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.txt, 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]
27 If set to vendor, prefer vendor specific driver
28 (e.g. thinkpad_acpi, sony_acpi, etc.) instead
29 of the ACPI video.ko driver.
31 acpi_force_32bit_fadt_addr
32 force FADT to use 32 bit addresses rather than the
33 64 bit X_* addresses. Some firmware have broken 64
34 bit addresses for force ACPI ignore these and use
35 the older legacy 32 bit addresses.
37 acpica_no_return_repair [HW, ACPI]
38 Disable AML predefined validation mechanism
39 This mechanism can repair the evaluation result to make
40 the return objects more ACPI specification compliant.
41 This option is useful for developers to identify the
42 root cause of an AML interpreter issue when the issue
43 has something to do with the repair mechanism.
45 acpi.debug_layer= [HW,ACPI,ACPI_DEBUG]
46 acpi.debug_level= [HW,ACPI,ACPI_DEBUG]
48 CONFIG_ACPI_DEBUG must be enabled to produce any ACPI
49 debug output. Bits in debug_layer correspond to a
50 _COMPONENT in an ACPI source file, e.g.,
51 #define _COMPONENT ACPI_PCI_COMPONENT
52 Bits in debug_level correspond to a level in
53 ACPI_DEBUG_PRINT statements, e.g.,
54 ACPI_DEBUG_PRINT((ACPI_DB_INFO, ...
55 The debug_level mask defaults to "info". See
56 Documentation/acpi/debug.txt for more information about
57 debug layers and levels.
59 Enable processor driver info messages:
60 acpi.debug_layer=0x20000000
61 Enable PCI/PCI interrupt routing info messages:
62 acpi.debug_layer=0x400000
63 Enable AML "Debug" output, i.e., stores to the Debug
64 object while interpreting AML:
65 acpi.debug_layer=0xffffffff acpi.debug_level=0x2
66 Enable all messages related to ACPI hardware:
67 acpi.debug_layer=0x2 acpi.debug_level=0xffffffff
69 Some values produce so much output that the system is
70 unusable. The "log_buf_len" parameter may be useful
71 if you need to capture more output.
73 acpi_enforce_resources= [ACPI]
75 Check for resource conflicts between native drivers
76 and ACPI OperationRegions (SystemIO and SystemMemory
77 only). IO ports and memory declared in ACPI might be
78 used by the ACPI subsystem in arbitrary AML code and
79 can interfere with legacy drivers.
80 strict (default): access to resources claimed by ACPI
81 is denied; legacy drivers trying to access reserved
82 resources will fail to bind to device using them.
83 lax: access to resources claimed by ACPI is allowed;
84 legacy drivers trying to access reserved resources
85 will bind successfully but a warning message is logged.
86 no: ACPI OperationRegions are not marked as reserved,
87 no further checks are performed.
89 acpi_force_table_verification [HW,ACPI]
90 Enable table checksum verification during early stage.
91 By default, this is disabled due to x86 early mapping
94 acpi_irq_balance [HW,ACPI]
95 ACPI will balance active IRQs
98 acpi_irq_nobalance [HW,ACPI]
99 ACPI will not move active IRQs (default)
102 acpi_irq_isa= [HW,ACPI] If irq_balance, mark listed IRQs used by ISA
103 Format: <irq>,<irq>...
105 acpi_irq_pci= [HW,ACPI] If irq_balance, clear listed IRQs for
107 Format: <irq>,<irq>...
109 acpi_mask_gpe= [HW,ACPI]
110 Due to the existence of _Lxx/_Exx, some GPEs triggered
111 by unsupported hardware/firmware features can result in
112 GPE floodings that cannot be automatically disabled by
114 This facility can be used to prevent such uncontrolled
118 acpi_no_auto_serialize [HW,ACPI]
119 Disable auto-serialization of AML methods
120 AML control methods that contain the opcodes to create
121 named objects will be marked as "Serialized" by the
122 auto-serialization feature.
123 This feature is enabled by default.
124 This option allows to turn off the feature.
126 acpi_no_memhotplug [ACPI] Disable memory hotplug. Useful for kdump
129 acpi_no_static_ssdt [HW,ACPI]
130 Disable installation of static SSDTs at early boot time
131 By default, SSDTs contained in the RSDT/XSDT will be
132 installed automatically and they will appear under
133 /sys/firmware/acpi/tables.
134 This option turns off this feature.
135 Note that specifying this option does not affect
136 dynamic table installation which will install SSDT
137 tables to /sys/firmware/acpi/tables/dynamic.
139 acpi_rsdp= [ACPI,EFI,KEXEC]
140 Pass the RSDP address to the kernel, mostly used
141 on machines running EFI runtime service to boot the
142 second kernel for kdump.
144 acpi_os_name= [HW,ACPI] Tell ACPI BIOS the name of the OS
145 Format: To spoof as Windows 98: ="Microsoft Windows"
147 acpi_rev_override [ACPI] Override the _REV object to return 5 (instead
148 of 2 which is mandated by ACPI 6) as the supported ACPI
149 specification revision (when using this switch, it may
150 be necessary to carry out a cold reboot _twice_ in a
151 row to make it take effect on the platform firmware).
153 acpi_osi= [HW,ACPI] Modify list of supported OS interface strings
154 acpi_osi="string1" # add string1
155 acpi_osi="!string2" # remove string2
156 acpi_osi=!* # remove all strings
157 acpi_osi=! # disable all built-in OS vendor
159 acpi_osi=!! # enable all built-in OS vendor
161 acpi_osi= # disable all strings
163 'acpi_osi=!' can be used in combination with single or
164 multiple 'acpi_osi="string1"' to support specific OS
165 vendor string(s). Note that such command can only
166 affect the default state of the OS vendor strings, thus
167 it cannot affect the default state of the feature group
168 strings and the current state of the OS vendor strings,
169 specifying it multiple times through kernel command line
170 is meaningless. This command is useful when one do not
171 care about the state of the feature group strings which
172 should be controlled by the OSPM.
174 1. 'acpi_osi=! acpi_osi="Windows 2000"' is equivalent
175 to 'acpi_osi="Windows 2000" acpi_osi=!', they all
176 can make '_OSI("Windows 2000")' TRUE.
178 'acpi_osi=' cannot be used in combination with other
179 'acpi_osi=' command lines, the _OSI method will not
180 exist in the ACPI namespace. NOTE that such command can
181 only affect the _OSI support state, thus specifying it
182 multiple times through kernel command line is also
185 1. 'acpi_osi=' can make 'CondRefOf(_OSI, Local1)'
188 'acpi_osi=!*' can be used in combination with single or
189 multiple 'acpi_osi="string1"' to support specific
190 string(s). Note that such command can affect the
191 current state of both the OS vendor strings and the
192 feature group strings, thus specifying it multiple times
193 through kernel command line is meaningful. But it may
194 still not able to affect the final state of a string if
195 there are quirks related to this string. This command
196 is useful when one want to control the state of the
197 feature group strings to debug BIOS issues related to
200 1. 'acpi_osi="Module Device" acpi_osi=!*' can make
201 '_OSI("Module Device")' FALSE.
202 2. 'acpi_osi=!* acpi_osi="Module Device"' can make
203 '_OSI("Module Device")' TRUE.
204 3. 'acpi_osi=! acpi_osi=!* acpi_osi="Windows 2000"' is
206 'acpi_osi=!* acpi_osi=! acpi_osi="Windows 2000"'
208 'acpi_osi=!* acpi_osi="Windows 2000" acpi_osi=!',
209 they all will make '_OSI("Windows 2000")' TRUE.
212 Override the pmtimer bug detection: force the kernel
213 to assume that this machine's pmtimer latches its value
214 and always returns good values.
216 acpi_sci= [HW,ACPI] ACPI System Control Interrupt trigger mode
217 Format: { level | edge | high | low }
219 acpi_skip_timer_override [HW,ACPI]
220 Recognize and ignore IRQ0/pin2 Interrupt Override.
221 For broken nForce2 BIOS resulting in XT-PIC timer.
223 acpi_sleep= [HW,ACPI] Sleep options
224 Format: { s3_bios, s3_mode, s3_beep, s4_nohwsig,
225 old_ordering, nonvs, sci_force_enable, nobl }
226 See Documentation/power/video.txt for information on
228 s3_beep is for debugging; it makes the PC's speaker beep
229 as soon as the kernel's real-mode entry point is called.
230 s4_nohwsig prevents ACPI hardware signature from being
231 used during resume from hibernation.
232 old_ordering causes the ACPI 1.0 ordering of the _PTS
233 control method, with respect to putting devices into
234 low power states, to be enforced (the ACPI 2.0 ordering
235 of _PTS is used by default).
236 nonvs prevents the kernel from saving/restoring the
237 ACPI NVS memory during suspend/hibernation and resume.
238 sci_force_enable causes the kernel to set SCI_EN directly
239 on resume from S1/S3 (which is against the ACPI spec,
240 but some broken systems don't work without it).
241 nobl causes the internal blacklist of systems known to
242 behave incorrectly in some ways with respect to system
243 suspend and resume to be ignored (use wisely).
245 acpi_use_timer_override [HW,ACPI]
246 Use timer override. For some broken Nvidia NF5 boards
247 that require a timer override, but don't have HPET
249 add_efi_memmap [EFI; X86] Include EFI memory map in
250 kernel's map of available physical RAM.
253 { off | try_unsupported }
254 off: disable AGP support
255 try_unsupported: try to drive unsupported chipsets
256 (may crash computer or cause data corruption)
259 See Documentation/sound/alsa-configuration.rst
262 Allow the default userspace alignment fault handler
263 behaviour to be specified. Bit 0 enables warnings,
264 bit 1 enables fixups, and bit 2 sends a segfault.
266 align_va_addr= [X86-64]
267 Align virtual addresses by clearing slice [14:12] when
268 allocating a VMA at process creation time. This option
269 gives you up to 3% performance improvement on AMD F15h
270 machines (where it is enabled by default) for a
271 CPU-intensive style benchmark, and it can vary highly in
272 a microbenchmark depending on workload and compiler.
274 32: only for 32-bit processes
275 64: only for 64-bit processes
276 on: enable for both 32- and 64-bit processes
277 off: disable for both 32- and 64-bit processes
279 alloc_snapshot [FTRACE]
280 Allocate the ftrace snapshot buffer on boot up when the
281 main buffer is allocated. This is handy if debugging
282 and you need to use tracing_snapshot() on boot up, and
283 do not want to use tracing_snapshot_alloc() as it needs
284 to be done where GFP_KERNEL allocations are allowed.
286 amd_iommu= [HW,X86-64]
287 Pass parameters to the AMD IOMMU driver in the system.
289 fullflush - enable flushing of IO/TLB entries when
290 they are unmapped. Otherwise they are
291 flushed before they will be reused, which
293 off - do not initialize any AMD IOMMU found in
295 force_isolation - Force device isolation for all
296 devices. The IOMMU driver is not
297 allowed anymore to lift isolation
298 requirements as needed. This option
299 does not override iommu=pt
301 amd_iommu_dump= [HW,X86-64]
302 Enable AMD IOMMU driver option to dump the ACPI table
303 for AMD IOMMU. With this option enabled, AMD IOMMU
304 driver will print ACPI tables for AMD IOMMU during
305 IOMMU initialization.
307 amd_iommu_intr= [HW,X86-64]
308 Specifies one of the following AMD IOMMU interrupt
310 legacy - Use legacy interrupt remapping mode.
311 vapic - Use virtual APIC mode, which allows IOMMU
312 to inject interrupts directly into guest.
313 This mode requires kvm-amd.avic=1.
314 (Default when IOMMU HW support is present.)
316 amijoy.map= [HW,JOY] Amiga joystick support
317 Map of devices attached to JOY0DAT and JOY1DAT
319 See also Documentation/input/joydev/joystick.rst
321 analog.map= [HW,JOY] Analog joystick and gamepad support
322 Specifies type or capabilities of an analog joystick
323 connected to one of 16 gameports
324 Format: <type1>,<type2>,..<type16>
327 Power management functions (SPARCstation-4/5 + deriv.)
329 Disable APC CPU standby support. SPARCstation-Fox does
330 not play well with APC CPU idle - disable it if you have
331 APC and your system crashes randomly.
333 apic= [APIC,X86] Advanced Programmable Interrupt Controller
334 Change the output verbosity whilst booting
335 Format: { quiet (default) | verbose | debug }
336 Change the amount of debugging information output
337 when initialising the APIC and IO-APIC components.
338 For X86-32, this can also be used to specify an APIC
340 Format: apic=driver_name
341 Examples: apic=bigsmp
343 apic_extnmi= [APIC,X86] External NMI delivery setting
344 Format: { bsp (default) | all | none }
345 bsp: External NMI is delivered only to CPU 0
346 all: External NMIs are broadcast to all CPUs as a
348 none: External NMI is masked for all CPUs. This is
349 useful so that a dump capture kernel won't be
353 See Documentation/networking/ipv6.txt.
355 show_lapic= [APIC,X86] Advanced Programmable Interrupt Controller
356 Limit apic dumping. The parameter defines the maximal
357 number of local apics being dumped. Also it is possible
358 to set it to "all" by meaning -- no limit here.
359 Format: { 1 (default) | 2 | ... | all }.
360 The parameter valid if only apic=debug or
361 apic=verbose is specified.
362 Example: apic=debug show_lapic=all
364 apm= [APM] Advanced Power Management
365 See header of arch/x86/kernel/apm_32.c.
367 arcrimi= [HW,NET] ARCnet - "RIM I" (entirely mem-mapped) cards
368 Format: <io>,<irq>,<nodeID>
372 atarimouse= [HW,MOUSE] Atari Mouse
374 atkbd.extra= [HW] Enable extra LEDs and keys on IBM RapidAccess,
375 EzKey and similar keyboards
377 atkbd.reset= [HW] Reset keyboard during initialization
379 atkbd.set= [HW] Select keyboard code set
380 Format: <int> (2 = AT (default), 3 = PS/2)
382 atkbd.scroll= [HW] Enable scroll wheel on MS Office and similar
385 atkbd.softraw= [HW] Choose between synthetic and real raw mode
386 Format: <bool> (0 = real, 1 = synthetic (default))
388 atkbd.softrepeat= [HW]
389 Use software keyboard repeat
391 audit= [KNL] Enable the audit sub-system
392 Format: { "0" | "1" | "off" | "on" }
393 0 | off - kernel audit is disabled and can not be
394 enabled until the next reboot
395 unset - kernel audit is initialized but disabled and
396 will be fully enabled by the userspace auditd.
397 1 | on - kernel audit is initialized and partially
398 enabled, storing at most audit_backlog_limit
399 messages in RAM until it is fully enabled by the
403 audit_backlog_limit= [KNL] Set the audit queue size limit.
404 Format: <int> (must be >=0)
407 bau= [X86_UV] Enable the BAU on SGI UV. The default
408 behavior is to disable the BAU (i.e. bau=0).
409 Format: { "0" | "1" }
412 unset - Disable the BAU.
414 baycom_epp= [HW,AX25]
417 baycom_par= [HW,AX25] BayCom Parallel Port AX.25 Modem
419 See header of drivers/net/hamradio/baycom_par.c.
421 baycom_ser_fdx= [HW,AX25]
422 BayCom Serial Port AX.25 Modem (Full Duplex Mode)
423 Format: <io>,<irq>,<mode>[,<baud>]
424 See header of drivers/net/hamradio/baycom_ser_fdx.c.
426 baycom_ser_hdx= [HW,AX25]
427 BayCom Serial Port AX.25 Modem (Half Duplex Mode)
428 Format: <io>,<irq>,<mode>
429 See header of drivers/net/hamradio/baycom_ser_hdx.c.
431 blkdevparts= Manual partition parsing of block device(s) for
432 embedded devices based on command line input.
433 See Documentation/block/cmdline-partition.txt
435 boot_delay= Milliseconds to delay each printk during boot.
436 Values larger than 10 seconds (10000) are changed to
440 bootmem_debug [KNL] Enable bootmem allocator debug messages.
443 Disable BERT OS support on buggy BIOSes.
445 bttv.card= [HW,V4L] bttv (bt848 + bt878 based grabber cards)
446 bttv.radio= Most important insmod options are available as
448 bttv.pll= See Documentation/media/v4l-drivers/bttv.rst
451 bulk_remove=off [PPC] This parameter disables the use of the pSeries
452 firmware feature for flushing multiple hpte entries
455 c101= [NET] Moxa C101 synchronous serial card
457 cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection.
458 Sometimes CPU hardware bugs make them report the cache
459 size incorrectly. The kernel will attempt work arounds
460 to fix known problems, but for some CPUs it is not
461 possible to determine what the correct size should be.
462 This option provides an override for these situations.
464 ca_keys= [KEYS] This parameter identifies a specific key(s) on
465 the system trusted keyring to be used for certificate
467 format: { id:<keyid> | builtin }
469 cca= [MIPS] Override the kernel pages' cache coherency
470 algorithm. Accepted values range from 0 to 7
471 inclusive. See arch/mips/include/asm/pgtable-bits.h
472 for platform specific values (SB1, Loongson3 and
475 ccw_timeout_log [S390]
476 See Documentation/s390/CommonIO for details.
478 cgroup_disable= [KNL] Disable a particular controller
479 Format: {name of the controller(s) to disable}
480 The effects of cgroup_disable=foo are:
481 - foo isn't auto-mounted if you mount all cgroups in
483 - foo isn't visible as an individually mountable
485 {Currently only "memory" controller deal with this and
486 cut the overhead, others just disable the usage. So
487 only cgroup_disable=memory is actually worthy}
489 cgroup_no_v1= [KNL] Disable one, multiple, all cgroup controllers in v1
490 Format: { controller[,controller...] | "all" }
491 Like cgroup_disable, but only applies to cgroup v1;
492 the blacklisted controllers remain available in cgroup2.
494 cgroup.memory= [KNL] Pass options to the cgroup memory controller.
496 nosocket -- Disable socket memory accounting.
497 nokmem -- Disable kernel memory accounting.
499 checkreqprot [SELINUX] Set initial checkreqprot flag value.
500 Format: { "0" | "1" }
501 See security/selinux/Kconfig help text.
502 0 -- check protection applied by kernel (includes
503 any implied execute protection).
504 1 -- check protection requested by application.
505 Default value is set via a kernel config option.
506 Value can be changed at runtime via
507 /selinux/checkreqprot.
510 See Documentation/s390/CommonIO for details.
513 Prevents the clock framework from automatically gating
514 clocks that have not been explicitly enabled by a Linux
515 device driver but are enabled in hardware at reset or
516 by the bootloader/firmware. Note that this does not
517 force such clocks to be always-on nor does it reserve
518 those clocks in any way. This parameter is useful for
519 debug and development, but should not be needed on a
520 platform with proper driver support. For more
521 information, see Documentation/driver-api/clk.rst.
523 clock= [BUGS=X86-32, HW] gettimeofday clocksource override.
525 Forces specified clocksource (if available) to be used
526 when calculating gettimeofday(). If specified
527 clocksource is not available, it defaults to PIT.
528 Format: { pit | tsc | cyclone | pmtmr }
530 clocksource= Override the default clocksource
532 Override the default clocksource and use the clocksource
533 with the name specified.
534 Some clocksource names to choose from, depending on
536 [all] jiffies (this is the base, fallback clocksource)
538 [ARM] imx_timer1,OSTS,netx_timer,mpu_timer2,
539 pxa_timer,timer3,32k_counter,timer0_1
540 [X86-32] pit,hpet,tsc;
541 scx200_hrt on Geode; cyclone on IBM x440
549 clocksource.arm_arch_timer.evtstrm=
552 Enable/disable the eventstream feature of the ARM
553 architected timer so that code using WFE-based polling
554 loops can be debugged more effectively on production
557 clearcpuid=BITNUM [X86]
558 Disable CPUID feature X for the kernel. See
559 arch/x86/include/asm/cpufeatures.h for the valid bit
560 numbers. Note the Linux specific bits are not necessarily
561 stable over kernel options, but the vendor specific
563 Also note that user programs calling CPUID directly
564 or using the feature without checking anything
565 will still see it. This just prevents it from
566 being used by the kernel or shown in /proc/cpuinfo.
567 Also note the kernel might malfunction if you disable
570 cma=nn[MG]@[start[MG][-end[MG]]]
572 Sets the size of kernel global memory area for
573 contiguous memory allocations and optionally the
574 placement constraint by the physical address range of
575 memory allocations. A value of 0 disables CMA
576 altogether. For more information, see
577 include/linux/dma-contiguous.h
579 cmo_free_hint= [PPC] Format: { yes | no }
580 Specify whether pages are marked as being inactive
581 when they are freed. This is used in CMO environments
582 to determine OS memory pressure for page stealing by
586 coherent_pool=nn[KMG] [ARM,KNL]
587 Sets the size of memory pool for coherent, atomic dma
588 allocations, by default set to 256K.
590 com20020= [HW,NET] ARCnet - COM20020 chipset
592 <io>[,<irq>[,<nodeID>[,<backplane>[,<ckp>[,<timeout>]]]]]
594 com90io= [HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers)
598 ARCnet - COM90xx chipset (memory-mapped buffers)
599 Format: <io>[,<irq>[,<memstart>]]
601 condev= [HW,S390] console device
604 console= [KNL] Output console device and options.
606 tty<n> Use the virtual console device <n>.
610 Use the specified serial port. The options are of
611 the form "bbbbpnf", where "bbbb" is the baud rate,
612 "p" is parity ("n", "o", or "e"), "n" is number of
613 bits, and "f" is flow control ("r" for RTS or
614 omit it). Default is "9600n8".
616 See Documentation/admin-guide/serial-console.rst for more
618 Documentation/networking/netconsole.txt for an
621 uart[8250],io,<addr>[,options]
622 uart[8250],mmio,<addr>[,options]
623 uart[8250],mmio16,<addr>[,options]
624 uart[8250],mmio32,<addr>[,options]
625 uart[8250],0x<addr>[,options]
626 Start an early, polled-mode console on the 8250/16550
627 UART at the specified I/O port or MMIO address,
628 switching to the matching ttyS device later.
629 MMIO inter-register address stride is either 8-bit
630 (mmio), 16-bit (mmio16), or 32-bit (mmio32).
631 If none of [io|mmio|mmio16|mmio32], <addr> is assumed
632 to be equivalent to 'mmio'. 'options' are specified in
633 the same format described for ttyS above; if unspecified,
634 the h/w is not re-initialized.
636 hvc<n> Use the hypervisor console device <n>. This is for
637 both Xen and PowerPC hypervisors.
639 If the device connected to the port is not a TTY but a braille
640 device, prepend "brl," before the device type, for instance
642 For now, only VisioBraille is supported.
645 [KNL] Change console messages format
647 By default we print messages on consoles in
648 "[time stamp] text\n" format (time stamp may not be
649 printed, depending on CONFIG_PRINTK_TIME or
650 `printk_time' param).
652 Switch to syslog format: "<%u>[time stamp] text\n"
653 IOW, each message will have a facility and loglevel
654 prefix. The format is similar to one used by syslog()
655 syscall, or to executing "dmesg -S --raw" or to reading
658 consoleblank= [KNL] The console blank (screen saver) timeout in
659 seconds. A value of 0 disables the blank timer.
663 [KNL] Change the default value for
664 /proc/<pid>/coredump_filter.
665 See also Documentation/filesystems/proc.txt.
667 coresight_cpu_debug.enable
670 Enable/disable the CPU sampling based debugging.
671 0: default value, disable debugging
672 1: enable debugging at boot time
674 cpuidle.off=1 [CPU_IDLE]
675 disable the cpuidle sub-system
677 cpufreq.off=1 [CPU_FREQ]
678 disable the cpufreq sub-system
681 [X86] Delay for N microsec between assert and de-assert
682 of APIC INIT to start processors. This delay occurs
683 on every CPU online, such as boot, and resume from suspend.
686 cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver
688 <first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>]
690 crashkernel=size[KMG][@offset[KMG]]
691 [KNL] Using kexec, Linux can switch to a 'crash kernel'
692 upon panic. This parameter reserves the physical
693 memory region [offset, offset + size] for that kernel
694 image. If '@offset' is omitted, then a suitable offset
695 is selected automatically. Check
696 Documentation/kdump/kdump.txt for further details.
698 crashkernel=range1:size1[,range2:size2,...][@offset]
699 [KNL] Same as above, but depends on the memory
700 in the running system. The syntax of range is
701 start-[end] where start and end are both
702 a memory unit (amount[KMG]). See also
703 Documentation/kdump/kdump.txt for an example.
705 crashkernel=size[KMG],high
706 [KNL, x86_64] range could be above 4G. Allow kernel
707 to allocate physical memory region from top, so could
708 be above 4G if system have more than 4G ram installed.
709 Otherwise memory region will be allocated below 4G, if
711 It will be ignored if crashkernel=X is specified.
712 crashkernel=size[KMG],low
713 [KNL, x86_64] range under 4G. When crashkernel=X,high
714 is passed, kernel could allocate physical memory region
715 above 4G, that cause second kernel crash on system
716 that require some amount of low memory, e.g. swiotlb
717 requires at least 64M+32K low memory, also enough extra
718 low memory is needed to make sure DMA buffers for 32-bit
719 devices won't run out. Kernel would try to allocate at
720 at least 256M below 4G automatically.
721 This one let user to specify own low range under 4G
722 for second kernel instead.
723 0: to disable low allocation.
724 It will be ignored when crashkernel=X,high is not used
725 or memory reserved is below 4G.
728 [KNL] Disable crypto self-tests
733 cs89x0_media= [HW,NET]
734 Format: { rj45 | aui | bnc }
737 See header of drivers/s390/block/dasd_devmap.c.
739 db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port
740 (one device per port)
741 Format: <port#>,<type>
742 See also Documentation/input/devices/joystick-parport.rst
744 ddebug_query= [KNL,DYNAMIC_DEBUG] Enable debug messages at early boot
746 Documentation/admin-guide/dynamic-debug-howto.rst for
747 details. Deprecated, see dyndbg.
749 debug [KNL] Enable kernel debugging (events log level).
752 [KNL] Enable printing [hashed] pointers early in the
753 boot sequence. If enabled, we use a weak hash instead
754 of siphash to hash pointers. Use this option if you are
755 seeing instances of '(___ptrval___)') and need to see a
756 value (hashed pointer) instead. Cryptographically
757 insecure, please do not use on production kernels.
760 [KNL] verbose self-tests
762 Print debugging info while doing the locking API
764 We default to 0 (no extra messages), setting it to
765 1 will print _a lot_ more information - normally
766 only useful to kernel developers.
768 debug_objects [KNL] Enable object debugging
771 [KNL] Disable object debugging
773 debug_guardpage_minorder=
774 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this
775 parameter allows control of the order of pages that will
776 be intentionally kept free (and hence protected) by the
777 buddy allocator. Bigger value increase the probability
778 of catching random memory corruption, but reduce the
779 amount of memory for normal system use. The maximum
780 possible value is MAX_ORDER/2. Setting this parameter
781 to 1 or 2 should be enough to identify most random
782 memory corruption problems caused by bugs in kernel or
783 driver code when a CPU writes to (or reads from) a
784 random memory location. Note that there exists a class
785 of memory corruptions problems caused by buggy H/W or
786 F/W or by drivers badly programing DMA (basically when
787 memory is written at bus level and the CPU MMU is
788 bypassed) which are not detectable by
789 CONFIG_DEBUG_PAGEALLOC, hence this option will not help
790 tracking down these problems.
793 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this
794 parameter enables the feature at boot time. In
795 default, it is disabled. We can avoid allocating huge
796 chunk of memory for debug pagealloc if we don't enable
797 it at boot time and the system will work mostly same
798 with the kernel built without CONFIG_DEBUG_PAGEALLOC.
799 on: enable the feature
801 debugpat [X86] Enable PAT debugging
803 decnet.addr= [HW,NET]
804 Format: <area>[,<node>]
805 See also Documentation/networking/decnet.txt.
808 [same as hugepagesz=] The size of the default
809 HugeTLB page size. This is the size represented by
810 the legacy /proc/ hugepages APIs, used for SHM, and
811 default size when mounting hugetlbfs filesystems.
812 Defaults to the default architecture's huge page size
815 deferred_probe_timeout=
816 [KNL] Debugging option to set a timeout in seconds for
817 deferred probe to give up waiting on dependencies to
818 probe. Only specific dependencies (subsystems or
819 drivers) that have opted in will be ignored. A timeout of 0
820 will timeout at the end of initcalls. This option will also
821 dump out devices still on the deferred probe list after
825 Set number of hash buckets for dentry cache.
827 disable_1tb_segments [PPC]
828 Disables the use of 1TB hash page table segments. This
829 causes the kernel to fall back to 256MB segments which
830 can be useful when debugging issues that require an SLB
834 See Documentation/networking/ipv6.txt.
837 [KNL] Under CONFIG_HARDENED_USERCOPY, whether
838 hardening is enabled for this boot. Hardened
839 usercopy checking is used to protect the kernel
840 from reading or writing beyond known memory
841 allocation boundaries as a proactive defense
842 against bounds-checking flaws in the kernel's
843 copy_to_user()/copy_from_user() interface.
844 on Perform hardened usercopy checks (default).
845 off Disable hardened usercopy checks.
848 Disable RADIX MMU mode on POWER9
850 disable_cpu_apicid= [X86,APIC,SMP]
852 The number of initial APIC ID for the
853 corresponding CPU to be disabled at boot,
854 mostly used for the kdump 2nd kernel to
855 disable BSP to wake up multiple CPUs without
856 causing system reset or hang due to sending
859 perf_v4_pmi= [X86,INTEL]
861 Disable Intel PMU counter freezing feature.
862 The feature only exists starting from
863 Arch Perfmon v4 (Skylake and newer).
865 disable_ddw [PPC/PSERIES]
866 Disable Dynamic DMA Window support. Use this if
867 to workaround buggy firmware.
870 See Documentation/networking/ipv6.txt.
872 disable_mtrr_cleanup [X86]
873 The kernel tries to adjust MTRR layout from continuous
874 to discrete, to make X server driver able to add WB
875 entry later. This parameter disables that.
877 disable_mtrr_trim [X86, Intel and AMD only]
878 By default the kernel will trim any uncacheable
879 memory out of your available memory pool based on
880 MTRR settings. This parameter disables that behavior,
881 possibly causing your machine to run very slowly.
883 disable_timer_pin_1 [X86]
884 Disable PIN 1 of APIC timer
885 Can be useful to work around chipset bugs.
887 dis_ucode_ldr [X86] Disable the microcode loader.
889 dma_debug=off If the kernel is compiled with DMA_API_DEBUG support,
890 this option disables the debugging code at boot.
892 dma_debug_entries=<number>
893 This option allows to tune the number of preallocated
894 entries for DMA-API debugging code. One entry is
895 required per DMA-API allocation. Use this if the
896 DMA-API debugging code disables itself because the
897 architectural default is too low.
899 dma_debug_driver=<driver_name>
900 With this option the DMA-API debugging driver
901 filter feature can be enabled at boot time. Just
902 pass the driver to filter for as the parameter.
903 The filter can be disabled or changed to another
904 driver later using sysfs.
906 drm.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>]
907 Broken monitors, graphic adapters, KVMs and EDIDless
908 panels may send no or incorrect EDID data sets.
909 This parameter allows to specify an EDID data sets
910 in the /lib/firmware directory that are used instead.
911 Generic built-in EDID data sets are used, if one of
912 edid/1024x768.bin, edid/1280x1024.bin,
913 edid/1680x1050.bin, or edid/1920x1080.bin is given
914 and no file with the same name exists. Details and
915 instructions how to build your own EDID data are
916 available in Documentation/EDID/HOWTO.txt. An EDID
917 data set will only be used for a particular connector,
918 if its name and a colon are prepended to the EDID
919 name. Each connector may use a unique EDID data
920 set by separating the files with a comma. An EDID
921 data set with no connector name will be used for
922 any connectors not explicitly specified.
927 Format: {"off" | "known"}
928 Control how the dt_cpu_ftrs device-tree binding is
929 used for CPU feature discovery and setup (if it
931 off: Do not use it, fall back to legacy cpu table.
932 known: Do not pass through unknown features to guests
933 or userspace, only those that the kernel is aware of.
935 dump_apple_properties [X86]
936 Dump name and content of EFI device properties on
937 x86 Macs. Useful for driver authors to determine
938 what data is available or for reverse-engineering.
940 dyndbg[="val"] [KNL,DYNAMIC_DEBUG]
941 module.dyndbg[="val"]
942 Enable debug messages at boot time. See
943 Documentation/admin-guide/dynamic-debug-howto.rst
946 nompx [X86] Disables Intel Memory Protection Extensions.
947 See Documentation/x86/intel_mpx.txt for more
948 information about the feature.
950 nopku [X86] Disable Memory Protection Keys CPU feature found
953 module.async_probe [KNL]
954 Enable asynchronous probe on this module.
956 early_ioremap_debug [KNL]
957 Enable debug messages in early_ioremap support. This
958 is useful for tracking down temporary early mappings
959 which are not unmapped.
961 earlycon= [KNL] Output early console device and options.
963 [ARM64] The early console is determined by the
964 stdout-path property in device tree's chosen node,
965 or determined by the ACPI SPCR table.
967 [X86] When used with no options the early console is
968 determined by the ACPI SPCR table.
970 cdns,<addr>[,options]
971 Start an early, polled-mode console on a Cadence
972 (xuartps) serial port at the specified address. Only
973 supported option is baud rate. If baud rate is not
974 specified, the serial port must already be setup and
977 uart[8250],io,<addr>[,options]
978 uart[8250],mmio,<addr>[,options]
979 uart[8250],mmio32,<addr>[,options]
980 uart[8250],mmio32be,<addr>[,options]
981 uart[8250],0x<addr>[,options]
982 Start an early, polled-mode console on the 8250/16550
983 UART at the specified I/O port or MMIO address.
984 MMIO inter-register address stride is either 8-bit
985 (mmio) or 32-bit (mmio32 or mmio32be).
986 If none of [io|mmio|mmio32|mmio32be], <addr> is assumed
987 to be equivalent to 'mmio'. 'options' are specified
988 in the same format described for "console=ttyS<n>"; if
989 unspecified, the h/w is not initialized.
993 Start an early, polled-mode console on a pl011 serial
994 port at the specified address. The pl011 serial port
995 must already be setup and configured. Options are not
996 yet supported. If 'mmio32' is specified, then only
997 the driver will use only 32-bit accessors to read/write
998 the device registers.
1001 Start an early, polled-mode console on a meson serial
1002 port at the specified address. The serial port must
1003 already be setup and configured. Options are not yet
1007 Start an early, polled-mode console on an msm serial
1008 port at the specified address. The serial port
1009 must already be setup and configured. Options are not
1012 msm_serial_dm,<addr>
1013 Start an early, polled-mode console on an msm serial
1014 dm port at the specified address. The serial port
1015 must already be setup and configured. Options are not
1019 Start an early, polled-mode console on a serial port
1020 of an Actions Semi SoC, such as S500 or S900, at the
1021 specified address. The serial port must already be
1022 setup and configured. Options are not yet supported.
1024 smh Use ARM semihosting calls for early console.
1032 Use early console provided by serial driver available
1033 on Samsung SoCs, requires selecting proper type and
1034 a correct base address of the selected UART port. The
1035 serial port must already be setup and configured.
1036 Options are not yet supported.
1039 Start an early, polled-mode console on a lantiq serial
1040 (lqasc) port at the specified address. The serial port
1041 must already be setup and configured. Options are not
1046 Use early console provided by Freescale LP UART driver
1047 found on Freescale Vybrid and QorIQ LS1021A processors.
1048 A valid base address must be provided, and the serial
1049 port must already be setup and configured.
1052 Start an early, polled-mode console on the
1053 Armada 3700 serial port at the specified
1054 address. The serial port must already be setup
1055 and configured. Options are not yet supported.
1058 Start an early, polled-mode console on a Qualcomm
1059 Generic Interface (GENI) based serial port at the
1060 specified address. The serial port must already be
1061 setup and configured. Options are not yet supported.
1063 earlyprintk= [X86,SH,ARM,M68k,S390]
1068 earlyprintk=serial[,ttySn[,baudrate]]
1069 earlyprintk=serial[,0x...[,baudrate]]
1070 earlyprintk=ttySn[,baudrate]
1071 earlyprintk=dbgp[debugController#]
1072 earlyprintk=pciserial[,force],bus:device.function[,baudrate]
1073 earlyprintk=xdbc[xhciController#]
1075 earlyprintk is useful when the kernel crashes before
1076 the normal console is initialized. It is not enabled by
1077 default because it has some cosmetic problems.
1079 Append ",keep" to not disable it when the real console
1082 Only one of vga, efi, serial, or usb debug port can
1085 Currently only ttyS0 and ttyS1 may be specified by
1086 name. Other I/O ports may be explicitly specified
1087 on some architectures (x86 and arm at least) by
1088 replacing ttySn with an I/O port address, like this:
1089 earlyprintk=serial,0x1008,115200
1090 You can find the port for a given device in
1091 /proc/tty/driver/serial:
1092 2: uart:ST16650V2 port:00001008 irq:18 ...
1094 Interaction with the standard serial driver is not
1097 The VGA and EFI output is eventually overwritten by
1100 The xen output can only be used by Xen PV guests.
1102 The sclp output can only be used on s390.
1104 The optional "force" to "pciserial" enables use of a
1105 PCI device even when its classcode is not of the
1108 edac_report= [HW,EDAC] Control how to report EDAC event
1109 Format: {"on" | "off" | "force"}
1110 on: enable EDAC to report H/W event. May be overridden
1111 by other higher priority error reporting module.
1112 off: disable H/W event reporting through EDAC.
1113 force: enforce the use of EDAC to report H/W event.
1116 ekgdboc= [X86,KGDB] Allow early kernel console debugging
1119 This is designed to be used in conjunction with
1120 the boot argument: earlyprintk=vga
1123 Format: {"off" | "on" | "skip[mbr]"}
1126 Format: { "old_map", "nochunk", "noruntime", "debug" }
1127 old_map [X86-64]: switch to the old ioremap-based EFI
1128 runtime services mapping. 32-bit still uses this one by
1130 nochunk: disable reading files in "chunks" in the EFI
1131 boot stub, as chunking can cause problems with some
1132 firmware implementations.
1133 noruntime : disable EFI runtime services support
1134 debug: enable misc debug output
1136 efi_no_storage_paranoia [EFI; X86]
1137 Using this parameter you can use more than 50% of
1138 your efi variable storage. Use this parameter only if
1139 you are really sure that your UEFI does sane gc and
1140 fulfills the spec otherwise your board may brick.
1142 efi_fake_mem= nn[KMG]@ss[KMG]:aa[,nn[KMG]@ss[KMG]:aa,..] [EFI; X86]
1143 Add arbitrary attribute to specific memory range by
1144 updating original EFI memory map.
1145 Region of memory which aa attribute is added to is
1147 If efi_fake_mem=2G@4G:0x10000,2G@0x10a0000000:0x10000
1148 is specified, EFI_MEMORY_MORE_RELIABLE(0x10000)
1149 attribute is added to range 0x100000000-0x180000000 and
1150 0x10a0000000-0x1120000000.
1152 Using this parameter you can do debugging of EFI memmap
1153 related feature. For example, you can do debugging of
1154 Address Range Mirroring feature even if your box
1157 efivar_ssdt= [EFI; X86] Name of an EFI variable that contains an SSDT
1158 that is to be dynamically loaded by Linux. If there are
1159 multiple variables with the same name but with different
1160 vendor GUIDs, all of them will be loaded. See
1161 Documentation/acpi/ssdt-overlays.txt for details.
1164 eisa_irq_edge= [PARISC,HW]
1165 See header of drivers/parisc/eisa.c.
1168 See comment before function elanfreq_setup() in
1169 arch/x86/kernel/cpu/cpufreq/elanfreq.c.
1172 Format: {"cfq" | "deadline" | "noop"}
1173 See Documentation/block/cfq-iosched.txt and
1174 Documentation/block/deadline-iosched.txt for details.
1176 elfcorehdr=[size[KMG]@]offset[KMG] [IA64,PPC,SH,X86,S390]
1177 Specifies physical address of start of kernel core
1178 image elf header and optionally the size. Generally
1179 kexec loader will pass this option to capture kernel.
1180 See Documentation/kdump/kdump.txt for details.
1182 enable_mtrr_cleanup [X86]
1183 The kernel tries to adjust MTRR layout from continuous
1184 to discrete, to make X server driver able to add WB
1185 entry later. This parameter enables that.
1187 enable_timer_pin_1 [X86]
1188 Enable PIN 1 of APIC timer
1189 Can be useful to work around chipset bugs
1190 (in particular on some ATI chipsets).
1191 The kernel tries to set a reasonable default.
1193 enforcing [SELINUX] Set initial enforcing status.
1195 See security/selinux/Kconfig help text.
1196 0 -- permissive (log only, no denials).
1197 1 -- enforcing (deny and log).
1199 Value can be changed at runtime via /selinux/enforce.
1202 Disable Error Record Serialization Table (ERST)
1205 ether= [HW,NET] Ethernet cards parameters
1206 This option is obsoleted by the "netdev=" option, which
1207 has equivalent usage. See its documentation for details.
1211 Permit 'security.evm' to be updated regardless of
1212 current integrity status.
1216 fail_make_request=[KNL]
1217 General fault injection mechanism.
1218 Format: <interval>,<probability>,<space>,<times>
1219 See also Documentation/fault-injection/.
1222 See Documentation/blockdev/floppy.txt.
1224 force_pal_cache_flush
1225 [IA-64] Avoid check_sal_cache_flush which may hang on
1226 buggy SAL_CACHE_FLUSH implementations. Using this
1227 parameter will force ia64_sal_cache_flush to call
1228 ia64_pal_cache_flush instead of SAL_CACHE_FLUSH.
1231 Forcefully enable Physical Address Extension (PAE).
1232 Many Pentium M systems disable PAE but may have a
1233 functionally usable PAE implementation.
1234 Warning: use of this parameter will taint the kernel
1235 and may cause unknown problems.
1238 [FTRACE] will set and start the specified tracer
1239 as early as possible in order to facilitate early
1242 ftrace_dump_on_oops[=orig_cpu]
1243 [FTRACE] will dump the trace buffers on oops.
1244 If no parameter is passed, ftrace will dump
1245 buffers of all CPUs, but if you pass orig_cpu, it will
1246 dump only the buffer of the CPU that triggered the
1249 ftrace_filter=[function-list]
1250 [FTRACE] Limit the functions traced by the function
1251 tracer at boot up. function-list is a comma separated
1252 list of functions. This list can be changed at run
1253 time by the set_ftrace_filter file in the debugfs
1256 ftrace_notrace=[function-list]
1257 [FTRACE] Do not trace the functions specified in
1258 function-list. This list can be changed at run time
1259 by the set_ftrace_notrace file in the debugfs
1262 ftrace_graph_filter=[function-list]
1263 [FTRACE] Limit the top level callers functions traced
1264 by the function graph tracer at boot up.
1265 function-list is a comma separated list of functions
1266 that can be changed at run time by the
1267 set_graph_function file in the debugfs tracing directory.
1269 ftrace_graph_notrace=[function-list]
1270 [FTRACE] Do not trace from the functions specified in
1271 function-list. This list is a comma separated list of
1272 functions that can be changed at run time by the
1273 set_graph_notrace file in the debugfs tracing directory.
1275 ftrace_graph_max_depth=<uint>
1276 [FTRACE] Used with the function graph tracer. This is
1277 the max depth it will trace into a function. This value
1278 can be changed at run time by the max_graph_depth file
1279 in the tracefs tracing directory. default: 0 (no limit)
1282 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad
1283 support via parallel port (up to 5 devices per port)
1284 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5>
1285 See also Documentation/input/devices/joystick-parport.rst
1289 gart_fix_e820= [X86_64] disable the fix e820 for K8 GART
1293 gcov_persist= [GCOV] When non-zero (default), profiling data for
1294 kernel modules is saved and remains accessible via
1295 debugfs, even when the module is unloaded/reloaded.
1296 When zero, profiling data is discarded and associated
1297 debugfs files are removed at module unload time.
1299 goldfish [X86] Enable the goldfish android emulator platform.
1300 Don't use this when you are not running on the
1303 gpt [EFI] Forces disk with valid GPT signature but
1304 invalid Protective MBR to be treated as GPT. If the
1305 primary GPT is corrupted, it enables the backup/alternate
1306 GPT to be used instead.
1308 grcan.enable0= [HW] Configuration of physical interface 0. Determines
1309 the "Enable 0" bit of the configuration register.
1312 grcan.enable1= [HW] Configuration of physical interface 1. Determines
1313 the "Enable 0" bit of the configuration register.
1316 grcan.select= [HW] Select which physical interface to use.
1319 grcan.txsize= [HW] Sets the size of the tx buffer.
1320 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0.
1322 grcan.rxsize= [HW] Sets the size of the rx buffer.
1323 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0.
1326 gpio-mockup.gpio_mockup_ranges
1327 [HW] Sets the ranges of gpiochip of for this device.
1328 Format: <start1>,<end1>,<start2>,<end2>...
1330 hardlockup_all_cpu_backtrace=
1331 [KNL] Should the hard-lockup detector generate
1332 backtraces on all cpus.
1335 hashdist= [KNL,NUMA] Large hashes allocated during boot
1336 are distributed across NUMA nodes. Defaults on
1337 for 64-bit NUMA, off otherwise.
1338 Format: 0 | 1 (for off | on)
1340 hcl= [IA-64] SGI's Hardware Graph compatibility layer
1342 hd= [EIDE] (E)IDE hard drive subsystem geometry
1343 Format: <cyl>,<head>,<sect>
1346 Disable Hardware Error Source Table (HEST) support;
1347 corresponding firmware-first mode error processing
1348 logic will be disabled.
1350 highmem=nn[KMG] [KNL,BOOT] forces the highmem zone to have an exact
1351 size of <nn>. This works even on boxes that have no
1352 highmem otherwise. This also works to reduce highmem
1353 size on bigger boxes.
1355 highres= [KNL] Enable/disable high resolution timer mode.
1356 Valid parameters: "on", "off"
1360 See Documentation/isdn/README.HiSax.
1364 hpet= [X86-32,HPET] option to control HPET usage
1365 Format: { enable (default) | disable | force |
1367 disable: disable HPET and use PIT instead
1368 force: allow force enabled of undocumented chips (ICH4,
1370 verbose: show contents of HPET registers during setup
1372 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET
1373 registers. Default set by CONFIG_HPET_MMAP_DEFAULT.
1375 hugepages= [HW,X86-32,IA-64] HugeTLB pages to allocate at boot.
1376 hugepagesz= [HW,IA-64,PPC,X86-64] The size of the HugeTLB pages.
1377 On x86-64 and powerpc, this option can be specified
1378 multiple times interleaved with hugepages= to reserve
1379 huge pages of different sizes. Valid pages sizes on
1380 x86-64 are 2M (when the CPU supports "pse") and 1G
1381 (when the CPU supports the "pdpe1gb" cpuinfo flag).
1384 [KNL] Should the hung task detector generate panics.
1387 A nonzero value instructs the kernel to panic when a
1388 hung task is detected. The default value is controlled
1389 by the CONFIG_BOOTPARAM_HUNG_TASK_PANIC build-time
1390 option. The value selected by this boot parameter can
1391 be changed later by the kernel.hung_task_panic sysctl.
1393 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC)
1394 terminal devices. Valid values: 0..8
1395 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs.
1396 If specified, z/VM IUCV HVC accepts connections
1397 from listed z/VM user IDs only.
1399 hv_nopvspin [X86,HYPER_V] Disables the paravirt spinlock optimizations
1400 which allow the hypervisor to 'idle' the
1401 guest on lock contention.
1404 Do not unregister boot console at start. This is only
1405 useful for debugging when something happens in the window
1406 between unregistering the boot console and initializing
1409 i2c_bus= [HW] Override the default board specific I2C bus speed
1410 or register an additional I2C bus that is not
1411 registered from board initialization code.
1415 i8042.debug [HW] Toggle i8042 debug mode
1416 i8042.unmask_kbd_data
1417 [HW] Enable printing of interrupt data from the KBD port
1418 (disabled by default, and as a pre-condition
1419 requires that i8042.debug=1 be enabled)
1420 i8042.direct [HW] Put keyboard port into non-translated mode
1421 i8042.dumbkbd [HW] Pretend that controller can only read data from
1422 keyboard and cannot control its state
1423 (Don't attempt to blink the leds)
1424 i8042.noaux [HW] Don't check for auxiliary (== mouse) port
1425 i8042.nokbd [HW] Don't check/create keyboard port
1426 i8042.noloop [HW] Disable the AUX Loopback command while probing
1428 i8042.nomux [HW] Don't check presence of an active multiplexing
1430 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX
1432 i8042.notimeout [HW] Ignore timeout condition signalled by controller
1433 i8042.reset [HW] Reset the controller during init, cleanup and
1434 suspend-to-ram transitions, only during s2r
1435 transitions, or never reset
1436 Format: { 1 | Y | y | 0 | N | n }
1437 1, Y, y: always reset controller
1438 0, N, n: don't ever reset controller
1439 Default: only on s2r transitions on x86; most other
1440 architectures force reset to be always executed
1441 i8042.unlock [HW] Unlock (ignore) the keylock
1442 i8042.kbdreset [HW] Reset device connected to KBD port
1446 i8k.ignore_dmi [HW] Continue probing hardware even if DMI data
1447 indicates that the driver is running on unsupported
1449 i8k.force [HW] Activate i8k driver even if SMM BIOS signature
1450 does not match list of supported models.
1452 [HW] Report power status in /proc/i8k
1453 (disabled by default)
1454 i8k.restricted [HW] Allow controlling fans only if SYS_ADMIN
1457 i915.invert_brightness=
1458 [DRM] Invert the sense of the variable that is used to
1459 set the brightness of the panel backlight. Normally a
1460 brightness value of 0 indicates backlight switched off,
1461 and the maximum of the brightness value sets the backlight
1462 to maximum brightness. If this parameter is set to 0
1463 (default) and the machine requires it, or this parameter
1464 is set to 1, a brightness value of 0 sets the backlight
1465 to maximum brightness, and the maximum of the brightness
1466 value switches the backlight off.
1467 -1 -- never invert brightness
1468 0 -- machine default
1469 1 -- force brightness inversion
1472 Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]]
1474 ide-core.nodma= [HW] (E)IDE subsystem
1475 Format: =0.0 to prevent dma on hda, =0.1 hdb =1.0 hdc
1476 .vlb_clock .pci_clock .noflush .nohpa .noprobe .nowerr
1477 .cdrom .chs .ignore_cable are additional options
1478 See Documentation/ide/ide.txt.
1480 ide-generic.probe-mask= [HW] (E)IDE subsystem
1482 Probe mask for legacy ISA IDE ports. Depending on
1483 platform up to 6 ports are supported, enabled by
1484 setting corresponding bits in the mask to 1. The
1485 default value is 0x0, which has a special meaning.
1486 On systems that have PCI, it triggers scanning the
1487 PCI bus for the first and the second port, which
1488 are then probed. On systems without PCI the value
1489 of 0x0 enables probing the two first ports as if it
1492 ide-pci-generic.all-generic-ide [HW] (E)IDE subsystem
1493 Claim all unknown PCI IDE storage controllers.
1496 Format: idle=poll, idle=halt, idle=nomwait
1497 Poll forces a polling idle loop that can slightly
1498 improve the performance of waking up a idle CPU, but
1499 will use a lot of power and make the system run hot.
1501 idle=halt: Halt is forced to be used for CPU idle.
1502 In such case C2/C3 won't be used again.
1503 idle=nomwait: Disable mwait for CPU C-states
1505 ieee754= [MIPS] Select IEEE Std 754 conformance mode
1506 Format: { strict | legacy | 2008 | relaxed }
1509 Choose which programs will be accepted for execution
1510 based on the IEEE 754 NaN encoding(s) supported by
1511 the FPU and the NaN encoding requested with the value
1512 of an ELF file header flag individually set by each
1513 binary. Hardware implementations are permitted to
1514 support either or both of the legacy and the 2008 NaN
1517 Available settings are as follows:
1518 strict accept binaries that request a NaN encoding
1519 supported by the FPU
1520 legacy only accept legacy-NaN binaries, if supported
1522 2008 only accept 2008-NaN binaries, if supported
1524 relaxed accept any binaries regardless of whether
1525 supported by the FPU
1527 The FPU emulator is always able to support both NaN
1528 encodings, so if no FPU hardware is present or it has
1529 been disabled with 'nofpu', then the settings of
1530 'legacy' and '2008' strap the emulator accordingly,
1531 'relaxed' straps the emulator for both legacy-NaN and
1532 2008-NaN, whereas 'strict' enables legacy-NaN only on
1533 legacy processors and both NaN encodings on MIPS32 or
1536 The setting for ABS.fmt/NEG.fmt instruction execution
1537 mode generally follows that for the NaN encoding,
1538 except where unsupported by hardware.
1540 ignore_loglevel [KNL]
1541 Ignore loglevel setting - this will print /all/
1542 kernel messages to the console. Useful for debugging.
1543 We also add it as printk module parameter, so users
1544 could change it dynamically, usually by
1545 /sys/module/printk/parameters/ignore_loglevel.
1548 Ignore RLIMIT_DATA setting for data mappings,
1549 print warning at first misuse. Can be changed via
1550 /sys/module/kernel/parameters/ignore_rlimit_data.
1552 ihash_entries= [KNL]
1553 Set number of hash buckets for inode cache.
1555 ima_appraise= [IMA] appraise integrity measurements
1556 Format: { "off" | "enforce" | "fix" | "log" }
1559 ima_appraise_tcb [IMA]
1560 The builtin appraise policy appraises all files
1563 ima_canonical_fmt [IMA]
1564 Use the canonical format for the binary runtime
1565 measurements, instead of host native format.
1568 Format: { md5 | sha1 | rmd160 | sha256 | sha384
1572 The list of supported hash algorithms is defined
1573 in crypto/hash_info.h.
1576 The builtin policies to load during IMA setup.
1577 Format: "tcb | appraise_tcb | secure_boot |
1580 The "tcb" policy measures all programs exec'd, files
1581 mmap'd for exec, and all files opened with the read
1582 mode bit set by either the effective uid (euid=0) or
1585 The "appraise_tcb" policy appraises the integrity of
1586 all files owned by root. (This is the equivalent
1587 of ima_appraise_tcb.)
1589 The "secure_boot" policy appraises the integrity
1590 of files (eg. kexec kernel image, kernel modules,
1591 firmware, policy, etc) based on file signatures.
1593 The "fail_securely" policy forces file signature
1594 verification failure also on privileged mounted
1595 filesystems with the SB_I_UNVERIFIABLE_SIGNATURE
1598 ima_tcb [IMA] Deprecated. Use ima_policy= instead.
1599 Load a policy which meets the needs of the Trusted
1600 Computing Base. This means IMA will measure all
1601 programs exec'd, files mmap'd for exec, and all files
1602 opened for read by uid=0.
1605 Select one of defined IMA measurements template formats.
1606 Formats: { "ima" | "ima-ng" | "ima-sig" }
1610 [IMA] Define a custom template format.
1611 Format: { "field1|...|fieldN" }
1613 ima.ahash_minsize= [IMA] Minimum file size for asynchronous hash usage
1614 Format: <min_file_size>
1615 Set the minimal file size for using asynchronous hash.
1616 If left unspecified, ahash usage is disabled.
1618 ahash performance varies for different data sizes on
1619 different crypto accelerators. This option can be used
1620 to achieve the best performance for a particular HW.
1622 ima.ahash_bufsize= [IMA] Asynchronous hash buffer size
1624 Set hashing buffer size. Default: 4k.
1626 ahash performance varies for different chunk sizes on
1627 different crypto accelerators. This option can be used
1628 to achieve best performance for particular HW.
1632 Run specified binary instead of /sbin/init as init
1635 initcall_debug [KNL] Trace initcalls as they are executed. Useful
1636 for working out where the kernel is dying during
1639 initcall_blacklist= [KNL] Do not execute a comma-separated list of
1640 initcall functions. Useful for debugging built-in
1641 modules and initcalls.
1643 initrd= [BOOT] Specify the location of the initial ramdisk
1645 init_pkru= [x86] Specify the default memory protection keys rights
1646 register contents for all processes. 0x55555554 by
1647 default (disallow access to all but pkey 0). Can
1648 override in debugfs after boot.
1650 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver
1653 int_pln_enable [x86] Enable power limit notification interrupt
1655 integrity_audit=[IMA]
1656 Format: { "0" | "1" }
1657 0 -- basic integrity auditing messages. (Default)
1658 1 -- additional integrity auditing messages.
1660 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option
1662 Enable intel iommu driver.
1664 Disable intel iommu driver.
1665 igfx_off [Default Off]
1666 By default, gfx is mapped as normal device. If a gfx
1667 device has a dedicated DMAR unit, the DMAR unit is
1668 bypassed by not enabling DMAR with this option. In
1669 this case, gfx device will use physical address for
1672 With this option iommu will not optimize to look
1673 for io virtual address below 32-bit forcing dual
1674 address cycle on pci bus for cards supporting greater
1675 than 32-bit addressing. The default is to look
1676 for translation below 32-bit and if not available
1677 then look in the higher range.
1678 strict [Default Off]
1679 With this option on every unmap_single operation will
1680 result in a hardware IOTLB flush operation as opposed
1681 to batching them for performance.
1682 sp_off [Default Off]
1683 By default, super page will be supported if Intel IOMMU
1684 has the capability. With this option, super page will
1686 ecs_off [Default Off]
1687 By default, extended context tables will be supported if
1688 the hardware advertises that it has support both for the
1689 extended tables themselves, and also PASID support. With
1690 this option set, extended tables will not be used even
1691 on hardware which claims to support them.
1692 tboot_noforce [Default Off]
1693 Do not force the Intel IOMMU enabled under tboot.
1694 By default, tboot will force Intel IOMMU on, which
1695 could harm performance of some high-throughput
1696 devices like 40GBit network cards, even if identity
1698 Note that using this option lowers the security
1699 provided by tboot because it makes the system
1700 vulnerable to DMA attacks.
1702 intel_idle.max_cstate= [KNL,HW,ACPI,X86]
1703 0 disables intel_idle and fall back on acpi_idle.
1704 1 to 9 specify maximum depth of C-state.
1708 Do not enable intel_pstate as the default
1709 scaling driver for the supported processors
1711 Use intel_pstate as a scaling driver, but configure it
1712 to work with generic cpufreq governors (instead of
1713 enabling its internal governor). This mode cannot be
1714 used along with the hardware-managed P-states (HWP)
1717 Enable intel_pstate on systems that prohibit it by default
1718 in favor of acpi-cpufreq. Forcing the intel_pstate driver
1719 instead of acpi-cpufreq may disable platform features, such
1720 as thermal controls and power capping, that rely on ACPI
1721 P-States information being indicated to OSPM and therefore
1722 should be used with caution. This option does not work with
1723 processors that aren't supported by the intel_pstate driver
1724 or on platforms that use pcc-cpufreq instead of acpi-cpufreq.
1726 Do not enable hardware P state control (HWP)
1729 Only load intel_pstate on systems which support
1730 hardware P state control (HWP) if available.
1732 Enforce ACPI _PPC performance limits. If the Fixed ACPI
1733 Description Table, specifies preferred power management
1734 profile as "Enterprise Server" or "Performance Server",
1735 then this feature is turned on by default.
1737 Allow per-logical-CPU P-State performance control limits using
1738 cpufreq sysfs interface
1740 intremap= [X86-64, Intel-IOMMU]
1741 on enable Interrupt Remapping (default)
1742 off disable Interrupt Remapping
1743 nosid disable Source ID checking
1745 BIOS x2APIC opt-out request will be ignored
1746 nopost disable Interrupt Posting
1748 iomem= Disable strict checking of access to MMIO memory
1749 strict regions from userspace.
1764 nobypass [PPC/POWERNV]
1765 Disable IOMMU bypass, using IOMMU for PCI devices.
1767 iommu.strict= [ARM64] Configure TLB invalidation behaviour
1768 Format: { "0" | "1" }
1770 Request that DMA unmap operations use deferred
1771 invalidation of hardware TLBs, for increased
1772 throughput at the cost of reduced device isolation.
1773 Will fall back to strict mode if not supported by
1774 the relevant IOMMU driver.
1775 1 - Strict mode (default).
1776 DMA unmap operations invalidate IOMMU hardware TLBs
1780 [ARM64] Configure DMA to bypass the IOMMU by default.
1781 Format: { "0" | "1" }
1782 0 - Use IOMMU translation for DMA.
1783 1 - Bypass the IOMMU for DMA.
1784 unset - Use value of CONFIG_IOMMU_DEFAULT_PASSTHROUGH.
1786 io7= [HW] IO7 for Marvel based alpha systems
1787 See comment before marvel_specify_io7 in
1788 arch/alpha/kernel/core_marvel.c.
1790 io_delay= [X86] I/O delay method
1792 Standard port 0x80 based delay
1794 Alternate port 0xed based delay (needed on some systems)
1796 Simple two microseconds delay
1801 See Documentation/filesystems/nfs/nfsroot.txt.
1803 irqaffinity= [SMP] Set the default irq affinity mask
1804 The argument is a cpu list, as described above.
1806 irqchip.gicv2_force_probe=
1809 Force the kernel to look for the second 4kB page
1810 of a GICv2 controller even if the memory range
1811 exposed by the device tree is too small.
1813 irqchip.gicv3_nolpi=
1815 Force the kernel to ignore the availability of
1816 LPIs (and by consequence ITSs). Intended for system
1817 that use the kernel as a bootloader, and thus want
1818 to let secondary kernels in charge of setting up
1822 When an interrupt is not handled search all handlers
1823 for it. Intended to get systems with badly broken
1827 When an interrupt is not handled search all handlers
1828 for it. Also check all handlers each timer
1829 interrupt. Intended to get systems with badly broken
1833 Format: <RDP>,<reset>,<pci_scan>,<verbosity>
1835 isolcpus= [KNL,SMP,ISOL] Isolate a given set of CPUs from disturbance.
1836 [Deprecated - use cpusets instead]
1837 Format: [flag-list,]<cpu-list>
1839 Specify one or more CPUs to isolate from disturbances
1840 specified in the flag list (default: domain):
1843 Disable the tick when a single task runs.
1845 A residual 1Hz tick is offloaded to workqueues, which you
1846 need to affine to housekeeping through the global
1847 workqueue's affinity configured via the
1848 /sys/devices/virtual/workqueue/cpumask sysfs file, or
1849 by using the 'domain' flag described below.
1851 NOTE: by default the global workqueue runs on all CPUs,
1852 so to protect individual CPUs the 'cpumask' file has to
1853 be configured manually after bootup.
1856 Isolate from the general SMP balancing and scheduling
1857 algorithms. Note that performing domain isolation this way
1858 is irreversible: it's not possible to bring back a CPU to
1859 the domains once isolated through isolcpus. It's strongly
1860 advised to use cpusets instead to disable scheduler load
1861 balancing through the "cpuset.sched_load_balance" file.
1862 It offers a much more flexible interface where CPUs can
1863 move in and out of an isolated set anytime.
1865 You can move a process onto or off an "isolated" CPU via
1866 the CPU affinity syscalls or cpuset.
1867 <cpu number> begins at 0 and the maximum value is
1868 "number of CPUs in system - 1".
1870 The format of <cpu-list> is described above.
1876 ivrs_ioapic [HW,X86_64]
1877 Provide an override to the IOAPIC-ID<->DEVICE-ID
1878 mapping provided in the IVRS ACPI table. For
1879 example, to map IOAPIC-ID decimal 10 to
1880 PCI device 00:14.0 write the parameter as:
1881 ivrs_ioapic[10]=00:14.0
1883 ivrs_hpet [HW,X86_64]
1884 Provide an override to the HPET-ID<->DEVICE-ID
1885 mapping provided in the IVRS ACPI table. For
1886 example, to map HPET-ID decimal 0 to
1887 PCI device 00:14.0 write the parameter as:
1888 ivrs_hpet[0]=00:14.0
1890 ivrs_acpihid [HW,X86_64]
1891 Provide an override to the ACPI-HID:UID<->DEVICE-ID
1892 mapping provided in the IVRS ACPI table. For
1893 example, to map UART-HID:UID AMD0020:0 to
1894 PCI device 00:14.5 write the parameter as:
1895 ivrs_acpihid[00:14.5]=AMD0020:0
1897 js= [HW,JOY] Analog joystick
1898 See Documentation/input/joydev/joystick.rst.
1901 When CONFIG_RANDOMIZE_BASE is set, this disables
1902 kernel and module base offset ASLR (Address Space
1903 Layout Randomization).
1906 [KNL] Enforce KASAN (Kernel Address Sanitizer) to print
1907 report on every invalid memory access. Without this
1908 parameter KASAN will print report only for the first
1913 kernelcore= [KNL,X86,IA-64,PPC]
1914 Format: nn[KMGTPE] | nn% | "mirror"
1915 This parameter specifies the amount of memory usable by
1916 the kernel for non-movable allocations. The requested
1917 amount is spread evenly throughout all nodes in the
1918 system as ZONE_NORMAL. The remaining memory is used for
1919 movable memory in its own zone, ZONE_MOVABLE. In the
1920 event, a node is too small to have both ZONE_NORMAL and
1921 ZONE_MOVABLE, kernelcore memory will take priority and
1922 other nodes will have a larger ZONE_MOVABLE.
1924 ZONE_MOVABLE is used for the allocation of pages that
1925 may be reclaimed or moved by the page migration
1926 subsystem. Note that allocations like PTEs-from-HighMem
1927 still use the HighMem zone if it exists, and the Normal
1928 zone if it does not.
1930 It is possible to specify the exact amount of memory in
1931 the form of "nn[KMGTPE]", a percentage of total system
1932 memory in the form of "nn%", or "mirror". If "mirror"
1933 option is specified, mirrored (reliable) memory is used
1934 for non-movable allocations and remaining memory is used
1935 for Movable pages. "nn[KMGTPE]", "nn%", and "mirror"
1936 are exclusive, so you cannot specify multiple forms.
1938 kgdbdbgp= [KGDB,HW] kgdb over EHCI usb debug port.
1939 Format: <Controller#>[,poll interval]
1940 The controller # is the number of the ehci usb debug
1941 port as it is probed via PCI. The poll interval is
1942 optional and is the number seconds in between
1943 each poll cycle to the debug port in case you need
1944 the functionality for interrupting the kernel with
1945 gdb or control-c on the dbgp connection. When
1946 not using this parameter you use sysrq-g to break into
1947 the kernel debugger.
1949 kgdboc= [KGDB,HW] kgdb over consoles.
1950 Requires a tty driver that supports console polling,
1951 or a supported polling keyboard driver (non-usb).
1952 Serial only format: <serial_device>[,baud]
1953 keyboard only format: kbd
1954 keyboard and serial format: kbd,<serial_device>[,baud]
1955 Optional Kernel mode setting:
1956 kms, kbd format: kms,kbd
1957 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud]
1959 kgdbwait [KGDB] Stop kernel execution and enter the
1960 kernel debugger at the earliest opportunity.
1962 kmac= [MIPS] korina ethernet MAC address.
1963 Configure the RouterBoard 532 series on-chip
1964 Ethernet adapter MAC address.
1966 kmemleak= [KNL] Boot-time kmemleak enable/disable
1967 Valid arguments: on, off
1969 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y,
1972 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs.
1973 Default is 0 (don't ignore, but inject #GP)
1975 kvm.enable_vmware_backdoor=[KVM] Support VMware backdoor PV interface.
1976 Default is false (don't support).
1978 kvm.mmu_audit= [KVM] This is a R/W parameter which allows audit
1982 kvm-amd.nested= [KVM,AMD] Allow nested virtualization in KVM/SVM.
1983 Default is 1 (enabled)
1985 kvm-amd.npt= [KVM,AMD] Disable nested paging (virtualized MMU)
1987 Default is 1 (enabled) if in 64-bit or 32-bit PAE mode.
1989 kvm-arm.vgic_v3_group0_trap=
1990 [KVM,ARM] Trap guest accesses to GICv3 group-0
1993 kvm-arm.vgic_v3_group1_trap=
1994 [KVM,ARM] Trap guest accesses to GICv3 group-1
1997 kvm-arm.vgic_v3_common_trap=
1998 [KVM,ARM] Trap guest accesses to GICv3 common
2001 kvm-arm.vgic_v4_enable=
2002 [KVM,ARM] Allow use of GICv4 for direct injection of
2005 kvm-intel.ept= [KVM,Intel] Disable extended page tables
2006 (virtualized MMU) support on capable Intel chips.
2007 Default is 1 (enabled)
2009 kvm-intel.emulate_invalid_guest_state=
2010 [KVM,Intel] Enable emulation of invalid guest states
2011 Default is 0 (disabled)
2013 kvm-intel.flexpriority=
2014 [KVM,Intel] Disable FlexPriority feature (TPR shadow).
2015 Default is 1 (enabled)
2018 [KVM,Intel] Enable VMX nesting (nVMX).
2019 Default is 0 (disabled)
2021 kvm-intel.unrestricted_guest=
2022 [KVM,Intel] Disable unrestricted guest feature
2023 (virtualized real and unpaged mode) on capable
2024 Intel chips. Default is 1 (enabled)
2026 kvm-intel.vmentry_l1d_flush=[KVM,Intel] Mitigation for L1 Terminal Fault
2029 Valid arguments: never, cond, always
2031 always: L1D cache flush on every VMENTER.
2032 cond: Flush L1D on VMENTER only when the code between
2033 VMEXIT and VMENTER can leak host memory.
2034 never: Disables the mitigation
2036 Default is cond (do L1 cache flush in specific instances)
2038 kvm-intel.vpid= [KVM,Intel] Disable Virtual Processor Identification
2039 feature (tagged TLBs) on capable Intel chips.
2040 Default is 1 (enabled)
2042 l1tf= [X86] Control mitigation of the L1TF vulnerability on
2045 The kernel PTE inversion protection is unconditionally
2046 enabled and cannot be disabled.
2049 Provides all available mitigations for the
2050 L1TF vulnerability. Disables SMT and
2051 enables all mitigations in the
2052 hypervisors, i.e. unconditional L1D flush.
2054 SMT control and L1D flush control via the
2055 sysfs interface is still possible after
2056 boot. Hypervisors will issue a warning
2057 when the first VM is started in a
2058 potentially insecure configuration,
2059 i.e. SMT enabled or L1D flush disabled.
2062 Same as 'full', but disables SMT and L1D
2063 flush runtime control. Implies the
2064 'nosmt=force' command line option.
2065 (i.e. sysfs control of SMT is disabled.)
2068 Leaves SMT enabled and enables the default
2069 hypervisor mitigation, i.e. conditional
2072 SMT control and L1D flush control via the
2073 sysfs interface is still possible after
2074 boot. Hypervisors will issue a warning
2075 when the first VM is started in a
2076 potentially insecure configuration,
2077 i.e. SMT enabled or L1D flush disabled.
2081 Disables SMT and enables the default
2082 hypervisor mitigation.
2084 SMT control and L1D flush control via the
2085 sysfs interface is still possible after
2086 boot. Hypervisors will issue a warning
2087 when the first VM is started in a
2088 potentially insecure configuration,
2089 i.e. SMT enabled or L1D flush disabled.
2092 Same as 'flush', but hypervisors will not
2093 warn when a VM is started in a potentially
2094 insecure configuration.
2097 Disables hypervisor mitigations and doesn't
2102 For details see: Documentation/admin-guide/l1tf.rst
2108 lapic [X86-32,APIC] Enable the local APIC even if BIOS
2111 lapic= [x86,APIC] "notscdeadline" Do not use TSC deadline
2112 value for LAPIC timer one-shot implementation. Default
2113 back to the programmable timer unit in the LAPIC.
2115 lapic_timer_c2_ok [X86,APIC] trust the local apic timer
2118 libata.dma= [LIBATA] DMA control
2119 libata.dma=0 Disable all PATA and SATA DMA
2120 libata.dma=1 PATA and SATA Disk DMA only
2121 libata.dma=2 ATAPI (CDROM) DMA only
2122 libata.dma=4 Compact Flash DMA only
2123 Combinations also work, so libata.dma=3 enables DMA
2124 for disks and CDROMs, but not CFs.
2126 libata.ignore_hpa= [LIBATA] Ignore HPA limit
2127 libata.ignore_hpa=0 keep BIOS limits (default)
2128 libata.ignore_hpa=1 ignore limits, using full disk
2130 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume
2134 libata.force= [LIBATA] Force configurations. The format is comma
2135 separated list of "[ID:]VAL" where ID is
2136 PORT[.DEVICE]. PORT and DEVICE are decimal numbers
2137 matching port, link or device. Basically, it matches
2138 the ATA ID string printed on console by libata. If
2139 the whole ID part is omitted, the last PORT and DEVICE
2140 values are used. If ID hasn't been specified yet, the
2141 configuration applies to all ports, links and devices.
2143 If only DEVICE is omitted, the parameter applies to
2144 the port and all links and devices behind it. DEVICE
2145 number of 0 either selects the first device or the
2146 first fan-out link behind PMP device. It does not
2147 select the host link. DEVICE number of 15 selects the
2148 host link and device attached to it.
2150 The VAL specifies the configuration to force. As long
2151 as there's no ambiguity shortcut notation is allowed.
2152 For example, both 1.5 and 1.5G would work for 1.5Gbps.
2153 The following configurations can be forced.
2155 * Cable type: 40c, 80c, short40c, unk, ign or sata.
2156 Any ID with matching PORT is used.
2158 * SATA link speed limit: 1.5Gbps or 3.0Gbps.
2160 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7].
2161 udma[/][16,25,33,44,66,100,133] notation is also
2164 * [no]ncq: Turn on or off NCQ.
2166 * [no]ncqtrim: Turn off queued DSM TRIM.
2168 * nohrst, nosrst, norst: suppress hard, soft
2171 * rstonce: only attempt one reset during
2172 hot-unplug link recovery
2174 * dump_id: dump IDENTIFY data.
2176 * atapi_dmadir: Enable ATAPI DMADIR bridge support
2178 * disable: Disable this device.
2180 If there are multiple matching configurations changing
2181 the same attribute, the last one is used.
2183 memblock=debug [KNL] Enable memblock debug messages.
2185 load_ramdisk= [RAM] List of ramdisks to load from floppy
2186 See Documentation/blockdev/ramdisk.txt.
2188 lockd.nlm_grace_period=P [NFS] Assign grace period.
2191 lockd.nlm_tcpport=N [NFS] Assign TCP port.
2194 lockd.nlm_timeout=T [NFS] Assign timeout value.
2197 lockd.nlm_udpport=M [NFS] Assign UDP port.
2200 locktorture.nreaders_stress= [KNL]
2201 Set the number of locking read-acquisition kthreads.
2202 Defaults to being automatically set based on the
2203 number of online CPUs.
2205 locktorture.nwriters_stress= [KNL]
2206 Set the number of locking write-acquisition kthreads.
2208 locktorture.onoff_holdoff= [KNL]
2209 Set time (s) after boot for CPU-hotplug testing.
2211 locktorture.onoff_interval= [KNL]
2212 Set time (s) between CPU-hotplug operations, or
2213 zero to disable CPU-hotplug testing.
2215 locktorture.shuffle_interval= [KNL]
2216 Set task-shuffle interval (jiffies). Shuffling
2217 tasks allows some CPUs to go into dyntick-idle
2218 mode during the locktorture test.
2220 locktorture.shutdown_secs= [KNL]
2221 Set time (s) after boot system shutdown. This
2222 is useful for hands-off automated testing.
2224 locktorture.stat_interval= [KNL]
2225 Time (s) between statistics printk()s.
2227 locktorture.stutter= [KNL]
2228 Time (s) to stutter testing, for example,
2229 specifying five seconds causes the test to run for
2230 five seconds, wait for five seconds, and so on.
2231 This tests the locking primitive's ability to
2232 transition abruptly to and from idle.
2234 locktorture.torture_type= [KNL]
2235 Specify the locking implementation to test.
2237 locktorture.verbose= [KNL]
2238 Enable additional printk() statements.
2240 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver
2243 loglevel= All Kernel Messages with a loglevel smaller than the
2244 console loglevel will be printed to the console. It can
2245 also be changed with klogd or other programs. The
2246 loglevels are defined as follows:
2248 0 (KERN_EMERG) system is unusable
2249 1 (KERN_ALERT) action must be taken immediately
2250 2 (KERN_CRIT) critical conditions
2251 3 (KERN_ERR) error conditions
2252 4 (KERN_WARNING) warning conditions
2253 5 (KERN_NOTICE) normal but significant condition
2254 6 (KERN_INFO) informational
2255 7 (KERN_DEBUG) debug-level messages
2257 log_buf_len=n[KMG] Sets the size of the printk ring buffer,
2258 in bytes. n must be a power of two and greater
2259 than the minimal size. The minimal size is defined
2260 by LOG_BUF_SHIFT kernel config parameter. There is
2261 also CONFIG_LOG_CPU_MAX_BUF_SHIFT config parameter
2262 that allows to increase the default size depending on
2263 the number of CPUs. See init/Kconfig for more details.
2265 logo.nologo [FB] Disables display of the built-in Linux logo.
2266 This may be used to provide more screen space for
2267 kernel log messages and is useful when debugging
2268 kernel boot problems.
2270 lp=0 [LP] Specify parallel ports to use, e.g,
2271 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses
2272 lp=reset first parallel port). 'lp=0' disables the
2273 lp=auto printer driver. 'lp=reset' (which can be
2274 specified in addition to the ports) causes
2275 attached printers to be reset. Using
2276 lp=port1,port2,... specifies the parallel ports
2277 to associate lp devices with, starting with
2278 lp0. A port specification may be 'none' to skip
2279 that lp device, or a parport name such as
2280 'parport0'. Specifying 'lp=auto' instead of a
2281 port specification list means that device IDs
2282 from each port should be examined, to see if
2283 an IEEE 1284-compliant printer is attached; if
2284 so, the driver will manage that printer.
2285 See also header of drivers/char/lp.c.
2288 Sets loops_per_jiffy to given constant, thus avoiding
2289 time-consuming boot-time autodetection (up to 250 ms per
2290 CPU). 0 enables autodetection (default). To determine
2291 the correct value for your kernel, boot with normal
2292 autodetection and see what value is printed. Note that
2293 on SMP systems the preset will be applied to all CPUs,
2294 which is likely to cause problems if your CPUs need
2295 significantly divergent settings. An incorrect value
2296 will cause delays in the kernel to be wrong, leading to
2297 unpredictable I/O errors and other breakage. Although
2298 unlikely, in the extreme case this might damage your
2302 Format: <io>,<irq>,<dma>
2304 lsm.debug [SECURITY] Enable LSM initialization debugging output.
2306 machvec= [IA-64] Force the use of a particular machine-vector
2307 (machvec) in a generic kernel.
2308 Example: machvec=hpzx1_swiotlb
2310 machtype= [Loongson] Share the same kernel image file between different
2312 Example: machtype=lemote-yeeloong-2f-7inch
2314 max_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory greater
2315 than or equal to this physical address is ignored.
2317 maxcpus= [SMP] Maximum number of processors that an SMP kernel
2318 will bring up during bootup. maxcpus=n : n >= 0 limits
2319 the kernel to bring up 'n' processors. Surely after
2320 bootup you can bring up the other plugged cpu by executing
2321 "echo 1 > /sys/devices/system/cpu/cpuX/online". So maxcpus
2322 only takes effect during system bootup.
2323 While n=0 is a special case, it is equivalent to "nosmp",
2324 which also disables the IO APIC.
2326 max_loop= [LOOP] The number of loop block devices that get
2327 (loop.max_loop) unconditionally pre-created at init time. The default
2328 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead
2329 of statically allocating a predefined number, loop
2330 devices can be requested on-demand with the
2331 /dev/loop-control interface.
2333 mce [X86-32] Machine Check Exception
2335 mce=option [X86-64] See Documentation/x86/x86_64/boot-options.txt
2337 md= [HW] RAID subsystems devices and level
2338 See Documentation/admin-guide/md.rst.
2341 Format: <first>,<last>
2342 Specifies range of consoles to be captured by the MDA.
2344 mem=nn[KMG] [KNL,BOOT] Force usage of a specific amount of memory
2345 Amount of memory to be used when the kernel is not able
2346 to see the whole system memory or for test.
2347 [X86] Work as limiting max address. Use together
2348 with memmap= to avoid physical address space collisions.
2349 Without memmap= PCI devices could be placed at addresses
2350 belonging to unused RAM.
2352 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel
2356 [KNL,SH] Allow user to override the default size for
2357 per-device physically contiguous DMA buffers.
2359 memhp_default_state=online/offline
2360 [KNL] Set the initial state for the memory hotplug
2361 onlining policy. If not specified, the default value is
2362 set according to the
2363 CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE kernel config
2365 See Documentation/memory-hotplug.txt.
2367 memmap=exactmap [KNL,X86] Enable setting of an exact
2368 E820 memory map, as specified by the user.
2369 Such memmap=exactmap lines can be constructed based on
2370 BIOS output or other requirements. See the memmap=nn@ss
2373 memmap=nn[KMG]@ss[KMG]
2374 [KNL] Force usage of a specific region of memory.
2375 Region of memory to be used is from ss to ss+nn.
2376 If @ss[KMG] is omitted, it is equivalent to mem=nn[KMG],
2377 which limits max address to nn[KMG].
2378 Multiple different regions can be specified,
2381 memmap=100M@2G,100M#3G,1G!1024G
2383 memmap=nn[KMG]#ss[KMG]
2384 [KNL,ACPI] Mark specific memory as ACPI data.
2385 Region of memory to be marked is from ss to ss+nn.
2387 memmap=nn[KMG]$ss[KMG]
2388 [KNL,ACPI] Mark specific memory as reserved.
2389 Region of memory to be reserved is from ss to ss+nn.
2390 Example: Exclude memory from 0x18690000-0x1869ffff
2391 memmap=64K$0x18690000
2393 memmap=0x10000$0x18690000
2394 Some bootloaders may need an escape character before '$',
2395 like Grub2, otherwise '$' and the following number
2398 memmap=nn[KMG]!ss[KMG]
2399 [KNL,X86] Mark specific memory as protected.
2400 Region of memory to be used, from ss to ss+nn.
2401 The memory region may be marked as e820 type 12 (0xc)
2402 and is NVDIMM or ADR memory.
2404 memmap=<size>%<offset>-<oldtype>+<newtype>
2405 [KNL,ACPI] Convert memory within the specified region
2406 from <oldtype> to <newtype>. If "-<oldtype>" is left
2407 out, the whole region will be marked as <newtype>,
2408 even if previously unavailable. If "+<newtype>" is left
2409 out, matching memory will be removed. Types are
2410 specified as e820 types, e.g., 1 = RAM, 2 = reserved,
2411 3 = ACPI, 12 = PRAM.
2413 memory_corruption_check=0/1 [X86]
2414 Some BIOSes seem to corrupt the first 64k of
2415 memory when doing things like suspend/resume.
2416 Setting this option will scan the memory
2417 looking for corruption. Enabling this will
2418 both detect corruption and prevent the kernel
2419 from using the memory being corrupted.
2420 However, its intended as a diagnostic tool; if
2421 repeatable BIOS-originated corruption always
2422 affects the same memory, you can use memmap=
2423 to prevent the kernel from using that memory.
2425 memory_corruption_check_size=size [X86]
2426 By default it checks for corruption in the low
2427 64k, making this memory unavailable for normal
2428 use. Use this parameter to scan for
2429 corruption in more or less memory.
2431 memory_corruption_check_period=seconds [X86]
2432 By default it checks for corruption every 60
2433 seconds. Use this parameter to check at some
2434 other rate. 0 disables periodic checking.
2436 memtest= [KNL,X86,ARM,PPC] Enable memtest
2438 default : 0 <disable>
2439 Specifies the number of memtest passes to be
2440 performed. Each pass selects another test
2441 pattern from a given set of patterns. Memtest
2442 fills the memory with this pattern, validates
2443 memory contents and reserves bad memory
2444 regions that are detected.
2446 mem_encrypt= [X86-64] AMD Secure Memory Encryption (SME) control
2447 Valid arguments: on, off
2448 Default (depends on kernel configuration option):
2449 on (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=y)
2450 off (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=n)
2451 mem_encrypt=on: Activate SME
2452 mem_encrypt=off: Do not activate SME
2454 Refer to Documentation/x86/amd-memory-encryption.txt
2455 for details on when memory encryption can be activated.
2457 mem_sleep_default= [SUSPEND] Default system suspend mode:
2458 s2idle - Suspend-To-Idle
2459 shallow - Power-On Suspend or equivalent (if supported)
2460 deep - Suspend-To-RAM or equivalent (if supported)
2461 See Documentation/admin-guide/pm/sleep-states.rst.
2463 meye.*= [HW] Set MotionEye Camera parameters
2464 See Documentation/media/v4l-drivers/meye.rst.
2466 mfgpt_irq= [IA-32] Specify the IRQ to use for the
2467 Multi-Function General Purpose Timers on AMD Geode
2470 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when
2471 the BIOS has incorrectly applied a workaround. TinyBIOS
2472 version 0.98 is known to be affected, 0.99 fixes the
2473 problem by letting the user disable the workaround.
2477 min_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory below this
2478 physical address is ignored.
2480 mini2440= [ARM,HW,KNL]
2481 Format:[0..2][b][c][t]
2483 MINI2440 configuration specification:
2484 0 - The attached screen is the 3.5" TFT
2485 1 - The attached screen is the 7" TFT
2486 2 - The VGA Shield is attached (1024x768)
2487 Leaving out the screen size parameter will not load
2488 the TFT driver, and the framebuffer will be left
2490 b - Enable backlight. The TFT backlight pin will be
2491 linked to the kernel VESA blanking code and a GPIO
2492 LED. This parameter is not necessary when using the
2494 c - Enable the s3c camera interface.
2495 t - Reserved for enabling touchscreen support. The
2496 touchscreen support is not enabled in the mainstream
2497 kernel as of 2.6.30, a preliminary port can be found
2498 in the "bleeding edge" mini2440 support kernel at
2499 http://repo.or.cz/w/linux-2.6/mini2440.git
2502 [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this
2503 parameter allows control of the logging verbosity for
2504 the additional memory initialisation checks. A value
2505 of 0 disables mminit logging and a level of 4 will
2506 log everything. Information is printed at KERN_DEBUG
2507 so loglevel=8 may also need to be specified.
2510 [KNL] When CONFIG_MODULE_SIG is set, this means that
2511 modules without (valid) signatures will fail to load.
2512 Note that if CONFIG_MODULE_SIG_FORCE is set, that
2513 is always true, so this option does nothing.
2515 module_blacklist= [KNL] Do not load a comma-separated list of
2516 modules. Useful for debugging problem modules.
2519 [MOUSE] Maximum time between finger touching and
2520 leaving touchpad surface for touch to be considered
2521 a tap and be reported as a left button click (for
2522 touchpads working in absolute mode only).
2524 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices
2525 reporting absolute coordinates, such as tablets
2526 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices
2527 reporting absolute coordinates, such as tablets
2529 movablecore= [KNL,X86,IA-64,PPC]
2530 Format: nn[KMGTPE] | nn%
2531 This parameter is the complement to kernelcore=, it
2532 specifies the amount of memory used for migratable
2533 allocations. If both kernelcore and movablecore is
2534 specified, then kernelcore will be at *least* the
2535 specified value but may be more. If movablecore on its
2536 own is specified, the administrator must be careful
2537 that the amount of memory usable for all allocations
2540 movable_node [KNL] Boot-time switch to make hotplugable memory
2541 NUMA nodes to be movable. This means that the memory
2542 of such nodes will be usable only for movable
2543 allocations which rules out almost all kernel
2544 allocations. Use with caution!
2546 MTD_Partition= [MTD]
2547 Format: <name>,<region-number>,<size>,<offset>
2549 MTD_Region= [MTD] Format:
2550 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>]
2553 See drivers/mtd/cmdlinepart.c.
2555 multitce=off [PPC] This parameter disables the use of the pSeries
2556 firmware feature for updating multiple TCE entries
2559 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration
2561 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock]
2563 boundary - index of last SLC block on Flex-OneNAND.
2564 The remaining blocks are configured as MLC blocks.
2565 lock - Configure if Flex-OneNAND boundary should be locked.
2566 Once locked, the boundary cannot be changed.
2567 1 indicates lock status, 0 indicates unlock status.
2570 ARM/S3C2412 JIVE boot control
2572 See arch/arm/mach-s3c2412/mach-jive.c
2574 mtouchusb.raw_coordinates=
2575 [HW] Make the MicroTouch USB driver use raw coordinates
2576 ('y', default) or cooked coordinates ('n')
2578 mtrr_chunk_size=nn[KMG] [X86]
2579 used for mtrr cleanup. It is largest continuous chunk
2580 that could hold holes aka. UC entries.
2582 mtrr_gran_size=nn[KMG] [X86]
2583 Used for mtrr cleanup. It is granularity of mtrr block.
2585 Large value could prevent small alignment from
2588 mtrr_spare_reg_nr=n [X86]
2590 Range: 0,7 : spare reg number
2592 Used for mtrr cleanup. It is spare mtrr entries number.
2593 Set to 2 or more if your graphical card needs more.
2595 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card
2597 netdev= [NET] Network devices parameters
2598 Format: <irq>,<io>,<mem_start>,<mem_end>,<name>
2599 Note that mem_start is often overloaded to mean
2600 something different and driver-specific.
2601 This usage is only documented in each driver source
2605 [NETFILTER] Enable connection tracking flow accounting
2606 0 to disable accounting
2607 1 to enable accounting
2610 nfsaddrs= [NFS] Deprecated. Use ip= instead.
2611 See Documentation/filesystems/nfs/nfsroot.txt.
2613 nfsroot= [NFS] nfs root filesystem for disk-less boxes.
2614 See Documentation/filesystems/nfs/nfsroot.txt.
2616 nfsrootdebug [NFS] enable nfsroot debugging messages.
2617 See Documentation/filesystems/nfs/nfsroot.txt.
2619 nfs.callback_nr_threads=
2620 [NFSv4] set the total number of threads that the
2621 NFS client will assign to service NFSv4 callback
2624 nfs.callback_tcpport=
2625 [NFS] set the TCP port on which the NFSv4 callback
2626 channel should listen.
2629 [NFS] sets the pathname to the program which is used
2630 to update the NFS client cache entries.
2632 nfs.cache_getent_timeout=
2633 [NFS] sets the timeout after which an attempt to
2634 update a cache entry is deemed to have failed.
2636 nfs.idmap_cache_timeout=
2637 [NFS] set the maximum lifetime for idmapper cache
2641 [NFS] enable 64-bit inode numbers.
2642 If zero, the NFS client will fake up a 32-bit inode
2643 number for the readdir() and stat() syscalls instead
2644 of returning the full 64-bit number.
2645 The default is to return 64-bit inode numbers.
2647 nfs.max_session_cb_slots=
2648 [NFSv4.1] Sets the maximum number of session
2649 slots the client will assign to the callback
2650 channel. This determines the maximum number of
2651 callbacks the client will process in parallel for
2652 a particular server.
2654 nfs.max_session_slots=
2655 [NFSv4.1] Sets the maximum number of session slots
2656 the client will attempt to negotiate with the server.
2657 This limits the number of simultaneous RPC requests
2658 that the client can send to the NFSv4.1 server.
2659 Note that there is little point in setting this
2660 value higher than the max_tcp_slot_table_limit.
2662 nfs.nfs4_disable_idmapping=
2663 [NFSv4] When set to the default of '1', this option
2664 ensures that both the RPC level authentication
2665 scheme and the NFS level operations agree to use
2666 numeric uids/gids if the mount is using the
2667 'sec=sys' security flavour. In effect it is
2668 disabling idmapping, which can make migration from
2669 legacy NFSv2/v3 systems to NFSv4 easier.
2670 Servers that do not support this mode of operation
2671 will be autodetected by the client, and it will fall
2672 back to using the idmapper.
2673 To turn off this behaviour, set the value to '0'.
2675 [NFS4] Specify an additional fixed unique ident-
2676 ification string that NFSv4 clients can insert into
2677 their nfs_client_id4 string. This is typically a
2678 UUID that is generated at system install time.
2680 nfs.send_implementation_id =
2681 [NFSv4.1] Send client implementation identification
2682 information in exchange_id requests.
2683 If zero, no implementation identification information
2685 The default is to send the implementation identification
2688 nfs.recover_lost_locks =
2689 [NFSv4] Attempt to recover locks that were lost due
2690 to a lease timeout on the server. Please note that
2691 doing this risks data corruption, since there are
2692 no guarantees that the file will remain unchanged
2693 after the locks are lost.
2694 If you want to enable the kernel legacy behaviour of
2695 attempting to recover these locks, then set this
2697 The default parameter value of '0' causes the kernel
2698 not to attempt recovery of lost locks.
2700 nfs4.layoutstats_timer =
2701 [NFSv4.2] Change the rate at which the kernel sends
2702 layoutstats to the pNFS metadata server.
2704 Setting this to value to 0 causes the kernel to use
2705 whatever value is the default set by the layout
2706 driver. A non-zero value sets the minimum interval
2707 in seconds between layoutstats transmissions.
2709 nfsd.nfs4_disable_idmapping=
2710 [NFSv4] When set to the default of '1', the NFSv4
2711 server will return only numeric uids and gids to
2712 clients using auth_sys, and will accept numeric uids
2713 and gids from such clients. This is intended to ease
2714 migration from NFSv2/v3.
2716 nmi_debug= [KNL,SH] Specify one or more actions to take
2717 when a NMI is triggered.
2718 Format: [state][,regs][,debounce][,die]
2720 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels
2721 Format: [panic,][nopanic,][num]
2723 0 - turn hardlockup detector in nmi_watchdog off
2724 1 - turn hardlockup detector in nmi_watchdog on
2725 When panic is specified, panic when an NMI watchdog
2726 timeout occurs (or 'nopanic' to override the opposite
2727 default). To disable both hard and soft lockup detectors,
2728 please see 'nowatchdog'.
2729 This is useful when you use a panic=... timeout and
2730 need the box quickly up again.
2732 These settings can be accessed at runtime via
2733 the nmi_watchdog and hardlockup_panic sysctls.
2735 netpoll.carrier_timeout=
2736 [NET] Specifies amount of time (in seconds) that
2737 netpoll should wait for a carrier. By default netpoll
2740 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths
2741 emulation library even if a 387 maths coprocessor
2744 no5lvl [X86-64] Disable 5-level paging mode. Forces
2745 kernel to use 4-level paging instead.
2748 [HW] Never suspend the console
2749 Disable suspending of consoles during suspend and
2750 hibernate operations. Once disabled, debugging
2751 messages can reach various consoles while the rest
2752 of the system is being put to sleep (ie, while
2753 debugging driver suspend/resume hooks). This may
2754 not work reliably with all consoles, but is known
2755 to work with serial and VGA consoles.
2756 To facilitate more flexible debugging, we also add
2757 console_suspend, a printk module parameter to control
2758 it. Users could use console_suspend (usually
2759 /sys/module/printk/parameters/console_suspend) to
2760 turn on/off it dynamically.
2762 noaliencache [MM, NUMA, SLAB] Disables the allocation of alien
2763 caches in the slab allocator. Saves per-node memory,
2764 but will impact performance.
2768 noaltinstr [S390] Disables alternative instructions patching
2769 (CPU alternatives feature).
2771 noapic [SMP,APIC] Tells the kernel to not make use of any
2772 IOAPICs that may be present in the system.
2774 noautogroup Disable scheduler automatic task group creation.
2776 nobats [PPC] Do not use BATs for mapping kernel lowmem
2777 on "Classic" PPC cores.
2781 noclflush [BUGS=X86] Don't use the CLFLUSH instruction
2783 nodelayacct [KNL] Disable per-task delay accounting
2785 nodsp [SH] Disable hardware DSP at boot time.
2787 noefi Disable EFI runtime services support.
2792 On X86-32 available only on PAE configured kernels.
2793 noexec=on: enable non-executable mappings (default)
2794 noexec=off: disable non-executable mappings
2797 Disable SMAP (Supervisor Mode Access Prevention)
2798 even if it is supported by processor.
2801 Disable SMEP (Supervisor Mode Execution Prevention)
2802 even if it is supported by processor.
2805 This affects only 32-bit executables.
2806 noexec32=on: enable non-executable mappings (default)
2807 read doesn't imply executable mappings
2808 noexec32=off: disable non-executable mappings
2809 read implies executable mappings
2811 nofpu [MIPS,SH] Disable hardware FPU at boot time.
2813 nofxsr [BUGS=X86-32] Disables x86 floating point extended
2814 register save and restore. The kernel will only save
2815 legacy floating-point registers on task switch.
2817 nohugeiomap [KNL,x86] Disable kernel huge I/O mappings.
2819 nosmt [KNL,S390] Disable symmetric multithreading (SMT).
2820 Equivalent to smt=1.
2822 [KNL,x86] Disable symmetric multithreading (SMT).
2823 nosmt=force: Force disable SMT, cannot be undone
2824 via the sysfs control file.
2826 nospectre_v1 [PPC] Disable mitigations for Spectre Variant 1 (bounds
2827 check bypass). With this option data leaks are possible
2830 nospectre_v2 [X86] Disable all mitigations for the Spectre variant 2
2831 (indirect branch prediction) vulnerability. System may
2832 allow data leaks with this option, which is equivalent
2835 nospec_store_bypass_disable
2836 [HW] Disable all mitigations for the Speculative Store Bypass vulnerability
2838 noxsave [BUGS=X86] Disables x86 extended register state save
2839 and restore using xsave. The kernel will fallback to
2840 enabling legacy floating-point and sse state.
2842 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended
2843 register states. The kernel will fall back to use
2844 xsave to save the states. By using this parameter,
2845 performance of saving the states is degraded because
2846 xsave doesn't support modified optimization while
2847 xsaveopt supports it on xsaveopt enabled systems.
2849 noxsaves [X86] Disables xsaves and xrstors used in saving and
2850 restoring x86 extended register state in compacted
2851 form of xsave area. The kernel will fall back to use
2852 xsaveopt and xrstor to save and restore the states
2853 in standard form of xsave area. By using this
2854 parameter, xsave area per process might occupy more
2855 memory on xsaves enabled systems.
2857 nohlt [BUGS=ARM,SH] Tells the kernel that the sleep(SH) or
2858 wfi(ARM) instruction doesn't work correctly and not to
2859 use it. This is also useful when using JTAG debugger.
2861 no_file_caps Tells the kernel not to honor file capabilities. The
2862 only way then for a file to be executed with privilege
2863 is to be setuid root or executed by root.
2865 nohalt [IA-64] Tells the kernel not to use the power saving
2866 function PAL_HALT_LIGHT when idle. This increases
2867 power-consumption. On the positive side, it reduces
2868 interrupt wake-up latency, which may improve performance
2869 in certain environments such as networked servers or
2872 nohibernate [HIBERNATION] Disable hibernation and resume.
2874 nohz= [KNL] Boottime enable/disable dynamic ticks
2875 Valid arguments: on, off
2878 nohz_full= [KNL,BOOT,SMP,ISOL]
2879 The argument is a cpu list, as described above.
2880 In kernels built with CONFIG_NO_HZ_FULL=y, set
2881 the specified list of CPUs whose tick will be stopped
2882 whenever possible. The boot CPU will be forced outside
2883 the range to maintain the timekeeping. Any CPUs
2884 in this list will have their RCU callbacks offloaded,
2885 just as if they had also been called out in the
2886 rcu_nocbs= boot parameter.
2888 noiotrap [SH] Disables trapped I/O port accesses.
2890 noirqdebug [X86-32] Disables the code which attempts to detect and
2891 disable unhandled interrupt sources.
2893 no_timer_check [X86,APIC] Disables the code which tests for
2894 broken timer IRQ sources.
2896 noisapnp [ISAPNP] Disables ISA PnP code.
2898 noinitrd [RAM] Tells the kernel not to load any configured
2901 nointremap [X86-64, Intel-IOMMU] Do not enable interrupt
2903 [Deprecated - use intremap=off]
2907 noinvpcid [X86] Disable the INVPCID cpu feature.
2909 nojitter [IA-64] Disables jitter checking for ITC timers.
2911 no-kvmclock [X86,KVM] Disable paravirtualized KVM clock driver
2913 no-kvmapf [X86,KVM] Disable paravirtualized asynchronous page
2917 [X86,PV_OPS] Disable paravirtualized VMware scheduler
2918 clock and use the default one.
2920 no-steal-acc [X86,KVM] Disable paravirtualized steal time accounting.
2921 steal time is computed, but won't influence scheduler
2924 nolapic [X86-32,APIC] Do not enable or use the local APIC.
2926 nolapic_timer [X86-32,APIC] Do not use the local APIC timer.
2928 noltlbs [PPC] Do not use large page/tlb entries for kernel
2929 lowmem mapping on PPC40x and PPC8xx
2931 nomca [IA-64] Disable machine check abort handling
2933 nomce [X86-32] Disable Machine Check Exception
2935 nomfgpt [X86-32] Disable Multi-Function General Purpose
2936 Timer usage (for AMD Geode machines).
2938 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to
2939 shutdown the other cpus. Instead use the REBOOT_VECTOR
2942 nomodule Disable module load
2944 nopat [X86] Disable PAT (page attribute table extension of
2945 pagetables) support.
2947 nopcid [X86-64] Disable the PCID cpu feature.
2949 norandmaps Don't use address space randomization. Equivalent to
2950 echo 0 > /proc/sys/kernel/randomize_va_space
2952 noreplace-smp [X86-32,SMP] Don't replace SMP instructions
2953 with UP alternatives
2955 nordrand [X86] Disable kernel use of the RDRAND and
2956 RDSEED instructions even if they are supported
2957 by the processor. RDRAND and RDSEED are still
2958 available to user space applications.
2960 noresume [SWSUSP] Disables resume and restores original swap
2963 no-scroll [VGA] Disables scrollback.
2964 This is required for the Braillex ib80-piezo Braille
2965 reader made by F.H. Papenmeier (Germany).
2969 nosep [BUGS=X86-32] Disables x86 SYSENTER/SYSEXIT support.
2971 nosmp [SMP] Tells an SMP kernel to act as a UP kernel,
2972 and disable the IO APIC. legacy for "maxcpus=0".
2974 nosoftlockup [KNL] Disable the soft-lockup detector.
2976 nosync [HW,M68K] Disables sync negotiation for all devices.
2978 nowatchdog [KNL] Disable both lockup detectors, i.e.
2979 soft-lockup and NMI watchdog (hard-lockup).
2983 nox2apic [X86-64,APIC] Do not enable x2APIC mode.
2985 cpu0_hotplug [X86] Turn on CPU0 hotplug feature when
2986 CONFIG_BOOTPARAM_HOTPLUG_CPU0 is off.
2987 Some features depend on CPU0. Known dependencies are:
2988 1. Resume from suspend/hibernate depends on CPU0.
2989 Suspend/hibernate will fail if CPU0 is offline and you
2990 need to online CPU0 before suspend/hibernate.
2991 2. PIC interrupts also depend on CPU0. CPU0 can't be
2992 removed if a PIC interrupt is detected.
2993 It's said poweroff/reboot may depend on CPU0 on some
2994 machines although I haven't seen such issues so far
2995 after CPU0 is offline on a few tested machines.
2996 If the dependencies are under your control, you can
2997 turn on cpu0_hotplug.
2999 nps_mtm_hs_ctr= [KNL,ARC]
3000 This parameter sets the maximum duration, in
3001 cycles, each HW thread of the CTOP can run
3002 without interruptions, before HW switches it.
3003 The actual maximum duration is 16 times this
3005 Format: integer between 1 and 255
3008 nptcg= [IA-64] Override max number of concurrent global TLB
3009 purges which is reported from either PAL_VM_SUMMARY or
3012 nr_cpus= [SMP] Maximum number of processors that an SMP kernel
3013 could support. nr_cpus=n : n >= 1 limits the kernel to
3014 support 'n' processors. It could be larger than the
3015 number of already plugged CPU during bootup, later in
3016 runtime you can physically add extra cpu until it reaches
3017 n. So during boot up some boot time memory for per-cpu
3018 variables need be pre-allocated for later physical cpu
3021 nr_uarts= [SERIAL] maximum number of UARTs to be registered.
3023 numa_balancing= [KNL,X86] Enable or disable automatic NUMA balancing.
3024 Allowed values are enable and disable
3026 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA.
3027 'node', 'default' can be specified
3028 This can be set from sysctl after boot.
3029 See Documentation/sysctl/vm.txt for details.
3031 ohci1394_dma=early [HW] enable debugging via the ohci1394 driver.
3032 See Documentation/debugging-via-ohci1394.txt for more
3035 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands
3036 Rather than timing out after 20 ms if an EC
3037 command is not properly ACKed, override the length
3038 of the timeout. We have interrupts disabled while
3039 waiting for the ACK, so if this is set too high
3040 interrupts *may* be lost!
3042 omap_mux= [OMAP] Override bootloader pin multiplexing.
3043 Format: <mux_mode0.mode_name=value>...
3044 For example, to override I2C bus2:
3045 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100
3047 oprofile.timer= [HW]
3048 Use timer interrupt instead of performance counters
3050 oprofile.cpu_type= Force an oprofile cpu type
3051 This might be useful if you have an older oprofile
3052 userland or if you want common events.
3053 Format: { arch_perfmon }
3054 arch_perfmon: [X86] Force use of architectural
3055 perfmon on Intel CPUs instead of the
3056 CPU specific event set.
3057 timer: [X86] Force use of architectural NMI
3058 timer mode (see also oprofile.timer
3059 for generic hr timer mode)
3061 oops=panic Always panic on oopses. Default is to just kill the
3062 process, but there is a small probability of
3063 deadlocking the machine.
3064 This will also cause panics on machine check exceptions.
3065 Useful together with panic=30 to trigger a reboot.
3067 page_owner= [KNL] Boot-time page_owner enabling option.
3068 Storage of the information about who allocated
3069 each page is disabled in default. With this switch,
3071 on: enable the feature
3073 page_poison= [KNL] Boot-time parameter changing the state of
3074 poisoning on the buddy allocator, available with
3075 CONFIG_PAGE_POISONING=y.
3076 off: turn off poisoning (default)
3077 on: turn on poisoning
3079 panic= [KNL] Kernel behaviour on panic: delay <timeout>
3080 timeout > 0: seconds before rebooting
3081 timeout = 0: wait forever
3082 timeout < 0: reboot immediately
3085 panic_on_warn panic() instead of WARN(). Useful to cause kdump
3088 crash_kexec_post_notifiers
3089 Run kdump after running panic-notifiers and dumping
3090 kmsg. This only for the users who doubt kdump always
3091 succeeds in any situation.
3092 Note that this also increases risks of kdump failure,
3093 because some panic notifiers can make the crashed
3094 kernel more unstable.
3096 parkbd.port= [HW] Parallel port number the keyboard adapter is
3097 connected to, default is 0.
3099 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation,
3100 0 for XT, 1 for AT (default is AT).
3103 parport= [HW,PPT] Specify parallel ports. 0 disables.
3104 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] }
3105 Use 'auto' to force the driver to use any
3106 IRQ/DMA settings detected (the default is to
3107 ignore detected IRQ/DMA settings because of
3108 possible conflicts). You can specify the base
3109 address, IRQ, and DMA settings; IRQ and DMA
3110 should be numbers, or 'auto' (for using detected
3111 settings on that particular port), or 'nofifo'
3112 (to avoid using a FIFO even if it is detected).
3113 Parallel ports are assigned in the order they
3114 are specified on the command line, starting
3117 parport_init_mode= [HW,PPT]
3118 Configure VIA parallel port to operate in
3119 a specific mode. This is necessary on Pegasos
3120 computer where firmware has no options for setting
3121 up parallel port mode and sets it to spp.
3122 Currently this function knows 686a and 8231 chips.
3123 Format: [spp|ps2|epp|ecp|ecpepp]
3126 Halt all CPUs after the first oops has been printed for
3127 the specified number of seconds. This is to be used if
3128 your oopses keep scrolling off the screen.
3133 See header of drivers/block/paride/pcd.c.
3134 See also Documentation/blockdev/paride.txt.
3136 pci=option[,option...] [PCI] various PCI subsystem options.
3138 Some options herein operate on a specific device
3139 or a set of devices (<pci_dev>). These are
3140 specified in one of the following formats:
3142 [<domain>:]<bus>:<dev>.<func>[/<dev>.<func>]*
3143 pci:<vendor>:<device>[:<subvendor>:<subdevice>]
3145 Note: the first format specifies a PCI
3146 bus/device/function address which may change
3147 if new hardware is inserted, if motherboard
3148 firmware changes, or due to changes caused
3149 by other kernel parameters. If the
3150 domain is left unspecified, it is
3151 taken to be zero. Optionally, a path
3152 to a device through multiple device/function
3153 addresses can be specified after the base
3154 address (this is more robust against
3155 renumbering issues). The second format
3156 selects devices using IDs from the
3157 configuration space which may match multiple
3158 devices in the system.
3160 earlydump dump PCI config space before the kernel
3162 off [X86] don't probe for the PCI bus
3163 bios [X86-32] force use of PCI BIOS, don't access
3164 the hardware directly. Use this if your machine
3165 has a non-standard PCI host bridge.
3166 nobios [X86-32] disallow use of PCI BIOS, only direct
3167 hardware access methods are allowed. Use this
3168 if you experience crashes upon bootup and you
3169 suspect they are caused by the BIOS.
3170 conf1 [X86] Force use of PCI Configuration Access
3171 Mechanism 1 (config address in IO port 0xCF8,
3172 data in IO port 0xCFC, both 32-bit).
3173 conf2 [X86] Force use of PCI Configuration Access
3174 Mechanism 2 (IO port 0xCF8 is an 8-bit port for
3175 the function, IO port 0xCFA, also 8-bit, sets
3176 bus number. The config space is then accessed
3177 through ports 0xC000-0xCFFF).
3178 See http://wiki.osdev.org/PCI for more info
3179 on the configuration access mechanisms.
3180 noaer [PCIE] If the PCIEAER kernel config parameter is
3181 enabled, this kernel boot option can be used to
3182 disable the use of PCIE advanced error reporting.
3183 nodomains [PCI] Disable support for multiple PCI
3184 root domains (aka PCI segments, in ACPI-speak).
3185 nommconf [X86] Disable use of MMCONFIG for PCI
3187 check_enable_amd_mmconf [X86] check for and enable
3188 properly configured MMIO access to PCI
3189 config space on AMD family 10h CPU
3190 nomsi [MSI] If the PCI_MSI kernel config parameter is
3191 enabled, this kernel boot option can be used to
3192 disable the use of MSI interrupts system-wide.
3193 noioapicquirk [APIC] Disable all boot interrupt quirks.
3194 Safety option to keep boot IRQs enabled. This
3195 should never be necessary.
3196 ioapicreroute [APIC] Enable rerouting of boot IRQs to the
3197 primary IO-APIC for bridges that cannot disable
3198 boot IRQs. This fixes a source of spurious IRQs
3199 when the system masks IRQs.
3200 noioapicreroute [APIC] Disable workaround that uses the
3201 boot IRQ equivalent of an IRQ that connects to
3202 a chipset where boot IRQs cannot be disabled.
3203 The opposite of ioapicreroute.
3204 biosirq [X86-32] Use PCI BIOS calls to get the interrupt
3205 routing table. These calls are known to be buggy
3206 on several machines and they hang the machine
3207 when used, but on other computers it's the only
3208 way to get the interrupt routing table. Try
3209 this option if the kernel is unable to allocate
3210 IRQs or discover secondary PCI buses on your
3212 rom [X86] Assign address space to expansion ROMs.
3213 Use with caution as certain devices share
3214 address decoders between ROMs and other
3216 norom [X86] Do not assign address space to
3217 expansion ROMs that do not already have
3218 BIOS assigned address ranges.
3219 nobar [X86] Do not assign address space to the
3220 BARs that weren't assigned by the BIOS.
3221 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be
3222 assigned automatically to PCI devices. You can
3223 make the kernel exclude IRQs of your ISA cards
3225 pirqaddr=0xAAAAA [X86] Specify the physical address
3226 of the PIRQ table (normally generated
3227 by the BIOS) if it is outside the
3228 F0000h-100000h range.
3229 lastbus=N [X86] Scan all buses thru bus #N. Can be
3230 useful if the kernel is unable to find your
3231 secondary buses and you want to tell it
3232 explicitly which ones they are.
3233 assign-busses [X86] Always assign all PCI bus
3234 numbers ourselves, overriding
3235 whatever the firmware may have done.
3236 usepirqmask [X86] Honor the possible IRQ mask stored
3237 in the BIOS $PIR table. This is needed on
3238 some systems with broken BIOSes, notably
3239 some HP Pavilion N5400 and Omnibook XE3
3240 notebooks. This will have no effect if ACPI
3241 IRQ routing is enabled.
3242 noacpi [X86] Do not use ACPI for IRQ routing
3243 or for PCI scanning.
3244 use_crs [X86] Use PCI host bridge window information
3245 from ACPI. On BIOSes from 2008 or later, this
3246 is enabled by default. If you need to use this,
3247 please report a bug.
3248 nocrs [X86] Ignore PCI host bridge windows from ACPI.
3249 If you need to use this, please report a bug.
3250 routeirq Do IRQ routing for all PCI devices.
3251 This is normally done in pci_enable_device(),
3252 so this option is a temporary workaround
3253 for broken drivers that don't call it.
3254 skip_isa_align [X86] do not align io start addr, so can
3255 handle more pci cards
3256 noearly [X86] Don't do any early type 1 scanning.
3257 This might help on some broken boards which
3258 machine check when some devices' config space
3259 is read. But various workarounds are disabled
3260 and some IOMMU drivers will not work.
3261 bfsort Sort PCI devices into breadth-first order.
3262 This sorting is done to get a device
3263 order compatible with older (<= 2.4) kernels.
3264 nobfsort Don't sort PCI devices into breadth-first order.
3265 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size)
3266 tuning and use the BIOS-configured MPS defaults.
3267 pcie_bus_safe Set every device's MPS to the largest value
3268 supported by all devices below the root complex.
3269 pcie_bus_perf Set device MPS to the largest allowable MPS
3270 based on its parent bus. Also set MRRS (Max
3271 Read Request Size) to the largest supported
3272 value (no larger than the MPS that the device
3273 or bus can support) for best performance.
3274 pcie_bus_peer2peer Set every device's MPS to 128B, which
3275 every device is guaranteed to support. This
3276 configuration allows peer-to-peer DMA between
3277 any pair of devices, possibly at the cost of
3278 reduced performance. This also guarantees
3279 that hot-added devices will work.
3280 cbiosize=nn[KMG] The fixed amount of bus space which is
3281 reserved for the CardBus bridge's IO window.
3282 The default value is 256 bytes.
3283 cbmemsize=nn[KMG] The fixed amount of bus space which is
3284 reserved for the CardBus bridge's memory
3285 window. The default value is 64 megabytes.
3288 [<order of align>@]<pci_dev>[; ...]
3289 Specifies alignment and device to reassign
3290 aligned memory resources. How to
3291 specify the device is described above.
3292 If <order of align> is not specified,
3293 PAGE_SIZE is used as alignment.
3294 PCI-PCI bridge can be specified, if resource
3295 windows need to be expanded.
3296 To specify the alignment for several
3297 instances of a device, the PCI vendor,
3298 device, subvendor, and subdevice may be
3299 specified, e.g., 4096@pci:8086:9c22:103c:198f
3300 ecrc= Enable/disable PCIe ECRC (transaction layer
3301 end-to-end CRC checking).
3302 bios: Use BIOS/firmware settings. This is the
3306 hpiosize=nn[KMG] The fixed amount of bus space which is
3307 reserved for hotplug bridge's IO window.
3308 Default size is 256 bytes.
3309 hpmemsize=nn[KMG] The fixed amount of bus space which is
3310 reserved for hotplug bridge's memory window.
3311 Default size is 2 megabytes.
3312 hpbussize=nn The minimum amount of additional bus numbers
3313 reserved for buses below a hotplug bridge.
3315 realloc= Enable/disable reallocating PCI bridge resources
3316 if allocations done by BIOS are too small to
3317 accommodate resources required by all child
3319 off: Turn realloc off
3321 realloc same as realloc=on
3322 noari do not use PCIe ARI.
3323 noats [PCIE, Intel-IOMMU, AMD-IOMMU]
3324 do not use PCIe ATS (and IOMMU device IOTLB).
3325 pcie_scan_all Scan all possible PCIe devices. Otherwise we
3326 only look for one device below a PCIe downstream
3328 big_root_window Try to add a big 64bit memory window to the PCIe
3329 root complex on AMD CPUs. Some GFX hardware
3330 can resize a BAR to allow access to all VRAM.
3331 Adding the window is slightly risky (it may
3332 conflict with unreported devices), so this
3334 disable_acs_redir=<pci_dev>[; ...]
3335 Specify one or more PCI devices (in the format
3336 specified above) separated by semicolons.
3337 Each device specified will have the PCI ACS
3338 redirect capabilities forced off which will
3339 allow P2P traffic between devices through
3340 bridges without forcing it upstream. Note:
3341 this removes isolation between devices and
3342 may put more devices in an IOMMU group.
3344 pcie_aspm= [PCIE] Forcibly enable or disable PCIe Active State Power
3347 force Enable ASPM even on devices that claim not to support it.
3348 WARNING: Forcing ASPM on may cause system lockups.
3350 pcie_ports= [PCIE] PCIe port services handling:
3351 native Use native PCIe services (PME, AER, DPC, PCIe hotplug)
3352 even if the platform doesn't give the OS permission to
3353 use them. This may cause conflicts if the platform
3354 also tries to use these services.
3355 compat Disable native PCIe services (PME, AER, DPC, PCIe
3358 pcie_port_pm= [PCIE] PCIe port power management handling:
3359 off Disable power management of all PCIe ports
3360 force Forcibly enable power management of all PCIe ports
3362 pcie_pme= [PCIE,PM] Native PCIe PME signaling options:
3363 nomsi Do not use MSI for native PCIe PME signaling (this makes
3364 all PCIe root ports use INTx for all services).
3366 pcmv= [HW,PCMCIA] BadgePAD 4
3370 Keep all power-domains already enabled by bootloader on,
3371 even if no driver has claimed them. This is useful
3372 for debug and development, but should not be
3373 needed on a platform with proper driver support.
3376 See Documentation/blockdev/paride.txt.
3378 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at
3381 See arch/parisc/kernel/pdc_chassis.c
3383 percpu_alloc= Select which percpu first chunk allocator to use.
3384 Currently supported values are "embed" and "page".
3385 Archs may support subset or none of the selections.
3386 See comments in mm/percpu.c for details on each
3387 allocator. This parameter is primarily for debugging
3388 and performance comparison.
3391 See Documentation/blockdev/paride.txt.
3394 See Documentation/blockdev/paride.txt.
3396 pirq= [SMP,APIC] Manual mp-table setup
3397 See Documentation/x86/i386/IO-APIC.txt.
3399 plip= [PPT,NET] Parallel port network link
3400 Format: { parport<nr> | timid | 0 }
3401 See also Documentation/admin-guide/parport.rst.
3403 pmtmr= [X86] Manual setup of pmtmr I/O Port.
3404 Override pmtimer IOPort with a hex value.
3408 Enable PNP debug messages (depends on the
3409 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time
3410 via /sys/module/pnp/parameters/debug. We always show
3411 current resource usage; turning this on also shows
3412 possible settings and some assignment information.
3418 { on | off | curr | res | no-curr | no-res }
3421 [ISAPNP] Exclude IRQs for the autoconfiguration
3424 [ISAPNP] Exclude DMAs for the autoconfiguration
3426 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration
3427 Ranges are in pairs (I/O port base and size).
3430 [ISAPNP] Exclude memory regions for the
3432 Ranges are in pairs (memory base and size).
3434 ports= [IP_VS_FTP] IPVS ftp helper module
3436 Up to 8 (IP_VS_APP_MAX_PORTS) ports
3438 Format: <port>,<port>....
3440 powersave=off [PPC] This option disables power saving features.
3441 It specifically disables cpuidle and sets the
3442 platform machine description specific power_save
3443 function to NULL. On Idle the CPU just reduces
3446 ppc_strict_facility_enable
3447 [PPC] This option catches any kernel floating point,
3448 Altivec, VSX and SPE outside of regions specifically
3449 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()).
3450 There is some performance impact when enabling this.
3454 Disable Hardware Transactional Memory
3456 print-fatal-signals=
3457 [KNL] debug: print fatal signals
3459 If enabled, warn about various signal handling
3460 related application anomalies: too many signals,
3461 too many POSIX.1 timers, fatal signals causing a
3464 If you hit the warning due to signal overflow,
3465 you might want to try "ulimit -i unlimited".
3469 printk.always_kmsg_dump=
3470 Trigger kmsg_dump for cases other than kernel oops or
3472 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
3475 printk.devkmsg={on,off,ratelimit}
3476 Control writing to /dev/kmsg.
3477 on - unlimited logging to /dev/kmsg from userspace
3478 off - logging to /dev/kmsg disabled
3479 ratelimit - ratelimit the logging
3482 printk.time= Show timing data prefixed to each printk message line
3483 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
3485 processor.max_cstate= [HW,ACPI]
3486 Limit processor to maximum C-state
3487 max_cstate=9 overrides any DMI blacklist limit.
3489 processor.nocst [HW,ACPI]
3490 Ignore the _CST method to determine C-states,
3491 instead using the legacy FADT method
3493 profile= [KNL] Enable kernel profiling via /proc/profile
3494 Format: [<profiletype>,]<number>
3495 Param: <profiletype>: "schedule", "sleep", or "kvm"
3496 [defaults to kernel profiling]
3497 Param: "schedule" - profile schedule points.
3498 Param: "sleep" - profile D-state sleeping (millisecs).
3499 Requires CONFIG_SCHEDSTATS
3500 Param: "kvm" - profile VM exits.
3501 Param: <number> - step/bucket size as a power of 2 for
3502 statistical time based profiling.
3504 prompt_ramdisk= [RAM] List of RAM disks to prompt for floppy disk
3506 See Documentation/blockdev/ramdisk.txt.
3508 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to
3509 probe for; one of (bare|imps|exps|lifebook|any).
3510 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports
3512 psmouse.resetafter= [HW,MOUSE]
3513 Try to reset the device after so many bad packets
3516 [HW,MOUSE] Set desired mouse resolution, in dpi.
3517 psmouse.smartscroll=
3518 [HW,MOUSE] Controls Logitech smartscroll autorepeat.
3519 0 = disabled, 1 = enabled (default).
3521 pstore.backend= Specify the name of the pstore backend to use
3524 See Documentation/blockdev/paride.txt.
3526 pti= [X86_64] Control Page Table Isolation of user and
3527 kernel address spaces. Disabling this feature
3528 removes hardening, but improves performance of
3529 system calls and interrupts.
3531 on - unconditionally enable
3532 off - unconditionally disable
3533 auto - kernel detects whether your CPU model is
3534 vulnerable to issues that PTI mitigates
3536 Not specifying this option is equivalent to pti=auto.
3539 Equivalent to pti=off
3542 [KNL] Number of legacy pty's. Overwrites compiled-in
3545 quiet [KNL] Disable most log messages
3550 See Documentation/admin-guide/md.rst.
3552 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes
3553 See Documentation/blockdev/ramdisk.txt.
3555 random.trust_cpu={on,off}
3556 [KNL] Enable or disable trusting the use of the
3557 CPU's random number generator (if available) to
3558 fully seed the kernel's CRNG. Default is controlled
3559 by CONFIG_RANDOM_TRUST_CPU.
3561 ras=option[,option,...] [KNL] RAS-specific options
3564 Disable the Correctable Errors Collector,
3565 see CONFIG_RAS_CEC help text.
3568 The argument is a cpu list, as described above.
3570 In kernels built with CONFIG_RCU_NOCB_CPU=y, set
3571 the specified list of CPUs to be no-callback CPUs.
3572 Invocation of these CPUs' RCU callbacks will be
3573 offloaded to "rcuox/N" kthreads created for that
3574 purpose, where "x" is "p" for RCU-preempt, and
3575 "s" for RCU-sched, and "N" is the CPU number.
3576 This reduces OS jitter on the offloaded CPUs,
3577 which can be useful for HPC and real-time
3578 workloads. It can also improve energy efficiency
3579 for asymmetric multiprocessors.
3582 Rather than requiring that offloaded CPUs
3583 (specified by rcu_nocbs= above) explicitly
3584 awaken the corresponding "rcuoN" kthreads,
3585 make these kthreads poll for callbacks.
3586 This improves the real-time response for the
3587 offloaded CPUs by relieving them of the need to
3588 wake up the corresponding kthread, but degrades
3589 energy efficiency by requiring that the kthreads
3590 periodically wake up to do the polling.
3592 rcutree.blimit= [KNL]
3593 Set maximum number of finished RCU callbacks to
3594 process in one batch.
3596 rcutree.dump_tree= [KNL]
3597 Dump the structure of the rcu_node combining tree
3598 out at early boot. This is used for diagnostic
3599 purposes, to verify correct tree setup.
3601 rcutree.gp_cleanup_delay= [KNL]
3602 Set the number of jiffies to delay each step of
3603 RCU grace-period cleanup.
3605 rcutree.gp_init_delay= [KNL]
3606 Set the number of jiffies to delay each step of
3607 RCU grace-period initialization.
3609 rcutree.gp_preinit_delay= [KNL]
3610 Set the number of jiffies to delay each step of
3611 RCU grace-period pre-initialization, that is,
3612 the propagation of recent CPU-hotplug changes up
3613 the rcu_node combining tree.
3615 rcutree.rcu_fanout_exact= [KNL]
3616 Disable autobalancing of the rcu_node combining
3617 tree. This is used by rcutorture, and might
3618 possibly be useful for architectures having high
3619 cache-to-cache transfer latencies.
3621 rcutree.rcu_fanout_leaf= [KNL]
3622 Change the number of CPUs assigned to each
3623 leaf rcu_node structure. Useful for very
3624 large systems, which will choose the value 64,
3625 and for NUMA systems with large remote-access
3626 latencies, which will choose a value aligned
3627 with the appropriate hardware boundaries.
3629 rcutree.jiffies_till_sched_qs= [KNL]
3630 Set required age in jiffies for a
3631 given grace period before RCU starts
3632 soliciting quiescent-state help from
3633 rcu_note_context_switch(). If not specified, the
3634 kernel will calculate a value based on the most
3635 recent settings of rcutree.jiffies_till_first_fqs
3636 and rcutree.jiffies_till_next_fqs.
3637 This calculated value may be viewed in
3638 rcutree.jiffies_to_sched_qs. Any attempt to
3639 set rcutree.jiffies_to_sched_qs will be
3640 cheerfully overwritten.
3642 rcutree.jiffies_till_first_fqs= [KNL]
3643 Set delay from grace-period initialization to
3644 first attempt to force quiescent states.
3645 Units are jiffies, minimum value is zero,
3646 and maximum value is HZ.
3648 rcutree.jiffies_till_next_fqs= [KNL]
3649 Set delay between subsequent attempts to force
3650 quiescent states. Units are jiffies, minimum
3651 value is one, and maximum value is HZ.
3653 rcutree.kthread_prio= [KNL,BOOT]
3654 Set the SCHED_FIFO priority of the RCU per-CPU
3655 kthreads (rcuc/N). This value is also used for
3656 the priority of the RCU boost threads (rcub/N)
3657 and for the RCU grace-period kthreads (rcu_bh,
3658 rcu_preempt, and rcu_sched). If RCU_BOOST is
3659 set, valid values are 1-99 and the default is 1
3660 (the least-favored priority). Otherwise, when
3661 RCU_BOOST is not set, valid values are 0-99 and
3662 the default is zero (non-realtime operation).
3664 rcutree.rcu_nocb_leader_stride= [KNL]
3665 Set the number of NOCB kthread groups, which
3666 defaults to the square root of the number of
3667 CPUs. Larger numbers reduces the wakeup overhead
3668 on the per-CPU grace-period kthreads, but increases
3669 that same overhead on each group's leader.
3671 rcutree.qhimark= [KNL]
3672 Set threshold of queued RCU callbacks beyond which
3673 batch limiting is disabled.
3675 rcutree.qlowmark= [KNL]
3676 Set threshold of queued RCU callbacks below which
3677 batch limiting is re-enabled.
3679 rcutree.rcu_idle_gp_delay= [KNL]
3680 Set wakeup interval for idle CPUs that have
3681 RCU callbacks (RCU_FAST_NO_HZ=y).
3683 rcutree.rcu_idle_lazy_gp_delay= [KNL]
3684 Set wakeup interval for idle CPUs that have
3685 only "lazy" RCU callbacks (RCU_FAST_NO_HZ=y).
3686 Lazy RCU callbacks are those which RCU can
3687 prove do nothing more than free memory.
3689 rcutree.rcu_kick_kthreads= [KNL]
3690 Cause the grace-period kthread to get an extra
3691 wake_up() if it sleeps three times longer than
3692 it should at force-quiescent-state time.
3693 This wake_up() will be accompanied by a
3694 WARN_ONCE() splat and an ftrace_dump().
3696 rcuperf.gp_async= [KNL]
3697 Measure performance of asynchronous
3698 grace-period primitives such as call_rcu().
3700 rcuperf.gp_async_max= [KNL]
3701 Specify the maximum number of outstanding
3702 callbacks per writer thread. When a writer
3703 thread exceeds this limit, it invokes the
3704 corresponding flavor of rcu_barrier() to allow
3705 previously posted callbacks to drain.
3707 rcuperf.gp_exp= [KNL]
3708 Measure performance of expedited synchronous
3709 grace-period primitives.
3711 rcuperf.holdoff= [KNL]
3712 Set test-start holdoff period. The purpose of
3713 this parameter is to delay the start of the
3714 test until boot completes in order to avoid
3717 rcuperf.nreaders= [KNL]
3718 Set number of RCU readers. The value -1 selects
3719 N, where N is the number of CPUs. A value
3720 "n" less than -1 selects N-n+1, where N is again
3721 the number of CPUs. For example, -2 selects N
3722 (the number of CPUs), -3 selects N+1, and so on.
3723 A value of "n" less than or equal to -N selects
3726 rcuperf.nwriters= [KNL]
3727 Set number of RCU writers. The values operate
3728 the same as for rcuperf.nreaders.
3729 N, where N is the number of CPUs
3731 rcuperf.perf_type= [KNL]
3732 Specify the RCU implementation to test.
3734 rcuperf.shutdown= [KNL]
3735 Shut the system down after performance tests
3736 complete. This is useful for hands-off automated
3739 rcuperf.verbose= [KNL]
3740 Enable additional printk() statements.
3742 rcuperf.writer_holdoff= [KNL]
3743 Write-side holdoff between grace periods,
3744 in microseconds. The default of zero says
3747 rcutorture.cbflood_inter_holdoff= [KNL]
3748 Set holdoff time (jiffies) between successive
3749 callback-flood tests.
3751 rcutorture.cbflood_intra_holdoff= [KNL]
3752 Set holdoff time (jiffies) between successive
3753 bursts of callbacks within a given callback-flood
3756 rcutorture.cbflood_n_burst= [KNL]
3757 Set the number of bursts making up a given
3758 callback-flood test. Set this to zero to
3759 disable callback-flood testing.
3761 rcutorture.cbflood_n_per_burst= [KNL]
3762 Set the number of callbacks to be registered
3763 in a given burst of a callback-flood test.
3765 rcutorture.fqs_duration= [KNL]
3766 Set duration of force_quiescent_state bursts
3769 rcutorture.fqs_holdoff= [KNL]
3770 Set holdoff time within force_quiescent_state bursts
3773 rcutorture.fqs_stutter= [KNL]
3774 Set wait time between force_quiescent_state bursts
3777 rcutorture.gp_cond= [KNL]
3778 Use conditional/asynchronous update-side
3779 primitives, if available.
3781 rcutorture.gp_exp= [KNL]
3782 Use expedited update-side primitives, if available.
3784 rcutorture.gp_normal= [KNL]
3785 Use normal (non-expedited) asynchronous
3786 update-side primitives, if available.
3788 rcutorture.gp_sync= [KNL]
3789 Use normal (non-expedited) synchronous
3790 update-side primitives, if available. If all
3791 of rcutorture.gp_cond=, rcutorture.gp_exp=,
3792 rcutorture.gp_normal=, and rcutorture.gp_sync=
3793 are zero, rcutorture acts as if is interpreted
3794 they are all non-zero.
3796 rcutorture.n_barrier_cbs= [KNL]
3797 Set callbacks/threads for rcu_barrier() testing.
3799 rcutorture.nfakewriters= [KNL]
3800 Set number of concurrent RCU writers. These just
3801 stress RCU, they don't participate in the actual
3802 test, hence the "fake".
3804 rcutorture.nreaders= [KNL]
3805 Set number of RCU readers. The value -1 selects
3806 N-1, where N is the number of CPUs. A value
3807 "n" less than -1 selects N-n-2, where N is again
3808 the number of CPUs. For example, -2 selects N
3809 (the number of CPUs), -3 selects N+1, and so on.
3811 rcutorture.object_debug= [KNL]
3812 Enable debug-object double-call_rcu() testing.
3814 rcutorture.onoff_holdoff= [KNL]
3815 Set time (s) after boot for CPU-hotplug testing.
3817 rcutorture.onoff_interval= [KNL]
3818 Set time (jiffies) between CPU-hotplug operations,
3819 or zero to disable CPU-hotplug testing.
3821 rcutorture.shuffle_interval= [KNL]
3822 Set task-shuffle interval (s). Shuffling tasks
3823 allows some CPUs to go into dyntick-idle mode
3824 during the rcutorture test.
3826 rcutorture.shutdown_secs= [KNL]
3827 Set time (s) after boot system shutdown. This
3828 is useful for hands-off automated testing.
3830 rcutorture.stall_cpu= [KNL]
3831 Duration of CPU stall (s) to test RCU CPU stall
3832 warnings, zero to disable.
3834 rcutorture.stall_cpu_holdoff= [KNL]
3835 Time to wait (s) after boot before inducing stall.
3837 rcutorture.stall_cpu_irqsoff= [KNL]
3838 Disable interrupts while stalling if set.
3840 rcutorture.stat_interval= [KNL]
3841 Time (s) between statistics printk()s.
3843 rcutorture.stutter= [KNL]
3844 Time (s) to stutter testing, for example, specifying
3845 five seconds causes the test to run for five seconds,
3846 wait for five seconds, and so on. This tests RCU's
3847 ability to transition abruptly to and from idle.
3849 rcutorture.test_boost= [KNL]
3850 Test RCU priority boosting? 0=no, 1=maybe, 2=yes.
3851 "Maybe" means test if the RCU implementation
3852 under test support RCU priority boosting.
3854 rcutorture.test_boost_duration= [KNL]
3855 Duration (s) of each individual boost test.
3857 rcutorture.test_boost_interval= [KNL]
3858 Interval (s) between each boost test.
3860 rcutorture.test_no_idle_hz= [KNL]
3861 Test RCU's dyntick-idle handling. See also the
3862 rcutorture.shuffle_interval parameter.
3864 rcutorture.torture_type= [KNL]
3865 Specify the RCU implementation to test.
3867 rcutorture.verbose= [KNL]
3868 Enable additional printk() statements.
3870 rcupdate.rcu_cpu_stall_suppress= [KNL]
3871 Suppress RCU CPU stall warning messages.
3873 rcupdate.rcu_cpu_stall_timeout= [KNL]
3874 Set timeout for RCU CPU stall warning messages.
3876 rcupdate.rcu_expedited= [KNL]
3877 Use expedited grace-period primitives, for
3878 example, synchronize_rcu_expedited() instead
3879 of synchronize_rcu(). This reduces latency,
3880 but can increase CPU utilization, degrade
3881 real-time latency, and degrade energy efficiency.
3882 No effect on CONFIG_TINY_RCU kernels.
3884 rcupdate.rcu_normal= [KNL]
3885 Use only normal grace-period primitives,
3886 for example, synchronize_rcu() instead of
3887 synchronize_rcu_expedited(). This improves
3888 real-time latency, CPU utilization, and
3889 energy efficiency, but can expose users to
3890 increased grace-period latency. This parameter
3891 overrides rcupdate.rcu_expedited. No effect on
3892 CONFIG_TINY_RCU kernels.
3894 rcupdate.rcu_normal_after_boot= [KNL]
3895 Once boot has completed (that is, after
3896 rcu_end_inkernel_boot() has been invoked), use
3897 only normal grace-period primitives. No effect
3898 on CONFIG_TINY_RCU kernels.
3900 rcupdate.rcu_task_stall_timeout= [KNL]
3901 Set timeout in jiffies for RCU task stall warning
3902 messages. Disable with a value less than or equal
3905 rcupdate.rcu_self_test= [KNL]
3906 Run the RCU early boot self tests
3910 Run specified binary instead of /init from the ramdisk,
3911 used for early userspace startup. See initrd.
3914 Turn on/off individual RDT features. List is:
3915 cmt, mbmtotal, mbmlocal, l3cat, l3cdp, l2cat, l2cdp,
3917 E.g. to turn on cmt and turn off mba use:
3921 Format (x86 or x86_64):
3922 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] \
3924 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \
3926 Where reboot_mode is one of warm (soft) or cold (hard) or gpio,
3927 reboot_type is one of bios, acpi, kbd, triple, efi, or pci,
3928 reboot_force is either force or not specified,
3929 reboot_cpu is s[mp]#### with #### being the processor
3930 to be used for rebooting.
3933 [KNL, SMP] Set scheduler's default relax_domain_level.
3934 See Documentation/cgroup-v1/cpusets.txt.
3936 reserve= [KNL,BUGS] Force kernel to ignore I/O ports or memory
3937 Format: <base1>,<size1>[,<base2>,<size2>,...]
3938 Reserve I/O ports or memory so the kernel won't use
3939 them. If <base> is less than 0x10000, the region
3940 is assumed to be I/O ports; otherwise it is memory.
3942 reservetop= [X86-32]
3944 Reserves a hole at the top of the kernel virtual
3949 Set the amount of memory to reserve for BIOS at
3950 the bottom of the address space.
3952 reset_devices [KNL] Force drivers to reset the underlying device
3953 during initialization.
3956 Specify the partition device for software suspend
3958 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>}
3960 resume_offset= [SWSUSP]
3961 Specify the offset from the beginning of the partition
3962 given by "resume=" at which the swap header is located,
3963 in <PAGE_SIZE> units (needed only for swap files).
3964 See Documentation/power/swsusp-and-swap-files.txt
3966 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to
3967 read the resume files
3969 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up.
3970 Useful for devices that are detected asynchronously
3971 (e.g. USB and MMC devices).
3973 hibernate= [HIBERNATION]
3974 noresume Don't check if there's a hibernation image
3975 present during boot.
3976 nocompress Don't compress/decompress hibernation images.
3977 no Disable hibernation and resume.
3978 protect_image Turn on image protection during restoration
3979 (that will set all pages holding image data
3980 during restoration read-only).
3982 retain_initrd [RAM] Keep initrd memory after extraction
3984 rfkill.default_state=
3985 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm,
3986 etc. communication is blocked by default.
3989 rfkill.master_switch_mode=
3990 0 The "airplane mode" button does nothing.
3991 1 The "airplane mode" button toggles between everything
3992 blocked and the previous configuration.
3993 2 The "airplane mode" button toggles between everything
3994 blocked and everything unblocked.
3996 rhash_entries= [KNL,NET]
3997 Set number of hash buckets for route cache
4000 [KNL] Disable ring 3 MONITOR/MWAIT feature on supported
4003 ro [KNL] Mount root device read-only on boot
4006 on Mark read-only kernel memory as read-only (default).
4007 off Leave read-only kernel memory writable for debugging.
4010 Enable the uart passthrough on the designated usb port
4011 on Rockchip SoCs. When active, the signals of the
4012 debug-uart get routed to the D+ and D- pins of the usb
4013 port and the regular usb controller gets disabled.
4015 root= [KNL] Root filesystem
4016 See name_to_dev_t comment in init/do_mounts.c.
4018 rootdelay= [KNL] Delay (in seconds) to pause before attempting to
4019 mount the root filesystem
4021 rootflags= [KNL] Set root filesystem mount option string
4023 rootfstype= [KNL] Set root filesystem type
4025 rootwait [KNL] Wait (indefinitely) for root device to show up.
4026 Useful for devices that are detected asynchronously
4027 (e.g. USB and MMC devices).
4029 rproc_mem=nn[KMG][@address]
4030 [KNL,ARM,CMA] Remoteproc physical memory block.
4031 Memory area to be used by remote processor image,
4034 rw [KNL] Mount root device read-write on boot
4036 S [KNL] Run init in single mode
4038 s390_iommu= [HW,S390]
4039 Set s390 IOTLB flushing mode
4041 With strict flushing every unmap operation will result in
4042 an IOTLB flush. Default is lazy flushing before reuse,
4046 See drivers/net/irda/sa1100_ir.c.
4048 sbni= [NET] Granch SBNI12 leased line adapter
4050 sched_debug [KNL] Enables verbose scheduler debug messages.
4052 schedstats= [KNL,X86] Enable or disable scheduled statistics.
4053 Allowed values are enable and disable. This feature
4054 incurs a small amount of overhead in the scheduler
4055 but is useful for debugging and performance tuning.
4057 skew_tick= [KNL] Offset the periodic timer tick per cpu to mitigate
4058 xtime_lock contention on larger systems, and/or RCU lock
4059 contention on all systems with CONFIG_MAXSMP set.
4060 Format: { "0" | "1" }
4061 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1"
4063 Note: increases power consumption, thus should only be
4064 enabled if running jitter sensitive (HPC/RT) workloads.
4066 security= [SECURITY] Choose a security module to enable at boot.
4067 If this boot parameter is not specified, only the first
4068 security module asking for security registration will be
4069 loaded. An invalid security module name will be treated
4070 as if no module has been chosen.
4072 selinux= [SELINUX] Disable or enable SELinux at boot time.
4073 Format: { "0" | "1" }
4074 See security/selinux/Kconfig help text.
4077 Default value is set via kernel config option.
4078 If enabled at boot time, /selinux/disable can be used
4079 later to disable prior to initial policy load.
4081 apparmor= [APPARMOR] Disable or enable AppArmor at boot time
4082 Format: { "0" | "1" }
4083 See security/apparmor/Kconfig help text
4086 Default value is set via kernel config option.
4088 serialnumber [BUGS=X86-32]
4091 Maximal number of shapers.
4099 Disable merging of slabs with similar size. May be
4100 necessary if there is some reason to distinguish
4101 allocs to different slabs, especially in hardened
4102 environments where the risk of heap overflows and
4103 layout control by attackers can usually be
4104 frustrated by disabling merging. This will reduce
4105 most of the exposure of a heap attack to a single
4106 cache (risks via metadata attacks are mostly
4107 unchanged). Debug options disable merging on their
4109 For more information see Documentation/vm/slub.rst.
4111 slab_max_order= [MM, SLAB]
4112 Determines the maximum allowed order for slabs.
4113 A high setting may cause OOMs due to memory
4114 fragmentation. Defaults to 1 for systems with
4115 more than 32MB of RAM, 0 otherwise.
4117 slub_debug[=options[,slabs]] [MM, SLUB]
4118 Enabling slub_debug allows one to determine the
4119 culprit if slab objects become corrupted. Enabling
4120 slub_debug can create guard zones around objects and
4121 may poison objects when not in use. Also tracks the
4122 last alloc / free. For more information see
4123 Documentation/vm/slub.rst.
4125 slub_memcg_sysfs= [MM, SLUB]
4126 Determines whether to enable sysfs directories for
4127 memory cgroup sub-caches. 1 to enable, 0 to disable.
4128 The default is determined by CONFIG_SLUB_MEMCG_SYSFS_ON.
4129 Enabling this can lead to a very high number of debug
4130 directories and files being created under
4133 slub_max_order= [MM, SLUB]
4134 Determines the maximum allowed order for slabs.
4135 A high setting may cause OOMs due to memory
4136 fragmentation. For more information see
4137 Documentation/vm/slub.rst.
4139 slub_min_objects= [MM, SLUB]
4140 The minimum number of objects per slab. SLUB will
4141 increase the slab order up to slub_max_order to
4142 generate a sufficiently large slab able to contain
4143 the number of objects indicated. The higher the number
4144 of objects the smaller the overhead of tracking slabs
4145 and the less frequently locks need to be acquired.
4146 For more information see Documentation/vm/slub.rst.
4148 slub_min_order= [MM, SLUB]
4149 Determines the minimum page order for slabs. Must be
4150 lower than slub_max_order.
4151 For more information see Documentation/vm/slub.rst.
4153 slub_nomerge [MM, SLUB]
4154 Same with slab_nomerge. This is supported for legacy.
4155 See slab_nomerge for more information.
4158 Format: <io1>[,<io2>[,...,<io8>]]
4160 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices
4161 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port
4162 smsc-ircc2.ircc_sir= [HW] SIR base I/O port
4163 smsc-ircc2.ircc_fir= [HW] FIR base I/O port
4164 smsc-ircc2.ircc_irq= [HW] IRQ line
4165 smsc-ircc2.ircc_dma= [HW] DMA channel
4166 smsc-ircc2.ircc_transceiver= [HW] Transceiver type:
4167 0: Toshiba Satellite 1800 (GP data pin select)
4168 1: Fast pin select (default)
4171 smt [KNL,S390] Set the maximum number of threads (logical
4172 CPUs) to use per physical CPU on systems capable of
4173 symmetric multithreading (SMT). Will be capped to the
4174 actual hardware limit.
4176 Default: -1 (no limit)
4179 [KNL] Should the soft-lockup detector generate panics.
4182 A nonzero value instructs the soft-lockup detector
4183 to panic the machine when a soft-lockup occurs. This
4184 is also controlled by CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC
4185 which is the respective build-time switch to that
4188 softlockup_all_cpu_backtrace=
4189 [KNL] Should the soft-lockup detector generate
4190 backtraces on all cpus.
4193 sonypi.*= [HW] Sony Programmable I/O Control Device driver
4194 See Documentation/laptops/sonypi.txt
4196 spectre_v2= [X86] Control mitigation of Spectre variant 2
4197 (indirect branch speculation) vulnerability.
4199 on - unconditionally enable
4200 off - unconditionally disable
4201 auto - kernel detects whether your CPU model is
4204 Selecting 'on' will, and 'auto' may, choose a
4205 mitigation method at run time according to the
4206 CPU, the available microcode, the setting of the
4207 CONFIG_RETPOLINE configuration option, and the
4208 compiler with which the kernel was built.
4210 Specific mitigations can also be selected manually:
4212 retpoline - replace indirect branches
4213 retpoline,generic - google's original retpoline
4214 retpoline,amd - AMD-specific minimal thunk
4216 Not specifying this option is equivalent to
4219 spec_store_bypass_disable=
4220 [HW] Control Speculative Store Bypass (SSB) Disable mitigation
4221 (Speculative Store Bypass vulnerability)
4223 Certain CPUs are vulnerable to an exploit against a
4224 a common industry wide performance optimization known
4225 as "Speculative Store Bypass" in which recent stores
4226 to the same memory location may not be observed by
4227 later loads during speculative execution. The idea
4228 is that such stores are unlikely and that they can
4229 be detected prior to instruction retirement at the
4230 end of a particular speculation execution window.
4232 In vulnerable processors, the speculatively forwarded
4233 store can be used in a cache side channel attack, for
4234 example to read memory to which the attacker does not
4235 directly have access (e.g. inside sandboxed code).
4237 This parameter controls whether the Speculative Store
4238 Bypass optimization is used.
4240 On x86 the options are:
4242 on - Unconditionally disable Speculative Store Bypass
4243 off - Unconditionally enable Speculative Store Bypass
4244 auto - Kernel detects whether the CPU model contains an
4245 implementation of Speculative Store Bypass and
4246 picks the most appropriate mitigation. If the
4247 CPU is not vulnerable, "off" is selected. If the
4248 CPU is vulnerable the default mitigation is
4249 architecture and Kconfig dependent. See below.
4250 prctl - Control Speculative Store Bypass per thread
4251 via prctl. Speculative Store Bypass is enabled
4252 for a process by default. The state of the control
4253 is inherited on fork.
4254 seccomp - Same as "prctl" above, but all seccomp threads
4255 will disable SSB unless they explicitly opt out.
4257 Default mitigations:
4258 X86: If CONFIG_SECCOMP=y "seccomp", otherwise "prctl"
4260 On powerpc the options are:
4262 on,auto - On Power8 and Power9 insert a store-forwarding
4263 barrier on kernel entry and exit. On Power7
4264 perform a software flush on kernel entry and
4268 Not specifying this option is equivalent to
4269 spec_store_bypass_disable=auto.
4271 spia_io_base= [HW,MTD]
4276 srcutree.counter_wrap_check [KNL]
4277 Specifies how frequently to check for
4278 grace-period sequence counter wrap for the
4279 srcu_data structure's ->srcu_gp_seq_needed field.
4280 The greater the number of bits set in this kernel
4281 parameter, the less frequently counter wrap will
4282 be checked for. Note that the bottom two bits
4285 srcutree.exp_holdoff [KNL]
4286 Specifies how many nanoseconds must elapse
4287 since the end of the last SRCU grace period for
4288 a given srcu_struct until the next normal SRCU
4289 grace period will be considered for automatic
4290 expediting. Set to zero to disable automatic
4294 Speculative Store Bypass Disable control
4296 On CPUs that are vulnerable to the Speculative
4297 Store Bypass vulnerability and offer a
4298 firmware based mitigation, this parameter
4299 indicates how the mitigation should be used:
4301 force-on: Unconditionally enable mitigation for
4302 for both kernel and userspace
4303 force-off: Unconditionally disable mitigation for
4304 for both kernel and userspace
4305 kernel: Always enable mitigation in the
4306 kernel, and offer a prctl interface
4307 to allow userspace to register its
4308 interest in being mitigated too.
4310 stack_guard_gap= [MM]
4311 override the default stack gap protection. The value
4312 is in page units and it defines how many pages prior
4313 to (for stacks growing down) resp. after (for stacks
4314 growing up) the main stack are reserved for no other
4315 mapping. Default value is 256 pages.
4318 Enabled the stack tracer on boot up.
4320 stacktrace_filter=[function-list]
4321 [FTRACE] Limit the functions that the stack tracer
4322 will trace at boot up. function-list is a comma separated
4323 list of functions. This list can be changed at run
4324 time by the stack_trace_filter file in the debugfs
4325 tracing directory. Note, this enables stack tracing
4326 and the stacktrace above is not needed.
4330 Set the STI (builtin display/keyboard on the HP-PARISC
4331 machines) console (graphic card) which should be used
4332 as the initial boot-console.
4333 See also comment in drivers/video/console/sticore.c.
4336 See comment in drivers/video/console/sticore.c.
4339 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]]
4341 sunrpc.min_resvport=
4342 sunrpc.max_resvport=
4344 SunRPC servers often require that client requests
4345 originate from a privileged port (i.e. a port in the
4346 range 0 < portnr < 1024).
4347 An administrator who wishes to reserve some of these
4348 ports for other uses may adjust the range that the
4349 kernel's sunrpc client considers to be privileged
4350 using these two parameters to set the minimum and
4351 maximum port values.
4353 sunrpc.svc_rpc_per_connection_limit=
4355 Limit the number of requests that the server will
4356 process in parallel from a single connection.
4357 The default value is 0 (no limit).
4361 Control how the NFS server code allocates CPUs to
4362 service thread pools. Depending on how many NICs
4363 you have and where their interrupts are bound, this
4364 option will affect which CPUs will do NFS serving.
4365 Note: this parameter cannot be changed while the
4366 NFS server is running.
4368 auto the server chooses an appropriate mode
4369 automatically using heuristics
4370 global a single global pool contains all CPUs
4371 percpu one pool for each CPU
4372 pernode one pool for each NUMA node (equivalent
4373 to global on non-NUMA machines)
4375 sunrpc.tcp_slot_table_entries=
4376 sunrpc.udp_slot_table_entries=
4378 Sets the upper limit on the number of simultaneous
4379 RPC calls that can be sent from the client to a
4380 server. Increasing these values may allow you to
4381 improve throughput, but will also increase the
4382 amount of memory reserved for use by the client.
4384 suspend.pm_test_delay=
4386 Sets the number of seconds to remain in a suspend test
4387 mode before resuming the system (see
4388 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG
4389 is set. Default value is 5.
4392 [KNL] Enable accounting of swap in memory resource
4393 controller if no parameter or 1 is given or disable
4394 it if 0 is given (See Documentation/cgroup-v1/memory.txt)
4396 swiotlb= [ARM,IA-64,PPC,MIPS,X86]
4397 Format: { <int> | force | noforce }
4398 <int> -- Number of I/O TLB slabs
4399 force -- force using of bounce buffers even if they
4400 wouldn't be automatically used by the kernel
4401 noforce -- Never use bounce buffers (for debugging)
4405 sysfs.deprecated=0|1 [KNL]
4406 Enable/disable old style sysfs layout for old udev
4407 on older distributions. When this option is enabled
4408 very new udev will not work anymore. When this option
4409 is disabled (or CONFIG_SYSFS_DEPRECATED not compiled)
4410 in older udev will not work anymore.
4411 Default depends on CONFIG_SYSFS_DEPRECATED_V2 set in
4412 the kernel configuration.
4414 sysrq_always_enabled
4416 Ignore sysrq setting - this boot parameter will
4417 neutralize any effect of /proc/sys/kernel/sysrq.
4418 Useful for debugging.
4420 tcpmhash_entries= [KNL,NET]
4421 Set the number of tcp_metrics_hash slots.
4422 Default value is 8192 or 16384 depending on total
4423 ram pages. This is used to specify the TCP metrics
4424 cache size. See Documentation/networking/ip-sysctl.txt
4425 "tcp_no_metrics_save" section for more details.
4429 test_suspend= [SUSPEND][,N]
4430 Specify "mem" (for Suspend-to-RAM) or "standby" (for
4431 standby suspend) or "freeze" (for suspend type freeze)
4432 as the system sleep state during system startup with
4433 the optional capability to repeat N number of times.
4434 The system is woken from this state using a
4435 wakeup-capable RTC alarm.
4437 thash_entries= [KNL,NET]
4438 Set number of hash buckets for TCP connection
4440 thermal.act= [HW,ACPI]
4441 -1: disable all active trip points in all thermal zones
4442 <degrees C>: override all lowest active trip points
4444 thermal.crt= [HW,ACPI]
4445 -1: disable all critical trip points in all thermal zones
4446 <degrees C>: override all critical trip points
4448 thermal.nocrt= [HW,ACPI]
4449 Set to disable actions on ACPI thermal zone
4450 critical and hot trip points.
4452 thermal.off= [HW,ACPI]
4453 1: disable ACPI thermal control
4455 thermal.psv= [HW,ACPI]
4456 -1: disable all passive trip points
4457 <degrees C>: override all passive trip points to this
4460 thermal.tzp= [HW,ACPI]
4461 Specify global default ACPI thermal zone polling rate
4462 <deci-seconds>: poll all this frequency
4463 0: no polling (default)
4466 Force threading of all interrupt handlers except those
4467 marked explicitly IRQF_NO_THREAD.
4470 Enable the Transcendent memory driver if built-in.
4472 tmem.cleancache=0|1 [KNL, XEN]
4473 Default is on (1). Disable the usage of the cleancache
4474 API to send anonymous pages to the hypervisor.
4476 tmem.frontswap=0|1 [KNL, XEN]
4477 Default is on (1). Disable the usage of the frontswap
4478 API to send swap pages to the hypervisor. If disabled
4479 the selfballooning and selfshrinking are force disabled.
4481 tmem.selfballooning=0|1 [KNL, XEN]
4482 Default is on (1). Disable the driving of swap pages
4485 tmem.selfshrinking=0|1 [KNL, XEN]
4486 Default is on (1). Partial swapoff that immediately
4487 transfers pages from Xen hypervisor back to the
4488 kernel based on different criteria.
4492 Specify if the kernel should make use of the cpu
4493 topology information if the hardware supports this.
4494 The scheduler will make use of this information and
4495 e.g. base its process migration decisions on it.
4498 topology_updates= [KNL, PPC, NUMA]
4500 Specify if the kernel should ignore (off)
4501 topology updates sent by the hypervisor to this
4506 tpm_suspend_pcr=[HW,TPM]
4507 Format: integer pcr id
4508 Specify that at suspend time, the tpm driver
4509 should extend the specified pcr with zeros,
4510 as a workaround for some chips which fail to
4511 flush the last written pcr on TPM_SaveState.
4512 This will guarantee that all the other pcrs
4515 trace_buf_size=nn[KMG]
4516 [FTRACE] will set tracing buffer size on each cpu.
4518 trace_event=[event-list]
4519 [FTRACE] Set and start specified trace events in order
4520 to facilitate early boot debugging. The event-list is a
4521 comma separated list of trace events to enable. See
4522 also Documentation/trace/events.rst
4524 trace_options=[option-list]
4525 [FTRACE] Enable or disable tracer options at boot.
4526 The option-list is a comma delimited list of options
4527 that can be enabled or disabled just as if you were
4528 to echo the option name into
4530 /sys/kernel/debug/tracing/trace_options
4532 For example, to enable stacktrace option (to dump the
4533 stack trace of each event), add to the command line:
4535 trace_options=stacktrace
4537 See also Documentation/trace/ftrace.rst "trace options"
4541 Have the tracepoints sent to printk as well as the
4542 tracing ring buffer. This is useful for early boot up
4543 where the system hangs or reboots and does not give the
4544 option for reading the tracing buffer or performing a
4545 ftrace_dump_on_oops.
4547 To turn off having tracepoints sent to printk,
4548 echo 0 > /proc/sys/kernel/tracepoint_printk
4549 Note, echoing 1 into this file without the
4550 tracepoint_printk kernel cmdline option has no effect.
4554 Having tracepoints sent to printk() and activating high
4555 frequency tracepoints such as irq or sched, can cause
4556 the system to live lock.
4559 [FTRACE] enable this option to disable tracing when a
4560 warning is hit. This turns off "tracing_on". Tracing can
4561 be enabled again by echoing '1' into the "tracing_on"
4562 file located in /sys/kernel/debug/tracing/
4564 This option is useful, as it disables the trace before
4565 the WARNING dump is called, which prevents the trace to
4566 be filled with content caused by the warning output.
4568 This option can also be set at run time via the sysctl
4569 option: kernel/traceoff_on_warning
4571 transparent_hugepage=
4573 Format: [always|madvise|never]
4574 Can be used to control the default behavior of the system
4575 with respect to transparent hugepages.
4576 See Documentation/admin-guide/mm/transhuge.rst
4579 tsc= Disable clocksource stability checks for TSC.
4581 [x86] reliable: mark tsc clocksource as reliable, this
4582 disables clocksource verification at runtime, as well
4583 as the stability checks done at bootup. Used to enable
4584 high-resolution timer mode on older hardware, and in
4585 virtualized environment.
4586 [x86] noirqtime: Do not use TSC to do irq accounting.
4587 Used to run time disable IRQ_TIME_ACCOUNTING on any
4588 platforms where RDTSC is slow and this accounting
4590 [x86] unstable: mark the TSC clocksource as unstable, this
4591 marks the TSC unconditionally unstable at bootup and
4592 avoids any further wobbles once the TSC watchdog notices.
4594 turbografx.map[2|3]= [HW,JOY]
4595 TurboGraFX parallel port interface
4597 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7>
4598 See also Documentation/input/devices/joystick-parport.rst
4600 udbg-immortal [PPC] When debugging early kernel crashes that
4601 happen after console_init() and before a proper
4602 console driver takes over, this boot options might
4603 help "seeing" what's going on.
4605 uhash_entries= [KNL,NET]
4606 Set number of hash buckets for UDP/UDP-Lite connections
4609 [USB] Ignore overcurrent events (default N).
4610 Some badly-designed motherboards generate lots of
4611 bogus events, for ports that aren't wired to
4612 anything. Set this parameter to avoid log spamming.
4613 Note that genuine overcurrent events won't be
4617 [X86] Cause panic on unknown NMI.
4619 usbcore.authorized_default=
4620 [USB] Default USB device authorization:
4621 (default -1 = authorized except for wireless USB,
4622 0 = not authorized, 1 = authorized)
4624 usbcore.autosuspend=
4625 [USB] The autosuspend time delay (in seconds) used
4626 for newly-detected USB devices (default 2). This
4627 is the time required before an idle device will be
4628 autosuspended. Devices for which the delay is set
4629 to a negative value won't be autosuspended at all.
4631 usbcore.usbfs_snoop=
4632 [USB] Set to log all usbfs traffic (default 0 = off).
4634 usbcore.usbfs_snoop_max=
4635 [USB] Maximum number of bytes to snoop in each URB
4638 usbcore.blinkenlights=
4639 [USB] Set to cycle leds on hubs (default 0 = off).
4641 usbcore.old_scheme_first=
4642 [USB] Start with the old device initialization
4643 scheme, applies only to low and full-speed devices
4646 usbcore.usbfs_memory_mb=
4647 [USB] Memory limit (in MB) for buffers allocated by
4648 usbfs (default = 16, 0 = max = 2047).
4650 usbcore.use_both_schemes=
4651 [USB] Try the other device initialization scheme
4652 if the first one fails (default 1 = enabled).
4654 usbcore.initial_descriptor_timeout=
4655 [USB] Specifies timeout for the initial 64-byte
4656 USB_REQ_GET_DESCRIPTOR request in milliseconds
4657 (default 5000 = 5.0 seconds).
4659 usbcore.nousb [USB] Disable the USB subsystem
4662 [USB] A list of quirk entries to augment the built-in
4663 usb core quirk list. List entries are separated by
4664 commas. Each entry has the form
4665 VendorID:ProductID:Flags. The IDs are 4-digit hex
4666 numbers and Flags is a set of letters. Each letter
4667 will change the built-in quirk; setting it if it is
4668 clear and clearing it if it is set. The letters have
4669 the following meanings:
4670 a = USB_QUIRK_STRING_FETCH_255 (string
4671 descriptors must not be fetched using
4673 b = USB_QUIRK_RESET_RESUME (device can't resume
4674 correctly so reset it instead);
4675 c = USB_QUIRK_NO_SET_INTF (device can't handle
4676 Set-Interface requests);
4677 d = USB_QUIRK_CONFIG_INTF_STRINGS (device can't
4678 handle its Configuration or Interface
4680 e = USB_QUIRK_RESET (device can't be reset
4681 (e.g morph devices), don't use reset);
4682 f = USB_QUIRK_HONOR_BNUMINTERFACES (device has
4683 more interface descriptions than the
4684 bNumInterfaces count, and can't handle
4685 talking to these interfaces);
4686 g = USB_QUIRK_DELAY_INIT (device needs a pause
4687 during initialization, after we read
4688 the device descriptor);
4689 h = USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL (For
4690 high speed and super speed interrupt
4691 endpoints, the USB 2.0 and USB 3.0 spec
4692 require the interval in microframes (1
4693 microframe = 125 microseconds) to be
4694 calculated as interval = 2 ^
4696 Devices with this quirk report their
4697 bInterval as the result of this
4698 calculation instead of the exponent
4699 variable used in the calculation);
4700 i = USB_QUIRK_DEVICE_QUALIFIER (device can't
4701 handle device_qualifier descriptor
4703 j = USB_QUIRK_IGNORE_REMOTE_WAKEUP (device
4704 generates spurious wakeup, ignore
4705 remote wakeup capability);
4706 k = USB_QUIRK_NO_LPM (device can't handle Link
4708 l = USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL
4709 (Device reports its bInterval as linear
4710 frames instead of the USB 2.0
4712 m = USB_QUIRK_DISCONNECT_SUSPEND (Device needs
4713 to be disconnected before suspend to
4714 prevent spurious wakeup);
4715 n = USB_QUIRK_DELAY_CTRL_MSG (Device needs a
4716 pause after every control message);
4717 o = USB_QUIRK_HUB_SLOW_RESET (Hub needs extra
4718 delay after resetting its port);
4719 Example: quirks=0781:5580:bk,0a5c:5834:gij
4722 [USBHID] The interval which mice are to be polled at.
4725 [USBHID] The interval which joysticks are to be polled at.
4728 [USBHID] The interval which keyboards are to be polled at.
4730 usb-storage.delay_use=
4731 [UMS] The delay in seconds before a new device is
4732 scanned for Logical Units (default 1).
4735 [UMS] A list of quirks entries to supplement or
4736 override the built-in unusual_devs list. List
4737 entries are separated by commas. Each entry has
4738 the form VID:PID:Flags where VID and PID are Vendor
4739 and Product ID values (4-digit hex numbers) and
4740 Flags is a set of characters, each corresponding
4741 to a common usb-storage quirk flag as follows:
4742 a = SANE_SENSE (collect more than 18 bytes
4744 b = BAD_SENSE (don't collect more than 18
4745 bytes of sense data);
4746 c = FIX_CAPACITY (decrease the reported
4747 device capacity by one sector);
4748 d = NO_READ_DISC_INFO (don't use
4749 READ_DISC_INFO command);
4750 e = NO_READ_CAPACITY_16 (don't use
4751 READ_CAPACITY_16 command);
4752 f = NO_REPORT_OPCODES (don't use report opcodes
4754 g = MAX_SECTORS_240 (don't transfer more than
4755 240 sectors at a time, uas only);
4756 h = CAPACITY_HEURISTICS (decrease the
4757 reported device capacity by one
4758 sector if the number is odd);
4759 i = IGNORE_DEVICE (don't bind to this
4761 j = NO_REPORT_LUNS (don't use report luns
4763 l = NOT_LOCKABLE (don't try to lock and
4764 unlock ejectable media);
4765 m = MAX_SECTORS_64 (don't transfer more
4766 than 64 sectors = 32 KB at a time);
4767 n = INITIAL_READ10 (force a retry of the
4768 initial READ(10) command);
4769 o = CAPACITY_OK (accept the capacity
4770 reported by the device);
4771 p = WRITE_CACHE (the device cache is ON
4773 r = IGNORE_RESIDUE (the device reports
4774 bogus residue values);
4775 s = SINGLE_LUN (the device has only one
4777 t = NO_ATA_1X (don't allow ATA(12) and ATA(16)
4778 commands, uas only);
4779 u = IGNORE_UAS (don't bind to the uas driver);
4780 w = NO_WP_DETECT (don't test whether the
4781 medium is write-protected).
4782 y = ALWAYS_SYNC (issue a SYNCHRONIZE_CACHE
4783 even if the device claims no cache)
4784 Example: quirks=0419:aaf5:rl,0421:0433:rc
4786 user_debug= [KNL,ARM]
4788 See arch/arm/Kconfig.debug help text.
4789 1 - undefined instruction events
4791 4 - invalid data aborts
4794 Example: user_debug=31
4797 [X86] Flags controlling user PTE allocations.
4799 nohigh = do not allocate PTE pages in
4800 HIGHMEM regardless of setting
4804 On X86_32, this is an alias for vdso32=. Otherwise:
4806 vdso=1: enable VDSO (the default)
4807 vdso=0: disable VDSO mapping
4809 vdso32= [X86] Control the 32-bit vDSO
4810 vdso32=1: enable 32-bit VDSO
4811 vdso32=0 or vdso32=2: disable 32-bit VDSO
4813 See the help text for CONFIG_COMPAT_VDSO for more
4814 details. If CONFIG_COMPAT_VDSO is set, the default is
4815 vdso32=0; otherwise, the default is vdso32=1.
4817 For compatibility with older kernels, vdso32=2 is an
4820 Try vdso32=0 if you encounter an error that says:
4821 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed!
4824 vector=percpu: enable percpu vector domain
4826 video= [FB] Frame buffer configuration
4827 See Documentation/fb/modedb.txt.
4829 video.brightness_switch_enabled= [0,1]
4830 If set to 1, on receiving an ACPI notify event
4831 generated by hotkey, video driver will adjust brightness
4832 level and then send out the event to user space through
4833 the allocated input device; If set to 0, video driver
4834 will only send out the event without touching backlight
4839 [VMMIO] Memory mapped virtio (platform) device.
4841 <size>@<baseaddr>:<irq>[:<id>]
4843 <size> := size (can use standard suffixes
4845 <baseaddr> := physical base address
4846 <irq> := interrupt number (as passed to
4848 <id> := (optional) platform device id
4850 virtio_mmio.device=1K@0x100b0000:48:7
4852 Can be used multiple times for multiple devices.
4854 vga= [BOOT,X86-32] Select a particular video mode
4855 See Documentation/x86/boot.txt and
4856 Documentation/svga.txt.
4857 Use vga=ask for menu.
4858 This is actually a boot loader parameter; the value is
4859 passed to the kernel using a special protocol.
4861 vm_debug[=options] [KNL] Available with CONFIG_DEBUG_VM=y.
4862 May slow down system boot speed, especially when
4863 enabled on systems with a large amount of memory.
4864 All options are enabled by default, and this
4865 interface is meant to allow for selectively
4866 enabling or disabling specific virtual memory
4869 Available options are:
4870 P Enable page structure init time poisoning
4871 - Disable all of the above options
4873 vmalloc=nn[KMG] [KNL,BOOT] Forces the vmalloc area to have an exact
4874 size of <nn>. This can be used to increase the
4875 minimum size (128MB on x86). It can also be used to
4876 decrease the size and leave more room for directly
4879 vmcp_cma=nn[MG] [KNL,S390]
4880 Sets the memory size reserved for contiguous memory
4881 allocations for the vmcp device driver.
4883 vmhalt= [KNL,S390] Perform z/VM CP command after system halt.
4886 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic.
4889 vmpoff= [KNL,S390] Perform z/VM CP command after power off.
4893 Controls the behavior of vsyscalls (i.e. calls to
4894 fixed addresses of 0xffffffffff600x00 from legacy
4895 code). Most statically-linked binaries and older
4896 versions of glibc use these calls. Because these
4897 functions are at fixed addresses, they make nice
4898 targets for exploits that can control RIP.
4900 emulate [default] Vsyscalls turn into traps and are
4901 emulated reasonably safely.
4903 native Vsyscalls are native syscall instructions.
4904 This is a little bit faster than trapping
4905 and makes a few dynamic recompilers work
4906 better than they would in emulation mode.
4907 It also makes exploits much easier to write.
4909 none Vsyscalls don't work at all. This makes
4910 them quite hard to use for exploits but
4911 might break your system.
4913 vt.color= [VT] Default text color.
4914 Format: 0xYX, X = foreground, Y = background.
4915 Default: 0x07 = light gray on black.
4917 vt.cur_default= [VT] Default cursor shape.
4918 Format: 0xCCBBAA, where AA, BB, and CC are the same as
4919 the parameters of the <Esc>[?A;B;Cc escape sequence;
4920 see VGA-softcursor.txt. Default: 2 = underline.
4922 vt.default_blu= [VT]
4923 Format: <blue0>,<blue1>,<blue2>,...,<blue15>
4924 Change the default blue palette of the console.
4925 This is a 16-member array composed of values
4928 vt.default_grn= [VT]
4929 Format: <green0>,<green1>,<green2>,...,<green15>
4930 Change the default green palette of the console.
4931 This is a 16-member array composed of values
4934 vt.default_red= [VT]
4935 Format: <red0>,<red1>,<red2>,...,<red15>
4936 Change the default red palette of the console.
4937 This is a 16-member array composed of values
4943 Set system-wide default UTF-8 mode for all tty's.
4944 Default is 1, i.e. UTF-8 mode is enabled for all
4945 newly opened terminals.
4947 vt.global_cursor_default=
4950 Set system-wide default for whether a cursor
4951 is shown on new VTs. Default is -1,
4952 i.e. cursors will be created by default unless
4953 overridden by individual drivers. 0 will hide
4954 cursors, 1 will display them.
4956 vt.italic= [VT] Default color for italic text; 0-15.
4959 vt.underline= [VT] Default color for underlined text; 0-15.
4962 watchdog timers [HW,WDT] For information on watchdog timers,
4963 see Documentation/watchdog/watchdog-parameters.txt
4964 or other driver-specific files in the
4965 Documentation/watchdog/ directory.
4967 workqueue.watchdog_thresh=
4968 If CONFIG_WQ_WATCHDOG is configured, workqueue can
4969 warn stall conditions and dump internal state to
4970 help debugging. 0 disables workqueue stall
4971 detection; otherwise, it's the stall threshold
4972 duration in seconds. The default value is 30 and
4973 it can be updated at runtime by writing to the
4974 corresponding sysfs file.
4976 workqueue.disable_numa
4977 By default, all work items queued to unbound
4978 workqueues are affine to the NUMA nodes they're
4979 issued on, which results in better behavior in
4980 general. If NUMA affinity needs to be disabled for
4981 whatever reason, this option can be used. Note
4982 that this also can be controlled per-workqueue for
4983 workqueues visible under /sys/bus/workqueue/.
4985 workqueue.power_efficient
4986 Per-cpu workqueues are generally preferred because
4987 they show better performance thanks to cache
4988 locality; unfortunately, per-cpu workqueues tend to
4989 be more power hungry than unbound workqueues.
4991 Enabling this makes the per-cpu workqueues which
4992 were observed to contribute significantly to power
4993 consumption unbound, leading to measurably lower
4994 power usage at the cost of small performance
4997 The default value of this parameter is determined by
4998 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT.
5000 workqueue.debug_force_rr_cpu
5001 Workqueue used to implicitly guarantee that work
5002 items queued without explicit CPU specified are put
5003 on the local CPU. This guarantee is no longer true
5004 and while local CPU is still preferred work items
5005 may be put on foreign CPUs. This debug option
5006 forces round-robin CPU selection to flush out
5007 usages which depend on the now broken guarantee.
5008 When enabled, memory and cache locality will be
5011 x2apic_phys [X86-64,APIC] Use x2apic physical mode instead of
5012 default x2apic cluster mode on platforms
5015 x86_intel_mid_timer= [X86-32,APBT]
5016 Choose timer option for x86 Intel MID platform.
5017 Two valid options are apbt timer only and lapic timer
5018 plus one apbt timer for broadcast timer.
5019 x86_intel_mid_timer=apbt_only | lapic_and_apbt
5021 xen_512gb_limit [KNL,X86-64,XEN]
5022 Restricts the kernel running paravirtualized under Xen
5023 to use only up to 512 GB of RAM. The reason to do so is
5024 crash analysis tools and Xen tools for doing domain
5025 save/restore/migration must be enabled to handle larger
5028 xen_emul_unplug= [HW,X86,XEN]
5029 Unplug Xen emulated devices
5030 Format: [unplug0,][unplug1]
5031 ide-disks -- unplug primary master IDE devices
5032 aux-ide-disks -- unplug non-primary-master IDE devices
5033 nics -- unplug network devices
5034 all -- unplug all emulated devices (NICs and IDE disks)
5035 unnecessary -- unplugging emulated devices is
5036 unnecessary even if the host did not respond to
5038 never -- do not unplug even if version check succeeds
5040 xen_nopvspin [X86,XEN]
5041 Disables the ticketlock slowpath using Xen PV
5045 Disables the PV optimizations forcing the HVM guest to
5046 run as generic HVM guest with no PV drivers.
5048 xen_scrub_pages= [XEN]
5049 Boolean option to control scrubbing pages before giving them back
5050 to Xen, for use by other domains. Can be also changed at runtime
5051 with /sys/devices/system/xen_memory/xen_memory0/scrub_pages.
5052 Default value controlled with CONFIG_XEN_SCRUB_PAGES_DEFAULT.
5054 xirc2ps_cs= [NET,PCMCIA]
5056 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]]
5058 xhci-hcd.quirks [USB,KNL]
5059 A hex value specifying bitmask with supplemental xhci
5060 host controller quirks. Meaning of each bit can be
5061 consulted in header drivers/usb/host/xhci.h.
projects / linux.git / blob