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.txt, 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 acpi_backlight=vendor 26 acpi_backlight=video 27 If set to vendor, prefer vendor specific driver 28 (e.g. thinkpad_acpi, sony_acpi, etc.) instead 29 of the ACPI video.ko driver. 30 31 acpi_force_32bit_fadt_addr 32 force FADT to use 32 bit addresses rather than the 33 64 bit X_* addresses. Some firmware have broken 64 34 bit addresses for force ACPI ignore these and use 35 the older legacy 32 bit addresses. 36 37 acpica_no_return_repair [HW, ACPI] 38 Disable AML predefined validation mechanism 39 This mechanism can repair the evaluation result to make 40 the return objects more ACPI specification compliant. 41 This option is useful for developers to identify the 42 root cause of an AML interpreter issue when the issue 43 has something to do with the repair mechanism. 44 45 acpi.debug_layer= [HW,ACPI,ACPI_DEBUG] 46 acpi.debug_level= [HW,ACPI,ACPI_DEBUG] 47 Format: <int> 48 CONFIG_ACPI_DEBUG must be enabled to produce any ACPI 49 debug output. Bits in debug_layer correspond to a 50 _COMPONENT in an ACPI source file, e.g., 51 #define _COMPONENT ACPI_PCI_COMPONENT 52 Bits in debug_level correspond to a level in 53 ACPI_DEBUG_PRINT statements, e.g., 54 ACPI_DEBUG_PRINT((ACPI_DB_INFO, ... 55 The debug_level mask defaults to "info". See 56 Documentation/acpi/debug.txt for more information about 57 debug layers and levels. 58 59 Enable processor driver info messages: 60 acpi.debug_layer=0x20000000 61 Enable PCI/PCI interrupt routing info messages: 62 acpi.debug_layer=0x400000 63 Enable AML "Debug" output, i.e., stores to the Debug 64 object while interpreting AML: 65 acpi.debug_layer=0xffffffff acpi.debug_level=0x2 66 Enable all messages related to ACPI hardware: 67 acpi.debug_layer=0x2 acpi.debug_level=0xffffffff 68 69 Some values produce so much output that the system is 70 unusable. The "log_buf_len" parameter may be useful 71 if you need to capture more output. 72 73 acpi_enforce_resources= [ACPI] 74 { strict | lax | no } 75 Check for resource conflicts between native drivers 76 and ACPI OperationRegions (SystemIO and SystemMemory 77 only). IO ports and memory declared in ACPI might be 78 used by the ACPI subsystem in arbitrary AML code and 79 can interfere with legacy drivers. 80 strict (default): access to resources claimed by ACPI 81 is denied; legacy drivers trying to access reserved 82 resources will fail to bind to device using them. 83 lax: access to resources claimed by ACPI is allowed; 84 legacy drivers trying to access reserved resources 85 will bind successfully but a warning message is logged. 86 no: ACPI OperationRegions are not marked as reserved, 87 no further checks are performed. 88 89 acpi_force_table_verification [HW,ACPI] 90 Enable table checksum verification during early stage. 91 By default, this is disabled due to x86 early mapping 92 size limitation. 93 94 acpi_irq_balance [HW,ACPI] 95 ACPI will balance active IRQs 96 default in APIC mode 97 98 acpi_irq_nobalance [HW,ACPI] 99 ACPI will not move active IRQs (default) 100 default in PIC mode 101 102 acpi_irq_isa= [HW,ACPI] If irq_balance, mark listed IRQs used by ISA 103 Format: <irq>,<irq>... 104 105 acpi_irq_pci= [HW,ACPI] If irq_balance, clear listed IRQs for 106 use by PCI 107 Format: <irq>,<irq>... 108 109 acpi_mask_gpe= [HW,ACPI] 110 Due to the existence of _Lxx/_Exx, some GPEs triggered 111 by unsupported hardware/firmware features can result in 112 GPE floodings that cannot be automatically disabled by 113 the GPE dispatcher. 114 This facility can be used to prevent such uncontrolled 115 GPE floodings. 116 Format: <int> 117 118 acpi_no_auto_serialize [HW,ACPI] 119 Disable auto-serialization of AML methods 120 AML control methods that contain the opcodes to create 121 named objects will be marked as "Serialized" by the 122 auto-serialization feature. 123 This feature is enabled by default. 124 This option allows to turn off the feature. 125 126 acpi_no_memhotplug [ACPI] Disable memory hotplug. Useful for kdump 127 kernels. 128 129 acpi_no_static_ssdt [HW,ACPI] 130 Disable installation of static SSDTs at early boot time 131 By default, SSDTs contained in the RSDT/XSDT will be 132 installed automatically and they will appear under 133 /sys/firmware/acpi/tables. 134 This option turns off this feature. 135 Note that specifying this option does not affect 136 dynamic table installation which will install SSDT 137 tables to /sys/firmware/acpi/tables/dynamic. 138 139 acpi_rsdp= [ACPI,EFI,KEXEC] 140 Pass the RSDP address to the kernel, mostly used 141 on machines running EFI runtime service to boot the 142 second kernel for kdump. 143 144 acpi_os_name= [HW,ACPI] Tell ACPI BIOS the name of the OS 145 Format: To spoof as Windows 98: ="Microsoft Windows" 146 147 acpi_rev_override [ACPI] Override the _REV object to return 5 (instead 148 of 2 which is mandated by ACPI 6) as the supported ACPI 149 specification revision (when using this switch, it may 150 be necessary to carry out a cold reboot _twice_ in a 151 row to make it take effect on the platform firmware). 152 153 acpi_osi= [HW,ACPI] Modify list of supported OS interface strings 154 acpi_osi="string1" # add string1 155 acpi_osi="!string2" # remove string2 156 acpi_osi=!* # remove all strings 157 acpi_osi=! # disable all built-in OS vendor 158 strings 159 acpi_osi=!! # enable all built-in OS vendor 160 strings 161 acpi_osi= # disable all strings 162 163 'acpi_osi=!' can be used in combination with single or 164 multiple 'acpi_osi="string1"' to support specific OS 165 vendor string(s). Note that such command can only 166 affect the default state of the OS vendor strings, thus 167 it cannot affect the default state of the feature group 168 strings and the current state of the OS vendor strings, 169 specifying it multiple times through kernel command line 170 is meaningless. This command is useful when one do not 171 care about the state of the feature group strings which 172 should be controlled by the OSPM. 173 Examples: 174 1. 'acpi_osi=! acpi_osi="Windows 2000"' is equivalent 175 to 'acpi_osi="Windows 2000" acpi_osi=!', they all 176 can make '_OSI("Windows 2000")' TRUE. 177 178 'acpi_osi=' cannot be used in combination with other 179 'acpi_osi=' command lines, the _OSI method will not 180 exist in the ACPI namespace. NOTE that such command can 181 only affect the _OSI support state, thus specifying it 182 multiple times through kernel command line is also 183 meaningless. 184 Examples: 185 1. 'acpi_osi=' can make 'CondRefOf(_OSI, Local1)' 186 FALSE. 187 188 'acpi_osi=!*' can be used in combination with single or 189 multiple 'acpi_osi="string1"' to support specific 190 string(s). Note that such command can affect the 191 current state of both the OS vendor strings and the 192 feature group strings, thus specifying it multiple times 193 through kernel command line is meaningful. But it may 194 still not able to affect the final state of a string if 195 there are quirks related to this string. This command 196 is useful when one want to control the state of the 197 feature group strings to debug BIOS issues related to 198 the OSPM features. 199 Examples: 200 1. 'acpi_osi="Module Device" acpi_osi=!*' can make 201 '_OSI("Module Device")' FALSE. 202 2. 'acpi_osi=!* acpi_osi="Module Device"' can make 203 '_OSI("Module Device")' TRUE. 204 3. 'acpi_osi=! acpi_osi=!* acpi_osi="Windows 2000"' is 205 equivalent to 206 'acpi_osi=!* acpi_osi=! acpi_osi="Windows 2000"' 207 and 208 'acpi_osi=!* acpi_osi="Windows 2000" acpi_osi=!', 209 they all will make '_OSI("Windows 2000")' TRUE. 210 211 acpi_pm_good [X86] 212 Override the pmtimer bug detection: force the kernel 213 to assume that this machine's pmtimer latches its value 214 and always returns good values. 215 216 acpi_sci= [HW,ACPI] ACPI System Control Interrupt trigger mode 217 Format: { level | edge | high | low } 218 219 acpi_skip_timer_override [HW,ACPI] 220 Recognize and ignore IRQ0/pin2 Interrupt Override. 221 For broken nForce2 BIOS resulting in XT-PIC timer. 222 223 acpi_sleep= [HW,ACPI] Sleep options 224 Format: { s3_bios, s3_mode, s3_beep, s4_nohwsig, 225 old_ordering, nonvs, sci_force_enable, nobl } 226 See Documentation/power/video.txt for information on 227 s3_bios and s3_mode. 228 s3_beep is for debugging; it makes the PC's speaker beep 229 as soon as the kernel's real-mode entry point is called. 230 s4_nohwsig prevents ACPI hardware signature from being 231 used during resume from hibernation. 232 old_ordering causes the ACPI 1.0 ordering of the _PTS 233 control method, with respect to putting devices into 234 low power states, to be enforced (the ACPI 2.0 ordering 235 of _PTS is used by default). 236 nonvs prevents the kernel from saving/restoring the 237 ACPI NVS memory during suspend/hibernation and resume. 238 sci_force_enable causes the kernel to set SCI_EN directly 239 on resume from S1/S3 (which is against the ACPI spec, 240 but some broken systems don't work without it). 241 nobl causes the internal blacklist of systems known to 242 behave incorrectly in some ways with respect to system 243 suspend and resume to be ignored (use wisely). 244 245 acpi_use_timer_override [HW,ACPI] 246 Use timer override. For some broken Nvidia NF5 boards 247 that require a timer override, but don't have HPET 248 249 add_efi_memmap [EFI; X86] Include EFI memory map in 250 kernel's map of available physical RAM. 251 252 agp= [AGP] 253 { off | try_unsupported } 254 off: disable AGP support 255 try_unsupported: try to drive unsupported chipsets 256 (may crash computer or cause data corruption) 257 258 ALSA [HW,ALSA] 259 See Documentation/sound/alsa-configuration.rst 260 261 alignment= [KNL,ARM] 262 Allow the default userspace alignment fault handler 263 behaviour to be specified. Bit 0 enables warnings, 264 bit 1 enables fixups, and bit 2 sends a segfault. 265 266 align_va_addr= [X86-64] 267 Align virtual addresses by clearing slice [14:12] when 268 allocating a VMA at process creation time. This option 269 gives you up to 3% performance improvement on AMD F15h 270 machines (where it is enabled by default) for a 271 CPU-intensive style benchmark, and it can vary highly in 272 a microbenchmark depending on workload and compiler. 273 274 32: only for 32-bit processes 275 64: only for 64-bit processes 276 on: enable for both 32- and 64-bit processes 277 off: disable for both 32- and 64-bit processes 278 279 alloc_snapshot [FTRACE] 280 Allocate the ftrace snapshot buffer on boot up when the 281 main buffer is allocated. This is handy if debugging 282 and you need to use tracing_snapshot() on boot up, and 283 do not want to use tracing_snapshot_alloc() as it needs 284 to be done where GFP_KERNEL allocations are allowed. 285 286 amd_iommu= [HW,X86-64] 287 Pass parameters to the AMD IOMMU driver in the system. 288 Possible values are: 289 fullflush - enable flushing of IO/TLB entries when 290 they are unmapped. Otherwise they are 291 flushed before they will be reused, which 292 is a lot of faster 293 off - do not initialize any AMD IOMMU found in 294 the system 295 force_isolation - Force device isolation for all 296 devices. The IOMMU driver is not 297 allowed anymore to lift isolation 298 requirements as needed. This option 299 does not override iommu=pt 300 301 amd_iommu_dump= [HW,X86-64] 302 Enable AMD IOMMU driver option to dump the ACPI table 303 for AMD IOMMU. With this option enabled, AMD IOMMU 304 driver will print ACPI tables for AMD IOMMU during 305 IOMMU initialization. 306 307 amd_iommu_intr= [HW,X86-64] 308 Specifies one of the following AMD IOMMU interrupt 309 remapping modes: 310 legacy - Use legacy interrupt remapping mode. 311 vapic - Use virtual APIC mode, which allows IOMMU 312 to inject interrupts directly into guest. 313 This mode requires kvm-amd.avic=1. 314 (Default when IOMMU HW support is present.) 315 316 amijoy.map= [HW,JOY] Amiga joystick support 317 Map of devices attached to JOY0DAT and JOY1DAT 318 Format: <a>,<b> 319 See also Documentation/input/joydev/joystick.rst 320 321 analog.map= [HW,JOY] Analog joystick and gamepad support 322 Specifies type or capabilities of an analog joystick 323 connected to one of 16 gameports 324 Format: <type1>,<type2>,..<type16> 325 326 apc= [HW,SPARC] 327 Power management functions (SPARCstation-4/5 + deriv.) 328 Format: noidle 329 Disable APC CPU standby support. SPARCstation-Fox does 330 not play well with APC CPU idle - disable it if you have 331 APC and your system crashes randomly. 332 333 apic= [APIC,X86] Advanced Programmable Interrupt Controller 334 Change the output verbosity while booting 335 Format: { quiet (default) | verbose | debug } 336 Change the amount of debugging information output 337 when initialising the APIC and IO-APIC components. 338 For X86-32, this can also be used to specify an APIC 339 driver name. 340 Format: apic=driver_name 341 Examples: apic=bigsmp 342 343 apic_extnmi= [APIC,X86] External NMI delivery setting 344 Format: { bsp (default) | all | none } 345 bsp: External NMI is delivered only to CPU 0 346 all: External NMIs are broadcast to all CPUs as a 347 backup of CPU 0 348 none: External NMI is masked for all CPUs. This is 349 useful so that a dump capture kernel won't be 350 shot down by NMI 351 352 autoconf= [IPV6] 353 See Documentation/networking/ipv6.txt. 354 355 show_lapic= [APIC,X86] Advanced Programmable Interrupt Controller 356 Limit apic dumping. The parameter defines the maximal 357 number of local apics being dumped. Also it is possible 358 to set it to "all" by meaning -- no limit here. 359 Format: { 1 (default) | 2 | ... | all }. 360 The parameter valid if only apic=debug or 361 apic=verbose is specified. 362 Example: apic=debug show_lapic=all 363 364 apm= [APM] Advanced Power Management 365 See header of arch/x86/kernel/apm_32.c. 366 367 arcrimi= [HW,NET] ARCnet - "RIM I" (entirely mem-mapped) cards 368 Format: <io>,<irq>,<nodeID> 369 370 ataflop= [HW,M68k] 371 372 atarimouse= [HW,MOUSE] Atari Mouse 373 374 atkbd.extra= [HW] Enable extra LEDs and keys on IBM RapidAccess, 375 EzKey and similar keyboards 376 377 atkbd.reset= [HW] Reset keyboard during initialization 378 379 atkbd.set= [HW] Select keyboard code set 380 Format: <int> (2 = AT (default), 3 = PS/2) 381 382 atkbd.scroll= [HW] Enable scroll wheel on MS Office and similar 383 keyboards 384 385 atkbd.softraw= [HW] Choose between synthetic and real raw mode 386 Format: <bool> (0 = real, 1 = synthetic (default)) 387 388 atkbd.softrepeat= [HW] 389 Use software keyboard repeat 390 391 audit= [KNL] Enable the audit sub-system 392 Format: { "0" | "1" | "off" | "on" } 393 0 | off - kernel audit is disabled and can not be 394 enabled until the next reboot 395 unset - kernel audit is initialized but disabled and 396 will be fully enabled by the userspace auditd. 397 1 | on - kernel audit is initialized and partially 398 enabled, storing at most audit_backlog_limit 399 messages in RAM until it is fully enabled by the 400 userspace auditd. 401 Default: unset 402 403 audit_backlog_limit= [KNL] Set the audit queue size limit. 404 Format: <int> (must be >=0) 405 Default: 64 406 407 bau= [X86_UV] Enable the BAU on SGI UV. The default 408 behavior is to disable the BAU (i.e. bau=0). 409 Format: { "0" | "1" } 410 0 - Disable the BAU. 411 1 - Enable the BAU. 412 unset - Disable the BAU. 413 414 baycom_epp= [HW,AX25] 415 Format: <io>,<mode> 416 417 baycom_par= [HW,AX25] BayCom Parallel Port AX.25 Modem 418 Format: <io>,<mode> 419 See header of drivers/net/hamradio/baycom_par.c. 420 421 baycom_ser_fdx= [HW,AX25] 422 BayCom Serial Port AX.25 Modem (Full Duplex Mode) 423 Format: <io>,<irq>,<mode>[,<baud>] 424 See header of drivers/net/hamradio/baycom_ser_fdx.c. 425 426 baycom_ser_hdx= [HW,AX25] 427 BayCom Serial Port AX.25 Modem (Half Duplex Mode) 428 Format: <io>,<irq>,<mode> 429 See header of drivers/net/hamradio/baycom_ser_hdx.c. 430 431 blkdevparts= Manual partition parsing of block device(s) for 432 embedded devices based on command line input. 433 See Documentation/block/cmdline-partition.txt 434 435 boot_delay= Milliseconds to delay each printk during boot. 436 Values larger than 10 seconds (10000) are changed to 437 no delay (0). 438 Format: integer 439 440 bootmem_debug [KNL] Enable bootmem allocator debug messages. 441 442 bert_disable [ACPI] 443 Disable BERT OS support on buggy BIOSes. 444 445 bttv.card= [HW,V4L] bttv (bt848 + bt878 based grabber cards) 446 bttv.radio= Most important insmod options are available as 447 kernel args too. 448 bttv.pll= See Documentation/media/v4l-drivers/bttv.rst 449 bttv.tuner= 450 451 bulk_remove=off [PPC] This parameter disables the use of the pSeries 452 firmware feature for flushing multiple hpte entries 453 at a time. 454 455 c101= [NET] Moxa C101 synchronous serial card 456 457 cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection. 458 Sometimes CPU hardware bugs make them report the cache 459 size incorrectly. The kernel will attempt work arounds 460 to fix known problems, but for some CPUs it is not 461 possible to determine what the correct size should be. 462 This option provides an override for these situations. 463 464 ca_keys= [KEYS] This parameter identifies a specific key(s) on 465 the system trusted keyring to be used for certificate 466 trust validation. 467 format: { id:<keyid> | builtin } 468 469 cca= [MIPS] Override the kernel pages' cache coherency 470 algorithm. Accepted values range from 0 to 7 471 inclusive. See arch/mips/include/asm/pgtable-bits.h 472 for platform specific values (SB1, Loongson3 and 473 others). 474 475 ccw_timeout_log [S390] 476 See Documentation/s390/CommonIO for details. 477 478 cgroup_disable= [KNL] Disable a particular controller 479 Format: {name of the controller(s) to disable} 480 The effects of cgroup_disable=foo are: 481 - foo isn't auto-mounted if you mount all cgroups in 482 a single hierarchy 483 - foo isn't visible as an individually mountable 484 subsystem 485 {Currently only "memory" controller deal with this and 486 cut the overhead, others just disable the usage. So 487 only cgroup_disable=memory is actually worthy} 488 489 cgroup_no_v1= [KNL] Disable cgroup controllers and named hierarchies in v1 490 Format: { { controller | "all" | "named" } 491 [,{ controller | "all" | "named" }...] } 492 Like cgroup_disable, but only applies to cgroup v1; 493 the blacklisted controllers remain available in cgroup2. 494 "all" blacklists all controllers and "named" disables 495 named mounts. Specifying both "all" and "named" disables 496 all v1 hierarchies. 497 498 cgroup.memory= [KNL] Pass options to the cgroup memory controller. 499 Format: <string> 500 nosocket -- Disable socket memory accounting. 501 nokmem -- Disable kernel memory accounting. 502 503 checkreqprot [SELINUX] Set initial checkreqprot flag value. 504 Format: { "0" | "1" } 505 See security/selinux/Kconfig help text. 506 0 -- check protection applied by kernel (includes 507 any implied execute protection). 508 1 -- check protection requested by application. 509 Default value is set via a kernel config option. 510 Value can be changed at runtime via 511 /selinux/checkreqprot. 512 513 cio_ignore= [S390] 514 See Documentation/s390/CommonIO for details. 515 clk_ignore_unused 516 [CLK] 517 Prevents the clock framework from automatically gating 518 clocks that have not been explicitly enabled by a Linux 519 device driver but are enabled in hardware at reset or 520 by the bootloader/firmware. Note that this does not 521 force such clocks to be always-on nor does it reserve 522 those clocks in any way. This parameter is useful for 523 debug and development, but should not be needed on a 524 platform with proper driver support. For more 525 information, see Documentation/driver-api/clk.rst. 526 527 clock= [BUGS=X86-32, HW] gettimeofday clocksource override. 528 [Deprecated] 529 Forces specified clocksource (if available) to be used 530 when calculating gettimeofday(). If specified 531 clocksource is not available, it defaults to PIT. 532 Format: { pit | tsc | cyclone | pmtmr } 533 534 clocksource= Override the default clocksource 535 Format: <string> 536 Override the default clocksource and use the clocksource 537 with the name specified. 538 Some clocksource names to choose from, depending on 539 the platform: 540 [all] jiffies (this is the base, fallback clocksource) 541 [ACPI] acpi_pm 542 [ARM] imx_timer1,OSTS,netx_timer,mpu_timer2, 543 pxa_timer,timer3,32k_counter,timer0_1 544 [X86-32] pit,hpet,tsc; 545 scx200_hrt on Geode; cyclone on IBM x440 546 [MIPS] MIPS 547 [PARISC] cr16 548 [S390] tod 549 [SH] SuperH 550 [SPARC64] tick 551 [X86-64] hpet,tsc 552 553 clocksource.arm_arch_timer.evtstrm= 554 [ARM,ARM64] 555 Format: <bool> 556 Enable/disable the eventstream feature of the ARM 557 architected timer so that code using WFE-based polling 558 loops can be debugged more effectively on production 559 systems. 560 561 clearcpuid=BITNUM [X86] 562 Disable CPUID feature X for the kernel. See 563 arch/x86/include/asm/cpufeatures.h for the valid bit 564 numbers. Note the Linux specific bits are not necessarily 565 stable over kernel options, but the vendor specific 566 ones should be. 567 Also note that user programs calling CPUID directly 568 or using the feature without checking anything 569 will still see it. This just prevents it from 570 being used by the kernel or shown in /proc/cpuinfo. 571 Also note the kernel might malfunction if you disable 572 some critical bits. 573 574 cma=nn[MG]@[start[MG][-end[MG]]] 575 [ARM,X86,KNL] 576 Sets the size of kernel global memory area for 577 contiguous memory allocations and optionally the 578 placement constraint by the physical address range of 579 memory allocations. A value of 0 disables CMA 580 altogether. For more information, see 581 include/linux/dma-contiguous.h 582 583 cmo_free_hint= [PPC] Format: { yes | no } 584 Specify whether pages are marked as being inactive 585 when they are freed. This is used in CMO environments 586 to determine OS memory pressure for page stealing by 587 a hypervisor. 588 Default: yes 589 590 coherent_pool=nn[KMG] [ARM,KNL] 591 Sets the size of memory pool for coherent, atomic dma 592 allocations, by default set to 256K. 593 594 com20020= [HW,NET] ARCnet - COM20020 chipset 595 Format: 596 <io>[,<irq>[,<nodeID>[,<backplane>[,<ckp>[,<timeout>]]]]] 597 598 com90io= [HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers) 599 Format: <io>[,<irq>] 600 601 com90xx= [HW,NET] 602 ARCnet - COM90xx chipset (memory-mapped buffers) 603 Format: <io>[,<irq>[,<memstart>]] 604 605 condev= [HW,S390] console device 606 conmode= 607 608 console= [KNL] Output console device and options. 609 610 tty<n> Use the virtual console device <n>. 611 612 ttyS<n>[,options] 613 ttyUSB0[,options] 614 Use the specified serial port. The options are of 615 the form "bbbbpnf", where "bbbb" is the baud rate, 616 "p" is parity ("n", "o", or "e"), "n" is number of 617 bits, and "f" is flow control ("r" for RTS or 618 omit it). Default is "9600n8". 619 620 See Documentation/admin-guide/serial-console.rst for more 621 information. See 622 Documentation/networking/netconsole.txt for an 623 alternative. 624 625 uart[8250],io,<addr>[,options] 626 uart[8250],mmio,<addr>[,options] 627 uart[8250],mmio16,<addr>[,options] 628 uart[8250],mmio32,<addr>[,options] 629 uart[8250],0x<addr>[,options] 630 Start an early, polled-mode console on the 8250/16550 631 UART at the specified I/O port or MMIO address, 632 switching to the matching ttyS device later. 633 MMIO inter-register address stride is either 8-bit 634 (mmio), 16-bit (mmio16), or 32-bit (mmio32). 635 If none of [io|mmio|mmio16|mmio32], <addr> is assumed 636 to be equivalent to 'mmio'. 'options' are specified in 637 the same format described for ttyS above; if unspecified, 638 the h/w is not re-initialized. 639 640 hvc<n> Use the hypervisor console device <n>. This is for 641 both Xen and PowerPC hypervisors. 642 643 If the device connected to the port is not a TTY but a braille 644 device, prepend "brl," before the device type, for instance 645 console=brl,ttyS0 646 For now, only VisioBraille is supported. 647 648 console_msg_format= 649 [KNL] Change console messages format 650 default 651 By default we print messages on consoles in 652 "[time stamp] text\n" format (time stamp may not be 653 printed, depending on CONFIG_PRINTK_TIME or 654 `printk_time' param). 655 syslog 656 Switch to syslog format: "<%u>[time stamp] text\n" 657 IOW, each message will have a facility and loglevel 658 prefix. The format is similar to one used by syslog() 659 syscall, or to executing "dmesg -S --raw" or to reading 660 from /proc/kmsg. 661 662 consoleblank= [KNL] The console blank (screen saver) timeout in 663 seconds. A value of 0 disables the blank timer. 664 Defaults to 0. 665 666 coredump_filter= 667 [KNL] Change the default value for 668 /proc/<pid>/coredump_filter. 669 See also Documentation/filesystems/proc.txt. 670 671 coresight_cpu_debug.enable 672 [ARM,ARM64] 673 Format: <bool> 674 Enable/disable the CPU sampling based debugging. 675 0: default value, disable debugging 676 1: enable debugging at boot time 677 678 cpuidle.off=1 [CPU_IDLE] 679 disable the cpuidle sub-system 680 681 cpuidle.governor= 682 [CPU_IDLE] Name of the cpuidle governor to use. 683 684 cpufreq.off=1 [CPU_FREQ] 685 disable the cpufreq sub-system 686 687 cpu_init_udelay=N 688 [X86] Delay for N microsec between assert and de-assert 689 of APIC INIT to start processors. This delay occurs 690 on every CPU online, such as boot, and resume from suspend. 691 Default: 10000 692 693 cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver 694 Format: 695 <first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>] 696 697 crashkernel=size[KMG][@offset[KMG]] 698 [KNL] Using kexec, Linux can switch to a 'crash kernel' 699 upon panic. This parameter reserves the physical 700 memory region [offset, offset + size] for that kernel 701 image. If '@offset' is omitted, then a suitable offset 702 is selected automatically. Check 703 Documentation/kdump/kdump.txt for further details. 704 705 crashkernel=range1:size1[,range2:size2,...][@offset] 706 [KNL] Same as above, but depends on the memory 707 in the running system. The syntax of range is 708 start-[end] where start and end are both 709 a memory unit (amount[KMG]). See also 710 Documentation/kdump/kdump.txt for an example. 711 712 crashkernel=size[KMG],high 713 [KNL, x86_64] range could be above 4G. Allow kernel 714 to allocate physical memory region from top, so could 715 be above 4G if system have more than 4G ram installed. 716 Otherwise memory region will be allocated below 4G, if 717 available. 718 It will be ignored if crashkernel=X is specified. 719 crashkernel=size[KMG],low 720 [KNL, x86_64] range under 4G. When crashkernel=X,high 721 is passed, kernel could allocate physical memory region 722 above 4G, that cause second kernel crash on system 723 that require some amount of low memory, e.g. swiotlb 724 requires at least 64M+32K low memory, also enough extra 725 low memory is needed to make sure DMA buffers for 32-bit 726 devices won't run out. Kernel would try to allocate at 727 at least 256M below 4G automatically. 728 This one let user to specify own low range under 4G 729 for second kernel instead. 730 0: to disable low allocation. 731 It will be ignored when crashkernel=X,high is not used 732 or memory reserved is below 4G. 733 734 cryptomgr.notests 735 [KNL] Disable crypto self-tests 736 737 cs89x0_dma= [HW,NET] 738 Format: <dma> 739 740 cs89x0_media= [HW,NET] 741 Format: { rj45 | aui | bnc } 742 743 dasd= [HW,NET] 744 See header of drivers/s390/block/dasd_devmap.c. 745 746 db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port 747 (one device per port) 748 Format: <port#>,<type> 749 See also Documentation/input/devices/joystick-parport.rst 750 751 ddebug_query= [KNL,DYNAMIC_DEBUG] Enable debug messages at early boot 752 time. See 753 Documentation/admin-guide/dynamic-debug-howto.rst for 754 details. Deprecated, see dyndbg. 755 756 debug [KNL] Enable kernel debugging (events log level). 757 758 debug_boot_weak_hash 759 [KNL] Enable printing [hashed] pointers early in the 760 boot sequence. If enabled, we use a weak hash instead 761 of siphash to hash pointers. Use this option if you are 762 seeing instances of '(___ptrval___)') and need to see a 763 value (hashed pointer) instead. Cryptographically 764 insecure, please do not use on production kernels. 765 766 debug_locks_verbose= 767 [KNL] verbose self-tests 768 Format=<0|1> 769 Print debugging info while doing the locking API 770 self-tests. 771 We default to 0 (no extra messages), setting it to 772 1 will print _a lot_ more information - normally 773 only useful to kernel developers. 774 775 debug_objects [KNL] Enable object debugging 776 777 no_debug_objects 778 [KNL] Disable object debugging 779 780 debug_guardpage_minorder= 781 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this 782 parameter allows control of the order of pages that will 783 be intentionally kept free (and hence protected) by the 784 buddy allocator. Bigger value increase the probability 785 of catching random memory corruption, but reduce the 786 amount of memory for normal system use. The maximum 787 possible value is MAX_ORDER/2. Setting this parameter 788 to 1 or 2 should be enough to identify most random 789 memory corruption problems caused by bugs in kernel or 790 driver code when a CPU writes to (or reads from) a 791 random memory location. Note that there exists a class 792 of memory corruptions problems caused by buggy H/W or 793 F/W or by drivers badly programing DMA (basically when 794 memory is written at bus level and the CPU MMU is 795 bypassed) which are not detectable by 796 CONFIG_DEBUG_PAGEALLOC, hence this option will not help 797 tracking down these problems. 798 799 debug_pagealloc= 800 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this 801 parameter enables the feature at boot time. In 802 default, it is disabled. We can avoid allocating huge 803 chunk of memory for debug pagealloc if we don't enable 804 it at boot time and the system will work mostly same 805 with the kernel built without CONFIG_DEBUG_PAGEALLOC. 806 on: enable the feature 807 808 debugpat [X86] Enable PAT debugging 809 810 decnet.addr= [HW,NET] 811 Format: <area>[,<node>] 812 See also Documentation/networking/decnet.txt. 813 814 default_hugepagesz= 815 [same as hugepagesz=] The size of the default 816 HugeTLB page size. This is the size represented by 817 the legacy /proc/ hugepages APIs, used for SHM, and 818 default size when mounting hugetlbfs filesystems. 819 Defaults to the default architecture's huge page size 820 if not specified. 821 822 deferred_probe_timeout= 823 [KNL] Debugging option to set a timeout in seconds for 824 deferred probe to give up waiting on dependencies to 825 probe. Only specific dependencies (subsystems or 826 drivers) that have opted in will be ignored. A timeout of 0 827 will timeout at the end of initcalls. This option will also 828 dump out devices still on the deferred probe list after 829 retrying. 830 831 dhash_entries= [KNL] 832 Set number of hash buckets for dentry cache. 833 834 disable_1tb_segments [PPC] 835 Disables the use of 1TB hash page table segments. This 836 causes the kernel to fall back to 256MB segments which 837 can be useful when debugging issues that require an SLB 838 miss to occur. 839 840 disable= [IPV6] 841 See Documentation/networking/ipv6.txt. 842 843 hardened_usercopy= 844 [KNL] Under CONFIG_HARDENED_USERCOPY, whether 845 hardening is enabled for this boot. Hardened 846 usercopy checking is used to protect the kernel 847 from reading or writing beyond known memory 848 allocation boundaries as a proactive defense 849 against bounds-checking flaws in the kernel's 850 copy_to_user()/copy_from_user() interface. 851 on Perform hardened usercopy checks (default). 852 off Disable hardened usercopy checks. 853 854 disable_radix [PPC] 855 Disable RADIX MMU mode on POWER9 856 857 disable_cpu_apicid= [X86,APIC,SMP] 858 Format: <int> 859 The number of initial APIC ID for the 860 corresponding CPU to be disabled at boot, 861 mostly used for the kdump 2nd kernel to 862 disable BSP to wake up multiple CPUs without 863 causing system reset or hang due to sending 864 INIT from AP to BSP. 865 866 perf_v4_pmi= [X86,INTEL] 867 Format: <bool> 868 Disable Intel PMU counter freezing feature. 869 The feature only exists starting from 870 Arch Perfmon v4 (Skylake and newer). 871 872 disable_ddw [PPC/PSERIES] 873 Disable Dynamic DMA Window support. Use this if 874 to workaround buggy firmware. 875 876 disable_ipv6= [IPV6] 877 See Documentation/networking/ipv6.txt. 878 879 disable_mtrr_cleanup [X86] 880 The kernel tries to adjust MTRR layout from continuous 881 to discrete, to make X server driver able to add WB 882 entry later. This parameter disables that. 883 884 disable_mtrr_trim [X86, Intel and AMD only] 885 By default the kernel will trim any uncacheable 886 memory out of your available memory pool based on 887 MTRR settings. This parameter disables that behavior, 888 possibly causing your machine to run very slowly. 889 890 disable_timer_pin_1 [X86] 891 Disable PIN 1 of APIC timer 892 Can be useful to work around chipset bugs. 893 894 dis_ucode_ldr [X86] Disable the microcode loader. 895 896 dma_debug=off If the kernel is compiled with DMA_API_DEBUG support, 897 this option disables the debugging code at boot. 898 899 dma_debug_entries=<number> 900 This option allows to tune the number of preallocated 901 entries for DMA-API debugging code. One entry is 902 required per DMA-API allocation. Use this if the 903 DMA-API debugging code disables itself because the 904 architectural default is too low. 905 906 dma_debug_driver=<driver_name> 907 With this option the DMA-API debugging driver 908 filter feature can be enabled at boot time. Just 909 pass the driver to filter for as the parameter. 910 The filter can be disabled or changed to another 911 driver later using sysfs. 912 913 drm.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>] 914 Broken monitors, graphic adapters, KVMs and EDIDless 915 panels may send no or incorrect EDID data sets. 916 This parameter allows to specify an EDID data sets 917 in the /lib/firmware directory that are used instead. 918 Generic built-in EDID data sets are used, if one of 919 edid/1024x768.bin, edid/1280x1024.bin, 920 edid/1680x1050.bin, or edid/1920x1080.bin is given 921 and no file with the same name exists. Details and 922 instructions how to build your own EDID data are 923 available in Documentation/EDID/HOWTO.txt. An EDID 924 data set will only be used for a particular connector, 925 if its name and a colon are prepended to the EDID 926 name. Each connector may use a unique EDID data 927 set by separating the files with a comma. An EDID 928 data set with no connector name will be used for 929 any connectors not explicitly specified. 930 931 dscc4.setup= [NET] 932 933 dt_cpu_ftrs= [PPC] 934 Format: {"off" | "known"} 935 Control how the dt_cpu_ftrs device-tree binding is 936 used for CPU feature discovery and setup (if it 937 exists). 938 off: Do not use it, fall back to legacy cpu table. 939 known: Do not pass through unknown features to guests 940 or userspace, only those that the kernel is aware of. 941 942 dump_apple_properties [X86] 943 Dump name and content of EFI device properties on 944 x86 Macs. Useful for driver authors to determine 945 what data is available or for reverse-engineering. 946 947 dyndbg[="val"] [KNL,DYNAMIC_DEBUG] 948 module.dyndbg[="val"] 949 Enable debug messages at boot time. See 950 Documentation/admin-guide/dynamic-debug-howto.rst 951 for details. 952 953 nompx [X86] Disables Intel Memory Protection Extensions. 954 See Documentation/x86/intel_mpx.txt for more 955 information about the feature. 956 957 nopku [X86] Disable Memory Protection Keys CPU feature found 958 in some Intel CPUs. 959 960 module.async_probe [KNL] 961 Enable asynchronous probe on this module. 962 963 early_ioremap_debug [KNL] 964 Enable debug messages in early_ioremap support. This 965 is useful for tracking down temporary early mappings 966 which are not unmapped. 967 968 earlycon= [KNL] Output early console device and options. 969 970 [ARM64] The early console is determined by the 971 stdout-path property in device tree's chosen node, 972 or determined by the ACPI SPCR table. 973 974 [X86] When used with no options the early console is 975 determined by the ACPI SPCR table. 976 977 cdns,<addr>[,options] 978 Start an early, polled-mode console on a Cadence 979 (xuartps) serial port at the specified address. Only 980 supported option is baud rate. If baud rate is not 981 specified, the serial port must already be setup and 982 configured. 983 984 uart[8250],io,<addr>[,options] 985 uart[8250],mmio,<addr>[,options] 986 uart[8250],mmio32,<addr>[,options] 987 uart[8250],mmio32be,<addr>[,options] 988 uart[8250],0x<addr>[,options] 989 Start an early, polled-mode console on the 8250/16550 990 UART at the specified I/O port or MMIO address. 991 MMIO inter-register address stride is either 8-bit 992 (mmio) or 32-bit (mmio32 or mmio32be). 993 If none of [io|mmio|mmio32|mmio32be], <addr> is assumed 994 to be equivalent to 'mmio'. 'options' are specified 995 in the same format described for "console=ttyS<n>"; if 996 unspecified, the h/w is not initialized. 997 998 pl011,<addr> 999 pl011,mmio32,<addr> 1000 Start an early, polled-mode console on a pl011 serial 1001 port at the specified address. The pl011 serial port 1002 must already be setup and configured. Options are not 1003 yet supported. If 'mmio32' is specified, then only 1004 the driver will use only 32-bit accessors to read/write 1005 the device registers. 1006 1007 meson,<addr> 1008 Start an early, polled-mode console on a meson serial 1009 port at the specified address. The serial port must 1010 already be setup and configured. Options are not yet 1011 supported. 1012 1013 msm_serial,<addr> 1014 Start an early, polled-mode console on an msm serial 1015 port at the specified address. The serial port 1016 must already be setup and configured. Options are not 1017 yet supported. 1018 1019 msm_serial_dm,<addr> 1020 Start an early, polled-mode console on an msm serial 1021 dm port at the specified address. The serial port 1022 must already be setup and configured. Options are not 1023 yet supported. 1024 1025 owl,<addr> 1026 Start an early, polled-mode console on a serial port 1027 of an Actions Semi SoC, such as S500 or S900, at the 1028 specified address. The serial port must already be 1029 setup and configured. Options are not yet supported. 1030 1031 smh Use ARM semihosting calls for early console. 1032 1033 s3c2410,<addr> 1034 s3c2412,<addr> 1035 s3c2440,<addr> 1036 s3c6400,<addr> 1037 s5pv210,<addr> 1038 exynos4210,<addr> 1039 Use early console provided by serial driver available 1040 on Samsung SoCs, requires selecting proper type and 1041 a correct base address of the selected UART port. The 1042 serial port must already be setup and configured. 1043 Options are not yet supported. 1044 1045 lantiq,<addr> 1046 Start an early, polled-mode console on a lantiq serial 1047 (lqasc) port at the specified address. The serial port 1048 must already be setup and configured. Options are not 1049 yet supported. 1050 1051 lpuart,<addr> 1052 lpuart32,<addr> 1053 Use early console provided by Freescale LP UART driver 1054 found on Freescale Vybrid and QorIQ LS1021A processors. 1055 A valid base address must be provided, and the serial 1056 port must already be setup and configured. 1057 1058 ar3700_uart,<addr> 1059 Start an early, polled-mode console on the 1060 Armada 3700 serial port at the specified 1061 address. The serial port must already be setup 1062 and configured. Options are not yet supported. 1063 1064 qcom_geni,<addr> 1065 Start an early, polled-mode console on a Qualcomm 1066 Generic Interface (GENI) based serial port at the 1067 specified address. The serial port must already be 1068 setup and configured. Options are not yet supported. 1069 1070 earlyprintk= [X86,SH,ARM,M68k,S390] 1071 earlyprintk=vga 1072 earlyprintk=efi 1073 earlyprintk=sclp 1074 earlyprintk=xen 1075 earlyprintk=serial[,ttySn[,baudrate]] 1076 earlyprintk=serial[,0x...[,baudrate]] 1077 earlyprintk=ttySn[,baudrate] 1078 earlyprintk=dbgp[debugController#] 1079 earlyprintk=pciserial[,force],bus:device.function[,baudrate] 1080 earlyprintk=xdbc[xhciController#] 1081 1082 earlyprintk is useful when the kernel crashes before 1083 the normal console is initialized. It is not enabled by 1084 default because it has some cosmetic problems. 1085 1086 Append ",keep" to not disable it when the real console 1087 takes over. 1088 1089 Only one of vga, efi, serial, or usb debug port can 1090 be used at a time. 1091 1092 Currently only ttyS0 and ttyS1 may be specified by 1093 name. Other I/O ports may be explicitly specified 1094 on some architectures (x86 and arm at least) by 1095 replacing ttySn with an I/O port address, like this: 1096 earlyprintk=serial,0x1008,115200 1097 You can find the port for a given device in 1098 /proc/tty/driver/serial: 1099 2: uart:ST16650V2 port:00001008 irq:18 ... 1100 1101 Interaction with the standard serial driver is not 1102 very good. 1103 1104 The VGA and EFI output is eventually overwritten by 1105 the real console. 1106 1107 The xen output can only be used by Xen PV guests. 1108 1109 The sclp output can only be used on s390. 1110 1111 The optional "force" to "pciserial" enables use of a 1112 PCI device even when its classcode is not of the 1113 UART class. 1114 1115 edac_report= [HW,EDAC] Control how to report EDAC event 1116 Format: {"on" | "off" | "force"} 1117 on: enable EDAC to report H/W event. May be overridden 1118 by other higher priority error reporting module. 1119 off: disable H/W event reporting through EDAC. 1120 force: enforce the use of EDAC to report H/W event. 1121 default: on. 1122 1123 ekgdboc= [X86,KGDB] Allow early kernel console debugging 1124 ekgdboc=kbd 1125 1126 This is designed to be used in conjunction with 1127 the boot argument: earlyprintk=vga 1128 1129 edd= [EDD] 1130 Format: {"off" | "on" | "skip[mbr]"} 1131 1132 efi= [EFI] 1133 Format: { "old_map", "nochunk", "noruntime", "debug" } 1134 old_map [X86-64]: switch to the old ioremap-based EFI 1135 runtime services mapping. 32-bit still uses this one by 1136 default. 1137 nochunk: disable reading files in "chunks" in the EFI 1138 boot stub, as chunking can cause problems with some 1139 firmware implementations. 1140 noruntime : disable EFI runtime services support 1141 debug: enable misc debug output 1142 1143 efi_no_storage_paranoia [EFI; X86] 1144 Using this parameter you can use more than 50% of 1145 your efi variable storage. Use this parameter only if 1146 you are really sure that your UEFI does sane gc and 1147 fulfills the spec otherwise your board may brick. 1148 1149 efi_fake_mem= nn[KMG]@ss[KMG]:aa[,nn[KMG]@ss[KMG]:aa,..] [EFI; X86] 1150 Add arbitrary attribute to specific memory range by 1151 updating original EFI memory map. 1152 Region of memory which aa attribute is added to is 1153 from ss to ss+nn. 1154 If efi_fake_mem=2G@4G:0x10000,2G@0x10a0000000:0x10000 1155 is specified, EFI_MEMORY_MORE_RELIABLE(0x10000) 1156 attribute is added to range 0x100000000-0x180000000 and 1157 0x10a0000000-0x1120000000. 1158 1159 Using this parameter you can do debugging of EFI memmap 1160 related feature. For example, you can do debugging of 1161 Address Range Mirroring feature even if your box 1162 doesn't support it. 1163 1164 efivar_ssdt= [EFI; X86] Name of an EFI variable that contains an SSDT 1165 that is to be dynamically loaded by Linux. If there are 1166 multiple variables with the same name but with different 1167 vendor GUIDs, all of them will be loaded. See 1168 Documentation/acpi/ssdt-overlays.txt for details. 1169 1170 1171 eisa_irq_edge= [PARISC,HW] 1172 See header of drivers/parisc/eisa.c. 1173 1174 elanfreq= [X86-32] 1175 See comment before function elanfreq_setup() in 1176 arch/x86/kernel/cpu/cpufreq/elanfreq.c. 1177 1178 elevator= [IOSCHED] 1179 Format: {"cfq" | "deadline" | "noop"} 1180 See Documentation/block/cfq-iosched.txt and 1181 Documentation/block/deadline-iosched.txt for details. 1182 1183 elfcorehdr=[size[KMG]@]offset[KMG] [IA64,PPC,SH,X86,S390] 1184 Specifies physical address of start of kernel core 1185 image elf header and optionally the size. Generally 1186 kexec loader will pass this option to capture kernel. 1187 See Documentation/kdump/kdump.txt for details. 1188 1189 enable_mtrr_cleanup [X86] 1190 The kernel tries to adjust MTRR layout from continuous 1191 to discrete, to make X server driver able to add WB 1192 entry later. This parameter enables that. 1193 1194 enable_timer_pin_1 [X86] 1195 Enable PIN 1 of APIC timer 1196 Can be useful to work around chipset bugs 1197 (in particular on some ATI chipsets). 1198 The kernel tries to set a reasonable default. 1199 1200 enforcing [SELINUX] Set initial enforcing status. 1201 Format: {"0" | "1"} 1202 See security/selinux/Kconfig help text. 1203 0 -- permissive (log only, no denials). 1204 1 -- enforcing (deny and log). 1205 Default value is 0. 1206 Value can be changed at runtime via /selinux/enforce. 1207 1208 erst_disable [ACPI] 1209 Disable Error Record Serialization Table (ERST) 1210 support. 1211 1212 ether= [HW,NET] Ethernet cards parameters 1213 This option is obsoleted by the "netdev=" option, which 1214 has equivalent usage. See its documentation for details. 1215 1216 evm= [EVM] 1217 Format: { "fix" } 1218 Permit 'security.evm' to be updated regardless of 1219 current integrity status. 1220 1221 failslab= 1222 fail_page_alloc= 1223 fail_make_request=[KNL] 1224 General fault injection mechanism. 1225 Format: <interval>,<probability>,<space>,<times> 1226 See also Documentation/fault-injection/. 1227 1228 floppy= [HW] 1229 See Documentation/blockdev/floppy.txt. 1230 1231 force_pal_cache_flush 1232 [IA-64] Avoid check_sal_cache_flush which may hang on 1233 buggy SAL_CACHE_FLUSH implementations. Using this 1234 parameter will force ia64_sal_cache_flush to call 1235 ia64_pal_cache_flush instead of SAL_CACHE_FLUSH. 1236 1237 forcepae [X86-32] 1238 Forcefully enable Physical Address Extension (PAE). 1239 Many Pentium M systems disable PAE but may have a 1240 functionally usable PAE implementation. 1241 Warning: use of this parameter will taint the kernel 1242 and may cause unknown problems. 1243 1244 ftrace=[tracer] 1245 [FTRACE] will set and start the specified tracer 1246 as early as possible in order to facilitate early 1247 boot debugging. 1248 1249 ftrace_dump_on_oops[=orig_cpu] 1250 [FTRACE] will dump the trace buffers on oops. 1251 If no parameter is passed, ftrace will dump 1252 buffers of all CPUs, but if you pass orig_cpu, it will 1253 dump only the buffer of the CPU that triggered the 1254 oops. 1255 1256 ftrace_filter=[function-list] 1257 [FTRACE] Limit the functions traced by the function 1258 tracer at boot up. function-list is a comma separated 1259 list of functions. This list can be changed at run 1260 time by the set_ftrace_filter file in the debugfs 1261 tracing directory. 1262 1263 ftrace_notrace=[function-list] 1264 [FTRACE] Do not trace the functions specified in 1265 function-list. This list can be changed at run time 1266 by the set_ftrace_notrace file in the debugfs 1267 tracing directory. 1268 1269 ftrace_graph_filter=[function-list] 1270 [FTRACE] Limit the top level callers functions traced 1271 by the function graph tracer at boot up. 1272 function-list is a comma separated list of functions 1273 that can be changed at run time by the 1274 set_graph_function file in the debugfs tracing directory. 1275 1276 ftrace_graph_notrace=[function-list] 1277 [FTRACE] Do not trace from the functions specified in 1278 function-list. This list is a comma separated list of 1279 functions that can be changed at run time by the 1280 set_graph_notrace file in the debugfs tracing directory. 1281 1282 ftrace_graph_max_depth=<uint> 1283 [FTRACE] Used with the function graph tracer. This is 1284 the max depth it will trace into a function. This value 1285 can be changed at run time by the max_graph_depth file 1286 in the tracefs tracing directory. default: 0 (no limit) 1287 1288 gamecon.map[2|3]= 1289 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad 1290 support via parallel port (up to 5 devices per port) 1291 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5> 1292 See also Documentation/input/devices/joystick-parport.rst 1293 1294 gamma= [HW,DRM] 1295 1296 gart_fix_e820= [X86_64] disable the fix e820 for K8 GART 1297 Format: off | on 1298 default: on 1299 1300 gcov_persist= [GCOV] When non-zero (default), profiling data for 1301 kernel modules is saved and remains accessible via 1302 debugfs, even when the module is unloaded/reloaded. 1303 When zero, profiling data is discarded and associated 1304 debugfs files are removed at module unload time. 1305 1306 goldfish [X86] Enable the goldfish android emulator platform. 1307 Don't use this when you are not running on the 1308 android emulator 1309 1310 gpt [EFI] Forces disk with valid GPT signature but 1311 invalid Protective MBR to be treated as GPT. If the 1312 primary GPT is corrupted, it enables the backup/alternate 1313 GPT to be used instead. 1314 1315 grcan.enable0= [HW] Configuration of physical interface 0. Determines 1316 the "Enable 0" bit of the configuration register. 1317 Format: 0 | 1 1318 Default: 0 1319 grcan.enable1= [HW] Configuration of physical interface 1. Determines 1320 the "Enable 0" bit of the configuration register. 1321 Format: 0 | 1 1322 Default: 0 1323 grcan.select= [HW] Select which physical interface to use. 1324 Format: 0 | 1 1325 Default: 0 1326 grcan.txsize= [HW] Sets the size of the tx buffer. 1327 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0. 1328 Default: 1024 1329 grcan.rxsize= [HW] Sets the size of the rx buffer. 1330 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0. 1331 Default: 1024 1332 1333 gpio-mockup.gpio_mockup_ranges 1334 [HW] Sets the ranges of gpiochip of for this device. 1335 Format: <start1>,<end1>,<start2>,<end2>... 1336 1337 hardlockup_all_cpu_backtrace= 1338 [KNL] Should the hard-lockup detector generate 1339 backtraces on all cpus. 1340 Format: <integer> 1341 1342 hashdist= [KNL,NUMA] Large hashes allocated during boot 1343 are distributed across NUMA nodes. Defaults on 1344 for 64-bit NUMA, off otherwise. 1345 Format: 0 | 1 (for off | on) 1346 1347 hcl= [IA-64] SGI's Hardware Graph compatibility layer 1348 1349 hd= [EIDE] (E)IDE hard drive subsystem geometry 1350 Format: <cyl>,<head>,<sect> 1351 1352 hest_disable [ACPI] 1353 Disable Hardware Error Source Table (HEST) support; 1354 corresponding firmware-first mode error processing 1355 logic will be disabled. 1356 1357 highmem=nn[KMG] [KNL,BOOT] forces the highmem zone to have an exact 1358 size of <nn>. This works even on boxes that have no 1359 highmem otherwise. This also works to reduce highmem 1360 size on bigger boxes. 1361 1362 highres= [KNL] Enable/disable high resolution timer mode. 1363 Valid parameters: "on", "off" 1364 Default: "on" 1365 1366 hisax= [HW,ISDN] 1367 See Documentation/isdn/README.HiSax. 1368 1369 hlt [BUGS=ARM,SH] 1370 1371 hpet= [X86-32,HPET] option to control HPET usage 1372 Format: { enable (default) | disable | force | 1373 verbose } 1374 disable: disable HPET and use PIT instead 1375 force: allow force enabled of undocumented chips (ICH4, 1376 VIA, nVidia) 1377 verbose: show contents of HPET registers during setup 1378 1379 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET 1380 registers. Default set by CONFIG_HPET_MMAP_DEFAULT. 1381 1382 hugepages= [HW,X86-32,IA-64] HugeTLB pages to allocate at boot. 1383 hugepagesz= [HW,IA-64,PPC,X86-64] The size of the HugeTLB pages. 1384 On x86-64 and powerpc, this option can be specified 1385 multiple times interleaved with hugepages= to reserve 1386 huge pages of different sizes. Valid pages sizes on 1387 x86-64 are 2M (when the CPU supports "pse") and 1G 1388 (when the CPU supports the "pdpe1gb" cpuinfo flag). 1389 1390 hung_task_panic= 1391 [KNL] Should the hung task detector generate panics. 1392 Format: <integer> 1393 1394 A nonzero value instructs the kernel to panic when a 1395 hung task is detected. The default value is controlled 1396 by the CONFIG_BOOTPARAM_HUNG_TASK_PANIC build-time 1397 option. The value selected by this boot parameter can 1398 be changed later by the kernel.hung_task_panic sysctl. 1399 1400 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC) 1401 terminal devices. Valid values: 0..8 1402 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs. 1403 If specified, z/VM IUCV HVC accepts connections 1404 from listed z/VM user IDs only. 1405 1406 hv_nopvspin [X86,HYPER_V] Disables the paravirt spinlock optimizations 1407 which allow the hypervisor to 'idle' the 1408 guest on lock contention. 1409 1410 keep_bootcon [KNL] 1411 Do not unregister boot console at start. This is only 1412 useful for debugging when something happens in the window 1413 between unregistering the boot console and initializing 1414 the real console. 1415 1416 i2c_bus= [HW] Override the default board specific I2C bus speed 1417 or register an additional I2C bus that is not 1418 registered from board initialization code. 1419 Format: 1420 <bus_id>,<clkrate> 1421 1422 i8042.debug [HW] Toggle i8042 debug mode 1423 i8042.unmask_kbd_data 1424 [HW] Enable printing of interrupt data from the KBD port 1425 (disabled by default, and as a pre-condition 1426 requires that i8042.debug=1 be enabled) 1427 i8042.direct [HW] Put keyboard port into non-translated mode 1428 i8042.dumbkbd [HW] Pretend that controller can only read data from 1429 keyboard and cannot control its state 1430 (Don't attempt to blink the leds) 1431 i8042.noaux [HW] Don't check for auxiliary (== mouse) port 1432 i8042.nokbd [HW] Don't check/create keyboard port 1433 i8042.noloop [HW] Disable the AUX Loopback command while probing 1434 for the AUX port 1435 i8042.nomux [HW] Don't check presence of an active multiplexing 1436 controller 1437 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX 1438 controllers 1439 i8042.notimeout [HW] Ignore timeout condition signalled by controller 1440 i8042.reset [HW] Reset the controller during init, cleanup and 1441 suspend-to-ram transitions, only during s2r 1442 transitions, or never reset 1443 Format: { 1 | Y | y | 0 | N | n } 1444 1, Y, y: always reset controller 1445 0, N, n: don't ever reset controller 1446 Default: only on s2r transitions on x86; most other 1447 architectures force reset to be always executed 1448 i8042.unlock [HW] Unlock (ignore) the keylock 1449 i8042.kbdreset [HW] Reset device connected to KBD port 1450 1451 i810= [HW,DRM] 1452 1453 i8k.ignore_dmi [HW] Continue probing hardware even if DMI data 1454 indicates that the driver is running on unsupported 1455 hardware. 1456 i8k.force [HW] Activate i8k driver even if SMM BIOS signature 1457 does not match list of supported models. 1458 i8k.power_status 1459 [HW] Report power status in /proc/i8k 1460 (disabled by default) 1461 i8k.restricted [HW] Allow controlling fans only if SYS_ADMIN 1462 capability is set. 1463 1464 i915.invert_brightness= 1465 [DRM] Invert the sense of the variable that is used to 1466 set the brightness of the panel backlight. Normally a 1467 brightness value of 0 indicates backlight switched off, 1468 and the maximum of the brightness value sets the backlight 1469 to maximum brightness. If this parameter is set to 0 1470 (default) and the machine requires it, or this parameter 1471 is set to 1, a brightness value of 0 sets the backlight 1472 to maximum brightness, and the maximum of the brightness 1473 value switches the backlight off. 1474 -1 -- never invert brightness 1475 0 -- machine default 1476 1 -- force brightness inversion 1477 1478 icn= [HW,ISDN] 1479 Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]] 1480 1481 ide-core.nodma= [HW] (E)IDE subsystem 1482 Format: =0.0 to prevent dma on hda, =0.1 hdb =1.0 hdc 1483 .vlb_clock .pci_clock .noflush .nohpa .noprobe .nowerr 1484 .cdrom .chs .ignore_cable are additional options 1485 See Documentation/ide/ide.txt. 1486 1487 ide-generic.probe-mask= [HW] (E)IDE subsystem 1488 Format: <int> 1489 Probe mask for legacy ISA IDE ports. Depending on 1490 platform up to 6 ports are supported, enabled by 1491 setting corresponding bits in the mask to 1. The 1492 default value is 0x0, which has a special meaning. 1493 On systems that have PCI, it triggers scanning the 1494 PCI bus for the first and the second port, which 1495 are then probed. On systems without PCI the value 1496 of 0x0 enables probing the two first ports as if it 1497 was 0x3. 1498 1499 ide-pci-generic.all-generic-ide [HW] (E)IDE subsystem 1500 Claim all unknown PCI IDE storage controllers. 1501 1502 idle= [X86] 1503 Format: idle=poll, idle=halt, idle=nomwait 1504 Poll forces a polling idle loop that can slightly 1505 improve the performance of waking up a idle CPU, but 1506 will use a lot of power and make the system run hot. 1507 Not recommended. 1508 idle=halt: Halt is forced to be used for CPU idle. 1509 In such case C2/C3 won't be used again. 1510 idle=nomwait: Disable mwait for CPU C-states 1511 1512 ieee754= [MIPS] Select IEEE Std 754 conformance mode 1513 Format: { strict | legacy | 2008 | relaxed } 1514 Default: strict 1515 1516 Choose which programs will be accepted for execution 1517 based on the IEEE 754 NaN encoding(s) supported by 1518 the FPU and the NaN encoding requested with the value 1519 of an ELF file header flag individually set by each 1520 binary. Hardware implementations are permitted to 1521 support either or both of the legacy and the 2008 NaN 1522 encoding mode. 1523 1524 Available settings are as follows: 1525 strict accept binaries that request a NaN encoding 1526 supported by the FPU 1527 legacy only accept legacy-NaN binaries, if supported 1528 by the FPU 1529 2008 only accept 2008-NaN binaries, if supported 1530 by the FPU 1531 relaxed accept any binaries regardless of whether 1532 supported by the FPU 1533 1534 The FPU emulator is always able to support both NaN 1535 encodings, so if no FPU hardware is present or it has 1536 been disabled with 'nofpu', then the settings of 1537 'legacy' and '2008' strap the emulator accordingly, 1538 'relaxed' straps the emulator for both legacy-NaN and 1539 2008-NaN, whereas 'strict' enables legacy-NaN only on 1540 legacy processors and both NaN encodings on MIPS32 or 1541 MIPS64 CPUs. 1542 1543 The setting for ABS.fmt/NEG.fmt instruction execution 1544 mode generally follows that for the NaN encoding, 1545 except where unsupported by hardware. 1546 1547 ignore_loglevel [KNL] 1548 Ignore loglevel setting - this will print /all/ 1549 kernel messages to the console. Useful for debugging. 1550 We also add it as printk module parameter, so users 1551 could change it dynamically, usually by 1552 /sys/module/printk/parameters/ignore_loglevel. 1553 1554 ignore_rlimit_data 1555 Ignore RLIMIT_DATA setting for data mappings, 1556 print warning at first misuse. Can be changed via 1557 /sys/module/kernel/parameters/ignore_rlimit_data. 1558 1559 ihash_entries= [KNL] 1560 Set number of hash buckets for inode cache. 1561 1562 ima_appraise= [IMA] appraise integrity measurements 1563 Format: { "off" | "enforce" | "fix" | "log" } 1564 default: "enforce" 1565 1566 ima_appraise_tcb [IMA] 1567 The builtin appraise policy appraises all files 1568 owned by uid=0. 1569 1570 ima_canonical_fmt [IMA] 1571 Use the canonical format for the binary runtime 1572 measurements, instead of host native format. 1573 1574 ima_hash= [IMA] 1575 Format: { md5 | sha1 | rmd160 | sha256 | sha384 1576 | sha512 | ... } 1577 default: "sha1" 1578 1579 The list of supported hash algorithms is defined 1580 in crypto/hash_info.h. 1581 1582 ima_policy= [IMA] 1583 The builtin policies to load during IMA setup. 1584 Format: "tcb | appraise_tcb | secure_boot | 1585 fail_securely" 1586 1587 The "tcb" policy measures all programs exec'd, files 1588 mmap'd for exec, and all files opened with the read 1589 mode bit set by either the effective uid (euid=0) or 1590 uid=0. 1591 1592 The "appraise_tcb" policy appraises the integrity of 1593 all files owned by root. (This is the equivalent 1594 of ima_appraise_tcb.) 1595 1596 The "secure_boot" policy appraises the integrity 1597 of files (eg. kexec kernel image, kernel modules, 1598 firmware, policy, etc) based on file signatures. 1599 1600 The "fail_securely" policy forces file signature 1601 verification failure also on privileged mounted 1602 filesystems with the SB_I_UNVERIFIABLE_SIGNATURE 1603 flag. 1604 1605 ima_tcb [IMA] Deprecated. Use ima_policy= instead. 1606 Load a policy which meets the needs of the Trusted 1607 Computing Base. This means IMA will measure all 1608 programs exec'd, files mmap'd for exec, and all files 1609 opened for read by uid=0. 1610 1611 ima_template= [IMA] 1612 Select one of defined IMA measurements template formats. 1613 Formats: { "ima" | "ima-ng" | "ima-sig" } 1614 Default: "ima-ng" 1615 1616 ima_template_fmt= 1617 [IMA] Define a custom template format. 1618 Format: { "field1|...|fieldN" } 1619 1620 ima.ahash_minsize= [IMA] Minimum file size for asynchronous hash usage 1621 Format: <min_file_size> 1622 Set the minimal file size for using asynchronous hash. 1623 If left unspecified, ahash usage is disabled. 1624 1625 ahash performance varies for different data sizes on 1626 different crypto accelerators. This option can be used 1627 to achieve the best performance for a particular HW. 1628 1629 ima.ahash_bufsize= [IMA] Asynchronous hash buffer size 1630 Format: <bufsize> 1631 Set hashing buffer size. Default: 4k. 1632 1633 ahash performance varies for different chunk sizes on 1634 different crypto accelerators. This option can be used 1635 to achieve best performance for particular HW. 1636 1637 init= [KNL] 1638 Format: <full_path> 1639 Run specified binary instead of /sbin/init as init 1640 process. 1641 1642 initcall_debug [KNL] Trace initcalls as they are executed. Useful 1643 for working out where the kernel is dying during 1644 startup. 1645 1646 initcall_blacklist= [KNL] Do not execute a comma-separated list of 1647 initcall functions. Useful for debugging built-in 1648 modules and initcalls. 1649 1650 initrd= [BOOT] Specify the location of the initial ramdisk 1651 1652 init_pkru= [x86] Specify the default memory protection keys rights 1653 register contents for all processes. 0x55555554 by 1654 default (disallow access to all but pkey 0). Can 1655 override in debugfs after boot. 1656 1657 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver 1658 Format: <irq> 1659 1660 int_pln_enable [x86] Enable power limit notification interrupt 1661 1662 integrity_audit=[IMA] 1663 Format: { "0" | "1" } 1664 0 -- basic integrity auditing messages. (Default) 1665 1 -- additional integrity auditing messages. 1666 1667 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option 1668 on 1669 Enable intel iommu driver. 1670 off 1671 Disable intel iommu driver. 1672 igfx_off [Default Off] 1673 By default, gfx is mapped as normal device. If a gfx 1674 device has a dedicated DMAR unit, the DMAR unit is 1675 bypassed by not enabling DMAR with this option. In 1676 this case, gfx device will use physical address for 1677 DMA. 1678 forcedac [x86_64] 1679 With this option iommu will not optimize to look 1680 for io virtual address below 32-bit forcing dual 1681 address cycle on pci bus for cards supporting greater 1682 than 32-bit addressing. The default is to look 1683 for translation below 32-bit and if not available 1684 then look in the higher range. 1685 strict [Default Off] 1686 With this option on every unmap_single operation will 1687 result in a hardware IOTLB flush operation as opposed 1688 to batching them for performance. 1689 sp_off [Default Off] 1690 By default, super page will be supported if Intel IOMMU 1691 has the capability. With this option, super page will 1692 not be supported. 1693 sm_off [Default Off] 1694 By default, scalable mode will be supported if the 1695 hardware advertises that it has support for the scalable 1696 mode translation. With this option set, scalable mode 1697 will not be used even on hardware which claims to support 1698 it. 1699 tboot_noforce [Default Off] 1700 Do not force the Intel IOMMU enabled under tboot. 1701 By default, tboot will force Intel IOMMU on, which 1702 could harm performance of some high-throughput 1703 devices like 40GBit network cards, even if identity 1704 mapping is enabled. 1705 Note that using this option lowers the security 1706 provided by tboot because it makes the system 1707 vulnerable to DMA attacks. 1708 1709 intel_idle.max_cstate= [KNL,HW,ACPI,X86] 1710 0 disables intel_idle and fall back on acpi_idle. 1711 1 to 9 specify maximum depth of C-state. 1712 1713 intel_pstate= [X86] 1714 disable 1715 Do not enable intel_pstate as the default 1716 scaling driver for the supported processors 1717 passive 1718 Use intel_pstate as a scaling driver, but configure it 1719 to work with generic cpufreq governors (instead of 1720 enabling its internal governor). This mode cannot be 1721 used along with the hardware-managed P-states (HWP) 1722 feature. 1723 force 1724 Enable intel_pstate on systems that prohibit it by default 1725 in favor of acpi-cpufreq. Forcing the intel_pstate driver 1726 instead of acpi-cpufreq may disable platform features, such 1727 as thermal controls and power capping, that rely on ACPI 1728 P-States information being indicated to OSPM and therefore 1729 should be used with caution. This option does not work with 1730 processors that aren't supported by the intel_pstate driver 1731 or on platforms that use pcc-cpufreq instead of acpi-cpufreq. 1732 no_hwp 1733 Do not enable hardware P state control (HWP) 1734 if available. 1735 hwp_only 1736 Only load intel_pstate on systems which support 1737 hardware P state control (HWP) if available. 1738 support_acpi_ppc 1739 Enforce ACPI _PPC performance limits. If the Fixed ACPI 1740 Description Table, specifies preferred power management 1741 profile as "Enterprise Server" or "Performance Server", 1742 then this feature is turned on by default. 1743 per_cpu_perf_limits 1744 Allow per-logical-CPU P-State performance control limits using 1745 cpufreq sysfs interface 1746 1747 intremap= [X86-64, Intel-IOMMU] 1748 on enable Interrupt Remapping (default) 1749 off disable Interrupt Remapping 1750 nosid disable Source ID checking 1751 no_x2apic_optout 1752 BIOS x2APIC opt-out request will be ignored 1753 nopost disable Interrupt Posting 1754 1755 iomem= Disable strict checking of access to MMIO memory 1756 strict regions from userspace. 1757 relaxed 1758 1759 iommu= [x86] 1760 off 1761 force 1762 noforce 1763 biomerge 1764 panic 1765 nopanic 1766 merge 1767 nomerge 1768 soft 1769 pt [x86] 1770 nopt [x86] 1771 nobypass [PPC/POWERNV] 1772 Disable IOMMU bypass, using IOMMU for PCI devices. 1773 1774 iommu.strict= [ARM64] Configure TLB invalidation behaviour 1775 Format: { "0" | "1" } 1776 0 - Lazy mode. 1777 Request that DMA unmap operations use deferred 1778 invalidation of hardware TLBs, for increased 1779 throughput at the cost of reduced device isolation. 1780 Will fall back to strict mode if not supported by 1781 the relevant IOMMU driver. 1782 1 - Strict mode (default). 1783 DMA unmap operations invalidate IOMMU hardware TLBs 1784 synchronously. 1785 1786 iommu.passthrough= 1787 [ARM64] Configure DMA to bypass the IOMMU by default. 1788 Format: { "0" | "1" } 1789 0 - Use IOMMU translation for DMA. 1790 1 - Bypass the IOMMU for DMA. 1791 unset - Use value of CONFIG_IOMMU_DEFAULT_PASSTHROUGH. 1792 1793 io7= [HW] IO7 for Marvel based alpha systems 1794 See comment before marvel_specify_io7 in 1795 arch/alpha/kernel/core_marvel.c. 1796 1797 io_delay= [X86] I/O delay method 1798 0x80 1799 Standard port 0x80 based delay 1800 0xed 1801 Alternate port 0xed based delay (needed on some systems) 1802 udelay 1803 Simple two microseconds delay 1804 none 1805 No delay 1806 1807 ip= [IP_PNP] 1808 See Documentation/filesystems/nfs/nfsroot.txt. 1809 1810 irqaffinity= [SMP] Set the default irq affinity mask 1811 The argument is a cpu list, as described above. 1812 1813 irqchip.gicv2_force_probe= 1814 [ARM, ARM64] 1815 Format: <bool> 1816 Force the kernel to look for the second 4kB page 1817 of a GICv2 controller even if the memory range 1818 exposed by the device tree is too small. 1819 1820 irqchip.gicv3_nolpi= 1821 [ARM, ARM64] 1822 Force the kernel to ignore the availability of 1823 LPIs (and by consequence ITSs). Intended for system 1824 that use the kernel as a bootloader, and thus want 1825 to let secondary kernels in charge of setting up 1826 LPIs. 1827 1828 irqfixup [HW] 1829 When an interrupt is not handled search all handlers 1830 for it. Intended to get systems with badly broken 1831 firmware running. 1832 1833 irqpoll [HW] 1834 When an interrupt is not handled search all handlers 1835 for it. Also check all handlers each timer 1836 interrupt. Intended to get systems with badly broken 1837 firmware running. 1838 1839 isapnp= [ISAPNP] 1840 Format: <RDP>,<reset>,<pci_scan>,<verbosity> 1841 1842 isolcpus= [KNL,SMP,ISOL] Isolate a given set of CPUs from disturbance. 1843 [Deprecated - use cpusets instead] 1844 Format: [flag-list,]<cpu-list> 1845 1846 Specify one or more CPUs to isolate from disturbances 1847 specified in the flag list (default: domain): 1848 1849 nohz 1850 Disable the tick when a single task runs. 1851 1852 A residual 1Hz tick is offloaded to workqueues, which you 1853 need to affine to housekeeping through the global 1854 workqueue's affinity configured via the 1855 /sys/devices/virtual/workqueue/cpumask sysfs file, or 1856 by using the 'domain' flag described below. 1857 1858 NOTE: by default the global workqueue runs on all CPUs, 1859 so to protect individual CPUs the 'cpumask' file has to 1860 be configured manually after bootup. 1861 1862 domain 1863 Isolate from the general SMP balancing and scheduling 1864 algorithms. Note that performing domain isolation this way 1865 is irreversible: it's not possible to bring back a CPU to 1866 the domains once isolated through isolcpus. It's strongly 1867 advised to use cpusets instead to disable scheduler load 1868 balancing through the "cpuset.sched_load_balance" file. 1869 It offers a much more flexible interface where CPUs can 1870 move in and out of an isolated set anytime. 1871 1872 You can move a process onto or off an "isolated" CPU via 1873 the CPU affinity syscalls or cpuset. 1874 <cpu number> begins at 0 and the maximum value is 1875 "number of CPUs in system - 1". 1876 1877 The format of <cpu-list> is described above. 1878 1879 1880 1881 iucv= [HW,NET] 1882 1883 ivrs_ioapic [HW,X86_64] 1884 Provide an override to the IOAPIC-ID<->DEVICE-ID 1885 mapping provided in the IVRS ACPI table. For 1886 example, to map IOAPIC-ID decimal 10 to 1887 PCI device 00:14.0 write the parameter as: 1888 ivrs_ioapic[10]=00:14.0 1889 1890 ivrs_hpet [HW,X86_64] 1891 Provide an override to the HPET-ID<->DEVICE-ID 1892 mapping provided in the IVRS ACPI table. For 1893 example, to map HPET-ID decimal 0 to 1894 PCI device 00:14.0 write the parameter as: 1895 ivrs_hpet[0]=00:14.0 1896 1897 ivrs_acpihid [HW,X86_64] 1898 Provide an override to the ACPI-HID:UID<->DEVICE-ID 1899 mapping provided in the IVRS ACPI table. For 1900 example, to map UART-HID:UID AMD0020:0 to 1901 PCI device 00:14.5 write the parameter as: 1902 ivrs_acpihid[00:14.5]=AMD0020:0 1903 1904 js= [HW,JOY] Analog joystick 1905 See Documentation/input/joydev/joystick.rst. 1906 1907 nokaslr [KNL] 1908 When CONFIG_RANDOMIZE_BASE is set, this disables 1909 kernel and module base offset ASLR (Address Space 1910 Layout Randomization). 1911 1912 kasan_multi_shot 1913 [KNL] Enforce KASAN (Kernel Address Sanitizer) to print 1914 report on every invalid memory access. Without this 1915 parameter KASAN will print report only for the first 1916 invalid access. 1917 1918 keepinitrd [HW,ARM] 1919 1920 kernelcore= [KNL,X86,IA-64,PPC] 1921 Format: nn[KMGTPE] | nn% | "mirror" 1922 This parameter specifies the amount of memory usable by 1923 the kernel for non-movable allocations. The requested 1924 amount is spread evenly throughout all nodes in the 1925 system as ZONE_NORMAL. The remaining memory is used for 1926 movable memory in its own zone, ZONE_MOVABLE. In the 1927 event, a node is too small to have both ZONE_NORMAL and 1928 ZONE_MOVABLE, kernelcore memory will take priority and 1929 other nodes will have a larger ZONE_MOVABLE. 1930 1931 ZONE_MOVABLE is used for the allocation of pages that 1932 may be reclaimed or moved by the page migration 1933 subsystem. Note that allocations like PTEs-from-HighMem 1934 still use the HighMem zone if it exists, and the Normal 1935 zone if it does not. 1936 1937 It is possible to specify the exact amount of memory in 1938 the form of "nn[KMGTPE]", a percentage of total system 1939 memory in the form of "nn%", or "mirror". If "mirror" 1940 option is specified, mirrored (reliable) memory is used 1941 for non-movable allocations and remaining memory is used 1942 for Movable pages. "nn[KMGTPE]", "nn%", and "mirror" 1943 are exclusive, so you cannot specify multiple forms. 1944 1945 kgdbdbgp= [KGDB,HW] kgdb over EHCI usb debug port. 1946 Format: <Controller#>[,poll interval] 1947 The controller # is the number of the ehci usb debug 1948 port as it is probed via PCI. The poll interval is 1949 optional and is the number seconds in between 1950 each poll cycle to the debug port in case you need 1951 the functionality for interrupting the kernel with 1952 gdb or control-c on the dbgp connection. When 1953 not using this parameter you use sysrq-g to break into 1954 the kernel debugger. 1955 1956 kgdboc= [KGDB,HW] kgdb over consoles. 1957 Requires a tty driver that supports console polling, 1958 or a supported polling keyboard driver (non-usb). 1959 Serial only format: <serial_device>[,baud] 1960 keyboard only format: kbd 1961 keyboard and serial format: kbd,<serial_device>[,baud] 1962 Optional Kernel mode setting: 1963 kms, kbd format: kms,kbd 1964 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud] 1965 1966 kgdbwait [KGDB] Stop kernel execution and enter the 1967 kernel debugger at the earliest opportunity. 1968 1969 kmac= [MIPS] korina ethernet MAC address. 1970 Configure the RouterBoard 532 series on-chip 1971 Ethernet adapter MAC address. 1972 1973 kmemleak= [KNL] Boot-time kmemleak enable/disable 1974 Valid arguments: on, off 1975 Default: on 1976 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y, 1977 the default is off. 1978 1979 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs. 1980 Default is 0 (don't ignore, but inject #GP) 1981 1982 kvm.enable_vmware_backdoor=[KVM] Support VMware backdoor PV interface. 1983 Default is false (don't support). 1984 1985 kvm.mmu_audit= [KVM] This is a R/W parameter which allows audit 1986 KVM MMU at runtime. 1987 Default is 0 (off) 1988 1989 kvm-amd.nested= [KVM,AMD] Allow nested virtualization in KVM/SVM. 1990 Default is 1 (enabled) 1991 1992 kvm-amd.npt= [KVM,AMD] Disable nested paging (virtualized MMU) 1993 for all guests. 1994 Default is 1 (enabled) if in 64-bit or 32-bit PAE mode. 1995 1996 kvm-arm.vgic_v3_group0_trap= 1997 [KVM,ARM] Trap guest accesses to GICv3 group-0 1998 system registers 1999 2000 kvm-arm.vgic_v3_group1_trap= 2001 [KVM,ARM] Trap guest accesses to GICv3 group-1 2002 system registers 2003 2004 kvm-arm.vgic_v3_common_trap= 2005 [KVM,ARM] Trap guest accesses to GICv3 common 2006 system registers 2007 2008 kvm-arm.vgic_v4_enable= 2009 [KVM,ARM] Allow use of GICv4 for direct injection of 2010 LPIs. 2011 2012 kvm-intel.ept= [KVM,Intel] Disable extended page tables 2013 (virtualized MMU) support on capable Intel chips. 2014 Default is 1 (enabled) 2015 2016 kvm-intel.emulate_invalid_guest_state= 2017 [KVM,Intel] Enable emulation of invalid guest states 2018 Default is 0 (disabled) 2019 2020 kvm-intel.flexpriority= 2021 [KVM,Intel] Disable FlexPriority feature (TPR shadow). 2022 Default is 1 (enabled) 2023 2024 kvm-intel.nested= 2025 [KVM,Intel] Enable VMX nesting (nVMX). 2026 Default is 0 (disabled) 2027 2028 kvm-intel.unrestricted_guest= 2029 [KVM,Intel] Disable unrestricted guest feature 2030 (virtualized real and unpaged mode) on capable 2031 Intel chips. Default is 1 (enabled) 2032 2033 kvm-intel.vmentry_l1d_flush=[KVM,Intel] Mitigation for L1 Terminal Fault 2034 CVE-2018-3620. 2035 2036 Valid arguments: never, cond, always 2037 2038 always: L1D cache flush on every VMENTER. 2039 cond: Flush L1D on VMENTER only when the code between 2040 VMEXIT and VMENTER can leak host memory. 2041 never: Disables the mitigation 2042 2043 Default is cond (do L1 cache flush in specific instances) 2044 2045 kvm-intel.vpid= [KVM,Intel] Disable Virtual Processor Identification 2046 feature (tagged TLBs) on capable Intel chips. 2047 Default is 1 (enabled) 2048 2049 l1tf= [X86] Control mitigation of the L1TF vulnerability on 2050 affected CPUs 2051 2052 The kernel PTE inversion protection is unconditionally 2053 enabled and cannot be disabled. 2054 2055 full 2056 Provides all available mitigations for the 2057 L1TF vulnerability. Disables SMT and 2058 enables all mitigations in the 2059 hypervisors, i.e. unconditional L1D flush. 2060 2061 SMT control and L1D flush control via the 2062 sysfs interface is still possible after 2063 boot. Hypervisors will issue a warning 2064 when the first VM is started in a 2065 potentially insecure configuration, 2066 i.e. SMT enabled or L1D flush disabled. 2067 2068 full,force 2069 Same as 'full', but disables SMT and L1D 2070 flush runtime control. Implies the 2071 'nosmt=force' command line option. 2072 (i.e. sysfs control of SMT is disabled.) 2073 2074 flush 2075 Leaves SMT enabled and enables the default 2076 hypervisor mitigation, i.e. conditional 2077 L1D flush. 2078 2079 SMT control and L1D flush control via the 2080 sysfs interface is still possible after 2081 boot. Hypervisors will issue a warning 2082 when the first VM is started in a 2083 potentially insecure configuration, 2084 i.e. SMT enabled or L1D flush disabled. 2085 2086 flush,nosmt 2087 2088 Disables SMT and enables the default 2089 hypervisor mitigation. 2090 2091 SMT control and L1D flush control via the 2092 sysfs interface is still possible after 2093 boot. Hypervisors will issue a warning 2094 when the first VM is started in a 2095 potentially insecure configuration, 2096 i.e. SMT enabled or L1D flush disabled. 2097 2098 flush,nowarn 2099 Same as 'flush', but hypervisors will not 2100 warn when a VM is started in a potentially 2101 insecure configuration. 2102 2103 off 2104 Disables hypervisor mitigations and doesn't 2105 emit any warnings. 2106 It also drops the swap size and available 2107 RAM limit restriction on both hypervisor and 2108 bare metal. 2109 2110 Default is 'flush'. 2111 2112 For details see: Documentation/admin-guide/l1tf.rst 2113 2114 l2cr= [PPC] 2115 2116 l3cr= [PPC] 2117 2118 lapic [X86-32,APIC] Enable the local APIC even if BIOS 2119 disabled it. 2120 2121 lapic= [x86,APIC] "notscdeadline" Do not use TSC deadline 2122 value for LAPIC timer one-shot implementation. Default 2123 back to the programmable timer unit in the LAPIC. 2124 2125 lapic_timer_c2_ok [X86,APIC] trust the local apic timer 2126 in C2 power state. 2127 2128 libata.dma= [LIBATA] DMA control 2129 libata.dma=0 Disable all PATA and SATA DMA 2130 libata.dma=1 PATA and SATA Disk DMA only 2131 libata.dma=2 ATAPI (CDROM) DMA only 2132 libata.dma=4 Compact Flash DMA only 2133 Combinations also work, so libata.dma=3 enables DMA 2134 for disks and CDROMs, but not CFs. 2135 2136 libata.ignore_hpa= [LIBATA] Ignore HPA limit 2137 libata.ignore_hpa=0 keep BIOS limits (default) 2138 libata.ignore_hpa=1 ignore limits, using full disk 2139 2140 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume 2141 when set. 2142 Format: <int> 2143 2144 libata.force= [LIBATA] Force configurations. The format is comma 2145 separated list of "[ID:]VAL" where ID is 2146 PORT[.DEVICE]. PORT and DEVICE are decimal numbers 2147 matching port, link or device. Basically, it matches 2148 the ATA ID string printed on console by libata. If 2149 the whole ID part is omitted, the last PORT and DEVICE 2150 values are used. If ID hasn't been specified yet, the 2151 configuration applies to all ports, links and devices. 2152 2153 If only DEVICE is omitted, the parameter applies to 2154 the port and all links and devices behind it. DEVICE 2155 number of 0 either selects the first device or the 2156 first fan-out link behind PMP device. It does not 2157 select the host link. DEVICE number of 15 selects the 2158 host link and device attached to it. 2159 2160 The VAL specifies the configuration to force. As long 2161 as there's no ambiguity shortcut notation is allowed. 2162 For example, both 1.5 and 1.5G would work for 1.5Gbps. 2163 The following configurations can be forced. 2164 2165 * Cable type: 40c, 80c, short40c, unk, ign or sata. 2166 Any ID with matching PORT is used. 2167 2168 * SATA link speed limit: 1.5Gbps or 3.0Gbps. 2169 2170 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7]. 2171 udma[/][16,25,33,44,66,100,133] notation is also 2172 allowed. 2173 2174 * [no]ncq: Turn on or off NCQ. 2175 2176 * [no]ncqtrim: Turn off queued DSM TRIM. 2177 2178 * nohrst, nosrst, norst: suppress hard, soft 2179 and both resets. 2180 2181 * rstonce: only attempt one reset during 2182 hot-unplug link recovery 2183 2184 * dump_id: dump IDENTIFY data. 2185 2186 * atapi_dmadir: Enable ATAPI DMADIR bridge support 2187 2188 * disable: Disable this device. 2189 2190 If there are multiple matching configurations changing 2191 the same attribute, the last one is used. 2192 2193 memblock=debug [KNL] Enable memblock debug messages. 2194 2195 load_ramdisk= [RAM] List of ramdisks to load from floppy 2196 See Documentation/blockdev/ramdisk.txt. 2197 2198 lockd.nlm_grace_period=P [NFS] Assign grace period. 2199 Format: <integer> 2200 2201 lockd.nlm_tcpport=N [NFS] Assign TCP port. 2202 Format: <integer> 2203 2204 lockd.nlm_timeout=T [NFS] Assign timeout value. 2205 Format: <integer> 2206 2207 lockd.nlm_udpport=M [NFS] Assign UDP port. 2208 Format: <integer> 2209 2210 locktorture.nreaders_stress= [KNL] 2211 Set the number of locking read-acquisition kthreads. 2212 Defaults to being automatically set based on the 2213 number of online CPUs. 2214 2215 locktorture.nwriters_stress= [KNL] 2216 Set the number of locking write-acquisition kthreads. 2217 2218 locktorture.onoff_holdoff= [KNL] 2219 Set time (s) after boot for CPU-hotplug testing. 2220 2221 locktorture.onoff_interval= [KNL] 2222 Set time (s) between CPU-hotplug operations, or 2223 zero to disable CPU-hotplug testing. 2224 2225 locktorture.shuffle_interval= [KNL] 2226 Set task-shuffle interval (jiffies). Shuffling 2227 tasks allows some CPUs to go into dyntick-idle 2228 mode during the locktorture test. 2229 2230 locktorture.shutdown_secs= [KNL] 2231 Set time (s) after boot system shutdown. This 2232 is useful for hands-off automated testing. 2233 2234 locktorture.stat_interval= [KNL] 2235 Time (s) between statistics printk()s. 2236 2237 locktorture.stutter= [KNL] 2238 Time (s) to stutter testing, for example, 2239 specifying five seconds causes the test to run for 2240 five seconds, wait for five seconds, and so on. 2241 This tests the locking primitive's ability to 2242 transition abruptly to and from idle. 2243 2244 locktorture.torture_type= [KNL] 2245 Specify the locking implementation to test. 2246 2247 locktorture.verbose= [KNL] 2248 Enable additional printk() statements. 2249 2250 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver 2251 Format: <irq> 2252 2253 loglevel= All Kernel Messages with a loglevel smaller than the 2254 console loglevel will be printed to the console. It can 2255 also be changed with klogd or other programs. The 2256 loglevels are defined as follows: 2257 2258 0 (KERN_EMERG) system is unusable 2259 1 (KERN_ALERT) action must be taken immediately 2260 2 (KERN_CRIT) critical conditions 2261 3 (KERN_ERR) error conditions 2262 4 (KERN_WARNING) warning conditions 2263 5 (KERN_NOTICE) normal but significant condition 2264 6 (KERN_INFO) informational 2265 7 (KERN_DEBUG) debug-level messages 2266 2267 log_buf_len=n[KMG] Sets the size of the printk ring buffer, 2268 in bytes. n must be a power of two and greater 2269 than the minimal size. The minimal size is defined 2270 by LOG_BUF_SHIFT kernel config parameter. There is 2271 also CONFIG_LOG_CPU_MAX_BUF_SHIFT config parameter 2272 that allows to increase the default size depending on 2273 the number of CPUs. See init/Kconfig for more details. 2274 2275 logo.nologo [FB] Disables display of the built-in Linux logo. 2276 This may be used to provide more screen space for 2277 kernel log messages and is useful when debugging 2278 kernel boot problems. 2279 2280 lp=0 [LP] Specify parallel ports to use, e.g, 2281 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses 2282 lp=reset first parallel port). 'lp=0' disables the 2283 lp=auto printer driver. 'lp=reset' (which can be 2284 specified in addition to the ports) causes 2285 attached printers to be reset. Using 2286 lp=port1,port2,... specifies the parallel ports 2287 to associate lp devices with, starting with 2288 lp0. A port specification may be 'none' to skip 2289 that lp device, or a parport name such as 2290 'parport0'. Specifying 'lp=auto' instead of a 2291 port specification list means that device IDs 2292 from each port should be examined, to see if 2293 an IEEE 1284-compliant printer is attached; if 2294 so, the driver will manage that printer. 2295 See also header of drivers/char/lp.c. 2296 2297 lpj=n [KNL] 2298 Sets loops_per_jiffy to given constant, thus avoiding 2299 time-consuming boot-time autodetection (up to 250 ms per 2300 CPU). 0 enables autodetection (default). To determine 2301 the correct value for your kernel, boot with normal 2302 autodetection and see what value is printed. Note that 2303 on SMP systems the preset will be applied to all CPUs, 2304 which is likely to cause problems if your CPUs need 2305 significantly divergent settings. An incorrect value 2306 will cause delays in the kernel to be wrong, leading to 2307 unpredictable I/O errors and other breakage. Although 2308 unlikely, in the extreme case this might damage your 2309 hardware. 2310 2311 ltpc= [NET] 2312 Format: <io>,<irq>,<dma> 2313 2314 lsm.debug [SECURITY] Enable LSM initialization debugging output. 2315 2316 machvec= [IA-64] Force the use of a particular machine-vector 2317 (machvec) in a generic kernel. 2318 Example: machvec=hpzx1_swiotlb 2319 2320 machtype= [Loongson] Share the same kernel image file between different 2321 yeeloong laptop. 2322 Example: machtype=lemote-yeeloong-2f-7inch 2323 2324 max_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory greater 2325 than or equal to this physical address is ignored. 2326 2327 maxcpus= [SMP] Maximum number of processors that an SMP kernel 2328 will bring up during bootup. maxcpus=n : n >= 0 limits 2329 the kernel to bring up 'n' processors. Surely after 2330 bootup you can bring up the other plugged cpu by executing 2331 "echo 1 > /sys/devices/system/cpu/cpuX/online". So maxcpus 2332 only takes effect during system bootup. 2333 While n=0 is a special case, it is equivalent to "nosmp", 2334 which also disables the IO APIC. 2335 2336 max_loop= [LOOP] The number of loop block devices that get 2337 (loop.max_loop) unconditionally pre-created at init time. The default 2338 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead 2339 of statically allocating a predefined number, loop 2340 devices can be requested on-demand with the 2341 /dev/loop-control interface. 2342 2343 mce [X86-32] Machine Check Exception 2344 2345 mce=option [X86-64] See Documentation/x86/x86_64/boot-options.txt 2346 2347 md= [HW] RAID subsystems devices and level 2348 See Documentation/admin-guide/md.rst. 2349 2350 mdacon= [MDA] 2351 Format: <first>,<last> 2352 Specifies range of consoles to be captured by the MDA. 2353 2354 mem=nn[KMG] [KNL,BOOT] Force usage of a specific amount of memory 2355 Amount of memory to be used when the kernel is not able 2356 to see the whole system memory or for test. 2357 [X86] Work as limiting max address. Use together 2358 with memmap= to avoid physical address space collisions. 2359 Without memmap= PCI devices could be placed at addresses 2360 belonging to unused RAM. 2361 2362 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel 2363 memory. 2364 2365 memchunk=nn[KMG] 2366 [KNL,SH] Allow user to override the default size for 2367 per-device physically contiguous DMA buffers. 2368 2369 memhp_default_state=online/offline 2370 [KNL] Set the initial state for the memory hotplug 2371 onlining policy. If not specified, the default value is 2372 set according to the 2373 CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE kernel config 2374 option. 2375 See Documentation/memory-hotplug.txt. 2376 2377 memmap=exactmap [KNL,X86] Enable setting of an exact 2378 E820 memory map, as specified by the user. 2379 Such memmap=exactmap lines can be constructed based on 2380 BIOS output or other requirements. See the memmap=nn@ss 2381 option description. 2382 2383 memmap=nn[KMG]@ss[KMG] 2384 [KNL] Force usage of a specific region of memory. 2385 Region of memory to be used is from ss to ss+nn. 2386 If @ss[KMG] is omitted, it is equivalent to mem=nn[KMG], 2387 which limits max address to nn[KMG]. 2388 Multiple different regions can be specified, 2389 comma delimited. 2390 Example: 2391 memmap=100M@2G,100M#3G,1G!1024G 2392 2393 memmap=nn[KMG]#ss[KMG] 2394 [KNL,ACPI] Mark specific memory as ACPI data. 2395 Region of memory to be marked is from ss to ss+nn. 2396 2397 memmap=nn[KMG]$ss[KMG] 2398 [KNL,ACPI] Mark specific memory as reserved. 2399 Region of memory to be reserved is from ss to ss+nn. 2400 Example: Exclude memory from 0x18690000-0x1869ffff 2401 memmap=64K$0x18690000 2402 or 2403 memmap=0x10000$0x18690000 2404 Some bootloaders may need an escape character before '$', 2405 like Grub2, otherwise '$' and the following number 2406 will be eaten. 2407 2408 memmap=nn[KMG]!ss[KMG] 2409 [KNL,X86] Mark specific memory as protected. 2410 Region of memory to be used, from ss to ss+nn. 2411 The memory region may be marked as e820 type 12 (0xc) 2412 and is NVDIMM or ADR memory. 2413 2414 memmap=<size>%<offset>-<oldtype>+<newtype> 2415 [KNL,ACPI] Convert memory within the specified region 2416 from <oldtype> to <newtype>. If "-<oldtype>" is left 2417 out, the whole region will be marked as <newtype>, 2418 even if previously unavailable. If "+<newtype>" is left 2419 out, matching memory will be removed. Types are 2420 specified as e820 types, e.g., 1 = RAM, 2 = reserved, 2421 3 = ACPI, 12 = PRAM. 2422 2423 memory_corruption_check=0/1 [X86] 2424 Some BIOSes seem to corrupt the first 64k of 2425 memory when doing things like suspend/resume. 2426 Setting this option will scan the memory 2427 looking for corruption. Enabling this will 2428 both detect corruption and prevent the kernel 2429 from using the memory being corrupted. 2430 However, its intended as a diagnostic tool; if 2431 repeatable BIOS-originated corruption always 2432 affects the same memory, you can use memmap= 2433 to prevent the kernel from using that memory. 2434 2435 memory_corruption_check_size=size [X86] 2436 By default it checks for corruption in the low 2437 64k, making this memory unavailable for normal 2438 use. Use this parameter to scan for 2439 corruption in more or less memory. 2440 2441 memory_corruption_check_period=seconds [X86] 2442 By default it checks for corruption every 60 2443 seconds. Use this parameter to check at some 2444 other rate. 0 disables periodic checking. 2445 2446 memtest= [KNL,X86,ARM,PPC] Enable memtest 2447 Format: <integer> 2448 default : 0 <disable> 2449 Specifies the number of memtest passes to be 2450 performed. Each pass selects another test 2451 pattern from a given set of patterns. Memtest 2452 fills the memory with this pattern, validates 2453 memory contents and reserves bad memory 2454 regions that are detected. 2455 2456 mem_encrypt= [X86-64] AMD Secure Memory Encryption (SME) control 2457 Valid arguments: on, off 2458 Default (depends on kernel configuration option): 2459 on (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=y) 2460 off (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=n) 2461 mem_encrypt=on: Activate SME 2462 mem_encrypt=off: Do not activate SME 2463 2464 Refer to Documentation/x86/amd-memory-encryption.txt 2465 for details on when memory encryption can be activated. 2466 2467 mem_sleep_default= [SUSPEND] Default system suspend mode: 2468 s2idle - Suspend-To-Idle 2469 shallow - Power-On Suspend or equivalent (if supported) 2470 deep - Suspend-To-RAM or equivalent (if supported) 2471 See Documentation/admin-guide/pm/sleep-states.rst. 2472 2473 meye.*= [HW] Set MotionEye Camera parameters 2474 See Documentation/media/v4l-drivers/meye.rst. 2475 2476 mfgpt_irq= [IA-32] Specify the IRQ to use for the 2477 Multi-Function General Purpose Timers on AMD Geode 2478 platforms. 2479 2480 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when 2481 the BIOS has incorrectly applied a workaround. TinyBIOS 2482 version 0.98 is known to be affected, 0.99 fixes the 2483 problem by letting the user disable the workaround. 2484 2485 mga= [HW,DRM] 2486 2487 min_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory below this 2488 physical address is ignored. 2489 2490 mini2440= [ARM,HW,KNL] 2491 Format:[0..2][b][c][t] 2492 Default: "0tb" 2493 MINI2440 configuration specification: 2494 0 - The attached screen is the 3.5" TFT 2495 1 - The attached screen is the 7" TFT 2496 2 - The VGA Shield is attached (1024x768) 2497 Leaving out the screen size parameter will not load 2498 the TFT driver, and the framebuffer will be left 2499 unconfigured. 2500 b - Enable backlight. The TFT backlight pin will be 2501 linked to the kernel VESA blanking code and a GPIO 2502 LED. This parameter is not necessary when using the 2503 VGA shield. 2504 c - Enable the s3c camera interface. 2505 t - Reserved for enabling touchscreen support. The 2506 touchscreen support is not enabled in the mainstream 2507 kernel as of 2.6.30, a preliminary port can be found 2508 in the "bleeding edge" mini2440 support kernel at 2509 http://repo.or.cz/w/linux-2.6/mini2440.git 2510 2511 mminit_loglevel= 2512 [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this 2513 parameter allows control of the logging verbosity for 2514 the additional memory initialisation checks. A value 2515 of 0 disables mminit logging and a level of 4 will 2516 log everything. Information is printed at KERN_DEBUG 2517 so loglevel=8 may also need to be specified. 2518 2519 module.sig_enforce 2520 [KNL] When CONFIG_MODULE_SIG is set, this means that 2521 modules without (valid) signatures will fail to load. 2522 Note that if CONFIG_MODULE_SIG_FORCE is set, that 2523 is always true, so this option does nothing. 2524 2525 module_blacklist= [KNL] Do not load a comma-separated list of 2526 modules. Useful for debugging problem modules. 2527 2528 mousedev.tap_time= 2529 [MOUSE] Maximum time between finger touching and 2530 leaving touchpad surface for touch to be considered 2531 a tap and be reported as a left button click (for 2532 touchpads working in absolute mode only). 2533 Format: <msecs> 2534 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices 2535 reporting absolute coordinates, such as tablets 2536 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices 2537 reporting absolute coordinates, such as tablets 2538 2539 movablecore= [KNL,X86,IA-64,PPC] 2540 Format: nn[KMGTPE] | nn% 2541 This parameter is the complement to kernelcore=, it 2542 specifies the amount of memory used for migratable 2543 allocations. If both kernelcore and movablecore is 2544 specified, then kernelcore will be at *least* the 2545 specified value but may be more. If movablecore on its 2546 own is specified, the administrator must be careful 2547 that the amount of memory usable for all allocations 2548 is not too small. 2549 2550 movable_node [KNL] Boot-time switch to make hotplugable memory 2551 NUMA nodes to be movable. This means that the memory 2552 of such nodes will be usable only for movable 2553 allocations which rules out almost all kernel 2554 allocations. Use with caution! 2555 2556 MTD_Partition= [MTD] 2557 Format: <name>,<region-number>,<size>,<offset> 2558 2559 MTD_Region= [MTD] Format: 2560 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>] 2561 2562 mtdparts= [MTD] 2563 See drivers/mtd/cmdlinepart.c. 2564 2565 multitce=off [PPC] This parameter disables the use of the pSeries 2566 firmware feature for updating multiple TCE entries 2567 at a time. 2568 2569 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration 2570 2571 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock] 2572 2573 boundary - index of last SLC block on Flex-OneNAND. 2574 The remaining blocks are configured as MLC blocks. 2575 lock - Configure if Flex-OneNAND boundary should be locked. 2576 Once locked, the boundary cannot be changed. 2577 1 indicates lock status, 0 indicates unlock status. 2578 2579 mtdset= [ARM] 2580 ARM/S3C2412 JIVE boot control 2581 2582 See arch/arm/mach-s3c2412/mach-jive.c 2583 2584 mtouchusb.raw_coordinates= 2585 [HW] Make the MicroTouch USB driver use raw coordinates 2586 ('y', default) or cooked coordinates ('n') 2587 2588 mtrr_chunk_size=nn[KMG] [X86] 2589 used for mtrr cleanup. It is largest continuous chunk 2590 that could hold holes aka. UC entries. 2591 2592 mtrr_gran_size=nn[KMG] [X86] 2593 Used for mtrr cleanup. It is granularity of mtrr block. 2594 Default is 1. 2595 Large value could prevent small alignment from 2596 using up MTRRs. 2597 2598 mtrr_spare_reg_nr=n [X86] 2599 Format: <integer> 2600 Range: 0,7 : spare reg number 2601 Default : 1 2602 Used for mtrr cleanup. It is spare mtrr entries number. 2603 Set to 2 or more if your graphical card needs more. 2604 2605 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card 2606 2607 netdev= [NET] Network devices parameters 2608 Format: <irq>,<io>,<mem_start>,<mem_end>,<name> 2609 Note that mem_start is often overloaded to mean 2610 something different and driver-specific. 2611 This usage is only documented in each driver source 2612 file if at all. 2613 2614 nf_conntrack.acct= 2615 [NETFILTER] Enable connection tracking flow accounting 2616 0 to disable accounting 2617 1 to enable accounting 2618 Default value is 0. 2619 2620 nfsaddrs= [NFS] Deprecated. Use ip= instead. 2621 See Documentation/filesystems/nfs/nfsroot.txt. 2622 2623 nfsroot= [NFS] nfs root filesystem for disk-less boxes. 2624 See Documentation/filesystems/nfs/nfsroot.txt. 2625 2626 nfsrootdebug [NFS] enable nfsroot debugging messages. 2627 See Documentation/filesystems/nfs/nfsroot.txt. 2628 2629 nfs.callback_nr_threads= 2630 [NFSv4] set the total number of threads that the 2631 NFS client will assign to service NFSv4 callback 2632 requests. 2633 2634 nfs.callback_tcpport= 2635 [NFS] set the TCP port on which the NFSv4 callback 2636 channel should listen. 2637 2638 nfs.cache_getent= 2639 [NFS] sets the pathname to the program which is used 2640 to update the NFS client cache entries. 2641 2642 nfs.cache_getent_timeout= 2643 [NFS] sets the timeout after which an attempt to 2644 update a cache entry is deemed to have failed. 2645 2646 nfs.idmap_cache_timeout= 2647 [NFS] set the maximum lifetime for idmapper cache 2648 entries. 2649 2650 nfs.enable_ino64= 2651 [NFS] enable 64-bit inode numbers. 2652 If zero, the NFS client will fake up a 32-bit inode 2653 number for the readdir() and stat() syscalls instead 2654 of returning the full 64-bit number. 2655 The default is to return 64-bit inode numbers. 2656 2657 nfs.max_session_cb_slots= 2658 [NFSv4.1] Sets the maximum number of session 2659 slots the client will assign to the callback 2660 channel. This determines the maximum number of 2661 callbacks the client will process in parallel for 2662 a particular server. 2663 2664 nfs.max_session_slots= 2665 [NFSv4.1] Sets the maximum number of session slots 2666 the client will attempt to negotiate with the server. 2667 This limits the number of simultaneous RPC requests 2668 that the client can send to the NFSv4.1 server. 2669 Note that there is little point in setting this 2670 value higher than the max_tcp_slot_table_limit. 2671 2672 nfs.nfs4_disable_idmapping= 2673 [NFSv4] When set to the default of '1', this option 2674 ensures that both the RPC level authentication 2675 scheme and the NFS level operations agree to use 2676 numeric uids/gids if the mount is using the 2677 'sec=sys' security flavour. In effect it is 2678 disabling idmapping, which can make migration from 2679 legacy NFSv2/v3 systems to NFSv4 easier. 2680 Servers that do not support this mode of operation 2681 will be autodetected by the client, and it will fall 2682 back to using the idmapper. 2683 To turn off this behaviour, set the value to '0'. 2684 nfs.nfs4_unique_id= 2685 [NFS4] Specify an additional fixed unique ident- 2686 ification string that NFSv4 clients can insert into 2687 their nfs_client_id4 string. This is typically a 2688 UUID that is generated at system install time. 2689 2690 nfs.send_implementation_id = 2691 [NFSv4.1] Send client implementation identification 2692 information in exchange_id requests. 2693 If zero, no implementation identification information 2694 will be sent. 2695 The default is to send the implementation identification 2696 information. 2697 2698 nfs.recover_lost_locks = 2699 [NFSv4] Attempt to recover locks that were lost due 2700 to a lease timeout on the server. Please note that 2701 doing this risks data corruption, since there are 2702 no guarantees that the file will remain unchanged 2703 after the locks are lost. 2704 If you want to enable the kernel legacy behaviour of 2705 attempting to recover these locks, then set this 2706 parameter to '1'. 2707 The default parameter value of '0' causes the kernel 2708 not to attempt recovery of lost locks. 2709 2710 nfs4.layoutstats_timer = 2711 [NFSv4.2] Change the rate at which the kernel sends 2712 layoutstats to the pNFS metadata server. 2713 2714 Setting this to value to 0 causes the kernel to use 2715 whatever value is the default set by the layout 2716 driver. A non-zero value sets the minimum interval 2717 in seconds between layoutstats transmissions. 2718 2719 nfsd.nfs4_disable_idmapping= 2720 [NFSv4] When set to the default of '1', the NFSv4 2721 server will return only numeric uids and gids to 2722 clients using auth_sys, and will accept numeric uids 2723 and gids from such clients. This is intended to ease 2724 migration from NFSv2/v3. 2725 2726 nmi_debug= [KNL,SH] Specify one or more actions to take 2727 when a NMI is triggered. 2728 Format: [state][,regs][,debounce][,die] 2729 2730 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels 2731 Format: [panic,][nopanic,][num] 2732 Valid num: 0 or 1 2733 0 - turn hardlockup detector in nmi_watchdog off 2734 1 - turn hardlockup detector in nmi_watchdog on 2735 When panic is specified, panic when an NMI watchdog 2736 timeout occurs (or 'nopanic' to override the opposite 2737 default). To disable both hard and soft lockup detectors, 2738 please see 'nowatchdog'. 2739 This is useful when you use a panic=... timeout and 2740 need the box quickly up again. 2741 2742 These settings can be accessed at runtime via 2743 the nmi_watchdog and hardlockup_panic sysctls. 2744 2745 netpoll.carrier_timeout= 2746 [NET] Specifies amount of time (in seconds) that 2747 netpoll should wait for a carrier. By default netpoll 2748 waits 4 seconds. 2749 2750 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths 2751 emulation library even if a 387 maths coprocessor 2752 is present. 2753 2754 no5lvl [X86-64] Disable 5-level paging mode. Forces 2755 kernel to use 4-level paging instead. 2756 2757 no_console_suspend 2758 [HW] Never suspend the console 2759 Disable suspending of consoles during suspend and 2760 hibernate operations. Once disabled, debugging 2761 messages can reach various consoles while the rest 2762 of the system is being put to sleep (ie, while 2763 debugging driver suspend/resume hooks). This may 2764 not work reliably with all consoles, but is known 2765 to work with serial and VGA consoles. 2766 To facilitate more flexible debugging, we also add 2767 console_suspend, a printk module parameter to control 2768 it. Users could use console_suspend (usually 2769 /sys/module/printk/parameters/console_suspend) to 2770 turn on/off it dynamically. 2771 2772 noaliencache [MM, NUMA, SLAB] Disables the allocation of alien 2773 caches in the slab allocator. Saves per-node memory, 2774 but will impact performance. 2775 2776 noalign [KNL,ARM] 2777 2778 noaltinstr [S390] Disables alternative instructions patching 2779 (CPU alternatives feature). 2780 2781 noapic [SMP,APIC] Tells the kernel to not make use of any 2782 IOAPICs that may be present in the system. 2783 2784 noautogroup Disable scheduler automatic task group creation. 2785 2786 nobats [PPC] Do not use BATs for mapping kernel lowmem 2787 on "Classic" PPC cores. 2788 2789 nocache [ARM] 2790 2791 noclflush [BUGS=X86] Don't use the CLFLUSH instruction 2792 2793 nodelayacct [KNL] Disable per-task delay accounting 2794 2795 nodsp [SH] Disable hardware DSP at boot time. 2796 2797 noefi Disable EFI runtime services support. 2798 2799 noexec [IA-64] 2800 2801 noexec [X86] 2802 On X86-32 available only on PAE configured kernels. 2803 noexec=on: enable non-executable mappings (default) 2804 noexec=off: disable non-executable mappings 2805 2806 nosmap [X86] 2807 Disable SMAP (Supervisor Mode Access Prevention) 2808 even if it is supported by processor. 2809 2810 nosmep [X86] 2811 Disable SMEP (Supervisor Mode Execution Prevention) 2812 even if it is supported by processor. 2813 2814 noexec32 [X86-64] 2815 This affects only 32-bit executables. 2816 noexec32=on: enable non-executable mappings (default) 2817 read doesn't imply executable mappings 2818 noexec32=off: disable non-executable mappings 2819 read implies executable mappings 2820 2821 nofpu [MIPS,SH] Disable hardware FPU at boot time. 2822 2823 nofxsr [BUGS=X86-32] Disables x86 floating point extended 2824 register save and restore. The kernel will only save 2825 legacy floating-point registers on task switch. 2826 2827 nohugeiomap [KNL,x86] Disable kernel huge I/O mappings. 2828 2829 nosmt [KNL,S390] Disable symmetric multithreading (SMT). 2830 Equivalent to smt=1. 2831 2832 [KNL,x86] Disable symmetric multithreading (SMT). 2833 nosmt=force: Force disable SMT, cannot be undone 2834 via the sysfs control file. 2835 2836 nospectre_v1 [PPC] Disable mitigations for Spectre Variant 1 (bounds 2837 check bypass). With this option data leaks are possible 2838 in the system. 2839 2840 nospectre_v2 [X86,PPC_FSL_BOOK3E] Disable all mitigations for the Spectre variant 2 2841 (indirect branch prediction) vulnerability. System may 2842 allow data leaks with this option, which is equivalent 2843 to spectre_v2=off. 2844 2845 nospec_store_bypass_disable 2846 [HW] Disable all mitigations for the Speculative Store Bypass vulnerability 2847 2848 noxsave [BUGS=X86] Disables x86 extended register state save 2849 and restore using xsave. The kernel will fallback to 2850 enabling legacy floating-point and sse state. 2851 2852 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended 2853 register states. The kernel will fall back to use 2854 xsave to save the states. By using this parameter, 2855 performance of saving the states is degraded because 2856 xsave doesn't support modified optimization while 2857 xsaveopt supports it on xsaveopt enabled systems. 2858 2859 noxsaves [X86] Disables xsaves and xrstors used in saving and 2860 restoring x86 extended register state in compacted 2861 form of xsave area. The kernel will fall back to use 2862 xsaveopt and xrstor to save and restore the states 2863 in standard form of xsave area. By using this 2864 parameter, xsave area per process might occupy more 2865 memory on xsaves enabled systems. 2866 2867 nohlt [BUGS=ARM,SH] Tells the kernel that the sleep(SH) or 2868 wfi(ARM) instruction doesn't work correctly and not to 2869 use it. This is also useful when using JTAG debugger. 2870 2871 no_file_caps Tells the kernel not to honor file capabilities. The 2872 only way then for a file to be executed with privilege 2873 is to be setuid root or executed by root. 2874 2875 nohalt [IA-64] Tells the kernel not to use the power saving 2876 function PAL_HALT_LIGHT when idle. This increases 2877 power-consumption. On the positive side, it reduces 2878 interrupt wake-up latency, which may improve performance 2879 in certain environments such as networked servers or 2880 real-time systems. 2881 2882 nohibernate [HIBERNATION] Disable hibernation and resume. 2883 2884 nohz= [KNL] Boottime enable/disable dynamic ticks 2885 Valid arguments: on, off 2886 Default: on 2887 2888 nohz_full= [KNL,BOOT,SMP,ISOL] 2889 The argument is a cpu list, as described above. 2890 In kernels built with CONFIG_NO_HZ_FULL=y, set 2891 the specified list of CPUs whose tick will be stopped 2892 whenever possible. The boot CPU will be forced outside 2893 the range to maintain the timekeeping. Any CPUs 2894 in this list will have their RCU callbacks offloaded, 2895 just as if they had also been called out in the 2896 rcu_nocbs= boot parameter. 2897 2898 noiotrap [SH] Disables trapped I/O port accesses. 2899 2900 noirqdebug [X86-32] Disables the code which attempts to detect and 2901 disable unhandled interrupt sources. 2902 2903 no_timer_check [X86,APIC] Disables the code which tests for 2904 broken timer IRQ sources. 2905 2906 noisapnp [ISAPNP] Disables ISA PnP code. 2907 2908 noinitrd [RAM] Tells the kernel not to load any configured 2909 initial RAM disk. 2910 2911 nointremap [X86-64, Intel-IOMMU] Do not enable interrupt 2912 remapping. 2913 [Deprecated - use intremap=off] 2914 2915 nointroute [IA-64] 2916 2917 noinvpcid [X86] Disable the INVPCID cpu feature. 2918 2919 nojitter [IA-64] Disables jitter checking for ITC timers. 2920 2921 no-kvmclock [X86,KVM] Disable paravirtualized KVM clock driver 2922 2923 no-kvmapf [X86,KVM] Disable paravirtualized asynchronous page 2924 fault handling. 2925 2926 no-vmw-sched-clock 2927 [X86,PV_OPS] Disable paravirtualized VMware scheduler 2928 clock and use the default one. 2929 2930 no-steal-acc [X86,KVM] Disable paravirtualized steal time accounting. 2931 steal time is computed, but won't influence scheduler 2932 behaviour 2933 2934 nolapic [X86-32,APIC] Do not enable or use the local APIC. 2935 2936 nolapic_timer [X86-32,APIC] Do not use the local APIC timer. 2937 2938 noltlbs [PPC] Do not use large page/tlb entries for kernel 2939 lowmem mapping on PPC40x and PPC8xx 2940 2941 nomca [IA-64] Disable machine check abort handling 2942 2943 nomce [X86-32] Disable Machine Check Exception 2944 2945 nomfgpt [X86-32] Disable Multi-Function General Purpose 2946 Timer usage (for AMD Geode machines). 2947 2948 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to 2949 shutdown the other cpus. Instead use the REBOOT_VECTOR 2950 irq. 2951 2952 nomodule Disable module load 2953 2954 nopat [X86] Disable PAT (page attribute table extension of 2955 pagetables) support. 2956 2957 nopcid [X86-64] Disable the PCID cpu feature. 2958 2959 norandmaps Don't use address space randomization. Equivalent to 2960 echo 0 > /proc/sys/kernel/randomize_va_space 2961 2962 noreplace-smp [X86-32,SMP] Don't replace SMP instructions 2963 with UP alternatives 2964 2965 nordrand [X86] Disable kernel use of the RDRAND and 2966 RDSEED instructions even if they are supported 2967 by the processor. RDRAND and RDSEED are still 2968 available to user space applications. 2969 2970 noresume [SWSUSP] Disables resume and restores original swap 2971 space. 2972 2973 no-scroll [VGA] Disables scrollback. 2974 This is required for the Braillex ib80-piezo Braille 2975 reader made by F.H. Papenmeier (Germany). 2976 2977 nosbagart [IA-64] 2978 2979 nosep [BUGS=X86-32] Disables x86 SYSENTER/SYSEXIT support. 2980 2981 nosmp [SMP] Tells an SMP kernel to act as a UP kernel, 2982 and disable the IO APIC. legacy for "maxcpus=0". 2983 2984 nosoftlockup [KNL] Disable the soft-lockup detector. 2985 2986 nosync [HW,M68K] Disables sync negotiation for all devices. 2987 2988 nowatchdog [KNL] Disable both lockup detectors, i.e. 2989 soft-lockup and NMI watchdog (hard-lockup). 2990 2991 nowb [ARM] 2992 2993 nox2apic [X86-64,APIC] Do not enable x2APIC mode. 2994 2995 cpu0_hotplug [X86] Turn on CPU0 hotplug feature when 2996 CONFIG_BOOTPARAM_HOTPLUG_CPU0 is off. 2997 Some features depend on CPU0. Known dependencies are: 2998 1. Resume from suspend/hibernate depends on CPU0. 2999 Suspend/hibernate will fail if CPU0 is offline and you 3000 need to online CPU0 before suspend/hibernate. 3001 2. PIC interrupts also depend on CPU0. CPU0 can't be 3002 removed if a PIC interrupt is detected. 3003 It's said poweroff/reboot may depend on CPU0 on some 3004 machines although I haven't seen such issues so far 3005 after CPU0 is offline on a few tested machines. 3006 If the dependencies are under your control, you can 3007 turn on cpu0_hotplug. 3008 3009 nps_mtm_hs_ctr= [KNL,ARC] 3010 This parameter sets the maximum duration, in 3011 cycles, each HW thread of the CTOP can run 3012 without interruptions, before HW switches it. 3013 The actual maximum duration is 16 times this 3014 parameter's value. 3015 Format: integer between 1 and 255 3016 Default: 255 3017 3018 nptcg= [IA-64] Override max number of concurrent global TLB 3019 purges which is reported from either PAL_VM_SUMMARY or 3020 SAL PALO. 3021 3022 nr_cpus= [SMP] Maximum number of processors that an SMP kernel 3023 could support. nr_cpus=n : n >= 1 limits the kernel to 3024 support 'n' processors. It could be larger than the 3025 number of already plugged CPU during bootup, later in 3026 runtime you can physically add extra cpu until it reaches 3027 n. So during boot up some boot time memory for per-cpu 3028 variables need be pre-allocated for later physical cpu 3029 hot plugging. 3030 3031 nr_uarts= [SERIAL] maximum number of UARTs to be registered. 3032 3033 numa_balancing= [KNL,X86] Enable or disable automatic NUMA balancing. 3034 Allowed values are enable and disable 3035 3036 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA. 3037 'node', 'default' can be specified 3038 This can be set from sysctl after boot. 3039 See Documentation/sysctl/vm.txt for details. 3040 3041 ohci1394_dma=early [HW] enable debugging via the ohci1394 driver. 3042 See Documentation/debugging-via-ohci1394.txt for more 3043 info. 3044 3045 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands 3046 Rather than timing out after 20 ms if an EC 3047 command is not properly ACKed, override the length 3048 of the timeout. We have interrupts disabled while 3049 waiting for the ACK, so if this is set too high 3050 interrupts *may* be lost! 3051 3052 omap_mux= [OMAP] Override bootloader pin multiplexing. 3053 Format: <mux_mode0.mode_name=value>... 3054 For example, to override I2C bus2: 3055 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100 3056 3057 oprofile.timer= [HW] 3058 Use timer interrupt instead of performance counters 3059 3060 oprofile.cpu_type= Force an oprofile cpu type 3061 This might be useful if you have an older oprofile 3062 userland or if you want common events. 3063 Format: { arch_perfmon } 3064 arch_perfmon: [X86] Force use of architectural 3065 perfmon on Intel CPUs instead of the 3066 CPU specific event set. 3067 timer: [X86] Force use of architectural NMI 3068 timer mode (see also oprofile.timer 3069 for generic hr timer mode) 3070 3071 oops=panic Always panic on oopses. Default is to just kill the 3072 process, but there is a small probability of 3073 deadlocking the machine. 3074 This will also cause panics on machine check exceptions. 3075 Useful together with panic=30 to trigger a reboot. 3076 3077 page_owner= [KNL] Boot-time page_owner enabling option. 3078 Storage of the information about who allocated 3079 each page is disabled in default. With this switch, 3080 we can turn it on. 3081 on: enable the feature 3082 3083 page_poison= [KNL] Boot-time parameter changing the state of 3084 poisoning on the buddy allocator, available with 3085 CONFIG_PAGE_POISONING=y. 3086 off: turn off poisoning (default) 3087 on: turn on poisoning 3088 3089 panic= [KNL] Kernel behaviour on panic: delay <timeout> 3090 timeout > 0: seconds before rebooting 3091 timeout = 0: wait forever 3092 timeout < 0: reboot immediately 3093 Format: <timeout> 3094 3095 panic_on_warn panic() instead of WARN(). Useful to cause kdump 3096 on a WARN(). 3097 3098 crash_kexec_post_notifiers 3099 Run kdump after running panic-notifiers and dumping 3100 kmsg. This only for the users who doubt kdump always 3101 succeeds in any situation. 3102 Note that this also increases risks of kdump failure, 3103 because some panic notifiers can make the crashed 3104 kernel more unstable. 3105 3106 parkbd.port= [HW] Parallel port number the keyboard adapter is 3107 connected to, default is 0. 3108 Format: <parport#> 3109 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation, 3110 0 for XT, 1 for AT (default is AT). 3111 Format: <mode> 3112 3113 parport= [HW,PPT] Specify parallel ports. 0 disables. 3114 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] } 3115 Use 'auto' to force the driver to use any 3116 IRQ/DMA settings detected (the default is to 3117 ignore detected IRQ/DMA settings because of 3118 possible conflicts). You can specify the base 3119 address, IRQ, and DMA settings; IRQ and DMA 3120 should be numbers, or 'auto' (for using detected 3121 settings on that particular port), or 'nofifo' 3122 (to avoid using a FIFO even if it is detected). 3123 Parallel ports are assigned in the order they 3124 are specified on the command line, starting 3125 with parport0. 3126 3127 parport_init_mode= [HW,PPT] 3128 Configure VIA parallel port to operate in 3129 a specific mode. This is necessary on Pegasos 3130 computer where firmware has no options for setting 3131 up parallel port mode and sets it to spp. 3132 Currently this function knows 686a and 8231 chips. 3133 Format: [spp|ps2|epp|ecp|ecpepp] 3134 3135 pause_on_oops= 3136 Halt all CPUs after the first oops has been printed for 3137 the specified number of seconds. This is to be used if 3138 your oopses keep scrolling off the screen. 3139 3140 pcbit= [HW,ISDN] 3141 3142 pcd. [PARIDE] 3143 See header of drivers/block/paride/pcd.c. 3144 See also Documentation/blockdev/paride.txt. 3145 3146 pci=option[,option...] [PCI] various PCI subsystem options. 3147 3148 Some options herein operate on a specific device 3149 or a set of devices (<pci_dev>). These are 3150 specified in one of the following formats: 3151 3152 [<domain>:]<bus>:<dev>.<func>[/<dev>.<func>]* 3153 pci:<vendor>:<device>[:<subvendor>:<subdevice>] 3154 3155 Note: the first format specifies a PCI 3156 bus/device/function address which may change 3157 if new hardware is inserted, if motherboard 3158 firmware changes, or due to changes caused 3159 by other kernel parameters. If the 3160 domain is left unspecified, it is 3161 taken to be zero. Optionally, a path 3162 to a device through multiple device/function 3163 addresses can be specified after the base 3164 address (this is more robust against 3165 renumbering issues). The second format 3166 selects devices using IDs from the 3167 configuration space which may match multiple 3168 devices in the system. 3169 3170 earlydump dump PCI config space before the kernel 3171 changes anything 3172 off [X86] don't probe for the PCI bus 3173 bios [X86-32] force use of PCI BIOS, don't access 3174 the hardware directly. Use this if your machine 3175 has a non-standard PCI host bridge. 3176 nobios [X86-32] disallow use of PCI BIOS, only direct 3177 hardware access methods are allowed. Use this 3178 if you experience crashes upon bootup and you 3179 suspect they are caused by the BIOS. 3180 conf1 [X86] Force use of PCI Configuration Access 3181 Mechanism 1 (config address in IO port 0xCF8, 3182 data in IO port 0xCFC, both 32-bit). 3183 conf2 [X86] Force use of PCI Configuration Access 3184 Mechanism 2 (IO port 0xCF8 is an 8-bit port for 3185 the function, IO port 0xCFA, also 8-bit, sets 3186 bus number. The config space is then accessed 3187 through ports 0xC000-0xCFFF). 3188 See http://wiki.osdev.org/PCI for more info 3189 on the configuration access mechanisms. 3190 noaer [PCIE] If the PCIEAER kernel config parameter is 3191 enabled, this kernel boot option can be used to 3192 disable the use of PCIE advanced error reporting. 3193 nodomains [PCI] Disable support for multiple PCI 3194 root domains (aka PCI segments, in ACPI-speak). 3195 nommconf [X86] Disable use of MMCONFIG for PCI 3196 Configuration 3197 check_enable_amd_mmconf [X86] check for and enable 3198 properly configured MMIO access to PCI 3199 config space on AMD family 10h CPU 3200 nomsi [MSI] If the PCI_MSI kernel config parameter is 3201 enabled, this kernel boot option can be used to 3202 disable the use of MSI interrupts system-wide. 3203 noioapicquirk [APIC] Disable all boot interrupt quirks. 3204 Safety option to keep boot IRQs enabled. This 3205 should never be necessary. 3206 ioapicreroute [APIC] Enable rerouting of boot IRQs to the 3207 primary IO-APIC for bridges that cannot disable 3208 boot IRQs. This fixes a source of spurious IRQs 3209 when the system masks IRQs. 3210 noioapicreroute [APIC] Disable workaround that uses the 3211 boot IRQ equivalent of an IRQ that connects to 3212 a chipset where boot IRQs cannot be disabled. 3213 The opposite of ioapicreroute. 3214 biosirq [X86-32] Use PCI BIOS calls to get the interrupt 3215 routing table. These calls are known to be buggy 3216 on several machines and they hang the machine 3217 when used, but on other computers it's the only 3218 way to get the interrupt routing table. Try 3219 this option if the kernel is unable to allocate 3220 IRQs or discover secondary PCI buses on your 3221 motherboard. 3222 rom [X86] Assign address space to expansion ROMs. 3223 Use with caution as certain devices share 3224 address decoders between ROMs and other 3225 resources. 3226 norom [X86] Do not assign address space to 3227 expansion ROMs that do not already have 3228 BIOS assigned address ranges. 3229 nobar [X86] Do not assign address space to the 3230 BARs that weren't assigned by the BIOS. 3231 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be 3232 assigned automatically to PCI devices. You can 3233 make the kernel exclude IRQs of your ISA cards 3234 this way. 3235 pirqaddr=0xAAAAA [X86] Specify the physical address 3236 of the PIRQ table (normally generated 3237 by the BIOS) if it is outside the 3238 F0000h-100000h range. 3239 lastbus=N [X86] Scan all buses thru bus #N. Can be 3240 useful if the kernel is unable to find your 3241 secondary buses and you want to tell it 3242 explicitly which ones they are. 3243 assign-busses [X86] Always assign all PCI bus 3244 numbers ourselves, overriding 3245 whatever the firmware may have done. 3246 usepirqmask [X86] Honor the possible IRQ mask stored 3247 in the BIOS $PIR table. This is needed on 3248 some systems with broken BIOSes, notably 3249 some HP Pavilion N5400 and Omnibook XE3 3250 notebooks. This will have no effect if ACPI 3251 IRQ routing is enabled. 3252 noacpi [X86] Do not use ACPI for IRQ routing 3253 or for PCI scanning. 3254 use_crs [X86] Use PCI host bridge window information 3255 from ACPI. On BIOSes from 2008 or later, this 3256 is enabled by default. If you need to use this, 3257 please report a bug. 3258 nocrs [X86] Ignore PCI host bridge windows from ACPI. 3259 If you need to use this, please report a bug. 3260 routeirq Do IRQ routing for all PCI devices. 3261 This is normally done in pci_enable_device(), 3262 so this option is a temporary workaround 3263 for broken drivers that don't call it. 3264 skip_isa_align [X86] do not align io start addr, so can 3265 handle more pci cards 3266 noearly [X86] Don't do any early type 1 scanning. 3267 This might help on some broken boards which 3268 machine check when some devices' config space 3269 is read. But various workarounds are disabled 3270 and some IOMMU drivers will not work. 3271 bfsort Sort PCI devices into breadth-first order. 3272 This sorting is done to get a device 3273 order compatible with older (<= 2.4) kernels. 3274 nobfsort Don't sort PCI devices into breadth-first order. 3275 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size) 3276 tuning and use the BIOS-configured MPS defaults. 3277 pcie_bus_safe Set every device's MPS to the largest value 3278 supported by all devices below the root complex. 3279 pcie_bus_perf Set device MPS to the largest allowable MPS 3280 based on its parent bus. Also set MRRS (Max 3281 Read Request Size) to the largest supported 3282 value (no larger than the MPS that the device 3283 or bus can support) for best performance. 3284 pcie_bus_peer2peer Set every device's MPS to 128B, which 3285 every device is guaranteed to support. This 3286 configuration allows peer-to-peer DMA between 3287 any pair of devices, possibly at the cost of 3288 reduced performance. This also guarantees 3289 that hot-added devices will work. 3290 cbiosize=nn[KMG] The fixed amount of bus space which is 3291 reserved for the CardBus bridge's IO window. 3292 The default value is 256 bytes. 3293 cbmemsize=nn[KMG] The fixed amount of bus space which is 3294 reserved for the CardBus bridge's memory 3295 window. The default value is 64 megabytes. 3296 resource_alignment= 3297 Format: 3298 [<order of align>@]<pci_dev>[; ...] 3299 Specifies alignment and device to reassign 3300 aligned memory resources. How to 3301 specify the device is described above. 3302 If <order of align> is not specified, 3303 PAGE_SIZE is used as alignment. 3304 PCI-PCI bridge can be specified, if resource 3305 windows need to be expanded. 3306 To specify the alignment for several 3307 instances of a device, the PCI vendor, 3308 device, subvendor, and subdevice may be 3309 specified, e.g., 4096@pci:8086:9c22:103c:198f 3310 ecrc= Enable/disable PCIe ECRC (transaction layer 3311 end-to-end CRC checking). 3312 bios: Use BIOS/firmware settings. This is the 3313 the default. 3314 off: Turn ECRC off 3315 on: Turn ECRC on. 3316 hpiosize=nn[KMG] The fixed amount of bus space which is 3317 reserved for hotplug bridge's IO window. 3318 Default size is 256 bytes. 3319 hpmemsize=nn[KMG] The fixed amount of bus space which is 3320 reserved for hotplug bridge's memory window. 3321 Default size is 2 megabytes. 3322 hpbussize=nn The minimum amount of additional bus numbers 3323 reserved for buses below a hotplug bridge. 3324 Default is 1. 3325 realloc= Enable/disable reallocating PCI bridge resources 3326 if allocations done by BIOS are too small to 3327 accommodate resources required by all child 3328 devices. 3329 off: Turn realloc off 3330 on: Turn realloc on 3331 realloc same as realloc=on 3332 noari do not use PCIe ARI. 3333 noats [PCIE, Intel-IOMMU, AMD-IOMMU] 3334 do not use PCIe ATS (and IOMMU device IOTLB). 3335 pcie_scan_all Scan all possible PCIe devices. Otherwise we 3336 only look for one device below a PCIe downstream 3337 port. 3338 big_root_window Try to add a big 64bit memory window to the PCIe 3339 root complex on AMD CPUs. Some GFX hardware 3340 can resize a BAR to allow access to all VRAM. 3341 Adding the window is slightly risky (it may 3342 conflict with unreported devices), so this 3343 taints the kernel. 3344 disable_acs_redir=<pci_dev>[; ...] 3345 Specify one or more PCI devices (in the format 3346 specified above) separated by semicolons. 3347 Each device specified will have the PCI ACS 3348 redirect capabilities forced off which will 3349 allow P2P traffic between devices through 3350 bridges without forcing it upstream. Note: 3351 this removes isolation between devices and 3352 may put more devices in an IOMMU group. 3353 3354 pcie_aspm= [PCIE] Forcibly enable or disable PCIe Active State Power 3355 Management. 3356 off Disable ASPM. 3357 force Enable ASPM even on devices that claim not to support it. 3358 WARNING: Forcing ASPM on may cause system lockups. 3359 3360 pcie_ports= [PCIE] PCIe port services handling: 3361 native Use native PCIe services (PME, AER, DPC, PCIe hotplug) 3362 even if the platform doesn't give the OS permission to 3363 use them. This may cause conflicts if the platform 3364 also tries to use these services. 3365 compat Disable native PCIe services (PME, AER, DPC, PCIe 3366 hotplug). 3367 3368 pcie_port_pm= [PCIE] PCIe port power management handling: 3369 off Disable power management of all PCIe ports 3370 force Forcibly enable power management of all PCIe ports 3371 3372 pcie_pme= [PCIE,PM] Native PCIe PME signaling options: 3373 nomsi Do not use MSI for native PCIe PME signaling (this makes 3374 all PCIe root ports use INTx for all services). 3375 3376 pcmv= [HW,PCMCIA] BadgePAD 4 3377 3378 pd_ignore_unused 3379 [PM] 3380 Keep all power-domains already enabled by bootloader on, 3381 even if no driver has claimed them. This is useful 3382 for debug and development, but should not be 3383 needed on a platform with proper driver support. 3384 3385 pd. [PARIDE] 3386 See Documentation/blockdev/paride.txt. 3387 3388 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at 3389 boot time. 3390 Format: { 0 | 1 } 3391 See arch/parisc/kernel/pdc_chassis.c 3392 3393 percpu_alloc= Select which percpu first chunk allocator to use. 3394 Currently supported values are "embed" and "page". 3395 Archs may support subset or none of the selections. 3396 See comments in mm/percpu.c for details on each 3397 allocator. This parameter is primarily for debugging 3398 and performance comparison. 3399 3400 pf. [PARIDE] 3401 See Documentation/blockdev/paride.txt. 3402 3403 pg. [PARIDE] 3404 See Documentation/blockdev/paride.txt. 3405 3406 pirq= [SMP,APIC] Manual mp-table setup 3407 See Documentation/x86/i386/IO-APIC.txt. 3408 3409 plip= [PPT,NET] Parallel port network link 3410 Format: { parport<nr> | timid | 0 } 3411 See also Documentation/admin-guide/parport.rst. 3412 3413 pmtmr= [X86] Manual setup of pmtmr I/O Port. 3414 Override pmtimer IOPort with a hex value. 3415 e.g. pmtmr=0x508 3416 3417 pnp.debug=1 [PNP] 3418 Enable PNP debug messages (depends on the 3419 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time 3420 via /sys/module/pnp/parameters/debug. We always show 3421 current resource usage; turning this on also shows 3422 possible settings and some assignment information. 3423 3424 pnpacpi= [ACPI] 3425 { off } 3426 3427 pnpbios= [ISAPNP] 3428 { on | off | curr | res | no-curr | no-res } 3429 3430 pnp_reserve_irq= 3431 [ISAPNP] Exclude IRQs for the autoconfiguration 3432 3433 pnp_reserve_dma= 3434 [ISAPNP] Exclude DMAs for the autoconfiguration 3435 3436 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration 3437 Ranges are in pairs (I/O port base and size). 3438 3439 pnp_reserve_mem= 3440 [ISAPNP] Exclude memory regions for the 3441 autoconfiguration. 3442 Ranges are in pairs (memory base and size). 3443 3444 ports= [IP_VS_FTP] IPVS ftp helper module 3445 Default is 21. 3446 Up to 8 (IP_VS_APP_MAX_PORTS) ports 3447 may be specified. 3448 Format: <port>,<port>.... 3449 3450 powersave=off [PPC] This option disables power saving features. 3451 It specifically disables cpuidle and sets the 3452 platform machine description specific power_save 3453 function to NULL. On Idle the CPU just reduces 3454 execution priority. 3455 3456 ppc_strict_facility_enable 3457 [PPC] This option catches any kernel floating point, 3458 Altivec, VSX and SPE outside of regions specifically 3459 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()). 3460 There is some performance impact when enabling this. 3461 3462 ppc_tm= [PPC] 3463 Format: {"off"} 3464 Disable Hardware Transactional Memory 3465 3466 print-fatal-signals= 3467 [KNL] debug: print fatal signals 3468 3469 If enabled, warn about various signal handling 3470 related application anomalies: too many signals, 3471 too many POSIX.1 timers, fatal signals causing a 3472 coredump - etc. 3473 3474 If you hit the warning due to signal overflow, 3475 you might want to try "ulimit -i unlimited". 3476 3477 default: off. 3478 3479 printk.always_kmsg_dump= 3480 Trigger kmsg_dump for cases other than kernel oops or 3481 panics 3482 Format: <bool> (1/Y/y=enable, 0/N/n=disable) 3483 default: disabled 3484 3485 printk.devkmsg={on,off,ratelimit} 3486 Control writing to /dev/kmsg. 3487 on - unlimited logging to /dev/kmsg from userspace 3488 off - logging to /dev/kmsg disabled 3489 ratelimit - ratelimit the logging 3490 Default: ratelimit 3491 3492 printk.time= Show timing data prefixed to each printk message line 3493 Format: <bool> (1/Y/y=enable, 0/N/n=disable) 3494 3495 processor.max_cstate= [HW,ACPI] 3496 Limit processor to maximum C-state 3497 max_cstate=9 overrides any DMI blacklist limit. 3498 3499 processor.nocst [HW,ACPI] 3500 Ignore the _CST method to determine C-states, 3501 instead using the legacy FADT method 3502 3503 profile= [KNL] Enable kernel profiling via /proc/profile 3504 Format: [<profiletype>,]<number> 3505 Param: <profiletype>: "schedule", "sleep", or "kvm" 3506 [defaults to kernel profiling] 3507 Param: "schedule" - profile schedule points. 3508 Param: "sleep" - profile D-state sleeping (millisecs). 3509 Requires CONFIG_SCHEDSTATS 3510 Param: "kvm" - profile VM exits. 3511 Param: <number> - step/bucket size as a power of 2 for 3512 statistical time based profiling. 3513 3514 prompt_ramdisk= [RAM] List of RAM disks to prompt for floppy disk 3515 before loading. 3516 See Documentation/blockdev/ramdisk.txt. 3517 3518 psi= [KNL] Enable or disable pressure stall information 3519 tracking. 3520 Format: <bool> 3521 3522 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to 3523 probe for; one of (bare|imps|exps|lifebook|any). 3524 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports 3525 per second. 3526 psmouse.resetafter= [HW,MOUSE] 3527 Try to reset the device after so many bad packets 3528 (0 = never). 3529 psmouse.resolution= 3530 [HW,MOUSE] Set desired mouse resolution, in dpi. 3531 psmouse.smartscroll= 3532 [HW,MOUSE] Controls Logitech smartscroll autorepeat. 3533 0 = disabled, 1 = enabled (default). 3534 3535 pstore.backend= Specify the name of the pstore backend to use 3536 3537 pt. [PARIDE] 3538 See Documentation/blockdev/paride.txt. 3539 3540 pti= [X86_64] Control Page Table Isolation of user and 3541 kernel address spaces. Disabling this feature 3542 removes hardening, but improves performance of 3543 system calls and interrupts. 3544 3545 on - unconditionally enable 3546 off - unconditionally disable 3547 auto - kernel detects whether your CPU model is 3548 vulnerable to issues that PTI mitigates 3549 3550 Not specifying this option is equivalent to pti=auto. 3551 3552 nopti [X86_64] 3553 Equivalent to pti=off 3554 3555 pty.legacy_count= 3556 [KNL] Number of legacy pty's. Overwrites compiled-in 3557 default number. 3558 3559 quiet [KNL] Disable most log messages 3560 3561 r128= [HW,DRM] 3562 3563 raid= [HW,RAID] 3564 See Documentation/admin-guide/md.rst. 3565 3566 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes 3567 See Documentation/blockdev/ramdisk.txt. 3568 3569 random.trust_cpu={on,off} 3570 [KNL] Enable or disable trusting the use of the 3571 CPU's random number generator (if available) to 3572 fully seed the kernel's CRNG. Default is controlled 3573 by CONFIG_RANDOM_TRUST_CPU. 3574 3575 ras=option[,option,...] [KNL] RAS-specific options 3576 3577 cec_disable [X86] 3578 Disable the Correctable Errors Collector, 3579 see CONFIG_RAS_CEC help text. 3580 3581 rcu_nocbs= [KNL] 3582 The argument is a cpu list, as described above. 3583 3584 In kernels built with CONFIG_RCU_NOCB_CPU=y, set 3585 the specified list of CPUs to be no-callback CPUs. 3586 Invocation of these CPUs' RCU callbacks will be 3587 offloaded to "rcuox/N" kthreads created for that 3588 purpose, where "x" is "p" for RCU-preempt, and 3589 "s" for RCU-sched, and "N" is the CPU number. 3590 This reduces OS jitter on the offloaded CPUs, 3591 which can be useful for HPC and real-time 3592 workloads. It can also improve energy efficiency 3593 for asymmetric multiprocessors. 3594 3595 rcu_nocb_poll [KNL] 3596 Rather than requiring that offloaded CPUs 3597 (specified by rcu_nocbs= above) explicitly 3598 awaken the corresponding "rcuoN" kthreads, 3599 make these kthreads poll for callbacks. 3600 This improves the real-time response for the 3601 offloaded CPUs by relieving them of the need to 3602 wake up the corresponding kthread, but degrades 3603 energy efficiency by requiring that the kthreads 3604 periodically wake up to do the polling. 3605 3606 rcutree.blimit= [KNL] 3607 Set maximum number of finished RCU callbacks to 3608 process in one batch. 3609 3610 rcutree.dump_tree= [KNL] 3611 Dump the structure of the rcu_node combining tree 3612 out at early boot. This is used for diagnostic 3613 purposes, to verify correct tree setup. 3614 3615 rcutree.gp_cleanup_delay= [KNL] 3616 Set the number of jiffies to delay each step of 3617 RCU grace-period cleanup. 3618 3619 rcutree.gp_init_delay= [KNL] 3620 Set the number of jiffies to delay each step of 3621 RCU grace-period initialization. 3622 3623 rcutree.gp_preinit_delay= [KNL] 3624 Set the number of jiffies to delay each step of 3625 RCU grace-period pre-initialization, that is, 3626 the propagation of recent CPU-hotplug changes up 3627 the rcu_node combining tree. 3628 3629 rcutree.rcu_fanout_exact= [KNL] 3630 Disable autobalancing of the rcu_node combining 3631 tree. This is used by rcutorture, and might 3632 possibly be useful for architectures having high 3633 cache-to-cache transfer latencies. 3634 3635 rcutree.rcu_fanout_leaf= [KNL] 3636 Change the number of CPUs assigned to each 3637 leaf rcu_node structure. Useful for very 3638 large systems, which will choose the value 64, 3639 and for NUMA systems with large remote-access 3640 latencies, which will choose a value aligned 3641 with the appropriate hardware boundaries. 3642 3643 rcutree.jiffies_till_sched_qs= [KNL] 3644 Set required age in jiffies for a 3645 given grace period before RCU starts 3646 soliciting quiescent-state help from 3647 rcu_note_context_switch(). If not specified, the 3648 kernel will calculate a value based on the most 3649 recent settings of rcutree.jiffies_till_first_fqs 3650 and rcutree.jiffies_till_next_fqs. 3651 This calculated value may be viewed in 3652 rcutree.jiffies_to_sched_qs. Any attempt to 3653 set rcutree.jiffies_to_sched_qs will be 3654 cheerfully overwritten. 3655 3656 rcutree.jiffies_till_first_fqs= [KNL] 3657 Set delay from grace-period initialization to 3658 first attempt to force quiescent states. 3659 Units are jiffies, minimum value is zero, 3660 and maximum value is HZ. 3661 3662 rcutree.jiffies_till_next_fqs= [KNL] 3663 Set delay between subsequent attempts to force 3664 quiescent states. Units are jiffies, minimum 3665 value is one, and maximum value is HZ. 3666 3667 rcutree.kthread_prio= [KNL,BOOT] 3668 Set the SCHED_FIFO priority of the RCU per-CPU 3669 kthreads (rcuc/N). This value is also used for 3670 the priority of the RCU boost threads (rcub/N) 3671 and for the RCU grace-period kthreads (rcu_bh, 3672 rcu_preempt, and rcu_sched). If RCU_BOOST is 3673 set, valid values are 1-99 and the default is 1 3674 (the least-favored priority). Otherwise, when 3675 RCU_BOOST is not set, valid values are 0-99 and 3676 the default is zero (non-realtime operation). 3677 3678 rcutree.rcu_nocb_leader_stride= [KNL] 3679 Set the number of NOCB kthread groups, which 3680 defaults to the square root of the number of 3681 CPUs. Larger numbers reduces the wakeup overhead 3682 on the per-CPU grace-period kthreads, but increases 3683 that same overhead on each group's leader. 3684 3685 rcutree.qhimark= [KNL] 3686 Set threshold of queued RCU callbacks beyond which 3687 batch limiting is disabled. 3688 3689 rcutree.qlowmark= [KNL] 3690 Set threshold of queued RCU callbacks below which 3691 batch limiting is re-enabled. 3692 3693 rcutree.rcu_idle_gp_delay= [KNL] 3694 Set wakeup interval for idle CPUs that have 3695 RCU callbacks (RCU_FAST_NO_HZ=y). 3696 3697 rcutree.rcu_idle_lazy_gp_delay= [KNL] 3698 Set wakeup interval for idle CPUs that have 3699 only "lazy" RCU callbacks (RCU_FAST_NO_HZ=y). 3700 Lazy RCU callbacks are those which RCU can 3701 prove do nothing more than free memory. 3702 3703 rcutree.rcu_kick_kthreads= [KNL] 3704 Cause the grace-period kthread to get an extra 3705 wake_up() if it sleeps three times longer than 3706 it should at force-quiescent-state time. 3707 This wake_up() will be accompanied by a 3708 WARN_ONCE() splat and an ftrace_dump(). 3709 3710 rcuperf.gp_async= [KNL] 3711 Measure performance of asynchronous 3712 grace-period primitives such as call_rcu(). 3713 3714 rcuperf.gp_async_max= [KNL] 3715 Specify the maximum number of outstanding 3716 callbacks per writer thread. When a writer 3717 thread exceeds this limit, it invokes the 3718 corresponding flavor of rcu_barrier() to allow 3719 previously posted callbacks to drain. 3720 3721 rcuperf.gp_exp= [KNL] 3722 Measure performance of expedited synchronous 3723 grace-period primitives. 3724 3725 rcuperf.holdoff= [KNL] 3726 Set test-start holdoff period. The purpose of 3727 this parameter is to delay the start of the 3728 test until boot completes in order to avoid 3729 interference. 3730 3731 rcuperf.nreaders= [KNL] 3732 Set number of RCU readers. The value -1 selects 3733 N, where N is the number of CPUs. A value 3734 "n" less than -1 selects N-n+1, where N is again 3735 the number of CPUs. For example, -2 selects N 3736 (the number of CPUs), -3 selects N+1, and so on. 3737 A value of "n" less than or equal to -N selects 3738 a single reader. 3739 3740 rcuperf.nwriters= [KNL] 3741 Set number of RCU writers. The values operate 3742 the same as for rcuperf.nreaders. 3743 N, where N is the number of CPUs 3744 3745 rcuperf.perf_type= [KNL] 3746 Specify the RCU implementation to test. 3747 3748 rcuperf.shutdown= [KNL] 3749 Shut the system down after performance tests 3750 complete. This is useful for hands-off automated 3751 testing. 3752 3753 rcuperf.verbose= [KNL] 3754 Enable additional printk() statements. 3755 3756 rcuperf.writer_holdoff= [KNL] 3757 Write-side holdoff between grace periods, 3758 in microseconds. The default of zero says 3759 no holdoff. 3760 3761 rcutorture.fqs_duration= [KNL] 3762 Set duration of force_quiescent_state bursts 3763 in microseconds. 3764 3765 rcutorture.fqs_holdoff= [KNL] 3766 Set holdoff time within force_quiescent_state bursts 3767 in microseconds. 3768 3769 rcutorture.fqs_stutter= [KNL] 3770 Set wait time between force_quiescent_state bursts 3771 in seconds. 3772 3773 rcutorture.fwd_progress= [KNL] 3774 Enable RCU grace-period forward-progress testing 3775 for the types of RCU supporting this notion. 3776 3777 rcutorture.fwd_progress_div= [KNL] 3778 Specify the fraction of a CPU-stall-warning 3779 period to do tight-loop forward-progress testing. 3780 3781 rcutorture.fwd_progress_holdoff= [KNL] 3782 Number of seconds to wait between successive 3783 forward-progress tests. 3784 3785 rcutorture.fwd_progress_need_resched= [KNL] 3786 Enclose cond_resched() calls within checks for 3787 need_resched() during tight-loop forward-progress 3788 testing. 3789 3790 rcutorture.gp_cond= [KNL] 3791 Use conditional/asynchronous update-side 3792 primitives, if available. 3793 3794 rcutorture.gp_exp= [KNL] 3795 Use expedited update-side primitives, if available. 3796 3797 rcutorture.gp_normal= [KNL] 3798 Use normal (non-expedited) asynchronous 3799 update-side primitives, if available. 3800 3801 rcutorture.gp_sync= [KNL] 3802 Use normal (non-expedited) synchronous 3803 update-side primitives, if available. If all 3804 of rcutorture.gp_cond=, rcutorture.gp_exp=, 3805 rcutorture.gp_normal=, and rcutorture.gp_sync= 3806 are zero, rcutorture acts as if is interpreted 3807 they are all non-zero. 3808 3809 rcutorture.n_barrier_cbs= [KNL] 3810 Set callbacks/threads for rcu_barrier() testing. 3811 3812 rcutorture.nfakewriters= [KNL] 3813 Set number of concurrent RCU writers. These just 3814 stress RCU, they don't participate in the actual 3815 test, hence the "fake". 3816 3817 rcutorture.nreaders= [KNL] 3818 Set number of RCU readers. The value -1 selects 3819 N-1, where N is the number of CPUs. A value 3820 "n" less than -1 selects N-n-2, where N is again 3821 the number of CPUs. For example, -2 selects N 3822 (the number of CPUs), -3 selects N+1, and so on. 3823 3824 rcutorture.object_debug= [KNL] 3825 Enable debug-object double-call_rcu() testing. 3826 3827 rcutorture.onoff_holdoff= [KNL] 3828 Set time (s) after boot for CPU-hotplug testing. 3829 3830 rcutorture.onoff_interval= [KNL] 3831 Set time (jiffies) between CPU-hotplug operations, 3832 or zero to disable CPU-hotplug testing. 3833 3834 rcutorture.shuffle_interval= [KNL] 3835 Set task-shuffle interval (s). Shuffling tasks 3836 allows some CPUs to go into dyntick-idle mode 3837 during the rcutorture test. 3838 3839 rcutorture.shutdown_secs= [KNL] 3840 Set time (s) after boot system shutdown. This 3841 is useful for hands-off automated testing. 3842 3843 rcutorture.stall_cpu= [KNL] 3844 Duration of CPU stall (s) to test RCU CPU stall 3845 warnings, zero to disable. 3846 3847 rcutorture.stall_cpu_holdoff= [KNL] 3848 Time to wait (s) after boot before inducing stall. 3849 3850 rcutorture.stall_cpu_irqsoff= [KNL] 3851 Disable interrupts while stalling if set. 3852 3853 rcutorture.stat_interval= [KNL] 3854 Time (s) between statistics printk()s. 3855 3856 rcutorture.stutter= [KNL] 3857 Time (s) to stutter testing, for example, specifying 3858 five seconds causes the test to run for five seconds, 3859 wait for five seconds, and so on. This tests RCU's 3860 ability to transition abruptly to and from idle. 3861 3862 rcutorture.test_boost= [KNL] 3863 Test RCU priority boosting? 0=no, 1=maybe, 2=yes. 3864 "Maybe" means test if the RCU implementation 3865 under test support RCU priority boosting. 3866 3867 rcutorture.test_boost_duration= [KNL] 3868 Duration (s) of each individual boost test. 3869 3870 rcutorture.test_boost_interval= [KNL] 3871 Interval (s) between each boost test. 3872 3873 rcutorture.test_no_idle_hz= [KNL] 3874 Test RCU's dyntick-idle handling. See also the 3875 rcutorture.shuffle_interval parameter. 3876 3877 rcutorture.torture_type= [KNL] 3878 Specify the RCU implementation to test. 3879 3880 rcutorture.verbose= [KNL] 3881 Enable additional printk() statements. 3882 3883 rcupdate.rcu_cpu_stall_suppress= [KNL] 3884 Suppress RCU CPU stall warning messages. 3885 3886 rcupdate.rcu_cpu_stall_timeout= [KNL] 3887 Set timeout for RCU CPU stall warning messages. 3888 3889 rcupdate.rcu_expedited= [KNL] 3890 Use expedited grace-period primitives, for 3891 example, synchronize_rcu_expedited() instead 3892 of synchronize_rcu(). This reduces latency, 3893 but can increase CPU utilization, degrade 3894 real-time latency, and degrade energy efficiency. 3895 No effect on CONFIG_TINY_RCU kernels. 3896 3897 rcupdate.rcu_normal= [KNL] 3898 Use only normal grace-period primitives, 3899 for example, synchronize_rcu() instead of 3900 synchronize_rcu_expedited(). This improves 3901 real-time latency, CPU utilization, and 3902 energy efficiency, but can expose users to 3903 increased grace-period latency. This parameter 3904 overrides rcupdate.rcu_expedited. No effect on 3905 CONFIG_TINY_RCU kernels. 3906 3907 rcupdate.rcu_normal_after_boot= [KNL] 3908 Once boot has completed (that is, after 3909 rcu_end_inkernel_boot() has been invoked), use 3910 only normal grace-period primitives. No effect 3911 on CONFIG_TINY_RCU kernels. 3912 3913 rcupdate.rcu_task_stall_timeout= [KNL] 3914 Set timeout in jiffies for RCU task stall warning 3915 messages. Disable with a value less than or equal 3916 to zero. 3917 3918 rcupdate.rcu_self_test= [KNL] 3919 Run the RCU early boot self tests 3920 3921 rdinit= [KNL] 3922 Format: <full_path> 3923 Run specified binary instead of /init from the ramdisk, 3924 used for early userspace startup. See initrd. 3925 3926 rdt= [HW,X86,RDT] 3927 Turn on/off individual RDT features. List is: 3928 cmt, mbmtotal, mbmlocal, l3cat, l3cdp, l2cat, l2cdp, 3929 mba. 3930 E.g. to turn on cmt and turn off mba use: 3931 rdt=cmt,!mba 3932 3933 reboot= [KNL] 3934 Format (x86 or x86_64): 3935 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] \ 3936 [[,]s[mp]#### \ 3937 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \ 3938 [[,]f[orce] 3939 Where reboot_mode is one of warm (soft) or cold (hard) or gpio, 3940 reboot_type is one of bios, acpi, kbd, triple, efi, or pci, 3941 reboot_force is either force or not specified, 3942 reboot_cpu is s[mp]#### with #### being the processor 3943 to be used for rebooting. 3944 3945 relax_domain_level= 3946 [KNL, SMP] Set scheduler's default relax_domain_level. 3947 See Documentation/cgroup-v1/cpusets.txt. 3948 3949 reserve= [KNL,BUGS] Force kernel to ignore I/O ports or memory 3950 Format: <base1>,<size1>[,<base2>,<size2>,...] 3951 Reserve I/O ports or memory so the kernel won't use 3952 them. If <base> is less than 0x10000, the region 3953 is assumed to be I/O ports; otherwise it is memory. 3954 3955 reservetop= [X86-32] 3956 Format: nn[KMG] 3957 Reserves a hole at the top of the kernel virtual 3958 address space. 3959 3960 reservelow= [X86] 3961 Format: nn[K] 3962 Set the amount of memory to reserve for BIOS at 3963 the bottom of the address space. 3964 3965 reset_devices [KNL] Force drivers to reset the underlying device 3966 during initialization. 3967 3968 resume= [SWSUSP] 3969 Specify the partition device for software suspend 3970 Format: 3971 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>} 3972 3973 resume_offset= [SWSUSP] 3974 Specify the offset from the beginning of the partition 3975 given by "resume=" at which the swap header is located, 3976 in <PAGE_SIZE> units (needed only for swap files). 3977 See Documentation/power/swsusp-and-swap-files.txt 3978 3979 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to 3980 read the resume files 3981 3982 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up. 3983 Useful for devices that are detected asynchronously 3984 (e.g. USB and MMC devices). 3985 3986 hibernate= [HIBERNATION] 3987 noresume Don't check if there's a hibernation image 3988 present during boot. 3989 nocompress Don't compress/decompress hibernation images. 3990 no Disable hibernation and resume. 3991 protect_image Turn on image protection during restoration 3992 (that will set all pages holding image data 3993 during restoration read-only). 3994 3995 retain_initrd [RAM] Keep initrd memory after extraction 3996 3997 rfkill.default_state= 3998 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm, 3999 etc. communication is blocked by default. 4000 1 Unblocked. 4001 4002 rfkill.master_switch_mode= 4003 0 The "airplane mode" button does nothing. 4004 1 The "airplane mode" button toggles between everything 4005 blocked and the previous configuration. 4006 2 The "airplane mode" button toggles between everything 4007 blocked and everything unblocked. 4008 4009 rhash_entries= [KNL,NET] 4010 Set number of hash buckets for route cache 4011 4012 ring3mwait=disable 4013 [KNL] Disable ring 3 MONITOR/MWAIT feature on supported 4014 CPUs. 4015 4016 ro [KNL] Mount root device read-only on boot 4017 4018 rodata= [KNL] 4019 on Mark read-only kernel memory as read-only (default). 4020 off Leave read-only kernel memory writable for debugging. 4021 4022 rockchip.usb_uart 4023 Enable the uart passthrough on the designated usb port 4024 on Rockchip SoCs. When active, the signals of the 4025 debug-uart get routed to the D+ and D- pins of the usb 4026 port and the regular usb controller gets disabled. 4027 4028 root= [KNL] Root filesystem 4029 See name_to_dev_t comment in init/do_mounts.c. 4030 4031 rootdelay= [KNL] Delay (in seconds) to pause before attempting to 4032 mount the root filesystem 4033 4034 rootflags= [KNL] Set root filesystem mount option string 4035 4036 rootfstype= [KNL] Set root filesystem type 4037 4038 rootwait [KNL] Wait (indefinitely) for root device to show up. 4039 Useful for devices that are detected asynchronously 4040 (e.g. USB and MMC devices). 4041 4042 rproc_mem=nn[KMG][@address] 4043 [KNL,ARM,CMA] Remoteproc physical memory block. 4044 Memory area to be used by remote processor image, 4045 managed by CMA. 4046 4047 rw [KNL] Mount root device read-write on boot 4048 4049 S [KNL] Run init in single mode 4050 4051 s390_iommu= [HW,S390] 4052 Set s390 IOTLB flushing mode 4053 strict 4054 With strict flushing every unmap operation will result in 4055 an IOTLB flush. Default is lazy flushing before reuse, 4056 which is faster. 4057 4058 sa1100ir [NET] 4059 See drivers/net/irda/sa1100_ir.c. 4060 4061 sbni= [NET] Granch SBNI12 leased line adapter 4062 4063 sched_debug [KNL] Enables verbose scheduler debug messages. 4064 4065 schedstats= [KNL,X86] Enable or disable scheduled statistics. 4066 Allowed values are enable and disable. This feature 4067 incurs a small amount of overhead in the scheduler 4068 but is useful for debugging and performance tuning. 4069 4070 skew_tick= [KNL] Offset the periodic timer tick per cpu to mitigate 4071 xtime_lock contention on larger systems, and/or RCU lock 4072 contention on all systems with CONFIG_MAXSMP set. 4073 Format: { "0" | "1" } 4074 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1" 4075 1 -- enable. 4076 Note: increases power consumption, thus should only be 4077 enabled if running jitter sensitive (HPC/RT) workloads. 4078 4079 security= [SECURITY] Choose a security module to enable at boot. 4080 If this boot parameter is not specified, only the first 4081 security module asking for security registration will be 4082 loaded. An invalid security module name will be treated 4083 as if no module has been chosen. 4084 4085 selinux= [SELINUX] Disable or enable SELinux at boot time. 4086 Format: { "0" | "1" } 4087 See security/selinux/Kconfig help text. 4088 0 -- disable. 4089 1 -- enable. 4090 Default value is set via kernel config option. 4091 If enabled at boot time, /selinux/disable can be used 4092 later to disable prior to initial policy load. 4093 4094 apparmor= [APPARMOR] Disable or enable AppArmor at boot time 4095 Format: { "0" | "1" } 4096 See security/apparmor/Kconfig help text 4097 0 -- disable. 4098 1 -- enable. 4099 Default value is set via kernel config option. 4100 4101 serialnumber [BUGS=X86-32] 4102 4103 shapers= [NET] 4104 Maximal number of shapers. 4105 4106 simeth= [IA-64] 4107 simscsi= 4108 4109 slram= [HW,MTD] 4110 4111 slab_nomerge [MM] 4112 Disable merging of slabs with similar size. May be 4113 necessary if there is some reason to distinguish 4114 allocs to different slabs, especially in hardened 4115 environments where the risk of heap overflows and 4116 layout control by attackers can usually be 4117 frustrated by disabling merging. This will reduce 4118 most of the exposure of a heap attack to a single 4119 cache (risks via metadata attacks are mostly 4120 unchanged). Debug options disable merging on their 4121 own. 4122 For more information see Documentation/vm/slub.rst. 4123 4124 slab_max_order= [MM, SLAB] 4125 Determines the maximum allowed order for slabs. 4126 A high setting may cause OOMs due to memory 4127 fragmentation. Defaults to 1 for systems with 4128 more than 32MB of RAM, 0 otherwise. 4129 4130 slub_debug[=options[,slabs]] [MM, SLUB] 4131 Enabling slub_debug allows one to determine the 4132 culprit if slab objects become corrupted. Enabling 4133 slub_debug can create guard zones around objects and 4134 may poison objects when not in use. Also tracks the 4135 last alloc / free. For more information see 4136 Documentation/vm/slub.rst. 4137 4138 slub_memcg_sysfs= [MM, SLUB] 4139 Determines whether to enable sysfs directories for 4140 memory cgroup sub-caches. 1 to enable, 0 to disable. 4141 The default is determined by CONFIG_SLUB_MEMCG_SYSFS_ON. 4142 Enabling this can lead to a very high number of debug 4143 directories and files being created under 4144 /sys/kernel/slub. 4145 4146 slub_max_order= [MM, SLUB] 4147 Determines the maximum allowed order for slabs. 4148 A high setting may cause OOMs due to memory 4149 fragmentation. For more information see 4150 Documentation/vm/slub.rst. 4151 4152 slub_min_objects= [MM, SLUB] 4153 The minimum number of objects per slab. SLUB will 4154 increase the slab order up to slub_max_order to 4155 generate a sufficiently large slab able to contain 4156 the number of objects indicated. The higher the number 4157 of objects the smaller the overhead of tracking slabs 4158 and the less frequently locks need to be acquired. 4159 For more information see Documentation/vm/slub.rst. 4160 4161 slub_min_order= [MM, SLUB] 4162 Determines the minimum page order for slabs. Must be 4163 lower than slub_max_order. 4164 For more information see Documentation/vm/slub.rst. 4165 4166 slub_nomerge [MM, SLUB] 4167 Same with slab_nomerge. This is supported for legacy. 4168 See slab_nomerge for more information. 4169 4170 smart2= [HW] 4171 Format: <io1>[,<io2>[,...,<io8>]] 4172 4173 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices 4174 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port 4175 smsc-ircc2.ircc_sir= [HW] SIR base I/O port 4176 smsc-ircc2.ircc_fir= [HW] FIR base I/O port 4177 smsc-ircc2.ircc_irq= [HW] IRQ line 4178 smsc-ircc2.ircc_dma= [HW] DMA channel 4179 smsc-ircc2.ircc_transceiver= [HW] Transceiver type: 4180 0: Toshiba Satellite 1800 (GP data pin select) 4181 1: Fast pin select (default) 4182 2: ATC IRMode 4183 4184 smt [KNL,S390] Set the maximum number of threads (logical 4185 CPUs) to use per physical CPU on systems capable of 4186 symmetric multithreading (SMT). Will be capped to the 4187 actual hardware limit. 4188 Format: <integer> 4189 Default: -1 (no limit) 4190 4191 softlockup_panic= 4192 [KNL] Should the soft-lockup detector generate panics. 4193 Format: <integer> 4194 4195 A nonzero value instructs the soft-lockup detector 4196 to panic the machine when a soft-lockup occurs. This 4197 is also controlled by CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC 4198 which is the respective build-time switch to that 4199 functionality. 4200 4201 softlockup_all_cpu_backtrace= 4202 [KNL] Should the soft-lockup detector generate 4203 backtraces on all cpus. 4204 Format: <integer> 4205 4206 sonypi.*= [HW] Sony Programmable I/O Control Device driver 4207 See Documentation/laptops/sonypi.txt 4208 4209 spectre_v2= [X86] Control mitigation of Spectre variant 2 4210 (indirect branch speculation) vulnerability. 4211 The default operation protects the kernel from 4212 user space attacks. 4213 4214 on - unconditionally enable, implies 4215 spectre_v2_user=on 4216 off - unconditionally disable, implies 4217 spectre_v2_user=off 4218 auto - kernel detects whether your CPU model is 4219 vulnerable 4220 4221 Selecting 'on' will, and 'auto' may, choose a 4222 mitigation method at run time according to the 4223 CPU, the available microcode, the setting of the 4224 CONFIG_RETPOLINE configuration option, and the 4225 compiler with which the kernel was built. 4226 4227 Selecting 'on' will also enable the mitigation 4228 against user space to user space task attacks. 4229 4230 Selecting 'off' will disable both the kernel and 4231 the user space protections. 4232 4233 Specific mitigations can also be selected manually: 4234 4235 retpoline - replace indirect branches 4236 retpoline,generic - google's original retpoline 4237 retpoline,amd - AMD-specific minimal thunk 4238 4239 Not specifying this option is equivalent to 4240 spectre_v2=auto. 4241 4242 spectre_v2_user= 4243 [X86] Control mitigation of Spectre variant 2 4244 (indirect branch speculation) vulnerability between 4245 user space tasks 4246 4247 on - Unconditionally enable mitigations. Is 4248 enforced by spectre_v2=on 4249 4250 off - Unconditionally disable mitigations. Is 4251 enforced by spectre_v2=off 4252 4253 prctl - Indirect branch speculation is enabled, 4254 but mitigation can be enabled via prctl 4255 per thread. The mitigation control state 4256 is inherited on fork. 4257 4258 prctl,ibpb 4259 - Like "prctl" above, but only STIBP is 4260 controlled per thread. IBPB is issued 4261 always when switching between different user 4262 space processes. 4263 4264 seccomp 4265 - Same as "prctl" above, but all seccomp 4266 threads will enable the mitigation unless 4267 they explicitly opt out. 4268 4269 seccomp,ibpb 4270 - Like "seccomp" above, but only STIBP is 4271 controlled per thread. IBPB is issued 4272 always when switching between different 4273 user space processes. 4274 4275 auto - Kernel selects the mitigation depending on 4276 the available CPU features and vulnerability. 4277 4278 Default mitigation: 4279 If CONFIG_SECCOMP=y then "seccomp", otherwise "prctl" 4280 4281 Not specifying this option is equivalent to 4282 spectre_v2_user=auto. 4283 4284 spec_store_bypass_disable= 4285 [HW] Control Speculative Store Bypass (SSB) Disable mitigation 4286 (Speculative Store Bypass vulnerability) 4287 4288 Certain CPUs are vulnerable to an exploit against a 4289 a common industry wide performance optimization known 4290 as "Speculative Store Bypass" in which recent stores 4291 to the same memory location may not be observed by 4292 later loads during speculative execution. The idea 4293 is that such stores are unlikely and that they can 4294 be detected prior to instruction retirement at the 4295 end of a particular speculation execution window. 4296 4297 In vulnerable processors, the speculatively forwarded 4298 store can be used in a cache side channel attack, for 4299 example to read memory to which the attacker does not 4300 directly have access (e.g. inside sandboxed code). 4301 4302 This parameter controls whether the Speculative Store 4303 Bypass optimization is used. 4304 4305 On x86 the options are: 4306 4307 on - Unconditionally disable Speculative Store Bypass 4308 off - Unconditionally enable Speculative Store Bypass 4309 auto - Kernel detects whether the CPU model contains an 4310 implementation of Speculative Store Bypass and 4311 picks the most appropriate mitigation. If the 4312 CPU is not vulnerable, "off" is selected. If the 4313 CPU is vulnerable the default mitigation is 4314 architecture and Kconfig dependent. See below. 4315 prctl - Control Speculative Store Bypass per thread 4316 via prctl. Speculative Store Bypass is enabled 4317 for a process by default. The state of the control 4318 is inherited on fork. 4319 seccomp - Same as "prctl" above, but all seccomp threads 4320 will disable SSB unless they explicitly opt out. 4321 4322 Default mitigations: 4323 X86: If CONFIG_SECCOMP=y "seccomp", otherwise "prctl" 4324 4325 On powerpc the options are: 4326 4327 on,auto - On Power8 and Power9 insert a store-forwarding 4328 barrier on kernel entry and exit. On Power7 4329 perform a software flush on kernel entry and 4330 exit. 4331 off - No action. 4332 4333 Not specifying this option is equivalent to 4334 spec_store_bypass_disable=auto. 4335 4336 spia_io_base= [HW,MTD] 4337 spia_fio_base= 4338 spia_pedr= 4339 spia_peddr= 4340 4341 srcutree.counter_wrap_check [KNL] 4342 Specifies how frequently to check for 4343 grace-period sequence counter wrap for the 4344 srcu_data structure's ->srcu_gp_seq_needed field. 4345 The greater the number of bits set in this kernel 4346 parameter, the less frequently counter wrap will 4347 be checked for. Note that the bottom two bits 4348 are ignored. 4349 4350 srcutree.exp_holdoff [KNL] 4351 Specifies how many nanoseconds must elapse 4352 since the end of the last SRCU grace period for 4353 a given srcu_struct until the next normal SRCU 4354 grace period will be considered for automatic 4355 expediting. Set to zero to disable automatic 4356 expediting. 4357 4358 ssbd= [ARM64,HW] 4359 Speculative Store Bypass Disable control 4360 4361 On CPUs that are vulnerable to the Speculative 4362 Store Bypass vulnerability and offer a 4363 firmware based mitigation, this parameter 4364 indicates how the mitigation should be used: 4365 4366 force-on: Unconditionally enable mitigation for 4367 for both kernel and userspace 4368 force-off: Unconditionally disable mitigation for 4369 for both kernel and userspace 4370 kernel: Always enable mitigation in the 4371 kernel, and offer a prctl interface 4372 to allow userspace to register its 4373 interest in being mitigated too. 4374 4375 stack_guard_gap= [MM] 4376 override the default stack gap protection. The value 4377 is in page units and it defines how many pages prior 4378 to (for stacks growing down) resp. after (for stacks 4379 growing up) the main stack are reserved for no other 4380 mapping. Default value is 256 pages. 4381 4382 stacktrace [FTRACE] 4383 Enabled the stack tracer on boot up. 4384 4385 stacktrace_filter=[function-list] 4386 [FTRACE] Limit the functions that the stack tracer 4387 will trace at boot up. function-list is a comma separated 4388 list of functions. This list can be changed at run 4389 time by the stack_trace_filter file in the debugfs 4390 tracing directory. Note, this enables stack tracing 4391 and the stacktrace above is not needed. 4392 4393 sti= [PARISC,HW] 4394 Format: <num> 4395 Set the STI (builtin display/keyboard on the HP-PARISC 4396 machines) console (graphic card) which should be used 4397 as the initial boot-console. 4398 See also comment in drivers/video/console/sticore.c. 4399 4400 sti_font= [HW] 4401 See comment in drivers/video/console/sticore.c. 4402 4403 stifb= [HW] 4404 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]] 4405 4406 sunrpc.min_resvport= 4407 sunrpc.max_resvport= 4408 [NFS,SUNRPC] 4409 SunRPC servers often require that client requests 4410 originate from a privileged port (i.e. a port in the 4411 range 0 < portnr < 1024). 4412 An administrator who wishes to reserve some of these 4413 ports for other uses may adjust the range that the 4414 kernel's sunrpc client considers to be privileged 4415 using these two parameters to set the minimum and 4416 maximum port values. 4417 4418 sunrpc.svc_rpc_per_connection_limit= 4419 [NFS,SUNRPC] 4420 Limit the number of requests that the server will 4421 process in parallel from a single connection. 4422 The default value is 0 (no limit). 4423 4424 sunrpc.pool_mode= 4425 [NFS] 4426 Control how the NFS server code allocates CPUs to 4427 service thread pools. Depending on how many NICs 4428 you have and where their interrupts are bound, this 4429 option will affect which CPUs will do NFS serving. 4430 Note: this parameter cannot be changed while the 4431 NFS server is running. 4432 4433 auto the server chooses an appropriate mode 4434 automatically using heuristics 4435 global a single global pool contains all CPUs 4436 percpu one pool for each CPU 4437 pernode one pool for each NUMA node (equivalent 4438 to global on non-NUMA machines) 4439 4440 sunrpc.tcp_slot_table_entries= 4441 sunrpc.udp_slot_table_entries= 4442 [NFS,SUNRPC] 4443 Sets the upper limit on the number of simultaneous 4444 RPC calls that can be sent from the client to a 4445 server. Increasing these values may allow you to 4446 improve throughput, but will also increase the 4447 amount of memory reserved for use by the client. 4448 4449 suspend.pm_test_delay= 4450 [SUSPEND] 4451 Sets the number of seconds to remain in a suspend test 4452 mode before resuming the system (see 4453 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG 4454 is set. Default value is 5. 4455 4456 swapaccount=[0|1] 4457 [KNL] Enable accounting of swap in memory resource 4458 controller if no parameter or 1 is given or disable 4459 it if 0 is given (See Documentation/cgroup-v1/memory.txt) 4460 4461 swiotlb= [ARM,IA-64,PPC,MIPS,X86] 4462 Format: { <int> | force | noforce } 4463 <int> -- Number of I/O TLB slabs 4464 force -- force using of bounce buffers even if they 4465 wouldn't be automatically used by the kernel 4466 noforce -- Never use bounce buffers (for debugging) 4467 4468 switches= [HW,M68k] 4469 4470 sysfs.deprecated=0|1 [KNL] 4471 Enable/disable old style sysfs layout for old udev 4472 on older distributions. When this option is enabled 4473 very new udev will not work anymore. When this option 4474 is disabled (or CONFIG_SYSFS_DEPRECATED not compiled) 4475 in older udev will not work anymore. 4476 Default depends on CONFIG_SYSFS_DEPRECATED_V2 set in 4477 the kernel configuration. 4478 4479 sysrq_always_enabled 4480 [KNL] 4481 Ignore sysrq setting - this boot parameter will 4482 neutralize any effect of /proc/sys/kernel/sysrq. 4483 Useful for debugging. 4484 4485 tcpmhash_entries= [KNL,NET] 4486 Set the number of tcp_metrics_hash slots. 4487 Default value is 8192 or 16384 depending on total 4488 ram pages. This is used to specify the TCP metrics 4489 cache size. See Documentation/networking/ip-sysctl.txt 4490 "tcp_no_metrics_save" section for more details. 4491 4492 tdfx= [HW,DRM] 4493 4494 test_suspend= [SUSPEND][,N] 4495 Specify "mem" (for Suspend-to-RAM) or "standby" (for 4496 standby suspend) or "freeze" (for suspend type freeze) 4497 as the system sleep state during system startup with 4498 the optional capability to repeat N number of times. 4499 The system is woken from this state using a 4500 wakeup-capable RTC alarm. 4501 4502 thash_entries= [KNL,NET] 4503 Set number of hash buckets for TCP connection 4504 4505 thermal.act= [HW,ACPI] 4506 -1: disable all active trip points in all thermal zones 4507 <degrees C>: override all lowest active trip points 4508 4509 thermal.crt= [HW,ACPI] 4510 -1: disable all critical trip points in all thermal zones 4511 <degrees C>: override all critical trip points 4512 4513 thermal.nocrt= [HW,ACPI] 4514 Set to disable actions on ACPI thermal zone 4515 critical and hot trip points. 4516 4517 thermal.off= [HW,ACPI] 4518 1: disable ACPI thermal control 4519 4520 thermal.psv= [HW,ACPI] 4521 -1: disable all passive trip points 4522 <degrees C>: override all passive trip points to this 4523 value 4524 4525 thermal.tzp= [HW,ACPI] 4526 Specify global default ACPI thermal zone polling rate 4527 <deci-seconds>: poll all this frequency 4528 0: no polling (default) 4529 4530 threadirqs [KNL] 4531 Force threading of all interrupt handlers except those 4532 marked explicitly IRQF_NO_THREAD. 4533 4534 tmem [KNL,XEN] 4535 Enable the Transcendent memory driver if built-in. 4536 4537 tmem.cleancache=0|1 [KNL, XEN] 4538 Default is on (1). Disable the usage of the cleancache 4539 API to send anonymous pages to the hypervisor. 4540 4541 tmem.frontswap=0|1 [KNL, XEN] 4542 Default is on (1). Disable the usage of the frontswap 4543 API to send swap pages to the hypervisor. If disabled 4544 the selfballooning and selfshrinking are force disabled. 4545 4546 tmem.selfballooning=0|1 [KNL, XEN] 4547 Default is on (1). Disable the driving of swap pages 4548 to the hypervisor. 4549 4550 tmem.selfshrinking=0|1 [KNL, XEN] 4551 Default is on (1). Partial swapoff that immediately 4552 transfers pages from Xen hypervisor back to the 4553 kernel based on different criteria. 4554 4555 topology= [S390] 4556 Format: {off | on} 4557 Specify if the kernel should make use of the cpu 4558 topology information if the hardware supports this. 4559 The scheduler will make use of this information and 4560 e.g. base its process migration decisions on it. 4561 Default is on. 4562 4563 topology_updates= [KNL, PPC, NUMA] 4564 Format: {off} 4565 Specify if the kernel should ignore (off) 4566 topology updates sent by the hypervisor to this 4567 LPAR. 4568 4569 tp720= [HW,PS2] 4570 4571 tpm_suspend_pcr=[HW,TPM] 4572 Format: integer pcr id 4573 Specify that at suspend time, the tpm driver 4574 should extend the specified pcr with zeros, 4575 as a workaround for some chips which fail to 4576 flush the last written pcr on TPM_SaveState. 4577 This will guarantee that all the other pcrs 4578 are saved. 4579 4580 trace_buf_size=nn[KMG] 4581 [FTRACE] will set tracing buffer size on each cpu. 4582 4583 trace_event=[event-list] 4584 [FTRACE] Set and start specified trace events in order 4585 to facilitate early boot debugging. The event-list is a 4586 comma separated list of trace events to enable. See 4587 also Documentation/trace/events.rst 4588 4589 trace_options=[option-list] 4590 [FTRACE] Enable or disable tracer options at boot. 4591 The option-list is a comma delimited list of options 4592 that can be enabled or disabled just as if you were 4593 to echo the option name into 4594 4595 /sys/kernel/debug/tracing/trace_options 4596 4597 For example, to enable stacktrace option (to dump the 4598 stack trace of each event), add to the command line: 4599 4600 trace_options=stacktrace 4601 4602 See also Documentation/trace/ftrace.rst "trace options" 4603 section. 4604 4605 tp_printk[FTRACE] 4606 Have the tracepoints sent to printk as well as the 4607 tracing ring buffer. This is useful for early boot up 4608 where the system hangs or reboots and does not give the 4609 option for reading the tracing buffer or performing a 4610 ftrace_dump_on_oops. 4611 4612 To turn off having tracepoints sent to printk, 4613 echo 0 > /proc/sys/kernel/tracepoint_printk 4614 Note, echoing 1 into this file without the 4615 tracepoint_printk kernel cmdline option has no effect. 4616 4617 ** CAUTION ** 4618 4619 Having tracepoints sent to printk() and activating high 4620 frequency tracepoints such as irq or sched, can cause 4621 the system to live lock. 4622 4623 traceoff_on_warning 4624 [FTRACE] enable this option to disable tracing when a 4625 warning is hit. This turns off "tracing_on". Tracing can 4626 be enabled again by echoing '1' into the "tracing_on" 4627 file located in /sys/kernel/debug/tracing/ 4628 4629 This option is useful, as it disables the trace before 4630 the WARNING dump is called, which prevents the trace to 4631 be filled with content caused by the warning output. 4632 4633 This option can also be set at run time via the sysctl 4634 option: kernel/traceoff_on_warning 4635 4636 transparent_hugepage= 4637 [KNL] 4638 Format: [always|madvise|never] 4639 Can be used to control the default behavior of the system 4640 with respect to transparent hugepages. 4641 See Documentation/admin-guide/mm/transhuge.rst 4642 for more details. 4643 4644 tsc= Disable clocksource stability checks for TSC. 4645 Format: <string> 4646 [x86] reliable: mark tsc clocksource as reliable, this 4647 disables clocksource verification at runtime, as well 4648 as the stability checks done at bootup. Used to enable 4649 high-resolution timer mode on older hardware, and in 4650 virtualized environment. 4651 [x86] noirqtime: Do not use TSC to do irq accounting. 4652 Used to run time disable IRQ_TIME_ACCOUNTING on any 4653 platforms where RDTSC is slow and this accounting 4654 can add overhead. 4655 [x86] unstable: mark the TSC clocksource as unstable, this 4656 marks the TSC unconditionally unstable at bootup and 4657 avoids any further wobbles once the TSC watchdog notices. 4658 4659 turbografx.map[2|3]= [HW,JOY] 4660 TurboGraFX parallel port interface 4661 Format: 4662 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7> 4663 See also Documentation/input/devices/joystick-parport.rst 4664 4665 udbg-immortal [PPC] When debugging early kernel crashes that 4666 happen after console_init() and before a proper 4667 console driver takes over, this boot options might 4668 help "seeing" what's going on. 4669 4670 uhash_entries= [KNL,NET] 4671 Set number of hash buckets for UDP/UDP-Lite connections 4672 4673 uhci-hcd.ignore_oc= 4674 [USB] Ignore overcurrent events (default N). 4675 Some badly-designed motherboards generate lots of 4676 bogus events, for ports that aren't wired to 4677 anything. Set this parameter to avoid log spamming. 4678 Note that genuine overcurrent events won't be 4679 reported either. 4680 4681 unknown_nmi_panic 4682 [X86] Cause panic on unknown NMI. 4683 4684 usbcore.authorized_default= 4685 [USB] Default USB device authorization: 4686 (default -1 = authorized except for wireless USB, 4687 0 = not authorized, 1 = authorized) 4688 4689 usbcore.autosuspend= 4690 [USB] The autosuspend time delay (in seconds) used 4691 for newly-detected USB devices (default 2). This 4692 is the time required before an idle device will be 4693 autosuspended. Devices for which the delay is set 4694 to a negative value won't be autosuspended at all. 4695 4696 usbcore.usbfs_snoop= 4697 [USB] Set to log all usbfs traffic (default 0 = off). 4698 4699 usbcore.usbfs_snoop_max= 4700 [USB] Maximum number of bytes to snoop in each URB 4701 (default = 65536). 4702 4703 usbcore.blinkenlights= 4704 [USB] Set to cycle leds on hubs (default 0 = off). 4705 4706 usbcore.old_scheme_first= 4707 [USB] Start with the old device initialization 4708 scheme, applies only to low and full-speed devices 4709 (default 0 = off). 4710 4711 usbcore.usbfs_memory_mb= 4712 [USB] Memory limit (in MB) for buffers allocated by 4713 usbfs (default = 16, 0 = max = 2047). 4714 4715 usbcore.use_both_schemes= 4716 [USB] Try the other device initialization scheme 4717 if the first one fails (default 1 = enabled). 4718 4719 usbcore.initial_descriptor_timeout= 4720 [USB] Specifies timeout for the initial 64-byte 4721 USB_REQ_GET_DESCRIPTOR request in milliseconds 4722 (default 5000 = 5.0 seconds). 4723 4724 usbcore.nousb [USB] Disable the USB subsystem 4725 4726 usbcore.quirks= 4727 [USB] A list of quirk entries to augment the built-in 4728 usb core quirk list. List entries are separated by 4729 commas. Each entry has the form 4730 VendorID:ProductID:Flags. The IDs are 4-digit hex 4731 numbers and Flags is a set of letters. Each letter 4732 will change the built-in quirk; setting it if it is 4733 clear and clearing it if it is set. The letters have 4734 the following meanings: 4735 a = USB_QUIRK_STRING_FETCH_255 (string 4736 descriptors must not be fetched using 4737 a 255-byte read); 4738 b = USB_QUIRK_RESET_RESUME (device can't resume 4739 correctly so reset it instead); 4740 c = USB_QUIRK_NO_SET_INTF (device can't handle 4741 Set-Interface requests); 4742 d = USB_QUIRK_CONFIG_INTF_STRINGS (device can't 4743 handle its Configuration or Interface 4744 strings); 4745 e = USB_QUIRK_RESET (device can't be reset 4746 (e.g morph devices), don't use reset); 4747 f = USB_QUIRK_HONOR_BNUMINTERFACES (device has 4748 more interface descriptions than the 4749 bNumInterfaces count, and can't handle 4750 talking to these interfaces); 4751 g = USB_QUIRK_DELAY_INIT (device needs a pause 4752 during initialization, after we read 4753 the device descriptor); 4754 h = USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL (For 4755 high speed and super speed interrupt 4756 endpoints, the USB 2.0 and USB 3.0 spec 4757 require the interval in microframes (1 4758 microframe = 125 microseconds) to be 4759 calculated as interval = 2 ^ 4760 (bInterval-1). 4761 Devices with this quirk report their 4762 bInterval as the result of this 4763 calculation instead of the exponent 4764 variable used in the calculation); 4765 i = USB_QUIRK_DEVICE_QUALIFIER (device can't 4766 handle device_qualifier descriptor 4767 requests); 4768 j = USB_QUIRK_IGNORE_REMOTE_WAKEUP (device 4769 generates spurious wakeup, ignore 4770 remote wakeup capability); 4771 k = USB_QUIRK_NO_LPM (device can't handle Link 4772 Power Management); 4773 l = USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL 4774 (Device reports its bInterval as linear 4775 frames instead of the USB 2.0 4776 calculation); 4777 m = USB_QUIRK_DISCONNECT_SUSPEND (Device needs 4778 to be disconnected before suspend to 4779 prevent spurious wakeup); 4780 n = USB_QUIRK_DELAY_CTRL_MSG (Device needs a 4781 pause after every control message); 4782 o = USB_QUIRK_HUB_SLOW_RESET (Hub needs extra 4783 delay after resetting its port); 4784 Example: quirks=0781:5580:bk,0a5c:5834:gij 4785 4786 usbhid.mousepoll= 4787 [USBHID] The interval which mice are to be polled at. 4788 4789 usbhid.jspoll= 4790 [USBHID] The interval which joysticks are to be polled at. 4791 4792 usbhid.kbpoll= 4793 [USBHID] The interval which keyboards are to be polled at. 4794 4795 usb-storage.delay_use= 4796 [UMS] The delay in seconds before a new device is 4797 scanned for Logical Units (default 1). 4798 4799 usb-storage.quirks= 4800 [UMS] A list of quirks entries to supplement or 4801 override the built-in unusual_devs list. List 4802 entries are separated by commas. Each entry has 4803 the form VID:PID:Flags where VID and PID are Vendor 4804 and Product ID values (4-digit hex numbers) and 4805 Flags is a set of characters, each corresponding 4806 to a common usb-storage quirk flag as follows: 4807 a = SANE_SENSE (collect more than 18 bytes 4808 of sense data); 4809 b = BAD_SENSE (don't collect more than 18 4810 bytes of sense data); 4811 c = FIX_CAPACITY (decrease the reported 4812 device capacity by one sector); 4813 d = NO_READ_DISC_INFO (don't use 4814 READ_DISC_INFO command); 4815 e = NO_READ_CAPACITY_16 (don't use 4816 READ_CAPACITY_16 command); 4817 f = NO_REPORT_OPCODES (don't use report opcodes 4818 command, uas only); 4819 g = MAX_SECTORS_240 (don't transfer more than 4820 240 sectors at a time, uas only); 4821 h = CAPACITY_HEURISTICS (decrease the 4822 reported device capacity by one 4823 sector if the number is odd); 4824 i = IGNORE_DEVICE (don't bind to this 4825 device); 4826 j = NO_REPORT_LUNS (don't use report luns 4827 command, uas only); 4828 l = NOT_LOCKABLE (don't try to lock and 4829 unlock ejectable media); 4830 m = MAX_SECTORS_64 (don't transfer more 4831 than 64 sectors = 32 KB at a time); 4832 n = INITIAL_READ10 (force a retry of the 4833 initial READ(10) command); 4834 o = CAPACITY_OK (accept the capacity 4835 reported by the device); 4836 p = WRITE_CACHE (the device cache is ON 4837 by default); 4838 r = IGNORE_RESIDUE (the device reports 4839 bogus residue values); 4840 s = SINGLE_LUN (the device has only one 4841 Logical Unit); 4842 t = NO_ATA_1X (don't allow ATA(12) and ATA(16) 4843 commands, uas only); 4844 u = IGNORE_UAS (don't bind to the uas driver); 4845 w = NO_WP_DETECT (don't test whether the 4846 medium is write-protected). 4847 y = ALWAYS_SYNC (issue a SYNCHRONIZE_CACHE 4848 even if the device claims no cache) 4849 Example: quirks=0419:aaf5:rl,0421:0433:rc 4850 4851 user_debug= [KNL,ARM] 4852 Format: <int> 4853 See arch/arm/Kconfig.debug help text. 4854 1 - undefined instruction events 4855 2 - system calls 4856 4 - invalid data aborts 4857 8 - SIGSEGV faults 4858 16 - SIGBUS faults 4859 Example: user_debug=31 4860 4861 userpte= 4862 [X86] Flags controlling user PTE allocations. 4863 4864 nohigh = do not allocate PTE pages in 4865 HIGHMEM regardless of setting 4866 of CONFIG_HIGHPTE. 4867 4868 vdso= [X86,SH] 4869 On X86_32, this is an alias for vdso32=. Otherwise: 4870 4871 vdso=1: enable VDSO (the default) 4872 vdso=0: disable VDSO mapping 4873 4874 vdso32= [X86] Control the 32-bit vDSO 4875 vdso32=1: enable 32-bit VDSO 4876 vdso32=0 or vdso32=2: disable 32-bit VDSO 4877 4878 See the help text for CONFIG_COMPAT_VDSO for more 4879 details. If CONFIG_COMPAT_VDSO is set, the default is 4880 vdso32=0; otherwise, the default is vdso32=1. 4881 4882 For compatibility with older kernels, vdso32=2 is an 4883 alias for vdso32=0. 4884 4885 Try vdso32=0 if you encounter an error that says: 4886 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed! 4887 4888 vector= [IA-64,SMP] 4889 vector=percpu: enable percpu vector domain 4890 4891 video= [FB] Frame buffer configuration 4892 See Documentation/fb/modedb.txt. 4893 4894 video.brightness_switch_enabled= [0,1] 4895 If set to 1, on receiving an ACPI notify event 4896 generated by hotkey, video driver will adjust brightness 4897 level and then send out the event to user space through 4898 the allocated input device; If set to 0, video driver 4899 will only send out the event without touching backlight 4900 brightness level. 4901 default: 1 4902 4903 virtio_mmio.device= 4904 [VMMIO] Memory mapped virtio (platform) device. 4905 4906 <size>@<baseaddr>:<irq>[:<id>] 4907 where: 4908 <size> := size (can use standard suffixes 4909 like K, M and G) 4910 <baseaddr> := physical base address 4911 <irq> := interrupt number (as passed to 4912 request_irq()) 4913 <id> := (optional) platform device id 4914 example: 4915 virtio_mmio.device=1K@0x100b0000:48:7 4916 4917 Can be used multiple times for multiple devices. 4918 4919 vga= [BOOT,X86-32] Select a particular video mode 4920 See Documentation/x86/boot.txt and 4921 Documentation/svga.txt. 4922 Use vga=ask for menu. 4923 This is actually a boot loader parameter; the value is 4924 passed to the kernel using a special protocol. 4925 4926 vm_debug[=options] [KNL] Available with CONFIG_DEBUG_VM=y. 4927 May slow down system boot speed, especially when 4928 enabled on systems with a large amount of memory. 4929 All options are enabled by default, and this 4930 interface is meant to allow for selectively 4931 enabling or disabling specific virtual memory 4932 debugging features. 4933 4934 Available options are: 4935 P Enable page structure init time poisoning 4936 - Disable all of the above options 4937 4938 vmalloc=nn[KMG] [KNL,BOOT] Forces the vmalloc area to have an exact 4939 size of <nn>. This can be used to increase the 4940 minimum size (128MB on x86). It can also be used to 4941 decrease the size and leave more room for directly 4942 mapped kernel RAM. 4943 4944 vmcp_cma=nn[MG] [KNL,S390] 4945 Sets the memory size reserved for contiguous memory 4946 allocations for the vmcp device driver. 4947 4948 vmhalt= [KNL,S390] Perform z/VM CP command after system halt. 4949 Format: <command> 4950 4951 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic. 4952 Format: <command> 4953 4954 vmpoff= [KNL,S390] Perform z/VM CP command after power off. 4955 Format: <command> 4956 4957 vsyscall= [X86-64] 4958 Controls the behavior of vsyscalls (i.e. calls to 4959 fixed addresses of 0xffffffffff600x00 from legacy 4960 code). Most statically-linked binaries and older 4961 versions of glibc use these calls. Because these 4962 functions are at fixed addresses, they make nice 4963 targets for exploits that can control RIP. 4964 4965 emulate [default] Vsyscalls turn into traps and are 4966 emulated reasonably safely. 4967 4968 native Vsyscalls are native syscall instructions. 4969 This is a little bit faster than trapping 4970 and makes a few dynamic recompilers work 4971 better than they would in emulation mode. 4972 It also makes exploits much easier to write. 4973 4974 none Vsyscalls don't work at all. This makes 4975 them quite hard to use for exploits but 4976 might break your system. 4977 4978 vt.color= [VT] Default text color. 4979 Format: 0xYX, X = foreground, Y = background. 4980 Default: 0x07 = light gray on black. 4981 4982 vt.cur_default= [VT] Default cursor shape. 4983 Format: 0xCCBBAA, where AA, BB, and CC are the same as 4984 the parameters of the <Esc>[?A;B;Cc escape sequence; 4985 see VGA-softcursor.txt. Default: 2 = underline. 4986 4987 vt.default_blu= [VT] 4988 Format: <blue0>,<blue1>,<blue2>,...,<blue15> 4989 Change the default blue palette of the console. 4990 This is a 16-member array composed of values 4991 ranging from 0-255. 4992 4993 vt.default_grn= [VT] 4994 Format: <green0>,<green1>,<green2>,...,<green15> 4995 Change the default green palette of the console. 4996 This is a 16-member array composed of values 4997 ranging from 0-255. 4998 4999 vt.default_red= [VT] 5000 Format: <red0>,<red1>,<red2>,...,<red15> 5001 Change the default red palette of the console. 5002 This is a 16-member array composed of values 5003 ranging from 0-255. 5004 5005 vt.default_utf8= 5006 [VT] 5007 Format=<0|1> 5008 Set system-wide default UTF-8 mode for all tty's. 5009 Default is 1, i.e. UTF-8 mode is enabled for all 5010 newly opened terminals. 5011 5012 vt.global_cursor_default= 5013 [VT] 5014 Format=<-1|0|1> 5015 Set system-wide default for whether a cursor 5016 is shown on new VTs. Default is -1, 5017 i.e. cursors will be created by default unless 5018 overridden by individual drivers. 0 will hide 5019 cursors, 1 will display them. 5020 5021 vt.italic= [VT] Default color for italic text; 0-15. 5022 Default: 2 = green. 5023 5024 vt.underline= [VT] Default color for underlined text; 0-15. 5025 Default: 3 = cyan. 5026 5027 watchdog timers [HW,WDT] For information on watchdog timers, 5028 see Documentation/watchdog/watchdog-parameters.txt 5029 or other driver-specific files in the 5030 Documentation/watchdog/ directory. 5031 5032 workqueue.watchdog_thresh= 5033 If CONFIG_WQ_WATCHDOG is configured, workqueue can 5034 warn stall conditions and dump internal state to 5035 help debugging. 0 disables workqueue stall 5036 detection; otherwise, it's the stall threshold 5037 duration in seconds. The default value is 30 and 5038 it can be updated at runtime by writing to the 5039 corresponding sysfs file. 5040 5041 workqueue.disable_numa 5042 By default, all work items queued to unbound 5043 workqueues are affine to the NUMA nodes they're 5044 issued on, which results in better behavior in 5045 general. If NUMA affinity needs to be disabled for 5046 whatever reason, this option can be used. Note 5047 that this also can be controlled per-workqueue for 5048 workqueues visible under /sys/bus/workqueue/. 5049 5050 workqueue.power_efficient 5051 Per-cpu workqueues are generally preferred because 5052 they show better performance thanks to cache 5053 locality; unfortunately, per-cpu workqueues tend to 5054 be more power hungry than unbound workqueues. 5055 5056 Enabling this makes the per-cpu workqueues which 5057 were observed to contribute significantly to power 5058 consumption unbound, leading to measurably lower 5059 power usage at the cost of small performance 5060 overhead. 5061 5062 The default value of this parameter is determined by 5063 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT. 5064 5065 workqueue.debug_force_rr_cpu 5066 Workqueue used to implicitly guarantee that work 5067 items queued without explicit CPU specified are put 5068 on the local CPU. This guarantee is no longer true 5069 and while local CPU is still preferred work items 5070 may be put on foreign CPUs. This debug option 5071 forces round-robin CPU selection to flush out 5072 usages which depend on the now broken guarantee. 5073 When enabled, memory and cache locality will be 5074 impacted. 5075 5076 x2apic_phys [X86-64,APIC] Use x2apic physical mode instead of 5077 default x2apic cluster mode on platforms 5078 supporting x2apic. 5079 5080 x86_intel_mid_timer= [X86-32,APBT] 5081 Choose timer option for x86 Intel MID platform. 5082 Two valid options are apbt timer only and lapic timer 5083 plus one apbt timer for broadcast timer. 5084 x86_intel_mid_timer=apbt_only | lapic_and_apbt 5085 5086 xen_512gb_limit [KNL,X86-64,XEN] 5087 Restricts the kernel running paravirtualized under Xen 5088 to use only up to 512 GB of RAM. The reason to do so is 5089 crash analysis tools and Xen tools for doing domain 5090 save/restore/migration must be enabled to handle larger 5091 domains. 5092 5093 xen_emul_unplug= [HW,X86,XEN] 5094 Unplug Xen emulated devices 5095 Format: [unplug0,][unplug1] 5096 ide-disks -- unplug primary master IDE devices 5097 aux-ide-disks -- unplug non-primary-master IDE devices 5098 nics -- unplug network devices 5099 all -- unplug all emulated devices (NICs and IDE disks) 5100 unnecessary -- unplugging emulated devices is 5101 unnecessary even if the host did not respond to 5102 the unplug protocol 5103 never -- do not unplug even if version check succeeds 5104 5105 xen_nopvspin [X86,XEN] 5106 Disables the ticketlock slowpath using Xen PV 5107 optimizations. 5108 5109 xen_nopv [X86] 5110 Disables the PV optimizations forcing the HVM guest to 5111 run as generic HVM guest with no PV drivers. 5112 5113 xen_scrub_pages= [XEN] 5114 Boolean option to control scrubbing pages before giving them back 5115 to Xen, for use by other domains. Can be also changed at runtime 5116 with /sys/devices/system/xen_memory/xen_memory0/scrub_pages. 5117 Default value controlled with CONFIG_XEN_SCRUB_PAGES_DEFAULT. 5118 5119 xirc2ps_cs= [NET,PCMCIA] 5120 Format: 5121 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]] 5122 5123 xhci-hcd.quirks [USB,KNL] 5124 A hex value specifying bitmask with supplemental xhci 5125 host controller quirks. Meaning of each bit can be 5126 consulted in header drivers/usb/host/xhci.h. 5127