1 acpi= [HW,ACPI,X86,ARM64,RISCV64] 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,riscv64] 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 and RISCV64, ONLY "acpi=off", "acpi=on" or 14 "acpi=force" are available 15 16 See also Documentation/power/runtime_pm.rst, pci=noacpi 17 18 acpi_apic_instance= [ACPI, IOAPIC] 19 Format: <int> 20 2: use 2nd APIC table, if available 21 1,0: use 1st APIC table 22 default: 0 23 24 acpi_backlight= [HW,ACPI] 25 { vendor | video | native | none } 26 If set to vendor, prefer vendor-specific driver 27 (e.g. thinkpad_acpi, sony_acpi, etc.) instead 28 of the ACPI video.ko driver. 29 If set to video, use the ACPI video.ko driver. 30 If set to native, use the device's native backlight mode. 31 If set to none, disable the ACPI backlight interface. 32 33 acpi_force_32bit_fadt_addr 34 force FADT to use 32 bit addresses rather than the 35 64 bit X_* addresses. Some firmware have broken 64 36 bit addresses for force ACPI ignore these and use 37 the older legacy 32 bit addresses. 38 39 acpica_no_return_repair [HW, ACPI] 40 Disable AML predefined validation mechanism 41 This mechanism can repair the evaluation result to make 42 the return objects more ACPI specification compliant. 43 This option is useful for developers to identify the 44 root cause of an AML interpreter issue when the issue 45 has something to do with the repair mechanism. 46 47 acpi.debug_layer= [HW,ACPI,ACPI_DEBUG] 48 acpi.debug_level= [HW,ACPI,ACPI_DEBUG] 49 Format: <int> 50 CONFIG_ACPI_DEBUG must be enabled to produce any ACPI 51 debug output. Bits in debug_layer correspond to a 52 _COMPONENT in an ACPI source file, e.g., 53 #define _COMPONENT ACPI_EVENTS 54 Bits in debug_level correspond to a level in 55 ACPI_DEBUG_PRINT statements, e.g., 56 ACPI_DEBUG_PRINT((ACPI_DB_INFO, ... 57 The debug_level mask defaults to "info". See 58 Documentation/firmware-guide/acpi/debug.rst for more information about 59 debug layers and levels. 60 61 Enable processor driver info messages: 62 acpi.debug_layer=0x20000000 63 Enable 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: <byte> or <bitmap-list> 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_no_watchdog [HW,ACPI,WDT] 140 Ignore the ACPI-based watchdog interface (WDAT) and let 141 a native driver control the watchdog device instead. 142 143 acpi_rsdp= [ACPI,EFI,KEXEC] 144 Pass the RSDP address to the kernel, mostly used 145 on machines running EFI runtime service to boot the 146 second kernel for kdump. 147 148 acpi_os_name= [HW,ACPI] Tell ACPI BIOS the name of the OS 149 Format: To spoof as Windows 98: ="Microsoft Windows" 150 151 acpi_rev_override [ACPI] Override the _REV object to return 5 (instead 152 of 2 which is mandated by ACPI 6) as the supported ACPI 153 specification revision (when using this switch, it may 154 be necessary to carry out a cold reboot _twice_ in a 155 row to make it take effect on the platform firmware). 156 157 acpi_osi= [HW,ACPI] Modify list of supported OS interface strings 158 acpi_osi="string1" # add string1 159 acpi_osi="!string2" # remove string2 160 acpi_osi=!* # remove all strings 161 acpi_osi=! # disable all built-in OS vendor 162 strings 163 acpi_osi=!! # enable all built-in OS vendor 164 strings 165 acpi_osi= # disable all strings 166 167 'acpi_osi=!' can be used in combination with single or 168 multiple 'acpi_osi="string1"' to support specific OS 169 vendor string(s). Note that such command can only 170 affect the default state of the OS vendor strings, thus 171 it cannot affect the default state of the feature group 172 strings and the current state of the OS vendor strings, 173 specifying it multiple times through kernel command line 174 is meaningless. This command is useful when one do not 175 care about the state of the feature group strings which 176 should be controlled by the OSPM. 177 Examples: 178 1. 'acpi_osi=! acpi_osi="Windows 2000"' is equivalent 179 to 'acpi_osi="Windows 2000" acpi_osi=!', they all 180 can make '_OSI("Windows 2000")' TRUE. 181 182 'acpi_osi=' cannot be used in combination with other 183 'acpi_osi=' command lines, the _OSI method will not 184 exist in the ACPI namespace. NOTE that such command can 185 only affect the _OSI support state, thus specifying it 186 multiple times through kernel command line is also 187 meaningless. 188 Examples: 189 1. 'acpi_osi=' can make 'CondRefOf(_OSI, Local1)' 190 FALSE. 191 192 'acpi_osi=!*' can be used in combination with single or 193 multiple 'acpi_osi="string1"' to support specific 194 string(s). Note that such command can affect the 195 current state of both the OS vendor strings and the 196 feature group strings, thus specifying it multiple times 197 through kernel command line is meaningful. But it may 198 still not able to affect the final state of a string if 199 there are quirks related to this string. This command 200 is useful when one want to control the state of the 201 feature group strings to debug BIOS issues related to 202 the OSPM features. 203 Examples: 204 1. 'acpi_osi="Module Device" acpi_osi=!*' can make 205 '_OSI("Module Device")' FALSE. 206 2. 'acpi_osi=!* acpi_osi="Module Device"' can make 207 '_OSI("Module Device")' TRUE. 208 3. 'acpi_osi=! acpi_osi=!* acpi_osi="Windows 2000"' is 209 equivalent to 210 'acpi_osi=!* acpi_osi=! acpi_osi="Windows 2000"' 211 and 212 'acpi_osi=!* acpi_osi="Windows 2000" acpi_osi=!', 213 they all will make '_OSI("Windows 2000")' TRUE. 214 215 acpi_pm_good [X86] 216 Override the pmtimer bug detection: force the kernel 217 to assume that this machine's pmtimer latches its value 218 and always returns good values. 219 220 acpi_sci= [HW,ACPI] ACPI System Control Interrupt trigger mode 221 Format: { level | edge | high | low } 222 223 acpi_skip_timer_override [HW,ACPI] 224 Recognize and ignore IRQ0/pin2 Interrupt Override. 225 For broken nForce2 BIOS resulting in XT-PIC timer. 226 227 acpi_sleep= [HW,ACPI] Sleep options 228 Format: { s3_bios, s3_mode, s3_beep, s4_hwsig, 229 s4_nohwsig, old_ordering, nonvs, 230 sci_force_enable, nobl } 231 See Documentation/power/video.rst for information on 232 s3_bios and s3_mode. 233 s3_beep is for debugging; it makes the PC's speaker beep 234 as soon as the kernel's real-mode entry point is called. 235 s4_hwsig causes the kernel to check the ACPI hardware 236 signature during resume from hibernation, and gracefully 237 refuse to resume if it has changed. This complies with 238 the ACPI specification but not with reality, since 239 Windows does not do this and many laptops do change it 240 on docking. So the default behaviour is to allow resume 241 and simply warn when the signature changes, unless the 242 s4_hwsig option is enabled. 243 s4_nohwsig prevents ACPI hardware signature from being 244 used (or even warned about) during resume. 245 old_ordering causes the ACPI 1.0 ordering of the _PTS 246 control method, with respect to putting devices into 247 low power states, to be enforced (the ACPI 2.0 ordering 248 of _PTS is used by default). 249 nonvs prevents the kernel from saving/restoring the 250 ACPI NVS memory during suspend/hibernation and resume. 251 sci_force_enable causes the kernel to set SCI_EN directly 252 on resume from S1/S3 (which is against the ACPI spec, 253 but some broken systems don't work without it). 254 nobl causes the internal blacklist of systems known to 255 behave incorrectly in some ways with respect to system 256 suspend and resume to be ignored (use wisely). 257 258 acpi_use_timer_override [HW,ACPI] 259 Use timer override. For some broken Nvidia NF5 boards 260 that require a timer override, but don't have HPET 261 262 add_efi_memmap [EFI; X86] Include EFI memory map in 263 kernel's map of available physical RAM. 264 265 agp= [AGP] 266 { off | try_unsupported } 267 off: disable AGP support 268 try_unsupported: try to drive unsupported chipsets 269 (may crash computer or cause data corruption) 270 271 ALSA [HW,ALSA] 272 See Documentation/sound/alsa-configuration.rst 273 274 alignment= [KNL,ARM] 275 Allow the default userspace alignment fault handler 276 behaviour to be specified. Bit 0 enables warnings, 277 bit 1 enables fixups, and bit 2 sends a segfault. 278 279 align_va_addr= [X86-64] 280 Align virtual addresses by clearing slice [14:12] when 281 allocating a VMA at process creation time. This option 282 gives you up to 3% performance improvement on AMD F15h 283 machines (where it is enabled by default) for a 284 CPU-intensive style benchmark, and it can vary highly in 285 a microbenchmark depending on workload and compiler. 286 287 32: only for 32-bit processes 288 64: only for 64-bit processes 289 on: enable for both 32- and 64-bit processes 290 off: disable for both 32- and 64-bit processes 291 292 alloc_snapshot [FTRACE] 293 Allocate the ftrace snapshot buffer on boot up when the 294 main buffer is allocated. This is handy if debugging 295 and you need to use tracing_snapshot() on boot up, and 296 do not want to use tracing_snapshot_alloc() as it needs 297 to be done where GFP_KERNEL allocations are allowed. 298 299 allow_mismatched_32bit_el0 [ARM64] 300 Allow execve() of 32-bit applications and setting of the 301 PER_LINUX32 personality on systems where only a strict 302 subset of the CPUs support 32-bit EL0. When this 303 parameter is present, the set of CPUs supporting 32-bit 304 EL0 is indicated by /sys/devices/system/cpu/aarch32_el0 305 and hot-unplug operations may be restricted. 306 307 See Documentation/arch/arm64/asymmetric-32bit.rst for more 308 information. 309 310 amd_iommu= [HW,X86-64] 311 Pass parameters to the AMD IOMMU driver in the system. 312 Possible values are: 313 fullflush - Deprecated, equivalent to iommu.strict=1 314 off - do not initialize any AMD IOMMU found in 315 the system 316 force_isolation - Force device isolation for all 317 devices. The IOMMU driver is not 318 allowed anymore to lift isolation 319 requirements as needed. This option 320 does not override iommu=pt 321 force_enable - Force enable the IOMMU on platforms known 322 to be buggy with IOMMU enabled. Use this 323 option with care. 324 pgtbl_v1 - Use v1 page table for DMA-API (Default). 325 pgtbl_v2 - Use v2 page table for DMA-API. 326 irtcachedis - Disable Interrupt Remapping Table (IRT) caching. 327 328 amd_iommu_dump= [HW,X86-64] 329 Enable AMD IOMMU driver option to dump the ACPI table 330 for AMD IOMMU. With this option enabled, AMD IOMMU 331 driver will print ACPI tables for AMD IOMMU during 332 IOMMU initialization. 333 334 amd_iommu_intr= [HW,X86-64] 335 Specifies one of the following AMD IOMMU interrupt 336 remapping modes: 337 legacy - Use legacy interrupt remapping mode. 338 vapic - Use virtual APIC mode, which allows IOMMU 339 to inject interrupts directly into guest. 340 This mode requires kvm-amd.avic=1. 341 (Default when IOMMU HW support is present.) 342 343 amd_pstate= [X86] 344 disable 345 Do not enable amd_pstate as the default 346 scaling driver for the supported processors 347 passive 348 Use amd_pstate with passive mode as a scaling driver. 349 In this mode autonomous selection is disabled. 350 Driver requests a desired performance level and platform 351 tries to match the same performance level if it is 352 satisfied by guaranteed performance level. 353 active 354 Use amd_pstate_epp driver instance as the scaling driver, 355 driver provides a hint to the hardware if software wants 356 to bias toward performance (0x0) or energy efficiency (0xff) 357 to the CPPC firmware. then CPPC power algorithm will 358 calculate the runtime workload and adjust the realtime cores 359 frequency. 360 guided 361 Activate guided autonomous mode. Driver requests minimum and 362 maximum performance level and the platform autonomously 363 selects a performance level in this range and appropriate 364 to the current workload. 365 366 amijoy.map= [HW,JOY] Amiga joystick support 367 Map of devices attached to JOY0DAT and JOY1DAT 368 Format: <a>,<b> 369 See also Documentation/input/joydev/joystick.rst 370 371 analog.map= [HW,JOY] Analog joystick and gamepad support 372 Specifies type or capabilities of an analog joystick 373 connected to one of 16 gameports 374 Format: <type1>,<type2>,..<type16> 375 376 apc= [HW,SPARC] 377 Power management functions (SPARCstation-4/5 + deriv.) 378 Format: noidle 379 Disable APC CPU standby support. SPARCstation-Fox does 380 not play well with APC CPU idle - disable it if you have 381 APC and your system crashes randomly. 382 383 apic= [APIC,X86] Advanced Programmable Interrupt Controller 384 Change the output verbosity while booting 385 Format: { quiet (default) | verbose | debug } 386 Change the amount of debugging information output 387 when initialising the APIC and IO-APIC components. 388 For X86-32, this can also be used to specify an APIC 389 driver name. 390 Format: apic=driver_name 391 Examples: apic=bigsmp 392 393 apic_extnmi= [APIC,X86] External NMI delivery setting 394 Format: { bsp (default) | all | none } 395 bsp: External NMI is delivered only to CPU 0 396 all: External NMIs are broadcast to all CPUs as a 397 backup of CPU 0 398 none: External NMI is masked for all CPUs. This is 399 useful so that a dump capture kernel won't be 400 shot down by NMI 401 402 autoconf= [IPV6] 403 See Documentation/networking/ipv6.rst. 404 405 apm= [APM] Advanced Power Management 406 See header of arch/x86/kernel/apm_32.c. 407 408 apparmor= [APPARMOR] Disable or enable AppArmor at boot time 409 Format: { "0" | "1" } 410 See security/apparmor/Kconfig help text 411 0 -- disable. 412 1 -- enable. 413 Default value is set via kernel config option. 414 415 arcrimi= [HW,NET] ARCnet - "RIM I" (entirely mem-mapped) cards 416 Format: <io>,<irq>,<nodeID> 417 418 arm64.nobti [ARM64] Unconditionally disable Branch Target 419 Identification support 420 421 arm64.nopauth [ARM64] Unconditionally disable Pointer Authentication 422 support 423 424 arm64.nomte [ARM64] Unconditionally disable Memory Tagging Extension 425 support 426 427 arm64.nosve [ARM64] Unconditionally disable Scalable Vector 428 Extension support 429 430 arm64.nosme [ARM64] Unconditionally disable Scalable Matrix 431 Extension support 432 433 arm64.nomops [ARM64] Unconditionally disable Memory Copy and Memory 434 Set instructions support 435 436 ataflop= [HW,M68k] 437 438 atarimouse= [HW,MOUSE] Atari Mouse 439 440 atkbd.extra= [HW] Enable extra LEDs and keys on IBM RapidAccess, 441 EzKey and similar keyboards 442 443 atkbd.reset= [HW] Reset keyboard during initialization 444 445 atkbd.set= [HW] Select keyboard code set 446 Format: <int> (2 = AT (default), 3 = PS/2) 447 448 atkbd.scroll= [HW] Enable scroll wheel on MS Office and similar 449 keyboards 450 451 atkbd.softraw= [HW] Choose between synthetic and real raw mode 452 Format: <bool> (0 = real, 1 = synthetic (default)) 453 454 atkbd.softrepeat= [HW] 455 Use software keyboard repeat 456 457 audit= [KNL] Enable the audit sub-system 458 Format: { "0" | "1" | "off" | "on" } 459 0 | off - kernel audit is disabled and can not be 460 enabled until the next reboot 461 unset - kernel audit is initialized but disabled and 462 will be fully enabled by the userspace auditd. 463 1 | on - kernel audit is initialized and partially 464 enabled, storing at most audit_backlog_limit 465 messages in RAM until it is fully enabled by the 466 userspace auditd. 467 Default: unset 468 469 audit_backlog_limit= [KNL] Set the audit queue size limit. 470 Format: <int> (must be >=0) 471 Default: 64 472 473 bau= [X86_UV] Enable the BAU on SGI UV. The default 474 behavior is to disable the BAU (i.e. bau=0). 475 Format: { "0" | "1" } 476 0 - Disable the BAU. 477 1 - Enable the BAU. 478 unset - Disable the BAU. 479 480 baycom_epp= [HW,AX25] 481 Format: <io>,<mode> 482 483 baycom_par= [HW,AX25] BayCom Parallel Port AX.25 Modem 484 Format: <io>,<mode> 485 See header of drivers/net/hamradio/baycom_par.c. 486 487 baycom_ser_fdx= [HW,AX25] 488 BayCom Serial Port AX.25 Modem (Full Duplex Mode) 489 Format: <io>,<irq>,<mode>[,<baud>] 490 See header of drivers/net/hamradio/baycom_ser_fdx.c. 491 492 baycom_ser_hdx= [HW,AX25] 493 BayCom Serial Port AX.25 Modem (Half Duplex Mode) 494 Format: <io>,<irq>,<mode> 495 See header of drivers/net/hamradio/baycom_ser_hdx.c. 496 497 bert_disable [ACPI] 498 Disable BERT OS support on buggy BIOSes. 499 500 bgrt_disable [ACPI][X86] 501 Disable BGRT to avoid flickering OEM logo. 502 503 blkdevparts= Manual partition parsing of block device(s) for 504 embedded devices based on command line input. 505 See Documentation/block/cmdline-partition.rst 506 507 boot_delay= Milliseconds to delay each printk during boot. 508 Only works if CONFIG_BOOT_PRINTK_DELAY is enabled, 509 and you may also have to specify "lpj=". Boot_delay 510 values larger than 10 seconds (10000) are assumed 511 erroneous and ignored. 512 Format: integer 513 514 bootconfig [KNL] 515 Extended command line options can be added to an initrd 516 and this will cause the kernel to look for it. 517 518 See Documentation/admin-guide/bootconfig.rst 519 520 bttv.card= [HW,V4L] bttv (bt848 + bt878 based grabber cards) 521 bttv.radio= Most important insmod options are available as 522 kernel args too. 523 bttv.pll= See Documentation/admin-guide/media/bttv.rst 524 bttv.tuner= 525 526 bulk_remove=off [PPC] This parameter disables the use of the pSeries 527 firmware feature for flushing multiple hpte entries 528 at a time. 529 530 c101= [NET] Moxa C101 synchronous serial card 531 532 cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection. 533 Sometimes CPU hardware bugs make them report the cache 534 size incorrectly. The kernel will attempt work arounds 535 to fix known problems, but for some CPUs it is not 536 possible to determine what the correct size should be. 537 This option provides an override for these situations. 538 539 carrier_timeout= 540 [NET] Specifies amount of time (in seconds) that 541 the kernel should wait for a network carrier. By default 542 it waits 120 seconds. 543 544 ca_keys= [KEYS] This parameter identifies a specific key(s) on 545 the system trusted keyring to be used for certificate 546 trust validation. 547 format: { id:<keyid> | builtin } 548 549 cca= [MIPS] Override the kernel pages' cache coherency 550 algorithm. Accepted values range from 0 to 7 551 inclusive. See arch/mips/include/asm/pgtable-bits.h 552 for platform specific values (SB1, Loongson3 and 553 others). 554 555 ccw_timeout_log [S390] 556 See Documentation/s390/common_io.rst for details. 557 558 cgroup_disable= [KNL] Disable a particular controller or optional feature 559 Format: {name of the controller(s) or feature(s) to disable} 560 The effects of cgroup_disable=foo are: 561 - foo isn't auto-mounted if you mount all cgroups in 562 a single hierarchy 563 - foo isn't visible as an individually mountable 564 subsystem 565 - if foo is an optional feature then the feature is 566 disabled and corresponding cgroup files are not 567 created 568 {Currently only "memory" controller deal with this and 569 cut the overhead, others just disable the usage. So 570 only cgroup_disable=memory is actually worthy} 571 Specifying "pressure" disables per-cgroup pressure 572 stall information accounting feature 573 574 cgroup_no_v1= [KNL] Disable cgroup controllers and named hierarchies in v1 575 Format: { { controller | "all" | "named" } 576 [,{ controller | "all" | "named" }...] } 577 Like cgroup_disable, but only applies to cgroup v1; 578 the blacklisted controllers remain available in cgroup2. 579 "all" blacklists all controllers and "named" disables 580 named mounts. Specifying both "all" and "named" disables 581 all v1 hierarchies. 582 583 cgroup.memory= [KNL] Pass options to the cgroup memory controller. 584 Format: <string> 585 nosocket -- Disable socket memory accounting. 586 nokmem -- Disable kernel memory accounting. 587 nobpf -- Disable BPF memory accounting. 588 589 checkreqprot= [SELINUX] Set initial checkreqprot flag value. 590 Format: { "0" | "1" } 591 See security/selinux/Kconfig help text. 592 0 -- check protection applied by kernel (includes 593 any implied execute protection). 594 1 -- check protection requested by application. 595 Default value is set via a kernel config option. 596 Value can be changed at runtime via 597 /sys/fs/selinux/checkreqprot. 598 Setting checkreqprot to 1 is deprecated. 599 600 cio_ignore= [S390] 601 See Documentation/s390/common_io.rst for details. 602 603 clearcpuid=X[,X...] [X86] 604 Disable CPUID feature X for the kernel. See 605 arch/x86/include/asm/cpufeatures.h for the valid bit 606 numbers X. Note the Linux-specific bits are not necessarily 607 stable over kernel options, but the vendor-specific 608 ones should be. 609 X can also be a string as appearing in the flags: line 610 in /proc/cpuinfo which does not have the above 611 instability issue. However, not all features have names 612 in /proc/cpuinfo. 613 Note that using this option will taint your kernel. 614 Also note that user programs calling CPUID directly 615 or using the feature without checking anything 616 will still see it. This just prevents it from 617 being used by the kernel or shown in /proc/cpuinfo. 618 Also note the kernel might malfunction if you disable 619 some critical bits. 620 621 clk_ignore_unused 622 [CLK] 623 Prevents the clock framework from automatically gating 624 clocks that have not been explicitly enabled by a Linux 625 device driver but are enabled in hardware at reset or 626 by the bootloader/firmware. Note that this does not 627 force such clocks to be always-on nor does it reserve 628 those clocks in any way. This parameter is useful for 629 debug and development, but should not be needed on a 630 platform with proper driver support. For more 631 information, see Documentation/driver-api/clk.rst. 632 633 clock= [BUGS=X86-32, HW] gettimeofday clocksource override. 634 [Deprecated] 635 Forces specified clocksource (if available) to be used 636 when calculating gettimeofday(). If specified 637 clocksource is not available, it defaults to PIT. 638 Format: { pit | tsc | cyclone | pmtmr } 639 640 clocksource= Override the default clocksource 641 Format: <string> 642 Override the default clocksource and use the clocksource 643 with the name specified. 644 Some clocksource names to choose from, depending on 645 the platform: 646 [all] jiffies (this is the base, fallback clocksource) 647 [ACPI] acpi_pm 648 [ARM] imx_timer1,OSTS,netx_timer,mpu_timer2, 649 pxa_timer,timer3,32k_counter,timer0_1 650 [X86-32] pit,hpet,tsc; 651 scx200_hrt on Geode; cyclone on IBM x440 652 [MIPS] MIPS 653 [PARISC] cr16 654 [S390] tod 655 [SH] SuperH 656 [SPARC64] tick 657 [X86-64] hpet,tsc 658 659 clocksource.arm_arch_timer.evtstrm= 660 [ARM,ARM64] 661 Format: <bool> 662 Enable/disable the eventstream feature of the ARM 663 architected timer so that code using WFE-based polling 664 loops can be debugged more effectively on production 665 systems. 666 667 clocksource.max_cswd_read_retries= [KNL] 668 Number of clocksource_watchdog() retries due to 669 external delays before the clock will be marked 670 unstable. Defaults to two retries, that is, 671 three attempts to read the clock under test. 672 673 clocksource.verify_n_cpus= [KNL] 674 Limit the number of CPUs checked for clocksources 675 marked with CLOCK_SOURCE_VERIFY_PERCPU that 676 are marked unstable due to excessive skew. 677 A negative value says to check all CPUs, while 678 zero says not to check any. Values larger than 679 nr_cpu_ids are silently truncated to nr_cpu_ids. 680 The actual CPUs are chosen randomly, with 681 no replacement if the same CPU is chosen twice. 682 683 clocksource-wdtest.holdoff= [KNL] 684 Set the time in seconds that the clocksource 685 watchdog test waits before commencing its tests. 686 Defaults to zero when built as a module and to 687 10 seconds when built into the kernel. 688 689 cma=nn[MG]@[start[MG][-end[MG]]] 690 [KNL,CMA] 691 Sets the size of kernel global memory area for 692 contiguous memory allocations and optionally the 693 placement constraint by the physical address range of 694 memory allocations. A value of 0 disables CMA 695 altogether. For more information, see 696 kernel/dma/contiguous.c 697 698 cma_pernuma=nn[MG] 699 [ARM64,KNL,CMA] 700 Sets the size of kernel per-numa memory area for 701 contiguous memory allocations. A value of 0 disables 702 per-numa CMA altogether. And If this option is not 703 specified, the default value is 0. 704 With per-numa CMA enabled, DMA users on node nid will 705 first try to allocate buffer from the pernuma area 706 which is located in node nid, if the allocation fails, 707 they will fallback to the global default memory area. 708 709 cmo_free_hint= [PPC] Format: { yes | no } 710 Specify whether pages are marked as being inactive 711 when they are freed. This is used in CMO environments 712 to determine OS memory pressure for page stealing by 713 a hypervisor. 714 Default: yes 715 716 coherent_pool=nn[KMG] [ARM,KNL] 717 Sets the size of memory pool for coherent, atomic dma 718 allocations, by default set to 256K. 719 720 com20020= [HW,NET] ARCnet - COM20020 chipset 721 Format: 722 <io>[,<irq>[,<nodeID>[,<backplane>[,<ckp>[,<timeout>]]]]] 723 724 com90io= [HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers) 725 Format: <io>[,<irq>] 726 727 com90xx= [HW,NET] 728 ARCnet - COM90xx chipset (memory-mapped buffers) 729 Format: <io>[,<irq>[,<memstart>]] 730 731 condev= [HW,S390] console device 732 conmode= 733 734 con3215_drop= [S390] 3215 console drop mode. 735 Format: y|n|Y|N|1|0 736 When set to true, drop data on the 3215 console when 737 the console buffer is full. In this case the 738 operator using a 3270 terminal emulator (for example 739 x3270) does not have to enter the clear key for the 740 console output to advance and the kernel to continue. 741 This leads to a much faster boot time when a 3270 742 terminal emulator is active. If no 3270 terminal 743 emulator is used, this parameter has no effect. 744 745 console= [KNL] Output console device and options. 746 747 tty<n> Use the virtual console device <n>. 748 749 ttyS<n>[,options] 750 ttyUSB0[,options] 751 Use the specified serial port. The options are of 752 the form "bbbbpnf", where "bbbb" is the baud rate, 753 "p" is parity ("n", "o", or "e"), "n" is number of 754 bits, and "f" is flow control ("r" for RTS or 755 omit it). Default is "9600n8". 756 757 See Documentation/admin-guide/serial-console.rst for more 758 information. See 759 Documentation/networking/netconsole.rst for an 760 alternative. 761 762 uart[8250],io,<addr>[,options] 763 uart[8250],mmio,<addr>[,options] 764 uart[8250],mmio16,<addr>[,options] 765 uart[8250],mmio32,<addr>[,options] 766 uart[8250],0x<addr>[,options] 767 Start an early, polled-mode console on the 8250/16550 768 UART at the specified I/O port or MMIO address, 769 switching to the matching ttyS device later. 770 MMIO inter-register address stride is either 8-bit 771 (mmio), 16-bit (mmio16), or 32-bit (mmio32). 772 If none of [io|mmio|mmio16|mmio32], <addr> is assumed 773 to be equivalent to 'mmio'. 'options' are specified in 774 the same format described for ttyS above; if unspecified, 775 the h/w is not re-initialized. 776 777 hvc<n> Use the hypervisor console device <n>. This is for 778 both Xen and PowerPC hypervisors. 779 780 { null | "" } 781 Use to disable console output, i.e., to have kernel 782 console messages discarded. 783 This must be the only console= parameter used on the 784 kernel command line. 785 786 If the device connected to the port is not a TTY but a braille 787 device, prepend "brl," before the device type, for instance 788 console=brl,ttyS0 789 For now, only VisioBraille is supported. 790 791 console_msg_format= 792 [KNL] Change console messages format 793 default 794 By default we print messages on consoles in 795 "[time stamp] text\n" format (time stamp may not be 796 printed, depending on CONFIG_PRINTK_TIME or 797 `printk_time' param). 798 syslog 799 Switch to syslog format: "<%u>[time stamp] text\n" 800 IOW, each message will have a facility and loglevel 801 prefix. The format is similar to one used by syslog() 802 syscall, or to executing "dmesg -S --raw" or to reading 803 from /proc/kmsg. 804 805 consoleblank= [KNL] The console blank (screen saver) timeout in 806 seconds. A value of 0 disables the blank timer. 807 Defaults to 0. 808 809 coredump_filter= 810 [KNL] Change the default value for 811 /proc/<pid>/coredump_filter. 812 See also Documentation/filesystems/proc.rst. 813 814 coresight_cpu_debug.enable 815 [ARM,ARM64] 816 Format: <bool> 817 Enable/disable the CPU sampling based debugging. 818 0: default value, disable debugging 819 1: enable debugging at boot time 820 821 cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver 822 Format: 823 <first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>] 824 825 cpuidle.off=1 [CPU_IDLE] 826 disable the cpuidle sub-system 827 828 cpuidle.governor= 829 [CPU_IDLE] Name of the cpuidle governor to use. 830 831 cpufreq.off=1 [CPU_FREQ] 832 disable the cpufreq sub-system 833 834 cpufreq.default_governor= 835 [CPU_FREQ] Name of the default cpufreq governor or 836 policy to use. This governor must be registered in the 837 kernel before the cpufreq driver probes. 838 839 cpu_init_udelay=N 840 [X86] Delay for N microsec between assert and de-assert 841 of APIC INIT to start processors. This delay occurs 842 on every CPU online, such as boot, and resume from suspend. 843 Default: 10000 844 845 cpuhp.parallel= 846 [SMP] Enable/disable parallel bringup of secondary CPUs 847 Format: <bool> 848 Default is enabled if CONFIG_HOTPLUG_PARALLEL=y. Otherwise 849 the parameter has no effect. 850 851 crash_kexec_post_notifiers 852 Run kdump after running panic-notifiers and dumping 853 kmsg. This only for the users who doubt kdump always 854 succeeds in any situation. 855 Note that this also increases risks of kdump failure, 856 because some panic notifiers can make the crashed 857 kernel more unstable. 858 859 crashkernel=size[KMG][@offset[KMG]] 860 [KNL] Using kexec, Linux can switch to a 'crash kernel' 861 upon panic. This parameter reserves the physical 862 memory region [offset, offset + size] for that kernel 863 image. If '@offset' is omitted, then a suitable offset 864 is selected automatically. 865 [KNL, X86-64, ARM64] Select a region under 4G first, and 866 fall back to reserve region above 4G when '@offset' 867 hasn't been specified. 868 See Documentation/admin-guide/kdump/kdump.rst for further details. 869 870 crashkernel=range1:size1[,range2:size2,...][@offset] 871 [KNL] Same as above, but depends on the memory 872 in the running system. The syntax of range is 873 start-[end] where start and end are both 874 a memory unit (amount[KMG]). See also 875 Documentation/admin-guide/kdump/kdump.rst for an example. 876 877 crashkernel=size[KMG],high 878 [KNL, X86-64, ARM64] range could be above 4G. Allow kernel 879 to allocate physical memory region from top, so could 880 be above 4G if system have more than 4G ram installed. 881 Otherwise memory region will be allocated below 4G, if 882 available. 883 It will be ignored if crashkernel=X is specified. 884 crashkernel=size[KMG],low 885 [KNL, X86-64, ARM64] range under 4G. When crashkernel=X,high 886 is passed, kernel could allocate physical memory region 887 above 4G, that cause second kernel crash on system 888 that require some amount of low memory, e.g. swiotlb 889 requires at least 64M+32K low memory, also enough extra 890 low memory is needed to make sure DMA buffers for 32-bit 891 devices won't run out. Kernel would try to allocate 892 default size of memory below 4G automatically. The default 893 size is platform dependent. 894 --> x86: max(swiotlb_size_or_default() + 8MiB, 256MiB) 895 --> arm64: 128MiB 896 This one lets the user specify own low range under 4G 897 for second kernel instead. 898 0: to disable low allocation. 899 It will be ignored when crashkernel=X,high is not used 900 or memory reserved is below 4G. 901 902 cryptomgr.notests 903 [KNL] Disable crypto self-tests 904 905 cs89x0_dma= [HW,NET] 906 Format: <dma> 907 908 cs89x0_media= [HW,NET] 909 Format: { rj45 | aui | bnc } 910 911 csdlock_debug= [KNL] Enable or disable debug add-ons of cross-CPU 912 function call handling. When switched on, 913 additional debug data is printed to the console 914 in case a hanging CPU is detected, and that 915 CPU is pinged again in order to try to resolve 916 the hang situation. The default value of this 917 option depends on the CSD_LOCK_WAIT_DEBUG_DEFAULT 918 Kconfig option. 919 920 dasd= [HW,NET] 921 See header of drivers/s390/block/dasd_devmap.c. 922 923 db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port 924 (one device per port) 925 Format: <port#>,<type> 926 See also Documentation/input/devices/joystick-parport.rst 927 928 debug [KNL] Enable kernel debugging (events log level). 929 930 debug_boot_weak_hash 931 [KNL] Enable printing [hashed] pointers early in the 932 boot sequence. If enabled, we use a weak hash instead 933 of siphash to hash pointers. Use this option if you are 934 seeing instances of '(___ptrval___)') and need to see a 935 value (hashed pointer) instead. Cryptographically 936 insecure, please do not use on production kernels. 937 938 debug_locks_verbose= 939 [KNL] verbose locking self-tests 940 Format: <int> 941 Print debugging info while doing the locking API 942 self-tests. 943 Bitmask for the various LOCKTYPE_ tests. Defaults to 0 944 (no extra messages), setting it to -1 (all bits set) 945 will print _a_lot_ more information - normally only 946 useful to lockdep developers. 947 948 debug_objects [KNL] Enable object debugging 949 950 debug_guardpage_minorder= 951 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this 952 parameter allows control of the order of pages that will 953 be intentionally kept free (and hence protected) by the 954 buddy allocator. Bigger value increase the probability 955 of catching random memory corruption, but reduce the 956 amount of memory for normal system use. The maximum 957 possible value is MAX_ORDER/2. Setting this parameter 958 to 1 or 2 should be enough to identify most random 959 memory corruption problems caused by bugs in kernel or 960 driver code when a CPU writes to (or reads from) a 961 random memory location. Note that there exists a class 962 of memory corruptions problems caused by buggy H/W or 963 F/W or by drivers badly programming DMA (basically when 964 memory is written at bus level and the CPU MMU is 965 bypassed) which are not detectable by 966 CONFIG_DEBUG_PAGEALLOC, hence this option will not help 967 tracking down these problems. 968 969 debug_pagealloc= 970 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this parameter 971 enables the feature at boot time. By default, it is 972 disabled and the system will work mostly the same as a 973 kernel built without CONFIG_DEBUG_PAGEALLOC. 974 Note: to get most of debug_pagealloc error reports, it's 975 useful to also enable the page_owner functionality. 976 on: enable the feature 977 978 debugfs= [KNL] This parameter enables what is exposed to userspace 979 and debugfs internal clients. 980 Format: { on, no-mount, off } 981 on: All functions are enabled. 982 no-mount: 983 Filesystem is not registered but kernel clients can 984 access APIs and a crashkernel can be used to read 985 its content. There is nothing to mount. 986 off: Filesystem is not registered and clients 987 get a -EPERM as result when trying to register files 988 or directories within debugfs. 989 This is equivalent of the runtime functionality if 990 debugfs was not enabled in the kernel at all. 991 Default value is set in build-time with a kernel configuration. 992 993 debugpat [X86] Enable PAT debugging 994 995 default_hugepagesz= 996 [HW] The size of the default HugeTLB page. This is 997 the size represented by the legacy /proc/ hugepages 998 APIs. In addition, this is the default hugetlb size 999 used for shmget(), mmap() and mounting hugetlbfs 1000 filesystems. If not specified, defaults to the 1001 architecture's default huge page size. Huge page 1002 sizes are architecture dependent. See also 1003 Documentation/admin-guide/mm/hugetlbpage.rst. 1004 Format: size[KMG] 1005 1006 deferred_probe_timeout= 1007 [KNL] Debugging option to set a timeout in seconds for 1008 deferred probe to give up waiting on dependencies to 1009 probe. Only specific dependencies (subsystems or 1010 drivers) that have opted in will be ignored. A timeout 1011 of 0 will timeout at the end of initcalls. If the time 1012 out hasn't expired, it'll be restarted by each 1013 successful driver registration. This option will also 1014 dump out devices still on the deferred probe list after 1015 retrying. 1016 1017 delayacct [KNL] Enable per-task delay accounting 1018 1019 dell_smm_hwmon.ignore_dmi= 1020 [HW] Continue probing hardware even if DMI data 1021 indicates that the driver is running on unsupported 1022 hardware. 1023 1024 dell_smm_hwmon.force= 1025 [HW] Activate driver even if SMM BIOS signature does 1026 not match list of supported models and enable otherwise 1027 blacklisted features. 1028 1029 dell_smm_hwmon.power_status= 1030 [HW] Report power status in /proc/i8k 1031 (disabled by default). 1032 1033 dell_smm_hwmon.restricted= 1034 [HW] Allow controlling fans only if SYS_ADMIN 1035 capability is set. 1036 1037 dell_smm_hwmon.fan_mult= 1038 [HW] Factor to multiply fan speed with. 1039 1040 dell_smm_hwmon.fan_max= 1041 [HW] Maximum configurable fan speed. 1042 1043 dfltcc= [HW,S390] 1044 Format: { on | off | def_only | inf_only | always } 1045 on: s390 zlib hardware support for compression on 1046 level 1 and decompression (default) 1047 off: No s390 zlib hardware support 1048 def_only: s390 zlib hardware support for deflate 1049 only (compression on level 1) 1050 inf_only: s390 zlib hardware support for inflate 1051 only (decompression) 1052 always: Same as 'on' but ignores the selected compression 1053 level always using hardware support (used for debugging) 1054 1055 dhash_entries= [KNL] 1056 Set number of hash buckets for dentry cache. 1057 1058 disable_1tb_segments [PPC] 1059 Disables the use of 1TB hash page table segments. This 1060 causes the kernel to fall back to 256MB segments which 1061 can be useful when debugging issues that require an SLB 1062 miss to occur. 1063 1064 disable= [IPV6] 1065 See Documentation/networking/ipv6.rst. 1066 1067 disable_radix [PPC] 1068 Disable RADIX MMU mode on POWER9 1069 1070 disable_tlbie [PPC] 1071 Disable TLBIE instruction. Currently does not work 1072 with KVM, with HASH MMU, or with coherent accelerators. 1073 1074 disable_cpu_apicid= [X86,APIC,SMP] 1075 Format: <int> 1076 The number of initial APIC ID for the 1077 corresponding CPU to be disabled at boot, 1078 mostly used for the kdump 2nd kernel to 1079 disable BSP to wake up multiple CPUs without 1080 causing system reset or hang due to sending 1081 INIT from AP to BSP. 1082 1083 disable_ddw [PPC/PSERIES] 1084 Disable Dynamic DMA Window support. Use this 1085 to workaround buggy firmware. 1086 1087 disable_ipv6= [IPV6] 1088 See Documentation/networking/ipv6.rst. 1089 1090 disable_mtrr_cleanup [X86] 1091 The kernel tries to adjust MTRR layout from continuous 1092 to discrete, to make X server driver able to add WB 1093 entry later. This parameter disables that. 1094 1095 disable_mtrr_trim [X86, Intel and AMD only] 1096 By default the kernel will trim any uncacheable 1097 memory out of your available memory pool based on 1098 MTRR settings. This parameter disables that behavior, 1099 possibly causing your machine to run very slowly. 1100 1101 disable_timer_pin_1 [X86] 1102 Disable PIN 1 of APIC timer 1103 Can be useful to work around chipset bugs. 1104 1105 dis_ucode_ldr [X86] Disable the microcode loader. 1106 1107 dma_debug=off If the kernel is compiled with DMA_API_DEBUG support, 1108 this option disables the debugging code at boot. 1109 1110 dma_debug_entries=<number> 1111 This option allows to tune the number of preallocated 1112 entries for DMA-API debugging code. One entry is 1113 required per DMA-API allocation. Use this if the 1114 DMA-API debugging code disables itself because the 1115 architectural default is too low. 1116 1117 dma_debug_driver=<driver_name> 1118 With this option the DMA-API debugging driver 1119 filter feature can be enabled at boot time. Just 1120 pass the driver to filter for as the parameter. 1121 The filter can be disabled or changed to another 1122 driver later using sysfs. 1123 1124 driver_async_probe= [KNL] 1125 List of driver names to be probed asynchronously. * 1126 matches with all driver names. If * is specified, the 1127 rest of the listed driver names are those that will NOT 1128 match the *. 1129 Format: <driver_name1>,<driver_name2>... 1130 1131 drm.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>] 1132 Broken monitors, graphic adapters, KVMs and EDIDless 1133 panels may send no or incorrect EDID data sets. 1134 This parameter allows to specify an EDID data sets 1135 in the /lib/firmware directory that are used instead. 1136 Generic built-in EDID data sets are used, if one of 1137 edid/1024x768.bin, edid/1280x1024.bin, 1138 edid/1680x1050.bin, or edid/1920x1080.bin is given 1139 and no file with the same name exists. Details and 1140 instructions how to build your own EDID data are 1141 available in Documentation/admin-guide/edid.rst. An EDID 1142 data set will only be used for a particular connector, 1143 if its name and a colon are prepended to the EDID 1144 name. Each connector may use a unique EDID data 1145 set by separating the files with a comma. An EDID 1146 data set with no connector name will be used for 1147 any connectors not explicitly specified. 1148 1149 dscc4.setup= [NET] 1150 1151 dt_cpu_ftrs= [PPC] 1152 Format: {"off" | "known"} 1153 Control how the dt_cpu_ftrs device-tree binding is 1154 used for CPU feature discovery and setup (if it 1155 exists). 1156 off: Do not use it, fall back to legacy cpu table. 1157 known: Do not pass through unknown features to guests 1158 or userspace, only those that the kernel is aware of. 1159 1160 dump_apple_properties [X86] 1161 Dump name and content of EFI device properties on 1162 x86 Macs. Useful for driver authors to determine 1163 what data is available or for reverse-engineering. 1164 1165 dyndbg[="val"] [KNL,DYNAMIC_DEBUG] 1166 <module>.dyndbg[="val"] 1167 Enable debug messages at boot time. See 1168 Documentation/admin-guide/dynamic-debug-howto.rst 1169 for details. 1170 1171 early_ioremap_debug [KNL] 1172 Enable debug messages in early_ioremap support. This 1173 is useful for tracking down temporary early mappings 1174 which are not unmapped. 1175 1176 earlycon= [KNL] Output early console device and options. 1177 1178 When used with no options, the early console is 1179 determined by stdout-path property in device tree's 1180 chosen node or the ACPI SPCR table if supported by 1181 the platform. 1182 1183 cdns,<addr>[,options] 1184 Start an early, polled-mode console on a Cadence 1185 (xuartps) serial port at the specified address. Only 1186 supported option is baud rate. If baud rate is not 1187 specified, the serial port must already be setup and 1188 configured. 1189 1190 uart[8250],io,<addr>[,options[,uartclk]] 1191 uart[8250],mmio,<addr>[,options[,uartclk]] 1192 uart[8250],mmio32,<addr>[,options[,uartclk]] 1193 uart[8250],mmio32be,<addr>[,options[,uartclk]] 1194 uart[8250],0x<addr>[,options] 1195 Start an early, polled-mode console on the 8250/16550 1196 UART at the specified I/O port or MMIO address. 1197 MMIO inter-register address stride is either 8-bit 1198 (mmio) or 32-bit (mmio32 or mmio32be). 1199 If none of [io|mmio|mmio32|mmio32be], <addr> is assumed 1200 to be equivalent to 'mmio'. 'options' are specified 1201 in the same format described for "console=ttyS<n>"; if 1202 unspecified, the h/w is not initialized. 'uartclk' is 1203 the uart clock frequency; if unspecified, it is set 1204 to 'BASE_BAUD' * 16. 1205 1206 pl011,<addr> 1207 pl011,mmio32,<addr> 1208 Start an early, polled-mode console on a pl011 serial 1209 port at the specified address. The pl011 serial port 1210 must already be setup and configured. Options are not 1211 yet supported. If 'mmio32' is specified, then only 1212 the driver will use only 32-bit accessors to read/write 1213 the device registers. 1214 1215 liteuart,<addr> 1216 Start an early console on a litex serial port at the 1217 specified address. The serial port must already be 1218 setup and configured. Options are not yet supported. 1219 1220 meson,<addr> 1221 Start an early, polled-mode console on a meson serial 1222 port at the specified address. The serial port must 1223 already be setup and configured. Options are not yet 1224 supported. 1225 1226 msm_serial,<addr> 1227 Start an early, polled-mode console on an msm serial 1228 port at the specified address. The serial port 1229 must already be setup and configured. Options are not 1230 yet supported. 1231 1232 msm_serial_dm,<addr> 1233 Start an early, polled-mode console on an msm serial 1234 dm port at the specified address. The serial port 1235 must already be setup and configured. Options are not 1236 yet supported. 1237 1238 owl,<addr> 1239 Start an early, polled-mode console on a serial port 1240 of an Actions Semi SoC, such as S500 or S900, at the 1241 specified address. The serial port must already be 1242 setup and configured. Options are not yet supported. 1243 1244 rda,<addr> 1245 Start an early, polled-mode console on a serial port 1246 of an RDA Micro SoC, such as RDA8810PL, at the 1247 specified address. The serial port must already be 1248 setup and configured. Options are not yet supported. 1249 1250 sbi 1251 Use RISC-V SBI (Supervisor Binary Interface) for early 1252 console. 1253 1254 smh Use ARM semihosting calls for early console. 1255 1256 s3c2410,<addr> 1257 s3c2412,<addr> 1258 s3c2440,<addr> 1259 s3c6400,<addr> 1260 s5pv210,<addr> 1261 exynos4210,<addr> 1262 Use early console provided by serial driver available 1263 on Samsung SoCs, requires selecting proper type and 1264 a correct base address of the selected UART port. The 1265 serial port must already be setup and configured. 1266 Options are not yet supported. 1267 1268 lantiq,<addr> 1269 Start an early, polled-mode console on a lantiq serial 1270 (lqasc) port at the specified address. The serial port 1271 must already be setup and configured. Options are not 1272 yet supported. 1273 1274 lpuart,<addr> 1275 lpuart32,<addr> 1276 Use early console provided by Freescale LP UART driver 1277 found on Freescale Vybrid and QorIQ LS1021A processors. 1278 A valid base address must be provided, and the serial 1279 port must already be setup and configured. 1280 1281 ec_imx21,<addr> 1282 ec_imx6q,<addr> 1283 Start an early, polled-mode, output-only console on the 1284 Freescale i.MX UART at the specified address. The UART 1285 must already be setup and configured. 1286 1287 ar3700_uart,<addr> 1288 Start an early, polled-mode console on the 1289 Armada 3700 serial port at the specified 1290 address. The serial port must already be setup 1291 and configured. Options are not yet supported. 1292 1293 qcom_geni,<addr> 1294 Start an early, polled-mode console on a Qualcomm 1295 Generic Interface (GENI) based serial port at the 1296 specified address. The serial port must already be 1297 setup and configured. Options are not yet supported. 1298 1299 efifb,[options] 1300 Start an early, unaccelerated console on the EFI 1301 memory mapped framebuffer (if available). On cache 1302 coherent non-x86 systems that use system memory for 1303 the framebuffer, pass the 'ram' option so that it is 1304 mapped with the correct attributes. 1305 1306 linflex,<addr> 1307 Use early console provided by Freescale LINFlexD UART 1308 serial driver for NXP S32V234 SoCs. A valid base 1309 address must be provided, and the serial port must 1310 already be setup and configured. 1311 1312 earlyprintk= [X86,SH,ARM,M68k,S390] 1313 earlyprintk=vga 1314 earlyprintk=sclp 1315 earlyprintk=xen 1316 earlyprintk=serial[,ttySn[,baudrate]] 1317 earlyprintk=serial[,0x...[,baudrate]] 1318 earlyprintk=ttySn[,baudrate] 1319 earlyprintk=dbgp[debugController#] 1320 earlyprintk=pciserial[,force],bus:device.function[,baudrate] 1321 earlyprintk=xdbc[xhciController#] 1322 1323 earlyprintk is useful when the kernel crashes before 1324 the normal console is initialized. It is not enabled by 1325 default because it has some cosmetic problems. 1326 1327 Append ",keep" to not disable it when the real console 1328 takes over. 1329 1330 Only one of vga, serial, or usb debug port can 1331 be used at a time. 1332 1333 Currently only ttyS0 and ttyS1 may be specified by 1334 name. Other I/O ports may be explicitly specified 1335 on some architectures (x86 and arm at least) by 1336 replacing ttySn with an I/O port address, like this: 1337 earlyprintk=serial,0x1008,115200 1338 You can find the port for a given device in 1339 /proc/tty/driver/serial: 1340 2: uart:ST16650V2 port:00001008 irq:18 ... 1341 1342 Interaction with the standard serial driver is not 1343 very good. 1344 1345 The VGA output is eventually overwritten by 1346 the real console. 1347 1348 The xen option can only be used in Xen domains. 1349 1350 The sclp output can only be used on s390. 1351 1352 The optional "force" to "pciserial" enables use of a 1353 PCI device even when its classcode is not of the 1354 UART class. 1355 1356 edac_report= [HW,EDAC] Control how to report EDAC event 1357 Format: {"on" | "off" | "force"} 1358 on: enable EDAC to report H/W event. May be overridden 1359 by other higher priority error reporting module. 1360 off: disable H/W event reporting through EDAC. 1361 force: enforce the use of EDAC to report H/W event. 1362 default: on. 1363 1364 edd= [EDD] 1365 Format: {"off" | "on" | "skip[mbr]"} 1366 1367 efi= [EFI] 1368 Format: { "debug", "disable_early_pci_dma", 1369 "nochunk", "noruntime", "nosoftreserve", 1370 "novamap", "no_disable_early_pci_dma" } 1371 debug: enable misc debug output. 1372 disable_early_pci_dma: disable the busmaster bit on all 1373 PCI bridges while in the EFI boot stub. 1374 nochunk: disable reading files in "chunks" in the EFI 1375 boot stub, as chunking can cause problems with some 1376 firmware implementations. 1377 noruntime : disable EFI runtime services support 1378 nosoftreserve: The EFI_MEMORY_SP (Specific Purpose) 1379 attribute may cause the kernel to reserve the 1380 memory range for a memory mapping driver to 1381 claim. Specify efi=nosoftreserve to disable this 1382 reservation and treat the memory by its base type 1383 (i.e. EFI_CONVENTIONAL_MEMORY / "System RAM"). 1384 novamap: do not call SetVirtualAddressMap(). 1385 no_disable_early_pci_dma: Leave the busmaster bit set 1386 on all PCI bridges while in the EFI boot stub 1387 1388 efi_no_storage_paranoia [EFI; X86] 1389 Using this parameter you can use more than 50% of 1390 your efi variable storage. Use this parameter only if 1391 you are really sure that your UEFI does sane gc and 1392 fulfills the spec otherwise your board may brick. 1393 1394 efi_fake_mem= nn[KMG]@ss[KMG]:aa[,nn[KMG]@ss[KMG]:aa,..] [EFI; X86] 1395 Add arbitrary attribute to specific memory range by 1396 updating original EFI memory map. 1397 Region of memory which aa attribute is added to is 1398 from ss to ss+nn. 1399 1400 If efi_fake_mem=2G@4G:0x10000,2G@0x10a0000000:0x10000 1401 is specified, EFI_MEMORY_MORE_RELIABLE(0x10000) 1402 attribute is added to range 0x100000000-0x180000000 and 1403 0x10a0000000-0x1120000000. 1404 1405 If efi_fake_mem=8G@9G:0x40000 is specified, the 1406 EFI_MEMORY_SP(0x40000) attribute is added to 1407 range 0x240000000-0x43fffffff. 1408 1409 Using this parameter you can do debugging of EFI memmap 1410 related features. For example, you can do debugging of 1411 Address Range Mirroring feature even if your box 1412 doesn't support it, or mark specific memory as 1413 "soft reserved". 1414 1415 efivar_ssdt= [EFI; X86] Name of an EFI variable that contains an SSDT 1416 that is to be dynamically loaded by Linux. If there are 1417 multiple variables with the same name but with different 1418 vendor GUIDs, all of them will be loaded. See 1419 Documentation/admin-guide/acpi/ssdt-overlays.rst for details. 1420 1421 1422 eisa_irq_edge= [PARISC,HW] 1423 See header of drivers/parisc/eisa.c. 1424 1425 ekgdboc= [X86,KGDB] Allow early kernel console debugging 1426 Format: ekgdboc=kbd 1427 1428 This is designed to be used in conjunction with 1429 the boot argument: earlyprintk=vga 1430 1431 This parameter works in place of the kgdboc parameter 1432 but can only be used if the backing tty is available 1433 very early in the boot process. For early debugging 1434 via a serial port see kgdboc_earlycon instead. 1435 1436 elanfreq= [X86-32] 1437 See comment before function elanfreq_setup() in 1438 arch/x86/kernel/cpu/cpufreq/elanfreq.c. 1439 1440 elfcorehdr=[size[KMG]@]offset[KMG] [IA64,PPC,SH,X86,S390] 1441 Specifies physical address of start of kernel core 1442 image elf header and optionally the size. Generally 1443 kexec loader will pass this option to capture kernel. 1444 See Documentation/admin-guide/kdump/kdump.rst for details. 1445 1446 enable_mtrr_cleanup [X86] 1447 The kernel tries to adjust MTRR layout from continuous 1448 to discrete, to make X server driver able to add WB 1449 entry later. This parameter enables that. 1450 1451 enable_timer_pin_1 [X86] 1452 Enable PIN 1 of APIC timer 1453 Can be useful to work around chipset bugs 1454 (in particular on some ATI chipsets). 1455 The kernel tries to set a reasonable default. 1456 1457 enforcing= [SELINUX] Set initial enforcing status. 1458 Format: {"0" | "1"} 1459 See security/selinux/Kconfig help text. 1460 0 -- permissive (log only, no denials). 1461 1 -- enforcing (deny and log). 1462 Default value is 0. 1463 Value can be changed at runtime via 1464 /sys/fs/selinux/enforce. 1465 1466 erst_disable [ACPI] 1467 Disable Error Record Serialization Table (ERST) 1468 support. 1469 1470 ether= [HW,NET] Ethernet cards parameters 1471 This option is obsoleted by the "netdev=" option, which 1472 has equivalent usage. See its documentation for details. 1473 1474 evm= [EVM] 1475 Format: { "fix" } 1476 Permit 'security.evm' to be updated regardless of 1477 current integrity status. 1478 1479 early_page_ext [KNL] Enforces page_ext initialization to earlier 1480 stages so cover more early boot allocations. 1481 Please note that as side effect some optimizations 1482 might be disabled to achieve that (e.g. parallelized 1483 memory initialization is disabled) so the boot process 1484 might take longer, especially on systems with a lot of 1485 memory. Available with CONFIG_PAGE_EXTENSION=y. 1486 1487 failslab= 1488 fail_usercopy= 1489 fail_page_alloc= 1490 fail_make_request=[KNL] 1491 General fault injection mechanism. 1492 Format: <interval>,<probability>,<space>,<times> 1493 See also Documentation/fault-injection/. 1494 1495 fb_tunnels= [NET] 1496 Format: { initns | none } 1497 See Documentation/admin-guide/sysctl/net.rst for 1498 fb_tunnels_only_for_init_ns 1499 1500 floppy= [HW] 1501 See Documentation/admin-guide/blockdev/floppy.rst. 1502 1503 force_pal_cache_flush 1504 [IA-64] Avoid check_sal_cache_flush which may hang on 1505 buggy SAL_CACHE_FLUSH implementations. Using this 1506 parameter will force ia64_sal_cache_flush to call 1507 ia64_pal_cache_flush instead of SAL_CACHE_FLUSH. 1508 1509 forcepae [X86-32] 1510 Forcefully enable Physical Address Extension (PAE). 1511 Many Pentium M systems disable PAE but may have a 1512 functionally usable PAE implementation. 1513 Warning: use of this parameter will taint the kernel 1514 and may cause unknown problems. 1515 1516 ftrace=[tracer] 1517 [FTRACE] will set and start the specified tracer 1518 as early as possible in order to facilitate early 1519 boot debugging. 1520 1521 ftrace_boot_snapshot 1522 [FTRACE] On boot up, a snapshot will be taken of the 1523 ftrace ring buffer that can be read at: 1524 /sys/kernel/tracing/snapshot. 1525 This is useful if you need tracing information from kernel 1526 boot up that is likely to be overridden by user space 1527 start up functionality. 1528 1529 Optionally, the snapshot can also be defined for a tracing 1530 instance that was created by the trace_instance= command 1531 line parameter. 1532 1533 trace_instance=foo,sched_switch ftrace_boot_snapshot=foo 1534 1535 The above will cause the "foo" tracing instance to trigger 1536 a snapshot at the end of boot up. 1537 1538 ftrace_dump_on_oops[=orig_cpu] 1539 [FTRACE] will dump the trace buffers on oops. 1540 If no parameter is passed, ftrace will dump 1541 buffers of all CPUs, but if you pass orig_cpu, it will 1542 dump only the buffer of the CPU that triggered the 1543 oops. 1544 1545 ftrace_filter=[function-list] 1546 [FTRACE] Limit the functions traced by the function 1547 tracer at boot up. function-list is a comma-separated 1548 list of functions. This list can be changed at run 1549 time by the set_ftrace_filter file in the debugfs 1550 tracing directory. 1551 1552 ftrace_notrace=[function-list] 1553 [FTRACE] Do not trace the functions specified in 1554 function-list. This list can be changed at run time 1555 by the set_ftrace_notrace file in the debugfs 1556 tracing directory. 1557 1558 ftrace_graph_filter=[function-list] 1559 [FTRACE] Limit the top level callers functions traced 1560 by the function graph tracer at boot up. 1561 function-list is a comma-separated list of functions 1562 that can be changed at run time by the 1563 set_graph_function file in the debugfs tracing directory. 1564 1565 ftrace_graph_notrace=[function-list] 1566 [FTRACE] Do not trace from the functions specified in 1567 function-list. This list is a comma-separated list of 1568 functions that can be changed at run time by the 1569 set_graph_notrace file in the debugfs tracing directory. 1570 1571 ftrace_graph_max_depth=<uint> 1572 [FTRACE] Used with the function graph tracer. This is 1573 the max depth it will trace into a function. This value 1574 can be changed at run time by the max_graph_depth file 1575 in the tracefs tracing directory. default: 0 (no limit) 1576 1577 fw_devlink= [KNL] Create device links between consumer and supplier 1578 devices by scanning the firmware to infer the 1579 consumer/supplier relationships. This feature is 1580 especially useful when drivers are loaded as modules as 1581 it ensures proper ordering of tasks like device probing 1582 (suppliers first, then consumers), supplier boot state 1583 clean up (only after all consumers have probed), 1584 suspend/resume & runtime PM (consumers first, then 1585 suppliers). 1586 Format: { off | permissive | on | rpm } 1587 off -- Don't create device links from firmware info. 1588 permissive -- Create device links from firmware info 1589 but use it only for ordering boot state clean 1590 up (sync_state() calls). 1591 on -- Create device links from firmware info and use it 1592 to enforce probe and suspend/resume ordering. 1593 rpm -- Like "on", but also use to order runtime PM. 1594 1595 fw_devlink.strict=<bool> 1596 [KNL] Treat all inferred dependencies as mandatory 1597 dependencies. This only applies for fw_devlink=on|rpm. 1598 Format: <bool> 1599 1600 fw_devlink.sync_state = 1601 [KNL] When all devices that could probe have finished 1602 probing, this parameter controls what to do with 1603 devices that haven't yet received their sync_state() 1604 calls. 1605 Format: { strict | timeout } 1606 strict -- Default. Continue waiting on consumers to 1607 probe successfully. 1608 timeout -- Give up waiting on consumers and call 1609 sync_state() on any devices that haven't yet 1610 received their sync_state() calls after 1611 deferred_probe_timeout has expired or by 1612 late_initcall() if !CONFIG_MODULES. 1613 1614 gamecon.map[2|3]= 1615 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad 1616 support via parallel port (up to 5 devices per port) 1617 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5> 1618 See also Documentation/input/devices/joystick-parport.rst 1619 1620 gamma= [HW,DRM] 1621 1622 gart_fix_e820= [X86-64] disable the fix e820 for K8 GART 1623 Format: off | on 1624 default: on 1625 1626 gather_data_sampling= 1627 [X86,INTEL] Control the Gather Data Sampling (GDS) 1628 mitigation. 1629 1630 Gather Data Sampling is a hardware vulnerability which 1631 allows unprivileged speculative access to data which was 1632 previously stored in vector registers. 1633 1634 This issue is mitigated by default in updated microcode. 1635 The mitigation may have a performance impact but can be 1636 disabled. On systems without the microcode mitigation 1637 disabling AVX serves as a mitigation. 1638 1639 force: Disable AVX to mitigate systems without 1640 microcode mitigation. No effect if the microcode 1641 mitigation is present. Known to cause crashes in 1642 userspace with buggy AVX enumeration. 1643 1644 off: Disable GDS mitigation. 1645 1646 gcov_persist= [GCOV] When non-zero (default), profiling data for 1647 kernel modules is saved and remains accessible via 1648 debugfs, even when the module is unloaded/reloaded. 1649 When zero, profiling data is discarded and associated 1650 debugfs files are removed at module unload time. 1651 1652 goldfish [X86] Enable the goldfish android emulator platform. 1653 Don't use this when you are not running on the 1654 android emulator 1655 1656 gpio-mockup.gpio_mockup_ranges 1657 [HW] Sets the ranges of gpiochip of for this device. 1658 Format: <start1>,<end1>,<start2>,<end2>... 1659 gpio-mockup.gpio_mockup_named_lines 1660 [HW] Let the driver know GPIO lines should be named. 1661 1662 gpt [EFI] Forces disk with valid GPT signature but 1663 invalid Protective MBR to be treated as GPT. If the 1664 primary GPT is corrupted, it enables the backup/alternate 1665 GPT to be used instead. 1666 1667 grcan.enable0= [HW] Configuration of physical interface 0. Determines 1668 the "Enable 0" bit of the configuration register. 1669 Format: 0 | 1 1670 Default: 0 1671 grcan.enable1= [HW] Configuration of physical interface 1. Determines 1672 the "Enable 0" bit of the configuration register. 1673 Format: 0 | 1 1674 Default: 0 1675 grcan.select= [HW] Select which physical interface to use. 1676 Format: 0 | 1 1677 Default: 0 1678 grcan.txsize= [HW] Sets the size of the tx buffer. 1679 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0. 1680 Default: 1024 1681 grcan.rxsize= [HW] Sets the size of the rx buffer. 1682 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0. 1683 Default: 1024 1684 1685 hardened_usercopy= 1686 [KNL] Under CONFIG_HARDENED_USERCOPY, whether 1687 hardening is enabled for this boot. Hardened 1688 usercopy checking is used to protect the kernel 1689 from reading or writing beyond known memory 1690 allocation boundaries as a proactive defense 1691 against bounds-checking flaws in the kernel's 1692 copy_to_user()/copy_from_user() interface. 1693 on Perform hardened usercopy checks (default). 1694 off Disable hardened usercopy checks. 1695 1696 hardlockup_all_cpu_backtrace= 1697 [KNL] Should the hard-lockup detector generate 1698 backtraces on all cpus. 1699 Format: 0 | 1 1700 1701 hashdist= [KNL,NUMA] Large hashes allocated during boot 1702 are distributed across NUMA nodes. Defaults on 1703 for 64-bit NUMA, off otherwise. 1704 Format: 0 | 1 (for off | on) 1705 1706 hcl= [IA-64] SGI's Hardware Graph compatibility layer 1707 1708 hd= [EIDE] (E)IDE hard drive subsystem geometry 1709 Format: <cyl>,<head>,<sect> 1710 1711 hest_disable [ACPI] 1712 Disable Hardware Error Source Table (HEST) support; 1713 corresponding firmware-first mode error processing 1714 logic will be disabled. 1715 1716 hibernate= [HIBERNATION] 1717 noresume Don't check if there's a hibernation image 1718 present during boot. 1719 nocompress Don't compress/decompress hibernation images. 1720 no Disable hibernation and resume. 1721 protect_image Turn on image protection during restoration 1722 (that will set all pages holding image data 1723 during restoration read-only). 1724 1725 highmem=nn[KMG] [KNL,BOOT] forces the highmem zone to have an exact 1726 size of <nn>. This works even on boxes that have no 1727 highmem otherwise. This also works to reduce highmem 1728 size on bigger boxes. 1729 1730 highres= [KNL] Enable/disable high resolution timer mode. 1731 Valid parameters: "on", "off" 1732 Default: "on" 1733 1734 hlt [BUGS=ARM,SH] 1735 1736 hostname= [KNL] Set the hostname (aka UTS nodename). 1737 Format: <string> 1738 This allows setting the system's hostname during early 1739 startup. This sets the name returned by gethostname. 1740 Using this parameter to set the hostname makes it 1741 possible to ensure the hostname is correctly set before 1742 any userspace processes run, avoiding the possibility 1743 that a process may call gethostname before the hostname 1744 has been explicitly set, resulting in the calling 1745 process getting an incorrect result. The string must 1746 not exceed the maximum allowed hostname length (usually 1747 64 characters) and will be truncated otherwise. 1748 1749 hpet= [X86-32,HPET] option to control HPET usage 1750 Format: { enable (default) | disable | force | 1751 verbose } 1752 disable: disable HPET and use PIT instead 1753 force: allow force enabled of undocumented chips (ICH4, 1754 VIA, nVidia) 1755 verbose: show contents of HPET registers during setup 1756 1757 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET 1758 registers. Default set by CONFIG_HPET_MMAP_DEFAULT. 1759 1760 hugepages= [HW] Number of HugeTLB pages to allocate at boot. 1761 If this follows hugepagesz (below), it specifies 1762 the number of pages of hugepagesz to be allocated. 1763 If this is the first HugeTLB parameter on the command 1764 line, it specifies the number of pages to allocate for 1765 the default huge page size. If using node format, the 1766 number of pages to allocate per-node can be specified. 1767 See also Documentation/admin-guide/mm/hugetlbpage.rst. 1768 Format: <integer> or (node format) 1769 <node>:<integer>[,<node>:<integer>] 1770 1771 hugepagesz= 1772 [HW] The size of the HugeTLB pages. This is used in 1773 conjunction with hugepages (above) to allocate huge 1774 pages of a specific size at boot. The pair 1775 hugepagesz=X hugepages=Y can be specified once for 1776 each supported huge page size. Huge page sizes are 1777 architecture dependent. See also 1778 Documentation/admin-guide/mm/hugetlbpage.rst. 1779 Format: size[KMG] 1780 1781 hugetlb_cma= [HW,CMA] The size of a CMA area used for allocation 1782 of gigantic hugepages. Or using node format, the size 1783 of a CMA area per node can be specified. 1784 Format: nn[KMGTPE] or (node format) 1785 <node>:nn[KMGTPE][,<node>:nn[KMGTPE]] 1786 1787 Reserve a CMA area of given size and allocate gigantic 1788 hugepages using the CMA allocator. If enabled, the 1789 boot-time allocation of gigantic hugepages is skipped. 1790 1791 hugetlb_free_vmemmap= 1792 [KNL] Requires CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP 1793 enabled. 1794 Control if HugeTLB Vmemmap Optimization (HVO) is enabled. 1795 Allows heavy hugetlb users to free up some more 1796 memory (7 * PAGE_SIZE for each 2MB hugetlb page). 1797 Format: { on | off (default) } 1798 1799 on: enable HVO 1800 off: disable HVO 1801 1802 Built with CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP_DEFAULT_ON=y, 1803 the default is on. 1804 1805 Note that the vmemmap pages may be allocated from the added 1806 memory block itself when memory_hotplug.memmap_on_memory is 1807 enabled, those vmemmap pages cannot be optimized even if this 1808 feature is enabled. Other vmemmap pages not allocated from 1809 the added memory block itself do not be affected. 1810 1811 hung_task_panic= 1812 [KNL] Should the hung task detector generate panics. 1813 Format: 0 | 1 1814 1815 A value of 1 instructs the kernel to panic when a 1816 hung task is detected. The default value is controlled 1817 by the CONFIG_BOOTPARAM_HUNG_TASK_PANIC build-time 1818 option. The value selected by this boot parameter can 1819 be changed later by the kernel.hung_task_panic sysctl. 1820 1821 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC) 1822 terminal devices. Valid values: 0..8 1823 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs. 1824 If specified, z/VM IUCV HVC accepts connections 1825 from listed z/VM user IDs only. 1826 1827 hv_nopvspin [X86,HYPER_V] Disables the paravirt spinlock optimizations 1828 which allow the hypervisor to 'idle' the 1829 guest on lock contention. 1830 1831 i2c_bus= [HW] Override the default board specific I2C bus speed 1832 or register an additional I2C bus that is not 1833 registered from board initialization code. 1834 Format: 1835 <bus_id>,<clkrate> 1836 1837 i8042.debug [HW] Toggle i8042 debug mode 1838 i8042.unmask_kbd_data 1839 [HW] Enable printing of interrupt data from the KBD port 1840 (disabled by default, and as a pre-condition 1841 requires that i8042.debug=1 be enabled) 1842 i8042.direct [HW] Put keyboard port into non-translated mode 1843 i8042.dumbkbd [HW] Pretend that controller can only read data from 1844 keyboard and cannot control its state 1845 (Don't attempt to blink the leds) 1846 i8042.noaux [HW] Don't check for auxiliary (== mouse) port 1847 i8042.nokbd [HW] Don't check/create keyboard port 1848 i8042.noloop [HW] Disable the AUX Loopback command while probing 1849 for the AUX port 1850 i8042.nomux [HW] Don't check presence of an active multiplexing 1851 controller 1852 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX 1853 controllers 1854 i8042.notimeout [HW] Ignore timeout condition signalled by controller 1855 i8042.reset [HW] Reset the controller during init, cleanup and 1856 suspend-to-ram transitions, only during s2r 1857 transitions, or never reset 1858 Format: { 1 | Y | y | 0 | N | n } 1859 1, Y, y: always reset controller 1860 0, N, n: don't ever reset controller 1861 Default: only on s2r transitions on x86; most other 1862 architectures force reset to be always executed 1863 i8042.unlock [HW] Unlock (ignore) the keylock 1864 i8042.kbdreset [HW] Reset device connected to KBD port 1865 i8042.probe_defer 1866 [HW] Allow deferred probing upon i8042 probe errors 1867 1868 i810= [HW,DRM] 1869 1870 i915.invert_brightness= 1871 [DRM] Invert the sense of the variable that is used to 1872 set the brightness of the panel backlight. Normally a 1873 brightness value of 0 indicates backlight switched off, 1874 and the maximum of the brightness value sets the backlight 1875 to maximum brightness. If this parameter is set to 0 1876 (default) and the machine requires it, or this parameter 1877 is set to 1, a brightness value of 0 sets the backlight 1878 to maximum brightness, and the maximum of the brightness 1879 value switches the backlight off. 1880 -1 -- never invert brightness 1881 0 -- machine default 1882 1 -- force brightness inversion 1883 1884 icn= [HW,ISDN] 1885 Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]] 1886 1887 1888 idle= [X86] 1889 Format: idle=poll, idle=halt, idle=nomwait 1890 Poll forces a polling idle loop that can slightly 1891 improve the performance of waking up a idle CPU, but 1892 will use a lot of power and make the system run hot. 1893 Not recommended. 1894 idle=halt: Halt is forced to be used for CPU idle. 1895 In such case C2/C3 won't be used again. 1896 idle=nomwait: Disable mwait for CPU C-states 1897 1898 idxd.sva= [HW] 1899 Format: <bool> 1900 Allow force disabling of Shared Virtual Memory (SVA) 1901 support for the idxd driver. By default it is set to 1902 true (1). 1903 1904 idxd.tc_override= [HW] 1905 Format: <bool> 1906 Allow override of default traffic class configuration 1907 for the device. By default it is set to false (0). 1908 1909 ieee754= [MIPS] Select IEEE Std 754 conformance mode 1910 Format: { strict | legacy | 2008 | relaxed } 1911 Default: strict 1912 1913 Choose which programs will be accepted for execution 1914 based on the IEEE 754 NaN encoding(s) supported by 1915 the FPU and the NaN encoding requested with the value 1916 of an ELF file header flag individually set by each 1917 binary. Hardware implementations are permitted to 1918 support either or both of the legacy and the 2008 NaN 1919 encoding mode. 1920 1921 Available settings are as follows: 1922 strict accept binaries that request a NaN encoding 1923 supported by the FPU 1924 legacy only accept legacy-NaN binaries, if supported 1925 by the FPU 1926 2008 only accept 2008-NaN binaries, if supported 1927 by the FPU 1928 relaxed accept any binaries regardless of whether 1929 supported by the FPU 1930 1931 The FPU emulator is always able to support both NaN 1932 encodings, so if no FPU hardware is present or it has 1933 been disabled with 'nofpu', then the settings of 1934 'legacy' and '2008' strap the emulator accordingly, 1935 'relaxed' straps the emulator for both legacy-NaN and 1936 2008-NaN, whereas 'strict' enables legacy-NaN only on 1937 legacy processors and both NaN encodings on MIPS32 or 1938 MIPS64 CPUs. 1939 1940 The setting for ABS.fmt/NEG.fmt instruction execution 1941 mode generally follows that for the NaN encoding, 1942 except where unsupported by hardware. 1943 1944 ignore_loglevel [KNL] 1945 Ignore loglevel setting - this will print /all/ 1946 kernel messages to the console. Useful for debugging. 1947 We also add it as printk module parameter, so users 1948 could change it dynamically, usually by 1949 /sys/module/printk/parameters/ignore_loglevel. 1950 1951 ignore_rlimit_data 1952 Ignore RLIMIT_DATA setting for data mappings, 1953 print warning at first misuse. Can be changed via 1954 /sys/module/kernel/parameters/ignore_rlimit_data. 1955 1956 ihash_entries= [KNL] 1957 Set number of hash buckets for inode cache. 1958 1959 ima_appraise= [IMA] appraise integrity measurements 1960 Format: { "off" | "enforce" | "fix" | "log" } 1961 default: "enforce" 1962 1963 ima_appraise_tcb [IMA] Deprecated. Use ima_policy= instead. 1964 The builtin appraise policy appraises all files 1965 owned by uid=0. 1966 1967 ima_canonical_fmt [IMA] 1968 Use the canonical format for the binary runtime 1969 measurements, instead of host native format. 1970 1971 ima_hash= [IMA] 1972 Format: { md5 | sha1 | rmd160 | sha256 | sha384 1973 | sha512 | ... } 1974 default: "sha1" 1975 1976 The list of supported hash algorithms is defined 1977 in crypto/hash_info.h. 1978 1979 ima_policy= [IMA] 1980 The builtin policies to load during IMA setup. 1981 Format: "tcb | appraise_tcb | secure_boot | 1982 fail_securely | critical_data" 1983 1984 The "tcb" policy measures all programs exec'd, files 1985 mmap'd for exec, and all files opened with the read 1986 mode bit set by either the effective uid (euid=0) or 1987 uid=0. 1988 1989 The "appraise_tcb" policy appraises the integrity of 1990 all files owned by root. 1991 1992 The "secure_boot" policy appraises the integrity 1993 of files (eg. kexec kernel image, kernel modules, 1994 firmware, policy, etc) based on file signatures. 1995 1996 The "fail_securely" policy forces file signature 1997 verification failure also on privileged mounted 1998 filesystems with the SB_I_UNVERIFIABLE_SIGNATURE 1999 flag. 2000 2001 The "critical_data" policy measures kernel integrity 2002 critical data. 2003 2004 ima_tcb [IMA] Deprecated. Use ima_policy= instead. 2005 Load a policy which meets the needs of the Trusted 2006 Computing Base. This means IMA will measure all 2007 programs exec'd, files mmap'd for exec, and all files 2008 opened for read by uid=0. 2009 2010 ima_template= [IMA] 2011 Select one of defined IMA measurements template formats. 2012 Formats: { "ima" | "ima-ng" | "ima-ngv2" | "ima-sig" | 2013 "ima-sigv2" } 2014 Default: "ima-ng" 2015 2016 ima_template_fmt= 2017 [IMA] Define a custom template format. 2018 Format: { "field1|...|fieldN" } 2019 2020 ima.ahash_minsize= [IMA] Minimum file size for asynchronous hash usage 2021 Format: <min_file_size> 2022 Set the minimal file size for using asynchronous hash. 2023 If left unspecified, ahash usage is disabled. 2024 2025 ahash performance varies for different data sizes on 2026 different crypto accelerators. This option can be used 2027 to achieve the best performance for a particular HW. 2028 2029 ima.ahash_bufsize= [IMA] Asynchronous hash buffer size 2030 Format: <bufsize> 2031 Set hashing buffer size. Default: 4k. 2032 2033 ahash performance varies for different chunk sizes on 2034 different crypto accelerators. This option can be used 2035 to achieve best performance for particular HW. 2036 2037 init= [KNL] 2038 Format: <full_path> 2039 Run specified binary instead of /sbin/init as init 2040 process. 2041 2042 initcall_debug [KNL] Trace initcalls as they are executed. Useful 2043 for working out where the kernel is dying during 2044 startup. 2045 2046 initcall_blacklist= [KNL] Do not execute a comma-separated list of 2047 initcall functions. Useful for debugging built-in 2048 modules and initcalls. 2049 2050 initramfs_async= [KNL] 2051 Format: <bool> 2052 Default: 1 2053 This parameter controls whether the initramfs 2054 image is unpacked asynchronously, concurrently 2055 with devices being probed and 2056 initialized. This should normally just work, 2057 but as a debugging aid, one can get the 2058 historical behaviour of the initramfs 2059 unpacking being completed before device_ and 2060 late_ initcalls. 2061 2062 initrd= [BOOT] Specify the location of the initial ramdisk 2063 2064 initrdmem= [KNL] Specify a physical address and size from which to 2065 load the initrd. If an initrd is compiled in or 2066 specified in the bootparams, it takes priority over this 2067 setting. 2068 Format: ss[KMG],nn[KMG] 2069 Default is 0, 0 2070 2071 init_on_alloc= [MM] Fill newly allocated pages and heap objects with 2072 zeroes. 2073 Format: 0 | 1 2074 Default set by CONFIG_INIT_ON_ALLOC_DEFAULT_ON. 2075 2076 init_on_free= [MM] Fill freed pages and heap objects with zeroes. 2077 Format: 0 | 1 2078 Default set by CONFIG_INIT_ON_FREE_DEFAULT_ON. 2079 2080 init_pkru= [X86] Specify the default memory protection keys rights 2081 register contents for all processes. 0x55555554 by 2082 default (disallow access to all but pkey 0). Can 2083 override in debugfs after boot. 2084 2085 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver 2086 Format: <irq> 2087 2088 int_pln_enable [X86] Enable power limit notification interrupt 2089 2090 integrity_audit=[IMA] 2091 Format: { "0" | "1" } 2092 0 -- basic integrity auditing messages. (Default) 2093 1 -- additional integrity auditing messages. 2094 2095 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option 2096 on 2097 Enable intel iommu driver. 2098 off 2099 Disable intel iommu driver. 2100 igfx_off [Default Off] 2101 By default, gfx is mapped as normal device. If a gfx 2102 device has a dedicated DMAR unit, the DMAR unit is 2103 bypassed by not enabling DMAR with this option. In 2104 this case, gfx device will use physical address for 2105 DMA. 2106 strict [Default Off] 2107 Deprecated, equivalent to iommu.strict=1. 2108 sp_off [Default Off] 2109 By default, super page will be supported if Intel IOMMU 2110 has the capability. With this option, super page will 2111 not be supported. 2112 sm_on 2113 Enable the Intel IOMMU scalable mode if the hardware 2114 advertises that it has support for the scalable mode 2115 translation. 2116 sm_off 2117 Disallow use of the Intel IOMMU scalable mode. 2118 tboot_noforce [Default Off] 2119 Do not force the Intel IOMMU enabled under tboot. 2120 By default, tboot will force Intel IOMMU on, which 2121 could harm performance of some high-throughput 2122 devices like 40GBit network cards, even if identity 2123 mapping is enabled. 2124 Note that using this option lowers the security 2125 provided by tboot because it makes the system 2126 vulnerable to DMA attacks. 2127 2128 intel_idle.max_cstate= [KNL,HW,ACPI,X86] 2129 0 disables intel_idle and fall back on acpi_idle. 2130 1 to 9 specify maximum depth of C-state. 2131 2132 intel_pstate= [X86] 2133 disable 2134 Do not enable intel_pstate as the default 2135 scaling driver for the supported processors 2136 active 2137 Use intel_pstate driver to bypass the scaling 2138 governors layer of cpufreq and provides it own 2139 algorithms for p-state selection. There are two 2140 P-state selection algorithms provided by 2141 intel_pstate in the active mode: powersave and 2142 performance. The way they both operate depends 2143 on whether or not the hardware managed P-states 2144 (HWP) feature has been enabled in the processor 2145 and possibly on the processor model. 2146 passive 2147 Use intel_pstate as a scaling driver, but configure it 2148 to work with generic cpufreq governors (instead of 2149 enabling its internal governor). This mode cannot be 2150 used along with the hardware-managed P-states (HWP) 2151 feature. 2152 force 2153 Enable intel_pstate on systems that prohibit it by default 2154 in favor of acpi-cpufreq. Forcing the intel_pstate driver 2155 instead of acpi-cpufreq may disable platform features, such 2156 as thermal controls and power capping, that rely on ACPI 2157 P-States information being indicated to OSPM and therefore 2158 should be used with caution. This option does not work with 2159 processors that aren't supported by the intel_pstate driver 2160 or on platforms that use pcc-cpufreq instead of acpi-cpufreq. 2161 no_hwp 2162 Do not enable hardware P state control (HWP) 2163 if available. 2164 hwp_only 2165 Only load intel_pstate on systems which support 2166 hardware P state control (HWP) if available. 2167 support_acpi_ppc 2168 Enforce ACPI _PPC performance limits. If the Fixed ACPI 2169 Description Table, specifies preferred power management 2170 profile as "Enterprise Server" or "Performance Server", 2171 then this feature is turned on by default. 2172 per_cpu_perf_limits 2173 Allow per-logical-CPU P-State performance control limits using 2174 cpufreq sysfs interface 2175 2176 intremap= [X86-64, Intel-IOMMU] 2177 on enable Interrupt Remapping (default) 2178 off disable Interrupt Remapping 2179 nosid disable Source ID checking 2180 no_x2apic_optout 2181 BIOS x2APIC opt-out request will be ignored 2182 nopost disable Interrupt Posting 2183 2184 iomem= Disable strict checking of access to MMIO memory 2185 strict regions from userspace. 2186 relaxed 2187 2188 iommu= [X86] 2189 off 2190 force 2191 noforce 2192 biomerge 2193 panic 2194 nopanic 2195 merge 2196 nomerge 2197 soft 2198 pt [X86] 2199 nopt [X86] 2200 nobypass [PPC/POWERNV] 2201 Disable IOMMU bypass, using IOMMU for PCI devices. 2202 2203 iommu.forcedac= [ARM64, X86] Control IOVA allocation for PCI devices. 2204 Format: { "0" | "1" } 2205 0 - Try to allocate a 32-bit DMA address first, before 2206 falling back to the full range if needed. 2207 1 - Allocate directly from the full usable range, 2208 forcing Dual Address Cycle for PCI cards supporting 2209 greater than 32-bit addressing. 2210 2211 iommu.strict= [ARM64, X86] Configure TLB invalidation behaviour 2212 Format: { "0" | "1" } 2213 0 - Lazy mode. 2214 Request that DMA unmap operations use deferred 2215 invalidation of hardware TLBs, for increased 2216 throughput at the cost of reduced device isolation. 2217 Will fall back to strict mode if not supported by 2218 the relevant IOMMU driver. 2219 1 - Strict mode. 2220 DMA unmap operations invalidate IOMMU hardware TLBs 2221 synchronously. 2222 unset - Use value of CONFIG_IOMMU_DEFAULT_DMA_{LAZY,STRICT}. 2223 Note: on x86, strict mode specified via one of the 2224 legacy driver-specific options takes precedence. 2225 2226 iommu.passthrough= 2227 [ARM64, X86] Configure DMA to bypass the IOMMU by default. 2228 Format: { "0" | "1" } 2229 0 - Use IOMMU translation for DMA. 2230 1 - Bypass the IOMMU for DMA. 2231 unset - Use value of CONFIG_IOMMU_DEFAULT_PASSTHROUGH. 2232 2233 io7= [HW] IO7 for Marvel-based Alpha systems 2234 See comment before marvel_specify_io7 in 2235 arch/alpha/kernel/core_marvel.c. 2236 2237 io_delay= [X86] I/O delay method 2238 0x80 2239 Standard port 0x80 based delay 2240 0xed 2241 Alternate port 0xed based delay (needed on some systems) 2242 udelay 2243 Simple two microseconds delay 2244 none 2245 No delay 2246 2247 ip= [IP_PNP] 2248 See Documentation/admin-guide/nfs/nfsroot.rst. 2249 2250 ipcmni_extend [KNL] Extend the maximum number of unique System V 2251 IPC identifiers from 32,768 to 16,777,216. 2252 2253 irqaffinity= [SMP] Set the default irq affinity mask 2254 The argument is a cpu list, as described above. 2255 2256 irqchip.gicv2_force_probe= 2257 [ARM, ARM64] 2258 Format: <bool> 2259 Force the kernel to look for the second 4kB page 2260 of a GICv2 controller even if the memory range 2261 exposed by the device tree is too small. 2262 2263 irqchip.gicv3_nolpi= 2264 [ARM, ARM64] 2265 Force the kernel to ignore the availability of 2266 LPIs (and by consequence ITSs). Intended for system 2267 that use the kernel as a bootloader, and thus want 2268 to let secondary kernels in charge of setting up 2269 LPIs. 2270 2271 irqchip.gicv3_pseudo_nmi= [ARM64] 2272 Enables support for pseudo-NMIs in the kernel. This 2273 requires the kernel to be built with 2274 CONFIG_ARM64_PSEUDO_NMI. 2275 2276 irqfixup [HW] 2277 When an interrupt is not handled search all handlers 2278 for it. Intended to get systems with badly broken 2279 firmware running. 2280 2281 irqpoll [HW] 2282 When an interrupt is not handled search all handlers 2283 for it. Also check all handlers each timer 2284 interrupt. Intended to get systems with badly broken 2285 firmware running. 2286 2287 isapnp= [ISAPNP] 2288 Format: <RDP>,<reset>,<pci_scan>,<verbosity> 2289 2290 isolcpus= [KNL,SMP,ISOL] Isolate a given set of CPUs from disturbance. 2291 [Deprecated - use cpusets instead] 2292 Format: [flag-list,]<cpu-list> 2293 2294 Specify one or more CPUs to isolate from disturbances 2295 specified in the flag list (default: domain): 2296 2297 nohz 2298 Disable the tick when a single task runs. 2299 2300 A residual 1Hz tick is offloaded to workqueues, which you 2301 need to affine to housekeeping through the global 2302 workqueue's affinity configured via the 2303 /sys/devices/virtual/workqueue/cpumask sysfs file, or 2304 by using the 'domain' flag described below. 2305 2306 NOTE: by default the global workqueue runs on all CPUs, 2307 so to protect individual CPUs the 'cpumask' file has to 2308 be configured manually after bootup. 2309 2310 domain 2311 Isolate from the general SMP balancing and scheduling 2312 algorithms. Note that performing domain isolation this way 2313 is irreversible: it's not possible to bring back a CPU to 2314 the domains once isolated through isolcpus. It's strongly 2315 advised to use cpusets instead to disable scheduler load 2316 balancing through the "cpuset.sched_load_balance" file. 2317 It offers a much more flexible interface where CPUs can 2318 move in and out of an isolated set anytime. 2319 2320 You can move a process onto or off an "isolated" CPU via 2321 the CPU affinity syscalls or cpuset. 2322 <cpu number> begins at 0 and the maximum value is 2323 "number of CPUs in system - 1". 2324 2325 managed_irq 2326 2327 Isolate from being targeted by managed interrupts 2328 which have an interrupt mask containing isolated 2329 CPUs. The affinity of managed interrupts is 2330 handled by the kernel and cannot be changed via 2331 the /proc/irq/* interfaces. 2332 2333 This isolation is best effort and only effective 2334 if the automatically assigned interrupt mask of a 2335 device queue contains isolated and housekeeping 2336 CPUs. If housekeeping CPUs are online then such 2337 interrupts are directed to the housekeeping CPU 2338 so that IO submitted on the housekeeping CPU 2339 cannot disturb the isolated CPU. 2340 2341 If a queue's affinity mask contains only isolated 2342 CPUs then this parameter has no effect on the 2343 interrupt routing decision, though interrupts are 2344 only delivered when tasks running on those 2345 isolated CPUs submit IO. IO submitted on 2346 housekeeping CPUs has no influence on those 2347 queues. 2348 2349 The format of <cpu-list> is described above. 2350 2351 iucv= [HW,NET] 2352 2353 ivrs_ioapic [HW,X86-64] 2354 Provide an override to the IOAPIC-ID<->DEVICE-ID 2355 mapping provided in the IVRS ACPI table. 2356 By default, PCI segment is 0, and can be omitted. 2357 2358 For example, to map IOAPIC-ID decimal 10 to 2359 PCI segment 0x1 and PCI device 00:14.0, 2360 write the parameter as: 2361 ivrs_ioapic=10@0001:00:14.0 2362 2363 Deprecated formats: 2364 * To map IOAPIC-ID decimal 10 to PCI device 00:14.0 2365 write the parameter as: 2366 ivrs_ioapic[10]=00:14.0 2367 * To map IOAPIC-ID decimal 10 to PCI segment 0x1 and 2368 PCI device 00:14.0 write the parameter as: 2369 ivrs_ioapic[10]=0001:00:14.0 2370 2371 ivrs_hpet [HW,X86-64] 2372 Provide an override to the HPET-ID<->DEVICE-ID 2373 mapping provided in the IVRS ACPI table. 2374 By default, PCI segment is 0, and can be omitted. 2375 2376 For example, to map HPET-ID decimal 10 to 2377 PCI segment 0x1 and PCI device 00:14.0, 2378 write the parameter as: 2379 ivrs_hpet=10@0001:00:14.0 2380 2381 Deprecated formats: 2382 * To map HPET-ID decimal 0 to PCI device 00:14.0 2383 write the parameter as: 2384 ivrs_hpet[0]=00:14.0 2385 * To map HPET-ID decimal 10 to PCI segment 0x1 and 2386 PCI device 00:14.0 write the parameter as: 2387 ivrs_ioapic[10]=0001:00:14.0 2388 2389 ivrs_acpihid [HW,X86-64] 2390 Provide an override to the ACPI-HID:UID<->DEVICE-ID 2391 mapping provided in the IVRS ACPI table. 2392 By default, PCI segment is 0, and can be omitted. 2393 2394 For example, to map UART-HID:UID AMD0020:0 to 2395 PCI segment 0x1 and PCI device ID 00:14.5, 2396 write the parameter as: 2397 ivrs_acpihid=AMD0020:0@0001:00:14.5 2398 2399 Deprecated formats: 2400 * To map UART-HID:UID AMD0020:0 to PCI segment is 0, 2401 PCI device ID 00:14.5, write the parameter as: 2402 ivrs_acpihid[00:14.5]=AMD0020:0 2403 * To map UART-HID:UID AMD0020:0 to PCI segment 0x1 and 2404 PCI device ID 00:14.5, write the parameter as: 2405 ivrs_acpihid[0001:00:14.5]=AMD0020:0 2406 2407 js= [HW,JOY] Analog joystick 2408 See Documentation/input/joydev/joystick.rst. 2409 2410 kasan_multi_shot 2411 [KNL] Enforce KASAN (Kernel Address Sanitizer) to print 2412 report on every invalid memory access. Without this 2413 parameter KASAN will print report only for the first 2414 invalid access. 2415 2416 keep_bootcon [KNL] 2417 Do not unregister boot console at start. This is only 2418 useful for debugging when something happens in the window 2419 between unregistering the boot console and initializing 2420 the real console. 2421 2422 keepinitrd [HW,ARM] 2423 2424 kernelcore= [KNL,X86,IA-64,PPC] 2425 Format: nn[KMGTPE] | nn% | "mirror" 2426 This parameter specifies the amount of memory usable by 2427 the kernel for non-movable allocations. The requested 2428 amount is spread evenly throughout all nodes in the 2429 system as ZONE_NORMAL. The remaining memory is used for 2430 movable memory in its own zone, ZONE_MOVABLE. In the 2431 event, a node is too small to have both ZONE_NORMAL and 2432 ZONE_MOVABLE, kernelcore memory will take priority and 2433 other nodes will have a larger ZONE_MOVABLE. 2434 2435 ZONE_MOVABLE is used for the allocation of pages that 2436 may be reclaimed or moved by the page migration 2437 subsystem. Note that allocations like PTEs-from-HighMem 2438 still use the HighMem zone if it exists, and the Normal 2439 zone if it does not. 2440 2441 It is possible to specify the exact amount of memory in 2442 the form of "nn[KMGTPE]", a percentage of total system 2443 memory in the form of "nn%", or "mirror". If "mirror" 2444 option is specified, mirrored (reliable) memory is used 2445 for non-movable allocations and remaining memory is used 2446 for Movable pages. "nn[KMGTPE]", "nn%", and "mirror" 2447 are exclusive, so you cannot specify multiple forms. 2448 2449 kgdbdbgp= [KGDB,HW] kgdb over EHCI usb debug port. 2450 Format: <Controller#>[,poll interval] 2451 The controller # is the number of the ehci usb debug 2452 port as it is probed via PCI. The poll interval is 2453 optional and is the number seconds in between 2454 each poll cycle to the debug port in case you need 2455 the functionality for interrupting the kernel with 2456 gdb or control-c on the dbgp connection. When 2457 not using this parameter you use sysrq-g to break into 2458 the kernel debugger. 2459 2460 kgdboc= [KGDB,HW] kgdb over consoles. 2461 Requires a tty driver that supports console polling, 2462 or a supported polling keyboard driver (non-usb). 2463 Serial only format: <serial_device>[,baud] 2464 keyboard only format: kbd 2465 keyboard and serial format: kbd,<serial_device>[,baud] 2466 Optional Kernel mode setting: 2467 kms, kbd format: kms,kbd 2468 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud] 2469 2470 kgdboc_earlycon= [KGDB,HW] 2471 If the boot console provides the ability to read 2472 characters and can work in polling mode, you can use 2473 this parameter to tell kgdb to use it as a backend 2474 until the normal console is registered. Intended to 2475 be used together with the kgdboc parameter which 2476 specifies the normal console to transition to. 2477 2478 The name of the early console should be specified 2479 as the value of this parameter. Note that the name of 2480 the early console might be different than the tty 2481 name passed to kgdboc. It's OK to leave the value 2482 blank and the first boot console that implements 2483 read() will be picked. 2484 2485 kgdbwait [KGDB] Stop kernel execution and enter the 2486 kernel debugger at the earliest opportunity. 2487 2488 kmac= [MIPS] Korina ethernet MAC address. 2489 Configure the RouterBoard 532 series on-chip 2490 Ethernet adapter MAC address. 2491 2492 kmemleak= [KNL] Boot-time kmemleak enable/disable 2493 Valid arguments: on, off 2494 Default: on 2495 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y, 2496 the default is off. 2497 2498 kprobe_event=[probe-list] 2499 [FTRACE] Add kprobe events and enable at boot time. 2500 The probe-list is a semicolon delimited list of probe 2501 definitions. Each definition is same as kprobe_events 2502 interface, but the parameters are comma delimited. 2503 For example, to add a kprobe event on vfs_read with 2504 arg1 and arg2, add to the command line; 2505 2506 kprobe_event=p,vfs_read,$arg1,$arg2 2507 2508 See also Documentation/trace/kprobetrace.rst "Kernel 2509 Boot Parameter" section. 2510 2511 kpti= [ARM64] Control page table isolation of user 2512 and kernel address spaces. 2513 Default: enabled on cores which need mitigation. 2514 0: force disabled 2515 1: force enabled 2516 2517 kunit.enable= [KUNIT] Enable executing KUnit tests. Requires 2518 CONFIG_KUNIT to be set to be fully enabled. The 2519 default value can be overridden via 2520 KUNIT_DEFAULT_ENABLED. 2521 Default is 1 (enabled) 2522 2523 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs. 2524 Default is 0 (don't ignore, but inject #GP) 2525 2526 kvm.eager_page_split= 2527 [KVM,X86] Controls whether or not KVM will try to 2528 proactively split all huge pages during dirty logging. 2529 Eager page splitting reduces interruptions to vCPU 2530 execution by eliminating the write-protection faults 2531 and MMU lock contention that would otherwise be 2532 required to split huge pages lazily. 2533 2534 VM workloads that rarely perform writes or that write 2535 only to a small region of VM memory may benefit from 2536 disabling eager page splitting to allow huge pages to 2537 still be used for reads. 2538 2539 The behavior of eager page splitting depends on whether 2540 KVM_DIRTY_LOG_INITIALLY_SET is enabled or disabled. If 2541 disabled, all huge pages in a memslot will be eagerly 2542 split when dirty logging is enabled on that memslot. If 2543 enabled, eager page splitting will be performed during 2544 the KVM_CLEAR_DIRTY ioctl, and only for the pages being 2545 cleared. 2546 2547 Eager page splitting is only supported when kvm.tdp_mmu=Y. 2548 2549 Default is Y (on). 2550 2551 kvm.enable_vmware_backdoor=[KVM] Support VMware backdoor PV interface. 2552 Default is false (don't support). 2553 2554 kvm.nx_huge_pages= 2555 [KVM] Controls the software workaround for the 2556 X86_BUG_ITLB_MULTIHIT bug. 2557 force : Always deploy workaround. 2558 off : Never deploy workaround. 2559 auto : Deploy workaround based on the presence of 2560 X86_BUG_ITLB_MULTIHIT. 2561 2562 Default is 'auto'. 2563 2564 If the software workaround is enabled for the host, 2565 guests do need not to enable it for nested guests. 2566 2567 kvm.nx_huge_pages_recovery_ratio= 2568 [KVM] Controls how many 4KiB pages are periodically zapped 2569 back to huge pages. 0 disables the recovery, otherwise if 2570 the value is N KVM will zap 1/Nth of the 4KiB pages every 2571 period (see below). The default is 60. 2572 2573 kvm.nx_huge_pages_recovery_period_ms= 2574 [KVM] Controls the time period at which KVM zaps 4KiB pages 2575 back to huge pages. If the value is a non-zero N, KVM will 2576 zap a portion (see ratio above) of the pages every N msecs. 2577 If the value is 0 (the default), KVM will pick a period based 2578 on the ratio, such that a page is zapped after 1 hour on average. 2579 2580 kvm-amd.nested= [KVM,AMD] Control nested virtualization feature in 2581 KVM/SVM. Default is 1 (enabled). 2582 2583 kvm-amd.npt= [KVM,AMD] Control KVM's use of Nested Page Tables, 2584 a.k.a. Two-Dimensional Page Tables. Default is 1 2585 (enabled). Disable by KVM if hardware lacks support 2586 for NPT. 2587 2588 kvm-arm.mode= 2589 [KVM,ARM] Select one of KVM/arm64's modes of operation. 2590 2591 none: Forcefully disable KVM. 2592 2593 nvhe: Standard nVHE-based mode, without support for 2594 protected guests. 2595 2596 protected: nVHE-based mode with support for guests whose 2597 state is kept private from the host. 2598 2599 nested: VHE-based mode with support for nested 2600 virtualization. Requires at least ARMv8.3 2601 hardware. 2602 2603 Defaults to VHE/nVHE based on hardware support. Setting 2604 mode to "protected" will disable kexec and hibernation 2605 for the host. "nested" is experimental and should be 2606 used with extreme caution. 2607 2608 kvm-arm.vgic_v3_group0_trap= 2609 [KVM,ARM] Trap guest accesses to GICv3 group-0 2610 system registers 2611 2612 kvm-arm.vgic_v3_group1_trap= 2613 [KVM,ARM] Trap guest accesses to GICv3 group-1 2614 system registers 2615 2616 kvm-arm.vgic_v3_common_trap= 2617 [KVM,ARM] Trap guest accesses to GICv3 common 2618 system registers 2619 2620 kvm-arm.vgic_v4_enable= 2621 [KVM,ARM] Allow use of GICv4 for direct injection of 2622 LPIs. 2623 2624 kvm_cma_resv_ratio=n [PPC] 2625 Reserves given percentage from system memory area for 2626 contiguous memory allocation for KVM hash pagetable 2627 allocation. 2628 By default it reserves 5% of total system memory. 2629 Format: <integer> 2630 Default: 5 2631 2632 kvm-intel.ept= [KVM,Intel] Control KVM's use of Extended Page Tables, 2633 a.k.a. Two-Dimensional Page Tables. Default is 1 2634 (enabled). Disable by KVM if hardware lacks support 2635 for EPT. 2636 2637 kvm-intel.emulate_invalid_guest_state= 2638 [KVM,Intel] Control whether to emulate invalid guest 2639 state. Ignored if kvm-intel.enable_unrestricted_guest=1, 2640 as guest state is never invalid for unrestricted 2641 guests. This param doesn't apply to nested guests (L2), 2642 as KVM never emulates invalid L2 guest state. 2643 Default is 1 (enabled). 2644 2645 kvm-intel.flexpriority= 2646 [KVM,Intel] Control KVM's use of FlexPriority feature 2647 (TPR shadow). Default is 1 (enabled). Disalbe by KVM if 2648 hardware lacks support for it. 2649 2650 kvm-intel.nested= 2651 [KVM,Intel] Control nested virtualization feature in 2652 KVM/VMX. Default is 1 (enabled). 2653 2654 kvm-intel.unrestricted_guest= 2655 [KVM,Intel] Control KVM's use of unrestricted guest 2656 feature (virtualized real and unpaged mode). Default 2657 is 1 (enabled). Disable by KVM if EPT is disabled or 2658 hardware lacks support for it. 2659 2660 kvm-intel.vmentry_l1d_flush=[KVM,Intel] Mitigation for L1 Terminal Fault 2661 CVE-2018-3620. 2662 2663 Valid arguments: never, cond, always 2664 2665 always: L1D cache flush on every VMENTER. 2666 cond: Flush L1D on VMENTER only when the code between 2667 VMEXIT and VMENTER can leak host memory. 2668 never: Disables the mitigation 2669 2670 Default is cond (do L1 cache flush in specific instances) 2671 2672 kvm-intel.vpid= [KVM,Intel] Control KVM's use of Virtual Processor 2673 Identification feature (tagged TLBs). Default is 1 2674 (enabled). Disable by KVM if hardware lacks support 2675 for it. 2676 2677 l1d_flush= [X86,INTEL] 2678 Control mitigation for L1D based snooping vulnerability. 2679 2680 Certain CPUs are vulnerable to an exploit against CPU 2681 internal buffers which can forward information to a 2682 disclosure gadget under certain conditions. 2683 2684 In vulnerable processors, the speculatively 2685 forwarded data can be used in a cache side channel 2686 attack, to access data to which the attacker does 2687 not have direct access. 2688 2689 This parameter controls the mitigation. The 2690 options are: 2691 2692 on - enable the interface for the mitigation 2693 2694 l1tf= [X86] Control mitigation of the L1TF vulnerability on 2695 affected CPUs 2696 2697 The kernel PTE inversion protection is unconditionally 2698 enabled and cannot be disabled. 2699 2700 full 2701 Provides all available mitigations for the 2702 L1TF vulnerability. Disables SMT and 2703 enables all mitigations in the 2704 hypervisors, i.e. unconditional L1D flush. 2705 2706 SMT control and L1D flush control via the 2707 sysfs interface is still possible after 2708 boot. Hypervisors will issue a warning 2709 when the first VM is started in a 2710 potentially insecure configuration, 2711 i.e. SMT enabled or L1D flush disabled. 2712 2713 full,force 2714 Same as 'full', but disables SMT and L1D 2715 flush runtime control. Implies the 2716 'nosmt=force' command line option. 2717 (i.e. sysfs control of SMT is disabled.) 2718 2719 flush 2720 Leaves SMT enabled and enables the default 2721 hypervisor mitigation, i.e. conditional 2722 L1D flush. 2723 2724 SMT control and L1D flush control via the 2725 sysfs interface is still possible after 2726 boot. Hypervisors will issue a warning 2727 when the first VM is started in a 2728 potentially insecure configuration, 2729 i.e. SMT enabled or L1D flush disabled. 2730 2731 flush,nosmt 2732 2733 Disables SMT and enables the default 2734 hypervisor mitigation. 2735 2736 SMT control and L1D flush control via the 2737 sysfs interface is still possible after 2738 boot. Hypervisors will issue a warning 2739 when the first VM is started in a 2740 potentially insecure configuration, 2741 i.e. SMT enabled or L1D flush disabled. 2742 2743 flush,nowarn 2744 Same as 'flush', but hypervisors will not 2745 warn when a VM is started in a potentially 2746 insecure configuration. 2747 2748 off 2749 Disables hypervisor mitigations and doesn't 2750 emit any warnings. 2751 It also drops the swap size and available 2752 RAM limit restriction on both hypervisor and 2753 bare metal. 2754 2755 Default is 'flush'. 2756 2757 For details see: Documentation/admin-guide/hw-vuln/l1tf.rst 2758 2759 l2cr= [PPC] 2760 2761 l3cr= [PPC] 2762 2763 lapic [X86-32,APIC] Enable the local APIC even if BIOS 2764 disabled it. 2765 2766 lapic= [X86,APIC] Do not use TSC deadline 2767 value for LAPIC timer one-shot implementation. Default 2768 back to the programmable timer unit in the LAPIC. 2769 Format: notscdeadline 2770 2771 lapic_timer_c2_ok [X86,APIC] trust the local apic timer 2772 in C2 power state. 2773 2774 libata.dma= [LIBATA] DMA control 2775 libata.dma=0 Disable all PATA and SATA DMA 2776 libata.dma=1 PATA and SATA Disk DMA only 2777 libata.dma=2 ATAPI (CDROM) DMA only 2778 libata.dma=4 Compact Flash DMA only 2779 Combinations also work, so libata.dma=3 enables DMA 2780 for disks and CDROMs, but not CFs. 2781 2782 libata.ignore_hpa= [LIBATA] Ignore HPA limit 2783 libata.ignore_hpa=0 keep BIOS limits (default) 2784 libata.ignore_hpa=1 ignore limits, using full disk 2785 2786 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume 2787 when set. 2788 Format: <int> 2789 2790 libata.force= [LIBATA] Force configurations. The format is a comma- 2791 separated list of "[ID:]VAL" where ID is PORT[.DEVICE]. 2792 PORT and DEVICE are decimal numbers matching port, link 2793 or device. Basically, it matches the ATA ID string 2794 printed on console by libata. If the whole ID part is 2795 omitted, the last PORT and DEVICE values are used. If 2796 ID hasn't been specified yet, the configuration applies 2797 to all ports, links and devices. 2798 2799 If only DEVICE is omitted, the parameter applies to 2800 the port and all links and devices behind it. DEVICE 2801 number of 0 either selects the first device or the 2802 first fan-out link behind PMP device. It does not 2803 select the host link. DEVICE number of 15 selects the 2804 host link and device attached to it. 2805 2806 The VAL specifies the configuration to force. As long 2807 as there is no ambiguity, shortcut notation is allowed. 2808 For example, both 1.5 and 1.5G would work for 1.5Gbps. 2809 The following configurations can be forced. 2810 2811 * Cable type: 40c, 80c, short40c, unk, ign or sata. 2812 Any ID with matching PORT is used. 2813 2814 * SATA link speed limit: 1.5Gbps or 3.0Gbps. 2815 2816 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7]. 2817 udma[/][16,25,33,44,66,100,133] notation is also 2818 allowed. 2819 2820 * nohrst, nosrst, norst: suppress hard, soft and both 2821 resets. 2822 2823 * rstonce: only attempt one reset during hot-unplug 2824 link recovery. 2825 2826 * [no]dbdelay: Enable or disable the extra 200ms delay 2827 before debouncing a link PHY and device presence 2828 detection. 2829 2830 * [no]ncq: Turn on or off NCQ. 2831 2832 * [no]ncqtrim: Enable or disable queued DSM TRIM. 2833 2834 * [no]ncqati: Enable or disable NCQ trim on ATI chipset. 2835 2836 * [no]trim: Enable or disable (unqueued) TRIM. 2837 2838 * trim_zero: Indicate that TRIM command zeroes data. 2839 2840 * max_trim_128m: Set 128M maximum trim size limit. 2841 2842 * [no]dma: Turn on or off DMA transfers. 2843 2844 * atapi_dmadir: Enable ATAPI DMADIR bridge support. 2845 2846 * atapi_mod16_dma: Enable the use of ATAPI DMA for 2847 commands that are not a multiple of 16 bytes. 2848 2849 * [no]dmalog: Enable or disable the use of the 2850 READ LOG DMA EXT command to access logs. 2851 2852 * [no]iddevlog: Enable or disable access to the 2853 identify device data log. 2854 2855 * [no]logdir: Enable or disable access to the general 2856 purpose log directory. 2857 2858 * max_sec_128: Set transfer size limit to 128 sectors. 2859 2860 * max_sec_1024: Set or clear transfer size limit to 2861 1024 sectors. 2862 2863 * max_sec_lba48: Set or clear transfer size limit to 2864 65535 sectors. 2865 2866 * [no]lpm: Enable or disable link power management. 2867 2868 * [no]setxfer: Indicate if transfer speed mode setting 2869 should be skipped. 2870 2871 * [no]fua: Disable or enable FUA (Force Unit Access) 2872 support for devices supporting this feature. 2873 2874 * dump_id: Dump IDENTIFY data. 2875 2876 * disable: Disable this device. 2877 2878 If there are multiple matching configurations changing 2879 the same attribute, the last one is used. 2880 2881 load_ramdisk= [RAM] [Deprecated] 2882 2883 lockd.nlm_grace_period=P [NFS] Assign grace period. 2884 Format: <integer> 2885 2886 lockd.nlm_tcpport=N [NFS] Assign TCP port. 2887 Format: <integer> 2888 2889 lockd.nlm_timeout=T [NFS] Assign timeout value. 2890 Format: <integer> 2891 2892 lockd.nlm_udpport=M [NFS] Assign UDP port. 2893 Format: <integer> 2894 2895 lockdown= [SECURITY] 2896 { integrity | confidentiality } 2897 Enable the kernel lockdown feature. If set to 2898 integrity, kernel features that allow userland to 2899 modify the running kernel are disabled. If set to 2900 confidentiality, kernel features that allow userland 2901 to extract confidential information from the kernel 2902 are also disabled. 2903 2904 locktorture.nreaders_stress= [KNL] 2905 Set the number of locking read-acquisition kthreads. 2906 Defaults to being automatically set based on the 2907 number of online CPUs. 2908 2909 locktorture.nwriters_stress= [KNL] 2910 Set the number of locking write-acquisition kthreads. 2911 2912 locktorture.onoff_holdoff= [KNL] 2913 Set time (s) after boot for CPU-hotplug testing. 2914 2915 locktorture.onoff_interval= [KNL] 2916 Set time (s) between CPU-hotplug operations, or 2917 zero to disable CPU-hotplug testing. 2918 2919 locktorture.shuffle_interval= [KNL] 2920 Set task-shuffle interval (jiffies). Shuffling 2921 tasks allows some CPUs to go into dyntick-idle 2922 mode during the locktorture test. 2923 2924 locktorture.shutdown_secs= [KNL] 2925 Set time (s) after boot system shutdown. This 2926 is useful for hands-off automated testing. 2927 2928 locktorture.stat_interval= [KNL] 2929 Time (s) between statistics printk()s. 2930 2931 locktorture.stutter= [KNL] 2932 Time (s) to stutter testing, for example, 2933 specifying five seconds causes the test to run for 2934 five seconds, wait for five seconds, and so on. 2935 This tests the locking primitive's ability to 2936 transition abruptly to and from idle. 2937 2938 locktorture.torture_type= [KNL] 2939 Specify the locking implementation to test. 2940 2941 locktorture.verbose= [KNL] 2942 Enable additional printk() statements. 2943 2944 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver 2945 Format: <irq> 2946 2947 loglevel= All Kernel Messages with a loglevel smaller than the 2948 console loglevel will be printed to the console. It can 2949 also be changed with klogd or other programs. The 2950 loglevels are defined as follows: 2951 2952 0 (KERN_EMERG) system is unusable 2953 1 (KERN_ALERT) action must be taken immediately 2954 2 (KERN_CRIT) critical conditions 2955 3 (KERN_ERR) error conditions 2956 4 (KERN_WARNING) warning conditions 2957 5 (KERN_NOTICE) normal but significant condition 2958 6 (KERN_INFO) informational 2959 7 (KERN_DEBUG) debug-level messages 2960 2961 log_buf_len=n[KMG] Sets the size of the printk ring buffer, 2962 in bytes. n must be a power of two and greater 2963 than the minimal size. The minimal size is defined 2964 by LOG_BUF_SHIFT kernel config parameter. There is 2965 also CONFIG_LOG_CPU_MAX_BUF_SHIFT config parameter 2966 that allows to increase the default size depending on 2967 the number of CPUs. See init/Kconfig for more details. 2968 2969 logo.nologo [FB] Disables display of the built-in Linux logo. 2970 This may be used to provide more screen space for 2971 kernel log messages and is useful when debugging 2972 kernel boot problems. 2973 2974 lp=0 [LP] Specify parallel ports to use, e.g, 2975 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses 2976 lp=reset first parallel port). 'lp=0' disables the 2977 lp=auto printer driver. 'lp=reset' (which can be 2978 specified in addition to the ports) causes 2979 attached printers to be reset. Using 2980 lp=port1,port2,... specifies the parallel ports 2981 to associate lp devices with, starting with 2982 lp0. A port specification may be 'none' to skip 2983 that lp device, or a parport name such as 2984 'parport0'. Specifying 'lp=auto' instead of a 2985 port specification list means that device IDs 2986 from each port should be examined, to see if 2987 an IEEE 1284-compliant printer is attached; if 2988 so, the driver will manage that printer. 2989 See also header of drivers/char/lp.c. 2990 2991 lpj=n [KNL] 2992 Sets loops_per_jiffy to given constant, thus avoiding 2993 time-consuming boot-time autodetection (up to 250 ms per 2994 CPU). 0 enables autodetection (default). To determine 2995 the correct value for your kernel, boot with normal 2996 autodetection and see what value is printed. Note that 2997 on SMP systems the preset will be applied to all CPUs, 2998 which is likely to cause problems if your CPUs need 2999 significantly divergent settings. An incorrect value 3000 will cause delays in the kernel to be wrong, leading to 3001 unpredictable I/O errors and other breakage. Although 3002 unlikely, in the extreme case this might damage your 3003 hardware. 3004 3005 ltpc= [NET] 3006 Format: <io>,<irq>,<dma> 3007 3008 lsm.debug [SECURITY] Enable LSM initialization debugging output. 3009 3010 lsm=lsm1,...,lsmN 3011 [SECURITY] Choose order of LSM initialization. This 3012 overrides CONFIG_LSM, and the "security=" parameter. 3013 3014 machvec= [IA-64] Force the use of a particular machine-vector 3015 (machvec) in a generic kernel. 3016 Example: machvec=hpzx1 3017 3018 machtype= [Loongson] Share the same kernel image file between 3019 different yeeloong laptops. 3020 Example: machtype=lemote-yeeloong-2f-7inch 3021 3022 max_addr=nn[KMG] [KNL,BOOT,IA-64] All physical memory greater 3023 than or equal to this physical address is ignored. 3024 3025 maxcpus= [SMP] Maximum number of processors that an SMP kernel 3026 will bring up during bootup. maxcpus=n : n >= 0 limits 3027 the kernel to bring up 'n' processors. Surely after 3028 bootup you can bring up the other plugged cpu by executing 3029 "echo 1 > /sys/devices/system/cpu/cpuX/online". So maxcpus 3030 only takes effect during system bootup. 3031 While n=0 is a special case, it is equivalent to "nosmp", 3032 which also disables the IO APIC. 3033 3034 max_loop= [LOOP] The number of loop block devices that get 3035 (loop.max_loop) unconditionally pre-created at init time. The default 3036 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead 3037 of statically allocating a predefined number, loop 3038 devices can be requested on-demand with the 3039 /dev/loop-control interface. 3040 3041 mce [X86-32] Machine Check Exception 3042 3043 mce=option [X86-64] See Documentation/arch/x86/x86_64/boot-options.rst 3044 3045 md= [HW] RAID subsystems devices and level 3046 See Documentation/admin-guide/md.rst. 3047 3048 mdacon= [MDA] 3049 Format: <first>,<last> 3050 Specifies range of consoles to be captured by the MDA. 3051 3052 mds= [X86,INTEL] 3053 Control mitigation for the Micro-architectural Data 3054 Sampling (MDS) vulnerability. 3055 3056 Certain CPUs are vulnerable to an exploit against CPU 3057 internal buffers which can forward information to a 3058 disclosure gadget under certain conditions. 3059 3060 In vulnerable processors, the speculatively 3061 forwarded data can be used in a cache side channel 3062 attack, to access data to which the attacker does 3063 not have direct access. 3064 3065 This parameter controls the MDS mitigation. The 3066 options are: 3067 3068 full - Enable MDS mitigation on vulnerable CPUs 3069 full,nosmt - Enable MDS mitigation and disable 3070 SMT on vulnerable CPUs 3071 off - Unconditionally disable MDS mitigation 3072 3073 On TAA-affected machines, mds=off can be prevented by 3074 an active TAA mitigation as both vulnerabilities are 3075 mitigated with the same mechanism so in order to disable 3076 this mitigation, you need to specify tsx_async_abort=off 3077 too. 3078 3079 Not specifying this option is equivalent to 3080 mds=full. 3081 3082 For details see: Documentation/admin-guide/hw-vuln/mds.rst 3083 3084 mem=nn[KMG] [HEXAGON] Set the memory size. 3085 Must be specified, otherwise memory size will be 0. 3086 3087 mem=nn[KMG] [KNL,BOOT] Force usage of a specific amount of memory 3088 Amount of memory to be used in cases as follows: 3089 3090 1 for test; 3091 2 when the kernel is not able to see the whole system memory; 3092 3 memory that lies after 'mem=' boundary is excluded from 3093 the hypervisor, then assigned to KVM guests. 3094 4 to limit the memory available for kdump kernel. 3095 3096 [ARC,MICROBLAZE] - the limit applies only to low memory, 3097 high memory is not affected. 3098 3099 [ARM64] - only limits memory covered by the linear 3100 mapping. The NOMAP regions are not affected. 3101 3102 [X86] Work as limiting max address. Use together 3103 with memmap= to avoid physical address space collisions. 3104 Without memmap= PCI devices could be placed at addresses 3105 belonging to unused RAM. 3106 3107 Note that this only takes effects during boot time since 3108 in above case 3, memory may need be hot added after boot 3109 if system memory of hypervisor is not sufficient. 3110 3111 mem=nn[KMG]@ss[KMG] 3112 [ARM,MIPS] - override the memory layout reported by 3113 firmware. 3114 Define a memory region of size nn[KMG] starting at 3115 ss[KMG]. 3116 Multiple different regions can be specified with 3117 multiple mem= parameters on the command line. 3118 3119 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel 3120 memory. 3121 3122 memblock=debug [KNL] Enable memblock debug messages. 3123 3124 memchunk=nn[KMG] 3125 [KNL,SH] Allow user to override the default size for 3126 per-device physically contiguous DMA buffers. 3127 3128 memhp_default_state=online/offline 3129 [KNL] Set the initial state for the memory hotplug 3130 onlining policy. If not specified, the default value is 3131 set according to the 3132 CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE kernel config 3133 option. 3134 See Documentation/admin-guide/mm/memory-hotplug.rst. 3135 3136 memmap=exactmap [KNL,X86] Enable setting of an exact 3137 E820 memory map, as specified by the user. 3138 Such memmap=exactmap lines can be constructed based on 3139 BIOS output or other requirements. See the memmap=nn@ss 3140 option description. 3141 3142 memmap=nn[KMG]@ss[KMG] 3143 [KNL, X86, MIPS, XTENSA] Force usage of a specific region of memory. 3144 Region of memory to be used is from ss to ss+nn. 3145 If @ss[KMG] is omitted, it is equivalent to mem=nn[KMG], 3146 which limits max address to nn[KMG]. 3147 Multiple different regions can be specified, 3148 comma delimited. 3149 Example: 3150 memmap=100M@2G,100M#3G,1G!1024G 3151 3152 memmap=nn[KMG]#ss[KMG] 3153 [KNL,ACPI] Mark specific memory as ACPI data. 3154 Region of memory to be marked is from ss to ss+nn. 3155 3156 memmap=nn[KMG]$ss[KMG] 3157 [KNL,ACPI] Mark specific memory as reserved. 3158 Region of memory to be reserved is from ss to ss+nn. 3159 Example: Exclude memory from 0x18690000-0x1869ffff 3160 memmap=64K$0x18690000 3161 or 3162 memmap=0x10000$0x18690000 3163 Some bootloaders may need an escape character before '$', 3164 like Grub2, otherwise '$' and the following number 3165 will be eaten. 3166 3167 memmap=nn[KMG]!ss[KMG] 3168 [KNL,X86] Mark specific memory as protected. 3169 Region of memory to be used, from ss to ss+nn. 3170 The memory region may be marked as e820 type 12 (0xc) 3171 and is NVDIMM or ADR memory. 3172 3173 memmap=<size>%<offset>-<oldtype>+<newtype> 3174 [KNL,ACPI] Convert memory within the specified region 3175 from <oldtype> to <newtype>. If "-<oldtype>" is left 3176 out, the whole region will be marked as <newtype>, 3177 even if previously unavailable. If "+<newtype>" is left 3178 out, matching memory will be removed. Types are 3179 specified as e820 types, e.g., 1 = RAM, 2 = reserved, 3180 3 = ACPI, 12 = PRAM. 3181 3182 memory_corruption_check=0/1 [X86] 3183 Some BIOSes seem to corrupt the first 64k of 3184 memory when doing things like suspend/resume. 3185 Setting this option will scan the memory 3186 looking for corruption. Enabling this will 3187 both detect corruption and prevent the kernel 3188 from using the memory being corrupted. 3189 However, its intended as a diagnostic tool; if 3190 repeatable BIOS-originated corruption always 3191 affects the same memory, you can use memmap= 3192 to prevent the kernel from using that memory. 3193 3194 memory_corruption_check_size=size [X86] 3195 By default it checks for corruption in the low 3196 64k, making this memory unavailable for normal 3197 use. Use this parameter to scan for 3198 corruption in more or less memory. 3199 3200 memory_corruption_check_period=seconds [X86] 3201 By default it checks for corruption every 60 3202 seconds. Use this parameter to check at some 3203 other rate. 0 disables periodic checking. 3204 3205 memory_hotplug.memmap_on_memory 3206 [KNL,X86,ARM] Boolean flag to enable this feature. 3207 Format: {on | off (default)} 3208 When enabled, runtime hotplugged memory will 3209 allocate its internal metadata (struct pages, 3210 those vmemmap pages cannot be optimized even 3211 if hugetlb_free_vmemmap is enabled) from the 3212 hotadded memory which will allow to hotadd a 3213 lot of memory without requiring additional 3214 memory to do so. 3215 This feature is disabled by default because it 3216 has some implication on large (e.g. GB) 3217 allocations in some configurations (e.g. small 3218 memory blocks). 3219 The state of the flag can be read in 3220 /sys/module/memory_hotplug/parameters/memmap_on_memory. 3221 Note that even when enabled, there are a few cases where 3222 the feature is not effective. 3223 3224 memtest= [KNL,X86,ARM,M68K,PPC,RISCV] Enable memtest 3225 Format: <integer> 3226 default : 0 <disable> 3227 Specifies the number of memtest passes to be 3228 performed. Each pass selects another test 3229 pattern from a given set of patterns. Memtest 3230 fills the memory with this pattern, validates 3231 memory contents and reserves bad memory 3232 regions that are detected. 3233 3234 mem_encrypt= [X86-64] AMD Secure Memory Encryption (SME) control 3235 Valid arguments: on, off 3236 Default (depends on kernel configuration option): 3237 on (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=y) 3238 off (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=n) 3239 mem_encrypt=on: Activate SME 3240 mem_encrypt=off: Do not activate SME 3241 3242 Refer to Documentation/virt/kvm/x86/amd-memory-encryption.rst 3243 for details on when memory encryption can be activated. 3244 3245 mem_sleep_default= [SUSPEND] Default system suspend mode: 3246 s2idle - Suspend-To-Idle 3247 shallow - Power-On Suspend or equivalent (if supported) 3248 deep - Suspend-To-RAM or equivalent (if supported) 3249 See Documentation/admin-guide/pm/sleep-states.rst. 3250 3251 mfgpt_irq= [IA-32] Specify the IRQ to use for the 3252 Multi-Function General Purpose Timers on AMD Geode 3253 platforms. 3254 3255 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when 3256 the BIOS has incorrectly applied a workaround. TinyBIOS 3257 version 0.98 is known to be affected, 0.99 fixes the 3258 problem by letting the user disable the workaround. 3259 3260 mga= [HW,DRM] 3261 3262 min_addr=nn[KMG] [KNL,BOOT,IA-64] All physical memory below this 3263 physical address is ignored. 3264 3265 mini2440= [ARM,HW,KNL] 3266 Format:[0..2][b][c][t] 3267 Default: "0tb" 3268 MINI2440 configuration specification: 3269 0 - The attached screen is the 3.5" TFT 3270 1 - The attached screen is the 7" TFT 3271 2 - The VGA Shield is attached (1024x768) 3272 Leaving out the screen size parameter will not load 3273 the TFT driver, and the framebuffer will be left 3274 unconfigured. 3275 b - Enable backlight. The TFT backlight pin will be 3276 linked to the kernel VESA blanking code and a GPIO 3277 LED. This parameter is not necessary when using the 3278 VGA shield. 3279 c - Enable the s3c camera interface. 3280 t - Reserved for enabling touchscreen support. The 3281 touchscreen support is not enabled in the mainstream 3282 kernel as of 2.6.30, a preliminary port can be found 3283 in the "bleeding edge" mini2440 support kernel at 3284 https://repo.or.cz/w/linux-2.6/mini2440.git 3285 3286 mitigations= 3287 [X86,PPC,S390,ARM64] Control optional mitigations for 3288 CPU vulnerabilities. This is a set of curated, 3289 arch-independent options, each of which is an 3290 aggregation of existing arch-specific options. 3291 3292 off 3293 Disable all optional CPU mitigations. This 3294 improves system performance, but it may also 3295 expose users to several CPU vulnerabilities. 3296 Equivalent to: if nokaslr then kpti=0 [ARM64] 3297 gather_data_sampling=off [X86] 3298 kvm.nx_huge_pages=off [X86] 3299 l1tf=off [X86] 3300 mds=off [X86] 3301 mmio_stale_data=off [X86] 3302 no_entry_flush [PPC] 3303 no_uaccess_flush [PPC] 3304 nobp=0 [S390] 3305 nopti [X86,PPC] 3306 nospectre_bhb [ARM64] 3307 nospectre_v1 [X86,PPC] 3308 nospectre_v2 [X86,PPC,S390,ARM64] 3309 retbleed=off [X86] 3310 spec_store_bypass_disable=off [X86,PPC] 3311 spectre_v2_user=off [X86] 3312 srbds=off [X86,INTEL] 3313 ssbd=force-off [ARM64] 3314 tsx_async_abort=off [X86] 3315 3316 Exceptions: 3317 This does not have any effect on 3318 kvm.nx_huge_pages when 3319 kvm.nx_huge_pages=force. 3320 3321 auto (default) 3322 Mitigate all CPU vulnerabilities, but leave SMT 3323 enabled, even if it's vulnerable. This is for 3324 users who don't want to be surprised by SMT 3325 getting disabled across kernel upgrades, or who 3326 have other ways of avoiding SMT-based attacks. 3327 Equivalent to: (default behavior) 3328 3329 auto,nosmt 3330 Mitigate all CPU vulnerabilities, disabling SMT 3331 if needed. This is for users who always want to 3332 be fully mitigated, even if it means losing SMT. 3333 Equivalent to: l1tf=flush,nosmt [X86] 3334 mds=full,nosmt [X86] 3335 tsx_async_abort=full,nosmt [X86] 3336 mmio_stale_data=full,nosmt [X86] 3337 retbleed=auto,nosmt [X86] 3338 3339 mminit_loglevel= 3340 [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this 3341 parameter allows control of the logging verbosity for 3342 the additional memory initialisation checks. A value 3343 of 0 disables mminit logging and a level of 4 will 3344 log everything. Information is printed at KERN_DEBUG 3345 so loglevel=8 may also need to be specified. 3346 3347 mmio_stale_data= 3348 [X86,INTEL] Control mitigation for the Processor 3349 MMIO Stale Data vulnerabilities. 3350 3351 Processor MMIO Stale Data is a class of 3352 vulnerabilities that may expose data after an MMIO 3353 operation. Exposed data could originate or end in 3354 the same CPU buffers as affected by MDS and TAA. 3355 Therefore, similar to MDS and TAA, the mitigation 3356 is to clear the affected CPU buffers. 3357 3358 This parameter controls the mitigation. The 3359 options are: 3360 3361 full - Enable mitigation on vulnerable CPUs 3362 3363 full,nosmt - Enable mitigation and disable SMT on 3364 vulnerable CPUs. 3365 3366 off - Unconditionally disable mitigation 3367 3368 On MDS or TAA affected machines, 3369 mmio_stale_data=off can be prevented by an active 3370 MDS or TAA mitigation as these vulnerabilities are 3371 mitigated with the same mechanism so in order to 3372 disable this mitigation, you need to specify 3373 mds=off and tsx_async_abort=off too. 3374 3375 Not specifying this option is equivalent to 3376 mmio_stale_data=full. 3377 3378 For details see: 3379 Documentation/admin-guide/hw-vuln/processor_mmio_stale_data.rst 3380 3381 <module>.async_probe[=<bool>] [KNL] 3382 If no <bool> value is specified or if the value 3383 specified is not a valid <bool>, enable asynchronous 3384 probe on this module. Otherwise, enable/disable 3385 asynchronous probe on this module as indicated by the 3386 <bool> value. See also: module.async_probe 3387 3388 module.async_probe=<bool> 3389 [KNL] When set to true, modules will use async probing 3390 by default. To enable/disable async probing for a 3391 specific module, use the module specific control that 3392 is documented under <module>.async_probe. When both 3393 module.async_probe and <module>.async_probe are 3394 specified, <module>.async_probe takes precedence for 3395 the specific module. 3396 3397 module.enable_dups_trace 3398 [KNL] When CONFIG_MODULE_DEBUG_AUTOLOAD_DUPS is set, 3399 this means that duplicate request_module() calls will 3400 trigger a WARN_ON() instead of a pr_warn(). Note that 3401 if MODULE_DEBUG_AUTOLOAD_DUPS_TRACE is set, WARN_ON()s 3402 will always be issued and this option does nothing. 3403 module.sig_enforce 3404 [KNL] When CONFIG_MODULE_SIG is set, this means that 3405 modules without (valid) signatures will fail to load. 3406 Note that if CONFIG_MODULE_SIG_FORCE is set, that 3407 is always true, so this option does nothing. 3408 3409 module_blacklist= [KNL] Do not load a comma-separated list of 3410 modules. Useful for debugging problem modules. 3411 3412 mousedev.tap_time= 3413 [MOUSE] Maximum time between finger touching and 3414 leaving touchpad surface for touch to be considered 3415 a tap and be reported as a left button click (for 3416 touchpads working in absolute mode only). 3417 Format: <msecs> 3418 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices 3419 reporting absolute coordinates, such as tablets 3420 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices 3421 reporting absolute coordinates, such as tablets 3422 3423 movablecore= [KNL,X86,IA-64,PPC] 3424 Format: nn[KMGTPE] | nn% 3425 This parameter is the complement to kernelcore=, it 3426 specifies the amount of memory used for migratable 3427 allocations. If both kernelcore and movablecore is 3428 specified, then kernelcore will be at *least* the 3429 specified value but may be more. If movablecore on its 3430 own is specified, the administrator must be careful 3431 that the amount of memory usable for all allocations 3432 is not too small. 3433 3434 movable_node [KNL] Boot-time switch to make hotplugable memory 3435 NUMA nodes to be movable. This means that the memory 3436 of such nodes will be usable only for movable 3437 allocations which rules out almost all kernel 3438 allocations. Use with caution! 3439 3440 MTD_Partition= [MTD] 3441 Format: <name>,<region-number>,<size>,<offset> 3442 3443 MTD_Region= [MTD] Format: 3444 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>] 3445 3446 mtdparts= [MTD] 3447 See drivers/mtd/parsers/cmdlinepart.c 3448 3449 mtdset= [ARM] 3450 ARM/S3C2412 JIVE boot control 3451 3452 See arch/arm/mach-s3c/mach-jive.c 3453 3454 mtouchusb.raw_coordinates= 3455 [HW] Make the MicroTouch USB driver use raw coordinates 3456 ('y', default) or cooked coordinates ('n') 3457 3458 mtrr=debug [X86] 3459 Enable printing debug information related to MTRR 3460 registers at boot time. 3461 3462 mtrr_chunk_size=nn[KMG] [X86] 3463 used for mtrr cleanup. It is largest continuous chunk 3464 that could hold holes aka. UC entries. 3465 3466 mtrr_gran_size=nn[KMG] [X86] 3467 Used for mtrr cleanup. It is granularity of mtrr block. 3468 Default is 1. 3469 Large value could prevent small alignment from 3470 using up MTRRs. 3471 3472 mtrr_spare_reg_nr=n [X86] 3473 Format: <integer> 3474 Range: 0,7 : spare reg number 3475 Default : 1 3476 Used for mtrr cleanup. It is spare mtrr entries number. 3477 Set to 2 or more if your graphical card needs more. 3478 3479 multitce=off [PPC] This parameter disables the use of the pSeries 3480 firmware feature for updating multiple TCE entries 3481 at a time. 3482 3483 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card 3484 3485 netdev= [NET] Network devices parameters 3486 Format: <irq>,<io>,<mem_start>,<mem_end>,<name> 3487 Note that mem_start is often overloaded to mean 3488 something different and driver-specific. 3489 This usage is only documented in each driver source 3490 file if at all. 3491 3492 netpoll.carrier_timeout= 3493 [NET] Specifies amount of time (in seconds) that 3494 netpoll should wait for a carrier. By default netpoll 3495 waits 4 seconds. 3496 3497 nf_conntrack.acct= 3498 [NETFILTER] Enable connection tracking flow accounting 3499 0 to disable accounting 3500 1 to enable accounting 3501 Default value is 0. 3502 3503 nfs.cache_getent= 3504 [NFS] sets the pathname to the program which is used 3505 to update the NFS client cache entries. 3506 3507 nfs.cache_getent_timeout= 3508 [NFS] sets the timeout after which an attempt to 3509 update a cache entry is deemed to have failed. 3510 3511 nfs.callback_nr_threads= 3512 [NFSv4] set the total number of threads that the 3513 NFS client will assign to service NFSv4 callback 3514 requests. 3515 3516 nfs.callback_tcpport= 3517 [NFS] set the TCP port on which the NFSv4 callback 3518 channel should listen. 3519 3520 nfs.enable_ino64= 3521 [NFS] enable 64-bit inode numbers. 3522 If zero, the NFS client will fake up a 32-bit inode 3523 number for the readdir() and stat() syscalls instead 3524 of returning the full 64-bit number. 3525 The default is to return 64-bit inode numbers. 3526 3527 nfs.idmap_cache_timeout= 3528 [NFS] set the maximum lifetime for idmapper cache 3529 entries. 3530 3531 nfs.max_session_cb_slots= 3532 [NFSv4.1] Sets the maximum number of session 3533 slots the client will assign to the callback 3534 channel. This determines the maximum number of 3535 callbacks the client will process in parallel for 3536 a particular server. 3537 3538 nfs.max_session_slots= 3539 [NFSv4.1] Sets the maximum number of session slots 3540 the client will attempt to negotiate with the server. 3541 This limits the number of simultaneous RPC requests 3542 that the client can send to the NFSv4.1 server. 3543 Note that there is little point in setting this 3544 value higher than the max_tcp_slot_table_limit. 3545 3546 nfs.nfs4_disable_idmapping= 3547 [NFSv4] When set to the default of '1', this option 3548 ensures that both the RPC level authentication 3549 scheme and the NFS level operations agree to use 3550 numeric uids/gids if the mount is using the 3551 'sec=sys' security flavour. In effect it is 3552 disabling idmapping, which can make migration from 3553 legacy NFSv2/v3 systems to NFSv4 easier. 3554 Servers that do not support this mode of operation 3555 will be autodetected by the client, and it will fall 3556 back to using the idmapper. 3557 To turn off this behaviour, set the value to '0'. 3558 3559 nfs.nfs4_unique_id= 3560 [NFS4] Specify an additional fixed unique ident- 3561 ification string that NFSv4 clients can insert into 3562 their nfs_client_id4 string. This is typically a 3563 UUID that is generated at system install time. 3564 3565 nfs.recover_lost_locks= 3566 [NFSv4] Attempt to recover locks that were lost due 3567 to a lease timeout on the server. Please note that 3568 doing this risks data corruption, since there are 3569 no guarantees that the file will remain unchanged 3570 after the locks are lost. 3571 If you want to enable the kernel legacy behaviour of 3572 attempting to recover these locks, then set this 3573 parameter to '1'. 3574 The default parameter value of '0' causes the kernel 3575 not to attempt recovery of lost locks. 3576 3577 nfs.send_implementation_id= 3578 [NFSv4.1] Send client implementation identification 3579 information in exchange_id requests. 3580 If zero, no implementation identification information 3581 will be sent. 3582 The default is to send the implementation identification 3583 information. 3584 3585 nfs4.layoutstats_timer= 3586 [NFSv4.2] Change the rate at which the kernel sends 3587 layoutstats to the pNFS metadata server. 3588 3589 Setting this to value to 0 causes the kernel to use 3590 whatever value is the default set by the layout 3591 driver. A non-zero value sets the minimum interval 3592 in seconds between layoutstats transmissions. 3593 3594 nfsd.inter_copy_offload_enable= 3595 [NFSv4.2] When set to 1, the server will support 3596 server-to-server copies for which this server is 3597 the destination of the copy. 3598 3599 nfsd.nfs4_disable_idmapping= 3600 [NFSv4] When set to the default of '1', the NFSv4 3601 server will return only numeric uids and gids to 3602 clients using auth_sys, and will accept numeric uids 3603 and gids from such clients. This is intended to ease 3604 migration from NFSv2/v3. 3605 3606 nfsd.nfsd4_ssc_umount_timeout= 3607 [NFSv4.2] When used as the destination of a 3608 server-to-server copy, knfsd temporarily mounts 3609 the source server. It caches the mount in case 3610 it will be needed again, and discards it if not 3611 used for the number of milliseconds specified by 3612 this parameter. 3613 3614 nfsaddrs= [NFS] Deprecated. Use ip= instead. 3615 See Documentation/admin-guide/nfs/nfsroot.rst. 3616 3617 nfsroot= [NFS] nfs root filesystem for disk-less boxes. 3618 See Documentation/admin-guide/nfs/nfsroot.rst. 3619 3620 nfsrootdebug [NFS] enable nfsroot debugging messages. 3621 See Documentation/admin-guide/nfs/nfsroot.rst. 3622 3623 nmi_backtrace.backtrace_idle [KNL] 3624 Dump stacks even of idle CPUs in response to an 3625 NMI stack-backtrace request. 3626 3627 nmi_debug= [KNL,SH] Specify one or more actions to take 3628 when a NMI is triggered. 3629 Format: [state][,regs][,debounce][,die] 3630 3631 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels 3632 Format: [panic,][nopanic,][num] 3633 Valid num: 0 or 1 3634 0 - turn hardlockup detector in nmi_watchdog off 3635 1 - turn hardlockup detector in nmi_watchdog on 3636 When panic is specified, panic when an NMI watchdog 3637 timeout occurs (or 'nopanic' to not panic on an NMI 3638 watchdog, if CONFIG_BOOTPARAM_HARDLOCKUP_PANIC is set) 3639 To disable both hard and soft lockup detectors, 3640 please see 'nowatchdog'. 3641 This is useful when you use a panic=... timeout and 3642 need the box quickly up again. 3643 3644 These settings can be accessed at runtime via 3645 the nmi_watchdog and hardlockup_panic sysctls. 3646 3647 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths 3648 emulation library even if a 387 maths coprocessor 3649 is present. 3650 3651 no4lvl [RISCV] Disable 4-level and 5-level paging modes. Forces 3652 kernel to use 3-level paging instead. 3653 3654 no5lvl [X86-64,RISCV] Disable 5-level paging mode. Forces 3655 kernel to use 4-level paging instead. 3656 3657 noaliencache [MM, NUMA, SLAB] Disables the allocation of alien 3658 caches in the slab allocator. Saves per-node memory, 3659 but will impact performance. 3660 3661 noalign [KNL,ARM] 3662 3663 noaltinstr [S390] Disables alternative instructions patching 3664 (CPU alternatives feature). 3665 3666 noapic [SMP,APIC] Tells the kernel to not make use of any 3667 IOAPICs that may be present in the system. 3668 3669 noautogroup Disable scheduler automatic task group creation. 3670 3671 nocache [ARM] 3672 3673 no_console_suspend 3674 [HW] Never suspend the console 3675 Disable suspending of consoles during suspend and 3676 hibernate operations. Once disabled, debugging 3677 messages can reach various consoles while the rest 3678 of the system is being put to sleep (ie, while 3679 debugging driver suspend/resume hooks). This may 3680 not work reliably with all consoles, but is known 3681 to work with serial and VGA consoles. 3682 To facilitate more flexible debugging, we also add 3683 console_suspend, a printk module parameter to control 3684 it. Users could use console_suspend (usually 3685 /sys/module/printk/parameters/console_suspend) to 3686 turn on/off it dynamically. 3687 3688 no_debug_objects 3689 [KNL] Disable object debugging 3690 3691 nodsp [SH] Disable hardware DSP at boot time. 3692 3693 noefi Disable EFI runtime services support. 3694 3695 no_entry_flush [PPC] Don't flush the L1-D cache when entering the kernel. 3696 3697 noexec [IA-64] 3698 3699 noexec32 [X86-64] 3700 This affects only 32-bit executables. 3701 noexec32=on: enable non-executable mappings (default) 3702 read doesn't imply executable mappings 3703 noexec32=off: disable non-executable mappings 3704 read implies executable mappings 3705 3706 no_file_caps Tells the kernel not to honor file capabilities. The 3707 only way then for a file to be executed with privilege 3708 is to be setuid root or executed by root. 3709 3710 nofpu [MIPS,SH] Disable hardware FPU at boot time. 3711 3712 nofsgsbase [X86] Disables FSGSBASE instructions. 3713 3714 nofxsr [BUGS=X86-32] Disables x86 floating point extended 3715 register save and restore. The kernel will only save 3716 legacy floating-point registers on task switch. 3717 3718 nohalt [IA-64] Tells the kernel not to use the power saving 3719 function PAL_HALT_LIGHT when idle. This increases 3720 power-consumption. On the positive side, it reduces 3721 interrupt wake-up latency, which may improve performance 3722 in certain environments such as networked servers or 3723 real-time systems. 3724 3725 no_hash_pointers 3726 Force pointers printed to the console or buffers to be 3727 unhashed. By default, when a pointer is printed via %p 3728 format string, that pointer is "hashed", i.e. obscured 3729 by hashing the pointer value. This is a security feature 3730 that hides actual kernel addresses from unprivileged 3731 users, but it also makes debugging the kernel more 3732 difficult since unequal pointers can no longer be 3733 compared. However, if this command-line option is 3734 specified, then all normal pointers will have their true 3735 value printed. This option should only be specified when 3736 debugging the kernel. Please do not use on production 3737 kernels. 3738 3739 nohibernate [HIBERNATION] Disable hibernation and resume. 3740 3741 nohlt [ARM,ARM64,MICROBLAZE,MIPS,SH] Forces the kernel to 3742 busy wait in do_idle() and not use the arch_cpu_idle() 3743 implementation; requires CONFIG_GENERIC_IDLE_POLL_SETUP 3744 to be effective. This is useful on platforms where the 3745 sleep(SH) or wfi(ARM,ARM64) instructions do not work 3746 correctly or when doing power measurements to evaluate 3747 the impact of the sleep instructions. This is also 3748 useful when using JTAG debugger. 3749 3750 nohugeiomap [KNL,X86,PPC,ARM64] Disable kernel huge I/O mappings. 3751 3752 nohugevmalloc [KNL,X86,PPC,ARM64] Disable kernel huge vmalloc mappings. 3753 3754 nohz= [KNL] Boottime enable/disable dynamic ticks 3755 Valid arguments: on, off 3756 Default: on 3757 3758 nohz_full= [KNL,BOOT,SMP,ISOL] 3759 The argument is a cpu list, as described above. 3760 In kernels built with CONFIG_NO_HZ_FULL=y, set 3761 the specified list of CPUs whose tick will be stopped 3762 whenever possible. The boot CPU will be forced outside 3763 the range to maintain the timekeeping. Any CPUs 3764 in this list will have their RCU callbacks offloaded, 3765 just as if they had also been called out in the 3766 rcu_nocbs= boot parameter. 3767 3768 Note that this argument takes precedence over 3769 the CONFIG_RCU_NOCB_CPU_DEFAULT_ALL option. 3770 3771 noinitrd [RAM] Tells the kernel not to load any configured 3772 initial RAM disk. 3773 3774 nointremap [X86-64, Intel-IOMMU] Do not enable interrupt 3775 remapping. 3776 [Deprecated - use intremap=off] 3777 3778 nointroute [IA-64] 3779 3780 noinvpcid [X86] Disable the INVPCID cpu feature. 3781 3782 noiotrap [SH] Disables trapped I/O port accesses. 3783 3784 noirqdebug [X86-32] Disables the code which attempts to detect and 3785 disable unhandled interrupt sources. 3786 3787 noisapnp [ISAPNP] Disables ISA PnP code. 3788 3789 nojitter [IA-64] Disables jitter checking for ITC timers. 3790 3791 nokaslr [KNL] 3792 When CONFIG_RANDOMIZE_BASE is set, this disables 3793 kernel and module base offset ASLR (Address Space 3794 Layout Randomization). 3795 3796 no-kvmapf [X86,KVM] Disable paravirtualized asynchronous page 3797 fault handling. 3798 3799 no-kvmclock [X86,KVM] Disable paravirtualized KVM clock driver 3800 3801 nolapic [X86-32,APIC] Do not enable or use the local APIC. 3802 3803 nolapic_timer [X86-32,APIC] Do not use the local APIC timer. 3804 3805 nomca [IA-64] Disable machine check abort handling 3806 3807 nomce [X86-32] Disable Machine Check Exception 3808 3809 nomfgpt [X86-32] Disable Multi-Function General Purpose 3810 Timer usage (for AMD Geode machines). 3811 3812 nomodeset Disable kernel modesetting. Most systems' firmware 3813 sets up a display mode and provides framebuffer memory 3814 for output. With nomodeset, DRM and fbdev drivers will 3815 not load if they could possibly displace the pre- 3816 initialized output. Only the system framebuffer will 3817 be available for use. The respective drivers will not 3818 perform display-mode changes or accelerated rendering. 3819 3820 Useful as error fallback, or for testing and debugging. 3821 3822 nomodule Disable module load 3823 3824 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to 3825 shutdown the other cpus. Instead use the REBOOT_VECTOR 3826 irq. 3827 3828 nopat [X86] Disable PAT (page attribute table extension of 3829 pagetables) support. 3830 3831 nopcid [X86-64] Disable the PCID cpu feature. 3832 3833 nopku [X86] Disable Memory Protection Keys CPU feature found 3834 in some Intel CPUs. 3835 3836 nopti [X86-64] 3837 Equivalent to pti=off 3838 3839 nopv= [X86,XEN,KVM,HYPER_V,VMWARE] 3840 Disables the PV optimizations forcing the guest to run 3841 as generic guest with no PV drivers. Currently support 3842 XEN HVM, KVM, HYPER_V and VMWARE guest. 3843 3844 nopvspin [X86,XEN,KVM] 3845 Disables the qspinlock slow path using PV optimizations 3846 which allow the hypervisor to 'idle' the guest on lock 3847 contention. 3848 3849 norandmaps Don't use address space randomization. Equivalent to 3850 echo 0 > /proc/sys/kernel/randomize_va_space 3851 3852 noreplace-smp [X86-32,SMP] Don't replace SMP instructions 3853 with UP alternatives 3854 3855 noresume [SWSUSP] Disables resume and restores original swap 3856 space. 3857 3858 nosbagart [IA-64] 3859 3860 no-scroll [VGA] Disables scrollback. 3861 This is required for the Braillex ib80-piezo Braille 3862 reader made by F.H. Papenmeier (Germany). 3863 3864 nosgx [X86-64,SGX] Disables Intel SGX kernel support. 3865 3866 nosmap [PPC] 3867 Disable SMAP (Supervisor Mode Access Prevention) 3868 even if it is supported by processor. 3869 3870 nosmep [PPC64s] 3871 Disable SMEP (Supervisor Mode Execution Prevention) 3872 even if it is supported by processor. 3873 3874 nosmp [SMP] Tells an SMP kernel to act as a UP kernel, 3875 and disable the IO APIC. legacy for "maxcpus=0". 3876 3877 nosmt [KNL,MIPS,S390] Disable symmetric multithreading (SMT). 3878 Equivalent to smt=1. 3879 3880 [KNL,X86] Disable symmetric multithreading (SMT). 3881 nosmt=force: Force disable SMT, cannot be undone 3882 via the sysfs control file. 3883 3884 nosoftlockup [KNL] Disable the soft-lockup detector. 3885 3886 nospec_store_bypass_disable 3887 [HW] Disable all mitigations for the Speculative Store Bypass vulnerability 3888 3889 nospectre_bhb [ARM64] Disable all mitigations for Spectre-BHB (branch 3890 history injection) vulnerability. System may allow data leaks 3891 with this option. 3892 3893 nospectre_v1 [X86,PPC] Disable mitigations for Spectre Variant 1 3894 (bounds check bypass). With this option data leaks are 3895 possible in the system. 3896 3897 nospectre_v2 [X86,PPC_E500,ARM64] Disable all mitigations for 3898 the Spectre variant 2 (indirect branch prediction) 3899 vulnerability. System may allow data leaks with this 3900 option. 3901 3902 no-steal-acc [X86,PV_OPS,ARM64,PPC/PSERIES] Disable paravirtualized 3903 steal time accounting. steal time is computed, but 3904 won't influence scheduler behaviour 3905 3906 nosync [HW,M68K] Disables sync negotiation for all devices. 3907 3908 no_timer_check [X86,APIC] Disables the code which tests for 3909 broken timer IRQ sources. 3910 3911 no_uaccess_flush 3912 [PPC] Don't flush the L1-D cache after accessing user data. 3913 3914 novmcoredd [KNL,KDUMP] 3915 Disable device dump. Device dump allows drivers to 3916 append dump data to vmcore so you can collect driver 3917 specified debug info. Drivers can append the data 3918 without any limit and this data is stored in memory, 3919 so this may cause significant memory stress. Disabling 3920 device dump can help save memory but the driver debug 3921 data will be no longer available. This parameter 3922 is only available when CONFIG_PROC_VMCORE_DEVICE_DUMP 3923 is set. 3924 3925 no-vmw-sched-clock 3926 [X86,PV_OPS] Disable paravirtualized VMware scheduler 3927 clock and use the default one. 3928 3929 nowatchdog [KNL] Disable both lockup detectors, i.e. 3930 soft-lockup and NMI watchdog (hard-lockup). 3931 3932 nowb [ARM] 3933 3934 nox2apic [X86-64,APIC] Do not enable x2APIC mode. 3935 3936 NOTE: this parameter will be ignored on systems with the 3937 LEGACY_XAPIC_DISABLED bit set in the 3938 IA32_XAPIC_DISABLE_STATUS MSR. 3939 3940 noxsave [BUGS=X86] Disables x86 extended register state save 3941 and restore using xsave. The kernel will fallback to 3942 enabling legacy floating-point and sse state. 3943 3944 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended 3945 register states. The kernel will fall back to use 3946 xsave to save the states. By using this parameter, 3947 performance of saving the states is degraded because 3948 xsave doesn't support modified optimization while 3949 xsaveopt supports it on xsaveopt enabled systems. 3950 3951 noxsaves [X86] Disables xsaves and xrstors used in saving and 3952 restoring x86 extended register state in compacted 3953 form of xsave area. The kernel will fall back to use 3954 xsaveopt and xrstor to save and restore the states 3955 in standard form of xsave area. By using this 3956 parameter, xsave area per process might occupy more 3957 memory on xsaves enabled systems. 3958 3959 nps_mtm_hs_ctr= [KNL,ARC] 3960 This parameter sets the maximum duration, in 3961 cycles, each HW thread of the CTOP can run 3962 without interruptions, before HW switches it. 3963 The actual maximum duration is 16 times this 3964 parameter's value. 3965 Format: integer between 1 and 255 3966 Default: 255 3967 3968 nptcg= [IA-64] Override max number of concurrent global TLB 3969 purges which is reported from either PAL_VM_SUMMARY or 3970 SAL PALO. 3971 3972 nr_cpus= [SMP] Maximum number of processors that an SMP kernel 3973 could support. nr_cpus=n : n >= 1 limits the kernel to 3974 support 'n' processors. It could be larger than the 3975 number of already plugged CPU during bootup, later in 3976 runtime you can physically add extra cpu until it reaches 3977 n. So during boot up some boot time memory for per-cpu 3978 variables need be pre-allocated for later physical cpu 3979 hot plugging. 3980 3981 nr_uarts= [SERIAL] maximum number of UARTs to be registered. 3982 3983 numa=off [KNL, ARM64, PPC, RISCV, SPARC, X86] Disable NUMA, Only 3984 set up a single NUMA node spanning all memory. 3985 3986 numa_balancing= [KNL,ARM64,PPC,RISCV,S390,X86] Enable or disable automatic 3987 NUMA balancing. 3988 Allowed values are enable and disable 3989 3990 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA. 3991 'node', 'default' can be specified 3992 This can be set from sysctl after boot. 3993 See Documentation/admin-guide/sysctl/vm.rst for details. 3994 3995 ohci1394_dma=early [HW] enable debugging via the ohci1394 driver. 3996 See Documentation/core-api/debugging-via-ohci1394.rst for more 3997 info. 3998 3999 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands 4000 Rather than timing out after 20 ms if an EC 4001 command is not properly ACKed, override the length 4002 of the timeout. We have interrupts disabled while 4003 waiting for the ACK, so if this is set too high 4004 interrupts *may* be lost! 4005 4006 omap_mux= [OMAP] Override bootloader pin multiplexing. 4007 Format: <mux_mode0.mode_name=value>... 4008 For example, to override I2C bus2: 4009 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100 4010 4011 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration 4012 4013 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock] 4014 4015 boundary - index of last SLC block on Flex-OneNAND. 4016 The remaining blocks are configured as MLC blocks. 4017 lock - Configure if Flex-OneNAND boundary should be locked. 4018 Once locked, the boundary cannot be changed. 4019 1 indicates lock status, 0 indicates unlock status. 4020 4021 oops=panic Always panic on oopses. Default is to just kill the 4022 process, but there is a small probability of 4023 deadlocking the machine. 4024 This will also cause panics on machine check exceptions. 4025 Useful together with panic=30 to trigger a reboot. 4026 4027 page_alloc.shuffle= 4028 [KNL] Boolean flag to control whether the page allocator 4029 should randomize its free lists. The randomization may 4030 be automatically enabled if the kernel detects it is 4031 running on a platform with a direct-mapped memory-side 4032 cache, and this parameter can be used to 4033 override/disable that behavior. The state of the flag 4034 can be read from sysfs at: 4035 /sys/module/page_alloc/parameters/shuffle. 4036 4037 page_owner= [KNL] Boot-time page_owner enabling option. 4038 Storage of the information about who allocated 4039 each page is disabled in default. With this switch, 4040 we can turn it on. 4041 on: enable the feature 4042 4043 page_poison= [KNL] Boot-time parameter changing the state of 4044 poisoning on the buddy allocator, available with 4045 CONFIG_PAGE_POISONING=y. 4046 off: turn off poisoning (default) 4047 on: turn on poisoning 4048 4049 page_reporting.page_reporting_order= 4050 [KNL] Minimal page reporting order 4051 Format: <integer> 4052 Adjust the minimal page reporting order. The page 4053 reporting is disabled when it exceeds MAX_ORDER. 4054 4055 panic= [KNL] Kernel behaviour on panic: delay <timeout> 4056 timeout > 0: seconds before rebooting 4057 timeout = 0: wait forever 4058 timeout < 0: reboot immediately 4059 Format: <timeout> 4060 4061 panic_print= Bitmask for printing system info when panic happens. 4062 User can chose combination of the following bits: 4063 bit 0: print all tasks info 4064 bit 1: print system memory info 4065 bit 2: print timer info 4066 bit 3: print locks info if CONFIG_LOCKDEP is on 4067 bit 4: print ftrace buffer 4068 bit 5: print all printk messages in buffer 4069 bit 6: print all CPUs backtrace (if available in the arch) 4070 *Be aware* that this option may print a _lot_ of lines, 4071 so there are risks of losing older messages in the log. 4072 Use this option carefully, maybe worth to setup a 4073 bigger log buffer with "log_buf_len" along with this. 4074 4075 panic_on_taint= Bitmask for conditionally calling panic() in add_taint() 4076 Format: <hex>[,nousertaint] 4077 Hexadecimal bitmask representing the set of TAINT flags 4078 that will cause the kernel to panic when add_taint() is 4079 called with any of the flags in this set. 4080 The optional switch "nousertaint" can be utilized to 4081 prevent userspace forced crashes by writing to sysctl 4082 /proc/sys/kernel/tainted any flagset matching with the 4083 bitmask set on panic_on_taint. 4084 See Documentation/admin-guide/tainted-kernels.rst for 4085 extra details on the taint flags that users can pick 4086 to compose the bitmask to assign to panic_on_taint. 4087 4088 panic_on_warn=1 panic() instead of WARN(). Useful to cause kdump 4089 on a WARN(). 4090 4091 parkbd.port= [HW] Parallel port number the keyboard adapter is 4092 connected to, default is 0. 4093 Format: <parport#> 4094 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation, 4095 0 for XT, 1 for AT (default is AT). 4096 Format: <mode> 4097 4098 parport= [HW,PPT] Specify parallel ports. 0 disables. 4099 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] } 4100 Use 'auto' to force the driver to use any 4101 IRQ/DMA settings detected (the default is to 4102 ignore detected IRQ/DMA settings because of 4103 possible conflicts). You can specify the base 4104 address, IRQ, and DMA settings; IRQ and DMA 4105 should be numbers, or 'auto' (for using detected 4106 settings on that particular port), or 'nofifo' 4107 (to avoid using a FIFO even if it is detected). 4108 Parallel ports are assigned in the order they 4109 are specified on the command line, starting 4110 with parport0. 4111 4112 parport_init_mode= [HW,PPT] 4113 Configure VIA parallel port to operate in 4114 a specific mode. This is necessary on Pegasos 4115 computer where firmware has no options for setting 4116 up parallel port mode and sets it to spp. 4117 Currently this function knows 686a and 8231 chips. 4118 Format: [spp|ps2|epp|ecp|ecpepp] 4119 4120 pata_legacy.all= [HW,LIBATA] 4121 Format: <int> 4122 Set to non-zero to probe primary and secondary ISA 4123 port ranges on PCI systems where no PCI PATA device 4124 has been found at either range. Disabled by default. 4125 4126 pata_legacy.autospeed= [HW,LIBATA] 4127 Format: <int> 4128 Set to non-zero if a chip is present that snoops speed 4129 changes. Disabled by default. 4130 4131 pata_legacy.ht6560a= [HW,LIBATA] 4132 Format: <int> 4133 Set to 1, 2, or 3 for HT 6560A on the primary channel, 4134 the secondary channel, or both channels respectively. 4135 Disabled by default. 4136 4137 pata_legacy.ht6560b= [HW,LIBATA] 4138 Format: <int> 4139 Set to 1, 2, or 3 for HT 6560B on the primary channel, 4140 the secondary channel, or both channels respectively. 4141 Disabled by default. 4142 4143 pata_legacy.iordy_mask= [HW,LIBATA] 4144 Format: <int> 4145 IORDY enable mask. Set individual bits to allow IORDY 4146 for the respective channel. Bit 0 is for the first 4147 legacy channel handled by this driver, bit 1 is for 4148 the second channel, and so on. The sequence will often 4149 correspond to the primary legacy channel, the secondary 4150 legacy channel, and so on, but the handling of a PCI 4151 bus and the use of other driver options may interfere 4152 with the sequence. By default IORDY is allowed across 4153 all channels. 4154 4155 pata_legacy.opti82c46x= [HW,LIBATA] 4156 Format: <int> 4157 Set to 1, 2, or 3 for Opti 82c611A on the primary 4158 channel, the secondary channel, or both channels 4159 respectively. Disabled by default. 4160 4161 pata_legacy.opti82c611a= [HW,LIBATA] 4162 Format: <int> 4163 Set to 1, 2, or 3 for Opti 82c465MV on the primary 4164 channel, the secondary channel, or both channels 4165 respectively. Disabled by default. 4166 4167 pata_legacy.pio_mask= [HW,LIBATA] 4168 Format: <int> 4169 PIO mode mask for autospeed devices. Set individual 4170 bits to allow the use of the respective PIO modes. 4171 Bit 0 is for mode 0, bit 1 is for mode 1, and so on. 4172 All modes allowed by default. 4173 4174 pata_legacy.probe_all= [HW,LIBATA] 4175 Format: <int> 4176 Set to non-zero to probe tertiary and further ISA 4177 port ranges on PCI systems. Disabled by default. 4178 4179 pata_legacy.probe_mask= [HW,LIBATA] 4180 Format: <int> 4181 Probe mask for legacy ISA PATA ports. Depending on 4182 platform configuration and the use of other driver 4183 options up to 6 legacy ports are supported: 0x1f0, 4184 0x170, 0x1e8, 0x168, 0x1e0, 0x160, however probing 4185 of individual ports can be disabled by setting the 4186 corresponding bits in the mask to 1. Bit 0 is for 4187 the first port in the list above (0x1f0), and so on. 4188 By default all supported ports are probed. 4189 4190 pata_legacy.qdi= [HW,LIBATA] 4191 Format: <int> 4192 Set to non-zero to probe QDI controllers. By default 4193 set to 1 if CONFIG_PATA_QDI_MODULE, 0 otherwise. 4194 4195 pata_legacy.winbond= [HW,LIBATA] 4196 Format: <int> 4197 Set to non-zero to probe Winbond controllers. Use 4198 the standard I/O port (0x130) if 1, otherwise the 4199 value given is the I/O port to use (typically 0x1b0). 4200 By default set to 1 if CONFIG_PATA_WINBOND_VLB_MODULE, 4201 0 otherwise. 4202 4203 pata_platform.pio_mask= [HW,LIBATA] 4204 Format: <int> 4205 Supported PIO mode mask. Set individual bits to allow 4206 the use of the respective PIO modes. Bit 0 is for 4207 mode 0, bit 1 is for mode 1, and so on. Mode 0 only 4208 allowed by default. 4209 4210 pause_on_oops= 4211 Halt all CPUs after the first oops has been printed for 4212 the specified number of seconds. This is to be used if 4213 your oopses keep scrolling off the screen. 4214 4215 pcbit= [HW,ISDN] 4216 4217 pci=option[,option...] [PCI] various PCI subsystem options. 4218 4219 Some options herein operate on a specific device 4220 or a set of devices (<pci_dev>). These are 4221 specified in one of the following formats: 4222 4223 [<domain>:]<bus>:<dev>.<func>[/<dev>.<func>]* 4224 pci:<vendor>:<device>[:<subvendor>:<subdevice>] 4225 4226 Note: the first format specifies a PCI 4227 bus/device/function address which may change 4228 if new hardware is inserted, if motherboard 4229 firmware changes, or due to changes caused 4230 by other kernel parameters. If the 4231 domain is left unspecified, it is 4232 taken to be zero. Optionally, a path 4233 to a device through multiple device/function 4234 addresses can be specified after the base 4235 address (this is more robust against 4236 renumbering issues). The second format 4237 selects devices using IDs from the 4238 configuration space which may match multiple 4239 devices in the system. 4240 4241 earlydump dump PCI config space before the kernel 4242 changes anything 4243 off [X86] don't probe for the PCI bus 4244 bios [X86-32] force use of PCI BIOS, don't access 4245 the hardware directly. Use this if your machine 4246 has a non-standard PCI host bridge. 4247 nobios [X86-32] disallow use of PCI BIOS, only direct 4248 hardware access methods are allowed. Use this 4249 if you experience crashes upon bootup and you 4250 suspect they are caused by the BIOS. 4251 conf1 [X86] Force use of PCI Configuration Access 4252 Mechanism 1 (config address in IO port 0xCF8, 4253 data in IO port 0xCFC, both 32-bit). 4254 conf2 [X86] Force use of PCI Configuration Access 4255 Mechanism 2 (IO port 0xCF8 is an 8-bit port for 4256 the function, IO port 0xCFA, also 8-bit, sets 4257 bus number. The config space is then accessed 4258 through ports 0xC000-0xCFFF). 4259 See http://wiki.osdev.org/PCI for more info 4260 on the configuration access mechanisms. 4261 noaer [PCIE] If the PCIEAER kernel config parameter is 4262 enabled, this kernel boot option can be used to 4263 disable the use of PCIE advanced error reporting. 4264 nodomains [PCI] Disable support for multiple PCI 4265 root domains (aka PCI segments, in ACPI-speak). 4266 nommconf [X86] Disable use of MMCONFIG for PCI 4267 Configuration 4268 check_enable_amd_mmconf [X86] check for and enable 4269 properly configured MMIO access to PCI 4270 config space on AMD family 10h CPU 4271 nomsi [MSI] If the PCI_MSI kernel config parameter is 4272 enabled, this kernel boot option can be used to 4273 disable the use of MSI interrupts system-wide. 4274 noioapicquirk [APIC] Disable all boot interrupt quirks. 4275 Safety option to keep boot IRQs enabled. This 4276 should never be necessary. 4277 ioapicreroute [APIC] Enable rerouting of boot IRQs to the 4278 primary IO-APIC for bridges that cannot disable 4279 boot IRQs. This fixes a source of spurious IRQs 4280 when the system masks IRQs. 4281 noioapicreroute [APIC] Disable workaround that uses the 4282 boot IRQ equivalent of an IRQ that connects to 4283 a chipset where boot IRQs cannot be disabled. 4284 The opposite of ioapicreroute. 4285 biosirq [X86-32] Use PCI BIOS calls to get the interrupt 4286 routing table. These calls are known to be buggy 4287 on several machines and they hang the machine 4288 when used, but on other computers it's the only 4289 way to get the interrupt routing table. Try 4290 this option if the kernel is unable to allocate 4291 IRQs or discover secondary PCI buses on your 4292 motherboard. 4293 rom [X86] Assign address space to expansion ROMs. 4294 Use with caution as certain devices share 4295 address decoders between ROMs and other 4296 resources. 4297 norom [X86] Do not assign address space to 4298 expansion ROMs that do not already have 4299 BIOS assigned address ranges. 4300 nobar [X86] Do not assign address space to the 4301 BARs that weren't assigned by the BIOS. 4302 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be 4303 assigned automatically to PCI devices. You can 4304 make the kernel exclude IRQs of your ISA cards 4305 this way. 4306 pirqaddr=0xAAAAA [X86] Specify the physical address 4307 of the PIRQ table (normally generated 4308 by the BIOS) if it is outside the 4309 F0000h-100000h range. 4310 lastbus=N [X86] Scan all buses thru bus #N. Can be 4311 useful if the kernel is unable to find your 4312 secondary buses and you want to tell it 4313 explicitly which ones they are. 4314 assign-busses [X86] Always assign all PCI bus 4315 numbers ourselves, overriding 4316 whatever the firmware may have done. 4317 usepirqmask [X86] Honor the possible IRQ mask stored 4318 in the BIOS $PIR table. This is needed on 4319 some systems with broken BIOSes, notably 4320 some HP Pavilion N5400 and Omnibook XE3 4321 notebooks. This will have no effect if ACPI 4322 IRQ routing is enabled. 4323 noacpi [X86] Do not use ACPI for IRQ routing 4324 or for PCI scanning. 4325 use_crs [X86] Use PCI host bridge window information 4326 from ACPI. On BIOSes from 2008 or later, this 4327 is enabled by default. If you need to use this, 4328 please report a bug. 4329 nocrs [X86] Ignore PCI host bridge windows from ACPI. 4330 If you need to use this, please report a bug. 4331 use_e820 [X86] Use E820 reservations to exclude parts of 4332 PCI host bridge windows. This is a workaround 4333 for BIOS defects in host bridge _CRS methods. 4334 If you need to use this, please report a bug to 4335 <linux-pci@vger.kernel.org>. 4336 no_e820 [X86] Ignore E820 reservations for PCI host 4337 bridge windows. This is the default on modern 4338 hardware. If you need to use this, please report 4339 a bug to <linux-pci@vger.kernel.org>. 4340 routeirq Do IRQ routing for all PCI devices. 4341 This is normally done in pci_enable_device(), 4342 so this option is a temporary workaround 4343 for broken drivers that don't call it. 4344 skip_isa_align [X86] do not align io start addr, so can 4345 handle more pci cards 4346 noearly [X86] Don't do any early type 1 scanning. 4347 This might help on some broken boards which 4348 machine check when some devices' config space 4349 is read. But various workarounds are disabled 4350 and some IOMMU drivers will not work. 4351 bfsort Sort PCI devices into breadth-first order. 4352 This sorting is done to get a device 4353 order compatible with older (<= 2.4) kernels. 4354 nobfsort Don't sort PCI devices into breadth-first order. 4355 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size) 4356 tuning and use the BIOS-configured MPS defaults. 4357 pcie_bus_safe Set every device's MPS to the largest value 4358 supported by all devices below the root complex. 4359 pcie_bus_perf Set device MPS to the largest allowable MPS 4360 based on its parent bus. Also set MRRS (Max 4361 Read Request Size) to the largest supported 4362 value (no larger than the MPS that the device 4363 or bus can support) for best performance. 4364 pcie_bus_peer2peer Set every device's MPS to 128B, which 4365 every device is guaranteed to support. This 4366 configuration allows peer-to-peer DMA between 4367 any pair of devices, possibly at the cost of 4368 reduced performance. This also guarantees 4369 that hot-added devices will work. 4370 cbiosize=nn[KMG] The fixed amount of bus space which is 4371 reserved for the CardBus bridge's IO window. 4372 The default value is 256 bytes. 4373 cbmemsize=nn[KMG] The fixed amount of bus space which is 4374 reserved for the CardBus bridge's memory 4375 window. The default value is 64 megabytes. 4376 resource_alignment= 4377 Format: 4378 [<order of align>@]<pci_dev>[; ...] 4379 Specifies alignment and device to reassign 4380 aligned memory resources. How to 4381 specify the device is described above. 4382 If <order of align> is not specified, 4383 PAGE_SIZE is used as alignment. 4384 A PCI-PCI bridge can be specified if resource 4385 windows need to be expanded. 4386 To specify the alignment for several 4387 instances of a device, the PCI vendor, 4388 device, subvendor, and subdevice may be 4389 specified, e.g., 12@pci:8086:9c22:103c:198f 4390 for 4096-byte alignment. 4391 ecrc= Enable/disable PCIe ECRC (transaction layer 4392 end-to-end CRC checking). Only effective if 4393 OS has native AER control (either granted by 4394 ACPI _OSC or forced via "pcie_ports=native") 4395 bios: Use BIOS/firmware settings. This is the 4396 the default. 4397 off: Turn ECRC off 4398 on: Turn ECRC on. 4399 hpiosize=nn[KMG] The fixed amount of bus space which is 4400 reserved for hotplug bridge's IO window. 4401 Default size is 256 bytes. 4402 hpmmiosize=nn[KMG] The fixed amount of bus space which is 4403 reserved for hotplug bridge's MMIO window. 4404 Default size is 2 megabytes. 4405 hpmmioprefsize=nn[KMG] The fixed amount of bus space which is 4406 reserved for hotplug bridge's MMIO_PREF window. 4407 Default size is 2 megabytes. 4408 hpmemsize=nn[KMG] The fixed amount of bus space which is 4409 reserved for hotplug bridge's MMIO and 4410 MMIO_PREF window. 4411 Default size is 2 megabytes. 4412 hpbussize=nn The minimum amount of additional bus numbers 4413 reserved for buses below a hotplug bridge. 4414 Default is 1. 4415 realloc= Enable/disable reallocating PCI bridge resources 4416 if allocations done by BIOS are too small to 4417 accommodate resources required by all child 4418 devices. 4419 off: Turn realloc off 4420 on: Turn realloc on 4421 realloc same as realloc=on 4422 noari do not use PCIe ARI. 4423 noats [PCIE, Intel-IOMMU, AMD-IOMMU] 4424 do not use PCIe ATS (and IOMMU device IOTLB). 4425 pcie_scan_all Scan all possible PCIe devices. Otherwise we 4426 only look for one device below a PCIe downstream 4427 port. 4428 big_root_window Try to add a big 64bit memory window to the PCIe 4429 root complex on AMD CPUs. Some GFX hardware 4430 can resize a BAR to allow access to all VRAM. 4431 Adding the window is slightly risky (it may 4432 conflict with unreported devices), so this 4433 taints the kernel. 4434 disable_acs_redir=<pci_dev>[; ...] 4435 Specify one or more PCI devices (in the format 4436 specified above) separated by semicolons. 4437 Each device specified will have the PCI ACS 4438 redirect capabilities forced off which will 4439 allow P2P traffic between devices through 4440 bridges without forcing it upstream. Note: 4441 this removes isolation between devices and 4442 may put more devices in an IOMMU group. 4443 force_floating [S390] Force usage of floating interrupts. 4444 nomio [S390] Do not use MIO instructions. 4445 norid [S390] ignore the RID field and force use of 4446 one PCI domain per PCI function 4447 4448 pcie_aspm= [PCIE] Forcibly enable or disable PCIe Active State Power 4449 Management. 4450 off Disable ASPM. 4451 force Enable ASPM even on devices that claim not to support it. 4452 WARNING: Forcing ASPM on may cause system lockups. 4453 4454 pcie_ports= [PCIE] PCIe port services handling: 4455 native Use native PCIe services (PME, AER, DPC, PCIe hotplug) 4456 even if the platform doesn't give the OS permission to 4457 use them. This may cause conflicts if the platform 4458 also tries to use these services. 4459 dpc-native Use native PCIe service for DPC only. May 4460 cause conflicts if firmware uses AER or DPC. 4461 compat Disable native PCIe services (PME, AER, DPC, PCIe 4462 hotplug). 4463 4464 pcie_port_pm= [PCIE] PCIe port power management handling: 4465 off Disable power management of all PCIe ports 4466 force Forcibly enable power management of all PCIe ports 4467 4468 pcie_pme= [PCIE,PM] Native PCIe PME signaling options: 4469 nomsi Do not use MSI for native PCIe PME signaling (this makes 4470 all PCIe root ports use INTx for all services). 4471 4472 pcmv= [HW,PCMCIA] BadgePAD 4 4473 4474 pd_ignore_unused 4475 [PM] 4476 Keep all power-domains already enabled by bootloader on, 4477 even if no driver has claimed them. This is useful 4478 for debug and development, but should not be 4479 needed on a platform with proper driver support. 4480 4481 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at 4482 boot time. 4483 Format: { 0 | 1 } 4484 See arch/parisc/kernel/pdc_chassis.c 4485 4486 percpu_alloc= Select which percpu first chunk allocator to use. 4487 Currently supported values are "embed" and "page". 4488 Archs may support subset or none of the selections. 4489 See comments in mm/percpu.c for details on each 4490 allocator. This parameter is primarily for debugging 4491 and performance comparison. 4492 4493 pirq= [SMP,APIC] Manual mp-table setup 4494 See Documentation/arch/x86/i386/IO-APIC.rst. 4495 4496 plip= [PPT,NET] Parallel port network link 4497 Format: { parport<nr> | timid | 0 } 4498 See also Documentation/admin-guide/parport.rst. 4499 4500 pmtmr= [X86] Manual setup of pmtmr I/O Port. 4501 Override pmtimer IOPort with a hex value. 4502 e.g. pmtmr=0x508 4503 4504 pmu_override= [PPC] Override the PMU. 4505 This option takes over the PMU facility, so it is no 4506 longer usable by perf. Setting this option starts the 4507 PMU counters by setting MMCR0 to 0 (the FC bit is 4508 cleared). If a number is given, then MMCR1 is set to 4509 that number, otherwise (e.g., 'pmu_override=on'), MMCR1 4510 remains 0. 4511 4512 pm_debug_messages [SUSPEND,KNL] 4513 Enable suspend/resume debug messages during boot up. 4514 4515 pnp.debug=1 [PNP] 4516 Enable PNP debug messages (depends on the 4517 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time 4518 via /sys/module/pnp/parameters/debug. We always show 4519 current resource usage; turning this on also shows 4520 possible settings and some assignment information. 4521 4522 pnpacpi= [ACPI] 4523 { off } 4524 4525 pnpbios= [ISAPNP] 4526 { on | off | curr | res | no-curr | no-res } 4527 4528 pnp_reserve_irq= 4529 [ISAPNP] Exclude IRQs for the autoconfiguration 4530 4531 pnp_reserve_dma= 4532 [ISAPNP] Exclude DMAs for the autoconfiguration 4533 4534 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration 4535 Ranges are in pairs (I/O port base and size). 4536 4537 pnp_reserve_mem= 4538 [ISAPNP] Exclude memory regions for the 4539 autoconfiguration. 4540 Ranges are in pairs (memory base and size). 4541 4542 ports= [IP_VS_FTP] IPVS ftp helper module 4543 Default is 21. 4544 Up to 8 (IP_VS_APP_MAX_PORTS) ports 4545 may be specified. 4546 Format: <port>,<port>.... 4547 4548 powersave=off [PPC] This option disables power saving features. 4549 It specifically disables cpuidle and sets the 4550 platform machine description specific power_save 4551 function to NULL. On Idle the CPU just reduces 4552 execution priority. 4553 4554 ppc_strict_facility_enable 4555 [PPC] This option catches any kernel floating point, 4556 Altivec, VSX and SPE outside of regions specifically 4557 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()). 4558 There is some performance impact when enabling this. 4559 4560 ppc_tm= [PPC] 4561 Format: {"off"} 4562 Disable Hardware Transactional Memory 4563 4564 preempt= [KNL] 4565 Select preemption mode if you have CONFIG_PREEMPT_DYNAMIC 4566 none - Limited to cond_resched() calls 4567 voluntary - Limited to cond_resched() and might_sleep() calls 4568 full - Any section that isn't explicitly preempt disabled 4569 can be preempted anytime. 4570 4571 print-fatal-signals= 4572 [KNL] debug: print fatal signals 4573 4574 If enabled, warn about various signal handling 4575 related application anomalies: too many signals, 4576 too many POSIX.1 timers, fatal signals causing a 4577 coredump - etc. 4578 4579 If you hit the warning due to signal overflow, 4580 you might want to try "ulimit -i unlimited". 4581 4582 default: off. 4583 4584 printk.always_kmsg_dump= 4585 Trigger kmsg_dump for cases other than kernel oops or 4586 panics 4587 Format: <bool> (1/Y/y=enable, 0/N/n=disable) 4588 default: disabled 4589 4590 printk.console_no_auto_verbose= 4591 Disable console loglevel raise on oops, panic 4592 or lockdep-detected issues (only if lock debug is on). 4593 With an exception to setups with low baudrate on 4594 serial console, keeping this 0 is a good choice 4595 in order to provide more debug information. 4596 Format: <bool> 4597 default: 0 (auto_verbose is enabled) 4598 4599 printk.devkmsg={on,off,ratelimit} 4600 Control writing to /dev/kmsg. 4601 on - unlimited logging to /dev/kmsg from userspace 4602 off - logging to /dev/kmsg disabled 4603 ratelimit - ratelimit the logging 4604 Default: ratelimit 4605 4606 printk.time= Show timing data prefixed to each printk message line 4607 Format: <bool> (1/Y/y=enable, 0/N/n=disable) 4608 4609 processor.max_cstate= [HW,ACPI] 4610 Limit processor to maximum C-state 4611 max_cstate=9 overrides any DMI blacklist limit. 4612 4613 processor.nocst [HW,ACPI] 4614 Ignore the _CST method to determine C-states, 4615 instead using the legacy FADT method 4616 4617 profile= [KNL] Enable kernel profiling via /proc/profile 4618 Format: [<profiletype>,]<number> 4619 Param: <profiletype>: "schedule", "sleep", or "kvm" 4620 [defaults to kernel profiling] 4621 Param: "schedule" - profile schedule points. 4622 Param: "sleep" - profile D-state sleeping (millisecs). 4623 Requires CONFIG_SCHEDSTATS 4624 Param: "kvm" - profile VM exits. 4625 Param: <number> - step/bucket size as a power of 2 for 4626 statistical time based profiling. 4627 4628 prompt_ramdisk= [RAM] [Deprecated] 4629 4630 prot_virt= [S390] enable hosting protected virtual machines 4631 isolated from the hypervisor (if hardware supports 4632 that). 4633 Format: <bool> 4634 4635 psi= [KNL] Enable or disable pressure stall information 4636 tracking. 4637 Format: <bool> 4638 4639 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to 4640 probe for; one of (bare|imps|exps|lifebook|any). 4641 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports 4642 per second. 4643 psmouse.resetafter= [HW,MOUSE] 4644 Try to reset the device after so many bad packets 4645 (0 = never). 4646 psmouse.resolution= 4647 [HW,MOUSE] Set desired mouse resolution, in dpi. 4648 psmouse.smartscroll= 4649 [HW,MOUSE] Controls Logitech smartscroll autorepeat. 4650 0 = disabled, 1 = enabled (default). 4651 4652 pstore.backend= Specify the name of the pstore backend to use 4653 4654 pti= [X86-64] Control Page Table Isolation of user and 4655 kernel address spaces. Disabling this feature 4656 removes hardening, but improves performance of 4657 system calls and interrupts. 4658 4659 on - unconditionally enable 4660 off - unconditionally disable 4661 auto - kernel detects whether your CPU model is 4662 vulnerable to issues that PTI mitigates 4663 4664 Not specifying this option is equivalent to pti=auto. 4665 4666 pty.legacy_count= 4667 [KNL] Number of legacy pty's. Overwrites compiled-in 4668 default number. 4669 4670 quiet [KNL] Disable most log messages 4671 4672 r128= [HW,DRM] 4673 4674 radix_hcall_invalidate=on [PPC/PSERIES] 4675 Disable RADIX GTSE feature and use hcall for TLB 4676 invalidate. 4677 4678 raid= [HW,RAID] 4679 See Documentation/admin-guide/md.rst. 4680 4681 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes 4682 See Documentation/admin-guide/blockdev/ramdisk.rst. 4683 4684 ramdisk_start= [RAM] RAM disk image start address 4685 4686 random.trust_cpu=off 4687 [KNL] Disable trusting the use of the CPU's 4688 random number generator (if available) to 4689 initialize the kernel's RNG. 4690 4691 random.trust_bootloader=off 4692 [KNL] Disable trusting the use of the a seed 4693 passed by the bootloader (if available) to 4694 initialize the kernel's RNG. 4695 4696 randomize_kstack_offset= 4697 [KNL] Enable or disable kernel stack offset 4698 randomization, which provides roughly 5 bits of 4699 entropy, frustrating memory corruption attacks 4700 that depend on stack address determinism or 4701 cross-syscall address exposures. This is only 4702 available on architectures that have defined 4703 CONFIG_HAVE_ARCH_RANDOMIZE_KSTACK_OFFSET. 4704 Format: <bool> (1/Y/y=enable, 0/N/n=disable) 4705 Default is CONFIG_RANDOMIZE_KSTACK_OFFSET_DEFAULT. 4706 4707 ras=option[,option,...] [KNL] RAS-specific options 4708 4709 cec_disable [X86] 4710 Disable the Correctable Errors Collector, 4711 see CONFIG_RAS_CEC help text. 4712 4713 rcu_nocbs[=cpu-list] 4714 [KNL] The optional argument is a cpu list, 4715 as described above. 4716 4717 In kernels built with CONFIG_RCU_NOCB_CPU=y, 4718 enable the no-callback CPU mode, which prevents 4719 such CPUs' callbacks from being invoked in 4720 softirq context. Invocation of such CPUs' RCU 4721 callbacks will instead be offloaded to "rcuox/N" 4722 kthreads created for that purpose, where "x" is 4723 "p" for RCU-preempt, "s" for RCU-sched, and "g" 4724 for the kthreads that mediate grace periods; and 4725 "N" is the CPU number. This reduces OS jitter on 4726 the offloaded CPUs, which can be useful for HPC 4727 and real-time workloads. It can also improve 4728 energy efficiency for asymmetric multiprocessors. 4729 4730 If a cpulist is passed as an argument, the specified 4731 list of CPUs is set to no-callback mode from boot. 4732 4733 Otherwise, if the '=' sign and the cpulist 4734 arguments are omitted, no CPU will be set to 4735 no-callback mode from boot but the mode may be 4736 toggled at runtime via cpusets. 4737 4738 Note that this argument takes precedence over 4739 the CONFIG_RCU_NOCB_CPU_DEFAULT_ALL option. 4740 4741 rcu_nocb_poll [KNL] 4742 Rather than requiring that offloaded CPUs 4743 (specified by rcu_nocbs= above) explicitly 4744 awaken the corresponding "rcuoN" kthreads, 4745 make these kthreads poll for callbacks. 4746 This improves the real-time response for the 4747 offloaded CPUs by relieving them of the need to 4748 wake up the corresponding kthread, but degrades 4749 energy efficiency by requiring that the kthreads 4750 periodically wake up to do the polling. 4751 4752 rcutree.blimit= [KNL] 4753 Set maximum number of finished RCU callbacks to 4754 process in one batch. 4755 4756 rcutree.dump_tree= [KNL] 4757 Dump the structure of the rcu_node combining tree 4758 out at early boot. This is used for diagnostic 4759 purposes, to verify correct tree setup. 4760 4761 rcutree.gp_cleanup_delay= [KNL] 4762 Set the number of jiffies to delay each step of 4763 RCU grace-period cleanup. 4764 4765 rcutree.gp_init_delay= [KNL] 4766 Set the number of jiffies to delay each step of 4767 RCU grace-period initialization. 4768 4769 rcutree.gp_preinit_delay= [KNL] 4770 Set the number of jiffies to delay each step of 4771 RCU grace-period pre-initialization, that is, 4772 the propagation of recent CPU-hotplug changes up 4773 the rcu_node combining tree. 4774 4775 rcutree.jiffies_till_first_fqs= [KNL] 4776 Set delay from grace-period initialization to 4777 first attempt to force quiescent states. 4778 Units are jiffies, minimum value is zero, 4779 and maximum value is HZ. 4780 4781 rcutree.jiffies_till_next_fqs= [KNL] 4782 Set delay between subsequent attempts to force 4783 quiescent states. Units are jiffies, minimum 4784 value is one, and maximum value is HZ. 4785 4786 rcutree.jiffies_till_sched_qs= [KNL] 4787 Set required age in jiffies for a 4788 given grace period before RCU starts 4789 soliciting quiescent-state help from 4790 rcu_note_context_switch() and cond_resched(). 4791 If not specified, the kernel will calculate 4792 a value based on the most recent settings 4793 of rcutree.jiffies_till_first_fqs 4794 and rcutree.jiffies_till_next_fqs. 4795 This calculated value may be viewed in 4796 rcutree.jiffies_to_sched_qs. Any attempt to set 4797 rcutree.jiffies_to_sched_qs will be cheerfully 4798 overwritten. 4799 4800 rcutree.kthread_prio= [KNL,BOOT] 4801 Set the SCHED_FIFO priority of the RCU per-CPU 4802 kthreads (rcuc/N). This value is also used for 4803 the priority of the RCU boost threads (rcub/N) 4804 and for the RCU grace-period kthreads (rcu_bh, 4805 rcu_preempt, and rcu_sched). If RCU_BOOST is 4806 set, valid values are 1-99 and the default is 1 4807 (the least-favored priority). Otherwise, when 4808 RCU_BOOST is not set, valid values are 0-99 and 4809 the default is zero (non-realtime operation). 4810 When RCU_NOCB_CPU is set, also adjust the 4811 priority of NOCB callback kthreads. 4812 4813 rcutree.nocb_nobypass_lim_per_jiffy= [KNL] 4814 On callback-offloaded (rcu_nocbs) CPUs, 4815 RCU reduces the lock contention that would 4816 otherwise be caused by callback floods through 4817 use of the ->nocb_bypass list. However, in the 4818 common non-flooded case, RCU queues directly to 4819 the main ->cblist in order to avoid the extra 4820 overhead of the ->nocb_bypass list and its lock. 4821 But if there are too many callbacks queued during 4822 a single jiffy, RCU pre-queues the callbacks into 4823 the ->nocb_bypass queue. The definition of "too 4824 many" is supplied by this kernel boot parameter. 4825 4826 rcutree.qhimark= [KNL] 4827 Set threshold of queued RCU callbacks beyond which 4828 batch limiting is disabled. 4829 4830 rcutree.qlowmark= [KNL] 4831 Set threshold of queued RCU callbacks below which 4832 batch limiting is re-enabled. 4833 4834 rcutree.qovld= [KNL] 4835 Set threshold of queued RCU callbacks beyond which 4836 RCU's force-quiescent-state scan will aggressively 4837 enlist help from cond_resched() and sched IPIs to 4838 help CPUs more quickly reach quiescent states. 4839 Set to less than zero to make this be set based 4840 on rcutree.qhimark at boot time and to zero to 4841 disable more aggressive help enlistment. 4842 4843 rcutree.rcu_delay_page_cache_fill_msec= [KNL] 4844 Set the page-cache refill delay (in milliseconds) 4845 in response to low-memory conditions. The range 4846 of permitted values is in the range 0:100000. 4847 4848 rcutree.rcu_divisor= [KNL] 4849 Set the shift-right count to use to compute 4850 the callback-invocation batch limit bl from 4851 the number of callbacks queued on this CPU. 4852 The result will be bounded below by the value of 4853 the rcutree.blimit kernel parameter. Every bl 4854 callbacks, the softirq handler will exit in 4855 order to allow the CPU to do other work. 4856 4857 Please note that this callback-invocation batch 4858 limit applies only to non-offloaded callback 4859 invocation. Offloaded callbacks are instead 4860 invoked in the context of an rcuoc kthread, which 4861 scheduler will preempt as it does any other task. 4862 4863 rcutree.rcu_fanout_exact= [KNL] 4864 Disable autobalancing of the rcu_node combining 4865 tree. This is used by rcutorture, and might 4866 possibly be useful for architectures having high 4867 cache-to-cache transfer latencies. 4868 4869 rcutree.rcu_fanout_leaf= [KNL] 4870 Change the number of CPUs assigned to each 4871 leaf rcu_node structure. Useful for very 4872 large systems, which will choose the value 64, 4873 and for NUMA systems with large remote-access 4874 latencies, which will choose a value aligned 4875 with the appropriate hardware boundaries. 4876 4877 rcutree.rcu_min_cached_objs= [KNL] 4878 Minimum number of objects which are cached and 4879 maintained per one CPU. Object size is equal 4880 to PAGE_SIZE. The cache allows to reduce the 4881 pressure to page allocator, also it makes the 4882 whole algorithm to behave better in low memory 4883 condition. 4884 4885 rcutree.rcu_nocb_gp_stride= [KNL] 4886 Set the number of NOCB callback kthreads in 4887 each group, which defaults to the square root 4888 of the number of CPUs. Larger numbers reduce 4889 the wakeup overhead on the global grace-period 4890 kthread, but increases that same overhead on 4891 each group's NOCB grace-period kthread. 4892 4893 rcutree.rcu_kick_kthreads= [KNL] 4894 Cause the grace-period kthread to get an extra 4895 wake_up() if it sleeps three times longer than 4896 it should at force-quiescent-state time. 4897 This wake_up() will be accompanied by a 4898 WARN_ONCE() splat and an ftrace_dump(). 4899 4900 rcutree.rcu_resched_ns= [KNL] 4901 Limit the time spend invoking a batch of RCU 4902 callbacks to the specified number of nanoseconds. 4903 By default, this limit is checked only once 4904 every 32 callbacks in order to limit the pain 4905 inflicted by local_clock() overhead. 4906 4907 rcutree.rcu_unlock_delay= [KNL] 4908 In CONFIG_RCU_STRICT_GRACE_PERIOD=y kernels, 4909 this specifies an rcu_read_unlock()-time delay 4910 in microseconds. This defaults to zero. 4911 Larger delays increase the probability of 4912 catching RCU pointer leaks, that is, buggy use 4913 of RCU-protected pointers after the relevant 4914 rcu_read_unlock() has completed. 4915 4916 rcutree.sysrq_rcu= [KNL] 4917 Commandeer a sysrq key to dump out Tree RCU's 4918 rcu_node tree with an eye towards determining 4919 why a new grace period has not yet started. 4920 4921 rcutree.use_softirq= [KNL] 4922 If set to zero, move all RCU_SOFTIRQ processing to 4923 per-CPU rcuc kthreads. Defaults to a non-zero 4924 value, meaning that RCU_SOFTIRQ is used by default. 4925 Specify rcutree.use_softirq=0 to use rcuc kthreads. 4926 4927 But note that CONFIG_PREEMPT_RT=y kernels disable 4928 this kernel boot parameter, forcibly setting it 4929 to zero. 4930 4931 rcuscale.gp_async= [KNL] 4932 Measure performance of asynchronous 4933 grace-period primitives such as call_rcu(). 4934 4935 rcuscale.gp_async_max= [KNL] 4936 Specify the maximum number of outstanding 4937 callbacks per writer thread. When a writer 4938 thread exceeds this limit, it invokes the 4939 corresponding flavor of rcu_barrier() to allow 4940 previously posted callbacks to drain. 4941 4942 rcuscale.gp_exp= [KNL] 4943 Measure performance of expedited synchronous 4944 grace-period primitives. 4945 4946 rcuscale.holdoff= [KNL] 4947 Set test-start holdoff period. The purpose of 4948 this parameter is to delay the start of the 4949 test until boot completes in order to avoid 4950 interference. 4951 4952 rcuscale.kfree_rcu_test= [KNL] 4953 Set to measure performance of kfree_rcu() flooding. 4954 4955 rcuscale.kfree_rcu_test_double= [KNL] 4956 Test the double-argument variant of kfree_rcu(). 4957 If this parameter has the same value as 4958 rcuscale.kfree_rcu_test_single, both the single- 4959 and double-argument variants are tested. 4960 4961 rcuscale.kfree_rcu_test_single= [KNL] 4962 Test the single-argument variant of kfree_rcu(). 4963 If this parameter has the same value as 4964 rcuscale.kfree_rcu_test_double, both the single- 4965 and double-argument variants are tested. 4966 4967 rcuscale.kfree_nthreads= [KNL] 4968 The number of threads running loops of kfree_rcu(). 4969 4970 rcuscale.kfree_alloc_num= [KNL] 4971 Number of allocations and frees done in an iteration. 4972 4973 rcuscale.kfree_loops= [KNL] 4974 Number of loops doing rcuscale.kfree_alloc_num number 4975 of allocations and frees. 4976 4977 rcuscale.nreaders= [KNL] 4978 Set number of RCU readers. The value -1 selects 4979 N, where N is the number of CPUs. A value 4980 "n" less than -1 selects N-n+1, where N is again 4981 the number of CPUs. For example, -2 selects N 4982 (the number of CPUs), -3 selects N+1, and so on. 4983 A value of "n" less than or equal to -N selects 4984 a single reader. 4985 4986 rcuscale.nwriters= [KNL] 4987 Set number of RCU writers. The values operate 4988 the same as for rcuscale.nreaders. 4989 N, where N is the number of CPUs 4990 4991 rcuscale.perf_type= [KNL] 4992 Specify the RCU implementation to test. 4993 4994 rcuscale.shutdown= [KNL] 4995 Shut the system down after performance tests 4996 complete. This is useful for hands-off automated 4997 testing. 4998 4999 rcuscale.verbose= [KNL] 5000 Enable additional printk() statements. 5001 5002 rcuscale.writer_holdoff= [KNL] 5003 Write-side holdoff between grace periods, 5004 in microseconds. The default of zero says 5005 no holdoff. 5006 5007 rcutorture.fqs_duration= [KNL] 5008 Set duration of force_quiescent_state bursts 5009 in microseconds. 5010 5011 rcutorture.fqs_holdoff= [KNL] 5012 Set holdoff time within force_quiescent_state bursts 5013 in microseconds. 5014 5015 rcutorture.fqs_stutter= [KNL] 5016 Set wait time between force_quiescent_state bursts 5017 in seconds. 5018 5019 rcutorture.fwd_progress= [KNL] 5020 Specifies the number of kthreads to be used 5021 for RCU grace-period forward-progress testing 5022 for the types of RCU supporting this notion. 5023 Defaults to 1 kthread, values less than zero or 5024 greater than the number of CPUs cause the number 5025 of CPUs to be used. 5026 5027 rcutorture.fwd_progress_div= [KNL] 5028 Specify the fraction of a CPU-stall-warning 5029 period to do tight-loop forward-progress testing. 5030 5031 rcutorture.fwd_progress_holdoff= [KNL] 5032 Number of seconds to wait between successive 5033 forward-progress tests. 5034 5035 rcutorture.fwd_progress_need_resched= [KNL] 5036 Enclose cond_resched() calls within checks for 5037 need_resched() during tight-loop forward-progress 5038 testing. 5039 5040 rcutorture.gp_cond= [KNL] 5041 Use conditional/asynchronous update-side 5042 primitives, if available. 5043 5044 rcutorture.gp_exp= [KNL] 5045 Use expedited update-side primitives, if available. 5046 5047 rcutorture.gp_normal= [KNL] 5048 Use normal (non-expedited) asynchronous 5049 update-side primitives, if available. 5050 5051 rcutorture.gp_sync= [KNL] 5052 Use normal (non-expedited) synchronous 5053 update-side primitives, if available. If all 5054 of rcutorture.gp_cond=, rcutorture.gp_exp=, 5055 rcutorture.gp_normal=, and rcutorture.gp_sync= 5056 are zero, rcutorture acts as if is interpreted 5057 they are all non-zero. 5058 5059 rcutorture.irqreader= [KNL] 5060 Run RCU readers from irq handlers, or, more 5061 accurately, from a timer handler. Not all RCU 5062 flavors take kindly to this sort of thing. 5063 5064 rcutorture.leakpointer= [KNL] 5065 Leak an RCU-protected pointer out of the reader. 5066 This can of course result in splats, and is 5067 intended to test the ability of things like 5068 CONFIG_RCU_STRICT_GRACE_PERIOD=y to detect 5069 such leaks. 5070 5071 rcutorture.n_barrier_cbs= [KNL] 5072 Set callbacks/threads for rcu_barrier() testing. 5073 5074 rcutorture.nfakewriters= [KNL] 5075 Set number of concurrent RCU writers. These just 5076 stress RCU, they don't participate in the actual 5077 test, hence the "fake". 5078 5079 rcutorture.nocbs_nthreads= [KNL] 5080 Set number of RCU callback-offload togglers. 5081 Zero (the default) disables toggling. 5082 5083 rcutorture.nocbs_toggle= [KNL] 5084 Set the delay in milliseconds between successive 5085 callback-offload toggling attempts. 5086 5087 rcutorture.nreaders= [KNL] 5088 Set number of RCU readers. The value -1 selects 5089 N-1, where N is the number of CPUs. A value 5090 "n" less than -1 selects N-n-2, where N is again 5091 the number of CPUs. For example, -2 selects N 5092 (the number of CPUs), -3 selects N+1, and so on. 5093 5094 rcutorture.object_debug= [KNL] 5095 Enable debug-object double-call_rcu() testing. 5096 5097 rcutorture.onoff_holdoff= [KNL] 5098 Set time (s) after boot for CPU-hotplug testing. 5099 5100 rcutorture.onoff_interval= [KNL] 5101 Set time (jiffies) between CPU-hotplug operations, 5102 or zero to disable CPU-hotplug testing. 5103 5104 rcutorture.read_exit= [KNL] 5105 Set the number of read-then-exit kthreads used 5106 to test the interaction of RCU updaters and 5107 task-exit processing. 5108 5109 rcutorture.read_exit_burst= [KNL] 5110 The number of times in a given read-then-exit 5111 episode that a set of read-then-exit kthreads 5112 is spawned. 5113 5114 rcutorture.read_exit_delay= [KNL] 5115 The delay, in seconds, between successive 5116 read-then-exit testing episodes. 5117 5118 rcutorture.shuffle_interval= [KNL] 5119 Set task-shuffle interval (s). Shuffling tasks 5120 allows some CPUs to go into dyntick-idle mode 5121 during the rcutorture test. 5122 5123 rcutorture.shutdown_secs= [KNL] 5124 Set time (s) after boot system shutdown. This 5125 is useful for hands-off automated testing. 5126 5127 rcutorture.stall_cpu= [KNL] 5128 Duration of CPU stall (s) to test RCU CPU stall 5129 warnings, zero to disable. 5130 5131 rcutorture.stall_cpu_block= [KNL] 5132 Sleep while stalling if set. This will result 5133 in warnings from preemptible RCU in addition to 5134 any other stall-related activity. Note that 5135 in kernels built with CONFIG_PREEMPTION=n and 5136 CONFIG_PREEMPT_COUNT=y, this parameter will 5137 cause the CPU to pass through a quiescent state. 5138 Given CONFIG_PREEMPTION=n, this will suppress 5139 RCU CPU stall warnings, but will instead result 5140 in scheduling-while-atomic splats. 5141 5142 Use of this module parameter results in splats. 5143 5144 5145 rcutorture.stall_cpu_holdoff= [KNL] 5146 Time to wait (s) after boot before inducing stall. 5147 5148 rcutorture.stall_cpu_irqsoff= [KNL] 5149 Disable interrupts while stalling if set. 5150 5151 rcutorture.stall_gp_kthread= [KNL] 5152 Duration (s) of forced sleep within RCU 5153 grace-period kthread to test RCU CPU stall 5154 warnings, zero to disable. If both stall_cpu 5155 and stall_gp_kthread are specified, the 5156 kthread is starved first, then the CPU. 5157 5158 rcutorture.stat_interval= [KNL] 5159 Time (s) between statistics printk()s. 5160 5161 rcutorture.stutter= [KNL] 5162 Time (s) to stutter testing, for example, specifying 5163 five seconds causes the test to run for five seconds, 5164 wait for five seconds, and so on. This tests RCU's 5165 ability to transition abruptly to and from idle. 5166 5167 rcutorture.test_boost= [KNL] 5168 Test RCU priority boosting? 0=no, 1=maybe, 2=yes. 5169 "Maybe" means test if the RCU implementation 5170 under test support RCU priority boosting. 5171 5172 rcutorture.test_boost_duration= [KNL] 5173 Duration (s) of each individual boost test. 5174 5175 rcutorture.test_boost_interval= [KNL] 5176 Interval (s) between each boost test. 5177 5178 rcutorture.test_no_idle_hz= [KNL] 5179 Test RCU's dyntick-idle handling. See also the 5180 rcutorture.shuffle_interval parameter. 5181 5182 rcutorture.torture_type= [KNL] 5183 Specify the RCU implementation to test. 5184 5185 rcutorture.verbose= [KNL] 5186 Enable additional printk() statements. 5187 5188 rcupdate.rcu_cpu_stall_ftrace_dump= [KNL] 5189 Dump ftrace buffer after reporting RCU CPU 5190 stall warning. 5191 5192 rcupdate.rcu_cpu_stall_suppress= [KNL] 5193 Suppress RCU CPU stall warning messages. 5194 5195 rcupdate.rcu_cpu_stall_suppress_at_boot= [KNL] 5196 Suppress RCU CPU stall warning messages and 5197 rcutorture writer stall warnings that occur 5198 during early boot, that is, during the time 5199 before the init task is spawned. 5200 5201 rcupdate.rcu_cpu_stall_timeout= [KNL] 5202 Set timeout for RCU CPU stall warning messages. 5203 The value is in seconds and the maximum allowed 5204 value is 300 seconds. 5205 5206 rcupdate.rcu_exp_cpu_stall_timeout= [KNL] 5207 Set timeout for expedited RCU CPU stall warning 5208 messages. The value is in milliseconds 5209 and the maximum allowed value is 21000 5210 milliseconds. Please note that this value is 5211 adjusted to an arch timer tick resolution. 5212 Setting this to zero causes the value from 5213 rcupdate.rcu_cpu_stall_timeout to be used (after 5214 conversion from seconds to milliseconds). 5215 5216 rcupdate.rcu_cpu_stall_cputime= [KNL] 5217 Provide statistics on the cputime and count of 5218 interrupts and tasks during the sampling period. For 5219 multiple continuous RCU stalls, all sampling periods 5220 begin at half of the first RCU stall timeout. 5221 5222 rcupdate.rcu_exp_stall_task_details= [KNL] 5223 Print stack dumps of any tasks blocking the 5224 current expedited RCU grace period during an 5225 expedited RCU CPU stall warning. 5226 5227 rcupdate.rcu_expedited= [KNL] 5228 Use expedited grace-period primitives, for 5229 example, synchronize_rcu_expedited() instead 5230 of synchronize_rcu(). This reduces latency, 5231 but can increase CPU utilization, degrade 5232 real-time latency, and degrade energy efficiency. 5233 No effect on CONFIG_TINY_RCU kernels. 5234 5235 rcupdate.rcu_normal= [KNL] 5236 Use only normal grace-period primitives, 5237 for example, synchronize_rcu() instead of 5238 synchronize_rcu_expedited(). This improves 5239 real-time latency, CPU utilization, and 5240 energy efficiency, but can expose users to 5241 increased grace-period latency. This parameter 5242 overrides rcupdate.rcu_expedited. No effect on 5243 CONFIG_TINY_RCU kernels. 5244 5245 rcupdate.rcu_normal_after_boot= [KNL] 5246 Once boot has completed (that is, after 5247 rcu_end_inkernel_boot() has been invoked), use 5248 only normal grace-period primitives. No effect 5249 on CONFIG_TINY_RCU kernels. 5250 5251 But note that CONFIG_PREEMPT_RT=y kernels enables 5252 this kernel boot parameter, forcibly setting 5253 it to the value one, that is, converting any 5254 post-boot attempt at an expedited RCU grace 5255 period to instead use normal non-expedited 5256 grace-period processing. 5257 5258 rcupdate.rcu_task_collapse_lim= [KNL] 5259 Set the maximum number of callbacks present 5260 at the beginning of a grace period that allows 5261 the RCU Tasks flavors to collapse back to using 5262 a single callback queue. This switching only 5263 occurs when rcupdate.rcu_task_enqueue_lim is 5264 set to the default value of -1. 5265 5266 rcupdate.rcu_task_contend_lim= [KNL] 5267 Set the minimum number of callback-queuing-time 5268 lock-contention events per jiffy required to 5269 cause the RCU Tasks flavors to switch to per-CPU 5270 callback queuing. This switching only occurs 5271 when rcupdate.rcu_task_enqueue_lim is set to 5272 the default value of -1. 5273 5274 rcupdate.rcu_task_enqueue_lim= [KNL] 5275 Set the number of callback queues to use for the 5276 RCU Tasks family of RCU flavors. The default 5277 of -1 allows this to be automatically (and 5278 dynamically) adjusted. This parameter is intended 5279 for use in testing. 5280 5281 rcupdate.rcu_task_ipi_delay= [KNL] 5282 Set time in jiffies during which RCU tasks will 5283 avoid sending IPIs, starting with the beginning 5284 of a given grace period. Setting a large 5285 number avoids disturbing real-time workloads, 5286 but lengthens grace periods. 5287 5288 rcupdate.rcu_task_stall_info= [KNL] 5289 Set initial timeout in jiffies for RCU task stall 5290 informational messages, which give some indication 5291 of the problem for those not patient enough to 5292 wait for ten minutes. Informational messages are 5293 only printed prior to the stall-warning message 5294 for a given grace period. Disable with a value 5295 less than or equal to zero. Defaults to ten 5296 seconds. A change in value does not take effect 5297 until the beginning of the next grace period. 5298 5299 rcupdate.rcu_task_stall_info_mult= [KNL] 5300 Multiplier for time interval between successive 5301 RCU task stall informational messages for a given 5302 RCU tasks grace period. This value is clamped 5303 to one through ten, inclusive. It defaults to 5304 the value three, so that the first informational 5305 message is printed 10 seconds into the grace 5306 period, the second at 40 seconds, the third at 5307 160 seconds, and then the stall warning at 600 5308 seconds would prevent a fourth at 640 seconds. 5309 5310 rcupdate.rcu_task_stall_timeout= [KNL] 5311 Set timeout in jiffies for RCU task stall 5312 warning messages. Disable with a value less 5313 than or equal to zero. Defaults to ten minutes. 5314 A change in value does not take effect until 5315 the beginning of the next grace period. 5316 5317 rcupdate.rcu_self_test= [KNL] 5318 Run the RCU early boot self tests 5319 5320 rdinit= [KNL] 5321 Format: <full_path> 5322 Run specified binary instead of /init from the ramdisk, 5323 used for early userspace startup. See initrd. 5324 5325 rdrand= [X86] 5326 force - Override the decision by the kernel to hide the 5327 advertisement of RDRAND support (this affects 5328 certain AMD processors because of buggy BIOS 5329 support, specifically around the suspend/resume 5330 path). 5331 5332 rdt= [HW,X86,RDT] 5333 Turn on/off individual RDT features. List is: 5334 cmt, mbmtotal, mbmlocal, l3cat, l3cdp, l2cat, l2cdp, 5335 mba, smba, bmec. 5336 E.g. to turn on cmt and turn off mba use: 5337 rdt=cmt,!mba 5338 5339 reboot= [KNL] 5340 Format (x86 or x86_64): 5341 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] | d[efault] \ 5342 [[,]s[mp]#### \ 5343 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \ 5344 [[,]f[orce] 5345 Where reboot_mode is one of warm (soft) or cold (hard) or gpio 5346 (prefix with 'panic_' to set mode for panic 5347 reboot only), 5348 reboot_type is one of bios, acpi, kbd, triple, efi, or pci, 5349 reboot_force is either force or not specified, 5350 reboot_cpu is s[mp]#### with #### being the processor 5351 to be used for rebooting. 5352 5353 refscale.holdoff= [KNL] 5354 Set test-start holdoff period. The purpose of 5355 this parameter is to delay the start of the 5356 test until boot completes in order to avoid 5357 interference. 5358 5359 refscale.loops= [KNL] 5360 Set the number of loops over the synchronization 5361 primitive under test. Increasing this number 5362 reduces noise due to loop start/end overhead, 5363 but the default has already reduced the per-pass 5364 noise to a handful of picoseconds on ca. 2020 5365 x86 laptops. 5366 5367 refscale.nreaders= [KNL] 5368 Set number of readers. The default value of -1 5369 selects N, where N is roughly 75% of the number 5370 of CPUs. A value of zero is an interesting choice. 5371 5372 refscale.nruns= [KNL] 5373 Set number of runs, each of which is dumped onto 5374 the console log. 5375 5376 refscale.readdelay= [KNL] 5377 Set the read-side critical-section duration, 5378 measured in microseconds. 5379 5380 refscale.scale_type= [KNL] 5381 Specify the read-protection implementation to test. 5382 5383 refscale.shutdown= [KNL] 5384 Shut down the system at the end of the performance 5385 test. This defaults to 1 (shut it down) when 5386 refscale is built into the kernel and to 0 (leave 5387 it running) when refscale is built as a module. 5388 5389 refscale.verbose= [KNL] 5390 Enable additional printk() statements. 5391 5392 refscale.verbose_batched= [KNL] 5393 Batch the additional printk() statements. If zero 5394 (the default) or negative, print everything. Otherwise, 5395 print every Nth verbose statement, where N is the value 5396 specified. 5397 5398 relax_domain_level= 5399 [KNL, SMP] Set scheduler's default relax_domain_level. 5400 See Documentation/admin-guide/cgroup-v1/cpusets.rst. 5401 5402 reserve= [KNL,BUGS] Force kernel to ignore I/O ports or memory 5403 Format: <base1>,<size1>[,<base2>,<size2>,...] 5404 Reserve I/O ports or memory so the kernel won't use 5405 them. If <base> is less than 0x10000, the region 5406 is assumed to be I/O ports; otherwise it is memory. 5407 5408 reservetop= [X86-32] 5409 Format: nn[KMG] 5410 Reserves a hole at the top of the kernel virtual 5411 address space. 5412 5413 reset_devices [KNL] Force drivers to reset the underlying device 5414 during initialization. 5415 5416 resume= [SWSUSP] 5417 Specify the partition device for software suspend 5418 Format: 5419 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>} 5420 5421 resume_offset= [SWSUSP] 5422 Specify the offset from the beginning of the partition 5423 given by "resume=" at which the swap header is located, 5424 in <PAGE_SIZE> units (needed only for swap files). 5425 See Documentation/power/swsusp-and-swap-files.rst 5426 5427 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to 5428 read the resume files 5429 5430 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up. 5431 Useful for devices that are detected asynchronously 5432 (e.g. USB and MMC devices). 5433 5434 retain_initrd [RAM] Keep initrd memory after extraction 5435 5436 retbleed= [X86] Control mitigation of RETBleed (Arbitrary 5437 Speculative Code Execution with Return Instructions) 5438 vulnerability. 5439 5440 AMD-based UNRET and IBPB mitigations alone do not stop 5441 sibling threads from influencing the predictions of other 5442 sibling threads. For that reason, STIBP is used on pro- 5443 cessors that support it, and mitigate SMT on processors 5444 that don't. 5445 5446 off - no mitigation 5447 auto - automatically select a migitation 5448 auto,nosmt - automatically select a mitigation, 5449 disabling SMT if necessary for 5450 the full mitigation (only on Zen1 5451 and older without STIBP). 5452 ibpb - On AMD, mitigate short speculation 5453 windows on basic block boundaries too. 5454 Safe, highest perf impact. It also 5455 enables STIBP if present. Not suitable 5456 on Intel. 5457 ibpb,nosmt - Like "ibpb" above but will disable SMT 5458 when STIBP is not available. This is 5459 the alternative for systems which do not 5460 have STIBP. 5461 unret - Force enable untrained return thunks, 5462 only effective on AMD f15h-f17h based 5463 systems. 5464 unret,nosmt - Like unret, but will disable SMT when STIBP 5465 is not available. This is the alternative for 5466 systems which do not have STIBP. 5467 5468 Selecting 'auto' will choose a mitigation method at run 5469 time according to the CPU. 5470 5471 Not specifying this option is equivalent to retbleed=auto. 5472 5473 rfkill.default_state= 5474 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm, 5475 etc. communication is blocked by default. 5476 1 Unblocked. 5477 5478 rfkill.master_switch_mode= 5479 0 The "airplane mode" button does nothing. 5480 1 The "airplane mode" button toggles between everything 5481 blocked and the previous configuration. 5482 2 The "airplane mode" button toggles between everything 5483 blocked and everything unblocked. 5484 5485 rhash_entries= [KNL,NET] 5486 Set number of hash buckets for route cache 5487 5488 ring3mwait=disable 5489 [KNL] Disable ring 3 MONITOR/MWAIT feature on supported 5490 CPUs. 5491 5492 ro [KNL] Mount root device read-only on boot 5493 5494 rodata= [KNL] 5495 on Mark read-only kernel memory as read-only (default). 5496 off Leave read-only kernel memory writable for debugging. 5497 full Mark read-only kernel memory and aliases as read-only 5498 [arm64] 5499 5500 rockchip.usb_uart 5501 Enable the uart passthrough on the designated usb port 5502 on Rockchip SoCs. When active, the signals of the 5503 debug-uart get routed to the D+ and D- pins of the usb 5504 port and the regular usb controller gets disabled. 5505 5506 root= [KNL] Root filesystem 5507 Usually this a a block device specifier of some kind, 5508 see the early_lookup_bdev comment in 5509 block/early-lookup.c for details. 5510 Alternatively this can be "ram" for the legacy initial 5511 ramdisk, "nfs" and "cifs" for root on a network file 5512 system, or "mtd" and "ubi" for mounting from raw flash. 5513 5514 rootdelay= [KNL] Delay (in seconds) to pause before attempting to 5515 mount the root filesystem 5516 5517 rootflags= [KNL] Set root filesystem mount option string 5518 5519 rootfstype= [KNL] Set root filesystem type 5520 5521 rootwait [KNL] Wait (indefinitely) for root device to show up. 5522 Useful for devices that are detected asynchronously 5523 (e.g. USB and MMC devices). 5524 5525 rproc_mem=nn[KMG][@address] 5526 [KNL,ARM,CMA] Remoteproc physical memory block. 5527 Memory area to be used by remote processor image, 5528 managed by CMA. 5529 5530 rw [KNL] Mount root device read-write on boot 5531 5532 S [KNL] Run init in single mode 5533 5534 s390_iommu= [HW,S390] 5535 Set s390 IOTLB flushing mode 5536 strict 5537 With strict flushing every unmap operation will result in 5538 an IOTLB flush. Default is lazy flushing before reuse, 5539 which is faster. 5540 5541 s390_iommu_aperture= [KNL,S390] 5542 Specifies the size of the per device DMA address space 5543 accessible through the DMA and IOMMU APIs as a decimal 5544 factor of the size of main memory. 5545 The default is 1 meaning that one can concurrently use 5546 as many DMA addresses as physical memory is installed, 5547 if supported by hardware, and thus map all of memory 5548 once. With a value of 2 one can map all of memory twice 5549 and so on. As a special case a factor of 0 imposes no 5550 restrictions other than those given by hardware at the 5551 cost of significant additional memory use for tables. 5552 5553 sa1100ir [NET] 5554 See drivers/net/irda/sa1100_ir.c. 5555 5556 sched_verbose [KNL] Enables verbose scheduler debug messages. 5557 5558 schedstats= [KNL,X86] Enable or disable scheduled statistics. 5559 Allowed values are enable and disable. This feature 5560 incurs a small amount of overhead in the scheduler 5561 but is useful for debugging and performance tuning. 5562 5563 sched_thermal_decay_shift= 5564 [KNL, SMP] Set a decay shift for scheduler thermal 5565 pressure signal. Thermal pressure signal follows the 5566 default decay period of other scheduler pelt 5567 signals(usually 32 ms but configurable). Setting 5568 sched_thermal_decay_shift will left shift the decay 5569 period for the thermal pressure signal by the shift 5570 value. 5571 i.e. with the default pelt decay period of 32 ms 5572 sched_thermal_decay_shift thermal pressure decay pr 5573 1 64 ms 5574 2 128 ms 5575 and so on. 5576 Format: integer between 0 and 10 5577 Default is 0. 5578 5579 scftorture.holdoff= [KNL] 5580 Number of seconds to hold off before starting 5581 test. Defaults to zero for module insertion and 5582 to 10 seconds for built-in smp_call_function() 5583 tests. 5584 5585 scftorture.longwait= [KNL] 5586 Request ridiculously long waits randomly selected 5587 up to the chosen limit in seconds. Zero (the 5588 default) disables this feature. Please note 5589 that requesting even small non-zero numbers of 5590 seconds can result in RCU CPU stall warnings, 5591 softlockup complaints, and so on. 5592 5593 scftorture.nthreads= [KNL] 5594 Number of kthreads to spawn to invoke the 5595 smp_call_function() family of functions. 5596 The default of -1 specifies a number of kthreads 5597 equal to the number of CPUs. 5598 5599 scftorture.onoff_holdoff= [KNL] 5600 Number seconds to wait after the start of the 5601 test before initiating CPU-hotplug operations. 5602 5603 scftorture.onoff_interval= [KNL] 5604 Number seconds to wait between successive 5605 CPU-hotplug operations. Specifying zero (which 5606 is the default) disables CPU-hotplug operations. 5607 5608 scftorture.shutdown_secs= [KNL] 5609 The number of seconds following the start of the 5610 test after which to shut down the system. The 5611 default of zero avoids shutting down the system. 5612 Non-zero values are useful for automated tests. 5613 5614 scftorture.stat_interval= [KNL] 5615 The number of seconds between outputting the 5616 current test statistics to the console. A value 5617 of zero disables statistics output. 5618 5619 scftorture.stutter_cpus= [KNL] 5620 The number of jiffies to wait between each change 5621 to the set of CPUs under test. 5622 5623 scftorture.use_cpus_read_lock= [KNL] 5624 Use use_cpus_read_lock() instead of the default 5625 preempt_disable() to disable CPU hotplug 5626 while invoking one of the smp_call_function*() 5627 functions. 5628 5629 scftorture.verbose= [KNL] 5630 Enable additional printk() statements. 5631 5632 scftorture.weight_single= [KNL] 5633 The probability weighting to use for the 5634 smp_call_function_single() function with a zero 5635 "wait" parameter. A value of -1 selects the 5636 default if all other weights are -1. However, 5637 if at least one weight has some other value, a 5638 value of -1 will instead select a weight of zero. 5639 5640 scftorture.weight_single_wait= [KNL] 5641 The probability weighting to use for the 5642 smp_call_function_single() function with a 5643 non-zero "wait" parameter. See weight_single. 5644 5645 scftorture.weight_many= [KNL] 5646 The probability weighting to use for the 5647 smp_call_function_many() function with a zero 5648 "wait" parameter. See weight_single. 5649 Note well that setting a high probability for 5650 this weighting can place serious IPI load 5651 on the system. 5652 5653 scftorture.weight_many_wait= [KNL] 5654 The probability weighting to use for the 5655 smp_call_function_many() function with a 5656 non-zero "wait" parameter. See weight_single 5657 and weight_many. 5658 5659 scftorture.weight_all= [KNL] 5660 The probability weighting to use for the 5661 smp_call_function_all() function with a zero 5662 "wait" parameter. See weight_single and 5663 weight_many. 5664 5665 scftorture.weight_all_wait= [KNL] 5666 The probability weighting to use for the 5667 smp_call_function_all() function with a 5668 non-zero "wait" parameter. See weight_single 5669 and weight_many. 5670 5671 skew_tick= [KNL] Offset the periodic timer tick per cpu to mitigate 5672 xtime_lock contention on larger systems, and/or RCU lock 5673 contention on all systems with CONFIG_MAXSMP set. 5674 Format: { "0" | "1" } 5675 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1" 5676 1 -- enable. 5677 Note: increases power consumption, thus should only be 5678 enabled if running jitter sensitive (HPC/RT) workloads. 5679 5680 security= [SECURITY] Choose a legacy "major" security module to 5681 enable at boot. This has been deprecated by the 5682 "lsm=" parameter. 5683 5684 selinux= [SELINUX] Disable or enable SELinux at boot time. 5685 Format: { "0" | "1" } 5686 See security/selinux/Kconfig help text. 5687 0 -- disable. 5688 1 -- enable. 5689 Default value is 1. 5690 5691 serialnumber [BUGS=X86-32] 5692 5693 sev=option[,option...] [X86-64] See Documentation/arch/x86/x86_64/boot-options.rst 5694 5695 shapers= [NET] 5696 Maximal number of shapers. 5697 5698 show_lapic= [APIC,X86] Advanced Programmable Interrupt Controller 5699 Limit apic dumping. The parameter defines the maximal 5700 number of local apics being dumped. Also it is possible 5701 to set it to "all" by meaning -- no limit here. 5702 Format: { 1 (default) | 2 | ... | all }. 5703 The parameter valid if only apic=debug or 5704 apic=verbose is specified. 5705 Example: apic=debug show_lapic=all 5706 5707 simeth= [IA-64] 5708 simscsi= 5709 5710 slram= [HW,MTD] 5711 5712 slab_merge [MM] 5713 Enable merging of slabs with similar size when the 5714 kernel is built without CONFIG_SLAB_MERGE_DEFAULT. 5715 5716 slab_nomerge [MM] 5717 Disable merging of slabs with similar size. May be 5718 necessary if there is some reason to distinguish 5719 allocs to different slabs, especially in hardened 5720 environments where the risk of heap overflows and 5721 layout control by attackers can usually be 5722 frustrated by disabling merging. This will reduce 5723 most of the exposure of a heap attack to a single 5724 cache (risks via metadata attacks are mostly 5725 unchanged). Debug options disable merging on their 5726 own. 5727 For more information see Documentation/mm/slub.rst. 5728 5729 slab_max_order= [MM, SLAB] 5730 Determines the maximum allowed order for slabs. 5731 A high setting may cause OOMs due to memory 5732 fragmentation. Defaults to 1 for systems with 5733 more than 32MB of RAM, 0 otherwise. 5734 5735 slub_debug[=options[,slabs][;[options[,slabs]]...] [MM, SLUB] 5736 Enabling slub_debug allows one to determine the 5737 culprit if slab objects become corrupted. Enabling 5738 slub_debug can create guard zones around objects and 5739 may poison objects when not in use. Also tracks the 5740 last alloc / free. For more information see 5741 Documentation/mm/slub.rst. 5742 5743 slub_max_order= [MM, SLUB] 5744 Determines the maximum allowed order for slabs. 5745 A high setting may cause OOMs due to memory 5746 fragmentation. For more information see 5747 Documentation/mm/slub.rst. 5748 5749 slub_min_objects= [MM, SLUB] 5750 The minimum number of objects per slab. SLUB will 5751 increase the slab order up to slub_max_order to 5752 generate a sufficiently large slab able to contain 5753 the number of objects indicated. The higher the number 5754 of objects the smaller the overhead of tracking slabs 5755 and the less frequently locks need to be acquired. 5756 For more information see Documentation/mm/slub.rst. 5757 5758 slub_min_order= [MM, SLUB] 5759 Determines the minimum page order for slabs. Must be 5760 lower than slub_max_order. 5761 For more information see Documentation/mm/slub.rst. 5762 5763 slub_merge [MM, SLUB] 5764 Same with slab_merge. 5765 5766 slub_nomerge [MM, SLUB] 5767 Same with slab_nomerge. This is supported for legacy. 5768 See slab_nomerge for more information. 5769 5770 smart2= [HW] 5771 Format: <io1>[,<io2>[,...,<io8>]] 5772 5773 smp.csd_lock_timeout= [KNL] 5774 Specify the period of time in milliseconds 5775 that smp_call_function() and friends will wait 5776 for a CPU to release the CSD lock. This is 5777 useful when diagnosing bugs involving CPUs 5778 disabling interrupts for extended periods 5779 of time. Defaults to 5,000 milliseconds, and 5780 setting a value of zero disables this feature. 5781 This feature may be more efficiently disabled 5782 using the csdlock_debug- kernel parameter. 5783 5784 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices 5785 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port 5786 smsc-ircc2.ircc_sir= [HW] SIR base I/O port 5787 smsc-ircc2.ircc_fir= [HW] FIR base I/O port 5788 smsc-ircc2.ircc_irq= [HW] IRQ line 5789 smsc-ircc2.ircc_dma= [HW] DMA channel 5790 smsc-ircc2.ircc_transceiver= [HW] Transceiver type: 5791 0: Toshiba Satellite 1800 (GP data pin select) 5792 1: Fast pin select (default) 5793 2: ATC IRMode 5794 5795 smt= [KNL,MIPS,S390] Set the maximum number of threads (logical 5796 CPUs) to use per physical CPU on systems capable of 5797 symmetric multithreading (SMT). Will be capped to the 5798 actual hardware limit. 5799 Format: <integer> 5800 Default: -1 (no limit) 5801 5802 softlockup_panic= 5803 [KNL] Should the soft-lockup detector generate panics. 5804 Format: 0 | 1 5805 5806 A value of 1 instructs the soft-lockup detector 5807 to panic the machine when a soft-lockup occurs. It is 5808 also controlled by the kernel.softlockup_panic sysctl 5809 and CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC, which is the 5810 respective build-time switch to that functionality. 5811 5812 softlockup_all_cpu_backtrace= 5813 [KNL] Should the soft-lockup detector generate 5814 backtraces on all cpus. 5815 Format: 0 | 1 5816 5817 sonypi.*= [HW] Sony Programmable I/O Control Device driver 5818 See Documentation/admin-guide/laptops/sonypi.rst 5819 5820 spectre_v2= [X86] Control mitigation of Spectre variant 2 5821 (indirect branch speculation) vulnerability. 5822 The default operation protects the kernel from 5823 user space attacks. 5824 5825 on - unconditionally enable, implies 5826 spectre_v2_user=on 5827 off - unconditionally disable, implies 5828 spectre_v2_user=off 5829 auto - kernel detects whether your CPU model is 5830 vulnerable 5831 5832 Selecting 'on' will, and 'auto' may, choose a 5833 mitigation method at run time according to the 5834 CPU, the available microcode, the setting of the 5835 CONFIG_RETPOLINE configuration option, and the 5836 compiler with which the kernel was built. 5837 5838 Selecting 'on' will also enable the mitigation 5839 against user space to user space task attacks. 5840 5841 Selecting 'off' will disable both the kernel and 5842 the user space protections. 5843 5844 Specific mitigations can also be selected manually: 5845 5846 retpoline - replace indirect branches 5847 retpoline,generic - Retpolines 5848 retpoline,lfence - LFENCE; indirect branch 5849 retpoline,amd - alias for retpoline,lfence 5850 eibrs - Enhanced/Auto IBRS 5851 eibrs,retpoline - Enhanced/Auto IBRS + Retpolines 5852 eibrs,lfence - Enhanced/Auto IBRS + LFENCE 5853 ibrs - use IBRS to protect kernel 5854 5855 Not specifying this option is equivalent to 5856 spectre_v2=auto. 5857 5858 spectre_v2_user= 5859 [X86] Control mitigation of Spectre variant 2 5860 (indirect branch speculation) vulnerability between 5861 user space tasks 5862 5863 on - Unconditionally enable mitigations. Is 5864 enforced by spectre_v2=on 5865 5866 off - Unconditionally disable mitigations. Is 5867 enforced by spectre_v2=off 5868 5869 prctl - Indirect branch speculation is enabled, 5870 but mitigation can be enabled via prctl 5871 per thread. The mitigation control state 5872 is inherited on fork. 5873 5874 prctl,ibpb 5875 - Like "prctl" above, but only STIBP is 5876 controlled per thread. IBPB is issued 5877 always when switching between different user 5878 space processes. 5879 5880 seccomp 5881 - Same as "prctl" above, but all seccomp 5882 threads will enable the mitigation unless 5883 they explicitly opt out. 5884 5885 seccomp,ibpb 5886 - Like "seccomp" above, but only STIBP is 5887 controlled per thread. IBPB is issued 5888 always when switching between different 5889 user space processes. 5890 5891 auto - Kernel selects the mitigation depending on 5892 the available CPU features and vulnerability. 5893 5894 Default mitigation: "prctl" 5895 5896 Not specifying this option is equivalent to 5897 spectre_v2_user=auto. 5898 5899 spec_rstack_overflow= 5900 [X86] Control RAS overflow mitigation on AMD Zen CPUs 5901 5902 off - Disable mitigation 5903 microcode - Enable microcode mitigation only 5904 safe-ret - Enable sw-only safe RET mitigation (default) 5905 ibpb - Enable mitigation by issuing IBPB on 5906 kernel entry 5907 ibpb-vmexit - Issue IBPB only on VMEXIT 5908 (cloud-specific mitigation) 5909 5910 spec_store_bypass_disable= 5911 [HW] Control Speculative Store Bypass (SSB) Disable mitigation 5912 (Speculative Store Bypass vulnerability) 5913 5914 Certain CPUs are vulnerable to an exploit against a 5915 a common industry wide performance optimization known 5916 as "Speculative Store Bypass" in which recent stores 5917 to the same memory location may not be observed by 5918 later loads during speculative execution. The idea 5919 is that such stores are unlikely and that they can 5920 be detected prior to instruction retirement at the 5921 end of a particular speculation execution window. 5922 5923 In vulnerable processors, the speculatively forwarded 5924 store can be used in a cache side channel attack, for 5925 example to read memory to which the attacker does not 5926 directly have access (e.g. inside sandboxed code). 5927 5928 This parameter controls whether the Speculative Store 5929 Bypass optimization is used. 5930 5931 On x86 the options are: 5932 5933 on - Unconditionally disable Speculative Store Bypass 5934 off - Unconditionally enable Speculative Store Bypass 5935 auto - Kernel detects whether the CPU model contains an 5936 implementation of Speculative Store Bypass and 5937 picks the most appropriate mitigation. If the 5938 CPU is not vulnerable, "off" is selected. If the 5939 CPU is vulnerable the default mitigation is 5940 architecture and Kconfig dependent. See below. 5941 prctl - Control Speculative Store Bypass per thread 5942 via prctl. Speculative Store Bypass is enabled 5943 for a process by default. The state of the control 5944 is inherited on fork. 5945 seccomp - Same as "prctl" above, but all seccomp threads 5946 will disable SSB unless they explicitly opt out. 5947 5948 Default mitigations: 5949 X86: "prctl" 5950 5951 On powerpc the options are: 5952 5953 on,auto - On Power8 and Power9 insert a store-forwarding 5954 barrier on kernel entry and exit. On Power7 5955 perform a software flush on kernel entry and 5956 exit. 5957 off - No action. 5958 5959 Not specifying this option is equivalent to 5960 spec_store_bypass_disable=auto. 5961 5962 spia_io_base= [HW,MTD] 5963 spia_fio_base= 5964 spia_pedr= 5965 spia_peddr= 5966 5967 split_lock_detect= 5968 [X86] Enable split lock detection or bus lock detection 5969 5970 When enabled (and if hardware support is present), atomic 5971 instructions that access data across cache line 5972 boundaries will result in an alignment check exception 5973 for split lock detection or a debug exception for 5974 bus lock detection. 5975 5976 off - not enabled 5977 5978 warn - the kernel will emit rate-limited warnings 5979 about applications triggering the #AC 5980 exception or the #DB exception. This mode is 5981 the default on CPUs that support split lock 5982 detection or bus lock detection. Default 5983 behavior is by #AC if both features are 5984 enabled in hardware. 5985 5986 fatal - the kernel will send SIGBUS to applications 5987 that trigger the #AC exception or the #DB 5988 exception. Default behavior is by #AC if 5989 both features are enabled in hardware. 5990 5991 ratelimit:N - 5992 Set system wide rate limit to N bus locks 5993 per second for bus lock detection. 5994 0 < N <= 1000. 5995 5996 N/A for split lock detection. 5997 5998 5999 If an #AC exception is hit in the kernel or in 6000 firmware (i.e. not while executing in user mode) 6001 the kernel will oops in either "warn" or "fatal" 6002 mode. 6003 6004 #DB exception for bus lock is triggered only when 6005 CPL > 0. 6006 6007 srbds= [X86,INTEL] 6008 Control the Special Register Buffer Data Sampling 6009 (SRBDS) mitigation. 6010 6011 Certain CPUs are vulnerable to an MDS-like 6012 exploit which can leak bits from the random 6013 number generator. 6014 6015 By default, this issue is mitigated by 6016 microcode. However, the microcode fix can cause 6017 the RDRAND and RDSEED instructions to become 6018 much slower. Among other effects, this will 6019 result in reduced throughput from /dev/urandom. 6020 6021 The microcode mitigation can be disabled with 6022 the following option: 6023 6024 off: Disable mitigation and remove 6025 performance impact to RDRAND and RDSEED 6026 6027 srcutree.big_cpu_lim [KNL] 6028 Specifies the number of CPUs constituting a 6029 large system, such that srcu_struct structures 6030 should immediately allocate an srcu_node array. 6031 This kernel-boot parameter defaults to 128, 6032 but takes effect only when the low-order four 6033 bits of srcutree.convert_to_big is equal to 3 6034 (decide at boot). 6035 6036 srcutree.convert_to_big [KNL] 6037 Specifies under what conditions an SRCU tree 6038 srcu_struct structure will be converted to big 6039 form, that is, with an rcu_node tree: 6040 6041 0: Never. 6042 1: At init_srcu_struct() time. 6043 2: When rcutorture decides to. 6044 3: Decide at boot time (default). 6045 0x1X: Above plus if high contention. 6046 6047 Either way, the srcu_node tree will be sized based 6048 on the actual runtime number of CPUs (nr_cpu_ids) 6049 instead of the compile-time CONFIG_NR_CPUS. 6050 6051 srcutree.counter_wrap_check [KNL] 6052 Specifies how frequently to check for 6053 grace-period sequence counter wrap for the 6054 srcu_data structure's ->srcu_gp_seq_needed field. 6055 The greater the number of bits set in this kernel 6056 parameter, the less frequently counter wrap will 6057 be checked for. Note that the bottom two bits 6058 are ignored. 6059 6060 srcutree.exp_holdoff [KNL] 6061 Specifies how many nanoseconds must elapse 6062 since the end of the last SRCU grace period for 6063 a given srcu_struct until the next normal SRCU 6064 grace period will be considered for automatic 6065 expediting. Set to zero to disable automatic 6066 expediting. 6067 6068 srcutree.srcu_max_nodelay [KNL] 6069 Specifies the number of no-delay instances 6070 per jiffy for which the SRCU grace period 6071 worker thread will be rescheduled with zero 6072 delay. Beyond this limit, worker thread will 6073 be rescheduled with a sleep delay of one jiffy. 6074 6075 srcutree.srcu_max_nodelay_phase [KNL] 6076 Specifies the per-grace-period phase, number of 6077 non-sleeping polls of readers. Beyond this limit, 6078 grace period worker thread will be rescheduled 6079 with a sleep delay of one jiffy, between each 6080 rescan of the readers, for a grace period phase. 6081 6082 srcutree.srcu_retry_check_delay [KNL] 6083 Specifies number of microseconds of non-sleeping 6084 delay between each non-sleeping poll of readers. 6085 6086 srcutree.small_contention_lim [KNL] 6087 Specifies the number of update-side contention 6088 events per jiffy will be tolerated before 6089 initiating a conversion of an srcu_struct 6090 structure to big form. Note that the value of 6091 srcutree.convert_to_big must have the 0x10 bit 6092 set for contention-based conversions to occur. 6093 6094 ssbd= [ARM64,HW] 6095 Speculative Store Bypass Disable control 6096 6097 On CPUs that are vulnerable to the Speculative 6098 Store Bypass vulnerability and offer a 6099 firmware based mitigation, this parameter 6100 indicates how the mitigation should be used: 6101 6102 force-on: Unconditionally enable mitigation for 6103 for both kernel and userspace 6104 force-off: Unconditionally disable mitigation for 6105 for both kernel and userspace 6106 kernel: Always enable mitigation in the 6107 kernel, and offer a prctl interface 6108 to allow userspace to register its 6109 interest in being mitigated too. 6110 6111 stack_guard_gap= [MM] 6112 override the default stack gap protection. The value 6113 is in page units and it defines how many pages prior 6114 to (for stacks growing down) resp. after (for stacks 6115 growing up) the main stack are reserved for no other 6116 mapping. Default value is 256 pages. 6117 6118 stack_depot_disable= [KNL] 6119 Setting this to true through kernel command line will 6120 disable the stack depot thereby saving the static memory 6121 consumed by the stack hash table. By default this is set 6122 to false. 6123 6124 stacktrace [FTRACE] 6125 Enabled the stack tracer on boot up. 6126 6127 stacktrace_filter=[function-list] 6128 [FTRACE] Limit the functions that the stack tracer 6129 will trace at boot up. function-list is a comma-separated 6130 list of functions. This list can be changed at run 6131 time by the stack_trace_filter file in the debugfs 6132 tracing directory. Note, this enables stack tracing 6133 and the stacktrace above is not needed. 6134 6135 sti= [PARISC,HW] 6136 Format: <num> 6137 Set the STI (builtin display/keyboard on the HP-PARISC 6138 machines) console (graphic card) which should be used 6139 as the initial boot-console. 6140 See also comment in drivers/video/console/sticore.c. 6141 6142 sti_font= [HW] 6143 See comment in drivers/video/console/sticore.c. 6144 6145 stifb= [HW] 6146 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]] 6147 6148 strict_sas_size= 6149 [X86] 6150 Format: <bool> 6151 Enable or disable strict sigaltstack size checks 6152 against the required signal frame size which 6153 depends on the supported FPU features. This can 6154 be used to filter out binaries which have 6155 not yet been made aware of AT_MINSIGSTKSZ. 6156 6157 stress_hpt [PPC] 6158 Limits the number of kernel HPT entries in the hash 6159 page table to increase the rate of hash page table 6160 faults on kernel addresses. 6161 6162 stress_slb [PPC] 6163 Limits the number of kernel SLB entries, and flushes 6164 them frequently to increase the rate of SLB faults 6165 on kernel addresses. 6166 6167 sunrpc.min_resvport= 6168 sunrpc.max_resvport= 6169 [NFS,SUNRPC] 6170 SunRPC servers often require that client requests 6171 originate from a privileged port (i.e. a port in the 6172 range 0 < portnr < 1024). 6173 An administrator who wishes to reserve some of these 6174 ports for other uses may adjust the range that the 6175 kernel's sunrpc client considers to be privileged 6176 using these two parameters to set the minimum and 6177 maximum port values. 6178 6179 sunrpc.svc_rpc_per_connection_limit= 6180 [NFS,SUNRPC] 6181 Limit the number of requests that the server will 6182 process in parallel from a single connection. 6183 The default value is 0 (no limit). 6184 6185 sunrpc.pool_mode= 6186 [NFS] 6187 Control how the NFS server code allocates CPUs to 6188 service thread pools. Depending on how many NICs 6189 you have and where their interrupts are bound, this 6190 option will affect which CPUs will do NFS serving. 6191 Note: this parameter cannot be changed while the 6192 NFS server is running. 6193 6194 auto the server chooses an appropriate mode 6195 automatically using heuristics 6196 global a single global pool contains all CPUs 6197 percpu one pool for each CPU 6198 pernode one pool for each NUMA node (equivalent 6199 to global on non-NUMA machines) 6200 6201 sunrpc.tcp_slot_table_entries= 6202 sunrpc.udp_slot_table_entries= 6203 [NFS,SUNRPC] 6204 Sets the upper limit on the number of simultaneous 6205 RPC calls that can be sent from the client to a 6206 server. Increasing these values may allow you to 6207 improve throughput, but will also increase the 6208 amount of memory reserved for use by the client. 6209 6210 suspend.pm_test_delay= 6211 [SUSPEND] 6212 Sets the number of seconds to remain in a suspend test 6213 mode before resuming the system (see 6214 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG 6215 is set. Default value is 5. 6216 6217 svm= [PPC] 6218 Format: { on | off | y | n | 1 | 0 } 6219 This parameter controls use of the Protected 6220 Execution Facility on pSeries. 6221 6222 swiotlb= [ARM,IA-64,PPC,MIPS,X86] 6223 Format: { <int> [,<int>] | force | noforce } 6224 <int> -- Number of I/O TLB slabs 6225 <int> -- Second integer after comma. Number of swiotlb 6226 areas with their own lock. Will be rounded up 6227 to a power of 2. 6228 force -- force using of bounce buffers even if they 6229 wouldn't be automatically used by the kernel 6230 noforce -- Never use bounce buffers (for debugging) 6231 6232 switches= [HW,M68k] 6233 6234 sysctl.*= [KNL] 6235 Set a sysctl parameter, right before loading the init 6236 process, as if the value was written to the respective 6237 /proc/sys/... file. Both '.' and '/' are recognized as 6238 separators. Unrecognized parameters and invalid values 6239 are reported in the kernel log. Sysctls registered 6240 later by a loaded module cannot be set this way. 6241 Example: sysctl.vm.swappiness=40 6242 6243 sysrq_always_enabled 6244 [KNL] 6245 Ignore sysrq setting - this boot parameter will 6246 neutralize any effect of /proc/sys/kernel/sysrq. 6247 Useful for debugging. 6248 6249 tcpmhash_entries= [KNL,NET] 6250 Set the number of tcp_metrics_hash slots. 6251 Default value is 8192 or 16384 depending on total 6252 ram pages. This is used to specify the TCP metrics 6253 cache size. See Documentation/networking/ip-sysctl.rst 6254 "tcp_no_metrics_save" section for more details. 6255 6256 tdfx= [HW,DRM] 6257 6258 test_suspend= [SUSPEND] 6259 Format: { "mem" | "standby" | "freeze" }[,N] 6260 Specify "mem" (for Suspend-to-RAM) or "standby" (for 6261 standby suspend) or "freeze" (for suspend type freeze) 6262 as the system sleep state during system startup with 6263 the optional capability to repeat N number of times. 6264 The system is woken from this state using a 6265 wakeup-capable RTC alarm. 6266 6267 thash_entries= [KNL,NET] 6268 Set number of hash buckets for TCP connection 6269 6270 thermal.act= [HW,ACPI] 6271 -1: disable all active trip points in all thermal zones 6272 <degrees C>: override all lowest active trip points 6273 6274 thermal.crt= [HW,ACPI] 6275 -1: disable all critical trip points in all thermal zones 6276 <degrees C>: override all critical trip points 6277 6278 thermal.nocrt= [HW,ACPI] 6279 Set to disable actions on ACPI thermal zone 6280 critical and hot trip points. 6281 6282 thermal.off= [HW,ACPI] 6283 1: disable ACPI thermal control 6284 6285 thermal.psv= [HW,ACPI] 6286 -1: disable all passive trip points 6287 <degrees C>: override all passive trip points to this 6288 value 6289 6290 thermal.tzp= [HW,ACPI] 6291 Specify global default ACPI thermal zone polling rate 6292 <deci-seconds>: poll all this frequency 6293 0: no polling (default) 6294 6295 threadirqs [KNL] 6296 Force threading of all interrupt handlers except those 6297 marked explicitly IRQF_NO_THREAD. 6298 6299 topology= [S390] 6300 Format: {off | on} 6301 Specify if the kernel should make use of the cpu 6302 topology information if the hardware supports this. 6303 The scheduler will make use of this information and 6304 e.g. base its process migration decisions on it. 6305 Default is on. 6306 6307 topology_updates= [KNL, PPC, NUMA] 6308 Format: {off} 6309 Specify if the kernel should ignore (off) 6310 topology updates sent by the hypervisor to this 6311 LPAR. 6312 6313 torture.disable_onoff_at_boot= [KNL] 6314 Prevent the CPU-hotplug component of torturing 6315 until after init has spawned. 6316 6317 torture.ftrace_dump_at_shutdown= [KNL] 6318 Dump the ftrace buffer at torture-test shutdown, 6319 even if there were no errors. This can be a 6320 very costly operation when many torture tests 6321 are running concurrently, especially on systems 6322 with rotating-rust storage. 6323 6324 torture.verbose_sleep_frequency= [KNL] 6325 Specifies how many verbose printk()s should be 6326 emitted between each sleep. The default of zero 6327 disables verbose-printk() sleeping. 6328 6329 torture.verbose_sleep_duration= [KNL] 6330 Duration of each verbose-printk() sleep in jiffies. 6331 6332 tp720= [HW,PS2] 6333 6334 tpm_suspend_pcr=[HW,TPM] 6335 Format: integer pcr id 6336 Specify that at suspend time, the tpm driver 6337 should extend the specified pcr with zeros, 6338 as a workaround for some chips which fail to 6339 flush the last written pcr on TPM_SaveState. 6340 This will guarantee that all the other pcrs 6341 are saved. 6342 6343 tp_printk [FTRACE] 6344 Have the tracepoints sent to printk as well as the 6345 tracing ring buffer. This is useful for early boot up 6346 where the system hangs or reboots and does not give the 6347 option for reading the tracing buffer or performing a 6348 ftrace_dump_on_oops. 6349 6350 To turn off having tracepoints sent to printk, 6351 echo 0 > /proc/sys/kernel/tracepoint_printk 6352 Note, echoing 1 into this file without the 6353 tracepoint_printk kernel cmdline option has no effect. 6354 6355 The tp_printk_stop_on_boot (see below) can also be used 6356 to stop the printing of events to console at 6357 late_initcall_sync. 6358 6359 ** CAUTION ** 6360 6361 Having tracepoints sent to printk() and activating high 6362 frequency tracepoints such as irq or sched, can cause 6363 the system to live lock. 6364 6365 tp_printk_stop_on_boot [FTRACE] 6366 When tp_printk (above) is set, it can cause a lot of noise 6367 on the console. It may be useful to only include the 6368 printing of events during boot up, as user space may 6369 make the system inoperable. 6370 6371 This command line option will stop the printing of events 6372 to console at the late_initcall_sync() time frame. 6373 6374 trace_buf_size=nn[KMG] 6375 [FTRACE] will set tracing buffer size on each cpu. 6376 6377 trace_clock= [FTRACE] Set the clock used for tracing events 6378 at boot up. 6379 local - Use the per CPU time stamp counter 6380 (converted into nanoseconds). Fast, but 6381 depending on the architecture, may not be 6382 in sync between CPUs. 6383 global - Event time stamps are synchronize across 6384 CPUs. May be slower than the local clock, 6385 but better for some race conditions. 6386 counter - Simple counting of events (1, 2, ..) 6387 note, some counts may be skipped due to the 6388 infrastructure grabbing the clock more than 6389 once per event. 6390 uptime - Use jiffies as the time stamp. 6391 perf - Use the same clock that perf uses. 6392 mono - Use ktime_get_mono_fast_ns() for time stamps. 6393 mono_raw - Use ktime_get_raw_fast_ns() for time 6394 stamps. 6395 boot - Use ktime_get_boot_fast_ns() for time stamps. 6396 Architectures may add more clocks. See 6397 Documentation/trace/ftrace.rst for more details. 6398 6399 trace_event=[event-list] 6400 [FTRACE] Set and start specified trace events in order 6401 to facilitate early boot debugging. The event-list is a 6402 comma-separated list of trace events to enable. See 6403 also Documentation/trace/events.rst 6404 6405 trace_instance=[instance-info] 6406 [FTRACE] Create a ring buffer instance early in boot up. 6407 This will be listed in: 6408 6409 /sys/kernel/tracing/instances 6410 6411 Events can be enabled at the time the instance is created 6412 via: 6413 6414 trace_instance=<name>,<system1>:<event1>,<system2>:<event2> 6415 6416 Note, the "<system*>:" portion is optional if the event is 6417 unique. 6418 6419 trace_instance=foo,sched:sched_switch,irq_handler_entry,initcall 6420 6421 will enable the "sched_switch" event (note, the "sched:" is optional, and 6422 the same thing would happen if it was left off). The irq_handler_entry 6423 event, and all events under the "initcall" system. 6424 6425 trace_options=[option-list] 6426 [FTRACE] Enable or disable tracer options at boot. 6427 The option-list is a comma delimited list of options 6428 that can be enabled or disabled just as if you were 6429 to echo the option name into 6430 6431 /sys/kernel/tracing/trace_options 6432 6433 For example, to enable stacktrace option (to dump the 6434 stack trace of each event), add to the command line: 6435 6436 trace_options=stacktrace 6437 6438 See also Documentation/trace/ftrace.rst "trace options" 6439 section. 6440 6441 trace_trigger=[trigger-list] 6442 [FTRACE] Add a event trigger on specific events. 6443 Set a trigger on top of a specific event, with an optional 6444 filter. 6445 6446 The format is is "trace_trigger=<event>.<trigger>[ if <filter>],..." 6447 Where more than one trigger may be specified that are comma deliminated. 6448 6449 For example: 6450 6451 trace_trigger="sched_switch.stacktrace if prev_state == 2" 6452 6453 The above will enable the "stacktrace" trigger on the "sched_switch" 6454 event but only trigger it if the "prev_state" of the "sched_switch" 6455 event is "2" (TASK_UNINTERUPTIBLE). 6456 6457 See also "Event triggers" in Documentation/trace/events.rst 6458 6459 6460 traceoff_on_warning 6461 [FTRACE] enable this option to disable tracing when a 6462 warning is hit. This turns off "tracing_on". Tracing can 6463 be enabled again by echoing '1' into the "tracing_on" 6464 file located in /sys/kernel/tracing/ 6465 6466 This option is useful, as it disables the trace before 6467 the WARNING dump is called, which prevents the trace to 6468 be filled with content caused by the warning output. 6469 6470 This option can also be set at run time via the sysctl 6471 option: kernel/traceoff_on_warning 6472 6473 transparent_hugepage= 6474 [KNL] 6475 Format: [always|madvise|never] 6476 Can be used to control the default behavior of the system 6477 with respect to transparent hugepages. 6478 See Documentation/admin-guide/mm/transhuge.rst 6479 for more details. 6480 6481 trusted.source= [KEYS] 6482 Format: <string> 6483 This parameter identifies the trust source as a backend 6484 for trusted keys implementation. Supported trust 6485 sources: 6486 - "tpm" 6487 - "tee" 6488 - "caam" 6489 If not specified then it defaults to iterating through 6490 the trust source list starting with TPM and assigns the 6491 first trust source as a backend which is initialized 6492 successfully during iteration. 6493 6494 trusted.rng= [KEYS] 6495 Format: <string> 6496 The RNG used to generate key material for trusted keys. 6497 Can be one of: 6498 - "kernel" 6499 - the same value as trusted.source: "tpm" or "tee" 6500 - "default" 6501 If not specified, "default" is used. In this case, 6502 the RNG's choice is left to each individual trust source. 6503 6504 tsc= Disable clocksource stability checks for TSC. 6505 Format: <string> 6506 [x86] reliable: mark tsc clocksource as reliable, this 6507 disables clocksource verification at runtime, as well 6508 as the stability checks done at bootup. Used to enable 6509 high-resolution timer mode on older hardware, and in 6510 virtualized environment. 6511 [x86] noirqtime: Do not use TSC to do irq accounting. 6512 Used to run time disable IRQ_TIME_ACCOUNTING on any 6513 platforms where RDTSC is slow and this accounting 6514 can add overhead. 6515 [x86] unstable: mark the TSC clocksource as unstable, this 6516 marks the TSC unconditionally unstable at bootup and 6517 avoids any further wobbles once the TSC watchdog notices. 6518 [x86] nowatchdog: disable clocksource watchdog. Used 6519 in situations with strict latency requirements (where 6520 interruptions from clocksource watchdog are not 6521 acceptable). 6522 [x86] recalibrate: force recalibration against a HW timer 6523 (HPET or PM timer) on systems whose TSC frequency was 6524 obtained from HW or FW using either an MSR or CPUID(0x15). 6525 Warn if the difference is more than 500 ppm. 6526 [x86] watchdog: Use TSC as the watchdog clocksource with 6527 which to check other HW timers (HPET or PM timer), but 6528 only on systems where TSC has been deemed trustworthy. 6529 This will be suppressed by an earlier tsc=nowatchdog and 6530 can be overridden by a later tsc=nowatchdog. A console 6531 message will flag any such suppression or overriding. 6532 6533 tsc_early_khz= [X86] Skip early TSC calibration and use the given 6534 value instead. Useful when the early TSC frequency discovery 6535 procedure is not reliable, such as on overclocked systems 6536 with CPUID.16h support and partial CPUID.15h support. 6537 Format: <unsigned int> 6538 6539 tsx= [X86] Control Transactional Synchronization 6540 Extensions (TSX) feature in Intel processors that 6541 support TSX control. 6542 6543 This parameter controls the TSX feature. The options are: 6544 6545 on - Enable TSX on the system. Although there are 6546 mitigations for all known security vulnerabilities, 6547 TSX has been known to be an accelerator for 6548 several previous speculation-related CVEs, and 6549 so there may be unknown security risks associated 6550 with leaving it enabled. 6551 6552 off - Disable TSX on the system. (Note that this 6553 option takes effect only on newer CPUs which are 6554 not vulnerable to MDS, i.e., have 6555 MSR_IA32_ARCH_CAPABILITIES.MDS_NO=1 and which get 6556 the new IA32_TSX_CTRL MSR through a microcode 6557 update. This new MSR allows for the reliable 6558 deactivation of the TSX functionality.) 6559 6560 auto - Disable TSX if X86_BUG_TAA is present, 6561 otherwise enable TSX on the system. 6562 6563 Not specifying this option is equivalent to tsx=off. 6564 6565 See Documentation/admin-guide/hw-vuln/tsx_async_abort.rst 6566 for more details. 6567 6568 tsx_async_abort= [X86,INTEL] Control mitigation for the TSX Async 6569 Abort (TAA) vulnerability. 6570 6571 Similar to Micro-architectural Data Sampling (MDS) 6572 certain CPUs that support Transactional 6573 Synchronization Extensions (TSX) are vulnerable to an 6574 exploit against CPU internal buffers which can forward 6575 information to a disclosure gadget under certain 6576 conditions. 6577 6578 In vulnerable processors, the speculatively forwarded 6579 data can be used in a cache side channel attack, to 6580 access data to which the attacker does not have direct 6581 access. 6582 6583 This parameter controls the TAA mitigation. The 6584 options are: 6585 6586 full - Enable TAA mitigation on vulnerable CPUs 6587 if TSX is enabled. 6588 6589 full,nosmt - Enable TAA mitigation and disable SMT on 6590 vulnerable CPUs. If TSX is disabled, SMT 6591 is not disabled because CPU is not 6592 vulnerable to cross-thread TAA attacks. 6593 off - Unconditionally disable TAA mitigation 6594 6595 On MDS-affected machines, tsx_async_abort=off can be 6596 prevented by an active MDS mitigation as both vulnerabilities 6597 are mitigated with the same mechanism so in order to disable 6598 this mitigation, you need to specify mds=off too. 6599 6600 Not specifying this option is equivalent to 6601 tsx_async_abort=full. On CPUs which are MDS affected 6602 and deploy MDS mitigation, TAA mitigation is not 6603 required and doesn't provide any additional 6604 mitigation. 6605 6606 For details see: 6607 Documentation/admin-guide/hw-vuln/tsx_async_abort.rst 6608 6609 turbografx.map[2|3]= [HW,JOY] 6610 TurboGraFX parallel port interface 6611 Format: 6612 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7> 6613 See also Documentation/input/devices/joystick-parport.rst 6614 6615 udbg-immortal [PPC] When debugging early kernel crashes that 6616 happen after console_init() and before a proper 6617 console driver takes over, this boot options might 6618 help "seeing" what's going on. 6619 6620 uhash_entries= [KNL,NET] 6621 Set number of hash buckets for UDP/UDP-Lite connections 6622 6623 uhci-hcd.ignore_oc= 6624 [USB] Ignore overcurrent events (default N). 6625 Some badly-designed motherboards generate lots of 6626 bogus events, for ports that aren't wired to 6627 anything. Set this parameter to avoid log spamming. 6628 Note that genuine overcurrent events won't be 6629 reported either. 6630 6631 unknown_nmi_panic 6632 [X86] Cause panic on unknown NMI. 6633 6634 unwind_debug [X86-64] 6635 Enable unwinder debug output. This can be 6636 useful for debugging certain unwinder error 6637 conditions, including corrupt stacks and 6638 bad/missing unwinder metadata. 6639 6640 usbcore.authorized_default= 6641 [USB] Default USB device authorization: 6642 (default -1 = authorized except for wireless USB, 6643 0 = not authorized, 1 = authorized, 2 = authorized 6644 if device connected to internal port) 6645 6646 usbcore.autosuspend= 6647 [USB] The autosuspend time delay (in seconds) used 6648 for newly-detected USB devices (default 2). This 6649 is the time required before an idle device will be 6650 autosuspended. Devices for which the delay is set 6651 to a negative value won't be autosuspended at all. 6652 6653 usbcore.usbfs_snoop= 6654 [USB] Set to log all usbfs traffic (default 0 = off). 6655 6656 usbcore.usbfs_snoop_max= 6657 [USB] Maximum number of bytes to snoop in each URB 6658 (default = 65536). 6659 6660 usbcore.blinkenlights= 6661 [USB] Set to cycle leds on hubs (default 0 = off). 6662 6663 usbcore.old_scheme_first= 6664 [USB] Start with the old device initialization 6665 scheme (default 0 = off). 6666 6667 usbcore.usbfs_memory_mb= 6668 [USB] Memory limit (in MB) for buffers allocated by 6669 usbfs (default = 16, 0 = max = 2047). 6670 6671 usbcore.use_both_schemes= 6672 [USB] Try the other device initialization scheme 6673 if the first one fails (default 1 = enabled). 6674 6675 usbcore.initial_descriptor_timeout= 6676 [USB] Specifies timeout for the initial 64-byte 6677 USB_REQ_GET_DESCRIPTOR request in milliseconds 6678 (default 5000 = 5.0 seconds). 6679 6680 usbcore.nousb [USB] Disable the USB subsystem 6681 6682 usbcore.quirks= 6683 [USB] A list of quirk entries to augment the built-in 6684 usb core quirk list. List entries are separated by 6685 commas. Each entry has the form 6686 VendorID:ProductID:Flags. The IDs are 4-digit hex 6687 numbers and Flags is a set of letters. Each letter 6688 will change the built-in quirk; setting it if it is 6689 clear and clearing it if it is set. The letters have 6690 the following meanings: 6691 a = USB_QUIRK_STRING_FETCH_255 (string 6692 descriptors must not be fetched using 6693 a 255-byte read); 6694 b = USB_QUIRK_RESET_RESUME (device can't resume 6695 correctly so reset it instead); 6696 c = USB_QUIRK_NO_SET_INTF (device can't handle 6697 Set-Interface requests); 6698 d = USB_QUIRK_CONFIG_INTF_STRINGS (device can't 6699 handle its Configuration or Interface 6700 strings); 6701 e = USB_QUIRK_RESET (device can't be reset 6702 (e.g morph devices), don't use reset); 6703 f = USB_QUIRK_HONOR_BNUMINTERFACES (device has 6704 more interface descriptions than the 6705 bNumInterfaces count, and can't handle 6706 talking to these interfaces); 6707 g = USB_QUIRK_DELAY_INIT (device needs a pause 6708 during initialization, after we read 6709 the device descriptor); 6710 h = USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL (For 6711 high speed and super speed interrupt 6712 endpoints, the USB 2.0 and USB 3.0 spec 6713 require the interval in microframes (1 6714 microframe = 125 microseconds) to be 6715 calculated as interval = 2 ^ 6716 (bInterval-1). 6717 Devices with this quirk report their 6718 bInterval as the result of this 6719 calculation instead of the exponent 6720 variable used in the calculation); 6721 i = USB_QUIRK_DEVICE_QUALIFIER (device can't 6722 handle device_qualifier descriptor 6723 requests); 6724 j = USB_QUIRK_IGNORE_REMOTE_WAKEUP (device 6725 generates spurious wakeup, ignore 6726 remote wakeup capability); 6727 k = USB_QUIRK_NO_LPM (device can't handle Link 6728 Power Management); 6729 l = USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL 6730 (Device reports its bInterval as linear 6731 frames instead of the USB 2.0 6732 calculation); 6733 m = USB_QUIRK_DISCONNECT_SUSPEND (Device needs 6734 to be disconnected before suspend to 6735 prevent spurious wakeup); 6736 n = USB_QUIRK_DELAY_CTRL_MSG (Device needs a 6737 pause after every control message); 6738 o = USB_QUIRK_HUB_SLOW_RESET (Hub needs extra 6739 delay after resetting its port); 6740 Example: quirks=0781:5580:bk,0a5c:5834:gij 6741 6742 usbhid.mousepoll= 6743 [USBHID] The interval which mice are to be polled at. 6744 6745 usbhid.jspoll= 6746 [USBHID] The interval which joysticks are to be polled at. 6747 6748 usbhid.kbpoll= 6749 [USBHID] The interval which keyboards are to be polled at. 6750 6751 usb-storage.delay_use= 6752 [UMS] The delay in seconds before a new device is 6753 scanned for Logical Units (default 1). 6754 6755 usb-storage.quirks= 6756 [UMS] A list of quirks entries to supplement or 6757 override the built-in unusual_devs list. List 6758 entries are separated by commas. Each entry has 6759 the form VID:PID:Flags where VID and PID are Vendor 6760 and Product ID values (4-digit hex numbers) and 6761 Flags is a set of characters, each corresponding 6762 to a common usb-storage quirk flag as follows: 6763 a = SANE_SENSE (collect more than 18 bytes 6764 of sense data, not on uas); 6765 b = BAD_SENSE (don't collect more than 18 6766 bytes of sense data, not on uas); 6767 c = FIX_CAPACITY (decrease the reported 6768 device capacity by one sector); 6769 d = NO_READ_DISC_INFO (don't use 6770 READ_DISC_INFO command, not on uas); 6771 e = NO_READ_CAPACITY_16 (don't use 6772 READ_CAPACITY_16 command); 6773 f = NO_REPORT_OPCODES (don't use report opcodes 6774 command, uas only); 6775 g = MAX_SECTORS_240 (don't transfer more than 6776 240 sectors at a time, uas only); 6777 h = CAPACITY_HEURISTICS (decrease the 6778 reported device capacity by one 6779 sector if the number is odd); 6780 i = IGNORE_DEVICE (don't bind to this 6781 device); 6782 j = NO_REPORT_LUNS (don't use report luns 6783 command, uas only); 6784 k = NO_SAME (do not use WRITE_SAME, uas only) 6785 l = NOT_LOCKABLE (don't try to lock and 6786 unlock ejectable media, not on uas); 6787 m = MAX_SECTORS_64 (don't transfer more 6788 than 64 sectors = 32 KB at a time, 6789 not on uas); 6790 n = INITIAL_READ10 (force a retry of the 6791 initial READ(10) command, not on uas); 6792 o = CAPACITY_OK (accept the capacity 6793 reported by the device, not on uas); 6794 p = WRITE_CACHE (the device cache is ON 6795 by default, not on uas); 6796 r = IGNORE_RESIDUE (the device reports 6797 bogus residue values, not on uas); 6798 s = SINGLE_LUN (the device has only one 6799 Logical Unit); 6800 t = NO_ATA_1X (don't allow ATA(12) and ATA(16) 6801 commands, uas only); 6802 u = IGNORE_UAS (don't bind to the uas driver); 6803 w = NO_WP_DETECT (don't test whether the 6804 medium is write-protected). 6805 y = ALWAYS_SYNC (issue a SYNCHRONIZE_CACHE 6806 even if the device claims no cache, 6807 not on uas) 6808 Example: quirks=0419:aaf5:rl,0421:0433:rc 6809 6810 user_debug= [KNL,ARM] 6811 Format: <int> 6812 See arch/arm/Kconfig.debug help text. 6813 1 - undefined instruction events 6814 2 - system calls 6815 4 - invalid data aborts 6816 8 - SIGSEGV faults 6817 16 - SIGBUS faults 6818 Example: user_debug=31 6819 6820 userpte= 6821 [X86] Flags controlling user PTE allocations. 6822 6823 nohigh = do not allocate PTE pages in 6824 HIGHMEM regardless of setting 6825 of CONFIG_HIGHPTE. 6826 6827 vdso= [X86,SH,SPARC] 6828 On X86_32, this is an alias for vdso32=. Otherwise: 6829 6830 vdso=1: enable VDSO (the default) 6831 vdso=0: disable VDSO mapping 6832 6833 vdso32= [X86] Control the 32-bit vDSO 6834 vdso32=1: enable 32-bit VDSO 6835 vdso32=0 or vdso32=2: disable 32-bit VDSO 6836 6837 See the help text for CONFIG_COMPAT_VDSO for more 6838 details. If CONFIG_COMPAT_VDSO is set, the default is 6839 vdso32=0; otherwise, the default is vdso32=1. 6840 6841 For compatibility with older kernels, vdso32=2 is an 6842 alias for vdso32=0. 6843 6844 Try vdso32=0 if you encounter an error that says: 6845 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed! 6846 6847 vector= [IA-64,SMP] 6848 vector=percpu: enable percpu vector domain 6849 6850 video= [FB] Frame buffer configuration 6851 See Documentation/fb/modedb.rst. 6852 6853 video.brightness_switch_enabled= [ACPI] 6854 Format: [0|1] 6855 If set to 1, on receiving an ACPI notify event 6856 generated by hotkey, video driver will adjust brightness 6857 level and then send out the event to user space through 6858 the allocated input device. If set to 0, video driver 6859 will only send out the event without touching backlight 6860 brightness level. 6861 default: 1 6862 6863 virtio_mmio.device= 6864 [VMMIO] Memory mapped virtio (platform) device. 6865 6866 <size>@<baseaddr>:<irq>[:<id>] 6867 where: 6868 <size> := size (can use standard suffixes 6869 like K, M and G) 6870 <baseaddr> := physical base address 6871 <irq> := interrupt number (as passed to 6872 request_irq()) 6873 <id> := (optional) platform device id 6874 example: 6875 virtio_mmio.device=1K@0x100b0000:48:7 6876 6877 Can be used multiple times for multiple devices. 6878 6879 vga= [BOOT,X86-32] Select a particular video mode 6880 See Documentation/arch/x86/boot.rst and 6881 Documentation/admin-guide/svga.rst. 6882 Use vga=ask for menu. 6883 This is actually a boot loader parameter; the value is 6884 passed to the kernel using a special protocol. 6885 6886 vm_debug[=options] [KNL] Available with CONFIG_DEBUG_VM=y. 6887 May slow down system boot speed, especially when 6888 enabled on systems with a large amount of memory. 6889 All options are enabled by default, and this 6890 interface is meant to allow for selectively 6891 enabling or disabling specific virtual memory 6892 debugging features. 6893 6894 Available options are: 6895 P Enable page structure init time poisoning 6896 - Disable all of the above options 6897 6898 vmalloc=nn[KMG] [KNL,BOOT] Forces the vmalloc area to have an exact 6899 size of <nn>. This can be used to increase the 6900 minimum size (128MB on x86). It can also be used to 6901 decrease the size and leave more room for directly 6902 mapped kernel RAM. 6903 6904 vmcp_cma=nn[MG] [KNL,S390] 6905 Sets the memory size reserved for contiguous memory 6906 allocations for the vmcp device driver. 6907 6908 vmhalt= [KNL,S390] Perform z/VM CP command after system halt. 6909 Format: <command> 6910 6911 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic. 6912 Format: <command> 6913 6914 vmpoff= [KNL,S390] Perform z/VM CP command after power off. 6915 Format: <command> 6916 6917 vsyscall= [X86-64] 6918 Controls the behavior of vsyscalls (i.e. calls to 6919 fixed addresses of 0xffffffffff600x00 from legacy 6920 code). Most statically-linked binaries and older 6921 versions of glibc use these calls. Because these 6922 functions are at fixed addresses, they make nice 6923 targets for exploits that can control RIP. 6924 6925 emulate Vsyscalls turn into traps and are emulated 6926 reasonably safely. The vsyscall page is 6927 readable. 6928 6929 xonly [default] Vsyscalls turn into traps and are 6930 emulated reasonably safely. The vsyscall 6931 page is not readable. 6932 6933 none Vsyscalls don't work at all. This makes 6934 them quite hard to use for exploits but 6935 might break your system. 6936 6937 vt.color= [VT] Default text color. 6938 Format: 0xYX, X = foreground, Y = background. 6939 Default: 0x07 = light gray on black. 6940 6941 vt.cur_default= [VT] Default cursor shape. 6942 Format: 0xCCBBAA, where AA, BB, and CC are the same as 6943 the parameters of the <Esc>[?A;B;Cc escape sequence; 6944 see VGA-softcursor.txt. Default: 2 = underline. 6945 6946 vt.default_blu= [VT] 6947 Format: <blue0>,<blue1>,<blue2>,...,<blue15> 6948 Change the default blue palette of the console. 6949 This is a 16-member array composed of values 6950 ranging from 0-255. 6951 6952 vt.default_grn= [VT] 6953 Format: <green0>,<green1>,<green2>,...,<green15> 6954 Change the default green palette of the console. 6955 This is a 16-member array composed of values 6956 ranging from 0-255. 6957 6958 vt.default_red= [VT] 6959 Format: <red0>,<red1>,<red2>,...,<red15> 6960 Change the default red palette of the console. 6961 This is a 16-member array composed of values 6962 ranging from 0-255. 6963 6964 vt.default_utf8= 6965 [VT] 6966 Format=<0|1> 6967 Set system-wide default UTF-8 mode for all tty's. 6968 Default is 1, i.e. UTF-8 mode is enabled for all 6969 newly opened terminals. 6970 6971 vt.global_cursor_default= 6972 [VT] 6973 Format=<-1|0|1> 6974 Set system-wide default for whether a cursor 6975 is shown on new VTs. Default is -1, 6976 i.e. cursors will be created by default unless 6977 overridden by individual drivers. 0 will hide 6978 cursors, 1 will display them. 6979 6980 vt.italic= [VT] Default color for italic text; 0-15. 6981 Default: 2 = green. 6982 6983 vt.underline= [VT] Default color for underlined text; 0-15. 6984 Default: 3 = cyan. 6985 6986 watchdog timers [HW,WDT] For information on watchdog timers, 6987 see Documentation/watchdog/watchdog-parameters.rst 6988 or other driver-specific files in the 6989 Documentation/watchdog/ directory. 6990 6991 watchdog_thresh= 6992 [KNL] 6993 Set the hard lockup detector stall duration 6994 threshold in seconds. The soft lockup detector 6995 threshold is set to twice the value. A value of 0 6996 disables both lockup detectors. Default is 10 6997 seconds. 6998 6999 workqueue.watchdog_thresh= 7000 If CONFIG_WQ_WATCHDOG is configured, workqueue can 7001 warn stall conditions and dump internal state to 7002 help debugging. 0 disables workqueue stall 7003 detection; otherwise, it's the stall threshold 7004 duration in seconds. The default value is 30 and 7005 it can be updated at runtime by writing to the 7006 corresponding sysfs file. 7007 7008 workqueue.cpu_intensive_thresh_us= 7009 Per-cpu work items which run for longer than this 7010 threshold are automatically considered CPU intensive 7011 and excluded from concurrency management to prevent 7012 them from noticeably delaying other per-cpu work 7013 items. Default is 10000 (10ms). 7014 7015 If CONFIG_WQ_CPU_INTENSIVE_REPORT is set, the kernel 7016 will report the work functions which violate this 7017 threshold repeatedly. They are likely good 7018 candidates for using WQ_UNBOUND workqueues instead. 7019 7020 workqueue.disable_numa 7021 By default, all work items queued to unbound 7022 workqueues are affine to the NUMA nodes they're 7023 issued on, which results in better behavior in 7024 general. If NUMA affinity needs to be disabled for 7025 whatever reason, this option can be used. Note 7026 that this also can be controlled per-workqueue for 7027 workqueues visible under /sys/bus/workqueue/. 7028 7029 workqueue.power_efficient 7030 Per-cpu workqueues are generally preferred because 7031 they show better performance thanks to cache 7032 locality; unfortunately, per-cpu workqueues tend to 7033 be more power hungry than unbound workqueues. 7034 7035 Enabling this makes the per-cpu workqueues which 7036 were observed to contribute significantly to power 7037 consumption unbound, leading to measurably lower 7038 power usage at the cost of small performance 7039 overhead. 7040 7041 The default value of this parameter is determined by 7042 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT. 7043 7044 workqueue.debug_force_rr_cpu 7045 Workqueue used to implicitly guarantee that work 7046 items queued without explicit CPU specified are put 7047 on the local CPU. This guarantee is no longer true 7048 and while local CPU is still preferred work items 7049 may be put on foreign CPUs. This debug option 7050 forces round-robin CPU selection to flush out 7051 usages which depend on the now broken guarantee. 7052 When enabled, memory and cache locality will be 7053 impacted. 7054 7055 writecombine= [LOONGARCH] Control the MAT (Memory Access Type) of 7056 ioremap_wc(). 7057 7058 on - Enable writecombine, use WUC for ioremap_wc() 7059 off - Disable writecombine, use SUC for ioremap_wc() 7060 7061 x2apic_phys [X86-64,APIC] Use x2apic physical mode instead of 7062 default x2apic cluster mode on platforms 7063 supporting x2apic. 7064 7065 xen_512gb_limit [KNL,X86-64,XEN] 7066 Restricts the kernel running paravirtualized under Xen 7067 to use only up to 512 GB of RAM. The reason to do so is 7068 crash analysis tools and Xen tools for doing domain 7069 save/restore/migration must be enabled to handle larger 7070 domains. 7071 7072 xen_emul_unplug= [HW,X86,XEN] 7073 Unplug Xen emulated devices 7074 Format: [unplug0,][unplug1] 7075 ide-disks -- unplug primary master IDE devices 7076 aux-ide-disks -- unplug non-primary-master IDE devices 7077 nics -- unplug network devices 7078 all -- unplug all emulated devices (NICs and IDE disks) 7079 unnecessary -- unplugging emulated devices is 7080 unnecessary even if the host did not respond to 7081 the unplug protocol 7082 never -- do not unplug even if version check succeeds 7083 7084 xen_legacy_crash [X86,XEN] 7085 Crash from Xen panic notifier, without executing late 7086 panic() code such as dumping handler. 7087 7088 xen_msr_safe= [X86,XEN] 7089 Format: <bool> 7090 Select whether to always use non-faulting (safe) MSR 7091 access functions when running as Xen PV guest. The 7092 default value is controlled by CONFIG_XEN_PV_MSR_SAFE. 7093 7094 xen_nopvspin [X86,XEN] 7095 Disables the qspinlock slowpath using Xen PV optimizations. 7096 This parameter is obsoleted by "nopvspin" parameter, which 7097 has equivalent effect for XEN platform. 7098 7099 xen_nopv [X86] 7100 Disables the PV optimizations forcing the HVM guest to 7101 run as generic HVM guest with no PV drivers. 7102 This option is obsoleted by the "nopv" option, which 7103 has equivalent effect for XEN platform. 7104 7105 xen_no_vector_callback 7106 [KNL,X86,XEN] Disable the vector callback for Xen 7107 event channel interrupts. 7108 7109 xen_scrub_pages= [XEN] 7110 Boolean option to control scrubbing pages before giving them back 7111 to Xen, for use by other domains. Can be also changed at runtime 7112 with /sys/devices/system/xen_memory/xen_memory0/scrub_pages. 7113 Default value controlled with CONFIG_XEN_SCRUB_PAGES_DEFAULT. 7114 7115 xen_timer_slop= [X86-64,XEN] 7116 Set the timer slop (in nanoseconds) for the virtual Xen 7117 timers (default is 100000). This adjusts the minimum 7118 delta of virtualized Xen timers, where lower values 7119 improve timer resolution at the expense of processing 7120 more timer interrupts. 7121 7122 xen.balloon_boot_timeout= [XEN] 7123 The time (in seconds) to wait before giving up to boot 7124 in case initial ballooning fails to free enough memory. 7125 Applies only when running as HVM or PVH guest and 7126 started with less memory configured than allowed at 7127 max. Default is 180. 7128 7129 xen.event_eoi_delay= [XEN] 7130 How long to delay EOI handling in case of event 7131 storms (jiffies). Default is 10. 7132 7133 xen.event_loop_timeout= [XEN] 7134 After which time (jiffies) the event handling loop 7135 should start to delay EOI handling. Default is 2. 7136 7137 xen.fifo_events= [XEN] 7138 Boolean parameter to disable using fifo event handling 7139 even if available. Normally fifo event handling is 7140 preferred over the 2-level event handling, as it is 7141 fairer and the number of possible event channels is 7142 much higher. Default is on (use fifo events). 7143 7144 xirc2ps_cs= [NET,PCMCIA] 7145 Format: 7146 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]] 7147 7148 xive= [PPC] 7149 By default on POWER9 and above, the kernel will 7150 natively use the XIVE interrupt controller. This option 7151 allows the fallback firmware mode to be used: 7152 7153 off Fallback to firmware control of XIVE interrupt 7154 controller on both pseries and powernv 7155 platforms. Only useful on POWER9 and above. 7156 7157 xive.store-eoi=off [PPC] 7158 By default on POWER10 and above, the kernel will use 7159 stores for EOI handling when the XIVE interrupt mode 7160 is active. This option allows the XIVE driver to use 7161 loads instead, as on POWER9. 7162 7163 xhci-hcd.quirks [USB,KNL] 7164 A hex value specifying bitmask with supplemental xhci 7165 host controller quirks. Meaning of each bit can be 7166 consulted in header drivers/usb/host/xhci.h. 7167 7168 xmon [PPC] 7169 Format: { early | on | rw | ro | off } 7170 Controls if xmon debugger is enabled. Default is off. 7171 Passing only "xmon" is equivalent to "xmon=early". 7172 early Call xmon as early as possible on boot; xmon 7173 debugger is called from setup_arch(). 7174 on xmon debugger hooks will be installed so xmon 7175 is only called on a kernel crash. Default mode, 7176 i.e. either "ro" or "rw" mode, is controlled 7177 with CONFIG_XMON_DEFAULT_RO_MODE. 7178 rw xmon debugger hooks will be installed so xmon 7179 is called only on a kernel crash, mode is write, 7180 meaning SPR registers, memory and, other data 7181 can be written using xmon commands. 7182 ro same as "rw" option above but SPR registers, 7183 memory, and other data can't be written using 7184 xmon commands. 7185 off xmon is disabled. 7186 7187