1 acpi= [HW,ACPI,X86,ARM64]
2 Advanced Configuration and Power Interface
3 Format: { force | on | off | strict | noirq | rsdt |
5 force -- enable ACPI if default was off
6 on -- enable ACPI but allow fallback to DT [arm64]
7 off -- disable ACPI if default was on
8 noirq -- do not use ACPI for IRQ routing
9 strict -- Be less tolerant of platforms that are not
10 strictly ACPI specification compliant.
11 rsdt -- prefer RSDT over (default) XSDT
12 copy_dsdt -- copy DSDT to memory
13 For ARM64, ONLY "acpi=off", "acpi=on" or "acpi=force"
16 See also Documentation/power/runtime_pm.rst, pci=noacpi
18 acpi_apic_instance= [ACPI, IOAPIC]
20 2: use 2nd APIC table, if available
21 1,0: use 1st APIC table
24 acpi_backlight= [HW,ACPI]
25 { vendor | video | native | none }
26 If set to vendor, prefer vendor-specific driver
27 (e.g. thinkpad_acpi, sony_acpi, etc.) instead
28 of the ACPI video.ko driver.
29 If set to video, use the ACPI video.ko driver.
30 If set to native, use the device's native backlight mode.
31 If set to none, disable the ACPI backlight interface.
33 acpi_force_32bit_fadt_addr
34 force FADT to use 32 bit addresses rather than the
35 64 bit X_* addresses. Some firmware have broken 64
36 bit addresses for force ACPI ignore these and use
37 the older legacy 32 bit addresses.
39 acpica_no_return_repair [HW, ACPI]
40 Disable AML predefined validation mechanism
41 This mechanism can repair the evaluation result to make
42 the return objects more ACPI specification compliant.
43 This option is useful for developers to identify the
44 root cause of an AML interpreter issue when the issue
45 has something to do with the repair mechanism.
47 acpi.debug_layer= [HW,ACPI,ACPI_DEBUG]
48 acpi.debug_level= [HW,ACPI,ACPI_DEBUG]
50 CONFIG_ACPI_DEBUG must be enabled to produce any ACPI
51 debug output. Bits in debug_layer correspond to a
52 _COMPONENT in an ACPI source file, e.g.,
53 #define _COMPONENT ACPI_EVENTS
54 Bits in debug_level correspond to a level in
55 ACPI_DEBUG_PRINT statements, e.g.,
56 ACPI_DEBUG_PRINT((ACPI_DB_INFO, ...
57 The debug_level mask defaults to "info". See
58 Documentation/firmware-guide/acpi/debug.rst for more information about
59 debug layers and levels.
61 Enable processor driver info messages:
62 acpi.debug_layer=0x20000000
63 Enable AML "Debug" output, i.e., stores to the Debug
64 object while interpreting AML:
65 acpi.debug_layer=0xffffffff acpi.debug_level=0x2
66 Enable all messages related to ACPI hardware:
67 acpi.debug_layer=0x2 acpi.debug_level=0xffffffff
69 Some values produce so much output that the system is
70 unusable. The "log_buf_len" parameter may be useful
71 if you need to capture more output.
73 acpi_enforce_resources= [ACPI]
75 Check for resource conflicts between native drivers
76 and ACPI OperationRegions (SystemIO and SystemMemory
77 only). IO ports and memory declared in ACPI might be
78 used by the ACPI subsystem in arbitrary AML code and
79 can interfere with legacy drivers.
80 strict (default): access to resources claimed by ACPI
81 is denied; legacy drivers trying to access reserved
82 resources will fail to bind to device using them.
83 lax: access to resources claimed by ACPI is allowed;
84 legacy drivers trying to access reserved resources
85 will bind successfully but a warning message is logged.
86 no: ACPI OperationRegions are not marked as reserved,
87 no further checks are performed.
89 acpi_force_table_verification [HW,ACPI]
90 Enable table checksum verification during early stage.
91 By default, this is disabled due to x86 early mapping
94 acpi_irq_balance [HW,ACPI]
95 ACPI will balance active IRQs
98 acpi_irq_nobalance [HW,ACPI]
99 ACPI will not move active IRQs (default)
102 acpi_irq_isa= [HW,ACPI] If irq_balance, mark listed IRQs used by ISA
103 Format: <irq>,<irq>...
105 acpi_irq_pci= [HW,ACPI] If irq_balance, clear listed IRQs for
107 Format: <irq>,<irq>...
109 acpi_mask_gpe= [HW,ACPI]
110 Due to the existence of _Lxx/_Exx, some GPEs triggered
111 by unsupported hardware/firmware features can result in
112 GPE floodings that cannot be automatically disabled by
114 This facility can be used to prevent such uncontrolled
116 Format: <byte> or <bitmap-list>
118 acpi_no_auto_serialize [HW,ACPI]
119 Disable auto-serialization of AML methods
120 AML control methods that contain the opcodes to create
121 named objects will be marked as "Serialized" by the
122 auto-serialization feature.
123 This feature is enabled by default.
124 This option allows to turn off the feature.
126 acpi_no_memhotplug [ACPI] Disable memory hotplug. Useful for kdump
129 acpi_no_static_ssdt [HW,ACPI]
130 Disable installation of static SSDTs at early boot time
131 By default, SSDTs contained in the RSDT/XSDT will be
132 installed automatically and they will appear under
133 /sys/firmware/acpi/tables.
134 This option turns off this feature.
135 Note that specifying this option does not affect
136 dynamic table installation which will install SSDT
137 tables to /sys/firmware/acpi/tables/dynamic.
139 acpi_no_watchdog [HW,ACPI,WDT]
140 Ignore the ACPI-based watchdog interface (WDAT) and let
141 a native driver control the watchdog device instead.
143 acpi_rsdp= [ACPI,EFI,KEXEC]
144 Pass the RSDP address to the kernel, mostly used
145 on machines running EFI runtime service to boot the
146 second kernel for kdump.
148 acpi_os_name= [HW,ACPI] Tell ACPI BIOS the name of the OS
149 Format: To spoof as Windows 98: ="Microsoft Windows"
151 acpi_rev_override [ACPI] Override the _REV object to return 5 (instead
152 of 2 which is mandated by ACPI 6) as the supported ACPI
153 specification revision (when using this switch, it may
154 be necessary to carry out a cold reboot _twice_ in a
155 row to make it take effect on the platform firmware).
157 acpi_osi= [HW,ACPI] Modify list of supported OS interface strings
158 acpi_osi="string1" # add string1
159 acpi_osi="!string2" # remove string2
160 acpi_osi=!* # remove all strings
161 acpi_osi=! # disable all built-in OS vendor
163 acpi_osi=!! # enable all built-in OS vendor
165 acpi_osi= # disable all strings
167 'acpi_osi=!' can be used in combination with single or
168 multiple 'acpi_osi="string1"' to support specific OS
169 vendor string(s). Note that such command can only
170 affect the default state of the OS vendor strings, thus
171 it cannot affect the default state of the feature group
172 strings and the current state of the OS vendor strings,
173 specifying it multiple times through kernel command line
174 is meaningless. This command is useful when one do not
175 care about the state of the feature group strings which
176 should be controlled by the OSPM.
178 1. 'acpi_osi=! acpi_osi="Windows 2000"' is equivalent
179 to 'acpi_osi="Windows 2000" acpi_osi=!', they all
180 can make '_OSI("Windows 2000")' TRUE.
182 'acpi_osi=' cannot be used in combination with other
183 'acpi_osi=' command lines, the _OSI method will not
184 exist in the ACPI namespace. NOTE that such command can
185 only affect the _OSI support state, thus specifying it
186 multiple times through kernel command line is also
189 1. 'acpi_osi=' can make 'CondRefOf(_OSI, Local1)'
192 'acpi_osi=!*' can be used in combination with single or
193 multiple 'acpi_osi="string1"' to support specific
194 string(s). Note that such command can affect the
195 current state of both the OS vendor strings and the
196 feature group strings, thus specifying it multiple times
197 through kernel command line is meaningful. But it may
198 still not able to affect the final state of a string if
199 there are quirks related to this string. This command
200 is useful when one want to control the state of the
201 feature group strings to debug BIOS issues related to
204 1. 'acpi_osi="Module Device" acpi_osi=!*' can make
205 '_OSI("Module Device")' FALSE.
206 2. 'acpi_osi=!* acpi_osi="Module Device"' can make
207 '_OSI("Module Device")' TRUE.
208 3. 'acpi_osi=! acpi_osi=!* acpi_osi="Windows 2000"' is
210 'acpi_osi=!* acpi_osi=! acpi_osi="Windows 2000"'
212 'acpi_osi=!* acpi_osi="Windows 2000" acpi_osi=!',
213 they all will make '_OSI("Windows 2000")' TRUE.
216 Override the pmtimer bug detection: force the kernel
217 to assume that this machine's pmtimer latches its value
218 and always returns good values.
220 acpi_sci= [HW,ACPI] ACPI System Control Interrupt trigger mode
221 Format: { level | edge | high | low }
223 acpi_skip_timer_override [HW,ACPI]
224 Recognize and ignore IRQ0/pin2 Interrupt Override.
225 For broken nForce2 BIOS resulting in XT-PIC timer.
227 acpi_sleep= [HW,ACPI] Sleep options
228 Format: { s3_bios, s3_mode, s3_beep, s4_hwsig,
229 s4_nohwsig, old_ordering, nonvs,
230 sci_force_enable, nobl }
231 See Documentation/power/video.rst for information on
233 s3_beep is for debugging; it makes the PC's speaker beep
234 as soon as the kernel's real-mode entry point is called.
235 s4_hwsig causes the kernel to check the ACPI hardware
236 signature during resume from hibernation, and gracefully
237 refuse to resume if it has changed. This complies with
238 the ACPI specification but not with reality, since
239 Windows does not do this and many laptops do change it
240 on docking. So the default behaviour is to allow resume
241 and simply warn when the signature changes, unless the
242 s4_hwsig option is enabled.
243 s4_nohwsig prevents ACPI hardware signature from being
244 used (or even warned about) during resume.
245 old_ordering causes the ACPI 1.0 ordering of the _PTS
246 control method, with respect to putting devices into
247 low power states, to be enforced (the ACPI 2.0 ordering
248 of _PTS is used by default).
249 nonvs prevents the kernel from saving/restoring the
250 ACPI NVS memory during suspend/hibernation and resume.
251 sci_force_enable causes the kernel to set SCI_EN directly
252 on resume from S1/S3 (which is against the ACPI spec,
253 but some broken systems don't work without it).
254 nobl causes the internal blacklist of systems known to
255 behave incorrectly in some ways with respect to system
256 suspend and resume to be ignored (use wisely).
258 acpi_use_timer_override [HW,ACPI]
259 Use timer override. For some broken Nvidia NF5 boards
260 that require a timer override, but don't have HPET
262 add_efi_memmap [EFI; X86] Include EFI memory map in
263 kernel's map of available physical RAM.
266 { off | try_unsupported }
267 off: disable AGP support
268 try_unsupported: try to drive unsupported chipsets
269 (may crash computer or cause data corruption)
272 See Documentation/sound/alsa-configuration.rst
275 Allow the default userspace alignment fault handler
276 behaviour to be specified. Bit 0 enables warnings,
277 bit 1 enables fixups, and bit 2 sends a segfault.
279 align_va_addr= [X86-64]
280 Align virtual addresses by clearing slice [14:12] when
281 allocating a VMA at process creation time. This option
282 gives you up to 3% performance improvement on AMD F15h
283 machines (where it is enabled by default) for a
284 CPU-intensive style benchmark, and it can vary highly in
285 a microbenchmark depending on workload and compiler.
287 32: only for 32-bit processes
288 64: only for 64-bit processes
289 on: enable for both 32- and 64-bit processes
290 off: disable for both 32- and 64-bit processes
292 alloc_snapshot [FTRACE]
293 Allocate the ftrace snapshot buffer on boot up when the
294 main buffer is allocated. This is handy if debugging
295 and you need to use tracing_snapshot() on boot up, and
296 do not want to use tracing_snapshot_alloc() as it needs
297 to be done where GFP_KERNEL allocations are allowed.
299 allow_mismatched_32bit_el0 [ARM64]
300 Allow execve() of 32-bit applications and setting of the
301 PER_LINUX32 personality on systems where only a strict
302 subset of the CPUs support 32-bit EL0. When this
303 parameter is present, the set of CPUs supporting 32-bit
304 EL0 is indicated by /sys/devices/system/cpu/aarch32_el0
305 and hot-unplug operations may be restricted.
307 See Documentation/arm64/asymmetric-32bit.rst for more
310 amd_iommu= [HW,X86-64]
311 Pass parameters to the AMD IOMMU driver in the system.
313 fullflush - Deprecated, equivalent to iommu.strict=1
314 off - do not initialize any AMD IOMMU found in
316 force_isolation - Force device isolation for all
317 devices. The IOMMU driver is not
318 allowed anymore to lift isolation
319 requirements as needed. This option
320 does not override iommu=pt
321 force_enable - Force enable the IOMMU on platforms known
322 to be buggy with IOMMU enabled. Use this
324 pgtbl_v1 - Use v1 page table for DMA-API (Default).
325 pgtbl_v2 - Use v2 page table for DMA-API.
327 amd_iommu_dump= [HW,X86-64]
328 Enable AMD IOMMU driver option to dump the ACPI table
329 for AMD IOMMU. With this option enabled, AMD IOMMU
330 driver will print ACPI tables for AMD IOMMU during
331 IOMMU initialization.
333 amd_iommu_intr= [HW,X86-64]
334 Specifies one of the following AMD IOMMU interrupt
336 legacy - Use legacy interrupt remapping mode.
337 vapic - Use virtual APIC mode, which allows IOMMU
338 to inject interrupts directly into guest.
339 This mode requires kvm-amd.avic=1.
340 (Default when IOMMU HW support is present.)
342 amijoy.map= [HW,JOY] Amiga joystick support
343 Map of devices attached to JOY0DAT and JOY1DAT
345 See also Documentation/input/joydev/joystick.rst
347 analog.map= [HW,JOY] Analog joystick and gamepad support
348 Specifies type or capabilities of an analog joystick
349 connected to one of 16 gameports
350 Format: <type1>,<type2>,..<type16>
353 Power management functions (SPARCstation-4/5 + deriv.)
355 Disable APC CPU standby support. SPARCstation-Fox does
356 not play well with APC CPU idle - disable it if you have
357 APC and your system crashes randomly.
359 apic= [APIC,X86] Advanced Programmable Interrupt Controller
360 Change the output verbosity while booting
361 Format: { quiet (default) | verbose | debug }
362 Change the amount of debugging information output
363 when initialising the APIC and IO-APIC components.
364 For X86-32, this can also be used to specify an APIC
366 Format: apic=driver_name
367 Examples: apic=bigsmp
369 apic_extnmi= [APIC,X86] External NMI delivery setting
370 Format: { bsp (default) | all | none }
371 bsp: External NMI is delivered only to CPU 0
372 all: External NMIs are broadcast to all CPUs as a
374 none: External NMI is masked for all CPUs. This is
375 useful so that a dump capture kernel won't be
379 See Documentation/networking/ipv6.rst.
381 apm= [APM] Advanced Power Management
382 See header of arch/x86/kernel/apm_32.c.
384 apparmor= [APPARMOR] Disable or enable AppArmor at boot time
385 Format: { "0" | "1" }
386 See security/apparmor/Kconfig help text
389 Default value is set via kernel config option.
391 arcrimi= [HW,NET] ARCnet - "RIM I" (entirely mem-mapped) cards
392 Format: <io>,<irq>,<nodeID>
394 arm64.nobti [ARM64] Unconditionally disable Branch Target
395 Identification support
397 arm64.nopauth [ARM64] Unconditionally disable Pointer Authentication
400 arm64.nomte [ARM64] Unconditionally disable Memory Tagging Extension
403 arm64.nosve [ARM64] Unconditionally disable Scalable Vector
406 arm64.nosme [ARM64] Unconditionally disable Scalable Matrix
411 atarimouse= [HW,MOUSE] Atari Mouse
413 atkbd.extra= [HW] Enable extra LEDs and keys on IBM RapidAccess,
414 EzKey and similar keyboards
416 atkbd.reset= [HW] Reset keyboard during initialization
418 atkbd.set= [HW] Select keyboard code set
419 Format: <int> (2 = AT (default), 3 = PS/2)
421 atkbd.scroll= [HW] Enable scroll wheel on MS Office and similar
424 atkbd.softraw= [HW] Choose between synthetic and real raw mode
425 Format: <bool> (0 = real, 1 = synthetic (default))
427 atkbd.softrepeat= [HW]
428 Use software keyboard repeat
430 audit= [KNL] Enable the audit sub-system
431 Format: { "0" | "1" | "off" | "on" }
432 0 | off - kernel audit is disabled and can not be
433 enabled until the next reboot
434 unset - kernel audit is initialized but disabled and
435 will be fully enabled by the userspace auditd.
436 1 | on - kernel audit is initialized and partially
437 enabled, storing at most audit_backlog_limit
438 messages in RAM until it is fully enabled by the
442 audit_backlog_limit= [KNL] Set the audit queue size limit.
443 Format: <int> (must be >=0)
446 bau= [X86_UV] Enable the BAU on SGI UV. The default
447 behavior is to disable the BAU (i.e. bau=0).
448 Format: { "0" | "1" }
451 unset - Disable the BAU.
453 baycom_epp= [HW,AX25]
456 baycom_par= [HW,AX25] BayCom Parallel Port AX.25 Modem
458 See header of drivers/net/hamradio/baycom_par.c.
460 baycom_ser_fdx= [HW,AX25]
461 BayCom Serial Port AX.25 Modem (Full Duplex Mode)
462 Format: <io>,<irq>,<mode>[,<baud>]
463 See header of drivers/net/hamradio/baycom_ser_fdx.c.
465 baycom_ser_hdx= [HW,AX25]
466 BayCom Serial Port AX.25 Modem (Half Duplex Mode)
467 Format: <io>,<irq>,<mode>
468 See header of drivers/net/hamradio/baycom_ser_hdx.c.
471 Disable BERT OS support on buggy BIOSes.
473 bgrt_disable [ACPI][X86]
474 Disable BGRT to avoid flickering OEM logo.
476 blkdevparts= Manual partition parsing of block device(s) for
477 embedded devices based on command line input.
478 See Documentation/block/cmdline-partition.rst
480 boot_delay= Milliseconds to delay each printk during boot.
481 Only works if CONFIG_BOOT_PRINTK_DELAY is enabled,
482 and you may also have to specify "lpj=". Boot_delay
483 values larger than 10 seconds (10000) are assumed
484 erroneous and ignored.
488 Extended command line options can be added to an initrd
489 and this will cause the kernel to look for it.
491 See Documentation/admin-guide/bootconfig.rst
493 bttv.card= [HW,V4L] bttv (bt848 + bt878 based grabber cards)
494 bttv.radio= Most important insmod options are available as
496 bttv.pll= See Documentation/admin-guide/media/bttv.rst
499 bulk_remove=off [PPC] This parameter disables the use of the pSeries
500 firmware feature for flushing multiple hpte entries
503 c101= [NET] Moxa C101 synchronous serial card
505 cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection.
506 Sometimes CPU hardware bugs make them report the cache
507 size incorrectly. The kernel will attempt work arounds
508 to fix known problems, but for some CPUs it is not
509 possible to determine what the correct size should be.
510 This option provides an override for these situations.
513 [NET] Specifies amount of time (in seconds) that
514 the kernel should wait for a network carrier. By default
515 it waits 120 seconds.
517 ca_keys= [KEYS] This parameter identifies a specific key(s) on
518 the system trusted keyring to be used for certificate
520 format: { id:<keyid> | builtin }
522 cca= [MIPS] Override the kernel pages' cache coherency
523 algorithm. Accepted values range from 0 to 7
524 inclusive. See arch/mips/include/asm/pgtable-bits.h
525 for platform specific values (SB1, Loongson3 and
528 ccw_timeout_log [S390]
529 See Documentation/s390/common_io.rst for details.
531 cgroup_disable= [KNL] Disable a particular controller or optional feature
532 Format: {name of the controller(s) or feature(s) to disable}
533 The effects of cgroup_disable=foo are:
534 - foo isn't auto-mounted if you mount all cgroups in
536 - foo isn't visible as an individually mountable
538 - if foo is an optional feature then the feature is
539 disabled and corresponding cgroup files are not
541 {Currently only "memory" controller deal with this and
542 cut the overhead, others just disable the usage. So
543 only cgroup_disable=memory is actually worthy}
544 Specifying "pressure" disables per-cgroup pressure
545 stall information accounting feature
547 cgroup_no_v1= [KNL] Disable cgroup controllers and named hierarchies in v1
548 Format: { { controller | "all" | "named" }
549 [,{ controller | "all" | "named" }...] }
550 Like cgroup_disable, but only applies to cgroup v1;
551 the blacklisted controllers remain available in cgroup2.
552 "all" blacklists all controllers and "named" disables
553 named mounts. Specifying both "all" and "named" disables
556 cgroup.memory= [KNL] Pass options to the cgroup memory controller.
558 nosocket -- Disable socket memory accounting.
559 nokmem -- Disable kernel memory accounting.
560 nobpf -- Disable BPF memory accounting.
562 checkreqprot= [SELINUX] Set initial checkreqprot flag value.
563 Format: { "0" | "1" }
564 See security/selinux/Kconfig help text.
565 0 -- check protection applied by kernel (includes
566 any implied execute protection).
567 1 -- check protection requested by application.
568 Default value is set via a kernel config option.
569 Value can be changed at runtime via
570 /sys/fs/selinux/checkreqprot.
571 Setting checkreqprot to 1 is deprecated.
574 See Documentation/s390/common_io.rst for details.
576 clearcpuid=X[,X...] [X86]
577 Disable CPUID feature X for the kernel. See
578 arch/x86/include/asm/cpufeatures.h for the valid bit
579 numbers X. Note the Linux-specific bits are not necessarily
580 stable over kernel options, but the vendor-specific
582 X can also be a string as appearing in the flags: line
583 in /proc/cpuinfo which does not have the above
584 instability issue. However, not all features have names
586 Note that using this option will taint your kernel.
587 Also note that user programs calling CPUID directly
588 or using the feature without checking anything
589 will still see it. This just prevents it from
590 being used by the kernel or shown in /proc/cpuinfo.
591 Also note the kernel might malfunction if you disable
596 Prevents the clock framework from automatically gating
597 clocks that have not been explicitly enabled by a Linux
598 device driver but are enabled in hardware at reset or
599 by the bootloader/firmware. Note that this does not
600 force such clocks to be always-on nor does it reserve
601 those clocks in any way. This parameter is useful for
602 debug and development, but should not be needed on a
603 platform with proper driver support. For more
604 information, see Documentation/driver-api/clk.rst.
606 clock= [BUGS=X86-32, HW] gettimeofday clocksource override.
608 Forces specified clocksource (if available) to be used
609 when calculating gettimeofday(). If specified
610 clocksource is not available, it defaults to PIT.
611 Format: { pit | tsc | cyclone | pmtmr }
613 clocksource= Override the default clocksource
615 Override the default clocksource and use the clocksource
616 with the name specified.
617 Some clocksource names to choose from, depending on
619 [all] jiffies (this is the base, fallback clocksource)
621 [ARM] imx_timer1,OSTS,netx_timer,mpu_timer2,
622 pxa_timer,timer3,32k_counter,timer0_1
623 [X86-32] pit,hpet,tsc;
624 scx200_hrt on Geode; cyclone on IBM x440
632 clocksource.arm_arch_timer.evtstrm=
635 Enable/disable the eventstream feature of the ARM
636 architected timer so that code using WFE-based polling
637 loops can be debugged more effectively on production
640 clocksource.max_cswd_read_retries= [KNL]
641 Number of clocksource_watchdog() retries due to
642 external delays before the clock will be marked
643 unstable. Defaults to two retries, that is,
644 three attempts to read the clock under test.
646 clocksource.verify_n_cpus= [KNL]
647 Limit the number of CPUs checked for clocksources
648 marked with CLOCK_SOURCE_VERIFY_PERCPU that
649 are marked unstable due to excessive skew.
650 A negative value says to check all CPUs, while
651 zero says not to check any. Values larger than
652 nr_cpu_ids are silently truncated to nr_cpu_ids.
653 The actual CPUs are chosen randomly, with
654 no replacement if the same CPU is chosen twice.
656 clocksource-wdtest.holdoff= [KNL]
657 Set the time in seconds that the clocksource
658 watchdog test waits before commencing its tests.
659 Defaults to zero when built as a module and to
660 10 seconds when built into the kernel.
662 cma=nn[MG]@[start[MG][-end[MG]]]
664 Sets the size of kernel global memory area for
665 contiguous memory allocations and optionally the
666 placement constraint by the physical address range of
667 memory allocations. A value of 0 disables CMA
668 altogether. For more information, see
669 kernel/dma/contiguous.c
673 Sets the size of kernel per-numa memory area for
674 contiguous memory allocations. A value of 0 disables
675 per-numa CMA altogether. And If this option is not
676 specified, the default value is 0.
677 With per-numa CMA enabled, DMA users on node nid will
678 first try to allocate buffer from the pernuma area
679 which is located in node nid, if the allocation fails,
680 they will fallback to the global default memory area.
682 cmo_free_hint= [PPC] Format: { yes | no }
683 Specify whether pages are marked as being inactive
684 when they are freed. This is used in CMO environments
685 to determine OS memory pressure for page stealing by
689 coherent_pool=nn[KMG] [ARM,KNL]
690 Sets the size of memory pool for coherent, atomic dma
691 allocations, by default set to 256K.
693 com20020= [HW,NET] ARCnet - COM20020 chipset
695 <io>[,<irq>[,<nodeID>[,<backplane>[,<ckp>[,<timeout>]]]]]
697 com90io= [HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers)
701 ARCnet - COM90xx chipset (memory-mapped buffers)
702 Format: <io>[,<irq>[,<memstart>]]
704 condev= [HW,S390] console device
707 con3215_drop= [S390] 3215 console drop mode.
709 When set to true, drop data on the 3215 console when
710 the console buffer is full. In this case the
711 operator using a 3270 terminal emulator (for example
712 x3270) does not have to enter the clear key for the
713 console output to advance and the kernel to continue.
714 This leads to a much faster boot time when a 3270
715 terminal emulator is active. If no 3270 terminal
716 emulator is used, this parameter has no effect.
718 console= [KNL] Output console device and options.
720 tty<n> Use the virtual console device <n>.
724 Use the specified serial port. The options are of
725 the form "bbbbpnf", where "bbbb" is the baud rate,
726 "p" is parity ("n", "o", or "e"), "n" is number of
727 bits, and "f" is flow control ("r" for RTS or
728 omit it). Default is "9600n8".
730 See Documentation/admin-guide/serial-console.rst for more
732 Documentation/networking/netconsole.rst for an
735 uart[8250],io,<addr>[,options]
736 uart[8250],mmio,<addr>[,options]
737 uart[8250],mmio16,<addr>[,options]
738 uart[8250],mmio32,<addr>[,options]
739 uart[8250],0x<addr>[,options]
740 Start an early, polled-mode console on the 8250/16550
741 UART at the specified I/O port or MMIO address,
742 switching to the matching ttyS device later.
743 MMIO inter-register address stride is either 8-bit
744 (mmio), 16-bit (mmio16), or 32-bit (mmio32).
745 If none of [io|mmio|mmio16|mmio32], <addr> is assumed
746 to be equivalent to 'mmio'. 'options' are specified in
747 the same format described for ttyS above; if unspecified,
748 the h/w is not re-initialized.
750 hvc<n> Use the hypervisor console device <n>. This is for
751 both Xen and PowerPC hypervisors.
754 Use to disable console output, i.e., to have kernel
755 console messages discarded.
756 This must be the only console= parameter used on the
759 If the device connected to the port is not a TTY but a braille
760 device, prepend "brl," before the device type, for instance
762 For now, only VisioBraille is supported.
765 [KNL] Change console messages format
767 By default we print messages on consoles in
768 "[time stamp] text\n" format (time stamp may not be
769 printed, depending on CONFIG_PRINTK_TIME or
770 `printk_time' param).
772 Switch to syslog format: "<%u>[time stamp] text\n"
773 IOW, each message will have a facility and loglevel
774 prefix. The format is similar to one used by syslog()
775 syscall, or to executing "dmesg -S --raw" or to reading
778 consoleblank= [KNL] The console blank (screen saver) timeout in
779 seconds. A value of 0 disables the blank timer.
783 [KNL] Change the default value for
784 /proc/<pid>/coredump_filter.
785 See also Documentation/filesystems/proc.rst.
787 coresight_cpu_debug.enable
790 Enable/disable the CPU sampling based debugging.
791 0: default value, disable debugging
792 1: enable debugging at boot time
794 cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver
796 <first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>]
798 cpu0_hotplug [X86] Turn on CPU0 hotplug feature when
799 CONFIG_BOOTPARAM_HOTPLUG_CPU0 is off.
800 Some features depend on CPU0. Known dependencies are:
801 1. Resume from suspend/hibernate depends on CPU0.
802 Suspend/hibernate will fail if CPU0 is offline and you
803 need to online CPU0 before suspend/hibernate.
804 2. PIC interrupts also depend on CPU0. CPU0 can't be
805 removed if a PIC interrupt is detected.
806 It's said poweroff/reboot may depend on CPU0 on some
807 machines although I haven't seen such issues so far
808 after CPU0 is offline on a few tested machines.
809 If the dependencies are under your control, you can
810 turn on cpu0_hotplug.
812 cpuidle.off=1 [CPU_IDLE]
813 disable the cpuidle sub-system
816 [CPU_IDLE] Name of the cpuidle governor to use.
818 cpufreq.off=1 [CPU_FREQ]
819 disable the cpufreq sub-system
821 cpufreq.default_governor=
822 [CPU_FREQ] Name of the default cpufreq governor or
823 policy to use. This governor must be registered in the
824 kernel before the cpufreq driver probes.
827 [X86] Delay for N microsec between assert and de-assert
828 of APIC INIT to start processors. This delay occurs
829 on every CPU online, such as boot, and resume from suspend.
832 crash_kexec_post_notifiers
833 Run kdump after running panic-notifiers and dumping
834 kmsg. This only for the users who doubt kdump always
835 succeeds in any situation.
836 Note that this also increases risks of kdump failure,
837 because some panic notifiers can make the crashed
838 kernel more unstable.
840 crashkernel=size[KMG][@offset[KMG]]
841 [KNL] Using kexec, Linux can switch to a 'crash kernel'
842 upon panic. This parameter reserves the physical
843 memory region [offset, offset + size] for that kernel
844 image. If '@offset' is omitted, then a suitable offset
845 is selected automatically.
846 [KNL, X86-64, ARM64] Select a region under 4G first, and
847 fall back to reserve region above 4G when '@offset'
848 hasn't been specified.
849 See Documentation/admin-guide/kdump/kdump.rst for further details.
851 crashkernel=range1:size1[,range2:size2,...][@offset]
852 [KNL] Same as above, but depends on the memory
853 in the running system. The syntax of range is
854 start-[end] where start and end are both
855 a memory unit (amount[KMG]). See also
856 Documentation/admin-guide/kdump/kdump.rst for an example.
858 crashkernel=size[KMG],high
859 [KNL, X86-64, ARM64] range could be above 4G. Allow kernel
860 to allocate physical memory region from top, so could
861 be above 4G if system have more than 4G ram installed.
862 Otherwise memory region will be allocated below 4G, if
864 It will be ignored if crashkernel=X is specified.
865 crashkernel=size[KMG],low
866 [KNL, X86-64, ARM64] range under 4G. When crashkernel=X,high
867 is passed, kernel could allocate physical memory region
868 above 4G, that cause second kernel crash on system
869 that require some amount of low memory, e.g. swiotlb
870 requires at least 64M+32K low memory, also enough extra
871 low memory is needed to make sure DMA buffers for 32-bit
872 devices won't run out. Kernel would try to allocate
873 default size of memory below 4G automatically. The default
874 size is platform dependent.
875 --> x86: max(swiotlb_size_or_default() + 8MiB, 256MiB)
877 This one lets the user specify own low range under 4G
878 for second kernel instead.
879 0: to disable low allocation.
880 It will be ignored when crashkernel=X,high is not used
881 or memory reserved is below 4G.
884 [KNL] Disable crypto self-tests
889 cs89x0_media= [HW,NET]
890 Format: { rj45 | aui | bnc }
892 csdlock_debug= [KNL] Enable debug add-ons of cross-CPU function call
893 handling. When switched on, additional debug data is
894 printed to the console in case a hanging CPU is
895 detected, and that CPU is pinged again in order to try
896 to resolve the hang situation.
897 0: disable csdlock debugging (default)
898 1: enable basic csdlock debugging (minor impact)
899 ext: enable extended csdlock debugging (more impact,
903 See header of drivers/s390/block/dasd_devmap.c.
905 db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port
906 (one device per port)
907 Format: <port#>,<type>
908 See also Documentation/input/devices/joystick-parport.rst
910 debug [KNL] Enable kernel debugging (events log level).
913 [KNL] Enable printing [hashed] pointers early in the
914 boot sequence. If enabled, we use a weak hash instead
915 of siphash to hash pointers. Use this option if you are
916 seeing instances of '(___ptrval___)') and need to see a
917 value (hashed pointer) instead. Cryptographically
918 insecure, please do not use on production kernels.
921 [KNL] verbose locking self-tests
923 Print debugging info while doing the locking API
925 Bitmask for the various LOCKTYPE_ tests. Defaults to 0
926 (no extra messages), setting it to -1 (all bits set)
927 will print _a_lot_ more information - normally only
928 useful to lockdep developers.
930 debug_objects [KNL] Enable object debugging
932 debug_guardpage_minorder=
933 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this
934 parameter allows control of the order of pages that will
935 be intentionally kept free (and hence protected) by the
936 buddy allocator. Bigger value increase the probability
937 of catching random memory corruption, but reduce the
938 amount of memory for normal system use. The maximum
939 possible value is MAX_ORDER/2. Setting this parameter
940 to 1 or 2 should be enough to identify most random
941 memory corruption problems caused by bugs in kernel or
942 driver code when a CPU writes to (or reads from) a
943 random memory location. Note that there exists a class
944 of memory corruptions problems caused by buggy H/W or
945 F/W or by drivers badly programming DMA (basically when
946 memory is written at bus level and the CPU MMU is
947 bypassed) which are not detectable by
948 CONFIG_DEBUG_PAGEALLOC, hence this option will not help
949 tracking down these problems.
952 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this parameter
953 enables the feature at boot time. By default, it is
954 disabled and the system will work mostly the same as a
955 kernel built without CONFIG_DEBUG_PAGEALLOC.
956 Note: to get most of debug_pagealloc error reports, it's
957 useful to also enable the page_owner functionality.
958 on: enable the feature
960 debugfs= [KNL] This parameter enables what is exposed to userspace
961 and debugfs internal clients.
962 Format: { on, no-mount, off }
963 on: All functions are enabled.
965 Filesystem is not registered but kernel clients can
966 access APIs and a crashkernel can be used to read
967 its content. There is nothing to mount.
968 off: Filesystem is not registered and clients
969 get a -EPERM as result when trying to register files
970 or directories within debugfs.
971 This is equivalent of the runtime functionality if
972 debugfs was not enabled in the kernel at all.
973 Default value is set in build-time with a kernel configuration.
975 debugpat [X86] Enable PAT debugging
978 [HW] The size of the default HugeTLB page. This is
979 the size represented by the legacy /proc/ hugepages
980 APIs. In addition, this is the default hugetlb size
981 used for shmget(), mmap() and mounting hugetlbfs
982 filesystems. If not specified, defaults to the
983 architecture's default huge page size. Huge page
984 sizes are architecture dependent. See also
985 Documentation/admin-guide/mm/hugetlbpage.rst.
988 deferred_probe_timeout=
989 [KNL] Debugging option to set a timeout in seconds for
990 deferred probe to give up waiting on dependencies to
991 probe. Only specific dependencies (subsystems or
992 drivers) that have opted in will be ignored. A timeout
993 of 0 will timeout at the end of initcalls. If the time
994 out hasn't expired, it'll be restarted by each
995 successful driver registration. This option will also
996 dump out devices still on the deferred probe list after
999 delayacct [KNL] Enable per-task delay accounting
1001 dell_smm_hwmon.ignore_dmi=
1002 [HW] Continue probing hardware even if DMI data
1003 indicates that the driver is running on unsupported
1006 dell_smm_hwmon.force=
1007 [HW] Activate driver even if SMM BIOS signature does
1008 not match list of supported models and enable otherwise
1009 blacklisted features.
1011 dell_smm_hwmon.power_status=
1012 [HW] Report power status in /proc/i8k
1013 (disabled by default).
1015 dell_smm_hwmon.restricted=
1016 [HW] Allow controlling fans only if SYS_ADMIN
1019 dell_smm_hwmon.fan_mult=
1020 [HW] Factor to multiply fan speed with.
1022 dell_smm_hwmon.fan_max=
1023 [HW] Maximum configurable fan speed.
1026 Format: { on | off | def_only | inf_only | always }
1027 on: s390 zlib hardware support for compression on
1028 level 1 and decompression (default)
1029 off: No s390 zlib hardware support
1030 def_only: s390 zlib hardware support for deflate
1031 only (compression on level 1)
1032 inf_only: s390 zlib hardware support for inflate
1033 only (decompression)
1034 always: Same as 'on' but ignores the selected compression
1035 level always using hardware support (used for debugging)
1037 dhash_entries= [KNL]
1038 Set number of hash buckets for dentry cache.
1040 disable_1tb_segments [PPC]
1041 Disables the use of 1TB hash page table segments. This
1042 causes the kernel to fall back to 256MB segments which
1043 can be useful when debugging issues that require an SLB
1047 See Documentation/networking/ipv6.rst.
1050 Disable RADIX MMU mode on POWER9
1053 Disable TLBIE instruction. Currently does not work
1054 with KVM, with HASH MMU, or with coherent accelerators.
1056 disable_cpu_apicid= [X86,APIC,SMP]
1058 The number of initial APIC ID for the
1059 corresponding CPU to be disabled at boot,
1060 mostly used for the kdump 2nd kernel to
1061 disable BSP to wake up multiple CPUs without
1062 causing system reset or hang due to sending
1063 INIT from AP to BSP.
1065 disable_ddw [PPC/PSERIES]
1066 Disable Dynamic DMA Window support. Use this
1067 to workaround buggy firmware.
1069 disable_ipv6= [IPV6]
1070 See Documentation/networking/ipv6.rst.
1072 disable_mtrr_cleanup [X86]
1073 The kernel tries to adjust MTRR layout from continuous
1074 to discrete, to make X server driver able to add WB
1075 entry later. This parameter disables that.
1077 disable_mtrr_trim [X86, Intel and AMD only]
1078 By default the kernel will trim any uncacheable
1079 memory out of your available memory pool based on
1080 MTRR settings. This parameter disables that behavior,
1081 possibly causing your machine to run very slowly.
1083 disable_timer_pin_1 [X86]
1084 Disable PIN 1 of APIC timer
1085 Can be useful to work around chipset bugs.
1087 dis_ucode_ldr [X86] Disable the microcode loader.
1089 dma_debug=off If the kernel is compiled with DMA_API_DEBUG support,
1090 this option disables the debugging code at boot.
1092 dma_debug_entries=<number>
1093 This option allows to tune the number of preallocated
1094 entries for DMA-API debugging code. One entry is
1095 required per DMA-API allocation. Use this if the
1096 DMA-API debugging code disables itself because the
1097 architectural default is too low.
1099 dma_debug_driver=<driver_name>
1100 With this option the DMA-API debugging driver
1101 filter feature can be enabled at boot time. Just
1102 pass the driver to filter for as the parameter.
1103 The filter can be disabled or changed to another
1104 driver later using sysfs.
1106 driver_async_probe= [KNL]
1107 List of driver names to be probed asynchronously. *
1108 matches with all driver names. If * is specified, the
1109 rest of the listed driver names are those that will NOT
1111 Format: <driver_name1>,<driver_name2>...
1113 drm.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>]
1114 Broken monitors, graphic adapters, KVMs and EDIDless
1115 panels may send no or incorrect EDID data sets.
1116 This parameter allows to specify an EDID data sets
1117 in the /lib/firmware directory that are used instead.
1118 Generic built-in EDID data sets are used, if one of
1119 edid/1024x768.bin, edid/1280x1024.bin,
1120 edid/1680x1050.bin, or edid/1920x1080.bin is given
1121 and no file with the same name exists. Details and
1122 instructions how to build your own EDID data are
1123 available in Documentation/admin-guide/edid.rst. An EDID
1124 data set will only be used for a particular connector,
1125 if its name and a colon are prepended to the EDID
1126 name. Each connector may use a unique EDID data
1127 set by separating the files with a comma. An EDID
1128 data set with no connector name will be used for
1129 any connectors not explicitly specified.
1134 Format: {"off" | "known"}
1135 Control how the dt_cpu_ftrs device-tree binding is
1136 used for CPU feature discovery and setup (if it
1138 off: Do not use it, fall back to legacy cpu table.
1139 known: Do not pass through unknown features to guests
1140 or userspace, only those that the kernel is aware of.
1142 dump_apple_properties [X86]
1143 Dump name and content of EFI device properties on
1144 x86 Macs. Useful for driver authors to determine
1145 what data is available or for reverse-engineering.
1147 dyndbg[="val"] [KNL,DYNAMIC_DEBUG]
1148 <module>.dyndbg[="val"]
1149 Enable debug messages at boot time. See
1150 Documentation/admin-guide/dynamic-debug-howto.rst
1153 early_ioremap_debug [KNL]
1154 Enable debug messages in early_ioremap support. This
1155 is useful for tracking down temporary early mappings
1156 which are not unmapped.
1158 earlycon= [KNL] Output early console device and options.
1160 When used with no options, the early console is
1161 determined by stdout-path property in device tree's
1162 chosen node or the ACPI SPCR table if supported by
1165 cdns,<addr>[,options]
1166 Start an early, polled-mode console on a Cadence
1167 (xuartps) serial port at the specified address. Only
1168 supported option is baud rate. If baud rate is not
1169 specified, the serial port must already be setup and
1172 uart[8250],io,<addr>[,options[,uartclk]]
1173 uart[8250],mmio,<addr>[,options[,uartclk]]
1174 uart[8250],mmio32,<addr>[,options[,uartclk]]
1175 uart[8250],mmio32be,<addr>[,options[,uartclk]]
1176 uart[8250],0x<addr>[,options]
1177 Start an early, polled-mode console on the 8250/16550
1178 UART at the specified I/O port or MMIO address.
1179 MMIO inter-register address stride is either 8-bit
1180 (mmio) or 32-bit (mmio32 or mmio32be).
1181 If none of [io|mmio|mmio32|mmio32be], <addr> is assumed
1182 to be equivalent to 'mmio'. 'options' are specified
1183 in the same format described for "console=ttyS<n>"; if
1184 unspecified, the h/w is not initialized. 'uartclk' is
1185 the uart clock frequency; if unspecified, it is set
1186 to 'BASE_BAUD' * 16.
1190 Start an early, polled-mode console on a pl011 serial
1191 port at the specified address. The pl011 serial port
1192 must already be setup and configured. Options are not
1193 yet supported. If 'mmio32' is specified, then only
1194 the driver will use only 32-bit accessors to read/write
1195 the device registers.
1198 Start an early console on a litex serial port at the
1199 specified address. The serial port must already be
1200 setup and configured. Options are not yet supported.
1203 Start an early, polled-mode console on a meson serial
1204 port at the specified address. The serial port must
1205 already be setup and configured. Options are not yet
1209 Start an early, polled-mode console on an msm serial
1210 port at the specified address. The serial port
1211 must already be setup and configured. Options are not
1214 msm_serial_dm,<addr>
1215 Start an early, polled-mode console on an msm serial
1216 dm port at the specified address. The serial port
1217 must already be setup and configured. Options are not
1221 Start an early, polled-mode console on a serial port
1222 of an Actions Semi SoC, such as S500 or S900, at the
1223 specified address. The serial port must already be
1224 setup and configured. Options are not yet supported.
1227 Start an early, polled-mode console on a serial port
1228 of an RDA Micro SoC, such as RDA8810PL, at the
1229 specified address. The serial port must already be
1230 setup and configured. Options are not yet supported.
1233 Use RISC-V SBI (Supervisor Binary Interface) for early
1236 smh Use ARM semihosting calls for early console.
1244 Use early console provided by serial driver available
1245 on Samsung SoCs, requires selecting proper type and
1246 a correct base address of the selected UART port. The
1247 serial port must already be setup and configured.
1248 Options are not yet supported.
1251 Start an early, polled-mode console on a lantiq serial
1252 (lqasc) port at the specified address. The serial port
1253 must already be setup and configured. Options are not
1258 Use early console provided by Freescale LP UART driver
1259 found on Freescale Vybrid and QorIQ LS1021A processors.
1260 A valid base address must be provided, and the serial
1261 port must already be setup and configured.
1265 Start an early, polled-mode, output-only console on the
1266 Freescale i.MX UART at the specified address. The UART
1267 must already be setup and configured.
1270 Start an early, polled-mode console on the
1271 Armada 3700 serial port at the specified
1272 address. The serial port must already be setup
1273 and configured. Options are not yet supported.
1276 Start an early, polled-mode console on a Qualcomm
1277 Generic Interface (GENI) based serial port at the
1278 specified address. The serial port must already be
1279 setup and configured. Options are not yet supported.
1282 Start an early, unaccelerated console on the EFI
1283 memory mapped framebuffer (if available). On cache
1284 coherent non-x86 systems that use system memory for
1285 the framebuffer, pass the 'ram' option so that it is
1286 mapped with the correct attributes.
1289 Use early console provided by Freescale LINFlexD UART
1290 serial driver for NXP S32V234 SoCs. A valid base
1291 address must be provided, and the serial port must
1292 already be setup and configured.
1294 earlyprintk= [X86,SH,ARM,M68k,S390]
1298 earlyprintk=serial[,ttySn[,baudrate]]
1299 earlyprintk=serial[,0x...[,baudrate]]
1300 earlyprintk=ttySn[,baudrate]
1301 earlyprintk=dbgp[debugController#]
1302 earlyprintk=pciserial[,force],bus:device.function[,baudrate]
1303 earlyprintk=xdbc[xhciController#]
1305 earlyprintk is useful when the kernel crashes before
1306 the normal console is initialized. It is not enabled by
1307 default because it has some cosmetic problems.
1309 Append ",keep" to not disable it when the real console
1312 Only one of vga, serial, or usb debug port can
1315 Currently only ttyS0 and ttyS1 may be specified by
1316 name. Other I/O ports may be explicitly specified
1317 on some architectures (x86 and arm at least) by
1318 replacing ttySn with an I/O port address, like this:
1319 earlyprintk=serial,0x1008,115200
1320 You can find the port for a given device in
1321 /proc/tty/driver/serial:
1322 2: uart:ST16650V2 port:00001008 irq:18 ...
1324 Interaction with the standard serial driver is not
1327 The VGA output is eventually overwritten by
1330 The xen option can only be used in Xen domains.
1332 The sclp output can only be used on s390.
1334 The optional "force" to "pciserial" enables use of a
1335 PCI device even when its classcode is not of the
1338 edac_report= [HW,EDAC] Control how to report EDAC event
1339 Format: {"on" | "off" | "force"}
1340 on: enable EDAC to report H/W event. May be overridden
1341 by other higher priority error reporting module.
1342 off: disable H/W event reporting through EDAC.
1343 force: enforce the use of EDAC to report H/W event.
1347 Format: {"off" | "on" | "skip[mbr]"}
1350 Format: { "debug", "disable_early_pci_dma",
1351 "nochunk", "noruntime", "nosoftreserve",
1352 "novamap", "no_disable_early_pci_dma" }
1353 debug: enable misc debug output.
1354 disable_early_pci_dma: disable the busmaster bit on all
1355 PCI bridges while in the EFI boot stub.
1356 nochunk: disable reading files in "chunks" in the EFI
1357 boot stub, as chunking can cause problems with some
1358 firmware implementations.
1359 noruntime : disable EFI runtime services support
1360 nosoftreserve: The EFI_MEMORY_SP (Specific Purpose)
1361 attribute may cause the kernel to reserve the
1362 memory range for a memory mapping driver to
1363 claim. Specify efi=nosoftreserve to disable this
1364 reservation and treat the memory by its base type
1365 (i.e. EFI_CONVENTIONAL_MEMORY / "System RAM").
1366 novamap: do not call SetVirtualAddressMap().
1367 no_disable_early_pci_dma: Leave the busmaster bit set
1368 on all PCI bridges while in the EFI boot stub
1370 efi_no_storage_paranoia [EFI; X86]
1371 Using this parameter you can use more than 50% of
1372 your efi variable storage. Use this parameter only if
1373 you are really sure that your UEFI does sane gc and
1374 fulfills the spec otherwise your board may brick.
1376 efi_fake_mem= nn[KMG]@ss[KMG]:aa[,nn[KMG]@ss[KMG]:aa,..] [EFI; X86]
1377 Add arbitrary attribute to specific memory range by
1378 updating original EFI memory map.
1379 Region of memory which aa attribute is added to is
1382 If efi_fake_mem=2G@4G:0x10000,2G@0x10a0000000:0x10000
1383 is specified, EFI_MEMORY_MORE_RELIABLE(0x10000)
1384 attribute is added to range 0x100000000-0x180000000 and
1385 0x10a0000000-0x1120000000.
1387 If efi_fake_mem=8G@9G:0x40000 is specified, the
1388 EFI_MEMORY_SP(0x40000) attribute is added to
1389 range 0x240000000-0x43fffffff.
1391 Using this parameter you can do debugging of EFI memmap
1392 related features. For example, you can do debugging of
1393 Address Range Mirroring feature even if your box
1394 doesn't support it, or mark specific memory as
1397 efivar_ssdt= [EFI; X86] Name of an EFI variable that contains an SSDT
1398 that is to be dynamically loaded by Linux. If there are
1399 multiple variables with the same name but with different
1400 vendor GUIDs, all of them will be loaded. See
1401 Documentation/admin-guide/acpi/ssdt-overlays.rst for details.
1404 eisa_irq_edge= [PARISC,HW]
1405 See header of drivers/parisc/eisa.c.
1407 ekgdboc= [X86,KGDB] Allow early kernel console debugging
1410 This is designed to be used in conjunction with
1411 the boot argument: earlyprintk=vga
1413 This parameter works in place of the kgdboc parameter
1414 but can only be used if the backing tty is available
1415 very early in the boot process. For early debugging
1416 via a serial port see kgdboc_earlycon instead.
1419 See comment before function elanfreq_setup() in
1420 arch/x86/kernel/cpu/cpufreq/elanfreq.c.
1422 elfcorehdr=[size[KMG]@]offset[KMG] [IA64,PPC,SH,X86,S390]
1423 Specifies physical address of start of kernel core
1424 image elf header and optionally the size. Generally
1425 kexec loader will pass this option to capture kernel.
1426 See Documentation/admin-guide/kdump/kdump.rst for details.
1428 enable_mtrr_cleanup [X86]
1429 The kernel tries to adjust MTRR layout from continuous
1430 to discrete, to make X server driver able to add WB
1431 entry later. This parameter enables that.
1433 enable_timer_pin_1 [X86]
1434 Enable PIN 1 of APIC timer
1435 Can be useful to work around chipset bugs
1436 (in particular on some ATI chipsets).
1437 The kernel tries to set a reasonable default.
1439 enforcing= [SELINUX] Set initial enforcing status.
1441 See security/selinux/Kconfig help text.
1442 0 -- permissive (log only, no denials).
1443 1 -- enforcing (deny and log).
1445 Value can be changed at runtime via
1446 /sys/fs/selinux/enforce.
1449 Disable Error Record Serialization Table (ERST)
1452 ether= [HW,NET] Ethernet cards parameters
1453 This option is obsoleted by the "netdev=" option, which
1454 has equivalent usage. See its documentation for details.
1458 Permit 'security.evm' to be updated regardless of
1459 current integrity status.
1461 early_page_ext [KNL] Enforces page_ext initialization to earlier
1462 stages so cover more early boot allocations.
1463 Please note that as side effect some optimizations
1464 might be disabled to achieve that (e.g. parallelized
1465 memory initialization is disabled) so the boot process
1466 might take longer, especially on systems with a lot of
1467 memory. Available with CONFIG_PAGE_EXTENSION=y.
1472 fail_make_request=[KNL]
1473 General fault injection mechanism.
1474 Format: <interval>,<probability>,<space>,<times>
1475 See also Documentation/fault-injection/.
1478 Format: { initns | none }
1479 See Documentation/admin-guide/sysctl/net.rst for
1480 fb_tunnels_only_for_init_ns
1483 See Documentation/admin-guide/blockdev/floppy.rst.
1485 force_pal_cache_flush
1486 [IA-64] Avoid check_sal_cache_flush which may hang on
1487 buggy SAL_CACHE_FLUSH implementations. Using this
1488 parameter will force ia64_sal_cache_flush to call
1489 ia64_pal_cache_flush instead of SAL_CACHE_FLUSH.
1492 Forcefully enable Physical Address Extension (PAE).
1493 Many Pentium M systems disable PAE but may have a
1494 functionally usable PAE implementation.
1495 Warning: use of this parameter will taint the kernel
1496 and may cause unknown problems.
1499 [FTRACE] will set and start the specified tracer
1500 as early as possible in order to facilitate early
1503 ftrace_boot_snapshot
1504 [FTRACE] On boot up, a snapshot will be taken of the
1505 ftrace ring buffer that can be read at:
1506 /sys/kernel/tracing/snapshot.
1507 This is useful if you need tracing information from kernel
1508 boot up that is likely to be overridden by user space
1509 start up functionality.
1511 Optionally, the snapshot can also be defined for a tracing
1512 instance that was created by the trace_instance= command
1515 trace_instance=foo,sched_switch ftrace_boot_snapshot=foo
1517 The above will cause the "foo" tracing instance to trigger
1518 a snapshot at the end of boot up.
1520 ftrace_dump_on_oops[=orig_cpu]
1521 [FTRACE] will dump the trace buffers on oops.
1522 If no parameter is passed, ftrace will dump
1523 buffers of all CPUs, but if you pass orig_cpu, it will
1524 dump only the buffer of the CPU that triggered the
1527 ftrace_filter=[function-list]
1528 [FTRACE] Limit the functions traced by the function
1529 tracer at boot up. function-list is a comma-separated
1530 list of functions. This list can be changed at run
1531 time by the set_ftrace_filter file in the debugfs
1534 ftrace_notrace=[function-list]
1535 [FTRACE] Do not trace the functions specified in
1536 function-list. This list can be changed at run time
1537 by the set_ftrace_notrace file in the debugfs
1540 ftrace_graph_filter=[function-list]
1541 [FTRACE] Limit the top level callers functions traced
1542 by the function graph tracer at boot up.
1543 function-list is a comma-separated list of functions
1544 that can be changed at run time by the
1545 set_graph_function file in the debugfs tracing directory.
1547 ftrace_graph_notrace=[function-list]
1548 [FTRACE] Do not trace from the functions specified in
1549 function-list. This list is a comma-separated list of
1550 functions that can be changed at run time by the
1551 set_graph_notrace file in the debugfs tracing directory.
1553 ftrace_graph_max_depth=<uint>
1554 [FTRACE] Used with the function graph tracer. This is
1555 the max depth it will trace into a function. This value
1556 can be changed at run time by the max_graph_depth file
1557 in the tracefs tracing directory. default: 0 (no limit)
1559 fw_devlink= [KNL] Create device links between consumer and supplier
1560 devices by scanning the firmware to infer the
1561 consumer/supplier relationships. This feature is
1562 especially useful when drivers are loaded as modules as
1563 it ensures proper ordering of tasks like device probing
1564 (suppliers first, then consumers), supplier boot state
1565 clean up (only after all consumers have probed),
1566 suspend/resume & runtime PM (consumers first, then
1568 Format: { off | permissive | on | rpm }
1569 off -- Don't create device links from firmware info.
1570 permissive -- Create device links from firmware info
1571 but use it only for ordering boot state clean
1572 up (sync_state() calls).
1573 on -- Create device links from firmware info and use it
1574 to enforce probe and suspend/resume ordering.
1575 rpm -- Like "on", but also use to order runtime PM.
1577 fw_devlink.strict=<bool>
1578 [KNL] Treat all inferred dependencies as mandatory
1579 dependencies. This only applies for fw_devlink=on|rpm.
1583 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad
1584 support via parallel port (up to 5 devices per port)
1585 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5>
1586 See also Documentation/input/devices/joystick-parport.rst
1590 gart_fix_e820= [X86-64] disable the fix e820 for K8 GART
1594 gcov_persist= [GCOV] When non-zero (default), profiling data for
1595 kernel modules is saved and remains accessible via
1596 debugfs, even when the module is unloaded/reloaded.
1597 When zero, profiling data is discarded and associated
1598 debugfs files are removed at module unload time.
1600 goldfish [X86] Enable the goldfish android emulator platform.
1601 Don't use this when you are not running on the
1604 gpio-mockup.gpio_mockup_ranges
1605 [HW] Sets the ranges of gpiochip of for this device.
1606 Format: <start1>,<end1>,<start2>,<end2>...
1607 gpio-mockup.gpio_mockup_named_lines
1608 [HW] Let the driver know GPIO lines should be named.
1610 gpt [EFI] Forces disk with valid GPT signature but
1611 invalid Protective MBR to be treated as GPT. If the
1612 primary GPT is corrupted, it enables the backup/alternate
1613 GPT to be used instead.
1615 grcan.enable0= [HW] Configuration of physical interface 0. Determines
1616 the "Enable 0" bit of the configuration register.
1619 grcan.enable1= [HW] Configuration of physical interface 1. Determines
1620 the "Enable 0" bit of the configuration register.
1623 grcan.select= [HW] Select which physical interface to use.
1626 grcan.txsize= [HW] Sets the size of the tx buffer.
1627 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0.
1629 grcan.rxsize= [HW] Sets the size of the rx buffer.
1630 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0.
1634 [KNL] Under CONFIG_HARDENED_USERCOPY, whether
1635 hardening is enabled for this boot. Hardened
1636 usercopy checking is used to protect the kernel
1637 from reading or writing beyond known memory
1638 allocation boundaries as a proactive defense
1639 against bounds-checking flaws in the kernel's
1640 copy_to_user()/copy_from_user() interface.
1641 on Perform hardened usercopy checks (default).
1642 off Disable hardened usercopy checks.
1644 hardlockup_all_cpu_backtrace=
1645 [KNL] Should the hard-lockup detector generate
1646 backtraces on all cpus.
1649 hashdist= [KNL,NUMA] Large hashes allocated during boot
1650 are distributed across NUMA nodes. Defaults on
1651 for 64-bit NUMA, off otherwise.
1652 Format: 0 | 1 (for off | on)
1654 hcl= [IA-64] SGI's Hardware Graph compatibility layer
1656 hd= [EIDE] (E)IDE hard drive subsystem geometry
1657 Format: <cyl>,<head>,<sect>
1660 Disable Hardware Error Source Table (HEST) support;
1661 corresponding firmware-first mode error processing
1662 logic will be disabled.
1664 hibernate= [HIBERNATION]
1665 noresume Don't check if there's a hibernation image
1666 present during boot.
1667 nocompress Don't compress/decompress hibernation images.
1668 no Disable hibernation and resume.
1669 protect_image Turn on image protection during restoration
1670 (that will set all pages holding image data
1671 during restoration read-only).
1673 highmem=nn[KMG] [KNL,BOOT] forces the highmem zone to have an exact
1674 size of <nn>. This works even on boxes that have no
1675 highmem otherwise. This also works to reduce highmem
1676 size on bigger boxes.
1678 highres= [KNL] Enable/disable high resolution timer mode.
1679 Valid parameters: "on", "off"
1684 hostname= [KNL] Set the hostname (aka UTS nodename).
1686 This allows setting the system's hostname during early
1687 startup. This sets the name returned by gethostname.
1688 Using this parameter to set the hostname makes it
1689 possible to ensure the hostname is correctly set before
1690 any userspace processes run, avoiding the possibility
1691 that a process may call gethostname before the hostname
1692 has been explicitly set, resulting in the calling
1693 process getting an incorrect result. The string must
1694 not exceed the maximum allowed hostname length (usually
1695 64 characters) and will be truncated otherwise.
1697 hpet= [X86-32,HPET] option to control HPET usage
1698 Format: { enable (default) | disable | force |
1700 disable: disable HPET and use PIT instead
1701 force: allow force enabled of undocumented chips (ICH4,
1703 verbose: show contents of HPET registers during setup
1705 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET
1706 registers. Default set by CONFIG_HPET_MMAP_DEFAULT.
1708 hugepages= [HW] Number of HugeTLB pages to allocate at boot.
1709 If this follows hugepagesz (below), it specifies
1710 the number of pages of hugepagesz to be allocated.
1711 If this is the first HugeTLB parameter on the command
1712 line, it specifies the number of pages to allocate for
1713 the default huge page size. If using node format, the
1714 number of pages to allocate per-node can be specified.
1715 See also Documentation/admin-guide/mm/hugetlbpage.rst.
1716 Format: <integer> or (node format)
1717 <node>:<integer>[,<node>:<integer>]
1720 [HW] The size of the HugeTLB pages. This is used in
1721 conjunction with hugepages (above) to allocate huge
1722 pages of a specific size at boot. The pair
1723 hugepagesz=X hugepages=Y can be specified once for
1724 each supported huge page size. Huge page sizes are
1725 architecture dependent. See also
1726 Documentation/admin-guide/mm/hugetlbpage.rst.
1729 hugetlb_cma= [HW,CMA] The size of a CMA area used for allocation
1730 of gigantic hugepages. Or using node format, the size
1731 of a CMA area per node can be specified.
1732 Format: nn[KMGTPE] or (node format)
1733 <node>:nn[KMGTPE][,<node>:nn[KMGTPE]]
1735 Reserve a CMA area of given size and allocate gigantic
1736 hugepages using the CMA allocator. If enabled, the
1737 boot-time allocation of gigantic hugepages is skipped.
1739 hugetlb_free_vmemmap=
1740 [KNL] Requires CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP
1742 Control if HugeTLB Vmemmap Optimization (HVO) is enabled.
1743 Allows heavy hugetlb users to free up some more
1744 memory (7 * PAGE_SIZE for each 2MB hugetlb page).
1745 Format: { on | off (default) }
1750 Built with CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP_DEFAULT_ON=y,
1753 Note that the vmemmap pages may be allocated from the added
1754 memory block itself when memory_hotplug.memmap_on_memory is
1755 enabled, those vmemmap pages cannot be optimized even if this
1756 feature is enabled. Other vmemmap pages not allocated from
1757 the added memory block itself do not be affected.
1760 [KNL] Should the hung task detector generate panics.
1763 A value of 1 instructs the kernel to panic when a
1764 hung task is detected. The default value is controlled
1765 by the CONFIG_BOOTPARAM_HUNG_TASK_PANIC build-time
1766 option. The value selected by this boot parameter can
1767 be changed later by the kernel.hung_task_panic sysctl.
1769 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC)
1770 terminal devices. Valid values: 0..8
1771 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs.
1772 If specified, z/VM IUCV HVC accepts connections
1773 from listed z/VM user IDs only.
1775 hv_nopvspin [X86,HYPER_V] Disables the paravirt spinlock optimizations
1776 which allow the hypervisor to 'idle' the
1777 guest on lock contention.
1779 i2c_bus= [HW] Override the default board specific I2C bus speed
1780 or register an additional I2C bus that is not
1781 registered from board initialization code.
1785 i8042.debug [HW] Toggle i8042 debug mode
1786 i8042.unmask_kbd_data
1787 [HW] Enable printing of interrupt data from the KBD port
1788 (disabled by default, and as a pre-condition
1789 requires that i8042.debug=1 be enabled)
1790 i8042.direct [HW] Put keyboard port into non-translated mode
1791 i8042.dumbkbd [HW] Pretend that controller can only read data from
1792 keyboard and cannot control its state
1793 (Don't attempt to blink the leds)
1794 i8042.noaux [HW] Don't check for auxiliary (== mouse) port
1795 i8042.nokbd [HW] Don't check/create keyboard port
1796 i8042.noloop [HW] Disable the AUX Loopback command while probing
1798 i8042.nomux [HW] Don't check presence of an active multiplexing
1800 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX
1802 i8042.notimeout [HW] Ignore timeout condition signalled by controller
1803 i8042.reset [HW] Reset the controller during init, cleanup and
1804 suspend-to-ram transitions, only during s2r
1805 transitions, or never reset
1806 Format: { 1 | Y | y | 0 | N | n }
1807 1, Y, y: always reset controller
1808 0, N, n: don't ever reset controller
1809 Default: only on s2r transitions on x86; most other
1810 architectures force reset to be always executed
1811 i8042.unlock [HW] Unlock (ignore) the keylock
1812 i8042.kbdreset [HW] Reset device connected to KBD port
1814 [HW] Allow deferred probing upon i8042 probe errors
1818 i915.invert_brightness=
1819 [DRM] Invert the sense of the variable that is used to
1820 set the brightness of the panel backlight. Normally a
1821 brightness value of 0 indicates backlight switched off,
1822 and the maximum of the brightness value sets the backlight
1823 to maximum brightness. If this parameter is set to 0
1824 (default) and the machine requires it, or this parameter
1825 is set to 1, a brightness value of 0 sets the backlight
1826 to maximum brightness, and the maximum of the brightness
1827 value switches the backlight off.
1828 -1 -- never invert brightness
1829 0 -- machine default
1830 1 -- force brightness inversion
1833 Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]]
1837 Format: idle=poll, idle=halt, idle=nomwait
1838 Poll forces a polling idle loop that can slightly
1839 improve the performance of waking up a idle CPU, but
1840 will use a lot of power and make the system run hot.
1842 idle=halt: Halt is forced to be used for CPU idle.
1843 In such case C2/C3 won't be used again.
1844 idle=nomwait: Disable mwait for CPU C-states
1848 Allow force disabling of Shared Virtual Memory (SVA)
1849 support for the idxd driver. By default it is set to
1852 idxd.tc_override= [HW]
1854 Allow override of default traffic class configuration
1855 for the device. By default it is set to false (0).
1857 ieee754= [MIPS] Select IEEE Std 754 conformance mode
1858 Format: { strict | legacy | 2008 | relaxed }
1861 Choose which programs will be accepted for execution
1862 based on the IEEE 754 NaN encoding(s) supported by
1863 the FPU and the NaN encoding requested with the value
1864 of an ELF file header flag individually set by each
1865 binary. Hardware implementations are permitted to
1866 support either or both of the legacy and the 2008 NaN
1869 Available settings are as follows:
1870 strict accept binaries that request a NaN encoding
1871 supported by the FPU
1872 legacy only accept legacy-NaN binaries, if supported
1874 2008 only accept 2008-NaN binaries, if supported
1876 relaxed accept any binaries regardless of whether
1877 supported by the FPU
1879 The FPU emulator is always able to support both NaN
1880 encodings, so if no FPU hardware is present or it has
1881 been disabled with 'nofpu', then the settings of
1882 'legacy' and '2008' strap the emulator accordingly,
1883 'relaxed' straps the emulator for both legacy-NaN and
1884 2008-NaN, whereas 'strict' enables legacy-NaN only on
1885 legacy processors and both NaN encodings on MIPS32 or
1888 The setting for ABS.fmt/NEG.fmt instruction execution
1889 mode generally follows that for the NaN encoding,
1890 except where unsupported by hardware.
1892 ignore_loglevel [KNL]
1893 Ignore loglevel setting - this will print /all/
1894 kernel messages to the console. Useful for debugging.
1895 We also add it as printk module parameter, so users
1896 could change it dynamically, usually by
1897 /sys/module/printk/parameters/ignore_loglevel.
1900 Ignore RLIMIT_DATA setting for data mappings,
1901 print warning at first misuse. Can be changed via
1902 /sys/module/kernel/parameters/ignore_rlimit_data.
1904 ihash_entries= [KNL]
1905 Set number of hash buckets for inode cache.
1907 ima_appraise= [IMA] appraise integrity measurements
1908 Format: { "off" | "enforce" | "fix" | "log" }
1911 ima_appraise_tcb [IMA] Deprecated. Use ima_policy= instead.
1912 The builtin appraise policy appraises all files
1915 ima_canonical_fmt [IMA]
1916 Use the canonical format for the binary runtime
1917 measurements, instead of host native format.
1920 Format: { md5 | sha1 | rmd160 | sha256 | sha384
1924 The list of supported hash algorithms is defined
1925 in crypto/hash_info.h.
1928 The builtin policies to load during IMA setup.
1929 Format: "tcb | appraise_tcb | secure_boot |
1930 fail_securely | critical_data"
1932 The "tcb" policy measures all programs exec'd, files
1933 mmap'd for exec, and all files opened with the read
1934 mode bit set by either the effective uid (euid=0) or
1937 The "appraise_tcb" policy appraises the integrity of
1938 all files owned by root.
1940 The "secure_boot" policy appraises the integrity
1941 of files (eg. kexec kernel image, kernel modules,
1942 firmware, policy, etc) based on file signatures.
1944 The "fail_securely" policy forces file signature
1945 verification failure also on privileged mounted
1946 filesystems with the SB_I_UNVERIFIABLE_SIGNATURE
1949 The "critical_data" policy measures kernel integrity
1952 ima_tcb [IMA] Deprecated. Use ima_policy= instead.
1953 Load a policy which meets the needs of the Trusted
1954 Computing Base. This means IMA will measure all
1955 programs exec'd, files mmap'd for exec, and all files
1956 opened for read by uid=0.
1959 Select one of defined IMA measurements template formats.
1960 Formats: { "ima" | "ima-ng" | "ima-ngv2" | "ima-sig" |
1965 [IMA] Define a custom template format.
1966 Format: { "field1|...|fieldN" }
1968 ima.ahash_minsize= [IMA] Minimum file size for asynchronous hash usage
1969 Format: <min_file_size>
1970 Set the minimal file size for using asynchronous hash.
1971 If left unspecified, ahash usage is disabled.
1973 ahash performance varies for different data sizes on
1974 different crypto accelerators. This option can be used
1975 to achieve the best performance for a particular HW.
1977 ima.ahash_bufsize= [IMA] Asynchronous hash buffer size
1979 Set hashing buffer size. Default: 4k.
1981 ahash performance varies for different chunk sizes on
1982 different crypto accelerators. This option can be used
1983 to achieve best performance for particular HW.
1987 Run specified binary instead of /sbin/init as init
1990 initcall_debug [KNL] Trace initcalls as they are executed. Useful
1991 for working out where the kernel is dying during
1994 initcall_blacklist= [KNL] Do not execute a comma-separated list of
1995 initcall functions. Useful for debugging built-in
1996 modules and initcalls.
1998 initramfs_async= [KNL]
2001 This parameter controls whether the initramfs
2002 image is unpacked asynchronously, concurrently
2003 with devices being probed and
2004 initialized. This should normally just work,
2005 but as a debugging aid, one can get the
2006 historical behaviour of the initramfs
2007 unpacking being completed before device_ and
2010 initrd= [BOOT] Specify the location of the initial ramdisk
2012 initrdmem= [KNL] Specify a physical address and size from which to
2013 load the initrd. If an initrd is compiled in or
2014 specified in the bootparams, it takes priority over this
2016 Format: ss[KMG],nn[KMG]
2019 init_on_alloc= [MM] Fill newly allocated pages and heap objects with
2022 Default set by CONFIG_INIT_ON_ALLOC_DEFAULT_ON.
2024 init_on_free= [MM] Fill freed pages and heap objects with zeroes.
2026 Default set by CONFIG_INIT_ON_FREE_DEFAULT_ON.
2028 init_pkru= [X86] Specify the default memory protection keys rights
2029 register contents for all processes. 0x55555554 by
2030 default (disallow access to all but pkey 0). Can
2031 override in debugfs after boot.
2033 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver
2036 int_pln_enable [X86] Enable power limit notification interrupt
2038 integrity_audit=[IMA]
2039 Format: { "0" | "1" }
2040 0 -- basic integrity auditing messages. (Default)
2041 1 -- additional integrity auditing messages.
2043 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option
2045 Enable intel iommu driver.
2047 Disable intel iommu driver.
2048 igfx_off [Default Off]
2049 By default, gfx is mapped as normal device. If a gfx
2050 device has a dedicated DMAR unit, the DMAR unit is
2051 bypassed by not enabling DMAR with this option. In
2052 this case, gfx device will use physical address for
2054 strict [Default Off]
2055 Deprecated, equivalent to iommu.strict=1.
2056 sp_off [Default Off]
2057 By default, super page will be supported if Intel IOMMU
2058 has the capability. With this option, super page will
2061 Enable the Intel IOMMU scalable mode if the hardware
2062 advertises that it has support for the scalable mode
2065 Disallow use of the Intel IOMMU scalable mode.
2066 tboot_noforce [Default Off]
2067 Do not force the Intel IOMMU enabled under tboot.
2068 By default, tboot will force Intel IOMMU on, which
2069 could harm performance of some high-throughput
2070 devices like 40GBit network cards, even if identity
2072 Note that using this option lowers the security
2073 provided by tboot because it makes the system
2074 vulnerable to DMA attacks.
2076 intel_idle.max_cstate= [KNL,HW,ACPI,X86]
2077 0 disables intel_idle and fall back on acpi_idle.
2078 1 to 9 specify maximum depth of C-state.
2082 Do not enable intel_pstate as the default
2083 scaling driver for the supported processors
2085 Use intel_pstate as a scaling driver, but configure it
2086 to work with generic cpufreq governors (instead of
2087 enabling its internal governor). This mode cannot be
2088 used along with the hardware-managed P-states (HWP)
2091 Enable intel_pstate on systems that prohibit it by default
2092 in favor of acpi-cpufreq. Forcing the intel_pstate driver
2093 instead of acpi-cpufreq may disable platform features, such
2094 as thermal controls and power capping, that rely on ACPI
2095 P-States information being indicated to OSPM and therefore
2096 should be used with caution. This option does not work with
2097 processors that aren't supported by the intel_pstate driver
2098 or on platforms that use pcc-cpufreq instead of acpi-cpufreq.
2100 Do not enable hardware P state control (HWP)
2103 Only load intel_pstate on systems which support
2104 hardware P state control (HWP) if available.
2106 Enforce ACPI _PPC performance limits. If the Fixed ACPI
2107 Description Table, specifies preferred power management
2108 profile as "Enterprise Server" or "Performance Server",
2109 then this feature is turned on by default.
2111 Allow per-logical-CPU P-State performance control limits using
2112 cpufreq sysfs interface
2114 intremap= [X86-64, Intel-IOMMU]
2115 on enable Interrupt Remapping (default)
2116 off disable Interrupt Remapping
2117 nosid disable Source ID checking
2119 BIOS x2APIC opt-out request will be ignored
2120 nopost disable Interrupt Posting
2122 iomem= Disable strict checking of access to MMIO memory
2123 strict regions from userspace.
2138 nobypass [PPC/POWERNV]
2139 Disable IOMMU bypass, using IOMMU for PCI devices.
2141 iommu.forcedac= [ARM64, X86] Control IOVA allocation for PCI devices.
2142 Format: { "0" | "1" }
2143 0 - Try to allocate a 32-bit DMA address first, before
2144 falling back to the full range if needed.
2145 1 - Allocate directly from the full usable range,
2146 forcing Dual Address Cycle for PCI cards supporting
2147 greater than 32-bit addressing.
2149 iommu.strict= [ARM64, X86] Configure TLB invalidation behaviour
2150 Format: { "0" | "1" }
2152 Request that DMA unmap operations use deferred
2153 invalidation of hardware TLBs, for increased
2154 throughput at the cost of reduced device isolation.
2155 Will fall back to strict mode if not supported by
2156 the relevant IOMMU driver.
2158 DMA unmap operations invalidate IOMMU hardware TLBs
2160 unset - Use value of CONFIG_IOMMU_DEFAULT_DMA_{LAZY,STRICT}.
2161 Note: on x86, strict mode specified via one of the
2162 legacy driver-specific options takes precedence.
2165 [ARM64, X86] Configure DMA to bypass the IOMMU by default.
2166 Format: { "0" | "1" }
2167 0 - Use IOMMU translation for DMA.
2168 1 - Bypass the IOMMU for DMA.
2169 unset - Use value of CONFIG_IOMMU_DEFAULT_PASSTHROUGH.
2171 io7= [HW] IO7 for Marvel-based Alpha systems
2172 See comment before marvel_specify_io7 in
2173 arch/alpha/kernel/core_marvel.c.
2175 io_delay= [X86] I/O delay method
2177 Standard port 0x80 based delay
2179 Alternate port 0xed based delay (needed on some systems)
2181 Simple two microseconds delay
2186 See Documentation/admin-guide/nfs/nfsroot.rst.
2188 ipcmni_extend [KNL] Extend the maximum number of unique System V
2189 IPC identifiers from 32,768 to 16,777,216.
2191 irqaffinity= [SMP] Set the default irq affinity mask
2192 The argument is a cpu list, as described above.
2194 irqchip.gicv2_force_probe=
2197 Force the kernel to look for the second 4kB page
2198 of a GICv2 controller even if the memory range
2199 exposed by the device tree is too small.
2201 irqchip.gicv3_nolpi=
2203 Force the kernel to ignore the availability of
2204 LPIs (and by consequence ITSs). Intended for system
2205 that use the kernel as a bootloader, and thus want
2206 to let secondary kernels in charge of setting up
2209 irqchip.gicv3_pseudo_nmi= [ARM64]
2210 Enables support for pseudo-NMIs in the kernel. This
2211 requires the kernel to be built with
2212 CONFIG_ARM64_PSEUDO_NMI.
2215 When an interrupt is not handled search all handlers
2216 for it. Intended to get systems with badly broken
2220 When an interrupt is not handled search all handlers
2221 for it. Also check all handlers each timer
2222 interrupt. Intended to get systems with badly broken
2226 Format: <RDP>,<reset>,<pci_scan>,<verbosity>
2228 isolcpus= [KNL,SMP,ISOL] Isolate a given set of CPUs from disturbance.
2229 [Deprecated - use cpusets instead]
2230 Format: [flag-list,]<cpu-list>
2232 Specify one or more CPUs to isolate from disturbances
2233 specified in the flag list (default: domain):
2236 Disable the tick when a single task runs.
2238 A residual 1Hz tick is offloaded to workqueues, which you
2239 need to affine to housekeeping through the global
2240 workqueue's affinity configured via the
2241 /sys/devices/virtual/workqueue/cpumask sysfs file, or
2242 by using the 'domain' flag described below.
2244 NOTE: by default the global workqueue runs on all CPUs,
2245 so to protect individual CPUs the 'cpumask' file has to
2246 be configured manually after bootup.
2249 Isolate from the general SMP balancing and scheduling
2250 algorithms. Note that performing domain isolation this way
2251 is irreversible: it's not possible to bring back a CPU to
2252 the domains once isolated through isolcpus. It's strongly
2253 advised to use cpusets instead to disable scheduler load
2254 balancing through the "cpuset.sched_load_balance" file.
2255 It offers a much more flexible interface where CPUs can
2256 move in and out of an isolated set anytime.
2258 You can move a process onto or off an "isolated" CPU via
2259 the CPU affinity syscalls or cpuset.
2260 <cpu number> begins at 0 and the maximum value is
2261 "number of CPUs in system - 1".
2265 Isolate from being targeted by managed interrupts
2266 which have an interrupt mask containing isolated
2267 CPUs. The affinity of managed interrupts is
2268 handled by the kernel and cannot be changed via
2269 the /proc/irq/* interfaces.
2271 This isolation is best effort and only effective
2272 if the automatically assigned interrupt mask of a
2273 device queue contains isolated and housekeeping
2274 CPUs. If housekeeping CPUs are online then such
2275 interrupts are directed to the housekeeping CPU
2276 so that IO submitted on the housekeeping CPU
2277 cannot disturb the isolated CPU.
2279 If a queue's affinity mask contains only isolated
2280 CPUs then this parameter has no effect on the
2281 interrupt routing decision, though interrupts are
2282 only delivered when tasks running on those
2283 isolated CPUs submit IO. IO submitted on
2284 housekeeping CPUs has no influence on those
2287 The format of <cpu-list> is described above.
2291 ivrs_ioapic [HW,X86-64]
2292 Provide an override to the IOAPIC-ID<->DEVICE-ID
2293 mapping provided in the IVRS ACPI table.
2294 By default, PCI segment is 0, and can be omitted.
2296 For example, to map IOAPIC-ID decimal 10 to
2297 PCI segment 0x1 and PCI device 00:14.0,
2298 write the parameter as:
2299 ivrs_ioapic=10@0001:00:14.0
2302 * To map IOAPIC-ID decimal 10 to PCI device 00:14.0
2303 write the parameter as:
2304 ivrs_ioapic[10]=00:14.0
2305 * To map IOAPIC-ID decimal 10 to PCI segment 0x1 and
2306 PCI device 00:14.0 write the parameter as:
2307 ivrs_ioapic[10]=0001:00:14.0
2309 ivrs_hpet [HW,X86-64]
2310 Provide an override to the HPET-ID<->DEVICE-ID
2311 mapping provided in the IVRS ACPI table.
2312 By default, PCI segment is 0, and can be omitted.
2314 For example, to map HPET-ID decimal 10 to
2315 PCI segment 0x1 and PCI device 00:14.0,
2316 write the parameter as:
2317 ivrs_hpet=10@0001:00:14.0
2320 * To map HPET-ID decimal 0 to PCI device 00:14.0
2321 write the parameter as:
2322 ivrs_hpet[0]=00:14.0
2323 * To map HPET-ID decimal 10 to PCI segment 0x1 and
2324 PCI device 00:14.0 write the parameter as:
2325 ivrs_ioapic[10]=0001:00:14.0
2327 ivrs_acpihid [HW,X86-64]
2328 Provide an override to the ACPI-HID:UID<->DEVICE-ID
2329 mapping provided in the IVRS ACPI table.
2330 By default, PCI segment is 0, and can be omitted.
2332 For example, to map UART-HID:UID AMD0020:0 to
2333 PCI segment 0x1 and PCI device ID 00:14.5,
2334 write the parameter as:
2335 ivrs_acpihid=AMD0020:0@0001:00:14.5
2338 * To map UART-HID:UID AMD0020:0 to PCI segment is 0,
2339 PCI device ID 00:14.5, write the parameter as:
2340 ivrs_acpihid[00:14.5]=AMD0020:0
2341 * To map UART-HID:UID AMD0020:0 to PCI segment 0x1 and
2342 PCI device ID 00:14.5, write the parameter as:
2343 ivrs_acpihid[0001:00:14.5]=AMD0020:0
2345 js= [HW,JOY] Analog joystick
2346 See Documentation/input/joydev/joystick.rst.
2349 [KNL] Enforce KASAN (Kernel Address Sanitizer) to print
2350 report on every invalid memory access. Without this
2351 parameter KASAN will print report only for the first
2355 Do not unregister boot console at start. This is only
2356 useful for debugging when something happens in the window
2357 between unregistering the boot console and initializing
2362 kernelcore= [KNL,X86,IA-64,PPC]
2363 Format: nn[KMGTPE] | nn% | "mirror"
2364 This parameter specifies the amount of memory usable by
2365 the kernel for non-movable allocations. The requested
2366 amount is spread evenly throughout all nodes in the
2367 system as ZONE_NORMAL. The remaining memory is used for
2368 movable memory in its own zone, ZONE_MOVABLE. In the
2369 event, a node is too small to have both ZONE_NORMAL and
2370 ZONE_MOVABLE, kernelcore memory will take priority and
2371 other nodes will have a larger ZONE_MOVABLE.
2373 ZONE_MOVABLE is used for the allocation of pages that
2374 may be reclaimed or moved by the page migration
2375 subsystem. Note that allocations like PTEs-from-HighMem
2376 still use the HighMem zone if it exists, and the Normal
2377 zone if it does not.
2379 It is possible to specify the exact amount of memory in
2380 the form of "nn[KMGTPE]", a percentage of total system
2381 memory in the form of "nn%", or "mirror". If "mirror"
2382 option is specified, mirrored (reliable) memory is used
2383 for non-movable allocations and remaining memory is used
2384 for Movable pages. "nn[KMGTPE]", "nn%", and "mirror"
2385 are exclusive, so you cannot specify multiple forms.
2387 kgdbdbgp= [KGDB,HW] kgdb over EHCI usb debug port.
2388 Format: <Controller#>[,poll interval]
2389 The controller # is the number of the ehci usb debug
2390 port as it is probed via PCI. The poll interval is
2391 optional and is the number seconds in between
2392 each poll cycle to the debug port in case you need
2393 the functionality for interrupting the kernel with
2394 gdb or control-c on the dbgp connection. When
2395 not using this parameter you use sysrq-g to break into
2396 the kernel debugger.
2398 kgdboc= [KGDB,HW] kgdb over consoles.
2399 Requires a tty driver that supports console polling,
2400 or a supported polling keyboard driver (non-usb).
2401 Serial only format: <serial_device>[,baud]
2402 keyboard only format: kbd
2403 keyboard and serial format: kbd,<serial_device>[,baud]
2404 Optional Kernel mode setting:
2405 kms, kbd format: kms,kbd
2406 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud]
2408 kgdboc_earlycon= [KGDB,HW]
2409 If the boot console provides the ability to read
2410 characters and can work in polling mode, you can use
2411 this parameter to tell kgdb to use it as a backend
2412 until the normal console is registered. Intended to
2413 be used together with the kgdboc parameter which
2414 specifies the normal console to transition to.
2416 The name of the early console should be specified
2417 as the value of this parameter. Note that the name of
2418 the early console might be different than the tty
2419 name passed to kgdboc. It's OK to leave the value
2420 blank and the first boot console that implements
2421 read() will be picked.
2423 kgdbwait [KGDB] Stop kernel execution and enter the
2424 kernel debugger at the earliest opportunity.
2426 kmac= [MIPS] Korina ethernet MAC address.
2427 Configure the RouterBoard 532 series on-chip
2428 Ethernet adapter MAC address.
2430 kmemleak= [KNL] Boot-time kmemleak enable/disable
2431 Valid arguments: on, off
2433 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y,
2436 kprobe_event=[probe-list]
2437 [FTRACE] Add kprobe events and enable at boot time.
2438 The probe-list is a semicolon delimited list of probe
2439 definitions. Each definition is same as kprobe_events
2440 interface, but the parameters are comma delimited.
2441 For example, to add a kprobe event on vfs_read with
2442 arg1 and arg2, add to the command line;
2444 kprobe_event=p,vfs_read,$arg1,$arg2
2446 See also Documentation/trace/kprobetrace.rst "Kernel
2447 Boot Parameter" section.
2449 kpti= [ARM64] Control page table isolation of user
2450 and kernel address spaces.
2451 Default: enabled on cores which need mitigation.
2455 kunit.enable= [KUNIT] Enable executing KUnit tests. Requires
2456 CONFIG_KUNIT to be set to be fully enabled. The
2457 default value can be overridden via
2458 KUNIT_DEFAULT_ENABLED.
2459 Default is 1 (enabled)
2461 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs.
2462 Default is 0 (don't ignore, but inject #GP)
2464 kvm.eager_page_split=
2465 [KVM,X86] Controls whether or not KVM will try to
2466 proactively split all huge pages during dirty logging.
2467 Eager page splitting reduces interruptions to vCPU
2468 execution by eliminating the write-protection faults
2469 and MMU lock contention that would otherwise be
2470 required to split huge pages lazily.
2472 VM workloads that rarely perform writes or that write
2473 only to a small region of VM memory may benefit from
2474 disabling eager page splitting to allow huge pages to
2475 still be used for reads.
2477 The behavior of eager page splitting depends on whether
2478 KVM_DIRTY_LOG_INITIALLY_SET is enabled or disabled. If
2479 disabled, all huge pages in a memslot will be eagerly
2480 split when dirty logging is enabled on that memslot. If
2481 enabled, eager page splitting will be performed during
2482 the KVM_CLEAR_DIRTY ioctl, and only for the pages being
2485 Eager page splitting is only supported when kvm.tdp_mmu=Y.
2489 kvm.enable_vmware_backdoor=[KVM] Support VMware backdoor PV interface.
2490 Default is false (don't support).
2493 [KVM] Controls the software workaround for the
2494 X86_BUG_ITLB_MULTIHIT bug.
2495 force : Always deploy workaround.
2496 off : Never deploy workaround.
2497 auto : Deploy workaround based on the presence of
2498 X86_BUG_ITLB_MULTIHIT.
2502 If the software workaround is enabled for the host,
2503 guests do need not to enable it for nested guests.
2505 kvm.nx_huge_pages_recovery_ratio=
2506 [KVM] Controls how many 4KiB pages are periodically zapped
2507 back to huge pages. 0 disables the recovery, otherwise if
2508 the value is N KVM will zap 1/Nth of the 4KiB pages every
2509 period (see below). The default is 60.
2511 kvm.nx_huge_pages_recovery_period_ms=
2512 [KVM] Controls the time period at which KVM zaps 4KiB pages
2513 back to huge pages. If the value is a non-zero N, KVM will
2514 zap a portion (see ratio above) of the pages every N msecs.
2515 If the value is 0 (the default), KVM will pick a period based
2516 on the ratio, such that a page is zapped after 1 hour on average.
2518 kvm-amd.nested= [KVM,AMD] Allow nested virtualization in KVM/SVM.
2519 Default is 1 (enabled)
2521 kvm-amd.npt= [KVM,AMD] Disable nested paging (virtualized MMU)
2523 Default is 1 (enabled) if in 64-bit or 32-bit PAE mode.
2526 [KVM,ARM] Select one of KVM/arm64's modes of operation.
2528 none: Forcefully disable KVM.
2530 nvhe: Standard nVHE-based mode, without support for
2533 protected: nVHE-based mode with support for guests whose
2534 state is kept private from the host.
2536 nested: VHE-based mode with support for nested
2537 virtualization. Requires at least ARMv8.3
2540 Defaults to VHE/nVHE based on hardware support. Setting
2541 mode to "protected" will disable kexec and hibernation
2542 for the host. "nested" is experimental and should be
2543 used with extreme caution.
2545 kvm-arm.vgic_v3_group0_trap=
2546 [KVM,ARM] Trap guest accesses to GICv3 group-0
2549 kvm-arm.vgic_v3_group1_trap=
2550 [KVM,ARM] Trap guest accesses to GICv3 group-1
2553 kvm-arm.vgic_v3_common_trap=
2554 [KVM,ARM] Trap guest accesses to GICv3 common
2557 kvm-arm.vgic_v4_enable=
2558 [KVM,ARM] Allow use of GICv4 for direct injection of
2561 kvm_cma_resv_ratio=n [PPC]
2562 Reserves given percentage from system memory area for
2563 contiguous memory allocation for KVM hash pagetable
2565 By default it reserves 5% of total system memory.
2569 kvm-intel.ept= [KVM,Intel] Disable extended page tables
2570 (virtualized MMU) support on capable Intel chips.
2571 Default is 1 (enabled)
2573 kvm-intel.emulate_invalid_guest_state=
2574 [KVM,Intel] Disable emulation of invalid guest state.
2575 Ignored if kvm-intel.enable_unrestricted_guest=1, as
2576 guest state is never invalid for unrestricted guests.
2577 This param doesn't apply to nested guests (L2), as KVM
2578 never emulates invalid L2 guest state.
2579 Default is 1 (enabled)
2581 kvm-intel.flexpriority=
2582 [KVM,Intel] Disable FlexPriority feature (TPR shadow).
2583 Default is 1 (enabled)
2586 [KVM,Intel] Enable VMX nesting (nVMX).
2587 Default is 0 (disabled)
2589 kvm-intel.unrestricted_guest=
2590 [KVM,Intel] Disable unrestricted guest feature
2591 (virtualized real and unpaged mode) on capable
2592 Intel chips. Default is 1 (enabled)
2594 kvm-intel.vmentry_l1d_flush=[KVM,Intel] Mitigation for L1 Terminal Fault
2597 Valid arguments: never, cond, always
2599 always: L1D cache flush on every VMENTER.
2600 cond: Flush L1D on VMENTER only when the code between
2601 VMEXIT and VMENTER can leak host memory.
2602 never: Disables the mitigation
2604 Default is cond (do L1 cache flush in specific instances)
2606 kvm-intel.vpid= [KVM,Intel] Disable Virtual Processor Identification
2607 feature (tagged TLBs) on capable Intel chips.
2608 Default is 1 (enabled)
2610 l1d_flush= [X86,INTEL]
2611 Control mitigation for L1D based snooping vulnerability.
2613 Certain CPUs are vulnerable to an exploit against CPU
2614 internal buffers which can forward information to a
2615 disclosure gadget under certain conditions.
2617 In vulnerable processors, the speculatively
2618 forwarded data can be used in a cache side channel
2619 attack, to access data to which the attacker does
2620 not have direct access.
2622 This parameter controls the mitigation. The
2625 on - enable the interface for the mitigation
2627 l1tf= [X86] Control mitigation of the L1TF vulnerability on
2630 The kernel PTE inversion protection is unconditionally
2631 enabled and cannot be disabled.
2634 Provides all available mitigations for the
2635 L1TF vulnerability. Disables SMT and
2636 enables all mitigations in the
2637 hypervisors, i.e. unconditional L1D flush.
2639 SMT control and L1D flush control via the
2640 sysfs interface is still possible after
2641 boot. Hypervisors will issue a warning
2642 when the first VM is started in a
2643 potentially insecure configuration,
2644 i.e. SMT enabled or L1D flush disabled.
2647 Same as 'full', but disables SMT and L1D
2648 flush runtime control. Implies the
2649 'nosmt=force' command line option.
2650 (i.e. sysfs control of SMT is disabled.)
2653 Leaves SMT enabled and enables the default
2654 hypervisor mitigation, i.e. conditional
2657 SMT control and L1D flush control via the
2658 sysfs interface is still possible after
2659 boot. Hypervisors will issue a warning
2660 when the first VM is started in a
2661 potentially insecure configuration,
2662 i.e. SMT enabled or L1D flush disabled.
2666 Disables SMT and enables the default
2667 hypervisor mitigation.
2669 SMT control and L1D flush control via the
2670 sysfs interface is still possible after
2671 boot. Hypervisors will issue a warning
2672 when the first VM is started in a
2673 potentially insecure configuration,
2674 i.e. SMT enabled or L1D flush disabled.
2677 Same as 'flush', but hypervisors will not
2678 warn when a VM is started in a potentially
2679 insecure configuration.
2682 Disables hypervisor mitigations and doesn't
2684 It also drops the swap size and available
2685 RAM limit restriction on both hypervisor and
2690 For details see: Documentation/admin-guide/hw-vuln/l1tf.rst
2696 lapic [X86-32,APIC] Enable the local APIC even if BIOS
2699 lapic= [X86,APIC] Do not use TSC deadline
2700 value for LAPIC timer one-shot implementation. Default
2701 back to the programmable timer unit in the LAPIC.
2702 Format: notscdeadline
2704 lapic_timer_c2_ok [X86,APIC] trust the local apic timer
2707 libata.dma= [LIBATA] DMA control
2708 libata.dma=0 Disable all PATA and SATA DMA
2709 libata.dma=1 PATA and SATA Disk DMA only
2710 libata.dma=2 ATAPI (CDROM) DMA only
2711 libata.dma=4 Compact Flash DMA only
2712 Combinations also work, so libata.dma=3 enables DMA
2713 for disks and CDROMs, but not CFs.
2715 libata.ignore_hpa= [LIBATA] Ignore HPA limit
2716 libata.ignore_hpa=0 keep BIOS limits (default)
2717 libata.ignore_hpa=1 ignore limits, using full disk
2719 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume
2723 libata.force= [LIBATA] Force configurations. The format is a comma-
2724 separated list of "[ID:]VAL" where ID is PORT[.DEVICE].
2725 PORT and DEVICE are decimal numbers matching port, link
2726 or device. Basically, it matches the ATA ID string
2727 printed on console by libata. If the whole ID part is
2728 omitted, the last PORT and DEVICE values are used. If
2729 ID hasn't been specified yet, the configuration applies
2730 to all ports, links and devices.
2732 If only DEVICE is omitted, the parameter applies to
2733 the port and all links and devices behind it. DEVICE
2734 number of 0 either selects the first device or the
2735 first fan-out link behind PMP device. It does not
2736 select the host link. DEVICE number of 15 selects the
2737 host link and device attached to it.
2739 The VAL specifies the configuration to force. As long
2740 as there is no ambiguity, shortcut notation is allowed.
2741 For example, both 1.5 and 1.5G would work for 1.5Gbps.
2742 The following configurations can be forced.
2744 * Cable type: 40c, 80c, short40c, unk, ign or sata.
2745 Any ID with matching PORT is used.
2747 * SATA link speed limit: 1.5Gbps or 3.0Gbps.
2749 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7].
2750 udma[/][16,25,33,44,66,100,133] notation is also
2753 * nohrst, nosrst, norst: suppress hard, soft and both
2756 * rstonce: only attempt one reset during hot-unplug
2759 * [no]dbdelay: Enable or disable the extra 200ms delay
2760 before debouncing a link PHY and device presence
2763 * [no]ncq: Turn on or off NCQ.
2765 * [no]ncqtrim: Enable or disable queued DSM TRIM.
2767 * [no]ncqati: Enable or disable NCQ trim on ATI chipset.
2769 * [no]trim: Enable or disable (unqueued) TRIM.
2771 * trim_zero: Indicate that TRIM command zeroes data.
2773 * max_trim_128m: Set 128M maximum trim size limit.
2775 * [no]dma: Turn on or off DMA transfers.
2777 * atapi_dmadir: Enable ATAPI DMADIR bridge support.
2779 * atapi_mod16_dma: Enable the use of ATAPI DMA for
2780 commands that are not a multiple of 16 bytes.
2782 * [no]dmalog: Enable or disable the use of the
2783 READ LOG DMA EXT command to access logs.
2785 * [no]iddevlog: Enable or disable access to the
2786 identify device data log.
2788 * [no]logdir: Enable or disable access to the general
2789 purpose log directory.
2791 * max_sec_128: Set transfer size limit to 128 sectors.
2793 * max_sec_1024: Set or clear transfer size limit to
2796 * max_sec_lba48: Set or clear transfer size limit to
2799 * [no]lpm: Enable or disable link power management.
2801 * [no]setxfer: Indicate if transfer speed mode setting
2804 * [no]fua: Disable or enable FUA (Force Unit Access)
2805 support for devices supporting this feature.
2807 * dump_id: Dump IDENTIFY data.
2809 * disable: Disable this device.
2811 If there are multiple matching configurations changing
2812 the same attribute, the last one is used.
2814 load_ramdisk= [RAM] [Deprecated]
2816 lockd.nlm_grace_period=P [NFS] Assign grace period.
2819 lockd.nlm_tcpport=N [NFS] Assign TCP port.
2822 lockd.nlm_timeout=T [NFS] Assign timeout value.
2825 lockd.nlm_udpport=M [NFS] Assign UDP port.
2828 lockdown= [SECURITY]
2829 { integrity | confidentiality }
2830 Enable the kernel lockdown feature. If set to
2831 integrity, kernel features that allow userland to
2832 modify the running kernel are disabled. If set to
2833 confidentiality, kernel features that allow userland
2834 to extract confidential information from the kernel
2837 locktorture.nreaders_stress= [KNL]
2838 Set the number of locking read-acquisition kthreads.
2839 Defaults to being automatically set based on the
2840 number of online CPUs.
2842 locktorture.nwriters_stress= [KNL]
2843 Set the number of locking write-acquisition kthreads.
2845 locktorture.onoff_holdoff= [KNL]
2846 Set time (s) after boot for CPU-hotplug testing.
2848 locktorture.onoff_interval= [KNL]
2849 Set time (s) between CPU-hotplug operations, or
2850 zero to disable CPU-hotplug testing.
2852 locktorture.shuffle_interval= [KNL]
2853 Set task-shuffle interval (jiffies). Shuffling
2854 tasks allows some CPUs to go into dyntick-idle
2855 mode during the locktorture test.
2857 locktorture.shutdown_secs= [KNL]
2858 Set time (s) after boot system shutdown. This
2859 is useful for hands-off automated testing.
2861 locktorture.stat_interval= [KNL]
2862 Time (s) between statistics printk()s.
2864 locktorture.stutter= [KNL]
2865 Time (s) to stutter testing, for example,
2866 specifying five seconds causes the test to run for
2867 five seconds, wait for five seconds, and so on.
2868 This tests the locking primitive's ability to
2869 transition abruptly to and from idle.
2871 locktorture.torture_type= [KNL]
2872 Specify the locking implementation to test.
2874 locktorture.verbose= [KNL]
2875 Enable additional printk() statements.
2877 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver
2880 loglevel= All Kernel Messages with a loglevel smaller than the
2881 console loglevel will be printed to the console. It can
2882 also be changed with klogd or other programs. The
2883 loglevels are defined as follows:
2885 0 (KERN_EMERG) system is unusable
2886 1 (KERN_ALERT) action must be taken immediately
2887 2 (KERN_CRIT) critical conditions
2888 3 (KERN_ERR) error conditions
2889 4 (KERN_WARNING) warning conditions
2890 5 (KERN_NOTICE) normal but significant condition
2891 6 (KERN_INFO) informational
2892 7 (KERN_DEBUG) debug-level messages
2894 log_buf_len=n[KMG] Sets the size of the printk ring buffer,
2895 in bytes. n must be a power of two and greater
2896 than the minimal size. The minimal size is defined
2897 by LOG_BUF_SHIFT kernel config parameter. There is
2898 also CONFIG_LOG_CPU_MAX_BUF_SHIFT config parameter
2899 that allows to increase the default size depending on
2900 the number of CPUs. See init/Kconfig for more details.
2902 logo.nologo [FB] Disables display of the built-in Linux logo.
2903 This may be used to provide more screen space for
2904 kernel log messages and is useful when debugging
2905 kernel boot problems.
2907 lp=0 [LP] Specify parallel ports to use, e.g,
2908 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses
2909 lp=reset first parallel port). 'lp=0' disables the
2910 lp=auto printer driver. 'lp=reset' (which can be
2911 specified in addition to the ports) causes
2912 attached printers to be reset. Using
2913 lp=port1,port2,... specifies the parallel ports
2914 to associate lp devices with, starting with
2915 lp0. A port specification may be 'none' to skip
2916 that lp device, or a parport name such as
2917 'parport0'. Specifying 'lp=auto' instead of a
2918 port specification list means that device IDs
2919 from each port should be examined, to see if
2920 an IEEE 1284-compliant printer is attached; if
2921 so, the driver will manage that printer.
2922 See also header of drivers/char/lp.c.
2925 Sets loops_per_jiffy to given constant, thus avoiding
2926 time-consuming boot-time autodetection (up to 250 ms per
2927 CPU). 0 enables autodetection (default). To determine
2928 the correct value for your kernel, boot with normal
2929 autodetection and see what value is printed. Note that
2930 on SMP systems the preset will be applied to all CPUs,
2931 which is likely to cause problems if your CPUs need
2932 significantly divergent settings. An incorrect value
2933 will cause delays in the kernel to be wrong, leading to
2934 unpredictable I/O errors and other breakage. Although
2935 unlikely, in the extreme case this might damage your
2939 Format: <io>,<irq>,<dma>
2941 lsm.debug [SECURITY] Enable LSM initialization debugging output.
2944 [SECURITY] Choose order of LSM initialization. This
2945 overrides CONFIG_LSM, and the "security=" parameter.
2947 machvec= [IA-64] Force the use of a particular machine-vector
2948 (machvec) in a generic kernel.
2949 Example: machvec=hpzx1
2951 machtype= [Loongson] Share the same kernel image file between
2952 different yeeloong laptops.
2953 Example: machtype=lemote-yeeloong-2f-7inch
2955 max_addr=nn[KMG] [KNL,BOOT,IA-64] All physical memory greater
2956 than or equal to this physical address is ignored.
2958 maxcpus= [SMP] Maximum number of processors that an SMP kernel
2959 will bring up during bootup. maxcpus=n : n >= 0 limits
2960 the kernel to bring up 'n' processors. Surely after
2961 bootup you can bring up the other plugged cpu by executing
2962 "echo 1 > /sys/devices/system/cpu/cpuX/online". So maxcpus
2963 only takes effect during system bootup.
2964 While n=0 is a special case, it is equivalent to "nosmp",
2965 which also disables the IO APIC.
2967 max_loop= [LOOP] The number of loop block devices that get
2968 (loop.max_loop) unconditionally pre-created at init time. The default
2969 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead
2970 of statically allocating a predefined number, loop
2971 devices can be requested on-demand with the
2972 /dev/loop-control interface.
2974 mce [X86-32] Machine Check Exception
2976 mce=option [X86-64] See Documentation/arch/x86/x86_64/boot-options.rst
2978 md= [HW] RAID subsystems devices and level
2979 See Documentation/admin-guide/md.rst.
2982 Format: <first>,<last>
2983 Specifies range of consoles to be captured by the MDA.
2986 Control mitigation for the Micro-architectural Data
2987 Sampling (MDS) vulnerability.
2989 Certain CPUs are vulnerable to an exploit against CPU
2990 internal buffers which can forward information to a
2991 disclosure gadget under certain conditions.
2993 In vulnerable processors, the speculatively
2994 forwarded data can be used in a cache side channel
2995 attack, to access data to which the attacker does
2996 not have direct access.
2998 This parameter controls the MDS mitigation. The
3001 full - Enable MDS mitigation on vulnerable CPUs
3002 full,nosmt - Enable MDS mitigation and disable
3003 SMT on vulnerable CPUs
3004 off - Unconditionally disable MDS mitigation
3006 On TAA-affected machines, mds=off can be prevented by
3007 an active TAA mitigation as both vulnerabilities are
3008 mitigated with the same mechanism so in order to disable
3009 this mitigation, you need to specify tsx_async_abort=off
3012 Not specifying this option is equivalent to
3015 For details see: Documentation/admin-guide/hw-vuln/mds.rst
3017 mem=nn[KMG] [HEXAGON] Set the memory size.
3018 Must be specified, otherwise memory size will be 0.
3020 mem=nn[KMG] [KNL,BOOT] Force usage of a specific amount of memory
3021 Amount of memory to be used in cases as follows:
3024 2 when the kernel is not able to see the whole system memory;
3025 3 memory that lies after 'mem=' boundary is excluded from
3026 the hypervisor, then assigned to KVM guests.
3027 4 to limit the memory available for kdump kernel.
3029 [ARC,MICROBLAZE] - the limit applies only to low memory,
3030 high memory is not affected.
3032 [ARM64] - only limits memory covered by the linear
3033 mapping. The NOMAP regions are not affected.
3035 [X86] Work as limiting max address. Use together
3036 with memmap= to avoid physical address space collisions.
3037 Without memmap= PCI devices could be placed at addresses
3038 belonging to unused RAM.
3040 Note that this only takes effects during boot time since
3041 in above case 3, memory may need be hot added after boot
3042 if system memory of hypervisor is not sufficient.
3045 [ARM,MIPS] - override the memory layout reported by
3047 Define a memory region of size nn[KMG] starting at
3049 Multiple different regions can be specified with
3050 multiple mem= parameters on the command line.
3052 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel
3055 memblock=debug [KNL] Enable memblock debug messages.
3058 [KNL,SH] Allow user to override the default size for
3059 per-device physically contiguous DMA buffers.
3061 memhp_default_state=online/offline
3062 [KNL] Set the initial state for the memory hotplug
3063 onlining policy. If not specified, the default value is
3064 set according to the
3065 CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE kernel config
3067 See Documentation/admin-guide/mm/memory-hotplug.rst.
3069 memmap=exactmap [KNL,X86] Enable setting of an exact
3070 E820 memory map, as specified by the user.
3071 Such memmap=exactmap lines can be constructed based on
3072 BIOS output or other requirements. See the memmap=nn@ss
3075 memmap=nn[KMG]@ss[KMG]
3076 [KNL, X86, MIPS, XTENSA] Force usage of a specific region of memory.
3077 Region of memory to be used is from ss to ss+nn.
3078 If @ss[KMG] is omitted, it is equivalent to mem=nn[KMG],
3079 which limits max address to nn[KMG].
3080 Multiple different regions can be specified,
3083 memmap=100M@2G,100M#3G,1G!1024G
3085 memmap=nn[KMG]#ss[KMG]
3086 [KNL,ACPI] Mark specific memory as ACPI data.
3087 Region of memory to be marked is from ss to ss+nn.
3089 memmap=nn[KMG]$ss[KMG]
3090 [KNL,ACPI] Mark specific memory as reserved.
3091 Region of memory to be reserved is from ss to ss+nn.
3092 Example: Exclude memory from 0x18690000-0x1869ffff
3093 memmap=64K$0x18690000
3095 memmap=0x10000$0x18690000
3096 Some bootloaders may need an escape character before '$',
3097 like Grub2, otherwise '$' and the following number
3100 memmap=nn[KMG]!ss[KMG]
3101 [KNL,X86] Mark specific memory as protected.
3102 Region of memory to be used, from ss to ss+nn.
3103 The memory region may be marked as e820 type 12 (0xc)
3104 and is NVDIMM or ADR memory.
3106 memmap=<size>%<offset>-<oldtype>+<newtype>
3107 [KNL,ACPI] Convert memory within the specified region
3108 from <oldtype> to <newtype>. If "-<oldtype>" is left
3109 out, the whole region will be marked as <newtype>,
3110 even if previously unavailable. If "+<newtype>" is left
3111 out, matching memory will be removed. Types are
3112 specified as e820 types, e.g., 1 = RAM, 2 = reserved,
3113 3 = ACPI, 12 = PRAM.
3115 memory_corruption_check=0/1 [X86]
3116 Some BIOSes seem to corrupt the first 64k of
3117 memory when doing things like suspend/resume.
3118 Setting this option will scan the memory
3119 looking for corruption. Enabling this will
3120 both detect corruption and prevent the kernel
3121 from using the memory being corrupted.
3122 However, its intended as a diagnostic tool; if
3123 repeatable BIOS-originated corruption always
3124 affects the same memory, you can use memmap=
3125 to prevent the kernel from using that memory.
3127 memory_corruption_check_size=size [X86]
3128 By default it checks for corruption in the low
3129 64k, making this memory unavailable for normal
3130 use. Use this parameter to scan for
3131 corruption in more or less memory.
3133 memory_corruption_check_period=seconds [X86]
3134 By default it checks for corruption every 60
3135 seconds. Use this parameter to check at some
3136 other rate. 0 disables periodic checking.
3138 memory_hotplug.memmap_on_memory
3139 [KNL,X86,ARM] Boolean flag to enable this feature.
3140 Format: {on | off (default)}
3141 When enabled, runtime hotplugged memory will
3142 allocate its internal metadata (struct pages,
3143 those vmemmap pages cannot be optimized even
3144 if hugetlb_free_vmemmap is enabled) from the
3145 hotadded memory which will allow to hotadd a
3146 lot of memory without requiring additional
3148 This feature is disabled by default because it
3149 has some implication on large (e.g. GB)
3150 allocations in some configurations (e.g. small
3152 The state of the flag can be read in
3153 /sys/module/memory_hotplug/parameters/memmap_on_memory.
3154 Note that even when enabled, there are a few cases where
3155 the feature is not effective.
3157 memtest= [KNL,X86,ARM,M68K,PPC,RISCV] Enable memtest
3159 default : 0 <disable>
3160 Specifies the number of memtest passes to be
3161 performed. Each pass selects another test
3162 pattern from a given set of patterns. Memtest
3163 fills the memory with this pattern, validates
3164 memory contents and reserves bad memory
3165 regions that are detected.
3167 mem_encrypt= [X86-64] AMD Secure Memory Encryption (SME) control
3168 Valid arguments: on, off
3169 Default (depends on kernel configuration option):
3170 on (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=y)
3171 off (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=n)
3172 mem_encrypt=on: Activate SME
3173 mem_encrypt=off: Do not activate SME
3175 Refer to Documentation/virt/kvm/x86/amd-memory-encryption.rst
3176 for details on when memory encryption can be activated.
3178 mem_sleep_default= [SUSPEND] Default system suspend mode:
3179 s2idle - Suspend-To-Idle
3180 shallow - Power-On Suspend or equivalent (if supported)
3181 deep - Suspend-To-RAM or equivalent (if supported)
3182 See Documentation/admin-guide/pm/sleep-states.rst.
3184 mfgpt_irq= [IA-32] Specify the IRQ to use for the
3185 Multi-Function General Purpose Timers on AMD Geode
3188 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when
3189 the BIOS has incorrectly applied a workaround. TinyBIOS
3190 version 0.98 is known to be affected, 0.99 fixes the
3191 problem by letting the user disable the workaround.
3195 min_addr=nn[KMG] [KNL,BOOT,IA-64] All physical memory below this
3196 physical address is ignored.
3198 mini2440= [ARM,HW,KNL]
3199 Format:[0..2][b][c][t]
3201 MINI2440 configuration specification:
3202 0 - The attached screen is the 3.5" TFT
3203 1 - The attached screen is the 7" TFT
3204 2 - The VGA Shield is attached (1024x768)
3205 Leaving out the screen size parameter will not load
3206 the TFT driver, and the framebuffer will be left
3208 b - Enable backlight. The TFT backlight pin will be
3209 linked to the kernel VESA blanking code and a GPIO
3210 LED. This parameter is not necessary when using the
3212 c - Enable the s3c camera interface.
3213 t - Reserved for enabling touchscreen support. The
3214 touchscreen support is not enabled in the mainstream
3215 kernel as of 2.6.30, a preliminary port can be found
3216 in the "bleeding edge" mini2440 support kernel at
3217 https://repo.or.cz/w/linux-2.6/mini2440.git
3220 [X86,PPC,S390,ARM64] Control optional mitigations for
3221 CPU vulnerabilities. This is a set of curated,
3222 arch-independent options, each of which is an
3223 aggregation of existing arch-specific options.
3226 Disable all optional CPU mitigations. This
3227 improves system performance, but it may also
3228 expose users to several CPU vulnerabilities.
3229 Equivalent to: nopti [X86,PPC]
3230 if nokaslr then kpti=0 [ARM64]
3231 nospectre_v1 [X86,PPC]
3233 nospectre_v2 [X86,PPC,S390,ARM64]
3234 spectre_v2_user=off [X86]
3235 spec_store_bypass_disable=off [X86,PPC]
3236 ssbd=force-off [ARM64]
3237 nospectre_bhb [ARM64]
3240 tsx_async_abort=off [X86]
3241 kvm.nx_huge_pages=off [X86]
3242 srbds=off [X86,INTEL]
3243 no_entry_flush [PPC]
3244 no_uaccess_flush [PPC]
3245 mmio_stale_data=off [X86]
3249 This does not have any effect on
3250 kvm.nx_huge_pages when
3251 kvm.nx_huge_pages=force.
3254 Mitigate all CPU vulnerabilities, but leave SMT
3255 enabled, even if it's vulnerable. This is for
3256 users who don't want to be surprised by SMT
3257 getting disabled across kernel upgrades, or who
3258 have other ways of avoiding SMT-based attacks.
3259 Equivalent to: (default behavior)
3262 Mitigate all CPU vulnerabilities, disabling SMT
3263 if needed. This is for users who always want to
3264 be fully mitigated, even if it means losing SMT.
3265 Equivalent to: l1tf=flush,nosmt [X86]
3266 mds=full,nosmt [X86]
3267 tsx_async_abort=full,nosmt [X86]
3268 mmio_stale_data=full,nosmt [X86]
3269 retbleed=auto,nosmt [X86]
3272 [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this
3273 parameter allows control of the logging verbosity for
3274 the additional memory initialisation checks. A value
3275 of 0 disables mminit logging and a level of 4 will
3276 log everything. Information is printed at KERN_DEBUG
3277 so loglevel=8 may also need to be specified.
3280 [X86,INTEL] Control mitigation for the Processor
3281 MMIO Stale Data vulnerabilities.
3283 Processor MMIO Stale Data is a class of
3284 vulnerabilities that may expose data after an MMIO
3285 operation. Exposed data could originate or end in
3286 the same CPU buffers as affected by MDS and TAA.
3287 Therefore, similar to MDS and TAA, the mitigation
3288 is to clear the affected CPU buffers.
3290 This parameter controls the mitigation. The
3293 full - Enable mitigation on vulnerable CPUs
3295 full,nosmt - Enable mitigation and disable SMT on
3298 off - Unconditionally disable mitigation
3300 On MDS or TAA affected machines,
3301 mmio_stale_data=off can be prevented by an active
3302 MDS or TAA mitigation as these vulnerabilities are
3303 mitigated with the same mechanism so in order to
3304 disable this mitigation, you need to specify
3305 mds=off and tsx_async_abort=off too.
3307 Not specifying this option is equivalent to
3308 mmio_stale_data=full.
3311 Documentation/admin-guide/hw-vuln/processor_mmio_stale_data.rst
3313 <module>.async_probe[=<bool>] [KNL]
3314 If no <bool> value is specified or if the value
3315 specified is not a valid <bool>, enable asynchronous
3316 probe on this module. Otherwise, enable/disable
3317 asynchronous probe on this module as indicated by the
3318 <bool> value. See also: module.async_probe
3320 module.async_probe=<bool>
3321 [KNL] When set to true, modules will use async probing
3322 by default. To enable/disable async probing for a
3323 specific module, use the module specific control that
3324 is documented under <module>.async_probe. When both
3325 module.async_probe and <module>.async_probe are
3326 specified, <module>.async_probe takes precedence for
3327 the specific module.
3330 [KNL] When CONFIG_MODULE_SIG is set, this means that
3331 modules without (valid) signatures will fail to load.
3332 Note that if CONFIG_MODULE_SIG_FORCE is set, that
3333 is always true, so this option does nothing.
3335 module_blacklist= [KNL] Do not load a comma-separated list of
3336 modules. Useful for debugging problem modules.
3339 [MOUSE] Maximum time between finger touching and
3340 leaving touchpad surface for touch to be considered
3341 a tap and be reported as a left button click (for
3342 touchpads working in absolute mode only).
3344 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices
3345 reporting absolute coordinates, such as tablets
3346 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices
3347 reporting absolute coordinates, such as tablets
3349 movablecore= [KNL,X86,IA-64,PPC]
3350 Format: nn[KMGTPE] | nn%
3351 This parameter is the complement to kernelcore=, it
3352 specifies the amount of memory used for migratable
3353 allocations. If both kernelcore and movablecore is
3354 specified, then kernelcore will be at *least* the
3355 specified value but may be more. If movablecore on its
3356 own is specified, the administrator must be careful
3357 that the amount of memory usable for all allocations
3360 movable_node [KNL] Boot-time switch to make hotplugable memory
3361 NUMA nodes to be movable. This means that the memory
3362 of such nodes will be usable only for movable
3363 allocations which rules out almost all kernel
3364 allocations. Use with caution!
3366 MTD_Partition= [MTD]
3367 Format: <name>,<region-number>,<size>,<offset>
3369 MTD_Region= [MTD] Format:
3370 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>]
3373 See drivers/mtd/parsers/cmdlinepart.c
3376 ARM/S3C2412 JIVE boot control
3378 See arch/arm/mach-s3c/mach-jive.c
3380 mtouchusb.raw_coordinates=
3381 [HW] Make the MicroTouch USB driver use raw coordinates
3382 ('y', default) or cooked coordinates ('n')
3384 mtrr_chunk_size=nn[KMG] [X86]
3385 used for mtrr cleanup. It is largest continuous chunk
3386 that could hold holes aka. UC entries.
3388 mtrr_gran_size=nn[KMG] [X86]
3389 Used for mtrr cleanup. It is granularity of mtrr block.
3391 Large value could prevent small alignment from
3394 mtrr_spare_reg_nr=n [X86]
3396 Range: 0,7 : spare reg number
3398 Used for mtrr cleanup. It is spare mtrr entries number.
3399 Set to 2 or more if your graphical card needs more.
3401 multitce=off [PPC] This parameter disables the use of the pSeries
3402 firmware feature for updating multiple TCE entries
3405 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card
3407 netdev= [NET] Network devices parameters
3408 Format: <irq>,<io>,<mem_start>,<mem_end>,<name>
3409 Note that mem_start is often overloaded to mean
3410 something different and driver-specific.
3411 This usage is only documented in each driver source
3414 netpoll.carrier_timeout=
3415 [NET] Specifies amount of time (in seconds) that
3416 netpoll should wait for a carrier. By default netpoll
3420 [NETFILTER] Enable connection tracking flow accounting
3421 0 to disable accounting
3422 1 to enable accounting
3426 [NFS] sets the pathname to the program which is used
3427 to update the NFS client cache entries.
3429 nfs.cache_getent_timeout=
3430 [NFS] sets the timeout after which an attempt to
3431 update a cache entry is deemed to have failed.
3433 nfs.callback_nr_threads=
3434 [NFSv4] set the total number of threads that the
3435 NFS client will assign to service NFSv4 callback
3438 nfs.callback_tcpport=
3439 [NFS] set the TCP port on which the NFSv4 callback
3440 channel should listen.
3443 [NFS] enable 64-bit inode numbers.
3444 If zero, the NFS client will fake up a 32-bit inode
3445 number for the readdir() and stat() syscalls instead
3446 of returning the full 64-bit number.
3447 The default is to return 64-bit inode numbers.
3449 nfs.idmap_cache_timeout=
3450 [NFS] set the maximum lifetime for idmapper cache
3453 nfs.max_session_cb_slots=
3454 [NFSv4.1] Sets the maximum number of session
3455 slots the client will assign to the callback
3456 channel. This determines the maximum number of
3457 callbacks the client will process in parallel for
3458 a particular server.
3460 nfs.max_session_slots=
3461 [NFSv4.1] Sets the maximum number of session slots
3462 the client will attempt to negotiate with the server.
3463 This limits the number of simultaneous RPC requests
3464 that the client can send to the NFSv4.1 server.
3465 Note that there is little point in setting this
3466 value higher than the max_tcp_slot_table_limit.
3468 nfs.nfs4_disable_idmapping=
3469 [NFSv4] When set to the default of '1', this option
3470 ensures that both the RPC level authentication
3471 scheme and the NFS level operations agree to use
3472 numeric uids/gids if the mount is using the
3473 'sec=sys' security flavour. In effect it is
3474 disabling idmapping, which can make migration from
3475 legacy NFSv2/v3 systems to NFSv4 easier.
3476 Servers that do not support this mode of operation
3477 will be autodetected by the client, and it will fall
3478 back to using the idmapper.
3479 To turn off this behaviour, set the value to '0'.
3482 [NFS4] Specify an additional fixed unique ident-
3483 ification string that NFSv4 clients can insert into
3484 their nfs_client_id4 string. This is typically a
3485 UUID that is generated at system install time.
3487 nfs.recover_lost_locks=
3488 [NFSv4] Attempt to recover locks that were lost due
3489 to a lease timeout on the server. Please note that
3490 doing this risks data corruption, since there are
3491 no guarantees that the file will remain unchanged
3492 after the locks are lost.
3493 If you want to enable the kernel legacy behaviour of
3494 attempting to recover these locks, then set this
3496 The default parameter value of '0' causes the kernel
3497 not to attempt recovery of lost locks.
3499 nfs.send_implementation_id=
3500 [NFSv4.1] Send client implementation identification
3501 information in exchange_id requests.
3502 If zero, no implementation identification information
3504 The default is to send the implementation identification
3507 nfs4.layoutstats_timer=
3508 [NFSv4.2] Change the rate at which the kernel sends
3509 layoutstats to the pNFS metadata server.
3511 Setting this to value to 0 causes the kernel to use
3512 whatever value is the default set by the layout
3513 driver. A non-zero value sets the minimum interval
3514 in seconds between layoutstats transmissions.
3516 nfsd.inter_copy_offload_enable=
3517 [NFSv4.2] When set to 1, the server will support
3518 server-to-server copies for which this server is
3519 the destination of the copy.
3521 nfsd.nfs4_disable_idmapping=
3522 [NFSv4] When set to the default of '1', the NFSv4
3523 server will return only numeric uids and gids to
3524 clients using auth_sys, and will accept numeric uids
3525 and gids from such clients. This is intended to ease
3526 migration from NFSv2/v3.
3528 nfsd.nfsd4_ssc_umount_timeout=
3529 [NFSv4.2] When used as the destination of a
3530 server-to-server copy, knfsd temporarily mounts
3531 the source server. It caches the mount in case
3532 it will be needed again, and discards it if not
3533 used for the number of milliseconds specified by
3536 nfsaddrs= [NFS] Deprecated. Use ip= instead.
3537 See Documentation/admin-guide/nfs/nfsroot.rst.
3539 nfsroot= [NFS] nfs root filesystem for disk-less boxes.
3540 See Documentation/admin-guide/nfs/nfsroot.rst.
3542 nfsrootdebug [NFS] enable nfsroot debugging messages.
3543 See Documentation/admin-guide/nfs/nfsroot.rst.
3545 nmi_backtrace.backtrace_idle [KNL]
3546 Dump stacks even of idle CPUs in response to an
3547 NMI stack-backtrace request.
3549 nmi_debug= [KNL,SH] Specify one or more actions to take
3550 when a NMI is triggered.
3551 Format: [state][,regs][,debounce][,die]
3553 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels
3554 Format: [panic,][nopanic,][num]
3556 0 - turn hardlockup detector in nmi_watchdog off
3557 1 - turn hardlockup detector in nmi_watchdog on
3558 When panic is specified, panic when an NMI watchdog
3559 timeout occurs (or 'nopanic' to not panic on an NMI
3560 watchdog, if CONFIG_BOOTPARAM_HARDLOCKUP_PANIC is set)
3561 To disable both hard and soft lockup detectors,
3562 please see 'nowatchdog'.
3563 This is useful when you use a panic=... timeout and
3564 need the box quickly up again.
3566 These settings can be accessed at runtime via
3567 the nmi_watchdog and hardlockup_panic sysctls.
3569 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths
3570 emulation library even if a 387 maths coprocessor
3573 no5lvl [X86-64] Disable 5-level paging mode. Forces
3574 kernel to use 4-level paging instead.
3576 noaliencache [MM, NUMA, SLAB] Disables the allocation of alien
3577 caches in the slab allocator. Saves per-node memory,
3578 but will impact performance.
3582 noaltinstr [S390] Disables alternative instructions patching
3583 (CPU alternatives feature).
3585 noapic [SMP,APIC] Tells the kernel to not make use of any
3586 IOAPICs that may be present in the system.
3588 noautogroup Disable scheduler automatic task group creation.
3593 [HW] Never suspend the console
3594 Disable suspending of consoles during suspend and
3595 hibernate operations. Once disabled, debugging
3596 messages can reach various consoles while the rest
3597 of the system is being put to sleep (ie, while
3598 debugging driver suspend/resume hooks). This may
3599 not work reliably with all consoles, but is known
3600 to work with serial and VGA consoles.
3601 To facilitate more flexible debugging, we also add
3602 console_suspend, a printk module parameter to control
3603 it. Users could use console_suspend (usually
3604 /sys/module/printk/parameters/console_suspend) to
3605 turn on/off it dynamically.
3608 [KNL] Disable object debugging
3610 nodsp [SH] Disable hardware DSP at boot time.
3612 noefi Disable EFI runtime services support.
3614 no_entry_flush [PPC] Don't flush the L1-D cache when entering the kernel.
3619 This affects only 32-bit executables.
3620 noexec32=on: enable non-executable mappings (default)
3621 read doesn't imply executable mappings
3622 noexec32=off: disable non-executable mappings
3623 read implies executable mappings
3625 no_file_caps Tells the kernel not to honor file capabilities. The
3626 only way then for a file to be executed with privilege
3627 is to be setuid root or executed by root.
3629 nofpu [MIPS,SH] Disable hardware FPU at boot time.
3631 nofsgsbase [X86] Disables FSGSBASE instructions.
3633 nofxsr [BUGS=X86-32] Disables x86 floating point extended
3634 register save and restore. The kernel will only save
3635 legacy floating-point registers on task switch.
3637 nohalt [IA-64] Tells the kernel not to use the power saving
3638 function PAL_HALT_LIGHT when idle. This increases
3639 power-consumption. On the positive side, it reduces
3640 interrupt wake-up latency, which may improve performance
3641 in certain environments such as networked servers or
3645 Force pointers printed to the console or buffers to be
3646 unhashed. By default, when a pointer is printed via %p
3647 format string, that pointer is "hashed", i.e. obscured
3648 by hashing the pointer value. This is a security feature
3649 that hides actual kernel addresses from unprivileged
3650 users, but it also makes debugging the kernel more
3651 difficult since unequal pointers can no longer be
3652 compared. However, if this command-line option is
3653 specified, then all normal pointers will have their true
3654 value printed. This option should only be specified when
3655 debugging the kernel. Please do not use on production
3658 nohibernate [HIBERNATION] Disable hibernation and resume.
3660 nohlt [ARM,ARM64,MICROBLAZE,SH] Forces the kernel to busy wait
3661 in do_idle() and not use the arch_cpu_idle()
3662 implementation; requires CONFIG_GENERIC_IDLE_POLL_SETUP
3663 to be effective. This is useful on platforms where the
3664 sleep(SH) or wfi(ARM,ARM64) instructions do not work
3665 correctly or when doing power measurements to evaluate
3666 the impact of the sleep instructions. This is also
3667 useful when using JTAG debugger.
3669 nohugeiomap [KNL,X86,PPC,ARM64] Disable kernel huge I/O mappings.
3671 nohugevmalloc [KNL,X86,PPC,ARM64] Disable kernel huge vmalloc mappings.
3673 nohz= [KNL] Boottime enable/disable dynamic ticks
3674 Valid arguments: on, off
3677 nohz_full= [KNL,BOOT,SMP,ISOL]
3678 The argument is a cpu list, as described above.
3679 In kernels built with CONFIG_NO_HZ_FULL=y, set
3680 the specified list of CPUs whose tick will be stopped
3681 whenever possible. The boot CPU will be forced outside
3682 the range to maintain the timekeeping. Any CPUs
3683 in this list will have their RCU callbacks offloaded,
3684 just as if they had also been called out in the
3685 rcu_nocbs= boot parameter.
3687 Note that this argument takes precedence over
3688 the CONFIG_RCU_NOCB_CPU_DEFAULT_ALL option.
3690 noinitrd [RAM] Tells the kernel not to load any configured
3693 nointremap [X86-64, Intel-IOMMU] Do not enable interrupt
3695 [Deprecated - use intremap=off]
3699 noinvpcid [X86] Disable the INVPCID cpu feature.
3701 noiotrap [SH] Disables trapped I/O port accesses.
3703 noirqdebug [X86-32] Disables the code which attempts to detect and
3704 disable unhandled interrupt sources.
3706 noisapnp [ISAPNP] Disables ISA PnP code.
3708 nojitter [IA-64] Disables jitter checking for ITC timers.
3711 When CONFIG_RANDOMIZE_BASE is set, this disables
3712 kernel and module base offset ASLR (Address Space
3713 Layout Randomization).
3715 no-kvmapf [X86,KVM] Disable paravirtualized asynchronous page
3718 no-kvmclock [X86,KVM] Disable paravirtualized KVM clock driver
3720 nolapic [X86-32,APIC] Do not enable or use the local APIC.
3722 nolapic_timer [X86-32,APIC] Do not use the local APIC timer.
3724 nomca [IA-64] Disable machine check abort handling
3726 nomce [X86-32] Disable Machine Check Exception
3728 nomfgpt [X86-32] Disable Multi-Function General Purpose
3729 Timer usage (for AMD Geode machines).
3731 nomodeset Disable kernel modesetting. Most systems' firmware
3732 sets up a display mode and provides framebuffer memory
3733 for output. With nomodeset, DRM and fbdev drivers will
3734 not load if they could possibly displace the pre-
3735 initialized output. Only the system framebuffer will
3736 be available for use. The respective drivers will not
3737 perform display-mode changes or accelerated rendering.
3739 Useful as error fallback, or for testing and debugging.
3741 nomodule Disable module load
3743 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to
3744 shutdown the other cpus. Instead use the REBOOT_VECTOR
3747 nopat [X86] Disable PAT (page attribute table extension of
3748 pagetables) support.
3750 nopcid [X86-64] Disable the PCID cpu feature.
3752 nopku [X86] Disable Memory Protection Keys CPU feature found
3756 Equivalent to pti=off
3758 nopv= [X86,XEN,KVM,HYPER_V,VMWARE]
3759 Disables the PV optimizations forcing the guest to run
3760 as generic guest with no PV drivers. Currently support
3761 XEN HVM, KVM, HYPER_V and VMWARE guest.
3763 nopvspin [X86,XEN,KVM]
3764 Disables the qspinlock slow path using PV optimizations
3765 which allow the hypervisor to 'idle' the guest on lock
3768 norandmaps Don't use address space randomization. Equivalent to
3769 echo 0 > /proc/sys/kernel/randomize_va_space
3771 noreplace-smp [X86-32,SMP] Don't replace SMP instructions
3772 with UP alternatives
3774 noresume [SWSUSP] Disables resume and restores original swap
3779 no-scroll [VGA] Disables scrollback.
3780 This is required for the Braillex ib80-piezo Braille
3781 reader made by F.H. Papenmeier (Germany).
3783 nosgx [X86-64,SGX] Disables Intel SGX kernel support.
3786 Disable SMAP (Supervisor Mode Access Prevention)
3787 even if it is supported by processor.
3790 Disable SMEP (Supervisor Mode Execution Prevention)
3791 even if it is supported by processor.
3793 nosmp [SMP] Tells an SMP kernel to act as a UP kernel,
3794 and disable the IO APIC. legacy for "maxcpus=0".
3796 nosmt [KNL,S390] Disable symmetric multithreading (SMT).
3797 Equivalent to smt=1.
3799 [KNL,X86] Disable symmetric multithreading (SMT).
3800 nosmt=force: Force disable SMT, cannot be undone
3801 via the sysfs control file.
3803 nosoftlockup [KNL] Disable the soft-lockup detector.
3805 nospec_store_bypass_disable
3806 [HW] Disable all mitigations for the Speculative Store Bypass vulnerability
3808 nospectre_bhb [ARM64] Disable all mitigations for Spectre-BHB (branch
3809 history injection) vulnerability. System may allow data leaks
3812 nospectre_v1 [X86,PPC] Disable mitigations for Spectre Variant 1
3813 (bounds check bypass). With this option data leaks are
3814 possible in the system.
3816 nospectre_v2 [X86,PPC_E500,ARM64] Disable all mitigations for
3817 the Spectre variant 2 (indirect branch prediction)
3818 vulnerability. System may allow data leaks with this
3821 no-steal-acc [X86,PV_OPS,ARM64,PPC/PSERIES] Disable paravirtualized
3822 steal time accounting. steal time is computed, but
3823 won't influence scheduler behaviour
3825 nosync [HW,M68K] Disables sync negotiation for all devices.
3827 no_timer_check [X86,APIC] Disables the code which tests for
3828 broken timer IRQ sources.
3831 [PPC] Don't flush the L1-D cache after accessing user data.
3833 novmcoredd [KNL,KDUMP]
3834 Disable device dump. Device dump allows drivers to
3835 append dump data to vmcore so you can collect driver
3836 specified debug info. Drivers can append the data
3837 without any limit and this data is stored in memory,
3838 so this may cause significant memory stress. Disabling
3839 device dump can help save memory but the driver debug
3840 data will be no longer available. This parameter
3841 is only available when CONFIG_PROC_VMCORE_DEVICE_DUMP
3845 [X86,PV_OPS] Disable paravirtualized VMware scheduler
3846 clock and use the default one.
3848 nowatchdog [KNL] Disable both lockup detectors, i.e.
3849 soft-lockup and NMI watchdog (hard-lockup).
3853 nox2apic [X86-64,APIC] Do not enable x2APIC mode.
3855 NOTE: this parameter will be ignored on systems with the
3856 LEGACY_XAPIC_DISABLED bit set in the
3857 IA32_XAPIC_DISABLE_STATUS MSR.
3859 noxsave [BUGS=X86] Disables x86 extended register state save
3860 and restore using xsave. The kernel will fallback to
3861 enabling legacy floating-point and sse state.
3863 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended
3864 register states. The kernel will fall back to use
3865 xsave to save the states. By using this parameter,
3866 performance of saving the states is degraded because
3867 xsave doesn't support modified optimization while
3868 xsaveopt supports it on xsaveopt enabled systems.
3870 noxsaves [X86] Disables xsaves and xrstors used in saving and
3871 restoring x86 extended register state in compacted
3872 form of xsave area. The kernel will fall back to use
3873 xsaveopt and xrstor to save and restore the states
3874 in standard form of xsave area. By using this
3875 parameter, xsave area per process might occupy more
3876 memory on xsaves enabled systems.
3878 nps_mtm_hs_ctr= [KNL,ARC]
3879 This parameter sets the maximum duration, in
3880 cycles, each HW thread of the CTOP can run
3881 without interruptions, before HW switches it.
3882 The actual maximum duration is 16 times this
3884 Format: integer between 1 and 255
3887 nptcg= [IA-64] Override max number of concurrent global TLB
3888 purges which is reported from either PAL_VM_SUMMARY or
3891 nr_cpus= [SMP] Maximum number of processors that an SMP kernel
3892 could support. nr_cpus=n : n >= 1 limits the kernel to
3893 support 'n' processors. It could be larger than the
3894 number of already plugged CPU during bootup, later in
3895 runtime you can physically add extra cpu until it reaches
3896 n. So during boot up some boot time memory for per-cpu
3897 variables need be pre-allocated for later physical cpu
3900 nr_uarts= [SERIAL] maximum number of UARTs to be registered.
3902 numa=off [KNL, ARM64, PPC, RISCV, SPARC, X86] Disable NUMA, Only
3903 set up a single NUMA node spanning all memory.
3905 numa_balancing= [KNL,ARM64,PPC,RISCV,S390,X86] Enable or disable automatic
3907 Allowed values are enable and disable
3909 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA.
3910 'node', 'default' can be specified
3911 This can be set from sysctl after boot.
3912 See Documentation/admin-guide/sysctl/vm.rst for details.
3914 ohci1394_dma=early [HW] enable debugging via the ohci1394 driver.
3915 See Documentation/core-api/debugging-via-ohci1394.rst for more
3918 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands
3919 Rather than timing out after 20 ms if an EC
3920 command is not properly ACKed, override the length
3921 of the timeout. We have interrupts disabled while
3922 waiting for the ACK, so if this is set too high
3923 interrupts *may* be lost!
3925 omap_mux= [OMAP] Override bootloader pin multiplexing.
3926 Format: <mux_mode0.mode_name=value>...
3927 For example, to override I2C bus2:
3928 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100
3930 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration
3932 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock]
3934 boundary - index of last SLC block on Flex-OneNAND.
3935 The remaining blocks are configured as MLC blocks.
3936 lock - Configure if Flex-OneNAND boundary should be locked.
3937 Once locked, the boundary cannot be changed.
3938 1 indicates lock status, 0 indicates unlock status.
3940 oops=panic Always panic on oopses. Default is to just kill the
3941 process, but there is a small probability of
3942 deadlocking the machine.
3943 This will also cause panics on machine check exceptions.
3944 Useful together with panic=30 to trigger a reboot.
3947 [KNL] Boolean flag to control whether the page allocator
3948 should randomize its free lists. The randomization may
3949 be automatically enabled if the kernel detects it is
3950 running on a platform with a direct-mapped memory-side
3951 cache, and this parameter can be used to
3952 override/disable that behavior. The state of the flag
3953 can be read from sysfs at:
3954 /sys/module/page_alloc/parameters/shuffle.
3956 page_owner= [KNL] Boot-time page_owner enabling option.
3957 Storage of the information about who allocated
3958 each page is disabled in default. With this switch,
3960 on: enable the feature
3962 page_poison= [KNL] Boot-time parameter changing the state of
3963 poisoning on the buddy allocator, available with
3964 CONFIG_PAGE_POISONING=y.
3965 off: turn off poisoning (default)
3966 on: turn on poisoning
3968 page_reporting.page_reporting_order=
3969 [KNL] Minimal page reporting order
3971 Adjust the minimal page reporting order. The page
3972 reporting is disabled when it exceeds (MAX_ORDER-1).
3974 panic= [KNL] Kernel behaviour on panic: delay <timeout>
3975 timeout > 0: seconds before rebooting
3976 timeout = 0: wait forever
3977 timeout < 0: reboot immediately
3980 panic_print= Bitmask for printing system info when panic happens.
3981 User can chose combination of the following bits:
3982 bit 0: print all tasks info
3983 bit 1: print system memory info
3984 bit 2: print timer info
3985 bit 3: print locks info if CONFIG_LOCKDEP is on
3986 bit 4: print ftrace buffer
3987 bit 5: print all printk messages in buffer
3988 bit 6: print all CPUs backtrace (if available in the arch)
3989 *Be aware* that this option may print a _lot_ of lines,
3990 so there are risks of losing older messages in the log.
3991 Use this option carefully, maybe worth to setup a
3992 bigger log buffer with "log_buf_len" along with this.
3994 panic_on_taint= Bitmask for conditionally calling panic() in add_taint()
3995 Format: <hex>[,nousertaint]
3996 Hexadecimal bitmask representing the set of TAINT flags
3997 that will cause the kernel to panic when add_taint() is
3998 called with any of the flags in this set.
3999 The optional switch "nousertaint" can be utilized to
4000 prevent userspace forced crashes by writing to sysctl
4001 /proc/sys/kernel/tainted any flagset matching with the
4002 bitmask set on panic_on_taint.
4003 See Documentation/admin-guide/tainted-kernels.rst for
4004 extra details on the taint flags that users can pick
4005 to compose the bitmask to assign to panic_on_taint.
4007 panic_on_warn panic() instead of WARN(). Useful to cause kdump
4010 parkbd.port= [HW] Parallel port number the keyboard adapter is
4011 connected to, default is 0.
4013 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation,
4014 0 for XT, 1 for AT (default is AT).
4017 parport= [HW,PPT] Specify parallel ports. 0 disables.
4018 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] }
4019 Use 'auto' to force the driver to use any
4020 IRQ/DMA settings detected (the default is to
4021 ignore detected IRQ/DMA settings because of
4022 possible conflicts). You can specify the base
4023 address, IRQ, and DMA settings; IRQ and DMA
4024 should be numbers, or 'auto' (for using detected
4025 settings on that particular port), or 'nofifo'
4026 (to avoid using a FIFO even if it is detected).
4027 Parallel ports are assigned in the order they
4028 are specified on the command line, starting
4031 parport_init_mode= [HW,PPT]
4032 Configure VIA parallel port to operate in
4033 a specific mode. This is necessary on Pegasos
4034 computer where firmware has no options for setting
4035 up parallel port mode and sets it to spp.
4036 Currently this function knows 686a and 8231 chips.
4037 Format: [spp|ps2|epp|ecp|ecpepp]
4039 pata_legacy.all= [HW,LIBATA]
4041 Set to non-zero to probe primary and secondary ISA
4042 port ranges on PCI systems where no PCI PATA device
4043 has been found at either range. Disabled by default.
4045 pata_legacy.autospeed= [HW,LIBATA]
4047 Set to non-zero if a chip is present that snoops speed
4048 changes. Disabled by default.
4050 pata_legacy.ht6560a= [HW,LIBATA]
4052 Set to 1, 2, or 3 for HT 6560A on the primary channel,
4053 the secondary channel, or both channels respectively.
4054 Disabled by default.
4056 pata_legacy.ht6560b= [HW,LIBATA]
4058 Set to 1, 2, or 3 for HT 6560B on the primary channel,
4059 the secondary channel, or both channels respectively.
4060 Disabled by default.
4062 pata_legacy.iordy_mask= [HW,LIBATA]
4064 IORDY enable mask. Set individual bits to allow IORDY
4065 for the respective channel. Bit 0 is for the first
4066 legacy channel handled by this driver, bit 1 is for
4067 the second channel, and so on. The sequence will often
4068 correspond to the primary legacy channel, the secondary
4069 legacy channel, and so on, but the handling of a PCI
4070 bus and the use of other driver options may interfere
4071 with the sequence. By default IORDY is allowed across
4074 pata_legacy.opti82c46x= [HW,LIBATA]
4076 Set to 1, 2, or 3 for Opti 82c611A on the primary
4077 channel, the secondary channel, or both channels
4078 respectively. Disabled by default.
4080 pata_legacy.opti82c611a= [HW,LIBATA]
4082 Set to 1, 2, or 3 for Opti 82c465MV on the primary
4083 channel, the secondary channel, or both channels
4084 respectively. Disabled by default.
4086 pata_legacy.pio_mask= [HW,LIBATA]
4088 PIO mode mask for autospeed devices. Set individual
4089 bits to allow the use of the respective PIO modes.
4090 Bit 0 is for mode 0, bit 1 is for mode 1, and so on.
4091 All modes allowed by default.
4093 pata_legacy.probe_all= [HW,LIBATA]
4095 Set to non-zero to probe tertiary and further ISA
4096 port ranges on PCI systems. Disabled by default.
4098 pata_legacy.probe_mask= [HW,LIBATA]
4100 Probe mask for legacy ISA PATA ports. Depending on
4101 platform configuration and the use of other driver
4102 options up to 6 legacy ports are supported: 0x1f0,
4103 0x170, 0x1e8, 0x168, 0x1e0, 0x160, however probing
4104 of individual ports can be disabled by setting the
4105 corresponding bits in the mask to 1. Bit 0 is for
4106 the first port in the list above (0x1f0), and so on.
4107 By default all supported ports are probed.
4109 pata_legacy.qdi= [HW,LIBATA]
4111 Set to non-zero to probe QDI controllers. By default
4112 set to 1 if CONFIG_PATA_QDI_MODULE, 0 otherwise.
4114 pata_legacy.winbond= [HW,LIBATA]
4116 Set to non-zero to probe Winbond controllers. Use
4117 the standard I/O port (0x130) if 1, otherwise the
4118 value given is the I/O port to use (typically 0x1b0).
4119 By default set to 1 if CONFIG_PATA_WINBOND_VLB_MODULE,
4122 pata_platform.pio_mask= [HW,LIBATA]
4124 Supported PIO mode mask. Set individual bits to allow
4125 the use of the respective PIO modes. Bit 0 is for
4126 mode 0, bit 1 is for mode 1, and so on. Mode 0 only
4130 Halt all CPUs after the first oops has been printed for
4131 the specified number of seconds. This is to be used if
4132 your oopses keep scrolling off the screen.
4136 pci=option[,option...] [PCI] various PCI subsystem options.
4138 Some options herein operate on a specific device
4139 or a set of devices (<pci_dev>). These are
4140 specified in one of the following formats:
4142 [<domain>:]<bus>:<dev>.<func>[/<dev>.<func>]*
4143 pci:<vendor>:<device>[:<subvendor>:<subdevice>]
4145 Note: the first format specifies a PCI
4146 bus/device/function address which may change
4147 if new hardware is inserted, if motherboard
4148 firmware changes, or due to changes caused
4149 by other kernel parameters. If the
4150 domain is left unspecified, it is
4151 taken to be zero. Optionally, a path
4152 to a device through multiple device/function
4153 addresses can be specified after the base
4154 address (this is more robust against
4155 renumbering issues). The second format
4156 selects devices using IDs from the
4157 configuration space which may match multiple
4158 devices in the system.
4160 earlydump dump PCI config space before the kernel
4162 off [X86] don't probe for the PCI bus
4163 bios [X86-32] force use of PCI BIOS, don't access
4164 the hardware directly. Use this if your machine
4165 has a non-standard PCI host bridge.
4166 nobios [X86-32] disallow use of PCI BIOS, only direct
4167 hardware access methods are allowed. Use this
4168 if you experience crashes upon bootup and you
4169 suspect they are caused by the BIOS.
4170 conf1 [X86] Force use of PCI Configuration Access
4171 Mechanism 1 (config address in IO port 0xCF8,
4172 data in IO port 0xCFC, both 32-bit).
4173 conf2 [X86] Force use of PCI Configuration Access
4174 Mechanism 2 (IO port 0xCF8 is an 8-bit port for
4175 the function, IO port 0xCFA, also 8-bit, sets
4176 bus number. The config space is then accessed
4177 through ports 0xC000-0xCFFF).
4178 See http://wiki.osdev.org/PCI for more info
4179 on the configuration access mechanisms.
4180 noaer [PCIE] If the PCIEAER kernel config parameter is
4181 enabled, this kernel boot option can be used to
4182 disable the use of PCIE advanced error reporting.
4183 nodomains [PCI] Disable support for multiple PCI
4184 root domains (aka PCI segments, in ACPI-speak).
4185 nommconf [X86] Disable use of MMCONFIG for PCI
4187 check_enable_amd_mmconf [X86] check for and enable
4188 properly configured MMIO access to PCI
4189 config space on AMD family 10h CPU
4190 nomsi [MSI] If the PCI_MSI kernel config parameter is
4191 enabled, this kernel boot option can be used to
4192 disable the use of MSI interrupts system-wide.
4193 noioapicquirk [APIC] Disable all boot interrupt quirks.
4194 Safety option to keep boot IRQs enabled. This
4195 should never be necessary.
4196 ioapicreroute [APIC] Enable rerouting of boot IRQs to the
4197 primary IO-APIC for bridges that cannot disable
4198 boot IRQs. This fixes a source of spurious IRQs
4199 when the system masks IRQs.
4200 noioapicreroute [APIC] Disable workaround that uses the
4201 boot IRQ equivalent of an IRQ that connects to
4202 a chipset where boot IRQs cannot be disabled.
4203 The opposite of ioapicreroute.
4204 biosirq [X86-32] Use PCI BIOS calls to get the interrupt
4205 routing table. These calls are known to be buggy
4206 on several machines and they hang the machine
4207 when used, but on other computers it's the only
4208 way to get the interrupt routing table. Try
4209 this option if the kernel is unable to allocate
4210 IRQs or discover secondary PCI buses on your
4212 rom [X86] Assign address space to expansion ROMs.
4213 Use with caution as certain devices share
4214 address decoders between ROMs and other
4216 norom [X86] Do not assign address space to
4217 expansion ROMs that do not already have
4218 BIOS assigned address ranges.
4219 nobar [X86] Do not assign address space to the
4220 BARs that weren't assigned by the BIOS.
4221 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be
4222 assigned automatically to PCI devices. You can
4223 make the kernel exclude IRQs of your ISA cards
4225 pirqaddr=0xAAAAA [X86] Specify the physical address
4226 of the PIRQ table (normally generated
4227 by the BIOS) if it is outside the
4228 F0000h-100000h range.
4229 lastbus=N [X86] Scan all buses thru bus #N. Can be
4230 useful if the kernel is unable to find your
4231 secondary buses and you want to tell it
4232 explicitly which ones they are.
4233 assign-busses [X86] Always assign all PCI bus
4234 numbers ourselves, overriding
4235 whatever the firmware may have done.
4236 usepirqmask [X86] Honor the possible IRQ mask stored
4237 in the BIOS $PIR table. This is needed on
4238 some systems with broken BIOSes, notably
4239 some HP Pavilion N5400 and Omnibook XE3
4240 notebooks. This will have no effect if ACPI
4241 IRQ routing is enabled.
4242 noacpi [X86] Do not use ACPI for IRQ routing
4243 or for PCI scanning.
4244 use_crs [X86] Use PCI host bridge window information
4245 from ACPI. On BIOSes from 2008 or later, this
4246 is enabled by default. If you need to use this,
4247 please report a bug.
4248 nocrs [X86] Ignore PCI host bridge windows from ACPI.
4249 If you need to use this, please report a bug.
4250 use_e820 [X86] Use E820 reservations to exclude parts of
4251 PCI host bridge windows. This is a workaround
4252 for BIOS defects in host bridge _CRS methods.
4253 If you need to use this, please report a bug to
4254 <linux-pci@vger.kernel.org>.
4255 no_e820 [X86] Ignore E820 reservations for PCI host
4256 bridge windows. This is the default on modern
4257 hardware. If you need to use this, please report
4258 a bug to <linux-pci@vger.kernel.org>.
4259 routeirq Do IRQ routing for all PCI devices.
4260 This is normally done in pci_enable_device(),
4261 so this option is a temporary workaround
4262 for broken drivers that don't call it.
4263 skip_isa_align [X86] do not align io start addr, so can
4264 handle more pci cards
4265 noearly [X86] Don't do any early type 1 scanning.
4266 This might help on some broken boards which
4267 machine check when some devices' config space
4268 is read. But various workarounds are disabled
4269 and some IOMMU drivers will not work.
4270 bfsort Sort PCI devices into breadth-first order.
4271 This sorting is done to get a device
4272 order compatible with older (<= 2.4) kernels.
4273 nobfsort Don't sort PCI devices into breadth-first order.
4274 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size)
4275 tuning and use the BIOS-configured MPS defaults.
4276 pcie_bus_safe Set every device's MPS to the largest value
4277 supported by all devices below the root complex.
4278 pcie_bus_perf Set device MPS to the largest allowable MPS
4279 based on its parent bus. Also set MRRS (Max
4280 Read Request Size) to the largest supported
4281 value (no larger than the MPS that the device
4282 or bus can support) for best performance.
4283 pcie_bus_peer2peer Set every device's MPS to 128B, which
4284 every device is guaranteed to support. This
4285 configuration allows peer-to-peer DMA between
4286 any pair of devices, possibly at the cost of
4287 reduced performance. This also guarantees
4288 that hot-added devices will work.
4289 cbiosize=nn[KMG] The fixed amount of bus space which is
4290 reserved for the CardBus bridge's IO window.
4291 The default value is 256 bytes.
4292 cbmemsize=nn[KMG] The fixed amount of bus space which is
4293 reserved for the CardBus bridge's memory
4294 window. The default value is 64 megabytes.
4297 [<order of align>@]<pci_dev>[; ...]
4298 Specifies alignment and device to reassign
4299 aligned memory resources. How to
4300 specify the device is described above.
4301 If <order of align> is not specified,
4302 PAGE_SIZE is used as alignment.
4303 A PCI-PCI bridge can be specified if resource
4304 windows need to be expanded.
4305 To specify the alignment for several
4306 instances of a device, the PCI vendor,
4307 device, subvendor, and subdevice may be
4308 specified, e.g., 12@pci:8086:9c22:103c:198f
4309 for 4096-byte alignment.
4310 ecrc= Enable/disable PCIe ECRC (transaction layer
4311 end-to-end CRC checking). Only effective if
4312 OS has native AER control (either granted by
4313 ACPI _OSC or forced via "pcie_ports=native")
4314 bios: Use BIOS/firmware settings. This is the
4318 hpiosize=nn[KMG] The fixed amount of bus space which is
4319 reserved for hotplug bridge's IO window.
4320 Default size is 256 bytes.
4321 hpmmiosize=nn[KMG] The fixed amount of bus space which is
4322 reserved for hotplug bridge's MMIO window.
4323 Default size is 2 megabytes.
4324 hpmmioprefsize=nn[KMG] The fixed amount of bus space which is
4325 reserved for hotplug bridge's MMIO_PREF window.
4326 Default size is 2 megabytes.
4327 hpmemsize=nn[KMG] The fixed amount of bus space which is
4328 reserved for hotplug bridge's MMIO and
4330 Default size is 2 megabytes.
4331 hpbussize=nn The minimum amount of additional bus numbers
4332 reserved for buses below a hotplug bridge.
4334 realloc= Enable/disable reallocating PCI bridge resources
4335 if allocations done by BIOS are too small to
4336 accommodate resources required by all child
4338 off: Turn realloc off
4340 realloc same as realloc=on
4341 noari do not use PCIe ARI.
4342 noats [PCIE, Intel-IOMMU, AMD-IOMMU]
4343 do not use PCIe ATS (and IOMMU device IOTLB).
4344 pcie_scan_all Scan all possible PCIe devices. Otherwise we
4345 only look for one device below a PCIe downstream
4347 big_root_window Try to add a big 64bit memory window to the PCIe
4348 root complex on AMD CPUs. Some GFX hardware
4349 can resize a BAR to allow access to all VRAM.
4350 Adding the window is slightly risky (it may
4351 conflict with unreported devices), so this
4353 disable_acs_redir=<pci_dev>[; ...]
4354 Specify one or more PCI devices (in the format
4355 specified above) separated by semicolons.
4356 Each device specified will have the PCI ACS
4357 redirect capabilities forced off which will
4358 allow P2P traffic between devices through
4359 bridges without forcing it upstream. Note:
4360 this removes isolation between devices and
4361 may put more devices in an IOMMU group.
4362 force_floating [S390] Force usage of floating interrupts.
4363 nomio [S390] Do not use MIO instructions.
4364 norid [S390] ignore the RID field and force use of
4365 one PCI domain per PCI function
4367 pcie_aspm= [PCIE] Forcibly enable or disable PCIe Active State Power
4370 force Enable ASPM even on devices that claim not to support it.
4371 WARNING: Forcing ASPM on may cause system lockups.
4373 pcie_ports= [PCIE] PCIe port services handling:
4374 native Use native PCIe services (PME, AER, DPC, PCIe hotplug)
4375 even if the platform doesn't give the OS permission to
4376 use them. This may cause conflicts if the platform
4377 also tries to use these services.
4378 dpc-native Use native PCIe service for DPC only. May
4379 cause conflicts if firmware uses AER or DPC.
4380 compat Disable native PCIe services (PME, AER, DPC, PCIe
4383 pcie_port_pm= [PCIE] PCIe port power management handling:
4384 off Disable power management of all PCIe ports
4385 force Forcibly enable power management of all PCIe ports
4387 pcie_pme= [PCIE,PM] Native PCIe PME signaling options:
4388 nomsi Do not use MSI for native PCIe PME signaling (this makes
4389 all PCIe root ports use INTx for all services).
4391 pcmv= [HW,PCMCIA] BadgePAD 4
4395 Keep all power-domains already enabled by bootloader on,
4396 even if no driver has claimed them. This is useful
4397 for debug and development, but should not be
4398 needed on a platform with proper driver support.
4400 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at
4403 See arch/parisc/kernel/pdc_chassis.c
4405 percpu_alloc= Select which percpu first chunk allocator to use.
4406 Currently supported values are "embed" and "page".
4407 Archs may support subset or none of the selections.
4408 See comments in mm/percpu.c for details on each
4409 allocator. This parameter is primarily for debugging
4410 and performance comparison.
4412 pirq= [SMP,APIC] Manual mp-table setup
4413 See Documentation/arch/x86/i386/IO-APIC.rst.
4415 plip= [PPT,NET] Parallel port network link
4416 Format: { parport<nr> | timid | 0 }
4417 See also Documentation/admin-guide/parport.rst.
4419 pmtmr= [X86] Manual setup of pmtmr I/O Port.
4420 Override pmtimer IOPort with a hex value.
4423 pmu_override= [PPC] Override the PMU.
4424 This option takes over the PMU facility, so it is no
4425 longer usable by perf. Setting this option starts the
4426 PMU counters by setting MMCR0 to 0 (the FC bit is
4427 cleared). If a number is given, then MMCR1 is set to
4428 that number, otherwise (e.g., 'pmu_override=on'), MMCR1
4431 pm_debug_messages [SUSPEND,KNL]
4432 Enable suspend/resume debug messages during boot up.
4435 Enable PNP debug messages (depends on the
4436 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time
4437 via /sys/module/pnp/parameters/debug. We always show
4438 current resource usage; turning this on also shows
4439 possible settings and some assignment information.
4445 { on | off | curr | res | no-curr | no-res }
4448 [ISAPNP] Exclude IRQs for the autoconfiguration
4451 [ISAPNP] Exclude DMAs for the autoconfiguration
4453 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration
4454 Ranges are in pairs (I/O port base and size).
4457 [ISAPNP] Exclude memory regions for the
4459 Ranges are in pairs (memory base and size).
4461 ports= [IP_VS_FTP] IPVS ftp helper module
4463 Up to 8 (IP_VS_APP_MAX_PORTS) ports
4465 Format: <port>,<port>....
4467 powersave=off [PPC] This option disables power saving features.
4468 It specifically disables cpuidle and sets the
4469 platform machine description specific power_save
4470 function to NULL. On Idle the CPU just reduces
4473 ppc_strict_facility_enable
4474 [PPC] This option catches any kernel floating point,
4475 Altivec, VSX and SPE outside of regions specifically
4476 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()).
4477 There is some performance impact when enabling this.
4481 Disable Hardware Transactional Memory
4484 Select preemption mode if you have CONFIG_PREEMPT_DYNAMIC
4485 none - Limited to cond_resched() calls
4486 voluntary - Limited to cond_resched() and might_sleep() calls
4487 full - Any section that isn't explicitly preempt disabled
4488 can be preempted anytime.
4490 print-fatal-signals=
4491 [KNL] debug: print fatal signals
4493 If enabled, warn about various signal handling
4494 related application anomalies: too many signals,
4495 too many POSIX.1 timers, fatal signals causing a
4498 If you hit the warning due to signal overflow,
4499 you might want to try "ulimit -i unlimited".
4503 printk.always_kmsg_dump=
4504 Trigger kmsg_dump for cases other than kernel oops or
4506 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
4509 printk.console_no_auto_verbose=
4510 Disable console loglevel raise on oops, panic
4511 or lockdep-detected issues (only if lock debug is on).
4512 With an exception to setups with low baudrate on
4513 serial console, keeping this 0 is a good choice
4514 in order to provide more debug information.
4516 default: 0 (auto_verbose is enabled)
4518 printk.devkmsg={on,off,ratelimit}
4519 Control writing to /dev/kmsg.
4520 on - unlimited logging to /dev/kmsg from userspace
4521 off - logging to /dev/kmsg disabled
4522 ratelimit - ratelimit the logging
4525 printk.time= Show timing data prefixed to each printk message line
4526 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
4528 processor.max_cstate= [HW,ACPI]
4529 Limit processor to maximum C-state
4530 max_cstate=9 overrides any DMI blacklist limit.
4532 processor.nocst [HW,ACPI]
4533 Ignore the _CST method to determine C-states,
4534 instead using the legacy FADT method
4536 profile= [KNL] Enable kernel profiling via /proc/profile
4537 Format: [<profiletype>,]<number>
4538 Param: <profiletype>: "schedule", "sleep", or "kvm"
4539 [defaults to kernel profiling]
4540 Param: "schedule" - profile schedule points.
4541 Param: "sleep" - profile D-state sleeping (millisecs).
4542 Requires CONFIG_SCHEDSTATS
4543 Param: "kvm" - profile VM exits.
4544 Param: <number> - step/bucket size as a power of 2 for
4545 statistical time based profiling.
4547 prompt_ramdisk= [RAM] [Deprecated]
4549 prot_virt= [S390] enable hosting protected virtual machines
4550 isolated from the hypervisor (if hardware supports
4554 psi= [KNL] Enable or disable pressure stall information
4558 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to
4559 probe for; one of (bare|imps|exps|lifebook|any).
4560 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports
4562 psmouse.resetafter= [HW,MOUSE]
4563 Try to reset the device after so many bad packets
4566 [HW,MOUSE] Set desired mouse resolution, in dpi.
4567 psmouse.smartscroll=
4568 [HW,MOUSE] Controls Logitech smartscroll autorepeat.
4569 0 = disabled, 1 = enabled (default).
4571 pstore.backend= Specify the name of the pstore backend to use
4573 pti= [X86-64] Control Page Table Isolation of user and
4574 kernel address spaces. Disabling this feature
4575 removes hardening, but improves performance of
4576 system calls and interrupts.
4578 on - unconditionally enable
4579 off - unconditionally disable
4580 auto - kernel detects whether your CPU model is
4581 vulnerable to issues that PTI mitigates
4583 Not specifying this option is equivalent to pti=auto.
4586 [KNL] Number of legacy pty's. Overwrites compiled-in
4589 quiet [KNL] Disable most log messages
4593 radix_hcall_invalidate=on [PPC/PSERIES]
4594 Disable RADIX GTSE feature and use hcall for TLB
4598 See Documentation/admin-guide/md.rst.
4600 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes
4601 See Documentation/admin-guide/blockdev/ramdisk.rst.
4603 ramdisk_start= [RAM] RAM disk image start address
4605 random.trust_cpu=off
4606 [KNL] Disable trusting the use of the CPU's
4607 random number generator (if available) to
4608 initialize the kernel's RNG.
4610 random.trust_bootloader=off
4611 [KNL] Disable trusting the use of the a seed
4612 passed by the bootloader (if available) to
4613 initialize the kernel's RNG.
4615 randomize_kstack_offset=
4616 [KNL] Enable or disable kernel stack offset
4617 randomization, which provides roughly 5 bits of
4618 entropy, frustrating memory corruption attacks
4619 that depend on stack address determinism or
4620 cross-syscall address exposures. This is only
4621 available on architectures that have defined
4622 CONFIG_HAVE_ARCH_RANDOMIZE_KSTACK_OFFSET.
4623 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
4624 Default is CONFIG_RANDOMIZE_KSTACK_OFFSET_DEFAULT.
4626 ras=option[,option,...] [KNL] RAS-specific options
4629 Disable the Correctable Errors Collector,
4630 see CONFIG_RAS_CEC help text.
4632 rcu_nocbs[=cpu-list]
4633 [KNL] The optional argument is a cpu list,
4636 In kernels built with CONFIG_RCU_NOCB_CPU=y,
4637 enable the no-callback CPU mode, which prevents
4638 such CPUs' callbacks from being invoked in
4639 softirq context. Invocation of such CPUs' RCU
4640 callbacks will instead be offloaded to "rcuox/N"
4641 kthreads created for that purpose, where "x" is
4642 "p" for RCU-preempt, "s" for RCU-sched, and "g"
4643 for the kthreads that mediate grace periods; and
4644 "N" is the CPU number. This reduces OS jitter on
4645 the offloaded CPUs, which can be useful for HPC
4646 and real-time workloads. It can also improve
4647 energy efficiency for asymmetric multiprocessors.
4649 If a cpulist is passed as an argument, the specified
4650 list of CPUs is set to no-callback mode from boot.
4652 Otherwise, if the '=' sign and the cpulist
4653 arguments are omitted, no CPU will be set to
4654 no-callback mode from boot but the mode may be
4655 toggled at runtime via cpusets.
4657 Note that this argument takes precedence over
4658 the CONFIG_RCU_NOCB_CPU_DEFAULT_ALL option.
4661 Rather than requiring that offloaded CPUs
4662 (specified by rcu_nocbs= above) explicitly
4663 awaken the corresponding "rcuoN" kthreads,
4664 make these kthreads poll for callbacks.
4665 This improves the real-time response for the
4666 offloaded CPUs by relieving them of the need to
4667 wake up the corresponding kthread, but degrades
4668 energy efficiency by requiring that the kthreads
4669 periodically wake up to do the polling.
4671 rcutree.blimit= [KNL]
4672 Set maximum number of finished RCU callbacks to
4673 process in one batch.
4675 rcutree.dump_tree= [KNL]
4676 Dump the structure of the rcu_node combining tree
4677 out at early boot. This is used for diagnostic
4678 purposes, to verify correct tree setup.
4680 rcutree.gp_cleanup_delay= [KNL]
4681 Set the number of jiffies to delay each step of
4682 RCU grace-period cleanup.
4684 rcutree.gp_init_delay= [KNL]
4685 Set the number of jiffies to delay each step of
4686 RCU grace-period initialization.
4688 rcutree.gp_preinit_delay= [KNL]
4689 Set the number of jiffies to delay each step of
4690 RCU grace-period pre-initialization, that is,
4691 the propagation of recent CPU-hotplug changes up
4692 the rcu_node combining tree.
4694 rcutree.use_softirq= [KNL]
4695 If set to zero, move all RCU_SOFTIRQ processing to
4696 per-CPU rcuc kthreads. Defaults to a non-zero
4697 value, meaning that RCU_SOFTIRQ is used by default.
4698 Specify rcutree.use_softirq=0 to use rcuc kthreads.
4700 But note that CONFIG_PREEMPT_RT=y kernels disable
4701 this kernel boot parameter, forcibly setting it
4704 rcutree.rcu_fanout_exact= [KNL]
4705 Disable autobalancing of the rcu_node combining
4706 tree. This is used by rcutorture, and might
4707 possibly be useful for architectures having high
4708 cache-to-cache transfer latencies.
4710 rcutree.rcu_fanout_leaf= [KNL]
4711 Change the number of CPUs assigned to each
4712 leaf rcu_node structure. Useful for very
4713 large systems, which will choose the value 64,
4714 and for NUMA systems with large remote-access
4715 latencies, which will choose a value aligned
4716 with the appropriate hardware boundaries.
4718 rcutree.rcu_min_cached_objs= [KNL]
4719 Minimum number of objects which are cached and
4720 maintained per one CPU. Object size is equal
4721 to PAGE_SIZE. The cache allows to reduce the
4722 pressure to page allocator, also it makes the
4723 whole algorithm to behave better in low memory
4726 rcutree.rcu_delay_page_cache_fill_msec= [KNL]
4727 Set the page-cache refill delay (in milliseconds)
4728 in response to low-memory conditions. The range
4729 of permitted values is in the range 0:100000.
4731 rcutree.jiffies_till_first_fqs= [KNL]
4732 Set delay from grace-period initialization to
4733 first attempt to force quiescent states.
4734 Units are jiffies, minimum value is zero,
4735 and maximum value is HZ.
4737 rcutree.jiffies_till_next_fqs= [KNL]
4738 Set delay between subsequent attempts to force
4739 quiescent states. Units are jiffies, minimum
4740 value is one, and maximum value is HZ.
4742 rcutree.jiffies_till_sched_qs= [KNL]
4743 Set required age in jiffies for a
4744 given grace period before RCU starts
4745 soliciting quiescent-state help from
4746 rcu_note_context_switch() and cond_resched().
4747 If not specified, the kernel will calculate
4748 a value based on the most recent settings
4749 of rcutree.jiffies_till_first_fqs
4750 and rcutree.jiffies_till_next_fqs.
4751 This calculated value may be viewed in
4752 rcutree.jiffies_to_sched_qs. Any attempt to set
4753 rcutree.jiffies_to_sched_qs will be cheerfully
4756 rcutree.kthread_prio= [KNL,BOOT]
4757 Set the SCHED_FIFO priority of the RCU per-CPU
4758 kthreads (rcuc/N). This value is also used for
4759 the priority of the RCU boost threads (rcub/N)
4760 and for the RCU grace-period kthreads (rcu_bh,
4761 rcu_preempt, and rcu_sched). If RCU_BOOST is
4762 set, valid values are 1-99 and the default is 1
4763 (the least-favored priority). Otherwise, when
4764 RCU_BOOST is not set, valid values are 0-99 and
4765 the default is zero (non-realtime operation).
4766 When RCU_NOCB_CPU is set, also adjust the
4767 priority of NOCB callback kthreads.
4769 rcutree.rcu_divisor= [KNL]
4770 Set the shift-right count to use to compute
4771 the callback-invocation batch limit bl from
4772 the number of callbacks queued on this CPU.
4773 The result will be bounded below by the value of
4774 the rcutree.blimit kernel parameter. Every bl
4775 callbacks, the softirq handler will exit in
4776 order to allow the CPU to do other work.
4778 Please note that this callback-invocation batch
4779 limit applies only to non-offloaded callback
4780 invocation. Offloaded callbacks are instead
4781 invoked in the context of an rcuoc kthread, which
4782 scheduler will preempt as it does any other task.
4784 rcutree.nocb_nobypass_lim_per_jiffy= [KNL]
4785 On callback-offloaded (rcu_nocbs) CPUs,
4786 RCU reduces the lock contention that would
4787 otherwise be caused by callback floods through
4788 use of the ->nocb_bypass list. However, in the
4789 common non-flooded case, RCU queues directly to
4790 the main ->cblist in order to avoid the extra
4791 overhead of the ->nocb_bypass list and its lock.
4792 But if there are too many callbacks queued during
4793 a single jiffy, RCU pre-queues the callbacks into
4794 the ->nocb_bypass queue. The definition of "too
4795 many" is supplied by this kernel boot parameter.
4797 rcutree.rcu_nocb_gp_stride= [KNL]
4798 Set the number of NOCB callback kthreads in
4799 each group, which defaults to the square root
4800 of the number of CPUs. Larger numbers reduce
4801 the wakeup overhead on the global grace-period
4802 kthread, but increases that same overhead on
4803 each group's NOCB grace-period kthread.
4805 rcutree.qhimark= [KNL]
4806 Set threshold of queued RCU callbacks beyond which
4807 batch limiting is disabled.
4809 rcutree.qlowmark= [KNL]
4810 Set threshold of queued RCU callbacks below which
4811 batch limiting is re-enabled.
4813 rcutree.qovld= [KNL]
4814 Set threshold of queued RCU callbacks beyond which
4815 RCU's force-quiescent-state scan will aggressively
4816 enlist help from cond_resched() and sched IPIs to
4817 help CPUs more quickly reach quiescent states.
4818 Set to less than zero to make this be set based
4819 on rcutree.qhimark at boot time and to zero to
4820 disable more aggressive help enlistment.
4822 rcutree.rcu_kick_kthreads= [KNL]
4823 Cause the grace-period kthread to get an extra
4824 wake_up() if it sleeps three times longer than
4825 it should at force-quiescent-state time.
4826 This wake_up() will be accompanied by a
4827 WARN_ONCE() splat and an ftrace_dump().
4829 rcutree.rcu_unlock_delay= [KNL]
4830 In CONFIG_RCU_STRICT_GRACE_PERIOD=y kernels,
4831 this specifies an rcu_read_unlock()-time delay
4832 in microseconds. This defaults to zero.
4833 Larger delays increase the probability of
4834 catching RCU pointer leaks, that is, buggy use
4835 of RCU-protected pointers after the relevant
4836 rcu_read_unlock() has completed.
4838 rcutree.sysrq_rcu= [KNL]
4839 Commandeer a sysrq key to dump out Tree RCU's
4840 rcu_node tree with an eye towards determining
4841 why a new grace period has not yet started.
4843 rcuscale.gp_async= [KNL]
4844 Measure performance of asynchronous
4845 grace-period primitives such as call_rcu().
4847 rcuscale.gp_async_max= [KNL]
4848 Specify the maximum number of outstanding
4849 callbacks per writer thread. When a writer
4850 thread exceeds this limit, it invokes the
4851 corresponding flavor of rcu_barrier() to allow
4852 previously posted callbacks to drain.
4854 rcuscale.gp_exp= [KNL]
4855 Measure performance of expedited synchronous
4856 grace-period primitives.
4858 rcuscale.holdoff= [KNL]
4859 Set test-start holdoff period. The purpose of
4860 this parameter is to delay the start of the
4861 test until boot completes in order to avoid
4864 rcuscale.kfree_rcu_test= [KNL]
4865 Set to measure performance of kfree_rcu() flooding.
4867 rcuscale.kfree_rcu_test_double= [KNL]
4868 Test the double-argument variant of kfree_rcu().
4869 If this parameter has the same value as
4870 rcuscale.kfree_rcu_test_single, both the single-
4871 and double-argument variants are tested.
4873 rcuscale.kfree_rcu_test_single= [KNL]
4874 Test the single-argument variant of kfree_rcu().
4875 If this parameter has the same value as
4876 rcuscale.kfree_rcu_test_double, both the single-
4877 and double-argument variants are tested.
4879 rcuscale.kfree_nthreads= [KNL]
4880 The number of threads running loops of kfree_rcu().
4882 rcuscale.kfree_alloc_num= [KNL]
4883 Number of allocations and frees done in an iteration.
4885 rcuscale.kfree_loops= [KNL]
4886 Number of loops doing rcuscale.kfree_alloc_num number
4887 of allocations and frees.
4889 rcuscale.nreaders= [KNL]
4890 Set number of RCU readers. The value -1 selects
4891 N, where N is the number of CPUs. A value
4892 "n" less than -1 selects N-n+1, where N is again
4893 the number of CPUs. For example, -2 selects N
4894 (the number of CPUs), -3 selects N+1, and so on.
4895 A value of "n" less than or equal to -N selects
4898 rcuscale.nwriters= [KNL]
4899 Set number of RCU writers. The values operate
4900 the same as for rcuscale.nreaders.
4901 N, where N is the number of CPUs
4903 rcuscale.perf_type= [KNL]
4904 Specify the RCU implementation to test.
4906 rcuscale.shutdown= [KNL]
4907 Shut the system down after performance tests
4908 complete. This is useful for hands-off automated
4911 rcuscale.verbose= [KNL]
4912 Enable additional printk() statements.
4914 rcuscale.writer_holdoff= [KNL]
4915 Write-side holdoff between grace periods,
4916 in microseconds. The default of zero says
4919 rcutorture.fqs_duration= [KNL]
4920 Set duration of force_quiescent_state bursts
4923 rcutorture.fqs_holdoff= [KNL]
4924 Set holdoff time within force_quiescent_state bursts
4927 rcutorture.fqs_stutter= [KNL]
4928 Set wait time between force_quiescent_state bursts
4931 rcutorture.fwd_progress= [KNL]
4932 Specifies the number of kthreads to be used
4933 for RCU grace-period forward-progress testing
4934 for the types of RCU supporting this notion.
4935 Defaults to 1 kthread, values less than zero or
4936 greater than the number of CPUs cause the number
4939 rcutorture.fwd_progress_div= [KNL]
4940 Specify the fraction of a CPU-stall-warning
4941 period to do tight-loop forward-progress testing.
4943 rcutorture.fwd_progress_holdoff= [KNL]
4944 Number of seconds to wait between successive
4945 forward-progress tests.
4947 rcutorture.fwd_progress_need_resched= [KNL]
4948 Enclose cond_resched() calls within checks for
4949 need_resched() during tight-loop forward-progress
4952 rcutorture.gp_cond= [KNL]
4953 Use conditional/asynchronous update-side
4954 primitives, if available.
4956 rcutorture.gp_exp= [KNL]
4957 Use expedited update-side primitives, if available.
4959 rcutorture.gp_normal= [KNL]
4960 Use normal (non-expedited) asynchronous
4961 update-side primitives, if available.
4963 rcutorture.gp_sync= [KNL]
4964 Use normal (non-expedited) synchronous
4965 update-side primitives, if available. If all
4966 of rcutorture.gp_cond=, rcutorture.gp_exp=,
4967 rcutorture.gp_normal=, and rcutorture.gp_sync=
4968 are zero, rcutorture acts as if is interpreted
4969 they are all non-zero.
4971 rcutorture.irqreader= [KNL]
4972 Run RCU readers from irq handlers, or, more
4973 accurately, from a timer handler. Not all RCU
4974 flavors take kindly to this sort of thing.
4976 rcutorture.leakpointer= [KNL]
4977 Leak an RCU-protected pointer out of the reader.
4978 This can of course result in splats, and is
4979 intended to test the ability of things like
4980 CONFIG_RCU_STRICT_GRACE_PERIOD=y to detect
4983 rcutorture.n_barrier_cbs= [KNL]
4984 Set callbacks/threads for rcu_barrier() testing.
4986 rcutorture.nfakewriters= [KNL]
4987 Set number of concurrent RCU writers. These just
4988 stress RCU, they don't participate in the actual
4989 test, hence the "fake".
4991 rcutorture.nocbs_nthreads= [KNL]
4992 Set number of RCU callback-offload togglers.
4993 Zero (the default) disables toggling.
4995 rcutorture.nocbs_toggle= [KNL]
4996 Set the delay in milliseconds between successive
4997 callback-offload toggling attempts.
4999 rcutorture.nreaders= [KNL]
5000 Set number of RCU readers. The value -1 selects
5001 N-1, where N is the number of CPUs. A value
5002 "n" less than -1 selects N-n-2, where N is again
5003 the number of CPUs. For example, -2 selects N
5004 (the number of CPUs), -3 selects N+1, and so on.
5006 rcutorture.object_debug= [KNL]
5007 Enable debug-object double-call_rcu() testing.
5009 rcutorture.onoff_holdoff= [KNL]
5010 Set time (s) after boot for CPU-hotplug testing.
5012 rcutorture.onoff_interval= [KNL]
5013 Set time (jiffies) between CPU-hotplug operations,
5014 or zero to disable CPU-hotplug testing.
5016 rcutorture.read_exit= [KNL]
5017 Set the number of read-then-exit kthreads used
5018 to test the interaction of RCU updaters and
5019 task-exit processing.
5021 rcutorture.read_exit_burst= [KNL]
5022 The number of times in a given read-then-exit
5023 episode that a set of read-then-exit kthreads
5026 rcutorture.read_exit_delay= [KNL]
5027 The delay, in seconds, between successive
5028 read-then-exit testing episodes.
5030 rcutorture.shuffle_interval= [KNL]
5031 Set task-shuffle interval (s). Shuffling tasks
5032 allows some CPUs to go into dyntick-idle mode
5033 during the rcutorture test.
5035 rcutorture.shutdown_secs= [KNL]
5036 Set time (s) after boot system shutdown. This
5037 is useful for hands-off automated testing.
5039 rcutorture.stall_cpu= [KNL]
5040 Duration of CPU stall (s) to test RCU CPU stall
5041 warnings, zero to disable.
5043 rcutorture.stall_cpu_block= [KNL]
5044 Sleep while stalling if set. This will result
5045 in warnings from preemptible RCU in addition
5046 to any other stall-related activity.
5048 rcutorture.stall_cpu_holdoff= [KNL]
5049 Time to wait (s) after boot before inducing stall.
5051 rcutorture.stall_cpu_irqsoff= [KNL]
5052 Disable interrupts while stalling if set.
5054 rcutorture.stall_gp_kthread= [KNL]
5055 Duration (s) of forced sleep within RCU
5056 grace-period kthread to test RCU CPU stall
5057 warnings, zero to disable. If both stall_cpu
5058 and stall_gp_kthread are specified, the
5059 kthread is starved first, then the CPU.
5061 rcutorture.stat_interval= [KNL]
5062 Time (s) between statistics printk()s.
5064 rcutorture.stutter= [KNL]
5065 Time (s) to stutter testing, for example, specifying
5066 five seconds causes the test to run for five seconds,
5067 wait for five seconds, and so on. This tests RCU's
5068 ability to transition abruptly to and from idle.
5070 rcutorture.test_boost= [KNL]
5071 Test RCU priority boosting? 0=no, 1=maybe, 2=yes.
5072 "Maybe" means test if the RCU implementation
5073 under test support RCU priority boosting.
5075 rcutorture.test_boost_duration= [KNL]
5076 Duration (s) of each individual boost test.
5078 rcutorture.test_boost_interval= [KNL]
5079 Interval (s) between each boost test.
5081 rcutorture.test_no_idle_hz= [KNL]
5082 Test RCU's dyntick-idle handling. See also the
5083 rcutorture.shuffle_interval parameter.
5085 rcutorture.torture_type= [KNL]
5086 Specify the RCU implementation to test.
5088 rcutorture.verbose= [KNL]
5089 Enable additional printk() statements.
5091 rcupdate.rcu_cpu_stall_ftrace_dump= [KNL]
5092 Dump ftrace buffer after reporting RCU CPU
5095 rcupdate.rcu_cpu_stall_suppress= [KNL]
5096 Suppress RCU CPU stall warning messages.
5098 rcupdate.rcu_cpu_stall_suppress_at_boot= [KNL]
5099 Suppress RCU CPU stall warning messages and
5100 rcutorture writer stall warnings that occur
5101 during early boot, that is, during the time
5102 before the init task is spawned.
5104 rcupdate.rcu_cpu_stall_timeout= [KNL]
5105 Set timeout for RCU CPU stall warning messages.
5106 The value is in seconds and the maximum allowed
5107 value is 300 seconds.
5109 rcupdate.rcu_exp_cpu_stall_timeout= [KNL]
5110 Set timeout for expedited RCU CPU stall warning
5111 messages. The value is in milliseconds
5112 and the maximum allowed value is 21000
5113 milliseconds. Please note that this value is
5114 adjusted to an arch timer tick resolution.
5115 Setting this to zero causes the value from
5116 rcupdate.rcu_cpu_stall_timeout to be used (after
5117 conversion from seconds to milliseconds).
5119 rcupdate.rcu_cpu_stall_cputime= [KNL]
5120 Provide statistics on the cputime and count of
5121 interrupts and tasks during the sampling period. For
5122 multiple continuous RCU stalls, all sampling periods
5123 begin at half of the first RCU stall timeout.
5125 rcupdate.rcu_exp_stall_task_details= [KNL]
5126 Print stack dumps of any tasks blocking the
5127 current expedited RCU grace period during an
5128 expedited RCU CPU stall warning.
5130 rcupdate.rcu_expedited= [KNL]
5131 Use expedited grace-period primitives, for
5132 example, synchronize_rcu_expedited() instead
5133 of synchronize_rcu(). This reduces latency,
5134 but can increase CPU utilization, degrade
5135 real-time latency, and degrade energy efficiency.
5136 No effect on CONFIG_TINY_RCU kernels.
5138 rcupdate.rcu_normal= [KNL]
5139 Use only normal grace-period primitives,
5140 for example, synchronize_rcu() instead of
5141 synchronize_rcu_expedited(). This improves
5142 real-time latency, CPU utilization, and
5143 energy efficiency, but can expose users to
5144 increased grace-period latency. This parameter
5145 overrides rcupdate.rcu_expedited. No effect on
5146 CONFIG_TINY_RCU kernels.
5148 rcupdate.rcu_normal_after_boot= [KNL]
5149 Once boot has completed (that is, after
5150 rcu_end_inkernel_boot() has been invoked), use
5151 only normal grace-period primitives. No effect
5152 on CONFIG_TINY_RCU kernels.
5154 But note that CONFIG_PREEMPT_RT=y kernels enables
5155 this kernel boot parameter, forcibly setting
5156 it to the value one, that is, converting any
5157 post-boot attempt at an expedited RCU grace
5158 period to instead use normal non-expedited
5159 grace-period processing.
5161 rcupdate.rcu_task_collapse_lim= [KNL]
5162 Set the maximum number of callbacks present
5163 at the beginning of a grace period that allows
5164 the RCU Tasks flavors to collapse back to using
5165 a single callback queue. This switching only
5166 occurs when rcupdate.rcu_task_enqueue_lim is
5167 set to the default value of -1.
5169 rcupdate.rcu_task_contend_lim= [KNL]
5170 Set the minimum number of callback-queuing-time
5171 lock-contention events per jiffy required to
5172 cause the RCU Tasks flavors to switch to per-CPU
5173 callback queuing. This switching only occurs
5174 when rcupdate.rcu_task_enqueue_lim is set to
5175 the default value of -1.
5177 rcupdate.rcu_task_enqueue_lim= [KNL]
5178 Set the number of callback queues to use for the
5179 RCU Tasks family of RCU flavors. The default
5180 of -1 allows this to be automatically (and
5181 dynamically) adjusted. This parameter is intended
5184 rcupdate.rcu_task_ipi_delay= [KNL]
5185 Set time in jiffies during which RCU tasks will
5186 avoid sending IPIs, starting with the beginning
5187 of a given grace period. Setting a large
5188 number avoids disturbing real-time workloads,
5189 but lengthens grace periods.
5191 rcupdate.rcu_task_stall_info= [KNL]
5192 Set initial timeout in jiffies for RCU task stall
5193 informational messages, which give some indication
5194 of the problem for those not patient enough to
5195 wait for ten minutes. Informational messages are
5196 only printed prior to the stall-warning message
5197 for a given grace period. Disable with a value
5198 less than or equal to zero. Defaults to ten
5199 seconds. A change in value does not take effect
5200 until the beginning of the next grace period.
5202 rcupdate.rcu_task_stall_info_mult= [KNL]
5203 Multiplier for time interval between successive
5204 RCU task stall informational messages for a given
5205 RCU tasks grace period. This value is clamped
5206 to one through ten, inclusive. It defaults to
5207 the value three, so that the first informational
5208 message is printed 10 seconds into the grace
5209 period, the second at 40 seconds, the third at
5210 160 seconds, and then the stall warning at 600
5211 seconds would prevent a fourth at 640 seconds.
5213 rcupdate.rcu_task_stall_timeout= [KNL]
5214 Set timeout in jiffies for RCU task stall
5215 warning messages. Disable with a value less
5216 than or equal to zero. Defaults to ten minutes.
5217 A change in value does not take effect until
5218 the beginning of the next grace period.
5220 rcupdate.rcu_self_test= [KNL]
5221 Run the RCU early boot self tests
5225 Run specified binary instead of /init from the ramdisk,
5226 used for early userspace startup. See initrd.
5229 force - Override the decision by the kernel to hide the
5230 advertisement of RDRAND support (this affects
5231 certain AMD processors because of buggy BIOS
5232 support, specifically around the suspend/resume
5236 Turn on/off individual RDT features. List is:
5237 cmt, mbmtotal, mbmlocal, l3cat, l3cdp, l2cat, l2cdp,
5239 E.g. to turn on cmt and turn off mba use:
5243 Format (x86 or x86_64):
5244 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] | d[efault] \
5246 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \
5248 Where reboot_mode is one of warm (soft) or cold (hard) or gpio
5249 (prefix with 'panic_' to set mode for panic
5251 reboot_type is one of bios, acpi, kbd, triple, efi, or pci,
5252 reboot_force is either force or not specified,
5253 reboot_cpu is s[mp]#### with #### being the processor
5254 to be used for rebooting.
5256 refscale.holdoff= [KNL]
5257 Set test-start holdoff period. The purpose of
5258 this parameter is to delay the start of the
5259 test until boot completes in order to avoid
5262 refscale.loops= [KNL]
5263 Set the number of loops over the synchronization
5264 primitive under test. Increasing this number
5265 reduces noise due to loop start/end overhead,
5266 but the default has already reduced the per-pass
5267 noise to a handful of picoseconds on ca. 2020
5270 refscale.nreaders= [KNL]
5271 Set number of readers. The default value of -1
5272 selects N, where N is roughly 75% of the number
5273 of CPUs. A value of zero is an interesting choice.
5275 refscale.nruns= [KNL]
5276 Set number of runs, each of which is dumped onto
5279 refscale.readdelay= [KNL]
5280 Set the read-side critical-section duration,
5281 measured in microseconds.
5283 refscale.scale_type= [KNL]
5284 Specify the read-protection implementation to test.
5286 refscale.shutdown= [KNL]
5287 Shut down the system at the end of the performance
5288 test. This defaults to 1 (shut it down) when
5289 refscale is built into the kernel and to 0 (leave
5290 it running) when refscale is built as a module.
5292 refscale.verbose= [KNL]
5293 Enable additional printk() statements.
5295 refscale.verbose_batched= [KNL]
5296 Batch the additional printk() statements. If zero
5297 (the default) or negative, print everything. Otherwise,
5298 print every Nth verbose statement, where N is the value
5302 [KNL, SMP] Set scheduler's default relax_domain_level.
5303 See Documentation/admin-guide/cgroup-v1/cpusets.rst.
5305 reserve= [KNL,BUGS] Force kernel to ignore I/O ports or memory
5306 Format: <base1>,<size1>[,<base2>,<size2>,...]
5307 Reserve I/O ports or memory so the kernel won't use
5308 them. If <base> is less than 0x10000, the region
5309 is assumed to be I/O ports; otherwise it is memory.
5311 reservetop= [X86-32]
5313 Reserves a hole at the top of the kernel virtual
5316 reset_devices [KNL] Force drivers to reset the underlying device
5317 during initialization.
5320 Specify the partition device for software suspend
5322 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>}
5324 resume_offset= [SWSUSP]
5325 Specify the offset from the beginning of the partition
5326 given by "resume=" at which the swap header is located,
5327 in <PAGE_SIZE> units (needed only for swap files).
5328 See Documentation/power/swsusp-and-swap-files.rst
5330 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to
5331 read the resume files
5333 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up.
5334 Useful for devices that are detected asynchronously
5335 (e.g. USB and MMC devices).
5337 retain_initrd [RAM] Keep initrd memory after extraction
5339 retbleed= [X86] Control mitigation of RETBleed (Arbitrary
5340 Speculative Code Execution with Return Instructions)
5343 AMD-based UNRET and IBPB mitigations alone do not stop
5344 sibling threads from influencing the predictions of other
5345 sibling threads. For that reason, STIBP is used on pro-
5346 cessors that support it, and mitigate SMT on processors
5350 auto - automatically select a migitation
5351 auto,nosmt - automatically select a mitigation,
5352 disabling SMT if necessary for
5353 the full mitigation (only on Zen1
5354 and older without STIBP).
5355 ibpb - On AMD, mitigate short speculation
5356 windows on basic block boundaries too.
5357 Safe, highest perf impact. It also
5358 enables STIBP if present. Not suitable
5360 ibpb,nosmt - Like "ibpb" above but will disable SMT
5361 when STIBP is not available. This is
5362 the alternative for systems which do not
5364 unret - Force enable untrained return thunks,
5365 only effective on AMD f15h-f17h based
5367 unret,nosmt - Like unret, but will disable SMT when STIBP
5368 is not available. This is the alternative for
5369 systems which do not have STIBP.
5371 Selecting 'auto' will choose a mitigation method at run
5372 time according to the CPU.
5374 Not specifying this option is equivalent to retbleed=auto.
5376 rfkill.default_state=
5377 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm,
5378 etc. communication is blocked by default.
5381 rfkill.master_switch_mode=
5382 0 The "airplane mode" button does nothing.
5383 1 The "airplane mode" button toggles between everything
5384 blocked and the previous configuration.
5385 2 The "airplane mode" button toggles between everything
5386 blocked and everything unblocked.
5388 rhash_entries= [KNL,NET]
5389 Set number of hash buckets for route cache
5392 [KNL] Disable ring 3 MONITOR/MWAIT feature on supported
5395 ro [KNL] Mount root device read-only on boot
5398 on Mark read-only kernel memory as read-only (default).
5399 off Leave read-only kernel memory writable for debugging.
5400 full Mark read-only kernel memory and aliases as read-only
5404 Enable the uart passthrough on the designated usb port
5405 on Rockchip SoCs. When active, the signals of the
5406 debug-uart get routed to the D+ and D- pins of the usb
5407 port and the regular usb controller gets disabled.
5409 root= [KNL] Root filesystem
5410 See name_to_dev_t comment in init/do_mounts.c.
5412 rootdelay= [KNL] Delay (in seconds) to pause before attempting to
5413 mount the root filesystem
5415 rootflags= [KNL] Set root filesystem mount option string
5417 rootfstype= [KNL] Set root filesystem type
5419 rootwait [KNL] Wait (indefinitely) for root device to show up.
5420 Useful for devices that are detected asynchronously
5421 (e.g. USB and MMC devices).
5423 rproc_mem=nn[KMG][@address]
5424 [KNL,ARM,CMA] Remoteproc physical memory block.
5425 Memory area to be used by remote processor image,
5428 rw [KNL] Mount root device read-write on boot
5430 S [KNL] Run init in single mode
5432 s390_iommu= [HW,S390]
5433 Set s390 IOTLB flushing mode
5435 With strict flushing every unmap operation will result in
5436 an IOTLB flush. Default is lazy flushing before reuse,
5439 s390_iommu_aperture= [KNL,S390]
5440 Specifies the size of the per device DMA address space
5441 accessible through the DMA and IOMMU APIs as a decimal
5442 factor of the size of main memory.
5443 The default is 1 meaning that one can concurrently use
5444 as many DMA addresses as physical memory is installed,
5445 if supported by hardware, and thus map all of memory
5446 once. With a value of 2 one can map all of memory twice
5447 and so on. As a special case a factor of 0 imposes no
5448 restrictions other than those given by hardware at the
5449 cost of significant additional memory use for tables.
5452 See drivers/net/irda/sa1100_ir.c.
5454 sched_verbose [KNL] Enables verbose scheduler debug messages.
5456 schedstats= [KNL,X86] Enable or disable scheduled statistics.
5457 Allowed values are enable and disable. This feature
5458 incurs a small amount of overhead in the scheduler
5459 but is useful for debugging and performance tuning.
5461 sched_thermal_decay_shift=
5462 [KNL, SMP] Set a decay shift for scheduler thermal
5463 pressure signal. Thermal pressure signal follows the
5464 default decay period of other scheduler pelt
5465 signals(usually 32 ms but configurable). Setting
5466 sched_thermal_decay_shift will left shift the decay
5467 period for the thermal pressure signal by the shift
5469 i.e. with the default pelt decay period of 32 ms
5470 sched_thermal_decay_shift thermal pressure decay pr
5474 Format: integer between 0 and 10
5477 scftorture.holdoff= [KNL]
5478 Number of seconds to hold off before starting
5479 test. Defaults to zero for module insertion and
5480 to 10 seconds for built-in smp_call_function()
5483 scftorture.longwait= [KNL]
5484 Request ridiculously long waits randomly selected
5485 up to the chosen limit in seconds. Zero (the
5486 default) disables this feature. Please note
5487 that requesting even small non-zero numbers of
5488 seconds can result in RCU CPU stall warnings,
5489 softlockup complaints, and so on.
5491 scftorture.nthreads= [KNL]
5492 Number of kthreads to spawn to invoke the
5493 smp_call_function() family of functions.
5494 The default of -1 specifies a number of kthreads
5495 equal to the number of CPUs.
5497 scftorture.onoff_holdoff= [KNL]
5498 Number seconds to wait after the start of the
5499 test before initiating CPU-hotplug operations.
5501 scftorture.onoff_interval= [KNL]
5502 Number seconds to wait between successive
5503 CPU-hotplug operations. Specifying zero (which
5504 is the default) disables CPU-hotplug operations.
5506 scftorture.shutdown_secs= [KNL]
5507 The number of seconds following the start of the
5508 test after which to shut down the system. The
5509 default of zero avoids shutting down the system.
5510 Non-zero values are useful for automated tests.
5512 scftorture.stat_interval= [KNL]
5513 The number of seconds between outputting the
5514 current test statistics to the console. A value
5515 of zero disables statistics output.
5517 scftorture.stutter_cpus= [KNL]
5518 The number of jiffies to wait between each change
5519 to the set of CPUs under test.
5521 scftorture.use_cpus_read_lock= [KNL]
5522 Use use_cpus_read_lock() instead of the default
5523 preempt_disable() to disable CPU hotplug
5524 while invoking one of the smp_call_function*()
5527 scftorture.verbose= [KNL]
5528 Enable additional printk() statements.
5530 scftorture.weight_single= [KNL]
5531 The probability weighting to use for the
5532 smp_call_function_single() function with a zero
5533 "wait" parameter. A value of -1 selects the
5534 default if all other weights are -1. However,
5535 if at least one weight has some other value, a
5536 value of -1 will instead select a weight of zero.
5538 scftorture.weight_single_wait= [KNL]
5539 The probability weighting to use for the
5540 smp_call_function_single() function with a
5541 non-zero "wait" parameter. See weight_single.
5543 scftorture.weight_many= [KNL]
5544 The probability weighting to use for the
5545 smp_call_function_many() function with a zero
5546 "wait" parameter. See weight_single.
5547 Note well that setting a high probability for
5548 this weighting can place serious IPI load
5551 scftorture.weight_many_wait= [KNL]
5552 The probability weighting to use for the
5553 smp_call_function_many() function with a
5554 non-zero "wait" parameter. See weight_single
5557 scftorture.weight_all= [KNL]
5558 The probability weighting to use for the
5559 smp_call_function_all() function with a zero
5560 "wait" parameter. See weight_single and
5563 scftorture.weight_all_wait= [KNL]
5564 The probability weighting to use for the
5565 smp_call_function_all() function with a
5566 non-zero "wait" parameter. See weight_single
5569 skew_tick= [KNL] Offset the periodic timer tick per cpu to mitigate
5570 xtime_lock contention on larger systems, and/or RCU lock
5571 contention on all systems with CONFIG_MAXSMP set.
5572 Format: { "0" | "1" }
5573 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1"
5575 Note: increases power consumption, thus should only be
5576 enabled if running jitter sensitive (HPC/RT) workloads.
5578 security= [SECURITY] Choose a legacy "major" security module to
5579 enable at boot. This has been deprecated by the
5582 selinux= [SELINUX] Disable or enable SELinux at boot time.
5583 Format: { "0" | "1" }
5584 See security/selinux/Kconfig help text.
5589 serialnumber [BUGS=X86-32]
5591 sev=option[,option...] [X86-64] See Documentation/arch/x86/x86_64/boot-options.rst
5594 Maximal number of shapers.
5596 show_lapic= [APIC,X86] Advanced Programmable Interrupt Controller
5597 Limit apic dumping. The parameter defines the maximal
5598 number of local apics being dumped. Also it is possible
5599 to set it to "all" by meaning -- no limit here.
5600 Format: { 1 (default) | 2 | ... | all }.
5601 The parameter valid if only apic=debug or
5602 apic=verbose is specified.
5603 Example: apic=debug show_lapic=all
5611 Enable merging of slabs with similar size when the
5612 kernel is built without CONFIG_SLAB_MERGE_DEFAULT.
5615 Disable merging of slabs with similar size. May be
5616 necessary if there is some reason to distinguish
5617 allocs to different slabs, especially in hardened
5618 environments where the risk of heap overflows and
5619 layout control by attackers can usually be
5620 frustrated by disabling merging. This will reduce
5621 most of the exposure of a heap attack to a single
5622 cache (risks via metadata attacks are mostly
5623 unchanged). Debug options disable merging on their
5625 For more information see Documentation/mm/slub.rst.
5627 slab_max_order= [MM, SLAB]
5628 Determines the maximum allowed order for slabs.
5629 A high setting may cause OOMs due to memory
5630 fragmentation. Defaults to 1 for systems with
5631 more than 32MB of RAM, 0 otherwise.
5633 slub_debug[=options[,slabs][;[options[,slabs]]...] [MM, SLUB]
5634 Enabling slub_debug allows one to determine the
5635 culprit if slab objects become corrupted. Enabling
5636 slub_debug can create guard zones around objects and
5637 may poison objects when not in use. Also tracks the
5638 last alloc / free. For more information see
5639 Documentation/mm/slub.rst.
5641 slub_max_order= [MM, SLUB]
5642 Determines the maximum allowed order for slabs.
5643 A high setting may cause OOMs due to memory
5644 fragmentation. For more information see
5645 Documentation/mm/slub.rst.
5647 slub_min_objects= [MM, SLUB]
5648 The minimum number of objects per slab. SLUB will
5649 increase the slab order up to slub_max_order to
5650 generate a sufficiently large slab able to contain
5651 the number of objects indicated. The higher the number
5652 of objects the smaller the overhead of tracking slabs
5653 and the less frequently locks need to be acquired.
5654 For more information see Documentation/mm/slub.rst.
5656 slub_min_order= [MM, SLUB]
5657 Determines the minimum page order for slabs. Must be
5658 lower than slub_max_order.
5659 For more information see Documentation/mm/slub.rst.
5661 slub_merge [MM, SLUB]
5662 Same with slab_merge.
5664 slub_nomerge [MM, SLUB]
5665 Same with slab_nomerge. This is supported for legacy.
5666 See slab_nomerge for more information.
5669 Format: <io1>[,<io2>[,...,<io8>]]
5671 smp.csd_lock_timeout= [KNL]
5672 Specify the period of time in milliseconds
5673 that smp_call_function() and friends will wait
5674 for a CPU to release the CSD lock. This is
5675 useful when diagnosing bugs involving CPUs
5676 disabling interrupts for extended periods
5677 of time. Defaults to 5,000 milliseconds, and
5678 setting a value of zero disables this feature.
5679 This feature may be more efficiently disabled
5680 using the csdlock_debug- kernel parameter.
5682 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices
5683 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port
5684 smsc-ircc2.ircc_sir= [HW] SIR base I/O port
5685 smsc-ircc2.ircc_fir= [HW] FIR base I/O port
5686 smsc-ircc2.ircc_irq= [HW] IRQ line
5687 smsc-ircc2.ircc_dma= [HW] DMA channel
5688 smsc-ircc2.ircc_transceiver= [HW] Transceiver type:
5689 0: Toshiba Satellite 1800 (GP data pin select)
5690 1: Fast pin select (default)
5693 smt= [KNL,S390] Set the maximum number of threads (logical
5694 CPUs) to use per physical CPU on systems capable of
5695 symmetric multithreading (SMT). Will be capped to the
5696 actual hardware limit.
5698 Default: -1 (no limit)
5701 [KNL] Should the soft-lockup detector generate panics.
5704 A value of 1 instructs the soft-lockup detector
5705 to panic the machine when a soft-lockup occurs. It is
5706 also controlled by the kernel.softlockup_panic sysctl
5707 and CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC, which is the
5708 respective build-time switch to that functionality.
5710 softlockup_all_cpu_backtrace=
5711 [KNL] Should the soft-lockup detector generate
5712 backtraces on all cpus.
5715 sonypi.*= [HW] Sony Programmable I/O Control Device driver
5716 See Documentation/admin-guide/laptops/sonypi.rst
5718 spectre_v2= [X86] Control mitigation of Spectre variant 2
5719 (indirect branch speculation) vulnerability.
5720 The default operation protects the kernel from
5723 on - unconditionally enable, implies
5725 off - unconditionally disable, implies
5727 auto - kernel detects whether your CPU model is
5730 Selecting 'on' will, and 'auto' may, choose a
5731 mitigation method at run time according to the
5732 CPU, the available microcode, the setting of the
5733 CONFIG_RETPOLINE configuration option, and the
5734 compiler with which the kernel was built.
5736 Selecting 'on' will also enable the mitigation
5737 against user space to user space task attacks.
5739 Selecting 'off' will disable both the kernel and
5740 the user space protections.
5742 Specific mitigations can also be selected manually:
5744 retpoline - replace indirect branches
5745 retpoline,generic - Retpolines
5746 retpoline,lfence - LFENCE; indirect branch
5747 retpoline,amd - alias for retpoline,lfence
5748 eibrs - Enhanced/Auto IBRS
5749 eibrs,retpoline - Enhanced/Auto IBRS + Retpolines
5750 eibrs,lfence - Enhanced/Auto IBRS + LFENCE
5751 ibrs - use IBRS to protect kernel
5753 Not specifying this option is equivalent to
5757 [X86] Control mitigation of Spectre variant 2
5758 (indirect branch speculation) vulnerability between
5761 on - Unconditionally enable mitigations. Is
5762 enforced by spectre_v2=on
5764 off - Unconditionally disable mitigations. Is
5765 enforced by spectre_v2=off
5767 prctl - Indirect branch speculation is enabled,
5768 but mitigation can be enabled via prctl
5769 per thread. The mitigation control state
5770 is inherited on fork.
5773 - Like "prctl" above, but only STIBP is
5774 controlled per thread. IBPB is issued
5775 always when switching between different user
5779 - Same as "prctl" above, but all seccomp
5780 threads will enable the mitigation unless
5781 they explicitly opt out.
5784 - Like "seccomp" above, but only STIBP is
5785 controlled per thread. IBPB is issued
5786 always when switching between different
5787 user space processes.
5789 auto - Kernel selects the mitigation depending on
5790 the available CPU features and vulnerability.
5792 Default mitigation: "prctl"
5794 Not specifying this option is equivalent to
5795 spectre_v2_user=auto.
5797 spec_store_bypass_disable=
5798 [HW] Control Speculative Store Bypass (SSB) Disable mitigation
5799 (Speculative Store Bypass vulnerability)
5801 Certain CPUs are vulnerable to an exploit against a
5802 a common industry wide performance optimization known
5803 as "Speculative Store Bypass" in which recent stores
5804 to the same memory location may not be observed by
5805 later loads during speculative execution. The idea
5806 is that such stores are unlikely and that they can
5807 be detected prior to instruction retirement at the
5808 end of a particular speculation execution window.
5810 In vulnerable processors, the speculatively forwarded
5811 store can be used in a cache side channel attack, for
5812 example to read memory to which the attacker does not
5813 directly have access (e.g. inside sandboxed code).
5815 This parameter controls whether the Speculative Store
5816 Bypass optimization is used.
5818 On x86 the options are:
5820 on - Unconditionally disable Speculative Store Bypass
5821 off - Unconditionally enable Speculative Store Bypass
5822 auto - Kernel detects whether the CPU model contains an
5823 implementation of Speculative Store Bypass and
5824 picks the most appropriate mitigation. If the
5825 CPU is not vulnerable, "off" is selected. If the
5826 CPU is vulnerable the default mitigation is
5827 architecture and Kconfig dependent. See below.
5828 prctl - Control Speculative Store Bypass per thread
5829 via prctl. Speculative Store Bypass is enabled
5830 for a process by default. The state of the control
5831 is inherited on fork.
5832 seccomp - Same as "prctl" above, but all seccomp threads
5833 will disable SSB unless they explicitly opt out.
5835 Default mitigations:
5838 On powerpc the options are:
5840 on,auto - On Power8 and Power9 insert a store-forwarding
5841 barrier on kernel entry and exit. On Power7
5842 perform a software flush on kernel entry and
5846 Not specifying this option is equivalent to
5847 spec_store_bypass_disable=auto.
5849 spia_io_base= [HW,MTD]
5855 [X86] Enable split lock detection or bus lock detection
5857 When enabled (and if hardware support is present), atomic
5858 instructions that access data across cache line
5859 boundaries will result in an alignment check exception
5860 for split lock detection or a debug exception for
5865 warn - the kernel will emit rate-limited warnings
5866 about applications triggering the #AC
5867 exception or the #DB exception. This mode is
5868 the default on CPUs that support split lock
5869 detection or bus lock detection. Default
5870 behavior is by #AC if both features are
5871 enabled in hardware.
5873 fatal - the kernel will send SIGBUS to applications
5874 that trigger the #AC exception or the #DB
5875 exception. Default behavior is by #AC if
5876 both features are enabled in hardware.
5879 Set system wide rate limit to N bus locks
5880 per second for bus lock detection.
5883 N/A for split lock detection.
5886 If an #AC exception is hit in the kernel or in
5887 firmware (i.e. not while executing in user mode)
5888 the kernel will oops in either "warn" or "fatal"
5891 #DB exception for bus lock is triggered only when
5895 Control the Special Register Buffer Data Sampling
5898 Certain CPUs are vulnerable to an MDS-like
5899 exploit which can leak bits from the random
5902 By default, this issue is mitigated by
5903 microcode. However, the microcode fix can cause
5904 the RDRAND and RDSEED instructions to become
5905 much slower. Among other effects, this will
5906 result in reduced throughput from /dev/urandom.
5908 The microcode mitigation can be disabled with
5909 the following option:
5911 off: Disable mitigation and remove
5912 performance impact to RDRAND and RDSEED
5914 srcutree.big_cpu_lim [KNL]
5915 Specifies the number of CPUs constituting a
5916 large system, such that srcu_struct structures
5917 should immediately allocate an srcu_node array.
5918 This kernel-boot parameter defaults to 128,
5919 but takes effect only when the low-order four
5920 bits of srcutree.convert_to_big is equal to 3
5923 srcutree.convert_to_big [KNL]
5924 Specifies under what conditions an SRCU tree
5925 srcu_struct structure will be converted to big
5926 form, that is, with an rcu_node tree:
5929 1: At init_srcu_struct() time.
5930 2: When rcutorture decides to.
5931 3: Decide at boot time (default).
5932 0x1X: Above plus if high contention.
5934 Either way, the srcu_node tree will be sized based
5935 on the actual runtime number of CPUs (nr_cpu_ids)
5936 instead of the compile-time CONFIG_NR_CPUS.
5938 srcutree.counter_wrap_check [KNL]
5939 Specifies how frequently to check for
5940 grace-period sequence counter wrap for the
5941 srcu_data structure's ->srcu_gp_seq_needed field.
5942 The greater the number of bits set in this kernel
5943 parameter, the less frequently counter wrap will
5944 be checked for. Note that the bottom two bits
5947 srcutree.exp_holdoff [KNL]
5948 Specifies how many nanoseconds must elapse
5949 since the end of the last SRCU grace period for
5950 a given srcu_struct until the next normal SRCU
5951 grace period will be considered for automatic
5952 expediting. Set to zero to disable automatic
5955 srcutree.srcu_max_nodelay [KNL]
5956 Specifies the number of no-delay instances
5957 per jiffy for which the SRCU grace period
5958 worker thread will be rescheduled with zero
5959 delay. Beyond this limit, worker thread will
5960 be rescheduled with a sleep delay of one jiffy.
5962 srcutree.srcu_max_nodelay_phase [KNL]
5963 Specifies the per-grace-period phase, number of
5964 non-sleeping polls of readers. Beyond this limit,
5965 grace period worker thread will be rescheduled
5966 with a sleep delay of one jiffy, between each
5967 rescan of the readers, for a grace period phase.
5969 srcutree.srcu_retry_check_delay [KNL]
5970 Specifies number of microseconds of non-sleeping
5971 delay between each non-sleeping poll of readers.
5973 srcutree.small_contention_lim [KNL]
5974 Specifies the number of update-side contention
5975 events per jiffy will be tolerated before
5976 initiating a conversion of an srcu_struct
5977 structure to big form. Note that the value of
5978 srcutree.convert_to_big must have the 0x10 bit
5979 set for contention-based conversions to occur.
5982 Speculative Store Bypass Disable control
5984 On CPUs that are vulnerable to the Speculative
5985 Store Bypass vulnerability and offer a
5986 firmware based mitigation, this parameter
5987 indicates how the mitigation should be used:
5989 force-on: Unconditionally enable mitigation for
5990 for both kernel and userspace
5991 force-off: Unconditionally disable mitigation for
5992 for both kernel and userspace
5993 kernel: Always enable mitigation in the
5994 kernel, and offer a prctl interface
5995 to allow userspace to register its
5996 interest in being mitigated too.
5998 stack_guard_gap= [MM]
5999 override the default stack gap protection. The value
6000 is in page units and it defines how many pages prior
6001 to (for stacks growing down) resp. after (for stacks
6002 growing up) the main stack are reserved for no other
6003 mapping. Default value is 256 pages.
6005 stack_depot_disable= [KNL]
6006 Setting this to true through kernel command line will
6007 disable the stack depot thereby saving the static memory
6008 consumed by the stack hash table. By default this is set
6012 Enabled the stack tracer on boot up.
6014 stacktrace_filter=[function-list]
6015 [FTRACE] Limit the functions that the stack tracer
6016 will trace at boot up. function-list is a comma-separated
6017 list of functions. This list can be changed at run
6018 time by the stack_trace_filter file in the debugfs
6019 tracing directory. Note, this enables stack tracing
6020 and the stacktrace above is not needed.
6024 Set the STI (builtin display/keyboard on the HP-PARISC
6025 machines) console (graphic card) which should be used
6026 as the initial boot-console.
6027 See also comment in drivers/video/console/sticore.c.
6030 See comment in drivers/video/console/sticore.c.
6033 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]]
6038 Enable or disable strict sigaltstack size checks
6039 against the required signal frame size which
6040 depends on the supported FPU features. This can
6041 be used to filter out binaries which have
6042 not yet been made aware of AT_MINSIGSTKSZ.
6045 Limits the number of kernel HPT entries in the hash
6046 page table to increase the rate of hash page table
6047 faults on kernel addresses.
6050 Limits the number of kernel SLB entries, and flushes
6051 them frequently to increase the rate of SLB faults
6052 on kernel addresses.
6054 sunrpc.min_resvport=
6055 sunrpc.max_resvport=
6057 SunRPC servers often require that client requests
6058 originate from a privileged port (i.e. a port in the
6059 range 0 < portnr < 1024).
6060 An administrator who wishes to reserve some of these
6061 ports for other uses may adjust the range that the
6062 kernel's sunrpc client considers to be privileged
6063 using these two parameters to set the minimum and
6064 maximum port values.
6066 sunrpc.svc_rpc_per_connection_limit=
6068 Limit the number of requests that the server will
6069 process in parallel from a single connection.
6070 The default value is 0 (no limit).
6074 Control how the NFS server code allocates CPUs to
6075 service thread pools. Depending on how many NICs
6076 you have and where their interrupts are bound, this
6077 option will affect which CPUs will do NFS serving.
6078 Note: this parameter cannot be changed while the
6079 NFS server is running.
6081 auto the server chooses an appropriate mode
6082 automatically using heuristics
6083 global a single global pool contains all CPUs
6084 percpu one pool for each CPU
6085 pernode one pool for each NUMA node (equivalent
6086 to global on non-NUMA machines)
6088 sunrpc.tcp_slot_table_entries=
6089 sunrpc.udp_slot_table_entries=
6091 Sets the upper limit on the number of simultaneous
6092 RPC calls that can be sent from the client to a
6093 server. Increasing these values may allow you to
6094 improve throughput, but will also increase the
6095 amount of memory reserved for use by the client.
6097 suspend.pm_test_delay=
6099 Sets the number of seconds to remain in a suspend test
6100 mode before resuming the system (see
6101 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG
6102 is set. Default value is 5.
6105 Format: { on | off | y | n | 1 | 0 }
6106 This parameter controls use of the Protected
6107 Execution Facility on pSeries.
6109 swiotlb= [ARM,IA-64,PPC,MIPS,X86]
6110 Format: { <int> [,<int>] | force | noforce }
6111 <int> -- Number of I/O TLB slabs
6112 <int> -- Second integer after comma. Number of swiotlb
6113 areas with their own lock. Will be rounded up
6115 force -- force using of bounce buffers even if they
6116 wouldn't be automatically used by the kernel
6117 noforce -- Never use bounce buffers (for debugging)
6122 Set a sysctl parameter, right before loading the init
6123 process, as if the value was written to the respective
6124 /proc/sys/... file. Both '.' and '/' are recognized as
6125 separators. Unrecognized parameters and invalid values
6126 are reported in the kernel log. Sysctls registered
6127 later by a loaded module cannot be set this way.
6128 Example: sysctl.vm.swappiness=40
6130 sysfs.deprecated=0|1 [KNL]
6131 Enable/disable old style sysfs layout for old udev
6132 on older distributions. When this option is enabled
6133 very new udev will not work anymore. When this option
6134 is disabled (or CONFIG_SYSFS_DEPRECATED not compiled)
6135 in older udev will not work anymore.
6136 Default depends on CONFIG_SYSFS_DEPRECATED_V2 set in
6137 the kernel configuration.
6139 sysrq_always_enabled
6141 Ignore sysrq setting - this boot parameter will
6142 neutralize any effect of /proc/sys/kernel/sysrq.
6143 Useful for debugging.
6145 tcpmhash_entries= [KNL,NET]
6146 Set the number of tcp_metrics_hash slots.
6147 Default value is 8192 or 16384 depending on total
6148 ram pages. This is used to specify the TCP metrics
6149 cache size. See Documentation/networking/ip-sysctl.rst
6150 "tcp_no_metrics_save" section for more details.
6154 test_suspend= [SUSPEND]
6155 Format: { "mem" | "standby" | "freeze" }[,N]
6156 Specify "mem" (for Suspend-to-RAM) or "standby" (for
6157 standby suspend) or "freeze" (for suspend type freeze)
6158 as the system sleep state during system startup with
6159 the optional capability to repeat N number of times.
6160 The system is woken from this state using a
6161 wakeup-capable RTC alarm.
6163 thash_entries= [KNL,NET]
6164 Set number of hash buckets for TCP connection
6166 thermal.act= [HW,ACPI]
6167 -1: disable all active trip points in all thermal zones
6168 <degrees C>: override all lowest active trip points
6170 thermal.crt= [HW,ACPI]
6171 -1: disable all critical trip points in all thermal zones
6172 <degrees C>: override all critical trip points
6174 thermal.nocrt= [HW,ACPI]
6175 Set to disable actions on ACPI thermal zone
6176 critical and hot trip points.
6178 thermal.off= [HW,ACPI]
6179 1: disable ACPI thermal control
6181 thermal.psv= [HW,ACPI]
6182 -1: disable all passive trip points
6183 <degrees C>: override all passive trip points to this
6186 thermal.tzp= [HW,ACPI]
6187 Specify global default ACPI thermal zone polling rate
6188 <deci-seconds>: poll all this frequency
6189 0: no polling (default)
6192 Force threading of all interrupt handlers except those
6193 marked explicitly IRQF_NO_THREAD.
6197 Specify if the kernel should make use of the cpu
6198 topology information if the hardware supports this.
6199 The scheduler will make use of this information and
6200 e.g. base its process migration decisions on it.
6203 topology_updates= [KNL, PPC, NUMA]
6205 Specify if the kernel should ignore (off)
6206 topology updates sent by the hypervisor to this
6209 torture.disable_onoff_at_boot= [KNL]
6210 Prevent the CPU-hotplug component of torturing
6211 until after init has spawned.
6213 torture.ftrace_dump_at_shutdown= [KNL]
6214 Dump the ftrace buffer at torture-test shutdown,
6215 even if there were no errors. This can be a
6216 very costly operation when many torture tests
6217 are running concurrently, especially on systems
6218 with rotating-rust storage.
6220 torture.verbose_sleep_frequency= [KNL]
6221 Specifies how many verbose printk()s should be
6222 emitted between each sleep. The default of zero
6223 disables verbose-printk() sleeping.
6225 torture.verbose_sleep_duration= [KNL]
6226 Duration of each verbose-printk() sleep in jiffies.
6230 tpm_suspend_pcr=[HW,TPM]
6231 Format: integer pcr id
6232 Specify that at suspend time, the tpm driver
6233 should extend the specified pcr with zeros,
6234 as a workaround for some chips which fail to
6235 flush the last written pcr on TPM_SaveState.
6236 This will guarantee that all the other pcrs
6240 Have the tracepoints sent to printk as well as the
6241 tracing ring buffer. This is useful for early boot up
6242 where the system hangs or reboots and does not give the
6243 option for reading the tracing buffer or performing a
6244 ftrace_dump_on_oops.
6246 To turn off having tracepoints sent to printk,
6247 echo 0 > /proc/sys/kernel/tracepoint_printk
6248 Note, echoing 1 into this file without the
6249 tracepoint_printk kernel cmdline option has no effect.
6251 The tp_printk_stop_on_boot (see below) can also be used
6252 to stop the printing of events to console at
6257 Having tracepoints sent to printk() and activating high
6258 frequency tracepoints such as irq or sched, can cause
6259 the system to live lock.
6261 tp_printk_stop_on_boot [FTRACE]
6262 When tp_printk (above) is set, it can cause a lot of noise
6263 on the console. It may be useful to only include the
6264 printing of events during boot up, as user space may
6265 make the system inoperable.
6267 This command line option will stop the printing of events
6268 to console at the late_initcall_sync() time frame.
6270 trace_buf_size=nn[KMG]
6271 [FTRACE] will set tracing buffer size on each cpu.
6273 trace_clock= [FTRACE] Set the clock used for tracing events
6275 local - Use the per CPU time stamp counter
6276 (converted into nanoseconds). Fast, but
6277 depending on the architecture, may not be
6278 in sync between CPUs.
6279 global - Event time stamps are synchronize across
6280 CPUs. May be slower than the local clock,
6281 but better for some race conditions.
6282 counter - Simple counting of events (1, 2, ..)
6283 note, some counts may be skipped due to the
6284 infrastructure grabbing the clock more than
6286 uptime - Use jiffies as the time stamp.
6287 perf - Use the same clock that perf uses.
6288 mono - Use ktime_get_mono_fast_ns() for time stamps.
6289 mono_raw - Use ktime_get_raw_fast_ns() for time
6291 boot - Use ktime_get_boot_fast_ns() for time stamps.
6292 Architectures may add more clocks. See
6293 Documentation/trace/ftrace.rst for more details.
6295 trace_event=[event-list]
6296 [FTRACE] Set and start specified trace events in order
6297 to facilitate early boot debugging. The event-list is a
6298 comma-separated list of trace events to enable. See
6299 also Documentation/trace/events.rst
6301 trace_instance=[instance-info]
6302 [FTRACE] Create a ring buffer instance early in boot up.
6303 This will be listed in:
6305 /sys/kernel/tracing/instances
6307 Events can be enabled at the time the instance is created
6310 trace_instance=<name>,<system1>:<event1>,<system2>:<event2>
6312 Note, the "<system*>:" portion is optional if the event is
6315 trace_instance=foo,sched:sched_switch,irq_handler_entry,initcall
6317 will enable the "sched_switch" event (note, the "sched:" is optional, and
6318 the same thing would happen if it was left off). The irq_handler_entry
6319 event, and all events under the "initcall" system.
6321 trace_options=[option-list]
6322 [FTRACE] Enable or disable tracer options at boot.
6323 The option-list is a comma delimited list of options
6324 that can be enabled or disabled just as if you were
6325 to echo the option name into
6327 /sys/kernel/tracing/trace_options
6329 For example, to enable stacktrace option (to dump the
6330 stack trace of each event), add to the command line:
6332 trace_options=stacktrace
6334 See also Documentation/trace/ftrace.rst "trace options"
6337 trace_trigger=[trigger-list]
6338 [FTRACE] Add a event trigger on specific events.
6339 Set a trigger on top of a specific event, with an optional
6342 The format is is "trace_trigger=<event>.<trigger>[ if <filter>],..."
6343 Where more than one trigger may be specified that are comma deliminated.
6347 trace_trigger="sched_switch.stacktrace if prev_state == 2"
6349 The above will enable the "stacktrace" trigger on the "sched_switch"
6350 event but only trigger it if the "prev_state" of the "sched_switch"
6351 event is "2" (TASK_UNINTERUPTIBLE).
6353 See also "Event triggers" in Documentation/trace/events.rst
6357 [FTRACE] enable this option to disable tracing when a
6358 warning is hit. This turns off "tracing_on". Tracing can
6359 be enabled again by echoing '1' into the "tracing_on"
6360 file located in /sys/kernel/tracing/
6362 This option is useful, as it disables the trace before
6363 the WARNING dump is called, which prevents the trace to
6364 be filled with content caused by the warning output.
6366 This option can also be set at run time via the sysctl
6367 option: kernel/traceoff_on_warning
6369 transparent_hugepage=
6371 Format: [always|madvise|never]
6372 Can be used to control the default behavior of the system
6373 with respect to transparent hugepages.
6374 See Documentation/admin-guide/mm/transhuge.rst
6377 trusted.source= [KEYS]
6379 This parameter identifies the trust source as a backend
6380 for trusted keys implementation. Supported trust
6385 If not specified then it defaults to iterating through
6386 the trust source list starting with TPM and assigns the
6387 first trust source as a backend which is initialized
6388 successfully during iteration.
6392 The RNG used to generate key material for trusted keys.
6395 - the same value as trusted.source: "tpm" or "tee"
6397 If not specified, "default" is used. In this case,
6398 the RNG's choice is left to each individual trust source.
6400 tsc= Disable clocksource stability checks for TSC.
6402 [x86] reliable: mark tsc clocksource as reliable, this
6403 disables clocksource verification at runtime, as well
6404 as the stability checks done at bootup. Used to enable
6405 high-resolution timer mode on older hardware, and in
6406 virtualized environment.
6407 [x86] noirqtime: Do not use TSC to do irq accounting.
6408 Used to run time disable IRQ_TIME_ACCOUNTING on any
6409 platforms where RDTSC is slow and this accounting
6411 [x86] unstable: mark the TSC clocksource as unstable, this
6412 marks the TSC unconditionally unstable at bootup and
6413 avoids any further wobbles once the TSC watchdog notices.
6414 [x86] nowatchdog: disable clocksource watchdog. Used
6415 in situations with strict latency requirements (where
6416 interruptions from clocksource watchdog are not
6418 [x86] recalibrate: force recalibration against a HW timer
6419 (HPET or PM timer) on systems whose TSC frequency was
6420 obtained from HW or FW using either an MSR or CPUID(0x15).
6421 Warn if the difference is more than 500 ppm.
6422 [x86] watchdog: Use TSC as the watchdog clocksource with
6423 which to check other HW timers (HPET or PM timer), but
6424 only on systems where TSC has been deemed trustworthy.
6425 This will be suppressed by an earlier tsc=nowatchdog and
6426 can be overridden by a later tsc=nowatchdog. A console
6427 message will flag any such suppression or overriding.
6429 tsc_early_khz= [X86] Skip early TSC calibration and use the given
6430 value instead. Useful when the early TSC frequency discovery
6431 procedure is not reliable, such as on overclocked systems
6432 with CPUID.16h support and partial CPUID.15h support.
6433 Format: <unsigned int>
6435 tsx= [X86] Control Transactional Synchronization
6436 Extensions (TSX) feature in Intel processors that
6437 support TSX control.
6439 This parameter controls the TSX feature. The options are:
6441 on - Enable TSX on the system. Although there are
6442 mitigations for all known security vulnerabilities,
6443 TSX has been known to be an accelerator for
6444 several previous speculation-related CVEs, and
6445 so there may be unknown security risks associated
6446 with leaving it enabled.
6448 off - Disable TSX on the system. (Note that this
6449 option takes effect only on newer CPUs which are
6450 not vulnerable to MDS, i.e., have
6451 MSR_IA32_ARCH_CAPABILITIES.MDS_NO=1 and which get
6452 the new IA32_TSX_CTRL MSR through a microcode
6453 update. This new MSR allows for the reliable
6454 deactivation of the TSX functionality.)
6456 auto - Disable TSX if X86_BUG_TAA is present,
6457 otherwise enable TSX on the system.
6459 Not specifying this option is equivalent to tsx=off.
6461 See Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
6464 tsx_async_abort= [X86,INTEL] Control mitigation for the TSX Async
6465 Abort (TAA) vulnerability.
6467 Similar to Micro-architectural Data Sampling (MDS)
6468 certain CPUs that support Transactional
6469 Synchronization Extensions (TSX) are vulnerable to an
6470 exploit against CPU internal buffers which can forward
6471 information to a disclosure gadget under certain
6474 In vulnerable processors, the speculatively forwarded
6475 data can be used in a cache side channel attack, to
6476 access data to which the attacker does not have direct
6479 This parameter controls the TAA mitigation. The
6482 full - Enable TAA mitigation on vulnerable CPUs
6485 full,nosmt - Enable TAA mitigation and disable SMT on
6486 vulnerable CPUs. If TSX is disabled, SMT
6487 is not disabled because CPU is not
6488 vulnerable to cross-thread TAA attacks.
6489 off - Unconditionally disable TAA mitigation
6491 On MDS-affected machines, tsx_async_abort=off can be
6492 prevented by an active MDS mitigation as both vulnerabilities
6493 are mitigated with the same mechanism so in order to disable
6494 this mitigation, you need to specify mds=off too.
6496 Not specifying this option is equivalent to
6497 tsx_async_abort=full. On CPUs which are MDS affected
6498 and deploy MDS mitigation, TAA mitigation is not
6499 required and doesn't provide any additional
6503 Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
6505 turbografx.map[2|3]= [HW,JOY]
6506 TurboGraFX parallel port interface
6508 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7>
6509 See also Documentation/input/devices/joystick-parport.rst
6511 udbg-immortal [PPC] When debugging early kernel crashes that
6512 happen after console_init() and before a proper
6513 console driver takes over, this boot options might
6514 help "seeing" what's going on.
6516 uhash_entries= [KNL,NET]
6517 Set number of hash buckets for UDP/UDP-Lite connections
6520 [USB] Ignore overcurrent events (default N).
6521 Some badly-designed motherboards generate lots of
6522 bogus events, for ports that aren't wired to
6523 anything. Set this parameter to avoid log spamming.
6524 Note that genuine overcurrent events won't be
6528 [X86] Cause panic on unknown NMI.
6530 usbcore.authorized_default=
6531 [USB] Default USB device authorization:
6532 (default -1 = authorized except for wireless USB,
6533 0 = not authorized, 1 = authorized, 2 = authorized
6534 if device connected to internal port)
6536 usbcore.autosuspend=
6537 [USB] The autosuspend time delay (in seconds) used
6538 for newly-detected USB devices (default 2). This
6539 is the time required before an idle device will be
6540 autosuspended. Devices for which the delay is set
6541 to a negative value won't be autosuspended at all.
6543 usbcore.usbfs_snoop=
6544 [USB] Set to log all usbfs traffic (default 0 = off).
6546 usbcore.usbfs_snoop_max=
6547 [USB] Maximum number of bytes to snoop in each URB
6550 usbcore.blinkenlights=
6551 [USB] Set to cycle leds on hubs (default 0 = off).
6553 usbcore.old_scheme_first=
6554 [USB] Start with the old device initialization
6555 scheme (default 0 = off).
6557 usbcore.usbfs_memory_mb=
6558 [USB] Memory limit (in MB) for buffers allocated by
6559 usbfs (default = 16, 0 = max = 2047).
6561 usbcore.use_both_schemes=
6562 [USB] Try the other device initialization scheme
6563 if the first one fails (default 1 = enabled).
6565 usbcore.initial_descriptor_timeout=
6566 [USB] Specifies timeout for the initial 64-byte
6567 USB_REQ_GET_DESCRIPTOR request in milliseconds
6568 (default 5000 = 5.0 seconds).
6570 usbcore.nousb [USB] Disable the USB subsystem
6573 [USB] A list of quirk entries to augment the built-in
6574 usb core quirk list. List entries are separated by
6575 commas. Each entry has the form
6576 VendorID:ProductID:Flags. The IDs are 4-digit hex
6577 numbers and Flags is a set of letters. Each letter
6578 will change the built-in quirk; setting it if it is
6579 clear and clearing it if it is set. The letters have
6580 the following meanings:
6581 a = USB_QUIRK_STRING_FETCH_255 (string
6582 descriptors must not be fetched using
6584 b = USB_QUIRK_RESET_RESUME (device can't resume
6585 correctly so reset it instead);
6586 c = USB_QUIRK_NO_SET_INTF (device can't handle
6587 Set-Interface requests);
6588 d = USB_QUIRK_CONFIG_INTF_STRINGS (device can't
6589 handle its Configuration or Interface
6591 e = USB_QUIRK_RESET (device can't be reset
6592 (e.g morph devices), don't use reset);
6593 f = USB_QUIRK_HONOR_BNUMINTERFACES (device has
6594 more interface descriptions than the
6595 bNumInterfaces count, and can't handle
6596 talking to these interfaces);
6597 g = USB_QUIRK_DELAY_INIT (device needs a pause
6598 during initialization, after we read
6599 the device descriptor);
6600 h = USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL (For
6601 high speed and super speed interrupt
6602 endpoints, the USB 2.0 and USB 3.0 spec
6603 require the interval in microframes (1
6604 microframe = 125 microseconds) to be
6605 calculated as interval = 2 ^
6607 Devices with this quirk report their
6608 bInterval as the result of this
6609 calculation instead of the exponent
6610 variable used in the calculation);
6611 i = USB_QUIRK_DEVICE_QUALIFIER (device can't
6612 handle device_qualifier descriptor
6614 j = USB_QUIRK_IGNORE_REMOTE_WAKEUP (device
6615 generates spurious wakeup, ignore
6616 remote wakeup capability);
6617 k = USB_QUIRK_NO_LPM (device can't handle Link
6619 l = USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL
6620 (Device reports its bInterval as linear
6621 frames instead of the USB 2.0
6623 m = USB_QUIRK_DISCONNECT_SUSPEND (Device needs
6624 to be disconnected before suspend to
6625 prevent spurious wakeup);
6626 n = USB_QUIRK_DELAY_CTRL_MSG (Device needs a
6627 pause after every control message);
6628 o = USB_QUIRK_HUB_SLOW_RESET (Hub needs extra
6629 delay after resetting its port);
6630 Example: quirks=0781:5580:bk,0a5c:5834:gij
6633 [USBHID] The interval which mice are to be polled at.
6636 [USBHID] The interval which joysticks are to be polled at.
6639 [USBHID] The interval which keyboards are to be polled at.
6641 usb-storage.delay_use=
6642 [UMS] The delay in seconds before a new device is
6643 scanned for Logical Units (default 1).
6646 [UMS] A list of quirks entries to supplement or
6647 override the built-in unusual_devs list. List
6648 entries are separated by commas. Each entry has
6649 the form VID:PID:Flags where VID and PID are Vendor
6650 and Product ID values (4-digit hex numbers) and
6651 Flags is a set of characters, each corresponding
6652 to a common usb-storage quirk flag as follows:
6653 a = SANE_SENSE (collect more than 18 bytes
6654 of sense data, not on uas);
6655 b = BAD_SENSE (don't collect more than 18
6656 bytes of sense data, not on uas);
6657 c = FIX_CAPACITY (decrease the reported
6658 device capacity by one sector);
6659 d = NO_READ_DISC_INFO (don't use
6660 READ_DISC_INFO command, not on uas);
6661 e = NO_READ_CAPACITY_16 (don't use
6662 READ_CAPACITY_16 command);
6663 f = NO_REPORT_OPCODES (don't use report opcodes
6665 g = MAX_SECTORS_240 (don't transfer more than
6666 240 sectors at a time, uas only);
6667 h = CAPACITY_HEURISTICS (decrease the
6668 reported device capacity by one
6669 sector if the number is odd);
6670 i = IGNORE_DEVICE (don't bind to this
6672 j = NO_REPORT_LUNS (don't use report luns
6674 k = NO_SAME (do not use WRITE_SAME, uas only)
6675 l = NOT_LOCKABLE (don't try to lock and
6676 unlock ejectable media, not on uas);
6677 m = MAX_SECTORS_64 (don't transfer more
6678 than 64 sectors = 32 KB at a time,
6680 n = INITIAL_READ10 (force a retry of the
6681 initial READ(10) command, not on uas);
6682 o = CAPACITY_OK (accept the capacity
6683 reported by the device, not on uas);
6684 p = WRITE_CACHE (the device cache is ON
6685 by default, not on uas);
6686 r = IGNORE_RESIDUE (the device reports
6687 bogus residue values, not on uas);
6688 s = SINGLE_LUN (the device has only one
6690 t = NO_ATA_1X (don't allow ATA(12) and ATA(16)
6691 commands, uas only);
6692 u = IGNORE_UAS (don't bind to the uas driver);
6693 w = NO_WP_DETECT (don't test whether the
6694 medium is write-protected).
6695 y = ALWAYS_SYNC (issue a SYNCHRONIZE_CACHE
6696 even if the device claims no cache,
6698 Example: quirks=0419:aaf5:rl,0421:0433:rc
6700 user_debug= [KNL,ARM]
6702 See arch/arm/Kconfig.debug help text.
6703 1 - undefined instruction events
6705 4 - invalid data aborts
6708 Example: user_debug=31
6711 [X86] Flags controlling user PTE allocations.
6713 nohigh = do not allocate PTE pages in
6714 HIGHMEM regardless of setting
6717 vdso= [X86,SH,SPARC]
6718 On X86_32, this is an alias for vdso32=. Otherwise:
6720 vdso=1: enable VDSO (the default)
6721 vdso=0: disable VDSO mapping
6723 vdso32= [X86] Control the 32-bit vDSO
6724 vdso32=1: enable 32-bit VDSO
6725 vdso32=0 or vdso32=2: disable 32-bit VDSO
6727 See the help text for CONFIG_COMPAT_VDSO for more
6728 details. If CONFIG_COMPAT_VDSO is set, the default is
6729 vdso32=0; otherwise, the default is vdso32=1.
6731 For compatibility with older kernels, vdso32=2 is an
6734 Try vdso32=0 if you encounter an error that says:
6735 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed!
6738 vector=percpu: enable percpu vector domain
6740 video= [FB] Frame buffer configuration
6741 See Documentation/fb/modedb.rst.
6743 video.brightness_switch_enabled= [ACPI]
6745 If set to 1, on receiving an ACPI notify event
6746 generated by hotkey, video driver will adjust brightness
6747 level and then send out the event to user space through
6748 the allocated input device. If set to 0, video driver
6749 will only send out the event without touching backlight
6754 [VMMIO] Memory mapped virtio (platform) device.
6756 <size>@<baseaddr>:<irq>[:<id>]
6758 <size> := size (can use standard suffixes
6760 <baseaddr> := physical base address
6761 <irq> := interrupt number (as passed to
6763 <id> := (optional) platform device id
6765 virtio_mmio.device=1K@0x100b0000:48:7
6767 Can be used multiple times for multiple devices.
6769 vga= [BOOT,X86-32] Select a particular video mode
6770 See Documentation/arch/x86/boot.rst and
6771 Documentation/admin-guide/svga.rst.
6772 Use vga=ask for menu.
6773 This is actually a boot loader parameter; the value is
6774 passed to the kernel using a special protocol.
6776 vm_debug[=options] [KNL] Available with CONFIG_DEBUG_VM=y.
6777 May slow down system boot speed, especially when
6778 enabled on systems with a large amount of memory.
6779 All options are enabled by default, and this
6780 interface is meant to allow for selectively
6781 enabling or disabling specific virtual memory
6784 Available options are:
6785 P Enable page structure init time poisoning
6786 - Disable all of the above options
6788 vmalloc=nn[KMG] [KNL,BOOT] Forces the vmalloc area to have an exact
6789 size of <nn>. This can be used to increase the
6790 minimum size (128MB on x86). It can also be used to
6791 decrease the size and leave more room for directly
6794 vmcp_cma=nn[MG] [KNL,S390]
6795 Sets the memory size reserved for contiguous memory
6796 allocations for the vmcp device driver.
6798 vmhalt= [KNL,S390] Perform z/VM CP command after system halt.
6801 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic.
6804 vmpoff= [KNL,S390] Perform z/VM CP command after power off.
6808 Controls the behavior of vsyscalls (i.e. calls to
6809 fixed addresses of 0xffffffffff600x00 from legacy
6810 code). Most statically-linked binaries and older
6811 versions of glibc use these calls. Because these
6812 functions are at fixed addresses, they make nice
6813 targets for exploits that can control RIP.
6815 emulate Vsyscalls turn into traps and are emulated
6816 reasonably safely. The vsyscall page is
6819 xonly [default] Vsyscalls turn into traps and are
6820 emulated reasonably safely. The vsyscall
6821 page is not readable.
6823 none Vsyscalls don't work at all. This makes
6824 them quite hard to use for exploits but
6825 might break your system.
6827 vt.color= [VT] Default text color.
6828 Format: 0xYX, X = foreground, Y = background.
6829 Default: 0x07 = light gray on black.
6831 vt.cur_default= [VT] Default cursor shape.
6832 Format: 0xCCBBAA, where AA, BB, and CC are the same as
6833 the parameters of the <Esc>[?A;B;Cc escape sequence;
6834 see VGA-softcursor.txt. Default: 2 = underline.
6836 vt.default_blu= [VT]
6837 Format: <blue0>,<blue1>,<blue2>,...,<blue15>
6838 Change the default blue palette of the console.
6839 This is a 16-member array composed of values
6842 vt.default_grn= [VT]
6843 Format: <green0>,<green1>,<green2>,...,<green15>
6844 Change the default green palette of the console.
6845 This is a 16-member array composed of values
6848 vt.default_red= [VT]
6849 Format: <red0>,<red1>,<red2>,...,<red15>
6850 Change the default red palette of the console.
6851 This is a 16-member array composed of values
6857 Set system-wide default UTF-8 mode for all tty's.
6858 Default is 1, i.e. UTF-8 mode is enabled for all
6859 newly opened terminals.
6861 vt.global_cursor_default=
6864 Set system-wide default for whether a cursor
6865 is shown on new VTs. Default is -1,
6866 i.e. cursors will be created by default unless
6867 overridden by individual drivers. 0 will hide
6868 cursors, 1 will display them.
6870 vt.italic= [VT] Default color for italic text; 0-15.
6873 vt.underline= [VT] Default color for underlined text; 0-15.
6876 watchdog timers [HW,WDT] For information on watchdog timers,
6877 see Documentation/watchdog/watchdog-parameters.rst
6878 or other driver-specific files in the
6879 Documentation/watchdog/ directory.
6883 Set the hard lockup detector stall duration
6884 threshold in seconds. The soft lockup detector
6885 threshold is set to twice the value. A value of 0
6886 disables both lockup detectors. Default is 10
6889 workqueue.watchdog_thresh=
6890 If CONFIG_WQ_WATCHDOG is configured, workqueue can
6891 warn stall conditions and dump internal state to
6892 help debugging. 0 disables workqueue stall
6893 detection; otherwise, it's the stall threshold
6894 duration in seconds. The default value is 30 and
6895 it can be updated at runtime by writing to the
6896 corresponding sysfs file.
6898 workqueue.disable_numa
6899 By default, all work items queued to unbound
6900 workqueues are affine to the NUMA nodes they're
6901 issued on, which results in better behavior in
6902 general. If NUMA affinity needs to be disabled for
6903 whatever reason, this option can be used. Note
6904 that this also can be controlled per-workqueue for
6905 workqueues visible under /sys/bus/workqueue/.
6907 workqueue.power_efficient
6908 Per-cpu workqueues are generally preferred because
6909 they show better performance thanks to cache
6910 locality; unfortunately, per-cpu workqueues tend to
6911 be more power hungry than unbound workqueues.
6913 Enabling this makes the per-cpu workqueues which
6914 were observed to contribute significantly to power
6915 consumption unbound, leading to measurably lower
6916 power usage at the cost of small performance
6919 The default value of this parameter is determined by
6920 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT.
6922 workqueue.debug_force_rr_cpu
6923 Workqueue used to implicitly guarantee that work
6924 items queued without explicit CPU specified are put
6925 on the local CPU. This guarantee is no longer true
6926 and while local CPU is still preferred work items
6927 may be put on foreign CPUs. This debug option
6928 forces round-robin CPU selection to flush out
6929 usages which depend on the now broken guarantee.
6930 When enabled, memory and cache locality will be
6933 writecombine= [LOONGARCH] Control the MAT (Memory Access Type) of
6936 on - Enable writecombine, use WUC for ioremap_wc()
6937 off - Disable writecombine, use SUC for ioremap_wc()
6939 x2apic_phys [X86-64,APIC] Use x2apic physical mode instead of
6940 default x2apic cluster mode on platforms
6943 xen_512gb_limit [KNL,X86-64,XEN]
6944 Restricts the kernel running paravirtualized under Xen
6945 to use only up to 512 GB of RAM. The reason to do so is
6946 crash analysis tools and Xen tools for doing domain
6947 save/restore/migration must be enabled to handle larger
6950 xen_emul_unplug= [HW,X86,XEN]
6951 Unplug Xen emulated devices
6952 Format: [unplug0,][unplug1]
6953 ide-disks -- unplug primary master IDE devices
6954 aux-ide-disks -- unplug non-primary-master IDE devices
6955 nics -- unplug network devices
6956 all -- unplug all emulated devices (NICs and IDE disks)
6957 unnecessary -- unplugging emulated devices is
6958 unnecessary even if the host did not respond to
6960 never -- do not unplug even if version check succeeds
6962 xen_legacy_crash [X86,XEN]
6963 Crash from Xen panic notifier, without executing late
6964 panic() code such as dumping handler.
6966 xen_msr_safe= [X86,XEN]
6968 Select whether to always use non-faulting (safe) MSR
6969 access functions when running as Xen PV guest. The
6970 default value is controlled by CONFIG_XEN_PV_MSR_SAFE.
6972 xen_nopvspin [X86,XEN]
6973 Disables the qspinlock slowpath using Xen PV optimizations.
6974 This parameter is obsoleted by "nopvspin" parameter, which
6975 has equivalent effect for XEN platform.
6978 Disables the PV optimizations forcing the HVM guest to
6979 run as generic HVM guest with no PV drivers.
6980 This option is obsoleted by the "nopv" option, which
6981 has equivalent effect for XEN platform.
6983 xen_no_vector_callback
6984 [KNL,X86,XEN] Disable the vector callback for Xen
6985 event channel interrupts.
6987 xen_scrub_pages= [XEN]
6988 Boolean option to control scrubbing pages before giving them back
6989 to Xen, for use by other domains. Can be also changed at runtime
6990 with /sys/devices/system/xen_memory/xen_memory0/scrub_pages.
6991 Default value controlled with CONFIG_XEN_SCRUB_PAGES_DEFAULT.
6993 xen_timer_slop= [X86-64,XEN]
6994 Set the timer slop (in nanoseconds) for the virtual Xen
6995 timers (default is 100000). This adjusts the minimum
6996 delta of virtualized Xen timers, where lower values
6997 improve timer resolution at the expense of processing
6998 more timer interrupts.
7000 xen.balloon_boot_timeout= [XEN]
7001 The time (in seconds) to wait before giving up to boot
7002 in case initial ballooning fails to free enough memory.
7003 Applies only when running as HVM or PVH guest and
7004 started with less memory configured than allowed at
7005 max. Default is 180.
7007 xen.event_eoi_delay= [XEN]
7008 How long to delay EOI handling in case of event
7009 storms (jiffies). Default is 10.
7011 xen.event_loop_timeout= [XEN]
7012 After which time (jiffies) the event handling loop
7013 should start to delay EOI handling. Default is 2.
7015 xen.fifo_events= [XEN]
7016 Boolean parameter to disable using fifo event handling
7017 even if available. Normally fifo event handling is
7018 preferred over the 2-level event handling, as it is
7019 fairer and the number of possible event channels is
7020 much higher. Default is on (use fifo events).
7022 xirc2ps_cs= [NET,PCMCIA]
7024 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]]
7027 By default on POWER9 and above, the kernel will
7028 natively use the XIVE interrupt controller. This option
7029 allows the fallback firmware mode to be used:
7031 off Fallback to firmware control of XIVE interrupt
7032 controller on both pseries and powernv
7033 platforms. Only useful on POWER9 and above.
7035 xive.store-eoi=off [PPC]
7036 By default on POWER10 and above, the kernel will use
7037 stores for EOI handling when the XIVE interrupt mode
7038 is active. This option allows the XIVE driver to use
7039 loads instead, as on POWER9.
7041 xhci-hcd.quirks [USB,KNL]
7042 A hex value specifying bitmask with supplemental xhci
7043 host controller quirks. Meaning of each bit can be
7044 consulted in header drivers/usb/host/xhci.h.
7047 Format: { early | on | rw | ro | off }
7048 Controls if xmon debugger is enabled. Default is off.
7049 Passing only "xmon" is equivalent to "xmon=early".
7050 early Call xmon as early as possible on boot; xmon
7051 debugger is called from setup_arch().
7052 on xmon debugger hooks will be installed so xmon
7053 is only called on a kernel crash. Default mode,
7054 i.e. either "ro" or "rw" mode, is controlled
7055 with CONFIG_XMON_DEFAULT_RO_MODE.
7056 rw xmon debugger hooks will be installed so xmon
7057 is called only on a kernel crash, mode is write,
7058 meaning SPR registers, memory and, other data
7059 can be written using xmon commands.
7060 ro same as "rw" option above but SPR registers,
7061 memory, and other data can't be written using
7063 off xmon is disabled.
7067 Do not enable amd_pstate as the default
7068 scaling driver for the supported processors
7070 Use amd_pstate as a scaling driver, driver requests a
7071 desired performance on this abstract scale and the power
7072 management firmware translates the requests into actual
7073 hardware states (core frequency, data fabric and memory
7076 Use amd_pstate_epp driver instance as the scaling driver,
7077 driver provides a hint to the hardware if software wants
7078 to bias toward performance (0x0) or energy efficiency (0xff)
7079 to the CPPC firmware. then CPPC power algorithm will
7080 calculate the runtime workload and adjust the realtime cores
projects / linux-block.git / blob