blob: a68b880de12d54750afc7fe3781b3b1805ffa9c0 [file] [log] [blame]
Mauro Carvalho Chehabe53dd082007-09-15 07:38:47 -03001 Kernel Parameters
2 ~~~~~~~~~~~~~~~~~
Linus Torvalds1da177e2005-04-16 15:20:36 -07003
Rusty Russell5888bcc2014-05-14 10:33:45 +09304The following is a consolidated list of the kernel parameters as
5implemented by the __setup(), core_param() and module_param() macros
6and sorted into English Dictionary order (defined as ignoring all
7punctuation and sorting digits before letters in a case insensitive
8manner), and with descriptions where known.
Linus Torvalds1da177e2005-04-16 15:20:36 -07009
Rusty Russell5888bcc2014-05-14 10:33:45 +093010The kernel parses parameters from the kernel command line up to "--";
11if it doesn't recognize a parameter and it doesn't contain a '.', the
12parameter gets passed to init: parameters with '=' go into init's
13environment, others are passed as command line arguments to init.
14Everything after "--" is passed as an argument to init.
Linus Torvalds1da177e2005-04-16 15:20:36 -070015
Rusty Russell5888bcc2014-05-14 10:33:45 +093016Module parameters can be specified in two ways: via the kernel command
17line with a module name prefix, or via modprobe, e.g.:
Linus Torvalds1da177e2005-04-16 15:20:36 -070018
Rusty Russell5888bcc2014-05-14 10:33:45 +093019 (kernel command line) usbcore.blinkenlights=1
20 (modprobe command line) modprobe usbcore blinkenlights=1
Linus Torvalds1da177e2005-04-16 15:20:36 -070021
Rusty Russell5888bcc2014-05-14 10:33:45 +093022Parameters for modules which are built into the kernel need to be
23specified on the kernel command line. modprobe looks through the
24kernel command line (/proc/cmdline) and collects module parameters
25when it loads a module, so the kernel command line can be used for
26loadable modules too.
Linus Torvalds1da177e2005-04-16 15:20:36 -070027
Randy Dunlapca1eda22009-05-06 16:02:58 -070028Hyphens (dashes) and underscores are equivalent in parameter names, so
29 log_buf_len=1M print-fatal-signals=1
30can also be entered as
31 log-buf-len=1M print_fatal_signals=1
32
Rusty Russell5888bcc2014-05-14 10:33:45 +093033Double-quotes can be used to protect spaces in values, e.g.:
34 param="spaces in here"
Randy Dunlapca1eda22009-05-06 16:02:58 -070035
Noam Camus2d13e6c2016-10-11 13:51:35 -070036cpu lists:
37----------
38
39Some kernel parameters take a list of CPUs as a value, e.g. isolcpus,
40nohz_full, irqaffinity, rcu_nocbs. The format of this list is:
41
42 <cpu number>,...,<cpu number>
43
44or
45
46 <cpu number>-<cpu number>
47 (must be a positive range in ascending order)
48
49or a mixture
50
51<cpu number>,...,<cpu number>-<cpu number>
52
53Note that for the special case of a range one can split the range into equal
54sized groups and for each group use some amount from the beginning of that
55group:
56
57 <cpu number>-cpu number>:<used size>/<group size>
58
59For example one can add to the command line following parameter:
60
61 isolcpus=1,2,10-20,100-2000:2/25
62
63where the final item represents CPUs 100,101,125,126,150,151,...
64
65
66
Stefan Richtera901ebb2006-04-01 01:43:18 +020067This document may not be entirely up to date and comprehensive. The command
68"modinfo -p ${modulename}" shows a current list of all parameters of a loadable
69module. Loadable modules, after being loaded into the running kernel, also
70reveal their parameters in /sys/module/${modulename}/parameters/. Some of these
71parameters may be changed at runtime by the command
72"echo -n ${value} > /sys/module/${modulename}/parameters/${parm}".
73
Stefan Richter6585fa82006-04-01 01:44:30 +020074The parameters listed below are only valid if certain kernel build options were
75enabled and if respective hardware is present. The text in square brackets at
76the beginning of each description states the restrictions within which a
77parameter is applicable:
Linus Torvalds1da177e2005-04-16 15:20:36 -070078
79 ACPI ACPI support is enabled.
Chuck Ebbertc99c1082007-07-27 10:46:20 +100080 AGP AGP (Accelerated Graphics Port) is enabled.
Linus Torvalds1da177e2005-04-16 15:20:36 -070081 ALSA ALSA sound support is enabled.
82 APIC APIC support is enabled.
83 APM Advanced Power Management support is enabled.
Randy Dunlap16290242011-08-13 12:34:52 -070084 ARM ARM architecture is enabled.
Haavard Skinnemoene7ba1762007-10-10 14:58:29 +020085 AVR32 AVR32 architecture is enabled.
Linus Torvalds1da177e2005-04-16 15:20:36 -070086 AX25 Appropriate AX.25 support is enabled.
Robin Getz0ae53642007-10-09 17:24:49 +080087 BLACKFIN Blackfin architecture is enabled.
Olof Johansson1e435252013-04-27 14:10:18 -070088 CLK Common clock infrastructure is enabled.
Robert Tivy5c71d612013-03-28 18:41:46 -070089 CMA Contiguous Memory Area support is enabled.
Will Drewrya058da82010-06-09 17:47:38 -050090 DM Device mapper support is enabled.
Alan Cox9cfe2682011-01-25 14:18:38 +000091 DRM Direct Rendering Management support is enabled.
92 DYNAMIC_DEBUG Build in debug messages and enable them at runtime
Linus Torvalds1da177e2005-04-16 15:20:36 -070093 EDD BIOS Enhanced Disk Drive Services (EDD) is enabled
94 EFI EFI Partitioning (GPT) is enabled
95 EIDE EIDE/ATAPI support is enabled.
Mimi Zohar7102ebc2011-05-12 18:33:20 -040096 EVM Extended Verification Module
Linus Torvalds1da177e2005-04-16 15:20:36 -070097 FB The frame buffer device is enabled.
Randy Dunlap16290242011-08-13 12:34:52 -070098 FTRACE Function tracing enabled.
Peter Oberparleiter2521f2c2009-06-17 16:28:08 -070099 GCOV GCOV profiling is enabled.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700100 HW Appropriate hardware is enabled.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700101 IA-64 IA-64 architecture is enabled.
Mimi Zohar6146f0d2009-02-04 09:06:57 -0500102 IMA Integrity measurement architecture is enabled.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700103 IOSCHED More than one I/O scheduler is enabled.
Adrian Bunk41e2e8b2005-07-12 13:58:33 -0700104 IP_PNP IP DHCP, BOOTP, or RARP is enabled.
Brian Haleyb0f83b22010-02-04 13:36:50 -0800105 IPV6 IPv6 support is enabled.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700106 ISAPNP ISA PnP code is enabled.
107 ISDN Appropriate ISDN support is enabled.
108 JOY Appropriate joystick support is enabled.
Jason Wessel84c08fd2010-05-20 21:04:24 -0500109 KGDB Kernel debugger support is enabled.
Andre Przywarafef07aa2009-07-10 14:20:35 +0200110 KVM Kernel Virtual Machine support is enabled.
Kristen Carlson Accardi11ef6972006-09-28 11:29:01 -0700111 LIBATA Libata driver is enabled
Linus Torvalds1da177e2005-04-16 15:20:36 -0700112 LP Printer support is enabled.
113 LOOP Loopback device support is enabled.
114 M68k M68k architecture is enabled.
115 These options have more detailed description inside of
116 Documentation/m68k/kernel-options.txt.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700117 MDA MDA console support is enabled.
Randy Dunlap16290242011-08-13 12:34:52 -0700118 MIPS MIPS architecture is enabled.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700119 MOUSE Appropriate mouse support is enabled.
Matthew Wilcox309e57d2006-03-05 22:33:34 -0700120 MSI Message Signaled Interrupts (PCI).
Randy Dunlapc8facbb2007-07-31 00:37:40 -0700121 MTD MTD (Memory Technology Device) support is enabled.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700122 NET Appropriate network support is enabled.
123 NUMA NUMA support is enabled.
124 NFS Appropriate NFS support is enabled.
125 OSS OSS sound support is enabled.
Randy Dunlapc8facbb2007-07-31 00:37:40 -0700126 PV_OPS A paravirtualized kernel is enabled.
127 PARIDE The ParIDE (parallel port IDE) subsystem is enabled.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700128 PARISC The PA-RISC architecture is enabled.
129 PCI PCI bus support is enabled.
Randy Dunlap7f785762007-10-05 13:17:58 -0700130 PCIE PCI Express support is enabled.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700131 PCMCIA The PCMCIA subsystem is enabled.
132 PNP Plug & Play support is enabled.
133 PPC PowerPC architecture is enabled.
134 PPT Parallel port support is enabled.
135 PS2 Appropriate PS/2 support is enabled.
136 RAM RAM disk support is enabled.
137 S390 S390 architecture is enabled.
138 SCSI Appropriate SCSI support is enabled.
Randy Dunlap163475f2010-08-14 12:36:14 -0700139 A lot of drivers have their options described inside
140 the Documentation/scsi/ sub-directory.
James Morris20510f22007-10-16 23:31:32 -0700141 SECURITY Different security models are enabled.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700142 SELINUX SELinux support is enabled.
John Johansenc1c124e2010-07-29 14:48:09 -0700143 APPARMOR AppArmor support is enabled.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700144 SERIAL Serial support is enabled.
Paul Mundte523d932007-02-28 18:30:01 +0900145 SH SuperH architecture is enabled.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700146 SMP The kernel is an SMP kernel.
147 SPARC Sparc architecture is enabled.
David Brownell77437fd2008-07-23 21:28:33 -0700148 SWSUSP Software suspend (hibernation) is enabled.
149 SUSPEND System suspend states are enabled.
Rajiv Andrade225a9be2010-03-25 00:55:32 -0300150 TPM TPM drivers are enabled.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700151 TS Appropriate touchscreen support is enabled.
Alan Sternd4f373e2008-11-10 14:07:45 -0500152 UMS USB Mass Storage support is enabled.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700153 USB USB support is enabled.
154 USBHID USB Human Interface Device support is enabled.
155 V4L Video For Linux support is enabled.
Pawel Moll81a054c2012-05-09 18:30:16 +0100156 VMMIO Driver for memory mapped virtio devices is enabled.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700157 VGA The VGA console has been enabled.
158 VT Virtual terminal support is enabled.
159 WDT Watchdog support is enabled.
160 XT IBM PC/XT MFM hard disk support is enabled.
Alan Coxcd4f0ef2007-07-31 00:37:59 -0700161 X86-32 X86-32, aka i386 architecture is enabled.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700162 X86-64 X86-64 architecture is enabled.
163 More X86-64 boot options can be found in
Uwe Hermann71cced62008-10-20 09:32:21 -0700164 Documentation/x86/x86_64/boot-options.txt .
Randy Dunlap16290242011-08-13 12:34:52 -0700165 X86 Either 32-bit or 64-bit x86 (same as X86-32+X86-64)
Alex Thorlton1c532e02016-03-31 14:18:29 -0500166 X86_UV SGI UV support is enabled.
Stefano Stabellinic1c54132010-05-14 12:44:30 +0100167 XEN Xen support is enabled
Linus Torvalds1da177e2005-04-16 15:20:36 -0700168
169In addition, the following text indicates that the option:
170
171 BUGS= Relates to possible processor bugs on the said processor.
172 KNL Is a kernel start-up parameter.
173 BOOT Is a boot loader parameter.
174
175Parameters denoted with BOOT are actually interpreted by the boot
176loader, and have no meaning to the kernel directly.
177Do not modify the syntax of boot loader parameters without extreme
Kyle McMartin954a8b82009-02-19 16:48:14 -0500178need or coordination with <Documentation/x86/boot.txt>.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700179
Karsten Weiss55588702007-02-13 13:26:21 +0100180There are also arch-specific kernel-parameters not documented here.
Uwe Hermann71cced62008-10-20 09:32:21 -0700181See for example <Documentation/x86/x86_64/boot-options.txt>.
Karsten Weiss55588702007-02-13 13:26:21 +0100182
Linus Torvalds1da177e2005-04-16 15:20:36 -0700183Note that ALL kernel parameters listed below are CASE SENSITIVE, and that
184a trailing = on the name of any parameter states that that parameter will
185be entered as an environment variable, whereas its absence indicates that
186it will appear as a kernel argument readable via /proc/cmdline by programs
187running once the system is up.
188
jens m. noedler9c4751f2006-09-29 02:00:27 -0700189The number of kernel parameters is not limited, but the length of the
190complete command line (parameters including spaces etc.) is limited to
191a fixed number of characters. This limit depends on the architecture
192and is between 256 and 4096 characters. It is defined in the file
193./include/asm/setup.h as COMMAND_LINE_SIZE.
194
Ahmed S. Darwish7a19a232011-02-20 20:08:35 -0800195Finally, the [KMG] suffix is commonly described after a number of kernel
196parameter values. These 'K', 'M', and 'G' letters represent the _binary_
197multipliers 'Kilo', 'Mega', and 'Giga', equalling 2^10, 2^20, and 2^30
198bytes respectively. Such letter suffixes can also be entirely omitted.
199
jens m. noedler9c4751f2006-09-29 02:00:27 -0700200
Al Stoneb10d79f2015-03-24 14:02:41 +0000201 acpi= [HW,ACPI,X86,ARM64]
Bernhard Walle03d926f2007-03-06 02:29:44 -0800202 Advanced Configuration and Power Interface
Ard Biesheuvel6a1f5472016-04-12 16:09:11 +0200203 Format: { force | on | off | strict | noirq | rsdt |
Rami Rosene58d1542015-09-26 19:27:57 +0300204 copy_dsdt }
Linus Torvalds1da177e2005-04-16 15:20:36 -0700205 force -- enable ACPI if default was off
Ard Biesheuvel6a1f5472016-04-12 16:09:11 +0200206 on -- enable ACPI but allow fallback to DT [arm64]
Linus Torvalds1da177e2005-04-16 15:20:36 -0700207 off -- disable ACPI if default was on
208 noirq -- do not use ACPI for IRQ routing
Randy Dunlapa9913042005-10-23 12:57:11 -0700209 strict -- Be less tolerant of platforms that are not
Linus Torvalds1da177e2005-04-16 15:20:36 -0700210 strictly ACPI specification compliant.
Zhao Yakui237889b2008-12-17 16:55:18 +0800211 rsdt -- prefer RSDT over (default) XSDT
Lin Mingaa2110c2010-04-08 14:34:27 +0800212 copy_dsdt -- copy DSDT to memory
Ard Biesheuvel6a1f5472016-04-12 16:09:11 +0200213 For ARM64, ONLY "acpi=off", "acpi=on" or "acpi=force"
214 are available
Linus Torvalds1da177e2005-04-16 15:20:36 -0700215
Paul Bolle395cf962011-08-15 02:02:26 +0200216 See also Documentation/power/runtime_pm.txt, pci=noacpi
Linus Torvalds1da177e2005-04-16 15:20:36 -0700217
Len Browna1fdcc02007-03-11 03:26:14 -0400218 acpi_apic_instance= [ACPI, IOAPIC]
219 Format: <int>
220 2: use 2nd APIC table, if available
221 1,0: use 1st APIC table
Len Brown4e381a42007-03-30 14:16:10 -0400222 default: 0
Len Browna1fdcc02007-03-11 03:26:14 -0400223
Thomas Renningerc3d6de62008-08-01 17:37:55 +0200224 acpi_backlight= [HW,ACPI]
225 acpi_backlight=vendor
226 acpi_backlight=video
227 If set to vendor, prefer vendor specific driver
228 (e.g. thinkpad_acpi, sony_acpi, etc.) instead
229 of the ACPI video.ko driver.
230
Colin Ian Kingb2ca5da2016-01-21 17:05:47 +0000231 acpi_force_32bit_fadt_addr
232 force FADT to use 32 bit addresses rather than the
233 64 bit X_* addresses. Some firmware have broken 64
234 bit addresses for force ACPI ignore these and use
235 the older legacy 32 bit addresses.
236
Dominik Brodowskief694492015-05-14 15:31:28 +0200237 acpica_no_return_repair [HW, ACPI]
238 Disable AML predefined validation mechanism
239 This mechanism can repair the evaluation result to make
240 the return objects more ACPI specification compliant.
241 This option is useful for developers to identify the
242 root cause of an AML interpreter issue when the issue
243 has something to do with the repair mechanism.
244
Bjorn Helgaasa0d84a92008-11-07 16:58:05 -0700245 acpi.debug_layer= [HW,ACPI,ACPI_DEBUG]
246 acpi.debug_level= [HW,ACPI,ACPI_DEBUG]
Linus Torvalds1da177e2005-04-16 15:20:36 -0700247 Format: <int>
Bjorn Helgaasa0d84a92008-11-07 16:58:05 -0700248 CONFIG_ACPI_DEBUG must be enabled to produce any ACPI
249 debug output. Bits in debug_layer correspond to a
250 _COMPONENT in an ACPI source file, e.g.,
251 #define _COMPONENT ACPI_PCI_COMPONENT
252 Bits in debug_level correspond to a level in
253 ACPI_DEBUG_PRINT statements, e.g.,
254 ACPI_DEBUG_PRINT((ACPI_DB_INFO, ...
Bjorn Helgaase76f4272008-11-13 17:30:13 -0600255 The debug_level mask defaults to "info". See
256 Documentation/acpi/debug.txt for more information about
257 debug layers and levels.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700258
Bjorn Helgaase76f4272008-11-13 17:30:13 -0600259 Enable processor driver info messages:
260 acpi.debug_layer=0x20000000
261 Enable PCI/PCI interrupt routing info messages:
262 acpi.debug_layer=0x400000
Bjorn Helgaasa0d84a92008-11-07 16:58:05 -0700263 Enable AML "Debug" output, i.e., stores to the Debug
264 object while interpreting AML:
265 acpi.debug_layer=0xffffffff acpi.debug_level=0x2
Bjorn Helgaasa0d84a92008-11-07 16:58:05 -0700266 Enable all messages related to ACPI hardware:
267 acpi.debug_layer=0x2 acpi.debug_level=0xffffffff
Thomas Renninger36eec5e2008-08-11 15:03:20 +0200268
Bjorn Helgaasa0d84a92008-11-07 16:58:05 -0700269 Some values produce so much output that the system is
270 unusable. The "log_buf_len" parameter may be useful
271 if you need to capture more output.
Zhang Ruif9891062007-04-24 13:53:22 +0800272
Dominik Brodowskief694492015-05-14 15:31:28 +0200273 acpi_enforce_resources= [ACPI]
274 { strict | lax | no }
275 Check for resource conflicts between native drivers
276 and ACPI OperationRegions (SystemIO and SystemMemory
277 only). IO ports and memory declared in ACPI might be
278 used by the ACPI subsystem in arbitrary AML code and
279 can interfere with legacy drivers.
280 strict (default): access to resources claimed by ACPI
281 is denied; legacy drivers trying to access reserved
282 resources will fail to bind to device using them.
283 lax: access to resources claimed by ACPI is allowed;
284 legacy drivers trying to access reserved resources
285 will bind successfully but a warning message is logged.
286 no: ACPI OperationRegions are not marked as reserved,
287 no further checks are performed.
288
Lv Zheng4fc0a7e2014-05-31 08:15:02 +0800289 acpi_force_table_verification [HW,ACPI]
290 Enable table checksum verification during early stage.
291 By default, this is disabled due to x86 early mapping
292 size limitation.
293
Randy Dunlap0cb55ad2009-04-05 15:55:22 -0700294 acpi_irq_balance [HW,ACPI]
295 ACPI will balance active IRQs
296 default in APIC mode
297
298 acpi_irq_nobalance [HW,ACPI]
299 ACPI will not move active IRQs (default)
300 default in PIC mode
301
302 acpi_irq_isa= [HW,ACPI] If irq_balance, mark listed IRQs used by ISA
303 Format: <irq>,<irq>...
304
305 acpi_irq_pci= [HW,ACPI] If irq_balance, clear listed IRQs for
306 use by PCI
307 Format: <irq>,<irq>...
308
Lv Zhengc1e94142017-04-04 19:32:17 +0000309 acpi_mask_gpe= [HW,ACPI]
310 Due to the existence of _Lxx/_Exx, some GPEs triggered
311 by unsupported hardware/firmware features can result in
312 GPE floodings that cannot be automatically disabled by
313 the GPE dispatcher.
314 This facility can be used to prevent such uncontrolled
315 GPE floodings.
316 Format: <int>
Lv Zhengc1e94142017-04-04 19:32:17 +0000317
Lv Zheng08e1d7c2014-03-24 14:49:22 +0800318 acpi_no_auto_serialize [HW,ACPI]
319 Disable auto-serialization of AML methods
Bob Moore22b5afc2014-03-24 14:49:00 +0800320 AML control methods that contain the opcodes to create
321 named objects will be marked as "Serialized" by the
322 auto-serialization feature.
Lv Zheng08e1d7c2014-03-24 14:49:22 +0800323 This feature is enabled by default.
324 This option allows to turn off the feature.
Bob Moore22b5afc2014-03-24 14:49:00 +0800325
Dominik Brodowskief694492015-05-14 15:31:28 +0200326 acpi_no_memhotplug [ACPI] Disable memory hotplug. Useful for kdump
327 kernels.
328
Lv Zhenga94e88c2014-04-04 12:39:11 +0800329 acpi_no_static_ssdt [HW,ACPI]
330 Disable installation of static SSDTs at early boot time
331 By default, SSDTs contained in the RSDT/XSDT will be
332 installed automatically and they will appear under
333 /sys/firmware/acpi/tables.
334 This option turns off this feature.
335 Note that specifying this option does not affect
336 dynamic table installation which will install SSDT
337 tables to /sys/firmware/acpi/tables/dynamic.
Randy Dunlap0cb55ad2009-04-05 15:55:22 -0700338
Dominik Brodowskief694492015-05-14 15:31:28 +0200339 acpi_rsdp= [ACPI,EFI,KEXEC]
340 Pass the RSDP address to the kernel, mostly used
341 on machines running EFI runtime service to boot the
342 second kernel for kdump.
Lv Zheng4dde5072014-02-11 11:01:52 +0800343
Randy Dunlap0cb55ad2009-04-05 15:55:22 -0700344 acpi_os_name= [HW,ACPI] Tell ACPI BIOS the name of the OS
345 Format: To spoof as Windows 98: ="Microsoft Windows"
346
Rafael J. Wysocki18d78b62015-07-03 01:06:00 +0200347 acpi_rev_override [ACPI] Override the _REV object to return 5 (instead
348 of 2 which is mandated by ACPI 6) as the supported ACPI
349 specification revision (when using this switch, it may
350 be necessary to carry out a cold reboot _twice_ in a
351 row to make it take effect on the platform firmware).
352
Randy Dunlap0cb55ad2009-04-05 15:55:22 -0700353 acpi_osi= [HW,ACPI] Modify list of supported OS interface strings
Lv Zheng5dc17982013-07-22 16:08:25 +0800354 acpi_osi="string1" # add string1
355 acpi_osi="!string2" # remove string2
Lv Zheng741d8122013-07-22 16:08:36 +0800356 acpi_osi=!* # remove all strings
Lv Zheng5dc17982013-07-22 16:08:25 +0800357 acpi_osi=! # disable all built-in OS vendor
358 strings
Lv Zhenga707ede2016-05-03 16:48:32 +0800359 acpi_osi=!! # enable all built-in OS vendor
360 strings
Randy Dunlap0cb55ad2009-04-05 15:55:22 -0700361 acpi_osi= # disable all strings
362
Lv Zheng5dc17982013-07-22 16:08:25 +0800363 'acpi_osi=!' can be used in combination with single or
364 multiple 'acpi_osi="string1"' to support specific OS
365 vendor string(s). Note that such command can only
366 affect the default state of the OS vendor strings, thus
367 it cannot affect the default state of the feature group
368 strings and the current state of the OS vendor strings,
369 specifying it multiple times through kernel command line
Lv Zheng741d8122013-07-22 16:08:36 +0800370 is meaningless. This command is useful when one do not
371 care about the state of the feature group strings which
372 should be controlled by the OSPM.
Lv Zheng5dc17982013-07-22 16:08:25 +0800373 Examples:
374 1. 'acpi_osi=! acpi_osi="Windows 2000"' is equivalent
375 to 'acpi_osi="Windows 2000" acpi_osi=!', they all
376 can make '_OSI("Windows 2000")' TRUE.
377
378 'acpi_osi=' cannot be used in combination with other
379 'acpi_osi=' command lines, the _OSI method will not
380 exist in the ACPI namespace. NOTE that such command can
381 only affect the _OSI support state, thus specifying it
382 multiple times through kernel command line is also
383 meaningless.
384 Examples:
385 1. 'acpi_osi=' can make 'CondRefOf(_OSI, Local1)'
386 FALSE.
387
Lv Zheng741d8122013-07-22 16:08:36 +0800388 'acpi_osi=!*' can be used in combination with single or
389 multiple 'acpi_osi="string1"' to support specific
390 string(s). Note that such command can affect the
391 current state of both the OS vendor strings and the
392 feature group strings, thus specifying it multiple times
393 through kernel command line is meaningful. But it may
394 still not able to affect the final state of a string if
395 there are quirks related to this string. This command
396 is useful when one want to control the state of the
397 feature group strings to debug BIOS issues related to
398 the OSPM features.
399 Examples:
400 1. 'acpi_osi="Module Device" acpi_osi=!*' can make
401 '_OSI("Module Device")' FALSE.
402 2. 'acpi_osi=!* acpi_osi="Module Device"' can make
403 '_OSI("Module Device")' TRUE.
404 3. 'acpi_osi=! acpi_osi=!* acpi_osi="Windows 2000"' is
405 equivalent to
406 'acpi_osi=!* acpi_osi=! acpi_osi="Windows 2000"'
407 and
408 'acpi_osi=!* acpi_osi="Windows 2000" acpi_osi=!',
409 they all will make '_OSI("Windows 2000")' TRUE.
410
Jaswinder Singh Rajput6cececf2009-04-14 14:03:43 +0530411 acpi_pm_good [X86]
Randy Dunlap0cb55ad2009-04-05 15:55:22 -0700412 Override the pmtimer bug detection: force the kernel
413 to assume that this machine's pmtimer latches its value
414 and always returns good values.
415
Randy Dunlap4af94f32009-04-17 18:30:28 -0700416 acpi_sci= [HW,ACPI] ACPI System Control Interrupt trigger mode
417 Format: { level | edge | high | low }
418
Randy Dunlap4af94f32009-04-17 18:30:28 -0700419 acpi_skip_timer_override [HW,ACPI]
420 Recognize and ignore IRQ0/pin2 Interrupt Override.
421 For broken nForce2 BIOS resulting in XT-PIC timer.
422
423 acpi_sleep= [HW,ACPI] Sleep options
424 Format: { s3_bios, s3_mode, s3_beep, s4_nohwsig,
Amerigo Wangc3b07952011-05-10 21:09:38 +0200425 old_ordering, nonvs, sci_force_enable }
Randy Dunlap4af94f32009-04-17 18:30:28 -0700426 See Documentation/power/video.txt for information on
427 s3_bios and s3_mode.
428 s3_beep is for debugging; it makes the PC's speaker beep
429 as soon as the kernel's real-mode entry point is called.
430 s4_nohwsig prevents ACPI hardware signature from being
431 used during resume from hibernation.
432 old_ordering causes the ACPI 1.0 ordering of the _PTS
433 control method, with respect to putting devices into
434 low power states, to be enforced (the ACPI 2.0 ordering
435 of _PTS is used by default).
Rafael J. Wysocki72ad5d72010-07-23 22:59:09 +0200436 nonvs prevents the kernel from saving/restoring the
437 ACPI NVS memory during suspend/hibernation and resume.
Zhang Ruid7f0eea2009-12-30 15:36:42 +0800438 sci_force_enable causes the kernel to set SCI_EN directly
439 on resume from S1/S3 (which is against the ACPI spec,
440 but some broken systems don't work without it).
Randy Dunlap4af94f32009-04-17 18:30:28 -0700441
442 acpi_use_timer_override [HW,ACPI]
443 Use timer override. For some broken Nvidia NF5 boards
444 that require a timer override, but don't have HPET
445
Randy Dunlap4af94f32009-04-17 18:30:28 -0700446 add_efi_memmap [EFI; X86] Include EFI memory map in
447 kernel's map of available physical RAM.
448
Randy Dunlap0cb55ad2009-04-05 15:55:22 -0700449 agp= [AGP]
450 { off | try_unsupported }
451 off: disable AGP support
452 try_unsupported: try to drive unsupported chipsets
453 (may crash computer or cause data corruption)
454
Randy Dunlapbcfde332010-06-07 17:10:38 -0700455 ALSA [HW,ALSA]
456 See Documentation/sound/alsa/alsa-parameters.txt
457
Russell Kingd944d542010-02-20 16:13:29 +0000458 alignment= [KNL,ARM]
459 Allow the default userspace alignment fault handler
460 behaviour to be specified. Bit 0 enables warnings,
461 bit 1 enables fixups, and bit 2 sends a segfault.
462
Borislav Petkovdfb09f92011-08-05 15:15:08 +0200463 align_va_addr= [X86-64]
464 Align virtual addresses by clearing slice [14:12] when
465 allocating a VMA at process creation time. This option
466 gives you up to 3% performance improvement on AMD F15h
467 machines (where it is enabled by default) for a
468 CPU-intensive style benchmark, and it can vary highly in
469 a microbenchmark depending on workload and compiler.
470
Borislav Petkov8360ee22011-11-21 12:10:19 +0100471 32: only for 32-bit processes
472 64: only for 64-bit processes
Borislav Petkovdfb09f92011-08-05 15:15:08 +0200473 on: enable for both 32- and 64-bit processes
474 off: disable for both 32- and 64-bit processes
475
Steven Rostedt (Red Hat)55034cd2013-03-07 22:48:09 -0500476 alloc_snapshot [FTRACE]
477 Allocate the ftrace snapshot buffer on boot up when the
478 main buffer is allocated. This is handy if debugging
479 and you need to use tracing_snapshot() on boot up, and
480 do not want to use tracing_snapshot_alloc() as it needs
481 to be done where GFP_KERNEL allocations are allowed.
482
Sedat Dilek89e0b9a2011-12-05 23:08:32 +0100483 amd_iommu= [HW,X86-64]
Joerg Roedel54b4cbd2008-06-26 21:28:10 +0200484 Pass parameters to the AMD IOMMU driver in the system.
485 Possible values are:
FUJITA Tomonoriafa9fdc2008-09-20 01:23:30 +0900486 fullflush - enable flushing of IO/TLB entries when
487 they are unmapped. Otherwise they are
488 flushed before they will be reused, which
489 is a lot of faster
Joerg Roedela5235722010-05-11 17:12:33 +0200490 off - do not initialize any AMD IOMMU found in
491 the system
Joerg Roedel5abcdba2011-12-01 15:49:45 +0100492 force_isolation - Force device isolation for all
493 devices. The IOMMU driver is not
494 allowed anymore to lift isolation
495 requirements as needed. This option
496 does not override iommu=pt
FUJITA Tomonoriafa9fdc2008-09-20 01:23:30 +0900497
Shuah Khanc099cf12012-05-24 15:58:25 -0600498 amd_iommu_dump= [HW,X86-64]
499 Enable AMD IOMMU driver option to dump the ACPI table
500 for AMD IOMMU. With this option enabled, AMD IOMMU
501 driver will print ACPI tables for AMD IOMMU during
502 IOMMU initialization.
503
Suravee Suthikulpanit3928aa32016-08-23 13:52:32 -0500504 amd_iommu_intr= [HW,X86-64]
505 Specifies one of the following AMD IOMMU interrupt
506 remapping modes:
507 legacy - Use legacy interrupt remapping mode.
508 vapic - Use virtual APIC mode, which allows IOMMU
509 to inject interrupts directly into guest.
510 This mode requires kvm-amd.avic=1.
511 (Default when IOMMU HW support is present.)
512
Linus Torvalds1da177e2005-04-16 15:20:36 -0700513 amijoy.map= [HW,JOY] Amiga joystick support
514 Map of devices attached to JOY0DAT and JOY1DAT
515 Format: <a>,<b>
Paul Bolle395cf962011-08-15 02:02:26 +0200516 See also Documentation/input/joystick.txt
Linus Torvalds1da177e2005-04-16 15:20:36 -0700517
518 analog.map= [HW,JOY] Analog joystick and gamepad support
519 Specifies type or capabilities of an analog joystick
520 connected to one of 16 gameports
521 Format: <type1>,<type2>,..<type16>
522
Randy Dunlapa9913042005-10-23 12:57:11 -0700523 apc= [HW,SPARC]
524 Power management functions (SPARCstation-4/5 + deriv.)
Linus Torvalds1da177e2005-04-16 15:20:36 -0700525 Format: noidle
526 Disable APC CPU standby support. SPARCstation-Fox does
527 not play well with APC CPU idle - disable it if you have
528 APC and your system crashes randomly.
529
Randy Dunlapca1eda22009-05-06 16:02:58 -0700530 apic= [APIC,X86-32] Advanced Programmable Interrupt Controller
Randy Dunlapc8facbb2007-07-31 00:37:40 -0700531 Change the output verbosity whilst booting
Linus Torvalds1da177e2005-04-16 15:20:36 -0700532 Format: { quiet (default) | verbose | debug }
533 Change the amount of debugging information output
534 when initialising the APIC and IO-APIC components.
Randy Dunlapa9913042005-10-23 12:57:11 -0700535
Hidehiro Kawaib7c49482015-12-14 11:19:12 +0100536 apic_extnmi= [APIC,X86] External NMI delivery setting
537 Format: { bsp (default) | all | none }
538 bsp: External NMI is delivered only to CPU 0
539 all: External NMIs are broadcast to all CPUs as a
540 backup of CPU 0
541 none: External NMI is masked for all CPUs. This is
542 useful so that a dump capture kernel won't be
543 shot down by NMI
544
Brian Haleyb0f83b22010-02-04 13:36:50 -0800545 autoconf= [IPV6]
546 See Documentation/networking/ipv6.txt.
547
Cyrill Gorcunov9636bc02009-10-14 19:09:04 +0400548 show_lapic= [APIC,X86] Advanced Programmable Interrupt Controller
549 Limit apic dumping. The parameter defines the maximal
550 number of local apics being dumped. Also it is possible
551 to set it to "all" by meaning -- no limit here.
552 Format: { 1 (default) | 2 | ... | all }.
553 The parameter valid if only apic=debug or
554 apic=verbose is specified.
555 Example: apic=debug show_lapic=all
556
Linus Torvalds1da177e2005-04-16 15:20:36 -0700557 apm= [APM] Advanced Power Management
Pavel Machek71f77052008-07-04 09:59:43 -0700558 See header of arch/x86/kernel/apm_32.c.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700559
Linus Torvalds1da177e2005-04-16 15:20:36 -0700560 arcrimi= [HW,NET] ARCnet - "RIM I" (entirely mem-mapped) cards
561 Format: <io>,<irq>,<nodeID>
562
563 ataflop= [HW,M68k]
564
565 atarimouse= [HW,MOUSE] Atari Mouse
566
Linus Torvalds1da177e2005-04-16 15:20:36 -0700567 atkbd.extra= [HW] Enable extra LEDs and keys on IBM RapidAccess,
568 EzKey and similar keyboards
569
570 atkbd.reset= [HW] Reset keyboard during initialization
571
Randy Dunlapa9913042005-10-23 12:57:11 -0700572 atkbd.set= [HW] Select keyboard code set
573 Format: <int> (2 = AT (default), 3 = PS/2)
Linus Torvalds1da177e2005-04-16 15:20:36 -0700574
575 atkbd.scroll= [HW] Enable scroll wheel on MS Office and similar
576 keyboards
577
578 atkbd.softraw= [HW] Choose between synthetic and real raw mode
579 Format: <bool> (0 = real, 1 = synthetic (default))
Randy Dunlapa9913042005-10-23 12:57:11 -0700580
581 atkbd.softrepeat= [HW]
582 Use software keyboard repeat
Linus Torvalds1da177e2005-04-16 15:20:36 -0700583
Richard Guy Briggsa106fb02013-09-23 21:53:35 -0400584 audit= [KNL] Enable the audit sub-system
585 Format: { "0" | "1" } (0 = disabled, 1 = enabled)
Eric Parisd7961142014-01-13 16:01:06 -0500586 0 - kernel audit is disabled and can not be enabled
587 until the next reboot
588 unset - kernel audit is initialized but disabled and
589 will be fully enabled by the userspace auditd.
590 1 - kernel audit is initialized and partially enabled,
591 storing at most audit_backlog_limit messages in
592 RAM until it is fully enabled by the userspace
593 auditd.
Richard Guy Briggsa106fb02013-09-23 21:53:35 -0400594 Default: unset
Richard Guy Briggsf3411cb2013-09-17 12:34:52 -0400595
Richard Guy Briggsf910fde2013-09-17 12:34:52 -0400596 audit_backlog_limit= [KNL] Set the audit queue size limit.
597 Format: <int> (must be >=0)
598 Default: 64
599
Alex Thorlton1c532e02016-03-31 14:18:29 -0500600 bau= [X86_UV] Enable the BAU on SGI UV. The default
601 behavior is to disable the BAU (i.e. bau=0).
602 Format: { "0" | "1" }
603 0 - Disable the BAU.
604 1 - Enable the BAU.
605 unset - Disable the BAU.
606
Linus Torvalds1da177e2005-04-16 15:20:36 -0700607 baycom_epp= [HW,AX25]
608 Format: <io>,<mode>
Randy Dunlapa9913042005-10-23 12:57:11 -0700609
Linus Torvalds1da177e2005-04-16 15:20:36 -0700610 baycom_par= [HW,AX25] BayCom Parallel Port AX.25 Modem
611 Format: <io>,<mode>
612 See header of drivers/net/hamradio/baycom_par.c.
613
Randy Dunlapa9913042005-10-23 12:57:11 -0700614 baycom_ser_fdx= [HW,AX25]
615 BayCom Serial Port AX.25 Modem (Full Duplex Mode)
Linus Torvalds1da177e2005-04-16 15:20:36 -0700616 Format: <io>,<irq>,<mode>[,<baud>]
617 See header of drivers/net/hamradio/baycom_ser_fdx.c.
618
Randy Dunlapa9913042005-10-23 12:57:11 -0700619 baycom_ser_hdx= [HW,AX25]
620 BayCom Serial Port AX.25 Modem (Half Duplex Mode)
Linus Torvalds1da177e2005-04-16 15:20:36 -0700621 Format: <io>,<irq>,<mode>
622 See header of drivers/net/hamradio/baycom_ser_hdx.c.
623
Paul Gortmaker080506a2013-09-30 13:45:19 -0700624 blkdevparts= Manual partition parsing of block device(s) for
625 embedded devices based on command line input.
626 See Documentation/block/cmdline-partition.txt
627
Randy Dunlapbfe8df32007-10-16 01:23:46 -0700628 boot_delay= Milliseconds to delay each printk during boot.
629 Values larger than 10 seconds (10000) are changed to
630 no delay (0).
631 Format: integer
632
Andreas Herrmann35fc9082008-08-20 14:08:58 -0700633 bootmem_debug [KNL] Enable bootmem allocator debug messages.
634
Huang Yinga3e2acc2016-06-29 13:04:29 -0700635 bert_disable [ACPI]
636 Disable BERT OS support on buggy BIOSes.
637
Linus Torvalds1da177e2005-04-16 15:20:36 -0700638 bttv.card= [HW,V4L] bttv (bt848 + bt878 based grabber cards)
Randy Dunlapa9913042005-10-23 12:57:11 -0700639 bttv.radio= Most important insmod options are available as
640 kernel args too.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700641 bttv.pll= See Documentation/video4linux/bttv/Insmod-options
Paul Bolle395cf962011-08-15 02:02:26 +0200642 bttv.tuner=
Linus Torvalds1da177e2005-04-16 15:20:36 -0700643
Will Schmidt4e89a2d2010-09-28 15:33:12 +0000644 bulk_remove=off [PPC] This parameter disables the use of the pSeries
645 firmware feature for flushing multiple hpte entries
646 at a time.
647
Linus Torvalds1da177e2005-04-16 15:20:36 -0700648 c101= [NET] Moxa C101 synchronous serial card
649
Alan Coxcd4f0ef2007-07-31 00:37:59 -0700650 cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700651 Sometimes CPU hardware bugs make them report the cache
652 size incorrectly. The kernel will attempt work arounds
653 to fix known problems, but for some CPUs it is not
654 possible to determine what the correct size should be.
655 This option provides an override for these situations.
656
Dmitry Kasatkinffb70f62014-06-17 11:56:58 +0300657 ca_keys= [KEYS] This parameter identifies a specific key(s) on
658 the system trusted keyring to be used for certificate
659 trust validation.
Dmitry Kasatkin32c47412014-06-17 11:56:59 +0300660 format: { id:<keyid> | builtin }
Dmitry Kasatkinffb70f62014-06-17 11:56:58 +0300661
Florian Fainellifd1bb4c2014-06-25 16:41:13 -0700662 cca= [MIPS] Override the kernel pages' cache coherency
663 algorithm. Accepted values range from 0 to 7
664 inclusive. See arch/mips/include/asm/pgtable-bits.h
665 for platform specific values (SB1, Loongson3 and
666 others).
667
Sebastian Ott14ff56b2008-01-26 14:10:37 +0100668 ccw_timeout_log [S390]
669 See Documentation/s390/CommonIO for details.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700670
Paul Menage8bab8dd2008-04-04 14:29:57 -0700671 cgroup_disable= [KNL] Disable a particular controller
672 Format: {name of the controller(s) to disable}
Qiang Huangca0bdbb2013-11-06 13:18:09 -0800673 The effects of cgroup_disable=foo are:
674 - foo isn't auto-mounted if you mount all cgroups in
675 a single hierarchy
676 - foo isn't visible as an individually mountable
677 subsystem
678 {Currently only "memory" controller deal with this and
679 cut the overhead, others just disable the usage. So
680 only cgroup_disable=memory is actually worthy}
Paul Menage8bab8dd2008-04-04 14:29:57 -0700681
Johannes Weiner1619b6d2016-02-16 13:21:14 -0500682 cgroup_no_v1= [KNL] Disable one, multiple, all cgroup controllers in v1
683 Format: { controller[,controller...] | "all" }
684 Like cgroup_disable, but only applies to cgroup v1;
685 the blacklisted controllers remain available in cgroup2.
686
Johannes Weinerf7e1cb62016-01-14 15:21:29 -0800687 cgroup.memory= [KNL] Pass options to the cgroup memory controller.
688 Format: <string>
689 nosocket -- Disable socket memory accounting.
Vladimir Davydov04823c82016-01-20 15:02:38 -0800690 nokmem -- Disable kernel memory accounting.
Johannes Weinerf7e1cb62016-01-14 15:21:29 -0800691
Linus Torvalds1da177e2005-04-16 15:20:36 -0700692 checkreqprot [SELINUX] Set initial checkreqprot flag value.
693 Format: { "0" | "1" }
694 See security/selinux/Kconfig help text.
Randy Dunlapa9913042005-10-23 12:57:11 -0700695 0 -- check protection applied by kernel (includes
696 any implied execute protection).
Linus Torvalds1da177e2005-04-16 15:20:36 -0700697 1 -- check protection requested by application.
698 Default value is set via a kernel config option.
Randy Dunlapa9913042005-10-23 12:57:11 -0700699 Value can be changed at runtime via
700 /selinux/checkreqprot.
701
Sebastian Ott661ca0d2008-01-26 14:10:36 +0100702 cio_ignore= [S390]
703 See Documentation/s390/CommonIO for details.
Olof Johansson1e435252013-04-27 14:10:18 -0700704 clk_ignore_unused
705 [CLK]
Mike Turquettee156ee52014-09-30 14:24:38 -0700706 Prevents the clock framework from automatically gating
707 clocks that have not been explicitly enabled by a Linux
708 device driver but are enabled in hardware at reset or
709 by the bootloader/firmware. Note that this does not
710 force such clocks to be always-on nor does it reserve
711 those clocks in any way. This parameter is useful for
712 debug and development, but should not be needed on a
713 platform with proper driver support. For more
714 information, see Documentation/clk.txt.
Sebastian Ott661ca0d2008-01-26 14:10:36 +0100715
Alan Coxcd4f0ef2007-07-31 00:37:59 -0700716 clock= [BUGS=X86-32, HW] gettimeofday clocksource override.
john stultz734efb42006-06-26 00:25:05 -0700717 [Deprecated]
Matt LaPlante3f6dee92006-10-03 22:45:33 +0200718 Forces specified clocksource (if available) to be used
john stultz734efb42006-06-26 00:25:05 -0700719 when calculating gettimeofday(). If specified
Matt LaPlante3f6dee92006-10-03 22:45:33 +0200720 clocksource is not available, it defaults to PIT.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700721 Format: { pit | tsc | cyclone | pmtmr }
722
John Stultz592913e2010-07-13 17:56:20 -0700723 clocksource= Override the default clocksource
Randy Dunlap3d6ac982007-05-23 13:58:16 -0700724 Format: <string>
725 Override the default clocksource and use the clocksource
726 with the name specified.
727 Some clocksource names to choose from, depending on
728 the platform:
729 [all] jiffies (this is the base, fallback clocksource)
730 [ACPI] acpi_pm
731 [ARM] imx_timer1,OSTS,netx_timer,mpu_timer2,
732 pxa_timer,timer3,32k_counter,timer0_1
733 [AVR32] avr32
Alok Kataria9863c902010-08-23 14:49:11 -0700734 [X86-32] pit,hpet,tsc;
Randy Dunlap3d6ac982007-05-23 13:58:16 -0700735 scx200_hrt on Geode; cyclone on IBM x440
736 [MIPS] MIPS
737 [PARISC] cr16
738 [S390] tod
739 [SH] SuperH
740 [SPARC64] tick
741 [X86-64] hpet,tsc
742
Will Deacon46fd5c62016-06-27 17:30:13 +0100743 clocksource.arm_arch_timer.evtstrm=
744 [ARM,ARM64]
745 Format: <bool>
746 Enable/disable the eventstream feature of the ARM
747 architected timer so that code using WFE-based polling
748 loops can be debugged more effectively on production
749 systems.
750
Scott Woodf6dc1572016-09-22 03:35:17 -0500751 clocksource.arm_arch_timer.fsl-a008585=
752 [ARM64]
753 Format: <bool>
754 Enable/disable the workaround of Freescale/NXP
755 erratum A-008585. This can be useful for KVM
756 guests, if the guest device tree doesn't show the
757 erratum. If unspecified, the workaround is
758 enabled based on the device tree.
759
Andi Kleenac72e782008-01-30 13:33:21 +0100760 clearcpuid=BITNUM [X86]
761 Disable CPUID feature X for the kernel. See
Borislav Petkovcd4d09e2016-01-26 22:12:04 +0100762 arch/x86/include/asm/cpufeatures.h for the valid bit
Randy Dunlap07983f02009-01-06 14:42:41 -0800763 numbers. Note the Linux specific bits are not necessarily
Andi Kleenac72e782008-01-30 13:33:21 +0100764 stable over kernel options, but the vendor specific
765 ones should be.
766 Also note that user programs calling CPUID directly
767 or using the feature without checking anything
768 will still see it. This just prevents it from
769 being used by the kernel or shown in /proc/cpuinfo.
770 Also note the kernel might malfunction if you disable
771 some critical bits.
772
Akinobu Mita5ea3b1b2014-06-04 16:06:54 -0700773 cma=nn[MG]@[start[MG][-end[MG]]]
774 [ARM,X86,KNL]
775 Sets the size of kernel global memory area for
776 contiguous memory allocations and optionally the
777 placement constraint by the physical address range of
Jean Delvaref0d6d1f2014-10-09 15:29:41 -0700778 memory allocations. A value of 0 disables CMA
779 altogether. For more information, see
Marek Szyprowskic64be2b2011-12-29 13:09:51 +0100780 include/linux/dma-contiguous.h
781
Robert Jennings14f966e2009-04-15 05:55:32 +0000782 cmo_free_hint= [PPC] Format: { yes | no }
783 Specify whether pages are marked as being inactive
784 when they are freed. This is used in CMO environments
785 to determine OS memory pressure for page stealing by
786 a hypervisor.
787 Default: yes
788
Marek Szyprowskic7909502011-12-29 13:09:51 +0100789 coherent_pool=nn[KMG] [ARM,KNL]
790 Sets the size of memory pool for coherent, atomic dma
Marek Szyprowskie9da6e92012-07-30 09:11:33 +0200791 allocations, by default set to 256K.
Marek Szyprowskic7909502011-12-29 13:09:51 +0100792
Jaswinder Singh Rajput6cececf2009-04-14 14:03:43 +0530793 code_bytes [X86] How many bytes of object code to print
Arjan van de Vena25bd942008-01-30 13:33:08 +0100794 in an oops report.
Chuck Ebbert86c41832007-02-13 13:26:25 +0100795 Range: 0 - 8192
796 Default: 64
797
Linus Torvalds1da177e2005-04-16 15:20:36 -0700798 com20020= [HW,NET] ARCnet - COM20020 chipset
Randy Dunlapa9913042005-10-23 12:57:11 -0700799 Format:
800 <io>[,<irq>[,<nodeID>[,<backplane>[,<ckp>[,<timeout>]]]]]
Linus Torvalds1da177e2005-04-16 15:20:36 -0700801
802 com90io= [HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers)
803 Format: <io>[,<irq>]
804
Randy Dunlapa9913042005-10-23 12:57:11 -0700805 com90xx= [HW,NET]
806 ARCnet - COM90xx chipset (memory-mapped buffers)
Linus Torvalds1da177e2005-04-16 15:20:36 -0700807 Format: <io>[,<irq>[,<memstart>]]
808
809 condev= [HW,S390] console device
810 conmode=
Randy Dunlapa9913042005-10-23 12:57:11 -0700811
Linus Torvalds1da177e2005-04-16 15:20:36 -0700812 console= [KNL] Output console device and options.
813
814 tty<n> Use the virtual console device <n>.
815
816 ttyS<n>[,options]
Randy Dunlapf1a1c2d2006-03-25 03:08:17 -0800817 ttyUSB0[,options]
Linus Torvalds1da177e2005-04-16 15:20:36 -0700818 Use the specified serial port. The options are of
Randy Dunlapf1a1c2d2006-03-25 03:08:17 -0800819 the form "bbbbpnf", where "bbbb" is the baud rate,
820 "p" is parity ("n", "o", or "e"), "n" is number of
821 bits, and "f" is flow control ("r" for RTS or
822 omit it). Default is "9600n8".
Linus Torvalds1da177e2005-04-16 15:20:36 -0700823
Randy Dunlapf1a1c2d2006-03-25 03:08:17 -0800824 See Documentation/serial-console.txt for more
825 information. See
826 Documentation/networking/netconsole.txt for an
827 alternative.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700828
Yinghai Lu18a8bd92007-07-15 23:37:59 -0700829 uart[8250],io,<addr>[,options]
830 uart[8250],mmio,<addr>[,options]
Masahiro Yamadabd94c402015-10-28 12:46:05 +0900831 uart[8250],mmio16,<addr>[,options]
Peter Hurleyca782f12015-04-06 10:52:39 -0400832 uart[8250],mmio32,<addr>[,options]
833 uart[8250],0x<addr>[,options]
Linus Torvalds1da177e2005-04-16 15:20:36 -0700834 Start an early, polled-mode console on the 8250/16550
835 UART at the specified I/O port or MMIO address,
Peter Hurleyca782f12015-04-06 10:52:39 -0400836 switching to the matching ttyS device later.
837 MMIO inter-register address stride is either 8-bit
Masahiro Yamadabd94c402015-10-28 12:46:05 +0900838 (mmio), 16-bit (mmio16), or 32-bit (mmio32).
839 If none of [io|mmio|mmio16|mmio32], <addr> is assumed
840 to be equivalent to 'mmio'. 'options' are specified in
841 the same format described for ttyS above; if unspecified,
Peter Hurleyca782f12015-04-06 10:52:39 -0400842 the h/w is not re-initialized.
843
Konrad Rzeszutek Wilka2fd6412013-02-25 15:54:09 -0500844 hvc<n> Use the hypervisor console device <n>. This is for
845 both Xen and PowerPC hypervisors.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700846
Samuel Thibaultf7511d52008-04-30 00:54:51 -0700847 If the device connected to the port is not a TTY but a braille
848 device, prepend "brl," before the device type, for instance
849 console=brl,ttyS0
850 For now, only VisioBraille is supported.
851
Daniel Mackf324edc2009-06-16 15:33:52 -0700852 consoleblank= [KNL] The console blank (screen saver) timeout in
853 seconds. Defaults to 10*60 = 10mins. A value of 0
854 disables the blank timer.
855
Hidehiro Kawai4cb0e112009-01-06 14:42:47 -0800856 coredump_filter=
857 [KNL] Change the default value for
858 /proc/<pid>/coredump_filter.
859 See also Documentation/filesystems/proc.txt.
860
Len Brown62027ae2011-04-01 18:13:10 -0400861 cpuidle.off=1 [CPU_IDLE]
862 disable the cpuidle sub-system
863
Len Brownd68921f2015-05-11 17:27:09 -0400864 cpu_init_udelay=N
865 [X86] Delay for N microsec between assert and de-assert
866 of APIC INIT to start processors. This delay occurs
867 on every CPU online, such as boot, and resume from suspend.
868 Default: 10000
869
Linus Torvalds1da177e2005-04-16 15:20:36 -0700870 cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver
Randy Dunlapa9913042005-10-23 12:57:11 -0700871 Format:
872 <first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>]
Linus Torvalds1da177e2005-04-16 15:20:36 -0700873
Ahmed S. Darwish6f21e642011-02-20 20:08:35 -0800874 crashkernel=size[KMG][@offset[KMG]]
875 [KNL] Using kexec, Linux can switch to a 'crash kernel'
876 upon panic. This parameter reserves the physical
877 memory region [offset, offset + size] for that kernel
878 image. If '@offset' is omitted, then a suitable offset
879 is selected automatically. Check
880 Documentation/kdump/kdump.txt for further details.
Eric W. Biedermandc009d92005-06-25 14:57:52 -0700881
Bernhard Wallefb391592007-10-18 23:41:02 -0700882 crashkernel=range1:size1[,range2:size2,...][@offset]
883 [KNL] Same as above, but depends on the memory
884 in the running system. The syntax of range is
885 start-[end] where start and end are both
886 a memory unit (amount[KMG]). See also
Ahmed S. Darwish6f21e642011-02-20 20:08:35 -0800887 Documentation/kdump/kdump.txt for an example.
Bernhard Wallefb391592007-10-18 23:41:02 -0700888
Yinghai Luadbc7422013-04-15 22:23:48 -0700889 crashkernel=size[KMG],high
Yinghai Lu55a20ee2013-04-15 22:23:47 -0700890 [KNL, x86_64] range could be above 4G. Allow kernel
891 to allocate physical memory region from top, so could
892 be above 4G if system have more than 4G ram installed.
893 Otherwise memory region will be allocated below 4G, if
894 available.
895 It will be ignored if crashkernel=X is specified.
Yinghai Luadbc7422013-04-15 22:23:48 -0700896 crashkernel=size[KMG],low
897 [KNL, x86_64] range under 4G. When crashkernel=X,high
898 is passed, kernel could allocate physical memory region
Yinghai Luc729de82013-04-15 22:23:45 -0700899 above 4G, that cause second kernel crash on system
900 that require some amount of low memory, e.g. swiotlb
Baoquan Hec6045032015-09-24 16:51:25 +0800901 requires at least 64M+32K low memory, also enough extra
902 low memory is needed to make sure DMA buffers for 32-bit
903 devices won't run out. Kernel would try to allocate at
904 at least 256M below 4G automatically.
Yinghai Luc729de82013-04-15 22:23:45 -0700905 This one let user to specify own low range under 4G
906 for second kernel instead.
907 0: to disable low allocation.
Yinghai Luadbc7422013-04-15 22:23:48 -0700908 It will be ignored when crashkernel=X,high is not used
Yinghai Lu55a20ee2013-04-15 22:23:47 -0700909 or memory reserved is below 4G.
Yinghai Luc729de82013-04-15 22:23:45 -0700910
Richard W.M. Jones9e5c9fe2016-05-03 10:00:17 +0100911 cryptomgr.notests
912 [KNL] Disable crypto self-tests
913
Linus Torvalds1da177e2005-04-16 15:20:36 -0700914 cs89x0_dma= [HW,NET]
915 Format: <dma>
916
917 cs89x0_media= [HW,NET]
918 Format: { rj45 | aui | bnc }
Randy Dunlapa9913042005-10-23 12:57:11 -0700919
Randy Dunlapa9913042005-10-23 12:57:11 -0700920 dasd= [HW,NET]
Linus Torvalds1da177e2005-04-16 15:20:36 -0700921 See header of drivers/s390/block/dasd_devmap.c.
922
923 db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port
924 (one device per port)
925 Format: <port#>,<type>
926 See also Documentation/input/joystick-parport.txt
927
Thomas Renningera648ec02010-08-06 16:11:02 +0200928 ddebug_query= [KNL,DYNAMIC_DEBUG] Enable debug messages at early boot
929 time. See Documentation/dynamic-debug-howto.txt for
Jim Cromie29e36c92012-04-27 14:30:41 -0600930 details. Deprecated, see dyndbg.
Thomas Renningera648ec02010-08-06 16:11:02 +0200931
Linus Torvalds1da177e2005-04-16 15:20:36 -0700932 debug [KNL] Enable kernel debugging (events log level).
933
Ingo Molnarcae2ed92006-07-03 00:24:48 -0700934 debug_locks_verbose=
935 [KNL] verbose self-tests
936 Format=<0|1>
937 Print debugging info while doing the locking API
938 self-tests.
939 We default to 0 (no extra messages), setting it to
940 1 will print _a lot_ more information - normally
941 only useful to kernel developers.
942
Thomas Gleixner3ac7fe52008-04-30 00:55:01 -0700943 debug_objects [KNL] Enable object debugging
944
Kyle McMartin3e8ebb52009-03-01 20:41:41 -0500945 no_debug_objects
946 [KNL] Disable object debugging
947
Stanislaw Gruszkac0a32fc52012-01-10 15:07:28 -0800948 debug_guardpage_minorder=
949 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this
950 parameter allows control of the order of pages that will
951 be intentionally kept free (and hence protected) by the
952 buddy allocator. Bigger value increase the probability
953 of catching random memory corruption, but reduce the
954 amount of memory for normal system use. The maximum
955 possible value is MAX_ORDER/2. Setting this parameter
956 to 1 or 2 should be enough to identify most random
957 memory corruption problems caused by bugs in kernel or
958 driver code when a CPU writes to (or reads from) a
959 random memory location. Note that there exists a class
960 of memory corruptions problems caused by buggy H/W or
961 F/W or by drivers badly programing DMA (basically when
962 memory is written at bus level and the CPU MMU is
963 bypassed) which are not detectable by
964 CONFIG_DEBUG_PAGEALLOC, hence this option will not help
965 tracking down these problems.
966
Joonsoo Kim031bc572014-12-12 16:55:52 -0800967 debug_pagealloc=
968 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this
969 parameter enables the feature at boot time. In
970 default, it is disabled. We can avoid allocating huge
971 chunk of memory for debug pagealloc if we don't enable
972 it at boot time and the system will work mostly same
973 with the kernel built without CONFIG_DEBUG_PAGEALLOC.
974 on: enable the feature
975
Thomas Gleixnerd3af01f2008-07-15 15:04:56 +0200976 debugpat [X86] Enable PAT debugging
977
Robert P. J. Day2d27a962008-02-03 15:18:45 +0200978 decnet.addr= [HW,NET]
Linus Torvalds1da177e2005-04-16 15:20:36 -0700979 Format: <area>[,<node>]
980 See also Documentation/networking/decnet.txt.
981
Randy Dunlap0cb55ad2009-04-05 15:55:22 -0700982 default_hugepagesz=
983 [same as hugepagesz=] The size of the default
984 HugeTLB page size. This is the size represented by
985 the legacy /proc/ hugepages APIs, used for SHM, and
986 default size when mounting hugetlbfs filesystems.
987 Defaults to the default architecture's huge page size
988 if not specified.
Antonino A. Daplas55ff9782007-05-08 00:38:53 -0700989
Linus Torvalds1da177e2005-04-16 15:20:36 -0700990 dhash_entries= [KNL]
991 Set number of hash buckets for dentry cache.
Randy Dunlapa9913042005-10-23 12:57:11 -0700992
Oliver O'Halloranfaf78822016-07-05 11:43:21 +1000993 disable_1tb_segments [PPC]
994 Disables the use of 1TB hash page table segments. This
995 causes the kernel to fall back to 256MB segments which
996 can be useful when debugging issues that require an SLB
997 miss to occur.
998
Brian Haleyb0f83b22010-02-04 13:36:50 -0800999 disable= [IPV6]
1000 See Documentation/networking/ipv6.txt.
1001
Aneesh Kumar K.Vb275bfb2016-07-13 15:05:31 +05301002 disable_radix [PPC]
1003 Disable RADIX MMU mode on POWER9
1004
HATAYAMA Daisuke151e0c72014-01-15 15:44:58 +09001005 disable_cpu_apicid= [X86,APIC,SMP]
1006 Format: <int>
1007 The number of initial APIC ID for the
1008 corresponding CPU to be disabled at boot,
1009 mostly used for the kdump 2nd kernel to
1010 disable BSP to wake up multiple CPUs without
1011 causing system reset or hang due to sending
1012 INIT from AP to BSP.
1013
Nishanth Aravamudan4e8b0cf2011-02-10 09:10:47 +00001014 disable_ddw [PPC/PSERIES]
1015 Disable Dynamic DMA Window support. Use this if
1016 to workaround buggy firmware.
1017
Brian Haleyb0f83b22010-02-04 13:36:50 -08001018 disable_ipv6= [IPV6]
1019 See Documentation/networking/ipv6.txt.
1020
Yinghai Lu95ffa242008-04-29 03:52:33 -07001021 disable_mtrr_cleanup [X86]
Yinghai Lu95ffa242008-04-29 03:52:33 -07001022 The kernel tries to adjust MTRR layout from continuous
1023 to discrete, to make X server driver able to add WB
Randy Dunlap0cb55ad2009-04-05 15:55:22 -07001024 entry later. This parameter disables that.
Yinghai Lu95ffa242008-04-29 03:52:33 -07001025
Yinghai Lu093af8d2008-01-30 13:33:32 +01001026 disable_mtrr_trim [X86, Intel and AMD only]
Jesse Barnes99fc8d42008-01-30 13:33:18 +01001027 By default the kernel will trim any uncacheable
1028 memory out of your available memory pool based on
1029 MTRR settings. This parameter disables that behavior,
1030 possibly causing your machine to run very slowly.
1031
Jaswinder Singh Rajput6cececf2009-04-14 14:03:43 +05301032 disable_timer_pin_1 [X86]
Randy Dunlap0cb55ad2009-04-05 15:55:22 -07001033 Disable PIN 1 of APIC timer
1034 Can be useful to work around chipset bugs.
1035
Prarit Bhargavace14c582015-08-25 13:34:53 -04001036 dis_ucode_ldr [X86] Disable the microcode loader.
1037
Will Drewrya058da82010-06-09 17:47:38 -05001038 dm= [DM] Allows early creation of a device-mapper device.
1039 See Documentation/device-mapper/boot.txt.
1040
1041 dmasound= [HW,OSS] Sound subsystem buff
1042
Randy Dunlap0cb55ad2009-04-05 15:55:22 -07001043 dma_debug=off If the kernel is compiled with DMA_API_DEBUG support,
1044 this option disables the debugging code at boot.
1045
1046 dma_debug_entries=<number>
1047 This option allows to tune the number of preallocated
1048 entries for DMA-API debugging code. One entry is
1049 required per DMA-API allocation. Use this if the
1050 DMA-API debugging code disables itself because the
1051 architectural default is too low.
1052
Joerg Roedel1745de52009-05-22 21:49:51 +02001053 dma_debug_driver=<driver_name>
1054 With this option the DMA-API debugging driver
1055 filter feature can be enabled at boot time. Just
1056 pass the driver to filter for as the parameter.
1057 The filter can be disabled or changed to another
1058 driver later using sysfs.
1059
Bob Paauwe96206e22015-08-27 10:04:13 -07001060 drm_kms_helper.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>]
1061 Broken monitors, graphic adapters, KVMs and EDIDless
1062 panels may send no or incorrect EDID data sets.
1063 This parameter allows to specify an EDID data sets
1064 in the /lib/firmware directory that are used instead.
Carsten Emdeda0df922012-03-18 22:37:33 +01001065 Generic built-in EDID data sets are used, if one of
1066 edid/1024x768.bin, edid/1280x1024.bin,
1067 edid/1680x1050.bin, or edid/1920x1080.bin is given
1068 and no file with the same name exists. Details and
1069 instructions how to build your own EDID data are
1070 available in Documentation/EDID/HOWTO.txt. An EDID
1071 data set will only be used for a particular connector,
1072 if its name and a colon are prepended to the EDID
Bob Paauwe96206e22015-08-27 10:04:13 -07001073 name. Each connector may use a unique EDID data
1074 set by separating the files with a comma. An EDID
1075 data set with no connector name will be used for
1076 any connectors not explicitly specified.
Carsten Emdeda0df922012-03-18 22:37:33 +01001077
Linus Torvalds1da177e2005-04-16 15:20:36 -07001078 dscc4.setup= [NET]
1079
Jim Cromie29e36c92012-04-27 14:30:41 -06001080 dyndbg[="val"] [KNL,DYNAMIC_DEBUG]
1081 module.dyndbg[="val"]
1082 Enable debug messages at boot time. See
1083 Documentation/dynamic-debug-howto.txt for details.
1084
Dave Hansen8c3641e2015-06-07 11:37:02 -07001085 nompx [X86] Disables Intel Memory Protection Extensions.
1086 See Documentation/x86/intel_mpx.txt for more
1087 information about the feature.
1088
Dave Hansen06976942016-02-12 13:02:29 -08001089 nopku [X86] Disable Memory Protection Keys CPU feature found
1090 in some Intel CPUs.
1091
Luis R. Rodriguezf2411da2015-03-30 16:20:05 -07001092 module.async_probe [KNL]
1093 Enable asynchronous probe on this module.
1094
Mark Salter56aeeba2014-04-07 15:39:53 -07001095 early_ioremap_debug [KNL]
1096 Enable debug messages in early_ioremap support. This
1097 is useful for tracking down temporary early mappings
1098 which are not unmapped.
1099
Randy Dunlap0cb55ad2009-04-05 15:55:22 -07001100 earlycon= [KNL] Output early console device and options.
Rob Herring0d3c6732014-04-18 17:19:57 -05001101
Scott Wood5664f762015-09-14 19:54:07 -05001102 When used with no options, the early console is
1103 determined by the stdout-path property in device
1104 tree's chosen node.
1105
Scott Telfordc41251b2016-09-22 16:58:16 +01001106 cdns,<addr>[,options]
1107 Start an early, polled-mode console on a Cadence
1108 (xuartps) serial port at the specified address. Only
1109 supported option is baud rate. If baud rate is not
1110 specified, the serial port must already be setup and
1111 configured.
Michal Simek6fa62fc2014-09-10 12:43:02 +02001112
Randy Dunlap0cb55ad2009-04-05 15:55:22 -07001113 uart[8250],io,<addr>[,options]
1114 uart[8250],mmio,<addr>[,options]
Samium Gromoff1917ac72010-07-20 15:26:51 -07001115 uart[8250],mmio32,<addr>[,options]
Noam Camus6e63be32015-05-25 06:54:28 +03001116 uart[8250],mmio32be,<addr>[,options]
Peter Hurleyca782f12015-04-06 10:52:39 -04001117 uart[8250],0x<addr>[,options]
Randy Dunlap0cb55ad2009-04-05 15:55:22 -07001118 Start an early, polled-mode console on the 8250/16550
1119 UART at the specified I/O port or MMIO address.
Randy Dunlap16290242011-08-13 12:34:52 -07001120 MMIO inter-register address stride is either 8-bit
Noam Camus6e63be32015-05-25 06:54:28 +03001121 (mmio) or 32-bit (mmio32 or mmio32be).
1122 If none of [io|mmio|mmio32|mmio32be], <addr> is assumed
1123 to be equivalent to 'mmio'. 'options' are specified
1124 in the same format described for "console=ttyS<n>"; if
Peter Hurleyca782f12015-04-06 10:52:39 -04001125 unspecified, the h/w is not initialized.
Randy Dunlap0cb55ad2009-04-05 15:55:22 -07001126
Rob Herring0d3c6732014-04-18 17:19:57 -05001127 pl011,<addr>
Timur Tabi3b78fae2016-01-04 15:37:42 -06001128 pl011,mmio32,<addr>
Rob Herring0d3c6732014-04-18 17:19:57 -05001129 Start an early, polled-mode console on a pl011 serial
1130 port at the specified address. The pl011 serial port
1131 must already be setup and configured. Options are not
Timur Tabi3b78fae2016-01-04 15:37:42 -06001132 yet supported. If 'mmio32' is specified, then only
1133 the driver will use only 32-bit accessors to read/write
1134 the device registers.
Rob Herring0d3c6732014-04-18 17:19:57 -05001135
Andreas Färber736d5532016-03-06 12:21:24 +01001136 meson,<addr>
1137 Start an early, polled-mode console on a meson serial
1138 port at the specified address. The serial port must
1139 already be setup and configured. Options are not yet
1140 supported.
1141
Stephen Boyd0efe7292014-09-15 17:22:51 -07001142 msm_serial,<addr>
1143 Start an early, polled-mode console on an msm serial
1144 port at the specified address. The serial port
1145 must already be setup and configured. Options are not
1146 yet supported.
1147
1148 msm_serial_dm,<addr>
1149 Start an early, polled-mode console on an msm serial
1150 dm port at the specified address. The serial port
1151 must already be setup and configured. Options are not
1152 yet supported.
1153
Rob Herringd50d7262014-04-18 17:19:58 -05001154 smh Use ARM semihosting calls for early console.
1155
Tomasz Figab94ba032015-01-23 14:47:41 +01001156 s3c2410,<addr>
1157 s3c2412,<addr>
1158 s3c2440,<addr>
1159 s3c6400,<addr>
1160 s5pv210,<addr>
1161 exynos4210,<addr>
1162 Use early console provided by serial driver available
1163 on Samsung SoCs, requires selecting proper type and
1164 a correct base address of the selected UART port. The
1165 serial port must already be setup and configured.
1166 Options are not yet supported.
1167
Stefan Agner1d59b382015-10-17 00:45:55 -07001168 lpuart,<addr>
1169 lpuart32,<addr>
1170 Use early console provided by Freescale LP UART driver
1171 found on Freescale Vybrid and QorIQ LS1021A processors.
1172 A valid base address must be provided, and the serial
1173 port must already be setup and configured.
1174
Wilson Ding30530792016-02-16 19:14:53 +01001175 armada3700_uart,<addr>
1176 Start an early, polled-mode console on the
1177 Armada 3700 serial port at the specified
1178 address. The serial port must already be setup
1179 and configured. Options are not yet supported.
1180
Finn Thain7913ad12014-04-12 23:48:56 +10001181 earlyprintk= [X86,SH,BLACKFIN,ARM,M68k]
Linus Torvalds1da177e2005-04-16 15:20:36 -07001182 earlyprintk=vga
Matt Fleming72548e82013-10-04 09:36:56 +01001183 earlyprintk=efi
Konrad Rzeszutek Wilk2482a922013-02-25 15:54:08 -05001184 earlyprintk=xen
Linus Torvalds1da177e2005-04-16 15:20:36 -07001185 earlyprintk=serial[,ttySn[,baudrate]]
Dave Hansen147ea092013-04-10 14:03:38 -07001186 earlyprintk=serial[,0x...[,baudrate]]
Jason Wesselea3acb12009-09-24 09:08:30 -05001187 earlyprintk=ttySn[,baudrate]
Jason Wessel9780bc42009-08-20 15:39:57 -05001188 earlyprintk=dbgp[debugController#]
Alexander Kuleshovc43088e2015-06-10 23:21:15 +06001189 earlyprintk=pciserial,bus:device.function[,baudrate]
Linus Torvalds1da177e2005-04-16 15:20:36 -07001190
Dave Hansen147ea092013-04-10 14:03:38 -07001191 earlyprintk is useful when the kernel crashes before
1192 the normal console is initialized. It is not enabled by
1193 default because it has some cosmetic problems.
1194
Randy Dunlapa9913042005-10-23 12:57:11 -07001195 Append ",keep" to not disable it when the real console
Linus Torvalds1da177e2005-04-16 15:20:36 -07001196 takes over.
1197
Matt Fleming72548e82013-10-04 09:36:56 +01001198 Only one of vga, efi, serial, or usb debug port can
1199 be used at a time.
Linus Torvalds1da177e2005-04-16 15:20:36 -07001200
Dave Hansen147ea092013-04-10 14:03:38 -07001201 Currently only ttyS0 and ttyS1 may be specified by
1202 name. Other I/O ports may be explicitly specified
1203 on some architectures (x86 and arm at least) by
1204 replacing ttySn with an I/O port address, like this:
1205 earlyprintk=serial,0x1008,115200
1206 You can find the port for a given device in
1207 /proc/tty/driver/serial:
1208 2: uart:ST16650V2 port:00001008 irq:18 ...
Linus Torvalds1da177e2005-04-16 15:20:36 -07001209
1210 Interaction with the standard serial driver is not
1211 very good.
1212
Matt Fleming72548e82013-10-04 09:36:56 +01001213 The VGA and EFI output is eventually overwritten by
1214 the real console.
Linus Torvalds1da177e2005-04-16 15:20:36 -07001215
Konrad Rzeszutek Wilk2482a922013-02-25 15:54:08 -05001216 The xen output can only be used by Xen PV guests.
1217
Chen, Gongc700f012013-12-06 01:17:08 -05001218 edac_report= [HW,EDAC] Control how to report EDAC event
1219 Format: {"on" | "off" | "force"}
1220 on: enable EDAC to report H/W event. May be overridden
1221 by other higher priority error reporting module.
1222 off: disable H/W event reporting through EDAC.
1223 force: enforce the use of EDAC to report H/W event.
1224 default: on.
1225
Jason Wessel97311912010-05-20 21:04:30 -05001226 ekgdboc= [X86,KGDB] Allow early kernel console debugging
1227 ekgdboc=kbd
1228
Lucas De Marchi25985ed2011-03-30 22:57:33 -03001229 This is designed to be used in conjunction with
Jason Wessel97311912010-05-20 21:04:30 -05001230 the boot argument: earlyprintk=vga
1231
Linus Torvalds1da177e2005-04-16 15:20:36 -07001232 edd= [EDD]
Tim Gardner8c4dd602008-04-29 01:02:45 -07001233 Format: {"off" | "on" | "skip[mbr]"}
Linus Torvalds1da177e2005-04-16 15:20:36 -07001234
Borislav Petkovd2f7cbe2013-10-31 17:25:08 +01001235 efi= [EFI]
Borislav Petkovfed6cef2015-02-05 11:44:41 +01001236 Format: { "old_map", "nochunk", "noruntime", "debug" }
Borislav Petkovd2f7cbe2013-10-31 17:25:08 +01001237 old_map [X86-64]: switch to the old ioremap-based EFI
1238 runtime services mapping. 32-bit still uses this one by
1239 default.
Matt Fleming5a17dae2014-08-05 11:52:11 +01001240 nochunk: disable reading files in "chunks" in the EFI
1241 boot stub, as chunking can cause problems with some
1242 firmware implementations.
Dave Young5ae36832014-08-14 17:15:28 +08001243 noruntime : disable EFI runtime services support
Borislav Petkovfed6cef2015-02-05 11:44:41 +01001244 debug: enable misc debug output
Borislav Petkovd2f7cbe2013-10-31 17:25:08 +01001245
Richard Weinberger8c58bf32013-04-17 01:00:53 +02001246 efi_no_storage_paranoia [EFI; X86]
1247 Using this parameter you can use more than 50% of
1248 your efi variable storage. Use this parameter only if
1249 you are really sure that your UEFI does sane gc and
1250 fulfills the spec otherwise your board may brick.
1251
Taku Izumi0f96a992015-09-30 23:01:56 +09001252 efi_fake_mem= nn[KMG]@ss[KMG]:aa[,nn[KMG]@ss[KMG]:aa,..] [EFI; X86]
1253 Add arbitrary attribute to specific memory range by
1254 updating original EFI memory map.
1255 Region of memory which aa attribute is added to is
1256 from ss to ss+nn.
1257 If efi_fake_mem=2G@4G:0x10000,2G@0x10a0000000:0x10000
1258 is specified, EFI_MEMORY_MORE_RELIABLE(0x10000)
1259 attribute is added to range 0x100000000-0x180000000 and
1260 0x10a0000000-0x1120000000.
1261
1262 Using this parameter you can do debugging of EFI memmap
1263 related feature. For example, you can do debugging of
1264 Address Range Mirroring feature even if your box
1265 doesn't support it.
1266
Octavian Purdila475fb4e2016-07-08 19:13:12 +03001267 efivar_ssdt= [EFI; X86] Name of an EFI variable that contains an SSDT
1268 that is to be dynamically loaded by Linux. If there are
1269 multiple variables with the same name but with different
1270 vendor GUIDs, all of them will be loaded. See
1271 Documentation/acpi/ssdt-overlays.txt for details.
1272
1273
Linus Torvalds1da177e2005-04-16 15:20:36 -07001274 eisa_irq_edge= [PARISC,HW]
1275 See header of drivers/parisc/eisa.c.
1276
Alan Coxcd4f0ef2007-07-31 00:37:59 -07001277 elanfreq= [X86-32]
Linus Torvalds1da177e2005-04-16 15:20:36 -07001278 See comment before function elanfreq_setup() in
Pavel Machek71f77052008-07-04 09:59:43 -07001279 arch/x86/kernel/cpu/cpufreq/elanfreq.c.
Linus Torvalds1da177e2005-04-16 15:20:36 -07001280
1281 elevator= [IOSCHED]
Randy Dunlap17a9e7b2010-11-11 12:09:59 +01001282 Format: {"cfq" | "deadline" | "noop"}
Paul Bolle395cf962011-08-15 02:02:26 +02001283 See Documentation/block/cfq-iosched.txt and
Randy Dunlapa9913042005-10-23 12:57:11 -07001284 Documentation/block/deadline-iosched.txt for details.
1285
Michael Holzheud3bf37952011-10-30 15:16:37 +01001286 elfcorehdr=[size[KMG]@]offset[KMG] [IA64,PPC,SH,X86,S390]
Randy Dunlapa9913042005-10-23 12:57:11 -07001287 Specifies physical address of start of kernel core
Michael Holzheud3bf37952011-10-30 15:16:37 +01001288 image elf header and optionally the size. Generally
1289 kexec loader will pass this option to capture kernel.
Vivek Goyalaac04b32006-01-09 20:51:47 -08001290 See Documentation/kdump/kdump.txt for details.
Linus Torvalds1da177e2005-04-16 15:20:36 -07001291
Randy Dunlap0cb55ad2009-04-05 15:55:22 -07001292 enable_mtrr_cleanup [X86]
1293 The kernel tries to adjust MTRR layout from continuous
1294 to discrete, to make X server driver able to add WB
1295 entry later. This parameter enables that.
1296
Randy Dunlapca1eda22009-05-06 16:02:58 -07001297 enable_timer_pin_1 [X86]
Randy Dunlap0cb55ad2009-04-05 15:55:22 -07001298 Enable PIN 1 of APIC timer
1299 Can be useful to work around chipset bugs
1300 (in particular on some ATI chipsets).
1301 The kernel tries to set a reasonable default.
1302
Linus Torvalds1da177e2005-04-16 15:20:36 -07001303 enforcing [SELINUX] Set initial enforcing status.
1304 Format: {"0" | "1"}
1305 See security/selinux/Kconfig help text.
1306 0 -- permissive (log only, no denials).
1307 1 -- enforcing (deny and log).
1308 Default value is 0.
1309 Value can be changed at runtime via /selinux/enforce.
1310
Huang Yinga08f82d2010-05-18 14:35:21 +08001311 erst_disable [ACPI]
1312 Disable Error Record Serialization Table (ERST)
1313 support.
1314
Linus Torvalds1da177e2005-04-16 15:20:36 -07001315 ether= [HW,NET] Ethernet cards parameters
1316 This option is obsoleted by the "netdev=" option, which
1317 has equivalent usage. See its documentation for details.
1318
Mimi Zohar7102ebc2011-05-12 18:33:20 -04001319 evm= [EVM]
1320 Format: { "fix" }
1321 Permit 'security.evm' to be updated regardless of
1322 current integrity status.
1323
Akinobu Mitade1ba092006-12-08 02:39:42 -08001324 failslab=
1325 fail_page_alloc=
1326 fail_make_request=[KNL]
1327 General fault injection mechanism.
1328 Format: <interval>,<probability>,<space>,<times>
Paul Bolle395cf962011-08-15 02:02:26 +02001329 See also Documentation/fault-injection/.
Akinobu Mitade1ba092006-12-08 02:39:42 -08001330
Linus Torvalds1da177e2005-04-16 15:20:36 -07001331 floppy= [HW]
Randy Dunlap31c00fc2008-11-13 21:33:24 +00001332 See Documentation/blockdev/floppy.txt.
Linus Torvalds1da177e2005-04-16 15:20:36 -07001333
Alex Chiangf13ae302008-05-08 14:03:23 -06001334 force_pal_cache_flush
1335 [IA-64] Avoid check_sal_cache_flush which may hang on
1336 buggy SAL_CACHE_FLUSH implementations. Using this
1337 parameter will force ia64_sal_cache_flush to call
1338 ia64_pal_cache_flush instead of SAL_CACHE_FLUSH.
1339
Chris Bainbridge69f23662014-03-07 18:40:42 +07001340 forcepae [X86-32]
1341 Forcefully enable Physical Address Extension (PAE).
1342 Many Pentium M systems disable PAE but may have a
1343 functionally usable PAE implementation.
1344 Warning: use of this parameter will taint the kernel
1345 and may cause unknown problems.
1346
Peter Zijlstrad9e54072008-11-01 19:57:37 +01001347 ftrace=[tracer]
Steven Rostedt2af15d62009-05-28 13:37:24 -04001348 [FTRACE] will set and start the specified tracer
Peter Zijlstrad9e54072008-11-01 19:57:37 +01001349 as early as possible in order to facilitate early
1350 boot debugging.
1351
Frederic Weisbeckercecbca92010-04-18 19:08:41 +02001352 ftrace_dump_on_oops[=orig_cpu]
Steven Rostedt2af15d62009-05-28 13:37:24 -04001353 [FTRACE] will dump the trace buffers on oops.
Frederic Weisbeckercecbca92010-04-18 19:08:41 +02001354 If no parameter is passed, ftrace will dump
1355 buffers of all CPUs, but if you pass orig_cpu, it will
1356 dump only the buffer of the CPU that triggered the
1357 oops.
Steven Rostedt2af15d62009-05-28 13:37:24 -04001358
1359 ftrace_filter=[function-list]
1360 [FTRACE] Limit the functions traced by the function
1361 tracer at boot up. function-list is a comma separated
1362 list of functions. This list can be changed at run
1363 time by the set_ftrace_filter file in the debugfs
Randy Dunlap16290242011-08-13 12:34:52 -07001364 tracing directory.
Steven Rostedt2af15d62009-05-28 13:37:24 -04001365
1366 ftrace_notrace=[function-list]
1367 [FTRACE] Do not trace the functions specified in
1368 function-list. This list can be changed at run time
1369 by the set_ftrace_notrace file in the debugfs
1370 tracing directory.
Peter Zijlstrad9e54072008-11-01 19:57:37 +01001371
Stefan Assmann369bc182009-10-12 22:17:21 +02001372 ftrace_graph_filter=[function-list]
1373 [FTRACE] Limit the top level callers functions traced
1374 by the function graph tracer at boot up.
1375 function-list is a comma separated list of functions
1376 that can be changed at run time by the
1377 set_graph_function file in the debugfs tracing directory.
1378
Namhyung Kim0d7d9a12014-06-13 01:23:50 +09001379 ftrace_graph_notrace=[function-list]
1380 [FTRACE] Do not trace from the functions specified in
1381 function-list. This list is a comma separated list of
1382 functions that can be changed at run time by the
1383 set_graph_notrace file in the debugfs tracing directory.
1384
Linus Torvalds1da177e2005-04-16 15:20:36 -07001385 gamecon.map[2|3]=
1386 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad
1387 support via parallel port (up to 5 devices per port)
1388 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5>
1389 See also Documentation/input/joystick-parport.txt
1390
1391 gamma= [HW,DRM]
1392
Yinghai Luaaf23042008-01-30 13:33:09 +01001393 gart_fix_e820= [X86_64] disable the fix e820 for K8 GART
1394 Format: off | on
1395 default: on
1396
Peter Oberparleiter2521f2c2009-06-17 16:28:08 -07001397 gcov_persist= [GCOV] When non-zero (default), profiling data for
1398 kernel modules is saved and remains accessible via
1399 debugfs, even when the module is unloaded/reloaded.
1400 When zero, profiling data is discarded and associated
1401 debugfs files are removed at module unload time.
1402
Thomas Gleixner2709c2a2017-02-15 11:11:50 +01001403 goldfish [X86] Enable the goldfish android emulator platform.
1404 Don't use this when you are not running on the
1405 android emulator
1406
Linus Torvalds1da177e2005-04-16 15:20:36 -07001407 gpt [EFI] Forces disk with valid GPT signature but
Davidlohr Bueso6c5de792014-01-23 15:56:03 -08001408 invalid Protective MBR to be treated as GPT. If the
1409 primary GPT is corrupted, it enables the backup/alternate
1410 GPT to be used instead.
Linus Torvalds1da177e2005-04-16 15:20:36 -07001411
Andreas Larsson6cec9b02012-11-15 08:47:14 +01001412 grcan.enable0= [HW] Configuration of physical interface 0. Determines
1413 the "Enable 0" bit of the configuration register.
1414 Format: 0 | 1
1415 Default: 0
1416 grcan.enable1= [HW] Configuration of physical interface 1. Determines
1417 the "Enable 0" bit of the configuration register.
1418 Format: 0 | 1
1419 Default: 0
1420 grcan.select= [HW] Select which physical interface to use.
1421 Format: 0 | 1
1422 Default: 0
1423 grcan.txsize= [HW] Sets the size of the tx buffer.
1424 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0.
1425 Default: 1024
1426 grcan.rxsize= [HW] Sets the size of the rx buffer.
1427 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0.
1428 Default: 1024
1429
Bamvor Jian Zhang0f98dd12016-08-31 11:45:46 +02001430 gpio-mockup.gpio_mockup_ranges
1431 [HW] Sets the ranges of gpiochip of for this device.
1432 Format: <start1>,<end1>,<start2>,<end2>...
1433
Jiri Kosina55537872015-11-05 18:44:41 -08001434 hardlockup_all_cpu_backtrace=
1435 [KNL] Should the hard-lockup detector generate
1436 backtraces on all cpus.
1437 Format: <integer>
1438
Linus Torvalds1da177e2005-04-16 15:20:36 -07001439 hashdist= [KNL,NUMA] Large hashes allocated during boot
1440 are distributed across NUMA nodes. Defaults on
Randy Dunlap16290242011-08-13 12:34:52 -07001441 for 64-bit NUMA, off otherwise.
Randy Dunlapa9913042005-10-23 12:57:11 -07001442 Format: 0 | 1 (for off | on)
Linus Torvalds1da177e2005-04-16 15:20:36 -07001443
1444 hcl= [IA-64] SGI's Hardware Graph compatibility layer
1445
1446 hd= [EIDE] (E)IDE hard drive subsystem geometry
1447 Format: <cyl>,<head>,<sect>
1448
Huang Yingea8c0712010-05-18 14:35:15 +08001449 hest_disable [ACPI]
1450 Disable Hardware Error Source Table (HEST) support;
1451 corresponding firmware-first mode error processing
1452 logic will be disabled.
1453
Linus Torvalds1da177e2005-04-16 15:20:36 -07001454 highmem=nn[KMG] [KNL,BOOT] forces the highmem zone to have an exact
1455 size of <nn>. This works even on boxes that have no
1456 highmem otherwise. This also works to reduce highmem
1457 size on bigger boxes.
1458
Thomas Gleixner54cdfdb2007-02-16 01:28:11 -08001459 highres= [KNL] Enable/disable high resolution timer mode.
1460 Valid parameters: "on", "off"
1461 Default: "on"
1462
Linus Torvalds1da177e2005-04-16 15:20:36 -07001463 hisax= [HW,ISDN]
1464 See Documentation/isdn/README.HiSax.
1465
Randy Dunlap0cb55ad2009-04-05 15:55:22 -07001466 hlt [BUGS=ARM,SH]
1467
1468 hpet= [X86-32,HPET] option to control HPET usage
1469 Format: { enable (default) | disable | force |
1470 verbose }
1471 disable: disable HPET and use PIT instead
1472 force: allow force enabled of undocumented chips (ICH4,
1473 VIA, nVidia)
1474 verbose: show contents of HPET registers during setup
1475
Prarit Bhargava3d035f52013-11-12 15:08:33 -08001476 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET
1477 registers. Default set by CONFIG_HPET_MMAP_DEFAULT.
1478
Andi Kleenb4718e62008-07-23 21:27:51 -07001479 hugepages= [HW,X86-32,IA-64] HugeTLB pages to allocate at boot.
1480 hugepagesz= [HW,IA-64,PPC,X86-64] The size of the HugeTLB pages.
Jon Tollefson0d9ea752008-07-23 21:27:56 -07001481 On x86-64 and powerpc, this option can be specified
1482 multiple times interleaved with hugepages= to reserve
1483 huge pages of different sizes. Valid pages sizes on
1484 x86-64 are 2M (when the CPU supports "pse") and 1G
Luiz Capitulino27ec26e2014-12-12 16:55:18 -08001485 (when the CPU supports the "pdpe1gb" cpuinfo flag).
Paul Mundt6902aa82008-09-21 17:14:42 +09001486
Hendrik Brueckner555d61d2009-01-09 12:15:02 +01001487 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC)
1488 terminal devices. Valid values: 0..8
Hendrik Brueckner431429f2009-03-26 15:23:55 +01001489 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs.
1490 If specified, z/VM IUCV HVC accepts connections
1491 from listed z/VM user IDs only.
Hendrik Bruecknercef71252008-12-25 13:39:55 +01001492
James Hoganfdabf522012-10-09 11:00:27 +01001493 hwthread_map= [METAG] Comma-separated list of Linux cpu id to
1494 hardware thread id mappings.
1495 Format: <cpu>:<hwthread>
1496
Fabio M. Di Nitto7bf69392011-03-22 16:34:20 -07001497 keep_bootcon [KNL]
1498 Do not unregister boot console at start. This is only
1499 useful for debugging when something happens in the window
1500 between unregistering the boot console and initializing
1501 the real console.
1502
Jarkko Nikula3a853fb2009-03-23 18:07:47 -07001503 i2c_bus= [HW] Override the default board specific I2C bus speed
Jarkko Nikula79547632009-03-23 18:07:48 -07001504 or register an additional I2C bus that is not
1505 registered from board initialization code.
Jarkko Nikula3a853fb2009-03-23 18:07:47 -07001506 Format:
1507 <bus_id>,<clkrate>
1508
Jiri Kosina36d95732008-10-06 02:51:09 -04001509 i8042.debug [HW] Toggle i8042 debug mode
Stephen Chandler Paule1443d22015-07-15 10:20:17 -07001510 i8042.unmask_kbd_data
1511 [HW] Enable printing of interrupt data from the KBD port
1512 (disabled by default, and as a pre-condition
1513 requires that i8042.debug=1 be enabled)
Linus Torvalds1da177e2005-04-16 15:20:36 -07001514 i8042.direct [HW] Put keyboard port into non-translated mode
Matt LaPlante84eb8d02006-10-03 22:53:09 +02001515 i8042.dumbkbd [HW] Pretend that controller can only read data from
1516 keyboard and cannot control its state
Linus Torvalds1da177e2005-04-16 15:20:36 -07001517 (Don't attempt to blink the leds)
1518 i8042.noaux [HW] Don't check for auxiliary (== mouse) port
Dmitry Torokhov945ef0d2005-09-04 01:42:00 -05001519 i8042.nokbd [HW] Don't check/create keyboard port
Jiri Kosina75d08c72008-03-13 16:13:59 -04001520 i8042.noloop [HW] Disable the AUX Loopback command while probing
1521 for the AUX port
Linus Torvalds1da177e2005-04-16 15:20:36 -07001522 i8042.nomux [HW] Don't check presence of an active multiplexing
Dmitry Torokhove55a3362014-10-31 09:35:53 -07001523 controller
Linus Torvalds1da177e2005-04-16 15:20:36 -07001524 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX
1525 controllers
Masanari Iida24775d62012-02-15 00:26:42 +09001526 i8042.notimeout [HW] Ignore timeout condition signalled by controller
Marcos Paulo de Souza930e1922016-10-01 12:07:35 -07001527 i8042.reset [HW] Reset the controller during init, cleanup and
1528 suspend-to-ram transitions, only during s2r
1529 transitions, or never reset
1530 Format: { 1 | Y | y | 0 | N | n }
1531 1, Y, y: always reset controller
1532 0, N, n: don't ever reset controller
1533 Default: only on s2r transitions on x86; most other
1534 architectures force reset to be always executed
Linus Torvalds1da177e2005-04-16 15:20:36 -07001535 i8042.unlock [HW] Unlock (ignore) the keylock
Srihari Vijayaraghavan148e9a72015-01-07 16:25:53 -08001536 i8042.kbdreset [HW] Reset device connected to KBD port
Linus Torvalds1da177e2005-04-16 15:20:36 -07001537
1538 i810= [HW,DRM]
1539
Dmitry Torokhove70c9d52005-06-25 14:54:25 -07001540 i8k.ignore_dmi [HW] Continue probing hardware even if DMI data
1541 indicates that the driver is running on unsupported
1542 hardware.
Linus Torvalds1da177e2005-04-16 15:20:36 -07001543 i8k.force [HW] Activate i8k driver even if SMM BIOS signature
1544 does not match list of supported models.
1545 i8k.power_status
1546 [HW] Report power status in /proc/i8k
1547 (disabled by default)
1548 i8k.restricted [HW] Allow controlling fans only if SYS_ADMIN
1549 capability is set.
1550
Carsten Emde4dca20e2012-03-15 15:56:26 +01001551 i915.invert_brightness=
Carsten Emde7bd90902012-03-15 15:56:25 +01001552 [DRM] Invert the sense of the variable that is used to
1553 set the brightness of the panel backlight. Normally a
Carsten Emde4dca20e2012-03-15 15:56:26 +01001554 brightness value of 0 indicates backlight switched off,
1555 and the maximum of the brightness value sets the backlight
1556 to maximum brightness. If this parameter is set to 0
1557 (default) and the machine requires it, or this parameter
1558 is set to 1, a brightness value of 0 sets the backlight
1559 to maximum brightness, and the maximum of the brightness
1560 value switches the backlight off.
1561 -1 -- never invert brightness
1562 0 -- machine default
1563 1 -- force brightness inversion
Carsten Emde7bd90902012-03-15 15:56:25 +01001564
Linus Torvalds1da177e2005-04-16 15:20:36 -07001565 icn= [HW,ISDN]
1566 Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]]
1567
David Fries0af80c02009-02-25 20:28:21 +01001568 ide-core.nodma= [HW] (E)IDE subsystem
1569 Format: =0.0 to prevent dma on hda, =0.1 hdb =1.0 hdc
Bartlomiej Zolnierkiewicz075affc2009-06-07 13:52:52 +02001570 .vlb_clock .pci_clock .noflush .nohpa .noprobe .nowerr
1571 .cdrom .chs .ignore_cable are additional options
Randy Dunlap1c10e932008-03-07 21:53:50 +01001572 See Documentation/ide/ide.txt.
Linus Torvalds1da177e2005-04-16 15:20:36 -07001573
Maciej W. Rozycki0f8b7f52014-10-25 17:03:52 +01001574 ide-generic.probe-mask= [HW] (E)IDE subsystem
1575 Format: <int>
1576 Probe mask for legacy ISA IDE ports. Depending on
1577 platform up to 6 ports are supported, enabled by
1578 setting corresponding bits in the mask to 1. The
1579 default value is 0x0, which has a special meaning.
1580 On systems that have PCI, it triggers scanning the
1581 PCI bus for the first and the second port, which
1582 are then probed. On systems without PCI the value
1583 of 0x0 enables probing the two first ports as if it
1584 was 0x3.
1585
Randy Dunlap0cb55ad2009-04-05 15:55:22 -07001586 ide-pci-generic.all-generic-ide [HW] (E)IDE subsystem
1587 Claim all unknown PCI IDE storage controllers.
1588
Andi Kleenf039b752007-05-02 19:27:12 +02001589 idle= [X86]
Len Brown69fb3672013-02-10 01:38:39 -05001590 Format: idle=poll, idle=halt, idle=nomwait
Randy Dunlapada9cfd2008-12-19 10:57:32 -08001591 Poll forces a polling idle loop that can slightly
1592 improve the performance of waking up a idle CPU, but
1593 will use a lot of power and make the system run hot.
1594 Not recommended.
Randy Dunlapada9cfd2008-12-19 10:57:32 -08001595 idle=halt: Halt is forced to be used for CPU idle.
Zhao Yakuic1e3b372008-06-24 17:58:53 +08001596 In such case C2/C3 won't be used again.
Randy Dunlapada9cfd2008-12-19 10:57:32 -08001597 idle=nomwait: Disable mwait for CPU C-states
Randy Dunlapa9913042005-10-23 12:57:11 -07001598
Maciej W. Rozycki503943e2015-11-13 00:48:29 +00001599 ieee754= [MIPS] Select IEEE Std 754 conformance mode
1600 Format: { strict | legacy | 2008 | relaxed }
1601 Default: strict
1602
1603 Choose which programs will be accepted for execution
1604 based on the IEEE 754 NaN encoding(s) supported by
1605 the FPU and the NaN encoding requested with the value
1606 of an ELF file header flag individually set by each
1607 binary. Hardware implementations are permitted to
1608 support either or both of the legacy and the 2008 NaN
1609 encoding mode.
1610
1611 Available settings are as follows:
1612 strict accept binaries that request a NaN encoding
1613 supported by the FPU
1614 legacy only accept legacy-NaN binaries, if supported
1615 by the FPU
1616 2008 only accept 2008-NaN binaries, if supported
1617 by the FPU
1618 relaxed accept any binaries regardless of whether
1619 supported by the FPU
1620
1621 The FPU emulator is always able to support both NaN
1622 encodings, so if no FPU hardware is present or it has
1623 been disabled with 'nofpu', then the settings of
1624 'legacy' and '2008' strap the emulator accordingly,
1625 'relaxed' straps the emulator for both legacy-NaN and
1626 2008-NaN, whereas 'strict' enables legacy-NaN only on
1627 legacy processors and both NaN encodings on MIPS32 or
1628 MIPS64 CPUs.
1629
1630 The setting for ABS.fmt/NEG.fmt instruction execution
1631 mode generally follows that for the NaN encoding,
1632 except where unsupported by hardware.
1633
Ingo Molnar792908222006-12-06 20:40:51 -08001634 ignore_loglevel [KNL]
1635 Ignore loglevel setting - this will print /all/
1636 kernel messages to the console. Useful for debugging.
Yanmin Zhang0eca6b72011-10-31 17:11:25 -07001637 We also add it as printk module parameter, so users
1638 could change it dynamically, usually by
1639 /sys/module/printk/parameters/ignore_loglevel.
Ingo Molnar792908222006-12-06 20:40:51 -08001640
Konstantin Khlebnikovd977d562016-02-02 16:57:43 -08001641 ignore_rlimit_data
1642 Ignore RLIMIT_DATA setting for data mappings,
1643 print warning at first misuse. Can be changed via
1644 /sys/module/kernel/parameters/ignore_rlimit_data.
1645
Linus Torvalds1da177e2005-04-16 15:20:36 -07001646 ihash_entries= [KNL]
1647 Set number of hash buckets for inode cache.
1648
Mimi Zohar2fe5d6d2012-02-13 10:15:05 -05001649 ima_appraise= [IMA] appraise integrity measurements
Dmitry Kasatkin2faa6ef2014-05-08 13:11:29 +03001650 Format: { "off" | "enforce" | "fix" | "log" }
Mimi Zohar2fe5d6d2012-02-13 10:15:05 -05001651 default: "enforce"
1652
Mimi Zohar07f6a792011-03-09 22:25:48 -05001653 ima_appraise_tcb [IMA]
1654 The builtin appraise policy appraises all files
1655 owned by uid=0.
1656
Mimi Zohar3323eec92009-02-04 09:06:58 -05001657 ima_hash= [IMA]
Mimi Zohare7a2ad72013-06-07 12:16:37 +02001658 Format: { md5 | sha1 | rmd160 | sha256 | sha384
1659 | sha512 | ... }
Mimi Zohar3323eec92009-02-04 09:06:58 -05001660 default: "sha1"
1661
Mimi Zohare7a2ad72013-06-07 12:16:37 +02001662 The list of supported hash algorithms is defined
1663 in crypto/hash_info.h.
1664
Mimi Zohar24fd03c2015-06-11 20:48:33 -04001665 ima_policy= [IMA]
1666 The builtin measurement policy to load during IMA
1667 setup. Specyfing "tcb" as the value, measures all
1668 programs exec'd, files mmap'd for exec, and all files
1669 opened with the read mode bit set by either the
1670 effective uid (euid=0) or uid=0.
1671 Format: "tcb"
1672
1673 ima_tcb [IMA] Deprecated. Use ima_policy= instead.
Eric Paris5789ba32009-05-21 15:47:06 -04001674 Load a policy which meets the needs of the Trusted
1675 Computing Base. This means IMA will measure all
1676 programs exec'd, files mmap'd for exec, and all files
1677 opened for read by uid=0.
1678
Roberto Sassu9b9d4ce2013-06-07 12:16:35 +02001679 ima_template= [IMA]
1680 Select one of defined IMA measurements template formats.
Roberto Sassu8265a2f2015-04-11 17:07:03 +02001681 Formats: { "ima" | "ima-ng" | "ima-sig" }
Roberto Sassu9b9d4ce2013-06-07 12:16:35 +02001682 Default: "ima-ng"
1683
Roberto Sassuc2426d22014-10-13 14:08:42 +02001684 ima_template_fmt=
1685 [IMA] Define a custom template format.
1686 Format: { "field1|...|fieldN" }
1687
Dmitry Kasatkin3bcced32014-02-26 17:05:20 +02001688 ima.ahash_minsize= [IMA] Minimum file size for asynchronous hash usage
1689 Format: <min_file_size>
1690 Set the minimal file size for using asynchronous hash.
1691 If left unspecified, ahash usage is disabled.
1692
1693 ahash performance varies for different data sizes on
1694 different crypto accelerators. This option can be used
1695 to achieve the best performance for a particular HW.
1696
Dmitry Kasatkin6edf7a82014-05-06 14:47:13 +03001697 ima.ahash_bufsize= [IMA] Asynchronous hash buffer size
1698 Format: <bufsize>
1699 Set hashing buffer size. Default: 4k.
1700
1701 ahash performance varies for different chunk sizes on
1702 different crypto accelerators. This option can be used
1703 to achieve best performance for particular HW.
1704
Linus Torvalds1da177e2005-04-16 15:20:36 -07001705 init= [KNL]
1706 Format: <full_path>
1707 Run specified binary instead of /sbin/init as init
1708 process.
1709
1710 initcall_debug [KNL] Trace initcalls as they are executed. Useful
1711 for working out where the kernel is dying during
1712 startup.
1713
Prarit Bhargava7b0b73d2014-06-04 16:12:17 -07001714 initcall_blacklist= [KNL] Do not execute a comma-separated list of
1715 initcall functions. Useful for debugging built-in
1716 modules and initcalls.
1717
Linus Torvalds1da177e2005-04-16 15:20:36 -07001718 initrd= [BOOT] Specify the location of the initial ramdisk
1719
Dave Hansenacd547b2016-07-29 09:30:21 -07001720 init_pkru= [x86] Specify the default memory protection keys rights
1721 register contents for all processes. 0x55555554 by
1722 default (disallow access to all but pkey 0). Can
1723 override in debugfs after boot.
1724
Linus Torvalds1da177e2005-04-16 15:20:36 -07001725 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver
1726 Format: <irq>
1727
Fenghua Yu6bb2ff82013-05-21 15:35:17 -04001728 int_pln_enable [x86] Enable power limit notification interrupt
1729
Mimi Zohard726d8d2013-03-18 14:48:02 -04001730 integrity_audit=[IMA]
1731 Format: { "0" | "1" }
1732 0 -- basic integrity auditing messages. (Default)
1733 1 -- additional integrity auditing messages.
1734
Keshavamurthy, Anil Sba395922007-10-21 16:41:49 -07001735 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option
Kyle McMartin0cd5c3c2009-02-04 14:29:19 -08001736 on
1737 Enable intel iommu driver.
Keshavamurthy, Anil Sba395922007-10-21 16:41:49 -07001738 off
1739 Disable intel iommu driver.
1740 igfx_off [Default Off]
1741 By default, gfx is mapped as normal device. If a gfx
1742 device has a dedicated DMAR unit, the DMAR unit is
1743 bypassed by not enabling DMAR with this option. In
1744 this case, gfx device will use physical address for
1745 DMA.
Keshavamurthy, Anil S7d3b03c2007-10-21 16:41:53 -07001746 forcedac [x86_64]
1747 With this option iommu will not optimize to look
Randy Dunlap16290242011-08-13 12:34:52 -07001748 for io virtual address below 32-bit forcing dual
Keshavamurthy, Anil S7d3b03c2007-10-21 16:41:53 -07001749 address cycle on pci bus for cards supporting greater
Randy Dunlap16290242011-08-13 12:34:52 -07001750 than 32-bit addressing. The default is to look
1751 for translation below 32-bit and if not available
Keshavamurthy, Anil S7d3b03c2007-10-21 16:41:53 -07001752 then look in the higher range.
mark gross5e0d2a62008-03-04 15:22:08 -08001753 strict [Default Off]
1754 With this option on every unmap_single operation will
1755 result in a hardware IOTLB flush operation as opposed
1756 to batching them for performance.
Youquan Song6dd9a7c2011-05-25 19:13:49 +01001757 sp_off [Default Off]
1758 By default, super page will be supported if Intel IOMMU
1759 has the capability. With this option, super page will
1760 not be supported.
David Woodhousec83b2f22015-06-12 10:15:49 +01001761 ecs_off [Default Off]
1762 By default, extended context tables will be supported if
1763 the hardware advertises that it has support both for the
1764 extended tables themselves, and also PASID support. With
1765 this option set, extended tables will not be used even
1766 on hardware which claims to support them.
Masanari Iida2e92c7a2011-12-15 01:18:52 +09001767
1768 intel_idle.max_cstate= [KNL,HW,ACPI,X86]
1769 0 disables intel_idle and fall back on acpi_idle.
baolex.ni22c6bbe2016-07-11 09:57:37 +08001770 1 to 9 specify maximum depth of C-state.
Masanari Iida2e92c7a2011-12-15 01:18:52 +09001771
Dirk Brandewie6be26492013-02-15 22:55:10 +01001772 intel_pstate= [X86]
1773 disable
1774 Do not enable intel_pstate as the default
1775 scaling driver for the supported processors
Ethan Zhaoaa4ea342014-12-09 10:43:19 +09001776 force
1777 Enable intel_pstate on systems that prohibit it by default
1778 in favor of acpi-cpufreq. Forcing the intel_pstate driver
1779 instead of acpi-cpufreq may disable platform features, such
1780 as thermal controls and power capping, that rely on ACPI
1781 P-States information being indicated to OSPM and therefore
1782 should be used with caution. This option does not work with
1783 processors that aren't supported by the intel_pstate driver
1784 or on platforms that use pcc-cpufreq instead of acpi-cpufreq.
Dirk Brandewie2f86dc42014-11-06 09:40:47 -08001785 no_hwp
1786 Do not enable hardware P state control (HWP)
1787 if available.
Kristen Carlson Accardid64c3b02015-02-06 13:41:55 -08001788 hwp_only
1789 Only load intel_pstate on systems which support
1790 hardware P state control (HWP) if available.
Srinivas Pandruvada9522a2f2016-04-27 15:48:06 -07001791 support_acpi_ppc
Srinivas Pandruvada2b3ec762016-04-27 15:48:08 -07001792 Enforce ACPI _PPC performance limits. If the Fixed ACPI
1793 Description Table, specifies preferred power management
1794 profile as "Enterprise Server" or "Performance Server",
1795 then this feature is turned on by default.
Dirk Brandewie6be26492013-02-15 22:55:10 +01001796
Chris Wrightd1423d52010-07-20 11:06:49 -07001797 intremap= [X86-64, Intel-IOMMU]
Chris Wrightd1423d52010-07-20 11:06:49 -07001798 on enable Interrupt Remapping (default)
1799 off disable Interrupt Remapping
1800 nosid disable Source ID checking
Suresh Siddha41750d32011-08-23 17:05:18 -07001801 no_x2apic_optout
1802 BIOS x2APIC opt-out request will be ignored
Feng Wub7d20632015-09-18 22:29:56 +08001803 nopost disable Interrupt Posting
Chris Wrightd1423d52010-07-20 11:06:49 -07001804
Randy Dunlap0cb55ad2009-04-05 15:55:22 -07001805 iomem= Disable strict checking of access to MMIO memory
1806 strict regions from userspace.
1807 relaxed
1808
1809 iommu= [x86]
1810 off
1811 force
1812 noforce
1813 biomerge
1814 panic
1815 nopanic
1816 merge
1817 nomerge
1818 forcesac
1819 soft
Alex Williamsonbcb71ab2011-10-21 15:56:24 -04001820 pt [x86, IA-64]
Thadeu Lima de Souza Cascardo4e287842014-10-23 19:19:35 -02001821 nobypass [PPC/POWERNV]
1822 Disable IOMMU bypass, using IOMMU for PCI devices.
Alex Williamsonbcb71ab2011-10-21 15:56:24 -04001823
Randy Dunlap0cb55ad2009-04-05 15:55:22 -07001824
1825 io7= [HW] IO7 for Marvel based alpha systems
1826 See comment before marvel_specify_io7 in
1827 arch/alpha/kernel/core_marvel.c.
1828
Jaswinder Singh Rajput6cececf2009-04-14 14:03:43 +05301829 io_delay= [X86] I/O delay method
Ingo Molnar6e7c4022008-01-30 13:30:05 +01001830 0x80
1831 Standard port 0x80 based delay
1832 0xed
1833 Alternate port 0xed based delay (needed on some systems)
Rene Hermanb02aae92008-01-30 13:30:05 +01001834 udelay
Ingo Molnar6e7c4022008-01-30 13:30:05 +01001835 Simple two microseconds delay
1836 none
1837 No delay
Rene Hermanb02aae92008-01-30 13:30:05 +01001838
Linus Torvalds1da177e2005-04-16 15:20:36 -07001839 ip= [IP_PNP]
J. Bruce Fieldsdc7a0812009-10-27 14:41:35 -04001840 See Documentation/filesystems/nfs/nfsroot.txt.
Linus Torvalds1da177e2005-04-16 15:20:36 -07001841
Thomas Gleixnerfbf19802016-02-03 19:52:23 +01001842 irqaffinity= [SMP] Set the default irq affinity mask
Noam Camus2d13e6c2016-10-11 13:51:35 -07001843 The argument is a cpu list, as described above.
Thomas Gleixnerfbf19802016-02-03 19:52:23 +01001844
Alan Cox200803d2005-06-28 20:45:18 -07001845 irqfixup [HW]
1846 When an interrupt is not handled search all handlers
1847 for it. Intended to get systems with badly broken
1848 firmware running.
1849
1850 irqpoll [HW]
1851 When an interrupt is not handled search all handlers
1852 for it. Also check all handlers each timer
1853 interrupt. Intended to get systems with badly broken
1854 firmware running.
1855
Linus Torvalds1da177e2005-04-16 15:20:36 -07001856 isapnp= [ISAPNP]
Randy Dunlapa9913042005-10-23 12:57:11 -07001857 Format: <RDP>,<reset>,<pci_scan>,<verbosity>
Linus Torvalds1da177e2005-04-16 15:20:36 -07001858
1859 isolcpus= [KNL,SMP] Isolate CPUs from the general scheduler.
Noam Camus2d13e6c2016-10-11 13:51:35 -07001860 The argument is a cpu list, as described above.
Li Zefanb225d442008-11-06 12:53:39 -08001861
Linus Torvalds1da177e2005-04-16 15:20:36 -07001862 This option can be used to specify one or more CPUs
1863 to isolate from the general SMP balancing and scheduling
Li Zefanb225d442008-11-06 12:53:39 -08001864 algorithms. You can move a process onto or off an
1865 "isolated" CPU via the CPU affinity syscalls or cpuset.
Linus Torvalds1da177e2005-04-16 15:20:36 -07001866 <cpu number> begins at 0 and the maximum value is
1867 "number of CPUs in system - 1".
1868
1869 This option is the preferred way to isolate CPUs. The
Randy Dunlapa9913042005-10-23 12:57:11 -07001870 alternative -- manually setting the CPU mask of all
1871 tasks in the system -- can cause problems and
1872 suboptimal load balancer performance.
Linus Torvalds1da177e2005-04-16 15:20:36 -07001873
Randy Dunlapa9913042005-10-23 12:57:11 -07001874 iucv= [HW,NET]
Linus Torvalds1da177e2005-04-16 15:20:36 -07001875
Joerg Roedel7d8bfa22013-04-09 21:27:19 +02001876 ivrs_ioapic [HW,X86_64]
1877 Provide an override to the IOAPIC-ID<->DEVICE-ID
1878 mapping provided in the IVRS ACPI table. For
1879 example, to map IOAPIC-ID decimal 10 to
1880 PCI device 00:14.0 write the parameter as:
1881 ivrs_ioapic[10]=00:14.0
1882
1883 ivrs_hpet [HW,X86_64]
1884 Provide an override to the HPET-ID<->DEVICE-ID
1885 mapping provided in the IVRS ACPI table. For
1886 example, to map HPET-ID decimal 0 to
1887 PCI device 00:14.0 write the parameter as:
1888 ivrs_hpet[0]=00:14.0
1889
Suravee Suthikulpanitca3bf5d2016-04-01 09:06:01 -04001890 ivrs_acpihid [HW,X86_64]
1891 Provide an override to the ACPI-HID:UID<->DEVICE-ID
1892 mapping provided in the IVRS ACPI table. For
1893 example, to map UART-HID:UID AMD0020:0 to
1894 PCI device 00:14.5 write the parameter as:
1895 ivrs_acpihid[00:14.5]=AMD0020:0
1896
Linus Torvalds1da177e2005-04-16 15:20:36 -07001897 js= [HW,JOY] Analog joystick
1898 See Documentation/input/joystick.txt.
1899
Kees Cook65fe9352016-06-13 15:10:02 -07001900 nokaslr [KNL]
1901 When CONFIG_RANDOMIZE_BASE is set, this disables
1902 kernel and module base offset ASLR (Address Space
1903 Layout Randomization).
Kees Cook24f2e022014-06-13 13:30:36 -07001904
Mark Rutland586b2bd2017-03-31 15:12:04 -07001905 kasan_multi_shot
1906 [KNL] Enforce KASAN (Kernel Address Sanitizer) to print
1907 report on every invalid memory access. Without this
1908 parameter KASAN will print report only for the first
1909 invalid access.
1910
Randy Dunlap0cb55ad2009-04-05 15:55:22 -07001911 keepinitrd [HW,ARM]
1912
Taku Izumi342332e2016-03-15 14:55:22 -07001913 kernelcore= [KNL,X86,IA-64,PPC]
1914 Format: nn[KMGTPE] | "mirror"
1915 This parameter
Mel Gormaned7ed362007-07-17 04:03:14 -07001916 specifies the amount of memory usable by the kernel
1917 for non-movable allocations. The requested amount is
1918 spread evenly throughout all nodes in the system. The
1919 remaining memory in each node is used for Movable
1920 pages. In the event, a node is too small to have both
1921 kernelcore and Movable pages, kernelcore pages will
1922 take priority and other nodes will have a larger number
Weiping Pan675217f2013-09-30 13:45:10 -07001923 of Movable pages. The Movable zone is used for the
Mel Gormaned7ed362007-07-17 04:03:14 -07001924 allocation of pages that may be reclaimed or moved
1925 by the page migration subsystem. This means that
1926 HugeTLB pages may not be allocated from this zone.
1927 Note that allocations like PTEs-from-HighMem still
1928 use the HighMem zone if it exists, and the Normal
1929 zone if it does not.
1930
Taku Izumi342332e2016-03-15 14:55:22 -07001931 Instead of specifying the amount of memory (nn[KMGTPE]),
1932 you can specify "mirror" option. In case "mirror"
1933 option is specified, mirrored (reliable) memory is used
1934 for non-movable allocations and remaining memory is used
1935 for Movable pages. nn[KMGTPE] and "mirror" are exclusive,
1936 so you can NOT specify nn[KMGTPE] and "mirror" at the same
1937 time.
1938
Jason Wessel4fe1da42010-05-20 21:04:31 -05001939 kgdbdbgp= [KGDB,HW] kgdb over EHCI usb debug port.
1940 Format: <Controller#>[,poll interval]
1941 The controller # is the number of the ehci usb debug
1942 port as it is probed via PCI. The poll interval is
1943 optional and is the number seconds in between
1944 each poll cycle to the debug port in case you need
1945 the functionality for interrupting the kernel with
1946 gdb or control-c on the dbgp connection. When
1947 not using this parameter you use sysrq-g to break into
1948 the kernel debugger.
1949
Jason Wessel84c08fd2010-05-20 21:04:24 -05001950 kgdboc= [KGDB,HW] kgdb over consoles.
Jason Wesselada64e42010-05-20 21:04:24 -05001951 Requires a tty driver that supports console polling,
1952 or a supported polling keyboard driver (non-usb).
Jason Wessel65b5ac12010-08-05 09:22:33 -05001953 Serial only format: <serial_device>[,baud]
1954 keyboard only format: kbd
1955 keyboard and serial format: kbd,<serial_device>[,baud]
1956 Optional Kernel mode setting:
1957 kms, kbd format: kms,kbd
1958 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud]
Jason Wessel6cdf6e02008-04-17 20:05:38 +02001959
Jason Wessel84c08fd2010-05-20 21:04:24 -05001960 kgdbwait [KGDB] Stop kernel execution and enter the
1961 kernel debugger at the earliest opportunity.
1962
Florian Fainelli9bed90c62008-08-23 18:54:37 +02001963 kmac= [MIPS] korina ethernet MAC address.
1964 Configure the RouterBoard 532 series on-chip
1965 Ethernet adapter MAC address.
1966
Catalin Marinas04f70332009-06-11 13:22:39 +01001967 kmemleak= [KNL] Boot-time kmemleak enable/disable
1968 Valid arguments: on, off
1969 Default: on
Masanari Iida47aeedd2014-10-24 21:24:59 +09001970 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y,
1971 the default is off.
Catalin Marinas04f70332009-06-11 13:22:39 +01001972
Xishi Qiuc3ac14b2014-01-23 15:53:33 -08001973 kmemcheck= [X86] Boot-time kmemcheck enable/disable/one-shot mode
1974 Valid arguments: 0, 1, 2
1975 kmemcheck=0 (disabled)
1976 kmemcheck=1 (enabled)
1977 kmemcheck=2 (one-shot mode)
1978 Default: 2 (one-shot mode)
1979
Jaswinder Singh Rajput6cececf2009-04-14 14:03:43 +05301980 kstack=N [X86] Print N words from the kernel stack
Randy Dunlap0cb55ad2009-04-05 15:55:22 -07001981 in oops dumps.
1982
Andre Przywarafef07aa2009-07-10 14:20:35 +02001983 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs.
1984 Default is 0 (don't ignore, but inject #GP)
1985
Xiao Guangronga182d872010-09-20 22:17:48 +08001986 kvm.mmu_audit= [KVM] This is a R/W parameter which allows audit
1987 KVM MMU at runtime.
Andre Przywarafef07aa2009-07-10 14:20:35 +02001988 Default is 0 (off)
1989
Paolo Bonzinia7ad7942019-11-04 12:22:02 +01001990 kvm.nx_huge_pages=
1991 [KVM] Controls the software workaround for the
1992 X86_BUG_ITLB_MULTIHIT bug.
1993 force : Always deploy workaround.
1994 off : Never deploy workaround.
1995 auto : Deploy workaround based on the presence of
1996 X86_BUG_ITLB_MULTIHIT.
1997
1998 Default is 'auto'.
1999
2000 If the software workaround is enabled for the host,
2001 guests do need not to enable it for nested guests.
2002
Junaid Shahidc6e94ac2019-11-04 12:22:03 +01002003 kvm.nx_huge_pages_recovery_ratio=
2004 [KVM] Controls how many 4KiB pages are periodically zapped
2005 back to huge pages. 0 disables the recovery, otherwise if
2006 the value is N KVM will zap 1/Nth of the 4KiB pages every
2007 minute. The default is 60.
2008
Andre Przywarafef07aa2009-07-10 14:20:35 +02002009 kvm-amd.nested= [KVM,AMD] Allow nested virtualization in KVM/SVM.
Xiao Guangrong8475f942010-09-20 22:16:45 +08002010 Default is 1 (enabled)
Andre Przywarafef07aa2009-07-10 14:20:35 +02002011
2012 kvm-amd.npt= [KVM,AMD] Disable nested paging (virtualized MMU)
2013 for all guests.
Randy Dunlap16290242011-08-13 12:34:52 -07002014 Default is 1 (enabled) if in 64-bit or 32-bit PAE mode.
Andre Przywarafef07aa2009-07-10 14:20:35 +02002015
Andre Przywarafef07aa2009-07-10 14:20:35 +02002016 kvm-intel.ept= [KVM,Intel] Disable extended page tables
2017 (virtualized MMU) support on capable Intel chips.
2018 Default is 1 (enabled)
2019
2020 kvm-intel.emulate_invalid_guest_state=
2021 [KVM,Intel] Enable emulation of invalid guest states
2022 Default is 0 (disabled)
2023
2024 kvm-intel.flexpriority=
2025 [KVM,Intel] Disable FlexPriority feature (TPR shadow).
2026 Default is 1 (enabled)
2027
Sasha Levine1a72ae2011-08-09 14:28:35 +03002028 kvm-intel.nested=
2029 [KVM,Intel] Enable VMX nesting (nVMX).
2030 Default is 0 (disabled)
2031
Andre Przywarafef07aa2009-07-10 14:20:35 +02002032 kvm-intel.unrestricted_guest=
2033 [KVM,Intel] Disable unrestricted guest feature
2034 (virtualized real and unpaged mode) on capable
2035 Intel chips. Default is 1 (enabled)
2036
Konrad Rzeszutek Wilkaf6ce922018-07-02 12:29:30 +02002037 kvm-intel.vmentry_l1d_flush=[KVM,Intel] Mitigation for L1 Terminal Fault
2038 CVE-2018-3620.
2039
2040 Valid arguments: never, cond, always
2041
2042 always: L1D cache flush on every VMENTER.
2043 cond: Flush L1D on VMENTER only when the code between
2044 VMEXIT and VMENTER can leak host memory.
2045 never: Disables the mitigation
2046
2047 Default is cond (do L1 cache flush in specific instances)
2048
Andre Przywarafef07aa2009-07-10 14:20:35 +02002049 kvm-intel.vpid= [KVM,Intel] Disable Virtual Processor Identification
2050 feature (tagged TLBs) on capable Intel chips.
2051 Default is 1 (enabled)
2052
Jiri Kosina2decbf52018-07-13 16:23:25 +02002053 l1tf= [X86] Control mitigation of the L1TF vulnerability on
2054 affected CPUs
2055
2056 The kernel PTE inversion protection is unconditionally
2057 enabled and cannot be disabled.
2058
2059 full
2060 Provides all available mitigations for the
2061 L1TF vulnerability. Disables SMT and
2062 enables all mitigations in the
2063 hypervisors, i.e. unconditional L1D flush.
2064
2065 SMT control and L1D flush control via the
2066 sysfs interface is still possible after
2067 boot. Hypervisors will issue a warning
2068 when the first VM is started in a
2069 potentially insecure configuration,
2070 i.e. SMT enabled or L1D flush disabled.
2071
2072 full,force
2073 Same as 'full', but disables SMT and L1D
2074 flush runtime control. Implies the
2075 'nosmt=force' command line option.
2076 (i.e. sysfs control of SMT is disabled.)
2077
2078 flush
2079 Leaves SMT enabled and enables the default
2080 hypervisor mitigation, i.e. conditional
2081 L1D flush.
2082
2083 SMT control and L1D flush control via the
2084 sysfs interface is still possible after
2085 boot. Hypervisors will issue a warning
2086 when the first VM is started in a
2087 potentially insecure configuration,
2088 i.e. SMT enabled or L1D flush disabled.
2089
2090 flush,nosmt
2091
2092 Disables SMT and enables the default
2093 hypervisor mitigation.
2094
2095 SMT control and L1D flush control via the
2096 sysfs interface is still possible after
2097 boot. Hypervisors will issue a warning
2098 when the first VM is started in a
2099 potentially insecure configuration,
2100 i.e. SMT enabled or L1D flush disabled.
2101
2102 flush,nowarn
2103 Same as 'flush', but hypervisors will not
2104 warn when a VM is started in a potentially
2105 insecure configuration.
2106
2107 off
2108 Disables hypervisor mitigations and doesn't
2109 emit any warnings.
Michal Hockoc3692582018-11-13 19:49:10 +01002110 It also drops the swap size and available
2111 RAM limit restriction on both hypervisor and
2112 bare metal.
Jiri Kosina2decbf52018-07-13 16:23:25 +02002113
2114 Default is 'flush'.
2115
Thomas Gleixnercb106032019-02-19 11:10:49 +01002116 For details see: Documentation/hw-vuln/l1tf.rst
Jiri Kosina2decbf52018-07-13 16:23:25 +02002117
Linus Torvalds1da177e2005-04-16 15:20:36 -07002118 l2cr= [PPC]
2119
Robert Brosea78bfbf2008-03-29 07:20:23 +11002120 l3cr= [PPC]
2121
Alan Coxcd4f0ef2007-07-31 00:37:59 -07002122 lapic [X86-32,APIC] Enable the local APIC even if BIOS
Randy Dunlapa9913042005-10-23 12:57:11 -07002123 disabled it.
Linus Torvalds1da177e2005-04-16 15:20:36 -07002124
Suresh Siddha279f1462012-10-22 14:37:58 -07002125 lapic= [x86,APIC] "notscdeadline" Do not use TSC deadline
2126 value for LAPIC timer one-shot implementation. Default
2127 back to the programmable timer unit in the LAPIC.
2128
Jaswinder Singh Rajput6cececf2009-04-14 14:03:43 +05302129 lapic_timer_c2_ok [X86,APIC] trust the local apic timer
Randy Dunlapada9cfd2008-12-19 10:57:32 -08002130 in C2 power state.
Thomas Gleixnere585bef2007-03-23 16:08:01 +01002131
FD Camifcb71f62008-01-06 19:08:56 +01002132 libata.dma= [LIBATA] DMA control
2133 libata.dma=0 Disable all PATA and SATA DMA
2134 libata.dma=1 PATA and SATA Disk DMA only
2135 libata.dma=2 ATAPI (CDROM) DMA only
Randy Dunlap16290242011-08-13 12:34:52 -07002136 libata.dma=4 Compact Flash DMA only
FD Camifcb71f62008-01-06 19:08:56 +01002137 Combinations also work, so libata.dma=3 enables DMA
2138 for disks and CDROMs, but not CFs.
Randy Dunlap16290242011-08-13 12:34:52 -07002139
Michael Prokop20308872009-08-06 00:14:10 +02002140 libata.ignore_hpa= [LIBATA] Ignore HPA limit
2141 libata.ignore_hpa=0 keep BIOS limits (default)
2142 libata.ignore_hpa=1 ignore limits, using full disk
FD Camifcb71f62008-01-06 19:08:56 +01002143
Dave Jones78e70c22007-09-27 11:50:13 -04002144 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume
2145 when set.
2146 Format: <int>
2147
Tejun Heo33267322008-02-13 09:15:09 +09002148 libata.force= [LIBATA] Force configurations. The format is comma
2149 separated list of "[ID:]VAL" where ID is
Roman Fietze4c44f302010-04-21 12:17:12 +02002150 PORT[.DEVICE]. PORT and DEVICE are decimal numbers
Tejun Heo33267322008-02-13 09:15:09 +09002151 matching port, link or device. Basically, it matches
2152 the ATA ID string printed on console by libata. If
2153 the whole ID part is omitted, the last PORT and DEVICE
2154 values are used. If ID hasn't been specified yet, the
2155 configuration applies to all ports, links and devices.
2156
2157 If only DEVICE is omitted, the parameter applies to
2158 the port and all links and devices behind it. DEVICE
2159 number of 0 either selects the first device or the
2160 first fan-out link behind PMP device. It does not
2161 select the host link. DEVICE number of 15 selects the
2162 host link and device attached to it.
2163
2164 The VAL specifies the configuration to force. As long
2165 as there's no ambiguity shortcut notation is allowed.
2166 For example, both 1.5 and 1.5G would work for 1.5Gbps.
2167 The following configurations can be forced.
2168
2169 * Cable type: 40c, 80c, short40c, unk, ign or sata.
2170 Any ID with matching PORT is used.
2171
2172 * SATA link speed limit: 1.5Gbps or 3.0Gbps.
2173
2174 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7].
2175 udma[/][16,25,33,44,66,100,133] notation is also
2176 allowed.
2177
2178 * [no]ncq: Turn on or off NCQ.
2179
Martin K. Petersend7b16e42015-05-04 21:54:18 -04002180 * [no]ncqtrim: Turn off queued DSM TRIM.
2181
Tejun Heo05944bd2008-08-13 20:19:09 +09002182 * nohrst, nosrst, norst: suppress hard, soft
2183 and both resets.
2184
Dan Williamsca6d43b2012-06-21 23:41:41 -07002185 * rstonce: only attempt one reset during
2186 hot-unplug link recovery
2187
Tejun Heo43c9c592010-05-23 12:59:11 +02002188 * dump_id: dump IDENTIFY data.
2189
Vincent Pelletier966fbe12013-05-21 22:30:58 +02002190 * atapi_dmadir: Enable ATAPI DMADIR bridge support
2191
Robin H. Johnsonb8bd6dc2013-12-16 09:31:19 -08002192 * disable: Disable this device.
2193
Tejun Heo33267322008-02-13 09:15:09 +09002194 If there are multiple matching configurations changing
2195 the same attribute, the last one is used.
2196
Yinghai Lu95f72d12010-07-12 14:36:09 +10002197 memblock=debug [KNL] Enable memblock debug messages.
Randy Dunlap7c4be2532009-01-06 14:42:44 -08002198
Linus Torvalds1da177e2005-04-16 15:20:36 -07002199 load_ramdisk= [RAM] List of ramdisks to load from floppy
Randy Dunlap31c00fc2008-11-13 21:33:24 +00002200 See Documentation/blockdev/ramdisk.txt.
Linus Torvalds1da177e2005-04-16 15:20:36 -07002201
Randy Dunlapa6b25b62006-01-14 13:21:19 -08002202 lockd.nlm_grace_period=P [NFS] Assign grace period.
2203 Format: <integer>
Linus Torvalds1da177e2005-04-16 15:20:36 -07002204
Randy Dunlapa6b25b62006-01-14 13:21:19 -08002205 lockd.nlm_tcpport=N [NFS] Assign TCP port.
2206 Format: <integer>
2207
2208 lockd.nlm_timeout=T [NFS] Assign timeout value.
2209 Format: <integer>
2210
2211 lockd.nlm_udpport=M [NFS] Assign UDP port.
2212 Format: <integer>
Linus Torvalds1da177e2005-04-16 15:20:36 -07002213
Paul E. McKenneyec4518a2014-09-12 10:50:01 -07002214 locktorture.nreaders_stress= [KNL]
2215 Set the number of locking read-acquisition kthreads.
2216 Defaults to being automatically set based on the
2217 number of online CPUs.
2218
2219 locktorture.nwriters_stress= [KNL]
2220 Set the number of locking write-acquisition kthreads.
2221
2222 locktorture.onoff_holdoff= [KNL]
2223 Set time (s) after boot for CPU-hotplug testing.
2224
2225 locktorture.onoff_interval= [KNL]
2226 Set time (s) between CPU-hotplug operations, or
2227 zero to disable CPU-hotplug testing.
2228
2229 locktorture.shuffle_interval= [KNL]
2230 Set task-shuffle interval (jiffies). Shuffling
2231 tasks allows some CPUs to go into dyntick-idle
2232 mode during the locktorture test.
2233
2234 locktorture.shutdown_secs= [KNL]
2235 Set time (s) after boot system shutdown. This
2236 is useful for hands-off automated testing.
2237
2238 locktorture.stat_interval= [KNL]
2239 Time (s) between statistics printk()s.
2240
2241 locktorture.stutter= [KNL]
2242 Time (s) to stutter testing, for example,
2243 specifying five seconds causes the test to run for
2244 five seconds, wait for five seconds, and so on.
2245 This tests the locking primitive's ability to
2246 transition abruptly to and from idle.
2247
2248 locktorture.torture_runnable= [BOOT]
2249 Start locktorture running at boot time.
2250
2251 locktorture.torture_type= [KNL]
2252 Specify the locking implementation to test.
2253
2254 locktorture.verbose= [KNL]
2255 Enable additional printk() statements.
2256
Linus Torvalds1da177e2005-04-16 15:20:36 -07002257 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver
2258 Format: <irq>
2259
2260 loglevel= All Kernel Messages with a loglevel smaller than the
2261 console loglevel will be printed to the console. It can
2262 also be changed with klogd or other programs. The
2263 loglevels are defined as follows:
2264
2265 0 (KERN_EMERG) system is unusable
2266 1 (KERN_ALERT) action must be taken immediately
2267 2 (KERN_CRIT) critical conditions
2268 3 (KERN_ERR) error conditions
2269 4 (KERN_WARNING) warning conditions
2270 5 (KERN_NOTICE) normal but significant condition
2271 6 (KERN_INFO) informational
2272 7 (KERN_DEBUG) debug-level messages
2273
Randy Dunlapc756d082011-02-20 20:08:35 -08002274 log_buf_len=n[KMG] Sets the size of the printk ring buffer,
Luis R. Rodriguez23b28992014-08-06 16:08:56 -07002275 in bytes. n must be a power of two and greater
2276 than the minimal size. The minimal size is defined
2277 by LOG_BUF_SHIFT kernel config parameter. There is
2278 also CONFIG_LOG_CPU_MAX_BUF_SHIFT config parameter
2279 that allows to increase the default size depending on
2280 the number of CPUs. See init/Kconfig for more details.
Linus Torvalds1da177e2005-04-16 15:20:36 -07002281
Randy Dunlapaccaa242007-10-16 01:29:37 -07002282 logo.nologo [FB] Disables display of the built-in Linux logo.
2283 This may be used to provide more screen space for
2284 kernel log messages and is useful when debugging
2285 kernel boot problems.
2286
Linus Torvalds1da177e2005-04-16 15:20:36 -07002287 lp=0 [LP] Specify parallel ports to use, e.g,
2288 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses
2289 lp=reset first parallel port). 'lp=0' disables the
2290 lp=auto printer driver. 'lp=reset' (which can be
2291 specified in addition to the ports) causes
2292 attached printers to be reset. Using
2293 lp=port1,port2,... specifies the parallel ports
2294 to associate lp devices with, starting with
2295 lp0. A port specification may be 'none' to skip
2296 that lp device, or a parport name such as
2297 'parport0'. Specifying 'lp=auto' instead of a
2298 port specification list means that device IDs
2299 from each port should be examined, to see if
2300 an IEEE 1284-compliant printer is attached; if
2301 so, the driver will manage that printer.
2302 See also header of drivers/char/lp.c.
2303
2304 lpj=n [KNL]
2305 Sets loops_per_jiffy to given constant, thus avoiding
2306 time-consuming boot-time autodetection (up to 250 ms per
2307 CPU). 0 enables autodetection (default). To determine
2308 the correct value for your kernel, boot with normal
2309 autodetection and see what value is printed. Note that
2310 on SMP systems the preset will be applied to all CPUs,
2311 which is likely to cause problems if your CPUs need
2312 significantly divergent settings. An incorrect value
2313 will cause delays in the kernel to be wrong, leading to
2314 unpredictable I/O errors and other breakage. Although
2315 unlikely, in the extreme case this might damage your
2316 hardware.
2317
2318 ltpc= [NET]
2319 Format: <io>,<irq>,<dma>
2320
Randy Dunlap16290242011-08-13 12:34:52 -07002321 machvec= [IA-64] Force the use of a particular machine-vector
Randy Dunlapa9913042005-10-23 12:57:11 -07002322 (machvec) in a generic kernel.
2323 Example: machvec=hpzx1_swiotlb
Linus Torvalds1da177e2005-04-16 15:20:36 -07002324
Wu Zhangjin3209e702009-07-02 23:27:12 +08002325 machtype= [Loongson] Share the same kernel image file between different
2326 yeeloong laptop.
2327 Example: machtype=lemote-yeeloong-2f-7inch
2328
Randy Dunlap0cb55ad2009-04-05 15:55:22 -07002329 max_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory greater
2330 than or equal to this physical address is ignored.
Linus Torvalds1da177e2005-04-16 15:20:36 -07002331
2332 maxcpus= [SMP] Maximum number of processors that an SMP kernel
Baoquan He7c142bf2016-08-24 13:06:45 +08002333 will bring up during bootup. maxcpus=n : n >= 0 limits
2334 the kernel to bring up 'n' processors. Surely after
2335 bootup you can bring up the other plugged cpu by executing
2336 "echo 1 > /sys/devices/system/cpu/cpuX/online". So maxcpus
2337 only takes effect during system bootup.
2338 While n=0 is a special case, it is equivalent to "nosmp",
2339 which also disables the IO APIC.
Linus Torvalds1da177e2005-04-16 15:20:36 -07002340
Kay Sieversd134b002011-07-31 22:08:04 +02002341 max_loop= [LOOP] The number of loop block devices that get
2342 (loop.max_loop) unconditionally pre-created at init time. The default
2343 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead
2344 of statically allocating a predefined number, loop
2345 devices can be requested on-demand with the
2346 /dev/loop-control interface.
Bob Picco2b2c3752005-06-29 18:00:00 -07002347
Alan Coxcd4f0ef2007-07-31 00:37:59 -07002348 mce [X86-32] Machine Check Exception
Linus Torvalds1da177e2005-04-16 15:20:36 -07002349
Uwe Hermann71cced62008-10-20 09:32:21 -07002350 mce=option [X86-64] See Documentation/x86/x86_64/boot-options.txt
Andi Kleen909dd322007-10-17 18:04:38 +02002351
Linus Torvalds1da177e2005-04-16 15:20:36 -07002352 md= [HW] RAID subsystems devices and level
2353 See Documentation/md.txt.
Randy Dunlapa9913042005-10-23 12:57:11 -07002354
Linus Torvalds1da177e2005-04-16 15:20:36 -07002355 mdacon= [MDA]
2356 Format: <first>,<last>
2357 Specifies range of consoles to be captured by the MDA.
Randy Dunlapa9913042005-10-23 12:57:11 -07002358
Thomas Gleixner4e722ae2019-02-18 22:04:08 +01002359 mds= [X86,INTEL]
2360 Control mitigation for the Micro-architectural Data
2361 Sampling (MDS) vulnerability.
2362
2363 Certain CPUs are vulnerable to an exploit against CPU
2364 internal buffers which can forward information to a
2365 disclosure gadget under certain conditions.
2366
2367 In vulnerable processors, the speculatively
2368 forwarded data can be used in a cache side channel
2369 attack, to access data to which the attacker does
2370 not have direct access.
2371
2372 This parameter controls the MDS mitigation. The
2373 options are:
2374
Josh Poimboeuff02eee62019-04-02 09:59:33 -05002375 full - Enable MDS mitigation on vulnerable CPUs
2376 full,nosmt - Enable MDS mitigation and disable
2377 SMT on vulnerable CPUs
2378 off - Unconditionally disable MDS mitigation
Thomas Gleixner4e722ae2019-02-18 22:04:08 +01002379
2380 Not specifying this option is equivalent to
2381 mds=full.
2382
Thomas Gleixner3880bc12019-02-19 00:02:31 +01002383 For details see: Documentation/hw-vuln/mds.rst
2384
Linus Torvalds1da177e2005-04-16 15:20:36 -07002385 mem=nn[KMG] [KNL,BOOT] Force usage of a specific amount of memory
2386 Amount of memory to be used when the kernel is not able
2387 to see the whole system memory or for test.
Wen Congyangfbb97d82012-12-17 15:59:29 -08002388 [X86] Work as limiting max address. Use together
2389 with memmap= to avoid physical address space collisions.
2390 Without memmap= PCI devices could be placed at addresses
2391 belonging to unused RAM.
Linus Torvalds1da177e2005-04-16 15:20:36 -07002392
Alan Coxcd4f0ef2007-07-31 00:37:59 -07002393 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel
Linus Torvalds1da177e2005-04-16 15:20:36 -07002394 memory.
2395
Paul Mundt6902aa82008-09-21 17:14:42 +09002396 memchunk=nn[KMG]
2397 [KNL,SH] Allow user to override the default size for
2398 per-device physically contiguous DMA buffers.
2399
Vitaly Kuznetsov86dd9952016-05-19 17:13:06 -07002400 memhp_default_state=online/offline
2401 [KNL] Set the initial state for the memory hotplug
2402 onlining policy. If not specified, the default value is
2403 set according to the
2404 CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE kernel config
2405 option.
2406 See Documentation/memory-hotplug.txt.
2407
Jaswinder Singh Rajput6cececf2009-04-14 14:03:43 +05302408 memmap=exactmap [KNL,X86] Enable setting of an exact
Linus Torvalds1da177e2005-04-16 15:20:36 -07002409 E820 memory map, as specified by the user.
2410 Such memmap=exactmap lines can be constructed based on
2411 BIOS output or other requirements. See the memmap=nn@ss
2412 option description.
2413
2414 memmap=nn[KMG]@ss[KMG]
Randy Dunlap277cba12014-02-06 12:04:19 -08002415 [KNL] Force usage of a specific region of memory.
2416 Region of memory to be used is from ss to ss+nn.
Linus Torvalds1da177e2005-04-16 15:20:36 -07002417
2418 memmap=nn[KMG]#ss[KMG]
2419 [KNL,ACPI] Mark specific memory as ACPI data.
Randy Dunlap277cba12014-02-06 12:04:19 -08002420 Region of memory to be marked is from ss to ss+nn.
Linus Torvalds1da177e2005-04-16 15:20:36 -07002421
2422 memmap=nn[KMG]$ss[KMG]
2423 [KNL,ACPI] Mark specific memory as reserved.
Randy Dunlap277cba12014-02-06 12:04:19 -08002424 Region of memory to be reserved is from ss to ss+nn.
Pavel Machek13128482008-03-24 12:29:43 -07002425 Example: Exclude memory from 0x18690000-0x1869ffff
2426 memmap=64K$0x18690000
2427 or
2428 memmap=0x10000$0x18690000
Linus Torvalds1da177e2005-04-16 15:20:36 -07002429
Christoph Hellwigec776ef2015-04-01 09:12:18 +02002430 memmap=nn[KMG]!ss[KMG]
2431 [KNL,X86] Mark specific memory as protected.
2432 Region of memory to be used, from ss to ss+nn.
2433 The memory region may be marked as e820 type 12 (0xc)
2434 and is NVDIMM or ADR memory.
2435
Jeremy Fitzhardinge9f077872008-09-07 01:51:34 -07002436 memory_corruption_check=0/1 [X86]
2437 Some BIOSes seem to corrupt the first 64k of
2438 memory when doing things like suspend/resume.
2439 Setting this option will scan the memory
2440 looking for corruption. Enabling this will
2441 both detect corruption and prevent the kernel
2442 from using the memory being corrupted.
2443 However, its intended as a diagnostic tool; if
2444 repeatable BIOS-originated corruption always
2445 affects the same memory, you can use memmap=
2446 to prevent the kernel from using that memory.
2447
2448 memory_corruption_check_size=size [X86]
2449 By default it checks for corruption in the low
2450 64k, making this memory unavailable for normal
2451 use. Use this parameter to scan for
2452 corruption in more or less memory.
2453
2454 memory_corruption_check_period=seconds [X86]
2455 By default it checks for corruption every 60
2456 seconds. Use this parameter to check at some
2457 other rate. 0 disables periodic checking.
2458
Vladimir Murzine4b0db72015-04-14 15:48:43 -07002459 memtest= [KNL,X86,ARM] Enable memtest
Yinghai Luc64df702008-03-21 18:56:19 -07002460 Format: <integer>
Yinghai Luc64df702008-03-21 18:56:19 -07002461 default : 0 <disable>
Andreas Herrmann9e5f6cf2009-02-25 11:30:45 +01002462 Specifies the number of memtest passes to be
2463 performed. Each pass selects another test
2464 pattern from a given set of patterns. Memtest
2465 fills the memory with this pattern, validates
2466 memory contents and reserves bad memory
2467 regions that are detected.
Yinghai Luc64df702008-03-21 18:56:19 -07002468
Linus Torvalds1da177e2005-04-16 15:20:36 -07002469 meye.*= [HW] Set MotionEye Camera parameters
2470 See Documentation/video4linux/meye.txt.
2471
Andres Salomon8f368812007-10-12 23:04:06 +02002472 mfgpt_irq= [IA-32] Specify the IRQ to use for the
2473 Multi-Function General Purpose Timers on AMD Geode
2474 platforms.
2475
Willy Tarreaue6c4dc62008-01-30 13:33:33 +01002476 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when
2477 the BIOS has incorrectly applied a workaround. TinyBIOS
2478 version 0.98 is known to be affected, 0.99 fixes the
2479 problem by letting the user disable the workaround.
2480
Linus Torvalds1da177e2005-04-16 15:20:36 -07002481 mga= [HW,DRM]
2482
Randy Dunlap1c207f92008-11-19 15:36:16 -08002483 min_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory below this
2484 physical address is ignored.
2485
Michel Pollet39f45d72009-05-20 11:10:31 +01002486 mini2440= [ARM,HW,KNL]
2487 Format:[0..2][b][c][t]
2488 Default: "0tb"
2489 MINI2440 configuration specification:
2490 0 - The attached screen is the 3.5" TFT
2491 1 - The attached screen is the 7" TFT
2492 2 - The VGA Shield is attached (1024x768)
2493 Leaving out the screen size parameter will not load
2494 the TFT driver, and the framebuffer will be left
2495 unconfigured.
2496 b - Enable backlight. The TFT backlight pin will be
2497 linked to the kernel VESA blanking code and a GPIO
2498 LED. This parameter is not necessary when using the
2499 VGA shield.
2500 c - Enable the s3c camera interface.
2501 t - Reserved for enabling touchscreen support. The
2502 touchscreen support is not enabled in the mainstream
2503 kernel as of 2.6.30, a preliminary port can be found
2504 in the "bleeding edge" mini2440 support kernel at
2505 http://repo.or.cz/w/linux-2.6/mini2440.git
2506
Josh Poimboeufedda9c32019-04-12 15:39:28 -05002507 mitigations=
Josh Poimboeuf17092842019-04-12 15:39:29 -05002508 [X86] Control optional mitigations for CPU
2509 vulnerabilities. This is a set of curated,
2510 arch-independent options, each of which is an
2511 aggregation of existing arch-specific options.
Josh Poimboeufedda9c32019-04-12 15:39:28 -05002512
2513 off
2514 Disable all optional CPU mitigations. This
2515 improves system performance, but it may also
2516 expose users to several CPU vulnerabilities.
Josh Poimboeuf17092842019-04-12 15:39:29 -05002517 Equivalent to: nopti [X86]
Josh Poimboeuf90d45f02019-07-08 11:52:26 -05002518 nospectre_v1 [X86]
Josh Poimboeuf17092842019-04-12 15:39:29 -05002519 nospectre_v2 [X86]
2520 spectre_v2_user=off [X86]
2521 spec_store_bypass_disable=off [X86]
2522 l1tf=off [X86]
Josh Poimboeuf025b9cf2019-04-17 16:39:02 -05002523 mds=off [X86]
Pawan Guptaba54aad2019-10-23 12:32:55 +02002524 tsx_async_abort=off [X86]
Paolo Bonzinia7ad7942019-11-04 12:22:02 +01002525 kvm.nx_huge_pages=off [X86]
2526
2527 Exceptions:
2528 This does not have any effect on
2529 kvm.nx_huge_pages when
2530 kvm.nx_huge_pages=force.
Josh Poimboeufedda9c32019-04-12 15:39:28 -05002531
2532 auto (default)
2533 Mitigate all CPU vulnerabilities, but leave SMT
2534 enabled, even if it's vulnerable. This is for
2535 users who don't want to be surprised by SMT
2536 getting disabled across kernel upgrades, or who
2537 have other ways of avoiding SMT-based attacks.
Josh Poimboeuf17092842019-04-12 15:39:29 -05002538 Equivalent to: (default behavior)
Josh Poimboeufedda9c32019-04-12 15:39:28 -05002539
2540 auto,nosmt
2541 Mitigate all CPU vulnerabilities, disabling SMT
2542 if needed. This is for users who always want to
2543 be fully mitigated, even if it means losing SMT.
Josh Poimboeuf17092842019-04-12 15:39:29 -05002544 Equivalent to: l1tf=flush,nosmt [X86]
Josh Poimboeuf025b9cf2019-04-17 16:39:02 -05002545 mds=full,nosmt [X86]
Pawan Guptaba54aad2019-10-23 12:32:55 +02002546 tsx_async_abort=full,nosmt [X86]
Josh Poimboeufedda9c32019-04-12 15:39:28 -05002547
Mel Gorman6b74ab92008-07-23 21:26:49 -07002548 mminit_loglevel=
2549 [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this
2550 parameter allows control of the logging verbosity for
2551 the additional memory initialisation checks. A value
2552 of 0 disables mminit logging and a level of 4 will
2553 log everything. Information is printed at KERN_DEBUG
2554 so loglevel=8 may also need to be specified.
2555
Rusty Russell106a4ee2012-09-26 10:09:40 +01002556 module.sig_enforce
2557 [KNL] When CONFIG_MODULE_SIG is set, this means that
2558 modules without (valid) signatures will fail to load.
Paul Bolle2a039be2013-03-25 20:42:06 +01002559 Note that if CONFIG_MODULE_SIG_FORCE is set, that
Rusty Russell106a4ee2012-09-26 10:09:40 +01002560 is always true, so this option does nothing.
2561
Prarit Bhargavabe7de5f2016-07-21 15:37:56 +09302562 module_blacklist= [KNL] Do not load a comma-separated list of
2563 modules. Useful for debugging problem modules.
2564
Linus Torvalds1da177e2005-04-16 15:20:36 -07002565 mousedev.tap_time=
2566 [MOUSE] Maximum time between finger touching and
2567 leaving touchpad surface for touch to be considered
2568 a tap and be reported as a left button click (for
2569 touchpads working in absolute mode only).
2570 Format: <msecs>
2571 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices
2572 reporting absolute coordinates, such as tablets
2573 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices
2574 reporting absolute coordinates, such as tablets
2575
Jaswinder Singh Rajput6cececf2009-04-14 14:03:43 +05302576 movablecore=nn[KMG] [KNL,X86,IA-64,PPC] This parameter
Randy Dunlap0cb55ad2009-04-05 15:55:22 -07002577 is similar to kernelcore except it specifies the
2578 amount of memory used for migratable allocations.
2579 If both kernelcore and movablecore is specified,
2580 then kernelcore will be at *least* the specified
2581 value but may be more. If movablecore on its own
2582 is specified, the administrator must be careful
2583 that the amount of memory usable for all allocations
2584 is not too small.
2585
Tang Chenc5320922013-11-12 15:08:10 -08002586 movable_node [KNL,X86] Boot-time switch to enable the effects
2587 of CONFIG_MOVABLE_NODE=y. See mm/Kconfig for details.
2588
Linus Torvalds1da177e2005-04-16 15:20:36 -07002589 MTD_Partition= [MTD]
2590 Format: <name>,<region-number>,<size>,<offset>
2591
Randy Dunlapa9913042005-10-23 12:57:11 -07002592 MTD_Region= [MTD] Format:
2593 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>]
Linus Torvalds1da177e2005-04-16 15:20:36 -07002594
2595 mtdparts= [MTD]
Randy Dunlapc8facbb2007-07-31 00:37:40 -07002596 See drivers/mtd/cmdlinepart.c.
Linus Torvalds1da177e2005-04-16 15:20:36 -07002597
Will Schmidt4e89a2d2010-09-28 15:33:12 +00002598 multitce=off [PPC] This parameter disables the use of the pSeries
2599 firmware feature for updating multiple TCE entries
2600 at a time.
2601
Rohit Hagargundgi5988af22009-05-12 13:46:57 -07002602 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration
2603
2604 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock]
2605
2606 boundary - index of last SLC block on Flex-OneNAND.
2607 The remaining blocks are configured as MLC blocks.
2608 lock - Configure if Flex-OneNAND boundary should be locked.
2609 Once locked, the boundary cannot be changed.
2610 1 indicates lock status, 0 indicates unlock status.
2611
Ben Dooks9db829f2008-07-03 11:24:29 +01002612 mtdset= [ARM]
2613 ARM/S3C2412 JIVE boot control
2614
2615 See arch/arm/mach-s3c2412/mach-jive.c
2616
Linus Torvalds1da177e2005-04-16 15:20:36 -07002617 mtouchusb.raw_coordinates=
Randy Dunlapa9913042005-10-23 12:57:11 -07002618 [HW] Make the MicroTouch USB driver use raw coordinates
2619 ('y', default) or cooked coordinates ('n')
Linus Torvalds1da177e2005-04-16 15:20:36 -07002620
Randy Dunlap0cb55ad2009-04-05 15:55:22 -07002621 mtrr_chunk_size=nn[KMG] [X86]
Matt LaPlante19f59462009-04-27 15:06:31 +02002622 used for mtrr cleanup. It is largest continuous chunk
Randy Dunlap0cb55ad2009-04-05 15:55:22 -07002623 that could hold holes aka. UC entries.
2624
2625 mtrr_gran_size=nn[KMG] [X86]
2626 Used for mtrr cleanup. It is granularity of mtrr block.
2627 Default is 1.
2628 Large value could prevent small alignment from
2629 using up MTRRs.
2630
2631 mtrr_spare_reg_nr=n [X86]
2632 Format: <integer>
2633 Range: 0,7 : spare reg number
2634 Default : 1
2635 Used for mtrr cleanup. It is spare mtrr entries number.
2636 Set to 2 or more if your graphical card needs more.
2637
Linus Torvalds1da177e2005-04-16 15:20:36 -07002638 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card
2639
Linus Torvalds1da177e2005-04-16 15:20:36 -07002640 netdev= [NET] Network devices parameters
2641 Format: <irq>,<io>,<mem_start>,<mem_end>,<name>
2642 Note that mem_start is often overloaded to mean
2643 something different and driver-specific.
Randy Dunlapa9913042005-10-23 12:57:11 -07002644 This usage is only documented in each driver source
2645 file if at all.
2646
Krzysztof Piotr Oledzki58401572008-07-21 10:01:34 -07002647 nf_conntrack.acct=
2648 [NETFILTER] Enable connection tracking flow accounting
2649 0 to disable accounting
2650 1 to enable accounting
Tim Gardnerd70a0112010-06-25 14:46:56 +02002651 Default value is 0.
Krzysztof Piotr Oledzki58401572008-07-21 10:01:34 -07002652
Chuck Lever306a0752010-09-17 10:54:37 -04002653 nfsaddrs= [NFS] Deprecated. Use ip= instead.
J. Bruce Fieldsdc7a0812009-10-27 14:41:35 -04002654 See Documentation/filesystems/nfs/nfsroot.txt.
Linus Torvalds1da177e2005-04-16 15:20:36 -07002655
2656 nfsroot= [NFS] nfs root filesystem for disk-less boxes.
J. Bruce Fieldsdc7a0812009-10-27 14:41:35 -04002657 See Documentation/filesystems/nfs/nfsroot.txt.
Linus Torvalds1da177e2005-04-16 15:20:36 -07002658
Chuck Lever306a0752010-09-17 10:54:37 -04002659 nfsrootdebug [NFS] enable nfsroot debugging messages.
2660 See Documentation/filesystems/nfs/nfsroot.txt.
2661
Trond Myklebust5405fc42016-08-29 20:03:52 -04002662 nfs.callback_nr_threads=
2663 [NFSv4] set the total number of threads that the
2664 NFS client will assign to service NFSv4 callback
2665 requests.
2666
Trond Myklebusta72b4422006-01-03 09:55:41 +01002667 nfs.callback_tcpport=
2668 [NFS] set the TCP port on which the NFSv4 callback
2669 channel should listen.
2670
Trond Myklebuste571cbf2009-08-19 18:12:27 -04002671 nfs.cache_getent=
2672 [NFS] sets the pathname to the program which is used
2673 to update the NFS client cache entries.
2674
2675 nfs.cache_getent_timeout=
2676 [NFS] sets the timeout after which an attempt to
2677 update a cache entry is deemed to have failed.
2678
Trond Myklebust58df0952006-01-03 09:55:57 +01002679 nfs.idmap_cache_timeout=
2680 [NFS] set the maximum lifetime for idmapper cache
2681 entries.
2682
Trond Myklebustf43bf0b2007-10-09 12:01:04 -04002683 nfs.enable_ino64=
2684 [NFS] enable 64-bit inode numbers.
2685 If zero, the NFS client will fake up a 32-bit inode
2686 number for the readdir() and stat() syscalls instead
2687 of returning the full 64-bit number.
2688 The default is to return 64-bit inode numbers.
2689
Trond Myklebust5405fc42016-08-29 20:03:52 -04002690 nfs.max_session_cb_slots=
2691 [NFSv4.1] Sets the maximum number of session
2692 slots the client will assign to the callback
2693 channel. This determines the maximum number of
2694 callbacks the client will process in parallel for
2695 a particular server.
2696
Trond Myklebustef159e92012-02-06 19:50:40 -05002697 nfs.max_session_slots=
2698 [NFSv4.1] Sets the maximum number of session slots
2699 the client will attempt to negotiate with the server.
2700 This limits the number of simultaneous RPC requests
2701 that the client can send to the NFSv4.1 server.
2702 Note that there is little point in setting this
2703 value higher than the max_tcp_slot_table_limit.
2704
Trond Myklebustb064eca22011-02-22 15:44:32 -08002705 nfs.nfs4_disable_idmapping=
Trond Myklebust074b1d12012-01-09 13:46:26 -05002706 [NFSv4] When set to the default of '1', this option
2707 ensures that both the RPC level authentication
2708 scheme and the NFS level operations agree to use
2709 numeric uids/gids if the mount is using the
2710 'sec=sys' security flavour. In effect it is
2711 disabling idmapping, which can make migration from
2712 legacy NFSv2/v3 systems to NFSv4 easier.
2713 Servers that do not support this mode of operation
2714 will be autodetected by the client, and it will fall
2715 back to using the idmapper.
2716 To turn off this behaviour, set the value to '0'.
Chuck Lever6f2ea7f2012-09-14 17:24:41 -04002717 nfs.nfs4_unique_id=
2718 [NFS4] Specify an additional fixed unique ident-
2719 ification string that NFSv4 clients can insert into
2720 their nfs_client_id4 string. This is typically a
2721 UUID that is generated at system install time.
Trond Myklebustb064eca22011-02-22 15:44:32 -08002722
Weston Andros Adamsondb8ac8b2012-02-17 15:20:24 -05002723 nfs.send_implementation_id =
2724 [NFSv4.1] Send client implementation identification
2725 information in exchange_id requests.
2726 If zero, no implementation identification information
2727 will be sent.
2728 The default is to send the implementation identification
2729 information.
Trond Myklebustf6de7a32013-09-04 10:08:54 -04002730
2731 nfs.recover_lost_locks =
2732 [NFSv4] Attempt to recover locks that were lost due
2733 to a lease timeout on the server. Please note that
2734 doing this risks data corruption, since there are
2735 no guarantees that the file will remain unchanged
2736 after the locks are lost.
2737 If you want to enable the kernel legacy behaviour of
2738 attempting to recover these locks, then set this
2739 parameter to '1'.
2740 The default parameter value of '0' causes the kernel
2741 not to attempt recovery of lost locks.
Weston Andros Adamsondb8ac8b2012-02-17 15:20:24 -05002742
Trond Myklebustbbf58bf2015-08-24 20:39:18 -04002743 nfs4.layoutstats_timer =
2744 [NFSv4.2] Change the rate at which the kernel sends
2745 layoutstats to the pNFS metadata server.
2746
2747 Setting this to value to 0 causes the kernel to use
2748 whatever value is the default set by the layout
2749 driver. A non-zero value sets the minimum interval
2750 in seconds between layoutstats transmissions.
2751
J. Bruce Fieldse9541ce2012-03-22 16:07:18 -04002752 nfsd.nfs4_disable_idmapping=
2753 [NFSv4] When set to the default of '1', the NFSv4
2754 server will return only numeric uids and gids to
2755 clients using auth_sys, and will accept numeric uids
2756 and gids from such clients. This is intended to ease
2757 migration from NFSv2/v3.
Weston Andros Adamsondb8ac8b2012-02-17 15:20:24 -05002758
Sachin Bhamare18d98f62012-03-19 20:47:58 -07002759 objlayoutdriver.osd_login_prog=
2760 [NFS] [OBJLAYOUT] sets the pathname to the program which
2761 is used to automatically discover and login into new
2762 osd-targets. Please see:
2763 Documentation/filesystems/pnfs.txt for more explanations
2764
Paul Mundt1e1030d2009-09-01 17:38:32 +09002765 nmi_debug= [KNL,AVR32,SH] Specify one or more actions to take
Haavard Skinnemoene7ba1762007-10-10 14:58:29 +02002766 when a NMI is triggered.
2767 Format: [state][,regs][,debounce][,die]
2768
Jaswinder Singh Rajput6cececf2009-04-14 14:03:43 +05302769 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels
Don Zickusfef2c9b2011-03-22 16:34:16 -07002770 Format: [panic,][nopanic,][num]
Ulrich Obergfell195daf62015-04-14 15:44:13 -07002771 Valid num: 0 or 1
Pranith Kumar334bb792015-10-10 15:40:42 -04002772 0 - turn hardlockup detector in nmi_watchdog off
2773 1 - turn hardlockup detector in nmi_watchdog on
Randy Dunlap0cb55ad2009-04-05 15:55:22 -07002774 When panic is specified, panic when an NMI watchdog
Don Zickusfef2c9b2011-03-22 16:34:16 -07002775 timeout occurs (or 'nopanic' to override the opposite
Pranith Kumar334bb792015-10-10 15:40:42 -04002776 default). To disable both hard and soft lockup detectors,
2777 please see 'nowatchdog'.
Randy Dunlap0cb55ad2009-04-05 15:55:22 -07002778 This is useful when you use a panic=... timeout and
2779 need the box quickly up again.
Linus Torvalds1da177e2005-04-16 15:20:36 -07002780
Anton Vorontsovbff38772009-07-08 11:10:56 -07002781 netpoll.carrier_timeout=
2782 [NET] Specifies amount of time (in seconds) that
2783 netpoll should wait for a carrier. By default netpoll
2784 waits 4 seconds.
2785
Alan Coxcd4f0ef2007-07-31 00:37:59 -07002786 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths
Linus Torvalds1da177e2005-04-16 15:20:36 -07002787 emulation library even if a 387 maths coprocessor
2788 is present.
2789
Randy Dunlap0cb55ad2009-04-05 15:55:22 -07002790 no_console_suspend
2791 [HW] Never suspend the console
2792 Disable suspending of consoles during suspend and
2793 hibernate operations. Once disabled, debugging
2794 messages can reach various consoles while the rest
2795 of the system is being put to sleep (ie, while
2796 debugging driver suspend/resume hooks). This may
2797 not work reliably with all consoles, but is known
2798 to work with serial and VGA consoles.
Yanmin Zhang134620f2011-10-31 17:11:27 -07002799 To facilitate more flexible debugging, we also add
2800 console_suspend, a printk module parameter to control
2801 it. Users could use console_suspend (usually
2802 /sys/module/printk/parameters/console_suspend) to
2803 turn on/off it dynamically.
Randy Dunlap0cb55ad2009-04-05 15:55:22 -07002804
Christoph Lameterc1aee212007-05-31 00:40:47 -07002805 noaliencache [MM, NUMA, SLAB] Disables the allocation of alien
2806 caches in the slab allocator. Saves per-node memory,
2807 but will impact performance.
Paul Menage3395ee02006-12-06 20:32:16 -08002808
Randy Dunlapa9913042005-10-23 12:57:11 -07002809 noalign [KNL,ARM]
2810
Vasily Gorbikd69aa5e2018-04-27 07:36:39 +02002811 noaltinstr [S390] Disables alternative instructions patching
2812 (CPU alternatives feature).
2813
Linus Torvalds1da177e2005-04-16 15:20:36 -07002814 noapic [SMP,APIC] Tells the kernel to not make use of any
2815 IOAPICs that may be present in the system.
2816
Mike Galbraith5091faa2010-11-30 14:18:03 +01002817 noautogroup Disable scheduler automatic task group creation.
2818
Linus Torvalds1da177e2005-04-16 15:20:36 -07002819 nobats [PPC] Do not use BATs for mapping kernel lowmem
2820 on "Classic" PPC cores.
2821
2822 nocache [ARM]
Randy Dunlapa9913042005-10-23 12:57:11 -07002823
Randy Dunlap0cb55ad2009-04-05 15:55:22 -07002824 noclflush [BUGS=X86] Don't use the CLFLUSH instruction
2825
Shailabh Nagar163ecdf2006-07-30 03:03:11 -07002826 nodelayacct [KNL] Disable per-task delay accounting
2827
Paul Mundt6902aa82008-09-21 17:14:42 +09002828 nodsp [SH] Disable hardware DSP at boot time.
2829
Dave Youngb2e0a542014-08-14 17:15:26 +08002830 noefi Disable EFI runtime services support.
Huang, Ying8b2cb7a2008-01-30 13:32:11 +01002831
Linus Torvalds1da177e2005-04-16 15:20:36 -07002832 noexec [IA-64]
2833
Jaswinder Singh Rajput6cececf2009-04-14 14:03:43 +05302834 noexec [X86]
Jiri Slabyf5a1b192008-04-12 10:28:25 +02002835 On X86-32 available only on PAE configured kernels.
Linus Torvalds1da177e2005-04-16 15:20:36 -07002836 noexec=on: enable non-executable mappings (default)
Jiri Slabyf5a1b192008-04-12 10:28:25 +02002837 noexec=off: disable non-executable mappings
2838
H. Peter Anvin52b61792012-09-21 12:43:13 -07002839 nosmap [X86]
2840 Disable SMAP (Supervisor Mode Access Prevention)
2841 even if it is supported by processor.
2842
Fenghua Yude5397a2011-05-11 16:51:05 -07002843 nosmep [X86]
H. Peter Anvin52b61792012-09-21 12:43:13 -07002844 Disable SMEP (Supervisor Mode Execution Prevention)
Fenghua Yude5397a2011-05-11 16:51:05 -07002845 even if it is supported by processor.
2846
Jiri Slabyf5a1b192008-04-12 10:28:25 +02002847 noexec32 [X86-64]
2848 This affects only 32-bit executables.
2849 noexec32=on: enable non-executable mappings (default)
2850 read doesn't imply executable mappings
2851 noexec32=off: disable non-executable mappings
2852 read implies executable mappings
Linus Torvalds1da177e2005-04-16 15:20:36 -07002853
Maciej W. Rozyckifab43ef2015-04-03 23:23:34 +01002854 nofpu [MIPS,SH] Disable hardware FPU at boot time.
Paul Mundt6902aa82008-09-21 17:14:42 +09002855
Alan Coxcd4f0ef2007-07-31 00:37:59 -07002856 nofxsr [BUGS=X86-32] Disables x86 floating point extended
Chuck Ebbert4f886512006-03-23 02:59:34 -08002857 register save and restore. The kernel will only save
2858 legacy floating-point registers on task switch.
Linus Torvalds1da177e2005-04-16 15:20:36 -07002859
Toshi Kani0ddab1d2015-04-14 15:47:20 -07002860 nohugeiomap [KNL,x86] Disable kernel huge I/O mappings.
2861
Sascha Silbe52c48c52016-04-05 12:53:38 +02002862 nosmt [KNL,S390] Disable symmetric multithreading (SMT).
2863 Equivalent to smt=1.
2864
Thomas Gleixnerf37486c2018-05-29 17:48:27 +02002865 [KNL,x86] Disable symmetric multithreading (SMT).
Thomas Gleixnerfe2a9552018-06-29 16:05:47 +02002866 nosmt=force: Force disable SMT, cannot be undone
2867 via the sysfs control file.
Thomas Gleixnerf37486c2018-05-29 17:48:27 +02002868
Josh Poimboeuf90d45f02019-07-08 11:52:26 -05002869 nospectre_v1 [X86,PPC] Disable mitigations for Spectre Variant 1
2870 (bounds check bypass). With this option data leaks are
2871 possible in the system.
2872
Diana Craciun0feb3712018-12-12 16:03:10 +02002873 nospectre_v2 [X86,PPC_FSL_BOOK3E] Disable all mitigations for the Spectre variant 2
David Woodhouse8f96937ee2018-01-11 21:46:26 +00002874 (indirect branch prediction) vulnerability. System may
2875 allow data leaks with this option, which is equivalent
2876 to spectre_v2=off.
2877
Konrad Rzeszutek Wilk6f70a552018-04-25 22:04:21 -04002878 nospec_store_bypass_disable
2879 [HW] Disable all mitigations for the Speculative Store Bypass vulnerability
2880
Suresh Siddha0c752a92009-05-22 12:17:45 -07002881 noxsave [BUGS=X86] Disables x86 extended register state save
2882 and restore using xsave. The kernel will fallback to
2883 enabling legacy floating-point and sse state.
2884
Fenghua Yub6f42a42014-05-29 11:12:31 -07002885 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended
2886 register states. The kernel will fall back to use
2887 xsave to save the states. By using this parameter,
2888 performance of saving the states is degraded because
2889 xsave doesn't support modified optimization while
2890 xsaveopt supports it on xsaveopt enabled systems.
2891
2892 noxsaves [X86] Disables xsaves and xrstors used in saving and
2893 restoring x86 extended register state in compacted
2894 form of xsave area. The kernel will fall back to use
2895 xsaveopt and xrstor to save and restore the states
2896 in standard form of xsave area. By using this
2897 parameter, xsave area per process might occupy more
2898 memory on xsaves enabled systems.
2899
Paulius Zaleckas01a24d22009-03-31 13:55:44 +01002900 nohlt [BUGS=ARM,SH] Tells the kernel that the sleep(SH) or
2901 wfi(ARM) instruction doesn't work correctly and not to
2902 use it. This is also useful when using JTAG debugger.
Randy Dunlapa9913042005-10-23 12:57:11 -07002903
Serge E. Hallyn1f29fae2008-11-05 16:08:52 -06002904 no_file_caps Tells the kernel not to honor file capabilities. The
2905 only way then for a file to be executed with privilege
2906 is to be setuid root or executed by root.
2907
Linus Torvalds1da177e2005-04-16 15:20:36 -07002908 nohalt [IA-64] Tells the kernel not to use the power saving
2909 function PAL_HALT_LIGHT when idle. This increases
2910 power-consumption. On the positive side, it reduces
2911 interrupt wake-up latency, which may improve performance
2912 in certain environments such as networked servers or
2913 real-time systems.
2914
Kees Cooka6e15a32014-06-13 13:30:35 -07002915 nohibernate [HIBERNATION] Disable hibernation and resume.
2916
Thomas Gleixner79bf2bb2007-02-16 01:28:03 -08002917 nohz= [KNL] Boottime enable/disable dynamic ticks
2918 Valid arguments: on, off
2919 Default: on
2920
Frederic Weisbeckerc5bfece2013-04-12 16:45:34 +02002921 nohz_full= [KNL,BOOT]
Noam Camus2d13e6c2016-10-11 13:51:35 -07002922 The argument is a cpu list, as described above.
Frederic Weisbeckerc5bfece2013-04-12 16:45:34 +02002923 In kernels built with CONFIG_NO_HZ_FULL=y, set
Frederic Weisbeckera8318812012-12-18 17:32:19 +01002924 the specified list of CPUs whose tick will be stopped
Frederic Weisbecker0453b432013-03-27 02:18:34 +01002925 whenever possible. The boot CPU will be forced outside
2926 the range to maintain the timekeeping.
Frederic Weisbeckerd1e43fa2013-03-26 23:47:24 +01002927 The CPUs in this range must also be included in the
2928 rcu_nocbs= set.
Frederic Weisbeckera8318812012-12-18 17:32:19 +01002929
Paul Mundteeee7852009-04-02 12:31:16 +09002930 noiotrap [SH] Disables trapped I/O port accesses.
2931
Alan Coxcd4f0ef2007-07-31 00:37:59 -07002932 noirqdebug [X86-32] Disables the code which attempts to detect and
Linus Torvalds1da177e2005-04-16 15:20:36 -07002933 disable unhandled interrupt sources.
2934
Jaswinder Singh Rajput6cececf2009-04-14 14:03:43 +05302935 no_timer_check [X86,APIC] Disables the code which tests for
Zachary Amsden8542b202006-12-07 02:14:09 +01002936 broken timer IRQ sources.
2937
Linus Torvalds1da177e2005-04-16 15:20:36 -07002938 noisapnp [ISAPNP] Disables ISA PnP code.
2939
2940 noinitrd [RAM] Tells the kernel not to load any configured
2941 initial RAM disk.
2942
Weidong Han03ea8152009-04-17 16:42:15 +08002943 nointremap [X86-64, Intel-IOMMU] Do not enable interrupt
2944 remapping.
Chris Wrightd1423d52010-07-20 11:06:49 -07002945 [Deprecated - use intremap=off]
Weidong Han03ea8152009-04-17 16:42:15 +08002946
Linus Torvalds1da177e2005-04-16 15:20:36 -07002947 nointroute [IA-64]
2948
Andy Lutomirskid12a72b2016-01-29 11:42:58 -08002949 noinvpcid [X86] Disable the INVPCID cpu feature.
2950
Randy Dunlap16290242011-08-13 12:34:52 -07002951 nojitter [IA-64] Disables jitter checking for ITC timers.
Tony Luck0aa366f2007-07-20 11:22:30 -07002952
Jiri Kosina9cf4c4f2010-08-16 17:51:20 +02002953 no-kvmclock [X86,KVM] Disable paravirtualized KVM clock driver
2954
Gleb Natapovfd10cde2010-10-14 11:22:51 +02002955 no-kvmapf [X86,KVM] Disable paravirtualized asynchronous page
2956 fault handling.
2957
Glauber Costad910f5c2011-07-11 15:28:19 -04002958 no-steal-acc [X86,KVM] Disable paravirtualized steal time accounting.
2959 steal time is computed, but won't influence scheduler
2960 behaviour
2961
Alan Coxcd4f0ef2007-07-31 00:37:59 -07002962 nolapic [X86-32,APIC] Do not enable or use the local APIC.
Linus Torvalds1da177e2005-04-16 15:20:36 -07002963
Alan Coxcd4f0ef2007-07-31 00:37:59 -07002964 nolapic_timer [X86-32,APIC] Do not use the local APIC timer.
Thomas Gleixnerad62ca22007-03-22 00:11:21 -08002965
Linus Torvalds1da177e2005-04-16 15:20:36 -07002966 noltlbs [PPC] Do not use large page/tlb entries for kernel
Christophe Leroyf15eea62016-02-09 17:07:52 +01002967 lowmem mapping on PPC40x and PPC8xx
Linus Torvalds1da177e2005-04-16 15:20:36 -07002968
Horms312f1f02006-02-22 09:57:55 +09002969 nomca [IA-64] Disable machine check abort handling
2970
Masanari Iida13696e02015-05-16 02:16:43 +09002971 nomce [X86-32] Disable Machine Check Exception
Hormsabe37e52006-04-01 01:36:09 +02002972
Andres Salomon83d73842007-10-12 23:04:06 +02002973 nomfgpt [X86-32] Disable Multi-Function General Purpose
2974 Timer usage (for AMD Geode machines).
2975
Don Zickusbda62632011-10-13 15:14:27 -04002976 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to
2977 shutdown the other cpus. Instead use the REBOOT_VECTOR
2978 irq.
2979
Dave Young02608be2012-02-01 10:33:14 +08002980 nomodule Disable module load
2981
Jiri Kosina016ddd92010-01-18 17:05:40 +01002982 nopat [X86] Disable PAT (page attribute table extension of
2983 pagetables) support.
2984
Andy Lutomirskie6a29322017-06-29 08:53:20 -07002985 nopcid [X86-64] Disable the PCID cpu feature.
2986
Randy Dunlap0cb55ad2009-04-05 15:55:22 -07002987 norandmaps Don't use address space randomization. Equivalent to
2988 echo 0 > /proc/sys/kernel/randomize_va_space
2989
Alan Coxcd4f0ef2007-07-31 00:37:59 -07002990 noreplace-smp [X86-32,SMP] Don't replace SMP instructions
Jeremy Fitzhardingeb7fb4af2007-05-02 19:27:13 +02002991 with UP alternatives
2992
H. Peter Anvin7a5091d2014-05-11 20:25:20 -07002993 nordrand [X86] Disable kernel use of the RDRAND and
2994 RDSEED instructions even if they are supported
2995 by the processor. RDRAND and RDSEED are still
2996 available to user space applications.
H. Peter Anvin49d859d2011-07-31 14:02:19 -07002997
Randy Dunlapa9913042005-10-23 12:57:11 -07002998 noresume [SWSUSP] Disables resume and restores original swap
2999 space.
3000
Linus Torvalds1da177e2005-04-16 15:20:36 -07003001 no-scroll [VGA] Disables scrollback.
3002 This is required for the Braillex ib80-piezo Braille
3003 reader made by F.H. Papenmeier (Germany).
3004
3005 nosbagart [IA-64]
3006
Alan Coxcd4f0ef2007-07-31 00:37:59 -07003007 nosep [BUGS=X86-32] Disables x86 SYSENTER/SYSEXIT support.
Chuck Ebbert4f886512006-03-23 02:59:34 -08003008
Len Brown61ec7562007-08-16 03:34:22 -04003009 nosmp [SMP] Tells an SMP kernel to act as a UP kernel,
3010 and disable the IO APIC. legacy for "maxcpus=0".
Linus Torvalds1da177e2005-04-16 15:20:36 -07003011
Dave Jones97842212007-07-15 23:41:05 -07003012 nosoftlockup [KNL] Disable the soft-lockup detector.
3013
Linus Torvalds1da177e2005-04-16 15:20:36 -07003014 nosync [HW,M68K] Disables sync negotiation for all devices.
3015
Alan Coxcd4f0ef2007-07-31 00:37:59 -07003016 notsc [BUGS=X86-32] Disable Time Stamp Counter
Linus Torvalds1da177e2005-04-16 15:20:36 -07003017
Ulrich Obergfell195daf62015-04-14 15:44:13 -07003018 nowatchdog [KNL] Disable both lockup detectors, i.e.
3019 soft-lockup and NMI watchdog (hard-lockup).
Don Zickus58687ac2010-05-07 17:11:44 -04003020
Linus Torvalds1da177e2005-04-16 15:20:36 -07003021 nowb [ARM]
Randy Dunlapa9913042005-10-23 12:57:11 -07003022
Weidong Han2b2fd872009-04-17 16:42:12 +08003023 nox2apic [X86-64,APIC] Do not enable x2APIC mode.
3024
Fenghua Yuf78cff42012-11-13 11:32:38 -08003025 cpu0_hotplug [X86] Turn on CPU0 hotplug feature when
3026 CONFIG_BOOTPARAM_HOTPLUG_CPU0 is off.
3027 Some features depend on CPU0. Known dependencies are:
3028 1. Resume from suspend/hibernate depends on CPU0.
3029 Suspend/hibernate will fail if CPU0 is offline and you
3030 need to online CPU0 before suspend/hibernate.
3031 2. PIC interrupts also depend on CPU0. CPU0 can't be
3032 removed if a PIC interrupt is detected.
3033 It's said poweroff/reboot may depend on CPU0 on some
3034 machines although I haven't seen such issues so far
3035 after CPU0 is offline on a few tested machines.
3036 If the dependencies are under your control, you can
3037 turn on cpu0_hotplug.
3038
Randy Dunlap16290242011-08-13 12:34:52 -07003039 nptcg= [IA-64] Override max number of concurrent global TLB
Fenghua Yua6c75b862008-03-14 13:57:08 -07003040 purges which is reported from either PAL_VM_SUMMARY or
3041 SAL PALO.
3042
Yinghai Lu2b633e32010-02-10 01:20:37 -08003043 nr_cpus= [SMP] Maximum number of processors that an SMP kernel
3044 could support. nr_cpus=n : n >= 1 limits the kernel to
Baoquan He7c142bf2016-08-24 13:06:45 +08003045 support 'n' processors. It could be larger than the
3046 number of already plugged CPU during bootup, later in
3047 runtime you can physically add extra cpu until it reaches
3048 n. So during boot up some boot time memory for per-cpu
3049 variables need be pre-allocated for later physical cpu
3050 hot plugging.
Yinghai Lu2b633e32010-02-10 01:20:37 -08003051
Randy Dunlap0cb55ad2009-04-05 15:55:22 -07003052 nr_uarts= [SERIAL] maximum number of UARTs to be registered.
3053
Mel Gorman1a687c22012-11-22 11:16:36 +00003054 numa_balancing= [KNL,X86] Enable or disable automatic NUMA balancing.
3055 Allowed values are enable and disable
3056
KAMEZAWA Hiroyukif0c0b2b2007-07-15 23:38:01 -07003057 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA.
3058 one of ['zone', 'node', 'default'] can be specified
3059 This can be set from sysctl after boot.
3060 See Documentation/sysctl/vm.txt for details.
3061
Randy Dunlap7c4be2532009-01-06 14:42:44 -08003062 ohci1394_dma=early [HW] enable debugging via the ohci1394 driver.
3063 See Documentation/debugging-via-ohci1394.txt for more
3064 info.
3065
Andres Salomon3ef0e1f2008-04-29 00:59:53 -07003066 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands
3067 Rather than timing out after 20 ms if an EC
3068 command is not properly ACKed, override the length
3069 of the timeout. We have interrupts disabled while
3070 waiting for the ACK, so if this is set too high
3071 interrupts *may* be lost!
3072
Tony Lindgren15ac7af2009-12-11 16:16:32 -08003073 omap_mux= [OMAP] Override bootloader pin multiplexing.
3074 Format: <mux_mode0.mode_name=value>...
3075 For example, to override I2C bus2:
3076 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100
3077
Linus Torvalds1da177e2005-04-16 15:20:36 -07003078 oprofile.timer= [HW]
3079 Use timer interrupt instead of performance counters
3080
Robert Richter7e4e0bd2009-05-06 12:10:23 +02003081 oprofile.cpu_type= Force an oprofile cpu type
3082 This might be useful if you have an older oprofile
3083 userland or if you want common events.
Robert Richter8d7ff4f2009-06-23 11:48:14 +02003084 Format: { arch_perfmon }
3085 arch_perfmon: [X86] Force use of architectural
Robert Richter7e4e0bd2009-05-06 12:10:23 +02003086 perfmon on Intel CPUs instead of the
3087 CPU specific event set.
Robert Richter159a80b2011-10-11 19:39:16 +02003088 timer: [X86] Force use of architectural NMI
3089 timer mode (see also oprofile.timer
3090 for generic hr timer mode)
Andi Kleen1dcdb5a2009-04-27 17:44:11 +02003091
Randy Dunlap44a4dcf2011-04-04 15:02:24 -07003092 oops=panic Always panic on oopses. Default is to just kill the
3093 process, but there is a small probability of
3094 deadlocking the machine.
Olaf Heringd404ab02011-03-22 16:34:04 -07003095 This will also cause panics on machine check exceptions.
3096 Useful together with panic=30 to trigger a reboot.
3097
Randy Dunlapbcfde332010-06-07 17:10:38 -07003098 OSS [HW,OSS]
3099 See Documentation/sound/oss/oss-parameters.txt
3100
Joonsoo Kim48c96a32014-12-12 16:56:01 -08003101 page_owner= [KNL] Boot-time page_owner enabling option.
3102 Storage of the information about who allocated
3103 each page is disabled in default. With this switch,
3104 we can turn it on.
3105 on: enable the feature
3106
Laura Abbott8823b1d2016-03-15 14:56:27 -07003107 page_poison= [KNL] Boot-time parameter changing the state of
3108 poisoning on the buddy allocator.
3109 off: turn off poisoning
3110 on: turn on poisoning
3111
Randy Dunlap44a4dcf2011-04-04 15:02:24 -07003112 panic= [KNL] Kernel behaviour on panic: delay <timeout>
Hugh Dickins4302fbc2011-07-26 16:08:52 -07003113 timeout > 0: seconds before rebooting
3114 timeout = 0: wait forever
3115 timeout < 0: reboot immediately
Linus Torvalds1da177e2005-04-16 15:20:36 -07003116 Format: <timeout>
3117
Prarit Bhargava9e3961a2014-12-10 15:45:50 -08003118 panic_on_warn panic() instead of WARN(). Useful to cause kdump
3119 on a WARN().
3120
Masami Hiramatsuf06e5152014-06-06 14:37:07 -07003121 crash_kexec_post_notifiers
3122 Run kdump after running panic-notifiers and dumping
3123 kmsg. This only for the users who doubt kdump always
3124 succeeds in any situation.
3125 Note that this also increases risks of kdump failure,
3126 because some panic notifiers can make the crashed
3127 kernel more unstable.
3128
Linus Torvalds1da177e2005-04-16 15:20:36 -07003129 parkbd.port= [HW] Parallel port number the keyboard adapter is
3130 connected to, default is 0.
3131 Format: <parport#>
3132 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation,
3133 0 for XT, 1 for AT (default is AT).
Randy Dunlapa9913042005-10-23 12:57:11 -07003134 Format: <mode>
Linus Torvalds1da177e2005-04-16 15:20:36 -07003135
Randy Dunlapa9913042005-10-23 12:57:11 -07003136 parport= [HW,PPT] Specify parallel ports. 0 disables.
3137 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] }
3138 Use 'auto' to force the driver to use any
3139 IRQ/DMA settings detected (the default is to
3140 ignore detected IRQ/DMA settings because of
3141 possible conflicts). You can specify the base
3142 address, IRQ, and DMA settings; IRQ and DMA
3143 should be numbers, or 'auto' (for using detected
3144 settings on that particular port), or 'nofifo'
3145 (to avoid using a FIFO even if it is detected).
3146 Parallel ports are assigned in the order they
3147 are specified on the command line, starting
3148 with parport0.
Linus Torvalds1da177e2005-04-16 15:20:36 -07003149
Randy Dunlapa9913042005-10-23 12:57:11 -07003150 parport_init_mode= [HW,PPT]
3151 Configure VIA parallel port to operate in
3152 a specific mode. This is necessary on Pegasos
3153 computer where firmware has no options for setting
3154 up parallel port mode and sets it to spp.
3155 Currently this function knows 686a and 8231 chips.
Linus Torvalds1da177e2005-04-16 15:20:36 -07003156 Format: [spp|ps2|epp|ecp|ecpepp]
3157
Andrew Mortondd287792006-03-23 03:00:57 -08003158 pause_on_oops=
3159 Halt all CPUs after the first oops has been printed for
3160 the specified number of seconds. This is to be used if
3161 your oopses keep scrolling off the screen.
3162
Linus Torvalds1da177e2005-04-16 15:20:36 -07003163 pcbit= [HW,ISDN]
3164
3165 pcd. [PARIDE]
3166 See header of drivers/block/paride/pcd.c.
Randy Dunlap31c00fc2008-11-13 21:33:24 +00003167 See also Documentation/blockdev/paride.txt.
Linus Torvalds1da177e2005-04-16 15:20:36 -07003168
Randy Dunlapa9913042005-10-23 12:57:11 -07003169 pci=option[,option...] [PCI] various PCI subsystem options:
Bjorn Helgaas1cc0ca22009-01-14 10:04:36 -07003170 earlydump [X86] dump PCI config space before the kernel
3171 changes anything
Andi Kleenc0115602008-08-22 09:53:39 +02003172 off [X86] don't probe for the PCI bus
Alan Coxcd4f0ef2007-07-31 00:37:59 -07003173 bios [X86-32] force use of PCI BIOS, don't access
Randy Dunlapa9913042005-10-23 12:57:11 -07003174 the hardware directly. Use this if your machine
3175 has a non-standard PCI host bridge.
Alan Coxcd4f0ef2007-07-31 00:37:59 -07003176 nobios [X86-32] disallow use of PCI BIOS, only direct
Randy Dunlapa9913042005-10-23 12:57:11 -07003177 hardware access methods are allowed. Use this
3178 if you experience crashes upon bootup and you
3179 suspect they are caused by the BIOS.
Borislav Petkovafd8c082016-01-13 16:48:51 +01003180 conf1 [X86] Force use of PCI Configuration Access
3181 Mechanism 1 (config address in IO port 0xCF8,
3182 data in IO port 0xCFC, both 32-bit).
3183 conf2 [X86] Force use of PCI Configuration Access
3184 Mechanism 2 (IO port 0xCF8 is an 8-bit port for
3185 the function, IO port 0xCFA, also 8-bit, sets
3186 bus number. The config space is then accessed
3187 through ports 0xC000-0xCFFF).
3188 See http://wiki.osdev.org/PCI for more info
3189 on the configuration access mechanisms.
Randy Dunlap7f785762007-10-05 13:17:58 -07003190 noaer [PCIE] If the PCIEAER kernel config parameter is
3191 enabled, this kernel boot option can be used to
3192 disable the use of PCIE advanced error reporting.
Jeff Garzik32a2eea2007-10-11 16:57:27 -04003193 nodomains [PCI] Disable support for multiple PCI
3194 root domains (aka PCI segments, in ACPI-speak).
Jaswinder Singh Rajput6cececf2009-04-14 14:03:43 +05303195 nommconf [X86] Disable use of MMCONFIG for PCI
Bjorn Helgaas61be6d62006-02-15 15:17:43 -08003196 Configuration
Andreas Herrmann12983072009-06-07 16:15:16 +02003197 check_enable_amd_mmconf [X86] check for and enable
3198 properly configured MMIO access to PCI
3199 config space on AMD family 10h CPU
Matthew Wilcox309e57d2006-03-05 22:33:34 -07003200 nomsi [MSI] If the PCI_MSI kernel config parameter is
3201 enabled, this kernel boot option can be used to
3202 disable the use of MSI interrupts system-wide.
Stefan Assmanna9322f62008-06-11 16:35:14 +02003203 noioapicquirk [APIC] Disable all boot interrupt quirks.
3204 Safety option to keep boot IRQs enabled. This
3205 should never be necessary.
Stefan Assmann91979792008-06-11 16:35:15 +02003206 ioapicreroute [APIC] Enable rerouting of boot IRQs to the
3207 primary IO-APIC for bridges that cannot disable
3208 boot IRQs. This fixes a source of spurious IRQs
3209 when the system masks IRQs.
Stefan Assmann41b9eb22008-07-15 13:48:55 +02003210 noioapicreroute [APIC] Disable workaround that uses the
3211 boot IRQ equivalent of an IRQ that connects to
3212 a chipset where boot IRQs cannot be disabled.
3213 The opposite of ioapicreroute.
Alan Coxcd4f0ef2007-07-31 00:37:59 -07003214 biosirq [X86-32] Use PCI BIOS calls to get the interrupt
Randy Dunlapa9913042005-10-23 12:57:11 -07003215 routing table. These calls are known to be buggy
3216 on several machines and they hang the machine
3217 when used, but on other computers it's the only
3218 way to get the interrupt routing table. Try
3219 this option if the kernel is unable to allocate
3220 IRQs or discover secondary PCI buses on your
3221 motherboard.
Andi Kleenc0115602008-08-22 09:53:39 +02003222 rom [X86] Assign address space to expansion ROMs.
Randy Dunlapa9913042005-10-23 12:57:11 -07003223 Use with caution as certain devices share
3224 address decoders between ROMs and other
3225 resources.
Andi Kleenc0115602008-08-22 09:53:39 +02003226 norom [X86] Do not assign address space to
Gary Hadebb71ad82008-05-12 13:57:46 -07003227 expansion ROMs that do not already have
3228 BIOS assigned address ranges.
Mike Habeck7bd1c362010-05-12 11:14:32 -07003229 nobar [X86] Do not assign address space to the
3230 BARs that weren't assigned by the BIOS.
Andi Kleenc0115602008-08-22 09:53:39 +02003231 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be
Randy Dunlapa9913042005-10-23 12:57:11 -07003232 assigned automatically to PCI devices. You can
3233 make the kernel exclude IRQs of your ISA cards
3234 this way.
Andi Kleenc0115602008-08-22 09:53:39 +02003235 pirqaddr=0xAAAAA [X86] Specify the physical address
Randy Dunlapa9913042005-10-23 12:57:11 -07003236 of the PIRQ table (normally generated
3237 by the BIOS) if it is outside the
3238 F0000h-100000h range.
Andi Kleenc0115602008-08-22 09:53:39 +02003239 lastbus=N [X86] Scan all buses thru bus #N. Can be
Randy Dunlapa9913042005-10-23 12:57:11 -07003240 useful if the kernel is unable to find your
3241 secondary buses and you want to tell it
3242 explicitly which ones they are.
Andi Kleenc0115602008-08-22 09:53:39 +02003243 assign-busses [X86] Always assign all PCI bus
Randy Dunlapa9913042005-10-23 12:57:11 -07003244 numbers ourselves, overriding
3245 whatever the firmware may have done.
Andi Kleenc0115602008-08-22 09:53:39 +02003246 usepirqmask [X86] Honor the possible IRQ mask stored
Randy Dunlapa9913042005-10-23 12:57:11 -07003247 in the BIOS $PIR table. This is needed on
3248 some systems with broken BIOSes, notably
3249 some HP Pavilion N5400 and Omnibook XE3
3250 notebooks. This will have no effect if ACPI
3251 IRQ routing is enabled.
Andi Kleenc0115602008-08-22 09:53:39 +02003252 noacpi [X86] Do not use ACPI for IRQ routing
Randy Dunlapa9913042005-10-23 12:57:11 -07003253 or for PCI scanning.
Bjorn Helgaas7bc5e3f2010-02-23 10:24:41 -07003254 use_crs [X86] Use PCI host bridge window information
3255 from ACPI. On BIOSes from 2008 or later, this
3256 is enabled by default. If you need to use this,
3257 please report a bug.
3258 nocrs [X86] Ignore PCI host bridge windows from ACPI.
3259 If you need to use this, please report a bug.
Randy Dunlapa9913042005-10-23 12:57:11 -07003260 routeirq Do IRQ routing for all PCI devices.
3261 This is normally done in pci_enable_device(),
3262 so this option is a temporary workaround
3263 for broken drivers that don't call it.
Yinghai Lu13a6ddb2008-03-27 01:31:18 -07003264 skip_isa_align [X86] do not align io start addr, so can
3265 handle more pci cards
Andi Kleen0637a702006-09-26 10:52:41 +02003266 noearly [X86] Don't do any early type 1 scanning.
3267 This might help on some broken boards which
3268 machine check when some devices' config space
3269 is read. But various workarounds are disabled
3270 and some IOMMU drivers will not work.
Matt Domsch6b4b78f2006-09-29 15:23:23 -05003271 bfsort Sort PCI devices into breadth-first order.
3272 This sorting is done to get a device
3273 order compatible with older (<= 2.4) kernels.
3274 nobfsort Don't sort PCI devices into breadth-first order.
Yijing Wangfa238712013-01-30 09:40:52 +08003275 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size)
3276 tuning and use the BIOS-configured MPS defaults.
3277 pcie_bus_safe Set every device's MPS to the largest value
3278 supported by all devices below the root complex.
3279 pcie_bus_perf Set device MPS to the largest allowable MPS
3280 based on its parent bus. Also set MRRS (Max
3281 Read Request Size) to the largest supported
3282 value (no larger than the MPS that the device
3283 or bus can support) for best performance.
3284 pcie_bus_peer2peer Set every device's MPS to 128B, which
3285 every device is guaranteed to support. This
3286 configuration allows peer-to-peer DMA between
3287 any pair of devices, possibly at the cost of
3288 reduced performance. This also guarantees
3289 that hot-added devices will work.
Atsushi Nemoto4516a612007-02-05 16:36:06 -08003290 cbiosize=nn[KMG] The fixed amount of bus space which is
3291 reserved for the CardBus bridge's IO window.
3292 The default value is 256 bytes.
3293 cbmemsize=nn[KMG] The fixed amount of bus space which is
3294 reserved for the CardBus bridge's memory
3295 window. The default value is 64 megabytes.
Yuji Shimada32a9a6822009-03-16 17:13:39 +09003296 resource_alignment=
3297 Format:
3298 [<order of align>@][<domain>:]<bus>:<slot>.<func>[; ...]
Koehrer Mathias (ETAS/ESW5)644a5442016-06-07 14:24:17 +00003299 [<order of align>@]pci:<vendor>:<device>\
3300 [:<subvendor>:<subdevice>][; ...]
Yuji Shimada32a9a6822009-03-16 17:13:39 +09003301 Specifies alignment and device to reassign
3302 aligned memory resources.
3303 If <order of align> is not specified,
3304 PAGE_SIZE is used as alignment.
3305 PCI-PCI bridge can be specified, if resource
3306 windows need to be expanded.
Mathias Koehrer8b078c62016-08-09 10:33:31 +02003307 To specify the alignment for several
3308 instances of a device, the PCI vendor,
3309 device, subvendor, and subdevice may be
3310 specified, e.g., 4096@pci:8086:9c22:103c:198f
Andrew Patterson43c16402009-04-22 16:52:09 -06003311 ecrc= Enable/disable PCIe ECRC (transaction layer
3312 end-to-end CRC checking).
3313 bios: Use BIOS/firmware settings. This is the
3314 the default.
3315 off: Turn ECRC off
3316 on: Turn ECRC on.
Yijing Wang8c8803c2013-01-23 20:29:06 +08003317 hpiosize=nn[KMG] The fixed amount of bus space which is
3318 reserved for hotplug bridge's IO window.
3319 Default size is 256 bytes.
3320 hpmemsize=nn[KMG] The fixed amount of bus space which is
3321 reserved for hotplug bridge's memory window.
3322 Default size is 2 megabytes.
Keith Busche16b4662016-07-21 21:40:28 -06003323 hpbussize=nn The minimum amount of additional bus numbers
3324 reserved for buses below a hotplug bridge.
3325 Default is 1.
Yinghai Lub55438f2012-02-23 19:23:30 -08003326 realloc= Enable/disable reallocating PCI bridge resources
3327 if allocations done by BIOS are too small to
3328 accommodate resources required by all child
3329 devices.
3330 off: Turn realloc off
3331 on: Turn realloc on
3332 realloc same as realloc=on
Rafael J. Wysocki6748dcc2012-03-01 00:06:33 +01003333 noari do not use PCIe ARI.
Bjorn Helgaas284f5f92012-04-30 15:21:02 -06003334 pcie_scan_all Scan all possible PCIe devices. Otherwise we
3335 only look for one device below a PCIe downstream
3336 port.
Matt Domsch6b4b78f2006-09-29 15:23:23 -05003337
Chuck Ebberte5665a42008-09-24 20:40:34 -04003338 pcie_aspm= [PCIE] Forcibly enable or disable PCIe Active State Power
3339 Management.
3340 off Disable ASPM.
3341 force Enable ASPM even on devices that claim not to support it.
3342 WARNING: Forcing ASPM on may cause system lockups.
3343
MUNEDA Takahiro7570a332012-02-02 11:09:22 -05003344 pcie_hp= [PCIE] PCI Express Hotplug driver options:
3345 nomsi Do not use MSI for PCI Express Native Hotplug (this
3346 makes all PCIe ports use INTx for hotplug services).
3347
Rafael J. Wysocki79dd9182010-08-21 01:51:44 +02003348 pcie_ports= [PCIE] PCIe ports handling:
Rafael J. Wysocki28eb5f22010-08-21 22:02:38 +02003349 auto Ask the BIOS whether or not to use native PCIe services
3350 associated with PCIe ports (PME, hot-plug, AER). Use
3351 them only if that is allowed by the BIOS.
3352 native Use native PCIe services associated with PCIe ports
3353 unconditionally.
Rafael J. Wysocki79dd9182010-08-21 01:51:44 +02003354 compat Treat PCIe ports as PCI-to-PCI bridges, disable the PCIe
3355 ports driver.
3356
Mika Westerberg9d26d3a2016-06-02 11:17:12 +03003357 pcie_port_pm= [PCIE] PCIe port power management handling:
3358 off Disable power management of all PCIe ports
3359 force Forcibly enable power management of all PCIe ports
3360
Rafael J. Wysockic7f48652010-02-17 23:39:08 +01003361 pcie_pme= [PCIE,PM] Native PCIe PME signaling options:
Rafael J. Wysockic39fae12010-02-17 23:40:07 +01003362 nomsi Do not use MSI for native PCIe PME signaling (this makes
Rafael J. Wysocki28eb5f22010-08-21 22:02:38 +02003363 all PCIe root ports use INTx for all services).
Rafael J. Wysockic7f48652010-02-17 23:39:08 +01003364
Linus Torvalds1da177e2005-04-16 15:20:36 -07003365 pcmv= [HW,PCMCIA] BadgePAD 4
3366
Tushar Behera39ac5ba2014-03-28 10:50:21 +05303367 pd_ignore_unused
3368 [PM]
3369 Keep all power-domains already enabled by bootloader on,
3370 even if no driver has claimed them. This is useful
3371 for debug and development, but should not be
3372 needed on a platform with proper driver support.
3373
Linus Torvalds1da177e2005-04-16 15:20:36 -07003374 pd. [PARIDE]
Randy Dunlap31c00fc2008-11-13 21:33:24 +00003375 See Documentation/blockdev/paride.txt.
Linus Torvalds1da177e2005-04-16 15:20:36 -07003376
3377 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at
3378 boot time.
3379 Format: { 0 | 1 }
3380 See arch/parisc/kernel/pdc_chassis.c
3381
Tejun Heof58dc012009-08-14 15:00:50 +09003382 percpu_alloc= Select which percpu first chunk allocator to use.
Tejun Heoe933a732009-08-14 15:00:53 +09003383 Currently supported values are "embed" and "page".
3384 Archs may support subset or none of the selections.
3385 See comments in mm/percpu.c for details on each
3386 allocator. This parameter is primarily for debugging
3387 and performance comparison.
Tejun Heofa8a7092009-06-22 11:56:24 +09003388
Linus Torvalds1da177e2005-04-16 15:20:36 -07003389 pf. [PARIDE]
Randy Dunlap31c00fc2008-11-13 21:33:24 +00003390 See Documentation/blockdev/paride.txt.
Linus Torvalds1da177e2005-04-16 15:20:36 -07003391
3392 pg. [PARIDE]
Randy Dunlap31c00fc2008-11-13 21:33:24 +00003393 See Documentation/blockdev/paride.txt.
Linus Torvalds1da177e2005-04-16 15:20:36 -07003394
3395 pirq= [SMP,APIC] Manual mp-table setup
Uwe Hermann71cced62008-10-20 09:32:21 -07003396 See Documentation/x86/i386/IO-APIC.txt.
Linus Torvalds1da177e2005-04-16 15:20:36 -07003397
3398 plip= [PPT,NET] Parallel port network link
3399 Format: { parport<nr> | timid | 0 }
3400 See also Documentation/parport.txt.
3401
Randy Dunlap16290242011-08-13 12:34:52 -07003402 pmtmr= [X86] Manual setup of pmtmr I/O Port.
Thomas Gleixnerde32a242008-07-12 05:33:30 +02003403 Override pmtimer IOPort with a hex value.
3404 e.g. pmtmr=0x508
3405
Bjorn Helgaas96242112011-08-11 12:14:05 -06003406 pnp.debug=1 [PNP]
3407 Enable PNP debug messages (depends on the
3408 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time
3409 via /sys/module/pnp/parameters/debug. We always show
3410 current resource usage; turning this on also shows
3411 possible settings and some assignment information.
Bjorn Helgaas97ef0622008-08-19 16:53:41 -06003412
Linus Torvalds1da177e2005-04-16 15:20:36 -07003413 pnpacpi= [ACPI]
3414 { off }
3415
3416 pnpbios= [ISAPNP]
3417 { on | off | curr | res | no-curr | no-res }
3418
3419 pnp_reserve_irq=
3420 [ISAPNP] Exclude IRQs for the autoconfiguration
3421
3422 pnp_reserve_dma=
3423 [ISAPNP] Exclude DMAs for the autoconfiguration
3424
3425 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration
Randy Dunlapa9913042005-10-23 12:57:11 -07003426 Ranges are in pairs (I/O port base and size).
Linus Torvalds1da177e2005-04-16 15:20:36 -07003427
3428 pnp_reserve_mem=
Randy Dunlapa9913042005-10-23 12:57:11 -07003429 [ISAPNP] Exclude memory regions for the
3430 autoconfiguration.
Linus Torvalds1da177e2005-04-16 15:20:36 -07003431 Ranges are in pairs (memory base and size).
3432
Randy Dunlap4af94f32009-04-17 18:30:28 -07003433 ports= [IP_VS_FTP] IPVS ftp helper module
3434 Default is 21.
3435 Up to 8 (IP_VS_APP_MAX_PORTS) ports
3436 may be specified.
3437 Format: <port>,<port>....
3438
Anton Blanchard3eb5d582015-10-29 11:44:06 +11003439 ppc_strict_facility_enable
3440 [PPC] This option catches any kernel floating point,
3441 Altivec, VSX and SPE outside of regions specifically
3442 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()).
3443 There is some performance impact when enabling this.
3444
Ingo Molnar45807a12007-07-15 23:40:10 -07003445 print-fatal-signals=
3446 [KNL] debug: print fatal signals
Naohiro Ooiwaf84d49b2009-11-09 00:46:42 +09003447
3448 If enabled, warn about various signal handling
3449 related application anomalies: too many signals,
3450 too many POSIX.1 timers, fatal signals causing a
3451 coredump - etc.
3452
3453 If you hit the warning due to signal overflow,
3454 you might want to try "ulimit -i unlimited".
3455
Ingo Molnar45807a12007-07-15 23:40:10 -07003456 default: off.
3457
Matthew Garrettc22ab332012-03-05 14:59:10 -08003458 printk.always_kmsg_dump=
3459 Trigger kmsg_dump for cases other than kernel oops or
3460 panics
3461 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
3462 default: disabled
3463
Borislav Petkov750afe72016-08-02 14:04:07 -07003464 printk.devkmsg={on,off,ratelimit}
3465 Control writing to /dev/kmsg.
3466 on - unlimited logging to /dev/kmsg from userspace
3467 off - logging to /dev/kmsg disabled
3468 ratelimit - ratelimit the logging
3469 Default: ratelimit
3470
Randy Dunlape84845c2007-07-15 23:40:25 -07003471 printk.time= Show timing data prefixed to each printk message line
3472 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
3473
Randy Dunlap0cb55ad2009-04-05 15:55:22 -07003474 processor.max_cstate= [HW,ACPI]
3475 Limit processor to maximum C-state
3476 max_cstate=9 overrides any DMI blacklist limit.
3477
3478 processor.nocst [HW,ACPI]
3479 Ignore the _CST method to determine C-states,
3480 instead using the legacy FADT method
3481
Linus Torvalds1da177e2005-04-16 15:20:36 -07003482 profile= [KNL] Enable kernel profiling via /proc/profile
Randy Dunlapa9913042005-10-23 12:57:11 -07003483 Format: [schedule,]<number>
3484 Param: "schedule" - profile schedule points.
3485 Param: <number> - step/bucket size as a power of 2 for
3486 statistical time based profiling.
Mel Gormanb3da2a72007-10-24 18:23:50 +02003487 Param: "sleep" - profile D-state sleeping (millisecs).
3488 Requires CONFIG_SCHEDSTATS
Dave Jonesc0fe2e62007-10-20 03:08:22 +02003489 Param: "kvm" - profile VM exits.
Linus Torvalds1da177e2005-04-16 15:20:36 -07003490
Linus Torvalds1da177e2005-04-16 15:20:36 -07003491 prompt_ramdisk= [RAM] List of RAM disks to prompt for floppy disk
3492 before loading.
Randy Dunlap31c00fc2008-11-13 21:33:24 +00003493 See Documentation/blockdev/ramdisk.txt.
Linus Torvalds1da177e2005-04-16 15:20:36 -07003494
Johannes Weinerc9f51ce2018-11-30 14:09:58 -08003495 psi= [KNL] Enable or disable pressure stall information
3496 tracking.
3497 Format: <bool>
3498
Randy Dunlapa9913042005-10-23 12:57:11 -07003499 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to
3500 probe for; one of (bare|imps|exps|lifebook|any).
Linus Torvalds1da177e2005-04-16 15:20:36 -07003501 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports
3502 per second.
Randy Dunlapa9913042005-10-23 12:57:11 -07003503 psmouse.resetafter= [HW,MOUSE]
3504 Try to reset the device after so many bad packets
Linus Torvalds1da177e2005-04-16 15:20:36 -07003505 (0 = never).
3506 psmouse.resolution=
3507 [HW,MOUSE] Set desired mouse resolution, in dpi.
3508 psmouse.smartscroll=
Randy Dunlapa9913042005-10-23 12:57:11 -07003509 [HW,MOUSE] Controls Logitech smartscroll autorepeat.
Linus Torvalds1da177e2005-04-16 15:20:36 -07003510 0 = disabled, 1 = enabled (default).
3511
Matthew Garrettdee28e72011-07-21 16:57:55 -04003512 pstore.backend= Specify the name of the pstore backend to use
3513
Linus Torvalds1da177e2005-04-16 15:20:36 -07003514 pt. [PARIDE]
Randy Dunlap31c00fc2008-11-13 21:33:24 +00003515 See Documentation/blockdev/paride.txt.
Linus Torvalds1da177e2005-04-16 15:20:36 -07003516
Dave Hansen4e6c2af2018-01-05 09:44:36 -08003517 pti= [X86_64] Control Page Table Isolation of user and
3518 kernel address spaces. Disabling this feature
3519 removes hardening, but improves performance of
3520 system calls and interrupts.
3521
3522 on - unconditionally enable
3523 off - unconditionally disable
3524 auto - kernel detects whether your CPU model is
3525 vulnerable to issues that PTI mitigates
3526
3527 Not specifying this option is equivalent to pti=auto.
3528
3529 nopti [X86_64]
3530 Equivalent to pti=off
Borislav Petkov80183072018-01-02 14:19:48 +01003531
Kay Sieversdc8c8582007-08-15 12:25:38 +02003532 pty.legacy_count=
3533 [KNL] Number of legacy pty's. Overwrites compiled-in
3534 default number.
3535
Randy Dunlap7d2c5022006-09-29 02:01:02 -07003536 quiet [KNL] Disable most log messages
Randy Dunlapa9913042005-10-23 12:57:11 -07003537
Linus Torvalds1da177e2005-04-16 15:20:36 -07003538 r128= [HW,DRM]
3539
3540 raid= [HW,RAID]
3541 See Documentation/md.txt.
3542
Linus Torvalds1da177e2005-04-16 15:20:36 -07003543 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes
Randy Dunlap31c00fc2008-11-13 21:33:24 +00003544 See Documentation/blockdev/ramdisk.txt.
Linus Torvalds1da177e2005-04-16 15:20:36 -07003545
Paul E. McKenney4102ada2013-10-08 20:23:47 -07003546 rcu_nocbs= [KNL]
Noam Camus2d13e6c2016-10-11 13:51:35 -07003547 The argument is a cpu list, as described above.
3548
Paul E. McKenney3fbfbf72012-08-19 21:35:53 -07003549 In kernels built with CONFIG_RCU_NOCB_CPU=y, set
3550 the specified list of CPUs to be no-callback CPUs.
3551 Invocation of these CPUs' RCU callbacks will
Paul E. McKenneya4889852012-12-03 08:16:28 -08003552 be offloaded to "rcuox/N" kthreads created for
3553 that purpose, where "x" is "b" for RCU-bh, "p"
3554 for RCU-preempt, and "s" for RCU-sched, and "N"
3555 is the CPU number. This reduces OS jitter on the
Paul E. McKenney3fbfbf72012-08-19 21:35:53 -07003556 offloaded CPUs, which can be useful for HPC and
3557 real-time workloads. It can also improve energy
3558 efficiency for asymmetric multiprocessors.
3559
Paul E. McKenney4102ada2013-10-08 20:23:47 -07003560 rcu_nocb_poll [KNL]
Paul E. McKenney3fbfbf72012-08-19 21:35:53 -07003561 Rather than requiring that offloaded CPUs
3562 (specified by rcu_nocbs= above) explicitly
3563 awaken the corresponding "rcuoN" kthreads,
3564 make these kthreads poll for callbacks.
3565 This improves the real-time response for the
3566 offloaded CPUs by relieving them of the need to
3567 wake up the corresponding kthread, but degrades
3568 energy efficiency by requiring that the kthreads
3569 periodically wake up to do the polling.
3570
Paul E. McKenney4102ada2013-10-08 20:23:47 -07003571 rcutree.blimit= [KNL]
Paul E. McKenney97e63f02013-10-27 09:44:03 -07003572 Set maximum number of finished RCU callbacks to
3573 process in one batch.
Dipankar Sarma21a1ea92006-03-07 21:55:33 -08003574
Paul E. McKenneya3dc2942015-04-20 11:40:50 -07003575 rcutree.dump_tree= [KNL]
3576 Dump the structure of the rcu_node combining tree
3577 out at early boot. This is used for diagnostic
3578 purposes, to verify correct tree setup.
3579
Paul E. McKenney0f41c0d2015-03-10 18:33:20 -07003580 rcutree.gp_cleanup_delay= [KNL]
3581 Set the number of jiffies to delay each step of
3582 RCU grace-period cleanup. This only has effect
3583 when CONFIG_RCU_TORTURE_TEST_SLOW_CLEANUP is set.
3584
Paul E. McKenney37745d22015-01-22 18:24:08 -08003585 rcutree.gp_init_delay= [KNL]
3586 Set the number of jiffies to delay each step of
3587 RCU grace-period initialization. This only has
Paul E. McKenney0f41c0d2015-03-10 18:33:20 -07003588 effect when CONFIG_RCU_TORTURE_TEST_SLOW_INIT
3589 is set.
3590
3591 rcutree.gp_preinit_delay= [KNL]
3592 Set the number of jiffies to delay each step of
3593 RCU grace-period pre-initialization, that is,
3594 the propagation of recent CPU-hotplug changes up
3595 the rcu_node combining tree. This only has effect
3596 when CONFIG_RCU_TORTURE_TEST_SLOW_PREINIT is set.
Paul E. McKenney37745d22015-01-22 18:24:08 -08003597
Paul E. McKenney7fa27002015-04-20 10:27:15 -07003598 rcutree.rcu_fanout_exact= [KNL]
3599 Disable autobalancing of the rcu_node combining
3600 tree. This is used by rcutorture, and might
3601 possibly be useful for architectures having high
3602 cache-to-cache transfer latencies.
Dipankar Sarma21a1ea92006-03-07 21:55:33 -08003603
Paul E. McKenney4102ada2013-10-08 20:23:47 -07003604 rcutree.rcu_fanout_leaf= [KNL]
Paul E. McKenneyee968ac2015-07-31 08:28:35 -07003605 Change the number of CPUs assigned to each
3606 leaf rcu_node structure. Useful for very
3607 large systems, which will choose the value 64,
3608 and for NUMA systems with large remote-access
3609 latencies, which will choose a value aligned
3610 with the appropriate hardware boundaries.
Paul E. McKenneyf885b7f2012-04-23 15:52:53 -07003611
Paul E. McKenney4a81e832014-06-20 16:49:01 -07003612 rcutree.jiffies_till_sched_qs= [KNL]
3613 Set required age in jiffies for a
3614 given grace period before RCU starts
3615 soliciting quiescent-state help from
3616 rcu_note_context_switch().
3617
Paul E. McKenney4102ada2013-10-08 20:23:47 -07003618 rcutree.jiffies_till_first_fqs= [KNL]
Paul E. McKenneyc0f4dfd2012-12-28 11:30:36 -08003619 Set delay from grace-period initialization to
3620 first attempt to force quiescent states.
3621 Units are jiffies, minimum value is zero,
3622 and maximum value is HZ.
3623
Paul E. McKenney4102ada2013-10-08 20:23:47 -07003624 rcutree.jiffies_till_next_fqs= [KNL]
Paul E. McKenneyc0f4dfd2012-12-28 11:30:36 -08003625 Set delay between subsequent attempts to force
3626 quiescent states. Units are jiffies, minimum
3627 value is one, and maximum value is HZ.
3628
Clark Williams21871d72014-09-12 21:21:09 -05003629 rcutree.kthread_prio= [KNL,BOOT]
Paul E. McKenneyd2af1ad2015-01-20 23:54:59 -08003630 Set the SCHED_FIFO priority of the RCU per-CPU
3631 kthreads (rcuc/N). This value is also used for
3632 the priority of the RCU boost threads (rcub/N)
3633 and for the RCU grace-period kthreads (rcu_bh,
3634 rcu_preempt, and rcu_sched). If RCU_BOOST is
3635 set, valid values are 1-99 and the default is 1
3636 (the least-favored priority). Otherwise, when
3637 RCU_BOOST is not set, valid values are 0-99 and
3638 the default is zero (non-realtime operation).
Clark Williams21871d72014-09-12 21:21:09 -05003639
Paul E. McKenneyfbce7492014-06-24 09:26:11 -07003640 rcutree.rcu_nocb_leader_stride= [KNL]
3641 Set the number of NOCB kthread groups, which
3642 defaults to the square root of the number of
3643 CPUs. Larger numbers reduces the wakeup overhead
3644 on the per-CPU grace-period kthreads, but increases
3645 that same overhead on each group's leader.
3646
Paul E. McKenney4102ada2013-10-08 20:23:47 -07003647 rcutree.qhimark= [KNL]
Paul E. McKenney97e63f02013-10-27 09:44:03 -07003648 Set threshold of queued RCU callbacks beyond which
3649 batch limiting is disabled.
Dipankar Sarma21a1ea92006-03-07 21:55:33 -08003650
Paul E. McKenney4102ada2013-10-08 20:23:47 -07003651 rcutree.qlowmark= [KNL]
Robert P. J. Day24aaef82008-02-03 15:20:26 +02003652 Set threshold of queued RCU callbacks below which
3653 batch limiting is re-enabled.
Dipankar Sarma21a1ea92006-03-07 21:55:33 -08003654
Paul E. McKenney4102ada2013-10-08 20:23:47 -07003655 rcutree.rcu_idle_gp_delay= [KNL]
Paul E. McKenneyc0f4dfd2012-12-28 11:30:36 -08003656 Set wakeup interval for idle CPUs that have
3657 RCU callbacks (RCU_FAST_NO_HZ=y).
Paul E. McKenneyd40011f2012-06-26 20:45:57 -07003658
Paul E. McKenney4102ada2013-10-08 20:23:47 -07003659 rcutree.rcu_idle_lazy_gp_delay= [KNL]
Paul E. McKenneyc0f4dfd2012-12-28 11:30:36 -08003660 Set wakeup interval for idle CPUs that have
3661 only "lazy" RCU callbacks (RCU_FAST_NO_HZ=y).
3662 Lazy RCU callbacks are those which RCU can
3663 prove do nothing more than free memory.
Paul E. McKenneyd40011f2012-06-26 20:45:57 -07003664
Paul E. McKenneybdea9e32016-01-01 13:47:19 -08003665 rcuperf.gp_exp= [KNL]
3666 Measure performance of expedited synchronous
3667 grace-period primitives.
3668
Paul E. McKenneydf37e662016-01-30 20:56:38 -08003669 rcuperf.holdoff= [KNL]
3670 Set test-start holdoff period. The purpose of
3671 this parameter is to delay the start of the
3672 test until boot completes in order to avoid
3673 interference.
3674
Paul E. McKenneybdea9e32016-01-01 13:47:19 -08003675 rcuperf.nreaders= [KNL]
3676 Set number of RCU readers. The value -1 selects
3677 N, where N is the number of CPUs. A value
3678 "n" less than -1 selects N-n+1, where N is again
3679 the number of CPUs. For example, -2 selects N
3680 (the number of CPUs), -3 selects N+1, and so on.
3681 A value of "n" less than or equal to -N selects
3682 a single reader.
3683
3684 rcuperf.nwriters= [KNL]
3685 Set number of RCU writers. The values operate
3686 the same as for rcuperf.nreaders.
3687 N, where N is the number of CPUs
3688
3689 rcuperf.perf_runnable= [BOOT]
3690 Start rcuperf running at boot time.
3691
3692 rcuperf.shutdown= [KNL]
3693 Shut the system down after performance tests
3694 complete. This is useful for hands-off automated
3695 testing.
3696
3697 rcuperf.perf_type= [KNL]
3698 Specify the RCU implementation to test.
3699
3700 rcuperf.verbose= [KNL]
3701 Enable additional printk() statements.
3702
Paul E. McKenney38706bc2014-08-18 21:12:17 -07003703 rcutorture.cbflood_inter_holdoff= [KNL]
3704 Set holdoff time (jiffies) between successive
3705 callback-flood tests.
3706
3707 rcutorture.cbflood_intra_holdoff= [KNL]
3708 Set holdoff time (jiffies) between successive
3709 bursts of callbacks within a given callback-flood
3710 test.
3711
3712 rcutorture.cbflood_n_burst= [KNL]
3713 Set the number of bursts making up a given
3714 callback-flood test. Set this to zero to
3715 disable callback-flood testing.
3716
3717 rcutorture.cbflood_n_per_burst= [KNL]
3718 Set the number of callbacks to be registered
3719 in a given burst of a callback-flood test.
3720
Paul E. McKenney4102ada2013-10-08 20:23:47 -07003721 rcutorture.fqs_duration= [KNL]
Paul E. McKenney21b05de2015-05-14 17:29:51 -07003722 Set duration of force_quiescent_state bursts
3723 in microseconds.
Paul E. McKenneydabb8aa2012-04-23 10:54:45 -07003724
Paul E. McKenney4102ada2013-10-08 20:23:47 -07003725 rcutorture.fqs_holdoff= [KNL]
Paul E. McKenney21b05de2015-05-14 17:29:51 -07003726 Set holdoff time within force_quiescent_state bursts
3727 in microseconds.
Paul E. McKenneydabb8aa2012-04-23 10:54:45 -07003728
Paul E. McKenney4102ada2013-10-08 20:23:47 -07003729 rcutorture.fqs_stutter= [KNL]
Paul E. McKenney21b05de2015-05-14 17:29:51 -07003730 Set wait time between force_quiescent_state bursts
3731 in seconds.
3732
3733 rcutorture.gp_cond= [KNL]
3734 Use conditional/asynchronous update-side
3735 primitives, if available.
Paul E. McKenneydabb8aa2012-04-23 10:54:45 -07003736
Paul E. McKenney4102ada2013-10-08 20:23:47 -07003737 rcutorture.gp_exp= [KNL]
Paul E. McKenney21b05de2015-05-14 17:29:51 -07003738 Use expedited update-side primitives, if available.
Paul E. McKenneydabb8aa2012-04-23 10:54:45 -07003739
Paul E. McKenney4102ada2013-10-08 20:23:47 -07003740 rcutorture.gp_normal= [KNL]
Paul E. McKenney21b05de2015-05-14 17:29:51 -07003741 Use normal (non-expedited) asynchronous
3742 update-side primitives, if available.
3743
3744 rcutorture.gp_sync= [KNL]
3745 Use normal (non-expedited) synchronous
3746 update-side primitives, if available. If all
3747 of rcutorture.gp_cond=, rcutorture.gp_exp=,
3748 rcutorture.gp_normal=, and rcutorture.gp_sync=
3749 are zero, rcutorture acts as if is interpreted
3750 they are all non-zero.
Paul E. McKenney4102ada2013-10-08 20:23:47 -07003751
3752 rcutorture.n_barrier_cbs= [KNL]
Paul E. McKenneydabb8aa2012-04-23 10:54:45 -07003753 Set callbacks/threads for rcu_barrier() testing.
3754
Paul E. McKenney4102ada2013-10-08 20:23:47 -07003755 rcutorture.nfakewriters= [KNL]
Paul E. McKenneydabb8aa2012-04-23 10:54:45 -07003756 Set number of concurrent RCU writers. These just
3757 stress RCU, they don't participate in the actual
3758 test, hence the "fake".
3759
Paul E. McKenney4102ada2013-10-08 20:23:47 -07003760 rcutorture.nreaders= [KNL]
Paul E. McKenney3838cc12015-03-12 13:55:48 -07003761 Set number of RCU readers. The value -1 selects
3762 N-1, where N is the number of CPUs. A value
3763 "n" less than -1 selects N-n-2, where N is again
3764 the number of CPUs. For example, -2 selects N
3765 (the number of CPUs), -3 selects N+1, and so on.
Paul E. McKenneydabb8aa2012-04-23 10:54:45 -07003766
Paul E. McKenney4102ada2013-10-08 20:23:47 -07003767 rcutorture.object_debug= [KNL]
3768 Enable debug-object double-call_rcu() testing.
3769
3770 rcutorture.onoff_holdoff= [KNL]
Paul E. McKenneydabb8aa2012-04-23 10:54:45 -07003771 Set time (s) after boot for CPU-hotplug testing.
3772
Paul E. McKenney4102ada2013-10-08 20:23:47 -07003773 rcutorture.onoff_interval= [KNL]
Paul E. McKenneydabb8aa2012-04-23 10:54:45 -07003774 Set time (s) between CPU-hotplug operations, or
3775 zero to disable CPU-hotplug testing.
3776
Paul E. McKenney4102ada2013-10-08 20:23:47 -07003777 rcutorture.shuffle_interval= [KNL]
Paul E. McKenneydabb8aa2012-04-23 10:54:45 -07003778 Set task-shuffle interval (s). Shuffling tasks
3779 allows some CPUs to go into dyntick-idle mode
3780 during the rcutorture test.
3781
Paul E. McKenney4102ada2013-10-08 20:23:47 -07003782 rcutorture.shutdown_secs= [KNL]
Paul E. McKenneydabb8aa2012-04-23 10:54:45 -07003783 Set time (s) after boot system shutdown. This
3784 is useful for hands-off automated testing.
3785
Paul E. McKenney4102ada2013-10-08 20:23:47 -07003786 rcutorture.stall_cpu= [KNL]
Paul E. McKenneydabb8aa2012-04-23 10:54:45 -07003787 Duration of CPU stall (s) to test RCU CPU stall
3788 warnings, zero to disable.
3789
Paul E. McKenney4102ada2013-10-08 20:23:47 -07003790 rcutorture.stall_cpu_holdoff= [KNL]
Paul E. McKenneydabb8aa2012-04-23 10:54:45 -07003791 Time to wait (s) after boot before inducing stall.
3792
Paul E. McKenney4102ada2013-10-08 20:23:47 -07003793 rcutorture.stat_interval= [KNL]
Paul E. McKenneydabb8aa2012-04-23 10:54:45 -07003794 Time (s) between statistics printk()s.
3795
Paul E. McKenney4102ada2013-10-08 20:23:47 -07003796 rcutorture.stutter= [KNL]
Paul E. McKenneydabb8aa2012-04-23 10:54:45 -07003797 Time (s) to stutter testing, for example, specifying
3798 five seconds causes the test to run for five seconds,
3799 wait for five seconds, and so on. This tests RCU's
3800 ability to transition abruptly to and from idle.
3801
Paul E. McKenney4102ada2013-10-08 20:23:47 -07003802 rcutorture.test_boost= [KNL]
Paul E. McKenneydabb8aa2012-04-23 10:54:45 -07003803 Test RCU priority boosting? 0=no, 1=maybe, 2=yes.
3804 "Maybe" means test if the RCU implementation
3805 under test support RCU priority boosting.
3806
Paul E. McKenney4102ada2013-10-08 20:23:47 -07003807 rcutorture.test_boost_duration= [KNL]
Paul E. McKenneydabb8aa2012-04-23 10:54:45 -07003808 Duration (s) of each individual boost test.
3809
Paul E. McKenney4102ada2013-10-08 20:23:47 -07003810 rcutorture.test_boost_interval= [KNL]
Paul E. McKenneydabb8aa2012-04-23 10:54:45 -07003811 Interval (s) between each boost test.
3812
Paul E. McKenney4102ada2013-10-08 20:23:47 -07003813 rcutorture.test_no_idle_hz= [KNL]
Paul E. McKenneydabb8aa2012-04-23 10:54:45 -07003814 Test RCU's dyntick-idle handling. See also the
3815 rcutorture.shuffle_interval parameter.
3816
Paul E. McKenney21b05de2015-05-14 17:29:51 -07003817 rcutorture.torture_runnable= [BOOT]
3818 Start rcutorture running at boot time.
3819
Paul E. McKenney4102ada2013-10-08 20:23:47 -07003820 rcutorture.torture_type= [KNL]
Paul E. McKenneydabb8aa2012-04-23 10:54:45 -07003821 Specify the RCU implementation to test.
3822
Paul E. McKenney4102ada2013-10-08 20:23:47 -07003823 rcutorture.verbose= [KNL]
Paul E. McKenneydabb8aa2012-04-23 10:54:45 -07003824 Enable additional printk() statements.
3825
Paul E. McKenney5a9be7c2015-11-24 15:44:06 -08003826 rcupdate.rcu_cpu_stall_suppress= [KNL]
3827 Suppress RCU CPU stall warning messages.
3828
3829 rcupdate.rcu_cpu_stall_timeout= [KNL]
3830 Set timeout for RCU CPU stall warning messages.
3831
Paul E. McKenney4102ada2013-10-08 20:23:47 -07003832 rcupdate.rcu_expedited= [KNL]
3833 Use expedited grace-period primitives, for
3834 example, synchronize_rcu_expedited() instead
3835 of synchronize_rcu(). This reduces latency,
3836 but can increase CPU utilization, degrade
3837 real-time latency, and degrade energy efficiency.
Paul E. McKenney79cfea02015-12-07 13:09:52 -08003838 No effect on CONFIG_TINY_RCU kernels.
Paul E. McKenney4102ada2013-10-08 20:23:47 -07003839
Paul E. McKenney5a9be7c2015-11-24 15:44:06 -08003840 rcupdate.rcu_normal= [KNL]
3841 Use only normal grace-period primitives,
3842 for example, synchronize_rcu() instead of
3843 synchronize_rcu_expedited(). This improves
Paul E. McKenney79cfea02015-12-07 13:09:52 -08003844 real-time latency, CPU utilization, and
3845 energy efficiency, but can expose users to
3846 increased grace-period latency. This parameter
3847 overrides rcupdate.rcu_expedited. No effect on
3848 CONFIG_TINY_RCU kernels.
Paul E. McKenney4102ada2013-10-08 20:23:47 -07003849
Paul E. McKenney3e42ec12015-11-25 18:56:00 -08003850 rcupdate.rcu_normal_after_boot= [KNL]
3851 Once boot has completed (that is, after
3852 rcu_end_inkernel_boot() has been invoked), use
Paul E. McKenney79cfea02015-12-07 13:09:52 -08003853 only normal grace-period primitives. No effect
3854 on CONFIG_TINY_RCU kernels.
Paul E. McKenney3e42ec12015-11-25 18:56:00 -08003855
Paul E. McKenney52db30a2014-07-01 18:16:30 -07003856 rcupdate.rcu_task_stall_timeout= [KNL]
3857 Set timeout in jiffies for RCU task stall warning
3858 messages. Disable with a value less than or equal
3859 to zero.
3860
Pranith Kumar74860fe2014-09-19 11:34:09 -04003861 rcupdate.rcu_self_test= [KNL]
3862 Run the RCU early boot self tests
3863
3864 rcupdate.rcu_self_test_bh= [KNL]
3865 Run the RCU bh early boot self tests
3866
3867 rcupdate.rcu_self_test_sched= [KNL]
3868 Run the RCU sched early boot self tests
3869
Olof Johanssonffdfc402005-09-06 15:17:19 -07003870 rdinit= [KNL]
3871 Format: <full_path>
3872 Run specified binary instead of /init from the ramdisk,
3873 used for early userspace startup. See initrd.
3874
Tom Lendacky9272c252019-08-19 15:52:35 +00003875 rdrand= [X86]
3876 force - Override the decision by the kernel to hide the
3877 advertisement of RDRAND support (this affects
3878 certain AMD processors because of buggy BIOS
3879 support, specifically around the suspend/resume
3880 path).
3881
Robin Holt1b3a5d02013-07-08 16:01:42 -07003882 reboot= [KNL]
3883 Format (x86 or x86_64):
3884 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] \
3885 [[,]s[mp]#### \
3886 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \
3887 [[,]f[orce]
3888 Where reboot_mode is one of warm (soft) or cold (hard) or gpio,
3889 reboot_type is one of bios, acpi, kbd, triple, efi, or pci,
3890 reboot_force is either force or not specified,
3891 reboot_cpu is s[mp]#### with #### being the processor
3892 to be used for rebooting.
Linus Torvalds1da177e2005-04-16 15:20:36 -07003893
Paul Jackson46b6d942008-07-04 10:00:09 -07003894 relax_domain_level=
3895 [KNL, SMP] Set scheduler's default relax_domain_level.
seokhoon.yoon09c3bcc2016-08-02 23:23:57 +09003896 See Documentation/cgroup-v1/cpusets.txt.
Paul Jackson46b6d942008-07-04 10:00:09 -07003897
Rafael J. Wysocki0399d4d2014-05-26 13:40:59 +02003898 relative_sleep_states=
3899 [SUSPEND] Use sleep state labeling where the deepest
3900 state available other than hibernation is always "mem".
3901 Format: { "0" | "1" }
3902 0 -- Traditional sleep state labels.
3903 1 -- Relative sleep state labels.
3904
Linus Torvalds1da177e2005-04-16 15:20:36 -07003905 reserve= [KNL,BUGS] Force the kernel to ignore some iomem area
3906
Alan Coxcd4f0ef2007-07-31 00:37:59 -07003907 reservetop= [X86-32]
Zachary Amsden461a9af2006-09-25 23:32:25 -07003908 Format: nn[KMG]
3909 Reserves a hole at the top of the kernel virtual
3910 address space.
3911
H. Peter Anvin9ea77bd2010-08-25 16:38:20 -07003912 reservelow= [X86]
3913 Format: nn[K]
3914 Set the amount of memory to reserve for BIOS at
3915 the bottom of the address space.
3916
Vivek Goyal7e962872006-09-27 01:50:44 -07003917 reset_devices [KNL] Force drivers to reset the underlying device
3918 during initialization.
3919
Randy Dunlapa9913042005-10-23 12:57:11 -07003920 resume= [SWSUSP]
3921 Specify the partition device for software suspend
Minho Ban2df83fa2012-05-14 21:45:31 +02003922 Format:
3923 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>}
Linus Torvalds1da177e2005-04-16 15:20:36 -07003924
Rafael J. Wysockiecbd0da2006-12-06 20:34:13 -08003925 resume_offset= [SWSUSP]
3926 Specify the offset from the beginning of the partition
3927 given by "resume=" at which the swap header is located,
3928 in <PAGE_SIZE> units (needed only for swap files).
3929 See Documentation/power/swsusp-and-swap-files.txt
3930
Barry Songf126f732011-10-10 23:38:41 +02003931 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to
3932 read the resume files
3933
Barry Song6f8d7022011-10-06 20:34:46 +02003934 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up.
3935 Useful for devices that are detected asynchronously
3936 (e.g. USB and MMC devices).
3937
Bojan Smojverf996fc92010-09-09 23:06:23 +02003938 hibernate= [HIBERNATION]
3939 noresume Don't check if there's a hibernation image
3940 present during boot.
3941 nocompress Don't compress/decompress hibernation images.
Kees Cooka6e15a32014-06-13 13:30:35 -07003942 no Disable hibernation and resume.
Rafael J. Wysocki4c0b6c12016-07-10 02:12:10 +02003943 protect_image Turn on image protection during restoration
3944 (that will set all pages holding image data
3945 during restoration read-only).
Bojan Smojverf996fc92010-09-09 23:06:23 +02003946
Michael Neuling0a7b35cb2007-02-10 01:44:33 -08003947 retain_initrd [RAM] Keep initrd memory after extraction
3948
Andrew Clausen0efbb782015-01-09 20:24:55 +00003949 rfkill.default_state=
3950 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm,
3951 etc. communication is blocked by default.
3952 1 Unblocked.
3953
3954 rfkill.master_switch_mode=
3955 0 The "airplane mode" button does nothing.
3956 1 The "airplane mode" button toggles between everything
3957 blocked and the previous configuration.
3958 2 The "airplane mode" button toggles between everything
3959 blocked and everything unblocked.
3960
Linus Torvalds1da177e2005-04-16 15:20:36 -07003961 rhash_entries= [KNL,NET]
3962 Set number of hash buckets for route cache
3963
Linus Torvalds1da177e2005-04-16 15:20:36 -07003964 ro [KNL] Mount root device read-only on boot
3965
Kees Cookd2aa1ac2016-02-17 14:41:13 -08003966 rodata= [KNL]
3967 on Mark read-only kernel memory as read-only (default).
3968 off Leave read-only kernel memory writable for debugging.
3969
Heiko Stuebner605df8a2016-02-22 12:55:01 +01003970 rockchip.usb_uart
3971 Enable the uart passthrough on the designated usb port
3972 on Rockchip SoCs. When active, the signals of the
3973 debug-uart get routed to the D+ and D- pins of the usb
3974 port and the regular usb controller gets disabled.
3975
Linus Torvalds1da177e2005-04-16 15:20:36 -07003976 root= [KNL] Root filesystem
Will Drewryf2d34fd2011-08-03 16:21:08 -07003977 See name_to_dev_t comment in init/do_mounts.c.
Linus Torvalds1da177e2005-04-16 15:20:36 -07003978
3979 rootdelay= [KNL] Delay (in seconds) to pause before attempting to
3980 mount the root filesystem
3981
3982 rootflags= [KNL] Set root filesystem mount option string
3983
3984 rootfstype= [KNL] Set root filesystem type
3985
Pierre Ossmancc1ed752007-07-15 23:40:35 -07003986 rootwait [KNL] Wait (indefinitely) for root device to show up.
3987 Useful for devices that are detected asynchronously
3988 (e.g. USB and MMC devices).
3989
Robert Tivy5c71d612013-03-28 18:41:46 -07003990 rproc_mem=nn[KMG][@address]
3991 [KNL,ARM,CMA] Remoteproc physical memory block.
3992 Memory area to be used by remote processor image,
3993 managed by CMA.
3994
Linus Torvalds1da177e2005-04-16 15:20:36 -07003995 rw [KNL] Mount root device read-write on boot
3996
3997 S [KNL] Run init in single mode
3998
Gerald Schaeferc60d1ae2014-07-18 17:37:08 +02003999 s390_iommu= [HW,S390]
4000 Set s390 IOTLB flushing mode
4001 strict
4002 With strict flushing every unmap operation will result in
4003 an IOTLB flush. Default is lazy flushing before reuse,
4004 which is faster.
4005
Linus Torvalds1da177e2005-04-16 15:20:36 -07004006 sa1100ir [NET]
4007 See drivers/net/irda/sa1100_ir.c.
4008
Linus Torvalds1da177e2005-04-16 15:20:36 -07004009 sbni= [NET] Granch SBNI12 leased line adapter
Randy Dunlapa9913042005-10-23 12:57:11 -07004010
Mike Travisf6630112009-11-17 18:22:15 -06004011 sched_debug [KNL] Enables verbose scheduler debug messages.
4012
Mel Gormancb251762016-02-05 09:08:36 +00004013 schedstats= [KNL,X86] Enable or disable scheduled statistics.
4014 Allowed values are enable and disable. This feature
4015 incurs a small amount of overhead in the scheduler
4016 but is useful for debugging and performance tuning.
4017
Mike Galbraith5307c952012-05-08 12:20:58 +02004018 skew_tick= [KNL] Offset the periodic timer tick per cpu to mitigate
4019 xtime_lock contention on larger systems, and/or RCU lock
4020 contention on all systems with CONFIG_MAXSMP set.
4021 Format: { "0" | "1" }
4022 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1"
4023 1 -- enable.
4024 Note: increases power consumption, thus should only be
4025 enabled if running jitter sensitive (HPC/RT) workloads.
4026
Randy Dunlap0cb55ad2009-04-05 15:55:22 -07004027 security= [SECURITY] Choose a security module to enable at boot.
4028 If this boot parameter is not specified, only the first
4029 security module asking for security registration will be
4030 loaded. An invalid security module name will be treated
4031 as if no module has been chosen.
4032
4033 selinux= [SELINUX] Disable or enable SELinux at boot time.
Linus Torvalds1da177e2005-04-16 15:20:36 -07004034 Format: { "0" | "1" }
4035 See security/selinux/Kconfig help text.
4036 0 -- disable.
4037 1 -- enable.
4038 Default value is set via kernel config option.
4039 If enabled at boot time, /selinux/disable can be used
4040 later to disable prior to initial policy load.
4041
John Johansenc1c124e2010-07-29 14:48:09 -07004042 apparmor= [APPARMOR] Disable or enable AppArmor at boot time
4043 Format: { "0" | "1" }
4044 See security/apparmor/Kconfig help text
4045 0 -- disable.
4046 1 -- enable.
4047 Default value is set via kernel config option.
4048
Alan Coxcd4f0ef2007-07-31 00:37:59 -07004049 serialnumber [BUGS=X86-32]
Linus Torvalds1da177e2005-04-16 15:20:36 -07004050
Linus Torvalds1da177e2005-04-16 15:20:36 -07004051 shapers= [NET]
4052 Maximal number of shapers.
Randy Dunlapa9913042005-10-23 12:57:11 -07004053
Yinghai Lub05f78f2008-08-22 01:32:50 -07004054 show_msr= [x86] show boot-time MSR settings
4055 Format: { <integer> }
4056 Show boot-time (BIOS-initialized) MSR settings.
4057 The parameter means the number of CPUs to show,
4058 for example 1 means boot CPU only.
4059
Linus Torvalds1da177e2005-04-16 15:20:36 -07004060 simeth= [IA-64]
4061 simscsi=
Randy Dunlapa9913042005-10-23 12:57:11 -07004062
Linus Torvalds1da177e2005-04-16 15:20:36 -07004063 slram= [HW,MTD]
4064
Joonsoo Kim423c9292014-10-09 15:26:22 -07004065 slab_nomerge [MM]
4066 Disable merging of slabs with similar size. May be
4067 necessary if there is some reason to distinguish
4068 allocs to different slabs. Debug options disable
4069 merging on their own.
4070 For more information see Documentation/vm/slub.txt.
4071
David Rientjes3df1ccc2011-10-18 22:09:28 -07004072 slab_max_order= [MM, SLAB]
4073 Determines the maximum allowed order for slabs.
4074 A high setting may cause OOMs due to memory
4075 fragmentation. Defaults to 1 for systems with
4076 more than 32MB of RAM, 0 otherwise.
4077
Christoph Lameterf0630ff2007-07-15 23:38:14 -07004078 slub_debug[=options[,slabs]] [MM, SLUB]
4079 Enabling slub_debug allows one to determine the
4080 culprit if slab objects become corrupted. Enabling
4081 slub_debug can create guard zones around objects and
4082 may poison objects when not in use. Also tracks the
4083 last alloc / free. For more information see
4084 Documentation/vm/slub.txt.
Christoph Lameterc1aee212007-05-31 00:40:47 -07004085
Tejun Heoa4ffb672018-08-24 13:22:21 +09004086 slub_memcg_sysfs= [MM, SLUB]
4087 Determines whether to enable sysfs directories for
4088 memory cgroup sub-caches. 1 to enable, 0 to disable.
4089 The default is determined by CONFIG_SLUB_MEMCG_SYSFS_ON.
4090 Enabling this can lead to a very high number of debug
4091 directories and files being created under
4092 /sys/kernel/slub.
4093
Christoph Lameterc1aee212007-05-31 00:40:47 -07004094 slub_max_order= [MM, SLUB]
Christoph Lameterf0630ff2007-07-15 23:38:14 -07004095 Determines the maximum allowed order for slabs.
4096 A high setting may cause OOMs due to memory
4097 fragmentation. For more information see
4098 Documentation/vm/slub.txt.
Christoph Lameterc1aee212007-05-31 00:40:47 -07004099
4100 slub_min_objects= [MM, SLUB]
Christoph Lameterf0630ff2007-07-15 23:38:14 -07004101 The minimum number of objects per slab. SLUB will
4102 increase the slab order up to slub_max_order to
4103 generate a sufficiently large slab able to contain
4104 the number of objects indicated. The higher the number
4105 of objects the smaller the overhead of tracking slabs
4106 and the less frequently locks need to be acquired.
Christoph Lameterc1aee212007-05-31 00:40:47 -07004107 For more information see Documentation/vm/slub.txt.
4108
4109 slub_min_order= [MM, SLUB]
Masanari Iida24775d62012-02-15 00:26:42 +09004110 Determines the minimum page order for slabs. Must be
Christoph Lameterf0630ff2007-07-15 23:38:14 -07004111 lower than slub_max_order.
Christoph Lameterc1aee212007-05-31 00:40:47 -07004112 For more information see Documentation/vm/slub.txt.
4113
4114 slub_nomerge [MM, SLUB]
Joonsoo Kim423c9292014-10-09 15:26:22 -07004115 Same with slab_nomerge. This is supported for legacy.
4116 See slab_nomerge for more information.
Christoph Lameterc1aee212007-05-31 00:40:47 -07004117
Linus Torvalds1da177e2005-04-16 15:20:36 -07004118 smart2= [HW]
4119 Format: <io1>[,<io2>[,...,<io8>]]
4120
Bjorn Helgaasd0d4f692007-05-08 00:36:05 -07004121 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices
4122 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port
4123 smsc-ircc2.ircc_sir= [HW] SIR base I/O port
4124 smsc-ircc2.ircc_fir= [HW] FIR base I/O port
4125 smsc-ircc2.ircc_irq= [HW] IRQ line
4126 smsc-ircc2.ircc_dma= [HW] DMA channel
4127 smsc-ircc2.ircc_transceiver= [HW] Transceiver type:
4128 0: Toshiba Satellite 1800 (GP data pin select)
4129 1: Fast pin select (default)
4130 2: ATC IRMode
4131
Sascha Silbe52c48c52016-04-05 12:53:38 +02004132 smt [KNL,S390] Set the maximum number of threads (logical
4133 CPUs) to use per physical CPU on systems capable of
4134 symmetric multithreading (SMT). Will be capped to the
4135 actual hardware limit.
4136 Format: <integer>
4137 Default: -1 (no limit)
4138
Ingo Molnar9c44bc02008-05-12 21:21:04 +02004139 softlockup_panic=
4140 [KNL] Should the soft-lockup detector generate panics.
Randy Dunlap44a4dcf2011-04-04 15:02:24 -07004141 Format: <integer>
Ingo Molnar9c44bc02008-05-12 21:21:04 +02004142
Aaron Tomlined235872014-06-23 13:22:05 -07004143 softlockup_all_cpu_backtrace=
4144 [KNL] Should the soft-lockup detector generate
4145 backtraces on all cpus.
4146 Format: <integer>
4147
Linus Torvalds1da177e2005-04-16 15:20:36 -07004148 sonypi.*= [HW] Sony Programmable I/O Control Device driver
Paul Bolle395cf962011-08-15 02:02:26 +02004149 See Documentation/laptops/sonypi.txt
Linus Torvalds1da177e2005-04-16 15:20:36 -07004150
David Woodhouse8f96937ee2018-01-11 21:46:26 +00004151 spectre_v2= [X86] Control mitigation of Spectre variant 2
4152 (indirect branch speculation) vulnerability.
Thomas Gleixnerdda365c2018-11-25 19:33:45 +01004153 The default operation protects the kernel from
4154 user space attacks.
David Woodhouse8f96937ee2018-01-11 21:46:26 +00004155
Thomas Gleixnerdda365c2018-11-25 19:33:45 +01004156 on - unconditionally enable, implies
4157 spectre_v2_user=on
4158 off - unconditionally disable, implies
4159 spectre_v2_user=off
David Woodhouse8f96937ee2018-01-11 21:46:26 +00004160 auto - kernel detects whether your CPU model is
4161 vulnerable
4162
4163 Selecting 'on' will, and 'auto' may, choose a
4164 mitigation method at run time according to the
4165 CPU, the available microcode, the setting of the
4166 CONFIG_RETPOLINE configuration option, and the
4167 compiler with which the kernel was built.
4168
Thomas Gleixnerdda365c2018-11-25 19:33:45 +01004169 Selecting 'on' will also enable the mitigation
4170 against user space to user space task attacks.
4171
4172 Selecting 'off' will disable both the kernel and
4173 the user space protections.
4174
David Woodhouse8f96937ee2018-01-11 21:46:26 +00004175 Specific mitigations can also be selected manually:
4176
4177 retpoline - replace indirect branches
4178 retpoline,generic - google's original retpoline
4179 retpoline,amd - AMD-specific minimal thunk
4180
4181 Not specifying this option is equivalent to
4182 spectre_v2=auto.
4183
Thomas Gleixnerdda365c2018-11-25 19:33:45 +01004184 spectre_v2_user=
4185 [X86] Control mitigation of Spectre variant 2
4186 (indirect branch speculation) vulnerability between
4187 user space tasks
4188
4189 on - Unconditionally enable mitigations. Is
4190 enforced by spectre_v2=on
4191
4192 off - Unconditionally disable mitigations. Is
4193 enforced by spectre_v2=off
4194
Thomas Gleixner91d9bbd2018-11-25 19:33:54 +01004195 prctl - Indirect branch speculation is enabled,
4196 but mitigation can be enabled via prctl
4197 per thread. The mitigation control state
4198 is inherited on fork.
4199
Thomas Gleixnere58cf372018-11-25 19:33:56 +01004200 prctl,ibpb
4201 - Like "prctl" above, but only STIBP is
4202 controlled per thread. IBPB is issued
4203 always when switching between different user
4204 space processes.
4205
Thomas Gleixner6f4b9252018-11-25 19:33:55 +01004206 seccomp
4207 - Same as "prctl" above, but all seccomp
4208 threads will enable the mitigation unless
4209 they explicitly opt out.
4210
Thomas Gleixnere58cf372018-11-25 19:33:56 +01004211 seccomp,ibpb
4212 - Like "seccomp" above, but only STIBP is
4213 controlled per thread. IBPB is issued
4214 always when switching between different
4215 user space processes.
4216
Thomas Gleixnerdda365c2018-11-25 19:33:45 +01004217 auto - Kernel selects the mitigation depending on
4218 the available CPU features and vulnerability.
Thomas Gleixner6f4b9252018-11-25 19:33:55 +01004219
4220 Default mitigation:
4221 If CONFIG_SECCOMP=y then "seccomp", otherwise "prctl"
Thomas Gleixnerdda365c2018-11-25 19:33:45 +01004222
4223 Not specifying this option is equivalent to
4224 spectre_v2_user=auto.
4225
Konrad Rzeszutek Wilk6f70a552018-04-25 22:04:21 -04004226 spec_store_bypass_disable=
4227 [HW] Control Speculative Store Bypass (SSB) Disable mitigation
4228 (Speculative Store Bypass vulnerability)
4229
4230 Certain CPUs are vulnerable to an exploit against a
4231 a common industry wide performance optimization known
4232 as "Speculative Store Bypass" in which recent stores
4233 to the same memory location may not be observed by
4234 later loads during speculative execution. The idea
4235 is that such stores are unlikely and that they can
4236 be detected prior to instruction retirement at the
4237 end of a particular speculation execution window.
4238
4239 In vulnerable processors, the speculatively forwarded
4240 store can be used in a cache side channel attack, for
4241 example to read memory to which the attacker does not
4242 directly have access (e.g. inside sandboxed code).
4243
4244 This parameter controls whether the Speculative Store
4245 Bypass optimization is used.
4246
Kees Cook05a85a32018-05-03 14:37:54 -07004247 on - Unconditionally disable Speculative Store Bypass
4248 off - Unconditionally enable Speculative Store Bypass
4249 auto - Kernel detects whether the CPU model contains an
4250 implementation of Speculative Store Bypass and
4251 picks the most appropriate mitigation. If the
4252 CPU is not vulnerable, "off" is selected. If the
4253 CPU is vulnerable the default mitigation is
4254 architecture and Kconfig dependent. See below.
4255 prctl - Control Speculative Store Bypass per thread
4256 via prctl. Speculative Store Bypass is enabled
4257 for a process by default. The state of the control
4258 is inherited on fork.
4259 seccomp - Same as "prctl" above, but all seccomp threads
4260 will disable SSB unless they explicitly opt out.
Konrad Rzeszutek Wilk6f70a552018-04-25 22:04:21 -04004261
4262 Not specifying this option is equivalent to
4263 spec_store_bypass_disable=auto.
4264
Kees Cook05a85a32018-05-03 14:37:54 -07004265 Default mitigations:
4266 X86: If CONFIG_SECCOMP=y "seccomp", otherwise "prctl"
4267
Linus Torvalds1da177e2005-04-16 15:20:36 -07004268 spia_io_base= [HW,MTD]
4269 spia_fio_base=
4270 spia_pedr=
4271 spia_peddr=
4272
Marc Zyngier3a64e6a2018-07-20 10:56:25 +01004273 ssbd= [ARM64,HW]
4274 Speculative Store Bypass Disable control
4275
4276 On CPUs that are vulnerable to the Speculative
4277 Store Bypass vulnerability and offer a
4278 firmware based mitigation, this parameter
4279 indicates how the mitigation should be used:
4280
4281 force-on: Unconditionally enable mitigation for
4282 for both kernel and userspace
4283 force-off: Unconditionally disable mitigation for
4284 for both kernel and userspace
4285 kernel: Always enable mitigation in the
4286 kernel, and offer a prctl interface
4287 to allow userspace to register its
4288 interest in being mitigated too.
4289
Hugh Dickinscfc0eb402017-06-19 04:03:24 -07004290 stack_guard_gap= [MM]
4291 override the default stack gap protection. The value
4292 is in page units and it defines how many pages prior
4293 to (for stacks growing down) resp. after (for stacks
4294 growing up) the main stack are reserved for no other
4295 mapping. Default value is 256 pages.
4296
Steven Rostedtf38f1d22008-12-16 23:06:40 -05004297 stacktrace [FTRACE]
4298 Enabled the stack tracer on boot up.
4299
Steven Rostedt762e1202011-12-19 22:01:00 -05004300 stacktrace_filter=[function-list]
4301 [FTRACE] Limit the functions that the stack tracer
4302 will trace at boot up. function-list is a comma separated
4303 list of functions. This list can be changed at run
4304 time by the stack_trace_filter file in the debugfs
4305 tracing directory. Note, this enables stack tracing
4306 and the stacktrace above is not needed.
4307
Linus Torvalds1da177e2005-04-16 15:20:36 -07004308 sti= [PARISC,HW]
4309 Format: <num>
4310 Set the STI (builtin display/keyboard on the HP-PARISC
4311 machines) console (graphic card) which should be used
4312 as the initial boot-console.
4313 See also comment in drivers/video/console/sticore.c.
4314
4315 sti_font= [HW]
4316 See comment in drivers/video/console/sticore.c.
4317
4318 stifb= [HW]
4319 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]]
4320
Trond Myklebustcbf11072009-08-09 15:06:19 -04004321 sunrpc.min_resvport=
4322 sunrpc.max_resvport=
4323 [NFS,SUNRPC]
4324 SunRPC servers often require that client requests
4325 originate from a privileged port (i.e. a port in the
4326 range 0 < portnr < 1024).
4327 An administrator who wishes to reserve some of these
4328 ports for other uses may adjust the range that the
4329 kernel's sunrpc client considers to be privileged
4330 using these two parameters to set the minimum and
4331 maximum port values.
4332
Trond Myklebustff3ac5c2016-06-24 10:55:50 -04004333 sunrpc.svc_rpc_per_connection_limit=
4334 [NFS,SUNRPC]
4335 Limit the number of requests that the server will
4336 process in parallel from a single connection.
4337 The default value is 0 (no limit).
4338
Greg Banks42a7fc42007-03-06 01:42:23 -08004339 sunrpc.pool_mode=
4340 [NFS]
4341 Control how the NFS server code allocates CPUs to
4342 service thread pools. Depending on how many NICs
4343 you have and where their interrupts are bound, this
4344 option will affect which CPUs will do NFS serving.
4345 Note: this parameter cannot be changed while the
4346 NFS server is running.
4347
4348 auto the server chooses an appropriate mode
4349 automatically using heuristics
4350 global a single global pool contains all CPUs
4351 percpu one pool for each CPU
4352 pernode one pool for each NUMA node (equivalent
4353 to global on non-NUMA machines)
4354
Trond Myklebustcbf11072009-08-09 15:06:19 -04004355 sunrpc.tcp_slot_table_entries=
4356 sunrpc.udp_slot_table_entries=
4357 [NFS,SUNRPC]
4358 Sets the upper limit on the number of simultaneous
4359 RPC calls that can be sent from the client to a
4360 server. Increasing these values may allow you to
4361 improve throughput, but will also increase the
4362 amount of memory reserved for use by the client.
4363
Brian Norris1d4a9c12015-02-22 21:16:49 -08004364 suspend.pm_test_delay=
4365 [SUSPEND]
4366 Sets the number of seconds to remain in a suspend test
4367 mode before resuming the system (see
4368 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG
4369 is set. Default value is 5.
4370
Michal Hocko07555ac2013-08-22 16:35:46 -07004371 swapaccount=[0|1]
Michal Hockoa42c3902010-11-24 12:57:08 -08004372 [KNL] Enable accounting of swap in memory resource
4373 controller if no parameter or 1 is given or disable
seokhoon.yoon09c3bcc2016-08-02 23:23:57 +09004374 it if 0 is given (See Documentation/cgroup-v1/memory.txt)
Michal Hockoa42c3902010-11-24 12:57:08 -08004375
Jiri Kosina91fec0f2013-11-27 13:48:09 +01004376 swiotlb= [ARM,IA-64,PPC,MIPS,X86]
Geert Uytterhoeven41c6b3e2016-12-16 14:28:42 +01004377 Format: { <int> | force | noforce }
Jiri Kosina91fec0f2013-11-27 13:48:09 +01004378 <int> -- Number of I/O TLB slabs
4379 force -- force using of bounce buffers even if they
4380 wouldn't be automatically used by the kernel
Geert Uytterhoeven41c6b3e2016-12-16 14:28:42 +01004381 noforce -- Never use bounce buffers (for debugging)
Randy Dunlapa9913042005-10-23 12:57:11 -07004382
Linus Torvalds1da177e2005-04-16 15:20:36 -07004383 switches= [HW,M68k]
4384
Andi Kleene52eec12010-09-08 16:54:17 +02004385 sysfs.deprecated=0|1 [KNL]
4386 Enable/disable old style sysfs layout for old udev
4387 on older distributions. When this option is enabled
4388 very new udev will not work anymore. When this option
4389 is disabled (or CONFIG_SYSFS_DEPRECATED not compiled)
4390 in older udev will not work anymore.
4391 Default depends on CONFIG_SYSFS_DEPRECATED_V2 set in
4392 the kernel configuration.
4393
Ingo Molnar5d6f6472006-12-13 00:34:36 -08004394 sysrq_always_enabled
4395 [KNL]
4396 Ignore sysrq setting - this boot parameter will
4397 neutralize any effect of /proc/sys/kernel/sysrq.
4398 Useful for debugging.
4399
Fabian Frederick747029a2014-11-06 19:46:50 +01004400 tcpmhash_entries= [KNL,NET]
4401 Set the number of tcp_metrics_hash slots.
4402 Default value is 8192 or 16384 depending on total
4403 ram pages. This is used to specify the TCP metrics
4404 cache size. See Documentation/networking/ip-sysctl.txt
4405 "tcp_no_metrics_save" section for more details.
4406
Linus Torvalds1da177e2005-04-16 15:20:36 -07004407 tdfx= [HW,DRM]
4408
Srinivas Pandruvadaacc82342014-09-02 11:54:41 -07004409 test_suspend= [SUSPEND][,N]
David Brownell77437fd2008-07-23 21:28:33 -07004410 Specify "mem" (for Suspend-to-RAM) or "standby" (for
Srinivas Pandruvadaacc82342014-09-02 11:54:41 -07004411 standby suspend) or "freeze" (for suspend type freeze)
4412 as the system sleep state during system startup with
4413 the optional capability to repeat N number of times.
4414 The system is woken from this state using a
4415 wakeup-capable RTC alarm.
David Brownell77437fd2008-07-23 21:28:33 -07004416
Linus Torvalds1da177e2005-04-16 15:20:36 -07004417 thash_entries= [KNL,NET]
4418 Set number of hash buckets for TCP connection
4419
Len Brownf8707ec2007-08-12 00:12:54 -04004420 thermal.act= [HW,ACPI]
4421 -1: disable all active trip points in all thermal zones
4422 <degrees C>: override all lowest active trip points
4423
Len Brownc52a7412007-08-14 15:49:32 -04004424 thermal.crt= [HW,ACPI]
4425 -1: disable all critical trip points in all thermal zones
Zhang Rui22a94d72008-10-17 02:41:20 -04004426 <degrees C>: override all critical trip points
Len Brownc52a7412007-08-14 15:49:32 -04004427
Len Brownf5487142007-08-12 00:12:44 -04004428 thermal.nocrt= [HW,ACPI]
4429 Set to disable actions on ACPI thermal zone
4430 critical and hot trip points.
4431
Len Brown72b33ef2007-08-12 00:12:17 -04004432 thermal.off= [HW,ACPI]
4433 1: disable ACPI thermal control
4434
Len Browna70cdc52007-08-12 00:12:35 -04004435 thermal.psv= [HW,ACPI]
4436 -1: disable all passive trip points
Randy Dunlapada9cfd2008-12-19 10:57:32 -08004437 <degrees C>: override all passive trip points to this
4438 value
Len Browna70cdc52007-08-12 00:12:35 -04004439
Len Brown730ff342007-08-12 00:12:26 -04004440 thermal.tzp= [HW,ACPI]
4441 Specify global default ACPI thermal zone polling rate
4442 <deci-seconds>: poll all this frequency
4443 0: no polling (default)
4444
Thomas Gleixner8d32a302011-02-23 23:52:23 +00004445 threadirqs [KNL]
4446 Force threading of all interrupt handlers except those
Masanari Iida24775d62012-02-15 00:26:42 +09004447 marked explicitly IRQF_NO_THREAD.
Thomas Gleixner8d32a302011-02-23 23:52:23 +00004448
Konrad Rzeszutek Wilk2ca62b02013-05-08 17:12:44 -04004449 tmem [KNL,XEN]
4450 Enable the Transcendent memory driver if built-in.
4451
4452 tmem.cleancache=0|1 [KNL, XEN]
4453 Default is on (1). Disable the usage of the cleancache
4454 API to send anonymous pages to the hypervisor.
4455
4456 tmem.frontswap=0|1 [KNL, XEN]
4457 Default is on (1). Disable the usage of the frontswap
Konrad Rzeszutek Wilk37d46e12013-05-14 13:56:42 -04004458 API to send swap pages to the hypervisor. If disabled
4459 the selfballooning and selfshrinking are force disabled.
Konrad Rzeszutek Wilk2ca62b02013-05-08 17:12:44 -04004460
4461 tmem.selfballooning=0|1 [KNL, XEN]
4462 Default is on (1). Disable the driving of swap pages
4463 to the hypervisor.
4464
4465 tmem.selfshrinking=0|1 [KNL, XEN]
4466 Default is on (1). Partial swapoff that immediately
4467 transfers pages from Xen hypervisor back to the
4468 kernel based on different criteria.
4469
Heiko Carstens2b1a61f2008-12-25 13:39:23 +01004470 topology= [S390]
4471 Format: {off | on}
4472 Specify if the kernel should make use of the cpu
Sylvestre Ledruf65e51d2011-04-04 15:04:46 -07004473 topology information if the hardware supports this.
4474 The scheduler will make use of this information and
Heiko Carstens2b1a61f2008-12-25 13:39:23 +01004475 e.g. base its process migration decisions on it.
Heiko Carstensc9af3fa2010-10-25 16:10:43 +02004476 Default is on.
Heiko Carstens2b1a61f2008-12-25 13:39:23 +01004477
Nishanth Aravamudan2d73bae2014-10-10 09:04:49 -07004478 topology_updates= [KNL, PPC, NUMA]
4479 Format: {off}
4480 Specify if the kernel should ignore (off)
4481 topology updates sent by the hypervisor to this
4482 LPAR.
4483
Linus Torvalds1da177e2005-04-16 15:20:36 -07004484 tp720= [HW,PS2]
4485
Rajiv Andrade225a9be2010-03-25 00:55:32 -03004486 tpm_suspend_pcr=[HW,TPM]
4487 Format: integer pcr id
4488 Specify that at suspend time, the tpm driver
4489 should extend the specified pcr with zeros,
4490 as a workaround for some chips which fail to
4491 flush the last written pcr on TPM_SaveState.
4492 This will guarantee that all the other pcrs
4493 are saved.
4494
Li Zefan9d612be2009-06-24 17:33:15 +08004495 trace_buf_size=nn[KMG]
Joonsoo Kim3e6fb8e2014-12-03 10:39:20 +09004496 [FTRACE] will set tracing buffer size on each cpu.
KOSAKI Motohiro631595f2009-03-10 13:57:10 +09004497
Li Zefan020e5f82009-07-01 10:47:05 +08004498 trace_event=[event-list]
4499 [FTRACE] Set and start specified trace events in order
Brian Norrisd81749e2016-05-23 13:37:58 -07004500 to facilitate early boot debugging. The event-list is a
4501 comma separated list of trace events to enable. See
4502 also Documentation/trace/events.txt
Li Zefan020e5f82009-07-01 10:47:05 +08004503
Steven Rostedt7bcfaf52012-11-01 22:56:07 -04004504 trace_options=[option-list]
4505 [FTRACE] Enable or disable tracer options at boot.
4506 The option-list is a comma delimited list of options
4507 that can be enabled or disabled just as if you were
4508 to echo the option name into
4509
4510 /sys/kernel/debug/tracing/trace_options
4511
4512 For example, to enable stacktrace option (to dump the
4513 stack trace of each event), add to the command line:
4514
4515 trace_options=stacktrace
4516
4517 See also Documentation/trace/ftrace.txt "trace options"
4518 section.
4519
Steven Rostedt (Red Hat)0daa23022014-12-12 22:27:10 -05004520 tp_printk[FTRACE]
4521 Have the tracepoints sent to printk as well as the
4522 tracing ring buffer. This is useful for early boot up
4523 where the system hangs or reboots and does not give the
4524 option for reading the tracing buffer or performing a
4525 ftrace_dump_on_oops.
4526
4527 To turn off having tracepoints sent to printk,
4528 echo 0 > /proc/sys/kernel/tracepoint_printk
4529 Note, echoing 1 into this file without the
4530 tracepoint_printk kernel cmdline option has no effect.
4531
4532 ** CAUTION **
4533
4534 Having tracepoints sent to printk() and activating high
4535 frequency tracepoints such as irq or sched, can cause
4536 the system to live lock.
4537
Steven Rostedt (Red Hat)de7edd32013-06-14 16:21:43 -04004538 traceoff_on_warning
4539 [FTRACE] enable this option to disable tracing when a
4540 warning is hit. This turns off "tracing_on". Tracing can
4541 be enabled again by echoing '1' into the "tracing_on"
4542 file located in /sys/kernel/debug/tracing/
4543
4544 This option is useful, as it disables the trace before
4545 the WARNING dump is called, which prevents the trace to
4546 be filled with content caused by the warning output.
4547
4548 This option can also be set at run time via the sysctl
4549 option: kernel/traceoff_on_warning
4550
Jiri Kosinafcf4d822012-03-21 16:34:02 -07004551 transparent_hugepage=
4552 [KNL]
4553 Format: [always|madvise|never]
4554 Can be used to control the default behavior of the system
4555 with respect to transparent hugepages.
4556 See Documentation/vm/transhuge.txt for more details.
4557
john stultzd3b8f882009-08-17 16:40:47 -07004558 tsc= Disable clocksource stability checks for TSC.
Alok Kataria395628e2008-10-24 17:22:01 -07004559 Format: <string>
4560 [x86] reliable: mark tsc clocksource as reliable, this
john stultzd3b8f882009-08-17 16:40:47 -07004561 disables clocksource verification at runtime, as well
4562 as the stability checks done at bootup. Used to enable
4563 high-resolution timer mode on older hardware, and in
4564 virtualized environment.
Venkatesh Pallipadie82b8e42010-10-04 17:03:20 -07004565 [x86] noirqtime: Do not use TSC to do irq accounting.
4566 Used to run time disable IRQ_TIME_ACCOUNTING on any
4567 platforms where RDTSC is slow and this accounting
4568 can add overhead.
Alok Kataria395628e2008-10-24 17:22:01 -07004569
Pawan Gupta21127882019-10-23 11:01:53 +02004570 tsx= [X86] Control Transactional Synchronization
4571 Extensions (TSX) feature in Intel processors that
4572 support TSX control.
4573
4574 This parameter controls the TSX feature. The options are:
4575
4576 on - Enable TSX on the system. Although there are
4577 mitigations for all known security vulnerabilities,
4578 TSX has been known to be an accelerator for
4579 several previous speculation-related CVEs, and
4580 so there may be unknown security risks associated
4581 with leaving it enabled.
4582
4583 off - Disable TSX on the system. (Note that this
4584 option takes effect only on newer CPUs which are
4585 not vulnerable to MDS, i.e., have
4586 MSR_IA32_ARCH_CAPABILITIES.MDS_NO=1 and which get
4587 the new IA32_TSX_CTRL MSR through a microcode
4588 update. This new MSR allows for the reliable
4589 deactivation of the TSX functionality.)
4590
Pawan Gupta562afad2019-10-23 12:28:57 +02004591 auto - Disable TSX if X86_BUG_TAA is present,
4592 otherwise enable TSX on the system.
4593
Pawan Gupta21127882019-10-23 11:01:53 +02004594 Not specifying this option is equivalent to tsx=off.
4595
4596 See Documentation/hw-vuln/tsx_async_abort.rst
4597 for more details.
4598
Pawan Guptaba54aad2019-10-23 12:32:55 +02004599 tsx_async_abort= [X86,INTEL] Control mitigation for the TSX Async
4600 Abort (TAA) vulnerability.
4601
4602 Similar to Micro-architectural Data Sampling (MDS)
4603 certain CPUs that support Transactional
4604 Synchronization Extensions (TSX) are vulnerable to an
4605 exploit against CPU internal buffers which can forward
4606 information to a disclosure gadget under certain
4607 conditions.
4608
4609 In vulnerable processors, the speculatively forwarded
4610 data can be used in a cache side channel attack, to
4611 access data to which the attacker does not have direct
4612 access.
4613
4614 This parameter controls the TAA mitigation. The
4615 options are:
4616
4617 full - Enable TAA mitigation on vulnerable CPUs
4618 if TSX is enabled.
4619
4620 full,nosmt - Enable TAA mitigation and disable SMT on
4621 vulnerable CPUs. If TSX is disabled, SMT
4622 is not disabled because CPU is not
4623 vulnerable to cross-thread TAA attacks.
4624 off - Unconditionally disable TAA mitigation
4625
4626 Not specifying this option is equivalent to
4627 tsx_async_abort=full. On CPUs which are MDS affected
4628 and deploy MDS mitigation, TAA mitigation is not
4629 required and doesn't provide any additional
4630 mitigation.
4631
4632 For details see:
4633 Documentation/hw-vuln/tsx_async_abort.rst
4634
Randy Dunlapa9913042005-10-23 12:57:11 -07004635 turbografx.map[2|3]= [HW,JOY]
4636 TurboGraFX parallel port interface
4637 Format:
4638 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7>
Linus Torvalds1da177e2005-04-16 15:20:36 -07004639 See also Documentation/input/joystick-parport.txt
4640
Christian Kujaub6935f82011-05-31 15:22:05 +00004641 udbg-immortal [PPC] When debugging early kernel crashes that
4642 happen after console_init() and before a proper
4643 console driver takes over, this boot options might
4644 help "seeing" what's going on.
4645
Eric Dumazetf86dcc52009-10-07 00:37:59 +00004646 uhash_entries= [KNL,NET]
4647 Set number of hash buckets for UDP/UDP-Lite connections
4648
Alan Stern5f8364b2006-12-05 16:29:55 -05004649 uhci-hcd.ignore_oc=
4650 [USB] Ignore overcurrent events (default N).
4651 Some badly-designed motherboards generate lots of
4652 bogus events, for ports that aren't wired to
4653 anything. Set this parameter to avoid log spamming.
4654 Note that genuine overcurrent events won't be
4655 reported either.
4656
Simon Arlotte3a61b02008-07-19 23:32:54 +01004657 unknown_nmi_panic
Randy Dunlap44a4dcf2011-04-04 15:02:24 -07004658 [X86] Cause panic on unknown NMI.
Simon Arlotte3a61b02008-07-19 23:32:54 +01004659
Carl-Daniel Hailfingerc4fc2342011-05-31 21:31:08 +02004660 usbcore.authorized_default=
4661 [USB] Default USB device authorization:
4662 (default -1 = authorized except for wireless USB,
4663 0 = not authorized, 1 = authorized)
4664
Alan Sternb5e795f2007-02-20 15:00:53 -05004665 usbcore.autosuspend=
4666 [USB] The autosuspend time delay (in seconds) used
4667 for newly-detected USB devices (default 2). This
4668 is the time required before an idle device will be
4669 autosuspended. Devices for which the delay is set
Alan Sterneaafbc32007-03-13 16:39:15 -04004670 to a negative value won't be autosuspended at all.
Alan Sternb5e795f2007-02-20 15:00:53 -05004671
Jaroslav Kyselafd7c5192008-10-10 16:24:45 +02004672 usbcore.usbfs_snoop=
4673 [USB] Set to log all usbfs traffic (default 0 = off).
4674
Alan Stern0290cc92015-11-20 13:53:22 -05004675 usbcore.usbfs_snoop_max=
4676 [USB] Maximum number of bytes to snoop in each URB
4677 (default = 65536).
4678
Jaroslav Kyselafd7c5192008-10-10 16:24:45 +02004679 usbcore.blinkenlights=
4680 [USB] Set to cycle leds on hubs (default 0 = off).
4681
4682 usbcore.old_scheme_first=
4683 [USB] Start with the old device initialization
4684 scheme (default 0 = off).
4685
Alan Stern3f5eb8d2011-11-17 16:41:35 -05004686 usbcore.usbfs_memory_mb=
4687 [USB] Memory limit (in MB) for buffers allocated by
4688 usbfs (default = 16, 0 = max = 2047).
4689
Jaroslav Kyselafd7c5192008-10-10 16:24:45 +02004690 usbcore.use_both_schemes=
4691 [USB] Try the other device initialization scheme
4692 if the first one fails (default 1 = enabled).
4693
4694 usbcore.initial_descriptor_timeout=
4695 [USB] Specifies timeout for the initial 64-byte
4696 USB_REQ_GET_DESCRIPTOR request in milliseconds
4697 (default 5000 = 5.0 seconds).
4698
Oliver Neukum40d58142015-12-03 15:03:32 +01004699 usbcore.nousb [USB] Disable the USB subsystem
4700
Linus Torvalds1da177e2005-04-16 15:20:36 -07004701 usbhid.mousepoll=
4702 [USBHID] The interval which mice are to be polled at.
Randy Dunlapa9913042005-10-23 12:57:11 -07004703
Alan Sternd4f373e2008-11-10 14:07:45 -05004704 usb-storage.delay_use=
4705 [UMS] The delay in seconds before a new device is
Mark Knibbs19101952014-11-04 13:00:15 +00004706 scanned for Logical Units (default 1).
Alan Sternd4f373e2008-11-10 14:07:45 -05004707
4708 usb-storage.quirks=
4709 [UMS] A list of quirks entries to supplement or
4710 override the built-in unusual_devs list. List
4711 entries are separated by commas. Each entry has
4712 the form VID:PID:Flags where VID and PID are Vendor
4713 and Product ID values (4-digit hex numbers) and
4714 Flags is a set of characters, each corresponding
4715 to a common usb-storage quirk flag as follows:
Alan Sternc838ea42008-12-15 10:40:06 -05004716 a = SANE_SENSE (collect more than 18 bytes
4717 of sense data);
Alan Sterna0bb1082009-12-07 16:39:16 -05004718 b = BAD_SENSE (don't collect more than 18
4719 bytes of sense data);
Alan Sternd4f373e2008-11-10 14:07:45 -05004720 c = FIX_CAPACITY (decrease the reported
4721 device capacity by one sector);
Karl Relton51169012011-05-18 21:42:34 +01004722 d = NO_READ_DISC_INFO (don't use
4723 READ_DISC_INFO command);
4724 e = NO_READ_CAPACITY_16 (don't use
4725 READ_CAPACITY_16 command);
Hans de Goede734016b2014-09-16 18:36:52 +02004726 f = NO_REPORT_OPCODES (don't use report opcodes
4727 command, uas only);
Hans de Goedeee136af2015-04-21 11:20:31 +02004728 g = MAX_SECTORS_240 (don't transfer more than
4729 240 sectors at a time, uas only);
Alan Sternc838ea42008-12-15 10:40:06 -05004730 h = CAPACITY_HEURISTICS (decrease the
4731 reported device capacity by one
4732 sector if the number is odd);
Alan Sternd4f373e2008-11-10 14:07:45 -05004733 i = IGNORE_DEVICE (don't bind to this
4734 device);
Hans de Goede13630742016-04-12 12:27:09 +02004735 j = NO_REPORT_LUNS (don't use report luns
4736 command, uas only);
Alan Sternd4f373e2008-11-10 14:07:45 -05004737 l = NOT_LOCKABLE (don't try to lock and
4738 unlock ejectable media);
4739 m = MAX_SECTORS_64 (don't transfer more
4740 than 64 sectors = 32 KB at a time);
Alan Stern21c13a42011-06-07 11:35:52 -04004741 n = INITIAL_READ10 (force a retry of the
4742 initial READ(10) command);
Alan Sternc838ea42008-12-15 10:40:06 -05004743 o = CAPACITY_OK (accept the capacity
4744 reported by the device);
Namjae Jeoneaa05df2012-07-07 23:05:28 -04004745 p = WRITE_CACHE (the device cache is ON
4746 by default);
Alan Sternd4f373e2008-11-10 14:07:45 -05004747 r = IGNORE_RESIDUE (the device reports
4748 bogus residue values);
4749 s = SINGLE_LUN (the device has only one
4750 Logical Unit);
Hans de Goede59307852014-09-15 16:04:12 +02004751 t = NO_ATA_1X (don't allow ATA(12) and ATA(16)
4752 commands, uas only);
Alan Sternb6089f12014-09-02 15:42:18 -04004753 u = IGNORE_UAS (don't bind to the uas driver);
Alan Sternd4f373e2008-11-10 14:07:45 -05004754 w = NO_WP_DETECT (don't test whether the
4755 medium is write-protected).
Oliver Neukum050bc4e2016-09-12 15:19:41 +02004756 y = ALWAYS_SYNC (issue a SYNCHRONIZE_CACHE
4757 even if the device claims no cache)
Alan Sternd4f373e2008-11-10 14:07:45 -05004758 Example: quirks=0419:aaf5:rl,0421:0433:rc
4759
Stephen Boydac1667d2011-08-13 12:34:50 -07004760 user_debug= [KNL,ARM]
4761 Format: <int>
4762 See arch/arm/Kconfig.debug help text.
4763 1 - undefined instruction events
4764 2 - system calls
4765 4 - invalid data aborts
4766 8 - SIGSEGV faults
4767 16 - SIGBUS faults
4768 Example: user_debug=31
4769
Ian Campbell14315592010-02-17 10:38:10 +00004770 userpte=
4771 [X86] Flags controlling user PTE allocations.
4772
4773 nohigh = do not allocate PTE pages in
4774 HIGHMEM regardless of setting
4775 of CONFIG_HIGHPTE.
4776
Jaswinder Singh Rajput6cececf2009-04-14 14:03:43 +05304777 vdso= [X86,SH]
Andy Lutomirskib0b49f22014-03-13 16:01:26 -07004778 On X86_32, this is an alias for vdso32=. Otherwise:
4779
4780 vdso=1: enable VDSO (the default)
Ingo Molnare6e54942006-06-27 02:53:50 -07004781 vdso=0: disable VDSO mapping
4782
Andy Lutomirskib0b49f22014-03-13 16:01:26 -07004783 vdso32= [X86] Control the 32-bit vDSO
4784 vdso32=1: enable 32-bit VDSO
4785 vdso32=0 or vdso32=2: disable 32-bit VDSO
4786
4787 See the help text for CONFIG_COMPAT_VDSO for more
4788 details. If CONFIG_COMPAT_VDSO is set, the default is
4789 vdso32=0; otherwise, the default is vdso32=1.
4790
4791 For compatibility with older kernels, vdso32=2 is an
4792 alias for vdso32=0.
4793
4794 Try vdso32=0 if you encounter an error that says:
4795 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed!
Roland McGrathaf65d642008-01-30 13:30:43 +01004796
Yasuaki Ishimatsud080d392007-07-17 21:22:55 +09004797 vector= [IA-64,SMP]
4798 vector=percpu: enable percpu vector domain
4799
Linus Torvalds1da177e2005-04-16 15:20:36 -07004800 video= [FB] Frame buffer configuration
4801 See Documentation/fb/modedb.txt.
4802
Aaron Lu3afe6da2013-06-20 15:08:55 +08004803 video.brightness_switch_enabled= [0,1]
4804 If set to 1, on receiving an ACPI notify event
4805 generated by hotkey, video driver will adjust brightness
4806 level and then send out the event to user space through
4807 the allocated input device; If set to 0, video driver
4808 will only send out the event without touching backlight
4809 brightness level.
Rafael J. Wysocki28437682014-07-14 19:35:45 +02004810 default: 1
Aaron Lu3afe6da2013-06-20 15:08:55 +08004811
Pawel Moll81a054c2012-05-09 18:30:16 +01004812 virtio_mmio.device=
4813 [VMMIO] Memory mapped virtio (platform) device.
4814
4815 <size>@<baseaddr>:<irq>[:<id>]
4816 where:
4817 <size> := size (can use standard suffixes
4818 like K, M and G)
4819 <baseaddr> := physical base address
4820 <irq> := interrupt number (as passed to
4821 request_irq())
4822 <id> := (optional) platform device id
4823 example:
4824 virtio_mmio.device=1K@0x100b0000:48:7
4825
4826 Can be used multiple times for multiple devices.
4827
Alan Coxcd4f0ef2007-07-31 00:37:59 -07004828 vga= [BOOT,X86-32] Select a particular video mode
Kyle McMartin954a8b82009-02-19 16:48:14 -05004829 See Documentation/x86/boot.txt and
Randy Dunlapa9913042005-10-23 12:57:11 -07004830 Documentation/svga.txt.
Linus Torvalds1da177e2005-04-16 15:20:36 -07004831 Use vga=ask for menu.
4832 This is actually a boot loader parameter; the value is
4833 passed to the kernel using a special protocol.
4834
Randy Dunlapa9913042005-10-23 12:57:11 -07004835 vmalloc=nn[KMG] [KNL,BOOT] Forces the vmalloc area to have an exact
Linus Torvalds1da177e2005-04-16 15:20:36 -07004836 size of <nn>. This can be used to increase the
4837 minimum size (128MB on x86). It can also be used to
4838 decrease the size and leave more room for directly
4839 mapped kernel RAM.
4840
Peter Oberparleiter585c3042006-06-29 15:08:25 +02004841 vmhalt= [KNL,S390] Perform z/VM CP command after system halt.
4842 Format: <command>
Linus Torvalds1da177e2005-04-16 15:20:36 -07004843
Peter Oberparleiter585c3042006-06-29 15:08:25 +02004844 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic.
4845 Format: <command>
4846
4847 vmpoff= [KNL,S390] Perform z/VM CP command after power off.
4848 Format: <command>
Randy Dunlapa9913042005-10-23 12:57:11 -07004849
Andy Lutomirski3ae36652011-08-10 11:15:32 -04004850 vsyscall= [X86-64]
4851 Controls the behavior of vsyscalls (i.e. calls to
4852 fixed addresses of 0xffffffffff600x00 from legacy
4853 code). Most statically-linked binaries and older
4854 versions of glibc use these calls. Because these
4855 functions are at fixed addresses, they make nice
4856 targets for exploits that can control RIP.
4857
Andy Lutomirski2e57ae02011-11-07 16:33:41 -08004858 emulate [default] Vsyscalls turn into traps and are
4859 emulated reasonably safely.
Andy Lutomirski3ae36652011-08-10 11:15:32 -04004860
Andy Lutomirski2e57ae02011-11-07 16:33:41 -08004861 native Vsyscalls are native syscall instructions.
Andy Lutomirski3ae36652011-08-10 11:15:32 -04004862 This is a little bit faster than trapping
4863 and makes a few dynamic recompilers work
4864 better than they would in emulation mode.
4865 It also makes exploits much easier to write.
4866
4867 none Vsyscalls don't work at all. This makes
4868 them quite hard to use for exploits but
4869 might break your system.
4870
Clemens Ladisch3855ae12013-08-04 13:09:50 +02004871 vt.color= [VT] Default text color.
4872 Format: 0xYX, X = foreground, Y = background.
4873 Default: 0x07 = light gray on black.
4874
Clemens Ladisch9ea9a882009-12-15 16:45:39 -08004875 vt.cur_default= [VT] Default cursor shape.
4876 Format: 0xCCBBAA, where AA, BB, and CC are the same as
4877 the parameters of the <Esc>[?A;B;Cc escape sequence;
4878 see VGA-softcursor.txt. Default: 2 = underline.
4879
Randy Dunlap0cb55ad2009-04-05 15:55:22 -07004880 vt.default_blu= [VT]
4881 Format: <blue0>,<blue1>,<blue2>,...,<blue15>
4882 Change the default blue palette of the console.
4883 This is a 16-member array composed of values
4884 ranging from 0-255.
4885
4886 vt.default_grn= [VT]
4887 Format: <green0>,<green1>,<green2>,...,<green15>
4888 Change the default green palette of the console.
4889 This is a 16-member array composed of values
4890 ranging from 0-255.
4891
4892 vt.default_red= [VT]
4893 Format: <red0>,<red1>,<red2>,...,<red15>
4894 Change the default red palette of the console.
4895 This is a 16-member array composed of values
4896 ranging from 0-255.
4897
4898 vt.default_utf8=
4899 [VT]
4900 Format=<0|1>
4901 Set system-wide default UTF-8 mode for all tty's.
4902 Default is 1, i.e. UTF-8 mode is enabled for all
4903 newly opened terminals.
4904
Matthew Garrettf6c06b62009-11-13 15:14:11 -05004905 vt.global_cursor_default=
4906 [VT]
4907 Format=<-1|0|1>
4908 Set system-wide default for whether a cursor
4909 is shown on new VTs. Default is -1,
4910 i.e. cursors will be created by default unless
4911 overridden by individual drivers. 0 will hide
4912 cursors, 1 will display them.
4913
Clemens Ladisch3855ae12013-08-04 13:09:50 +02004914 vt.italic= [VT] Default color for italic text; 0-15.
4915 Default: 2 = green.
4916
4917 vt.underline= [VT] Default color for underlined text; 0-15.
4918 Default: 3 = cyan.
4919
Randy Dunlap4724ba572010-05-03 11:42:52 -07004920 watchdog timers [HW,WDT] For information on watchdog timers,
4921 see Documentation/watchdog/watchdog-parameters.txt
4922 or other driver-specific files in the
4923 Documentation/watchdog/ directory.
Linus Torvalds1da177e2005-04-16 15:20:36 -07004924
Tejun Heo82607adc2015-12-08 11:28:04 -05004925 workqueue.watchdog_thresh=
4926 If CONFIG_WQ_WATCHDOG is configured, workqueue can
4927 warn stall conditions and dump internal state to
4928 help debugging. 0 disables workqueue stall
4929 detection; otherwise, it's the stall threshold
4930 duration in seconds. The default value is 30 and
4931 it can be updated at runtime by writing to the
4932 corresponding sysfs file.
4933
Tejun Heod55262c2013-04-01 11:23:38 -07004934 workqueue.disable_numa
4935 By default, all work items queued to unbound
4936 workqueues are affine to the NUMA nodes they're
4937 issued on, which results in better behavior in
4938 general. If NUMA affinity needs to be disabled for
4939 whatever reason, this option can be used. Note
4940 that this also can be controlled per-workqueue for
4941 workqueues visible under /sys/bus/workqueue/.
4942
Viresh Kumarcee22a12013-04-08 16:45:40 +05304943 workqueue.power_efficient
4944 Per-cpu workqueues are generally preferred because
4945 they show better performance thanks to cache
4946 locality; unfortunately, per-cpu workqueues tend to
4947 be more power hungry than unbound workqueues.
4948
4949 Enabling this makes the per-cpu workqueues which
4950 were observed to contribute significantly to power
4951 consumption unbound, leading to measurably lower
4952 power usage at the cost of small performance
4953 overhead.
4954
4955 The default value of this parameter is determined by
4956 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT.
4957
Tejun Heof303fcc2016-02-09 17:59:38 -05004958 workqueue.debug_force_rr_cpu
4959 Workqueue used to implicitly guarantee that work
4960 items queued without explicit CPU specified are put
4961 on the local CPU. This guarantee is no longer true
4962 and while local CPU is still preferred work items
4963 may be put on foreign CPUs. This debug option
4964 forces round-robin CPU selection to flush out
4965 usages which depend on the now broken guarantee.
4966 When enabled, memory and cache locality will be
4967 impacted.
4968
Randy Dunlap0cb55ad2009-04-05 15:55:22 -07004969 x2apic_phys [X86-64,APIC] Use x2apic physical mode instead of
4970 default x2apic cluster mode on platforms
4971 supporting x2apic.
4972
Kuppuswamy Sathyanarayanan712b6aa2013-10-17 15:35:29 -07004973 x86_intel_mid_timer= [X86-32,APBT]
4974 Choose timer option for x86 Intel MID platform.
Jacob Panbb24c472009-09-02 07:37:17 -07004975 Two valid options are apbt timer only and lapic timer
4976 plus one apbt timer for broadcast timer.
Kuppuswamy Sathyanarayanan712b6aa2013-10-17 15:35:29 -07004977 x86_intel_mid_timer=apbt_only | lapic_and_apbt
Jacob Panbb24c472009-09-02 07:37:17 -07004978
Juergen Grossc70727a2015-07-17 06:51:36 +02004979 xen_512gb_limit [KNL,X86-64,XEN]
4980 Restricts the kernel running paravirtualized under Xen
4981 to use only up to 512 GB of RAM. The reason to do so is
4982 crash analysis tools and Xen tools for doing domain
4983 save/restore/migration must be enabled to handle larger
4984 domains.
4985
Stefano Stabellinic1c54132010-05-14 12:44:30 +01004986 xen_emul_unplug= [HW,X86,XEN]
4987 Unplug Xen emulated devices
4988 Format: [unplug0,][unplug1]
4989 ide-disks -- unplug primary master IDE devices
4990 aux-ide-disks -- unplug non-primary-master IDE devices
4991 nics -- unplug network devices
4992 all -- unplug all emulated devices (NICs and IDE disks)
Ian Campbell1dc7ce92010-08-23 11:59:29 +01004993 unnecessary -- unplugging emulated devices is
4994 unnecessary even if the host did not respond to
4995 the unplug protocol
Ian Campbellc93a4df2010-08-23 11:59:28 +01004996 never -- do not unplug even if version check succeeds
Stefano Stabellinic1c54132010-05-14 12:44:30 +01004997
Konrad Rzeszutek Wilk15a3eac2013-09-25 10:07:20 -04004998 xen_nopvspin [X86,XEN]
4999 Disables the ticketlock slowpath using Xen PV
5000 optimizations.
5001
Konrad Rzeszutek Wilk8d693b92014-07-11 11:51:35 -04005002 xen_nopv [X86]
5003 Disables the PV optimizations forcing the HVM guest to
5004 run as generic HVM guest with no PV drivers.
5005
Linus Torvalds1da177e2005-04-16 15:20:36 -07005006 xirc2ps_cs= [NET,PCMCIA]
Randy Dunlapa9913042005-10-23 12:57:11 -07005007 Format:
5008 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]]
Linus Torvalds1da177e2005-04-16 15:20:36 -07005009
Randy Dunlapa9913042005-10-23 12:57:11 -07005010______________________________________________________________________
Linus Torvalds1da177e2005-04-16 15:20:36 -07005011
5012TODO:
5013
Linus Torvalds1da177e2005-04-16 15:20:36 -07005014 Add more DRM drivers.