1 acpi= [HW,ACPI,X86,ARM64] 2 Advanced Configuration and Power Interface 3 Format: { force | on | off | strict | noirq | rsdt | 4 copy_dsdt } 5 force -- enable ACPI if default was off 6 on -- enable ACPI but allow fallback to DT [arm64] 7 off -- disable ACPI if default was on 8 noirq -- do not use ACPI for IRQ routing 9 strict -- Be less tolerant of platforms that are not 10 strictly ACPI specification compliant. 11 rsdt -- prefer RSDT over (default) XSDT 12 copy_dsdt -- copy DSDT to memory 13 For ARM64, ONLY "acpi=off", "acpi=on" or "acpi=force" 14 are available 15 16 See also Documentation/power/runtime_pm.rst, pci=noacpi 17 18 acpi_apic_instance= [ACPI, IOAPIC] 19 Format: <int> 20 2: use 2nd APIC table, if available 21 1,0: use 1st APIC table 22 default: 0 23 24 acpi_backlight= [HW,ACPI] 25 { vendor | video | native | none } 26 If set to vendor, prefer vendor-specific driver 27 (e.g. thinkpad_acpi, sony_acpi, etc.) instead 28 of the ACPI video.ko driver. 29 If set to video, use the ACPI video.ko driver. 30 If set to native, use the device's native backlight mode. 31 If set to none, disable the ACPI backlight interface. 32 33 acpi_force_32bit_fadt_addr 34 force FADT to use 32 bit addresses rather than the 35 64 bit X_* addresses. Some firmware have broken 64 36 bit addresses for force ACPI ignore these and use 37 the older legacy 32 bit addresses. 38 39 acpica_no_return_repair [HW, ACPI] 40 Disable AML predefined validation mechanism 41 This mechanism can repair the evaluation result to make 42 the return objects more ACPI specification compliant. 43 This option is useful for developers to identify the 44 root cause of an AML interpreter issue when the issue 45 has something to do with the repair mechanism. 46 47 acpi.debug_layer= [HW,ACPI,ACPI_DEBUG] 48 acpi.debug_level= [HW,ACPI,ACPI_DEBUG] 49 Format: <int> 50 CONFIG_ACPI_DEBUG must be enabled to produce any ACPI 51 debug output. Bits in debug_layer correspond to a 52 _COMPONENT in an ACPI source file, e.g., 53 #define _COMPONENT ACPI_PCI_COMPONENT 54 Bits in debug_level correspond to a level in 55 ACPI_DEBUG_PRINT statements, e.g., 56 ACPI_DEBUG_PRINT((ACPI_DB_INFO, ... 57 The debug_level mask defaults to "info". See 58 Documentation/firmware-guide/acpi/debug.rst for more information about 59 debug layers and levels. 60 61 Enable processor driver info messages: 62 acpi.debug_layer=0x20000000 63 Enable PCI/PCI interrupt routing info messages: 64 acpi.debug_layer=0x400000 65 Enable AML "Debug" output, i.e., stores to the Debug 66 object while interpreting AML: 67 acpi.debug_layer=0xffffffff acpi.debug_level=0x2 68 Enable all messages related to ACPI hardware: 69 acpi.debug_layer=0x2 acpi.debug_level=0xffffffff 70 71 Some values produce so much output that the system is 72 unusable. The "log_buf_len" parameter may be useful 73 if you need to capture more output. 74 75 acpi_enforce_resources= [ACPI] 76 { strict | lax | no } 77 Check for resource conflicts between native drivers 78 and ACPI OperationRegions (SystemIO and SystemMemory 79 only). IO ports and memory declared in ACPI might be 80 used by the ACPI subsystem in arbitrary AML code and 81 can interfere with legacy drivers. 82 strict (default): access to resources claimed by ACPI 83 is denied; legacy drivers trying to access reserved 84 resources will fail to bind to device using them. 85 lax: access to resources claimed by ACPI is allowed; 86 legacy drivers trying to access reserved resources 87 will bind successfully but a warning message is logged. 88 no: ACPI OperationRegions are not marked as reserved, 89 no further checks are performed. 90 91 acpi_force_table_verification [HW,ACPI] 92 Enable table checksum verification during early stage. 93 By default, this is disabled due to x86 early mapping 94 size limitation. 95 96 acpi_irq_balance [HW,ACPI] 97 ACPI will balance active IRQs 98 default in APIC mode 99 100 acpi_irq_nobalance [HW,ACPI] 101 ACPI will not move active IRQs (default) 102 default in PIC mode 103 104 acpi_irq_isa= [HW,ACPI] If irq_balance, mark listed IRQs used by ISA 105 Format: <irq>,<irq>... 106 107 acpi_irq_pci= [HW,ACPI] If irq_balance, clear listed IRQs for 108 use by PCI 109 Format: <irq>,<irq>... 110 111 acpi_mask_gpe= [HW,ACPI] 112 Due to the existence of _Lxx/_Exx, some GPEs triggered 113 by unsupported hardware/firmware features can result in 114 GPE floodings that cannot be automatically disabled by 115 the GPE dispatcher. 116 This facility can be used to prevent such uncontrolled 117 GPE floodings. 118 Format: <byte> 119 120 acpi_no_auto_serialize [HW,ACPI] 121 Disable auto-serialization of AML methods 122 AML control methods that contain the opcodes to create 123 named objects will be marked as "Serialized" by the 124 auto-serialization feature. 125 This feature is enabled by default. 126 This option allows to turn off the feature. 127 128 acpi_no_memhotplug [ACPI] Disable memory hotplug. Useful for kdump 129 kernels. 130 131 acpi_no_static_ssdt [HW,ACPI] 132 Disable installation of static SSDTs at early boot time 133 By default, SSDTs contained in the RSDT/XSDT will be 134 installed automatically and they will appear under 135 /sys/firmware/acpi/tables. 136 This option turns off this feature. 137 Note that specifying this option does not affect 138 dynamic table installation which will install SSDT 139 tables to /sys/firmware/acpi/tables/dynamic. 140 141 acpi_no_watchdog [HW,ACPI,WDT] 142 Ignore the ACPI-based watchdog interface (WDAT) and let 143 a native driver control the watchdog device instead. 144 145 acpi_rsdp= [ACPI,EFI,KEXEC] 146 Pass the RSDP address to the kernel, mostly used 147 on machines running EFI runtime service to boot the 148 second kernel for kdump. 149 150 acpi_os_name= [HW,ACPI] Tell ACPI BIOS the name of the OS 151 Format: To spoof as Windows 98: ="Microsoft Windows" 152 153 acpi_rev_override [ACPI] Override the _REV object to return 5 (instead 154 of 2 which is mandated by ACPI 6) as the supported ACPI 155 specification revision (when using this switch, it may 156 be necessary to carry out a cold reboot _twice_ in a 157 row to make it take effect on the platform firmware). 158 159 acpi_osi= [HW,ACPI] Modify list of supported OS interface strings 160 acpi_osi="string1" # add string1 161 acpi_osi="!string2" # remove string2 162 acpi_osi=!* # remove all strings 163 acpi_osi=! # disable all built-in OS vendor 164 strings 165 acpi_osi=!! # enable all built-in OS vendor 166 strings 167 acpi_osi= # disable all strings 168 169 'acpi_osi=!' can be used in combination with single or 170 multiple 'acpi_osi="string1"' to support specific OS 171 vendor string(s). Note that such command can only 172 affect the default state of the OS vendor strings, thus 173 it cannot affect the default state of the feature group 174 strings and the current state of the OS vendor strings, 175 specifying it multiple times through kernel command line 176 is meaningless. This command is useful when one do not 177 care about the state of the feature group strings which 178 should be controlled by the OSPM. 179 Examples: 180 1. 'acpi_osi=! acpi_osi="Windows 2000"' is equivalent 181 to 'acpi_osi="Windows 2000" acpi_osi=!', they all 182 can make '_OSI("Windows 2000")' TRUE. 183 184 'acpi_osi=' cannot be used in combination with other 185 'acpi_osi=' command lines, the _OSI method will not 186 exist in the ACPI namespace. NOTE that such command can 187 only affect the _OSI support state, thus specifying it 188 multiple times through kernel command line is also 189 meaningless. 190 Examples: 191 1. 'acpi_osi=' can make 'CondRefOf(_OSI, Local1)' 192 FALSE. 193 194 'acpi_osi=!*' can be used in combination with single or 195 multiple 'acpi_osi="string1"' to support specific 196 string(s). Note that such command can affect the 197 current state of both the OS vendor strings and the 198 feature group strings, thus specifying it multiple times 199 through kernel command line is meaningful. But it may 200 still not able to affect the final state of a string if 201 there are quirks related to this string. This command 202 is useful when one want to control the state of the 203 feature group strings to debug BIOS issues related to 204 the OSPM features. 205 Examples: 206 1. 'acpi_osi="Module Device" acpi_osi=!*' can make 207 '_OSI("Module Device")' FALSE. 208 2. 'acpi_osi=!* acpi_osi="Module Device"' can make 209 '_OSI("Module Device")' TRUE. 210 3. 'acpi_osi=! acpi_osi=!* acpi_osi="Windows 2000"' is 211 equivalent to 212 'acpi_osi=!* acpi_osi=! acpi_osi="Windows 2000"' 213 and 214 'acpi_osi=!* acpi_osi="Windows 2000" acpi_osi=!', 215 they all will make '_OSI("Windows 2000")' TRUE. 216 217 acpi_pm_good [X86] 218 Override the pmtimer bug detection: force the kernel 219 to assume that this machine's pmtimer latches its value 220 and always returns good values. 221 222 acpi_sci= [HW,ACPI] ACPI System Control Interrupt trigger mode 223 Format: { level | edge | high | low } 224 225 acpi_skip_timer_override [HW,ACPI] 226 Recognize and ignore IRQ0/pin2 Interrupt Override. 227 For broken nForce2 BIOS resulting in XT-PIC timer. 228 229 acpi_sleep= [HW,ACPI] Sleep options 230 Format: { s3_bios, s3_mode, s3_beep, s4_nohwsig, 231 old_ordering, nonvs, sci_force_enable, nobl } 232 See Documentation/power/video.rst for information on 233 s3_bios and s3_mode. 234 s3_beep is for debugging; it makes the PC's speaker beep 235 as soon as the kernel's real-mode entry point is called. 236 s4_nohwsig prevents ACPI hardware signature from being 237 used during resume from hibernation. 238 old_ordering causes the ACPI 1.0 ordering of the _PTS 239 control method, with respect to putting devices into 240 low power states, to be enforced (the ACPI 2.0 ordering 241 of _PTS is used by default). 242 nonvs prevents the kernel from saving/restoring the 243 ACPI NVS memory during suspend/hibernation and resume. 244 sci_force_enable causes the kernel to set SCI_EN directly 245 on resume from S1/S3 (which is against the ACPI spec, 246 but some broken systems don't work without it). 247 nobl causes the internal blacklist of systems known to 248 behave incorrectly in some ways with respect to system 249 suspend and resume to be ignored (use wisely). 250 251 acpi_use_timer_override [HW,ACPI] 252 Use timer override. For some broken Nvidia NF5 boards 253 that require a timer override, but don't have HPET 254 255 add_efi_memmap [EFI; X86] Include EFI memory map in 256 kernel's map of available physical RAM. 257 258 agp= [AGP] 259 { off | try_unsupported } 260 off: disable AGP support 261 try_unsupported: try to drive unsupported chipsets 262 (may crash computer or cause data corruption) 263 264 ALSA [HW,ALSA] 265 See Documentation/sound/alsa-configuration.rst 266 267 alignment= [KNL,ARM] 268 Allow the default userspace alignment fault handler 269 behaviour to be specified. Bit 0 enables warnings, 270 bit 1 enables fixups, and bit 2 sends a segfault. 271 272 align_va_addr= [X86-64] 273 Align virtual addresses by clearing slice [14:12] when 274 allocating a VMA at process creation time. This option 275 gives you up to 3% performance improvement on AMD F15h 276 machines (where it is enabled by default) for a 277 CPU-intensive style benchmark, and it can vary highly in 278 a microbenchmark depending on workload and compiler. 279 280 32: only for 32-bit processes 281 64: only for 64-bit processes 282 on: enable for both 32- and 64-bit processes 283 off: disable for both 32- and 64-bit processes 284 285 alloc_snapshot [FTRACE] 286 Allocate the ftrace snapshot buffer on boot up when the 287 main buffer is allocated. This is handy if debugging 288 and you need to use tracing_snapshot() on boot up, and 289 do not want to use tracing_snapshot_alloc() as it needs 290 to be done where GFP_KERNEL allocations are allowed. 291 292 amd_iommu= [HW,X86-64] 293 Pass parameters to the AMD IOMMU driver in the system. 294 Possible values are: 295 fullflush - enable flushing of IO/TLB entries when 296 they are unmapped. Otherwise they are 297 flushed before they will be reused, which 298 is a lot of faster 299 off - do not initialize any AMD IOMMU found in 300 the system 301 force_isolation - Force device isolation for all 302 devices. The IOMMU driver is not 303 allowed anymore to lift isolation 304 requirements as needed. This option 305 does not override iommu=pt 306 307 amd_iommu_dump= [HW,X86-64] 308 Enable AMD IOMMU driver option to dump the ACPI table 309 for AMD IOMMU. With this option enabled, AMD IOMMU 310 driver will print ACPI tables for AMD IOMMU during 311 IOMMU initialization. 312 313 amd_iommu_intr= [HW,X86-64] 314 Specifies one of the following AMD IOMMU interrupt 315 remapping modes: 316 legacy - Use legacy interrupt remapping mode. 317 vapic - Use virtual APIC mode, which allows IOMMU 318 to inject interrupts directly into guest. 319 This mode requires kvm-amd.avic=1. 320 (Default when IOMMU HW support is present.) 321 322 amijoy.map= [HW,JOY] Amiga joystick support 323 Map of devices attached to JOY0DAT and JOY1DAT 324 Format: <a>,<b> 325 See also Documentation/input/joydev/joystick.rst 326 327 analog.map= [HW,JOY] Analog joystick and gamepad support 328 Specifies type or capabilities of an analog joystick 329 connected to one of 16 gameports 330 Format: <type1>,<type2>,..<type16> 331 332 apc= [HW,SPARC] 333 Power management functions (SPARCstation-4/5 + deriv.) 334 Format: noidle 335 Disable APC CPU standby support. SPARCstation-Fox does 336 not play well with APC CPU idle - disable it if you have 337 APC and your system crashes randomly. 338 339 apic= [APIC,X86] Advanced Programmable Interrupt Controller 340 Change the output verbosity while booting 341 Format: { quiet (default) | verbose | debug } 342 Change the amount of debugging information output 343 when initialising the APIC and IO-APIC components. 344 For X86-32, this can also be used to specify an APIC 345 driver name. 346 Format: apic=driver_name 347 Examples: apic=bigsmp 348 349 apic_extnmi= [APIC,X86] External NMI delivery setting 350 Format: { bsp (default) | all | none } 351 bsp: External NMI is delivered only to CPU 0 352 all: External NMIs are broadcast to all CPUs as a 353 backup of CPU 0 354 none: External NMI is masked for all CPUs. This is 355 useful so that a dump capture kernel won't be 356 shot down by NMI 357 358 autoconf= [IPV6] 359 See Documentation/networking/ipv6.rst. 360 361 show_lapic= [APIC,X86] Advanced Programmable Interrupt Controller 362 Limit apic dumping. The parameter defines the maximal 363 number of local apics being dumped. Also it is possible 364 to set it to "all" by meaning -- no limit here. 365 Format: { 1 (default) | 2 | ... | all }. 366 The parameter valid if only apic=debug or 367 apic=verbose is specified. 368 Example: apic=debug show_lapic=all 369 370 apm= [APM] Advanced Power Management 371 See header of arch/x86/kernel/apm_32.c. 372 373 arcrimi= [HW,NET] ARCnet - "RIM I" (entirely mem-mapped) cards 374 Format: <io>,<irq>,<nodeID> 375 376 ataflop= [HW,M68k] 377 378 atarimouse= [HW,MOUSE] Atari Mouse 379 380 atkbd.extra= [HW] Enable extra LEDs and keys on IBM RapidAccess, 381 EzKey and similar keyboards 382 383 atkbd.reset= [HW] Reset keyboard during initialization 384 385 atkbd.set= [HW] Select keyboard code set 386 Format: <int> (2 = AT (default), 3 = PS/2) 387 388 atkbd.scroll= [HW] Enable scroll wheel on MS Office and similar 389 keyboards 390 391 atkbd.softraw= [HW] Choose between synthetic and real raw mode 392 Format: <bool> (0 = real, 1 = synthetic (default)) 393 394 atkbd.softrepeat= [HW] 395 Use software keyboard repeat 396 397 audit= [KNL] Enable the audit sub-system 398 Format: { "0" | "1" | "off" | "on" } 399 0 | off - kernel audit is disabled and can not be 400 enabled until the next reboot 401 unset - kernel audit is initialized but disabled and 402 will be fully enabled by the userspace auditd. 403 1 | on - kernel audit is initialized and partially 404 enabled, storing at most audit_backlog_limit 405 messages in RAM until it is fully enabled by the 406 userspace auditd. 407 Default: unset 408 409 audit_backlog_limit= [KNL] Set the audit queue size limit. 410 Format: <int> (must be >=0) 411 Default: 64 412 413 bau= [X86_UV] Enable the BAU on SGI UV. The default 414 behavior is to disable the BAU (i.e. bau=0). 415 Format: { "0" | "1" } 416 0 - Disable the BAU. 417 1 - Enable the BAU. 418 unset - Disable the BAU. 419 420 baycom_epp= [HW,AX25] 421 Format: <io>,<mode> 422 423 baycom_par= [HW,AX25] BayCom Parallel Port AX.25 Modem 424 Format: <io>,<mode> 425 See header of drivers/net/hamradio/baycom_par.c. 426 427 baycom_ser_fdx= [HW,AX25] 428 BayCom Serial Port AX.25 Modem (Full Duplex Mode) 429 Format: <io>,<irq>,<mode>[,<baud>] 430 See header of drivers/net/hamradio/baycom_ser_fdx.c. 431 432 baycom_ser_hdx= [HW,AX25] 433 BayCom Serial Port AX.25 Modem (Half Duplex Mode) 434 Format: <io>,<irq>,<mode> 435 See header of drivers/net/hamradio/baycom_ser_hdx.c. 436 437 blkdevparts= Manual partition parsing of block device(s) for 438 embedded devices based on command line input. 439 See Documentation/block/cmdline-partition.rst 440 441 boot_delay= Milliseconds to delay each printk during boot. 442 Values larger than 10 seconds (10000) are changed to 443 no delay (0). 444 Format: integer 445 446 bootconfig [KNL] 447 Extended command line options can be added to an initrd 448 and this will cause the kernel to look for it. 449 450 See Documentation/admin-guide/bootconfig.rst 451 452 bert_disable [ACPI] 453 Disable BERT OS support on buggy BIOSes. 454 455 bgrt_disable [ACPI][X86] 456 Disable BGRT to avoid flickering OEM logo. 457 458 bttv.card= [HW,V4L] bttv (bt848 + bt878 based grabber cards) 459 bttv.radio= Most important insmod options are available as 460 kernel args too. 461 bttv.pll= See Documentation/admin-guide/media/bttv.rst 462 bttv.tuner= 463 464 bulk_remove=off [PPC] This parameter disables the use of the pSeries 465 firmware feature for flushing multiple hpte entries 466 at a time. 467 468 c101= [NET] Moxa C101 synchronous serial card 469 470 cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection. 471 Sometimes CPU hardware bugs make them report the cache 472 size incorrectly. The kernel will attempt work arounds 473 to fix known problems, but for some CPUs it is not 474 possible to determine what the correct size should be. 475 This option provides an override for these situations. 476 477 carrier_timeout= 478 [NET] Specifies amount of time (in seconds) that 479 the kernel should wait for a network carrier. By default 480 it waits 120 seconds. 481 482 ca_keys= [KEYS] This parameter identifies a specific key(s) on 483 the system trusted keyring to be used for certificate 484 trust validation. 485 format: { id:<keyid> | builtin } 486 487 cca= [MIPS] Override the kernel pages' cache coherency 488 algorithm. Accepted values range from 0 to 7 489 inclusive. See arch/mips/include/asm/pgtable-bits.h 490 for platform specific values (SB1, Loongson3 and 491 others). 492 493 ccw_timeout_log [S390] 494 See Documentation/s390/common_io.rst for details. 495 496 cgroup_disable= [KNL] Disable a particular controller 497 Format: {name of the controller(s) to disable} 498 The effects of cgroup_disable=foo are: 499 - foo isn't auto-mounted if you mount all cgroups in 500 a single hierarchy 501 - foo isn't visible as an individually mountable 502 subsystem 503 {Currently only "memory" controller deal with this and 504 cut the overhead, others just disable the usage. So 505 only cgroup_disable=memory is actually worthy} 506 507 cgroup_no_v1= [KNL] Disable cgroup controllers and named hierarchies in v1 508 Format: { { controller | "all" | "named" } 509 [,{ controller | "all" | "named" }...] } 510 Like cgroup_disable, but only applies to cgroup v1; 511 the blacklisted controllers remain available in cgroup2. 512 "all" blacklists all controllers and "named" disables 513 named mounts. Specifying both "all" and "named" disables 514 all v1 hierarchies. 515 516 cgroup.memory= [KNL] Pass options to the cgroup memory controller. 517 Format: <string> 518 nosocket -- Disable socket memory accounting. 519 nokmem -- Disable kernel memory accounting. 520 521 checkreqprot [SELINUX] Set initial checkreqprot flag value. 522 Format: { "0" | "1" } 523 See security/selinux/Kconfig help text. 524 0 -- check protection applied by kernel (includes 525 any implied execute protection). 526 1 -- check protection requested by application. 527 Default value is set via a kernel config option. 528 Value can be changed at runtime via 529 /sys/fs/selinux/checkreqprot. 530 Setting checkreqprot to 1 is deprecated. 531 532 cio_ignore= [S390] 533 See Documentation/s390/common_io.rst for details. 534 clk_ignore_unused 535 [CLK] 536 Prevents the clock framework from automatically gating 537 clocks that have not been explicitly enabled by a Linux 538 device driver but are enabled in hardware at reset or 539 by the bootloader/firmware. Note that this does not 540 force such clocks to be always-on nor does it reserve 541 those clocks in any way. This parameter is useful for 542 debug and development, but should not be needed on a 543 platform with proper driver support. For more 544 information, see Documentation/driver-api/clk.rst. 545 546 clock= [BUGS=X86-32, HW] gettimeofday clocksource override. 547 [Deprecated] 548 Forces specified clocksource (if available) to be used 549 when calculating gettimeofday(). If specified 550 clocksource is not available, it defaults to PIT. 551 Format: { pit | tsc | cyclone | pmtmr } 552 553 clocksource= Override the default clocksource 554 Format: <string> 555 Override the default clocksource and use the clocksource 556 with the name specified. 557 Some clocksource names to choose from, depending on 558 the platform: 559 [all] jiffies (this is the base, fallback clocksource) 560 [ACPI] acpi_pm 561 [ARM] imx_timer1,OSTS,netx_timer,mpu_timer2, 562 pxa_timer,timer3,32k_counter,timer0_1 563 [X86-32] pit,hpet,tsc; 564 scx200_hrt on Geode; cyclone on IBM x440 565 [MIPS] MIPS 566 [PARISC] cr16 567 [S390] tod 568 [SH] SuperH 569 [SPARC64] tick 570 [X86-64] hpet,tsc 571 572 clocksource.arm_arch_timer.evtstrm= 573 [ARM,ARM64] 574 Format: <bool> 575 Enable/disable the eventstream feature of the ARM 576 architected timer so that code using WFE-based polling 577 loops can be debugged more effectively on production 578 systems. 579 580 clearcpuid=BITNUM [X86] 581 Disable CPUID feature X for the kernel. See 582 arch/x86/include/asm/cpufeatures.h for the valid bit 583 numbers. Note the Linux specific bits are not necessarily 584 stable over kernel options, but the vendor specific 585 ones should be. 586 Also note that user programs calling CPUID directly 587 or using the feature without checking anything 588 will still see it. This just prevents it from 589 being used by the kernel or shown in /proc/cpuinfo. 590 Also note the kernel might malfunction if you disable 591 some critical bits. 592 593 cma=nn[MG]@[start[MG][-end[MG]]] 594 [ARM,X86,KNL] 595 Sets the size of kernel global memory area for 596 contiguous memory allocations and optionally the 597 placement constraint by the physical address range of 598 memory allocations. A value of 0 disables CMA 599 altogether. For more information, see 600 include/linux/dma-contiguous.h 601 602 cmo_free_hint= [PPC] Format: { yes | no } 603 Specify whether pages are marked as being inactive 604 when they are freed. This is used in CMO environments 605 to determine OS memory pressure for page stealing by 606 a hypervisor. 607 Default: yes 608 609 coherent_pool=nn[KMG] [ARM,KNL] 610 Sets the size of memory pool for coherent, atomic dma 611 allocations, by default set to 256K. 612 613 com20020= [HW,NET] ARCnet - COM20020 chipset 614 Format: 615 <io>[,<irq>[,<nodeID>[,<backplane>[,<ckp>[,<timeout>]]]]] 616 617 com90io= [HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers) 618 Format: <io>[,<irq>] 619 620 com90xx= [HW,NET] 621 ARCnet - COM90xx chipset (memory-mapped buffers) 622 Format: <io>[,<irq>[,<memstart>]] 623 624 condev= [HW,S390] console device 625 conmode= 626 627 console= [KNL] Output console device and options. 628 629 tty<n> Use the virtual console device <n>. 630 631 ttyS<n>[,options] 632 ttyUSB0[,options] 633 Use the specified serial port. The options are of 634 the form "bbbbpnf", where "bbbb" is the baud rate, 635 "p" is parity ("n", "o", or "e"), "n" is number of 636 bits, and "f" is flow control ("r" for RTS or 637 omit it). Default is "9600n8". 638 639 See Documentation/admin-guide/serial-console.rst for more 640 information. See 641 Documentation/networking/netconsole.rst for an 642 alternative. 643 644 uart[8250],io,<addr>[,options] 645 uart[8250],mmio,<addr>[,options] 646 uart[8250],mmio16,<addr>[,options] 647 uart[8250],mmio32,<addr>[,options] 648 uart[8250],0x<addr>[,options] 649 Start an early, polled-mode console on the 8250/16550 650 UART at the specified I/O port or MMIO address, 651 switching to the matching ttyS device later. 652 MMIO inter-register address stride is either 8-bit 653 (mmio), 16-bit (mmio16), or 32-bit (mmio32). 654 If none of [io|mmio|mmio16|mmio32], <addr> is assumed 655 to be equivalent to 'mmio'. 'options' are specified in 656 the same format described for ttyS above; if unspecified, 657 the h/w is not re-initialized. 658 659 hvc<n> Use the hypervisor console device <n>. This is for 660 both Xen and PowerPC hypervisors. 661 662 If the device connected to the port is not a TTY but a braille 663 device, prepend "brl," before the device type, for instance 664 console=brl,ttyS0 665 For now, only VisioBraille is supported. 666 667 console_msg_format= 668 [KNL] Change console messages format 669 default 670 By default we print messages on consoles in 671 "[time stamp] text\n" format (time stamp may not be 672 printed, depending on CONFIG_PRINTK_TIME or 673 `printk_time' param). 674 syslog 675 Switch to syslog format: "<%u>[time stamp] text\n" 676 IOW, each message will have a facility and loglevel 677 prefix. The format is similar to one used by syslog() 678 syscall, or to executing "dmesg -S --raw" or to reading 679 from /proc/kmsg. 680 681 consoleblank= [KNL] The console blank (screen saver) timeout in 682 seconds. A value of 0 disables the blank timer. 683 Defaults to 0. 684 685 coredump_filter= 686 [KNL] Change the default value for 687 /proc/<pid>/coredump_filter. 688 See also Documentation/filesystems/proc.rst. 689 690 coresight_cpu_debug.enable 691 [ARM,ARM64] 692 Format: <bool> 693 Enable/disable the CPU sampling based debugging. 694 0: default value, disable debugging 695 1: enable debugging at boot time 696 697 cpuidle.off=1 [CPU_IDLE] 698 disable the cpuidle sub-system 699 700 cpuidle.governor= 701 [CPU_IDLE] Name of the cpuidle governor to use. 702 703 cpufreq.off=1 [CPU_FREQ] 704 disable the cpufreq sub-system 705 706 cpu_init_udelay=N 707 [X86] Delay for N microsec between assert and de-assert 708 of APIC INIT to start processors. This delay occurs 709 on every CPU online, such as boot, and resume from suspend. 710 Default: 10000 711 712 cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver 713 Format: 714 <first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>] 715 716 crashkernel=size[KMG][@offset[KMG]] 717 [KNL] Using kexec, Linux can switch to a 'crash kernel' 718 upon panic. This parameter reserves the physical 719 memory region [offset, offset + size] for that kernel 720 image. If '@offset' is omitted, then a suitable offset 721 is selected automatically. 722 [KNL, x86_64] select a region under 4G first, and 723 fall back to reserve region above 4G when '@offset' 724 hasn't been specified. 725 See Documentation/admin-guide/kdump/kdump.rst for further details. 726 727 crashkernel=range1:size1[,range2:size2,...][@offset] 728 [KNL] Same as above, but depends on the memory 729 in the running system. The syntax of range is 730 start-[end] where start and end are both 731 a memory unit (amount[KMG]). See also 732 Documentation/admin-guide/kdump/kdump.rst for an example. 733 734 crashkernel=size[KMG],high 735 [KNL, x86_64] range could be above 4G. Allow kernel 736 to allocate physical memory region from top, so could 737 be above 4G if system have more than 4G ram installed. 738 Otherwise memory region will be allocated below 4G, if 739 available. 740 It will be ignored if crashkernel=X is specified. 741 crashkernel=size[KMG],low 742 [KNL, x86_64] range under 4G. When crashkernel=X,high 743 is passed, kernel could allocate physical memory region 744 above 4G, that cause second kernel crash on system 745 that require some amount of low memory, e.g. swiotlb 746 requires at least 64M+32K low memory, also enough extra 747 low memory is needed to make sure DMA buffers for 32-bit 748 devices won't run out. Kernel would try to allocate at 749 at least 256M below 4G automatically. 750 This one let user to specify own low range under 4G 751 for second kernel instead. 752 0: to disable low allocation. 753 It will be ignored when crashkernel=X,high is not used 754 or memory reserved is below 4G. 755 756 cryptomgr.notests 757 [KNL] Disable crypto self-tests 758 759 cs89x0_dma= [HW,NET] 760 Format: <dma> 761 762 cs89x0_media= [HW,NET] 763 Format: { rj45 | aui | bnc } 764 765 dasd= [HW,NET] 766 See header of drivers/s390/block/dasd_devmap.c. 767 768 db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port 769 (one device per port) 770 Format: <port#>,<type> 771 See also Documentation/input/devices/joystick-parport.rst 772 773 ddebug_query= [KNL,DYNAMIC_DEBUG] Enable debug messages at early boot 774 time. See 775 Documentation/admin-guide/dynamic-debug-howto.rst for 776 details. Deprecated, see dyndbg. 777 778 debug [KNL] Enable kernel debugging (events log level). 779 780 debug_boot_weak_hash 781 [KNL] Enable printing [hashed] pointers early in the 782 boot sequence. If enabled, we use a weak hash instead 783 of siphash to hash pointers. Use this option if you are 784 seeing instances of '(___ptrval___)') and need to see a 785 value (hashed pointer) instead. Cryptographically 786 insecure, please do not use on production kernels. 787 788 debug_locks_verbose= 789 [KNL] verbose self-tests 790 Format=<0|1> 791 Print debugging info while doing the locking API 792 self-tests. 793 We default to 0 (no extra messages), setting it to 794 1 will print _a lot_ more information - normally 795 only useful to kernel developers. 796 797 debug_objects [KNL] Enable object debugging 798 799 no_debug_objects 800 [KNL] Disable object debugging 801 802 debug_guardpage_minorder= 803 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this 804 parameter allows control of the order of pages that will 805 be intentionally kept free (and hence protected) by the 806 buddy allocator. Bigger value increase the probability 807 of catching random memory corruption, but reduce the 808 amount of memory for normal system use. The maximum 809 possible value is MAX_ORDER/2. Setting this parameter 810 to 1 or 2 should be enough to identify most random 811 memory corruption problems caused by bugs in kernel or 812 driver code when a CPU writes to (or reads from) a 813 random memory location. Note that there exists a class 814 of memory corruptions problems caused by buggy H/W or 815 F/W or by drivers badly programing DMA (basically when 816 memory is written at bus level and the CPU MMU is 817 bypassed) which are not detectable by 818 CONFIG_DEBUG_PAGEALLOC, hence this option will not help 819 tracking down these problems. 820 821 debug_pagealloc= 822 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this parameter 823 enables the feature at boot time. By default, it is 824 disabled and the system will work mostly the same as a 825 kernel built without CONFIG_DEBUG_PAGEALLOC. 826 Note: to get most of debug_pagealloc error reports, it's 827 useful to also enable the page_owner functionality. 828 on: enable the feature 829 830 debugpat [X86] Enable PAT debugging 831 832 decnet.addr= [HW,NET] 833 Format: <area>[,<node>] 834 See also Documentation/networking/decnet.rst. 835 836 default_hugepagesz= 837 [HW] The size of the default HugeTLB page. This is 838 the size represented by the legacy /proc/ hugepages 839 APIs. In addition, this is the default hugetlb size 840 used for shmget(), mmap() and mounting hugetlbfs 841 filesystems. If not specified, defaults to the 842 architecture's default huge page size. Huge page 843 sizes are architecture dependent. See also 844 Documentation/admin-guide/mm/hugetlbpage.rst. 845 Format: size[KMG] 846 847 deferred_probe_timeout= 848 [KNL] Debugging option to set a timeout in seconds for 849 deferred probe to give up waiting on dependencies to 850 probe. Only specific dependencies (subsystems or 851 drivers) that have opted in will be ignored. A timeout of 0 852 will timeout at the end of initcalls. This option will also 853 dump out devices still on the deferred probe list after 854 retrying. 855 856 dfltcc= [HW,S390] 857 Format: { on | off | def_only | inf_only | always } 858 on: s390 zlib hardware support for compression on 859 level 1 and decompression (default) 860 off: No s390 zlib hardware support 861 def_only: s390 zlib hardware support for deflate 862 only (compression on level 1) 863 inf_only: s390 zlib hardware support for inflate 864 only (decompression) 865 always: Same as 'on' but ignores the selected compression 866 level always using hardware support (used for debugging) 867 868 dhash_entries= [KNL] 869 Set number of hash buckets for dentry cache. 870 871 disable_1tb_segments [PPC] 872 Disables the use of 1TB hash page table segments. This 873 causes the kernel to fall back to 256MB segments which 874 can be useful when debugging issues that require an SLB 875 miss to occur. 876 877 stress_slb [PPC] 878 Limits the number of kernel SLB entries, and flushes 879 them frequently to increase the rate of SLB faults 880 on kernel addresses. 881 882 disable= [IPV6] 883 See Documentation/networking/ipv6.rst. 884 885 hardened_usercopy= 886 [KNL] Under CONFIG_HARDENED_USERCOPY, whether 887 hardening is enabled for this boot. Hardened 888 usercopy checking is used to protect the kernel 889 from reading or writing beyond known memory 890 allocation boundaries as a proactive defense 891 against bounds-checking flaws in the kernel's 892 copy_to_user()/copy_from_user() interface. 893 on Perform hardened usercopy checks (default). 894 off Disable hardened usercopy checks. 895 896 disable_radix [PPC] 897 Disable RADIX MMU mode on POWER9 898 899 disable_tlbie [PPC] 900 Disable TLBIE instruction. Currently does not work 901 with KVM, with HASH MMU, or with coherent accelerators. 902 903 disable_cpu_apicid= [X86,APIC,SMP] 904 Format: <int> 905 The number of initial APIC ID for the 906 corresponding CPU to be disabled at boot, 907 mostly used for the kdump 2nd kernel to 908 disable BSP to wake up multiple CPUs without 909 causing system reset or hang due to sending 910 INIT from AP to BSP. 911 912 perf_v4_pmi= [X86,INTEL] 913 Format: <bool> 914 Disable Intel PMU counter freezing feature. 915 The feature only exists starting from 916 Arch Perfmon v4 (Skylake and newer). 917 918 disable_ddw [PPC/PSERIES] 919 Disable Dynamic DMA Window support. Use this if 920 to workaround buggy firmware. 921 922 disable_ipv6= [IPV6] 923 See Documentation/networking/ipv6.rst. 924 925 disable_mtrr_cleanup [X86] 926 The kernel tries to adjust MTRR layout from continuous 927 to discrete, to make X server driver able to add WB 928 entry later. This parameter disables that. 929 930 disable_mtrr_trim [X86, Intel and AMD only] 931 By default the kernel will trim any uncacheable 932 memory out of your available memory pool based on 933 MTRR settings. This parameter disables that behavior, 934 possibly causing your machine to run very slowly. 935 936 disable_timer_pin_1 [X86] 937 Disable PIN 1 of APIC timer 938 Can be useful to work around chipset bugs. 939 940 dis_ucode_ldr [X86] Disable the microcode loader. 941 942 dma_debug=off If the kernel is compiled with DMA_API_DEBUG support, 943 this option disables the debugging code at boot. 944 945 dma_debug_entries=<number> 946 This option allows to tune the number of preallocated 947 entries for DMA-API debugging code. One entry is 948 required per DMA-API allocation. Use this if the 949 DMA-API debugging code disables itself because the 950 architectural default is too low. 951 952 dma_debug_driver=<driver_name> 953 With this option the DMA-API debugging driver 954 filter feature can be enabled at boot time. Just 955 pass the driver to filter for as the parameter. 956 The filter can be disabled or changed to another 957 driver later using sysfs. 958 959 driver_async_probe= [KNL] 960 List of driver names to be probed asynchronously. 961 Format: <driver_name1>,<driver_name2>... 962 963 drm.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>] 964 Broken monitors, graphic adapters, KVMs and EDIDless 965 panels may send no or incorrect EDID data sets. 966 This parameter allows to specify an EDID data sets 967 in the /lib/firmware directory that are used instead. 968 Generic built-in EDID data sets are used, if one of 969 edid/1024x768.bin, edid/1280x1024.bin, 970 edid/1680x1050.bin, or edid/1920x1080.bin is given 971 and no file with the same name exists. Details and 972 instructions how to build your own EDID data are 973 available in Documentation/admin-guide/edid.rst. An EDID 974 data set will only be used for a particular connector, 975 if its name and a colon are prepended to the EDID 976 name. Each connector may use a unique EDID data 977 set by separating the files with a comma. An EDID 978 data set with no connector name will be used for 979 any connectors not explicitly specified. 980 981 dscc4.setup= [NET] 982 983 dt_cpu_ftrs= [PPC] 984 Format: {"off" | "known"} 985 Control how the dt_cpu_ftrs device-tree binding is 986 used for CPU feature discovery and setup (if it 987 exists). 988 off: Do not use it, fall back to legacy cpu table. 989 known: Do not pass through unknown features to guests 990 or userspace, only those that the kernel is aware of. 991 992 dump_apple_properties [X86] 993 Dump name and content of EFI device properties on 994 x86 Macs. Useful for driver authors to determine 995 what data is available or for reverse-engineering. 996 997 dyndbg[="val"] [KNL,DYNAMIC_DEBUG] 998 module.dyndbg[="val"] 999 Enable debug messages at boot time. See 1000 Documentation/admin-guide/dynamic-debug-howto.rst 1001 for details. 1002 1003 nopku [X86] Disable Memory Protection Keys CPU feature found 1004 in some Intel CPUs. 1005 1006 module.async_probe [KNL] 1007 Enable asynchronous probe on this module. 1008 1009 early_ioremap_debug [KNL] 1010 Enable debug messages in early_ioremap support. This 1011 is useful for tracking down temporary early mappings 1012 which are not unmapped. 1013 1014 earlycon= [KNL] Output early console device and options. 1015 1016 When used with no options, the early console is 1017 determined by stdout-path property in device tree's 1018 chosen node or the ACPI SPCR table if supported by 1019 the platform. 1020 1021 cdns,<addr>[,options] 1022 Start an early, polled-mode console on a Cadence 1023 (xuartps) serial port at the specified address. Only 1024 supported option is baud rate. If baud rate is not 1025 specified, the serial port must already be setup and 1026 configured. 1027 1028 uart[8250],io,<addr>[,options] 1029 uart[8250],mmio,<addr>[,options] 1030 uart[8250],mmio32,<addr>[,options] 1031 uart[8250],mmio32be,<addr>[,options] 1032 uart[8250],0x<addr>[,options] 1033 Start an early, polled-mode console on the 8250/16550 1034 UART at the specified I/O port or MMIO address. 1035 MMIO inter-register address stride is either 8-bit 1036 (mmio) or 32-bit (mmio32 or mmio32be). 1037 If none of [io|mmio|mmio32|mmio32be], <addr> is assumed 1038 to be equivalent to 'mmio'. 'options' are specified 1039 in the same format described for "console=ttyS<n>"; if 1040 unspecified, the h/w is not initialized. 1041 1042 pl011,<addr> 1043 pl011,mmio32,<addr> 1044 Start an early, polled-mode console on a pl011 serial 1045 port at the specified address. The pl011 serial port 1046 must already be setup and configured. Options are not 1047 yet supported. If 'mmio32' is specified, then only 1048 the driver will use only 32-bit accessors to read/write 1049 the device registers. 1050 1051 meson,<addr> 1052 Start an early, polled-mode console on a meson serial 1053 port at the specified address. The serial port must 1054 already be setup and configured. Options are not yet 1055 supported. 1056 1057 msm_serial,<addr> 1058 Start an early, polled-mode console on an msm serial 1059 port at the specified address. The serial port 1060 must already be setup and configured. Options are not 1061 yet supported. 1062 1063 msm_serial_dm,<addr> 1064 Start an early, polled-mode console on an msm serial 1065 dm port at the specified address. The serial port 1066 must already be setup and configured. Options are not 1067 yet supported. 1068 1069 owl,<addr> 1070 Start an early, polled-mode console on a serial port 1071 of an Actions Semi SoC, such as S500 or S900, at the 1072 specified address. The serial port must already be 1073 setup and configured. Options are not yet supported. 1074 1075 rda,<addr> 1076 Start an early, polled-mode console on a serial port 1077 of an RDA Micro SoC, such as RDA8810PL, at the 1078 specified address. The serial port must already be 1079 setup and configured. Options are not yet supported. 1080 1081 sbi 1082 Use RISC-V SBI (Supervisor Binary Interface) for early 1083 console. 1084 1085 smh Use ARM semihosting calls for early console. 1086 1087 s3c2410,<addr> 1088 s3c2412,<addr> 1089 s3c2440,<addr> 1090 s3c6400,<addr> 1091 s5pv210,<addr> 1092 exynos4210,<addr> 1093 Use early console provided by serial driver available 1094 on Samsung SoCs, requires selecting proper type and 1095 a correct base address of the selected UART port. The 1096 serial port must already be setup and configured. 1097 Options are not yet supported. 1098 1099 lantiq,<addr> 1100 Start an early, polled-mode console on a lantiq serial 1101 (lqasc) port at the specified address. The serial port 1102 must already be setup and configured. Options are not 1103 yet supported. 1104 1105 lpuart,<addr> 1106 lpuart32,<addr> 1107 Use early console provided by Freescale LP UART driver 1108 found on Freescale Vybrid and QorIQ LS1021A processors. 1109 A valid base address must be provided, and the serial 1110 port must already be setup and configured. 1111 1112 ec_imx21,<addr> 1113 ec_imx6q,<addr> 1114 Start an early, polled-mode, output-only console on the 1115 Freescale i.MX UART at the specified address. The UART 1116 must already be setup and configured. 1117 1118 ar3700_uart,<addr> 1119 Start an early, polled-mode console on the 1120 Armada 3700 serial port at the specified 1121 address. The serial port must already be setup 1122 and configured. Options are not yet supported. 1123 1124 qcom_geni,<addr> 1125 Start an early, polled-mode console on a Qualcomm 1126 Generic Interface (GENI) based serial port at the 1127 specified address. The serial port must already be 1128 setup and configured. Options are not yet supported. 1129 1130 efifb,[options] 1131 Start an early, unaccelerated console on the EFI 1132 memory mapped framebuffer (if available). On cache 1133 coherent non-x86 systems that use system memory for 1134 the framebuffer, pass the 'ram' option so that it is 1135 mapped with the correct attributes. 1136 1137 linflex,<addr> 1138 Use early console provided by Freescale LINFlexD UART 1139 serial driver for NXP S32V234 SoCs. A valid base 1140 address must be provided, and the serial port must 1141 already be setup and configured. 1142 1143 earlyprintk= [X86,SH,ARM,M68k,S390] 1144 earlyprintk=vga 1145 earlyprintk=sclp 1146 earlyprintk=xen 1147 earlyprintk=serial[,ttySn[,baudrate]] 1148 earlyprintk=serial[,0x...[,baudrate]] 1149 earlyprintk=ttySn[,baudrate] 1150 earlyprintk=dbgp[debugController#] 1151 earlyprintk=pciserial[,force],bus:device.function[,baudrate] 1152 earlyprintk=xdbc[xhciController#] 1153 1154 earlyprintk is useful when the kernel crashes before 1155 the normal console is initialized. It is not enabled by 1156 default because it has some cosmetic problems. 1157 1158 Append ",keep" to not disable it when the real console 1159 takes over. 1160 1161 Only one of vga, efi, serial, or usb debug port can 1162 be used at a time. 1163 1164 Currently only ttyS0 and ttyS1 may be specified by 1165 name. Other I/O ports may be explicitly specified 1166 on some architectures (x86 and arm at least) by 1167 replacing ttySn with an I/O port address, like this: 1168 earlyprintk=serial,0x1008,115200 1169 You can find the port for a given device in 1170 /proc/tty/driver/serial: 1171 2: uart:ST16650V2 port:00001008 irq:18 ... 1172 1173 Interaction with the standard serial driver is not 1174 very good. 1175 1176 The VGA and EFI output is eventually overwritten by 1177 the real console. 1178 1179 The xen output can only be used by Xen PV guests. 1180 1181 The sclp output can only be used on s390. 1182 1183 The optional "force" to "pciserial" enables use of a 1184 PCI device even when its classcode is not of the 1185 UART class. 1186 1187 edac_report= [HW,EDAC] Control how to report EDAC event 1188 Format: {"on" | "off" | "force"} 1189 on: enable EDAC to report H/W event. May be overridden 1190 by other higher priority error reporting module. 1191 off: disable H/W event reporting through EDAC. 1192 force: enforce the use of EDAC to report H/W event. 1193 default: on. 1194 1195 ekgdboc= [X86,KGDB] Allow early kernel console debugging 1196 ekgdboc=kbd 1197 1198 This is designed to be used in conjunction with 1199 the boot argument: earlyprintk=vga 1200 1201 This parameter works in place of the kgdboc parameter 1202 but can only be used if the backing tty is available 1203 very early in the boot process. For early debugging 1204 via a serial port see kgdboc_earlycon instead. 1205 1206 edd= [EDD] 1207 Format: {"off" | "on" | "skip[mbr]"} 1208 1209 efi= [EFI] 1210 Format: { "debug", "disable_early_pci_dma", 1211 "nochunk", "noruntime", "nosoftreserve", 1212 "novamap", "no_disable_early_pci_dma", 1213 "old_map" } 1214 debug: enable misc debug output. 1215 disable_early_pci_dma: disable the busmaster bit on all 1216 PCI bridges while in the EFI boot stub. 1217 nochunk: disable reading files in "chunks" in the EFI 1218 boot stub, as chunking can cause problems with some 1219 firmware implementations. 1220 noruntime : disable EFI runtime services support 1221 nosoftreserve: The EFI_MEMORY_SP (Specific Purpose) 1222 attribute may cause the kernel to reserve the 1223 memory range for a memory mapping driver to 1224 claim. Specify efi=nosoftreserve to disable this 1225 reservation and treat the memory by its base type 1226 (i.e. EFI_CONVENTIONAL_MEMORY / "System RAM"). 1227 novamap: do not call SetVirtualAddressMap(). 1228 no_disable_early_pci_dma: Leave the busmaster bit set 1229 on all PCI bridges while in the EFI boot stub 1230 old_map [X86-64]: switch to the old ioremap-based EFI 1231 runtime services mapping. [Needs CONFIG_X86_UV=y] 1232 1233 efi_no_storage_paranoia [EFI; X86] 1234 Using this parameter you can use more than 50% of 1235 your efi variable storage. Use this parameter only if 1236 you are really sure that your UEFI does sane gc and 1237 fulfills the spec otherwise your board may brick. 1238 1239 efi_fake_mem= nn[KMG]@ss[KMG]:aa[,nn[KMG]@ss[KMG]:aa,..] [EFI; X86] 1240 Add arbitrary attribute to specific memory range by 1241 updating original EFI memory map. 1242 Region of memory which aa attribute is added to is 1243 from ss to ss+nn. 1244 1245 If efi_fake_mem=2G@4G:0x10000,2G@0x10a0000000:0x10000 1246 is specified, EFI_MEMORY_MORE_RELIABLE(0x10000) 1247 attribute is added to range 0x100000000-0x180000000 and 1248 0x10a0000000-0x1120000000. 1249 1250 If efi_fake_mem=8G@9G:0x40000 is specified, the 1251 EFI_MEMORY_SP(0x40000) attribute is added to 1252 range 0x240000000-0x43fffffff. 1253 1254 Using this parameter you can do debugging of EFI memmap 1255 related features. For example, you can do debugging of 1256 Address Range Mirroring feature even if your box 1257 doesn't support it, or mark specific memory as 1258 "soft reserved". 1259 1260 efivar_ssdt= [EFI; X86] Name of an EFI variable that contains an SSDT 1261 that is to be dynamically loaded by Linux. If there are 1262 multiple variables with the same name but with different 1263 vendor GUIDs, all of them will be loaded. See 1264 Documentation/admin-guide/acpi/ssdt-overlays.rst for details. 1265 1266 1267 eisa_irq_edge= [PARISC,HW] 1268 See header of drivers/parisc/eisa.c. 1269 1270 elanfreq= [X86-32] 1271 See comment before function elanfreq_setup() in 1272 arch/x86/kernel/cpu/cpufreq/elanfreq.c. 1273 1274 elfcorehdr=[size[KMG]@]offset[KMG] [IA64,PPC,SH,X86,S390] 1275 Specifies physical address of start of kernel core 1276 image elf header and optionally the size. Generally 1277 kexec loader will pass this option to capture kernel. 1278 See Documentation/admin-guide/kdump/kdump.rst for details. 1279 1280 enable_mtrr_cleanup [X86] 1281 The kernel tries to adjust MTRR layout from continuous 1282 to discrete, to make X server driver able to add WB 1283 entry later. This parameter enables that. 1284 1285 enable_timer_pin_1 [X86] 1286 Enable PIN 1 of APIC timer 1287 Can be useful to work around chipset bugs 1288 (in particular on some ATI chipsets). 1289 The kernel tries to set a reasonable default. 1290 1291 enforcing [SELINUX] Set initial enforcing status. 1292 Format: {"0" | "1"} 1293 See security/selinux/Kconfig help text. 1294 0 -- permissive (log only, no denials). 1295 1 -- enforcing (deny and log). 1296 Default value is 0. 1297 Value can be changed at runtime via 1298 /sys/fs/selinux/enforce. 1299 1300 erst_disable [ACPI] 1301 Disable Error Record Serialization Table (ERST) 1302 support. 1303 1304 ether= [HW,NET] Ethernet cards parameters 1305 This option is obsoleted by the "netdev=" option, which 1306 has equivalent usage. See its documentation for details. 1307 1308 evm= [EVM] 1309 Format: { "fix" } 1310 Permit 'security.evm' to be updated regardless of 1311 current integrity status. 1312 1313 failslab= 1314 fail_page_alloc= 1315 fail_make_request=[KNL] 1316 General fault injection mechanism. 1317 Format: <interval>,<probability>,<space>,<times> 1318 See also Documentation/fault-injection/. 1319 1320 floppy= [HW] 1321 See Documentation/admin-guide/blockdev/floppy.rst. 1322 1323 force_pal_cache_flush 1324 [IA-64] Avoid check_sal_cache_flush which may hang on 1325 buggy SAL_CACHE_FLUSH implementations. Using this 1326 parameter will force ia64_sal_cache_flush to call 1327 ia64_pal_cache_flush instead of SAL_CACHE_FLUSH. 1328 1329 forcepae [X86-32] 1330 Forcefully enable Physical Address Extension (PAE). 1331 Many Pentium M systems disable PAE but may have a 1332 functionally usable PAE implementation. 1333 Warning: use of this parameter will taint the kernel 1334 and may cause unknown problems. 1335 1336 ftrace=[tracer] 1337 [FTRACE] will set and start the specified tracer 1338 as early as possible in order to facilitate early 1339 boot debugging. 1340 1341 ftrace_dump_on_oops[=orig_cpu] 1342 [FTRACE] will dump the trace buffers on oops. 1343 If no parameter is passed, ftrace will dump 1344 buffers of all CPUs, but if you pass orig_cpu, it will 1345 dump only the buffer of the CPU that triggered the 1346 oops. 1347 1348 ftrace_filter=[function-list] 1349 [FTRACE] Limit the functions traced by the function 1350 tracer at boot up. function-list is a comma separated 1351 list of functions. This list can be changed at run 1352 time by the set_ftrace_filter file in the debugfs 1353 tracing directory. 1354 1355 ftrace_notrace=[function-list] 1356 [FTRACE] Do not trace the functions specified in 1357 function-list. This list can be changed at run time 1358 by the set_ftrace_notrace file in the debugfs 1359 tracing directory. 1360 1361 ftrace_graph_filter=[function-list] 1362 [FTRACE] Limit the top level callers functions traced 1363 by the function graph tracer at boot up. 1364 function-list is a comma separated list of functions 1365 that can be changed at run time by the 1366 set_graph_function file in the debugfs tracing directory. 1367 1368 ftrace_graph_notrace=[function-list] 1369 [FTRACE] Do not trace from the functions specified in 1370 function-list. This list is a comma separated list of 1371 functions that can be changed at run time by the 1372 set_graph_notrace file in the debugfs tracing directory. 1373 1374 ftrace_graph_max_depth=<uint> 1375 [FTRACE] Used with the function graph tracer. This is 1376 the max depth it will trace into a function. This value 1377 can be changed at run time by the max_graph_depth file 1378 in the tracefs tracing directory. default: 0 (no limit) 1379 1380 fw_devlink= [KNL] Create device links between consumer and supplier 1381 devices by scanning the firmware to infer the 1382 consumer/supplier relationships. This feature is 1383 especially useful when drivers are loaded as modules as 1384 it ensures proper ordering of tasks like device probing 1385 (suppliers first, then consumers), supplier boot state 1386 clean up (only after all consumers have probed), 1387 suspend/resume & runtime PM (consumers first, then 1388 suppliers). 1389 Format: { off | permissive | on | rpm } 1390 off -- Don't create device links from firmware info. 1391 permissive -- Create device links from firmware info 1392 but use it only for ordering boot state clean 1393 up (sync_state() calls). 1394 on -- Create device links from firmware info and use it 1395 to enforce probe and suspend/resume ordering. 1396 rpm -- Like "on", but also use to order runtime PM. 1397 1398 gamecon.map[2|3]= 1399 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad 1400 support via parallel port (up to 5 devices per port) 1401 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5> 1402 See also Documentation/input/devices/joystick-parport.rst 1403 1404 gamma= [HW,DRM] 1405 1406 gart_fix_e820= [X86_64] disable the fix e820 for K8 GART 1407 Format: off | on 1408 default: on 1409 1410 gcov_persist= [GCOV] When non-zero (default), profiling data for 1411 kernel modules is saved and remains accessible via 1412 debugfs, even when the module is unloaded/reloaded. 1413 When zero, profiling data is discarded and associated 1414 debugfs files are removed at module unload time. 1415 1416 goldfish [X86] Enable the goldfish android emulator platform. 1417 Don't use this when you are not running on the 1418 android emulator 1419 1420 gpt [EFI] Forces disk with valid GPT signature but 1421 invalid Protective MBR to be treated as GPT. If the 1422 primary GPT is corrupted, it enables the backup/alternate 1423 GPT to be used instead. 1424 1425 grcan.enable0= [HW] Configuration of physical interface 0. Determines 1426 the "Enable 0" bit of the configuration register. 1427 Format: 0 | 1 1428 Default: 0 1429 grcan.enable1= [HW] Configuration of physical interface 1. Determines 1430 the "Enable 0" bit of the configuration register. 1431 Format: 0 | 1 1432 Default: 0 1433 grcan.select= [HW] Select which physical interface to use. 1434 Format: 0 | 1 1435 Default: 0 1436 grcan.txsize= [HW] Sets the size of the tx buffer. 1437 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0. 1438 Default: 1024 1439 grcan.rxsize= [HW] Sets the size of the rx buffer. 1440 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0. 1441 Default: 1024 1442 1443 gpio-mockup.gpio_mockup_ranges 1444 [HW] Sets the ranges of gpiochip of for this device. 1445 Format: <start1>,<end1>,<start2>,<end2>... 1446 1447 hardlockup_all_cpu_backtrace= 1448 [KNL] Should the hard-lockup detector generate 1449 backtraces on all cpus. 1450 Format: 0 | 1 1451 1452 hashdist= [KNL,NUMA] Large hashes allocated during boot 1453 are distributed across NUMA nodes. Defaults on 1454 for 64-bit NUMA, off otherwise. 1455 Format: 0 | 1 (for off | on) 1456 1457 hcl= [IA-64] SGI's Hardware Graph compatibility layer 1458 1459 hd= [EIDE] (E)IDE hard drive subsystem geometry 1460 Format: <cyl>,<head>,<sect> 1461 1462 hest_disable [ACPI] 1463 Disable Hardware Error Source Table (HEST) support; 1464 corresponding firmware-first mode error processing 1465 logic will be disabled. 1466 1467 highmem=nn[KMG] [KNL,BOOT] forces the highmem zone to have an exact 1468 size of <nn>. This works even on boxes that have no 1469 highmem otherwise. This also works to reduce highmem 1470 size on bigger boxes. 1471 1472 highres= [KNL] Enable/disable high resolution timer mode. 1473 Valid parameters: "on", "off" 1474 Default: "on" 1475 1476 hlt [BUGS=ARM,SH] 1477 1478 hpet= [X86-32,HPET] option to control HPET usage 1479 Format: { enable (default) | disable | force | 1480 verbose } 1481 disable: disable HPET and use PIT instead 1482 force: allow force enabled of undocumented chips (ICH4, 1483 VIA, nVidia) 1484 verbose: show contents of HPET registers during setup 1485 1486 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET 1487 registers. Default set by CONFIG_HPET_MMAP_DEFAULT. 1488 1489 hugetlb_cma= [HW] The size of a cma area used for allocation 1490 of gigantic hugepages. 1491 Format: nn[KMGTPE] 1492 1493 Reserve a cma area of given size and allocate gigantic 1494 hugepages using the cma allocator. If enabled, the 1495 boot-time allocation of gigantic hugepages is skipped. 1496 1497 hugepages= [HW] Number of HugeTLB pages to allocate at boot. 1498 If this follows hugepagesz (below), it specifies 1499 the number of pages of hugepagesz to be allocated. 1500 If this is the first HugeTLB parameter on the command 1501 line, it specifies the number of pages to allocate for 1502 the default huge page size. See also 1503 Documentation/admin-guide/mm/hugetlbpage.rst. 1504 Format: <integer> 1505 1506 hugepagesz= 1507 [HW] The size of the HugeTLB pages. This is used in 1508 conjunction with hugepages (above) to allocate huge 1509 pages of a specific size at boot. The pair 1510 hugepagesz=X hugepages=Y can be specified once for 1511 each supported huge page size. Huge page sizes are 1512 architecture dependent. See also 1513 Documentation/admin-guide/mm/hugetlbpage.rst. 1514 Format: size[KMG] 1515 1516 hung_task_panic= 1517 [KNL] Should the hung task detector generate panics. 1518 Format: 0 | 1 1519 1520 A value of 1 instructs the kernel to panic when a 1521 hung task is detected. The default value is controlled 1522 by the CONFIG_BOOTPARAM_HUNG_TASK_PANIC build-time 1523 option. The value selected by this boot parameter can 1524 be changed later by the kernel.hung_task_panic sysctl. 1525 1526 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC) 1527 terminal devices. Valid values: 0..8 1528 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs. 1529 If specified, z/VM IUCV HVC accepts connections 1530 from listed z/VM user IDs only. 1531 1532 hv_nopvspin [X86,HYPER_V] Disables the paravirt spinlock optimizations 1533 which allow the hypervisor to 'idle' the 1534 guest on lock contention. 1535 1536 keep_bootcon [KNL] 1537 Do not unregister boot console at start. This is only 1538 useful for debugging when something happens in the window 1539 between unregistering the boot console and initializing 1540 the real console. 1541 1542 i2c_bus= [HW] Override the default board specific I2C bus speed 1543 or register an additional I2C bus that is not 1544 registered from board initialization code. 1545 Format: 1546 <bus_id>,<clkrate> 1547 1548 i8042.debug [HW] Toggle i8042 debug mode 1549 i8042.unmask_kbd_data 1550 [HW] Enable printing of interrupt data from the KBD port 1551 (disabled by default, and as a pre-condition 1552 requires that i8042.debug=1 be enabled) 1553 i8042.direct [HW] Put keyboard port into non-translated mode 1554 i8042.dumbkbd [HW] Pretend that controller can only read data from 1555 keyboard and cannot control its state 1556 (Don't attempt to blink the leds) 1557 i8042.noaux [HW] Don't check for auxiliary (== mouse) port 1558 i8042.nokbd [HW] Don't check/create keyboard port 1559 i8042.noloop [HW] Disable the AUX Loopback command while probing 1560 for the AUX port 1561 i8042.nomux [HW] Don't check presence of an active multiplexing 1562 controller 1563 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX 1564 controllers 1565 i8042.notimeout [HW] Ignore timeout condition signalled by controller 1566 i8042.reset [HW] Reset the controller during init, cleanup and 1567 suspend-to-ram transitions, only during s2r 1568 transitions, or never reset 1569 Format: { 1 | Y | y | 0 | N | n } 1570 1, Y, y: always reset controller 1571 0, N, n: don't ever reset controller 1572 Default: only on s2r transitions on x86; most other 1573 architectures force reset to be always executed 1574 i8042.unlock [HW] Unlock (ignore) the keylock 1575 i8042.kbdreset [HW] Reset device connected to KBD port 1576 1577 i810= [HW,DRM] 1578 1579 i8k.ignore_dmi [HW] Continue probing hardware even if DMI data 1580 indicates that the driver is running on unsupported 1581 hardware. 1582 i8k.force [HW] Activate i8k driver even if SMM BIOS signature 1583 does not match list of supported models. 1584 i8k.power_status 1585 [HW] Report power status in /proc/i8k 1586 (disabled by default) 1587 i8k.restricted [HW] Allow controlling fans only if SYS_ADMIN 1588 capability is set. 1589 1590 i915.invert_brightness= 1591 [DRM] Invert the sense of the variable that is used to 1592 set the brightness of the panel backlight. Normally a 1593 brightness value of 0 indicates backlight switched off, 1594 and the maximum of the brightness value sets the backlight 1595 to maximum brightness. If this parameter is set to 0 1596 (default) and the machine requires it, or this parameter 1597 is set to 1, a brightness value of 0 sets the backlight 1598 to maximum brightness, and the maximum of the brightness 1599 value switches the backlight off. 1600 -1 -- never invert brightness 1601 0 -- machine default 1602 1 -- force brightness inversion 1603 1604 icn= [HW,ISDN] 1605 Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]] 1606 1607 ide-core.nodma= [HW] (E)IDE subsystem 1608 Format: =0.0 to prevent dma on hda, =0.1 hdb =1.0 hdc 1609 .vlb_clock .pci_clock .noflush .nohpa .noprobe .nowerr 1610 .cdrom .chs .ignore_cable are additional options 1611 See Documentation/ide/ide.rst. 1612 1613 ide-generic.probe-mask= [HW] (E)IDE subsystem 1614 Format: <int> 1615 Probe mask for legacy ISA IDE ports. Depending on 1616 platform up to 6 ports are supported, enabled by 1617 setting corresponding bits in the mask to 1. The 1618 default value is 0x0, which has a special meaning. 1619 On systems that have PCI, it triggers scanning the 1620 PCI bus for the first and the second port, which 1621 are then probed. On systems without PCI the value 1622 of 0x0 enables probing the two first ports as if it 1623 was 0x3. 1624 1625 ide-pci-generic.all-generic-ide [HW] (E)IDE subsystem 1626 Claim all unknown PCI IDE storage controllers. 1627 1628 idle= [X86] 1629 Format: idle=poll, idle=halt, idle=nomwait 1630 Poll forces a polling idle loop that can slightly 1631 improve the performance of waking up a idle CPU, but 1632 will use a lot of power and make the system run hot. 1633 Not recommended. 1634 idle=halt: Halt is forced to be used for CPU idle. 1635 In such case C2/C3 won't be used again. 1636 idle=nomwait: Disable mwait for CPU C-states 1637 1638 ieee754= [MIPS] Select IEEE Std 754 conformance mode 1639 Format: { strict | legacy | 2008 | relaxed } 1640 Default: strict 1641 1642 Choose which programs will be accepted for execution 1643 based on the IEEE 754 NaN encoding(s) supported by 1644 the FPU and the NaN encoding requested with the value 1645 of an ELF file header flag individually set by each 1646 binary. Hardware implementations are permitted to 1647 support either or both of the legacy and the 2008 NaN 1648 encoding mode. 1649 1650 Available settings are as follows: 1651 strict accept binaries that request a NaN encoding 1652 supported by the FPU 1653 legacy only accept legacy-NaN binaries, if supported 1654 by the FPU 1655 2008 only accept 2008-NaN binaries, if supported 1656 by the FPU 1657 relaxed accept any binaries regardless of whether 1658 supported by the FPU 1659 1660 The FPU emulator is always able to support both NaN 1661 encodings, so if no FPU hardware is present or it has 1662 been disabled with 'nofpu', then the settings of 1663 'legacy' and '2008' strap the emulator accordingly, 1664 'relaxed' straps the emulator for both legacy-NaN and 1665 2008-NaN, whereas 'strict' enables legacy-NaN only on 1666 legacy processors and both NaN encodings on MIPS32 or 1667 MIPS64 CPUs. 1668 1669 The setting for ABS.fmt/NEG.fmt instruction execution 1670 mode generally follows that for the NaN encoding, 1671 except where unsupported by hardware. 1672 1673 ignore_loglevel [KNL] 1674 Ignore loglevel setting - this will print /all/ 1675 kernel messages to the console. Useful for debugging. 1676 We also add it as printk module parameter, so users 1677 could change it dynamically, usually by 1678 /sys/module/printk/parameters/ignore_loglevel. 1679 1680 ignore_rlimit_data 1681 Ignore RLIMIT_DATA setting for data mappings, 1682 print warning at first misuse. Can be changed via 1683 /sys/module/kernel/parameters/ignore_rlimit_data. 1684 1685 ihash_entries= [KNL] 1686 Set number of hash buckets for inode cache. 1687 1688 ima_appraise= [IMA] appraise integrity measurements 1689 Format: { "off" | "enforce" | "fix" | "log" } 1690 default: "enforce" 1691 1692 ima_appraise_tcb [IMA] Deprecated. Use ima_policy= instead. 1693 The builtin appraise policy appraises all files 1694 owned by uid=0. 1695 1696 ima_canonical_fmt [IMA] 1697 Use the canonical format for the binary runtime 1698 measurements, instead of host native format. 1699 1700 ima_hash= [IMA] 1701 Format: { md5 | sha1 | rmd160 | sha256 | sha384 1702 | sha512 | ... } 1703 default: "sha1" 1704 1705 The list of supported hash algorithms is defined 1706 in crypto/hash_info.h. 1707 1708 ima_policy= [IMA] 1709 The builtin policies to load during IMA setup. 1710 Format: "tcb | appraise_tcb | secure_boot | 1711 fail_securely" 1712 1713 The "tcb" policy measures all programs exec'd, files 1714 mmap'd for exec, and all files opened with the read 1715 mode bit set by either the effective uid (euid=0) or 1716 uid=0. 1717 1718 The "appraise_tcb" policy appraises the integrity of 1719 all files owned by root. 1720 1721 The "secure_boot" policy appraises the integrity 1722 of files (eg. kexec kernel image, kernel modules, 1723 firmware, policy, etc) based on file signatures. 1724 1725 The "fail_securely" policy forces file signature 1726 verification failure also on privileged mounted 1727 filesystems with the SB_I_UNVERIFIABLE_SIGNATURE 1728 flag. 1729 1730 ima_tcb [IMA] Deprecated. Use ima_policy= instead. 1731 Load a policy which meets the needs of the Trusted 1732 Computing Base. This means IMA will measure all 1733 programs exec'd, files mmap'd for exec, and all files 1734 opened for read by uid=0. 1735 1736 ima_template= [IMA] 1737 Select one of defined IMA measurements template formats. 1738 Formats: { "ima" | "ima-ng" | "ima-sig" } 1739 Default: "ima-ng" 1740 1741 ima_template_fmt= 1742 [IMA] Define a custom template format. 1743 Format: { "field1|...|fieldN" } 1744 1745 ima.ahash_minsize= [IMA] Minimum file size for asynchronous hash usage 1746 Format: <min_file_size> 1747 Set the minimal file size for using asynchronous hash. 1748 If left unspecified, ahash usage is disabled. 1749 1750 ahash performance varies for different data sizes on 1751 different crypto accelerators. This option can be used 1752 to achieve the best performance for a particular HW. 1753 1754 ima.ahash_bufsize= [IMA] Asynchronous hash buffer size 1755 Format: <bufsize> 1756 Set hashing buffer size. Default: 4k. 1757 1758 ahash performance varies for different chunk sizes on 1759 different crypto accelerators. This option can be used 1760 to achieve best performance for particular HW. 1761 1762 init= [KNL] 1763 Format: <full_path> 1764 Run specified binary instead of /sbin/init as init 1765 process. 1766 1767 initcall_debug [KNL] Trace initcalls as they are executed. Useful 1768 for working out where the kernel is dying during 1769 startup. 1770 1771 initcall_blacklist= [KNL] Do not execute a comma-separated list of 1772 initcall functions. Useful for debugging built-in 1773 modules and initcalls. 1774 1775 initrd= [BOOT] Specify the location of the initial ramdisk 1776 1777 initrdmem= [KNL] Specify a physical address and size from which to 1778 load the initrd. If an initrd is compiled in or 1779 specified in the bootparams, it takes priority over this 1780 setting. 1781 Format: ss[KMG],nn[KMG] 1782 Default is 0, 0 1783 1784 init_on_alloc= [MM] Fill newly allocated pages and heap objects with 1785 zeroes. 1786 Format: 0 | 1 1787 Default set by CONFIG_INIT_ON_ALLOC_DEFAULT_ON. 1788 1789 init_on_free= [MM] Fill freed pages and heap objects with zeroes. 1790 Format: 0 | 1 1791 Default set by CONFIG_INIT_ON_FREE_DEFAULT_ON. 1792 1793 init_pkru= [x86] Specify the default memory protection keys rights 1794 register contents for all processes. 0x55555554 by 1795 default (disallow access to all but pkey 0). Can 1796 override in debugfs after boot. 1797 1798 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver 1799 Format: <irq> 1800 1801 int_pln_enable [x86] Enable power limit notification interrupt 1802 1803 integrity_audit=[IMA] 1804 Format: { "0" | "1" } 1805 0 -- basic integrity auditing messages. (Default) 1806 1 -- additional integrity auditing messages. 1807 1808 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option 1809 on 1810 Enable intel iommu driver. 1811 off 1812 Disable intel iommu driver. 1813 igfx_off [Default Off] 1814 By default, gfx is mapped as normal device. If a gfx 1815 device has a dedicated DMAR unit, the DMAR unit is 1816 bypassed by not enabling DMAR with this option. In 1817 this case, gfx device will use physical address for 1818 DMA. 1819 forcedac [x86_64] 1820 With this option iommu will not optimize to look 1821 for io virtual address below 32-bit forcing dual 1822 address cycle on pci bus for cards supporting greater 1823 than 32-bit addressing. The default is to look 1824 for translation below 32-bit and if not available 1825 then look in the higher range. 1826 strict [Default Off] 1827 With this option on every unmap_single operation will 1828 result in a hardware IOTLB flush operation as opposed 1829 to batching them for performance. 1830 sp_off [Default Off] 1831 By default, super page will be supported if Intel IOMMU 1832 has the capability. With this option, super page will 1833 not be supported. 1834 sm_on [Default Off] 1835 By default, scalable mode will be disabled even if the 1836 hardware advertises that it has support for the scalable 1837 mode translation. With this option set, scalable mode 1838 will be used on hardware which claims to support it. 1839 tboot_noforce [Default Off] 1840 Do not force the Intel IOMMU enabled under tboot. 1841 By default, tboot will force Intel IOMMU on, which 1842 could harm performance of some high-throughput 1843 devices like 40GBit network cards, even if identity 1844 mapping is enabled. 1845 Note that using this option lowers the security 1846 provided by tboot because it makes the system 1847 vulnerable to DMA attacks. 1848 nobounce [Default off] 1849 Disable bounce buffer for untrusted devices such as 1850 the Thunderbolt devices. This will treat the untrusted 1851 devices as the trusted ones, hence might expose security 1852 risks of DMA attacks. 1853 1854 intel_idle.max_cstate= [KNL,HW,ACPI,X86] 1855 0 disables intel_idle and fall back on acpi_idle. 1856 1 to 9 specify maximum depth of C-state. 1857 1858 intel_pstate= [X86] 1859 disable 1860 Do not enable intel_pstate as the default 1861 scaling driver for the supported processors 1862 passive 1863 Use intel_pstate as a scaling driver, but configure it 1864 to work with generic cpufreq governors (instead of 1865 enabling its internal governor). This mode cannot be 1866 used along with the hardware-managed P-states (HWP) 1867 feature. 1868 force 1869 Enable intel_pstate on systems that prohibit it by default 1870 in favor of acpi-cpufreq. Forcing the intel_pstate driver 1871 instead of acpi-cpufreq may disable platform features, such 1872 as thermal controls and power capping, that rely on ACPI 1873 P-States information being indicated to OSPM and therefore 1874 should be used with caution. This option does not work with 1875 processors that aren't supported by the intel_pstate driver 1876 or on platforms that use pcc-cpufreq instead of acpi-cpufreq. 1877 no_hwp 1878 Do not enable hardware P state control (HWP) 1879 if available. 1880 hwp_only 1881 Only load intel_pstate on systems which support 1882 hardware P state control (HWP) if available. 1883 support_acpi_ppc 1884 Enforce ACPI _PPC performance limits. If the Fixed ACPI 1885 Description Table, specifies preferred power management 1886 profile as "Enterprise Server" or "Performance Server", 1887 then this feature is turned on by default. 1888 per_cpu_perf_limits 1889 Allow per-logical-CPU P-State performance control limits using 1890 cpufreq sysfs interface 1891 1892 intremap= [X86-64, Intel-IOMMU] 1893 on enable Interrupt Remapping (default) 1894 off disable Interrupt Remapping 1895 nosid disable Source ID checking 1896 no_x2apic_optout 1897 BIOS x2APIC opt-out request will be ignored 1898 nopost disable Interrupt Posting 1899 1900 iomem= Disable strict checking of access to MMIO memory 1901 strict regions from userspace. 1902 relaxed 1903 1904 iommu= [x86] 1905 off 1906 force 1907 noforce 1908 biomerge 1909 panic 1910 nopanic 1911 merge 1912 nomerge 1913 soft 1914 pt [x86] 1915 nopt [x86] 1916 nobypass [PPC/POWERNV] 1917 Disable IOMMU bypass, using IOMMU for PCI devices. 1918 1919 iommu.strict= [ARM64] Configure TLB invalidation behaviour 1920 Format: { "0" | "1" } 1921 0 - Lazy mode. 1922 Request that DMA unmap operations use deferred 1923 invalidation of hardware TLBs, for increased 1924 throughput at the cost of reduced device isolation. 1925 Will fall back to strict mode if not supported by 1926 the relevant IOMMU driver. 1927 1 - Strict mode (default). 1928 DMA unmap operations invalidate IOMMU hardware TLBs 1929 synchronously. 1930 1931 iommu.passthrough= 1932 [ARM64, X86] Configure DMA to bypass the IOMMU by default. 1933 Format: { "0" | "1" } 1934 0 - Use IOMMU translation for DMA. 1935 1 - Bypass the IOMMU for DMA. 1936 unset - Use value of CONFIG_IOMMU_DEFAULT_PASSTHROUGH. 1937 1938 io7= [HW] IO7 for Marvel based alpha systems 1939 See comment before marvel_specify_io7 in 1940 arch/alpha/kernel/core_marvel.c. 1941 1942 io_delay= [X86] I/O delay method 1943 0x80 1944 Standard port 0x80 based delay 1945 0xed 1946 Alternate port 0xed based delay (needed on some systems) 1947 udelay 1948 Simple two microseconds delay 1949 none 1950 No delay 1951 1952 ip= [IP_PNP] 1953 See Documentation/admin-guide/nfs/nfsroot.rst. 1954 1955 ipcmni_extend [KNL] Extend the maximum number of unique System V 1956 IPC identifiers from 32,768 to 16,777,216. 1957 1958 irqaffinity= [SMP] Set the default irq affinity mask 1959 The argument is a cpu list, as described above. 1960 1961 irqchip.gicv2_force_probe= 1962 [ARM, ARM64] 1963 Format: <bool> 1964 Force the kernel to look for the second 4kB page 1965 of a GICv2 controller even if the memory range 1966 exposed by the device tree is too small. 1967 1968 irqchip.gicv3_nolpi= 1969 [ARM, ARM64] 1970 Force the kernel to ignore the availability of 1971 LPIs (and by consequence ITSs). Intended for system 1972 that use the kernel as a bootloader, and thus want 1973 to let secondary kernels in charge of setting up 1974 LPIs. 1975 1976 irqchip.gicv3_pseudo_nmi= [ARM64] 1977 Enables support for pseudo-NMIs in the kernel. This 1978 requires the kernel to be built with 1979 CONFIG_ARM64_PSEUDO_NMI. 1980 1981 irqfixup [HW] 1982 When an interrupt is not handled search all handlers 1983 for it. Intended to get systems with badly broken 1984 firmware running. 1985 1986 irqpoll [HW] 1987 When an interrupt is not handled search all handlers 1988 for it. Also check all handlers each timer 1989 interrupt. Intended to get systems with badly broken 1990 firmware running. 1991 1992 isapnp= [ISAPNP] 1993 Format: <RDP>,<reset>,<pci_scan>,<verbosity> 1994 1995 isolcpus= [KNL,SMP,ISOL] Isolate a given set of CPUs from disturbance. 1996 [Deprecated - use cpusets instead] 1997 Format: [flag-list,]<cpu-list> 1998 1999 Specify one or more CPUs to isolate from disturbances 2000 specified in the flag list (default: domain): 2001 2002 nohz 2003 Disable the tick when a single task runs. 2004 2005 A residual 1Hz tick is offloaded to workqueues, which you 2006 need to affine to housekeeping through the global 2007 workqueue's affinity configured via the 2008 /sys/devices/virtual/workqueue/cpumask sysfs file, or 2009 by using the 'domain' flag described below. 2010 2011 NOTE: by default the global workqueue runs on all CPUs, 2012 so to protect individual CPUs the 'cpumask' file has to 2013 be configured manually after bootup. 2014 2015 domain 2016 Isolate from the general SMP balancing and scheduling 2017 algorithms. Note that performing domain isolation this way 2018 is irreversible: it's not possible to bring back a CPU to 2019 the domains once isolated through isolcpus. It's strongly 2020 advised to use cpusets instead to disable scheduler load 2021 balancing through the "cpuset.sched_load_balance" file. 2022 It offers a much more flexible interface where CPUs can 2023 move in and out of an isolated set anytime. 2024 2025 You can move a process onto or off an "isolated" CPU via 2026 the CPU affinity syscalls or cpuset. 2027 <cpu number> begins at 0 and the maximum value is 2028 "number of CPUs in system - 1". 2029 2030 managed_irq 2031 2032 Isolate from being targeted by managed interrupts 2033 which have an interrupt mask containing isolated 2034 CPUs. The affinity of managed interrupts is 2035 handled by the kernel and cannot be changed via 2036 the /proc/irq/* interfaces. 2037 2038 This isolation is best effort and only effective 2039 if the automatically assigned interrupt mask of a 2040 device queue contains isolated and housekeeping 2041 CPUs. If housekeeping CPUs are online then such 2042 interrupts are directed to the housekeeping CPU 2043 so that IO submitted on the housekeeping CPU 2044 cannot disturb the isolated CPU. 2045 2046 If a queue's affinity mask contains only isolated 2047 CPUs then this parameter has no effect on the 2048 interrupt routing decision, though interrupts are 2049 only delivered when tasks running on those 2050 isolated CPUs submit IO. IO submitted on 2051 housekeeping CPUs has no influence on those 2052 queues. 2053 2054 The format of <cpu-list> is described above. 2055 2056 iucv= [HW,NET] 2057 2058 ivrs_ioapic [HW,X86_64] 2059 Provide an override to the IOAPIC-ID<->DEVICE-ID 2060 mapping provided in the IVRS ACPI table. For 2061 example, to map IOAPIC-ID decimal 10 to 2062 PCI device 00:14.0 write the parameter as: 2063 ivrs_ioapic[10]=00:14.0 2064 2065 ivrs_hpet [HW,X86_64] 2066 Provide an override to the HPET-ID<->DEVICE-ID 2067 mapping provided in the IVRS ACPI table. For 2068 example, to map HPET-ID decimal 0 to 2069 PCI device 00:14.0 write the parameter as: 2070 ivrs_hpet[0]=00:14.0 2071 2072 ivrs_acpihid [HW,X86_64] 2073 Provide an override to the ACPI-HID:UID<->DEVICE-ID 2074 mapping provided in the IVRS ACPI table. For 2075 example, to map UART-HID:UID AMD0020:0 to 2076 PCI device 00:14.5 write the parameter as: 2077 ivrs_acpihid[00:14.5]=AMD0020:0 2078 2079 js= [HW,JOY] Analog joystick 2080 See Documentation/input/joydev/joystick.rst. 2081 2082 nokaslr [KNL] 2083 When CONFIG_RANDOMIZE_BASE is set, this disables 2084 kernel and module base offset ASLR (Address Space 2085 Layout Randomization). 2086 2087 kasan_multi_shot 2088 [KNL] Enforce KASAN (Kernel Address Sanitizer) to print 2089 report on every invalid memory access. Without this 2090 parameter KASAN will print report only for the first 2091 invalid access. 2092 2093 keepinitrd [HW,ARM] 2094 2095 kernelcore= [KNL,X86,IA-64,PPC] 2096 Format: nn[KMGTPE] | nn% | "mirror" 2097 This parameter specifies the amount of memory usable by 2098 the kernel for non-movable allocations. The requested 2099 amount is spread evenly throughout all nodes in the 2100 system as ZONE_NORMAL. The remaining memory is used for 2101 movable memory in its own zone, ZONE_MOVABLE. In the 2102 event, a node is too small to have both ZONE_NORMAL and 2103 ZONE_MOVABLE, kernelcore memory will take priority and 2104 other nodes will have a larger ZONE_MOVABLE. 2105 2106 ZONE_MOVABLE is used for the allocation of pages that 2107 may be reclaimed or moved by the page migration 2108 subsystem. Note that allocations like PTEs-from-HighMem 2109 still use the HighMem zone if it exists, and the Normal 2110 zone if it does not. 2111 2112 It is possible to specify the exact amount of memory in 2113 the form of "nn[KMGTPE]", a percentage of total system 2114 memory in the form of "nn%", or "mirror". If "mirror" 2115 option is specified, mirrored (reliable) memory is used 2116 for non-movable allocations and remaining memory is used 2117 for Movable pages. "nn[KMGTPE]", "nn%", and "mirror" 2118 are exclusive, so you cannot specify multiple forms. 2119 2120 kgdbdbgp= [KGDB,HW] kgdb over EHCI usb debug port. 2121 Format: <Controller#>[,poll interval] 2122 The controller # is the number of the ehci usb debug 2123 port as it is probed via PCI. The poll interval is 2124 optional and is the number seconds in between 2125 each poll cycle to the debug port in case you need 2126 the functionality for interrupting the kernel with 2127 gdb or control-c on the dbgp connection. When 2128 not using this parameter you use sysrq-g to break into 2129 the kernel debugger. 2130 2131 kgdboc= [KGDB,HW] kgdb over consoles. 2132 Requires a tty driver that supports console polling, 2133 or a supported polling keyboard driver (non-usb). 2134 Serial only format: <serial_device>[,baud] 2135 keyboard only format: kbd 2136 keyboard and serial format: kbd,<serial_device>[,baud] 2137 Optional Kernel mode setting: 2138 kms, kbd format: kms,kbd 2139 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud] 2140 2141 kgdboc_earlycon= [KGDB,HW] 2142 If the boot console provides the ability to read 2143 characters and can work in polling mode, you can use 2144 this parameter to tell kgdb to use it as a backend 2145 until the normal console is registered. Intended to 2146 be used together with the kgdboc parameter which 2147 specifies the normal console to transition to. 2148 2149 The name of the early console should be specified 2150 as the value of this parameter. Note that the name of 2151 the early console might be different than the tty 2152 name passed to kgdboc. It's OK to leave the value 2153 blank and the first boot console that implements 2154 read() will be picked. 2155 2156 kgdbwait [KGDB] Stop kernel execution and enter the 2157 kernel debugger at the earliest opportunity. 2158 2159 kmac= [MIPS] korina ethernet MAC address. 2160 Configure the RouterBoard 532 series on-chip 2161 Ethernet adapter MAC address. 2162 2163 kmemleak= [KNL] Boot-time kmemleak enable/disable 2164 Valid arguments: on, off 2165 Default: on 2166 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y, 2167 the default is off. 2168 2169 kprobe_event=[probe-list] 2170 [FTRACE] Add kprobe events and enable at boot time. 2171 The probe-list is a semicolon delimited list of probe 2172 definitions. Each definition is same as kprobe_events 2173 interface, but the parameters are comma delimited. 2174 For example, to add a kprobe event on vfs_read with 2175 arg1 and arg2, add to the command line; 2176 2177 kprobe_event=p,vfs_read,$arg1,$arg2 2178 2179 See also Documentation/trace/kprobetrace.rst "Kernel 2180 Boot Parameter" section. 2181 2182 kpti= [ARM64] Control page table isolation of user 2183 and kernel address spaces. 2184 Default: enabled on cores which need mitigation. 2185 0: force disabled 2186 1: force enabled 2187 2188 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs. 2189 Default is 0 (don't ignore, but inject #GP) 2190 2191 kvm.enable_vmware_backdoor=[KVM] Support VMware backdoor PV interface. 2192 Default is false (don't support). 2193 2194 kvm.mmu_audit= [KVM] This is a R/W parameter which allows audit 2195 KVM MMU at runtime. 2196 Default is 0 (off) 2197 2198 kvm.nx_huge_pages= 2199 [KVM] Controls the software workaround for the 2200 X86_BUG_ITLB_MULTIHIT bug. 2201 force : Always deploy workaround. 2202 off : Never deploy workaround. 2203 auto : Deploy workaround based on the presence of 2204 X86_BUG_ITLB_MULTIHIT. 2205 2206 Default is 'auto'. 2207 2208 If the software workaround is enabled for the host, 2209 guests do need not to enable it for nested guests. 2210 2211 kvm.nx_huge_pages_recovery_ratio= 2212 [KVM] Controls how many 4KiB pages are periodically zapped 2213 back to huge pages. 0 disables the recovery, otherwise if 2214 the value is N KVM will zap 1/Nth of the 4KiB pages every 2215 minute. The default is 60. 2216 2217 kvm-amd.nested= [KVM,AMD] Allow nested virtualization in KVM/SVM. 2218 Default is 1 (enabled) 2219 2220 kvm-amd.npt= [KVM,AMD] Disable nested paging (virtualized MMU) 2221 for all guests. 2222 Default is 1 (enabled) if in 64-bit or 32-bit PAE mode. 2223 2224 kvm-arm.vgic_v3_group0_trap= 2225 [KVM,ARM] Trap guest accesses to GICv3 group-0 2226 system registers 2227 2228 kvm-arm.vgic_v3_group1_trap= 2229 [KVM,ARM] Trap guest accesses to GICv3 group-1 2230 system registers 2231 2232 kvm-arm.vgic_v3_common_trap= 2233 [KVM,ARM] Trap guest accesses to GICv3 common 2234 system registers 2235 2236 kvm-arm.vgic_v4_enable= 2237 [KVM,ARM] Allow use of GICv4 for direct injection of 2238 LPIs. 2239 2240 kvm-intel.ept= [KVM,Intel] Disable extended page tables 2241 (virtualized MMU) support on capable Intel chips. 2242 Default is 1 (enabled) 2243 2244 kvm-intel.emulate_invalid_guest_state= 2245 [KVM,Intel] Enable emulation of invalid guest states 2246 Default is 0 (disabled) 2247 2248 kvm-intel.flexpriority= 2249 [KVM,Intel] Disable FlexPriority feature (TPR shadow). 2250 Default is 1 (enabled) 2251 2252 kvm-intel.nested= 2253 [KVM,Intel] Enable VMX nesting (nVMX). 2254 Default is 0 (disabled) 2255 2256 kvm-intel.unrestricted_guest= 2257 [KVM,Intel] Disable unrestricted guest feature 2258 (virtualized real and unpaged mode) on capable 2259 Intel chips. Default is 1 (enabled) 2260 2261 kvm-intel.vmentry_l1d_flush=[KVM,Intel] Mitigation for L1 Terminal Fault 2262 CVE-2018-3620. 2263 2264 Valid arguments: never, cond, always 2265 2266 always: L1D cache flush on every VMENTER. 2267 cond: Flush L1D on VMENTER only when the code between 2268 VMEXIT and VMENTER can leak host memory. 2269 never: Disables the mitigation 2270 2271 Default is cond (do L1 cache flush in specific instances) 2272 2273 kvm-intel.vpid= [KVM,Intel] Disable Virtual Processor Identification 2274 feature (tagged TLBs) on capable Intel chips. 2275 Default is 1 (enabled) 2276 2277 l1tf= [X86] Control mitigation of the L1TF vulnerability on 2278 affected CPUs 2279 2280 The kernel PTE inversion protection is unconditionally 2281 enabled and cannot be disabled. 2282 2283 full 2284 Provides all available mitigations for the 2285 L1TF vulnerability. Disables SMT and 2286 enables all mitigations in the 2287 hypervisors, i.e. unconditional L1D flush. 2288 2289 SMT control and L1D flush control via the 2290 sysfs interface is still possible after 2291 boot. Hypervisors will issue a warning 2292 when the first VM is started in a 2293 potentially insecure configuration, 2294 i.e. SMT enabled or L1D flush disabled. 2295 2296 full,force 2297 Same as 'full', but disables SMT and L1D 2298 flush runtime control. Implies the 2299 'nosmt=force' command line option. 2300 (i.e. sysfs control of SMT is disabled.) 2301 2302 flush 2303 Leaves SMT enabled and enables the default 2304 hypervisor mitigation, i.e. conditional 2305 L1D flush. 2306 2307 SMT control and L1D flush control via the 2308 sysfs interface is still possible after 2309 boot. Hypervisors will issue a warning 2310 when the first VM is started in a 2311 potentially insecure configuration, 2312 i.e. SMT enabled or L1D flush disabled. 2313 2314 flush,nosmt 2315 2316 Disables SMT and enables the default 2317 hypervisor mitigation. 2318 2319 SMT control and L1D flush control via the 2320 sysfs interface is still possible after 2321 boot. Hypervisors will issue a warning 2322 when the first VM is started in a 2323 potentially insecure configuration, 2324 i.e. SMT enabled or L1D flush disabled. 2325 2326 flush,nowarn 2327 Same as 'flush', but hypervisors will not 2328 warn when a VM is started in a potentially 2329 insecure configuration. 2330 2331 off 2332 Disables hypervisor mitigations and doesn't 2333 emit any warnings. 2334 It also drops the swap size and available 2335 RAM limit restriction on both hypervisor and 2336 bare metal. 2337 2338 Default is 'flush'. 2339 2340 For details see: Documentation/admin-guide/hw-vuln/l1tf.rst 2341 2342 l2cr= [PPC] 2343 2344 l3cr= [PPC] 2345 2346 lapic [X86-32,APIC] Enable the local APIC even if BIOS 2347 disabled it. 2348 2349 lapic= [x86,APIC] "notscdeadline" Do not use TSC deadline 2350 value for LAPIC timer one-shot implementation. Default 2351 back to the programmable timer unit in the LAPIC. 2352 2353 lapic_timer_c2_ok [X86,APIC] trust the local apic timer 2354 in C2 power state. 2355 2356 libata.dma= [LIBATA] DMA control 2357 libata.dma=0 Disable all PATA and SATA DMA 2358 libata.dma=1 PATA and SATA Disk DMA only 2359 libata.dma=2 ATAPI (CDROM) DMA only 2360 libata.dma=4 Compact Flash DMA only 2361 Combinations also work, so libata.dma=3 enables DMA 2362 for disks and CDROMs, but not CFs. 2363 2364 libata.ignore_hpa= [LIBATA] Ignore HPA limit 2365 libata.ignore_hpa=0 keep BIOS limits (default) 2366 libata.ignore_hpa=1 ignore limits, using full disk 2367 2368 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume 2369 when set. 2370 Format: <int> 2371 2372 libata.force= [LIBATA] Force configurations. The format is comma 2373 separated list of "[ID:]VAL" where ID is 2374 PORT[.DEVICE]. PORT and DEVICE are decimal numbers 2375 matching port, link or device. Basically, it matches 2376 the ATA ID string printed on console by libata. If 2377 the whole ID part is omitted, the last PORT and DEVICE 2378 values are used. If ID hasn't been specified yet, the 2379 configuration applies to all ports, links and devices. 2380 2381 If only DEVICE is omitted, the parameter applies to 2382 the port and all links and devices behind it. DEVICE 2383 number of 0 either selects the first device or the 2384 first fan-out link behind PMP device. It does not 2385 select the host link. DEVICE number of 15 selects the 2386 host link and device attached to it. 2387 2388 The VAL specifies the configuration to force. As long 2389 as there's no ambiguity shortcut notation is allowed. 2390 For example, both 1.5 and 1.5G would work for 1.5Gbps. 2391 The following configurations can be forced. 2392 2393 * Cable type: 40c, 80c, short40c, unk, ign or sata. 2394 Any ID with matching PORT is used. 2395 2396 * SATA link speed limit: 1.5Gbps or 3.0Gbps. 2397 2398 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7]. 2399 udma[/][16,25,33,44,66,100,133] notation is also 2400 allowed. 2401 2402 * [no]ncq: Turn on or off NCQ. 2403 2404 * [no]ncqtrim: Turn off queued DSM TRIM. 2405 2406 * nohrst, nosrst, norst: suppress hard, soft 2407 and both resets. 2408 2409 * rstonce: only attempt one reset during 2410 hot-unplug link recovery 2411 2412 * dump_id: dump IDENTIFY data. 2413 2414 * atapi_dmadir: Enable ATAPI DMADIR bridge support 2415 2416 * disable: Disable this device. 2417 2418 If there are multiple matching configurations changing 2419 the same attribute, the last one is used. 2420 2421 memblock=debug [KNL] Enable memblock debug messages. 2422 2423 load_ramdisk= [RAM] List of ramdisks to load from floppy 2424 See Documentation/admin-guide/blockdev/ramdisk.rst. 2425 2426 lockd.nlm_grace_period=P [NFS] Assign grace period. 2427 Format: <integer> 2428 2429 lockd.nlm_tcpport=N [NFS] Assign TCP port. 2430 Format: <integer> 2431 2432 lockd.nlm_timeout=T [NFS] Assign timeout value. 2433 Format: <integer> 2434 2435 lockd.nlm_udpport=M [NFS] Assign UDP port. 2436 Format: <integer> 2437 2438 lockdown= [SECURITY] 2439 { integrity | confidentiality } 2440 Enable the kernel lockdown feature. If set to 2441 integrity, kernel features that allow userland to 2442 modify the running kernel are disabled. If set to 2443 confidentiality, kernel features that allow userland 2444 to extract confidential information from the kernel 2445 are also disabled. 2446 2447 locktorture.nreaders_stress= [KNL] 2448 Set the number of locking read-acquisition kthreads. 2449 Defaults to being automatically set based on the 2450 number of online CPUs. 2451 2452 locktorture.nwriters_stress= [KNL] 2453 Set the number of locking write-acquisition kthreads. 2454 2455 locktorture.onoff_holdoff= [KNL] 2456 Set time (s) after boot for CPU-hotplug testing. 2457 2458 locktorture.onoff_interval= [KNL] 2459 Set time (s) between CPU-hotplug operations, or 2460 zero to disable CPU-hotplug testing. 2461 2462 locktorture.shuffle_interval= [KNL] 2463 Set task-shuffle interval (jiffies). Shuffling 2464 tasks allows some CPUs to go into dyntick-idle 2465 mode during the locktorture test. 2466 2467 locktorture.shutdown_secs= [KNL] 2468 Set time (s) after boot system shutdown. This 2469 is useful for hands-off automated testing. 2470 2471 locktorture.stat_interval= [KNL] 2472 Time (s) between statistics printk()s. 2473 2474 locktorture.stutter= [KNL] 2475 Time (s) to stutter testing, for example, 2476 specifying five seconds causes the test to run for 2477 five seconds, wait for five seconds, and so on. 2478 This tests the locking primitive's ability to 2479 transition abruptly to and from idle. 2480 2481 locktorture.torture_type= [KNL] 2482 Specify the locking implementation to test. 2483 2484 locktorture.verbose= [KNL] 2485 Enable additional printk() statements. 2486 2487 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver 2488 Format: <irq> 2489 2490 loglevel= All Kernel Messages with a loglevel smaller than the 2491 console loglevel will be printed to the console. It can 2492 also be changed with klogd or other programs. The 2493 loglevels are defined as follows: 2494 2495 0 (KERN_EMERG) system is unusable 2496 1 (KERN_ALERT) action must be taken immediately 2497 2 (KERN_CRIT) critical conditions 2498 3 (KERN_ERR) error conditions 2499 4 (KERN_WARNING) warning conditions 2500 5 (KERN_NOTICE) normal but significant condition 2501 6 (KERN_INFO) informational 2502 7 (KERN_DEBUG) debug-level messages 2503 2504 log_buf_len=n[KMG] Sets the size of the printk ring buffer, 2505 in bytes. n must be a power of two and greater 2506 than the minimal size. The minimal size is defined 2507 by LOG_BUF_SHIFT kernel config parameter. There is 2508 also CONFIG_LOG_CPU_MAX_BUF_SHIFT config parameter 2509 that allows to increase the default size depending on 2510 the number of CPUs. See init/Kconfig for more details. 2511 2512 logo.nologo [FB] Disables display of the built-in Linux logo. 2513 This may be used to provide more screen space for 2514 kernel log messages and is useful when debugging 2515 kernel boot problems. 2516 2517 lp=0 [LP] Specify parallel ports to use, e.g, 2518 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses 2519 lp=reset first parallel port). 'lp=0' disables the 2520 lp=auto printer driver. 'lp=reset' (which can be 2521 specified in addition to the ports) causes 2522 attached printers to be reset. Using 2523 lp=port1,port2,... specifies the parallel ports 2524 to associate lp devices with, starting with 2525 lp0. A port specification may be 'none' to skip 2526 that lp device, or a parport name such as 2527 'parport0'. Specifying 'lp=auto' instead of a 2528 port specification list means that device IDs 2529 from each port should be examined, to see if 2530 an IEEE 1284-compliant printer is attached; if 2531 so, the driver will manage that printer. 2532 See also header of drivers/char/lp.c. 2533 2534 lpj=n [KNL] 2535 Sets loops_per_jiffy to given constant, thus avoiding 2536 time-consuming boot-time autodetection (up to 250 ms per 2537 CPU). 0 enables autodetection (default). To determine 2538 the correct value for your kernel, boot with normal 2539 autodetection and see what value is printed. Note that 2540 on SMP systems the preset will be applied to all CPUs, 2541 which is likely to cause problems if your CPUs need 2542 significantly divergent settings. An incorrect value 2543 will cause delays in the kernel to be wrong, leading to 2544 unpredictable I/O errors and other breakage. Although 2545 unlikely, in the extreme case this might damage your 2546 hardware. 2547 2548 ltpc= [NET] 2549 Format: <io>,<irq>,<dma> 2550 2551 lsm.debug [SECURITY] Enable LSM initialization debugging output. 2552 2553 lsm=lsm1,...,lsmN 2554 [SECURITY] Choose order of LSM initialization. This 2555 overrides CONFIG_LSM, and the "security=" parameter. 2556 2557 machvec= [IA-64] Force the use of a particular machine-vector 2558 (machvec) in a generic kernel. 2559 Example: machvec=hpzx1 2560 2561 machtype= [Loongson] Share the same kernel image file between different 2562 yeeloong laptop. 2563 Example: machtype=lemote-yeeloong-2f-7inch 2564 2565 max_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory greater 2566 than or equal to this physical address is ignored. 2567 2568 maxcpus= [SMP] Maximum number of processors that an SMP kernel 2569 will bring up during bootup. maxcpus=n : n >= 0 limits 2570 the kernel to bring up 'n' processors. Surely after 2571 bootup you can bring up the other plugged cpu by executing 2572 "echo 1 > /sys/devices/system/cpu/cpuX/online". So maxcpus 2573 only takes effect during system bootup. 2574 While n=0 is a special case, it is equivalent to "nosmp", 2575 which also disables the IO APIC. 2576 2577 max_loop= [LOOP] The number of loop block devices that get 2578 (loop.max_loop) unconditionally pre-created at init time. The default 2579 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead 2580 of statically allocating a predefined number, loop 2581 devices can be requested on-demand with the 2582 /dev/loop-control interface. 2583 2584 mce [X86-32] Machine Check Exception 2585 2586 mce=option [X86-64] See Documentation/x86/x86_64/boot-options.rst 2587 2588 md= [HW] RAID subsystems devices and level 2589 See Documentation/admin-guide/md.rst. 2590 2591 mdacon= [MDA] 2592 Format: <first>,<last> 2593 Specifies range of consoles to be captured by the MDA. 2594 2595 mds= [X86,INTEL] 2596 Control mitigation for the Micro-architectural Data 2597 Sampling (MDS) vulnerability. 2598 2599 Certain CPUs are vulnerable to an exploit against CPU 2600 internal buffers which can forward information to a 2601 disclosure gadget under certain conditions. 2602 2603 In vulnerable processors, the speculatively 2604 forwarded data can be used in a cache side channel 2605 attack, to access data to which the attacker does 2606 not have direct access. 2607 2608 This parameter controls the MDS mitigation. The 2609 options are: 2610 2611 full - Enable MDS mitigation on vulnerable CPUs 2612 full,nosmt - Enable MDS mitigation and disable 2613 SMT on vulnerable CPUs 2614 off - Unconditionally disable MDS mitigation 2615 2616 On TAA-affected machines, mds=off can be prevented by 2617 an active TAA mitigation as both vulnerabilities are 2618 mitigated with the same mechanism so in order to disable 2619 this mitigation, you need to specify tsx_async_abort=off 2620 too. 2621 2622 Not specifying this option is equivalent to 2623 mds=full. 2624 2625 For details see: Documentation/admin-guide/hw-vuln/mds.rst 2626 2627 mem=nn[KMG] [KNL,BOOT] Force usage of a specific amount of memory 2628 Amount of memory to be used in cases as follows: 2629 2630 1 for test; 2631 2 when the kernel is not able to see the whole system memory; 2632 3 memory that lies after 'mem=' boundary is excluded from 2633 the hypervisor, then assigned to KVM guests. 2634 2635 [X86] Work as limiting max address. Use together 2636 with memmap= to avoid physical address space collisions. 2637 Without memmap= PCI devices could be placed at addresses 2638 belonging to unused RAM. 2639 2640 Note that this only takes effects during boot time since 2641 in above case 3, memory may need be hot added after boot 2642 if system memory of hypervisor is not sufficient. 2643 2644 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel 2645 memory. 2646 2647 memchunk=nn[KMG] 2648 [KNL,SH] Allow user to override the default size for 2649 per-device physically contiguous DMA buffers. 2650 2651 memhp_default_state=online/offline 2652 [KNL] Set the initial state for the memory hotplug 2653 onlining policy. If not specified, the default value is 2654 set according to the 2655 CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE kernel config 2656 option. 2657 See Documentation/admin-guide/mm/memory-hotplug.rst. 2658 2659 memmap=exactmap [KNL,X86] Enable setting of an exact 2660 E820 memory map, as specified by the user. 2661 Such memmap=exactmap lines can be constructed based on 2662 BIOS output or other requirements. See the memmap=nn@ss 2663 option description. 2664 2665 memmap=nn[KMG]@ss[KMG] 2666 [KNL] Force usage of a specific region of memory. 2667 Region of memory to be used is from ss to ss+nn. 2668 If @ss[KMG] is omitted, it is equivalent to mem=nn[KMG], 2669 which limits max address to nn[KMG]. 2670 Multiple different regions can be specified, 2671 comma delimited. 2672 Example: 2673 memmap=100M@2G,100M#3G,1G!1024G 2674 2675 memmap=nn[KMG]#ss[KMG] 2676 [KNL,ACPI] Mark specific memory as ACPI data. 2677 Region of memory to be marked is from ss to ss+nn. 2678 2679 memmap=nn[KMG]$ss[KMG] 2680 [KNL,ACPI] Mark specific memory as reserved. 2681 Region of memory to be reserved is from ss to ss+nn. 2682 Example: Exclude memory from 0x18690000-0x1869ffff 2683 memmap=64K$0x18690000 2684 or 2685 memmap=0x10000$0x18690000 2686 Some bootloaders may need an escape character before '$', 2687 like Grub2, otherwise '$' and the following number 2688 will be eaten. 2689 2690 memmap=nn[KMG]!ss[KMG] 2691 [KNL,X86] Mark specific memory as protected. 2692 Region of memory to be used, from ss to ss+nn. 2693 The memory region may be marked as e820 type 12 (0xc) 2694 and is NVDIMM or ADR memory. 2695 2696 memmap=<size>%<offset>-<oldtype>+<newtype> 2697 [KNL,ACPI] Convert memory within the specified region 2698 from <oldtype> to <newtype>. If "-<oldtype>" is left 2699 out, the whole region will be marked as <newtype>, 2700 even if previously unavailable. If "+<newtype>" is left 2701 out, matching memory will be removed. Types are 2702 specified as e820 types, e.g., 1 = RAM, 2 = reserved, 2703 3 = ACPI, 12 = PRAM. 2704 2705 memory_corruption_check=0/1 [X86] 2706 Some BIOSes seem to corrupt the first 64k of 2707 memory when doing things like suspend/resume. 2708 Setting this option will scan the memory 2709 looking for corruption. Enabling this will 2710 both detect corruption and prevent the kernel 2711 from using the memory being corrupted. 2712 However, its intended as a diagnostic tool; if 2713 repeatable BIOS-originated corruption always 2714 affects the same memory, you can use memmap= 2715 to prevent the kernel from using that memory. 2716 2717 memory_corruption_check_size=size [X86] 2718 By default it checks for corruption in the low 2719 64k, making this memory unavailable for normal 2720 use. Use this parameter to scan for 2721 corruption in more or less memory. 2722 2723 memory_corruption_check_period=seconds [X86] 2724 By default it checks for corruption every 60 2725 seconds. Use this parameter to check at some 2726 other rate. 0 disables periodic checking. 2727 2728 memtest= [KNL,X86,ARM,PPC] Enable memtest 2729 Format: <integer> 2730 default : 0 <disable> 2731 Specifies the number of memtest passes to be 2732 performed. Each pass selects another test 2733 pattern from a given set of patterns. Memtest 2734 fills the memory with this pattern, validates 2735 memory contents and reserves bad memory 2736 regions that are detected. 2737 2738 mem_encrypt= [X86-64] AMD Secure Memory Encryption (SME) control 2739 Valid arguments: on, off 2740 Default (depends on kernel configuration option): 2741 on (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=y) 2742 off (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=n) 2743 mem_encrypt=on: Activate SME 2744 mem_encrypt=off: Do not activate SME 2745 2746 Refer to Documentation/virt/kvm/amd-memory-encryption.rst 2747 for details on when memory encryption can be activated. 2748 2749 mem_sleep_default= [SUSPEND] Default system suspend mode: 2750 s2idle - Suspend-To-Idle 2751 shallow - Power-On Suspend or equivalent (if supported) 2752 deep - Suspend-To-RAM or equivalent (if supported) 2753 See Documentation/admin-guide/pm/sleep-states.rst. 2754 2755 meye.*= [HW] Set MotionEye Camera parameters 2756 See Documentation/admin-guide/media/meye.rst. 2757 2758 mfgpt_irq= [IA-32] Specify the IRQ to use for the 2759 Multi-Function General Purpose Timers on AMD Geode 2760 platforms. 2761 2762 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when 2763 the BIOS has incorrectly applied a workaround. TinyBIOS 2764 version 0.98 is known to be affected, 0.99 fixes the 2765 problem by letting the user disable the workaround. 2766 2767 mga= [HW,DRM] 2768 2769 min_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory below this 2770 physical address is ignored. 2771 2772 mini2440= [ARM,HW,KNL] 2773 Format:[0..2][b][c][t] 2774 Default: "0tb" 2775 MINI2440 configuration specification: 2776 0 - The attached screen is the 3.5" TFT 2777 1 - The attached screen is the 7" TFT 2778 2 - The VGA Shield is attached (1024x768) 2779 Leaving out the screen size parameter will not load 2780 the TFT driver, and the framebuffer will be left 2781 unconfigured. 2782 b - Enable backlight. The TFT backlight pin will be 2783 linked to the kernel VESA blanking code and a GPIO 2784 LED. This parameter is not necessary when using the 2785 VGA shield. 2786 c - Enable the s3c camera interface. 2787 t - Reserved for enabling touchscreen support. The 2788 touchscreen support is not enabled in the mainstream 2789 kernel as of 2.6.30, a preliminary port can be found 2790 in the "bleeding edge" mini2440 support kernel at 2791 http://repo.or.cz/w/linux-2.6/mini2440.git 2792 2793 mitigations= 2794 [X86,PPC,S390,ARM64] Control optional mitigations for 2795 CPU vulnerabilities. This is a set of curated, 2796 arch-independent options, each of which is an 2797 aggregation of existing arch-specific options. 2798 2799 off 2800 Disable all optional CPU mitigations. This 2801 improves system performance, but it may also 2802 expose users to several CPU vulnerabilities. 2803 Equivalent to: nopti [X86,PPC] 2804 kpti=0 [ARM64] 2805 nospectre_v1 [X86,PPC] 2806 nobp=0 [S390] 2807 nospectre_v2 [X86,PPC,S390,ARM64] 2808 spectre_v2_user=off [X86] 2809 spec_store_bypass_disable=off [X86,PPC] 2810 ssbd=force-off [ARM64] 2811 l1tf=off [X86] 2812 mds=off [X86] 2813 tsx_async_abort=off [X86] 2814 kvm.nx_huge_pages=off [X86] 2815 2816 Exceptions: 2817 This does not have any effect on 2818 kvm.nx_huge_pages when 2819 kvm.nx_huge_pages=force. 2820 2821 auto (default) 2822 Mitigate all CPU vulnerabilities, but leave SMT 2823 enabled, even if it's vulnerable. This is for 2824 users who don't want to be surprised by SMT 2825 getting disabled across kernel upgrades, or who 2826 have other ways of avoiding SMT-based attacks. 2827 Equivalent to: (default behavior) 2828 2829 auto,nosmt 2830 Mitigate all CPU vulnerabilities, disabling SMT 2831 if needed. This is for users who always want to 2832 be fully mitigated, even if it means losing SMT. 2833 Equivalent to: l1tf=flush,nosmt [X86] 2834 mds=full,nosmt [X86] 2835 tsx_async_abort=full,nosmt [X86] 2836 2837 mminit_loglevel= 2838 [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this 2839 parameter allows control of the logging verbosity for 2840 the additional memory initialisation checks. A value 2841 of 0 disables mminit logging and a level of 4 will 2842 log everything. Information is printed at KERN_DEBUG 2843 so loglevel=8 may also need to be specified. 2844 2845 module.sig_enforce 2846 [KNL] When CONFIG_MODULE_SIG is set, this means that 2847 modules without (valid) signatures will fail to load. 2848 Note that if CONFIG_MODULE_SIG_FORCE is set, that 2849 is always true, so this option does nothing. 2850 2851 module_blacklist= [KNL] Do not load a comma-separated list of 2852 modules. Useful for debugging problem modules. 2853 2854 mousedev.tap_time= 2855 [MOUSE] Maximum time between finger touching and 2856 leaving touchpad surface for touch to be considered 2857 a tap and be reported as a left button click (for 2858 touchpads working in absolute mode only). 2859 Format: <msecs> 2860 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices 2861 reporting absolute coordinates, such as tablets 2862 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices 2863 reporting absolute coordinates, such as tablets 2864 2865 movablecore= [KNL,X86,IA-64,PPC] 2866 Format: nn[KMGTPE] | nn% 2867 This parameter is the complement to kernelcore=, it 2868 specifies the amount of memory used for migratable 2869 allocations. If both kernelcore and movablecore is 2870 specified, then kernelcore will be at *least* the 2871 specified value but may be more. If movablecore on its 2872 own is specified, the administrator must be careful 2873 that the amount of memory usable for all allocations 2874 is not too small. 2875 2876 movable_node [KNL] Boot-time switch to make hotplugable memory 2877 NUMA nodes to be movable. This means that the memory 2878 of such nodes will be usable only for movable 2879 allocations which rules out almost all kernel 2880 allocations. Use with caution! 2881 2882 MTD_Partition= [MTD] 2883 Format: <name>,<region-number>,<size>,<offset> 2884 2885 MTD_Region= [MTD] Format: 2886 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>] 2887 2888 mtdparts= [MTD] 2889 See drivers/mtd/parsers/cmdlinepart.c 2890 2891 multitce=off [PPC] This parameter disables the use of the pSeries 2892 firmware feature for updating multiple TCE entries 2893 at a time. 2894 2895 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration 2896 2897 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock] 2898 2899 boundary - index of last SLC block on Flex-OneNAND. 2900 The remaining blocks are configured as MLC blocks. 2901 lock - Configure if Flex-OneNAND boundary should be locked. 2902 Once locked, the boundary cannot be changed. 2903 1 indicates lock status, 0 indicates unlock status. 2904 2905 mtdset= [ARM] 2906 ARM/S3C2412 JIVE boot control 2907 2908 See arch/arm/mach-s3c2412/mach-jive.c 2909 2910 mtouchusb.raw_coordinates= 2911 [HW] Make the MicroTouch USB driver use raw coordinates 2912 ('y', default) or cooked coordinates ('n') 2913 2914 mtrr_chunk_size=nn[KMG] [X86] 2915 used for mtrr cleanup. It is largest continuous chunk 2916 that could hold holes aka. UC entries. 2917 2918 mtrr_gran_size=nn[KMG] [X86] 2919 Used for mtrr cleanup. It is granularity of mtrr block. 2920 Default is 1. 2921 Large value could prevent small alignment from 2922 using up MTRRs. 2923 2924 mtrr_spare_reg_nr=n [X86] 2925 Format: <integer> 2926 Range: 0,7 : spare reg number 2927 Default : 1 2928 Used for mtrr cleanup. It is spare mtrr entries number. 2929 Set to 2 or more if your graphical card needs more. 2930 2931 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card 2932 2933 netdev= [NET] Network devices parameters 2934 Format: <irq>,<io>,<mem_start>,<mem_end>,<name> 2935 Note that mem_start is often overloaded to mean 2936 something different and driver-specific. 2937 This usage is only documented in each driver source 2938 file if at all. 2939 2940 nf_conntrack.acct= 2941 [NETFILTER] Enable connection tracking flow accounting 2942 0 to disable accounting 2943 1 to enable accounting 2944 Default value is 0. 2945 2946 nfsaddrs= [NFS] Deprecated. Use ip= instead. 2947 See Documentation/admin-guide/nfs/nfsroot.rst. 2948 2949 nfsroot= [NFS] nfs root filesystem for disk-less boxes. 2950 See Documentation/admin-guide/nfs/nfsroot.rst. 2951 2952 nfsrootdebug [NFS] enable nfsroot debugging messages. 2953 See Documentation/admin-guide/nfs/nfsroot.rst. 2954 2955 nfs.callback_nr_threads= 2956 [NFSv4] set the total number of threads that the 2957 NFS client will assign to service NFSv4 callback 2958 requests. 2959 2960 nfs.callback_tcpport= 2961 [NFS] set the TCP port on which the NFSv4 callback 2962 channel should listen. 2963 2964 nfs.cache_getent= 2965 [NFS] sets the pathname to the program which is used 2966 to update the NFS client cache entries. 2967 2968 nfs.cache_getent_timeout= 2969 [NFS] sets the timeout after which an attempt to 2970 update a cache entry is deemed to have failed. 2971 2972 nfs.idmap_cache_timeout= 2973 [NFS] set the maximum lifetime for idmapper cache 2974 entries. 2975 2976 nfs.enable_ino64= 2977 [NFS] enable 64-bit inode numbers. 2978 If zero, the NFS client will fake up a 32-bit inode 2979 number for the readdir() and stat() syscalls instead 2980 of returning the full 64-bit number. 2981 The default is to return 64-bit inode numbers. 2982 2983 nfs.max_session_cb_slots= 2984 [NFSv4.1] Sets the maximum number of session 2985 slots the client will assign to the callback 2986 channel. This determines the maximum number of 2987 callbacks the client will process in parallel for 2988 a particular server. 2989 2990 nfs.max_session_slots= 2991 [NFSv4.1] Sets the maximum number of session slots 2992 the client will attempt to negotiate with the server. 2993 This limits the number of simultaneous RPC requests 2994 that the client can send to the NFSv4.1 server. 2995 Note that there is little point in setting this 2996 value higher than the max_tcp_slot_table_limit. 2997 2998 nfs.nfs4_disable_idmapping= 2999 [NFSv4] When set to the default of '1', this option 3000 ensures that both the RPC level authentication 3001 scheme and the NFS level operations agree to use 3002 numeric uids/gids if the mount is using the 3003 'sec=sys' security flavour. In effect it is 3004 disabling idmapping, which can make migration from 3005 legacy NFSv2/v3 systems to NFSv4 easier. 3006 Servers that do not support this mode of operation 3007 will be autodetected by the client, and it will fall 3008 back to using the idmapper. 3009 To turn off this behaviour, set the value to '0'. 3010 nfs.nfs4_unique_id= 3011 [NFS4] Specify an additional fixed unique ident- 3012 ification string that NFSv4 clients can insert into 3013 their nfs_client_id4 string. This is typically a 3014 UUID that is generated at system install time. 3015 3016 nfs.send_implementation_id = 3017 [NFSv4.1] Send client implementation identification 3018 information in exchange_id requests. 3019 If zero, no implementation identification information 3020 will be sent. 3021 The default is to send the implementation identification 3022 information. 3023 3024 nfs.recover_lost_locks = 3025 [NFSv4] Attempt to recover locks that were lost due 3026 to a lease timeout on the server. Please note that 3027 doing this risks data corruption, since there are 3028 no guarantees that the file will remain unchanged 3029 after the locks are lost. 3030 If you want to enable the kernel legacy behaviour of 3031 attempting to recover these locks, then set this 3032 parameter to '1'. 3033 The default parameter value of '0' causes the kernel 3034 not to attempt recovery of lost locks. 3035 3036 nfs4.layoutstats_timer = 3037 [NFSv4.2] Change the rate at which the kernel sends 3038 layoutstats to the pNFS metadata server. 3039 3040 Setting this to value to 0 causes the kernel to use 3041 whatever value is the default set by the layout 3042 driver. A non-zero value sets the minimum interval 3043 in seconds between layoutstats transmissions. 3044 3045 nfsd.nfs4_disable_idmapping= 3046 [NFSv4] When set to the default of '1', the NFSv4 3047 server will return only numeric uids and gids to 3048 clients using auth_sys, and will accept numeric uids 3049 and gids from such clients. This is intended to ease 3050 migration from NFSv2/v3. 3051 3052 nmi_debug= [KNL,SH] Specify one or more actions to take 3053 when a NMI is triggered. 3054 Format: [state][,regs][,debounce][,die] 3055 3056 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels 3057 Format: [panic,][nopanic,][num] 3058 Valid num: 0 or 1 3059 0 - turn hardlockup detector in nmi_watchdog off 3060 1 - turn hardlockup detector in nmi_watchdog on 3061 When panic is specified, panic when an NMI watchdog 3062 timeout occurs (or 'nopanic' to not panic on an NMI 3063 watchdog, if CONFIG_BOOTPARAM_HARDLOCKUP_PANIC is set) 3064 To disable both hard and soft lockup detectors, 3065 please see 'nowatchdog'. 3066 This is useful when you use a panic=... timeout and 3067 need the box quickly up again. 3068 3069 These settings can be accessed at runtime via 3070 the nmi_watchdog and hardlockup_panic sysctls. 3071 3072 netpoll.carrier_timeout= 3073 [NET] Specifies amount of time (in seconds) that 3074 netpoll should wait for a carrier. By default netpoll 3075 waits 4 seconds. 3076 3077 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths 3078 emulation library even if a 387 maths coprocessor 3079 is present. 3080 3081 no5lvl [X86-64] Disable 5-level paging mode. Forces 3082 kernel to use 4-level paging instead. 3083 3084 no_console_suspend 3085 [HW] Never suspend the console 3086 Disable suspending of consoles during suspend and 3087 hibernate operations. Once disabled, debugging 3088 messages can reach various consoles while the rest 3089 of the system is being put to sleep (ie, while 3090 debugging driver suspend/resume hooks). This may 3091 not work reliably with all consoles, but is known 3092 to work with serial and VGA consoles. 3093 To facilitate more flexible debugging, we also add 3094 console_suspend, a printk module parameter to control 3095 it. Users could use console_suspend (usually 3096 /sys/module/printk/parameters/console_suspend) to 3097 turn on/off it dynamically. 3098 3099 novmcoredd [KNL,KDUMP] 3100 Disable device dump. Device dump allows drivers to 3101 append dump data to vmcore so you can collect driver 3102 specified debug info. Drivers can append the data 3103 without any limit and this data is stored in memory, 3104 so this may cause significant memory stress. Disabling 3105 device dump can help save memory but the driver debug 3106 data will be no longer available. This parameter 3107 is only available when CONFIG_PROC_VMCORE_DEVICE_DUMP 3108 is set. 3109 3110 noaliencache [MM, NUMA, SLAB] Disables the allocation of alien 3111 caches in the slab allocator. Saves per-node memory, 3112 but will impact performance. 3113 3114 noalign [KNL,ARM] 3115 3116 noaltinstr [S390] Disables alternative instructions patching 3117 (CPU alternatives feature). 3118 3119 noapic [SMP,APIC] Tells the kernel to not make use of any 3120 IOAPICs that may be present in the system. 3121 3122 noautogroup Disable scheduler automatic task group creation. 3123 3124 nobats [PPC] Do not use BATs for mapping kernel lowmem 3125 on "Classic" PPC cores. 3126 3127 nocache [ARM] 3128 3129 noclflush [BUGS=X86] Don't use the CLFLUSH instruction 3130 3131 nodelayacct [KNL] Disable per-task delay accounting 3132 3133 nodsp [SH] Disable hardware DSP at boot time. 3134 3135 noefi Disable EFI runtime services support. 3136 3137 noexec [IA-64] 3138 3139 noexec [X86] 3140 On X86-32 available only on PAE configured kernels. 3141 noexec=on: enable non-executable mappings (default) 3142 noexec=off: disable non-executable mappings 3143 3144 nosmap [X86,PPC] 3145 Disable SMAP (Supervisor Mode Access Prevention) 3146 even if it is supported by processor. 3147 3148 nosmep [X86,PPC] 3149 Disable SMEP (Supervisor Mode Execution Prevention) 3150 even if it is supported by processor. 3151 3152 noexec32 [X86-64] 3153 This affects only 32-bit executables. 3154 noexec32=on: enable non-executable mappings (default) 3155 read doesn't imply executable mappings 3156 noexec32=off: disable non-executable mappings 3157 read implies executable mappings 3158 3159 nofpu [MIPS,SH] Disable hardware FPU at boot time. 3160 3161 nofxsr [BUGS=X86-32] Disables x86 floating point extended 3162 register save and restore. The kernel will only save 3163 legacy floating-point registers on task switch. 3164 3165 nohugeiomap [KNL,x86,PPC] Disable kernel huge I/O mappings. 3166 3167 nosmt [KNL,S390] Disable symmetric multithreading (SMT). 3168 Equivalent to smt=1. 3169 3170 [KNL,x86] Disable symmetric multithreading (SMT). 3171 nosmt=force: Force disable SMT, cannot be undone 3172 via the sysfs control file. 3173 3174 nospectre_v1 [X86,PPC] Disable mitigations for Spectre Variant 1 3175 (bounds check bypass). With this option data leaks are 3176 possible in the system. 3177 3178 nospectre_v2 [X86,PPC_FSL_BOOK3E,ARM64] Disable all mitigations for 3179 the Spectre variant 2 (indirect branch prediction) 3180 vulnerability. System may allow data leaks with this 3181 option. 3182 3183 nospec_store_bypass_disable 3184 [HW] Disable all mitigations for the Speculative Store Bypass vulnerability 3185 3186 noxsave [BUGS=X86] Disables x86 extended register state save 3187 and restore using xsave. The kernel will fallback to 3188 enabling legacy floating-point and sse state. 3189 3190 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended 3191 register states. The kernel will fall back to use 3192 xsave to save the states. By using this parameter, 3193 performance of saving the states is degraded because 3194 xsave doesn't support modified optimization while 3195 xsaveopt supports it on xsaveopt enabled systems. 3196 3197 noxsaves [X86] Disables xsaves and xrstors used in saving and 3198 restoring x86 extended register state in compacted 3199 form of xsave area. The kernel will fall back to use 3200 xsaveopt and xrstor to save and restore the states 3201 in standard form of xsave area. By using this 3202 parameter, xsave area per process might occupy more 3203 memory on xsaves enabled systems. 3204 3205 nohlt [BUGS=ARM,SH] Tells the kernel that the sleep(SH) or 3206 wfi(ARM) instruction doesn't work correctly and not to 3207 use it. This is also useful when using JTAG debugger. 3208 3209 no_file_caps Tells the kernel not to honor file capabilities. The 3210 only way then for a file to be executed with privilege 3211 is to be setuid root or executed by root. 3212 3213 nohalt [IA-64] Tells the kernel not to use the power saving 3214 function PAL_HALT_LIGHT when idle. This increases 3215 power-consumption. On the positive side, it reduces 3216 interrupt wake-up latency, which may improve performance 3217 in certain environments such as networked servers or 3218 real-time systems. 3219 3220 nohibernate [HIBERNATION] Disable hibernation and resume. 3221 3222 nohz= [KNL] Boottime enable/disable dynamic ticks 3223 Valid arguments: on, off 3224 Default: on 3225 3226 nohz_full= [KNL,BOOT,SMP,ISOL] 3227 The argument is a cpu list, as described above. 3228 In kernels built with CONFIG_NO_HZ_FULL=y, set 3229 the specified list of CPUs whose tick will be stopped 3230 whenever possible. The boot CPU will be forced outside 3231 the range to maintain the timekeeping. Any CPUs 3232 in this list will have their RCU callbacks offloaded, 3233 just as if they had also been called out in the 3234 rcu_nocbs= boot parameter. 3235 3236 noiotrap [SH] Disables trapped I/O port accesses. 3237 3238 noirqdebug [X86-32] Disables the code which attempts to detect and 3239 disable unhandled interrupt sources. 3240 3241 no_timer_check [X86,APIC] Disables the code which tests for 3242 broken timer IRQ sources. 3243 3244 noisapnp [ISAPNP] Disables ISA PnP code. 3245 3246 noinitrd [RAM] Tells the kernel not to load any configured 3247 initial RAM disk. 3248 3249 nointremap [X86-64, Intel-IOMMU] Do not enable interrupt 3250 remapping. 3251 [Deprecated - use intremap=off] 3252 3253 nointroute [IA-64] 3254 3255 noinvpcid [X86] Disable the INVPCID cpu feature. 3256 3257 nojitter [IA-64] Disables jitter checking for ITC timers. 3258 3259 no-kvmclock [X86,KVM] Disable paravirtualized KVM clock driver 3260 3261 no-kvmapf [X86,KVM] Disable paravirtualized asynchronous page 3262 fault handling. 3263 3264 no-vmw-sched-clock 3265 [X86,PV_OPS] Disable paravirtualized VMware scheduler 3266 clock and use the default one. 3267 3268 no-steal-acc [X86,PV_OPS,ARM64] Disable paravirtualized steal time 3269 accounting. steal time is computed, but won't 3270 influence scheduler behaviour 3271 3272 nolapic [X86-32,APIC] Do not enable or use the local APIC. 3273 3274 nolapic_timer [X86-32,APIC] Do not use the local APIC timer. 3275 3276 noltlbs [PPC] Do not use large page/tlb entries for kernel 3277 lowmem mapping on PPC40x and PPC8xx 3278 3279 nomca [IA-64] Disable machine check abort handling 3280 3281 nomce [X86-32] Disable Machine Check Exception 3282 3283 nomfgpt [X86-32] Disable Multi-Function General Purpose 3284 Timer usage (for AMD Geode machines). 3285 3286 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to 3287 shutdown the other cpus. Instead use the REBOOT_VECTOR 3288 irq. 3289 3290 nomodule Disable module load 3291 3292 nopat [X86] Disable PAT (page attribute table extension of 3293 pagetables) support. 3294 3295 nopcid [X86-64] Disable the PCID cpu feature. 3296 3297 norandmaps Don't use address space randomization. Equivalent to 3298 echo 0 > /proc/sys/kernel/randomize_va_space 3299 3300 noreplace-smp [X86-32,SMP] Don't replace SMP instructions 3301 with UP alternatives 3302 3303 nordrand [X86] Disable kernel use of the RDRAND and 3304 RDSEED instructions even if they are supported 3305 by the processor. RDRAND and RDSEED are still 3306 available to user space applications. 3307 3308 noresume [SWSUSP] Disables resume and restores original swap 3309 space. 3310 3311 no-scroll [VGA] Disables scrollback. 3312 This is required for the Braillex ib80-piezo Braille 3313 reader made by F.H. Papenmeier (Germany). 3314 3315 nosbagart [IA-64] 3316 3317 nosep [BUGS=X86-32] Disables x86 SYSENTER/SYSEXIT support. 3318 3319 nosmp [SMP] Tells an SMP kernel to act as a UP kernel, 3320 and disable the IO APIC. legacy for "maxcpus=0". 3321 3322 nosoftlockup [KNL] Disable the soft-lockup detector. 3323 3324 nosync [HW,M68K] Disables sync negotiation for all devices. 3325 3326 nowatchdog [KNL] Disable both lockup detectors, i.e. 3327 soft-lockup and NMI watchdog (hard-lockup). 3328 3329 nowb [ARM] 3330 3331 nox2apic [X86-64,APIC] Do not enable x2APIC mode. 3332 3333 cpu0_hotplug [X86] Turn on CPU0 hotplug feature when 3334 CONFIG_BOOTPARAM_HOTPLUG_CPU0 is off. 3335 Some features depend on CPU0. Known dependencies are: 3336 1. Resume from suspend/hibernate depends on CPU0. 3337 Suspend/hibernate will fail if CPU0 is offline and you 3338 need to online CPU0 before suspend/hibernate. 3339 2. PIC interrupts also depend on CPU0. CPU0 can't be 3340 removed if a PIC interrupt is detected. 3341 It's said poweroff/reboot may depend on CPU0 on some 3342 machines although I haven't seen such issues so far 3343 after CPU0 is offline on a few tested machines. 3344 If the dependencies are under your control, you can 3345 turn on cpu0_hotplug. 3346 3347 nps_mtm_hs_ctr= [KNL,ARC] 3348 This parameter sets the maximum duration, in 3349 cycles, each HW thread of the CTOP can run 3350 without interruptions, before HW switches it. 3351 The actual maximum duration is 16 times this 3352 parameter's value. 3353 Format: integer between 1 and 255 3354 Default: 255 3355 3356 nptcg= [IA-64] Override max number of concurrent global TLB 3357 purges which is reported from either PAL_VM_SUMMARY or 3358 SAL PALO. 3359 3360 nr_cpus= [SMP] Maximum number of processors that an SMP kernel 3361 could support. nr_cpus=n : n >= 1 limits the kernel to 3362 support 'n' processors. It could be larger than the 3363 number of already plugged CPU during bootup, later in 3364 runtime you can physically add extra cpu until it reaches 3365 n. So during boot up some boot time memory for per-cpu 3366 variables need be pre-allocated for later physical cpu 3367 hot plugging. 3368 3369 nr_uarts= [SERIAL] maximum number of UARTs to be registered. 3370 3371 numa_balancing= [KNL,X86] Enable or disable automatic NUMA balancing. 3372 Allowed values are enable and disable 3373 3374 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA. 3375 'node', 'default' can be specified 3376 This can be set from sysctl after boot. 3377 See Documentation/admin-guide/sysctl/vm.rst for details. 3378 3379 ohci1394_dma=early [HW] enable debugging via the ohci1394 driver. 3380 See Documentation/core-api/debugging-via-ohci1394.rst for more 3381 info. 3382 3383 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands 3384 Rather than timing out after 20 ms if an EC 3385 command is not properly ACKed, override the length 3386 of the timeout. We have interrupts disabled while 3387 waiting for the ACK, so if this is set too high 3388 interrupts *may* be lost! 3389 3390 omap_mux= [OMAP] Override bootloader pin multiplexing. 3391 Format: <mux_mode0.mode_name=value>... 3392 For example, to override I2C bus2: 3393 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100 3394 3395 oprofile.timer= [HW] 3396 Use timer interrupt instead of performance counters 3397 3398 oprofile.cpu_type= Force an oprofile cpu type 3399 This might be useful if you have an older oprofile 3400 userland or if you want common events. 3401 Format: { arch_perfmon } 3402 arch_perfmon: [X86] Force use of architectural 3403 perfmon on Intel CPUs instead of the 3404 CPU specific event set. 3405 timer: [X86] Force use of architectural NMI 3406 timer mode (see also oprofile.timer 3407 for generic hr timer mode) 3408 3409 oops=panic Always panic on oopses. Default is to just kill the 3410 process, but there is a small probability of 3411 deadlocking the machine. 3412 This will also cause panics on machine check exceptions. 3413 Useful together with panic=30 to trigger a reboot. 3414 3415 page_alloc.shuffle= 3416 [KNL] Boolean flag to control whether the page allocator 3417 should randomize its free lists. The randomization may 3418 be automatically enabled if the kernel detects it is 3419 running on a platform with a direct-mapped memory-side 3420 cache, and this parameter can be used to 3421 override/disable that behavior. The state of the flag 3422 can be read from sysfs at: 3423 /sys/module/page_alloc/parameters/shuffle. 3424 3425 page_owner= [KNL] Boot-time page_owner enabling option. 3426 Storage of the information about who allocated 3427 each page is disabled in default. With this switch, 3428 we can turn it on. 3429 on: enable the feature 3430 3431 page_poison= [KNL] Boot-time parameter changing the state of 3432 poisoning on the buddy allocator, available with 3433 CONFIG_PAGE_POISONING=y. 3434 off: turn off poisoning (default) 3435 on: turn on poisoning 3436 3437 panic= [KNL] Kernel behaviour on panic: delay <timeout> 3438 timeout > 0: seconds before rebooting 3439 timeout = 0: wait forever 3440 timeout < 0: reboot immediately 3441 Format: <timeout> 3442 3443 panic_print= Bitmask for printing system info when panic happens. 3444 User can chose combination of the following bits: 3445 bit 0: print all tasks info 3446 bit 1: print system memory info 3447 bit 2: print timer info 3448 bit 3: print locks info if CONFIG_LOCKDEP is on 3449 bit 4: print ftrace buffer 3450 bit 5: print all printk messages in buffer 3451 3452 panic_on_taint= Bitmask for conditionally calling panic() in add_taint() 3453 Format: <hex>[,nousertaint] 3454 Hexadecimal bitmask representing the set of TAINT flags 3455 that will cause the kernel to panic when add_taint() is 3456 called with any of the flags in this set. 3457 The optional switch "nousertaint" can be utilized to 3458 prevent userspace forced crashes by writing to sysctl 3459 /proc/sys/kernel/tainted any flagset matching with the 3460 bitmask set on panic_on_taint. 3461 See Documentation/admin-guide/tainted-kernels.rst for 3462 extra details on the taint flags that users can pick 3463 to compose the bitmask to assign to panic_on_taint. 3464 3465 panic_on_warn panic() instead of WARN(). Useful to cause kdump 3466 on a WARN(). 3467 3468 crash_kexec_post_notifiers 3469 Run kdump after running panic-notifiers and dumping 3470 kmsg. This only for the users who doubt kdump always 3471 succeeds in any situation. 3472 Note that this also increases risks of kdump failure, 3473 because some panic notifiers can make the crashed 3474 kernel more unstable. 3475 3476 parkbd.port= [HW] Parallel port number the keyboard adapter is 3477 connected to, default is 0. 3478 Format: <parport#> 3479 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation, 3480 0 for XT, 1 for AT (default is AT). 3481 Format: <mode> 3482 3483 parport= [HW,PPT] Specify parallel ports. 0 disables. 3484 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] } 3485 Use 'auto' to force the driver to use any 3486 IRQ/DMA settings detected (the default is to 3487 ignore detected IRQ/DMA settings because of 3488 possible conflicts). You can specify the base 3489 address, IRQ, and DMA settings; IRQ and DMA 3490 should be numbers, or 'auto' (for using detected 3491 settings on that particular port), or 'nofifo' 3492 (to avoid using a FIFO even if it is detected). 3493 Parallel ports are assigned in the order they 3494 are specified on the command line, starting 3495 with parport0. 3496 3497 parport_init_mode= [HW,PPT] 3498 Configure VIA parallel port to operate in 3499 a specific mode. This is necessary on Pegasos 3500 computer where firmware has no options for setting 3501 up parallel port mode and sets it to spp. 3502 Currently this function knows 686a and 8231 chips. 3503 Format: [spp|ps2|epp|ecp|ecpepp] 3504 3505 pause_on_oops= 3506 Halt all CPUs after the first oops has been printed for 3507 the specified number of seconds. This is to be used if 3508 your oopses keep scrolling off the screen. 3509 3510 pcbit= [HW,ISDN] 3511 3512 pcd. [PARIDE] 3513 See header of drivers/block/paride/pcd.c. 3514 See also Documentation/admin-guide/blockdev/paride.rst. 3515 3516 pci=option[,option...] [PCI] various PCI subsystem options. 3517 3518 Some options herein operate on a specific device 3519 or a set of devices (<pci_dev>). These are 3520 specified in one of the following formats: 3521 3522 [<domain>:]<bus>:<dev>.<func>[/<dev>.<func>]* 3523 pci:<vendor>:<device>[:<subvendor>:<subdevice>] 3524 3525 Note: the first format specifies a PCI 3526 bus/device/function address which may change 3527 if new hardware is inserted, if motherboard 3528 firmware changes, or due to changes caused 3529 by other kernel parameters. If the 3530 domain is left unspecified, it is 3531 taken to be zero. Optionally, a path 3532 to a device through multiple device/function 3533 addresses can be specified after the base 3534 address (this is more robust against 3535 renumbering issues). The second format 3536 selects devices using IDs from the 3537 configuration space which may match multiple 3538 devices in the system. 3539 3540 earlydump dump PCI config space before the kernel 3541 changes anything 3542 off [X86] don't probe for the PCI bus 3543 bios [X86-32] force use of PCI BIOS, don't access 3544 the hardware directly. Use this if your machine 3545 has a non-standard PCI host bridge. 3546 nobios [X86-32] disallow use of PCI BIOS, only direct 3547 hardware access methods are allowed. Use this 3548 if you experience crashes upon bootup and you 3549 suspect they are caused by the BIOS. 3550 conf1 [X86] Force use of PCI Configuration Access 3551 Mechanism 1 (config address in IO port 0xCF8, 3552 data in IO port 0xCFC, both 32-bit). 3553 conf2 [X86] Force use of PCI Configuration Access 3554 Mechanism 2 (IO port 0xCF8 is an 8-bit port for 3555 the function, IO port 0xCFA, also 8-bit, sets 3556 bus number. The config space is then accessed 3557 through ports 0xC000-0xCFFF). 3558 See http://wiki.osdev.org/PCI for more info 3559 on the configuration access mechanisms. 3560 noaer [PCIE] If the PCIEAER kernel config parameter is 3561 enabled, this kernel boot option can be used to 3562 disable the use of PCIE advanced error reporting. 3563 nodomains [PCI] Disable support for multiple PCI 3564 root domains (aka PCI segments, in ACPI-speak). 3565 nommconf [X86] Disable use of MMCONFIG for PCI 3566 Configuration 3567 check_enable_amd_mmconf [X86] check for and enable 3568 properly configured MMIO access to PCI 3569 config space on AMD family 10h CPU 3570 nomsi [MSI] If the PCI_MSI kernel config parameter is 3571 enabled, this kernel boot option can be used to 3572 disable the use of MSI interrupts system-wide. 3573 noioapicquirk [APIC] Disable all boot interrupt quirks. 3574 Safety option to keep boot IRQs enabled. This 3575 should never be necessary. 3576 ioapicreroute [APIC] Enable rerouting of boot IRQs to the 3577 primary IO-APIC for bridges that cannot disable 3578 boot IRQs. This fixes a source of spurious IRQs 3579 when the system masks IRQs. 3580 noioapicreroute [APIC] Disable workaround that uses the 3581 boot IRQ equivalent of an IRQ that connects to 3582 a chipset where boot IRQs cannot be disabled. 3583 The opposite of ioapicreroute. 3584 biosirq [X86-32] Use PCI BIOS calls to get the interrupt 3585 routing table. These calls are known to be buggy 3586 on several machines and they hang the machine 3587 when used, but on other computers it's the only 3588 way to get the interrupt routing table. Try 3589 this option if the kernel is unable to allocate 3590 IRQs or discover secondary PCI buses on your 3591 motherboard. 3592 rom [X86] Assign address space to expansion ROMs. 3593 Use with caution as certain devices share 3594 address decoders between ROMs and other 3595 resources. 3596 norom [X86] Do not assign address space to 3597 expansion ROMs that do not already have 3598 BIOS assigned address ranges. 3599 nobar [X86] Do not assign address space to the 3600 BARs that weren't assigned by the BIOS. 3601 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be 3602 assigned automatically to PCI devices. You can 3603 make the kernel exclude IRQs of your ISA cards 3604 this way. 3605 pirqaddr=0xAAAAA [X86] Specify the physical address 3606 of the PIRQ table (normally generated 3607 by the BIOS) if it is outside the 3608 F0000h-100000h range. 3609 lastbus=N [X86] Scan all buses thru bus #N. Can be 3610 useful if the kernel is unable to find your 3611 secondary buses and you want to tell it 3612 explicitly which ones they are. 3613 assign-busses [X86] Always assign all PCI bus 3614 numbers ourselves, overriding 3615 whatever the firmware may have done. 3616 usepirqmask [X86] Honor the possible IRQ mask stored 3617 in the BIOS $PIR table. This is needed on 3618 some systems with broken BIOSes, notably 3619 some HP Pavilion N5400 and Omnibook XE3 3620 notebooks. This will have no effect if ACPI 3621 IRQ routing is enabled. 3622 noacpi [X86] Do not use ACPI for IRQ routing 3623 or for PCI scanning. 3624 use_crs [X86] Use PCI host bridge window information 3625 from ACPI. On BIOSes from 2008 or later, this 3626 is enabled by default. If you need to use this, 3627 please report a bug. 3628 nocrs [X86] Ignore PCI host bridge windows from ACPI. 3629 If you need to use this, please report a bug. 3630 routeirq Do IRQ routing for all PCI devices. 3631 This is normally done in pci_enable_device(), 3632 so this option is a temporary workaround 3633 for broken drivers that don't call it. 3634 skip_isa_align [X86] do not align io start addr, so can 3635 handle more pci cards 3636 noearly [X86] Don't do any early type 1 scanning. 3637 This might help on some broken boards which 3638 machine check when some devices' config space 3639 is read. But various workarounds are disabled 3640 and some IOMMU drivers will not work. 3641 bfsort Sort PCI devices into breadth-first order. 3642 This sorting is done to get a device 3643 order compatible with older (<= 2.4) kernels. 3644 nobfsort Don't sort PCI devices into breadth-first order. 3645 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size) 3646 tuning and use the BIOS-configured MPS defaults. 3647 pcie_bus_safe Set every device's MPS to the largest value 3648 supported by all devices below the root complex. 3649 pcie_bus_perf Set device MPS to the largest allowable MPS 3650 based on its parent bus. Also set MRRS (Max 3651 Read Request Size) to the largest supported 3652 value (no larger than the MPS that the device 3653 or bus can support) for best performance. 3654 pcie_bus_peer2peer Set every device's MPS to 128B, which 3655 every device is guaranteed to support. This 3656 configuration allows peer-to-peer DMA between 3657 any pair of devices, possibly at the cost of 3658 reduced performance. This also guarantees 3659 that hot-added devices will work. 3660 cbiosize=nn[KMG] The fixed amount of bus space which is 3661 reserved for the CardBus bridge's IO window. 3662 The default value is 256 bytes. 3663 cbmemsize=nn[KMG] The fixed amount of bus space which is 3664 reserved for the CardBus bridge's memory 3665 window. The default value is 64 megabytes. 3666 resource_alignment= 3667 Format: 3668 [<order of align>@]<pci_dev>[; ...] 3669 Specifies alignment and device to reassign 3670 aligned memory resources. How to 3671 specify the device is described above. 3672 If <order of align> is not specified, 3673 PAGE_SIZE is used as alignment. 3674 A PCI-PCI bridge can be specified if resource 3675 windows need to be expanded. 3676 To specify the alignment for several 3677 instances of a device, the PCI vendor, 3678 device, subvendor, and subdevice may be 3679 specified, e.g., 12@pci:8086:9c22:103c:198f 3680 for 4096-byte alignment. 3681 ecrc= Enable/disable PCIe ECRC (transaction layer 3682 end-to-end CRC checking). 3683 bios: Use BIOS/firmware settings. This is the 3684 the default. 3685 off: Turn ECRC off 3686 on: Turn ECRC on. 3687 hpiosize=nn[KMG] The fixed amount of bus space which is 3688 reserved for hotplug bridge's IO window. 3689 Default size is 256 bytes. 3690 hpmmiosize=nn[KMG] The fixed amount of bus space which is 3691 reserved for hotplug bridge's MMIO window. 3692 Default size is 2 megabytes. 3693 hpmmioprefsize=nn[KMG] The fixed amount of bus space which is 3694 reserved for hotplug bridge's MMIO_PREF window. 3695 Default size is 2 megabytes. 3696 hpmemsize=nn[KMG] The fixed amount of bus space which is 3697 reserved for hotplug bridge's MMIO and 3698 MMIO_PREF window. 3699 Default size is 2 megabytes. 3700 hpbussize=nn The minimum amount of additional bus numbers 3701 reserved for buses below a hotplug bridge. 3702 Default is 1. 3703 realloc= Enable/disable reallocating PCI bridge resources 3704 if allocations done by BIOS are too small to 3705 accommodate resources required by all child 3706 devices. 3707 off: Turn realloc off 3708 on: Turn realloc on 3709 realloc same as realloc=on 3710 noari do not use PCIe ARI. 3711 noats [PCIE, Intel-IOMMU, AMD-IOMMU] 3712 do not use PCIe ATS (and IOMMU device IOTLB). 3713 pcie_scan_all Scan all possible PCIe devices. Otherwise we 3714 only look for one device below a PCIe downstream 3715 port. 3716 big_root_window Try to add a big 64bit memory window to the PCIe 3717 root complex on AMD CPUs. Some GFX hardware 3718 can resize a BAR to allow access to all VRAM. 3719 Adding the window is slightly risky (it may 3720 conflict with unreported devices), so this 3721 taints the kernel. 3722 disable_acs_redir=<pci_dev>[; ...] 3723 Specify one or more PCI devices (in the format 3724 specified above) separated by semicolons. 3725 Each device specified will have the PCI ACS 3726 redirect capabilities forced off which will 3727 allow P2P traffic between devices through 3728 bridges without forcing it upstream. Note: 3729 this removes isolation between devices and 3730 may put more devices in an IOMMU group. 3731 force_floating [S390] Force usage of floating interrupts. 3732 nomio [S390] Do not use MIO instructions. 3733 norid [S390] ignore the RID field and force use of 3734 one PCI domain per PCI function 3735 3736 pcie_aspm= [PCIE] Forcibly enable or disable PCIe Active State Power 3737 Management. 3738 off Disable ASPM. 3739 force Enable ASPM even on devices that claim not to support it. 3740 WARNING: Forcing ASPM on may cause system lockups. 3741 3742 pcie_ports= [PCIE] PCIe port services handling: 3743 native Use native PCIe services (PME, AER, DPC, PCIe hotplug) 3744 even if the platform doesn't give the OS permission to 3745 use them. This may cause conflicts if the platform 3746 also tries to use these services. 3747 dpc-native Use native PCIe service for DPC only. May 3748 cause conflicts if firmware uses AER or DPC. 3749 compat Disable native PCIe services (PME, AER, DPC, PCIe 3750 hotplug). 3751 3752 pcie_port_pm= [PCIE] PCIe port power management handling: 3753 off Disable power management of all PCIe ports 3754 force Forcibly enable power management of all PCIe ports 3755 3756 pcie_pme= [PCIE,PM] Native PCIe PME signaling options: 3757 nomsi Do not use MSI for native PCIe PME signaling (this makes 3758 all PCIe root ports use INTx for all services). 3759 3760 pcmv= [HW,PCMCIA] BadgePAD 4 3761 3762 pd_ignore_unused 3763 [PM] 3764 Keep all power-domains already enabled by bootloader on, 3765 even if no driver has claimed them. This is useful 3766 for debug and development, but should not be 3767 needed on a platform with proper driver support. 3768 3769 pd. [PARIDE] 3770 See Documentation/admin-guide/blockdev/paride.rst. 3771 3772 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at 3773 boot time. 3774 Format: { 0 | 1 } 3775 See arch/parisc/kernel/pdc_chassis.c 3776 3777 percpu_alloc= Select which percpu first chunk allocator to use. 3778 Currently supported values are "embed" and "page". 3779 Archs may support subset or none of the selections. 3780 See comments in mm/percpu.c for details on each 3781 allocator. This parameter is primarily for debugging 3782 and performance comparison. 3783 3784 pf. [PARIDE] 3785 See Documentation/admin-guide/blockdev/paride.rst. 3786 3787 pg. [PARIDE] 3788 See Documentation/admin-guide/blockdev/paride.rst. 3789 3790 pirq= [SMP,APIC] Manual mp-table setup 3791 See Documentation/x86/i386/IO-APIC.rst. 3792 3793 plip= [PPT,NET] Parallel port network link 3794 Format: { parport<nr> | timid | 0 } 3795 See also Documentation/admin-guide/parport.rst. 3796 3797 pmtmr= [X86] Manual setup of pmtmr I/O Port. 3798 Override pmtimer IOPort with a hex value. 3799 e.g. pmtmr=0x508 3800 3801 pm_debug_messages [SUSPEND,KNL] 3802 Enable suspend/resume debug messages during boot up. 3803 3804 pnp.debug=1 [PNP] 3805 Enable PNP debug messages (depends on the 3806 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time 3807 via /sys/module/pnp/parameters/debug. We always show 3808 current resource usage; turning this on also shows 3809 possible settings and some assignment information. 3810 3811 pnpacpi= [ACPI] 3812 { off } 3813 3814 pnpbios= [ISAPNP] 3815 { on | off | curr | res | no-curr | no-res } 3816 3817 pnp_reserve_irq= 3818 [ISAPNP] Exclude IRQs for the autoconfiguration 3819 3820 pnp_reserve_dma= 3821 [ISAPNP] Exclude DMAs for the autoconfiguration 3822 3823 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration 3824 Ranges are in pairs (I/O port base and size). 3825 3826 pnp_reserve_mem= 3827 [ISAPNP] Exclude memory regions for the 3828 autoconfiguration. 3829 Ranges are in pairs (memory base and size). 3830 3831 ports= [IP_VS_FTP] IPVS ftp helper module 3832 Default is 21. 3833 Up to 8 (IP_VS_APP_MAX_PORTS) ports 3834 may be specified. 3835 Format: <port>,<port>.... 3836 3837 powersave=off [PPC] This option disables power saving features. 3838 It specifically disables cpuidle and sets the 3839 platform machine description specific power_save 3840 function to NULL. On Idle the CPU just reduces 3841 execution priority. 3842 3843 ppc_strict_facility_enable 3844 [PPC] This option catches any kernel floating point, 3845 Altivec, VSX and SPE outside of regions specifically 3846 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()). 3847 There is some performance impact when enabling this. 3848 3849 ppc_tm= [PPC] 3850 Format: {"off"} 3851 Disable Hardware Transactional Memory 3852 3853 print-fatal-signals= 3854 [KNL] debug: print fatal signals 3855 3856 If enabled, warn about various signal handling 3857 related application anomalies: too many signals, 3858 too many POSIX.1 timers, fatal signals causing a 3859 coredump - etc. 3860 3861 If you hit the warning due to signal overflow, 3862 you might want to try "ulimit -i unlimited". 3863 3864 default: off. 3865 3866 printk.always_kmsg_dump= 3867 Trigger kmsg_dump for cases other than kernel oops or 3868 panics 3869 Format: <bool> (1/Y/y=enable, 0/N/n=disable) 3870 default: disabled 3871 3872 printk.devkmsg={on,off,ratelimit} 3873 Control writing to /dev/kmsg. 3874 on - unlimited logging to /dev/kmsg from userspace 3875 off - logging to /dev/kmsg disabled 3876 ratelimit - ratelimit the logging 3877 Default: ratelimit 3878 3879 printk.time= Show timing data prefixed to each printk message line 3880 Format: <bool> (1/Y/y=enable, 0/N/n=disable) 3881 3882 processor.max_cstate= [HW,ACPI] 3883 Limit processor to maximum C-state 3884 max_cstate=9 overrides any DMI blacklist limit. 3885 3886 processor.nocst [HW,ACPI] 3887 Ignore the _CST method to determine C-states, 3888 instead using the legacy FADT method 3889 3890 profile= [KNL] Enable kernel profiling via /proc/profile 3891 Format: [<profiletype>,]<number> 3892 Param: <profiletype>: "schedule", "sleep", or "kvm" 3893 [defaults to kernel profiling] 3894 Param: "schedule" - profile schedule points. 3895 Param: "sleep" - profile D-state sleeping (millisecs). 3896 Requires CONFIG_SCHEDSTATS 3897 Param: "kvm" - profile VM exits. 3898 Param: <number> - step/bucket size as a power of 2 for 3899 statistical time based profiling. 3900 3901 prompt_ramdisk= [RAM] List of RAM disks to prompt for floppy disk 3902 before loading. 3903 See Documentation/admin-guide/blockdev/ramdisk.rst. 3904 3905 prot_virt= [S390] enable hosting protected virtual machines 3906 isolated from the hypervisor (if hardware supports 3907 that). 3908 Format: <bool> 3909 3910 psi= [KNL] Enable or disable pressure stall information 3911 tracking. 3912 Format: <bool> 3913 3914 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to 3915 probe for; one of (bare|imps|exps|lifebook|any). 3916 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports 3917 per second. 3918 psmouse.resetafter= [HW,MOUSE] 3919 Try to reset the device after so many bad packets 3920 (0 = never). 3921 psmouse.resolution= 3922 [HW,MOUSE] Set desired mouse resolution, in dpi. 3923 psmouse.smartscroll= 3924 [HW,MOUSE] Controls Logitech smartscroll autorepeat. 3925 0 = disabled, 1 = enabled (default). 3926 3927 pstore.backend= Specify the name of the pstore backend to use 3928 3929 pt. [PARIDE] 3930 See Documentation/admin-guide/blockdev/paride.rst. 3931 3932 pti= [X86_64] Control Page Table Isolation of user and 3933 kernel address spaces. Disabling this feature 3934 removes hardening, but improves performance of 3935 system calls and interrupts. 3936 3937 on - unconditionally enable 3938 off - unconditionally disable 3939 auto - kernel detects whether your CPU model is 3940 vulnerable to issues that PTI mitigates 3941 3942 Not specifying this option is equivalent to pti=auto. 3943 3944 nopti [X86_64] 3945 Equivalent to pti=off 3946 3947 pty.legacy_count= 3948 [KNL] Number of legacy pty's. Overwrites compiled-in 3949 default number. 3950 3951 quiet [KNL] Disable most log messages 3952 3953 r128= [HW,DRM] 3954 3955 raid= [HW,RAID] 3956 See Documentation/admin-guide/md.rst. 3957 3958 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes 3959 See Documentation/admin-guide/blockdev/ramdisk.rst. 3960 3961 random.trust_cpu={on,off} 3962 [KNL] Enable or disable trusting the use of the 3963 CPU's random number generator (if available) to 3964 fully seed the kernel's CRNG. Default is controlled 3965 by CONFIG_RANDOM_TRUST_CPU. 3966 3967 ras=option[,option,...] [KNL] RAS-specific options 3968 3969 cec_disable [X86] 3970 Disable the Correctable Errors Collector, 3971 see CONFIG_RAS_CEC help text. 3972 3973 rcu_nocbs= [KNL] 3974 The argument is a cpu list, as described above, 3975 except that the string "all" can be used to 3976 specify every CPU on the system. 3977 3978 In kernels built with CONFIG_RCU_NOCB_CPU=y, set 3979 the specified list of CPUs to be no-callback CPUs. 3980 Invocation of these CPUs' RCU callbacks will be 3981 offloaded to "rcuox/N" kthreads created for that 3982 purpose, where "x" is "p" for RCU-preempt, and 3983 "s" for RCU-sched, and "N" is the CPU number. 3984 This reduces OS jitter on the offloaded CPUs, 3985 which can be useful for HPC and real-time 3986 workloads. It can also improve energy efficiency 3987 for asymmetric multiprocessors. 3988 3989 rcu_nocb_poll [KNL] 3990 Rather than requiring that offloaded CPUs 3991 (specified by rcu_nocbs= above) explicitly 3992 awaken the corresponding "rcuoN" kthreads, 3993 make these kthreads poll for callbacks. 3994 This improves the real-time response for the 3995 offloaded CPUs by relieving them of the need to 3996 wake up the corresponding kthread, but degrades 3997 energy efficiency by requiring that the kthreads 3998 periodically wake up to do the polling. 3999 4000 rcutree.blimit= [KNL] 4001 Set maximum number of finished RCU callbacks to 4002 process in one batch. 4003 4004 rcutree.dump_tree= [KNL] 4005 Dump the structure of the rcu_node combining tree 4006 out at early boot. This is used for diagnostic 4007 purposes, to verify correct tree setup. 4008 4009 rcutree.gp_cleanup_delay= [KNL] 4010 Set the number of jiffies to delay each step of 4011 RCU grace-period cleanup. 4012 4013 rcutree.gp_init_delay= [KNL] 4014 Set the number of jiffies to delay each step of 4015 RCU grace-period initialization. 4016 4017 rcutree.gp_preinit_delay= [KNL] 4018 Set the number of jiffies to delay each step of 4019 RCU grace-period pre-initialization, that is, 4020 the propagation of recent CPU-hotplug changes up 4021 the rcu_node combining tree. 4022 4023 rcutree.use_softirq= [KNL] 4024 If set to zero, move all RCU_SOFTIRQ processing to 4025 per-CPU rcuc kthreads. Defaults to a non-zero 4026 value, meaning that RCU_SOFTIRQ is used by default. 4027 Specify rcutree.use_softirq=0 to use rcuc kthreads. 4028 4029 rcutree.rcu_fanout_exact= [KNL] 4030 Disable autobalancing of the rcu_node combining 4031 tree. This is used by rcutorture, and might 4032 possibly be useful for architectures having high 4033 cache-to-cache transfer latencies. 4034 4035 rcutree.rcu_fanout_leaf= [KNL] 4036 Change the number of CPUs assigned to each 4037 leaf rcu_node structure. Useful for very 4038 large systems, which will choose the value 64, 4039 and for NUMA systems with large remote-access 4040 latencies, which will choose a value aligned 4041 with the appropriate hardware boundaries. 4042 4043 rcutree.jiffies_till_first_fqs= [KNL] 4044 Set delay from grace-period initialization to 4045 first attempt to force quiescent states. 4046 Units are jiffies, minimum value is zero, 4047 and maximum value is HZ. 4048 4049 rcutree.jiffies_till_next_fqs= [KNL] 4050 Set delay between subsequent attempts to force 4051 quiescent states. Units are jiffies, minimum 4052 value is one, and maximum value is HZ. 4053 4054 rcutree.jiffies_till_sched_qs= [KNL] 4055 Set required age in jiffies for a 4056 given grace period before RCU starts 4057 soliciting quiescent-state help from 4058 rcu_note_context_switch() and cond_resched(). 4059 If not specified, the kernel will calculate 4060 a value based on the most recent settings 4061 of rcutree.jiffies_till_first_fqs 4062 and rcutree.jiffies_till_next_fqs. 4063 This calculated value may be viewed in 4064 rcutree.jiffies_to_sched_qs. Any attempt to set 4065 rcutree.jiffies_to_sched_qs will be cheerfully 4066 overwritten. 4067 4068 rcutree.kthread_prio= [KNL,BOOT] 4069 Set the SCHED_FIFO priority of the RCU per-CPU 4070 kthreads (rcuc/N). This value is also used for 4071 the priority of the RCU boost threads (rcub/N) 4072 and for the RCU grace-period kthreads (rcu_bh, 4073 rcu_preempt, and rcu_sched). If RCU_BOOST is 4074 set, valid values are 1-99 and the default is 1 4075 (the least-favored priority). Otherwise, when 4076 RCU_BOOST is not set, valid values are 0-99 and 4077 the default is zero (non-realtime operation). 4078 4079 rcutree.rcu_nocb_gp_stride= [KNL] 4080 Set the number of NOCB callback kthreads in 4081 each group, which defaults to the square root 4082 of the number of CPUs. Larger numbers reduce 4083 the wakeup overhead on the global grace-period 4084 kthread, but increases that same overhead on 4085 each group's NOCB grace-period kthread. 4086 4087 rcutree.qhimark= [KNL] 4088 Set threshold of queued RCU callbacks beyond which 4089 batch limiting is disabled. 4090 4091 rcutree.qlowmark= [KNL] 4092 Set threshold of queued RCU callbacks below which 4093 batch limiting is re-enabled. 4094 4095 rcutree.qovld= [KNL] 4096 Set threshold of queued RCU callbacks beyond which 4097 RCU's force-quiescent-state scan will aggressively 4098 enlist help from cond_resched() and sched IPIs to 4099 help CPUs more quickly reach quiescent states. 4100 Set to less than zero to make this be set based 4101 on rcutree.qhimark at boot time and to zero to 4102 disable more aggressive help enlistment. 4103 4104 rcutree.rcu_idle_gp_delay= [KNL] 4105 Set wakeup interval for idle CPUs that have 4106 RCU callbacks (RCU_FAST_NO_HZ=y). 4107 4108 rcutree.rcu_idle_lazy_gp_delay= [KNL] 4109 Set wakeup interval for idle CPUs that have 4110 only "lazy" RCU callbacks (RCU_FAST_NO_HZ=y). 4111 Lazy RCU callbacks are those which RCU can 4112 prove do nothing more than free memory. 4113 4114 rcutree.rcu_kick_kthreads= [KNL] 4115 Cause the grace-period kthread to get an extra 4116 wake_up() if it sleeps three times longer than 4117 it should at force-quiescent-state time. 4118 This wake_up() will be accompanied by a 4119 WARN_ONCE() splat and an ftrace_dump(). 4120 4121 rcutree.sysrq_rcu= [KNL] 4122 Commandeer a sysrq key to dump out Tree RCU's 4123 rcu_node tree with an eye towards determining 4124 why a new grace period has not yet started. 4125 4126 rcuperf.gp_async= [KNL] 4127 Measure performance of asynchronous 4128 grace-period primitives such as call_rcu(). 4129 4130 rcuperf.gp_async_max= [KNL] 4131 Specify the maximum number of outstanding 4132 callbacks per writer thread. When a writer 4133 thread exceeds this limit, it invokes the 4134 corresponding flavor of rcu_barrier() to allow 4135 previously posted callbacks to drain. 4136 4137 rcuperf.gp_exp= [KNL] 4138 Measure performance of expedited synchronous 4139 grace-period primitives. 4140 4141 rcuperf.holdoff= [KNL] 4142 Set test-start holdoff period. The purpose of 4143 this parameter is to delay the start of the 4144 test until boot completes in order to avoid 4145 interference. 4146 4147 rcuperf.kfree_rcu_test= [KNL] 4148 Set to measure performance of kfree_rcu() flooding. 4149 4150 rcuperf.kfree_nthreads= [KNL] 4151 The number of threads running loops of kfree_rcu(). 4152 4153 rcuperf.kfree_alloc_num= [KNL] 4154 Number of allocations and frees done in an iteration. 4155 4156 rcuperf.kfree_loops= [KNL] 4157 Number of loops doing rcuperf.kfree_alloc_num number 4158 of allocations and frees. 4159 4160 rcuperf.nreaders= [KNL] 4161 Set number of RCU readers. The value -1 selects 4162 N, where N is the number of CPUs. A value 4163 "n" less than -1 selects N-n+1, where N is again 4164 the number of CPUs. For example, -2 selects N 4165 (the number of CPUs), -3 selects N+1, and so on. 4166 A value of "n" less than or equal to -N selects 4167 a single reader. 4168 4169 rcuperf.nwriters= [KNL] 4170 Set number of RCU writers. The values operate 4171 the same as for rcuperf.nreaders. 4172 N, where N is the number of CPUs 4173 4174 rcuperf.perf_type= [KNL] 4175 Specify the RCU implementation to test. 4176 4177 rcuperf.shutdown= [KNL] 4178 Shut the system down after performance tests 4179 complete. This is useful for hands-off automated 4180 testing. 4181 4182 rcuperf.verbose= [KNL] 4183 Enable additional printk() statements. 4184 4185 rcuperf.writer_holdoff= [KNL] 4186 Write-side holdoff between grace periods, 4187 in microseconds. The default of zero says 4188 no holdoff. 4189 4190 rcutorture.fqs_duration= [KNL] 4191 Set duration of force_quiescent_state bursts 4192 in microseconds. 4193 4194 rcutorture.fqs_holdoff= [KNL] 4195 Set holdoff time within force_quiescent_state bursts 4196 in microseconds. 4197 4198 rcutorture.fqs_stutter= [KNL] 4199 Set wait time between force_quiescent_state bursts 4200 in seconds. 4201 4202 rcutorture.fwd_progress= [KNL] 4203 Enable RCU grace-period forward-progress testing 4204 for the types of RCU supporting this notion. 4205 4206 rcutorture.fwd_progress_div= [KNL] 4207 Specify the fraction of a CPU-stall-warning 4208 period to do tight-loop forward-progress testing. 4209 4210 rcutorture.fwd_progress_holdoff= [KNL] 4211 Number of seconds to wait between successive 4212 forward-progress tests. 4213 4214 rcutorture.fwd_progress_need_resched= [KNL] 4215 Enclose cond_resched() calls within checks for 4216 need_resched() during tight-loop forward-progress 4217 testing. 4218 4219 rcutorture.gp_cond= [KNL] 4220 Use conditional/asynchronous update-side 4221 primitives, if available. 4222 4223 rcutorture.gp_exp= [KNL] 4224 Use expedited update-side primitives, if available. 4225 4226 rcutorture.gp_normal= [KNL] 4227 Use normal (non-expedited) asynchronous 4228 update-side primitives, if available. 4229 4230 rcutorture.gp_sync= [KNL] 4231 Use normal (non-expedited) synchronous 4232 update-side primitives, if available. If all 4233 of rcutorture.gp_cond=, rcutorture.gp_exp=, 4234 rcutorture.gp_normal=, and rcutorture.gp_sync= 4235 are zero, rcutorture acts as if is interpreted 4236 they are all non-zero. 4237 4238 rcutorture.n_barrier_cbs= [KNL] 4239 Set callbacks/threads for rcu_barrier() testing. 4240 4241 rcutorture.nfakewriters= [KNL] 4242 Set number of concurrent RCU writers. These just 4243 stress RCU, they don't participate in the actual 4244 test, hence the "fake". 4245 4246 rcutorture.nreaders= [KNL] 4247 Set number of RCU readers. The value -1 selects 4248 N-1, where N is the number of CPUs. A value 4249 "n" less than -1 selects N-n-2, where N is again 4250 the number of CPUs. For example, -2 selects N 4251 (the number of CPUs), -3 selects N+1, and so on. 4252 4253 rcutorture.object_debug= [KNL] 4254 Enable debug-object double-call_rcu() testing. 4255 4256 rcutorture.onoff_holdoff= [KNL] 4257 Set time (s) after boot for CPU-hotplug testing. 4258 4259 rcutorture.onoff_interval= [KNL] 4260 Set time (jiffies) between CPU-hotplug operations, 4261 or zero to disable CPU-hotplug testing. 4262 4263 rcutorture.shuffle_interval= [KNL] 4264 Set task-shuffle interval (s). Shuffling tasks 4265 allows some CPUs to go into dyntick-idle mode 4266 during the rcutorture test. 4267 4268 rcutorture.shutdown_secs= [KNL] 4269 Set time (s) after boot system shutdown. This 4270 is useful for hands-off automated testing. 4271 4272 rcutorture.stall_cpu= [KNL] 4273 Duration of CPU stall (s) to test RCU CPU stall 4274 warnings, zero to disable. 4275 4276 rcutorture.stall_cpu_block= [KNL] 4277 Sleep while stalling if set. This will result 4278 in warnings from preemptible RCU in addition 4279 to any other stall-related activity. 4280 4281 rcutorture.stall_cpu_holdoff= [KNL] 4282 Time to wait (s) after boot before inducing stall. 4283 4284 rcutorture.stall_cpu_irqsoff= [KNL] 4285 Disable interrupts while stalling if set. 4286 4287 rcutorture.stall_gp_kthread= [KNL] 4288 Duration (s) of forced sleep within RCU 4289 grace-period kthread to test RCU CPU stall 4290 warnings, zero to disable. If both stall_cpu 4291 and stall_gp_kthread are specified, the 4292 kthread is starved first, then the CPU. 4293 4294 rcutorture.stat_interval= [KNL] 4295 Time (s) between statistics printk()s. 4296 4297 rcutorture.stutter= [KNL] 4298 Time (s) to stutter testing, for example, specifying 4299 five seconds causes the test to run for five seconds, 4300 wait for five seconds, and so on. This tests RCU's 4301 ability to transition abruptly to and from idle. 4302 4303 rcutorture.test_boost= [KNL] 4304 Test RCU priority boosting? 0=no, 1=maybe, 2=yes. 4305 "Maybe" means test if the RCU implementation 4306 under test support RCU priority boosting. 4307 4308 rcutorture.test_boost_duration= [KNL] 4309 Duration (s) of each individual boost test. 4310 4311 rcutorture.test_boost_interval= [KNL] 4312 Interval (s) between each boost test. 4313 4314 rcutorture.test_no_idle_hz= [KNL] 4315 Test RCU's dyntick-idle handling. See also the 4316 rcutorture.shuffle_interval parameter. 4317 4318 rcutorture.torture_type= [KNL] 4319 Specify the RCU implementation to test. 4320 4321 rcutorture.verbose= [KNL] 4322 Enable additional printk() statements. 4323 4324 rcupdate.rcu_cpu_stall_ftrace_dump= [KNL] 4325 Dump ftrace buffer after reporting RCU CPU 4326 stall warning. 4327 4328 rcupdate.rcu_cpu_stall_suppress= [KNL] 4329 Suppress RCU CPU stall warning messages. 4330 4331 rcupdate.rcu_cpu_stall_suppress_at_boot= [KNL] 4332 Suppress RCU CPU stall warning messages and 4333 rcutorture writer stall warnings that occur 4334 during early boot, that is, during the time 4335 before the init task is spawned. 4336 4337 rcupdate.rcu_cpu_stall_timeout= [KNL] 4338 Set timeout for RCU CPU stall warning messages. 4339 4340 rcupdate.rcu_expedited= [KNL] 4341 Use expedited grace-period primitives, for 4342 example, synchronize_rcu_expedited() instead 4343 of synchronize_rcu(). This reduces latency, 4344 but can increase CPU utilization, degrade 4345 real-time latency, and degrade energy efficiency. 4346 No effect on CONFIG_TINY_RCU kernels. 4347 4348 rcupdate.rcu_normal= [KNL] 4349 Use only normal grace-period primitives, 4350 for example, synchronize_rcu() instead of 4351 synchronize_rcu_expedited(). This improves 4352 real-time latency, CPU utilization, and 4353 energy efficiency, but can expose users to 4354 increased grace-period latency. This parameter 4355 overrides rcupdate.rcu_expedited. No effect on 4356 CONFIG_TINY_RCU kernels. 4357 4358 rcupdate.rcu_normal_after_boot= [KNL] 4359 Once boot has completed (that is, after 4360 rcu_end_inkernel_boot() has been invoked), use 4361 only normal grace-period primitives. No effect 4362 on CONFIG_TINY_RCU kernels. 4363 4364 rcupdate.rcu_task_ipi_delay= [KNL] 4365 Set time in jiffies during which RCU tasks will 4366 avoid sending IPIs, starting with the beginning 4367 of a given grace period. Setting a large 4368 number avoids disturbing real-time workloads, 4369 but lengthens grace periods. 4370 4371 rcupdate.rcu_task_stall_timeout= [KNL] 4372 Set timeout in jiffies for RCU task stall warning 4373 messages. Disable with a value less than or equal 4374 to zero. 4375 4376 rcupdate.rcu_self_test= [KNL] 4377 Run the RCU early boot self tests 4378 4379 rdinit= [KNL] 4380 Format: <full_path> 4381 Run specified binary instead of /init from the ramdisk, 4382 used for early userspace startup. See initrd. 4383 4384 rdrand= [X86] 4385 force - Override the decision by the kernel to hide the 4386 advertisement of RDRAND support (this affects 4387 certain AMD processors because of buggy BIOS 4388 support, specifically around the suspend/resume 4389 path). 4390 4391 rdt= [HW,X86,RDT] 4392 Turn on/off individual RDT features. List is: 4393 cmt, mbmtotal, mbmlocal, l3cat, l3cdp, l2cat, l2cdp, 4394 mba. 4395 E.g. to turn on cmt and turn off mba use: 4396 rdt=cmt,!mba 4397 4398 reboot= [KNL] 4399 Format (x86 or x86_64): 4400 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] \ 4401 [[,]s[mp]#### \ 4402 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \ 4403 [[,]f[orce] 4404 Where reboot_mode is one of warm (soft) or cold (hard) or gpio 4405 (prefix with 'panic_' to set mode for panic 4406 reboot only), 4407 reboot_type is one of bios, acpi, kbd, triple, efi, or pci, 4408 reboot_force is either force or not specified, 4409 reboot_cpu is s[mp]#### with #### being the processor 4410 to be used for rebooting. 4411 4412 relax_domain_level= 4413 [KNL, SMP] Set scheduler's default relax_domain_level. 4414 See Documentation/admin-guide/cgroup-v1/cpusets.rst. 4415 4416 reserve= [KNL,BUGS] Force kernel to ignore I/O ports or memory 4417 Format: <base1>,<size1>[,<base2>,<size2>,...] 4418 Reserve I/O ports or memory so the kernel won't use 4419 them. If <base> is less than 0x10000, the region 4420 is assumed to be I/O ports; otherwise it is memory. 4421 4422 reservetop= [X86-32] 4423 Format: nn[KMG] 4424 Reserves a hole at the top of the kernel virtual 4425 address space. 4426 4427 reservelow= [X86] 4428 Format: nn[K] 4429 Set the amount of memory to reserve for BIOS at 4430 the bottom of the address space. 4431 4432 reset_devices [KNL] Force drivers to reset the underlying device 4433 during initialization. 4434 4435 resume= [SWSUSP] 4436 Specify the partition device for software suspend 4437 Format: 4438 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>} 4439 4440 resume_offset= [SWSUSP] 4441 Specify the offset from the beginning of the partition 4442 given by "resume=" at which the swap header is located, 4443 in <PAGE_SIZE> units (needed only for swap files). 4444 See Documentation/power/swsusp-and-swap-files.rst 4445 4446 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to 4447 read the resume files 4448 4449 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up. 4450 Useful for devices that are detected asynchronously 4451 (e.g. USB and MMC devices). 4452 4453 hibernate= [HIBERNATION] 4454 noresume Don't check if there's a hibernation image 4455 present during boot. 4456 nocompress Don't compress/decompress hibernation images. 4457 no Disable hibernation and resume. 4458 protect_image Turn on image protection during restoration 4459 (that will set all pages holding image data 4460 during restoration read-only). 4461 4462 retain_initrd [RAM] Keep initrd memory after extraction 4463 4464 rfkill.default_state= 4465 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm, 4466 etc. communication is blocked by default. 4467 1 Unblocked. 4468 4469 rfkill.master_switch_mode= 4470 0 The "airplane mode" button does nothing. 4471 1 The "airplane mode" button toggles between everything 4472 blocked and the previous configuration. 4473 2 The "airplane mode" button toggles between everything 4474 blocked and everything unblocked. 4475 4476 rhash_entries= [KNL,NET] 4477 Set number of hash buckets for route cache 4478 4479 ring3mwait=disable 4480 [KNL] Disable ring 3 MONITOR/MWAIT feature on supported 4481 CPUs. 4482 4483 ro [KNL] Mount root device read-only on boot 4484 4485 rodata= [KNL] 4486 on Mark read-only kernel memory as read-only (default). 4487 off Leave read-only kernel memory writable for debugging. 4488 4489 rockchip.usb_uart 4490 Enable the uart passthrough on the designated usb port 4491 on Rockchip SoCs. When active, the signals of the 4492 debug-uart get routed to the D+ and D- pins of the usb 4493 port and the regular usb controller gets disabled. 4494 4495 root= [KNL] Root filesystem 4496 See name_to_dev_t comment in init/do_mounts.c. 4497 4498 rootdelay= [KNL] Delay (in seconds) to pause before attempting to 4499 mount the root filesystem 4500 4501 rootflags= [KNL] Set root filesystem mount option string 4502 4503 rootfstype= [KNL] Set root filesystem type 4504 4505 rootwait [KNL] Wait (indefinitely) for root device to show up. 4506 Useful for devices that are detected asynchronously 4507 (e.g. USB and MMC devices). 4508 4509 rproc_mem=nn[KMG][@address] 4510 [KNL,ARM,CMA] Remoteproc physical memory block. 4511 Memory area to be used by remote processor image, 4512 managed by CMA. 4513 4514 rw [KNL] Mount root device read-write on boot 4515 4516 S [KNL] Run init in single mode 4517 4518 s390_iommu= [HW,S390] 4519 Set s390 IOTLB flushing mode 4520 strict 4521 With strict flushing every unmap operation will result in 4522 an IOTLB flush. Default is lazy flushing before reuse, 4523 which is faster. 4524 4525 sa1100ir [NET] 4526 See drivers/net/irda/sa1100_ir.c. 4527 4528 sbni= [NET] Granch SBNI12 leased line adapter 4529 4530 sched_debug [KNL] Enables verbose scheduler debug messages. 4531 4532 schedstats= [KNL,X86] Enable or disable scheduled statistics. 4533 Allowed values are enable and disable. This feature 4534 incurs a small amount of overhead in the scheduler 4535 but is useful for debugging and performance tuning. 4536 4537 sched_thermal_decay_shift= 4538 [KNL, SMP] Set a decay shift for scheduler thermal 4539 pressure signal. Thermal pressure signal follows the 4540 default decay period of other scheduler pelt 4541 signals(usually 32 ms but configurable). Setting 4542 sched_thermal_decay_shift will left shift the decay 4543 period for the thermal pressure signal by the shift 4544 value. 4545 i.e. with the default pelt decay period of 32 ms 4546 sched_thermal_decay_shift thermal pressure decay pr 4547 1 64 ms 4548 2 128 ms 4549 and so on. 4550 Format: integer between 0 and 10 4551 Default is 0. 4552 4553 skew_tick= [KNL] Offset the periodic timer tick per cpu to mitigate 4554 xtime_lock contention on larger systems, and/or RCU lock 4555 contention on all systems with CONFIG_MAXSMP set. 4556 Format: { "0" | "1" } 4557 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1" 4558 1 -- enable. 4559 Note: increases power consumption, thus should only be 4560 enabled if running jitter sensitive (HPC/RT) workloads. 4561 4562 security= [SECURITY] Choose a legacy "major" security module to 4563 enable at boot. This has been deprecated by the 4564 "lsm=" parameter. 4565 4566 selinux= [SELINUX] Disable or enable SELinux at boot time. 4567 Format: { "0" | "1" } 4568 See security/selinux/Kconfig help text. 4569 0 -- disable. 4570 1 -- enable. 4571 Default value is 1. 4572 4573 apparmor= [APPARMOR] Disable or enable AppArmor at boot time 4574 Format: { "0" | "1" } 4575 See security/apparmor/Kconfig help text 4576 0 -- disable. 4577 1 -- enable. 4578 Default value is set via kernel config option. 4579 4580 serialnumber [BUGS=X86-32] 4581 4582 shapers= [NET] 4583 Maximal number of shapers. 4584 4585 simeth= [IA-64] 4586 simscsi= 4587 4588 slram= [HW,MTD] 4589 4590 slab_nomerge [MM] 4591 Disable merging of slabs with similar size. May be 4592 necessary if there is some reason to distinguish 4593 allocs to different slabs, especially in hardened 4594 environments where the risk of heap overflows and 4595 layout control by attackers can usually be 4596 frustrated by disabling merging. This will reduce 4597 most of the exposure of a heap attack to a single 4598 cache (risks via metadata attacks are mostly 4599 unchanged). Debug options disable merging on their 4600 own. 4601 For more information see Documentation/vm/slub.rst. 4602 4603 slab_max_order= [MM, SLAB] 4604 Determines the maximum allowed order for slabs. 4605 A high setting may cause OOMs due to memory 4606 fragmentation. Defaults to 1 for systems with 4607 more than 32MB of RAM, 0 otherwise. 4608 4609 slub_debug[=options[,slabs]] [MM, SLUB] 4610 Enabling slub_debug allows one to determine the 4611 culprit if slab objects become corrupted. Enabling 4612 slub_debug can create guard zones around objects and 4613 may poison objects when not in use. Also tracks the 4614 last alloc / free. For more information see 4615 Documentation/vm/slub.rst. 4616 4617 slub_memcg_sysfs= [MM, SLUB] 4618 Determines whether to enable sysfs directories for 4619 memory cgroup sub-caches. 1 to enable, 0 to disable. 4620 The default is determined by CONFIG_SLUB_MEMCG_SYSFS_ON. 4621 Enabling this can lead to a very high number of debug 4622 directories and files being created under 4623 /sys/kernel/slub. 4624 4625 slub_max_order= [MM, SLUB] 4626 Determines the maximum allowed order for slabs. 4627 A high setting may cause OOMs due to memory 4628 fragmentation. For more information see 4629 Documentation/vm/slub.rst. 4630 4631 slub_min_objects= [MM, SLUB] 4632 The minimum number of objects per slab. SLUB will 4633 increase the slab order up to slub_max_order to 4634 generate a sufficiently large slab able to contain 4635 the number of objects indicated. The higher the number 4636 of objects the smaller the overhead of tracking slabs 4637 and the less frequently locks need to be acquired. 4638 For more information see Documentation/vm/slub.rst. 4639 4640 slub_min_order= [MM, SLUB] 4641 Determines the minimum page order for slabs. Must be 4642 lower than slub_max_order. 4643 For more information see Documentation/vm/slub.rst. 4644 4645 slub_nomerge [MM, SLUB] 4646 Same with slab_nomerge. This is supported for legacy. 4647 See slab_nomerge for more information. 4648 4649 smart2= [HW] 4650 Format: <io1>[,<io2>[,...,<io8>]] 4651 4652 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices 4653 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port 4654 smsc-ircc2.ircc_sir= [HW] SIR base I/O port 4655 smsc-ircc2.ircc_fir= [HW] FIR base I/O port 4656 smsc-ircc2.ircc_irq= [HW] IRQ line 4657 smsc-ircc2.ircc_dma= [HW] DMA channel 4658 smsc-ircc2.ircc_transceiver= [HW] Transceiver type: 4659 0: Toshiba Satellite 1800 (GP data pin select) 4660 1: Fast pin select (default) 4661 2: ATC IRMode 4662 4663 smt [KNL,S390] Set the maximum number of threads (logical 4664 CPUs) to use per physical CPU on systems capable of 4665 symmetric multithreading (SMT). Will be capped to the 4666 actual hardware limit. 4667 Format: <integer> 4668 Default: -1 (no limit) 4669 4670 softlockup_panic= 4671 [KNL] Should the soft-lockup detector generate panics. 4672 Format: 0 | 1 4673 4674 A value of 1 instructs the soft-lockup detector 4675 to panic the machine when a soft-lockup occurs. It is 4676 also controlled by the kernel.softlockup_panic sysctl 4677 and CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC, which is the 4678 respective build-time switch to that functionality. 4679 4680 softlockup_all_cpu_backtrace= 4681 [KNL] Should the soft-lockup detector generate 4682 backtraces on all cpus. 4683 Format: 0 | 1 4684 4685 sonypi.*= [HW] Sony Programmable I/O Control Device driver 4686 See Documentation/admin-guide/laptops/sonypi.rst 4687 4688 spectre_v2= [X86] Control mitigation of Spectre variant 2 4689 (indirect branch speculation) vulnerability. 4690 The default operation protects the kernel from 4691 user space attacks. 4692 4693 on - unconditionally enable, implies 4694 spectre_v2_user=on 4695 off - unconditionally disable, implies 4696 spectre_v2_user=off 4697 auto - kernel detects whether your CPU model is 4698 vulnerable 4699 4700 Selecting 'on' will, and 'auto' may, choose a 4701 mitigation method at run time according to the 4702 CPU, the available microcode, the setting of the 4703 CONFIG_RETPOLINE configuration option, and the 4704 compiler with which the kernel was built. 4705 4706 Selecting 'on' will also enable the mitigation 4707 against user space to user space task attacks. 4708 4709 Selecting 'off' will disable both the kernel and 4710 the user space protections. 4711 4712 Specific mitigations can also be selected manually: 4713 4714 retpoline - replace indirect branches 4715 retpoline,generic - google's original retpoline 4716 retpoline,amd - AMD-specific minimal thunk 4717 4718 Not specifying this option is equivalent to 4719 spectre_v2=auto. 4720 4721 spectre_v2_user= 4722 [X86] Control mitigation of Spectre variant 2 4723 (indirect branch speculation) vulnerability between 4724 user space tasks 4725 4726 on - Unconditionally enable mitigations. Is 4727 enforced by spectre_v2=on 4728 4729 off - Unconditionally disable mitigations. Is 4730 enforced by spectre_v2=off 4731 4732 prctl - Indirect branch speculation is enabled, 4733 but mitigation can be enabled via prctl 4734 per thread. The mitigation control state 4735 is inherited on fork. 4736 4737 prctl,ibpb 4738 - Like "prctl" above, but only STIBP is 4739 controlled per thread. IBPB is issued 4740 always when switching between different user 4741 space processes. 4742 4743 seccomp 4744 - Same as "prctl" above, but all seccomp 4745 threads will enable the mitigation unless 4746 they explicitly opt out. 4747 4748 seccomp,ibpb 4749 - Like "seccomp" above, but only STIBP is 4750 controlled per thread. IBPB is issued 4751 always when switching between different 4752 user space processes. 4753 4754 auto - Kernel selects the mitigation depending on 4755 the available CPU features and vulnerability. 4756 4757 Default mitigation: 4758 If CONFIG_SECCOMP=y then "seccomp", otherwise "prctl" 4759 4760 Not specifying this option is equivalent to 4761 spectre_v2_user=auto. 4762 4763 spec_store_bypass_disable= 4764 [HW] Control Speculative Store Bypass (SSB) Disable mitigation 4765 (Speculative Store Bypass vulnerability) 4766 4767 Certain CPUs are vulnerable to an exploit against a 4768 a common industry wide performance optimization known 4769 as "Speculative Store Bypass" in which recent stores 4770 to the same memory location may not be observed by 4771 later loads during speculative execution. The idea 4772 is that such stores are unlikely and that they can 4773 be detected prior to instruction retirement at the 4774 end of a particular speculation execution window. 4775 4776 In vulnerable processors, the speculatively forwarded 4777 store can be used in a cache side channel attack, for 4778 example to read memory to which the attacker does not 4779 directly have access (e.g. inside sandboxed code). 4780 4781 This parameter controls whether the Speculative Store 4782 Bypass optimization is used. 4783 4784 On x86 the options are: 4785 4786 on - Unconditionally disable Speculative Store Bypass 4787 off - Unconditionally enable Speculative Store Bypass 4788 auto - Kernel detects whether the CPU model contains an 4789 implementation of Speculative Store Bypass and 4790 picks the most appropriate mitigation. If the 4791 CPU is not vulnerable, "off" is selected. If the 4792 CPU is vulnerable the default mitigation is 4793 architecture and Kconfig dependent. See below. 4794 prctl - Control Speculative Store Bypass per thread 4795 via prctl. Speculative Store Bypass is enabled 4796 for a process by default. The state of the control 4797 is inherited on fork. 4798 seccomp - Same as "prctl" above, but all seccomp threads 4799 will disable SSB unless they explicitly opt out. 4800 4801 Default mitigations: 4802 X86: If CONFIG_SECCOMP=y "seccomp", otherwise "prctl" 4803 4804 On powerpc the options are: 4805 4806 on,auto - On Power8 and Power9 insert a store-forwarding 4807 barrier on kernel entry and exit. On Power7 4808 perform a software flush on kernel entry and 4809 exit. 4810 off - No action. 4811 4812 Not specifying this option is equivalent to 4813 spec_store_bypass_disable=auto. 4814 4815 spia_io_base= [HW,MTD] 4816 spia_fio_base= 4817 spia_pedr= 4818 spia_peddr= 4819 4820 split_lock_detect= 4821 [X86] Enable split lock detection 4822 4823 When enabled (and if hardware support is present), atomic 4824 instructions that access data across cache line 4825 boundaries will result in an alignment check exception. 4826 4827 off - not enabled 4828 4829 warn - the kernel will emit rate limited warnings 4830 about applications triggering the #AC 4831 exception. This mode is the default on CPUs 4832 that supports split lock detection. 4833 4834 fatal - the kernel will send SIGBUS to applications 4835 that trigger the #AC exception. 4836 4837 If an #AC exception is hit in the kernel or in 4838 firmware (i.e. not while executing in user mode) 4839 the kernel will oops in either "warn" or "fatal" 4840 mode. 4841 4842 srbds= [X86,INTEL] 4843 Control the Special Register Buffer Data Sampling 4844 (SRBDS) mitigation. 4845 4846 Certain CPUs are vulnerable to an MDS-like 4847 exploit which can leak bits from the random 4848 number generator. 4849 4850 By default, this issue is mitigated by 4851 microcode. However, the microcode fix can cause 4852 the RDRAND and RDSEED instructions to become 4853 much slower. Among other effects, this will 4854 result in reduced throughput from /dev/urandom. 4855 4856 The microcode mitigation can be disabled with 4857 the following option: 4858 4859 off: Disable mitigation and remove 4860 performance impact to RDRAND and RDSEED 4861 4862 srcutree.counter_wrap_check [KNL] 4863 Specifies how frequently to check for 4864 grace-period sequence counter wrap for the 4865 srcu_data structure's ->srcu_gp_seq_needed field. 4866 The greater the number of bits set in this kernel 4867 parameter, the less frequently counter wrap will 4868 be checked for. Note that the bottom two bits 4869 are ignored. 4870 4871 srcutree.exp_holdoff [KNL] 4872 Specifies how many nanoseconds must elapse 4873 since the end of the last SRCU grace period for 4874 a given srcu_struct until the next normal SRCU 4875 grace period will be considered for automatic 4876 expediting. Set to zero to disable automatic 4877 expediting. 4878 4879 ssbd= [ARM64,HW] 4880 Speculative Store Bypass Disable control 4881 4882 On CPUs that are vulnerable to the Speculative 4883 Store Bypass vulnerability and offer a 4884 firmware based mitigation, this parameter 4885 indicates how the mitigation should be used: 4886 4887 force-on: Unconditionally enable mitigation for 4888 for both kernel and userspace 4889 force-off: Unconditionally disable mitigation for 4890 for both kernel and userspace 4891 kernel: Always enable mitigation in the 4892 kernel, and offer a prctl interface 4893 to allow userspace to register its 4894 interest in being mitigated too. 4895 4896 stack_guard_gap= [MM] 4897 override the default stack gap protection. The value 4898 is in page units and it defines how many pages prior 4899 to (for stacks growing down) resp. after (for stacks 4900 growing up) the main stack are reserved for no other 4901 mapping. Default value is 256 pages. 4902 4903 stacktrace [FTRACE] 4904 Enabled the stack tracer on boot up. 4905 4906 stacktrace_filter=[function-list] 4907 [FTRACE] Limit the functions that the stack tracer 4908 will trace at boot up. function-list is a comma separated 4909 list of functions. This list can be changed at run 4910 time by the stack_trace_filter file in the debugfs 4911 tracing directory. Note, this enables stack tracing 4912 and the stacktrace above is not needed. 4913 4914 sti= [PARISC,HW] 4915 Format: <num> 4916 Set the STI (builtin display/keyboard on the HP-PARISC 4917 machines) console (graphic card) which should be used 4918 as the initial boot-console. 4919 See also comment in drivers/video/console/sticore.c. 4920 4921 sti_font= [HW] 4922 See comment in drivers/video/console/sticore.c. 4923 4924 stifb= [HW] 4925 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]] 4926 4927 sunrpc.min_resvport= 4928 sunrpc.max_resvport= 4929 [NFS,SUNRPC] 4930 SunRPC servers often require that client requests 4931 originate from a privileged port (i.e. a port in the 4932 range 0 < portnr < 1024). 4933 An administrator who wishes to reserve some of these 4934 ports for other uses may adjust the range that the 4935 kernel's sunrpc client considers to be privileged 4936 using these two parameters to set the minimum and 4937 maximum port values. 4938 4939 sunrpc.svc_rpc_per_connection_limit= 4940 [NFS,SUNRPC] 4941 Limit the number of requests that the server will 4942 process in parallel from a single connection. 4943 The default value is 0 (no limit). 4944 4945 sunrpc.pool_mode= 4946 [NFS] 4947 Control how the NFS server code allocates CPUs to 4948 service thread pools. Depending on how many NICs 4949 you have and where their interrupts are bound, this 4950 option will affect which CPUs will do NFS serving. 4951 Note: this parameter cannot be changed while the 4952 NFS server is running. 4953 4954 auto the server chooses an appropriate mode 4955 automatically using heuristics 4956 global a single global pool contains all CPUs 4957 percpu one pool for each CPU 4958 pernode one pool for each NUMA node (equivalent 4959 to global on non-NUMA machines) 4960 4961 sunrpc.tcp_slot_table_entries= 4962 sunrpc.udp_slot_table_entries= 4963 [NFS,SUNRPC] 4964 Sets the upper limit on the number of simultaneous 4965 RPC calls that can be sent from the client to a 4966 server. Increasing these values may allow you to 4967 improve throughput, but will also increase the 4968 amount of memory reserved for use by the client. 4969 4970 suspend.pm_test_delay= 4971 [SUSPEND] 4972 Sets the number of seconds to remain in a suspend test 4973 mode before resuming the system (see 4974 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG 4975 is set. Default value is 5. 4976 4977 svm= [PPC] 4978 Format: { on | off | y | n | 1 | 0 } 4979 This parameter controls use of the Protected 4980 Execution Facility on pSeries. 4981 4982 swapaccount=[0|1] 4983 [KNL] Enable accounting of swap in memory resource 4984 controller if no parameter or 1 is given or disable 4985 it if 0 is given (See Documentation/admin-guide/cgroup-v1/memory.rst) 4986 4987 swiotlb= [ARM,IA-64,PPC,MIPS,X86] 4988 Format: { <int> | force | noforce } 4989 <int> -- Number of I/O TLB slabs 4990 force -- force using of bounce buffers even if they 4991 wouldn't be automatically used by the kernel 4992 noforce -- Never use bounce buffers (for debugging) 4993 4994 switches= [HW,M68k] 4995 4996 sysctl.*= [KNL] 4997 Set a sysctl parameter, right before loading the init 4998 process, as if the value was written to the respective 4999 /proc/sys/... file. Both '.' and '/' are recognized as 5000 separators. Unrecognized parameters and invalid values 5001 are reported in the kernel log. Sysctls registered 5002 later by a loaded module cannot be set this way. 5003 Example: sysctl.vm.swappiness=40 5004 5005 sysfs.deprecated=0|1 [KNL] 5006 Enable/disable old style sysfs layout for old udev 5007 on older distributions. When this option is enabled 5008 very new udev will not work anymore. When this option 5009 is disabled (or CONFIG_SYSFS_DEPRECATED not compiled) 5010 in older udev will not work anymore. 5011 Default depends on CONFIG_SYSFS_DEPRECATED_V2 set in 5012 the kernel configuration. 5013 5014 sysrq_always_enabled 5015 [KNL] 5016 Ignore sysrq setting - this boot parameter will 5017 neutralize any effect of /proc/sys/kernel/sysrq. 5018 Useful for debugging. 5019 5020 tcpmhash_entries= [KNL,NET] 5021 Set the number of tcp_metrics_hash slots. 5022 Default value is 8192 or 16384 depending on total 5023 ram pages. This is used to specify the TCP metrics 5024 cache size. See Documentation/networking/ip-sysctl.rst 5025 "tcp_no_metrics_save" section for more details. 5026 5027 tdfx= [HW,DRM] 5028 5029 test_suspend= [SUSPEND][,N] 5030 Specify "mem" (for Suspend-to-RAM) or "standby" (for 5031 standby suspend) or "freeze" (for suspend type freeze) 5032 as the system sleep state during system startup with 5033 the optional capability to repeat N number of times. 5034 The system is woken from this state using a 5035 wakeup-capable RTC alarm. 5036 5037 thash_entries= [KNL,NET] 5038 Set number of hash buckets for TCP connection 5039 5040 thermal.act= [HW,ACPI] 5041 -1: disable all active trip points in all thermal zones 5042 <degrees C>: override all lowest active trip points 5043 5044 thermal.crt= [HW,ACPI] 5045 -1: disable all critical trip points in all thermal zones 5046 <degrees C>: override all critical trip points 5047 5048 thermal.nocrt= [HW,ACPI] 5049 Set to disable actions on ACPI thermal zone 5050 critical and hot trip points. 5051 5052 thermal.off= [HW,ACPI] 5053 1: disable ACPI thermal control 5054 5055 thermal.psv= [HW,ACPI] 5056 -1: disable all passive trip points 5057 <degrees C>: override all passive trip points to this 5058 value 5059 5060 thermal.tzp= [HW,ACPI] 5061 Specify global default ACPI thermal zone polling rate 5062 <deci-seconds>: poll all this frequency 5063 0: no polling (default) 5064 5065 threadirqs [KNL] 5066 Force threading of all interrupt handlers except those 5067 marked explicitly IRQF_NO_THREAD. 5068 5069 topology= [S390] 5070 Format: {off | on} 5071 Specify if the kernel should make use of the cpu 5072 topology information if the hardware supports this. 5073 The scheduler will make use of this information and 5074 e.g. base its process migration decisions on it. 5075 Default is on. 5076 5077 topology_updates= [KNL, PPC, NUMA] 5078 Format: {off} 5079 Specify if the kernel should ignore (off) 5080 topology updates sent by the hypervisor to this 5081 LPAR. 5082 5083 torture.disable_onoff_at_boot= [KNL] 5084 Prevent the CPU-hotplug component of torturing 5085 until after init has spawned. 5086 5087 tp720= [HW,PS2] 5088 5089 tpm_suspend_pcr=[HW,TPM] 5090 Format: integer pcr id 5091 Specify that at suspend time, the tpm driver 5092 should extend the specified pcr with zeros, 5093 as a workaround for some chips which fail to 5094 flush the last written pcr on TPM_SaveState. 5095 This will guarantee that all the other pcrs 5096 are saved. 5097 5098 trace_buf_size=nn[KMG] 5099 [FTRACE] will set tracing buffer size on each cpu. 5100 5101 trace_event=[event-list] 5102 [FTRACE] Set and start specified trace events in order 5103 to facilitate early boot debugging. The event-list is a 5104 comma separated list of trace events to enable. See 5105 also Documentation/trace/events.rst 5106 5107 trace_options=[option-list] 5108 [FTRACE] Enable or disable tracer options at boot. 5109 The option-list is a comma delimited list of options 5110 that can be enabled or disabled just as if you were 5111 to echo the option name into 5112 5113 /sys/kernel/debug/tracing/trace_options 5114 5115 For example, to enable stacktrace option (to dump the 5116 stack trace of each event), add to the command line: 5117 5118 trace_options=stacktrace 5119 5120 See also Documentation/trace/ftrace.rst "trace options" 5121 section. 5122 5123 tp_printk[FTRACE] 5124 Have the tracepoints sent to printk as well as the 5125 tracing ring buffer. This is useful for early boot up 5126 where the system hangs or reboots and does not give the 5127 option for reading the tracing buffer or performing a 5128 ftrace_dump_on_oops. 5129 5130 To turn off having tracepoints sent to printk, 5131 echo 0 > /proc/sys/kernel/tracepoint_printk 5132 Note, echoing 1 into this file without the 5133 tracepoint_printk kernel cmdline option has no effect. 5134 5135 ** CAUTION ** 5136 5137 Having tracepoints sent to printk() and activating high 5138 frequency tracepoints such as irq or sched, can cause 5139 the system to live lock. 5140 5141 traceoff_on_warning 5142 [FTRACE] enable this option to disable tracing when a 5143 warning is hit. This turns off "tracing_on". Tracing can 5144 be enabled again by echoing '1' into the "tracing_on" 5145 file located in /sys/kernel/debug/tracing/ 5146 5147 This option is useful, as it disables the trace before 5148 the WARNING dump is called, which prevents the trace to 5149 be filled with content caused by the warning output. 5150 5151 This option can also be set at run time via the sysctl 5152 option: kernel/traceoff_on_warning 5153 5154 transparent_hugepage= 5155 [KNL] 5156 Format: [always|madvise|never] 5157 Can be used to control the default behavior of the system 5158 with respect to transparent hugepages. 5159 See Documentation/admin-guide/mm/transhuge.rst 5160 for more details. 5161 5162 tsc= Disable clocksource stability checks for TSC. 5163 Format: <string> 5164 [x86] reliable: mark tsc clocksource as reliable, this 5165 disables clocksource verification at runtime, as well 5166 as the stability checks done at bootup. Used to enable 5167 high-resolution timer mode on older hardware, and in 5168 virtualized environment. 5169 [x86] noirqtime: Do not use TSC to do irq accounting. 5170 Used to run time disable IRQ_TIME_ACCOUNTING on any 5171 platforms where RDTSC is slow and this accounting 5172 can add overhead. 5173 [x86] unstable: mark the TSC clocksource as unstable, this 5174 marks the TSC unconditionally unstable at bootup and 5175 avoids any further wobbles once the TSC watchdog notices. 5176 [x86] nowatchdog: disable clocksource watchdog. Used 5177 in situations with strict latency requirements (where 5178 interruptions from clocksource watchdog are not 5179 acceptable). 5180 5181 tsc_early_khz= [X86] Skip early TSC calibration and use the given 5182 value instead. Useful when the early TSC frequency discovery 5183 procedure is not reliable, such as on overclocked systems 5184 with CPUID.16h support and partial CPUID.15h support. 5185 Format: <unsigned int> 5186 5187 tsx= [X86] Control Transactional Synchronization 5188 Extensions (TSX) feature in Intel processors that 5189 support TSX control. 5190 5191 This parameter controls the TSX feature. The options are: 5192 5193 on - Enable TSX on the system. Although there are 5194 mitigations for all known security vulnerabilities, 5195 TSX has been known to be an accelerator for 5196 several previous speculation-related CVEs, and 5197 so there may be unknown security risks associated 5198 with leaving it enabled. 5199 5200 off - Disable TSX on the system. (Note that this 5201 option takes effect only on newer CPUs which are 5202 not vulnerable to MDS, i.e., have 5203 MSR_IA32_ARCH_CAPABILITIES.MDS_NO=1 and which get 5204 the new IA32_TSX_CTRL MSR through a microcode 5205 update. This new MSR allows for the reliable 5206 deactivation of the TSX functionality.) 5207 5208 auto - Disable TSX if X86_BUG_TAA is present, 5209 otherwise enable TSX on the system. 5210 5211 Not specifying this option is equivalent to tsx=off. 5212 5213 See Documentation/admin-guide/hw-vuln/tsx_async_abort.rst 5214 for more details. 5215 5216 tsx_async_abort= [X86,INTEL] Control mitigation for the TSX Async 5217 Abort (TAA) vulnerability. 5218 5219 Similar to Micro-architectural Data Sampling (MDS) 5220 certain CPUs that support Transactional 5221 Synchronization Extensions (TSX) are vulnerable to an 5222 exploit against CPU internal buffers which can forward 5223 information to a disclosure gadget under certain 5224 conditions. 5225 5226 In vulnerable processors, the speculatively forwarded 5227 data can be used in a cache side channel attack, to 5228 access data to which the attacker does not have direct 5229 access. 5230 5231 This parameter controls the TAA mitigation. The 5232 options are: 5233 5234 full - Enable TAA mitigation on vulnerable CPUs 5235 if TSX is enabled. 5236 5237 full,nosmt - Enable TAA mitigation and disable SMT on 5238 vulnerable CPUs. If TSX is disabled, SMT 5239 is not disabled because CPU is not 5240 vulnerable to cross-thread TAA attacks. 5241 off - Unconditionally disable TAA mitigation 5242 5243 On MDS-affected machines, tsx_async_abort=off can be 5244 prevented by an active MDS mitigation as both vulnerabilities 5245 are mitigated with the same mechanism so in order to disable 5246 this mitigation, you need to specify mds=off too. 5247 5248 Not specifying this option is equivalent to 5249 tsx_async_abort=full. On CPUs which are MDS affected 5250 and deploy MDS mitigation, TAA mitigation is not 5251 required and doesn't provide any additional 5252 mitigation. 5253 5254 For details see: 5255 Documentation/admin-guide/hw-vuln/tsx_async_abort.rst 5256 5257 turbografx.map[2|3]= [HW,JOY] 5258 TurboGraFX parallel port interface 5259 Format: 5260 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7> 5261 See also Documentation/input/devices/joystick-parport.rst 5262 5263 udbg-immortal [PPC] When debugging early kernel crashes that 5264 happen after console_init() and before a proper 5265 console driver takes over, this boot options might 5266 help "seeing" what's going on. 5267 5268 uhash_entries= [KNL,NET] 5269 Set number of hash buckets for UDP/UDP-Lite connections 5270 5271 uhci-hcd.ignore_oc= 5272 [USB] Ignore overcurrent events (default N). 5273 Some badly-designed motherboards generate lots of 5274 bogus events, for ports that aren't wired to 5275 anything. Set this parameter to avoid log spamming. 5276 Note that genuine overcurrent events won't be 5277 reported either. 5278 5279 unknown_nmi_panic 5280 [X86] Cause panic on unknown NMI. 5281 5282 usbcore.authorized_default= 5283 [USB] Default USB device authorization: 5284 (default -1 = authorized except for wireless USB, 5285 0 = not authorized, 1 = authorized, 2 = authorized 5286 if device connected to internal port) 5287 5288 usbcore.autosuspend= 5289 [USB] The autosuspend time delay (in seconds) used 5290 for newly-detected USB devices (default 2). This 5291 is the time required before an idle device will be 5292 autosuspended. Devices for which the delay is set 5293 to a negative value won't be autosuspended at all. 5294 5295 usbcore.usbfs_snoop= 5296 [USB] Set to log all usbfs traffic (default 0 = off). 5297 5298 usbcore.usbfs_snoop_max= 5299 [USB] Maximum number of bytes to snoop in each URB 5300 (default = 65536). 5301 5302 usbcore.blinkenlights= 5303 [USB] Set to cycle leds on hubs (default 0 = off). 5304 5305 usbcore.old_scheme_first= 5306 [USB] Start with the old device initialization 5307 scheme (default 0 = off). 5308 5309 usbcore.usbfs_memory_mb= 5310 [USB] Memory limit (in MB) for buffers allocated by 5311 usbfs (default = 16, 0 = max = 2047). 5312 5313 usbcore.use_both_schemes= 5314 [USB] Try the other device initialization scheme 5315 if the first one fails (default 1 = enabled). 5316 5317 usbcore.initial_descriptor_timeout= 5318 [USB] Specifies timeout for the initial 64-byte 5319 USB_REQ_GET_DESCRIPTOR request in milliseconds 5320 (default 5000 = 5.0 seconds). 5321 5322 usbcore.nousb [USB] Disable the USB subsystem 5323 5324 usbcore.quirks= 5325 [USB] A list of quirk entries to augment the built-in 5326 usb core quirk list. List entries are separated by 5327 commas. Each entry has the form 5328 VendorID:ProductID:Flags. The IDs are 4-digit hex 5329 numbers and Flags is a set of letters. Each letter 5330 will change the built-in quirk; setting it if it is 5331 clear and clearing it if it is set. The letters have 5332 the following meanings: 5333 a = USB_QUIRK_STRING_FETCH_255 (string 5334 descriptors must not be fetched using 5335 a 255-byte read); 5336 b = USB_QUIRK_RESET_RESUME (device can't resume 5337 correctly so reset it instead); 5338 c = USB_QUIRK_NO_SET_INTF (device can't handle 5339 Set-Interface requests); 5340 d = USB_QUIRK_CONFIG_INTF_STRINGS (device can't 5341 handle its Configuration or Interface 5342 strings); 5343 e = USB_QUIRK_RESET (device can't be reset 5344 (e.g morph devices), don't use reset); 5345 f = USB_QUIRK_HONOR_BNUMINTERFACES (device has 5346 more interface descriptions than the 5347 bNumInterfaces count, and can't handle 5348 talking to these interfaces); 5349 g = USB_QUIRK_DELAY_INIT (device needs a pause 5350 during initialization, after we read 5351 the device descriptor); 5352 h = USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL (For 5353 high speed and super speed interrupt 5354 endpoints, the USB 2.0 and USB 3.0 spec 5355 require the interval in microframes (1 5356 microframe = 125 microseconds) to be 5357 calculated as interval = 2 ^ 5358 (bInterval-1). 5359 Devices with this quirk report their 5360 bInterval as the result of this 5361 calculation instead of the exponent 5362 variable used in the calculation); 5363 i = USB_QUIRK_DEVICE_QUALIFIER (device can't 5364 handle device_qualifier descriptor 5365 requests); 5366 j = USB_QUIRK_IGNORE_REMOTE_WAKEUP (device 5367 generates spurious wakeup, ignore 5368 remote wakeup capability); 5369 k = USB_QUIRK_NO_LPM (device can't handle Link 5370 Power Management); 5371 l = USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL 5372 (Device reports its bInterval as linear 5373 frames instead of the USB 2.0 5374 calculation); 5375 m = USB_QUIRK_DISCONNECT_SUSPEND (Device needs 5376 to be disconnected before suspend to 5377 prevent spurious wakeup); 5378 n = USB_QUIRK_DELAY_CTRL_MSG (Device needs a 5379 pause after every control message); 5380 o = USB_QUIRK_HUB_SLOW_RESET (Hub needs extra 5381 delay after resetting its port); 5382 Example: quirks=0781:5580:bk,0a5c:5834:gij 5383 5384 usbhid.mousepoll= 5385 [USBHID] The interval which mice are to be polled at. 5386 5387 usbhid.jspoll= 5388 [USBHID] The interval which joysticks are to be polled at. 5389 5390 usbhid.kbpoll= 5391 [USBHID] The interval which keyboards are to be polled at. 5392 5393 usb-storage.delay_use= 5394 [UMS] The delay in seconds before a new device is 5395 scanned for Logical Units (default 1). 5396 5397 usb-storage.quirks= 5398 [UMS] A list of quirks entries to supplement or 5399 override the built-in unusual_devs list. List 5400 entries are separated by commas. Each entry has 5401 the form VID:PID:Flags where VID and PID are Vendor 5402 and Product ID values (4-digit hex numbers) and 5403 Flags is a set of characters, each corresponding 5404 to a common usb-storage quirk flag as follows: 5405 a = SANE_SENSE (collect more than 18 bytes 5406 of sense data, not on uas); 5407 b = BAD_SENSE (don't collect more than 18 5408 bytes of sense data, not on uas); 5409 c = FIX_CAPACITY (decrease the reported 5410 device capacity by one sector); 5411 d = NO_READ_DISC_INFO (don't use 5412 READ_DISC_INFO command, not on uas); 5413 e = NO_READ_CAPACITY_16 (don't use 5414 READ_CAPACITY_16 command); 5415 f = NO_REPORT_OPCODES (don't use report opcodes 5416 command, uas only); 5417 g = MAX_SECTORS_240 (don't transfer more than 5418 240 sectors at a time, uas only); 5419 h = CAPACITY_HEURISTICS (decrease the 5420 reported device capacity by one 5421 sector if the number is odd); 5422 i = IGNORE_DEVICE (don't bind to this 5423 device); 5424 j = NO_REPORT_LUNS (don't use report luns 5425 command, uas only); 5426 l = NOT_LOCKABLE (don't try to lock and 5427 unlock ejectable media, not on uas); 5428 m = MAX_SECTORS_64 (don't transfer more 5429 than 64 sectors = 32 KB at a time, 5430 not on uas); 5431 n = INITIAL_READ10 (force a retry of the 5432 initial READ(10) command, not on uas); 5433 o = CAPACITY_OK (accept the capacity 5434 reported by the device, not on uas); 5435 p = WRITE_CACHE (the device cache is ON 5436 by default, not on uas); 5437 r = IGNORE_RESIDUE (the device reports 5438 bogus residue values, not on uas); 5439 s = SINGLE_LUN (the device has only one 5440 Logical Unit); 5441 t = NO_ATA_1X (don't allow ATA(12) and ATA(16) 5442 commands, uas only); 5443 u = IGNORE_UAS (don't bind to the uas driver); 5444 w = NO_WP_DETECT (don't test whether the 5445 medium is write-protected). 5446 y = ALWAYS_SYNC (issue a SYNCHRONIZE_CACHE 5447 even if the device claims no cache, 5448 not on uas) 5449 Example: quirks=0419:aaf5:rl,0421:0433:rc 5450 5451 user_debug= [KNL,ARM] 5452 Format: <int> 5453 See arch/arm/Kconfig.debug help text. 5454 1 - undefined instruction events 5455 2 - system calls 5456 4 - invalid data aborts 5457 8 - SIGSEGV faults 5458 16 - SIGBUS faults 5459 Example: user_debug=31 5460 5461 userpte= 5462 [X86] Flags controlling user PTE allocations. 5463 5464 nohigh = do not allocate PTE pages in 5465 HIGHMEM regardless of setting 5466 of CONFIG_HIGHPTE. 5467 5468 vdso= [X86,SH] 5469 On X86_32, this is an alias for vdso32=. Otherwise: 5470 5471 vdso=1: enable VDSO (the default) 5472 vdso=0: disable VDSO mapping 5473 5474 vdso32= [X86] Control the 32-bit vDSO 5475 vdso32=1: enable 32-bit VDSO 5476 vdso32=0 or vdso32=2: disable 32-bit VDSO 5477 5478 See the help text for CONFIG_COMPAT_VDSO for more 5479 details. If CONFIG_COMPAT_VDSO is set, the default is 5480 vdso32=0; otherwise, the default is vdso32=1. 5481 5482 For compatibility with older kernels, vdso32=2 is an 5483 alias for vdso32=0. 5484 5485 Try vdso32=0 if you encounter an error that says: 5486 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed! 5487 5488 vector= [IA-64,SMP] 5489 vector=percpu: enable percpu vector domain 5490 5491 video= [FB] Frame buffer configuration 5492 See Documentation/fb/modedb.rst. 5493 5494 video.brightness_switch_enabled= [0,1] 5495 If set to 1, on receiving an ACPI notify event 5496 generated by hotkey, video driver will adjust brightness 5497 level and then send out the event to user space through 5498 the allocated input device; If set to 0, video driver 5499 will only send out the event without touching backlight 5500 brightness level. 5501 default: 1 5502 5503 virtio_mmio.device= 5504 [VMMIO] Memory mapped virtio (platform) device. 5505 5506 <size>@<baseaddr>:<irq>[:<id>] 5507 where: 5508 <size> := size (can use standard suffixes 5509 like K, M and G) 5510 <baseaddr> := physical base address 5511 <irq> := interrupt number (as passed to 5512 request_irq()) 5513 <id> := (optional) platform device id 5514 example: 5515 virtio_mmio.device=1K@0x100b0000:48:7 5516 5517 Can be used multiple times for multiple devices. 5518 5519 vga= [BOOT,X86-32] Select a particular video mode 5520 See Documentation/x86/boot.rst and 5521 Documentation/admin-guide/svga.rst. 5522 Use vga=ask for menu. 5523 This is actually a boot loader parameter; the value is 5524 passed to the kernel using a special protocol. 5525 5526 vm_debug[=options] [KNL] Available with CONFIG_DEBUG_VM=y. 5527 May slow down system boot speed, especially when 5528 enabled on systems with a large amount of memory. 5529 All options are enabled by default, and this 5530 interface is meant to allow for selectively 5531 enabling or disabling specific virtual memory 5532 debugging features. 5533 5534 Available options are: 5535 P Enable page structure init time poisoning 5536 - Disable all of the above options 5537 5538 vmalloc=nn[KMG] [KNL,BOOT] Forces the vmalloc area to have an exact 5539 size of <nn>. This can be used to increase the 5540 minimum size (128MB on x86). It can also be used to 5541 decrease the size and leave more room for directly 5542 mapped kernel RAM. 5543 5544 vmcp_cma=nn[MG] [KNL,S390] 5545 Sets the memory size reserved for contiguous memory 5546 allocations for the vmcp device driver. 5547 5548 vmhalt= [KNL,S390] Perform z/VM CP command after system halt. 5549 Format: <command> 5550 5551 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic. 5552 Format: <command> 5553 5554 vmpoff= [KNL,S390] Perform z/VM CP command after power off. 5555 Format: <command> 5556 5557 vsyscall= [X86-64] 5558 Controls the behavior of vsyscalls (i.e. calls to 5559 fixed addresses of 0xffffffffff600x00 from legacy 5560 code). Most statically-linked binaries and older 5561 versions of glibc use these calls. Because these 5562 functions are at fixed addresses, they make nice 5563 targets for exploits that can control RIP. 5564 5565 emulate [default] Vsyscalls turn into traps and are 5566 emulated reasonably safely. The vsyscall 5567 page is readable. 5568 5569 xonly Vsyscalls turn into traps and are 5570 emulated reasonably safely. The vsyscall 5571 page is not readable. 5572 5573 none Vsyscalls don't work at all. This makes 5574 them quite hard to use for exploits but 5575 might break your system. 5576 5577 vt.color= [VT] Default text color. 5578 Format: 0xYX, X = foreground, Y = background. 5579 Default: 0x07 = light gray on black. 5580 5581 vt.cur_default= [VT] Default cursor shape. 5582 Format: 0xCCBBAA, where AA, BB, and CC are the same as 5583 the parameters of the <Esc>[?A;B;Cc escape sequence; 5584 see VGA-softcursor.txt. Default: 2 = underline. 5585 5586 vt.default_blu= [VT] 5587 Format: <blue0>,<blue1>,<blue2>,...,<blue15> 5588 Change the default blue palette of the console. 5589 This is a 16-member array composed of values 5590 ranging from 0-255. 5591 5592 vt.default_grn= [VT] 5593 Format: <green0>,<green1>,<green2>,...,<green15> 5594 Change the default green palette of the console. 5595 This is a 16-member array composed of values 5596 ranging from 0-255. 5597 5598 vt.default_red= [VT] 5599 Format: <red0>,<red1>,<red2>,...,<red15> 5600 Change the default red palette of the console. 5601 This is a 16-member array composed of values 5602 ranging from 0-255. 5603 5604 vt.default_utf8= 5605 [VT] 5606 Format=<0|1> 5607 Set system-wide default UTF-8 mode for all tty's. 5608 Default is 1, i.e. UTF-8 mode is enabled for all 5609 newly opened terminals. 5610 5611 vt.global_cursor_default= 5612 [VT] 5613 Format=<-1|0|1> 5614 Set system-wide default for whether a cursor 5615 is shown on new VTs. Default is -1, 5616 i.e. cursors will be created by default unless 5617 overridden by individual drivers. 0 will hide 5618 cursors, 1 will display them. 5619 5620 vt.italic= [VT] Default color for italic text; 0-15. 5621 Default: 2 = green. 5622 5623 vt.underline= [VT] Default color for underlined text; 0-15. 5624 Default: 3 = cyan. 5625 5626 watchdog timers [HW,WDT] For information on watchdog timers, 5627 see Documentation/watchdog/watchdog-parameters.rst 5628 or other driver-specific files in the 5629 Documentation/watchdog/ directory. 5630 5631 watchdog_thresh= 5632 [KNL] 5633 Set the hard lockup detector stall duration 5634 threshold in seconds. The soft lockup detector 5635 threshold is set to twice the value. A value of 0 5636 disables both lockup detectors. Default is 10 5637 seconds. 5638 5639 workqueue.watchdog_thresh= 5640 If CONFIG_WQ_WATCHDOG is configured, workqueue can 5641 warn stall conditions and dump internal state to 5642 help debugging. 0 disables workqueue stall 5643 detection; otherwise, it's the stall threshold 5644 duration in seconds. The default value is 30 and 5645 it can be updated at runtime by writing to the 5646 corresponding sysfs file. 5647 5648 workqueue.disable_numa 5649 By default, all work items queued to unbound 5650 workqueues are affine to the NUMA nodes they're 5651 issued on, which results in better behavior in 5652 general. If NUMA affinity needs to be disabled for 5653 whatever reason, this option can be used. Note 5654 that this also can be controlled per-workqueue for 5655 workqueues visible under /sys/bus/workqueue/. 5656 5657 workqueue.power_efficient 5658 Per-cpu workqueues are generally preferred because 5659 they show better performance thanks to cache 5660 locality; unfortunately, per-cpu workqueues tend to 5661 be more power hungry than unbound workqueues. 5662 5663 Enabling this makes the per-cpu workqueues which 5664 were observed to contribute significantly to power 5665 consumption unbound, leading to measurably lower 5666 power usage at the cost of small performance 5667 overhead. 5668 5669 The default value of this parameter is determined by 5670 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT. 5671 5672 workqueue.debug_force_rr_cpu 5673 Workqueue used to implicitly guarantee that work 5674 items queued without explicit CPU specified are put 5675 on the local CPU. This guarantee is no longer true 5676 and while local CPU is still preferred work items 5677 may be put on foreign CPUs. This debug option 5678 forces round-robin CPU selection to flush out 5679 usages which depend on the now broken guarantee. 5680 When enabled, memory and cache locality will be 5681 impacted. 5682 5683 x2apic_phys [X86-64,APIC] Use x2apic physical mode instead of 5684 default x2apic cluster mode on platforms 5685 supporting x2apic. 5686 5687 x86_intel_mid_timer= [X86-32,APBT] 5688 Choose timer option for x86 Intel MID platform. 5689 Two valid options are apbt timer only and lapic timer 5690 plus one apbt timer for broadcast timer. 5691 x86_intel_mid_timer=apbt_only | lapic_and_apbt 5692 5693 xen_512gb_limit [KNL,X86-64,XEN] 5694 Restricts the kernel running paravirtualized under Xen 5695 to use only up to 512 GB of RAM. The reason to do so is 5696 crash analysis tools and Xen tools for doing domain 5697 save/restore/migration must be enabled to handle larger 5698 domains. 5699 5700 xen_emul_unplug= [HW,X86,XEN] 5701 Unplug Xen emulated devices 5702 Format: [unplug0,][unplug1] 5703 ide-disks -- unplug primary master IDE devices 5704 aux-ide-disks -- unplug non-primary-master IDE devices 5705 nics -- unplug network devices 5706 all -- unplug all emulated devices (NICs and IDE disks) 5707 unnecessary -- unplugging emulated devices is 5708 unnecessary even if the host did not respond to 5709 the unplug protocol 5710 never -- do not unplug even if version check succeeds 5711 5712 xen_legacy_crash [X86,XEN] 5713 Crash from Xen panic notifier, without executing late 5714 panic() code such as dumping handler. 5715 5716 xen_nopvspin [X86,XEN] 5717 Disables the ticketlock slowpath using Xen PV 5718 optimizations. 5719 5720 xen_nopv [X86] 5721 Disables the PV optimizations forcing the HVM guest to 5722 run as generic HVM guest with no PV drivers. 5723 This option is obsoleted by the "nopv" option, which 5724 has equivalent effect for XEN platform. 5725 5726 xen_scrub_pages= [XEN] 5727 Boolean option to control scrubbing pages before giving them back 5728 to Xen, for use by other domains. Can be also changed at runtime 5729 with /sys/devices/system/xen_memory/xen_memory0/scrub_pages. 5730 Default value controlled with CONFIG_XEN_SCRUB_PAGES_DEFAULT. 5731 5732 xen_timer_slop= [X86-64,XEN] 5733 Set the timer slop (in nanoseconds) for the virtual Xen 5734 timers (default is 100000). This adjusts the minimum 5735 delta of virtualized Xen timers, where lower values 5736 improve timer resolution at the expense of processing 5737 more timer interrupts. 5738 5739 nopv= [X86,XEN,KVM,HYPER_V,VMWARE] 5740 Disables the PV optimizations forcing the guest to run 5741 as generic guest with no PV drivers. Currently support 5742 XEN HVM, KVM, HYPER_V and VMWARE guest. 5743 5744 xirc2ps_cs= [NET,PCMCIA] 5745 Format: 5746 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]] 5747 5748 xive= [PPC] 5749 By default on POWER9 and above, the kernel will 5750 natively use the XIVE interrupt controller. This option 5751 allows the fallback firmware mode to be used: 5752 5753 off Fallback to firmware control of XIVE interrupt 5754 controller on both pseries and powernv 5755 platforms. Only useful on POWER9 and above. 5756 5757 xhci-hcd.quirks [USB,KNL] 5758 A hex value specifying bitmask with supplemental xhci 5759 host controller quirks. Meaning of each bit can be 5760 consulted in header drivers/usb/host/xhci.h. 5761 5762 xmon [PPC] 5763 Format: { early | on | rw | ro | off } 5764 Controls if xmon debugger is enabled. Default is off. 5765 Passing only "xmon" is equivalent to "xmon=early". 5766 early Call xmon as early as possible on boot; xmon 5767 debugger is called from setup_arch(). 5768 on xmon debugger hooks will be installed so xmon 5769 is only called on a kernel crash. Default mode, 5770 i.e. either "ro" or "rw" mode, is controlled 5771 with CONFIG_XMON_DEFAULT_RO_MODE. 5772 rw xmon debugger hooks will be installed so xmon 5773 is called only on a kernel crash, mode is write, 5774 meaning SPR registers, memory and, other data 5775 can be written using xmon commands. 5776 ro same as "rw" option above but SPR registers, 5777 memory, and other data can't be written using 5778 xmon commands. 5779 off xmon is disabled. 5780