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/alsa-parameters.txt
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] verbose self-tests
754 Print debugging info while doing the locking API
756 We default to 0 (no extra messages), setting it to
757 1 will print _a lot_ more information - normally
758 only useful to kernel developers.
760 debug_objects [KNL] Enable object debugging
763 [KNL] Disable object debugging
765 debug_guardpage_minorder=
766 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this
767 parameter allows control of the order of pages that will
768 be intentionally kept free (and hence protected) by the
769 buddy allocator. Bigger value increase the probability
770 of catching random memory corruption, but reduce the
771 amount of memory for normal system use. The maximum
772 possible value is MAX_ORDER/2. Setting this parameter
773 to 1 or 2 should be enough to identify most random
774 memory corruption problems caused by bugs in kernel or
775 driver code when a CPU writes to (or reads from) a
776 random memory location. Note that there exists a class
777 of memory corruptions problems caused by buggy H/W or
778 F/W or by drivers badly programing DMA (basically when
779 memory is written at bus level and the CPU MMU is
780 bypassed) which are not detectable by
781 CONFIG_DEBUG_PAGEALLOC, hence this option will not help
782 tracking down these problems.
785 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this
786 parameter enables the feature at boot time. In
787 default, it is disabled. We can avoid allocating huge
788 chunk of memory for debug pagealloc if we don't enable
789 it at boot time and the system will work mostly same
790 with the kernel built without CONFIG_DEBUG_PAGEALLOC.
791 on: enable the feature
793 debugpat [X86] Enable PAT debugging
795 decnet.addr= [HW,NET]
796 Format: <area>[,<node>]
797 See also Documentation/networking/decnet.txt.
800 [same as hugepagesz=] The size of the default
801 HugeTLB page size. This is the size represented by
802 the legacy /proc/ hugepages APIs, used for SHM, and
803 default size when mounting hugetlbfs filesystems.
804 Defaults to the default architecture's huge page size
808 Set number of hash buckets for dentry cache.
810 disable_1tb_segments [PPC]
811 Disables the use of 1TB hash page table segments. This
812 causes the kernel to fall back to 256MB segments which
813 can be useful when debugging issues that require an SLB
817 See Documentation/networking/ipv6.txt.
820 Disable RADIX MMU mode on POWER9
822 disable_cpu_apicid= [X86,APIC,SMP]
824 The number of initial APIC ID for the
825 corresponding CPU to be disabled at boot,
826 mostly used for the kdump 2nd kernel to
827 disable BSP to wake up multiple CPUs without
828 causing system reset or hang due to sending
831 disable_ddw [PPC/PSERIES]
832 Disable Dynamic DMA Window support. Use this if
833 to workaround buggy firmware.
836 See Documentation/networking/ipv6.txt.
838 disable_mtrr_cleanup [X86]
839 The kernel tries to adjust MTRR layout from continuous
840 to discrete, to make X server driver able to add WB
841 entry later. This parameter disables that.
843 disable_mtrr_trim [X86, Intel and AMD only]
844 By default the kernel will trim any uncacheable
845 memory out of your available memory pool based on
846 MTRR settings. This parameter disables that behavior,
847 possibly causing your machine to run very slowly.
849 disable_timer_pin_1 [X86]
850 Disable PIN 1 of APIC timer
851 Can be useful to work around chipset bugs.
853 dis_ucode_ldr [X86] Disable the microcode loader.
855 dma_debug=off If the kernel is compiled with DMA_API_DEBUG support,
856 this option disables the debugging code at boot.
858 dma_debug_entries=<number>
859 This option allows to tune the number of preallocated
860 entries for DMA-API debugging code. One entry is
861 required per DMA-API allocation. Use this if the
862 DMA-API debugging code disables itself because the
863 architectural default is too low.
865 dma_debug_driver=<driver_name>
866 With this option the DMA-API debugging driver
867 filter feature can be enabled at boot time. Just
868 pass the driver to filter for as the parameter.
869 The filter can be disabled or changed to another
870 driver later using sysfs.
872 drm.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>]
873 Broken monitors, graphic adapters, KVMs and EDIDless
874 panels may send no or incorrect EDID data sets.
875 This parameter allows to specify an EDID data sets
876 in the /lib/firmware directory that are used instead.
877 Generic built-in EDID data sets are used, if one of
878 edid/1024x768.bin, edid/1280x1024.bin,
879 edid/1680x1050.bin, or edid/1920x1080.bin is given
880 and no file with the same name exists. Details and
881 instructions how to build your own EDID data are
882 available in Documentation/EDID/HOWTO.txt. An EDID
883 data set will only be used for a particular connector,
884 if its name and a colon are prepended to the EDID
885 name. Each connector may use a unique EDID data
886 set by separating the files with a comma. An EDID
887 data set with no connector name will be used for
888 any connectors not explicitly specified.
893 Format: {"off" | "known"}
894 Control how the dt_cpu_ftrs device-tree binding is
895 used for CPU feature discovery and setup (if it
897 off: Do not use it, fall back to legacy cpu table.
898 known: Do not pass through unknown features to guests
899 or userspace, only those that the kernel is aware of.
901 dump_apple_properties [X86]
902 Dump name and content of EFI device properties on
903 x86 Macs. Useful for driver authors to determine
904 what data is available or for reverse-engineering.
906 dyndbg[="val"] [KNL,DYNAMIC_DEBUG]
907 module.dyndbg[="val"]
908 Enable debug messages at boot time. See
909 Documentation/admin-guide/dynamic-debug-howto.rst
912 nompx [X86] Disables Intel Memory Protection Extensions.
913 See Documentation/x86/intel_mpx.txt for more
914 information about the feature.
916 nopku [X86] Disable Memory Protection Keys CPU feature found
919 module.async_probe [KNL]
920 Enable asynchronous probe on this module.
922 early_ioremap_debug [KNL]
923 Enable debug messages in early_ioremap support. This
924 is useful for tracking down temporary early mappings
925 which are not unmapped.
927 earlycon= [KNL] Output early console device and options.
929 [ARM64] The early console is determined by the
930 stdout-path property in device tree's chosen node,
931 or determined by the ACPI SPCR table.
933 [X86] When used with no options the early console is
934 determined by the ACPI SPCR table.
936 cdns,<addr>[,options]
937 Start an early, polled-mode console on a Cadence
938 (xuartps) serial port at the specified address. Only
939 supported option is baud rate. If baud rate is not
940 specified, the serial port must already be setup and
943 uart[8250],io,<addr>[,options]
944 uart[8250],mmio,<addr>[,options]
945 uart[8250],mmio32,<addr>[,options]
946 uart[8250],mmio32be,<addr>[,options]
947 uart[8250],0x<addr>[,options]
948 Start an early, polled-mode console on the 8250/16550
949 UART at the specified I/O port or MMIO address.
950 MMIO inter-register address stride is either 8-bit
951 (mmio) or 32-bit (mmio32 or mmio32be).
952 If none of [io|mmio|mmio32|mmio32be], <addr> is assumed
953 to be equivalent to 'mmio'. 'options' are specified
954 in the same format described for "console=ttyS<n>"; if
955 unspecified, the h/w is not initialized.
959 Start an early, polled-mode console on a pl011 serial
960 port at the specified address. The pl011 serial port
961 must already be setup and configured. Options are not
962 yet supported. If 'mmio32' is specified, then only
963 the driver will use only 32-bit accessors to read/write
964 the device registers.
967 Start an early, polled-mode console on a meson serial
968 port at the specified address. The serial port must
969 already be setup and configured. Options are not yet
973 Start an early, polled-mode console on an msm serial
974 port at the specified address. The serial port
975 must already be setup and configured. Options are not
979 Start an early, polled-mode console on an msm serial
980 dm port at the specified address. The serial port
981 must already be setup and configured. Options are not
985 Start an early, polled-mode console on a serial port
986 of an Actions Semi SoC, such as S500 or S900, at the
987 specified address. The serial port must already be
988 setup and configured. Options are not yet supported.
990 smh Use ARM semihosting calls for early console.
998 Use early console provided by serial driver available
999 on Samsung SoCs, requires selecting proper type and
1000 a correct base address of the selected UART port. The
1001 serial port must already be setup and configured.
1002 Options are not yet supported.
1005 Start an early, polled-mode console on a lantiq serial
1006 (lqasc) port at the specified address. The serial port
1007 must already be setup and configured. Options are not
1012 Use early console provided by Freescale LP UART driver
1013 found on Freescale Vybrid and QorIQ LS1021A processors.
1014 A valid base address must be provided, and the serial
1015 port must already be setup and configured.
1018 Start an early, polled-mode console on the
1019 Armada 3700 serial port at the specified
1020 address. The serial port must already be setup
1021 and configured. Options are not yet supported.
1023 earlyprintk= [X86,SH,ARM,M68k,S390]
1028 earlyprintk=serial[,ttySn[,baudrate]]
1029 earlyprintk=serial[,0x...[,baudrate]]
1030 earlyprintk=ttySn[,baudrate]
1031 earlyprintk=dbgp[debugController#]
1032 earlyprintk=pciserial,bus:device.function[,baudrate]
1033 earlyprintk=xdbc[xhciController#]
1035 earlyprintk is useful when the kernel crashes before
1036 the normal console is initialized. It is not enabled by
1037 default because it has some cosmetic problems.
1039 Append ",keep" to not disable it when the real console
1042 Only one of vga, efi, serial, or usb debug port can
1045 Currently only ttyS0 and ttyS1 may be specified by
1046 name. Other I/O ports may be explicitly specified
1047 on some architectures (x86 and arm at least) by
1048 replacing ttySn with an I/O port address, like this:
1049 earlyprintk=serial,0x1008,115200
1050 You can find the port for a given device in
1051 /proc/tty/driver/serial:
1052 2: uart:ST16650V2 port:00001008 irq:18 ...
1054 Interaction with the standard serial driver is not
1057 The VGA and EFI output is eventually overwritten by
1060 The xen output can only be used by Xen PV guests.
1062 The sclp output can only be used on s390.
1064 edac_report= [HW,EDAC] Control how to report EDAC event
1065 Format: {"on" | "off" | "force"}
1066 on: enable EDAC to report H/W event. May be overridden
1067 by other higher priority error reporting module.
1068 off: disable H/W event reporting through EDAC.
1069 force: enforce the use of EDAC to report H/W event.
1072 ekgdboc= [X86,KGDB] Allow early kernel console debugging
1075 This is designed to be used in conjunction with
1076 the boot argument: earlyprintk=vga
1079 Format: {"off" | "on" | "skip[mbr]"}
1082 Format: { "old_map", "nochunk", "noruntime", "debug" }
1083 old_map [X86-64]: switch to the old ioremap-based EFI
1084 runtime services mapping. 32-bit still uses this one by
1086 nochunk: disable reading files in "chunks" in the EFI
1087 boot stub, as chunking can cause problems with some
1088 firmware implementations.
1089 noruntime : disable EFI runtime services support
1090 debug: enable misc debug output
1092 efi_no_storage_paranoia [EFI; X86]
1093 Using this parameter you can use more than 50% of
1094 your efi variable storage. Use this parameter only if
1095 you are really sure that your UEFI does sane gc and
1096 fulfills the spec otherwise your board may brick.
1098 efi_fake_mem= nn[KMG]@ss[KMG]:aa[,nn[KMG]@ss[KMG]:aa,..] [EFI; X86]
1099 Add arbitrary attribute to specific memory range by
1100 updating original EFI memory map.
1101 Region of memory which aa attribute is added to is
1103 If efi_fake_mem=2G@4G:0x10000,2G@0x10a0000000:0x10000
1104 is specified, EFI_MEMORY_MORE_RELIABLE(0x10000)
1105 attribute is added to range 0x100000000-0x180000000 and
1106 0x10a0000000-0x1120000000.
1108 Using this parameter you can do debugging of EFI memmap
1109 related feature. For example, you can do debugging of
1110 Address Range Mirroring feature even if your box
1113 efivar_ssdt= [EFI; X86] Name of an EFI variable that contains an SSDT
1114 that is to be dynamically loaded by Linux. If there are
1115 multiple variables with the same name but with different
1116 vendor GUIDs, all of them will be loaded. See
1117 Documentation/acpi/ssdt-overlays.txt for details.
1120 eisa_irq_edge= [PARISC,HW]
1121 See header of drivers/parisc/eisa.c.
1124 See comment before function elanfreq_setup() in
1125 arch/x86/kernel/cpu/cpufreq/elanfreq.c.
1128 Format: {"cfq" | "deadline" | "noop"}
1129 See Documentation/block/cfq-iosched.txt and
1130 Documentation/block/deadline-iosched.txt for details.
1132 elfcorehdr=[size[KMG]@]offset[KMG] [IA64,PPC,SH,X86,S390]
1133 Specifies physical address of start of kernel core
1134 image elf header and optionally the size. Generally
1135 kexec loader will pass this option to capture kernel.
1136 See Documentation/kdump/kdump.txt for details.
1138 enable_mtrr_cleanup [X86]
1139 The kernel tries to adjust MTRR layout from continuous
1140 to discrete, to make X server driver able to add WB
1141 entry later. This parameter enables that.
1143 enable_timer_pin_1 [X86]
1144 Enable PIN 1 of APIC timer
1145 Can be useful to work around chipset bugs
1146 (in particular on some ATI chipsets).
1147 The kernel tries to set a reasonable default.
1149 enforcing [SELINUX] Set initial enforcing status.
1151 See security/selinux/Kconfig help text.
1152 0 -- permissive (log only, no denials).
1153 1 -- enforcing (deny and log).
1155 Value can be changed at runtime via /selinux/enforce.
1158 Disable Error Record Serialization Table (ERST)
1161 ether= [HW,NET] Ethernet cards parameters
1162 This option is obsoleted by the "netdev=" option, which
1163 has equivalent usage. See its documentation for details.
1167 Permit 'security.evm' to be updated regardless of
1168 current integrity status.
1172 fail_make_request=[KNL]
1173 General fault injection mechanism.
1174 Format: <interval>,<probability>,<space>,<times>
1175 See also Documentation/fault-injection/.
1178 See Documentation/blockdev/floppy.txt.
1180 force_pal_cache_flush
1181 [IA-64] Avoid check_sal_cache_flush which may hang on
1182 buggy SAL_CACHE_FLUSH implementations. Using this
1183 parameter will force ia64_sal_cache_flush to call
1184 ia64_pal_cache_flush instead of SAL_CACHE_FLUSH.
1187 Forcefully enable Physical Address Extension (PAE).
1188 Many Pentium M systems disable PAE but may have a
1189 functionally usable PAE implementation.
1190 Warning: use of this parameter will taint the kernel
1191 and may cause unknown problems.
1194 [FTRACE] will set and start the specified tracer
1195 as early as possible in order to facilitate early
1198 ftrace_dump_on_oops[=orig_cpu]
1199 [FTRACE] will dump the trace buffers on oops.
1200 If no parameter is passed, ftrace will dump
1201 buffers of all CPUs, but if you pass orig_cpu, it will
1202 dump only the buffer of the CPU that triggered the
1205 ftrace_filter=[function-list]
1206 [FTRACE] Limit the functions traced by the function
1207 tracer at boot up. function-list is a comma separated
1208 list of functions. This list can be changed at run
1209 time by the set_ftrace_filter file in the debugfs
1212 ftrace_notrace=[function-list]
1213 [FTRACE] Do not trace the functions specified in
1214 function-list. This list can be changed at run time
1215 by the set_ftrace_notrace file in the debugfs
1218 ftrace_graph_filter=[function-list]
1219 [FTRACE] Limit the top level callers functions traced
1220 by the function graph tracer at boot up.
1221 function-list is a comma separated list of functions
1222 that can be changed at run time by the
1223 set_graph_function file in the debugfs tracing directory.
1225 ftrace_graph_notrace=[function-list]
1226 [FTRACE] Do not trace from the functions specified in
1227 function-list. This list is a comma separated list of
1228 functions that can be changed at run time by the
1229 set_graph_notrace file in the debugfs tracing directory.
1231 ftrace_graph_max_depth=<uint>
1232 [FTRACE] Used with the function graph tracer. This is
1233 the max depth it will trace into a function. This value
1234 can be changed at run time by the max_graph_depth file
1235 in the tracefs tracing directory. default: 0 (no limit)
1238 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad
1239 support via parallel port (up to 5 devices per port)
1240 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5>
1241 See also Documentation/input/devices/joystick-parport.rst
1245 gart_fix_e820= [X86_64] disable the fix e820 for K8 GART
1249 gcov_persist= [GCOV] When non-zero (default), profiling data for
1250 kernel modules is saved and remains accessible via
1251 debugfs, even when the module is unloaded/reloaded.
1252 When zero, profiling data is discarded and associated
1253 debugfs files are removed at module unload time.
1255 goldfish [X86] Enable the goldfish android emulator platform.
1256 Don't use this when you are not running on the
1259 gpt [EFI] Forces disk with valid GPT signature but
1260 invalid Protective MBR to be treated as GPT. If the
1261 primary GPT is corrupted, it enables the backup/alternate
1262 GPT to be used instead.
1264 grcan.enable0= [HW] Configuration of physical interface 0. Determines
1265 the "Enable 0" bit of the configuration register.
1268 grcan.enable1= [HW] Configuration of physical interface 1. Determines
1269 the "Enable 0" bit of the configuration register.
1272 grcan.select= [HW] Select which physical interface to use.
1275 grcan.txsize= [HW] Sets the size of the tx buffer.
1276 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0.
1278 grcan.rxsize= [HW] Sets the size of the rx buffer.
1279 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0.
1282 gpio-mockup.gpio_mockup_ranges
1283 [HW] Sets the ranges of gpiochip of for this device.
1284 Format: <start1>,<end1>,<start2>,<end2>...
1286 hardlockup_all_cpu_backtrace=
1287 [KNL] Should the hard-lockup detector generate
1288 backtraces on all cpus.
1291 hashdist= [KNL,NUMA] Large hashes allocated during boot
1292 are distributed across NUMA nodes. Defaults on
1293 for 64-bit NUMA, off otherwise.
1294 Format: 0 | 1 (for off | on)
1296 hcl= [IA-64] SGI's Hardware Graph compatibility layer
1298 hd= [EIDE] (E)IDE hard drive subsystem geometry
1299 Format: <cyl>,<head>,<sect>
1302 Disable Hardware Error Source Table (HEST) support;
1303 corresponding firmware-first mode error processing
1304 logic will be disabled.
1306 highmem=nn[KMG] [KNL,BOOT] forces the highmem zone to have an exact
1307 size of <nn>. This works even on boxes that have no
1308 highmem otherwise. This also works to reduce highmem
1309 size on bigger boxes.
1311 highres= [KNL] Enable/disable high resolution timer mode.
1312 Valid parameters: "on", "off"
1316 See Documentation/isdn/README.HiSax.
1320 hpet= [X86-32,HPET] option to control HPET usage
1321 Format: { enable (default) | disable | force |
1323 disable: disable HPET and use PIT instead
1324 force: allow force enabled of undocumented chips (ICH4,
1326 verbose: show contents of HPET registers during setup
1328 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET
1329 registers. Default set by CONFIG_HPET_MMAP_DEFAULT.
1331 hugepages= [HW,X86-32,IA-64] HugeTLB pages to allocate at boot.
1332 hugepagesz= [HW,IA-64,PPC,X86-64] The size of the HugeTLB pages.
1333 On x86-64 and powerpc, this option can be specified
1334 multiple times interleaved with hugepages= to reserve
1335 huge pages of different sizes. Valid pages sizes on
1336 x86-64 are 2M (when the CPU supports "pse") and 1G
1337 (when the CPU supports the "pdpe1gb" cpuinfo flag).
1340 [KNL] Should the hung task detector generate panics.
1343 A nonzero value instructs the kernel to panic when a
1344 hung task is detected. The default value is controlled
1345 by the CONFIG_BOOTPARAM_HUNG_TASK_PANIC build-time
1346 option. The value selected by this boot parameter can
1347 be changed later by the kernel.hung_task_panic sysctl.
1349 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC)
1350 terminal devices. Valid values: 0..8
1351 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs.
1352 If specified, z/VM IUCV HVC accepts connections
1353 from listed z/VM user IDs only.
1355 Do not unregister boot console at start. This is only
1356 useful for debugging when something happens in the window
1357 between unregistering the boot console and initializing
1360 i2c_bus= [HW] Override the default board specific I2C bus speed
1361 or register an additional I2C bus that is not
1362 registered from board initialization code.
1366 i8042.debug [HW] Toggle i8042 debug mode
1367 i8042.unmask_kbd_data
1368 [HW] Enable printing of interrupt data from the KBD port
1369 (disabled by default, and as a pre-condition
1370 requires that i8042.debug=1 be enabled)
1371 i8042.direct [HW] Put keyboard port into non-translated mode
1372 i8042.dumbkbd [HW] Pretend that controller can only read data from
1373 keyboard and cannot control its state
1374 (Don't attempt to blink the leds)
1375 i8042.noaux [HW] Don't check for auxiliary (== mouse) port
1376 i8042.nokbd [HW] Don't check/create keyboard port
1377 i8042.noloop [HW] Disable the AUX Loopback command while probing
1379 i8042.nomux [HW] Don't check presence of an active multiplexing
1381 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX
1383 i8042.notimeout [HW] Ignore timeout condition signalled by controller
1384 i8042.reset [HW] Reset the controller during init, cleanup and
1385 suspend-to-ram transitions, only during s2r
1386 transitions, or never reset
1387 Format: { 1 | Y | y | 0 | N | n }
1388 1, Y, y: always reset controller
1389 0, N, n: don't ever reset controller
1390 Default: only on s2r transitions on x86; most other
1391 architectures force reset to be always executed
1392 i8042.unlock [HW] Unlock (ignore) the keylock
1393 i8042.kbdreset [HW] Reset device connected to KBD port
1397 i8k.ignore_dmi [HW] Continue probing hardware even if DMI data
1398 indicates that the driver is running on unsupported
1400 i8k.force [HW] Activate i8k driver even if SMM BIOS signature
1401 does not match list of supported models.
1403 [HW] Report power status in /proc/i8k
1404 (disabled by default)
1405 i8k.restricted [HW] Allow controlling fans only if SYS_ADMIN
1408 i915.invert_brightness=
1409 [DRM] Invert the sense of the variable that is used to
1410 set the brightness of the panel backlight. Normally a
1411 brightness value of 0 indicates backlight switched off,
1412 and the maximum of the brightness value sets the backlight
1413 to maximum brightness. If this parameter is set to 0
1414 (default) and the machine requires it, or this parameter
1415 is set to 1, a brightness value of 0 sets the backlight
1416 to maximum brightness, and the maximum of the brightness
1417 value switches the backlight off.
1418 -1 -- never invert brightness
1419 0 -- machine default
1420 1 -- force brightness inversion
1423 Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]]
1425 ide-core.nodma= [HW] (E)IDE subsystem
1426 Format: =0.0 to prevent dma on hda, =0.1 hdb =1.0 hdc
1427 .vlb_clock .pci_clock .noflush .nohpa .noprobe .nowerr
1428 .cdrom .chs .ignore_cable are additional options
1429 See Documentation/ide/ide.txt.
1431 ide-generic.probe-mask= [HW] (E)IDE subsystem
1433 Probe mask for legacy ISA IDE ports. Depending on
1434 platform up to 6 ports are supported, enabled by
1435 setting corresponding bits in the mask to 1. The
1436 default value is 0x0, which has a special meaning.
1437 On systems that have PCI, it triggers scanning the
1438 PCI bus for the first and the second port, which
1439 are then probed. On systems without PCI the value
1440 of 0x0 enables probing the two first ports as if it
1443 ide-pci-generic.all-generic-ide [HW] (E)IDE subsystem
1444 Claim all unknown PCI IDE storage controllers.
1447 Format: idle=poll, idle=halt, idle=nomwait
1448 Poll forces a polling idle loop that can slightly
1449 improve the performance of waking up a idle CPU, but
1450 will use a lot of power and make the system run hot.
1452 idle=halt: Halt is forced to be used for CPU idle.
1453 In such case C2/C3 won't be used again.
1454 idle=nomwait: Disable mwait for CPU C-states
1456 ieee754= [MIPS] Select IEEE Std 754 conformance mode
1457 Format: { strict | legacy | 2008 | relaxed }
1460 Choose which programs will be accepted for execution
1461 based on the IEEE 754 NaN encoding(s) supported by
1462 the FPU and the NaN encoding requested with the value
1463 of an ELF file header flag individually set by each
1464 binary. Hardware implementations are permitted to
1465 support either or both of the legacy and the 2008 NaN
1468 Available settings are as follows:
1469 strict accept binaries that request a NaN encoding
1470 supported by the FPU
1471 legacy only accept legacy-NaN binaries, if supported
1473 2008 only accept 2008-NaN binaries, if supported
1475 relaxed accept any binaries regardless of whether
1476 supported by the FPU
1478 The FPU emulator is always able to support both NaN
1479 encodings, so if no FPU hardware is present or it has
1480 been disabled with 'nofpu', then the settings of
1481 'legacy' and '2008' strap the emulator accordingly,
1482 'relaxed' straps the emulator for both legacy-NaN and
1483 2008-NaN, whereas 'strict' enables legacy-NaN only on
1484 legacy processors and both NaN encodings on MIPS32 or
1487 The setting for ABS.fmt/NEG.fmt instruction execution
1488 mode generally follows that for the NaN encoding,
1489 except where unsupported by hardware.
1491 ignore_loglevel [KNL]
1492 Ignore loglevel setting - this will print /all/
1493 kernel messages to the console. Useful for debugging.
1494 We also add it as printk module parameter, so users
1495 could change it dynamically, usually by
1496 /sys/module/printk/parameters/ignore_loglevel.
1499 Ignore RLIMIT_DATA setting for data mappings,
1500 print warning at first misuse. Can be changed via
1501 /sys/module/kernel/parameters/ignore_rlimit_data.
1503 ihash_entries= [KNL]
1504 Set number of hash buckets for inode cache.
1506 ima_appraise= [IMA] appraise integrity measurements
1507 Format: { "off" | "enforce" | "fix" | "log" }
1510 ima_appraise_tcb [IMA]
1511 The builtin appraise policy appraises all files
1514 ima_canonical_fmt [IMA]
1515 Use the canonical format for the binary runtime
1516 measurements, instead of host native format.
1519 Format: { md5 | sha1 | rmd160 | sha256 | sha384
1523 The list of supported hash algorithms is defined
1524 in crypto/hash_info.h.
1527 The builtin policies to load during IMA setup.
1528 Format: "tcb | appraise_tcb | secure_boot |
1531 The "tcb" policy measures all programs exec'd, files
1532 mmap'd for exec, and all files opened with the read
1533 mode bit set by either the effective uid (euid=0) or
1536 The "appraise_tcb" policy appraises the integrity of
1537 all files owned by root. (This is the equivalent
1538 of ima_appraise_tcb.)
1540 The "secure_boot" policy appraises the integrity
1541 of files (eg. kexec kernel image, kernel modules,
1542 firmware, policy, etc) based on file signatures.
1544 The "fail_securely" policy forces file signature
1545 verification failure also on privileged mounted
1546 filesystems with the SB_I_UNVERIFIABLE_SIGNATURE
1549 ima_tcb [IMA] Deprecated. Use ima_policy= instead.
1550 Load a policy which meets the needs of the Trusted
1551 Computing Base. This means IMA will measure all
1552 programs exec'd, files mmap'd for exec, and all files
1553 opened for read by uid=0.
1556 Select one of defined IMA measurements template formats.
1557 Formats: { "ima" | "ima-ng" | "ima-sig" }
1561 [IMA] Define a custom template format.
1562 Format: { "field1|...|fieldN" }
1564 ima.ahash_minsize= [IMA] Minimum file size for asynchronous hash usage
1565 Format: <min_file_size>
1566 Set the minimal file size for using asynchronous hash.
1567 If left unspecified, ahash usage is disabled.
1569 ahash performance varies for different data sizes on
1570 different crypto accelerators. This option can be used
1571 to achieve the best performance for a particular HW.
1573 ima.ahash_bufsize= [IMA] Asynchronous hash buffer size
1575 Set hashing buffer size. Default: 4k.
1577 ahash performance varies for different chunk sizes on
1578 different crypto accelerators. This option can be used
1579 to achieve best performance for particular HW.
1583 Run specified binary instead of /sbin/init as init
1586 initcall_debug [KNL] Trace initcalls as they are executed. Useful
1587 for working out where the kernel is dying during
1590 initcall_blacklist= [KNL] Do not execute a comma-separated list of
1591 initcall functions. Useful for debugging built-in
1592 modules and initcalls.
1594 initrd= [BOOT] Specify the location of the initial ramdisk
1596 init_pkru= [x86] Specify the default memory protection keys rights
1597 register contents for all processes. 0x55555554 by
1598 default (disallow access to all but pkey 0). Can
1599 override in debugfs after boot.
1601 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver
1604 int_pln_enable [x86] Enable power limit notification interrupt
1606 integrity_audit=[IMA]
1607 Format: { "0" | "1" }
1608 0 -- basic integrity auditing messages. (Default)
1609 1 -- additional integrity auditing messages.
1611 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option
1613 Enable intel iommu driver.
1615 Disable intel iommu driver.
1616 igfx_off [Default Off]
1617 By default, gfx is mapped as normal device. If a gfx
1618 device has a dedicated DMAR unit, the DMAR unit is
1619 bypassed by not enabling DMAR with this option. In
1620 this case, gfx device will use physical address for
1623 With this option iommu will not optimize to look
1624 for io virtual address below 32-bit forcing dual
1625 address cycle on pci bus for cards supporting greater
1626 than 32-bit addressing. The default is to look
1627 for translation below 32-bit and if not available
1628 then look in the higher range.
1629 strict [Default Off]
1630 With this option on every unmap_single operation will
1631 result in a hardware IOTLB flush operation as opposed
1632 to batching them for performance.
1633 sp_off [Default Off]
1634 By default, super page will be supported if Intel IOMMU
1635 has the capability. With this option, super page will
1637 ecs_off [Default Off]
1638 By default, extended context tables will be supported if
1639 the hardware advertises that it has support both for the
1640 extended tables themselves, and also PASID support. With
1641 this option set, extended tables will not be used even
1642 on hardware which claims to support them.
1643 tboot_noforce [Default Off]
1644 Do not force the Intel IOMMU enabled under tboot.
1645 By default, tboot will force Intel IOMMU on, which
1646 could harm performance of some high-throughput
1647 devices like 40GBit network cards, even if identity
1649 Note that using this option lowers the security
1650 provided by tboot because it makes the system
1651 vulnerable to DMA attacks.
1653 intel_idle.max_cstate= [KNL,HW,ACPI,X86]
1654 0 disables intel_idle and fall back on acpi_idle.
1655 1 to 9 specify maximum depth of C-state.
1659 Do not enable intel_pstate as the default
1660 scaling driver for the supported processors
1662 Use intel_pstate as a scaling driver, but configure it
1663 to work with generic cpufreq governors (instead of
1664 enabling its internal governor). This mode cannot be
1665 used along with the hardware-managed P-states (HWP)
1668 Enable intel_pstate on systems that prohibit it by default
1669 in favor of acpi-cpufreq. Forcing the intel_pstate driver
1670 instead of acpi-cpufreq may disable platform features, such
1671 as thermal controls and power capping, that rely on ACPI
1672 P-States information being indicated to OSPM and therefore
1673 should be used with caution. This option does not work with
1674 processors that aren't supported by the intel_pstate driver
1675 or on platforms that use pcc-cpufreq instead of acpi-cpufreq.
1677 Do not enable hardware P state control (HWP)
1680 Only load intel_pstate on systems which support
1681 hardware P state control (HWP) if available.
1683 Enforce ACPI _PPC performance limits. If the Fixed ACPI
1684 Description Table, specifies preferred power management
1685 profile as "Enterprise Server" or "Performance Server",
1686 then this feature is turned on by default.
1688 Allow per-logical-CPU P-State performance control limits using
1689 cpufreq sysfs interface
1691 intremap= [X86-64, Intel-IOMMU]
1692 on enable Interrupt Remapping (default)
1693 off disable Interrupt Remapping
1694 nosid disable Source ID checking
1696 BIOS x2APIC opt-out request will be ignored
1697 nopost disable Interrupt Posting
1699 iomem= Disable strict checking of access to MMIO memory
1700 strict regions from userspace.
1714 nobypass [PPC/POWERNV]
1715 Disable IOMMU bypass, using IOMMU for PCI devices.
1718 [ARM64] Configure DMA to bypass the IOMMU by default.
1719 Format: { "0" | "1" }
1720 0 - Use IOMMU translation for DMA.
1721 1 - Bypass the IOMMU for DMA.
1722 unset - Use IOMMU translation for DMA.
1724 io7= [HW] IO7 for Marvel based alpha systems
1725 See comment before marvel_specify_io7 in
1726 arch/alpha/kernel/core_marvel.c.
1728 io_delay= [X86] I/O delay method
1730 Standard port 0x80 based delay
1732 Alternate port 0xed based delay (needed on some systems)
1734 Simple two microseconds delay
1739 See Documentation/filesystems/nfs/nfsroot.txt.
1741 irqaffinity= [SMP] Set the default irq affinity mask
1742 The argument is a cpu list, as described above.
1744 irqchip.gicv2_force_probe=
1747 Force the kernel to look for the second 4kB page
1748 of a GICv2 controller even if the memory range
1749 exposed by the device tree is too small.
1751 irqchip.gicv3_nolpi=
1753 Force the kernel to ignore the availability of
1754 LPIs (and by consequence ITSs). Intended for system
1755 that use the kernel as a bootloader, and thus want
1756 to let secondary kernels in charge of setting up
1760 When an interrupt is not handled search all handlers
1761 for it. Intended to get systems with badly broken
1765 When an interrupt is not handled search all handlers
1766 for it. Also check all handlers each timer
1767 interrupt. Intended to get systems with badly broken
1771 Format: <RDP>,<reset>,<pci_scan>,<verbosity>
1773 isolcpus= [KNL,SMP,ISOL] Isolate a given set of CPUs from disturbance.
1774 [Deprecated - use cpusets instead]
1775 Format: [flag-list,]<cpu-list>
1777 Specify one or more CPUs to isolate from disturbances
1778 specified in the flag list (default: domain):
1781 Disable the tick when a single task runs.
1783 A residual 1Hz tick is offloaded to workqueues, which you
1784 need to affine to housekeeping through the global
1785 workqueue's affinity configured via the
1786 /sys/devices/virtual/workqueue/cpumask sysfs file, or
1787 by using the 'domain' flag described below.
1789 NOTE: by default the global workqueue runs on all CPUs,
1790 so to protect individual CPUs the 'cpumask' file has to
1791 be configured manually after bootup.
1794 Isolate from the general SMP balancing and scheduling
1795 algorithms. Note that performing domain isolation this way
1796 is irreversible: it's not possible to bring back a CPU to
1797 the domains once isolated through isolcpus. It's strongly
1798 advised to use cpusets instead to disable scheduler load
1799 balancing through the "cpuset.sched_load_balance" file.
1800 It offers a much more flexible interface where CPUs can
1801 move in and out of an isolated set anytime.
1803 You can move a process onto or off an "isolated" CPU via
1804 the CPU affinity syscalls or cpuset.
1805 <cpu number> begins at 0 and the maximum value is
1806 "number of CPUs in system - 1".
1808 The format of <cpu-list> is described above.
1814 ivrs_ioapic [HW,X86_64]
1815 Provide an override to the IOAPIC-ID<->DEVICE-ID
1816 mapping provided in the IVRS ACPI table. For
1817 example, to map IOAPIC-ID decimal 10 to
1818 PCI device 00:14.0 write the parameter as:
1819 ivrs_ioapic[10]=00:14.0
1821 ivrs_hpet [HW,X86_64]
1822 Provide an override to the HPET-ID<->DEVICE-ID
1823 mapping provided in the IVRS ACPI table. For
1824 example, to map HPET-ID decimal 0 to
1825 PCI device 00:14.0 write the parameter as:
1826 ivrs_hpet[0]=00:14.0
1828 ivrs_acpihid [HW,X86_64]
1829 Provide an override to the ACPI-HID:UID<->DEVICE-ID
1830 mapping provided in the IVRS ACPI table. For
1831 example, to map UART-HID:UID AMD0020:0 to
1832 PCI device 00:14.5 write the parameter as:
1833 ivrs_acpihid[00:14.5]=AMD0020:0
1835 js= [HW,JOY] Analog joystick
1836 See Documentation/input/joydev/joystick.rst.
1839 When CONFIG_RANDOMIZE_BASE is set, this disables
1840 kernel and module base offset ASLR (Address Space
1841 Layout Randomization).
1844 [KNL] Enforce KASAN (Kernel Address Sanitizer) to print
1845 report on every invalid memory access. Without this
1846 parameter KASAN will print report only for the first
1851 kernelcore= [KNL,X86,IA-64,PPC]
1852 Format: nn[KMGTPE] | nn% | "mirror"
1853 This parameter specifies the amount of memory usable by
1854 the kernel for non-movable allocations. The requested
1855 amount is spread evenly throughout all nodes in the
1856 system as ZONE_NORMAL. The remaining memory is used for
1857 movable memory in its own zone, ZONE_MOVABLE. In the
1858 event, a node is too small to have both ZONE_NORMAL and
1859 ZONE_MOVABLE, kernelcore memory will take priority and
1860 other nodes will have a larger ZONE_MOVABLE.
1862 ZONE_MOVABLE is used for the allocation of pages that
1863 may be reclaimed or moved by the page migration
1864 subsystem. Note that allocations like PTEs-from-HighMem
1865 still use the HighMem zone if it exists, and the Normal
1866 zone if it does not.
1868 It is possible to specify the exact amount of memory in
1869 the form of "nn[KMGTPE]", a percentage of total system
1870 memory in the form of "nn%", or "mirror". If "mirror"
1871 option is specified, mirrored (reliable) memory is used
1872 for non-movable allocations and remaining memory is used
1873 for Movable pages. "nn[KMGTPE]", "nn%", and "mirror"
1874 are exclusive, so you cannot specify multiple forms.
1876 kgdbdbgp= [KGDB,HW] kgdb over EHCI usb debug port.
1877 Format: <Controller#>[,poll interval]
1878 The controller # is the number of the ehci usb debug
1879 port as it is probed via PCI. The poll interval is
1880 optional and is the number seconds in between
1881 each poll cycle to the debug port in case you need
1882 the functionality for interrupting the kernel with
1883 gdb or control-c on the dbgp connection. When
1884 not using this parameter you use sysrq-g to break into
1885 the kernel debugger.
1887 kgdboc= [KGDB,HW] kgdb over consoles.
1888 Requires a tty driver that supports console polling,
1889 or a supported polling keyboard driver (non-usb).
1890 Serial only format: <serial_device>[,baud]
1891 keyboard only format: kbd
1892 keyboard and serial format: kbd,<serial_device>[,baud]
1893 Optional Kernel mode setting:
1894 kms, kbd format: kms,kbd
1895 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud]
1897 kgdbwait [KGDB] Stop kernel execution and enter the
1898 kernel debugger at the earliest opportunity.
1900 kmac= [MIPS] korina ethernet MAC address.
1901 Configure the RouterBoard 532 series on-chip
1902 Ethernet adapter MAC address.
1904 kmemleak= [KNL] Boot-time kmemleak enable/disable
1905 Valid arguments: on, off
1907 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y,
1910 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs.
1911 Default is 0 (don't ignore, but inject #GP)
1913 kvm.enable_vmware_backdoor=[KVM] Support VMware backdoor PV interface.
1914 Default is false (don't support).
1916 kvm.mmu_audit= [KVM] This is a R/W parameter which allows audit
1920 kvm-amd.nested= [KVM,AMD] Allow nested virtualization in KVM/SVM.
1921 Default is 1 (enabled)
1923 kvm-amd.npt= [KVM,AMD] Disable nested paging (virtualized MMU)
1925 Default is 1 (enabled) if in 64-bit or 32-bit PAE mode.
1927 kvm-arm.vgic_v3_group0_trap=
1928 [KVM,ARM] Trap guest accesses to GICv3 group-0
1931 kvm-arm.vgic_v3_group1_trap=
1932 [KVM,ARM] Trap guest accesses to GICv3 group-1
1935 kvm-arm.vgic_v3_common_trap=
1936 [KVM,ARM] Trap guest accesses to GICv3 common
1939 kvm-arm.vgic_v4_enable=
1940 [KVM,ARM] Allow use of GICv4 for direct injection of
1943 kvm-intel.ept= [KVM,Intel] Disable extended page tables
1944 (virtualized MMU) support on capable Intel chips.
1945 Default is 1 (enabled)
1947 kvm-intel.emulate_invalid_guest_state=
1948 [KVM,Intel] Enable emulation of invalid guest states
1949 Default is 0 (disabled)
1951 kvm-intel.flexpriority=
1952 [KVM,Intel] Disable FlexPriority feature (TPR shadow).
1953 Default is 1 (enabled)
1956 [KVM,Intel] Enable VMX nesting (nVMX).
1957 Default is 0 (disabled)
1959 kvm-intel.unrestricted_guest=
1960 [KVM,Intel] Disable unrestricted guest feature
1961 (virtualized real and unpaged mode) on capable
1962 Intel chips. Default is 1 (enabled)
1964 kvm-intel.vpid= [KVM,Intel] Disable Virtual Processor Identification
1965 feature (tagged TLBs) on capable Intel chips.
1966 Default is 1 (enabled)
1972 lapic [X86-32,APIC] Enable the local APIC even if BIOS
1975 lapic= [x86,APIC] "notscdeadline" Do not use TSC deadline
1976 value for LAPIC timer one-shot implementation. Default
1977 back to the programmable timer unit in the LAPIC.
1979 lapic_timer_c2_ok [X86,APIC] trust the local apic timer
1982 libata.dma= [LIBATA] DMA control
1983 libata.dma=0 Disable all PATA and SATA DMA
1984 libata.dma=1 PATA and SATA Disk DMA only
1985 libata.dma=2 ATAPI (CDROM) DMA only
1986 libata.dma=4 Compact Flash DMA only
1987 Combinations also work, so libata.dma=3 enables DMA
1988 for disks and CDROMs, but not CFs.
1990 libata.ignore_hpa= [LIBATA] Ignore HPA limit
1991 libata.ignore_hpa=0 keep BIOS limits (default)
1992 libata.ignore_hpa=1 ignore limits, using full disk
1994 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume
1998 libata.force= [LIBATA] Force configurations. The format is comma
1999 separated list of "[ID:]VAL" where ID is
2000 PORT[.DEVICE]. PORT and DEVICE are decimal numbers
2001 matching port, link or device. Basically, it matches
2002 the ATA ID string printed on console by libata. If
2003 the whole ID part is omitted, the last PORT and DEVICE
2004 values are used. If ID hasn't been specified yet, the
2005 configuration applies to all ports, links and devices.
2007 If only DEVICE is omitted, the parameter applies to
2008 the port and all links and devices behind it. DEVICE
2009 number of 0 either selects the first device or the
2010 first fan-out link behind PMP device. It does not
2011 select the host link. DEVICE number of 15 selects the
2012 host link and device attached to it.
2014 The VAL specifies the configuration to force. As long
2015 as there's no ambiguity shortcut notation is allowed.
2016 For example, both 1.5 and 1.5G would work for 1.5Gbps.
2017 The following configurations can be forced.
2019 * Cable type: 40c, 80c, short40c, unk, ign or sata.
2020 Any ID with matching PORT is used.
2022 * SATA link speed limit: 1.5Gbps or 3.0Gbps.
2024 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7].
2025 udma[/][16,25,33,44,66,100,133] notation is also
2028 * [no]ncq: Turn on or off NCQ.
2030 * [no]ncqtrim: Turn off queued DSM TRIM.
2032 * nohrst, nosrst, norst: suppress hard, soft
2035 * rstonce: only attempt one reset during
2036 hot-unplug link recovery
2038 * dump_id: dump IDENTIFY data.
2040 * atapi_dmadir: Enable ATAPI DMADIR bridge support
2042 * disable: Disable this device.
2044 If there are multiple matching configurations changing
2045 the same attribute, the last one is used.
2047 memblock=debug [KNL] Enable memblock debug messages.
2049 load_ramdisk= [RAM] List of ramdisks to load from floppy
2050 See Documentation/blockdev/ramdisk.txt.
2052 lockd.nlm_grace_period=P [NFS] Assign grace period.
2055 lockd.nlm_tcpport=N [NFS] Assign TCP port.
2058 lockd.nlm_timeout=T [NFS] Assign timeout value.
2061 lockd.nlm_udpport=M [NFS] Assign UDP port.
2064 locktorture.nreaders_stress= [KNL]
2065 Set the number of locking read-acquisition kthreads.
2066 Defaults to being automatically set based on the
2067 number of online CPUs.
2069 locktorture.nwriters_stress= [KNL]
2070 Set the number of locking write-acquisition kthreads.
2072 locktorture.onoff_holdoff= [KNL]
2073 Set time (s) after boot for CPU-hotplug testing.
2075 locktorture.onoff_interval= [KNL]
2076 Set time (s) between CPU-hotplug operations, or
2077 zero to disable CPU-hotplug testing.
2079 locktorture.shuffle_interval= [KNL]
2080 Set task-shuffle interval (jiffies). Shuffling
2081 tasks allows some CPUs to go into dyntick-idle
2082 mode during the locktorture test.
2084 locktorture.shutdown_secs= [KNL]
2085 Set time (s) after boot system shutdown. This
2086 is useful for hands-off automated testing.
2088 locktorture.stat_interval= [KNL]
2089 Time (s) between statistics printk()s.
2091 locktorture.stutter= [KNL]
2092 Time (s) to stutter testing, for example,
2093 specifying five seconds causes the test to run for
2094 five seconds, wait for five seconds, and so on.
2095 This tests the locking primitive's ability to
2096 transition abruptly to and from idle.
2098 locktorture.torture_type= [KNL]
2099 Specify the locking implementation to test.
2101 locktorture.verbose= [KNL]
2102 Enable additional printk() statements.
2104 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver
2107 loglevel= All Kernel Messages with a loglevel smaller than the
2108 console loglevel will be printed to the console. It can
2109 also be changed with klogd or other programs. The
2110 loglevels are defined as follows:
2112 0 (KERN_EMERG) system is unusable
2113 1 (KERN_ALERT) action must be taken immediately
2114 2 (KERN_CRIT) critical conditions
2115 3 (KERN_ERR) error conditions
2116 4 (KERN_WARNING) warning conditions
2117 5 (KERN_NOTICE) normal but significant condition
2118 6 (KERN_INFO) informational
2119 7 (KERN_DEBUG) debug-level messages
2121 log_buf_len=n[KMG] Sets the size of the printk ring buffer,
2122 in bytes. n must be a power of two and greater
2123 than the minimal size. The minimal size is defined
2124 by LOG_BUF_SHIFT kernel config parameter. There is
2125 also CONFIG_LOG_CPU_MAX_BUF_SHIFT config parameter
2126 that allows to increase the default size depending on
2127 the number of CPUs. See init/Kconfig for more details.
2129 logo.nologo [FB] Disables display of the built-in Linux logo.
2130 This may be used to provide more screen space for
2131 kernel log messages and is useful when debugging
2132 kernel boot problems.
2134 lp=0 [LP] Specify parallel ports to use, e.g,
2135 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses
2136 lp=reset first parallel port). 'lp=0' disables the
2137 lp=auto printer driver. 'lp=reset' (which can be
2138 specified in addition to the ports) causes
2139 attached printers to be reset. Using
2140 lp=port1,port2,... specifies the parallel ports
2141 to associate lp devices with, starting with
2142 lp0. A port specification may be 'none' to skip
2143 that lp device, or a parport name such as
2144 'parport0'. Specifying 'lp=auto' instead of a
2145 port specification list means that device IDs
2146 from each port should be examined, to see if
2147 an IEEE 1284-compliant printer is attached; if
2148 so, the driver will manage that printer.
2149 See also header of drivers/char/lp.c.
2152 Sets loops_per_jiffy to given constant, thus avoiding
2153 time-consuming boot-time autodetection (up to 250 ms per
2154 CPU). 0 enables autodetection (default). To determine
2155 the correct value for your kernel, boot with normal
2156 autodetection and see what value is printed. Note that
2157 on SMP systems the preset will be applied to all CPUs,
2158 which is likely to cause problems if your CPUs need
2159 significantly divergent settings. An incorrect value
2160 will cause delays in the kernel to be wrong, leading to
2161 unpredictable I/O errors and other breakage. Although
2162 unlikely, in the extreme case this might damage your
2166 Format: <io>,<irq>,<dma>
2168 machvec= [IA-64] Force the use of a particular machine-vector
2169 (machvec) in a generic kernel.
2170 Example: machvec=hpzx1_swiotlb
2172 machtype= [Loongson] Share the same kernel image file between different
2174 Example: machtype=lemote-yeeloong-2f-7inch
2176 max_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory greater
2177 than or equal to this physical address is ignored.
2179 maxcpus= [SMP] Maximum number of processors that an SMP kernel
2180 will bring up during bootup. maxcpus=n : n >= 0 limits
2181 the kernel to bring up 'n' processors. Surely after
2182 bootup you can bring up the other plugged cpu by executing
2183 "echo 1 > /sys/devices/system/cpu/cpuX/online". So maxcpus
2184 only takes effect during system bootup.
2185 While n=0 is a special case, it is equivalent to "nosmp",
2186 which also disables the IO APIC.
2188 max_loop= [LOOP] The number of loop block devices that get
2189 (loop.max_loop) unconditionally pre-created at init time. The default
2190 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead
2191 of statically allocating a predefined number, loop
2192 devices can be requested on-demand with the
2193 /dev/loop-control interface.
2195 mce [X86-32] Machine Check Exception
2197 mce=option [X86-64] See Documentation/x86/x86_64/boot-options.txt
2199 md= [HW] RAID subsystems devices and level
2200 See Documentation/admin-guide/md.rst.
2203 Format: <first>,<last>
2204 Specifies range of consoles to be captured by the MDA.
2206 mem=nn[KMG] [KNL,BOOT] Force usage of a specific amount of memory
2207 Amount of memory to be used when the kernel is not able
2208 to see the whole system memory or for test.
2209 [X86] Work as limiting max address. Use together
2210 with memmap= to avoid physical address space collisions.
2211 Without memmap= PCI devices could be placed at addresses
2212 belonging to unused RAM.
2214 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel
2218 [KNL,SH] Allow user to override the default size for
2219 per-device physically contiguous DMA buffers.
2221 memhp_default_state=online/offline
2222 [KNL] Set the initial state for the memory hotplug
2223 onlining policy. If not specified, the default value is
2224 set according to the
2225 CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE kernel config
2227 See Documentation/memory-hotplug.txt.
2229 memmap=exactmap [KNL,X86] Enable setting of an exact
2230 E820 memory map, as specified by the user.
2231 Such memmap=exactmap lines can be constructed based on
2232 BIOS output or other requirements. See the memmap=nn@ss
2235 memmap=nn[KMG]@ss[KMG]
2236 [KNL] Force usage of a specific region of memory.
2237 Region of memory to be used is from ss to ss+nn.
2238 If @ss[KMG] is omitted, it is equivalent to mem=nn[KMG],
2239 which limits max address to nn[KMG].
2240 Multiple different regions can be specified,
2243 memmap=100M@2G,100M#3G,1G!1024G
2245 memmap=nn[KMG]#ss[KMG]
2246 [KNL,ACPI] Mark specific memory as ACPI data.
2247 Region of memory to be marked is from ss to ss+nn.
2249 memmap=nn[KMG]$ss[KMG]
2250 [KNL,ACPI] Mark specific memory as reserved.
2251 Region of memory to be reserved is from ss to ss+nn.
2252 Example: Exclude memory from 0x18690000-0x1869ffff
2253 memmap=64K$0x18690000
2255 memmap=0x10000$0x18690000
2256 Some bootloaders may need an escape character before '$',
2257 like Grub2, otherwise '$' and the following number
2260 memmap=nn[KMG]!ss[KMG]
2261 [KNL,X86] Mark specific memory as protected.
2262 Region of memory to be used, from ss to ss+nn.
2263 The memory region may be marked as e820 type 12 (0xc)
2264 and is NVDIMM or ADR memory.
2266 memmap=<size>%<offset>-<oldtype>+<newtype>
2267 [KNL,ACPI] Convert memory within the specified region
2268 from <oldtype> to <newtype>. If "-<oldtype>" is left
2269 out, the whole region will be marked as <newtype>,
2270 even if previously unavailable. If "+<newtype>" is left
2271 out, matching memory will be removed. Types are
2272 specified as e820 types, e.g., 1 = RAM, 2 = reserved,
2273 3 = ACPI, 12 = PRAM.
2275 memory_corruption_check=0/1 [X86]
2276 Some BIOSes seem to corrupt the first 64k of
2277 memory when doing things like suspend/resume.
2278 Setting this option will scan the memory
2279 looking for corruption. Enabling this will
2280 both detect corruption and prevent the kernel
2281 from using the memory being corrupted.
2282 However, its intended as a diagnostic tool; if
2283 repeatable BIOS-originated corruption always
2284 affects the same memory, you can use memmap=
2285 to prevent the kernel from using that memory.
2287 memory_corruption_check_size=size [X86]
2288 By default it checks for corruption in the low
2289 64k, making this memory unavailable for normal
2290 use. Use this parameter to scan for
2291 corruption in more or less memory.
2293 memory_corruption_check_period=seconds [X86]
2294 By default it checks for corruption every 60
2295 seconds. Use this parameter to check at some
2296 other rate. 0 disables periodic checking.
2298 memtest= [KNL,X86,ARM] Enable memtest
2300 default : 0 <disable>
2301 Specifies the number of memtest passes to be
2302 performed. Each pass selects another test
2303 pattern from a given set of patterns. Memtest
2304 fills the memory with this pattern, validates
2305 memory contents and reserves bad memory
2306 regions that are detected.
2308 mem_encrypt= [X86-64] AMD Secure Memory Encryption (SME) control
2309 Valid arguments: on, off
2310 Default (depends on kernel configuration option):
2311 on (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=y)
2312 off (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=n)
2313 mem_encrypt=on: Activate SME
2314 mem_encrypt=off: Do not activate SME
2316 Refer to Documentation/x86/amd-memory-encryption.txt
2317 for details on when memory encryption can be activated.
2319 mem_sleep_default= [SUSPEND] Default system suspend mode:
2320 s2idle - Suspend-To-Idle
2321 shallow - Power-On Suspend or equivalent (if supported)
2322 deep - Suspend-To-RAM or equivalent (if supported)
2323 See Documentation/admin-guide/pm/sleep-states.rst.
2325 meye.*= [HW] Set MotionEye Camera parameters
2326 See Documentation/media/v4l-drivers/meye.rst.
2328 mfgpt_irq= [IA-32] Specify the IRQ to use for the
2329 Multi-Function General Purpose Timers on AMD Geode
2332 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when
2333 the BIOS has incorrectly applied a workaround. TinyBIOS
2334 version 0.98 is known to be affected, 0.99 fixes the
2335 problem by letting the user disable the workaround.
2339 min_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory below this
2340 physical address is ignored.
2342 mini2440= [ARM,HW,KNL]
2343 Format:[0..2][b][c][t]
2345 MINI2440 configuration specification:
2346 0 - The attached screen is the 3.5" TFT
2347 1 - The attached screen is the 7" TFT
2348 2 - The VGA Shield is attached (1024x768)
2349 Leaving out the screen size parameter will not load
2350 the TFT driver, and the framebuffer will be left
2352 b - Enable backlight. The TFT backlight pin will be
2353 linked to the kernel VESA blanking code and a GPIO
2354 LED. This parameter is not necessary when using the
2356 c - Enable the s3c camera interface.
2357 t - Reserved for enabling touchscreen support. The
2358 touchscreen support is not enabled in the mainstream
2359 kernel as of 2.6.30, a preliminary port can be found
2360 in the "bleeding edge" mini2440 support kernel at
2361 http://repo.or.cz/w/linux-2.6/mini2440.git
2364 [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this
2365 parameter allows control of the logging verbosity for
2366 the additional memory initialisation checks. A value
2367 of 0 disables mminit logging and a level of 4 will
2368 log everything. Information is printed at KERN_DEBUG
2369 so loglevel=8 may also need to be specified.
2372 [KNL] When CONFIG_MODULE_SIG is set, this means that
2373 modules without (valid) signatures will fail to load.
2374 Note that if CONFIG_MODULE_SIG_FORCE is set, that
2375 is always true, so this option does nothing.
2377 module_blacklist= [KNL] Do not load a comma-separated list of
2378 modules. Useful for debugging problem modules.
2381 [MOUSE] Maximum time between finger touching and
2382 leaving touchpad surface for touch to be considered
2383 a tap and be reported as a left button click (for
2384 touchpads working in absolute mode only).
2386 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices
2387 reporting absolute coordinates, such as tablets
2388 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices
2389 reporting absolute coordinates, such as tablets
2391 movablecore= [KNL,X86,IA-64,PPC]
2392 Format: nn[KMGTPE] | nn%
2393 This parameter is the complement to kernelcore=, it
2394 specifies the amount of memory used for migratable
2395 allocations. If both kernelcore and movablecore is
2396 specified, then kernelcore will be at *least* the
2397 specified value but may be more. If movablecore on its
2398 own is specified, the administrator must be careful
2399 that the amount of memory usable for all allocations
2402 movable_node [KNL] Boot-time switch to make hotplugable memory
2403 NUMA nodes to be movable. This means that the memory
2404 of such nodes will be usable only for movable
2405 allocations which rules out almost all kernel
2406 allocations. Use with caution!
2408 MTD_Partition= [MTD]
2409 Format: <name>,<region-number>,<size>,<offset>
2411 MTD_Region= [MTD] Format:
2412 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>]
2415 See drivers/mtd/cmdlinepart.c.
2417 multitce=off [PPC] This parameter disables the use of the pSeries
2418 firmware feature for updating multiple TCE entries
2421 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration
2423 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock]
2425 boundary - index of last SLC block on Flex-OneNAND.
2426 The remaining blocks are configured as MLC blocks.
2427 lock - Configure if Flex-OneNAND boundary should be locked.
2428 Once locked, the boundary cannot be changed.
2429 1 indicates lock status, 0 indicates unlock status.
2432 ARM/S3C2412 JIVE boot control
2434 See arch/arm/mach-s3c2412/mach-jive.c
2436 mtouchusb.raw_coordinates=
2437 [HW] Make the MicroTouch USB driver use raw coordinates
2438 ('y', default) or cooked coordinates ('n')
2440 mtrr_chunk_size=nn[KMG] [X86]
2441 used for mtrr cleanup. It is largest continuous chunk
2442 that could hold holes aka. UC entries.
2444 mtrr_gran_size=nn[KMG] [X86]
2445 Used for mtrr cleanup. It is granularity of mtrr block.
2447 Large value could prevent small alignment from
2450 mtrr_spare_reg_nr=n [X86]
2452 Range: 0,7 : spare reg number
2454 Used for mtrr cleanup. It is spare mtrr entries number.
2455 Set to 2 or more if your graphical card needs more.
2457 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card
2459 netdev= [NET] Network devices parameters
2460 Format: <irq>,<io>,<mem_start>,<mem_end>,<name>
2461 Note that mem_start is often overloaded to mean
2462 something different and driver-specific.
2463 This usage is only documented in each driver source
2467 [NETFILTER] Enable connection tracking flow accounting
2468 0 to disable accounting
2469 1 to enable accounting
2472 nfsaddrs= [NFS] Deprecated. Use ip= instead.
2473 See Documentation/filesystems/nfs/nfsroot.txt.
2475 nfsroot= [NFS] nfs root filesystem for disk-less boxes.
2476 See Documentation/filesystems/nfs/nfsroot.txt.
2478 nfsrootdebug [NFS] enable nfsroot debugging messages.
2479 See Documentation/filesystems/nfs/nfsroot.txt.
2481 nfs.callback_nr_threads=
2482 [NFSv4] set the total number of threads that the
2483 NFS client will assign to service NFSv4 callback
2486 nfs.callback_tcpport=
2487 [NFS] set the TCP port on which the NFSv4 callback
2488 channel should listen.
2491 [NFS] sets the pathname to the program which is used
2492 to update the NFS client cache entries.
2494 nfs.cache_getent_timeout=
2495 [NFS] sets the timeout after which an attempt to
2496 update a cache entry is deemed to have failed.
2498 nfs.idmap_cache_timeout=
2499 [NFS] set the maximum lifetime for idmapper cache
2503 [NFS] enable 64-bit inode numbers.
2504 If zero, the NFS client will fake up a 32-bit inode
2505 number for the readdir() and stat() syscalls instead
2506 of returning the full 64-bit number.
2507 The default is to return 64-bit inode numbers.
2509 nfs.max_session_cb_slots=
2510 [NFSv4.1] Sets the maximum number of session
2511 slots the client will assign to the callback
2512 channel. This determines the maximum number of
2513 callbacks the client will process in parallel for
2514 a particular server.
2516 nfs.max_session_slots=
2517 [NFSv4.1] Sets the maximum number of session slots
2518 the client will attempt to negotiate with the server.
2519 This limits the number of simultaneous RPC requests
2520 that the client can send to the NFSv4.1 server.
2521 Note that there is little point in setting this
2522 value higher than the max_tcp_slot_table_limit.
2524 nfs.nfs4_disable_idmapping=
2525 [NFSv4] When set to the default of '1', this option
2526 ensures that both the RPC level authentication
2527 scheme and the NFS level operations agree to use
2528 numeric uids/gids if the mount is using the
2529 'sec=sys' security flavour. In effect it is
2530 disabling idmapping, which can make migration from
2531 legacy NFSv2/v3 systems to NFSv4 easier.
2532 Servers that do not support this mode of operation
2533 will be autodetected by the client, and it will fall
2534 back to using the idmapper.
2535 To turn off this behaviour, set the value to '0'.
2537 [NFS4] Specify an additional fixed unique ident-
2538 ification string that NFSv4 clients can insert into
2539 their nfs_client_id4 string. This is typically a
2540 UUID that is generated at system install time.
2542 nfs.send_implementation_id =
2543 [NFSv4.1] Send client implementation identification
2544 information in exchange_id requests.
2545 If zero, no implementation identification information
2547 The default is to send the implementation identification
2550 nfs.recover_lost_locks =
2551 [NFSv4] Attempt to recover locks that were lost due
2552 to a lease timeout on the server. Please note that
2553 doing this risks data corruption, since there are
2554 no guarantees that the file will remain unchanged
2555 after the locks are lost.
2556 If you want to enable the kernel legacy behaviour of
2557 attempting to recover these locks, then set this
2559 The default parameter value of '0' causes the kernel
2560 not to attempt recovery of lost locks.
2562 nfs4.layoutstats_timer =
2563 [NFSv4.2] Change the rate at which the kernel sends
2564 layoutstats to the pNFS metadata server.
2566 Setting this to value to 0 causes the kernel to use
2567 whatever value is the default set by the layout
2568 driver. A non-zero value sets the minimum interval
2569 in seconds between layoutstats transmissions.
2571 nfsd.nfs4_disable_idmapping=
2572 [NFSv4] When set to the default of '1', the NFSv4
2573 server will return only numeric uids and gids to
2574 clients using auth_sys, and will accept numeric uids
2575 and gids from such clients. This is intended to ease
2576 migration from NFSv2/v3.
2578 nmi_debug= [KNL,SH] Specify one or more actions to take
2579 when a NMI is triggered.
2580 Format: [state][,regs][,debounce][,die]
2582 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels
2583 Format: [panic,][nopanic,][num]
2585 0 - turn hardlockup detector in nmi_watchdog off
2586 1 - turn hardlockup detector in nmi_watchdog on
2587 When panic is specified, panic when an NMI watchdog
2588 timeout occurs (or 'nopanic' to override the opposite
2589 default). To disable both hard and soft lockup detectors,
2590 please see 'nowatchdog'.
2591 This is useful when you use a panic=... timeout and
2592 need the box quickly up again.
2594 These settings can be accessed at runtime via
2595 the nmi_watchdog and hardlockup_panic sysctls.
2597 netpoll.carrier_timeout=
2598 [NET] Specifies amount of time (in seconds) that
2599 netpoll should wait for a carrier. By default netpoll
2602 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths
2603 emulation library even if a 387 maths coprocessor
2606 no5lvl [X86-64] Disable 5-level paging mode. Forces
2607 kernel to use 4-level paging instead.
2610 [HW] Never suspend the console
2611 Disable suspending of consoles during suspend and
2612 hibernate operations. Once disabled, debugging
2613 messages can reach various consoles while the rest
2614 of the system is being put to sleep (ie, while
2615 debugging driver suspend/resume hooks). This may
2616 not work reliably with all consoles, but is known
2617 to work with serial and VGA consoles.
2618 To facilitate more flexible debugging, we also add
2619 console_suspend, a printk module parameter to control
2620 it. Users could use console_suspend (usually
2621 /sys/module/printk/parameters/console_suspend) to
2622 turn on/off it dynamically.
2624 noaliencache [MM, NUMA, SLAB] Disables the allocation of alien
2625 caches in the slab allocator. Saves per-node memory,
2626 but will impact performance.
2630 noaltinstr [S390] Disables alternative instructions patching
2631 (CPU alternatives feature).
2633 noapic [SMP,APIC] Tells the kernel to not make use of any
2634 IOAPICs that may be present in the system.
2636 noautogroup Disable scheduler automatic task group creation.
2638 nobats [PPC] Do not use BATs for mapping kernel lowmem
2639 on "Classic" PPC cores.
2643 noclflush [BUGS=X86] Don't use the CLFLUSH instruction
2645 nodelayacct [KNL] Disable per-task delay accounting
2647 nodsp [SH] Disable hardware DSP at boot time.
2649 noefi Disable EFI runtime services support.
2654 On X86-32 available only on PAE configured kernels.
2655 noexec=on: enable non-executable mappings (default)
2656 noexec=off: disable non-executable mappings
2659 Disable SMAP (Supervisor Mode Access Prevention)
2660 even if it is supported by processor.
2663 Disable SMEP (Supervisor Mode Execution Prevention)
2664 even if it is supported by processor.
2667 This affects only 32-bit executables.
2668 noexec32=on: enable non-executable mappings (default)
2669 read doesn't imply executable mappings
2670 noexec32=off: disable non-executable mappings
2671 read implies executable mappings
2673 nofpu [MIPS,SH] Disable hardware FPU at boot time.
2675 nofxsr [BUGS=X86-32] Disables x86 floating point extended
2676 register save and restore. The kernel will only save
2677 legacy floating-point registers on task switch.
2679 nohugeiomap [KNL,x86] Disable kernel huge I/O mappings.
2681 nosmt [KNL,S390] Disable symmetric multithreading (SMT).
2682 Equivalent to smt=1.
2684 nospectre_v2 [X86] Disable all mitigations for the Spectre variant 2
2685 (indirect branch prediction) vulnerability. System may
2686 allow data leaks with this option, which is equivalent
2689 nospec_store_bypass_disable
2690 [HW] Disable all mitigations for the Speculative Store Bypass vulnerability
2692 noxsave [BUGS=X86] Disables x86 extended register state save
2693 and restore using xsave. The kernel will fallback to
2694 enabling legacy floating-point and sse state.
2696 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended
2697 register states. The kernel will fall back to use
2698 xsave to save the states. By using this parameter,
2699 performance of saving the states is degraded because
2700 xsave doesn't support modified optimization while
2701 xsaveopt supports it on xsaveopt enabled systems.
2703 noxsaves [X86] Disables xsaves and xrstors used in saving and
2704 restoring x86 extended register state in compacted
2705 form of xsave area. The kernel will fall back to use
2706 xsaveopt and xrstor to save and restore the states
2707 in standard form of xsave area. By using this
2708 parameter, xsave area per process might occupy more
2709 memory on xsaves enabled systems.
2711 nohlt [BUGS=ARM,SH] Tells the kernel that the sleep(SH) or
2712 wfi(ARM) instruction doesn't work correctly and not to
2713 use it. This is also useful when using JTAG debugger.
2715 no_file_caps Tells the kernel not to honor file capabilities. The
2716 only way then for a file to be executed with privilege
2717 is to be setuid root or executed by root.
2719 nohalt [IA-64] Tells the kernel not to use the power saving
2720 function PAL_HALT_LIGHT when idle. This increases
2721 power-consumption. On the positive side, it reduces
2722 interrupt wake-up latency, which may improve performance
2723 in certain environments such as networked servers or
2726 nohibernate [HIBERNATION] Disable hibernation and resume.
2728 nohz= [KNL] Boottime enable/disable dynamic ticks
2729 Valid arguments: on, off
2732 nohz_full= [KNL,BOOT,SMP,ISOL]
2733 The argument is a cpu list, as described above.
2734 In kernels built with CONFIG_NO_HZ_FULL=y, set
2735 the specified list of CPUs whose tick will be stopped
2736 whenever possible. The boot CPU will be forced outside
2737 the range to maintain the timekeeping. Any CPUs
2738 in this list will have their RCU callbacks offloaded,
2739 just as if they had also been called out in the
2740 rcu_nocbs= boot parameter.
2742 noiotrap [SH] Disables trapped I/O port accesses.
2744 noirqdebug [X86-32] Disables the code which attempts to detect and
2745 disable unhandled interrupt sources.
2747 no_timer_check [X86,APIC] Disables the code which tests for
2748 broken timer IRQ sources.
2750 noisapnp [ISAPNP] Disables ISA PnP code.
2752 noinitrd [RAM] Tells the kernel not to load any configured
2755 nointremap [X86-64, Intel-IOMMU] Do not enable interrupt
2757 [Deprecated - use intremap=off]
2761 noinvpcid [X86] Disable the INVPCID cpu feature.
2763 nojitter [IA-64] Disables jitter checking for ITC timers.
2765 no-kvmclock [X86,KVM] Disable paravirtualized KVM clock driver
2767 no-kvmapf [X86,KVM] Disable paravirtualized asynchronous page
2771 [X86,PV_OPS] Disable paravirtualized VMware scheduler
2772 clock and use the default one.
2774 no-steal-acc [X86,KVM] Disable paravirtualized steal time accounting.
2775 steal time is computed, but won't influence scheduler
2778 nolapic [X86-32,APIC] Do not enable or use the local APIC.
2780 nolapic_timer [X86-32,APIC] Do not use the local APIC timer.
2782 noltlbs [PPC] Do not use large page/tlb entries for kernel
2783 lowmem mapping on PPC40x and PPC8xx
2785 nomca [IA-64] Disable machine check abort handling
2787 nomce [X86-32] Disable Machine Check Exception
2789 nomfgpt [X86-32] Disable Multi-Function General Purpose
2790 Timer usage (for AMD Geode machines).
2792 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to
2793 shutdown the other cpus. Instead use the REBOOT_VECTOR
2796 nomodule Disable module load
2798 nopat [X86] Disable PAT (page attribute table extension of
2799 pagetables) support.
2801 nopcid [X86-64] Disable the PCID cpu feature.
2803 norandmaps Don't use address space randomization. Equivalent to
2804 echo 0 > /proc/sys/kernel/randomize_va_space
2806 noreplace-smp [X86-32,SMP] Don't replace SMP instructions
2807 with UP alternatives
2809 nordrand [X86] Disable kernel use of the RDRAND and
2810 RDSEED instructions even if they are supported
2811 by the processor. RDRAND and RDSEED are still
2812 available to user space applications.
2814 noresume [SWSUSP] Disables resume and restores original swap
2817 no-scroll [VGA] Disables scrollback.
2818 This is required for the Braillex ib80-piezo Braille
2819 reader made by F.H. Papenmeier (Germany).
2823 nosep [BUGS=X86-32] Disables x86 SYSENTER/SYSEXIT support.
2825 nosmp [SMP] Tells an SMP kernel to act as a UP kernel,
2826 and disable the IO APIC. legacy for "maxcpus=0".
2828 nosoftlockup [KNL] Disable the soft-lockup detector.
2830 nosync [HW,M68K] Disables sync negotiation for all devices.
2832 notsc [BUGS=X86-32] Disable Time Stamp Counter
2834 nowatchdog [KNL] Disable both lockup detectors, i.e.
2835 soft-lockup and NMI watchdog (hard-lockup).
2839 nox2apic [X86-64,APIC] Do not enable x2APIC mode.
2841 cpu0_hotplug [X86] Turn on CPU0 hotplug feature when
2842 CONFIG_BOOTPARAM_HOTPLUG_CPU0 is off.
2843 Some features depend on CPU0. Known dependencies are:
2844 1. Resume from suspend/hibernate depends on CPU0.
2845 Suspend/hibernate will fail if CPU0 is offline and you
2846 need to online CPU0 before suspend/hibernate.
2847 2. PIC interrupts also depend on CPU0. CPU0 can't be
2848 removed if a PIC interrupt is detected.
2849 It's said poweroff/reboot may depend on CPU0 on some
2850 machines although I haven't seen such issues so far
2851 after CPU0 is offline on a few tested machines.
2852 If the dependencies are under your control, you can
2853 turn on cpu0_hotplug.
2855 nps_mtm_hs_ctr= [KNL,ARC]
2856 This parameter sets the maximum duration, in
2857 cycles, each HW thread of the CTOP can run
2858 without interruptions, before HW switches it.
2859 The actual maximum duration is 16 times this
2861 Format: integer between 1 and 255
2864 nptcg= [IA-64] Override max number of concurrent global TLB
2865 purges which is reported from either PAL_VM_SUMMARY or
2868 nr_cpus= [SMP] Maximum number of processors that an SMP kernel
2869 could support. nr_cpus=n : n >= 1 limits the kernel to
2870 support 'n' processors. It could be larger than the
2871 number of already plugged CPU during bootup, later in
2872 runtime you can physically add extra cpu until it reaches
2873 n. So during boot up some boot time memory for per-cpu
2874 variables need be pre-allocated for later physical cpu
2877 nr_uarts= [SERIAL] maximum number of UARTs to be registered.
2879 numa_balancing= [KNL,X86] Enable or disable automatic NUMA balancing.
2880 Allowed values are enable and disable
2882 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA.
2883 'node', 'default' can be specified
2884 This can be set from sysctl after boot.
2885 See Documentation/sysctl/vm.txt for details.
2887 ohci1394_dma=early [HW] enable debugging via the ohci1394 driver.
2888 See Documentation/debugging-via-ohci1394.txt for more
2891 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands
2892 Rather than timing out after 20 ms if an EC
2893 command is not properly ACKed, override the length
2894 of the timeout. We have interrupts disabled while
2895 waiting for the ACK, so if this is set too high
2896 interrupts *may* be lost!
2898 omap_mux= [OMAP] Override bootloader pin multiplexing.
2899 Format: <mux_mode0.mode_name=value>...
2900 For example, to override I2C bus2:
2901 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100
2903 oprofile.timer= [HW]
2904 Use timer interrupt instead of performance counters
2906 oprofile.cpu_type= Force an oprofile cpu type
2907 This might be useful if you have an older oprofile
2908 userland or if you want common events.
2909 Format: { arch_perfmon }
2910 arch_perfmon: [X86] Force use of architectural
2911 perfmon on Intel CPUs instead of the
2912 CPU specific event set.
2913 timer: [X86] Force use of architectural NMI
2914 timer mode (see also oprofile.timer
2915 for generic hr timer mode)
2917 oops=panic Always panic on oopses. Default is to just kill the
2918 process, but there is a small probability of
2919 deadlocking the machine.
2920 This will also cause panics on machine check exceptions.
2921 Useful together with panic=30 to trigger a reboot.
2924 See Documentation/sound/oss/oss-parameters.txt
2926 page_owner= [KNL] Boot-time page_owner enabling option.
2927 Storage of the information about who allocated
2928 each page is disabled in default. With this switch,
2930 on: enable the feature
2932 page_poison= [KNL] Boot-time parameter changing the state of
2933 poisoning on the buddy allocator.
2934 off: turn off poisoning
2935 on: turn on poisoning
2937 panic= [KNL] Kernel behaviour on panic: delay <timeout>
2938 timeout > 0: seconds before rebooting
2939 timeout = 0: wait forever
2940 timeout < 0: reboot immediately
2943 panic_on_warn panic() instead of WARN(). Useful to cause kdump
2946 crash_kexec_post_notifiers
2947 Run kdump after running panic-notifiers and dumping
2948 kmsg. This only for the users who doubt kdump always
2949 succeeds in any situation.
2950 Note that this also increases risks of kdump failure,
2951 because some panic notifiers can make the crashed
2952 kernel more unstable.
2954 parkbd.port= [HW] Parallel port number the keyboard adapter is
2955 connected to, default is 0.
2957 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation,
2958 0 for XT, 1 for AT (default is AT).
2961 parport= [HW,PPT] Specify parallel ports. 0 disables.
2962 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] }
2963 Use 'auto' to force the driver to use any
2964 IRQ/DMA settings detected (the default is to
2965 ignore detected IRQ/DMA settings because of
2966 possible conflicts). You can specify the base
2967 address, IRQ, and DMA settings; IRQ and DMA
2968 should be numbers, or 'auto' (for using detected
2969 settings on that particular port), or 'nofifo'
2970 (to avoid using a FIFO even if it is detected).
2971 Parallel ports are assigned in the order they
2972 are specified on the command line, starting
2975 parport_init_mode= [HW,PPT]
2976 Configure VIA parallel port to operate in
2977 a specific mode. This is necessary on Pegasos
2978 computer where firmware has no options for setting
2979 up parallel port mode and sets it to spp.
2980 Currently this function knows 686a and 8231 chips.
2981 Format: [spp|ps2|epp|ecp|ecpepp]
2984 Halt all CPUs after the first oops has been printed for
2985 the specified number of seconds. This is to be used if
2986 your oopses keep scrolling off the screen.
2991 See header of drivers/block/paride/pcd.c.
2992 See also Documentation/blockdev/paride.txt.
2994 pci=option[,option...] [PCI] various PCI subsystem options:
2995 earlydump [X86] dump PCI config space before the kernel
2997 off [X86] don't probe for the PCI bus
2998 bios [X86-32] force use of PCI BIOS, don't access
2999 the hardware directly. Use this if your machine
3000 has a non-standard PCI host bridge.
3001 nobios [X86-32] disallow use of PCI BIOS, only direct
3002 hardware access methods are allowed. Use this
3003 if you experience crashes upon bootup and you
3004 suspect they are caused by the BIOS.
3005 conf1 [X86] Force use of PCI Configuration Access
3006 Mechanism 1 (config address in IO port 0xCF8,
3007 data in IO port 0xCFC, both 32-bit).
3008 conf2 [X86] Force use of PCI Configuration Access
3009 Mechanism 2 (IO port 0xCF8 is an 8-bit port for
3010 the function, IO port 0xCFA, also 8-bit, sets
3011 bus number. The config space is then accessed
3012 through ports 0xC000-0xCFFF).
3013 See http://wiki.osdev.org/PCI for more info
3014 on the configuration access mechanisms.
3015 noaer [PCIE] If the PCIEAER kernel config parameter is
3016 enabled, this kernel boot option can be used to
3017 disable the use of PCIE advanced error reporting.
3018 nodomains [PCI] Disable support for multiple PCI
3019 root domains (aka PCI segments, in ACPI-speak).
3020 nommconf [X86] Disable use of MMCONFIG for PCI
3022 check_enable_amd_mmconf [X86] check for and enable
3023 properly configured MMIO access to PCI
3024 config space on AMD family 10h CPU
3025 nomsi [MSI] If the PCI_MSI kernel config parameter is
3026 enabled, this kernel boot option can be used to
3027 disable the use of MSI interrupts system-wide.
3028 noioapicquirk [APIC] Disable all boot interrupt quirks.
3029 Safety option to keep boot IRQs enabled. This
3030 should never be necessary.
3031 ioapicreroute [APIC] Enable rerouting of boot IRQs to the
3032 primary IO-APIC for bridges that cannot disable
3033 boot IRQs. This fixes a source of spurious IRQs
3034 when the system masks IRQs.
3035 noioapicreroute [APIC] Disable workaround that uses the
3036 boot IRQ equivalent of an IRQ that connects to
3037 a chipset where boot IRQs cannot be disabled.
3038 The opposite of ioapicreroute.
3039 biosirq [X86-32] Use PCI BIOS calls to get the interrupt
3040 routing table. These calls are known to be buggy
3041 on several machines and they hang the machine
3042 when used, but on other computers it's the only
3043 way to get the interrupt routing table. Try
3044 this option if the kernel is unable to allocate
3045 IRQs or discover secondary PCI buses on your
3047 rom [X86] Assign address space to expansion ROMs.
3048 Use with caution as certain devices share
3049 address decoders between ROMs and other
3051 norom [X86] Do not assign address space to
3052 expansion ROMs that do not already have
3053 BIOS assigned address ranges.
3054 nobar [X86] Do not assign address space to the
3055 BARs that weren't assigned by the BIOS.
3056 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be
3057 assigned automatically to PCI devices. You can
3058 make the kernel exclude IRQs of your ISA cards
3060 pirqaddr=0xAAAAA [X86] Specify the physical address
3061 of the PIRQ table (normally generated
3062 by the BIOS) if it is outside the
3063 F0000h-100000h range.
3064 lastbus=N [X86] Scan all buses thru bus #N. Can be
3065 useful if the kernel is unable to find your
3066 secondary buses and you want to tell it
3067 explicitly which ones they are.
3068 assign-busses [X86] Always assign all PCI bus
3069 numbers ourselves, overriding
3070 whatever the firmware may have done.
3071 usepirqmask [X86] Honor the possible IRQ mask stored
3072 in the BIOS $PIR table. This is needed on
3073 some systems with broken BIOSes, notably
3074 some HP Pavilion N5400 and Omnibook XE3
3075 notebooks. This will have no effect if ACPI
3076 IRQ routing is enabled.
3077 noacpi [X86] Do not use ACPI for IRQ routing
3078 or for PCI scanning.
3079 use_crs [X86] Use PCI host bridge window information
3080 from ACPI. On BIOSes from 2008 or later, this
3081 is enabled by default. If you need to use this,
3082 please report a bug.
3083 nocrs [X86] Ignore PCI host bridge windows from ACPI.
3084 If you need to use this, please report a bug.
3085 routeirq Do IRQ routing for all PCI devices.
3086 This is normally done in pci_enable_device(),
3087 so this option is a temporary workaround
3088 for broken drivers that don't call it.
3089 skip_isa_align [X86] do not align io start addr, so can
3090 handle more pci cards
3091 noearly [X86] Don't do any early type 1 scanning.
3092 This might help on some broken boards which
3093 machine check when some devices' config space
3094 is read. But various workarounds are disabled
3095 and some IOMMU drivers will not work.
3096 bfsort Sort PCI devices into breadth-first order.
3097 This sorting is done to get a device
3098 order compatible with older (<= 2.4) kernels.
3099 nobfsort Don't sort PCI devices into breadth-first order.
3100 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size)
3101 tuning and use the BIOS-configured MPS defaults.
3102 pcie_bus_safe Set every device's MPS to the largest value
3103 supported by all devices below the root complex.
3104 pcie_bus_perf Set device MPS to the largest allowable MPS
3105 based on its parent bus. Also set MRRS (Max
3106 Read Request Size) to the largest supported
3107 value (no larger than the MPS that the device
3108 or bus can support) for best performance.
3109 pcie_bus_peer2peer Set every device's MPS to 128B, which
3110 every device is guaranteed to support. This
3111 configuration allows peer-to-peer DMA between
3112 any pair of devices, possibly at the cost of
3113 reduced performance. This also guarantees
3114 that hot-added devices will work.
3115 cbiosize=nn[KMG] The fixed amount of bus space which is
3116 reserved for the CardBus bridge's IO window.
3117 The default value is 256 bytes.
3118 cbmemsize=nn[KMG] The fixed amount of bus space which is
3119 reserved for the CardBus bridge's memory
3120 window. The default value is 64 megabytes.
3123 [<order of align>@][<domain>:]<bus>:<slot>.<func>[; ...]
3124 [<order of align>@]pci:<vendor>:<device>\
3125 [:<subvendor>:<subdevice>][; ...]
3126 Specifies alignment and device to reassign
3127 aligned memory resources.
3128 If <order of align> is not specified,
3129 PAGE_SIZE is used as alignment.
3130 PCI-PCI bridge can be specified, if resource
3131 windows need to be expanded.
3132 To specify the alignment for several
3133 instances of a device, the PCI vendor,
3134 device, subvendor, and subdevice may be
3135 specified, e.g., 4096@pci:8086:9c22:103c:198f
3136 ecrc= Enable/disable PCIe ECRC (transaction layer
3137 end-to-end CRC checking).
3138 bios: Use BIOS/firmware settings. This is the
3142 hpiosize=nn[KMG] The fixed amount of bus space which is
3143 reserved for hotplug bridge's IO window.
3144 Default size is 256 bytes.
3145 hpmemsize=nn[KMG] The fixed amount of bus space which is
3146 reserved for hotplug bridge's memory window.
3147 Default size is 2 megabytes.
3148 hpbussize=nn The minimum amount of additional bus numbers
3149 reserved for buses below a hotplug bridge.
3151 realloc= Enable/disable reallocating PCI bridge resources
3152 if allocations done by BIOS are too small to
3153 accommodate resources required by all child
3155 off: Turn realloc off
3157 realloc same as realloc=on
3158 noari do not use PCIe ARI.
3159 pcie_scan_all Scan all possible PCIe devices. Otherwise we
3160 only look for one device below a PCIe downstream
3162 big_root_window Try to add a big 64bit memory window to the PCIe
3163 root complex on AMD CPUs. Some GFX hardware
3164 can resize a BAR to allow access to all VRAM.
3165 Adding the window is slightly risky (it may
3166 conflict with unreported devices), so this
3169 pcie_aspm= [PCIE] Forcibly enable or disable PCIe Active State Power
3172 force Enable ASPM even on devices that claim not to support it.
3173 WARNING: Forcing ASPM on may cause system lockups.
3175 pcie_ports= [PCIE] PCIe port services handling:
3176 native Use native PCIe services (PME, AER, DPC, PCIe hotplug)
3177 even if the platform doesn't give the OS permission to
3178 use them. This may cause conflicts if the platform
3179 also tries to use these services.
3180 compat Disable native PCIe services (PME, AER, DPC, PCIe
3183 pcie_port_pm= [PCIE] PCIe port power management handling:
3184 off Disable power management of all PCIe ports
3185 force Forcibly enable power management of all PCIe ports
3187 pcie_pme= [PCIE,PM] Native PCIe PME signaling options:
3188 nomsi Do not use MSI for native PCIe PME signaling (this makes
3189 all PCIe root ports use INTx for all services).
3191 pcmv= [HW,PCMCIA] BadgePAD 4
3195 Keep all power-domains already enabled by bootloader on,
3196 even if no driver has claimed them. This is useful
3197 for debug and development, but should not be
3198 needed on a platform with proper driver support.
3201 See Documentation/blockdev/paride.txt.
3203 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at
3206 See arch/parisc/kernel/pdc_chassis.c
3208 percpu_alloc= Select which percpu first chunk allocator to use.
3209 Currently supported values are "embed" and "page".
3210 Archs may support subset or none of the selections.
3211 See comments in mm/percpu.c for details on each
3212 allocator. This parameter is primarily for debugging
3213 and performance comparison.
3216 See Documentation/blockdev/paride.txt.
3219 See Documentation/blockdev/paride.txt.
3221 pirq= [SMP,APIC] Manual mp-table setup
3222 See Documentation/x86/i386/IO-APIC.txt.
3224 plip= [PPT,NET] Parallel port network link
3225 Format: { parport<nr> | timid | 0 }
3226 See also Documentation/admin-guide/parport.rst.
3228 pmtmr= [X86] Manual setup of pmtmr I/O Port.
3229 Override pmtimer IOPort with a hex value.
3233 Enable PNP debug messages (depends on the
3234 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time
3235 via /sys/module/pnp/parameters/debug. We always show
3236 current resource usage; turning this on also shows
3237 possible settings and some assignment information.
3243 { on | off | curr | res | no-curr | no-res }
3246 [ISAPNP] Exclude IRQs for the autoconfiguration
3249 [ISAPNP] Exclude DMAs for the autoconfiguration
3251 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration
3252 Ranges are in pairs (I/O port base and size).
3255 [ISAPNP] Exclude memory regions for the
3257 Ranges are in pairs (memory base and size).
3259 ports= [IP_VS_FTP] IPVS ftp helper module
3261 Up to 8 (IP_VS_APP_MAX_PORTS) ports
3263 Format: <port>,<port>....
3265 powersave=off [PPC] This option disables power saving features.
3266 It specifically disables cpuidle and sets the
3267 platform machine description specific power_save
3268 function to NULL. On Idle the CPU just reduces
3271 ppc_strict_facility_enable
3272 [PPC] This option catches any kernel floating point,
3273 Altivec, VSX and SPE outside of regions specifically
3274 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()).
3275 There is some performance impact when enabling this.
3279 Disable Hardware Transactional Memory
3281 print-fatal-signals=
3282 [KNL] debug: print fatal signals
3284 If enabled, warn about various signal handling
3285 related application anomalies: too many signals,
3286 too many POSIX.1 timers, fatal signals causing a
3289 If you hit the warning due to signal overflow,
3290 you might want to try "ulimit -i unlimited".
3294 printk.always_kmsg_dump=
3295 Trigger kmsg_dump for cases other than kernel oops or
3297 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
3300 printk.devkmsg={on,off,ratelimit}
3301 Control writing to /dev/kmsg.
3302 on - unlimited logging to /dev/kmsg from userspace
3303 off - logging to /dev/kmsg disabled
3304 ratelimit - ratelimit the logging
3307 printk.time= Show timing data prefixed to each printk message line
3308 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
3310 processor.max_cstate= [HW,ACPI]
3311 Limit processor to maximum C-state
3312 max_cstate=9 overrides any DMI blacklist limit.
3314 processor.nocst [HW,ACPI]
3315 Ignore the _CST method to determine C-states,
3316 instead using the legacy FADT method
3318 profile= [KNL] Enable kernel profiling via /proc/profile
3319 Format: [<profiletype>,]<number>
3320 Param: <profiletype>: "schedule", "sleep", or "kvm"
3321 [defaults to kernel profiling]
3322 Param: "schedule" - profile schedule points.
3323 Param: "sleep" - profile D-state sleeping (millisecs).
3324 Requires CONFIG_SCHEDSTATS
3325 Param: "kvm" - profile VM exits.
3326 Param: <number> - step/bucket size as a power of 2 for
3327 statistical time based profiling.
3329 prompt_ramdisk= [RAM] List of RAM disks to prompt for floppy disk
3331 See Documentation/blockdev/ramdisk.txt.
3333 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to
3334 probe for; one of (bare|imps|exps|lifebook|any).
3335 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports
3337 psmouse.resetafter= [HW,MOUSE]
3338 Try to reset the device after so many bad packets
3341 [HW,MOUSE] Set desired mouse resolution, in dpi.
3342 psmouse.smartscroll=
3343 [HW,MOUSE] Controls Logitech smartscroll autorepeat.
3344 0 = disabled, 1 = enabled (default).
3346 pstore.backend= Specify the name of the pstore backend to use
3349 See Documentation/blockdev/paride.txt.
3351 pti= [X86_64] Control Page Table Isolation of user and
3352 kernel address spaces. Disabling this feature
3353 removes hardening, but improves performance of
3354 system calls and interrupts.
3356 on - unconditionally enable
3357 off - unconditionally disable
3358 auto - kernel detects whether your CPU model is
3359 vulnerable to issues that PTI mitigates
3361 Not specifying this option is equivalent to pti=auto.
3364 Equivalent to pti=off
3367 [KNL] Number of legacy pty's. Overwrites compiled-in
3370 quiet [KNL] Disable most log messages
3375 See Documentation/admin-guide/md.rst.
3377 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes
3378 See Documentation/blockdev/ramdisk.txt.
3380 ras=option[,option,...] [KNL] RAS-specific options
3383 Disable the Correctable Errors Collector,
3384 see CONFIG_RAS_CEC help text.
3387 The argument is a cpu list, as described above.
3389 In kernels built with CONFIG_RCU_NOCB_CPU=y, set
3390 the specified list of CPUs to be no-callback CPUs.
3391 Invocation of these CPUs' RCU callbacks will
3392 be offloaded to "rcuox/N" kthreads created for
3393 that purpose, where "x" is "b" for RCU-bh, "p"
3394 for RCU-preempt, and "s" for RCU-sched, and "N"
3395 is the CPU number. This reduces OS jitter on the
3396 offloaded CPUs, which can be useful for HPC and
3397 real-time workloads. It can also improve energy
3398 efficiency for asymmetric multiprocessors.
3401 Rather than requiring that offloaded CPUs
3402 (specified by rcu_nocbs= above) explicitly
3403 awaken the corresponding "rcuoN" kthreads,
3404 make these kthreads poll for callbacks.
3405 This improves the real-time response for the
3406 offloaded CPUs by relieving them of the need to
3407 wake up the corresponding kthread, but degrades
3408 energy efficiency by requiring that the kthreads
3409 periodically wake up to do the polling.
3411 rcutree.blimit= [KNL]
3412 Set maximum number of finished RCU callbacks to
3413 process in one batch.
3415 rcutree.dump_tree= [KNL]
3416 Dump the structure of the rcu_node combining tree
3417 out at early boot. This is used for diagnostic
3418 purposes, to verify correct tree setup.
3420 rcutree.gp_cleanup_delay= [KNL]
3421 Set the number of jiffies to delay each step of
3422 RCU grace-period cleanup.
3424 rcutree.gp_init_delay= [KNL]
3425 Set the number of jiffies to delay each step of
3426 RCU grace-period initialization.
3428 rcutree.gp_preinit_delay= [KNL]
3429 Set the number of jiffies to delay each step of
3430 RCU grace-period pre-initialization, that is,
3431 the propagation of recent CPU-hotplug changes up
3432 the rcu_node combining tree.
3434 rcutree.rcu_fanout_exact= [KNL]
3435 Disable autobalancing of the rcu_node combining
3436 tree. This is used by rcutorture, and might
3437 possibly be useful for architectures having high
3438 cache-to-cache transfer latencies.
3440 rcutree.rcu_fanout_leaf= [KNL]
3441 Change the number of CPUs assigned to each
3442 leaf rcu_node structure. Useful for very
3443 large systems, which will choose the value 64,
3444 and for NUMA systems with large remote-access
3445 latencies, which will choose a value aligned
3446 with the appropriate hardware boundaries.
3448 rcutree.jiffies_till_sched_qs= [KNL]
3449 Set required age in jiffies for a
3450 given grace period before RCU starts
3451 soliciting quiescent-state help from
3452 rcu_note_context_switch().
3454 rcutree.jiffies_till_first_fqs= [KNL]
3455 Set delay from grace-period initialization to
3456 first attempt to force quiescent states.
3457 Units are jiffies, minimum value is zero,
3458 and maximum value is HZ.
3460 rcutree.jiffies_till_next_fqs= [KNL]
3461 Set delay between subsequent attempts to force
3462 quiescent states. Units are jiffies, minimum
3463 value is one, and maximum value is HZ.
3465 rcutree.kthread_prio= [KNL,BOOT]
3466 Set the SCHED_FIFO priority of the RCU per-CPU
3467 kthreads (rcuc/N). This value is also used for
3468 the priority of the RCU boost threads (rcub/N)
3469 and for the RCU grace-period kthreads (rcu_bh,
3470 rcu_preempt, and rcu_sched). If RCU_BOOST is
3471 set, valid values are 1-99 and the default is 1
3472 (the least-favored priority). Otherwise, when
3473 RCU_BOOST is not set, valid values are 0-99 and
3474 the default is zero (non-realtime operation).
3476 rcutree.rcu_nocb_leader_stride= [KNL]
3477 Set the number of NOCB kthread groups, which
3478 defaults to the square root of the number of
3479 CPUs. Larger numbers reduces the wakeup overhead
3480 on the per-CPU grace-period kthreads, but increases
3481 that same overhead on each group's leader.
3483 rcutree.qhimark= [KNL]
3484 Set threshold of queued RCU callbacks beyond which
3485 batch limiting is disabled.
3487 rcutree.qlowmark= [KNL]
3488 Set threshold of queued RCU callbacks below which
3489 batch limiting is re-enabled.
3491 rcutree.rcu_idle_gp_delay= [KNL]
3492 Set wakeup interval for idle CPUs that have
3493 RCU callbacks (RCU_FAST_NO_HZ=y).
3495 rcutree.rcu_idle_lazy_gp_delay= [KNL]
3496 Set wakeup interval for idle CPUs that have
3497 only "lazy" RCU callbacks (RCU_FAST_NO_HZ=y).
3498 Lazy RCU callbacks are those which RCU can
3499 prove do nothing more than free memory.
3501 rcutree.rcu_kick_kthreads= [KNL]
3502 Cause the grace-period kthread to get an extra
3503 wake_up() if it sleeps three times longer than
3504 it should at force-quiescent-state time.
3505 This wake_up() will be accompanied by a
3506 WARN_ONCE() splat and an ftrace_dump().
3508 rcuperf.gp_async= [KNL]
3509 Measure performance of asynchronous
3510 grace-period primitives such as call_rcu().
3512 rcuperf.gp_async_max= [KNL]
3513 Specify the maximum number of outstanding
3514 callbacks per writer thread. When a writer
3515 thread exceeds this limit, it invokes the
3516 corresponding flavor of rcu_barrier() to allow
3517 previously posted callbacks to drain.
3519 rcuperf.gp_exp= [KNL]
3520 Measure performance of expedited synchronous
3521 grace-period primitives.
3523 rcuperf.holdoff= [KNL]
3524 Set test-start holdoff period. The purpose of
3525 this parameter is to delay the start of the
3526 test until boot completes in order to avoid
3529 rcuperf.nreaders= [KNL]
3530 Set number of RCU readers. The value -1 selects
3531 N, where N is the number of CPUs. A value
3532 "n" less than -1 selects N-n+1, where N is again
3533 the number of CPUs. For example, -2 selects N
3534 (the number of CPUs), -3 selects N+1, and so on.
3535 A value of "n" less than or equal to -N selects
3538 rcuperf.nwriters= [KNL]
3539 Set number of RCU writers. The values operate
3540 the same as for rcuperf.nreaders.
3541 N, where N is the number of CPUs
3543 rcuperf.perf_type= [KNL]
3544 Specify the RCU implementation to test.
3546 rcuperf.shutdown= [KNL]
3547 Shut the system down after performance tests
3548 complete. This is useful for hands-off automated
3551 rcuperf.verbose= [KNL]
3552 Enable additional printk() statements.
3554 rcuperf.writer_holdoff= [KNL]
3555 Write-side holdoff between grace periods,
3556 in microseconds. The default of zero says
3559 rcutorture.cbflood_inter_holdoff= [KNL]
3560 Set holdoff time (jiffies) between successive
3561 callback-flood tests.
3563 rcutorture.cbflood_intra_holdoff= [KNL]
3564 Set holdoff time (jiffies) between successive
3565 bursts of callbacks within a given callback-flood
3568 rcutorture.cbflood_n_burst= [KNL]
3569 Set the number of bursts making up a given
3570 callback-flood test. Set this to zero to
3571 disable callback-flood testing.
3573 rcutorture.cbflood_n_per_burst= [KNL]
3574 Set the number of callbacks to be registered
3575 in a given burst of a callback-flood test.
3577 rcutorture.fqs_duration= [KNL]
3578 Set duration of force_quiescent_state bursts
3581 rcutorture.fqs_holdoff= [KNL]
3582 Set holdoff time within force_quiescent_state bursts
3585 rcutorture.fqs_stutter= [KNL]
3586 Set wait time between force_quiescent_state bursts
3589 rcutorture.gp_cond= [KNL]
3590 Use conditional/asynchronous update-side
3591 primitives, if available.
3593 rcutorture.gp_exp= [KNL]
3594 Use expedited update-side primitives, if available.
3596 rcutorture.gp_normal= [KNL]
3597 Use normal (non-expedited) asynchronous
3598 update-side primitives, if available.
3600 rcutorture.gp_sync= [KNL]
3601 Use normal (non-expedited) synchronous
3602 update-side primitives, if available. If all
3603 of rcutorture.gp_cond=, rcutorture.gp_exp=,
3604 rcutorture.gp_normal=, and rcutorture.gp_sync=
3605 are zero, rcutorture acts as if is interpreted
3606 they are all non-zero.
3608 rcutorture.n_barrier_cbs= [KNL]
3609 Set callbacks/threads for rcu_barrier() testing.
3611 rcutorture.nfakewriters= [KNL]
3612 Set number of concurrent RCU writers. These just
3613 stress RCU, they don't participate in the actual
3614 test, hence the "fake".
3616 rcutorture.nreaders= [KNL]
3617 Set number of RCU readers. The value -1 selects
3618 N-1, where N is the number of CPUs. A value
3619 "n" less than -1 selects N-n-2, where N is again
3620 the number of CPUs. For example, -2 selects N
3621 (the number of CPUs), -3 selects N+1, and so on.
3623 rcutorture.object_debug= [KNL]
3624 Enable debug-object double-call_rcu() testing.
3626 rcutorture.onoff_holdoff= [KNL]
3627 Set time (s) after boot for CPU-hotplug testing.
3629 rcutorture.onoff_interval= [KNL]
3630 Set time (s) between CPU-hotplug operations, or
3631 zero to disable CPU-hotplug testing.
3633 rcutorture.shuffle_interval= [KNL]
3634 Set task-shuffle interval (s). Shuffling tasks
3635 allows some CPUs to go into dyntick-idle mode
3636 during the rcutorture test.
3638 rcutorture.shutdown_secs= [KNL]
3639 Set time (s) after boot system shutdown. This
3640 is useful for hands-off automated testing.
3642 rcutorture.stall_cpu= [KNL]
3643 Duration of CPU stall (s) to test RCU CPU stall
3644 warnings, zero to disable.
3646 rcutorture.stall_cpu_holdoff= [KNL]
3647 Time to wait (s) after boot before inducing stall.
3649 rcutorture.stall_cpu_irqsoff= [KNL]
3650 Disable interrupts while stalling if set.
3652 rcutorture.stat_interval= [KNL]
3653 Time (s) between statistics printk()s.
3655 rcutorture.stutter= [KNL]
3656 Time (s) to stutter testing, for example, specifying
3657 five seconds causes the test to run for five seconds,
3658 wait for five seconds, and so on. This tests RCU's
3659 ability to transition abruptly to and from idle.
3661 rcutorture.test_boost= [KNL]
3662 Test RCU priority boosting? 0=no, 1=maybe, 2=yes.
3663 "Maybe" means test if the RCU implementation
3664 under test support RCU priority boosting.
3666 rcutorture.test_boost_duration= [KNL]
3667 Duration (s) of each individual boost test.
3669 rcutorture.test_boost_interval= [KNL]
3670 Interval (s) between each boost test.
3672 rcutorture.test_no_idle_hz= [KNL]
3673 Test RCU's dyntick-idle handling. See also the
3674 rcutorture.shuffle_interval parameter.
3676 rcutorture.torture_type= [KNL]
3677 Specify the RCU implementation to test.
3679 rcutorture.verbose= [KNL]
3680 Enable additional printk() statements.
3682 rcupdate.rcu_cpu_stall_suppress= [KNL]
3683 Suppress RCU CPU stall warning messages.
3685 rcupdate.rcu_cpu_stall_timeout= [KNL]
3686 Set timeout for RCU CPU stall warning messages.
3688 rcupdate.rcu_expedited= [KNL]
3689 Use expedited grace-period primitives, for
3690 example, synchronize_rcu_expedited() instead
3691 of synchronize_rcu(). This reduces latency,
3692 but can increase CPU utilization, degrade
3693 real-time latency, and degrade energy efficiency.
3694 No effect on CONFIG_TINY_RCU kernels.
3696 rcupdate.rcu_normal= [KNL]
3697 Use only normal grace-period primitives,
3698 for example, synchronize_rcu() instead of
3699 synchronize_rcu_expedited(). This improves
3700 real-time latency, CPU utilization, and
3701 energy efficiency, but can expose users to
3702 increased grace-period latency. This parameter
3703 overrides rcupdate.rcu_expedited. No effect on
3704 CONFIG_TINY_RCU kernels.
3706 rcupdate.rcu_normal_after_boot= [KNL]
3707 Once boot has completed (that is, after
3708 rcu_end_inkernel_boot() has been invoked), use
3709 only normal grace-period primitives. No effect
3710 on CONFIG_TINY_RCU kernels.
3712 rcupdate.rcu_task_stall_timeout= [KNL]
3713 Set timeout in jiffies for RCU task stall warning
3714 messages. Disable with a value less than or equal
3717 rcupdate.rcu_self_test= [KNL]
3718 Run the RCU early boot self tests
3720 rcupdate.rcu_self_test_bh= [KNL]
3721 Run the RCU bh early boot self tests
3723 rcupdate.rcu_self_test_sched= [KNL]
3724 Run the RCU sched early boot self tests
3728 Run specified binary instead of /init from the ramdisk,
3729 used for early userspace startup. See initrd.
3732 Turn on/off individual RDT features. List is:
3733 cmt, mbmtotal, mbmlocal, l3cat, l3cdp, l2cat, l2cdp,
3735 E.g. to turn on cmt and turn off mba use:
3739 Format (x86 or x86_64):
3740 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] \
3742 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \
3744 Where reboot_mode is one of warm (soft) or cold (hard) or gpio,
3745 reboot_type is one of bios, acpi, kbd, triple, efi, or pci,
3746 reboot_force is either force or not specified,
3747 reboot_cpu is s[mp]#### with #### being the processor
3748 to be used for rebooting.
3751 [KNL, SMP] Set scheduler's default relax_domain_level.
3752 See Documentation/cgroup-v1/cpusets.txt.
3754 reserve= [KNL,BUGS] Force kernel to ignore I/O ports or memory
3755 Format: <base1>,<size1>[,<base2>,<size2>,...]
3756 Reserve I/O ports or memory so the kernel won't use
3757 them. If <base> is less than 0x10000, the region
3758 is assumed to be I/O ports; otherwise it is memory.
3760 reservetop= [X86-32]
3762 Reserves a hole at the top of the kernel virtual
3767 Set the amount of memory to reserve for BIOS at
3768 the bottom of the address space.
3770 reset_devices [KNL] Force drivers to reset the underlying device
3771 during initialization.
3774 Specify the partition device for software suspend
3776 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>}
3778 resume_offset= [SWSUSP]
3779 Specify the offset from the beginning of the partition
3780 given by "resume=" at which the swap header is located,
3781 in <PAGE_SIZE> units (needed only for swap files).
3782 See Documentation/power/swsusp-and-swap-files.txt
3784 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to
3785 read the resume files
3787 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up.
3788 Useful for devices that are detected asynchronously
3789 (e.g. USB and MMC devices).
3791 hibernate= [HIBERNATION]
3792 noresume Don't check if there's a hibernation image
3793 present during boot.
3794 nocompress Don't compress/decompress hibernation images.
3795 no Disable hibernation and resume.
3796 protect_image Turn on image protection during restoration
3797 (that will set all pages holding image data
3798 during restoration read-only).
3800 retain_initrd [RAM] Keep initrd memory after extraction
3802 rfkill.default_state=
3803 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm,
3804 etc. communication is blocked by default.
3807 rfkill.master_switch_mode=
3808 0 The "airplane mode" button does nothing.
3809 1 The "airplane mode" button toggles between everything
3810 blocked and the previous configuration.
3811 2 The "airplane mode" button toggles between everything
3812 blocked and everything unblocked.
3814 rhash_entries= [KNL,NET]
3815 Set number of hash buckets for route cache
3818 [KNL] Disable ring 3 MONITOR/MWAIT feature on supported
3821 ro [KNL] Mount root device read-only on boot
3824 on Mark read-only kernel memory as read-only (default).
3825 off Leave read-only kernel memory writable for debugging.
3828 Enable the uart passthrough on the designated usb port
3829 on Rockchip SoCs. When active, the signals of the
3830 debug-uart get routed to the D+ and D- pins of the usb
3831 port and the regular usb controller gets disabled.
3833 root= [KNL] Root filesystem
3834 See name_to_dev_t comment in init/do_mounts.c.
3836 rootdelay= [KNL] Delay (in seconds) to pause before attempting to
3837 mount the root filesystem
3839 rootflags= [KNL] Set root filesystem mount option string
3841 rootfstype= [KNL] Set root filesystem type
3843 rootwait [KNL] Wait (indefinitely) for root device to show up.
3844 Useful for devices that are detected asynchronously
3845 (e.g. USB and MMC devices).
3847 rproc_mem=nn[KMG][@address]
3848 [KNL,ARM,CMA] Remoteproc physical memory block.
3849 Memory area to be used by remote processor image,
3852 rw [KNL] Mount root device read-write on boot
3854 S [KNL] Run init in single mode
3856 s390_iommu= [HW,S390]
3857 Set s390 IOTLB flushing mode
3859 With strict flushing every unmap operation will result in
3860 an IOTLB flush. Default is lazy flushing before reuse,
3864 See drivers/net/irda/sa1100_ir.c.
3866 sbni= [NET] Granch SBNI12 leased line adapter
3868 sched_debug [KNL] Enables verbose scheduler debug messages.
3870 schedstats= [KNL,X86] Enable or disable scheduled statistics.
3871 Allowed values are enable and disable. This feature
3872 incurs a small amount of overhead in the scheduler
3873 but is useful for debugging and performance tuning.
3875 skew_tick= [KNL] Offset the periodic timer tick per cpu to mitigate
3876 xtime_lock contention on larger systems, and/or RCU lock
3877 contention on all systems with CONFIG_MAXSMP set.
3878 Format: { "0" | "1" }
3879 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1"
3881 Note: increases power consumption, thus should only be
3882 enabled if running jitter sensitive (HPC/RT) workloads.
3884 security= [SECURITY] Choose a security module to enable at boot.
3885 If this boot parameter is not specified, only the first
3886 security module asking for security registration will be
3887 loaded. An invalid security module name will be treated
3888 as if no module has been chosen.
3890 selinux= [SELINUX] Disable or enable SELinux at boot time.
3891 Format: { "0" | "1" }
3892 See security/selinux/Kconfig help text.
3895 Default value is set via kernel config option.
3896 If enabled at boot time, /selinux/disable can be used
3897 later to disable prior to initial policy load.
3899 apparmor= [APPARMOR] Disable or enable AppArmor at boot time
3900 Format: { "0" | "1" }
3901 See security/apparmor/Kconfig help text
3904 Default value is set via kernel config option.
3906 serialnumber [BUGS=X86-32]
3909 Maximal number of shapers.
3917 Disable merging of slabs with similar size. May be
3918 necessary if there is some reason to distinguish
3919 allocs to different slabs, especially in hardened
3920 environments where the risk of heap overflows and
3921 layout control by attackers can usually be
3922 frustrated by disabling merging. This will reduce
3923 most of the exposure of a heap attack to a single
3924 cache (risks via metadata attacks are mostly
3925 unchanged). Debug options disable merging on their
3927 For more information see Documentation/vm/slub.rst.
3929 slab_max_order= [MM, SLAB]
3930 Determines the maximum allowed order for slabs.
3931 A high setting may cause OOMs due to memory
3932 fragmentation. Defaults to 1 for systems with
3933 more than 32MB of RAM, 0 otherwise.
3935 slub_debug[=options[,slabs]] [MM, SLUB]
3936 Enabling slub_debug allows one to determine the
3937 culprit if slab objects become corrupted. Enabling
3938 slub_debug can create guard zones around objects and
3939 may poison objects when not in use. Also tracks the
3940 last alloc / free. For more information see
3941 Documentation/vm/slub.rst.
3943 slub_memcg_sysfs= [MM, SLUB]
3944 Determines whether to enable sysfs directories for
3945 memory cgroup sub-caches. 1 to enable, 0 to disable.
3946 The default is determined by CONFIG_SLUB_MEMCG_SYSFS_ON.
3947 Enabling this can lead to a very high number of debug
3948 directories and files being created under
3951 slub_max_order= [MM, SLUB]
3952 Determines the maximum allowed order for slabs.
3953 A high setting may cause OOMs due to memory
3954 fragmentation. For more information see
3955 Documentation/vm/slub.rst.
3957 slub_min_objects= [MM, SLUB]
3958 The minimum number of objects per slab. SLUB will
3959 increase the slab order up to slub_max_order to
3960 generate a sufficiently large slab able to contain
3961 the number of objects indicated. The higher the number
3962 of objects the smaller the overhead of tracking slabs
3963 and the less frequently locks need to be acquired.
3964 For more information see Documentation/vm/slub.rst.
3966 slub_min_order= [MM, SLUB]
3967 Determines the minimum page order for slabs. Must be
3968 lower than slub_max_order.
3969 For more information see Documentation/vm/slub.rst.
3971 slub_nomerge [MM, SLUB]
3972 Same with slab_nomerge. This is supported for legacy.
3973 See slab_nomerge for more information.
3976 Format: <io1>[,<io2>[,...,<io8>]]
3978 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices
3979 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port
3980 smsc-ircc2.ircc_sir= [HW] SIR base I/O port
3981 smsc-ircc2.ircc_fir= [HW] FIR base I/O port
3982 smsc-ircc2.ircc_irq= [HW] IRQ line
3983 smsc-ircc2.ircc_dma= [HW] DMA channel
3984 smsc-ircc2.ircc_transceiver= [HW] Transceiver type:
3985 0: Toshiba Satellite 1800 (GP data pin select)
3986 1: Fast pin select (default)
3989 smt [KNL,S390] Set the maximum number of threads (logical
3990 CPUs) to use per physical CPU on systems capable of
3991 symmetric multithreading (SMT). Will be capped to the
3992 actual hardware limit.
3994 Default: -1 (no limit)
3997 [KNL] Should the soft-lockup detector generate panics.
4000 A nonzero value instructs the soft-lockup detector
4001 to panic the machine when a soft-lockup occurs. This
4002 is also controlled by CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC
4003 which is the respective build-time switch to that
4006 softlockup_all_cpu_backtrace=
4007 [KNL] Should the soft-lockup detector generate
4008 backtraces on all cpus.
4011 sonypi.*= [HW] Sony Programmable I/O Control Device driver
4012 See Documentation/laptops/sonypi.txt
4014 spectre_v2= [X86] Control mitigation of Spectre variant 2
4015 (indirect branch speculation) vulnerability.
4017 on - unconditionally enable
4018 off - unconditionally disable
4019 auto - kernel detects whether your CPU model is
4022 Selecting 'on' will, and 'auto' may, choose a
4023 mitigation method at run time according to the
4024 CPU, the available microcode, the setting of the
4025 CONFIG_RETPOLINE configuration option, and the
4026 compiler with which the kernel was built.
4028 Specific mitigations can also be selected manually:
4030 retpoline - replace indirect branches
4031 retpoline,generic - google's original retpoline
4032 retpoline,amd - AMD-specific minimal thunk
4034 Not specifying this option is equivalent to
4037 spec_store_bypass_disable=
4038 [HW] Control Speculative Store Bypass (SSB) Disable mitigation
4039 (Speculative Store Bypass vulnerability)
4041 Certain CPUs are vulnerable to an exploit against a
4042 a common industry wide performance optimization known
4043 as "Speculative Store Bypass" in which recent stores
4044 to the same memory location may not be observed by
4045 later loads during speculative execution. The idea
4046 is that such stores are unlikely and that they can
4047 be detected prior to instruction retirement at the
4048 end of a particular speculation execution window.
4050 In vulnerable processors, the speculatively forwarded
4051 store can be used in a cache side channel attack, for
4052 example to read memory to which the attacker does not
4053 directly have access (e.g. inside sandboxed code).
4055 This parameter controls whether the Speculative Store
4056 Bypass optimization is used.
4058 on - Unconditionally disable Speculative Store Bypass
4059 off - Unconditionally enable Speculative Store Bypass
4060 auto - Kernel detects whether the CPU model contains an
4061 implementation of Speculative Store Bypass and
4062 picks the most appropriate mitigation. If the
4063 CPU is not vulnerable, "off" is selected. If the
4064 CPU is vulnerable the default mitigation is
4065 architecture and Kconfig dependent. See below.
4066 prctl - Control Speculative Store Bypass per thread
4067 via prctl. Speculative Store Bypass is enabled
4068 for a process by default. The state of the control
4069 is inherited on fork.
4070 seccomp - Same as "prctl" above, but all seccomp threads
4071 will disable SSB unless they explicitly opt out.
4073 Not specifying this option is equivalent to
4074 spec_store_bypass_disable=auto.
4076 Default mitigations:
4077 X86: If CONFIG_SECCOMP=y "seccomp", otherwise "prctl"
4079 spia_io_base= [HW,MTD]
4084 srcutree.counter_wrap_check [KNL]
4085 Specifies how frequently to check for
4086 grace-period sequence counter wrap for the
4087 srcu_data structure's ->srcu_gp_seq_needed field.
4088 The greater the number of bits set in this kernel
4089 parameter, the less frequently counter wrap will
4090 be checked for. Note that the bottom two bits
4093 srcutree.exp_holdoff [KNL]
4094 Specifies how many nanoseconds must elapse
4095 since the end of the last SRCU grace period for
4096 a given srcu_struct until the next normal SRCU
4097 grace period will be considered for automatic
4098 expediting. Set to zero to disable automatic
4101 stack_guard_gap= [MM]
4102 override the default stack gap protection. The value
4103 is in page units and it defines how many pages prior
4104 to (for stacks growing down) resp. after (for stacks
4105 growing up) the main stack are reserved for no other
4106 mapping. Default value is 256 pages.
4109 Enabled the stack tracer on boot up.
4111 stacktrace_filter=[function-list]
4112 [FTRACE] Limit the functions that the stack tracer
4113 will trace at boot up. function-list is a comma separated
4114 list of functions. This list can be changed at run
4115 time by the stack_trace_filter file in the debugfs
4116 tracing directory. Note, this enables stack tracing
4117 and the stacktrace above is not needed.
4121 Set the STI (builtin display/keyboard on the HP-PARISC
4122 machines) console (graphic card) which should be used
4123 as the initial boot-console.
4124 See also comment in drivers/video/console/sticore.c.
4127 See comment in drivers/video/console/sticore.c.
4130 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]]
4132 sunrpc.min_resvport=
4133 sunrpc.max_resvport=
4135 SunRPC servers often require that client requests
4136 originate from a privileged port (i.e. a port in the
4137 range 0 < portnr < 1024).
4138 An administrator who wishes to reserve some of these
4139 ports for other uses may adjust the range that the
4140 kernel's sunrpc client considers to be privileged
4141 using these two parameters to set the minimum and
4142 maximum port values.
4144 sunrpc.svc_rpc_per_connection_limit=
4146 Limit the number of requests that the server will
4147 process in parallel from a single connection.
4148 The default value is 0 (no limit).
4152 Control how the NFS server code allocates CPUs to
4153 service thread pools. Depending on how many NICs
4154 you have and where their interrupts are bound, this
4155 option will affect which CPUs will do NFS serving.
4156 Note: this parameter cannot be changed while the
4157 NFS server is running.
4159 auto the server chooses an appropriate mode
4160 automatically using heuristics
4161 global a single global pool contains all CPUs
4162 percpu one pool for each CPU
4163 pernode one pool for each NUMA node (equivalent
4164 to global on non-NUMA machines)
4166 sunrpc.tcp_slot_table_entries=
4167 sunrpc.udp_slot_table_entries=
4169 Sets the upper limit on the number of simultaneous
4170 RPC calls that can be sent from the client to a
4171 server. Increasing these values may allow you to
4172 improve throughput, but will also increase the
4173 amount of memory reserved for use by the client.
4175 suspend.pm_test_delay=
4177 Sets the number of seconds to remain in a suspend test
4178 mode before resuming the system (see
4179 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG
4180 is set. Default value is 5.
4183 [KNL] Enable accounting of swap in memory resource
4184 controller if no parameter or 1 is given or disable
4185 it if 0 is given (See Documentation/cgroup-v1/memory.txt)
4187 swiotlb= [ARM,IA-64,PPC,MIPS,X86]
4188 Format: { <int> | force | noforce }
4189 <int> -- Number of I/O TLB slabs
4190 force -- force using of bounce buffers even if they
4191 wouldn't be automatically used by the kernel
4192 noforce -- Never use bounce buffers (for debugging)
4196 sysfs.deprecated=0|1 [KNL]
4197 Enable/disable old style sysfs layout for old udev
4198 on older distributions. When this option is enabled
4199 very new udev will not work anymore. When this option
4200 is disabled (or CONFIG_SYSFS_DEPRECATED not compiled)
4201 in older udev will not work anymore.
4202 Default depends on CONFIG_SYSFS_DEPRECATED_V2 set in
4203 the kernel configuration.
4205 sysrq_always_enabled
4207 Ignore sysrq setting - this boot parameter will
4208 neutralize any effect of /proc/sys/kernel/sysrq.
4209 Useful for debugging.
4211 tcpmhash_entries= [KNL,NET]
4212 Set the number of tcp_metrics_hash slots.
4213 Default value is 8192 or 16384 depending on total
4214 ram pages. This is used to specify the TCP metrics
4215 cache size. See Documentation/networking/ip-sysctl.txt
4216 "tcp_no_metrics_save" section for more details.
4220 test_suspend= [SUSPEND][,N]
4221 Specify "mem" (for Suspend-to-RAM) or "standby" (for
4222 standby suspend) or "freeze" (for suspend type freeze)
4223 as the system sleep state during system startup with
4224 the optional capability to repeat N number of times.
4225 The system is woken from this state using a
4226 wakeup-capable RTC alarm.
4228 thash_entries= [KNL,NET]
4229 Set number of hash buckets for TCP connection
4231 thermal.act= [HW,ACPI]
4232 -1: disable all active trip points in all thermal zones
4233 <degrees C>: override all lowest active trip points
4235 thermal.crt= [HW,ACPI]
4236 -1: disable all critical trip points in all thermal zones
4237 <degrees C>: override all critical trip points
4239 thermal.nocrt= [HW,ACPI]
4240 Set to disable actions on ACPI thermal zone
4241 critical and hot trip points.
4243 thermal.off= [HW,ACPI]
4244 1: disable ACPI thermal control
4246 thermal.psv= [HW,ACPI]
4247 -1: disable all passive trip points
4248 <degrees C>: override all passive trip points to this
4251 thermal.tzp= [HW,ACPI]
4252 Specify global default ACPI thermal zone polling rate
4253 <deci-seconds>: poll all this frequency
4254 0: no polling (default)
4257 Force threading of all interrupt handlers except those
4258 marked explicitly IRQF_NO_THREAD.
4261 Enable the Transcendent memory driver if built-in.
4263 tmem.cleancache=0|1 [KNL, XEN]
4264 Default is on (1). Disable the usage of the cleancache
4265 API to send anonymous pages to the hypervisor.
4267 tmem.frontswap=0|1 [KNL, XEN]
4268 Default is on (1). Disable the usage of the frontswap
4269 API to send swap pages to the hypervisor. If disabled
4270 the selfballooning and selfshrinking are force disabled.
4272 tmem.selfballooning=0|1 [KNL, XEN]
4273 Default is on (1). Disable the driving of swap pages
4276 tmem.selfshrinking=0|1 [KNL, XEN]
4277 Default is on (1). Partial swapoff that immediately
4278 transfers pages from Xen hypervisor back to the
4279 kernel based on different criteria.
4283 Specify if the kernel should make use of the cpu
4284 topology information if the hardware supports this.
4285 The scheduler will make use of this information and
4286 e.g. base its process migration decisions on it.
4289 topology_updates= [KNL, PPC, NUMA]
4291 Specify if the kernel should ignore (off)
4292 topology updates sent by the hypervisor to this
4297 tpm_suspend_pcr=[HW,TPM]
4298 Format: integer pcr id
4299 Specify that at suspend time, the tpm driver
4300 should extend the specified pcr with zeros,
4301 as a workaround for some chips which fail to
4302 flush the last written pcr on TPM_SaveState.
4303 This will guarantee that all the other pcrs
4306 trace_buf_size=nn[KMG]
4307 [FTRACE] will set tracing buffer size on each cpu.
4309 trace_event=[event-list]
4310 [FTRACE] Set and start specified trace events in order
4311 to facilitate early boot debugging. The event-list is a
4312 comma separated list of trace events to enable. See
4313 also Documentation/trace/events.txt
4315 trace_options=[option-list]
4316 [FTRACE] Enable or disable tracer options at boot.
4317 The option-list is a comma delimited list of options
4318 that can be enabled or disabled just as if you were
4319 to echo the option name into
4321 /sys/kernel/debug/tracing/trace_options
4323 For example, to enable stacktrace option (to dump the
4324 stack trace of each event), add to the command line:
4326 trace_options=stacktrace
4328 See also Documentation/trace/ftrace.txt "trace options"
4332 Have the tracepoints sent to printk as well as the
4333 tracing ring buffer. This is useful for early boot up
4334 where the system hangs or reboots and does not give the
4335 option for reading the tracing buffer or performing a
4336 ftrace_dump_on_oops.
4338 To turn off having tracepoints sent to printk,
4339 echo 0 > /proc/sys/kernel/tracepoint_printk
4340 Note, echoing 1 into this file without the
4341 tracepoint_printk kernel cmdline option has no effect.
4345 Having tracepoints sent to printk() and activating high
4346 frequency tracepoints such as irq or sched, can cause
4347 the system to live lock.
4350 [FTRACE] enable this option to disable tracing when a
4351 warning is hit. This turns off "tracing_on". Tracing can
4352 be enabled again by echoing '1' into the "tracing_on"
4353 file located in /sys/kernel/debug/tracing/
4355 This option is useful, as it disables the trace before
4356 the WARNING dump is called, which prevents the trace to
4357 be filled with content caused by the warning output.
4359 This option can also be set at run time via the sysctl
4360 option: kernel/traceoff_on_warning
4362 transparent_hugepage=
4364 Format: [always|madvise|never]
4365 Can be used to control the default behavior of the system
4366 with respect to transparent hugepages.
4367 See Documentation/admin-guide/mm/transhuge.rst
4370 tsc= Disable clocksource stability checks for TSC.
4372 [x86] reliable: mark tsc clocksource as reliable, this
4373 disables clocksource verification at runtime, as well
4374 as the stability checks done at bootup. Used to enable
4375 high-resolution timer mode on older hardware, and in
4376 virtualized environment.
4377 [x86] noirqtime: Do not use TSC to do irq accounting.
4378 Used to run time disable IRQ_TIME_ACCOUNTING on any
4379 platforms where RDTSC is slow and this accounting
4381 [x86] unstable: mark the TSC clocksource as unstable, this
4382 marks the TSC unconditionally unstable at bootup and
4383 avoids any further wobbles once the TSC watchdog notices.
4385 turbografx.map[2|3]= [HW,JOY]
4386 TurboGraFX parallel port interface
4388 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7>
4389 See also Documentation/input/devices/joystick-parport.rst
4391 udbg-immortal [PPC] When debugging early kernel crashes that
4392 happen after console_init() and before a proper
4393 console driver takes over, this boot options might
4394 help "seeing" what's going on.
4396 uhash_entries= [KNL,NET]
4397 Set number of hash buckets for UDP/UDP-Lite connections
4400 [USB] Ignore overcurrent events (default N).
4401 Some badly-designed motherboards generate lots of
4402 bogus events, for ports that aren't wired to
4403 anything. Set this parameter to avoid log spamming.
4404 Note that genuine overcurrent events won't be
4408 [X86] Cause panic on unknown NMI.
4410 usbcore.authorized_default=
4411 [USB] Default USB device authorization:
4412 (default -1 = authorized except for wireless USB,
4413 0 = not authorized, 1 = authorized)
4415 usbcore.autosuspend=
4416 [USB] The autosuspend time delay (in seconds) used
4417 for newly-detected USB devices (default 2). This
4418 is the time required before an idle device will be
4419 autosuspended. Devices for which the delay is set
4420 to a negative value won't be autosuspended at all.
4422 usbcore.usbfs_snoop=
4423 [USB] Set to log all usbfs traffic (default 0 = off).
4425 usbcore.usbfs_snoop_max=
4426 [USB] Maximum number of bytes to snoop in each URB
4429 usbcore.blinkenlights=
4430 [USB] Set to cycle leds on hubs (default 0 = off).
4432 usbcore.old_scheme_first=
4433 [USB] Start with the old device initialization
4434 scheme (default 0 = off).
4436 usbcore.usbfs_memory_mb=
4437 [USB] Memory limit (in MB) for buffers allocated by
4438 usbfs (default = 16, 0 = max = 2047).
4440 usbcore.use_both_schemes=
4441 [USB] Try the other device initialization scheme
4442 if the first one fails (default 1 = enabled).
4444 usbcore.initial_descriptor_timeout=
4445 [USB] Specifies timeout for the initial 64-byte
4446 USB_REQ_GET_DESCRIPTOR request in milliseconds
4447 (default 5000 = 5.0 seconds).
4449 usbcore.nousb [USB] Disable the USB subsystem
4452 [USB] A list of quirk entries to augment the built-in
4453 usb core quirk list. List entries are separated by
4454 commas. Each entry has the form
4455 VendorID:ProductID:Flags. The IDs are 4-digit hex
4456 numbers and Flags is a set of letters. Each letter
4457 will change the built-in quirk; setting it if it is
4458 clear and clearing it if it is set. The letters have
4459 the following meanings:
4460 a = USB_QUIRK_STRING_FETCH_255 (string
4461 descriptors must not be fetched using
4463 b = USB_QUIRK_RESET_RESUME (device can't resume
4464 correctly so reset it instead);
4465 c = USB_QUIRK_NO_SET_INTF (device can't handle
4466 Set-Interface requests);
4467 d = USB_QUIRK_CONFIG_INTF_STRINGS (device can't
4468 handle its Configuration or Interface
4470 e = USB_QUIRK_RESET (device can't be reset
4471 (e.g morph devices), don't use reset);
4472 f = USB_QUIRK_HONOR_BNUMINTERFACES (device has
4473 more interface descriptions than the
4474 bNumInterfaces count, and can't handle
4475 talking to these interfaces);
4476 g = USB_QUIRK_DELAY_INIT (device needs a pause
4477 during initialization, after we read
4478 the device descriptor);
4479 h = USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL (For
4480 high speed and super speed interrupt
4481 endpoints, the USB 2.0 and USB 3.0 spec
4482 require the interval in microframes (1
4483 microframe = 125 microseconds) to be
4484 calculated as interval = 2 ^
4486 Devices with this quirk report their
4487 bInterval as the result of this
4488 calculation instead of the exponent
4489 variable used in the calculation);
4490 i = USB_QUIRK_DEVICE_QUALIFIER (device can't
4491 handle device_qualifier descriptor
4493 j = USB_QUIRK_IGNORE_REMOTE_WAKEUP (device
4494 generates spurious wakeup, ignore
4495 remote wakeup capability);
4496 k = USB_QUIRK_NO_LPM (device can't handle Link
4498 l = USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL
4499 (Device reports its bInterval as linear
4500 frames instead of the USB 2.0
4502 m = USB_QUIRK_DISCONNECT_SUSPEND (Device needs
4503 to be disconnected before suspend to
4504 prevent spurious wakeup);
4505 n = USB_QUIRK_DELAY_CTRL_MSG (Device needs a
4506 pause after every control message);
4507 Example: quirks=0781:5580:bk,0a5c:5834:gij
4510 [USBHID] The interval which mice are to be polled at.
4513 [USBHID] The interval which joysticks are to be polled at.
4516 [USBHID] The interval which keyboards are to be polled at.
4518 usb-storage.delay_use=
4519 [UMS] The delay in seconds before a new device is
4520 scanned for Logical Units (default 1).
4523 [UMS] A list of quirks entries to supplement or
4524 override the built-in unusual_devs list. List
4525 entries are separated by commas. Each entry has
4526 the form VID:PID:Flags where VID and PID are Vendor
4527 and Product ID values (4-digit hex numbers) and
4528 Flags is a set of characters, each corresponding
4529 to a common usb-storage quirk flag as follows:
4530 a = SANE_SENSE (collect more than 18 bytes
4532 b = BAD_SENSE (don't collect more than 18
4533 bytes of sense data);
4534 c = FIX_CAPACITY (decrease the reported
4535 device capacity by one sector);
4536 d = NO_READ_DISC_INFO (don't use
4537 READ_DISC_INFO command);
4538 e = NO_READ_CAPACITY_16 (don't use
4539 READ_CAPACITY_16 command);
4540 f = NO_REPORT_OPCODES (don't use report opcodes
4542 g = MAX_SECTORS_240 (don't transfer more than
4543 240 sectors at a time, uas only);
4544 h = CAPACITY_HEURISTICS (decrease the
4545 reported device capacity by one
4546 sector if the number is odd);
4547 i = IGNORE_DEVICE (don't bind to this
4549 j = NO_REPORT_LUNS (don't use report luns
4551 l = NOT_LOCKABLE (don't try to lock and
4552 unlock ejectable media);
4553 m = MAX_SECTORS_64 (don't transfer more
4554 than 64 sectors = 32 KB at a time);
4555 n = INITIAL_READ10 (force a retry of the
4556 initial READ(10) command);
4557 o = CAPACITY_OK (accept the capacity
4558 reported by the device);
4559 p = WRITE_CACHE (the device cache is ON
4561 r = IGNORE_RESIDUE (the device reports
4562 bogus residue values);
4563 s = SINGLE_LUN (the device has only one
4565 t = NO_ATA_1X (don't allow ATA(12) and ATA(16)
4566 commands, uas only);
4567 u = IGNORE_UAS (don't bind to the uas driver);
4568 w = NO_WP_DETECT (don't test whether the
4569 medium is write-protected).
4570 y = ALWAYS_SYNC (issue a SYNCHRONIZE_CACHE
4571 even if the device claims no cache)
4572 Example: quirks=0419:aaf5:rl,0421:0433:rc
4574 user_debug= [KNL,ARM]
4576 See arch/arm/Kconfig.debug help text.
4577 1 - undefined instruction events
4579 4 - invalid data aborts
4582 Example: user_debug=31
4585 [X86] Flags controlling user PTE allocations.
4587 nohigh = do not allocate PTE pages in
4588 HIGHMEM regardless of setting
4592 On X86_32, this is an alias for vdso32=. Otherwise:
4594 vdso=1: enable VDSO (the default)
4595 vdso=0: disable VDSO mapping
4597 vdso32= [X86] Control the 32-bit vDSO
4598 vdso32=1: enable 32-bit VDSO
4599 vdso32=0 or vdso32=2: disable 32-bit VDSO
4601 See the help text for CONFIG_COMPAT_VDSO for more
4602 details. If CONFIG_COMPAT_VDSO is set, the default is
4603 vdso32=0; otherwise, the default is vdso32=1.
4605 For compatibility with older kernels, vdso32=2 is an
4608 Try vdso32=0 if you encounter an error that says:
4609 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed!
4612 vector=percpu: enable percpu vector domain
4614 video= [FB] Frame buffer configuration
4615 See Documentation/fb/modedb.txt.
4617 video.brightness_switch_enabled= [0,1]
4618 If set to 1, on receiving an ACPI notify event
4619 generated by hotkey, video driver will adjust brightness
4620 level and then send out the event to user space through
4621 the allocated input device; If set to 0, video driver
4622 will only send out the event without touching backlight
4627 [VMMIO] Memory mapped virtio (platform) device.
4629 <size>@<baseaddr>:<irq>[:<id>]
4631 <size> := size (can use standard suffixes
4633 <baseaddr> := physical base address
4634 <irq> := interrupt number (as passed to
4636 <id> := (optional) platform device id
4638 virtio_mmio.device=1K@0x100b0000:48:7
4640 Can be used multiple times for multiple devices.
4642 vga= [BOOT,X86-32] Select a particular video mode
4643 See Documentation/x86/boot.txt and
4644 Documentation/svga.txt.
4645 Use vga=ask for menu.
4646 This is actually a boot loader parameter; the value is
4647 passed to the kernel using a special protocol.
4649 vmalloc=nn[KMG] [KNL,BOOT] Forces the vmalloc area to have an exact
4650 size of <nn>. This can be used to increase the
4651 minimum size (128MB on x86). It can also be used to
4652 decrease the size and leave more room for directly
4655 vmcp_cma=nn[MG] [KNL,S390]
4656 Sets the memory size reserved for contiguous memory
4657 allocations for the vmcp device driver.
4659 vmhalt= [KNL,S390] Perform z/VM CP command after system halt.
4662 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic.
4665 vmpoff= [KNL,S390] Perform z/VM CP command after power off.
4669 Controls the behavior of vsyscalls (i.e. calls to
4670 fixed addresses of 0xffffffffff600x00 from legacy
4671 code). Most statically-linked binaries and older
4672 versions of glibc use these calls. Because these
4673 functions are at fixed addresses, they make nice
4674 targets for exploits that can control RIP.
4676 emulate [default] Vsyscalls turn into traps and are
4677 emulated reasonably safely.
4679 native Vsyscalls are native syscall instructions.
4680 This is a little bit faster than trapping
4681 and makes a few dynamic recompilers work
4682 better than they would in emulation mode.
4683 It also makes exploits much easier to write.
4685 none Vsyscalls don't work at all. This makes
4686 them quite hard to use for exploits but
4687 might break your system.
4689 vt.color= [VT] Default text color.
4690 Format: 0xYX, X = foreground, Y = background.
4691 Default: 0x07 = light gray on black.
4693 vt.cur_default= [VT] Default cursor shape.
4694 Format: 0xCCBBAA, where AA, BB, and CC are the same as
4695 the parameters of the <Esc>[?A;B;Cc escape sequence;
4696 see VGA-softcursor.txt. Default: 2 = underline.
4698 vt.default_blu= [VT]
4699 Format: <blue0>,<blue1>,<blue2>,...,<blue15>
4700 Change the default blue palette of the console.
4701 This is a 16-member array composed of values
4704 vt.default_grn= [VT]
4705 Format: <green0>,<green1>,<green2>,...,<green15>
4706 Change the default green palette of the console.
4707 This is a 16-member array composed of values
4710 vt.default_red= [VT]
4711 Format: <red0>,<red1>,<red2>,...,<red15>
4712 Change the default red palette of the console.
4713 This is a 16-member array composed of values
4719 Set system-wide default UTF-8 mode for all tty's.
4720 Default is 1, i.e. UTF-8 mode is enabled for all
4721 newly opened terminals.
4723 vt.global_cursor_default=
4726 Set system-wide default for whether a cursor
4727 is shown on new VTs. Default is -1,
4728 i.e. cursors will be created by default unless
4729 overridden by individual drivers. 0 will hide
4730 cursors, 1 will display them.
4732 vt.italic= [VT] Default color for italic text; 0-15.
4735 vt.underline= [VT] Default color for underlined text; 0-15.
4738 watchdog timers [HW,WDT] For information on watchdog timers,
4739 see Documentation/watchdog/watchdog-parameters.txt
4740 or other driver-specific files in the
4741 Documentation/watchdog/ directory.
4743 workqueue.watchdog_thresh=
4744 If CONFIG_WQ_WATCHDOG is configured, workqueue can
4745 warn stall conditions and dump internal state to
4746 help debugging. 0 disables workqueue stall
4747 detection; otherwise, it's the stall threshold
4748 duration in seconds. The default value is 30 and
4749 it can be updated at runtime by writing to the
4750 corresponding sysfs file.
4752 workqueue.disable_numa
4753 By default, all work items queued to unbound
4754 workqueues are affine to the NUMA nodes they're
4755 issued on, which results in better behavior in
4756 general. If NUMA affinity needs to be disabled for
4757 whatever reason, this option can be used. Note
4758 that this also can be controlled per-workqueue for
4759 workqueues visible under /sys/bus/workqueue/.
4761 workqueue.power_efficient
4762 Per-cpu workqueues are generally preferred because
4763 they show better performance thanks to cache
4764 locality; unfortunately, per-cpu workqueues tend to
4765 be more power hungry than unbound workqueues.
4767 Enabling this makes the per-cpu workqueues which
4768 were observed to contribute significantly to power
4769 consumption unbound, leading to measurably lower
4770 power usage at the cost of small performance
4773 The default value of this parameter is determined by
4774 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT.
4776 workqueue.debug_force_rr_cpu
4777 Workqueue used to implicitly guarantee that work
4778 items queued without explicit CPU specified are put
4779 on the local CPU. This guarantee is no longer true
4780 and while local CPU is still preferred work items
4781 may be put on foreign CPUs. This debug option
4782 forces round-robin CPU selection to flush out
4783 usages which depend on the now broken guarantee.
4784 When enabled, memory and cache locality will be
4787 x2apic_phys [X86-64,APIC] Use x2apic physical mode instead of
4788 default x2apic cluster mode on platforms
4791 x86_intel_mid_timer= [X86-32,APBT]
4792 Choose timer option for x86 Intel MID platform.
4793 Two valid options are apbt timer only and lapic timer
4794 plus one apbt timer for broadcast timer.
4795 x86_intel_mid_timer=apbt_only | lapic_and_apbt
4797 xen_512gb_limit [KNL,X86-64,XEN]
4798 Restricts the kernel running paravirtualized under Xen
4799 to use only up to 512 GB of RAM. The reason to do so is
4800 crash analysis tools and Xen tools for doing domain
4801 save/restore/migration must be enabled to handle larger
4804 xen_emul_unplug= [HW,X86,XEN]
4805 Unplug Xen emulated devices
4806 Format: [unplug0,][unplug1]
4807 ide-disks -- unplug primary master IDE devices
4808 aux-ide-disks -- unplug non-primary-master IDE devices
4809 nics -- unplug network devices
4810 all -- unplug all emulated devices (NICs and IDE disks)
4811 unnecessary -- unplugging emulated devices is
4812 unnecessary even if the host did not respond to
4814 never -- do not unplug even if version check succeeds
4816 xen_nopvspin [X86,XEN]
4817 Disables the ticketlock slowpath using Xen PV
4821 Disables the PV optimizations forcing the HVM guest to
4822 run as generic HVM guest with no PV drivers.
4824 xirc2ps_cs= [NET,PCMCIA]
4826 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]]
projects / linux-2.6-microblaze.git / blob