]>
Commit | Line | Data |
---|---|---|
b10d79f7 | 1 | acpi= [HW,ACPI,X86,ARM64] |
03d926f8 | 2 | Advanced Configuration and Power Interface |
6a1f5471 | 3 | Format: { force | on | off | strict | noirq | rsdt | |
e58d154b | 4 | copy_dsdt } |
1da177e4 | 5 | force -- enable ACPI if default was off |
6a1f5471 | 6 | on -- enable ACPI but allow fallback to DT [arm64] |
1da177e4 LT |
7 | off -- disable ACPI if default was on |
8 | noirq -- do not use ACPI for IRQ routing | |
a9913044 | 9 | strict -- Be less tolerant of platforms that are not |
1da177e4 | 10 | strictly ACPI specification compliant. |
237889bf | 11 | rsdt -- prefer RSDT over (default) XSDT |
aa2110cb | 12 | copy_dsdt -- copy DSDT to memory |
6a1f5471 AB |
13 | For ARM64, ONLY "acpi=off", "acpi=on" or "acpi=force" |
14 | are available | |
1da177e4 | 15 | |
151f4e2b | 16 | See also Documentation/power/runtime_pm.rst, pci=noacpi |
1da177e4 | 17 | |
a1fdcc0d LB |
18 | acpi_apic_instance= [ACPI, IOAPIC] |
19 | Format: <int> | |
20 | 2: use 2nd APIC table, if available | |
21 | 1,0: use 1st APIC table | |
4e381a4f | 22 | default: 0 |
a1fdcc0d | 23 | |
c3d6de69 | 24 | acpi_backlight= [HW,ACPI] |
5fd769c2 RD |
25 | { vendor | video | native | none } |
26 | If set to vendor, prefer vendor-specific driver | |
c3d6de69 TR |
27 | (e.g. thinkpad_acpi, sony_acpi, etc.) instead |
28 | of the ACPI video.ko driver. | |
5fd769c2 RD |
29 | If set to video, use the ACPI video.ko driver. |
30 | If set to native, use the device's native backlight mode. | |
31 | If set to none, disable the ACPI backlight interface. | |
c3d6de69 | 32 | |
b2ca5dae CIK |
33 | acpi_force_32bit_fadt_addr |
34 | force FADT to use 32 bit addresses rather than the | |
35 | 64 bit X_* addresses. Some firmware have broken 64 | |
36 | bit addresses for force ACPI ignore these and use | |
37 | the older legacy 32 bit addresses. | |
38 | ||
ef69449b DB |
39 | acpica_no_return_repair [HW, ACPI] |
40 | Disable AML predefined validation mechanism | |
41 | This mechanism can repair the evaluation result to make | |
42 | the return objects more ACPI specification compliant. | |
43 | This option is useful for developers to identify the | |
44 | root cause of an AML interpreter issue when the issue | |
45 | has something to do with the repair mechanism. | |
46 | ||
a0d84a92 BH |
47 | acpi.debug_layer= [HW,ACPI,ACPI_DEBUG] |
48 | acpi.debug_level= [HW,ACPI,ACPI_DEBUG] | |
1da177e4 | 49 | Format: <int> |
a0d84a92 BH |
50 | CONFIG_ACPI_DEBUG must be enabled to produce any ACPI |
51 | debug output. Bits in debug_layer correspond to a | |
52 | _COMPONENT in an ACPI source file, e.g., | |
53 | #define _COMPONENT ACPI_PCI_COMPONENT | |
54 | Bits in debug_level correspond to a level in | |
55 | ACPI_DEBUG_PRINT statements, e.g., | |
56 | ACPI_DEBUG_PRINT((ACPI_DB_INFO, ... | |
e76f4276 | 57 | The debug_level mask defaults to "info". See |
cb1aaebe | 58 | Documentation/firmware-guide/acpi/debug.rst for more information about |
e76f4276 | 59 | debug layers and levels. |
a0d84a92 | 60 | |
e76f4276 BH |
61 | Enable processor driver info messages: |
62 | acpi.debug_layer=0x20000000 | |
63 | Enable PCI/PCI interrupt routing info messages: | |
64 | acpi.debug_layer=0x400000 | |
a0d84a92 BH |
65 | Enable AML "Debug" output, i.e., stores to the Debug |
66 | object while interpreting AML: | |
67 | acpi.debug_layer=0xffffffff acpi.debug_level=0x2 | |
a0d84a92 BH |
68 | Enable all messages related to ACPI hardware: |
69 | acpi.debug_layer=0x2 acpi.debug_level=0xffffffff | |
70 | ||
71 | Some values produce so much output that the system is | |
72 | unusable. The "log_buf_len" parameter may be useful | |
73 | if you need to capture more output. | |
f989106c | 74 | |
ef69449b DB |
75 | acpi_enforce_resources= [ACPI] |
76 | { strict | lax | no } | |
77 | Check for resource conflicts between native drivers | |
78 | and ACPI OperationRegions (SystemIO and SystemMemory | |
79 | only). IO ports and memory declared in ACPI might be | |
80 | used by the ACPI subsystem in arbitrary AML code and | |
81 | can interfere with legacy drivers. | |
82 | strict (default): access to resources claimed by ACPI | |
83 | is denied; legacy drivers trying to access reserved | |
84 | resources will fail to bind to device using them. | |
85 | lax: access to resources claimed by ACPI is allowed; | |
86 | legacy drivers trying to access reserved resources | |
87 | will bind successfully but a warning message is logged. | |
88 | no: ACPI OperationRegions are not marked as reserved, | |
89 | no further checks are performed. | |
90 | ||
4fc0a7e8 LZ |
91 | acpi_force_table_verification [HW,ACPI] |
92 | Enable table checksum verification during early stage. | |
93 | By default, this is disabled due to x86 early mapping | |
94 | size limitation. | |
95 | ||
0cb55ad2 RD |
96 | acpi_irq_balance [HW,ACPI] |
97 | ACPI will balance active IRQs | |
98 | default in APIC mode | |
99 | ||
100 | acpi_irq_nobalance [HW,ACPI] | |
101 | ACPI will not move active IRQs (default) | |
102 | default in PIC mode | |
103 | ||
104 | acpi_irq_isa= [HW,ACPI] If irq_balance, mark listed IRQs used by ISA | |
105 | Format: <irq>,<irq>... | |
106 | ||
107 | acpi_irq_pci= [HW,ACPI] If irq_balance, clear listed IRQs for | |
108 | use by PCI | |
109 | Format: <irq>,<irq>... | |
110 | ||
6dddd7a7 | 111 | acpi_mask_gpe= [HW,ACPI] |
9c4aa1ee LZ |
112 | Due to the existence of _Lxx/_Exx, some GPEs triggered |
113 | by unsupported hardware/firmware features can result in | |
6dddd7a7 TB |
114 | GPE floodings that cannot be automatically disabled by |
115 | the GPE dispatcher. | |
9c4aa1ee LZ |
116 | This facility can be used to prevent such uncontrolled |
117 | GPE floodings. | |
a7583e72 | 118 | Format: <byte> |
9c4aa1ee | 119 | |
08e1d7c0 LZ |
120 | acpi_no_auto_serialize [HW,ACPI] |
121 | Disable auto-serialization of AML methods | |
22b5afce BM |
122 | AML control methods that contain the opcodes to create |
123 | named objects will be marked as "Serialized" by the | |
124 | auto-serialization feature. | |
08e1d7c0 LZ |
125 | This feature is enabled by default. |
126 | This option allows to turn off the feature. | |
22b5afce | 127 | |
ef69449b DB |
128 | acpi_no_memhotplug [ACPI] Disable memory hotplug. Useful for kdump |
129 | kernels. | |
130 | ||
a94e88cd LZ |
131 | acpi_no_static_ssdt [HW,ACPI] |
132 | Disable installation of static SSDTs at early boot time | |
133 | By default, SSDTs contained in the RSDT/XSDT will be | |
134 | installed automatically and they will appear under | |
135 | /sys/firmware/acpi/tables. | |
136 | This option turns off this feature. | |
137 | Note that specifying this option does not affect | |
138 | dynamic table installation which will install SSDT | |
139 | tables to /sys/firmware/acpi/tables/dynamic. | |
0cb55ad2 | 140 | |
3f9e12e0 JD |
141 | acpi_no_watchdog [HW,ACPI,WDT] |
142 | Ignore the ACPI-based watchdog interface (WDAT) and let | |
143 | a native driver control the watchdog device instead. | |
144 | ||
ef69449b DB |
145 | acpi_rsdp= [ACPI,EFI,KEXEC] |
146 | Pass the RSDP address to the kernel, mostly used | |
147 | on machines running EFI runtime service to boot the | |
148 | second kernel for kdump. | |
4dde507f | 149 | |
0cb55ad2 RD |
150 | acpi_os_name= [HW,ACPI] Tell ACPI BIOS the name of the OS |
151 | Format: To spoof as Windows 98: ="Microsoft Windows" | |
152 | ||
18d78b64 RW |
153 | acpi_rev_override [ACPI] Override the _REV object to return 5 (instead |
154 | of 2 which is mandated by ACPI 6) as the supported ACPI | |
155 | specification revision (when using this switch, it may | |
156 | be necessary to carry out a cold reboot _twice_ in a | |
157 | row to make it take effect on the platform firmware). | |
158 | ||
0cb55ad2 | 159 | acpi_osi= [HW,ACPI] Modify list of supported OS interface strings |
5dc17986 LZ |
160 | acpi_osi="string1" # add string1 |
161 | acpi_osi="!string2" # remove string2 | |
741d8128 | 162 | acpi_osi=!* # remove all strings |
5dc17986 LZ |
163 | acpi_osi=! # disable all built-in OS vendor |
164 | strings | |
a707edeb LZ |
165 | acpi_osi=!! # enable all built-in OS vendor |
166 | strings | |
0cb55ad2 RD |
167 | acpi_osi= # disable all strings |
168 | ||
5dc17986 LZ |
169 | 'acpi_osi=!' can be used in combination with single or |
170 | multiple 'acpi_osi="string1"' to support specific OS | |
171 | vendor string(s). Note that such command can only | |
172 | affect the default state of the OS vendor strings, thus | |
173 | it cannot affect the default state of the feature group | |
174 | strings and the current state of the OS vendor strings, | |
175 | specifying it multiple times through kernel command line | |
741d8128 LZ |
176 | is meaningless. This command is useful when one do not |
177 | care about the state of the feature group strings which | |
178 | should be controlled by the OSPM. | |
5dc17986 LZ |
179 | Examples: |
180 | 1. 'acpi_osi=! acpi_osi="Windows 2000"' is equivalent | |
181 | to 'acpi_osi="Windows 2000" acpi_osi=!', they all | |
182 | can make '_OSI("Windows 2000")' TRUE. | |
183 | ||
184 | 'acpi_osi=' cannot be used in combination with other | |
185 | 'acpi_osi=' command lines, the _OSI method will not | |
186 | exist in the ACPI namespace. NOTE that such command can | |
187 | only affect the _OSI support state, thus specifying it | |
188 | multiple times through kernel command line is also | |
189 | meaningless. | |
190 | Examples: | |
191 | 1. 'acpi_osi=' can make 'CondRefOf(_OSI, Local1)' | |
192 | FALSE. | |
193 | ||
741d8128 LZ |
194 | 'acpi_osi=!*' can be used in combination with single or |
195 | multiple 'acpi_osi="string1"' to support specific | |
196 | string(s). Note that such command can affect the | |
197 | current state of both the OS vendor strings and the | |
198 | feature group strings, thus specifying it multiple times | |
199 | through kernel command line is meaningful. But it may | |
200 | still not able to affect the final state of a string if | |
201 | there are quirks related to this string. This command | |
202 | is useful when one want to control the state of the | |
203 | feature group strings to debug BIOS issues related to | |
204 | the OSPM features. | |
205 | Examples: | |
206 | 1. 'acpi_osi="Module Device" acpi_osi=!*' can make | |
207 | '_OSI("Module Device")' FALSE. | |
208 | 2. 'acpi_osi=!* acpi_osi="Module Device"' can make | |
209 | '_OSI("Module Device")' TRUE. | |
210 | 3. 'acpi_osi=! acpi_osi=!* acpi_osi="Windows 2000"' is | |
211 | equivalent to | |
212 | 'acpi_osi=!* acpi_osi=! acpi_osi="Windows 2000"' | |
213 | and | |
214 | 'acpi_osi=!* acpi_osi="Windows 2000" acpi_osi=!', | |
215 | they all will make '_OSI("Windows 2000")' TRUE. | |
216 | ||
6cececfc | 217 | acpi_pm_good [X86] |
0cb55ad2 RD |
218 | Override the pmtimer bug detection: force the kernel |
219 | to assume that this machine's pmtimer latches its value | |
220 | and always returns good values. | |
221 | ||
4af94f39 RD |
222 | acpi_sci= [HW,ACPI] ACPI System Control Interrupt trigger mode |
223 | Format: { level | edge | high | low } | |
224 | ||
4af94f39 RD |
225 | acpi_skip_timer_override [HW,ACPI] |
226 | Recognize and ignore IRQ0/pin2 Interrupt Override. | |
227 | For broken nForce2 BIOS resulting in XT-PIC timer. | |
228 | ||
229 | acpi_sleep= [HW,ACPI] Sleep options | |
230 | Format: { s3_bios, s3_mode, s3_beep, s4_nohwsig, | |
57044031 | 231 | old_ordering, nonvs, sci_force_enable, nobl } |
151f4e2b | 232 | See Documentation/power/video.rst for information on |
4af94f39 RD |
233 | s3_bios and s3_mode. |
234 | s3_beep is for debugging; it makes the PC's speaker beep | |
235 | as soon as the kernel's real-mode entry point is called. | |
236 | s4_nohwsig prevents ACPI hardware signature from being | |
237 | used during resume from hibernation. | |
238 | old_ordering causes the ACPI 1.0 ordering of the _PTS | |
239 | control method, with respect to putting devices into | |
240 | low power states, to be enforced (the ACPI 2.0 ordering | |
241 | of _PTS is used by default). | |
72ad5d77 RW |
242 | nonvs prevents the kernel from saving/restoring the |
243 | ACPI NVS memory during suspend/hibernation and resume. | |
d7f0eea9 ZR |
244 | sci_force_enable causes the kernel to set SCI_EN directly |
245 | on resume from S1/S3 (which is against the ACPI spec, | |
246 | but some broken systems don't work without it). | |
57044031 RW |
247 | nobl causes the internal blacklist of systems known to |
248 | behave incorrectly in some ways with respect to system | |
249 | suspend and resume to be ignored (use wisely). | |
4af94f39 RD |
250 | |
251 | acpi_use_timer_override [HW,ACPI] | |
252 | Use timer override. For some broken Nvidia NF5 boards | |
253 | that require a timer override, but don't have HPET | |
254 | ||
4af94f39 RD |
255 | add_efi_memmap [EFI; X86] Include EFI memory map in |
256 | kernel's map of available physical RAM. | |
257 | ||
0cb55ad2 RD |
258 | agp= [AGP] |
259 | { off | try_unsupported } | |
260 | off: disable AGP support | |
261 | try_unsupported: try to drive unsupported chipsets | |
262 | (may crash computer or cause data corruption) | |
263 | ||
bcfde334 | 264 | ALSA [HW,ALSA] |
1ca2c806 | 265 | See Documentation/sound/alsa-configuration.rst |
bcfde334 | 266 | |
d944d549 RK |
267 | alignment= [KNL,ARM] |
268 | Allow the default userspace alignment fault handler | |
269 | behaviour to be specified. Bit 0 enables warnings, | |
270 | bit 1 enables fixups, and bit 2 sends a segfault. | |
271 | ||
dfb09f9b BP |
272 | align_va_addr= [X86-64] |
273 | Align virtual addresses by clearing slice [14:12] when | |
274 | allocating a VMA at process creation time. This option | |
275 | gives you up to 3% performance improvement on AMD F15h | |
276 | machines (where it is enabled by default) for a | |
277 | CPU-intensive style benchmark, and it can vary highly in | |
278 | a microbenchmark depending on workload and compiler. | |
279 | ||
8360ee2f BP |
280 | 32: only for 32-bit processes |
281 | 64: only for 64-bit processes | |
dfb09f9b BP |
282 | on: enable for both 32- and 64-bit processes |
283 | off: disable for both 32- and 64-bit processes | |
284 | ||
55034cd6 SRRH |
285 | alloc_snapshot [FTRACE] |
286 | Allocate the ftrace snapshot buffer on boot up when the | |
287 | main buffer is allocated. This is handy if debugging | |
288 | and you need to use tracing_snapshot() on boot up, and | |
289 | do not want to use tracing_snapshot_alloc() as it needs | |
290 | to be done where GFP_KERNEL allocations are allowed. | |
291 | ||
89e0b9a3 | 292 | amd_iommu= [HW,X86-64] |
54b4cbd2 JR |
293 | Pass parameters to the AMD IOMMU driver in the system. |
294 | Possible values are: | |
afa9fdc2 FT |
295 | fullflush - enable flushing of IO/TLB entries when |
296 | they are unmapped. Otherwise they are | |
297 | flushed before they will be reused, which | |
298 | is a lot of faster | |
a5235725 JR |
299 | off - do not initialize any AMD IOMMU found in |
300 | the system | |
5abcdba4 JR |
301 | force_isolation - Force device isolation for all |
302 | devices. The IOMMU driver is not | |
303 | allowed anymore to lift isolation | |
304 | requirements as needed. This option | |
305 | does not override iommu=pt | |
afa9fdc2 | 306 | |
c099cf17 SK |
307 | amd_iommu_dump= [HW,X86-64] |
308 | Enable AMD IOMMU driver option to dump the ACPI table | |
309 | for AMD IOMMU. With this option enabled, AMD IOMMU | |
310 | driver will print ACPI tables for AMD IOMMU during | |
311 | IOMMU initialization. | |
312 | ||
3928aa3f SS |
313 | amd_iommu_intr= [HW,X86-64] |
314 | Specifies one of the following AMD IOMMU interrupt | |
315 | remapping modes: | |
316 | legacy - Use legacy interrupt remapping mode. | |
317 | vapic - Use virtual APIC mode, which allows IOMMU | |
318 | to inject interrupts directly into guest. | |
319 | This mode requires kvm-amd.avic=1. | |
320 | (Default when IOMMU HW support is present.) | |
321 | ||
1da177e4 LT |
322 | amijoy.map= [HW,JOY] Amiga joystick support |
323 | Map of devices attached to JOY0DAT and JOY1DAT | |
324 | Format: <a>,<b> | |
1752118d | 325 | See also Documentation/input/joydev/joystick.rst |
1da177e4 LT |
326 | |
327 | analog.map= [HW,JOY] Analog joystick and gamepad support | |
328 | Specifies type or capabilities of an analog joystick | |
329 | connected to one of 16 gameports | |
330 | Format: <type1>,<type2>,..<type16> | |
331 | ||
a9913044 RD |
332 | apc= [HW,SPARC] |
333 | Power management functions (SPARCstation-4/5 + deriv.) | |
1da177e4 LT |
334 | Format: noidle |
335 | Disable APC CPU standby support. SPARCstation-Fox does | |
336 | not play well with APC CPU idle - disable it if you have | |
337 | APC and your system crashes randomly. | |
338 | ||
64e05d11 | 339 | apic= [APIC,X86] Advanced Programmable Interrupt Controller |
806654a9 | 340 | Change the output verbosity while booting |
1da177e4 LT |
341 | Format: { quiet (default) | verbose | debug } |
342 | Change the amount of debugging information output | |
343 | when initialising the APIC and IO-APIC components. | |
64e05d11 DL |
344 | For X86-32, this can also be used to specify an APIC |
345 | driver name. | |
346 | Format: apic=driver_name | |
347 | Examples: apic=bigsmp | |
a9913044 | 348 | |
b7c4948e HK |
349 | apic_extnmi= [APIC,X86] External NMI delivery setting |
350 | Format: { bsp (default) | all | none } | |
351 | bsp: External NMI is delivered only to CPU 0 | |
352 | all: External NMIs are broadcast to all CPUs as a | |
353 | backup of CPU 0 | |
354 | none: External NMI is masked for all CPUs. This is | |
355 | useful so that a dump capture kernel won't be | |
356 | shot down by NMI | |
357 | ||
b0f83b28 | 358 | autoconf= [IPV6] |
19093313 | 359 | See Documentation/networking/ipv6.rst. |
b0f83b28 | 360 | |
9636bc05 CG |
361 | show_lapic= [APIC,X86] Advanced Programmable Interrupt Controller |
362 | Limit apic dumping. The parameter defines the maximal | |
363 | number of local apics being dumped. Also it is possible | |
364 | to set it to "all" by meaning -- no limit here. | |
365 | Format: { 1 (default) | 2 | ... | all }. | |
366 | The parameter valid if only apic=debug or | |
367 | apic=verbose is specified. | |
368 | Example: apic=debug show_lapic=all | |
369 | ||
1da177e4 | 370 | apm= [APM] Advanced Power Management |
71f77055 | 371 | See header of arch/x86/kernel/apm_32.c. |
1da177e4 | 372 | |
1da177e4 LT |
373 | arcrimi= [HW,NET] ARCnet - "RIM I" (entirely mem-mapped) cards |
374 | Format: <io>,<irq>,<nodeID> | |
375 | ||
376 | ataflop= [HW,M68k] | |
377 | ||
378 | atarimouse= [HW,MOUSE] Atari Mouse | |
379 | ||
1da177e4 LT |
380 | atkbd.extra= [HW] Enable extra LEDs and keys on IBM RapidAccess, |
381 | EzKey and similar keyboards | |
382 | ||
383 | atkbd.reset= [HW] Reset keyboard during initialization | |
384 | ||
a9913044 RD |
385 | atkbd.set= [HW] Select keyboard code set |
386 | Format: <int> (2 = AT (default), 3 = PS/2) | |
1da177e4 LT |
387 | |
388 | atkbd.scroll= [HW] Enable scroll wheel on MS Office and similar | |
389 | keyboards | |
390 | ||
391 | atkbd.softraw= [HW] Choose between synthetic and real raw mode | |
392 | Format: <bool> (0 = real, 1 = synthetic (default)) | |
a9913044 RD |
393 | |
394 | atkbd.softrepeat= [HW] | |
395 | Use software keyboard repeat | |
1da177e4 | 396 | |
a106fb0c | 397 | audit= [KNL] Enable the audit sub-system |
11dd2666 GE |
398 | Format: { "0" | "1" | "off" | "on" } |
399 | 0 | off - kernel audit is disabled and can not be | |
400 | enabled until the next reboot | |
d7961148 EP |
401 | unset - kernel audit is initialized but disabled and |
402 | will be fully enabled by the userspace auditd. | |
11dd2666 GE |
403 | 1 | on - kernel audit is initialized and partially |
404 | enabled, storing at most audit_backlog_limit | |
405 | messages in RAM until it is fully enabled by the | |
406 | userspace auditd. | |
a106fb0c | 407 | Default: unset |
f3411cb2 | 408 | |
f910fde7 RGB |
409 | audit_backlog_limit= [KNL] Set the audit queue size limit. |
410 | Format: <int> (must be >=0) | |
411 | Default: 64 | |
412 | ||
1c532e00 AT |
413 | bau= [X86_UV] Enable the BAU on SGI UV. The default |
414 | behavior is to disable the BAU (i.e. bau=0). | |
415 | Format: { "0" | "1" } | |
416 | 0 - Disable the BAU. | |
417 | 1 - Enable the BAU. | |
418 | unset - Disable the BAU. | |
419 | ||
1da177e4 LT |
420 | baycom_epp= [HW,AX25] |
421 | Format: <io>,<mode> | |
a9913044 | 422 | |
1da177e4 LT |
423 | baycom_par= [HW,AX25] BayCom Parallel Port AX.25 Modem |
424 | Format: <io>,<mode> | |
425 | See header of drivers/net/hamradio/baycom_par.c. | |
426 | ||
a9913044 RD |
427 | baycom_ser_fdx= [HW,AX25] |
428 | BayCom Serial Port AX.25 Modem (Full Duplex Mode) | |
1da177e4 LT |
429 | Format: <io>,<irq>,<mode>[,<baud>] |
430 | See header of drivers/net/hamradio/baycom_ser_fdx.c. | |
431 | ||
a9913044 RD |
432 | baycom_ser_hdx= [HW,AX25] |
433 | BayCom Serial Port AX.25 Modem (Half Duplex Mode) | |
1da177e4 LT |
434 | Format: <io>,<irq>,<mode> |
435 | See header of drivers/net/hamradio/baycom_ser_hdx.c. | |
436 | ||
080506ad PG |
437 | blkdevparts= Manual partition parsing of block device(s) for |
438 | embedded devices based on command line input. | |
898bd37a | 439 | See Documentation/block/cmdline-partition.rst |
080506ad | 440 | |
bfe8df3d RD |
441 | boot_delay= Milliseconds to delay each printk during boot. |
442 | Values larger than 10 seconds (10000) are changed to | |
443 | no delay (0). | |
444 | Format: integer | |
445 | ||
7495e092 SRV |
446 | bootconfig [KNL] |
447 | Extended command line options can be added to an initrd | |
448 | and this will cause the kernel to look for it. | |
449 | ||
450 | See Documentation/admin-guide/bootconfig.rst | |
451 | ||
a3e2acc5 HY |
452 | bert_disable [ACPI] |
453 | Disable BERT OS support on buggy BIOSes. | |
454 | ||
1ffb8d03 AH |
455 | bgrt_disable [ACPI][X86] |
456 | Disable BGRT to avoid flickering OEM logo. | |
457 | ||
1da177e4 | 458 | bttv.card= [HW,V4L] bttv (bt848 + bt878 based grabber cards) |
a9913044 RD |
459 | bttv.radio= Most important insmod options are available as |
460 | kernel args too. | |
32e2eae2 | 461 | bttv.pll= See Documentation/admin-guide/media/bttv.rst |
395cf969 | 462 | bttv.tuner= |
1da177e4 | 463 | |
4e89a2d8 WS |
464 | bulk_remove=off [PPC] This parameter disables the use of the pSeries |
465 | firmware feature for flushing multiple hpte entries | |
466 | at a time. | |
467 | ||
1da177e4 LT |
468 | c101= [NET] Moxa C101 synchronous serial card |
469 | ||
cd4f0ef7 | 470 | cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection. |
1da177e4 LT |
471 | Sometimes CPU hardware bugs make them report the cache |
472 | size incorrectly. The kernel will attempt work arounds | |
473 | to fix known problems, but for some CPUs it is not | |
474 | possible to determine what the correct size should be. | |
475 | This option provides an override for these situations. | |
476 | ||
3fc46fc9 MK |
477 | carrier_timeout= |
478 | [NET] Specifies amount of time (in seconds) that | |
479 | the kernel should wait for a network carrier. By default | |
480 | it waits 120 seconds. | |
481 | ||
ffb70f61 DK |
482 | ca_keys= [KEYS] This parameter identifies a specific key(s) on |
483 | the system trusted keyring to be used for certificate | |
484 | trust validation. | |
32c4741c | 485 | format: { id:<keyid> | builtin } |
ffb70f61 | 486 | |
fd1bb4c9 FF |
487 | cca= [MIPS] Override the kernel pages' cache coherency |
488 | algorithm. Accepted values range from 0 to 7 | |
489 | inclusive. See arch/mips/include/asm/pgtable-bits.h | |
490 | for platform specific values (SB1, Loongson3 and | |
491 | others). | |
492 | ||
6dddd7a7 | 493 | ccw_timeout_log [S390] |
8b4a503d | 494 | See Documentation/s390/common_io.rst for details. |
1da177e4 | 495 | |
6dddd7a7 | 496 | cgroup_disable= [KNL] Disable a particular controller |
8bab8dde | 497 | Format: {name of the controller(s) to disable} |
ca0bdbb5 QH |
498 | The effects of cgroup_disable=foo are: |
499 | - foo isn't auto-mounted if you mount all cgroups in | |
500 | a single hierarchy | |
501 | - foo isn't visible as an individually mountable | |
502 | subsystem | |
503 | {Currently only "memory" controller deal with this and | |
504 | cut the overhead, others just disable the usage. So | |
505 | only cgroup_disable=memory is actually worthy} | |
8bab8dde | 506 | |
3fc9c12d TH |
507 | cgroup_no_v1= [KNL] Disable cgroup controllers and named hierarchies in v1 |
508 | Format: { { controller | "all" | "named" } | |
509 | [,{ controller | "all" | "named" }...] } | |
1619b6d4 JW |
510 | Like cgroup_disable, but only applies to cgroup v1; |
511 | the blacklisted controllers remain available in cgroup2. | |
3fc9c12d TH |
512 | "all" blacklists all controllers and "named" disables |
513 | named mounts. Specifying both "all" and "named" disables | |
514 | all v1 hierarchies. | |
1619b6d4 | 515 | |
f7e1cb6e JW |
516 | cgroup.memory= [KNL] Pass options to the cgroup memory controller. |
517 | Format: <string> | |
518 | nosocket -- Disable socket memory accounting. | |
04823c83 | 519 | nokmem -- Disable kernel memory accounting. |
f7e1cb6e | 520 | |
1da177e4 LT |
521 | checkreqprot [SELINUX] Set initial checkreqprot flag value. |
522 | Format: { "0" | "1" } | |
523 | See security/selinux/Kconfig help text. | |
a9913044 RD |
524 | 0 -- check protection applied by kernel (includes |
525 | any implied execute protection). | |
1da177e4 LT |
526 | 1 -- check protection requested by application. |
527 | Default value is set via a kernel config option. | |
a9913044 | 528 | Value can be changed at runtime via |
d41415eb | 529 | /sys/fs/selinux/checkreqprot. |
e9c38f9f | 530 | Setting checkreqprot to 1 is deprecated. |
a9913044 | 531 | |
661ca0da | 532 | cio_ignore= [S390] |
8b4a503d | 533 | See Documentation/s390/common_io.rst for details. |
1e435256 OJ |
534 | clk_ignore_unused |
535 | [CLK] | |
e156ee56 MT |
536 | Prevents the clock framework from automatically gating |
537 | clocks that have not been explicitly enabled by a Linux | |
538 | device driver but are enabled in hardware at reset or | |
539 | by the bootloader/firmware. Note that this does not | |
540 | force such clocks to be always-on nor does it reserve | |
541 | those clocks in any way. This parameter is useful for | |
542 | debug and development, but should not be needed on a | |
543 | platform with proper driver support. For more | |
18bcaa4e | 544 | information, see Documentation/driver-api/clk.rst. |
661ca0da | 545 | |
cd4f0ef7 | 546 | clock= [BUGS=X86-32, HW] gettimeofday clocksource override. |
734efb46 | 547 | [Deprecated] |
3f6dee9b | 548 | Forces specified clocksource (if available) to be used |
734efb46 | 549 | when calculating gettimeofday(). If specified |
3f6dee9b | 550 | clocksource is not available, it defaults to PIT. |
1da177e4 LT |
551 | Format: { pit | tsc | cyclone | pmtmr } |
552 | ||
592913ec | 553 | clocksource= Override the default clocksource |
3d6ac984 RD |
554 | Format: <string> |
555 | Override the default clocksource and use the clocksource | |
556 | with the name specified. | |
557 | Some clocksource names to choose from, depending on | |
558 | the platform: | |
559 | [all] jiffies (this is the base, fallback clocksource) | |
560 | [ACPI] acpi_pm | |
561 | [ARM] imx_timer1,OSTS,netx_timer,mpu_timer2, | |
562 | pxa_timer,timer3,32k_counter,timer0_1 | |
9863c90f | 563 | [X86-32] pit,hpet,tsc; |
3d6ac984 RD |
564 | scx200_hrt on Geode; cyclone on IBM x440 |
565 | [MIPS] MIPS | |
566 | [PARISC] cr16 | |
567 | [S390] tod | |
568 | [SH] SuperH | |
569 | [SPARC64] tick | |
570 | [X86-64] hpet,tsc | |
571 | ||
46fd5c6b WD |
572 | clocksource.arm_arch_timer.evtstrm= |
573 | [ARM,ARM64] | |
574 | Format: <bool> | |
575 | Enable/disable the eventstream feature of the ARM | |
576 | architected timer so that code using WFE-based polling | |
577 | loops can be debugged more effectively on production | |
578 | systems. | |
579 | ||
ac72e788 AK |
580 | clearcpuid=BITNUM [X86] |
581 | Disable CPUID feature X for the kernel. See | |
cd4d09ec | 582 | arch/x86/include/asm/cpufeatures.h for the valid bit |
07983f0e | 583 | numbers. Note the Linux specific bits are not necessarily |
ac72e788 AK |
584 | stable over kernel options, but the vendor specific |
585 | ones should be. | |
586 | Also note that user programs calling CPUID directly | |
587 | or using the feature without checking anything | |
588 | will still see it. This just prevents it from | |
589 | being used by the kernel or shown in /proc/cpuinfo. | |
590 | Also note the kernel might malfunction if you disable | |
591 | some critical bits. | |
592 | ||
5ea3b1b2 | 593 | cma=nn[MG]@[start[MG][-end[MG]]] |
c372e741 | 594 | [KNL,CMA] |
5ea3b1b2 AM |
595 | Sets the size of kernel global memory area for |
596 | contiguous memory allocations and optionally the | |
597 | placement constraint by the physical address range of | |
f0d6d1f6 JD |
598 | memory allocations. A value of 0 disables CMA |
599 | altogether. For more information, see | |
c64be2bb MS |
600 | include/linux/dma-contiguous.h |
601 | ||
14f966e7 RJ |
602 | cmo_free_hint= [PPC] Format: { yes | no } |
603 | Specify whether pages are marked as being inactive | |
604 | when they are freed. This is used in CMO environments | |
605 | to determine OS memory pressure for page stealing by | |
606 | a hypervisor. | |
607 | Default: yes | |
608 | ||
c7909509 MS |
609 | coherent_pool=nn[KMG] [ARM,KNL] |
610 | Sets the size of memory pool for coherent, atomic dma | |
e9da6e99 | 611 | allocations, by default set to 256K. |
c7909509 | 612 | |
1da177e4 | 613 | com20020= [HW,NET] ARCnet - COM20020 chipset |
a9913044 RD |
614 | Format: |
615 | <io>[,<irq>[,<nodeID>[,<backplane>[,<ckp>[,<timeout>]]]]] | |
1da177e4 LT |
616 | |
617 | com90io= [HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers) | |
618 | Format: <io>[,<irq>] | |
619 | ||
a9913044 RD |
620 | com90xx= [HW,NET] |
621 | ARCnet - COM90xx chipset (memory-mapped buffers) | |
1da177e4 LT |
622 | Format: <io>[,<irq>[,<memstart>]] |
623 | ||
624 | condev= [HW,S390] console device | |
625 | conmode= | |
a9913044 | 626 | |
1da177e4 LT |
627 | console= [KNL] Output console device and options. |
628 | ||
629 | tty<n> Use the virtual console device <n>. | |
630 | ||
631 | ttyS<n>[,options] | |
f1a1c2dc | 632 | ttyUSB0[,options] |
1da177e4 | 633 | Use the specified serial port. The options are of |
f1a1c2dc RD |
634 | the form "bbbbpnf", where "bbbb" is the baud rate, |
635 | "p" is parity ("n", "o", or "e"), "n" is number of | |
636 | bits, and "f" is flow control ("r" for RTS or | |
637 | omit it). Default is "9600n8". | |
638 | ||
e52347bd | 639 | See Documentation/admin-guide/serial-console.rst for more |
f1a1c2dc | 640 | information. See |
d9d6ef25 | 641 | Documentation/networking/netconsole.rst for an |
f1a1c2dc | 642 | alternative. |
1da177e4 | 643 | |
18a8bd94 YL |
644 | uart[8250],io,<addr>[,options] |
645 | uart[8250],mmio,<addr>[,options] | |
bd94c407 | 646 | uart[8250],mmio16,<addr>[,options] |
ca782f16 PH |
647 | uart[8250],mmio32,<addr>[,options] |
648 | uart[8250],0x<addr>[,options] | |
1da177e4 LT |
649 | Start an early, polled-mode console on the 8250/16550 |
650 | UART at the specified I/O port or MMIO address, | |
ca782f16 PH |
651 | switching to the matching ttyS device later. |
652 | MMIO inter-register address stride is either 8-bit | |
bd94c407 MY |
653 | (mmio), 16-bit (mmio16), or 32-bit (mmio32). |
654 | If none of [io|mmio|mmio16|mmio32], <addr> is assumed | |
655 | to be equivalent to 'mmio'. 'options' are specified in | |
656 | the same format described for ttyS above; if unspecified, | |
ca782f16 PH |
657 | the h/w is not re-initialized. |
658 | ||
a2fd6419 KRW |
659 | hvc<n> Use the hypervisor console device <n>. This is for |
660 | both Xen and PowerPC hypervisors. | |
1da177e4 | 661 | |
6dddd7a7 TB |
662 | If the device connected to the port is not a TTY but a braille |
663 | device, prepend "brl," before the device type, for instance | |
f7511d5f ST |
664 | console=brl,ttyS0 |
665 | For now, only VisioBraille is supported. | |
666 | ||
cca10d58 SS |
667 | console_msg_format= |
668 | [KNL] Change console messages format | |
669 | default | |
670 | By default we print messages on consoles in | |
671 | "[time stamp] text\n" format (time stamp may not be | |
672 | printed, depending on CONFIG_PRINTK_TIME or | |
673 | `printk_time' param). | |
674 | syslog | |
675 | Switch to syslog format: "<%u>[time stamp] text\n" | |
676 | IOW, each message will have a facility and loglevel | |
677 | prefix. The format is similar to one used by syslog() | |
678 | syscall, or to executing "dmesg -S --raw" or to reading | |
679 | from /proc/kmsg. | |
680 | ||
f324edc8 | 681 | consoleblank= [KNL] The console blank (screen saver) timeout in |
ac0a314c | 682 | seconds. A value of 0 disables the blank timer. |
6dddd7a7 | 683 | Defaults to 0. |
f324edc8 | 684 | |
4cb0e11b HK |
685 | coredump_filter= |
686 | [KNL] Change the default value for | |
687 | /proc/<pid>/coredump_filter. | |
cd4ca341 | 688 | See also Documentation/filesystems/proc.rst. |
4cb0e11b | 689 | |
62a31ce1 LY |
690 | coresight_cpu_debug.enable |
691 | [ARM,ARM64] | |
692 | Format: <bool> | |
693 | Enable/disable the CPU sampling based debugging. | |
694 | 0: default value, disable debugging | |
695 | 1: enable debugging at boot time | |
696 | ||
62027aea LB |
697 | cpuidle.off=1 [CPU_IDLE] |
698 | disable the cpuidle sub-system | |
699 | ||
61cb5758 RW |
700 | cpuidle.governor= |
701 | [CPU_IDLE] Name of the cpuidle governor to use. | |
702 | ||
d82f2692 LB |
703 | cpufreq.off=1 [CPU_FREQ] |
704 | disable the cpufreq sub-system | |
705 | ||
8412b456 QP |
706 | cpufreq.default_governor= |
707 | [CPU_FREQ] Name of the default cpufreq governor or | |
708 | policy to use. This governor must be registered in the | |
709 | kernel before the cpufreq driver probes. | |
710 | ||
d68921f9 LB |
711 | cpu_init_udelay=N |
712 | [X86] Delay for N microsec between assert and de-assert | |
713 | of APIC INIT to start processors. This delay occurs | |
714 | on every CPU online, such as boot, and resume from suspend. | |
715 | Default: 10000 | |
716 | ||
1da177e4 | 717 | cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver |
a9913044 RD |
718 | Format: |
719 | <first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>] | |
1da177e4 | 720 | |
6f21e646 AD |
721 | crashkernel=size[KMG][@offset[KMG]] |
722 | [KNL] Using kexec, Linux can switch to a 'crash kernel' | |
723 | upon panic. This parameter reserves the physical | |
724 | memory region [offset, offset + size] for that kernel | |
725 | image. If '@offset' is omitted, then a suitable offset | |
b9ac3849 | 726 | is selected automatically. |
be3a5b0e | 727 | [KNL, X86-64] Select a region under 4G first, and |
b9ac3849 DY |
728 | fall back to reserve region above 4G when '@offset' |
729 | hasn't been specified. | |
330d4810 | 730 | See Documentation/admin-guide/kdump/kdump.rst for further details. |
dc009d92 | 731 | |
fb391599 BW |
732 | crashkernel=range1:size1[,range2:size2,...][@offset] |
733 | [KNL] Same as above, but depends on the memory | |
734 | in the running system. The syntax of range is | |
735 | start-[end] where start and end are both | |
736 | a memory unit (amount[KMG]). See also | |
330d4810 | 737 | Documentation/admin-guide/kdump/kdump.rst for an example. |
fb391599 | 738 | |
adbc742b | 739 | crashkernel=size[KMG],high |
be3a5b0e | 740 | [KNL, X86-64] range could be above 4G. Allow kernel |
55a20ee7 YL |
741 | to allocate physical memory region from top, so could |
742 | be above 4G if system have more than 4G ram installed. | |
743 | Otherwise memory region will be allocated below 4G, if | |
744 | available. | |
745 | It will be ignored if crashkernel=X is specified. | |
adbc742b | 746 | crashkernel=size[KMG],low |
be3a5b0e | 747 | [KNL, X86-64] range under 4G. When crashkernel=X,high |
adbc742b | 748 | is passed, kernel could allocate physical memory region |
c729de8f YL |
749 | above 4G, that cause second kernel crash on system |
750 | that require some amount of low memory, e.g. swiotlb | |
c6045031 BH |
751 | requires at least 64M+32K low memory, also enough extra |
752 | low memory is needed to make sure DMA buffers for 32-bit | |
753 | devices won't run out. Kernel would try to allocate at | |
754 | at least 256M below 4G automatically. | |
c729de8f YL |
755 | This one let user to specify own low range under 4G |
756 | for second kernel instead. | |
757 | 0: to disable low allocation. | |
adbc742b | 758 | It will be ignored when crashkernel=X,high is not used |
55a20ee7 | 759 | or memory reserved is below 4G. |
c729de8f | 760 | |
9e5c9fe4 | 761 | cryptomgr.notests |
6dddd7a7 | 762 | [KNL] Disable crypto self-tests |
9e5c9fe4 | 763 | |
1da177e4 LT |
764 | cs89x0_dma= [HW,NET] |
765 | Format: <dma> | |
766 | ||
767 | cs89x0_media= [HW,NET] | |
768 | Format: { rj45 | aui | bnc } | |
a9913044 | 769 | |
a9913044 | 770 | dasd= [HW,NET] |
1da177e4 LT |
771 | See header of drivers/s390/block/dasd_devmap.c. |
772 | ||
773 | db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port | |
774 | (one device per port) | |
775 | Format: <port#>,<type> | |
1752118d | 776 | See also Documentation/input/devices/joystick-parport.rst |
1da177e4 | 777 | |
6dddd7a7 | 778 | ddebug_query= [KNL,DYNAMIC_DEBUG] Enable debug messages at early boot |
787e3075 SM |
779 | time. See |
780 | Documentation/admin-guide/dynamic-debug-howto.rst for | |
29e36c9f | 781 | details. Deprecated, see dyndbg. |
a648ec05 | 782 | |
1da177e4 LT |
783 | debug [KNL] Enable kernel debugging (events log level). |
784 | ||
3672476e TH |
785 | debug_boot_weak_hash |
786 | [KNL] Enable printing [hashed] pointers early in the | |
787 | boot sequence. If enabled, we use a weak hash instead | |
788 | of siphash to hash pointers. Use this option if you are | |
789 | seeing instances of '(___ptrval___)') and need to see a | |
790 | value (hashed pointer) instead. Cryptographically | |
791 | insecure, please do not use on production kernels. | |
792 | ||
cae2ed9a IM |
793 | debug_locks_verbose= |
794 | [KNL] verbose self-tests | |
795 | Format=<0|1> | |
796 | Print debugging info while doing the locking API | |
797 | self-tests. | |
798 | We default to 0 (no extra messages), setting it to | |
799 | 1 will print _a lot_ more information - normally | |
800 | only useful to kernel developers. | |
801 | ||
3ac7fe5a TG |
802 | debug_objects [KNL] Enable object debugging |
803 | ||
3e8ebb5c KM |
804 | no_debug_objects |
805 | [KNL] Disable object debugging | |
806 | ||
c0a32fc5 SG |
807 | debug_guardpage_minorder= |
808 | [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this | |
809 | parameter allows control of the order of pages that will | |
810 | be intentionally kept free (and hence protected) by the | |
811 | buddy allocator. Bigger value increase the probability | |
812 | of catching random memory corruption, but reduce the | |
813 | amount of memory for normal system use. The maximum | |
814 | possible value is MAX_ORDER/2. Setting this parameter | |
815 | to 1 or 2 should be enough to identify most random | |
816 | memory corruption problems caused by bugs in kernel or | |
817 | driver code when a CPU writes to (or reads from) a | |
818 | random memory location. Note that there exists a class | |
819 | of memory corruptions problems caused by buggy H/W or | |
820 | F/W or by drivers badly programing DMA (basically when | |
821 | memory is written at bus level and the CPU MMU is | |
822 | bypassed) which are not detectable by | |
823 | CONFIG_DEBUG_PAGEALLOC, hence this option will not help | |
824 | tracking down these problems. | |
825 | ||
031bc574 | 826 | debug_pagealloc= |
3972f6bb VB |
827 | [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this parameter |
828 | enables the feature at boot time. By default, it is | |
829 | disabled and the system will work mostly the same as a | |
830 | kernel built without CONFIG_DEBUG_PAGEALLOC. | |
8974558f VB |
831 | Note: to get most of debug_pagealloc error reports, it's |
832 | useful to also enable the page_owner functionality. | |
031bc574 JK |
833 | on: enable the feature |
834 | ||
a24c6f7b PE |
835 | debugfs= [KNL] This parameter enables what is exposed to userspace |
836 | and debugfs internal clients. | |
837 | Format: { on, no-mount, off } | |
838 | on: All functions are enabled. | |
839 | no-mount: | |
840 | Filesystem is not registered but kernel clients can | |
841 | access APIs and a crashkernel can be used to read | |
842 | its content. There is nothing to mount. | |
843 | off: Filesystem is not registered and clients | |
844 | get a -EPERM as result when trying to register files | |
845 | or directories within debugfs. | |
846 | This is equivalent of the runtime functionality if | |
847 | debugfs was not enabled in the kernel at all. | |
848 | Default value is set in build-time with a kernel configuration. | |
849 | ||
d3af01f1 TG |
850 | debugpat [X86] Enable PAT debugging |
851 | ||
2d27a966 | 852 | decnet.addr= [HW,NET] |
1da177e4 | 853 | Format: <area>[,<node>] |
9a69fb9c | 854 | See also Documentation/networking/decnet.rst. |
1da177e4 | 855 | |
0cb55ad2 | 856 | default_hugepagesz= |
282f4214 MK |
857 | [HW] The size of the default HugeTLB page. This is |
858 | the size represented by the legacy /proc/ hugepages | |
859 | APIs. In addition, this is the default hugetlb size | |
860 | used for shmget(), mmap() and mounting hugetlbfs | |
861 | filesystems. If not specified, defaults to the | |
862 | architecture's default huge page size. Huge page | |
863 | sizes are architecture dependent. See also | |
864 | Documentation/admin-guide/mm/hugetlbpage.rst. | |
865 | Format: size[KMG] | |
55ff9780 | 866 | |
25b4e70d RH |
867 | deferred_probe_timeout= |
868 | [KNL] Debugging option to set a timeout in seconds for | |
869 | deferred probe to give up waiting on dependencies to | |
870 | probe. Only specific dependencies (subsystems or | |
871 | drivers) that have opted in will be ignored. A timeout of 0 | |
872 | will timeout at the end of initcalls. This option will also | |
873 | dump out devices still on the deferred probe list after | |
874 | retrying. | |
875 | ||
c65e6815 MZ |
876 | dfltcc= [HW,S390] |
877 | Format: { on | off | def_only | inf_only | always } | |
878 | on: s390 zlib hardware support for compression on | |
879 | level 1 and decompression (default) | |
880 | off: No s390 zlib hardware support | |
881 | def_only: s390 zlib hardware support for deflate | |
882 | only (compression on level 1) | |
883 | inf_only: s390 zlib hardware support for inflate | |
884 | only (decompression) | |
885 | always: Same as 'on' but ignores the selected compression | |
886 | level always using hardware support (used for debugging) | |
887 | ||
1da177e4 LT |
888 | dhash_entries= [KNL] |
889 | Set number of hash buckets for dentry cache. | |
a9913044 | 890 | |
faf78829 OH |
891 | disable_1tb_segments [PPC] |
892 | Disables the use of 1TB hash page table segments. This | |
893 | causes the kernel to fall back to 256MB segments which | |
894 | can be useful when debugging issues that require an SLB | |
895 | miss to occur. | |
896 | ||
82a1b8ed NP |
897 | stress_slb [PPC] |
898 | Limits the number of kernel SLB entries, and flushes | |
899 | them frequently to increase the rate of SLB faults | |
900 | on kernel addresses. | |
901 | ||
b0f83b28 | 902 | disable= [IPV6] |
19093313 | 903 | See Documentation/networking/ipv6.rst. |
b0f83b28 | 904 | |
b5cb15d9 CR |
905 | hardened_usercopy= |
906 | [KNL] Under CONFIG_HARDENED_USERCOPY, whether | |
907 | hardening is enabled for this boot. Hardened | |
908 | usercopy checking is used to protect the kernel | |
909 | from reading or writing beyond known memory | |
910 | allocation boundaries as a proactive defense | |
911 | against bounds-checking flaws in the kernel's | |
912 | copy_to_user()/copy_from_user() interface. | |
913 | on Perform hardened usercopy checks (default). | |
914 | off Disable hardened usercopy checks. | |
915 | ||
b275bfb2 AK |
916 | disable_radix [PPC] |
917 | Disable RADIX MMU mode on POWER9 | |
918 | ||
bf6b7661 AK |
919 | radix_hcall_invalidate=on [PPC/PSERIES] |
920 | Disable RADIX GTSE feature and use hcall for TLB | |
921 | invalidate. | |
922 | ||
2275d7b5 NP |
923 | disable_tlbie [PPC] |
924 | Disable TLBIE instruction. Currently does not work | |
925 | with KVM, with HASH MMU, or with coherent accelerators. | |
926 | ||
151e0c7d HD |
927 | disable_cpu_apicid= [X86,APIC,SMP] |
928 | Format: <int> | |
929 | The number of initial APIC ID for the | |
930 | corresponding CPU to be disabled at boot, | |
931 | mostly used for the kdump 2nd kernel to | |
932 | disable BSP to wake up multiple CPUs without | |
933 | causing system reset or hang due to sending | |
934 | INIT from AP to BSP. | |
935 | ||
2a5bf23d PZ |
936 | perf_v4_pmi= [X86,INTEL] |
937 | Format: <bool> | |
af3bdb99 AK |
938 | Disable Intel PMU counter freezing feature. |
939 | The feature only exists starting from | |
940 | Arch Perfmon v4 (Skylake and newer). | |
941 | ||
6dddd7a7 | 942 | disable_ddw [PPC/PSERIES] |
255bf90f | 943 | Disable Dynamic DMA Window support. Use this |
4e8b0cf4 NA |
944 | to workaround buggy firmware. |
945 | ||
b0f83b28 | 946 | disable_ipv6= [IPV6] |
19093313 | 947 | See Documentation/networking/ipv6.rst. |
b0f83b28 | 948 | |
95ffa243 | 949 | disable_mtrr_cleanup [X86] |
95ffa243 YL |
950 | The kernel tries to adjust MTRR layout from continuous |
951 | to discrete, to make X server driver able to add WB | |
0cb55ad2 | 952 | entry later. This parameter disables that. |
95ffa243 | 953 | |
093af8d7 | 954 | disable_mtrr_trim [X86, Intel and AMD only] |
99fc8d42 JB |
955 | By default the kernel will trim any uncacheable |
956 | memory out of your available memory pool based on | |
957 | MTRR settings. This parameter disables that behavior, | |
958 | possibly causing your machine to run very slowly. | |
959 | ||
6cececfc | 960 | disable_timer_pin_1 [X86] |
0cb55ad2 RD |
961 | Disable PIN 1 of APIC timer |
962 | Can be useful to work around chipset bugs. | |
963 | ||
ce14c583 PB |
964 | dis_ucode_ldr [X86] Disable the microcode loader. |
965 | ||
0cb55ad2 RD |
966 | dma_debug=off If the kernel is compiled with DMA_API_DEBUG support, |
967 | this option disables the debugging code at boot. | |
968 | ||
969 | dma_debug_entries=<number> | |
970 | This option allows to tune the number of preallocated | |
971 | entries for DMA-API debugging code. One entry is | |
972 | required per DMA-API allocation. Use this if the | |
973 | DMA-API debugging code disables itself because the | |
974 | architectural default is too low. | |
975 | ||
1745de5e JR |
976 | dma_debug_driver=<driver_name> |
977 | With this option the DMA-API debugging driver | |
978 | filter feature can be enabled at boot time. Just | |
979 | pass the driver to filter for as the parameter. | |
980 | The filter can be disabled or changed to another | |
981 | driver later using sysfs. | |
982 | ||
1ea61b68 FT |
983 | driver_async_probe= [KNL] |
984 | List of driver names to be probed asynchronously. | |
985 | Format: <driver_name1>,<driver_name2>... | |
986 | ||
53fd40a9 | 987 | drm.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>] |
96206e29 BP |
988 | Broken monitors, graphic adapters, KVMs and EDIDless |
989 | panels may send no or incorrect EDID data sets. | |
990 | This parameter allows to specify an EDID data sets | |
991 | in the /lib/firmware directory that are used instead. | |
da0df92b CE |
992 | Generic built-in EDID data sets are used, if one of |
993 | edid/1024x768.bin, edid/1280x1024.bin, | |
994 | edid/1680x1050.bin, or edid/1920x1080.bin is given | |
995 | and no file with the same name exists. Details and | |
996 | instructions how to build your own EDID data are | |
cd4ca341 | 997 | available in Documentation/admin-guide/edid.rst. An EDID |
da0df92b CE |
998 | data set will only be used for a particular connector, |
999 | if its name and a colon are prepended to the EDID | |
96206e29 BP |
1000 | name. Each connector may use a unique EDID data |
1001 | set by separating the files with a comma. An EDID | |
1002 | data set with no connector name will be used for | |
1003 | any connectors not explicitly specified. | |
da0df92b | 1004 | |
1da177e4 LT |
1005 | dscc4.setup= [NET] |
1006 | ||
a2b05b7a NP |
1007 | dt_cpu_ftrs= [PPC] |
1008 | Format: {"off" | "known"} | |
1009 | Control how the dt_cpu_ftrs device-tree binding is | |
1010 | used for CPU feature discovery and setup (if it | |
1011 | exists). | |
1012 | off: Do not use it, fall back to legacy cpu table. | |
1013 | known: Do not pass through unknown features to guests | |
1014 | or userspace, only those that the kernel is aware of. | |
1015 | ||
58c5475a LW |
1016 | dump_apple_properties [X86] |
1017 | Dump name and content of EFI device properties on | |
1018 | x86 Macs. Useful for driver authors to determine | |
1019 | what data is available or for reverse-engineering. | |
1020 | ||
29e36c9f | 1021 | dyndbg[="val"] [KNL,DYNAMIC_DEBUG] |
307e3ee9 | 1022 | <module>.dyndbg[="val"] |
29e36c9f | 1023 | Enable debug messages at boot time. See |
787e3075 SM |
1024 | Documentation/admin-guide/dynamic-debug-howto.rst |
1025 | for details. | |
29e36c9f | 1026 | |
06976945 DH |
1027 | nopku [X86] Disable Memory Protection Keys CPU feature found |
1028 | in some Intel CPUs. | |
1029 | ||
307e3ee9 | 1030 | <module>.async_probe [KNL] |
f2411da7 LR |
1031 | Enable asynchronous probe on this module. |
1032 | ||
56aeeba8 MS |
1033 | early_ioremap_debug [KNL] |
1034 | Enable debug messages in early_ioremap support. This | |
1035 | is useful for tracking down temporary early mappings | |
1036 | which are not unmapped. | |
1037 | ||
0cb55ad2 | 1038 | earlycon= [KNL] Output early console device and options. |
0d3c673e | 1039 | |
e18409c0 CH |
1040 | When used with no options, the early console is |
1041 | determined by stdout-path property in device tree's | |
1042 | chosen node or the ACPI SPCR table if supported by | |
1043 | the platform. | |
5664f764 | 1044 | |
c41251b1 ST |
1045 | cdns,<addr>[,options] |
1046 | Start an early, polled-mode console on a Cadence | |
1047 | (xuartps) serial port at the specified address. Only | |
1048 | supported option is baud rate. If baud rate is not | |
1049 | specified, the serial port must already be setup and | |
1050 | configured. | |
6fa62fc4 | 1051 | |
0cb55ad2 RD |
1052 | uart[8250],io,<addr>[,options] |
1053 | uart[8250],mmio,<addr>[,options] | |
1917ac76 | 1054 | uart[8250],mmio32,<addr>[,options] |
6e63be3f | 1055 | uart[8250],mmio32be,<addr>[,options] |
ca782f16 | 1056 | uart[8250],0x<addr>[,options] |
0cb55ad2 RD |
1057 | Start an early, polled-mode console on the 8250/16550 |
1058 | UART at the specified I/O port or MMIO address. | |
16290246 | 1059 | MMIO inter-register address stride is either 8-bit |
6e63be3f NC |
1060 | (mmio) or 32-bit (mmio32 or mmio32be). |
1061 | If none of [io|mmio|mmio32|mmio32be], <addr> is assumed | |
1062 | to be equivalent to 'mmio'. 'options' are specified | |
1063 | in the same format described for "console=ttyS<n>"; if | |
ca782f16 | 1064 | unspecified, the h/w is not initialized. |
0cb55ad2 | 1065 | |
0d3c673e | 1066 | pl011,<addr> |
3b78fae7 | 1067 | pl011,mmio32,<addr> |
0d3c673e RH |
1068 | Start an early, polled-mode console on a pl011 serial |
1069 | port at the specified address. The pl011 serial port | |
1070 | must already be setup and configured. Options are not | |
3b78fae7 TT |
1071 | yet supported. If 'mmio32' is specified, then only |
1072 | the driver will use only 32-bit accessors to read/write | |
1073 | the device registers. | |
0d3c673e | 1074 | |
736d5538 AF |
1075 | meson,<addr> |
1076 | Start an early, polled-mode console on a meson serial | |
1077 | port at the specified address. The serial port must | |
1078 | already be setup and configured. Options are not yet | |
1079 | supported. | |
1080 | ||
0efe7296 SB |
1081 | msm_serial,<addr> |
1082 | Start an early, polled-mode console on an msm serial | |
1083 | port at the specified address. The serial port | |
1084 | must already be setup and configured. Options are not | |
1085 | yet supported. | |
1086 | ||
1087 | msm_serial_dm,<addr> | |
1088 | Start an early, polled-mode console on an msm serial | |
1089 | dm port at the specified address. The serial port | |
1090 | must already be setup and configured. Options are not | |
1091 | yet supported. | |
1092 | ||
e36361d7 AF |
1093 | owl,<addr> |
1094 | Start an early, polled-mode console on a serial port | |
1095 | of an Actions Semi SoC, such as S500 or S900, at the | |
1096 | specified address. The serial port must already be | |
1097 | setup and configured. Options are not yet supported. | |
1098 | ||
c10b1332 MS |
1099 | rda,<addr> |
1100 | Start an early, polled-mode console on a serial port | |
1101 | of an RDA Micro SoC, such as RDA8810PL, at the | |
1102 | specified address. The serial port must already be | |
e36361d7 AF |
1103 | setup and configured. Options are not yet supported. |
1104 | ||
82f12ab3 PD |
1105 | sbi |
1106 | Use RISC-V SBI (Supervisor Binary Interface) for early | |
1107 | console. | |
1108 | ||
d50d7269 RH |
1109 | smh Use ARM semihosting calls for early console. |
1110 | ||
b94ba032 TF |
1111 | s3c2410,<addr> |
1112 | s3c2412,<addr> | |
1113 | s3c2440,<addr> | |
1114 | s3c6400,<addr> | |
1115 | s5pv210,<addr> | |
1116 | exynos4210,<addr> | |
1117 | Use early console provided by serial driver available | |
1118 | on Samsung SoCs, requires selecting proper type and | |
1119 | a correct base address of the selected UART port. The | |
1120 | serial port must already be setup and configured. | |
1121 | Options are not yet supported. | |
1122 | ||
ec84aa0a MB |
1123 | lantiq,<addr> |
1124 | Start an early, polled-mode console on a lantiq serial | |
1125 | (lqasc) port at the specified address. The serial port | |
1126 | must already be setup and configured. Options are not | |
1127 | yet supported. | |
1128 | ||
1d59b382 SA |
1129 | lpuart,<addr> |
1130 | lpuart32,<addr> | |
1131 | Use early console provided by Freescale LP UART driver | |
1132 | found on Freescale Vybrid and QorIQ LS1021A processors. | |
1133 | A valid base address must be provided, and the serial | |
1134 | port must already be setup and configured. | |
1135 | ||
7fe068db JN |
1136 | ec_imx21,<addr> |
1137 | ec_imx6q,<addr> | |
1138 | Start an early, polled-mode, output-only console on the | |
1139 | Freescale i.MX UART at the specified address. The UART | |
1140 | must already be setup and configured. | |
1141 | ||
f7c864e7 | 1142 | ar3700_uart,<addr> |
30530791 WD |
1143 | Start an early, polled-mode console on the |
1144 | Armada 3700 serial port at the specified | |
1145 | address. The serial port must already be setup | |
1146 | and configured. Options are not yet supported. | |
1147 | ||
43f1831b KR |
1148 | qcom_geni,<addr> |
1149 | Start an early, polled-mode console on a Qualcomm | |
1150 | Generic Interface (GENI) based serial port at the | |
1151 | specified address. The serial port must already be | |
1152 | setup and configured. Options are not yet supported. | |
1153 | ||
69c1f396 AB |
1154 | efifb,[options] |
1155 | Start an early, unaccelerated console on the EFI | |
1156 | memory mapped framebuffer (if available). On cache | |
1157 | coherent non-x86 systems that use system memory for | |
1158 | the framebuffer, pass the 'ram' option so that it is | |
1159 | mapped with the correct attributes. | |
1160 | ||
09864c1c | 1161 | linflex,<addr> |
9905f32a | 1162 | Use early console provided by Freescale LINFlexD UART |
09864c1c SM |
1163 | serial driver for NXP S32V234 SoCs. A valid base |
1164 | address must be provided, and the serial port must | |
1165 | already be setup and configured. | |
1166 | ||
4ba66a97 | 1167 | earlyprintk= [X86,SH,ARM,M68k,S390] |
1da177e4 | 1168 | earlyprintk=vga |
89175cf7 | 1169 | earlyprintk=sclp |
2482a92e | 1170 | earlyprintk=xen |
1da177e4 | 1171 | earlyprintk=serial[,ttySn[,baudrate]] |
147ea091 | 1172 | earlyprintk=serial[,0x...[,baudrate]] |
ea3acb19 | 1173 | earlyprintk=ttySn[,baudrate] |
9780bc41 | 1174 | earlyprintk=dbgp[debugController#] |
d2266bbf | 1175 | earlyprintk=pciserial[,force],bus:device.function[,baudrate] |
1b5aeebf | 1176 | earlyprintk=xdbc[xhciController#] |
1da177e4 | 1177 | |
147ea091 DH |
1178 | earlyprintk is useful when the kernel crashes before |
1179 | the normal console is initialized. It is not enabled by | |
1180 | default because it has some cosmetic problems. | |
1181 | ||
a9913044 | 1182 | Append ",keep" to not disable it when the real console |
1da177e4 LT |
1183 | takes over. |
1184 | ||
72548e83 MF |
1185 | Only one of vga, efi, serial, or usb debug port can |
1186 | be used at a time. | |
1da177e4 | 1187 | |
147ea091 DH |
1188 | Currently only ttyS0 and ttyS1 may be specified by |
1189 | name. Other I/O ports may be explicitly specified | |
1190 | on some architectures (x86 and arm at least) by | |
1191 | replacing ttySn with an I/O port address, like this: | |
1192 | earlyprintk=serial,0x1008,115200 | |
1193 | You can find the port for a given device in | |
1194 | /proc/tty/driver/serial: | |
1195 | 2: uart:ST16650V2 port:00001008 irq:18 ... | |
1da177e4 LT |
1196 | |
1197 | Interaction with the standard serial driver is not | |
1198 | very good. | |
1199 | ||
72548e83 MF |
1200 | The VGA and EFI output is eventually overwritten by |
1201 | the real console. | |
1da177e4 | 1202 | |
2482a92e KRW |
1203 | The xen output can only be used by Xen PV guests. |
1204 | ||
89175cf7 HC |
1205 | The sclp output can only be used on s390. |
1206 | ||
d2266bbf FT |
1207 | The optional "force" to "pciserial" enables use of a |
1208 | PCI device even when its classcode is not of the | |
1209 | UART class. | |
1210 | ||
c700f013 CG |
1211 | edac_report= [HW,EDAC] Control how to report EDAC event |
1212 | Format: {"on" | "off" | "force"} | |
1213 | on: enable EDAC to report H/W event. May be overridden | |
1214 | by other higher priority error reporting module. | |
1215 | off: disable H/W event reporting through EDAC. | |
1216 | force: enforce the use of EDAC to report H/W event. | |
1217 | default: on. | |
1218 | ||
9731191f JW |
1219 | ekgdboc= [X86,KGDB] Allow early kernel console debugging |
1220 | ekgdboc=kbd | |
1221 | ||
25985edc | 1222 | This is designed to be used in conjunction with |
9731191f JW |
1223 | the boot argument: earlyprintk=vga |
1224 | ||
f71fc3bc DA |
1225 | This parameter works in place of the kgdboc parameter |
1226 | but can only be used if the backing tty is available | |
1227 | very early in the boot process. For early debugging | |
1228 | via a serial port see kgdboc_earlycon instead. | |
1229 | ||
1da177e4 | 1230 | edd= [EDD] |
8c4dd606 | 1231 | Format: {"off" | "on" | "skip[mbr]"} |
1da177e4 | 1232 | |
d2f7cbe7 | 1233 | efi= [EFI] |
c03f739f HS |
1234 | Format: { "debug", "disable_early_pci_dma", |
1235 | "nochunk", "noruntime", "nosoftreserve", | |
1236 | "novamap", "no_disable_early_pci_dma", | |
1237 | "old_map" } | |
1238 | debug: enable misc debug output. | |
1239 | disable_early_pci_dma: disable the busmaster bit on all | |
1240 | PCI bridges while in the EFI boot stub. | |
5a17dae4 MF |
1241 | nochunk: disable reading files in "chunks" in the EFI |
1242 | boot stub, as chunking can cause problems with some | |
1243 | firmware implementations. | |
5ae3683c | 1244 | noruntime : disable EFI runtime services support |
b617c526 DW |
1245 | nosoftreserve: The EFI_MEMORY_SP (Specific Purpose) |
1246 | attribute may cause the kernel to reserve the | |
1247 | memory range for a memory mapping driver to | |
1248 | claim. Specify efi=nosoftreserve to disable this | |
1249 | reservation and treat the memory by its base type | |
1250 | (i.e. EFI_CONVENTIONAL_MEMORY / "System RAM"). | |
c03f739f | 1251 | novamap: do not call SetVirtualAddressMap(). |
4444f854 MG |
1252 | no_disable_early_pci_dma: Leave the busmaster bit set |
1253 | on all PCI bridges while in the EFI boot stub | |
c03f739f HS |
1254 | old_map [X86-64]: switch to the old ioremap-based EFI |
1255 | runtime services mapping. [Needs CONFIG_X86_UV=y] | |
d2f7cbe7 | 1256 | |
8c58bf3e RW |
1257 | efi_no_storage_paranoia [EFI; X86] |
1258 | Using this parameter you can use more than 50% of | |
1259 | your efi variable storage. Use this parameter only if | |
1260 | you are really sure that your UEFI does sane gc and | |
1261 | fulfills the spec otherwise your board may brick. | |
1262 | ||
0f96a99d TI |
1263 | efi_fake_mem= nn[KMG]@ss[KMG]:aa[,nn[KMG]@ss[KMG]:aa,..] [EFI; X86] |
1264 | Add arbitrary attribute to specific memory range by | |
1265 | updating original EFI memory map. | |
1266 | Region of memory which aa attribute is added to is | |
1267 | from ss to ss+nn. | |
199c8471 | 1268 | |
0f96a99d TI |
1269 | If efi_fake_mem=2G@4G:0x10000,2G@0x10a0000000:0x10000 |
1270 | is specified, EFI_MEMORY_MORE_RELIABLE(0x10000) | |
1271 | attribute is added to range 0x100000000-0x180000000 and | |
1272 | 0x10a0000000-0x1120000000. | |
1273 | ||
199c8471 DW |
1274 | If efi_fake_mem=8G@9G:0x40000 is specified, the |
1275 | EFI_MEMORY_SP(0x40000) attribute is added to | |
1276 | range 0x240000000-0x43fffffff. | |
1277 | ||
0f96a99d | 1278 | Using this parameter you can do debugging of EFI memmap |
199c8471 | 1279 | related features. For example, you can do debugging of |
0f96a99d | 1280 | Address Range Mirroring feature even if your box |
199c8471 DW |
1281 | doesn't support it, or mark specific memory as |
1282 | "soft reserved". | |
0f96a99d | 1283 | |
475fb4e8 OP |
1284 | efivar_ssdt= [EFI; X86] Name of an EFI variable that contains an SSDT |
1285 | that is to be dynamically loaded by Linux. If there are | |
1286 | multiple variables with the same name but with different | |
1287 | vendor GUIDs, all of them will be loaded. See | |
cb1aaebe | 1288 | Documentation/admin-guide/acpi/ssdt-overlays.rst for details. |
475fb4e8 OP |
1289 | |
1290 | ||
1da177e4 LT |
1291 | eisa_irq_edge= [PARISC,HW] |
1292 | See header of drivers/parisc/eisa.c. | |
1293 | ||
cd4f0ef7 | 1294 | elanfreq= [X86-32] |
1da177e4 | 1295 | See comment before function elanfreq_setup() in |
71f77055 | 1296 | arch/x86/kernel/cpu/cpufreq/elanfreq.c. |
1da177e4 | 1297 | |
d3bf3795 | 1298 | elfcorehdr=[size[KMG]@]offset[KMG] [IA64,PPC,SH,X86,S390] |
a9913044 | 1299 | Specifies physical address of start of kernel core |
d3bf3795 MH |
1300 | image elf header and optionally the size. Generally |
1301 | kexec loader will pass this option to capture kernel. | |
330d4810 | 1302 | See Documentation/admin-guide/kdump/kdump.rst for details. |
1da177e4 | 1303 | |
0cb55ad2 RD |
1304 | enable_mtrr_cleanup [X86] |
1305 | The kernel tries to adjust MTRR layout from continuous | |
1306 | to discrete, to make X server driver able to add WB | |
1307 | entry later. This parameter enables that. | |
1308 | ||
ca1eda2d | 1309 | enable_timer_pin_1 [X86] |
0cb55ad2 RD |
1310 | Enable PIN 1 of APIC timer |
1311 | Can be useful to work around chipset bugs | |
1312 | (in particular on some ATI chipsets). | |
1313 | The kernel tries to set a reasonable default. | |
1314 | ||
1da177e4 LT |
1315 | enforcing [SELINUX] Set initial enforcing status. |
1316 | Format: {"0" | "1"} | |
1317 | See security/selinux/Kconfig help text. | |
1318 | 0 -- permissive (log only, no denials). | |
1319 | 1 -- enforcing (deny and log). | |
1320 | Default value is 0. | |
d41415eb SS |
1321 | Value can be changed at runtime via |
1322 | /sys/fs/selinux/enforce. | |
1da177e4 | 1323 | |
a08f82d0 HY |
1324 | erst_disable [ACPI] |
1325 | Disable Error Record Serialization Table (ERST) | |
1326 | support. | |
1327 | ||
1da177e4 LT |
1328 | ether= [HW,NET] Ethernet cards parameters |
1329 | This option is obsoleted by the "netdev=" option, which | |
1330 | has equivalent usage. See its documentation for details. | |
1331 | ||
7102ebcd MZ |
1332 | evm= [EVM] |
1333 | Format: { "fix" } | |
1334 | Permit 'security.evm' to be updated regardless of | |
1335 | current integrity status. | |
1336 | ||
de1ba09b AM |
1337 | failslab= |
1338 | fail_page_alloc= | |
1339 | fail_make_request=[KNL] | |
1340 | General fault injection mechanism. | |
1341 | Format: <interval>,<probability>,<space>,<times> | |
395cf969 | 1342 | See also Documentation/fault-injection/. |
de1ba09b | 1343 | |
1da177e4 | 1344 | floppy= [HW] |
e7751617 | 1345 | See Documentation/admin-guide/blockdev/floppy.rst. |
1da177e4 | 1346 | |
f13ae30e AC |
1347 | force_pal_cache_flush |
1348 | [IA-64] Avoid check_sal_cache_flush which may hang on | |
1349 | buggy SAL_CACHE_FLUSH implementations. Using this | |
1350 | parameter will force ia64_sal_cache_flush to call | |
1351 | ia64_pal_cache_flush instead of SAL_CACHE_FLUSH. | |
1352 | ||
6dddd7a7 | 1353 | forcepae [X86-32] |
69f2366c CB |
1354 | Forcefully enable Physical Address Extension (PAE). |
1355 | Many Pentium M systems disable PAE but may have a | |
1356 | functionally usable PAE implementation. | |
1357 | Warning: use of this parameter will taint the kernel | |
1358 | and may cause unknown problems. | |
1359 | ||
d9e54076 | 1360 | ftrace=[tracer] |
2af15d6a | 1361 | [FTRACE] will set and start the specified tracer |
d9e54076 PZ |
1362 | as early as possible in order to facilitate early |
1363 | boot debugging. | |
1364 | ||
cecbca96 | 1365 | ftrace_dump_on_oops[=orig_cpu] |
2af15d6a | 1366 | [FTRACE] will dump the trace buffers on oops. |
cecbca96 FW |
1367 | If no parameter is passed, ftrace will dump |
1368 | buffers of all CPUs, but if you pass orig_cpu, it will | |
1369 | dump only the buffer of the CPU that triggered the | |
1370 | oops. | |
2af15d6a SR |
1371 | |
1372 | ftrace_filter=[function-list] | |
1373 | [FTRACE] Limit the functions traced by the function | |
1374 | tracer at boot up. function-list is a comma separated | |
1375 | list of functions. This list can be changed at run | |
1376 | time by the set_ftrace_filter file in the debugfs | |
16290246 | 1377 | tracing directory. |
2af15d6a SR |
1378 | |
1379 | ftrace_notrace=[function-list] | |
1380 | [FTRACE] Do not trace the functions specified in | |
1381 | function-list. This list can be changed at run time | |
1382 | by the set_ftrace_notrace file in the debugfs | |
1383 | tracing directory. | |
d9e54076 | 1384 | |
369bc18f SA |
1385 | ftrace_graph_filter=[function-list] |
1386 | [FTRACE] Limit the top level callers functions traced | |
1387 | by the function graph tracer at boot up. | |
1388 | function-list is a comma separated list of functions | |
1389 | that can be changed at run time by the | |
1390 | set_graph_function file in the debugfs tracing directory. | |
1391 | ||
0d7d9a16 NK |
1392 | ftrace_graph_notrace=[function-list] |
1393 | [FTRACE] Do not trace from the functions specified in | |
1394 | function-list. This list is a comma separated list of | |
1395 | functions that can be changed at run time by the | |
1396 | set_graph_notrace file in the debugfs tracing directory. | |
1397 | ||
65a50c65 TB |
1398 | ftrace_graph_max_depth=<uint> |
1399 | [FTRACE] Used with the function graph tracer. This is | |
1400 | the max depth it will trace into a function. This value | |
1401 | can be changed at run time by the max_graph_depth file | |
1402 | in the tracefs tracing directory. default: 0 (no limit) | |
1403 | ||
8375e74f SK |
1404 | fw_devlink= [KNL] Create device links between consumer and supplier |
1405 | devices by scanning the firmware to infer the | |
1406 | consumer/supplier relationships. This feature is | |
1407 | especially useful when drivers are loaded as modules as | |
1408 | it ensures proper ordering of tasks like device probing | |
1409 | (suppliers first, then consumers), supplier boot state | |
1410 | clean up (only after all consumers have probed), | |
1411 | suspend/resume & runtime PM (consumers first, then | |
1412 | suppliers). | |
1413 | Format: { off | permissive | on | rpm } | |
1414 | off -- Don't create device links from firmware info. | |
1415 | permissive -- Create device links from firmware info | |
1416 | but use it only for ordering boot state clean | |
1417 | up (sync_state() calls). | |
1418 | on -- Create device links from firmware info and use it | |
1419 | to enforce probe and suspend/resume ordering. | |
1420 | rpm -- Like "on", but also use to order runtime PM. | |
1421 | ||
1da177e4 LT |
1422 | gamecon.map[2|3]= |
1423 | [HW,JOY] Multisystem joystick and NES/SNES/PSX pad | |
1424 | support via parallel port (up to 5 devices per port) | |
1425 | Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5> | |
1752118d | 1426 | See also Documentation/input/devices/joystick-parport.rst |
1da177e4 LT |
1427 | |
1428 | gamma= [HW,DRM] | |
1429 | ||
be3a5b0e | 1430 | gart_fix_e820= [X86-64] disable the fix e820 for K8 GART |
aaf23042 YL |
1431 | Format: off | on |
1432 | default: on | |
1433 | ||
2521f2c2 PO |
1434 | gcov_persist= [GCOV] When non-zero (default), profiling data for |
1435 | kernel modules is saved and remains accessible via | |
1436 | debugfs, even when the module is unloaded/reloaded. | |
1437 | When zero, profiling data is discarded and associated | |
1438 | debugfs files are removed at module unload time. | |
1439 | ||
47512cfd TG |
1440 | goldfish [X86] Enable the goldfish android emulator platform. |
1441 | Don't use this when you are not running on the | |
1442 | android emulator | |
1443 | ||
1da177e4 | 1444 | gpt [EFI] Forces disk with valid GPT signature but |
6c5de79b DB |
1445 | invalid Protective MBR to be treated as GPT. If the |
1446 | primary GPT is corrupted, it enables the backup/alternate | |
1447 | GPT to be used instead. | |
1da177e4 | 1448 | |
6cec9b07 AL |
1449 | grcan.enable0= [HW] Configuration of physical interface 0. Determines |
1450 | the "Enable 0" bit of the configuration register. | |
1451 | Format: 0 | 1 | |
1452 | Default: 0 | |
1453 | grcan.enable1= [HW] Configuration of physical interface 1. Determines | |
1454 | the "Enable 0" bit of the configuration register. | |
1455 | Format: 0 | 1 | |
1456 | Default: 0 | |
1457 | grcan.select= [HW] Select which physical interface to use. | |
1458 | Format: 0 | 1 | |
1459 | Default: 0 | |
1460 | grcan.txsize= [HW] Sets the size of the tx buffer. | |
1461 | Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0. | |
1462 | Default: 1024 | |
1463 | grcan.rxsize= [HW] Sets the size of the rx buffer. | |
1464 | Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0. | |
1465 | Default: 1024 | |
1466 | ||
0f98dd1b BJZ |
1467 | gpio-mockup.gpio_mockup_ranges |
1468 | [HW] Sets the ranges of gpiochip of for this device. | |
1469 | Format: <start1>,<end1>,<start2>,<end2>... | |
1470 | ||
55537871 JK |
1471 | hardlockup_all_cpu_backtrace= |
1472 | [KNL] Should the hard-lockup detector generate | |
1473 | backtraces on all cpus. | |
f117955a | 1474 | Format: 0 | 1 |
55537871 | 1475 | |
1da177e4 LT |
1476 | hashdist= [KNL,NUMA] Large hashes allocated during boot |
1477 | are distributed across NUMA nodes. Defaults on | |
16290246 | 1478 | for 64-bit NUMA, off otherwise. |
a9913044 | 1479 | Format: 0 | 1 (for off | on) |
1da177e4 LT |
1480 | |
1481 | hcl= [IA-64] SGI's Hardware Graph compatibility layer | |
1482 | ||
1483 | hd= [EIDE] (E)IDE hard drive subsystem geometry | |
1484 | Format: <cyl>,<head>,<sect> | |
1485 | ||
ea8c071c HY |
1486 | hest_disable [ACPI] |
1487 | Disable Hardware Error Source Table (HEST) support; | |
1488 | corresponding firmware-first mode error processing | |
1489 | logic will be disabled. | |
1490 | ||
1da177e4 LT |
1491 | highmem=nn[KMG] [KNL,BOOT] forces the highmem zone to have an exact |
1492 | size of <nn>. This works even on boxes that have no | |
1493 | highmem otherwise. This also works to reduce highmem | |
1494 | size on bigger boxes. | |
1495 | ||
54cdfdb4 TG |
1496 | highres= [KNL] Enable/disable high resolution timer mode. |
1497 | Valid parameters: "on", "off" | |
1498 | Default: "on" | |
1499 | ||
0cb55ad2 RD |
1500 | hlt [BUGS=ARM,SH] |
1501 | ||
1502 | hpet= [X86-32,HPET] option to control HPET usage | |
1503 | Format: { enable (default) | disable | force | | |
1504 | verbose } | |
1505 | disable: disable HPET and use PIT instead | |
1506 | force: allow force enabled of undocumented chips (ICH4, | |
1507 | VIA, nVidia) | |
1508 | verbose: show contents of HPET registers during setup | |
1509 | ||
3d035f58 PB |
1510 | hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET |
1511 | registers. Default set by CONFIG_HPET_MMAP_DEFAULT. | |
1512 | ||
cf11e85f RG |
1513 | hugetlb_cma= [HW] The size of a cma area used for allocation |
1514 | of gigantic hugepages. | |
1515 | Format: nn[KMGTPE] | |
1516 | ||
1517 | Reserve a cma area of given size and allocate gigantic | |
1518 | hugepages using the cma allocator. If enabled, the | |
1519 | boot-time allocation of gigantic hugepages is skipped. | |
1520 | ||
282f4214 MK |
1521 | hugepages= [HW] Number of HugeTLB pages to allocate at boot. |
1522 | If this follows hugepagesz (below), it specifies | |
1523 | the number of pages of hugepagesz to be allocated. | |
1524 | If this is the first HugeTLB parameter on the command | |
1525 | line, it specifies the number of pages to allocate for | |
1526 | the default huge page size. See also | |
1527 | Documentation/admin-guide/mm/hugetlbpage.rst. | |
1528 | Format: <integer> | |
1529 | ||
1530 | hugepagesz= | |
1531 | [HW] The size of the HugeTLB pages. This is used in | |
1532 | conjunction with hugepages (above) to allocate huge | |
1533 | pages of a specific size at boot. The pair | |
1534 | hugepagesz=X hugepages=Y can be specified once for | |
1535 | each supported huge page size. Huge page sizes are | |
1536 | architecture dependent. See also | |
1537 | Documentation/admin-guide/mm/hugetlbpage.rst. | |
1538 | Format: size[KMG] | |
6902aa84 | 1539 | |
a49d9c0a OS |
1540 | hung_task_panic= |
1541 | [KNL] Should the hung task detector generate panics. | |
f117955a | 1542 | Format: 0 | 1 |
cef7125d | 1543 | |
b467f3ef | 1544 | A value of 1 instructs the kernel to panic when a |
a49d9c0a OS |
1545 | hung task is detected. The default value is controlled |
1546 | by the CONFIG_BOOTPARAM_HUNG_TASK_PANIC build-time | |
1547 | option. The value selected by this boot parameter can | |
1548 | be changed later by the kernel.hung_task_panic sysctl. | |
1549 | ||
6dddd7a7 TB |
1550 | hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC) |
1551 | terminal devices. Valid values: 0..8 | |
1552 | hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs. | |
1553 | If specified, z/VM IUCV HVC accepts connections | |
1554 | from listed z/VM user IDs only. | |
3a025de6 YS |
1555 | |
1556 | hv_nopvspin [X86,HYPER_V] Disables the paravirt spinlock optimizations | |
1557 | which allow the hypervisor to 'idle' the | |
1558 | guest on lock contention. | |
1559 | ||
7bf69395 FDN |
1560 | keep_bootcon [KNL] |
1561 | Do not unregister boot console at start. This is only | |
1562 | useful for debugging when something happens in the window | |
1563 | between unregistering the boot console and initializing | |
1564 | the real console. | |
1565 | ||
6dddd7a7 TB |
1566 | i2c_bus= [HW] Override the default board specific I2C bus speed |
1567 | or register an additional I2C bus that is not | |
1568 | registered from board initialization code. | |
1569 | Format: | |
1570 | <bus_id>,<clkrate> | |
3a853fb9 | 1571 | |
36d95739 | 1572 | i8042.debug [HW] Toggle i8042 debug mode |
e1443d28 SCP |
1573 | i8042.unmask_kbd_data |
1574 | [HW] Enable printing of interrupt data from the KBD port | |
1575 | (disabled by default, and as a pre-condition | |
1576 | requires that i8042.debug=1 be enabled) | |
1da177e4 | 1577 | i8042.direct [HW] Put keyboard port into non-translated mode |
84eb8d06 ML |
1578 | i8042.dumbkbd [HW] Pretend that controller can only read data from |
1579 | keyboard and cannot control its state | |
1da177e4 LT |
1580 | (Don't attempt to blink the leds) |
1581 | i8042.noaux [HW] Don't check for auxiliary (== mouse) port | |
945ef0d4 | 1582 | i8042.nokbd [HW] Don't check/create keyboard port |
75d08c78 JK |
1583 | i8042.noloop [HW] Disable the AUX Loopback command while probing |
1584 | for the AUX port | |
1da177e4 | 1585 | i8042.nomux [HW] Don't check presence of an active multiplexing |
e55a3366 | 1586 | controller |
1da177e4 LT |
1587 | i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX |
1588 | controllers | |
24775d65 | 1589 | i8042.notimeout [HW] Ignore timeout condition signalled by controller |
930e1924 MPS |
1590 | i8042.reset [HW] Reset the controller during init, cleanup and |
1591 | suspend-to-ram transitions, only during s2r | |
1592 | transitions, or never reset | |
1593 | Format: { 1 | Y | y | 0 | N | n } | |
1594 | 1, Y, y: always reset controller | |
1595 | 0, N, n: don't ever reset controller | |
1596 | Default: only on s2r transitions on x86; most other | |
1597 | architectures force reset to be always executed | |
1da177e4 | 1598 | i8042.unlock [HW] Unlock (ignore) the keylock |
6dddd7a7 | 1599 | i8042.kbdreset [HW] Reset device connected to KBD port |
1da177e4 LT |
1600 | |
1601 | i810= [HW,DRM] | |
1602 | ||
e70c9d5e DT |
1603 | i8k.ignore_dmi [HW] Continue probing hardware even if DMI data |
1604 | indicates that the driver is running on unsupported | |
1605 | hardware. | |
1da177e4 LT |
1606 | i8k.force [HW] Activate i8k driver even if SMM BIOS signature |
1607 | does not match list of supported models. | |
1608 | i8k.power_status | |
1609 | [HW] Report power status in /proc/i8k | |
1610 | (disabled by default) | |
1611 | i8k.restricted [HW] Allow controlling fans only if SYS_ADMIN | |
1612 | capability is set. | |
1613 | ||
4dca20ef | 1614 | i915.invert_brightness= |
7bd90909 CE |
1615 | [DRM] Invert the sense of the variable that is used to |
1616 | set the brightness of the panel backlight. Normally a | |
4dca20ef CE |
1617 | brightness value of 0 indicates backlight switched off, |
1618 | and the maximum of the brightness value sets the backlight | |
1619 | to maximum brightness. If this parameter is set to 0 | |
1620 | (default) and the machine requires it, or this parameter | |
1621 | is set to 1, a brightness value of 0 sets the backlight | |
1622 | to maximum brightness, and the maximum of the brightness | |
1623 | value switches the backlight off. | |
1624 | -1 -- never invert brightness | |
1625 | 0 -- machine default | |
1626 | 1 -- force brightness inversion | |
7bd90909 | 1627 | |
1da177e4 LT |
1628 | icn= [HW,ISDN] |
1629 | Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]] | |
1630 | ||
0af80c04 DF |
1631 | ide-core.nodma= [HW] (E)IDE subsystem |
1632 | Format: =0.0 to prevent dma on hda, =0.1 hdb =1.0 hdc | |
075affcb BZ |
1633 | .vlb_clock .pci_clock .noflush .nohpa .noprobe .nowerr |
1634 | .cdrom .chs .ignore_cable are additional options | |
d7b461c5 | 1635 | See Documentation/ide/ide.rst. |
1da177e4 | 1636 | |
0f8b7f5d MR |
1637 | ide-generic.probe-mask= [HW] (E)IDE subsystem |
1638 | Format: <int> | |
1639 | Probe mask for legacy ISA IDE ports. Depending on | |
1640 | platform up to 6 ports are supported, enabled by | |
1641 | setting corresponding bits in the mask to 1. The | |
1642 | default value is 0x0, which has a special meaning. | |
1643 | On systems that have PCI, it triggers scanning the | |
1644 | PCI bus for the first and the second port, which | |
1645 | are then probed. On systems without PCI the value | |
1646 | of 0x0 enables probing the two first ports as if it | |
1647 | was 0x3. | |
1648 | ||
0cb55ad2 RD |
1649 | ide-pci-generic.all-generic-ide [HW] (E)IDE subsystem |
1650 | Claim all unknown PCI IDE storage controllers. | |
1651 | ||
f039b754 | 1652 | idle= [X86] |
69fb3676 | 1653 | Format: idle=poll, idle=halt, idle=nomwait |
ada9cfdd RD |
1654 | Poll forces a polling idle loop that can slightly |
1655 | improve the performance of waking up a idle CPU, but | |
1656 | will use a lot of power and make the system run hot. | |
1657 | Not recommended. | |
ada9cfdd | 1658 | idle=halt: Halt is forced to be used for CPU idle. |
c1e3b377 | 1659 | In such case C2/C3 won't be used again. |
ada9cfdd | 1660 | idle=nomwait: Disable mwait for CPU C-states |
a9913044 | 1661 | |
503943e0 MR |
1662 | ieee754= [MIPS] Select IEEE Std 754 conformance mode |
1663 | Format: { strict | legacy | 2008 | relaxed } | |
1664 | Default: strict | |
1665 | ||
1666 | Choose which programs will be accepted for execution | |
1667 | based on the IEEE 754 NaN encoding(s) supported by | |
1668 | the FPU and the NaN encoding requested with the value | |
1669 | of an ELF file header flag individually set by each | |
1670 | binary. Hardware implementations are permitted to | |
1671 | support either or both of the legacy and the 2008 NaN | |
1672 | encoding mode. | |
1673 | ||
1674 | Available settings are as follows: | |
1675 | strict accept binaries that request a NaN encoding | |
1676 | supported by the FPU | |
1677 | legacy only accept legacy-NaN binaries, if supported | |
1678 | by the FPU | |
1679 | 2008 only accept 2008-NaN binaries, if supported | |
1680 | by the FPU | |
1681 | relaxed accept any binaries regardless of whether | |
1682 | supported by the FPU | |
1683 | ||
1684 | The FPU emulator is always able to support both NaN | |
1685 | encodings, so if no FPU hardware is present or it has | |
1686 | been disabled with 'nofpu', then the settings of | |
1687 | 'legacy' and '2008' strap the emulator accordingly, | |
1688 | 'relaxed' straps the emulator for both legacy-NaN and | |
1689 | 2008-NaN, whereas 'strict' enables legacy-NaN only on | |
1690 | legacy processors and both NaN encodings on MIPS32 or | |
1691 | MIPS64 CPUs. | |
1692 | ||
1693 | The setting for ABS.fmt/NEG.fmt instruction execution | |
1694 | mode generally follows that for the NaN encoding, | |
1695 | except where unsupported by hardware. | |
1696 | ||
79290822 IM |
1697 | ignore_loglevel [KNL] |
1698 | Ignore loglevel setting - this will print /all/ | |
1699 | kernel messages to the console. Useful for debugging. | |
0eca6b7c YZ |
1700 | We also add it as printk module parameter, so users |
1701 | could change it dynamically, usually by | |
1702 | /sys/module/printk/parameters/ignore_loglevel. | |
79290822 | 1703 | |
d977d56c KK |
1704 | ignore_rlimit_data |
1705 | Ignore RLIMIT_DATA setting for data mappings, | |
1706 | print warning at first misuse. Can be changed via | |
1707 | /sys/module/kernel/parameters/ignore_rlimit_data. | |
1708 | ||
1da177e4 LT |
1709 | ihash_entries= [KNL] |
1710 | Set number of hash buckets for inode cache. | |
1711 | ||
2fe5d6de | 1712 | ima_appraise= [IMA] appraise integrity measurements |
2faa6ef3 | 1713 | Format: { "off" | "enforce" | "fix" | "log" } |
2fe5d6de MZ |
1714 | default: "enforce" |
1715 | ||
41475a3e | 1716 | ima_appraise_tcb [IMA] Deprecated. Use ima_policy= instead. |
07f6a794 MZ |
1717 | The builtin appraise policy appraises all files |
1718 | owned by uid=0. | |
1719 | ||
d68a6fe9 MZ |
1720 | ima_canonical_fmt [IMA] |
1721 | Use the canonical format for the binary runtime | |
1722 | measurements, instead of host native format. | |
1723 | ||
3323eec9 | 1724 | ima_hash= [IMA] |
e7a2ad7e MZ |
1725 | Format: { md5 | sha1 | rmd160 | sha256 | sha384 |
1726 | | sha512 | ... } | |
3323eec9 MZ |
1727 | default: "sha1" |
1728 | ||
e7a2ad7e MZ |
1729 | The list of supported hash algorithms is defined |
1730 | in crypto/hash_info.h. | |
1731 | ||
24fd03c8 | 1732 | ima_policy= [IMA] |
33ce9549 | 1733 | The builtin policies to load during IMA setup. |
9e67028e MZ |
1734 | Format: "tcb | appraise_tcb | secure_boot | |
1735 | fail_securely" | |
33ce9549 MZ |
1736 | |
1737 | The "tcb" policy measures all programs exec'd, files | |
1738 | mmap'd for exec, and all files opened with the read | |
1739 | mode bit set by either the effective uid (euid=0) or | |
1740 | uid=0. | |
1741 | ||
1742 | The "appraise_tcb" policy appraises the integrity of | |
41475a3e | 1743 | all files owned by root. |
24fd03c8 | 1744 | |
503ceaef MZ |
1745 | The "secure_boot" policy appraises the integrity |
1746 | of files (eg. kexec kernel image, kernel modules, | |
1747 | firmware, policy, etc) based on file signatures. | |
24fd03c8 | 1748 | |
9e67028e MZ |
1749 | The "fail_securely" policy forces file signature |
1750 | verification failure also on privileged mounted | |
1751 | filesystems with the SB_I_UNVERIFIABLE_SIGNATURE | |
1752 | flag. | |
1753 | ||
24fd03c8 | 1754 | ima_tcb [IMA] Deprecated. Use ima_policy= instead. |
5789ba3b EP |
1755 | Load a policy which meets the needs of the Trusted |
1756 | Computing Base. This means IMA will measure all | |
1757 | programs exec'd, files mmap'd for exec, and all files | |
1758 | opened for read by uid=0. | |
1759 | ||
6dddd7a7 | 1760 | ima_template= [IMA] |
9b9d4ce5 | 1761 | Select one of defined IMA measurements template formats. |
8265a2f8 | 1762 | Formats: { "ima" | "ima-ng" | "ima-sig" } |
9b9d4ce5 RS |
1763 | Default: "ima-ng" |
1764 | ||
c2426d2a | 1765 | ima_template_fmt= |
6dddd7a7 | 1766 | [IMA] Define a custom template format. |
c2426d2a RS |
1767 | Format: { "field1|...|fieldN" } |
1768 | ||
3bcced39 DK |
1769 | ima.ahash_minsize= [IMA] Minimum file size for asynchronous hash usage |
1770 | Format: <min_file_size> | |
1771 | Set the minimal file size for using asynchronous hash. | |
1772 | If left unspecified, ahash usage is disabled. | |
1773 | ||
1774 | ahash performance varies for different data sizes on | |
1775 | different crypto accelerators. This option can be used | |
1776 | to achieve the best performance for a particular HW. | |
1777 | ||
6edf7a89 DK |
1778 | ima.ahash_bufsize= [IMA] Asynchronous hash buffer size |
1779 | Format: <bufsize> | |
1780 | Set hashing buffer size. Default: 4k. | |
1781 | ||
1782 | ahash performance varies for different chunk sizes on | |
1783 | different crypto accelerators. This option can be used | |
1784 | to achieve best performance for particular HW. | |
1785 | ||
1da177e4 LT |
1786 | init= [KNL] |
1787 | Format: <full_path> | |
1788 | Run specified binary instead of /sbin/init as init | |
1789 | process. | |
1790 | ||
1791 | initcall_debug [KNL] Trace initcalls as they are executed. Useful | |
1792 | for working out where the kernel is dying during | |
1793 | startup. | |
1794 | ||
7b0b73d7 PB |
1795 | initcall_blacklist= [KNL] Do not execute a comma-separated list of |
1796 | initcall functions. Useful for debugging built-in | |
1797 | modules and initcalls. | |
1798 | ||
1da177e4 LT |
1799 | initrd= [BOOT] Specify the location of the initial ramdisk |
1800 | ||
694cfd87 RM |
1801 | initrdmem= [KNL] Specify a physical address and size from which to |
1802 | load the initrd. If an initrd is compiled in or | |
1803 | specified in the bootparams, it takes priority over this | |
1804 | setting. | |
1805 | Format: ss[KMG],nn[KMG] | |
1806 | Default is 0, 0 | |
1807 | ||
6471384a AP |
1808 | init_on_alloc= [MM] Fill newly allocated pages and heap objects with |
1809 | zeroes. | |
1810 | Format: 0 | 1 | |
1811 | Default set by CONFIG_INIT_ON_ALLOC_DEFAULT_ON. | |
1812 | ||
1813 | init_on_free= [MM] Fill freed pages and heap objects with zeroes. | |
1814 | Format: 0 | 1 | |
1815 | Default set by CONFIG_INIT_ON_FREE_DEFAULT_ON. | |
1816 | ||
be3a5b0e | 1817 | init_pkru= [X86] Specify the default memory protection keys rights |
acd547b2 DH |
1818 | register contents for all processes. 0x55555554 by |
1819 | default (disallow access to all but pkey 0). Can | |
1820 | override in debugfs after boot. | |
1821 | ||
1da177e4 LT |
1822 | inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver |
1823 | Format: <irq> | |
1824 | ||
be3a5b0e | 1825 | int_pln_enable [X86] Enable power limit notification interrupt |
6bb2ff84 | 1826 | |
d726d8d7 MZ |
1827 | integrity_audit=[IMA] |
1828 | Format: { "0" | "1" } | |
1829 | 0 -- basic integrity auditing messages. (Default) | |
1830 | 1 -- additional integrity auditing messages. | |
1831 | ||
ba395927 | 1832 | intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option |
0cd5c3c8 KM |
1833 | on |
1834 | Enable intel iommu driver. | |
ba395927 KA |
1835 | off |
1836 | Disable intel iommu driver. | |
1837 | igfx_off [Default Off] | |
1838 | By default, gfx is mapped as normal device. If a gfx | |
1839 | device has a dedicated DMAR unit, the DMAR unit is | |
1840 | bypassed by not enabling DMAR with this option. In | |
1841 | this case, gfx device will use physical address for | |
1842 | DMA. | |
be3a5b0e | 1843 | forcedac [X86-64] |
7d3b03ce | 1844 | With this option iommu will not optimize to look |
16290246 | 1845 | for io virtual address below 32-bit forcing dual |
7d3b03ce | 1846 | address cycle on pci bus for cards supporting greater |
16290246 RD |
1847 | than 32-bit addressing. The default is to look |
1848 | for translation below 32-bit and if not available | |
7d3b03ce | 1849 | then look in the higher range. |
5e0d2a6f | 1850 | strict [Default Off] |
1851 | With this option on every unmap_single operation will | |
1852 | result in a hardware IOTLB flush operation as opposed | |
1853 | to batching them for performance. | |
6dd9a7c7 YS |
1854 | sp_off [Default Off] |
1855 | By default, super page will be supported if Intel IOMMU | |
1856 | has the capability. With this option, super page will | |
1857 | not be supported. | |
8950dcd8 LB |
1858 | sm_on [Default Off] |
1859 | By default, scalable mode will be disabled even if the | |
765b6a98 LB |
1860 | hardware advertises that it has support for the scalable |
1861 | mode translation. With this option set, scalable mode | |
8950dcd8 | 1862 | will be used on hardware which claims to support it. |
bfd20f1c SL |
1863 | tboot_noforce [Default Off] |
1864 | Do not force the Intel IOMMU enabled under tboot. | |
1865 | By default, tboot will force Intel IOMMU on, which | |
1866 | could harm performance of some high-throughput | |
1867 | devices like 40GBit network cards, even if identity | |
1868 | mapping is enabled. | |
1869 | Note that using this option lowers the security | |
1870 | provided by tboot because it makes the system | |
1871 | vulnerable to DMA attacks. | |
e5e04d05 | 1872 | nobounce [Default off] |
bf347b9d | 1873 | Disable bounce buffer for untrusted devices such as |
e5e04d05 LB |
1874 | the Thunderbolt devices. This will treat the untrusted |
1875 | devices as the trusted ones, hence might expose security | |
1876 | risks of DMA attacks. | |
2e92c7ad MI |
1877 | |
1878 | intel_idle.max_cstate= [KNL,HW,ACPI,X86] | |
1879 | 0 disables intel_idle and fall back on acpi_idle. | |
22c6bbe4 | 1880 | 1 to 9 specify maximum depth of C-state. |
2e92c7ad | 1881 | |
6dddd7a7 TB |
1882 | intel_pstate= [X86] |
1883 | disable | |
1884 | Do not enable intel_pstate as the default | |
1885 | scaling driver for the supported processors | |
1886 | passive | |
1887 | Use intel_pstate as a scaling driver, but configure it | |
1888 | to work with generic cpufreq governors (instead of | |
1889 | enabling its internal governor). This mode cannot be | |
1890 | used along with the hardware-managed P-states (HWP) | |
1891 | feature. | |
1892 | force | |
1893 | Enable intel_pstate on systems that prohibit it by default | |
1894 | in favor of acpi-cpufreq. Forcing the intel_pstate driver | |
1895 | instead of acpi-cpufreq may disable platform features, such | |
1896 | as thermal controls and power capping, that rely on ACPI | |
1897 | P-States information being indicated to OSPM and therefore | |
1898 | should be used with caution. This option does not work with | |
1899 | processors that aren't supported by the intel_pstate driver | |
1900 | or on platforms that use pcc-cpufreq instead of acpi-cpufreq. | |
1901 | no_hwp | |
1902 | Do not enable hardware P state control (HWP) | |
1903 | if available. | |
1904 | hwp_only | |
1905 | Only load intel_pstate on systems which support | |
1906 | hardware P state control (HWP) if available. | |
1907 | support_acpi_ppc | |
1908 | Enforce ACPI _PPC performance limits. If the Fixed ACPI | |
1909 | Description Table, specifies preferred power management | |
1910 | profile as "Enterprise Server" or "Performance Server", | |
1911 | then this feature is turned on by default. | |
1912 | per_cpu_perf_limits | |
1913 | Allow per-logical-CPU P-State performance control limits using | |
1914 | cpufreq sysfs interface | |
6be26498 | 1915 | |
d1423d56 | 1916 | intremap= [X86-64, Intel-IOMMU] |
d1423d56 CW |
1917 | on enable Interrupt Remapping (default) |
1918 | off disable Interrupt Remapping | |
1919 | nosid disable Source ID checking | |
41750d31 SS |
1920 | no_x2apic_optout |
1921 | BIOS x2APIC opt-out request will be ignored | |
b7d20631 | 1922 | nopost disable Interrupt Posting |
d1423d56 | 1923 | |
0cb55ad2 RD |
1924 | iomem= Disable strict checking of access to MMIO memory |
1925 | strict regions from userspace. | |
1926 | relaxed | |
1927 | ||
be3a5b0e | 1928 | iommu= [X86] |
0cb55ad2 RD |
1929 | off |
1930 | force | |
1931 | noforce | |
1932 | biomerge | |
1933 | panic | |
1934 | nopanic | |
1935 | merge | |
1936 | nomerge | |
0cb55ad2 | 1937 | soft |
be3a5b0e RD |
1938 | pt [X86] |
1939 | nopt [X86] | |
4e287840 TLSC |
1940 | nobypass [PPC/POWERNV] |
1941 | Disable IOMMU bypass, using IOMMU for PCI devices. | |
bcb71abe | 1942 | |
68a6efe8 ZL |
1943 | iommu.strict= [ARM64] Configure TLB invalidation behaviour |
1944 | Format: { "0" | "1" } | |
1945 | 0 - Lazy mode. | |
1946 | Request that DMA unmap operations use deferred | |
1947 | invalidation of hardware TLBs, for increased | |
1948 | throughput at the cost of reduced device isolation. | |
1949 | Will fall back to strict mode if not supported by | |
1950 | the relevant IOMMU driver. | |
1951 | 1 - Strict mode (default). | |
1952 | DMA unmap operations invalidate IOMMU hardware TLBs | |
1953 | synchronously. | |
1954 | ||
fccb4e3b | 1955 | iommu.passthrough= |
c8fb436b | 1956 | [ARM64, X86] Configure DMA to bypass the IOMMU by default. |
fccb4e3b WD |
1957 | Format: { "0" | "1" } |
1958 | 0 - Use IOMMU translation for DMA. | |
1959 | 1 - Bypass the IOMMU for DMA. | |
9d723b4c | 1960 | unset - Use value of CONFIG_IOMMU_DEFAULT_PASSTHROUGH. |
0cb55ad2 | 1961 | |
7c42376e | 1962 | io7= [HW] IO7 for Marvel-based Alpha systems |
0cb55ad2 RD |
1963 | See comment before marvel_specify_io7 in |
1964 | arch/alpha/kernel/core_marvel.c. | |
1965 | ||
6cececfc | 1966 | io_delay= [X86] I/O delay method |
6e7c4025 IM |
1967 | 0x80 |
1968 | Standard port 0x80 based delay | |
1969 | 0xed | |
1970 | Alternate port 0xed based delay (needed on some systems) | |
b02aae9c | 1971 | udelay |
6e7c4025 IM |
1972 | Simple two microseconds delay |
1973 | none | |
1974 | No delay | |
b02aae9c | 1975 | |
1da177e4 | 1976 | ip= [IP_PNP] |
3eb30c51 | 1977 | See Documentation/admin-guide/nfs/nfsroot.rst. |
1da177e4 | 1978 | |
5ac893b8 WL |
1979 | ipcmni_extend [KNL] Extend the maximum number of unique System V |
1980 | IPC identifiers from 32,768 to 16,777,216. | |
1981 | ||
fbf19803 | 1982 | irqaffinity= [SMP] Set the default irq affinity mask |
2d13e6ca | 1983 | The argument is a cpu list, as described above. |
fbf19803 | 1984 | |
0962289b MZ |
1985 | irqchip.gicv2_force_probe= |
1986 | [ARM, ARM64] | |
1987 | Format: <bool> | |
1988 | Force the kernel to look for the second 4kB page | |
1989 | of a GICv2 controller even if the memory range | |
1990 | exposed by the device tree is too small. | |
1991 | ||
f736d65d MZ |
1992 | irqchip.gicv3_nolpi= |
1993 | [ARM, ARM64] | |
1994 | Force the kernel to ignore the availability of | |
1995 | LPIs (and by consequence ITSs). Intended for system | |
1996 | that use the kernel as a bootloader, and thus want | |
1997 | to let secondary kernels in charge of setting up | |
1998 | LPIs. | |
1999 | ||
bc3c03cc JT |
2000 | irqchip.gicv3_pseudo_nmi= [ARM64] |
2001 | Enables support for pseudo-NMIs in the kernel. This | |
2002 | requires the kernel to be built with | |
2003 | CONFIG_ARM64_PSEUDO_NMI. | |
2004 | ||
200803df AC |
2005 | irqfixup [HW] |
2006 | When an interrupt is not handled search all handlers | |
2007 | for it. Intended to get systems with badly broken | |
2008 | firmware running. | |
2009 | ||
2010 | irqpoll [HW] | |
2011 | When an interrupt is not handled search all handlers | |
2012 | for it. Also check all handlers each timer | |
2013 | interrupt. Intended to get systems with badly broken | |
2014 | firmware running. | |
2015 | ||
1da177e4 | 2016 | isapnp= [ISAPNP] |
a9913044 | 2017 | Format: <RDP>,<reset>,<pci_scan>,<verbosity> |
1da177e4 | 2018 | |
d94d1053 | 2019 | isolcpus= [KNL,SMP,ISOL] Isolate a given set of CPUs from disturbance. |
b0d40d2b FW |
2020 | [Deprecated - use cpusets instead] |
2021 | Format: [flag-list,]<cpu-list> | |
2022 | ||
2023 | Specify one or more CPUs to isolate from disturbances | |
2024 | specified in the flag list (default: domain): | |
2025 | ||
2026 | nohz | |
2027 | Disable the tick when a single task runs. | |
083c6eea FW |
2028 | |
2029 | A residual 1Hz tick is offloaded to workqueues, which you | |
2030 | need to affine to housekeeping through the global | |
2031 | workqueue's affinity configured via the | |
2032 | /sys/devices/virtual/workqueue/cpumask sysfs file, or | |
2033 | by using the 'domain' flag described below. | |
2034 | ||
2035 | NOTE: by default the global workqueue runs on all CPUs, | |
2036 | so to protect individual CPUs the 'cpumask' file has to | |
2037 | be configured manually after bootup. | |
2038 | ||
b0d40d2b FW |
2039 | domain |
2040 | Isolate from the general SMP balancing and scheduling | |
2041 | algorithms. Note that performing domain isolation this way | |
2042 | is irreversible: it's not possible to bring back a CPU to | |
2043 | the domains once isolated through isolcpus. It's strongly | |
2044 | advised to use cpusets instead to disable scheduler load | |
2045 | balancing through the "cpuset.sched_load_balance" file. | |
2046 | It offers a much more flexible interface where CPUs can | |
2047 | move in and out of an isolated set anytime. | |
2048 | ||
2049 | You can move a process onto or off an "isolated" CPU via | |
2050 | the CPU affinity syscalls or cpuset. | |
2051 | <cpu number> begins at 0 and the maximum value is | |
2052 | "number of CPUs in system - 1". | |
2053 | ||
11ea68f5 ML |
2054 | managed_irq |
2055 | ||
2056 | Isolate from being targeted by managed interrupts | |
2057 | which have an interrupt mask containing isolated | |
2058 | CPUs. The affinity of managed interrupts is | |
2059 | handled by the kernel and cannot be changed via | |
2060 | the /proc/irq/* interfaces. | |
2061 | ||
2062 | This isolation is best effort and only effective | |
2063 | if the automatically assigned interrupt mask of a | |
2064 | device queue contains isolated and housekeeping | |
2065 | CPUs. If housekeeping CPUs are online then such | |
2066 | interrupts are directed to the housekeeping CPU | |
2067 | so that IO submitted on the housekeeping CPU | |
2068 | cannot disturb the isolated CPU. | |
2069 | ||
2070 | If a queue's affinity mask contains only isolated | |
2071 | CPUs then this parameter has no effect on the | |
2072 | interrupt routing decision, though interrupts are | |
2073 | only delivered when tasks running on those | |
2074 | isolated CPUs submit IO. IO submitted on | |
2075 | housekeeping CPUs has no influence on those | |
2076 | queues. | |
1da177e4 | 2077 | |
11ea68f5 | 2078 | The format of <cpu-list> is described above. |
1da177e4 | 2079 | |
a9913044 | 2080 | iucv= [HW,NET] |
1da177e4 | 2081 | |
be3a5b0e | 2082 | ivrs_ioapic [HW,X86-64] |
7d8bfa26 JR |
2083 | Provide an override to the IOAPIC-ID<->DEVICE-ID |
2084 | mapping provided in the IVRS ACPI table. For | |
2085 | example, to map IOAPIC-ID decimal 10 to | |
2086 | PCI device 00:14.0 write the parameter as: | |
2087 | ivrs_ioapic[10]=00:14.0 | |
2088 | ||
be3a5b0e | 2089 | ivrs_hpet [HW,X86-64] |
7d8bfa26 JR |
2090 | Provide an override to the HPET-ID<->DEVICE-ID |
2091 | mapping provided in the IVRS ACPI table. For | |
2092 | example, to map HPET-ID decimal 0 to | |
2093 | PCI device 00:14.0 write the parameter as: | |
2094 | ivrs_hpet[0]=00:14.0 | |
2095 | ||
be3a5b0e | 2096 | ivrs_acpihid [HW,X86-64] |
ca3bf5d4 SS |
2097 | Provide an override to the ACPI-HID:UID<->DEVICE-ID |
2098 | mapping provided in the IVRS ACPI table. For | |
2099 | example, to map UART-HID:UID AMD0020:0 to | |
2100 | PCI device 00:14.5 write the parameter as: | |
2101 | ivrs_acpihid[00:14.5]=AMD0020:0 | |
2102 | ||
1da177e4 | 2103 | js= [HW,JOY] Analog joystick |
1752118d | 2104 | See Documentation/input/joydev/joystick.rst. |
1da177e4 | 2105 | |
65fe935d KC |
2106 | nokaslr [KNL] |
2107 | When CONFIG_RANDOMIZE_BASE is set, this disables | |
2108 | kernel and module base offset ASLR (Address Space | |
2109 | Layout Randomization). | |
24f2e027 | 2110 | |
b0845ce5 MR |
2111 | kasan_multi_shot |
2112 | [KNL] Enforce KASAN (Kernel Address Sanitizer) to print | |
2113 | report on every invalid memory access. Without this | |
2114 | parameter KASAN will print report only for the first | |
2115 | invalid access. | |
2116 | ||
0cb55ad2 RD |
2117 | keepinitrd [HW,ARM] |
2118 | ||
342332e6 | 2119 | kernelcore= [KNL,X86,IA-64,PPC] |
a5c6d650 DR |
2120 | Format: nn[KMGTPE] | nn% | "mirror" |
2121 | This parameter specifies the amount of memory usable by | |
2122 | the kernel for non-movable allocations. The requested | |
2123 | amount is spread evenly throughout all nodes in the | |
2124 | system as ZONE_NORMAL. The remaining memory is used for | |
2125 | movable memory in its own zone, ZONE_MOVABLE. In the | |
2126 | event, a node is too small to have both ZONE_NORMAL and | |
2127 | ZONE_MOVABLE, kernelcore memory will take priority and | |
2128 | other nodes will have a larger ZONE_MOVABLE. | |
2129 | ||
2130 | ZONE_MOVABLE is used for the allocation of pages that | |
2131 | may be reclaimed or moved by the page migration | |
2132 | subsystem. Note that allocations like PTEs-from-HighMem | |
2133 | still use the HighMem zone if it exists, and the Normal | |
ed7ed365 MG |
2134 | zone if it does not. |
2135 | ||
a5c6d650 DR |
2136 | It is possible to specify the exact amount of memory in |
2137 | the form of "nn[KMGTPE]", a percentage of total system | |
2138 | memory in the form of "nn%", or "mirror". If "mirror" | |
342332e6 TI |
2139 | option is specified, mirrored (reliable) memory is used |
2140 | for non-movable allocations and remaining memory is used | |
a5c6d650 DR |
2141 | for Movable pages. "nn[KMGTPE]", "nn%", and "mirror" |
2142 | are exclusive, so you cannot specify multiple forms. | |
ed7ed365 | 2143 | |
4fe1da4e JW |
2144 | kgdbdbgp= [KGDB,HW] kgdb over EHCI usb debug port. |
2145 | Format: <Controller#>[,poll interval] | |
2146 | The controller # is the number of the ehci usb debug | |
2147 | port as it is probed via PCI. The poll interval is | |
2148 | optional and is the number seconds in between | |
2149 | each poll cycle to the debug port in case you need | |
2150 | the functionality for interrupting the kernel with | |
2151 | gdb or control-c on the dbgp connection. When | |
2152 | not using this parameter you use sysrq-g to break into | |
2153 | the kernel debugger. | |
2154 | ||
84c08fd6 | 2155 | kgdboc= [KGDB,HW] kgdb over consoles. |
ada64e4c JW |
2156 | Requires a tty driver that supports console polling, |
2157 | or a supported polling keyboard driver (non-usb). | |
65b5ac14 JW |
2158 | Serial only format: <serial_device>[,baud] |
2159 | keyboard only format: kbd | |
2160 | keyboard and serial format: kbd,<serial_device>[,baud] | |
2161 | Optional Kernel mode setting: | |
2162 | kms, kbd format: kms,kbd | |
2163 | kms, kbd and serial format: kms,kbd,<ser_dev>[,baud] | |
6cdf6e06 | 2164 | |
f71fc3bc DA |
2165 | kgdboc_earlycon= [KGDB,HW] |
2166 | If the boot console provides the ability to read | |
2167 | characters and can work in polling mode, you can use | |
2168 | this parameter to tell kgdb to use it as a backend | |
2169 | until the normal console is registered. Intended to | |
2170 | be used together with the kgdboc parameter which | |
2171 | specifies the normal console to transition to. | |
2172 | ||
2173 | The name of the early console should be specified | |
2174 | as the value of this parameter. Note that the name of | |
2175 | the early console might be different than the tty | |
2176 | name passed to kgdboc. It's OK to leave the value | |
2177 | blank and the first boot console that implements | |
2178 | read() will be picked. | |
2179 | ||
84c08fd6 JW |
2180 | kgdbwait [KGDB] Stop kernel execution and enter the |
2181 | kernel debugger at the earliest opportunity. | |
2182 | ||
497de97e | 2183 | kmac= [MIPS] Korina ethernet MAC address. |
9bed90c6 FF |
2184 | Configure the RouterBoard 532 series on-chip |
2185 | Ethernet adapter MAC address. | |
2186 | ||
04f70336 CM |
2187 | kmemleak= [KNL] Boot-time kmemleak enable/disable |
2188 | Valid arguments: on, off | |
2189 | Default: on | |
47aeeddc MI |
2190 | Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y, |
2191 | the default is off. | |
04f70336 | 2192 | |
970988e1 MH |
2193 | kprobe_event=[probe-list] |
2194 | [FTRACE] Add kprobe events and enable at boot time. | |
2195 | The probe-list is a semicolon delimited list of probe | |
2196 | definitions. Each definition is same as kprobe_events | |
2197 | interface, but the parameters are comma delimited. | |
2198 | For example, to add a kprobe event on vfs_read with | |
2199 | arg1 and arg2, add to the command line; | |
2200 | ||
2201 | kprobe_event=p,vfs_read,$arg1,$arg2 | |
2202 | ||
2203 | See also Documentation/trace/kprobetrace.rst "Kernel | |
2204 | Boot Parameter" section. | |
2205 | ||
de190555 JL |
2206 | kpti= [ARM64] Control page table isolation of user |
2207 | and kernel address spaces. | |
2208 | Default: enabled on cores which need mitigation. | |
2209 | 0: force disabled | |
2210 | 1: force enabled | |
2211 | ||
fef07aae AP |
2212 | kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs. |
2213 | Default is 0 (don't ignore, but inject #GP) | |
2214 | ||
c4ae60e4 LA |
2215 | kvm.enable_vmware_backdoor=[KVM] Support VMware backdoor PV interface. |
2216 | Default is false (don't support). | |
2217 | ||
a182d873 XG |
2218 | kvm.mmu_audit= [KVM] This is a R/W parameter which allows audit |
2219 | KVM MMU at runtime. | |
fef07aae AP |
2220 | Default is 0 (off) |
2221 | ||
b8e8c830 PB |
2222 | kvm.nx_huge_pages= |
2223 | [KVM] Controls the software workaround for the | |
2224 | X86_BUG_ITLB_MULTIHIT bug. | |
2225 | force : Always deploy workaround. | |
2226 | off : Never deploy workaround. | |
2227 | auto : Deploy workaround based on the presence of | |
2228 | X86_BUG_ITLB_MULTIHIT. | |
2229 | ||
2230 | Default is 'auto'. | |
2231 | ||
2232 | If the software workaround is enabled for the host, | |
2233 | guests do need not to enable it for nested guests. | |
2234 | ||
1aa9b957 JS |
2235 | kvm.nx_huge_pages_recovery_ratio= |
2236 | [KVM] Controls how many 4KiB pages are periodically zapped | |
2237 | back to huge pages. 0 disables the recovery, otherwise if | |
2238 | the value is N KVM will zap 1/Nth of the 4KiB pages every | |
2239 | minute. The default is 60. | |
2240 | ||
fef07aae | 2241 | kvm-amd.nested= [KVM,AMD] Allow nested virtualization in KVM/SVM. |
8475f94a | 2242 | Default is 1 (enabled) |
fef07aae AP |
2243 | |
2244 | kvm-amd.npt= [KVM,AMD] Disable nested paging (virtualized MMU) | |
2245 | for all guests. | |
16290246 | 2246 | Default is 1 (enabled) if in 64-bit or 32-bit PAE mode. |
fef07aae | 2247 | |
e23f62f7 MZ |
2248 | kvm-arm.vgic_v3_group0_trap= |
2249 | [KVM,ARM] Trap guest accesses to GICv3 group-0 | |
2250 | system registers | |
2251 | ||
182936ee MZ |
2252 | kvm-arm.vgic_v3_group1_trap= |
2253 | [KVM,ARM] Trap guest accesses to GICv3 group-1 | |
2254 | system registers | |
2255 | ||
ff89511e MZ |
2256 | kvm-arm.vgic_v3_common_trap= |
2257 | [KVM,ARM] Trap guest accesses to GICv3 common | |
2258 | system registers | |
2259 | ||
a7546054 MZ |
2260 | kvm-arm.vgic_v4_enable= |
2261 | [KVM,ARM] Allow use of GICv4 for direct injection of | |
2262 | LPIs. | |
2263 | ||
aed26eeb SR |
2264 | kvm_cma_resv_ratio=n [PPC] |
2265 | Reserves given percentage from system memory area for | |
2266 | contiguous memory allocation for KVM hash pagetable | |
2267 | allocation. | |
2268 | By default it reserves 5% of total system memory. | |
2269 | Format: <integer> | |
2270 | Default: 5 | |
2271 | ||
fef07aae AP |
2272 | kvm-intel.ept= [KVM,Intel] Disable extended page tables |
2273 | (virtualized MMU) support on capable Intel chips. | |
2274 | Default is 1 (enabled) | |
2275 | ||
2276 | kvm-intel.emulate_invalid_guest_state= | |
2277 | [KVM,Intel] Enable emulation of invalid guest states | |
2278 | Default is 0 (disabled) | |
2279 | ||
2280 | kvm-intel.flexpriority= | |
2281 | [KVM,Intel] Disable FlexPriority feature (TPR shadow). | |
2282 | Default is 1 (enabled) | |
2283 | ||
e1a72ae2 SL |
2284 | kvm-intel.nested= |
2285 | [KVM,Intel] Enable VMX nesting (nVMX). | |
2286 | Default is 0 (disabled) | |
2287 | ||
fef07aae AP |
2288 | kvm-intel.unrestricted_guest= |
2289 | [KVM,Intel] Disable unrestricted guest feature | |
2290 | (virtualized real and unpaged mode) on capable | |
2291 | Intel chips. Default is 1 (enabled) | |
2292 | ||
a399477e KRW |
2293 | kvm-intel.vmentry_l1d_flush=[KVM,Intel] Mitigation for L1 Terminal Fault |
2294 | CVE-2018-3620. | |
2295 | ||
2296 | Valid arguments: never, cond, always | |
2297 | ||
2298 | always: L1D cache flush on every VMENTER. | |
2299 | cond: Flush L1D on VMENTER only when the code between | |
2300 | VMEXIT and VMENTER can leak host memory. | |
2301 | never: Disables the mitigation | |
2302 | ||
2303 | Default is cond (do L1 cache flush in specific instances) | |
2304 | ||
fef07aae AP |
2305 | kvm-intel.vpid= [KVM,Intel] Disable Virtual Processor Identification |
2306 | feature (tagged TLBs) on capable Intel chips. | |
2307 | Default is 1 (enabled) | |
2308 | ||
d90a7a0e JK |
2309 | l1tf= [X86] Control mitigation of the L1TF vulnerability on |
2310 | affected CPUs | |
2311 | ||
2312 | The kernel PTE inversion protection is unconditionally | |
2313 | enabled and cannot be disabled. | |
2314 | ||
2315 | full | |
2316 | Provides all available mitigations for the | |
2317 | L1TF vulnerability. Disables SMT and | |
2318 | enables all mitigations in the | |
2319 | hypervisors, i.e. unconditional L1D flush. | |
2320 | ||
2321 | SMT control and L1D flush control via the | |
2322 | sysfs interface is still possible after | |
2323 | boot. Hypervisors will issue a warning | |
2324 | when the first VM is started in a | |
2325 | potentially insecure configuration, | |
2326 | i.e. SMT enabled or L1D flush disabled. | |
2327 | ||
2328 | full,force | |
2329 | Same as 'full', but disables SMT and L1D | |
2330 | flush runtime control. Implies the | |
2331 | 'nosmt=force' command line option. | |
2332 | (i.e. sysfs control of SMT is disabled.) | |
2333 | ||
2334 | flush | |
2335 | Leaves SMT enabled and enables the default | |
2336 | hypervisor mitigation, i.e. conditional | |
2337 | L1D flush. | |
2338 | ||
2339 | SMT control and L1D flush control via the | |
2340 | sysfs interface is still possible after | |
2341 | boot. Hypervisors will issue a warning | |
2342 | when the first VM is started in a | |
2343 | potentially insecure configuration, | |
2344 | i.e. SMT enabled or L1D flush disabled. | |
2345 | ||
2346 | flush,nosmt | |
2347 | ||
2348 | Disables SMT and enables the default | |
2349 | hypervisor mitigation. | |
2350 | ||
2351 | SMT control and L1D flush control via the | |
2352 | sysfs interface is still possible after | |
2353 | boot. Hypervisors will issue a warning | |
2354 | when the first VM is started in a | |
2355 | potentially insecure configuration, | |
2356 | i.e. SMT enabled or L1D flush disabled. | |
2357 | ||
2358 | flush,nowarn | |
2359 | Same as 'flush', but hypervisors will not | |
2360 | warn when a VM is started in a potentially | |
2361 | insecure configuration. | |
2362 | ||
2363 | off | |
2364 | Disables hypervisor mitigations and doesn't | |
2365 | emit any warnings. | |
5b5e4d62 MH |
2366 | It also drops the swap size and available |
2367 | RAM limit restriction on both hypervisor and | |
2368 | bare metal. | |
d90a7a0e JK |
2369 | |
2370 | Default is 'flush'. | |
2371 | ||
65fd4cb6 | 2372 | For details see: Documentation/admin-guide/hw-vuln/l1tf.rst |
d90a7a0e | 2373 | |
1da177e4 LT |
2374 | l2cr= [PPC] |
2375 | ||
a78bfbfc RB |
2376 | l3cr= [PPC] |
2377 | ||
cd4f0ef7 | 2378 | lapic [X86-32,APIC] Enable the local APIC even if BIOS |
a9913044 | 2379 | disabled it. |
1da177e4 | 2380 | |
622381e6 | 2381 | lapic= [X86,APIC] Do not use TSC deadline |
279f1461 SS |
2382 | value for LAPIC timer one-shot implementation. Default |
2383 | back to the programmable timer unit in the LAPIC. | |
622381e6 | 2384 | Format: notscdeadline |
279f1461 | 2385 | |
6cececfc | 2386 | lapic_timer_c2_ok [X86,APIC] trust the local apic timer |
ada9cfdd | 2387 | in C2 power state. |
e585bef8 | 2388 | |
fcb71f6f FC |
2389 | libata.dma= [LIBATA] DMA control |
2390 | libata.dma=0 Disable all PATA and SATA DMA | |
2391 | libata.dma=1 PATA and SATA Disk DMA only | |
2392 | libata.dma=2 ATAPI (CDROM) DMA only | |
16290246 | 2393 | libata.dma=4 Compact Flash DMA only |
fcb71f6f FC |
2394 | Combinations also work, so libata.dma=3 enables DMA |
2395 | for disks and CDROMs, but not CFs. | |
16290246 | 2396 | |
20308871 MP |
2397 | libata.ignore_hpa= [LIBATA] Ignore HPA limit |
2398 | libata.ignore_hpa=0 keep BIOS limits (default) | |
2399 | libata.ignore_hpa=1 ignore limits, using full disk | |
fcb71f6f | 2400 | |
78e70c23 DJ |
2401 | libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume |
2402 | when set. | |
2403 | Format: <int> | |
2404 | ||
33267325 TH |
2405 | libata.force= [LIBATA] Force configurations. The format is comma |
2406 | separated list of "[ID:]VAL" where ID is | |
4c44f309 | 2407 | PORT[.DEVICE]. PORT and DEVICE are decimal numbers |
33267325 TH |
2408 | matching port, link or device. Basically, it matches |
2409 | the ATA ID string printed on console by libata. If | |
2410 | the whole ID part is omitted, the last PORT and DEVICE | |
2411 | values are used. If ID hasn't been specified yet, the | |
2412 | configuration applies to all ports, links and devices. | |
2413 | ||
2414 | If only DEVICE is omitted, the parameter applies to | |
2415 | the port and all links and devices behind it. DEVICE | |
2416 | number of 0 either selects the first device or the | |
2417 | first fan-out link behind PMP device. It does not | |
2418 | select the host link. DEVICE number of 15 selects the | |
2419 | host link and device attached to it. | |
2420 | ||
2421 | The VAL specifies the configuration to force. As long | |
2422 | as there's no ambiguity shortcut notation is allowed. | |
2423 | For example, both 1.5 and 1.5G would work for 1.5Gbps. | |
2424 | The following configurations can be forced. | |
2425 | ||
2426 | * Cable type: 40c, 80c, short40c, unk, ign or sata. | |
2427 | Any ID with matching PORT is used. | |
2428 | ||
2429 | * SATA link speed limit: 1.5Gbps or 3.0Gbps. | |
2430 | ||
2431 | * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7]. | |
2432 | udma[/][16,25,33,44,66,100,133] notation is also | |
2433 | allowed. | |
2434 | ||
2435 | * [no]ncq: Turn on or off NCQ. | |
2436 | ||
d7b16e4f MP |
2437 | * [no]ncqtrim: Turn off queued DSM TRIM. |
2438 | ||
05944bdf | 2439 | * nohrst, nosrst, norst: suppress hard, soft |
6dddd7a7 | 2440 | and both resets. |
05944bdf | 2441 | |
ca6d43b0 DW |
2442 | * rstonce: only attempt one reset during |
2443 | hot-unplug link recovery | |
2444 | ||
43c9c591 TH |
2445 | * dump_id: dump IDENTIFY data. |
2446 | ||
966fbe19 VP |
2447 | * atapi_dmadir: Enable ATAPI DMADIR bridge support |
2448 | ||
b8bd6dc3 RJ |
2449 | * disable: Disable this device. |
2450 | ||
33267325 TH |
2451 | If there are multiple matching configurations changing |
2452 | the same attribute, the last one is used. | |
2453 | ||
95f72d1e | 2454 | memblock=debug [KNL] Enable memblock debug messages. |
7c4be253 | 2455 | |
6b99e6e6 | 2456 | load_ramdisk= [RAM] [Deprecated] |
1da177e4 | 2457 | |
a6b25b67 RD |
2458 | lockd.nlm_grace_period=P [NFS] Assign grace period. |
2459 | Format: <integer> | |
1da177e4 | 2460 | |
a6b25b67 RD |
2461 | lockd.nlm_tcpport=N [NFS] Assign TCP port. |
2462 | Format: <integer> | |
2463 | ||
2464 | lockd.nlm_timeout=T [NFS] Assign timeout value. | |
2465 | Format: <integer> | |
2466 | ||
2467 | lockd.nlm_udpport=M [NFS] Assign UDP port. | |
2468 | Format: <integer> | |
1da177e4 | 2469 | |
000d388e MG |
2470 | lockdown= [SECURITY] |
2471 | { integrity | confidentiality } | |
2472 | Enable the kernel lockdown feature. If set to | |
2473 | integrity, kernel features that allow userland to | |
2474 | modify the running kernel are disabled. If set to | |
2475 | confidentiality, kernel features that allow userland | |
2476 | to extract confidential information from the kernel | |
2477 | are also disabled. | |
2478 | ||
ec4518aa PM |
2479 | locktorture.nreaders_stress= [KNL] |
2480 | Set the number of locking read-acquisition kthreads. | |
2481 | Defaults to being automatically set based on the | |
2482 | number of online CPUs. | |
2483 | ||
2484 | locktorture.nwriters_stress= [KNL] | |
2485 | Set the number of locking write-acquisition kthreads. | |
2486 | ||
2487 | locktorture.onoff_holdoff= [KNL] | |
2488 | Set time (s) after boot for CPU-hotplug testing. | |
2489 | ||
2490 | locktorture.onoff_interval= [KNL] | |
2491 | Set time (s) between CPU-hotplug operations, or | |
2492 | zero to disable CPU-hotplug testing. | |
2493 | ||
2494 | locktorture.shuffle_interval= [KNL] | |
2495 | Set task-shuffle interval (jiffies). Shuffling | |
2496 | tasks allows some CPUs to go into dyntick-idle | |
2497 | mode during the locktorture test. | |
2498 | ||
2499 | locktorture.shutdown_secs= [KNL] | |
2500 | Set time (s) after boot system shutdown. This | |
2501 | is useful for hands-off automated testing. | |
2502 | ||
2503 | locktorture.stat_interval= [KNL] | |
2504 | Time (s) between statistics printk()s. | |
2505 | ||
2506 | locktorture.stutter= [KNL] | |
2507 | Time (s) to stutter testing, for example, | |
2508 | specifying five seconds causes the test to run for | |
2509 | five seconds, wait for five seconds, and so on. | |
2510 | This tests the locking primitive's ability to | |
2511 | transition abruptly to and from idle. | |
2512 | ||
ec4518aa PM |
2513 | locktorture.torture_type= [KNL] |
2514 | Specify the locking implementation to test. | |
2515 | ||
2516 | locktorture.verbose= [KNL] | |
2517 | Enable additional printk() statements. | |
2518 | ||
1da177e4 LT |
2519 | logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver |
2520 | Format: <irq> | |
2521 | ||
2522 | loglevel= All Kernel Messages with a loglevel smaller than the | |
2523 | console loglevel will be printed to the console. It can | |
2524 | also be changed with klogd or other programs. The | |
2525 | loglevels are defined as follows: | |
2526 | ||
2527 | 0 (KERN_EMERG) system is unusable | |
2528 | 1 (KERN_ALERT) action must be taken immediately | |
2529 | 2 (KERN_CRIT) critical conditions | |
2530 | 3 (KERN_ERR) error conditions | |
2531 | 4 (KERN_WARNING) warning conditions | |
2532 | 5 (KERN_NOTICE) normal but significant condition | |
2533 | 6 (KERN_INFO) informational | |
2534 | 7 (KERN_DEBUG) debug-level messages | |
2535 | ||
c756d08a | 2536 | log_buf_len=n[KMG] Sets the size of the printk ring buffer, |
23b2899f LR |
2537 | in bytes. n must be a power of two and greater |
2538 | than the minimal size. The minimal size is defined | |
2539 | by LOG_BUF_SHIFT kernel config parameter. There is | |
2540 | also CONFIG_LOG_CPU_MAX_BUF_SHIFT config parameter | |
2541 | that allows to increase the default size depending on | |
2542 | the number of CPUs. See init/Kconfig for more details. | |
1da177e4 | 2543 | |
accaa24c RD |
2544 | logo.nologo [FB] Disables display of the built-in Linux logo. |
2545 | This may be used to provide more screen space for | |
2546 | kernel log messages and is useful when debugging | |
2547 | kernel boot problems. | |
2548 | ||
1da177e4 LT |
2549 | lp=0 [LP] Specify parallel ports to use, e.g, |
2550 | lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses | |
2551 | lp=reset first parallel port). 'lp=0' disables the | |
2552 | lp=auto printer driver. 'lp=reset' (which can be | |
2553 | specified in addition to the ports) causes | |
2554 | attached printers to be reset. Using | |
2555 | lp=port1,port2,... specifies the parallel ports | |
2556 | to associate lp devices with, starting with | |
2557 | lp0. A port specification may be 'none' to skip | |
2558 | that lp device, or a parport name such as | |
2559 | 'parport0'. Specifying 'lp=auto' instead of a | |
2560 | port specification list means that device IDs | |
2561 | from each port should be examined, to see if | |
2562 | an IEEE 1284-compliant printer is attached; if | |
2563 | so, the driver will manage that printer. | |
2564 | See also header of drivers/char/lp.c. | |
2565 | ||
2566 | lpj=n [KNL] | |
2567 | Sets loops_per_jiffy to given constant, thus avoiding | |
2568 | time-consuming boot-time autodetection (up to 250 ms per | |
2569 | CPU). 0 enables autodetection (default). To determine | |
2570 | the correct value for your kernel, boot with normal | |
2571 | autodetection and see what value is printed. Note that | |
2572 | on SMP systems the preset will be applied to all CPUs, | |
2573 | which is likely to cause problems if your CPUs need | |
2574 | significantly divergent settings. An incorrect value | |
2575 | will cause delays in the kernel to be wrong, leading to | |
2576 | unpredictable I/O errors and other breakage. Although | |
2577 | unlikely, in the extreme case this might damage your | |
2578 | hardware. | |
2579 | ||
2580 | ltpc= [NET] | |
2581 | Format: <io>,<irq>,<dma> | |
2582 | ||
9b8c7c14 KC |
2583 | lsm.debug [SECURITY] Enable LSM initialization debugging output. |
2584 | ||
79f7865d KC |
2585 | lsm=lsm1,...,lsmN |
2586 | [SECURITY] Choose order of LSM initialization. This | |
89a9684e | 2587 | overrides CONFIG_LSM, and the "security=" parameter. |
79f7865d | 2588 | |
16290246 | 2589 | machvec= [IA-64] Force the use of a particular machine-vector |
a9913044 | 2590 | (machvec) in a generic kernel. |
df43acac | 2591 | Example: machvec=hpzx1 |
1da177e4 | 2592 | |
42769488 RD |
2593 | machtype= [Loongson] Share the same kernel image file between |
2594 | different yeeloong laptops. | |
3209e70e WZ |
2595 | Example: machtype=lemote-yeeloong-2f-7inch |
2596 | ||
0cb55ad2 RD |
2597 | max_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory greater |
2598 | than or equal to this physical address is ignored. | |
1da177e4 LT |
2599 | |
2600 | maxcpus= [SMP] Maximum number of processors that an SMP kernel | |
7c142bfe BH |
2601 | will bring up during bootup. maxcpus=n : n >= 0 limits |
2602 | the kernel to bring up 'n' processors. Surely after | |
2603 | bootup you can bring up the other plugged cpu by executing | |
2604 | "echo 1 > /sys/devices/system/cpu/cpuX/online". So maxcpus | |
2605 | only takes effect during system bootup. | |
2606 | While n=0 is a special case, it is equivalent to "nosmp", | |
2607 | which also disables the IO APIC. | |
1da177e4 | 2608 | |
d134b00b KS |
2609 | max_loop= [LOOP] The number of loop block devices that get |
2610 | (loop.max_loop) unconditionally pre-created at init time. The default | |
2611 | number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead | |
2612 | of statically allocating a predefined number, loop | |
2613 | devices can be requested on-demand with the | |
2614 | /dev/loop-control interface. | |
2b2c3750 | 2615 | |
cd4f0ef7 | 2616 | mce [X86-32] Machine Check Exception |
1da177e4 | 2617 | |
cb1aaebe | 2618 | mce=option [X86-64] See Documentation/x86/x86_64/boot-options.rst |
909dd324 | 2619 | |
1da177e4 | 2620 | md= [HW] RAID subsystems devices and level |
e52347bd | 2621 | See Documentation/admin-guide/md.rst. |
a9913044 | 2622 | |
1da177e4 LT |
2623 | mdacon= [MDA] |
2624 | Format: <first>,<last> | |
2625 | Specifies range of consoles to be captured by the MDA. | |
a9913044 | 2626 | |
bc124170 TG |
2627 | mds= [X86,INTEL] |
2628 | Control mitigation for the Micro-architectural Data | |
2629 | Sampling (MDS) vulnerability. | |
2630 | ||
2631 | Certain CPUs are vulnerable to an exploit against CPU | |
2632 | internal buffers which can forward information to a | |
2633 | disclosure gadget under certain conditions. | |
2634 | ||
2635 | In vulnerable processors, the speculatively | |
2636 | forwarded data can be used in a cache side channel | |
2637 | attack, to access data to which the attacker does | |
2638 | not have direct access. | |
2639 | ||
2640 | This parameter controls the MDS mitigation. The | |
2641 | options are: | |
2642 | ||
d71eb0ce JP |
2643 | full - Enable MDS mitigation on vulnerable CPUs |
2644 | full,nosmt - Enable MDS mitigation and disable | |
2645 | SMT on vulnerable CPUs | |
2646 | off - Unconditionally disable MDS mitigation | |
bc124170 | 2647 | |
64870ed1 WL |
2648 | On TAA-affected machines, mds=off can be prevented by |
2649 | an active TAA mitigation as both vulnerabilities are | |
2650 | mitigated with the same mechanism so in order to disable | |
2651 | this mitigation, you need to specify tsx_async_abort=off | |
2652 | too. | |
2653 | ||
bc124170 TG |
2654 | Not specifying this option is equivalent to |
2655 | mds=full. | |
2656 | ||
5999bbe7 TG |
2657 | For details see: Documentation/admin-guide/hw-vuln/mds.rst |
2658 | ||
1da177e4 | 2659 | mem=nn[KMG] [KNL,BOOT] Force usage of a specific amount of memory |
f3cd4c86 BH |
2660 | Amount of memory to be used in cases as follows: |
2661 | ||
2662 | 1 for test; | |
2663 | 2 when the kernel is not able to see the whole system memory; | |
2664 | 3 memory that lies after 'mem=' boundary is excluded from | |
2665 | the hypervisor, then assigned to KVM guests. | |
2666 | ||
fbb97d87 WC |
2667 | [X86] Work as limiting max address. Use together |
2668 | with memmap= to avoid physical address space collisions. | |
2669 | Without memmap= PCI devices could be placed at addresses | |
2670 | belonging to unused RAM. | |
1da177e4 | 2671 | |
f3cd4c86 BH |
2672 | Note that this only takes effects during boot time since |
2673 | in above case 3, memory may need be hot added after boot | |
2674 | if system memory of hypervisor is not sufficient. | |
2675 | ||
cd4f0ef7 | 2676 | mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel |
1da177e4 LT |
2677 | memory. |
2678 | ||
6902aa84 PM |
2679 | memchunk=nn[KMG] |
2680 | [KNL,SH] Allow user to override the default size for | |
2681 | per-device physically contiguous DMA buffers. | |
2682 | ||
6dddd7a7 | 2683 | memhp_default_state=online/offline |
86dd995d VK |
2684 | [KNL] Set the initial state for the memory hotplug |
2685 | onlining policy. If not specified, the default value is | |
2686 | set according to the | |
2687 | CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE kernel config | |
2688 | option. | |
cb1aaebe | 2689 | See Documentation/admin-guide/mm/memory-hotplug.rst. |
86dd995d | 2690 | |
6cececfc | 2691 | memmap=exactmap [KNL,X86] Enable setting of an exact |
1da177e4 LT |
2692 | E820 memory map, as specified by the user. |
2693 | Such memmap=exactmap lines can be constructed based on | |
2694 | BIOS output or other requirements. See the memmap=nn@ss | |
2695 | option description. | |
2696 | ||
2697 | memmap=nn[KMG]@ss[KMG] | |
277cba1d RD |
2698 | [KNL] Force usage of a specific region of memory. |
2699 | Region of memory to be used is from ss to ss+nn. | |
8fcc9bc3 BH |
2700 | If @ss[KMG] is omitted, it is equivalent to mem=nn[KMG], |
2701 | which limits max address to nn[KMG]. | |
2702 | Multiple different regions can be specified, | |
2703 | comma delimited. | |
2704 | Example: | |
2705 | memmap=100M@2G,100M#3G,1G!1024G | |
1da177e4 LT |
2706 | |
2707 | memmap=nn[KMG]#ss[KMG] | |
2708 | [KNL,ACPI] Mark specific memory as ACPI data. | |
277cba1d | 2709 | Region of memory to be marked is from ss to ss+nn. |
1da177e4 LT |
2710 | |
2711 | memmap=nn[KMG]$ss[KMG] | |
2712 | [KNL,ACPI] Mark specific memory as reserved. | |
277cba1d | 2713 | Region of memory to be reserved is from ss to ss+nn. |
1312848e PM |
2714 | Example: Exclude memory from 0x18690000-0x1869ffff |
2715 | memmap=64K$0x18690000 | |
2716 | or | |
2717 | memmap=0x10000$0x18690000 | |
8fcc9bc3 BH |
2718 | Some bootloaders may need an escape character before '$', |
2719 | like Grub2, otherwise '$' and the following number | |
2720 | will be eaten. | |
1da177e4 | 2721 | |
ec776ef6 CH |
2722 | memmap=nn[KMG]!ss[KMG] |
2723 | [KNL,X86] Mark specific memory as protected. | |
2724 | Region of memory to be used, from ss to ss+nn. | |
2725 | The memory region may be marked as e820 type 12 (0xc) | |
2726 | and is NVDIMM or ADR memory. | |
2727 | ||
ef61f8a3 JS |
2728 | memmap=<size>%<offset>-<oldtype>+<newtype> |
2729 | [KNL,ACPI] Convert memory within the specified region | |
2730 | from <oldtype> to <newtype>. If "-<oldtype>" is left | |
2731 | out, the whole region will be marked as <newtype>, | |
2732 | even if previously unavailable. If "+<newtype>" is left | |
2733 | out, matching memory will be removed. Types are | |
2734 | specified as e820 types, e.g., 1 = RAM, 2 = reserved, | |
2735 | 3 = ACPI, 12 = PRAM. | |
2736 | ||
9f077871 JF |
2737 | memory_corruption_check=0/1 [X86] |
2738 | Some BIOSes seem to corrupt the first 64k of | |
2739 | memory when doing things like suspend/resume. | |
2740 | Setting this option will scan the memory | |
2741 | looking for corruption. Enabling this will | |
2742 | both detect corruption and prevent the kernel | |
2743 | from using the memory being corrupted. | |
2744 | However, its intended as a diagnostic tool; if | |
2745 | repeatable BIOS-originated corruption always | |
2746 | affects the same memory, you can use memmap= | |
2747 | to prevent the kernel from using that memory. | |
2748 | ||
2749 | memory_corruption_check_size=size [X86] | |
2750 | By default it checks for corruption in the low | |
2751 | 64k, making this memory unavailable for normal | |
2752 | use. Use this parameter to scan for | |
2753 | corruption in more or less memory. | |
2754 | ||
2755 | memory_corruption_check_period=seconds [X86] | |
2756 | By default it checks for corruption every 60 | |
2757 | seconds. Use this parameter to check at some | |
2758 | other rate. 0 disables periodic checking. | |
2759 | ||
d90fe2ac | 2760 | memtest= [KNL,X86,ARM,PPC] Enable memtest |
c64df707 | 2761 | Format: <integer> |
c64df707 | 2762 | default : 0 <disable> |
9e5f6cf5 AH |
2763 | Specifies the number of memtest passes to be |
2764 | performed. Each pass selects another test | |
2765 | pattern from a given set of patterns. Memtest | |
2766 | fills the memory with this pattern, validates | |
2767 | memory contents and reserves bad memory | |
2768 | regions that are detected. | |
c64df707 | 2769 | |
c262f3b9 TL |
2770 | mem_encrypt= [X86-64] AMD Secure Memory Encryption (SME) control |
2771 | Valid arguments: on, off | |
2772 | Default (depends on kernel configuration option): | |
2773 | on (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=y) | |
2774 | off (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=n) | |
2775 | mem_encrypt=on: Activate SME | |
2776 | mem_encrypt=off: Do not activate SME | |
2777 | ||
2f5947df | 2778 | Refer to Documentation/virt/kvm/amd-memory-encryption.rst |
c262f3b9 TL |
2779 | for details on when memory encryption can be activated. |
2780 | ||
406e7938 RW |
2781 | mem_sleep_default= [SUSPEND] Default system suspend mode: |
2782 | s2idle - Suspend-To-Idle | |
2783 | shallow - Power-On Suspend or equivalent (if supported) | |
2784 | deep - Suspend-To-RAM or equivalent (if supported) | |
58e7cb9e | 2785 | See Documentation/admin-guide/pm/sleep-states.rst. |
406e7938 | 2786 | |
1da177e4 | 2787 | meye.*= [HW] Set MotionEye Camera parameters |
32e2eae2 | 2788 | See Documentation/admin-guide/media/meye.rst. |
1da177e4 | 2789 | |
8f36881b AS |
2790 | mfgpt_irq= [IA-32] Specify the IRQ to use for the |
2791 | Multi-Function General Purpose Timers on AMD Geode | |
2792 | platforms. | |
2793 | ||
e6c4dc6c WT |
2794 | mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when |
2795 | the BIOS has incorrectly applied a workaround. TinyBIOS | |
2796 | version 0.98 is known to be affected, 0.99 fixes the | |
2797 | problem by letting the user disable the workaround. | |
2798 | ||
1da177e4 LT |
2799 | mga= [HW,DRM] |
2800 | ||
1c207f95 RD |
2801 | min_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory below this |
2802 | physical address is ignored. | |
2803 | ||
39f45d7b MP |
2804 | mini2440= [ARM,HW,KNL] |
2805 | Format:[0..2][b][c][t] | |
2806 | Default: "0tb" | |
2807 | MINI2440 configuration specification: | |
2808 | 0 - The attached screen is the 3.5" TFT | |
2809 | 1 - The attached screen is the 7" TFT | |
2810 | 2 - The VGA Shield is attached (1024x768) | |
2811 | Leaving out the screen size parameter will not load | |
2812 | the TFT driver, and the framebuffer will be left | |
2813 | unconfigured. | |
2814 | b - Enable backlight. The TFT backlight pin will be | |
2815 | linked to the kernel VESA blanking code and a GPIO | |
2816 | LED. This parameter is not necessary when using the | |
2817 | VGA shield. | |
2818 | c - Enable the s3c camera interface. | |
2819 | t - Reserved for enabling touchscreen support. The | |
2820 | touchscreen support is not enabled in the mainstream | |
2821 | kernel as of 2.6.30, a preliminary port can be found | |
2822 | in the "bleeding edge" mini2440 support kernel at | |
6b2484e1 | 2823 | https://repo.or.cz/w/linux-2.6/mini2440.git |
39f45d7b | 2824 | |
98af8452 | 2825 | mitigations= |
a111b7c0 JP |
2826 | [X86,PPC,S390,ARM64] Control optional mitigations for |
2827 | CPU vulnerabilities. This is a set of curated, | |
d68be4c4 JP |
2828 | arch-independent options, each of which is an |
2829 | aggregation of existing arch-specific options. | |
98af8452 JP |
2830 | |
2831 | off | |
2832 | Disable all optional CPU mitigations. This | |
2833 | improves system performance, but it may also | |
2834 | expose users to several CPU vulnerabilities. | |
782e69ef | 2835 | Equivalent to: nopti [X86,PPC] |
a111b7c0 | 2836 | kpti=0 [ARM64] |
a2059825 | 2837 | nospectre_v1 [X86,PPC] |
0336e04a | 2838 | nobp=0 [S390] |
a111b7c0 | 2839 | nospectre_v2 [X86,PPC,S390,ARM64] |
d68be4c4 | 2840 | spectre_v2_user=off [X86] |
782e69ef | 2841 | spec_store_bypass_disable=off [X86,PPC] |
a111b7c0 | 2842 | ssbd=force-off [ARM64] |
d68be4c4 | 2843 | l1tf=off [X86] |
5c14068f | 2844 | mds=off [X86] |
a7a248c5 | 2845 | tsx_async_abort=off [X86] |
b8e8c830 PB |
2846 | kvm.nx_huge_pages=off [X86] |
2847 | ||
2848 | Exceptions: | |
2849 | This does not have any effect on | |
2850 | kvm.nx_huge_pages when | |
2851 | kvm.nx_huge_pages=force. | |
98af8452 JP |
2852 | |
2853 | auto (default) | |
2854 | Mitigate all CPU vulnerabilities, but leave SMT | |
2855 | enabled, even if it's vulnerable. This is for | |
2856 | users who don't want to be surprised by SMT | |
2857 | getting disabled across kernel upgrades, or who | |
2858 | have other ways of avoiding SMT-based attacks. | |
d68be4c4 | 2859 | Equivalent to: (default behavior) |
98af8452 JP |
2860 | |
2861 | auto,nosmt | |
2862 | Mitigate all CPU vulnerabilities, disabling SMT | |
2863 | if needed. This is for users who always want to | |
2864 | be fully mitigated, even if it means losing SMT. | |
d68be4c4 | 2865 | Equivalent to: l1tf=flush,nosmt [X86] |
5c14068f | 2866 | mds=full,nosmt [X86] |
a7a248c5 | 2867 | tsx_async_abort=full,nosmt [X86] |
98af8452 | 2868 | |
6b74ab97 MG |
2869 | mminit_loglevel= |
2870 | [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this | |
2871 | parameter allows control of the logging verbosity for | |
2872 | the additional memory initialisation checks. A value | |
2873 | of 0 disables mminit logging and a level of 4 will | |
2874 | log everything. Information is printed at KERN_DEBUG | |
2875 | so loglevel=8 may also need to be specified. | |
2876 | ||
106a4ee2 RR |
2877 | module.sig_enforce |
2878 | [KNL] When CONFIG_MODULE_SIG is set, this means that | |
2879 | modules without (valid) signatures will fail to load. | |
2a039be7 | 2880 | Note that if CONFIG_MODULE_SIG_FORCE is set, that |
106a4ee2 RR |
2881 | is always true, so this option does nothing. |
2882 | ||
be7de5f9 PB |
2883 | module_blacklist= [KNL] Do not load a comma-separated list of |
2884 | modules. Useful for debugging problem modules. | |
2885 | ||
1da177e4 LT |
2886 | mousedev.tap_time= |
2887 | [MOUSE] Maximum time between finger touching and | |
2888 | leaving touchpad surface for touch to be considered | |
2889 | a tap and be reported as a left button click (for | |
2890 | touchpads working in absolute mode only). | |
2891 | Format: <msecs> | |
2892 | mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices | |
2893 | reporting absolute coordinates, such as tablets | |
2894 | mousedev.yres= [MOUSE] Vertical screen resolution, used for devices | |
2895 | reporting absolute coordinates, such as tablets | |
2896 | ||
a5c6d650 DR |
2897 | movablecore= [KNL,X86,IA-64,PPC] |
2898 | Format: nn[KMGTPE] | nn% | |
2899 | This parameter is the complement to kernelcore=, it | |
2900 | specifies the amount of memory used for migratable | |
2901 | allocations. If both kernelcore and movablecore is | |
2902 | specified, then kernelcore will be at *least* the | |
2903 | specified value but may be more. If movablecore on its | |
2904 | own is specified, the administrator must be careful | |
0cb55ad2 RD |
2905 | that the amount of memory usable for all allocations |
2906 | is not too small. | |
2907 | ||
f70029bb MH |
2908 | movable_node [KNL] Boot-time switch to make hotplugable memory |
2909 | NUMA nodes to be movable. This means that the memory | |
2910 | of such nodes will be usable only for movable | |
2911 | allocations which rules out almost all kernel | |
2912 | allocations. Use with caution! | |
c5320926 | 2913 | |
1da177e4 LT |
2914 | MTD_Partition= [MTD] |
2915 | Format: <name>,<region-number>,<size>,<offset> | |
2916 | ||
a9913044 RD |
2917 | MTD_Region= [MTD] Format: |
2918 | <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>] | |
1da177e4 LT |
2919 | |
2920 | mtdparts= [MTD] | |
fb251124 | 2921 | See drivers/mtd/parsers/cmdlinepart.c |
1da177e4 | 2922 | |
4e89a2d8 WS |
2923 | multitce=off [PPC] This parameter disables the use of the pSeries |
2924 | firmware feature for updating multiple TCE entries | |
2925 | at a time. | |
2926 | ||
5988af23 RH |
2927 | onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration |
2928 | ||
2929 | Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock] | |
2930 | ||
2931 | boundary - index of last SLC block on Flex-OneNAND. | |
2932 | The remaining blocks are configured as MLC blocks. | |
2933 | lock - Configure if Flex-OneNAND boundary should be locked. | |
2934 | Once locked, the boundary cannot be changed. | |
2935 | 1 indicates lock status, 0 indicates unlock status. | |
2936 | ||
9db829f4 BD |
2937 | mtdset= [ARM] |
2938 | ARM/S3C2412 JIVE boot control | |
2939 | ||
2940 | See arch/arm/mach-s3c2412/mach-jive.c | |
2941 | ||
1da177e4 | 2942 | mtouchusb.raw_coordinates= |
a9913044 RD |
2943 | [HW] Make the MicroTouch USB driver use raw coordinates |
2944 | ('y', default) or cooked coordinates ('n') | |
1da177e4 | 2945 | |
0cb55ad2 | 2946 | mtrr_chunk_size=nn[KMG] [X86] |
19f59460 | 2947 | used for mtrr cleanup. It is largest continuous chunk |
0cb55ad2 RD |
2948 | that could hold holes aka. UC entries. |
2949 | ||
2950 | mtrr_gran_size=nn[KMG] [X86] | |
2951 | Used for mtrr cleanup. It is granularity of mtrr block. | |
2952 | Default is 1. | |
2953 | Large value could prevent small alignment from | |
2954 | using up MTRRs. | |
2955 | ||
2956 | mtrr_spare_reg_nr=n [X86] | |
2957 | Format: <integer> | |
2958 | Range: 0,7 : spare reg number | |
2959 | Default : 1 | |
2960 | Used for mtrr cleanup. It is spare mtrr entries number. | |
2961 | Set to 2 or more if your graphical card needs more. | |
2962 | ||
1da177e4 LT |
2963 | n2= [NET] SDL Inc. RISCom/N2 synchronous serial card |
2964 | ||
1da177e4 LT |
2965 | netdev= [NET] Network devices parameters |
2966 | Format: <irq>,<io>,<mem_start>,<mem_end>,<name> | |
2967 | Note that mem_start is often overloaded to mean | |
2968 | something different and driver-specific. | |
a9913044 RD |
2969 | This usage is only documented in each driver source |
2970 | file if at all. | |
2971 | ||
58401572 KPO |
2972 | nf_conntrack.acct= |
2973 | [NETFILTER] Enable connection tracking flow accounting | |
2974 | 0 to disable accounting | |
2975 | 1 to enable accounting | |
d70a011d | 2976 | Default value is 0. |
58401572 | 2977 | |
306a0753 | 2978 | nfsaddrs= [NFS] Deprecated. Use ip= instead. |
3eb30c51 | 2979 | See Documentation/admin-guide/nfs/nfsroot.rst. |
1da177e4 LT |
2980 | |
2981 | nfsroot= [NFS] nfs root filesystem for disk-less boxes. | |
3eb30c51 | 2982 | See Documentation/admin-guide/nfs/nfsroot.rst. |
1da177e4 | 2983 | |
306a0753 | 2984 | nfsrootdebug [NFS] enable nfsroot debugging messages. |
3eb30c51 | 2985 | See Documentation/admin-guide/nfs/nfsroot.rst. |
306a0753 | 2986 | |
5405fc44 TM |
2987 | nfs.callback_nr_threads= |
2988 | [NFSv4] set the total number of threads that the | |
2989 | NFS client will assign to service NFSv4 callback | |
2990 | requests. | |
2991 | ||
a72b4422 TM |
2992 | nfs.callback_tcpport= |
2993 | [NFS] set the TCP port on which the NFSv4 callback | |
2994 | channel should listen. | |
2995 | ||
e571cbf1 TM |
2996 | nfs.cache_getent= |
2997 | [NFS] sets the pathname to the program which is used | |
2998 | to update the NFS client cache entries. | |
2999 | ||
3000 | nfs.cache_getent_timeout= | |
3001 | [NFS] sets the timeout after which an attempt to | |
3002 | update a cache entry is deemed to have failed. | |
3003 | ||
58df095b TM |
3004 | nfs.idmap_cache_timeout= |
3005 | [NFS] set the maximum lifetime for idmapper cache | |
3006 | entries. | |
3007 | ||
f43bf0be TM |
3008 | nfs.enable_ino64= |
3009 | [NFS] enable 64-bit inode numbers. | |
3010 | If zero, the NFS client will fake up a 32-bit inode | |
3011 | number for the readdir() and stat() syscalls instead | |
3012 | of returning the full 64-bit number. | |
3013 | The default is to return 64-bit inode numbers. | |
3014 | ||
5405fc44 TM |
3015 | nfs.max_session_cb_slots= |
3016 | [NFSv4.1] Sets the maximum number of session | |
3017 | slots the client will assign to the callback | |
3018 | channel. This determines the maximum number of | |
3019 | callbacks the client will process in parallel for | |
3020 | a particular server. | |
3021 | ||
ef159e91 TM |
3022 | nfs.max_session_slots= |
3023 | [NFSv4.1] Sets the maximum number of session slots | |
3024 | the client will attempt to negotiate with the server. | |
3025 | This limits the number of simultaneous RPC requests | |
3026 | that the client can send to the NFSv4.1 server. | |
3027 | Note that there is little point in setting this | |
3028 | value higher than the max_tcp_slot_table_limit. | |
3029 | ||
b064eca2 | 3030 | nfs.nfs4_disable_idmapping= |
074b1d12 TM |
3031 | [NFSv4] When set to the default of '1', this option |
3032 | ensures that both the RPC level authentication | |
3033 | scheme and the NFS level operations agree to use | |
3034 | numeric uids/gids if the mount is using the | |
3035 | 'sec=sys' security flavour. In effect it is | |
3036 | disabling idmapping, which can make migration from | |
3037 | legacy NFSv2/v3 systems to NFSv4 easier. | |
3038 | Servers that do not support this mode of operation | |
3039 | will be autodetected by the client, and it will fall | |
3040 | back to using the idmapper. | |
3041 | To turn off this behaviour, set the value to '0'. | |
6f2ea7f2 CL |
3042 | nfs.nfs4_unique_id= |
3043 | [NFS4] Specify an additional fixed unique ident- | |
3044 | ification string that NFSv4 clients can insert into | |
3045 | their nfs_client_id4 string. This is typically a | |
3046 | UUID that is generated at system install time. | |
b064eca2 | 3047 | |
db8ac8ba WAA |
3048 | nfs.send_implementation_id = |
3049 | [NFSv4.1] Send client implementation identification | |
3050 | information in exchange_id requests. | |
3051 | If zero, no implementation identification information | |
3052 | will be sent. | |
3053 | The default is to send the implementation identification | |
3054 | information. | |
e52347bd | 3055 | |
f6de7a39 TM |
3056 | nfs.recover_lost_locks = |
3057 | [NFSv4] Attempt to recover locks that were lost due | |
3058 | to a lease timeout on the server. Please note that | |
3059 | doing this risks data corruption, since there are | |
3060 | no guarantees that the file will remain unchanged | |
3061 | after the locks are lost. | |
3062 | If you want to enable the kernel legacy behaviour of | |
3063 | attempting to recover these locks, then set this | |
3064 | parameter to '1'. | |
3065 | The default parameter value of '0' causes the kernel | |
3066 | not to attempt recovery of lost locks. | |
db8ac8ba | 3067 | |
bbf58bf3 TM |
3068 | nfs4.layoutstats_timer = |
3069 | [NFSv4.2] Change the rate at which the kernel sends | |
3070 | layoutstats to the pNFS metadata server. | |
3071 | ||
3072 | Setting this to value to 0 causes the kernel to use | |
3073 | whatever value is the default set by the layout | |
3074 | driver. A non-zero value sets the minimum interval | |
3075 | in seconds between layoutstats transmissions. | |
3076 | ||
e9541ce8 BF |
3077 | nfsd.nfs4_disable_idmapping= |
3078 | [NFSv4] When set to the default of '1', the NFSv4 | |
3079 | server will return only numeric uids and gids to | |
3080 | clients using auth_sys, and will accept numeric uids | |
3081 | and gids from such clients. This is intended to ease | |
3082 | migration from NFSv2/v3. | |
db8ac8ba | 3083 | |
c0c74acb | 3084 | nmi_debug= [KNL,SH] Specify one or more actions to take |
e7ba176b HS |
3085 | when a NMI is triggered. |
3086 | Format: [state][,regs][,debounce][,die] | |
3087 | ||
6cececfc | 3088 | nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels |
fef2c9bc | 3089 | Format: [panic,][nopanic,][num] |
195daf66 | 3090 | Valid num: 0 or 1 |
334bb79c PK |
3091 | 0 - turn hardlockup detector in nmi_watchdog off |
3092 | 1 - turn hardlockup detector in nmi_watchdog on | |
0cb55ad2 | 3093 | When panic is specified, panic when an NMI watchdog |
93285c01 ZD |
3094 | timeout occurs (or 'nopanic' to not panic on an NMI |
3095 | watchdog, if CONFIG_BOOTPARAM_HARDLOCKUP_PANIC is set) | |
3096 | To disable both hard and soft lockup detectors, | |
334bb79c | 3097 | please see 'nowatchdog'. |
0cb55ad2 RD |
3098 | This is useful when you use a panic=... timeout and |
3099 | need the box quickly up again. | |
1da177e4 | 3100 | |
d22881dc SW |
3101 | These settings can be accessed at runtime via |
3102 | the nmi_watchdog and hardlockup_panic sysctls. | |
3103 | ||
bff38771 AV |
3104 | netpoll.carrier_timeout= |
3105 | [NET] Specifies amount of time (in seconds) that | |
3106 | netpoll should wait for a carrier. By default netpoll | |
3107 | waits 4 seconds. | |
3108 | ||
cd4f0ef7 | 3109 | no387 [BUGS=X86-32] Tells the kernel to use the 387 maths |
1da177e4 LT |
3110 | emulation library even if a 387 maths coprocessor |
3111 | is present. | |
3112 | ||
372fddf7 KS |
3113 | no5lvl [X86-64] Disable 5-level paging mode. Forces |
3114 | kernel to use 4-level paging instead. | |
3115 | ||
b745cfba | 3116 | nofsgsbase [X86] Disables FSGSBASE instructions. |
dd649bd0 | 3117 | |
0cb55ad2 RD |
3118 | no_console_suspend |
3119 | [HW] Never suspend the console | |
3120 | Disable suspending of consoles during suspend and | |
3121 | hibernate operations. Once disabled, debugging | |
3122 | messages can reach various consoles while the rest | |
3123 | of the system is being put to sleep (ie, while | |
3124 | debugging driver suspend/resume hooks). This may | |
3125 | not work reliably with all consoles, but is known | |
3126 | to work with serial and VGA consoles. | |
134620f7 YZ |
3127 | To facilitate more flexible debugging, we also add |
3128 | console_suspend, a printk module parameter to control | |
3129 | it. Users could use console_suspend (usually | |
3130 | /sys/module/printk/parameters/console_suspend) to | |
3131 | turn on/off it dynamically. | |
0cb55ad2 | 3132 | |
c6c40533 KS |
3133 | novmcoredd [KNL,KDUMP] |
3134 | Disable device dump. Device dump allows drivers to | |
3135 | append dump data to vmcore so you can collect driver | |
3136 | specified debug info. Drivers can append the data | |
3137 | without any limit and this data is stored in memory, | |
3138 | so this may cause significant memory stress. Disabling | |
3139 | device dump can help save memory but the driver debug | |
3140 | data will be no longer available. This parameter | |
3141 | is only available when CONFIG_PROC_VMCORE_DEVICE_DUMP | |
3142 | is set. | |
3143 | ||
c1aee215 CL |
3144 | noaliencache [MM, NUMA, SLAB] Disables the allocation of alien |
3145 | caches in the slab allocator. Saves per-node memory, | |
3146 | but will impact performance. | |
3395ee05 | 3147 | |
a9913044 RD |
3148 | noalign [KNL,ARM] |
3149 | ||
686140a1 VG |
3150 | noaltinstr [S390] Disables alternative instructions patching |
3151 | (CPU alternatives feature). | |
3152 | ||
1da177e4 LT |
3153 | noapic [SMP,APIC] Tells the kernel to not make use of any |
3154 | IOAPICs that may be present in the system. | |
3155 | ||
5091faa4 MG |
3156 | noautogroup Disable scheduler automatic task group creation. |
3157 | ||
1da177e4 LT |
3158 | nobats [PPC] Do not use BATs for mapping kernel lowmem |
3159 | on "Classic" PPC cores. | |
3160 | ||
3161 | nocache [ARM] | |
a9913044 | 3162 | |
0cb55ad2 RD |
3163 | noclflush [BUGS=X86] Don't use the CLFLUSH instruction |
3164 | ||
163ecdff SN |
3165 | nodelayacct [KNL] Disable per-task delay accounting |
3166 | ||
6902aa84 PM |
3167 | nodsp [SH] Disable hardware DSP at boot time. |
3168 | ||
b2e0a54a | 3169 | noefi Disable EFI runtime services support. |
8b2cb7a8 | 3170 | |
1da177e4 LT |
3171 | noexec [IA-64] |
3172 | ||
6cececfc | 3173 | noexec [X86] |
f5a1b191 | 3174 | On X86-32 available only on PAE configured kernels. |
1da177e4 | 3175 | noexec=on: enable non-executable mappings (default) |
f5a1b191 JS |
3176 | noexec=off: disable non-executable mappings |
3177 | ||
de78a9c4 | 3178 | nosmap [X86,PPC] |
52b6179a PA |
3179 | Disable SMAP (Supervisor Mode Access Prevention) |
3180 | even if it is supported by processor. | |
3181 | ||
0fb1c25a | 3182 | nosmep [X86,PPC] |
52b6179a | 3183 | Disable SMEP (Supervisor Mode Execution Prevention) |
de5397ad FY |
3184 | even if it is supported by processor. |
3185 | ||
f5a1b191 JS |
3186 | noexec32 [X86-64] |
3187 | This affects only 32-bit executables. | |
3188 | noexec32=on: enable non-executable mappings (default) | |
3189 | read doesn't imply executable mappings | |
3190 | noexec32=off: disable non-executable mappings | |
3191 | read implies executable mappings | |
1da177e4 | 3192 | |
fab43ef4 | 3193 | nofpu [MIPS,SH] Disable hardware FPU at boot time. |
6902aa84 | 3194 | |
cd4f0ef7 | 3195 | nofxsr [BUGS=X86-32] Disables x86 floating point extended |
4f886511 CE |
3196 | register save and restore. The kernel will only save |
3197 | legacy floating-point registers on task switch. | |
1da177e4 | 3198 | |
5b280ed4 | 3199 | nohugeiomap [KNL,X86,PPC,ARM64] Disable kernel huge I/O mappings. |
0ddab1d2 | 3200 | |
52c48c51 SS |
3201 | nosmt [KNL,S390] Disable symmetric multithreading (SMT). |
3202 | Equivalent to smt=1. | |
3203 | ||
be3a5b0e | 3204 | [KNL,X86] Disable symmetric multithreading (SMT). |
506a66f3 TG |
3205 | nosmt=force: Force disable SMT, cannot be undone |
3206 | via the sysfs control file. | |
5e2d059b | 3207 | |
a2059825 JP |
3208 | nospectre_v1 [X86,PPC] Disable mitigations for Spectre Variant 1 |
3209 | (bounds check bypass). With this option data leaks are | |
3210 | possible in the system. | |
05736e4a | 3211 | |
e5ce5e72 JL |
3212 | nospectre_v2 [X86,PPC_FSL_BOOK3E,ARM64] Disable all mitigations for |
3213 | the Spectre variant 2 (indirect branch prediction) | |
3214 | vulnerability. System may allow data leaks with this | |
3215 | option. | |
da285121 | 3216 | |
24f7fc83 KRW |
3217 | nospec_store_bypass_disable |
3218 | [HW] Disable all mitigations for the Speculative Store Bypass vulnerability | |
3219 | ||
0c752a93 SS |
3220 | noxsave [BUGS=X86] Disables x86 extended register state save |
3221 | and restore using xsave. The kernel will fallback to | |
3222 | enabling legacy floating-point and sse state. | |
3223 | ||
b6f42a4a FY |
3224 | noxsaveopt [X86] Disables xsaveopt used in saving x86 extended |
3225 | register states. The kernel will fall back to use | |
3226 | xsave to save the states. By using this parameter, | |
3227 | performance of saving the states is degraded because | |
3228 | xsave doesn't support modified optimization while | |
3229 | xsaveopt supports it on xsaveopt enabled systems. | |
3230 | ||
3231 | noxsaves [X86] Disables xsaves and xrstors used in saving and | |
3232 | restoring x86 extended register state in compacted | |
3233 | form of xsave area. The kernel will fall back to use | |
3234 | xsaveopt and xrstor to save and restore the states | |
3235 | in standard form of xsave area. By using this | |
3236 | parameter, xsave area per process might occupy more | |
3237 | memory on xsaves enabled systems. | |
3238 | ||
01a24d2b PZ |
3239 | nohlt [BUGS=ARM,SH] Tells the kernel that the sleep(SH) or |
3240 | wfi(ARM) instruction doesn't work correctly and not to | |
3241 | use it. This is also useful when using JTAG debugger. | |
a9913044 | 3242 | |
1f29fae2 SH |
3243 | no_file_caps Tells the kernel not to honor file capabilities. The |
3244 | only way then for a file to be executed with privilege | |
3245 | is to be setuid root or executed by root. | |
3246 | ||
1da177e4 LT |
3247 | nohalt [IA-64] Tells the kernel not to use the power saving |
3248 | function PAL_HALT_LIGHT when idle. This increases | |
3249 | power-consumption. On the positive side, it reduces | |
3250 | interrupt wake-up latency, which may improve performance | |
3251 | in certain environments such as networked servers or | |
3252 | real-time systems. | |
3253 | ||
a6e15a39 KC |
3254 | nohibernate [HIBERNATION] Disable hibernation and resume. |
3255 | ||
79bf2bb3 TG |
3256 | nohz= [KNL] Boottime enable/disable dynamic ticks |
3257 | Valid arguments: on, off | |
3258 | Default: on | |
3259 | ||
d94d1053 | 3260 | nohz_full= [KNL,BOOT,SMP,ISOL] |
2d13e6ca | 3261 | The argument is a cpu list, as described above. |
c5bfece2 | 3262 | In kernels built with CONFIG_NO_HZ_FULL=y, set |
a831881b | 3263 | the specified list of CPUs whose tick will be stopped |
0453b435 | 3264 | whenever possible. The boot CPU will be forced outside |
f99bcb2c PM |
3265 | the range to maintain the timekeeping. Any CPUs |
3266 | in this list will have their RCU callbacks offloaded, | |
3267 | just as if they had also been called out in the | |
3268 | rcu_nocbs= boot parameter. | |
a831881b | 3269 | |
eeee7853 PM |
3270 | noiotrap [SH] Disables trapped I/O port accesses. |
3271 | ||
cd4f0ef7 | 3272 | noirqdebug [X86-32] Disables the code which attempts to detect and |
1da177e4 LT |
3273 | disable unhandled interrupt sources. |
3274 | ||
6cececfc | 3275 | no_timer_check [X86,APIC] Disables the code which tests for |
8542b200 ZA |
3276 | broken timer IRQ sources. |
3277 | ||
1da177e4 LT |
3278 | noisapnp [ISAPNP] Disables ISA PnP code. |
3279 | ||
3280 | noinitrd [RAM] Tells the kernel not to load any configured | |
3281 | initial RAM disk. | |
3282 | ||
03ea8155 WH |
3283 | nointremap [X86-64, Intel-IOMMU] Do not enable interrupt |
3284 | remapping. | |
d1423d56 | 3285 | [Deprecated - use intremap=off] |
03ea8155 | 3286 | |
1da177e4 LT |
3287 | nointroute [IA-64] |
3288 | ||
d12a72b8 AL |
3289 | noinvpcid [X86] Disable the INVPCID cpu feature. |
3290 | ||
16290246 | 3291 | nojitter [IA-64] Disables jitter checking for ITC timers. |
0aa366f3 | 3292 | |
9cf4c4fc JK |
3293 | no-kvmclock [X86,KVM] Disable paravirtualized KVM clock driver |
3294 | ||
fd10cde9 GN |
3295 | no-kvmapf [X86,KVM] Disable paravirtualized asynchronous page |
3296 | fault handling. | |
3297 | ||
80e9a4f2 AM |
3298 | no-vmw-sched-clock |
3299 | [X86,PV_OPS] Disable paravirtualized VMware scheduler | |
3300 | clock and use the default one. | |
3301 | ||
e73a8f38 | 3302 | no-steal-acc [X86,PV_OPS,ARM64] Disable paravirtualized steal time |
e0685fa2 SP |
3303 | accounting. steal time is computed, but won't |
3304 | influence scheduler behaviour | |
d910f5c1 | 3305 | |
cd4f0ef7 | 3306 | nolapic [X86-32,APIC] Do not enable or use the local APIC. |
1da177e4 | 3307 | |
cd4f0ef7 | 3308 | nolapic_timer [X86-32,APIC] Do not use the local APIC timer. |
ad62ca2b | 3309 | |
1da177e4 | 3310 | noltlbs [PPC] Do not use large page/tlb entries for kernel |
f15eea66 | 3311 | lowmem mapping on PPC40x and PPC8xx |
1da177e4 | 3312 | |
312f1f01 H |
3313 | nomca [IA-64] Disable machine check abort handling |
3314 | ||
13696e0a | 3315 | nomce [X86-32] Disable Machine Check Exception |
abe37e5a | 3316 | |
83d7384f AS |
3317 | nomfgpt [X86-32] Disable Multi-Function General Purpose |
3318 | Timer usage (for AMD Geode machines). | |
3319 | ||
bda62633 DZ |
3320 | nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to |
3321 | shutdown the other cpus. Instead use the REBOOT_VECTOR | |
3322 | irq. | |
3323 | ||
02608bef DY |
3324 | nomodule Disable module load |
3325 | ||
016ddd9b JK |
3326 | nopat [X86] Disable PAT (page attribute table extension of |
3327 | pagetables) support. | |
3328 | ||
0790c9aa AL |
3329 | nopcid [X86-64] Disable the PCID cpu feature. |
3330 | ||
0cb55ad2 RD |
3331 | norandmaps Don't use address space randomization. Equivalent to |
3332 | echo 0 > /proc/sys/kernel/randomize_va_space | |
3333 | ||
cd4f0ef7 | 3334 | noreplace-smp [X86-32,SMP] Don't replace SMP instructions |
b7fb4af0 JF |
3335 | with UP alternatives |
3336 | ||
7a5091d5 PA |
3337 | nordrand [X86] Disable kernel use of the RDRAND and |
3338 | RDSEED instructions even if they are supported | |
3339 | by the processor. RDRAND and RDSEED are still | |
3340 | available to user space applications. | |
49d859d7 | 3341 | |
a9913044 RD |
3342 | noresume [SWSUSP] Disables resume and restores original swap |
3343 | space. | |
3344 | ||
1da177e4 LT |
3345 | no-scroll [VGA] Disables scrollback. |
3346 | This is required for the Braillex ib80-piezo Braille | |
3347 | reader made by F.H. Papenmeier (Germany). | |
3348 | ||
3349 | nosbagart [IA-64] | |
3350 | ||
cd4f0ef7 | 3351 | nosep [BUGS=X86-32] Disables x86 SYSENTER/SYSEXIT support. |
4f886511 | 3352 | |
61ec7567 LB |
3353 | nosmp [SMP] Tells an SMP kernel to act as a UP kernel, |
3354 | and disable the IO APIC. legacy for "maxcpus=0". | |
1da177e4 | 3355 | |
97842216 DJ |
3356 | nosoftlockup [KNL] Disable the soft-lockup detector. |
3357 | ||
1da177e4 LT |
3358 | nosync [HW,M68K] Disables sync negotiation for all devices. |
3359 | ||
195daf66 | 3360 | nowatchdog [KNL] Disable both lockup detectors, i.e. |
6dddd7a7 | 3361 | soft-lockup and NMI watchdog (hard-lockup). |
58687acb | 3362 | |
1da177e4 | 3363 | nowb [ARM] |
a9913044 | 3364 | |
2b2fd87a WH |
3365 | nox2apic [X86-64,APIC] Do not enable x2APIC mode. |
3366 | ||
f78cff48 FY |
3367 | cpu0_hotplug [X86] Turn on CPU0 hotplug feature when |
3368 | CONFIG_BOOTPARAM_HOTPLUG_CPU0 is off. | |
3369 | Some features depend on CPU0. Known dependencies are: | |
3370 | 1. Resume from suspend/hibernate depends on CPU0. | |
3371 | Suspend/hibernate will fail if CPU0 is offline and you | |
3372 | need to online CPU0 before suspend/hibernate. | |
3373 | 2. PIC interrupts also depend on CPU0. CPU0 can't be | |
3374 | removed if a PIC interrupt is detected. | |
3375 | It's said poweroff/reboot may depend on CPU0 on some | |
3376 | machines although I haven't seen such issues so far | |
3377 | after CPU0 is offline on a few tested machines. | |
3378 | If the dependencies are under your control, you can | |
3379 | turn on cpu0_hotplug. | |
3380 | ||
6dddd7a7 | 3381 | nps_mtm_hs_ctr= [KNL,ARC] |
35b55ef2 NC |
3382 | This parameter sets the maximum duration, in |
3383 | cycles, each HW thread of the CTOP can run | |
3384 | without interruptions, before HW switches it. | |
3385 | The actual maximum duration is 16 times this | |
3386 | parameter's value. | |
3387 | Format: integer between 1 and 255 | |
3388 | Default: 255 | |
3389 | ||
16290246 | 3390 | nptcg= [IA-64] Override max number of concurrent global TLB |
a6c75b86 FY |
3391 | purges which is reported from either PAL_VM_SUMMARY or |
3392 | SAL PALO. | |
3393 | ||
2b633e3f YL |
3394 | nr_cpus= [SMP] Maximum number of processors that an SMP kernel |
3395 | could support. nr_cpus=n : n >= 1 limits the kernel to | |
7c142bfe BH |
3396 | support 'n' processors. It could be larger than the |
3397 | number of already plugged CPU during bootup, later in | |
3398 | runtime you can physically add extra cpu until it reaches | |
3399 | n. So during boot up some boot time memory for per-cpu | |
3400 | variables need be pre-allocated for later physical cpu | |
3401 | hot plugging. | |
2b633e3f | 3402 | |
0cb55ad2 RD |
3403 | nr_uarts= [SERIAL] maximum number of UARTs to be registered. |
3404 | ||
1a687c2e MG |
3405 | numa_balancing= [KNL,X86] Enable or disable automatic NUMA balancing. |
3406 | Allowed values are enable and disable | |
3407 | ||
f0c0b2b8 | 3408 | numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA. |
c9bff3ee | 3409 | 'node', 'default' can be specified |
f0c0b2b8 | 3410 | This can be set from sysctl after boot. |
57043247 | 3411 | See Documentation/admin-guide/sysctl/vm.rst for details. |
f0c0b2b8 | 3412 | |
7c4be253 | 3413 | ohci1394_dma=early [HW] enable debugging via the ohci1394 driver. |
a74e2a22 | 3414 | See Documentation/core-api/debugging-via-ohci1394.rst for more |
7c4be253 RD |
3415 | info. |
3416 | ||
3ef0e1f8 AS |
3417 | olpc_ec_timeout= [OLPC] ms delay when issuing EC commands |
3418 | Rather than timing out after 20 ms if an EC | |
3419 | command is not properly ACKed, override the length | |
3420 | of the timeout. We have interrupts disabled while | |
3421 | waiting for the ACK, so if this is set too high | |
3422 | interrupts *may* be lost! | |
3423 | ||
15ac7afe TL |
3424 | omap_mux= [OMAP] Override bootloader pin multiplexing. |
3425 | Format: <mux_mode0.mode_name=value>... | |
3426 | For example, to override I2C bus2: | |
3427 | omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100 | |
3428 | ||
1da177e4 LT |
3429 | oprofile.timer= [HW] |
3430 | Use timer interrupt instead of performance counters | |
3431 | ||
7e4e0bd5 RR |
3432 | oprofile.cpu_type= Force an oprofile cpu type |
3433 | This might be useful if you have an older oprofile | |
3434 | userland or if you want common events. | |
8d7ff4f2 RR |
3435 | Format: { arch_perfmon } |
3436 | arch_perfmon: [X86] Force use of architectural | |
7e4e0bd5 RR |
3437 | perfmon on Intel CPUs instead of the |
3438 | CPU specific event set. | |
159a80b2 RR |
3439 | timer: [X86] Force use of architectural NMI |
3440 | timer mode (see also oprofile.timer | |
3441 | for generic hr timer mode) | |
1dcdb5a9 | 3442 | |
44a4dcf7 RD |
3443 | oops=panic Always panic on oopses. Default is to just kill the |
3444 | process, but there is a small probability of | |
3445 | deadlocking the machine. | |
d404ab0a OH |
3446 | This will also cause panics on machine check exceptions. |
3447 | Useful together with panic=30 to trigger a reboot. | |
3448 | ||
e900a918 DW |
3449 | page_alloc.shuffle= |
3450 | [KNL] Boolean flag to control whether the page allocator | |
3451 | should randomize its free lists. The randomization may | |
3452 | be automatically enabled if the kernel detects it is | |
3453 | running on a platform with a direct-mapped memory-side | |
3454 | cache, and this parameter can be used to | |
3455 | override/disable that behavior. The state of the flag | |
3456 | can be read from sysfs at: | |
3457 | /sys/module/page_alloc/parameters/shuffle. | |
3458 | ||
48c96a36 JK |
3459 | page_owner= [KNL] Boot-time page_owner enabling option. |
3460 | Storage of the information about who allocated | |
3461 | each page is disabled in default. With this switch, | |
3462 | we can turn it on. | |
3463 | on: enable the feature | |
3464 | ||
8823b1db | 3465 | page_poison= [KNL] Boot-time parameter changing the state of |
8c9a134c KC |
3466 | poisoning on the buddy allocator, available with |
3467 | CONFIG_PAGE_POISONING=y. | |
3468 | off: turn off poisoning (default) | |
8823b1db LA |
3469 | on: turn on poisoning |
3470 | ||
44a4dcf7 | 3471 | panic= [KNL] Kernel behaviour on panic: delay <timeout> |
4302fbc8 HD |
3472 | timeout > 0: seconds before rebooting |
3473 | timeout = 0: wait forever | |
3474 | timeout < 0: reboot immediately | |
1da177e4 LT |
3475 | Format: <timeout> |
3476 | ||
d999bd93 FT |
3477 | panic_print= Bitmask for printing system info when panic happens. |
3478 | User can chose combination of the following bits: | |
3479 | bit 0: print all tasks info | |
3480 | bit 1: print system memory info | |
3481 | bit 2: print timer info | |
3482 | bit 3: print locks info if CONFIG_LOCKDEP is on | |
3483 | bit 4: print ftrace buffer | |
de6da1e8 | 3484 | bit 5: print all printk messages in buffer |
d999bd93 | 3485 | |
db38d5c1 RA |
3486 | panic_on_taint= Bitmask for conditionally calling panic() in add_taint() |
3487 | Format: <hex>[,nousertaint] | |
3488 | Hexadecimal bitmask representing the set of TAINT flags | |
3489 | that will cause the kernel to panic when add_taint() is | |
3490 | called with any of the flags in this set. | |
3491 | The optional switch "nousertaint" can be utilized to | |
3492 | prevent userspace forced crashes by writing to sysctl | |
3493 | /proc/sys/kernel/tainted any flagset matching with the | |
3494 | bitmask set on panic_on_taint. | |
3495 | See Documentation/admin-guide/tainted-kernels.rst for | |
3496 | extra details on the taint flags that users can pick | |
3497 | to compose the bitmask to assign to panic_on_taint. | |
3498 | ||
9e3961a0 PB |
3499 | panic_on_warn panic() instead of WARN(). Useful to cause kdump |
3500 | on a WARN(). | |
3501 | ||
f06e5153 MH |
3502 | crash_kexec_post_notifiers |
3503 | Run kdump after running panic-notifiers and dumping | |
3504 | kmsg. This only for the users who doubt kdump always | |
3505 | succeeds in any situation. | |
3506 | Note that this also increases risks of kdump failure, | |
3507 | because some panic notifiers can make the crashed | |
3508 | kernel more unstable. | |
3509 | ||
1da177e4 LT |
3510 | parkbd.port= [HW] Parallel port number the keyboard adapter is |
3511 | connected to, default is 0. | |
3512 | Format: <parport#> | |
3513 | parkbd.mode= [HW] Parallel port keyboard adapter mode of operation, | |
3514 | 0 for XT, 1 for AT (default is AT). | |
a9913044 RD |
3515 | Format: <mode> |
3516 | ||
3517 | parport= [HW,PPT] Specify parallel ports. 0 disables. | |
3518 | Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] } | |
3519 | Use 'auto' to force the driver to use any | |
3520 | IRQ/DMA settings detected (the default is to | |
3521 | ignore detected IRQ/DMA settings because of | |
3522 | possible conflicts). You can specify the base | |
3523 | address, IRQ, and DMA settings; IRQ and DMA | |
3524 | should be numbers, or 'auto' (for using detected | |
3525 | settings on that particular port), or 'nofifo' | |
3526 | (to avoid using a FIFO even if it is detected). | |
3527 | Parallel ports are assigned in the order they | |
3528 | are specified on the command line, starting | |
3529 | with parport0. | |
3530 | ||
3531 | parport_init_mode= [HW,PPT] | |
3532 | Configure VIA parallel port to operate in | |
3533 | a specific mode. This is necessary on Pegasos | |
3534 | computer where firmware has no options for setting | |
3535 | up parallel port mode and sets it to spp. | |
3536 | Currently this function knows 686a and 8231 chips. | |
1da177e4 LT |
3537 | Format: [spp|ps2|epp|ecp|ecpepp] |
3538 | ||
dd287796 AM |
3539 | pause_on_oops= |
3540 | Halt all CPUs after the first oops has been printed for | |
3541 | the specified number of seconds. This is to be used if | |
3542 | your oopses keep scrolling off the screen. | |
3543 | ||
1da177e4 LT |
3544 | pcbit= [HW,ISDN] |
3545 | ||
3546 | pcd. [PARIDE] | |
3547 | See header of drivers/block/paride/pcd.c. | |
e7751617 | 3548 | See also Documentation/admin-guide/blockdev/paride.rst. |
1da177e4 | 3549 | |
07d8d7e5 LG |
3550 | pci=option[,option...] [PCI] various PCI subsystem options. |
3551 | ||
3552 | Some options herein operate on a specific device | |
3553 | or a set of devices (<pci_dev>). These are | |
3554 | specified in one of the following formats: | |
3555 | ||
45db3370 | 3556 | [<domain>:]<bus>:<dev>.<func>[/<dev>.<func>]* |
07d8d7e5 LG |
3557 | pci:<vendor>:<device>[:<subvendor>:<subdevice>] |
3558 | ||
3559 | Note: the first format specifies a PCI | |
3560 | bus/device/function address which may change | |
3561 | if new hardware is inserted, if motherboard | |
3562 | firmware changes, or due to changes caused | |
3563 | by other kernel parameters. If the | |
3564 | domain is left unspecified, it is | |
45db3370 LG |
3565 | taken to be zero. Optionally, a path |
3566 | to a device through multiple device/function | |
3567 | addresses can be specified after the base | |
3568 | address (this is more robust against | |
3569 | renumbering issues). The second format | |
07d8d7e5 LG |
3570 | selects devices using IDs from the |
3571 | configuration space which may match multiple | |
3572 | devices in the system. | |
3573 | ||
11eb0e0e | 3574 | earlydump dump PCI config space before the kernel |
6dddd7a7 | 3575 | changes anything |
c0115606 | 3576 | off [X86] don't probe for the PCI bus |
cd4f0ef7 | 3577 | bios [X86-32] force use of PCI BIOS, don't access |
a9913044 RD |
3578 | the hardware directly. Use this if your machine |
3579 | has a non-standard PCI host bridge. | |
cd4f0ef7 | 3580 | nobios [X86-32] disallow use of PCI BIOS, only direct |
a9913044 RD |
3581 | hardware access methods are allowed. Use this |
3582 | if you experience crashes upon bootup and you | |
3583 | suspect they are caused by the BIOS. | |
afd8c084 BP |
3584 | conf1 [X86] Force use of PCI Configuration Access |
3585 | Mechanism 1 (config address in IO port 0xCF8, | |
3586 | data in IO port 0xCFC, both 32-bit). | |
3587 | conf2 [X86] Force use of PCI Configuration Access | |
3588 | Mechanism 2 (IO port 0xCF8 is an 8-bit port for | |
3589 | the function, IO port 0xCFA, also 8-bit, sets | |
3590 | bus number. The config space is then accessed | |
3591 | through ports 0xC000-0xCFFF). | |
3592 | See http://wiki.osdev.org/PCI for more info | |
3593 | on the configuration access mechanisms. | |
7f785763 RD |
3594 | noaer [PCIE] If the PCIEAER kernel config parameter is |
3595 | enabled, this kernel boot option can be used to | |
3596 | disable the use of PCIE advanced error reporting. | |
32a2eea7 JG |
3597 | nodomains [PCI] Disable support for multiple PCI |
3598 | root domains (aka PCI segments, in ACPI-speak). | |
6cececfc | 3599 | nommconf [X86] Disable use of MMCONFIG for PCI |
61be6d66 | 3600 | Configuration |
12983077 AH |
3601 | check_enable_amd_mmconf [X86] check for and enable |
3602 | properly configured MMIO access to PCI | |
3603 | config space on AMD family 10h CPU | |
309e57df MW |
3604 | nomsi [MSI] If the PCI_MSI kernel config parameter is |
3605 | enabled, this kernel boot option can be used to | |
3606 | disable the use of MSI interrupts system-wide. | |
a9322f64 SA |
3607 | noioapicquirk [APIC] Disable all boot interrupt quirks. |
3608 | Safety option to keep boot IRQs enabled. This | |
3609 | should never be necessary. | |
9197979b SA |
3610 | ioapicreroute [APIC] Enable rerouting of boot IRQs to the |
3611 | primary IO-APIC for bridges that cannot disable | |
3612 | boot IRQs. This fixes a source of spurious IRQs | |
3613 | when the system masks IRQs. | |
41b9eb26 SA |
3614 | noioapicreroute [APIC] Disable workaround that uses the |
3615 | boot IRQ equivalent of an IRQ that connects to | |
3616 | a chipset where boot IRQs cannot be disabled. | |
3617 | The opposite of ioapicreroute. | |
cd4f0ef7 | 3618 | biosirq [X86-32] Use PCI BIOS calls to get the interrupt |
a9913044 RD |
3619 | routing table. These calls are known to be buggy |
3620 | on several machines and they hang the machine | |
3621 | when used, but on other computers it's the only | |
3622 | way to get the interrupt routing table. Try | |
3623 | this option if the kernel is unable to allocate | |
3624 | IRQs or discover secondary PCI buses on your | |
3625 | motherboard. | |
c0115606 | 3626 | rom [X86] Assign address space to expansion ROMs. |
a9913044 RD |
3627 | Use with caution as certain devices share |
3628 | address decoders between ROMs and other | |
3629 | resources. | |
c0115606 | 3630 | norom [X86] Do not assign address space to |
bb71ad88 GH |
3631 | expansion ROMs that do not already have |
3632 | BIOS assigned address ranges. | |
7bd1c365 MH |
3633 | nobar [X86] Do not assign address space to the |
3634 | BARs that weren't assigned by the BIOS. | |
c0115606 | 3635 | irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be |
a9913044 RD |
3636 | assigned automatically to PCI devices. You can |
3637 | make the kernel exclude IRQs of your ISA cards | |
3638 | this way. | |
c0115606 | 3639 | pirqaddr=0xAAAAA [X86] Specify the physical address |
a9913044 RD |
3640 | of the PIRQ table (normally generated |
3641 | by the BIOS) if it is outside the | |
3642 | F0000h-100000h range. | |
c0115606 | 3643 | lastbus=N [X86] Scan all buses thru bus #N. Can be |
a9913044 RD |
3644 | useful if the kernel is unable to find your |
3645 | secondary buses and you want to tell it | |
3646 | explicitly which ones they are. | |
c0115606 | 3647 | assign-busses [X86] Always assign all PCI bus |
a9913044 RD |
3648 | numbers ourselves, overriding |
3649 | whatever the firmware may have done. | |
c0115606 | 3650 | usepirqmask [X86] Honor the possible IRQ mask stored |
a9913044 RD |
3651 | in the BIOS $PIR table. This is needed on |
3652 | some systems with broken BIOSes, notably | |
3653 | some HP Pavilion N5400 and Omnibook XE3 | |
3654 | notebooks. This will have no effect if ACPI | |
3655 | IRQ routing is enabled. | |
c0115606 | 3656 | noacpi [X86] Do not use ACPI for IRQ routing |
a9913044 | 3657 | or for PCI scanning. |
7bc5e3f2 BH |
3658 | use_crs [X86] Use PCI host bridge window information |
3659 | from ACPI. On BIOSes from 2008 or later, this | |
3660 | is enabled by default. If you need to use this, | |
3661 | please report a bug. | |
3662 | nocrs [X86] Ignore PCI host bridge windows from ACPI. | |
6dddd7a7 | 3663 | If you need to use this, please report a bug. |
a9913044 RD |
3664 | routeirq Do IRQ routing for all PCI devices. |
3665 | This is normally done in pci_enable_device(), | |
3666 | so this option is a temporary workaround | |
3667 | for broken drivers that don't call it. | |
13a6ddb0 YL |
3668 | skip_isa_align [X86] do not align io start addr, so can |
3669 | handle more pci cards | |
0637a70a AK |
3670 | noearly [X86] Don't do any early type 1 scanning. |
3671 | This might help on some broken boards which | |
3672 | machine check when some devices' config space | |
3673 | is read. But various workarounds are disabled | |
3674 | and some IOMMU drivers will not work. | |
6b4b78fe MD |
3675 | bfsort Sort PCI devices into breadth-first order. |
3676 | This sorting is done to get a device | |
3677 | order compatible with older (<= 2.4) kernels. | |
3678 | nobfsort Don't sort PCI devices into breadth-first order. | |
fa238712 YW |
3679 | pcie_bus_tune_off Disable PCIe MPS (Max Payload Size) |
3680 | tuning and use the BIOS-configured MPS defaults. | |
3681 | pcie_bus_safe Set every device's MPS to the largest value | |
3682 | supported by all devices below the root complex. | |
3683 | pcie_bus_perf Set device MPS to the largest allowable MPS | |
3684 | based on its parent bus. Also set MRRS (Max | |
3685 | Read Request Size) to the largest supported | |
3686 | value (no larger than the MPS that the device | |
3687 | or bus can support) for best performance. | |
3688 | pcie_bus_peer2peer Set every device's MPS to 128B, which | |
3689 | every device is guaranteed to support. This | |
3690 | configuration allows peer-to-peer DMA between | |
3691 | any pair of devices, possibly at the cost of | |
3692 | reduced performance. This also guarantees | |
3693 | that hot-added devices will work. | |
4516a618 AN |
3694 | cbiosize=nn[KMG] The fixed amount of bus space which is |
3695 | reserved for the CardBus bridge's IO window. | |
3696 | The default value is 256 bytes. | |
3697 | cbmemsize=nn[KMG] The fixed amount of bus space which is | |
3698 | reserved for the CardBus bridge's memory | |
3699 | window. The default value is 64 megabytes. | |
32a9a682 YS |
3700 | resource_alignment= |
3701 | Format: | |
07d8d7e5 | 3702 | [<order of align>@]<pci_dev>[; ...] |
32a9a682 | 3703 | Specifies alignment and device to reassign |
07d8d7e5 LG |
3704 | aligned memory resources. How to |
3705 | specify the device is described above. | |
32a9a682 YS |
3706 | If <order of align> is not specified, |
3707 | PAGE_SIZE is used as alignment. | |
3b1b1ce3 | 3708 | A PCI-PCI bridge can be specified if resource |
32a9a682 | 3709 | windows need to be expanded. |
8b078c60 MK |
3710 | To specify the alignment for several |
3711 | instances of a device, the PCI vendor, | |
3712 | device, subvendor, and subdevice may be | |
3b1b1ce3 AK |
3713 | specified, e.g., 12@pci:8086:9c22:103c:198f |
3714 | for 4096-byte alignment. | |
43c16408 AP |
3715 | ecrc= Enable/disable PCIe ECRC (transaction layer |
3716 | end-to-end CRC checking). | |
3717 | bios: Use BIOS/firmware settings. This is the | |
3718 | the default. | |
3719 | off: Turn ECRC off | |
3720 | on: Turn ECRC on. | |
8c8803c5 YW |
3721 | hpiosize=nn[KMG] The fixed amount of bus space which is |
3722 | reserved for hotplug bridge's IO window. | |
3723 | Default size is 256 bytes. | |
d7b8a217 NJ |
3724 | hpmmiosize=nn[KMG] The fixed amount of bus space which is |
3725 | reserved for hotplug bridge's MMIO window. | |
3726 | Default size is 2 megabytes. | |
3727 | hpmmioprefsize=nn[KMG] The fixed amount of bus space which is | |
3728 | reserved for hotplug bridge's MMIO_PREF window. | |
3729 | Default size is 2 megabytes. | |
8c8803c5 | 3730 | hpmemsize=nn[KMG] The fixed amount of bus space which is |
d7b8a217 NJ |
3731 | reserved for hotplug bridge's MMIO and |
3732 | MMIO_PREF window. | |
8c8803c5 | 3733 | Default size is 2 megabytes. |
e16b4660 KB |
3734 | hpbussize=nn The minimum amount of additional bus numbers |
3735 | reserved for buses below a hotplug bridge. | |
3736 | Default is 1. | |
b55438fd YL |
3737 | realloc= Enable/disable reallocating PCI bridge resources |
3738 | if allocations done by BIOS are too small to | |
3739 | accommodate resources required by all child | |
3740 | devices. | |
3741 | off: Turn realloc off | |
3742 | on: Turn realloc on | |
3743 | realloc same as realloc=on | |
6748dcc2 | 3744 | noari do not use PCIe ARI. |
cef74409 GK |
3745 | noats [PCIE, Intel-IOMMU, AMD-IOMMU] |
3746 | do not use PCIe ATS (and IOMMU device IOTLB). | |
284f5f9d BH |
3747 | pcie_scan_all Scan all possible PCIe devices. Otherwise we |
3748 | only look for one device below a PCIe downstream | |
3749 | port. | |
f32ab754 UCCB |
3750 | big_root_window Try to add a big 64bit memory window to the PCIe |
3751 | root complex on AMD CPUs. Some GFX hardware | |
3752 | can resize a BAR to allow access to all VRAM. | |
3753 | Adding the window is slightly risky (it may | |
3754 | conflict with unreported devices), so this | |
3755 | taints the kernel. | |
aaca43fd LG |
3756 | disable_acs_redir=<pci_dev>[; ...] |
3757 | Specify one or more PCI devices (in the format | |
3758 | specified above) separated by semicolons. | |
3759 | Each device specified will have the PCI ACS | |
3760 | redirect capabilities forced off which will | |
3761 | allow P2P traffic between devices through | |
3762 | bridges without forcing it upstream. Note: | |
3763 | this removes isolation between devices and | |
3764 | may put more devices in an IOMMU group. | |
fbfe07d4 | 3765 | force_floating [S390] Force usage of floating interrupts. |
56271303 | 3766 | nomio [S390] Do not use MIO instructions. |
de267a7c PM |
3767 | norid [S390] ignore the RID field and force use of |
3768 | one PCI domain per PCI function | |
6b4b78fe | 3769 | |
e5665a45 CE |
3770 | pcie_aspm= [PCIE] Forcibly enable or disable PCIe Active State Power |
3771 | Management. | |
3772 | off Disable ASPM. | |
3773 | force Enable ASPM even on devices that claim not to support it. | |
3774 | WARNING: Forcing ASPM on may cause system lockups. | |
3775 | ||
4c0fd764 BH |
3776 | pcie_ports= [PCIE] PCIe port services handling: |
3777 | native Use native PCIe services (PME, AER, DPC, PCIe hotplug) | |
3778 | even if the platform doesn't give the OS permission to | |
3779 | use them. This may cause conflicts if the platform | |
3780 | also tries to use these services. | |
35a0b237 OJ |
3781 | dpc-native Use native PCIe service for DPC only. May |
3782 | cause conflicts if firmware uses AER or DPC. | |
4c0fd764 BH |
3783 | compat Disable native PCIe services (PME, AER, DPC, PCIe |
3784 | hotplug). | |
79dd9182 | 3785 | |
9d26d3a8 MW |
3786 | pcie_port_pm= [PCIE] PCIe port power management handling: |
3787 | off Disable power management of all PCIe ports | |
3788 | force Forcibly enable power management of all PCIe ports | |
3789 | ||
c7f48656 | 3790 | pcie_pme= [PCIE,PM] Native PCIe PME signaling options: |
c39fae14 | 3791 | nomsi Do not use MSI for native PCIe PME signaling (this makes |
28eb5f27 | 3792 | all PCIe root ports use INTx for all services). |
c7f48656 | 3793 | |
1da177e4 LT |
3794 | pcmv= [HW,PCMCIA] BadgePAD 4 |
3795 | ||
39ac5ba5 TB |
3796 | pd_ignore_unused |
3797 | [PM] | |
3798 | Keep all power-domains already enabled by bootloader on, | |
3799 | even if no driver has claimed them. This is useful | |
3800 | for debug and development, but should not be | |
3801 | needed on a platform with proper driver support. | |
3802 | ||
1da177e4 | 3803 | pd. [PARIDE] |
e7751617 | 3804 | See Documentation/admin-guide/blockdev/paride.rst. |
1da177e4 LT |
3805 | |
3806 | pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at | |
3807 | boot time. | |
3808 | Format: { 0 | 1 } | |
3809 | See arch/parisc/kernel/pdc_chassis.c | |
3810 | ||
f58dc01b | 3811 | percpu_alloc= Select which percpu first chunk allocator to use. |
e933a73f TH |
3812 | Currently supported values are "embed" and "page". |
3813 | Archs may support subset or none of the selections. | |
3814 | See comments in mm/percpu.c for details on each | |
3815 | allocator. This parameter is primarily for debugging | |
3816 | and performance comparison. | |
fa8a7094 | 3817 | |
1da177e4 | 3818 | pf. [PARIDE] |
e7751617 | 3819 | See Documentation/admin-guide/blockdev/paride.rst. |
1da177e4 LT |
3820 | |
3821 | pg. [PARIDE] | |
e7751617 | 3822 | See Documentation/admin-guide/blockdev/paride.rst. |
1da177e4 LT |
3823 | |
3824 | pirq= [SMP,APIC] Manual mp-table setup | |
cb1aaebe | 3825 | See Documentation/x86/i386/IO-APIC.rst. |
1da177e4 LT |
3826 | |
3827 | plip= [PPT,NET] Parallel port network link | |
3828 | Format: { parport<nr> | timid | 0 } | |
3ba9b1b8 | 3829 | See also Documentation/admin-guide/parport.rst. |
1da177e4 | 3830 | |
16290246 | 3831 | pmtmr= [X86] Manual setup of pmtmr I/O Port. |
de32a243 TG |
3832 | Override pmtimer IOPort with a hex value. |
3833 | e.g. pmtmr=0x508 | |
3834 | ||
db96a759 CY |
3835 | pm_debug_messages [SUSPEND,KNL] |
3836 | Enable suspend/resume debug messages during boot up. | |
3837 | ||
96242116 BH |
3838 | pnp.debug=1 [PNP] |
3839 | Enable PNP debug messages (depends on the | |
3840 | CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time | |
3841 | via /sys/module/pnp/parameters/debug. We always show | |
3842 | current resource usage; turning this on also shows | |
3843 | possible settings and some assignment information. | |
97ef062b | 3844 | |
1da177e4 LT |
3845 | pnpacpi= [ACPI] |
3846 | { off } | |
3847 | ||
3848 | pnpbios= [ISAPNP] | |
3849 | { on | off | curr | res | no-curr | no-res } | |
3850 | ||
3851 | pnp_reserve_irq= | |
3852 | [ISAPNP] Exclude IRQs for the autoconfiguration | |
3853 | ||
3854 | pnp_reserve_dma= | |
3855 | [ISAPNP] Exclude DMAs for the autoconfiguration | |
3856 | ||
3857 | pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration | |
a9913044 | 3858 | Ranges are in pairs (I/O port base and size). |
1da177e4 LT |
3859 | |
3860 | pnp_reserve_mem= | |
a9913044 RD |
3861 | [ISAPNP] Exclude memory regions for the |
3862 | autoconfiguration. | |
1da177e4 LT |
3863 | Ranges are in pairs (memory base and size). |
3864 | ||
4af94f39 RD |
3865 | ports= [IP_VS_FTP] IPVS ftp helper module |
3866 | Default is 21. | |
3867 | Up to 8 (IP_VS_APP_MAX_PORTS) ports | |
3868 | may be specified. | |
3869 | Format: <port>,<port>.... | |
3870 | ||
c3cbd075 BS |
3871 | powersave=off [PPC] This option disables power saving features. |
3872 | It specifically disables cpuidle and sets the | |
3873 | platform machine description specific power_save | |
3874 | function to NULL. On Idle the CPU just reduces | |
3875 | execution priority. | |
3876 | ||
3eb5d588 AB |
3877 | ppc_strict_facility_enable |
3878 | [PPC] This option catches any kernel floating point, | |
3879 | Altivec, VSX and SPE outside of regions specifically | |
3880 | allowed (eg kernel_enable_fpu()/kernel_disable_fpu()). | |
3881 | There is some performance impact when enabling this. | |
3882 | ||
07fd1761 CB |
3883 | ppc_tm= [PPC] |
3884 | Format: {"off"} | |
3885 | Disable Hardware Transactional Memory | |
3886 | ||
45807a1d IM |
3887 | print-fatal-signals= |
3888 | [KNL] debug: print fatal signals | |
f84d49b2 NO |
3889 | |
3890 | If enabled, warn about various signal handling | |
3891 | related application anomalies: too many signals, | |
3892 | too many POSIX.1 timers, fatal signals causing a | |
3893 | coredump - etc. | |
3894 | ||
3895 | If you hit the warning due to signal overflow, | |
3896 | you might want to try "ulimit -i unlimited". | |
3897 | ||
45807a1d IM |
3898 | default: off. |
3899 | ||
c22ab332 MG |
3900 | printk.always_kmsg_dump= |
3901 | Trigger kmsg_dump for cases other than kernel oops or | |
3902 | panics | |
3903 | Format: <bool> (1/Y/y=enable, 0/N/n=disable) | |
3904 | default: disabled | |
3905 | ||
750afe7b BP |
3906 | printk.devkmsg={on,off,ratelimit} |
3907 | Control writing to /dev/kmsg. | |
3908 | on - unlimited logging to /dev/kmsg from userspace | |
3909 | off - logging to /dev/kmsg disabled | |
3910 | ratelimit - ratelimit the logging | |
3911 | Default: ratelimit | |
3912 | ||
e84845c4 RD |
3913 | printk.time= Show timing data prefixed to each printk message line |
3914 | Format: <bool> (1/Y/y=enable, 0/N/n=disable) | |
3915 | ||
0cb55ad2 RD |
3916 | processor.max_cstate= [HW,ACPI] |
3917 | Limit processor to maximum C-state | |
3918 | max_cstate=9 overrides any DMI blacklist limit. | |
3919 | ||
3920 | processor.nocst [HW,ACPI] | |
3921 | Ignore the _CST method to determine C-states, | |
3922 | instead using the legacy FADT method | |
3923 | ||
1da177e4 | 3924 | profile= [KNL] Enable kernel profiling via /proc/profile |
e7e61fc0 RD |
3925 | Format: [<profiletype>,]<number> |
3926 | Param: <profiletype>: "schedule", "sleep", or "kvm" | |
3927 | [defaults to kernel profiling] | |
a9913044 | 3928 | Param: "schedule" - profile schedule points. |
b3da2a73 MG |
3929 | Param: "sleep" - profile D-state sleeping (millisecs). |
3930 | Requires CONFIG_SCHEDSTATS | |
c0fe2e69 | 3931 | Param: "kvm" - profile VM exits. |
e7e61fc0 RD |
3932 | Param: <number> - step/bucket size as a power of 2 for |
3933 | statistical time based profiling. | |
1da177e4 | 3934 | |
6b99e6e6 | 3935 | prompt_ramdisk= [RAM] [Deprecated] |
1da177e4 | 3936 | |
ecdc5d84 VG |
3937 | prot_virt= [S390] enable hosting protected virtual machines |
3938 | isolated from the hypervisor (if hardware supports | |
3939 | that). | |
3940 | Format: <bool> | |
3941 | ||
e0c27447 JW |
3942 | psi= [KNL] Enable or disable pressure stall information |
3943 | tracking. | |
3944 | Format: <bool> | |
3945 | ||
a9913044 RD |
3946 | psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to |
3947 | probe for; one of (bare|imps|exps|lifebook|any). | |
1da177e4 LT |
3948 | psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports |
3949 | per second. | |
a9913044 RD |
3950 | psmouse.resetafter= [HW,MOUSE] |
3951 | Try to reset the device after so many bad packets | |
1da177e4 LT |
3952 | (0 = never). |
3953 | psmouse.resolution= | |
3954 | [HW,MOUSE] Set desired mouse resolution, in dpi. | |
3955 | psmouse.smartscroll= | |
a9913044 | 3956 | [HW,MOUSE] Controls Logitech smartscroll autorepeat. |
1da177e4 LT |
3957 | 0 = disabled, 1 = enabled (default). |
3958 | ||
dee28e72 MG |
3959 | pstore.backend= Specify the name of the pstore backend to use |
3960 | ||
1da177e4 | 3961 | pt. [PARIDE] |
e7751617 | 3962 | See Documentation/admin-guide/blockdev/paride.rst. |
1da177e4 | 3963 | |
be3a5b0e | 3964 | pti= [X86-64] Control Page Table Isolation of user and |
01c9b17b DH |
3965 | kernel address spaces. Disabling this feature |
3966 | removes hardening, but improves performance of | |
3967 | system calls and interrupts. | |
3968 | ||
3969 | on - unconditionally enable | |
3970 | off - unconditionally disable | |
3971 | auto - kernel detects whether your CPU model is | |
3972 | vulnerable to issues that PTI mitigates | |
3973 | ||
3974 | Not specifying this option is equivalent to pti=auto. | |
3975 | ||
be3a5b0e | 3976 | nopti [X86-64] |
01c9b17b | 3977 | Equivalent to pti=off |
41f4c20b | 3978 | |
dc8c8587 KS |
3979 | pty.legacy_count= |
3980 | [KNL] Number of legacy pty's. Overwrites compiled-in | |
3981 | default number. | |
3982 | ||
7d2c502f | 3983 | quiet [KNL] Disable most log messages |
a9913044 | 3984 | |
1da177e4 LT |
3985 | r128= [HW,DRM] |
3986 | ||
3987 | raid= [HW,RAID] | |
e52347bd | 3988 | See Documentation/admin-guide/md.rst. |
1da177e4 | 3989 | |
1da177e4 | 3990 | ramdisk_size= [RAM] Sizes of RAM disks in kilobytes |
e7751617 | 3991 | See Documentation/admin-guide/blockdev/ramdisk.rst. |
1da177e4 | 3992 | |
6b99e6e6 RD |
3993 | ramdisk_start= [RAM] RAM disk image start address |
3994 | ||
9b254366 KC |
3995 | random.trust_cpu={on,off} |
3996 | [KNL] Enable or disable trusting the use of the | |
3997 | CPU's random number generator (if available) to | |
3998 | fully seed the kernel's CRNG. Default is controlled | |
3999 | by CONFIG_RANDOM_TRUST_CPU. | |
4000 | ||
011d8261 BP |
4001 | ras=option[,option,...] [KNL] RAS-specific options |
4002 | ||
4003 | cec_disable [X86] | |
4004 | Disable the Correctable Errors Collector, | |
4005 | see CONFIG_RAS_CEC help text. | |
4006 | ||
4102adab | 4007 | rcu_nocbs= [KNL] |
da8739f2 PM |
4008 | The argument is a cpu list, as described above, |
4009 | except that the string "all" can be used to | |
4010 | specify every CPU on the system. | |
2d13e6ca | 4011 | |
3fbfbf7a PM |
4012 | In kernels built with CONFIG_RCU_NOCB_CPU=y, set |
4013 | the specified list of CPUs to be no-callback CPUs. | |
77095901 PM |
4014 | Invocation of these CPUs' RCU callbacks will be |
4015 | offloaded to "rcuox/N" kthreads created for that | |
4016 | purpose, where "x" is "p" for RCU-preempt, and | |
4017 | "s" for RCU-sched, and "N" is the CPU number. | |
4018 | This reduces OS jitter on the offloaded CPUs, | |
4019 | which can be useful for HPC and real-time | |
4020 | workloads. It can also improve energy efficiency | |
4021 | for asymmetric multiprocessors. | |
3fbfbf7a | 4022 | |
4102adab | 4023 | rcu_nocb_poll [KNL] |
3fbfbf7a PM |
4024 | Rather than requiring that offloaded CPUs |
4025 | (specified by rcu_nocbs= above) explicitly | |
4026 | awaken the corresponding "rcuoN" kthreads, | |
4027 | make these kthreads poll for callbacks. | |
4028 | This improves the real-time response for the | |
4029 | offloaded CPUs by relieving them of the need to | |
4030 | wake up the corresponding kthread, but degrades | |
4031 | energy efficiency by requiring that the kthreads | |
4032 | periodically wake up to do the polling. | |
4033 | ||
4102adab | 4034 | rcutree.blimit= [KNL] |
97e63f0c PM |
4035 | Set maximum number of finished RCU callbacks to |
4036 | process in one batch. | |
21a1ea9e | 4037 | |
a3dc2948 PM |
4038 | rcutree.dump_tree= [KNL] |
4039 | Dump the structure of the rcu_node combining tree | |
4040 | out at early boot. This is used for diagnostic | |
4041 | purposes, to verify correct tree setup. | |
4042 | ||
0f41c0dd PM |
4043 | rcutree.gp_cleanup_delay= [KNL] |
4044 | Set the number of jiffies to delay each step of | |
90040c9e | 4045 | RCU grace-period cleanup. |
0f41c0dd | 4046 | |
37745d28 PM |
4047 | rcutree.gp_init_delay= [KNL] |
4048 | Set the number of jiffies to delay each step of | |
90040c9e | 4049 | RCU grace-period initialization. |
0f41c0dd PM |
4050 | |
4051 | rcutree.gp_preinit_delay= [KNL] | |
4052 | Set the number of jiffies to delay each step of | |
4053 | RCU grace-period pre-initialization, that is, | |
4054 | the propagation of recent CPU-hotplug changes up | |
90040c9e | 4055 | the rcu_node combining tree. |
37745d28 | 4056 | |
48d07c04 SAS |
4057 | rcutree.use_softirq= [KNL] |
4058 | If set to zero, move all RCU_SOFTIRQ processing to | |
4059 | per-CPU rcuc kthreads. Defaults to a non-zero | |
4060 | value, meaning that RCU_SOFTIRQ is used by default. | |
4061 | Specify rcutree.use_softirq=0 to use rcuc kthreads. | |
4062 | ||
7fa27001 PM |
4063 | rcutree.rcu_fanout_exact= [KNL] |
4064 | Disable autobalancing of the rcu_node combining | |
4065 | tree. This is used by rcutorture, and might | |
4066 | possibly be useful for architectures having high | |
4067 | cache-to-cache transfer latencies. | |
37745d28 | 4068 | |
4102adab | 4069 | rcutree.rcu_fanout_leaf= [KNL] |
ee968ac6 PM |
4070 | Change the number of CPUs assigned to each |
4071 | leaf rcu_node structure. Useful for very | |
4072 | large systems, which will choose the value 64, | |
4073 | and for NUMA systems with large remote-access | |
4074 | latencies, which will choose a value aligned | |
4075 | with the appropriate hardware boundaries. | |
f885b7f2 | 4076 | |
53c72b59 URS |
4077 | rcutree.rcu_min_cached_objs= [KNL] |
4078 | Minimum number of objects which are cached and | |
4079 | maintained per one CPU. Object size is equal | |
4080 | to PAGE_SIZE. The cache allows to reduce the | |
4081 | pressure to page allocator, also it makes the | |
4082 | whole algorithm to behave better in low memory | |
4083 | condition. | |
4084 | ||
4102adab | 4085 | rcutree.jiffies_till_first_fqs= [KNL] |
c0f4dfd4 PM |
4086 | Set delay from grace-period initialization to |
4087 | first attempt to force quiescent states. | |
4088 | Units are jiffies, minimum value is zero, | |
4089 | and maximum value is HZ. | |
4090 | ||
4102adab | 4091 | rcutree.jiffies_till_next_fqs= [KNL] |
c0f4dfd4 PM |
4092 | Set delay between subsequent attempts to force |
4093 | quiescent states. Units are jiffies, minimum | |
4094 | value is one, and maximum value is HZ. | |
4095 | ||
1a4762b9 PM |
4096 | rcutree.jiffies_till_sched_qs= [KNL] |
4097 | Set required age in jiffies for a | |
4098 | given grace period before RCU starts | |
4099 | soliciting quiescent-state help from | |
4100 | rcu_note_context_switch() and cond_resched(). | |
4101 | If not specified, the kernel will calculate | |
4102 | a value based on the most recent settings | |
4103 | of rcutree.jiffies_till_first_fqs | |
4104 | and rcutree.jiffies_till_next_fqs. | |
4105 | This calculated value may be viewed in | |
4106 | rcutree.jiffies_to_sched_qs. Any attempt to set | |
4107 | rcutree.jiffies_to_sched_qs will be cheerfully | |
4108 | overwritten. | |
4109 | ||
21871d7e | 4110 | rcutree.kthread_prio= [KNL,BOOT] |
d2af1ad7 PM |
4111 | Set the SCHED_FIFO priority of the RCU per-CPU |
4112 | kthreads (rcuc/N). This value is also used for | |
4113 | the priority of the RCU boost threads (rcub/N) | |
4114 | and for the RCU grace-period kthreads (rcu_bh, | |
4115 | rcu_preempt, and rcu_sched). If RCU_BOOST is | |
4116 | set, valid values are 1-99 and the default is 1 | |
4117 | (the least-favored priority). Otherwise, when | |
4118 | RCU_BOOST is not set, valid values are 0-99 and | |
4119 | the default is zero (non-realtime operation). | |
21871d7e | 4120 | |
f7c612b0 PM |
4121 | rcutree.rcu_nocb_gp_stride= [KNL] |
4122 | Set the number of NOCB callback kthreads in | |
4123 | each group, which defaults to the square root | |
4124 | of the number of CPUs. Larger numbers reduce | |
4125 | the wakeup overhead on the global grace-period | |
4126 | kthread, but increases that same overhead on | |
4127 | each group's NOCB grace-period kthread. | |
fbce7497 | 4128 | |
4102adab | 4129 | rcutree.qhimark= [KNL] |
97e63f0c PM |
4130 | Set threshold of queued RCU callbacks beyond which |
4131 | batch limiting is disabled. | |
21a1ea9e | 4132 | |
4102adab | 4133 | rcutree.qlowmark= [KNL] |
24aaef8d RD |
4134 | Set threshold of queued RCU callbacks below which |
4135 | batch limiting is re-enabled. | |
21a1ea9e | 4136 | |
b2b00ddf PM |
4137 | rcutree.qovld= [KNL] |
4138 | Set threshold of queued RCU callbacks beyond which | |
4139 | RCU's force-quiescent-state scan will aggressively | |
4140 | enlist help from cond_resched() and sched IPIs to | |
4141 | help CPUs more quickly reach quiescent states. | |
4142 | Set to less than zero to make this be set based | |
4143 | on rcutree.qhimark at boot time and to zero to | |
4144 | disable more aggressive help enlistment. | |
4145 | ||
4102adab | 4146 | rcutree.rcu_idle_gp_delay= [KNL] |
c0f4dfd4 PM |
4147 | Set wakeup interval for idle CPUs that have |
4148 | RCU callbacks (RCU_FAST_NO_HZ=y). | |
d40011f6 | 4149 | |
4102adab | 4150 | rcutree.rcu_idle_lazy_gp_delay= [KNL] |
c0f4dfd4 PM |
4151 | Set wakeup interval for idle CPUs that have |
4152 | only "lazy" RCU callbacks (RCU_FAST_NO_HZ=y). | |
4153 | Lazy RCU callbacks are those which RCU can | |
4154 | prove do nothing more than free memory. | |
d40011f6 | 4155 | |
e3c50dfb PM |
4156 | rcutree.rcu_kick_kthreads= [KNL] |
4157 | Cause the grace-period kthread to get an extra | |
4158 | wake_up() if it sleeps three times longer than | |
4159 | it should at force-quiescent-state time. | |
4160 | This wake_up() will be accompanied by a | |
4161 | WARN_ONCE() splat and an ftrace_dump(). | |
4162 | ||
2ccaff10 PM |
4163 | rcutree.sysrq_rcu= [KNL] |
4164 | Commandeer a sysrq key to dump out Tree RCU's | |
4165 | rcu_node tree with an eye towards determining | |
4166 | why a new grace period has not yet started. | |
4167 | ||
881ed593 PM |
4168 | rcuperf.gp_async= [KNL] |
4169 | Measure performance of asynchronous | |
4170 | grace-period primitives such as call_rcu(). | |
4171 | ||
4172 | rcuperf.gp_async_max= [KNL] | |
4173 | Specify the maximum number of outstanding | |
4174 | callbacks per writer thread. When a writer | |
4175 | thread exceeds this limit, it invokes the | |
4176 | corresponding flavor of rcu_barrier() to allow | |
4177 | previously posted callbacks to drain. | |
4178 | ||
bdea9e34 PM |
4179 | rcuperf.gp_exp= [KNL] |
4180 | Measure performance of expedited synchronous | |
4181 | grace-period primitives. | |
4182 | ||
df37e66b PM |
4183 | rcuperf.holdoff= [KNL] |
4184 | Set test-start holdoff period. The purpose of | |
4185 | this parameter is to delay the start of the | |
4186 | test until boot completes in order to avoid | |
4187 | interference. | |
4188 | ||
e6e78b00 JFG |
4189 | rcuperf.kfree_rcu_test= [KNL] |
4190 | Set to measure performance of kfree_rcu() flooding. | |
4191 | ||
4192 | rcuperf.kfree_nthreads= [KNL] | |
4193 | The number of threads running loops of kfree_rcu(). | |
4194 | ||
4195 | rcuperf.kfree_alloc_num= [KNL] | |
4196 | Number of allocations and frees done in an iteration. | |
4197 | ||
4198 | rcuperf.kfree_loops= [KNL] | |
4199 | Number of loops doing rcuperf.kfree_alloc_num number | |
4200 | of allocations and frees. | |
4201 | ||
bdea9e34 PM |
4202 | rcuperf.nreaders= [KNL] |
4203 | Set number of RCU readers. The value -1 selects | |
4204 | N, where N is the number of CPUs. A value | |
4205 | "n" less than -1 selects N-n+1, where N is again | |
4206 | the number of CPUs. For example, -2 selects N | |
4207 | (the number of CPUs), -3 selects N+1, and so on. | |
4208 | A value of "n" less than or equal to -N selects | |
4209 | a single reader. | |
4210 | ||
4211 | rcuperf.nwriters= [KNL] | |
4212 | Set number of RCU writers. The values operate | |
4213 | the same as for rcuperf.nreaders. | |
4214 | N, where N is the number of CPUs | |
4215 | ||
820687a7 PM |
4216 | rcuperf.perf_type= [KNL] |
4217 | Specify the RCU implementation to test. | |
4218 | ||
bdea9e34 PM |
4219 | rcuperf.shutdown= [KNL] |
4220 | Shut the system down after performance tests | |
4221 | complete. This is useful for hands-off automated | |
4222 | testing. | |
4223 | ||
bdea9e34 PM |
4224 | rcuperf.verbose= [KNL] |
4225 | Enable additional printk() statements. | |
4226 | ||
820687a7 PM |
4227 | rcuperf.writer_holdoff= [KNL] |
4228 | Write-side holdoff between grace periods, | |
4229 | in microseconds. The default of zero says | |
4230 | no holdoff. | |
4231 | ||
4102adab | 4232 | rcutorture.fqs_duration= [KNL] |
21b05de4 PM |
4233 | Set duration of force_quiescent_state bursts |
4234 | in microseconds. | |
dabb8aa9 | 4235 | |
4102adab | 4236 | rcutorture.fqs_holdoff= [KNL] |
21b05de4 PM |
4237 | Set holdoff time within force_quiescent_state bursts |
4238 | in microseconds. | |
dabb8aa9 | 4239 | |
4102adab | 4240 | rcutorture.fqs_stutter= [KNL] |
21b05de4 PM |
4241 | Set wait time between force_quiescent_state bursts |
4242 | in seconds. | |
4243 | ||
ed8f6fb2 PM |
4244 | rcutorture.fwd_progress= [KNL] |
4245 | Enable RCU grace-period forward-progress testing | |
4246 | for the types of RCU supporting this notion. | |
4247 | ||
4248 | rcutorture.fwd_progress_div= [KNL] | |
4249 | Specify the fraction of a CPU-stall-warning | |
4250 | period to do tight-loop forward-progress testing. | |
4251 | ||
4252 | rcutorture.fwd_progress_holdoff= [KNL] | |
4253 | Number of seconds to wait between successive | |
4254 | forward-progress tests. | |
4255 | ||
4256 | rcutorture.fwd_progress_need_resched= [KNL] | |
4257 | Enclose cond_resched() calls within checks for | |
4258 | need_resched() during tight-loop forward-progress | |
4259 | testing. | |
4260 | ||
21b05de4 PM |
4261 | rcutorture.gp_cond= [KNL] |
4262 | Use conditional/asynchronous update-side | |
4263 | primitives, if available. | |
dabb8aa9 | 4264 | |
4102adab | 4265 | rcutorture.gp_exp= [KNL] |
21b05de4 | 4266 | Use expedited update-side primitives, if available. |
4102adab PM |
4267 | |
4268 | rcutorture.gp_normal= [KNL] | |
21b05de4 PM |
4269 | Use normal (non-expedited) asynchronous |
4270 | update-side primitives, if available. | |
4271 | ||
4272 | rcutorture.gp_sync= [KNL] | |
4273 | Use normal (non-expedited) synchronous | |
4274 | update-side primitives, if available. If all | |
4275 | of rcutorture.gp_cond=, rcutorture.gp_exp=, | |
4276 | rcutorture.gp_normal=, and rcutorture.gp_sync= | |
4277 | are zero, rcutorture acts as if is interpreted | |
4278 | they are all non-zero. | |
dabb8aa9 | 4279 | |
4102adab | 4280 | rcutorture.n_barrier_cbs= [KNL] |
dabb8aa9 PM |
4281 | Set callbacks/threads for rcu_barrier() testing. |
4282 | ||
4102adab | 4283 | rcutorture.nfakewriters= [KNL] |
dabb8aa9 PM |
4284 | Set number of concurrent RCU writers. These just |
4285 | stress RCU, they don't participate in the actual | |
4286 | test, hence the "fake". | |
4287 | ||
4102adab | 4288 | rcutorture.nreaders= [KNL] |
3838cc18 PM |
4289 | Set number of RCU readers. The value -1 selects |
4290 | N-1, where N is the number of CPUs. A value | |
4291 | "n" less than -1 selects N-n-2, where N is again | |
4292 | the number of CPUs. For example, -2 selects N | |
4293 | (the number of CPUs), -3 selects N+1, and so on. | |
dabb8aa9 | 4294 | |
4102adab PM |
4295 | rcutorture.object_debug= [KNL] |
4296 | Enable debug-object double-call_rcu() testing. | |
4297 | ||
4298 | rcutorture.onoff_holdoff= [KNL] | |
dabb8aa9 PM |
4299 | Set time (s) after boot for CPU-hotplug testing. |
4300 | ||
4102adab | 4301 | rcutorture.onoff_interval= [KNL] |
028be12b PM |
4302 | Set time (jiffies) between CPU-hotplug operations, |
4303 | or zero to disable CPU-hotplug testing. | |
dabb8aa9 | 4304 | |
4a5f133c PM |
4305 | rcutorture.read_exit= [KNL] |
4306 | Set the number of read-then-exit kthreads used | |
4307 | to test the interaction of RCU updaters and | |
4308 | task-exit processing. | |
4309 | ||
4310 | rcutorture.read_exit_burst= [KNL] | |
4311 | The number of times in a given read-then-exit | |
4312 | episode that a set of read-then-exit kthreads | |
4313 | is spawned. | |
4314 | ||
4315 | rcutorture.read_exit_delay= [KNL] | |
4316 | The delay, in seconds, between successive | |
4317 | read-then-exit testing episodes. | |
4318 | ||
4102adab | 4319 | rcutorture.shuffle_interval= [KNL] |
dabb8aa9 PM |
4320 | Set task-shuffle interval (s). Shuffling tasks |
4321 | allows some CPUs to go into dyntick-idle mode | |
4322 | during the rcutorture test. | |
4323 | ||
4102adab | 4324 | rcutorture.shutdown_secs= [KNL] |
dabb8aa9 PM |
4325 | Set time (s) after boot system shutdown. This |
4326 | is useful for hands-off automated testing. | |
4327 | ||
4102adab | 4328 | rcutorture.stall_cpu= [KNL] |
dabb8aa9 PM |
4329 | Duration of CPU stall (s) to test RCU CPU stall |
4330 | warnings, zero to disable. | |
4331 | ||
19a8ff95 PM |
4332 | rcutorture.stall_cpu_block= [KNL] |
4333 | Sleep while stalling if set. This will result | |
4334 | in warnings from preemptible RCU in addition | |
4335 | to any other stall-related activity. | |
4336 | ||
4102adab | 4337 | rcutorture.stall_cpu_holdoff= [KNL] |
dabb8aa9 PM |
4338 | Time to wait (s) after boot before inducing stall. |
4339 | ||
2b1516e5 PM |
4340 | rcutorture.stall_cpu_irqsoff= [KNL] |
4341 | Disable interrupts while stalling if set. | |
4342 | ||
55b2dcf5 PM |
4343 | rcutorture.stall_gp_kthread= [KNL] |
4344 | Duration (s) of forced sleep within RCU | |
4345 | grace-period kthread to test RCU CPU stall | |
4346 | warnings, zero to disable. If both stall_cpu | |
4347 | and stall_gp_kthread are specified, the | |
4348 | kthread is starved first, then the CPU. | |
4349 | ||
4102adab | 4350 | rcutorture.stat_interval= [KNL] |
dabb8aa9 PM |
4351 | Time (s) between statistics printk()s. |
4352 | ||
4102adab | 4353 | rcutorture.stutter= [KNL] |
dabb8aa9 PM |
4354 | Time (s) to stutter testing, for example, specifying |
4355 | five seconds causes the test to run for five seconds, | |
4356 | wait for five seconds, and so on. This tests RCU's | |
4357 | ability to transition abruptly to and from idle. | |
4358 | ||
4102adab | 4359 | rcutorture.test_boost= [KNL] |
dabb8aa9 PM |
4360 | Test RCU priority boosting? 0=no, 1=maybe, 2=yes. |
4361 | "Maybe" means test if the RCU implementation | |
4362 | under test support RCU priority boosting. | |
4363 | ||
4102adab | 4364 | rcutorture.test_boost_duration= [KNL] |
dabb8aa9 PM |
4365 | Duration (s) of each individual boost test. |
4366 | ||
4102adab | 4367 | rcutorture.test_boost_interval= [KNL] |
dabb8aa9 PM |
4368 | Interval (s) between each boost test. |
4369 | ||
4102adab | 4370 | rcutorture.test_no_idle_hz= [KNL] |
dabb8aa9 PM |
4371 | Test RCU's dyntick-idle handling. See also the |
4372 | rcutorture.shuffle_interval parameter. | |
4373 | ||
4102adab | 4374 | rcutorture.torture_type= [KNL] |
dabb8aa9 PM |
4375 | Specify the RCU implementation to test. |
4376 | ||
4102adab | 4377 | rcutorture.verbose= [KNL] |
dabb8aa9 PM |
4378 | Enable additional printk() statements. |
4379 | ||
cdc694b2 PM |
4380 | rcupdate.rcu_cpu_stall_ftrace_dump= [KNL] |
4381 | Dump ftrace buffer after reporting RCU CPU | |
4382 | stall warning. | |
4383 | ||
5a9be7c6 PM |
4384 | rcupdate.rcu_cpu_stall_suppress= [KNL] |
4385 | Suppress RCU CPU stall warning messages. | |
4386 | ||
58c53360 PM |
4387 | rcupdate.rcu_cpu_stall_suppress_at_boot= [KNL] |
4388 | Suppress RCU CPU stall warning messages and | |
4389 | rcutorture writer stall warnings that occur | |
4390 | during early boot, that is, during the time | |
4391 | before the init task is spawned. | |
4392 | ||
5a9be7c6 PM |
4393 | rcupdate.rcu_cpu_stall_timeout= [KNL] |
4394 | Set timeout for RCU CPU stall warning messages. | |
4395 | ||
4102adab PM |
4396 | rcupdate.rcu_expedited= [KNL] |
4397 | Use expedited grace-period primitives, for | |
4398 | example, synchronize_rcu_expedited() instead | |
4399 | of synchronize_rcu(). This reduces latency, | |
4400 | but can increase CPU utilization, degrade | |
4401 | real-time latency, and degrade energy efficiency. | |
79cfea02 | 4402 | No effect on CONFIG_TINY_RCU kernels. |
4102adab | 4403 | |
5a9be7c6 PM |
4404 | rcupdate.rcu_normal= [KNL] |
4405 | Use only normal grace-period primitives, | |
4406 | for example, synchronize_rcu() instead of | |
4407 | synchronize_rcu_expedited(). This improves | |
79cfea02 PM |
4408 | real-time latency, CPU utilization, and |
4409 | energy efficiency, but can expose users to | |
4410 | increased grace-period latency. This parameter | |
4411 | overrides rcupdate.rcu_expedited. No effect on | |
4412 | CONFIG_TINY_RCU kernels. | |
4102adab | 4413 | |
3e42ec1a PM |
4414 | rcupdate.rcu_normal_after_boot= [KNL] |
4415 | Once boot has completed (that is, after | |
4416 | rcu_end_inkernel_boot() has been invoked), use | |
79cfea02 PM |
4417 | only normal grace-period primitives. No effect |
4418 | on CONFIG_TINY_RCU kernels. | |
3e42ec1a | 4419 | |
b0afa0f0 PM |
4420 | rcupdate.rcu_task_ipi_delay= [KNL] |
4421 | Set time in jiffies during which RCU tasks will | |
4422 | avoid sending IPIs, starting with the beginning | |
4423 | of a given grace period. Setting a large | |
4424 | number avoids disturbing real-time workloads, | |
4425 | but lengthens grace periods. | |
4426 | ||
52db30ab PM |
4427 | rcupdate.rcu_task_stall_timeout= [KNL] |
4428 | Set timeout in jiffies for RCU task stall warning | |
4429 | messages. Disable with a value less than or equal | |
4430 | to zero. | |
4431 | ||
74860fee PK |
4432 | rcupdate.rcu_self_test= [KNL] |
4433 | Run the RCU early boot self tests | |
4434 | ||
ffdfc409 OJ |
4435 | rdinit= [KNL] |
4436 | Format: <full_path> | |
4437 | Run specified binary instead of /init from the ramdisk, | |
4438 | used for early userspace startup. See initrd. | |
4439 | ||
c49a0a80 TL |
4440 | rdrand= [X86] |
4441 | force - Override the decision by the kernel to hide the | |
4442 | advertisement of RDRAND support (this affects | |
4443 | certain AMD processors because of buggy BIOS | |
4444 | support, specifically around the suspend/resume | |
4445 | path). | |
4446 | ||
1d9807fc TL |
4447 | rdt= [HW,X86,RDT] |
4448 | Turn on/off individual RDT features. List is: | |
31516de3 FY |
4449 | cmt, mbmtotal, mbmlocal, l3cat, l3cdp, l2cat, l2cdp, |
4450 | mba. | |
1d9807fc TL |
4451 | E.g. to turn on cmt and turn off mba use: |
4452 | rdt=cmt,!mba | |
4453 | ||
1b3a5d02 RH |
4454 | reboot= [KNL] |
4455 | Format (x86 or x86_64): | |
4456 | [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] \ | |
4457 | [[,]s[mp]#### \ | |
4458 | [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \ | |
4459 | [[,]f[orce] | |
b287a25a AK |
4460 | Where reboot_mode is one of warm (soft) or cold (hard) or gpio |
4461 | (prefix with 'panic_' to set mode for panic | |
4462 | reboot only), | |
1b3a5d02 RH |
4463 | reboot_type is one of bios, acpi, kbd, triple, efi, or pci, |
4464 | reboot_force is either force or not specified, | |
4465 | reboot_cpu is s[mp]#### with #### being the processor | |
4466 | to be used for rebooting. | |
1da177e4 | 4467 | |
1fbeb3a8 | 4468 | refscale.holdoff= [KNL] |
847dd70a PM |
4469 | Set test-start holdoff period. The purpose of |
4470 | this parameter is to delay the start of the | |
4471 | test until boot completes in order to avoid | |
4472 | interference. | |
4473 | ||
1fbeb3a8 | 4474 | refscale.loops= [KNL] |
847dd70a PM |
4475 | Set the number of loops over the synchronization |
4476 | primitive under test. Increasing this number | |
4477 | reduces noise due to loop start/end overhead, | |
4478 | but the default has already reduced the per-pass | |
4479 | noise to a handful of picoseconds on ca. 2020 | |
4480 | x86 laptops. | |
4481 | ||
1fbeb3a8 | 4482 | refscale.nreaders= [KNL] |
847dd70a PM |
4483 | Set number of readers. The default value of -1 |
4484 | selects N, where N is roughly 75% of the number | |
4485 | of CPUs. A value of zero is an interesting choice. | |
4486 | ||
1fbeb3a8 | 4487 | refscale.nruns= [KNL] |
847dd70a PM |
4488 | Set number of runs, each of which is dumped onto |
4489 | the console log. | |
4490 | ||
1fbeb3a8 | 4491 | refscale.readdelay= [KNL] |
847dd70a PM |
4492 | Set the read-side critical-section duration, |
4493 | measured in microseconds. | |
4494 | ||
1fbeb3a8 PM |
4495 | refscale.scale_type= [KNL] |
4496 | Specify the read-protection implementation to test. | |
4497 | ||
4498 | refscale.shutdown= [KNL] | |
847dd70a PM |
4499 | Shut down the system at the end of the performance |
4500 | test. This defaults to 1 (shut it down) when | |
4501 | rcuperf is built into the kernel and to 0 (leave | |
4502 | it running) when rcuperf is built as a module. | |
4503 | ||
1fbeb3a8 | 4504 | refscale.verbose= [KNL] |
847dd70a PM |
4505 | Enable additional printk() statements. |
4506 | ||
46b6d94e PJ |
4507 | relax_domain_level= |
4508 | [KNL, SMP] Set scheduler's default relax_domain_level. | |
da82c92f | 4509 | See Documentation/admin-guide/cgroup-v1/cpusets.rst. |
46b6d94e | 4510 | |
ffd2e8df BH |
4511 | reserve= [KNL,BUGS] Force kernel to ignore I/O ports or memory |
4512 | Format: <base1>,<size1>[,<base2>,<size2>,...] | |
4513 | Reserve I/O ports or memory so the kernel won't use | |
4514 | them. If <base> is less than 0x10000, the region | |
4515 | is assumed to be I/O ports; otherwise it is memory. | |
1da177e4 | 4516 | |
cd4f0ef7 | 4517 | reservetop= [X86-32] |
461a9aff ZA |
4518 | Format: nn[KMG] |
4519 | Reserves a hole at the top of the kernel virtual | |
4520 | address space. | |
4521 | ||
9ea77bdb PA |
4522 | reservelow= [X86] |
4523 | Format: nn[K] | |
4524 | Set the amount of memory to reserve for BIOS at | |
4525 | the bottom of the address space. | |
4526 | ||
7e96287d VG |
4527 | reset_devices [KNL] Force drivers to reset the underlying device |
4528 | during initialization. | |
4529 | ||
a9913044 RD |
4530 | resume= [SWSUSP] |
4531 | Specify the partition device for software suspend | |
2df83fa4 MB |
4532 | Format: |
4533 | {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>} | |
1da177e4 | 4534 | |
ecbd0da1 RW |
4535 | resume_offset= [SWSUSP] |
4536 | Specify the offset from the beginning of the partition | |
4537 | given by "resume=" at which the swap header is located, | |
4538 | in <PAGE_SIZE> units (needed only for swap files). | |
151f4e2b | 4539 | See Documentation/power/swsusp-and-swap-files.rst |
ecbd0da1 | 4540 | |
f126f733 BS |
4541 | resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to |
4542 | read the resume files | |
4543 | ||
6f8d7022 BS |
4544 | resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up. |
4545 | Useful for devices that are detected asynchronously | |
4546 | (e.g. USB and MMC devices). | |
4547 | ||
f996fc96 BS |
4548 | hibernate= [HIBERNATION] |
4549 | noresume Don't check if there's a hibernation image | |
4550 | present during boot. | |
4551 | nocompress Don't compress/decompress hibernation images. | |
a6e15a39 | 4552 | no Disable hibernation and resume. |
4c0b6c10 RW |
4553 | protect_image Turn on image protection during restoration |
4554 | (that will set all pages holding image data | |
4555 | during restoration read-only). | |
f996fc96 | 4556 | |
0a7b35cb MN |
4557 | retain_initrd [RAM] Keep initrd memory after extraction |
4558 | ||
0efbb786 AC |
4559 | rfkill.default_state= |
4560 | 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm, | |
4561 | etc. communication is blocked by default. | |
4562 | 1 Unblocked. | |
4563 | ||
4564 | rfkill.master_switch_mode= | |
4565 | 0 The "airplane mode" button does nothing. | |
4566 | 1 The "airplane mode" button toggles between everything | |
4567 | blocked and the previous configuration. | |
4568 | 2 The "airplane mode" button toggles between everything | |
4569 | blocked and everything unblocked. | |
4570 | ||
1da177e4 LT |
4571 | rhash_entries= [KNL,NET] |
4572 | Set number of hash buckets for route cache | |
4573 | ||
e16fd002 GA |
4574 | ring3mwait=disable |
4575 | [KNL] Disable ring 3 MONITOR/MWAIT feature on supported | |
4576 | CPUs. | |
4577 | ||
1da177e4 LT |
4578 | ro [KNL] Mount root device read-only on boot |
4579 | ||
d2aa1aca KC |
4580 | rodata= [KNL] |
4581 | on Mark read-only kernel memory as read-only (default). | |
4582 | off Leave read-only kernel memory writable for debugging. | |
4583 | ||
605df8af HS |
4584 | rockchip.usb_uart |
4585 | Enable the uart passthrough on the designated usb port | |
4586 | on Rockchip SoCs. When active, the signals of the | |
4587 | debug-uart get routed to the D+ and D- pins of the usb | |
4588 | port and the regular usb controller gets disabled. | |
4589 | ||
1da177e4 | 4590 | root= [KNL] Root filesystem |
f2d34fd9 | 4591 | See name_to_dev_t comment in init/do_mounts.c. |
1da177e4 LT |
4592 | |
4593 | rootdelay= [KNL] Delay (in seconds) to pause before attempting to | |
4594 | mount the root filesystem | |
4595 | ||
4596 | rootflags= [KNL] Set root filesystem mount option string | |
4597 | ||
4598 | rootfstype= [KNL] Set root filesystem type | |
4599 | ||
cc1ed754 PO |
4600 | rootwait [KNL] Wait (indefinitely) for root device to show up. |
4601 | Useful for devices that are detected asynchronously | |
4602 | (e.g. USB and MMC devices). | |
4603 | ||
5c71d618 RT |
4604 | rproc_mem=nn[KMG][@address] |
4605 | [KNL,ARM,CMA] Remoteproc physical memory block. | |
4606 | Memory area to be used by remote processor image, | |
4607 | managed by CMA. | |
4608 | ||
1da177e4 LT |
4609 | rw [KNL] Mount root device read-write on boot |
4610 | ||
4611 | S [KNL] Run init in single mode | |
4612 | ||
c60d1ae4 GS |
4613 | s390_iommu= [HW,S390] |
4614 | Set s390 IOTLB flushing mode | |
4615 | strict | |
4616 | With strict flushing every unmap operation will result in | |
4617 | an IOTLB flush. Default is lazy flushing before reuse, | |
4618 | which is faster. | |
4619 | ||
1da177e4 LT |
4620 | sa1100ir [NET] |
4621 | See drivers/net/irda/sa1100_ir.c. | |
4622 | ||
1da177e4 | 4623 | sbni= [NET] Granch SBNI12 leased line adapter |
a9913044 | 4624 | |
f6630114 MT |
4625 | sched_debug [KNL] Enables verbose scheduler debug messages. |
4626 | ||
cb251765 MG |
4627 | schedstats= [KNL,X86] Enable or disable scheduled statistics. |
4628 | Allowed values are enable and disable. This feature | |
4629 | incurs a small amount of overhead in the scheduler | |
4630 | but is useful for debugging and performance tuning. | |
f6630114 | 4631 | |
05289b90 TG |
4632 | sched_thermal_decay_shift= |
4633 | [KNL, SMP] Set a decay shift for scheduler thermal | |
4634 | pressure signal. Thermal pressure signal follows the | |
4635 | default decay period of other scheduler pelt | |
4636 | signals(usually 32 ms but configurable). Setting | |
4637 | sched_thermal_decay_shift will left shift the decay | |
4638 | period for the thermal pressure signal by the shift | |
4639 | value. | |
4640 | i.e. with the default pelt decay period of 32 ms | |
4641 | sched_thermal_decay_shift thermal pressure decay pr | |
4642 | 1 64 ms | |
4643 | 2 128 ms | |
4644 | and so on. | |
4645 | Format: integer between 0 and 10 | |
4646 | Default is 0. | |
4647 | ||
5307c955 MG |
4648 | skew_tick= [KNL] Offset the periodic timer tick per cpu to mitigate |
4649 | xtime_lock contention on larger systems, and/or RCU lock | |
4650 | contention on all systems with CONFIG_MAXSMP set. | |
4651 | Format: { "0" | "1" } | |
4652 | 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1" | |
4653 | 1 -- enable. | |
4654 | Note: increases power consumption, thus should only be | |
4655 | enabled if running jitter sensitive (HPC/RT) workloads. | |
4656 | ||
89a9684e KC |
4657 | security= [SECURITY] Choose a legacy "major" security module to |
4658 | enable at boot. This has been deprecated by the | |
4659 | "lsm=" parameter. | |
0cb55ad2 RD |
4660 | |
4661 | selinux= [SELINUX] Disable or enable SELinux at boot time. | |
1da177e4 LT |
4662 | Format: { "0" | "1" } |
4663 | See security/selinux/Kconfig help text. | |
4664 | 0 -- disable. | |
4665 | 1 -- enable. | |
d41415eb | 4666 | Default value is 1. |
1da177e4 | 4667 | |
c1c124e9 JJ |
4668 | apparmor= [APPARMOR] Disable or enable AppArmor at boot time |
4669 | Format: { "0" | "1" } | |
4670 | See security/apparmor/Kconfig help text | |
4671 | 0 -- disable. | |
4672 | 1 -- enable. | |
4673 | Default value is set via kernel config option. | |
4674 | ||
cd4f0ef7 | 4675 | serialnumber [BUGS=X86-32] |
1da177e4 | 4676 | |
1da177e4 LT |
4677 | shapers= [NET] |
4678 | Maximal number of shapers. | |
a9913044 | 4679 | |
1da177e4 LT |
4680 | simeth= [IA-64] |
4681 | simscsi= | |
a9913044 | 4682 | |
1da177e4 LT |
4683 | slram= [HW,MTD] |
4684 | ||
423c929c JK |
4685 | slab_nomerge [MM] |
4686 | Disable merging of slabs with similar size. May be | |
4687 | necessary if there is some reason to distinguish | |
7660a6fd KC |
4688 | allocs to different slabs, especially in hardened |
4689 | environments where the risk of heap overflows and | |
4690 | layout control by attackers can usually be | |
4691 | frustrated by disabling merging. This will reduce | |
4692 | most of the exposure of a heap attack to a single | |
4693 | cache (risks via metadata attacks are mostly | |
4694 | unchanged). Debug options disable merging on their | |
4695 | own. | |
ad56b738 | 4696 | For more information see Documentation/vm/slub.rst. |
423c929c | 4697 | |
3df1cccd DR |
4698 | slab_max_order= [MM, SLAB] |
4699 | Determines the maximum allowed order for slabs. | |
4700 | A high setting may cause OOMs due to memory | |
4701 | fragmentation. Defaults to 1 for systems with | |
4702 | more than 32MB of RAM, 0 otherwise. | |
4703 | ||
e17f1dfb | 4704 | slub_debug[=options[,slabs][;[options[,slabs]]...] [MM, SLUB] |
f0630fff CL |
4705 | Enabling slub_debug allows one to determine the |
4706 | culprit if slab objects become corrupted. Enabling | |
4707 | slub_debug can create guard zones around objects and | |
4708 | may poison objects when not in use. Also tracks the | |
4709 | last alloc / free. For more information see | |
ad56b738 | 4710 | Documentation/vm/slub.rst. |
c1aee215 | 4711 | |
1663f26d TH |
4712 | slub_memcg_sysfs= [MM, SLUB] |
4713 | Determines whether to enable sysfs directories for | |
4714 | memory cgroup sub-caches. 1 to enable, 0 to disable. | |
4715 | The default is determined by CONFIG_SLUB_MEMCG_SYSFS_ON. | |
4716 | Enabling this can lead to a very high number of debug | |
4717 | directories and files being created under | |
4718 | /sys/kernel/slub. | |
4719 | ||
c1aee215 | 4720 | slub_max_order= [MM, SLUB] |
f0630fff CL |
4721 | Determines the maximum allowed order for slabs. |
4722 | A high setting may cause OOMs due to memory | |
4723 | fragmentation. For more information see | |
ad56b738 | 4724 | Documentation/vm/slub.rst. |
c1aee215 CL |
4725 | |
4726 | slub_min_objects= [MM, SLUB] | |
f0630fff CL |
4727 | The minimum number of objects per slab. SLUB will |
4728 | increase the slab order up to slub_max_order to | |
4729 | generate a sufficiently large slab able to contain | |
4730 | the number of objects indicated. The higher the number | |
4731 | of objects the smaller the overhead of tracking slabs | |
4732 | and the less frequently locks need to be acquired. | |
ad56b738 | 4733 | For more information see Documentation/vm/slub.rst. |
c1aee215 CL |
4734 | |
4735 | slub_min_order= [MM, SLUB] | |
24775d65 | 4736 | Determines the minimum page order for slabs. Must be |
f0630fff | 4737 | lower than slub_max_order. |
ad56b738 | 4738 | For more information see Documentation/vm/slub.rst. |
c1aee215 CL |
4739 | |
4740 | slub_nomerge [MM, SLUB] | |
423c929c JK |
4741 | Same with slab_nomerge. This is supported for legacy. |
4742 | See slab_nomerge for more information. | |
c1aee215 | 4743 | |
1da177e4 LT |
4744 | smart2= [HW] |
4745 | Format: <io1>[,<io2>[,...,<io8>]] | |
4746 | ||
d0d4f69b BH |
4747 | smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices |
4748 | smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port | |
4749 | smsc-ircc2.ircc_sir= [HW] SIR base I/O port | |
4750 | smsc-ircc2.ircc_fir= [HW] FIR base I/O port | |
4751 | smsc-ircc2.ircc_irq= [HW] IRQ line | |
4752 | smsc-ircc2.ircc_dma= [HW] DMA channel | |
4753 | smsc-ircc2.ircc_transceiver= [HW] Transceiver type: | |
4754 | 0: Toshiba Satellite 1800 (GP data pin select) | |
4755 | 1: Fast pin select (default) | |
4756 | 2: ATC IRMode | |
4757 | ||
52c48c51 SS |
4758 | smt [KNL,S390] Set the maximum number of threads (logical |
4759 | CPUs) to use per physical CPU on systems capable of | |
4760 | symmetric multithreading (SMT). Will be capped to the | |
4761 | actual hardware limit. | |
4762 | Format: <integer> | |
4763 | Default: -1 (no limit) | |
4764 | ||
9c44bc03 IM |
4765 | softlockup_panic= |
4766 | [KNL] Should the soft-lockup detector generate panics. | |
f117955a | 4767 | Format: 0 | 1 |
9c44bc03 | 4768 | |
f117955a | 4769 | A value of 1 instructs the soft-lockup detector |
0a07bef6 GP |
4770 | to panic the machine when a soft-lockup occurs. It is |
4771 | also controlled by the kernel.softlockup_panic sysctl | |
4772 | and CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC, which is the | |
4773 | respective build-time switch to that functionality. | |
3ce62385 | 4774 | |
ed235875 AT |
4775 | softlockup_all_cpu_backtrace= |
4776 | [KNL] Should the soft-lockup detector generate | |
4777 | backtraces on all cpus. | |
f117955a | 4778 | Format: 0 | 1 |
ed235875 | 4779 | |
1da177e4 | 4780 | sonypi.*= [HW] Sony Programmable I/O Control Device driver |
9e1cbede | 4781 | See Documentation/admin-guide/laptops/sonypi.rst |
1da177e4 | 4782 | |
da285121 DW |
4783 | spectre_v2= [X86] Control mitigation of Spectre variant 2 |
4784 | (indirect branch speculation) vulnerability. | |
fa1202ef TG |
4785 | The default operation protects the kernel from |
4786 | user space attacks. | |
da285121 | 4787 | |
fa1202ef TG |
4788 | on - unconditionally enable, implies |
4789 | spectre_v2_user=on | |
4790 | off - unconditionally disable, implies | |
4791 | spectre_v2_user=off | |
da285121 DW |
4792 | auto - kernel detects whether your CPU model is |
4793 | vulnerable | |
4794 | ||
4795 | Selecting 'on' will, and 'auto' may, choose a | |
4796 | mitigation method at run time according to the | |
4797 | CPU, the available microcode, the setting of the | |
4798 | CONFIG_RETPOLINE configuration option, and the | |
4799 | compiler with which the kernel was built. | |
4800 | ||
fa1202ef TG |
4801 | Selecting 'on' will also enable the mitigation |
4802 | against user space to user space task attacks. | |
4803 | ||
4804 | Selecting 'off' will disable both the kernel and | |
4805 | the user space protections. | |
4806 | ||
da285121 DW |
4807 | Specific mitigations can also be selected manually: |
4808 | ||
4809 | retpoline - replace indirect branches | |
4810 | retpoline,generic - google's original retpoline | |
4811 | retpoline,amd - AMD-specific minimal thunk | |
4812 | ||
4813 | Not specifying this option is equivalent to | |
4814 | spectre_v2=auto. | |
4815 | ||
fa1202ef TG |
4816 | spectre_v2_user= |
4817 | [X86] Control mitigation of Spectre variant 2 | |
4818 | (indirect branch speculation) vulnerability between | |
4819 | user space tasks | |
4820 | ||
4821 | on - Unconditionally enable mitigations. Is | |
4822 | enforced by spectre_v2=on | |
4823 | ||
4824 | off - Unconditionally disable mitigations. Is | |
4825 | enforced by spectre_v2=off | |
4826 | ||
7cc765a6 TG |
4827 | prctl - Indirect branch speculation is enabled, |
4828 | but mitigation can be enabled via prctl | |
4829 | per thread. The mitigation control state | |
4830 | is inherited on fork. | |
4831 | ||
55a97402 TG |
4832 | prctl,ibpb |
4833 | - Like "prctl" above, but only STIBP is | |
4834 | controlled per thread. IBPB is issued | |
4835 | always when switching between different user | |
4836 | space processes. | |
4837 | ||
6b3e64c2 TG |
4838 | seccomp |
4839 | - Same as "prctl" above, but all seccomp | |
4840 | threads will enable the mitigation unless | |
4841 | they explicitly opt out. | |
4842 | ||
55a97402 TG |
4843 | seccomp,ibpb |
4844 | - Like "seccomp" above, but only STIBP is | |
4845 | controlled per thread. IBPB is issued | |
4846 | always when switching between different | |
4847 | user space processes. | |
4848 | ||
fa1202ef TG |
4849 | auto - Kernel selects the mitigation depending on |
4850 | the available CPU features and vulnerability. | |
6b3e64c2 TG |
4851 | |
4852 | Default mitigation: | |
4853 | If CONFIG_SECCOMP=y then "seccomp", otherwise "prctl" | |
fa1202ef TG |
4854 | |
4855 | Not specifying this option is equivalent to | |
4856 | spectre_v2_user=auto. | |
4857 | ||
24f7fc83 KRW |
4858 | spec_store_bypass_disable= |
4859 | [HW] Control Speculative Store Bypass (SSB) Disable mitigation | |
4860 | (Speculative Store Bypass vulnerability) | |
4861 | ||
4862 | Certain CPUs are vulnerable to an exploit against a | |
4863 | a common industry wide performance optimization known | |
4864 | as "Speculative Store Bypass" in which recent stores | |
4865 | to the same memory location may not be observed by | |
4866 | later loads during speculative execution. The idea | |
4867 | is that such stores are unlikely and that they can | |
4868 | be detected prior to instruction retirement at the | |
4869 | end of a particular speculation execution window. | |
4870 | ||
4871 | In vulnerable processors, the speculatively forwarded | |
4872 | store can be used in a cache side channel attack, for | |
4873 | example to read memory to which the attacker does not | |
4874 | directly have access (e.g. inside sandboxed code). | |
4875 | ||
4876 | This parameter controls whether the Speculative Store | |
4877 | Bypass optimization is used. | |
4878 | ||
6b4c1360 ME |
4879 | On x86 the options are: |
4880 | ||
f21b53b2 KC |
4881 | on - Unconditionally disable Speculative Store Bypass |
4882 | off - Unconditionally enable Speculative Store Bypass | |
4883 | auto - Kernel detects whether the CPU model contains an | |
4884 | implementation of Speculative Store Bypass and | |
4885 | picks the most appropriate mitigation. If the | |
4886 | CPU is not vulnerable, "off" is selected. If the | |
4887 | CPU is vulnerable the default mitigation is | |
4888 | architecture and Kconfig dependent. See below. | |
4889 | prctl - Control Speculative Store Bypass per thread | |
4890 | via prctl. Speculative Store Bypass is enabled | |
4891 | for a process by default. The state of the control | |
4892 | is inherited on fork. | |
4893 | seccomp - Same as "prctl" above, but all seccomp threads | |
4894 | will disable SSB unless they explicitly opt out. | |
24f7fc83 | 4895 | |
f21b53b2 KC |
4896 | Default mitigations: |
4897 | X86: If CONFIG_SECCOMP=y "seccomp", otherwise "prctl" | |
4898 | ||
6b4c1360 ME |
4899 | On powerpc the options are: |
4900 | ||
4901 | on,auto - On Power8 and Power9 insert a store-forwarding | |
4902 | barrier on kernel entry and exit. On Power7 | |
4903 | perform a software flush on kernel entry and | |
4904 | exit. | |
4905 | off - No action. | |
4906 | ||
4907 | Not specifying this option is equivalent to | |
4908 | spec_store_bypass_disable=auto. | |
4909 | ||
1da177e4 LT |
4910 | spia_io_base= [HW,MTD] |
4911 | spia_fio_base= | |
4912 | spia_pedr= | |
4913 | spia_peddr= | |
4914 | ||
6650cdd9 PZI |
4915 | split_lock_detect= |
4916 | [X86] Enable split lock detection | |
4917 | ||
4918 | When enabled (and if hardware support is present), atomic | |
4919 | instructions that access data across cache line | |
4920 | boundaries will result in an alignment check exception. | |
4921 | ||
4922 | off - not enabled | |
4923 | ||
4924 | warn - the kernel will emit rate limited warnings | |
4925 | about applications triggering the #AC | |
4926 | exception. This mode is the default on CPUs | |
4927 | that supports split lock detection. | |
4928 | ||
4929 | fatal - the kernel will send SIGBUS to applications | |
4930 | that trigger the #AC exception. | |
4931 | ||
4932 | If an #AC exception is hit in the kernel or in | |
4933 | firmware (i.e. not while executing in user mode) | |
4934 | the kernel will oops in either "warn" or "fatal" | |
4935 | mode. | |
4936 | ||
7e5b3c26 MG |
4937 | srbds= [X86,INTEL] |
4938 | Control the Special Register Buffer Data Sampling | |
4939 | (SRBDS) mitigation. | |
4940 | ||
4941 | Certain CPUs are vulnerable to an MDS-like | |
4942 | exploit which can leak bits from the random | |
4943 | number generator. | |
4944 | ||
4945 | By default, this issue is mitigated by | |
4946 | microcode. However, the microcode fix can cause | |
4947 | the RDRAND and RDSEED instructions to become | |
4948 | much slower. Among other effects, this will | |
4949 | result in reduced throughput from /dev/urandom. | |
4950 | ||
4951 | The microcode mitigation can be disabled with | |
4952 | the following option: | |
4953 | ||
4954 | off: Disable mitigation and remove | |
4955 | performance impact to RDRAND and RDSEED | |
4956 | ||
c350c008 PM |
4957 | srcutree.counter_wrap_check [KNL] |
4958 | Specifies how frequently to check for | |
4959 | grace-period sequence counter wrap for the | |
4960 | srcu_data structure's ->srcu_gp_seq_needed field. | |
4961 | The greater the number of bits set in this kernel | |
4962 | parameter, the less frequently counter wrap will | |
4963 | be checked for. Note that the bottom two bits | |
4964 | are ignored. | |
4965 | ||
22607d66 PM |
4966 | srcutree.exp_holdoff [KNL] |
4967 | Specifies how many nanoseconds must elapse | |
4968 | since the end of the last SRCU grace period for | |
4969 | a given srcu_struct until the next normal SRCU | |
4970 | grace period will be considered for automatic | |
4971 | expediting. Set to zero to disable automatic | |
4972 | expediting. | |
4973 | ||
a43ae4df MZ |
4974 | ssbd= [ARM64,HW] |
4975 | Speculative Store Bypass Disable control | |
4976 | ||
4977 | On CPUs that are vulnerable to the Speculative | |
4978 | Store Bypass vulnerability and offer a | |
4979 | firmware based mitigation, this parameter | |
4980 | indicates how the mitigation should be used: | |
4981 | ||
4982 | force-on: Unconditionally enable mitigation for | |
4983 | for both kernel and userspace | |
4984 | force-off: Unconditionally disable mitigation for | |
4985 | for both kernel and userspace | |
4986 | kernel: Always enable mitigation in the | |
4987 | kernel, and offer a prctl interface | |
4988 | to allow userspace to register its | |
4989 | interest in being mitigated too. | |
4990 | ||
1be7107f HD |
4991 | stack_guard_gap= [MM] |
4992 | override the default stack gap protection. The value | |
4993 | is in page units and it defines how many pages prior | |
4994 | to (for stacks growing down) resp. after (for stacks | |
4995 | growing up) the main stack are reserved for no other | |
4996 | mapping. Default value is 256 pages. | |
4997 | ||
f38f1d2a SR |
4998 | stacktrace [FTRACE] |
4999 | Enabled the stack tracer on boot up. | |
5000 | ||
762e1207 SR |
5001 | stacktrace_filter=[function-list] |
5002 | [FTRACE] Limit the functions that the stack tracer | |
5003 | will trace at boot up. function-list is a comma separated | |
5004 | list of functions. This list can be changed at run | |
5005 | time by the stack_trace_filter file in the debugfs | |
5006 | tracing directory. Note, this enables stack tracing | |
5007 | and the stacktrace above is not needed. | |
5008 | ||
1da177e4 LT |
5009 | sti= [PARISC,HW] |
5010 | Format: <num> | |
5011 | Set the STI (builtin display/keyboard on the HP-PARISC | |
5012 | machines) console (graphic card) which should be used | |
5013 | as the initial boot-console. | |
5014 | See also comment in drivers/video/console/sticore.c. | |
5015 | ||
5016 | sti_font= [HW] | |
5017 | See comment in drivers/video/console/sticore.c. | |
5018 | ||
5019 | stifb= [HW] | |
5020 | Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]] | |
5021 | ||
cbf11071 TM |
5022 | sunrpc.min_resvport= |
5023 | sunrpc.max_resvport= | |
5024 | [NFS,SUNRPC] | |
5025 | SunRPC servers often require that client requests | |
5026 | originate from a privileged port (i.e. a port in the | |
5027 | range 0 < portnr < 1024). | |
5028 | An administrator who wishes to reserve some of these | |
5029 | ports for other uses may adjust the range that the | |
5030 | kernel's sunrpc client considers to be privileged | |
5031 | using these two parameters to set the minimum and | |
5032 | maximum port values. | |
5033 | ||
ff3ac5c3 TM |
5034 | sunrpc.svc_rpc_per_connection_limit= |
5035 | [NFS,SUNRPC] | |
5036 | Limit the number of requests that the server will | |
5037 | process in parallel from a single connection. | |
5038 | The default value is 0 (no limit). | |
5039 | ||
42a7fc4a GB |
5040 | sunrpc.pool_mode= |
5041 | [NFS] | |
5042 | Control how the NFS server code allocates CPUs to | |
5043 | service thread pools. Depending on how many NICs | |
5044 | you have and where their interrupts are bound, this | |
5045 | option will affect which CPUs will do NFS serving. | |
5046 | Note: this parameter cannot be changed while the | |
5047 | NFS server is running. | |
5048 | ||
5049 | auto the server chooses an appropriate mode | |
5050 | automatically using heuristics | |
5051 | global a single global pool contains all CPUs | |
5052 | percpu one pool for each CPU | |
5053 | pernode one pool for each NUMA node (equivalent | |
5054 | to global on non-NUMA machines) | |
5055 | ||
cbf11071 TM |
5056 | sunrpc.tcp_slot_table_entries= |
5057 | sunrpc.udp_slot_table_entries= | |
5058 | [NFS,SUNRPC] | |
5059 | Sets the upper limit on the number of simultaneous | |
5060 | RPC calls that can be sent from the client to a | |
5061 | server. Increasing these values may allow you to | |
5062 | improve throughput, but will also increase the | |
5063 | amount of memory reserved for use by the client. | |
5064 | ||
1d4a9c17 BN |
5065 | suspend.pm_test_delay= |
5066 | [SUSPEND] | |
5067 | Sets the number of seconds to remain in a suspend test | |
5068 | mode before resuming the system (see | |
5069 | /sys/power/pm_test). Only available when CONFIG_PM_DEBUG | |
5070 | is set. Default value is 5. | |
5071 | ||
6a9c930b RP |
5072 | svm= [PPC] |
5073 | Format: { on | off | y | n | 1 | 0 } | |
5074 | This parameter controls use of the Protected | |
5075 | Execution Facility on pSeries. | |
5076 | ||
07555ac1 | 5077 | swapaccount=[0|1] |
a42c390c MH |
5078 | [KNL] Enable accounting of swap in memory resource |
5079 | controller if no parameter or 1 is given or disable | |
da82c92f | 5080 | it if 0 is given (See Documentation/admin-guide/cgroup-v1/memory.rst) |
a42c390c | 5081 | |
91fec0f5 | 5082 | swiotlb= [ARM,IA-64,PPC,MIPS,X86] |
fff5d992 | 5083 | Format: { <int> | force | noforce } |
91fec0f5 JK |
5084 | <int> -- Number of I/O TLB slabs |
5085 | force -- force using of bounce buffers even if they | |
5086 | wouldn't be automatically used by the kernel | |
fff5d992 | 5087 | noforce -- Never use bounce buffers (for debugging) |
a9913044 | 5088 | |
1da177e4 LT |
5089 | switches= [HW,M68k] |
5090 | ||
3db978d4 VB |
5091 | sysctl.*= [KNL] |
5092 | Set a sysctl parameter, right before loading the init | |
5093 | process, as if the value was written to the respective | |
5094 | /proc/sys/... file. Both '.' and '/' are recognized as | |
5095 | separators. Unrecognized parameters and invalid values | |
5096 | are reported in the kernel log. Sysctls registered | |
5097 | later by a loaded module cannot be set this way. | |
5098 | Example: sysctl.vm.swappiness=40 | |
5099 | ||
e52eec13 AK |
5100 | sysfs.deprecated=0|1 [KNL] |
5101 | Enable/disable old style sysfs layout for old udev | |
5102 | on older distributions. When this option is enabled | |
5103 | very new udev will not work anymore. When this option | |
5104 | is disabled (or CONFIG_SYSFS_DEPRECATED not compiled) | |
5105 | in older udev will not work anymore. | |
5106 | Default depends on CONFIG_SYSFS_DEPRECATED_V2 set in | |
5107 | the kernel configuration. | |
5108 | ||
5d6f647f IM |
5109 | sysrq_always_enabled |
5110 | [KNL] | |
5111 | Ignore sysrq setting - this boot parameter will | |
5112 | neutralize any effect of /proc/sys/kernel/sysrq. | |
5113 | Useful for debugging. | |
5114 | ||
747029a5 FF |
5115 | tcpmhash_entries= [KNL,NET] |
5116 | Set the number of tcp_metrics_hash slots. | |
5117 | Default value is 8192 or 16384 depending on total | |
5118 | ram pages. This is used to specify the TCP metrics | |
1cec2cac | 5119 | cache size. See Documentation/networking/ip-sysctl.rst |
747029a5 FF |
5120 | "tcp_no_metrics_save" section for more details. |
5121 | ||
1da177e4 LT |
5122 | tdfx= [HW,DRM] |
5123 | ||
acc82342 | 5124 | test_suspend= [SUSPEND][,N] |
77437fd4 | 5125 | Specify "mem" (for Suspend-to-RAM) or "standby" (for |
acc82342 SP |
5126 | standby suspend) or "freeze" (for suspend type freeze) |
5127 | as the system sleep state during system startup with | |
5128 | the optional capability to repeat N number of times. | |
5129 | The system is woken from this state using a | |
5130 | wakeup-capable RTC alarm. | |
77437fd4 | 5131 | |
1da177e4 LT |
5132 | thash_entries= [KNL,NET] |
5133 | Set number of hash buckets for TCP connection | |
5134 | ||
f8707ec9 LB |
5135 | thermal.act= [HW,ACPI] |
5136 | -1: disable all active trip points in all thermal zones | |
5137 | <degrees C>: override all lowest active trip points | |
5138 | ||
c52a7419 LB |
5139 | thermal.crt= [HW,ACPI] |
5140 | -1: disable all critical trip points in all thermal zones | |
22a94d79 | 5141 | <degrees C>: override all critical trip points |
c52a7419 | 5142 | |
f5487145 LB |
5143 | thermal.nocrt= [HW,ACPI] |
5144 | Set to disable actions on ACPI thermal zone | |
5145 | critical and hot trip points. | |
5146 | ||
72b33ef8 LB |
5147 | thermal.off= [HW,ACPI] |
5148 | 1: disable ACPI thermal control | |
5149 | ||
a70cdc52 LB |
5150 | thermal.psv= [HW,ACPI] |
5151 | -1: disable all passive trip points | |
ada9cfdd RD |
5152 | <degrees C>: override all passive trip points to this |
5153 | value | |
a70cdc52 | 5154 | |
730ff34d LB |
5155 | thermal.tzp= [HW,ACPI] |
5156 | Specify global default ACPI thermal zone polling rate | |
5157 | <deci-seconds>: poll all this frequency | |
5158 | 0: no polling (default) | |
5159 | ||
8d32a307 TG |
5160 | threadirqs [KNL] |
5161 | Force threading of all interrupt handlers except those | |
24775d65 | 5162 | marked explicitly IRQF_NO_THREAD. |
8d32a307 | 5163 | |
2b1a61f0 HC |
5164 | topology= [S390] |
5165 | Format: {off | on} | |
5166 | Specify if the kernel should make use of the cpu | |
f65e51d7 SL |
5167 | topology information if the hardware supports this. |
5168 | The scheduler will make use of this information and | |
2b1a61f0 | 5169 | e.g. base its process migration decisions on it. |
c9af3fa9 | 5170 | Default is on. |
2b1a61f0 | 5171 | |
2d73bae1 NA |
5172 | topology_updates= [KNL, PPC, NUMA] |
5173 | Format: {off} | |
5174 | Specify if the kernel should ignore (off) | |
5175 | topology updates sent by the hypervisor to this | |
5176 | LPAR. | |
5177 | ||
8171d3e0 PM |
5178 | torture.disable_onoff_at_boot= [KNL] |
5179 | Prevent the CPU-hotplug component of torturing | |
5180 | until after init has spawned. | |
5181 | ||
2102ad29 PM |
5182 | torture.ftrace_dump_at_shutdown= [KNL] |
5183 | Dump the ftrace buffer at torture-test shutdown, | |
5184 | even if there were no errors. This can be a | |
5185 | very costly operation when many torture tests | |
5186 | are running concurrently, especially on systems | |
5187 | with rotating-rust storage. | |
5188 | ||
1da177e4 LT |
5189 | tp720= [HW,PS2] |
5190 | ||
225a9be2 RA |
5191 | tpm_suspend_pcr=[HW,TPM] |
5192 | Format: integer pcr id | |
5193 | Specify that at suspend time, the tpm driver | |
5194 | should extend the specified pcr with zeros, | |
5195 | as a workaround for some chips which fail to | |
5196 | flush the last written pcr on TPM_SaveState. | |
5197 | This will guarantee that all the other pcrs | |
5198 | are saved. | |
5199 | ||
9d612bef | 5200 | trace_buf_size=nn[KMG] |
3e6fb8e9 | 5201 | [FTRACE] will set tracing buffer size on each cpu. |
631595fb | 5202 | |
020e5f85 LZ |
5203 | trace_event=[event-list] |
5204 | [FTRACE] Set and start specified trace events in order | |
d81749ea BN |
5205 | to facilitate early boot debugging. The event-list is a |
5206 | comma separated list of trace events to enable. See | |
5fb94e9c | 5207 | also Documentation/trace/events.rst |
020e5f85 | 5208 | |
7bcfaf54 SR |
5209 | trace_options=[option-list] |
5210 | [FTRACE] Enable or disable tracer options at boot. | |
5211 | The option-list is a comma delimited list of options | |
5212 | that can be enabled or disabled just as if you were | |
5213 | to echo the option name into | |
5214 | ||
5215 | /sys/kernel/debug/tracing/trace_options | |
5216 | ||
5217 | For example, to enable stacktrace option (to dump the | |
5218 | stack trace of each event), add to the command line: | |
5219 | ||
5220 | trace_options=stacktrace | |
5221 | ||
5fb94e9c | 5222 | See also Documentation/trace/ftrace.rst "trace options" |
7bcfaf54 SR |
5223 | section. |
5224 | ||
0daa2302 SRRH |
5225 | tp_printk[FTRACE] |
5226 | Have the tracepoints sent to printk as well as the | |
5227 | tracing ring buffer. This is useful for early boot up | |
5228 | where the system hangs or reboots and does not give the | |
5229 | option for reading the tracing buffer or performing a | |
5230 | ftrace_dump_on_oops. | |
5231 | ||
5232 | To turn off having tracepoints sent to printk, | |
5233 | echo 0 > /proc/sys/kernel/tracepoint_printk | |
5234 | Note, echoing 1 into this file without the | |
5235 | tracepoint_printk kernel cmdline option has no effect. | |
5236 | ||
5237 | ** CAUTION ** | |
5238 | ||
5239 | Having tracepoints sent to printk() and activating high | |
5240 | frequency tracepoints such as irq or sched, can cause | |
5241 | the system to live lock. | |
5242 | ||
de7edd31 SRRH |
5243 | traceoff_on_warning |
5244 | [FTRACE] enable this option to disable tracing when a | |
5245 | warning is hit. This turns off "tracing_on". Tracing can | |
5246 | be enabled again by echoing '1' into the "tracing_on" | |
5247 | file located in /sys/kernel/debug/tracing/ | |
5248 | ||
5249 | This option is useful, as it disables the trace before | |
5250 | the WARNING dump is called, which prevents the trace to | |
5251 | be filled with content caused by the warning output. | |
5252 | ||
5253 | This option can also be set at run time via the sysctl | |
5254 | option: kernel/traceoff_on_warning | |
5255 | ||
fcf4d821 JK |
5256 | transparent_hugepage= |
5257 | [KNL] | |
5258 | Format: [always|madvise|never] | |
5259 | Can be used to control the default behavior of the system | |
5260 | with respect to transparent hugepages. | |
45c9a74f MR |
5261 | See Documentation/admin-guide/mm/transhuge.rst |
5262 | for more details. | |
fcf4d821 | 5263 | |
d3b8f889 | 5264 | tsc= Disable clocksource stability checks for TSC. |
395628ef AK |
5265 | Format: <string> |
5266 | [x86] reliable: mark tsc clocksource as reliable, this | |
d3b8f889 JS |
5267 | disables clocksource verification at runtime, as well |
5268 | as the stability checks done at bootup. Used to enable | |
5269 | high-resolution timer mode on older hardware, and in | |
5270 | virtualized environment. | |
e82b8e4e VP |
5271 | [x86] noirqtime: Do not use TSC to do irq accounting. |
5272 | Used to run time disable IRQ_TIME_ACCOUNTING on any | |
5273 | platforms where RDTSC is slow and this accounting | |
5274 | can add overhead. | |
6be53520 DL |
5275 | [x86] unstable: mark the TSC clocksource as unstable, this |
5276 | marks the TSC unconditionally unstable at bootup and | |
5277 | avoids any further wobbles once the TSC watchdog notices. | |
0f0b7e1c JL |
5278 | [x86] nowatchdog: disable clocksource watchdog. Used |
5279 | in situations with strict latency requirements (where | |
5280 | interruptions from clocksource watchdog are not | |
5281 | acceptable). | |
395628ef | 5282 | |
bd35c77e KP |
5283 | tsc_early_khz= [X86] Skip early TSC calibration and use the given |
5284 | value instead. Useful when the early TSC frequency discovery | |
5285 | procedure is not reliable, such as on overclocked systems | |
5286 | with CPUID.16h support and partial CPUID.15h support. | |
5287 | Format: <unsigned int> | |
5288 | ||
95c5824f PG |
5289 | tsx= [X86] Control Transactional Synchronization |
5290 | Extensions (TSX) feature in Intel processors that | |
5291 | support TSX control. | |
5292 | ||
5293 | This parameter controls the TSX feature. The options are: | |
5294 | ||
5295 | on - Enable TSX on the system. Although there are | |
5296 | mitigations for all known security vulnerabilities, | |
5297 | TSX has been known to be an accelerator for | |
5298 | several previous speculation-related CVEs, and | |
5299 | so there may be unknown security risks associated | |
5300 | with leaving it enabled. | |
5301 | ||
5302 | off - Disable TSX on the system. (Note that this | |
5303 | option takes effect only on newer CPUs which are | |
5304 | not vulnerable to MDS, i.e., have | |
5305 | MSR_IA32_ARCH_CAPABILITIES.MDS_NO=1 and which get | |
5306 | the new IA32_TSX_CTRL MSR through a microcode | |
5307 | update. This new MSR allows for the reliable | |
5308 | deactivation of the TSX functionality.) | |
5309 | ||
7531a359 PG |
5310 | auto - Disable TSX if X86_BUG_TAA is present, |
5311 | otherwise enable TSX on the system. | |
5312 | ||
95c5824f PG |
5313 | Not specifying this option is equivalent to tsx=off. |
5314 | ||
5315 | See Documentation/admin-guide/hw-vuln/tsx_async_abort.rst | |
5316 | for more details. | |
5317 | ||
a7a248c5 PG |
5318 | tsx_async_abort= [X86,INTEL] Control mitigation for the TSX Async |
5319 | Abort (TAA) vulnerability. | |
5320 | ||
5321 | Similar to Micro-architectural Data Sampling (MDS) | |
5322 | certain CPUs that support Transactional | |
5323 | Synchronization Extensions (TSX) are vulnerable to an | |
5324 | exploit against CPU internal buffers which can forward | |
5325 | information to a disclosure gadget under certain | |
5326 | conditions. | |
5327 | ||
5328 | In vulnerable processors, the speculatively forwarded | |
5329 | data can be used in a cache side channel attack, to | |
5330 | access data to which the attacker does not have direct | |
5331 | access. | |
5332 | ||
5333 | This parameter controls the TAA mitigation. The | |
5334 | options are: | |
5335 | ||
5336 | full - Enable TAA mitigation on vulnerable CPUs | |
5337 | if TSX is enabled. | |
5338 | ||
5339 | full,nosmt - Enable TAA mitigation and disable SMT on | |
5340 | vulnerable CPUs. If TSX is disabled, SMT | |
5341 | is not disabled because CPU is not | |
5342 | vulnerable to cross-thread TAA attacks. | |
5343 | off - Unconditionally disable TAA mitigation | |
5344 | ||
64870ed1 WL |
5345 | On MDS-affected machines, tsx_async_abort=off can be |
5346 | prevented by an active MDS mitigation as both vulnerabilities | |
5347 | are mitigated with the same mechanism so in order to disable | |
5348 | this mitigation, you need to specify mds=off too. | |
5349 | ||
a7a248c5 PG |
5350 | Not specifying this option is equivalent to |
5351 | tsx_async_abort=full. On CPUs which are MDS affected | |
5352 | and deploy MDS mitigation, TAA mitigation is not | |
5353 | required and doesn't provide any additional | |
5354 | mitigation. | |
5355 | ||
5356 | For details see: | |
5357 | Documentation/admin-guide/hw-vuln/tsx_async_abort.rst | |
5358 | ||
a9913044 RD |
5359 | turbografx.map[2|3]= [HW,JOY] |
5360 | TurboGraFX parallel port interface | |
5361 | Format: | |
5362 | <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7> | |
1752118d | 5363 | See also Documentation/input/devices/joystick-parport.rst |
1da177e4 | 5364 | |
b6935f8c | 5365 | udbg-immortal [PPC] When debugging early kernel crashes that |
e52347bd | 5366 | happen after console_init() and before a proper |
b6935f8c CK |
5367 | console driver takes over, this boot options might |
5368 | help "seeing" what's going on. | |
5369 | ||
f86dcc5a ED |
5370 | uhash_entries= [KNL,NET] |
5371 | Set number of hash buckets for UDP/UDP-Lite connections | |
5372 | ||
5f8364b7 AS |
5373 | uhci-hcd.ignore_oc= |
5374 | [USB] Ignore overcurrent events (default N). | |
5375 | Some badly-designed motherboards generate lots of | |
5376 | bogus events, for ports that aren't wired to | |
5377 | anything. Set this parameter to avoid log spamming. | |
5378 | Note that genuine overcurrent events won't be | |
5379 | reported either. | |
5380 | ||
e3a61b0a | 5381 | unknown_nmi_panic |
44a4dcf7 | 5382 | [X86] Cause panic on unknown NMI. |
e3a61b0a | 5383 | |
c4fc2342 CDH |
5384 | usbcore.authorized_default= |
5385 | [USB] Default USB device authorization: | |
5386 | (default -1 = authorized except for wireless USB, | |
7bae0432 DT |
5387 | 0 = not authorized, 1 = authorized, 2 = authorized |
5388 | if device connected to internal port) | |
c4fc2342 | 5389 | |
b5e795f8 AS |
5390 | usbcore.autosuspend= |
5391 | [USB] The autosuspend time delay (in seconds) used | |
5392 | for newly-detected USB devices (default 2). This | |
5393 | is the time required before an idle device will be | |
5394 | autosuspended. Devices for which the delay is set | |
eaafbc3a | 5395 | to a negative value won't be autosuspended at all. |
b5e795f8 | 5396 | |
fd7c519d JK |
5397 | usbcore.usbfs_snoop= |
5398 | [USB] Set to log all usbfs traffic (default 0 = off). | |
5399 | ||
0290cc9f AS |
5400 | usbcore.usbfs_snoop_max= |
5401 | [USB] Maximum number of bytes to snoop in each URB | |
5402 | (default = 65536). | |
5403 | ||
fd7c519d JK |
5404 | usbcore.blinkenlights= |
5405 | [USB] Set to cycle leds on hubs (default 0 = off). | |
5406 | ||
5407 | usbcore.old_scheme_first= | |
5408 | [USB] Start with the old device initialization | |
3155f4f4 | 5409 | scheme (default 0 = off). |
fd7c519d | 5410 | |
3f5eb8d5 AS |
5411 | usbcore.usbfs_memory_mb= |
5412 | [USB] Memory limit (in MB) for buffers allocated by | |
5413 | usbfs (default = 16, 0 = max = 2047). | |
5414 | ||
fd7c519d JK |
5415 | usbcore.use_both_schemes= |
5416 | [USB] Try the other device initialization scheme | |
5417 | if the first one fails (default 1 = enabled). | |
5418 | ||
5419 | usbcore.initial_descriptor_timeout= | |
5420 | [USB] Specifies timeout for the initial 64-byte | |
6dddd7a7 | 5421 | USB_REQ_GET_DESCRIPTOR request in milliseconds |
fd7c519d JK |
5422 | (default 5000 = 5.0 seconds). |
5423 | ||
40d58148 ON |
5424 | usbcore.nousb [USB] Disable the USB subsystem |
5425 | ||
027bd6ca KHF |
5426 | usbcore.quirks= |
5427 | [USB] A list of quirk entries to augment the built-in | |
5428 | usb core quirk list. List entries are separated by | |
5429 | commas. Each entry has the form | |
5430 | VendorID:ProductID:Flags. The IDs are 4-digit hex | |
5431 | numbers and Flags is a set of letters. Each letter | |
5432 | will change the built-in quirk; setting it if it is | |
5433 | clear and clearing it if it is set. The letters have | |
5434 | the following meanings: | |
5435 | a = USB_QUIRK_STRING_FETCH_255 (string | |
5436 | descriptors must not be fetched using | |
5437 | a 255-byte read); | |
5438 | b = USB_QUIRK_RESET_RESUME (device can't resume | |
5439 | correctly so reset it instead); | |
5440 | c = USB_QUIRK_NO_SET_INTF (device can't handle | |
5441 | Set-Interface requests); | |
5442 | d = USB_QUIRK_CONFIG_INTF_STRINGS (device can't | |
5443 | handle its Configuration or Interface | |
5444 | strings); | |
5445 | e = USB_QUIRK_RESET (device can't be reset | |
5446 | (e.g morph devices), don't use reset); | |
5447 | f = USB_QUIRK_HONOR_BNUMINTERFACES (device has | |
5448 | more interface descriptions than the | |
5449 | bNumInterfaces count, and can't handle | |
5450 | talking to these interfaces); | |
5451 | g = USB_QUIRK_DELAY_INIT (device needs a pause | |
5452 | during initialization, after we read | |
5453 | the device descriptor); | |
5454 | h = USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL (For | |
5455 | high speed and super speed interrupt | |
5456 | endpoints, the USB 2.0 and USB 3.0 spec | |
5457 | require the interval in microframes (1 | |
5458 | microframe = 125 microseconds) to be | |
5459 | calculated as interval = 2 ^ | |
5460 | (bInterval-1). | |
5461 | Devices with this quirk report their | |
5462 | bInterval as the result of this | |
5463 | calculation instead of the exponent | |
5464 | variable used in the calculation); | |
5465 | i = USB_QUIRK_DEVICE_QUALIFIER (device can't | |
5466 | handle device_qualifier descriptor | |
5467 | requests); | |
5468 | j = USB_QUIRK_IGNORE_REMOTE_WAKEUP (device | |
5469 | generates spurious wakeup, ignore | |
5470 | remote wakeup capability); | |
5471 | k = USB_QUIRK_NO_LPM (device can't handle Link | |
5472 | Power Management); | |
5473 | l = USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL | |
5474 | (Device reports its bInterval as linear | |
5475 | frames instead of the USB 2.0 | |
5476 | calculation); | |
5477 | m = USB_QUIRK_DISCONNECT_SUSPEND (Device needs | |
5478 | to be disconnected before suspend to | |
4d8d5a39 KHF |
5479 | prevent spurious wakeup); |
5480 | n = USB_QUIRK_DELAY_CTRL_MSG (Device needs a | |
5481 | pause after every control message); | |
781f0766 KHF |
5482 | o = USB_QUIRK_HUB_SLOW_RESET (Hub needs extra |
5483 | delay after resetting its port); | |
027bd6ca KHF |
5484 | Example: quirks=0781:5580:bk,0a5c:5834:gij |
5485 | ||
1da177e4 LT |
5486 | usbhid.mousepoll= |
5487 | [USBHID] The interval which mice are to be polled at. | |
a9913044 | 5488 | |
933bfe4d TJ |
5489 | usbhid.jspoll= |
5490 | [USBHID] The interval which joysticks are to be polled at. | |
5491 | ||
2ddc8e2d FA |
5492 | usbhid.kbpoll= |
5493 | [USBHID] The interval which keyboards are to be polled at. | |
5494 | ||
d4f373e5 AS |
5495 | usb-storage.delay_use= |
5496 | [UMS] The delay in seconds before a new device is | |
19101954 | 5497 | scanned for Logical Units (default 1). |
d4f373e5 AS |
5498 | |
5499 | usb-storage.quirks= | |
5500 | [UMS] A list of quirks entries to supplement or | |
5501 | override the built-in unusual_devs list. List | |
5502 | entries are separated by commas. Each entry has | |
5503 | the form VID:PID:Flags where VID and PID are Vendor | |
5504 | and Product ID values (4-digit hex numbers) and | |
5505 | Flags is a set of characters, each corresponding | |
5506 | to a common usb-storage quirk flag as follows: | |
c838ea46 | 5507 | a = SANE_SENSE (collect more than 18 bytes |
65cc8bf9 | 5508 | of sense data, not on uas); |
a0bb1081 | 5509 | b = BAD_SENSE (don't collect more than 18 |
65cc8bf9 | 5510 | bytes of sense data, not on uas); |
d4f373e5 AS |
5511 | c = FIX_CAPACITY (decrease the reported |
5512 | device capacity by one sector); | |
5116901d | 5513 | d = NO_READ_DISC_INFO (don't use |
65cc8bf9 | 5514 | READ_DISC_INFO command, not on uas); |
5116901d KR |
5515 | e = NO_READ_CAPACITY_16 (don't use |
5516 | READ_CAPACITY_16 command); | |
734016b0 HG |
5517 | f = NO_REPORT_OPCODES (don't use report opcodes |
5518 | command, uas only); | |
ee136af4 HG |
5519 | g = MAX_SECTORS_240 (don't transfer more than |
5520 | 240 sectors at a time, uas only); | |
c838ea46 AS |
5521 | h = CAPACITY_HEURISTICS (decrease the |
5522 | reported device capacity by one | |
5523 | sector if the number is odd); | |
d4f373e5 AS |
5524 | i = IGNORE_DEVICE (don't bind to this |
5525 | device); | |
13630746 HG |
5526 | j = NO_REPORT_LUNS (don't use report luns |
5527 | command, uas only); | |
d4f373e5 | 5528 | l = NOT_LOCKABLE (don't try to lock and |
65cc8bf9 | 5529 | unlock ejectable media, not on uas); |
d4f373e5 | 5530 | m = MAX_SECTORS_64 (don't transfer more |
65cc8bf9 ON |
5531 | than 64 sectors = 32 KB at a time, |
5532 | not on uas); | |
21c13a4f | 5533 | n = INITIAL_READ10 (force a retry of the |
65cc8bf9 | 5534 | initial READ(10) command, not on uas); |
c838ea46 | 5535 | o = CAPACITY_OK (accept the capacity |
65cc8bf9 | 5536 | reported by the device, not on uas); |
eaa05dfc | 5537 | p = WRITE_CACHE (the device cache is ON |
65cc8bf9 | 5538 | by default, not on uas); |
d4f373e5 | 5539 | r = IGNORE_RESIDUE (the device reports |
65cc8bf9 | 5540 | bogus residue values, not on uas); |
d4f373e5 AS |
5541 | s = SINGLE_LUN (the device has only one |
5542 | Logical Unit); | |
59307852 HG |
5543 | t = NO_ATA_1X (don't allow ATA(12) and ATA(16) |
5544 | commands, uas only); | |
b6089f19 | 5545 | u = IGNORE_UAS (don't bind to the uas driver); |
d4f373e5 AS |
5546 | w = NO_WP_DETECT (don't test whether the |
5547 | medium is write-protected). | |
050bc4e8 | 5548 | y = ALWAYS_SYNC (issue a SYNCHRONIZE_CACHE |
65cc8bf9 ON |
5549 | even if the device claims no cache, |
5550 | not on uas) | |
d4f373e5 AS |
5551 | Example: quirks=0419:aaf5:rl,0421:0433:rc |
5552 | ||
ac1667db SB |
5553 | user_debug= [KNL,ARM] |
5554 | Format: <int> | |
5555 | See arch/arm/Kconfig.debug help text. | |
5556 | 1 - undefined instruction events | |
5557 | 2 - system calls | |
5558 | 4 - invalid data aborts | |
5559 | 8 - SIGSEGV faults | |
5560 | 16 - SIGBUS faults | |
5561 | Example: user_debug=31 | |
5562 | ||
14315592 IC |
5563 | userpte= |
5564 | [X86] Flags controlling user PTE allocations. | |
5565 | ||
5566 | nohigh = do not allocate PTE pages in | |
5567 | HIGHMEM regardless of setting | |
5568 | of CONFIG_HIGHPTE. | |
5569 | ||
6cececfc | 5570 | vdso= [X86,SH] |
b0b49f26 AL |
5571 | On X86_32, this is an alias for vdso32=. Otherwise: |
5572 | ||
5573 | vdso=1: enable VDSO (the default) | |
e6e5494c IM |
5574 | vdso=0: disable VDSO mapping |
5575 | ||
b0b49f26 AL |
5576 | vdso32= [X86] Control the 32-bit vDSO |
5577 | vdso32=1: enable 32-bit VDSO | |
5578 | vdso32=0 or vdso32=2: disable 32-bit VDSO | |
5579 | ||
5580 | See the help text for CONFIG_COMPAT_VDSO for more | |
5581 | details. If CONFIG_COMPAT_VDSO is set, the default is | |
5582 | vdso32=0; otherwise, the default is vdso32=1. | |
5583 | ||
5584 | For compatibility with older kernels, vdso32=2 is an | |
5585 | alias for vdso32=0. | |
5586 | ||
5587 | Try vdso32=0 if you encounter an error that says: | |
5588 | dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed! | |
af65d648 | 5589 | |
d080d397 YI |
5590 | vector= [IA-64,SMP] |
5591 | vector=percpu: enable percpu vector domain | |
5592 | ||
1da177e4 | 5593 | video= [FB] Frame buffer configuration |
ab42b818 | 5594 | See Documentation/fb/modedb.rst. |
1da177e4 | 5595 | |
3afe6dab AL |
5596 | video.brightness_switch_enabled= [0,1] |
5597 | If set to 1, on receiving an ACPI notify event | |
5598 | generated by hotkey, video driver will adjust brightness | |
5599 | level and then send out the event to user space through | |
5600 | the allocated input device; If set to 0, video driver | |
5601 | will only send out the event without touching backlight | |
5602 | brightness level. | |
2843768b | 5603 | default: 1 |
3afe6dab | 5604 | |
81a054ce PM |
5605 | virtio_mmio.device= |
5606 | [VMMIO] Memory mapped virtio (platform) device. | |
5607 | ||
5608 | <size>@<baseaddr>:<irq>[:<id>] | |
5609 | where: | |
5610 | <size> := size (can use standard suffixes | |
5611 | like K, M and G) | |
5612 | <baseaddr> := physical base address | |
5613 | <irq> := interrupt number (as passed to | |
5614 | request_irq()) | |
5615 | <id> := (optional) platform device id | |
5616 | example: | |
5617 | virtio_mmio.device=1K@0x100b0000:48:7 | |
5618 | ||
5619 | Can be used multiple times for multiple devices. | |
5620 | ||
cd4f0ef7 | 5621 | vga= [BOOT,X86-32] Select a particular video mode |
cb1aaebe | 5622 | See Documentation/x86/boot.rst and |
4f4cfa6c | 5623 | Documentation/admin-guide/svga.rst. |
1da177e4 LT |
5624 | Use vga=ask for menu. |
5625 | This is actually a boot loader parameter; the value is | |
5626 | passed to the kernel using a special protocol. | |
5627 | ||
f682a97a AD |
5628 | vm_debug[=options] [KNL] Available with CONFIG_DEBUG_VM=y. |
5629 | May slow down system boot speed, especially when | |
5630 | enabled on systems with a large amount of memory. | |
5631 | All options are enabled by default, and this | |
5632 | interface is meant to allow for selectively | |
5633 | enabling or disabling specific virtual memory | |
5634 | debugging features. | |
5635 | ||
5636 | Available options are: | |
5637 | P Enable page structure init time poisoning | |
5638 | - Disable all of the above options | |
5639 | ||
a9913044 | 5640 | vmalloc=nn[KMG] [KNL,BOOT] Forces the vmalloc area to have an exact |
1da177e4 LT |
5641 | size of <nn>. This can be used to increase the |
5642 | minimum size (128MB on x86). It can also be used to | |
5643 | decrease the size and leave more room for directly | |
5644 | mapped kernel RAM. | |
5645 | ||
3f429842 HC |
5646 | vmcp_cma=nn[MG] [KNL,S390] |
5647 | Sets the memory size reserved for contiguous memory | |
5648 | allocations for the vmcp device driver. | |
5649 | ||
585c3047 PO |
5650 | vmhalt= [KNL,S390] Perform z/VM CP command after system halt. |
5651 | Format: <command> | |
1da177e4 | 5652 | |
585c3047 PO |
5653 | vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic. |
5654 | Format: <command> | |
5655 | ||
5656 | vmpoff= [KNL,S390] Perform z/VM CP command after power off. | |
5657 | Format: <command> | |
a9913044 | 5658 | |
3ae36655 AL |
5659 | vsyscall= [X86-64] |
5660 | Controls the behavior of vsyscalls (i.e. calls to | |
5661 | fixed addresses of 0xffffffffff600x00 from legacy | |
5662 | code). Most statically-linked binaries and older | |
5663 | versions of glibc use these calls. Because these | |
5664 | functions are at fixed addresses, they make nice | |
5665 | targets for exploits that can control RIP. | |
5666 | ||
2e57ae05 | 5667 | emulate [default] Vsyscalls turn into traps and are |
bd49e16e AL |
5668 | emulated reasonably safely. The vsyscall |
5669 | page is readable. | |
3ae36655 | 5670 | |
bd49e16e AL |
5671 | xonly Vsyscalls turn into traps and are |
5672 | emulated reasonably safely. The vsyscall | |
5673 | page is not readable. | |
3ae36655 AL |
5674 | |
5675 | none Vsyscalls don't work at all. This makes | |
5676 | them quite hard to use for exploits but | |
5677 | might break your system. | |
5678 | ||
3855ae1c CL |
5679 | vt.color= [VT] Default text color. |
5680 | Format: 0xYX, X = foreground, Y = background. | |
5681 | Default: 0x07 = light gray on black. | |
5682 | ||
9ea9a886 CL |
5683 | vt.cur_default= [VT] Default cursor shape. |
5684 | Format: 0xCCBBAA, where AA, BB, and CC are the same as | |
5685 | the parameters of the <Esc>[?A;B;Cc escape sequence; | |
5686 | see VGA-softcursor.txt. Default: 2 = underline. | |
5687 | ||
0cb55ad2 RD |
5688 | vt.default_blu= [VT] |
5689 | Format: <blue0>,<blue1>,<blue2>,...,<blue15> | |
5690 | Change the default blue palette of the console. | |
5691 | This is a 16-member array composed of values | |
5692 | ranging from 0-255. | |
5693 | ||
5694 | vt.default_grn= [VT] | |
5695 | Format: <green0>,<green1>,<green2>,...,<green15> | |
5696 | Change the default green palette of the console. | |
5697 | This is a 16-member array composed of values | |
5698 | ranging from 0-255. | |
5699 | ||
5700 | vt.default_red= [VT] | |
5701 | Format: <red0>,<red1>,<red2>,...,<red15> | |
5702 | Change the default red palette of the console. | |
5703 | This is a 16-member array composed of values | |
5704 | ranging from 0-255. | |
5705 | ||
5706 | vt.default_utf8= | |
5707 | [VT] | |
5708 | Format=<0|1> | |
5709 | Set system-wide default UTF-8 mode for all tty's. | |
5710 | Default is 1, i.e. UTF-8 mode is enabled for all | |
5711 | newly opened terminals. | |
5712 | ||
f6c06b68 MG |
5713 | vt.global_cursor_default= |
5714 | [VT] | |
5715 | Format=<-1|0|1> | |
5716 | Set system-wide default for whether a cursor | |
5717 | is shown on new VTs. Default is -1, | |
5718 | i.e. cursors will be created by default unless | |
5719 | overridden by individual drivers. 0 will hide | |
5720 | cursors, 1 will display them. | |
5721 | ||
3855ae1c CL |
5722 | vt.italic= [VT] Default color for italic text; 0-15. |
5723 | Default: 2 = green. | |
5724 | ||
5725 | vt.underline= [VT] Default color for underlined text; 0-15. | |
5726 | Default: 3 = cyan. | |
5727 | ||
4724ba57 | 5728 | watchdog timers [HW,WDT] For information on watchdog timers, |
cc2a2d19 | 5729 | see Documentation/watchdog/watchdog-parameters.rst |
4724ba57 RD |
5730 | or other driver-specific files in the |
5731 | Documentation/watchdog/ directory. | |
1da177e4 | 5732 | |
11295055 LO |
5733 | watchdog_thresh= |
5734 | [KNL] | |
5735 | Set the hard lockup detector stall duration | |
5736 | threshold in seconds. The soft lockup detector | |
5737 | threshold is set to twice the value. A value of 0 | |
5738 | disables both lockup detectors. Default is 10 | |
5739 | seconds. | |
5740 | ||
82607adc TH |
5741 | workqueue.watchdog_thresh= |
5742 | If CONFIG_WQ_WATCHDOG is configured, workqueue can | |
5743 | warn stall conditions and dump internal state to | |
5744 | help debugging. 0 disables workqueue stall | |
5745 | detection; otherwise, it's the stall threshold | |
5746 | duration in seconds. The default value is 30 and | |
5747 | it can be updated at runtime by writing to the | |
5748 | corresponding sysfs file. | |
5749 | ||
d55262c4 TH |
5750 | workqueue.disable_numa |
5751 | By default, all work items queued to unbound | |
5752 | workqueues are affine to the NUMA nodes they're | |
5753 | issued on, which results in better behavior in | |
5754 | general. If NUMA affinity needs to be disabled for | |
5755 | whatever reason, this option can be used. Note | |
5756 | that this also can be controlled per-workqueue for | |
5757 | workqueues visible under /sys/bus/workqueue/. | |
5758 | ||
cee22a15 VK |
5759 | workqueue.power_efficient |
5760 | Per-cpu workqueues are generally preferred because | |
5761 | they show better performance thanks to cache | |
5762 | locality; unfortunately, per-cpu workqueues tend to | |
5763 | be more power hungry than unbound workqueues. | |
5764 | ||
5765 | Enabling this makes the per-cpu workqueues which | |
5766 | were observed to contribute significantly to power | |
5767 | consumption unbound, leading to measurably lower | |
5768 | power usage at the cost of small performance | |
5769 | overhead. | |
5770 | ||
5771 | The default value of this parameter is determined by | |
5772 | the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT. | |
5773 | ||
f303fccb TH |
5774 | workqueue.debug_force_rr_cpu |
5775 | Workqueue used to implicitly guarantee that work | |
5776 | items queued without explicit CPU specified are put | |
5777 | on the local CPU. This guarantee is no longer true | |
5778 | and while local CPU is still preferred work items | |
5779 | may be put on foreign CPUs. This debug option | |
5780 | forces round-robin CPU selection to flush out | |
5781 | usages which depend on the now broken guarantee. | |
5782 | When enabled, memory and cache locality will be | |
5783 | impacted. | |
5784 | ||
0cb55ad2 RD |
5785 | x2apic_phys [X86-64,APIC] Use x2apic physical mode instead of |
5786 | default x2apic cluster mode on platforms | |
5787 | supporting x2apic. | |
5788 | ||
712b6aa8 KS |
5789 | x86_intel_mid_timer= [X86-32,APBT] |
5790 | Choose timer option for x86 Intel MID platform. | |
bb24c471 JP |
5791 | Two valid options are apbt timer only and lapic timer |
5792 | plus one apbt timer for broadcast timer. | |
712b6aa8 | 5793 | x86_intel_mid_timer=apbt_only | lapic_and_apbt |
bb24c471 | 5794 | |
c70727a5 JG |
5795 | xen_512gb_limit [KNL,X86-64,XEN] |
5796 | Restricts the kernel running paravirtualized under Xen | |
5797 | to use only up to 512 GB of RAM. The reason to do so is | |
5798 | crash analysis tools and Xen tools for doing domain | |
5799 | save/restore/migration must be enabled to handle larger | |
5800 | domains. | |
5801 | ||
c1c5413a SS |
5802 | xen_emul_unplug= [HW,X86,XEN] |
5803 | Unplug Xen emulated devices | |
5804 | Format: [unplug0,][unplug1] | |
5805 | ide-disks -- unplug primary master IDE devices | |
5806 | aux-ide-disks -- unplug non-primary-master IDE devices | |
5807 | nics -- unplug network devices | |
5808 | all -- unplug all emulated devices (NICs and IDE disks) | |
1dc7ce99 IC |
5809 | unnecessary -- unplugging emulated devices is |
5810 | unnecessary even if the host did not respond to | |
5811 | the unplug protocol | |
c93a4dfb | 5812 | never -- do not unplug even if version check succeeds |
c1c5413a | 5813 | |
c6875f3a BO |
5814 | xen_legacy_crash [X86,XEN] |
5815 | Crash from Xen panic notifier, without executing late | |
5816 | panic() code such as dumping handler. | |
5817 | ||
15a3eac0 | 5818 | xen_nopvspin [X86,XEN] |
9a3c05e6 ZD |
5819 | Disables the qspinlock slowpath using Xen PV optimizations. |
5820 | This parameter is obsoleted by "nopvspin" parameter, which | |
5821 | has equivalent effect for XEN platform. | |
15a3eac0 | 5822 | |
8d693b91 KRW |
5823 | xen_nopv [X86] |
5824 | Disables the PV optimizations forcing the HVM guest to | |
5825 | run as generic HVM guest with no PV drivers. | |
b39b0497 ZD |
5826 | This option is obsoleted by the "nopv" option, which |
5827 | has equivalent effect for XEN platform. | |
8d693b91 | 5828 | |
197ecb38 MMG |
5829 | xen_scrub_pages= [XEN] |
5830 | Boolean option to control scrubbing pages before giving them back | |
5831 | to Xen, for use by other domains. Can be also changed at runtime | |
5832 | with /sys/devices/system/xen_memory/xen_memory0/scrub_pages. | |
5833 | Default value controlled with CONFIG_XEN_SCRUB_PAGES_DEFAULT. | |
5834 | ||
2ec16bc0 RT |
5835 | xen_timer_slop= [X86-64,XEN] |
5836 | Set the timer slop (in nanoseconds) for the virtual Xen | |
5837 | timers (default is 100000). This adjusts the minimum | |
5838 | delta of virtualized Xen timers, where lower values | |
5839 | improve timer resolution at the expense of processing | |
5840 | more timer interrupts. | |
5841 | ||
30978346 ZD |
5842 | nopv= [X86,XEN,KVM,HYPER_V,VMWARE] |
5843 | Disables the PV optimizations forcing the guest to run | |
5844 | as generic guest with no PV drivers. Currently support | |
5845 | XEN HVM, KVM, HYPER_V and VMWARE guest. | |
5846 | ||
9a3c05e6 | 5847 | nopvspin [X86,XEN,KVM] |
05eee619 ZD |
5848 | Disables the qspinlock slow path using PV optimizations |
5849 | which allow the hypervisor to 'idle' the guest on lock | |
5850 | contention. | |
5851 | ||
1da177e4 | 5852 | xirc2ps_cs= [NET,PCMCIA] |
a9913044 RD |
5853 | Format: |
5854 | <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]] | |
c0addc9a | 5855 | |
ba45cff6 MN |
5856 | xive= [PPC] |
5857 | By default on POWER9 and above, the kernel will | |
5858 | natively use the XIVE interrupt controller. This option | |
5859 | allows the fallback firmware mode to be used: | |
5860 | ||
5861 | off Fallback to firmware control of XIVE interrupt | |
5862 | controller on both pseries and powernv | |
5863 | platforms. Only useful on POWER9 and above. | |
5864 | ||
c0addc9a LT |
5865 | xhci-hcd.quirks [USB,KNL] |
5866 | A hex value specifying bitmask with supplemental xhci | |
5867 | host controller quirks. Meaning of each bit can be | |
5868 | consulted in header drivers/usb/host/xhci.h. | |
6278f55b GR |
5869 | |
5870 | xmon [PPC] | |
5871 | Format: { early | on | rw | ro | off } | |
5872 | Controls if xmon debugger is enabled. Default is off. | |
5873 | Passing only "xmon" is equivalent to "xmon=early". | |
5874 | early Call xmon as early as possible on boot; xmon | |
5875 | debugger is called from setup_arch(). | |
5876 | on xmon debugger hooks will be installed so xmon | |
5877 | is only called on a kernel crash. Default mode, | |
5878 | i.e. either "ro" or "rw" mode, is controlled | |
5879 | with CONFIG_XMON_DEFAULT_RO_MODE. | |
5880 | rw xmon debugger hooks will be installed so xmon | |
5881 | is called only on a kernel crash, mode is write, | |
5882 | meaning SPR registers, memory and, other data | |
5883 | can be written using xmon commands. | |
5884 | ro same as "rw" option above but SPR registers, | |
5885 | memory, and other data can't be written using | |
5886 | xmon commands. | |
5887 | off xmon is disabled. |