← Documents Documentation/admin-guide/kernel-parameters.txt GitHub 원문 ↗

Linux 6.18.37 · Administration / Boot Parameters

Linux kernel command-line parameters

Linux v6.18.37의 전체 kernel command-line parameter를 architecture·subsystem 표기, 형식, 기본값, 보안·성능 영향과 함께 원문 순서대로 번역합니다.

Source pathDocumentation/admin-guide/kernel-parameters.txt
Source versionLinux v6.18.37
TranslationDUJINLABS 전문 번역 + 해설

요약·해설과 원문, 전문 번역을 서로 분리했습니다. API 이름, symbol, source path는 원문 표기를 사용합니다.

1. 요약·해설

원문의 핵심 논리와 kernel programming 관점의 보충 설명입니다. 아래의 전문 번역과는 별도로 작성했습니다.

전체 번역 구간 1: L1-702

kernel-parameters.txt:1-702

문서 표제와 공통 표기법부터 ACPI·APIC·ARM·cgroup·CPUID까지 A–C 초반 boot parameter를 번역합니다.

  • Architecture·hardware·debug tag와 option 형식을 원문 순서대로 유지합니다.

전체 번역 구간 2: L703-1316

kernel-parameters.txt:703-1316

C–E 구간의 CPU, crash, device, DMA, driver, DRM, dynamic-debug parameter를 번역합니다.

  • Serial mapping과 device별 format·source path를 그대로 보존합니다.

전체 번역 구간 3: L1317-2038

kernel-parameters.txt:1317-2038

E–H 구간의 EFI, filesystem, firmware, hugepage, Hyper-V parameter를 번역합니다.

  • HugeTLB 크기·node 형식과 firmware 기능 이름을 원문과 대조할 수 있습니다.

전체 번역 구간 4: L2039-2759

kernel-parameters.txt:2039-2759

I 구간의 init, IOMMU, IRQ, isolation, IVRS mapping parameter를 번역합니다.

  • IOMMU와 CPU isolation option은 성능·안정성 trade-off를 함께 설명합니다.

전체 번역 구간 5: L2760-3321

kernel-parameters.txt:2760-3321

J–L 구간의 KASAN/KVM, L1TF, LAPIC와 libata 강제 구성 parameter를 번역합니다.

  • 보안 완화 mode와 `libata.force=` 선택지를 symbol·철자 그대로 보존합니다.

전체 번역 구간 6: L3322-3954

kernel-parameters.txt:3322-3954

L–M 구간의 lockup, memory, MIPS, 통합 CPU mitigation과 MMIO stale-data parameter를 번역합니다.

  • Memory sizing과 취약점 완화의 기본값·상호작용을 구분해 설명합니다.

전체 번역 구간 7: L3955-4555

kernel-parameters.txt:3955-4555

M–N 구간의 module, network, NFS, NMI, no-* switch와 NUMA parameter를 번역합니다.

  • Feature disable switch와 NUMA topology 형식을 원문 순서대로 제공합니다.

전체 번역 구간 8: L4556-5268

kernel-parameters.txt:4556-5268

O–P 구간의 OOM, PCI, perf, printk, processor와 pstore parameter를 번역합니다.

  • PCI routing·resource option과 printk/pstore 진단 설정을 함께 해설합니다.

전체 번역 구간 9: L5269-5726

kernel-parameters.txt:5269-5726

Q–R 구간의 QLogic, RCU, reboot, reserve-memory와 RISC-V parameter를 번역합니다.

  • RCU timing·stall option과 reboot path의 architecture 차이를 보존합니다.

전체 번역 구간 10: L5727-6194

kernel-parameters.txt:5727-6194

R–S 구간의 root filesystem, SCSI, security, serial, slab과 software-I/O parameter를 번역합니다.

  • Storage identifier·queue·timeout 단위와 security mode를 정확히 유지합니다.

전체 번역 구간 11: L6195-6655

kernel-parameters.txt:6195-6655

S 구간의 scheduler, SELinux, S390, serial console, slab과 softlockup parameter를 번역합니다.

  • Scheduler·console 형식과 S390 command option을 source symbol과 함께 제공합니다.

전체 번역 구간 12: L6656-7116

kernel-parameters.txt:6656-7116

S 구간의 spectrum, speculation, SPI, suspend, storage와 systemd 관련 parameter를 번역합니다.

  • Speculation control과 suspend/storage 정책의 적용 범위를 구분합니다.

전체 번역 구간 13: L7117-7601

kernel-parameters.txt:7117-7601

S–T 구간의 SSBD, stack, SunRPC, SWIOTLB, thermal, THP, TPM과 boot tracing parameter를 번역합니다.

  • Persistent trace buffer와 trigger 형식의 원문 예제를 보존합니다.

전체 번역 구간 14: L7602-8409

kernel-parameters.txt:7602-8409

T–X 구간의 THP, TSC/TSX, USB, vDSO/VM, VT, watchdog/workqueue, Xen과 xmon parameter를 번역합니다.

  • USB quirk symbol과 Xen event·timer, workqueue affinity의 동작을 상세히 설명합니다.

2. 영어 원문 전체

번역 기준이 된 Linux v6.18.37 원문입니다. 줄 번호는 이 버전의 파일 좌표입니다.

원문 전체 펼치기
1 accept_memory= [MM]
2 Format: { eager | lazy }
3 default: lazy
4 By default, unaccepted memory is accepted lazily to
5 avoid prolonged boot times. The lazy option will add
6 some runtime overhead until all memory is eventually
7 accepted. In most cases the overhead is negligible.
8 For some workloads or for debugging purposes
9 accept_memory=eager can be used to accept all memory
10 at once during boot.
11
12 acpi= [HW,ACPI,X86,ARM64,RISCV64,EARLY]
13 Advanced Configuration and Power Interface
14 Format: { force | on | off | strict | noirq | rsdt |
15 copy_dsdt | nospcr }
16 force -- enable ACPI if default was off
17 on -- enable ACPI but allow fallback to DT [arm64,riscv64]
18 off -- disable ACPI if default was on
19 noirq -- do not use ACPI for IRQ routing
20 strict -- Be less tolerant of platforms that are not
21 strictly ACPI specification compliant.
22 rsdt -- prefer RSDT over (default) XSDT
23 copy_dsdt -- copy DSDT to memory
24 nocmcff -- Disable firmware first mode for corrected
25 errors. This disables parsing the HEST CMC error
26 source to check if firmware has set the FF flag. This
27 may result in duplicate corrected error reports.
28 nospcr -- disable console in ACPI SPCR table as
29 default _serial_ console on ARM64
30 For ARM64, ONLY "acpi=off", "acpi=on", "acpi=force" or
31 "acpi=nospcr" are available
32 For RISCV64, ONLY "acpi=off", "acpi=on" or "acpi=force"
33 are available
34
35 See also Documentation/power/runtime_pm.rst, pci=noacpi
36
37 acpi_apic_instance= [ACPI,IOAPIC,EARLY]
38 Format: <int>
39 2: use 2nd APIC table, if available
40 1,0: use 1st APIC table
41 default: 0
42
43 acpi_backlight= [HW,ACPI]
44 { vendor | video | native | none }
45 If set to vendor, prefer vendor-specific driver
46 (e.g. thinkpad_acpi, sony_acpi, etc.) instead
47 of the ACPI video.ko driver.
48 If set to video, use the ACPI video.ko driver.
49 If set to native, use the device's native backlight mode.
50 If set to none, disable the ACPI backlight interface.
51
52 acpi_force_32bit_fadt_addr [ACPI,EARLY]
53 force FADT to use 32 bit addresses rather than the
54 64 bit X_* addresses. Some firmware have broken 64
55 bit addresses for force ACPI ignore these and use
56 the older legacy 32 bit addresses.
57
58 acpica_no_return_repair [HW, ACPI]
59 Disable AML predefined validation mechanism
60 This mechanism can repair the evaluation result to make
61 the return objects more ACPI specification compliant.
62 This option is useful for developers to identify the
63 root cause of an AML interpreter issue when the issue
64 has something to do with the repair mechanism.
65
66 acpi.debug_layer= [HW,ACPI,ACPI_DEBUG]
67 acpi.debug_level= [HW,ACPI,ACPI_DEBUG]
68 Format: <int>
69 CONFIG_ACPI_DEBUG must be enabled to produce any ACPI
70 debug output. Bits in debug_layer correspond to a
71 _COMPONENT in an ACPI source file, e.g.,
72 #define _COMPONENT ACPI_EVENTS
73 Bits in debug_level correspond to a level in
74 ACPI_DEBUG_PRINT statements, e.g.,
75 ACPI_DEBUG_PRINT((ACPI_DB_INFO, ...
76 The debug_level mask defaults to "info". See
77 Documentation/firmware-guide/acpi/debug.rst for more information about
78 debug layers and levels.
79
80 Enable processor driver info messages:
81 acpi.debug_layer=0x20000000
82 Enable AML "Debug" output, i.e., stores to the Debug
83 object while interpreting AML:
84 acpi.debug_layer=0xffffffff acpi.debug_level=0x2
85 Enable all messages related to ACPI hardware:
86 acpi.debug_layer=0x2 acpi.debug_level=0xffffffff
87
88 Some values produce so much output that the system is
89 unusable. The "log_buf_len" parameter may be useful
90 if you need to capture more output.
91
92 acpi_enforce_resources= [ACPI]
93 { strict | lax | no }
94 Check for resource conflicts between native drivers
95 and ACPI OperationRegions (SystemIO and SystemMemory
96 only). IO ports and memory declared in ACPI might be
97 used by the ACPI subsystem in arbitrary AML code and
98 can interfere with legacy drivers.
99 strict (default): access to resources claimed by ACPI
100 is denied; legacy drivers trying to access reserved
101 resources will fail to bind to device using them.
102 lax: access to resources claimed by ACPI is allowed;
103 legacy drivers trying to access reserved resources
104 will bind successfully but a warning message is logged.
105 no: ACPI OperationRegions are not marked as reserved,
106 no further checks are performed.
107
108 acpi_force_table_verification [HW,ACPI,EARLY]
109 Enable table checksum verification during early stage.
110 By default, this is disabled due to x86 early mapping
111 size limitation.
112
113 acpi_irq_balance [HW,ACPI]
114 ACPI will balance active IRQs
115 default in APIC mode
116
117 acpi_irq_nobalance [HW,ACPI]
118 ACPI will not move active IRQs (default)
119 default in PIC mode
120
121 acpi_irq_isa= [HW,ACPI] If irq_balance, mark listed IRQs used by ISA
122 Format: <irq>,<irq>...
123
124 acpi_irq_pci= [HW,ACPI] If irq_balance, clear listed IRQs for
125 use by PCI
126 Format: <irq>,<irq>...
127
128 acpi_mask_gpe= [HW,ACPI]
129 Due to the existence of _Lxx/_Exx, some GPEs triggered
130 by unsupported hardware/firmware features can result in
131 GPE floodings that cannot be automatically disabled by
132 the GPE dispatcher.
133 This facility can be used to prevent such uncontrolled
134 GPE floodings.
135 Format: <byte> or <bitmap-list>
136
137 acpi_no_auto_serialize [HW,ACPI]
138 Disable auto-serialization of AML methods
139 AML control methods that contain the opcodes to create
140 named objects will be marked as "Serialized" by the
141 auto-serialization feature.
142 This feature is enabled by default.
143 This option allows to turn off the feature.
144
145 acpi_no_memhotplug [ACPI] Disable memory hotplug. Useful for kdump
146 kernels.
147
148 acpi_no_static_ssdt [HW,ACPI,EARLY]
149 Disable installation of static SSDTs at early boot time
150 By default, SSDTs contained in the RSDT/XSDT will be
151 installed automatically and they will appear under
152 /sys/firmware/acpi/tables.
153 This option turns off this feature.
154 Note that specifying this option does not affect
155 dynamic table installation which will install SSDT
156 tables to /sys/firmware/acpi/tables/dynamic.
157
158 acpi_no_watchdog [HW,ACPI,WDT]
159 Ignore the ACPI-based watchdog interface (WDAT) and let
160 a native driver control the watchdog device instead.
161
162 acpi_rsdp= [ACPI,EFI,KEXEC,EARLY]
163 Pass the RSDP address to the kernel, mostly used
164 on machines running EFI runtime service to boot the
165 second kernel for kdump.
166
167 acpi_os_name= [HW,ACPI] Tell ACPI BIOS the name of the OS
168 Format: To spoof as Windows 98: ="Microsoft Windows"
169
170 acpi_rev_override [ACPI] Override the _REV object to return 5 (instead
171 of 2 which is mandated by ACPI 6) as the supported ACPI
172 specification revision (when using this switch, it may
173 be necessary to carry out a cold reboot _twice_ in a
174 row to make it take effect on the platform firmware).
175
176 acpi_osi= [HW,ACPI] Modify list of supported OS interface strings
177 acpi_osi="string1" # add string1
178 acpi_osi="!string2" # remove string2
179 acpi_osi=!* # remove all strings
180 acpi_osi=! # disable all built-in OS vendor
181 strings
182 acpi_osi=!! # enable all built-in OS vendor
183 strings
184 acpi_osi= # disable all strings
185
186 'acpi_osi=!' can be used in combination with single or
187 multiple 'acpi_osi="string1"' to support specific OS
188 vendor string(s). Note that such command can only
189 affect the default state of the OS vendor strings, thus
190 it cannot affect the default state of the feature group
191 strings and the current state of the OS vendor strings,
192 specifying it multiple times through kernel command line
193 is meaningless. This command is useful when one do not
194 care about the state of the feature group strings which
195 should be controlled by the OSPM.
196 Examples:
197 1. 'acpi_osi=! acpi_osi="Windows 2000"' is equivalent
198 to 'acpi_osi="Windows 2000" acpi_osi=!', they all
199 can make '_OSI("Windows 2000")' TRUE.
200
201 'acpi_osi=' cannot be used in combination with other
202 'acpi_osi=' command lines, the _OSI method will not
203 exist in the ACPI namespace. NOTE that such command can
204 only affect the _OSI support state, thus specifying it
205 multiple times through kernel command line is also
206 meaningless.
207 Examples:
208 1. 'acpi_osi=' can make 'CondRefOf(_OSI, Local1)'
209 FALSE.
210
211 'acpi_osi=!*' can be used in combination with single or
212 multiple 'acpi_osi="string1"' to support specific
213 string(s). Note that such command can affect the
214 current state of both the OS vendor strings and the
215 feature group strings, thus specifying it multiple times
216 through kernel command line is meaningful. But it may
217 still not able to affect the final state of a string if
218 there are quirks related to this string. This command
219 is useful when one want to control the state of the
220 feature group strings to debug BIOS issues related to
221 the OSPM features.
222 Examples:
223 1. 'acpi_osi="Module Device" acpi_osi=!*' can make
224 '_OSI("Module Device")' FALSE.
225 2. 'acpi_osi=!* acpi_osi="Module Device"' can make
226 '_OSI("Module Device")' TRUE.
227 3. 'acpi_osi=! acpi_osi=!* acpi_osi="Windows 2000"' is
228 equivalent to
229 'acpi_osi=!* acpi_osi=! acpi_osi="Windows 2000"'
230 and
231 'acpi_osi=!* acpi_osi="Windows 2000" acpi_osi=!',
232 they all will make '_OSI("Windows 2000")' TRUE.
233
234 acpi_pm_good [X86]
235 Override the pmtimer bug detection: force the kernel
236 to assume that this machine's pmtimer latches its value
237 and always returns good values.
238
239 acpi_sci= [HW,ACPI,EARLY] ACPI System Control Interrupt trigger mode
240 Format: { level | edge | high | low }
241
242 acpi_skip_timer_override [HW,ACPI,EARLY]
243 Recognize and ignore IRQ0/pin2 Interrupt Override.
244 For broken nForce2 BIOS resulting in XT-PIC timer.
245
246 acpi_sleep= [HW,ACPI] Sleep options
247 Format: { s3_bios, s3_mode, s3_beep, s4_hwsig,
248 s4_nohwsig, old_ordering, nonvs,
249 sci_force_enable, nobl }
250 See Documentation/power/video.rst for information on
251 s3_bios and s3_mode.
252 s3_beep is for debugging; it makes the PC's speaker beep
253 as soon as the kernel's real-mode entry point is called.
254 s4_hwsig causes the kernel to check the ACPI hardware
255 signature during resume from hibernation, and gracefully
256 refuse to resume if it has changed. This complies with
257 the ACPI specification but not with reality, since
258 Windows does not do this and many laptops do change it
259 on docking. So the default behaviour is to allow resume
260 and simply warn when the signature changes, unless the
261 s4_hwsig option is enabled.
262 s4_nohwsig prevents ACPI hardware signature from being
263 used (or even warned about) during resume.
264 old_ordering causes the ACPI 1.0 ordering of the _PTS
265 control method, with respect to putting devices into
266 low power states, to be enforced (the ACPI 2.0 ordering
267 of _PTS is used by default).
268 nonvs prevents the kernel from saving/restoring the
269 ACPI NVS memory during suspend/hibernation and resume.
270 sci_force_enable causes the kernel to set SCI_EN directly
271 on resume from S1/S3 (which is against the ACPI spec,
272 but some broken systems don't work without it).
273 nobl causes the internal blacklist of systems known to
274 behave incorrectly in some ways with respect to system
275 suspend and resume to be ignored (use wisely).
276
277 acpi_use_timer_override [HW,ACPI,EARLY]
278 Use timer override. For some broken Nvidia NF5 boards
279 that require a timer override, but don't have HPET
280
281 add_efi_memmap [EFI,X86,EARLY] Include EFI memory map in
282 kernel's map of available physical RAM.
283
284 agp= [AGP]
285 { off | try_unsupported }
286 off: disable AGP support
287 try_unsupported: try to drive unsupported chipsets
288 (may crash computer or cause data corruption)
289
290 ALSA [HW,ALSA]
291 See Documentation/sound/alsa-configuration.rst
292
293 alignment= [KNL,ARM]
294 Allow the default userspace alignment fault handler
295 behaviour to be specified. Bit 0 enables warnings,
296 bit 1 enables fixups, and bit 2 sends a segfault.
297
298 align_va_addr= [X86-64]
299 Align virtual addresses by clearing slice [14:12] when
300 allocating a VMA at process creation time. This option
301 gives you up to 3% performance improvement on AMD F15h
302 machines (where it is enabled by default) for a
303 CPU-intensive style benchmark, and it can vary highly in
304 a microbenchmark depending on workload and compiler.
305
306 32: only for 32-bit processes
307 64: only for 64-bit processes
308 on: enable for both 32- and 64-bit processes
309 off: disable for both 32- and 64-bit processes
310
311 alloc_snapshot [FTRACE]
312 Allocate the ftrace snapshot buffer on boot up when the
313 main buffer is allocated. This is handy if debugging
314 and you need to use tracing_snapshot() on boot up, and
315 do not want to use tracing_snapshot_alloc() as it needs
316 to be done where GFP_KERNEL allocations are allowed.
317
318 allow_mismatched_32bit_el0 [ARM64,EARLY]
319 Allow execve() of 32-bit applications and setting of the
320 PER_LINUX32 personality on systems where only a strict
321 subset of the CPUs support 32-bit EL0. When this
322 parameter is present, the set of CPUs supporting 32-bit
323 EL0 is indicated by /sys/devices/system/cpu/aarch32_el0
324 and hot-unplug operations may be restricted.
325
326 See Documentation/arch/arm64/asymmetric-32bit.rst for more
327 information.
328
329 amd_iommu= [HW,X86-64]
330 Pass parameters to the AMD IOMMU driver in the system.
331 Possible values are:
332 fullflush - Deprecated, equivalent to iommu.strict=1
333 off - do not initialize any AMD IOMMU found in
334 the system
335 force_isolation - Force device isolation for all
336 devices. The IOMMU driver is not
337 allowed anymore to lift isolation
338 requirements as needed. This option
339 does not override iommu=pt
340 force_enable - Force enable the IOMMU on platforms known
341 to be buggy with IOMMU enabled. Use this
342 option with care.
343 pgtbl_v1 - Use v1 page table for DMA-API (Default).
344 pgtbl_v2 - Use v2 page table for DMA-API.
345 irtcachedis - Disable Interrupt Remapping Table (IRT) caching.
346 nohugepages - Limit page-sizes used for v1 page-tables
347 to 4 KiB.
348 v2_pgsizes_only - Limit page-sizes used for v1 page-tables
349 to 4KiB/2Mib/1GiB.
350
351
352 amd_iommu_dump= [HW,X86-64]
353 Enable AMD IOMMU driver option to dump the ACPI table
354 for AMD IOMMU. With this option enabled, AMD IOMMU
355 driver will print ACPI tables for AMD IOMMU during
356 IOMMU initialization.
357
358 amd_iommu_intr= [HW,X86-64]
359 Specifies one of the following AMD IOMMU interrupt
360 remapping modes:
361 legacy - Use legacy interrupt remapping mode.
362 vapic - Use virtual APIC mode, which allows IOMMU
363 to inject interrupts directly into guest.
364 This mode requires kvm-amd.avic=1.
365 (Default when IOMMU HW support is present.)
366
367 amd_pstate= [X86,EARLY]
368 disable
369 Do not enable amd_pstate as the default
370 scaling driver for the supported processors
371 passive
372 Use amd_pstate with passive mode as a scaling driver.
373 In this mode autonomous selection is disabled.
374 Driver requests a desired performance level and platform
375 tries to match the same performance level if it is
376 satisfied by guaranteed performance level.
377 active
378 Use amd_pstate_epp driver instance as the scaling driver,
379 driver provides a hint to the hardware if software wants
380 to bias toward performance (0x0) or energy efficiency (0xff)
381 to the CPPC firmware. then CPPC power algorithm will
382 calculate the runtime workload and adjust the realtime cores
383 frequency.
384 guided
385 Activate guided autonomous mode. Driver requests minimum and
386 maximum performance level and the platform autonomously
387 selects a performance level in this range and appropriate
388 to the current workload.
389
390 amd_prefcore=
391 [X86]
392 disable
393 Disable amd-pstate preferred core.
394
395 amijoy.map= [HW,JOY] Amiga joystick support
396 Map of devices attached to JOY0DAT and JOY1DAT
397 Format: <a>,<b>
398 See also Documentation/input/joydev/joystick.rst
399
400 analog.map= [HW,JOY] Analog joystick and gamepad support
401 Specifies type or capabilities of an analog joystick
402 connected to one of 16 gameports
403 Format: <type1>,<type2>,..<type16>
404
405 apc= [HW,SPARC]
406 Power management functions (SPARCstation-4/5 + deriv.)
407 Format: noidle
408 Disable APC CPU standby support. SPARCstation-Fox does
409 not play well with APC CPU idle - disable it if you have
410 APC and your system crashes randomly.
411
412 apic [APIC,X86-64] Use IO-APIC. Default.
413
414 apic= [APIC,X86,EARLY] Advanced Programmable Interrupt Controller
415 Change the output verbosity while booting
416 Format: { quiet (default) | verbose | debug }
417 Change the amount of debugging information output
418 when initialising the APIC and IO-APIC components.
419
420 apic_extnmi= [APIC,X86,EARLY] External NMI delivery setting
421 Format: { bsp (default) | all | none }
422 bsp: External NMI is delivered only to CPU 0
423 all: External NMIs are broadcast to all CPUs as a
424 backup of CPU 0
425 none: External NMI is masked for all CPUs. This is
426 useful so that a dump capture kernel won't be
427 shot down by NMI
428
429 apicpmtimer Do APIC timer calibration using the pmtimer. Implies
430 apicmaintimer. Useful when your PIT timer is totally
431 broken.
432
433 autoconf= [IPV6]
434 See Documentation/networking/ipv6.rst.
435
436 apm= [APM] Advanced Power Management
437 See header of arch/x86/kernel/apm_32.c.
438
439 apparmor= [APPARMOR] Disable or enable AppArmor at boot time
440 Format: { "0" | "1" }
441 See security/apparmor/Kconfig help text
442 0 -- disable.
443 1 -- enable.
444 Default value is set via kernel config option.
445
446 arcrimi= [HW,NET] ARCnet - "RIM I" (entirely mem-mapped) cards
447 Format: <io>,<irq>,<nodeID>
448
449 arm64.no32bit_el0 [ARM64] Unconditionally disable the execution of
450 32 bit applications.
451
452 arm64.nobti [ARM64] Unconditionally disable Branch Target
453 Identification support
454
455 arm64.nogcs [ARM64] Unconditionally disable Guarded Control Stack
456 support
457
458 arm64.nomops [ARM64] Unconditionally disable Memory Copy and Memory
459 Set instructions support
460
461 arm64.nompam [ARM64] Unconditionally disable Memory Partitioning And
462 Monitoring support
463
464 arm64.nomte [ARM64] Unconditionally disable Memory Tagging Extension
465 support
466
467 arm64.nopauth [ARM64] Unconditionally disable Pointer Authentication
468 support
469
470 arm64.nosme [ARM64] Unconditionally disable Scalable Matrix
471 Extension support
472
473 arm64.nosve [ARM64] Unconditionally disable Scalable Vector
474 Extension support
475
476 ataflop= [HW,M68k]
477
478 atarimouse= [HW,MOUSE] Atari Mouse
479
480 atkbd.extra= [HW] Enable extra LEDs and keys on IBM RapidAccess,
481 EzKey and similar keyboards
482
483 atkbd.reset= [HW] Reset keyboard during initialization
484
485 atkbd.set= [HW] Select keyboard code set
486 Format: <int> (2 = AT (default), 3 = PS/2)
487
488 atkbd.scroll= [HW] Enable scroll wheel on MS Office and similar
489 keyboards
490
491 atkbd.softraw= [HW] Choose between synthetic and real raw mode
492 Format: <bool> (0 = real, 1 = synthetic (default))
493
494 atkbd.softrepeat= [HW]
495 Use software keyboard repeat
496
497 audit= [KNL] Enable the audit sub-system
498 Format: { "0" | "1" | "off" | "on" }
499 0 | off - kernel audit is disabled and can not be
500 enabled until the next reboot
501 unset - kernel audit is initialized but disabled and
502 will be fully enabled by the userspace auditd.
503 1 | on - kernel audit is initialized and partially
504 enabled, storing at most audit_backlog_limit
505 messages in RAM until it is fully enabled by the
506 userspace auditd.
507 Default: unset
508
509 audit_backlog_limit= [KNL] Set the audit queue size limit.
510 Format: <int> (must be >=0)
511 Default: 64
512
513 bau= [X86_UV] Enable the BAU on SGI UV. The default
514 behavior is to disable the BAU (i.e. bau=0).
515 Format: { "0" | "1" }
516 0 - Disable the BAU.
517 1 - Enable the BAU.
518 unset - Disable the BAU.
519
520 baycom_epp= [HW,AX25]
521 Format: <io>,<mode>
522
523 baycom_par= [HW,AX25] BayCom Parallel Port AX.25 Modem
524 Format: <io>,<mode>
525 See header of drivers/net/hamradio/baycom_par.c.
526
527 baycom_ser_fdx= [HW,AX25]
528 BayCom Serial Port AX.25 Modem (Full Duplex Mode)
529 Format: <io>,<irq>,<mode>[,<baud>]
530 See header of drivers/net/hamradio/baycom_ser_fdx.c.
531
532 baycom_ser_hdx= [HW,AX25]
533 BayCom Serial Port AX.25 Modem (Half Duplex Mode)
534 Format: <io>,<irq>,<mode>
535 See header of drivers/net/hamradio/baycom_ser_hdx.c.
536
537 bdev_allow_write_mounted=
538 Format: <bool>
539 Control the ability to open a mounted block device
540 for writing, i.e., allow / disallow writes that bypass
541 the FS. This was implemented as a means to prevent
542 fuzzers from crashing the kernel by overwriting the
543 metadata underneath a mounted FS without its awareness.
544 This also prevents destructive formatting of mounted
545 filesystems by naive storage tooling that don't use
546 O_EXCL. Default is Y and can be changed through the
547 Kconfig option CONFIG_BLK_DEV_WRITE_MOUNTED.
548
549 bert_disable [ACPI]
550 Disable BERT OS support on buggy BIOSes.
551
552 bgrt_disable [ACPI,X86,EARLY]
553 Disable BGRT to avoid flickering OEM logo.
554
555 blkdevparts= Manual partition parsing of block device(s) for
556 embedded devices based on command line input.
557 See Documentation/block/cmdline-partition.rst
558
559 boot_delay= [KNL,EARLY]
560 Milliseconds to delay each printk during boot.
561 Only works if CONFIG_BOOT_PRINTK_DELAY is enabled,
562 and you may also have to specify "lpj=". Boot_delay
563 values larger than 10 seconds (10000) are assumed
564 erroneous and ignored.
565 Format: integer
566
567 bootconfig [KNL,EARLY]
568 Extended command line options can be added to an initrd
569 and this will cause the kernel to look for it.
570
571 See Documentation/admin-guide/bootconfig.rst
572
573 bttv.card= [HW,V4L] bttv (bt848 + bt878 based grabber cards)
574 bttv.radio= Most important insmod options are available as
575 kernel args too.
576 bttv.pll= See Documentation/admin-guide/media/bttv.rst
577 bttv.tuner=
578
579 bulk_remove=off [PPC] This parameter disables the use of the pSeries
580 firmware feature for flushing multiple hpte entries
581 at a time.
582
583 c101= [NET] Moxa C101 synchronous serial card
584
585 cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection.
586 Sometimes CPU hardware bugs make them report the cache
587 size incorrectly. The kernel will attempt work arounds
588 to fix known problems, but for some CPUs it is not
589 possible to determine what the correct size should be.
590 This option provides an override for these situations.
591
592 carrier_timeout=
593 [NET] Specifies amount of time (in seconds) that
594 the kernel should wait for a network carrier. By default
595 it waits 120 seconds.
596
597 ca_keys= [KEYS] This parameter identifies a specific key(s) on
598 the system trusted keyring to be used for certificate
599 trust validation.
600 format: { id:<keyid> | builtin }
601
602 cca= [MIPS,EARLY] Override the kernel pages' cache coherency
603 algorithm. Accepted values range from 0 to 7
604 inclusive. See arch/mips/include/asm/pgtable-bits.h
605 for platform specific values (SB1, Loongson3 and
606 others).
607
608 ccw_timeout_log [S390]
609 See Documentation/arch/s390/common_io.rst for details.
610
611 cfi= [X86-64] Set Control Flow Integrity checking features
612 when CONFIG_FINEIBT is enabled.
613 Format: feature[,feature...]
614 Default: auto
615
616 auto: Use FineIBT if IBT available, otherwise kCFI.
617 Under FineIBT, enable "paranoid" mode when
618 FRED is not available.
619 off: Turn off CFI checking.
620 kcfi: Use kCFI (disable FineIBT).
621 fineibt: Use FineIBT (even if IBT not available).
622 norand: Do not re-randomize CFI hashes.
623 paranoid: Add caller hash checking under FineIBT.
624 bhi: Enable register poisoning to stop speculation
625 across FineIBT. (Disabled by default.)
626 warn: Do not enforce CFI checking: warn only.
627 debug: Report CFI initialization details.
628
629 cgroup_disable= [KNL] Disable a particular controller or optional feature
630 Format: {name of the controller(s) or feature(s) to disable}
631 The effects of cgroup_disable=foo are:
632 - foo isn't auto-mounted if you mount all cgroups in
633 a single hierarchy
634 - foo isn't visible as an individually mountable
635 subsystem
636 - if foo is an optional feature then the feature is
637 disabled and corresponding cgroup files are not
638 created
639 {Currently only "memory" controller deal with this and
640 cut the overhead, others just disable the usage. So
641 only cgroup_disable=memory is actually worthy}
642 Specifying "pressure" disables per-cgroup pressure
643 stall information accounting feature
644
645 cgroup_no_v1= [KNL] Disable cgroup controllers and named hierarchies in v1
646 Format: { { controller | "all" | "named" }
647 [,{ controller | "all" | "named" }...] }
648 Like cgroup_disable, but only applies to cgroup v1;
649 the blacklisted controllers remain available in cgroup2.
650 "all" blacklists all controllers and "named" disables
651 named mounts. Specifying both "all" and "named" disables
652 all v1 hierarchies.
653
654 cgroup_v1_proc= [KNL] Show also missing controllers in /proc/cgroups
655 Format: { "true" | "false" }
656 /proc/cgroups lists only v1 controllers by default.
657 This compatibility option enables listing also v2
658 controllers (whose v1 code is not compiled!), so that
659 semi-legacy software can check this file to decide
660 about usage of v2 (sic) controllers.
661
662 cgroup_favordynmods= [KNL] Enable or Disable favordynmods.
663 Format: { "true" | "false" }
664 Defaults to the value of CONFIG_CGROUP_FAVOR_DYNMODS.
665
666 cgroup.memory= [KNL] Pass options to the cgroup memory controller.
667 Format: <string>
668 nosocket -- Disable socket memory accounting.
669 nokmem -- Disable kernel memory accounting.
670 nobpf -- Disable BPF memory accounting.
671
672 checkreqprot= [SELINUX] Set initial checkreqprot flag value.
673 Format: { "0" | "1" }
674 See security/selinux/Kconfig help text.
675 0 -- check protection applied by kernel (includes
676 any implied execute protection).
677 1 -- check protection requested by application.
678 Default value is set via a kernel config option.
679 Value can be changed at runtime via
680 /sys/fs/selinux/checkreqprot.
681 Setting checkreqprot to 1 is deprecated.
682
683 cio_ignore= [S390]
684 See Documentation/arch/s390/common_io.rst for details.
685
686 clearcpuid=X[,X...] [X86]
687 Disable CPUID feature X for the kernel. See
688 arch/x86/include/asm/cpufeatures.h for the valid bit
689 numbers X. Note the Linux-specific bits are not necessarily
690 stable over kernel options, but the vendor-specific
691 ones should be.
692 X can also be a string as appearing in the flags: line
693 in /proc/cpuinfo which does not have the above
694 instability issue. However, not all features have names
695 in /proc/cpuinfo.
696 Note that using this option will taint your kernel.
697 Also note that user programs calling CPUID directly
698 or using the feature without checking anything
699 will still see it. This just prevents it from
700 being used by the kernel or shown in /proc/cpuinfo.
701 Also note the kernel might malfunction if you disable
702 some critical bits.
703
704 clk_ignore_unused
705 [CLK]
706 Prevents the clock framework from automatically gating
707 clocks that have not been explicitly enabled by a Linux
708 device driver but are enabled in hardware at reset or
709 by the bootloader/firmware. Note that this does not
710 force such clocks to be always-on nor does it reserve
711 those clocks in any way. This parameter is useful for
712 debug and development, but should not be needed on a
713 platform with proper driver support. For more
714 information, see Documentation/driver-api/clk.rst.
715
716 clock= [BUGS=X86-32, HW] gettimeofday clocksource override.
717 [Deprecated]
718 Forces specified clocksource (if available) to be used
719 when calculating gettimeofday(). If specified
720 clocksource is not available, it defaults to PIT.
721 Format: { pit | tsc | cyclone | pmtmr }
722
723 clocksource= Override the default clocksource
724 Format: <string>
725 Override the default clocksource and use the clocksource
726 with the name specified.
727 Some clocksource names to choose from, depending on
728 the platform:
729 [all] jiffies (this is the base, fallback clocksource)
730 [ACPI] acpi_pm
731 [ARM] imx_timer1,OSTS,netx_timer,mpu_timer2,
732 pxa_timer,timer3,32k_counter,timer0_1
733 [X86-32] pit,hpet,tsc;
734 scx200_hrt on Geode; cyclone on IBM x440
735 [MIPS] MIPS
736 [PARISC] cr16
737 [S390] tod
738 [SH] SuperH
739 [SPARC64] tick
740 [X86-64] hpet,tsc
741
742 clocksource.arm_arch_timer.evtstrm=
743 [ARM,ARM64,EARLY]
744 Format: <bool>
745 Enable/disable the eventstream feature of the ARM
746 architected timer so that code using WFE-based polling
747 loops can be debugged more effectively on production
748 systems.
749
750 clocksource.verify_n_cpus= [KNL]
751 Limit the number of CPUs checked for clocksources
752 marked with CLOCK_SOURCE_VERIFY_PERCPU that
753 are marked unstable due to excessive skew.
754 A negative value says to check all CPUs, while
755 zero says not to check any. Values larger than
756 nr_cpu_ids are silently truncated to nr_cpu_ids.
757 The actual CPUs are chosen randomly, with
758 no replacement if the same CPU is chosen twice.
759
760 clocksource-wdtest.holdoff= [KNL]
761 Set the time in seconds that the clocksource
762 watchdog test waits before commencing its tests.
763 Defaults to zero when built as a module and to
764 10 seconds when built into the kernel.
765
766 cma=nn[MG]@[start[MG][-end[MG]]]
767 [KNL,CMA,EARLY]
768 Sets the size of kernel global memory area for
769 contiguous memory allocations and optionally the
770 placement constraint by the physical address range of
771 memory allocations. A value of 0 disables CMA
772 altogether. For more information, see
773 kernel/dma/contiguous.c
774
775 cma_pernuma=nn[MG]
776 [KNL,CMA,EARLY]
777 Sets the size of kernel per-numa memory area for
778 contiguous memory allocations. A value of 0 disables
779 per-numa CMA altogether. And If this option is not
780 specified, the default value is 0.
781 With per-numa CMA enabled, DMA users on node nid will
782 first try to allocate buffer from the pernuma area
783 which is located in node nid, if the allocation fails,
784 they will fallback to the global default memory area.
785
786 numa_cma=<node>:nn[MG][,<node>:nn[MG]]
787 [KNL,CMA,EARLY]
788 Sets the size of kernel numa memory area for
789 contiguous memory allocations. It will reserve CMA
790 area for the specified node.
791
792 With numa CMA enabled, DMA users on node nid will
793 first try to allocate buffer from the numa area
794 which is located in node nid, if the allocation fails,
795 they will fallback to the global default memory area.
796
797 cmo_free_hint= [PPC] Format: { yes | no }
798 Specify whether pages are marked as being inactive
799 when they are freed. This is used in CMO environments
800 to determine OS memory pressure for page stealing by
801 a hypervisor.
802 Default: yes
803
804 coherent_pool=nn[KMG] [ARM,KNL,EARLY]
805 Sets the size of memory pool for coherent, atomic dma
806 allocations, by default set to 256K.
807
808 com20020= [HW,NET] ARCnet - COM20020 chipset
809 Format:
810 <io>[,<irq>[,<nodeID>[,<backplane>[,<ckp>[,<timeout>]]]]]
811
812 com90io= [HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers)
813 Format: <io>[,<irq>]
814
815 com90xx= [HW,NET]
816 ARCnet - COM90xx chipset (memory-mapped buffers)
817 Format: <io>[,<irq>[,<memstart>]]
818
819 condev= [HW,S390] console device
820 conmode=
821
822 con3215_drop= [S390,EARLY] 3215 console drop mode.
823 Format: y|n|Y|N|1|0
824 When set to true, drop data on the 3215 console when
825 the console buffer is full. In this case the
826 operator using a 3270 terminal emulator (for example
827 x3270) does not have to enter the clear key for the
828 console output to advance and the kernel to continue.
829 This leads to a much faster boot time when a 3270
830 terminal emulator is active. If no 3270 terminal
831 emulator is used, this parameter has no effect.
832
833 console= [KNL] Output console device and options.
834
835 tty<n> Use the virtual console device <n>.
836
837 ttyS<n>[,options]
838 ttyUSB0[,options]
839 Use the specified serial port. The options are of
840 the form "bbbbpnf", where "bbbb" is the baud rate,
841 "p" is parity ("n", "o", or "e"), "n" is number of
842 bits, and "f" is flow control ("r" for RTS or
843 omit it). Default is "9600n8".
844
845 See Documentation/admin-guide/serial-console.rst for more
846 information. See
847 Documentation/networking/netconsole.rst for an
848 alternative.
849
850 <DEVNAME>:<n>.<n>[,options]
851 Use the specified serial port on the serial core bus.
852 The addressing uses DEVNAME of the physical serial port
853 device, followed by the serial core controller instance,
854 and the serial port instance. The options are the same
855 as documented for the ttyS addressing above.
856
857 The mapping of the serial ports to the tty instances
858 can be viewed with:
859
860 $ ls -d /sys/bus/serial-base/devices/*:*.*/tty/*
861 /sys/bus/serial-base/devices/00:04:0.0/tty/ttyS0
862
863 In the above example, the console can be addressed with
864 console=00:04:0.0. Note that a console addressed this
865 way will only get added when the related device driver
866 is ready. The use of an earlycon parameter in addition to
867 the console may be desired for console output early on.
868
869 uart[8250],io,<addr>[,options]
870 uart[8250],mmio,<addr>[,options]
871 uart[8250],mmio16,<addr>[,options]
872 uart[8250],mmio32,<addr>[,options]
873 uart[8250],0x<addr>[,options]
874 Start an early, polled-mode console on the 8250/16550
875 UART at the specified I/O port or MMIO address,
876 switching to the matching ttyS device later.
877 MMIO inter-register address stride is either 8-bit
878 (mmio), 16-bit (mmio16), or 32-bit (mmio32).
879 If none of [io|mmio|mmio16|mmio32], <addr> is assumed
880 to be equivalent to 'mmio'. 'options' are specified in
881 the same format described for ttyS above; if unspecified,
882 the h/w is not re-initialized.
883
884 hvc<n> Use the hypervisor console device <n>. This is for
885 both Xen and PowerPC hypervisors.
886
887 { null | "" }
888 Use to disable console output, i.e., to have kernel
889 console messages discarded.
890 This must be the only console= parameter used on the
891 kernel command line.
892
893 If the device connected to the port is not a TTY but a braille
894 device, prepend "brl," before the device type, for instance
895 console=brl,ttyS0
896 For now, only VisioBraille is supported.
897
898 console_msg_format=
899 [KNL] Change console messages format
900 default
901 By default we print messages on consoles in
902 "[time stamp] text\n" format (time stamp may not be
903 printed, depending on CONFIG_PRINTK_TIME or
904 `printk_time' param).
905 syslog
906 Switch to syslog format: "<%u>[time stamp] text\n"
907 IOW, each message will have a facility and loglevel
908 prefix. The format is similar to one used by syslog()
909 syscall, or to executing "dmesg -S --raw" or to reading
910 from /proc/kmsg.
911
912 consoleblank= [KNL] The console blank (screen saver) timeout in
913 seconds. A value of 0 disables the blank timer.
914 Defaults to 0.
915
916 coredump_filter=
917 [KNL] Change the default value for
918 /proc/<pid>/coredump_filter.
919 See also Documentation/filesystems/proc.rst.
920
921 coresight_cpu_debug.enable
922 [ARM,ARM64]
923 Format: <bool>
924 Enable/disable the CPU sampling based debugging.
925 0: default value, disable debugging
926 1: enable debugging at boot time
927
928 cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver
929 Format:
930 <first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>]
931
932 cpuidle.off=1 [CPU_IDLE]
933 disable the cpuidle sub-system
934
935 cpuidle.governor=
936 [CPU_IDLE] Name of the cpuidle governor to use.
937
938 cpufreq.off=1 [CPU_FREQ]
939 disable the cpufreq sub-system
940
941 cpufreq.default_governor=
942 [CPU_FREQ] Name of the default cpufreq governor or
943 policy to use. This governor must be registered in the
944 kernel before the cpufreq driver probes.
945
946 cpu_init_udelay=N
947 [X86,EARLY] Delay for N microsec between assert and de-assert
948 of APIC INIT to start processors. This delay occurs
949 on every CPU online, such as boot, and resume from suspend.
950 Default: 10000
951
952 cpuhp.parallel=
953 [SMP] Enable/disable parallel bringup of secondary CPUs
954 Format: <bool>
955 Default is enabled if CONFIG_HOTPLUG_PARALLEL=y. Otherwise
956 the parameter has no effect.
957
958 crash_kexec_post_notifiers
959 Only jump to kdump kernel after running the panic
960 notifiers and dumping kmsg. This option increases
961 the risks of a kdump failure, since some panic
962 notifiers can make the crashed kernel more unstable.
963 In configurations where kdump may not be reliable,
964 running the panic notifiers could allow collecting
965 more data on dmesg, like stack traces from other CPUS
966 or extra data dumped by panic_print. Note that some
967 configurations enable this option unconditionally,
968 like Hyper-V, PowerPC (fadump) and AMD SEV-SNP.
969
970 crashkernel=size[KMG][@offset[KMG]]
971 [KNL,EARLY] Using kexec, Linux can switch to a 'crash kernel'
972 upon panic. This parameter reserves the physical
973 memory region [offset, offset + size] for that kernel
974 image. If '@offset' is omitted, then a suitable offset
975 is selected automatically.
976 [KNL, X86-64, ARM64, RISCV, LoongArch] Select a region
977 under 4G first, and fall back to reserve region above
978 4G when '@offset' hasn't been specified.
979 See Documentation/admin-guide/kdump/kdump.rst for further details.
980
981 crashkernel=range1:size1[,range2:size2,...][@offset]
982 [KNL] Same as above, but depends on the memory
983 in the running system. The syntax of range is
984 start-[end] where start and end are both
985 a memory unit (amount[KMG]). See also
986 Documentation/admin-guide/kdump/kdump.rst for an example.
987
988 crashkernel=size[KMG],high
989 [KNL, X86-64, ARM64, RISCV, LoongArch] range could be
990 above 4G.
991 Allow kernel to allocate physical memory region from top,
992 so could be above 4G if system have more than 4G ram
993 installed. Otherwise memory region will be allocated
994 below 4G, if available.
995 It will be ignored if crashkernel=X is specified.
996 crashkernel=size[KMG],low
997 [KNL, X86-64, ARM64, RISCV, LoongArch] range under 4G.
998 When crashkernel=X,high is passed, kernel could allocate
999 physical memory region above 4G, that cause second kernel
1000 crash on system that require some amount of low memory,
1001 e.g. swiotlb requires at least 64M+32K low memory, also
1002 enough extra low memory is needed to make sure DMA buffers
1003 for 32-bit devices won't run out. Kernel would try to allocate
1004 default size of memory below 4G automatically. The default
1005 size is platform dependent.
1006 --> x86: max(swiotlb_size_or_default() + 8MiB, 256MiB)
1007 --> arm64: 128MiB
1008 --> riscv: 128MiB
1009 --> loongarch: 128MiB
1010 This one lets the user specify own low range under 4G
1011 for second kernel instead.
1012 0: to disable low allocation.
1013 It will be ignored when crashkernel=X,high is not used
1014 or memory reserved is below 4G.
1015 crashkernel=size[KMG],cma
1016 [KNL, X86] Reserve additional crash kernel memory from
1017 CMA. This reservation is usable by the first system's
1018 userspace memory and kernel movable allocations (memory
1019 balloon, zswap). Pages allocated from this memory range
1020 will not be included in the vmcore so this should not
1021 be used if dumping of userspace memory is intended and
1022 it has to be expected that some movable kernel pages
1023 may be missing from the dump.
1025 A standard crashkernel reservation, as described above,
1026 is still needed to hold the crash kernel and initrd.
1028 This option increases the risk of a kdump failure: DMA
1029 transfers configured by the first kernel may end up
1030 corrupting the second kernel's memory.
1032 This reservation method is intended for systems that
1033 can't afford to sacrifice enough memory for standard
1034 crashkernel reservation and where less reliable and
1035 possibly incomplete kdump is preferable to no kdump at
1036 all.
1038 cryptomgr.notests
1039 [KNL] Disable crypto self-tests
1041 cs89x0_dma= [HW,NET]
1042 Format: <dma>
1044 cs89x0_media= [HW,NET]
1045 Format: { rj45 | aui | bnc }
1047 csdlock_debug= [KNL] Enable or disable debug add-ons of cross-CPU
1048 function call handling. When switched on,
1049 additional debug data is printed to the console
1050 in case a hanging CPU is detected, and that
1051 CPU is pinged again in order to try to resolve
1052 the hang situation. The default value of this
1053 option depends on the CSD_LOCK_WAIT_DEBUG_DEFAULT
1054 Kconfig option.
1056 dasd= [HW,NET]
1057 See header of drivers/s390/block/dasd_devmap.c.
1059 db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port
1060 (one device per port)
1061 Format: <port#>,<type>
1062 See also Documentation/input/devices/joystick-parport.rst
1064 debug [KNL,EARLY] Enable kernel debugging (events log level).
1066 debug_boot_weak_hash
1067 [KNL,EARLY] Enable printing [hashed] pointers early in the
1068 boot sequence. If enabled, we use a weak hash instead
1069 of siphash to hash pointers. Use this option if you are
1070 seeing instances of '(___ptrval___)') and need to see a
1071 value (hashed pointer) instead. Cryptographically
1072 insecure, please do not use on production kernels.
1074 debug_locks_verbose=
1075 [KNL] verbose locking self-tests
1076 Format: <int>
1077 Print debugging info while doing the locking API
1078 self-tests.
1079 Bitmask for the various LOCKTYPE_ tests. Defaults to 0
1080 (no extra messages), setting it to -1 (all bits set)
1081 will print _a_lot_ more information - normally only
1082 useful to lockdep developers.
1084 debug_objects [KNL,EARLY] Enable object debugging
1086 debug_guardpage_minorder=
1087 [KNL,EARLY] When CONFIG_DEBUG_PAGEALLOC is set, this
1088 parameter allows control of the order of pages that will
1089 be intentionally kept free (and hence protected) by the
1090 buddy allocator. Bigger value increase the probability
1091 of catching random memory corruption, but reduce the
1092 amount of memory for normal system use. The maximum
1093 possible value is MAX_PAGE_ORDER/2. Setting this
1094 parameter to 1 or 2 should be enough to identify most
1095 random memory corruption problems caused by bugs in
1096 kernel or driver code when a CPU writes to (or reads
1097 from) a random memory location. Note that there exists
1098 a class of memory corruptions problems caused by buggy
1099 H/W or F/W or by drivers badly programming DMA
1100 (basically when memory is written at bus level and the
1101 CPU MMU is bypassed) which are not detectable by
1102 CONFIG_DEBUG_PAGEALLOC, hence this option will not
1103 help tracking down these problems.
1105 debug_pagealloc=
1106 [KNL,EARLY] When CONFIG_DEBUG_PAGEALLOC is set, this parameter
1107 enables the feature at boot time. By default, it is
1108 disabled and the system will work mostly the same as a
1109 kernel built without CONFIG_DEBUG_PAGEALLOC.
1110 Note: to get most of debug_pagealloc error reports, it's
1111 useful to also enable the page_owner functionality.
1112 on: enable the feature
1114 debugfs= [KNL,EARLY] This parameter enables what is exposed to
1115 userspace and debugfs internal clients.
1116 Format: { on, no-mount, off }
1117 on: All functions are enabled.
1118 no-mount:
1119 Filesystem is not registered but kernel clients can
1120 access APIs and a crashkernel can be used to read
1121 its content. There is nothing to mount.
1122 off: Filesystem is not registered and clients
1123 get a -EPERM as result when trying to register files
1124 or directories within debugfs.
1125 This is equivalent of the runtime functionality if
1126 debugfs was not enabled in the kernel at all.
1127 Default value is set in build-time with a kernel configuration.
1129 debugpat [X86] Enable PAT debugging
1131 default_hugepagesz=
1132 [HW] The size of the default HugeTLB page. This is
1133 the size represented by the legacy /proc/ hugepages
1134 APIs. In addition, this is the default hugetlb size
1135 used for shmget(), mmap() and mounting hugetlbfs
1136 filesystems. If not specified, defaults to the
1137 architecture's default huge page size. Huge page
1138 sizes are architecture dependent. See also
1139 Documentation/admin-guide/mm/hugetlbpage.rst.
1140 Format: size[KMG]
1142 deferred_probe_timeout=
1143 [KNL] Debugging option to set a timeout in seconds for
1144 deferred probe to give up waiting on dependencies to
1145 probe. Only specific dependencies (subsystems or
1146 drivers) that have opted in will be ignored. A timeout
1147 of 0 will timeout at the end of initcalls. If the time
1148 out hasn't expired, it'll be restarted by each
1149 successful driver registration. This option will also
1150 dump out devices still on the deferred probe list after
1151 retrying.
1153 delayacct [KNL] Enable per-task delay accounting
1155 dell_smm_hwmon.ignore_dmi=
1156 [HW] Continue probing hardware even if DMI data
1157 indicates that the driver is running on unsupported
1158 hardware.
1160 dell_smm_hwmon.force=
1161 [HW] Activate driver even if SMM BIOS signature does
1162 not match list of supported models and enable otherwise
1163 blacklisted features.
1165 dell_smm_hwmon.power_status=
1166 [HW] Report power status in /proc/i8k
1167 (disabled by default).
1169 dell_smm_hwmon.restricted=
1170 [HW] Allow controlling fans only if SYS_ADMIN
1171 capability is set.
1173 dell_smm_hwmon.fan_mult=
1174 [HW] Factor to multiply fan speed with.
1176 dell_smm_hwmon.fan_max=
1177 [HW] Maximum configurable fan speed.
1179 dfltcc= [HW,S390]
1180 Format: { on | off | def_only | inf_only | always }
1181 on: s390 zlib hardware support for compression on
1182 level 1 and decompression (default)
1183 off: No s390 zlib hardware support
1184 def_only: s390 zlib hardware support for deflate
1185 only (compression on level 1)
1186 inf_only: s390 zlib hardware support for inflate
1187 only (decompression)
1188 always: Same as 'on' but ignores the selected compression
1189 level always using hardware support (used for debugging)
1191 dhash_entries= [KNL]
1192 Set number of hash buckets for dentry cache.
1194 disable_1tb_segments [PPC,EARLY]
1195 Disables the use of 1TB hash page table segments. This
1196 causes the kernel to fall back to 256MB segments which
1197 can be useful when debugging issues that require an SLB
1198 miss to occur.
1200 disable= [IPV6]
1201 See Documentation/networking/ipv6.rst.
1203 disable_radix [PPC,EARLY]
1204 Disable RADIX MMU mode on POWER9
1206 disable_tlbie [PPC]
1207 Disable TLBIE instruction. Currently does not work
1208 with KVM, with HASH MMU, or with coherent accelerators.
1210 disable_ddw [PPC/PSERIES,EARLY]
1211 Disable Dynamic DMA Window support. Use this
1212 to workaround buggy firmware.
1214 disable_ipv6= [IPV6]
1215 See Documentation/networking/ipv6.rst.
1217 disable_mtrr_cleanup [X86,EARLY]
1218 The kernel tries to adjust MTRR layout from continuous
1219 to discrete, to make X server driver able to add WB
1220 entry later. This parameter disables that.
1222 disable_mtrr_trim [X86, Intel and AMD only,EARLY]
1223 By default the kernel will trim any uncacheable
1224 memory out of your available memory pool based on
1225 MTRR settings. This parameter disables that behavior,
1226 possibly causing your machine to run very slowly.
1228 disable_timer_pin_1 [X86,EARLY]
1229 Disable PIN 1 of APIC timer
1230 Can be useful to work around chipset bugs.
1232 dis_ucode_ldr [X86] Disable the microcode loader.
1234 dma_debug=off If the kernel is compiled with DMA_API_DEBUG support,
1235 this option disables the debugging code at boot.
1237 dma_debug_entries=<number>
1238 This option allows to tune the number of preallocated
1239 entries for DMA-API debugging code. One entry is
1240 required per DMA-API allocation. Use this if the
1241 DMA-API debugging code disables itself because the
1242 architectural default is too low.
1244 dma_debug_driver=<driver_name>
1245 With this option the DMA-API debugging driver
1246 filter feature can be enabled at boot time. Just
1247 pass the driver to filter for as the parameter.
1248 The filter can be disabled or changed to another
1249 driver later using sysfs.
1251 reg_file_data_sampling=
1252 [X86] Controls mitigation for Register File Data
1253 Sampling (RFDS) vulnerability. RFDS is a CPU
1254 vulnerability which may allow userspace to infer
1255 kernel data values previously stored in floating point
1256 registers, vector registers, or integer registers.
1257 RFDS only affects Intel Atom processors.
1259 on: Turns ON the mitigation.
1260 off: Turns OFF the mitigation.
1262 This parameter overrides the compile time default set
1263 by CONFIG_MITIGATION_RFDS. Mitigation cannot be
1264 disabled when other VERW based mitigations (like MDS)
1265 are enabled. In order to disable RFDS mitigation all
1266 VERW based mitigations need to be disabled.
1268 For details see:
1269 Documentation/admin-guide/hw-vuln/reg-file-data-sampling.rst
1271 driver_async_probe= [KNL]
1272 List of driver names to be probed asynchronously. *
1273 matches with all driver names. If * is specified, the
1274 rest of the listed driver names are those that will NOT
1275 match the *.
1276 Format: <driver_name1>,<driver_name2>...
1278 drm.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>]
1279 Broken monitors, graphic adapters, KVMs and EDIDless
1280 panels may send no or incorrect EDID data sets.
1281 This parameter allows to specify an EDID data sets
1282 in the /lib/firmware directory that are used instead.
1283 An EDID data set will only be used for a particular
1284 connector, if its name and a colon are prepended to
1285 the EDID name. Each connector may use a unique EDID
1286 data set by separating the files with a comma. An EDID
1287 data set with no connector name will be used for
1288 any connectors not explicitly specified.
1290 dscc4.setup= [NET]
1292 dt_cpu_ftrs= [PPC,EARLY]
1293 Format: {"off" | "known"}
1294 Control how the dt_cpu_ftrs device-tree binding is
1295 used for CPU feature discovery and setup (if it
1296 exists).
1297 off: Do not use it, fall back to legacy cpu table.
1298 known: Do not pass through unknown features to guests
1299 or userspace, only those that the kernel is aware of.
1301 dump_apple_properties [X86]
1302 Dump name and content of EFI device properties on
1303 x86 Macs. Useful for driver authors to determine
1304 what data is available or for reverse-engineering.
1306 dyndbg[="val"] [KNL,DYNAMIC_DEBUG]
1307 <module>.dyndbg[="val"]
1308 Enable debug messages at boot time. See
1309 Documentation/admin-guide/dynamic-debug-howto.rst
1310 for details.
1312 early_ioremap_debug [KNL,EARLY]
1313 Enable debug messages in early_ioremap support. This
1314 is useful for tracking down temporary early mappings
1315 which are not unmapped.
1317 earlycon= [KNL,EARLY] Output early console device and options.
1319 When used with no options, the early console is
1320 determined by stdout-path property in device tree's
1321 chosen node or the ACPI SPCR table if supported by
1322 the platform.
1324 cdns,<addr>[,options]
1325 Start an early, polled-mode console on a Cadence
1326 (xuartps) serial port at the specified address. Only
1327 supported option is baud rate. If baud rate is not
1328 specified, the serial port must already be setup and
1329 configured.
1331 uart[8250],io,<addr>[,options[,uartclk]]
1332 uart[8250],mmio,<addr>[,options[,uartclk]]
1333 uart[8250],mmio32,<addr>[,options[,uartclk]]
1334 uart[8250],mmio32be,<addr>[,options[,uartclk]]
1335 uart[8250],0x<addr>[,options]
1336 Start an early, polled-mode console on the 8250/16550
1337 UART at the specified I/O port or MMIO address.
1338 MMIO inter-register address stride is either 8-bit
1339 (mmio) or 32-bit (mmio32 or mmio32be).
1340 If none of [io|mmio|mmio32|mmio32be], <addr> is assumed
1341 to be equivalent to 'mmio'. 'options' are specified
1342 in the same format described for "console=ttyS<n>"; if
1343 unspecified, the h/w is not initialized. 'uartclk' is
1344 the uart clock frequency; if unspecified, it is set
1345 to 'BASE_BAUD' * 16.
1347 pl011,<addr>
1348 pl011,mmio32,<addr>
1349 Start an early, polled-mode console on a pl011 serial
1350 port at the specified address. The pl011 serial port
1351 must already be setup and configured. Options are not
1352 yet supported. If 'mmio32' is specified, then only
1353 the driver will use only 32-bit accessors to read/write
1354 the device registers.
1356 liteuart,<addr>
1357 Start an early console on a litex serial port at the
1358 specified address. The serial port must already be
1359 setup and configured. Options are not yet supported.
1361 meson,<addr>
1362 Start an early, polled-mode console on a meson serial
1363 port at the specified address. The serial port must
1364 already be setup and configured. Options are not yet
1365 supported.
1367 msm_serial,<addr>
1368 Start an early, polled-mode console on an msm serial
1369 port at the specified address. The serial port
1370 must already be setup and configured. Options are not
1371 yet supported.
1373 msm_serial_dm,<addr>
1374 Start an early, polled-mode console on an msm serial
1375 dm port at the specified address. The serial port
1376 must already be setup and configured. Options are not
1377 yet supported.
1379 owl,<addr>
1380 Start an early, polled-mode console on a serial port
1381 of an Actions Semi SoC, such as S500 or S900, at the
1382 specified address. The serial port must already be
1383 setup and configured. Options are not yet supported.
1385 rda,<addr>
1386 Start an early, polled-mode console on a serial port
1387 of an RDA Micro SoC, such as RDA8810PL, at the
1388 specified address. The serial port must already be
1389 setup and configured. Options are not yet supported.
1391 sbi
1392 Use RISC-V SBI (Supervisor Binary Interface) for early
1393 console.
1395 smh Use ARM semihosting calls for early console.
1397 s3c2410,<addr>
1398 s3c2412,<addr>
1399 s3c2440,<addr>
1400 s3c6400,<addr>
1401 s5pv210,<addr>
1402 exynos4210,<addr>
1403 Use early console provided by serial driver available
1404 on Samsung SoCs, requires selecting proper type and
1405 a correct base address of the selected UART port. The
1406 serial port must already be setup and configured.
1407 Options are not yet supported.
1409 lantiq,<addr>
1410 Start an early, polled-mode console on a lantiq serial
1411 (lqasc) port at the specified address. The serial port
1412 must already be setup and configured. Options are not
1413 yet supported.
1415 lpuart,<addr>
1416 lpuart32,<addr>
1417 Use early console provided by Freescale LP UART driver
1418 found on Freescale Vybrid and QorIQ LS1021A processors.
1419 A valid base address must be provided, and the serial
1420 port must already be setup and configured.
1422 ec_imx21,<addr>
1423 ec_imx6q,<addr>
1424 Start an early, polled-mode, output-only console on the
1425 Freescale i.MX UART at the specified address. The UART
1426 must already be setup and configured.
1428 ar3700_uart,<addr>
1429 Start an early, polled-mode console on the
1430 Armada 3700 serial port at the specified
1431 address. The serial port must already be setup
1432 and configured. Options are not yet supported.
1434 qcom_geni,<addr>
1435 Start an early, polled-mode console on a Qualcomm
1436 Generic Interface (GENI) based serial port at the
1437 specified address. The serial port must already be
1438 setup and configured. Options are not yet supported.
1440 efifb,[options]
1441 Start an early, unaccelerated console on the EFI
1442 memory mapped framebuffer (if available). On cache
1443 coherent non-x86 systems that use system memory for
1444 the framebuffer, pass the 'ram' option so that it is
1445 mapped with the correct attributes.
1447 linflex,<addr>
1448 Use early console provided by Freescale LINFlexD UART
1449 serial driver for NXP S32V234 SoCs. A valid base
1450 address must be provided, and the serial port must
1451 already be setup and configured.
1453 earlyprintk= [X86,SH,ARM,M68k,S390,UM,EARLY]
1454 earlyprintk=vga
1455 earlyprintk=sclp
1456 earlyprintk=xen
1457 earlyprintk=serial[,ttySn[,baudrate]]
1458 earlyprintk=serial[,0x...[,baudrate]]
1459 earlyprintk=ttySn[,baudrate]
1460 earlyprintk=dbgp[debugController#]
1461 earlyprintk=mmio32,membase[,{nocfg|baudrate}]
1462 earlyprintk=pciserial[,force],bus:device.function[,{nocfg|baudrate}]
1463 earlyprintk=xdbc[xhciController#]
1464 earlyprintk=bios
1466 earlyprintk is useful when the kernel crashes before
1467 the normal console is initialized. It is not enabled by
1468 default because it has some cosmetic problems.
1470 Use "nocfg" to skip UART configuration, assume
1471 BIOS/firmware has configured UART correctly.
1473 Append ",keep" to not disable it when the real console
1474 takes over.
1476 Only one of vga, serial, or usb debug port can
1477 be used at a time.
1479 Currently only ttyS0 and ttyS1 may be specified by
1480 name. Other I/O ports may be explicitly specified
1481 on some architectures (x86 and arm at least) by
1482 replacing ttySn with an I/O port address, like this:
1483 earlyprintk=serial,0x1008,115200
1484 You can find the port for a given device in
1485 /proc/tty/driver/serial:
1486 2: uart:ST16650V2 port:00001008 irq:18 ...
1488 Interaction with the standard serial driver is not
1489 very good.
1491 The VGA output is eventually overwritten by
1492 the real console.
1494 The xen option can only be used in Xen domains.
1496 The sclp output can only be used on s390.
1498 The bios output can only be used on SuperH.
1500 The optional "force" to "pciserial" enables use of a
1501 PCI device even when its classcode is not of the
1502 UART class.
1504 edac_report= [HW,EDAC] Control how to report EDAC event
1505 Format: {"on" | "off" | "force"}
1506 on: enable EDAC to report H/W event. May be overridden
1507 by other higher priority error reporting module.
1508 off: disable H/W event reporting through EDAC.
1509 force: enforce the use of EDAC to report H/W event.
1510 default: on.
1512 edd= [EDD]
1513 Format: {"off" | "on" | "skip[mbr]"}
1515 efi= [EFI,EARLY]
1516 Format: { "debug", "disable_early_pci_dma",
1517 "nochunk", "noruntime", "nosoftreserve",
1518 "novamap", "no_disable_early_pci_dma" }
1519 debug: enable misc debug output.
1520 disable_early_pci_dma: disable the busmaster bit on all
1521 PCI bridges while in the EFI boot stub.
1522 nochunk: disable reading files in "chunks" in the EFI
1523 boot stub, as chunking can cause problems with some
1524 firmware implementations.
1525 noruntime : disable EFI runtime services support
1526 nosoftreserve: The EFI_MEMORY_SP (Specific Purpose)
1527 attribute may cause the kernel to reserve the
1528 memory range for a memory mapping driver to
1529 claim. Specify efi=nosoftreserve to disable this
1530 reservation and treat the memory by its base type
1531 (i.e. EFI_CONVENTIONAL_MEMORY / "System RAM").
1532 novamap: do not call SetVirtualAddressMap().
1533 no_disable_early_pci_dma: Leave the busmaster bit set
1534 on all PCI bridges while in the EFI boot stub
1536 efi_no_storage_paranoia [EFI,X86,EARLY]
1537 Using this parameter you can use more than 50% of
1538 your efi variable storage. Use this parameter only if
1539 you are really sure that your UEFI does sane gc and
1540 fulfills the spec otherwise your board may brick.
1542 efivar_ssdt= [EFI; X86] Name of an EFI variable that contains an SSDT
1543 that is to be dynamically loaded by Linux. If there are
1544 multiple variables with the same name but with different
1545 vendor GUIDs, all of them will be loaded. See
1546 Documentation/admin-guide/acpi/ssdt-overlays.rst for details.
1549 eisa_irq_edge= [PARISC,HW]
1550 See header of drivers/parisc/eisa.c.
1552 ekgdboc= [X86,KGDB,EARLY] Allow early kernel console debugging
1553 Format: ekgdboc=kbd
1555 This is designed to be used in conjunction with
1556 the boot argument: earlyprintk=vga
1558 This parameter works in place of the kgdboc parameter
1559 but can only be used if the backing tty is available
1560 very early in the boot process. For early debugging
1561 via a serial port see kgdboc_earlycon instead.
1563 elanfreq= [X86-32]
1564 See comment before function elanfreq_setup() in
1565 arch/x86/kernel/cpu/cpufreq/elanfreq.c.
1567 elfcorehdr=[size[KMG]@]offset[KMG] [PPC,SH,X86,S390,EARLY]
1568 Specifies physical address of start of kernel core
1569 image elf header and optionally the size. Generally
1570 kexec loader will pass this option to capture kernel.
1571 See Documentation/admin-guide/kdump/kdump.rst for details.
1573 enable_mtrr_cleanup [X86,EARLY]
1574 The kernel tries to adjust MTRR layout from continuous
1575 to discrete, to make X server driver able to add WB
1576 entry later. This parameter enables that.
1578 enable_timer_pin_1 [X86]
1579 Enable PIN 1 of APIC timer
1580 Can be useful to work around chipset bugs
1581 (in particular on some ATI chipsets).
1582 The kernel tries to set a reasonable default.
1584 enforcing= [SELINUX] Set initial enforcing status.
1585 Format: {"0" | "1"}
1586 See security/selinux/Kconfig help text.
1587 0 -- permissive (log only, no denials).
1588 1 -- enforcing (deny and log).
1589 Default value is 0.
1590 Value can be changed at runtime via
1591 /sys/fs/selinux/enforce.
1593 erst_disable [ACPI]
1594 Disable Error Record Serialization Table (ERST)
1595 support.
1597 ether= [HW,NET] Ethernet cards parameters
1598 This option is obsoleted by the "netdev=" option, which
1599 has equivalent usage. See its documentation for details.
1601 evm= [EVM]
1602 Format: { "fix" }
1603 Permit 'security.evm' to be updated regardless of
1604 current integrity status.
1606 early_page_ext [KNL,EARLY] Enforces page_ext initialization to earlier
1607 stages so cover more early boot allocations.
1608 Please note that as side effect some optimizations
1609 might be disabled to achieve that (e.g. parallelized
1610 memory initialization is disabled) so the boot process
1611 might take longer, especially on systems with a lot of
1612 memory. Available with CONFIG_PAGE_EXTENSION=y.
1614 failslab=
1615 fail_usercopy=
1616 fail_page_alloc=
1617 fail_skb_realloc=
1618 fail_make_request=[KNL]
1619 General fault injection mechanism.
1620 Format: <interval>,<probability>,<space>,<times>
1621 See also Documentation/fault-injection/.
1623 fb_tunnels= [NET]
1624 Format: { initns | none }
1625 See Documentation/admin-guide/sysctl/net.rst for
1626 fb_tunnels_only_for_init_ns
1628 floppy= [HW]
1629 See Documentation/admin-guide/blockdev/floppy.rst.
1631 forcepae [X86-32]
1632 Forcefully enable Physical Address Extension (PAE).
1633 Many Pentium M systems disable PAE but may have a
1634 functionally usable PAE implementation.
1635 Warning: use of this parameter will taint the kernel
1636 and may cause unknown problems.
1638 fred= [X86-64]
1639 Enable/disable Flexible Return and Event Delivery.
1640 Format: { on | off }
1641 on: enable FRED when it's present.
1642 off: disable FRED, the default setting.
1644 ftrace=[tracer]
1645 [FTRACE] will set and start the specified tracer
1646 as early as possible in order to facilitate early
1647 boot debugging.
1649 ftrace_boot_snapshot
1650 [FTRACE] On boot up, a snapshot will be taken of the
1651 ftrace ring buffer that can be read at:
1652 /sys/kernel/tracing/snapshot.
1653 This is useful if you need tracing information from kernel
1654 boot up that is likely to be overridden by user space
1655 start up functionality.
1657 Optionally, the snapshot can also be defined for a tracing
1658 instance that was created by the trace_instance= command
1659 line parameter.
1661 trace_instance=foo,sched_switch ftrace_boot_snapshot=foo
1663 The above will cause the "foo" tracing instance to trigger
1664 a snapshot at the end of boot up.
1666 ftrace_dump_on_oops[=2(orig_cpu) | =<instance>][,<instance> |
1667 ,<instance>=2(orig_cpu)]
1668 [FTRACE] will dump the trace buffers on oops.
1669 If no parameter is passed, ftrace will dump global
1670 buffers of all CPUs, if you pass 2 or orig_cpu, it
1671 will dump only the buffer of the CPU that triggered
1672 the oops, or the specific instance will be dumped if
1673 its name is passed. Multiple instance dump is also
1674 supported, and instances are separated by commas. Each
1675 instance supports only dump on CPU that triggered the
1676 oops by passing 2 or orig_cpu to it.
1678 ftrace_dump_on_oops=foo=orig_cpu
1680 The above will dump only the buffer of "foo" instance
1681 on CPU that triggered the oops.
1683 ftrace_dump_on_oops,foo,bar=orig_cpu
1685 The above will dump global buffer on all CPUs, the
1686 buffer of "foo" instance on all CPUs and the buffer
1687 of "bar" instance on CPU that triggered the oops.
1689 ftrace_filter=[function-list]
1690 [FTRACE] Limit the functions traced by the function
1691 tracer at boot up. function-list is a comma-separated
1692 list of functions. This list can be changed at run
1693 time by the set_ftrace_filter file in the debugfs
1694 tracing directory.
1696 ftrace_notrace=[function-list]
1697 [FTRACE] Do not trace the functions specified in
1698 function-list. This list can be changed at run time
1699 by the set_ftrace_notrace file in the debugfs
1700 tracing directory.
1702 ftrace_graph_filter=[function-list]
1703 [FTRACE] Limit the top level callers functions traced
1704 by the function graph tracer at boot up.
1705 function-list is a comma-separated list of functions
1706 that can be changed at run time by the
1707 set_graph_function file in the debugfs tracing directory.
1709 ftrace_graph_notrace=[function-list]
1710 [FTRACE] Do not trace from the functions specified in
1711 function-list. This list is a comma-separated list of
1712 functions that can be changed at run time by the
1713 set_graph_notrace file in the debugfs tracing directory.
1715 ftrace_graph_max_depth=<uint>
1716 [FTRACE] Used with the function graph tracer. This is
1717 the max depth it will trace into a function. This value
1718 can be changed at run time by the max_graph_depth file
1719 in the tracefs tracing directory. default: 0 (no limit)
1721 fw_devlink= [KNL,EARLY] Create device links between consumer and supplier
1722 devices by scanning the firmware to infer the
1723 consumer/supplier relationships. This feature is
1724 especially useful when drivers are loaded as modules as
1725 it ensures proper ordering of tasks like device probing
1726 (suppliers first, then consumers), supplier boot state
1727 clean up (only after all consumers have probed),
1728 suspend/resume & runtime PM (consumers first, then
1729 suppliers).
1730 Format: { off | permissive | on | rpm }
1731 off -- Don't create device links from firmware info.
1732 permissive -- Create device links from firmware info
1733 but use it only for ordering boot state clean
1734 up (sync_state() calls).
1735 on -- Create device links from firmware info and use it
1736 to enforce probe and suspend/resume ordering.
1737 rpm -- Like "on", but also use to order runtime PM.
1739 fw_devlink.strict=<bool>
1740 [KNL,EARLY] Treat all inferred dependencies as mandatory
1741 dependencies. This only applies for fw_devlink=on|rpm.
1742 Format: <bool>
1744 fw_devlink.sync_state =
1745 [KNL,EARLY] When all devices that could probe have finished
1746 probing, this parameter controls what to do with
1747 devices that haven't yet received their sync_state()
1748 calls.
1749 Format: { strict | timeout }
1750 strict -- Default. Continue waiting on consumers to
1751 probe successfully.
1752 timeout -- Give up waiting on consumers and call
1753 sync_state() on any devices that haven't yet
1754 received their sync_state() calls after
1755 deferred_probe_timeout has expired or by
1756 late_initcall() if !CONFIG_MODULES.
1758 gamecon.map[2|3]=
1759 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad
1760 support via parallel port (up to 5 devices per port)
1761 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5>
1762 See also Documentation/input/devices/joystick-parport.rst
1764 gamma= [HW,DRM]
1766 gart_fix_e820= [X86-64,EARLY] disable the fix e820 for K8 GART
1767 Format: off | on
1768 default: on
1770 gather_data_sampling=
1771 [X86,INTEL,EARLY] Control the Gather Data Sampling (GDS)
1772 mitigation.
1774 Gather Data Sampling is a hardware vulnerability which
1775 allows unprivileged speculative access to data which was
1776 previously stored in vector registers.
1778 This issue is mitigated by default in updated microcode.
1779 The mitigation may have a performance impact but can be
1780 disabled. On systems without the microcode mitigation
1781 disabling AVX serves as a mitigation.
1783 force: Disable AVX to mitigate systems without
1784 microcode mitigation. No effect if the microcode
1785 mitigation is present. Known to cause crashes in
1786 userspace with buggy AVX enumeration.
1788 off: Disable GDS mitigation.
1790 gbpages [X86] Use GB pages for kernel direct mappings.
1792 gcov_persist= [GCOV] When non-zero (default), profiling data for
1793 kernel modules is saved and remains accessible via
1794 debugfs, even when the module is unloaded/reloaded.
1795 When zero, profiling data is discarded and associated
1796 debugfs files are removed at module unload time.
1798 goldfish [X86] Enable the goldfish android emulator platform.
1799 Don't use this when you are not running on the
1800 android emulator
1802 gpio-mockup.gpio_mockup_ranges
1803 [HW] Sets the ranges of gpiochip of for this device.
1804 Format: <start1>,<end1>,<start2>,<end2>...
1805 gpio-mockup.gpio_mockup_named_lines
1806 [HW] Let the driver know GPIO lines should be named.
1808 gpt [EFI] Forces disk with valid GPT signature but
1809 invalid Protective MBR to be treated as GPT. If the
1810 primary GPT is corrupted, it enables the backup/alternate
1811 GPT to be used instead.
1813 grcan.enable0= [HW] Configuration of physical interface 0. Determines
1814 the "Enable 0" bit of the configuration register.
1815 Format: 0 | 1
1816 Default: 0
1817 grcan.enable1= [HW] Configuration of physical interface 1. Determines
1818 the "Enable 0" bit of the configuration register.
1819 Format: 0 | 1
1820 Default: 0
1821 grcan.select= [HW] Select which physical interface to use.
1822 Format: 0 | 1
1823 Default: 0
1824 grcan.txsize= [HW] Sets the size of the tx buffer.
1825 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0.
1826 Default: 1024
1827 grcan.rxsize= [HW] Sets the size of the rx buffer.
1828 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0.
1829 Default: 1024
1831 hardened_usercopy=
1832 [KNL] Under CONFIG_HARDENED_USERCOPY, whether
1833 hardening is enabled for this boot. Hardened
1834 usercopy checking is used to protect the kernel
1835 from reading or writing beyond known memory
1836 allocation boundaries as a proactive defense
1837 against bounds-checking flaws in the kernel's
1838 copy_to_user()/copy_from_user() interface.
1839 The default is determined by
1840 CONFIG_HARDENED_USERCOPY_DEFAULT_ON.
1841 on Perform hardened usercopy checks.
1842 off Disable hardened usercopy checks.
1844 hardlockup_all_cpu_backtrace=
1845 [KNL] Should the hard-lockup detector generate
1846 backtraces on all cpus.
1847 Format: 0 | 1
1849 hash_pointers=
1850 [KNL,EARLY]
1851 By default, when pointers are printed to the console
1852 or buffers via the %p format string, that pointer is
1853 "hashed", i.e. obscured by hashing the pointer value.
1854 This is a security feature that hides actual kernel
1855 addresses from unprivileged users, but it also makes
1856 debugging the kernel more difficult since unequal
1857 pointers can no longer be compared. The choices are:
1858 Format: { auto | always | never }
1859 Default: auto
1861 auto - Hash pointers unless slab_debug is enabled.
1862 always - Always hash pointers (even if slab_debug is
1863 enabled).
1864 never - Never hash pointers. This option should only
1865 be specified when debugging the kernel. Do
1866 not use on production kernels. The boot
1867 param "no_hash_pointers" is an alias for
1868 this mode.
1870 hashdist= [KNL,NUMA] Large hashes allocated during boot
1871 are distributed across NUMA nodes. Defaults on
1872 for 64-bit NUMA, off otherwise.
1873 Format: 0 | 1 (for off | on)
1875 hd= [EIDE] (E)IDE hard drive subsystem geometry
1876 Format: <cyl>,<head>,<sect>
1878 hest_disable [ACPI]
1879 Disable Hardware Error Source Table (HEST) support;
1880 corresponding firmware-first mode error processing
1881 logic will be disabled.
1883 hibernate= [HIBERNATION]
1884 noresume Don't check if there's a hibernation image
1885 present during boot.
1886 nocompress Don't compress/decompress hibernation images.
1887 no Disable hibernation and resume.
1888 protect_image Turn on image protection during restoration
1889 (that will set all pages holding image data
1890 during restoration read-only).
1892 hibernate.compressor= [HIBERNATION] Compression algorithm to be
1893 used with hibernation.
1894 Format: { lzo | lz4 }
1895 Default: lzo
1897 lzo: Select LZO compression algorithm to
1898 compress/decompress hibernation image.
1900 lz4: Select LZ4 compression algorithm to
1901 compress/decompress hibernation image.
1903 hibernate.pm_test_delay=
1904 [HIBERNATION]
1905 Sets the number of seconds to remain in a hibernation test
1906 mode before resuming the system (see
1907 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG
1908 is set. Default value is 5.
1910 highmem=nn[KMG] [KNL,BOOT,EARLY] forces the highmem zone to have an exact
1911 size of <nn>. This works even on boxes that have no
1912 highmem otherwise. This also works to reduce highmem
1913 size on bigger boxes.
1915 highres= [KNL] Enable/disable high resolution timer mode.
1916 Valid parameters: "on", "off"
1917 Default: "on"
1919 hlt [BUGS=ARM,SH]
1921 hostname= [KNL,EARLY] Set the hostname (aka UTS nodename).
1922 Format: <string>
1923 This allows setting the system's hostname during early
1924 startup. This sets the name returned by gethostname.
1925 Using this parameter to set the hostname makes it
1926 possible to ensure the hostname is correctly set before
1927 any userspace processes run, avoiding the possibility
1928 that a process may call gethostname before the hostname
1929 has been explicitly set, resulting in the calling
1930 process getting an incorrect result. The string must
1931 not exceed the maximum allowed hostname length (usually
1932 64 characters) and will be truncated otherwise.
1934 hpet= [X86-32,HPET] option to control HPET usage
1935 Format: { enable (default) | disable | force |
1936 verbose }
1937 disable: disable HPET and use PIT instead
1938 force: allow force enabled of undocumented chips (ICH4,
1939 VIA, nVidia)
1940 verbose: show contents of HPET registers during setup
1942 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET
1943 registers. Default set by CONFIG_HPET_MMAP_DEFAULT.
1945 hugepages= [HW,EARLY] Number of HugeTLB pages to allocate at boot.
1946 If this follows hugepagesz (below), it specifies
1947 the number of pages of hugepagesz to be allocated.
1948 If this is the first HugeTLB parameter on the command
1949 line, it specifies the number of pages to allocate for
1950 the default huge page size. If using node format, the
1951 number of pages to allocate per-node can be specified.
1952 See also Documentation/admin-guide/mm/hugetlbpage.rst.
1953 Format: <integer> or (node format)
1954 <node>:<integer>[,<node>:<integer>]
1956 hugepagesz=
1957 [HW,EARLY] The size of the HugeTLB pages. This is
1958 used in conjunction with hugepages (above) to
1959 allocate huge pages of a specific size at boot. The
1960 pair hugepagesz=X hugepages=Y can be specified once
1961 for each supported huge page size. Huge page sizes
1962 are architecture dependent. See also
1963 Documentation/admin-guide/mm/hugetlbpage.rst.
1964 Format: size[KMG]
1966 hugepage_alloc_threads=
1967 [HW] The number of threads that should be used to
1968 allocate hugepages during boot. This option can be
1969 used to improve system bootup time when allocating
1970 a large amount of huge pages.
1971 The default value is 25% of the available hardware threads.
1973 Note that this parameter only applies to non-gigantic huge pages.
1975 hugetlb_cma= [HW,CMA,EARLY] The size of a CMA area used for allocation
1976 of gigantic hugepages. Or using node format, the size
1977 of a CMA area per node can be specified.
1978 Format: nn[KMGTPE] or (node format)
1979 <node>:nn[KMGTPE][,<node>:nn[KMGTPE]]
1981 Reserve a CMA area of given size and allocate gigantic
1982 hugepages using the CMA allocator. If enabled, the
1983 boot-time allocation of gigantic hugepages is skipped.
1985 hugetlb_cma_only=
1986 [HW,CMA,EARLY] When allocating new HugeTLB pages, only
1987 try to allocate from the CMA areas.
1989 This option does nothing if hugetlb_cma= is not also
1990 specified.
1992 hugetlb_free_vmemmap=
1993 [KNL] Requires CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP
1994 enabled.
1995 Control if HugeTLB Vmemmap Optimization (HVO) is enabled.
1996 Allows heavy hugetlb users to free up some more
1997 memory (7 * PAGE_SIZE for each 2MB hugetlb page).
1998 Format: { on | off (default) }
2000 on: enable HVO
2001 off: disable HVO
2003 Built with CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP_DEFAULT_ON=y,
2004 the default is on.
2006 Note that the vmemmap pages may be allocated from the added
2007 memory block itself when memory_hotplug.memmap_on_memory is
2008 enabled, those vmemmap pages cannot be optimized even if this
2009 feature is enabled. Other vmemmap pages not allocated from
2010 the added memory block itself do not be affected.
2012 hung_task_panic=
2013 [KNL] Should the hung task detector generate panics.
2014 Format: 0 | 1
2016 A value of 1 instructs the kernel to panic when a
2017 hung task is detected. The default value is controlled
2018 by the CONFIG_BOOTPARAM_HUNG_TASK_PANIC build-time
2019 option. The value selected by this boot parameter can
2020 be changed later by the kernel.hung_task_panic sysctl.
2022 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC)
2023 terminal devices. Valid values: 0..8
2024 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs.
2025 If specified, z/VM IUCV HVC accepts connections
2026 from listed z/VM user IDs only.
2028 hv_nopvspin [X86,HYPER_V,EARLY]
2029 Disables the paravirt spinlock optimizations
2030 which allow the hypervisor to 'idle' the guest
2031 on lock contention.
2033 hw_protection= [HW]
2034 Format: reboot | shutdown
2036 Hardware protection action taken on critical events like
2037 overtemperature or imminent voltage loss.
2039 i2c_bus= [HW] Override the default board specific I2C bus speed
2040 or register an additional I2C bus that is not
2041 registered from board initialization code.
2042 Format:
2043 <bus_id>,<clkrate>
2045 i2c_touchscreen_props= [HW,ACPI,X86]
2046 Set device-properties for ACPI-enumerated I2C-attached
2047 touchscreen, to e.g. fix coordinates of upside-down
2048 mounted touchscreens. If you need this option please
2049 submit a drivers/platform/x86/touchscreen_dmi.c patch
2050 adding a DMI quirk for this.
2052 Format:
2053 <ACPI_HW_ID>:<prop_name>=<val>[:prop_name=val][:...]
2054 Where <val> is one of:
2055 Omit "=<val>" entirely Set a boolean device-property
2056 Unsigned number Set a u32 device-property
2057 Anything else Set a string device-property
2059 Examples (split over multiple lines):
2060 i2c_touchscreen_props=GDIX1001:touchscreen-inverted-x:
2061 touchscreen-inverted-y
2063 i2c_touchscreen_props=MSSL1680:touchscreen-size-x=1920:
2064 touchscreen-size-y=1080:touchscreen-inverted-y:
2065 firmware-name=gsl1680-vendor-model.fw:silead,home-button
2067 i8042.debug [HW] Toggle i8042 debug mode
2068 i8042.unmask_kbd_data
2069 [HW] Enable printing of interrupt data from the KBD port
2070 (disabled by default, and as a pre-condition
2071 requires that i8042.debug=1 be enabled)
2072 i8042.direct [HW] Put keyboard port into non-translated mode
2073 i8042.dumbkbd [HW] Pretend that controller can only read data from
2074 keyboard and cannot control its state
2075 (Don't attempt to blink the leds)
2076 i8042.noaux [HW] Don't check for auxiliary (== mouse) port
2077 i8042.nokbd [HW] Don't check/create keyboard port
2078 i8042.noloop [HW] Disable the AUX Loopback command while probing
2079 for the AUX port
2080 i8042.nomux [HW] Don't check presence of an active multiplexing
2081 controller
2082 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX
2083 controllers
2084 i8042.notimeout [HW] Ignore timeout condition signalled by controller
2085 i8042.reset [HW] Reset the controller during init, cleanup and
2086 suspend-to-ram transitions, only during s2r
2087 transitions, or never reset
2088 Format: { 1 | Y | y | 0 | N | n }
2089 1, Y, y: always reset controller
2090 0, N, n: don't ever reset controller
2091 Default: only on s2r transitions on x86; most other
2092 architectures force reset to be always executed
2093 i8042.unlock [HW] Unlock (ignore) the keylock
2094 i8042.kbdreset [HW] Reset device connected to KBD port
2095 i8042.probe_defer
2096 [HW] Allow deferred probing upon i8042 probe errors
2098 i810= [HW,DRM]
2100 i915.invert_brightness=
2101 [DRM] Invert the sense of the variable that is used to
2102 set the brightness of the panel backlight. Normally a
2103 brightness value of 0 indicates backlight switched off,
2104 and the maximum of the brightness value sets the backlight
2105 to maximum brightness. If this parameter is set to 0
2106 (default) and the machine requires it, or this parameter
2107 is set to 1, a brightness value of 0 sets the backlight
2108 to maximum brightness, and the maximum of the brightness
2109 value switches the backlight off.
2110 -1 -- never invert brightness
2111 0 -- machine default
2112 1 -- force brightness inversion
2114 ia32_emulation= [X86-64]
2115 Format: <bool>
2116 When true, allows loading 32-bit programs and executing 32-bit
2117 syscalls, essentially overriding IA32_EMULATION_DEFAULT_DISABLED at
2118 boot time. When false, unconditionally disables IA32 emulation.
2120 icn= [HW,ISDN]
2121 Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]]
2124 idle= [X86,EARLY]
2125 Format: idle=poll, idle=halt, idle=nomwait
2127 idle=poll: Don't do power saving in the idle loop
2128 using HLT, but poll for rescheduling event. This will
2129 make the CPUs eat a lot more power, but may be useful
2130 to get slightly better performance in multiprocessor
2131 benchmarks. It also makes some profiling using
2132 performance counters more accurate. Please note that
2133 on systems with MONITOR/MWAIT support (like Intel
2134 EM64T CPUs) this option has no performance advantage
2135 over the normal idle loop. It may also interact badly
2136 with hyperthreading.
2138 idle=halt: Halt is forced to be used for CPU idle.
2139 In such case C2/C3 won't be used again.
2141 idle=nomwait: Disable mwait for CPU C-states
2143 idxd.sva= [HW]
2144 Format: <bool>
2145 Allow force disabling of Shared Virtual Memory (SVA)
2146 support for the idxd driver. By default it is set to
2147 true (1).
2149 idxd.tc_override= [HW]
2150 Format: <bool>
2151 Allow override of default traffic class configuration
2152 for the device. By default it is set to false (0).
2154 ieee754= [MIPS] Select IEEE Std 754 conformance mode
2155 Format: { strict | legacy | 2008 | relaxed | emulated }
2156 Default: strict
2158 Choose which programs will be accepted for execution
2159 based on the IEEE 754 NaN encoding(s) supported by
2160 the FPU and the NaN encoding requested with the value
2161 of an ELF file header flag individually set by each
2162 binary. Hardware implementations are permitted to
2163 support either or both of the legacy and the 2008 NaN
2164 encoding mode.
2166 Available settings are as follows:
2167 strict accept binaries that request a NaN encoding
2168 supported by the FPU
2169 legacy only accept legacy-NaN binaries, if supported
2170 by the FPU
2171 2008 only accept 2008-NaN binaries, if supported
2172 by the FPU
2173 relaxed accept any binaries regardless of whether
2174 supported by the FPU
2175 emulated accept any binaries but enable FPU emulator
2176 if binary mode is unsupported by the FPU.
2178 The FPU emulator is always able to support both NaN
2179 encodings, so if no FPU hardware is present or it has
2180 been disabled with 'nofpu', then the settings of
2181 'legacy' and '2008' strap the emulator accordingly,
2182 'relaxed' straps the emulator for both legacy-NaN and
2183 2008-NaN, whereas 'strict' enables legacy-NaN only on
2184 legacy processors and both NaN encodings on MIPS32 or
2185 MIPS64 CPUs.
2187 The setting for ABS.fmt/NEG.fmt instruction execution
2188 mode generally follows that for the NaN encoding,
2189 except where unsupported by hardware.
2191 ignore_loglevel [KNL,EARLY]
2192 Ignore loglevel setting - this will print /all/
2193 kernel messages to the console. Useful for debugging.
2194 We also add it as printk module parameter, so users
2195 could change it dynamically, usually by
2196 /sys/module/printk/parameters/ignore_loglevel.
2198 ignore_rlimit_data
2199 Ignore RLIMIT_DATA setting for data mappings,
2200 print warning at first misuse. Can be changed via
2201 /sys/module/kernel/parameters/ignore_rlimit_data.
2203 ihash_entries= [KNL]
2204 Set number of hash buckets for inode cache.
2206 ima_appraise= [IMA] appraise integrity measurements
2207 Format: { "off" | "enforce" | "fix" | "log" }
2208 default: "enforce"
2210 ima_appraise_tcb [IMA] Deprecated. Use ima_policy= instead.
2211 The builtin appraise policy appraises all files
2212 owned by uid=0.
2214 ima_canonical_fmt [IMA]
2215 Use the canonical format for the binary runtime
2216 measurements, instead of host native format.
2218 ima_hash= [IMA]
2219 Format: { md5 | sha1 | rmd160 | sha256 | sha384
2220 | sha512 | ... }
2221 default: "sha1"
2223 The list of supported hash algorithms is defined
2224 in crypto/hash_info.h.
2226 ima_policy= [IMA]
2227 The builtin policies to load during IMA setup.
2228 Format: "tcb | appraise_tcb | secure_boot |
2229 fail_securely | critical_data"
2231 The "tcb" policy measures all programs exec'd, files
2232 mmap'd for exec, and all files opened with the read
2233 mode bit set by either the effective uid (euid=0) or
2234 uid=0.
2236 The "appraise_tcb" policy appraises the integrity of
2237 all files owned by root.
2239 The "secure_boot" policy appraises the integrity
2240 of files (eg. kexec kernel image, kernel modules,
2241 firmware, policy, etc) based on file signatures.
2243 The "fail_securely" policy forces file signature
2244 verification failure also on privileged mounted
2245 filesystems with the SB_I_UNVERIFIABLE_SIGNATURE
2246 flag.
2248 The "critical_data" policy measures kernel integrity
2249 critical data.
2251 ima_tcb [IMA] Deprecated. Use ima_policy= instead.
2252 Load a policy which meets the needs of the Trusted
2253 Computing Base. This means IMA will measure all
2254 programs exec'd, files mmap'd for exec, and all files
2255 opened for read by uid=0.
2257 ima_template= [IMA]
2258 Select one of defined IMA measurements template formats.
2259 Formats: { "ima" | "ima-ng" | "ima-ngv2" | "ima-sig" |
2260 "ima-sigv2" }
2261 Default: "ima-ng"
2263 ima_template_fmt=
2264 [IMA] Define a custom template format.
2265 Format: { "field1|...|fieldN" }
2267 ima.ahash_minsize= [IMA] Minimum file size for asynchronous hash usage
2268 Format: <min_file_size>
2269 Set the minimal file size for using asynchronous hash.
2270 If left unspecified, ahash usage is disabled.
2272 ahash performance varies for different data sizes on
2273 different crypto accelerators. This option can be used
2274 to achieve the best performance for a particular HW.
2276 ima.ahash_bufsize= [IMA] Asynchronous hash buffer size
2277 Format: <bufsize>
2278 Set hashing buffer size. Default: 4k.
2280 ahash performance varies for different chunk sizes on
2281 different crypto accelerators. This option can be used
2282 to achieve best performance for particular HW.
2284 ima= [IMA] Enable or disable IMA
2285 Format: { "off" | "on" }
2286 Default: "on"
2287 Note that disabling IMA is limited to kdump kernel.
2289 indirect_target_selection= [X86,Intel] Mitigation control for Indirect
2290 Target Selection(ITS) bug in Intel CPUs. Updated
2291 microcode is also required for a fix in IBPB.
2293 on: Enable mitigation (default).
2294 off: Disable mitigation.
2295 force: Force the ITS bug and deploy default
2296 mitigation.
2297 vmexit: Only deploy mitigation if CPU is affected by
2298 guest/host isolation part of ITS.
2299 stuff: Deploy RSB-fill mitigation when retpoline is
2300 also deployed. Otherwise, deploy the default
2301 mitigation.
2303 For details see:
2304 Documentation/admin-guide/hw-vuln/indirect-target-selection.rst
2306 init= [KNL]
2307 Format: <full_path>
2308 Run specified binary instead of /sbin/init as init
2309 process.
2311 initcall_debug [KNL] Trace initcalls as they are executed. Useful
2312 for working out where the kernel is dying during
2313 startup.
2315 initcall_blacklist= [KNL] Do not execute a comma-separated list of
2316 initcall functions. Useful for debugging built-in
2317 modules and initcalls.
2319 initramfs_async= [KNL]
2320 Format: <bool>
2321 Default: 1
2322 This parameter controls whether the initramfs
2323 image is unpacked asynchronously, concurrently
2324 with devices being probed and
2325 initialized. This should normally just work,
2326 but as a debugging aid, one can get the
2327 historical behaviour of the initramfs
2328 unpacking being completed before device_ and
2329 late_ initcalls.
2331 initrd= [BOOT,EARLY] Specify the location of the initial ramdisk
2333 initrdmem= [KNL,EARLY] Specify a physical address and size from which to
2334 load the initrd. If an initrd is compiled in or
2335 specified in the bootparams, it takes priority over this
2336 setting.
2337 Format: ss[KMG],nn[KMG]
2338 Default is 0, 0
2340 init_on_alloc= [MM,EARLY] Fill newly allocated pages and heap objects with
2341 zeroes.
2342 Format: 0 | 1
2343 Default set by CONFIG_INIT_ON_ALLOC_DEFAULT_ON.
2345 init_on_free= [MM,EARLY] Fill freed pages and heap objects with zeroes.
2346 Format: 0 | 1
2347 Default set by CONFIG_INIT_ON_FREE_DEFAULT_ON.
2349 init_pkru= [X86] Specify the default memory protection keys rights
2350 register contents for all processes. 0x55555554 by
2351 default (disallow access to all but pkey 0). Can
2352 override in debugfs after boot.
2354 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver
2355 Format: <irq>
2357 int_pln_enable [X86] Enable power limit notification interrupt
2359 integrity_audit=[IMA]
2360 Format: { "0" | "1" }
2361 0 -- basic integrity auditing messages. (Default)
2362 1 -- additional integrity auditing messages.
2364 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option
2365 on
2366 Enable intel iommu driver.
2367 off
2368 Disable intel iommu driver.
2369 igfx_off [Default Off]
2370 By default, gfx is mapped as normal device. If a gfx
2371 device has a dedicated DMAR unit, the DMAR unit is
2372 bypassed by not enabling DMAR with this option. In
2373 this case, gfx device will use physical address for
2374 DMA.
2375 strict [Default Off]
2376 Deprecated, equivalent to iommu.strict=1.
2377 sp_off [Default Off]
2378 By default, super page will be supported if Intel IOMMU
2379 has the capability. With this option, super page will
2380 not be supported.
2381 sm_on
2382 Enable the Intel IOMMU scalable mode if the hardware
2383 advertises that it has support for the scalable mode
2384 translation.
2385 sm_off
2386 Disallow use of the Intel IOMMU scalable mode.
2387 tboot_noforce [Default Off]
2388 Do not force the Intel IOMMU enabled under tboot.
2389 By default, tboot will force Intel IOMMU on, which
2390 could harm performance of some high-throughput
2391 devices like 40GBit network cards, even if identity
2392 mapping is enabled.
2393 Note that using this option lowers the security
2394 provided by tboot because it makes the system
2395 vulnerable to DMA attacks.
2397 intel_idle.max_cstate= [KNL,HW,ACPI,X86]
2398 0 disables intel_idle and fall back on acpi_idle.
2399 1 to 9 specify maximum depth of C-state.
2401 intel_pstate= [X86,EARLY]
2402 disable
2403 Do not enable intel_pstate as the default
2404 scaling driver for the supported processors
2405 active
2406 Use intel_pstate driver to bypass the scaling
2407 governors layer of cpufreq and provides it own
2408 algorithms for p-state selection. There are two
2409 P-state selection algorithms provided by
2410 intel_pstate in the active mode: powersave and
2411 performance. The way they both operate depends
2412 on whether or not the hardware managed P-states
2413 (HWP) feature has been enabled in the processor
2414 and possibly on the processor model.
2415 passive
2416 Use intel_pstate as a scaling driver, but configure it
2417 to work with generic cpufreq governors (instead of
2418 enabling its internal governor). This mode cannot be
2419 used along with the hardware-managed P-states (HWP)
2420 feature.
2421 force
2422 Enable intel_pstate on systems that prohibit it by default
2423 in favor of acpi-cpufreq. Forcing the intel_pstate driver
2424 instead of acpi-cpufreq may disable platform features, such
2425 as thermal controls and power capping, that rely on ACPI
2426 P-States information being indicated to OSPM and therefore
2427 should be used with caution. This option does not work with
2428 processors that aren't supported by the intel_pstate driver
2429 or on platforms that use pcc-cpufreq instead of acpi-cpufreq.
2430 no_hwp
2431 Do not enable hardware P state control (HWP)
2432 if available.
2433 hwp_only
2434 Only load intel_pstate on systems which support
2435 hardware P state control (HWP) if available.
2436 support_acpi_ppc
2437 Enforce ACPI _PPC performance limits. If the Fixed ACPI
2438 Description Table, specifies preferred power management
2439 profile as "Enterprise Server" or "Performance Server",
2440 then this feature is turned on by default.
2441 per_cpu_perf_limits
2442 Allow per-logical-CPU P-State performance control limits using
2443 cpufreq sysfs interface
2444 no_cas
2445 Do not enable capacity-aware scheduling (CAS) on
2446 hybrid systems
2448 intremap= [X86-64,Intel-IOMMU,EARLY]
2449 on enable Interrupt Remapping (default)
2450 off disable Interrupt Remapping
2451 nosid disable Source ID checking
2452 no_x2apic_optout
2453 BIOS x2APIC opt-out request will be ignored
2454 nopost disable Interrupt Posting
2455 posted_msi
2456 enable MSIs delivered as posted interrupts
2458 iomem= Disable strict checking of access to MMIO memory
2459 strict regions from userspace.
2460 relaxed
2462 iommu= [X86,EARLY]
2464 off
2465 Don't initialize and use any kind of IOMMU.
2467 force
2468 Force the use of the hardware IOMMU even when
2469 it is not actually needed (e.g. because < 3 GB
2470 memory).
2472 noforce
2473 Don't force hardware IOMMU usage when it is not
2474 needed. (default).
2476 biomerge
2477 panic
2478 nopanic
2479 merge
2480 nomerge
2482 soft
2483 Use software bounce buffering (SWIOTLB) (default for
2484 Intel machines). This can be used to prevent the usage
2485 of an available hardware IOMMU.
2487 [X86]
2488 pt
2489 [X86]
2490 nopt
2491 [PPC/POWERNV]
2492 nobypass
2493 Disable IOMMU bypass, using IOMMU for PCI devices.
2495 [X86]
2496 AMD Gart HW IOMMU-specific options:
2498 <size>
2499 Set the size of the remapping area in bytes.
2501 allowed
2502 Overwrite iommu off workarounds for specific chipsets
2504 fullflush
2505 Flush IOMMU on each allocation (default).
2507 nofullflush
2508 Don't use IOMMU fullflush.
2510 memaper[=<order>]
2511 Allocate an own aperture over RAM with size
2512 32MB<<order. (default: order=1, i.e. 64MB)
2514 merge
2515 Do scatter-gather (SG) merging. Implies "force"
2516 (experimental).
2518 nomerge
2519 Don't do scatter-gather (SG) merging.
2521 noaperture
2522 Ask the IOMMU not to touch the aperture for AGP.
2524 noagp
2525 Don't initialize the AGP driver and use full aperture.
2527 panic
2528 Always panic when IOMMU overflows.
2530 iommu.forcedac= [ARM64,X86,EARLY] Control IOVA allocation for PCI devices.
2531 Format: { "0" | "1" }
2532 0 - Try to allocate a 32-bit DMA address first, before
2533 falling back to the full range if needed.
2534 1 - Allocate directly from the full usable range,
2535 forcing Dual Address Cycle for PCI cards supporting
2536 greater than 32-bit addressing.
2538 iommu.strict= [ARM64,X86,S390,EARLY] Configure TLB invalidation behaviour
2539 Format: { "0" | "1" }
2540 0 - Lazy mode.
2541 Request that DMA unmap operations use deferred
2542 invalidation of hardware TLBs, for increased
2543 throughput at the cost of reduced device isolation.
2544 Will fall back to strict mode if not supported by
2545 the relevant IOMMU driver.
2546 1 - Strict mode.
2547 DMA unmap operations invalidate IOMMU hardware TLBs
2548 synchronously.
2549 unset - Use value of CONFIG_IOMMU_DEFAULT_DMA_{LAZY,STRICT}.
2550 Note: on x86, strict mode specified via one of the
2551 legacy driver-specific options takes precedence.
2553 iommu.passthrough=
2554 [ARM64,X86,EARLY] Configure DMA to bypass the IOMMU by default.
2555 Format: { "0" | "1" }
2556 0 - Use IOMMU translation for DMA.
2557 1 - Bypass the IOMMU for DMA.
2558 unset - Use value of CONFIG_IOMMU_DEFAULT_PASSTHROUGH.
2560 io7= [HW] IO7 for Marvel-based Alpha systems
2561 See comment before marvel_specify_io7 in
2562 arch/alpha/kernel/core_marvel.c.
2564 io_delay= [X86,EARLY] I/O delay method
2565 0x80
2566 Standard port 0x80 based delay
2567 0xed
2568 Alternate port 0xed based delay (needed on some systems)
2569 udelay
2570 Simple two microseconds delay
2571 none
2572 No delay
2574 ip= [IP_PNP]
2575 See Documentation/admin-guide/nfs/nfsroot.rst.
2577 ipcmni_extend [KNL,EARLY] Extend the maximum number of unique System V
2578 IPC identifiers from 32,768 to 16,777,216.
2580 ipe.enforce= [IPE]
2581 Format: <bool>
2582 Determine whether IPE starts in permissive (0) or
2583 enforce (1) mode. The default is enforce.
2585 ipe.success_audit=
2586 [IPE]
2587 Format: <bool>
2588 Start IPE with success auditing enabled, emitting
2589 an audit event when a binary is allowed. The default
2590 is 0.
2592 irqaffinity= [SMP] Set the default irq affinity mask
2593 The argument is a cpu list, as described above.
2595 irqchip.gicv2_force_probe=
2596 [ARM,ARM64,EARLY]
2597 Format: <bool>
2598 Force the kernel to look for the second 4kB page
2599 of a GICv2 controller even if the memory range
2600 exposed by the device tree is too small.
2602 irqchip.gicv3_nolpi=
2603 [ARM,ARM64,EARLY]
2604 Force the kernel to ignore the availability of
2605 LPIs (and by consequence ITSs). Intended for system
2606 that use the kernel as a bootloader, and thus want
2607 to let secondary kernels in charge of setting up
2608 LPIs.
2610 irqchip.gicv3_pseudo_nmi= [ARM64,EARLY]
2611 Enables support for pseudo-NMIs in the kernel. This
2612 requires the kernel to be built with
2613 CONFIG_ARM64_PSEUDO_NMI.
2615 irqchip.riscv_imsic_noipi
2616 [RISC-V,EARLY]
2617 Force the kernel to not use IMSIC software injected MSIs
2618 as IPIs. Intended for system where IMSIC is trap-n-emulated,
2619 and thus want to reduce MMIO traps when triggering IPIs
2620 to multiple harts.
2622 irqfixup [HW]
2623 When an interrupt is not handled search all handlers
2624 for it. Intended to get systems with badly broken
2625 firmware running.
2627 irqhandler.duration_warn_us= [KNL]
2628 Warn if an IRQ handler exceeds the specified duration
2629 threshold in microseconds. Useful for identifying
2630 long-running IRQs in the system.
2632 irqpoll [HW]
2633 When an interrupt is not handled search all handlers
2634 for it. Also check all handlers each timer
2635 interrupt. Intended to get systems with badly broken
2636 firmware running.
2638 isapnp= [ISAPNP]
2639 Format: <RDP>,<reset>,<pci_scan>,<verbosity>
2641 isolcpus= [KNL,SMP,ISOL] Isolate a given set of CPUs from disturbance.
2642 [Deprecated - use cpusets instead]
2643 Format: [flag-list,]<cpu-list>
2645 Specify one or more CPUs to isolate from disturbances
2646 specified in the flag list (default: domain):
2648 nohz
2649 Disable the tick when a single task runs as well as
2650 disabling other kernel noises like having RCU callbacks
2651 offloaded. This is equivalent to the nohz_full parameter.
2653 A residual 1Hz tick is offloaded to workqueues, which you
2654 need to affine to housekeeping through the global
2655 workqueue's affinity configured via the
2656 /sys/devices/virtual/workqueue/cpumask sysfs file, or
2657 by using the 'domain' flag described below.
2659 NOTE: by default the global workqueue runs on all CPUs,
2660 so to protect individual CPUs the 'cpumask' file has to
2661 be configured manually after bootup.
2663 domain
2664 Isolate from the general SMP balancing and scheduling
2665 algorithms. Note that performing domain isolation this way
2666 is irreversible: it's not possible to bring back a CPU to
2667 the domains once isolated through isolcpus. It's strongly
2668 advised to use cpusets instead to disable scheduler load
2669 balancing through the "cpuset.sched_load_balance" file.
2670 It offers a much more flexible interface where CPUs can
2671 move in and out of an isolated set anytime.
2673 You can move a process onto or off an "isolated" CPU via
2674 the CPU affinity syscalls or cpuset.
2675 <cpu number> begins at 0 and the maximum value is
2676 "number of CPUs in system - 1".
2678 managed_irq
2680 Isolate from being targeted by managed interrupts
2681 which have an interrupt mask containing isolated
2682 CPUs. The affinity of managed interrupts is
2683 handled by the kernel and cannot be changed via
2684 the /proc/irq/* interfaces.
2686 This isolation is best effort and only effective
2687 if the automatically assigned interrupt mask of a
2688 device queue contains isolated and housekeeping
2689 CPUs. If housekeeping CPUs are online then such
2690 interrupts are directed to the housekeeping CPU
2691 so that IO submitted on the housekeeping CPU
2692 cannot disturb the isolated CPU.
2694 If a queue's affinity mask contains only isolated
2695 CPUs then this parameter has no effect on the
2696 interrupt routing decision, though interrupts are
2697 only delivered when tasks running on those
2698 isolated CPUs submit IO. IO submitted on
2699 housekeeping CPUs has no influence on those
2700 queues.
2702 The format of <cpu-list> is described above.
2704 iucv= [HW,NET]
2706 ivrs_ioapic [HW,X86-64]
2707 Provide an override to the IOAPIC-ID<->DEVICE-ID
2708 mapping provided in the IVRS ACPI table.
2709 By default, PCI segment is 0, and can be omitted.
2711 For example, to map IOAPIC-ID decimal 10 to
2712 PCI segment 0x1 and PCI device 00:14.0,
2713 write the parameter as:
2714 ivrs_ioapic=10@0001:00:14.0
2716 Deprecated formats:
2717 * To map IOAPIC-ID decimal 10 to PCI device 00:14.0
2718 write the parameter as:
2719 ivrs_ioapic[10]=00:14.0
2720 * To map IOAPIC-ID decimal 10 to PCI segment 0x1 and
2721 PCI device 00:14.0 write the parameter as:
2722 ivrs_ioapic[10]=0001:00:14.0
2724 ivrs_hpet [HW,X86-64]
2725 Provide an override to the HPET-ID<->DEVICE-ID
2726 mapping provided in the IVRS ACPI table.
2727 By default, PCI segment is 0, and can be omitted.
2729 For example, to map HPET-ID decimal 10 to
2730 PCI segment 0x1 and PCI device 00:14.0,
2731 write the parameter as:
2732 ivrs_hpet=10@0001:00:14.0
2734 Deprecated formats:
2735 * To map HPET-ID decimal 0 to PCI device 00:14.0
2736 write the parameter as:
2737 ivrs_hpet[0]=00:14.0
2738 * To map HPET-ID decimal 10 to PCI segment 0x1 and
2739 PCI device 00:14.0 write the parameter as:
2740 ivrs_ioapic[10]=0001:00:14.0
2742 ivrs_acpihid [HW,X86-64]
2743 Provide an override to the ACPI-HID:UID<->DEVICE-ID
2744 mapping provided in the IVRS ACPI table.
2745 By default, PCI segment is 0, and can be omitted.
2747 For example, to map UART-HID:UID AMD0020:0 to
2748 PCI segment 0x1 and PCI device ID 00:14.5,
2749 write the parameter as:
2750 ivrs_acpihid=AMD0020:0@0001:00:14.5
2752 Deprecated formats:
2753 * To map UART-HID:UID AMD0020:0 to PCI segment is 0,
2754 PCI device ID 00:14.5, write the parameter as:
2755 ivrs_acpihid[00:14.5]=AMD0020:0
2756 * To map UART-HID:UID AMD0020:0 to PCI segment 0x1 and
2757 PCI device ID 00:14.5, write the parameter as:
2758 ivrs_acpihid[0001:00:14.5]=AMD0020:0
2760 js= [HW,JOY] Analog joystick
2761 See Documentation/input/joydev/joystick.rst.
2763 kasan_multi_shot
2764 [KNL] Enforce KASAN (Kernel Address Sanitizer) to print
2765 report on every invalid memory access. Without this
2766 parameter KASAN will print report only for the first
2767 invalid access.
2769 keep_bootcon [KNL,EARLY]
2770 Do not unregister boot console at start. This is only
2771 useful for debugging when something happens in the window
2772 between unregistering the boot console and initializing
2773 the real console.
2775 keepinitrd [HW,ARM] See retain_initrd.
2777 kernelcore= [KNL,X86,PPC,EARLY]
2778 Format: nn[KMGTPE] | nn% | "mirror"
2779 This parameter specifies the amount of memory usable by
2780 the kernel for non-movable allocations. The requested
2781 amount is spread evenly throughout all nodes in the
2782 system as ZONE_NORMAL. The remaining memory is used for
2783 movable memory in its own zone, ZONE_MOVABLE. In the
2784 event, a node is too small to have both ZONE_NORMAL and
2785 ZONE_MOVABLE, kernelcore memory will take priority and
2786 other nodes will have a larger ZONE_MOVABLE.
2788 ZONE_MOVABLE is used for the allocation of pages that
2789 may be reclaimed or moved by the page migration
2790 subsystem. Note that allocations like PTEs-from-HighMem
2791 still use the HighMem zone if it exists, and the Normal
2792 zone if it does not.
2794 It is possible to specify the exact amount of memory in
2795 the form of "nn[KMGTPE]", a percentage of total system
2796 memory in the form of "nn%", or "mirror". If "mirror"
2797 option is specified, mirrored (reliable) memory is used
2798 for non-movable allocations and remaining memory is used
2799 for Movable pages. "nn[KMGTPE]", "nn%", and "mirror"
2800 are exclusive, so you cannot specify multiple forms.
2802 kgdbdbgp= [KGDB,HW,EARLY] kgdb over EHCI usb debug port.
2803 Format: <Controller#>[,poll interval]
2804 The controller # is the number of the ehci usb debug
2805 port as it is probed via PCI. The poll interval is
2806 optional and is the number seconds in between
2807 each poll cycle to the debug port in case you need
2808 the functionality for interrupting the kernel with
2809 gdb or control-c on the dbgp connection. When
2810 not using this parameter you use sysrq-g to break into
2811 the kernel debugger.
2813 kgdboc= [KGDB,HW] kgdb over consoles.
2814 Requires a tty driver that supports console polling,
2815 or a supported polling keyboard driver (non-usb).
2816 Serial only format: <serial_device>[,baud]
2817 keyboard only format: kbd
2818 keyboard and serial format: kbd,<serial_device>[,baud]
2819 Optional Kernel mode setting:
2820 kms, kbd format: kms,kbd
2821 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud]
2823 kgdboc_earlycon= [KGDB,HW,EARLY]
2824 If the boot console provides the ability to read
2825 characters and can work in polling mode, you can use
2826 this parameter to tell kgdb to use it as a backend
2827 until the normal console is registered. Intended to
2828 be used together with the kgdboc parameter which
2829 specifies the normal console to transition to.
2831 The name of the early console should be specified
2832 as the value of this parameter. Note that the name of
2833 the early console might be different than the tty
2834 name passed to kgdboc. It's OK to leave the value
2835 blank and the first boot console that implements
2836 read() will be picked.
2838 kgdbwait [KGDB,EARLY] Stop kernel execution and enter the
2839 kernel debugger at the earliest opportunity.
2841 kho= [KEXEC,EARLY]
2842 Format: { "0" | "1" | "off" | "on" | "y" | "n" }
2843 Enables or disables Kexec HandOver.
2844 "0" | "off" | "n" - kexec handover is disabled
2845 "1" | "on" | "y" - kexec handover is enabled
2847 kho_scratch= [KEXEC,EARLY]
2848 Format: ll[KMG],mm[KMG],nn[KMG] | nn%
2849 Defines the size of the KHO scratch region. The KHO
2850 scratch regions are physically contiguous memory
2851 ranges that can only be used for non-kernel
2852 allocations. That way, even when memory is heavily
2853 fragmented with handed over memory, the kexeced
2854 kernel will always have enough contiguous ranges to
2855 bootstrap itself.
2857 It is possible to specify the exact amount of
2858 memory in the form of "ll[KMG],mm[KMG],nn[KMG]"
2859 where the first parameter defines the size of a low
2860 memory scratch area, the second parameter defines
2861 the size of a global scratch area and the third
2862 parameter defines the size of additional per-node
2863 scratch areas. The form "nn%" defines scale factor
2864 (in percents) of memory that was used during boot.
2866 kmac= [MIPS] Korina ethernet MAC address.
2867 Configure the RouterBoard 532 series on-chip
2868 Ethernet adapter MAC address.
2870 kmemleak= [KNL,EARLY] Boot-time kmemleak enable/disable
2871 Valid arguments: on, off
2872 Default: on
2873 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y,
2874 the default is off.
2876 kprobe_event=[probe-list]
2877 [FTRACE] Add kprobe events and enable at boot time.
2878 The probe-list is a semicolon delimited list of probe
2879 definitions. Each definition is same as kprobe_events
2880 interface, but the parameters are comma delimited.
2881 For example, to add a kprobe event on vfs_read with
2882 arg1 and arg2, add to the command line;
2884 kprobe_event=p,vfs_read,$arg1,$arg2
2886 See also Documentation/trace/kprobetrace.rst "Kernel
2887 Boot Parameter" section.
2889 kpti= [ARM64,EARLY] Control page table isolation of
2890 user and kernel address spaces.
2891 Default: enabled on cores which need mitigation.
2892 0: force disabled
2893 1: force enabled
2895 kunit.enable= [KUNIT] Enable executing KUnit tests. Requires
2896 CONFIG_KUNIT to be set to be fully enabled. The
2897 default value can be overridden via
2898 KUNIT_DEFAULT_ENABLED.
2899 Default is 1 (enabled)
2901 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs.
2902 Default is 0 (don't ignore, but inject #GP)
2904 kvm.eager_page_split=
2905 [KVM,X86] Controls whether or not KVM will try to
2906 proactively split all huge pages during dirty logging.
2907 Eager page splitting reduces interruptions to vCPU
2908 execution by eliminating the write-protection faults
2909 and MMU lock contention that would otherwise be
2910 required to split huge pages lazily.
2912 VM workloads that rarely perform writes or that write
2913 only to a small region of VM memory may benefit from
2914 disabling eager page splitting to allow huge pages to
2915 still be used for reads.
2917 The behavior of eager page splitting depends on whether
2918 KVM_DIRTY_LOG_INITIALLY_SET is enabled or disabled. If
2919 disabled, all huge pages in a memslot will be eagerly
2920 split when dirty logging is enabled on that memslot. If
2921 enabled, eager page splitting will be performed during
2922 the KVM_CLEAR_DIRTY ioctl, and only for the pages being
2923 cleared.
2925 Eager page splitting is only supported when kvm.tdp_mmu=Y.
2927 Default is Y (on).
2929 kvm.enable_virt_at_load=[KVM,ARM64,LOONGARCH,MIPS,RISCV,X86]
2930 If enabled, KVM will enable virtualization in hardware
2931 when KVM is loaded, and disable virtualization when KVM
2932 is unloaded (if KVM is built as a module).
2934 If disabled, KVM will dynamically enable and disable
2935 virtualization on-demand when creating and destroying
2936 VMs, i.e. on the 0=>1 and 1=>0 transitions of the
2937 number of VMs.
2939 Enabling virtualization at module load avoids potential
2940 latency for creation of the 0=>1 VM, as KVM serializes
2941 virtualization enabling across all online CPUs. The
2942 "cost" of enabling virtualization when KVM is loaded,
2943 is that doing so may interfere with using out-of-tree
2944 hypervisors that want to "own" virtualization hardware.
2946 kvm.enable_vmware_backdoor=[KVM] Support VMware backdoor PV interface.
2947 Default is false (don't support).
2949 kvm.nx_huge_pages=
2950 [KVM] Controls the software workaround for the
2951 X86_BUG_ITLB_MULTIHIT bug.
2952 force : Always deploy workaround.
2953 off : Never deploy workaround.
2954 auto : Deploy workaround based on the presence of
2955 X86_BUG_ITLB_MULTIHIT.
2957 Default is 'auto'.
2959 If the software workaround is enabled for the host,
2960 guests do need not to enable it for nested guests.
2962 kvm.nx_huge_pages_recovery_ratio=
2963 [KVM] Controls how many 4KiB pages are periodically zapped
2964 back to huge pages. 0 disables the recovery, otherwise if
2965 the value is N KVM will zap 1/Nth of the 4KiB pages every
2966 period (see below). The default is 60.
2968 kvm.nx_huge_pages_recovery_period_ms=
2969 [KVM] Controls the time period at which KVM zaps 4KiB pages
2970 back to huge pages. If the value is a non-zero N, KVM will
2971 zap a portion (see ratio above) of the pages every N msecs.
2972 If the value is 0 (the default), KVM will pick a period based
2973 on the ratio, such that a page is zapped after 1 hour on average.
2975 kvm-amd.nested= [KVM,AMD] Control nested virtualization feature in
2976 KVM/SVM. Default is 1 (enabled).
2978 kvm-amd.npt= [KVM,AMD] Control KVM's use of Nested Page Tables,
2979 a.k.a. Two-Dimensional Page Tables. Default is 1
2980 (enabled). Disable by KVM if hardware lacks support
2981 for NPT.
2983 kvm-amd.ciphertext_hiding_asids=
2984 [KVM,AMD] Ciphertext hiding prevents disallowed accesses
2985 to SNP private memory from reading ciphertext. Instead,
2986 reads will see constant default values (0xff).
2988 If ciphertext hiding is enabled, the joint SEV-ES and
2989 SEV-SNP ASID space is partitioned into separate SEV-ES
2990 and SEV-SNP ASID ranges, with the SEV-SNP range being
2991 [1..max_snp_asid] and the SEV-ES range being
2992 (max_snp_asid..min_sev_asid), where min_sev_asid is
2993 enumerated by CPUID.0x.8000_001F[EDX].
2995 A non-zero value enables SEV-SNP ciphertext hiding and
2996 adjusts the ASID ranges for SEV-ES and SEV-SNP guests.
2997 KVM caps the number of SEV-SNP ASIDs at the maximum
2998 possible value, e.g. specifying -1u will assign all
2999 joint SEV-ES and SEV-SNP ASIDs to SEV-SNP. Note,
3000 assigning all joint ASIDs to SEV-SNP, i.e. configuring
3001 max_snp_asid == min_sev_asid-1, will effectively make
3002 SEV-ES unusable.
3004 kvm-arm.mode=
3005 [KVM,ARM,EARLY] Select one of KVM/arm64's modes of
3006 operation.
3008 none: Forcefully disable KVM.
3010 nvhe: Standard nVHE-based mode, without support for
3011 protected guests.
3013 protected: Mode with support for guests whose state is
3014 kept private from the host, using VHE or
3015 nVHE depending on HW support.
3017 nested: VHE-based mode with support for nested
3018 virtualization. Requires at least ARMv8.4
3019 hardware (with FEAT_NV2).
3021 Defaults to VHE/nVHE based on hardware support. Setting
3022 mode to "protected" will disable kexec and hibernation
3023 for the host. To force nVHE on VHE hardware, add
3024 "arm64_sw.hvhe=0 id_aa64mmfr1.vh=0" to the
3025 command-line.
3026 "nested" is experimental and should be used with
3027 extreme caution.
3029 kvm-arm.vgic_v3_group0_trap=
3030 [KVM,ARM,EARLY] Trap guest accesses to GICv3 group-0
3031 system registers
3033 kvm-arm.vgic_v3_group1_trap=
3034 [KVM,ARM,EARLY] Trap guest accesses to GICv3 group-1
3035 system registers
3037 kvm-arm.vgic_v3_common_trap=
3038 [KVM,ARM,EARLY] Trap guest accesses to GICv3 common
3039 system registers
3041 kvm-arm.vgic_v4_enable=
3042 [KVM,ARM,EARLY] Allow use of GICv4 for direct
3043 injection of LPIs.
3045 kvm-arm.wfe_trap_policy=
3046 [KVM,ARM] Control when to set WFE instruction trap for
3047 KVM VMs. Traps are allowed but not guaranteed by the
3048 CPU architecture.
3050 trap: set WFE instruction trap
3052 notrap: clear WFE instruction trap
3054 kvm-arm.wfi_trap_policy=
3055 [KVM,ARM] Control when to set WFI instruction trap for
3056 KVM VMs. Traps are allowed but not guaranteed by the
3057 CPU architecture.
3059 trap: set WFI instruction trap
3061 notrap: clear WFI instruction trap
3063 kvm_cma_resv_ratio=n [PPC,EARLY]
3064 Reserves given percentage from system memory area for
3065 contiguous memory allocation for KVM hash pagetable
3066 allocation.
3067 By default it reserves 5% of total system memory.
3068 Format: <integer>
3069 Default: 5
3071 kvm-intel.ept= [KVM,Intel] Control KVM's use of Extended Page Tables,
3072 a.k.a. Two-Dimensional Page Tables. Default is 1
3073 (enabled). Disable by KVM if hardware lacks support
3074 for EPT.
3076 kvm-intel.emulate_invalid_guest_state=
3077 [KVM,Intel] Control whether to emulate invalid guest
3078 state. Ignored if kvm-intel.enable_unrestricted_guest=1,
3079 as guest state is never invalid for unrestricted
3080 guests. This param doesn't apply to nested guests (L2),
3081 as KVM never emulates invalid L2 guest state.
3082 Default is 1 (enabled).
3084 kvm-intel.flexpriority=
3085 [KVM,Intel] Control KVM's use of FlexPriority feature
3086 (TPR shadow). Default is 1 (enabled). Disable by KVM if
3087 hardware lacks support for it.
3089 kvm-intel.nested=
3090 [KVM,Intel] Control nested virtualization feature in
3091 KVM/VMX. Default is 1 (enabled).
3093 kvm-intel.unrestricted_guest=
3094 [KVM,Intel] Control KVM's use of unrestricted guest
3095 feature (virtualized real and unpaged mode). Default
3096 is 1 (enabled). Disable by KVM if EPT is disabled or
3097 hardware lacks support for it.
3099 kvm-intel.vmentry_l1d_flush=[KVM,Intel] Mitigation for L1 Terminal Fault
3100 CVE-2018-3620.
3102 Valid arguments: never, cond, always
3104 always: L1D cache flush on every VMENTER.
3105 cond: Flush L1D on VMENTER only when the code between
3106 VMEXIT and VMENTER can leak host memory.
3107 never: Disables the mitigation
3109 Default is cond (do L1 cache flush in specific instances)
3111 kvm-intel.vpid= [KVM,Intel] Control KVM's use of Virtual Processor
3112 Identification feature (tagged TLBs). Default is 1
3113 (enabled). Disable by KVM if hardware lacks support
3114 for it.
3116 l1d_flush= [X86,INTEL,EARLY]
3117 Control mitigation for L1D based snooping vulnerability.
3119 Certain CPUs are vulnerable to an exploit against CPU
3120 internal buffers which can forward information to a
3121 disclosure gadget under certain conditions.
3123 In vulnerable processors, the speculatively
3124 forwarded data can be used in a cache side channel
3125 attack, to access data to which the attacker does
3126 not have direct access.
3128 This parameter controls the mitigation. The
3129 options are:
3131 on - enable the interface for the mitigation
3133 l1tf= [X86,EARLY] Control mitigation of the L1TF vulnerability on
3134 affected CPUs
3136 The kernel PTE inversion protection is unconditionally
3137 enabled and cannot be disabled.
3139 full
3140 Provides all available mitigations for the
3141 L1TF vulnerability. Disables SMT and
3142 enables all mitigations in the
3143 hypervisors, i.e. unconditional L1D flush.
3145 SMT control and L1D flush control via the
3146 sysfs interface is still possible after
3147 boot. Hypervisors will issue a warning
3148 when the first VM is started in a
3149 potentially insecure configuration,
3150 i.e. SMT enabled or L1D flush disabled.
3152 full,force
3153 Same as 'full', but disables SMT and L1D
3154 flush runtime control. Implies the
3155 'nosmt=force' command line option.
3156 (i.e. sysfs control of SMT is disabled.)
3158 flush
3159 Leaves SMT enabled and enables the default
3160 hypervisor mitigation, i.e. conditional
3161 L1D flush.
3163 SMT control and L1D flush control via the
3164 sysfs interface is still possible after
3165 boot. Hypervisors will issue a warning
3166 when the first VM is started in a
3167 potentially insecure configuration,
3168 i.e. SMT enabled or L1D flush disabled.
3170 flush,nosmt
3172 Disables SMT and enables the default
3173 hypervisor mitigation.
3175 SMT control and L1D flush control via the
3176 sysfs interface is still possible after
3177 boot. Hypervisors will issue a warning
3178 when the first VM is started in a
3179 potentially insecure configuration,
3180 i.e. SMT enabled or L1D flush disabled.
3182 flush,nowarn
3183 Same as 'flush', but hypervisors will not
3184 warn when a VM is started in a potentially
3185 insecure configuration.
3187 off
3188 Disables hypervisor mitigations and doesn't
3189 emit any warnings.
3190 It also drops the swap size and available
3191 RAM limit restriction on both hypervisor and
3192 bare metal.
3194 Default is 'flush'.
3196 For details see: Documentation/admin-guide/hw-vuln/l1tf.rst
3198 l2cr= [PPC]
3200 l3cr= [PPC]
3202 lapic [X86-32,APIC,EARLY] Enable the local APIC even if BIOS
3203 disabled it.
3205 lapic= [X86,APIC] Do not use TSC deadline
3206 value for LAPIC timer one-shot implementation. Default
3207 back to the programmable timer unit in the LAPIC.
3208 Format: notscdeadline
3210 lapic_timer_c2_ok [X86,APIC,EARLY] trust the local apic timer
3211 in C2 power state.
3213 libata.dma= [LIBATA] DMA control
3214 libata.dma=0 Disable all PATA and SATA DMA
3215 libata.dma=1 PATA and SATA Disk DMA only
3216 libata.dma=2 ATAPI (CDROM) DMA only
3217 libata.dma=4 Compact Flash DMA only
3218 Combinations also work, so libata.dma=3 enables DMA
3219 for disks and CDROMs, but not CFs.
3221 libata.ignore_hpa= [LIBATA] Ignore HPA limit
3222 libata.ignore_hpa=0 keep BIOS limits (default)
3223 libata.ignore_hpa=1 ignore limits, using full disk
3225 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume
3226 when set.
3227 Format: <int>
3229 libata.force= [LIBATA] Force configurations. The format is a comma-
3230 separated list of "[ID:]VAL" where ID is PORT[.DEVICE].
3231 PORT and DEVICE are decimal numbers matching port, link
3232 or device. Basically, it matches the ATA ID string
3233 printed on console by libata. If the whole ID part is
3234 omitted, the last PORT and DEVICE values are used. If
3235 ID hasn't been specified yet, the configuration applies
3236 to all ports, links and devices.
3238 If only DEVICE is omitted, the parameter applies to
3239 the port and all links and devices behind it. DEVICE
3240 number of 0 either selects the first device or the
3241 first fan-out link behind PMP device. It does not
3242 select the host link. DEVICE number of 15 selects the
3243 host link and device attached to it.
3245 The VAL specifies the configuration to force. As long
3246 as there is no ambiguity, shortcut notation is allowed.
3247 For example, both 1.5 and 1.5G would work for 1.5Gbps.
3248 The following configurations can be forced.
3250 * Cable type: 40c, 80c, short40c, unk, ign or sata.
3251 Any ID with matching PORT is used.
3253 * SATA link speed limit: 1.5Gbps or 3.0Gbps.
3255 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7].
3256 udma[/][16,25,33,44,66,100,133] notation is also
3257 allowed.
3259 * nohrst, nosrst, norst: suppress hard, soft and both
3260 resets.
3262 * rstonce: only attempt one reset during hot-unplug
3263 link recovery.
3265 * [no]dbdelay: Enable or disable the extra 200ms delay
3266 before debouncing a link PHY and device presence
3267 detection.
3269 * [no]ncq: Turn on or off NCQ.
3271 * [no]ncqtrim: Enable or disable queued DSM TRIM.
3273 * [no]ncqati: Enable or disable NCQ trim on ATI chipset.
3275 * [no]trim: Enable or disable (unqueued) TRIM.
3277 * trim_zero: Indicate that TRIM command zeroes data.
3279 * max_trim_128m: Set 128M maximum trim size limit.
3281 * [no]dma: Turn on or off DMA transfers.
3283 * atapi_dmadir: Enable ATAPI DMADIR bridge support.
3285 * atapi_mod16_dma: Enable the use of ATAPI DMA for
3286 commands that are not a multiple of 16 bytes.
3288 * [no]dmalog: Enable or disable the use of the
3289 READ LOG DMA EXT command to access logs.
3291 * [no]iddevlog: Enable or disable access to the
3292 identify device data log.
3294 * [no]logdir: Enable or disable access to the general
3295 purpose log directory.
3297 * max_sec_128: Set transfer size limit to 128 sectors.
3299 * max_sec_1024: Set or clear transfer size limit to
3300 1024 sectors.
3302 * max_sec_lba48: Set or clear transfer size limit to
3303 65535 sectors.
3305 * external: Mark port as external (hotplug-capable).
3307 * [no]lpm: Enable or disable link power management.
3309 * [no]setxfer: Indicate if transfer speed mode setting
3310 should be skipped.
3312 * [no]fua: Disable or enable FUA (Force Unit Access)
3313 support for devices supporting this feature.
3315 * dump_id: Dump IDENTIFY data.
3317 * disable: Disable this device.
3319 If there are multiple matching configurations changing
3320 the same attribute, the last one is used.
3322 load_ramdisk= [RAM] [Deprecated]
3324 lockd.nlm_grace_period=P [NFS] Assign grace period.
3325 Format: <integer>
3327 lockd.nlm_tcpport=N [NFS] Assign TCP port.
3328 Format: <integer>
3330 lockd.nlm_timeout=T [NFS] Assign timeout value.
3331 Format: <integer>
3333 lockd.nlm_udpport=M [NFS] Assign UDP port.
3334 Format: <integer>
3336 lockdown= [SECURITY,EARLY]
3337 { integrity | confidentiality }
3338 Enable the kernel lockdown feature. If set to
3339 integrity, kernel features that allow userland to
3340 modify the running kernel are disabled. If set to
3341 confidentiality, kernel features that allow userland
3342 to extract confidential information from the kernel
3343 are also disabled.
3345 locktorture.acq_writer_lim= [KNL]
3346 Set the time limit in jiffies for a lock
3347 acquisition. Acquisitions exceeding this limit
3348 will result in a splat once they do complete.
3350 locktorture.bind_readers= [KNL]
3351 Specify the list of CPUs to which the readers are
3352 to be bound.
3354 locktorture.bind_writers= [KNL]
3355 Specify the list of CPUs to which the writers are
3356 to be bound.
3358 locktorture.call_rcu_chains= [KNL]
3359 Specify the number of self-propagating call_rcu()
3360 chains to set up. These are used to ensure that
3361 there is a high probability of an RCU grace period
3362 in progress at any given time. Defaults to 0,
3363 which disables these call_rcu() chains.
3365 locktorture.long_hold= [KNL]
3366 Specify the duration in milliseconds for the
3367 occasional long-duration lock hold time. Defaults
3368 to 100 milliseconds. Select 0 to disable.
3370 locktorture.nested_locks= [KNL]
3371 Specify the maximum lock nesting depth that
3372 locktorture is to exercise, up to a limit of 8
3373 (MAX_NESTED_LOCKS). Specify zero to disable.
3374 Note that this parameter is ineffective on types
3375 of locks that do not support nested acquisition.
3377 locktorture.nreaders_stress= [KNL]
3378 Set the number of locking read-acquisition kthreads.
3379 Defaults to being automatically set based on the
3380 number of online CPUs.
3382 locktorture.nwriters_stress= [KNL]
3383 Set the number of locking write-acquisition kthreads.
3385 locktorture.onoff_holdoff= [KNL]
3386 Set time (s) after boot for CPU-hotplug testing.
3388 locktorture.onoff_interval= [KNL]
3389 Set time (s) between CPU-hotplug operations, or
3390 zero to disable CPU-hotplug testing.
3392 locktorture.rt_boost= [KNL]
3393 Do periodic testing of real-time lock priority
3394 boosting. Select 0 to disable, 1 to boost
3395 only rt_mutex, and 2 to boost unconditionally.
3396 Defaults to 2, which might seem to be an
3397 odd choice, but which should be harmless for
3398 non-real-time spinlocks, due to their disabling
3399 of preemption. Note that non-realtime mutexes
3400 disable boosting.
3402 locktorture.rt_boost_factor= [KNL]
3403 Number that determines how often and for how
3404 long priority boosting is exercised. This is
3405 scaled down by the number of writers, so that the
3406 number of boosts per unit time remains roughly
3407 constant as the number of writers increases.
3408 On the other hand, the duration of each boost
3409 increases with the number of writers.
3411 locktorture.shuffle_interval= [KNL]
3412 Set task-shuffle interval (jiffies). Shuffling
3413 tasks allows some CPUs to go into dyntick-idle
3414 mode during the locktorture test.
3416 locktorture.shutdown_secs= [KNL]
3417 Set time (s) after boot system shutdown. This
3418 is useful for hands-off automated testing.
3420 locktorture.stat_interval= [KNL]
3421 Time (s) between statistics printk()s.
3423 locktorture.stutter= [KNL]
3424 Time (s) to stutter testing, for example,
3425 specifying five seconds causes the test to run for
3426 five seconds, wait for five seconds, and so on.
3427 This tests the locking primitive's ability to
3428 transition abruptly to and from idle.
3430 locktorture.torture_type= [KNL]
3431 Specify the locking implementation to test.
3433 locktorture.verbose= [KNL]
3434 Enable additional printk() statements.
3436 locktorture.writer_fifo= [KNL]
3437 Run the write-side locktorture kthreads at
3438 sched_set_fifo() real-time priority.
3440 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver
3441 Format: <irq>
3443 loglevel= [KNL,EARLY]
3444 All Kernel Messages with a loglevel smaller than the
3445 console loglevel will be printed to the console. It can
3446 also be changed with klogd or other programs. The
3447 loglevels are defined as follows:
3449 0 (KERN_EMERG) system is unusable
3450 1 (KERN_ALERT) action must be taken immediately
3451 2 (KERN_CRIT) critical conditions
3452 3 (KERN_ERR) error conditions
3453 4 (KERN_WARNING) warning conditions
3454 5 (KERN_NOTICE) normal but significant condition
3455 6 (KERN_INFO) informational
3456 7 (KERN_DEBUG) debug-level messages
3458 log_buf_len=n[KMG] [KNL,EARLY]
3459 Sets the size of the printk ring buffer, in bytes.
3460 n must be a power of two and greater than the
3461 minimal size. The minimal size is defined by
3462 LOG_BUF_SHIFT kernel config parameter. There
3463 is also CONFIG_LOG_CPU_MAX_BUF_SHIFT config
3464 parameter that allows to increase the default size
3465 depending on the number of CPUs. See init/Kconfig
3466 for more details.
3468 logo.nologo [FB] Disables display of the built-in Linux logo.
3469 This may be used to provide more screen space for
3470 kernel log messages and is useful when debugging
3471 kernel boot problems.
3473 lp=0 [LP] Specify parallel ports to use, e.g,
3474 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses
3475 lp=reset first parallel port). 'lp=0' disables the
3476 lp=auto printer driver. 'lp=reset' (which can be
3477 specified in addition to the ports) causes
3478 attached printers to be reset. Using
3479 lp=port1,port2,... specifies the parallel ports
3480 to associate lp devices with, starting with
3481 lp0. A port specification may be 'none' to skip
3482 that lp device, or a parport name such as
3483 'parport0'. Specifying 'lp=auto' instead of a
3484 port specification list means that device IDs
3485 from each port should be examined, to see if
3486 an IEEE 1284-compliant printer is attached; if
3487 so, the driver will manage that printer.
3488 See also header of drivers/char/lp.c.
3490 lpj=n [KNL]
3491 Sets loops_per_jiffy to given constant, thus avoiding
3492 time-consuming boot-time autodetection (up to 250 ms per
3493 CPU). 0 enables autodetection (default). To determine
3494 the correct value for your kernel, boot with normal
3495 autodetection and see what value is printed. Note that
3496 on SMP systems the preset will be applied to all CPUs,
3497 which is likely to cause problems if your CPUs need
3498 significantly divergent settings. An incorrect value
3499 will cause delays in the kernel to be wrong, leading to
3500 unpredictable I/O errors and other breakage. Although
3501 unlikely, in the extreme case this might damage your
3502 hardware.
3504 lsm.debug [SECURITY] Enable LSM initialization debugging output.
3506 lsm=lsm1,...,lsmN
3507 [SECURITY] Choose order of LSM initialization. This
3508 overrides CONFIG_LSM, and the "security=" parameter.
3510 machtype= [Loongson] Share the same kernel image file between
3511 different yeeloong laptops.
3512 Example: machtype=lemote-yeeloong-2f-7inch
3514 maxcpus= [SMP,EARLY] Maximum number of processors that an SMP kernel
3515 will bring up during bootup. maxcpus=n : n >= 0 limits
3516 the kernel to bring up 'n' processors. Surely after
3517 bootup you can bring up the other plugged cpu by executing
3518 "echo 1 > /sys/devices/system/cpu/cpuX/online". So maxcpus
3519 only takes effect during system bootup.
3520 While n=0 is a special case, it is equivalent to "nosmp",
3521 which also disables the IO APIC.
3523 max_loop= [LOOP] The number of loop block devices that get
3524 (loop.max_loop) unconditionally pre-created at init time. The default
3525 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead
3526 of statically allocating a predefined number, loop
3527 devices can be requested on-demand with the
3528 /dev/loop-control interface.
3530 mce= [X86-{32,64}]
3532 Please see Documentation/arch/x86/x86_64/machinecheck.rst for sysfs runtime tunables.
3534 off
3535 disable machine check
3537 no_cmci
3538 disable CMCI(Corrected Machine Check Interrupt) that
3539 Intel processor supports. Usually this disablement is
3540 not recommended, but it might be handy if your
3541 hardware is misbehaving.
3543 Note that you'll get more problems without CMCI than
3544 with due to the shared banks, i.e. you might get
3545 duplicated error logs.
3547 dont_log_ce
3548 don't make logs for corrected errors. All events
3549 reported as corrected are silently cleared by OS. This
3550 option will be useful if you have no interest in any
3551 of corrected errors.
3553 ignore_ce
3554 disable features for corrected errors, e.g.
3555 polling timer and CMCI. All events reported as
3556 corrected are not cleared by OS and remained in its
3557 error banks.
3559 Usually this disablement is not recommended, however
3560 if there is an agent checking/clearing corrected
3561 errors (e.g. BIOS or hardware monitoring
3562 applications), conflicting with OS's error handling,
3563 and you cannot deactivate the agent, then this option
3564 will be a help.
3566 no_lmce
3567 do not opt-in to Local MCE delivery. Use legacy method
3568 to broadcast MCEs.
3570 bootlog
3571 enable logging of machine checks left over from
3572 booting. Disabled by default on AMD Fam10h and older
3573 because some BIOS leave bogus ones.
3575 If your BIOS doesn't do that it's a good idea to
3576 enable though to make sure you log even machine check
3577 events that result in a reboot. On Intel systems it is
3578 enabled by default.
3580 nobootlog
3581 disable boot machine check logging.
3583 monarchtimeout (number)
3584 sets the time in us to wait for other CPUs on machine
3585 checks. 0 to disable.
3587 bios_cmci_threshold
3588 don't overwrite the bios-set CMCI threshold. This boot
3589 option prevents Linux from overwriting the CMCI
3590 threshold set by the bios. Without this option, Linux
3591 always sets the CMCI threshold to 1. Enabling this may
3592 make memory predictive failure analysis less effective
3593 if the bios sets thresholds for memory errors since we
3594 will not see details for all errors.
3596 recovery
3597 force-enable recoverable machine check code paths
3599 Everything else is in sysfs now.
3602 md= [HW] RAID subsystems devices and level
3603 See Documentation/admin-guide/md.rst.
3605 mdacon= [MDA]
3606 Format: <first>,<last>
3607 Specifies range of consoles to be captured by the MDA.
3609 mds= [X86,INTEL,EARLY]
3610 Control mitigation for the Micro-architectural Data
3611 Sampling (MDS) vulnerability.
3613 Certain CPUs are vulnerable to an exploit against CPU
3614 internal buffers which can forward information to a
3615 disclosure gadget under certain conditions.
3617 In vulnerable processors, the speculatively
3618 forwarded data can be used in a cache side channel
3619 attack, to access data to which the attacker does
3620 not have direct access.
3622 This parameter controls the MDS mitigation. The
3623 options are:
3625 full - Enable MDS mitigation on vulnerable CPUs
3626 full,nosmt - Enable MDS mitigation and disable
3627 SMT on vulnerable CPUs
3628 off - Unconditionally disable MDS mitigation
3630 On TAA-affected machines, mds=off can be prevented by
3631 an active TAA mitigation as both vulnerabilities are
3632 mitigated with the same mechanism so in order to disable
3633 this mitigation, you need to specify tsx_async_abort=off
3634 too.
3636 Not specifying this option is equivalent to
3637 mds=full.
3639 For details see: Documentation/admin-guide/hw-vuln/mds.rst
3641 mem=nn[KMG] [HEXAGON,EARLY] Set the memory size.
3642 Must be specified, otherwise memory size will be 0.
3644 mem=nn[KMG] [KNL,BOOT,EARLY] Force usage of a specific amount
3645 of memory Amount of memory to be used in cases
3646 as follows:
3648 1 for test;
3649 2 when the kernel is not able to see the whole system memory;
3650 3 memory that lies after 'mem=' boundary is excluded from
3651 the hypervisor, then assigned to KVM guests.
3652 4 to limit the memory available for kdump kernel.
3654 [ARC,MICROBLAZE] - the limit applies only to low memory,
3655 high memory is not affected.
3657 [ARM64] - only limits memory covered by the linear
3658 mapping. The NOMAP regions are not affected.
3660 [X86] Work as limiting max address. Use together
3661 with memmap= to avoid physical address space collisions.
3662 Without memmap= PCI devices could be placed at addresses
3663 belonging to unused RAM.
3665 Note that this only takes effects during boot time since
3666 in above case 3, memory may need be hot added after boot
3667 if system memory of hypervisor is not sufficient.
3669 mem=nn[KMG]@ss[KMG]
3670 [ARM,MIPS,EARLY] - override the memory layout
3671 reported by firmware.
3672 Define a memory region of size nn[KMG] starting at
3673 ss[KMG].
3674 Multiple different regions can be specified with
3675 multiple mem= parameters on the command line.
3677 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel
3678 memory.
3680 memblock=debug [KNL,EARLY] Enable memblock debug messages.
3682 memchunk=nn[KMG]
3683 [KNL,SH] Allow user to override the default size for
3684 per-device physically contiguous DMA buffers.
3686 memhp_default_state=online/offline/online_kernel/online_movable
3687 [KNL] Set the initial state for the memory hotplug
3688 onlining policy. If not specified, the default value is
3689 set according to the
3690 CONFIG_MHP_DEFAULT_ONLINE_TYPE kernel config
3691 options.
3692 See Documentation/admin-guide/mm/memory-hotplug.rst.
3694 memmap=exactmap [KNL,X86,EARLY] Enable setting of an exact
3695 E820 memory map, as specified by the user.
3696 Such memmap=exactmap lines can be constructed based on
3697 BIOS output or other requirements. See the memmap=nn@ss
3698 option description.
3700 memmap=nn[KMG]@ss[KMG]
3701 [KNL, X86,MIPS,XTENSA,EARLY] Force usage of a specific region of memory.
3702 Region of memory to be used is from ss to ss+nn.
3703 If @ss[KMG] is omitted, it is equivalent to mem=nn[KMG],
3704 which limits max address to nn[KMG].
3705 Multiple different regions can be specified,
3706 comma delimited.
3707 Example:
3708 memmap=100M@2G,100M#3G,1G!1024G
3710 memmap=nn[KMG]#ss[KMG]
3711 [KNL,ACPI,EARLY] Mark specific memory as ACPI data.
3712 Region of memory to be marked is from ss to ss+nn.
3714 memmap=nn[KMG]$ss[KMG]
3715 [KNL,ACPI,EARLY] Mark specific memory as reserved.
3716 Region of memory to be reserved is from ss to ss+nn.
3717 Example: Exclude memory from 0x18690000-0x1869ffff
3718 memmap=64K$0x18690000
3719 or
3720 memmap=0x10000$0x18690000
3721 Some bootloaders may need an escape character before '$',
3722 like Grub2, otherwise '$' and the following number
3723 will be eaten.
3725 memmap=nn[KMG]!ss[KMG,EARLY]
3726 [KNL,X86] Mark specific memory as protected.
3727 Region of memory to be used, from ss to ss+nn.
3728 The memory region may be marked as e820 type 12 (0xc)
3729 and is NVDIMM or ADR memory.
3731 memmap=<size>%<offset>-<oldtype>+<newtype>
3732 [KNL,ACPI,EARLY] Convert memory within the specified region
3733 from <oldtype> to <newtype>. If "-<oldtype>" is left
3734 out, the whole region will be marked as <newtype>,
3735 even if previously unavailable. If "+<newtype>" is left
3736 out, matching memory will be removed. Types are
3737 specified as e820 types, e.g., 1 = RAM, 2 = reserved,
3738 3 = ACPI, 12 = PRAM.
3740 memory_corruption_check=0/1 [X86,EARLY]
3741 Some BIOSes seem to corrupt the first 64k of
3742 memory when doing things like suspend/resume.
3743 Setting this option will scan the memory
3744 looking for corruption. Enabling this will
3745 both detect corruption and prevent the kernel
3746 from using the memory being corrupted.
3747 However, it's intended as a diagnostic tool; if
3748 repeatable BIOS-originated corruption always
3749 affects the same memory, you can use memmap=
3750 to prevent the kernel from using that memory.
3752 memory_corruption_check_size=size [X86,EARLY]
3753 By default it checks for corruption in the low
3754 64k, making this memory unavailable for normal
3755 use. Use this parameter to scan for
3756 corruption in more or less memory.
3758 memory_corruption_check_period=seconds [X86,EARLY]
3759 By default it checks for corruption every 60
3760 seconds. Use this parameter to check at some
3761 other rate. 0 disables periodic checking.
3763 memory_hotplug.memmap_on_memory
3764 [KNL,X86,ARM] Boolean flag to enable this feature.
3765 Format: {on | off (default)}
3766 When enabled, runtime hotplugged memory will
3767 allocate its internal metadata (struct pages,
3768 those vmemmap pages cannot be optimized even
3769 if hugetlb_free_vmemmap is enabled) from the
3770 hotadded memory which will allow to hotadd a
3771 lot of memory without requiring additional
3772 memory to do so.
3773 This feature is disabled by default because it
3774 has some implication on large (e.g. GB)
3775 allocations in some configurations (e.g. small
3776 memory blocks).
3777 The state of the flag can be read in
3778 /sys/module/memory_hotplug/parameters/memmap_on_memory.
3779 Note that even when enabled, there are a few cases where
3780 the feature is not effective.
3782 memtest= [KNL,X86,ARM,M68K,PPC,RISCV,EARLY] Enable memtest
3783 Format: <integer>
3784 default : 0 <disable>
3785 Specifies the number of memtest passes to be
3786 performed. Each pass selects another test
3787 pattern from a given set of patterns. Memtest
3788 fills the memory with this pattern, validates
3789 memory contents and reserves bad memory
3790 regions that are detected.
3792 mem_encrypt= [X86-64] AMD Secure Memory Encryption (SME) control
3793 Valid arguments: on, off
3794 Default: off
3795 mem_encrypt=on: Activate SME
3796 mem_encrypt=off: Do not activate SME
3798 Refer to Documentation/virt/kvm/x86/amd-memory-encryption.rst
3799 for details on when memory encryption can be activated.
3801 mem_sleep_default= [SUSPEND] Default system suspend mode:
3802 s2idle - Suspend-To-Idle
3803 shallow - Power-On Suspend or equivalent (if supported)
3804 deep - Suspend-To-RAM or equivalent (if supported)
3805 See Documentation/admin-guide/pm/sleep-states.rst.
3807 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when
3808 the BIOS has incorrectly applied a workaround. TinyBIOS
3809 version 0.98 is known to be affected, 0.99 fixes the
3810 problem by letting the user disable the workaround.
3812 mga= [HW,DRM]
3814 microcode= [X86] Control the behavior of the microcode loader.
3815 Available options, comma separated:
3817 base_rev=X - with <X> with format: <u32>
3818 Set the base microcode revision of each thread when in
3819 debug mode.
3821 dis_ucode_ldr: disable the microcode loader
3823 force_minrev:
3824 Enable or disable the microcode minimal revision
3825 enforcement for the runtime microcode loader.
3827 mini2440= [ARM,HW,KNL]
3828 Format:[0..2][b][c][t]
3829 Default: "0tb"
3830 MINI2440 configuration specification:
3831 0 - The attached screen is the 3.5" TFT
3832 1 - The attached screen is the 7" TFT
3833 2 - The VGA Shield is attached (1024x768)
3834 Leaving out the screen size parameter will not load
3835 the TFT driver, and the framebuffer will be left
3836 unconfigured.
3837 b - Enable backlight. The TFT backlight pin will be
3838 linked to the kernel VESA blanking code and a GPIO
3839 LED. This parameter is not necessary when using the
3840 VGA shield.
3841 c - Enable the s3c camera interface.
3842 t - Reserved for enabling touchscreen support. The
3843 touchscreen support is not enabled in the mainstream
3844 kernel as of 2.6.30, a preliminary port can be found
3845 in the "bleeding edge" mini2440 support kernel at
3846 https://repo.or.cz/w/linux-2.6/mini2440.git
3848 mitigations=
3849 [X86,PPC,S390,ARM64,EARLY] Control optional mitigations for
3850 CPU vulnerabilities. This is a set of curated,
3851 arch-independent options, each of which is an
3852 aggregation of existing arch-specific options.
3854 Note, "mitigations" is supported if and only if the
3855 kernel was built with CPU_MITIGATIONS=y.
3857 off
3858 Disable all optional CPU mitigations. This
3859 improves system performance, but it may also
3860 expose users to several CPU vulnerabilities.
3861 Equivalent to: if nokaslr then kpti=0 [ARM64]
3862 gather_data_sampling=off [X86]
3863 indirect_target_selection=off [X86]
3864 kvm.nx_huge_pages=off [X86]
3865 l1tf=off [X86]
3866 mds=off [X86]
3867 mmio_stale_data=off [X86]
3868 no_entry_flush [PPC]
3869 no_uaccess_flush [PPC]
3870 nobp=0 [S390]
3871 nopti [X86,PPC]
3872 nospectre_bhb [ARM64]
3873 nospectre_v1 [X86,PPC]
3874 nospectre_v2 [X86,PPC,S390,ARM64]
3875 reg_file_data_sampling=off [X86]
3876 retbleed=off [X86]
3877 spec_rstack_overflow=off [X86]
3878 spec_store_bypass_disable=off [X86,PPC]
3879 spectre_bhi=off [X86]
3880 spectre_v2_user=off [X86]
3881 srbds=off [X86,INTEL]
3882 ssbd=force-off [ARM64]
3883 tsx_async_abort=off [X86]
3884 vmscape=off [X86]
3886 Exceptions:
3887 This does not have any effect on
3888 kvm.nx_huge_pages when
3889 kvm.nx_huge_pages=force.
3891 auto (default)
3892 Mitigate all CPU vulnerabilities, but leave SMT
3893 enabled, even if it's vulnerable. This is for
3894 users who don't want to be surprised by SMT
3895 getting disabled across kernel upgrades, or who
3896 have other ways of avoiding SMT-based attacks.
3897 Equivalent to: (default behavior)
3899 auto,nosmt
3900 Mitigate all CPU vulnerabilities, disabling SMT
3901 if needed. This is for users who always want to
3902 be fully mitigated, even if it means losing SMT.
3903 Equivalent to: l1tf=flush,nosmt [X86]
3904 mds=full,nosmt [X86]
3905 tsx_async_abort=full,nosmt [X86]
3906 mmio_stale_data=full,nosmt [X86]
3907 retbleed=auto,nosmt [X86]
3909 [X86] After one of the above options, additionally
3910 supports attack-vector based controls as documented in
3911 Documentation/admin-guide/hw-vuln/attack_vector_controls.rst
3913 mminit_loglevel=
3914 [KNL,EARLY] When CONFIG_DEBUG_MEMORY_INIT is set, this
3915 parameter allows control of the logging verbosity for
3916 the additional memory initialisation checks. A value
3917 of 0 disables mminit logging and a level of 4 will
3918 log everything. Information is printed at KERN_DEBUG
3919 so loglevel=8 may also need to be specified.
3921 mmio_stale_data=
3922 [X86,INTEL,EARLY] Control mitigation for the Processor
3923 MMIO Stale Data vulnerabilities.
3925 Processor MMIO Stale Data is a class of
3926 vulnerabilities that may expose data after an MMIO
3927 operation. Exposed data could originate or end in
3928 the same CPU buffers as affected by MDS and TAA.
3929 Therefore, similar to MDS and TAA, the mitigation
3930 is to clear the affected CPU buffers.
3932 This parameter controls the mitigation. The
3933 options are:
3935 full - Enable mitigation on vulnerable CPUs
3937 full,nosmt - Enable mitigation and disable SMT on
3938 vulnerable CPUs.
3940 off - Unconditionally disable mitigation
3942 On MDS or TAA affected machines,
3943 mmio_stale_data=off can be prevented by an active
3944 MDS or TAA mitigation as these vulnerabilities are
3945 mitigated with the same mechanism so in order to
3946 disable this mitigation, you need to specify
3947 mds=off and tsx_async_abort=off too.
3949 Not specifying this option is equivalent to
3950 mmio_stale_data=full.
3952 For details see:
3953 Documentation/admin-guide/hw-vuln/processor_mmio_stale_data.rst
3955 <module>.async_probe[=<bool>] [KNL]
3956 If no <bool> value is specified or if the value
3957 specified is not a valid <bool>, enable asynchronous
3958 probe on this module. Otherwise, enable/disable
3959 asynchronous probe on this module as indicated by the
3960 <bool> value. See also: module.async_probe
3962 module.async_probe=<bool>
3963 [KNL] When set to true, modules will use async probing
3964 by default. To enable/disable async probing for a
3965 specific module, use the module specific control that
3966 is documented under <module>.async_probe. When both
3967 module.async_probe and <module>.async_probe are
3968 specified, <module>.async_probe takes precedence for
3969 the specific module.
3971 module.enable_dups_trace
3972 [KNL] When CONFIG_MODULE_DEBUG_AUTOLOAD_DUPS is set,
3973 this means that duplicate request_module() calls will
3974 trigger a WARN_ON() instead of a pr_warn(). Note that
3975 if MODULE_DEBUG_AUTOLOAD_DUPS_TRACE is set, WARN_ON()s
3976 will always be issued and this option does nothing.
3977 module.sig_enforce
3978 [KNL] When CONFIG_MODULE_SIG is set, this means that
3979 modules without (valid) signatures will fail to load.
3980 Note that if CONFIG_MODULE_SIG_FORCE is set, that
3981 is always true, so this option does nothing.
3983 module_blacklist= [KNL] Do not load a comma-separated list of
3984 modules. Useful for debugging problem modules.
3986 mousedev.tap_time=
3987 [MOUSE] Maximum time between finger touching and
3988 leaving touchpad surface for touch to be considered
3989 a tap and be reported as a left button click (for
3990 touchpads working in absolute mode only).
3991 Format: <msecs>
3992 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices
3993 reporting absolute coordinates, such as tablets
3994 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices
3995 reporting absolute coordinates, such as tablets
3997 movablecore= [KNL,X86,PPC,EARLY]
3998 Format: nn[KMGTPE] | nn%
3999 This parameter is the complement to kernelcore=, it
4000 specifies the amount of memory used for migratable
4001 allocations. If both kernelcore and movablecore is
4002 specified, then kernelcore will be at *least* the
4003 specified value but may be more. If movablecore on its
4004 own is specified, the administrator must be careful
4005 that the amount of memory usable for all allocations
4006 is not too small.
4008 movable_node [KNL,EARLY] Boot-time switch to make hotplugable memory
4009 NUMA nodes to be movable. This means that the memory
4010 of such nodes will be usable only for movable
4011 allocations which rules out almost all kernel
4012 allocations. Use with caution!
4014 MTD_Partition= [MTD]
4015 Format: <name>,<region-number>,<size>,<offset>
4017 MTD_Region= [MTD] Format:
4018 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>]
4020 mtdparts= [MTD]
4021 See drivers/mtd/parsers/cmdlinepart.c
4023 mtouchusb.raw_coordinates=
4024 [HW] Make the MicroTouch USB driver use raw coordinates
4025 ('y', default) or cooked coordinates ('n')
4027 mtrr=debug [X86,EARLY]
4028 Enable printing debug information related to MTRR
4029 registers at boot time.
4031 mtrr_chunk_size=nn[KMG,X86,EARLY]
4032 used for mtrr cleanup. It is largest continuous chunk
4033 that could hold holes aka. UC entries.
4035 mtrr_gran_size=nn[KMG,X86,EARLY]
4036 Used for mtrr cleanup. It is granularity of mtrr block.
4037 Default is 1.
4038 Large value could prevent small alignment from
4039 using up MTRRs.
4041 mtrr_spare_reg_nr=n [X86,EARLY]
4042 Format: <integer>
4043 Range: 0,7 : spare reg number
4044 Default : 1
4045 Used for mtrr cleanup. It is spare mtrr entries number.
4046 Set to 2 or more if your graphical card needs more.
4048 multitce=off [PPC] This parameter disables the use of the pSeries
4049 firmware feature for updating multiple TCE entries
4050 at a time.
4052 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card
4054 netdev= [NET] Network devices parameters
4055 Format: <irq>,<io>,<mem_start>,<mem_end>,<name>
4056 Note that mem_start is often overloaded to mean
4057 something different and driver-specific.
4058 This usage is only documented in each driver source
4059 file if at all.
4061 netpoll.carrier_timeout=
4062 [NET] Specifies amount of time (in seconds) that
4063 netpoll should wait for a carrier. By default netpoll
4064 waits 4 seconds.
4066 nf_conntrack.acct=
4067 [NETFILTER] Enable connection tracking flow accounting
4068 0 to disable accounting
4069 1 to enable accounting
4070 Default value is 0.
4072 nfs.cache_getent=
4073 [NFS] sets the pathname to the program which is used
4074 to update the NFS client cache entries.
4076 nfs.cache_getent_timeout=
4077 [NFS] sets the timeout after which an attempt to
4078 update a cache entry is deemed to have failed.
4080 nfs.callback_nr_threads=
4081 [NFSv4] set the total number of threads that the
4082 NFS client will assign to service NFSv4 callback
4083 requests.
4085 nfs.callback_tcpport=
4086 [NFS] set the TCP port on which the NFSv4 callback
4087 channel should listen.
4089 nfs.delay_retrans=
4090 [NFS] specifies the number of times the NFSv4 client
4091 retries the request before returning an EAGAIN error,
4092 after a reply of NFS4ERR_DELAY from the server.
4093 Only applies if the softerr mount option is enabled,
4094 and the specified value is >= 0.
4096 nfs.enable_ino64=
4097 [NFS] enable 64-bit inode numbers.
4098 If zero, the NFS client will fake up a 32-bit inode
4099 number for the readdir() and stat() syscalls instead
4100 of returning the full 64-bit number.
4101 The default is to return 64-bit inode numbers.
4103 nfs.idmap_cache_timeout=
4104 [NFS] set the maximum lifetime for idmapper cache
4105 entries.
4107 nfs.max_session_cb_slots=
4108 [NFSv4.1] Sets the maximum number of session
4109 slots the client will assign to the callback
4110 channel. This determines the maximum number of
4111 callbacks the client will process in parallel for
4112 a particular server.
4114 nfs.max_session_slots=
4115 [NFSv4.1] Sets the maximum number of session slots
4116 the client will attempt to negotiate with the server.
4117 This limits the number of simultaneous RPC requests
4118 that the client can send to the NFSv4.1 server.
4119 Note that there is little point in setting this
4120 value higher than the max_tcp_slot_table_limit.
4122 nfs.nfs4_disable_idmapping=
4123 [NFSv4] When set to the default of '1', this option
4124 ensures that both the RPC level authentication
4125 scheme and the NFS level operations agree to use
4126 numeric uids/gids if the mount is using the
4127 'sec=sys' security flavour. In effect it is
4128 disabling idmapping, which can make migration from
4129 legacy NFSv2/v3 systems to NFSv4 easier.
4130 Servers that do not support this mode of operation
4131 will be autodetected by the client, and it will fall
4132 back to using the idmapper.
4133 To turn off this behaviour, set the value to '0'.
4135 nfs.nfs4_unique_id=
4136 [NFS4] Specify an additional fixed unique ident-
4137 ification string that NFSv4 clients can insert into
4138 their nfs_client_id4 string. This is typically a
4139 UUID that is generated at system install time.
4141 nfs.recover_lost_locks=
4142 [NFSv4] Attempt to recover locks that were lost due
4143 to a lease timeout on the server. Please note that
4144 doing this risks data corruption, since there are
4145 no guarantees that the file will remain unchanged
4146 after the locks are lost.
4147 If you want to enable the kernel legacy behaviour of
4148 attempting to recover these locks, then set this
4149 parameter to '1'.
4150 The default parameter value of '0' causes the kernel
4151 not to attempt recovery of lost locks.
4153 nfs.send_implementation_id=
4154 [NFSv4.1] Send client implementation identification
4155 information in exchange_id requests.
4156 If zero, no implementation identification information
4157 will be sent.
4158 The default is to send the implementation identification
4159 information.
4161 nfs4.layoutstats_timer=
4162 [NFSv4.2] Change the rate at which the kernel sends
4163 layoutstats to the pNFS metadata server.
4165 Setting this to value to 0 causes the kernel to use
4166 whatever value is the default set by the layout
4167 driver. A non-zero value sets the minimum interval
4168 in seconds between layoutstats transmissions.
4170 nfsd.inter_copy_offload_enable=
4171 [NFSv4.2] When set to 1, the server will support
4172 server-to-server copies for which this server is
4173 the destination of the copy.
4175 nfsd.nfs4_disable_idmapping=
4176 [NFSv4] When set to the default of '1', the NFSv4
4177 server will return only numeric uids and gids to
4178 clients using auth_sys, and will accept numeric uids
4179 and gids from such clients. This is intended to ease
4180 migration from NFSv2/v3.
4182 nfsd.nfsd4_ssc_umount_timeout=
4183 [NFSv4.2] When used as the destination of a
4184 server-to-server copy, knfsd temporarily mounts
4185 the source server. It caches the mount in case
4186 it will be needed again, and discards it if not
4187 used for the number of milliseconds specified by
4188 this parameter.
4190 nfsaddrs= [NFS] Deprecated. Use ip= instead.
4191 See Documentation/admin-guide/nfs/nfsroot.rst.
4193 nfsroot= [NFS] nfs root filesystem for disk-less boxes.
4194 See Documentation/admin-guide/nfs/nfsroot.rst.
4196 nfsrootdebug [NFS] enable nfsroot debugging messages.
4197 See Documentation/admin-guide/nfs/nfsroot.rst.
4199 nmi_backtrace.backtrace_idle [KNL]
4200 Dump stacks even of idle CPUs in response to an
4201 NMI stack-backtrace request.
4203 nmi_debug= [KNL,SH] Specify one or more actions to take
4204 when a NMI is triggered.
4205 Format: [state][,regs][,debounce][,die]
4207 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels
4208 Format: [panic,][nopanic,][rNNN,][num]
4209 Valid num: 0 or 1
4210 0 - turn hardlockup detector in nmi_watchdog off
4211 1 - turn hardlockup detector in nmi_watchdog on
4212 rNNN - configure the watchdog with raw perf event 0xNNN
4214 When panic is specified, panic when an NMI watchdog
4215 timeout occurs (or 'nopanic' to not panic on an NMI
4216 watchdog, if CONFIG_BOOTPARAM_HARDLOCKUP_PANIC is set)
4217 To disable both hard and soft lockup detectors,
4218 please see 'nowatchdog'.
4219 This is useful when you use a panic=... timeout and
4220 need the box quickly up again.
4222 These settings can be accessed at runtime via
4223 the nmi_watchdog and hardlockup_panic sysctls.
4225 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths
4226 emulation library even if a 387 maths coprocessor
4227 is present.
4229 no4lvl [RISCV,EARLY] Disable 4-level and 5-level paging modes.
4230 Forces kernel to use 3-level paging instead.
4232 no5lvl [X86-64,RISCV,EARLY] Disable 5-level paging mode. Forces
4233 kernel to use 4-level paging instead.
4235 noalign [KNL,ARM]
4237 noapic [SMP,APIC,EARLY] Tells the kernel to not make use of any
4238 IOAPICs that may be present in the system.
4240 noapictimer [APIC,X86] Don't set up the APIC timer
4242 noautogroup Disable scheduler automatic task group creation.
4244 nocache [ARM,EARLY]
4246 no_console_suspend
4247 [HW] Never suspend the console
4248 Disable suspending of consoles during suspend and
4249 hibernate operations. Once disabled, debugging
4250 messages can reach various consoles while the rest
4251 of the system is being put to sleep (ie, while
4252 debugging driver suspend/resume hooks). This may
4253 not work reliably with all consoles, but is known
4254 to work with serial and VGA consoles.
4255 To facilitate more flexible debugging, we also add
4256 console_suspend, a printk module parameter to control
4257 it. Users could use console_suspend (usually
4258 /sys/module/printk/parameters/console_suspend) to
4259 turn on/off it dynamically.
4261 no_debug_objects
4262 [KNL,EARLY] Disable object debugging
4264 nodsp [SH] Disable hardware DSP at boot time.
4266 noefi [EFI,EARLY] Disable EFI runtime services support.
4268 no_entry_flush [PPC,EARLY] Don't flush the L1-D cache when entering the kernel.
4270 noexec32 [X86-64]
4271 This affects only 32-bit executables.
4272 noexec32=on: enable non-executable mappings (default)
4273 read doesn't imply executable mappings
4274 noexec32=off: disable non-executable mappings
4275 read implies executable mappings
4277 no_file_caps Tells the kernel not to honor file capabilities. The
4278 only way then for a file to be executed with privilege
4279 is to be setuid root or executed by root.
4281 nofpu [MIPS,SH] Disable hardware FPU at boot time.
4283 nofsgsbase [X86] Disables FSGSBASE instructions.
4285 nofxsr [BUGS=X86-32] Disables x86 floating point extended
4286 register save and restore. The kernel will only save
4287 legacy floating-point registers on task switch.
4289 nogbpages [X86] Do not use GB pages for kernel direct mappings.
4291 no_hash_pointers
4292 [KNL,EARLY]
4293 Alias for "hash_pointers=never".
4295 nohibernate [HIBERNATION] Disable hibernation and resume.
4297 nohlt [ARM,ARM64,MICROBLAZE,MIPS,PPC,RISCV,SH] Forces the kernel to
4298 busy wait in do_idle() and not use the arch_cpu_idle()
4299 implementation; requires CONFIG_GENERIC_IDLE_POLL_SETUP
4300 to be effective. This is useful on platforms where the
4301 sleep(SH) or wfi(ARM,ARM64) instructions do not work
4302 correctly or when doing power measurements to evaluate
4303 the impact of the sleep instructions. This is also
4304 useful when using JTAG debugger.
4306 nohpet [X86] Don't use the HPET timer.
4308 nohugeiomap [KNL,X86,PPC,ARM64,EARLY] Disable kernel huge I/O mappings.
4310 nohugevmalloc [KNL,X86,PPC,ARM64,EARLY] Disable kernel huge vmalloc mappings.
4312 nohz= [KNL] Boottime enable/disable dynamic ticks
4313 Valid arguments: on, off
4314 Default: on
4316 nohz_full= [KNL,BOOT,SMP,ISOL]
4317 The argument is a cpu list, as described above.
4318 In kernels built with CONFIG_NO_HZ_FULL=y, set
4319 the specified list of CPUs whose tick will be stopped
4320 whenever possible. The boot CPU will be forced outside
4321 the range to maintain the timekeeping. Any CPUs
4322 in this list will have their RCU callbacks offloaded,
4323 just as if they had also been called out in the
4324 rcu_nocbs= boot parameter.
4326 Note that this argument takes precedence over
4327 the CONFIG_RCU_NOCB_CPU_DEFAULT_ALL option.
4329 noinitrd [RAM] Tells the kernel not to load any configured
4330 initial RAM disk.
4332 nointremap [X86-64,Intel-IOMMU,EARLY] Do not enable interrupt
4333 remapping.
4334 [Deprecated - use intremap=off]
4336 noinvpcid [X86,EARLY] Disable the INVPCID cpu feature.
4338 noiotrap [SH] Disables trapped I/O port accesses.
4340 noirqdebug [X86-32] Disables the code which attempts to detect and
4341 disable unhandled interrupt sources.
4343 noisapnp [ISAPNP] Disables ISA PnP code.
4345 nokaslr [KNL,EARLY]
4346 When CONFIG_RANDOMIZE_BASE is set, this disables
4347 kernel and module base offset ASLR (Address Space
4348 Layout Randomization).
4350 no-kvmapf [X86,KVM,EARLY] Disable paravirtualized asynchronous page
4351 fault handling.
4353 no-kvmclock [X86,KVM,EARLY] Disable paravirtualized KVM clock driver
4355 nolapic [X86-32,APIC,EARLY] Do not enable or use the local APIC.
4357 nolapic_timer [X86-32,APIC,EARLY] Do not use the local APIC timer.
4359 nomce [X86-32] Disable Machine Check Exception
4361 nomfgpt [X86-32] Disable Multi-Function General Purpose
4362 Timer usage (for AMD Geode machines).
4364 nomodeset Disable kernel modesetting. Most systems' firmware
4365 sets up a display mode and provides framebuffer memory
4366 for output. With nomodeset, DRM and fbdev drivers will
4367 not load if they could possibly displace the pre-
4368 initialized output. Only the system framebuffer will
4369 be available for use. The respective drivers will not
4370 perform display-mode changes or accelerated rendering.
4372 Useful as error fallback, or for testing and debugging.
4374 nomodule Disable module load
4376 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to
4377 shutdown the other cpus. Instead use the REBOOT_VECTOR
4378 irq.
4380 nopat [X86,EARLY] Disable PAT (page attribute table extension of
4381 pagetables) support.
4383 nopcid [X86-64,EARLY] Disable the PCID cpu feature.
4385 nopku [X86] Disable Memory Protection Keys CPU feature found
4386 in some Intel CPUs.
4388 nopti [X86-64,EARLY]
4389 Equivalent to pti=off
4391 nopv= [X86,XEN,KVM,HYPER_V,VMWARE,EARLY]
4392 Disables the PV optimizations forcing the guest to run
4393 as generic guest with no PV drivers. Currently support
4394 XEN HVM, KVM, HYPER_V and VMWARE guest.
4396 nopvspin [X86,XEN,KVM,EARLY]
4397 Disables the qspinlock slow path using PV optimizations
4398 which allow the hypervisor to 'idle' the guest on lock
4399 contention.
4401 norandmaps Don't use address space randomization. Equivalent to
4402 echo 0 > /proc/sys/kernel/randomize_va_space
4404 noreplace-smp [X86-32,SMP] Don't replace SMP instructions
4405 with UP alternatives
4407 noresume [SWSUSP] Disables resume and restores original swap
4408 space.
4410 no-scroll [VGA] Disables scrollback.
4411 This is required for the Braillex ib80-piezo Braille
4412 reader made by F.H. Papenmeier (Germany).
4414 nosgx [X86-64,SGX,EARLY] Disables Intel SGX kernel support.
4416 nosmap [PPC,EARLY]
4417 Disable SMAP (Supervisor Mode Access Prevention)
4418 even if it is supported by processor.
4420 nosmep [PPC64s,EARLY]
4421 Disable SMEP (Supervisor Mode Execution Prevention)
4422 even if it is supported by processor.
4424 nosmp [SMP,EARLY] Tells an SMP kernel to act as a UP kernel,
4425 and disable the IO APIC. legacy for "maxcpus=0".
4427 nosmt [KNL,MIPS,PPC,EARLY] Disable symmetric multithreading (SMT).
4428 Equivalent to smt=1.
4430 [KNL,X86,PPC,S390] Disable symmetric multithreading (SMT).
4431 nosmt=force: Force disable SMT, cannot be undone
4432 via the sysfs control file.
4434 nosoftlockup [KNL] Disable the soft-lockup detector.
4436 nospec_store_bypass_disable
4437 [HW,EARLY] Disable all mitigations for the Speculative
4438 Store Bypass vulnerability
4440 nospectre_bhb [ARM64,EARLY] Disable all mitigations for Spectre-BHB (branch
4441 history injection) vulnerability. System may allow data leaks
4442 with this option.
4444 nospectre_v1 [X86,PPC,EARLY] Disable mitigations for Spectre Variant 1
4445 (bounds check bypass). With this option data leaks are
4446 possible in the system.
4448 nospectre_v2 [X86,PPC_E500,ARM64,EARLY] Disable all mitigations
4449 for the Spectre variant 2 (indirect branch
4450 prediction) vulnerability. System may allow data
4451 leaks with this option.
4453 no-steal-acc [X86,PV_OPS,ARM64,PPC/PSERIES,RISCV,LOONGARCH,EARLY]
4454 Disable paravirtualized steal time accounting. steal time
4455 is computed, but won't influence scheduler behaviour
4457 nosync [HW,M68K] Disables sync negotiation for all devices.
4459 no_timer_check [X86,APIC] Disables the code which tests for broken
4460 timer IRQ sources, i.e., the IO-APIC timer. This can
4461 work around problems with incorrect timer
4462 initialization on some boards.
4464 no_uaccess_flush
4465 [PPC,EARLY] Don't flush the L1-D cache after accessing user data.
4467 novmcoredd [KNL,KDUMP]
4468 Disable device dump. Device dump allows drivers to
4469 append dump data to vmcore so you can collect driver
4470 specified debug info. Drivers can append the data
4471 without any limit and this data is stored in memory,
4472 so this may cause significant memory stress. Disabling
4473 device dump can help save memory but the driver debug
4474 data will be no longer available. This parameter
4475 is only available when CONFIG_PROC_VMCORE_DEVICE_DUMP
4476 is set.
4478 no-vmw-sched-clock
4479 [X86,PV_OPS,EARLY] Disable paravirtualized VMware
4480 scheduler clock and use the default one.
4482 nowatchdog [KNL] Disable both lockup detectors, i.e.
4483 soft-lockup and NMI watchdog (hard-lockup).
4485 nowb [ARM,EARLY]
4487 nox2apic [X86-64,APIC,EARLY] Do not enable x2APIC mode.
4489 NOTE: this parameter will be ignored on systems with the
4490 LEGACY_XAPIC_DISABLED bit set in the
4491 IA32_XAPIC_DISABLE_STATUS MSR.
4493 noxsave [BUGS=X86] Disables x86 extended register state save
4494 and restore using xsave. The kernel will fallback to
4495 enabling legacy floating-point and sse state.
4497 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended
4498 register states. The kernel will fall back to use
4499 xsave to save the states. By using this parameter,
4500 performance of saving the states is degraded because
4501 xsave doesn't support modified optimization while
4502 xsaveopt supports it on xsaveopt enabled systems.
4504 noxsaves [X86] Disables xsaves and xrstors used in saving and
4505 restoring x86 extended register state in compacted
4506 form of xsave area. The kernel will fall back to use
4507 xsaveopt and xrstor to save and restore the states
4508 in standard form of xsave area. By using this
4509 parameter, xsave area per process might occupy more
4510 memory on xsaves enabled systems.
4512 nr_cpus= [SMP,EARLY] Maximum number of processors that an SMP kernel
4513 could support. nr_cpus=n : n >= 1 limits the kernel to
4514 support 'n' processors. It could be larger than the
4515 number of already plugged CPU during bootup, later in
4516 runtime you can physically add extra cpu until it reaches
4517 n. So during boot up some boot time memory for per-cpu
4518 variables need be pre-allocated for later physical cpu
4519 hot plugging.
4521 nr_uarts= [SERIAL] maximum number of UARTs to be registered.
4523 numa=off [KNL, ARM64, PPC, RISCV, SPARC, X86, EARLY]
4524 Disable NUMA, Only set up a single NUMA node
4525 spanning all memory.
4527 numa=fake=<size>[MG]
4528 [KNL, ARM64, RISCV, X86, EARLY]
4529 If given as a memory unit, fills all system RAM with
4530 nodes of size interleaved over physical nodes.
4532 numa=fake=<N>
4533 [KNL, ARM64, RISCV, X86, EARLY]
4534 If given as an integer, fills all system RAM with N
4535 fake nodes interleaved over physical nodes.
4537 numa=fake=<N>U
4538 [KNL, ARM64, RISCV, X86, EARLY]
4539 If given as an integer followed by 'U', it will
4540 divide each physical node into N emulated nodes.
4542 numa=noacpi [X86] Don't parse the SRAT table for NUMA setup
4544 numa=nohmat [X86] Don't parse the HMAT table for NUMA setup, or
4545 soft-reserved memory partitioning.
4547 numa_balancing= [KNL,ARM64,PPC,RISCV,S390,X86] Enable or disable automatic
4548 NUMA balancing.
4549 Allowed values are enable and disable
4551 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA.
4552 'node', 'default' can be specified
4553 This can be set from sysctl after boot.
4554 See Documentation/admin-guide/sysctl/vm.rst for details.
4556 ohci1394_dma=early [HW,EARLY] enable debugging via the ohci1394 driver.
4557 See Documentation/core-api/debugging-via-ohci1394.rst for more
4558 info.
4560 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands
4561 Rather than timing out after 20 ms if an EC
4562 command is not properly ACKed, override the length
4563 of the timeout. We have interrupts disabled while
4564 waiting for the ACK, so if this is set too high
4565 interrupts *may* be lost!
4567 omap_mux= [OMAP] Override bootloader pin multiplexing.
4568 Format: <mux_mode0.mode_name=value>...
4569 For example, to override I2C bus2:
4570 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100
4572 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration
4574 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock]
4576 boundary - index of last SLC block on Flex-OneNAND.
4577 The remaining blocks are configured as MLC blocks.
4578 lock - Configure if Flex-OneNAND boundary should be locked.
4579 Once locked, the boundary cannot be changed.
4580 1 indicates lock status, 0 indicates unlock status.
4582 oops=panic [KNL,EARLY]
4583 Always panic on oopses. Default is to just kill the
4584 process, but there is a small probability of
4585 deadlocking the machine.
4586 This will also cause panics on machine check exceptions.
4587 Useful together with panic=30 to trigger a reboot.
4589 page_alloc.shuffle=
4590 [KNL] Boolean flag to control whether the page allocator
4591 should randomize its free lists. This parameter can be
4592 used to enable/disable page randomization. The state of
4593 the flag can be read from sysfs at:
4594 /sys/module/page_alloc/parameters/shuffle.
4595 This parameter is only available if CONFIG_SHUFFLE_PAGE_ALLOCATOR=y.
4597 page_owner= [KNL,EARLY] Boot-time page_owner enabling option.
4598 Storage of the information about who allocated
4599 each page is disabled in default. With this switch,
4600 we can turn it on.
4601 on: enable the feature
4603 page_poison= [KNL,EARLY] Boot-time parameter changing the state of
4604 poisoning on the buddy allocator, available with
4605 CONFIG_PAGE_POISONING=y.
4606 off: turn off poisoning (default)
4607 on: turn on poisoning
4609 page_reporting.page_reporting_order=
4610 [KNL] Minimal page reporting order
4611 Format: <integer>
4612 Adjust the minimal page reporting order. The page
4613 reporting is disabled when it exceeds MAX_PAGE_ORDER.
4615 panic= [KNL] Kernel behaviour on panic: delay <timeout>
4616 timeout > 0: seconds before rebooting
4617 timeout = 0: wait forever
4618 timeout < 0: reboot immediately
4619 Format: <timeout>
4621 panic_on_taint= [KNL,EARLY]
4622 Bitmask for conditionally calling panic() in add_taint()
4623 Format: <hex>[,nousertaint]
4624 Hexadecimal bitmask representing the set of TAINT flags
4625 that will cause the kernel to panic when add_taint() is
4626 called with any of the flags in this set.
4627 The optional switch "nousertaint" can be utilized to
4628 prevent userspace forced crashes by writing to sysctl
4629 /proc/sys/kernel/tainted any flagset matching with the
4630 bitmask set on panic_on_taint.
4631 See Documentation/admin-guide/tainted-kernels.rst for
4632 extra details on the taint flags that users can pick
4633 to compose the bitmask to assign to panic_on_taint.
4635 panic_on_warn=1 panic() instead of WARN(). Useful to cause kdump
4636 on a WARN().
4638 panic_print= Bitmask for printing system info when panic happens.
4639 User can chose combination of the following bits:
4640 bit 0: print all tasks info
4641 bit 1: print system memory info
4642 bit 2: print timer info
4643 bit 3: print locks info if CONFIG_LOCKDEP is on
4644 bit 4: print ftrace buffer
4645 bit 5: replay all kernel messages on consoles at the end of panic
4646 bit 6: print all CPUs backtrace (if available in the arch)
4647 bit 7: print only tasks in uninterruptible (blocked) state
4648 *Be aware* that this option may print a _lot_ of lines,
4649 so there are risks of losing older messages in the log.
4650 Use this option carefully, maybe worth to setup a
4651 bigger log buffer with "log_buf_len" along with this.
4653 panic_sys_info= A comma separated list of extra information to be dumped
4654 on panic.
4655 Format: val[,val...]
4656 Where @val can be any of the following:
4658 tasks: print all tasks info
4659 mem: print system memory info
4660 timers: print timers info
4661 locks: print locks info if CONFIG_LOCKDEP is on
4662 ftrace: print ftrace buffer
4663 all_bt: print all CPUs backtrace (if available in the arch)
4664 blocked_tasks: print only tasks in uninterruptible (blocked) state
4666 This is a human readable alternative to the 'panic_print' option.
4668 panic_console_replay
4669 When panic happens, replay all kernel messages on
4670 consoles at the end of panic.
4672 parkbd.port= [HW] Parallel port number the keyboard adapter is
4673 connected to, default is 0.
4674 Format: <parport#>
4675 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation,
4676 0 for XT, 1 for AT (default is AT).
4677 Format: <mode>
4679 parport= [HW,PPT] Specify parallel ports. 0 disables.
4680 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] }
4681 Use 'auto' to force the driver to use any
4682 IRQ/DMA settings detected (the default is to
4683 ignore detected IRQ/DMA settings because of
4684 possible conflicts). You can specify the base
4685 address, IRQ, and DMA settings; IRQ and DMA
4686 should be numbers, or 'auto' (for using detected
4687 settings on that particular port), or 'nofifo'
4688 (to avoid using a FIFO even if it is detected).
4689 Parallel ports are assigned in the order they
4690 are specified on the command line, starting
4691 with parport0.
4693 parport_init_mode= [HW,PPT]
4694 Configure VIA parallel port to operate in
4695 a specific mode. This is necessary on Pegasos
4696 computer where firmware has no options for setting
4697 up parallel port mode and sets it to spp.
4698 Currently this function knows 686a and 8231 chips.
4699 Format: [spp|ps2|epp|ecp|ecpepp]
4701 pata_legacy.all= [HW,LIBATA]
4702 Format: <int>
4703 Set to non-zero to probe primary and secondary ISA
4704 port ranges on PCI systems where no PCI PATA device
4705 has been found at either range. Disabled by default.
4707 pata_legacy.autospeed= [HW,LIBATA]
4708 Format: <int>
4709 Set to non-zero if a chip is present that snoops speed
4710 changes. Disabled by default.
4712 pata_legacy.ht6560a= [HW,LIBATA]
4713 Format: <int>
4714 Set to 1, 2, or 3 for HT 6560A on the primary channel,
4715 the secondary channel, or both channels respectively.
4716 Disabled by default.
4718 pata_legacy.ht6560b= [HW,LIBATA]
4719 Format: <int>
4720 Set to 1, 2, or 3 for HT 6560B on the primary channel,
4721 the secondary channel, or both channels respectively.
4722 Disabled by default.
4724 pata_legacy.iordy_mask= [HW,LIBATA]
4725 Format: <int>
4726 IORDY enable mask. Set individual bits to allow IORDY
4727 for the respective channel. Bit 0 is for the first
4728 legacy channel handled by this driver, bit 1 is for
4729 the second channel, and so on. The sequence will often
4730 correspond to the primary legacy channel, the secondary
4731 legacy channel, and so on, but the handling of a PCI
4732 bus and the use of other driver options may interfere
4733 with the sequence. By default IORDY is allowed across
4734 all channels.
4736 pata_legacy.opti82c46x= [HW,LIBATA]
4737 Format: <int>
4738 Set to 1, 2, or 3 for Opti 82c611A on the primary
4739 channel, the secondary channel, or both channels
4740 respectively. Disabled by default.
4742 pata_legacy.opti82c611a= [HW,LIBATA]
4743 Format: <int>
4744 Set to 1, 2, or 3 for Opti 82c465MV on the primary
4745 channel, the secondary channel, or both channels
4746 respectively. Disabled by default.
4748 pata_legacy.pio_mask= [HW,LIBATA]
4749 Format: <int>
4750 PIO mode mask for autospeed devices. Set individual
4751 bits to allow the use of the respective PIO modes.
4752 Bit 0 is for mode 0, bit 1 is for mode 1, and so on.
4753 All modes allowed by default.
4755 pata_legacy.probe_all= [HW,LIBATA]
4756 Format: <int>
4757 Set to non-zero to probe tertiary and further ISA
4758 port ranges on PCI systems. Disabled by default.
4760 pata_legacy.probe_mask= [HW,LIBATA]
4761 Format: <int>
4762 Probe mask for legacy ISA PATA ports. Depending on
4763 platform configuration and the use of other driver
4764 options up to 6 legacy ports are supported: 0x1f0,
4765 0x170, 0x1e8, 0x168, 0x1e0, 0x160, however probing
4766 of individual ports can be disabled by setting the
4767 corresponding bits in the mask to 1. Bit 0 is for
4768 the first port in the list above (0x1f0), and so on.
4769 By default all supported ports are probed.
4771 pata_legacy.qdi= [HW,LIBATA]
4772 Format: <int>
4773 Set to non-zero to probe QDI controllers. By default
4774 set to 1 if CONFIG_PATA_QDI_MODULE, 0 otherwise.
4776 pata_legacy.winbond= [HW,LIBATA]
4777 Format: <int>
4778 Set to non-zero to probe Winbond controllers. Use
4779 the standard I/O port (0x130) if 1, otherwise the
4780 value given is the I/O port to use (typically 0x1b0).
4781 By default set to 1 if CONFIG_PATA_WINBOND_VLB_MODULE,
4782 0 otherwise.
4784 pata_platform.pio_mask= [HW,LIBATA]
4785 Format: <int>
4786 Supported PIO mode mask. Set individual bits to allow
4787 the use of the respective PIO modes. Bit 0 is for
4788 mode 0, bit 1 is for mode 1, and so on. Mode 0 only
4789 allowed by default.
4791 pause_on_oops=<int>
4792 Halt all CPUs after the first oops has been printed for
4793 the specified number of seconds. This is to be used if
4794 your oopses keep scrolling off the screen.
4796 pcbit= [HW,ISDN]
4798 pci=option[,option...] [PCI,EARLY] various PCI subsystem options.
4800 Some options herein operate on a specific device
4801 or a set of devices (<pci_dev>). These are
4802 specified in one of the following formats:
4804 [<domain>:]<bus>:<dev>.<func>[/<dev>.<func>]*
4805 pci:<vendor>:<device>[:<subvendor>:<subdevice>]
4807 Note: the first format specifies a PCI
4808 bus/device/function address which may change
4809 if new hardware is inserted, if motherboard
4810 firmware changes, or due to changes caused
4811 by other kernel parameters. If the
4812 domain is left unspecified, it is
4813 taken to be zero. Optionally, a path
4814 to a device through multiple device/function
4815 addresses can be specified after the base
4816 address (this is more robust against
4817 renumbering issues). The second format
4818 selects devices using IDs from the
4819 configuration space which may match multiple
4820 devices in the system.
4822 earlydump dump PCI config space before the kernel
4823 changes anything
4824 off [X86] don't probe for the PCI bus
4825 bios [X86-32] force use of PCI BIOS, don't access
4826 the hardware directly. Use this if your machine
4827 has a non-standard PCI host bridge.
4828 nobios [X86-32] disallow use of PCI BIOS, only direct
4829 hardware access methods are allowed. Use this
4830 if you experience crashes upon bootup and you
4831 suspect they are caused by the BIOS.
4832 conf1 [X86] Force use of PCI Configuration Access
4833 Mechanism 1 (config address in IO port 0xCF8,
4834 data in IO port 0xCFC, both 32-bit).
4835 conf2 [X86] Force use of PCI Configuration Access
4836 Mechanism 2 (IO port 0xCF8 is an 8-bit port for
4837 the function, IO port 0xCFA, also 8-bit, sets
4838 bus number. The config space is then accessed
4839 through ports 0xC000-0xCFFF).
4840 See http://wiki.osdev.org/PCI for more info
4841 on the configuration access mechanisms.
4842 noaer [PCIE] If the PCIEAER kernel config parameter is
4843 enabled, this kernel boot option can be used to
4844 disable the use of PCIE advanced error reporting.
4845 nodomains [PCI] Disable support for multiple PCI
4846 root domains (aka PCI segments, in ACPI-speak).
4847 nommconf [X86] Disable use of MMCONFIG for PCI
4848 Configuration
4849 check_enable_amd_mmconf [X86] check for and enable
4850 properly configured MMIO access to PCI
4851 config space on AMD family 10h CPU
4852 nomsi [MSI] If the PCI_MSI kernel config parameter is
4853 enabled, this kernel boot option can be used to
4854 disable the use of MSI interrupts system-wide.
4855 noioapicquirk [APIC] Disable all boot interrupt quirks.
4856 Safety option to keep boot IRQs enabled. This
4857 should never be necessary.
4858 ioapicreroute [APIC] Enable rerouting of boot IRQs to the
4859 primary IO-APIC for bridges that cannot disable
4860 boot IRQs. This fixes a source of spurious IRQs
4861 when the system masks IRQs.
4862 noioapicreroute [APIC] Disable workaround that uses the
4863 boot IRQ equivalent of an IRQ that connects to
4864 a chipset where boot IRQs cannot be disabled.
4865 The opposite of ioapicreroute.
4866 biosirq [X86-32] Use PCI BIOS calls to get the interrupt
4867 routing table. These calls are known to be buggy
4868 on several machines and they hang the machine
4869 when used, but on other computers it's the only
4870 way to get the interrupt routing table. Try
4871 this option if the kernel is unable to allocate
4872 IRQs or discover secondary PCI buses on your
4873 motherboard.
4874 rom [X86] Assign address space to expansion ROMs.
4875 Use with caution as certain devices share
4876 address decoders between ROMs and other
4877 resources.
4878 norom [X86] Do not assign address space to
4879 expansion ROMs that do not already have
4880 BIOS assigned address ranges.
4881 nobar [X86] Do not assign address space to the
4882 BARs that weren't assigned by the BIOS.
4883 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be
4884 assigned automatically to PCI devices. You can
4885 make the kernel exclude IRQs of your ISA cards
4886 this way.
4887 pirqaddr=0xAAAAA [X86] Specify the physical address
4888 of the PIRQ table (normally generated
4889 by the BIOS) if it is outside the
4890 F0000h-100000h range.
4891 lastbus=N [X86] Scan all buses thru bus #N. Can be
4892 useful if the kernel is unable to find your
4893 secondary buses and you want to tell it
4894 explicitly which ones they are.
4895 assign-busses [X86] Always assign all PCI bus
4896 numbers ourselves, overriding
4897 whatever the firmware may have done.
4898 usepirqmask [X86] Honor the possible IRQ mask stored
4899 in the BIOS $PIR table. This is needed on
4900 some systems with broken BIOSes, notably
4901 some HP Pavilion N5400 and Omnibook XE3
4902 notebooks. This will have no effect if ACPI
4903 IRQ routing is enabled.
4904 noacpi [X86] Do not use ACPI for IRQ routing
4905 or for PCI scanning.
4906 use_crs [X86] Use PCI host bridge window information
4907 from ACPI. On BIOSes from 2008 or later, this
4908 is enabled by default. If you need to use this,
4909 please report a bug.
4910 nocrs [X86] Ignore PCI host bridge windows from ACPI.
4911 If you need to use this, please report a bug.
4912 use_e820 [X86] Use E820 reservations to exclude parts of
4913 PCI host bridge windows. This is a workaround
4914 for BIOS defects in host bridge _CRS methods.
4915 If you need to use this, please report a bug to
4916 <linux-pci@vger.kernel.org>.
4917 no_e820 [X86] Ignore E820 reservations for PCI host
4918 bridge windows. This is the default on modern
4919 hardware. If you need to use this, please report
4920 a bug to <linux-pci@vger.kernel.org>.
4921 routeirq Do IRQ routing for all PCI devices.
4922 This is normally done in pci_enable_device(),
4923 so this option is a temporary workaround
4924 for broken drivers that don't call it.
4925 skip_isa_align [X86] do not align io start addr, so can
4926 handle more pci cards
4927 noearly [X86] Don't do any early type 1 scanning.
4928 This might help on some broken boards which
4929 machine check when some devices' config space
4930 is read. But various workarounds are disabled
4931 and some IOMMU drivers will not work.
4932 bfsort Sort PCI devices into breadth-first order.
4933 This sorting is done to get a device
4934 order compatible with older (<= 2.4) kernels.
4935 nobfsort Don't sort PCI devices into breadth-first order.
4936 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size)
4937 tuning and use the BIOS-configured MPS defaults.
4938 pcie_bus_safe Set every device's MPS to the largest value
4939 supported by all devices below the root complex.
4940 pcie_bus_perf Set device MPS to the largest allowable MPS
4941 based on its parent bus. Also set MRRS (Max
4942 Read Request Size) to the largest supported
4943 value (no larger than the MPS that the device
4944 or bus can support) for best performance.
4945 pcie_bus_peer2peer Set every device's MPS to 128B, which
4946 every device is guaranteed to support. This
4947 configuration allows peer-to-peer DMA between
4948 any pair of devices, possibly at the cost of
4949 reduced performance. This also guarantees
4950 that hot-added devices will work.
4951 cbiosize=nn[KMG] The fixed amount of bus space which is
4952 reserved for the CardBus bridge's IO window.
4953 The default value is 256 bytes.
4954 cbmemsize=nn[KMG] The fixed amount of bus space which is
4955 reserved for the CardBus bridge's memory
4956 window. The default value is 64 megabytes.
4957 resource_alignment=
4958 Format:
4959 [<order of align>@]<pci_dev>[; ...]
4960 Specifies alignment and device to reassign
4961 aligned memory resources. How to
4962 specify the device is described above.
4963 If <order of align> is not specified,
4964 PAGE_SIZE is used as alignment.
4965 A PCI-PCI bridge can be specified if resource
4966 windows need to be expanded.
4967 To specify the alignment for several
4968 instances of a device, the PCI vendor,
4969 device, subvendor, and subdevice may be
4970 specified, e.g., 12@pci:8086:9c22:103c:198f
4971 for 4096-byte alignment.
4972 ecrc= Enable/disable PCIe ECRC (transaction layer
4973 end-to-end CRC checking). Only effective if
4974 OS has native AER control (either granted by
4975 ACPI _OSC or forced via "pcie_ports=native")
4976 bios: Use BIOS/firmware settings. This is the
4977 the default.
4978 off: Turn ECRC off
4979 on: Turn ECRC on.
4980 hpiosize=nn[KMG] The fixed amount of bus space which is
4981 reserved for hotplug bridge's IO window.
4982 Default size is 256 bytes.
4983 hpmmiosize=nn[KMG] The fixed amount of bus space which is
4984 reserved for hotplug bridge's MMIO window.
4985 Default size is 2 megabytes.
4986 hpmmioprefsize=nn[KMG] The fixed amount of bus space which is
4987 reserved for hotplug bridge's MMIO_PREF window.
4988 Default size is 2 megabytes.
4989 hpmemsize=nn[KMG] The fixed amount of bus space which is
4990 reserved for hotplug bridge's MMIO and
4991 MMIO_PREF window.
4992 Default size is 2 megabytes.
4993 hpbussize=nn The minimum amount of additional bus numbers
4994 reserved for buses below a hotplug bridge.
4995 Default is 1.
4996 realloc= Enable/disable reallocating PCI bridge resources
4997 if allocations done by BIOS are too small to
4998 accommodate resources required by all child
4999 devices.
5000 off: Turn realloc off
5001 on: Turn realloc on
5002 realloc same as realloc=on
5003 noari do not use PCIe ARI.
5004 noats [PCIE, Intel-IOMMU, AMD-IOMMU]
5005 do not use PCIe ATS (and IOMMU device IOTLB).
5006 pcie_scan_all Scan all possible PCIe devices. Otherwise we
5007 only look for one device below a PCIe downstream
5008 port.
5009 big_root_window Try to add a big 64bit memory window to the PCIe
5010 root complex on AMD CPUs. Some GFX hardware
5011 can resize a BAR to allow access to all VRAM.
5012 Adding the window is slightly risky (it may
5013 conflict with unreported devices), so this
5014 taints the kernel.
5015 disable_acs_redir=<pci_dev>[; ...]
5016 Specify one or more PCI devices (in the format
5017 specified above) separated by semicolons.
5018 Each device specified will have the PCI ACS
5019 redirect capabilities forced off which will
5020 allow P2P traffic between devices through
5021 bridges without forcing it upstream. Note:
5022 this removes isolation between devices and
5023 may put more devices in an IOMMU group.
5024 config_acs=
5025 Format:
5026 <ACS flags>@<pci_dev>[; ...]
5027 Specify one or more PCI devices (in the format
5028 specified above) optionally prepended with flags
5029 and separated by semicolons. The respective
5030 capabilities will be enabled, disabled or
5031 unchanged based on what is specified in
5032 flags.
5034 ACS Flags is defined as follows:
5035 bit-0 : ACS Source Validation
5036 bit-1 : ACS Translation Blocking
5037 bit-2 : ACS P2P Request Redirect
5038 bit-3 : ACS P2P Completion Redirect
5039 bit-4 : ACS Upstream Forwarding
5040 bit-5 : ACS P2P Egress Control
5041 bit-6 : ACS Direct Translated P2P
5042 Each bit can be marked as:
5043 '0' – force disabled
5044 '1' – force enabled
5045 'x' – unchanged
5046 For example,
5047 pci=config_acs=10x@pci:0:0
5048 would configure all devices that support
5049 ACS to enable P2P Request Redirect, disable
5050 Translation Blocking, and leave Source
5051 Validation unchanged from whatever power-up
5052 or firmware set it to.
5054 Note: this may remove isolation between devices
5055 and may put more devices in an IOMMU group.
5056 force_floating [S390] Force usage of floating interrupts.
5057 nomio [S390] Do not use MIO instructions.
5058 norid [S390] ignore the RID field and force use of
5059 one PCI domain per PCI function
5060 notph [PCIE] If the PCIE_TPH kernel config parameter
5061 is enabled, this kernel boot option can be used
5062 to disable PCIe TLP Processing Hints support
5063 system-wide.
5065 pcie_aspm= [PCIE] Forcibly enable or ignore PCIe Active State Power
5066 Management.
5067 off Don't touch ASPM configuration at all. Leave any
5068 configuration done by firmware unchanged.
5069 force Enable ASPM even on devices that claim not to support it.
5070 WARNING: Forcing ASPM on may cause system lockups.
5072 pcie_ports= [PCIE] PCIe port services handling:
5073 native Use native PCIe services (PME, AER, DPC, PCIe hotplug)
5074 even if the platform doesn't give the OS permission to
5075 use them. This may cause conflicts if the platform
5076 also tries to use these services.
5077 dpc-native Use native PCIe service for DPC only. May
5078 cause conflicts if firmware uses AER or DPC.
5079 compat Disable native PCIe services (PME, AER, DPC, PCIe
5080 hotplug).
5082 pcie_port_pm= [PCIE] PCIe port power management handling:
5083 off Disable power management of all PCIe ports
5084 force Forcibly enable power management of all PCIe ports
5086 pcie_pme= [PCIE,PM] Native PCIe PME signaling options:
5087 nomsi Do not use MSI for native PCIe PME signaling (this makes
5088 all PCIe root ports use INTx for all services).
5090 pcmv= [HW,PCMCIA] BadgePAD 4
5092 pd_ignore_unused
5093 [PM]
5094 Keep all power-domains already enabled by bootloader on,
5095 even if no driver has claimed them. This is useful
5096 for debug and development, but should not be
5097 needed on a platform with proper driver support.
5099 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at
5100 boot time.
5101 Format: { 0 | 1 }
5102 See arch/parisc/kernel/pdc_chassis.c
5104 percpu_alloc= [MM,EARLY]
5105 Select which percpu first chunk allocator to use.
5106 Currently supported values are "embed" and "page".
5107 Archs may support subset or none of the selections.
5108 See comments in mm/percpu.c for details on each
5109 allocator. This parameter is primarily for debugging
5110 and performance comparison.
5112 pirq= [SMP,APIC] Manual mp-table setup
5113 See Documentation/arch/x86/i386/IO-APIC.rst.
5115 plip= [PPT,NET] Parallel port network link
5116 Format: { parport<nr> | timid | 0 }
5117 See also Documentation/admin-guide/parport.rst.
5119 pmtmr= [X86] Manual setup of pmtmr I/O Port.
5120 Override pmtimer IOPort with a hex value.
5121 e.g. pmtmr=0x508
5123 pmu_override= [PPC] Override the PMU.
5124 This option takes over the PMU facility, so it is no
5125 longer usable by perf. Setting this option starts the
5126 PMU counters by setting MMCR0 to 0 (the FC bit is
5127 cleared). If a number is given, then MMCR1 is set to
5128 that number, otherwise (e.g., 'pmu_override=on'), MMCR1
5129 remains 0.
5131 pm_async= [PM]
5132 Format: off
5133 This parameter sets the initial value of the
5134 /sys/power/pm_async sysfs knob at boot time.
5135 If set to "off", disables asynchronous suspend and
5136 resume of devices during system-wide power transitions.
5137 This can be useful on platforms where device
5138 dependencies are not well-defined, or for debugging
5139 power management issues. Asynchronous operations are
5140 enabled by default.
5143 pm_debug_messages [SUSPEND,KNL]
5144 Enable suspend/resume debug messages during boot up.
5146 pnp.debug=1 [PNP]
5147 Enable PNP debug messages (depends on the
5148 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time
5149 via /sys/module/pnp/parameters/debug. We always show
5150 current resource usage; turning this on also shows
5151 possible settings and some assignment information.
5153 pnpacpi= [ACPI]
5154 { off }
5156 pnpbios= [ISAPNP]
5157 { on | off | curr | res | no-curr | no-res }
5159 pnp_reserve_irq=
5160 [ISAPNP] Exclude IRQs for the autoconfiguration
5162 pnp_reserve_dma=
5163 [ISAPNP] Exclude DMAs for the autoconfiguration
5165 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration
5166 Ranges are in pairs (I/O port base and size).
5168 pnp_reserve_mem=
5169 [ISAPNP] Exclude memory regions for the
5170 autoconfiguration.
5171 Ranges are in pairs (memory base and size).
5173 ports= [IP_VS_FTP] IPVS ftp helper module
5174 Default is 21.
5175 Up to 8 (IP_VS_APP_MAX_PORTS) ports
5176 may be specified.
5177 Format: <port>,<port>....
5179 possible_cpus= [SMP,S390,X86]
5180 Format: <unsigned int>
5181 Set the number of possible CPUs, overriding the
5182 regular discovery mechanisms (such as ACPI/FW, etc).
5184 powersave=off [PPC] This option disables power saving features.
5185 It specifically disables cpuidle and sets the
5186 platform machine description specific power_save
5187 function to NULL. On Idle the CPU just reduces
5188 execution priority.
5190 ppc_strict_facility_enable
5191 [PPC,ENABLE] This option catches any kernel floating point,
5192 Altivec, VSX and SPE outside of regions specifically
5193 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()).
5194 There is some performance impact when enabling this.
5196 ppc_tm= [PPC,EARLY]
5197 Format: {"off"}
5198 Disable Hardware Transactional Memory
5200 preempt= [KNL]
5201 Select preemption mode if you have CONFIG_PREEMPT_DYNAMIC
5202 none - Limited to cond_resched() calls
5203 voluntary - Limited to cond_resched() and might_sleep() calls
5204 full - Any section that isn't explicitly preempt disabled
5205 can be preempted anytime. Tasks will also yield
5206 contended spinlocks (if the critical section isn't
5207 explicitly preempt disabled beyond the lock itself).
5208 lazy - Scheduler controlled. Similar to full but instead
5209 of preempting the task immediately, the task gets
5210 one HZ tick time to yield itself before the
5211 preemption will be forced. One preemption is when the
5212 task returns to user space.
5214 print-fatal-signals=
5215 [KNL] debug: print fatal signals
5217 If enabled, warn about various signal handling
5218 related application anomalies: too many signals,
5219 too many POSIX.1 timers, fatal signals causing a
5220 coredump - etc.
5222 If you hit the warning due to signal overflow,
5223 you might want to try "ulimit -i unlimited".
5225 default: off.
5227 printk.always_kmsg_dump=
5228 Trigger kmsg_dump for cases other than kernel oops or
5229 panics
5230 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
5231 default: disabled
5233 printk.console_no_auto_verbose=
5234 Disable console loglevel raise on oops, panic
5235 or lockdep-detected issues (only if lock debug is on).
5236 With an exception to setups with low baudrate on
5237 serial console, keeping this 0 is a good choice
5238 in order to provide more debug information.
5239 Format: <bool>
5240 default: 0 (auto_verbose is enabled)
5242 printk.debug_non_panic_cpus=
5243 Allows storing messages from non-panic CPUs into
5244 the printk log buffer during panic(). They are
5245 flushed to consoles by the panic-CPU on
5246 a best-effort basis.
5247 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
5248 Default: disabled
5250 printk.devkmsg={on,off,ratelimit}
5251 Control writing to /dev/kmsg.
5252 on - unlimited logging to /dev/kmsg from userspace
5253 off - logging to /dev/kmsg disabled
5254 ratelimit - ratelimit the logging
5255 Default: ratelimit
5257 printk.time= Show timing data prefixed to each printk message line
5258 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
5260 proc_mem.force_override= [KNL]
5261 Format: {always | ptrace | never}
5262 Traditionally /proc/pid/mem allows memory permissions to be
5263 overridden without restrictions. This option may be set to
5264 restrict that. Can be one of:
5265 - 'always': traditional behavior always allows mem overrides.
5266 - 'ptrace': only allow mem overrides for active ptracers.
5267 - 'never': never allow mem overrides.
5268 If not specified, default is the CONFIG_PROC_MEM_* choice.
5270 processor.max_cstate= [HW,ACPI]
5271 Limit processor to maximum C-state
5272 max_cstate=9 overrides any DMI blacklist limit.
5274 processor.nocst [HW,ACPI]
5275 Ignore the _CST method to determine C-states,
5276 instead using the legacy FADT method
5278 profile= [KNL] Enable kernel profiling via /proc/profile
5279 Format: [<profiletype>,]<number>
5280 Param: <profiletype>: "schedule" or "kvm"
5281 [defaults to kernel profiling]
5282 Param: "schedule" - profile schedule points.
5283 Param: "kvm" - profile VM exits.
5284 Param: <number> - step/bucket size as a power of 2 for
5285 statistical time based profiling.
5287 prompt_ramdisk= [RAM] [Deprecated]
5289 prot_virt= [S390] enable hosting protected virtual machines
5290 isolated from the hypervisor (if hardware supports
5291 that). If enabled, the default kernel base address
5292 might be overridden even when Kernel Address Space
5293 Layout Randomization is disabled.
5294 Format: <bool>
5296 psi= [KNL] Enable or disable pressure stall information
5297 tracking.
5298 Format: <bool>
5300 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to
5301 probe for; one of (bare|imps|exps|lifebook|any).
5302 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports
5303 per second.
5304 psmouse.resetafter= [HW,MOUSE]
5305 Try to reset the device after so many bad packets
5306 (0 = never).
5307 psmouse.resolution=
5308 [HW,MOUSE] Set desired mouse resolution, in dpi.
5309 psmouse.smartscroll=
5310 [HW,MOUSE] Controls Logitech smartscroll autorepeat.
5311 0 = disabled, 1 = enabled (default).
5313 pstore.backend= Specify the name of the pstore backend to use
5315 pti= [X86-64] Control Page Table Isolation of user and
5316 kernel address spaces. Disabling this feature
5317 removes hardening, but improves performance of
5318 system calls and interrupts.
5320 on - unconditionally enable
5321 off - unconditionally disable
5322 auto - kernel detects whether your CPU model is
5323 vulnerable to issues that PTI mitigates
5325 Not specifying this option is equivalent to pti=auto.
5327 pty.legacy_count=
5328 [KNL] Number of legacy pty's. Overwrites compiled-in
5329 default number.
5331 quiet [KNL,EARLY] Disable most log messages
5333 r128= [HW,DRM]
5335 radix_hcall_invalidate=on [PPC/PSERIES]
5336 Disable RADIX GTSE feature and use hcall for TLB
5337 invalidate.
5339 raid= [HW,RAID]
5340 See Documentation/admin-guide/md.rst.
5342 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes
5343 See Documentation/admin-guide/blockdev/ramdisk.rst.
5345 ramdisk_start= [RAM] RAM disk image start address
5347 random.trust_cpu=off
5348 [KNL,EARLY] Disable trusting the use of the CPU's
5349 random number generator (if available) to
5350 initialize the kernel's RNG.
5352 random.trust_bootloader=off
5353 [KNL,EARLY] Disable trusting the use of the a seed
5354 passed by the bootloader (if available) to
5355 initialize the kernel's RNG.
5357 randomize_kstack_offset=
5358 [KNL,EARLY] Enable or disable kernel stack offset
5359 randomization, which provides roughly 5 bits of
5360 entropy, frustrating memory corruption attacks
5361 that depend on stack address determinism or
5362 cross-syscall address exposures. This is only
5363 available on architectures that have defined
5364 CONFIG_HAVE_ARCH_RANDOMIZE_KSTACK_OFFSET.
5365 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
5366 Default is CONFIG_RANDOMIZE_KSTACK_OFFSET_DEFAULT.
5368 ras=option[,option,...] [KNL] RAS-specific options
5370 cec_disable [X86]
5371 Disable the Correctable Errors Collector,
5372 see CONFIG_RAS_CEC help text.
5374 rcu_nocbs[=cpu-list]
5375 [KNL] The optional argument is a cpu list,
5376 as described above.
5378 In kernels built with CONFIG_RCU_NOCB_CPU=y,
5379 enable the no-callback CPU mode, which prevents
5380 such CPUs' callbacks from being invoked in
5381 softirq context. Invocation of such CPUs' RCU
5382 callbacks will instead be offloaded to "rcuox/N"
5383 kthreads created for that purpose, where "x" is
5384 "p" for RCU-preempt, "s" for RCU-sched, and "g"
5385 for the kthreads that mediate grace periods; and
5386 "N" is the CPU number. This reduces OS jitter on
5387 the offloaded CPUs, which can be useful for HPC
5388 and real-time workloads. It can also improve
5389 energy efficiency for asymmetric multiprocessors.
5391 If a cpulist is passed as an argument, the specified
5392 list of CPUs is set to no-callback mode from boot.
5394 Otherwise, if the '=' sign and the cpulist
5395 arguments are omitted, no CPU will be set to
5396 no-callback mode from boot but the mode may be
5397 toggled at runtime via cpusets.
5399 Note that this argument takes precedence over
5400 the CONFIG_RCU_NOCB_CPU_DEFAULT_ALL option.
5402 rcu_nocb_poll [KNL]
5403 Rather than requiring that offloaded CPUs
5404 (specified by rcu_nocbs= above) explicitly
5405 awaken the corresponding "rcuoN" kthreads,
5406 make these kthreads poll for callbacks.
5407 This improves the real-time response for the
5408 offloaded CPUs by relieving them of the need to
5409 wake up the corresponding kthread, but degrades
5410 energy efficiency by requiring that the kthreads
5411 periodically wake up to do the polling.
5413 rcutree.blimit= [KNL]
5414 Set maximum number of finished RCU callbacks to
5415 process in one batch.
5417 rcutree.csd_lock_suppress_rcu_stall= [KNL]
5418 Do only a one-line RCU CPU stall warning when
5419 there is an ongoing too-long CSD-lock wait.
5421 rcutree.do_rcu_barrier= [KNL]
5422 Request a call to rcu_barrier(). This is
5423 throttled so that userspace tests can safely
5424 hammer on the sysfs variable if they so choose.
5425 If triggered before the RCU grace-period machinery
5426 is fully active, this will error out with EAGAIN.
5428 rcutree.dump_tree= [KNL]
5429 Dump the structure of the rcu_node combining tree
5430 out at early boot. This is used for diagnostic
5431 purposes, to verify correct tree setup.
5433 rcutree.gp_cleanup_delay= [KNL]
5434 Set the number of jiffies to delay each step of
5435 RCU grace-period cleanup.
5437 rcutree.gp_init_delay= [KNL]
5438 Set the number of jiffies to delay each step of
5439 RCU grace-period initialization.
5441 rcutree.gp_preinit_delay= [KNL]
5442 Set the number of jiffies to delay each step of
5443 RCU grace-period pre-initialization, that is,
5444 the propagation of recent CPU-hotplug changes up
5445 the rcu_node combining tree.
5447 rcutree.jiffies_till_first_fqs= [KNL]
5448 Set delay from grace-period initialization to
5449 first attempt to force quiescent states.
5450 Units are jiffies, minimum value is zero,
5451 and maximum value is HZ.
5453 rcutree.jiffies_till_next_fqs= [KNL]
5454 Set delay between subsequent attempts to force
5455 quiescent states. Units are jiffies, minimum
5456 value is one, and maximum value is HZ.
5458 rcutree.jiffies_till_sched_qs= [KNL]
5459 Set required age in jiffies for a
5460 given grace period before RCU starts
5461 soliciting quiescent-state help from
5462 rcu_note_context_switch() and cond_resched().
5463 If not specified, the kernel will calculate
5464 a value based on the most recent settings
5465 of rcutree.jiffies_till_first_fqs
5466 and rcutree.jiffies_till_next_fqs.
5467 This calculated value may be viewed in
5468 rcutree.jiffies_to_sched_qs. Any attempt to set
5469 rcutree.jiffies_to_sched_qs will be cheerfully
5470 overwritten.
5472 rcutree.kthread_prio= [KNL,BOOT]
5473 Set the SCHED_FIFO priority of the RCU per-CPU
5474 kthreads (rcuc/N). This value is also used for
5475 the priority of the RCU boost threads (rcub/N)
5476 and for the RCU grace-period kthreads (rcu_bh,
5477 rcu_preempt, and rcu_sched). If RCU_BOOST is
5478 set, valid values are 1-99 and the default is 1
5479 (the least-favored priority). Otherwise, when
5480 RCU_BOOST is not set, valid values are 0-99 and
5481 the default is zero (non-realtime operation).
5482 When RCU_NOCB_CPU is set, also adjust the
5483 priority of NOCB callback kthreads.
5485 rcutree.nocb_nobypass_lim_per_jiffy= [KNL]
5486 On callback-offloaded (rcu_nocbs) CPUs,
5487 RCU reduces the lock contention that would
5488 otherwise be caused by callback floods through
5489 use of the ->nocb_bypass list. However, in the
5490 common non-flooded case, RCU queues directly to
5491 the main ->cblist in order to avoid the extra
5492 overhead of the ->nocb_bypass list and its lock.
5493 But if there are too many callbacks queued during
5494 a single jiffy, RCU pre-queues the callbacks into
5495 the ->nocb_bypass queue. The definition of "too
5496 many" is supplied by this kernel boot parameter.
5498 rcutree.nohz_full_patience_delay= [KNL]
5499 On callback-offloaded (rcu_nocbs) CPUs, avoid
5500 disturbing RCU unless the grace period has
5501 reached the specified age in milliseconds.
5502 Defaults to zero. Large values will be capped
5503 at five seconds. All values will be rounded down
5504 to the nearest value representable by jiffies.
5506 rcutree.qhimark= [KNL]
5507 Set threshold of queued RCU callbacks beyond which
5508 batch limiting is disabled.
5510 rcutree.qlowmark= [KNL]
5511 Set threshold of queued RCU callbacks below which
5512 batch limiting is re-enabled.
5514 rcutree.qovld= [KNL]
5515 Set threshold of queued RCU callbacks beyond which
5516 RCU's force-quiescent-state scan will aggressively
5517 enlist help from cond_resched() and sched IPIs to
5518 help CPUs more quickly reach quiescent states.
5519 Set to less than zero to make this be set based
5520 on rcutree.qhimark at boot time and to zero to
5521 disable more aggressive help enlistment.
5523 rcutree.rcu_delay_page_cache_fill_msec= [KNL]
5524 Set the page-cache refill delay (in milliseconds)
5525 in response to low-memory conditions. The range
5526 of permitted values is in the range 0:100000.
5528 rcutree.rcu_divisor= [KNL]
5529 Set the shift-right count to use to compute
5530 the callback-invocation batch limit bl from
5531 the number of callbacks queued on this CPU.
5532 The result will be bounded below by the value of
5533 the rcutree.blimit kernel parameter. Every bl
5534 callbacks, the softirq handler will exit in
5535 order to allow the CPU to do other work.
5537 Please note that this callback-invocation batch
5538 limit applies only to non-offloaded callback
5539 invocation. Offloaded callbacks are instead
5540 invoked in the context of an rcuoc kthread, which
5541 scheduler will preempt as it does any other task.
5543 rcutree.rcu_fanout_exact= [KNL]
5544 Disable autobalancing of the rcu_node combining
5545 tree. This is used by rcutorture, and might
5546 possibly be useful for architectures having high
5547 cache-to-cache transfer latencies.
5549 rcutree.rcu_fanout_leaf= [KNL]
5550 Change the number of CPUs assigned to each
5551 leaf rcu_node structure. Useful for very
5552 large systems, which will choose the value 64,
5553 and for NUMA systems with large remote-access
5554 latencies, which will choose a value aligned
5555 with the appropriate hardware boundaries.
5557 rcutree.rcu_min_cached_objs= [KNL]
5558 Minimum number of objects which are cached and
5559 maintained per one CPU. Object size is equal
5560 to PAGE_SIZE. The cache allows to reduce the
5561 pressure to page allocator, also it makes the
5562 whole algorithm to behave better in low memory
5563 condition.
5565 rcutree.rcu_nocb_gp_stride= [KNL]
5566 Set the number of NOCB callback kthreads in
5567 each group, which defaults to the square root
5568 of the number of CPUs. Larger numbers reduce
5569 the wakeup overhead on the global grace-period
5570 kthread, but increases that same overhead on
5571 each group's NOCB grace-period kthread.
5573 rcutree.rcu_kick_kthreads= [KNL]
5574 Cause the grace-period kthread to get an extra
5575 wake_up() if it sleeps three times longer than
5576 it should at force-quiescent-state time.
5577 This wake_up() will be accompanied by a
5578 WARN_ONCE() splat and an ftrace_dump().
5580 rcutree.rcu_resched_ns= [KNL]
5581 Limit the time spend invoking a batch of RCU
5582 callbacks to the specified number of nanoseconds.
5583 By default, this limit is checked only once
5584 every 32 callbacks in order to limit the pain
5585 inflicted by local_clock() overhead.
5587 rcutree.rcu_unlock_delay= [KNL]
5588 In CONFIG_RCU_STRICT_GRACE_PERIOD=y kernels,
5589 this specifies an rcu_read_unlock()-time delay
5590 in microseconds. This defaults to zero.
5591 Larger delays increase the probability of
5592 catching RCU pointer leaks, that is, buggy use
5593 of RCU-protected pointers after the relevant
5594 rcu_read_unlock() has completed.
5596 rcutree.sysrq_rcu= [KNL]
5597 Commandeer a sysrq key to dump out Tree RCU's
5598 rcu_node tree with an eye towards determining
5599 why a new grace period has not yet started.
5601 rcutree.use_softirq= [KNL]
5602 If set to zero, move all RCU_SOFTIRQ processing to
5603 per-CPU rcuc kthreads. Defaults to a non-zero
5604 value, meaning that RCU_SOFTIRQ is used by default.
5605 Specify rcutree.use_softirq=0 to use rcuc kthreads.
5607 But note that CONFIG_PREEMPT_RT=y kernels disable
5608 this kernel boot parameter, forcibly setting it
5609 to zero.
5611 rcutree.enable_rcu_lazy= [KNL]
5612 To save power, batch RCU callbacks and flush after
5613 delay, memory pressure or callback list growing too
5614 big.
5616 rcutree.rcu_normal_wake_from_gp= [KNL]
5617 Reduces a latency of synchronize_rcu() call. This approach
5618 maintains its own track of synchronize_rcu() callers, so it
5619 does not interact with regular callbacks because it does not
5620 use a call_rcu[_hurry]() path. Please note, this is for a
5621 normal grace period.
5623 How to enable it:
5625 echo 1 > /sys/module/rcutree/parameters/rcu_normal_wake_from_gp
5626 or pass a boot parameter "rcutree.rcu_normal_wake_from_gp=1"
5628 Default is 1 if num_possible_cpus() <= 16 and it is not explicitly
5629 disabled by the boot parameter passing 0.
5631 rcuscale.gp_async= [KNL]
5632 Measure performance of asynchronous
5633 grace-period primitives such as call_rcu().
5635 rcuscale.gp_async_max= [KNL]
5636 Specify the maximum number of outstanding
5637 callbacks per writer thread. When a writer
5638 thread exceeds this limit, it invokes the
5639 corresponding flavor of rcu_barrier() to allow
5640 previously posted callbacks to drain.
5642 rcuscale.gp_exp= [KNL]
5643 Measure performance of expedited synchronous
5644 grace-period primitives.
5646 rcuscale.holdoff= [KNL]
5647 Set test-start holdoff period. The purpose of
5648 this parameter is to delay the start of the
5649 test until boot completes in order to avoid
5650 interference.
5652 rcuscale.kfree_by_call_rcu= [KNL]
5653 In kernels built with CONFIG_RCU_LAZY=y, test
5654 call_rcu() instead of kfree_rcu().
5656 rcuscale.kfree_mult= [KNL]
5657 Instead of allocating an object of size kfree_obj,
5658 allocate one of kfree_mult * sizeof(kfree_obj).
5659 Defaults to 1.
5661 rcuscale.kfree_rcu_test= [KNL]
5662 Set to measure performance of kfree_rcu() flooding.
5664 rcuscale.kfree_rcu_test_double= [KNL]
5665 Test the double-argument variant of kfree_rcu().
5666 If this parameter has the same value as
5667 rcuscale.kfree_rcu_test_single, both the single-
5668 and double-argument variants are tested.
5670 rcuscale.kfree_rcu_test_single= [KNL]
5671 Test the single-argument variant of kfree_rcu().
5672 If this parameter has the same value as
5673 rcuscale.kfree_rcu_test_double, both the single-
5674 and double-argument variants are tested.
5676 rcuscale.kfree_nthreads= [KNL]
5677 The number of threads running loops of kfree_rcu().
5679 rcuscale.kfree_alloc_num= [KNL]
5680 Number of allocations and frees done in an iteration.
5682 rcuscale.kfree_loops= [KNL]
5683 Number of loops doing rcuscale.kfree_alloc_num number
5684 of allocations and frees.
5686 rcuscale.minruntime= [KNL]
5687 Set the minimum test run time in seconds. This
5688 does not affect the data-collection interval,
5689 but instead allows better measurement of things
5690 like CPU consumption.
5692 rcuscale.nreaders= [KNL]
5693 Set number of RCU readers. The value -1 selects
5694 N, where N is the number of CPUs. A value
5695 "n" less than -1 selects N-n+1, where N is again
5696 the number of CPUs. For example, -2 selects N
5697 (the number of CPUs), -3 selects N+1, and so on.
5698 A value of "n" less than or equal to -N selects
5699 a single reader.
5701 rcuscale.nwriters= [KNL]
5702 Set number of RCU writers. The values operate
5703 the same as for rcuscale.nreaders.
5704 N, where N is the number of CPUs
5706 rcuscale.scale_type= [KNL]
5707 Specify the RCU implementation to test.
5709 rcuscale.shutdown= [KNL]
5710 Shut the system down after performance tests
5711 complete. This is useful for hands-off automated
5712 testing.
5714 rcuscale.verbose= [KNL]
5715 Enable additional printk() statements.
5717 rcuscale.writer_holdoff= [KNL]
5718 Write-side holdoff between grace periods,
5719 in microseconds. The default of zero says
5720 no holdoff.
5722 rcuscale.writer_holdoff_jiffies= [KNL]
5723 Additional write-side holdoff between grace
5724 periods, but in jiffies. The default of zero
5725 says no holdoff.
5727 rcutorture.fqs_duration= [KNL]
5728 Set duration of force_quiescent_state bursts
5729 in microseconds.
5731 rcutorture.fqs_holdoff= [KNL]
5732 Set holdoff time within force_quiescent_state bursts
5733 in microseconds.
5735 rcutorture.fqs_stutter= [KNL]
5736 Set wait time between force_quiescent_state bursts
5737 in seconds.
5739 rcutorture.fwd_progress= [KNL]
5740 Specifies the number of kthreads to be used
5741 for RCU grace-period forward-progress testing
5742 for the types of RCU supporting this notion.
5743 Defaults to 1 kthread, values less than zero or
5744 greater than the number of CPUs cause the number
5745 of CPUs to be used.
5747 rcutorture.fwd_progress_div= [KNL]
5748 Specify the fraction of a CPU-stall-warning
5749 period to do tight-loop forward-progress testing.
5751 rcutorture.fwd_progress_holdoff= [KNL]
5752 Number of seconds to wait between successive
5753 forward-progress tests.
5755 rcutorture.fwd_progress_need_resched= [KNL]
5756 Enclose cond_resched() calls within checks for
5757 need_resched() during tight-loop forward-progress
5758 testing.
5760 rcutorture.gp_cond= [KNL]
5761 Use conditional/asynchronous update-side
5762 normal-grace-period primitives, if available.
5764 rcutorture.gp_cond_exp= [KNL]
5765 Use conditional/asynchronous update-side
5766 expedited-grace-period primitives, if available.
5768 rcutorture.gp_cond_full= [KNL]
5769 Use conditional/asynchronous update-side
5770 normal-grace-period primitives that also take
5771 concurrent expedited grace periods into account,
5772 if available.
5774 rcutorture.gp_cond_exp_full= [KNL]
5775 Use conditional/asynchronous update-side
5776 expedited-grace-period primitives that also take
5777 concurrent normal grace periods into account,
5778 if available.
5780 rcutorture.gp_cond_wi= [KNL]
5781 Nominal wait interval for normal conditional
5782 grace periods (specified by rcutorture's
5783 gp_cond and gp_cond_full module parameters),
5784 in microseconds. The actual wait interval will
5785 be randomly selected to nanosecond granularity up
5786 to this wait interval. Defaults to 16 jiffies,
5787 for example, 16,000 microseconds on a system
5788 with HZ=1000.
5790 rcutorture.gp_cond_wi_exp= [KNL]
5791 Nominal wait interval for expedited conditional
5792 grace periods (specified by rcutorture's
5793 gp_cond_exp and gp_cond_exp_full module
5794 parameters), in microseconds. The actual wait
5795 interval will be randomly selected to nanosecond
5796 granularity up to this wait interval. Defaults to
5797 128 microseconds.
5799 rcutorture.gp_exp= [KNL]
5800 Use expedited update-side primitives, if available.
5802 rcutorture.gp_normal= [KNL]
5803 Use normal (non-expedited) asynchronous
5804 update-side primitives, if available.
5806 rcutorture.gp_poll= [KNL]
5807 Use polled update-side normal-grace-period
5808 primitives, if available.
5810 rcutorture.gp_poll_exp= [KNL]
5811 Use polled update-side expedited-grace-period
5812 primitives, if available.
5814 rcutorture.gp_poll_full= [KNL]
5815 Use polled update-side normal-grace-period
5816 primitives that also take concurrent expedited
5817 grace periods into account, if available.
5819 rcutorture.gp_poll_exp_full= [KNL]
5820 Use polled update-side expedited-grace-period
5821 primitives that also take concurrent normal
5822 grace periods into account, if available.
5824 rcutorture.gp_poll_wi= [KNL]
5825 Nominal wait interval for normal conditional
5826 grace periods (specified by rcutorture's
5827 gp_poll and gp_poll_full module parameters),
5828 in microseconds. The actual wait interval will
5829 be randomly selected to nanosecond granularity up
5830 to this wait interval. Defaults to 16 jiffies,
5831 for example, 16,000 microseconds on a system
5832 with HZ=1000.
5834 rcutorture.gp_poll_wi_exp= [KNL]
5835 Nominal wait interval for expedited conditional
5836 grace periods (specified by rcutorture's
5837 gp_poll_exp and gp_poll_exp_full module
5838 parameters), in microseconds. The actual wait
5839 interval will be randomly selected to nanosecond
5840 granularity up to this wait interval. Defaults to
5841 128 microseconds.
5843 rcutorture.gp_sync= [KNL]
5844 Use normal (non-expedited) synchronous
5845 update-side primitives, if available. If all
5846 of rcutorture.gp_cond=, rcutorture.gp_exp=,
5847 rcutorture.gp_normal=, and rcutorture.gp_sync=
5848 are zero, rcutorture acts as if is interpreted
5849 they are all non-zero.
5851 rcutorture.gpwrap_lag= [KNL]
5852 Enable grace-period wrap lag testing. Setting
5853 to false prevents the gpwrap lag test from
5854 running. Default is true.
5856 rcutorture.gpwrap_lag_gps= [KNL]
5857 Set the value for grace-period wrap lag during
5858 active lag testing periods. This controls how many
5859 grace periods differences we tolerate between
5860 rdp and rnp's gp_seq before setting overflow flag.
5861 The default is always set to 8.
5863 rcutorture.gpwrap_lag_cycle_mins= [KNL]
5864 Set the total cycle duration for gpwrap lag
5865 testing in minutes. This is the total time for
5866 one complete cycle of active and inactive
5867 testing periods. Default is 30 minutes.
5869 rcutorture.gpwrap_lag_active_mins= [KNL]
5870 Set the duration for which gpwrap lag is active
5871 within each cycle, in minutes. During this time,
5872 the grace-period wrap lag will be set to the
5873 value specified by gpwrap_lag_gps. Default is
5874 5 minutes.
5876 rcutorture.irqreader= [KNL]
5877 Run RCU readers from irq handlers, or, more
5878 accurately, from a timer handler. Not all RCU
5879 flavors take kindly to this sort of thing.
5881 rcutorture.leakpointer= [KNL]
5882 Leak an RCU-protected pointer out of the reader.
5883 This can of course result in splats, and is
5884 intended to test the ability of things like
5885 CONFIG_RCU_STRICT_GRACE_PERIOD=y to detect
5886 such leaks.
5888 rcutorture.n_barrier_cbs= [KNL]
5889 Set callbacks/threads for rcu_barrier() testing.
5891 rcutorture.nfakewriters= [KNL]
5892 Set number of concurrent RCU writers. These just
5893 stress RCU, they don't participate in the actual
5894 test, hence the "fake".
5896 rcutorture.nocbs_nthreads= [KNL]
5897 Set number of RCU callback-offload togglers.
5898 Zero (the default) disables toggling.
5900 rcutorture.nocbs_toggle= [KNL]
5901 Set the delay in milliseconds between successive
5902 callback-offload toggling attempts.
5904 rcutorture.nreaders= [KNL]
5905 Set number of RCU readers. The value -1 selects
5906 N-1, where N is the number of CPUs. A value
5907 "n" less than -1 selects N-n-2, where N is again
5908 the number of CPUs. For example, -2 selects N
5909 (the number of CPUs), -3 selects N+1, and so on.
5911 rcutorture.object_debug= [KNL]
5912 Enable debug-object double-call_rcu() testing.
5914 rcutorture.onoff_holdoff= [KNL]
5915 Set time (s) after boot for CPU-hotplug testing.
5917 rcutorture.onoff_interval= [KNL]
5918 Set time (jiffies) between CPU-hotplug operations,
5919 or zero to disable CPU-hotplug testing.
5921 rcutorture.preempt_duration= [KNL]
5922 Set duration (in milliseconds) of preemptions
5923 by a high-priority FIFO real-time task. Set to
5924 zero (the default) to disable. The CPUs to
5925 preempt are selected randomly from the set that
5926 are online at a given point in time. Races with
5927 CPUs going offline are ignored, with that attempt
5928 at preemption skipped.
5930 rcutorture.preempt_interval= [KNL]
5931 Set interval (in milliseconds, defaulting to one
5932 second) between preemptions by a high-priority
5933 FIFO real-time task. This delay is mediated
5934 by an hrtimer and is further fuzzed to avoid
5935 inadvertent synchronizations.
5937 rcutorture.read_exit_burst= [KNL]
5938 The number of times in a given read-then-exit
5939 episode that a set of read-then-exit kthreads
5940 is spawned.
5942 rcutorture.read_exit_delay= [KNL]
5943 The delay, in seconds, between successive
5944 read-then-exit testing episodes.
5946 rcutorture.reader_flavor= [KNL]
5947 A bit mask indicating which readers to use.
5948 If there is more than one bit set, the readers
5949 are entered from low-order bit up, and are
5950 exited in the opposite order. For SRCU, the
5951 0x1 bit is normal readers, 0x2 NMI-safe readers,
5952 and 0x4 light-weight readers.
5954 rcutorture.shuffle_interval= [KNL]
5955 Set task-shuffle interval (s). Shuffling tasks
5956 allows some CPUs to go into dyntick-idle mode
5957 during the rcutorture test.
5959 rcutorture.shutdown_secs= [KNL]
5960 Set time (s) after boot system shutdown. This
5961 is useful for hands-off automated testing.
5963 rcutorture.stall_cpu= [KNL]
5964 Duration of CPU stall (s) to test RCU CPU stall
5965 warnings, zero to disable.
5967 rcutorture.stall_cpu_block= [KNL]
5968 Sleep while stalling if set. This will result
5969 in warnings from preemptible RCU in addition to
5970 any other stall-related activity. Note that
5971 in kernels built with CONFIG_PREEMPTION=n and
5972 CONFIG_PREEMPT_COUNT=y, this parameter will
5973 cause the CPU to pass through a quiescent state.
5974 Given CONFIG_PREEMPTION=n, this will suppress
5975 RCU CPU stall warnings, but will instead result
5976 in scheduling-while-atomic splats.
5978 Use of this module parameter results in splats.
5981 rcutorture.stall_cpu_holdoff= [KNL]
5982 Time to wait (s) after boot before inducing stall.
5984 rcutorture.stall_cpu_irqsoff= [KNL]
5985 Disable interrupts while stalling if set, but only
5986 on the first stall in the set.
5988 rcutorture.stall_cpu_repeat= [KNL]
5989 Number of times to repeat the stall sequence,
5990 so that rcutorture.stall_cpu_repeat=3 will result
5991 in four stall sequences.
5993 rcutorture.stall_gp_kthread= [KNL]
5994 Duration (s) of forced sleep within RCU
5995 grace-period kthread to test RCU CPU stall
5996 warnings, zero to disable. If both stall_cpu
5997 and stall_gp_kthread are specified, the
5998 kthread is starved first, then the CPU.
6000 rcutorture.stat_interval= [KNL]
6001 Time (s) between statistics printk()s.
6003 rcutorture.stutter= [KNL]
6004 Time (s) to stutter testing, for example, specifying
6005 five seconds causes the test to run for five seconds,
6006 wait for five seconds, and so on. This tests RCU's
6007 ability to transition abruptly to and from idle.
6009 rcutorture.test_boost= [KNL]
6010 Test RCU priority boosting? 0=no, 1=maybe, 2=yes.
6011 "Maybe" means test if the RCU implementation
6012 under test support RCU priority boosting.
6014 rcutorture.test_boost_duration= [KNL]
6015 Duration (s) of each individual boost test.
6017 rcutorture.test_boost_holdoff= [KNL]
6018 Holdoff time (s) from start of test to the start
6019 of RCU priority-boost testing. Defaults to zero,
6020 that is, no holdoff.
6022 rcutorture.test_boost_interval= [KNL]
6023 Interval (s) between each boost test.
6025 rcutorture.test_no_idle_hz= [KNL]
6026 Test RCU's dyntick-idle handling. See also the
6027 rcutorture.shuffle_interval parameter.
6029 rcutorture.torture_type= [KNL]
6030 Specify the RCU implementation to test.
6032 rcutorture.verbose= [KNL]
6033 Enable additional printk() statements.
6035 rcupdate.rcu_cpu_stall_ftrace_dump= [KNL]
6036 Dump ftrace buffer after reporting RCU CPU
6037 stall warning.
6039 rcupdate.rcu_cpu_stall_notifiers= [KNL]
6040 Provide RCU CPU stall notifiers, but see the
6041 warnings in the RCU_CPU_STALL_NOTIFIER Kconfig
6042 option's help text. TL;DR: You almost certainly
6043 do not want rcupdate.rcu_cpu_stall_notifiers.
6045 rcupdate.rcu_cpu_stall_suppress= [KNL]
6046 Suppress RCU CPU stall warning messages.
6048 rcupdate.rcu_cpu_stall_suppress_at_boot= [KNL]
6049 Suppress RCU CPU stall warning messages and
6050 rcutorture writer stall warnings that occur
6051 during early boot, that is, during the time
6052 before the init task is spawned.
6054 rcupdate.rcu_cpu_stall_timeout= [KNL]
6055 Set timeout for RCU CPU stall warning messages.
6056 The value is in seconds and the maximum allowed
6057 value is 300 seconds.
6059 rcupdate.rcu_exp_cpu_stall_timeout= [KNL]
6060 Set timeout for expedited RCU CPU stall warning
6061 messages. The value is in milliseconds
6062 and the maximum allowed value is 21000
6063 milliseconds. Please note that this value is
6064 adjusted to an arch timer tick resolution.
6065 Setting this to zero causes the value from
6066 rcupdate.rcu_cpu_stall_timeout to be used (after
6067 conversion from seconds to milliseconds).
6069 rcupdate.rcu_cpu_stall_cputime= [KNL]
6070 Provide statistics on the cputime and count of
6071 interrupts and tasks during the sampling period. For
6072 multiple continuous RCU stalls, all sampling periods
6073 begin at half of the first RCU stall timeout.
6075 rcupdate.rcu_exp_stall_task_details= [KNL]
6076 Print stack dumps of any tasks blocking the
6077 current expedited RCU grace period during an
6078 expedited RCU CPU stall warning.
6080 rcupdate.rcu_expedited= [KNL]
6081 Use expedited grace-period primitives, for
6082 example, synchronize_rcu_expedited() instead
6083 of synchronize_rcu(). This reduces latency,
6084 but can increase CPU utilization, degrade
6085 real-time latency, and degrade energy efficiency.
6086 No effect on CONFIG_TINY_RCU kernels.
6088 rcupdate.rcu_normal= [KNL]
6089 Use only normal grace-period primitives,
6090 for example, synchronize_rcu() instead of
6091 synchronize_rcu_expedited(). This improves
6092 real-time latency, CPU utilization, and
6093 energy efficiency, but can expose users to
6094 increased grace-period latency. This parameter
6095 overrides rcupdate.rcu_expedited. No effect on
6096 CONFIG_TINY_RCU kernels.
6098 rcupdate.rcu_normal_after_boot= [KNL]
6099 Once boot has completed (that is, after
6100 rcu_end_inkernel_boot() has been invoked), use
6101 only normal grace-period primitives. No effect
6102 on CONFIG_TINY_RCU kernels.
6104 But note that CONFIG_PREEMPT_RT=y kernels enables
6105 this kernel boot parameter, forcibly setting
6106 it to the value one, that is, converting any
6107 post-boot attempt at an expedited RCU grace
6108 period to instead use normal non-expedited
6109 grace-period processing.
6111 rcupdate.rcu_task_collapse_lim= [KNL]
6112 Set the maximum number of callbacks present
6113 at the beginning of a grace period that allows
6114 the RCU Tasks flavors to collapse back to using
6115 a single callback queue. This switching only
6116 occurs when rcupdate.rcu_task_enqueue_lim is
6117 set to the default value of -1.
6119 rcupdate.rcu_task_contend_lim= [KNL]
6120 Set the minimum number of callback-queuing-time
6121 lock-contention events per jiffy required to
6122 cause the RCU Tasks flavors to switch to per-CPU
6123 callback queuing. This switching only occurs
6124 when rcupdate.rcu_task_enqueue_lim is set to
6125 the default value of -1.
6127 rcupdate.rcu_task_enqueue_lim= [KNL]
6128 Set the number of callback queues to use for the
6129 RCU Tasks family of RCU flavors. The default
6130 of -1 allows this to be automatically (and
6131 dynamically) adjusted. This parameter is intended
6132 for use in testing.
6134 rcupdate.rcu_task_ipi_delay= [KNL]
6135 Set time in jiffies during which RCU tasks will
6136 avoid sending IPIs, starting with the beginning
6137 of a given grace period. Setting a large
6138 number avoids disturbing real-time workloads,
6139 but lengthens grace periods.
6141 rcupdate.rcu_task_lazy_lim= [KNL]
6142 Number of callbacks on a given CPU that will
6143 cancel laziness on that CPU. Use -1 to disable
6144 cancellation of laziness, but be advised that
6145 doing so increases the danger of OOM due to
6146 callback flooding.
6148 rcupdate.rcu_task_stall_info= [KNL]
6149 Set initial timeout in jiffies for RCU task stall
6150 informational messages, which give some indication
6151 of the problem for those not patient enough to
6152 wait for ten minutes. Informational messages are
6153 only printed prior to the stall-warning message
6154 for a given grace period. Disable with a value
6155 less than or equal to zero. Defaults to ten
6156 seconds. A change in value does not take effect
6157 until the beginning of the next grace period.
6159 rcupdate.rcu_task_stall_info_mult= [KNL]
6160 Multiplier for time interval between successive
6161 RCU task stall informational messages for a given
6162 RCU tasks grace period. This value is clamped
6163 to one through ten, inclusive. It defaults to
6164 the value three, so that the first informational
6165 message is printed 10 seconds into the grace
6166 period, the second at 40 seconds, the third at
6167 160 seconds, and then the stall warning at 600
6168 seconds would prevent a fourth at 640 seconds.
6170 rcupdate.rcu_task_stall_timeout= [KNL]
6171 Set timeout in jiffies for RCU task stall
6172 warning messages. Disable with a value less
6173 than or equal to zero. Defaults to ten minutes.
6174 A change in value does not take effect until
6175 the beginning of the next grace period.
6177 rcupdate.rcu_tasks_lazy_ms= [KNL]
6178 Set timeout in milliseconds RCU Tasks asynchronous
6179 callback batching for call_rcu_tasks().
6180 A negative value will take the default. A value
6181 of zero will disable batching. Batching is
6182 always disabled for synchronize_rcu_tasks().
6184 rcupdate.rcu_tasks_trace_lazy_ms= [KNL]
6185 Set timeout in milliseconds RCU Tasks
6186 Trace asynchronous callback batching for
6187 call_rcu_tasks_trace(). A negative value
6188 will take the default. A value of zero will
6189 disable batching. Batching is always disabled
6190 for synchronize_rcu_tasks_trace().
6192 rcupdate.rcu_self_test= [KNL]
6193 Run the RCU early boot self tests
6195 rdinit= [KNL]
6196 Format: <full_path>
6197 Run specified binary instead of /init from the ramdisk,
6198 used for early userspace startup. See initrd.
6200 rdrand= [X86,EARLY]
6201 force - Override the decision by the kernel to hide the
6202 advertisement of RDRAND support (this affects
6203 certain AMD processors because of buggy BIOS
6204 support, specifically around the suspend/resume
6205 path).
6207 rdt= [HW,X86,RDT]
6208 Turn on/off individual RDT features. List is:
6209 cmt, mbmtotal, mbmlocal, l3cat, l3cdp, l2cat, l2cdp,
6210 mba, smba, bmec, abmc.
6211 E.g. to turn on cmt and turn off mba use:
6212 rdt=cmt,!mba
6214 reboot= [KNL]
6215 Format (x86 or x86_64):
6216 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] | d[efault] \
6217 [[,]s[mp]#### \
6218 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \
6219 [[,]f[orce]
6220 Where reboot_mode is one of warm (soft) or cold (hard) or gpio
6221 (prefix with 'panic_' to set mode for panic
6222 reboot only),
6223 reboot_type is one of bios, acpi, kbd, triple, efi, or pci,
6224 reboot_force is either force or not specified,
6225 reboot_cpu is s[mp]#### with #### being the processor
6226 to be used for rebooting.
6228 acpi
6229 Use the ACPI RESET_REG in the FADT. If ACPI is not
6230 configured or the ACPI reset does not work, the reboot
6231 path attempts the reset using the keyboard controller.
6233 bios
6234 Use the CPU reboot vector for warm reset
6236 cold
6237 Set the cold reboot flag
6239 default
6240 There are some built-in platform specific "quirks"
6241 - you may see: "reboot: <name> series board detected.
6242 Selecting <type> for reboots." In the case where you
6243 think the quirk is in error (e.g. you have newer BIOS,
6244 or newer board) using this option will ignore the
6245 built-in quirk table, and use the generic default
6246 reboot actions.
6248 efi
6249 Use efi reset_system runtime service. If EFI is not
6250 configured or the EFI reset does not work, the reboot
6251 path attempts the reset using the keyboard controller.
6253 force
6254 Don't stop other CPUs on reboot. This can make reboot
6255 more reliable in some cases.
6257 kbd
6258 Use the keyboard controller. cold reset (default)
6260 pci
6261 Use a write to the PCI config space register 0xcf9 to
6262 trigger reboot.
6264 triple
6265 Force a triple fault (init)
6267 warm
6268 Don't set the cold reboot flag
6270 Using warm reset will be much faster especially on big
6271 memory systems because the BIOS will not go through
6272 the memory check. Disadvantage is that not all
6273 hardware will be completely reinitialized on reboot so
6274 there may be boot problems on some systems.
6277 refscale.holdoff= [KNL]
6278 Set test-start holdoff period. The purpose of
6279 this parameter is to delay the start of the
6280 test until boot completes in order to avoid
6281 interference.
6283 refscale.lookup_instances= [KNL]
6284 Number of data elements to use for the forms of
6285 SLAB_TYPESAFE_BY_RCU testing. A negative number
6286 is negated and multiplied by nr_cpu_ids, while
6287 zero specifies nr_cpu_ids.
6289 refscale.loops= [KNL]
6290 Set the number of loops over the synchronization
6291 primitive under test. Increasing this number
6292 reduces noise due to loop start/end overhead,
6293 but the default has already reduced the per-pass
6294 noise to a handful of picoseconds on ca. 2020
6295 x86 laptops.
6297 refscale.nreaders= [KNL]
6298 Set number of readers. The default value of -1
6299 selects N, where N is roughly 75% of the number
6300 of CPUs. A value of zero is an interesting choice.
6302 refscale.nruns= [KNL]
6303 Set number of runs, each of which is dumped onto
6304 the console log.
6306 refscale.readdelay= [KNL]
6307 Set the read-side critical-section duration,
6308 measured in microseconds.
6310 refscale.scale_type= [KNL]
6311 Specify the read-protection implementation to test.
6313 refscale.shutdown= [KNL]
6314 Shut down the system at the end of the performance
6315 test. This defaults to 1 (shut it down) when
6316 refscale is built into the kernel and to 0 (leave
6317 it running) when refscale is built as a module.
6319 refscale.verbose= [KNL]
6320 Enable additional printk() statements.
6322 refscale.verbose_batched= [KNL]
6323 Batch the additional printk() statements. If zero
6324 (the default) or negative, print everything. Otherwise,
6325 print every Nth verbose statement, where N is the value
6326 specified.
6328 regulator_ignore_unused
6329 [REGULATOR]
6330 Prevents regulator framework from disabling regulators
6331 that are unused, due no driver claiming them. This may
6332 be useful for debug and development, but should not be
6333 needed on a platform with proper driver support.
6335 relax_domain_level=
6336 [KNL, SMP] Set scheduler's default relax_domain_level.
6337 See Documentation/admin-guide/cgroup-v1/cpusets.rst.
6339 reserve= [KNL,BUGS] Force kernel to ignore I/O ports or memory
6340 Format: <base1>,<size1>[,<base2>,<size2>,...]
6341 Reserve I/O ports or memory so the kernel won't use
6342 them. If <base> is less than 0x10000, the region
6343 is assumed to be I/O ports; otherwise it is memory.
6345 reserve_mem= [RAM]
6346 Format: nn[KMG]:<align>:<label>
6347 Reserve physical memory and label it with a name that
6348 other subsystems can use to access it. This is typically
6349 used for systems that do not wipe the RAM, and this command
6350 line will try to reserve the same physical memory on
6351 soft reboots. Note, it is not guaranteed to be the same
6352 location. For example, if anything about the system changes
6353 or if booting a different kernel. It can also fail if KASLR
6354 places the kernel at the location of where the RAM reservation
6355 was from a previous boot, the new reservation will be at a
6356 different location.
6357 Any subsystem using this feature must add a way to verify
6358 that the contents of the physical memory is from a previous
6359 boot, as there may be cases where the memory will not be
6360 located at the same location.
6362 The format is size:align:label for example, to request
6363 12 megabytes of 4096 alignment for ramoops:
6365 reserve_mem=12M:4096:oops ramoops.mem_name=oops
6367 reservetop= [X86-32,EARLY]
6368 Format: nn[KMG]
6369 Reserves a hole at the top of the kernel virtual
6370 address space.
6372 reset_devices [KNL] Force drivers to reset the underlying device
6373 during initialization.
6375 resume= [SWSUSP]
6376 Specify the partition device for software suspend
6377 Format:
6378 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>}
6380 resume_offset= [SWSUSP]
6381 Specify the offset from the beginning of the partition
6382 given by "resume=" at which the swap header is located,
6383 in <PAGE_SIZE> units (needed only for swap files).
6384 See Documentation/power/swsusp-and-swap-files.rst
6386 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to
6387 read the resume files
6389 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up.
6390 Useful for devices that are detected asynchronously
6391 (e.g. USB and MMC devices).
6393 retain_initrd [RAM] Keep initrd memory after extraction. After boot, it will
6394 be accessible via /sys/firmware/initrd.
6396 retbleed= [X86] Control mitigation of RETBleed (Arbitrary
6397 Speculative Code Execution with Return Instructions)
6398 vulnerability.
6400 AMD-based UNRET and IBPB mitigations alone do not stop
6401 sibling threads from influencing the predictions of other
6402 sibling threads. For that reason, STIBP is used on pro-
6403 cessors that support it, and mitigate SMT on processors
6404 that don't.
6406 off - no mitigation
6407 auto - automatically select a migitation
6408 auto,nosmt - automatically select a mitigation,
6409 disabling SMT if necessary for
6410 the full mitigation (only on Zen1
6411 and older without STIBP).
6412 ibpb - On AMD, mitigate short speculation
6413 windows on basic block boundaries too.
6414 Safe, highest perf impact. It also
6415 enables STIBP if present. Not suitable
6416 on Intel.
6417 ibpb,nosmt - Like "ibpb" above but will disable SMT
6418 when STIBP is not available. This is
6419 the alternative for systems which do not
6420 have STIBP.
6421 unret - Force enable untrained return thunks,
6422 only effective on AMD f15h-f17h based
6423 systems.
6424 unret,nosmt - Like unret, but will disable SMT when STIBP
6425 is not available. This is the alternative for
6426 systems which do not have STIBP.
6428 Selecting 'auto' will choose a mitigation method at run
6429 time according to the CPU.
6431 Not specifying this option is equivalent to retbleed=auto.
6433 rfkill.default_state=
6434 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm,
6435 etc. communication is blocked by default.
6436 1 Unblocked.
6438 rfkill.master_switch_mode=
6439 0 The "airplane mode" button does nothing.
6440 1 The "airplane mode" button toggles between everything
6441 blocked and the previous configuration.
6442 2 The "airplane mode" button toggles between everything
6443 blocked and everything unblocked.
6445 ring3mwait=disable
6446 [KNL] Disable ring 3 MONITOR/MWAIT feature on supported
6447 CPUs.
6449 riscv_isa_fallback [RISCV,EARLY]
6450 When CONFIG_RISCV_ISA_FALLBACK is not enabled, permit
6451 falling back to detecting extension support by parsing
6452 "riscv,isa" property on devicetree systems when the
6453 replacement properties are not found. See the Kconfig
6454 entry for RISCV_ISA_FALLBACK.
6456 ro [KNL] Mount root device read-only on boot
6458 rodata= [KNL,EARLY]
6459 on Mark read-only kernel memory as read-only (default).
6460 off Leave read-only kernel memory writable for debugging.
6461 noalias Mark read-only kernel memory as read-only but retain
6462 writable aliases in the direct map for regions outside
6463 of the kernel image. [arm64]
6465 rockchip.usb_uart
6466 [EARLY]
6467 Enable the uart passthrough on the designated usb port
6468 on Rockchip SoCs. When active, the signals of the
6469 debug-uart get routed to the D+ and D- pins of the usb
6470 port and the regular usb controller gets disabled.
6472 root= [KNL] Root filesystem
6473 Usually this is a block device specifier of some kind,
6474 see the early_lookup_bdev comment in
6475 block/early-lookup.c for details.
6476 Alternatively this can be "ram" for the legacy initial
6477 ramdisk, "nfs" and "cifs" for root on a network file
6478 system, or "mtd" and "ubi" for mounting from raw flash.
6480 rootdelay= [KNL] Delay (in seconds) to pause before attempting to
6481 mount the root filesystem
6483 rootflags= [KNL] Set root filesystem mount option string
6485 initramfs_options= [KNL]
6486 Specify mount options for for the initramfs mount.
6488 rootfstype= [KNL] Set root filesystem type
6490 rootwait [KNL] Wait (indefinitely) for root device to show up.
6491 Useful for devices that are detected asynchronously
6492 (e.g. USB and MMC devices).
6494 rootwait= [KNL] Maximum time (in seconds) to wait for root device
6495 to show up before attempting to mount the root
6496 filesystem.
6498 rproc_mem=nn[KMG][@address]
6499 [KNL,ARM,CMA] Remoteproc physical memory block.
6500 Memory area to be used by remote processor image,
6501 managed by CMA.
6503 rt_group_sched= [KNL] Enable or disable SCHED_RR/FIFO group scheduling
6504 when CONFIG_RT_GROUP_SCHED=y. Defaults to
6505 !CONFIG_RT_GROUP_SCHED_DEFAULT_DISABLED.
6506 Format: <bool>
6508 rw [KNL] Mount root device read-write on boot
6510 S [KNL] Run init in single mode
6512 s390_iommu= [HW,S390]
6513 Set s390 IOTLB flushing mode
6514 strict
6515 With strict flushing every unmap operation will result
6516 in an IOTLB flush. Default is lazy flushing before
6517 reuse, which is faster. Deprecated, equivalent to
6518 iommu.strict=1.
6520 s390_iommu_aperture= [KNL,S390]
6521 Specifies the size of the per device DMA address space
6522 accessible through the DMA and IOMMU APIs as a decimal
6523 factor of the size of main memory.
6524 The default is 1 meaning that one can concurrently use
6525 as many DMA addresses as physical memory is installed,
6526 if supported by hardware, and thus map all of memory
6527 once. With a value of 2 one can map all of memory twice
6528 and so on. As a special case a factor of 0 imposes no
6529 restrictions other than those given by hardware at the
6530 cost of significant additional memory use for tables.
6532 sa1100ir [NET]
6533 See drivers/net/irda/sa1100_ir.c.
6535 sched_proxy_exec= [KNL]
6536 Enables or disables "proxy execution" style
6537 solution to mutex-based priority inversion.
6538 Format: <bool>
6540 sched_verbose [KNL,EARLY] Enables verbose scheduler debug messages.
6542 schedstats= [KNL,X86] Enable or disable scheduled statistics.
6543 Allowed values are enable and disable. This feature
6544 incurs a small amount of overhead in the scheduler
6545 but is useful for debugging and performance tuning.
6547 sched_thermal_decay_shift=
6548 [Deprecated]
6549 [KNL, SMP] Set a decay shift for scheduler thermal
6550 pressure signal. Thermal pressure signal follows the
6551 default decay period of other scheduler pelt
6552 signals(usually 32 ms but configurable). Setting
6553 sched_thermal_decay_shift will left shift the decay
6554 period for the thermal pressure signal by the shift
6555 value.
6556 i.e. with the default pelt decay period of 32 ms
6557 sched_thermal_decay_shift thermal pressure decay pr
6558 1 64 ms
6559 2 128 ms
6560 and so on.
6561 Format: integer between 0 and 10
6562 Default is 0.
6564 scftorture.holdoff= [KNL]
6565 Number of seconds to hold off before starting
6566 test. Defaults to zero for module insertion and
6567 to 10 seconds for built-in smp_call_function()
6568 tests.
6570 scftorture.longwait= [KNL]
6571 Request ridiculously long waits randomly selected
6572 up to the chosen limit in seconds. Zero (the
6573 default) disables this feature. Please note
6574 that requesting even small non-zero numbers of
6575 seconds can result in RCU CPU stall warnings,
6576 softlockup complaints, and so on.
6578 scftorture.nthreads= [KNL]
6579 Number of kthreads to spawn to invoke the
6580 smp_call_function() family of functions.
6581 The default of -1 specifies a number of kthreads
6582 equal to the number of CPUs.
6584 scftorture.onoff_holdoff= [KNL]
6585 Number seconds to wait after the start of the
6586 test before initiating CPU-hotplug operations.
6588 scftorture.onoff_interval= [KNL]
6589 Number seconds to wait between successive
6590 CPU-hotplug operations. Specifying zero (which
6591 is the default) disables CPU-hotplug operations.
6593 scftorture.shutdown_secs= [KNL]
6594 The number of seconds following the start of the
6595 test after which to shut down the system. The
6596 default of zero avoids shutting down the system.
6597 Non-zero values are useful for automated tests.
6599 scftorture.stat_interval= [KNL]
6600 The number of seconds between outputting the
6601 current test statistics to the console. A value
6602 of zero disables statistics output.
6604 scftorture.stutter_cpus= [KNL]
6605 The number of jiffies to wait between each change
6606 to the set of CPUs under test.
6608 scftorture.use_cpus_read_lock= [KNL]
6609 Use use_cpus_read_lock() instead of the default
6610 preempt_disable() to disable CPU hotplug
6611 while invoking one of the smp_call_function*()
6612 functions.
6614 scftorture.verbose= [KNL]
6615 Enable additional printk() statements.
6617 scftorture.weight_single= [KNL]
6618 The probability weighting to use for the
6619 smp_call_function_single() function with a zero
6620 "wait" parameter. A value of -1 selects the
6621 default if all other weights are -1. However,
6622 if at least one weight has some other value, a
6623 value of -1 will instead select a weight of zero.
6625 scftorture.weight_single_wait= [KNL]
6626 The probability weighting to use for the
6627 smp_call_function_single() function with a
6628 non-zero "wait" parameter. See weight_single.
6630 scftorture.weight_many= [KNL]
6631 The probability weighting to use for the
6632 smp_call_function_many() function with a zero
6633 "wait" parameter. See weight_single.
6634 Note well that setting a high probability for
6635 this weighting can place serious IPI load
6636 on the system.
6638 scftorture.weight_many_wait= [KNL]
6639 The probability weighting to use for the
6640 smp_call_function_many() function with a
6641 non-zero "wait" parameter. See weight_single
6642 and weight_many.
6644 scftorture.weight_all= [KNL]
6645 The probability weighting to use for the
6646 smp_call_function_all() function with a zero
6647 "wait" parameter. See weight_single and
6648 weight_many.
6650 scftorture.weight_all_wait= [KNL]
6651 The probability weighting to use for the
6652 smp_call_function_all() function with a
6653 non-zero "wait" parameter. See weight_single
6654 and weight_many.
6656 sdw_mclk_divider=[SDW]
6657 Specify the MCLK divider for Intel SoundWire buses in
6658 case the BIOS does not provide the clock rate properly.
6660 skew_tick= [KNL,EARLY] Offset the periodic timer tick per cpu to mitigate
6661 xtime_lock contention on larger systems, and/or RCU lock
6662 contention on all systems with CONFIG_MAXSMP set.
6663 Format: { "0" | "1" }
6664 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1"
6665 1 -- enable.
6666 Note: increases power consumption, thus should only be
6667 enabled if running jitter sensitive (HPC/RT) workloads.
6669 security= [SECURITY] Choose a legacy "major" security module to
6670 enable at boot. This has been deprecated by the
6671 "lsm=" parameter.
6673 selinux= [SELINUX] Disable or enable SELinux at boot time.
6674 Format: { "0" | "1" }
6675 See security/selinux/Kconfig help text.
6676 0 -- disable.
6677 1 -- enable.
6678 Default value is 1.
6680 serialnumber [BUGS=X86-32]
6682 sev=option[,option...] [X86-64]
6684 debug
6685 Enable debug messages.
6687 nosnp
6688 Do not enable SEV-SNP (applies to host/hypervisor
6689 only). Setting 'nosnp' avoids the RMP check overhead
6690 in memory accesses when users do not want to run
6691 SEV-SNP guests.
6693 shapers= [NET]
6694 Maximal number of shapers.
6696 show_lapic= [APIC,X86] Advanced Programmable Interrupt Controller
6697 Limit apic dumping. The parameter defines the maximal
6698 number of local apics being dumped. Also it is possible
6699 to set it to "all" by meaning -- no limit here.
6700 Format: { 1 (default) | 2 | ... | all }.
6701 The parameter valid if only apic=debug or
6702 apic=verbose is specified.
6703 Example: apic=debug show_lapic=all
6705 slab_debug[=options[,slabs][;[options[,slabs]]...] [MM]
6706 Enabling slab_debug allows one to determine the
6707 culprit if slab objects become corrupted. Enabling
6708 slab_debug can create guard zones around objects and
6709 may poison objects when not in use. Also tracks the
6710 last alloc / free. For more information see
6711 Documentation/admin-guide/mm/slab.rst.
6712 (slub_debug legacy name also accepted for now)
6714 Using this option implies the "no_hash_pointers"
6715 option which can be undone by adding the
6716 "hash_pointers=always" option.
6718 slab_max_order= [MM]
6719 Determines the maximum allowed order for slabs.
6720 A high setting may cause OOMs due to memory
6721 fragmentation. For more information see
6722 Documentation/admin-guide/mm/slab.rst.
6723 (slub_max_order legacy name also accepted for now)
6725 slab_merge [MM]
6726 Enable merging of slabs with similar size when the
6727 kernel is built without CONFIG_SLAB_MERGE_DEFAULT.
6728 (slub_merge legacy name also accepted for now)
6730 slab_min_objects= [MM]
6731 The minimum number of objects per slab. SLUB will
6732 increase the slab order up to slab_max_order to
6733 generate a sufficiently large slab able to contain
6734 the number of objects indicated. The higher the number
6735 of objects the smaller the overhead of tracking slabs
6736 and the less frequently locks need to be acquired.
6737 For more information see
6738 Documentation/admin-guide/mm/slab.rst.
6739 (slub_min_objects legacy name also accepted for now)
6741 slab_min_order= [MM]
6742 Determines the minimum page order for slabs. Must be
6743 lower or equal to slab_max_order. For more information see
6744 Documentation/admin-guide/mm/slab.rst.
6745 (slub_min_order legacy name also accepted for now)
6747 slab_nomerge [MM]
6748 Disable merging of slabs with similar size. May be
6749 necessary if there is some reason to distinguish
6750 allocs to different slabs, especially in hardened
6751 environments where the risk of heap overflows and
6752 layout control by attackers can usually be
6753 frustrated by disabling merging. This will reduce
6754 most of the exposure of a heap attack to a single
6755 cache (risks via metadata attacks are mostly
6756 unchanged). Debug options disable merging on their
6757 own.
6758 For more information see
6759 Documentation/admin-guide/mm/slab.rst.
6760 (slub_nomerge legacy name also accepted for now)
6762 slab_strict_numa [MM]
6763 Support memory policies on a per object level
6764 in the slab allocator. The default is for memory
6765 policies to be applied at the folio level when
6766 a new folio is needed or a partial folio is
6767 retrieved from the lists. Increases overhead
6768 in the slab fastpaths but gains more accurate
6769 NUMA kernel object placement which helps with slow
6770 interconnects in NUMA systems.
6772 slram= [HW,MTD]
6774 smart2= [HW]
6775 Format: <io1>[,<io2>[,...,<io8>]]
6777 smp.csd_lock_timeout= [KNL]
6778 Specify the period of time in milliseconds
6779 that smp_call_function() and friends will wait
6780 for a CPU to release the CSD lock. This is
6781 useful when diagnosing bugs involving CPUs
6782 disabling interrupts for extended periods
6783 of time. Defaults to 5,000 milliseconds, and
6784 setting a value of zero disables this feature.
6785 This feature may be more efficiently disabled
6786 using the csdlock_debug- kernel parameter.
6788 smp.panic_on_ipistall= [KNL]
6789 If a csd_lock_timeout extends for more than
6790 the specified number of milliseconds, panic the
6791 system. By default, let CSD-lock acquisition
6792 take as long as they take. Specifying 300,000
6793 for this value provides a 5-minute timeout.
6795 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices
6796 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port
6797 smsc-ircc2.ircc_sir= [HW] SIR base I/O port
6798 smsc-ircc2.ircc_fir= [HW] FIR base I/O port
6799 smsc-ircc2.ircc_irq= [HW] IRQ line
6800 smsc-ircc2.ircc_dma= [HW] DMA channel
6801 smsc-ircc2.ircc_transceiver= [HW] Transceiver type:
6802 0: Toshiba Satellite 1800 (GP data pin select)
6803 1: Fast pin select (default)
6804 2: ATC IRMode
6806 smt= [KNL,MIPS,S390,EARLY] Set the maximum number of threads
6807 (logical CPUs) to use per physical CPU on systems
6808 capable of symmetric multithreading (SMT). Will
6809 be capped to the actual hardware limit.
6810 Format: <integer>
6811 Default: -1 (no limit)
6813 softlockup_panic=
6814 [KNL] Should the soft-lockup detector generate panics.
6815 Format: 0 | 1
6817 A value of 1 instructs the soft-lockup detector
6818 to panic the machine when a soft-lockup occurs. It is
6819 also controlled by the kernel.softlockup_panic sysctl
6820 and CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC, which is the
6821 respective build-time switch to that functionality.
6823 softlockup_all_cpu_backtrace=
6824 [KNL] Should the soft-lockup detector generate
6825 backtraces on all cpus.
6826 Format: 0 | 1
6828 sonypi.*= [HW] Sony Programmable I/O Control Device driver
6829 See Documentation/admin-guide/laptops/sonypi.rst
6831 spectre_bhi= [X86] Control mitigation of Branch History Injection
6832 (BHI) vulnerability. This setting affects the
6833 deployment of the HW BHI control and the SW BHB
6834 clearing sequence.
6836 on - (default) Enable the HW or SW mitigation as
6837 needed. This protects the kernel from
6838 both syscalls and VMs.
6839 vmexit - On systems which don't have the HW mitigation
6840 available, enable the SW mitigation on vmexit
6841 ONLY. On such systems, the host kernel is
6842 protected from VM-originated BHI attacks, but
6843 may still be vulnerable to syscall attacks.
6844 off - Disable the mitigation.
6846 spectre_v2= [X86,EARLY] Control mitigation of Spectre variant 2
6847 (indirect branch speculation) vulnerability.
6848 The default operation protects the kernel from
6849 user space attacks.
6851 on - unconditionally enable, implies
6852 spectre_v2_user=on
6853 off - unconditionally disable, implies
6854 spectre_v2_user=off
6855 auto - kernel detects whether your CPU model is
6856 vulnerable
6858 Selecting 'on' will, and 'auto' may, choose a
6859 mitigation method at run time according to the
6860 CPU, the available microcode, the setting of the
6861 CONFIG_MITIGATION_RETPOLINE configuration option,
6862 and the compiler with which the kernel was built.
6864 Selecting 'on' will also enable the mitigation
6865 against user space to user space task attacks.
6866 Selecting specific mitigation does not force enable
6867 user mitigations.
6869 Selecting 'off' will disable both the kernel and
6870 the user space protections.
6872 Specific mitigations can also be selected manually:
6874 retpoline - replace indirect branches
6875 retpoline,generic - Retpolines
6876 retpoline,lfence - LFENCE; indirect branch
6877 retpoline,amd - alias for retpoline,lfence
6878 eibrs - Enhanced/Auto IBRS
6879 eibrs,retpoline - Enhanced/Auto IBRS + Retpolines
6880 eibrs,lfence - Enhanced/Auto IBRS + LFENCE
6881 ibrs - use IBRS to protect kernel
6883 Not specifying this option is equivalent to
6884 spectre_v2=auto.
6886 spectre_v2_user=
6887 [X86] Control mitigation of Spectre variant 2
6888 (indirect branch speculation) vulnerability between
6889 user space tasks
6891 on - Unconditionally enable mitigations. Is
6892 enforced by spectre_v2=on
6894 off - Unconditionally disable mitigations. Is
6895 enforced by spectre_v2=off
6897 prctl - Indirect branch speculation is enabled,
6898 but mitigation can be enabled via prctl
6899 per thread. The mitigation control state
6900 is inherited on fork.
6902 prctl,ibpb
6903 - Like "prctl" above, but only STIBP is
6904 controlled per thread. IBPB is issued
6905 always when switching between different user
6906 space processes.
6908 seccomp
6909 - Same as "prctl" above, but all seccomp
6910 threads will enable the mitigation unless
6911 they explicitly opt out.
6913 seccomp,ibpb
6914 - Like "seccomp" above, but only STIBP is
6915 controlled per thread. IBPB is issued
6916 always when switching between different
6917 user space processes.
6919 auto - Kernel selects the mitigation depending on
6920 the available CPU features and vulnerability.
6922 Default mitigation: "prctl"
6924 Not specifying this option is equivalent to
6925 spectre_v2_user=auto.
6927 spec_rstack_overflow=
6928 [X86,EARLY] Control RAS overflow mitigation on AMD Zen CPUs
6930 off - Disable mitigation
6931 microcode - Enable microcode mitigation only
6932 safe-ret - Enable sw-only safe RET mitigation (default)
6933 ibpb - Enable mitigation by issuing IBPB on
6934 kernel entry
6935 ibpb-vmexit - Issue IBPB only on VMEXIT
6936 (cloud-specific mitigation)
6938 spec_store_bypass_disable=
6939 [HW,EARLY] Control Speculative Store Bypass (SSB) Disable mitigation
6940 (Speculative Store Bypass vulnerability)
6942 Certain CPUs are vulnerable to an exploit against a
6943 a common industry wide performance optimization known
6944 as "Speculative Store Bypass" in which recent stores
6945 to the same memory location may not be observed by
6946 later loads during speculative execution. The idea
6947 is that such stores are unlikely and that they can
6948 be detected prior to instruction retirement at the
6949 end of a particular speculation execution window.
6951 In vulnerable processors, the speculatively forwarded
6952 store can be used in a cache side channel attack, for
6953 example to read memory to which the attacker does not
6954 directly have access (e.g. inside sandboxed code).
6956 This parameter controls whether the Speculative Store
6957 Bypass optimization is used.
6959 On x86 the options are:
6961 on - Unconditionally disable Speculative Store Bypass
6962 off - Unconditionally enable Speculative Store Bypass
6963 auto - Kernel detects whether the CPU model contains an
6964 implementation of Speculative Store Bypass and
6965 picks the most appropriate mitigation. If the
6966 CPU is not vulnerable, "off" is selected. If the
6967 CPU is vulnerable the default mitigation is
6968 architecture and Kconfig dependent. See below.
6969 prctl - Control Speculative Store Bypass per thread
6970 via prctl. Speculative Store Bypass is enabled
6971 for a process by default. The state of the control
6972 is inherited on fork.
6973 seccomp - Same as "prctl" above, but all seccomp threads
6974 will disable SSB unless they explicitly opt out.
6976 Default mitigations:
6977 X86: "prctl"
6979 On powerpc the options are:
6981 on,auto - On Power8 and Power9 insert a store-forwarding
6982 barrier on kernel entry and exit. On Power7
6983 perform a software flush on kernel entry and
6984 exit.
6985 off - No action.
6987 Not specifying this option is equivalent to
6988 spec_store_bypass_disable=auto.
6990 split_lock_detect=
6991 [X86] Enable split lock detection or bus lock detection
6993 When enabled (and if hardware support is present), atomic
6994 instructions that access data across cache line
6995 boundaries will result in an alignment check exception
6996 for split lock detection or a debug exception for
6997 bus lock detection.
6999 off - not enabled
7001 warn - the kernel will emit rate-limited warnings
7002 about applications triggering the #AC
7003 exception or the #DB exception. This mode is
7004 the default on CPUs that support split lock
7005 detection or bus lock detection. Default
7006 behavior is by #AC if both features are
7007 enabled in hardware.
7009 fatal - the kernel will send SIGBUS to applications
7010 that trigger the #AC exception or the #DB
7011 exception. Default behavior is by #AC if
7012 both features are enabled in hardware.
7014 ratelimit:N -
7015 Set system wide rate limit to N bus locks
7016 per second for bus lock detection.
7017 0 < N <= 1000.
7019 N/A for split lock detection.
7022 If an #AC exception is hit in the kernel or in
7023 firmware (i.e. not while executing in user mode)
7024 the kernel will oops in either "warn" or "fatal"
7025 mode.
7027 #DB exception for bus lock is triggered only when
7028 CPL > 0.
7030 srbds= [X86,INTEL,EARLY]
7031 Control the Special Register Buffer Data Sampling
7032 (SRBDS) mitigation.
7034 Certain CPUs are vulnerable to an MDS-like
7035 exploit which can leak bits from the random
7036 number generator.
7038 By default, this issue is mitigated by
7039 microcode. However, the microcode fix can cause
7040 the RDRAND and RDSEED instructions to become
7041 much slower. Among other effects, this will
7042 result in reduced throughput from /dev/urandom.
7044 The microcode mitigation can be disabled with
7045 the following option:
7047 off: Disable mitigation and remove
7048 performance impact to RDRAND and RDSEED
7050 srcutree.big_cpu_lim [KNL]
7051 Specifies the number of CPUs constituting a
7052 large system, such that srcu_struct structures
7053 should immediately allocate an srcu_node array.
7054 This kernel-boot parameter defaults to 128,
7055 but takes effect only when the low-order four
7056 bits of srcutree.convert_to_big is equal to 3
7057 (decide at boot).
7059 srcutree.convert_to_big [KNL]
7060 Specifies under what conditions an SRCU tree
7061 srcu_struct structure will be converted to big
7062 form, that is, with an rcu_node tree:
7064 0: Never.
7065 1: At init_srcu_struct() time.
7066 2: When rcutorture decides to.
7067 3: Decide at boot time (default).
7068 0x1X: Above plus if high contention.
7070 Either way, the srcu_node tree will be sized based
7071 on the actual runtime number of CPUs (nr_cpu_ids)
7072 instead of the compile-time CONFIG_NR_CPUS.
7074 srcutree.counter_wrap_check [KNL]
7075 Specifies how frequently to check for
7076 grace-period sequence counter wrap for the
7077 srcu_data structure's ->srcu_gp_seq_needed field.
7078 The greater the number of bits set in this kernel
7079 parameter, the less frequently counter wrap will
7080 be checked for. Note that the bottom two bits
7081 are ignored.
7083 srcutree.exp_holdoff [KNL]
7084 Specifies how many nanoseconds must elapse
7085 since the end of the last SRCU grace period for
7086 a given srcu_struct until the next normal SRCU
7087 grace period will be considered for automatic
7088 expediting. Set to zero to disable automatic
7089 expediting.
7091 srcutree.srcu_max_nodelay [KNL]
7092 Specifies the number of no-delay instances
7093 per jiffy for which the SRCU grace period
7094 worker thread will be rescheduled with zero
7095 delay. Beyond this limit, worker thread will
7096 be rescheduled with a sleep delay of one jiffy.
7098 srcutree.srcu_max_nodelay_phase [KNL]
7099 Specifies the per-grace-period phase, number of
7100 non-sleeping polls of readers. Beyond this limit,
7101 grace period worker thread will be rescheduled
7102 with a sleep delay of one jiffy, between each
7103 rescan of the readers, for a grace period phase.
7105 srcutree.srcu_retry_check_delay [KNL]
7106 Specifies number of microseconds of non-sleeping
7107 delay between each non-sleeping poll of readers.
7109 srcutree.small_contention_lim [KNL]
7110 Specifies the number of update-side contention
7111 events per jiffy will be tolerated before
7112 initiating a conversion of an srcu_struct
7113 structure to big form. Note that the value of
7114 srcutree.convert_to_big must have the 0x10 bit
7115 set for contention-based conversions to occur.
7117 ssbd= [ARM64,HW,EARLY]
7118 Speculative Store Bypass Disable control
7120 On CPUs that are vulnerable to the Speculative
7121 Store Bypass vulnerability and offer a
7122 firmware based mitigation, this parameter
7123 indicates how the mitigation should be used:
7125 force-on: Unconditionally enable mitigation for
7126 for both kernel and userspace
7127 force-off: Unconditionally disable mitigation for
7128 for both kernel and userspace
7129 kernel: Always enable mitigation in the
7130 kernel, and offer a prctl interface
7131 to allow userspace to register its
7132 interest in being mitigated too.
7134 stack_guard_gap= [MM]
7135 override the default stack gap protection. The value
7136 is in page units and it defines how many pages prior
7137 to (for stacks growing down) resp. after (for stacks
7138 growing up) the main stack are reserved for no other
7139 mapping. Default value is 256 pages.
7141 stack_depot_disable= [KNL,EARLY]
7142 Setting this to true through kernel command line will
7143 disable the stack depot thereby saving the static memory
7144 consumed by the stack hash table. By default this is set
7145 to false.
7147 stack_depot_max_pools= [KNL,EARLY]
7148 Specify the maximum number of pools to use for storing
7149 stack traces. Pools are allocated on-demand up to this
7150 limit. Default value is 8191 pools.
7152 stacktrace [FTRACE]
7153 Enabled the stack tracer on boot up.
7155 stacktrace_filter=[function-list]
7156 [FTRACE] Limit the functions that the stack tracer
7157 will trace at boot up. function-list is a comma-separated
7158 list of functions. This list can be changed at run
7159 time by the stack_trace_filter file in the debugfs
7160 tracing directory. Note, this enables stack tracing
7161 and the stacktrace above is not needed.
7163 sti= [PARISC,HW]
7164 Format: <num>
7165 Set the STI (builtin display/keyboard on the HP-PARISC
7166 machines) console (graphic card) which should be used
7167 as the initial boot-console.
7168 See also comment in drivers/video/console/sticore.c.
7170 sti_font= [HW]
7171 See comment in drivers/video/console/sticore.c.
7173 stifb= [HW]
7174 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]]
7176 strict_sas_size=
7177 [X86]
7178 Format: <bool>
7179 Enable or disable strict sigaltstack size checks
7180 against the required signal frame size which
7181 depends on the supported FPU features. This can
7182 be used to filter out binaries which have
7183 not yet been made aware of AT_MINSIGSTKSZ.
7185 stress_hpt [PPC,EARLY]
7186 Limits the number of kernel HPT entries in the hash
7187 page table to increase the rate of hash page table
7188 faults on kernel addresses.
7190 stress_slb [PPC,EARLY]
7191 Limits the number of kernel SLB entries, and flushes
7192 them frequently to increase the rate of SLB faults
7193 on kernel addresses.
7195 sunrpc.min_resvport=
7196 sunrpc.max_resvport=
7197 [NFS,SUNRPC]
7198 SunRPC servers often require that client requests
7199 originate from a privileged port (i.e. a port in the
7200 range 0 < portnr < 1024).
7201 An administrator who wishes to reserve some of these
7202 ports for other uses may adjust the range that the
7203 kernel's sunrpc client considers to be privileged
7204 using these two parameters to set the minimum and
7205 maximum port values.
7207 sunrpc.svc_rpc_per_connection_limit=
7208 [NFS,SUNRPC]
7209 Limit the number of requests that the server will
7210 process in parallel from a single connection.
7211 The default value is 0 (no limit).
7213 sunrpc.pool_mode=
7214 [NFS]
7215 Control how the NFS server code allocates CPUs to
7216 service thread pools. Depending on how many NICs
7217 you have and where their interrupts are bound, this
7218 option will affect which CPUs will do NFS serving.
7219 Note: this parameter cannot be changed while the
7220 NFS server is running.
7222 auto the server chooses an appropriate mode
7223 automatically using heuristics
7224 global a single global pool contains all CPUs
7225 percpu one pool for each CPU
7226 pernode one pool for each NUMA node (equivalent
7227 to global on non-NUMA machines)
7229 sunrpc.tcp_slot_table_entries=
7230 sunrpc.udp_slot_table_entries=
7231 [NFS,SUNRPC]
7232 Sets the upper limit on the number of simultaneous
7233 RPC calls that can be sent from the client to a
7234 server. Increasing these values may allow you to
7235 improve throughput, but will also increase the
7236 amount of memory reserved for use by the client.
7238 suspend.pm_test_delay=
7239 [SUSPEND]
7240 Sets the number of seconds to remain in a suspend test
7241 mode before resuming the system (see
7242 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG
7243 is set. Default value is 5.
7245 svm= [PPC]
7246 Format: { on | off | y | n | 1 | 0 }
7247 This parameter controls use of the Protected
7248 Execution Facility on pSeries.
7250 swiotlb= [ARM,PPC,MIPS,X86,S390,EARLY]
7251 Format: { <int> [,<int>] | force | noforce }
7252 <int> -- Number of I/O TLB slabs
7253 <int> -- Second integer after comma. Number of swiotlb
7254 areas with their own lock. Will be rounded up
7255 to a power of 2.
7256 force -- force using of bounce buffers even if they
7257 wouldn't be automatically used by the kernel
7258 noforce -- Never use bounce buffers (for debugging)
7260 switches= [HW,M68k,EARLY]
7262 sysctl.*= [KNL]
7263 Set a sysctl parameter, right before loading the init
7264 process, as if the value was written to the respective
7265 /proc/sys/... file. Both '.' and '/' are recognized as
7266 separators. Unrecognized parameters and invalid values
7267 are reported in the kernel log. Sysctls registered
7268 later by a loaded module cannot be set this way.
7269 Example: sysctl.vm.swappiness=40
7271 sysrq_always_enabled
7272 [KNL]
7273 Ignore sysrq setting - this boot parameter will
7274 neutralize any effect of /proc/sys/kernel/sysrq.
7275 Useful for debugging.
7277 tcpmhash_entries= [KNL,NET]
7278 Set the number of tcp_metrics_hash slots.
7279 Default value is 8192 or 16384 depending on total
7280 ram pages. This is used to specify the TCP metrics
7281 cache size. See Documentation/networking/ip-sysctl.rst
7282 "tcp_no_metrics_save" section for more details.
7284 tdfx= [HW,DRM]
7286 test_suspend= [SUSPEND]
7287 Format: { "mem" | "standby" | "freeze" }[,N]
7288 Specify "mem" (for Suspend-to-RAM) or "standby" (for
7289 standby suspend) or "freeze" (for suspend type freeze)
7290 as the system sleep state during system startup with
7291 the optional capability to repeat N number of times.
7292 The system is woken from this state using a
7293 wakeup-capable RTC alarm.
7295 thash_entries= [KNL,NET]
7296 Set number of hash buckets for TCP connection
7298 thermal.act= [HW,ACPI]
7299 -1: disable all active trip points in all thermal zones
7300 <degrees C>: override all lowest active trip points
7302 thermal.crt= [HW,ACPI]
7303 -1: disable all critical trip points in all thermal zones
7304 <degrees C>: override all critical trip points
7306 thermal.off= [HW,ACPI]
7307 1: disable ACPI thermal control
7309 thermal.psv= [HW,ACPI]
7310 -1: disable all passive trip points
7311 <degrees C>: override all passive trip points to this
7312 value
7314 thermal.tzp= [HW,ACPI]
7315 Specify global default ACPI thermal zone polling rate
7316 <deci-seconds>: poll all this frequency
7317 0: no polling (default)
7319 thp_anon= [KNL]
7320 Format: <size>[KMG],<size>[KMG]:<state>;<size>[KMG]-<size>[KMG]:<state>
7321 state is one of "always", "madvise", "never" or "inherit".
7322 Control the default behavior of the system with respect
7323 to anonymous transparent hugepages.
7324 Can be used multiple times for multiple anon THP sizes.
7325 See Documentation/admin-guide/mm/transhuge.rst for more
7326 details.
7328 threadirqs [KNL,EARLY]
7329 Force threading of all interrupt handlers except those
7330 marked explicitly IRQF_NO_THREAD.
7332 thp_shmem= [KNL]
7333 Format: <size>[KMG],<size>[KMG]:<policy>;<size>[KMG]-<size>[KMG]:<policy>
7334 Control the default policy of each hugepage size for the
7335 internal shmem mount. <policy> is one of policies available
7336 for the shmem mount ("always", "inherit", "never", "within_size",
7337 and "advise").
7338 It can be used multiple times for multiple shmem THP sizes.
7339 See Documentation/admin-guide/mm/transhuge.rst for more
7340 details.
7342 topology= [S390,EARLY]
7343 Format: {off | on}
7344 Specify if the kernel should make use of the cpu
7345 topology information if the hardware supports this.
7346 The scheduler will make use of this information and
7347 e.g. base its process migration decisions on it.
7348 Default is on.
7350 torture.disable_onoff_at_boot= [KNL]
7351 Prevent the CPU-hotplug component of torturing
7352 until after init has spawned.
7354 torture.ftrace_dump_at_shutdown= [KNL]
7355 Dump the ftrace buffer at torture-test shutdown,
7356 even if there were no errors. This can be a
7357 very costly operation when many torture tests
7358 are running concurrently, especially on systems
7359 with rotating-rust storage.
7361 torture.verbose_sleep_frequency= [KNL]
7362 Specifies how many verbose printk()s should be
7363 emitted between each sleep. The default of zero
7364 disables verbose-printk() sleeping.
7366 torture.verbose_sleep_duration= [KNL]
7367 Duration of each verbose-printk() sleep in jiffies.
7369 tpm.disable_pcr_integrity= [HW,TPM]
7370 Do not protect PCR registers from unintended physical
7371 access, or interposers in the bus by the means of
7372 having an integrity protected session wrapped around
7373 TPM2_PCR_Extend command. Consider this in a situation
7374 where TPM is heavily utilized by IMA, thus protection
7375 causing a major performance hit, and the space where
7376 machines are deployed is by other means guarded.
7378 tpm_crb_ffa.busy_timeout_ms= [ARM64,TPM]
7379 Maximum time in milliseconds to retry sending a message
7380 to the TPM service before giving up. This parameter controls
7381 how long the system will continue retrying when the TPM
7382 service is busy.
7383 Format: <unsigned int>
7384 Default: 2000 (2 seconds)
7386 tpm_suspend_pcr=[HW,TPM]
7387 Format: integer pcr id
7388 Specify that at suspend time, the tpm driver
7389 should extend the specified pcr with zeros,
7390 as a workaround for some chips which fail to
7391 flush the last written pcr on TPM_SaveState.
7392 This will guarantee that all the other pcrs
7393 are saved.
7395 tpm_tis.interrupts= [HW,TPM]
7396 Enable interrupts for the MMIO based physical layer
7397 for the FIFO interface. By default it is set to false
7398 (0). For more information about TPM hardware interfaces
7399 defined by Trusted Computing Group (TCG) see
7400 https://trustedcomputinggroup.org/resource/pc-client-platform-tpm-profile-ptp-specification/
7402 tp_printk [FTRACE]
7403 Have the tracepoints sent to printk as well as the
7404 tracing ring buffer. This is useful for early boot up
7405 where the system hangs or reboots and does not give the
7406 option for reading the tracing buffer or performing a
7407 ftrace_dump_on_oops.
7409 To turn off having tracepoints sent to printk,
7410 echo 0 > /proc/sys/kernel/tracepoint_printk
7411 Note, echoing 1 into this file without the
7412 tp_printk kernel cmdline option has no effect.
7414 The tp_printk_stop_on_boot (see below) can also be used
7415 to stop the printing of events to console at
7416 late_initcall_sync.
7418 ** CAUTION **
7420 Having tracepoints sent to printk() and activating high
7421 frequency tracepoints such as irq or sched, can cause
7422 the system to live lock.
7424 tp_printk_stop_on_boot [FTRACE]
7425 When tp_printk (above) is set, it can cause a lot of noise
7426 on the console. It may be useful to only include the
7427 printing of events during boot up, as user space may
7428 make the system inoperable.
7430 This command line option will stop the printing of events
7431 to console at the late_initcall_sync() time frame.
7433 trace_buf_size=nn[KMG]
7434 [FTRACE] will set tracing buffer size on each cpu.
7436 trace_clock= [FTRACE] Set the clock used for tracing events
7437 at boot up.
7438 local - Use the per CPU time stamp counter
7439 (converted into nanoseconds). Fast, but
7440 depending on the architecture, may not be
7441 in sync between CPUs.
7442 global - Event time stamps are synchronized across
7443 CPUs. May be slower than the local clock,
7444 but better for some race conditions.
7445 counter - Simple counting of events (1, 2, ..)
7446 note, some counts may be skipped due to the
7447 infrastructure grabbing the clock more than
7448 once per event.
7449 uptime - Use jiffies as the time stamp.
7450 perf - Use the same clock that perf uses.
7451 mono - Use ktime_get_mono_fast_ns() for time stamps.
7452 mono_raw - Use ktime_get_raw_fast_ns() for time
7453 stamps.
7454 boot - Use ktime_get_boot_fast_ns() for time stamps.
7455 Architectures may add more clocks. See
7456 Documentation/trace/ftrace.rst for more details.
7458 trace_event=[event-list]
7459 [FTRACE] Set and start specified trace events in order
7460 to facilitate early boot debugging. The event-list is a
7461 comma-separated list of trace events to enable. See
7462 also Documentation/trace/events.rst
7464 To enable modules, use :mod: keyword:
7466 trace_event=:mod:<module>
7468 The value before :mod: will only enable specific events
7469 that are part of the module. See the above mentioned
7470 document for more information.
7472 trace_instance=[instance-info]
7473 [FTRACE] Create a ring buffer instance early in boot up.
7474 This will be listed in:
7476 /sys/kernel/tracing/instances
7478 Events can be enabled at the time the instance is created
7479 via:
7481 trace_instance=<name>,<system1>:<event1>,<system2>:<event2>
7483 Note, the "<system*>:" portion is optional if the event is
7484 unique.
7486 trace_instance=foo,sched:sched_switch,irq_handler_entry,initcall
7488 will enable the "sched_switch" event (note, the "sched:" is optional, and
7489 the same thing would happen if it was left off). The irq_handler_entry
7490 event, and all events under the "initcall" system.
7492 Flags can be added to the instance to modify its behavior when it is
7493 created. The flags are separated by '^'.
7495 The available flags are:
7497 traceoff - Have the tracing instance tracing disabled after it is created.
7498 traceprintk - Have trace_printk() write into this trace instance
7499 (note, "printk" and "trace_printk" can also be used)
7501 trace_instance=foo^traceoff^traceprintk,sched,irq
7503 The flags must come before the defined events.
7505 If memory has been reserved (see memmap for x86), the instance
7506 can use that memory:
7508 memmap=12M$0x284500000 trace_instance=boot_map@0x284500000:12M
7510 The above will create a "boot_map" instance that uses the physical
7511 memory at 0x284500000 that is 12Megs. The per CPU buffers of that
7512 instance will be split up accordingly.
7514 Alternatively, the memory can be reserved by the reserve_mem option:
7516 reserve_mem=12M:4096:trace trace_instance=boot_map@trace
7518 This will reserve 12 megabytes at boot up with a 4096 byte alignment
7519 and place the ring buffer in this memory. Note that due to KASLR, the
7520 memory may not be the same location each time, which will not preserve
7521 the buffer content.
7523 Also note that the layout of the ring buffer data may change between
7524 kernel versions where the validator will fail and reset the ring buffer
7525 if the layout is not the same as the previous kernel.
7527 If the ring buffer is used for persistent bootups and has events enabled,
7528 it is recommend to disable tracing so that events from a previous boot do not
7529 mix with events of the current boot (unless you are debugging a random crash
7530 at boot up).
7532 reserve_mem=12M:4096:trace trace_instance=boot_map^traceoff^traceprintk@trace,sched,irq
7534 Note, saving the trace buffer across reboots does require that the system
7535 is set up to not wipe memory. For instance, CONFIG_RESET_ATTACK_MITIGATION
7536 can force a memory reset on boot which will clear any trace that was stored.
7537 This is just one of many ways that can clear memory. Make sure your system
7538 keeps the content of memory across reboots before relying on this option.
7540 NB: Both the mapped address and size must be page aligned for the architecture.
7542 See also Documentation/trace/debugging.rst
7545 trace_options=[option-list]
7546 [FTRACE] Enable or disable tracer options at boot.
7547 The option-list is a comma delimited list of options
7548 that can be enabled or disabled just as if you were
7549 to echo the option name into
7551 /sys/kernel/tracing/trace_options
7553 For example, to enable stacktrace option (to dump the
7554 stack trace of each event), add to the command line:
7556 trace_options=stacktrace
7558 See also Documentation/trace/ftrace.rst "trace options"
7559 section.
7561 trace_trigger=[trigger-list]
7562 [FTRACE] Add an event trigger on specific events.
7563 Set a trigger on top of a specific event, with an optional
7564 filter.
7566 The format is "trace_trigger=<event>.<trigger>[ if <filter>],..."
7567 Where more than one trigger may be specified that are comma delimited.
7569 For example:
7571 trace_trigger="sched_switch.stacktrace if prev_state == 2"
7573 The above will enable the "stacktrace" trigger on the "sched_switch"
7574 event but only trigger it if the "prev_state" of the "sched_switch"
7575 event is "2" (TASK_UNINTERRUPTIBLE).
7577 See also "Event triggers" in Documentation/trace/events.rst
7580 traceoff_after_boot
7581 [FTRACE] Sometimes tracing is used to debug issues
7582 during the boot process. Since the trace buffer has a
7583 limited amount of storage, it may be prudent to
7584 disable tracing after the boot is finished, otherwise
7585 the critical information may be overwritten. With this
7586 option, the main tracing buffer will be turned off at
7587 the end of the boot process.
7589 traceoff_on_warning
7590 [FTRACE] enable this option to disable tracing when a
7591 warning is hit. This turns off "tracing_on". Tracing can
7592 be enabled again by echoing '1' into the "tracing_on"
7593 file located in /sys/kernel/tracing/
7595 This option is useful, as it disables the trace before
7596 the WARNING dump is called, which prevents the trace to
7597 be filled with content caused by the warning output.
7599 This option can also be set at run time via the sysctl
7600 option: kernel/traceoff_on_warning
7602 transparent_hugepage=
7603 [KNL]
7604 Format: [always|madvise|never]
7605 Can be used to control the default behavior of the system
7606 with respect to transparent hugepages.
7607 See Documentation/admin-guide/mm/transhuge.rst
7608 for more details.
7610 transparent_hugepage_shmem= [KNL]
7611 Format: [always|within_size|advise|never|deny|force]
7612 Can be used to control the hugepage allocation policy for
7613 the internal shmem mount.
7614 See Documentation/admin-guide/mm/transhuge.rst
7615 for more details.
7617 transparent_hugepage_tmpfs= [KNL]
7618 Format: [always|within_size|advise|never]
7619 Can be used to control the default hugepage allocation policy
7620 for the tmpfs mount.
7621 See Documentation/admin-guide/mm/transhuge.rst
7622 for more details.
7624 trusted.source= [KEYS]
7625 Format: <string>
7626 This parameter identifies the trust source as a backend
7627 for trusted keys implementation. Supported trust
7628 sources:
7629 - "tpm"
7630 - "tee"
7631 - "caam"
7632 - "dcp"
7633 If not specified then it defaults to iterating through
7634 the trust source list starting with TPM and assigns the
7635 first trust source as a backend which is initialized
7636 successfully during iteration.
7638 trusted.rng= [KEYS]
7639 Format: <string>
7640 The RNG used to generate key material for trusted keys.
7641 Can be one of:
7642 - "kernel"
7643 - the same value as trusted.source: "tpm" or "tee"
7644 - "default"
7645 If not specified, "default" is used. In this case,
7646 the RNG's choice is left to each individual trust source.
7648 trusted.dcp_use_otp_key
7649 This is intended to be used in combination with
7650 trusted.source=dcp and will select the DCP OTP key
7651 instead of the DCP UNIQUE key blob encryption.
7653 trusted.dcp_skip_zk_test
7654 This is intended to be used in combination with
7655 trusted.source=dcp and will disable the check if the
7656 blob key is all zeros. This is helpful for situations where
7657 having this key zero'ed is acceptable. E.g. in testing
7658 scenarios.
7660 tsa= [X86] Control mitigation for Transient Scheduler
7661 Attacks on AMD CPUs. Search the following in your
7662 favourite search engine for more details:
7664 "Technical guidance for mitigating transient scheduler
7665 attacks".
7667 off - disable the mitigation
7668 on - enable the mitigation (default)
7669 user - mitigate only user/kernel transitions
7670 vm - mitigate only guest/host transitions
7673 tsc= Disable clocksource stability checks for TSC.
7674 Format: <string>
7675 [x86] reliable: mark tsc clocksource as reliable, this
7676 disables clocksource verification at runtime, as well
7677 as the stability checks done at bootup. Used to enable
7678 high-resolution timer mode on older hardware, and in
7679 virtualized environment.
7680 [x86] noirqtime: Do not use TSC to do irq accounting.
7681 Used to run time disable IRQ_TIME_ACCOUNTING on any
7682 platforms where RDTSC is slow and this accounting
7683 can add overhead.
7684 [x86] unstable: mark the TSC clocksource as unstable, this
7685 marks the TSC unconditionally unstable at bootup and
7686 avoids any further wobbles once the TSC watchdog notices.
7687 [x86] nowatchdog: disable clocksource watchdog. Used
7688 in situations with strict latency requirements (where
7689 interruptions from clocksource watchdog are not
7690 acceptable).
7691 [x86] recalibrate: force recalibration against a HW timer
7692 (HPET or PM timer) on systems whose TSC frequency was
7693 obtained from HW or FW using either an MSR or CPUID(0x15).
7694 Warn if the difference is more than 500 ppm.
7695 [x86] watchdog: Use TSC as the watchdog clocksource with
7696 which to check other HW timers (HPET or PM timer), but
7697 only on systems where TSC has been deemed trustworthy.
7698 This will be suppressed by an earlier tsc=nowatchdog and
7699 can be overridden by a later tsc=nowatchdog. A console
7700 message will flag any such suppression or overriding.
7702 tsc_early_khz= [X86,EARLY] Skip early TSC calibration and use the given
7703 value instead. Useful when the early TSC frequency discovery
7704 procedure is not reliable, such as on overclocked systems
7705 with CPUID.16h support and partial CPUID.15h support.
7706 Format: <unsigned int>
7708 tsx= [X86] Control Transactional Synchronization
7709 Extensions (TSX) feature in Intel processors that
7710 support TSX control.
7712 This parameter controls the TSX feature. The options are:
7714 on - Enable TSX on the system. Although there are
7715 mitigations for all known security vulnerabilities,
7716 TSX has been known to be an accelerator for
7717 several previous speculation-related CVEs, and
7718 so there may be unknown security risks associated
7719 with leaving it enabled.
7721 off - Disable TSX on the system. (Note that this
7722 option takes effect only on newer CPUs which are
7723 not vulnerable to MDS, i.e., have
7724 MSR_IA32_ARCH_CAPABILITIES.MDS_NO=1 and which get
7725 the new IA32_TSX_CTRL MSR through a microcode
7726 update. This new MSR allows for the reliable
7727 deactivation of the TSX functionality.)
7729 auto - Disable TSX if X86_BUG_TAA is present,
7730 otherwise enable TSX on the system.
7732 Not specifying this option is equivalent to tsx=off.
7734 See Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
7735 for more details.
7737 tsx_async_abort= [X86,INTEL,EARLY] Control mitigation for the TSX Async
7738 Abort (TAA) vulnerability.
7740 Similar to Micro-architectural Data Sampling (MDS)
7741 certain CPUs that support Transactional
7742 Synchronization Extensions (TSX) are vulnerable to an
7743 exploit against CPU internal buffers which can forward
7744 information to a disclosure gadget under certain
7745 conditions.
7747 In vulnerable processors, the speculatively forwarded
7748 data can be used in a cache side channel attack, to
7749 access data to which the attacker does not have direct
7750 access.
7752 This parameter controls the TAA mitigation. The
7753 options are:
7755 full - Enable TAA mitigation on vulnerable CPUs
7756 if TSX is enabled.
7758 full,nosmt - Enable TAA mitigation and disable SMT on
7759 vulnerable CPUs. If TSX is disabled, SMT
7760 is not disabled because CPU is not
7761 vulnerable to cross-thread TAA attacks.
7762 off - Unconditionally disable TAA mitigation
7764 On MDS-affected machines, tsx_async_abort=off can be
7765 prevented by an active MDS mitigation as both vulnerabilities
7766 are mitigated with the same mechanism so in order to disable
7767 this mitigation, you need to specify mds=off too.
7769 Not specifying this option is equivalent to
7770 tsx_async_abort=full. On CPUs which are MDS affected
7771 and deploy MDS mitigation, TAA mitigation is not
7772 required and doesn't provide any additional
7773 mitigation.
7775 For details see:
7776 Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
7778 turbografx.map[2|3]= [HW,JOY]
7779 TurboGraFX parallel port interface
7780 Format:
7781 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7>
7782 See also Documentation/input/devices/joystick-parport.rst
7784 udbg-immortal [PPC] When debugging early kernel crashes that
7785 happen after console_init() and before a proper
7786 console driver takes over, this boot options might
7787 help "seeing" what's going on.
7789 uhash_entries= [KNL,NET]
7790 Set number of hash buckets for UDP/UDP-Lite connections
7792 uhci-hcd.ignore_oc=
7793 [USB] Ignore overcurrent events (default N).
7794 Some badly-designed motherboards generate lots of
7795 bogus events, for ports that aren't wired to
7796 anything. Set this parameter to avoid log spamming.
7797 Note that genuine overcurrent events won't be
7798 reported either.
7800 unaligned_scalar_speed=
7801 [RISCV]
7802 Format: {slow | fast | unsupported}
7803 Allow skipping scalar unaligned access speed tests. This
7804 is useful for testing alternative code paths and to skip
7805 the tests in environments where they run too slowly. All
7806 CPUs must have the same scalar unaligned access speed.
7808 unaligned_vector_speed=
7809 [RISCV]
7810 Format: {slow | fast | unsupported}
7811 Allow skipping vector unaligned access speed tests. This
7812 is useful for testing alternative code paths and to skip
7813 the tests in environments where they run too slowly. All
7814 CPUs must have the same vector unaligned access speed.
7816 unknown_nmi_panic
7817 [X86] Cause panic on unknown NMI.
7819 unwind_debug [X86-64,EARLY]
7820 Enable unwinder debug output. This can be
7821 useful for debugging certain unwinder error
7822 conditions, including corrupt stacks and
7823 bad/missing unwinder metadata.
7825 usbcore.authorized_default=
7826 [USB] Default USB device authorization:
7827 (default -1 = authorized (same as 1),
7828 0 = not authorized, 1 = authorized, 2 = authorized
7829 if device connected to internal port)
7831 usbcore.autosuspend=
7832 [USB] The autosuspend time delay (in seconds) used
7833 for newly-detected USB devices (default 2). This
7834 is the time required before an idle device will be
7835 autosuspended. Devices for which the delay is set
7836 to a negative value won't be autosuspended at all.
7838 usbcore.usbfs_snoop=
7839 [USB] Set to log all usbfs traffic (default 0 = off).
7841 usbcore.usbfs_snoop_max=
7842 [USB] Maximum number of bytes to snoop in each URB
7843 (default = 65536).
7845 usbcore.blinkenlights=
7846 [USB] Set to cycle leds on hubs (default 0 = off).
7848 usbcore.old_scheme_first=
7849 [USB] Start with the old device initialization
7850 scheme (default 0 = off).
7852 usbcore.usbfs_memory_mb=
7853 [USB] Memory limit (in MB) for buffers allocated by
7854 usbfs (default = 16, 0 = max = 2047).
7856 usbcore.use_both_schemes=
7857 [USB] Try the other device initialization scheme
7858 if the first one fails (default 1 = enabled).
7860 usbcore.initial_descriptor_timeout=
7861 [USB] Specifies timeout for the initial 64-byte
7862 USB_REQ_GET_DESCRIPTOR request in milliseconds
7863 (default 5000 = 5.0 seconds).
7865 usbcore.nousb [USB] Disable the USB subsystem
7867 usbcore.quirks=
7868 [USB] A list of quirk entries to augment the built-in
7869 usb core quirk list. List entries are separated by
7870 commas. Each entry has the form
7871 VendorID:ProductID:Flags. The IDs are 4-digit hex
7872 numbers and Flags is a set of letters. Each letter
7873 will change the built-in quirk; setting it if it is
7874 clear and clearing it if it is set. The letters have
7875 the following meanings:
7876 a = USB_QUIRK_STRING_FETCH_255 (string
7877 descriptors must not be fetched using
7878 a 255-byte read);
7879 b = USB_QUIRK_RESET_RESUME (device can't resume
7880 correctly so reset it instead);
7881 c = USB_QUIRK_NO_SET_INTF (device can't handle
7882 Set-Interface requests);
7883 d = USB_QUIRK_CONFIG_INTF_STRINGS (device can't
7884 handle its Configuration or Interface
7885 strings);
7886 e = USB_QUIRK_RESET (device can't be reset
7887 (e.g morph devices), don't use reset);
7888 f = USB_QUIRK_HONOR_BNUMINTERFACES (device has
7889 more interface descriptions than the
7890 bNumInterfaces count, and can't handle
7891 talking to these interfaces);
7892 g = USB_QUIRK_DELAY_INIT (device needs a pause
7893 during initialization, after we read
7894 the device descriptor);
7895 h = USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL (For
7896 high speed and super speed interrupt
7897 endpoints, the USB 2.0 and USB 3.0 spec
7898 require the interval in microframes (1
7899 microframe = 125 microseconds) to be
7900 calculated as interval = 2 ^
7901 (bInterval-1).
7902 Devices with this quirk report their
7903 bInterval as the result of this
7904 calculation instead of the exponent
7905 variable used in the calculation);
7906 i = USB_QUIRK_DEVICE_QUALIFIER (device can't
7907 handle device_qualifier descriptor
7908 requests);
7909 j = USB_QUIRK_IGNORE_REMOTE_WAKEUP (device
7910 generates spurious wakeup, ignore
7911 remote wakeup capability);
7912 k = USB_QUIRK_NO_LPM (device can't handle Link
7913 Power Management);
7914 l = USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL
7915 (Device reports its bInterval as linear
7916 frames instead of the USB 2.0
7917 calculation);
7918 m = USB_QUIRK_DISCONNECT_SUSPEND (Device needs
7919 to be disconnected before suspend to
7920 prevent spurious wakeup);
7921 n = USB_QUIRK_DELAY_CTRL_MSG (Device needs a
7922 pause after every control message);
7923 o = USB_QUIRK_HUB_SLOW_RESET (Hub needs extra
7924 delay after resetting its port);
7925 p = USB_QUIRK_SHORT_SET_ADDRESS_REQ_TIMEOUT
7926 (Reduce timeout of the SET_ADDRESS
7927 request from 5000 ms to 500 ms);
7928 q = USB_QUIRK_FORCE_ONE_CONFIG (Device
7929 claims zero configurations,
7930 forcing to 1);
7931 Example: quirks=0781:5580:bk,0a5c:5834:gij
7933 usbhid.mousepoll=
7934 [USBHID] The interval which mice are to be polled at.
7936 usbhid.jspoll=
7937 [USBHID] The interval which joysticks are to be polled at.
7939 usbhid.kbpoll=
7940 [USBHID] The interval which keyboards are to be polled at.
7942 usb-storage.delay_use=
7943 [UMS] The delay in seconds before a new device is
7944 scanned for Logical Units (default 1).
7945 Optionally the delay in milliseconds if the value has
7946 suffix with "ms".
7947 Example: delay_use=2567ms
7949 usb-storage.quirks=
7950 [UMS] A list of quirks entries to supplement or
7951 override the built-in unusual_devs list. List
7952 entries are separated by commas. Each entry has
7953 the form VID:PID:Flags where VID and PID are Vendor
7954 and Product ID values (4-digit hex numbers) and
7955 Flags is a set of characters, each corresponding
7956 to a common usb-storage quirk flag as follows:
7957 a = SANE_SENSE (collect more than 18 bytes
7958 of sense data, not on uas);
7959 b = BAD_SENSE (don't collect more than 18
7960 bytes of sense data, not on uas);
7961 c = FIX_CAPACITY (decrease the reported
7962 device capacity by one sector);
7963 d = NO_READ_DISC_INFO (don't use
7964 READ_DISC_INFO command, not on uas);
7965 e = NO_READ_CAPACITY_16 (don't use
7966 READ_CAPACITY_16 command);
7967 f = NO_REPORT_OPCODES (don't use report opcodes
7968 command, uas only);
7969 g = MAX_SECTORS_240 (don't transfer more than
7970 240 sectors at a time, uas only);
7971 h = CAPACITY_HEURISTICS (decrease the
7972 reported device capacity by one
7973 sector if the number is odd);
7974 i = IGNORE_DEVICE (don't bind to this
7975 device);
7976 j = NO_REPORT_LUNS (don't use report luns
7977 command, uas only);
7978 k = NO_SAME (do not use WRITE_SAME, uas only)
7979 l = NOT_LOCKABLE (don't try to lock and
7980 unlock ejectable media, not on uas);
7981 m = MAX_SECTORS_64 (don't transfer more
7982 than 64 sectors = 32 KB at a time,
7983 not on uas);
7984 n = INITIAL_READ10 (force a retry of the
7985 initial READ(10) command, not on uas);
7986 o = CAPACITY_OK (accept the capacity
7987 reported by the device, not on uas);
7988 p = WRITE_CACHE (the device cache is ON
7989 by default, not on uas);
7990 r = IGNORE_RESIDUE (the device reports
7991 bogus residue values, not on uas);
7992 s = SINGLE_LUN (the device has only one
7993 Logical Unit);
7994 t = NO_ATA_1X (don't allow ATA(12) and ATA(16)
7995 commands, uas only);
7996 u = IGNORE_UAS (don't bind to the uas driver);
7997 w = NO_WP_DETECT (don't test whether the
7998 medium is write-protected).
7999 y = ALWAYS_SYNC (issue a SYNCHRONIZE_CACHE
8000 even if the device claims no cache,
8001 not on uas)
8002 Example: quirks=0419:aaf5:rl,0421:0433:rc
8004 user_debug= [KNL,ARM]
8005 Format: <int>
8006 See arch/arm/Kconfig.debug help text.
8007 1 - undefined instruction events
8008 2 - system calls
8009 4 - invalid data aborts
8010 8 - SIGSEGV faults
8011 16 - SIGBUS faults
8012 Example: user_debug=31
8014 vdso= [X86,SH,SPARC]
8015 On X86_32, this is an alias for vdso32=. Otherwise:
8017 vdso=1: enable VDSO (the default)
8018 vdso=0: disable VDSO mapping
8020 vdso32= [X86] Control the 32-bit vDSO
8021 vdso32=1: enable 32-bit VDSO
8022 vdso32=0 or vdso32=2: disable 32-bit VDSO
8024 See the help text for CONFIG_COMPAT_VDSO for more
8025 details. If CONFIG_COMPAT_VDSO is set, the default is
8026 vdso32=0; otherwise, the default is vdso32=1.
8028 For compatibility with older kernels, vdso32=2 is an
8029 alias for vdso32=0.
8031 Try vdso32=0 if you encounter an error that says:
8032 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed!
8034 video= [FB,EARLY] Frame buffer configuration
8035 See Documentation/fb/modedb.rst.
8037 video.brightness_switch_enabled= [ACPI]
8038 Format: [0|1]
8039 If set to 1, on receiving an ACPI notify event
8040 generated by hotkey, video driver will adjust brightness
8041 level and then send out the event to user space through
8042 the allocated input device. If set to 0, video driver
8043 will only send out the event without touching backlight
8044 brightness level.
8045 default: 1
8047 virtio_mmio.device=
8048 [VMMIO] Memory mapped virtio (platform) device.
8050 <size>@<baseaddr>:<irq>[:<id>]
8051 where:
8052 <size> := size (can use standard suffixes
8053 like K, M and G)
8054 <baseaddr> := physical base address
8055 <irq> := interrupt number (as passed to
8056 request_irq())
8057 <id> := (optional) platform device id
8058 example:
8059 virtio_mmio.device=1K@0x100b0000:48:7
8061 Can be used multiple times for multiple devices.
8063 vga= [BOOT,X86-32] Select a particular video mode
8064 See Documentation/arch/x86/boot.rst and
8065 Documentation/admin-guide/svga.rst.
8066 Use vga=ask for menu.
8067 This is actually a boot loader parameter; the value is
8068 passed to the kernel using a special protocol.
8070 vm_debug[=options] [KNL] Available with CONFIG_DEBUG_VM=y.
8071 May slow down system boot speed, especially when
8072 enabled on systems with a large amount of memory.
8073 All options are enabled by default, and this
8074 interface is meant to allow for selectively
8075 enabling or disabling specific virtual memory
8076 debugging features.
8078 Available options are:
8079 P Enable page structure init time poisoning
8080 - Disable all of the above options
8082 vmalloc=nn[KMG] [KNL,BOOT,EARLY] Forces the vmalloc area to have an
8083 exact size of <nn>. This can be used to increase
8084 the minimum size (128MB on x86, arm32 platforms).
8085 It can also be used to decrease the size and leave more room
8086 for directly mapped kernel RAM. Note that this parameter does
8087 not exist on many other platforms (including arm64, alpha,
8088 loongarch, arc, csky, hexagon, microblaze, mips, nios2, openrisc,
8089 parisc, m64k, powerpc, riscv, sh, um, xtensa, s390, sparc).
8091 vmcp_cma=nn[MG] [KNL,S390,EARLY]
8092 Sets the memory size reserved for contiguous memory
8093 allocations for the vmcp device driver.
8095 vmhalt= [KNL,S390] Perform z/VM CP command after system halt.
8096 Format: <command>
8098 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic.
8099 Format: <command>
8101 vmpoff= [KNL,S390] Perform z/VM CP command after power off.
8102 Format: <command>
8104 vmscape= [X86] Controls mitigation for VMscape attacks.
8105 VMscape attacks can leak information from a userspace
8106 hypervisor to a guest via speculative side-channels.
8108 off - disable the mitigation
8109 ibpb - use Indirect Branch Prediction Barrier
8110 (IBPB) mitigation (default)
8111 force - force vulnerability detection even on
8112 unaffected processors
8114 vsyscall= [X86-64,EARLY]
8115 Controls the behavior of vsyscalls (i.e. calls to
8116 fixed addresses of 0xffffffffff600x00 from legacy
8117 code). Most statically-linked binaries and older
8118 versions of glibc use these calls. Because these
8119 functions are at fixed addresses, they make nice
8120 targets for exploits that can control RIP.
8122 emulate Vsyscalls turn into traps and are emulated
8123 reasonably safely. The vsyscall page is
8124 readable.
8126 xonly [default] Vsyscalls turn into traps and are
8127 emulated reasonably safely. The vsyscall
8128 page is not readable.
8130 none Vsyscalls don't work at all. This makes
8131 them quite hard to use for exploits but
8132 might break your system.
8134 vt.color= [VT] Default text color.
8135 Format: 0xYX, X = foreground, Y = background.
8136 Default: 0x07 = light gray on black.
8138 vt.cur_default= [VT] Default cursor shape.
8139 Format: 0xCCBBAA, where AA, BB, and CC are the same as
8140 the parameters of the <Esc>[?A;B;Cc escape sequence;
8141 see vga-softcursor.rst. Default: 2 = underline.
8143 vt.default_blu= [VT]
8144 Format: <blue0>,<blue1>,<blue2>,...,<blue15>
8145 Change the default blue palette of the console.
8146 This is a 16-member array composed of values
8147 ranging from 0-255.
8149 vt.default_grn= [VT]
8150 Format: <green0>,<green1>,<green2>,...,<green15>
8151 Change the default green palette of the console.
8152 This is a 16-member array composed of values
8153 ranging from 0-255.
8155 vt.default_red= [VT]
8156 Format: <red0>,<red1>,<red2>,...,<red15>
8157 Change the default red palette of the console.
8158 This is a 16-member array composed of values
8159 ranging from 0-255.
8161 vt.default_utf8=
8162 [VT]
8163 Format=<0|1>
8164 Set system-wide default UTF-8 mode for all tty's.
8165 Default is 1, i.e. UTF-8 mode is enabled for all
8166 newly opened terminals.
8168 vt.global_cursor_default=
8169 [VT]
8170 Format=<-1|0|1>
8171 Set system-wide default for whether a cursor
8172 is shown on new VTs. Default is -1,
8173 i.e. cursors will be created by default unless
8174 overridden by individual drivers. 0 will hide
8175 cursors, 1 will display them.
8177 vt.italic= [VT] Default color for italic text; 0-15.
8178 Default: 2 = green.
8180 vt.underline= [VT] Default color for underlined text; 0-15.
8181 Default: 3 = cyan.
8183 watchdog timers [HW,WDT] For information on watchdog timers,
8184 see Documentation/watchdog/watchdog-parameters.rst
8185 or other driver-specific files in the
8186 Documentation/watchdog/ directory.
8188 watchdog_thresh=
8189 [KNL]
8190 Set the hard lockup detector stall duration
8191 threshold in seconds. The soft lockup detector
8192 threshold is set to twice the value. A value of 0
8193 disables both lockup detectors. Default is 10
8194 seconds.
8196 workqueue.unbound_cpus=
8197 [KNL,SMP] Specify to constrain one or some CPUs
8198 to use in unbound workqueues.
8199 Format: <cpu-list>
8200 By default, all online CPUs are available for
8201 unbound workqueues.
8203 workqueue.watchdog_thresh=
8204 If CONFIG_WQ_WATCHDOG is configured, workqueue can
8205 warn stall conditions and dump internal state to
8206 help debugging. 0 disables workqueue stall
8207 detection; otherwise, it's the stall threshold
8208 duration in seconds. The default value is 30 and
8209 it can be updated at runtime by writing to the
8210 corresponding sysfs file.
8212 workqueue.panic_on_stall=<uint>
8213 Panic when workqueue stall is detected by
8214 CONFIG_WQ_WATCHDOG. It sets the number times of the
8215 stall to trigger panic.
8217 The default is 0, which disables the panic on stall.
8219 workqueue.cpu_intensive_thresh_us=
8220 Per-cpu work items which run for longer than this
8221 threshold are automatically considered CPU intensive
8222 and excluded from concurrency management to prevent
8223 them from noticeably delaying other per-cpu work
8224 items. Default is 10000 (10ms).
8226 If CONFIG_WQ_CPU_INTENSIVE_REPORT is set, the kernel
8227 will report the work functions which violate this
8228 threshold repeatedly. They are likely good
8229 candidates for using WQ_UNBOUND workqueues instead.
8231 workqueue.cpu_intensive_warning_thresh=<uint>
8232 If CONFIG_WQ_CPU_INTENSIVE_REPORT is set, the kernel
8233 will report the work functions which violate the
8234 intensive_threshold_us repeatedly. In order to prevent
8235 spurious warnings, start printing only after a work
8236 function has violated this threshold number of times.
8238 The default is 4 times. 0 disables the warning.
8240 workqueue.power_efficient
8241 Per-cpu workqueues are generally preferred because
8242 they show better performance thanks to cache
8243 locality; unfortunately, per-cpu workqueues tend to
8244 be more power hungry than unbound workqueues.
8246 Enabling this makes the per-cpu workqueues which
8247 were observed to contribute significantly to power
8248 consumption unbound, leading to measurably lower
8249 power usage at the cost of small performance
8250 overhead.
8252 The default value of this parameter is determined by
8253 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT.
8255 workqueue.default_affinity_scope=
8256 Select the default affinity scope to use for unbound
8257 workqueues. Can be one of "cpu", "smt", "cache",
8258 "numa" and "system". Default is "cache". For more
8259 information, see the Affinity Scopes section in
8260 Documentation/core-api/workqueue.rst.
8262 This can be changed after boot by writing to the
8263 matching /sys/module/workqueue/parameters file. All
8264 workqueues with the "default" affinity scope will be
8265 updated accordingly.
8267 workqueue.debug_force_rr_cpu
8268 Workqueue used to implicitly guarantee that work
8269 items queued without explicit CPU specified are put
8270 on the local CPU. This guarantee is no longer true
8271 and while local CPU is still preferred work items
8272 may be put on foreign CPUs. This debug option
8273 forces round-robin CPU selection to flush out
8274 usages which depend on the now broken guarantee.
8275 When enabled, memory and cache locality will be
8276 impacted.
8278 writecombine= [LOONGARCH,EARLY] Control the MAT (Memory Access
8279 Type) of ioremap_wc().
8281 on - Enable writecombine, use WUC for ioremap_wc()
8282 off - Disable writecombine, use SUC for ioremap_wc()
8284 x2apic_phys [X86-64,APIC,EARLY] Use x2apic physical mode instead of
8285 default x2apic cluster mode on platforms
8286 supporting x2apic.
8288 xen_512gb_limit [KNL,X86-64,XEN]
8289 Restricts the kernel running paravirtualized under Xen
8290 to use only up to 512 GB of RAM. The reason to do so is
8291 crash analysis tools and Xen tools for doing domain
8292 save/restore/migration must be enabled to handle larger
8293 domains.
8295 xen_emul_unplug= [HW,X86,XEN,EARLY]
8296 Unplug Xen emulated devices
8297 Format: [unplug0,][unplug1]
8298 ide-disks -- unplug primary master IDE devices
8299 aux-ide-disks -- unplug non-primary-master IDE devices
8300 nics -- unplug network devices
8301 all -- unplug all emulated devices (NICs and IDE disks)
8302 unnecessary -- unplugging emulated devices is
8303 unnecessary even if the host did not respond to
8304 the unplug protocol
8305 never -- do not unplug even if version check succeeds
8307 xen_legacy_crash [X86,XEN,EARLY]
8308 Crash from Xen panic notifier, without executing late
8309 panic() code such as dumping handler.
8311 xen_mc_debug [X86,XEN,EARLY]
8312 Enable multicall debugging when running as a Xen PV guest.
8313 Enabling this feature will reduce performance a little
8314 bit, so it should only be enabled for obtaining extended
8315 debug data in case of multicall errors.
8317 xen_msr_safe= [X86,XEN,EARLY]
8318 Format: <bool>
8319 Select whether to always use non-faulting (safe) MSR
8320 access functions when running as Xen PV guest. The
8321 default value is controlled by CONFIG_XEN_PV_MSR_SAFE.
8323 xen_nopv [X86]
8324 Disables the PV optimizations forcing the HVM guest to
8325 run as generic HVM guest with no PV drivers.
8326 This option is obsoleted by the "nopv" option, which
8327 has equivalent effect for XEN platform.
8329 xen_no_vector_callback
8330 [KNL,X86,XEN,EARLY] Disable the vector callback for Xen
8331 event channel interrupts.
8333 xen_scrub_pages= [XEN]
8334 Boolean option to control scrubbing pages before giving them back
8335 to Xen, for use by other domains. Can be also changed at runtime
8336 with /sys/devices/system/xen_memory/xen_memory0/scrub_pages.
8337 Default value controlled with CONFIG_XEN_SCRUB_PAGES_DEFAULT.
8339 xen_timer_slop= [X86-64,XEN,EARLY]
8340 Set the timer slop (in nanoseconds) for the virtual Xen
8341 timers (default is 100000). This adjusts the minimum
8342 delta of virtualized Xen timers, where lower values
8343 improve timer resolution at the expense of processing
8344 more timer interrupts.
8346 xen.balloon_boot_timeout= [XEN]
8347 The time (in seconds) to wait before giving up to boot
8348 in case initial ballooning fails to free enough memory.
8349 Applies only when running as HVM or PVH guest and
8350 started with less memory configured than allowed at
8351 max. Default is 180.
8353 xen.event_eoi_delay= [XEN]
8354 How long to delay EOI handling in case of event
8355 storms (jiffies). Default is 10.
8357 xen.event_loop_timeout= [XEN]
8358 After which time (jiffies) the event handling loop
8359 should start to delay EOI handling. Default is 2.
8361 xen.fifo_events= [XEN]
8362 Boolean parameter to disable using fifo event handling
8363 even if available. Normally fifo event handling is
8364 preferred over the 2-level event handling, as it is
8365 fairer and the number of possible event channels is
8366 much higher. Default is on (use fifo events).
8368 xirc2ps_cs= [NET,PCMCIA]
8369 Format:
8370 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]]
8372 xive= [PPC]
8373 By default on POWER9 and above, the kernel will
8374 natively use the XIVE interrupt controller. This option
8375 allows the fallback firmware mode to be used:
8377 off Fallback to firmware control of XIVE interrupt
8378 controller on both pseries and powernv
8379 platforms. Only useful on POWER9 and above.
8381 xive.store-eoi=off [PPC]
8382 By default on POWER10 and above, the kernel will use
8383 stores for EOI handling when the XIVE interrupt mode
8384 is active. This option allows the XIVE driver to use
8385 loads instead, as on POWER9.
8387 xhci-hcd.quirks [USB,KNL]
8388 A hex value specifying bitmask with supplemental xhci
8389 host controller quirks. Meaning of each bit can be
8390 consulted in header drivers/usb/host/xhci.h.
8392 xmon [PPC,EARLY]
8393 Format: { early | on | rw | ro | off }
8394 Controls if xmon debugger is enabled. Default is off.
8395 Passing only "xmon" is equivalent to "xmon=early".
8396 early Call xmon as early as possible on boot; xmon
8397 debugger is called from setup_arch().
8398 on xmon debugger hooks will be installed so xmon
8399 is only called on a kernel crash. Default mode,
8400 i.e. either "ro" or "rw" mode, is controlled
8401 with CONFIG_XMON_DEFAULT_RO_MODE.
8402 rw xmon debugger hooks will be installed so xmon
8403 is called only on a kernel crash, mode is write,
8404 meaning SPR registers, memory and, other data
8405 can be written using xmon commands.
8406 ro same as "rw" option above but SPR registers,
8407 memory, and other data can't be written using
8408 xmon commands.
8409 off xmon is disabled.

3. 한국어 전문 번역

영어 원문의 문단 순서와 의미를 유지한 전체 번역입니다. 코드, 함수명, symbol과 URL은 원문 표기를 유지합니다.

메모리 수락과 ACPI 기본 동작

1-36
매개변수적용·형식한국어 설명
accept_memory=[MM] `{ eager | lazy }`, 기본 `lazy`기본값은 부팅 지연을 줄이기 위해 unaccepted memory를 지연 수락합니다. 모든 메모리가 수락될 때까지 약간의 런타임 오버헤드가 생기지만 대부분 무시할 만합니다. 특정 워크로드나 디버깅에서는 `accept_memory=eager`로 부팅 중 모든 메모리를 한 번에 수락할 수 있습니다.
acpi=[HW,ACPI,X86,ARM64,RISCV64,EARLY] `{ force | on | off | strict | noirq | rsdt | copy_dsdt | nospcr }``force`는 기본이 off인 ACPI를 강제로 켭니다. `on`은 ACPI를 켜되 arm64·riscv64에서 DT fallback을 허용합니다. `off`는 ACPI를 끄고, `noirq`는 IRQ routing에 ACPI를 쓰지 않으며, `strict`는 ACPI 규격을 엄격히 지키지 않는 플랫폼을 덜 관대하게 처리합니다. `rsdt`는 기본 XSDT보다 RSDT를 우선하고 `copy_dsdt`는 DSDT를 메모리에 복사합니다. `nocmcff`는 corrected error의 firmware-first 모드를 꺼 중복 보고가 생길 수 있습니다. `nospcr`은 ARM64에서 ACPI SPCR 콘솔을 기본 serial 콘솔로 쓰지 않습니다. ARM64는 `off/on/force/nospcr`, RISCV64는 `off/on/force`만 지원합니다. `Documentation/power/runtime_pm.rst`와 `pci=noacpi`도 참조하십시오.

ACPI 테이블과 AML 반환값

37-64
매개변수적용·형식한국어 설명
acpi_apic_instance=[ACPI,IOAPIC,EARLY] `<int>`, 기본 `0``2`는 가능하면 두 번째 APIC 테이블을 사용하고 `1` 또는 `0`은 첫 번째 APIC 테이블을 사용합니다.
acpi_backlight=[HW,ACPI] `{ vendor | video | native | none }``vendor`는 `video.ko` 대신 `thinkpad_acpi`, `sony_acpi` 같은 제조사 드라이버를 우선합니다. `video`는 ACPI `video.ko`, `native`는 장치의 native backlight 모드를 사용하고 `none`은 ACPI backlight 인터페이스를 끕니다.
acpi_force_32bit_fadt_addr[ACPI,EARLY]FADT가 64비트 `X_*` 주소 대신 32비트 주소를 쓰게 합니다. 손상된 64비트 주소를 제공하는 펌웨어에서 이를 무시하고 구형 32비트 주소를 사용합니다.
acpica_no_return_repair[HW,ACPI]AML predefined validation의 반환값 복구 기능을 끕니다. 이 기능은 평가 결과를 ACPI 규격에 더 맞게 고칠 수 있습니다. 복구 기능과 관련된 AML interpreter 문제의 근본 원인을 찾는 개발자에게 유용합니다.

ACPI 디버그와 리소스 충돌

65-107

`acpi.debug_layer`와 `acpi.debug_level`의 원문 코드 예시는 그대로 보존합니다.

			    #define _COMPONENT ACPI_EVENTS
			    ACPI_DEBUG_PRINT((ACPI_DB_INFO, ...
			    acpi.debug_layer=0x20000000
			    acpi.debug_layer=0xffffffff acpi.debug_level=0x2
			    acpi.debug_layer=0x2 acpi.debug_level=0xffffffff
매개변수적용·형식한국어 설명
acpi.debug_layer=[HW,ACPI,ACPI_DEBUG] `<int>`; `acpi.debug_level=`과 함께 사용`CONFIG_ACPI_DEBUG`가 켜져 있어야 출력합니다. `debug_layer` 비트는 ACPI 소스의 `_COMPONENT`에, `debug_level` 비트는 `ACPI_DEBUG_PRINT`의 수준에 대응합니다. 기본 level mask는 `info`입니다. 자세한 내용은 `Documentation/firmware-guide/acpi/debug.rst`를 참조하십시오. 일부 값은 시스템을 못 쓸 만큼 많은 로그를 만들 수 있으므로 필요하면 `log_buf_len`을 늘립니다.
acpi_enforce_resources=[ACPI] `{ strict | lax | no }`native driver와 ACPI OperationRegion(SystemIO·SystemMemory)의 리소스 충돌을 검사합니다. `strict`(기본)는 ACPI가 점유한 리소스 접근을 거부해 legacy driver bind가 실패합니다. `lax`는 접근과 bind를 허용하되 경고를 기록합니다. `no`는 OperationRegion을 예약으로 표시하지 않고 추가 검사도 하지 않습니다.

ACPI 검증, IRQ와 테이블 설치

108-166
매개변수적용·형식한국어 설명
acpi_force_table_verification[HW,ACPI,EARLY]초기 단계에서 ACPI 테이블 checksum 검증을 켭니다. x86 초기 mapping 크기 제한 때문에 기본은 꺼져 있습니다.
acpi_irq_balance[HW,ACPI]ACPI가 활성 IRQ의 균형을 맞춥니다. APIC 모드의 기본값입니다.
acpi_irq_nobalance[HW,ACPI]ACPI가 활성 IRQ를 이동하지 않습니다. PIC 모드의 기본값입니다.
acpi_irq_isa=[HW,ACPI] `<irq>,<irq>...``irq_balance` 사용 시 나열한 IRQ를 ISA가 사용하는 것으로 표시합니다.
acpi_irq_pci=[HW,ACPI] `<irq>,<irq>...``irq_balance` 사용 시 나열한 IRQ를 PCI가 사용할 수 있도록 비웁니다.
acpi_mask_gpe=[HW,ACPI] `<byte>` 또는 `<bitmap-list>`지원하지 않는 하드웨어·펌웨어 기능의 `_Lxx`/`_Exx` 때문에 GPE dispatcher가 자동으로 끄지 못하는 GPE flood가 생길 때 해당 GPE를 mask하여 폭주를 막습니다.
acpi_no_auto_serialize[HW,ACPI]named object 생성 opcode를 포함한 AML control method를 자동으로 `Serialized` 표시하는 기본 기능을 끕니다.
acpi_no_memhotplug[ACPI]memory hotplug를 끕니다. Kdump 커널에 유용합니다.
acpi_no_static_ssdt[HW,ACPI,EARLY]초기 부팅에서 정적 SSDT 설치를 끕니다. 기본적으로 RSDT/XSDT의 SSDT는 자동 설치되어 `/sys/firmware/acpi/tables`에 나타납니다. 동적 테이블 설치에는 영향이 없으며 동적 SSDT는 `/sys/firmware/acpi/tables/dynamic`에 설치됩니다.
acpi_no_watchdog[HW,ACPI,WDT]ACPI 기반 watchdog 인터페이스(WDAT)를 무시하고 native driver가 watchdog 장치를 제어하게 합니다.
acpi_rsdp=[ACPI,EFI,KEXEC,EARLY]RSDP 주소를 커널에 전달합니다. 주로 EFI runtime service를 사용하는 시스템에서 Kdump용 두 번째 커널을 부팅할 때 사용합니다.

ACPI OS 식별과 _OSI 제어

167-233

지원 OS interface 문자열을 바꾸는 기본 명령 형식은 다음과 같습니다.

			acpi_osi="string1"	# add string1
			acpi_osi="!string2"	# remove string2
			acpi_osi=!*		# remove all strings
			acpi_osi=!		# disable all built-in OS vendor
						  strings
			acpi_osi=!!		# enable all built-in OS vendor
						  strings
			acpi_osi=		# disable all strings
매개변수적용·형식한국어 설명
acpi_os_name=[HW,ACPI] 예: `="Microsoft Windows"`ACPI BIOS에 OS 이름을 알립니다. 예시는 Windows 98로 가장합니다.
acpi_rev_override[ACPI]`_REV`가 ACPI 6 규정값 2 대신 지원 ACPI 규격 revision 5를 반환하게 합니다. 플랫폼 펌웨어에 적용하려면 연속으로 두 번 cold reboot가 필요할 수 있습니다.
acpi_osi=[HW,ACPI] OS interface 문자열 목록 수정`"string1"`은 문자열 추가, `"!string2"`는 제거, `!*`는 모든 문자열 제거, `!`는 내장 OS vendor 문자열 비활성화, `!!`는 모두 활성화, 빈 값은 모든 문자열을 비활성화하고 `_OSI` method 자체를 없앱니다. `!`는 특정 vendor 문자열을 다시 추가하는 명령과 조합할 수 있지만 feature group의 기본·현재 상태에는 영향을 주지 않습니다. `!*`는 vendor와 feature group의 현재 상태를 모두 바꾸며 명령 순서와 반복이 의미가 있지만 관련 quirk가 최종 상태를 다시 바꿀 수 있습니다. OSPM 기능 관련 BIOS 문제를 디버깅할 때 사용합니다.

ACPI 타이머, SCI와 절전

234-280
매개변수적용·형식한국어 설명
acpi_pm_good[X86]PM timer 버그 탐지를 무시하고 이 시스템의 pmtimer가 값을 latch하며 항상 올바른 값을 반환한다고 간주합니다.
acpi_sci=[HW,ACPI,EARLY] `{ level | edge | high | low }`ACPI System Control Interrupt의 trigger mode를 지정합니다.
acpi_skip_timer_override[HW,ACPI,EARLY]IRQ0/pin2 Interrupt Override를 인식하되 무시합니다. XT-PIC timer 문제를 일으키는 손상된 nForce2 BIOS용입니다.
acpi_sleep=[HW,ACPI] `{ s3_bios, s3_mode, s3_beep, s4_hwsig, s4_nohwsig, old_ordering, nonvs, sci_force_enable, nobl }``s3_bios`와 `s3_mode`는 `Documentation/power/video.rst`를 참조하십시오. `s3_beep`은 real-mode 진입점 호출 시 PC speaker를 울립니다. `s4_hwsig`는 hibernation resume 때 ACPI hardware signature가 바뀌면 복귀를 거부하며, 기본은 변경을 경고하고 복귀를 허용합니다. `s4_nohwsig`는 signature 사용과 경고를 모두 막습니다. `old_ordering`은 장치 저전력 전환에 ACPI 1.0 `_PTS` 순서를 강제합니다. `nonvs`는 suspend·hibernation·resume에서 ACPI NVS 저장·복원을 막습니다. `sci_force_enable`은 S1/S3 resume 때 `SCI_EN`을 직접 설정합니다. `nobl`은 suspend·resume 문제 시스템의 내부 blacklist를 무시하므로 신중히 사용해야 합니다.
acpi_use_timer_override[HW,ACPI,EARLY]timer override가 필요하지만 HPET가 없는 일부 손상된 Nvidia NF5 보드에서 override를 사용합니다.

EFI, AGP, ARM 정렬과 AMD IOMMU

281-351
매개변수적용·형식한국어 설명
add_efi_memmap[EFI,X86,EARLY]EFI 메모리 map을 커널의 사용 가능한 물리 RAM map에 포함합니다.
agp=[AGP] `{ off | try_unsupported }``off`는 AGP 지원을 끕니다. `try_unsupported`는 지원하지 않는 chipset 구동을 시도하며 시스템 crash나 데이터 손상을 일으킬 수 있습니다.
ALSA[HW,ALSA]`Documentation/sound/alsa-configuration.rst`를 참조하십시오.
alignment=[KNL,ARM]기본 사용자 공간 alignment fault handler 동작을 지정합니다. bit 0은 경고, bit 1은 fixup, bit 2는 segfault 전송을 활성화합니다.
align_va_addr=[X86-64] `{ 32 | 64 | on | off }`프로세스 생성 시 VMA를 할당할 때 가상 주소의 slice `[14:12]`를 지워 정렬합니다. AMD F15h에서 CPU 집약 benchmark 성능이 최대 3% 좋아질 수 있으나 워크로드와 컴파일러에 따라 크게 달라집니다. `32`는 32비트 프로세스, `64`는 64비트 프로세스, `on`은 둘 다 활성화하고 `off`는 둘 다 끕니다.
alloc_snapshot[FTRACE]main buffer 할당 시 ftrace snapshot buffer도 부팅 중 할당합니다. 부팅 초기에 `tracing_snapshot()`이 필요하지만 `GFP_KERNEL` 할당이 가능한 위치에서 해야 하는 `tracing_snapshot_alloc()`을 쓰고 싶지 않을 때 유용합니다.
allow_mismatched_32bit_el0[ARM64,EARLY]CPU 일부만 32비트 EL0를 지원하는 시스템에서 32비트 응용 프로그램의 `execve()`와 `PER_LINUX32` personality 설정을 허용합니다. 지원 CPU 집합은 `/sys/devices/system/cpu/aarch32_el0`에 표시되며 CPU hot-unplug가 제한될 수 있습니다. `Documentation/arch/arm64/asymmetric-32bit.rst`를 참조하십시오.
amd_iommu=[HW,X86-64]AMD IOMMU driver 옵션입니다. `fullflush`는 폐기되었고 `iommu.strict=1`과 같습니다. `off`는 IOMMU를 초기화하지 않습니다. `force_isolation`은 모든 장치 격리를 강제하지만 `iommu=pt`를 덮어쓰지 않습니다. `force_enable`은 문제가 알려진 플랫폼에서도 IOMMU를 강제하므로 주의해야 합니다. `pgtbl_v1`(기본)과 `pgtbl_v2`는 DMA-API page table 버전을 고릅니다. `irtcachedis`는 IRT cache를 끕니다. `nohugepages`는 v1 page table을 4KiB로 제한하고 `v2_pgsizes_only`는 4KiB/2MiB/1GiB로 제한합니다.

AMD IOMMU 진단과 P-State

352-394
매개변수적용·형식한국어 설명
amd_iommu_dump=[HW,X86-64]AMD IOMMU driver가 초기화 중 AMD IOMMU용 ACPI 테이블을 출력하게 합니다.
amd_iommu_intr=[HW,X86-64] `{ legacy | vapic }``legacy`는 legacy interrupt remapping을 사용합니다. `vapic`은 IOMMU가 guest에 직접 interrupt를 주입하는 virtual APIC 모드이며 `kvm-amd.avic=1`이 필요합니다. IOMMU 하드웨어가 지원되면 기본값입니다.
amd_pstate=[X86,EARLY] `{ disable | passive | active | guided }``disable`은 `amd_pstate`를 기본 scaling driver로 사용하지 않습니다. `passive`는 자율 선택을 끄고 driver가 희망 성능 수준을 요청합니다. `active`는 `amd_pstate_epp`가 CPPC 펌웨어에 성능(`0x0`) 또는 에너지 효율(`0xff`) 편향을 알려 하드웨어가 실시간 주파수를 조정합니다. `guided`는 driver가 최소·최대 성능을 요청하고 플랫폼이 현재 workload에 맞는 값을 범위 안에서 자율 선택합니다.
amd_prefcore=[X86] `disable``amd-pstate` preferred core 기능을 끕니다.

입력 장치, APIC와 AppArmor

395-445
매개변수적용·형식한국어 설명
amijoy.map=[HW,JOY] `<a>,<b>`Amiga joystick의 `JOY0DAT`와 `JOY1DAT`에 연결한 장치 mapping입니다. `Documentation/input/joydev/joystick.rst`를 참조하십시오.
analog.map=[HW,JOY] `<type1>,<type2>,..<type16>`16개 gameport 중 하나에 연결한 analog joystick·gamepad의 유형 또는 기능을 지정합니다.
apc=[HW,SPARC] `noidle`SPARCstation-4/5 계열의 전원 관리에서 APC CPU standby를 끕니다. SPARCstation-Fox에서 무작위 crash가 발생하면 사용합니다.
apic[APIC,X86-64]기본값으로 IO-APIC을 사용합니다.
apic=[APIC,X86,EARLY] `{ quiet | verbose | debug }`, 기본 `quiet`APIC·IO-APIC 초기화 중 출력하는 디버그 정보의 양을 조절합니다.
apic_extnmi=[APIC,X86,EARLY] `{ bsp | all | none }`, 기본 `bsp``bsp`는 external NMI를 CPU 0에만 전달합니다. `all`은 CPU 0의 backup으로 모든 CPU에 broadcast합니다. `none`은 모든 CPU에서 mask하며 dump capture kernel이 NMI로 종료되는 것을 막는 데 유용합니다.
apicpmtimerX86 timerpmtimer로 APIC timer를 보정하며 `apicmaintimer`를 암시합니다. PIT timer가 완전히 손상됐을 때 유용합니다.
autoconf=[IPV6]`Documentation/networking/ipv6.rst`를 참조하십시오.
apm=[APM]`arch/x86/kernel/apm_32.c` 머리말을 참조하십시오.
apparmor=[APPARMOR] `{ 0 | 1 }`부팅 시 AppArmor를 끄거나 켭니다. `0`은 비활성화, `1`은 활성화이며 기본값은 커널 config가 정합니다. `security/apparmor/Kconfig` 도움말을 참조하십시오.

ARCnet, ARM64 기능과 키보드

446-496
매개변수적용·형식한국어 설명
arcrimi=[HW,NET] `<io>,<irq>,<nodeID>`메모리 mapping 방식 ARCnet `RIM I` 카드 설정입니다.
arm64.no32bit_el0[ARM64]32비트 응용 프로그램 실행을 무조건 금지합니다.
arm64.nobti[ARM64]Branch Target Identification 지원을 무조건 끕니다.
arm64.nogcs[ARM64]Guarded Control Stack 지원을 무조건 끕니다.
arm64.nomops[ARM64]Memory Copy와 Memory Set instruction 지원을 무조건 끕니다.
arm64.nompam[ARM64]Memory Partitioning And Monitoring 지원을 무조건 끕니다.
arm64.nomte[ARM64]Memory Tagging Extension 지원을 무조건 끕니다.
arm64.nopauth[ARM64]Pointer Authentication 지원을 무조건 끕니다.
arm64.nosme[ARM64]Scalable Matrix Extension 지원을 무조건 끕니다.
arm64.nosve[ARM64]Scalable Vector Extension 지원을 무조건 끕니다.
ataflop=[HW,M68k]Atari floppy 관련 매개변수입니다.
atarimouse=[HW,MOUSE]Atari mouse 관련 매개변수입니다.
atkbd.extra=[HW]IBM RapidAccess, EzKey 및 유사 키보드의 추가 LED와 key를 활성화합니다.
atkbd.reset=[HW]초기화 중 키보드를 reset합니다.
atkbd.set=[HW] `<int>`키보드 code set을 선택합니다. `2`는 AT(기본), `3`은 PS/2입니다.
atkbd.scroll=[HW]MS Office 및 유사 키보드의 scroll wheel을 활성화합니다.
atkbd.softraw=[HW] `<bool>``0`은 실제 raw, `1`은 synthetic raw(기본) 모드입니다.
atkbd.softrepeat=[HW]software keyboard repeat를 사용합니다.

감사, BAU, BayCom과 mounted block 쓰기

497-548
매개변수적용·형식한국어 설명
audit=[KNL] `{ 0 | 1 | off | on }`, 기본 unset`0/off`는 다음 재부팅까지 kernel audit를 완전히 끕니다. unset은 audit를 초기화하되 비활성화하고 사용자 공간 `auditd`가 완전히 켜게 합니다. `1/on`은 부분 활성화하여 `auditd`가 완전히 켤 때까지 RAM에 최대 `audit_backlog_limit`개 메시지를 저장합니다.
audit_backlog_limit=[KNL] `<int>`(0 이상), 기본 `64`audit queue 크기 제한을 설정합니다.
bau=[X86_UV] `{ 0 | 1 }`SGI UV의 BAU를 제어합니다. `1`은 활성화하고 `0` 또는 미지정은 비활성화합니다.
baycom_epp=[HW,AX25] `<io>,<mode>`BayCom EPP AX.25 modem 설정입니다.
baycom_par=[HW,AX25] `<io>,<mode>`BayCom parallel port AX.25 modem 설정입니다. `drivers/net/hamradio/baycom_par.c` 머리말을 참조하십시오.
baycom_ser_fdx=[HW,AX25] `<io>,<irq>,<mode>[,<baud>]`BayCom serial port AX.25 modem의 full duplex 모드입니다. `drivers/net/hamradio/baycom_ser_fdx.c` 머리말을 참조하십시오.
baycom_ser_hdx=[HW,AX25] `<io>,<irq>,<mode>`BayCom serial port AX.25 modem의 half duplex 모드입니다. `drivers/net/hamradio/baycom_ser_hdx.c` 머리말을 참조하십시오.
bdev_allow_write_mounted=`<bool>`, 기본 Y; `CONFIG_BLK_DEV_WRITE_MOUNTED`mounted block device를 쓰기 모드로 열어 파일시스템을 우회하는 쓰기를 허용할지 제어합니다. fuzzer가 mounted FS의 metadata를 몰래 덮어 커널을 crash시키거나 `O_EXCL`을 쓰지 않는 단순 저장 도구가 mounted filesystem을 파괴적으로 format하는 일을 막기 위해 도입됐습니다.

펌웨어 표, 부팅과 초기 C 항목

549-610
매개변수적용·형식한국어 설명
bert_disable[ACPI]문제가 있는 BIOS에서 BERT OS 지원을 끕니다.
bgrt_disable[ACPI,X86,EARLY]OEM logo 깜빡임을 피하도록 BGRT를 끕니다.
blkdevparts=embedded block device partition명령줄 입력으로 embedded block device의 파티션을 수동 해석합니다. `Documentation/block/cmdline-partition.rst`를 참조하십시오.
boot_delay=[KNL,EARLY] 정수 밀리초부팅 중 각 `printk`를 지정한 밀리초만큼 지연합니다. `CONFIG_BOOT_PRINTK_DELAY`가 필요하고 `lpj=`도 필요할 수 있습니다. 10초(10000) 초과 값은 오류로 보고 무시합니다.
bootconfig[KNL,EARLY]initrd에 확장 명령줄 옵션을 추가하고 커널이 이를 찾게 합니다. `Documentation/admin-guide/bootconfig.rst`를 참조하십시오.
bttv.card=[HW,V4L] `bttv.radio=`, `bttv.pll=`, `bttv.tuner=` 포함bt848·bt878 기반 grabber card용 bttv 옵션입니다. 주요 insmod 옵션을 커널 인수로도 쓸 수 있습니다. `Documentation/admin-guide/media/bttv.rst`를 참조하십시오.
bulk_remove=off[PPC]여러 HPTE 항목을 한 번에 flush하는 pSeries 펌웨어 기능을 사용하지 않습니다.
c101=[NET]Moxa C101 synchronous serial card 설정입니다.
cachesize=[BUGS=X86-32]level 2 CPU cache 크기 탐지값을 재정의합니다. 하드웨어 버그로 cache 크기를 잘못 보고하고 커널이 정확한 값을 결정할 수 없을 때 사용합니다.
carrier_timeout=[NET] 초 단위, 기본 `120`네트워크 carrier를 기다릴 시간을 지정합니다.
ca_keys=[KEYS] `{ id:<keyid> | builtin }`certificate trust 검증에 사용할 시스템 trusted keyring의 특정 key를 식별합니다.
cca=[MIPS,EARLY] `0`~`7`커널 페이지의 cache coherency algorithm을 재정의합니다. SB1·Loongson3 등 플랫폼별 값은 `arch/mips/include/asm/pgtable-bits.h`를 참조하십시오.
ccw_timeout_log[S390]`Documentation/arch/s390/common_io.rst`를 참조하십시오.

CFI와 cgroup·SELinux

611-681
매개변수적용·형식한국어 설명
cfi=[X86-64], `CONFIG_FINEIBT`, `feature[,feature...]`, 기본 `auto``auto`는 IBT가 있으면 FineIBT, 아니면 kCFI를 쓰며 FRED가 없으면 paranoid mode를 켭니다. `off`는 CFI 검사를 끄고 `kcfi`는 kCFI만, `fineibt`는 IBT가 없어도 FineIBT를 사용합니다. `norand`는 CFI hash 재무작위화를 막고 `paranoid`는 caller hash 검사를 추가합니다. `bhi`는 FineIBT를 가로지르는 speculation을 막도록 register poisoning을 켭니다. `warn`은 강제하지 않고 경고만 하며 `debug`는 초기화 세부 정보를 보고합니다.
cgroup_disable=[KNL] 비활성화할 controller·feature 이름지정한 controller를 전체 cgroup 단일 hierarchy 자동 mount에서 제외하고 개별 mount 가능한 subsystem으로도 보이지 않게 합니다. optional feature면 기능과 관련 cgroup 파일 생성을 끕니다. 현재 실질적인 overhead 절감은 `cgroup_disable=memory`만 제공합니다. `pressure`를 지정하면 cgroup별 pressure stall information 회계를 끕니다.
cgroup_no_v1=[KNL] controller, `all`, `named` 목록`cgroup_disable`과 비슷하지만 cgroup v1에만 적용해 차단된 controller를 cgroup2에서는 계속 사용할 수 있습니다. `all`은 모든 controller, `named`는 named mount를 막으며 둘을 함께 지정하면 모든 v1 hierarchy를 끕니다.
cgroup_v1_proc=[KNL] `{ true | false }`기본 `/proc/cgroups`는 v1 controller만 나열합니다. 이 호환 옵션은 v1 코드가 빌드되지 않은 v2 controller도 표시하여 준 legacy 소프트웨어가 v2 사용 여부를 판단하게 합니다.
cgroup_favordynmods=[KNL] `{ true | false }``favordynmods`를 켜거나 끕니다. 기본값은 `CONFIG_CGROUP_FAVOR_DYNMODS`입니다.
cgroup.memory=[KNL] `<string>`cgroup memory controller 옵션입니다. `nosocket`은 socket memory accounting, `nokmem`은 kernel memory accounting, `nobpf`는 BPF memory accounting을 끕니다.
checkreqprot=[SELINUX] `{ 0 | 1 }`초기 `checkreqprot` 값을 설정합니다. `0`은 implied execute protection을 포함해 커널이 적용한 보호를 검사하고 `1`은 응용 프로그램이 요청한 보호를 검사합니다. 기본은 커널 config가 정하며 `/sys/fs/selinux/checkreqprot`에서 런타임 변경할 수 있습니다. 값 1은 폐기 예정입니다.

s390 I/O와 CPUID 기능 차단

682-702
매개변수적용·형식한국어 설명
cio_ignore=[S390]`Documentation/arch/s390/common_io.rst`를 참조하십시오.
clearcpuid=X[,X...][X86] feature bit 번호 또는 `/proc/cpuinfo` flags 이름커널에서 CPUID 기능 X를 끕니다. 유효 bit 번호는 `arch/x86/include/asm/cpufeatures.h`를 참조하십시오. Linux 전용 bit는 커널 옵션 사이에서 안정적이지 않을 수 있지만 vendor bit는 안정적이어야 합니다. `/proc/cpuinfo`의 `flags:` 이름도 쓸 수 있으나 모든 기능에 이름이 있는 것은 아닙니다. 이 옵션은 커널을 taint하며, CPUID를 직접 호출하거나 검사 없이 기능을 쓰는 사용자 프로그램에는 기능이 계속 보입니다. 커널이 기능을 사용하거나 `/proc/cpuinfo`에 표시하는 것만 막습니다. 중요한 bit를 끄면 커널이 오동작할 수 있습니다.

Clocksource와 CMA

703-803
매개변수적용·형식한국어 설명
clk_ignore_unused[CLK]Linux device driver가 명시적으로 켜지 않았지만 reset 또는 bootloader·firmware가 하드웨어에서 켠 clock을 clock framework가 자동 gating하지 않게 합니다. clock을 항상 켜 두거나 예약하는 옵션은 아닙니다. 디버그·개발용이며 driver 지원이 올바른 플랫폼에는 필요 없어야 합니다. `Documentation/driver-api/clk.rst`를 참조하십시오.
clock=[BUGS=X86-32,HW] `{ pit | tsc | cyclone | pmtmr }`, 폐기 예정`gettimeofday()` 계산에 지정 clocksource를 강제합니다. 사용할 수 없으면 PIT로 돌아갑니다.
clocksource=`<string>`기본 clocksource를 이름으로 재정의합니다. 플랫폼별 후보에는 공통 fallback `jiffies`, ACPI `acpi_pm`, ARM의 `imx_timer1`, `OSTS`, `netx_timer`, `mpu_timer2`, `pxa_timer`, `timer3`, `32k_counter`, `timer0_1`, x86의 `pit`, `hpet`, `tsc`, Geode의 `scx200_hrt`, IBM x440의 `cyclone`, MIPS, PARISC `cr16`, S390 `tod`, SH `SuperH`, SPARC64 `tick`이 있습니다.
clocksource.arm_arch_timer.evtstrm=[ARM,ARM64,EARLY] `<bool>`ARM architected timer의 eventstream 기능을 켜거나 꺼 WFE 기반 polling loop를 운영 시스템에서도 더 효과적으로 디버깅하게 합니다.
clocksource.verify_n_cpus=[KNL]`CLOCK_SOURCE_VERIFY_PERCPU` 표시 clocksource의 과도한 skew를 검사할 CPU 수를 제한합니다. 음수는 모든 CPU, 0은 검사 안 함이며 `nr_cpu_ids`보다 큰 값은 그 값으로 잘립니다. 실제 CPU는 중복 없이 무작위로 선택합니다.
clocksource-wdtest.holdoff=[KNL] 초clocksource watchdog test가 시험을 시작하기 전 기다리는 시간을 정합니다. 모듈이면 기본 0초, 커널 내장이면 10초입니다.
cma=nn[MG]@[start[MG][-end[MG]]][KNL,CMA,EARLY]연속 메모리 할당용 전역 CMA 크기와 선택적인 물리 주소 배치 범위를 설정합니다. 0은 CMA를 완전히 끕니다. `kernel/dma/contiguous.c`를 참조하십시오.
cma_pernuma=nn[MG][KNL,CMA,EARLY], 기본 `0`NUMA 노드마다 연속 메모리 할당용 CMA 크기를 정합니다. 0은 per-NUMA CMA를 끕니다. 활성화하면 노드 nid의 DMA 사용자가 먼저 같은 노드 영역에서 buffer를 할당하고 실패하면 전역 기본 영역으로 돌아갑니다.
numa_cma=<node>:nn[MG][,<node>:nn[MG]][KNL,CMA,EARLY]지정한 노드에 NUMA CMA 영역을 예약합니다. 노드 nid의 DMA 사용자는 해당 노드 영역을 먼저 사용하고 실패하면 전역 기본 영역으로 돌아갑니다.
cmo_free_hint=[PPC] `{ yes | no }`, 기본 `yes`해제한 페이지를 inactive로 표시할지 정합니다. CMO 환경에서 hypervisor가 page stealing에 사용할 OS memory pressure를 판단합니다.

DMA pool, ARCnet와 s390 콘솔

804-832
매개변수적용·형식한국어 설명
coherent_pool=nn[KMG][ARM,KNL,EARLY], 기본 `256K`coherent atomic DMA 할당용 메모리 pool 크기를 설정합니다.
com20020=[HW,NET] `<io>[,<irq>[,<nodeID>[,<backplane>[,<ckp>[,<timeout>]]]]]`COM20020 chipset 기반 ARCnet 카드 설정입니다.
com90io=[HW,NET] `<io>[,<irq>]`I/O-mapped buffer를 쓰는 COM90xx ARCnet chipset 설정입니다.
com90xx=[HW,NET] `<io>[,<irq>[,<memstart>]]`memory-mapped buffer를 쓰는 COM90xx ARCnet chipset 설정입니다.
condev=[HW,S390], `conmode=` 포함s390 콘솔 장치와 콘솔 모드를 지정합니다.
con3215_drop=[S390,EARLY] `{ y | n | Y | N | 1 | 0 }`true이면 3215 console buffer가 찼을 때 데이터를 버립니다. x3270 같은 3270 terminal emulator에서 clear key를 눌러야 출력과 커널이 진행되는 상황을 피하므로 부팅이 훨씬 빨라집니다. 3270 emulator를 쓰지 않으면 영향이 없습니다.

출력 콘솔과 메시지 형식

833-920

serial core bus의 tty mapping 확인 예시는 원문 그대로 보존합니다.

			$ ls -d /sys/bus/serial-base/devices/*:*.*/tty/*
			/sys/bus/serial-base/devices/00:04:0.0/tty/ttyS0
매개변수적용·형식한국어 설명
console=[KNL] console device와 options`tty<n>`은 virtual console을 사용합니다. `ttyS<n>[,options]`와 `ttyUSB0[,options]`는 serial port를 사용하며 options는 baud·parity(`n/o/e`)·bit 수·RTS flow control을 나타내는 `bbbbpnf`, 기본 `9600n8` 형식입니다. `<DEVNAME>:<n>.<n>[,options]`는 물리 serial 장치 이름, controller instance와 port instance로 serial core bus의 port를 지정하며 driver가 준비된 뒤 등록됩니다. 조기 출력에는 `earlycon`을 함께 쓸 수 있습니다. `uart[8250],{io|mmio|mmio16|mmio32},<addr>[,options]` 또는 `uart[8250],0x<addr>[,options]`는 8250/16550 UART에서 polled early console을 시작한 뒤 일치하는 ttyS로 전환합니다. stride는 8·16·32비트이며 주소 유형 생략 시 `mmio`입니다. options를 생략하면 하드웨어를 다시 초기화하지 않습니다. `hvc<n>`은 Xen·PowerPC hypervisor console입니다. `null` 또는 빈 문자열은 console 출력을 버리며 유일한 `console=`이어야 합니다. Braille 장치는 `console=brl,ttyS0`처럼 `brl,`을 앞에 붙이며 현재 VisioBraille만 지원합니다. `Documentation/admin-guide/serial-console.rst`와 대안인 `Documentation/networking/netconsole.rst`를 참조하십시오.
console_msg_format=[KNL] `{ default | syslog }``default`는 `[time stamp] text\n` 형식이며 `CONFIG_PRINTK_TIME` 또는 `printk_time`에 따라 timestamp를 생략할 수 있습니다. `syslog`는 `<%u>[time stamp] text\n` 형식으로 facility·loglevel 접두사를 붙이며 `syslog()` 호출, `dmesg -S --raw`, `/proc/kmsg`와 비슷합니다.
consoleblank=[KNL] 초, 기본 `0`console blank(screen saver) timeout입니다. 0은 blank timer를 끕니다.
coredump_filter=[KNL]`/proc/<pid>/coredump_filter`의 기본값을 바꿉니다. `Documentation/filesystems/proc.rst`를 참조하십시오.

CoreSight, CPU idle·frequency와 hotplug

921-957
매개변수적용·형식한국어 설명
coresight_cpu_debug.enable[ARM,ARM64] `<bool>`CPU sampling 기반 디버깅을 제어합니다. 0은 기본 비활성화, 1은 부팅 시 활성화입니다.
cpcihp_generic=[HW,PCI] `<first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>]`generic port I/O CompactPCI driver 설정입니다.
cpuidle.off=1[CPU_IDLE]cpuidle 하위 시스템을 끕니다.
cpuidle.governor=[CPU_IDLE] governor 이름사용할 cpuidle governor를 지정합니다.
cpufreq.off=1[CPU_FREQ]cpufreq 하위 시스템을 끕니다.
cpufreq.default_governor=[CPU_FREQ] governor·policy 이름기본 cpufreq governor 또는 policy를 지정합니다. cpufreq driver probe 전에 커널에 등록되어 있어야 합니다.
cpu_init_udelay=N[X86,EARLY] 마이크로초, 기본 `10000`프로세서를 시작하도록 APIC INIT을 assert한 뒤 de-assert하기 전 지연합니다. 부팅과 suspend resume 등 CPU online 때마다 적용됩니다.
cpuhp.parallel=[SMP] `<bool>`secondary CPU의 병렬 bringup을 제어합니다. `CONFIG_HOTPLUG_PARALLEL=y`이면 기본 활성화되며 그렇지 않으면 효과가 없습니다.

Kdump와 crashkernel 예약

958-1037
매개변수적용·형식한국어 설명
crash_kexec_post_notifiersKdump panic 경로panic notifier 실행과 kmsg dump가 끝난 뒤에만 Kdump 커널로 이동합니다. notifier가 crash 커널을 더 불안정하게 해 실패 위험이 커지지만, Kdump 신뢰성이 낮은 구성에서는 다른 CPU stack trace나 `panic_print` 데이터처럼 dmesg 정보를 더 수집할 수 있습니다. Hyper-V, PowerPC fadump, AMD SEV-SNP 같은 일부 구성은 항상 활성화합니다.
crashkernel=size[KMG][@offset[KMG]][KNL,EARLY]panic 시 kexec로 전환할 crash kernel을 위해 물리 영역 `[offset, offset + size]`를 예약합니다. offset을 생략하면 자동 선택합니다. X86-64, ARM64, RISCV, LoongArch는 offset 미지정 시 4G 아래를 먼저 찾고 실패하면 4G 위를 예약합니다. `Documentation/admin-guide/kdump/kdump.rst`를 참조하십시오.
crashkernel=range1:size1[,range2:size2,...][@offset][KNL], 범위 `start-[end]`실행 시스템 메모리 크기에 따라 예약량을 고릅니다. start와 end는 `amount[KMG]` 단위입니다. 예시는 Kdump 문서를 참조하십시오.
crashkernel=size[KMG],high[KNL,X86-64,ARM64,RISCV,LoongArch]; `,low`와 `,cma` 포함`,high`는 위쪽부터 할당해 RAM이 4G보다 많으면 4G 위 영역을 허용하고, 그렇지 않으면 가능한 4G 아래를 사용합니다. `crashkernel=X`가 있으면 무시됩니다. `crashkernel=size[KMG],low`는 high 예약과 함께 두 번째 커널·swiotlb·32비트 DMA용 4G 아래 메모리를 사용자가 지정합니다. 기본은 x86에서 `max(swiotlb_size_or_default() + 8MiB, 256MiB)`, arm64·riscv·loongarch에서 128MiB이며 0은 low 할당을 끕니다. high를 쓰지 않거나 예약이 이미 4G 아래면 무시됩니다. X86의 `crashkernel=size[KMG],cma`는 첫 시스템 사용자 공간과 movable kernel allocation(memory balloon, zswap)이 쓸 추가 CMA를 예약합니다. 여기서 할당한 페이지는 vmcore에 빠지며 표준 crashkernel 예약도 별도로 필요합니다. 첫 커널 DMA가 두 번째 커널 메모리를 손상시킬 수 있어 실패 위험이 커지지만 충분한 전용 메모리를 희생할 수 없는 시스템을 위한 선택입니다.
Crash kernel memory reservation 구조
예약주소·기본값역할과 제약
`crashkernel=...,high`Top-down, 4 GiB 위 허용Crash kernel 본체; `crashkernel=X`가 있으면 무시
`crashkernel=...,low` · x864 GiB 아래 · max(SWIOTLB + 8 MiB, 256 MiB)SWIOTLB와 32-bit DMA buffer
`crashkernel=...,low` · arm64/riscv/loongarch4 GiB 아래 · 128 MiBLow-memory 보조 예약; 0이면 비활성화
`crashkernel=...,cma` · x86CMA movable 영역첫 kernel과 공유 가능, vmcore에서 page 누락 가능

`high`는 crash kernel 본체를 4 GiB 위까지 배치하고 `low`는 DMA·SWIOTLB에 필요한 낮은 주소 공간을 보충합니다. `cma`는 첫 kernel과 공유할 수 있지만 vmcore 누락과 DMA 손상 위험이 있어 표준 예약을 대신하지 않습니다.

Crypto, 네트워크 장치와 기본 디버그

1038-1063
매개변수적용·형식한국어 설명
cryptomgr.notests[KNL]crypto self-test를 끕니다.
cs89x0_dma=[HW,NET] `<dma>`CS89x0 network device의 DMA 설정입니다.
cs89x0_media=[HW,NET] `{ rj45 | aui | bnc }`CS89x0 network media 유형을 지정합니다.
csdlock_debug=[KNL]cross-CPU function call 처리의 디버그 확장을 제어합니다. 활성화하면 CPU hang 탐지 시 추가 데이터를 console에 출력하고 해당 CPU를 다시 ping해 hang 해소를 시도합니다. 기본값은 `CSD_LOCK_WAIT_DEBUG_DEFAULT` Kconfig 옵션에 따릅니다.
dasd=[HW,NET]`drivers/s390/block/dasd_devmap.c` 머리말을 참조하십시오.
db9.dev[2|3]=[HW,JOY] `<port#>,<type>`parallel port를 통한 Multisystem joystick을 port당 하나 설정합니다. `Documentation/input/devices/joystick-parport.rst`를 참조하십시오.

커널·객체·페이지와 debugfs 디버깅

1064-1130
매개변수적용·형식한국어 설명
debug[KNL,EARLY]커널 디버깅을 활성화해 event log level을 높입니다.
debug_boot_weak_hash[KNL,EARLY]부팅 초기에 hash된 pointer를 출력합니다. `(___ptrval___)` 대신 값이 필요할 때 siphash 대신 약한 hash를 사용합니다. 암호학적으로 안전하지 않으므로 운영 커널에는 쓰지 마십시오.
debug_locks_verbose=[KNL] `<int>` bitmask, 기본 `0`locking API self-test 중 디버그 정보를 출력합니다. 여러 `LOCKTYPE_` 시험의 bitmask이며 -1은 모든 bit를 켜 매우 많은 정보를 출력하므로 보통 lockdep 개발자용입니다.
debug_objects[KNL,EARLY]object debugging을 활성화합니다.
debug_guardpage_minorder=[KNL,EARLY], `CONFIG_DEBUG_PAGEALLOC`buddy allocator가 의도적으로 비워 보호할 페이지 order를 제어합니다. 값이 크면 무작위 메모리 손상 탐지 확률은 높지만 일반 시스템 메모리는 줄어듭니다. 최대 `MAX_PAGE_ORDER/2`이며 1이나 2로 대부분의 kernel·driver 무작위 CPU 접근 버그를 찾을 수 있습니다. MMU를 우회하는 불량 하드웨어·펌웨어·DMA 쓰기는 탐지하지 못합니다.
debug_pagealloc=[KNL,EARLY], `CONFIG_DEBUG_PAGEALLOC`, `on`부팅 시 page allocation 디버깅을 켭니다. 기본은 비활성화입니다. 오류 보고를 최대한 얻으려면 `page_owner`도 활성화하는 것이 유용합니다.
debugfs=[KNL,EARLY] `{ on | no-mount | off }``on`은 모든 기능을 켭니다. `no-mount`는 filesystem을 등록하지 않아 mount할 수 없지만 커널 client는 API에 접근하고 crashkernel이 내용을 읽을 수 있습니다. `off`는 filesystem을 등록하지 않고 client의 파일·디렉터리 등록에 `-EPERM`을 반환해 debugfs가 빌드되지 않은 것과 같습니다. 기본은 build-time config로 정합니다.
debugpat[X86]PAT 디버깅을 활성화합니다.

HugeTLB, deferred probe와 Dell SMM

1131-1178
매개변수적용·형식한국어 설명
default_hugepagesz=[HW] `size[KMG]`legacy `/proc/` hugepages API가 나타내고 `shmget()`, `mmap()`, hugetlbfs mount가 기본으로 쓰는 HugeTLB page 크기입니다. 생략하면 아키텍처 기본 huge page 크기입니다. `Documentation/admin-guide/mm/hugetlbpage.rst`를 참조하십시오.
deferred_probe_timeout=[KNL] 초deferred probe가 dependency를 기다리다 포기할 timeout입니다. 명시적으로 참여한 subsystem·driver dependency만 무시됩니다. 0은 initcall 끝에서 timeout합니다. 만료 전 성공한 driver 등록 때마다 다시 시작하며 retry 후에도 deferred 목록에 남은 장치를 출력합니다.
delayacct[KNL]task별 delay accounting을 활성화합니다.
dell_smm_hwmon.ignore_dmi=[HW]DMI가 미지원 하드웨어라고 해도 probe를 계속합니다.
dell_smm_hwmon.force=[HW]SMM BIOS signature가 지원 모델과 맞지 않아도 driver와 blacklist 기능을 강제로 활성화합니다.
dell_smm_hwmon.power_status=[HW], 기본 비활성`/proc/i8k`에 power status를 보고합니다.
dell_smm_hwmon.restricted=[HW]`SYS_ADMIN` capability가 있을 때만 fan 제어를 허용합니다.
dell_smm_hwmon.fan_mult=[HW]fan speed에 곱할 계수입니다.
dell_smm_hwmon.fan_max=[HW]설정 가능한 최대 fan speed입니다.

s390 압축과 기능 비활성화

1179-1233
매개변수적용·형식한국어 설명
dfltcc=[HW,S390] `{ on | off | def_only | inf_only | always }``on`(기본)은 level 1 압축과 압축 해제에 s390 zlib 하드웨어를 사용합니다. `off`는 하드웨어를 쓰지 않고 `def_only`는 deflate만, `inf_only`는 inflate만 사용합니다. `always`는 `on`과 같지만 선택 압축 level을 무시하고 항상 하드웨어를 쓰는 디버그 모드입니다.
dhash_entries=[KNL]dentry cache의 hash bucket 수를 설정합니다.
disable_1tb_segments[PPC,EARLY]1TB hash page table segment를 끄고 256MB segment로 돌아갑니다. SLB miss가 필요한 문제를 디버깅할 때 유용합니다.
disable=[IPV6]`Documentation/networking/ipv6.rst`를 참조하십시오.
disable_radix[PPC,EARLY]POWER9의 RADIX MMU 모드를 끕니다.
disable_tlbie[PPC]TLBIE instruction을 끕니다. 현재 KVM, HASH MMU, coherent accelerator와 함께 동작하지 않습니다.
disable_ddw[PPC/PSERIES,EARLY]손상된 펌웨어를 우회하도록 Dynamic DMA Window 지원을 끕니다.
disable_ipv6=[IPV6]`Documentation/networking/ipv6.rst`를 참조하십시오.
disable_mtrr_cleanup[X86,EARLY]X server driver가 나중에 WB 항목을 추가할 수 있도록 커널이 연속 MTRR 배치를 discrete하게 조정하는 기능을 끕니다.
disable_mtrr_trim[X86,Intel·AMD,EARLY]기본적으로 MTRR 설정에 따라 uncacheable memory를 사용 가능 pool에서 잘라내는 동작을 끕니다. 시스템이 매우 느려질 수 있습니다.
disable_timer_pin_1[X86,EARLY]APIC timer의 PIN 1을 끕니다. chipset bug 우회에 유용할 수 있습니다.
dis_ucode_ldr[X86]microcode loader를 끕니다.

DMA API 디버깅과 RFDS 완화

1234-1270
매개변수적용·형식한국어 설명
dma_debug=off`DMA_API_DEBUG`커널이 DMA API 디버그 지원으로 빌드됐어도 부팅 시 디버그 코드를 끕니다.
dma_debug_entries=<number>DMA API debug entry 수DMA API 할당마다 하나 필요한 사전 할당 entry 수를 조절합니다. 아키텍처 기본값이 너무 작아 디버그 코드가 스스로 꺼질 때 사용합니다.
dma_debug_driver=<driver_name>driver filter부팅 시 지정 driver만 대상으로 DMA API 디버그 filter를 켭니다. 나중에 sysfs로 끄거나 다른 driver로 바꿀 수 있습니다.
reg_file_data_sampling=[X86] `{ on | off }`Intel Atom에만 영향을 주는 Register File Data Sampling(RFDS) 취약점 완화를 제어합니다. RFDS는 사용자 공간이 floating point·vector·integer register에 이전에 있던 커널 데이터를 추론하게 할 수 있습니다. 이 값은 `CONFIG_MITIGATION_RFDS` 기본을 덮어쓰지만 MDS처럼 다른 VERW 기반 완화가 켜져 있으면 끌 수 없습니다. 모두 끄려면 모든 VERW 기반 완화를 꺼야 합니다. `Documentation/admin-guide/hw-vuln/reg-file-data-sampling.rst`를 참조하십시오.

비동기 probe, EDID와 device tree

1271-1305
매개변수적용·형식한국어 설명
driver_async_probe=[KNL] `<driver_name1>,<driver_name2>...`비동기로 probe할 driver 이름 목록입니다. `*`는 모든 이름과 일치하고, 함께 나열한 나머지 이름은 `*`에서 제외됩니다.
drm.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>]EDID firmware 대체손상된 monitor·graphic adapter·KVM·EDID 없는 panel 대신 `/lib/firmware`의 EDID data set을 사용합니다. `connector:`를 앞에 붙이면 해당 connector에만 적용하며 쉼표로 각 connector의 고유 파일을 지정할 수 있습니다. connector 이름이 없는 data set은 명시되지 않은 모든 connector에 적용됩니다.
dscc4.setup=[NET]DSCC4 network driver 설정입니다.
dt_cpu_ftrs=[PPC,EARLY] `{ off | known }`존재하는 경우 CPU feature 발견·설정에 `dt_cpu_ftrs` device-tree binding을 쓰는 방식을 제어합니다. `off`는 사용하지 않고 legacy CPU table로 돌아갑니다. `known`은 커널이 아는 기능만 guest·사용자 공간에 전달합니다.
dump_apple_properties[X86]x86 Mac의 EFI device property 이름과 내용을 출력합니다. driver 작성자가 사용 가능한 데이터를 확인하거나 reverse engineering할 때 유용합니다.

Dynamic debug와 early ioremap

1306-1316
매개변수적용·형식한국어 설명
dyndbg[="val"][KNL,DYNAMIC_DEBUG], `<module>.dyndbg[="val"]`부팅 시 debug message를 활성화합니다. `Documentation/admin-guide/dynamic-debug-howto.rst`를 참조하십시오.
early_ioremap_debug[KNL,EARLY]early ioremap 지원의 디버그 메시지를 켭니다. 해제되지 않은 임시 early mapping을 추적할 때 유용합니다.

초기 콘솔 장치

1317-1452

`earlycon=`은 정규 console이 준비되기 전 사용할 출력 장치와 옵션을 정합니다. 옵션 없이 쓰면 platform이 지원하는 경우 device tree `chosen` node의 `stdout-path` property 또는 ACPI SPCR table에서 초기 콘솔을 결정합니다.

		cdns,<addr>[,options]
			Start an early, polled-mode console on a Cadence
			(xuartps) serial port at the specified address. Only
			supported option is baud rate. If baud rate is not
			specified, the serial port must already be setup and
			configured.

		uart[8250],io,<addr>[,options[,uartclk]]
		uart[8250],mmio,<addr>[,options[,uartclk]]
		uart[8250],mmio32,<addr>[,options[,uartclk]]
		uart[8250],mmio32be,<addr>[,options[,uartclk]]
		uart[8250],0x<addr>[,options]
			Start an early, polled-mode console on the 8250/16550
			UART at the specified I/O port or MMIO address.
			MMIO inter-register address stride is either 8-bit
			(mmio) or 32-bit (mmio32 or mmio32be).
			If none of [io|mmio|mmio32|mmio32be], <addr> is assumed
			to be equivalent to 'mmio'. 'options' are specified
			in the same format described for "console=ttyS<n>"; if
			unspecified, the h/w is not initialized. 'uartclk' is
			the uart clock frequency; if unspecified, it is set
			to 'BASE_BAUD' * 16.

		pl011,<addr>
		pl011,mmio32,<addr>
			Start an early, polled-mode console on a pl011 serial
			port at the specified address. The pl011 serial port
			must already be setup and configured. Options are not
			yet supported.  If 'mmio32' is specified, then only
			the driver will use only 32-bit accessors to read/write
			the device registers.

		liteuart,<addr>
			Start an early console on a litex serial port at the
			specified address. The serial port must already be
			setup and configured. Options are not yet supported.

		meson,<addr>
			Start an early, polled-mode console on a meson serial
			port at the specified address. The serial port must
			already be setup and configured. Options are not yet
			supported.

		msm_serial,<addr>
			Start an early, polled-mode console on an msm serial
			port at the specified address. The serial port
			must already be setup and configured. Options are not
			yet supported.

		msm_serial_dm,<addr>
			Start an early, polled-mode console on an msm serial
			dm port at the specified address. The serial port
			must already be setup and configured. Options are not
			yet supported.

		owl,<addr>
			Start an early, polled-mode console on a serial port
			of an Actions Semi SoC, such as S500 or S900, at the
			specified address. The serial port must already be
			setup and configured. Options are not yet supported.

		rda,<addr>
			Start an early, polled-mode console on a serial port
			of an RDA Micro SoC, such as RDA8810PL, at the
			specified address. The serial port must already be
			setup and configured. Options are not yet supported.

		sbi
			Use RISC-V SBI (Supervisor Binary Interface) for early
			console.

		smh	Use ARM semihosting calls for early console.

		s3c2410,<addr>
		s3c2412,<addr>
		s3c2440,<addr>
		s3c6400,<addr>
		s5pv210,<addr>
		exynos4210,<addr>
			Use early console provided by serial driver available
			on Samsung SoCs, requires selecting proper type and
			a correct base address of the selected UART port. The
			serial port must already be setup and configured.
			Options are not yet supported.

		lantiq,<addr>
			Start an early, polled-mode console on a lantiq serial
			(lqasc) port at the specified address. The serial port
			must already be setup and configured. Options are not
			yet supported.

		lpuart,<addr>
		lpuart32,<addr>
			Use early console provided by Freescale LP UART driver
			found on Freescale Vybrid and QorIQ LS1021A processors.
			A valid base address must be provided, and the serial
			port must already be setup and configured.

		ec_imx21,<addr>
		ec_imx6q,<addr>
			Start an early, polled-mode, output-only console on the
			Freescale i.MX UART at the specified address. The UART
			must already be setup and configured.

		ar3700_uart,<addr>
			Start an early, polled-mode console on the
			Armada 3700 serial port at the specified
			address. The serial port must already be setup
			and configured. Options are not yet supported.

		qcom_geni,<addr>
			Start an early, polled-mode console on a Qualcomm
			Generic Interface (GENI) based serial port at the
			specified address. The serial port must already be
			setup and configured. Options are not yet supported.

		efifb,[options]
			Start an early, unaccelerated console on the EFI
			memory mapped framebuffer (if available). On cache
			coherent non-x86 systems that use system memory for
			the framebuffer, pass the 'ram' option so that it is
			mapped with the correct attributes.

		linflex,<addr>
매개변수적용·형식한국어 설명
earlycon=[KNL,EARLY]부팅 극초기에 사용할 console backend를 선택합니다. 아래 주소와 형식은 원문 코드 블록에 그대로 보존했습니다.
cdns,<addr>[,options]Cadence xuartps지정 주소의 Cadence serial port에서 polling 방식 초기 콘솔을 시작합니다. 지원 옵션은 baud rate뿐이며, 생략하면 port가 이미 설정·구성되어 있어야 합니다.
uart[8250],{io|mmio|mmio32|mmio32be},<addr>[,options[,uartclk]]8250/16550 UARTI/O port 또는 MMIO 주소에서 polling console을 시작합니다. register stride는 `mmio`가 8bit, `mmio32`·`mmio32be`가 32bit입니다. 주소 유형을 생략한 `uart[8250],0x<addr>[,options]`는 `mmio`와 같습니다. options는 `console=ttyS<n>` 형식이며 생략하면 hardware를 초기화하지 않습니다. `uartclk`를 생략하면 `BASE_BAUD * 16`을 사용합니다.
pl011,<addr> | pl011,mmio32,<addr>PL011이미 설정된 PL011 port에서 polling console을 시작합니다. 아직 options는 지원하지 않으며 `mmio32`는 device register 읽기·쓰기에 32bit accessor만 사용하게 합니다.
liteuart,<addr>LiteX이미 설정·구성된 LiteX serial port를 초기 콘솔로 사용합니다. options는 아직 지원하지 않습니다.
meson,<addr>Meson이미 설정·구성된 Meson serial port에서 polling console을 시작합니다. options는 아직 지원하지 않습니다.
msm_serial,<addr>Qualcomm MSM이미 설정·구성된 MSM serial port에서 polling console을 시작합니다. options는 아직 지원하지 않습니다.
msm_serial_dm,<addr>Qualcomm MSM DM이미 설정·구성된 MSM serial DM port에서 polling console을 시작합니다. options는 아직 지원하지 않습니다.
owl,<addr>Actions Semi S500·S900지정 주소의 Actions Semi SoC serial port를 사용합니다. port가 이미 구성되어 있어야 하며 options는 아직 지원하지 않습니다.
rda,<addr>RDA Micro RDA8810PL 등지정 주소의 RDA Micro SoC serial port를 사용합니다. port가 이미 구성되어 있어야 하며 options는 아직 지원하지 않습니다.
sbiRISC-VRISC-V SBI(Supervisor Binary Interface)를 초기 콘솔로 사용합니다.
smhARMARM semihosting 호출을 초기 콘솔로 사용합니다.
s3c2410|s3c2412|s3c2440|s3c6400|s5pv210|exynos4210,<addr>Samsung SoCSamsung SoC serial driver의 초기 콘솔을 사용합니다. 올바른 유형과 선택한 UART port의 base address가 필요하고 port가 이미 구성되어 있어야 합니다. options는 아직 지원하지 않습니다.
lantiq,<addr>Lantiq lqasc이미 설정·구성된 Lantiq serial port에서 polling console을 시작합니다. options는 아직 지원하지 않습니다.
lpuart,<addr> | lpuart32,<addr>Freescale Vybrid·QorIQ LS1021AFreescale LP UART driver의 초기 콘솔을 사용합니다. 유효한 base address가 필요하며 port가 이미 설정·구성되어 있어야 합니다.
ec_imx21,<addr> | ec_imx6q,<addr>Freescale i.MX지정 주소의 i.MX UART에서 polling 방식의 출력 전용 초기 콘솔을 시작합니다. UART가 이미 설정·구성되어 있어야 합니다.
ar3700_uart,<addr>Armada 3700이미 구성된 Armada 3700 serial port에서 polling console을 시작합니다. options는 아직 지원하지 않습니다.
qcom_geni,<addr>Qualcomm GENI이미 구성된 Qualcomm Generic Interface 기반 serial port에서 polling console을 시작합니다. options는 아직 지원하지 않습니다.
efifb,[options]EFI framebuffer사용 가능한 EFI memory-mapped framebuffer에서 가속 없는 초기 콘솔을 시작합니다. system memory를 framebuffer로 쓰는 cache-coherent 비-x86 시스템은 올바른 속성으로 mapping하도록 `ram` option을 전달합니다.
linflex,<addr>NXP S32V234Freescale LINFlexD UART driver의 초기 콘솔을 사용합니다. 유효한 base address가 필요하며 port가 이미 설정·구성되어 있어야 합니다.

earlyprintk

1453-1503
			earlyprintk=vga
			earlyprintk=sclp
			earlyprintk=xen
			earlyprintk=serial[,ttySn[,baudrate]]
			earlyprintk=serial[,0x...[,baudrate]]
			earlyprintk=ttySn[,baudrate]
			earlyprintk=dbgp[debugController#]
			earlyprintk=mmio32,membase[,{nocfg|baudrate}]
			earlyprintk=pciserial[,force],bus:device.function[,{nocfg|baudrate}]
			earlyprintk=xdbc[xhciController#]
			earlyprintk=bios

`earlyprintk=`는 정규 console 초기화 전에 kernel이 crash할 때 유용합니다. 화면 표시상 문제가 있어 기본으로 켜지지 않습니다. `nocfg`는 BIOS·firmware가 UART를 올바르게 구성했다고 보고 UART 설정을 건너뛰며, `,keep`을 붙이면 실제 console이 인계한 뒤에도 끄지 않습니다. VGA, serial, USB debug port 가운데 한 번에 하나만 쓸 수 있습니다.

이름으로 지정할 수 있는 것은 현재 `ttyS0`과 `ttyS1`뿐입니다. x86·ARM 등에서는 `ttySn` 대신 I/O port 주소를 직접 줄 수 있고, 장치 port는 `/proc/tty/driver/serial`에서 찾습니다.

				earlyprintk=serial,0x1008,115200
			You can find the port for a given device in
			/proc/tty/driver/serial:
				2: uart:ST16650V2 port:00001008 irq:18 ...

표준 serial driver와의 연동은 좋지 않으며 VGA 출력은 결국 실제 console이 덮어씁니다. `xen`은 Xen domain, `sclp`는 s390, `bios`는 SuperH에서만 사용할 수 있습니다. `pciserial`의 선택적 `force`는 PCI classcode가 UART가 아니어도 장치를 사용하게 합니다.

매개변수적용·형식한국어 설명
earlyprintk=[X86,SH,ARM,M68k,S390,UM,EARLY]정규 console 이전의 조기 printk 출력 경로를 선택합니다. 지원 문법은 위 원문 코드 블록과 같습니다.

EDAC와 EFI

1504-1548
매개변수적용·형식한국어 설명
edac_report=[HW,EDAC] `{ on | off | force }`, 기본 `on``on`은 EDAC가 hardware event를 보고하게 하지만 우선순위가 높은 다른 error reporting module이 덮어쓸 수 있습니다. `off`는 EDAC 보고를 끄고 `force`는 EDAC 사용을 강제합니다.
edd=[EDD] `{ off | on | skip[mbr] }`Enhanced Disk Drive 정보 처리 방식을 선택합니다.
efi=[EFI,EARLY]`debug`는 기타 debug 출력을 켭니다. `disable_early_pci_dma`는 EFI boot stub에서 모든 PCI bridge의 busmaster bit를 끕니다. `nochunk`는 일부 firmware와 충돌할 수 있는 chunk 단위 파일 읽기를 끕니다. `noruntime`은 EFI runtime service 지원을 끕니다. `nosoftreserve`는 `EFI_MEMORY_SP` memory range 예약을 막아 base type, 즉 `EFI_CONVENTIONAL_MEMORY` 또는 System RAM으로 취급합니다. `novamap`은 `SetVirtualAddressMap()`을 호출하지 않습니다. `no_disable_early_pci_dma`는 boot stub에서 PCI bridge busmaster bit를 유지합니다.
efi_no_storage_paranoia[EFI,X86,EARLY]EFI variable storage의 50%를 넘게 사용할 수 있게 합니다. UEFI의 garbage collection이 정상이고 규격을 지킨다는 확신이 있을 때만 써야 하며, 그렇지 않으면 board를 사용할 수 없게 만들 수 있습니다.
efivar_ssdt=[EFI,X86] EFI variable 이름Linux가 동적으로 load할 SSDT를 담은 EFI variable 이름입니다. 같은 이름에 vendor GUID가 다른 variable이 여러 개면 모두 load합니다. `Documentation/admin-guide/acpi/ssdt-overlays.rst`를 참조하십시오.

EISA, KGDB, SELinux와 EVM

1549-1605
매개변수적용·형식한국어 설명
eisa_irq_edge=[PARISC,HW]`drivers/parisc/eisa.c`의 머리말을 참조하십시오.
ekgdboc=[X86,KGDB,EARLY] `ekgdboc=kbd``earlyprintk=vga` boot argument와 함께 쓰는 초기 kernel console debugging입니다. `kgdboc` 대신 동작하지만 backing tty가 부팅 매우 초기에 있어야 합니다. serial port를 통한 초기 debugging에는 `kgdboc_earlycon`을 사용합니다.
elanfreq=[X86-32]`arch/x86/kernel/cpu/cpufreq/elanfreq.c`의 `elanfreq_setup()` 앞 주석을 참조하십시오.
elfcorehdr=[size[KMG]@]offset[KMG][PPC,SH,X86,S390,EARLY]kernel core image ELF header가 시작하는 물리 주소와 선택적 크기를 지정합니다. 보통 kexec loader가 capture kernel에 전달합니다. `Documentation/admin-guide/kdump/kdump.rst`를 참조하십시오.
enable_mtrr_cleanup[X86,EARLY]X server driver가 나중에 WB entry를 추가할 수 있도록 MTRR 배치를 연속형에서 discrete 형태로 조정하는 기능을 켭니다.
enable_timer_pin_1[X86]APIC timer의 PIN 1을 켭니다. 특히 일부 ATI chipset bug 우회에 유용할 수 있으며 kernel이 합리적인 기본값을 고릅니다.
enforcing=[SELINUX] `{ 0 | 1 }`, 기본 `0`초기 SELinux enforcing 상태를 정합니다. 0은 permissive로 거부 없이 기록만 하고, 1은 enforcing으로 거부하고 기록합니다. 실행 중 `/sys/fs/selinux/enforce`에서 바꿀 수 있습니다. `security/selinux/Kconfig` 도움말도 참조하십시오.
erst_disable[ACPI]Error Record Serialization Table(ERST) 지원을 끕니다.
ether=[HW,NET]Ethernet card parameter입니다. 같은 용법의 `netdev=`로 대체되었으므로 해당 설명을 참조하십시오.
evm=[EVM] `{ fix }`현재 integrity 상태와 관계없이 `security.evm`을 갱신할 수 있게 합니다.

초기 page 확장, fault injection과 FRED

1606-1643
매개변수적용·형식한국어 설명
early_page_ext[KNL,EARLY], `CONFIG_PAGE_EXTENSION=y`더 많은 초기 boot allocation을 포괄하도록 `page_ext` 초기화를 앞 단계로 당깁니다. parallel memory initialization 같은 최적화가 꺼질 수 있어, 특히 memory가 많은 시스템에서는 부팅이 더 오래 걸릴 수 있습니다.
failslab=[KNL] `<interval>,<probability>,<space>,<times>`slab allocation용 일반 fault injection mechanism입니다. `Documentation/fault-injection/`을 참조하십시오.
fail_usercopy=[KNL] `<interval>,<probability>,<space>,<times>`usercopy용 일반 fault injection mechanism입니다. `Documentation/fault-injection/`을 참조하십시오.
fail_page_alloc=[KNL] `<interval>,<probability>,<space>,<times>`page allocation용 일반 fault injection mechanism입니다. `Documentation/fault-injection/`을 참조하십시오.
fail_skb_realloc=[KNL] `<interval>,<probability>,<space>,<times>`SKB 재할당용 일반 fault injection mechanism입니다. `Documentation/fault-injection/`을 참조하십시오.
fail_make_request=[KNL] `<interval>,<probability>,<space>,<times>`block request용 일반 fault injection mechanism입니다. `Documentation/fault-injection/`을 참조하십시오.
fb_tunnels=[NET] `{ initns | none }``Documentation/admin-guide/sysctl/net.rst`의 `fb_tunnels_only_for_init_ns`를 참조하십시오.
floppy=[HW]`Documentation/admin-guide/blockdev/floppy.rst`를 참조하십시오.
forcepae[X86-32]Physical Address Extension(PAE)을 강제로 켭니다. 많은 Pentium M 시스템이 PAE를 꺼 두었어도 기능상 쓸 수 있을 수 있습니다. 이 매개변수는 kernel을 taint하며 알 수 없는 문제를 일으킬 수 있습니다.
fred=[X86-64] `{ on | off }`, 기본 `off`Flexible Return and Event Delivery를 제어합니다. `on`은 기능이 있을 때 FRED를 켜고 `off`는 끕니다.

부팅 초기 ftrace

1644-1720
매개변수적용·형식한국어 설명
ftrace=[tracer][FTRACE]초기 boot debugging을 위해 지정 tracer를 가능한 한 일찍 설정하고 시작합니다.
ftrace_boot_snapshot[FTRACE], 선택적 instance 이름부팅 때 ftrace ring buffer snapshot을 만들어 `/sys/kernel/tracing/snapshot`에서 읽게 합니다. 사용자 공간 시작 과정에 덮일 수 있는 초기 boot trace를 보존할 때 유용합니다. `trace_instance=`로 만든 instance 이름도 지정할 수 있습니다.
ftrace_dump_on_oops[=2(orig_cpu) | =<instance>][,<instance> | ,<instance>=2(orig_cpu)][FTRACE]oops 때 trace buffer를 dump합니다. 인수가 없으면 모든 CPU의 global buffer를, `2` 또는 `orig_cpu`면 oops를 일으킨 CPU buffer만 dump합니다. instance 이름을 주면 해당 instance를 dump하고, 쉼표로 여러 instance를 나열할 수 있습니다. 각 instance에 `2` 또는 `orig_cpu`를 붙이면 문제 CPU의 buffer만 dump합니다.
ftrace_filter=[function-list][FTRACE] 쉼표 구분 함수 목록부팅 때 function tracer가 추적할 함수를 제한합니다. 실행 중 debugfs tracing directory의 `set_ftrace_filter`로 바꿀 수 있습니다.
ftrace_notrace=[function-list][FTRACE] 쉼표 구분 함수 목록목록의 함수를 추적하지 않습니다. 실행 중 `set_ftrace_notrace`로 바꿀 수 있습니다.
ftrace_graph_filter=[function-list][FTRACE] 쉼표 구분 함수 목록function graph tracer가 추적할 최상위 caller 함수를 제한합니다. 실행 중 `set_graph_function`으로 바꿀 수 있습니다.
ftrace_graph_notrace=[function-list][FTRACE] 쉼표 구분 함수 목록목록의 함수에서 시작하는 추적을 하지 않습니다. 실행 중 `set_graph_notrace`로 바꿀 수 있습니다.
ftrace_graph_max_depth=<uint>[FTRACE], 기본 `0`(제한 없음)function graph tracer가 함수 안으로 추적하는 최대 깊이입니다. 실행 중 tracefs tracing directory의 `max_graph_depth`로 바꿀 수 있습니다.

`foo` instance가 부팅 끝에 snapshot을 만들게 하는 예입니다.

			trace_instance=foo,sched_switch ftrace_boot_snapshot=foo

oops를 일으킨 CPU에 있는 `foo` instance buffer만 dump하는 예입니다.

			ftrace_dump_on_oops=foo=orig_cpu

모든 CPU의 global buffer와 `foo` buffer, 그리고 문제 CPU의 `bar` buffer를 dump하는 예입니다.

			ftrace_dump_on_oops,foo,bar=orig_cpu

입력 장치와 GDS 완화

1758-1791
매개변수적용·형식한국어 설명
gamecon.map[2|3]=[HW,JOY] `<port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5>`parallel port 하나당 최대 5개의 multisystem joystick 또는 NES·SNES·PSX pad를 지원합니다. `Documentation/input/devices/joystick-parport.rst`를 참조하십시오.
gamma=[HW,DRM]DRM gamma 관련 설정입니다.
gart_fix_e820=[X86-64,EARLY] `{ off | on }`, 기본 `on`K8 GART를 위한 e820 수정 기능을 제어합니다.
gather_data_sampling=[X86,INTEL,EARLY] `{ force | off }`Gather Data Sampling(GDS) 완화를 제어합니다. GDS는 권한 없는 speculative access가 vector register에 이전에 저장된 데이터를 읽게 할 수 있는 hardware 취약점입니다. 갱신된 microcode가 기본으로 완화하지만 성능 영향이 있을 수 있습니다. `force`는 microcode 완화가 없는 시스템에서 AVX를 꺼 완화하며, microcode 완화가 있으면 효과가 없습니다. 잘못된 AVX 열거를 하는 사용자 공간이 crash할 수 있습니다. `off`는 GDS 완화를 끕니다.
gbpages[X86]kernel direct mapping에 GB page를 사용합니다.

GCOV, GPIO와 GRCAN

1792-1830
매개변수적용·형식한국어 설명
gcov_persist=[GCOV], 기본 non-zero0이 아니면 kernel module profiling data를 저장해 module unload·reload 뒤에도 debugfs에서 볼 수 있습니다. 0이면 unload 때 profiling data와 관련 debugfs file을 제거합니다.
goldfish[X86]Goldfish Android emulator platform을 켭니다. Android emulator에서 실행 중이 아닐 때는 사용하지 마십시오.
gpio-mockup.gpio_mockup_ranges[HW] `<start1>,<end1>,<start2>,<end2>...`이 device가 만들 gpiochip의 범위를 지정합니다.
gpio-mockup.gpio_mockup_named_lines[HW]GPIO line에 이름을 붙여야 함을 driver에 알립니다.
gpt[EFI]유효한 GPT signature가 있지만 Protective MBR이 잘못된 disk를 GPT로 취급합니다. primary GPT가 손상되면 backup·alternate GPT를 사용하게 합니다.
grcan.enable0=[HW] `{ 0 | 1 }`, 기본 `0`configuration register의 `Enable 0` bit, 즉 physical interface 0 사용 여부를 정합니다.
grcan.enable1=[HW] `{ 0 | 1 }`, 기본 `0`원문 표현대로 configuration register의 `Enable 0` bit로 설명된 physical interface 1 설정입니다.
grcan.select=[HW] `{ 0 | 1 }`, 기본 `0`사용할 physical interface를 선택합니다.
grcan.txsize=[HW] `<unsigned int>`, 기본 `1024``(txsize & ~0x1fffc0) == 0`을 만족하는 tx buffer 크기를 설정합니다.
grcan.rxsize=[HW] `<unsigned int>`, 기본 `1024``(rxsize & ~0x1fffc0) == 0`을 만족하는 rx buffer 크기를 설정합니다.

Usercopy hardening과 pointer hash

1831-1882
매개변수적용·형식한국어 설명
hardened_usercopy=[KNL] `{ on | off }``CONFIG_HARDENED_USERCOPY` 아래에서 이번 부팅의 hardening 사용 여부를 정합니다. `copy_to_user()`·`copy_from_user()`가 알려진 memory allocation 경계를 넘어 읽고 쓰는 것을 검사해 bounds-checking 결함을 방어합니다. 기본은 `CONFIG_HARDENED_USERCOPY_DEFAULT_ON`이 정합니다. `on`은 검사를 수행하고 `off`는 끕니다.
hardlockup_all_cpu_backtrace=[KNL] `{ 0 | 1 }`hard-lockup detector가 모든 CPU의 backtrace를 만들지 정합니다.
hash_pointers=[KNL,EARLY] `{ auto | always | never }`, 기본 `auto``%p` format으로 console·buffer에 출력하는 pointer 값을 hash해 실제 kernel address를 권한 없는 사용자에게 숨깁니다. 서로 다른 pointer 비교가 어려워져 debugging에는 불편할 수 있습니다. `auto`는 `slab_debug`가 켜지지 않았을 때 hash하고, `always`는 항상 hash합니다. `never`는 debugging 때만 hash를 끄며 production kernel에서는 쓰지 않아야 합니다. `no_hash_pointers`는 `never`의 alias입니다.
hashdist=[KNL,NUMA] `{ 0 | 1 }`부팅 중 할당하는 큰 hash를 NUMA node에 분산합니다. 64bit NUMA에서는 기본 on, 그 밖에는 off입니다.
hd=[EIDE] `<cyl>,<head>,<sect>`(E)IDE hard drive subsystem geometry를 지정합니다.
hest_disable[ACPI]Hardware Error Source Table(HEST) 지원과 이에 대응하는 firmware-first mode error 처리 logic을 끕니다.

최대 절전과 HPET

1883-1944
매개변수적용·형식한국어 설명
hibernate=[HIBERNATION] `{ noresume | nocompress | no | protect_image }``noresume`은 부팅 때 hibernation image를 찾지 않습니다. `nocompress`는 image를 압축·해제하지 않습니다. `no`는 hibernation과 resume을 끕니다. `protect_image`는 복원 중 image data를 담은 모든 page를 read-only로 만들어 image 보호를 켭니다.
hibernate.compressor=[HIBERNATION] `{ lzo | lz4 }`, 기본 `lzo`hibernation image 압축·해제 algorithm으로 LZO 또는 LZ4를 선택합니다.
hibernate.pm_test_delay=[HIBERNATION], 초, 기본 `5`hibernation test mode에서 system을 resume하기 전 머무는 시간을 정합니다. `/sys/power/pm_test`를 참조하며 `CONFIG_PM_DEBUG`일 때만 쓸 수 있습니다.
highmem=nn[KMG][KNL,BOOT,EARLY]highmem zone을 정확히 `<nn>` 크기로 강제합니다. 원래 highmem이 없는 시스템에도 만들 수 있고 큰 시스템에서는 highmem 크기를 줄일 수도 있습니다.
highres=[KNL] `{ on | off }`, 기본 `on`고해상도 timer mode를 켜거나 끕니다.
hlt[BUGS=ARM,SH]ARM·SuperH의 HLT 관련 bug 우회 설정입니다.
hostname=[KNL,EARLY] `<string>`초기 시작 때 hostname, 즉 UTS nodename을 설정하며 `gethostname()` 반환값이 됩니다. 사용자 공간 process가 명시적 설정 전에 잘못된 이름을 읽는 일을 막습니다. 보통 64자인 최대 hostname 길이를 넘으면 잘립니다.
hpet=[X86-32,HPET] `{ enable | disable | force | verbose }`, 기본 `enable``disable`은 HPET를 끄고 PIT를 씁니다. `force`는 ICH4·VIA·nVidia 같은 문서화되지 않은 chip에서도 강제로 켭니다. `verbose`는 setup 중 HPET register 내용을 표시합니다.
hpet_mmap=[X86,HPET_MMAP]사용자 공간이 HPET register를 `mmap`하게 합니다. 기본값은 `CONFIG_HPET_MMAP_DEFAULT`가 정합니다.

HugeTLB, hung task와 Hyper-V

1945-2038
매개변수적용·형식한국어 설명
hugepages=[HW,EARLY] `<integer>` 또는 `<node>:<integer>[,<node>:<integer>]`부팅 때 할당할 HugeTLB page 수입니다. `hugepagesz=` 뒤에 오면 그 크기의 page 수를 뜻합니다. command line의 첫 HugeTLB parameter라면 기본 huge page 크기의 수를 뜻합니다. node 형식으로 node별 수를 지정할 수 있습니다. `Documentation/admin-guide/mm/hugetlbpage.rst`를 참조하십시오.
hugepagesz=[HW,EARLY] `size[KMG]``hugepages=`와 함께 특정 크기의 huge page를 부팅 때 할당합니다. 지원하는 huge page 크기마다 `hugepagesz=X hugepages=Y` 쌍을 한 번씩 지정할 수 있으며 크기는 architecture마다 다릅니다.
hugepage_alloc_threads=[HW], 기본 available hardware thread의 25%부팅 중 huge page 할당에 쓸 thread 수입니다. 많은 huge page를 할당할 때 부팅 시간을 줄일 수 있습니다. gigantic huge page에는 적용되지 않습니다.
hugetlb_cma=[HW,CMA,EARLY] `nn[KMGTPE]` 또는 `<node>:nn[KMGTPE][,<node>:nn[KMGTPE]]`gigantic huge page 할당에 쓸 CMA area를 전체 또는 node별로 예약합니다. 켜면 gigantic huge page의 일반적인 boot-time allocation을 건너뛰고 CMA allocator로 할당합니다.
hugetlb_cma_only=[HW,CMA,EARLY]새 HugeTLB page를 할당할 때 CMA area만 시도합니다. `hugetlb_cma=`도 지정해야 효과가 있습니다.
hugetlb_free_vmemmap=[KNL] `{ on | off }`, 기본 `off``CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP`이 필요합니다. HugeTLB Vmemmap Optimization(HVO)을 제어해 2MB hugetlb page마다 `7 * PAGE_SIZE`를 돌려받을 수 있습니다. `CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP_DEFAULT_ON=y`로 빌드하면 기본이 on입니다. `memory_hotplug.memmap_on_memory`가 켜져 추가 memory block 자체에서 할당된 vmemmap page는 최적화할 수 없지만, 그 밖의 vmemmap page에는 영향이 없습니다.
hung_task_panic=[KNL] `{ 0 | 1 }`1이면 hung task를 찾았을 때 kernel panic을 일으킵니다. 기본은 `CONFIG_BOOTPARAM_HUNG_TASK_PANIC`이 정하며, 나중에 `kernel.hung_task_panic` sysctl로 바꿀 수 있습니다.
hvc_iucv=[S390] `0..8`z/VM IUCV hypervisor console(HVC) terminal device 수를 정합니다.
hvc_iucv_allow=[S390] 쉼표 구분 z/VM user ID지정하면 z/VM IUCV HVC가 나열된 user ID의 연결만 받아들입니다.
hv_nopvspin[X86,HYPER_V,EARLY]lock contention 때 hypervisor가 guest를 idle 상태로 둘 수 있게 하는 paravirtual spinlock 최적화를 끕니다.
hw_protection=[HW] `{ reboot | shutdown }`과열이나 임박한 전압 손실 같은 critical event에서 취할 hardware protection action을 정합니다.

I2C bus와 touchscreen property

2039-2066
매개변수적용·형식한국어 설명
i2c_bus=[HW] `<bus_id>,<clkrate>`board별 기본 I2C bus 속도를 덮어쓰거나 board initialization code가 등록하지 않은 I2C bus를 추가로 등록합니다.
i2c_touchscreen_props=[HW,ACPI,X86] `<ACPI_HW_ID>:<prop_name>=<val>[:prop_name=val][:...]`ACPI가 열거한 I2C touchscreen에 device property를 설정해 거꾸로 장착된 panel의 좌표 등을 보정합니다. 값 전체를 생략하면 boolean property, unsigned number면 u32 property, 그 밖의 값이면 string property입니다. 이 옵션이 필요한 장치는 `drivers/platform/x86/touchscreen_dmi.c`에 DMI quirk를 추가하는 patch를 제출해야 합니다.

지원 형식과 여러 줄로 나눈 두 예제를 원문 그대로 보존합니다.

			Where <val> is one of:
			Omit "=<val>" entirely	Set a boolean device-property
			Unsigned number		Set a u32 device-property
			Anything else		Set a string device-property

			Examples (split over multiple lines):
			i2c_touchscreen_props=GDIX1001:touchscreen-inverted-x:
			touchscreen-inverted-y

			i2c_touchscreen_props=MSSL1680:touchscreen-size-x=1920:
			touchscreen-size-y=1080:touchscreen-inverted-y:
			firmware-name=gsl1680-vendor-model.fw:silead,home-button

i8042 keyboard·mouse controller

2067-2099
매개변수적용·형식한국어 설명
i8042.debug[HW]i8042 debug mode를 전환합니다.
i8042.unmask_kbd_data[HW], 기본 offKBD port interrupt data 출력을 켭니다. 선행 조건으로 `i8042.debug=1`이 필요합니다.
i8042.direct[HW]keyboard port를 scan code 변환이 없는 mode로 둡니다.
i8042.dumbkbd[HW]controller가 keyboard data만 읽고 상태를 제어하지 못한다고 가정해 LED 점멸 같은 제어를 시도하지 않습니다.
i8042.noaux[HW]보조 장치, 즉 mouse port를 검사하지 않습니다.
i8042.nokbd[HW]keyboard port를 검사하거나 만들지 않습니다.
i8042.noloop[HW]AUX port probe 중 AUX Loopback command를 사용하지 않습니다.
i8042.nomux[HW]활성 multiplexing controller가 있는지 검사하지 않습니다.
i8042.nopnp[HW]KBD·AUX controller 발견에 ACPIPnP 또는 PnPBIOS를 사용하지 않습니다.
i8042.notimeout[HW]controller가 알린 timeout 상태를 무시합니다.
i8042.reset[HW] `{ 1 | Y | y | 0 | N | n }`init·cleanup·suspend-to-RAM 전환 때 controller reset 방식을 정합니다. 1·Y·y는 항상 reset하고 0·N·n은 절대 reset하지 않습니다. x86 기본값은 s2r 전환 때만 reset하며, 다른 대부분의 architecture는 항상 reset을 강제합니다.
i8042.unlock[HW]keylock을 무시해 잠금을 해제합니다.
i8042.kbdreset[HW]KBD port에 연결된 장치를 reset합니다.
i8042.probe_defer[HW]i8042 probe error가 발생하면 deferred probing을 허용합니다.
i810=[HW,DRM]Intel i810 DRM 관련 설정입니다.

i915 밝기, IA32와 CPU idle

2100-2142
매개변수적용·형식한국어 설명
i915.invert_brightness=[DRM] `{ -1 | 0 | 1 }`, 기본 `0`panel backlight 밝기 값의 의미를 뒤집습니다. 보통 0은 꺼짐, 최대값은 최대 밝기지만 반전 시 0이 최대 밝기이고 최대값이 꺼짐입니다. -1은 절대 반전하지 않고, 0은 machine 기본 동작, 1은 반전을 강제합니다.
ia32_emulation=[X86-64] `<bool>`true면 `IA32_EMULATION_DEFAULT_DISABLED`를 부팅 때 덮어써 32bit program load와 32bit syscall 실행을 허용합니다. false면 IA32 emulation을 조건 없이 끕니다.
icn=[HW,ISDN] `<io>[,<membase>[,<icn_id>[,<icn_id2>]]]`ICN ISDN adapter의 I/O, memory base와 card ID를 지정합니다.
idle=[X86,EARLY] `{ idle=poll | idle=halt | idle=nomwait }``idle=poll`은 idle loop에서 HLT 절전 대신 reschedule event를 polling합니다. 전력은 훨씬 많이 들지만 multiprocessor benchmark 성능이 약간 좋아지거나 performance counter profiling이 더 정확해질 수 있습니다. MONITOR/MWAIT를 지원하는 Intel EM64T 같은 시스템에서는 정상 idle loop보다 이점이 없고 hyperthreading과 나쁘게 상호작용할 수 있습니다. `idle=halt`는 CPU idle에 halt를 강제해 C2·C3를 쓰지 못하게 합니다. `idle=nomwait`는 CPU C-state에서 mwait를 끕니다.

IDXD와 MIPS IEEE 754

2143-2190
매개변수적용·형식한국어 설명
idxd.sva=[HW] `<bool>`, 기본 `1`idxd driver의 Shared Virtual Memory(SVA) 지원을 강제로 끌 수 있게 합니다.
idxd.tc_override=[HW] `<bool>`, 기본 `0`장치의 기본 traffic class 구성을 덮어쓸 수 있게 합니다.
ieee754=[MIPS] `{ strict | legacy | 2008 | relaxed | emulated }`, 기본 `strict`FPU가 지원하는 IEEE 754 NaN encoding과 각 binary의 ELF header flag가 요구하는 encoding을 기준으로 실행할 program을 고릅니다. hardware는 legacy와 2008 NaN mode 중 하나 또는 둘 다 지원할 수 있습니다. `strict`는 FPU가 지원하는 encoding을 요구한 binary만, `legacy`와 `2008`은 해당 mode binary만 받아들입니다. `relaxed`는 FPU 지원 여부와 상관없이 모두 받아들이고, `emulated`도 모두 받아들이되 지원하지 않는 mode는 FPU emulator를 켭니다. emulator는 두 encoding을 모두 지원합니다. FPU가 없거나 `nofpu`로 껐을 때 `legacy`·`2008`은 emulator를 해당 mode로 고정하고 `relaxed`는 둘 다 허용합니다. `strict`는 legacy processor에서 legacy NaN만, MIPS32·MIPS64에서 둘 다 허용합니다. `ABS.fmt`·`NEG.fmt` instruction mode도 hardware가 지원하지 않는 경우를 빼면 대체로 NaN mode를 따릅니다.

Loglevel, inode hash와 IMA

2191-2288
매개변수적용·형식한국어 설명
ignore_loglevel[KNL,EARLY]loglevel 설정을 무시하고 모든 kernel message를 console에 출력합니다. debugging에 유용하며 printk module parameter로도 제공되어 `/sys/module/printk/parameters/ignore_loglevel`에서 동적으로 바꿀 수 있습니다.
ignore_rlimit_data[KNL]data mapping의 `RLIMIT_DATA`를 무시하고 처음 잘못 쓸 때 warning을 냅니다. `/sys/module/kernel/parameters/ignore_rlimit_data`에서 바꿀 수 있습니다.
ihash_entries=[KNL]inode cache의 hash bucket 수를 설정합니다.
ima_appraise=[IMA] `{ off | enforce | fix | log }`, 기본 `enforce`integrity measurement appraisal 동작을 정합니다.
ima_appraise_tcb[IMA], 폐기됨내장 appraisal policy로 uid 0이 소유한 모든 file을 평가합니다. 대신 `ima_policy=`를 사용하십시오.
ima_canonical_fmt[IMA]binary runtime measurement에 host native 형식 대신 canonical 형식을 사용합니다.
ima_hash=[IMA] `{ md5 | sha1 | rmd160 | sha256 | sha384 | sha512 | ... }`, 기본 `sha1`IMA hash algorithm을 선택합니다. 지원 목록은 `crypto/hash_info.h`에 정의됩니다.
ima_policy=[IMA] `{ tcb | appraise_tcb | secure_boot | fail_securely | critical_data }`IMA setup 중 load할 내장 policy입니다. `tcb`는 effective uid 또는 uid가 0인 사용자가 실행한 program, 실행용으로 mmap한 file, read mode로 연 file을 측정합니다. `appraise_tcb`는 root 소유 file의 integrity를 평가합니다. `secure_boot`는 file signature를 바탕으로 kexec kernel image, module, firmware, policy 등의 integrity를 평가합니다. `fail_securely`는 `SB_I_UNVERIFIABLE_SIGNATURE` flag가 있는 privileged mounted filesystem에서도 signature verification failure를 강제합니다. `critical_data`는 kernel integrity에 중요한 data를 측정합니다.
ima_tcb[IMA], 폐기됨Trusted Computing Base 요구에 맞는 policy를 load해 uid 0이 실행한 program, 실행용 mmap file과 read로 연 file을 측정합니다. 대신 `ima_policy=`를 사용하십시오.
ima_template=[IMA] `{ ima | ima-ng | ima-ngv2 | ima-sig | ima-sigv2 }`, 기본 `ima-ng`정의된 IMA measurement template 형식 하나를 선택합니다.
ima_template_fmt=[IMA] `field1|...|fieldN`사용자 정의 template 형식을 정합니다.
ima.ahash_minsize=[IMA] `<min_file_size>`asynchronous hash를 사용할 최소 file 크기입니다. 생략하면 ahash를 쓰지 않습니다. accelerator마다 data 크기에 따른 성능이 달라 특정 hardware에 맞게 조정할 수 있습니다.
ima.ahash_bufsize=[IMA] `<bufsize>`, 기본 `4k`asynchronous hash buffer 크기입니다. accelerator마다 chunk 크기에 따른 성능이 달라 특정 hardware에 맞게 조정할 수 있습니다.
ima=[IMA] `{ off | on }`, 기본 `on`IMA를 켜거나 끕니다. IMA 비활성화는 kdump kernel에서만 허용됩니다.

ITS 완화와 초기화

2289-2358
매개변수적용·형식한국어 설명
indirect_target_selection=[X86,Intel] `{ on | off | force | vmexit | stuff }`Intel CPU의 Indirect Target Selection(ITS) bug 완화를 제어합니다. IBPB 수정에는 갱신된 microcode도 필요합니다. `on`은 기본 완화, `off`는 비활성화, `force`는 ITS bug가 있다고 강제해 기본 완화를 배치합니다. `vmexit`은 CPU가 ITS의 guest·host isolation 부분에 영향받을 때만 완화합니다. `stuff`는 retpoline도 사용 중이면 RSB-fill을, 아니면 기본 완화를 배치합니다. `Documentation/admin-guide/hw-vuln/indirect-target-selection.rst`를 참조하십시오.
init=[KNL] `<full_path>`init process로 `/sbin/init` 대신 지정 binary를 실행합니다.
initcall_debug[KNL]initcall 실행을 trace합니다. startup 중 kernel이 멈추는 위치를 찾을 때 유용합니다.
initcall_blacklist=[KNL] 쉼표 구분 initcall 함수 목록나열한 initcall을 실행하지 않습니다. built-in module과 initcall debugging에 유용합니다.
initramfs_async=[KNL] `<bool>`, 기본 `1`device probe·initialization과 동시에 initramfs image를 비동기로 풀지 정합니다. 일반적으로 그대로 동작하지만 debugging을 위해 device initcall과 late initcall 전에 unpack을 끝내던 과거 동작으로 돌릴 수 있습니다.
initrd=[BOOT,EARLY]initial ramdisk 위치를 지정합니다.
initrdmem=[KNL,EARLY] `ss[KMG],nn[KMG]`, 기본 `0,0`initrd를 load할 물리 주소와 크기를 지정합니다. built-in initrd 또는 boot parameter로 지정한 initrd가 있으면 그것이 우선합니다.
init_on_alloc=[MM,EARLY] `{ 0 | 1 }`새로 할당한 page와 heap object를 0으로 채웁니다. 기본은 `CONFIG_INIT_ON_ALLOC_DEFAULT_ON`이 정합니다.
init_on_free=[MM,EARLY] `{ 0 | 1 }`해제한 page와 heap object를 0으로 채웁니다. 기본은 `CONFIG_INIT_ON_FREE_DEFAULT_ON`이 정합니다.
init_pkru=[X86], 기본 `0x55555554`모든 process의 기본 memory protection key rights register 값을 지정합니다. 기본은 pkey 0 이외의 접근을 막으며 부팅 뒤 debugfs에서 덮어쓸 수 있습니다.
inport.irq=[HW] `<irq>`Inport(ATI XL·Microsoft) busmouse driver의 IRQ를 지정합니다.
int_pln_enable[X86]power limit notification interrupt를 켭니다.

Integrity audit와 Intel IOMMU

2359-2396
매개변수적용·형식한국어 설명
integrity_audit=[IMA] `{ 0 | 1 }`, 기본 `0`0은 기본 integrity audit message, 1은 추가 integrity audit message를 출력합니다.
intel_iommu=[DMAR]`on`은 Intel IOMMU driver를 켜고 `off`는 끕니다. `igfx_off`는 전용 DMAR unit이 있는 graphic device의 DMAR을 우회해 DMA에 물리 주소를 쓰게 합니다. 폐기된 `strict`는 `iommu.strict=1`과 같습니다. `sp_off`는 super page 지원을 끕니다. `sm_on`은 hardware가 광고하는 scalable mode translation을 켜고 `sm_off`는 scalable mode를 금지합니다. `tboot_noforce`는 tboot 아래에서 Intel IOMMU 강제 활성화를 막아 고처리량 장치의 성능 저하를 피할 수 있지만 DMA attack 방어를 낮춥니다.

Intel IOMMU의 전체 option spelling과 기본값을 원문 그대로 보존합니다.

		on
			Enable intel iommu driver.
		off
			Disable intel iommu driver.
		igfx_off [Default Off]
			By default, gfx is mapped as normal device. If a gfx
			device has a dedicated DMAR unit, the DMAR unit is
			bypassed by not enabling DMAR with this option. In
			this case, gfx device will use physical address for
			DMA.
		strict [Default Off]
			Deprecated, equivalent to iommu.strict=1.
		sp_off [Default Off]
			By default, super page will be supported if Intel IOMMU
			has the capability. With this option, super page will
			not be supported.
		sm_on
			Enable the Intel IOMMU scalable mode if the hardware
			advertises that it has support for the scalable mode
			translation.
		sm_off
			Disallow use of the Intel IOMMU scalable mode.
		tboot_noforce [Default Off]
			Do not force the Intel IOMMU enabled under tboot.
			By default, tboot will force Intel IOMMU on, which
			could harm performance of some high-throughput
			devices like 40GBit network cards, even if identity
			mapping is enabled.
			Note that using this option lowers the security
			provided by tboot because it makes the system
			vulnerable to DMA attacks.

Intel idle·P-state와 interrupt remapping

2397-2457
매개변수적용·형식한국어 설명
intel_idle.max_cstate=[KNL,HW,ACPI,X86] `0..9`0은 `intel_idle`을 끄고 `acpi_idle`로 돌아갑니다. 1부터 9는 최대 C-state 깊이를 지정합니다.
intel_pstate=[X86,EARLY]`disable`은 지원 processor의 기본 scaling driver로 쓰지 않습니다. `active`는 generic cpufreq governor layer를 우회하고 HWP 지원 여부와 model에 따라 내부 powersave·performance algorithm으로 P-state를 고릅니다. `passive`는 generic cpufreq governor와 함께 scaling driver로 쓰며 HWP와 동시에 쓸 수 없습니다. `force`는 기본적으로 acpi-cpufreq를 선호하는 시스템에서도 강제하지만 ACPI P-state 정보에 의존하는 thermal control·power capping을 끌 수 있어 주의해야 하며 미지원 CPU 또는 `pcc-cpufreq` platform에서는 동작하지 않습니다. `no_hwp`는 HWP를 끄고 `hwp_only`는 HWP 지원 시스템에서만 driver를 load합니다. `support_acpi_ppc`는 ACPI `_PPC` limit을 강제하며 Enterprise Server·Performance Server profile에서는 기본으로 켜집니다. `per_cpu_perf_limits`는 cpufreq sysfs를 통한 logical CPU별 P-state limit을 허용합니다. `no_cas`는 hybrid system의 capacity-aware scheduling을 끕니다.
intremap=[X86-64,Intel-IOMMU,EARLY]`on`은 Interrupt Remapping을 켜고 기본값이며 `off`는 끕니다. `nosid`는 Source ID 검사를 끄고 `no_x2apic_optout`은 BIOS의 x2APIC opt-out 요청을 무시합니다. `nopost`는 Interrupt Posting을 끄고 `posted_msi`는 MSI를 posted interrupt로 전달합니다.
iomem=`{ strict | relaxed }`사용자 공간의 MMIO memory 접근에 대한 strict 검사를 제어합니다.
			disable
			  Do not enable intel_pstate as the default
			  scaling driver for the supported processors
                        active
                          Use intel_pstate driver to bypass the scaling
                          governors layer of cpufreq and provides it own
                          algorithms for p-state selection. There are two
                          P-state selection algorithms provided by
                          intel_pstate in the active mode: powersave and
                          performance.  The way they both operate depends
                          on whether or not the hardware managed P-states
                          (HWP) feature has been enabled in the processor
                          and possibly on the processor model.
			passive
			  Use intel_pstate as a scaling driver, but configure it
			  to work with generic cpufreq governors (instead of
			  enabling its internal governor).  This mode cannot be
			  used along with the hardware-managed P-states (HWP)
			  feature.
			force
			  Enable intel_pstate on systems that prohibit it by default
			  in favor of acpi-cpufreq. Forcing the intel_pstate driver
			  instead of acpi-cpufreq may disable platform features, such
			  as thermal controls and power capping, that rely on ACPI
			  P-States information being indicated to OSPM and therefore
			  should be used with caution. This option does not work with
			  processors that aren't supported by the intel_pstate driver
			  or on platforms that use pcc-cpufreq instead of acpi-cpufreq.
			no_hwp
			  Do not enable hardware P state control (HWP)
			  if available.
			hwp_only
			  Only load intel_pstate on systems which support
			  hardware P state control (HWP) if available.
			support_acpi_ppc
			  Enforce ACPI _PPC performance limits. If the Fixed ACPI
			  Description Table, specifies preferred power management
			  profile as "Enterprise Server" or "Performance Server",
			  then this feature is turned on by default.
			per_cpu_perf_limits
			  Allow per-logical-CPU P-State performance control limits using
			  cpufreq sysfs interface
			no_cas
			  Do not enable capacity-aware scheduling (CAS) on
			  hybrid systems
			on	enable Interrupt Remapping (default)
			off	disable Interrupt Remapping
			nosid	disable Source ID checking
			no_x2apic_optout
				BIOS x2APIC opt-out request will be ignored
			nopost	disable Interrupt Posting
			posted_msi
				enable MSIs delivered as posted interrupts

공통·legacy IOMMU 설정

2458-2559
매개변수적용·형식한국어 설명
iommu=[X86,EARLY]`off`는 모든 IOMMU 초기화·사용을 막습니다. `force`는 3GB 미만 memory처럼 필요하지 않아도 hardware IOMMU를 강제하고 `noforce`는 필요 없을 때 강제하지 않는 기본값입니다. `soft`는 hardware IOMMU 대신 software bounce buffering인 SWIOTLB를 사용하며 Intel machine 기본입니다. `nobypass`는 PCI device가 IOMMU를 우회하지 못하게 합니다. 그 밖에 `biomerge`, `panic`, `nopanic`, `merge`, `nomerge`, x86 `pt`, PPC·POWERNV `nopt`가 있습니다. AMD GART 전용으로 `<size>`는 remapping area byte 크기, `allowed`는 특정 chipset의 IOMMU-off workaround를 덮어씁니다. `fullflush`는 allocation마다 flush하는 기본값, `nofullflush`는 이를 끕니다. `memaper[=<order>]`는 RAM 위에 `32MB << order` 크기 aperture를 만들며 기본 order 1은 64MB입니다. `merge`는 experimental SG merge와 `force`를 함께 켜고 `nomerge`는 SG merge를 끕니다. `noaperture`는 AGP aperture를 건드리지 않게 하고 `noagp`는 AGP driver 초기화를 막고 full aperture를 씁니다. `panic`은 IOMMU overflow 때 항상 panic합니다.
iommu.forcedac=[ARM64,X86,EARLY] `{ 0 | 1 }`PCI device IOVA 할당을 제어합니다. 0은 먼저 32bit DMA 주소를 시도한 뒤 필요하면 전체 range로 fallback합니다. 1은 전체 usable range에서 바로 할당해 32bit 초과 addressing card에 Dual Address Cycle을 강제합니다.
iommu.strict=[ARM64,X86,S390,EARLY] `{ 0 | 1 }`0은 DMA unmap의 hardware TLB invalidation을 미뤄 isolation을 일부 낮추는 대신 throughput을 높이는 lazy mode이며 driver가 지원하지 않으면 strict로 돌아갑니다. 1은 unmap 때 IOMMU hardware TLB를 동기적으로 invalidate하는 strict mode입니다. 생략하면 `CONFIG_IOMMU_DEFAULT_DMA_{LAZY,STRICT}`를 따릅니다. x86에서는 legacy driver별 strict option이 우선합니다.
iommu.passthrough=[ARM64,X86,EARLY] `{ 0 | 1 }`0은 DMA에 IOMMU translation을 쓰고 1은 기본적으로 IOMMU를 우회합니다. 생략하면 `CONFIG_IOMMU_DEFAULT_PASSTHROUGH`를 따릅니다.

legacy `iommu=`의 architecture별 option과 AMD GART 문법은 원문 그대로 보존합니다.

		off
			Don't initialize and use any kind of IOMMU.

		force
			Force the use of the hardware IOMMU even when
			it is not actually needed (e.g. because < 3 GB
			memory).

		noforce
			Don't force hardware IOMMU usage when it is not
			needed. (default).

		biomerge
		panic
		nopanic
		merge
		nomerge

		soft
			Use software bounce buffering (SWIOTLB) (default for
			Intel machines). This can be used to prevent the usage
			of an available hardware IOMMU.

			[X86]
		pt
			[X86]
		nopt
			[PPC/POWERNV]
		nobypass
			Disable IOMMU bypass, using IOMMU for PCI devices.

		[X86]
		AMD Gart HW IOMMU-specific options:

		<size>
			Set the size of the remapping area in bytes.

		allowed
			Overwrite iommu off workarounds for specific chipsets

		fullflush
			Flush IOMMU on each allocation (default).

		nofullflush
			Don't use IOMMU fullflush.

		memaper[=<order>]
			Allocate an own aperture over RAM with size
			32MB<<order.  (default: order=1, i.e. 64MB)

		merge
			Do scatter-gather (SG) merging. Implies "force"
			(experimental).

		nomerge
			Don't do scatter-gather (SG) merging.

		noaperture
			Ask the IOMMU not to touch the aperture for AGP.

		noagp
			Don't initialize the AGP driver and use full aperture.

		panic
			Always panic when IOMMU overflows.

I/O, IPE와 interrupt 처리

2560-2640
매개변수적용·형식한국어 설명
io7=[HW] Marvel 기반 Alpha`arch/alpha/kernel/core_marvel.c`의 `marvel_specify_io7` 앞 주석을 참조하십시오.
io_delay=[X86,EARLY] `{ 0x80 | 0xed | udelay | none }`I/O delay 방식을 고릅니다. `0x80`은 표준 port 0x80, `0xed`는 일부 시스템에 필요한 대체 port, `udelay`는 단순 2us 지연, `none`은 지연 없음입니다.
ip=[IP_PNP]`Documentation/admin-guide/nfs/nfsroot.rst`를 참조하십시오.
ipcmni_extend[KNL,EARLY]고유 System V IPC identifier 최대 수를 32,768에서 16,777,216으로 늘립니다.
ipe.enforce=[IPE] `<bool>`, 기본 `1`IPE를 permissive(0) 또는 enforce(1) mode로 시작할지 정합니다.
ipe.success_audit=[IPE] `<bool>`, 기본 `0`허용된 binary마다 audit event를 내는 success auditing을 켠 상태로 IPE를 시작합니다.
irqaffinity=[SMP] CPU 목록기본 IRQ affinity mask를 앞서 설명한 CPU list 형식으로 지정합니다.
irqchip.gicv2_force_probe=[ARM,ARM64,EARLY] `<bool>`device tree가 노출한 memory range가 너무 작아도 GICv2 controller의 두 번째 4KB page를 강제로 찾습니다.
irqchip.gicv3_nolpi=[ARM,ARM64,EARLY]LPI와 그에 따른 ITS가 있음을 무시합니다. kernel을 bootloader로 써 secondary kernel이 LPI를 설정하게 하는 시스템용입니다.
irqchip.gicv3_pseudo_nmi=[ARM64,EARLY], `CONFIG_ARM64_PSEUDO_NMI`kernel의 pseudo-NMI 지원을 켭니다.
irqchip.riscv_imsic_noipi[RISC-V,EARLY]IMSIC software-injected MSI를 IPI로 쓰지 않습니다. IMSIC가 trap-and-emulate되는 시스템에서 여러 hart에 IPI를 보낼 때 MMIO trap을 줄이려는 옵션입니다.
irqfixup[HW]처리되지 않은 interrupt가 생기면 모든 handler를 검색합니다. 심하게 잘못된 firmware 시스템을 작동시키기 위한 옵션입니다.
irqhandler.duration_warn_us=[KNL] microsecondIRQ handler가 지정 duration을 넘으면 warning을 내 long-running IRQ를 찾게 합니다.
irqpoll[HW]처리되지 않은 interrupt 때 모든 handler를 검색하고 timer interrupt마다 모든 handler도 검사합니다. 심하게 잘못된 firmware 시스템용입니다.
isapnp=[ISAPNP] `<RDP>,<reset>,<pci_scan>,<verbosity>`ISA Plug and Play probe 설정을 지정합니다.

I/O delay backend 이름과 설명을 원문 그대로 보존합니다.

		0x80
			Standard port 0x80 based delay
		0xed
			Alternate port 0xed based delay (needed on some systems)
		udelay
			Simple two microseconds delay
		none
			No delay

CPU 격리

2641-2705
매개변수적용·형식한국어 설명
isolcpus=[KNL,SMP,ISOL] `[flag-list,]<cpu-list>`, 폐기 예정지정 CPU를 선택한 disturbance에서 격리하며 기본 flag는 `domain`입니다. 새 구성은 cpuset 사용을 권장합니다. `nohz`는 단일 task 실행 때 tick을 끄고 RCU callback offload 같은 kernel noise도 줄여 `nohz_full`과 같은 효과를 냅니다. 남은 1Hz tick은 workqueue로 offload되므로 global workqueue affinity를 `/sys/devices/virtual/workqueue/cpumask` 또는 `domain` flag로 housekeeping CPU에 묶어야 합니다. 기본 global workqueue는 모든 CPU에서 실행되므로 개별 CPU 보호에는 부팅 뒤 cpumask 설정이 필요합니다. `domain`은 일반 SMP balancing·scheduling domain에서 격리하며 되돌릴 수 없습니다. 대신 CPU를 언제든 옮길 수 있는 cpuset의 `cpuset.sched_load_balance` 사용을 강하게 권장합니다. process는 CPU affinity syscall 또는 cpuset으로 isolated CPU에 넣거나 뺄 수 있고 CPU 번호는 0부터 전체 CPU 수보다 1 작은 값까지입니다. `managed_irq`는 isolated CPU를 포함하는 mask의 managed interrupt target에서 격리합니다. managed affinity는 kernel이 관리해 `/proc/irq/*`로 바꿀 수 없습니다. device queue mask에 isolated CPU와 housekeeping CPU가 함께 있고 housekeeping CPU가 online일 때 interrupt를 housekeeping CPU로 보내는 best-effort 동작입니다. mask가 isolated CPU만 담으면 routing에는 효과가 없지만 그 CPU task가 I/O를 제출할 때만 interrupt가 전달되며 housekeeping CPU의 I/O는 그 queue에 영향을 주지 않습니다. `<cpu-list>`는 문서 앞부분의 형식을 따릅니다.
iucv=[HW,NET]z/VM IUCV network 관련 설정입니다.

IVRS mapping override

2706-2759
매개변수적용·형식한국어 설명
ivrs_ioapic[HW,X86-64]IVRS ACPI table의 IOAPIC-ID와 DEVICE-ID mapping을 덮어씁니다. 기본 PCI segment는 0이며 생략할 수 있습니다. 새 형식은 `ivrs_ioapic=<ioapic-id>@<segment>:<bus>:<device>.<function>`입니다. 대괄호를 쓰는 두 legacy 형식은 폐기됐습니다.
ivrs_hpet[HW,X86-64]IVRS ACPI table의 HPET-ID와 DEVICE-ID mapping을 덮어씁니다. 기본 PCI segment는 0이며 생략할 수 있습니다. 새 형식은 `ivrs_hpet=<hpet-id>@<segment>:<bus>:<device>.<function>`입니다. 대괄호를 쓰는 두 legacy 형식은 폐기됐습니다. 원문의 마지막 legacy 예제는 `ivrs_ioapic[10]` 표기를 그대로 보존합니다.
ivrs_acpihid[HW,X86-64]IVRS ACPI table의 `ACPI-HID:UID`와 DEVICE-ID mapping을 덮어씁니다. 기본 PCI segment는 0이며 생략할 수 있습니다. 새 형식은 `ivrs_acpihid=<HID>:<UID>@<segment>:<bus>:<device>.<function>`이며 대괄호 legacy 형식은 폐기됐습니다.

IOAPIC-ID 10 mapping과 두 deprecated 형식을 원문 그대로 보존합니다.

			For example, to map IOAPIC-ID decimal 10 to
			PCI segment 0x1 and PCI device 00:14.0,
			write the parameter as:
				ivrs_ioapic=10@0001:00:14.0

			Deprecated formats:
			* To map IOAPIC-ID decimal 10 to PCI device 00:14.0
			  write the parameter as:
				ivrs_ioapic[10]=00:14.0
			* To map IOAPIC-ID decimal 10 to PCI segment 0x1 and
			  PCI device 00:14.0 write the parameter as:
				ivrs_ioapic[10]=0001:00:14.0

HPET-ID 10 mapping과 두 deprecated 형식을 원문 그대로 보존합니다.

			For example, to map HPET-ID decimal 10 to
			PCI segment 0x1 and PCI device 00:14.0,
			write the parameter as:
				ivrs_hpet=10@0001:00:14.0

			Deprecated formats:
			* To map HPET-ID decimal 0 to PCI device 00:14.0
			  write the parameter as:
				ivrs_hpet[0]=00:14.0
			* To map HPET-ID decimal 10 to PCI segment 0x1 and
			  PCI device 00:14.0 write the parameter as:
				ivrs_ioapic[10]=0001:00:14.0

ACPI HID `AMD0020:0` mapping과 두 deprecated 형식을 원문 그대로 보존합니다.

			For example, to map UART-HID:UID AMD0020:0 to
			PCI segment 0x1 and PCI device ID 00:14.5,
			write the parameter as:
				ivrs_acpihid=AMD0020:0@0001:00:14.5

			Deprecated formats:
			* To map UART-HID:UID AMD0020:0 to PCI segment is 0,
			  PCI device ID 00:14.5, write the parameter as:
				ivrs_acpihid[00:14.5]=AMD0020:0
			* To map UART-HID:UID AMD0020:0 to PCI segment 0x1 and
			  PCI device ID 00:14.5, write the parameter as:
				ivrs_acpihid[0001:00:14.5]=AMD0020:0

Joystick, KASAN과 memory zone

2760-2801
매개변수적용·형식한국어 설명
js=[HW,JOY]analog joystick 설정입니다. `Documentation/input/joydev/joystick.rst`를 참조하십시오.
kasan_multi_shot[KNL]KASAN이 잘못된 memory access를 처음 한 번만이 아니라 매번 보고하게 강제합니다.
keep_bootcon[KNL,EARLY]시작할 때 boot console을 등록 해제하지 않습니다. boot console 해제와 실제 console 초기화 사이에서 생기는 문제를 debugging할 때만 유용합니다.
keepinitrd[HW,ARM]`retain_initrd`를 참조하십시오.
kernelcore=[KNL,X86,PPC,EARLY] `nn[KMGTPE] | nn% | mirror`kernel의 non-movable allocation에 쓸 memory 양을 정합니다. 요청량은 모든 node에 `ZONE_NORMAL`로 고르게 분배되고 나머지는 `ZONE_MOVABLE`이 됩니다. 한 node가 두 zone을 모두 갖기 너무 작으면 kernelcore가 우선되고 다른 node의 `ZONE_MOVABLE`이 커집니다. `ZONE_MOVABLE`은 page migration subsystem이 회수·이동할 수 있는 page에 쓰지만 HighMem의 PTE 같은 allocation은 HighMem zone이 있으면 그대로 HighMem을, 없으면 Normal을 씁니다. `nn[KMGTPE]`는 정확한 크기, `nn%`는 전체 memory 비율입니다. `mirror`는 mirrored reliable memory를 non-movable allocation에 쓰고 나머지를 movable page로 씁니다. 세 형식은 서로 배타적입니다.

KGDB console backend

2802-2840
매개변수적용·형식한국어 설명
kgdbdbgp=[KGDB,HW,EARLY] `<Controller#>[,poll interval]`EHCI USB debug port를 통한 KGDB입니다. controller 번호는 PCI probe 순서의 EHCI debug port 번호입니다. 선택적 poll interval은 dbgp 연결에서 gdb 또는 Control-C로 kernel을 interrupt하기 위해 debug port를 polling하는 주기(초)입니다. 이 매개변수를 쓰지 않으면 `sysrq-g`로 debugger에 들어갑니다.
kgdboc=[KGDB,HW]console polling을 지원하는 tty driver 또는 지원되는 비-USB polling keyboard driver를 KGDB backend로 씁니다. serial, keyboard, 두 장치 조합과 선택적 kernel mode setting 형식을 지원합니다.
kgdboc_earlycon=[KGDB,HW,EARLY] 초기 console 이름boot console이 문자를 읽고 polling mode로 동작할 수 있으면 정규 console 등록 전까지 KGDB backend로 씁니다. 전환할 정규 console을 지정하는 `kgdboc`와 함께 사용합니다. 초기 console 이름은 `kgdboc`의 tty 이름과 다를 수 있습니다. 값을 비우면 `read()`를 구현한 첫 boot console을 고릅니다.
kgdbwait[KGDB,EARLY]가능한 가장 이른 시점에 kernel 실행을 멈추고 kernel debugger로 들어갑니다.

KGDB console 조합의 정확한 문법을 원문 그대로 보존합니다.

			or a supported polling keyboard driver (non-usb).
			 Serial only format: <serial_device>[,baud]
			 keyboard only format: kbd
			 keyboard and serial format: kbd,<serial_device>[,baud]
			Optional Kernel mode setting:
			 kms, kbd format: kms,kbd
			 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud]

Kexec HandOver와 kmemleak

2841-2875
매개변수적용·형식한국어 설명
kho=[KEXEC,EARLY] `{ 0 | 1 | off | on | y | n }``0`·`off`·`n`은 Kexec HandOver를 끄고 `1`·`on`·`y`는 켭니다.
kho_scratch=[KEXEC,EARLY] `ll[KMG],mm[KMG],nn[KMG] | nn%`KHO scratch region 크기를 정합니다. 이 물리적으로 연속된 memory range는 non-kernel allocation에만 써서 handed-over memory가 심하게 조각나도 새 kexec kernel이 bootstrap할 연속 range를 확보하게 합니다. 세 크기 형식은 low-memory scratch, global scratch, node별 추가 scratch 크기 순서입니다. `nn%`는 boot 중 사용한 memory에 대한 비율입니다.
kmac=[MIPS] MAC addressRouterBoard 532 series의 on-chip Korina Ethernet adapter MAC address를 설정합니다.
kmemleak=[KNL,EARLY] `{ on | off }`, 기본 `on`부팅 때 kmemleak을 켜거나 끕니다. `CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y`로 빌드하면 기본값은 off입니다.

Kprobe, KPTI와 KUnit

2876-2903
매개변수적용·형식한국어 설명
kprobe_event=[probe-list][FTRACE] 세미콜론 구분 probe 정의 목록부팅 때 kprobe event를 추가하고 켭니다. 각 정의는 `kprobe_events` interface와 같지만 parameter는 쉼표로 구분합니다. `Documentation/trace/kprobetrace.rst`의 Kernel Boot Parameter 절을 참조하십시오.
kpti=[ARM64,EARLY] `{ 0 | 1 }`user·kernel address space의 page table isolation을 제어합니다. 완화가 필요한 core에서는 기본 enabled이며 0은 강제 비활성화, 1은 강제 활성화입니다.
kunit.enable=[KUNIT] `{ 0 | 1 }`, 기본 `1`KUnit test 실행을 켭니다. 완전히 사용하려면 `CONFIG_KUNIT`이 필요하며 기본은 `KUNIT_DEFAULT_ENABLED`로 덮어쓸 수 있습니다.
kvm.ignore_msrs=[KVM], 기본 `0`처리하지 못한 MSR에 대한 guest 접근을 무시합니다. 기본은 무시하지 않고 `#GP`를 inject합니다.

`vfs_read`에 `$arg1`, `$arg2`를 전달하는 kprobe 예제를 보존합니다.

			arg1 and arg2, add to the command line;

			      kprobe_event=p,vfs_read,$arg1,$arg2

KVM 공통 동작

2904-2948
매개변수적용·형식한국어 설명
kvm.eager_page_split=[KVM,X86], 기본 `Y`dirty logging 중 모든 huge page를 미리 split할지 정합니다. eager split은 lazy split에 필요한 write-protection fault와 MMU lock contention을 없애 vCPU interruption을 줄입니다. write가 드물거나 VM memory의 작은 영역만 쓰는 workload는 끄면 read에 huge page를 계속 써 이득일 수 있습니다. `KVM_DIRTY_LOG_INITIALLY_SET`이 꺼져 있으면 memslot의 dirty logging을 켤 때 모든 huge page를 split합니다. 켜져 있으면 `KVM_CLEAR_DIRTY` ioctl 중 clear하는 page만 split합니다. `kvm.tdp_mmu=Y`에서만 지원합니다.
kvm.enable_virt_at_load=[KVM,ARM64,LOONGARCH,MIPS,RISCV,X86]켜면 KVM load 때 hardware virtualization을 켜고 module unload 때 끕니다. 끄면 VM 수가 0에서 1, 1에서 0으로 바뀔 때 demand 방식으로 켜고 끕니다. load 때 활성화하면 모든 online CPU에 걸친 직렬화로 인해 첫 VM 생성에 생기는 latency를 피하지만 virtualization hardware를 소유하려는 out-of-tree hypervisor와 충돌할 수 있습니다.
kvm.enable_vmware_backdoor=[KVM] `<bool>`, 기본 `false`VMware backdoor paravirtual interface 지원을 제어합니다.

KVM NX huge page와 AMD

2949-3003
매개변수적용·형식한국어 설명
kvm.nx_huge_pages=[KVM] `{ force | off | auto }`, 기본 `auto``X86_BUG_ITLB_MULTIHIT`의 software workaround를 제어합니다. `force`는 항상, `off`는 절대 배치하지 않고 `auto`는 bug 존재 여부에 따릅니다. host에 workaround가 켜져 있으면 guest가 nested guest용 완화를 따로 켤 필요는 없습니다.
kvm.nx_huge_pages_recovery_ratio=[KVM], 기본 `60`주기적으로 4KiB page를 huge page로 되돌리는 비율입니다. 0은 recovery를 끄고 N이면 매 period마다 4KiB page의 1/N을 zap합니다.
kvm.nx_huge_pages_recovery_period_ms=[KVM], 기본 `0`4KiB page를 huge page로 되돌리는 period입니다. N이면 N ms마다 ratio만큼 zap합니다. 0이면 각 page가 평균 한 시간 뒤 zap되도록 ratio에서 period를 정합니다.
kvm-amd.nested=[KVM,AMD], 기본 `1`KVM·SVM nested virtualization을 제어합니다.
kvm-amd.npt=[KVM,AMD], 기본 `1`Nested Page Tables, 즉 Two-Dimensional Page Tables 사용을 제어합니다. hardware가 NPT를 지원하지 않으면 KVM이 끕니다.
kvm-amd.ciphertext_hiding_asids=[KVM,AMD], non-zeroSNP private memory에 허용되지 않은 read가 ciphertext 대신 고정값 `0xff`를 보게 합니다. 켜면 공동 SEV-ES·SEV-SNP ASID 공간을 SEV-SNP `[1..max_snp_asid]`와 SEV-ES `(max_snp_asid..min_sev_asid)`로 나눕니다. `min_sev_asid`는 `CPUID.0x8000_001F[EDX]`에서 열거합니다. non-zero 값은 hiding을 켜고 두 guest 종류의 range를 조정합니다. KVM은 SEV-SNP ASID 수를 가능한 최대치로 제한하므로 `-1u`는 공동 ASID 전체를 SEV-SNP에 줍니다. `max_snp_asid == min_sev_asid-1`이면 SEV-ES는 사실상 쓸 수 없습니다.

KVM arm64 mode와 GIC trap

3004-3062
매개변수적용·형식한국어 설명
kvm-arm.mode=[KVM,ARM,EARLY] `{ none | nvhe | protected | nested }``none`은 KVM을 강제로 끕니다. `nvhe`는 protected guest가 없는 표준 nVHE mode입니다. `protected`는 hardware 지원에 따라 VHE 또는 nVHE를 쓰면서 guest state를 host에서 비공개로 유지합니다. `nested`는 ARMv8.4 `FEAT_NV2` 이상에서 nested virtualization을 지원하는 VHE mode입니다. 기본은 hardware에 따른 VHE·nVHE입니다. protected mode는 host의 kexec와 hibernation을 끕니다. VHE hardware에서 nVHE를 강제하려면 `arm64_sw.hvhe=0 id_aa64mmfr1.vh=0`을 command line에 추가합니다. nested는 experimental입니다.
kvm-arm.vgic_v3_group0_trap=[KVM,ARM,EARLY]guest의 GICv3 group-0 system register 접근을 trap합니다.
kvm-arm.vgic_v3_group1_trap=[KVM,ARM,EARLY]guest의 GICv3 group-1 system register 접근을 trap합니다.
kvm-arm.vgic_v3_common_trap=[KVM,ARM,EARLY]guest의 GICv3 common system register 접근을 trap합니다.
kvm-arm.vgic_v4_enable=[KVM,ARM,EARLY]LPI direct injection에 GICv4 사용을 허용합니다.
kvm-arm.wfe_trap_policy=[KVM,ARM] `{ trap | notrap }`KVM VM의 WFE instruction trap 설정 시점을 제어합니다. architecture가 trap을 허용하지만 보장하지는 않습니다. `trap`은 설정하고 `notrap`은 해제합니다.
kvm-arm.wfi_trap_policy=[KVM,ARM] `{ trap | notrap }`KVM VM의 WFI instruction trap 설정 시점을 제어합니다. architecture가 trap을 허용하지만 보장하지는 않습니다. `trap`은 설정하고 `notrap`은 해제합니다.

KVM arm64 mode의 원문 선택지와 주의사항을 그대로 보존합니다.

	kvm-arm.mode=
			[KVM,ARM,EARLY] Select one of KVM/arm64's modes of
			operation.

			none: Forcefully disable KVM.

			nvhe: Standard nVHE-based mode, without support for
			      protected guests.

			protected: Mode with support for guests whose state is
				   kept private from the host, using VHE or
				   nVHE depending on HW support.

			nested: VHE-based mode with support for nested
				virtualization. Requires at least ARMv8.4
				hardware (with FEAT_NV2).

			Defaults to VHE/nVHE based on hardware support. Setting
			mode to "protected" will disable kexec and hibernation
			for the host. To force nVHE on VHE hardware, add
			"arm64_sw.hvhe=0 id_aa64mmfr1.vh=0" to the
			command-line.
			"nested" is experimental and should be used with
			extreme caution.

KVM PPC·Intel 설정

3063-3115
매개변수적용·형식한국어 설명
kvm_cma_resv_ratio=n[PPC,EARLY] `<integer>`, 기본 `5`KVM hash page table의 contiguous allocation을 위해 system memory의 지정 비율을 예약합니다.
kvm-intel.ept=[KVM,Intel], 기본 `1`Extended Page Tables, 즉 Two-Dimensional Page Tables 사용을 제어합니다. hardware가 EPT를 지원하지 않으면 KVM이 끕니다.
kvm-intel.emulate_invalid_guest_state=[KVM,Intel], 기본 `1`잘못된 guest state를 emulate할지 정합니다. `kvm-intel.enable_unrestricted_guest=1`이면 guest state가 invalid하지 않아 무시됩니다. KVM은 invalid L2 guest state를 emulate하지 않으므로 nested guest에는 적용되지 않습니다.
kvm-intel.flexpriority=[KVM,Intel], 기본 `1`FlexPriority, 즉 TPR shadow 사용을 제어합니다. hardware가 지원하지 않으면 KVM이 끕니다.
kvm-intel.nested=[KVM,Intel], 기본 `1`KVM·VMX nested virtualization을 제어합니다.
kvm-intel.unrestricted_guest=[KVM,Intel], 기본 `1`virtualized real·unpaged mode인 unrestricted guest를 제어합니다. EPT가 꺼져 있거나 hardware가 지원하지 않으면 KVM이 끕니다.
kvm-intel.vmentry_l1d_flush=[KVM,Intel] `{ never | cond | always }`, 기본 `cond`L1 Terminal Fault `CVE-2018-3620` 완화입니다. `always`는 모든 VMENTER, `cond`는 VMEXIT와 VMENTER 사이 code가 host memory를 leak할 수 있을 때만 L1D cache를 flush하고 `never`는 완화를 끕니다.
kvm-intel.vpid=[KVM,Intel], 기본 `1`tagged TLB인 Virtual Processor Identification 사용을 제어합니다. hardware가 지원하지 않으면 KVM이 끕니다.

L1D snooping과 L1TF 완화

3116-3197
매개변수적용·형식한국어 설명
l1d_flush=[X86,INTEL,EARLY] `{ on }`L1D 기반 snooping 취약점 완화를 제어합니다. 영향받는 CPU에서는 내부 buffer가 특정 조건에서 정보를 disclosure gadget으로 전달하고, speculative data가 cache side-channel 공격에 쓰여 공격자가 직접 접근할 수 없는 data를 노출할 수 있습니다. `on`은 완화 interface를 켭니다.
l1tf=[X86,EARLY] `{ full | full,force | flush | flush,nosmt | flush,nowarn | off }`, 기본 `flush`영향받는 CPU의 L1TF 완화를 제어합니다. kernel PTE inversion 보호는 항상 켜져 있고 끌 수 없습니다. `full`은 SMT를 끄고 hypervisor에서 unconditional L1D flush 등 모든 완화를 켜며 부팅 뒤 sysfs로 SMT·flush를 바꿀 수 있습니다. insecure 구성에서 첫 VM 시작 시 warning을 냅니다. `full,force`는 runtime 제어도 막고 `nosmt=force`를 뜻합니다. `flush`는 SMT를 유지하면서 conditional L1D flush를 켜고 sysfs 제어·warning을 유지합니다. `flush,nosmt`는 SMT를 끄고 기본 hypervisor 완화를 켭니다. `flush,nowarn`은 `flush`와 같지만 insecure VM warning을 내지 않습니다. `off`는 hypervisor 완화와 warning을 끄고 hypervisor·bare metal 모두의 swap 크기와 available RAM 제한도 제거합니다. `Documentation/admin-guide/hw-vuln/l1tf.rst`를 참조하십시오.

L1TF mode별 상세 동작을 원문 그대로 보존합니다.

	l1tf=           [X86,EARLY] Control mitigation of the L1TF vulnerability on
			      affected CPUs

			The kernel PTE inversion protection is unconditionally
			enabled and cannot be disabled.

			full
				Provides all available mitigations for the
				L1TF vulnerability. Disables SMT and
				enables all mitigations in the
				hypervisors, i.e. unconditional L1D flush.

				SMT control and L1D flush control via the
				sysfs interface is still possible after
				boot.  Hypervisors will issue a warning
				when the first VM is started in a
				potentially insecure configuration,
				i.e. SMT enabled or L1D flush disabled.

			full,force
				Same as 'full', but disables SMT and L1D
				flush runtime control. Implies the
				'nosmt=force' command line option.
				(i.e. sysfs control of SMT is disabled.)

			flush
				Leaves SMT enabled and enables the default
				hypervisor mitigation, i.e. conditional
				L1D flush.

				SMT control and L1D flush control via the
				sysfs interface is still possible after
				boot.  Hypervisors will issue a warning
				when the first VM is started in a
				potentially insecure configuration,
				i.e. SMT enabled or L1D flush disabled.

			flush,nosmt

				Disables SMT and enables the default
				hypervisor mitigation.

				SMT control and L1D flush control via the
				sysfs interface is still possible after
				boot.  Hypervisors will issue a warning
				when the first VM is started in a
				potentially insecure configuration,
				i.e. SMT enabled or L1D flush disabled.

			flush,nowarn
				Same as 'flush', but hypervisors will not
				warn when a VM is started in a potentially
				insecure configuration.

			off
				Disables hypervisor mitigations and doesn't
				emit any warnings.
				It also drops the swap size and available
				RAM limit restriction on both hypervisor and
				bare metal.

			Default is 'flush'.

			For details see: Documentation/admin-guide/hw-vuln/l1tf.rst

PowerPC control register와 LAPIC

3198-3212
매개변수적용·형식한국어 설명
l2cr=[PPC]PowerPC L2 control register 관련 설정입니다.
l3cr=[PPC]PowerPC L3 control register 관련 설정입니다.
lapic[X86-32,APIC,EARLY]BIOS가 껐어도 local APIC를 켭니다.
lapic=[X86,APIC] `notscdeadline`LAPIC timer one-shot 구현에 TSC deadline 값을 쓰지 않고 programmable LAPIC timer unit으로 돌아갑니다.
lapic_timer_c2_ok[X86,APIC,EARLY]C2 power state에서도 local APIC timer를 신뢰합니다.

libata DMA·HPA·ACPI

3213-3228
매개변수적용·형식한국어 설명
libata.dma=[LIBATA] bitmask `0..7`0은 모든 PATA·SATA DMA, 1은 disk DMA, 2는 ATAPI CD-ROM DMA, 4는 CompactFlash DMA를 뜻합니다. 조합할 수 있어 3은 disk와 CD-ROM DMA를 켜고 CF DMA는 끕니다.
libata.ignore_hpa=[LIBATA] `{ 0 | 1 }`, 기본 `0`0은 BIOS HPA limit을 유지하고 1은 limit을 무시해 disk 전체를 씁니다.
libata.noacpi[LIBATA] `<int>`설정하면 libata suspend·resume에서 ACPI 사용을 끕니다.

libata 강제 구성

3229-3321
매개변수적용·형식한국어 설명
libata.force=[LIBATA] 쉼표 구분 `[ID:]VAL`, ID는 `PORT[.DEVICE]`console에 출력된 ATA ID에 맞춰 port·link·device 설정을 강제합니다. ID 전체를 생략하면 마지막 PORT·DEVICE를 재사용하고 아직 ID가 없으면 모든 대상에 적용합니다. DEVICE만 생략하면 port와 뒤의 모든 link·device에 적용합니다. DEVICE 0은 첫 device 또는 PMP 뒤 첫 fan-out link, 15는 host link와 연결 device를 선택합니다. VAL은 모호하지 않으면 `1.5`처럼 단축할 수 있습니다. 강제 항목은 cable `40c|80c|short40c|unk|ign|sata`, SATA speed `1.5Gbps|3.0Gbps`, transfer mode `pio[0-7]|mwdma[0-4]|udma[0-7]` 및 `udma[/][16,25,33,44,66,100,133]`입니다. `nohrst|nosrst|norst`는 hard·soft·모든 reset을 억제하고 `rstonce`는 hot-unplug recovery 때 reset을 한 번만 시도합니다. `[no]dbdelay`는 PHY·device presence debounce 전 200ms 지연, `[no]ncq`, `[no]ncqtrim`, `[no]ncqati`, `[no]trim`은 각각 NCQ, queued DSM TRIM, ATI NCQ TRIM, unqueued TRIM을 제어합니다. `trim_zero`는 TRIM이 data를 zero로 만든다고 표시하고 `max_trim_128m`은 최대 trim 크기를 128M으로 제한합니다. `[no]dma`, `atapi_dmadir`, `atapi_mod16_dma`, `[no]dmalog`, `[no]iddevlog`, `[no]logdir`는 DMA와 ATAPI bridge·16byte 비배수 command DMA, READ LOG DMA EXT, identify-device data log, general-purpose log directory를 제어합니다. `max_sec_128`, `max_sec_1024`, `max_sec_lba48`은 transfer를 각각 128, 1024, 65,535 sector로 제한하거나 해제합니다. `external`은 hotplug 가능 외부 port로 표시하고 `[no]lpm`, `[no]setxfer`, `[no]fua`는 link power management, transfer-speed mode 설정 건너뛰기, Force Unit Access를 제어합니다. `dump_id`는 IDENTIFY data를 dump하고 `disable`은 device를 끕니다. 같은 attribute를 바꾸는 matching 설정이 여럿이면 마지막 값이 우선합니다.

모든 `libata.force=` 선택지와 철자를 원문 그대로 보존합니다.

	libata.force=	[LIBATA] Force configurations.  The format is a comma-
			separated list of "[ID:]VAL" where ID is PORT[.DEVICE].
			PORT and DEVICE are decimal numbers matching port, link
			or device.  Basically, it matches the ATA ID string
			printed on console by libata.  If the whole ID part is
			omitted, the last PORT and DEVICE values are used.  If
			ID hasn't been specified yet, the configuration applies
			to all ports, links and devices.

			If only DEVICE is omitted, the parameter applies to
			the port and all links and devices behind it.  DEVICE
			number of 0 either selects the first device or the
			first fan-out link behind PMP device.  It does not
			select the host link.  DEVICE number of 15 selects the
			host link and device attached to it.

			The VAL specifies the configuration to force.  As long
			as there is no ambiguity, shortcut notation is allowed.
			For example, both 1.5 and 1.5G would work for 1.5Gbps.
			The following configurations can be forced.

			* Cable type: 40c, 80c, short40c, unk, ign or sata.
			  Any ID with matching PORT is used.

			* SATA link speed limit: 1.5Gbps or 3.0Gbps.

			* Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7].
			  udma[/][16,25,33,44,66,100,133] notation is also
			  allowed.

			* nohrst, nosrst, norst: suppress hard, soft and both
			  resets.

			* rstonce: only attempt one reset during hot-unplug
			  link recovery.

			* [no]dbdelay: Enable or disable the extra 200ms delay
			  before debouncing a link PHY and device presence
			  detection.

			* [no]ncq: Turn on or off NCQ.

			* [no]ncqtrim: Enable or disable queued DSM TRIM.

			* [no]ncqati: Enable or disable NCQ trim on ATI chipset.

			* [no]trim: Enable or disable (unqueued) TRIM.

			* trim_zero: Indicate that TRIM command zeroes data.

			* max_trim_128m: Set 128M maximum trim size limit.

			* [no]dma: Turn on or off DMA transfers.

			* atapi_dmadir: Enable ATAPI DMADIR bridge support.

			* atapi_mod16_dma: Enable the use of ATAPI DMA for
			  commands that are not a multiple of 16 bytes.

			* [no]dmalog: Enable or disable the use of the
			  READ LOG DMA EXT command to access logs.

			* [no]iddevlog: Enable or disable access to the
			  identify device data log.

			* [no]logdir: Enable or disable access to the general
			  purpose log directory.

			* max_sec_128: Set transfer size limit to 128 sectors.

			* max_sec_1024: Set or clear transfer size limit to
			  1024 sectors.

			* max_sec_lba48: Set or clear transfer size limit to
			  65535 sectors.

			* external: Mark port as external (hotplug-capable).

			* [no]lpm: Enable or disable link power management.

			* [no]setxfer: Indicate if transfer speed mode setting
			  should be skipped.

			* [no]fua: Disable or enable FUA (Force Unit Access)
			  support for devices supporting this feature.

			* dump_id: Dump IDENTIFY data.

			* disable: Disable this device.

			If there are multiple matching configurations changing
			the same attribute, the last one is used.

Ramdisk, NFS lock과 lockdown

3322-3344
매개변수적용·형식한국어 설명
load_ramdisk=[RAM], 폐기됨폐기된 initial ramdisk load 설정입니다.
lockd.nlm_grace_period=P[NFS] `<integer>`NLM grace period를 지정합니다.
lockd.nlm_tcpport=N[NFS] `<integer>`NLM TCP port를 지정합니다.
lockd.nlm_timeout=T[NFS] `<integer>`NLM timeout 값을 지정합니다.
lockd.nlm_udpport=M[NFS] `<integer>`NLM UDP port를 지정합니다.
lockdown=[SECURITY,EARLY] `{ integrity | confidentiality }`kernel lockdown을 켭니다. `integrity`는 사용자 공간이 실행 중 kernel을 수정할 수 있게 하는 기능을 끕니다. `confidentiality`는 여기에 사용자 공간이 kernel의 confidential information을 추출할 수 있게 하는 기능까지 끕니다.

Locktorture 설정

3345-3439
매개변수적용·형식한국어 설명
locktorture.acq_writer_lim=[KNL] jiffieslock acquisition의 시간 제한입니다. 넘긴 acquisition은 완료된 뒤 splat을 냅니다.
locktorture.bind_readers=[KNL] CPU 목록reader를 bind할 CPU 목록입니다.
locktorture.bind_writers=[KNL] CPU 목록writer를 bind할 CPU 목록입니다.
locktorture.call_rcu_chains=[KNL], 기본 `0`self-propagating `call_rcu()` chain 수입니다. 어느 시점이든 RCU grace period가 진행 중일 가능성을 높입니다. 0은 chain을 끕니다.
locktorture.long_hold=[KNL] ms, 기본 `100`간헐적으로 lock을 오래 보유할 duration입니다. 0은 끕니다.
locktorture.nested_locks=[KNL] `0..8`시험할 최대 lock nesting 깊이이며 상한은 `MAX_NESTED_LOCKS` 8입니다. 0은 끕니다. nested acquisition을 지원하지 않는 lock에는 효과가 없습니다.
locktorture.nreaders_stress=[KNL]read acquisition kthread 수입니다. 기본은 online CPU 수에 따라 자동 설정됩니다.
locktorture.nwriters_stress=[KNL]write acquisition kthread 수입니다.
locktorture.onoff_holdoff=[KNL] 초부팅 뒤 CPU-hotplug test를 시작하기 전 대기 시간입니다.
locktorture.onoff_interval=[KNL] 초CPU-hotplug operation 사이 간격이며 0은 test를 끕니다.
locktorture.rt_boost=[KNL] `{ 0 | 1 | 2 }`, 기본 `2`real-time lock priority boosting을 주기적으로 시험합니다. 0은 끄고 1은 `rt_mutex`만, 2는 무조건 boost합니다. non-real-time spinlock은 preemption을 끄므로 2여도 해가 없어야 합니다. non-realtime mutex는 boosting을 끕니다.
locktorture.rt_boost_factor=[KNL]priority boosting 빈도와 duration을 정하는 수입니다. writer 수가 늘어도 단위 시간당 boost 수가 대략 일정하도록 빈도를 낮추고, 각 boost duration은 늘립니다.
locktorture.shuffle_interval=[KNL] jiffiestask shuffle 간격입니다. test 중 일부 CPU가 dyntick-idle에 들어갈 수 있게 합니다.
locktorture.shutdown_secs=[KNL] 초부팅 뒤 system shutdown까지 시간입니다. 무인 자동 test에 유용합니다.
locktorture.stat_interval=[KNL] 초통계 `printk()` 사이 간격입니다.
locktorture.stutter=[KNL] 초test의 실행·대기를 번갈아 반복하는 시간입니다. 예를 들어 5는 5초 실행하고 5초 대기합니다. lock primitive가 idle과 active 사이를 급히 전환하는 능력을 시험합니다.
locktorture.torture_type=[KNL]시험할 locking 구현을 지정합니다.
locktorture.verbose=[KNL]추가 `printk()` 문을 켭니다.
locktorture.writer_fifo=[KNL]write-side locktorture kthread를 `sched_set_fifo()` real-time priority로 실행합니다.

Kernel log와 parallel printer

3440-3489
매개변수적용·형식한국어 설명
logibm.irq=[HW,MOUSE] `<irq>`Logitech Bus Mouse driver IRQ를 지정합니다.
loglevel=[KNL,EARLY] `0..7`console loglevel보다 숫자가 작은 kernel message를 console에 출력합니다. `klogd` 등으로 바꿀 수도 있습니다. 0 `KERN_EMERG`는 system unusable, 1 `KERN_ALERT`는 즉시 조치, 2 `KERN_CRIT`는 critical, 3 `KERN_ERR`는 error, 4 `KERN_WARNING`은 warning, 5 `KERN_NOTICE`는 정상이나 중요한 상태, 6 `KERN_INFO`는 정보, 7 `KERN_DEBUG`는 debug message입니다.
log_buf_len=n[KMG][KNL,EARLY], 2의 거듭제곱 byteprintk ring buffer 크기입니다. `n`은 `LOG_BUF_SHIFT`가 정한 최소값보다 큰 2의 거듭제곱이어야 합니다. `CONFIG_LOG_CPU_MAX_BUF_SHIFT`는 CPU 수에 따라 기본값을 늘릴 수 있습니다. `init/Kconfig`를 참조하십시오.
logo.nologo[FB]내장 Linux logo 표시를 끕니다. kernel log 공간을 늘려 boot 문제 debugging에 유용합니다.
lp=0[LP]printer driver를 끕니다.
lp=port[,port...][LP]`lp0`부터 연결할 parallel port 목록입니다. `none`은 해당 lp device를 건너뛰고 `parport0` 같은 이름을 쓸 수 있습니다.
lp=reset[LP]port 지정과 함께 사용할 수 있으며 연결된 printer를 reset합니다.
lp=auto[LP]각 port의 device ID를 검사해 IEEE 1284 printer가 있으면 driver가 관리합니다. `drivers/char/lp.c` 머리말도 참조하십시오.

loglevel 정의와 parallel-printer 문법을 원문 그대로 보존합니다.

			0 (KERN_EMERG)		system is unusable
			1 (KERN_ALERT)		action must be taken immediately
			2 (KERN_CRIT)		critical conditions
			3 (KERN_ERR)		error conditions
			4 (KERN_WARNING)	warning conditions
			5 (KERN_NOTICE)		normal but significant condition
			6 (KERN_INFO)		informational
			7 (KERN_DEBUG)		debug-level messages
	lp=0		[LP]	Specify parallel ports to use, e.g,
	lp=port[,port...]	lp=none,parport0 (lp0 not configured, lp1 uses
	lp=reset		first parallel port). 'lp=0' disables the
	lp=auto			printer driver. 'lp=reset' (which can be
				specified in addition to the ports) causes
				attached printers to be reset. Using
				lp=port1,port2,... specifies the parallel ports
				to associate lp devices with, starting with
				lp0. A port specification may be 'none' to skip
				that lp device, or a parport name such as
				'parport0'. Specifying 'lp=auto' instead of a
				port specification list means that device IDs
				from each port should be examined, to see if
				an IEEE 1284-compliant printer is attached; if
				so, the driver will manage that printer.
				See also header of drivers/char/lp.c.

Loop calibration, LSM과 CPU 수

3490-3529
매개변수적용·형식한국어 설명
lpj=n[KNL], 기본 `0``loops_per_jiffy`를 상수로 정해 CPU마다 최대 250ms 걸릴 수 있는 boot-time autodetection을 피합니다. 0은 자동 감지입니다. 정상 자동 감지로 부팅해 출력값을 확인해야 합니다. SMP에서는 모든 CPU에 같은 값을 써 CPU 특성이 크게 다르면 문제가 생길 수 있습니다. 잘못된 값은 delay를 틀리게 해 예측할 수 없는 I/O error나 손상, 극단적으로 hardware damage를 일으킬 수 있습니다.
lsm.debug[SECURITY]LSM initialization debug 출력을 켭니다.
lsm=lsm1,...,lsmN[SECURITY]LSM initialization 순서를 고릅니다. `CONFIG_LSM`과 `security=`를 덮어씁니다.
machtype=[Loongson]서로 다른 Yeeloong laptop이 같은 kernel image를 공유하게 합니다. 예: `machtype=lemote-yeeloong-2f-7inch`.
maxcpus=[SMP,EARLY] `n >= 0`SMP kernel이 boot 중 올릴 processor 최대 수입니다. 부팅 뒤 `echo 1 > /sys/devices/system/cpu/cpuX/online`으로 나머지를 올릴 수 있어 boot에만 적용됩니다. 0은 `nosmp`와 같고 IO APIC도 끕니다.
max_loop=[LOOP]init 때 조건 없이 미리 만들 loop block device 수입니다. 기본은 `BLK_DEV_LOOP_MIN_COUNT`입니다. 고정 수 대신 `/dev/loop-control`로 필요할 때 요청할 수 있습니다.
(loop.max_loop)[LOOP] alias`max_loop=`의 module-style 별칭입니다.

x86 Machine Check

3530-3601
매개변수적용·형식한국어 설명
mce=[X86-32,X86-64]`off`는 machine check를 끕니다. `no_cmci`는 Intel Corrected Machine Check Interrupt를 끄며 보통 권장하지 않습니다. shared bank 때문에 중복 error log가 더 생길 수 있습니다. `dont_log_ce`는 corrected error event를 조용히 clear하고 기록하지 않습니다. `ignore_ce`는 polling timer·CMCI 같은 corrected-error 기능을 끄고 event를 OS가 clear하지 않아 error bank에 남깁니다. BIOS·hardware monitor가 corrected error를 검사·clear해 OS와 충돌하지만 그 agent를 끌 수 없을 때 도움이 될 수 있습니다. `no_lmce`는 Local MCE delivery opt-in을 하지 않고 legacy broadcast를 씁니다. `bootlog`는 boot 전에 남은 machine check 기록을 켭니다. AMD Fam10h 이하는 잘못된 BIOS record 때문에 기본 off이고 Intel은 기본 on입니다. `nobootlog`는 이를 끕니다. `monarchtimeout=<us>`는 다른 CPU를 기다리는 시간이고 0은 대기를 끕니다. `bios_cmci_threshold`는 BIOS가 정한 CMCI threshold를 Linux가 1로 덮어쓰지 않게 하지만 memory predictive failure 분석을 약화할 수 있습니다. `recovery`는 recoverable machine-check code path를 강제로 켭니다. 나머지 runtime 조정은 sysfs에 있으며 `Documentation/arch/x86/x86_64/machinecheck.rst`를 참조하십시오.
		off
			disable machine check

		no_cmci
			disable CMCI(Corrected Machine Check Interrupt) that
			Intel processor supports.  Usually this disablement is
			not recommended, but it might be handy if your
			hardware is misbehaving.

			Note that you'll get more problems without CMCI than
			with due to the shared banks, i.e. you might get
			duplicated error logs.

		dont_log_ce
			don't make logs for corrected errors.  All events
			reported as corrected are silently cleared by OS. This
			option will be useful if you have no interest in any
			of corrected errors.

		ignore_ce
			disable features for corrected errors, e.g.
			polling timer and CMCI.  All events reported as
			corrected are not cleared by OS and remained in its
			error banks.

			Usually this disablement is not recommended, however
			if there is an agent checking/clearing corrected
			errors (e.g. BIOS or hardware monitoring
			applications), conflicting with OS's error handling,
			and you cannot deactivate the agent, then this option
			will be a help.

		no_lmce
			do not opt-in to Local MCE delivery. Use legacy method
			to broadcast MCEs.

		bootlog
			enable logging of machine checks left over from
			booting. Disabled by default on AMD Fam10h and older
			because some BIOS leave bogus ones.

			If your BIOS doesn't do that it's a good idea to
			enable though to make sure you log even machine check
			events that result in a reboot. On Intel systems it is
			enabled by default.

		nobootlog
			disable boot machine check logging.

		monarchtimeout (number)
			sets the time in us to wait for other CPUs on machine
			checks. 0 to disable.

		bios_cmci_threshold
			don't overwrite the bios-set CMCI threshold. This boot
			option prevents Linux from overwriting the CMCI
			threshold set by the bios.  Without this option, Linux
			always sets the CMCI threshold to 1. Enabling this may
			make memory predictive failure analysis less effective
			if the bios sets thresholds for memory errors since we
			will not see details for all errors.

		recovery
			force-enable recoverable machine check code paths

			Everything else is in sysfs now.

RAID와 MDS 완화

3602-3640
매개변수적용·형식한국어 설명
md=[HW] RAID device·level`Documentation/admin-guide/md.rst`를 참조하십시오.
mdacon=[MDA] `<first>,<last>`MDA가 capture할 console 범위를 지정합니다.
mds=[X86,INTEL,EARLY] `{ full | full,nosmt | off }`, 기본 `full`Micro-architectural Data Sampling 완화를 제어합니다. 영향받는 CPU 내부 buffer의 speculative data가 cache side channel로 직접 접근할 수 없는 정보를 노출할 수 있습니다. `full`은 취약 CPU에서 완화, `full,nosmt`는 완화와 SMT 비활성화, `off`는 조건 없이 완화를 끕니다. TAA도 영향받는 machine에서는 두 취약점이 같은 mechanism으로 완화되어 active TAA 완화가 `mds=off`를 막을 수 있으므로 완전히 끄려면 `tsx_async_abort=off`도 필요합니다. `Documentation/admin-guide/hw-vuln/mds.rst`를 참조하십시오.
	mds=		[X86,INTEL,EARLY]
			Control mitigation for the Micro-architectural Data
			Sampling (MDS) vulnerability.

			Certain CPUs are vulnerable to an exploit against CPU
			internal buffers which can forward information to a
			disclosure gadget under certain conditions.

			In vulnerable processors, the speculatively
			forwarded data can be used in a cache side channel
			attack, to access data to which the attacker does
			not have direct access.

			This parameter controls the MDS mitigation. The
			options are:

			full       - Enable MDS mitigation on vulnerable CPUs
			full,nosmt - Enable MDS mitigation and disable
				     SMT on vulnerable CPUs
			off        - Unconditionally disable MDS mitigation

			On TAA-affected machines, mds=off can be prevented by
			an active TAA mitigation as both vulnerabilities are
			mitigated with the same mechanism so in order to disable
			this mitigation, you need to specify tsx_async_abort=off
			too.

			Not specifying this option is equivalent to
			mds=full.

			For details see: Documentation/admin-guide/hw-vuln/mds.rst

Memory 크기와 hotplug 기본 정책

3641-3693
매개변수적용·형식한국어 설명
mem=nn[KMG][HEXAGON,EARLY]memory 크기를 정합니다. 반드시 지정해야 하며 생략하면 0입니다.
mem=nn[KMG][KNL,BOOT,EARLY]kernel이 사용할 memory 양 또는 x86의 최대 주소를 제한합니다. test, kernel이 전체 memory를 못 볼 때, 경계 밖 memory를 hypervisor에서 제외해 KVM guest에 줄 때, kdump kernel memory 제한에 씁니다. ARC·MicroBlaze는 low memory만 제한하고 high memory에는 영향이 없습니다. ARM64는 linear mapping에 포함된 memory만 제한하고 NOMAP region에는 영향이 없습니다. x86은 physical address collision을 피하도록 `memmap=`과 함께 써야 하며, 그렇지 않으면 PCI device가 사용하지 않는 RAM 주소에 배치될 수 있습니다. hypervisor memory가 부족하면 제외한 memory를 boot 뒤 hot-add해야 할 수 있으므로 이 제한은 boot 때만 작용합니다.
mem=nn[KMG]@ss[KMG][ARM,MIPS,EARLY]firmware가 보고한 memory layout을 덮어써 `ss`에서 시작하는 크기 `nn`의 region을 정의합니다. command line에 여러 `mem=`으로 여러 region을 지정할 수 있습니다.
mem=nopentium[BUGS=X86-32]kernel memory에 4MB page를 사용하지 않습니다.
memblock=debug[KNL,EARLY]memblock debug message를 켭니다.
memchunk=nn[KMG][KNL,SH]device별 physically contiguous DMA buffer의 기본 크기를 덮어씁니다.
memhp_default_state=online/offline/online_kernel/online_movable[KNL]memory hotplug onlining policy의 초기 상태입니다. 생략하면 `CONFIG_MHP_DEFAULT_ONLINE_TYPE`이 정합니다. `Documentation/admin-guide/mm/memory-hotplug.rst`를 참조하십시오.

E820 memory map override

3694-3739
매개변수적용·형식한국어 설명
memmap=exactmap[KNL,X86,EARLY]사용자가 지정한 정확한 E820 memory map을 사용합니다. BIOS 출력 또는 요구사항에서 `memmap=exactmap` line을 만들 수 있으며 `memmap=nn@ss` 설명을 참조하십시오.
memmap=nn[KMG]@ss[KMG][KNL,X86,MIPS,XTENSA,EARLY]`ss`부터 `ss+nn`까지 사용할 memory region을 강제합니다. `@ss`를 생략하면 최대 주소를 `nn`으로 제한하는 `mem=nn`과 같습니다. 쉼표로 여러 region을 지정할 수 있습니다.
memmap=nn[KMG]#ss[KMG][KNL,ACPI,EARLY]`ss`부터 `ss+nn`까지를 ACPI data로 표시합니다.
memmap=nn[KMG]$ss[KMG][KNL,ACPI,EARLY]`ss`부터 `ss+nn`까지를 reserved로 표시합니다. 일부 bootloader, 예를 들어 Grub2는 `$`와 뒤 숫자를 소비하지 않도록 `$` 앞에 escape가 필요합니다.
memmap=nn[KMG]!ss[KMG][KNL,X86,EARLY]`ss`부터 `ss+nn`까지를 protected memory로 표시합니다. e820 type 12(0xc), NVDIMM 또는 ADR memory일 수 있습니다.
memmap=<size>%<offset>-<oldtype>+<newtype>[KNL,ACPI,EARLY]지정 region의 memory를 old e820 type에서 new type으로 바꿉니다. `-<oldtype>`을 생략하면 이전 상태와 무관하게 region 전체를 new type으로 표시하고, `+<newtype>`을 생략하면 일치 memory를 제거합니다. 예: 1 RAM, 2 reserved, 3 ACPI, 12 PRAM.

모든 memmap 문법과 예제를 원문 그대로 보존합니다.

	memmap=exactmap	[KNL,X86,EARLY] Enable setting of an exact
			E820 memory map, as specified by the user.
			Such memmap=exactmap lines can be constructed based on
			BIOS output or other requirements. See the memmap=nn@ss
			option description.

	memmap=nn[KMG]@ss[KMG]
			[KNL, X86,MIPS,XTENSA,EARLY] Force usage of a specific region of memory.
			Region of memory to be used is from ss to ss+nn.
			If @ss[KMG] is omitted, it is equivalent to mem=nn[KMG],
			which limits max address to nn[KMG].
			Multiple different regions can be specified,
			comma delimited.
			Example:
				memmap=100M@2G,100M#3G,1G!1024G

	memmap=nn[KMG]#ss[KMG]
			[KNL,ACPI,EARLY] Mark specific memory as ACPI data.
			Region of memory to be marked is from ss to ss+nn.

	memmap=nn[KMG]$ss[KMG]
			[KNL,ACPI,EARLY] Mark specific memory as reserved.
			Region of memory to be reserved is from ss to ss+nn.
			Example: Exclude memory from 0x18690000-0x1869ffff
			         memmap=64K$0x18690000
			         or
			         memmap=0x10000$0x18690000
			Some bootloaders may need an escape character before '$',
			like Grub2, otherwise '$' and the following number
			will be eaten.

	memmap=nn[KMG]!ss[KMG,EARLY]
			[KNL,X86] Mark specific memory as protected.
			Region of memory to be used, from ss to ss+nn.
			The memory region may be marked as e820 type 12 (0xc)
			and is NVDIMM or ADR memory.

	memmap=<size>%<offset>-<oldtype>+<newtype>
			[KNL,ACPI,EARLY] Convert memory within the specified region
			from <oldtype> to <newtype>. If "-<oldtype>" is left
			out, the whole region will be marked as <newtype>,
			even if previously unavailable. If "+<newtype>" is left
			out, matching memory will be removed. Types are
			specified as e820 types, e.g., 1 = RAM, 2 = reserved,
			3 = ACPI, 12 = PRAM.

Memory corruption 검사와 hotplug metadata

3740-3781
매개변수적용·형식한국어 설명
memory_corruption_check=0/1[X86,EARLY]일부 BIOS가 suspend·resume 같은 동작에서 low 64K를 손상하는지 scan합니다. corruption을 감지하고 해당 memory를 kernel이 쓰지 못하게 합니다. 진단 도구이며 손상 위치가 반복 가능하면 `memmap=`으로 제외할 수 있습니다.
memory_corruption_check_size=size[X86,EARLY], 기본 low `64K`corruption을 검사하고 일반 사용에서 제외할 low-memory 크기를 바꿉니다.
memory_corruption_check_period=seconds[X86,EARLY], 기본 `60`corruption 검사 주기를 바꾸며 0은 periodic check를 끕니다.
memory_hotplug.memmap_on_memory[KNL,X86,ARM] `{ on | off }`, 기본 `off`runtime hotplug memory의 `struct page` 등 내부 vmemmap metadata를 hot-added memory 자체에서 할당해 추가 memory 없이 큰 hot-add를 가능하게 합니다. 이 vmemmap page는 `hugetlb_free_vmemmap`이 켜져도 optimize할 수 없습니다. 작은 memory block 같은 구성에서 GB급 allocation에 영향을 줄 수 있어 기본 off입니다. 상태는 `/sys/module/memory_hotplug/parameters/memmap_on_memory`에서 읽습니다. 켜도 효과가 없는 몇 가지 경우가 있습니다.

Memtest, microcode와 MINI2440

3782-3847
매개변수적용·형식한국어 설명
memtest=[KNL,X86,ARM,M68K,PPC,RISCV,EARLY] `<integer>`, 기본 `0`memory test pass 수입니다. pass마다 pattern set에서 다른 pattern을 골라 memory를 채우고 검증하며 발견한 bad region을 예약합니다. 0은 끕니다.
mem_encrypt=[X86-64] `{ on | off }`, 기본 `off`AMD Secure Memory Encryption(SME)을 켜거나 끕니다. 활성화 조건은 `Documentation/virt/kvm/x86/amd-memory-encryption.rst`를 참조하십시오.
mem_sleep_default=[SUSPEND] `{ s2idle | shallow | deep }`기본 suspend mode입니다. `s2idle`은 Suspend-To-Idle, `shallow`는 지원되는 Power-On Suspend 등가 mode, `deep`은 Suspend-To-RAM 등가 mode입니다. `Documentation/admin-guide/pm/sleep-states.rst`를 참조하십시오.
mfgptfix[X86-32]BIOS가 workaround를 잘못 적용한 AMD Geode platform의 MFGPT timer를 수정합니다. TinyBIOS 0.98은 영향받고 0.99는 사용자가 workaround를 끌 수 있어 수정됐습니다.
mga=[HW,DRM]Matrox DRM 관련 설정입니다.
microcode=[X86] 쉼표 구분 option`base_rev=X`는 debug mode에서 각 thread의 base microcode revision을 u32 X로 정합니다. `dis_ucode_ldr`는 loader를 끕니다. `force_minrev`는 runtime loader의 minimal revision 강제를 켜거나 끕니다.
mini2440=[ARM,HW,KNL] `[0..2][b][c][t]`, 기본 `0tb`0은 3.5inch TFT, 1은 7inch TFT, 2는 1024x768 VGA shield입니다. screen size를 생략하면 TFT driver를 load하지 않고 framebuffer를 구성하지 않습니다. `b`는 backlight를 켜 VESA blanking code와 GPIO LED에 연결하며 VGA shield에서는 필요 없습니다. `c`는 s3c camera interface를 켭니다. `t`는 touchscreen 지원 예약 문자입니다. mainstream kernel 2.6.30 당시 지원되지 않았고 예비 port는 원문 URL에 있습니다.
	mini2440=	[ARM,HW,KNL]
			Format:[0..2][b][c][t]
			Default: "0tb"
			MINI2440 configuration specification:
			0 - The attached screen is the 3.5" TFT
			1 - The attached screen is the 7" TFT
			2 - The VGA Shield is attached (1024x768)
			Leaving out the screen size parameter will not load
			the TFT driver, and the framebuffer will be left
			unconfigured.
			b - Enable backlight. The TFT backlight pin will be
			linked to the kernel VESA blanking code and a GPIO
			LED. This parameter is not necessary when using the
			VGA shield.
			c - Enable the s3c camera interface.
			t - Reserved for enabling touchscreen support. The
			touchscreen support is not enabled in the mainstream
			kernel as of 2.6.30, a preliminary port can be found
			in the "bleeding edge" mini2440 support kernel at
			https://repo.or.cz/w/linux-2.6/mini2440.git

통합 CPU 취약점 완화

3848-3912
매개변수적용·형식한국어 설명
mitigations=[X86,PPC,S390,ARM64,EARLY] `{ off | auto | auto,nosmt }``CPU_MITIGATIONS=y`로 빌드했을 때만 지원되는 architecture-independent 완화 묶음입니다. `off`는 성능을 높이는 대신 optional CPU 완화를 모두 꺼 여러 취약점에 노출할 수 있습니다. ARM64의 KASLR이 없을 때 `kpti=0`, x86의 GDS·ITS·NX huge page·L1TF·MDS·MMIO stale data·RFDS·Retbleed·Speculative RAS Stack Overflow·Spectre BHI·Spectre v2 user·SRBDS·TAA·VMSCAPE 완화 off, PPC의 entry·uaccess flush off, S390 branch prediction 설정, 여러 architecture의 PTI·Spectre·SSBD 설정 등을 조합합니다. 단 `kvm.nx_huge_pages=force`이면 해당 완화에는 영향이 없습니다. 기본 `auto`는 모든 취약점을 완화하되 SMT가 취약해도 유지해 kernel upgrade에서 SMT가 갑자기 꺼지는 일을 피합니다. `auto,nosmt`는 필요하면 SMT까지 꺼 완전 완화를 택하며 `l1tf=flush,nosmt`, `mds=full,nosmt`, `tsx_async_abort=full,nosmt`, `mmio_stale_data=full,nosmt`, `retbleed=auto,nosmt`와 같습니다. x86은 뒤에 `Documentation/admin-guide/hw-vuln/attack_vector_controls.rst`의 attack-vector control도 추가할 수 있습니다.
	mitigations=
			[X86,PPC,S390,ARM64,EARLY] Control optional mitigations for
			CPU vulnerabilities.  This is a set of curated,
			arch-independent options, each of which is an
			aggregation of existing arch-specific options.

			Note, "mitigations" is supported if and only if the
			kernel was built with CPU_MITIGATIONS=y.

			off
				Disable all optional CPU mitigations.  This
				improves system performance, but it may also
				expose users to several CPU vulnerabilities.
				Equivalent to: if nokaslr then kpti=0 [ARM64]
					       gather_data_sampling=off [X86]
					       indirect_target_selection=off [X86]
					       kvm.nx_huge_pages=off [X86]
					       l1tf=off [X86]
					       mds=off [X86]
					       mmio_stale_data=off [X86]
					       no_entry_flush [PPC]
					       no_uaccess_flush [PPC]
					       nobp=0 [S390]
					       nopti [X86,PPC]
					       nospectre_bhb [ARM64]
					       nospectre_v1 [X86,PPC]
					       nospectre_v2 [X86,PPC,S390,ARM64]
					       reg_file_data_sampling=off [X86]
					       retbleed=off [X86]
					       spec_rstack_overflow=off [X86]
					       spec_store_bypass_disable=off [X86,PPC]
					       spectre_bhi=off [X86]
					       spectre_v2_user=off [X86]
					       srbds=off [X86,INTEL]
					       ssbd=force-off [ARM64]
					       tsx_async_abort=off [X86]
					       vmscape=off [X86]

				Exceptions:
					       This does not have any effect on
					       kvm.nx_huge_pages when
					       kvm.nx_huge_pages=force.

			auto (default)
				Mitigate all CPU vulnerabilities, but leave SMT
				enabled, even if it's vulnerable.  This is for
				users who don't want to be surprised by SMT
				getting disabled across kernel upgrades, or who
				have other ways of avoiding SMT-based attacks.
				Equivalent to: (default behavior)

			auto,nosmt
				Mitigate all CPU vulnerabilities, disabling SMT
				if needed.  This is for users who always want to
				be fully mitigated, even if it means losing SMT.
				Equivalent to: l1tf=flush,nosmt [X86]
					       mds=full,nosmt [X86]
					       tsx_async_abort=full,nosmt [X86]
					       mmio_stale_data=full,nosmt [X86]
					       retbleed=auto,nosmt [X86]

			[X86] After one of the above options, additionally
			supports attack-vector based controls as documented in
			Documentation/admin-guide/hw-vuln/attack_vector_controls.rst

Memory-init log와 MMIO stale data

3913-3954
매개변수적용·형식한국어 설명
mminit_loglevel=[KNL,EARLY], `CONFIG_DEBUG_MEMORY_INIT`, `0..4`추가 memory initialization check의 log verbosity입니다. 0은 log를 끄고 4는 모두 기록합니다. `KERN_DEBUG`로 출력되므로 `loglevel=8`도 필요할 수 있습니다.
mmio_stale_data=[X86,INTEL,EARLY] `{ full | full,nosmt | off }`, 기본 `full`Processor MMIO Stale Data 취약점 완화를 제어합니다. MMIO operation 뒤 data가 노출될 수 있고 MDS·TAA와 같은 CPU buffer에서 시작하거나 끝날 수 있어 해당 buffer를 clear하는 방식으로 완화합니다. `full`은 취약 CPU에서 완화, `full,nosmt`는 완화와 SMT 비활성화, `off`는 조건 없이 끕니다. MDS·TAA 영향 machine에서는 같은 mechanism의 active 완화가 `off`를 막을 수 있어 완전히 끄려면 `mds=off`와 `tsx_async_abort=off`도 필요합니다. `Documentation/admin-guide/hw-vuln/processor_mmio_stale_data.rst`를 참조하십시오.
	mmio_stale_data=
			[X86,INTEL,EARLY] Control mitigation for the Processor
			MMIO Stale Data vulnerabilities.

			Processor MMIO Stale Data is a class of
			vulnerabilities that may expose data after an MMIO
			operation. Exposed data could originate or end in
			the same CPU buffers as affected by MDS and TAA.
			Therefore, similar to MDS and TAA, the mitigation
			is to clear the affected CPU buffers.

			This parameter controls the mitigation. The
			options are:

			full       - Enable mitigation on vulnerable CPUs

			full,nosmt - Enable mitigation and disable SMT on
				     vulnerable CPUs.

			off        - Unconditionally disable mitigation

			On MDS or TAA affected machines,
			mmio_stale_data=off can be prevented by an active
			MDS or TAA mitigation as these vulnerabilities are
			mitigated with the same mechanism so in order to
			disable this mitigation, you need to specify
			mds=off and tsx_async_abort=off too.

			Not specifying this option is equivalent to
			mmio_stale_data=full.

			For details see:
			Documentation/admin-guide/hw-vuln/processor_mmio_stale_data.rst

Module probe·signature와 mouse

3955-3996
매개변수적용·형식한국어 설명
<module>.async_probe[=<bool>][KNL]특정 module의 asynchronous probe를 제어합니다. bool을 생략하거나 잘못된 bool이면 켜고, 유효한 값이면 그 값대로 켜거나 끕니다. `module.async_probe`도 참조하십시오.
module.async_probe=<bool>[KNL]true면 module이 기본적으로 async probe를 사용합니다. 특정 module의 `<module>.async_probe`를 함께 지정하면 그 module에는 개별 설정이 우선합니다.
module.enable_dups_trace[KNL], `CONFIG_MODULE_DEBUG_AUTOLOAD_DUPS`중복 `request_module()` 호출이 `pr_warn()` 대신 `WARN_ON()`을 일으키게 합니다. `MODULE_DEBUG_AUTOLOAD_DUPS_TRACE`가 설정되면 항상 WARN_ON을 내므로 효과가 없습니다.
module.sig_enforce[KNL], `CONFIG_MODULE_SIG`유효한 signature가 없는 module load를 실패시킵니다. `CONFIG_MODULE_SIG_FORCE`이면 언제나 강제되므로 효과가 없습니다.
module_blacklist=[KNL] 쉼표 구분 module 목록나열한 module을 load하지 않습니다. 문제 module debugging에 유용합니다.
mousedev.tap_time=[MOUSE] `<msecs>`absolute mode touchpad에서 손가락 접촉부터 떼기까지 tap으로 보아 left click으로 보고할 최대 시간입니다.
mousedev.xres=[MOUSE]tablet처럼 absolute coordinate를 보고하는 장치의 수평 screen resolution입니다.
mousedev.yres=[MOUSE]tablet처럼 absolute coordinate를 보고하는 장치의 수직 screen resolution입니다.

Movable memory, MTD와 MTRR

3997-4051
매개변수적용·형식한국어 설명
movablecore=[KNL,X86,PPC,EARLY] `nn[KMGTPE] | nn%``kernelcore=`의 보완으로 migratable allocation에 쓸 memory 양을 정합니다. 둘 다 지정하면 kernelcore는 지정값 이상일 수 있습니다. movablecore만 지정하면 모든 allocation에 쓸 memory가 너무 적지 않도록 주의해야 합니다.
movable_node[KNL,EARLY]hotpluggable memory NUMA node를 movable로 만들어 해당 memory를 movable allocation에만 사용하게 합니다. 거의 모든 kernel allocation을 배제하므로 주의해야 합니다.
MTD_Partition=[MTD] `<name>,<region-number>,<size>,<offset>`MTD partition을 지정합니다.
MTD_Region=[MTD] `<name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>]`MTD region을 지정합니다.
mtdparts=[MTD]`drivers/mtd/parsers/cmdlinepart.c`를 참조하십시오.
mtouchusb.raw_coordinates=[HW] `{ y | n }`, 기본 `y`MicroTouch USB driver가 raw coordinate를 쓰게 하거나 `n`으로 cooked coordinate를 쓰게 합니다.
mtrr=debug[X86,EARLY]boot 때 MTRR register 관련 debug 정보를 출력합니다.
mtrr_chunk_size=nn[KMG][X86,EARLY]MTRR cleanup에서 hole, 즉 UC entry를 담을 수 있는 가장 큰 연속 chunk 크기입니다.
mtrr_gran_size=nn[KMG][X86,EARLY], 기본 `1`MTRR block granularity입니다. 큰 값은 작은 alignment가 MTRR을 소모하는 일을 막을 수 있습니다.
mtrr_spare_reg_nr=n[X86,EARLY] `<integer>`, 범위 `0..7`, 기본 `1`MTRR cleanup에 남겨둘 spare entry 수입니다. graphic card가 더 필요하면 2 이상으로 설정합니다.
multitce=off[PPC]여러 TCE entry를 한 번에 갱신하는 pSeries firmware 기능을 끕니다.

Network device와 conntrack

4052-4071
매개변수적용·형식한국어 설명
n2=[NET]SDL Inc. RISCom/N2 synchronous serial card 설정입니다.
netdev=[NET] `<irq>,<io>,<mem_start>,<mem_end>,<name>`network device parameter입니다. `mem_start`는 driver별 다른 의미로 자주 재사용되며, 필요한 경우 각 driver source에만 문서화됩니다.
netpoll.carrier_timeout=[NET] 초, 기본 `4`netpoll이 carrier를 기다리는 시간입니다.
nf_conntrack.acct=[NETFILTER] `{ 0 | 1 }`, 기본 `0`connection-tracking flow accounting을 끄거나 켭니다.

NFS client cache와 session

4072-4134
매개변수적용·형식한국어 설명
nfs.cache_getent=[NFS] pathNFS client cache entry를 갱신하는 program pathname입니다.
nfs.cache_getent_timeout=[NFS]cache entry 갱신 시도가 실패로 간주되는 timeout입니다.
nfs.callback_nr_threads=[NFSv4]NFSv4 callback request 처리에 client가 배정할 전체 thread 수입니다.
nfs.callback_tcpport=[NFS]NFSv4 callback channel이 listen할 TCP port입니다.
nfs.delay_retrans=[NFS] `>= 0`server가 `NFS4ERR_DELAY`를 응답한 뒤 NFSv4 client가 `EAGAIN`을 반환하기 전 retry 수입니다. `softerr` mount option이 켜졌을 때만 적용됩니다.
nfs.enable_ino64=[NFS], 기본 enabled64bit inode 번호를 사용합니다. 0이면 `readdir()`·`stat()`에 full 64bit 대신 가짜 32bit inode를 반환합니다.
nfs.idmap_cache_timeout=[NFS]idmapper cache entry의 최대 수명입니다.
nfs.max_session_cb_slots=[NFSv4.1]client가 callback channel에 배정할 최대 session slot 수로 특정 server에 대해 병렬 처리할 callback 최대 수를 정합니다.
nfs.max_session_slots=[NFSv4.1]client가 server와 협상할 최대 session slot 수로 동시 RPC request 수를 제한합니다. `max_tcp_slot_table_limit`보다 높게 둘 실익은 거의 없습니다.
nfs.nfs4_disable_idmapping=[NFSv4], 기본 `1``sec=sys` mount에서 RPC authentication과 NFS operation 모두 numeric uid·gid를 쓰게 해 idmapping을 사실상 끄고 NFSv2·v3에서 NFSv4로 migration하기 쉽게 합니다. 지원하지 않는 server는 client가 자동 감지해 idmapper로 fallback합니다. 0은 이 동작을 끕니다.

NFS identity, lock recovery와 server

4135-4198
매개변수적용·형식한국어 설명
nfs.nfs4_unique_id=[NFSv4] stringNFSv4 client가 `nfs_client_id4`에 넣을 고정 unique identification string입니다. 보통 설치 때 만든 UUID입니다.
nfs.recover_lost_locks=[NFSv4] `{ 0 | 1 }`, 기본 `0`server lease timeout으로 잃은 lock을 복구하려 시도합니다. lock을 잃은 뒤 file이 그대로라는 보장이 없어 data corruption 위험이 있습니다. 1은 legacy recovery를 켭니다.
nfs.send_implementation_id=[NFSv4.1], 기본 enabled`exchange_id` request에 client implementation 식별 정보를 보냅니다. 0은 보내지 않습니다.
nfs4.layoutstats_timer=[NFSv4.2] 초pNFS metadata server로 layoutstats를 보내는 주기를 바꿉니다. 0은 layout driver 기본값, non-zero는 전송 최소 간격입니다.
nfsd.inter_copy_offload_enable=[NFSv4.2] `{ 0 | 1 }`1이면 이 server가 destination인 server-to-server copy를 지원합니다.
nfsd.nfs4_disable_idmapping=[NFSv4], 기본 `1`auth_sys client에 numeric uid·gid만 반환하고 받아들여 NFSv2·v3 migration을 쉽게 합니다.
nfsd.nfsd4_ssc_umount_timeout=[NFSv4.2] msserver-to-server copy destination일 때 knfsd가 임시 mount한 source server를 재사용하도록 cache하고, 지정 시간 동안 쓰지 않으면 버립니다.
nfsaddrs=[NFS], 폐기됨`ip=`를 사용하십시오. `Documentation/admin-guide/nfs/nfsroot.rst`를 참조하십시오.
nfsroot=[NFS]diskless machine의 NFS root filesystem입니다. `Documentation/admin-guide/nfs/nfsroot.rst`를 참조하십시오.
nfsrootdebug[NFS]nfsroot debug message를 켭니다. `Documentation/admin-guide/nfs/nfsroot.rst`를 참조하십시오.

NMI backtrace와 watchdog

4199-4224
매개변수적용·형식한국어 설명
nmi_backtrace.backtrace_idle[KNL]NMI stack-backtrace request에 idle CPU stack도 dump합니다.
nmi_debug=[KNL,SH] `[state][,regs][,debounce][,die]`NMI가 trigger될 때 수행할 동작을 하나 이상 지정합니다.
nmi_watchdog=[KNL,BUGS=X86] `[panic,][nopanic,][rNNN,][0|1]`SMP kernel hard-lockup detector를 제어합니다. 0은 끄고 1은 켭니다. `rNNN`은 raw perf event `0xNNN`을 사용합니다. `panic`은 watchdog timeout 때 panic하고 `nopanic`은 `CONFIG_BOOTPARAM_HARDLOCKUP_PANIC`이어도 panic하지 않습니다. hard·soft detector를 모두 끄려면 `nowatchdog`를 씁니다. `panic=...`와 함께 빠른 reboot가 필요할 때 유용합니다. 실행 중 `nmi_watchdog`와 `hardlockup_panic` sysctl로 접근할 수 있습니다.
	nmi_watchdog=	[KNL,BUGS=X86] Debugging features for SMP kernels
			Format: [panic,][nopanic,][rNNN,][num]
			Valid num: 0 or 1
			0 - turn hardlockup detector in nmi_watchdog off
			1 - turn hardlockup detector in nmi_watchdog on
			rNNN - configure the watchdog with raw perf event 0xNNN

			When panic is specified, panic when an NMI watchdog
			timeout occurs (or 'nopanic' to not panic on an NMI
			watchdog, if CONFIG_BOOTPARAM_HARDLOCKUP_PANIC is set)
			To disable both hard and soft lockup detectors,
			please see 'nowatchdog'.
			This is useful when you use a panic=... timeout and
			need the box quickly up again.

			These settings can be accessed at runtime via
			the nmi_watchdog and hardlockup_panic sysctls.

CPU·APIC와 console 비활성화

4225-4296
매개변수적용·형식한국어 설명
no387[BUGS=X86-32]387 math coprocessor가 있어도 387 math emulation library를 씁니다.
no4lvl[RISCV,EARLY]4·5 level paging을 끄고 3 level paging을 강제합니다.
no5lvl[X86-64,RISCV,EARLY]5 level paging을 끄고 4 level paging을 강제합니다.
noalign[KNL,ARM]ARM alignment 관련 동작을 끕니다.
noapic[SMP,APIC,EARLY]system의 IOAPIC를 사용하지 않습니다.
noapictimer[APIC,X86]APIC timer를 설정하지 않습니다.
noautogroupschedulerscheduler automatic task group 생성을 끕니다.
nocache[ARM,EARLY]ARM cache 사용을 끕니다.
no_console_suspend[HW]suspend·hibernate 때 console을 절대 suspend하지 않아 나머지 system이 sleep으로 들어가는 동안 driver suspend·resume hook debug message를 보낼 수 있게 합니다. 모든 console에서 신뢰할 수 있는 것은 아니지만 serial·VGA에서는 동작합니다. printk module parameter `console_suspend`, 보통 `/sys/module/printk/parameters/console_suspend`에서 동적으로 바꿀 수 있습니다.
no_debug_objects[KNL,EARLY]object debugging을 끕니다.
nodsp[SH]boot 때 hardware DSP를 끕니다.
noefi[EFI,EARLY]EFI runtime service 지원을 끕니다.
no_entry_flush[PPC,EARLY]kernel 진입 때 L1-D cache를 flush하지 않습니다.
noexec32[X86-64] `{ on | off }`32bit executable에만 적용됩니다. 기본 `on`은 read mapping에 execute를 암시하지 않는 non-executable mapping을 켜고, `off`는 read가 execute를 암시하게 합니다.
no_file_capssecurityfile capability를 인정하지 않습니다. file이 privilege로 실행되는 길은 setuid root 또는 root 실행뿐입니다.
nofpu[MIPS,SH]boot 때 hardware FPU를 끕니다.
nofsgsbase[X86]FSGSBASE instruction을 끕니다.
nofxsr[BUGS=X86-32]x86 extended floating-point register save·restore를 끄고 task switch 때 legacy floating-point register만 저장합니다.
nogbpages[X86]kernel direct mapping에 GB page를 쓰지 않습니다.
no_hash_pointers[KNL,EARLY]`hash_pointers=never`의 alias입니다.
nohibernate[HIBERNATION]hibernation과 resume을 끕니다.

Idle·tick·initrd와 interrupt 비활성화

4297-4363
매개변수적용·형식한국어 설명
nohlt[ARM,ARM64,MICROBLAZE,MIPS,PPC,RISCV,SH]`do_idle()`에서 busy-wait하고 `arch_cpu_idle()` 구현을 쓰지 않습니다. `CONFIG_GENERIC_IDLE_POLL_SETUP`이 필요합니다. sleep·WFI instruction이 잘못 동작하는 platform, sleep instruction 영향 측정, JTAG debugging에 유용합니다.
nohpet[X86]HPET timer를 쓰지 않습니다.
nohugeiomap[KNL,X86,PPC,ARM64,EARLY]kernel huge I/O mapping을 끕니다.
nohugevmalloc[KNL,X86,PPC,ARM64,EARLY]kernel huge vmalloc mapping을 끕니다.
nohz=[KNL] `{ on | off }`, 기본 `on`boot 때 dynamic tick을 켜거나 끕니다.
nohz_full=[KNL,BOOT,SMP,ISOL] CPU 목록`CONFIG_NO_HZ_FULL=y` kernel에서 가능한 때 tick을 멈출 CPU 목록입니다. boot CPU는 timekeeping을 위해 범위 밖으로 강제됩니다. 나열한 CPU의 RCU callback은 `rcu_nocbs=`처럼 offload됩니다. 이 값은 `CONFIG_RCU_NOCB_CPU_DEFAULT_ALL`보다 우선합니다.
noinitrd[RAM]구성된 initial RAM disk를 load하지 않습니다.
nointremap[X86-64,Intel-IOMMU,EARLY], 폐기됨interrupt remapping을 켜지 않습니다. 대신 `intremap=off`를 사용하십시오.
noinvpcid[X86,EARLY]INVPCID CPU feature를 끕니다.
noiotrap[SH]trap된 I/O port access를 끕니다.
noirqdebug[X86-32]처리되지 않은 interrupt source를 찾아 끄는 code를 비활성화합니다.
noisapnp[ISAPNP]ISA PnP code를 끕니다.
nokaslr[KNL,EARLY], `CONFIG_RANDOMIZE_BASE`kernel과 module base offset ASLR을 끕니다.
no-kvmapf[X86,KVM,EARLY]paravirtualized asynchronous page fault 처리를 끕니다.
no-kvmclock[X86,KVM,EARLY]paravirtualized KVM clock driver를 끕니다.
nolapic[X86-32,APIC,EARLY]local APIC를 켜거나 사용하지 않습니다.
nolapic_timer[X86-32,APIC,EARLY]local APIC timer를 사용하지 않습니다.
nomce[X86-32]Machine Check Exception을 끕니다.
nomfgpt[X86-32]AMD Geode의 Multi-Function General Purpose Timer를 사용하지 않습니다.
	nohz_full=	[KNL,BOOT,SMP,ISOL]
			The argument is a cpu list, as described above.
			In kernels built with CONFIG_NO_HZ_FULL=y, set
			the specified list of CPUs whose tick will be stopped
			whenever possible. The boot CPU will be forced outside
			the range to maintain the timekeeping.  Any CPUs
			in this list will have their RCU callbacks offloaded,
			just as if they had also been called out in the
			rcu_nocbs= boot parameter.

			Note that this argument takes precedence over
			the CONFIG_RCU_NOCB_CPU_DEFAULT_ALL option.

Display·PV·보안 기능 비활성화

4364-4423
매개변수적용·형식한국어 설명
nomodesetdisplaykernel modesetting을 끕니다. firmware가 미리 만든 display mode·framebuffer를 밀어낼 수 있는 DRM·fbdev driver는 load하지 않아 system framebuffer만 남습니다. display mode 변경과 accelerated rendering도 하지 않습니다. error fallback이나 test·debugging에 유용합니다.
nomodulemodulemodule load를 끕니다.
nonmi_ipi[X86]panic·reboot 때 다른 CPU shutdown에 NMI IPI 대신 `REBOOT_VECTOR` IRQ를 씁니다.
nopat[X86,EARLY]page table의 Page Attribute Table extension 지원을 끕니다.
nopcid[X86-64,EARLY]PCID CPU feature를 끕니다.
nopku[X86]일부 Intel CPU의 Memory Protection Keys feature를 끕니다.
nopti[X86-64,EARLY]`pti=off`와 같습니다.
nopv=[X86,XEN,KVM,HYPER_V,VMWARE,EARLY]PV optimization과 driver를 끄고 XEN HVM·KVM·Hyper-V·VMware guest를 generic guest로 실행합니다.
nopvspin[X86,XEN,KVM,EARLY]lock contention 때 hypervisor가 guest를 idle하게 하는 PV qspinlock slow-path optimization을 끕니다.
norandmapsASLR`echo 0 > /proc/sys/kernel/randomize_va_space`와 같이 address-space randomization을 끕니다.
noreplace-smp[X86-32,SMP]SMP instruction을 UP alternative로 바꾸지 않습니다.
noresume[SWSUSP]resume을 끄고 원래 swap space를 복원합니다.
no-scroll[VGA]scrollback을 끕니다. F.H. Papenmeier의 Braillex ib80-piezo Braille reader에 필요합니다.
nosgx[X86-64,SGX,EARLY]Intel SGX kernel 지원을 끕니다.
nosmap[PPC,EARLY]processor가 지원해도 Supervisor Mode Access Prevention을 끕니다.
nosmep[PPC64s,EARLY]processor가 지원해도 Supervisor Mode Execution Prevention을 끕니다.

SMT·Spectre·watchdog 관련 비활성화

4424-4486
매개변수적용·형식한국어 설명
nosmp[SMP,EARLY]SMP kernel을 UP kernel처럼 동작시키고 IO APIC를 끕니다. legacy `maxcpus=0`입니다.
nosmt[KNL,MIPS,PPC,X86,S390,EARLY]symmetric multithreading을 끕니다. MIPS·PPC의 `nosmt`는 `smt=1`과 같습니다. `nosmt=force`는 sysfs로 되돌릴 수 없게 강제로 끕니다.
nosoftlockup[KNL]soft-lockup detector를 끕니다.
nospec_store_bypass_disable[HW,EARLY]Speculative Store Bypass 취약점의 모든 완화를 끕니다.
nospectre_bhb[ARM64,EARLY]Spectre-BHB branch-history injection 완화를 모두 끕니다. data leak을 허용할 수 있습니다.
nospectre_v1[X86,PPC,EARLY]Spectre Variant 1 bounds-check-bypass 완화를 끕니다. data leak이 가능해집니다.
nospectre_v2[X86,PPC_E500,ARM64,EARLY]Spectre Variant 2 indirect-branch-prediction 완화를 모두 끕니다. data leak을 허용할 수 있습니다.
no-steal-acc[X86,PV_OPS,ARM64,PPC/PSERIES,RISCV,LOONGARCH,EARLY]paravirtualized steal-time accounting의 scheduler 영향을 끕니다. steal time 자체는 계속 계산합니다.
nosync[HW,M68K]모든 장치의 sync negotiation을 끕니다.
no_timer_check[X86,APIC]고장 난 timer IRQ source, 즉 IO-APIC timer 검사 code를 끕니다. 일부 board의 잘못된 timer initialization을 우회할 수 있습니다.
no_uaccess_flush[PPC,EARLY]user data 접근 뒤 L1-D cache를 flush하지 않습니다.
novmcoredd[KNL,KDUMP], `CONFIG_PROC_VMCORE_DEVICE_DUMP`driver가 debug data를 vmcore에 제한 없이 붙이는 device dump를 끕니다. memory stress를 줄이지만 driver debug data를 잃습니다.
no-vmw-sched-clock[X86,PV_OPS,EARLY]paravirtualized VMware scheduler clock을 끄고 기본 clock을 씁니다.
nowatchdog[KNL]soft-lockup detector와 NMI hard-lockup watchdog을 모두 끕니다.
nowb[ARM,EARLY]ARM write buffer 관련 동작을 끕니다.
	nosmp		[SMP,EARLY] Tells an SMP kernel to act as a UP kernel,
			and disable the IO APIC.  legacy for "maxcpus=0".

	nosmt		[KNL,MIPS,PPC,EARLY] Disable symmetric multithreading (SMT).
			Equivalent to smt=1.

			[KNL,X86,PPC,S390] Disable symmetric multithreading (SMT).
			nosmt=force: Force disable SMT, cannot be undone
				     via the sysfs control file.

x2APIC·XSAVE와 CPU 상한

4487-4522
매개변수적용·형식한국어 설명
nox2apic[X86-64,APIC,EARLY]x2APIC mode를 켜지 않습니다. `IA32_XAPIC_DISABLE_STATUS` MSR의 `LEGACY_XAPIC_DISABLED` bit가 설정된 system에서는 무시됩니다.
noxsave[BUGS=X86]xsave를 통한 x86 extended register-state save·restore를 끄고 legacy floating-point·SSE state를 사용합니다.
noxsaveopt[X86]xsaveopt를 끄고 xsave를 씁니다. xsave는 modified optimization을 지원하지 않아 state 저장 성능이 낮아집니다.
noxsaves[X86]compacted xsave area의 xsaves·xrstors를 끄고 standard form의 xsaveopt·xrstor로 돌아갑니다. process별 xsave area가 더 많은 memory를 차지할 수 있습니다.
nr_cpus=[SMP,EARLY] `n >= 1`SMP kernel이 지원할 processor 최대 수입니다. boot 때 꽂힌 CPU보다 크게 정해 runtime에 n까지 물리 CPU를 추가할 수 있습니다. 이후 hotplug용 per-CPU variable memory를 boot 때 미리 할당합니다.
nr_uarts=[SERIAL]등록할 UART 최대 수입니다.

NUMA topology와 balancing

4523-4555
매개변수적용·형식한국어 설명
numa=off[KNL,ARM64,PPC,RISCV,SPARC,X86,EARLY]NUMA를 끄고 모든 memory를 덮는 단일 NUMA node만 만듭니다.
numa=fake=<size>[MG][KNL,ARM64,RISCV,X86,EARLY]memory 단위로 주면 system RAM 전체를 물리 node에 interleave된 지정 크기 fake node로 채웁니다.
numa=fake=<N>[KNL,ARM64,RISCV,X86,EARLY]정수 N으로 주면 system RAM 전체를 물리 node에 interleave된 N개 fake node로 채웁니다.
numa=fake=<N>U[KNL,ARM64,RISCV,X86,EARLY]정수 뒤 U를 붙이면 각 물리 node를 N개 emulated node로 나눕니다.
numa=noacpi[X86]NUMA setup에 SRAT table을 parse하지 않습니다.
numa=nohmat[X86]NUMA setup 또는 soft-reserved memory partitioning에 HMAT table을 parse하지 않습니다.
numa_balancing=[KNL,ARM64,PPC,RISCV,S390,X86] `{ enable | disable }`automatic NUMA balancing을 켜거나 끕니다.
numa_zonelist_order=[KNL,BOOT] `{ node | default }`NUMA zonelist 순서를 고릅니다. boot 뒤 sysctl로 바꿀 수 있으며 `Documentation/admin-guide/sysctl/vm.rst`를 참조하십시오.

OHCI·OLPC·OMAP·OneNAND와 oops

4556-4587
매개변수적용·형식한국어 설명
ohci1394_dma=early[HW,EARLY]ohci1394 드라이버를 통한 조기 디버깅을 활성화합니다. 자세한 내용은 `Documentation/core-api/debugging-via-ohci1394.rst`를 참조하십시오.
olpc_ec_timeout=[OLPC] 밀리초EC 명령을 보낸 뒤 ACK를 기다리는 제한 시간을 기본 20ms 대신 지정합니다. ACK 대기 중에는 인터럽트가 비활성화되므로 값을 지나치게 높이면 인터럽트가 유실될 수 있습니다.
omap_mux=[OMAP] `<mux_mode0.mode_name=value>...`부트로더가 설정한 핀 멀티플렉싱을 덮어씁니다. I2C bus2 예시는 `omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100`입니다.
onenand.bdry=[HW,MTD] `[die0_boundary][,die0_lock][,die1_boundary][,die1_lock]`Flex-OneNAND 경계를 구성합니다. `boundary`는 각 die에서 마지막 SLC 블록의 인덱스이며 나머지 블록은 MLC로 구성됩니다. `lock`은 경계 잠금 여부로, 1은 잠금이고 0은 잠금 해제입니다. 잠긴 경계는 다시 변경할 수 없습니다.
oops=panic[KNL,EARLY]oops가 발생할 때 항상 panic을 일으킵니다. 기본 동작은 해당 프로세스만 종료하지만 교착 가능성이 조금 남습니다. 이 옵션은 machine check exception에도 panic을 일으키며, `panic=30`과 함께 사용하면 재부팅을 유도할 수 있습니다.

페이지 할당·소유·poison·보고

4588-4614
매개변수적용·형식한국어 설명
page_alloc.shuffle=[KNL] bool, `CONFIG_SHUFFLE_PAGE_ALLOCATOR=y`page allocator의 free list 무작위화를 켜거나 끕니다. 현재 값은 `/sys/module/page_alloc/parameters/shuffle`에서 읽을 수 있습니다.
page_owner=[KNL,EARLY] `on`기본적으로 꺼져 있는 페이지별 할당자 정보 저장 기능을 부팅 시 활성화합니다.
page_poison=[KNL,EARLY] `{ off | on }`, `CONFIG_PAGE_POISONING=y`buddy allocator의 page poisoning 상태를 정합니다. `off`가 기본값이며 `on`은 poisoning을 활성화합니다.
page_reporting.page_reporting_order=[KNL] 정수보고할 최소 page order를 조정합니다. 값이 `MAX_PAGE_ORDER`보다 크면 page reporting이 비활성화됩니다.

panic 동작과 진단 출력

4615-4671
매개변수적용·형식한국어 설명
panic=[KNL] `<timeout>`panic 이후 동작을 초 단위로 정합니다. 양수는 지정 시간 뒤 재부팅, 0은 무한 대기, 음수는 즉시 재부팅입니다.
panic_on_taint=[KNL,EARLY] `<hex>[,nousertaint]``add_taint()`에 전달된 TAINT 플래그가 지정한 16진수 비트마스크와 겹치면 panic을 일으킵니다. `nousertaint`를 붙이면 사용자가 `/proc/sys/kernel/tainted`에 일치하는 플래그를 써서 강제로 crash시키는 것을 막습니다. 선택 가능한 플래그는 `Documentation/admin-guide/tainted-kernels.rst`를 참조하십시오.
panic_on_warn=1kernel 경고`WARN()`에서 계속 실행하는 대신 `panic()`을 호출합니다. WARN을 kdump로 수집할 때 유용합니다.
panic_print=비트마스크panic 시 출력할 정보를 조합합니다. bit 0은 모든 task, bit 1은 시스템 메모리, bit 2는 timer, bit 3은 `CONFIG_LOCKDEP` 사용 시 lock, bit 4는 ftrace buffer, bit 5는 panic 끝에 모든 kernel message를 console로 재생, bit 6은 지원되는 경우 모든 CPU backtrace, bit 7은 uninterruptible 상태의 task만 출력합니다. 출력량이 매우 많아 이전 log가 밀릴 수 있으므로 필요하면 `log_buf_len`도 키우십시오.
panic_sys_info=`val[,val...]``panic_print`의 사람이 읽기 쉬운 대안입니다. `tasks`는 모든 task, `mem`은 시스템 메모리, `timers`는 timer, `locks`는 `CONFIG_LOCKDEP` 사용 시 lock, `ftrace`는 ftrace buffer, `all_bt`는 지원되는 경우 모든 CPU backtrace, `blocked_tasks`는 uninterruptible 상태의 task만 출력합니다.
panic_console_replaypanic consolepanic 끝에 모든 kernel message를 console로 다시 출력합니다.

병렬 포트 키보드와 parport

4672-4700
매개변수적용·형식한국어 설명
parkbd.port=[HW] `<parport#>`, 기본 `0`키보드 어댑터가 연결된 병렬 포트 번호를 지정합니다.
parkbd.mode=[HW] `<mode>`병렬 포트 키보드 어댑터 모드를 0(XT) 또는 1(AT)로 정합니다. 기본값은 AT입니다.
parport=[HW,PPT] `{ 0 | auto | 0xBBB[,IRQ[,DMA]] }`병렬 포트를 지정하며 0은 비활성화합니다. `auto`는 감지된 IRQ와 DMA를 사용하게 합니다. base address, IRQ, DMA를 직접 지정할 수 있고 IRQ·DMA에는 숫자, 감지값을 쓰는 `auto`, FIFO를 피하는 `nofifo`를 사용할 수 있습니다. 명령행에 나온 순서대로 `parport0`부터 배정됩니다. 기본 동작은 충돌 가능성 때문에 감지된 IRQ·DMA를 무시합니다.
parport_init_mode=[HW,PPT] `[spp|ps2|epp|ecp|ecpepp]`VIA 병렬 포트의 동작 모드를 지정합니다. firmware에서 모드를 설정할 수 없어 SPP로 두는 Pegasos에 필요하며 686a와 8231 chip을 지원합니다.

legacy PATA 탐색과 oops 일시 정지

4701-4795
매개변수적용·형식한국어 설명
pata_legacy.all=[HW,LIBATA] 정수PCI PATA 장치를 찾지 못한 PCI 시스템에서 primary·secondary ISA 포트 범위를 탐색합니다. 0이 아닌 값으로 켜며 기본값은 꺼짐입니다.
pata_legacy.autospeed=[HW,LIBATA] 정수속도 변경을 감시하는 chip이 있음을 알립니다. 0이 아닌 값으로 켜며 기본값은 꺼짐입니다.
pata_legacy.ht6560a=[HW,LIBATA] `{ 1 | 2 | 3 }`HT 6560A를 primary channel, secondary channel, 두 channel 모두에 각각 1, 2, 3으로 지정합니다. 기본값은 꺼짐입니다.
pata_legacy.ht6560b=[HW,LIBATA] `{ 1 | 2 | 3 }`HT 6560B를 primary channel, secondary channel, 두 channel 모두에 각각 1, 2, 3으로 지정합니다. 기본값은 꺼짐입니다.
pata_legacy.iordy_mask=[HW,LIBATA] 정수 bitmaskchannel별 IORDY 허용 mask입니다. bit 0은 이 driver가 다루는 첫 legacy channel, bit 1은 두 번째 channel에 대응합니다. 보통 primary, secondary 순서지만 PCI bus 처리와 다른 driver 옵션 때문에 달라질 수 있습니다. 기본값은 모든 channel 허용입니다.
pata_legacy.opti82c46x=[HW,LIBATA] `{ 1 | 2 | 3 }`Opti 82c611A를 primary, secondary, 두 channel 모두에 각각 1, 2, 3으로 지정합니다. 기본값은 꺼짐입니다.
pata_legacy.opti82c611a=[HW,LIBATA] `{ 1 | 2 | 3 }`Opti 82c465MV를 primary, secondary, 두 channel 모두에 각각 1, 2, 3으로 지정합니다. 기본값은 꺼짐입니다.
pata_legacy.pio_mask=[HW,LIBATA] 정수 bitmaskautospeed 장치에서 허용할 PIO mode mask입니다. bit 0은 mode 0, bit 1은 mode 1처럼 대응하며 기본값은 모든 mode 허용입니다.
pata_legacy.probe_all=[HW,LIBATA] 정수PCI 시스템에서 세 번째 이후 ISA 포트 범위도 탐색합니다. 0이 아닌 값으로 켜며 기본값은 꺼짐입니다.
pata_legacy.probe_mask=[HW,LIBATA] 정수 bitmasklegacy ISA PATA 포트 탐색 mask입니다. 최대 여섯 포트 `0x1f0`, `0x170`, `0x1e8`, `0x168`, `0x1e0`, `0x160`을 지원하며 각 bit를 1로 두면 해당 포트의 탐색을 끕니다. bit 0은 첫 포트에 대응하고 기본값은 모든 지원 포트 탐색입니다.
pata_legacy.qdi=[HW,LIBATA] 정수0이 아닌 값이면 QDI controller를 탐색합니다. 기본값은 `CONFIG_PATA_QDI_MODULE`이면 1, 아니면 0입니다.
pata_legacy.winbond=[HW,LIBATA] 정수0이 아닌 값이면 Winbond controller를 탐색합니다. 1은 표준 I/O port `0x130`, 그 밖의 값은 직접 지정한 port(일반적으로 `0x1b0`)입니다. 기본값은 `CONFIG_PATA_WINBOND_VLB_MODULE`이면 1, 아니면 0입니다.
pata_platform.pio_mask=[HW,LIBATA] 정수 bitmask지원할 PIO mode mask입니다. bit 0은 mode 0, bit 1은 mode 1처럼 대응하며 기본값은 mode 0만 허용입니다.
pause_on_oops=<int>첫 oops를 출력한 뒤 지정한 시간 동안 모든 CPU를 멈춥니다. 연속 oops가 화면에서 너무 빨리 밀려날 때 사용합니다.

ISDN과 PCI 장치 지정 형식

4796-4821
매개변수적용·형식한국어 설명
pcbit=[HW,ISDN]PCBIT ISDN hardware driver 매개변수입니다.
pci=option[,option...][PCI,EARLY]PCI subsystem 옵션을 쉼표로 나열합니다. 일부 옵션의 `<pci_dev>`는 bus/device/function 주소 또는 configuration space ID로 특정 장치나 장치 집합을 고릅니다. 첫 형식의 주소는 hardware 추가, motherboard firmware 변경, 다른 kernel 매개변수에 따른 재번호화로 바뀔 수 있습니다. domain을 생략하면 0이며, base 주소 뒤에 여러 device/function 경로를 붙이면 재번호화에 더 강합니다. 두 번째 형식은 configuration space ID로 선택하므로 여러 장치와 일치할 수 있습니다.

`<pci_dev>`의 두 원문 형식은 다음과 같습니다.

				[<domain>:]<bus>:<dev>.<func>[/<dev>.<func>]*
				pci:<vendor>:<device>[:<subvendor>:<subdevice>]

PCI 탐색·설정 공간·IRQ 라우팅

4822-4890
매개변수적용·형식한국어 설명
`pci=earlydump`PCIkernel이 값을 바꾸기 전에 PCI configuration space를 dump합니다.
`pci=off`[X86]PCI bus를 탐색하지 않습니다.
`pci=bios`[X86-32]PCI BIOS 사용을 강제하고 hardware에 직접 접근하지 않습니다. 비표준 PCI host bridge에서 사용할 수 있습니다.
`pci=nobios`[X86-32]PCI BIOS 사용을 금지하고 직접 hardware 접근만 허용합니다. 부팅 crash가 BIOS 때문이라고 의심될 때 사용할 수 있습니다.
`pci=conf1`[X86]PCI Configuration Access Mechanism 1을 강제합니다. config address는 I/O port `0xCF8`, data는 `0xCFC`이며 둘 다 32bit입니다.
`pci=conf2`[X86]Mechanism 2를 강제합니다. `0xCF8`은 function용 8bit port, `0xCFA`는 bus 번호용 8bit port이며 configuration space는 `0xC000-0xCFFF`를 통해 접근합니다. 자세한 내용은 원문의 `http://wiki.osdev.org/PCI`를 참조하십시오.
`pci=noaer`[PCIE]`PCIEAER` kernel 구성이 켜져 있어도 PCIe Advanced Error Reporting을 사용하지 않습니다.
`pci=nodomains`[PCI]ACPI에서 PCI segment라 부르는 여러 PCI root domain 지원을 끕니다.
`pci=nommconf`[X86]PCI configuration에 MMCONFIG를 사용하지 않습니다.
`pci=check_enable_amd_mmconf`[X86]AMD family 10h CPU에서 PCI configuration space에 대한 올바른 MMIO 접근을 검사하고 활성화합니다.
`pci=nomsi`[MSI]`PCI_MSI` kernel 구성이 켜져 있어도 시스템 전체에서 MSI interrupt를 사용하지 않습니다.
`pci=noioapicquirk`[APIC]모든 boot interrupt quirk를 끕니다. boot IRQ를 계속 켜 두는 안전 옵션이지만 일반적으로 필요하지 않습니다.
`pci=ioapicreroute`[APIC]boot IRQ를 끌 수 없는 bridge에서 primary IO-APIC으로 boot IRQ를 reroute하여 IRQ mask 시 발생하는 spurious IRQ를 막습니다.
`pci=noioapicreroute`[APIC]boot IRQ를 끌 수 없는 chipset에 연결된 IRQ의 boot IRQ equivalent를 쓰는 workaround를 끕니다. `ioapicreroute`의 반대입니다.
`pci=biosirq`[X86-32]PCI BIOS 호출로 interrupt routing table을 얻습니다. 일부 장치에서는 hang을 일으키지만 kernel이 IRQ를 배정하지 못하거나 secondary PCI bus를 찾지 못할 때 유일한 방법일 수 있습니다.
`pci=rom`[X86]expansion ROM에 address space를 배정합니다. 일부 장치가 ROM과 다른 resource의 address decoder를 공유하므로 주의해야 합니다.
`pci=norom`[X86]BIOS가 이미 주소 범위를 배정하지 않은 expansion ROM에는 address space를 배정하지 않습니다.
`pci=nobar`[X86]BIOS가 배정하지 않은 BAR에는 address space를 배정하지 않습니다.
`pci=irqmask=0xMMMM`[X86]PCI 장치에 자동 배정할 수 있는 IRQ mask를 정합니다. ISA card의 IRQ를 제외할 때 사용할 수 있습니다.
`pci=pirqaddr=0xAAAAA`[X86]PIRQ table이 일반적인 `F0000h-100000h` 범위 밖에 있을 때 그 물리 주소를 지정합니다.

PCI bus 번호·window·payload

4891-4956
매개변수적용·형식한국어 설명
`pci=lastbus=N`[X86]bus N까지 모두 scan합니다. kernel이 secondary bus를 찾지 못할 때 마지막 bus를 명시합니다.
`pci=assign-busses`[X86]firmware가 정한 값을 덮어쓰고 모든 PCI bus 번호를 kernel이 직접 배정합니다.
`pci=usepirqmask`[X86]BIOS `$PIR` table의 가능한 IRQ mask를 따릅니다. 일부 HP Pavilion N5400·Omnibook XE3처럼 손상된 BIOS에 필요하며 ACPI IRQ routing이 켜져 있으면 효과가 없습니다.
`pci=noacpi`[X86]IRQ routing과 PCI scanning에 ACPI를 사용하지 않습니다.
`pci=use_crs`[X86]ACPI의 PCI host bridge window 정보를 사용합니다. 2008년 이후 BIOS에서는 기본으로 켜져 있으며, 별도 지정이 필요했다면 bug를 보고해야 합니다.
`pci=nocrs`[X86]ACPI의 PCI host bridge window를 무시합니다. 이 옵션이 필요하면 bug를 보고해야 합니다.
`pci=use_e820`[X86]E820 reservation으로 PCI host bridge window 일부를 제외합니다. 잘못된 host bridge `_CRS` method를 우회하며 필요하면 `linux-pci@vger.kernel.org`에 보고해야 합니다.
`pci=no_e820`[X86]PCI host bridge window에 E820 reservation을 무시합니다. 최신 hardware의 기본값이며 별도 지정이 필요하면 `linux-pci@vger.kernel.org`에 보고해야 합니다.
`pci=routeirq`PCI모든 PCI 장치에 IRQ routing을 수행합니다. 보통 `pci_enable_device()`에서 처리하므로 이를 호출하지 않는 잘못된 driver를 위한 임시 workaround입니다.
`pci=skip_isa_align`[X86]I/O 시작 주소를 ISA 기준으로 정렬하지 않아 더 많은 PCI card를 처리할 수 있게 합니다.
`pci=noearly`[X86]조기 type 1 scanning을 하지 않습니다. configuration space 읽기에서 machine check가 나는 board에 도움이 될 수 있지만 여러 workaround가 꺼지고 일부 IOMMU driver가 동작하지 않습니다.
`pci=bfsort`PCIPCI 장치를 breadth-first 순서로 정렬하여 Linux 2.4 이하와 호환되는 장치 순서를 만듭니다.
`pci=nobfsort`PCIPCI 장치를 breadth-first 순서로 정렬하지 않습니다.
`pci=pcie_bus_tune_off`PCIePCIe MPS(Max Payload Size) tuning을 끄고 BIOS가 설정한 MPS를 그대로 사용합니다.
`pci=pcie_bus_safe`PCIeroot complex 아래 모든 장치가 지원하는 범위에서 가장 큰 MPS를 각 장치에 설정합니다.
`pci=pcie_bus_perf`PCIeparent bus를 기준으로 허용되는 가장 큰 MPS를 설정하고, 장치나 bus가 지원하는 MPS보다 크지 않은 최대 MRRS(Max Read Request Size)를 설정해 성능을 높입니다.
`pci=pcie_bus_peer2peer`PCIe모든 장치의 MPS를 보장된 최소값 128B로 맞춥니다. 성능이 낮아질 수 있지만 모든 장치 쌍의 peer-to-peer DMA와 hot-add 장치 동작을 보장합니다.
`pci=cbiosize=nn[KMG]`CardBus, 기본 256 bytesCardBus bridge I/O window에 예약할 고정 bus space를 정합니다.
`pci=cbmemsize=nn[KMG]`CardBus, 기본 64MBCardBus bridge memory window에 예약할 고정 bus space를 정합니다.

PCI resource 재배치와 hotplug

4957-5014
매개변수적용·형식한국어 설명
`pci=resource_alignment=``[<order of align>@]<pci_dev>[; ...]`memory resource를 다시 배정할 장치와 정렬을 지정합니다. 정렬 order를 생략하면 `PAGE_SIZE`를 사용합니다. resource window 확장이 필요하면 PCI-PCI bridge도 지정할 수 있습니다. 같은 장치의 여러 instance에는 vendor, device, subvendor, subdevice를 쓸 수 있으며 예를 들어 `12@pci:8086:9c22:103c:198f`는 4096byte 정렬입니다.
`pci=ecrc=``{ bios | off | on }`PCIe ECRC(transaction layer end-to-end CRC)를 제어합니다. OS가 ACPI `_OSC`로 native AER 제어권을 받았거나 `pcie_ports=native`로 강제한 경우에만 효과가 있습니다. `bios`는 firmware 설정을 따르는 기본값, `off`는 끔, `on`은 켬입니다.
`pci=hpiosize=nn[KMG]`hotplug, 기본 256 byteshotplug bridge의 I/O window에 예약할 고정 bus space입니다.
`pci=hpmmiosize=nn[KMG]`hotplug, 기본 2MBhotplug bridge의 MMIO window에 예약할 고정 bus space입니다.
`pci=hpmmioprefsize=nn[KMG]`hotplug, 기본 2MBhotplug bridge의 MMIO_PREF window에 예약할 고정 bus space입니다.
`pci=hpmemsize=nn[KMG]`hotplug, 기본 2MBhotplug bridge의 MMIO와 MMIO_PREF window에 예약할 고정 bus space입니다.
`pci=hpbussize=nn`hotplug, 기본 `1`hotplug bridge 아래 bus에 추가로 예약할 최소 bus 번호 수입니다.
`pci=realloc=``{ off | on }`BIOS가 배정한 PCI bridge resource가 모든 child 장치에 필요한 크기보다 작을 때 재할당을 켜거나 끕니다.
`pci=realloc``realloc=on` 별칭PCI bridge resource 재할당을 활성화합니다.
`pci=noari`PCIePCIe ARI를 사용하지 않습니다.
`pci=noats`[PCIE,Intel-IOMMU,AMD-IOMMU]PCIe ATS와 IOMMU device IOTLB를 사용하지 않습니다.
`pci=pcie_scan_all`PCIe가능한 모든 PCIe 장치를 scan합니다. 지정하지 않으면 PCIe downstream port 아래에서 장치 하나만 찾습니다.
`pci=big_root_window`AMD CPUAMD CPU의 PCIe root complex에 큰 64bit memory window를 추가하려고 시도합니다. 일부 GFX hardware는 BAR를 늘려 전체 VRAM에 접근할 수 있습니다. 보고되지 않은 장치와 충돌할 위험이 조금 있어 kernel을 taint합니다.

`resource_alignment=`의 원문 형식은 다음과 같습니다.

				[<order of align>@]<pci_dev>[; ...]

PCI ACS·격리와 s390 옵션

5015-5064
매개변수적용·형식한국어 설명
`pci=disable_acs_redir=<pci_dev>[; ...]`세미콜론 구분 장치 목록지정한 장치의 PCI ACS redirect capability를 강제로 끕니다. bridge를 거치는 P2P traffic을 upstream으로 강제하지 않지만 장치 사이 격리를 제거하고 더 많은 장치를 같은 IOMMU group에 넣을 수 있습니다.
`pci=config_acs=``<ACS flags>@<pci_dev>[; ...]`장치마다 ACS capability를 켜거나 끄거나 유지합니다. flag bit 0은 Source Validation, 1은 Translation Blocking, 2는 P2P Request Redirect, 3은 P2P Completion Redirect, 4는 Upstream Forwarding, 5는 P2P Egress Control, 6은 Direct Translated P2P입니다. 각 bit의 `0`은 강제 비활성화, `1`은 강제 활성화, `x`는 변경하지 않음입니다. `pci=config_acs=10x@pci:0:0`은 P2P Request Redirect를 켜고 Translation Blocking을 끄며 Source Validation은 firmware 상태를 유지합니다. 격리를 제거하고 IOMMU group을 합칠 수 있으므로 주의해야 합니다.
`pci=force_floating`[S390]floating interrupt 사용을 강제합니다.
`pci=nomio`[S390]MIO instruction을 사용하지 않습니다.
`pci=norid`[S390]RID field를 무시하고 PCI function마다 PCI domain 하나를 사용하도록 강제합니다.
`pci=notph`[PCIE], `PCIE_TPH`kernel 구성이 지원하더라도 시스템 전체에서 PCIe TLP Processing Hints를 비활성화합니다.

ACS bit와 예시의 원문 표기는 다음과 같습니다.

				ACS Flags is defined as follows:
				  bit-0 : ACS Source Validation
				  bit-1 : ACS Translation Blocking
				  bit-2 : ACS P2P Request Redirect
				  bit-3 : ACS P2P Completion Redirect
				  bit-4 : ACS Upstream Forwarding
				  bit-5 : ACS P2P Egress Control
				  bit-6 : ACS Direct Translated P2P
				Each bit can be marked as:
				  '0' – force disabled
				  '1' – force enabled
				  'x' – unchanged
				For example,
				  pci=config_acs=10x@pci:0:0

PCIe service·전원·percpu·PMU

5065-5130
매개변수적용·형식한국어 설명
pcie_aspm=[PCIE] `{ off | force }`PCIe Active State Power Management를 제어합니다. `off`는 firmware 설정을 건드리지 않고, `force`는 지원하지 않는다고 알리는 장치에도 ASPM을 강제로 켭니다. 강제 활성화는 system lockup을 일으킬 수 있습니다.
pcie_ports=[PCIE] `{ native | dpc-native | compat }``native`는 platform 허가 없이도 PME, AER, DPC, PCIe hotplug service를 native로 사용해 platform과 충돌할 수 있습니다. `dpc-native`는 DPC만 native로 사용하며 firmware AER·DPC와 충돌할 수 있습니다. `compat`은 이 native PCIe service들을 끕니다.
pcie_port_pm=[PCIE] `{ off | force }``off`는 모든 PCIe port 전원 관리를 끄고 `force`는 모두 강제로 켭니다.
pcie_pme=[PCIE,PM] `nomsi`native PCIe PME signaling에 MSI를 사용하지 않습니다. 모든 PCIe root port가 모든 service에 INTx를 사용하게 됩니다.
pcmv=[HW,PCMCIA]BadgePAD 4용 PCMCIA 매개변수입니다.
pd_ignore_unused[PM]driver가 claim하지 않았더라도 bootloader가 켜 둔 모든 power domain을 계속 켭니다. debug·개발용이며 driver가 올바른 platform에는 필요하지 않아야 합니다.
pdcchassis=[PARISC,HW] `{ 0 | 1 }`부팅 시 PDC Chassis Status code를 끄거나 켭니다. 구현은 `arch/parisc/kernel/pdc_chassis.c`를 참조하십시오.
percpu_alloc=[MM,EARLY] `{ embed | page }`percpu first chunk allocator를 고릅니다. architecture에 따라 일부 또는 전혀 지원하지 않을 수 있습니다. 각 allocator의 설명은 `mm/percpu.c` 주석에 있으며 주로 debug와 성능 비교에 사용합니다.
pirq=[SMP,APIC]mp-table을 수동 설정합니다. `Documentation/arch/x86/i386/IO-APIC.rst`를 참조하십시오.
plip=[PPT,NET] `{ parport<nr> | timid | 0 }`병렬 포트 network link를 정합니다. `Documentation/admin-guide/parport.rst`도 참조하십시오.
pmtmr=[X86] 16진수 I/O portpmtimer I/O port를 수동으로 덮어씁니다. 예시는 `pmtmr=0x508`입니다.
pmu_override=[PPC] 숫자 또는 `on`PMU facility를 가져오므로 perf가 더 이상 사용할 수 없습니다. MMCR0을 0으로 두어 FC bit를 지우고 counter를 시작합니다. 숫자를 주면 MMCR1을 그 값으로 설정하고 `on`처럼 숫자가 아니면 MMCR1은 0으로 남습니다.

전원·PnP·preemption·printk·proc_mem

5131-5268
매개변수적용·형식한국어 설명
pm_async=[PM] `off`부팅 시 `/sys/power/pm_async`의 초기값을 정합니다. `off`는 시스템 전체 전원 전환 중 장치의 비동기 suspend·resume을 끕니다. 장치 의존성이 명확하지 않은 platform이나 전원 관리 debug에 유용하며 기본값은 비동기 동작 활성화입니다.
pm_debug_messages[SUSPEND,KNL]부팅 과정에서 suspend·resume debug message를 활성화합니다.
pnp.debug=1[PNP], `CONFIG_PNP_DEBUG_MESSAGES`PNP debug message를 켭니다. 실행 중에는 `/sys/module/pnp/parameters/debug`에서 바꿀 수 있습니다. 현재 resource 사용량은 항상 보이며, 이 옵션은 가능한 설정과 일부 할당 정보도 표시합니다.
pnpacpi=[ACPI] `off`ACPI PnP 지원을 끕니다.
pnpbios=[ISAPNP] `{ on | off | curr | res | no-curr | no-res }`PnP BIOS의 활성화와 current configuration·resource 정보 사용 여부를 제어합니다.
pnp_reserve_irq=[ISAPNP]자동 구성에서 제외할 IRQ를 예약합니다.
pnp_reserve_dma=[ISAPNP]자동 구성에서 제외할 DMA channel을 예약합니다.
pnp_reserve_io=[ISAPNP] base와 size 쌍자동 구성에서 제외할 I/O port 범위를 예약합니다.
pnp_reserve_mem=[ISAPNP] base와 size 쌍자동 구성에서 제외할 memory region을 예약합니다.
ports=[IP_VS_FTP] `<port>,<port>...`IPVS FTP helper module의 port를 최대 `IP_VS_APP_MAX_PORTS`인 8개까지 지정합니다. 기본값은 21입니다.
possible_cpus=[SMP,S390,X86] unsigned intACPI·firmware 같은 일반 탐지 결과를 덮어쓰고 가능한 CPU 수를 지정합니다.
powersave=off[PPC]cpuidle을 끄고 platform machine description의 `power_save` 함수를 `NULL`로 만들어 절전 기능을 비활성화합니다. idle에서 CPU는 실행 우선순위만 낮춥니다.
ppc_strict_facility_enable[PPC,ENABLE]`kernel_enable_fpu()`·`kernel_disable_fpu()`처럼 명시적으로 허용한 구역 밖에서 kernel이 floating point, Altivec, VSX, SPE를 사용하는지 잡아냅니다. 활성화하면 성능 비용이 있습니다.
ppc_tm=[PPC,EARLY] `off`Hardware Transactional Memory를 비활성화합니다.
preempt=[KNL], `CONFIG_PREEMPT_DYNAMIC`preemption mode를 고릅니다. `none`은 `cond_resched()` 지점만, `voluntary`는 `cond_resched()`와 `might_sleep()` 지점만 선점합니다. `full`은 명시적으로 preemption을 끄지 않은 모든 구역을 언제든 선점하고, critical section이 lock 자체보다 넓게 preemption을 끄지 않았다면 경합 중 spinlock에서도 task가 양보합니다. `lazy`는 scheduler가 제어하며 full과 비슷하지만 즉시 선점하지 않고 task가 스스로 양보할 시간을 HZ tick 하나만큼 준 뒤 강제합니다. 또 하나의 선점 지점은 user space로 돌아갈 때입니다.
print-fatal-signals=[KNL] bool, 기본 offsignal이 너무 많거나 POSIX.1 timer가 너무 많거나 fatal signal로 coredump가 발생하는 등 signal 처리 관련 application 이상을 경고합니다. signal overflow 경고라면 `ulimit -i unlimited`를 고려할 수 있습니다.
printk.always_kmsg_dump=bool, 기본 disabledkernel oops나 panic 이외의 경우에도 `kmsg_dump`를 호출합니다.
printk.console_no_auto_verbose=bool, 기본 `0`oops, panic, lockdep 문제에서 console loglevel을 자동으로 올리지 않게 합니다. 낮은 baudrate serial console을 제외하면 더 많은 debug 정보를 위해 기본 0, 즉 auto verbose 활성화를 유지하는 편이 좋습니다.
printk.debug_non_panic_cpus=bool, 기본 disabled`panic()` 중 panic을 일으키지 않은 CPU의 message도 printk log buffer에 저장하게 합니다. panic CPU가 최선의 노력으로 이를 console에 flush합니다.
printk.devkmsg={on,off,ratelimit}기본 `ratelimit``/dev/kmsg` 쓰기를 제어합니다. `on`은 userspace의 무제한 logging, `off`는 logging 금지, `ratelimit`은 rate limit 적용입니다.
printk.time=bool각 printk message line 앞에 timing data를 붙입니다.
proc_mem.force_override=[KNL] `{ always | ptrace | never }`전통적으로 `/proc/pid/mem`은 제한 없이 memory permission을 덮어쓸 수 있습니다. `always`는 이 동작을 항상 허용하고, `ptrace`는 활성 ptracer에만 허용하며, `never`는 허용하지 않습니다. 생략하면 `CONFIG_PROC_MEM_*` 선택을 따릅니다.

Processor C-state·profiling·PS/2·PTI

5269-5330
매개변수적용·형식한국어 설명
processor.max_cstate=[HW,ACPI]processor가 진입할 최대 C-state를 제한합니다. `max_cstate=9`는 DMI blacklist의 제한도 덮어씁니다.
processor.nocst[HW,ACPI]C-state를 결정할 때 ACPI `_CST` method를 무시하고 legacy FADT method를 사용합니다.
profile=[KNL] `[<profiletype>,]<number>``/proc/profile`을 통한 kernel profiling을 활성화합니다. profile type은 `schedule` 또는 `kvm`이며 생략하면 일반 kernel profiling입니다. `schedule`은 schedule 지점, `kvm`은 VM exit를 profile합니다. 숫자는 통계적 시간 기반 profiling의 2의 거듭제곱 step/bucket 크기입니다.
prompt_ramdisk=[RAM] deprecated더 이상 권장되지 않는 RAM disk prompt 매개변수입니다.
prot_virt=[S390] boolhardware가 지원하면 hypervisor로부터 격리된 protected virtual machine hosting을 활성화합니다. 이 기능을 켜면 KASLR이 꺼져 있어도 기본 kernel base address가 바뀔 수 있습니다.
psi=[KNL] boolPSI(Pressure Stall Information) 추적을 켜거나 끕니다.
psmouse.proto=[HW,MOUSE] `{ bare | imps | exps | lifebook | any }`탐색할 가장 높은 PS/2 mouse protocol extension을 지정합니다.
psmouse.rate=[HW,MOUSE] 초당 report 수원하는 mouse report rate를 지정합니다.
psmouse.resetafter=[HW,MOUSE] 정수지정한 수만큼 잘못된 packet이 발생하면 장치를 reset합니다. 0은 reset하지 않습니다.
psmouse.resolution=[HW,MOUSE] dpi원하는 mouse 해상도를 지정합니다.
psmouse.smartscroll=[HW,MOUSE] `{ 0 | 1 }`, 기본 `1`Logitech smartscroll autorepeat를 제어합니다. 0은 비활성화, 1은 활성화입니다.
pstore.backend=backend 이름사용할 pstore backend의 이름을 지정합니다.
pti=[X86-64] `{ on | off | auto }`user와 kernel address space의 Page Table Isolation을 제어합니다. 끄면 보안 강화가 사라지지만 system call과 interrupt 성능은 좋아집니다. `on`은 무조건 활성화, `off`는 무조건 비활성화, `auto`는 CPU model이 PTI가 완화하는 취약점에 영향받는지 kernel이 판단합니다. 생략은 `pti=auto`와 같습니다.
pty.legacy_count=[KNL] 정수legacy PTY 수를 지정하여 compile-time 기본값을 덮어씁니다.

Quiet·RAM disk·random·RAS

5331-5373
매개변수적용·형식한국어 설명
quiet[KNL,EARLY]대부분의 log message를 비활성화합니다.
r128=[HW,DRM]Rage 128 DRM driver 매개변수입니다.
radix_hcall_invalidate=on[PPC/PSERIES]RADIX GTSE 기능을 끄고 TLB invalidate에 hypercall을 사용합니다.
raid=[HW,RAID]RAID 설정입니다. `Documentation/admin-guide/md.rst`를 참조하십시오.
ramdisk_size=[RAM] KBRAM disk 크기를 KB 단위로 정합니다. `Documentation/admin-guide/blockdev/ramdisk.rst`를 참조하십시오.
ramdisk_start=[RAM] 주소RAM disk image의 시작 주소를 지정합니다.
random.trust_cpu=off[KNL,EARLY]사용 가능한 경우에도 kernel RNG 초기화에 CPU random number generator를 신뢰하지 않습니다.
random.trust_bootloader=off[KNL,EARLY]사용 가능한 경우에도 kernel RNG 초기화에 bootloader가 전달한 seed를 신뢰하지 않습니다.
randomize_kstack_offset=[KNL,EARLY] boolkernel stack offset randomization을 켜거나 끕니다. 약 5bit entropy를 제공하여 stack 주소 결정성이나 system call 사이의 주소 노출에 의존하는 memory corruption 공격을 어렵게 합니다. `CONFIG_HAVE_ARCH_RANDOMIZE_KSTACK_OFFSET`을 정의한 architecture에서만 사용할 수 있고 기본값은 `CONFIG_RANDOMIZE_KSTACK_OFFSET_DEFAULT`입니다.
ras=option[,option,...][KNL] `cec_disable`RAS 전용 옵션입니다. X86의 `cec_disable`은 Correctable Errors Collector를 비활성화하며 자세한 내용은 `CONFIG_RAS_CEC` 도움말을 참조하십시오.

RCU callback offload

5374-5412
매개변수적용·형식한국어 설명
rcu_nocbs[=cpu-list][KNL], `CONFIG_RCU_NOCB_CPU=y`선택한 CPU를 no-callback mode로 두어 해당 CPU의 RCU callback을 softirq context에서 실행하지 않고 전용 `rcuox/N` kthread로 offload합니다. `x`는 RCU-preempt의 `p`, RCU-sched의 `s`, grace period 중재 thread의 `g`이고 N은 CPU 번호입니다. offload CPU의 OS jitter를 줄여 HPC·real-time workload에 유용하며 asymmetric multiprocessor의 energy efficiency도 높일 수 있습니다. cpu-list가 있으면 부팅부터 해당 CPU에 적용합니다. `=`와 목록을 생략하면 부팅 시에는 적용하지 않고 cpuset으로 실행 중 전환할 수 있습니다. 이 옵션은 `CONFIG_RCU_NOCB_CPU_DEFAULT_ALL`보다 우선합니다.
rcu_nocb_poll[KNL]`rcu_nocbs=`로 offload한 CPU가 대응하는 `rcuoN` kthread를 직접 깨우는 대신 kthread가 callback을 polling하게 합니다. wakeup 부담을 없애 real-time response는 좋아지지만 kthread가 주기적으로 깨어나므로 energy efficiency는 낮아집니다.

RCU tree batch·barrier·grace period 지연

5413-5440
매개변수적용·형식한국어 설명
rcutree.blimit=[KNL] 정수한 batch에서 처리할 완료된 RCU callback의 최대 수를 정합니다.
rcutree.csd_lock_suppress_rcu_stall=[KNL]지나치게 긴 CSD-lock 대기가 진행 중이면 RCU CPU stall warning을 한 줄만 출력합니다.
rcutree.do_rcu_barrier=[KNL]`rcu_barrier()` 호출을 요청합니다. userspace test가 sysfs 변수를 반복해도 안전하도록 throttle됩니다. RCU grace-period machinery가 완전히 활성화되기 전에 trigger하면 `EAGAIN`으로 실패합니다.
rcutree.dump_tree=[KNL]초기 부팅에서 `rcu_node` combining tree 구조를 dump하여 tree 설정이 올바른지 진단합니다.
rcutree.gp_cleanup_delay=[KNL] jiffiesRCU grace-period cleanup의 각 단계를 지연할 jiffy 수를 정합니다.
rcutree.gp_init_delay=[KNL] jiffiesRCU grace-period initialization의 각 단계를 지연할 jiffy 수를 정합니다.

RCU FQS timing과 kthread priority

5441-5484
매개변수적용·형식한국어 설명
rcutree.gp_preinit_delay=[KNL] jiffies최근 CPU hotplug 변경을 `rcu_node` combining tree 위로 전파하는 grace-period pre-initialization의 각 단계를 지연합니다.
rcutree.jiffies_till_first_fqs=[KNL] `0..HZ`grace-period initialization부터 첫 force-quiescent-state 시도까지의 지연을 jiffy로 지정합니다.
rcutree.jiffies_till_next_fqs=[KNL] `1..HZ`이후 force-quiescent-state 시도 사이의 지연을 jiffy로 지정합니다.
rcutree.jiffies_till_sched_qs=[KNL] jiffiesRCU가 `rcu_note_context_switch()`와 `cond_resched()`에 quiescent-state 도움을 요청하기 전에 grace period가 도달해야 하는 나이를 지정합니다. 생략하면 최근 `jiffies_till_first_fqs`와 `jiffies_till_next_fqs`에서 계산합니다. 계산값은 `rcutree.jiffies_to_sched_qs`에서 볼 수 있으며 이를 직접 설정해도 덮어씁니다.
rcutree.kthread_prio=[KNL,BOOT] priorityRCU per-CPU `rcuc/N` kthread의 `SCHED_FIFO` priority를 정하며 `rcub/N` boost thread와 `rcu_bh`, `rcu_preempt`, `rcu_sched` grace-period kthread에도 적용합니다. `RCU_BOOST`이면 범위 1-99, 기본 1이고 아니면 범위 0-99, 기본 0인 non-realtime입니다. `RCU_NOCB_CPU`이면 NOCB callback kthread priority도 조정합니다.

NOCB queue와 callback 임계값

5485-5522
매개변수적용·형식한국어 설명
rcutree.nocb_nobypass_lim_per_jiffy=[KNL] callback 수/jiffycallback-offload CPU에서 flood가 아닐 때는 추가 overhead를 피하려고 main `->cblist`에 직접 queue하고, 한 jiffy에 이 임계값보다 많은 callback이 들어오면 lock contention을 줄이기 위해 `->nocb_bypass` queue에 먼저 넣습니다.
rcutree.nohz_full_patience_delay=[KNL] ms, 기본 `0`, 최대 5초callback-offload CPU에서 grace period가 지정한 나이에 도달하기 전에는 RCU를 방해하지 않습니다. 큰 값은 5초로 제한되고 모든 값은 jiffy로 표현 가능한 가장 가까운 아래 값으로 반올림됩니다.
rcutree.qhimark=[KNL] callback 수queue된 RCU callback이 이 임계값을 넘으면 batch 제한을 비활성화합니다.
rcutree.qlowmark=[KNL] callback 수queue된 RCU callback이 이 임계값 아래로 내려가면 batch 제한을 다시 활성화합니다.
rcutree.qovld=[KNL] callback 수queue된 callback이 이 임계값을 넘으면 RCU force-quiescent-state scan이 `cond_resched()`와 scheduler IPI의 도움을 적극 요청하여 CPU가 더 빨리 quiescent state에 도달하게 합니다. 음수는 부팅 시 `qhimark`에 따라 정하고 0은 적극적인 도움 요청을 끕니다.
NOCB callback queue 선택
rcu_nocbs CPU callbackcallbacks/jiffy ≤ limitmain ->cblist
rcu_nocbs CPU callbackcallbacks/jiffy > limitpre-queue ->nocb_bypassmain ->cblist

한 jiffy의 callback 유입량이 임계값 이하이면 짧은 경로인 `->cblist`로 직접 넣고, flood로 판단되면 lock contention을 분산하기 위해 `->nocb_bypass`에 먼저 모읍니다.

RCU cache refill·batch 계산·tree fanout

5523-5564
매개변수적용·형식한국어 설명
rcutree.rcu_delay_page_cache_fill_msec=[KNL] `0..100000` mslow-memory 상황에 대응해 page-cache를 다시 채우기까지의 지연을 정합니다.
rcutree.rcu_divisor=[KNL] shift-right count이 CPU에 queue된 callback 수에서 callback 실행 batch limit `bl`을 계산할 때 사용할 right shift 횟수입니다. 결과는 `rcutree.blimit`보다 작아지지 않습니다. softirq handler는 `bl`개 callback마다 종료해 CPU가 다른 일을 할 기회를 줍니다. 이 제한은 offload하지 않은 callback에만 적용되고, offload callback은 scheduler가 일반 task처럼 선점하는 `rcuoc` kthread context에서 실행됩니다.
rcutree.rcu_fanout_exact=[KNL]`rcu_node` combining tree의 auto-balancing을 끕니다. `rcutorture`와 cache-to-cache transfer latency가 큰 architecture에 유용할 수 있습니다.
rcutree.rcu_fanout_leaf=[KNL] CPU 수각 leaf `rcu_node`에 배정할 CPU 수를 바꿉니다. 매우 큰 system은 64를 선택하고, remote access latency가 큰 NUMA system은 적절한 hardware boundary에 맞춘 값을 선택합니다.
rcutree.rcu_min_cached_objs=[KNL] object 수/CPUCPU마다 cache하고 유지할 최소 object 수를 정합니다. object 크기는 `PAGE_SIZE`입니다. 이 cache는 page allocator pressure를 줄이고 low-memory 상황의 동작을 개선합니다.

RCU NOCB group·callback 시간·softirq

5565-5610
매개변수적용·형식한국어 설명
rcutree.rcu_nocb_gp_stride=[KNL] kthread 수/group각 group의 NOCB callback kthread 수를 정합니다. 기본값은 CPU 수의 제곱근입니다. 큰 값은 global grace-period kthread의 wakeup overhead를 줄이지만 각 group NOCB grace-period kthread의 overhead는 늘립니다.
rcutree.rcu_kick_kthreads=[KNL]grace-period kthread가 force-quiescent-state 시점에 기대보다 세 배 오래 잠들면 추가 `wake_up()`을 보냅니다. 이때 `WARN_ONCE()`와 `ftrace_dump()`도 발생합니다.
rcutree.rcu_resched_ns=[KNL] nsRCU callback batch 실행 시간을 지정한 nanosecond로 제한합니다. `local_clock()` overhead를 줄이기 위해 기본적으로 32 callback마다 한 번만 제한을 검사합니다.
rcutree.rcu_unlock_delay=[KNL] microseconds, 기본 `0`, `CONFIG_RCU_STRICT_GRACE_PERIOD=y``rcu_read_unlock()` 시점에 지연을 넣습니다. 지연이 클수록 unlock 뒤 RCU-protected pointer를 잘못 사용하는 pointer leak을 잡을 가능성이 높아집니다.
rcutree.sysrq_rcu=[KNL]새 grace period가 시작되지 않는 이유를 조사할 수 있도록 sysrq key 하나를 Tree RCU의 `rcu_node` tree dump에 할당합니다.
rcutree.use_softirq=[KNL] bool0이면 모든 `RCU_SOFTIRQ` 처리를 per-CPU `rcuc` kthread로 옮깁니다. 기본값은 0이 아니어서 softirq를 사용합니다. `CONFIG_PREEMPT_RT=y` kernel은 이 boot parameter를 비활성화하고 강제로 0으로 설정합니다.

RCU lazy callback과 synchronize_rcu wakeup

5611-5630
매개변수적용·형식한국어 설명
rcutree.enable_rcu_lazy=[KNL]전력을 아끼기 위해 RCU callback을 batch로 모으고 지연 만료, memory pressure, callback list 과대 성장 시 flush합니다.
rcutree.rcu_normal_wake_from_gp=[KNL] bool`synchronize_rcu()` 호출의 latency를 줄입니다. 일반 callback의 `call_rcu[_hurry]()` 경로를 쓰지 않고 `synchronize_rcu()` caller를 별도로 추적하며 normal grace period에만 적용됩니다. 기본값은 `num_possible_cpus() <= 16`이면 1이지만 boot parameter로 0을 명시하면 꺼집니다.

실행 중 sysfs로 켜거나 boot parameter로 전달하는 원문 예시는 다음과 같습니다.

			echo 1 > /sys/module/rcutree/parameters/rcu_normal_wake_from_gp
			or pass a boot parameter "rcutree.rcu_normal_wake_from_gp=1"

rcuscale grace-period와 kfree 크기

5631-5660
매개변수적용·형식한국어 설명
rcuscale.gp_async=[KNL]`call_rcu()` 같은 asynchronous grace-period primitive의 성능을 측정합니다.
rcuscale.gp_async_max=[KNL] callback 수/writerwriter thread마다 미완료 상태로 둘 수 있는 최대 callback 수를 정합니다. 한도를 넘으면 해당 RCU flavor의 `rcu_barrier()`를 호출해 이전 callback을 drain합니다.
rcuscale.gp_exp=[KNL]expedited synchronous grace-period primitive의 성능을 측정합니다.
rcuscale.holdoff=[KNL] 시간부팅 완료와 test의 간섭을 피하도록 test 시작을 지연합니다.
rcuscale.kfree_by_call_rcu=[KNL], `CONFIG_RCU_LAZY=y``kfree_rcu()` 대신 `call_rcu()`를 test합니다.
rcuscale.kfree_mult=[KNL] 배수, 기본 `1``kfree_obj` 크기 대신 `kfree_mult * sizeof(kfree_obj)` 크기의 object를 할당합니다.

rcuscale kfree flood와 reader 수

5661-5700
매개변수적용·형식한국어 설명
rcuscale.kfree_rcu_test=[KNL]`kfree_rcu()` flooding 성능 측정을 활성화합니다.
rcuscale.kfree_rcu_test_double=[KNL]`kfree_rcu()`의 두 argument variant를 test합니다. 값이 `kfree_rcu_test_single`과 같으면 single·double variant를 모두 test합니다.
rcuscale.kfree_rcu_test_single=[KNL]`kfree_rcu()`의 한 argument variant를 test합니다. 값이 `kfree_rcu_test_double`과 같으면 single·double variant를 모두 test합니다.
rcuscale.kfree_nthreads=[KNL] thread 수`kfree_rcu()` loop를 실행할 thread 수를 정합니다.
rcuscale.kfree_alloc_num=[KNL] 횟수한 iteration에서 수행할 allocation과 free 수를 정합니다.
rcuscale.kfree_loops=[KNL] loop 수각 loop에서 `kfree_alloc_num`회 allocation과 free를 수행할 반복 횟수를 정합니다.
rcuscale.minruntime=[KNL] 초최소 test 실행 시간을 정합니다. data collection interval에는 영향이 없고 CPU 사용량 같은 값을 더 잘 측정하게 합니다.
rcuscale.nreaders=[KNL] 정수RCU reader 수를 정합니다. -1은 CPU 수 N을 선택하고 -1보다 작은 n은 `N-n+1`을 선택합니다. 원문 예시대로 -2는 N, -3은 N+1이며 n이 `-N` 이하이면 reader 하나를 선택합니다.

rcuscale writer·종료·holdoff

5701-5726
매개변수적용·형식한국어 설명
rcuscale.nwriters=[KNL] 정수RCU writer 수를 정합니다. 값의 해석은 `rcuscale.nreaders`와 같습니다.
rcuscale.scale_type=[KNL] RCU implementationtest할 RCU 구현을 지정합니다.
rcuscale.shutdown=[KNL] boolperformance test 완료 뒤 system을 종료합니다. 사람 개입 없는 자동 test에 유용합니다.
rcuscale.verbose=[KNL] bool추가 `printk()` statement를 활성화합니다.
rcuscale.writer_holdoff=[KNL] microseconds, 기본 `0`grace period 사이 write-side holdoff를 microsecond로 정합니다. 0은 holdoff 없음입니다.
rcuscale.writer_holdoff_jiffies=[KNL] jiffies, 기본 `0`grace period 사이 추가 write-side holdoff를 jiffy로 정합니다. 0은 holdoff 없음입니다.

rcutorture FQS와 forward progress

5727-5759
매개변수적용·형식한국어 설명
rcutorture.fqs_duration=[KNL] microseconds`force_quiescent_state` burst의 지속 시간을 정합니다.
rcutorture.fqs_holdoff=[KNL] microseconds`force_quiescent_state` burst 내부의 holdoff 시간을 정합니다.
rcutorture.fqs_stutter=[KNL] seconds`force_quiescent_state` burst 사이의 대기 시간을 정합니다.
rcutorture.fwd_progress=[KNL] kthread 수, 기본 `1`이 개념을 지원하는 RCU type의 grace-period forward-progress test에 사용할 kthread 수를 정합니다. 0보다 작거나 CPU 수보다 큰 값은 CPU 수를 사용하게 합니다.
rcutorture.fwd_progress_div=[KNL] 비율tight-loop forward-progress test를 수행할 CPU-stall-warning 기간의 비율을 정합니다.
rcutorture.fwd_progress_holdoff=[KNL] seconds연속 forward-progress test 사이에 기다릴 시간을 정합니다.
rcutorture.fwd_progress_need_resched=[KNL] booltight-loop forward-progress test의 `cond_resched()` 호출을 `need_resched()` 검사 안에 넣습니다.

조건부 grace-period primitive

5760-5798
매개변수적용·형식한국어 설명
rcutorture.gp_cond=[KNL]지원되는 경우 update-side conditional·asynchronous normal grace-period primitive를 사용합니다.
rcutorture.gp_cond_exp=[KNL]지원되는 경우 update-side conditional·asynchronous expedited grace-period primitive를 사용합니다.
rcutorture.gp_cond_full=[KNL]동시에 진행되는 expedited grace period도 고려하는 conditional·asynchronous normal grace-period primitive를 사용합니다.
rcutorture.gp_cond_exp_full=[KNL]동시에 진행되는 normal grace period도 고려하는 conditional·asynchronous expedited grace-period primitive를 사용합니다.
rcutorture.gp_cond_wi=[KNL] microseconds`gp_cond`·`gp_cond_full` normal conditional grace period의 nominal wait interval입니다. 실제 값은 이 상한까지 nanosecond 단위로 무작위 선택합니다. 기본값은 16 jiffy이며 `HZ=1000`에서는 예를 들어 16,000microseconds입니다.
rcutorture.gp_cond_wi_exp=[KNL] microseconds, 기본 `128``gp_cond_exp`·`gp_cond_exp_full` expedited conditional grace period의 nominal wait interval입니다. 실제 값은 이 상한까지 nanosecond 단위로 무작위 선택합니다.

Expedited·poll·synchronous grace period

5799-5849
매개변수적용·형식한국어 설명
rcutorture.gp_exp=[KNL]지원되는 경우 expedited update-side primitive를 사용합니다.
rcutorture.gp_normal=[KNL]지원되는 경우 normal, 즉 non-expedited asynchronous update-side primitive를 사용합니다.
rcutorture.gp_poll=[KNL]지원되는 경우 polled update-side normal grace-period primitive를 사용합니다.
rcutorture.gp_poll_exp=[KNL]지원되는 경우 polled update-side expedited grace-period primitive를 사용합니다.
rcutorture.gp_poll_full=[KNL]동시 expedited grace period도 고려하는 polled normal grace-period primitive를 사용합니다.
rcutorture.gp_poll_exp_full=[KNL]동시 normal grace period도 고려하는 polled expedited grace-period primitive를 사용합니다.
rcutorture.gp_poll_wi=[KNL] microseconds`gp_poll`·`gp_poll_full` normal grace period의 nominal wait interval입니다. 실제 값은 이 상한까지 nanosecond 단위로 무작위 선택합니다. 기본값은 16 jiffy이며 `HZ=1000`에서는 예를 들어 16,000microseconds입니다.
rcutorture.gp_poll_wi_exp=[KNL] microseconds, 기본 `128``gp_poll_exp`·`gp_poll_exp_full` expedited grace period의 nominal wait interval입니다. 실제 값은 이 상한까지 nanosecond 단위로 무작위 선택합니다.
rcutorture.gp_sync=[KNL]지원되는 경우 normal synchronous update-side primitive를 사용합니다. `gp_cond`, `gp_exp`, `gp_normal`, `gp_sync`가 모두 0이면 rcutorture는 모두 0이 아닌 것처럼 동작합니다.

Grace-period wrap lag·IRQ reader·NOCB

5850-5903
매개변수적용·형식한국어 설명
rcutorture.gpwrap_lag=[KNL] bool, 기본 truegrace-period wrap lag test를 켭니다. false이면 실행하지 않습니다.
rcutorture.gpwrap_lag_gps=[KNL] grace period 수, 기본 `8`active lag test에서 사용할 grace-period wrap lag 값을 정합니다. overflow flag를 세우기 전에 허용할 `rdp`와 `rnp`의 `gp_seq` 차이를 제어합니다.
rcutorture.gpwrap_lag_cycle_mins=[KNL] minutes, 기본 `30`active·inactive test 기간을 합친 gpwrap lag 한 cycle의 전체 시간을 정합니다.
rcutorture.gpwrap_lag_active_mins=[KNL] minutes, 기본 `5`각 cycle에서 `gpwrap_lag_gps` 값을 적용해 gpwrap lag를 활성화할 시간을 정합니다.
rcutorture.irqreader=[KNL] boolIRQ handler, 더 정확히는 timer handler에서 RCU reader를 실행합니다. 일부 RCU flavor는 이 사용 방식을 견디지 못할 수 있습니다.
rcutorture.leakpointer=[KNL] boolreader 밖으로 RCU-protected pointer를 일부러 유출합니다. splat이 발생할 수 있으며 `CONFIG_RCU_STRICT_GRACE_PERIOD=y` 같은 기능이 leak을 탐지하는지 test합니다.
rcutorture.n_barrier_cbs=[KNL] callback/thread 수`rcu_barrier()` test의 callback과 thread 수를 정합니다.
rcutorture.nfakewriters=[KNL] writer 수실제 test에는 참여하지 않고 RCU에 stress만 주는 동시 fake writer 수를 정합니다.
rcutorture.nocbs_nthreads=[KNL] thread 수, 기본 `0`RCU callback-offload를 전환할 toggler 수를 정합니다. 0은 전환을 비활성화합니다.
rcutorture.nocbs_toggle=[KNL] milliseconds연속 callback-offload 전환 시도 사이의 지연을 정합니다.

Reader 수·CPU hotplug·강제 선점

5904-5953
매개변수적용·형식한국어 설명
rcutorture.nreaders=[KNL] 정수RCU reader 수를 정합니다. CPU 수를 N이라 할 때 -1은 N-1, -1보다 작은 n은 `N-n-2`를 선택합니다. 원문 예시대로 -2는 N, -3은 N+1입니다.
rcutorture.object_debug=[KNL] booldebug-object의 double-`call_rcu()` test를 활성화합니다.
rcutorture.onoff_holdoff=[KNL] seconds부팅 뒤 CPU-hotplug test를 시작하기 전 대기 시간을 정합니다.
rcutorture.onoff_interval=[KNL] jiffiesCPU-hotplug 동작 사이 시간을 정합니다. 0은 hotplug test를 끕니다.
rcutorture.preempt_duration=[KNL] milliseconds, 기본 `0`high-priority FIFO real-time task가 선점할 지속 시간을 정합니다. 0은 비활성화입니다. 대상 CPU는 그 시점의 online 집합에서 무작위 선택하며 offline 전환과의 race가 나면 해당 선점 시도를 건너뜁니다.
rcutorture.preempt_interval=[KNL] milliseconds, 기본 1초high-priority FIFO real-time task의 선점 사이 간격을 정합니다. hrtimer가 지연을 중재하고 의도치 않은 동기화를 피하도록 추가 무작위화를 적용합니다.
rcutorture.read_exit_burst=[KNL] spawn 횟수한 read-then-exit episode에서 read-then-exit kthread 집합을 생성할 횟수를 정합니다.
rcutorture.read_exit_delay=[KNL] seconds연속 read-then-exit test episode 사이의 지연을 정합니다.
rcutorture.reader_flavor=[KNL] bitmask사용할 reader type을 지정합니다. 여러 bit가 설정되면 낮은 bit부터 reader에 진입하고 반대 순서로 나옵니다. SRCU에서 `0x1`은 normal reader, `0x2`는 NMI-safe reader, `0x4`는 lightweight reader입니다.

Task shuffle·CPU stall·stutter

5954-6008
매개변수적용·형식한국어 설명
rcutorture.shuffle_interval=[KNL] secondstask shuffle 간격을 정합니다. shuffle은 test 중 일부 CPU가 dyntick-idle mode에 들어가게 합니다.
rcutorture.shutdown_secs=[KNL] seconds부팅 뒤 지정 시간이 지나면 system을 종료합니다. 무인 자동 test에 유용합니다.
rcutorture.stall_cpu=[KNL] seconds, 0은 비활성화RCU CPU stall warning을 test할 CPU stall 지속 시간을 정합니다.
rcutorture.stall_cpu_block=[KNL] boolstall 중 sleep합니다. preemptible RCU warning과 다른 stall 동작을 일으킵니다. `CONFIG_PREEMPTION=n`, `CONFIG_PREEMPT_COUNT=y` kernel에서는 CPU가 quiescent state를 통과해 RCU CPU stall warning이 억제되고 대신 scheduling-while-atomic splat이 발생합니다. 이 매개변수는 의도적으로 splat을 만듭니다.
rcutorture.stall_cpu_holdoff=[KNL] seconds부팅 후 stall을 유발하기 전 대기 시간을 정합니다.
rcutorture.stall_cpu_irqsoff=[KNL] boolstall 집합의 첫 stall에서만 interrupt를 비활성화합니다.
rcutorture.stall_cpu_repeat=[KNL] 반복 수stall sequence의 추가 반복 횟수입니다. 예를 들어 `stall_cpu_repeat=3`은 총 네 번의 stall sequence를 만듭니다.
rcutorture.stall_gp_kthread=[KNL] seconds, 0은 비활성화RCU grace-period kthread를 강제로 sleep시켜 stall warning을 test합니다. `stall_cpu`도 지정하면 먼저 kthread를 굶기고 다음에 CPU를 stall합니다.
rcutorture.stat_interval=[KNL] secondsstatistics `printk()` 사이 시간을 정합니다.
rcutorture.stutter=[KNL] secondstest 실행과 대기를 번갈아 반복할 시간을 정합니다. 예를 들어 5초면 5초 실행, 5초 대기를 반복하여 RCU가 idle 상태로 급격히 들어가고 나오는 능력을 test합니다.

RCU boost·dyntick·test 종료 옵션

6009-6034
매개변수적용·형식한국어 설명
rcutorture.test_boost=[KNL] `{ 0 | 1 | 2 }`RCU priority boosting을 test합니다. 0은 안 함, 1은 해당 RCU 구현이 지원할 때만, 2는 항상 test입니다.
rcutorture.test_boost_duration=[KNL] seconds각 개별 boost test의 지속 시간을 정합니다.
rcutorture.test_boost_holdoff=[KNL] seconds, 기본 `0`test 시작부터 RCU priority-boost test 시작까지의 대기 시간입니다.
rcutorture.test_boost_interval=[KNL] seconds각 boost test 사이의 간격을 정합니다.
rcutorture.test_no_idle_hz=[KNL] boolRCU의 dyntick-idle 처리를 test합니다. `rcutorture.shuffle_interval`도 참조하십시오.
rcutorture.torture_type=[KNL] RCU 구현test할 RCU 구현을 지정합니다.
rcutorture.verbose=[KNL] bool추가 `printk()` statement를 활성화합니다.

RCU CPU stall 경고·timeout·통계

6035-6074
매개변수적용·형식한국어 설명
rcupdate.rcu_cpu_stall_ftrace_dump=[KNL] boolRCU CPU stall warning을 보고한 뒤 ftrace buffer를 dump합니다.
rcupdate.rcu_cpu_stall_notifiers=[KNL] boolRCU CPU stall notifier를 제공합니다. `RCU_CPU_STALL_NOTIFIER` Kconfig 도움말의 경고대로 거의 모든 경우 이 옵션을 원하지 않을 가능성이 큽니다.
rcupdate.rcu_cpu_stall_suppress=[KNL] boolRCU CPU stall warning message를 억제합니다.
rcupdate.rcu_cpu_stall_suppress_at_boot=[KNL] boolinit task가 생성되기 전 early boot 동안 발생하는 RCU CPU stall warning과 rcutorture writer stall warning을 억제합니다.
rcupdate.rcu_cpu_stall_timeout=[KNL] seconds, 최대 `300`RCU CPU stall warning의 timeout을 정합니다.
rcupdate.rcu_exp_cpu_stall_timeout=[KNL] milliseconds, 최대 `21000`expedited RCU CPU stall warning timeout을 정하며 architecture timer tick 해상도에 맞게 조정됩니다. 0이면 `rcu_cpu_stall_timeout`의 초 값을 millisecond로 변환해 사용합니다.
rcupdate.rcu_cpu_stall_cputime=[KNL] boolsampling 기간의 CPU time과 interrupt·task 횟수 통계를 제공합니다. 연속 RCU stall이 여러 번이면 모든 sampling 기간은 첫 stall timeout의 절반 시점에서 시작합니다.

Expedited stall task와 grace-period 정책

6075-6110
매개변수적용·형식한국어 설명
rcupdate.rcu_exp_stall_task_details=[KNL] boolexpedited RCU CPU stall warning에서 현재 expedited grace period를 막는 task의 stack dump를 출력합니다.
rcupdate.rcu_expedited=[KNL]`synchronize_rcu()` 대신 `synchronize_rcu_expedited()` 같은 expedited grace-period primitive를 사용합니다. latency는 줄지만 CPU 사용량이 늘고 real-time latency와 energy efficiency가 나빠질 수 있습니다. `CONFIG_TINY_RCU`에는 효과가 없습니다.
rcupdate.rcu_normal=[KNL]`synchronize_rcu_expedited()` 대신 `synchronize_rcu()` 같은 normal grace-period primitive만 사용합니다. real-time latency, CPU 사용량, energy efficiency는 좋아지지만 grace-period latency가 늘 수 있습니다. `rcu_expedited`보다 우선하며 `CONFIG_TINY_RCU`에는 효과가 없습니다.
rcupdate.rcu_normal_after_boot=[KNL]`rcu_end_inkernel_boot()` 호출 뒤에는 normal grace-period primitive만 사용합니다. `CONFIG_TINY_RCU`에는 효과가 없습니다. `CONFIG_PREEMPT_RT=y` kernel은 이 값을 강제로 1로 두어 부팅 뒤 expedited 요청을 normal 처리로 바꿉니다.

RCU Tasks callback queue와 IPI

6111-6147
매개변수적용·형식한국어 설명
rcupdate.rcu_task_collapse_lim=[KNL] callback 수grace period 시작 시 callback 수가 이 값 이하면 RCU Tasks flavor가 callback queue 하나로 다시 합칠 수 있습니다. 이 동적 전환은 `rcu_task_enqueue_lim`이 기본 -1일 때만 일어납니다.
rcupdate.rcu_task_contend_lim=[KNL] contention/jiffyRCU Tasks flavor가 per-CPU callback queue로 전환하기 위해 필요한 최소 callback-queue lock contention event 수를 정합니다. `rcu_task_enqueue_lim=-1`일 때만 동적 전환합니다.
rcupdate.rcu_task_enqueue_lim=[KNL] queue 수, 기본 `-1`RCU Tasks family가 사용할 callback queue 수를 정합니다. -1은 자동·동적 조정을 허용하며 이 매개변수는 test용입니다.
rcupdate.rcu_task_ipi_delay=[KNL] jiffiesgrace period 시작 뒤 RCU Tasks가 IPI를 보내지 않을 시간을 정합니다. 큰 값은 real-time workload 방해를 줄이지만 grace period를 늘립니다.
rcupdate.rcu_task_lazy_lim=[KNL] callback 수해당 CPU에서 lazy 상태를 취소할 callback 수입니다. -1은 lazy 취소를 막지만 callback flood로 OOM이 날 위험을 높입니다.

RCU Tasks stall 정보와 lazy batching

6148-6191
매개변수적용·형식한국어 설명
rcupdate.rcu_task_stall_info=[KNL] jiffies, 기본 10초RCU task stall informational message의 초기 timeout입니다. 0 이하이면 끕니다. 같은 grace period의 stall warning 전에만 출력하며 값 변경은 다음 grace period 시작부터 적용됩니다.
rcupdate.rcu_task_stall_info_mult=[KNL] `1..10`, 기본 `3`한 RCU Tasks grace period에서 연속 informational message 간격의 배수입니다. 기본값에서는 첫 message가 10초, 두 번째가 40초, 세 번째가 160초에 나오며 600초 stall warning 때문에 640초의 네 번째 message는 나오지 않습니다.
rcupdate.rcu_task_stall_timeout=[KNL] jiffies, 기본 10분RCU task stall warning timeout을 정합니다. 0 이하이면 끄며 변경은 다음 grace period 시작부터 적용됩니다.
rcupdate.rcu_tasks_lazy_ms=[KNL] milliseconds`call_rcu_tasks()`의 asynchronous callback batching timeout입니다. 음수는 기본값, 0은 batching 비활성화입니다. `synchronize_rcu_tasks()`에는 batching을 항상 사용하지 않습니다.
rcupdate.rcu_tasks_trace_lazy_ms=[KNL] milliseconds`call_rcu_tasks_trace()`의 asynchronous callback batching timeout입니다. 음수는 기본값, 0은 batching 비활성화입니다. `synchronize_rcu_tasks_trace()`에는 batching을 항상 사용하지 않습니다.

RCU early boot self-test

6192-6194
매개변수적용·형식한국어 설명
rcupdate.rcu_self_test=[KNL] boolRCU early boot self-test를 실행합니다.

Ramdisk init·RDRAND·RDT

6195-6213
매개변수적용·형식한국어 설명
rdinit=[KNL] `<full_path>`early userspace 시작에 ramdisk의 `/init` 대신 지정한 binary를 실행합니다. `initrd`도 참조하십시오.
rdrand=[X86,EARLY] `force`특정 AMD processor의 잘못된 BIOS suspend·resume 처리 때문에 kernel이 RDRAND 지원 표시를 숨기기로 한 결정을 덮어씁니다.
rdt=[HW,X86,RDT] feature 목록개별 RDT 기능을 켜거나 끕니다. 목록은 `cmt`, `mbmtotal`, `mbmlocal`, `l3cat`, `l3cdp`, `l2cat`, `l2cdp`, `mba`, `smba`, `bmec`, `abmc`입니다. 예를 들어 `rdt=cmt,!mba`는 cmt를 켜고 mba를 끕니다.

x86 reboot 문법과 기본 경로

6214-6246
매개변수적용·형식한국어 설명
reboot=[KNL] x86·x86_64`reboot_mode`는 warm(soft), cold(hard), gpio 중 하나이며 `panic_` prefix를 붙이면 panic reboot에만 적용합니다. `reboot_type`은 bios, acpi, kbd, triple, efi, pci 중 하나이고 `force`는 다른 CPU를 멈추지 않게 합니다. `smp####`는 reboot에 사용할 processor 번호를 정합니다. `acpi`는 FADT의 ACPI `RESET_REG`를 사용하고 실패하면 keyboard controller를 시도합니다. `bios`는 warm reset용 CPU reboot vector, `cold`는 cold reboot flag를 설정합니다. `default`는 platform quirk table을 무시하고 generic default reboot 동작을 사용합니다.

x86·x86_64의 원문 형식은 다음과 같습니다.

			Format (x86 or x86_64):
				[w[arm] | c[old] | h[ard] | s[oft] | g[pio]] | d[efault] \
				[[,]s[mp]#### \
				[[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \
				[[,]f[orce]

EFI·keyboard·PCI·warm reboot

6247-6276
매개변수적용·형식한국어 설명
`reboot=efi`EFIEFI `reset_system` runtime service를 사용합니다. EFI가 없거나 reset이 실패하면 keyboard controller를 시도합니다.
`reboot=force`rebootreboot할 때 다른 CPU를 멈추지 않습니다. 일부 system에서 reboot 신뢰성을 높일 수 있습니다.
`reboot=kbd`기본 cold resetkeyboard controller를 사용합니다.
`reboot=pci`PCIPCI configuration space register `0xcf9`에 써서 reboot를 trigger합니다.
`reboot=triple`x86triple fault(init)를 강제로 일으킵니다.
`reboot=warm`warm resetcold reboot flag를 설정하지 않습니다. BIOS memory check를 생략하므로 memory가 큰 system에서 훨씬 빠르지만 hardware가 완전히 다시 초기화되지 않아 일부 system에서 부팅 문제가 생길 수 있습니다.

refscale performance test

6277-6327
매개변수적용·형식한국어 설명
refscale.holdoff=[KNL] 시간부팅 완료와 test 간섭을 피하도록 test 시작을 지연합니다.
refscale.lookup_instances=[KNL] data element 수`SLAB_TYPESAFE_BY_RCU` 형식 test에 사용할 data element 수입니다. 음수는 절댓값에 `nr_cpu_ids`를 곱하고 0은 `nr_cpu_ids`를 사용합니다.
refscale.loops=[KNL] loop 수test 대상 synchronization primitive를 반복할 횟수입니다. 값을 늘리면 loop 시작·끝 overhead의 noise가 줄지만 기본값도 2020년경 x86 laptop에서 pass당 noise를 수 picosecond로 낮춥니다.
refscale.nreaders=[KNL] reader 수, 기본 `-1`reader 수를 정합니다. -1은 CPU 수의 약 75%인 N을 선택하며 0도 의미 있는 test 값입니다.
refscale.nruns=[KNL] run 수각 결과를 console log에 dump하는 run 횟수를 정합니다.
refscale.readdelay=[KNL] microsecondsread-side critical section의 지속 시간을 정합니다.
refscale.scale_type=[KNL] 구현 이름test할 read-protection 구현을 지정합니다.
refscale.shutdown=[KNL] boolperformance test 끝에 system을 종료합니다. built-in이면 기본 1, module이면 기본 0입니다.
refscale.verbose=[KNL] bool추가 `printk()` statement를 활성화합니다.
refscale.verbose_batched=[KNL] 정수추가 `printk()`를 batch 처리합니다. 0 또는 음수이면 모두 출력하고, 양수 N이면 N번째 verbose statement마다 출력합니다.

Regulator·resource 예약·software resume

6328-6385
매개변수적용·형식한국어 설명
regulator_ignore_unused[REGULATOR]driver가 claim하지 않은 regulator를 regulator framework가 끄지 않게 합니다. debug·개발용이며 driver 지원이 올바른 platform에는 필요하지 않아야 합니다.
relax_domain_level=[KNL,SMP]scheduler의 기본 `relax_domain_level`을 정합니다. `Documentation/admin-guide/cgroup-v1/cpusets.rst`를 참조하십시오.
reserve=[KNL,BUGS] `<base1>,<size1>[,<base2>,<size2>,...]`kernel이 사용하지 않도록 I/O port나 memory를 예약합니다. base가 `0x10000`보다 작으면 I/O port, 그 이상이면 memory로 봅니다.
reserve_mem=[RAM] `nn[KMG]:<align>:<label>`다른 subsystem이 이름으로 접근할 physical memory를 예약합니다. RAM을 지우지 않는 system에서 soft reboot 뒤 같은 memory를 다시 잡으려는 용도로 사용합니다. system 변경, 다른 kernel, KASLR 배치 때문에 같은 위치를 보장하지 않으므로 사용 subsystem은 내용이 이전 boot에서 온 것인지 검증해야 합니다. `12M:4096:oops`는 ramoops용 12MB를 4096 정렬로 요청합니다.
reservetop=[X86-32,EARLY] `nn[KMG]`kernel virtual address space의 맨 위에 hole을 예약합니다.
reset_devices[KNL]driver가 초기화 중 underlying device를 reset하도록 강제합니다.
resume=[SWSUSP] 장치 지정software suspend에서 resume image가 있는 partition device를 지정합니다.
resume_offset=[SWSUSP] `PAGE_SIZE` 단위`resume=` partition 시작부터 swap header까지의 offset을 지정합니다. swap file에만 필요하며 `Documentation/power/swsusp-and-swap-files.rst`를 참조하십시오.

`reserve_mem` 예시와 `resume=` 장치 형식의 원문은 다음과 같습니다.

			reserve_mem=12M:4096:oops ramoops.mem_name=oops
			{/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>}

Hibernation·RETBleed·rfkill

6386-6444
매개변수적용·형식한국어 설명
resumedelay=[HIBERNATION] secondsresume file 읽기를 시도하기 전 대기 시간을 정합니다.
resumewait[HIBERNATION]resume device가 나타날 때까지 무한 대기합니다. USB·MMC처럼 비동기 탐지 장치에 유용합니다.
retain_initrd[RAM]압축 해제 뒤에도 initrd memory를 보존합니다. 부팅 뒤 `/sys/firmware/initrd`에서 접근할 수 있습니다.
retbleed=[X86] mitigation modeRETBleed를 완화합니다. AMD의 UNRET·IBPB만으로는 sibling thread 간 branch prediction 영향을 막지 못하므로 STIBP가 있으면 사용하고 없으면 SMT를 완화할 수 있습니다. `off`는 완화 없음, `auto`는 CPU에 따라 자동 선택, `auto,nosmt`는 완전한 완화에 필요하면 Zen1 이하에서 SMT를 끕니다. `ibpb`는 AMD basic-block boundary의 짧은 speculation window까지 막아 가장 안전하지만 성능 비용이 크고 STIBP도 켜며 Intel에는 적합하지 않습니다. `ibpb,nosmt`는 STIBP가 없으면 SMT도 끕니다. `unret`은 AMD family 15h-17h에서 untrained return thunk를 강제하고 `unret,nosmt`는 STIBP가 없으면 SMT도 끕니다. 생략은 `retbleed=auto`와 같습니다.
rfkill.default_state=`{ 0 | 1 }`0은 기본 airplane mode로 Wi-Fi, Bluetooth, WiMAX, GPS, FM 등 통신을 모두 막고 1은 차단하지 않습니다.
rfkill.master_switch_mode=`{ 0 | 1 | 2 }`airplane mode button 동작입니다. 0은 아무 동작도 하지 않고, 1은 전체 차단과 이전 구성 사이를 전환하며, 2는 전체 차단과 전체 해제 사이를 전환합니다.

RISC-V fallback·root filesystem·RT group

6445-6509
매개변수적용·형식한국어 설명
ring3mwait=disable[KNL]지원 CPU에서 ring 3 `MONITOR/MWAIT` 기능을 비활성화합니다.
riscv_isa_fallback[RISCV,EARLY]`CONFIG_RISCV_ISA_FALLBACK`이 꺼져 있어도 대체 property를 찾지 못한 devicetree system에서 `riscv,isa` property를 parse해 extension 지원을 탐지하는 fallback을 허용합니다. `RISCV_ISA_FALLBACK` Kconfig를 참조하십시오.
ro[KNL]부팅 시 root device를 read-only로 mount합니다.
rodata=[KNL,EARLY] `{ on | off | noalias }``on`은 read-only kernel memory를 실제 read-only로 표시하는 기본값입니다. `off`는 debug를 위해 writable로 두고, arm64의 `noalias`는 read-only로 표시하되 kernel image 밖의 direct-map region에는 writable alias를 유지합니다.
rockchip.usb_uart[EARLY]Rockchip SoC의 지정 USB port에 UART passthrough를 활성화합니다. debug UART signal을 USB D+·D- pin으로 route하고 일반 USB controller를 비활성화합니다.
root=[KNL] root filesystem보통 block device specifier이며 자세한 형식은 `block/early-lookup.c`의 `early_lookup_bdev` 주석을 참조하십시오. legacy initial ramdisk는 `ram`, network root는 `nfs`·`cifs`, raw flash mount는 `mtd`·`ubi`를 사용할 수 있습니다.
rootdelay=[KNL] secondsroot filesystem mount를 시도하기 전 대기 시간을 정합니다.
rootflags=[KNL] mount option stringroot filesystem mount option 문자열을 지정합니다.
initramfs_options=[KNL] mount option stringinitramfs mount option을 지정합니다.
rootfstype=[KNL] filesystem typeroot filesystem type을 지정합니다.
rootwait[KNL]root device가 나타날 때까지 무한 대기합니다. USB·MMC 같은 비동기 탐지 장치에 유용합니다.
rootwait=[KNL] secondsroot device가 나타나기를 기다릴 최대 시간을 정한 뒤 root filesystem mount를 시도합니다.
rproc_mem=nn[KMG][@address][KNL,ARM,CMA]remote processor image가 사용할 physical memory block을 CMA로 관리합니다.
rt_group_sched=[KNL] bool, `CONFIG_RT_GROUP_SCHED=y``SCHED_RR`·`SCHED_FIFO` group scheduling을 켜거나 끕니다. 기본값은 `!CONFIG_RT_GROUP_SCHED_DEFAULT_DISABLED`입니다.
rw[KNL]부팅 시 root device를 read-write로 mount합니다.

Single mode·s390 IOMMU·SA1100 IrDA

6510-6534
매개변수적용·형식한국어 설명
S[KNL]init을 single mode로 실행합니다.
s390_iommu=[HW,S390] `strict`s390 IOTLB flush mode를 정합니다. `strict`는 unmap마다 IOTLB를 flush합니다. 기본 lazy flush는 reuse 직전에 flush하여 더 빠릅니다. 이 옵션은 deprecated이며 `iommu.strict=1`과 같습니다.
s390_iommu_aperture=[KNL,S390] 10진수 배수장치별 DMA address space 크기를 main memory 크기의 배수로 지정합니다. 기본 1은 hardware가 지원하면 installed physical memory 전체를 한 번 mapping할 수 있고 2는 두 번 mapping할 수 있습니다. 0은 hardware 제한 외에는 제한하지 않지만 table memory를 크게 더 사용합니다.
sa1100ir[NET]SA1100 IrDA driver 설정입니다. `drivers/net/irda/sa1100_ir.c`를 참조하십시오.

Scheduler proxy·통계·thermal decay

6535-6563
매개변수적용·형식한국어 설명
sched_proxy_exec=[KNL] boolmutex 기반 priority inversion을 해결하는 proxy execution 방식을 켜거나 끕니다.
sched_verbose[KNL,EARLY]상세 scheduler debug message를 활성화합니다.
schedstats=[KNL,X86] `{ enable | disable }`scheduler 통계를 켜거나 끕니다. 약간의 overhead가 있지만 debug와 performance tuning에 유용합니다.
sched_thermal_decay_shift=[KNL,SMP] deprecated, `0..10`, 기본 `0`scheduler thermal-pressure signal의 decay period를 지정한 값만큼 left shift합니다. 다른 PELT signal의 기본 decay가 32ms라면 shift 1은 64ms, shift 2는 128ms입니다.

smp_call_function torture 기본 제어

6564-6603
매개변수적용·형식한국어 설명
scftorture.holdoff=[KNL] secondstest 시작 전 대기 시간입니다. module insertion에서는 기본 0, built-in `smp_call_function()` test에서는 기본 10초입니다.
scftorture.longwait=[KNL] seconds, 기본 `0`선택한 상한까지 무작위로 매우 긴 대기를 요청합니다. 작은 양수도 RCU CPU stall warning과 softlockup을 일으킬 수 있습니다.
scftorture.nthreads=[KNL] kthread 수, 기본 `-1``smp_call_function()` family를 호출할 kthread 수입니다. -1은 CPU 수와 같은 수를 사용합니다.
scftorture.onoff_holdoff=[KNL] secondstest 시작 뒤 CPU-hotplug 동작을 시작하기 전 대기 시간입니다.
scftorture.onoff_interval=[KNL] seconds, 기본 `0`연속 CPU-hotplug 동작 사이의 대기입니다. 0은 hotplug를 비활성화합니다.
scftorture.shutdown_secs=[KNL] seconds, 기본 `0`test 시작 뒤 system을 종료할 시간을 정합니다. 양수는 자동 test에 유용합니다.
scftorture.stat_interval=[KNL] seconds현재 test 통계를 console에 출력할 간격입니다. 0은 통계 출력을 끕니다.

CPU 집합 전환과 smp_call_function weight

6604-6643
매개변수적용·형식한국어 설명
scftorture.stutter_cpus=[KNL] jiffiestest 대상 CPU 집합을 바꾸는 각 동작 사이의 대기 시간을 정합니다.
scftorture.use_cpus_read_lock=[KNL] bool`smp_call_function*()` 호출 중 CPU hotplug를 막을 때 기본 `preempt_disable()` 대신 `use_cpus_read_lock()`을 사용합니다.
scftorture.verbose=[KNL] bool추가 `printk()` statement를 활성화합니다.
scftorture.weight_single=[KNL] probability weight`wait=0`인 `smp_call_function_single()`의 선택 가중치입니다. 모든 weight가 -1이면 -1은 기본값을 고르지만 하나라도 다른 값이면 -1은 weight 0을 뜻합니다.
scftorture.weight_single_wait=[KNL] probability weight`wait`가 0이 아닌 `smp_call_function_single()`의 가중치입니다. `weight_single` 규칙을 따릅니다.
scftorture.weight_many=[KNL] probability weight`wait=0`인 `smp_call_function_many()`의 가중치입니다. 높은 값은 system에 매우 큰 IPI 부하를 줄 수 있습니다.
scftorture.weight_many_wait=[KNL] probability weight`wait`가 0이 아닌 `smp_call_function_many()`의 가중치입니다. `weight_single`·`weight_many` 규칙을 따릅니다.

전체 CPU 호출 weight

6644-6655
매개변수적용·형식한국어 설명
scftorture.weight_all=[KNL] probability weight`wait=0`인 `smp_call_function_all()`의 가중치입니다. `weight_single`·`weight_many` 규칙을 따릅니다.
scftorture.weight_all_wait=[KNL] probability weight`wait`가 0이 아닌 `smp_call_function_all()`의 가중치입니다. `weight_single`·`weight_many` 규칙을 따릅니다.

SoundWire·tick skew·security·SELinux

6656-6681
매개변수적용·형식한국어 설명
sdw_mclk_divider=[SDW]Intel SoundWireBIOS가 clock rate를 올바르게 제공하지 않을 때 Intel SoundWire bus의 MCLK divider를 지정합니다.
skew_tick=[KNL,EARLY] `{ 0 | 1 }`CPU별 periodic timer tick을 엇갈리게 하여 큰 system의 `xtime_lock` contention과 `CONFIG_MAXSMP` system의 RCU lock contention을 줄입니다. 0은 비활성화, 1은 활성화이며 `CONFIG_CMDLINE="skew_tick=1"`로 기본값이 1일 수 있습니다. 전력 소비가 늘므로 jitter에 민감한 HPC·RT workload에서만 사용해야 합니다.
security=[SECURITY] legacy major LSM부팅 시 사용할 legacy major security module을 고릅니다. `lsm=`으로 대체되어 deprecated입니다.
selinux=[SELINUX] `{ 0 | 1 }`, 기본 `1`부팅 시 SELinux를 끄거나 켭니다. `security/selinux/Kconfig` 도움말을 참조하십시오.
serialnumber[BUGS=X86-32]x86-32 processor serial number 관련 workaround 매개변수입니다.

SEV·network shaper·local APIC dump

6682-6704
매개변수적용·형식한국어 설명
sev=option[,option...][X86-64] `{ debug | nosnp }``debug`는 debug message를 켭니다. host·hypervisor에만 적용되는 `nosnp`는 SEV-SNP를 활성화하지 않아 SEV-SNP guest를 실행하지 않을 때 memory access의 RMP check overhead를 피합니다.
shapers=[NET] 정수최대 network shaper 수를 지정합니다.
show_lapic=[APIC,X86] `{ 1 | 2 | ... | all }`, 기본 `1`dump할 local APIC의 최대 수를 정하며 `all`은 제한하지 않습니다. `apic=debug` 또는 `apic=verbose`를 지정했을 때만 유효합니다. 예시는 `apic=debug show_lapic=all`입니다.

SLAB·SLUB debug와 merge·NUMA 정책

6705-6771
매개변수적용·형식한국어 설명
slab_debug[=options[,slabs][;[options[,slabs]]...][MM]slab object corruption 원인을 찾도록 guard zone, unused object poisoning, 마지막 alloc/free 추적을 활성화할 수 있습니다. `Documentation/admin-guide/mm/slab.rst`를 참조하십시오. legacy 이름 `slub_debug`도 현재 허용됩니다. 이 옵션은 `no_hash_pointers`를 암시하며 `hash_pointers=always`를 추가하면 되돌릴 수 있습니다.
slab_max_order=[MM] page orderslab에 허용할 최대 order입니다. 값이 높으면 memory fragmentation으로 OOM이 날 수 있습니다. legacy `slub_max_order`도 허용되며 자세한 내용은 `Documentation/admin-guide/mm/slab.rst`를 참조하십시오.
slab_merge[MM]`CONFIG_SLAB_MERGE_DEFAULT` 없이 build한 kernel에서 크기가 비슷한 slab의 merge를 활성화합니다. legacy `slub_merge`도 허용됩니다.
slab_min_objects=[MM] object 수slab당 최소 object 수입니다. SLUB는 이 수를 담을 수 있도록 `slab_max_order`까지 order를 높입니다. object 수가 많을수록 slab 추적 overhead와 lock 획득 빈도가 줄어듭니다. legacy `slub_min_objects`도 허용됩니다.
slab_min_order=[MM] page orderslab의 최소 page order이며 `slab_max_order` 이하여야 합니다. legacy `slub_min_order`도 허용됩니다.
slab_nomerge[MM]크기가 비슷한 slab의 merge를 끕니다. 서로 다른 slab allocation을 구분해야 할 때, 특히 heap overflow와 attacker의 layout 제어 위험을 줄이는 hardened 환경에 유용합니다. heap 공격 노출을 대체로 한 cache로 제한하지만 metadata 공격 위험은 거의 변하지 않습니다. debug 옵션은 자체적으로 merge를 끕니다. legacy `slub_nomerge`도 허용됩니다.
slab_strict_numa[MM]slab allocator의 object 단위 memory policy를 지원합니다. 기본은 새 folio가 필요하거나 partial folio를 list에서 가져올 때 folio 단위로 policy를 적용합니다. slab fast path overhead는 늘지만 느린 interconnect의 NUMA system에서 kernel object 배치를 더 정확하게 합니다.

Legacy hardware·CSD lock·SMC IrDA

6772-6805
매개변수적용·형식한국어 설명
slram=[HW,MTD]SLRAM MTD driver 매개변수입니다.
smart2=[HW] `<io1>[,<io2>[,...,<io8>]]`Smart Array controller의 I/O 주소를 최대 8개 지정합니다.
smp.csd_lock_timeout=[KNL] milliseconds, 기본 `5000``smp_call_function()` 계열이 CPU의 CSD lock 해제를 기다릴 시간을 정합니다. CPU가 interrupt를 오래 끄는 bug 진단에 유용하고 0은 기능을 끕니다. `csdlock_debug-` kernel parameter로 더 효율적으로 끌 수 있습니다.
smp.panic_on_ipistall=[KNL] milliseconds`csd_lock_timeout`이 지정 시간보다 길어지면 panic을 일으킵니다. 기본은 제한 없이 기다립니다. `300000`은 5분 timeout입니다.
smsc-ircc2.nopnp[HW]SMC 장치 탐지에 PNP를 사용하지 않습니다.
smsc-ircc2.ircc_cfg=[HW] I/O port장치 configuration I/O port를 지정합니다.
smsc-ircc2.ircc_sir=[HW] I/O portSIR base I/O port를 지정합니다.
smsc-ircc2.ircc_fir=[HW] I/O portFIR base I/O port를 지정합니다.
smsc-ircc2.ircc_irq=[HW] IRQIRQ line을 지정합니다.
smsc-ircc2.ircc_dma=[HW] DMADMA channel을 지정합니다.
smsc-ircc2.ircc_transceiver=[HW] `{ 0 | 1 | 2 }`transceiver type을 정합니다. 0은 Toshiba Satellite 1800의 GP data pin select, 1은 기본 Fast pin select, 2는 ATC IRMode입니다.

SMT·soft-lockup·Sony I/O

6806-6830
매개변수적용·형식한국어 설명
smt=[KNL,MIPS,S390,EARLY] 정수, 기본 `-1`SMT system에서 physical CPU당 사용할 최대 thread, 즉 logical CPU 수를 정합니다. 실제 hardware 한도를 넘을 수 없으며 -1은 제한 없음입니다.
softlockup_panic=[KNL] `{ 0 | 1 }`1이면 soft-lockup detector가 soft lockup에서 machine을 panic시킵니다. `kernel.softlockup_panic` sysctl과 build-time `CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC`도 이 기능을 제어합니다.
softlockup_all_cpu_backtrace=[KNL] `{ 0 | 1 }`soft-lockup detector가 모든 CPU의 backtrace를 생성할지 정합니다.
sonypi.*=[HW]Sony Programmable I/O Control Device driver 설정입니다. `Documentation/admin-guide/laptops/sonypi.rst`를 참조하십시오.

BHI와 kernel Spectre v2 완화

6831-6885
매개변수적용·형식한국어 설명
spectre_bhi=[X86] `{ on | vmexit | off }`Branch History Injection 완화에서 hardware BHI control과 software BHB clearing sequence 배치를 제어합니다. 기본 `on`은 필요에 따라 HW 또는 SW 완화를 켜 kernel을 system call과 VM 양쪽 공격에서 보호합니다. `vmexit`은 HW 완화가 없는 system에서 vmexit에만 SW 완화를 적용하여 VM 발 BHI는 막지만 system call 공격에는 취약할 수 있습니다. `off`는 완화를 끕니다.
spectre_v2=[X86,EARLY] mitigation modeSpectre variant 2 indirect branch speculation을 완화합니다. `on`은 무조건 켜고 `spectre_v2_user=on`을 암시합니다. `off`는 모두 끄고 `spectre_v2_user=off`를 암시합니다. `auto`는 CPU 취약성을 탐지합니다. `on`과 경우에 따라 `auto`는 CPU, microcode, `CONFIG_MITIGATION_RETPOLINE`, compiler에 따라 runtime 완화를 고릅니다. 직접 선택은 `retpoline`, `retpoline,generic`, `retpoline,lfence`, 그 alias인 `retpoline,amd`, `eibrs`, `eibrs,retpoline`, `eibrs,lfence`, kernel 보호용 `ibrs`입니다. 생략은 `spectre_v2=auto`와 같습니다.

Userspace Spectre v2와 RAS overflow

6886-6937
매개변수적용·형식한국어 설명
spectre_v2_user=[X86] mitigation modeuserspace task 사이 Spectre v2를 제어합니다. `on`은 무조건 활성화되고 `spectre_v2=on`이 강제합니다. `off`는 무조건 비활성화되고 `spectre_v2=off`가 강제합니다. `prctl`은 기본적으로 indirect branch speculation을 허용하되 thread별 prctl로 완화를 켜며 상태는 fork에 상속됩니다. `prctl,ibpb`는 STIBP만 thread별로 제어하고 서로 다른 userspace process 전환 때 항상 IBPB를 발행합니다. `seccomp`는 prctl 방식에 더해 모든 seccomp thread가 명시적으로 opt-out하지 않으면 완화를 켭니다. `seccomp,ibpb`도 STIBP만 thread별 제어하고 process 전환마다 IBPB를 발행합니다. `auto`는 CPU 기능과 취약성에 따라 선택합니다. 기본 완화는 `prctl`이며 생략은 `spectre_v2_user=auto`입니다.
spec_rstack_overflow=[X86,EARLY] AMD ZenRAS return-address stack overflow 완화를 제어합니다. `off`는 끔, `microcode`는 microcode 완화만, 기본 `safe-ret`은 software-only safe RET, `ibpb`는 kernel 진입 시 IBPB, `ibpb-vmexit`은 cloud용으로 VMEXIT에서만 IBPB를 발행합니다.

Speculative Store Bypass 완화

6938-6989
매개변수적용·형식한국어 설명
spec_store_bypass_disable=[HW,EARLY] architecture별 modeSpeculative Store Bypass 최적화 사용을 제어합니다. 취약 CPU에서는 speculative store forwarding이 cache side-channel로 보호 memory를 읽는 데 악용될 수 있습니다. x86의 `on`은 SSB를 무조건 끄고, `off`는 무조건 켜며, `auto`는 CPU 구현과 취약성을 탐지해 architecture·Kconfig에 맞는 완화를 고릅니다. `prctl`은 process 기본 SSB 활성 상태에서 thread별 제어하고 fork에 상속합니다. `seccomp`는 명시적으로 opt-out하지 않은 모든 seccomp thread에서 SSB를 끕니다. x86 기본 완화는 `prctl`입니다. PowerPC의 `on,auto`는 Power8·Power9의 kernel 진입·종료에 store-forwarding barrier를 넣고 Power7에서는 software flush를 수행하며 `off`는 아무 작업도 하지 않습니다. 생략은 `spec_store_bypass_disable=auto`와 같습니다.

Split lock·bus lock·SRBDS

6990-7049
매개변수적용·형식한국어 설명
split_lock_detect=[X86] modehardware가 지원하면 cache line을 가로지르는 atomic access의 split lock에는 alignment-check `#AC`, bus lock에는 debug `#DB` exception을 사용합니다. `off`는 비활성화합니다. 기본 `warn`은 application의 `#AC`·`#DB`를 rate-limit해 경고하며 둘 다 지원하면 기본적으로 `#AC`를 사용합니다. `fatal`은 exception을 일으킨 application에 `SIGBUS`를 보내며 같은 우선순위를 따릅니다. `ratelimit:N`은 bus-lock detection을 초당 N회, `0 < N <= 1000`으로 제한하고 split-lock에는 적용되지 않습니다. kernel이나 firmware에서 `#AC`가 나면 warn·fatal 모두 kernel oops가 발생하며 bus-lock `#DB`는 `CPL > 0`일 때만 발생합니다.
srbds=[X86,INTEL,EARLY] `off`RNG bit를 유출할 수 있는 MDS 유사 SRBDS(Special Register Buffer Data Sampling)를 제어합니다. 기본 microcode 완화는 `RDRAND`·`RDSEED`를 크게 느리게 하고 `/dev/urandom` throughput도 줄일 수 있습니다. `off`는 완화를 끄고 이 성능 비용을 제거합니다.

SRCU big form 전환

7050-7073
매개변수적용·형식한국어 설명
srcutree.big_cpu_lim[KNL] CPU 수, 기본 `128`large system으로 간주해 `srcu_struct`가 즉시 `srcu_node` array를 할당할 CPU 수입니다. `srcutree.convert_to_big`의 하위 4bit가 3, 즉 boot-time 결정일 때만 효과가 있습니다.
srcutree.convert_to_big[KNL] modeSRCU tree `srcu_struct`를 `rcu_node` tree를 가진 big form으로 바꿀 조건입니다. 0은 never, 1은 `init_srcu_struct()` 시점, 2는 rcutorture가 결정, 기본 3은 boot-time 결정입니다. `0x1X`는 해당 조건에 high contention 전환을 추가합니다. tree 크기는 compile-time `CONFIG_NR_CPUS`가 아니라 runtime `nr_cpu_ids`를 기준으로 합니다.

SRCU wrap·expedite·poll delay

7074-7108
매개변수적용·형식한국어 설명
srcutree.counter_wrap_check[KNL] bit pattern`srcu_data`의 `->srcu_gp_seq_needed` grace-period sequence counter wrap 검사 빈도를 정합니다. 설정된 bit가 많을수록 덜 자주 검사하며 하위 2bit는 무시합니다.
srcutree.exp_holdoff[KNL] nanoseconds한 `srcu_struct`의 마지막 SRCU grace period가 끝난 뒤 다음 normal grace period를 자동 expedite 후보로 보기까지의 시간을 정합니다. 0은 자동 expedite를 끕니다.
srcutree.srcu_max_nodelay[KNL] instance/jiffySRCU grace-period worker를 지연 0으로 reschedule할 jiffy당 no-delay instance 수입니다. 한도를 넘으면 1 jiffy sleep delay로 reschedule합니다.
srcutree.srcu_max_nodelay_phase[KNL] poll 수/phasegrace-period phase마다 sleep 없이 reader를 poll할 횟수입니다. 한도를 넘으면 reader rescan 사이마다 1 jiffy sleep delay로 worker를 reschedule합니다.
srcutree.srcu_retry_check_delay[KNL] microsecondsreader를 sleep 없이 연속 poll할 때 각 poll 사이의 non-sleeping delay를 정합니다.

SRCU contention 기반 big 전환

7109-7116
매개변수적용·형식한국어 설명
srcutree.small_contention_lim[KNL] event/jiffy`srcu_struct`를 big form으로 바꾸기 전에 허용할 jiffy당 update-side contention event 수입니다. contention 기반 전환에는 `srcutree.convert_to_big`의 `0x10` bit가 설정되어 있어야 합니다.

ARM64 Speculative Store Bypass Disable

7117-7133
매개변수적용·형식한국어 설명
ssbd=[ARM64,HW,EARLY] `{ force-on | force-off | kernel }`firmware 기반 완화를 제공하는 SSB 취약 CPU에서 사용 방식을 정합니다. `force-on`은 kernel과 userspace 모두에 무조건 완화를 켜고, `force-off`는 모두 끕니다. `kernel`은 kernel에서는 항상 켜고 userspace가 완화를 요청할 수 있도록 prctl interface를 제공합니다.

Stack guard·depot·tracer

7134-7162
매개변수적용·형식한국어 설명
stack_guard_gap=[MM] page 수, 기본 `256`기본 stack gap 보호를 덮어씁니다. 아래로 자라는 stack 앞 또는 위로 자라는 stack 뒤에서 다른 mapping에 쓰지 않도록 예약할 page 수입니다.
stack_depot_disable=[KNL,EARLY] bool, 기본 falsecommand line에서 true로 두면 stack depot을 비활성화하여 stack hash table이 소비하는 static memory를 아낍니다.
stack_depot_max_pools=[KNL,EARLY] pool 수, 기본 `8191`stack trace 저장에 사용할 최대 pool 수입니다. pool은 이 한도까지 필요할 때 할당합니다.
stacktrace[FTRACE]부팅 시 stack tracer를 활성화합니다.
stacktrace_filter=[function-list][FTRACE] 쉼표 구분 함수부팅 시 stack tracer가 추적할 함수를 제한합니다. 실행 중에는 debugfs tracing directory의 `stack_trace_filter`에서 바꿀 수 있습니다. 이 옵션 자체가 stack tracing을 켜므로 별도 `stacktrace`는 필요하지 않습니다.

HP-PARISC STI와 PPC stress

7163-7194
매개변수적용·형식한국어 설명
sti=[PARISC,HW] `<num>`HP-PARISC의 built-in display·keyboard인 STI console 중 초기 boot console로 사용할 graphic card를 정합니다. `drivers/video/console/sticore.c` 주석도 참조하십시오.
sti_font=[HW]STI console font 설정입니다. `drivers/video/console/sticore.c` 주석을 참조하십시오.
stifb=[HW] `bpp:<bpp1>[:<bpp2>[:<bpp3>...]]`STI framebuffer의 bits-per-pixel 값을 지정합니다.
strict_sas_size=[X86] bool지원 FPU 기능에 따라 달라지는 필수 signal frame 크기를 기준으로 strict sigaltstack size 검사를 켜거나 끕니다. 아직 `AT_MINSIGSTKSZ`를 인식하지 못하는 binary를 걸러낼 수 있습니다.
stress_hpt[PPC,EARLY]hash page table의 kernel HPT entry 수를 제한하여 kernel address의 hash-page-table fault 빈도를 높입니다.
stress_slb[PPC,EARLY]kernel SLB entry 수를 제한하고 자주 flush하여 kernel address의 SLB fault 빈도를 높입니다.

SunRPC port·pool·slot table

7195-7237
매개변수적용·형식한국어 설명
sunrpc.min_resvport=[NFS,SUNRPC] portSunRPC client가 privileged로 간주할 port 범위의 최솟값을 정합니다. server는 흔히 `0 < portnr < 1024`에서 온 request를 요구하며 일부 port를 다른 용도로 예약할 때 범위를 조정합니다.
sunrpc.max_resvport=[NFS,SUNRPC] portSunRPC client가 privileged로 간주할 port 범위의 최댓값을 정합니다.
sunrpc.svc_rpc_per_connection_limit=[NFS,SUNRPC] request 수, 기본 `0`server가 connection 하나에서 병렬 처리할 request 수를 제한합니다. 0은 제한 없음입니다.
sunrpc.pool_mode=[NFS] `{ auto | global | percpu | pernode }`NFS server service thread pool에 CPU를 배정하는 방법입니다. NIC 수와 interrupt binding에 따라 실제 serving CPU가 달라집니다. server 실행 중에는 바꿀 수 없습니다. `auto`는 heuristic 자동 선택, `global`은 모든 CPU를 한 pool에, `percpu`는 CPU별 pool, `pernode`는 NUMA node별 pool이며 non-NUMA에서는 global과 같습니다.
sunrpc.tcp_slot_table_entries=[NFS,SUNRPC] 동시 RPC 수client가 server로 보낼 수 있는 동시 TCP RPC call 상한입니다. 값을 늘리면 throughput이 좋아질 수 있지만 client 예약 memory도 늘어납니다.
sunrpc.udp_slot_table_entries=[NFS,SUNRPC] 동시 RPC 수client가 server로 보낼 수 있는 동시 UDP RPC call 상한입니다. 값을 늘리면 throughput과 예약 memory가 함께 늘어납니다.

Suspend test·SWIOTLB·sysctl·SysRq

7238-7276
매개변수적용·형식한국어 설명
suspend.pm_test_delay=[SUSPEND] seconds, 기본 `5`, `CONFIG_PM_DEBUG``/sys/power/pm_test` suspend test mode에서 resume하기 전 머무를 시간을 정합니다.
svm=[PPC] `{ on | off | y | n | 1 | 0 }`pSeries의 Protected Execution Facility 사용을 제어합니다.
swiotlb=[ARM,PPC,MIPS,X86,S390,EARLY] `{ <int>[,<int>] | force | noforce }`첫 정수는 I/O TLB slab 수, 쉼표 뒤 두 번째 정수는 각자 lock을 가진 SWIOTLB area 수이며 2의 거듭제곱으로 올림합니다. `force`는 kernel이 자동으로 쓰지 않을 상황에도 bounce buffer를 강제하고 `noforce`는 debug 목적으로 절대 사용하지 않습니다.
switches=[HW,M68k,EARLY]M68k hardware switch 설정입니다.
sysctl.*=[KNL]init process를 load하기 직전에 대응하는 `/proc/sys/...`에 쓴 것처럼 sysctl을 설정합니다. `.`과 `/`를 separator로 모두 인식합니다. 알 수 없는 parameter와 잘못된 값은 kernel log에 보고되며 나중에 module이 등록하는 sysctl은 설정할 수 없습니다. 예시는 `sysctl.vm.swappiness=40`입니다.
sysrq_always_enabled[KNL]`/proc/sys/kernel/sysrq` 설정의 효과를 무시하고 SysRq를 항상 허용합니다. debug에 유용합니다.

TCP hash·suspend test·ACPI thermal

7277-7318
매개변수적용·형식한국어 설명
tcpmhash_entries=[KNL,NET] slot 수TCP metrics cache의 `tcp_metrics_hash` slot 수입니다. 총 RAM page에 따라 기본 8192 또는 16384입니다. `Documentation/networking/ip-sysctl.rst`의 `tcp_no_metrics_save`를 참조하십시오.
tdfx=[HW,DRM]3Dfx DRM driver 매개변수입니다.
test_suspend=[SUSPEND] `{ mem | standby | freeze }[,N]`system startup 중 Suspend-to-RAM, standby, freeze 중 지정한 sleep state에 들어가며 선택적으로 N회 반복합니다. wakeup-capable RTC alarm으로 깨웁니다.
thash_entries=[KNL,NET] bucket 수TCP connection hash bucket 수를 정합니다.
thermal.act=[HW,ACPI] `-1` 또는 섭씨-1은 모든 thermal zone의 active trip point를 끄고 온도 값은 모든 lowest active trip point를 덮어씁니다.
thermal.crt=[HW,ACPI] `-1` 또는 섭씨-1은 모든 critical trip point를 끄고 온도 값은 모두 덮어씁니다.
thermal.off=[HW,ACPI] `1`ACPI thermal control을 비활성화합니다.
thermal.psv=[HW,ACPI] `-1` 또는 섭씨-1은 모든 passive trip point를 끄고 온도 값은 모두 덮어씁니다.
thermal.tzp=[HW,ACPI] deciseconds, 기본 `0`전체 ACPI thermal zone polling 주기를 정합니다. 0은 polling하지 않습니다.

THP·threaded IRQ·topology·torture

7319-7368
매개변수적용·형식한국어 설명
thp_anon=[KNL] size range와 stateanonymous transparent hugepage의 크기별 기본 동작을 정합니다. 형식은 `<size>[KMG],<size>[KMG]:<state>;<size>[KMG]-<size>[KMG]:<state>`이고 state는 `always`, `madvise`, `never`, `inherit`입니다. 여러 크기에 반복 지정할 수 있으며 `Documentation/admin-guide/mm/transhuge.rst`를 참조하십시오.
threadirqs[KNL,EARLY]`IRQF_NO_THREAD`로 명시한 handler를 제외한 모든 interrupt handler를 thread화합니다.
thp_shmem=[KNL] size range와 policyinternal shmem mount의 hugepage 크기별 기본 policy를 정합니다. policy는 `always`, `inherit`, `never`, `within_size`, `advise`이며 여러 크기에 반복 지정할 수 있습니다. `Documentation/admin-guide/mm/transhuge.rst`를 참조하십시오.
topology=[S390,EARLY] `{ off | on }`, 기본 `on`hardware가 지원하면 kernel과 scheduler가 CPU topology 정보를 사용해 process migration 같은 결정을 내리도록 합니다.
torture.disable_onoff_at_boot=[KNL] boolinit이 생성될 때까지 torture test의 CPU-hotplug component를 막습니다.
torture.ftrace_dump_at_shutdown=[KNL] boolerror가 없어도 torture-test 종료 시 ftrace buffer를 dump합니다. 여러 test가 동시에 실행되고 회전식 storage를 쓰면 비용이 매우 클 수 있습니다.
torture.verbose_sleep_frequency=[KNL] printk 수, 기본 `0`sleep 사이에 출력할 verbose `printk()` 수입니다. 0은 verbose-printk sleep을 끕니다.
torture.verbose_sleep_duration=[KNL] jiffies각 verbose-printk sleep의 지속 시간을 정합니다.

TPM과 tracepoint printk

7369-7423
매개변수적용·형식한국어 설명
tpm.disable_pcr_integrity=[HW,TPM] bool`TPM2_PCR_Extend`를 integrity-protected session으로 감싸 PCR register를 의도치 않은 physical access나 bus interposer로부터 보호하는 기능을 끕니다. IMA가 TPM을 많이 써 성능 비용이 크고 물리 환경이 다른 수단으로 보호될 때 고려할 수 있습니다.
tpm_crb_ffa.busy_timeout_ms=[ARM64,TPM] unsigned int, 기본 `2000`TPM service가 busy일 때 message 전송을 포기하기 전까지 재시도할 최대 시간을 millisecond로 정합니다.
tpm_suspend_pcr=[HW,TPM]PCR ID 정수suspend 시 지정 PCR을 0으로 extend합니다. `TPM_SaveState`에서 마지막 PCR write를 flush하지 못하는 chip의 workaround이며 다른 PCR이 모두 저장되도록 보장합니다.
tpm_tis.interrupts=[HW,TPM] bool, 기본 falseFIFO interface의 MMIO physical layer interrupt를 활성화합니다. TCG hardware interface는 `https://trustedcomputinggroup.org/resource/pc-client-platform-tpm-profile-ptp-specification/`를 참조하십시오.
tp_printk[FTRACE]tracepoint를 tracing ring buffer뿐 아니라 printk에도 보냅니다. buffer를 읽거나 `ftrace_dump_on_oops`를 수행하기 전에 hang·reboot되는 early boot debug에 유용합니다. `tp_printk_stop_on_boot`으로 `late_initcall_sync` 시점에 console 출력을 멈출 수 있습니다. irq·sched 같은 고빈도 tracepoint와 함께 쓰면 live lock이 날 수 있으므로 주의해야 합니다.

tracepoint의 printk 전달을 실행 중 끄는 원문 명령은 다음과 같습니다.

			 echo 0 > /proc/sys/kernel/tracepoint_printk

Boot trace 출력과 clock

7424-7457
매개변수적용·형식한국어 설명
tp_printk_stop_on_boot[FTRACE]`tp_printk`의 많은 console 출력을 boot 중에만 허용하고 `late_initcall_sync()` 시점에 멈춥니다.
trace_buf_size=nn[KMG][FTRACE] CPU별 크기각 CPU의 tracing buffer 크기를 정합니다.
trace_clock=[FTRACE] clock 이름boot trace event의 clock을 고릅니다. `local`은 빠른 per-CPU timestamp지만 CPU 간 동기화되지 않을 수 있습니다. `global`은 CPU 간 동기화되지만 느릴 수 있습니다. `counter`는 event를 1, 2처럼 세며 clock을 여러 번 읽어 count가 건너뛸 수 있습니다. `uptime`은 jiffies, `perf`는 perf clock, `mono`는 `ktime_get_mono_fast_ns()`, `mono_raw`는 `ktime_get_raw_fast_ns()`, `boot`는 `ktime_get_boot_fast_ns()`를 사용합니다. architecture가 clock을 추가할 수 있으며 `Documentation/trace/ftrace.rst`를 참조하십시오.

Early boot trace event

7458-7471
매개변수적용·형식한국어 설명
trace_event=[event-list][FTRACE] 쉼표 구분 eventearly boot debug를 위해 지정 trace event를 설정하고 시작합니다. module event는 `:mod:<module>` keyword를 사용하며 그 앞에 값을 두면 해당 module의 특정 event만 활성화합니다. `Documentation/trace/events.rst`를 참조하십시오.

module event 활성화의 원문 형식은 다음과 같습니다.

			trace_event=:mod:<module>

Early tracing instance와 persistent buffer

7472-7544
매개변수적용·형식한국어 설명
trace_instance=[instance-info][FTRACE]early boot에 ring-buffer instance를 만들며 `/sys/kernel/tracing/instances`에 나타납니다. 생성과 함께 `<name>,<system>:<event>` 형식으로 event를 켤 수 있고 event가 unique하면 system prefix를 생략할 수 있습니다. `sched:sched_switch,irq_handler_entry,initcall` 예시는 sched_switch, irq_handler_entry, initcall system의 모든 event를 켭니다. instance 이름 뒤 `^traceoff`는 생성 직후 tracing을 끄고 `^traceprintk`는 `trace_printk()`를 이 instance에 기록하며 flag는 event보다 앞에 와야 합니다. x86 `memmap`으로 예약한 page-aligned physical memory 또는 `reserve_mem` label을 buffer에 사용할 수 있습니다. KASLR로 주소가 달라질 수 있고 kernel version 사이 ring-buffer layout이 바뀌면 validator가 buffer를 reset할 수 있습니다. reboot 사이 event가 섞이지 않도록 persistent boot buffer에는 `traceoff`를 권장합니다. memory를 지우지 않는 system이어야 하며 `CONFIG_RESET_ATTACK_MITIGATION` 같은 기능은 저장 trace를 지울 수 있습니다. `Documentation/trace/debugging.rst`를 참조하십시오.

instance event·flag·예약 memory의 원문 예시는 다음과 같습니다.

				trace_instance=<name>,<system1>:<event1>,<system2>:<event2>
				trace_instance=foo,sched:sched_switch,irq_handler_entry,initcall
				trace_instance=foo^traceoff^traceprintk,sched,irq
				memmap=12M$0x284500000 trace_instance=boot_map@0x284500000:12M
				reserve_mem=12M:4096:trace trace_instance=boot_map@trace
				reserve_mem=12M:4096:trace trace_instance=boot_map^traceoff^traceprintk@trace,sched,irq

Trace option·trigger·boot 종료 정책

7545-7601
매개변수적용·형식한국어 설명
trace_options=[option-list][FTRACE] 쉼표 구분 optionboot 시 tracer option을 켜거나 끕니다. `/sys/kernel/tracing/trace_options`에 option 이름을 쓰는 것과 같습니다. 예를 들어 `stacktrace`는 각 event의 stack trace를 dump합니다. `Documentation/trace/ftrace.rst`의 trace options를 참조하십시오.
trace_trigger=[trigger-list][FTRACE] `trace_trigger=<event>.<trigger>[ if <filter>],...`특정 event에 optional filter가 있는 trigger를 추가합니다. `sched_switch.stacktrace if prev_state == 2`는 `sched_switch`의 `prev_state`가 2, 즉 `TASK_UNINTERRUPTIBLE`일 때만 stacktrace trigger를 실행합니다. `Documentation/trace/events.rst`의 Event triggers를 참조하십시오.
traceoff_after_boot[FTRACE]boot가 끝나면 main tracing buffer를 꺼 limited buffer의 중요한 boot 정보가 이후 event로 덮이지 않게 합니다.
traceoff_on_warning[FTRACE]warning에서 `tracing_on`을 꺼 warning 출력이 trace를 채우기 전에 tracing을 멈춥니다. `/sys/kernel/tracing/tracing_on`에 1을 써 다시 켤 수 있으며 실행 중 sysctl `kernel/traceoff_on_warning`으로도 설정할 수 있습니다.

trace option 경로·예시와 filtered trigger 원문은 다음과 같습니다.

			    /sys/kernel/tracing/trace_options
			      trace_options=stacktrace
			  trace_trigger="sched_switch.stacktrace if prev_state == 2"

Transparent hugepage와 trusted key backend

7602-7659
매개변수적용·형식한국어 설명
transparent_hugepage=[KNL] `{ always | madvise | never }`시스템의 transparent hugepage 기본 동작을 제어합니다. `always`는 가능한 mapping에 항상 THP를 시도하고, `madvise`는 명시적으로 요청한 영역에 적용하며, `never`는 기본 사용을 막습니다. 자세한 내용은 `Documentation/admin-guide/mm/transhuge.rst`를 참조하십시오.
transparent_hugepage_shmem=[KNL] `{ always | within_size | advise | never | deny | force }`내부 shmem mount의 hugepage 할당 정책을 제어합니다. `within_size`는 파일 크기 안에서만, `advise`는 요청된 경우에 적용하며 `deny`와 `force`를 포함한 정책 의미는 `Documentation/admin-guide/mm/transhuge.rst`에 설명되어 있습니다.
transparent_hugepage_tmpfs=[KNL] `{ always | within_size | advise | never }`tmpfs mount의 기본 hugepage 할당 정책을 설정합니다. 세부 정책과 page 크기별 동작은 `Documentation/admin-guide/mm/transhuge.rst`를 참조하십시오.
trusted.source=[KEYS] string: `tpm`, `tee`, `caam`, `dcp`Trusted key 구현이 사용할 신뢰 원천 backend를 지정합니다. 생략하면 TPM부터 신뢰 원천 목록을 순회하며 초기화에 처음 성공한 backend를 선택합니다.
trusted.rng=[KEYS] string: `kernel`, `tpm`, `tee`, `default`Trusted key의 key material을 생성할 RNG를 선택합니다. `tpm` 또는 `tee`는 `trusted.source`와 같은 값을 사용할 수 있습니다. 생략 시 `default`이며, 이 경우 각 신뢰 원천이 RNG를 선택합니다.
trusted.dcp_use_otp_key[KEYS] `trusted.source=dcp`와 함께 사용DCP UNIQUE key blob 암호화 대신 DCP OTP key를 선택합니다.
trusted.dcp_skip_zk_test[KEYS] `trusted.source=dcp`와 함께 사용Blob key가 모두 0인지 검사하는 절차를 비활성화합니다. 0 key가 허용되는 시험 환경 등에 유용합니다.

AMD TSA와 TSC clocksource

7660-7707
매개변수적용·형식한국어 설명
tsa=[X86] `{ off | on | user | vm }`AMD CPU의 Transient Scheduler Attack 완화를 제어합니다. `off`는 완화를 끄고, 기본값 `on`은 켭니다. `user`는 user/kernel 전환만, `vm`은 guest/host 전환만 완화합니다. 원문은 추가 기술 지침으로 “Technical guidance for mitigating transient scheduler attacks”를 검색하도록 안내합니다.
tsc=[X86] stringTSC clocksource 정책을 정합니다. `reliable`은 boot와 runtime 안정성 검사를 생략해 오래된 hardware나 가상 환경에서 high-resolution timer를 허용합니다. `noirqtime`은 느린 `RDTSC` 때문에 생기는 IRQ accounting 비용을 피하고, `unstable`은 TSC를 무조건 불안정으로 표시합니다. `nowatchdog`는 엄격한 latency 환경에서 clocksource watchdog을 끕니다. `recalibrate`는 MSR 또는 `CPUID(0x15)`로 얻은 주파수를 HPET/PM timer에 맞춰 재보정하고 500 ppm을 넘는 차이를 경고합니다. `watchdog`는 신뢰할 수 있는 TSC로 다른 timer를 검사하며 앞뒤 `tsc=nowatchdog`의 순서에 따라 억제하거나 다시 덮어쓸 수 있고 그 결과를 console에 알립니다.
tsc_early_khz=[X86,EARLY] unsigned int초기 TSC 보정을 건너뛰고 주어진 kHz 값을 사용합니다. `CPUID.16h`는 지원하지만 `CPUID.15h` 지원은 불완전한 overclocked system처럼 초기 주파수 탐색이 신뢰할 수 없는 환경에 유용합니다.

Intel TSX와 TAA 완화

7708-7777
매개변수적용·형식한국어 설명
tsx=[X86] `{ on | off | auto }`, 생략 시 `off`TSX control을 지원하는 Intel processor의 Transactional Synchronization Extensions를 제어합니다. `on`은 알려진 취약점 완화가 있더라도 TSX가 과거 speculative-execution CVE를 가속한 전력이 있어 알려지지 않은 위험이 남을 수 있습니다. `off`는 TSX를 끄며, MDS 비취약 CPU에서 microcode가 제공한 `IA32_TSX_CTRL` MSR로 신뢰성 있게 비활성화할 수 있을 때 적용됩니다. `auto`는 `X86_BUG_TAA`가 있으면 끄고 그렇지 않으면 켭니다. 자세한 내용은 `Documentation/admin-guide/hw-vuln/tsx_async_abort.rst`를 참조하십시오.
tsx_async_abort=[X86,INTEL,EARLY] `{ full | full,nosmt | off }`CPU 내부 buffer의 speculatively forwarded data를 cache side channel로 노출할 수 있는 TSX Async Abort(TAA) 취약점 완화를 제어합니다. 기본 `full`은 TSX가 활성화된 취약 CPU에서 완화를 켭니다. `full,nosmt`는 완화와 함께 SMT를 끄지만, TSX가 비활성화되어 cross-thread TAA에 취약하지 않으면 SMT를 끄지 않습니다. `off`는 TAA 완화를 무조건 끄려 하지만 MDS와 같은 mechanism을 쓰므로 MDS 완화가 활성화된 machine에서는 `mds=off`도 필요합니다. MDS 완화를 적용 중인 MDS 취약 CPU에는 별도 TAA 완화가 추가 이익을 주지 않습니다.

TurboGrafx, UDP, RISC-V unaligned와 unwinder

7778-7824
매개변수적용·형식한국어 설명
turbografx.map[2|3]=[HW,JOY] `<port#>,<js1>,...,<js7>`TurboGraFX parallel-port interface의 port와 최대 7개 joystick mapping을 지정합니다. `Documentation/input/devices/joystick-parport.rst`도 참조하십시오.
udbg-immortal[PPC]`console_init()` 뒤 proper console driver가 인계하기 전에 발생하는 early kernel crash를 디버깅할 때 udbg console을 계속 살려 두어 상황을 볼 수 있게 돕습니다.
uhash_entries=[KNL,NET] bucket 수UDP와 UDP-Lite connection hash bucket 수를 설정합니다.
uhci-hcd.ignore_oc=[USB] bool, 기본 `N`Overcurrent event를 무시합니다. 연결되지 않은 port에서 가짜 event를 많이 만드는 잘못 설계된 motherboard의 log 폭주를 막을 수 있지만 실제 overcurrent도 보고되지 않습니다.
unaligned_scalar_speed=[RISCV] `{ slow | fast | unsupported }`Scalar unaligned-access 속도 시험을 건너뛰고 결과를 강제로 지정합니다. 대체 code path 시험이나 시험이 지나치게 느린 환경에 유용하며 모든 CPU가 같은 scalar unaligned-access 속도를 가져야 합니다.
unaligned_vector_speed=[RISCV] `{ slow | fast | unsupported }`Vector unaligned-access 속도 시험을 건너뛰고 결과를 강제로 지정합니다. 모든 CPU가 같은 vector unaligned-access 속도를 가져야 합니다.
unknown_nmi_panic[X86]알 수 없는 NMI를 받으면 kernel panic을 일으킵니다.
unwind_debug[X86-64,EARLY]Unwinder debug 출력을 켭니다. 손상된 stack, 잘못되거나 누락된 unwinder metadata 등 unwinder 오류 조건을 진단할 때 유용합니다.

USB core authorization, 초기화와 quirk

7825-7932
매개변수적용·형식한국어 설명
usbcore.authorized_default=[USB] `-1`, `0`, `1`, `2`새 USB device의 기본 authorization을 정합니다. `-1`과 `1`은 허용, `0`은 불허, `2`는 internal port에 연결된 device만 허용입니다.
usbcore.autosuspend=[USB] seconds, 기본 `2`새로 감지한 USB device가 idle 상태에 들어간 뒤 autosuspend하기까지의 시간을 정합니다. 음수면 autosuspend하지 않습니다.
usbcore.usbfs_snoop=[USB] bool, 기본 `0`모든 usbfs traffic을 log에 기록합니다.
usbcore.usbfs_snoop_max=[USB] bytes, 기본 `65536`각 URB에서 snoop할 최대 byte 수를 정합니다.
usbcore.blinkenlights=[USB] bool, 기본 `0`Hub LED를 순환 점등합니다.
usbcore.old_scheme_first=[USB] bool, 기본 `0`오래된 device 초기화 scheme을 먼저 시도합니다.
usbcore.usbfs_memory_mb=[USB] MiB, 기본 `16`; `0`은 최대 `2047`usbfs가 할당하는 buffer의 memory 한도를 정합니다.
usbcore.use_both_schemes=[USB] bool, 기본 `1`첫 device 초기화 scheme이 실패하면 다른 scheme도 시도합니다.
usbcore.initial_descriptor_timeout=[USB] milliseconds, 기본 `5000`초기 64-byte `USB_REQ_GET_DESCRIPTOR` request의 timeout을 정합니다.
usbcore.nousb[USB]USB subsystem 전체를 비활성화합니다.
usbcore.quirks=[USB] `VendorID:ProductID:Flags,...`Built-in USB core quirk를 보충하거나 반전하는 목록입니다. ID는 4자리 hex이고 각 flag 문자는 현재 quirk bit를 toggle합니다. `a` `USB_QUIRK_STRING_FETCH_255`는 255-byte string descriptor fetch 금지, `b` `USB_QUIRK_RESET_RESUME`은 resume 대신 reset, `c` `USB_QUIRK_NO_SET_INTF`는 Set-Interface 금지, `d` `USB_QUIRK_CONFIG_INTF_STRINGS`는 Configuration/Interface string 회피, `e` `USB_QUIRK_RESET`은 reset 금지, `f` `USB_QUIRK_HONOR_BNUMINTERFACES`는 `bNumInterfaces`보다 많은 interface description을 존중합니다. `g` `USB_QUIRK_DELAY_INIT`은 descriptor read 뒤 초기화 지연, `h` `USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL`은 high/super-speed interrupt endpoint의 `bInterval`을 exponent가 아닌 계산된 microframe 간격으로 취급, `i` `USB_QUIRK_DEVICE_QUALIFIER`는 device-qualifier request 금지, `j` `USB_QUIRK_IGNORE_REMOTE_WAKEUP`은 가짜 remote wakeup 무시, `k` `USB_QUIRK_NO_LPM`은 Link Power Management 금지, `l` `USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL`은 `bInterval`을 linear frame으로 취급합니다. `m` `USB_QUIRK_DISCONNECT_SUSPEND`는 suspend 전에 disconnect, `n` `USB_QUIRK_DELAY_CTRL_MSG`는 control message마다 지연, `o` `USB_QUIRK_HUB_SLOW_RESET`은 hub port reset 뒤 추가 지연, `p` `USB_QUIRK_SHORT_SET_ADDRESS_REQ_TIMEOUT`은 SET_ADDRESS timeout을 5000 ms에서 500 ms로 단축, `q` `USB_QUIRK_FORCE_ONE_CONFIG`는 configuration 수가 0이어도 1로 강제합니다. 예: `quirks=0781:5580:bk,0a5c:5834:gij`.

USB HID polling과 mass-storage quirk

7933-8003
매개변수적용·형식한국어 설명
usbhid.mousepoll=[USBHID] polling intervalMouse polling 간격을 설정합니다.
usbhid.jspoll=[USBHID] polling intervalJoystick polling 간격을 설정합니다.
usbhid.kbpoll=[USBHID] polling intervalKeyboard polling 간격을 설정합니다.
usb-storage.delay_use=[UMS] seconds 또는 `ms` suffix, 기본 `1`초새 device에서 Logical Unit scan을 시작하기 전 지연입니다. `delay_use=2567ms`처럼 millisecond를 지정할 수 있습니다.
usb-storage.quirks=[UMS] `VID:PID:Flags,...``unusual_devs` 목록을 보충하거나 덮어씁니다. `a` SANE_SENSE는 18 byte보다 많은 sense data 수집(no UAS), `b` BAD_SENSE는 18 byte 초과 수집 금지(no UAS), `c` FIX_CAPACITY는 보고 capacity를 한 sector 줄이고, `d` NO_READ_DISC_INFO는 `READ_DISC_INFO`를 쓰지 않습니다(no UAS). `e`는 `READ_CAPACITY_16` 금지, `f`는 report-opcodes 금지(UAS), `g`는 한 번에 240 sector 이하(UAS), `h`는 odd capacity일 때 한 sector 감소, `i`는 device binding 무시, `j`는 report-luns 금지(UAS), `k`는 `WRITE_SAME` 금지(UAS), `l`은 ejectable media lock/unlock 금지(no UAS), `m`은 64 sector=32 KiB 이하(no UAS), `n`은 최초 `READ(10)` 재시도(no UAS), `o`는 device 보고 capacity 수용(no UAS), `p`는 기본 write cache가 켜졌다고 간주(no UAS), `r`은 잘못된 residue 무시(no UAS), `s`는 single LUN, `t`는 `ATA(12)`와 `ATA(16)` 금지(UAS), `u`는 UAS driver binding 금지, `w`는 write-protect 검사 금지, `y`는 cache가 없다고 주장해도 `SYNCHRONIZE_CACHE` 발행(no UAS)입니다. 예: `quirks=0419:aaf5:rl,0421:0433:rc`.

ARM user debug, vDSO, video와 virtio-mmio

8004-8062
매개변수적용·형식한국어 설명
user_debug=[KNL,ARM] integer bitmask`arch/arm/Kconfig.debug`의 user fault debug event bitmask입니다. 1은 undefined instruction, 2는 system call, 4는 invalid data abort, 8은 `SIGSEGV`, 16은 `SIGBUS`이며 `user_debug=31`은 모두 켭니다.
vdso=[X86,SH,SPARC] `0` 또는 `1`X86_32에서는 `vdso32=`의 alias입니다. 그 밖의 architecture에서 `1`은 VDSO mapping을 켜는 기본값이고 `0`은 끕니다.
vdso32=[X86] `0`, `1`, `2`32-bit vDSO를 제어합니다. `1`은 활성화, `0`과 호환 alias `2`는 비활성화입니다. `CONFIG_COMPAT_VDSO`가 설정되면 기본 `0`, 아니면 기본 `1`입니다. `dl_main: Assertion '(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed!` 오류가 나면 `vdso32=0`을 시도합니다.
video=[FB,EARLY]Framebuffer mode를 설정합니다. `Documentation/fb/modedb.rst`를 참조하십시오.
video.brightness_switch_enabled=[ACPI] `0` 또는 `1`, 기본 `1`Hotkey ACPI notify event를 받았을 때 `1`이면 video driver가 brightness를 조정한 뒤 input device를 통해 userspace에 event를 보냅니다. `0`이면 backlight를 바꾸지 않고 event만 보냅니다.
virtio_mmio.device=[VMMIO] `<size>@<baseaddr>:<irq>[:<id>]`Memory-mapped virtio platform device를 등록합니다. `size`에는 K/M/G suffix를 쓸 수 있고, `baseaddr`는 physical base address, `irq`는 `request_irq()`에 넘길 interrupt number, `id`는 optional platform-device ID입니다. 여러 device를 위해 매개변수를 반복할 수 있습니다.

ARM user debug bit와 virtio-mmio device 형식·예시는 원문 줄을 그대로 보존합니다.

				 1 - undefined instruction events
				 2 - system calls
				 4 - invalid data aborts
				 8 - SIGSEGV faults
				16 - SIGBUS faults
			Example: user_debug=31
				<size>@<baseaddr>:<irq>[:<id>]
			where:
				<size>     := size (can use standard suffixes
						like K, M and G)
				<baseaddr> := physical base address
				<irq>      := interrupt number (as passed to
						request_irq())
				<id>       := (optional) platform device id
			example:
				virtio_mmio.device=1K@0x100b0000:48:7

VGA, VM debug, S390 VM command와 VMscape

8063-8113
매개변수적용·형식한국어 설명
vga=[BOOT,X86-32] video mode특정 video mode를 선택합니다. `vga=ask`는 menu를 표시합니다. 실제로는 boot loader 매개변수이며 특별한 protocol로 kernel에 전달됩니다. `Documentation/arch/x86/boot.rst`와 `Documentation/admin-guide/svga.rst`를 참조하십시오.
vm_debug[=options][KNL] `CONFIG_DEBUG_VM=y`; `P`, `-`Virtual-memory debug feature를 선택적으로 켜거나 끕니다. 기본적으로 모든 option이 켜져 있고 큰 memory system에서는 boot가 느려질 수 있습니다. `P`는 page structure 초기화 시 poisoning을 켜며 `-`는 위 option을 모두 끕니다.
vmalloc=nn[KMG][KNL,BOOT,EARLY] exact sizevmalloc 영역을 정확히 `<nn>` 크기로 강제합니다. x86/arm32의 최소 128 MiB보다 늘릴 수 있고, 줄여서 direct-mapped kernel RAM 공간을 늘릴 수도 있습니다. arm64, alpha, loongarch, arc, csky, hexagon, microblaze, mips, nios2, openrisc, parisc, m64k, powerpc, riscv, sh, um, xtensa, s390, sparc 등 많은 platform에는 이 매개변수가 없습니다.
vmcp_cma=nn[MG][KNL,S390,EARLY]`vmcp` device driver의 contiguous memory allocation에 예약할 memory 크기를 설정합니다.
vmhalt=[KNL,S390] commandSystem halt 뒤 실행할 z/VM CP command를 지정합니다.
vmpanic=[KNL,S390] commandKernel panic 뒤 실행할 z/VM CP command를 지정합니다.
vmpoff=[KNL,S390] commandPower off 뒤 실행할 z/VM CP command를 지정합니다.
vmscape=[X86] `{ off | ibpb | force }`Userspace hypervisor에서 guest로 speculative side channel을 통해 정보를 유출할 수 있는 VMscape 공격 완화를 제어합니다. `off`는 끄고, 기본 `ibpb`는 Indirect Branch Prediction Barrier(IBPB)를 사용하며, `force`는 영향받지 않는 processor에서도 취약성 감지를 강제합니다.

x86 vsyscall과 VT console 기본값

8114-8182
매개변수적용·형식한국어 설명
vsyscall=[X86-64,EARLY] `{ emulate | xonly | none }`Legacy code가 고정 주소 `0xffffffffff600x00`으로 호출하는 vsyscall의 동작을 제어합니다. 고정 주소는 RIP를 통제하는 exploit 표적이 되기 쉽습니다. `emulate`는 trap으로 바꿔 비교적 안전하게 emulate하며 page를 읽을 수 있습니다. 기본 `xonly`도 emulate하지만 page를 읽을 수 없습니다. `none`은 완전히 막아 exploit을 어렵게 하지만 오래된 static binary나 glibc를 깨뜨릴 수 있습니다.
vt.color=[VT] `0xYX`, 기본 `0x07`Console 기본 text color를 정합니다. X는 foreground, Y는 background이며 기본은 검정 배경의 밝은 회색입니다.
vt.cur_default=[VT] `0xCCBBAA`, 기본 `2`기본 cursor 모양을 정합니다. AA, BB, CC는 `<Esc>[?A;B;Cc` escape sequence의 같은 매개변수이며 `vga-softcursor.rst`를 참조하십시오. 기본 2는 underline입니다.
vt.default_blu=[VT] `<blue0>,...,<blue15>`0~255 값 16개로 console 기본 blue palette를 바꿉니다.
vt.default_grn=[VT] `<green0>,...,<green15>`0~255 값 16개로 console 기본 green palette를 바꿉니다.
vt.default_red=[VT] `<red0>,...,<red15>`0~255 값 16개로 console 기본 red palette를 바꿉니다.
vt.default_utf8=[VT] `0` 또는 `1`, 기본 `1`모든 tty의 system-wide 기본 UTF-8 mode를 정합니다. 기본적으로 새 terminal은 UTF-8 mode로 열립니다.
vt.global_cursor_default=[VT] `-1`, `0`, `1`; 기본 `-1`새 VT에 cursor를 표시할지 system-wide 기본값을 정합니다. `-1`은 driver가 덮어쓰지 않는 한 생성, `0`은 숨김, `1`은 표시입니다.
vt.italic=[VT] color `0`~`15`, 기본 `2`Italic text의 기본 color를 정하며 기본 2는 green입니다.
vt.underline=[VT] color `0`~`15`, 기본 `3`Underlined text의 기본 color를 정하며 기본 3은 cyan입니다.

Lockup watchdog와 workqueue 정책

8183-8254
매개변수적용·형식한국어 설명
watchdog timers[HW,WDT]Watchdog timer 매개변수는 `Documentation/watchdog/watchdog-parameters.rst`와 `Documentation/watchdog/` 아래 driver별 문서를 참조하십시오.
watchdog_thresh=[KNL] seconds, 기본 `10`Hard-lockup detector의 stall threshold를 정합니다. Soft-lockup threshold는 이 값의 두 배이며 `0`은 두 detector를 모두 끕니다.
workqueue.unbound_cpus=[KNL,SMP] CPU listUnbound workqueue가 사용할 수 있는 CPU를 제한합니다. 기본은 모든 online CPU입니다.
workqueue.watchdog_thresh=`CONFIG_WQ_WATCHDOG`, seconds, 기본 `30`Workqueue stall 경고와 내부 상태 dump의 threshold를 설정합니다. `0`은 감지를 끄며 대응 sysfs file에서 runtime에 바꿀 수 있습니다.
workqueue.panic_on_stall=<uint>`CONFIG_WQ_WATCHDOG`, count, 기본 `0`Workqueue stall이 지정 횟수 감지되면 panic합니다. 기본 0은 stall panic을 끕니다.
workqueue.cpu_intensive_thresh_us=microseconds, 기본 `10000`이 시간보다 오래 실행되는 per-CPU work item을 자동으로 CPU-intensive로 간주해 concurrency management에서 제외하고 다른 per-CPU work가 눈에 띄게 지연되지 않게 합니다. `CONFIG_WQ_CPU_INTENSIVE_REPORT`가 있으면 반복 위반 function을 보고하며, 이런 function은 `WQ_UNBOUND` 후보일 수 있습니다.
workqueue.cpu_intensive_warning_thresh=<uint>`CONFIG_WQ_CPU_INTENSIVE_REPORT`, count, 기본 `4`CPU-intensive threshold를 이 횟수만큼 위반한 뒤 warning을 출력해 우발적 경고를 줄입니다. `0`은 warning을 끕니다.
workqueue.power_efficientbool; 기본은 `CONFIG_WQ_POWER_EFFICIENT_DEFAULT`Cache locality 성능 때문에 보통 선호되는 per-CPU workqueue 중 전력 소비에 크게 기여하는 것을 unbound로 바꿉니다. 작은 성능 비용으로 측정 가능한 전력 절감을 얻습니다.

Workqueue affinity와 LoongArch write-combine

8255-8283
매개변수적용·형식한국어 설명
workqueue.default_affinity_scope=`cpu`, `smt`, `cache`, `numa`, `system`; 기본 `cache`Unbound workqueue의 기본 affinity scope를 고릅니다. `Documentation/core-api/workqueue.rst`의 Affinity Scopes 절을 참조하십시오. Boot 뒤 대응 `/sys/module/workqueue/parameters` file에 쓰면 `default` scope인 모든 workqueue가 갱신됩니다.
workqueue.debug_force_rr_cpudebug boolCPU를 명시하지 않은 work item이 local CPU에 놓인다는 과거의 암묵적 보장에 의존하는 코드를 찾기 위해 round-robin CPU 선택을 강제합니다. 현재도 local CPU를 선호하지만 foreign CPU에 갈 수 있습니다. 활성화하면 memory와 cache locality가 나빠집니다.
writecombine=[LOONGARCH,EARLY] `{ on | off }``ioremap_wc()`의 MAT(Memory Access Type)를 제어합니다. `on`은 write-combine을 켜고 WUC를, `off`는 끄고 SUC를 사용합니다.

x2APIC과 Xen 초기·PV 제어

8284-8332
매개변수적용·형식한국어 설명
x2apic_phys[X86-64,APIC,EARLY]x2APIC 지원 platform에서 기본 cluster mode 대신 physical mode를 사용합니다.
xen_512gb_limit[KNL,X86-64,XEN]Xen에서 paravirtualized로 실행하는 kernel이 최대 512 GiB RAM만 쓰게 제한합니다. Crash-analysis tool과 Xen domain save/restore/migration tool이 더 큰 domain을 처리하도록 준비되지 않은 경우를 위한 제한입니다.
xen_emul_unplug=[HW,X86,XEN,EARLY] comma-separated optionXen emulated device를 unplug합니다. `ide-disks`는 primary-master IDE, `aux-ide-disks`는 그 밖의 IDE, `nics`는 network device, `all`은 NIC와 IDE 전부입니다. `unnecessary`는 host가 unplug protocol에 응답하지 않아도 unplug이 불필요하다고 표시하고, `never`는 version check에 성공해도 unplug하지 않습니다.
xen_legacy_crash[X86,XEN,EARLY]Xen panic notifier에서 dump handler 같은 후기 `panic()` code를 실행하지 않고 즉시 crash합니다.
xen_mc_debug[X86,XEN,EARLY]Xen PV guest의 multicall debug를 켭니다. 성능이 조금 낮아지므로 multicall 오류의 확장 debug data가 필요할 때만 사용합니다.
xen_msr_safe=[X86,XEN,EARLY] boolXen PV guest에서 항상 fault를 내지 않는 safe MSR access function을 사용할지 정합니다. 기본값은 `CONFIG_XEN_PV_MSR_SAFE`가 제어합니다.
xen_nopv[X86]PV optimization을 끄고 HVM guest를 PV driver 없는 일반 HVM guest로 실행합니다. Xen platform에서 같은 효과를 내는 `nopv` option으로 대체되었습니다.
xen_no_vector_callback[KNL,X86,XEN,EARLY]Xen event-channel interrupt의 vector callback을 비활성화합니다.

Xen memory·timer·event와 Xircom PCMCIA

8333-8371
매개변수적용·형식한국어 설명
xen_scrub_pages=[XEN] boolPage를 다른 domain에서 쓰도록 Xen에 돌려주기 전에 scrub할지 제어합니다. Runtime에는 `/sys/devices/system/xen_memory/xen_memory0/scrub_pages`에서 바꿀 수 있고 기본값은 `CONFIG_XEN_SCRUB_PAGES_DEFAULT`가 정합니다.
xen_timer_slop=[X86-64,XEN,EARLY] nanoseconds, 기본 `100000`Virtual Xen timer의 최소 delta를 조정합니다. 값을 낮추면 timer resolution이 좋아지지만 interrupt 처리량이 늘어납니다.
xen.balloon_boot_timeout=[XEN] seconds, 기본 `180`초기 ballooning이 충분한 memory를 해제하지 못할 때 boot를 포기하기까지 기다립니다. 허용 최대보다 적은 memory로 시작한 HVM 또는 PVH guest에만 적용됩니다.
xen.event_eoi_delay=[XEN] jiffies, 기본 `10`Event storm 때 EOI 처리를 지연할 시간을 설정합니다.
xen.event_loop_timeout=[XEN] jiffies, 기본 `2`Event handling loop가 EOI 지연을 시작할 때까지의 시간을 정합니다.
xen.fifo_events=[XEN] bool, 기본 on사용 가능해도 FIFO event handling을 쓰지 않도록 할 수 있습니다. FIFO는 2-level 방식보다 공정하고 가능한 event channel 수도 훨씬 많아 기본적으로 선호됩니다.
xirc2ps_cs=[NET,PCMCIA] IRQ·I/O option listXircom CreditCard Ethernet/Modem PCMCIA driver의 IRQ, IRQ mask, I/O, full-duplex, sound, lockup workaround와 추가 IRQ를 지정합니다.

Xircom 매개변수의 쉼표·대괄호 중첩 형식은 원문 그대로 보존합니다.

			<irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]]

POWER XIVE, xHCI quirk와 xmon

8372-8409
매개변수적용·형식한국어 설명
xive=[PPC] `off`POWER9 이상에서 기본적으로 native XIVE interrupt controller를 사용합니다. `off`는 pseries와 powernv 모두에서 firmware-controlled XIVE mode로 fallback하며 POWER9 이상에서만 유용합니다.
xive.store-eoi=off[PPC]POWER10 이상에서 XIVE interrupt mode의 EOI 처리는 기본적으로 store를 사용합니다. 이 option은 POWER9처럼 load를 사용하게 합니다.
xhci-hcd.quirks[USB,KNL] hex bitmask추가 xHCI host-controller quirk bitmask를 지정합니다. 각 bit의 의미는 `drivers/usb/host/xhci.h`를 참조하십시오.
xmon[PPC,EARLY] `{ early | on | rw | ro | off }`, 기본 `off`xmon debugger를 제어합니다. 값 없이 `xmon`만 주면 `xmon=early`와 같아 `setup_arch()`에서 가능한 한 일찍 호출합니다. `on`은 crash 때만 호출하는 hook을 설치하며 기본 read-only/read-write mode는 `CONFIG_XMON_DEFAULT_RO_MODE`가 정합니다. `rw`는 crash 때 debugger를 열어 SPR register와 memory 등을 쓸 수 있고, `ro`는 같은 hook을 read-only로 사용합니다. `off`는 xmon을 끕니다.