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 carrier_timeout= 465 [NET] Specifies amount of time (in seconds) that 466 the kernel should wait for a network carrier. By default 467 it waits 120 seconds. 468 469 ca_keys= [KEYS] This parameter identifies a specific key(s) on 470 the system trusted keyring to be used for certificate 471 trust validation. 472 format: { id:<keyid> | builtin } 473 474 cca= [MIPS] Override the kernel pages' cache coherency 475 algorithm. Accepted values range from 0 to 7 476 inclusive. See arch/mips/include/asm/pgtable-bits.h 477 for platform specific values (SB1, Loongson3 and 478 others). 479 480 ccw_timeout_log [S390] 481 See Documentation/s390/CommonIO for details. 482 483 cgroup_disable= [KNL] Disable a particular controller 484 Format: {name of the controller(s) to disable} 485 The effects of cgroup_disable=foo are: 486 - foo isn't auto-mounted if you mount all cgroups in 487 a single hierarchy 488 - foo isn't visible as an individually mountable 489 subsystem 490 {Currently only "memory" controller deal with this and 491 cut the overhead, others just disable the usage. So 492 only cgroup_disable=memory is actually worthy} 493 494 cgroup_no_v1= [KNL] Disable cgroup controllers and named hierarchies in v1 495 Format: { { controller | "all" | "named" } 496 [,{ controller | "all" | "named" }...] } 497 Like cgroup_disable, but only applies to cgroup v1; 498 the blacklisted controllers remain available in cgroup2. 499 "all" blacklists all controllers and "named" disables 500 named mounts. Specifying both "all" and "named" disables 501 all v1 hierarchies. 502 503 cgroup.memory= [KNL] Pass options to the cgroup memory controller. 504 Format: <string> 505 nosocket -- Disable socket memory accounting. 506 nokmem -- Disable kernel memory accounting. 507 508 checkreqprot [SELINUX] Set initial checkreqprot flag value. 509 Format: { "0" | "1" } 510 See security/selinux/Kconfig help text. 511 0 -- check protection applied by kernel (includes 512 any implied execute protection). 513 1 -- check protection requested by application. 514 Default value is set via a kernel config option. 515 Value can be changed at runtime via 516 /selinux/checkreqprot. 517 518 cio_ignore= [S390] 519 See Documentation/s390/CommonIO for details. 520 clk_ignore_unused 521 [CLK] 522 Prevents the clock framework from automatically gating 523 clocks that have not been explicitly enabled by a Linux 524 device driver but are enabled in hardware at reset or 525 by the bootloader/firmware. Note that this does not 526 force such clocks to be always-on nor does it reserve 527 those clocks in any way. This parameter is useful for 528 debug and development, but should not be needed on a 529 platform with proper driver support. For more 530 information, see Documentation/driver-api/clk.rst. 531 532 clock= [BUGS=X86-32, HW] gettimeofday clocksource override. 533 [Deprecated] 534 Forces specified clocksource (if available) to be used 535 when calculating gettimeofday(). If specified 536 clocksource is not available, it defaults to PIT. 537 Format: { pit | tsc | cyclone | pmtmr } 538 539 clocksource= Override the default clocksource 540 Format: <string> 541 Override the default clocksource and use the clocksource 542 with the name specified. 543 Some clocksource names to choose from, depending on 544 the platform: 545 [all] jiffies (this is the base, fallback clocksource) 546 [ACPI] acpi_pm 547 [ARM] imx_timer1,OSTS,netx_timer,mpu_timer2, 548 pxa_timer,timer3,32k_counter,timer0_1 549 [X86-32] pit,hpet,tsc; 550 scx200_hrt on Geode; cyclone on IBM x440 551 [MIPS] MIPS 552 [PARISC] cr16 553 [S390] tod 554 [SH] SuperH 555 [SPARC64] tick 556 [X86-64] hpet,tsc 557 558 clocksource.arm_arch_timer.evtstrm= 559 [ARM,ARM64] 560 Format: <bool> 561 Enable/disable the eventstream feature of the ARM 562 architected timer so that code using WFE-based polling 563 loops can be debugged more effectively on production 564 systems. 565 566 clearcpuid=BITNUM [X86] 567 Disable CPUID feature X for the kernel. See 568 arch/x86/include/asm/cpufeatures.h for the valid bit 569 numbers. Note the Linux specific bits are not necessarily 570 stable over kernel options, but the vendor specific 571 ones should be. 572 Also note that user programs calling CPUID directly 573 or using the feature without checking anything 574 will still see it. This just prevents it from 575 being used by the kernel or shown in /proc/cpuinfo. 576 Also note the kernel might malfunction if you disable 577 some critical bits. 578 579 cma=nn[MG]@[start[MG][-end[MG]]] 580 [ARM,X86,KNL] 581 Sets the size of kernel global memory area for 582 contiguous memory allocations and optionally the 583 placement constraint by the physical address range of 584 memory allocations. A value of 0 disables CMA 585 altogether. For more information, see 586 include/linux/dma-contiguous.h 587 588 cmo_free_hint= [PPC] Format: { yes | no } 589 Specify whether pages are marked as being inactive 590 when they are freed. This is used in CMO environments 591 to determine OS memory pressure for page stealing by 592 a hypervisor. 593 Default: yes 594 595 coherent_pool=nn[KMG] [ARM,KNL] 596 Sets the size of memory pool for coherent, atomic dma 597 allocations, by default set to 256K. 598 599 com20020= [HW,NET] ARCnet - COM20020 chipset 600 Format: 601 <io>[,<irq>[,<nodeID>[,<backplane>[,<ckp>[,<timeout>]]]]] 602 603 com90io= [HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers) 604 Format: <io>[,<irq>] 605 606 com90xx= [HW,NET] 607 ARCnet - COM90xx chipset (memory-mapped buffers) 608 Format: <io>[,<irq>[,<memstart>]] 609 610 condev= [HW,S390] console device 611 conmode= 612 613 console= [KNL] Output console device and options. 614 615 tty<n> Use the virtual console device <n>. 616 617 ttyS<n>[,options] 618 ttyUSB0[,options] 619 Use the specified serial port. The options are of 620 the form "bbbbpnf", where "bbbb" is the baud rate, 621 "p" is parity ("n", "o", or "e"), "n" is number of 622 bits, and "f" is flow control ("r" for RTS or 623 omit it). Default is "9600n8". 624 625 See Documentation/admin-guide/serial-console.rst for more 626 information. See 627 Documentation/networking/netconsole.txt for an 628 alternative. 629 630 uart[8250],io,<addr>[,options] 631 uart[8250],mmio,<addr>[,options] 632 uart[8250],mmio16,<addr>[,options] 633 uart[8250],mmio32,<addr>[,options] 634 uart[8250],0x<addr>[,options] 635 Start an early, polled-mode console on the 8250/16550 636 UART at the specified I/O port or MMIO address, 637 switching to the matching ttyS device later. 638 MMIO inter-register address stride is either 8-bit 639 (mmio), 16-bit (mmio16), or 32-bit (mmio32). 640 If none of [io|mmio|mmio16|mmio32], <addr> is assumed 641 to be equivalent to 'mmio'. 'options' are specified in 642 the same format described for ttyS above; if unspecified, 643 the h/w is not re-initialized. 644 645 hvc<n> Use the hypervisor console device <n>. This is for 646 both Xen and PowerPC hypervisors. 647 648 If the device connected to the port is not a TTY but a braille 649 device, prepend "brl," before the device type, for instance 650 console=brl,ttyS0 651 For now, only VisioBraille is supported. 652 653 console_msg_format= 654 [KNL] Change console messages format 655 default 656 By default we print messages on consoles in 657 "[time stamp] text\n" format (time stamp may not be 658 printed, depending on CONFIG_PRINTK_TIME or 659 `printk_time' param). 660 syslog 661 Switch to syslog format: "<%u>[time stamp] text\n" 662 IOW, each message will have a facility and loglevel 663 prefix. The format is similar to one used by syslog() 664 syscall, or to executing "dmesg -S --raw" or to reading 665 from /proc/kmsg. 666 667 consoleblank= [KNL] The console blank (screen saver) timeout in 668 seconds. A value of 0 disables the blank timer. 669 Defaults to 0. 670 671 coredump_filter= 672 [KNL] Change the default value for 673 /proc/<pid>/coredump_filter. 674 See also Documentation/filesystems/proc.txt. 675 676 coresight_cpu_debug.enable 677 [ARM,ARM64] 678 Format: <bool> 679 Enable/disable the CPU sampling based debugging. 680 0: default value, disable debugging 681 1: enable debugging at boot time 682 683 cpuidle.off=1 [CPU_IDLE] 684 disable the cpuidle sub-system 685 686 cpuidle.governor= 687 [CPU_IDLE] Name of the cpuidle governor to use. 688 689 cpufreq.off=1 [CPU_FREQ] 690 disable the cpufreq sub-system 691 692 cpu_init_udelay=N 693 [X86] Delay for N microsec between assert and de-assert 694 of APIC INIT to start processors. This delay occurs 695 on every CPU online, such as boot, and resume from suspend. 696 Default: 10000 697 698 cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver 699 Format: 700 <first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>] 701 702 crashkernel=size[KMG][@offset[KMG]] 703 [KNL] Using kexec, Linux can switch to a 'crash kernel' 704 upon panic. This parameter reserves the physical 705 memory region [offset, offset + size] for that kernel 706 image. If '@offset' is omitted, then a suitable offset 707 is selected automatically. Check 708 Documentation/kdump/kdump.txt for further details. 709 710 crashkernel=range1:size1[,range2:size2,...][@offset] 711 [KNL] Same as above, but depends on the memory 712 in the running system. The syntax of range is 713 start-[end] where start and end are both 714 a memory unit (amount[KMG]). See also 715 Documentation/kdump/kdump.txt for an example. 716 717 crashkernel=size[KMG],high 718 [KNL, x86_64] range could be above 4G. Allow kernel 719 to allocate physical memory region from top, so could 720 be above 4G if system have more than 4G ram installed. 721 Otherwise memory region will be allocated below 4G, if 722 available. 723 It will be ignored if crashkernel=X is specified. 724 crashkernel=size[KMG],low 725 [KNL, x86_64] range under 4G. When crashkernel=X,high 726 is passed, kernel could allocate physical memory region 727 above 4G, that cause second kernel crash on system 728 that require some amount of low memory, e.g. swiotlb 729 requires at least 64M+32K low memory, also enough extra 730 low memory is needed to make sure DMA buffers for 32-bit 731 devices won't run out. Kernel would try to allocate at 732 at least 256M below 4G automatically. 733 This one let user to specify own low range under 4G 734 for second kernel instead. 735 0: to disable low allocation. 736 It will be ignored when crashkernel=X,high is not used 737 or memory reserved is below 4G. 738 739 cryptomgr.notests 740 [KNL] Disable crypto self-tests 741 742 cs89x0_dma= [HW,NET] 743 Format: <dma> 744 745 cs89x0_media= [HW,NET] 746 Format: { rj45 | aui | bnc } 747 748 dasd= [HW,NET] 749 See header of drivers/s390/block/dasd_devmap.c. 750 751 db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port 752 (one device per port) 753 Format: <port#>,<type> 754 See also Documentation/input/devices/joystick-parport.rst 755 756 ddebug_query= [KNL,DYNAMIC_DEBUG] Enable debug messages at early boot 757 time. See 758 Documentation/admin-guide/dynamic-debug-howto.rst for 759 details. Deprecated, see dyndbg. 760 761 debug [KNL] Enable kernel debugging (events log level). 762 763 debug_boot_weak_hash 764 [KNL] Enable printing [hashed] pointers early in the 765 boot sequence. If enabled, we use a weak hash instead 766 of siphash to hash pointers. Use this option if you are 767 seeing instances of '(___ptrval___)') and need to see a 768 value (hashed pointer) instead. Cryptographically 769 insecure, please do not use on production kernels. 770 771 debug_locks_verbose= 772 [KNL] verbose self-tests 773 Format=<0|1> 774 Print debugging info while doing the locking API 775 self-tests. 776 We default to 0 (no extra messages), setting it to 777 1 will print _a lot_ more information - normally 778 only useful to kernel developers. 779 780 debug_objects [KNL] Enable object debugging 781 782 no_debug_objects 783 [KNL] Disable object debugging 784 785 debug_guardpage_minorder= 786 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this 787 parameter allows control of the order of pages that will 788 be intentionally kept free (and hence protected) by the 789 buddy allocator. Bigger value increase the probability 790 of catching random memory corruption, but reduce the 791 amount of memory for normal system use. The maximum 792 possible value is MAX_ORDER/2. Setting this parameter 793 to 1 or 2 should be enough to identify most random 794 memory corruption problems caused by bugs in kernel or 795 driver code when a CPU writes to (or reads from) a 796 random memory location. Note that there exists a class 797 of memory corruptions problems caused by buggy H/W or 798 F/W or by drivers badly programing DMA (basically when 799 memory is written at bus level and the CPU MMU is 800 bypassed) which are not detectable by 801 CONFIG_DEBUG_PAGEALLOC, hence this option will not help 802 tracking down these problems. 803 804 debug_pagealloc= 805 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this 806 parameter enables the feature at boot time. In 807 default, it is disabled. We can avoid allocating huge 808 chunk of memory for debug pagealloc if we don't enable 809 it at boot time and the system will work mostly same 810 with the kernel built without CONFIG_DEBUG_PAGEALLOC. 811 on: enable the feature 812 813 debugpat [X86] Enable PAT debugging 814 815 decnet.addr= [HW,NET] 816 Format: <area>[,<node>] 817 See also Documentation/networking/decnet.txt. 818 819 default_hugepagesz= 820 [same as hugepagesz=] The size of the default 821 HugeTLB page size. This is the size represented by 822 the legacy /proc/ hugepages APIs, used for SHM, and 823 default size when mounting hugetlbfs filesystems. 824 Defaults to the default architecture's huge page size 825 if not specified. 826 827 deferred_probe_timeout= 828 [KNL] Debugging option to set a timeout in seconds for 829 deferred probe to give up waiting on dependencies to 830 probe. Only specific dependencies (subsystems or 831 drivers) that have opted in will be ignored. A timeout of 0 832 will timeout at the end of initcalls. This option will also 833 dump out devices still on the deferred probe list after 834 retrying. 835 836 dhash_entries= [KNL] 837 Set number of hash buckets for dentry cache. 838 839 disable_1tb_segments [PPC] 840 Disables the use of 1TB hash page table segments. This 841 causes the kernel to fall back to 256MB segments which 842 can be useful when debugging issues that require an SLB 843 miss to occur. 844 845 disable= [IPV6] 846 See Documentation/networking/ipv6.txt. 847 848 hardened_usercopy= 849 [KNL] Under CONFIG_HARDENED_USERCOPY, whether 850 hardening is enabled for this boot. Hardened 851 usercopy checking is used to protect the kernel 852 from reading or writing beyond known memory 853 allocation boundaries as a proactive defense 854 against bounds-checking flaws in the kernel's 855 copy_to_user()/copy_from_user() interface. 856 on Perform hardened usercopy checks (default). 857 off Disable hardened usercopy checks. 858 859 disable_radix [PPC] 860 Disable RADIX MMU mode on POWER9 861 862 disable_cpu_apicid= [X86,APIC,SMP] 863 Format: <int> 864 The number of initial APIC ID for the 865 corresponding CPU to be disabled at boot, 866 mostly used for the kdump 2nd kernel to 867 disable BSP to wake up multiple CPUs without 868 causing system reset or hang due to sending 869 INIT from AP to BSP. 870 871 perf_v4_pmi= [X86,INTEL] 872 Format: <bool> 873 Disable Intel PMU counter freezing feature. 874 The feature only exists starting from 875 Arch Perfmon v4 (Skylake and newer). 876 877 disable_ddw [PPC/PSERIES] 878 Disable Dynamic DMA Window support. Use this if 879 to workaround buggy firmware. 880 881 disable_ipv6= [IPV6] 882 See Documentation/networking/ipv6.txt. 883 884 disable_mtrr_cleanup [X86] 885 The kernel tries to adjust MTRR layout from continuous 886 to discrete, to make X server driver able to add WB 887 entry later. This parameter disables that. 888 889 disable_mtrr_trim [X86, Intel and AMD only] 890 By default the kernel will trim any uncacheable 891 memory out of your available memory pool based on 892 MTRR settings. This parameter disables that behavior, 893 possibly causing your machine to run very slowly. 894 895 disable_timer_pin_1 [X86] 896 Disable PIN 1 of APIC timer 897 Can be useful to work around chipset bugs. 898 899 dis_ucode_ldr [X86] Disable the microcode loader. 900 901 dma_debug=off If the kernel is compiled with DMA_API_DEBUG support, 902 this option disables the debugging code at boot. 903 904 dma_debug_entries=<number> 905 This option allows to tune the number of preallocated 906 entries for DMA-API debugging code. One entry is 907 required per DMA-API allocation. Use this if the 908 DMA-API debugging code disables itself because the 909 architectural default is too low. 910 911 dma_debug_driver=<driver_name> 912 With this option the DMA-API debugging driver 913 filter feature can be enabled at boot time. Just 914 pass the driver to filter for as the parameter. 915 The filter can be disabled or changed to another 916 driver later using sysfs. 917 918 driver_async_probe= [KNL] 919 List of driver names to be probed asynchronously. 920 Format: <driver_name1>,<driver_name2>... 921 922 drm.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>] 923 Broken monitors, graphic adapters, KVMs and EDIDless 924 panels may send no or incorrect EDID data sets. 925 This parameter allows to specify an EDID data sets 926 in the /lib/firmware directory that are used instead. 927 Generic built-in EDID data sets are used, if one of 928 edid/1024x768.bin, edid/1280x1024.bin, 929 edid/1680x1050.bin, or edid/1920x1080.bin is given 930 and no file with the same name exists. Details and 931 instructions how to build your own EDID data are 932 available in Documentation/EDID/HOWTO.txt. An EDID 933 data set will only be used for a particular connector, 934 if its name and a colon are prepended to the EDID 935 name. Each connector may use a unique EDID data 936 set by separating the files with a comma. An EDID 937 data set with no connector name will be used for 938 any connectors not explicitly specified. 939 940 dscc4.setup= [NET] 941 942 dt_cpu_ftrs= [PPC] 943 Format: {"off" | "known"} 944 Control how the dt_cpu_ftrs device-tree binding is 945 used for CPU feature discovery and setup (if it 946 exists). 947 off: Do not use it, fall back to legacy cpu table. 948 known: Do not pass through unknown features to guests 949 or userspace, only those that the kernel is aware of. 950 951 dump_apple_properties [X86] 952 Dump name and content of EFI device properties on 953 x86 Macs. Useful for driver authors to determine 954 what data is available or for reverse-engineering. 955 956 dyndbg[="val"] [KNL,DYNAMIC_DEBUG] 957 module.dyndbg[="val"] 958 Enable debug messages at boot time. See 959 Documentation/admin-guide/dynamic-debug-howto.rst 960 for details. 961 962 nompx [X86] Disables Intel Memory Protection Extensions. 963 See Documentation/x86/intel_mpx.txt for more 964 information about the feature. 965 966 nopku [X86] Disable Memory Protection Keys CPU feature found 967 in some Intel CPUs. 968 969 module.async_probe [KNL] 970 Enable asynchronous probe on this module. 971 972 early_ioremap_debug [KNL] 973 Enable debug messages in early_ioremap support. This 974 is useful for tracking down temporary early mappings 975 which are not unmapped. 976 977 earlycon= [KNL] Output early console device and options. 978 979 [ARM64] The early console is determined by the 980 stdout-path property in device tree's chosen node, 981 or determined by the ACPI SPCR table. 982 983 [X86] When used with no options the early console is 984 determined by the ACPI SPCR table. 985 986 cdns,<addr>[,options] 987 Start an early, polled-mode console on a Cadence 988 (xuartps) serial port at the specified address. Only 989 supported option is baud rate. If baud rate is not 990 specified, the serial port must already be setup and 991 configured. 992 993 uart[8250],io,<addr>[,options] 994 uart[8250],mmio,<addr>[,options] 995 uart[8250],mmio32,<addr>[,options] 996 uart[8250],mmio32be,<addr>[,options] 997 uart[8250],0x<addr>[,options] 998 Start an early, polled-mode console on the 8250/16550 999 UART at the specified I/O port or MMIO address. 1000 MMIO inter-register address stride is either 8-bit 1001 (mmio) or 32-bit (mmio32 or mmio32be). 1002 If none of [io|mmio|mmio32|mmio32be], <addr> is assumed 1003 to be equivalent to 'mmio'. 'options' are specified 1004 in the same format described for "console=ttyS<n>"; if 1005 unspecified, the h/w is not initialized. 1006 1007 pl011,<addr> 1008 pl011,mmio32,<addr> 1009 Start an early, polled-mode console on a pl011 serial 1010 port at the specified address. The pl011 serial port 1011 must already be setup and configured. Options are not 1012 yet supported. If 'mmio32' is specified, then only 1013 the driver will use only 32-bit accessors to read/write 1014 the device registers. 1015 1016 meson,<addr> 1017 Start an early, polled-mode console on a meson serial 1018 port at the specified address. The serial port must 1019 already be setup and configured. Options are not yet 1020 supported. 1021 1022 msm_serial,<addr> 1023 Start an early, polled-mode console on an msm serial 1024 port at the specified address. The serial port 1025 must already be setup and configured. Options are not 1026 yet supported. 1027 1028 msm_serial_dm,<addr> 1029 Start an early, polled-mode console on an msm serial 1030 dm port at the specified address. The serial port 1031 must already be setup and configured. Options are not 1032 yet supported. 1033 1034 owl,<addr> 1035 Start an early, polled-mode console on a serial port 1036 of an Actions Semi SoC, such as S500 or S900, at the 1037 specified address. The serial port must already be 1038 setup and configured. Options are not yet supported. 1039 1040 rda,<addr> 1041 Start an early, polled-mode console on a serial port 1042 of an RDA Micro SoC, such as RDA8810PL, at the 1043 specified address. The serial port must already be 1044 setup and configured. Options are not yet supported. 1045 1046 smh Use ARM semihosting calls for early console. 1047 1048 s3c2410,<addr> 1049 s3c2412,<addr> 1050 s3c2440,<addr> 1051 s3c6400,<addr> 1052 s5pv210,<addr> 1053 exynos4210,<addr> 1054 Use early console provided by serial driver available 1055 on Samsung SoCs, requires selecting proper type and 1056 a correct base address of the selected UART port. The 1057 serial port must already be setup and configured. 1058 Options are not yet supported. 1059 1060 lantiq,<addr> 1061 Start an early, polled-mode console on a lantiq serial 1062 (lqasc) port at the specified address. The serial port 1063 must already be setup and configured. Options are not 1064 yet supported. 1065 1066 lpuart,<addr> 1067 lpuart32,<addr> 1068 Use early console provided by Freescale LP UART driver 1069 found on Freescale Vybrid and QorIQ LS1021A processors. 1070 A valid base address must be provided, and the serial 1071 port must already be setup and configured. 1072 1073 ar3700_uart,<addr> 1074 Start an early, polled-mode console on the 1075 Armada 3700 serial port at the specified 1076 address. The serial port must already be setup 1077 and configured. Options are not yet supported. 1078 1079 qcom_geni,<addr> 1080 Start an early, polled-mode console on a Qualcomm 1081 Generic Interface (GENI) based serial port at the 1082 specified address. The serial port must already be 1083 setup and configured. Options are not yet supported. 1084 1085 efifb,[options] 1086 Start an early, unaccelerated console on the EFI 1087 memory mapped framebuffer (if available). On cache 1088 coherent non-x86 systems that use system memory for 1089 the framebuffer, pass the 'ram' option so that it is 1090 mapped with the correct attributes. 1091 1092 earlyprintk= [X86,SH,ARM,M68k,S390] 1093 earlyprintk=vga 1094 earlyprintk=sclp 1095 earlyprintk=xen 1096 earlyprintk=serial[,ttySn[,baudrate]] 1097 earlyprintk=serial[,0x...[,baudrate]] 1098 earlyprintk=ttySn[,baudrate] 1099 earlyprintk=dbgp[debugController#] 1100 earlyprintk=pciserial[,force],bus:device.function[,baudrate] 1101 earlyprintk=xdbc[xhciController#] 1102 1103 earlyprintk is useful when the kernel crashes before 1104 the normal console is initialized. It is not enabled by 1105 default because it has some cosmetic problems. 1106 1107 Append ",keep" to not disable it when the real console 1108 takes over. 1109 1110 Only one of vga, efi, serial, or usb debug port can 1111 be used at a time. 1112 1113 Currently only ttyS0 and ttyS1 may be specified by 1114 name. Other I/O ports may be explicitly specified 1115 on some architectures (x86 and arm at least) by 1116 replacing ttySn with an I/O port address, like this: 1117 earlyprintk=serial,0x1008,115200 1118 You can find the port for a given device in 1119 /proc/tty/driver/serial: 1120 2: uart:ST16650V2 port:00001008 irq:18 ... 1121 1122 Interaction with the standard serial driver is not 1123 very good. 1124 1125 The VGA and EFI output is eventually overwritten by 1126 the real console. 1127 1128 The xen output can only be used by Xen PV guests. 1129 1130 The sclp output can only be used on s390. 1131 1132 The optional "force" to "pciserial" enables use of a 1133 PCI device even when its classcode is not of the 1134 UART class. 1135 1136 edac_report= [HW,EDAC] Control how to report EDAC event 1137 Format: {"on" | "off" | "force"} 1138 on: enable EDAC to report H/W event. May be overridden 1139 by other higher priority error reporting module. 1140 off: disable H/W event reporting through EDAC. 1141 force: enforce the use of EDAC to report H/W event. 1142 default: on. 1143 1144 ekgdboc= [X86,KGDB] Allow early kernel console debugging 1145 ekgdboc=kbd 1146 1147 This is designed to be used in conjunction with 1148 the boot argument: earlyprintk=vga 1149 1150 edd= [EDD] 1151 Format: {"off" | "on" | "skip[mbr]"} 1152 1153 efi= [EFI] 1154 Format: { "old_map", "nochunk", "noruntime", "debug" } 1155 old_map [X86-64]: switch to the old ioremap-based EFI 1156 runtime services mapping. 32-bit still uses this one by 1157 default. 1158 nochunk: disable reading files in "chunks" in the EFI 1159 boot stub, as chunking can cause problems with some 1160 firmware implementations. 1161 noruntime : disable EFI runtime services support 1162 debug: enable misc debug output 1163 1164 efi_no_storage_paranoia [EFI; X86] 1165 Using this parameter you can use more than 50% of 1166 your efi variable storage. Use this parameter only if 1167 you are really sure that your UEFI does sane gc and 1168 fulfills the spec otherwise your board may brick. 1169 1170 efi_fake_mem= nn[KMG]@ss[KMG]:aa[,nn[KMG]@ss[KMG]:aa,..] [EFI; X86] 1171 Add arbitrary attribute to specific memory range by 1172 updating original EFI memory map. 1173 Region of memory which aa attribute is added to is 1174 from ss to ss+nn. 1175 If efi_fake_mem=2G@4G:0x10000,2G@0x10a0000000:0x10000 1176 is specified, EFI_MEMORY_MORE_RELIABLE(0x10000) 1177 attribute is added to range 0x100000000-0x180000000 and 1178 0x10a0000000-0x1120000000. 1179 1180 Using this parameter you can do debugging of EFI memmap 1181 related feature. For example, you can do debugging of 1182 Address Range Mirroring feature even if your box 1183 doesn't support it. 1184 1185 efivar_ssdt= [EFI; X86] Name of an EFI variable that contains an SSDT 1186 that is to be dynamically loaded by Linux. If there are 1187 multiple variables with the same name but with different 1188 vendor GUIDs, all of them will be loaded. See 1189 Documentation/acpi/ssdt-overlays.txt for details. 1190 1191 1192 eisa_irq_edge= [PARISC,HW] 1193 See header of drivers/parisc/eisa.c. 1194 1195 elanfreq= [X86-32] 1196 See comment before function elanfreq_setup() in 1197 arch/x86/kernel/cpu/cpufreq/elanfreq.c. 1198 1199 elevator= [IOSCHED] 1200 Format: { "mq-deadline" | "kyber" | "bfq" } 1201 See Documentation/block/deadline-iosched.txt, 1202 Documentation/block/kyber-iosched.txt and 1203 Documentation/block/bfq-iosched.txt for details. 1204 1205 elfcorehdr=[size[KMG]@]offset[KMG] [IA64,PPC,SH,X86,S390] 1206 Specifies physical address of start of kernel core 1207 image elf header and optionally the size. Generally 1208 kexec loader will pass this option to capture kernel. 1209 See Documentation/kdump/kdump.txt for details. 1210 1211 enable_mtrr_cleanup [X86] 1212 The kernel tries to adjust MTRR layout from continuous 1213 to discrete, to make X server driver able to add WB 1214 entry later. This parameter enables that. 1215 1216 enable_timer_pin_1 [X86] 1217 Enable PIN 1 of APIC timer 1218 Can be useful to work around chipset bugs 1219 (in particular on some ATI chipsets). 1220 The kernel tries to set a reasonable default. 1221 1222 enforcing [SELINUX] Set initial enforcing status. 1223 Format: {"0" | "1"} 1224 See security/selinux/Kconfig help text. 1225 0 -- permissive (log only, no denials). 1226 1 -- enforcing (deny and log). 1227 Default value is 0. 1228 Value can be changed at runtime via /selinux/enforce. 1229 1230 erst_disable [ACPI] 1231 Disable Error Record Serialization Table (ERST) 1232 support. 1233 1234 ether= [HW,NET] Ethernet cards parameters 1235 This option is obsoleted by the "netdev=" option, which 1236 has equivalent usage. See its documentation for details. 1237 1238 evm= [EVM] 1239 Format: { "fix" } 1240 Permit 'security.evm' to be updated regardless of 1241 current integrity status. 1242 1243 failslab= 1244 fail_page_alloc= 1245 fail_make_request=[KNL] 1246 General fault injection mechanism. 1247 Format: <interval>,<probability>,<space>,<times> 1248 See also Documentation/fault-injection/. 1249 1250 floppy= [HW] 1251 See Documentation/blockdev/floppy.txt. 1252 1253 force_pal_cache_flush 1254 [IA-64] Avoid check_sal_cache_flush which may hang on 1255 buggy SAL_CACHE_FLUSH implementations. Using this 1256 parameter will force ia64_sal_cache_flush to call 1257 ia64_pal_cache_flush instead of SAL_CACHE_FLUSH. 1258 1259 forcepae [X86-32] 1260 Forcefully enable Physical Address Extension (PAE). 1261 Many Pentium M systems disable PAE but may have a 1262 functionally usable PAE implementation. 1263 Warning: use of this parameter will taint the kernel 1264 and may cause unknown problems. 1265 1266 ftrace=[tracer] 1267 [FTRACE] will set and start the specified tracer 1268 as early as possible in order to facilitate early 1269 boot debugging. 1270 1271 ftrace_dump_on_oops[=orig_cpu] 1272 [FTRACE] will dump the trace buffers on oops. 1273 If no parameter is passed, ftrace will dump 1274 buffers of all CPUs, but if you pass orig_cpu, it will 1275 dump only the buffer of the CPU that triggered the 1276 oops. 1277 1278 ftrace_filter=[function-list] 1279 [FTRACE] Limit the functions traced by the function 1280 tracer at boot up. function-list is a comma separated 1281 list of functions. This list can be changed at run 1282 time by the set_ftrace_filter file in the debugfs 1283 tracing directory. 1284 1285 ftrace_notrace=[function-list] 1286 [FTRACE] Do not trace the functions specified in 1287 function-list. This list can be changed at run time 1288 by the set_ftrace_notrace file in the debugfs 1289 tracing directory. 1290 1291 ftrace_graph_filter=[function-list] 1292 [FTRACE] Limit the top level callers functions traced 1293 by the function graph tracer at boot up. 1294 function-list is a comma separated list of functions 1295 that can be changed at run time by the 1296 set_graph_function file in the debugfs tracing directory. 1297 1298 ftrace_graph_notrace=[function-list] 1299 [FTRACE] Do not trace from the functions specified in 1300 function-list. This list is a comma separated list of 1301 functions that can be changed at run time by the 1302 set_graph_notrace file in the debugfs tracing directory. 1303 1304 ftrace_graph_max_depth=<uint> 1305 [FTRACE] Used with the function graph tracer. This is 1306 the max depth it will trace into a function. This value 1307 can be changed at run time by the max_graph_depth file 1308 in the tracefs tracing directory. default: 0 (no limit) 1309 1310 gamecon.map[2|3]= 1311 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad 1312 support via parallel port (up to 5 devices per port) 1313 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5> 1314 See also Documentation/input/devices/joystick-parport.rst 1315 1316 gamma= [HW,DRM] 1317 1318 gart_fix_e820= [X86_64] disable the fix e820 for K8 GART 1319 Format: off | on 1320 default: on 1321 1322 gcov_persist= [GCOV] When non-zero (default), profiling data for 1323 kernel modules is saved and remains accessible via 1324 debugfs, even when the module is unloaded/reloaded. 1325 When zero, profiling data is discarded and associated 1326 debugfs files are removed at module unload time. 1327 1328 goldfish [X86] Enable the goldfish android emulator platform. 1329 Don't use this when you are not running on the 1330 android emulator 1331 1332 gpt [EFI] Forces disk with valid GPT signature but 1333 invalid Protective MBR to be treated as GPT. If the 1334 primary GPT is corrupted, it enables the backup/alternate 1335 GPT to be used instead. 1336 1337 grcan.enable0= [HW] Configuration of physical interface 0. Determines 1338 the "Enable 0" bit of the configuration register. 1339 Format: 0 | 1 1340 Default: 0 1341 grcan.enable1= [HW] Configuration of physical interface 1. Determines 1342 the "Enable 0" bit of the configuration register. 1343 Format: 0 | 1 1344 Default: 0 1345 grcan.select= [HW] Select which physical interface to use. 1346 Format: 0 | 1 1347 Default: 0 1348 grcan.txsize= [HW] Sets the size of the tx buffer. 1349 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0. 1350 Default: 1024 1351 grcan.rxsize= [HW] Sets the size of the rx buffer. 1352 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0. 1353 Default: 1024 1354 1355 gpio-mockup.gpio_mockup_ranges 1356 [HW] Sets the ranges of gpiochip of for this device. 1357 Format: <start1>,<end1>,<start2>,<end2>... 1358 1359 hardlockup_all_cpu_backtrace= 1360 [KNL] Should the hard-lockup detector generate 1361 backtraces on all cpus. 1362 Format: <integer> 1363 1364 hashdist= [KNL,NUMA] Large hashes allocated during boot 1365 are distributed across NUMA nodes. Defaults on 1366 for 64-bit NUMA, off otherwise. 1367 Format: 0 | 1 (for off | on) 1368 1369 hcl= [IA-64] SGI's Hardware Graph compatibility layer 1370 1371 hd= [EIDE] (E)IDE hard drive subsystem geometry 1372 Format: <cyl>,<head>,<sect> 1373 1374 hest_disable [ACPI] 1375 Disable Hardware Error Source Table (HEST) support; 1376 corresponding firmware-first mode error processing 1377 logic will be disabled. 1378 1379 highmem=nn[KMG] [KNL,BOOT] forces the highmem zone to have an exact 1380 size of <nn>. This works even on boxes that have no 1381 highmem otherwise. This also works to reduce highmem 1382 size on bigger boxes. 1383 1384 highres= [KNL] Enable/disable high resolution timer mode. 1385 Valid parameters: "on", "off" 1386 Default: "on" 1387 1388 hisax= [HW,ISDN] 1389 See Documentation/isdn/README.HiSax. 1390 1391 hlt [BUGS=ARM,SH] 1392 1393 hpet= [X86-32,HPET] option to control HPET usage 1394 Format: { enable (default) | disable | force | 1395 verbose } 1396 disable: disable HPET and use PIT instead 1397 force: allow force enabled of undocumented chips (ICH4, 1398 VIA, nVidia) 1399 verbose: show contents of HPET registers during setup 1400 1401 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET 1402 registers. Default set by CONFIG_HPET_MMAP_DEFAULT. 1403 1404 hugepages= [HW,X86-32,IA-64] HugeTLB pages to allocate at boot. 1405 hugepagesz= [HW,IA-64,PPC,X86-64] The size of the HugeTLB pages. 1406 On x86-64 and powerpc, this option can be specified 1407 multiple times interleaved with hugepages= to reserve 1408 huge pages of different sizes. Valid pages sizes on 1409 x86-64 are 2M (when the CPU supports "pse") and 1G 1410 (when the CPU supports the "pdpe1gb" cpuinfo flag). 1411 1412 hung_task_panic= 1413 [KNL] Should the hung task detector generate panics. 1414 Format: <integer> 1415 1416 A nonzero value instructs the kernel to panic when a 1417 hung task is detected. The default value is controlled 1418 by the CONFIG_BOOTPARAM_HUNG_TASK_PANIC build-time 1419 option. The value selected by this boot parameter can 1420 be changed later by the kernel.hung_task_panic sysctl. 1421 1422 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC) 1423 terminal devices. Valid values: 0..8 1424 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs. 1425 If specified, z/VM IUCV HVC accepts connections 1426 from listed z/VM user IDs only. 1427 1428 hv_nopvspin [X86,HYPER_V] Disables the paravirt spinlock optimizations 1429 which allow the hypervisor to 'idle' the 1430 guest on lock contention. 1431 1432 keep_bootcon [KNL] 1433 Do not unregister boot console at start. This is only 1434 useful for debugging when something happens in the window 1435 between unregistering the boot console and initializing 1436 the real console. 1437 1438 i2c_bus= [HW] Override the default board specific I2C bus speed 1439 or register an additional I2C bus that is not 1440 registered from board initialization code. 1441 Format: 1442 <bus_id>,<clkrate> 1443 1444 i8042.debug [HW] Toggle i8042 debug mode 1445 i8042.unmask_kbd_data 1446 [HW] Enable printing of interrupt data from the KBD port 1447 (disabled by default, and as a pre-condition 1448 requires that i8042.debug=1 be enabled) 1449 i8042.direct [HW] Put keyboard port into non-translated mode 1450 i8042.dumbkbd [HW] Pretend that controller can only read data from 1451 keyboard and cannot control its state 1452 (Don't attempt to blink the leds) 1453 i8042.noaux [HW] Don't check for auxiliary (== mouse) port 1454 i8042.nokbd [HW] Don't check/create keyboard port 1455 i8042.noloop [HW] Disable the AUX Loopback command while probing 1456 for the AUX port 1457 i8042.nomux [HW] Don't check presence of an active multiplexing 1458 controller 1459 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX 1460 controllers 1461 i8042.notimeout [HW] Ignore timeout condition signalled by controller 1462 i8042.reset [HW] Reset the controller during init, cleanup and 1463 suspend-to-ram transitions, only during s2r 1464 transitions, or never reset 1465 Format: { 1 | Y | y | 0 | N | n } 1466 1, Y, y: always reset controller 1467 0, N, n: don't ever reset controller 1468 Default: only on s2r transitions on x86; most other 1469 architectures force reset to be always executed 1470 i8042.unlock [HW] Unlock (ignore) the keylock 1471 i8042.kbdreset [HW] Reset device connected to KBD port 1472 1473 i810= [HW,DRM] 1474 1475 i8k.ignore_dmi [HW] Continue probing hardware even if DMI data 1476 indicates that the driver is running on unsupported 1477 hardware. 1478 i8k.force [HW] Activate i8k driver even if SMM BIOS signature 1479 does not match list of supported models. 1480 i8k.power_status 1481 [HW] Report power status in /proc/i8k 1482 (disabled by default) 1483 i8k.restricted [HW] Allow controlling fans only if SYS_ADMIN 1484 capability is set. 1485 1486 i915.invert_brightness= 1487 [DRM] Invert the sense of the variable that is used to 1488 set the brightness of the panel backlight. Normally a 1489 brightness value of 0 indicates backlight switched off, 1490 and the maximum of the brightness value sets the backlight 1491 to maximum brightness. If this parameter is set to 0 1492 (default) and the machine requires it, or this parameter 1493 is set to 1, a brightness value of 0 sets the backlight 1494 to maximum brightness, and the maximum of the brightness 1495 value switches the backlight off. 1496 -1 -- never invert brightness 1497 0 -- machine default 1498 1 -- force brightness inversion 1499 1500 icn= [HW,ISDN] 1501 Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]] 1502 1503 ide-core.nodma= [HW] (E)IDE subsystem 1504 Format: =0.0 to prevent dma on hda, =0.1 hdb =1.0 hdc 1505 .vlb_clock .pci_clock .noflush .nohpa .noprobe .nowerr 1506 .cdrom .chs .ignore_cable are additional options 1507 See Documentation/ide/ide.txt. 1508 1509 ide-generic.probe-mask= [HW] (E)IDE subsystem 1510 Format: <int> 1511 Probe mask for legacy ISA IDE ports. Depending on 1512 platform up to 6 ports are supported, enabled by 1513 setting corresponding bits in the mask to 1. The 1514 default value is 0x0, which has a special meaning. 1515 On systems that have PCI, it triggers scanning the 1516 PCI bus for the first and the second port, which 1517 are then probed. On systems without PCI the value 1518 of 0x0 enables probing the two first ports as if it 1519 was 0x3. 1520 1521 ide-pci-generic.all-generic-ide [HW] (E)IDE subsystem 1522 Claim all unknown PCI IDE storage controllers. 1523 1524 idle= [X86] 1525 Format: idle=poll, idle=halt, idle=nomwait 1526 Poll forces a polling idle loop that can slightly 1527 improve the performance of waking up a idle CPU, but 1528 will use a lot of power and make the system run hot. 1529 Not recommended. 1530 idle=halt: Halt is forced to be used for CPU idle. 1531 In such case C2/C3 won't be used again. 1532 idle=nomwait: Disable mwait for CPU C-states 1533 1534 ieee754= [MIPS] Select IEEE Std 754 conformance mode 1535 Format: { strict | legacy | 2008 | relaxed } 1536 Default: strict 1537 1538 Choose which programs will be accepted for execution 1539 based on the IEEE 754 NaN encoding(s) supported by 1540 the FPU and the NaN encoding requested with the value 1541 of an ELF file header flag individually set by each 1542 binary. Hardware implementations are permitted to 1543 support either or both of the legacy and the 2008 NaN 1544 encoding mode. 1545 1546 Available settings are as follows: 1547 strict accept binaries that request a NaN encoding 1548 supported by the FPU 1549 legacy only accept legacy-NaN binaries, if supported 1550 by the FPU 1551 2008 only accept 2008-NaN binaries, if supported 1552 by the FPU 1553 relaxed accept any binaries regardless of whether 1554 supported by the FPU 1555 1556 The FPU emulator is always able to support both NaN 1557 encodings, so if no FPU hardware is present or it has 1558 been disabled with 'nofpu', then the settings of 1559 'legacy' and '2008' strap the emulator accordingly, 1560 'relaxed' straps the emulator for both legacy-NaN and 1561 2008-NaN, whereas 'strict' enables legacy-NaN only on 1562 legacy processors and both NaN encodings on MIPS32 or 1563 MIPS64 CPUs. 1564 1565 The setting for ABS.fmt/NEG.fmt instruction execution 1566 mode generally follows that for the NaN encoding, 1567 except where unsupported by hardware. 1568 1569 ignore_loglevel [KNL] 1570 Ignore loglevel setting - this will print /all/ 1571 kernel messages to the console. Useful for debugging. 1572 We also add it as printk module parameter, so users 1573 could change it dynamically, usually by 1574 /sys/module/printk/parameters/ignore_loglevel. 1575 1576 ignore_rlimit_data 1577 Ignore RLIMIT_DATA setting for data mappings, 1578 print warning at first misuse. Can be changed via 1579 /sys/module/kernel/parameters/ignore_rlimit_data. 1580 1581 ihash_entries= [KNL] 1582 Set number of hash buckets for inode cache. 1583 1584 ima_appraise= [IMA] appraise integrity measurements 1585 Format: { "off" | "enforce" | "fix" | "log" } 1586 default: "enforce" 1587 1588 ima_appraise_tcb [IMA] 1589 The builtin appraise policy appraises all files 1590 owned by uid=0. 1591 1592 ima_canonical_fmt [IMA] 1593 Use the canonical format for the binary runtime 1594 measurements, instead of host native format. 1595 1596 ima_hash= [IMA] 1597 Format: { md5 | sha1 | rmd160 | sha256 | sha384 1598 | sha512 | ... } 1599 default: "sha1" 1600 1601 The list of supported hash algorithms is defined 1602 in crypto/hash_info.h. 1603 1604 ima_policy= [IMA] 1605 The builtin policies to load during IMA setup. 1606 Format: "tcb | appraise_tcb | secure_boot | 1607 fail_securely" 1608 1609 The "tcb" policy measures all programs exec'd, files 1610 mmap'd for exec, and all files opened with the read 1611 mode bit set by either the effective uid (euid=0) or 1612 uid=0. 1613 1614 The "appraise_tcb" policy appraises the integrity of 1615 all files owned by root. (This is the equivalent 1616 of ima_appraise_tcb.) 1617 1618 The "secure_boot" policy appraises the integrity 1619 of files (eg. kexec kernel image, kernel modules, 1620 firmware, policy, etc) based on file signatures. 1621 1622 The "fail_securely" policy forces file signature 1623 verification failure also on privileged mounted 1624 filesystems with the SB_I_UNVERIFIABLE_SIGNATURE 1625 flag. 1626 1627 ima_tcb [IMA] Deprecated. Use ima_policy= instead. 1628 Load a policy which meets the needs of the Trusted 1629 Computing Base. This means IMA will measure all 1630 programs exec'd, files mmap'd for exec, and all files 1631 opened for read by uid=0. 1632 1633 ima_template= [IMA] 1634 Select one of defined IMA measurements template formats. 1635 Formats: { "ima" | "ima-ng" | "ima-sig" } 1636 Default: "ima-ng" 1637 1638 ima_template_fmt= 1639 [IMA] Define a custom template format. 1640 Format: { "field1|...|fieldN" } 1641 1642 ima.ahash_minsize= [IMA] Minimum file size for asynchronous hash usage 1643 Format: <min_file_size> 1644 Set the minimal file size for using asynchronous hash. 1645 If left unspecified, ahash usage is disabled. 1646 1647 ahash performance varies for different data sizes on 1648 different crypto accelerators. This option can be used 1649 to achieve the best performance for a particular HW. 1650 1651 ima.ahash_bufsize= [IMA] Asynchronous hash buffer size 1652 Format: <bufsize> 1653 Set hashing buffer size. Default: 4k. 1654 1655 ahash performance varies for different chunk sizes on 1656 different crypto accelerators. This option can be used 1657 to achieve best performance for particular HW. 1658 1659 init= [KNL] 1660 Format: <full_path> 1661 Run specified binary instead of /sbin/init as init 1662 process. 1663 1664 initcall_debug [KNL] Trace initcalls as they are executed. Useful 1665 for working out where the kernel is dying during 1666 startup. 1667 1668 initcall_blacklist= [KNL] Do not execute a comma-separated list of 1669 initcall functions. Useful for debugging built-in 1670 modules and initcalls. 1671 1672 initrd= [BOOT] Specify the location of the initial ramdisk 1673 1674 init_pkru= [x86] Specify the default memory protection keys rights 1675 register contents for all processes. 0x55555554 by 1676 default (disallow access to all but pkey 0). Can 1677 override in debugfs after boot. 1678 1679 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver 1680 Format: <irq> 1681 1682 int_pln_enable [x86] Enable power limit notification interrupt 1683 1684 integrity_audit=[IMA] 1685 Format: { "0" | "1" } 1686 0 -- basic integrity auditing messages. (Default) 1687 1 -- additional integrity auditing messages. 1688 1689 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option 1690 on 1691 Enable intel iommu driver. 1692 off 1693 Disable intel iommu driver. 1694 igfx_off [Default Off] 1695 By default, gfx is mapped as normal device. If a gfx 1696 device has a dedicated DMAR unit, the DMAR unit is 1697 bypassed by not enabling DMAR with this option. In 1698 this case, gfx device will use physical address for 1699 DMA. 1700 forcedac [x86_64] 1701 With this option iommu will not optimize to look 1702 for io virtual address below 32-bit forcing dual 1703 address cycle on pci bus for cards supporting greater 1704 than 32-bit addressing. The default is to look 1705 for translation below 32-bit and if not available 1706 then look in the higher range. 1707 strict [Default Off] 1708 With this option on every unmap_single operation will 1709 result in a hardware IOTLB flush operation as opposed 1710 to batching them for performance. 1711 sp_off [Default Off] 1712 By default, super page will be supported if Intel IOMMU 1713 has the capability. With this option, super page will 1714 not be supported. 1715 sm_on [Default Off] 1716 By default, scalable mode will be disabled even if the 1717 hardware advertises that it has support for the scalable 1718 mode translation. With this option set, scalable mode 1719 will be used on hardware which claims to support it. 1720 tboot_noforce [Default Off] 1721 Do not force the Intel IOMMU enabled under tboot. 1722 By default, tboot will force Intel IOMMU on, which 1723 could harm performance of some high-throughput 1724 devices like 40GBit network cards, even if identity 1725 mapping is enabled. 1726 Note that using this option lowers the security 1727 provided by tboot because it makes the system 1728 vulnerable to DMA attacks. 1729 1730 intel_idle.max_cstate= [KNL,HW,ACPI,X86] 1731 0 disables intel_idle and fall back on acpi_idle. 1732 1 to 9 specify maximum depth of C-state. 1733 1734 intel_pstate= [X86] 1735 disable 1736 Do not enable intel_pstate as the default 1737 scaling driver for the supported processors 1738 passive 1739 Use intel_pstate as a scaling driver, but configure it 1740 to work with generic cpufreq governors (instead of 1741 enabling its internal governor). This mode cannot be 1742 used along with the hardware-managed P-states (HWP) 1743 feature. 1744 force 1745 Enable intel_pstate on systems that prohibit it by default 1746 in favor of acpi-cpufreq. Forcing the intel_pstate driver 1747 instead of acpi-cpufreq may disable platform features, such 1748 as thermal controls and power capping, that rely on ACPI 1749 P-States information being indicated to OSPM and therefore 1750 should be used with caution. This option does not work with 1751 processors that aren't supported by the intel_pstate driver 1752 or on platforms that use pcc-cpufreq instead of acpi-cpufreq. 1753 no_hwp 1754 Do not enable hardware P state control (HWP) 1755 if available. 1756 hwp_only 1757 Only load intel_pstate on systems which support 1758 hardware P state control (HWP) if available. 1759 support_acpi_ppc 1760 Enforce ACPI _PPC performance limits. If the Fixed ACPI 1761 Description Table, specifies preferred power management 1762 profile as "Enterprise Server" or "Performance Server", 1763 then this feature is turned on by default. 1764 per_cpu_perf_limits 1765 Allow per-logical-CPU P-State performance control limits using 1766 cpufreq sysfs interface 1767 1768 intremap= [X86-64, Intel-IOMMU] 1769 on enable Interrupt Remapping (default) 1770 off disable Interrupt Remapping 1771 nosid disable Source ID checking 1772 no_x2apic_optout 1773 BIOS x2APIC opt-out request will be ignored 1774 nopost disable Interrupt Posting 1775 1776 iomem= Disable strict checking of access to MMIO memory 1777 strict regions from userspace. 1778 relaxed 1779 1780 iommu= [x86] 1781 off 1782 force 1783 noforce 1784 biomerge 1785 panic 1786 nopanic 1787 merge 1788 nomerge 1789 soft 1790 pt [x86] 1791 nopt [x86] 1792 nobypass [PPC/POWERNV] 1793 Disable IOMMU bypass, using IOMMU for PCI devices. 1794 1795 iommu.strict= [ARM64] Configure TLB invalidation behaviour 1796 Format: { "0" | "1" } 1797 0 - Lazy mode. 1798 Request that DMA unmap operations use deferred 1799 invalidation of hardware TLBs, for increased 1800 throughput at the cost of reduced device isolation. 1801 Will fall back to strict mode if not supported by 1802 the relevant IOMMU driver. 1803 1 - Strict mode (default). 1804 DMA unmap operations invalidate IOMMU hardware TLBs 1805 synchronously. 1806 1807 iommu.passthrough= 1808 [ARM64] Configure DMA to bypass the IOMMU by default. 1809 Format: { "0" | "1" } 1810 0 - Use IOMMU translation for DMA. 1811 1 - Bypass the IOMMU for DMA. 1812 unset - Use value of CONFIG_IOMMU_DEFAULT_PASSTHROUGH. 1813 1814 io7= [HW] IO7 for Marvel based alpha systems 1815 See comment before marvel_specify_io7 in 1816 arch/alpha/kernel/core_marvel.c. 1817 1818 io_delay= [X86] I/O delay method 1819 0x80 1820 Standard port 0x80 based delay 1821 0xed 1822 Alternate port 0xed based delay (needed on some systems) 1823 udelay 1824 Simple two microseconds delay 1825 none 1826 No delay 1827 1828 ip= [IP_PNP] 1829 See Documentation/filesystems/nfs/nfsroot.txt. 1830 1831 irqaffinity= [SMP] Set the default irq affinity mask 1832 The argument is a cpu list, as described above. 1833 1834 irqchip.gicv2_force_probe= 1835 [ARM, ARM64] 1836 Format: <bool> 1837 Force the kernel to look for the second 4kB page 1838 of a GICv2 controller even if the memory range 1839 exposed by the device tree is too small. 1840 1841 irqchip.gicv3_nolpi= 1842 [ARM, ARM64] 1843 Force the kernel to ignore the availability of 1844 LPIs (and by consequence ITSs). Intended for system 1845 that use the kernel as a bootloader, and thus want 1846 to let secondary kernels in charge of setting up 1847 LPIs. 1848 1849 irqchip.gicv3_pseudo_nmi= [ARM64] 1850 Enables support for pseudo-NMIs in the kernel. This 1851 requires the kernel to be built with 1852 CONFIG_ARM64_PSEUDO_NMI. 1853 1854 irqfixup [HW] 1855 When an interrupt is not handled search all handlers 1856 for it. Intended to get systems with badly broken 1857 firmware running. 1858 1859 irqpoll [HW] 1860 When an interrupt is not handled search all handlers 1861 for it. Also check all handlers each timer 1862 interrupt. Intended to get systems with badly broken 1863 firmware running. 1864 1865 isapnp= [ISAPNP] 1866 Format: <RDP>,<reset>,<pci_scan>,<verbosity> 1867 1868 isolcpus= [KNL,SMP,ISOL] Isolate a given set of CPUs from disturbance. 1869 [Deprecated - use cpusets instead] 1870 Format: [flag-list,]<cpu-list> 1871 1872 Specify one or more CPUs to isolate from disturbances 1873 specified in the flag list (default: domain): 1874 1875 nohz 1876 Disable the tick when a single task runs. 1877 1878 A residual 1Hz tick is offloaded to workqueues, which you 1879 need to affine to housekeeping through the global 1880 workqueue's affinity configured via the 1881 /sys/devices/virtual/workqueue/cpumask sysfs file, or 1882 by using the 'domain' flag described below. 1883 1884 NOTE: by default the global workqueue runs on all CPUs, 1885 so to protect individual CPUs the 'cpumask' file has to 1886 be configured manually after bootup. 1887 1888 domain 1889 Isolate from the general SMP balancing and scheduling 1890 algorithms. Note that performing domain isolation this way 1891 is irreversible: it's not possible to bring back a CPU to 1892 the domains once isolated through isolcpus. It's strongly 1893 advised to use cpusets instead to disable scheduler load 1894 balancing through the "cpuset.sched_load_balance" file. 1895 It offers a much more flexible interface where CPUs can 1896 move in and out of an isolated set anytime. 1897 1898 You can move a process onto or off an "isolated" CPU via 1899 the CPU affinity syscalls or cpuset. 1900 <cpu number> begins at 0 and the maximum value is 1901 "number of CPUs in system - 1". 1902 1903 The format of <cpu-list> is described above. 1904 1905 1906 1907 iucv= [HW,NET] 1908 1909 ivrs_ioapic [HW,X86_64] 1910 Provide an override to the IOAPIC-ID<->DEVICE-ID 1911 mapping provided in the IVRS ACPI table. For 1912 example, to map IOAPIC-ID decimal 10 to 1913 PCI device 00:14.0 write the parameter as: 1914 ivrs_ioapic[10]=00:14.0 1915 1916 ivrs_hpet [HW,X86_64] 1917 Provide an override to the HPET-ID<->DEVICE-ID 1918 mapping provided in the IVRS ACPI table. For 1919 example, to map HPET-ID decimal 0 to 1920 PCI device 00:14.0 write the parameter as: 1921 ivrs_hpet[0]=00:14.0 1922 1923 ivrs_acpihid [HW,X86_64] 1924 Provide an override to the ACPI-HID:UID<->DEVICE-ID 1925 mapping provided in the IVRS ACPI table. For 1926 example, to map UART-HID:UID AMD0020:0 to 1927 PCI device 00:14.5 write the parameter as: 1928 ivrs_acpihid[00:14.5]=AMD0020:0 1929 1930 js= [HW,JOY] Analog joystick 1931 See Documentation/input/joydev/joystick.rst. 1932 1933 nokaslr [KNL] 1934 When CONFIG_RANDOMIZE_BASE is set, this disables 1935 kernel and module base offset ASLR (Address Space 1936 Layout Randomization). 1937 1938 kasan_multi_shot 1939 [KNL] Enforce KASAN (Kernel Address Sanitizer) to print 1940 report on every invalid memory access. Without this 1941 parameter KASAN will print report only for the first 1942 invalid access. 1943 1944 keepinitrd [HW,ARM] 1945 1946 kernelcore= [KNL,X86,IA-64,PPC] 1947 Format: nn[KMGTPE] | nn% | "mirror" 1948 This parameter specifies the amount of memory usable by 1949 the kernel for non-movable allocations. The requested 1950 amount is spread evenly throughout all nodes in the 1951 system as ZONE_NORMAL. The remaining memory is used for 1952 movable memory in its own zone, ZONE_MOVABLE. In the 1953 event, a node is too small to have both ZONE_NORMAL and 1954 ZONE_MOVABLE, kernelcore memory will take priority and 1955 other nodes will have a larger ZONE_MOVABLE. 1956 1957 ZONE_MOVABLE is used for the allocation of pages that 1958 may be reclaimed or moved by the page migration 1959 subsystem. Note that allocations like PTEs-from-HighMem 1960 still use the HighMem zone if it exists, and the Normal 1961 zone if it does not. 1962 1963 It is possible to specify the exact amount of memory in 1964 the form of "nn[KMGTPE]", a percentage of total system 1965 memory in the form of "nn%", or "mirror". If "mirror" 1966 option is specified, mirrored (reliable) memory is used 1967 for non-movable allocations and remaining memory is used 1968 for Movable pages. "nn[KMGTPE]", "nn%", and "mirror" 1969 are exclusive, so you cannot specify multiple forms. 1970 1971 kgdbdbgp= [KGDB,HW] kgdb over EHCI usb debug port. 1972 Format: <Controller#>[,poll interval] 1973 The controller # is the number of the ehci usb debug 1974 port as it is probed via PCI. The poll interval is 1975 optional and is the number seconds in between 1976 each poll cycle to the debug port in case you need 1977 the functionality for interrupting the kernel with 1978 gdb or control-c on the dbgp connection. When 1979 not using this parameter you use sysrq-g to break into 1980 the kernel debugger. 1981 1982 kgdboc= [KGDB,HW] kgdb over consoles. 1983 Requires a tty driver that supports console polling, 1984 or a supported polling keyboard driver (non-usb). 1985 Serial only format: <serial_device>[,baud] 1986 keyboard only format: kbd 1987 keyboard and serial format: kbd,<serial_device>[,baud] 1988 Optional Kernel mode setting: 1989 kms, kbd format: kms,kbd 1990 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud] 1991 1992 kgdbwait [KGDB] Stop kernel execution and enter the 1993 kernel debugger at the earliest opportunity. 1994 1995 kmac= [MIPS] korina ethernet MAC address. 1996 Configure the RouterBoard 532 series on-chip 1997 Ethernet adapter MAC address. 1998 1999 kmemleak= [KNL] Boot-time kmemleak enable/disable 2000 Valid arguments: on, off 2001 Default: on 2002 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y, 2003 the default is off. 2004 2005 kpti= [ARM64] Control page table isolation of user 2006 and kernel address spaces. 2007 Default: enabled on cores which need mitigation. 2008 0: force disabled 2009 1: force enabled 2010 2011 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs. 2012 Default is 0 (don't ignore, but inject #GP) 2013 2014 kvm.enable_vmware_backdoor=[KVM] Support VMware backdoor PV interface. 2015 Default is false (don't support). 2016 2017 kvm.mmu_audit= [KVM] This is a R/W parameter which allows audit 2018 KVM MMU at runtime. 2019 Default is 0 (off) 2020 2021 kvm-amd.nested= [KVM,AMD] Allow nested virtualization in KVM/SVM. 2022 Default is 1 (enabled) 2023 2024 kvm-amd.npt= [KVM,AMD] Disable nested paging (virtualized MMU) 2025 for all guests. 2026 Default is 1 (enabled) if in 64-bit or 32-bit PAE mode. 2027 2028 kvm-arm.vgic_v3_group0_trap= 2029 [KVM,ARM] Trap guest accesses to GICv3 group-0 2030 system registers 2031 2032 kvm-arm.vgic_v3_group1_trap= 2033 [KVM,ARM] Trap guest accesses to GICv3 group-1 2034 system registers 2035 2036 kvm-arm.vgic_v3_common_trap= 2037 [KVM,ARM] Trap guest accesses to GICv3 common 2038 system registers 2039 2040 kvm-arm.vgic_v4_enable= 2041 [KVM,ARM] Allow use of GICv4 for direct injection of 2042 LPIs. 2043 2044 kvm-intel.ept= [KVM,Intel] Disable extended page tables 2045 (virtualized MMU) support on capable Intel chips. 2046 Default is 1 (enabled) 2047 2048 kvm-intel.emulate_invalid_guest_state= 2049 [KVM,Intel] Enable emulation of invalid guest states 2050 Default is 0 (disabled) 2051 2052 kvm-intel.flexpriority= 2053 [KVM,Intel] Disable FlexPriority feature (TPR shadow). 2054 Default is 1 (enabled) 2055 2056 kvm-intel.nested= 2057 [KVM,Intel] Enable VMX nesting (nVMX). 2058 Default is 0 (disabled) 2059 2060 kvm-intel.unrestricted_guest= 2061 [KVM,Intel] Disable unrestricted guest feature 2062 (virtualized real and unpaged mode) on capable 2063 Intel chips. Default is 1 (enabled) 2064 2065 kvm-intel.vmentry_l1d_flush=[KVM,Intel] Mitigation for L1 Terminal Fault 2066 CVE-2018-3620. 2067 2068 Valid arguments: never, cond, always 2069 2070 always: L1D cache flush on every VMENTER. 2071 cond: Flush L1D on VMENTER only when the code between 2072 VMEXIT and VMENTER can leak host memory. 2073 never: Disables the mitigation 2074 2075 Default is cond (do L1 cache flush in specific instances) 2076 2077 kvm-intel.vpid= [KVM,Intel] Disable Virtual Processor Identification 2078 feature (tagged TLBs) on capable Intel chips. 2079 Default is 1 (enabled) 2080 2081 l1tf= [X86] Control mitigation of the L1TF vulnerability on 2082 affected CPUs 2083 2084 The kernel PTE inversion protection is unconditionally 2085 enabled and cannot be disabled. 2086 2087 full 2088 Provides all available mitigations for the 2089 L1TF vulnerability. Disables SMT and 2090 enables all mitigations in the 2091 hypervisors, i.e. unconditional L1D flush. 2092 2093 SMT control and L1D flush control via the 2094 sysfs interface is still possible after 2095 boot. Hypervisors will issue a warning 2096 when the first VM is started in a 2097 potentially insecure configuration, 2098 i.e. SMT enabled or L1D flush disabled. 2099 2100 full,force 2101 Same as 'full', but disables SMT and L1D 2102 flush runtime control. Implies the 2103 'nosmt=force' command line option. 2104 (i.e. sysfs control of SMT is disabled.) 2105 2106 flush 2107 Leaves SMT enabled and enables the default 2108 hypervisor mitigation, i.e. conditional 2109 L1D flush. 2110 2111 SMT control and L1D flush control via the 2112 sysfs interface is still possible after 2113 boot. Hypervisors will issue a warning 2114 when the first VM is started in a 2115 potentially insecure configuration, 2116 i.e. SMT enabled or L1D flush disabled. 2117 2118 flush,nosmt 2119 2120 Disables SMT and enables the default 2121 hypervisor mitigation. 2122 2123 SMT control and L1D flush control via the 2124 sysfs interface is still possible after 2125 boot. Hypervisors will issue a warning 2126 when the first VM is started in a 2127 potentially insecure configuration, 2128 i.e. SMT enabled or L1D flush disabled. 2129 2130 flush,nowarn 2131 Same as 'flush', but hypervisors will not 2132 warn when a VM is started in a potentially 2133 insecure configuration. 2134 2135 off 2136 Disables hypervisor mitigations and doesn't 2137 emit any warnings. 2138 It also drops the swap size and available 2139 RAM limit restriction on both hypervisor and 2140 bare metal. 2141 2142 Default is 'flush'. 2143 2144 For details see: Documentation/admin-guide/l1tf.rst 2145 2146 l2cr= [PPC] 2147 2148 l3cr= [PPC] 2149 2150 lapic [X86-32,APIC] Enable the local APIC even if BIOS 2151 disabled it. 2152 2153 lapic= [x86,APIC] "notscdeadline" Do not use TSC deadline 2154 value for LAPIC timer one-shot implementation. Default 2155 back to the programmable timer unit in the LAPIC. 2156 2157 lapic_timer_c2_ok [X86,APIC] trust the local apic timer 2158 in C2 power state. 2159 2160 libata.dma= [LIBATA] DMA control 2161 libata.dma=0 Disable all PATA and SATA DMA 2162 libata.dma=1 PATA and SATA Disk DMA only 2163 libata.dma=2 ATAPI (CDROM) DMA only 2164 libata.dma=4 Compact Flash DMA only 2165 Combinations also work, so libata.dma=3 enables DMA 2166 for disks and CDROMs, but not CFs. 2167 2168 libata.ignore_hpa= [LIBATA] Ignore HPA limit 2169 libata.ignore_hpa=0 keep BIOS limits (default) 2170 libata.ignore_hpa=1 ignore limits, using full disk 2171 2172 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume 2173 when set. 2174 Format: <int> 2175 2176 libata.force= [LIBATA] Force configurations. The format is comma 2177 separated list of "[ID:]VAL" where ID is 2178 PORT[.DEVICE]. PORT and DEVICE are decimal numbers 2179 matching port, link or device. Basically, it matches 2180 the ATA ID string printed on console by libata. If 2181 the whole ID part is omitted, the last PORT and DEVICE 2182 values are used. If ID hasn't been specified yet, the 2183 configuration applies to all ports, links and devices. 2184 2185 If only DEVICE is omitted, the parameter applies to 2186 the port and all links and devices behind it. DEVICE 2187 number of 0 either selects the first device or the 2188 first fan-out link behind PMP device. It does not 2189 select the host link. DEVICE number of 15 selects the 2190 host link and device attached to it. 2191 2192 The VAL specifies the configuration to force. As long 2193 as there's no ambiguity shortcut notation is allowed. 2194 For example, both 1.5 and 1.5G would work for 1.5Gbps. 2195 The following configurations can be forced. 2196 2197 * Cable type: 40c, 80c, short40c, unk, ign or sata. 2198 Any ID with matching PORT is used. 2199 2200 * SATA link speed limit: 1.5Gbps or 3.0Gbps. 2201 2202 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7]. 2203 udma[/][16,25,33,44,66,100,133] notation is also 2204 allowed. 2205 2206 * [no]ncq: Turn on or off NCQ. 2207 2208 * [no]ncqtrim: Turn off queued DSM TRIM. 2209 2210 * nohrst, nosrst, norst: suppress hard, soft 2211 and both resets. 2212 2213 * rstonce: only attempt one reset during 2214 hot-unplug link recovery 2215 2216 * dump_id: dump IDENTIFY data. 2217 2218 * atapi_dmadir: Enable ATAPI DMADIR bridge support 2219 2220 * disable: Disable this device. 2221 2222 If there are multiple matching configurations changing 2223 the same attribute, the last one is used. 2224 2225 memblock=debug [KNL] Enable memblock debug messages. 2226 2227 load_ramdisk= [RAM] List of ramdisks to load from floppy 2228 See Documentation/blockdev/ramdisk.txt. 2229 2230 lockd.nlm_grace_period=P [NFS] Assign grace period. 2231 Format: <integer> 2232 2233 lockd.nlm_tcpport=N [NFS] Assign TCP port. 2234 Format: <integer> 2235 2236 lockd.nlm_timeout=T [NFS] Assign timeout value. 2237 Format: <integer> 2238 2239 lockd.nlm_udpport=M [NFS] Assign UDP port. 2240 Format: <integer> 2241 2242 locktorture.nreaders_stress= [KNL] 2243 Set the number of locking read-acquisition kthreads. 2244 Defaults to being automatically set based on the 2245 number of online CPUs. 2246 2247 locktorture.nwriters_stress= [KNL] 2248 Set the number of locking write-acquisition kthreads. 2249 2250 locktorture.onoff_holdoff= [KNL] 2251 Set time (s) after boot for CPU-hotplug testing. 2252 2253 locktorture.onoff_interval= [KNL] 2254 Set time (s) between CPU-hotplug operations, or 2255 zero to disable CPU-hotplug testing. 2256 2257 locktorture.shuffle_interval= [KNL] 2258 Set task-shuffle interval (jiffies). Shuffling 2259 tasks allows some CPUs to go into dyntick-idle 2260 mode during the locktorture test. 2261 2262 locktorture.shutdown_secs= [KNL] 2263 Set time (s) after boot system shutdown. This 2264 is useful for hands-off automated testing. 2265 2266 locktorture.stat_interval= [KNL] 2267 Time (s) between statistics printk()s. 2268 2269 locktorture.stutter= [KNL] 2270 Time (s) to stutter testing, for example, 2271 specifying five seconds causes the test to run for 2272 five seconds, wait for five seconds, and so on. 2273 This tests the locking primitive's ability to 2274 transition abruptly to and from idle. 2275 2276 locktorture.torture_type= [KNL] 2277 Specify the locking implementation to test. 2278 2279 locktorture.verbose= [KNL] 2280 Enable additional printk() statements. 2281 2282 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver 2283 Format: <irq> 2284 2285 loglevel= All Kernel Messages with a loglevel smaller than the 2286 console loglevel will be printed to the console. It can 2287 also be changed with klogd or other programs. The 2288 loglevels are defined as follows: 2289 2290 0 (KERN_EMERG) system is unusable 2291 1 (KERN_ALERT) action must be taken immediately 2292 2 (KERN_CRIT) critical conditions 2293 3 (KERN_ERR) error conditions 2294 4 (KERN_WARNING) warning conditions 2295 5 (KERN_NOTICE) normal but significant condition 2296 6 (KERN_INFO) informational 2297 7 (KERN_DEBUG) debug-level messages 2298 2299 log_buf_len=n[KMG] Sets the size of the printk ring buffer, 2300 in bytes. n must be a power of two and greater 2301 than the minimal size. The minimal size is defined 2302 by LOG_BUF_SHIFT kernel config parameter. There is 2303 also CONFIG_LOG_CPU_MAX_BUF_SHIFT config parameter 2304 that allows to increase the default size depending on 2305 the number of CPUs. See init/Kconfig for more details. 2306 2307 logo.nologo [FB] Disables display of the built-in Linux logo. 2308 This may be used to provide more screen space for 2309 kernel log messages and is useful when debugging 2310 kernel boot problems. 2311 2312 lp=0 [LP] Specify parallel ports to use, e.g, 2313 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses 2314 lp=reset first parallel port). 'lp=0' disables the 2315 lp=auto printer driver. 'lp=reset' (which can be 2316 specified in addition to the ports) causes 2317 attached printers to be reset. Using 2318 lp=port1,port2,... specifies the parallel ports 2319 to associate lp devices with, starting with 2320 lp0. A port specification may be 'none' to skip 2321 that lp device, or a parport name such as 2322 'parport0'. Specifying 'lp=auto' instead of a 2323 port specification list means that device IDs 2324 from each port should be examined, to see if 2325 an IEEE 1284-compliant printer is attached; if 2326 so, the driver will manage that printer. 2327 See also header of drivers/char/lp.c. 2328 2329 lpj=n [KNL] 2330 Sets loops_per_jiffy to given constant, thus avoiding 2331 time-consuming boot-time autodetection (up to 250 ms per 2332 CPU). 0 enables autodetection (default). To determine 2333 the correct value for your kernel, boot with normal 2334 autodetection and see what value is printed. Note that 2335 on SMP systems the preset will be applied to all CPUs, 2336 which is likely to cause problems if your CPUs need 2337 significantly divergent settings. An incorrect value 2338 will cause delays in the kernel to be wrong, leading to 2339 unpredictable I/O errors and other breakage. Although 2340 unlikely, in the extreme case this might damage your 2341 hardware. 2342 2343 ltpc= [NET] 2344 Format: <io>,<irq>,<dma> 2345 2346 lsm.debug [SECURITY] Enable LSM initialization debugging output. 2347 2348 lsm=lsm1,...,lsmN 2349 [SECURITY] Choose order of LSM initialization. This 2350 overrides CONFIG_LSM, and the "security=" parameter. 2351 2352 machvec= [IA-64] Force the use of a particular machine-vector 2353 (machvec) in a generic kernel. 2354 Example: machvec=hpzx1_swiotlb 2355 2356 machtype= [Loongson] Share the same kernel image file between different 2357 yeeloong laptop. 2358 Example: machtype=lemote-yeeloong-2f-7inch 2359 2360 max_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory greater 2361 than or equal to this physical address is ignored. 2362 2363 maxcpus= [SMP] Maximum number of processors that an SMP kernel 2364 will bring up during bootup. maxcpus=n : n >= 0 limits 2365 the kernel to bring up 'n' processors. Surely after 2366 bootup you can bring up the other plugged cpu by executing 2367 "echo 1 > /sys/devices/system/cpu/cpuX/online". So maxcpus 2368 only takes effect during system bootup. 2369 While n=0 is a special case, it is equivalent to "nosmp", 2370 which also disables the IO APIC. 2371 2372 max_loop= [LOOP] The number of loop block devices that get 2373 (loop.max_loop) unconditionally pre-created at init time. The default 2374 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead 2375 of statically allocating a predefined number, loop 2376 devices can be requested on-demand with the 2377 /dev/loop-control interface. 2378 2379 mce [X86-32] Machine Check Exception 2380 2381 mce=option [X86-64] See Documentation/x86/x86_64/boot-options.txt 2382 2383 md= [HW] RAID subsystems devices and level 2384 See Documentation/admin-guide/md.rst. 2385 2386 mdacon= [MDA] 2387 Format: <first>,<last> 2388 Specifies range of consoles to be captured by the MDA. 2389 2390 mem=nn[KMG] [KNL,BOOT] Force usage of a specific amount of memory 2391 Amount of memory to be used when the kernel is not able 2392 to see the whole system memory or for test. 2393 [X86] Work as limiting max address. Use together 2394 with memmap= to avoid physical address space collisions. 2395 Without memmap= PCI devices could be placed at addresses 2396 belonging to unused RAM. 2397 2398 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel 2399 memory. 2400 2401 memchunk=nn[KMG] 2402 [KNL,SH] Allow user to override the default size for 2403 per-device physically contiguous DMA buffers. 2404 2405 memhp_default_state=online/offline 2406 [KNL] Set the initial state for the memory hotplug 2407 onlining policy. If not specified, the default value is 2408 set according to the 2409 CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE kernel config 2410 option. 2411 See Documentation/memory-hotplug.txt. 2412 2413 memmap=exactmap [KNL,X86] Enable setting of an exact 2414 E820 memory map, as specified by the user. 2415 Such memmap=exactmap lines can be constructed based on 2416 BIOS output or other requirements. See the memmap=nn@ss 2417 option description. 2418 2419 memmap=nn[KMG]@ss[KMG] 2420 [KNL] Force usage of a specific region of memory. 2421 Region of memory to be used is from ss to ss+nn. 2422 If @ss[KMG] is omitted, it is equivalent to mem=nn[KMG], 2423 which limits max address to nn[KMG]. 2424 Multiple different regions can be specified, 2425 comma delimited. 2426 Example: 2427 memmap=100M@2G,100M#3G,1G!1024G 2428 2429 memmap=nn[KMG]#ss[KMG] 2430 [KNL,ACPI] Mark specific memory as ACPI data. 2431 Region of memory to be marked is from ss to ss+nn. 2432 2433 memmap=nn[KMG]$ss[KMG] 2434 [KNL,ACPI] Mark specific memory as reserved. 2435 Region of memory to be reserved is from ss to ss+nn. 2436 Example: Exclude memory from 0x18690000-0x1869ffff 2437 memmap=64K$0x18690000 2438 or 2439 memmap=0x10000$0x18690000 2440 Some bootloaders may need an escape character before '$', 2441 like Grub2, otherwise '$' and the following number 2442 will be eaten. 2443 2444 memmap=nn[KMG]!ss[KMG] 2445 [KNL,X86] Mark specific memory as protected. 2446 Region of memory to be used, from ss to ss+nn. 2447 The memory region may be marked as e820 type 12 (0xc) 2448 and is NVDIMM or ADR memory. 2449 2450 memmap=<size>%<offset>-<oldtype>+<newtype> 2451 [KNL,ACPI] Convert memory within the specified region 2452 from <oldtype> to <newtype>. If "-<oldtype>" is left 2453 out, the whole region will be marked as <newtype>, 2454 even if previously unavailable. If "+<newtype>" is left 2455 out, matching memory will be removed. Types are 2456 specified as e820 types, e.g., 1 = RAM, 2 = reserved, 2457 3 = ACPI, 12 = PRAM. 2458 2459 memory_corruption_check=0/1 [X86] 2460 Some BIOSes seem to corrupt the first 64k of 2461 memory when doing things like suspend/resume. 2462 Setting this option will scan the memory 2463 looking for corruption. Enabling this will 2464 both detect corruption and prevent the kernel 2465 from using the memory being corrupted. 2466 However, its intended as a diagnostic tool; if 2467 repeatable BIOS-originated corruption always 2468 affects the same memory, you can use memmap= 2469 to prevent the kernel from using that memory. 2470 2471 memory_corruption_check_size=size [X86] 2472 By default it checks for corruption in the low 2473 64k, making this memory unavailable for normal 2474 use. Use this parameter to scan for 2475 corruption in more or less memory. 2476 2477 memory_corruption_check_period=seconds [X86] 2478 By default it checks for corruption every 60 2479 seconds. Use this parameter to check at some 2480 other rate. 0 disables periodic checking. 2481 2482 memtest= [KNL,X86,ARM,PPC] Enable memtest 2483 Format: <integer> 2484 default : 0 <disable> 2485 Specifies the number of memtest passes to be 2486 performed. Each pass selects another test 2487 pattern from a given set of patterns. Memtest 2488 fills the memory with this pattern, validates 2489 memory contents and reserves bad memory 2490 regions that are detected. 2491 2492 mem_encrypt= [X86-64] AMD Secure Memory Encryption (SME) control 2493 Valid arguments: on, off 2494 Default (depends on kernel configuration option): 2495 on (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=y) 2496 off (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=n) 2497 mem_encrypt=on: Activate SME 2498 mem_encrypt=off: Do not activate SME 2499 2500 Refer to Documentation/x86/amd-memory-encryption.txt 2501 for details on when memory encryption can be activated. 2502 2503 mem_sleep_default= [SUSPEND] Default system suspend mode: 2504 s2idle - Suspend-To-Idle 2505 shallow - Power-On Suspend or equivalent (if supported) 2506 deep - Suspend-To-RAM or equivalent (if supported) 2507 See Documentation/admin-guide/pm/sleep-states.rst. 2508 2509 meye.*= [HW] Set MotionEye Camera parameters 2510 See Documentation/media/v4l-drivers/meye.rst. 2511 2512 mfgpt_irq= [IA-32] Specify the IRQ to use for the 2513 Multi-Function General Purpose Timers on AMD Geode 2514 platforms. 2515 2516 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when 2517 the BIOS has incorrectly applied a workaround. TinyBIOS 2518 version 0.98 is known to be affected, 0.99 fixes the 2519 problem by letting the user disable the workaround. 2520 2521 mga= [HW,DRM] 2522 2523 min_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory below this 2524 physical address is ignored. 2525 2526 mini2440= [ARM,HW,KNL] 2527 Format:[0..2][b][c][t] 2528 Default: "0tb" 2529 MINI2440 configuration specification: 2530 0 - The attached screen is the 3.5" TFT 2531 1 - The attached screen is the 7" TFT 2532 2 - The VGA Shield is attached (1024x768) 2533 Leaving out the screen size parameter will not load 2534 the TFT driver, and the framebuffer will be left 2535 unconfigured. 2536 b - Enable backlight. The TFT backlight pin will be 2537 linked to the kernel VESA blanking code and a GPIO 2538 LED. This parameter is not necessary when using the 2539 VGA shield. 2540 c - Enable the s3c camera interface. 2541 t - Reserved for enabling touchscreen support. The 2542 touchscreen support is not enabled in the mainstream 2543 kernel as of 2.6.30, a preliminary port can be found 2544 in the "bleeding edge" mini2440 support kernel at 2545 http://repo.or.cz/w/linux-2.6/mini2440.git 2546 2547 mminit_loglevel= 2548 [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this 2549 parameter allows control of the logging verbosity for 2550 the additional memory initialisation checks. A value 2551 of 0 disables mminit logging and a level of 4 will 2552 log everything. Information is printed at KERN_DEBUG 2553 so loglevel=8 may also need to be specified. 2554 2555 module.sig_enforce 2556 [KNL] When CONFIG_MODULE_SIG is set, this means that 2557 modules without (valid) signatures will fail to load. 2558 Note that if CONFIG_MODULE_SIG_FORCE is set, that 2559 is always true, so this option does nothing. 2560 2561 module_blacklist= [KNL] Do not load a comma-separated list of 2562 modules. Useful for debugging problem modules. 2563 2564 mousedev.tap_time= 2565 [MOUSE] Maximum time between finger touching and 2566 leaving touchpad surface for touch to be considered 2567 a tap and be reported as a left button click (for 2568 touchpads working in absolute mode only). 2569 Format: <msecs> 2570 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices 2571 reporting absolute coordinates, such as tablets 2572 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices 2573 reporting absolute coordinates, such as tablets 2574 2575 movablecore= [KNL,X86,IA-64,PPC] 2576 Format: nn[KMGTPE] | nn% 2577 This parameter is the complement to kernelcore=, it 2578 specifies the amount of memory used for migratable 2579 allocations. If both kernelcore and movablecore is 2580 specified, then kernelcore will be at *least* the 2581 specified value but may be more. If movablecore on its 2582 own is specified, the administrator must be careful 2583 that the amount of memory usable for all allocations 2584 is not too small. 2585 2586 movable_node [KNL] Boot-time switch to make hotplugable memory 2587 NUMA nodes to be movable. This means that the memory 2588 of such nodes will be usable only for movable 2589 allocations which rules out almost all kernel 2590 allocations. Use with caution! 2591 2592 MTD_Partition= [MTD] 2593 Format: <name>,<region-number>,<size>,<offset> 2594 2595 MTD_Region= [MTD] Format: 2596 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>] 2597 2598 mtdparts= [MTD] 2599 See drivers/mtd/cmdlinepart.c. 2600 2601 multitce=off [PPC] This parameter disables the use of the pSeries 2602 firmware feature for updating multiple TCE entries 2603 at a time. 2604 2605 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration 2606 2607 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock] 2608 2609 boundary - index of last SLC block on Flex-OneNAND. 2610 The remaining blocks are configured as MLC blocks. 2611 lock - Configure if Flex-OneNAND boundary should be locked. 2612 Once locked, the boundary cannot be changed. 2613 1 indicates lock status, 0 indicates unlock status. 2614 2615 mtdset= [ARM] 2616 ARM/S3C2412 JIVE boot control 2617 2618 See arch/arm/mach-s3c2412/mach-jive.c 2619 2620 mtouchusb.raw_coordinates= 2621 [HW] Make the MicroTouch USB driver use raw coordinates 2622 ('y', default) or cooked coordinates ('n') 2623 2624 mtrr_chunk_size=nn[KMG] [X86] 2625 used for mtrr cleanup. It is largest continuous chunk 2626 that could hold holes aka. UC entries. 2627 2628 mtrr_gran_size=nn[KMG] [X86] 2629 Used for mtrr cleanup. It is granularity of mtrr block. 2630 Default is 1. 2631 Large value could prevent small alignment from 2632 using up MTRRs. 2633 2634 mtrr_spare_reg_nr=n [X86] 2635 Format: <integer> 2636 Range: 0,7 : spare reg number 2637 Default : 1 2638 Used for mtrr cleanup. It is spare mtrr entries number. 2639 Set to 2 or more if your graphical card needs more. 2640 2641 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card 2642 2643 netdev= [NET] Network devices parameters 2644 Format: <irq>,<io>,<mem_start>,<mem_end>,<name> 2645 Note that mem_start is often overloaded to mean 2646 something different and driver-specific. 2647 This usage is only documented in each driver source 2648 file if at all. 2649 2650 nf_conntrack.acct= 2651 [NETFILTER] Enable connection tracking flow accounting 2652 0 to disable accounting 2653 1 to enable accounting 2654 Default value is 0. 2655 2656 nfsaddrs= [NFS] Deprecated. Use ip= instead. 2657 See Documentation/filesystems/nfs/nfsroot.txt. 2658 2659 nfsroot= [NFS] nfs root filesystem for disk-less boxes. 2660 See Documentation/filesystems/nfs/nfsroot.txt. 2661 2662 nfsrootdebug [NFS] enable nfsroot debugging messages. 2663 See Documentation/filesystems/nfs/nfsroot.txt. 2664 2665 nfs.callback_nr_threads= 2666 [NFSv4] set the total number of threads that the 2667 NFS client will assign to service NFSv4 callback 2668 requests. 2669 2670 nfs.callback_tcpport= 2671 [NFS] set the TCP port on which the NFSv4 callback 2672 channel should listen. 2673 2674 nfs.cache_getent= 2675 [NFS] sets the pathname to the program which is used 2676 to update the NFS client cache entries. 2677 2678 nfs.cache_getent_timeout= 2679 [NFS] sets the timeout after which an attempt to 2680 update a cache entry is deemed to have failed. 2681 2682 nfs.idmap_cache_timeout= 2683 [NFS] set the maximum lifetime for idmapper cache 2684 entries. 2685 2686 nfs.enable_ino64= 2687 [NFS] enable 64-bit inode numbers. 2688 If zero, the NFS client will fake up a 32-bit inode 2689 number for the readdir() and stat() syscalls instead 2690 of returning the full 64-bit number. 2691 The default is to return 64-bit inode numbers. 2692 2693 nfs.max_session_cb_slots= 2694 [NFSv4.1] Sets the maximum number of session 2695 slots the client will assign to the callback 2696 channel. This determines the maximum number of 2697 callbacks the client will process in parallel for 2698 a particular server. 2699 2700 nfs.max_session_slots= 2701 [NFSv4.1] Sets the maximum number of session slots 2702 the client will attempt to negotiate with the server. 2703 This limits the number of simultaneous RPC requests 2704 that the client can send to the NFSv4.1 server. 2705 Note that there is little point in setting this 2706 value higher than the max_tcp_slot_table_limit. 2707 2708 nfs.nfs4_disable_idmapping= 2709 [NFSv4] When set to the default of '1', this option 2710 ensures that both the RPC level authentication 2711 scheme and the NFS level operations agree to use 2712 numeric uids/gids if the mount is using the 2713 'sec=sys' security flavour. In effect it is 2714 disabling idmapping, which can make migration from 2715 legacy NFSv2/v3 systems to NFSv4 easier. 2716 Servers that do not support this mode of operation 2717 will be autodetected by the client, and it will fall 2718 back to using the idmapper. 2719 To turn off this behaviour, set the value to '0'. 2720 nfs.nfs4_unique_id= 2721 [NFS4] Specify an additional fixed unique ident- 2722 ification string that NFSv4 clients can insert into 2723 their nfs_client_id4 string. This is typically a 2724 UUID that is generated at system install time. 2725 2726 nfs.send_implementation_id = 2727 [NFSv4.1] Send client implementation identification 2728 information in exchange_id requests. 2729 If zero, no implementation identification information 2730 will be sent. 2731 The default is to send the implementation identification 2732 information. 2733 2734 nfs.recover_lost_locks = 2735 [NFSv4] Attempt to recover locks that were lost due 2736 to a lease timeout on the server. Please note that 2737 doing this risks data corruption, since there are 2738 no guarantees that the file will remain unchanged 2739 after the locks are lost. 2740 If you want to enable the kernel legacy behaviour of 2741 attempting to recover these locks, then set this 2742 parameter to '1'. 2743 The default parameter value of '0' causes the kernel 2744 not to attempt recovery of lost locks. 2745 2746 nfs4.layoutstats_timer = 2747 [NFSv4.2] Change the rate at which the kernel sends 2748 layoutstats to the pNFS metadata server. 2749 2750 Setting this to value to 0 causes the kernel to use 2751 whatever value is the default set by the layout 2752 driver. A non-zero value sets the minimum interval 2753 in seconds between layoutstats transmissions. 2754 2755 nfsd.nfs4_disable_idmapping= 2756 [NFSv4] When set to the default of '1', the NFSv4 2757 server will return only numeric uids and gids to 2758 clients using auth_sys, and will accept numeric uids 2759 and gids from such clients. This is intended to ease 2760 migration from NFSv2/v3. 2761 2762 nmi_debug= [KNL,SH] Specify one or more actions to take 2763 when a NMI is triggered. 2764 Format: [state][,regs][,debounce][,die] 2765 2766 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels 2767 Format: [panic,][nopanic,][num] 2768 Valid num: 0 or 1 2769 0 - turn hardlockup detector in nmi_watchdog off 2770 1 - turn hardlockup detector in nmi_watchdog on 2771 When panic is specified, panic when an NMI watchdog 2772 timeout occurs (or 'nopanic' to override the opposite 2773 default). To disable both hard and soft lockup detectors, 2774 please see 'nowatchdog'. 2775 This is useful when you use a panic=... timeout and 2776 need the box quickly up again. 2777 2778 These settings can be accessed at runtime via 2779 the nmi_watchdog and hardlockup_panic sysctls. 2780 2781 netpoll.carrier_timeout= 2782 [NET] Specifies amount of time (in seconds) that 2783 netpoll should wait for a carrier. By default netpoll 2784 waits 4 seconds. 2785 2786 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths 2787 emulation library even if a 387 maths coprocessor 2788 is present. 2789 2790 no5lvl [X86-64] Disable 5-level paging mode. Forces 2791 kernel to use 4-level paging instead. 2792 2793 no_console_suspend 2794 [HW] Never suspend the console 2795 Disable suspending of consoles during suspend and 2796 hibernate operations. Once disabled, debugging 2797 messages can reach various consoles while the rest 2798 of the system is being put to sleep (ie, while 2799 debugging driver suspend/resume hooks). This may 2800 not work reliably with all consoles, but is known 2801 to work with serial and VGA consoles. 2802 To facilitate more flexible debugging, we also add 2803 console_suspend, a printk module parameter to control 2804 it. Users could use console_suspend (usually 2805 /sys/module/printk/parameters/console_suspend) to 2806 turn on/off it dynamically. 2807 2808 noaliencache [MM, NUMA, SLAB] Disables the allocation of alien 2809 caches in the slab allocator. Saves per-node memory, 2810 but will impact performance. 2811 2812 noalign [KNL,ARM] 2813 2814 noaltinstr [S390] Disables alternative instructions patching 2815 (CPU alternatives feature). 2816 2817 noapic [SMP,APIC] Tells the kernel to not make use of any 2818 IOAPICs that may be present in the system. 2819 2820 noautogroup Disable scheduler automatic task group creation. 2821 2822 nobats [PPC] Do not use BATs for mapping kernel lowmem 2823 on "Classic" PPC cores. 2824 2825 nocache [ARM] 2826 2827 noclflush [BUGS=X86] Don't use the CLFLUSH instruction 2828 2829 nodelayacct [KNL] Disable per-task delay accounting 2830 2831 nodsp [SH] Disable hardware DSP at boot time. 2832 2833 noefi Disable EFI runtime services support. 2834 2835 noexec [IA-64] 2836 2837 noexec [X86] 2838 On X86-32 available only on PAE configured kernels. 2839 noexec=on: enable non-executable mappings (default) 2840 noexec=off: disable non-executable mappings 2841 2842 nosmap [X86] 2843 Disable SMAP (Supervisor Mode Access Prevention) 2844 even if it is supported by processor. 2845 2846 nosmep [X86,PPC] 2847 Disable SMEP (Supervisor Mode Execution Prevention) 2848 even if it is supported by processor. 2849 2850 noexec32 [X86-64] 2851 This affects only 32-bit executables. 2852 noexec32=on: enable non-executable mappings (default) 2853 read doesn't imply executable mappings 2854 noexec32=off: disable non-executable mappings 2855 read implies executable mappings 2856 2857 nofpu [MIPS,SH] Disable hardware FPU at boot time. 2858 2859 nofxsr [BUGS=X86-32] Disables x86 floating point extended 2860 register save and restore. The kernel will only save 2861 legacy floating-point registers on task switch. 2862 2863 nohugeiomap [KNL,x86] Disable kernel huge I/O mappings. 2864 2865 nosmt [KNL,S390] Disable symmetric multithreading (SMT). 2866 Equivalent to smt=1. 2867 2868 [KNL,x86] Disable symmetric multithreading (SMT). 2869 nosmt=force: Force disable SMT, cannot be undone 2870 via the sysfs control file. 2871 2872 nospectre_v1 [PPC] Disable mitigations for Spectre Variant 1 (bounds 2873 check bypass). With this option data leaks are possible 2874 in the system. 2875 2876 nospectre_v2 [X86,PPC_FSL_BOOK3E] Disable all mitigations for the Spectre variant 2 2877 (indirect branch prediction) vulnerability. System may 2878 allow data leaks with this option, which is equivalent 2879 to spectre_v2=off. 2880 2881 nospec_store_bypass_disable 2882 [HW] Disable all mitigations for the Speculative Store Bypass vulnerability 2883 2884 noxsave [BUGS=X86] Disables x86 extended register state save 2885 and restore using xsave. The kernel will fallback to 2886 enabling legacy floating-point and sse state. 2887 2888 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended 2889 register states. The kernel will fall back to use 2890 xsave to save the states. By using this parameter, 2891 performance of saving the states is degraded because 2892 xsave doesn't support modified optimization while 2893 xsaveopt supports it on xsaveopt enabled systems. 2894 2895 noxsaves [X86] Disables xsaves and xrstors used in saving and 2896 restoring x86 extended register state in compacted 2897 form of xsave area. The kernel will fall back to use 2898 xsaveopt and xrstor to save and restore the states 2899 in standard form of xsave area. By using this 2900 parameter, xsave area per process might occupy more 2901 memory on xsaves enabled systems. 2902 2903 nohlt [BUGS=ARM,SH] Tells the kernel that the sleep(SH) or 2904 wfi(ARM) instruction doesn't work correctly and not to 2905 use it. This is also useful when using JTAG debugger. 2906 2907 no_file_caps Tells the kernel not to honor file capabilities. The 2908 only way then for a file to be executed with privilege 2909 is to be setuid root or executed by root. 2910 2911 nohalt [IA-64] Tells the kernel not to use the power saving 2912 function PAL_HALT_LIGHT when idle. This increases 2913 power-consumption. On the positive side, it reduces 2914 interrupt wake-up latency, which may improve performance 2915 in certain environments such as networked servers or 2916 real-time systems. 2917 2918 nohibernate [HIBERNATION] Disable hibernation and resume. 2919 2920 nohz= [KNL] Boottime enable/disable dynamic ticks 2921 Valid arguments: on, off 2922 Default: on 2923 2924 nohz_full= [KNL,BOOT,SMP,ISOL] 2925 The argument is a cpu list, as described above. 2926 In kernels built with CONFIG_NO_HZ_FULL=y, set 2927 the specified list of CPUs whose tick will be stopped 2928 whenever possible. The boot CPU will be forced outside 2929 the range to maintain the timekeeping. Any CPUs 2930 in this list will have their RCU callbacks offloaded, 2931 just as if they had also been called out in the 2932 rcu_nocbs= boot parameter. 2933 2934 noiotrap [SH] Disables trapped I/O port accesses. 2935 2936 noirqdebug [X86-32] Disables the code which attempts to detect and 2937 disable unhandled interrupt sources. 2938 2939 no_timer_check [X86,APIC] Disables the code which tests for 2940 broken timer IRQ sources. 2941 2942 noisapnp [ISAPNP] Disables ISA PnP code. 2943 2944 noinitrd [RAM] Tells the kernel not to load any configured 2945 initial RAM disk. 2946 2947 nointremap [X86-64, Intel-IOMMU] Do not enable interrupt 2948 remapping. 2949 [Deprecated - use intremap=off] 2950 2951 nointroute [IA-64] 2952 2953 noinvpcid [X86] Disable the INVPCID cpu feature. 2954 2955 nojitter [IA-64] Disables jitter checking for ITC timers. 2956 2957 no-kvmclock [X86,KVM] Disable paravirtualized KVM clock driver 2958 2959 no-kvmapf [X86,KVM] Disable paravirtualized asynchronous page 2960 fault handling. 2961 2962 no-vmw-sched-clock 2963 [X86,PV_OPS] Disable paravirtualized VMware scheduler 2964 clock and use the default one. 2965 2966 no-steal-acc [X86,KVM] Disable paravirtualized steal time accounting. 2967 steal time is computed, but won't influence scheduler 2968 behaviour 2969 2970 nolapic [X86-32,APIC] Do not enable or use the local APIC. 2971 2972 nolapic_timer [X86-32,APIC] Do not use the local APIC timer. 2973 2974 noltlbs [PPC] Do not use large page/tlb entries for kernel 2975 lowmem mapping on PPC40x and PPC8xx 2976 2977 nomca [IA-64] Disable machine check abort handling 2978 2979 nomce [X86-32] Disable Machine Check Exception 2980 2981 nomfgpt [X86-32] Disable Multi-Function General Purpose 2982 Timer usage (for AMD Geode machines). 2983 2984 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to 2985 shutdown the other cpus. Instead use the REBOOT_VECTOR 2986 irq. 2987 2988 nomodule Disable module load 2989 2990 nopat [X86] Disable PAT (page attribute table extension of 2991 pagetables) support. 2992 2993 nopcid [X86-64] Disable the PCID cpu feature. 2994 2995 norandmaps Don't use address space randomization. Equivalent to 2996 echo 0 > /proc/sys/kernel/randomize_va_space 2997 2998 noreplace-smp [X86-32,SMP] Don't replace SMP instructions 2999 with UP alternatives 3000 3001 nordrand [X86] Disable kernel use of the RDRAND and 3002 RDSEED instructions even if they are supported 3003 by the processor. RDRAND and RDSEED are still 3004 available to user space applications. 3005 3006 noresume [SWSUSP] Disables resume and restores original swap 3007 space. 3008 3009 no-scroll [VGA] Disables scrollback. 3010 This is required for the Braillex ib80-piezo Braille 3011 reader made by F.H. Papenmeier (Germany). 3012 3013 nosbagart [IA-64] 3014 3015 nosep [BUGS=X86-32] Disables x86 SYSENTER/SYSEXIT support. 3016 3017 nosmp [SMP] Tells an SMP kernel to act as a UP kernel, 3018 and disable the IO APIC. legacy for "maxcpus=0". 3019 3020 nosoftlockup [KNL] Disable the soft-lockup detector. 3021 3022 nosync [HW,M68K] Disables sync negotiation for all devices. 3023 3024 nowatchdog [KNL] Disable both lockup detectors, i.e. 3025 soft-lockup and NMI watchdog (hard-lockup). 3026 3027 nowb [ARM] 3028 3029 nox2apic [X86-64,APIC] Do not enable x2APIC mode. 3030 3031 cpu0_hotplug [X86] Turn on CPU0 hotplug feature when 3032 CONFIG_BOOTPARAM_HOTPLUG_CPU0 is off. 3033 Some features depend on CPU0. Known dependencies are: 3034 1. Resume from suspend/hibernate depends on CPU0. 3035 Suspend/hibernate will fail if CPU0 is offline and you 3036 need to online CPU0 before suspend/hibernate. 3037 2. PIC interrupts also depend on CPU0. CPU0 can't be 3038 removed if a PIC interrupt is detected. 3039 It's said poweroff/reboot may depend on CPU0 on some 3040 machines although I haven't seen such issues so far 3041 after CPU0 is offline on a few tested machines. 3042 If the dependencies are under your control, you can 3043 turn on cpu0_hotplug. 3044 3045 nps_mtm_hs_ctr= [KNL,ARC] 3046 This parameter sets the maximum duration, in 3047 cycles, each HW thread of the CTOP can run 3048 without interruptions, before HW switches it. 3049 The actual maximum duration is 16 times this 3050 parameter's value. 3051 Format: integer between 1 and 255 3052 Default: 255 3053 3054 nptcg= [IA-64] Override max number of concurrent global TLB 3055 purges which is reported from either PAL_VM_SUMMARY or 3056 SAL PALO. 3057 3058 nr_cpus= [SMP] Maximum number of processors that an SMP kernel 3059 could support. nr_cpus=n : n >= 1 limits the kernel to 3060 support 'n' processors. It could be larger than the 3061 number of already plugged CPU during bootup, later in 3062 runtime you can physically add extra cpu until it reaches 3063 n. So during boot up some boot time memory for per-cpu 3064 variables need be pre-allocated for later physical cpu 3065 hot plugging. 3066 3067 nr_uarts= [SERIAL] maximum number of UARTs to be registered. 3068 3069 numa_balancing= [KNL,X86] Enable or disable automatic NUMA balancing. 3070 Allowed values are enable and disable 3071 3072 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA. 3073 'node', 'default' can be specified 3074 This can be set from sysctl after boot. 3075 See Documentation/sysctl/vm.txt for details. 3076 3077 ohci1394_dma=early [HW] enable debugging via the ohci1394 driver. 3078 See Documentation/debugging-via-ohci1394.txt for more 3079 info. 3080 3081 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands 3082 Rather than timing out after 20 ms if an EC 3083 command is not properly ACKed, override the length 3084 of the timeout. We have interrupts disabled while 3085 waiting for the ACK, so if this is set too high 3086 interrupts *may* be lost! 3087 3088 omap_mux= [OMAP] Override bootloader pin multiplexing. 3089 Format: <mux_mode0.mode_name=value>... 3090 For example, to override I2C bus2: 3091 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100 3092 3093 oprofile.timer= [HW] 3094 Use timer interrupt instead of performance counters 3095 3096 oprofile.cpu_type= Force an oprofile cpu type 3097 This might be useful if you have an older oprofile 3098 userland or if you want common events. 3099 Format: { arch_perfmon } 3100 arch_perfmon: [X86] Force use of architectural 3101 perfmon on Intel CPUs instead of the 3102 CPU specific event set. 3103 timer: [X86] Force use of architectural NMI 3104 timer mode (see also oprofile.timer 3105 for generic hr timer mode) 3106 3107 oops=panic Always panic on oopses. Default is to just kill the 3108 process, but there is a small probability of 3109 deadlocking the machine. 3110 This will also cause panics on machine check exceptions. 3111 Useful together with panic=30 to trigger a reboot. 3112 3113 page_owner= [KNL] Boot-time page_owner enabling option. 3114 Storage of the information about who allocated 3115 each page is disabled in default. With this switch, 3116 we can turn it on. 3117 on: enable the feature 3118 3119 page_poison= [KNL] Boot-time parameter changing the state of 3120 poisoning on the buddy allocator, available with 3121 CONFIG_PAGE_POISONING=y. 3122 off: turn off poisoning (default) 3123 on: turn on poisoning 3124 3125 panic= [KNL] Kernel behaviour on panic: delay <timeout> 3126 timeout > 0: seconds before rebooting 3127 timeout = 0: wait forever 3128 timeout < 0: reboot immediately 3129 Format: <timeout> 3130 3131 panic_print= Bitmask for printing system info when panic happens. 3132 User can chose combination of the following bits: 3133 bit 0: print all tasks info 3134 bit 1: print system memory info 3135 bit 2: print timer info 3136 bit 3: print locks info if CONFIG_LOCKDEP is on 3137 bit 4: print ftrace buffer 3138 3139 panic_on_warn panic() instead of WARN(). Useful to cause kdump 3140 on a WARN(). 3141 3142 crash_kexec_post_notifiers 3143 Run kdump after running panic-notifiers and dumping 3144 kmsg. This only for the users who doubt kdump always 3145 succeeds in any situation. 3146 Note that this also increases risks of kdump failure, 3147 because some panic notifiers can make the crashed 3148 kernel more unstable. 3149 3150 parkbd.port= [HW] Parallel port number the keyboard adapter is 3151 connected to, default is 0. 3152 Format: <parport#> 3153 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation, 3154 0 for XT, 1 for AT (default is AT). 3155 Format: <mode> 3156 3157 parport= [HW,PPT] Specify parallel ports. 0 disables. 3158 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] } 3159 Use 'auto' to force the driver to use any 3160 IRQ/DMA settings detected (the default is to 3161 ignore detected IRQ/DMA settings because of 3162 possible conflicts). You can specify the base 3163 address, IRQ, and DMA settings; IRQ and DMA 3164 should be numbers, or 'auto' (for using detected 3165 settings on that particular port), or 'nofifo' 3166 (to avoid using a FIFO even if it is detected). 3167 Parallel ports are assigned in the order they 3168 are specified on the command line, starting 3169 with parport0. 3170 3171 parport_init_mode= [HW,PPT] 3172 Configure VIA parallel port to operate in 3173 a specific mode. This is necessary on Pegasos 3174 computer where firmware has no options for setting 3175 up parallel port mode and sets it to spp. 3176 Currently this function knows 686a and 8231 chips. 3177 Format: [spp|ps2|epp|ecp|ecpepp] 3178 3179 pause_on_oops= 3180 Halt all CPUs after the first oops has been printed for 3181 the specified number of seconds. This is to be used if 3182 your oopses keep scrolling off the screen. 3183 3184 pcbit= [HW,ISDN] 3185 3186 pcd. [PARIDE] 3187 See header of drivers/block/paride/pcd.c. 3188 See also Documentation/blockdev/paride.txt. 3189 3190 pci=option[,option...] [PCI] various PCI subsystem options. 3191 3192 Some options herein operate on a specific device 3193 or a set of devices (<pci_dev>). These are 3194 specified in one of the following formats: 3195 3196 [<domain>:]<bus>:<dev>.<func>[/<dev>.<func>]* 3197 pci:<vendor>:<device>[:<subvendor>:<subdevice>] 3198 3199 Note: the first format specifies a PCI 3200 bus/device/function address which may change 3201 if new hardware is inserted, if motherboard 3202 firmware changes, or due to changes caused 3203 by other kernel parameters. If the 3204 domain is left unspecified, it is 3205 taken to be zero. Optionally, a path 3206 to a device through multiple device/function 3207 addresses can be specified after the base 3208 address (this is more robust against 3209 renumbering issues). The second format 3210 selects devices using IDs from the 3211 configuration space which may match multiple 3212 devices in the system. 3213 3214 earlydump dump PCI config space before the kernel 3215 changes anything 3216 off [X86] don't probe for the PCI bus 3217 bios [X86-32] force use of PCI BIOS, don't access 3218 the hardware directly. Use this if your machine 3219 has a non-standard PCI host bridge. 3220 nobios [X86-32] disallow use of PCI BIOS, only direct 3221 hardware access methods are allowed. Use this 3222 if you experience crashes upon bootup and you 3223 suspect they are caused by the BIOS. 3224 conf1 [X86] Force use of PCI Configuration Access 3225 Mechanism 1 (config address in IO port 0xCF8, 3226 data in IO port 0xCFC, both 32-bit). 3227 conf2 [X86] Force use of PCI Configuration Access 3228 Mechanism 2 (IO port 0xCF8 is an 8-bit port for 3229 the function, IO port 0xCFA, also 8-bit, sets 3230 bus number. The config space is then accessed 3231 through ports 0xC000-0xCFFF). 3232 See http://wiki.osdev.org/PCI for more info 3233 on the configuration access mechanisms. 3234 noaer [PCIE] If the PCIEAER kernel config parameter is 3235 enabled, this kernel boot option can be used to 3236 disable the use of PCIE advanced error reporting. 3237 nodomains [PCI] Disable support for multiple PCI 3238 root domains (aka PCI segments, in ACPI-speak). 3239 nommconf [X86] Disable use of MMCONFIG for PCI 3240 Configuration 3241 check_enable_amd_mmconf [X86] check for and enable 3242 properly configured MMIO access to PCI 3243 config space on AMD family 10h CPU 3244 nomsi [MSI] If the PCI_MSI kernel config parameter is 3245 enabled, this kernel boot option can be used to 3246 disable the use of MSI interrupts system-wide. 3247 noioapicquirk [APIC] Disable all boot interrupt quirks. 3248 Safety option to keep boot IRQs enabled. This 3249 should never be necessary. 3250 ioapicreroute [APIC] Enable rerouting of boot IRQs to the 3251 primary IO-APIC for bridges that cannot disable 3252 boot IRQs. This fixes a source of spurious IRQs 3253 when the system masks IRQs. 3254 noioapicreroute [APIC] Disable workaround that uses the 3255 boot IRQ equivalent of an IRQ that connects to 3256 a chipset where boot IRQs cannot be disabled. 3257 The opposite of ioapicreroute. 3258 biosirq [X86-32] Use PCI BIOS calls to get the interrupt 3259 routing table. These calls are known to be buggy 3260 on several machines and they hang the machine 3261 when used, but on other computers it's the only 3262 way to get the interrupt routing table. Try 3263 this option if the kernel is unable to allocate 3264 IRQs or discover secondary PCI buses on your 3265 motherboard. 3266 rom [X86] Assign address space to expansion ROMs. 3267 Use with caution as certain devices share 3268 address decoders between ROMs and other 3269 resources. 3270 norom [X86] Do not assign address space to 3271 expansion ROMs that do not already have 3272 BIOS assigned address ranges. 3273 nobar [X86] Do not assign address space to the 3274 BARs that weren't assigned by the BIOS. 3275 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be 3276 assigned automatically to PCI devices. You can 3277 make the kernel exclude IRQs of your ISA cards 3278 this way. 3279 pirqaddr=0xAAAAA [X86] Specify the physical address 3280 of the PIRQ table (normally generated 3281 by the BIOS) if it is outside the 3282 F0000h-100000h range. 3283 lastbus=N [X86] Scan all buses thru bus #N. Can be 3284 useful if the kernel is unable to find your 3285 secondary buses and you want to tell it 3286 explicitly which ones they are. 3287 assign-busses [X86] Always assign all PCI bus 3288 numbers ourselves, overriding 3289 whatever the firmware may have done. 3290 usepirqmask [X86] Honor the possible IRQ mask stored 3291 in the BIOS $PIR table. This is needed on 3292 some systems with broken BIOSes, notably 3293 some HP Pavilion N5400 and Omnibook XE3 3294 notebooks. This will have no effect if ACPI 3295 IRQ routing is enabled. 3296 noacpi [X86] Do not use ACPI for IRQ routing 3297 or for PCI scanning. 3298 use_crs [X86] Use PCI host bridge window information 3299 from ACPI. On BIOSes from 2008 or later, this 3300 is enabled by default. If you need to use this, 3301 please report a bug. 3302 nocrs [X86] Ignore PCI host bridge windows from ACPI. 3303 If you need to use this, please report a bug. 3304 routeirq Do IRQ routing for all PCI devices. 3305 This is normally done in pci_enable_device(), 3306 so this option is a temporary workaround 3307 for broken drivers that don't call it. 3308 skip_isa_align [X86] do not align io start addr, so can 3309 handle more pci cards 3310 noearly [X86] Don't do any early type 1 scanning. 3311 This might help on some broken boards which 3312 machine check when some devices' config space 3313 is read. But various workarounds are disabled 3314 and some IOMMU drivers will not work. 3315 bfsort Sort PCI devices into breadth-first order. 3316 This sorting is done to get a device 3317 order compatible with older (<= 2.4) kernels. 3318 nobfsort Don't sort PCI devices into breadth-first order. 3319 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size) 3320 tuning and use the BIOS-configured MPS defaults. 3321 pcie_bus_safe Set every device's MPS to the largest value 3322 supported by all devices below the root complex. 3323 pcie_bus_perf Set device MPS to the largest allowable MPS 3324 based on its parent bus. Also set MRRS (Max 3325 Read Request Size) to the largest supported 3326 value (no larger than the MPS that the device 3327 or bus can support) for best performance. 3328 pcie_bus_peer2peer Set every device's MPS to 128B, which 3329 every device is guaranteed to support. This 3330 configuration allows peer-to-peer DMA between 3331 any pair of devices, possibly at the cost of 3332 reduced performance. This also guarantees 3333 that hot-added devices will work. 3334 cbiosize=nn[KMG] The fixed amount of bus space which is 3335 reserved for the CardBus bridge's IO window. 3336 The default value is 256 bytes. 3337 cbmemsize=nn[KMG] The fixed amount of bus space which is 3338 reserved for the CardBus bridge's memory 3339 window. The default value is 64 megabytes. 3340 resource_alignment= 3341 Format: 3342 [<order of align>@]<pci_dev>[; ...] 3343 Specifies alignment and device to reassign 3344 aligned memory resources. How to 3345 specify the device is described above. 3346 If <order of align> is not specified, 3347 PAGE_SIZE is used as alignment. 3348 PCI-PCI bridge can be specified, if resource 3349 windows need to be expanded. 3350 To specify the alignment for several 3351 instances of a device, the PCI vendor, 3352 device, subvendor, and subdevice may be 3353 specified, e.g., 4096@pci:8086:9c22:103c:198f 3354 ecrc= Enable/disable PCIe ECRC (transaction layer 3355 end-to-end CRC checking). 3356 bios: Use BIOS/firmware settings. This is the 3357 the default. 3358 off: Turn ECRC off 3359 on: Turn ECRC on. 3360 hpiosize=nn[KMG] The fixed amount of bus space which is 3361 reserved for hotplug bridge's IO window. 3362 Default size is 256 bytes. 3363 hpmemsize=nn[KMG] The fixed amount of bus space which is 3364 reserved for hotplug bridge's memory window. 3365 Default size is 2 megabytes. 3366 hpbussize=nn The minimum amount of additional bus numbers 3367 reserved for buses below a hotplug bridge. 3368 Default is 1. 3369 realloc= Enable/disable reallocating PCI bridge resources 3370 if allocations done by BIOS are too small to 3371 accommodate resources required by all child 3372 devices. 3373 off: Turn realloc off 3374 on: Turn realloc on 3375 realloc same as realloc=on 3376 noari do not use PCIe ARI. 3377 noats [PCIE, Intel-IOMMU, AMD-IOMMU] 3378 do not use PCIe ATS (and IOMMU device IOTLB). 3379 pcie_scan_all Scan all possible PCIe devices. Otherwise we 3380 only look for one device below a PCIe downstream 3381 port. 3382 big_root_window Try to add a big 64bit memory window to the PCIe 3383 root complex on AMD CPUs. Some GFX hardware 3384 can resize a BAR to allow access to all VRAM. 3385 Adding the window is slightly risky (it may 3386 conflict with unreported devices), so this 3387 taints the kernel. 3388 disable_acs_redir=<pci_dev>[; ...] 3389 Specify one or more PCI devices (in the format 3390 specified above) separated by semicolons. 3391 Each device specified will have the PCI ACS 3392 redirect capabilities forced off which will 3393 allow P2P traffic between devices through 3394 bridges without forcing it upstream. Note: 3395 this removes isolation between devices and 3396 may put more devices in an IOMMU group. 3397 3398 pcie_aspm= [PCIE] Forcibly enable or disable PCIe Active State Power 3399 Management. 3400 off Disable ASPM. 3401 force Enable ASPM even on devices that claim not to support it. 3402 WARNING: Forcing ASPM on may cause system lockups. 3403 3404 pcie_ports= [PCIE] PCIe port services handling: 3405 native Use native PCIe services (PME, AER, DPC, PCIe hotplug) 3406 even if the platform doesn't give the OS permission to 3407 use them. This may cause conflicts if the platform 3408 also tries to use these services. 3409 compat Disable native PCIe services (PME, AER, DPC, PCIe 3410 hotplug). 3411 3412 pcie_port_pm= [PCIE] PCIe port power management handling: 3413 off Disable power management of all PCIe ports 3414 force Forcibly enable power management of all PCIe ports 3415 3416 pcie_pme= [PCIE,PM] Native PCIe PME signaling options: 3417 nomsi Do not use MSI for native PCIe PME signaling (this makes 3418 all PCIe root ports use INTx for all services). 3419 3420 pcmv= [HW,PCMCIA] BadgePAD 4 3421 3422 pd_ignore_unused 3423 [PM] 3424 Keep all power-domains already enabled by bootloader on, 3425 even if no driver has claimed them. This is useful 3426 for debug and development, but should not be 3427 needed on a platform with proper driver support. 3428 3429 pd. [PARIDE] 3430 See Documentation/blockdev/paride.txt. 3431 3432 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at 3433 boot time. 3434 Format: { 0 | 1 } 3435 See arch/parisc/kernel/pdc_chassis.c 3436 3437 percpu_alloc= Select which percpu first chunk allocator to use. 3438 Currently supported values are "embed" and "page". 3439 Archs may support subset or none of the selections. 3440 See comments in mm/percpu.c for details on each 3441 allocator. This parameter is primarily for debugging 3442 and performance comparison. 3443 3444 pf. [PARIDE] 3445 See Documentation/blockdev/paride.txt. 3446 3447 pg. [PARIDE] 3448 See Documentation/blockdev/paride.txt. 3449 3450 pirq= [SMP,APIC] Manual mp-table setup 3451 See Documentation/x86/i386/IO-APIC.txt. 3452 3453 plip= [PPT,NET] Parallel port network link 3454 Format: { parport<nr> | timid | 0 } 3455 See also Documentation/admin-guide/parport.rst. 3456 3457 pmtmr= [X86] Manual setup of pmtmr I/O Port. 3458 Override pmtimer IOPort with a hex value. 3459 e.g. pmtmr=0x508 3460 3461 pnp.debug=1 [PNP] 3462 Enable PNP debug messages (depends on the 3463 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time 3464 via /sys/module/pnp/parameters/debug. We always show 3465 current resource usage; turning this on also shows 3466 possible settings and some assignment information. 3467 3468 pnpacpi= [ACPI] 3469 { off } 3470 3471 pnpbios= [ISAPNP] 3472 { on | off | curr | res | no-curr | no-res } 3473 3474 pnp_reserve_irq= 3475 [ISAPNP] Exclude IRQs for the autoconfiguration 3476 3477 pnp_reserve_dma= 3478 [ISAPNP] Exclude DMAs for the autoconfiguration 3479 3480 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration 3481 Ranges are in pairs (I/O port base and size). 3482 3483 pnp_reserve_mem= 3484 [ISAPNP] Exclude memory regions for the 3485 autoconfiguration. 3486 Ranges are in pairs (memory base and size). 3487 3488 ports= [IP_VS_FTP] IPVS ftp helper module 3489 Default is 21. 3490 Up to 8 (IP_VS_APP_MAX_PORTS) ports 3491 may be specified. 3492 Format: <port>,<port>.... 3493 3494 powersave=off [PPC] This option disables power saving features. 3495 It specifically disables cpuidle and sets the 3496 platform machine description specific power_save 3497 function to NULL. On Idle the CPU just reduces 3498 execution priority. 3499 3500 ppc_strict_facility_enable 3501 [PPC] This option catches any kernel floating point, 3502 Altivec, VSX and SPE outside of regions specifically 3503 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()). 3504 There is some performance impact when enabling this. 3505 3506 ppc_tm= [PPC] 3507 Format: {"off"} 3508 Disable Hardware Transactional Memory 3509 3510 print-fatal-signals= 3511 [KNL] debug: print fatal signals 3512 3513 If enabled, warn about various signal handling 3514 related application anomalies: too many signals, 3515 too many POSIX.1 timers, fatal signals causing a 3516 coredump - etc. 3517 3518 If you hit the warning due to signal overflow, 3519 you might want to try "ulimit -i unlimited". 3520 3521 default: off. 3522 3523 printk.always_kmsg_dump= 3524 Trigger kmsg_dump for cases other than kernel oops or 3525 panics 3526 Format: <bool> (1/Y/y=enable, 0/N/n=disable) 3527 default: disabled 3528 3529 printk.devkmsg={on,off,ratelimit} 3530 Control writing to /dev/kmsg. 3531 on - unlimited logging to /dev/kmsg from userspace 3532 off - logging to /dev/kmsg disabled 3533 ratelimit - ratelimit the logging 3534 Default: ratelimit 3535 3536 printk.time= Show timing data prefixed to each printk message line 3537 Format: <bool> (1/Y/y=enable, 0/N/n=disable) 3538 3539 processor.max_cstate= [HW,ACPI] 3540 Limit processor to maximum C-state 3541 max_cstate=9 overrides any DMI blacklist limit. 3542 3543 processor.nocst [HW,ACPI] 3544 Ignore the _CST method to determine C-states, 3545 instead using the legacy FADT method 3546 3547 profile= [KNL] Enable kernel profiling via /proc/profile 3548 Format: [<profiletype>,]<number> 3549 Param: <profiletype>: "schedule", "sleep", or "kvm" 3550 [defaults to kernel profiling] 3551 Param: "schedule" - profile schedule points. 3552 Param: "sleep" - profile D-state sleeping (millisecs). 3553 Requires CONFIG_SCHEDSTATS 3554 Param: "kvm" - profile VM exits. 3555 Param: <number> - step/bucket size as a power of 2 for 3556 statistical time based profiling. 3557 3558 prompt_ramdisk= [RAM] List of RAM disks to prompt for floppy disk 3559 before loading. 3560 See Documentation/blockdev/ramdisk.txt. 3561 3562 psi= [KNL] Enable or disable pressure stall information 3563 tracking. 3564 Format: <bool> 3565 3566 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to 3567 probe for; one of (bare|imps|exps|lifebook|any). 3568 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports 3569 per second. 3570 psmouse.resetafter= [HW,MOUSE] 3571 Try to reset the device after so many bad packets 3572 (0 = never). 3573 psmouse.resolution= 3574 [HW,MOUSE] Set desired mouse resolution, in dpi. 3575 psmouse.smartscroll= 3576 [HW,MOUSE] Controls Logitech smartscroll autorepeat. 3577 0 = disabled, 1 = enabled (default). 3578 3579 pstore.backend= Specify the name of the pstore backend to use 3580 3581 pt. [PARIDE] 3582 See Documentation/blockdev/paride.txt. 3583 3584 pti= [X86_64] Control Page Table Isolation of user and 3585 kernel address spaces. Disabling this feature 3586 removes hardening, but improves performance of 3587 system calls and interrupts. 3588 3589 on - unconditionally enable 3590 off - unconditionally disable 3591 auto - kernel detects whether your CPU model is 3592 vulnerable to issues that PTI mitigates 3593 3594 Not specifying this option is equivalent to pti=auto. 3595 3596 nopti [X86_64] 3597 Equivalent to pti=off 3598 3599 pty.legacy_count= 3600 [KNL] Number of legacy pty's. Overwrites compiled-in 3601 default number. 3602 3603 quiet [KNL] Disable most log messages 3604 3605 r128= [HW,DRM] 3606 3607 raid= [HW,RAID] 3608 See Documentation/admin-guide/md.rst. 3609 3610 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes 3611 See Documentation/blockdev/ramdisk.txt. 3612 3613 random.trust_cpu={on,off} 3614 [KNL] Enable or disable trusting the use of the 3615 CPU's random number generator (if available) to 3616 fully seed the kernel's CRNG. Default is controlled 3617 by CONFIG_RANDOM_TRUST_CPU. 3618 3619 ras=option[,option,...] [KNL] RAS-specific options 3620 3621 cec_disable [X86] 3622 Disable the Correctable Errors Collector, 3623 see CONFIG_RAS_CEC help text. 3624 3625 rcu_nocbs= [KNL] 3626 The argument is a cpu list, as described above. 3627 3628 In kernels built with CONFIG_RCU_NOCB_CPU=y, set 3629 the specified list of CPUs to be no-callback CPUs. 3630 Invocation of these CPUs' RCU callbacks will be 3631 offloaded to "rcuox/N" kthreads created for that 3632 purpose, where "x" is "p" for RCU-preempt, and 3633 "s" for RCU-sched, and "N" is the CPU number. 3634 This reduces OS jitter on the offloaded CPUs, 3635 which can be useful for HPC and real-time 3636 workloads. It can also improve energy efficiency 3637 for asymmetric multiprocessors. 3638 3639 rcu_nocb_poll [KNL] 3640 Rather than requiring that offloaded CPUs 3641 (specified by rcu_nocbs= above) explicitly 3642 awaken the corresponding "rcuoN" kthreads, 3643 make these kthreads poll for callbacks. 3644 This improves the real-time response for the 3645 offloaded CPUs by relieving them of the need to 3646 wake up the corresponding kthread, but degrades 3647 energy efficiency by requiring that the kthreads 3648 periodically wake up to do the polling. 3649 3650 rcutree.blimit= [KNL] 3651 Set maximum number of finished RCU callbacks to 3652 process in one batch. 3653 3654 rcutree.dump_tree= [KNL] 3655 Dump the structure of the rcu_node combining tree 3656 out at early boot. This is used for diagnostic 3657 purposes, to verify correct tree setup. 3658 3659 rcutree.gp_cleanup_delay= [KNL] 3660 Set the number of jiffies to delay each step of 3661 RCU grace-period cleanup. 3662 3663 rcutree.gp_init_delay= [KNL] 3664 Set the number of jiffies to delay each step of 3665 RCU grace-period initialization. 3666 3667 rcutree.gp_preinit_delay= [KNL] 3668 Set the number of jiffies to delay each step of 3669 RCU grace-period pre-initialization, that is, 3670 the propagation of recent CPU-hotplug changes up 3671 the rcu_node combining tree. 3672 3673 rcutree.rcu_fanout_exact= [KNL] 3674 Disable autobalancing of the rcu_node combining 3675 tree. This is used by rcutorture, and might 3676 possibly be useful for architectures having high 3677 cache-to-cache transfer latencies. 3678 3679 rcutree.rcu_fanout_leaf= [KNL] 3680 Change the number of CPUs assigned to each 3681 leaf rcu_node structure. Useful for very 3682 large systems, which will choose the value 64, 3683 and for NUMA systems with large remote-access 3684 latencies, which will choose a value aligned 3685 with the appropriate hardware boundaries. 3686 3687 rcutree.jiffies_till_first_fqs= [KNL] 3688 Set delay from grace-period initialization to 3689 first attempt to force quiescent states. 3690 Units are jiffies, minimum value is zero, 3691 and maximum value is HZ. 3692 3693 rcutree.jiffies_till_next_fqs= [KNL] 3694 Set delay between subsequent attempts to force 3695 quiescent states. Units are jiffies, minimum 3696 value is one, and maximum value is HZ. 3697 3698 rcutree.jiffies_till_sched_qs= [KNL] 3699 Set required age in jiffies for a 3700 given grace period before RCU starts 3701 soliciting quiescent-state help from 3702 rcu_note_context_switch() and cond_resched(). 3703 If not specified, the kernel will calculate 3704 a value based on the most recent settings 3705 of rcutree.jiffies_till_first_fqs 3706 and rcutree.jiffies_till_next_fqs. 3707 This calculated value may be viewed in 3708 rcutree.jiffies_to_sched_qs. Any attempt to set 3709 rcutree.jiffies_to_sched_qs will be cheerfully 3710 overwritten. 3711 3712 rcutree.kthread_prio= [KNL,BOOT] 3713 Set the SCHED_FIFO priority of the RCU per-CPU 3714 kthreads (rcuc/N). This value is also used for 3715 the priority of the RCU boost threads (rcub/N) 3716 and for the RCU grace-period kthreads (rcu_bh, 3717 rcu_preempt, and rcu_sched). If RCU_BOOST is 3718 set, valid values are 1-99 and the default is 1 3719 (the least-favored priority). Otherwise, when 3720 RCU_BOOST is not set, valid values are 0-99 and 3721 the default is zero (non-realtime operation). 3722 3723 rcutree.rcu_nocb_leader_stride= [KNL] 3724 Set the number of NOCB kthread groups, which 3725 defaults to the square root of the number of 3726 CPUs. Larger numbers reduces the wakeup overhead 3727 on the per-CPU grace-period kthreads, but increases 3728 that same overhead on each group's leader. 3729 3730 rcutree.qhimark= [KNL] 3731 Set threshold of queued RCU callbacks beyond which 3732 batch limiting is disabled. 3733 3734 rcutree.qlowmark= [KNL] 3735 Set threshold of queued RCU callbacks below which 3736 batch limiting is re-enabled. 3737 3738 rcutree.rcu_idle_gp_delay= [KNL] 3739 Set wakeup interval for idle CPUs that have 3740 RCU callbacks (RCU_FAST_NO_HZ=y). 3741 3742 rcutree.rcu_idle_lazy_gp_delay= [KNL] 3743 Set wakeup interval for idle CPUs that have 3744 only "lazy" RCU callbacks (RCU_FAST_NO_HZ=y). 3745 Lazy RCU callbacks are those which RCU can 3746 prove do nothing more than free memory. 3747 3748 rcutree.rcu_kick_kthreads= [KNL] 3749 Cause the grace-period kthread to get an extra 3750 wake_up() if it sleeps three times longer than 3751 it should at force-quiescent-state time. 3752 This wake_up() will be accompanied by a 3753 WARN_ONCE() splat and an ftrace_dump(). 3754 3755 rcutree.sysrq_rcu= [KNL] 3756 Commandeer a sysrq key to dump out Tree RCU's 3757 rcu_node tree with an eye towards determining 3758 why a new grace period has not yet started. 3759 3760 rcuperf.gp_async= [KNL] 3761 Measure performance of asynchronous 3762 grace-period primitives such as call_rcu(). 3763 3764 rcuperf.gp_async_max= [KNL] 3765 Specify the maximum number of outstanding 3766 callbacks per writer thread. When a writer 3767 thread exceeds this limit, it invokes the 3768 corresponding flavor of rcu_barrier() to allow 3769 previously posted callbacks to drain. 3770 3771 rcuperf.gp_exp= [KNL] 3772 Measure performance of expedited synchronous 3773 grace-period primitives. 3774 3775 rcuperf.holdoff= [KNL] 3776 Set test-start holdoff period. The purpose of 3777 this parameter is to delay the start of the 3778 test until boot completes in order to avoid 3779 interference. 3780 3781 rcuperf.nreaders= [KNL] 3782 Set number of RCU readers. The value -1 selects 3783 N, where N is the number of CPUs. A value 3784 "n" less than -1 selects N-n+1, where N is again 3785 the number of CPUs. For example, -2 selects N 3786 (the number of CPUs), -3 selects N+1, and so on. 3787 A value of "n" less than or equal to -N selects 3788 a single reader. 3789 3790 rcuperf.nwriters= [KNL] 3791 Set number of RCU writers. The values operate 3792 the same as for rcuperf.nreaders. 3793 N, where N is the number of CPUs 3794 3795 rcuperf.perf_type= [KNL] 3796 Specify the RCU implementation to test. 3797 3798 rcuperf.shutdown= [KNL] 3799 Shut the system down after performance tests 3800 complete. This is useful for hands-off automated 3801 testing. 3802 3803 rcuperf.verbose= [KNL] 3804 Enable additional printk() statements. 3805 3806 rcuperf.writer_holdoff= [KNL] 3807 Write-side holdoff between grace periods, 3808 in microseconds. The default of zero says 3809 no holdoff. 3810 3811 rcutorture.fqs_duration= [KNL] 3812 Set duration of force_quiescent_state bursts 3813 in microseconds. 3814 3815 rcutorture.fqs_holdoff= [KNL] 3816 Set holdoff time within force_quiescent_state bursts 3817 in microseconds. 3818 3819 rcutorture.fqs_stutter= [KNL] 3820 Set wait time between force_quiescent_state bursts 3821 in seconds. 3822 3823 rcutorture.fwd_progress= [KNL] 3824 Enable RCU grace-period forward-progress testing 3825 for the types of RCU supporting this notion. 3826 3827 rcutorture.fwd_progress_div= [KNL] 3828 Specify the fraction of a CPU-stall-warning 3829 period to do tight-loop forward-progress testing. 3830 3831 rcutorture.fwd_progress_holdoff= [KNL] 3832 Number of seconds to wait between successive 3833 forward-progress tests. 3834 3835 rcutorture.fwd_progress_need_resched= [KNL] 3836 Enclose cond_resched() calls within checks for 3837 need_resched() during tight-loop forward-progress 3838 testing. 3839 3840 rcutorture.gp_cond= [KNL] 3841 Use conditional/asynchronous update-side 3842 primitives, if available. 3843 3844 rcutorture.gp_exp= [KNL] 3845 Use expedited update-side primitives, if available. 3846 3847 rcutorture.gp_normal= [KNL] 3848 Use normal (non-expedited) asynchronous 3849 update-side primitives, if available. 3850 3851 rcutorture.gp_sync= [KNL] 3852 Use normal (non-expedited) synchronous 3853 update-side primitives, if available. If all 3854 of rcutorture.gp_cond=, rcutorture.gp_exp=, 3855 rcutorture.gp_normal=, and rcutorture.gp_sync= 3856 are zero, rcutorture acts as if is interpreted 3857 they are all non-zero. 3858 3859 rcutorture.n_barrier_cbs= [KNL] 3860 Set callbacks/threads for rcu_barrier() testing. 3861 3862 rcutorture.nfakewriters= [KNL] 3863 Set number of concurrent RCU writers. These just 3864 stress RCU, they don't participate in the actual 3865 test, hence the "fake". 3866 3867 rcutorture.nreaders= [KNL] 3868 Set number of RCU readers. The value -1 selects 3869 N-1, where N is the number of CPUs. A value 3870 "n" less than -1 selects N-n-2, where N is again 3871 the number of CPUs. For example, -2 selects N 3872 (the number of CPUs), -3 selects N+1, and so on. 3873 3874 rcutorture.object_debug= [KNL] 3875 Enable debug-object double-call_rcu() testing. 3876 3877 rcutorture.onoff_holdoff= [KNL] 3878 Set time (s) after boot for CPU-hotplug testing. 3879 3880 rcutorture.onoff_interval= [KNL] 3881 Set time (jiffies) between CPU-hotplug operations, 3882 or zero to disable CPU-hotplug testing. 3883 3884 rcutorture.shuffle_interval= [KNL] 3885 Set task-shuffle interval (s). Shuffling tasks 3886 allows some CPUs to go into dyntick-idle mode 3887 during the rcutorture test. 3888 3889 rcutorture.shutdown_secs= [KNL] 3890 Set time (s) after boot system shutdown. This 3891 is useful for hands-off automated testing. 3892 3893 rcutorture.stall_cpu= [KNL] 3894 Duration of CPU stall (s) to test RCU CPU stall 3895 warnings, zero to disable. 3896 3897 rcutorture.stall_cpu_holdoff= [KNL] 3898 Time to wait (s) after boot before inducing stall. 3899 3900 rcutorture.stall_cpu_irqsoff= [KNL] 3901 Disable interrupts while stalling if set. 3902 3903 rcutorture.stat_interval= [KNL] 3904 Time (s) between statistics printk()s. 3905 3906 rcutorture.stutter= [KNL] 3907 Time (s) to stutter testing, for example, specifying 3908 five seconds causes the test to run for five seconds, 3909 wait for five seconds, and so on. This tests RCU's 3910 ability to transition abruptly to and from idle. 3911 3912 rcutorture.test_boost= [KNL] 3913 Test RCU priority boosting? 0=no, 1=maybe, 2=yes. 3914 "Maybe" means test if the RCU implementation 3915 under test support RCU priority boosting. 3916 3917 rcutorture.test_boost_duration= [KNL] 3918 Duration (s) of each individual boost test. 3919 3920 rcutorture.test_boost_interval= [KNL] 3921 Interval (s) between each boost test. 3922 3923 rcutorture.test_no_idle_hz= [KNL] 3924 Test RCU's dyntick-idle handling. See also the 3925 rcutorture.shuffle_interval parameter. 3926 3927 rcutorture.torture_type= [KNL] 3928 Specify the RCU implementation to test. 3929 3930 rcutorture.verbose= [KNL] 3931 Enable additional printk() statements. 3932 3933 rcupdate.rcu_cpu_stall_suppress= [KNL] 3934 Suppress RCU CPU stall warning messages. 3935 3936 rcupdate.rcu_cpu_stall_timeout= [KNL] 3937 Set timeout for RCU CPU stall warning messages. 3938 3939 rcupdate.rcu_expedited= [KNL] 3940 Use expedited grace-period primitives, for 3941 example, synchronize_rcu_expedited() instead 3942 of synchronize_rcu(). This reduces latency, 3943 but can increase CPU utilization, degrade 3944 real-time latency, and degrade energy efficiency. 3945 No effect on CONFIG_TINY_RCU kernels. 3946 3947 rcupdate.rcu_normal= [KNL] 3948 Use only normal grace-period primitives, 3949 for example, synchronize_rcu() instead of 3950 synchronize_rcu_expedited(). This improves 3951 real-time latency, CPU utilization, and 3952 energy efficiency, but can expose users to 3953 increased grace-period latency. This parameter 3954 overrides rcupdate.rcu_expedited. No effect on 3955 CONFIG_TINY_RCU kernels. 3956 3957 rcupdate.rcu_normal_after_boot= [KNL] 3958 Once boot has completed (that is, after 3959 rcu_end_inkernel_boot() has been invoked), use 3960 only normal grace-period primitives. No effect 3961 on CONFIG_TINY_RCU kernels. 3962 3963 rcupdate.rcu_task_stall_timeout= [KNL] 3964 Set timeout in jiffies for RCU task stall warning 3965 messages. Disable with a value less than or equal 3966 to zero. 3967 3968 rcupdate.rcu_self_test= [KNL] 3969 Run the RCU early boot self tests 3970 3971 rdinit= [KNL] 3972 Format: <full_path> 3973 Run specified binary instead of /init from the ramdisk, 3974 used for early userspace startup. See initrd. 3975 3976 rdt= [HW,X86,RDT] 3977 Turn on/off individual RDT features. List is: 3978 cmt, mbmtotal, mbmlocal, l3cat, l3cdp, l2cat, l2cdp, 3979 mba. 3980 E.g. to turn on cmt and turn off mba use: 3981 rdt=cmt,!mba 3982 3983 reboot= [KNL] 3984 Format (x86 or x86_64): 3985 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] \ 3986 [[,]s[mp]#### \ 3987 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \ 3988 [[,]f[orce] 3989 Where reboot_mode is one of warm (soft) or cold (hard) or gpio, 3990 reboot_type is one of bios, acpi, kbd, triple, efi, or pci, 3991 reboot_force is either force or not specified, 3992 reboot_cpu is s[mp]#### with #### being the processor 3993 to be used for rebooting. 3994 3995 relax_domain_level= 3996 [KNL, SMP] Set scheduler's default relax_domain_level. 3997 See Documentation/cgroup-v1/cpusets.txt. 3998 3999 reserve= [KNL,BUGS] Force kernel to ignore I/O ports or memory 4000 Format: <base1>,<size1>[,<base2>,<size2>,...] 4001 Reserve I/O ports or memory so the kernel won't use 4002 them. If <base> is less than 0x10000, the region 4003 is assumed to be I/O ports; otherwise it is memory. 4004 4005 reservetop= [X86-32] 4006 Format: nn[KMG] 4007 Reserves a hole at the top of the kernel virtual 4008 address space. 4009 4010 reservelow= [X86] 4011 Format: nn[K] 4012 Set the amount of memory to reserve for BIOS at 4013 the bottom of the address space. 4014 4015 reset_devices [KNL] Force drivers to reset the underlying device 4016 during initialization. 4017 4018 resume= [SWSUSP] 4019 Specify the partition device for software suspend 4020 Format: 4021 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>} 4022 4023 resume_offset= [SWSUSP] 4024 Specify the offset from the beginning of the partition 4025 given by "resume=" at which the swap header is located, 4026 in <PAGE_SIZE> units (needed only for swap files). 4027 See Documentation/power/swsusp-and-swap-files.txt 4028 4029 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to 4030 read the resume files 4031 4032 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up. 4033 Useful for devices that are detected asynchronously 4034 (e.g. USB and MMC devices). 4035 4036 hibernate= [HIBERNATION] 4037 noresume Don't check if there's a hibernation image 4038 present during boot. 4039 nocompress Don't compress/decompress hibernation images. 4040 no Disable hibernation and resume. 4041 protect_image Turn on image protection during restoration 4042 (that will set all pages holding image data 4043 during restoration read-only). 4044 4045 retain_initrd [RAM] Keep initrd memory after extraction 4046 4047 rfkill.default_state= 4048 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm, 4049 etc. communication is blocked by default. 4050 1 Unblocked. 4051 4052 rfkill.master_switch_mode= 4053 0 The "airplane mode" button does nothing. 4054 1 The "airplane mode" button toggles between everything 4055 blocked and the previous configuration. 4056 2 The "airplane mode" button toggles between everything 4057 blocked and everything unblocked. 4058 4059 rhash_entries= [KNL,NET] 4060 Set number of hash buckets for route cache 4061 4062 ring3mwait=disable 4063 [KNL] Disable ring 3 MONITOR/MWAIT feature on supported 4064 CPUs. 4065 4066 ro [KNL] Mount root device read-only on boot 4067 4068 rodata= [KNL] 4069 on Mark read-only kernel memory as read-only (default). 4070 off Leave read-only kernel memory writable for debugging. 4071 4072 rockchip.usb_uart 4073 Enable the uart passthrough on the designated usb port 4074 on Rockchip SoCs. When active, the signals of the 4075 debug-uart get routed to the D+ and D- pins of the usb 4076 port and the regular usb controller gets disabled. 4077 4078 root= [KNL] Root filesystem 4079 See name_to_dev_t comment in init/do_mounts.c. 4080 4081 rootdelay= [KNL] Delay (in seconds) to pause before attempting to 4082 mount the root filesystem 4083 4084 rootflags= [KNL] Set root filesystem mount option string 4085 4086 rootfstype= [KNL] Set root filesystem type 4087 4088 rootwait [KNL] Wait (indefinitely) for root device to show up. 4089 Useful for devices that are detected asynchronously 4090 (e.g. USB and MMC devices). 4091 4092 rproc_mem=nn[KMG][@address] 4093 [KNL,ARM,CMA] Remoteproc physical memory block. 4094 Memory area to be used by remote processor image, 4095 managed by CMA. 4096 4097 rw [KNL] Mount root device read-write on boot 4098 4099 S [KNL] Run init in single mode 4100 4101 s390_iommu= [HW,S390] 4102 Set s390 IOTLB flushing mode 4103 strict 4104 With strict flushing every unmap operation will result in 4105 an IOTLB flush. Default is lazy flushing before reuse, 4106 which is faster. 4107 4108 sa1100ir [NET] 4109 See drivers/net/irda/sa1100_ir.c. 4110 4111 sbni= [NET] Granch SBNI12 leased line adapter 4112 4113 sched_debug [KNL] Enables verbose scheduler debug messages. 4114 4115 schedstats= [KNL,X86] Enable or disable scheduled statistics. 4116 Allowed values are enable and disable. This feature 4117 incurs a small amount of overhead in the scheduler 4118 but is useful for debugging and performance tuning. 4119 4120 skew_tick= [KNL] Offset the periodic timer tick per cpu to mitigate 4121 xtime_lock contention on larger systems, and/or RCU lock 4122 contention on all systems with CONFIG_MAXSMP set. 4123 Format: { "0" | "1" } 4124 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1" 4125 1 -- enable. 4126 Note: increases power consumption, thus should only be 4127 enabled if running jitter sensitive (HPC/RT) workloads. 4128 4129 security= [SECURITY] Choose a legacy "major" security module to 4130 enable at boot. This has been deprecated by the 4131 "lsm=" parameter. 4132 4133 selinux= [SELINUX] Disable or enable SELinux at boot time. 4134 Format: { "0" | "1" } 4135 See security/selinux/Kconfig help text. 4136 0 -- disable. 4137 1 -- enable. 4138 Default value is set via kernel config option. 4139 If enabled at boot time, /selinux/disable can be used 4140 later to disable prior to initial policy load. 4141 4142 apparmor= [APPARMOR] Disable or enable AppArmor at boot time 4143 Format: { "0" | "1" } 4144 See security/apparmor/Kconfig help text 4145 0 -- disable. 4146 1 -- enable. 4147 Default value is set via kernel config option. 4148 4149 serialnumber [BUGS=X86-32] 4150 4151 shapers= [NET] 4152 Maximal number of shapers. 4153 4154 simeth= [IA-64] 4155 simscsi= 4156 4157 slram= [HW,MTD] 4158 4159 slab_nomerge [MM] 4160 Disable merging of slabs with similar size. May be 4161 necessary if there is some reason to distinguish 4162 allocs to different slabs, especially in hardened 4163 environments where the risk of heap overflows and 4164 layout control by attackers can usually be 4165 frustrated by disabling merging. This will reduce 4166 most of the exposure of a heap attack to a single 4167 cache (risks via metadata attacks are mostly 4168 unchanged). Debug options disable merging on their 4169 own. 4170 For more information see Documentation/vm/slub.rst. 4171 4172 slab_max_order= [MM, SLAB] 4173 Determines the maximum allowed order for slabs. 4174 A high setting may cause OOMs due to memory 4175 fragmentation. Defaults to 1 for systems with 4176 more than 32MB of RAM, 0 otherwise. 4177 4178 slub_debug[=options[,slabs]] [MM, SLUB] 4179 Enabling slub_debug allows one to determine the 4180 culprit if slab objects become corrupted. Enabling 4181 slub_debug can create guard zones around objects and 4182 may poison objects when not in use. Also tracks the 4183 last alloc / free. For more information see 4184 Documentation/vm/slub.rst. 4185 4186 slub_memcg_sysfs= [MM, SLUB] 4187 Determines whether to enable sysfs directories for 4188 memory cgroup sub-caches. 1 to enable, 0 to disable. 4189 The default is determined by CONFIG_SLUB_MEMCG_SYSFS_ON. 4190 Enabling this can lead to a very high number of debug 4191 directories and files being created under 4192 /sys/kernel/slub. 4193 4194 slub_max_order= [MM, SLUB] 4195 Determines the maximum allowed order for slabs. 4196 A high setting may cause OOMs due to memory 4197 fragmentation. For more information see 4198 Documentation/vm/slub.rst. 4199 4200 slub_min_objects= [MM, SLUB] 4201 The minimum number of objects per slab. SLUB will 4202 increase the slab order up to slub_max_order to 4203 generate a sufficiently large slab able to contain 4204 the number of objects indicated. The higher the number 4205 of objects the smaller the overhead of tracking slabs 4206 and the less frequently locks need to be acquired. 4207 For more information see Documentation/vm/slub.rst. 4208 4209 slub_min_order= [MM, SLUB] 4210 Determines the minimum page order for slabs. Must be 4211 lower than slub_max_order. 4212 For more information see Documentation/vm/slub.rst. 4213 4214 slub_nomerge [MM, SLUB] 4215 Same with slab_nomerge. This is supported for legacy. 4216 See slab_nomerge for more information. 4217 4218 smart2= [HW] 4219 Format: <io1>[,<io2>[,...,<io8>]] 4220 4221 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices 4222 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port 4223 smsc-ircc2.ircc_sir= [HW] SIR base I/O port 4224 smsc-ircc2.ircc_fir= [HW] FIR base I/O port 4225 smsc-ircc2.ircc_irq= [HW] IRQ line 4226 smsc-ircc2.ircc_dma= [HW] DMA channel 4227 smsc-ircc2.ircc_transceiver= [HW] Transceiver type: 4228 0: Toshiba Satellite 1800 (GP data pin select) 4229 1: Fast pin select (default) 4230 2: ATC IRMode 4231 4232 smt [KNL,S390] Set the maximum number of threads (logical 4233 CPUs) to use per physical CPU on systems capable of 4234 symmetric multithreading (SMT). Will be capped to the 4235 actual hardware limit. 4236 Format: <integer> 4237 Default: -1 (no limit) 4238 4239 softlockup_panic= 4240 [KNL] Should the soft-lockup detector generate panics. 4241 Format: <integer> 4242 4243 A nonzero value instructs the soft-lockup detector 4244 to panic the machine when a soft-lockup occurs. This 4245 is also controlled by CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC 4246 which is the respective build-time switch to that 4247 functionality. 4248 4249 softlockup_all_cpu_backtrace= 4250 [KNL] Should the soft-lockup detector generate 4251 backtraces on all cpus. 4252 Format: <integer> 4253 4254 sonypi.*= [HW] Sony Programmable I/O Control Device driver 4255 See Documentation/laptops/sonypi.txt 4256 4257 spectre_v2= [X86] Control mitigation of Spectre variant 2 4258 (indirect branch speculation) vulnerability. 4259 The default operation protects the kernel from 4260 user space attacks. 4261 4262 on - unconditionally enable, implies 4263 spectre_v2_user=on 4264 off - unconditionally disable, implies 4265 spectre_v2_user=off 4266 auto - kernel detects whether your CPU model is 4267 vulnerable 4268 4269 Selecting 'on' will, and 'auto' may, choose a 4270 mitigation method at run time according to the 4271 CPU, the available microcode, the setting of the 4272 CONFIG_RETPOLINE configuration option, and the 4273 compiler with which the kernel was built. 4274 4275 Selecting 'on' will also enable the mitigation 4276 against user space to user space task attacks. 4277 4278 Selecting 'off' will disable both the kernel and 4279 the user space protections. 4280 4281 Specific mitigations can also be selected manually: 4282 4283 retpoline - replace indirect branches 4284 retpoline,generic - google's original retpoline 4285 retpoline,amd - AMD-specific minimal thunk 4286 4287 Not specifying this option is equivalent to 4288 spectre_v2=auto. 4289 4290 spectre_v2_user= 4291 [X86] Control mitigation of Spectre variant 2 4292 (indirect branch speculation) vulnerability between 4293 user space tasks 4294 4295 on - Unconditionally enable mitigations. Is 4296 enforced by spectre_v2=on 4297 4298 off - Unconditionally disable mitigations. Is 4299 enforced by spectre_v2=off 4300 4301 prctl - Indirect branch speculation is enabled, 4302 but mitigation can be enabled via prctl 4303 per thread. The mitigation control state 4304 is inherited on fork. 4305 4306 prctl,ibpb 4307 - Like "prctl" above, but only STIBP is 4308 controlled per thread. IBPB is issued 4309 always when switching between different user 4310 space processes. 4311 4312 seccomp 4313 - Same as "prctl" above, but all seccomp 4314 threads will enable the mitigation unless 4315 they explicitly opt out. 4316 4317 seccomp,ibpb 4318 - Like "seccomp" above, but only STIBP is 4319 controlled per thread. IBPB is issued 4320 always when switching between different 4321 user space processes. 4322 4323 auto - Kernel selects the mitigation depending on 4324 the available CPU features and vulnerability. 4325 4326 Default mitigation: 4327 If CONFIG_SECCOMP=y then "seccomp", otherwise "prctl" 4328 4329 Not specifying this option is equivalent to 4330 spectre_v2_user=auto. 4331 4332 spec_store_bypass_disable= 4333 [HW] Control Speculative Store Bypass (SSB) Disable mitigation 4334 (Speculative Store Bypass vulnerability) 4335 4336 Certain CPUs are vulnerable to an exploit against a 4337 a common industry wide performance optimization known 4338 as "Speculative Store Bypass" in which recent stores 4339 to the same memory location may not be observed by 4340 later loads during speculative execution. The idea 4341 is that such stores are unlikely and that they can 4342 be detected prior to instruction retirement at the 4343 end of a particular speculation execution window. 4344 4345 In vulnerable processors, the speculatively forwarded 4346 store can be used in a cache side channel attack, for 4347 example to read memory to which the attacker does not 4348 directly have access (e.g. inside sandboxed code). 4349 4350 This parameter controls whether the Speculative Store 4351 Bypass optimization is used. 4352 4353 On x86 the options are: 4354 4355 on - Unconditionally disable Speculative Store Bypass 4356 off - Unconditionally enable Speculative Store Bypass 4357 auto - Kernel detects whether the CPU model contains an 4358 implementation of Speculative Store Bypass and 4359 picks the most appropriate mitigation. If the 4360 CPU is not vulnerable, "off" is selected. If the 4361 CPU is vulnerable the default mitigation is 4362 architecture and Kconfig dependent. See below. 4363 prctl - Control Speculative Store Bypass per thread 4364 via prctl. Speculative Store Bypass is enabled 4365 for a process by default. The state of the control 4366 is inherited on fork. 4367 seccomp - Same as "prctl" above, but all seccomp threads 4368 will disable SSB unless they explicitly opt out. 4369 4370 Default mitigations: 4371 X86: If CONFIG_SECCOMP=y "seccomp", otherwise "prctl" 4372 4373 On powerpc the options are: 4374 4375 on,auto - On Power8 and Power9 insert a store-forwarding 4376 barrier on kernel entry and exit. On Power7 4377 perform a software flush on kernel entry and 4378 exit. 4379 off - No action. 4380 4381 Not specifying this option is equivalent to 4382 spec_store_bypass_disable=auto. 4383 4384 spia_io_base= [HW,MTD] 4385 spia_fio_base= 4386 spia_pedr= 4387 spia_peddr= 4388 4389 srcutree.counter_wrap_check [KNL] 4390 Specifies how frequently to check for 4391 grace-period sequence counter wrap for the 4392 srcu_data structure's ->srcu_gp_seq_needed field. 4393 The greater the number of bits set in this kernel 4394 parameter, the less frequently counter wrap will 4395 be checked for. Note that the bottom two bits 4396 are ignored. 4397 4398 srcutree.exp_holdoff [KNL] 4399 Specifies how many nanoseconds must elapse 4400 since the end of the last SRCU grace period for 4401 a given srcu_struct until the next normal SRCU 4402 grace period will be considered for automatic 4403 expediting. Set to zero to disable automatic 4404 expediting. 4405 4406 ssbd= [ARM64,HW] 4407 Speculative Store Bypass Disable control 4408 4409 On CPUs that are vulnerable to the Speculative 4410 Store Bypass vulnerability and offer a 4411 firmware based mitigation, this parameter 4412 indicates how the mitigation should be used: 4413 4414 force-on: Unconditionally enable mitigation for 4415 for both kernel and userspace 4416 force-off: Unconditionally disable mitigation for 4417 for both kernel and userspace 4418 kernel: Always enable mitigation in the 4419 kernel, and offer a prctl interface 4420 to allow userspace to register its 4421 interest in being mitigated too. 4422 4423 stack_guard_gap= [MM] 4424 override the default stack gap protection. The value 4425 is in page units and it defines how many pages prior 4426 to (for stacks growing down) resp. after (for stacks 4427 growing up) the main stack are reserved for no other 4428 mapping. Default value is 256 pages. 4429 4430 stacktrace [FTRACE] 4431 Enabled the stack tracer on boot up. 4432 4433 stacktrace_filter=[function-list] 4434 [FTRACE] Limit the functions that the stack tracer 4435 will trace at boot up. function-list is a comma separated 4436 list of functions. This list can be changed at run 4437 time by the stack_trace_filter file in the debugfs 4438 tracing directory. Note, this enables stack tracing 4439 and the stacktrace above is not needed. 4440 4441 sti= [PARISC,HW] 4442 Format: <num> 4443 Set the STI (builtin display/keyboard on the HP-PARISC 4444 machines) console (graphic card) which should be used 4445 as the initial boot-console. 4446 See also comment in drivers/video/console/sticore.c. 4447 4448 sti_font= [HW] 4449 See comment in drivers/video/console/sticore.c. 4450 4451 stifb= [HW] 4452 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]] 4453 4454 sunrpc.min_resvport= 4455 sunrpc.max_resvport= 4456 [NFS,SUNRPC] 4457 SunRPC servers often require that client requests 4458 originate from a privileged port (i.e. a port in the 4459 range 0 < portnr < 1024). 4460 An administrator who wishes to reserve some of these 4461 ports for other uses may adjust the range that the 4462 kernel's sunrpc client considers to be privileged 4463 using these two parameters to set the minimum and 4464 maximum port values. 4465 4466 sunrpc.svc_rpc_per_connection_limit= 4467 [NFS,SUNRPC] 4468 Limit the number of requests that the server will 4469 process in parallel from a single connection. 4470 The default value is 0 (no limit). 4471 4472 sunrpc.pool_mode= 4473 [NFS] 4474 Control how the NFS server code allocates CPUs to 4475 service thread pools. Depending on how many NICs 4476 you have and where their interrupts are bound, this 4477 option will affect which CPUs will do NFS serving. 4478 Note: this parameter cannot be changed while the 4479 NFS server is running. 4480 4481 auto the server chooses an appropriate mode 4482 automatically using heuristics 4483 global a single global pool contains all CPUs 4484 percpu one pool for each CPU 4485 pernode one pool for each NUMA node (equivalent 4486 to global on non-NUMA machines) 4487 4488 sunrpc.tcp_slot_table_entries= 4489 sunrpc.udp_slot_table_entries= 4490 [NFS,SUNRPC] 4491 Sets the upper limit on the number of simultaneous 4492 RPC calls that can be sent from the client to a 4493 server. Increasing these values may allow you to 4494 improve throughput, but will also increase the 4495 amount of memory reserved for use by the client. 4496 4497 suspend.pm_test_delay= 4498 [SUSPEND] 4499 Sets the number of seconds to remain in a suspend test 4500 mode before resuming the system (see 4501 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG 4502 is set. Default value is 5. 4503 4504 swapaccount=[0|1] 4505 [KNL] Enable accounting of swap in memory resource 4506 controller if no parameter or 1 is given or disable 4507 it if 0 is given (See Documentation/cgroup-v1/memory.txt) 4508 4509 swiotlb= [ARM,IA-64,PPC,MIPS,X86] 4510 Format: { <int> | force | noforce } 4511 <int> -- Number of I/O TLB slabs 4512 force -- force using of bounce buffers even if they 4513 wouldn't be automatically used by the kernel 4514 noforce -- Never use bounce buffers (for debugging) 4515 4516 switches= [HW,M68k] 4517 4518 sysfs.deprecated=0|1 [KNL] 4519 Enable/disable old style sysfs layout for old udev 4520 on older distributions. When this option is enabled 4521 very new udev will not work anymore. When this option 4522 is disabled (or CONFIG_SYSFS_DEPRECATED not compiled) 4523 in older udev will not work anymore. 4524 Default depends on CONFIG_SYSFS_DEPRECATED_V2 set in 4525 the kernel configuration. 4526 4527 sysrq_always_enabled 4528 [KNL] 4529 Ignore sysrq setting - this boot parameter will 4530 neutralize any effect of /proc/sys/kernel/sysrq. 4531 Useful for debugging. 4532 4533 tcpmhash_entries= [KNL,NET] 4534 Set the number of tcp_metrics_hash slots. 4535 Default value is 8192 or 16384 depending on total 4536 ram pages. This is used to specify the TCP metrics 4537 cache size. See Documentation/networking/ip-sysctl.txt 4538 "tcp_no_metrics_save" section for more details. 4539 4540 tdfx= [HW,DRM] 4541 4542 test_suspend= [SUSPEND][,N] 4543 Specify "mem" (for Suspend-to-RAM) or "standby" (for 4544 standby suspend) or "freeze" (for suspend type freeze) 4545 as the system sleep state during system startup with 4546 the optional capability to repeat N number of times. 4547 The system is woken from this state using a 4548 wakeup-capable RTC alarm. 4549 4550 thash_entries= [KNL,NET] 4551 Set number of hash buckets for TCP connection 4552 4553 thermal.act= [HW,ACPI] 4554 -1: disable all active trip points in all thermal zones 4555 <degrees C>: override all lowest active trip points 4556 4557 thermal.crt= [HW,ACPI] 4558 -1: disable all critical trip points in all thermal zones 4559 <degrees C>: override all critical trip points 4560 4561 thermal.nocrt= [HW,ACPI] 4562 Set to disable actions on ACPI thermal zone 4563 critical and hot trip points. 4564 4565 thermal.off= [HW,ACPI] 4566 1: disable ACPI thermal control 4567 4568 thermal.psv= [HW,ACPI] 4569 -1: disable all passive trip points 4570 <degrees C>: override all passive trip points to this 4571 value 4572 4573 thermal.tzp= [HW,ACPI] 4574 Specify global default ACPI thermal zone polling rate 4575 <deci-seconds>: poll all this frequency 4576 0: no polling (default) 4577 4578 threadirqs [KNL] 4579 Force threading of all interrupt handlers except those 4580 marked explicitly IRQF_NO_THREAD. 4581 4582 tmem [KNL,XEN] 4583 Enable the Transcendent memory driver if built-in. 4584 4585 tmem.cleancache=0|1 [KNL, XEN] 4586 Default is on (1). Disable the usage of the cleancache 4587 API to send anonymous pages to the hypervisor. 4588 4589 tmem.frontswap=0|1 [KNL, XEN] 4590 Default is on (1). Disable the usage of the frontswap 4591 API to send swap pages to the hypervisor. If disabled 4592 the selfballooning and selfshrinking are force disabled. 4593 4594 tmem.selfballooning=0|1 [KNL, XEN] 4595 Default is on (1). Disable the driving of swap pages 4596 to the hypervisor. 4597 4598 tmem.selfshrinking=0|1 [KNL, XEN] 4599 Default is on (1). Partial swapoff that immediately 4600 transfers pages from Xen hypervisor back to the 4601 kernel based on different criteria. 4602 4603 topology= [S390] 4604 Format: {off | on} 4605 Specify if the kernel should make use of the cpu 4606 topology information if the hardware supports this. 4607 The scheduler will make use of this information and 4608 e.g. base its process migration decisions on it. 4609 Default is on. 4610 4611 topology_updates= [KNL, PPC, NUMA] 4612 Format: {off} 4613 Specify if the kernel should ignore (off) 4614 topology updates sent by the hypervisor to this 4615 LPAR. 4616 4617 tp720= [HW,PS2] 4618 4619 tpm_suspend_pcr=[HW,TPM] 4620 Format: integer pcr id 4621 Specify that at suspend time, the tpm driver 4622 should extend the specified pcr with zeros, 4623 as a workaround for some chips which fail to 4624 flush the last written pcr on TPM_SaveState. 4625 This will guarantee that all the other pcrs 4626 are saved. 4627 4628 trace_buf_size=nn[KMG] 4629 [FTRACE] will set tracing buffer size on each cpu. 4630 4631 trace_event=[event-list] 4632 [FTRACE] Set and start specified trace events in order 4633 to facilitate early boot debugging. The event-list is a 4634 comma separated list of trace events to enable. See 4635 also Documentation/trace/events.rst 4636 4637 trace_options=[option-list] 4638 [FTRACE] Enable or disable tracer options at boot. 4639 The option-list is a comma delimited list of options 4640 that can be enabled or disabled just as if you were 4641 to echo the option name into 4642 4643 /sys/kernel/debug/tracing/trace_options 4644 4645 For example, to enable stacktrace option (to dump the 4646 stack trace of each event), add to the command line: 4647 4648 trace_options=stacktrace 4649 4650 See also Documentation/trace/ftrace.rst "trace options" 4651 section. 4652 4653 tp_printk[FTRACE] 4654 Have the tracepoints sent to printk as well as the 4655 tracing ring buffer. This is useful for early boot up 4656 where the system hangs or reboots and does not give the 4657 option for reading the tracing buffer or performing a 4658 ftrace_dump_on_oops. 4659 4660 To turn off having tracepoints sent to printk, 4661 echo 0 > /proc/sys/kernel/tracepoint_printk 4662 Note, echoing 1 into this file without the 4663 tracepoint_printk kernel cmdline option has no effect. 4664 4665 ** CAUTION ** 4666 4667 Having tracepoints sent to printk() and activating high 4668 frequency tracepoints such as irq or sched, can cause 4669 the system to live lock. 4670 4671 traceoff_on_warning 4672 [FTRACE] enable this option to disable tracing when a 4673 warning is hit. This turns off "tracing_on". Tracing can 4674 be enabled again by echoing '1' into the "tracing_on" 4675 file located in /sys/kernel/debug/tracing/ 4676 4677 This option is useful, as it disables the trace before 4678 the WARNING dump is called, which prevents the trace to 4679 be filled with content caused by the warning output. 4680 4681 This option can also be set at run time via the sysctl 4682 option: kernel/traceoff_on_warning 4683 4684 transparent_hugepage= 4685 [KNL] 4686 Format: [always|madvise|never] 4687 Can be used to control the default behavior of the system 4688 with respect to transparent hugepages. 4689 See Documentation/admin-guide/mm/transhuge.rst 4690 for more details. 4691 4692 tsc= Disable clocksource stability checks for TSC. 4693 Format: <string> 4694 [x86] reliable: mark tsc clocksource as reliable, this 4695 disables clocksource verification at runtime, as well 4696 as the stability checks done at bootup. Used to enable 4697 high-resolution timer mode on older hardware, and in 4698 virtualized environment. 4699 [x86] noirqtime: Do not use TSC to do irq accounting. 4700 Used to run time disable IRQ_TIME_ACCOUNTING on any 4701 platforms where RDTSC is slow and this accounting 4702 can add overhead. 4703 [x86] unstable: mark the TSC clocksource as unstable, this 4704 marks the TSC unconditionally unstable at bootup and 4705 avoids any further wobbles once the TSC watchdog notices. 4706 4707 turbografx.map[2|3]= [HW,JOY] 4708 TurboGraFX parallel port interface 4709 Format: 4710 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7> 4711 See also Documentation/input/devices/joystick-parport.rst 4712 4713 udbg-immortal [PPC] When debugging early kernel crashes that 4714 happen after console_init() and before a proper 4715 console driver takes over, this boot options might 4716 help "seeing" what's going on. 4717 4718 uhash_entries= [KNL,NET] 4719 Set number of hash buckets for UDP/UDP-Lite connections 4720 4721 uhci-hcd.ignore_oc= 4722 [USB] Ignore overcurrent events (default N). 4723 Some badly-designed motherboards generate lots of 4724 bogus events, for ports that aren't wired to 4725 anything. Set this parameter to avoid log spamming. 4726 Note that genuine overcurrent events won't be 4727 reported either. 4728 4729 unknown_nmi_panic 4730 [X86] Cause panic on unknown NMI. 4731 4732 usbcore.authorized_default= 4733 [USB] Default USB device authorization: 4734 (default -1 = authorized except for wireless USB, 4735 0 = not authorized, 1 = authorized, 2 = authorized 4736 if device connected to internal port) 4737 4738 usbcore.autosuspend= 4739 [USB] The autosuspend time delay (in seconds) used 4740 for newly-detected USB devices (default 2). This 4741 is the time required before an idle device will be 4742 autosuspended. Devices for which the delay is set 4743 to a negative value won't be autosuspended at all. 4744 4745 usbcore.usbfs_snoop= 4746 [USB] Set to log all usbfs traffic (default 0 = off). 4747 4748 usbcore.usbfs_snoop_max= 4749 [USB] Maximum number of bytes to snoop in each URB 4750 (default = 65536). 4751 4752 usbcore.blinkenlights= 4753 [USB] Set to cycle leds on hubs (default 0 = off). 4754 4755 usbcore.old_scheme_first= 4756 [USB] Start with the old device initialization 4757 scheme, applies only to low and full-speed devices 4758 (default 0 = off). 4759 4760 usbcore.usbfs_memory_mb= 4761 [USB] Memory limit (in MB) for buffers allocated by 4762 usbfs (default = 16, 0 = max = 2047). 4763 4764 usbcore.use_both_schemes= 4765 [USB] Try the other device initialization scheme 4766 if the first one fails (default 1 = enabled). 4767 4768 usbcore.initial_descriptor_timeout= 4769 [USB] Specifies timeout for the initial 64-byte 4770 USB_REQ_GET_DESCRIPTOR request in milliseconds 4771 (default 5000 = 5.0 seconds). 4772 4773 usbcore.nousb [USB] Disable the USB subsystem 4774 4775 usbcore.quirks= 4776 [USB] A list of quirk entries to augment the built-in 4777 usb core quirk list. List entries are separated by 4778 commas. Each entry has the form 4779 VendorID:ProductID:Flags. The IDs are 4-digit hex 4780 numbers and Flags is a set of letters. Each letter 4781 will change the built-in quirk; setting it if it is 4782 clear and clearing it if it is set. The letters have 4783 the following meanings: 4784 a = USB_QUIRK_STRING_FETCH_255 (string 4785 descriptors must not be fetched using 4786 a 255-byte read); 4787 b = USB_QUIRK_RESET_RESUME (device can't resume 4788 correctly so reset it instead); 4789 c = USB_QUIRK_NO_SET_INTF (device can't handle 4790 Set-Interface requests); 4791 d = USB_QUIRK_CONFIG_INTF_STRINGS (device can't 4792 handle its Configuration or Interface 4793 strings); 4794 e = USB_QUIRK_RESET (device can't be reset 4795 (e.g morph devices), don't use reset); 4796 f = USB_QUIRK_HONOR_BNUMINTERFACES (device has 4797 more interface descriptions than the 4798 bNumInterfaces count, and can't handle 4799 talking to these interfaces); 4800 g = USB_QUIRK_DELAY_INIT (device needs a pause 4801 during initialization, after we read 4802 the device descriptor); 4803 h = USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL (For 4804 high speed and super speed interrupt 4805 endpoints, the USB 2.0 and USB 3.0 spec 4806 require the interval in microframes (1 4807 microframe = 125 microseconds) to be 4808 calculated as interval = 2 ^ 4809 (bInterval-1). 4810 Devices with this quirk report their 4811 bInterval as the result of this 4812 calculation instead of the exponent 4813 variable used in the calculation); 4814 i = USB_QUIRK_DEVICE_QUALIFIER (device can't 4815 handle device_qualifier descriptor 4816 requests); 4817 j = USB_QUIRK_IGNORE_REMOTE_WAKEUP (device 4818 generates spurious wakeup, ignore 4819 remote wakeup capability); 4820 k = USB_QUIRK_NO_LPM (device can't handle Link 4821 Power Management); 4822 l = USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL 4823 (Device reports its bInterval as linear 4824 frames instead of the USB 2.0 4825 calculation); 4826 m = USB_QUIRK_DISCONNECT_SUSPEND (Device needs 4827 to be disconnected before suspend to 4828 prevent spurious wakeup); 4829 n = USB_QUIRK_DELAY_CTRL_MSG (Device needs a 4830 pause after every control message); 4831 o = USB_QUIRK_HUB_SLOW_RESET (Hub needs extra 4832 delay after resetting its port); 4833 Example: quirks=0781:5580:bk,0a5c:5834:gij 4834 4835 usbhid.mousepoll= 4836 [USBHID] The interval which mice are to be polled at. 4837 4838 usbhid.jspoll= 4839 [USBHID] The interval which joysticks are to be polled at. 4840 4841 usbhid.kbpoll= 4842 [USBHID] The interval which keyboards are to be polled at. 4843 4844 usb-storage.delay_use= 4845 [UMS] The delay in seconds before a new device is 4846 scanned for Logical Units (default 1). 4847 4848 usb-storage.quirks= 4849 [UMS] A list of quirks entries to supplement or 4850 override the built-in unusual_devs list. List 4851 entries are separated by commas. Each entry has 4852 the form VID:PID:Flags where VID and PID are Vendor 4853 and Product ID values (4-digit hex numbers) and 4854 Flags is a set of characters, each corresponding 4855 to a common usb-storage quirk flag as follows: 4856 a = SANE_SENSE (collect more than 18 bytes 4857 of sense data); 4858 b = BAD_SENSE (don't collect more than 18 4859 bytes of sense data); 4860 c = FIX_CAPACITY (decrease the reported 4861 device capacity by one sector); 4862 d = NO_READ_DISC_INFO (don't use 4863 READ_DISC_INFO command); 4864 e = NO_READ_CAPACITY_16 (don't use 4865 READ_CAPACITY_16 command); 4866 f = NO_REPORT_OPCODES (don't use report opcodes 4867 command, uas only); 4868 g = MAX_SECTORS_240 (don't transfer more than 4869 240 sectors at a time, uas only); 4870 h = CAPACITY_HEURISTICS (decrease the 4871 reported device capacity by one 4872 sector if the number is odd); 4873 i = IGNORE_DEVICE (don't bind to this 4874 device); 4875 j = NO_REPORT_LUNS (don't use report luns 4876 command, uas only); 4877 l = NOT_LOCKABLE (don't try to lock and 4878 unlock ejectable media); 4879 m = MAX_SECTORS_64 (don't transfer more 4880 than 64 sectors = 32 KB at a time); 4881 n = INITIAL_READ10 (force a retry of the 4882 initial READ(10) command); 4883 o = CAPACITY_OK (accept the capacity 4884 reported by the device); 4885 p = WRITE_CACHE (the device cache is ON 4886 by default); 4887 r = IGNORE_RESIDUE (the device reports 4888 bogus residue values); 4889 s = SINGLE_LUN (the device has only one 4890 Logical Unit); 4891 t = NO_ATA_1X (don't allow ATA(12) and ATA(16) 4892 commands, uas only); 4893 u = IGNORE_UAS (don't bind to the uas driver); 4894 w = NO_WP_DETECT (don't test whether the 4895 medium is write-protected). 4896 y = ALWAYS_SYNC (issue a SYNCHRONIZE_CACHE 4897 even if the device claims no cache) 4898 Example: quirks=0419:aaf5:rl,0421:0433:rc 4899 4900 user_debug= [KNL,ARM] 4901 Format: <int> 4902 See arch/arm/Kconfig.debug help text. 4903 1 - undefined instruction events 4904 2 - system calls 4905 4 - invalid data aborts 4906 8 - SIGSEGV faults 4907 16 - SIGBUS faults 4908 Example: user_debug=31 4909 4910 userpte= 4911 [X86] Flags controlling user PTE allocations. 4912 4913 nohigh = do not allocate PTE pages in 4914 HIGHMEM regardless of setting 4915 of CONFIG_HIGHPTE. 4916 4917 vdso= [X86,SH] 4918 On X86_32, this is an alias for vdso32=. Otherwise: 4919 4920 vdso=1: enable VDSO (the default) 4921 vdso=0: disable VDSO mapping 4922 4923 vdso32= [X86] Control the 32-bit vDSO 4924 vdso32=1: enable 32-bit VDSO 4925 vdso32=0 or vdso32=2: disable 32-bit VDSO 4926 4927 See the help text for CONFIG_COMPAT_VDSO for more 4928 details. If CONFIG_COMPAT_VDSO is set, the default is 4929 vdso32=0; otherwise, the default is vdso32=1. 4930 4931 For compatibility with older kernels, vdso32=2 is an 4932 alias for vdso32=0. 4933 4934 Try vdso32=0 if you encounter an error that says: 4935 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed! 4936 4937 vector= [IA-64,SMP] 4938 vector=percpu: enable percpu vector domain 4939 4940 video= [FB] Frame buffer configuration 4941 See Documentation/fb/modedb.txt. 4942 4943 video.brightness_switch_enabled= [0,1] 4944 If set to 1, on receiving an ACPI notify event 4945 generated by hotkey, video driver will adjust brightness 4946 level and then send out the event to user space through 4947 the allocated input device; If set to 0, video driver 4948 will only send out the event without touching backlight 4949 brightness level. 4950 default: 1 4951 4952 virtio_mmio.device= 4953 [VMMIO] Memory mapped virtio (platform) device. 4954 4955 <size>@<baseaddr>:<irq>[:<id>] 4956 where: 4957 <size> := size (can use standard suffixes 4958 like K, M and G) 4959 <baseaddr> := physical base address 4960 <irq> := interrupt number (as passed to 4961 request_irq()) 4962 <id> := (optional) platform device id 4963 example: 4964 virtio_mmio.device=1K@0x100b0000:48:7 4965 4966 Can be used multiple times for multiple devices. 4967 4968 vga= [BOOT,X86-32] Select a particular video mode 4969 See Documentation/x86/boot.txt and 4970 Documentation/svga.txt. 4971 Use vga=ask for menu. 4972 This is actually a boot loader parameter; the value is 4973 passed to the kernel using a special protocol. 4974 4975 vm_debug[=options] [KNL] Available with CONFIG_DEBUG_VM=y. 4976 May slow down system boot speed, especially when 4977 enabled on systems with a large amount of memory. 4978 All options are enabled by default, and this 4979 interface is meant to allow for selectively 4980 enabling or disabling specific virtual memory 4981 debugging features. 4982 4983 Available options are: 4984 P Enable page structure init time poisoning 4985 - Disable all of the above options 4986 4987 vmalloc=nn[KMG] [KNL,BOOT] Forces the vmalloc area to have an exact 4988 size of <nn>. This can be used to increase the 4989 minimum size (128MB on x86). It can also be used to 4990 decrease the size and leave more room for directly 4991 mapped kernel RAM. 4992 4993 vmcp_cma=nn[MG] [KNL,S390] 4994 Sets the memory size reserved for contiguous memory 4995 allocations for the vmcp device driver. 4996 4997 vmhalt= [KNL,S390] Perform z/VM CP command after system halt. 4998 Format: <command> 4999 5000 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic. 5001 Format: <command> 5002 5003 vmpoff= [KNL,S390] Perform z/VM CP command after power off. 5004 Format: <command> 5005 5006 vsyscall= [X86-64] 5007 Controls the behavior of vsyscalls (i.e. calls to 5008 fixed addresses of 0xffffffffff600x00 from legacy 5009 code). Most statically-linked binaries and older 5010 versions of glibc use these calls. Because these 5011 functions are at fixed addresses, they make nice 5012 targets for exploits that can control RIP. 5013 5014 emulate [default] Vsyscalls turn into traps and are 5015 emulated reasonably safely. 5016 5017 native Vsyscalls are native syscall instructions. 5018 This is a little bit faster than trapping 5019 and makes a few dynamic recompilers work 5020 better than they would in emulation mode. 5021 It also makes exploits much easier to write. 5022 5023 none Vsyscalls don't work at all. This makes 5024 them quite hard to use for exploits but 5025 might break your system. 5026 5027 vt.color= [VT] Default text color. 5028 Format: 0xYX, X = foreground, Y = background. 5029 Default: 0x07 = light gray on black. 5030 5031 vt.cur_default= [VT] Default cursor shape. 5032 Format: 0xCCBBAA, where AA, BB, and CC are the same as 5033 the parameters of the <Esc>[?A;B;Cc escape sequence; 5034 see VGA-softcursor.txt. Default: 2 = underline. 5035 5036 vt.default_blu= [VT] 5037 Format: <blue0>,<blue1>,<blue2>,...,<blue15> 5038 Change the default blue palette of the console. 5039 This is a 16-member array composed of values 5040 ranging from 0-255. 5041 5042 vt.default_grn= [VT] 5043 Format: <green0>,<green1>,<green2>,...,<green15> 5044 Change the default green palette of the console. 5045 This is a 16-member array composed of values 5046 ranging from 0-255. 5047 5048 vt.default_red= [VT] 5049 Format: <red0>,<red1>,<red2>,...,<red15> 5050 Change the default red palette of the console. 5051 This is a 16-member array composed of values 5052 ranging from 0-255. 5053 5054 vt.default_utf8= 5055 [VT] 5056 Format=<0|1> 5057 Set system-wide default UTF-8 mode for all tty's. 5058 Default is 1, i.e. UTF-8 mode is enabled for all 5059 newly opened terminals. 5060 5061 vt.global_cursor_default= 5062 [VT] 5063 Format=<-1|0|1> 5064 Set system-wide default for whether a cursor 5065 is shown on new VTs. Default is -1, 5066 i.e. cursors will be created by default unless 5067 overridden by individual drivers. 0 will hide 5068 cursors, 1 will display them. 5069 5070 vt.italic= [VT] Default color for italic text; 0-15. 5071 Default: 2 = green. 5072 5073 vt.underline= [VT] Default color for underlined text; 0-15. 5074 Default: 3 = cyan. 5075 5076 watchdog timers [HW,WDT] For information on watchdog timers, 5077 see Documentation/watchdog/watchdog-parameters.txt 5078 or other driver-specific files in the 5079 Documentation/watchdog/ directory. 5080 5081 watchdog_thresh= 5082 [KNL] 5083 Set the hard lockup detector stall duration 5084 threshold in seconds. The soft lockup detector 5085 threshold is set to twice the value. A value of 0 5086 disables both lockup detectors. Default is 10 5087 seconds. 5088 5089 workqueue.watchdog_thresh= 5090 If CONFIG_WQ_WATCHDOG is configured, workqueue can 5091 warn stall conditions and dump internal state to 5092 help debugging. 0 disables workqueue stall 5093 detection; otherwise, it's the stall threshold 5094 duration in seconds. The default value is 30 and 5095 it can be updated at runtime by writing to the 5096 corresponding sysfs file. 5097 5098 workqueue.disable_numa 5099 By default, all work items queued to unbound 5100 workqueues are affine to the NUMA nodes they're 5101 issued on, which results in better behavior in 5102 general. If NUMA affinity needs to be disabled for 5103 whatever reason, this option can be used. Note 5104 that this also can be controlled per-workqueue for 5105 workqueues visible under /sys/bus/workqueue/. 5106 5107 workqueue.power_efficient 5108 Per-cpu workqueues are generally preferred because 5109 they show better performance thanks to cache 5110 locality; unfortunately, per-cpu workqueues tend to 5111 be more power hungry than unbound workqueues. 5112 5113 Enabling this makes the per-cpu workqueues which 5114 were observed to contribute significantly to power 5115 consumption unbound, leading to measurably lower 5116 power usage at the cost of small performance 5117 overhead. 5118 5119 The default value of this parameter is determined by 5120 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT. 5121 5122 workqueue.debug_force_rr_cpu 5123 Workqueue used to implicitly guarantee that work 5124 items queued without explicit CPU specified are put 5125 on the local CPU. This guarantee is no longer true 5126 and while local CPU is still preferred work items 5127 may be put on foreign CPUs. This debug option 5128 forces round-robin CPU selection to flush out 5129 usages which depend on the now broken guarantee. 5130 When enabled, memory and cache locality will be 5131 impacted. 5132 5133 x2apic_phys [X86-64,APIC] Use x2apic physical mode instead of 5134 default x2apic cluster mode on platforms 5135 supporting x2apic. 5136 5137 x86_intel_mid_timer= [X86-32,APBT] 5138 Choose timer option for x86 Intel MID platform. 5139 Two valid options are apbt timer only and lapic timer 5140 plus one apbt timer for broadcast timer. 5141 x86_intel_mid_timer=apbt_only | lapic_and_apbt 5142 5143 xen_512gb_limit [KNL,X86-64,XEN] 5144 Restricts the kernel running paravirtualized under Xen 5145 to use only up to 512 GB of RAM. The reason to do so is 5146 crash analysis tools and Xen tools for doing domain 5147 save/restore/migration must be enabled to handle larger 5148 domains. 5149 5150 xen_emul_unplug= [HW,X86,XEN] 5151 Unplug Xen emulated devices 5152 Format: [unplug0,][unplug1] 5153 ide-disks -- unplug primary master IDE devices 5154 aux-ide-disks -- unplug non-primary-master IDE devices 5155 nics -- unplug network devices 5156 all -- unplug all emulated devices (NICs and IDE disks) 5157 unnecessary -- unplugging emulated devices is 5158 unnecessary even if the host did not respond to 5159 the unplug protocol 5160 never -- do not unplug even if version check succeeds 5161 5162 xen_nopvspin [X86,XEN] 5163 Disables the ticketlock slowpath using Xen PV 5164 optimizations. 5165 5166 xen_nopv [X86] 5167 Disables the PV optimizations forcing the HVM guest to 5168 run as generic HVM guest with no PV drivers. 5169 5170 xen_scrub_pages= [XEN] 5171 Boolean option to control scrubbing pages before giving them back 5172 to Xen, for use by other domains. Can be also changed at runtime 5173 with /sys/devices/system/xen_memory/xen_memory0/scrub_pages. 5174 Default value controlled with CONFIG_XEN_SCRUB_PAGES_DEFAULT. 5175 5176 xirc2ps_cs= [NET,PCMCIA] 5177 Format: 5178 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]] 5179 5180 xhci-hcd.quirks [USB,KNL] 5181 A hex value specifying bitmask with supplemental xhci 5182 host controller quirks. Meaning of each bit can be 5183 consulted in header drivers/usb/host/xhci.h. 5184