layout from continuous
963 to discrete, to make X server driver able to add WB
964 entry later. This parameter disables that.
966 disable_mtrr_trim [X86, Intel and AMD only]
967 By default the kernel will trim any uncacheable
968 memory out of your available memory pool based on
969 MTRR settings. This parameter disables that behavior,
970 possibly causing your machine to run very slowly.
972 disable_timer_pin_1 [X86]
973 Disable PIN 1 of APIC timer
974 Can be useful to work around chipset bugs.
976 dis_ucode_ldr [X86] Disable the microcode loader.
978 dma_debug=off If the kernel is compiled with DMA_API_DEBUG support,
979 this option disables the debugging code at boot.
981 dma_debug_entries=<number>
982 This option allows to tune the number of preallocated
983 entries for DMA-API debugging code. One entry is
984 required per DMA-API allocation. Use this if the
985 DMA-API debugging code disables itself because the
986 architectural default is too low.
988 dma_debug_driver=<driver_name>
989 With this option the DMA-API debugging driver
990 filter feature can be enabled at boot time. Just
991 pass the driver to filter for as the parameter.
992 The filter can be disabled or changed to another
993 driver later using sysfs.
995 driver_async_probe= [KNL]
996 List of driver names to be probed asynchronously.
997 Format: <driver_name1>,<driver_name2>...
999 drm.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>]
1000 Broken monitors, graphic adapters, KVMs and EDIDless
1001 panels may send no or incorrect EDID data sets.
1002 This parameter allows to specify an EDID data sets
1003 in the /lib/firmware directory that are used instead.
1004 Generic built-in EDID data sets are used, if one of
1005 edid/1024x768.bin, edid/1280x1024.bin,
1006 edid/1680x1050.bin, or edid/1920x1080.bin is given
1007 and no file with the same name exists. Details and
1008 instructions how to build your own EDID data are
1009 available in Documentation/admin-guide/edid.rst. An EDID
1010 data set will only be used for a particular connector,
1011 if its name and a colon are prepended to the EDID
1012 name. Each connector may use a unique EDID data
1013 set by separating the files with a comma. An EDID
1014 data set with no connector name will be used for
1015 any connectors not explicitly specified.
1020 Format: {"off" | "known"}
1021 Control how the dt_cpu_ftrs device-tree binding is
1022 used for CPU feature discovery and setup (if it
1024 off: Do not use it, fall back to legacy cpu table.
1025 known: Do not pass through unknown features to guests
1026 or userspace, only those that the kernel is aware of.
1028 dump_apple_properties [X86]
1029 Dump name and content of EFI device properties on
1030 x86 Macs. Useful for driver authors to determine
1031 what data is available or for reverse-engineering.
1033 dyndbg[="val"] [KNL,DYNAMIC_DEBUG]
1034 <module>.dyndbg[="val"]
1035 Enable debug messages at boot time. See
1036 Documentation/admin-guide/dynamic-debug-howto.rst
1039 nopku [X86] Disable Memory Protection Keys CPU feature found
1042 <module>.async_probe [KNL]
1043 Enable asynchronous probe on this module.
1045 early_ioremap_debug [KNL]
1046 Enable debug messages in early_ioremap support. This
1047 is useful for tracking down temporary early mappings
1048 which are not unmapped.
1050 earlycon= [KNL] Output early console device and options.
1052 When used with no options, the early console is
1053 determined by stdout-path property in device tree's
1054 chosen node or the ACPI SPCR table if supported by
1057 cdns,<addr>[,options]
1058 Start an early, polled-mode console on a Cadence
1059 (xuartps) serial port at the specified address. Only
1060 supported option is baud rate. If baud rate is not
1061 specified, the serial port must already be setup and
1064 uart[8250],io,<addr>[,options]
1065 uart[8250],mmio,<addr>[,options]
1066 uart[8250],mmio32,<addr>[,options]
1067 uart[8250],mmio32be,<addr>[,options]
1068 uart[8250],0x<addr>[,options]
1069 Start an early, polled-mode console on the 8250/16550
1070 UART at the specified I/O port or MMIO address.
1071 MMIO inter-register address stride is either 8-bit
1072 (mmio) or 32-bit (mmio32 or mmio32be).
1073 If none of [io|mmio|mmio32|mmio32be], <addr> is assumed
1074 to be equivalent to 'mmio'. 'options' are specified
1075 in the same format described for "console=ttyS<n>"; if
1076 unspecified, the h/w is not initialized.
1080 Start an early, polled-mode console on a pl011 serial
1081 port at the specified address. The pl011 serial port
1082 must already be setup and configured. Options are not
1083 yet supported. If 'mmio32' is specified, then only
1084 the driver will use only 32-bit accessors to read/write
1085 the device registers.
1088 Start an early, polled-mode console on a meson serial
1089 port at the specified address. The serial port must
1090 already be setup and configured. Options are not yet
1094 Start an early, polled-mode console on an msm serial
1095 port at the specified address. The serial port
1096 must already be setup and configured. Options are not
1099 msm_serial_dm,<addr>
1100 Start an early, polled-mode console on an msm serial
1101 dm port at the specified address. The serial port
1102 must already be setup and configured. Options are not
1106 Start an early, polled-mode console on a serial port
1107 of an Actions Semi SoC, such as S500 or S900, at the
1108 specified address. The serial port must already be
1109 setup and configured. Options are not yet supported.
1112 Start an early, polled-mode console on a serial port
1113 of an RDA Micro SoC, such as RDA8810PL, at the
1114 specified address. The serial port must already be
1115 setup and configured. Options are not yet supported.
1118 Use RISC-V SBI (Supervisor Binary Interface) for early
1121 smh Use ARM semihosting calls for early console.
1129 Use early console provided by serial driver available
1130 on Samsung SoCs, requires selecting proper type and
1131 a correct base address of the selected UART port. The
1132 serial port must already be setup and configured.
1133 Options are not yet supported.
1136 Start an early, polled-mode console on a lantiq serial
1137 (lqasc) port at the specified address. The serial port
1138 must already be setup and configured. Options are not
1143 Use early console provided by Freescale LP UART driver
1144 found on Freescale Vybrid and QorIQ LS1021A processors.
1145 A valid base address must be provided, and the serial
1146 port must already be setup and configured.
1150 Start an early, polled-mode, output-only console on the
1151 Freescale i.MX UART at the specified address. The UART
1152 must already be setup and configured.
1155 Start an early, polled-mode console on the
1156 Armada 3700 serial port at the specified
1157 address. The serial port must already be setup
1158 and configured. Options are not yet supported.
1161 Start an early, polled-mode console on a Qualcomm
1162 Generic Interface (GENI) based serial port at the
1163 specified address. The serial port must already be
1164 setup and configured. Options are not yet supported.
1167 Start an early, unaccelerated console on the EFI
1168 memory mapped framebuffer (if available). On cache
1169 coherent non-x86 systems that use system memory for
1170 the framebuffer, pass the 'ram' option so that it is
1171 mapped with the correct attributes.
1174 Use early console provided by Freescale LINFlexD UART
1175 serial driver for NXP S32V234 SoCs. A valid base
1176 address must be provided, and the serial port must
1177 already be setup and configured.
1179 earlyprintk= [X86,SH,ARM,M68k,S390]
1183 earlyprintk=serial[,ttySn[,baudrate]]
1184 earlyprintk=serial[,0x...[,baudrate]]
1185 earlyprintk=ttySn[,baudrate]
1186 earlyprintk=dbgp[debugController#]
1187 earlyprintk=pciserial[,force],bus:device.function[,baudrate]
1188 earlyprintk=xdbc[xhciController#]
1190 earlyprintk is useful when the kernel crashes before
1191 the normal console is initialized. It is not enabled by
1192 default because it has some cosmetic problems.
1194 Append ",keep" to not disable it when the real console
1197 Only one of vga, efi, serial, or usb debug port can
1200 Currently only ttyS0 and ttyS1 may be specified by
1201 name. Other I/O ports may be explicitly specified
1202 on some architectures (x86 and arm at least) by
1203 replacing ttySn with an I/O port address, like this:
1204 earlyprintk=serial,0x1008,115200
1205 You can find the port for a given device in
1206 /proc/tty/driver/serial:
1207 2: uart:ST16650V2 port:00001008 irq:18 ...
1209 Interaction with the standard serial driver is not
1212 The VGA and EFI output is eventually overwritten by
1215 The xen output can only be used by Xen PV guests.
1217 The sclp output can only be used on s390.
1219 The optional "force" to "pciserial" enables use of a
1220 PCI device even when its classcode is not of the
1223 edac_report= [HW,EDAC] Control how to report EDAC event
1224 Format: {"on" | "off" | "force"}
1225 on: enable EDAC to report H/W event. May be overridden
1226 by other higher priority error reporting module.
1227 off: disable H/W event reporting through EDAC.
1228 force: enforce the use of EDAC to report H/W event.
1231 ekgdboc= [X86,KGDB] Allow early kernel console debugging
1234 This is designed to be used in conjunction with
1235 the boot argument: earlyprintk=vga
1237 This parameter works in place of the kgdboc parameter
1238 but can only be used if the backing tty is available
1239 very early in the boot process. For early debugging
1240 via a serial port see kgdboc_earlycon instead.
1243 Format: {"off" | "on" | "skip[mbr]"}
1246 Format: { "debug", "disable_early_pci_dma",
1247 "nochunk", "noruntime", "nosoftreserve",
1248 "novamap", "no_disable_early_pci_dma" }
1249 debug: enable misc debug output.
1250 disable_early_pci_dma: disable the busmaster bit on all
1251 PCI bridges while in the EFI boot stub.
1252 nochunk: disable reading files in "chunks" in the EFI
1253 boot stub, as chunking can cause problems with some
1254 firmware implementations.
1255 noruntime : disable EFI runtime services support
1256 nosoftreserve: The EFI_MEMORY_SP (Specific Purpose)
1257 attribute may cause the kernel to reserve the
1258 memory range for a memory mapping driver to
1259 claim. Specify efi=nosoftreserve to disable this
1260 reservation and treat the memory by its base type
1261 (i.e. EFI_CONVENTIONAL_MEMORY / "System RAM").
1262 novamap: do not call SetVirtualAddressMap().
1263 no_disable_early_pci_dma: Leave the busmaster bit set
1264 on all PCI bridges while in the EFI boot stub
1266 efi_no_storage_paranoia [EFI; X86]
1267 Using this parameter you can use more than 50% of
1268 your efi variable storage. Use this parameter only if
1269 you are really sure that your UEFI does sane gc and
1270 fulfills the spec otherwise your board may brick.
1272 efi_fake_mem= nn[KMG]@ss[KMG]:aa[,nn[KMG]@ss[KMG]:aa,..] [EFI; X86]
1273 Add arbitrary attribute to specific memory range by
1274 updating original EFI memory map.
1275 Region of memory which aa attribute is added to is
1278 If efi_fake_mem=2G@4G:0x10000,2G@0x10a0000000:0x10000
1279 is specified, EFI_MEMORY_MORE_RELIABLE(0x10000)
1280 attribute is added to range 0x100000000-0x180000000 and
1281 0x10a0000000-0x1120000000.
1283 If efi_fake_mem=8G@9G:0x40000 is specified, the
1284 EFI_MEMORY_SP(0x40000) attribute is added to
1285 range 0x240000000-0x43fffffff.
1287 Using this parameter you can do debugging of EFI memmap
1288 related features. For example, you can do debugging of
1289 Address Range Mirroring feature even if your box
1290 doesn't support it, or mark specific memory as
1293 efivar_ssdt= [EFI; X86] Name of an EFI variable that contains an SSDT
1294 that is to be dynamically loaded by Linux. If there are
1295 multiple variables with the same name but with different
1296 vendor GUIDs, all of them will be loaded. See
1297 Documentation/admin-guide/acpi/ssdt-overlays.rst for details.
1300 eisa_irq_edge= [PARISC,HW]
1301 See header of drivers/parisc/eisa.c.
1304 See comment before function elanfreq_setup() in
1305 arch/x86/kernel/cpu/cpufreq/elanfreq.c.
1307 elfcorehdr=[size[KMG]@]offset[KMG] [IA64,PPC,SH,X86,S390]
1308 Specifies physical address of start of kernel core
1309 image elf header and optionally the size. Generally
1310 kexec loader will pass this option to capture kernel.
1311 See Documentation/admin-guide/kdump/kdump.rst for details.
1313 enable_mtrr_cleanup [X86]
1314 The kernel tries to adjust MTRR layout from continuous
1315 to discrete, to make X server driver able to add WB
1316 entry later. This parameter enables that.
1318 enable_timer_pin_1 [X86]
1319 Enable PIN 1 of APIC timer
1320 Can be useful to work around chipset bugs
1321 (in particular on some ATI chipsets).
1322 The kernel tries to set a reasonable default.
1324 enforcing [SELINUX] Set initial enforcing status.
1326 See security/selinux/Kconfig help text.
1327 0 -- permissive (log only, no denials).
1328 1 -- enforcing (deny and log).
1330 Value can be changed at runtime via
1331 /sys/fs/selinux/enforce.
1334 Disable Error Record Serialization Table (ERST)
1337 ether= [HW,NET] Ethernet cards parameters
1338 This option is obsoleted by the "netdev=" option, which
1339 has equivalent usage. See its documentation for details.
1343 Permit 'security.evm' to be updated regardless of
1344 current integrity status.
1349 fail_make_request=[KNL]
1350 General fault injection mechanism.
1351 Format: <interval>,<probability>,<space>,<times>
1352 See also Documentation/fault-injection/.
1355 Format: { initns | none }
1356 See Documentation/admin-guide/sysctl/net.rst for
1357 fb_tunnels_only_for_init_ns
1360 See Documentation/admin-guide/blockdev/floppy.rst.
1362 force_pal_cache_flush
1363 [IA-64] Avoid check_sal_cache_flush which may hang on
1364 buggy SAL_CACHE_FLUSH implementations. Using this
1365 parameter will force ia64_sal_cache_flush to call
1366 ia64_pal_cache_flush instead of SAL_CACHE_FLUSH.
1369 Forcefully enable Physical Address Extension (PAE).
1370 Many Pentium M systems disable PAE but may have a
1371 functionally usable PAE implementation.
1372 Warning: use of this parameter will taint the kernel
1373 and may cause unknown problems.
1376 [FTRACE] will set and start the specified tracer
1377 as early as possible in order to facilitate early
1380 ftrace_dump_on_oops[=orig_cpu]
1381 [FTRACE] will dump the trace buffers on oops.
1382 If no parameter is passed, ftrace will dump
1383 buffers of all CPUs, but if you pass orig_cpu, it will
1384 dump only the buffer of the CPU that triggered the
1387 ftrace_filter=[function-list]
1388 [FTRACE] Limit the functions traced by the function
1389 tracer at boot up. function-list is a comma-separated
1390 list of functions. This list can be changed at run
1391 time by the set_ftrace_filter file in the debugfs
1394 ftrace_notrace=[function-list]
1395 [FTRACE] Do not trace the functions specified in
1396 function-list. This list can be changed at run time
1397 by the set_ftrace_notrace file in the debugfs
1400 ftrace_graph_filter=[function-list]
1401 [FTRACE] Limit the top level callers functions traced
1402 by the function graph tracer at boot up.
1403 function-list is a comma-separated list of functions
1404 that can be changed at run time by the
1405 set_graph_function file in the debugfs tracing directory.
1407 ftrace_graph_notrace=[function-list]
1408 [FTRACE] Do not trace from the functions specified in
1409 function-list. This list is a comma-separated list of
1410 functions that can be changed at run time by the
1411 set_graph_notrace file in the debugfs tracing directory.
1413 ftrace_graph_max_depth=<uint>
1414 [FTRACE] Used with the function graph tracer. This is
1415 the max depth it will trace into a function. This value
1416 can be changed at run time by the max_graph_depth file
1417 in the tracefs tracing directory. default: 0 (no limit)
1419 fw_devlink= [KNL] Create device links between consumer and supplier
1420 devices by scanning the firmware to infer the
1421 consumer/supplier relationships. This feature is
1422 especially useful when drivers are loaded as modules as
1423 it ensures proper ordering of tasks like device probing
1424 (suppliers first, then consumers), supplier boot state
1425 clean up (only after all consumers have probed),
1426 suspend/resume & runtime PM (consumers first, then
1428 Format: { off | permissive | on | rpm }
1429 off -- Don't create device links from firmware info.
1430 permissive -- Create device links from firmware info
1431 but use it only for ordering boot state clean
1432 up (sync_state() calls).
1433 on -- Create device links from firmware info and use it
1434 to enforce probe and suspend/resume ordering.
1435 rpm -- Like "on", but also use to order runtime PM.
1438 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad
1439 support via parallel port (up to 5 devices per port)
1440 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5>
1441 See also Documentation/input/devices/joystick-parport.rst
1445 gart_fix_e820= [X86-64] disable the fix e820 for K8 GART
1449 gcov_persist= [GCOV] When non-zero (default), profiling data for
1450 kernel modules is saved and remains accessible via
1451 debugfs, even when the module is unloaded/reloaded.
1452 When zero, profiling data is discarded and associated
1453 debugfs files are removed at module unload time.
1455 goldfish [X86] Enable the goldfish android emulator platform.
1456 Don't use this when you are not running on the
1459 gpt [EFI] Forces disk with valid GPT signature but
1460 invalid Protective MBR to be treated as GPT. If the
1461 primary GPT is corrupted, it enables the backup/alternate
1462 GPT to be used instead.
1464 grcan.enable0= [HW] Configuration of physical interface 0. Determines
1465 the "Enable 0" bit of the configuration register.
1468 grcan.enable1= [HW] Configuration of physical interface 1. Determines
1469 the "Enable 0" bit of the configuration register.
1472 grcan.select= [HW] Select which physical interface to use.
1475 grcan.txsize= [HW] Sets the size of the tx buffer.
1476 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0.
1478 grcan.rxsize= [HW] Sets the size of the rx buffer.
1479 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0.
1482 gpio-mockup.gpio_mockup_ranges
1483 [HW] Sets the ranges of gpiochip of for this device.
1484 Format: <start1>,<end1>,<start2>,<end2>...
1486 hardlockup_all_cpu_backtrace=
1487 [KNL] Should the hard-lockup detector generate
1488 backtraces on all cpus.
1491 hashdist= [KNL,NUMA] Large hashes allocated during boot
1492 are distributed across NUMA nodes. Defaults on
1493 for 64-bit NUMA, off otherwise.
1494 Format: 0 | 1 (for off | on)
1496 hcl= [IA-64] SGI's Hardware Graph compatibility layer
1498 hd= [EIDE] (E)IDE hard drive subsystem geometry
1499 Format: <cyl>,<head>,<sect>
1502 Disable Hardware Error Source Table (HEST) support;
1503 corresponding firmware-first mode error processing
1504 logic will be disabled.
1506 highmem=nn[KMG] [KNL,BOOT] forces the highmem zone to have an exact
1507 size of <nn>. This works even on boxes that have no
1508 highmem otherwise. This also works to reduce highmem
1509 size on bigger boxes.
1511 highres= [KNL] Enable/disable high resolution timer mode.
1512 Valid parameters: "on", "off"
1517 hpet= [X86-32,HPET] option to control HPET usage
1518 Format: { enable (default) | disable | force |
1520 disable: disable HPET and use PIT instead
1521 force: allow force enabled of undocumented chips (ICH4,
1523 verbose: show contents of HPET registers during setup
1525 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET
1526 registers. Default set by CONFIG_HPET_MMAP_DEFAULT.
1528 hugetlb_cma= [HW,CMA] The size of a CMA area used for allocation
1529 of gigantic hugepages.
1532 Reserve a CMA area of given size and allocate gigantic
1533 hugepages using the CMA allocator. If enabled, the
1534 boot-time allocation of gigantic hugepages is skipped.
1536 hugepages= [HW] Number of HugeTLB pages to allocate at boot.
1537 If this follows hugepagesz (below), it specifies
1538 the number of pages of hugepagesz to be allocated.
1539 If this is the first HugeTLB parameter on the command
1540 line, it specifies the number of pages to allocate for
1541 the default huge page size. See also
1542 Documentation/admin-guide/mm/hugetlbpage.rst.
1546 [HW] The size of the HugeTLB pages. This is used in
1547 conjunction with hugepages (above) to allocate huge
1548 pages of a specific size at boot. The pair
1549 hugepagesz=X hugepages=Y can be specified once for
1550 each supported huge page size. Huge page sizes are
1551 architecture dependent. See also
1552 Documentation/admin-guide/mm/hugetlbpage.rst.
1556 [KNL] Should the hung task detector generate panics.
1559 A value of 1 instructs the kernel to panic when a
1560 hung task is detected. The default value is controlled
1561 by the CONFIG_BOOTPARAM_HUNG_TASK_PANIC build-time
1562 option. The value selected by this boot parameter can
1563 be changed later by the kernel.hung_task_panic sysctl.
1565 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC)
1566 terminal devices. Valid values: 0..8
1567 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs.
1568 If specified, z/VM IUCV HVC accepts connections
1569 from listed z/VM user IDs only.
1571 hv_nopvspin [X86,HYPER_V] Disables the paravirt spinlock optimizations
1572 which allow the hypervisor to 'idle' the
1573 guest on lock contention.
1576 Do not unregister boot console at start. This is only
1577 useful for debugging when something happens in the window
1578 between unregistering the boot console and initializing
1581 i2c_bus= [HW] Override the default board specific I2C bus speed
1582 or register an additional I2C bus that is not
1583 registered from board initialization code.
1587 i8042.debug [HW] Toggle i8042 debug mode
1588 i8042.unmask_kbd_data
1589 [HW] Enable printing of interrupt data from the KBD port
1590 (disabled by default, and as a pre-condition
1591 requires that i8042.debug=1 be enabled)
1592 i8042.direct [HW] Put keyboard port into non-translated mode
1593 i8042.dumbkbd [HW] Pretend that controller can only read data from
1594 keyboard and cannot control its state
1595 (Don't attempt to blink the leds)
1596 i8042.noaux [HW] Don't check for auxiliary (== mouse) port
1597 i8042.nokbd [HW] Don't check/create keyboard port
1598 i8042.noloop [HW] Disable the AUX Loopback command while probing
1600 i8042.nomux [HW] Don't check presence of an active multiplexing
1602 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX
1604 i8042.notimeout [HW] Ignore timeout condition signalled by controller
1605 i8042.reset [HW] Reset the controller during init, cleanup and
1606 suspend-to-ram transitions, only during s2r
1607 transitions, or never reset
1608 Format: { 1 | Y | y | 0 | N | n }
1609 1, Y, y: always reset controller
1610 0, N, n: don't ever reset controller
1611 Default: only on s2r transitions on x86; most other
1612 architectures force reset to be always executed
1613 i8042.unlock [HW] Unlock (ignore) the keylock
1614 i8042.kbdreset [HW] Reset device connected to KBD port
1618 i8k.ignore_dmi [HW] Continue probing hardware even if DMI data
1619 indicates that the driver is running on unsupported
1621 i8k.force [HW] Activate i8k driver even if SMM BIOS signature
1622 does not match list of supported models.
1624 [HW] Report power status in /proc/i8k
1625 (disabled by default)
1626 i8k.restricted [HW] Allow controlling fans only if SYS_ADMIN
1629 i915.invert_brightness=
1630 [DRM] Invert the sense of the variable that is used to
1631 set the brightness of the panel backlight. Normally a
1632 brightness value of 0 indicates backlight switched off,
1633 and the maximum of the brightness value sets the backlight
1634 to maximum brightness. If this parameter is set to 0
1635 (default) and the machine requires it, or this parameter
1636 is set to 1, a brightness value of 0 sets the backlight
1637 to maximum brightness, and the maximum of the brightness
1638 value switches the backlight off.
1639 -1 -- never invert brightness
1640 0 -- machine default
1641 1 -- force brightness inversion
1644 Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]]
1646 ide-core.nodma= [HW] (E)IDE subsystem
1647 Format: =0.0 to prevent dma on hda, =0.1 hdb =1.0 hdc
1648 .vlb_clock .pci_clock .noflush .nohpa .noprobe .nowerr
1649 .cdrom .chs .ignore_cable are additional options
1650 See Documentation/ide/ide.rst.
1652 ide-generic.probe-mask= [HW] (E)IDE subsystem
1654 Probe mask for legacy ISA IDE ports. Depending on
1655 platform up to 6 ports are supported, enabled by
1656 setting corresponding bits in the mask to 1. The
1657 default value is 0x0, which has a special meaning.
1658 On systems that have PCI, it triggers scanning the
1659 PCI bus for the first and the second port, which
1660 are then probed. On systems without PCI the value
1661 of 0x0 enables probing the two first ports as if it
1664 ide-pci-generic.all-generic-ide [HW] (E)IDE subsystem
1665 Claim all unknown PCI IDE storage controllers.
1668 Format: idle=poll, idle=halt, idle=nomwait
1669 Poll forces a polling idle loop that can slightly
1670 improve the performance of waking up a idle CPU, but
1671 will use a lot of power and make the system run hot.
1673 idle=halt: Halt is forced to be used for CPU idle.
1674 In such case C2/C3 won't be used again.
1675 idle=nomwait: Disable mwait for CPU C-states
1677 ieee754= [MIPS] Select IEEE Std 754 conformance mode
1678 Format: { strict | legacy | 2008 | relaxed }
1681 Choose which programs will be accepted for execution
1682 based on the IEEE 754 NaN encoding(s) supported by
1683 the FPU and the NaN encoding requested with the value
1684 of an ELF file header flag individually set by each
1685 binary. Hardware implementations are permitted to
1686 support either or both of the legacy and the 2008 NaN
1689 Available settings are as follows:
1690 strict accept binaries that request a NaN encoding
1691 supported by the FPU
1692 legacy only accept legacy-NaN binaries, if supported
1694 2008 only accept 2008-NaN binaries, if supported
1696 relaxed accept any binaries regardless of whether
1697 supported by the FPU
1699 The FPU emulator is always able to support both NaN
1700 encodings, so if no FPU hardware is present or it has
1701 been disabled with 'nofpu', then the settings of
1702 'legacy' and '2008' strap the emulator accordingly,
1703 'relaxed' straps the emulator for both legacy-NaN and
1704 2008-NaN, whereas 'strict' enables legacy-NaN only on
1705 legacy processors and both NaN encodings on MIPS32 or
1708 The setting for ABS.fmt/NEG.fmt instruction execution
1709 mode generally follows that for the NaN encoding,
1710 except where unsupported by hardware.
1712 ignore_loglevel [KNL]
1713 Ignore loglevel setting - this will print /all/
1714 kernel messages to the console. Useful for debugging.
1715 We also add it as printk module parameter, so users
1716 could change it dynamically, usually by
1717 /sys/module/printk/parameters/ignore_loglevel.
1720 Ignore RLIMIT_DATA setting for data mappings,
1721 print warning at first misuse. Can be changed via
1722 /sys/module/kernel/parameters/ignore_rlimit_data.
1724 ihash_entries= [KNL]
1725 Set number of hash buckets for inode cache.
1727 ima_appraise= [IMA] appraise integrity measurements
1728 Format: { "off" | "enforce" | "fix" | "log" }
1731 ima_appraise_tcb [IMA] Deprecated. Use ima_policy= instead.
1732 The builtin appraise policy appraises all files
1735 ima_canonical_fmt [IMA]
1736 Use the canonical format for the binary runtime
1737 measurements, instead of host native format.
1740 Format: { md5 | sha1 | rmd160 | sha256 | sha384
1744 The list of supported hash algorithms is defined
1745 in crypto/hash_info.h.
1748 The builtin policies to load during IMA setup.
1749 Format: "tcb | appraise_tcb | secure_boot |
1750 fail_securely | critical_data"
1752 The "tcb" policy measures all programs exec'd, files
1753 mmap'd for exec, and all files opened with the read
1754 mode bit set by either the effective uid (euid=0) or
1757 The "appraise_tcb" policy appraises the integrity of
1758 all files owned by root.
1760 The "secure_boot" policy appraises the integrity
1761 of files (eg. kexec kernel image, kernel modules,
1762 firmware, policy, etc) based on file signatures.
1764 The "fail_securely" policy forces file signature
1765 verification failure also on privileged mounted
1766 filesystems with the SB_I_UNVERIFIABLE_SIGNATURE
1769 The "critical_data" policy measures kernel integrity
1772 ima_tcb [IMA] Deprecated. Use ima_policy= instead.
1773 Load a policy which meets the needs of the Trusted
1774 Computing Base. This means IMA will measure all
1775 programs exec'd, files mmap'd for exec, and all files
1776 opened for read by uid=0.
1779 Select one of defined IMA measurements template formats.
1780 Formats: { "ima" | "ima-ng" | "ima-sig" }
1784 [IMA] Define a custom template format.
1785 Format: { "field1|...|fieldN" }
1787 ima.ahash_minsize= [IMA] Minimum file size for asynchronous hash usage
1788 Format: <min_file_size>
1789 Set the minimal file size for using asynchronous hash.
1790 If left unspecified, ahash usage is disabled.
1792 ahash performance varies for different data sizes on
1793 different crypto accelerators. This option can be used
1794 to achieve the best performance for a particular HW.
1796 ima.ahash_bufsize= [IMA] Asynchronous hash buffer size
1798 Set hashing buffer size. Default: 4k.
1800 ahash performance varies for different chunk sizes on
1801 different crypto accelerators. This option can be used
1802 to achieve best performance for particular HW.
1806 Run specified binary instead of /sbin/init as init
1809 initcall_debug [KNL] Trace initcalls as they are executed. Useful
1810 for working out where the kernel is dying during
1813 initcall_blacklist= [KNL] Do not execute a comma-separated list of
1814 initcall functions. Useful for debugging built-in
1815 modules and initcalls.
1817 initrd= [BOOT] Specify the location of the initial ramdisk
1819 initrdmem= [KNL] Specify a physical address and size from which to
1820 load the initrd. If an initrd is compiled in or
1821 specified in the bootparams, it takes priority over this
1823 Format: ss[KMG],nn[KMG]
1826 init_on_alloc= [MM] Fill newly allocated pages and heap objects with
1829 Default set by CONFIG_INIT_ON_ALLOC_DEFAULT_ON.
1831 init_on_free= [MM] Fill freed pages and heap objects with zeroes.
1833 Default set by CONFIG_INIT_ON_FREE_DEFAULT_ON.
1835 init_pkru= [X86] Specify the default memory protection keys rights
1836 register contents for all processes. 0x55555554 by
1837 default (disallow access to all but pkey 0). Can
1838 override in debugfs after boot.
1840 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver
1843 int_pln_enable [X86] Enable power limit notification interrupt
1845 integrity_audit=[IMA]
1846 Format: { "0" | "1" }
1847 0 -- basic integrity auditing messages. (Default)
1848 1 -- additional integrity auditing messages.
1850 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option
1852 Enable intel iommu driver.
1854 Disable intel iommu driver.
1855 igfx_off [Default Off]
1856 By default, gfx is mapped as normal device. If a gfx
1857 device has a dedicated DMAR unit, the DMAR unit is
1858 bypassed by not enabling DMAR with this option. In
1859 this case, gfx device will use physical address for
1862 With this option iommu will not optimize to look
1863 for io virtual address below 32-bit forcing dual
1864 address cycle on pci bus for cards supporting greater
1865 than 32-bit addressing. The default is to look
1866 for translation below 32-bit and if not available
1867 then look in the higher range.
1868 strict [Default Off]
1869 With this option on every unmap_single operation will
1870 result in a hardware IOTLB flush operation as opposed
1871 to batching them for performance.
1872 sp_off [Default Off]
1873 By default, super page will be supported if Intel IOMMU
1874 has the capability. With this option, super page will
1877 By default, scalable mode will be disabled even if the
1878 hardware advertises that it has support for the scalable
1879 mode translation. With this option set, scalable mode
1880 will be used on hardware which claims to support it.
1881 tboot_noforce [Default Off]
1882 Do not force the Intel IOMMU enabled under tboot.
1883 By default, tboot will force Intel IOMMU on, which
1884 could harm performance of some high-throughput
1885 devices like 40GBit network cards, even if identity
1887 Note that using this option lowers the security
1888 provided by tboot because it makes the system
1889 vulnerable to DMA attacks.
1891 intel_idle.max_cstate= [KNL,HW,ACPI,X86]
1892 0 disables intel_idle and fall back on acpi_idle.
1893 1 to 9 specify maximum depth of C-state.
1897 Do not enable intel_pstate as the default
1898 scaling driver for the supported processors
1900 Use intel_pstate as a scaling driver, but configure it
1901 to work with generic cpufreq governors (instead of
1902 enabling its internal governor). This mode cannot be
1903 used along with the hardware-managed P-states (HWP)
1906 Enable intel_pstate on systems that prohibit it by default
1907 in favor of acpi-cpufreq. Forcing the intel_pstate driver
1908 instead of acpi-cpufreq may disable platform features, such
1909 as thermal controls and power capping, that rely on ACPI
1910 P-States information being indicated to OSPM and therefore
1911 should be used with caution. This option does not work with
1912 processors that aren't supported by the intel_pstate driver
1913 or on platforms that use pcc-cpufreq instead of acpi-cpufreq.
1915 Do not enable hardware P state control (HWP)
1918 Only load intel_pstate on systems which support
1919 hardware P state control (HWP) if available.
1921 Enforce ACPI _PPC performance limits. If the Fixed ACPI
1922 Description Table, specifies preferred power management
1923 profile as "Enterprise Server" or "Performance Server",
1924 then this feature is turned on by default.
1926 Allow per-logical-CPU P-State performance control limits using
1927 cpufreq sysfs interface
1929 intremap= [X86-64, Intel-IOMMU]
1930 on enable Interrupt Remapping (default)
1931 off disable Interrupt Remapping
1932 nosid disable Source ID checking
1934 BIOS x2APIC opt-out request will be ignored
1935 nopost disable Interrupt Posting
1937 iomem= Disable strict checking of access to MMIO memory
1938 strict regions from userspace.
1953 nobypass [PPC/POWERNV]
1954 Disable IOMMU bypass, using IOMMU for PCI devices.
1956 iommu.strict= [ARM64] Configure TLB invalidation behaviour
1957 Format: { "0" | "1" }
1959 Request that DMA unmap operations use deferred
1960 invalidation of hardware TLBs, for increased
1961 throughput at the cost of reduced device isolation.
1962 Will fall back to strict mode if not supported by
1963 the relevant IOMMU driver.
1964 1 - Strict mode (default).
1965 DMA unmap operations invalidate IOMMU hardware TLBs
1969 [ARM64, X86] Configure DMA to bypass the IOMMU by default.
1970 Format: { "0" | "1" }
1971 0 - Use IOMMU translation for DMA.
1972 1 - Bypass the IOMMU for DMA.
1973 unset - Use value of CONFIG_IOMMU_DEFAULT_PASSTHROUGH.
1975 io7= [HW] IO7 for Marvel-based Alpha systems
1976 See comment before marvel_specify_io7 in
1977 arch/alpha/kernel/core_marvel.c.
1979 io_delay= [X86] I/O delay method
1981 Standard port 0x80 based delay
1983 Alternate port 0xed based delay (needed on some systems)
1985 Simple two microseconds delay
1990 See Documentation/admin-guide/nfs/nfsroot.rst.
1992 ipcmni_extend [KNL] Extend the maximum number of unique System V
1993 IPC identifiers from 32,768 to 16,777,216.
1995 irqaffinity= [SMP] Set the default irq affinity mask
1996 The argument is a cpu list, as described above.
1998 irqchip.gicv2_force_probe=
2001 Force the kernel to look for the second 4kB page
2002 of a GICv2 controller even if the memory range
2003 exposed by the device tree is too small.
2005 irqchip.gicv3_nolpi=
2007 Force the kernel to ignore the availability of
2008 LPIs (and by consequence ITSs). Intended for system
2009 that use the kernel as a bootloader, and thus want
2010 to let secondary kernels in charge of setting up
2013 irqchip.gicv3_pseudo_nmi= [ARM64]
2014 Enables support for pseudo-NMIs in the kernel. This
2015 requires the kernel to be built with
2016 CONFIG_ARM64_PSEUDO_NMI.
2019 When an interrupt is not handled search all handlers
2020 for it. Intended to get systems with badly broken
2024 When an interrupt is not handled search all handlers
2025 for it. Also check all handlers each timer
2026 interrupt. Intended to get systems with badly broken
2030 Format: <RDP>,<reset>,<pci_scan>,<verbosity>
2032 isolcpus= [KNL,SMP,ISOL] Isolate a given set of CPUs from disturbance.
2033 [Deprecated - use cpusets instead]
2034 Format: [flag-list,]<cpu-list>
2036 Specify one or more CPUs to isolate from disturbances
2037 specified in the flag list (default: domain):
2040 Disable the tick when a single task runs.
2042 A residual 1Hz tick is offloaded to workqueues, which you
2043 need to affine to housekeeping through the global
2044 workqueue's affinity configured via the
2045 /sys/devices/virtual/workqueue/cpumask sysfs file, or
2046 by using the 'domain' flag described below.
2048 NOTE: by default the global workqueue runs on all CPUs,
2049 so to protect individual CPUs the 'cpumask' file has to
2050 be configured manually after bootup.
2053 Isolate from the general SMP balancing and scheduling
2054 algorithms. Note that performing domain isolation this way
2055 is irreversible: it's not possible to bring back a CPU to
2056 the domains once isolated through isolcpus. It's strongly
2057 advised to use cpusets instead to disable scheduler load
2058 balancing through the "cpuset.sched_load_balance" file.
2059 It offers a much more flexible interface where CPUs can
2060 move in and out of an isolated set anytime.
2062 You can move a process onto or off an "isolated" CPU via
2063 the CPU affinity syscalls or cpuset.
2064 <cpu number> begins at 0 and the maximum value is
2065 "number of CPUs in system - 1".
2069 Isolate from being targeted by managed interrupts
2070 which have an interrupt mask containing isolated
2071 CPUs. The affinity of managed interrupts is
2072 handled by the kernel and cannot be changed via
2073 the /proc/irq/* interfaces.
2075 This isolation is best effort and only effective
2076 if the automatically assigned interrupt mask of a
2077 device queue contains isolated and housekeeping
2078 CPUs. If housekeeping CPUs are online then such
2079 interrupts are directed to the housekeeping CPU
2080 so that IO submitted on the housekeeping CPU
2081 cannot disturb the isolated CPU.
2083 If a queue's affinity mask contains only isolated
2084 CPUs then this parameter has no effect on the
2085 interrupt routing decision, though interrupts are
2086 only delivered when tasks running on those
2087 isolated CPUs submit IO. IO submitted on
2088 housekeeping CPUs has no influence on those
2091 The format of <cpu-list> is described above.
2095 ivrs_ioapic [HW,X86-64]
2096 Provide an override to the IOAPIC-ID<->DEVICE-ID
2097 mapping provided in the IVRS ACPI table. For
2098 example, to map IOAPIC-ID decimal 10 to
2099 PCI device 00:14.0 write the parameter as:
2100 ivrs_ioapic[10]=00:14.0
2102 ivrs_hpet [HW,X86-64]
2103 Provide an override to the HPET-ID<->DEVICE-ID
2104 mapping provided in the IVRS ACPI table. For
2105 example, to map HPET-ID decimal 0 to
2106 PCI device 00:14.0 write the parameter as:
2107 ivrs_hpet[0]=00:14.0
2109 ivrs_acpihid [HW,X86-64]
2110 Provide an override to the ACPI-HID:UID<->DEVICE-ID
2111 mapping provided in the IVRS ACPI table. For
2112 example, to map UART-HID:UID AMD0020:0 to
2113 PCI device 00:14.5 write the parameter as:
2114 ivrs_acpihid[00:14.5]=AMD0020:0
2116 js= [HW,JOY] Analog joystick
2117 See Documentation/input/joydev/joystick.rst.
2120 When CONFIG_RANDOMIZE_BASE is set, this disables
2121 kernel and module base offset ASLR (Address Space
2122 Layout Randomization).
2125 [KNL] Enforce KASAN (Kernel Address Sanitizer) to print
2126 report on every invalid memory access. Without this
2127 parameter KASAN will print report only for the first
2132 kernelcore= [KNL,X86,IA-64,PPC]
2133 Format: nn[KMGTPE] | nn% | "mirror"
2134 This parameter specifies the amount of memory usable by
2135 the kernel for non-movable allocations. The requested
2136 amount is spread evenly throughout all nodes in the
2137 system as ZONE_NORMAL. The remaining memory is used for
2138 movable memory in its own zone, ZONE_MOVABLE. In the
2139 event, a node is too small to have both ZONE_NORMAL and
2140 ZONE_MOVABLE, kernelcore memory will take priority and
2141 other nodes will have a larger ZONE_MOVABLE.
2143 ZONE_MOVABLE is used for the allocation of pages that
2144 may be reclaimed or moved by the page migration
2145 subsystem. Note that allocations like PTEs-from-HighMem
2146 still use the HighMem zone if it exists, and the Normal
2147 zone if it does not.
2149 It is possible to specify the exact amount of memory in
2150 the form of "nn[KMGTPE]", a percentage of total system
2151 memory in the form of "nn%", or "mirror". If "mirror"
2152 option is specified, mirrored (reliable) memory is used
2153 for non-movable allocations and remaining memory is used
2154 for Movable pages. "nn[KMGTPE]", "nn%", and "mirror"
2155 are exclusive, so you cannot specify multiple forms.
2157 kgdbdbgp= [KGDB,HW] kgdb over EHCI usb debug port.
2158 Format: <Controller#>[,poll interval]
2159 The controller # is the number of the ehci usb debug
2160 port as it is probed via PCI. The poll interval is
2161 optional and is the number seconds in between
2162 each poll cycle to the debug port in case you need
2163 the functionality for interrupting the kernel with
2164 gdb or control-c on the dbgp connection. When
2165 not using this parameter you use sysrq-g to break into
2166 the kernel debugger.
2168 kgdboc= [KGDB,HW] kgdb over consoles.
2169 Requires a tty driver that supports console polling,
2170 or a supported polling keyboard driver (non-usb).
2171 Serial only format: <serial_device>[,baud]
2172 keyboard only format: kbd
2173 keyboard and serial format: kbd,<serial_device>[,baud]
2174 Optional Kernel mode setting:
2175 kms, kbd format: kms,kbd
2176 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud]
2178 kgdboc_earlycon= [KGDB,HW]
2179 If the boot console provides the ability to read
2180 characters and can work in polling mode, you can use
2181 this parameter to tell kgdb to use it as a backend
2182 until the normal console is registered. Intended to
2183 be used together with the kgdboc parameter which
2184 specifies the normal console to transition to.
2186 The name of the early console should be specified
2187 as the value of this parameter. Note that the name of
2188 the early console might be different than the tty
2189 name passed to kgdboc. It's OK to leave the value
2190 blank and the first boot console that implements
2191 read() will be picked.
2193 kgdbwait [KGDB] Stop kernel execution and enter the
2194 kernel debugger at the earliest opportunity.
2196 kmac= [MIPS] Korina ethernet MAC address.
2197 Configure the RouterBoard 532 series on-chip
2198 Ethernet adapter MAC address.
2200 kmemleak= [KNL] Boot-time kmemleak enable/disable
2201 Valid arguments: on, off
2203 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y,
2206 kprobe_event=[probe-list]
2207 [FTRACE] Add kprobe events and enable at boot time.
2208 The probe-list is a semicolon delimited list of probe
2209 definitions. Each definition is same as kprobe_events
2210 interface, but the parameters are comma delimited.
2211 For example, to add a kprobe event on vfs_read with
2212 arg1 and arg2, add to the command line;
2214 kprobe_event=p,vfs_read,$arg1,$arg2
2216 See also Documentation/trace/kprobetrace.rst "Kernel
2217 Boot Parameter" section.
2219 kpti= [ARM64] Control page table isolation of user
2220 and kernel address spaces.
2221 Default: enabled on cores which need mitigation.
2225 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs.
2226 Default is 0 (don't ignore, but inject #GP)
2228 kvm.enable_vmware_backdoor=[KVM] Support VMware backdoor PV interface.
2229 Default is false (don't support).
2231 kvm.mmu_audit= [KVM] This is a R/W parameter which allows audit
2236 [KVM] Controls the software workaround for the
2237 X86_BUG_ITLB_MULTIHIT bug.
2238 force : Always deploy workaround.
2239 off : Never deploy workaround.
2240 auto : Deploy workaround based on the presence of
2241 X86_BUG_ITLB_MULTIHIT.
2245 If the software workaround is enabled for the host,
2246 guests do need not to enable it for nested guests.
2248 kvm.nx_huge_pages_recovery_ratio=
2249 [KVM] Controls how many 4KiB pages are periodically zapped
2250 back to huge pages. 0 disables the recovery, otherwise if
2251 the value is N KVM will zap 1/Nth of the 4KiB pages every
2252 minute. The default is 60.
2254 kvm-amd.nested= [KVM,AMD] Allow nested virtualization in KVM/SVM.
2255 Default is 1 (enabled)
2257 kvm-amd.npt= [KVM,AMD] Disable nested paging (virtualized MMU)
2259 Default is 1 (enabled) if in 64-bit or 32-bit PAE mode.
2262 [KVM,ARM] Select one of KVM/arm64's modes of operation.
2264 nvhe: Standard nVHE-based mode, without support for
2267 protected: nVHE-based mode with support for guests whose
2268 state is kept private from the host.
2269 Not valid if the kernel is running in EL2.
2271 Defaults to VHE/nVHE based on hardware support and
2272 the value of CONFIG_ARM64_VHE.
2274 kvm-arm.vgic_v3_group0_trap=
2275 [KVM,ARM] Trap guest accesses to GICv3 group-0
2278 kvm-arm.vgic_v3_group1_trap=
2279 [KVM,ARM] Trap guest accesses to GICv3 group-1
2282 kvm-arm.vgic_v3_common_trap=
2283 [KVM,ARM] Trap guest accesses to GICv3 common
2286 kvm-arm.vgic_v4_enable=
2287 [KVM,ARM] Allow use of GICv4 for direct injection of
2290 kvm_cma_resv_ratio=n [PPC]
2291 Reserves given percentage from system memory area for
2292 contiguous memory allocation for KVM hash pagetable
2294 By default it reserves 5% of total system memory.
2298 kvm-intel.ept= [KVM,Intel] Disable extended page tables
2299 (virtualized MMU) support on capable Intel chips.
2300 Default is 1 (enabled)
2302 kvm-intel.emulate_invalid_guest_state=
2303 [KVM,Intel] Enable emulation of invalid guest states
2304 Default is 0 (disabled)
2306 kvm-intel.flexpriority=
2307 [KVM,Intel] Disable FlexPriority feature (TPR shadow).
2308 Default is 1 (enabled)
2311 [KVM,Intel] Enable VMX nesting (nVMX).
2312 Default is 0 (disabled)
2314 kvm-intel.unrestricted_guest=
2315 [KVM,Intel] Disable unrestricted guest feature
2316 (virtualized real and unpaged mode) on capable
2317 Intel chips. Default is 1 (enabled)
2319 kvm-intel.vmentry_l1d_flush=[KVM,Intel] Mitigation for L1 Terminal Fault
2322 Valid arguments: never, cond, always
2324 always: L1D cache flush on every VMENTER.
2325 cond: Flush L1D on VMENTER only when the code between
2326 VMEXIT and VMENTER can leak host memory.
2327 never: Disables the mitigation
2329 Default is cond (do L1 cache flush in specific instances)
2331 kvm-intel.vpid= [KVM,Intel] Disable Virtual Processor Identification
2332 feature (tagged TLBs) on capable Intel chips.
2333 Default is 1 (enabled)
2335 l1tf= [X86] Control mitigation of the L1TF vulnerability on
2338 The kernel PTE inversion protection is unconditionally
2339 enabled and cannot be disabled.
2342 Provides all available mitigations for the
2343 L1TF vulnerability. Disables SMT and
2344 enables all mitigations in the
2345 hypervisors, i.e. unconditional L1D flush.
2347 SMT control and L1D flush control via the
2348 sysfs interface is still possible after
2349 boot. Hypervisors will issue a warning
2350 when the first VM is started in a
2351 potentially insecure configuration,
2352 i.e. SMT enabled or L1D flush disabled.
2355 Same as 'full', but disables SMT and L1D
2356 flush runtime control. Implies the
2357 'nosmt=force' command line option.
2358 (i.e. sysfs control of SMT is disabled.)
2361 Leaves SMT enabled and enables the default
2362 hypervisor mitigation, i.e. conditional
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.
2374 Disables SMT and enables the default
2375 hypervisor mitigation.
2377 SMT control and L1D flush control via the
2378 sysfs interface is still possible after
2379 boot. Hypervisors will issue a warning
2380 when the first VM is started in a
2381 potentially insecure configuration,
2382 i.e. SMT enabled or L1D flush disabled.
2385 Same as 'flush', but hypervisors will not
2386 warn when a VM is started in a potentially
2387 insecure configuration.
2390 Disables hypervisor mitigations and doesn't
2392 It also drops the swap size and available
2393 RAM limit restriction on both hypervisor and
2398 For details see: Documentation/admin-guide/hw-vuln/l1tf.rst
2404 lapic [X86-32,APIC] Enable the local APIC even if BIOS
2407 lapic= [X86,APIC] Do not use TSC deadline
2408 value for LAPIC timer one-shot implementation. Default
2409 back to the programmable timer unit in the LAPIC.
2410 Format: notscdeadline
2412 lapic_timer_c2_ok [X86,APIC] trust the local apic timer
2415 libata.dma= [LIBATA] DMA control
2416 libata.dma=0 Disable all PATA and SATA DMA
2417 libata.dma=1 PATA and SATA Disk DMA only
2418 libata.dma=2 ATAPI (CDROM) DMA only
2419 libata.dma=4 Compact Flash DMA only
2420 Combinations also work, so libata.dma=3 enables DMA
2421 for disks and CDROMs, but not CFs.
2423 libata.ignore_hpa= [LIBATA] Ignore HPA limit
2424 libata.ignore_hpa=0 keep BIOS limits (default)
2425 libata.ignore_hpa=1 ignore limits, using full disk
2427 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume
2431 libata.force= [LIBATA] Force configurations. The format is comma-
2432 separated list of "[ID:]VAL" where ID is
2433 PORT[.DEVICE]. PORT and DEVICE are decimal numbers
2434 matching port, link or device. Basically, it matches
2435 the ATA ID string printed on console by libata. If
2436 the whole ID part is omitted, the last PORT and DEVICE
2437 values are used. If ID hasn't been specified yet, the
2438 configuration applies to all ports, links and devices.
2440 If only DEVICE is omitted, the parameter applies to
2441 the port and all links and devices behind it. DEVICE
2442 number of 0 either selects the first device or the
2443 first fan-out link behind PMP device. It does not
2444 select the host link. DEVICE number of 15 selects the
2445 host link and device attached to it.
2447 The VAL specifies the configuration to force. As long
2448 as there's no ambiguity shortcut notation is allowed.
2449 For example, both 1.5 and 1.5G would work for 1.5Gbps.
2450 The following configurations can be forced.
2452 * Cable type: 40c, 80c, short40c, unk, ign or sata.
2453 Any ID with matching PORT is used.
2455 * SATA link speed limit: 1.5Gbps or 3.0Gbps.
2457 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7].
2458 udma[/][16,25,33,44,66,100,133] notation is also
2461 * [no]ncq: Turn on or off NCQ.
2463 * [no]ncqtrim: Turn off queued DSM TRIM.
2465 * nohrst, nosrst, norst: suppress hard, soft
2468 * rstonce: only attempt one reset during
2469 hot-unplug link recovery
2471 * dump_id: dump IDENTIFY data.
2473 * atapi_dmadir: Enable ATAPI DMADIR bridge support
2475 * disable: Disable this device.
2477 If there are multiple matching configurations changing
2478 the same attribute, the last one is used.
2480 memblock=debug [KNL] Enable memblock debug messages.
2482 load_ramdisk= [RAM] [Deprecated]
2484 lockd.nlm_grace_period=P [NFS] Assign grace period.
2487 lockd.nlm_tcpport=N [NFS] Assign TCP port.
2490 lockd.nlm_timeout=T [NFS] Assign timeout value.
2493 lockd.nlm_udpport=M [NFS] Assign UDP port.
2496 lockdown= [SECURITY]
2497 { integrity | confidentiality }
2498 Enable the kernel lockdown feature. If set to
2499 integrity, kernel features that allow userland to
2500 modify the running kernel are disabled. If set to
2501 confidentiality, kernel features that allow userland
2502 to extract confidential information from the kernel
2505 locktorture.nreaders_stress= [KNL]
2506 Set the number of locking read-acquisition kthreads.
2507 Defaults to being automatically set based on the
2508 number of online CPUs.
2510 locktorture.nwriters_stress= [KNL]
2511 Set the number of locking write-acquisition kthreads.
2513 locktorture.onoff_holdoff= [KNL]
2514 Set time (s) after boot for CPU-hotplug testing.
2516 locktorture.onoff_interval= [KNL]
2517 Set time (s) between CPU-hotplug operations, or
2518 zero to disable CPU-hotplug testing.
2520 locktorture.shuffle_interval= [KNL]
2521 Set task-shuffle interval (jiffies). Shuffling
2522 tasks allows some CPUs to go into dyntick-idle
2523 mode during the locktorture test.
2525 locktorture.shutdown_secs= [KNL]
2526 Set time (s) after boot system shutdown. This
2527 is useful for hands-off automated testing.
2529 locktorture.stat_interval= [KNL]
2530 Time (s) between statistics printk()s.
2532 locktorture.stutter= [KNL]
2533 Time (s) to stutter testing, for example,
2534 specifying five seconds causes the test to run for
2535 five seconds, wait for five seconds, and so on.
2536 This tests the locking primitive's ability to
2537 transition abruptly to and from idle.
2539 locktorture.torture_type= [KNL]
2540 Specify the locking implementation to test.
2542 locktorture.verbose= [KNL]
2543 Enable additional printk() statements.
2545 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver
2548 loglevel= All Kernel Messages with a loglevel smaller than the
2549 console loglevel will be printed to the console. It can
2550 also be changed with klogd or other programs. The
2551 loglevels are defined as follows:
2553 0 (KERN_EMERG) system is unusable
2554 1 (KERN_ALERT) action must be taken immediately
2555 2 (KERN_CRIT) critical conditions
2556 3 (KERN_ERR) error conditions
2557 4 (KERN_WARNING) warning conditions
2558 5 (KERN_NOTICE) normal but significant condition
2559 6 (KERN_INFO) informational
2560 7 (KERN_DEBUG) debug-level messages
2562 log_buf_len=n[KMG] Sets the size of the printk ring buffer,
2563 in bytes. n must be a power of two and greater
2564 than the minimal size. The minimal size is defined
2565 by LOG_BUF_SHIFT kernel config parameter. There is
2566 also CONFIG_LOG_CPU_MAX_BUF_SHIFT config parameter
2567 that allows to increase the default size depending on
2568 the number of CPUs. See init/Kconfig for more details.
2570 logo.nologo [FB] Disables display of the built-in Linux logo.
2571 This may be used to provide more screen space for
2572 kernel log messages and is useful when debugging
2573 kernel boot problems.
2575 lp=0 [LP] Specify parallel ports to use, e.g,
2576 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses
2577 lp=reset first parallel port). 'lp=0' disables the
2578 lp=auto printer driver. 'lp=reset' (which can be
2579 specified in addition to the ports) causes
2580 attached printers to be reset. Using
2581 lp=port1,port2,... specifies the parallel ports
2582 to associate lp devices with, starting with
2583 lp0. A port specification may be 'none' to skip
2584 that lp device, or a parport name such as
2585 'parport0'. Specifying 'lp=auto' instead of a
2586 port specification list means that device IDs
2587 from each port should be examined, to see if
2588 an IEEE 1284-compliant printer is attached; if
2589 so, the driver will manage that printer.
2590 See also header of drivers/char/lp.c.
2593 Sets loops_per_jiffy to given constant, thus avoiding
2594 time-consuming boot-time autodetection (up to 250 ms per
2595 CPU). 0 enables autodetection (default). To determine
2596 the correct value for your kernel, boot with normal
2597 autodetection and see what value is printed. Note that
2598 on SMP systems the preset will be applied to all CPUs,
2599 which is likely to cause problems if your CPUs need
2600 significantly divergent settings. An incorrect value
2601 will cause delays in the kernel to be wrong, leading to
2602 unpredictable I/O errors and other breakage. Although
2603 unlikely, in the extreme case this might damage your
2607 Format: <io>,<irq>,<dma>
2609 lsm.debug [SECURITY] Enable LSM initialization debugging output.
2612 [SECURITY] Choose order of LSM initialization. This
2613 overrides CONFIG_LSM, and the "security=" parameter.
2615 machvec= [IA-64] Force the use of a particular machine-vector
2616 (machvec) in a generic kernel.
2617 Example: machvec=hpzx1
2619 machtype= [Loongson] Share the same kernel image file between
2620 different yeeloong laptops.
2621 Example: machtype=lemote-yeeloong-2f-7inch
2623 max_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory greater
2624 than or equal to this physical address is ignored.
2626 maxcpus= [SMP] Maximum number of processors that an SMP kernel
2627 will bring up during bootup. maxcpus=n : n >= 0 limits
2628 the kernel to bring up 'n' processors. Surely after
2629 bootup you can bring up the other plugged cpu by executing
2630 "echo 1 > /sys/devices/system/cpu/cpuX/online". So maxcpus
2631 only takes effect during system bootup.
2632 While n=0 is a special case, it is equivalent to "nosmp",
2633 which also disables the IO APIC.
2635 max_loop= [LOOP] The number of loop block devices that get
2636 (loop.max_loop) unconditionally pre-created at init time. The default
2637 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead
2638 of statically allocating a predefined number, loop
2639 devices can be requested on-demand with the
2640 /dev/loop-control interface.
2642 mce [X86-32] Machine Check Exception
2644 mce=option [X86-64] See Documentation/x86/x86_64/boot-options.rst
2646 md= [HW] RAID subsystems devices and level
2647 See Documentation/admin-guide/md.rst.
2650 Format: <first>,<last>
2651 Specifies range of consoles to be captured by the MDA.
2654 Control mitigation for the Micro-architectural Data
2655 Sampling (MDS) vulnerability.
2657 Certain CPUs are vulnerable to an exploit against CPU
2658 internal buffers which can forward information to a
2659 disclosure gadget under certain conditions.
2661 In vulnerable processors, the speculatively
2662 forwarded data can be used in a cache side channel
2663 attack, to access data to which the attacker does
2664 not have direct access.
2666 This parameter controls the MDS mitigation. The
2669 full - Enable MDS mitigation on vulnerable CPUs
2670 full,nosmt - Enable MDS mitigation and disable
2671 SMT on vulnerable CPUs
2672 off - Unconditionally disable MDS mitigation
2674 On TAA-affected machines, mds=off can be prevented by
2675 an active TAA mitigation as both vulnerabilities are
2676 mitigated with the same mechanism so in order to disable
2677 this mitigation, you need to specify tsx_async_abort=off
2680 Not specifying this option is equivalent to
2683 For details see: Documentation/admin-guide/hw-vuln/mds.rst
2685 mem=nn[KMG] [KNL,BOOT] Force usage of a specific amount of memory
2686 Amount of memory to be used in cases as follows:
2689 2 when the kernel is not able to see the whole system memory;
2690 3 memory that lies after 'mem=' boundary is excluded from
2691 the hypervisor, then assigned to KVM guests.
2693 [X86] Work as limiting max address. Use together
2694 with memmap= to avoid physical address space collisions.
2695 Without memmap= PCI devices could be placed at addresses
2696 belonging to unused RAM.
2698 Note that this only takes effects during boot time since
2699 in above case 3, memory may need be hot added after boot
2700 if system memory of hypervisor is not sufficient.
2702 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel
2706 [KNL,SH] Allow user to override the default size for
2707 per-device physically contiguous DMA buffers.
2709 memhp_default_state=online/offline
2710 [KNL] Set the initial state for the memory hotplug
2711 onlining policy. If not specified, the default value is
2712 set according to the
2713 CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE kernel config
2715 See Documentation/admin-guide/mm/memory-hotplug.rst.
2717 memmap=exactmap [KNL,X86] Enable setting of an exact
2718 E820 memory map, as specified by the user.
2719 Such memmap=exactmap lines can be constructed based on
2720 BIOS output or other requirements. See the memmap=nn@ss
2723 memmap=nn[KMG]@ss[KMG]
2724 [KNL, X86, MIPS, XTENSA] Force usage of a specific region of memory.
2725 Region of memory to be used is from ss to ss+nn.
2726 If @ss[KMG] is omitted, it is equivalent to mem=nn[KMG],
2727 which limits max address to nn[KMG].
2728 Multiple different regions can be specified,
2731 memmap=100M@2G,100M#3G,1G!1024G
2733 memmap=nn[KMG]#ss[KMG]
2734 [KNL,ACPI] Mark specific memory as ACPI data.
2735 Region of memory to be marked is from ss to ss+nn.
2737 memmap=nn[KMG]$ss[KMG]
2738 [KNL,ACPI] Mark specific memory as reserved.
2739 Region of memory to be reserved is from ss to ss+nn.
2740 Example: Exclude memory from 0x18690000-0x1869ffff
2741 memmap=64K$0x18690000
2743 memmap=0x10000$0x18690000
2744 Some bootloaders may need an escape character before '$',
2745 like Grub2, otherwise '$' and the following number
2748 memmap=nn[KMG]!ss[KMG]
2749 [KNL,X86] Mark specific memory as protected.
2750 Region of memory to be used, from ss to ss+nn.
2751 The memory region may be marked as e820 type 12 (0xc)
2752 and is NVDIMM or ADR memory.
2754 memmap=<size>%<offset>-<oldtype>+<newtype>
2755 [KNL,ACPI] Convert memory within the specified region
2756 from <oldtype> to <newtype>. If "-<oldtype>" is left
2757 out, the whole region will be marked as <newtype>,
2758 even if previously unavailable. If "+<newtype>" is left
2759 out, matching memory will be removed. Types are
2760 specified as e820 types, e.g., 1 = RAM, 2 = reserved,
2761 3 = ACPI, 12 = PRAM.
2763 memory_corruption_check=0/1 [X86]
2764 Some BIOSes seem to corrupt the first 64k of
2765 memory when doing things like suspend/resume.
2766 Setting this option will scan the memory
2767 looking for corruption. Enabling this will
2768 both detect corruption and prevent the kernel
2769 from using the memory being corrupted.
2770 However, its intended as a diagnostic tool; if
2771 repeatable BIOS-originated corruption always
2772 affects the same memory, you can use memmap=
2773 to prevent the kernel from using that memory.
2775 memory_corruption_check_size=size [X86]
2776 By default it checks for corruption in the low
2777 64k, making this memory unavailable for normal
2778 use. Use this parameter to scan for
2779 corruption in more or less memory.
2781 memory_corruption_check_period=seconds [X86]
2782 By default it checks for corruption every 60
2783 seconds. Use this parameter to check at some
2784 other rate. 0 disables periodic checking.
2786 memtest= [KNL,X86,ARM,PPC] Enable memtest
2788 default : 0 <disable>
2789 Specifies the number of memtest passes to be
2790 performed. Each pass selects another test
2791 pattern from a given set of patterns. Memtest
2792 fills the memory with this pattern, validates
2793 memory contents and reserves bad memory
2794 regions that are detected.
2796 mem_encrypt= [X86-64] AMD Secure Memory Encryption (SME) control
2797 Valid arguments: on, off
2798 Default (depends on kernel configuration option):
2799 on (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=y)
2800 off (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=n)
2801 mem_encrypt=on: Activate SME
2802 mem_encrypt=off: Do not activate SME
2804 Refer to Documentation/virt/kvm/amd-memory-encryption.rst
2805 for details on when memory encryption can be activated.
2807 mem_sleep_default= [SUSPEND] Default system suspend mode:
2808 s2idle - Suspend-To-Idle
2809 shallow - Power-On Suspend or equivalent (if supported)
2810 deep - Suspend-To-RAM or equivalent (if supported)
2811 See Documentation/admin-guide/pm/sleep-states.rst.
2813 meye.*= [HW] Set MotionEye Camera parameters
2814 See Documentation/admin-guide/media/meye.rst.
2816 mfgpt_irq= [IA-32] Specify the IRQ to use for the
2817 Multi-Function General Purpose Timers on AMD Geode
2820 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when
2821 the BIOS has incorrectly applied a workaround. TinyBIOS
2822 version 0.98 is known to be affected, 0.99 fixes the
2823 problem by letting the user disable the workaround.
2827 min_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory below this
2828 physical address is ignored.
2830 mini2440= [ARM,HW,KNL]
2831 Format:[0..2][b][c][t]
2833 MINI2440 configuration specification:
2834 0 - The attached screen is the 3.5" TFT
2835 1 - The attached screen is the 7" TFT
2836 2 - The VGA Shield is attached (1024x768)
2837 Leaving out the screen size parameter will not load
2838 the TFT driver, and the framebuffer will be left
2840 b - Enable backlight. The TFT backlight pin will be
2841 linked to the kernel VESA blanking code and a GPIO
2842 LED. This parameter is not necessary when using the
2844 c - Enable the s3c camera interface.
2845 t - Reserved for enabling touchscreen support. The
2846 touchscreen support is not enabled in the mainstream
2847 kernel as of 2.6.30, a preliminary port can be found
2848 in the "bleeding edge" mini2440 support kernel at
2849 https://repo.or.cz/w/linux-2.6/mini2440.git
2852 [X86,PPC,S390,ARM64] Control optional mitigations for
2853 CPU vulnerabilities. This is a set of curated,
2854 arch-independent options, each of which is an
2855 aggregation of existing arch-specific options.
2858 Disable all optional CPU mitigations. This
2859 improves system performance, but it may also
2860 expose users to several CPU vulnerabilities.
2861 Equivalent to: nopti [X86,PPC]
2863 nospectre_v1 [X86,PPC]
2865 nospectre_v2 [X86,PPC,S390,ARM64]
2866 spectre_v2_user=off [X86]
2867 spec_store_bypass_disable=off [X86,PPC]
2868 ssbd=force-off [ARM64]
2871 tsx_async_abort=off [X86]
2872 kvm.nx_huge_pages=off [X86]
2873 no_entry_flush [PPC]
2874 no_uaccess_flush [PPC]
2877 This does not have any effect on
2878 kvm.nx_huge_pages when
2879 kvm.nx_huge_pages=force.
2882 Mitigate all CPU vulnerabilities, but leave SMT
2883 enabled, even if it's vulnerable. This is for
2884 users who don't want to be surprised by SMT
2885 getting disabled across kernel upgrades, or who
2886 have other ways of avoiding SMT-based attacks.
2887 Equivalent to: (default behavior)
2890 Mitigate all CPU vulnerabilities, disabling SMT
2891 if needed. This is for users who always want to
2892 be fully mitigated, even if it means losing SMT.
2893 Equivalent to: l1tf=flush,nosmt [X86]
2894 mds=full,nosmt [X86]
2895 tsx_async_abort=full,nosmt [X86]
2898 [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this
2899 parameter allows control of the logging verbosity for
2900 the additional memory initialisation checks. A value
2901 of 0 disables mminit logging and a level of 4 will
2902 log everything. Information is printed at KERN_DEBUG
2903 so loglevel=8 may also need to be specified.
2906 [KNL] When CONFIG_MODULE_SIG is set, this means that
2907 modules without (valid) signatures will fail to load.
2908 Note that if CONFIG_MODULE_SIG_FORCE is set, that
2909 is always true, so this option does nothing.
2911 module_blacklist= [KNL] Do not load a comma-separated list of
2912 modules. Useful for debugging problem modules.
2915 [MOUSE] Maximum time between finger touching and
2916 leaving touchpad surface for touch to be considered
2917 a tap and be reported as a left button click (for
2918 touchpads working in absolute mode only).
2920 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices
2921 reporting absolute coordinates, such as tablets
2922 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices
2923 reporting absolute coordinates, such as tablets
2925 movablecore= [KNL,X86,IA-64,PPC]
2926 Format: nn[KMGTPE] | nn%
2927 This parameter is the complement to kernelcore=, it
2928 specifies the amount of memory used for migratable
2929 allocations. If both kernelcore and movablecore is
2930 specified, then kernelcore will be at *least* the
2931 specified value but may be more. If movablecore on its
2932 own is specified, the administrator must be careful
2933 that the amount of memory usable for all allocations
2936 movable_node [KNL] Boot-time switch to make hotplugable memory
2937 NUMA nodes to be movable. This means that the memory
2938 of such nodes will be usable only for movable
2939 allocations which rules out almost all kernel
2940 allocations. Use with caution!
2942 MTD_Partition= [MTD]
2943 Format: <name>,<region-number>,<size>,<offset>
2945 MTD_Region= [MTD] Format:
2946 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>]
2949 See drivers/mtd/parsers/cmdlinepart.c
2951 multitce=off [PPC] This parameter disables the use of the pSeries
2952 firmware feature for updating multiple TCE entries
2955 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration
2957 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock]
2959 boundary - index of last SLC block on Flex-OneNAND.
2960 The remaining blocks are configured as MLC blocks.
2961 lock - Configure if Flex-OneNAND boundary should be locked.
2962 Once locked, the boundary cannot be changed.
2963 1 indicates lock status, 0 indicates unlock status.
2966 ARM/S3C2412 JIVE boot control
2968 See arch/arm/mach-s3c/mach-jive.c
2970 mtouchusb.raw_coordinates=
2971 [HW] Make the MicroTouch USB driver use raw coordinates
2972 ('y', default) or cooked coordinates ('n')
2974 mtrr_chunk_size=nn[KMG] [X86]
2975 used for mtrr cleanup. It is largest continuous chunk
2976 that could hold holes aka. UC entries.
2978 mtrr_gran_size=nn[KMG] [X86]
2979 Used for mtrr cleanup. It is granularity of mtrr block.
2981 Large value could prevent small alignment from
2984 mtrr_spare_reg_nr=n [X86]
2986 Range: 0,7 : spare reg number
2988 Used for mtrr cleanup. It is spare mtrr entries number.
2989 Set to 2 or more if your graphical card needs more.
2991 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card
2993 netdev= [NET] Network devices parameters
2994 Format: <irq>,<io>,<mem_start>,<mem_end>,<name>
2995 Note that mem_start is often overloaded to mean
2996 something different and driver-specific.
2997 This usage is only documented in each driver source
3001 [NETFILTER] Enable connection tracking flow accounting
3002 0 to disable accounting
3003 1 to enable accounting
3006 nfsaddrs= [NFS] Deprecated. Use ip= instead.
3007 See Documentation/admin-guide/nfs/nfsroot.rst.
3009 nfsroot= [NFS] nfs root filesystem for disk-less boxes.
3010 See Documentation/admin-guide/nfs/nfsroot.rst.
3012 nfsrootdebug [NFS] enable nfsroot debugging messages.
3013 See Documentation/admin-guide/nfs/nfsroot.rst.
3015 nfs.callback_nr_threads=
3016 [NFSv4] set the total number of threads that the
3017 NFS client will assign to service NFSv4 callback
3020 nfs.callback_tcpport=
3021 [NFS] set the TCP port on which the NFSv4 callback
3022 channel should listen.
3025 [NFS] sets the pathname to the program which is used
3026 to update the NFS client cache entries.
3028 nfs.cache_getent_timeout=
3029 [NFS] sets the timeout after which an attempt to
3030 update a cache entry is deemed to have failed.
3032 nfs.idmap_cache_timeout=
3033 [NFS] set the maximum lifetime for idmapper cache
3037 [NFS] enable 64-bit inode numbers.
3038 If zero, the NFS client will fake up a 32-bit inode
3039 number for the readdir() and stat() syscalls instead
3040 of returning the full 64-bit number.
3041 The default is to return 64-bit inode numbers.
3043 nfs.max_session_cb_slots=
3044 [NFSv4.1] Sets the maximum number of session
3045 slots the client will assign to the callback
3046 channel. This determines the maximum number of
3047 callbacks the client will process in parallel for
3048 a particular server.
3050 nfs.max_session_slots=
3051 [NFSv4.1] Sets the maximum number of session slots
3052 the client will attempt to negotiate with the server.
3053 This limits the number of simultaneous RPC requests
3054 that the client can send to the NFSv4.1 server.
3055 Note that there is little point in setting this
3056 value higher than the max_tcp_slot_table_limit.
3058 nfs.nfs4_disable_idmapping=
3059 [NFSv4] When set to the default of '1', this option
3060 ensures that both the RPC level authentication
3061 scheme and the NFS level operations agree to use
3062 numeric uids/gids if the mount is using the
3063 'sec=sys' security flavour. In effect it is
3064 disabling idmapping, which can make migration from
3065 legacy NFSv2/v3 systems to NFSv4 easier.
3066 Servers that do not support this mode of operation
3067 will be autodetected by the client, and it will fall
3068 back to using the idmapper.
3069 To turn off this behaviour, set the value to '0'.
3071 [NFS4] Specify an additional fixed unique ident-
3072 ification string that NFSv4 clients can insert into
3073 their nfs_client_id4 string. This is typically a
3074 UUID that is generated at system install time.
3076 nfs.send_implementation_id =
3077 [NFSv4.1] Send client implementation identification
3078 information in exchange_id requests.
3079 If zero, no implementation identification information
3081 The default is to send the implementation identification
3084 nfs.recover_lost_locks =
3085 [NFSv4] Attempt to recover locks that were lost due
3086 to a lease timeout on the server. Please note that
3087 doing this risks data corruption, since there are
3088 no guarantees that the file will remain unchanged
3089 after the locks are lost.
3090 If you want to enable the kernel legacy behaviour of
3091 attempting to recover these locks, then set this
3093 The default parameter value of '0' causes the kernel
3094 not to attempt recovery of lost locks.
3096 nfs4.layoutstats_timer =
3097 [NFSv4.2] Change the rate at which the kernel sends
3098 layoutstats to the pNFS metadata server.
3100 Setting this to value to 0 causes the kernel to use
3101 whatever value is the default set by the layout
3102 driver. A non-zero value sets the minimum interval
3103 in seconds between layoutstats transmissions.
3105 nfsd.nfs4_disable_idmapping=
3106 [NFSv4] When set to the default of '1', the NFSv4
3107 server will return only numeric uids and gids to
3108 clients using auth_sys, and will accept numeric uids
3109 and gids from such clients. This is intended to ease
3110 migration from NFSv2/v3.
3112 nmi_backtrace.backtrace_idle [KNL]
3113 Dump stacks even of idle CPUs in response to an
3114 NMI stack-backtrace request.
3116 nmi_debug= [KNL,SH] Specify one or more actions to take
3117 when a NMI is triggered.
3118 Format: [state][,regs][,debounce][,die]
3120 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels
3121 Format: [panic,][nopanic,][num]
3123 0 - turn hardlockup detector in nmi_watchdog off
3124 1 - turn hardlockup detector in nmi_watchdog on
3125 When panic is specified, panic when an NMI watchdog
3126 timeout occurs (or 'nopanic' to not panic on an NMI
3127 watchdog, if CONFIG_BOOTPARAM_HARDLOCKUP_PANIC is set)
3128 To disable both hard and soft lockup detectors,
3129 please see 'nowatchdog'.
3130 This is useful when you use a panic=... timeout and
3131 need the box quickly up again.
3133 These settings can be accessed at runtime via
3134 the nmi_watchdog and hardlockup_panic sysctls.
3136 netpoll.carrier_timeout=
3137 [NET] Specifies amount of time (in seconds) that
3138 netpoll should wait for a carrier. By default netpoll
3141 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths
3142 emulation library even if a 387 maths coprocessor
3145 no5lvl [X86-64] Disable 5-level paging mode. Forces
3146 kernel to use 4-level paging instead.
3148 nofsgsbase [X86] Disables FSGSBASE instructions.
3151 [HW] Never suspend the console
3152 Disable suspending of consoles during suspend and
3153 hibernate operations. Once disabled, debugging
3154 messages can reach various consoles while the rest
3155 of the system is being put to sleep (ie, while
3156 debugging driver suspend/resume hooks). This may
3157 not work reliably with all consoles, but is known
3158 to work with serial and VGA consoles.
3159 To facilitate more flexible debugging, we also add
3160 console_suspend, a printk module parameter to control
3161 it. Users could use console_suspend (usually
3162 /sys/module/printk/parameters/console_suspend) to
3163 turn on/off it dynamically.
3165 novmcoredd [KNL,KDUMP]
3166 Disable device dump. Device dump allows drivers to
3167 append dump data to vmcore so you can collect driver
3168 specified debug info. Drivers can append the data
3169 without any limit and this data is stored in memory,
3170 so this may cause significant memory stress. Disabling
3171 device dump can help save memory but the driver debug
3172 data will be no longer available. This parameter
3173 is only available when CONFIG_PROC_VMCORE_DEVICE_DUMP
3176 noaliencache [MM, NUMA, SLAB] Disables the allocation of alien
3177 caches in the slab allocator. Saves per-node memory,
3178 but will impact performance.
3182 noaltinstr [S390] Disables alternative instructions patching
3183 (CPU alternatives feature).
3185 noapic [SMP,APIC] Tells the kernel to not make use of any
3186 IOAPICs that may be present in the system.
3188 noautogroup Disable scheduler automatic task group creation.
3190 nobats [PPC] Do not use BATs for mapping kernel lowmem
3191 on "Classic" PPC cores.
3195 noclflush [BUGS=X86] Don't use the CLFLUSH instruction
3197 nodelayacct [KNL] Disable per-task delay accounting
3199 nodsp [SH] Disable hardware DSP at boot time.
3201 noefi Disable EFI runtime services support.
3203 no_entry_flush [PPC] Don't flush the L1-D cache when entering the kernel.
3208 On X86-32 available only on PAE configured kernels.
3209 noexec=on: enable non-executable mappings (default)
3210 noexec=off: disable non-executable mappings
3213 Disable SMAP (Supervisor Mode Access Prevention)
3214 even if it is supported by processor.
3217 Disable SMEP (Supervisor Mode Execution Prevention)
3218 even if it is supported by processor.
3221 This affects only 32-bit executables.
3222 noexec32=on: enable non-executable mappings (default)
3223 read doesn't imply executable mappings
3224 noexec32=off: disable non-executable mappings
3225 read implies executable mappings
3227 nofpu [MIPS,SH] Disable hardware FPU at boot time.
3229 nofxsr [BUGS=X86-32] Disables x86 floating point extended
3230 register save and restore. The kernel will only save
3231 legacy floating-point registers on task switch.
3233 nohugeiomap [KNL,X86,PPC,ARM64] Disable kernel huge I/O mappings.
3235 nosmt [KNL,S390] Disable symmetric multithreading (SMT).
3236 Equivalent to smt=1.
3238 [KNL,X86] Disable symmetric multithreading (SMT).
3239 nosmt=force: Force disable SMT, cannot be undone
3240 via the sysfs control file.
3242 nospectre_v1 [X86,PPC] Disable mitigations for Spectre Variant 1
3243 (bounds check bypass). With this option data leaks are
3244 possible in the system.
3246 nospectre_v2 [X86,PPC_FSL_BOOK3E,ARM64] Disable all mitigations for
3247 the Spectre variant 2 (indirect branch prediction)
3248 vulnerability. System may allow data leaks with this
3251 nospec_store_bypass_disable
3252 [HW] Disable all mitigations for the Speculative Store Bypass vulnerability
3255 [PPC] Don't flush the L1-D cache after accessing user data.
3257 noxsave [BUGS=X86] Disables x86 extended register state save
3258 and restore using xsave. The kernel will fallback to
3259 enabling legacy floating-point and sse state.
3261 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended
3262 register states. The kernel will fall back to use
3263 xsave to save the states. By using this parameter,
3264 performance of saving the states is degraded because
3265 xsave doesn't support modified optimization while
3266 xsaveopt supports it on xsaveopt enabled systems.
3268 noxsaves [X86] Disables xsaves and xrstors used in saving and
3269 restoring x86 extended register state in compacted
3270 form of xsave area. The kernel will fall back to use
3271 xsaveopt and xrstor to save and restore the states
3272 in standard form of xsave area. By using this
3273 parameter, xsave area per process might occupy more
3274 memory on xsaves enabled systems.
3276 nohlt [ARM,ARM64,MICROBLAZE,SH] Forces the kernel to busy wait
3277 in do_idle() and not use the arch_cpu_idle()
3278 implementation; requires CONFIG_GENERIC_IDLE_POLL_SETUP
3279 to be effective. This is useful on platforms where the
3280 sleep(SH) or wfi(ARM,ARM64) instructions do not work
3281 correctly or when doing power measurements to evalute
3282 the impact of the sleep instructions. This is also
3283 useful when using JTAG debugger.
3285 no_file_caps Tells the kernel not to honor file capabilities. The
3286 only way then for a file to be executed with privilege
3287 is to be setuid root or executed by root.
3289 nohalt [IA-64] Tells the kernel not to use the power saving
3290 function PAL_HALT_LIGHT when idle. This increases
3291 power-consumption. On the positive side, it reduces
3292 interrupt wake-up latency, which may improve performance
3293 in certain environments such as networked servers or
3297 Force pointers printed to the console or buffers to be
3298 unhashed. By default, when a pointer is printed via %p
3299 format string, that pointer is "hashed", i.e. obscured
3300 by hashing the pointer value. This is a security feature
3301 that hides actual kernel addresses from unprivileged
3302 users, but it also makes debugging the kernel more
3303 difficult since unequal pointers can no longer be
3304 compared. However, if this command-line option is
3305 specified, then all normal pointers will have their true
3306 value printed. Pointers printed via %pK may still be
3307 hashed. This option should only be specified when
3308 debugging the kernel. Please do not use on production
3311 nohibernate [HIBERNATION] Disable hibernation and resume.
3313 nohz= [KNL] Boottime enable/disable dynamic ticks
3314 Valid arguments: on, off
3317 nohz_full= [KNL,BOOT,SMP,ISOL]
3318 The argument is a cpu list, as described above.
3319 In kernels built with CONFIG_NO_HZ_FULL=y, set
3320 the specified list of CPUs whose tick will be stopped
3321 whenever possible. The boot CPU will be forced outside
3322 the range to maintain the timekeeping. Any CPUs
3323 in this list will have their RCU callbacks offloaded,
3324 just as if they had also been called out in the
3325 rcu_nocbs= boot parameter.
3327 noiotrap [SH] Disables trapped I/O port accesses.
3329 noirqdebug [X86-32] Disables the code which attempts to detect and
3330 disable unhandled interrupt sources.
3332 no_timer_check [X86,APIC] Disables the code which tests for
3333 broken timer IRQ sources.
3335 noisapnp [ISAPNP] Disables ISA PnP code.
3337 noinitrd [RAM] Tells the kernel not to load any configured
3340 nointremap [X86-64, Intel-IOMMU] Do not enable interrupt
3342 [Deprecated - use intremap=off]
3346 noinvpcid [X86] Disable the INVPCID cpu feature.
3348 nojitter [IA-64] Disables jitter checking for ITC timers.
3350 no-kvmclock [X86,KVM] Disable paravirtualized KVM clock driver
3352 no-kvmapf [X86,KVM] Disable paravirtualized asynchronous page
3356 [X86,PV_OPS] Disable paravirtualized VMware scheduler
3357 clock and use the default one.
3359 no-steal-acc [X86,PV_OPS,ARM64] Disable paravirtualized steal time
3360 accounting. steal time is computed, but won't
3361 influence scheduler behaviour
3363 nolapic [X86-32,APIC] Do not enable or use the local APIC.
3365 nolapic_timer [X86-32,APIC] Do not use the local APIC timer.
3367 noltlbs [PPC] Do not use large page/tlb entries for kernel
3368 lowmem mapping on PPC40x and PPC8xx
3370 nomca [IA-64] Disable machine check abort handling
3372 nomce [X86-32] Disable Machine Check Exception
3374 nomfgpt [X86-32] Disable Multi-Function General Purpose
3375 Timer usage (for AMD Geode machines).
3377 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to
3378 shutdown the other cpus. Instead use the REBOOT_VECTOR
3381 nomodule Disable module load
3383 nopat [X86] Disable PAT (page attribute table extension of
3384 pagetables) support.
3386 nopcid [X86-64] Disable the PCID cpu feature.
3388 norandmaps Don't use address space randomization. Equivalent to
3389 echo 0 > /proc/sys/kernel/randomize_va_space
3391 noreplace-smp [X86-32,SMP] Don't replace SMP instructions
3392 with UP alternatives
3394 nordrand [X86] Disable kernel use of the RDRAND and
3395 RDSEED instructions even if they are supported
3396 by the processor. RDRAND and RDSEED are still
3397 available to user space applications.
3399 noresume [SWSUSP] Disables resume and restores original swap
3402 no-scroll [VGA] Disables scrollback.
3403 This is required for the Braillex ib80-piezo Braille
3404 reader made by F.H. Papenmeier (Germany).
3408 nosep [BUGS=X86-32] Disables x86 SYSENTER/SYSEXIT support.
3410 nosgx [X86-64,SGX] Disables Intel SGX kernel support.
3412 nosmp [SMP] Tells an SMP kernel to act as a UP kernel,
3413 and disable the IO APIC. legacy for "maxcpus=0".
3415 nosoftlockup [KNL] Disable the soft-lockup detector.
3417 nosync [HW,M68K] Disables sync negotiation for all devices.
3419 nowatchdog [KNL] Disable both lockup detectors, i.e.
3420 soft-lockup and NMI watchdog (hard-lockup).
3424 nox2apic [X86-64,APIC] Do not enable x2APIC mode.
3426 cpu0_hotplug [X86] Turn on CPU0 hotplug feature when
3427 CONFIG_BOOTPARAM_HOTPLUG_CPU0 is off.
3428 Some features depend on CPU0. Known dependencies are:
3429 1. Resume from suspend/hibernate depends on CPU0.
3430 Suspend/hibernate will fail if CPU0 is offline and you
3431 need to online CPU0 before suspend/hibernate.
3432 2. PIC interrupts also depend on CPU0. CPU0 can't be
3433 removed if a PIC interrupt is detected.
3434 It's said poweroff/reboot may depend on CPU0 on some
3435 machines although I haven't seen such issues so far
3436 after CPU0 is offline on a few tested machines.
3437 If the dependencies are under your control, you can
3438 turn on cpu0_hotplug.
3440 nps_mtm_hs_ctr= [KNL,ARC]
3441 This parameter sets the maximum duration, in
3442 cycles, each HW thread of the CTOP can run
3443 without interruptions, before HW switches it.
3444 The actual maximum duration is 16 times this
3446 Format: integer between 1 and 255
3449 nptcg= [IA-64] Override max number of concurrent global TLB
3450 purges which is reported from either PAL_VM_SUMMARY or
3453 nr_cpus= [SMP] Maximum number of processors that an SMP kernel
3454 could support. nr_cpus=n : n >= 1 limits the kernel to
3455 support 'n' processors. It could be larger than the
3456 number of already plugged CPU during bootup, later in
3457 runtime you can physically add extra cpu until it reaches
3458 n. So during boot up some boot time memory for per-cpu
3459 variables need be pre-allocated for later physical cpu
3462 nr_uarts= [SERIAL] maximum number of UARTs to be registered.
3464 numa_balancing= [KNL,X86] Enable or disable automatic NUMA balancing.
3465 Allowed values are enable and disable
3467 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA.
3468 'node', 'default' can be specified
3469 This can be set from sysctl after boot.
3470 See Documentation/admin-guide/sysctl/vm.rst for details.
3472 ohci1394_dma=early [HW] enable debugging via the ohci1394 driver.
3473 See Documentation/core-api/debugging-via-ohci1394.rst for more
3476 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands
3477 Rather than timing out after 20 ms if an EC
3478 command is not properly ACKed, override the length
3479 of the timeout. We have interrupts disabled while
3480 waiting for the ACK, so if this is set too high
3481 interrupts *may* be lost!
3483 omap_mux= [OMAP] Override bootloader pin multiplexing.
3484 Format: <mux_mode0.mode_name=value>...
3485 For example, to override I2C bus2:
3486 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100
3488 oops=panic Always panic on oopses. Default is to just kill the
3489 process, but there is a small probability of
3490 deadlocking the machine.
3491 This will also cause panics on machine check exceptions.
3492 Useful together with panic=30 to trigger a reboot.
3495 [KNL] Boolean flag to control whether the page allocator
3496 should randomize its free lists. The randomization may
3497 be automatically enabled if the kernel detects it is
3498 running on a platform with a direct-mapped memory-side
3499 cache, and this parameter can be used to
3500 override/disable that behavior. The state of the flag
3501 can be read from sysfs at:
3502 /sys/module/page_alloc/parameters/shuffle.
3504 page_owner= [KNL] Boot-time page_owner enabling option.
3505 Storage of the information about who allocated
3506 each page is disabled in default. With this switch,
3508 on: enable the feature
3510 page_poison= [KNL] Boot-time parameter changing the state of
3511 poisoning on the buddy allocator, available with
3512 CONFIG_PAGE_POISONING=y.
3513 off: turn off poisoning (default)
3514 on: turn on poisoning
3516 panic= [KNL] Kernel behaviour on panic: delay <timeout>
3517 timeout > 0: seconds before rebooting
3518 timeout = 0: wait forever
3519 timeout < 0: reboot immediately
3522 panic_print= Bitmask for printing system info when panic happens.
3523 User can chose combination of the following bits:
3524 bit 0: print all tasks info
3525 bit 1: print system memory info
3526 bit 2: print timer info
3527 bit 3: print locks info if CONFIG_LOCKDEP is on
3528 bit 4: print ftrace buffer
3529 bit 5: print all printk messages in buffer
3531 panic_on_taint= Bitmask for conditionally calling panic() in add_taint()
3532 Format: <hex>[,nousertaint]
3533 Hexadecimal bitmask representing the set of TAINT flags
3534 that will cause the kernel to panic when add_taint() is
3535 called with any of the flags in this set.
3536 The optional switch "nousertaint" can be utilized to
3537 prevent userspace forced crashes by writing to sysctl
3538 /proc/sys/kernel/tainted any flagset matching with the
3539 bitmask set on panic_on_taint.
3540 See Documentation/admin-guide/tainted-kernels.rst for
3541 extra details on the taint flags that users can pick
3542 to compose the bitmask to assign to panic_on_taint.
3544 panic_on_warn panic() instead of WARN(). Useful to cause kdump
3547 crash_kexec_post_notifiers
3548 Run kdump after running panic-notifiers and dumping
3549 kmsg. This only for the users who doubt kdump always
3550 succeeds in any situation.
3551 Note that this also increases risks of kdump failure,
3552 because some panic notifiers can make the crashed
3553 kernel more unstable.
3555 parkbd.port= [HW] Parallel port number the keyboard adapter is
3556 connected to, default is 0.
3558 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation,
3559 0 for XT, 1 for AT (default is AT).
3562 parport= [HW,PPT] Specify parallel ports. 0 disables.
3563 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] }
3564 Use 'auto' to force the driver to use any
3565 IRQ/DMA settings detected (the default is to
3566 ignore detected IRQ/DMA settings because of
3567 possible conflicts). You can specify the base
3568 address, IRQ, and DMA settings; IRQ and DMA
3569 should be numbers, or 'auto' (for using detected
3570 settings on that particular port), or 'nofifo'
3571 (to avoid using a FIFO even if it is detected).
3572 Parallel ports are assigned in the order they
3573 are specified on the command line, starting
3576 parport_init_mode= [HW,PPT]
3577 Configure VIA parallel port to operate in
3578 a specific mode. This is necessary on Pegasos
3579 computer where firmware has no options for setting
3580 up parallel port mode and sets it to spp.
3581 Currently this function knows 686a and 8231 chips.
3582 Format: [spp|ps2|epp|ecp|ecpepp]
3585 Halt all CPUs after the first oops has been printed for
3586 the specified number of seconds. This is to be used if
3587 your oopses keep scrolling off the screen.
3592 See header of drivers/block/paride/pcd.c.
3593 See also Documentation/admin-guide/blockdev/paride.rst.
3595 pci=option[,option...] [PCI] various PCI subsystem options.
3597 Some options herein operate on a specific device
3598 or a set of devices (<pci_dev>). These are
3599 specified in one of the following formats:
3601 [<domain>:]<bus>:<dev>.<func>[/<dev>.<func>]*
3602 pci:<vendor>:<device>[:<subvendor>:<subdevice>]
3604 Note: the first format specifies a PCI
3605 bus/device/function address which may change
3606 if new hardware is inserted, if motherboard
3607 firmware changes, or due to changes caused
3608 by other kernel parameters. If the
3609 domain is left unspecified, it is
3610 taken to be zero. Optionally, a path
3611 to a device through multiple device/function
3612 addresses can be specified after the base
3613 address (this is more robust against
3614 renumbering issues). The second format
3615 selects devices using IDs from the
3616 configuration space which may match multiple
3617 devices in the system.
3619 earlydump dump PCI config space before the kernel
3621 off [X86] don't probe for the PCI bus
3622 bios [X86-32] force use of PCI BIOS, don't access
3623 the hardware directly. Use this if your machine
3624 has a non-standard PCI host bridge.
3625 nobios [X86-32] disallow use of PCI BIOS, only direct
3626 hardware access methods are allowed. Use this
3627 if you experience crashes upon bootup and you
3628 suspect they are caused by the BIOS.
3629 conf1 [X86] Force use of PCI Configuration Access
3630 Mechanism 1 (config address in IO port 0xCF8,
3631 data in IO port 0xCFC, both 32-bit).
3632 conf2 [X86] Force use of PCI Configuration Access
3633 Mechanism 2 (IO port 0xCF8 is an 8-bit port for
3634 the function, IO port 0xCFA, also 8-bit, sets
3635 bus number. The config space is then accessed
3636 through ports 0xC000-0xCFFF).
3637 See http://wiki.osdev.org/PCI for more info
3638 on the configuration access mechanisms.
3639 noaer [PCIE] If the PCIEAER kernel config parameter is
3640 enabled, this kernel boot option can be used to
3641 disable the use of PCIE advanced error reporting.
3642 nodomains [PCI] Disable support for multiple PCI
3643 root domains (aka PCI segments, in ACPI-speak).
3644 nommconf [X86] Disable use of MMCONFIG for PCI
3646 check_enable_amd_mmconf [X86] check for and enable
3647 properly configured MMIO access to PCI
3648 config space on AMD family 10h CPU
3649 nomsi [MSI] If the PCI_MSI kernel config parameter is
3650 enabled, this kernel boot option can be used to
3651 disable the use of MSI interrupts system-wide.
3652 noioapicquirk [APIC] Disable all boot interrupt quirks.
3653 Safety option to keep boot IRQs enabled. This
3654 should never be necessary.
3655 ioapicreroute [APIC] Enable rerouting of boot IRQs to the
3656 primary IO-APIC for bridges that cannot disable
3657 boot IRQs. This fixes a source of spurious IRQs
3658 when the system masks IRQs.
3659 noioapicreroute [APIC] Disable workaround that uses the
3660 boot IRQ equivalent of an IRQ that connects to
3661 a chipset where boot IRQs cannot be disabled.
3662 The opposite of ioapicreroute.
3663 biosirq [X86-32] Use PCI BIOS calls to get the interrupt
3664 routing table. These calls are known to be buggy
3665 on several machines and they hang the machine
3666 when used, but on other computers it's the only
3667 way to get the interrupt routing table. Try
3668 this option if the kernel is unable to allocate
3669 IRQs or discover secondary PCI buses on your
3671 rom [X86] Assign address space to expansion ROMs.
3672 Use with caution as certain devices share
3673 address decoders between ROMs and other
3675 norom [X86] Do not assign address space to
3676 expansion ROMs that do not already have
3677 BIOS assigned address ranges.
3678 nobar [X86] Do not assign address space to the
3679 BARs that weren't assigned by the BIOS.
3680 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be
3681 assigned automatically to PCI devices. You can
3682 make the kernel exclude IRQs of your ISA cards
3684 pirqaddr=0xAAAAA [X86] Specify the physical address
3685 of the PIRQ table (normally generated
3686 by the BIOS) if it is outside the
3687 F0000h-100000h range.
3688 lastbus=N [X86] Scan all buses thru bus #N. Can be
3689 useful if the kernel is unable to find your
3690 secondary buses and you want to tell it
3691 explicitly which ones they are.
3692 assign-busses [X86] Always assign all PCI bus
3693 numbers ourselves, overriding
3694 whatever the firmware may have done.
3695 usepirqmask [X86] Honor the possible IRQ mask stored
3696 in the BIOS $PIR table. This is needed on
3697 some systems with broken BIOSes, notably
3698 some HP Pavilion N5400 and Omnibook XE3
3699 notebooks. This will have no effect if ACPI
3700 IRQ routing is enabled.
3701 noacpi [X86] Do not use ACPI for IRQ routing
3702 or for PCI scanning.
3703 use_crs [X86] Use PCI host bridge window information
3704 from ACPI. On BIOSes from 2008 or later, this
3705 is enabled by default. If you need to use this,
3706 please report a bug.
3707 nocrs [X86] Ignore PCI host bridge windows from ACPI.
3708 If you need to use this, please report a bug.
3709 routeirq Do IRQ routing for all PCI devices.
3710 This is normally done in pci_enable_device(),
3711 so this option is a temporary workaround
3712 for broken drivers that don't call it.
3713 skip_isa_align [X86] do not align io start addr, so can
3714 handle more pci cards
3715 noearly [X86] Don't do any early type 1 scanning.
3716 This might help on some broken boards which
3717 machine check when some devices' config space
3718 is read. But various workarounds are disabled
3719 and some IOMMU drivers will not work.
3720 bfsort Sort PCI devices into breadth-first order.
3721 This sorting is done to get a device
3722 order compatible with older (<= 2.4) kernels.
3723 nobfsort Don't sort PCI devices into breadth-first order.
3724 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size)
3725 tuning and use the BIOS-configured MPS defaults.
3726 pcie_bus_safe Set every device's MPS to the largest value
3727 supported by all devices below the root complex.
3728 pcie_bus_perf Set device MPS to the largest allowable MPS
3729 based on its parent bus. Also set MRRS (Max
3730 Read Request Size) to the largest supported
3731 value (no larger than the MPS that the device
3732 or bus can support) for best performance.
3733 pcie_bus_peer2peer Set every device's MPS to 128B, which
3734 every device is guaranteed to support. This
3735 configuration allows peer-to-peer DMA between
3736 any pair of devices, possibly at the cost of
3737 reduced performance. This also guarantees
3738 that hot-added devices will work.
3739 cbiosize=nn[KMG] The fixed amount of bus space which is
3740 reserved for the CardBus bridge's IO window.
3741 The default value is 256 bytes.
3742 cbmemsize=nn[KMG] The fixed amount of bus space which is
3743 reserved for the CardBus bridge's memory
3744 window. The default value is 64 megabytes.
3747 [<order of align>@]<pci_dev>[; ...]
3748 Specifies alignment and device to reassign
3749 aligned memory resources. How to
3750 specify the device is described above.
3751 If <order of align> is not specified,
3752 PAGE_SIZE is used as alignment.
3753 A PCI-PCI bridge can be specified if resource
3754 windows need to be expanded.
3755 To specify the alignment for several
3756 instances of a device, the PCI vendor,
3757 device, subvendor, and subdevice may be
3758 specified, e.g., 12@pci:8086:9c22:103c:198f
3759 for 4096-byte alignment.
3760 ecrc= Enable/disable PCIe ECRC (transaction layer
3761 end-to-end CRC checking).
3762 bios: Use BIOS/firmware settings. This is the
3766 hpiosize=nn[KMG] The fixed amount of bus space which is
3767 reserved for hotplug bridge's IO window.
3768 Default size is 256 bytes.
3769 hpmmiosize=nn[KMG] The fixed amount of bus space which is
3770 reserved for hotplug bridge's MMIO window.
3771 Default size is 2 megabytes.
3772 hpmmioprefsize=nn[KMG] The fixed amount of bus space which is
3773 reserved for hotplug bridge's MMIO_PREF window.
3774 Default size is 2 megabytes.
3775 hpmemsize=nn[KMG] The fixed amount of bus space which is
3776 reserved for hotplug bridge's MMIO and
3778 Default size is 2 megabytes.
3779 hpbussize=nn The minimum amount of additional bus numbers
3780 reserved for buses below a hotplug bridge.
3782 realloc= Enable/disable reallocating PCI bridge resources
3783 if allocations done by BIOS are too small to
3784 accommodate resources required by all child
3786 off: Turn realloc off
3788 realloc same as realloc=on
3789 noari do not use PCIe ARI.
3790 noats [PCIE, Intel-IOMMU, AMD-IOMMU]
3791 do not use PCIe ATS (and IOMMU device IOTLB).
3792 pcie_scan_all Scan all possible PCIe devices. Otherwise we
3793 only look for one device below a PCIe downstream
3795 big_root_window Try to add a big 64bit memory window to the PCIe
3796 root complex on AMD CPUs. Some GFX hardware
3797 can resize a BAR to allow access to all VRAM.
3798 Adding the window is slightly risky (it may
3799 conflict with unreported devices), so this
3801 disable_acs_redir=<pci_dev>[; ...]
3802 Specify one or more PCI devices (in the format
3803 specified above) separated by semicolons.
3804 Each device specified will have the PCI ACS
3805 redirect capabilities forced off which will
3806 allow P2P traffic between devices through
3807 bridges without forcing it upstream. Note:
3808 this removes isolation between devices and
3809 may put more devices in an IOMMU group.
3810 force_floating [S390] Force usage of floating interrupts.
3811 nomio [S390] Do not use MIO instructions.
3812 norid [S390] ignore the RID field and force use of
3813 one PCI domain per PCI function
3815 pcie_aspm= [PCIE] Forcibly enable or disable PCIe Active State Power
3818 force Enable ASPM even on devices that claim not to support it.
3819 WARNING: Forcing ASPM on may cause system lockups.
3821 pcie_ports= [PCIE] PCIe port services handling:
3822 native Use native PCIe services (PME, AER, DPC, PCIe hotplug)
3823 even if the platform doesn't give the OS permission to
3824 use them. This may cause conflicts if the platform
3825 also tries to use these services.
3826 dpc-native Use native PCIe service for DPC only. May
3827 cause conflicts if firmware uses AER or DPC.
3828 compat Disable native PCIe services (PME, AER, DPC, PCIe
3831 pcie_port_pm= [PCIE] PCIe port power management handling:
3832 off Disable power management of all PCIe ports
3833 force Forcibly enable power management of all PCIe ports
3835 pcie_pme= [PCIE,PM] Native PCIe PME signaling options:
3836 nomsi Do not use MSI for native PCIe PME signaling (this makes
3837 all PCIe root ports use INTx for all services).
3839 pcmv= [HW,PCMCIA] BadgePAD 4
3843 Keep all power-domains already enabled by bootloader on,
3844 even if no driver has claimed them. This is useful
3845 for debug and development, but should not be
3846 needed on a platform with proper driver support.
3849 See Documentation/admin-guide/blockdev/paride.rst.
3851 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at
3854 See arch/parisc/kernel/pdc_chassis.c
3856 percpu_alloc= Select which percpu first chunk allocator to use.
3857 Currently supported values are "embed" and "page".
3858 Archs may support subset or none of the selections.
3859 See comments in mm/percpu.c for details on each
3860 allocator. This parameter is primarily for debugging
3861 and performance comparison.
3864 See Documentation/admin-guide/blockdev/paride.rst.
3867 See Documentation/admin-guide/blockdev/paride.rst.
3869 pirq= [SMP,APIC] Manual mp-table setup
3870 See Documentation/x86/i386/IO-APIC.rst.
3872 plip= [PPT,NET] Parallel port network link
3873 Format: { parport<nr> | timid | 0 }
3874 See also Documentation/admin-guide/parport.rst.
3876 pmtmr= [X86] Manual setup of pmtmr I/O Port.
3877 Override pmtimer IOPort with a hex value.
3880 pm_debug_messages [SUSPEND,KNL]
3881 Enable suspend/resume debug messages during boot up.
3884 Enable PNP debug messages (depends on the
3885 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time
3886 via /sys/module/pnp/parameters/debug. We always show
3887 current resource usage; turning this on also shows
3888 possible settings and some assignment information.
3894 { on | off | curr | res | no-curr | no-res }
3897 [ISAPNP] Exclude IRQs for the autoconfiguration
3900 [ISAPNP] Exclude DMAs for the autoconfiguration
3902 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration
3903 Ranges are in pairs (I/O port base and size).
3906 [ISAPNP] Exclude memory regions for the
3908 Ranges are in pairs (memory base and size).
3910 ports= [IP_VS_FTP] IPVS ftp helper module
3912 Up to 8 (IP_VS_APP_MAX_PORTS) ports
3914 Format: <port>,<port>....
3916 powersave=off [PPC] This option disables power saving features.
3917 It specifically disables cpuidle and sets the
3918 platform machine description specific power_save
3919 function to NULL. On Idle the CPU just reduces
3922 ppc_strict_facility_enable
3923 [PPC] This option catches any kernel floating point,
3924 Altivec, VSX and SPE outside of regions specifically
3925 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()).
3926 There is some performance impact when enabling this.
3930 Disable Hardware Transactional Memory
3933 Select preemption mode if you have CONFIG_PREEMPT_DYNAMIC
3934 none - Limited to cond_resched() calls
3935 voluntary - Limited to cond_resched() and might_sleep() calls
3936 full - Any section that isn't explicitly preempt disabled
3937 can be preempted anytime.
3939 print-fatal-signals=
3940 [KNL] debug: print fatal signals
3942 If enabled, warn about various signal handling
3943 related application anomalies: too many signals,
3944 too many POSIX.1 timers, fatal signals causing a
3947 If you hit the warning due to signal overflow,
3948 you might want to try "ulimit -i unlimited".
3952 printk.always_kmsg_dump=
3953 Trigger kmsg_dump for cases other than kernel oops or
3955 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
3958 printk.devkmsg={on,off,ratelimit}
3959 Control writing to /dev/kmsg.
3960 on - unlimited logging to /dev/kmsg from userspace
3961 off - logging to /dev/kmsg disabled
3962 ratelimit - ratelimit the logging
3965 printk.time= Show timing data prefixed to each printk message line
3966 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
3968 processor.max_cstate= [HW,ACPI]
3969 Limit processor to maximum C-state
3970 max_cstate=9 overrides any DMI blacklist limit.
3972 processor.nocst [HW,ACPI]
3973 Ignore the _CST method to determine C-states,
3974 instead using the legacy FADT method
3976 profile= [KNL] Enable kernel profiling via /proc/profile
3977 Format: [<profiletype>,]<number>
3978 Param: <profiletype>: "schedule", "sleep", or "kvm"
3979 [defaults to kernel profiling]
3980 Param: "schedule" - profile schedule points.
3981 Param: "sleep" - profile D-state sleeping (millisecs).
3982 Requires CONFIG_SCHEDSTATS
3983 Param: "kvm" - profile VM exits.
3984 Param: <number> - step/bucket size as a power of 2 for
3985 statistical time based profiling.
3987 prompt_ramdisk= [RAM] [Deprecated]
3989 prot_virt= [S390] enable hosting protected virtual machines
3990 isolated from the hypervisor (if hardware supports
3994 psi= [KNL] Enable or disable pressure stall information
3998 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to
3999 probe for; one of (bare|imps|exps|lifebook|any).
4000 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports
4002 psmouse.resetafter= [HW,MOUSE]
4003 Try to reset the device after so many bad packets
4006 [HW,MOUSE] Set desired mouse resolution, in dpi.
4007 psmouse.smartscroll=
4008 [HW,MOUSE] Controls Logitech smartscroll autorepeat.
4009 0 = disabled, 1 = enabled (default).
4011 pstore.backend= Specify the name of the pstore backend to use
4014 See Documentation/admin-guide/blockdev/paride.rst.
4016 pti= [X86-64] Control Page Table Isolation of user and
4017 kernel address spaces. Disabling this feature
4018 removes hardening, but improves performance of
4019 system calls and interrupts.
4021 on - unconditionally enable
4022 off - unconditionally disable
4023 auto - kernel detects whether your CPU model is
4024 vulnerable to issues that PTI mitigates
4026 Not specifying this option is equivalent to pti=auto.
4029 Equivalent to pti=off
4032 [KNL] Number of legacy pty's. Overwrites compiled-in
4035 quiet [KNL] Disable most log messages
4040 See Documentation/admin-guide/md.rst.
4042 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes
4043 See Documentation/admin-guide/blockdev/ramdisk.rst.
4045 ramdisk_start= [RAM] RAM disk image start address
4047 random.trust_cpu={on,off}
4048 [KNL] Enable or disable trusting the use of the
4049 CPU's random number generator (if available) to
4050 fully seed the kernel's CRNG. Default is controlled
4051 by CONFIG_RANDOM_TRUST_CPU.
4053 ras=option[,option,...] [KNL] RAS-specific options
4056 Disable the Correctable Errors Collector,
4057 see CONFIG_RAS_CEC help text.
4060 The argument is a cpu list, as described above,
4061 except that the string "all" can be used to
4062 specify every CPU on the system.
4064 In kernels built with CONFIG_RCU_NOCB_CPU=y, set
4065 the specified list of CPUs to be no-callback CPUs.
4066 Invocation of these CPUs' RCU callbacks will be
4067 offloaded to "rcuox/N" kthreads created for that
4068 purpose, where "x" is "p" for RCU-preempt, and
4069 "s" for RCU-sched, and "N" is the CPU number.
4070 This reduces OS jitter on the offloaded CPUs,
4071 which can be useful for HPC and real-time
4072 workloads. It can also improve energy efficiency
4073 for asymmetric multiprocessors.
4076 Rather than requiring that offloaded CPUs
4077 (specified by rcu_nocbs= above) explicitly
4078 awaken the corresponding "rcuoN" kthreads,
4079 make these kthreads poll for callbacks.
4080 This improves the real-time response for the
4081 offloaded CPUs by relieving them of the need to
4082 wake up the corresponding kthread, but degrades
4083 energy efficiency by requiring that the kthreads
4084 periodically wake up to do the polling.
4086 rcutree.blimit= [KNL]
4087 Set maximum number of finished RCU callbacks to
4088 process in one batch.
4090 rcutree.dump_tree= [KNL]
4091 Dump the structure of the rcu_node combining tree
4092 out at early boot. This is used for diagnostic
4093 purposes, to verify correct tree setup.
4095 rcutree.gp_cleanup_delay= [KNL]
4096 Set the number of jiffies to delay each step of
4097 RCU grace-period cleanup.
4099 rcutree.gp_init_delay= [KNL]
4100 Set the number of jiffies to delay each step of
4101 RCU grace-period initialization.
4103 rcutree.gp_preinit_delay= [KNL]
4104 Set the number of jiffies to delay each step of
4105 RCU grace-period pre-initialization, that is,
4106 the propagation of recent CPU-hotplug changes up
4107 the rcu_node combining tree.
4109 rcutree.use_softirq= [KNL]
4110 If set to zero, move all RCU_SOFTIRQ processing to
4111 per-CPU rcuc kthreads. Defaults to a non-zero
4112 value, meaning that RCU_SOFTIRQ is used by default.
4113 Specify rcutree.use_softirq=0 to use rcuc kthreads.
4115 But note that CONFIG_PREEMPT_RT=y kernels disable
4116 this kernel boot parameter, forcibly setting it
4119 rcutree.rcu_fanout_exact= [KNL]
4120 Disable autobalancing of the rcu_node combining
4121 tree. This is used by rcutorture, and might
4122 possibly be useful for architectures having high
4123 cache-to-cache transfer latencies.
4125 rcutree.rcu_fanout_leaf= [KNL]
4126 Change the number of CPUs assigned to each
4127 leaf rcu_node structure. Useful for very
4128 large systems, which will choose the value 64,
4129 and for NUMA systems with large remote-access
4130 latencies, which will choose a value aligned
4131 with the appropriate hardware boundaries.
4133 rcutree.rcu_min_cached_objs= [KNL]
4134 Minimum number of objects which are cached and
4135 maintained per one CPU. Object size is equal
4136 to PAGE_SIZE. The cache allows to reduce the
4137 pressure to page allocator, also it makes the
4138 whole algorithm to behave better in low memory
4141 rcutree.jiffies_till_first_fqs= [KNL]
4142 Set delay from grace-period initialization to
4143 first attempt to force quiescent states.
4144 Units are jiffies, minimum value is zero,
4145 and maximum value is HZ.
4147 rcutree.jiffies_till_next_fqs= [KNL]
4148 Set delay between subsequent attempts to force
4149 quiescent states. Units are jiffies, minimum
4150 value is one, and maximum value is HZ.
4152 rcutree.jiffies_till_sched_qs= [KNL]
4153 Set required age in jiffies for a
4154 given grace period before RCU starts
4155 soliciting quiescent-state help from
4156 rcu_note_context_switch() and cond_resched().
4157 If not specified, the kernel will calculate
4158 a value based on the most recent settings
4159 of rcutree.jiffies_till_first_fqs
4160 and rcutree.jiffies_till_next_fqs.
4161 This calculated value may be viewed in
4162 rcutree.jiffies_to_sched_qs. Any attempt to set
4163 rcutree.jiffies_to_sched_qs will be cheerfully
4166 rcutree.kthread_prio= [KNL,BOOT]
4167 Set the SCHED_FIFO priority of the RCU per-CPU
4168 kthreads (rcuc/N). This value is also used for
4169 the priority of the RCU boost threads (rcub/N)
4170 and for the RCU grace-period kthreads (rcu_bh,
4171 rcu_preempt, and rcu_sched). If RCU_BOOST is
4172 set, valid values are 1-99 and the default is 1
4173 (the least-favored priority). Otherwise, when
4174 RCU_BOOST is not set, valid values are 0-99 and
4175 the default is zero (non-realtime operation).
4177 rcutree.rcu_nocb_gp_stride= [KNL]
4178 Set the number of NOCB callback kthreads in
4179 each group, which defaults to the square root
4180 of the number of CPUs. Larger numbers reduce
4181 the wakeup overhead on the global grace-period
4182 kthread, but increases that same overhead on
4183 each group's NOCB grace-period kthread.
4185 rcutree.qhimark= [KNL]
4186 Set threshold of queued RCU callbacks beyond which
4187 batch limiting is disabled.
4189 rcutree.qlowmark= [KNL]
4190 Set threshold of queued RCU callbacks below which
4191 batch limiting is re-enabled.
4193 rcutree.qovld= [KNL]
4194 Set threshold of queued RCU callbacks beyond which
4195 RCU's force-quiescent-state scan will aggressively
4196 enlist help from cond_resched() and sched IPIs to
4197 help CPUs more quickly reach quiescent states.
4198 Set to less than zero to make this be set based
4199 on rcutree.qhimark at boot time and to zero to
4200 disable more aggressive help enlistment.
4202 rcutree.rcu_idle_gp_delay= [KNL]
4203 Set wakeup interval for idle CPUs that have
4204 RCU callbacks (RCU_FAST_NO_HZ=y).
4206 rcutree.rcu_kick_kthreads= [KNL]
4207 Cause the grace-period kthread to get an extra
4208 wake_up() if it sleeps three times longer than
4209 it should at force-quiescent-state time.
4210 This wake_up() will be accompanied by a
4211 WARN_ONCE() splat and an ftrace_dump().
4213 rcutree.rcu_unlock_delay= [KNL]
4214 In CONFIG_RCU_STRICT_GRACE_PERIOD=y kernels,
4215 this specifies an rcu_read_unlock()-time delay
4216 in microseconds. This defaults to zero.
4217 Larger delays increase the probability of
4218 catching RCU pointer leaks, that is, buggy use
4219 of RCU-protected pointers after the relevant
4220 rcu_read_unlock() has completed.
4222 rcutree.sysrq_rcu= [KNL]
4223 Commandeer a sysrq key to dump out Tree RCU's
4224 rcu_node tree with an eye towards determining
4225 why a new grace period has not yet started.
4227 rcuscale.gp_async= [KNL]
4228 Measure performance of asynchronous
4229 grace-period primitives such as call_rcu().
4231 rcuscale.gp_async_max= [KNL]
4232 Specify the maximum number of outstanding
4233 callbacks per writer thread. When a writer
4234 thread exceeds this limit, it invokes the
4235 corresponding flavor of rcu_barrier() to allow
4236 previously posted callbacks to drain.
4238 rcuscale.gp_exp= [KNL]
4239 Measure performance of expedited synchronous
4240 grace-period primitives.
4242 rcuscale.holdoff= [KNL]
4243 Set test-start holdoff period. The purpose of
4244 this parameter is to delay the start of the
4245 test until boot completes in order to avoid
4248 rcuscale.kfree_rcu_test= [KNL]
4249 Set to measure performance of kfree_rcu() flooding.
4251 rcuscale.kfree_nthreads= [KNL]
4252 The number of threads running loops of kfree_rcu().
4254 rcuscale.kfree_alloc_num= [KNL]
4255 Number of allocations and frees done in an iteration.
4257 rcuscale.kfree_loops= [KNL]
4258 Number of loops doing rcuscale.kfree_alloc_num number
4259 of allocations and frees.
4261 rcuscale.nreaders= [KNL]
4262 Set number of RCU readers. The value -1 selects
4263 N, where N is the number of CPUs. A value
4264 "n" less than -1 selects N-n+1, where N is again
4265 the number of CPUs. For example, -2 selects N
4266 (the number of CPUs), -3 selects N+1, and so on.
4267 A value of "n" less than or equal to -N selects
4270 rcuscale.nwriters= [KNL]
4271 Set number of RCU writers. The values operate
4272 the same as for rcuscale.nreaders.
4273 N, where N is the number of CPUs
4275 rcuscale.perf_type= [KNL]
4276 Specify the RCU implementation to test.
4278 rcuscale.shutdown= [KNL]
4279 Shut the system down after performance tests
4280 complete. This is useful for hands-off automated
4283 rcuscale.verbose= [KNL]
4284 Enable additional printk() statements.
4286 rcuscale.writer_holdoff= [KNL]
4287 Write-side holdoff between grace periods,
4288 in microseconds. The default of zero says
4291 rcutorture.fqs_duration= [KNL]
4292 Set duration of force_quiescent_state bursts
4295 rcutorture.fqs_holdoff= [KNL]
4296 Set holdoff time within force_quiescent_state bursts
4299 rcutorture.fqs_stutter= [KNL]
4300 Set wait time between force_quiescent_state bursts
4303 rcutorture.fwd_progress= [KNL]
4304 Enable RCU grace-period forward-progress testing
4305 for the types of RCU supporting this notion.
4307 rcutorture.fwd_progress_div= [KNL]
4308 Specify the fraction of a CPU-stall-warning
4309 period to do tight-loop forward-progress testing.
4311 rcutorture.fwd_progress_holdoff= [KNL]
4312 Number of seconds to wait between successive
4313 forward-progress tests.
4315 rcutorture.fwd_progress_need_resched= [KNL]
4316 Enclose cond_resched() calls within checks for
4317 need_resched() during tight-loop forward-progress
4320 rcutorture.gp_cond= [KNL]
4321 Use conditional/asynchronous update-side
4322 primitives, if available.
4324 rcutorture.gp_exp= [KNL]
4325 Use expedited update-side primitives, if available.
4327 rcutorture.gp_normal= [KNL]
4328 Use normal (non-expedited) asynchronous
4329 update-side primitives, if available.
4331 rcutorture.gp_sync= [KNL]
4332 Use normal (non-expedited) synchronous
4333 update-side primitives, if available. If all
4334 of rcutorture.gp_cond=, rcutorture.gp_exp=,
4335 rcutorture.gp_normal=, and rcutorture.gp_sync=
4336 are zero, rcutorture acts as if is interpreted
4337 they are all non-zero.
4339 rcutorture.irqreader= [KNL]
4340 Run RCU readers from irq handlers, or, more
4341 accurately, from a timer handler. Not all RCU
4342 flavors take kindly to this sort of thing.
4344 rcutorture.leakpointer= [KNL]
4345 Leak an RCU-protected pointer out of the reader.
4346 This can of course result in splats, and is
4347 intended to test the ability of things like
4348 CONFIG_RCU_STRICT_GRACE_PERIOD=y to detect
4351 rcutorture.n_barrier_cbs= [KNL]
4352 Set callbacks/threads for rcu_barrier() testing.
4354 rcutorture.nfakewriters= [KNL]
4355 Set number of concurrent RCU writers. These just
4356 stress RCU, they don't participate in the actual
4357 test, hence the "fake".
4359 rcutorture.nocbs_nthreads= [KNL]
4360 Set number of RCU callback-offload togglers.
4361 Zero (the default) disables toggling.
4363 rcutorture.nocbs_toggle= [KNL]
4364 Set the delay in milliseconds between successive
4365 callback-offload toggling attempts.
4367 rcutorture.nreaders= [KNL]
4368 Set number of RCU readers. The value -1 selects
4369 N-1, where N is the number of CPUs. A value
4370 "n" less than -1 selects N-n-2, where N is again
4371 the number of CPUs. For example, -2 selects N
4372 (the number of CPUs), -3 selects N+1, and so on.
4374 rcutorture.object_debug= [KNL]
4375 Enable debug-object double-call_rcu() testing.
4377 rcutorture.onoff_holdoff= [KNL]
4378 Set time (s) after boot for CPU-hotplug testing.
4380 rcutorture.onoff_interval= [KNL]
4381 Set time (jiffies) between CPU-hotplug operations,
4382 or zero to disable CPU-hotplug testing.
4384 rcutorture.read_exit= [KNL]
4385 Set the number of read-then-exit kthreads used
4386 to test the interaction of RCU updaters and
4387 task-exit processing.
4389 rcutorture.read_exit_burst= [KNL]
4390 The number of times in a given read-then-exit
4391 episode that a set of read-then-exit kthreads
4394 rcutorture.read_exit_delay= [KNL]
4395 The delay, in seconds, between successive
4396 read-then-exit testing episodes.
4398 rcutorture.shuffle_interval= [KNL]
4399 Set task-shuffle interval (s). Shuffling tasks
4400 allows some CPUs to go into dyntick-idle mode
4401 during the rcutorture test.
4403 rcutorture.shutdown_secs= [KNL]
4404 Set time (s) after boot system shutdown. This
4405 is useful for hands-off automated testing.
4407 rcutorture.stall_cpu= [KNL]
4408 Duration of CPU stall (s) to test RCU CPU stall
4409 warnings, zero to disable.
4411 rcutorture.stall_cpu_block= [KNL]
4412 Sleep while stalling if set. This will result
4413 in warnings from preemptible RCU in addition
4414 to any other stall-related activity.
4416 rcutorture.stall_cpu_holdoff= [KNL]
4417 Time to wait (s) after boot before inducing stall.
4419 rcutorture.stall_cpu_irqsoff= [KNL]
4420 Disable interrupts while stalling if set.
4422 rcutorture.stall_gp_kthread= [KNL]
4423 Duration (s) of forced sleep within RCU
4424 grace-period kthread to test RCU CPU stall
4425 warnings, zero to disable. If both stall_cpu
4426 and stall_gp_kthread are specified, the
4427 kthread is starved first, then the CPU.
4429 rcutorture.stat_interval= [KNL]
4430 Time (s) between statistics printk()s.
4432 rcutorture.stutter= [KNL]
4433 Time (s) to stutter testing, for example, specifying
4434 five seconds causes the test to run for five seconds,
4435 wait for five seconds, and so on. This tests RCU's
4436 ability to transition abruptly to and from idle.
4438 rcutorture.test_boost= [KNL]
4439 Test RCU priority boosting? 0=no, 1=maybe, 2=yes.
4440 "Maybe" means test if the RCU implementation
4441 under test support RCU priority boosting.
4443 rcutorture.test_boost_duration= [KNL]
4444 Duration (s) of each individual boost test.
4446 rcutorture.test_boost_interval= [KNL]
4447 Interval (s) between each boost test.
4449 rcutorture.test_no_idle_hz= [KNL]
4450 Test RCU's dyntick-idle handling. See also the
4451 rcutorture.shuffle_interval parameter.
4453 rcutorture.torture_type= [KNL]
4454 Specify the RCU implementation to test.
4456 rcutorture.verbose= [KNL]
4457 Enable additional printk() statements.
4459 rcupdate.rcu_cpu_stall_ftrace_dump= [KNL]
4460 Dump ftrace buffer after reporting RCU CPU
4463 rcupdate.rcu_cpu_stall_suppress= [KNL]
4464 Suppress RCU CPU stall warning messages.
4466 rcupdate.rcu_cpu_stall_suppress_at_boot= [KNL]
4467 Suppress RCU CPU stall warning messages and
4468 rcutorture writer stall warnings that occur
4469 during early boot, that is, during the time
4470 before the init task is spawned.
4472 rcupdate.rcu_cpu_stall_timeout= [KNL]
4473 Set timeout for RCU CPU stall warning messages.
4475 rcupdate.rcu_expedited= [KNL]
4476 Use expedited grace-period primitives, for
4477 example, synchronize_rcu_expedited() instead
4478 of synchronize_rcu(). This reduces latency,
4479 but can increase CPU utilization, degrade
4480 real-time latency, and degrade energy efficiency.
4481 No effect on CONFIG_TINY_RCU kernels.
4483 rcupdate.rcu_normal= [KNL]
4484 Use only normal grace-period primitives,
4485 for example, synchronize_rcu() instead of
4486 synchronize_rcu_expedited(). This improves
4487 real-time latency, CPU utilization, and
4488 energy efficiency, but can expose users to
4489 increased grace-period latency. This parameter
4490 overrides rcupdate.rcu_expedited. No effect on
4491 CONFIG_TINY_RCU kernels.
4493 rcupdate.rcu_normal_after_boot= [KNL]
4494 Once boot has completed (that is, after
4495 rcu_end_inkernel_boot() has been invoked), use
4496 only normal grace-period primitives. No effect
4497 on CONFIG_TINY_RCU kernels.
4499 But note that CONFIG_PREEMPT_RT=y kernels enables
4500 this kernel boot parameter, forcibly setting
4501 it to the value one, that is, converting any
4502 post-boot attempt at an expedited RCU grace
4503 period to instead use normal non-expedited
4504 grace-period processing.
4506 rcupdate.rcu_task_ipi_delay= [KNL]
4507 Set time in jiffies during which RCU tasks will
4508 avoid sending IPIs, starting with the beginning
4509 of a given grace period. Setting a large
4510 number avoids disturbing real-time workloads,
4511 but lengthens grace periods.
4513 rcupdate.rcu_task_stall_timeout= [KNL]
4514 Set timeout in jiffies for RCU task stall warning
4515 messages. Disable with a value less than or equal
4518 rcupdate.rcu_self_test= [KNL]
4519 Run the RCU early boot self tests
4523 Run specified binary instead of /init from the ramdisk,
4524 used for early userspace startup. See initrd.
4527 force - Override the decision by the kernel to hide the
4528 advertisement of RDRAND support (this affects
4529 certain AMD processors because of buggy BIOS
4530 support, specifically around the suspend/resume
4534 Turn on/off individual RDT features. List is:
4535 cmt, mbmtotal, mbmlocal, l3cat, l3cdp, l2cat, l2cdp,
4537 E.g. to turn on cmt and turn off mba use:
4541 Format (x86 or x86_64):
4542 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] \
4544 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \
4546 Where reboot_mode is one of warm (soft) or cold (hard) or gpio
4547 (prefix with 'panic_' to set mode for panic
4549 reboot_type is one of bios, acpi, kbd, triple, efi, or pci,
4550 reboot_force is either force or not specified,
4551 reboot_cpu is s[mp]#### with #### being the processor
4552 to be used for rebooting.
4554 refscale.holdoff= [KNL]
4555 Set test-start holdoff period. The purpose of
4556 this parameter is to delay the start of the
4557 test until boot completes in order to avoid
4560 refscale.loops= [KNL]
4561 Set the number of loops over the synchronization
4562 primitive under test. Increasing this number
4563 reduces noise due to loop start/end overhead,
4564 but the default has already reduced the per-pass
4565 noise to a handful of picoseconds on ca. 2020
4568 refscale.nreaders= [KNL]
4569 Set number of readers. The default value of -1
4570 selects N, where N is roughly 75% of the number
4571 of CPUs. A value of zero is an interesting choice.
4573 refscale.nruns= [KNL]
4574 Set number of runs, each of which is dumped onto
4577 refscale.readdelay= [KNL]
4578 Set the read-side critical-section duration,
4579 measured in microseconds.
4581 refscale.scale_type= [KNL]
4582 Specify the read-protection implementation to test.
4584 refscale.shutdown= [KNL]
4585 Shut down the system at the end of the performance
4586 test. This defaults to 1 (shut it down) when
4587 refscale is built into the kernel and to 0 (leave
4588 it running) when refscale is built as a module.
4590 refscale.verbose= [KNL]
4591 Enable additional printk() statements.
4593 refscale.verbose_batched= [KNL]
4594 Batch the additional printk() statements. If zero
4595 (the default) or negative, print everything. Otherwise,
4596 print every Nth verbose statement, where N is the value
4600 [KNL, SMP] Set scheduler's default relax_domain_level.
4601 See Documentation/admin-guide/cgroup-v1/cpusets.rst.
4603 reserve= [KNL,BUGS] Force kernel to ignore I/O ports or memory
4604 Format: <base1>,<size1>[,<base2>,<size2>,...]
4605 Reserve I/O ports or memory so the kernel won't use
4606 them. If <base> is less than 0x10000, the region
4607 is assumed to be I/O ports; otherwise it is memory.
4609 reservetop= [X86-32]
4611 Reserves a hole at the top of the kernel virtual
4616 Set the amount of memory to reserve for BIOS at
4617 the bottom of the address space.
4619 reset_devices [KNL] Force drivers to reset the underlying device
4620 during initialization.
4623 Specify the partition device for software suspend
4625 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>}
4627 resume_offset= [SWSUSP]
4628 Specify the offset from the beginning of the partition
4629 given by "resume=" at which the swap header is located,
4630 in <PAGE_SIZE> units (needed only for swap files).
4631 See Documentation/power/swsusp-and-swap-files.rst
4633 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to
4634 read the resume files
4636 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up.
4637 Useful for devices that are detected asynchronously
4638 (e.g. USB and MMC devices).
4640 hibernate= [HIBERNATION]
4641 noresume Don't check if there's a hibernation image
4642 present during boot.
4643 nocompress Don't compress/decompress hibernation images.
4644 no Disable hibernation and resume.
4645 protect_image Turn on image protection during restoration
4646 (that will set all pages holding image data
4647 during restoration read-only).
4649 retain_initrd [RAM] Keep initrd memory after extraction
4651 rfkill.default_state=
4652 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm,
4653 etc. communication is blocked by default.
4656 rfkill.master_switch_mode=
4657 0 The "airplane mode" button does nothing.
4658 1 The "airplane mode" button toggles between everything
4659 blocked and the previous configuration.
4660 2 The "airplane mode" button toggles between everything
4661 blocked and everything unblocked.
4663 rhash_entries= [KNL,NET]
4664 Set number of hash buckets for route cache
4667 [KNL] Disable ring 3 MONITOR/MWAIT feature on supported
4670 ro [KNL] Mount root device read-only on boot
4673 on Mark read-only kernel memory as read-only (default).
4674 off Leave read-only kernel memory writable for debugging.
4677 Enable the uart passthrough on the designated usb port
4678 on Rockchip SoCs. When active, the signals of the
4679 debug-uart get routed to the D+ and D- pins of the usb
4680 port and the regular usb controller gets disabled.
4682 root= [KNL] Root filesystem
4683 See name_to_dev_t comment in init/do_mounts.c.
4685 rootdelay= [KNL] Delay (in seconds) to pause before attempting to
4686 mount the root filesystem
4688 rootflags= [KNL] Set root filesystem mount option string
4690 rootfstype= [KNL] Set root filesystem type
4692 rootwait [KNL] Wait (indefinitely) for root device to show up.
4693 Useful for devices that are detected asynchronously
4694 (e.g. USB and MMC devices).
4696 rproc_mem=nn[KMG][@address]
4697 [KNL,ARM,CMA] Remoteproc physical memory block.
4698 Memory area to be used by remote processor image,
4701 rw [KNL] Mount root device read-write on boot
4703 S [KNL] Run init in single mode
4705 s390_iommu= [HW,S390]
4706 Set s390 IOTLB flushing mode
4708 With strict flushing every unmap operation will result in
4709 an IOTLB flush. Default is lazy flushing before reuse,
4713 See drivers/net/irda/sa1100_ir.c.
4715 sbni= [NET] Granch SBNI12 leased line adapter
4717 sched_debug [KNL] Enables verbose scheduler debug messages.
4719 schedstats= [KNL,X86] Enable or disable scheduled statistics.
4720 Allowed values are enable and disable. This feature
4721 incurs a small amount of overhead in the scheduler
4722 but is useful for debugging and performance tuning.
4724 sched_thermal_decay_shift=
4725 [KNL, SMP] Set a decay shift for scheduler thermal
4726 pressure signal. Thermal pressure signal follows the
4727 default decay period of other scheduler pelt
4728 signals(usually 32 ms but configurable). Setting
4729 sched_thermal_decay_shift will left shift the decay
4730 period for the thermal pressure signal by the shift
4732 i.e. with the default pelt decay period of 32 ms
4733 sched_thermal_decay_shift thermal pressure decay pr
4737 Format: integer between 0 and 10
4740 scftorture.holdoff= [KNL]
4741 Number of seconds to hold off before starting
4742 test. Defaults to zero for module insertion and
4743 to 10 seconds for built-in smp_call_function()
4746 scftorture.longwait= [KNL]
4747 Request ridiculously long waits randomly selected
4748 up to the chosen limit in seconds. Zero (the
4749 default) disables this feature. Please note
4750 that requesting even small non-zero numbers of
4751 seconds can result in RCU CPU stall warnings,
4752 softlockup complaints, and so on.
4754 scftorture.nthreads= [KNL]
4755 Number of kthreads to spawn to invoke the
4756 smp_call_function() family of functions.
4757 The default of -1 specifies a number of kthreads
4758 equal to the number of CPUs.
4760 scftorture.onoff_holdoff= [KNL]
4761 Number seconds to wait after the start of the
4762 test before initiating CPU-hotplug operations.
4764 scftorture.onoff_interval= [KNL]
4765 Number seconds to wait between successive
4766 CPU-hotplug operations. Specifying zero (which
4767 is the default) disables CPU-hotplug operations.
4769 scftorture.shutdown_secs= [KNL]
4770 The number of seconds following the start of the
4771 test after which to shut down the system. The
4772 default of zero avoids shutting down the system.
4773 Non-zero values are useful for automated tests.
4775 scftorture.stat_interval= [KNL]
4776 The number of seconds between outputting the
4777 current test statistics to the console. A value
4778 of zero disables statistics output.
4780 scftorture.stutter_cpus= [KNL]
4781 The number of jiffies to wait between each change
4782 to the set of CPUs under test.
4784 scftorture.use_cpus_read_lock= [KNL]
4785 Use use_cpus_read_lock() instead of the default
4786 preempt_disable() to disable CPU hotplug
4787 while invoking one of the smp_call_function*()
4790 scftorture.verbose= [KNL]
4791 Enable additional printk() statements.
4793 scftorture.weight_single= [KNL]
4794 The probability weighting to use for the
4795 smp_call_function_single() function with a zero
4796 "wait" parameter. A value of -1 selects the
4797 default if all other weights are -1. However,
4798 if at least one weight has some other value, a
4799 value of -1 will instead select a weight of zero.
4801 scftorture.weight_single_wait= [KNL]
4802 The probability weighting to use for the
4803 smp_call_function_single() function with a
4804 non-zero "wait" parameter. See weight_single.
4806 scftorture.weight_many= [KNL]
4807 The probability weighting to use for the
4808 smp_call_function_many() function with a zero
4809 "wait" parameter. See weight_single.
4810 Note well that setting a high probability for
4811 this weighting can place serious IPI load
4814 scftorture.weight_many_wait= [KNL]
4815 The probability weighting to use for the
4816 smp_call_function_many() function with a
4817 non-zero "wait" parameter. See weight_single
4820 scftorture.weight_all= [KNL]
4821 The probability weighting to use for the
4822 smp_call_function_all() function with a zero
4823 "wait" parameter. See weight_single and
4826 scftorture.weight_all_wait= [KNL]
4827 The probability weighting to use for the
4828 smp_call_function_all() function with a
4829 non-zero "wait" parameter. See weight_single
4832 skew_tick= [KNL] Offset the periodic timer tick per cpu to mitigate
4833 xtime_lock contention on larger systems, and/or RCU lock
4834 contention on all systems with CONFIG_MAXSMP set.
4835 Format: { "0" | "1" }
4836 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1"
4838 Note: increases power consumption, thus should only be
4839 enabled if running jitter sensitive (HPC/RT) workloads.
4841 security= [SECURITY] Choose a legacy "major" security module to
4842 enable at boot. This has been deprecated by the
4845 selinux= [SELINUX] Disable or enable SELinux at boot time.
4846 Format: { "0" | "1" }
4847 See security/selinux/Kconfig help text.
4852 apparmor= [APPARMOR] Disable or enable AppArmor at boot time
4853 Format: { "0" | "1" }
4854 See security/apparmor/Kconfig help text
4857 Default value is set via kernel config option.
4859 serialnumber [BUGS=X86-32]
4862 Maximal number of shapers.
4870 Disable merging of slabs with similar size. May be
4871 necessary if there is some reason to distinguish
4872 allocs to different slabs, especially in hardened
4873 environments where the risk of heap overflows and
4874 layout control by attackers can usually be
4875 frustrated by disabling merging. This will reduce
4876 most of the exposure of a heap attack to a single
4877 cache (risks via metadata attacks are mostly
4878 unchanged). Debug options disable merging on their
4880 For more information see Documentation/vm/slub.rst.
4882 slab_max_order= [MM, SLAB]
4883 Determines the maximum allowed order for slabs.
4884 A high setting may cause OOMs due to memory
4885 fragmentation. Defaults to 1 for systems with
4886 more than 32MB of RAM, 0 otherwise.
4888 slub_debug[=options[,slabs][;[options[,slabs]]...] [MM, SLUB]
4889 Enabling slub_debug allows one to determine the
4890 culprit if slab objects become corrupted. Enabling
4891 slub_debug can create guard zones around objects and
4892 may poison objects when not in use. Also tracks the
4893 last alloc / free. For more information see
4894 Documentation/vm/slub.rst.
4896 slub_memcg_sysfs= [MM, SLUB]
4897 Determines whether to enable sysfs directories for
4898 memory cgroup sub-caches. 1 to enable, 0 to disable.
4899 The default is determined by CONFIG_SLUB_MEMCG_SYSFS_ON.
4900 Enabling this can lead to a very high number of debug
4901 directories and files being created under
4904 slub_max_order= [MM, SLUB]
4905 Determines the maximum allowed order for slabs.
4906 A high setting may cause OOMs due to memory
4907 fragmentation. For more information see
4908 Documentation/vm/slub.rst.
4910 slub_min_objects= [MM, SLUB]
4911 The minimum number of objects per slab. SLUB will
4912 increase the slab order up to slub_max_order to
4913 generate a sufficiently large slab able to contain
4914 the number of objects indicated. The higher the number
4915 of objects the smaller the overhead of tracking slabs
4916 and the less frequently locks need to be acquired.
4917 For more information see Documentation/vm/slub.rst.
4919 slub_min_order= [MM, SLUB]
4920 Determines the minimum page order for slabs. Must be
4921 lower than slub_max_order.
4922 For more information see Documentation/vm/slub.rst.
4924 slub_nomerge [MM, SLUB]
4925 Same with slab_nomerge. This is supported for legacy.
4926 See slab_nomerge for more information.
4929 Format: <io1>[,<io2>[,...,<io8>]]
4931 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices
4932 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port
4933 smsc-ircc2.ircc_sir= [HW] SIR base I/O port
4934 smsc-ircc2.ircc_fir= [HW] FIR base I/O port
4935 smsc-ircc2.ircc_irq= [HW] IRQ line
4936 smsc-ircc2.ircc_dma= [HW] DMA channel
4937 smsc-ircc2.ircc_transceiver= [HW] Transceiver type:
4938 0: Toshiba Satellite 1800 (GP data pin select)
4939 1: Fast pin select (default)
4942 smt [KNL,S390] Set the maximum number of threads (logical
4943 CPUs) to use per physical CPU on systems capable of
4944 symmetric multithreading (SMT). Will be capped to the
4945 actual hardware limit.
4947 Default: -1 (no limit)
4950 [KNL] Should the soft-lockup detector generate panics.
4953 A value of 1 instructs the soft-lockup detector
4954 to panic the machine when a soft-lockup occurs. It is
4955 also controlled by the kernel.softlockup_panic sysctl
4956 and CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC, which is the
4957 respective build-time switch to that functionality.
4959 softlockup_all_cpu_backtrace=
4960 [KNL] Should the soft-lockup detector generate
4961 backtraces on all cpus.
4964 sonypi.*= [HW] Sony Programmable I/O Control Device driver
4965 See Documentation/admin-guide/laptops/sonypi.rst
4967 spectre_v2= [X86] Control mitigation of Spectre variant 2
4968 (indirect branch speculation) vulnerability.
4969 The default operation protects the kernel from
4972 on - unconditionally enable, implies
4974 off - unconditionally disable, implies
4976 auto - kernel detects whether your CPU model is
4979 Selecting 'on' will, and 'auto' may, choose a
4980 mitigation method at run time according to the
4981 CPU, the available microcode, the setting of the
4982 CONFIG_RETPOLINE configuration option, and the
4983 compiler with which the kernel was built.
4985 Selecting 'on' will also enable the mitigation
4986 against user space to user space task attacks.
4988 Selecting 'off' will disable both the kernel and
4989 the user space protections.
4991 Specific mitigations can also be selected manually:
4993 retpoline - replace indirect branches
4994 retpoline,generic - google's original retpoline
4995 retpoline,amd - AMD-specific minimal thunk
4997 Not specifying this option is equivalent to
5001 [X86] Control mitigation of Spectre variant 2
5002 (indirect branch speculation) vulnerability between
5005 on - Unconditionally enable mitigations. Is
5006 enforced by spectre_v2=on
5008 off - Unconditionally disable mitigations. Is
5009 enforced by spectre_v2=off
5011 prctl - Indirect branch speculation is enabled,
5012 but mitigation can be enabled via prctl
5013 per thread. The mitigation control state
5014 is inherited on fork.
5017 - Like "prctl" above, but only STIBP is
5018 controlled per thread. IBPB is issued
5019 always when switching between different user
5023 - Same as "prctl" above, but all seccomp
5024 threads will enable the mitigation unless
5025 they explicitly opt out.
5028 - Like "seccomp" above, but only STIBP is
5029 controlled per thread. IBPB is issued
5030 always when switching between different
5031 user space processes.
5033 auto - Kernel selects the mitigation depending on
5034 the available CPU features and vulnerability.
5037 If CONFIG_SECCOMP=y then "seccomp", otherwise "prctl"
5039 Not specifying this option is equivalent to
5040 spectre_v2_user=auto.
5042 spec_store_bypass_disable=
5043 [HW] Control Speculative Store Bypass (SSB) Disable mitigation
5044 (Speculative Store Bypass vulnerability)
5046 Certain CPUs are vulnerable to an exploit against a
5047 a common industry wide performance optimization known
5048 as "Speculative Store Bypass" in which recent stores
5049 to the same memory location may not be observed by
5050 later loads during speculative execution. The idea
5051 is that such stores are unlikely and that they can
5052 be detected prior to instruction retirement at the
5053 end of a particular speculation execution window.
5055 In vulnerable processors, the speculatively forwarded
5056 store can be used in a cache side channel attack, for
5057 example to read memory to which the attacker does not
5058 directly have access (e.g. inside sandboxed code).
5060 This parameter controls whether the Speculative Store
5061 Bypass optimization is used.
5063 On x86 the options are:
5065 on - Unconditionally disable Speculative Store Bypass
5066 off - Unconditionally enable Speculative Store Bypass
5067 auto - Kernel detects whether the CPU model contains an
5068 implementation of Speculative Store Bypass and
5069 picks the most appropriate mitigation. If the
5070 CPU is not vulnerable, "off" is selected. If the
5071 CPU is vulnerable the default mitigation is
5072 architecture and Kconfig dependent. See below.
5073 prctl - Control Speculative Store Bypass per thread
5074 via prctl. Speculative Store Bypass is enabled
5075 for a process by default. The state of the control
5076 is inherited on fork.
5077 seccomp - Same as "prctl" above, but all seccomp threads
5078 will disable SSB unless they explicitly opt out.
5080 Default mitigations:
5081 X86: If CONFIG_SECCOMP=y "seccomp", otherwise "prctl"
5083 On powerpc the options are:
5085 on,auto - On Power8 and Power9 insert a store-forwarding
5086 barrier on kernel entry and exit. On Power7
5087 perform a software flush on kernel entry and
5091 Not specifying this option is equivalent to
5092 spec_store_bypass_disable=auto.
5094 spia_io_base= [HW,MTD]
5100 [X86] Enable split lock detection
5102 When enabled (and if hardware support is present), atomic
5103 instructions that access data across cache line
5104 boundaries will result in an alignment check exception.
5108 warn - the kernel will emit rate limited warnings
5109 about applications triggering the #AC
5110 exception. This mode is the default on CPUs
5111 that supports split lock detection.
5113 fatal - the kernel will send SIGBUS to applications
5114 that trigger the #AC exception.
5116 If an #AC exception is hit in the kernel or in
5117 firmware (i.e. not while executing in user mode)
5118 the kernel will oops in either "warn" or "fatal"
5122 Control the Special Register Buffer Data Sampling
5125 Certain CPUs are vulnerable to an MDS-like
5126 exploit which can leak bits from the random
5129 By default, this issue is mitigated by
5130 microcode. However, the microcode fix can cause
5131 the RDRAND and RDSEED instructions to become
5132 much slower. Among other effects, this will
5133 result in reduced throughput from /dev/urandom.
5135 The microcode mitigation can be disabled with
5136 the following option:
5138 off: Disable mitigation and remove
5139 performance impact to RDRAND and RDSEED
5141 srcutree.counter_wrap_check [KNL]
5142 Specifies how frequently to check for
5143 grace-period sequence counter wrap for the
5144 srcu_data structure's ->srcu_gp_seq_needed field.
5145 The greater the number of bits set in this kernel
5146 parameter, the less frequently counter wrap will
5147 be checked for. Note that the bottom two bits
5150 srcutree.exp_holdoff [KNL]
5151 Specifies how many nanoseconds must elapse
5152 since the end of the last SRCU grace period for
5153 a given srcu_struct until the next normal SRCU
5154 grace period will be considered for automatic
5155 expediting. Set to zero to disable automatic
5159 Speculative Store Bypass Disable control
5161 On CPUs that are vulnerable to the Speculative
5162 Store Bypass vulnerability and offer a
5163 firmware based mitigation, this parameter
5164 indicates how the mitigation should be used:
5166 force-on: Unconditionally enable mitigation for
5167 for both kernel and userspace
5168 force-off: Unconditionally disable mitigation for
5169 for both kernel and userspace
5170 kernel: Always enable mitigation in the
5171 kernel, and offer a prctl interface
5172 to allow userspace to register its
5173 interest in being mitigated too.
5175 stack_guard_gap= [MM]
5176 override the default stack gap protection. The value
5177 is in page units and it defines how many pages prior
5178 to (for stacks growing down) resp. after (for stacks
5179 growing up) the main stack are reserved for no other
5180 mapping. Default value is 256 pages.
5183 Enabled the stack tracer on boot up.
5185 stacktrace_filter=[function-list]
5186 [FTRACE] Limit the functions that the stack tracer
5187 will trace at boot up. function-list is a comma-separated
5188 list of functions. This list can be changed at run
5189 time by the stack_trace_filter file in the debugfs
5190 tracing directory. Note, this enables stack tracing
5191 and the stacktrace above is not needed.
5195 Set the STI (builtin display/keyboard on the HP-PARISC
5196 machines) console (graphic card) which should be used
5197 as the initial boot-console.
5198 See also comment in drivers/video/console/sticore.c.
5201 See comment in drivers/video/console/sticore.c.
5204 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]]
5206 sunrpc.min_resvport=
5207 sunrpc.max_resvport=
5209 SunRPC servers often require that client requests
5210 originate from a privileged port (i.e. a port in the
5211 range 0 < portnr < 1024).
5212 An administrator who wishes to reserve some of these
5213 ports for other uses may adjust the range that the
5214 kernel's sunrpc client considers to be privileged
5215 using these two parameters to set the minimum and
5216 maximum port values.
5218 sunrpc.svc_rpc_per_connection_limit=
5220 Limit the number of requests that the server will
5221 process in parallel from a single connection.
5222 The default value is 0 (no limit).
5226 Control how the NFS server code allocates CPUs to
5227 service thread pools. Depending on how many NICs
5228 you have and where their interrupts are bound, this
5229 option will affect which CPUs will do NFS serving.
5230 Note: this parameter cannot be changed while the
5231 NFS server is running.
5233 auto the server chooses an appropriate mode
5234 automatically using heuristics
5235 global a single global pool contains all CPUs
5236 percpu one pool for each CPU
5237 pernode one pool for each NUMA node (equivalent
5238 to global on non-NUMA machines)
5240 sunrpc.tcp_slot_table_entries=
5241 sunrpc.udp_slot_table_entries=
5243 Sets the upper limit on the number of simultaneous
5244 RPC calls that can be sent from the client to a
5245 server. Increasing these values may allow you to
5246 improve throughput, but will also increase the
5247 amount of memory reserved for use by the client.
5249 suspend.pm_test_delay=
5251 Sets the number of seconds to remain in a suspend test
5252 mode before resuming the system (see
5253 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG
5254 is set. Default value is 5.
5257 Format: { on | off | y | n | 1 | 0 }
5258 This parameter controls use of the Protected
5259 Execution Facility on pSeries.
5262 [KNL] Enable accounting of swap in memory resource
5263 controller if no parameter or 1 is given or disable
5264 it if 0 is given (See Documentation/admin-guide/cgroup-v1/memory.rst)
5266 swiotlb= [ARM,IA-64,PPC,MIPS,X86]
5267 Format: { <int> | force | noforce }
5268 <int> -- Number of I/O TLB slabs
5269 force -- force using of bounce buffers even if they
5270 wouldn't be automatically used by the kernel
5271 noforce -- Never use bounce buffers (for debugging)
5276 Set a sysctl parameter, right before loading the init
5277 process, as if the value was written to the respective
5278 /proc/sys/... file. Both '.' and '/' are recognized as
5279 separators. Unrecognized parameters and invalid values
5280 are reported in the kernel log. Sysctls registered
5281 later by a loaded module cannot be set this way.
5282 Example: sysctl.vm.swappiness=40
5284 sysfs.deprecated=0|1 [KNL]
5285 Enable/disable old style sysfs layout for old udev
5286 on older distributions. When this option is enabled
5287 very new udev will not work anymore. When this option
5288 is disabled (or CONFIG_SYSFS_DEPRECATED not compiled)
5289 in older udev will not work anymore.
5290 Default depends on CONFIG_SYSFS_DEPRECATED_V2 set in
5291 the kernel configuration.
5293 sysrq_always_enabled
5295 Ignore sysrq setting - this boot parameter will
5296 neutralize any effect of /proc/sys/kernel/sysrq.
5297 Useful for debugging.
5299 tcpmhash_entries= [KNL,NET]
5300 Set the number of tcp_metrics_hash slots.
5301 Default value is 8192 or 16384 depending on total
5302 ram pages. This is used to specify the TCP metrics
5303 cache size. See Documentation/networking/ip-sysctl.rst
5304 "tcp_no_metrics_save" section for more details.
5308 test_suspend= [SUSPEND][,N]
5309 Specify "mem" (for Suspend-to-RAM) or "standby" (for
5310 standby suspend) or "freeze" (for suspend type freeze)
5311 as the system sleep state during system startup with
5312 the optional capability to repeat N number of times.
5313 The system is woken from this state using a
5314 wakeup-capable RTC alarm.
5316 thash_entries= [KNL,NET]
5317 Set number of hash buckets for TCP connection
5319 thermal.act= [HW,ACPI]
5320 -1: disable all active trip points in all thermal zones
5321 <degrees C>: override all lowest active trip points
5323 thermal.crt= [HW,ACPI]
5324 -1: disable all critical trip points in all thermal zones
5325 <degrees C>: override all critical trip points
5327 thermal.nocrt= [HW,ACPI]
5328 Set to disable actions on ACPI thermal zone
5329 critical and hot trip points.
5331 thermal.off= [HW,ACPI]
5332 1: disable ACPI thermal control
5334 thermal.psv= [HW,ACPI]
5335 -1: disable all passive trip points
5336 <degrees C>: override all passive trip points to this
5339 thermal.tzp= [HW,ACPI]
5340 Specify global default ACPI thermal zone polling rate
5341 <deci-seconds>: poll all this frequency
5342 0: no polling (default)
5345 Force threading of all interrupt handlers except those
5346 marked explicitly IRQF_NO_THREAD.
5350 Specify if the kernel should make use of the cpu
5351 topology information if the hardware supports this.
5352 The scheduler will make use of this information and
5353 e.g. base its process migration decisions on it.
5356 topology_updates= [KNL, PPC, NUMA]
5358 Specify if the kernel should ignore (off)
5359 topology updates sent by the hypervisor to this
5362 torture.disable_onoff_at_boot= [KNL]
5363 Prevent the CPU-hotplug component of torturing
5364 until after init has spawned.
5366 torture.ftrace_dump_at_shutdown= [KNL]
5367 Dump the ftrace buffer at torture-test shutdown,
5368 even if there were no errors. This can be a
5369 very costly operation when many torture tests
5370 are running concurrently, especially on systems
5371 with rotating-rust storage.
5373 torture.verbose_sleep_frequency= [KNL]
5374 Specifies how many verbose printk()s should be
5375 emitted between each sleep. The default of zero
5376 disables verbose-printk() sleeping.
5378 torture.verbose_sleep_duration= [KNL]
5379 Duration of each verbose-printk() sleep in jiffies.
5383 tpm_suspend_pcr=[HW,TPM]
5384 Format: integer pcr id
5385 Specify that at suspend time, the tpm driver
5386 should extend the specified pcr with zeros,
5387 as a workaround for some chips which fail to
5388 flush the last written pcr on TPM_SaveState.
5389 This will guarantee that all the other pcrs
5392 trace_buf_size=nn[KMG]
5393 [FTRACE] will set tracing buffer size on each cpu.
5395 trace_event=[event-list]
5396 [FTRACE] Set and start specified trace events in order
5397 to facilitate early boot debugging. The event-list is a
5398 comma-separated list of trace events to enable. See
5399 also Documentation/trace/events.rst
5401 trace_options=[option-list]
5402 [FTRACE] Enable or disable tracer options at boot.
5403 The option-list is a comma delimited list of options
5404 that can be enabled or disabled just as if you were
5405 to echo the option name into
5407 /sys/kernel/debug/tracing/trace_options
5409 For example, to enable stacktrace option (to dump the
5410 stack trace of each event), add to the command line:
5412 trace_options=stacktrace
5414 See also Documentation/trace/ftrace.rst "trace options"
5418 Have the tracepoints sent to printk as well as the
5419 tracing ring buffer. This is useful for early boot up
5420 where the system hangs or reboots and does not give the
5421 option for reading the tracing buffer or performing a
5422 ftrace_dump_on_oops.
5424 To turn off having tracepoints sent to printk,
5425 echo 0 > /proc/sys/kernel/tracepoint_printk
5426 Note, echoing 1 into this file without the
5427 tracepoint_printk kernel cmdline option has no effect.
5431 Having tracepoints sent to printk() and activating high
5432 frequency tracepoints such as irq or sched, can cause
5433 the system to live lock.
5436 [FTRACE] enable this option to disable tracing when a
5437 warning is hit. This turns off "tracing_on". Tracing can
5438 be enabled again by echoing '1' into the "tracing_on"
5439 file located in /sys/kernel/debug/tracing/
5441 This option is useful, as it disables the trace before
5442 the WARNING dump is called, which prevents the trace to
5443 be filled with content caused by the warning output.
5445 This option can also be set at run time via the sysctl
5446 option: kernel/traceoff_on_warning
5448 transparent_hugepage=
5450 Format: [always|madvise|never]
5451 Can be used to control the default behavior of the system
5452 with respect to transparent hugepages.
5453 See Documentation/admin-guide/mm/transhuge.rst
5456 tsc= Disable clocksource stability checks for TSC.
5458 [x86] reliable: mark tsc clocksource as reliable, this
5459 disables clocksource verification at runtime, as well
5460 as the stability checks done at bootup. Used to enable
5461 high-resolution timer mode on older hardware, and in
5462 virtualized environment.
5463 [x86] noirqtime: Do not use TSC to do irq accounting.
5464 Used to run time disable IRQ_TIME_ACCOUNTING on any
5465 platforms where RDTSC is slow and this accounting
5467 [x86] unstable: mark the TSC clocksource as unstable, this
5468 marks the TSC unconditionally unstable at bootup and
5469 avoids any further wobbles once the TSC watchdog notices.
5470 [x86] nowatchdog: disable clocksource watchdog. Used
5471 in situations with strict latency requirements (where
5472 interruptions from clocksource watchdog are not
5475 tsc_early_khz= [X86] Skip early TSC calibration and use the given
5476 value instead. Useful when the early TSC frequency discovery
5477 procedure is not reliable, such as on overclocked systems
5478 with CPUID.16h support and partial CPUID.15h support.
5479 Format: <unsigned int>
5481 tsx= [X86] Control Transactional Synchronization
5482 Extensions (TSX) feature in Intel processors that
5483 support TSX control.
5485 This parameter controls the TSX feature. The options are:
5487 on - Enable TSX on the system. Although there are
5488 mitigations for all known security vulnerabilities,
5489 TSX has been known to be an accelerator for
5490 several previous speculation-related CVEs, and
5491 so there may be unknown security risks associated
5492 with leaving it enabled.
5494 off - Disable TSX on the system. (Note that this
5495 option takes effect only on newer CPUs which are
5496 not vulnerable to MDS, i.e., have
5497 MSR_IA32_ARCH_CAPABILITIES.MDS_NO=1 and which get
5498 the new IA32_TSX_CTRL MSR through a microcode
5499 update. This new MSR allows for the reliable
5500 deactivation of the TSX functionality.)
5502 auto - Disable TSX if X86_BUG_TAA is present,
5503 otherwise enable TSX on the system.
5505 Not specifying this option is equivalent to tsx=off.
5507 See Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
5510 tsx_async_abort= [X86,INTEL] Control mitigation for the TSX Async
5511 Abort (TAA) vulnerability.
5513 Similar to Micro-architectural Data Sampling (MDS)
5514 certain CPUs that support Transactional
5515 Synchronization Extensions (TSX) are vulnerable to an
5516 exploit against CPU internal buffers which can forward
5517 information to a disclosure gadget under certain
5520 In vulnerable processors, the speculatively forwarded
5521 data can be used in a cache side channel attack, to
5522 access data to which the attacker does not have direct
5525 This parameter controls the TAA mitigation. The
5528 full - Enable TAA mitigation on vulnerable CPUs
5531 full,nosmt - Enable TAA mitigation and disable SMT on
5532 vulnerable CPUs. If TSX is disabled, SMT
5533 is not disabled because CPU is not
5534 vulnerable to cross-thread TAA attacks.
5535 off - Unconditionally disable TAA mitigation
5537 On MDS-affected machines, tsx_async_abort=off can be
5538 prevented by an active MDS mitigation as both vulnerabilities
5539 are mitigated with the same mechanism so in order to disable
5540 this mitigation, you need to specify mds=off too.
5542 Not specifying this option is equivalent to
5543 tsx_async_abort=full. On CPUs which are MDS affected
5544 and deploy MDS mitigation, TAA mitigation is not
5545 required and doesn't provide any additional
5549 Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
5551 turbografx.map[2|3]= [HW,JOY]
5552 TurboGraFX parallel port interface
5554 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7>
5555 See also Documentation/input/devices/joystick-parport.rst
5557 udbg-immortal [PPC] When debugging early kernel crashes that
5558 happen after console_init() and before a proper
5559 console driver takes over, this boot options might
5560 help "seeing" what's going on.
5562 uhash_entries= [KNL,NET]
5563 Set number of hash buckets for UDP/UDP-Lite connections
5566 [USB] Ignore overcurrent events (default N).
5567 Some badly-designed motherboards generate lots of
5568 bogus events, for ports that aren't wired to
5569 anything. Set this parameter to avoid log spamming.
5570 Note that genuine overcurrent events won't be
5574 [X86] Cause panic on unknown NMI.
5576 usbcore.authorized_default=
5577 [USB] Default USB device authorization:
5578 (default -1 = authorized except for wireless USB,
5579 0 = not authorized, 1 = authorized, 2 = authorized
5580 if device connected to internal port)
5582 usbcore.autosuspend=
5583 [USB] The autosuspend time delay (in seconds) used
5584 for newly-detected USB devices (default 2). This
5585 is the time required before an idle device will be
5586 autosuspended. Devices for which the delay is set
5587 to a negative value won't be autosuspended at all.
5589 usbcore.usbfs_snoop=
5590 [USB] Set to log all usbfs traffic (default 0 = off).
5592 usbcore.usbfs_snoop_max=
5593 [USB] Maximum number of bytes to snoop in each URB
5596 usbcore.blinkenlights=
5597 [USB] Set to cycle leds on hubs (default 0 = off).
5599 usbcore.old_scheme_first=
5600 [USB] Start with the old device initialization
5601 scheme (default 0 = off).
5603 usbcore.usbfs_memory_mb=
5604 [USB] Memory limit (in MB) for buffers allocated by
5605 usbfs (default = 16, 0 = max = 2047).
5607 usbcore.use_both_schemes=
5608 [USB] Try the other device initialization scheme
5609 if the first one fails (default 1 = enabled).
5611 usbcore.initial_descriptor_timeout=
5612 [USB] Specifies timeout for the initial 64-byte
5613 USB_REQ_GET_DESCRIPTOR request in milliseconds
5614 (default 5000 = 5.0 seconds).
5616 usbcore.nousb [USB] Disable the USB subsystem
5619 [USB] A list of quirk entries to augment the built-in
5620 usb core quirk list. List entries are separated by
5621 commas. Each entry has the form
5622 VendorID:ProductID:Flags. The IDs are 4-digit hex
5623 numbers and Flags is a set of letters. Each letter
5624 will change the built-in quirk; setting it if it is
5625 clear and clearing it if it is set. The letters have
5626 the following meanings:
5627 a = USB_QUIRK_STRING_FETCH_255 (string
5628 descriptors must not be fetched using
5630 b = USB_QUIRK_RESET_RESUME (device can't resume
5631 correctly so reset it instead);
5632 c = USB_QUIRK_NO_SET_INTF (device can't handle
5633 Set-Interface requests);
5634 d = USB_QUIRK_CONFIG_INTF_STRINGS (device can't
5635 handle its Configuration or Interface
5637 e = USB_QUIRK_RESET (device can't be reset
5638 (e.g morph devices), don't use reset);
5639 f = USB_QUIRK_HONOR_BNUMINTERFACES (device has
5640 more interface descriptions than the
5641 bNumInterfaces count, and can't handle
5642 talking to these interfaces);
5643 g = USB_QUIRK_DELAY_INIT (device needs a pause
5644 during initialization, after we read
5645 the device descriptor);
5646 h = USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL (For
5647 high speed and super speed interrupt
5648 endpoints, the USB 2.0 and USB 3.0 spec
5649 require the interval in microframes (1
5650 microframe = 125 microseconds) to be
5651 calculated as interval = 2 ^
5653 Devices with this quirk report their
5654 bInterval as the result of this
5655 calculation instead of the exponent
5656 variable used in the calculation);
5657 i = USB_QUIRK_DEVICE_QUALIFIER (device can't
5658 handle device_qualifier descriptor
5660 j = USB_QUIRK_IGNORE_REMOTE_WAKEUP (device
5661 generates spurious wakeup, ignore
5662 remote wakeup capability);
5663 k = USB_QUIRK_NO_LPM (device can't handle Link
5665 l = USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL
5666 (Device reports its bInterval as linear
5667 frames instead of the USB 2.0
5669 m = USB_QUIRK_DISCONNECT_SUSPEND (Device needs
5670 to be disconnected before suspend to
5671 prevent spurious wakeup);
5672 n = USB_QUIRK_DELAY_CTRL_MSG (Device needs a
5673 pause after every control message);
5674 o = USB_QUIRK_HUB_SLOW_RESET (Hub needs extra
5675 delay after resetting its port);
5676 Example: quirks=0781:5580:bk,0a5c:5834:gij
5679 [USBHID] The interval which mice are to be polled at.
5682 [USBHID] The interval which joysticks are to be polled at.
5685 [USBHID] The interval which keyboards are to be polled at.
5687 usb-storage.delay_use=
5688 [UMS] The delay in seconds before a new device is
5689 scanned for Logical Units (default 1).
5692 [UMS] A list of quirks entries to supplement or
5693 override the built-in unusual_devs list. List
5694 entries are separated by commas. Each entry has
5695 the form VID:PID:Flags where VID and PID are Vendor
5696 and Product ID values (4-digit hex numbers) and
5697 Flags is a set of characters, each corresponding
5698 to a common usb-storage quirk flag as follows:
5699 a = SANE_SENSE (collect more than 18 bytes
5700 of sense data, not on uas);
5701 b = BAD_SENSE (don't collect more than 18
5702 bytes of sense data, not on uas);
5703 c = FIX_CAPACITY (decrease the reported
5704 device capacity by one sector);
5705 d = NO_READ_DISC_INFO (don't use
5706 READ_DISC_INFO command, not on uas);
5707 e = NO_READ_CAPACITY_16 (don't use
5708 READ_CAPACITY_16 command);
5709 f = NO_REPORT_OPCODES (don't use report opcodes
5711 g = MAX_SECTORS_240 (don't transfer more than
5712 240 sectors at a time, uas only);
5713 h = CAPACITY_HEURISTICS (decrease the
5714 reported device capacity by one
5715 sector if the number is odd);
5716 i = IGNORE_DEVICE (don't bind to this
5718 j = NO_REPORT_LUNS (don't use report luns
5720 k = NO_SAME (do not use WRITE_SAME, uas only)
5721 l = NOT_LOCKABLE (don't try to lock and
5722 unlock ejectable media, not on uas);
5723 m = MAX_SECTORS_64 (don't transfer more
5724 than 64 sectors = 32 KB at a time,
5726 n = INITIAL_READ10 (force a retry of the
5727 initial READ(10) command, not on uas);
5728 o = CAPACITY_OK (accept the capacity
5729 reported by the device, not on uas);
5730 p = WRITE_CACHE (the device cache is ON
5731 by default, not on uas);
5732 r = IGNORE_RESIDUE (the device reports
5733 bogus residue values, not on uas);
5734 s = SINGLE_LUN (the device has only one
5736 t = NO_ATA_1X (don't allow ATA(12) and ATA(16)
5737 commands, uas only);
5738 u = IGNORE_UAS (don't bind to the uas driver);
5739 w = NO_WP_DETECT (don't test whether the
5740 medium is write-protected).
5741 y = ALWAYS_SYNC (issue a SYNCHRONIZE_CACHE
5742 even if the device claims no cache,
5744 Example: quirks=0419:aaf5:rl,0421:0433:rc
5746 user_debug= [KNL,ARM]
5748 See arch/arm/Kconfig.debug help text.
5749 1 - undefined instruction events
5751 4 - invalid data aborts
5754 Example: user_debug=31
5757 [X86] Flags controlling user PTE allocations.
5759 nohigh = do not allocate PTE pages in
5760 HIGHMEM regardless of setting
5764 On X86_32, this is an alias for vdso32=. Otherwise:
5766 vdso=1: enable VDSO (the default)
5767 vdso=0: disable VDSO mapping
5769 vdso32= [X86] Control the 32-bit vDSO
5770 vdso32=1: enable 32-bit VDSO
5771 vdso32=0 or vdso32=2: disable 32-bit VDSO
5773 See the help text for CONFIG_COMPAT_VDSO for more
5774 details. If CONFIG_COMPAT_VDSO is set, the default is
5775 vdso32=0; otherwise, the default is vdso32=1.
5777 For compatibility with older kernels, vdso32=2 is an
5780 Try vdso32=0 if you encounter an error that says:
5781 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed!
5784 vector=percpu: enable percpu vector domain
5786 video= [FB] Frame buffer configuration
5787 See Documentation/fb/modedb.rst.
5789 video.brightness_switch_enabled= [0,1]
5790 If set to 1, on receiving an ACPI notify event
5791 generated by hotkey, video driver will adjust brightness
5792 level and then send out the event to user space through
5793 the allocated input device; If set to 0, video driver
5794 will only send out the event without touching backlight
5799 [VMMIO] Memory mapped virtio (platform) device.
5801 <size>@<baseaddr>:<irq>[:<id>]
5803 <size> := size (can use standard suffixes
5805 <baseaddr> := physical base address
5806 <irq> := interrupt number (as passed to
5808 <id> := (optional) platform device id
5810 virtio_mmio.device=1K@0x100b0000:48:7
5812 Can be used multiple times for multiple devices.
5814 vga= [BOOT,X86-32] Select a particular video mode
5815 See Documentation/x86/boot.rst and
5816 Documentation/admin-guide/svga.rst.
5817 Use vga=ask for menu.
5818 This is actually a boot loader parameter; the value is
5819 passed to the kernel using a special protocol.
5821 vm_debug[=options] [KNL] Available with CONFIG_DEBUG_VM=y.
5822 May slow down system boot speed, especially when
5823 enabled on systems with a large amount of memory.
5824 All options are enabled by default, and this
5825 interface is meant to allow for selectively
5826 enabling or disabling specific virtual memory
5829 Available options are:
5830 P Enable page structure init time poisoning
5831 - Disable all of the above options
5833 vmalloc=nn[KMG] [KNL,BOOT] Forces the vmalloc area to have an exact
5834 size of <nn>. This can be used to increase the
5835 minimum size (128MB on x86). It can also be used to
5836 decrease the size and leave more room for directly
5839 vmcp_cma=nn[MG] [KNL,S390]
5840 Sets the memory size reserved for contiguous memory
5841 allocations for the vmcp device driver.
5843 vmhalt= [KNL,S390] Perform z/VM CP command after system halt.
5846 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic.
5849 vmpoff= [KNL,S390] Perform z/VM CP command after power off.
5853 Controls the behavior of vsyscalls (i.e. calls to
5854 fixed addresses of 0xffffffffff600x00 from legacy
5855 code). Most statically-linked binaries and older
5856 versions of glibc use these calls. Because these
5857 functions are at fixed addresses, they make nice
5858 targets for exploits that can control RIP.
5860 emulate [default] Vsyscalls turn into traps and are
5861 emulated reasonably safely. The vsyscall
5864 xonly Vsyscalls turn into traps and are
5865 emulated reasonably safely. The vsyscall
5866 page is not readable.
5868 none Vsyscalls don't work at all. This makes
5869 them quite hard to use for exploits but
5870 might break your system.
5872 vt.color= [VT] Default text color.
5873 Format: 0xYX, X = foreground, Y = background.
5874 Default: 0x07 = light gray on black.
5876 vt.cur_default= [VT] Default cursor shape.
5877 Format: 0xCCBBAA, where AA, BB, and CC are the same as
5878 the parameters of the <Esc>[?A;B;Cc escape sequence;
5879 see VGA-softcursor.txt. Default: 2 = underline.
5881 vt.default_blu= [VT]
5882 Format: <blue0>,<blue1>,<blue2>,...,<blue15>
5883 Change the default blue palette of the console.
5884 This is a 16-member array composed of values
5887 vt.default_grn= [VT]
5888 Format: <green0>,<green1>,<green2>,...,<green15>
5889 Change the default green palette of the console.
5890 This is a 16-member array composed of values
5893 vt.default_red= [VT]
5894 Format: <red0>,<red1>,<red2>,...,<red15>
5895 Change the default red palette of the console.
5896 This is a 16-member array composed of values
5902 Set system-wide default UTF-8 mode for all tty's.
5903 Default is 1, i.e. UTF-8 mode is enabled for all
5904 newly opened terminals.
5906 vt.global_cursor_default=
5909 Set system-wide default for whether a cursor
5910 is shown on new VTs. Default is -1,
5911 i.e. cursors will be created by default unless
5912 overridden by individual drivers. 0 will hide
5913 cursors, 1 will display them.
5915 vt.italic= [VT] Default color for italic text; 0-15.
5918 vt.underline= [VT] Default color for underlined text; 0-15.
5921 watchdog timers [HW,WDT] For information on watchdog timers,
5922 see Documentation/watchdog/watchdog-parameters.rst
5923 or other driver-specific files in the
5924 Documentation/watchdog/ directory.
5928 Set the hard lockup detector stall duration
5929 threshold in seconds. The soft lockup detector
5930 threshold is set to twice the value. A value of 0
5931 disables both lockup detectors. Default is 10
5934 workqueue.watchdog_thresh=
5935 If CONFIG_WQ_WATCHDOG is configured, workqueue can
5936 warn stall conditions and dump internal state to
5937 help debugging. 0 disables workqueue stall
5938 detection; otherwise, it's the stall threshold
5939 duration in seconds. The default value is 30 and
5940 it can be updated at runtime by writing to the
5941 corresponding sysfs file.
5943 workqueue.disable_numa
5944 By default, all work items queued to unbound
5945 workqueues are affine to the NUMA nodes they're
5946 issued on, which results in better behavior in
5947 general. If NUMA affinity needs to be disabled for
5948 whatever reason, this option can be used. Note
5949 that this also can be controlled per-workqueue for
5950 workqueues visible under /sys/bus/workqueue/.
5952 workqueue.power_efficient
5953 Per-cpu workqueues are generally preferred because
5954 they show better performance thanks to cache
5955 locality; unfortunately, per-cpu workqueues tend to
5956 be more power hungry than unbound workqueues.
5958 Enabling this makes the per-cpu workqueues which
5959 were observed to contribute significantly to power
5960 consumption unbound, leading to measurably lower
5961 power usage at the cost of small performance
5964 The default value of this parameter is determined by
5965 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT.
5967 workqueue.debug_force_rr_cpu
5968 Workqueue used to implicitly guarantee that work
5969 items queued without explicit CPU specified are put
5970 on the local CPU. This guarantee is no longer true
5971 and while local CPU is still preferred work items
5972 may be put on foreign CPUs. This debug option
5973 forces round-robin CPU selection to flush out
5974 usages which depend on the now broken guarantee.
5975 When enabled, memory and cache locality will be
5978 x2apic_phys [X86-64,APIC] Use x2apic physical mode instead of
5979 default x2apic cluster mode on platforms
5982 x86_intel_mid_timer= [X86-32,APBT]
5983 Choose timer option for x86 Intel MID platform.
5984 Two valid options are apbt timer only and lapic timer
5985 plus one apbt timer for broadcast timer.
5986 x86_intel_mid_timer=apbt_only | lapic_and_apbt
5988 xen_512gb_limit [KNL,X86-64,XEN]
5989 Restricts the kernel running paravirtualized under Xen
5990 to use only up to 512 GB of RAM. The reason to do so is
5991 crash analysis tools and Xen tools for doing domain
5992 save/restore/migration must be enabled to handle larger
5995 xen_emul_unplug= [HW,X86,XEN]
5996 Unplug Xen emulated devices
5997 Format: [unplug0,][unplug1]
5998 ide-disks -- unplug primary master IDE devices
5999 aux-ide-disks -- unplug non-primary-master IDE devices
6000 nics -- unplug network devices
6001 all -- unplug all emulated devices (NICs and IDE disks)
6002 unnecessary -- unplugging emulated devices is
6003 unnecessary even if the host did not respond to
6005 never -- do not unplug even if version check succeeds
6007 xen_legacy_crash [X86,XEN]
6008 Crash from Xen panic notifier, without executing late
6009 panic() code such as dumping handler.
6011 xen_nopvspin [X86,XEN]
6012 Disables the qspinlock slowpath using Xen PV optimizations.
6013 This parameter is obsoleted by "nopvspin" parameter, which
6014 has equivalent effect for XEN platform.
6017 Disables the PV optimizations forcing the HVM guest to
6018 run as generic HVM guest with no PV drivers.
6019 This option is obsoleted by the "nopv" option, which
6020 has equivalent effect for XEN platform.
6022 xen_no_vector_callback
6023 [KNL,X86,XEN] Disable the vector callback for Xen
6024 event channel interrupts.
6026 xen_scrub_pages= [XEN]
6027 Boolean option to control scrubbing pages before giving them back
6028 to Xen, for use by other domains. Can be also changed at runtime
6029 with /sys/devices/system/xen_memory/xen_memory0/scrub_pages.
6030 Default value controlled with CONFIG_XEN_SCRUB_PAGES_DEFAULT.
6032 xen_timer_slop= [X86-64,XEN]
6033 Set the timer slop (in nanoseconds) for the virtual Xen
6034 timers (default is 100000). This adjusts the minimum
6035 delta of virtualized Xen timers, where lower values
6036 improve timer resolution at the expense of processing
6037 more timer interrupts.
6039 xen.event_eoi_delay= [XEN]
6040 How long to delay EOI handling in case of event
6041 storms (jiffies). Default is 10.
6043 xen.event_loop_timeout= [XEN]
6044 After which time (jiffies) the event handling loop
6045 should start to delay EOI handling. Default is 2.
6047 xen.fifo_events= [XEN]
6048 Boolean parameter to disable using fifo event handling
6049 even if available. Normally fifo event handling is
6050 preferred over the 2-level event handling, as it is
6051 fairer and the number of possible event channels is
6052 much higher. Default is on (use fifo events).
6054 nopv= [X86,XEN,KVM,HYPER_V,VMWARE]
6055 Disables the PV optimizations forcing the guest to run
6056 as generic guest with no PV drivers. Currently support
6057 XEN HVM, KVM, HYPER_V and VMWARE guest.
6059 nopvspin [X86,XEN,KVM]
6060 Disables the qspinlock slow path using PV optimizations
6061 which allow the hypervisor to 'idle' the guest on lock
6064 xirc2ps_cs= [NET,PCMCIA]
6066 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]]
6069 By default on POWER9 and above, the kernel will
6070 natively use the XIVE interrupt controller. This option
6071 allows the fallback firmware mode to be used:
6073 off Fallback to firmware control of XIVE interrupt
6074 controller on both pseries and powernv
6075 platforms. Only useful on POWER9 and above.
6077 xhci-hcd.quirks [USB,KNL]
6078 A hex value specifying bitmask with supplemental xhci
6079 host controller quirks. Meaning of each bit can be
6080 consulted in header drivers/usb/host/xhci.h.
6083 Format: { early | on | rw | ro | off }
6084 Controls if xmon debugger is enabled. Default is off.
6085 Passing only "xmon" is equivalent to "xmon=early".
6086 early Call xmon as early as possible on boot; xmon
6087 debugger is called from setup_arch().
6088 on xmon debugger hooks will be installed so xmon
6089 is only called on a kernel crash. Default mode,
6090 i.e. either "ro" or "rw" mode, is controlled
6091 with CONFIG_XMON_DEFAULT_RO_MODE.
6092 rw xmon debugger hooks will be installed so xmon
6093 is called only on a kernel crash, mode is write,
6094 meaning SPR registers, memory and, other data
6095 can be written using xmon commands.
6096 ro same as "rw" option above but SPR registers,
6097 memory, and other data can't be written using
6099 off xmon is disabled.