1 ACPI ACPI support is enabled. 2 AGP AGP (Accelerated Graphics Port) is enabled. 3 ALSA ALSA sound support is enabled. 4 APIC APIC support is enabled. 5 APM Advanced Power Management support is enabled. 6 APPARMOR AppArmor support is enabled. 7 ARM ARM architecture is enabled. 8 ARM64 ARM64 architecture is enabled. 9 CLK Common clock infrastructure is enabled. 10 CMA Contiguous Memory Area support is enabled. 11 DRM Direct Rendering Management support is enabled. 12 DYNAMIC_DEBUG Build in debug messages and enable them at runtime 13 EARLY Parameter processed too early to be embedded in initrd. 14 EDD BIOS Enhanced Disk Drive Services (EDD) is enabled 15 EFI EFI Partitioning (GPT) is enabled 16 EVM Extended Verification Module 17 FB The frame buffer device is enabled. 18 FTRACE Function tracing enabled. 19 GCOV GCOV profiling is enabled. 20 HIBERNATION HIBERNATION is enabled. 21 HW Appropriate hardware is enabled. 22 HYPER_V HYPERV support is enabled. 23 IMA Integrity measurement architecture is enabled. 24 IP_PNP IP DHCP, BOOTP, or RARP is enabled. 25 IPV6 IPv6 support is enabled. 26 ISAPNP ISA PnP code is enabled. 27 ISOL CPU Isolation is enabled. 28 JOY Appropriate joystick support is enabled. 29 KGDB Kernel debugger support is enabled. 30 KVM Kernel Virtual Machine support is enabled. 31 LIBATA Libata driver is enabled 32 LOONGARCH LoongArch architecture is enabled. 33 LOOP Loopback device support is enabled. 34 LP Printer support is enabled. 35 M68k M68k architecture is enabled. 36 These options have more detailed description inside of 37 Documentation/arch/m68k/kernel-options.rst. 38 MIPS MIPS architecture is enabled. 39 MOUSE Appropriate mouse support is enabled. 40 MSI Message Signaled Interrupts (PCI). 41 MTD MTD (Memory Technology Device) support is enabled. 42 NET Appropriate network support is enabled. 43 NFS Appropriate NFS support is enabled. 44 NUMA NUMA support is enabled. 45 OF Devicetree is enabled. 46 PARISC The PA-RISC architecture is enabled. 47 PCI PCI bus support is enabled. 48 PCIE PCI Express support is enabled. 49 PCMCIA The PCMCIA subsystem is enabled. 50 PNP Plug & Play support is enabled. 51 PPC PowerPC architecture is enabled. 52 PPT Parallel port support is enabled. 53 PS2 Appropriate PS/2 support is enabled. 54 PV_OPS A paravirtualized kernel is enabled. 55 RAM RAM disk support is enabled. 56 RDT Intel Resource Director Technology. 57 RISCV RISCV architecture is enabled. 58 S390 S390 architecture is enabled. 59 SCSI Appropriate SCSI support is enabled. 60 A lot of drivers have their options described inside 61 the Documentation/scsi/ sub-directory. 62 SDW SoundWire support is enabled. 63 SECURITY Different security models are enabled. 64 SELINUX SELinux support is enabled. 65 SERIAL Serial support is enabled. 66 SH SuperH architecture is enabled. 67 SMP The kernel is an SMP kernel. 68 SPARC Sparc architecture is enabled. 69 SUSPEND System suspend states are enabled. 70 SWSUSP Software suspend (hibernation) is enabled. 71 TPM TPM drivers are enabled. 72 UMS USB Mass Storage support is enabled. 73 USB USB support is enabled. 74 NVME NVMe support is enabled 75 USBHID USB Human Interface Device support is enabled. 76 V4L Video For Linux support is enabled. 77 VGA The VGA console has been enabled. 78 VMMIO Driver for memory mapped virtio devices is enabled. 79 VT Virtual terminal support is enabled. 80 WDT Watchdog support is enabled. 81 X86-32 X86-32, aka i386 architecture is enabled. 82 X86-64 X86-64 architecture is enabled. 83 X86 Either 32-bit or 64-bit x86 (same as X86-32+X86-64) 84 X86_UV SGI UV support is enabled. 85 XEN Xen support is enabled 86 XTENSA xtensa architecture is enabled. 87 88In addition, the following text indicates that the option 89 90 BOOT Is a boot loader parameter. 91 BUGS= Relates to possible processor bugs on the said processor. 92 KNL Is a kernel start-up parameter. 93 94 95Kernel parameters 96 97 accept_memory= [MM] 98 Format: { eager | lazy } 99 default: lazy 100 By default, unaccepted memory is accepted lazily to 101 avoid prolonged boot times. The lazy option will add 102 some runtime overhead until all memory is eventually 103 accepted. In most cases the overhead is negligible. 104 For some workloads or for debugging purposes 105 accept_memory=eager can be used to accept all memory 106 at once during boot. 107 108 acpi= [HW,ACPI,X86,ARM64,RISCV64,EARLY] 109 Advanced Configuration and Power Interface 110 Format: { force | on | off | strict | noirq | rsdt | 111 copy_dsdt | nospcr } 112 force -- enable ACPI if default was off 113 on -- enable ACPI but allow fallback to DT [arm64,riscv64] 114 off -- disable ACPI if default was on 115 noirq -- do not use ACPI for IRQ routing 116 strict -- Be less tolerant of platforms that are not 117 strictly ACPI specification compliant. 118 rsdt -- prefer RSDT over (default) XSDT 119 copy_dsdt -- copy DSDT to memory 120 nocmcff -- Disable firmware first mode for corrected 121 errors. This disables parsing the HEST CMC error 122 source to check if firmware has set the FF flag. This 123 may result in duplicate corrected error reports. 124 nospcr -- disable console in ACPI SPCR table as 125 default _serial_ console on ARM64 126 spcr -- enable console in ACPI SPCR table as 127 default _serial_ console on x86 128 For ARM64, ONLY "acpi=off", "acpi=on", "acpi=force" or 129 "acpi=nospcr" are available 130 For RISCV64, ONLY "acpi=off", "acpi=on" or "acpi=force" 131 are available 132 133 See also Documentation/power/runtime_pm.rst, pci=noacpi 134 135 acpi_apic_instance= [ACPI,IOAPIC,EARLY] 136 Format: <int> 137 2: use 2nd APIC table, if available 138 1,0: use 1st APIC table 139 default: 0 140 141 acpi_backlight= [HW,ACPI] 142 { vendor | video | native | none } 143 If set to vendor, prefer vendor-specific driver 144 (e.g. thinkpad_acpi, sony_acpi, etc.) instead 145 of the ACPI video.ko driver. 146 If set to video, use the ACPI video.ko driver. 147 If set to native, use the device's native backlight mode. 148 If set to none, disable the ACPI backlight interface. 149 150 acpi_force_32bit_fadt_addr [ACPI,EARLY] 151 force FADT to use 32 bit addresses rather than the 152 64 bit X_* addresses. Some firmware have broken 64 153 bit addresses for force ACPI ignore these and use 154 the older legacy 32 bit addresses. 155 156 acpica_no_return_repair [HW, ACPI] 157 Disable AML predefined validation mechanism 158 This mechanism can repair the evaluation result to make 159 the return objects more ACPI specification compliant. 160 This option is useful for developers to identify the 161 root cause of an AML interpreter issue when the issue 162 has something to do with the repair mechanism. 163 164 acpi.debug_layer= [HW,ACPI,ACPI_DEBUG] 165 acpi.debug_level= [HW,ACPI,ACPI_DEBUG] 166 Format: <int> 167 CONFIG_ACPI_DEBUG must be enabled to produce any ACPI 168 debug output. Bits in debug_layer correspond to a 169 _COMPONENT in an ACPI source file, e.g., 170 #define _COMPONENT ACPI_EVENTS 171 Bits in debug_level correspond to a level in 172 ACPI_DEBUG_PRINT statements, e.g., 173 ACPI_DEBUG_PRINT((ACPI_DB_INFO, ... 174 The debug_level mask defaults to "info". See 175 Documentation/firmware-guide/acpi/debug.rst for more information about 176 debug layers and levels. 177 178 Enable processor driver info messages: 179 acpi.debug_layer=0x20000000 180 Enable AML "Debug" output, i.e., stores to the Debug 181 object while interpreting AML: 182 acpi.debug_layer=0xffffffff acpi.debug_level=0x2 183 Enable all messages related to ACPI hardware: 184 acpi.debug_layer=0x2 acpi.debug_level=0xffffffff 185 186 Some values produce so much output that the system is 187 unusable. The "log_buf_len" parameter may be useful 188 if you need to capture more output. 189 190 acpi.poweroff_on_fatal= [ACPI] 191 {0 | 1} 192 Causes the system to poweroff when the ACPI bytecode signals 193 a fatal error. The default value of this setting is 1. 194 Overriding this value should only be done for diagnosing 195 ACPI firmware problems, as the system might behave erratically 196 after having encountered a fatal ACPI error. 197 198 acpi_enforce_resources= [ACPI] 199 { strict | lax | no } 200 Check for resource conflicts between native drivers 201 and ACPI OperationRegions (SystemIO and SystemMemory 202 only). IO ports and memory declared in ACPI might be 203 used by the ACPI subsystem in arbitrary AML code and 204 can interfere with legacy drivers. 205 strict (default): access to resources claimed by ACPI 206 is denied; legacy drivers trying to access reserved 207 resources will fail to bind to device using them. 208 lax: access to resources claimed by ACPI is allowed; 209 legacy drivers trying to access reserved resources 210 will bind successfully but a warning message is logged. 211 no: ACPI OperationRegions are not marked as reserved, 212 no further checks are performed. 213 214 acpi_force_table_verification [HW,ACPI,EARLY] 215 Enable table checksum verification during early stage. 216 By default, this is disabled due to x86 early mapping 217 size limitation. 218 219 acpi_irq_balance [HW,ACPI] 220 ACPI will balance active IRQs 221 default in APIC mode 222 223 acpi_irq_nobalance [HW,ACPI] 224 ACPI will not move active IRQs (default) 225 default in PIC mode 226 227 acpi_irq_isa= [HW,ACPI] If irq_balance, mark listed IRQs used by ISA 228 Format: <irq>,<irq>... 229 230 acpi_irq_pci= [HW,ACPI] If irq_balance, clear listed IRQs for 231 use by PCI 232 Format: <irq>,<irq>... 233 234 acpi_mask_gpe= [HW,ACPI] 235 Due to the existence of _Lxx/_Exx, some GPEs triggered 236 by unsupported hardware/firmware features can result in 237 GPE floodings that cannot be automatically disabled by 238 the GPE dispatcher. 239 This facility can be used to prevent such uncontrolled 240 GPE floodings. 241 Format: <byte> or <bitmap-list> 242 243 acpi_no_auto_serialize [HW,ACPI] 244 Disable auto-serialization of AML methods 245 AML control methods that contain the opcodes to create 246 named objects will be marked as "Serialized" by the 247 auto-serialization feature. 248 This feature is enabled by default. 249 This option allows to turn off the feature. 250 251 acpi_no_memhotplug [ACPI] Disable memory hotplug. Useful for kdump 252 kernels. 253 254 acpi_no_static_ssdt [HW,ACPI,EARLY] 255 Disable installation of static SSDTs at early boot time 256 By default, SSDTs contained in the RSDT/XSDT will be 257 installed automatically and they will appear under 258 /sys/firmware/acpi/tables. 259 This option turns off this feature. 260 Note that specifying this option does not affect 261 dynamic table installation which will install SSDT 262 tables to /sys/firmware/acpi/tables/dynamic. 263 264 acpi_no_watchdog [HW,ACPI,WDT] 265 Ignore the ACPI-based watchdog interface (WDAT) and let 266 a native driver control the watchdog device instead. 267 268 acpi_rsdp= [ACPI,EFI,KEXEC,EARLY] 269 Pass the RSDP address to the kernel, mostly used 270 on machines running EFI runtime service to boot the 271 second kernel for kdump. 272 273 acpi_os_name= [HW,ACPI] Tell ACPI BIOS the name of the OS 274 Format: To spoof as Windows 98: ="Microsoft Windows" 275 276 acpi_rev_override [ACPI] Override the _REV object to return 5 (instead 277 of 2 which is mandated by ACPI 6) as the supported ACPI 278 specification revision (when using this switch, it may 279 be necessary to carry out a cold reboot _twice_ in a 280 row to make it take effect on the platform firmware). 281 282 acpi_osi= [HW,ACPI] Modify list of supported OS interface strings 283 acpi_osi="string1" # add string1 284 acpi_osi="!string2" # remove string2 285 acpi_osi=!* # remove all strings 286 acpi_osi=! # disable all built-in OS vendor 287 strings 288 acpi_osi=!! # enable all built-in OS vendor 289 strings 290 acpi_osi= # disable all strings 291 292 'acpi_osi=!' can be used in combination with single or 293 multiple 'acpi_osi="string1"' to support specific OS 294 vendor string(s). Note that such command can only 295 affect the default state of the OS vendor strings, thus 296 it cannot affect the default state of the feature group 297 strings and the current state of the OS vendor strings, 298 specifying it multiple times through kernel command line 299 is meaningless. This command is useful when one do not 300 care about the state of the feature group strings which 301 should be controlled by the OSPM. 302 Examples: 303 1. 'acpi_osi=! acpi_osi="Windows 2000"' is equivalent 304 to 'acpi_osi="Windows 2000" acpi_osi=!', they all 305 can make '_OSI("Windows 2000")' TRUE. 306 307 'acpi_osi=' cannot be used in combination with other 308 'acpi_osi=' command lines, the _OSI method will not 309 exist in the ACPI namespace. NOTE that such command can 310 only affect the _OSI support state, thus specifying it 311 multiple times through kernel command line is also 312 meaningless. 313 Examples: 314 1. 'acpi_osi=' can make 'CondRefOf(_OSI, Local1)' 315 FALSE. 316 317 'acpi_osi=!*' can be used in combination with single or 318 multiple 'acpi_osi="string1"' to support specific 319 string(s). Note that such command can affect the 320 current state of both the OS vendor strings and the 321 feature group strings, thus specifying it multiple times 322 through kernel command line is meaningful. But it may 323 still not able to affect the final state of a string if 324 there are quirks related to this string. This command 325 is useful when one want to control the state of the 326 feature group strings to debug BIOS issues related to 327 the OSPM features. 328 Examples: 329 1. 'acpi_osi="Module Device" acpi_osi=!*' can make 330 '_OSI("Module Device")' FALSE. 331 2. 'acpi_osi=!* acpi_osi="Module Device"' can make 332 '_OSI("Module Device")' TRUE. 333 3. 'acpi_osi=! acpi_osi=!* acpi_osi="Windows 2000"' is 334 equivalent to 335 'acpi_osi=!* acpi_osi=! acpi_osi="Windows 2000"' 336 and 337 'acpi_osi=!* acpi_osi="Windows 2000" acpi_osi=!', 338 they all will make '_OSI("Windows 2000")' TRUE. 339 340 acpi_pm_good [X86] 341 Override the pmtimer bug detection: force the kernel 342 to assume that this machine's pmtimer latches its value 343 and always returns good values. 344 345 acpi_sci= [HW,ACPI,EARLY] ACPI System Control Interrupt trigger mode 346 Format: { level | edge | high | low } 347 348 acpi_skip_timer_override [HW,ACPI,EARLY] 349 Recognize and ignore IRQ0/pin2 Interrupt Override. 350 For broken nForce2 BIOS resulting in XT-PIC timer. 351 352 acpi_sleep= [HW,ACPI] Sleep options 353 Format: { s3_bios, s3_mode, s3_beep, s4_hwsig, 354 s4_nohwsig, old_ordering, nonvs, 355 sci_force_enable, nobl } 356 See Documentation/power/video.rst for information on 357 s3_bios and s3_mode. 358 s3_beep is for debugging; it makes the PC's speaker beep 359 as soon as the kernel's real-mode entry point is called. 360 s4_hwsig causes the kernel to check the ACPI hardware 361 signature during resume from hibernation, and gracefully 362 refuse to resume if it has changed. This complies with 363 the ACPI specification but not with reality, since 364 Windows does not do this and many laptops do change it 365 on docking. So the default behaviour is to allow resume 366 and simply warn when the signature changes, unless the 367 s4_hwsig option is enabled. 368 s4_nohwsig prevents ACPI hardware signature from being 369 used (or even warned about) during resume. 370 old_ordering causes the ACPI 1.0 ordering of the _PTS 371 control method, with respect to putting devices into 372 low power states, to be enforced (the ACPI 2.0 ordering 373 of _PTS is used by default). 374 nonvs prevents the kernel from saving/restoring the 375 ACPI NVS memory during suspend/hibernation and resume. 376 sci_force_enable causes the kernel to set SCI_EN directly 377 on resume from S1/S3 (which is against the ACPI spec, 378 but some broken systems don't work without it). 379 nobl causes the internal blacklist of systems known to 380 behave incorrectly in some ways with respect to system 381 suspend and resume to be ignored (use wisely). 382 383 acpi_use_timer_override [HW,ACPI,EARLY] 384 Use timer override. For some broken Nvidia NF5 boards 385 that require a timer override, but don't have HPET 386 387 add_efi_memmap [EFI,X86,EARLY] Include EFI memory map in 388 kernel's map of available physical RAM. 389 390 agp= [AGP] 391 { off | try_unsupported } 392 off: disable AGP support 393 try_unsupported: try to drive unsupported chipsets 394 (may crash computer or cause data corruption) 395 396 ALSA [HW,ALSA] 397 See Documentation/sound/alsa-configuration.rst 398 399 alignment= [KNL,ARM] 400 Allow the default userspace alignment fault handler 401 behaviour to be specified. Bit 0 enables warnings, 402 bit 1 enables fixups, and bit 2 sends a segfault. 403 404 align_va_addr= [X86-64] 405 Align virtual addresses by clearing slice [14:12] when 406 allocating a VMA at process creation time. This option 407 gives you up to 3% performance improvement on AMD F15h 408 machines (where it is enabled by default) for a 409 CPU-intensive style benchmark, and it can vary highly in 410 a microbenchmark depending on workload and compiler. 411 412 32: only for 32-bit processes 413 64: only for 64-bit processes 414 on: enable for both 32- and 64-bit processes 415 off: disable for both 32- and 64-bit processes 416 417 alloc_snapshot [FTRACE] 418 Allocate the ftrace snapshot buffer on boot up when the 419 main buffer is allocated. This is handy if debugging 420 and you need to use tracing_snapshot() on boot up, and 421 do not want to use tracing_snapshot_alloc() as it needs 422 to be done where GFP_KERNEL allocations are allowed. 423 424 allow_mismatched_32bit_el0 [ARM64,EARLY] 425 Allow execve() of 32-bit applications and setting of the 426 PER_LINUX32 personality on systems where only a strict 427 subset of the CPUs support 32-bit EL0. When this 428 parameter is present, the set of CPUs supporting 32-bit 429 EL0 is indicated by /sys/devices/system/cpu/aarch32_el0 430 and hot-unplug operations may be restricted. 431 432 See Documentation/arch/arm64/asymmetric-32bit.rst for more 433 information. 434 435 amd_iommu= [HW,X86-64] 436 Pass parameters to the AMD IOMMU driver in the system. 437 Possible values are: 438 fullflush - Deprecated, equivalent to iommu.strict=1 439 off - do not initialize any AMD IOMMU found in 440 the system 441 force_isolation - Force device isolation for all 442 devices. The IOMMU driver is not 443 allowed anymore to lift isolation 444 requirements as needed. This option 445 does not override iommu=pt 446 force_enable - Force enable the IOMMU on platforms known 447 to be buggy with IOMMU enabled. Use this 448 option with care. 449 pgtbl_v1 - Use v1 page table for DMA-API (Default). 450 pgtbl_v2 - Use v2 page table for DMA-API. 451 irtcachedis - Disable Interrupt Remapping Table (IRT) caching. 452 nohugepages - Limit page-sizes used for v1 page-tables 453 to 4 KiB. 454 v2_pgsizes_only - Limit page-sizes used for v1 page-tables 455 to 4KiB/2Mib/1GiB. 456 457 458 amd_iommu_dump= [HW,X86-64] 459 Enable AMD IOMMU driver option to dump the ACPI table 460 for AMD IOMMU. With this option enabled, AMD IOMMU 461 driver will print ACPI tables for AMD IOMMU during 462 IOMMU initialization. 463 464 amd_iommu_intr= [HW,X86-64] 465 Specifies one of the following AMD IOMMU interrupt 466 remapping modes: 467 legacy - Use legacy interrupt remapping mode. 468 vapic - Use virtual APIC mode, which allows IOMMU 469 to inject interrupts directly into guest. 470 This mode requires kvm-amd.avic=1. 471 (Default when IOMMU HW support is present.) 472 473 amd_pstate= [X86,EARLY] 474 disable 475 Do not enable amd_pstate as the default 476 scaling driver for the supported processors 477 passive 478 Use amd_pstate with passive mode as a scaling driver. 479 In this mode autonomous selection is disabled. 480 Driver requests a desired performance level and platform 481 tries to match the same performance level if it is 482 satisfied by guaranteed performance level. 483 active 484 Use amd_pstate_epp driver instance as the scaling driver, 485 driver provides a hint to the hardware if software wants 486 to bias toward performance (0x0) or energy efficiency (0xff) 487 to the CPPC firmware. then CPPC power algorithm will 488 calculate the runtime workload and adjust the realtime cores 489 frequency. 490 guided 491 Activate guided autonomous mode. Driver requests minimum and 492 maximum performance level and the platform autonomously 493 selects a performance level in this range and appropriate 494 to the current workload. 495 496 amd_prefcore= 497 [X86] 498 disable 499 Disable amd-pstate preferred core. 500 501 amd_dynamic_epp= 502 [X86] 503 disable 504 Disable amd-pstate dynamic EPP. 505 enable 506 Enable amd-pstate dynamic EPP. 507 508 amijoy.map= [HW,JOY] Amiga joystick support 509 Map of devices attached to JOY0DAT and JOY1DAT 510 Format: <a>,<b> 511 See also Documentation/input/joydev/joystick.rst 512 513 analog.map= [HW,JOY] Analog joystick and gamepad support 514 Specifies type or capabilities of an analog joystick 515 connected to one of 16 gameports 516 Format: <type1>,<type2>,..<type16> 517 518 apc= [HW,SPARC] 519 Power management functions (SPARCstation-4/5 + deriv.) 520 Format: noidle 521 Disable APC CPU standby support. SPARCstation-Fox does 522 not play well with APC CPU idle - disable it if you have 523 APC and your system crashes randomly. 524 525 apic [APIC,X86-64] Use IO-APIC. Default. 526 527 apic= [APIC,X86,EARLY] Advanced Programmable Interrupt Controller 528 Change the output verbosity while booting 529 Format: { quiet (default) | verbose | debug } 530 Change the amount of debugging information output 531 when initialising the APIC and IO-APIC components. 532 533 apic_extnmi= [APIC,X86,EARLY] External NMI delivery setting 534 Format: { bsp (default) | all | none } 535 bsp: External NMI is delivered only to CPU 0 536 all: External NMIs are broadcast to all CPUs as a 537 backup of CPU 0 538 none: External NMI is masked for all CPUs. This is 539 useful so that a dump capture kernel won't be 540 shot down by NMI 541 542 apicpmtimer Do APIC timer calibration using the pmtimer. Implies 543 apicmaintimer. Useful when your PIT timer is totally 544 broken. 545 546 autoconf= [IPV6] 547 See Documentation/networking/ipv6.rst. 548 549 apm= [APM] Advanced Power Management 550 See header of arch/x86/kernel/apm_32.c. 551 552 apparmor= [APPARMOR] Disable or enable AppArmor at boot time 553 Format: { "0" | "1" } 554 See security/apparmor/Kconfig help text 555 0 -- disable. 556 1 -- enable. 557 Default value is set via kernel config option. 558 559 arm64.no32bit_el0 [ARM64] Unconditionally disable the execution of 560 32 bit applications. 561 562 arm64.nobti [ARM64] Unconditionally disable Branch Target 563 Identification support 564 565 arm64.nogcs [ARM64] Unconditionally disable Guarded Control Stack 566 support 567 568 arm64.nomops [ARM64] Unconditionally disable Memory Copy and Memory 569 Set instructions support 570 571 arm64.nompam [ARM64] Unconditionally disable Memory Partitioning And 572 Monitoring support 573 574 arm64.nomte [ARM64] Unconditionally disable Memory Tagging Extension 575 support 576 577 arm64.nopauth [ARM64] Unconditionally disable Pointer Authentication 578 support 579 580 arm64.nosme [ARM64] Unconditionally disable Scalable Matrix 581 Extension support 582 583 arm64.nosve [ARM64] Unconditionally disable Scalable Vector 584 Extension support 585 586 ataflop= [HW,M68k] 587 588 atarimouse= [HW,MOUSE] Atari Mouse 589 590 atkbd.extra= [HW] Enable extra LEDs and keys on IBM RapidAccess, 591 EzKey and similar keyboards 592 593 atkbd.reset= [HW] Reset keyboard during initialization 594 595 atkbd.set= [HW] Select keyboard code set 596 Format: <int> (2 = AT (default), 3 = PS/2) 597 598 atkbd.scroll= [HW] Enable scroll wheel on MS Office and similar 599 keyboards 600 601 atkbd.softraw= [HW] Choose between synthetic and real raw mode 602 Format: <bool> (0 = real, 1 = synthetic (default)) 603 604 atkbd.softrepeat= [HW] 605 Use software keyboard repeat 606 607 audit= [KNL] Enable the audit sub-system 608 Format: { "0" | "1" | "off" | "on" } 609 0 | off - kernel audit is disabled and can not be 610 enabled until the next reboot 611 unset - kernel audit is initialized but disabled and 612 will be fully enabled by the userspace auditd. 613 1 | on - kernel audit is initialized and partially 614 enabled, storing at most audit_backlog_limit 615 messages in RAM until it is fully enabled by the 616 userspace auditd. 617 Default: unset 618 619 audit_backlog_limit= [KNL] Set the audit queue size limit. 620 Format: <int> (must be >=0) 621 Default: 64 622 623 bau= [X86_UV] Enable the BAU on SGI UV. The default 624 behavior is to disable the BAU (i.e. bau=0). 625 Format: { "0" | "1" } 626 0 - Disable the BAU. 627 1 - Enable the BAU. 628 unset - Disable the BAU. 629 630 bdev_allow_write_mounted= 631 Format: <bool> 632 Control the ability to open a mounted block device 633 for writing, i.e., allow / disallow writes that bypass 634 the FS. This was implemented as a means to prevent 635 fuzzers from crashing the kernel by overwriting the 636 metadata underneath a mounted FS without its awareness. 637 This also prevents destructive formatting of mounted 638 filesystems by naive storage tooling that don't use 639 O_EXCL. Default is Y and can be changed through the 640 Kconfig option CONFIG_BLK_DEV_WRITE_MOUNTED. 641 642 bert_disable [ACPI] 643 Disable BERT OS support on buggy BIOSes. 644 645 bgrt_disable [ACPI,X86,EARLY] 646 Disable BGRT to avoid flickering OEM logo. 647 648 blkdevparts= Manual partition parsing of block device(s) for 649 embedded devices based on command line input. 650 See Documentation/block/cmdline-partition.rst 651 652 boot_delay= [KNL,EARLY] 653 Milliseconds to delay each printk during boot. 654 Only works if CONFIG_BOOT_PRINTK_DELAY is enabled, 655 and you may also have to specify "lpj=". Boot_delay 656 values larger than 10 seconds (10000) are assumed 657 erroneous and ignored. 658 Format: integer 659 660 bootconfig [KNL,EARLY] 661 Extended command line options can be added to an initrd 662 and this will cause the kernel to look for it. 663 664 See Documentation/admin-guide/bootconfig.rst 665 666 bttv.card= [HW,V4L] bttv (bt848 + bt878 based grabber cards) 667 bttv.radio= Most important insmod options are available as 668 kernel args too. 669 bttv.pll= See Documentation/admin-guide/media/bttv.rst 670 bttv.tuner= 671 672 bulk_remove=off [PPC] This parameter disables the use of the pSeries 673 firmware feature for flushing multiple hpte entries 674 at a time. 675 676 c101= [NET] Moxa C101 synchronous serial card 677 678 cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection. 679 Sometimes CPU hardware bugs make them report the cache 680 size incorrectly. The kernel will attempt work arounds 681 to fix known problems, but for some CPUs it is not 682 possible to determine what the correct size should be. 683 This option provides an override for these situations. 684 685 carrier_timeout= 686 [NET] Specifies amount of time (in seconds) that 687 the kernel should wait for a network carrier. By default 688 it waits 120 seconds. 689 690 ca_keys= [KEYS] This parameter identifies a specific key(s) on 691 the system trusted keyring to be used for certificate 692 trust validation. 693 format: { id:<keyid> | builtin } 694 695 cca= [MIPS,EARLY] Override the kernel pages' cache coherency 696 algorithm. Accepted values range from 0 to 7 697 inclusive. See arch/mips/include/asm/pgtable-bits.h 698 for platform specific values (SB1, Loongson3 and 699 others). 700 701 ccw_timeout_log [S390] 702 See Documentation/arch/s390/common_io.rst for details. 703 704 cfi= [X86-64] Set Control Flow Integrity checking features 705 when CONFIG_FINEIBT is enabled. 706 Format: feature[,feature...] 707 Default: auto 708 709 auto: Use FineIBT if IBT available, otherwise kCFI. 710 Under FineIBT, enable "paranoid" mode when 711 FRED is not available. 712 off: Turn off CFI checking. 713 kcfi: Use kCFI (disable FineIBT). 714 fineibt: Use FineIBT (even if IBT not available). 715 norand: Do not re-randomize CFI hashes. 716 paranoid: Add caller hash checking under FineIBT. 717 bhi: Enable register poisoning to stop speculation 718 across FineIBT. (Disabled by default.) 719 warn: Do not enforce CFI checking: warn only. 720 debug: Report CFI initialization details. 721 722 cgroup_disable= [KNL] Disable a particular controller or optional feature 723 Format: {name of the controller(s) or feature(s) to disable} 724 The effects of cgroup_disable=foo are: 725 - foo isn't auto-mounted if you mount all cgroups in 726 a single hierarchy 727 - foo isn't visible as an individually mountable 728 subsystem 729 - if foo is an optional feature then the feature is 730 disabled and corresponding cgroup files are not 731 created 732 {Currently only "memory" controller deal with this and 733 cut the overhead, others just disable the usage. So 734 only cgroup_disable=memory is actually worthy} 735 Specifying "pressure" disables per-cgroup pressure 736 stall information accounting feature 737 738 cgroup_no_v1= [KNL] Disable cgroup controllers and named hierarchies in v1 739 Format: { { controller | "all" | "named" } 740 [,{ controller | "all" | "named" }...] } 741 Like cgroup_disable, but only applies to cgroup v1; 742 the blacklisted controllers remain available in cgroup2. 743 "all" blacklists all controllers and "named" disables 744 named mounts. Specifying both "all" and "named" disables 745 all v1 hierarchies. 746 747 cgroup_v1_proc= [KNL] Show also missing controllers in /proc/cgroups 748 Format: { "true" | "false" } 749 /proc/cgroups lists only v1 controllers by default. 750 This compatibility option enables listing also v2 751 controllers (whose v1 code is not compiled!), so that 752 semi-legacy software can check this file to decide 753 about usage of v2 (sic) controllers. 754 755 cgroup_favordynmods= [KNL] Enable or Disable favordynmods. 756 Format: { "true" | "false" } 757 Defaults to the value of CONFIG_CGROUP_FAVOR_DYNMODS. 758 759 cgroup.memory= [KNL] Pass options to the cgroup memory controller. 760 Format: <string> 761 nosocket -- Disable socket memory accounting. 762 nokmem -- Disable kernel memory accounting. 763 nobpf -- Disable BPF memory accounting. 764 765 check_pages= [MM,EARLY] Enable sanity checking of pages after 766 allocations / before freeing. This adds checks to catch 767 double-frees, use-after-frees, and other sources of 768 page corruption by inspecting page internals (flags, 769 mapcount/refcount, memcg_data, etc.). 770 Format: { "0" | "1" } 771 Default: 0 (1 if CONFIG_DEBUG_VM is set) 772 773 checkreqprot= [SELINUX] Set initial checkreqprot flag value. 774 Format: { "0" | "1" } 775 See security/selinux/Kconfig help text. 776 0 -- check protection applied by kernel (includes 777 any implied execute protection). 778 1 -- check protection requested by application. 779 Default value is set via a kernel config option. 780 Value can be changed at runtime via 781 /sys/fs/selinux/checkreqprot. 782 Setting checkreqprot to 1 is deprecated. 783 784 cio_ignore= [S390] 785 See Documentation/arch/s390/common_io.rst for details. 786 787 clk_ignore_unused 788 [CLK] 789 Prevents the clock framework from automatically gating 790 clocks that have not been explicitly enabled by a Linux 791 device driver but are enabled in hardware at reset or 792 by the bootloader/firmware. Note that this does not 793 force such clocks to be always-on nor does it reserve 794 those clocks in any way. This parameter is useful for 795 debug and development, but should not be needed on a 796 platform with proper driver support. For more 797 information, see Documentation/driver-api/clk.rst. 798 799 clock= [BUGS=X86-32, HW] gettimeofday clocksource override. 800 [Deprecated] 801 Forces specified clocksource (if available) to be used 802 when calculating gettimeofday(). If specified 803 clocksource is not available, it defaults to PIT. 804 Format: { pit | tsc | cyclone | pmtmr } 805 806 clocksource= Override the default clocksource 807 Format: <string> 808 Override the default clocksource and use the clocksource 809 with the name specified. 810 Some clocksource names to choose from, depending on 811 the platform: 812 [all] jiffies (this is the base, fallback clocksource) 813 [ACPI] acpi_pm 814 [ARM] imx_timer1,OSTS,netx_timer,mpu_timer2, 815 pxa_timer,timer3,32k_counter,timer0_1 816 [X86-32] pit,hpet,tsc; 817 scx200_hrt on Geode; cyclone on IBM x440 818 [MIPS] MIPS 819 [PARISC] cr16 820 [S390] tod 821 [SH] SuperH 822 [SPARC64] tick 823 [X86-64] hpet,tsc 824 825 clocksource.arm_arch_timer.evtstrm= 826 [ARM,ARM64,EARLY] 827 Format: <bool> 828 Enable/disable the eventstream feature of the ARM 829 architected timer so that code using WFE-based polling 830 loops can be debugged more effectively on production 831 systems. 832 833 clocksource.verify_n_cpus= [KNL] 834 Limit the number of CPUs checked for clocksources 835 marked with CLOCK_SOURCE_VERIFY_PERCPU that 836 are marked unstable due to excessive skew. 837 A negative value says to check all CPUs, while 838 zero says not to check any. Values larger than 839 nr_cpu_ids are silently truncated to nr_cpu_ids. 840 The actual CPUs are chosen randomly, with 841 no replacement if the same CPU is chosen twice. 842 843 clocksource-wdtest.holdoff= [KNL] 844 Set the time in seconds that the clocksource 845 watchdog test waits before commencing its tests. 846 Defaults to zero when built as a module and to 847 10 seconds when built into the kernel. 848 849 cma=nn[MG]@[start[MG][-end[MG]]] 850 [KNL,CMA,EARLY] 851 Sets the size of kernel global memory area for 852 contiguous memory allocations and optionally the 853 placement constraint by the physical address range of 854 memory allocations. A value of 0 disables CMA 855 altogether. For more information, see 856 kernel/dma/contiguous.c 857 858 cma_pernuma=nn[MG] 859 [KNL,CMA,EARLY] 860 Sets the size of kernel per-numa memory area for 861 contiguous memory allocations. A value of 0 disables 862 per-numa CMA altogether. And If this option is not 863 specified, the default value is 0. 864 With per-numa CMA enabled, DMA users on node nid will 865 first try to allocate buffer from the pernuma area 866 which is located in node nid, if the allocation fails, 867 they will fallback to the global default memory area. 868 869 numa_cma=<node>:nn[MG][,<node>:nn[MG]] 870 [KNL,CMA,EARLY] 871 Sets the size of kernel numa memory area for 872 contiguous memory allocations. It will reserve CMA 873 area for the specified node. 874 875 If it is setup together with upper 'cmd_pernuma=' 876 (unlikely), its size setting takes priority for the 877 specified numa nodes. 878 879 With numa CMA enabled, DMA users on node nid will 880 first try to allocate buffer from the numa area 881 which is located in node nid, if the allocation fails, 882 they will fallback to the global default memory area. 883 884 cmo_free_hint= [PPC] Format: { yes | no } 885 Specify whether pages are marked as being inactive 886 when they are freed. This is used in CMO environments 887 to determine OS memory pressure for page stealing by 888 a hypervisor. 889 Default: yes 890 891 coherent_pool=nn[KMG] [ARM,KNL,EARLY] 892 Sets the size of memory pool for coherent, atomic dma 893 allocations, by default set to 256K. 894 895 condev= [HW,S390] console device 896 conmode= 897 898 con3215_drop= [S390,EARLY] 3215 console drop mode. 899 Format: y|n|Y|N|1|0 900 When set to true, drop data on the 3215 console when 901 the console buffer is full. In this case the 902 operator using a 3270 terminal emulator (for example 903 x3270) does not have to enter the clear key for the 904 console output to advance and the kernel to continue. 905 This leads to a much faster boot time when a 3270 906 terminal emulator is active. If no 3270 terminal 907 emulator is used, this parameter has no effect. 908 909 console= [KNL] Output console device and options. 910 911 tty<n> Use the virtual console device <n>. 912 913 ttyS<n>[,options] 914 ttyUSB0[,options] 915 Use the specified serial port. The options are of 916 the form "bbbbpnf", where "bbbb" is the baud rate, 917 "p" is parity ("n", "o", or "e"), "n" is number of 918 bits, and "f" is flow control ("r" for RTS or 919 omit it). Default is "9600n8". 920 921 See Documentation/admin-guide/serial-console.rst for more 922 information. See 923 Documentation/networking/netconsole.rst for an 924 alternative. 925 926 <DEVNAME>:<n>.<n>[,options] 927 Use the specified serial port on the serial core bus. 928 The addressing uses DEVNAME of the physical serial port 929 device, followed by the serial core controller instance, 930 and the serial port instance. The options are the same 931 as documented for the ttyS addressing above. 932 933 The mapping of the serial ports to the tty instances 934 can be viewed with: 935 936 $ ls -d /sys/bus/serial-base/devices/*:*.*/tty/* 937 /sys/bus/serial-base/devices/00:04:0.0/tty/ttyS0 938 939 In the above example, the console can be addressed with 940 console=00:04:0.0. Note that a console addressed this 941 way will only get added when the related device driver 942 is ready. The use of an earlycon parameter in addition to 943 the console may be desired for console output early on. 944 945 uart[8250],io,<addr>[,options] 946 uart[8250],mmio,<addr>[,options] 947 uart[8250],mmio16,<addr>[,options] 948 uart[8250],mmio32,<addr>[,options] 949 uart[8250],0x<addr>[,options] 950 Start an early, polled-mode console on the 8250/16550 951 UART at the specified I/O port or MMIO address, 952 switching to the matching ttyS device later. 953 MMIO inter-register address stride is either 8-bit 954 (mmio), 16-bit (mmio16), or 32-bit (mmio32). 955 If none of [io|mmio|mmio16|mmio32], <addr> is assumed 956 to be equivalent to 'mmio'. 'options' are specified in 957 the same format described for ttyS above; if unspecified, 958 the h/w is not re-initialized. 959 960 hvc<n> Use the hypervisor console device <n>. This is for 961 both Xen and PowerPC hypervisors. 962 963 { null | "" } 964 Use to disable console output, i.e., to have kernel 965 console messages discarded. 966 This must be the only console= parameter used on the 967 kernel command line. 968 969 If the device connected to the port is not a TTY but a braille 970 device, prepend "brl," before the device type, for instance 971 console=brl,ttyS0 972 For now, only VisioBraille is supported. 973 974 console_msg_format= 975 [KNL] Change console messages format 976 default 977 By default we print messages on consoles in 978 "[time stamp] text\n" format (time stamp may not be 979 printed, depending on CONFIG_PRINTK_TIME or 980 `printk_time' param). 981 syslog 982 Switch to syslog format: "<%u>[time stamp] text\n" 983 IOW, each message will have a facility and loglevel 984 prefix. The format is similar to one used by syslog() 985 syscall, or to executing "dmesg -S --raw" or to reading 986 from /proc/kmsg. 987 988 consoleblank= [KNL] The console blank (screen saver) timeout in 989 seconds. A value of 0 disables the blank timer. 990 Defaults to 0. 991 992 coredump_filter= 993 [KNL] Change the default value for 994 /proc/<pid>/coredump_filter. 995 See also Documentation/filesystems/proc.rst. 996 997 coresight_cpu_debug.enable 998 [ARM,ARM64] 999 Format: <bool> 1000 Enable/disable the CPU sampling based debugging. 1001 0: default value, disable debugging 1002 1: enable debugging at boot time 1003 1004 cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver 1005 Format: 1006 <first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>] 1007 1008 cpuidle.off=1 [CPU_IDLE] 1009 disable the cpuidle sub-system 1010 1011 cpuidle.governor= 1012 [CPU_IDLE] Name of the cpuidle governor to use. 1013 1014 cpufreq.off=1 [CPU_FREQ] 1015 disable the cpufreq sub-system 1016 1017 cpufreq.default_governor= 1018 [CPU_FREQ] Name of the default cpufreq governor or 1019 policy to use. This governor must be registered in the 1020 kernel before the cpufreq driver probes. 1021 1022 cpu_init_udelay=N 1023 [X86,EARLY] Delay for N microsec between assert and de-assert 1024 of APIC INIT to start processors. This delay occurs 1025 on every CPU online, such as boot, and resume from suspend. 1026 Default: 10000 1027 1028 cpuhp.parallel= 1029 [SMP] Enable/disable parallel bringup of secondary CPUs 1030 Format: <bool> 1031 Default is enabled if CONFIG_HOTPLUG_PARALLEL=y. Otherwise 1032 the parameter has no effect. 1033 1034 crash_kexec_post_notifiers 1035 Only jump to kdump kernel after running the panic 1036 notifiers and dumping kmsg. This option increases 1037 the risks of a kdump failure, since some panic 1038 notifiers can make the crashed kernel more unstable. 1039 In configurations where kdump may not be reliable, 1040 running the panic notifiers could allow collecting 1041 more data on dmesg, like stack traces from other CPUS 1042 or extra data dumped by panic_print. Note that some 1043 configurations enable this option unconditionally, 1044 like Hyper-V, PowerPC (fadump) and AMD SEV-SNP. 1045 1046 crashkernel=size[KMG][@offset[KMG]] 1047 [KNL,EARLY] Using kexec, Linux can switch to a 'crash kernel' 1048 upon panic. This parameter reserves the physical 1049 memory region [offset, offset + size] for that kernel 1050 image. If '@offset' is omitted, then a suitable offset 1051 is selected automatically. 1052 [KNL, X86-64, ARM64, RISCV, LoongArch] Select a region 1053 under 4G first, and fall back to reserve region above 1054 4G when '@offset' hasn't been specified. 1055 See Documentation/admin-guide/kdump/kdump.rst for further details. 1056 1057 crashkernel=range1:size1[,range2:size2,...][@offset] 1058 [KNL] Same as above, but depends on the memory 1059 in the running system. The syntax of range is 1060 start-[end] where start and end are both 1061 a memory unit (amount[KMG]). See also 1062 Documentation/admin-guide/kdump/kdump.rst for an example. 1063 1064 crashkernel=size[KMG],high 1065 [KNL, X86-64, ARM64, RISCV, LoongArch] range could be 1066 above 4G. 1067 Allow kernel to allocate physical memory region from top, 1068 so could be above 4G if system have more than 4G ram 1069 installed. Otherwise memory region will be allocated 1070 below 4G, if available. 1071 It will be ignored if crashkernel=X is specified. 1072 crashkernel=size[KMG],low 1073 [KNL, X86-64, ARM64, RISCV, LoongArch] range under 4G. 1074 When crashkernel=X,high is passed, kernel could allocate 1075 physical memory region above 4G, that cause second kernel 1076 crash on system that require some amount of low memory, 1077 e.g. swiotlb requires at least 64M+32K low memory, also 1078 enough extra low memory is needed to make sure DMA buffers 1079 for 32-bit devices won't run out. Kernel would try to allocate 1080 default size of memory below 4G automatically. The default 1081 size is platform dependent. 1082 --> x86: max(swiotlb_size_or_default() + 8MiB, 256MiB) 1083 --> arm64: 128MiB 1084 --> riscv: 128MiB 1085 --> loongarch: 128MiB 1086 This one lets the user specify own low range under 4G 1087 for second kernel instead. 1088 0: to disable low allocation. 1089 It will be ignored when crashkernel=X,high is not used 1090 or memory reserved is below 4G. 1091 crashkernel=size[KMG],cma 1092 [KNL, X86, ppc] Reserve additional crash kernel memory from 1093 CMA. This reservation is usable by the first system's 1094 userspace memory and kernel movable allocations (memory 1095 balloon, zswap). Pages allocated from this memory range 1096 will not be included in the vmcore so this should not 1097 be used if dumping of userspace memory is intended and 1098 it has to be expected that some movable kernel pages 1099 may be missing from the dump. 1100 1101 A standard crashkernel reservation, as described above, 1102 is still needed to hold the crash kernel and initrd. 1103 1104 This option increases the risk of a kdump failure: DMA 1105 transfers configured by the first kernel may end up 1106 corrupting the second kernel's memory. 1107 1108 This reservation method is intended for systems that 1109 can't afford to sacrifice enough memory for standard 1110 crashkernel reservation and where less reliable and 1111 possibly incomplete kdump is preferable to no kdump at 1112 all. 1113 1114 cryptomgr.notests 1115 [KNL] Disable crypto self-tests 1116 1117 cs89x0_dma= [HW,NET] 1118 Format: <dma> 1119 1120 cs89x0_media= [HW,NET] 1121 Format: { rj45 | aui | bnc } 1122 1123 csdlock_debug= [KNL] Enable or disable debug add-ons of cross-CPU 1124 function call handling. When switched on, 1125 additional debug data is printed to the console 1126 in case a hanging CPU is detected, and that 1127 CPU is pinged again in order to try to resolve 1128 the hang situation. The default value of this 1129 option depends on the CSD_LOCK_WAIT_DEBUG_DEFAULT 1130 Kconfig option. 1131 1132 dasd= [HW,NET] 1133 See header of drivers/s390/block/dasd_devmap.c. 1134 1135 db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port 1136 (one device per port) 1137 Format: <port#>,<type> 1138 See also Documentation/input/devices/joystick-parport.rst 1139 1140 debug [KNL,EARLY] Enable kernel debugging (events log level). 1141 1142 debug_boot_weak_hash 1143 [KNL,EARLY] Enable printing [hashed] pointers early in the 1144 boot sequence. If enabled, we use a weak hash instead 1145 of siphash to hash pointers. Use this option if you are 1146 seeing instances of '(___ptrval___)') and need to see a 1147 value (hashed pointer) instead. Cryptographically 1148 insecure, please do not use on production kernels. 1149 1150 debug_locks_verbose= 1151 [KNL] verbose locking self-tests 1152 Format: <int> 1153 Print debugging info while doing the locking API 1154 self-tests. 1155 Bitmask for the various LOCKTYPE_ tests. Defaults to 0 1156 (no extra messages), setting it to -1 (all bits set) 1157 will print _a_lot_ more information - normally only 1158 useful to lockdep developers. 1159 1160 debug_objects [KNL,EARLY] Enable object debugging 1161 1162 debug_guardpage_minorder= 1163 [KNL,EARLY] When CONFIG_DEBUG_PAGEALLOC is set, this 1164 parameter allows control of the order of pages that will 1165 be intentionally kept free (and hence protected) by the 1166 buddy allocator. Bigger value increase the probability 1167 of catching random memory corruption, but reduce the 1168 amount of memory for normal system use. The maximum 1169 possible value is MAX_PAGE_ORDER/2. Setting this 1170 parameter to 1 or 2 should be enough to identify most 1171 random memory corruption problems caused by bugs in 1172 kernel or driver code when a CPU writes to (or reads 1173 from) a random memory location. Note that there exists 1174 a class of memory corruptions problems caused by buggy 1175 H/W or F/W or by drivers badly programming DMA 1176 (basically when memory is written at bus level and the 1177 CPU MMU is bypassed) which are not detectable by 1178 CONFIG_DEBUG_PAGEALLOC, hence this option will not 1179 help tracking down these problems. 1180 1181 debug_pagealloc= 1182 [KNL,EARLY] When CONFIG_DEBUG_PAGEALLOC is set, this parameter 1183 enables the feature at boot time. By default, it is 1184 disabled and the system will work mostly the same as a 1185 kernel built without CONFIG_DEBUG_PAGEALLOC. 1186 Note: to get most of debug_pagealloc error reports, it's 1187 useful to also enable the page_owner functionality. 1188 on: enable the feature 1189 1190 debugfs= [KNL,EARLY] This parameter enables what is exposed to 1191 userspace and debugfs internal clients. 1192 Format: { on, off } 1193 on: All functions are enabled. 1194 off: Filesystem is not registered and clients 1195 get a -EPERM as result when trying to register files 1196 or directories within debugfs. 1197 This is equivalent of the runtime functionality if 1198 debugfs was not enabled in the kernel at all. 1199 Default value is set in build-time with a kernel configuration. 1200 1201 debugpat [X86] Enable PAT debugging 1202 1203 default_hugepagesz= 1204 [HW] The size of the default HugeTLB page. This is 1205 the size represented by the legacy /proc/ hugepages 1206 APIs. In addition, this is the default hugetlb size 1207 used for shmget(), mmap() and mounting hugetlbfs 1208 filesystems. If not specified, defaults to the 1209 architecture's default huge page size. Huge page 1210 sizes are architecture dependent. See also 1211 Documentation/admin-guide/mm/hugetlbpage.rst. 1212 Format: size[KMG] 1213 1214 deferred_probe_timeout= 1215 [KNL] Debugging option to set a timeout in seconds for 1216 deferred probe to give up waiting on dependencies to 1217 probe. Only specific dependencies (subsystems or 1218 drivers) that have opted in will be ignored. A timeout 1219 of 0 will timeout at the end of initcalls; a negative 1220 value is treated as an infinite timeout value. If the 1221 timeout hasn't expired, it'll be restarted by each 1222 successful driver registration. This option will also 1223 dump out devices still on the deferred probe list after 1224 retrying. 1225 1226 delayacct [KNL] Enable per-task delay accounting 1227 1228 dell_smm_hwmon.ignore_dmi= 1229 [HW] Continue probing hardware even if DMI data 1230 indicates that the driver is running on unsupported 1231 hardware. 1232 1233 dell_smm_hwmon.force= 1234 [HW] Activate driver even if SMM BIOS signature does 1235 not match list of supported models and enable otherwise 1236 blacklisted features. 1237 1238 dell_smm_hwmon.power_status= 1239 [HW] Report power status in /proc/i8k 1240 (disabled by default). 1241 1242 dell_smm_hwmon.restricted= 1243 [HW] Allow controlling fans only if SYS_ADMIN 1244 capability is set. 1245 1246 dell_smm_hwmon.fan_mult= 1247 [HW] Factor to multiply fan speed with. 1248 1249 dell_smm_hwmon.fan_max= 1250 [HW] Maximum configurable fan speed. 1251 1252 dfltcc= [HW,S390] 1253 Format: { on | off | def_only | inf_only | always } 1254 on: s390 zlib hardware support for compression on 1255 level 1 and decompression (default) 1256 off: No s390 zlib hardware support 1257 def_only: s390 zlib hardware support for deflate 1258 only (compression on level 1) 1259 inf_only: s390 zlib hardware support for inflate 1260 only (decompression) 1261 always: Same as 'on' but ignores the selected compression 1262 level always using hardware support (used for debugging) 1263 1264 dhash_entries= [KNL] 1265 Set number of hash buckets for dentry cache. 1266 1267 disable_1tb_segments [PPC,EARLY] 1268 Disables the use of 1TB hash page table segments. This 1269 causes the kernel to fall back to 256MB segments which 1270 can be useful when debugging issues that require an SLB 1271 miss to occur. 1272 1273 disable= [IPV6] 1274 See Documentation/networking/ipv6.rst. 1275 1276 disable_radix [PPC,EARLY] 1277 Disable RADIX MMU mode on POWER9 1278 1279 disable_tlbie [PPC] 1280 Disable TLBIE instruction. Currently does not work 1281 with KVM, with HASH MMU, or with coherent accelerators. 1282 1283 disable_ddw [PPC/PSERIES,EARLY] 1284 Disable Dynamic DMA Window support. Use this 1285 to workaround buggy firmware. 1286 1287 disable_ipv6= [IPV6] 1288 See Documentation/networking/ipv6.rst. 1289 1290 disable_mtrr_cleanup [X86,EARLY] 1291 The kernel tries to adjust MTRR layout from continuous 1292 to discrete, to make X server driver able to add WB 1293 entry later. This parameter disables that. 1294 1295 disable_mtrr_trim [X86, Intel and AMD only,EARLY] 1296 By default the kernel will trim any uncacheable 1297 memory out of your available memory pool based on 1298 MTRR settings. This parameter disables that behavior, 1299 possibly causing your machine to run very slowly. 1300 1301 disable_timer_pin_1 [X86,EARLY] 1302 Disable PIN 1 of APIC timer 1303 Can be useful to work around chipset bugs. 1304 1305 dis_ucode_ldr [X86] Disable the microcode loader. 1306 1307 dma_debug=off If the kernel is compiled with DMA_API_DEBUG support, 1308 this option disables the debugging code at boot. 1309 1310 dma_debug_entries=<number> 1311 This option allows to tune the number of preallocated 1312 entries for DMA-API debugging code. One entry is 1313 required per DMA-API allocation. Use this if the 1314 DMA-API debugging code disables itself because the 1315 architectural default is too low. 1316 1317 dma_debug_driver=<driver_name> 1318 With this option the DMA-API debugging driver 1319 filter feature can be enabled at boot time. Just 1320 pass the driver to filter for as the parameter. 1321 The filter can be disabled or changed to another 1322 driver later using sysfs. 1323 1324 reg_file_data_sampling= 1325 [X86] Controls mitigation for Register File Data 1326 Sampling (RFDS) vulnerability. RFDS is a CPU 1327 vulnerability which may allow userspace to infer 1328 kernel data values previously stored in floating point 1329 registers, vector registers, or integer registers. 1330 RFDS only affects Intel Atom processors. 1331 1332 on: Turns ON the mitigation. 1333 off: Turns OFF the mitigation. 1334 1335 This parameter overrides the compile time default set 1336 by CONFIG_MITIGATION_RFDS. Mitigation cannot be 1337 disabled when other VERW based mitigations (like MDS) 1338 are enabled. In order to disable RFDS mitigation all 1339 VERW based mitigations need to be disabled. 1340 1341 For details see: 1342 Documentation/admin-guide/hw-vuln/reg-file-data-sampling.rst 1343 1344 dm_verity.keyring_unsealed= 1345 [KNL] When set to 1, leave the dm-verity keyring 1346 unsealed after initialization so userspace can 1347 provision keys. Once the keyring is restricted 1348 it becomes active and is searched during signature 1349 verification. 1350 1351 driver_async_probe= [KNL] 1352 List of driver names to be probed asynchronously. * 1353 matches with all driver names. If * is specified, the 1354 rest of the listed driver names are those that will NOT 1355 match the *. 1356 Format: <driver_name1>,<driver_name2>... 1357 1358 drm.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>] 1359 Broken monitors, graphic adapters, KVMs and EDIDless 1360 panels may send no or incorrect EDID data sets. 1361 This parameter allows to specify an EDID data sets 1362 in the /lib/firmware directory that are used instead. 1363 An EDID data set will only be used for a particular 1364 connector, if its name and a colon are prepended to 1365 the EDID name. Each connector may use a unique EDID 1366 data set by separating the files with a comma. An EDID 1367 data set with no connector name will be used for 1368 any connectors not explicitly specified. 1369 1370 dscc4.setup= [NET] 1371 1372 dt_cpu_ftrs= [PPC,EARLY] 1373 Format: {"off" | "known"} 1374 Control how the dt_cpu_ftrs device-tree binding is 1375 used for CPU feature discovery and setup (if it 1376 exists). 1377 off: Do not use it, fall back to legacy cpu table. 1378 known: Do not pass through unknown features to guests 1379 or userspace, only those that the kernel is aware of. 1380 1381 dump_apple_properties [X86] 1382 Dump name and content of EFI device properties on 1383 x86 Macs. Useful for driver authors to determine 1384 what data is available or for reverse-engineering. 1385 1386 dyndbg[="val"] [KNL,DYNAMIC_DEBUG] 1387 <module>.dyndbg[="val"] 1388 Enable debug messages at boot time. See 1389 Documentation/admin-guide/dynamic-debug-howto.rst 1390 for details. 1391 1392 early_ioremap_debug [KNL,EARLY] 1393 Enable debug messages in early_ioremap support. This 1394 is useful for tracking down temporary early mappings 1395 which are not unmapped. 1396 1397 earlycon= [KNL,EARLY] Output early console device and options. 1398 1399 When used with no options, the early console is 1400 determined by stdout-path property in device tree's 1401 chosen node or the ACPI SPCR table if supported by 1402 the platform. 1403 1404 cdns,<addr>[,options] 1405 Start an early, polled-mode console on a Cadence 1406 (xuartps) serial port at the specified address. Only 1407 supported option is baud rate. If baud rate is not 1408 specified, the serial port must already be setup and 1409 configured. 1410 1411 uart[8250],io,<addr>[,options[,uartclk]] 1412 uart[8250],mmio,<addr>[,options[,uartclk]] 1413 uart[8250],mmio32,<addr>[,options[,uartclk]] 1414 uart[8250],mmio32be,<addr>[,options[,uartclk]] 1415 uart[8250],0x<addr>[,options] 1416 Start an early, polled-mode console on the 8250/16550 1417 UART at the specified I/O port or MMIO address. 1418 MMIO inter-register address stride is either 8-bit 1419 (mmio) or 32-bit (mmio32 or mmio32be). 1420 If none of [io|mmio|mmio32|mmio32be], <addr> is assumed 1421 to be equivalent to 'mmio'. 'options' are specified 1422 in the same format described for "console=ttyS<n>"; if 1423 unspecified, the h/w is not initialized. 'uartclk' is 1424 the uart clock frequency; if unspecified, it is set 1425 to 'BASE_BAUD' * 16. 1426 1427 pl011,<addr> 1428 pl011,mmio32,<addr> 1429 Start an early, polled-mode console on a pl011 serial 1430 port at the specified address. The pl011 serial port 1431 must already be setup and configured. Options are not 1432 yet supported. If 'mmio32' is specified, then only 1433 the driver will use only 32-bit accessors to read/write 1434 the device registers. 1435 1436 liteuart,<addr> 1437 Start an early console on a litex serial port at the 1438 specified address. The serial port must already be 1439 setup and configured. Options are not yet supported. 1440 1441 meson,<addr> 1442 Start an early, polled-mode console on a meson serial 1443 port at the specified address. The serial port must 1444 already be setup and configured. Options are not yet 1445 supported. 1446 1447 msm_serial,<addr> 1448 Start an early, polled-mode console on an msm serial 1449 port at the specified address. The serial port 1450 must already be setup and configured. Options are not 1451 yet supported. 1452 1453 msm_serial_dm,<addr> 1454 Start an early, polled-mode console on an msm serial 1455 dm port at the specified address. The serial port 1456 must already be setup and configured. Options are not 1457 yet supported. 1458 1459 owl,<addr> 1460 Start an early, polled-mode console on a serial port 1461 of an Actions Semi SoC, such as S500 or S900, at the 1462 specified address. The serial port must already be 1463 setup and configured. Options are not yet supported. 1464 1465 rda,<addr> 1466 Start an early, polled-mode console on a serial port 1467 of an RDA Micro SoC, such as RDA8810PL, at the 1468 specified address. The serial port must already be 1469 setup and configured. Options are not yet supported. 1470 1471 sbi 1472 Use RISC-V SBI (Supervisor Binary Interface) for early 1473 console. 1474 1475 smh Use ARM semihosting calls for early console. 1476 1477 s3c2410,<addr> 1478 s3c2412,<addr> 1479 s3c2440,<addr> 1480 s3c6400,<addr> 1481 s5pv210,<addr> 1482 exynos4210,<addr> 1483 Use early console provided by serial driver available 1484 on Samsung SoCs, requires selecting proper type and 1485 a correct base address of the selected UART port. The 1486 serial port must already be setup and configured. 1487 Options are not yet supported. 1488 1489 lantiq,<addr> 1490 Start an early, polled-mode console on a lantiq serial 1491 (lqasc) port at the specified address. The serial port 1492 must already be setup and configured. Options are not 1493 yet supported. 1494 1495 lpuart,<addr> 1496 lpuart32,<addr> 1497 Use early console provided by Freescale LP UART driver 1498 found on Freescale Vybrid and QorIQ LS1021A processors. 1499 A valid base address must be provided, and the serial 1500 port must already be setup and configured. 1501 1502 ec_imx21,<addr> 1503 ec_imx6q,<addr> 1504 Start an early, polled-mode, output-only console on the 1505 Freescale i.MX UART at the specified address. The UART 1506 must already be setup and configured. 1507 1508 ar3700_uart,<addr> 1509 Start an early, polled-mode console on the 1510 Armada 3700 serial port at the specified 1511 address. The serial port must already be setup 1512 and configured. Options are not yet supported. 1513 1514 qcom_geni,<addr> 1515 Start an early, polled-mode console on a Qualcomm 1516 Generic Interface (GENI) based serial port at the 1517 specified address. The serial port must already be 1518 setup and configured. Options are not yet supported. 1519 1520 efifb,[options] 1521 Start an early, unaccelerated console on the EFI 1522 memory mapped framebuffer (if available). On cache 1523 coherent non-x86 systems that use system memory for 1524 the framebuffer, pass the 'ram' option so that it is 1525 mapped with the correct attributes. 1526 1527 linflex,<addr> 1528 Use early console provided by Freescale LINFlexD UART 1529 serial driver for NXP S32V234 SoCs. A valid base 1530 address must be provided, and the serial port must 1531 already be setup and configured. 1532 1533 earlyprintk= [X86,SH,ARM,M68k,S390,UM,EARLY] 1534 earlyprintk=vga 1535 earlyprintk=sclp 1536 earlyprintk=xen 1537 earlyprintk=serial[,ttySn[,baudrate]] 1538 earlyprintk=serial[,0x...[,baudrate]] 1539 earlyprintk=ttySn[,baudrate] 1540 earlyprintk=dbgp[debugController#] 1541 earlyprintk=mmio32,membase[,{nocfg|baudrate}] 1542 earlyprintk=pciserial[,force],bus:device.function[,{nocfg|baudrate}] 1543 earlyprintk=xdbc[xhciController#] 1544 earlyprintk=bios 1545 1546 earlyprintk is useful when the kernel crashes before 1547 the normal console is initialized. It is not enabled by 1548 default because it has some cosmetic problems. 1549 1550 Use "nocfg" to skip UART configuration, assume 1551 BIOS/firmware has configured UART correctly. 1552 1553 Append ",keep" to not disable it when the real console 1554 takes over. 1555 1556 Only one of vga, serial, or usb debug port can 1557 be used at a time. 1558 1559 Currently only ttyS0 and ttyS1 may be specified by 1560 name. Other I/O ports may be explicitly specified 1561 on some architectures (x86 and arm at least) by 1562 replacing ttySn with an I/O port address, like this: 1563 earlyprintk=serial,0x1008,115200 1564 You can find the port for a given device in 1565 /proc/tty/driver/serial: 1566 2: uart:ST16650V2 port:00001008 irq:18 ... 1567 1568 Interaction with the standard serial driver is not 1569 very good. 1570 1571 The VGA output is eventually overwritten by 1572 the real console. 1573 1574 The xen option can only be used in Xen domains. 1575 1576 The sclp output can only be used on s390. 1577 1578 The bios output can only be used on SuperH. 1579 1580 The optional "force" to "pciserial" enables use of a 1581 PCI device even when its classcode is not of the 1582 UART class. 1583 1584 edac_report= [HW,EDAC] Control how to report EDAC event 1585 Format: {"on" | "off" | "force"} 1586 on: enable EDAC to report H/W event. May be overridden 1587 by other higher priority error reporting module. 1588 off: disable H/W event reporting through EDAC. 1589 force: enforce the use of EDAC to report H/W event. 1590 default: on. 1591 1592 edd= [EDD] 1593 Format: {"off" | "on" | "skip[mbr]"} 1594 1595 efi= [EFI,EARLY] 1596 Format: { "debug", "disable_early_pci_dma", 1597 "nochunk", "noruntime", "nosoftreserve", 1598 "novamap", "no_disable_early_pci_dma" } 1599 debug: enable misc debug output. 1600 disable_early_pci_dma: disable the busmaster bit on all 1601 PCI bridges while in the EFI boot stub. 1602 nochunk: disable reading files in "chunks" in the EFI 1603 boot stub, as chunking can cause problems with some 1604 firmware implementations. 1605 noruntime : disable EFI runtime services support 1606 nosoftreserve: The EFI_MEMORY_SP (Specific Purpose) 1607 attribute may cause the kernel to reserve the 1608 memory range for a memory mapping driver to 1609 claim. Specify efi=nosoftreserve to disable this 1610 reservation and treat the memory by its base type 1611 (i.e. EFI_CONVENTIONAL_MEMORY / "System RAM"). 1612 novamap: do not call SetVirtualAddressMap(). 1613 no_disable_early_pci_dma: Leave the busmaster bit set 1614 on all PCI bridges while in the EFI boot stub 1615 1616 efi_no_storage_paranoia [EFI,X86,EARLY] 1617 Using this parameter you can use more than 50% of 1618 your efi variable storage. Use this parameter only if 1619 you are really sure that your UEFI does sane gc and 1620 fulfills the spec otherwise your board may brick. 1621 1622 efivar_ssdt= [EFI; X86] Name of an EFI variable that contains an SSDT 1623 that is to be dynamically loaded by Linux. If there are 1624 multiple variables with the same name but with different 1625 vendor GUIDs, all of them will be loaded. See 1626 Documentation/admin-guide/acpi/ssdt-overlays.rst for details. 1627 1628 1629 eisa_irq_edge= [PARISC,HW] 1630 See header of drivers/parisc/eisa.c. 1631 1632 ekgdboc= [X86,KGDB,EARLY] Allow early kernel console debugging 1633 Format: ekgdboc=kbd 1634 1635 This is designed to be used in conjunction with 1636 the boot argument: earlyprintk=vga 1637 1638 This parameter works in place of the kgdboc parameter 1639 but can only be used if the backing tty is available 1640 very early in the boot process. For early debugging 1641 via a serial port see kgdboc_earlycon instead. 1642 1643 elfcorehdr=[size[KMG]@]offset[KMG] [PPC,SH,X86,S390,EARLY] 1644 Specifies physical address of start of kernel core 1645 image elf header and optionally the size. Generally 1646 kexec loader will pass this option to capture kernel. 1647 See Documentation/admin-guide/kdump/kdump.rst for details. 1648 1649 enable_mtrr_cleanup [X86,EARLY] 1650 The kernel tries to adjust MTRR layout from continuous 1651 to discrete, to make X server driver able to add WB 1652 entry later. This parameter enables that. 1653 1654 enable_timer_pin_1 [X86] 1655 Enable PIN 1 of APIC timer 1656 Can be useful to work around chipset bugs 1657 (in particular on some ATI chipsets). 1658 The kernel tries to set a reasonable default. 1659 1660 enforcing= [SELINUX] Set initial enforcing status. 1661 Format: {"0" | "1"} 1662 See security/selinux/Kconfig help text. 1663 0 -- permissive (log only, no denials). 1664 1 -- enforcing (deny and log). 1665 Default value is 0. 1666 Value can be changed at runtime via 1667 /sys/fs/selinux/enforce. 1668 1669 erst_disable [ACPI] 1670 Disable Error Record Serialization Table (ERST) 1671 support. 1672 1673 ether= [HW,NET] Ethernet cards parameters 1674 This option is obsoleted by the "netdev=" option, which 1675 has equivalent usage. See its documentation for details. 1676 1677 evm= [EVM] 1678 Format: { "fix" } 1679 Permit 'security.evm' to be updated regardless of 1680 current integrity status. 1681 1682 early_page_ext [KNL,EARLY] Enforces page_ext initialization to earlier 1683 stages so cover more early boot allocations. 1684 Please note that as side effect some optimizations 1685 might be disabled to achieve that (e.g. parallelized 1686 memory initialization is disabled) so the boot process 1687 might take longer, especially on systems with a lot of 1688 memory. Available with CONFIG_PAGE_EXTENSION=y. 1689 1690 failslab= 1691 fail_usercopy= 1692 fail_page_alloc= 1693 fail_skb_realloc= 1694 fail_make_request=[KNL] 1695 General fault injection mechanism. 1696 Format: <interval>,<probability>,<space>,<times> 1697 See also Documentation/fault-injection/. 1698 1699 fb_tunnels= [NET] 1700 Format: { initns | none } 1701 See Documentation/admin-guide/sysctl/net.rst for 1702 fb_tunnels_only_for_init_ns 1703 1704 floppy= [HW] 1705 See Documentation/admin-guide/blockdev/floppy.rst. 1706 1707 forcepae [X86-32] 1708 Forcefully enable Physical Address Extension (PAE). 1709 Many Pentium M systems disable PAE but may have a 1710 functionally usable PAE implementation. 1711 Warning: use of this parameter will taint the kernel 1712 and may cause unknown problems. 1713 1714 fred= [X86-64] 1715 Enable/disable Flexible Return and Event Delivery. 1716 Format: { on | off } 1717 on: enable FRED when it's present, the default setting. 1718 off: disable FRED. 1719 1720 ftrace=[tracer] 1721 [FTRACE] will set and start the specified tracer 1722 as early as possible in order to facilitate early 1723 boot debugging. 1724 1725 ftrace_boot_snapshot 1726 [FTRACE] On boot up, a snapshot will be taken of the 1727 ftrace ring buffer that can be read at: 1728 /sys/kernel/tracing/snapshot. 1729 This is useful if you need tracing information from kernel 1730 boot up that is likely to be overridden by user space 1731 start up functionality. 1732 1733 Optionally, the snapshot can also be defined for a tracing 1734 instance that was created by the trace_instance= command 1735 line parameter. 1736 1737 trace_instance=foo,sched_switch ftrace_boot_snapshot=foo 1738 1739 The above will cause the "foo" tracing instance to trigger 1740 a snapshot at the end of boot up. 1741 1742 ftrace_dump_on_oops[=2(orig_cpu) | =<instance>][,<instance> | 1743 ,<instance>=2(orig_cpu)] 1744 [FTRACE] will dump the trace buffers on oops. 1745 If no parameter is passed, ftrace will dump global 1746 buffers of all CPUs, if you pass 2 or orig_cpu, it 1747 will dump only the buffer of the CPU that triggered 1748 the oops, or the specific instance will be dumped if 1749 its name is passed. Multiple instance dump is also 1750 supported, and instances are separated by commas. Each 1751 instance supports only dump on CPU that triggered the 1752 oops by passing 2 or orig_cpu to it. 1753 1754 ftrace_dump_on_oops=foo=orig_cpu 1755 1756 The above will dump only the buffer of "foo" instance 1757 on CPU that triggered the oops. 1758 1759 ftrace_dump_on_oops,foo,bar=orig_cpu 1760 1761 The above will dump global buffer on all CPUs, the 1762 buffer of "foo" instance on all CPUs and the buffer 1763 of "bar" instance on CPU that triggered the oops. 1764 1765 ftrace_filter=[function-list] 1766 [FTRACE] Limit the functions traced by the function 1767 tracer at boot up. function-list is a comma-separated 1768 list of functions. This list can be changed at run 1769 time by the set_ftrace_filter file in the debugfs 1770 tracing directory. 1771 1772 ftrace_notrace=[function-list] 1773 [FTRACE] Do not trace the functions specified in 1774 function-list. This list can be changed at run time 1775 by the set_ftrace_notrace file in the debugfs 1776 tracing directory. 1777 1778 ftrace_graph_filter=[function-list] 1779 [FTRACE] Limit the top level callers functions traced 1780 by the function graph tracer at boot up. 1781 function-list is a comma-separated list of functions 1782 that can be changed at run time by the 1783 set_graph_function file in the debugfs tracing directory. 1784 1785 ftrace_graph_notrace=[function-list] 1786 [FTRACE] Do not trace from the functions specified in 1787 function-list. This list is a comma-separated list of 1788 functions that can be changed at run time by the 1789 set_graph_notrace file in the debugfs tracing directory. 1790 1791 ftrace_graph_max_depth=<uint> 1792 [FTRACE] Used with the function graph tracer. This is 1793 the max depth it will trace into a function. This value 1794 can be changed at run time by the max_graph_depth file 1795 in the tracefs tracing directory. default: 0 (no limit) 1796 1797 fw_devlink= [KNL,EARLY] Create device links between consumer and supplier 1798 devices by scanning the firmware to infer the 1799 consumer/supplier relationships. This feature is 1800 especially useful when drivers are loaded as modules as 1801 it ensures proper ordering of tasks like device probing 1802 (suppliers first, then consumers), supplier boot state 1803 clean up (only after all consumers have probed), 1804 suspend/resume & runtime PM (consumers first, then 1805 suppliers). 1806 Format: { off | permissive | on | rpm } 1807 off -- Don't create device links from firmware info. 1808 permissive -- Create device links from firmware info 1809 but use it only for ordering boot state clean 1810 up (sync_state() calls). 1811 on -- Create device links from firmware info and use it 1812 to enforce probe and suspend/resume ordering. 1813 rpm -- Like "on", but also use to order runtime PM. 1814 1815 fw_devlink.strict=<bool> 1816 [KNL,EARLY] Treat all inferred dependencies as mandatory 1817 dependencies. This only applies for fw_devlink=on|rpm. 1818 Format: <bool> 1819 1820 fw_devlink.sync_state = 1821 [KNL,EARLY] When all devices that could probe have finished 1822 probing, this parameter controls what to do with 1823 devices that haven't yet received their sync_state() 1824 calls. 1825 Format: { strict | timeout } 1826 strict -- Default. Continue waiting on consumers to 1827 probe successfully. 1828 timeout -- Give up waiting on consumers and call 1829 sync_state() on any devices that haven't yet 1830 received their sync_state() calls after 1831 deferred_probe_timeout has expired or by 1832 late_initcall() if !CONFIG_MODULES. 1833 1834 gamecon.map[2|3]= 1835 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad 1836 support via parallel port (up to 5 devices per port) 1837 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5> 1838 See also Documentation/input/devices/joystick-parport.rst 1839 1840 gamma= [HW,DRM] 1841 1842 gart_fix_e820= [X86-64,EARLY] disable the fix e820 for K8 GART 1843 Format: off | on 1844 default: on 1845 1846 gather_data_sampling= 1847 [X86,INTEL,EARLY] Control the Gather Data Sampling (GDS) 1848 mitigation. 1849 1850 Gather Data Sampling is a hardware vulnerability which 1851 allows unprivileged speculative access to data which was 1852 previously stored in vector registers. 1853 1854 This issue is mitigated by default in updated microcode. 1855 The mitigation may have a performance impact but can be 1856 disabled. On systems without the microcode mitigation 1857 disabling AVX serves as a mitigation. 1858 1859 force: Disable AVX to mitigate systems without 1860 microcode mitigation. No effect if the microcode 1861 mitigation is present. Known to cause crashes in 1862 userspace with buggy AVX enumeration. 1863 1864 off: Disable GDS mitigation. 1865 1866 gbpages [X86] Use GB pages for kernel direct mappings. 1867 1868 gcov_persist= [GCOV] When non-zero (default), profiling data for 1869 kernel modules is saved and remains accessible via 1870 debugfs, even when the module is unloaded/reloaded. 1871 When zero, profiling data is discarded and associated 1872 debugfs files are removed at module unload time. 1873 1874 goldfish [X86] Enable the goldfish android emulator platform. 1875 Don't use this when you are not running on the 1876 android emulator 1877 1878 gpio-mockup.gpio_mockup_ranges 1879 [HW] Sets the ranges of gpiochip of for this device. 1880 Format: <start1>,<end1>,<start2>,<end2>... 1881 gpio-mockup.gpio_mockup_named_lines 1882 [HW] Let the driver know GPIO lines should be named. 1883 1884 gpt [EFI] Forces disk with valid GPT signature but 1885 invalid Protective MBR to be treated as GPT. If the 1886 primary GPT is corrupted, it enables the backup/alternate 1887 GPT to be used instead. 1888 1889 grcan.enable0= [HW] Configuration of physical interface 0. Determines 1890 the "Enable 0" bit of the configuration register. 1891 Format: 0 | 1 1892 Default: 0 1893 grcan.enable1= [HW] Configuration of physical interface 1. Determines 1894 the "Enable 0" bit of the configuration register. 1895 Format: 0 | 1 1896 Default: 0 1897 grcan.select= [HW] Select which physical interface to use. 1898 Format: 0 | 1 1899 Default: 0 1900 grcan.txsize= [HW] Sets the size of the tx buffer. 1901 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0. 1902 Default: 1024 1903 grcan.rxsize= [HW] Sets the size of the rx buffer. 1904 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0. 1905 Default: 1024 1906 1907 hardened_usercopy= 1908 [KNL] Under CONFIG_HARDENED_USERCOPY, whether 1909 hardening is enabled for this boot. Hardened 1910 usercopy checking is used to protect the kernel 1911 from reading or writing beyond known memory 1912 allocation boundaries as a proactive defense 1913 against bounds-checking flaws in the kernel's 1914 copy_to_user()/copy_from_user() interface. 1915 The default is determined by 1916 CONFIG_HARDENED_USERCOPY_DEFAULT_ON. 1917 on Perform hardened usercopy checks. 1918 off Disable hardened usercopy checks. 1919 1920 hardlockup_all_cpu_backtrace= 1921 [KNL] Should the hard-lockup detector generate 1922 backtraces on all cpus. 1923 Format: 0 | 1 1924 1925 hash_pointers= 1926 [KNL,EARLY] 1927 By default, when pointers are printed to the console 1928 or buffers via the %p format string, that pointer is 1929 "hashed", i.e. obscured by hashing the pointer value. 1930 This is a security feature that hides actual kernel 1931 addresses from unprivileged users, but it also makes 1932 debugging the kernel more difficult since unequal 1933 pointers can no longer be compared. The choices are: 1934 Format: { auto | always | never } 1935 Default: auto 1936 1937 auto - Hash pointers unless slab_debug is enabled. 1938 always - Always hash pointers (even if slab_debug is 1939 enabled). 1940 never - Never hash pointers. This option should only 1941 be specified when debugging the kernel. Do 1942 not use on production kernels. The boot 1943 param "no_hash_pointers" is an alias for 1944 this mode. 1945 1946 For controlling hashing dynamically at runtime, 1947 use the "kernel.kptr_restrict" sysctl instead. 1948 1949 hashdist= [KNL,NUMA] Large hashes allocated during boot 1950 are distributed across NUMA nodes. Defaults on 1951 for 64-bit NUMA, off otherwise. 1952 Format: 0 | 1 (for off | on) 1953 1954 hd= [EIDE] (E)IDE hard drive subsystem geometry 1955 Format: <cyl>,<head>,<sect> 1956 1957 hest_disable [ACPI] 1958 Disable Hardware Error Source Table (HEST) support; 1959 corresponding firmware-first mode error processing 1960 logic will be disabled. 1961 1962 hibernate= [HIBERNATION] 1963 noresume Don't check if there's a hibernation image 1964 present during boot. 1965 nocompress Don't compress/decompress hibernation images. 1966 no Disable hibernation and resume. 1967 protect_image Turn on image protection during restoration 1968 (that will set all pages holding image data 1969 during restoration read-only). 1970 1971 hibernate.compressor= [HIBERNATION] Compression algorithm to be 1972 used with hibernation. 1973 Format: { lzo | lz4 } 1974 Default: lzo 1975 1976 lzo: Select LZO compression algorithm to 1977 compress/decompress hibernation image. 1978 1979 lz4: Select LZ4 compression algorithm to 1980 compress/decompress hibernation image. 1981 1982 hibernate.pm_test_delay= 1983 [HIBERNATION] 1984 Sets the number of seconds to remain in a hibernation test 1985 mode before resuming the system (see 1986 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG 1987 is set. Default value is 5. 1988 1989 hibernate_compression_threads= 1990 [HIBERNATION] 1991 Set the number of threads used for compressing or decompressing 1992 hibernation images. 1993 1994 Format: <integer> 1995 Default: 3 1996 Minimum: 1 1997 Example: hibernate_compression_threads=4 1998 1999 highmem=nn[KMG] [KNL,BOOT,EARLY] forces the highmem zone to have an exact 2000 size of <nn>. This works even on boxes that have no 2001 highmem otherwise. This also works to reduce highmem 2002 size on bigger boxes. 2003 2004 highres= [KNL] Enable/disable high resolution timer mode. 2005 Valid parameters: "on", "off" 2006 Default: "on" 2007 2008 hlt [BUGS=ARM,SH] 2009 2010 hostname= [KNL,EARLY] Set the hostname (aka UTS nodename). 2011 Format: <string> 2012 This allows setting the system's hostname during early 2013 startup. This sets the name returned by gethostname. 2014 Using this parameter to set the hostname makes it 2015 possible to ensure the hostname is correctly set before 2016 any userspace processes run, avoiding the possibility 2017 that a process may call gethostname before the hostname 2018 has been explicitly set, resulting in the calling 2019 process getting an incorrect result. The string must 2020 not exceed the maximum allowed hostname length (usually 2021 64 characters) and will be truncated otherwise. 2022 2023 hpet= [X86-32,HPET] option to control HPET usage 2024 Format: { enable (default) | disable | force | 2025 verbose } 2026 disable: disable HPET and use PIT instead 2027 force: allow force enabled of undocumented chips (ICH4, 2028 VIA, nVidia) 2029 verbose: show contents of HPET registers during setup 2030 2031 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET 2032 registers. Default set by CONFIG_HPET_MMAP_DEFAULT. 2033 2034 hugepages= [HW,EARLY] Number of HugeTLB pages to allocate at boot. 2035 If this follows hugepagesz (below), it specifies 2036 the number of pages of hugepagesz to be allocated. 2037 If this is the first HugeTLB parameter on the command 2038 line, it specifies the number of pages to allocate for 2039 the default huge page size. If using node format, the 2040 number of pages to allocate per-node can be specified. 2041 See also Documentation/admin-guide/mm/hugetlbpage.rst. 2042 Format: <integer> or (node format) 2043 <node>:<integer>[,<node>:<integer>] 2044 2045 hugepagesz= 2046 [HW,EARLY] The size of the HugeTLB pages. This is 2047 used in conjunction with hugepages (above) to 2048 allocate huge pages of a specific size at boot. The 2049 pair hugepagesz=X hugepages=Y can be specified once 2050 for each supported huge page size. Huge page sizes 2051 are architecture dependent. See also 2052 Documentation/admin-guide/mm/hugetlbpage.rst. 2053 Format: size[KMG] 2054 2055 hugepage_alloc_threads= 2056 [HW] The number of threads that should be used to 2057 allocate hugepages during boot. This option can be 2058 used to improve system bootup time when allocating 2059 a large amount of huge pages. 2060 The default value is 25% of the available hardware threads. 2061 2062 Note that this parameter only applies to non-gigantic huge pages. 2063 2064 hugetlb_cma= [HW,CMA,EARLY] The size of a CMA area used for allocation 2065 of gigantic hugepages. Or using node format, the size 2066 of a CMA area per node can be specified. 2067 Format: nn[KMGTPE] or (node format) 2068 <node>:nn[KMGTPE][,<node>:nn[KMGTPE]] 2069 2070 The size must be a multiple of the gigantic page size. 2071 When using node format, this applies to each per-node size. 2072 Missaligned values are dropped with a warning. 2073 2074 Reserve a CMA area of given size and allocate gigantic 2075 hugepages using the CMA allocator. If enabled, the 2076 boot-time allocation of gigantic hugepages is skipped. 2077 2078 hugetlb_cma_only= 2079 [HW,CMA,EARLY] When allocating new HugeTLB pages, only 2080 try to allocate from the CMA areas. 2081 2082 This option does nothing if hugetlb_cma= is not also 2083 specified. 2084 2085 hugetlb_free_vmemmap= 2086 [KNL] Requires CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP 2087 enabled. 2088 Control if HugeTLB Vmemmap Optimization (HVO) is enabled. 2089 Allows heavy hugetlb users to free up some more 2090 memory (7 * PAGE_SIZE for each 2MB hugetlb page). 2091 Format: { on | off (default) } 2092 2093 on: enable HVO 2094 off: disable HVO 2095 2096 Built with CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP_DEFAULT_ON=y, 2097 the default is on. 2098 2099 Note that the vmemmap pages may be allocated from the added 2100 memory block itself when memory_hotplug.memmap_on_memory is 2101 enabled, those vmemmap pages cannot be optimized even if this 2102 feature is enabled. Other vmemmap pages not allocated from 2103 the added memory block itself do not be affected. 2104 2105 hung_task_panic= 2106 [KNL] Number of hung tasks to trigger kernel panic. 2107 Format: <int> 2108 2109 When set to a non-zero value, a kernel panic will be triggered if 2110 the number of detected hung tasks reaches this value. 2111 2112 0: don't panic 2113 1: panic immediately on first hung task 2114 N: panic after N hung tasks are detected in a single scan 2115 2116 The default value is controlled by the 2117 CONFIG_BOOTPARAM_HUNG_TASK_PANIC build-time option. The value 2118 selected by this boot parameter can be changed later by the 2119 kernel.hung_task_panic sysctl. 2120 2121 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC) 2122 terminal devices. Valid values: 0..8 2123 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs. 2124 If specified, z/VM IUCV HVC accepts connections 2125 from listed z/VM user IDs only. 2126 2127 hv_nopvspin [X86,HYPER_V,EARLY] 2128 Disables the paravirt spinlock optimizations 2129 which allow the hypervisor to 'idle' the guest 2130 on lock contention. 2131 2132 hw_protection= [HW] 2133 Format: reboot | shutdown 2134 2135 Hardware protection action taken on critical events like 2136 overtemperature or imminent voltage loss. 2137 2138 i2c_bus= [HW] Override the default board specific I2C bus speed 2139 or register an additional I2C bus that is not 2140 registered from board initialization code. 2141 Format: 2142 <bus_id>,<clkrate> 2143 2144 i2c_touchscreen_props= [HW,ACPI,X86] 2145 Set device-properties for ACPI-enumerated I2C-attached 2146 touchscreen, to e.g. fix coordinates of upside-down 2147 mounted touchscreens. If you need this option please 2148 submit a drivers/platform/x86/touchscreen_dmi.c patch 2149 adding a DMI quirk for this. 2150 2151 Format: 2152 <ACPI_HW_ID>:<prop_name>=<val>[:prop_name=val][:...] 2153 Where <val> is one of: 2154 Omit "=<val>" entirely Set a boolean device-property 2155 Unsigned number Set a u32 device-property 2156 Anything else Set a string device-property 2157 2158 Examples (split over multiple lines): 2159 i2c_touchscreen_props=GDIX1001:touchscreen-inverted-x: 2160 touchscreen-inverted-y 2161 2162 i2c_touchscreen_props=MSSL1680:touchscreen-size-x=1920: 2163 touchscreen-size-y=1080:touchscreen-inverted-y: 2164 firmware-name=gsl1680-vendor-model.fw:silead,home-button 2165 2166 i8042.debug [HW] Toggle i8042 debug mode 2167 i8042.unmask_kbd_data 2168 [HW] Enable printing of interrupt data from the KBD port 2169 (disabled by default, and as a pre-condition 2170 requires that i8042.debug=1 be enabled) 2171 i8042.direct [HW] Put keyboard port into non-translated mode 2172 i8042.dumbkbd [HW] Pretend that controller can only read data from 2173 keyboard and cannot control its state 2174 (Don't attempt to blink the leds) 2175 i8042.noaux [HW] Don't check for auxiliary (== mouse) port 2176 i8042.nokbd [HW] Don't check/create keyboard port 2177 i8042.noloop [HW] Disable the AUX Loopback command while probing 2178 for the AUX port 2179 i8042.nomux [HW] Don't check presence of an active multiplexing 2180 controller 2181 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX 2182 controllers 2183 i8042.notimeout [HW] Ignore timeout condition signalled by controller 2184 i8042.reset [HW] Reset the controller during init, cleanup and 2185 suspend-to-ram transitions, only during s2r 2186 transitions, or never reset 2187 Format: { 1 | Y | y | 0 | N | n } 2188 1, Y, y: always reset controller 2189 0, N, n: don't ever reset controller 2190 Default: only on s2r transitions on x86; most other 2191 architectures force reset to be always executed 2192 i8042.unlock [HW] Unlock (ignore) the keylock 2193 i8042.kbdreset [HW] Reset device connected to KBD port 2194 i8042.probe_defer 2195 [HW] Allow deferred probing upon i8042 probe errors 2196 2197 i810= [HW,DRM] 2198 2199 i915.invert_brightness= 2200 [DRM] Invert the sense of the variable that is used to 2201 set the brightness of the panel backlight. Normally a 2202 brightness value of 0 indicates backlight switched off, 2203 and the maximum of the brightness value sets the backlight 2204 to maximum brightness. If this parameter is set to 0 2205 (default) and the machine requires it, or this parameter 2206 is set to 1, a brightness value of 0 sets the backlight 2207 to maximum brightness, and the maximum of the brightness 2208 value switches the backlight off. 2209 -1 -- never invert brightness 2210 0 -- machine default 2211 1 -- force brightness inversion 2212 2213 ia32_emulation= [X86-64] 2214 Format: <bool> 2215 When true, allows loading 32-bit programs and executing 32-bit 2216 syscalls, essentially overriding IA32_EMULATION_DEFAULT_DISABLED at 2217 boot time. When false, unconditionally disables IA32 emulation. 2218 2219 2220 idle= [X86,EARLY] 2221 Format: idle=poll, idle=halt, idle=nomwait 2222 2223 idle=poll: Don't do power saving in the idle loop 2224 using HLT, but poll for rescheduling event. This will 2225 make the CPUs eat a lot more power, but may be useful 2226 to get slightly better performance in multiprocessor 2227 benchmarks. It also makes some profiling using 2228 performance counters more accurate. Please note that 2229 on systems with MONITOR/MWAIT support (like Intel 2230 EM64T CPUs) this option has no performance advantage 2231 over the normal idle loop. It may also interact badly 2232 with hyperthreading. 2233 2234 idle=halt: Halt is forced to be used for CPU idle. 2235 In such case C2/C3 won't be used again. 2236 2237 idle=nomwait: Disable mwait for CPU C-states 2238 2239 idxd.sva= [HW] 2240 Format: <bool> 2241 Allow force disabling of Shared Virtual Memory (SVA) 2242 support for the idxd driver. By default it is set to 2243 true (1). 2244 2245 idxd.tc_override= [HW] 2246 Format: <bool> 2247 Allow override of default traffic class configuration 2248 for the device. By default it is set to false (0). 2249 2250 ieee754= [MIPS] Select IEEE Std 754 conformance mode 2251 Format: { strict | legacy | 2008 | relaxed | emulated } 2252 Default: strict 2253 2254 Choose which programs will be accepted for execution 2255 based on the IEEE 754 NaN encoding(s) supported by 2256 the FPU and the NaN encoding requested with the value 2257 of an ELF file header flag individually set by each 2258 binary. Hardware implementations are permitted to 2259 support either or both of the legacy and the 2008 NaN 2260 encoding mode. 2261 2262 Available settings are as follows: 2263 strict accept binaries that request a NaN encoding 2264 supported by the FPU 2265 legacy only accept legacy-NaN binaries, if supported 2266 by the FPU 2267 2008 only accept 2008-NaN binaries, if supported 2268 by the FPU 2269 relaxed accept any binaries regardless of whether 2270 supported by the FPU 2271 emulated accept any binaries but enable FPU emulator 2272 if binary mode is unsupported by the FPU. 2273 2274 The FPU emulator is always able to support both NaN 2275 encodings, so if no FPU hardware is present or it has 2276 been disabled with 'nofpu', then the settings of 2277 'legacy' and '2008' strap the emulator accordingly, 2278 'relaxed' straps the emulator for both legacy-NaN and 2279 2008-NaN, whereas 'strict' enables legacy-NaN only on 2280 legacy processors and both NaN encodings on MIPS32 or 2281 MIPS64 CPUs. 2282 2283 The setting for ABS.fmt/NEG.fmt instruction execution 2284 mode generally follows that for the NaN encoding, 2285 except where unsupported by hardware. 2286 2287 ignore_loglevel [KNL,EARLY] 2288 Ignore loglevel setting - this will print /all/ 2289 kernel messages to the console. Useful for debugging. 2290 We also add it as printk module parameter, so users 2291 could change it dynamically, usually by 2292 /sys/module/printk/parameters/ignore_loglevel. 2293 2294 ignore_rlimit_data 2295 Ignore RLIMIT_DATA setting for data mappings, 2296 print warning at first misuse. Can be changed via 2297 /sys/module/kernel/parameters/ignore_rlimit_data. 2298 2299 ihash_entries= [KNL] 2300 Set number of hash buckets for inode cache. 2301 2302 ima_appraise= [IMA] appraise integrity measurements 2303 Format: { "off" | "enforce" | "fix" | "log" } 2304 default: "enforce" 2305 2306 ima_appraise_tcb [IMA] Deprecated. Use ima_policy= instead. 2307 The builtin appraise policy appraises all files 2308 owned by uid=0. 2309 2310 ima_canonical_fmt [IMA] 2311 Use the canonical format for the binary runtime 2312 measurements, instead of host native format. 2313 2314 ima_flush_htable [IMA] 2315 Flush the IMA hash table when deleting all the 2316 staged measurement records, to achieve maximum 2317 memory saving at the cost of having duplicate 2318 records across the staged measurement lists. 2319 2320 ima_hash= [IMA] 2321 Format: { md5 | sha1 | rmd160 | sha256 | sha384 2322 | sha512 | ... } 2323 default: "sha1" 2324 2325 The list of supported hash algorithms is defined 2326 in crypto/hash_info.h. 2327 2328 ima_policy= [IMA] 2329 The builtin policies to load during IMA setup. 2330 Format: "tcb | appraise_tcb | secure_boot | 2331 fail_securely | critical_data" 2332 2333 The "tcb" policy measures all programs exec'd, files 2334 mmap'd for exec, and all files opened with the read 2335 mode bit set by either the effective uid (euid=0) or 2336 uid=0. 2337 2338 The "appraise_tcb" policy appraises the integrity of 2339 all files owned by root. 2340 2341 The "secure_boot" policy appraises the integrity 2342 of files (eg. kexec kernel image, kernel modules, 2343 firmware, policy, etc) based on file signatures. 2344 2345 The "fail_securely" policy forces file signature 2346 verification failure also on privileged mounted 2347 filesystems with the SB_I_UNVERIFIABLE_SIGNATURE 2348 flag. 2349 2350 The "critical_data" policy measures kernel integrity 2351 critical data. 2352 2353 ima_tcb [IMA] Deprecated. Use ima_policy= instead. 2354 Load a policy which meets the needs of the Trusted 2355 Computing Base. This means IMA will measure all 2356 programs exec'd, files mmap'd for exec, and all files 2357 opened for read by uid=0. 2358 2359 ima_template= [IMA] 2360 Select one of defined IMA measurements template formats. 2361 Formats: { "ima" | "ima-ng" | "ima-ngv2" | "ima-sig" | 2362 "ima-sigv2" } 2363 Default: "ima-ng" 2364 2365 ima_template_fmt= 2366 [IMA] Define a custom template format. 2367 Format: { "field1|...|fieldN" } 2368 2369 ima= [IMA] Enable or disable IMA 2370 Format: { "off" | "on" } 2371 Default: "on" 2372 Note that disabling IMA is limited to kdump kernel. 2373 2374 indirect_target_selection= [X86,Intel] Mitigation control for Indirect 2375 Target Selection(ITS) bug in Intel CPUs. Updated 2376 microcode is also required for a fix in IBPB. 2377 2378 on: Enable mitigation (default). 2379 off: Disable mitigation. 2380 force: Force the ITS bug and deploy default 2381 mitigation. 2382 vmexit: Only deploy mitigation if CPU is affected by 2383 guest/host isolation part of ITS. 2384 stuff: Deploy RSB-fill mitigation when retpoline is 2385 also deployed. Otherwise, deploy the default 2386 mitigation. 2387 2388 For details see: 2389 Documentation/admin-guide/hw-vuln/indirect-target-selection.rst 2390 2391 init= [KNL] 2392 Format: <full_path> 2393 Run specified binary instead of /sbin/init as init 2394 process. 2395 2396 initcall_debug [KNL] Trace initcalls as they are executed. Useful 2397 for working out where the kernel is dying during 2398 startup. 2399 2400 initcall_blacklist= [KNL] Do not execute a comma-separated list of 2401 initcall functions. Useful for debugging built-in 2402 modules and initcalls. 2403 2404 initramfs_async= [KNL] 2405 Format: <bool> 2406 Default: 1 2407 This parameter controls whether the initramfs 2408 image is unpacked asynchronously, concurrently 2409 with devices being probed and 2410 initialized. This should normally just work, 2411 but as a debugging aid, one can get the 2412 historical behaviour of the initramfs 2413 unpacking being completed before device_ and 2414 late_ initcalls. 2415 2416 initrd= [BOOT,EARLY] Specify the location of the initial ramdisk 2417 2418 initrdmem= [KNL,EARLY] Specify a physical address and size from which to 2419 load the initrd. If an initrd is compiled in or 2420 specified in the bootparams, it takes priority over this 2421 setting. 2422 Format: ss[KMG],nn[KMG] 2423 Default is 0, 0 2424 2425 init_on_alloc= [MM,EARLY] Fill newly allocated pages and heap objects with 2426 zeroes. 2427 Format: 0 | 1 2428 Default set by CONFIG_INIT_ON_ALLOC_DEFAULT_ON. 2429 2430 init_on_free= [MM,EARLY] Fill freed pages and heap objects with zeroes. 2431 Format: 0 | 1 2432 Default set by CONFIG_INIT_ON_FREE_DEFAULT_ON. 2433 2434 init_pkru= [X86] Specify the default memory protection keys rights 2435 register contents for all processes. 0x55555554 by 2436 default (disallow access to all but pkey 0). Can 2437 override in debugfs after boot. 2438 2439 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver 2440 Format: <irq> 2441 2442 int_pln_enable [X86] Enable power limit notification interrupt 2443 2444 integrity_audit=[IMA] 2445 Format: { "0" | "1" } 2446 0 -- basic integrity auditing messages. (Default) 2447 1 -- additional integrity auditing messages. 2448 2449 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option 2450 on 2451 Enable intel iommu driver. 2452 off 2453 Disable intel iommu driver. 2454 igfx_off [Default Off] 2455 By default, gfx is mapped as normal device. If a gfx 2456 device has a dedicated DMAR unit, the DMAR unit is 2457 bypassed by not enabling DMAR with this option. In 2458 this case, gfx device will use physical address for 2459 DMA. 2460 strict [Default Off] 2461 Deprecated, equivalent to iommu.strict=1. 2462 sp_off [Default Off] 2463 By default, super page will be supported if Intel IOMMU 2464 has the capability. With this option, super page will 2465 not be supported. 2466 sm_on 2467 Enable the Intel IOMMU scalable mode if the hardware 2468 advertises that it has support for the scalable mode 2469 translation. 2470 sm_off 2471 Disallow use of the Intel IOMMU scalable mode. 2472 tboot_noforce [Default Off] 2473 Do not force the Intel IOMMU enabled under tboot. 2474 By default, tboot will force Intel IOMMU on, which 2475 could harm performance of some high-throughput 2476 devices like 40GBit network cards, even if identity 2477 mapping is enabled. 2478 Note that using this option lowers the security 2479 provided by tboot because it makes the system 2480 vulnerable to DMA attacks. 2481 2482 intel_idle.max_cstate= [KNL,HW,ACPI,X86] 2483 0 disables intel_idle and fall back on acpi_idle. 2484 1 to 9 specify maximum depth of C-state. 2485 2486 intel_pstate= [X86,EARLY] 2487 disable 2488 Do not enable intel_pstate as the default 2489 scaling driver for the supported processors 2490 active 2491 Use intel_pstate driver to bypass the scaling 2492 governors layer of cpufreq and provides it own 2493 algorithms for p-state selection. There are two 2494 P-state selection algorithms provided by 2495 intel_pstate in the active mode: powersave and 2496 performance. The way they both operate depends 2497 on whether or not the hardware managed P-states 2498 (HWP) feature has been enabled in the processor 2499 and possibly on the processor model. 2500 passive 2501 Use intel_pstate as a scaling driver, but configure it 2502 to work with generic cpufreq governors (instead of 2503 enabling its internal governor). This mode cannot be 2504 used along with the hardware-managed P-states (HWP) 2505 feature. 2506 force 2507 Enable intel_pstate on systems that prohibit it by default 2508 in favor of acpi-cpufreq. Forcing the intel_pstate driver 2509 instead of acpi-cpufreq may disable platform features, such 2510 as thermal controls and power capping, that rely on ACPI 2511 P-States information being indicated to OSPM and therefore 2512 should be used with caution. This option does not work with 2513 processors that aren't supported by the intel_pstate driver 2514 or on platforms that use pcc-cpufreq instead of acpi-cpufreq. 2515 no_hwp 2516 Do not enable hardware P state control (HWP) 2517 if available. 2518 hwp_only 2519 Only load intel_pstate on systems which support 2520 hardware P state control (HWP) if available. 2521 support_acpi_ppc 2522 Enforce ACPI _PPC performance limits. If the Fixed ACPI 2523 Description Table, specifies preferred power management 2524 profile as "Enterprise Server" or "Performance Server", 2525 then this feature is turned on by default. 2526 per_cpu_perf_limits 2527 Allow per-logical-CPU P-State performance control limits using 2528 cpufreq sysfs interface 2529 no_cas 2530 Do not enable capacity-aware scheduling (CAS) on 2531 hybrid systems 2532 2533 intremap= [X86-64,Intel-IOMMU,EARLY] 2534 on enable Interrupt Remapping (default) 2535 off disable Interrupt Remapping 2536 nosid disable Source ID checking 2537 no_x2apic_optout 2538 BIOS x2APIC opt-out request will be ignored 2539 nopost disable Interrupt Posting 2540 posted_msi 2541 enable MSIs delivered as posted interrupts 2542 2543 iomem= Disable strict checking of access to MMIO memory 2544 strict regions from userspace. 2545 relaxed 2546 2547 iommu= [X86,EARLY] 2548 2549 off 2550 Don't initialize and use any kind of IOMMU. 2551 2552 force 2553 Force the use of the hardware IOMMU even when 2554 it is not actually needed (e.g. because < 3 GB 2555 memory). 2556 2557 noforce 2558 Don't force hardware IOMMU usage when it is not 2559 needed. (default). 2560 2561 merge 2562 Do scatter-gather (SG) merging. Implies "force" 2563 (experimental). 2564 2565 nomerge 2566 Don't do scatter-gather (SG) merging. 2567 2568 biomerge 2569 Do scatter-gather (SG) merging. Implies "force" 2570 (experimental). [same as "merge"] 2571 2572 panic 2573 Always panic when IOMMU overflows. 2574 2575 nopanic 2576 Don't panic on IOMMU overflows. 2577 2578 pt 2579 Use passththrough mode by default 2580 (Equivalent to iommu.passthrough=1) 2581 2582 nopt 2583 Use translated mode for DMA by default 2584 (Equivalent to iommu.passthrough=0) 2585 2586 soft 2587 Use software bounce buffering (SWIOTLB) (default for 2588 Intel machines). This can be used to prevent the usage 2589 of an available hardware IOMMU. 2590 2591 usedac 2592 Use the DAC on VIA PCI bridge 2593 (default: disable the VIA PCI bridge DAC) 2594 2595 AMD Gart HW IOMMU-specific options: (CONFIG_GART_IOMMU) 2596 2597 <size> 2598 Set the size of the remapping area in bytes. 2599 2600 allowed 2601 Overwrite iommu off workarounds for specific chipsets 2602 2603 force 2604 Overwrite iommu off workarounds for specific chipsets 2605 2606 fullflush 2607 Flush IOMMU on each allocation (default). 2608 2609 nofullflush 2610 Don't use IOMMU fullflush. 2611 2612 memaper[=<order>] 2613 Allocate an own aperture over RAM with size 2614 32MB<<order. (default: order=1, i.e. 64MB) 2615 2616 noaperture 2617 Ask the IOMMU not to touch the aperture for AGP. 2618 2619 noagp 2620 Don't initialize the AGP driver and use full aperture. 2621 2622 iommu= [PPC/POWERNV] 2623 2624 nobypass 2625 Disable IOMMU bypass, using IOMMU for PCI devices. 2626 2627 iommu.forcedac= [ARM64,X86,EARLY] Control IOVA allocation for PCI devices. 2628 Format: { "0" | "1" } 2629 0 - Try to allocate a 32-bit DMA address first, before 2630 falling back to the full range if needed. 2631 1 - Allocate directly from the full usable range, 2632 forcing Dual Address Cycle for PCI cards supporting 2633 greater than 32-bit addressing. 2634 2635 iommu.strict= [ARM64,X86,S390,EARLY] Configure TLB invalidation behaviour 2636 Format: { "0" | "1" } 2637 0 - Lazy mode. 2638 Request that DMA unmap operations use deferred 2639 invalidation of hardware TLBs, for increased 2640 throughput at the cost of reduced device isolation. 2641 Will fall back to strict mode if not supported by 2642 the relevant IOMMU driver. 2643 1 - Strict mode. 2644 DMA unmap operations invalidate IOMMU hardware TLBs 2645 synchronously. 2646 unset - Use value of CONFIG_IOMMU_DEFAULT_DMA_{LAZY,STRICT}. 2647 Note: on x86, strict mode specified via one of the 2648 legacy driver-specific options takes precedence. 2649 2650 iommu.passthrough= 2651 [ARM64,X86,EARLY] Configure DMA to bypass the IOMMU by default. 2652 Format: { "0" | "1" } 2653 0 - Use IOMMU translation for DMA. 2654 1 - Bypass the IOMMU for DMA. 2655 unset - Use value of CONFIG_IOMMU_DEFAULT_PASSTHROUGH. 2656 2657 iommu.debug_pagealloc= 2658 [KNL,EARLY] When CONFIG_IOMMU_DEBUG_PAGEALLOC is set, this 2659 parameter enables the feature at boot time. By default, it 2660 is disabled and the system behaves the same way as a kernel 2661 built without CONFIG_IOMMU_DEBUG_PAGEALLOC. 2662 Format: { "0" | "1" } 2663 0 - Sanitizer disabled. 2664 1 - Sanitizer enabled, expect runtime overhead. 2665 2666 io7= [HW] IO7 for Marvel-based Alpha systems 2667 See comment before marvel_specify_io7 in 2668 arch/alpha/kernel/core_marvel.c. 2669 2670 io_delay= [X86,EARLY] I/O delay method 2671 0x80 2672 Standard port 0x80 based delay 2673 0xed 2674 Alternate port 0xed based delay (needed on some systems) 2675 udelay 2676 Simple two microseconds delay 2677 none 2678 No delay 2679 2680 ip= [IP_PNP] 2681 See Documentation/admin-guide/nfs/nfsroot.rst. 2682 2683 ipcmni_extend [KNL,EARLY] Extend the maximum number of unique System V 2684 IPC identifiers from 32,768 to 16,777,216. 2685 2686 ipe.enforce= [IPE] 2687 Format: <bool> 2688 Determine whether IPE starts in permissive (0) or 2689 enforce (1) mode. The default is enforce. 2690 2691 ipe.success_audit= 2692 [IPE] 2693 Format: <bool> 2694 Start IPE with success auditing enabled, emitting 2695 an audit event when a binary is allowed. The default 2696 is 0. 2697 2698 irqaffinity= [SMP] Set the default irq affinity mask 2699 The argument is a cpu list, as described above. 2700 2701 irqchip.gicv2_force_probe= 2702 [ARM,ARM64,EARLY] 2703 Format: <bool> 2704 Force the kernel to look for the second 4kB page 2705 of a GICv2 controller even if the memory range 2706 exposed by the device tree is too small. 2707 2708 irqchip.gicv3_nolpi= 2709 [ARM,ARM64,EARLY] 2710 Force the kernel to ignore the availability of 2711 LPIs (and by consequence ITSs). Intended for system 2712 that use the kernel as a bootloader, and thus want 2713 to let secondary kernels in charge of setting up 2714 LPIs. 2715 2716 irqchip.gicv3_pseudo_nmi= [ARM64,EARLY] 2717 Enables support for pseudo-NMIs in the kernel. This 2718 requires the kernel to be built with 2719 CONFIG_ARM64_PSEUDO_NMI. 2720 2721 irqchip.riscv_imsic_noipi 2722 [RISC-V,EARLY] 2723 Force the kernel to not use IMSIC software injected MSIs 2724 as IPIs. Intended for system where IMSIC is trap-n-emulated, 2725 and thus want to reduce MMIO traps when triggering IPIs 2726 to multiple harts. 2727 2728 irqfixup [HW] 2729 When an interrupt is not handled search all handlers 2730 for it. Intended to get systems with badly broken 2731 firmware running. 2732 2733 irqhandler.duration_warn_us= [KNL] 2734 Warn if an IRQ handler exceeds the specified duration 2735 threshold in microseconds. Useful for identifying 2736 long-running IRQs in the system. 2737 2738 irqpoll [HW] 2739 When an interrupt is not handled search all handlers 2740 for it. Also check all handlers each timer 2741 interrupt. Intended to get systems with badly broken 2742 firmware running. 2743 2744 isapnp= [ISAPNP] 2745 Format: <RDP>,<reset>,<pci_scan>,<verbosity> 2746 2747 isolcpus= [KNL,SMP,ISOL] Isolate a given set of CPUs from disturbance. 2748 Format: [flag-list,]<cpu-list> 2749 2750 Specify one or more CPUs to isolate from disturbances 2751 specified in the flag list (default: domain): 2752 2753 nohz 2754 Disable the tick when a single task runs as well as 2755 disabling other kernel noises like having RCU callbacks 2756 offloaded. This is equivalent to the nohz_full parameter. 2757 2758 A residual 1Hz tick is offloaded to workqueues, which you 2759 need to affine to housekeeping through the global 2760 workqueue's affinity configured via the 2761 /sys/devices/virtual/workqueue/cpumask sysfs file, or 2762 by using the 'domain' flag described below. 2763 2764 NOTE: by default the global workqueue runs on all CPUs, 2765 so to protect individual CPUs the 'cpumask' file has to 2766 be configured manually after bootup. 2767 2768 domain 2769 Isolate from the general SMP balancing and scheduling 2770 algorithms. Note that performing domain isolation this way 2771 is irreversible: it's not possible to bring back a CPU to 2772 the domains once isolated through this boot time 2773 configuration. Use cpusets for a dynamic configuration 2774 which can be altered at runtime. For details see 2775 Documentation/admin-guide/cpu-isolation.rst. 2776 2777 You can move a process onto or off an "isolated" CPU via 2778 the CPU affinity syscalls or cpuset. 2779 <cpu number> begins at 0 and the maximum value is 2780 "number of CPUs in system - 1". 2781 2782 managed_irq 2783 2784 Isolate from being targeted by managed interrupts 2785 which have an interrupt mask containing isolated 2786 CPUs. The affinity of managed interrupts is 2787 handled by the kernel and cannot be changed via 2788 the /proc/irq/* interfaces. 2789 2790 This isolation is best effort and only effective 2791 if the automatically assigned interrupt mask of a 2792 device queue contains isolated and housekeeping 2793 CPUs. If housekeeping CPUs are online then such 2794 interrupts are directed to the housekeeping CPU 2795 so that IO submitted on the housekeeping CPU 2796 cannot disturb the isolated CPU. 2797 2798 If a queue's affinity mask contains only isolated 2799 CPUs then this parameter has no effect on the 2800 interrupt routing decision, though interrupts are 2801 only delivered when tasks running on those 2802 isolated CPUs submit IO. IO submitted on 2803 housekeeping CPUs has no influence on those 2804 queues. 2805 2806 The format of <cpu-list> is described above. 2807 2808 iucv= [HW,NET] 2809 2810 ivrs_ioapic [HW,X86-64] 2811 Provide an override to the IOAPIC-ID<->DEVICE-ID 2812 mapping provided in the IVRS ACPI table. 2813 By default, PCI segment is 0, and can be omitted. 2814 2815 For example, to map IOAPIC-ID decimal 10 to 2816 PCI segment 0x1 and PCI device 00:14.0, 2817 write the parameter as: 2818 ivrs_ioapic=10@0001:00:14.0 2819 2820 Deprecated formats: 2821 * To map IOAPIC-ID decimal 10 to PCI device 00:14.0 2822 write the parameter as: 2823 ivrs_ioapic[10]=00:14.0 2824 * To map IOAPIC-ID decimal 10 to PCI segment 0x1 and 2825 PCI device 00:14.0 write the parameter as: 2826 ivrs_ioapic[10]=0001:00:14.0 2827 2828 ivrs_hpet [HW,X86-64] 2829 Provide an override to the HPET-ID<->DEVICE-ID 2830 mapping provided in the IVRS ACPI table. 2831 By default, PCI segment is 0, and can be omitted. 2832 2833 For example, to map HPET-ID decimal 10 to 2834 PCI segment 0x1 and PCI device 00:14.0, 2835 write the parameter as: 2836 ivrs_hpet=10@0001:00:14.0 2837 2838 Deprecated formats: 2839 * To map HPET-ID decimal 0 to PCI device 00:14.0 2840 write the parameter as: 2841 ivrs_hpet[0]=00:14.0 2842 * To map HPET-ID decimal 10 to PCI segment 0x1 and 2843 PCI device 00:14.0 write the parameter as: 2844 ivrs_ioapic[10]=0001:00:14.0 2845 2846 ivrs_acpihid [HW,X86-64] 2847 Provide an override to the ACPI-HID:UID<->DEVICE-ID 2848 mapping provided in the IVRS ACPI table. 2849 By default, PCI segment is 0, and can be omitted. 2850 2851 For example, to map UART-HID:UID AMD0020:0 to 2852 PCI segment 0x1 and PCI device ID 00:14.5, 2853 write the parameter as: 2854 ivrs_acpihid=AMD0020:0@0001:00:14.5 2855 2856 Deprecated formats: 2857 * To map UART-HID:UID AMD0020:0 to PCI segment is 0, 2858 PCI device ID 00:14.5, write the parameter as: 2859 ivrs_acpihid[00:14.5]=AMD0020:0 2860 * To map UART-HID:UID AMD0020:0 to PCI segment 0x1 and 2861 PCI device ID 00:14.5, write the parameter as: 2862 ivrs_acpihid[0001:00:14.5]=AMD0020:0 2863 2864 js= [HW,JOY] Analog joystick 2865 See Documentation/input/joydev/joystick.rst. 2866 2867 kasan_multi_shot 2868 [KNL] Enforce KASAN (Kernel Address Sanitizer) to print 2869 report on every invalid memory access. Without this 2870 parameter KASAN will print report only for the first 2871 invalid access. 2872 2873 keep_bootcon [KNL,EARLY] 2874 Do not unregister boot console at start. This is only 2875 useful for debugging when something happens in the window 2876 between unregistering the boot console and initializing 2877 the real console. 2878 2879 keepinitrd [HW,ARM] See retain_initrd. 2880 2881 kernelcore= [KNL,X86,PPC,EARLY] 2882 Format: nn[KMGTPE] | nn% | "mirror" 2883 This parameter specifies the amount of memory usable by 2884 the kernel for non-movable allocations. The requested 2885 amount is spread evenly throughout all nodes in the 2886 system as ZONE_NORMAL. The remaining memory is used for 2887 movable memory in its own zone, ZONE_MOVABLE. In the 2888 event, a node is too small to have both ZONE_NORMAL and 2889 ZONE_MOVABLE, kernelcore memory will take priority and 2890 other nodes will have a larger ZONE_MOVABLE. 2891 2892 ZONE_MOVABLE is used for the allocation of pages that 2893 may be reclaimed or moved by the page migration 2894 subsystem. Note that allocations like PTEs-from-HighMem 2895 still use the HighMem zone if it exists, and the Normal 2896 zone if it does not. 2897 2898 It is possible to specify the exact amount of memory in 2899 the form of "nn[KMGTPE]", a percentage of total system 2900 memory in the form of "nn%", or "mirror". If "mirror" 2901 option is specified, mirrored (reliable) memory is used 2902 for non-movable allocations and remaining memory is used 2903 for Movable pages. "nn[KMGTPE]", "nn%", and "mirror" 2904 are exclusive, so you cannot specify multiple forms. 2905 2906 kfence.burst= [MM,KFENCE] The number of additional successive 2907 allocations to be attempted through KFENCE for each 2908 sample interval. 2909 Format: <unsigned integer> 2910 Default: 0 2911 2912 kfence.check_on_panic= 2913 [MM,KFENCE] Whether to check all KFENCE-managed objects' 2914 canaries on panic. 2915 Format: <bool> 2916 Default: false 2917 2918 kfence.deferrable= 2919 [MM,KFENCE] Whether to use a deferrable timer to trigger 2920 allocations. This avoids forcing CPU wake-ups if the 2921 system is idle, at the risk of a less predictable 2922 sample interval. 2923 Format: <bool> 2924 Default: CONFIG_KFENCE_DEFERRABLE 2925 2926 kfence.fault= [MM,KFENCE] Controls the behavior when a KFENCE 2927 error is detected. 2928 report - print the error report and continue (default). 2929 oops - print the error report and oops. 2930 panic - print the error report and panic. 2931 2932 kfence.sample_interval= 2933 [MM,KFENCE] KFENCE's sample interval in milliseconds. 2934 Format: <unsigned integer> 2935 0 - Disable KFENCE. 2936 >0 - Enabled KFENCE with given sample interval. 2937 Default: CONFIG_KFENCE_SAMPLE_INTERVAL 2938 2939 kfence.skip_covered_thresh= 2940 [MM,KFENCE] If pool utilization reaches this threshold 2941 (pool usage%), KFENCE limits currently covered 2942 allocations of the same source from further filling 2943 up the pool. 2944 Format: <unsigned integer> 2945 Default: 75 2946 2947 kgdbdbgp= [KGDB,HW,EARLY] kgdb over EHCI usb debug port. 2948 Format: <Controller#>[,poll interval] 2949 The controller # is the number of the ehci usb debug 2950 port as it is probed via PCI. The poll interval is 2951 optional and is the number seconds in between 2952 each poll cycle to the debug port in case you need 2953 the functionality for interrupting the kernel with 2954 gdb or control-c on the dbgp connection. When 2955 not using this parameter you use sysrq-g to break into 2956 the kernel debugger. 2957 2958 kgdboc= [KGDB,HW] kgdb over consoles. 2959 Requires a tty driver that supports console polling, 2960 or a supported polling keyboard driver (non-usb). 2961 Serial only format: <serial_device>[,baud] 2962 keyboard only format: kbd 2963 keyboard and serial format: kbd,<serial_device>[,baud] 2964 Optional Kernel mode setting: 2965 kms, kbd format: kms,kbd 2966 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud] 2967 2968 kgdboc_earlycon= [KGDB,HW,EARLY] 2969 If the boot console provides the ability to read 2970 characters and can work in polling mode, you can use 2971 this parameter to tell kgdb to use it as a backend 2972 until the normal console is registered. Intended to 2973 be used together with the kgdboc parameter which 2974 specifies the normal console to transition to. 2975 2976 The name of the early console should be specified 2977 as the value of this parameter. Note that the name of 2978 the early console might be different than the tty 2979 name passed to kgdboc. It's OK to leave the value 2980 blank and the first boot console that implements 2981 read() will be picked. 2982 2983 kgdbwait [KGDB,EARLY] Stop kernel execution and enter the 2984 kernel debugger at the earliest opportunity. 2985 2986 kho= [KEXEC,EARLY] 2987 Format: { "0" | "1" | "off" | "on" | "y" | "n" } 2988 Enables or disables Kexec HandOver. 2989 "0" | "off" | "n" - kexec handover is disabled 2990 "1" | "on" | "y" - kexec handover is enabled 2991 2992 kho_scratch= [KEXEC,EARLY] 2993 Format: ll[KMG],mm[KMG],nn[KMG] | nn% 2994 Defines the size of the KHO scratch region. The KHO 2995 scratch regions are physically contiguous memory 2996 ranges that can only be used for non-kernel 2997 allocations. That way, even when memory is heavily 2998 fragmented with handed over memory, the kexeced 2999 kernel will always have enough contiguous ranges to 3000 bootstrap itself. 3001 3002 It is possible to specify the exact amount of 3003 memory in the form of "ll[KMG],mm[KMG],nn[KMG]" 3004 where the first parameter defines the size of a low 3005 memory scratch area, the second parameter defines 3006 the size of a global scratch area and the third 3007 parameter defines the size of additional per-node 3008 scratch areas. The form "nn%" defines scale factor 3009 (in percents) of memory that was used during boot. 3010 3011 kmac= [MIPS] Korina ethernet MAC address. 3012 Configure the RouterBoard 532 series on-chip 3013 Ethernet adapter MAC address. 3014 3015 kmemleak= [KNL,EARLY] Boot-time kmemleak enable/disable 3016 Valid arguments: on, off 3017 Default: on 3018 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y, 3019 the default is off. 3020 3021 kprobe_event=[probe-list] 3022 [FTRACE] Add kprobe events and enable at boot time. 3023 The probe-list is a semicolon delimited list of probe 3024 definitions. Each definition is same as kprobe_events 3025 interface, but the parameters are comma delimited. 3026 For example, to add a kprobe event on vfs_read with 3027 arg1 and arg2, add to the command line; 3028 3029 kprobe_event=p,vfs_read,$arg1,$arg2 3030 3031 See also Documentation/trace/kprobetrace.rst "Kernel 3032 Boot Parameter" section. 3033 3034 kpti= [ARM64,EARLY] Control page table isolation of 3035 user and kernel address spaces. 3036 Default: enabled on cores which need mitigation. 3037 0: force disabled 3038 1: force enabled 3039 3040 kunit.enable= [KUNIT] Enable executing KUnit tests. Requires 3041 CONFIG_KUNIT to be set to be fully enabled. The 3042 default value can be overridden via 3043 KUNIT_DEFAULT_ENABLED. 3044 Default is 1 (enabled) 3045 3046 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs. 3047 Default is 0 (don't ignore, but inject #GP) 3048 3049 kvm.eager_page_split= 3050 [KVM,X86,RISCV] Controls whether or not KVM will try to 3051 proactively split all huge pages during dirty logging. 3052 Eager page splitting reduces interruptions to vCPU 3053 execution by eliminating the write-protection faults 3054 and MMU lock contention that would otherwise be 3055 required to split huge pages lazily. 3056 3057 VM workloads that rarely perform writes or that write 3058 only to a small region of VM memory may benefit from 3059 disabling eager page splitting to allow huge pages to 3060 still be used for reads. 3061 3062 The behavior of eager page splitting depends on whether 3063 KVM_DIRTY_LOG_INITIALLY_SET is enabled or disabled. If 3064 disabled, all huge pages in a memslot will be eagerly 3065 split when dirty logging is enabled on that memslot. If 3066 enabled, eager page splitting will be performed during 3067 the KVM_CLEAR_DIRTY ioctl, and only for the pages being 3068 cleared. 3069 3070 On x86, eager page splitting is only supported when 3071 kvm.tdp_mmu=Y. 3072 3073 On RISCV, eager page splitting is supported by default. 3074 3075 Default is Y (on). 3076 3077 kvm.enable_pmu=[KVM,X86] 3078 If enabled, KVM will virtualize PMU functionality based 3079 on the virtual CPU model defined by userspace. This 3080 can be overridden on a per-VM basis via 3081 KVM_CAP_PMU_CAPABILITY. 3082 3083 If disabled, KVM will not virtualize PMU functionality, 3084 e.g. MSRs, PMCs, PMIs, etc., even if userspace defines 3085 a virtual CPU model that contains PMU assets. 3086 3087 Note, KVM's vPMU support implicitly requires running 3088 with an in-kernel local APIC, e.g. to deliver PMIs to 3089 the guest. Running without an in-kernel local APIC is 3090 not supported, though KVM will allow such a combination 3091 (with severely degraded functionality). 3092 3093 See also enable_mediated_pmu. 3094 3095 Default is Y (on). 3096 3097 kvm.enable_virt_at_load=[KVM,ARM64,LOONGARCH,MIPS,RISCV,X86] 3098 If enabled, KVM will enable virtualization in hardware 3099 when KVM is loaded, and disable virtualization when KVM 3100 is unloaded (if KVM is built as a module). 3101 3102 If disabled, KVM will dynamically enable and disable 3103 virtualization on-demand when creating and destroying 3104 VMs, i.e. on the 0=>1 and 1=>0 transitions of the 3105 number of VMs. 3106 3107 Enabling virtualization at module load avoids potential 3108 latency for creation of the 0=>1 VM, as KVM serializes 3109 virtualization enabling across all online CPUs. The 3110 "cost" of enabling virtualization when KVM is loaded, 3111 is that doing so may interfere with using out-of-tree 3112 hypervisors that want to "own" virtualization hardware. 3113 3114 kvm.enable_vmware_backdoor=[KVM] Support VMware backdoor PV interface. 3115 Default is false (don't support). 3116 3117 kvm.nx_huge_pages= 3118 [KVM] Controls the software workaround for the 3119 X86_BUG_ITLB_MULTIHIT bug. 3120 force : Always deploy workaround. 3121 off : Never deploy workaround. 3122 auto : Deploy workaround based on the presence of 3123 X86_BUG_ITLB_MULTIHIT. 3124 3125 Default is 'auto'. 3126 3127 If the software workaround is enabled for the host, 3128 guests do need not to enable it for nested guests. 3129 3130 kvm.nx_huge_pages_recovery_ratio= 3131 [KVM] Controls how many 4KiB pages are periodically zapped 3132 back to huge pages. 0 disables the recovery, otherwise if 3133 the value is N KVM will zap 1/Nth of the 4KiB pages every 3134 period (see below). The default is 60. 3135 3136 kvm.nx_huge_pages_recovery_period_ms= 3137 [KVM] Controls the time period at which KVM zaps 4KiB pages 3138 back to huge pages. If the value is a non-zero N, KVM will 3139 zap a portion (see ratio above) of the pages every N msecs. 3140 If the value is 0 (the default), KVM will pick a period based 3141 on the ratio, such that a page is zapped after 1 hour on average. 3142 3143 kvm-{amd,intel}.enable_mediated_pmu=[KVM,AMD,INTEL] 3144 If enabled, KVM will provide a mediated virtual PMU, 3145 instead of the default perf-based virtual PMU (if 3146 kvm.enable_pmu is true and PMU is enumerated via the 3147 virtual CPU model). 3148 3149 With a perf-based vPMU, KVM operates as a user of perf, 3150 i.e. emulates guest PMU counters using perf events. 3151 KVM-created perf events are managed by perf as regular 3152 (guest-only) events, e.g. are scheduled in/out, contend 3153 for hardware resources, etc. Using a perf-based vPMU 3154 allows guest and host usage of the PMU to co-exist, but 3155 incurs non-trivial overhead and can result in silently 3156 dropped guest events (due to resource contention). 3157 3158 With a mediated vPMU, hardware PMU state is context 3159 switched around the world switch to/from the guest. 3160 KVM mediates which events the guest can utilize, but 3161 gives the guest direct access to all other PMU assets 3162 when possible (KVM may intercept some accesses if the 3163 virtual CPU model provides a subset of hardware PMU 3164 functionality). Using a mediated vPMU significantly 3165 reduces PMU virtualization overhead and eliminates lost 3166 guest events, but is mutually exclusive with using perf 3167 to profile KVM guests and adds latency to most VM-Exits 3168 (to context switch PMU state). 3169 3170 Default is N (off). 3171 3172 kvm-amd.nested= [KVM,AMD] Control nested virtualization feature in 3173 KVM/SVM. Default is 1 (enabled). 3174 3175 kvm-amd.npt= [KVM,AMD] Control KVM's use of Nested Page Tables, 3176 a.k.a. Two-Dimensional Page Tables. Default is 1 3177 (enabled). Disable by KVM if hardware lacks support 3178 for NPT. 3179 3180 kvm-amd.ciphertext_hiding_asids= 3181 [KVM,AMD] Ciphertext hiding prevents disallowed accesses 3182 to SNP private memory from reading ciphertext. Instead, 3183 reads will see constant default values (0xff). 3184 3185 If ciphertext hiding is enabled, the joint SEV-ES and 3186 SEV-SNP ASID space is partitioned into separate SEV-ES 3187 and SEV-SNP ASID ranges, with the SEV-SNP range being 3188 [1..max_snp_asid] and the SEV-ES range being 3189 (max_snp_asid..min_sev_asid), where min_sev_asid is 3190 enumerated by CPUID.0x.8000_001F[EDX]. 3191 3192 A non-zero value enables SEV-SNP ciphertext hiding and 3193 adjusts the ASID ranges for SEV-ES and SEV-SNP guests. 3194 KVM caps the number of SEV-SNP ASIDs at the maximum 3195 possible value, e.g. specifying -1u will assign all 3196 joint SEV-ES and SEV-SNP ASIDs to SEV-SNP. Note, 3197 assigning all joint ASIDs to SEV-SNP, i.e. configuring 3198 max_snp_asid == min_sev_asid-1, will effectively make 3199 SEV-ES unusable. 3200 3201 kvm-arm.mode= 3202 [KVM,ARM,EARLY] Select one of KVM/arm64's modes of 3203 operation. 3204 3205 none: Forcefully disable KVM. 3206 3207 nvhe: Standard nVHE-based mode, without support for 3208 protected guests. 3209 3210 protected: Mode with support for guests whose state is 3211 kept private from the host, using VHE or 3212 nVHE depending on HW support. 3213 3214 nested: VHE-based mode with support for nested 3215 virtualization. Requires at least ARMv8.4 3216 hardware (with FEAT_NV2). 3217 3218 Defaults to VHE/nVHE based on hardware support. Setting 3219 mode to "protected" will disable kexec and hibernation 3220 for the host. To force nVHE on VHE hardware, add 3221 "arm64_sw.hvhe=0 id_aa64mmfr1.vh=0" to the 3222 command-line. 3223 "nested" and "protected" are experimental and should be 3224 used with extreme caution. 3225 3226 kvm-arm.vgic_v3_group0_trap= 3227 [KVM,ARM,EARLY] Trap guest accesses to GICv3 group-0 3228 system registers 3229 3230 kvm-arm.vgic_v3_group1_trap= 3231 [KVM,ARM,EARLY] Trap guest accesses to GICv3 group-1 3232 system registers 3233 3234 kvm-arm.vgic_v3_common_trap= 3235 [KVM,ARM,EARLY] Trap guest accesses to GICv3 common 3236 system registers 3237 3238 kvm-arm.vgic_v4_enable= 3239 [KVM,ARM,EARLY] Allow use of GICv4 for direct 3240 injection of LPIs. 3241 3242 kvm-arm.wfe_trap_policy= 3243 [KVM,ARM] Control when to set WFE instruction trap for 3244 KVM VMs. Traps are allowed but not guaranteed by the 3245 CPU architecture. 3246 3247 trap: set WFE instruction trap 3248 3249 notrap: clear WFE instruction trap 3250 3251 kvm-arm.wfi_trap_policy= 3252 [KVM,ARM] Control when to set WFI instruction trap for 3253 KVM VMs. Traps are allowed but not guaranteed by the 3254 CPU architecture. 3255 3256 trap: set WFI instruction trap 3257 3258 notrap: clear WFI instruction trap 3259 3260 kvm_cma_resv_ratio=n [PPC,EARLY] 3261 Reserves given percentage from system memory area for 3262 contiguous memory allocation for KVM hash pagetable 3263 allocation. 3264 By default it reserves 5% of total system memory. 3265 Format: <integer> 3266 Default: 5 3267 3268 kvm-intel.ept= [KVM,Intel] Control KVM's use of Extended Page Tables, 3269 a.k.a. Two-Dimensional Page Tables. Default is 1 3270 (enabled). Disable by KVM if hardware lacks support 3271 for EPT. 3272 3273 kvm-intel.emulate_invalid_guest_state= 3274 [KVM,Intel] Control whether to emulate invalid guest 3275 state. Ignored if kvm-intel.enable_unrestricted_guest=1, 3276 as guest state is never invalid for unrestricted 3277 guests. This param doesn't apply to nested guests (L2), 3278 as KVM never emulates invalid L2 guest state. 3279 Default is 1 (enabled). 3280 3281 kvm-intel.flexpriority= 3282 [KVM,Intel] Control KVM's use of FlexPriority feature 3283 (TPR shadow). Default is 1 (enabled). Disable by KVM if 3284 hardware lacks support for it. 3285 3286 kvm-intel.nested= 3287 [KVM,Intel] Control nested virtualization feature in 3288 KVM/VMX. Default is 1 (enabled). 3289 3290 kvm-intel.unrestricted_guest= 3291 [KVM,Intel] Control KVM's use of unrestricted guest 3292 feature (virtualized real and unpaged mode). Default 3293 is 1 (enabled). Disable by KVM if EPT is disabled or 3294 hardware lacks support for it. 3295 3296 kvm-intel.vmentry_l1d_flush=[KVM,Intel] Mitigation for L1 Terminal Fault 3297 CVE-2018-3620. 3298 3299 Valid arguments: never, cond, always 3300 3301 always: L1D cache flush on every VMENTER. 3302 cond: Flush L1D on VMENTER only when the code between 3303 VMEXIT and VMENTER can leak host memory. 3304 never: Disables the mitigation 3305 3306 Default is cond (do L1 cache flush in specific instances) 3307 3308 kvm-intel.vpid= [KVM,Intel] Control KVM's use of Virtual Processor 3309 Identification feature (tagged TLBs). Default is 1 3310 (enabled). Disable by KVM if hardware lacks support 3311 for it. 3312 3313 l1d_flush= [X86,INTEL,EARLY] 3314 Control mitigation for L1D based snooping vulnerability. 3315 3316 Certain CPUs are vulnerable to an exploit against CPU 3317 internal buffers which can forward information to a 3318 disclosure gadget under certain conditions. 3319 3320 In vulnerable processors, the speculatively 3321 forwarded data can be used in a cache side channel 3322 attack, to access data to which the attacker does 3323 not have direct access. 3324 3325 This parameter controls the mitigation. The 3326 options are: 3327 3328 on - enable the interface for the mitigation 3329 3330 l1tf= [X86,EARLY] Control mitigation of the L1TF vulnerability on 3331 affected CPUs 3332 3333 The kernel PTE inversion protection is unconditionally 3334 enabled and cannot be disabled. 3335 3336 full 3337 Provides all available mitigations for the 3338 L1TF vulnerability. Disables SMT and 3339 enables all mitigations in the 3340 hypervisors, i.e. unconditional L1D flush. 3341 3342 SMT control and L1D flush control via the 3343 sysfs interface is still possible after 3344 boot. Hypervisors will issue a warning 3345 when the first VM is started in a 3346 potentially insecure configuration, 3347 i.e. SMT enabled or L1D flush disabled. 3348 3349 full,force 3350 Same as 'full', but disables SMT and L1D 3351 flush runtime control. Implies the 3352 'nosmt=force' command line option. 3353 (i.e. sysfs control of SMT is disabled.) 3354 3355 flush 3356 Leaves SMT enabled and enables the default 3357 hypervisor mitigation, i.e. conditional 3358 L1D flush. 3359 3360 SMT control and L1D flush control via the 3361 sysfs interface is still possible after 3362 boot. Hypervisors will issue a warning 3363 when the first VM is started in a 3364 potentially insecure configuration, 3365 i.e. SMT enabled or L1D flush disabled. 3366 3367 flush,nosmt 3368 3369 Disables SMT and enables the default 3370 hypervisor mitigation. 3371 3372 SMT control and L1D flush control via the 3373 sysfs interface is still possible after 3374 boot. Hypervisors will issue a warning 3375 when the first VM is started in a 3376 potentially insecure configuration, 3377 i.e. SMT enabled or L1D flush disabled. 3378 3379 flush,nowarn 3380 Same as 'flush', but hypervisors will not 3381 warn when a VM is started in a potentially 3382 insecure configuration. 3383 3384 off 3385 Disables hypervisor mitigations and doesn't 3386 emit any warnings. 3387 It also drops the swap size and available 3388 RAM limit restriction on both hypervisor and 3389 bare metal. 3390 3391 Default is 'flush'. 3392 3393 For details see: Documentation/admin-guide/hw-vuln/l1tf.rst 3394 3395 l2cr= [PPC] 3396 3397 l3cr= [PPC] 3398 3399 lapic [X86-32,APIC,EARLY] Enable the local APIC even if BIOS 3400 disabled it. 3401 3402 lapic= [X86,APIC] Do not use TSC deadline 3403 value for LAPIC timer one-shot implementation. Default 3404 back to the programmable timer unit in the LAPIC. 3405 Format: notscdeadline 3406 3407 lapic_timer_c2_ok [X86,APIC,EARLY] trust the local apic timer 3408 in C2 power state. 3409 3410 libata.dma= [LIBATA] DMA control 3411 libata.dma=0 Disable all PATA and SATA DMA 3412 libata.dma=1 PATA and SATA Disk DMA only 3413 libata.dma=2 ATAPI (CDROM) DMA only 3414 libata.dma=4 Compact Flash DMA only 3415 Combinations also work, so libata.dma=3 enables DMA 3416 for disks and CDROMs, but not CFs. 3417 3418 libata.ignore_hpa= [LIBATA] Ignore HPA limit 3419 libata.ignore_hpa=0 keep BIOS limits (default) 3420 libata.ignore_hpa=1 ignore limits, using full disk 3421 3422 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume 3423 when set. 3424 Format: <int> 3425 3426 libata.force= [LIBATA] Force configurations. The format is a comma- 3427 separated list of "[ID:]VAL" where ID is PORT[.DEVICE]. 3428 PORT and DEVICE are decimal numbers matching port, link 3429 or device. Basically, it matches the ATA ID string 3430 printed on console by libata. If the whole ID part is 3431 omitted, the last PORT and DEVICE values are used. If 3432 ID hasn't been specified yet, the configuration applies 3433 to all ports, links and devices. 3434 3435 If only DEVICE is omitted, the parameter applies to 3436 the port and all links and devices behind it. DEVICE 3437 number of 0 either selects the first device or the 3438 first fan-out link behind PMP device. It does not 3439 select the host link. DEVICE number of 15 selects the 3440 host link and device attached to it. 3441 3442 The VAL specifies the configuration to force. As long 3443 as there is no ambiguity, shortcut notation is allowed. 3444 For example, both 1.5 and 1.5G would work for 1.5Gbps. 3445 The following configurations can be forced. 3446 3447 * Cable type: 40c, 80c, short40c, unk, ign or sata. 3448 Any ID with matching PORT is used. 3449 3450 * SATA link speed limit: 1.5Gbps or 3.0Gbps. 3451 3452 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7]. 3453 udma[/][16,25,33,44,66,100,133] notation is also 3454 allowed. 3455 3456 * nohrst, nosrst, norst: suppress hard, soft and both 3457 resets. 3458 3459 * rstonce: only attempt one reset during hot-unplug 3460 link recovery. 3461 3462 * [no]dbdelay: Enable or disable the extra 200ms delay 3463 before debouncing a link PHY and device presence 3464 detection. 3465 3466 * [no]ncq: Turn on or off NCQ. 3467 3468 * [no]ncqtrim: Enable or disable queued DSM TRIM. 3469 3470 * [no]ncqati: Enable or disable NCQ trim on ATI chipset. 3471 3472 * [no]trim: Enable or disable (unqueued) TRIM. 3473 3474 * trim_zero: Indicate that TRIM command zeroes data. 3475 3476 * max_trim_128m: Set 128M maximum trim size limit. 3477 3478 * [no]dma: Turn on or off DMA transfers. 3479 3480 * atapi_dmadir: Enable ATAPI DMADIR bridge support. 3481 3482 * atapi_mod16_dma: Enable the use of ATAPI DMA for 3483 commands that are not a multiple of 16 bytes. 3484 3485 * [no]dmalog: Enable or disable the use of the 3486 READ LOG DMA EXT command to access logs. 3487 3488 * [no]iddevlog: Enable or disable access to the 3489 identify device data log. 3490 3491 * [no]logdir: Enable or disable access to the general 3492 purpose log directory. 3493 3494 * max_sec=<sectors>: Set the transfer size limit, in 3495 number of 512-byte sectors, to the value specified in 3496 <sectors>. The value specified in <sectors> has to be 3497 a non-zero positive integer. 3498 3499 * max_sec_128: Set transfer size limit to 128 sectors. 3500 3501 * max_sec_1024: Set or clear transfer size limit to 3502 1024 sectors. 3503 3504 * max_sec_lba48: Set or clear transfer size limit to 3505 65535 sectors. 3506 3507 * external: Mark port as external (hotplug-capable). 3508 3509 * [no]lpm: Enable or disable link power management. 3510 3511 * [no]setxfer: Indicate if transfer speed mode setting 3512 should be skipped. 3513 3514 * [no]fua: Disable or enable FUA (Force Unit Access) 3515 support for devices supporting this feature. 3516 3517 * dump_id: Dump IDENTIFY data. 3518 3519 * disable: Disable this device. 3520 3521 If there are multiple matching configurations changing 3522 the same attribute, the last one is used. 3523 3524 liveupdate= [KNL,EARLY] 3525 Format: <bool> 3526 Enable Live Update Orchestrator (LUO). 3527 Default: off. 3528 3529 lockd.nlm_grace_period=P [NFS] Assign grace period. 3530 Format: <integer> 3531 3532 lockd.nlm_tcpport=N [NFS] Assign TCP port. 3533 Format: <integer> 3534 3535 lockd.nlm_timeout=T [NFS] Assign timeout value. 3536 Format: <integer> 3537 3538 lockd.nlm_udpport=M [NFS] Assign UDP port. 3539 Format: <integer> 3540 3541 lockdown= [SECURITY,EARLY] 3542 { integrity | confidentiality } 3543 Enable the kernel lockdown feature. If set to 3544 integrity, kernel features that allow userland to 3545 modify the running kernel are disabled. If set to 3546 confidentiality, kernel features that allow userland 3547 to extract confidential information from the kernel 3548 are also disabled. 3549 3550 locktorture.acq_writer_lim= [KNL] 3551 Set the time limit in jiffies for a lock 3552 acquisition. Acquisitions exceeding this limit 3553 will result in a splat once they do complete. 3554 3555 locktorture.bind_readers= [KNL] 3556 Specify the list of CPUs to which the readers are 3557 to be bound. 3558 3559 locktorture.bind_writers= [KNL] 3560 Specify the list of CPUs to which the writers are 3561 to be bound. 3562 3563 locktorture.call_rcu_chains= [KNL] 3564 Specify the number of self-propagating call_rcu() 3565 chains to set up. These are used to ensure that 3566 there is a high probability of an RCU grace period 3567 in progress at any given time. Defaults to 0, 3568 which disables these call_rcu() chains. 3569 3570 locktorture.long_hold= [KNL] 3571 Specify the duration in milliseconds for the 3572 occasional long-duration lock hold time. Defaults 3573 to 100 milliseconds. Select 0 to disable. 3574 3575 locktorture.nested_locks= [KNL] 3576 Specify the maximum lock nesting depth that 3577 locktorture is to exercise, up to a limit of 8 3578 (MAX_NESTED_LOCKS). Specify zero to disable. 3579 Note that this parameter is ineffective on types 3580 of locks that do not support nested acquisition. 3581 3582 locktorture.nreaders_stress= [KNL] 3583 Set the number of locking read-acquisition kthreads. 3584 Defaults to being automatically set based on the 3585 number of online CPUs. 3586 3587 locktorture.nwriters_stress= [KNL] 3588 Set the number of locking write-acquisition kthreads. 3589 3590 locktorture.onoff_holdoff= [KNL] 3591 Set time (s) after boot for CPU-hotplug testing. 3592 3593 locktorture.onoff_interval= [KNL] 3594 Set time (s) between CPU-hotplug operations, or 3595 zero to disable CPU-hotplug testing. 3596 3597 locktorture.rt_boost= [KNL] 3598 Do periodic testing of real-time lock priority 3599 boosting. Select 0 to disable, 1 to boost 3600 only rt_mutex, and 2 to boost unconditionally. 3601 Defaults to 2, which might seem to be an 3602 odd choice, but which should be harmless for 3603 non-real-time spinlocks, due to their disabling 3604 of preemption. Note that non-realtime mutexes 3605 disable boosting. 3606 3607 locktorture.rt_boost_factor= [KNL] 3608 Number that determines how often and for how 3609 long priority boosting is exercised. This is 3610 scaled down by the number of writers, so that the 3611 number of boosts per unit time remains roughly 3612 constant as the number of writers increases. 3613 On the other hand, the duration of each boost 3614 increases with the number of writers. 3615 3616 locktorture.shuffle_interval= [KNL] 3617 Set task-shuffle interval (jiffies). Shuffling 3618 tasks allows some CPUs to go into dyntick-idle 3619 mode during the locktorture test. 3620 3621 locktorture.shutdown_secs= [KNL] 3622 Set time (s) after boot system shutdown. This 3623 is useful for hands-off automated testing. 3624 3625 locktorture.stat_interval= [KNL] 3626 Time (s) between statistics printk()s. 3627 3628 locktorture.stutter= [KNL] 3629 Time (s) to stutter testing, for example, 3630 specifying five seconds causes the test to run for 3631 five seconds, wait for five seconds, and so on. 3632 This tests the locking primitive's ability to 3633 transition abruptly to and from idle. 3634 3635 locktorture.torture_type= [KNL] 3636 Specify the locking implementation to test. 3637 3638 locktorture.verbose= [KNL] 3639 Enable additional printk() statements. 3640 3641 locktorture.writer_fifo= [KNL] 3642 Run the write-side locktorture kthreads at 3643 sched_set_fifo() real-time priority. 3644 3645 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver 3646 Format: <irq> 3647 3648 loglevel= [KNL,EARLY] 3649 All Kernel Messages with a loglevel smaller than the 3650 console loglevel will be printed to the console. It can 3651 also be changed with klogd or other programs. The 3652 loglevels are defined as follows: 3653 3654 0 (KERN_EMERG) system is unusable 3655 1 (KERN_ALERT) action must be taken immediately 3656 2 (KERN_CRIT) critical conditions 3657 3 (KERN_ERR) error conditions 3658 4 (KERN_WARNING) warning conditions 3659 5 (KERN_NOTICE) normal but significant condition 3660 6 (KERN_INFO) informational 3661 7 (KERN_DEBUG) debug-level messages 3662 3663 log_buf_len=n[KMG] [KNL,EARLY] 3664 Sets the size of the printk ring buffer, in bytes. 3665 n must be a power of two and greater than the 3666 minimal size. The minimal size is defined by 3667 LOG_BUF_SHIFT kernel config parameter. There 3668 is also CONFIG_LOG_CPU_MAX_BUF_SHIFT config 3669 parameter that allows to increase the default size 3670 depending on the number of CPUs. See init/Kconfig 3671 for more details. 3672 3673 logo.nologo [FB] Disables display of the built-in Linux logo. 3674 This may be used to provide more screen space for 3675 kernel log messages and is useful when debugging 3676 kernel boot problems. 3677 3678 lp=0 [LP] Specify parallel ports to use, e.g, 3679 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses 3680 lp=reset first parallel port). 'lp=0' disables the 3681 lp=auto printer driver. 'lp=reset' (which can be 3682 specified in addition to the ports) causes 3683 attached printers to be reset. Using 3684 lp=port1,port2,... specifies the parallel ports 3685 to associate lp devices with, starting with 3686 lp0. A port specification may be 'none' to skip 3687 that lp device, or a parport name such as 3688 'parport0'. Specifying 'lp=auto' instead of a 3689 port specification list means that device IDs 3690 from each port should be examined, to see if 3691 an IEEE 1284-compliant printer is attached; if 3692 so, the driver will manage that printer. 3693 See also header of drivers/char/lp.c. 3694 3695 lpj=n [KNL] 3696 Sets loops_per_jiffy to given constant, thus avoiding 3697 time-consuming boot-time autodetection (up to 250 ms per 3698 CPU). 0 enables autodetection (default). To determine 3699 the correct value for your kernel, boot with normal 3700 autodetection and see what value is printed. Note that 3701 on SMP systems the preset will be applied to all CPUs, 3702 which is likely to cause problems if your CPUs need 3703 significantly divergent settings. An incorrect value 3704 will cause delays in the kernel to be wrong, leading to 3705 unpredictable I/O errors and other breakage. Although 3706 unlikely, in the extreme case this might damage your 3707 hardware. 3708 3709 lsm.debug [SECURITY] Enable LSM initialization debugging output. 3710 3711 lsm=lsm1,...,lsmN 3712 [SECURITY] Choose order of LSM initialization. This 3713 overrides CONFIG_LSM, and the "security=" parameter. 3714 3715 machtype= [Loongson] Share the same kernel image file between 3716 different yeeloong laptops. 3717 Example: machtype=lemote-yeeloong-2f-7inch 3718 3719 maxcpus= [SMP,EARLY] Maximum number of processors that an SMP kernel 3720 will bring up during bootup. maxcpus=n : n >= 0 limits 3721 the kernel to bring up 'n' processors. Surely after 3722 bootup you can bring up the other plugged cpu by executing 3723 "echo 1 > /sys/devices/system/cpu/cpuX/online". So maxcpus 3724 only takes effect during system bootup. 3725 While n=0 is a special case, it is equivalent to "nosmp", 3726 which also disables the IO APIC. 3727 3728 max_loop= [LOOP] The number of loop block devices that get 3729 (loop.max_loop) unconditionally pre-created at init time. The default 3730 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead 3731 of statically allocating a predefined number, loop 3732 devices can be requested on-demand with the 3733 /dev/loop-control interface. 3734 3735 mce= [X86-{32,64}] 3736 3737 Please see Documentation/arch/x86/x86_64/machinecheck.rst for sysfs runtime tunables. 3738 3739 off 3740 disable machine check 3741 3742 no_cmci 3743 disable CMCI(Corrected Machine Check Interrupt) that 3744 Intel processor supports. Usually this disablement is 3745 not recommended, but it might be handy if your 3746 hardware is misbehaving. 3747 3748 Note that you'll get more problems without CMCI than 3749 with due to the shared banks, i.e. you might get 3750 duplicated error logs. 3751 3752 dont_log_ce 3753 don't make logs for corrected errors. All events 3754 reported as corrected are silently cleared by OS. This 3755 option will be useful if you have no interest in any 3756 of corrected errors. 3757 3758 ignore_ce 3759 disable features for corrected errors, e.g. 3760 polling timer and CMCI. All events reported as 3761 corrected are not cleared by OS and remained in its 3762 error banks. 3763 3764 Usually this disablement is not recommended, however 3765 if there is an agent checking/clearing corrected 3766 errors (e.g. BIOS or hardware monitoring 3767 applications), conflicting with OS's error handling, 3768 and you cannot deactivate the agent, then this option 3769 will be a help. 3770 3771 no_lmce 3772 do not opt-in to Local MCE delivery. Use legacy method 3773 to broadcast MCEs. 3774 3775 bootlog 3776 enable logging of machine checks left over from 3777 booting. Disabled by default on AMD Fam10h and older 3778 because some BIOS leave bogus ones. 3779 3780 If your BIOS doesn't do that it's a good idea to 3781 enable though to make sure you log even machine check 3782 events that result in a reboot. On Intel systems it is 3783 enabled by default. 3784 3785 nobootlog 3786 disable boot machine check logging. 3787 3788 monarchtimeout (number) 3789 sets the time in us to wait for other CPUs on machine 3790 checks. 0 to disable. 3791 3792 bios_cmci_threshold 3793 don't overwrite the bios-set CMCI threshold. This boot 3794 option prevents Linux from overwriting the CMCI 3795 threshold set by the bios. Without this option, Linux 3796 always sets the CMCI threshold to 1. Enabling this may 3797 make memory predictive failure analysis less effective 3798 if the bios sets thresholds for memory errors since we 3799 will not see details for all errors. 3800 3801 recovery 3802 force-enable recoverable machine check code paths 3803 3804 Everything else is in sysfs now. 3805 3806 3807 md= [HW] RAID subsystems devices and level 3808 See Documentation/admin-guide/md.rst. 3809 3810 mds= [X86,INTEL,EARLY] 3811 Control mitigation for the Micro-architectural Data 3812 Sampling (MDS) vulnerability. 3813 3814 Certain CPUs are vulnerable to an exploit against CPU 3815 internal buffers which can forward information to a 3816 disclosure gadget under certain conditions. 3817 3818 In vulnerable processors, the speculatively 3819 forwarded data can be used in a cache side channel 3820 attack, to access data to which the attacker does 3821 not have direct access. 3822 3823 This parameter controls the MDS mitigation. The 3824 options are: 3825 3826 full - Enable MDS mitigation on vulnerable CPUs 3827 full,nosmt - Enable MDS mitigation and disable 3828 SMT on vulnerable CPUs 3829 off - Unconditionally disable MDS mitigation 3830 3831 On TAA-affected machines, mds=off can be prevented by 3832 an active TAA mitigation as both vulnerabilities are 3833 mitigated with the same mechanism so in order to disable 3834 this mitigation, you need to specify tsx_async_abort=off 3835 too. 3836 3837 Not specifying this option is equivalent to 3838 mds=full. 3839 3840 For details see: Documentation/admin-guide/hw-vuln/mds.rst 3841 3842 mem=nn[KMG] [HEXAGON,EARLY] Set the memory size. 3843 Must be specified, otherwise memory size will be 0. 3844 3845 mem=nn[KMG] [KNL,BOOT,EARLY] Force usage of a specific amount 3846 of memory Amount of memory to be used in cases 3847 as follows: 3848 3849 1 for test; 3850 2 when the kernel is not able to see the whole system memory; 3851 3 memory that lies after 'mem=' boundary is excluded from 3852 the hypervisor, then assigned to KVM guests. 3853 4 to limit the memory available for kdump kernel. 3854 3855 [ARC,MICROBLAZE] - the limit applies only to low memory, 3856 high memory is not affected. 3857 3858 [ARM64] - only limits memory covered by the linear 3859 mapping. The NOMAP regions are not affected. 3860 3861 [X86] Work as limiting max address. Use together 3862 with memmap= to avoid physical address space collisions. 3863 Without memmap= PCI devices could be placed at addresses 3864 belonging to unused RAM. 3865 3866 Note that this only takes effects during boot time since 3867 in above case 3, memory may need be hot added after boot 3868 if system memory of hypervisor is not sufficient. 3869 3870 mem=nn[KMG]@ss[KMG] 3871 [ARM,MIPS,EARLY] - override the memory layout 3872 reported by firmware. 3873 Define a memory region of size nn[KMG] starting at 3874 ss[KMG]. 3875 Multiple different regions can be specified with 3876 multiple mem= parameters on the command line. 3877 3878 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel 3879 memory. 3880 3881 memblock=debug [KNL,EARLY] Enable memblock debug messages. 3882 3883 memchunk=nn[KMG] 3884 [KNL,SH] Allow user to override the default size for 3885 per-device physically contiguous DMA buffers. 3886 3887 memhp_default_state=online/offline/online_kernel/online_movable 3888 [KNL] Set the initial state for the memory hotplug 3889 onlining policy. If not specified, the default value is 3890 set according to the 3891 CONFIG_MHP_DEFAULT_ONLINE_TYPE kernel config 3892 options. 3893 See Documentation/admin-guide/mm/memory-hotplug.rst. 3894 3895 memmap=exactmap [KNL,X86,EARLY] Enable setting of an exact 3896 E820 memory map, as specified by the user. 3897 Such memmap=exactmap lines can be constructed based on 3898 BIOS output or other requirements. See the memmap=nn@ss 3899 option description. 3900 3901 memmap=nn[KMG]@ss[KMG] 3902 [KNL, X86,MIPS,XTENSA,EARLY] Force usage of a specific region of memory. 3903 Region of memory to be used is from ss to ss+nn. 3904 If @ss[KMG] is omitted, it is equivalent to mem=nn[KMG], 3905 which limits max address to nn[KMG]. 3906 Multiple different regions can be specified, 3907 comma delimited. 3908 Example: 3909 memmap=100M@2G,100M#3G,1G!1024G 3910 3911 memmap=nn[KMG]#ss[KMG] 3912 [KNL,ACPI,EARLY] Mark specific memory as ACPI data. 3913 Region of memory to be marked is from ss to ss+nn. 3914 3915 memmap=nn[KMG]$ss[KMG] 3916 [KNL,ACPI,EARLY] Mark specific memory as reserved. 3917 Region of memory to be reserved is from ss to ss+nn. 3918 Example: Exclude memory from 0x18690000-0x1869ffff 3919 memmap=64K$0x18690000 3920 or 3921 memmap=0x10000$0x18690000 3922 Some bootloaders may need an escape character before '$', 3923 like Grub2, otherwise '$' and the following number 3924 will be eaten. 3925 3926 memmap=nn[KMG]!ss[KMG,EARLY] 3927 [KNL,X86] Mark specific memory as protected. 3928 Region of memory to be used, from ss to ss+nn. 3929 The memory region may be marked as e820 type 12 (0xc) 3930 and is NVDIMM or ADR memory. 3931 3932 memmap=<size>%<offset>-<oldtype>+<newtype> 3933 [KNL,ACPI,EARLY] Convert memory within the specified region 3934 from <oldtype> to <newtype>. If "-<oldtype>" is left 3935 out, the whole region will be marked as <newtype>, 3936 even if previously unavailable. If "+<newtype>" is left 3937 out, matching memory will be removed. Types are 3938 specified as e820 types, e.g., 1 = RAM, 2 = reserved, 3939 3 = ACPI, 12 = PRAM. 3940 3941 memory_corruption_check=0/1 [X86,EARLY] 3942 Some BIOSes seem to corrupt the first 64k of 3943 memory when doing things like suspend/resume. 3944 Setting this option will scan the memory 3945 looking for corruption. Enabling this will 3946 both detect corruption and prevent the kernel 3947 from using the memory being corrupted. 3948 However, it's intended as a diagnostic tool; if 3949 repeatable BIOS-originated corruption always 3950 affects the same memory, you can use memmap= 3951 to prevent the kernel from using that memory. 3952 3953 memory_corruption_check_size=size [X86,EARLY] 3954 By default it checks for corruption in the low 3955 64k, making this memory unavailable for normal 3956 use. Use this parameter to scan for 3957 corruption in more or less memory. 3958 3959 memory_corruption_check_period=seconds [X86,EARLY] 3960 By default it checks for corruption every 60 3961 seconds. Use this parameter to check at some 3962 other rate. 0 disables periodic checking. 3963 3964 memory_hotplug.memmap_on_memory 3965 [KNL,X86,ARM] Boolean flag to enable this feature. 3966 Format: {on | off (default)} 3967 When enabled, runtime hotplugged memory will 3968 allocate its internal metadata (struct pages, 3969 those vmemmap pages cannot be optimized even 3970 if hugetlb_free_vmemmap is enabled) from the 3971 hotadded memory which will allow to hotadd a 3972 lot of memory without requiring additional 3973 memory to do so. 3974 This feature is disabled by default because it 3975 has some implication on large (e.g. GB) 3976 allocations in some configurations (e.g. small 3977 memory blocks). 3978 The state of the flag can be read in 3979 /sys/module/memory_hotplug/parameters/memmap_on_memory. 3980 Note that even when enabled, there are a few cases where 3981 the feature is not effective. 3982 3983 memtest= [KNL,X86,ARM,M68K,PPC,RISCV,EARLY] Enable memtest 3984 Format: <integer> 3985 default : 0 <disable> 3986 Specifies the number of memtest passes to be 3987 performed. Each pass selects another test 3988 pattern from a given set of patterns. Memtest 3989 fills the memory with this pattern, validates 3990 memory contents and reserves bad memory 3991 regions that are detected. 3992 3993 mem_encrypt= [X86-64] AMD Secure Memory Encryption (SME) control 3994 Valid arguments: on, off 3995 Default: off 3996 mem_encrypt=on: Activate SME 3997 mem_encrypt=off: Do not activate SME 3998 3999 Refer to Documentation/virt/kvm/x86/amd-memory-encryption.rst 4000 for details on when memory encryption can be activated. 4001 4002 mem_sleep_default= [SUSPEND] Default system suspend mode: 4003 s2idle - Suspend-To-Idle 4004 shallow - Power-On Suspend or equivalent (if supported) 4005 deep - Suspend-To-RAM or equivalent (if supported) 4006 See Documentation/admin-guide/pm/sleep-states.rst. 4007 4008 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when 4009 the BIOS has incorrectly applied a workaround. TinyBIOS 4010 version 0.98 is known to be affected, 0.99 fixes the 4011 problem by letting the user disable the workaround. 4012 4013 mga= [HW,DRM] 4014 4015 microcode= [X86] Control the behavior of the microcode loader. 4016 Available options, comma separated: 4017 4018 base_rev=X - with <X> with format: <u32> 4019 Set the base microcode revision of each thread when in 4020 debug mode. 4021 4022 dis_ucode_ldr: disable the microcode loader 4023 4024 force_minrev: 4025 Enable or disable the microcode minimal revision 4026 enforcement for the runtime microcode loader. 4027 4028 mini2440= [ARM,HW,KNL] 4029 Format:[0..2][b][c][t] 4030 Default: "0tb" 4031 MINI2440 configuration specification: 4032 0 - The attached screen is the 3.5" TFT 4033 1 - The attached screen is the 7" TFT 4034 2 - The VGA Shield is attached (1024x768) 4035 Leaving out the screen size parameter will not load 4036 the TFT driver, and the framebuffer will be left 4037 unconfigured. 4038 b - Enable backlight. The TFT backlight pin will be 4039 linked to the kernel VESA blanking code and a GPIO 4040 LED. This parameter is not necessary when using the 4041 VGA shield. 4042 c - Enable the s3c camera interface. 4043 t - Reserved for enabling touchscreen support. The 4044 touchscreen support is not enabled in the mainstream 4045 kernel as of 2.6.30, a preliminary port can be found 4046 in the "bleeding edge" mini2440 support kernel at 4047 https://repo.or.cz/w/linux-2.6/mini2440.git 4048 4049 mitigations= 4050 [X86,PPC,S390,ARM64,EARLY] Control optional mitigations for 4051 CPU vulnerabilities. This is a set of curated, 4052 arch-independent options, each of which is an 4053 aggregation of existing arch-specific options. 4054 4055 Note, "mitigations" is supported if and only if the 4056 kernel was built with CPU_MITIGATIONS=y. 4057 4058 off 4059 Disable all optional CPU mitigations. This 4060 improves system performance, but it may also 4061 expose users to several CPU vulnerabilities. 4062 Equivalent to: if nokaslr then kpti=0 [ARM64] 4063 gather_data_sampling=off [X86] 4064 indirect_target_selection=off [X86] 4065 kvm.nx_huge_pages=off [X86] 4066 l1tf=off [X86] 4067 mds=off [X86] 4068 mmio_stale_data=off [X86] 4069 no_entry_flush [PPC] 4070 no_uaccess_flush [PPC] 4071 nobp=0 [S390] 4072 nopti [X86,PPC] 4073 nospectre_bhb [ARM64] 4074 nospectre_v1 [X86,PPC] 4075 nospectre_v2 [X86,PPC,S390,ARM64] 4076 reg_file_data_sampling=off [X86] 4077 retbleed=off [X86] 4078 spec_rstack_overflow=off [X86] 4079 spec_store_bypass_disable=off [X86,PPC] 4080 spectre_bhi=off [X86] 4081 spectre_v2_user=off [X86] 4082 srbds=off [X86,INTEL] 4083 ssbd=force-off [ARM64] 4084 tsa=off [X86,AMD] 4085 tsx_async_abort=off [X86] 4086 vmscape=off [X86] 4087 4088 Exceptions: 4089 This does not have any effect on 4090 kvm.nx_huge_pages when 4091 kvm.nx_huge_pages=force. 4092 4093 auto (default) 4094 Mitigate all CPU vulnerabilities, but leave SMT 4095 enabled, even if it's vulnerable. This is for 4096 users who don't want to be surprised by SMT 4097 getting disabled across kernel upgrades, or who 4098 have other ways of avoiding SMT-based attacks. 4099 Equivalent to: (default behavior) 4100 4101 auto,nosmt 4102 Mitigate all CPU vulnerabilities, disabling SMT 4103 if needed. This is for users who always want to 4104 be fully mitigated, even if it means losing SMT. 4105 Equivalent to: l1tf=flush,nosmt [X86] 4106 mds=full,nosmt [X86] 4107 tsx_async_abort=full,nosmt [X86] 4108 mmio_stale_data=full,nosmt [X86] 4109 retbleed=auto,nosmt [X86] 4110 4111 [X86] After one of the above options, additionally 4112 supports attack-vector based controls as documented in 4113 Documentation/admin-guide/hw-vuln/attack_vector_controls.rst 4114 4115 mminit_loglevel= 4116 [KNL,EARLY] When CONFIG_DEBUG_MEMORY_INIT is set, this 4117 parameter allows control of the logging verbosity for 4118 the additional memory initialisation checks. A value 4119 of 0 disables mminit logging and a level of 4 will 4120 log everything. Information is printed at KERN_DEBUG 4121 so loglevel=8 may also need to be specified. 4122 4123 mmio_stale_data= 4124 [X86,INTEL,EARLY] Control mitigation for the Processor 4125 MMIO Stale Data vulnerabilities. 4126 4127 Processor MMIO Stale Data is a class of 4128 vulnerabilities that may expose data after an MMIO 4129 operation. Exposed data could originate or end in 4130 the same CPU buffers as affected by MDS and TAA. 4131 Therefore, similar to MDS and TAA, the mitigation 4132 is to clear the affected CPU buffers. 4133 4134 This parameter controls the mitigation. The 4135 options are: 4136 4137 full - Enable mitigation on vulnerable CPUs 4138 4139 full,nosmt - Enable mitigation and disable SMT on 4140 vulnerable CPUs. 4141 4142 off - Unconditionally disable mitigation 4143 4144 On MDS or TAA affected machines, 4145 mmio_stale_data=off can be prevented by an active 4146 MDS or TAA mitigation as these vulnerabilities are 4147 mitigated with the same mechanism so in order to 4148 disable this mitigation, you need to specify 4149 mds=off and tsx_async_abort=off too. 4150 4151 Not specifying this option is equivalent to 4152 mmio_stale_data=full. 4153 4154 For details see: 4155 Documentation/admin-guide/hw-vuln/processor_mmio_stale_data.rst 4156 4157 <module>.async_probe[=<bool>] [KNL] 4158 If no <bool> value is specified or if the value 4159 specified is not a valid <bool>, enable asynchronous 4160 probe on this module. Otherwise, enable/disable 4161 asynchronous probe on this module as indicated by the 4162 <bool> value. See also: module.async_probe 4163 4164 module.async_probe=<bool> 4165 [KNL] When set to true, modules will use async probing 4166 by default. To enable/disable async probing for a 4167 specific module, use the module specific control that 4168 is documented under <module>.async_probe. When both 4169 module.async_probe and <module>.async_probe are 4170 specified, <module>.async_probe takes precedence for 4171 the specific module. 4172 4173 module.enable_dups_trace 4174 [KNL] When CONFIG_MODULE_DEBUG_AUTOLOAD_DUPS is set, 4175 this means that duplicate request_module() calls will 4176 trigger a WARN_ON() instead of a pr_warn(). Note that 4177 if MODULE_DEBUG_AUTOLOAD_DUPS_TRACE is set, WARN_ON()s 4178 will always be issued and this option does nothing. 4179 module.sig_enforce 4180 [KNL] When CONFIG_MODULE_SIG is set, this means that 4181 modules without (valid) signatures will fail to load. 4182 Note that if CONFIG_MODULE_SIG_FORCE is set, that 4183 is always true, so this option does nothing. 4184 4185 module_blacklist= [KNL] Do not load a comma-separated list of 4186 modules. Useful for debugging problem modules. 4187 4188 mousedev.tap_time= 4189 [MOUSE] Maximum time between finger touching and 4190 leaving touchpad surface for touch to be considered 4191 a tap and be reported as a left button click (for 4192 touchpads working in absolute mode only). 4193 Format: <msecs> 4194 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices 4195 reporting absolute coordinates, such as tablets 4196 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices 4197 reporting absolute coordinates, such as tablets 4198 4199 movablecore= [KNL,X86,PPC,EARLY] 4200 Format: nn[KMGTPE] | nn% 4201 This parameter is the complement to kernelcore=, it 4202 specifies the amount of memory used for migratable 4203 allocations. If both kernelcore and movablecore is 4204 specified, then kernelcore will be at *least* the 4205 specified value but may be more. If movablecore on its 4206 own is specified, the administrator must be careful 4207 that the amount of memory usable for all allocations 4208 is not too small. 4209 4210 movable_node [KNL,EARLY] Boot-time switch to make hotplugable memory 4211 NUMA nodes to be movable. This means that the memory 4212 of such nodes will be usable only for movable 4213 allocations which rules out almost all kernel 4214 allocations. Use with caution! 4215 4216 MTD_Partition= [MTD] 4217 Format: <name>,<region-number>,<size>,<offset> 4218 4219 MTD_Region= [MTD] Format: 4220 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>] 4221 4222 mtdparts= [MTD] 4223 See drivers/mtd/parsers/cmdlinepart.c 4224 4225 mtouchusb.raw_coordinates= 4226 [HW] Make the MicroTouch USB driver use raw coordinates 4227 ('y', default) or cooked coordinates ('n') 4228 4229 mtrr=debug [X86,EARLY] 4230 Enable printing debug information related to MTRR 4231 registers at boot time. 4232 4233 mtrr_chunk_size=nn[KMG,X86,EARLY] 4234 used for mtrr cleanup. It is largest continuous chunk 4235 that could hold holes aka. UC entries. 4236 4237 mtrr_gran_size=nn[KMG,X86,EARLY] 4238 Used for mtrr cleanup. It is granularity of mtrr block. 4239 Default is 1. 4240 Large value could prevent small alignment from 4241 using up MTRRs. 4242 4243 mtrr_spare_reg_nr=n [X86,EARLY] 4244 Format: <integer> 4245 Range: 0,7 : spare reg number 4246 Default : 1 4247 Used for mtrr cleanup. It is spare mtrr entries number. 4248 Set to 2 or more if your graphical card needs more. 4249 4250 multitce=off [PPC] This parameter disables the use of the pSeries 4251 firmware feature for updating multiple TCE entries 4252 at a time. 4253 4254 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card 4255 4256 netdev= [NET] NE2000 ISA network devices parameters 4257 Format: <irq>,<io>,<mem_start>,<mem_end>,<name> 4258 4259 netpoll.carrier_timeout= 4260 [NET] Specifies amount of time (in seconds) that 4261 netpoll should wait for a carrier. By default netpoll 4262 waits 4 seconds. 4263 4264 nf_conntrack.acct= 4265 [NETFILTER] Enable connection tracking flow accounting 4266 0 to disable accounting 4267 1 to enable accounting 4268 Default value is 0. 4269 4270 nfs.cache_getent= 4271 [NFS] sets the pathname to the program which is used 4272 to update the NFS client cache entries. 4273 4274 nfs.cache_getent_timeout= 4275 [NFS] sets the timeout after which an attempt to 4276 update a cache entry is deemed to have failed. 4277 4278 nfs.callback_nr_threads= 4279 [NFSv4] set the total number of threads that the 4280 NFS client will assign to service NFSv4 callback 4281 requests. 4282 4283 nfs.callback_tcpport= 4284 [NFS] set the TCP port on which the NFSv4 callback 4285 channel should listen. 4286 4287 nfs.delay_retrans= 4288 [NFS] specifies the number of times the NFSv4 client 4289 retries the request before returning an EAGAIN error, 4290 after a reply of NFS4ERR_DELAY from the server. 4291 Only applies if the softerr mount option is enabled, 4292 and the specified value is >= 0. 4293 4294 nfs.idmap_cache_timeout= 4295 [NFS] set the maximum lifetime for idmapper cache 4296 entries. 4297 4298 nfs.max_session_cb_slots= 4299 [NFSv4.1] Sets the maximum number of session 4300 slots the client will assign to the callback 4301 channel. This determines the maximum number of 4302 callbacks the client will process in parallel for 4303 a particular server. 4304 4305 nfs.max_session_slots= 4306 [NFSv4.1] Sets the maximum number of session slots 4307 the client will attempt to negotiate with the server. 4308 This limits the number of simultaneous RPC requests 4309 that the client can send to the NFSv4.1 server. 4310 Note that there is little point in setting this 4311 value higher than the max_tcp_slot_table_limit. 4312 4313 nfs.nfs4_disable_idmapping= 4314 [NFSv4] When set to the default of '1', this option 4315 ensures that both the RPC level authentication 4316 scheme and the NFS level operations agree to use 4317 numeric uids/gids if the mount is using the 4318 'sec=sys' security flavour. In effect it is 4319 disabling idmapping, which can make migration from 4320 legacy NFSv2/v3 systems to NFSv4 easier. 4321 Servers that do not support this mode of operation 4322 will be autodetected by the client, and it will fall 4323 back to using the idmapper. 4324 To turn off this behaviour, set the value to '0'. 4325 4326 nfs.nfs4_unique_id= 4327 [NFS4] Specify an additional fixed unique ident- 4328 ification string that NFSv4 clients can insert into 4329 their nfs_client_id4 string. This is typically a 4330 UUID that is generated at system install time. 4331 4332 nfs.recover_lost_locks= 4333 [NFSv4] Attempt to recover locks that were lost due 4334 to a lease timeout on the server. Please note that 4335 doing this risks data corruption, since there are 4336 no guarantees that the file will remain unchanged 4337 after the locks are lost. 4338 If you want to enable the kernel legacy behaviour of 4339 attempting to recover these locks, then set this 4340 parameter to '1'. 4341 The default parameter value of '0' causes the kernel 4342 not to attempt recovery of lost locks. 4343 4344 nfs.send_implementation_id= 4345 [NFSv4.1] Send client implementation identification 4346 information in exchange_id requests. 4347 If zero, no implementation identification information 4348 will be sent. 4349 The default is to send the implementation identification 4350 information. 4351 4352 nfs4.layoutstats_timer= 4353 [NFSv4.2] Change the rate at which the kernel sends 4354 layoutstats to the pNFS metadata server. 4355 4356 Setting this to value to 0 causes the kernel to use 4357 whatever value is the default set by the layout 4358 driver. A non-zero value sets the minimum interval 4359 in seconds between layoutstats transmissions. 4360 4361 nfsd.inter_copy_offload_enable= 4362 [NFSv4.2] When set to 1, the server will support 4363 server-to-server copies for which this server is 4364 the destination of the copy. 4365 4366 nfsd.nfs4_disable_idmapping= 4367 [NFSv4] When set to the default of '1', the NFSv4 4368 server will return only numeric uids and gids to 4369 clients using auth_sys, and will accept numeric uids 4370 and gids from such clients. This is intended to ease 4371 migration from NFSv2/v3. 4372 4373 nfsd.nfsd4_ssc_umount_timeout= 4374 [NFSv4.2] When used as the destination of a 4375 server-to-server copy, knfsd temporarily mounts 4376 the source server. It caches the mount in case 4377 it will be needed again, and discards it if not 4378 used for the number of milliseconds specified by 4379 this parameter. 4380 4381 nfsaddrs= [NFS] Deprecated. Use ip= instead. 4382 See Documentation/admin-guide/nfs/nfsroot.rst. 4383 4384 nfsroot= [NFS] nfs root filesystem for disk-less boxes. 4385 See Documentation/admin-guide/nfs/nfsroot.rst. 4386 4387 nfsrootdebug [NFS] enable nfsroot debugging messages. 4388 See Documentation/admin-guide/nfs/nfsroot.rst. 4389 4390 nmi_backtrace.backtrace_idle [KNL] 4391 Dump stacks even of idle CPUs in response to an 4392 NMI stack-backtrace request. 4393 4394 nmi_debug= [KNL,SH] Specify one or more actions to take 4395 when a NMI is triggered. 4396 Format: [state][,regs][,debounce][,die] 4397 4398 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels 4399 Format: [panic,][nopanic,][rNNN,][num] 4400 Valid num: 0 or 1 4401 0 - turn hardlockup detector in nmi_watchdog off 4402 1 - turn hardlockup detector in nmi_watchdog on 4403 rNNN - configure the watchdog with raw perf event 0xNNN 4404 4405 When panic is specified, panic when an NMI watchdog 4406 timeout occurs (or 'nopanic' to not panic on an NMI 4407 watchdog, if CONFIG_BOOTPARAM_HARDLOCKUP_PANIC is set) 4408 To disable both hard and soft lockup detectors, 4409 please see 'nowatchdog'. 4410 This is useful when you use a panic=... timeout and 4411 need the box quickly up again. 4412 4413 These settings can be accessed at runtime via 4414 the nmi_watchdog and hardlockup_panic sysctls. 4415 4416 no4lvl [RISCV,EARLY] Disable 4-level and 5-level paging modes. 4417 Forces kernel to use 3-level paging instead. 4418 4419 no5lvl [X86-64,RISCV,EARLY] Disable 5-level paging mode. Forces 4420 kernel to use 4-level paging instead. 4421 4422 noalign [KNL,ARM] 4423 4424 noapic [SMP,APIC,EARLY] Tells the kernel to not make use of any 4425 IOAPICs that may be present in the system. 4426 4427 noapictimer [APIC,X86] Don't set up the APIC timer 4428 4429 noautogroup Disable scheduler automatic task group creation. 4430 4431 nocache [ARM,EARLY] 4432 4433 no_console_suspend 4434 [HW] Never suspend the console 4435 Disable suspending of consoles during suspend and 4436 hibernate operations. Once disabled, debugging 4437 messages can reach various consoles while the rest 4438 of the system is being put to sleep (ie, while 4439 debugging driver suspend/resume hooks). This may 4440 not work reliably with all consoles, but is known 4441 to work with serial and VGA consoles. 4442 To facilitate more flexible debugging, we also add 4443 console_suspend, a printk module parameter to control 4444 it. Users could use console_suspend (usually 4445 /sys/module/printk/parameters/console_suspend) to 4446 turn on/off it dynamically. 4447 4448 no_debug_objects 4449 [KNL,EARLY] Disable object debugging 4450 4451 nodsp [SH] Disable hardware DSP at boot time. 4452 4453 noefi [EFI,EARLY] Disable EFI runtime services support. 4454 4455 no_entry_flush [PPC,EARLY] Don't flush the L1-D cache when entering the kernel. 4456 4457 noexec32 [X86-64] 4458 This affects only 32-bit executables. 4459 noexec32=on: enable non-executable mappings (default) 4460 read doesn't imply executable mappings 4461 noexec32=off: disable non-executable mappings 4462 read implies executable mappings 4463 4464 no_file_caps Tells the kernel not to honor file capabilities. The 4465 only way then for a file to be executed with privilege 4466 is to be setuid root or executed by root. 4467 4468 nofpu [MIPS,SH] Disable hardware FPU at boot time. 4469 4470 nofsgsbase [X86] Disables FSGSBASE instructions. 4471 4472 nofxsr [BUGS=X86-32] Disables x86 floating point extended 4473 register save and restore. The kernel will only save 4474 legacy floating-point registers on task switch. 4475 4476 nogbpages [X86] Do not use GB pages for kernel direct mappings. 4477 4478 no_hash_pointers 4479 [KNL,EARLY] 4480 Alias for "hash_pointers=never". 4481 4482 nohibernate [HIBERNATION] Disable hibernation and resume. 4483 4484 nohlt [ARM,ARM64,MICROBLAZE,MIPS,PPC,RISCV,SH] Forces the kernel to 4485 busy wait in do_idle() and not use the arch_cpu_idle() 4486 implementation; requires CONFIG_GENERIC_IDLE_POLL_SETUP 4487 to be effective. This is useful on platforms where the 4488 sleep(SH) or wfi(ARM,ARM64) instructions do not work 4489 correctly or when doing power measurements to evaluate 4490 the impact of the sleep instructions. This is also 4491 useful when using JTAG debugger. 4492 4493 nohpet [X86] Don't use the HPET timer. 4494 4495 nohugeiomap [KNL,X86,PPC,ARM64,EARLY] Disable kernel huge I/O mappings. 4496 4497 nohugevmalloc [KNL,X86,PPC,ARM64,EARLY] Disable kernel huge vmalloc mappings. 4498 4499 nohz= [KNL] Boottime enable/disable dynamic ticks 4500 Valid arguments: on, off 4501 Default: on 4502 4503 nohz_full= [KNL,BOOT,SMP,ISOL] 4504 The argument is a cpu list, as described above. 4505 In kernels built with CONFIG_NO_HZ_FULL=y, set 4506 the specified list of CPUs whose tick will be stopped 4507 whenever possible. The boot CPU will be forced outside 4508 the range to maintain the timekeeping. Any CPUs 4509 in this list will have their RCU callbacks offloaded, 4510 just as if they had also been called out in the 4511 rcu_nocbs= boot parameter. 4512 4513 Note that this argument takes precedence over 4514 the CONFIG_RCU_NOCB_CPU_DEFAULT_ALL option. 4515 4516 noinitrd [Deprecated,RAM] Tells the kernel not to load any configured 4517 initial RAM disk. Currently this parameter applies to 4518 initrd only, not to initramfs. But it applies to both 4519 in EFI mode. 4520 4521 nointremap [X86-64,Intel-IOMMU,EARLY] Do not enable interrupt 4522 remapping. 4523 [Deprecated - use intremap=off] 4524 4525 noinvpcid [X86,EARLY] Disable the INVPCID cpu feature. 4526 4527 noiotrap [SH] Disables trapped I/O port accesses. 4528 4529 noirqdebug [X86-32] Disables the code which attempts to detect and 4530 disable unhandled interrupt sources. 4531 4532 noisapnp [ISAPNP] Disables ISA PnP code. 4533 4534 nokaslr [KNL,EARLY] 4535 When CONFIG_RANDOMIZE_BASE is set, this disables 4536 kernel and module base offset ASLR (Address Space 4537 Layout Randomization). 4538 4539 no-kvmapf [X86,KVM,EARLY] Disable paravirtualized asynchronous page 4540 fault handling. 4541 4542 no-kvmclock [X86,KVM,EARLY] Disable paravirtualized KVM clock driver 4543 4544 nolapic [X86-32,APIC,EARLY] Do not enable or use the local APIC. 4545 4546 nolapic_timer [X86-32,APIC,EARLY] Do not use the local APIC timer. 4547 4548 nomce [X86-32] Disable Machine Check Exception 4549 4550 nomfgpt [X86-32] Disable Multi-Function General Purpose 4551 Timer usage (for AMD Geode machines). 4552 4553 nomodeset Disable kernel modesetting. Most systems' firmware 4554 sets up a display mode and provides framebuffer memory 4555 for output. With nomodeset, DRM and fbdev drivers will 4556 not load if they could possibly displace the pre- 4557 initialized output. Only the system framebuffer will 4558 be available for use. The respective drivers will not 4559 perform display-mode changes or accelerated rendering. 4560 4561 Useful as error fallback, or for testing and debugging. 4562 4563 nomodule Disable module load 4564 4565 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to 4566 shutdown the other cpus. Instead use the REBOOT_VECTOR 4567 irq. 4568 4569 nopat [X86,EARLY] Disable PAT (page attribute table extension of 4570 pagetables) support. 4571 4572 nopcid [X86-64,EARLY] Disable the PCID cpu feature. 4573 4574 nopku [X86] Disable Memory Protection Keys CPU feature found 4575 in some Intel CPUs. 4576 4577 nopti [X86-64,EARLY] 4578 Equivalent to pti=off 4579 4580 nopv= [X86,XEN,KVM,HYPER_V,VMWARE,EARLY] 4581 Disables the PV optimizations forcing the guest to run 4582 as generic guest with no PV drivers. Currently support 4583 XEN HVM, KVM, HYPER_V and VMWARE guest. 4584 4585 nopvspin [X86,XEN,KVM,EARLY] 4586 Disables the qspinlock slow path using PV optimizations 4587 which allow the hypervisor to 'idle' the guest on lock 4588 contention. 4589 4590 norandmaps Don't use address space randomization. Equivalent to 4591 echo 0 > /proc/sys/kernel/randomize_va_space 4592 4593 noreplace-smp [X86-32,SMP] Don't replace SMP instructions 4594 with UP alternatives 4595 4596 noresume [SWSUSP] Disables resume and restores original swap 4597 space. 4598 4599 no-scroll [VGA] Disables scrollback. 4600 This is required for the Braillex ib80-piezo Braille 4601 reader made by F.H. Papenmeier (Germany). 4602 4603 nosgx [X86-64,SGX,EARLY] Disables Intel SGX kernel support. 4604 4605 nosmap [PPC,EARLY] 4606 Disable SMAP (Supervisor Mode Access Prevention) 4607 even if it is supported by processor. 4608 4609 nosmep [PPC64s,EARLY] 4610 Disable SMEP (Supervisor Mode Execution Prevention) 4611 even if it is supported by processor. 4612 4613 nosmp [SMP,EARLY] Tells an SMP kernel to act as a UP kernel, 4614 and disable the IO APIC. legacy for "maxcpus=0". 4615 4616 nosmt [KNL,MIPS,PPC,EARLY] Disable symmetric multithreading (SMT). 4617 Equivalent to smt=1. 4618 4619 [KNL,LOONGARCH,X86,ARM64,PPC,S390] Disable symmetric multithreading (SMT). 4620 nosmt=force: Force disable SMT, cannot be undone 4621 via the sysfs control file. 4622 4623 nosoftlockup [KNL] Disable the soft-lockup detector. 4624 4625 nospec_store_bypass_disable 4626 [HW,EARLY] Disable all mitigations for the Speculative 4627 Store Bypass vulnerability 4628 4629 nospectre_bhb [ARM64,EARLY] Disable all mitigations for Spectre-BHB (branch 4630 history injection) vulnerability. System may allow data leaks 4631 with this option. 4632 4633 nospectre_v1 [X86,PPC,EARLY] Disable mitigations for Spectre Variant 1 4634 (bounds check bypass). With this option data leaks are 4635 possible in the system. 4636 4637 nospectre_v2 [X86,PPC_E500,ARM64,EARLY] Disable all mitigations 4638 for the Spectre variant 2 (indirect branch 4639 prediction) vulnerability. System may allow data 4640 leaks with this option. 4641 4642 no-steal-acc [X86,PV_OPS,ARM64,PPC/PSERIES,RISCV,LOONGARCH,EARLY] 4643 Disable paravirtualized steal time accounting. steal time 4644 is computed, but won't influence scheduler behaviour 4645 4646 nosync [HW,M68K] Disables sync negotiation for all devices. 4647 4648 no_timer_check [X86,APIC] Disables the code which tests for broken 4649 timer IRQ sources, i.e., the IO-APIC timer. This can 4650 work around problems with incorrect timer 4651 initialization on some boards. 4652 4653 no_uaccess_flush 4654 [PPC,EARLY] Don't flush the L1-D cache after accessing user data. 4655 4656 novmcoredd [KNL,KDUMP] 4657 Disable device dump. Device dump allows drivers to 4658 append dump data to vmcore so you can collect driver 4659 specified debug info. Drivers can append the data 4660 without any limit and this data is stored in memory, 4661 so this may cause significant memory stress. Disabling 4662 device dump can help save memory but the driver debug 4663 data will be no longer available. This parameter 4664 is only available when CONFIG_PROC_VMCORE_DEVICE_DUMP 4665 is set. 4666 4667 no-vmw-sched-clock 4668 [X86,PV_OPS,EARLY] Disable paravirtualized VMware 4669 scheduler clock and use the default one. 4670 4671 nowatchdog [KNL] Disable both lockup detectors, i.e. 4672 soft-lockup and NMI watchdog (hard-lockup). 4673 4674 nowb [ARM,EARLY] 4675 4676 nox2apic [X86-64,APIC,EARLY] Do not enable x2APIC mode. 4677 4678 NOTE: this parameter will be ignored on systems with the 4679 LEGACY_XAPIC_DISABLED bit set in the 4680 IA32_XAPIC_DISABLE_STATUS MSR. 4681 4682 noxsave [BUGS=X86] Disables x86 extended register state save 4683 and restore using xsave. The kernel will fallback to 4684 enabling legacy floating-point and sse state. 4685 4686 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended 4687 register states. The kernel will fall back to use 4688 xsave to save the states. By using this parameter, 4689 performance of saving the states is degraded because 4690 xsave doesn't support modified optimization while 4691 xsaveopt supports it on xsaveopt enabled systems. 4692 4693 noxsaves [X86] Disables xsaves and xrstors used in saving and 4694 restoring x86 extended register state in compacted 4695 form of xsave area. The kernel will fall back to use 4696 xsaveopt and xrstor to save and restore the states 4697 in standard form of xsave area. By using this 4698 parameter, xsave area per process might occupy more 4699 memory on xsaves enabled systems. 4700 4701 nr_cpus= [SMP,EARLY] Maximum number of processors that an SMP kernel 4702 could support. nr_cpus=n : n >= 1 limits the kernel to 4703 support 'n' processors. It could be larger than the 4704 number of already plugged CPU during bootup, later in 4705 runtime you can physically add extra cpu until it reaches 4706 n. So during boot up some boot time memory for per-cpu 4707 variables need be pre-allocated for later physical cpu 4708 hot plugging. 4709 4710 nr_uarts= [SERIAL] maximum number of UARTs to be registered. 4711 4712 numa=off [KNL, ARM64, PPC, RISCV, SPARC, X86, EARLY] 4713 Disable NUMA, Only set up a single NUMA node 4714 spanning all memory. 4715 4716 numa=fake=<size>[MG] 4717 [KNL, ARM64, RISCV, X86, EARLY] 4718 If given as a memory unit, fills all system RAM with 4719 nodes of size interleaved over physical nodes. 4720 4721 numa=fake=<N> 4722 [KNL, ARM64, RISCV, X86, EARLY] 4723 If given as an integer, fills all system RAM with N 4724 fake nodes interleaved over physical nodes. 4725 4726 numa=fake=<N>U 4727 [KNL, ARM64, RISCV, X86, EARLY] 4728 If given as an integer followed by 'U', it will 4729 divide each physical node into N emulated nodes. 4730 4731 numa=noacpi [X86] Don't parse the SRAT table for NUMA setup 4732 4733 numa=nohmat [X86] Don't parse the HMAT table for NUMA setup, or 4734 soft-reserved memory partitioning. 4735 4736 numa_balancing= [KNL,ARM64,PPC,RISCV,S390,X86] Enable or disable automatic 4737 NUMA balancing. 4738 Allowed values are enable and disable 4739 4740 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA. 4741 'node', 'default' can be specified 4742 This can be set from sysctl after boot. 4743 See Documentation/admin-guide/sysctl/vm.rst for details. 4744 4745 nvme.quirks= [NVME] A list of quirk entries to augment the built-in 4746 nvme quirk list. List entries are separated by a 4747 '-' character. 4748 Each entry has the form VendorID:ProductID:quirk_names. 4749 The IDs are 4-digits hex numbers and quirk_names is a 4750 list of quirk names separated by commas. A quirk name 4751 can be prefixed by '^', meaning that the specified 4752 quirk must be disabled. 4753 4754 Example: 4755 nvme.quirks=7710:2267:bogus_nid,^identify_cns-9900:7711:broken_msi 4756 4757 ohci1394_dma=early [HW,EARLY] enable debugging via the ohci1394 driver. 4758 See Documentation/core-api/debugging-via-ohci1394.rst for more 4759 info. 4760 4761 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands 4762 Rather than timing out after 20 ms if an EC 4763 command is not properly ACKed, override the length 4764 of the timeout. We have interrupts disabled while 4765 waiting for the ACK, so if this is set too high 4766 interrupts *may* be lost! 4767 4768 omap_mux= [OMAP] Override bootloader pin multiplexing. 4769 Format: <mux_mode0.mode_name=value>... 4770 For example, to override I2C bus2: 4771 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100 4772 4773 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration 4774 4775 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock] 4776 4777 boundary - index of last SLC block on Flex-OneNAND. 4778 The remaining blocks are configured as MLC blocks. 4779 lock - Configure if Flex-OneNAND boundary should be locked. 4780 Once locked, the boundary cannot be changed. 4781 1 indicates lock status, 0 indicates unlock status. 4782 4783 oops=panic [KNL,EARLY] 4784 Always panic on oopses. Default is to just kill the 4785 process, but there is a small probability of 4786 deadlocking the machine. 4787 This will also cause panics on machine check exceptions. 4788 Useful together with panic=30 to trigger a reboot. 4789 4790 page_alloc.shuffle= 4791 [KNL] Boolean flag to control whether the page allocator 4792 should randomize its free lists. This parameter can be 4793 used to enable/disable page randomization. The state of 4794 the flag can be read from sysfs at: 4795 /sys/module/page_alloc/parameters/shuffle. 4796 This parameter is only available if CONFIG_SHUFFLE_PAGE_ALLOCATOR=y. 4797 4798 page_owner= [KNL,EARLY] Boot-time page_owner enabling option. 4799 Storage of the information about who allocated 4800 each page is disabled in default. With this switch, 4801 we can turn it on. 4802 on: enable the feature 4803 4804 page_poison= [KNL,EARLY] Boot-time parameter changing the state of 4805 poisoning on the buddy allocator, available with 4806 CONFIG_PAGE_POISONING=y. 4807 off: turn off poisoning (default) 4808 on: turn on poisoning 4809 4810 page_reporting.page_reporting_order= 4811 [KNL] Minimal page reporting order 4812 Format: <integer> 4813 Adjust the minimal page reporting order. The page 4814 reporting is disabled when it exceeds MAX_PAGE_ORDER. 4815 4816 panic= [KNL] Kernel behaviour on panic: delay <timeout> 4817 timeout > 0: seconds before rebooting 4818 timeout = 0: wait forever 4819 timeout < 0: reboot immediately 4820 Format: <timeout> 4821 4822 panic_on_taint= [KNL,EARLY] 4823 Bitmask for conditionally calling panic() in add_taint() 4824 Format: <hex>[,nousertaint] 4825 Hexadecimal bitmask representing the set of TAINT flags 4826 that will cause the kernel to panic when add_taint() is 4827 called with any of the flags in this set. 4828 The optional switch "nousertaint" can be utilized to 4829 prevent userspace forced crashes by writing to sysctl 4830 /proc/sys/kernel/tainted any flagset matching with the 4831 bitmask set on panic_on_taint. 4832 See Documentation/admin-guide/tainted-kernels.rst for 4833 extra details on the taint flags that users can pick 4834 to compose the bitmask to assign to panic_on_taint. 4835 4836 panic_on_warn=1 panic() instead of WARN(). Useful to cause kdump 4837 on a WARN(). 4838 4839 panic_force_cpu= 4840 [KNL,SMP] Force panic handling to execute on a specific CPU. 4841 Format: <cpu number> 4842 Some platforms require panic handling to occur on a 4843 specific CPU for the crash kernel to function correctly. 4844 This can be due to firmware limitations, interrupt routing 4845 constraints, or platform-specific requirements where only 4846 a particular CPU can safely enter the crash kernel. 4847 When set, panic() will redirect execution to the specified 4848 CPU before proceeding with the normal panic and kexec flow. 4849 If the target CPU is offline or unavailable, panic proceeds 4850 on the current CPU. 4851 This option should only be used for systems with the above 4852 constraints as it might cause the panic operation to be less reliable. 4853 4854 panic_print= Bitmask for printing system info when panic happens. 4855 User can chose combination of the following bits: 4856 bit 0: print all tasks info 4857 bit 1: print system memory info 4858 bit 2: print timer info 4859 bit 3: print locks info if CONFIG_LOCKDEP is on 4860 bit 4: print ftrace buffer 4861 bit 5: replay all kernel messages on consoles at the end of panic 4862 bit 6: print all CPUs backtrace (if available in the arch) 4863 bit 7: print only tasks in uninterruptible (blocked) state 4864 *Be aware* that this option may print a _lot_ of lines, 4865 so there are risks of losing older messages in the log. 4866 Use this option carefully, maybe worth to setup a 4867 bigger log buffer with "log_buf_len" along with this. 4868 4869 panic_sys_info= A comma separated list of extra information to be dumped 4870 on panic. 4871 Format: val[,val...] 4872 Where @val can be any of the following: 4873 4874 tasks: print all tasks info 4875 mem: print system memory info 4876 timers: print timers info 4877 locks: print locks info if CONFIG_LOCKDEP is on 4878 ftrace: print ftrace buffer 4879 all_bt: print all CPUs backtrace (if available in the arch) 4880 blocked_tasks: print only tasks in uninterruptible (blocked) state 4881 4882 This is a human readable alternative to the 'panic_print' option. 4883 4884 panic_console_replay 4885 When panic happens, replay all kernel messages on 4886 consoles at the end of panic. 4887 4888 parkbd.port= [HW] Parallel port number the keyboard adapter is 4889 connected to, default is 0. 4890 Format: <parport#> 4891 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation, 4892 0 for XT, 1 for AT (default is AT). 4893 Format: <mode> 4894 4895 parport= [HW,PPT] Specify parallel ports. 0 disables. 4896 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] } 4897 Use 'auto' to force the driver to use any 4898 IRQ/DMA settings detected (the default is to 4899 ignore detected IRQ/DMA settings because of 4900 possible conflicts). You can specify the base 4901 address, IRQ, and DMA settings; IRQ and DMA 4902 should be numbers, or 'auto' (for using detected 4903 settings on that particular port), or 'nofifo' 4904 (to avoid using a FIFO even if it is detected). 4905 Parallel ports are assigned in the order they 4906 are specified on the command line, starting 4907 with parport0. 4908 4909 parport_init_mode= [HW,PPT] 4910 Configure VIA parallel port to operate in 4911 a specific mode. This is necessary on Pegasos 4912 computer where firmware has no options for setting 4913 up parallel port mode and sets it to spp. 4914 Currently this function knows 686a and 8231 chips. 4915 Format: [spp|ps2|epp|ecp|ecpepp] 4916 4917 pata_legacy.all= [HW,LIBATA] 4918 Format: <int> 4919 Set to non-zero to probe primary and secondary ISA 4920 port ranges on PCI systems where no PCI PATA device 4921 has been found at either range. Disabled by default. 4922 4923 pata_legacy.autospeed= [HW,LIBATA] 4924 Format: <int> 4925 Set to non-zero if a chip is present that snoops speed 4926 changes. Disabled by default. 4927 4928 pata_legacy.ht6560a= [HW,LIBATA] 4929 Format: <int> 4930 Set to 1, 2, or 3 for HT 6560A on the primary channel, 4931 the secondary channel, or both channels respectively. 4932 Disabled by default. 4933 4934 pata_legacy.ht6560b= [HW,LIBATA] 4935 Format: <int> 4936 Set to 1, 2, or 3 for HT 6560B on the primary channel, 4937 the secondary channel, or both channels respectively. 4938 Disabled by default. 4939 4940 pata_legacy.iordy_mask= [HW,LIBATA] 4941 Format: <int> 4942 IORDY enable mask. Set individual bits to allow IORDY 4943 for the respective channel. Bit 0 is for the first 4944 legacy channel handled by this driver, bit 1 is for 4945 the second channel, and so on. The sequence will often 4946 correspond to the primary legacy channel, the secondary 4947 legacy channel, and so on, but the handling of a PCI 4948 bus and the use of other driver options may interfere 4949 with the sequence. By default IORDY is allowed across 4950 all channels. 4951 4952 pata_legacy.opti82c46x= [HW,LIBATA] 4953 Format: <int> 4954 Set to 1, 2, or 3 for Opti 82c611A on the primary 4955 channel, the secondary channel, or both channels 4956 respectively. Disabled by default. 4957 4958 pata_legacy.opti82c611a= [HW,LIBATA] 4959 Format: <int> 4960 Set to 1, 2, or 3 for Opti 82c465MV on the primary 4961 channel, the secondary channel, or both channels 4962 respectively. Disabled by default. 4963 4964 pata_legacy.pio_mask= [HW,LIBATA] 4965 Format: <int> 4966 PIO mode mask for autospeed devices. Set individual 4967 bits to allow the use of the respective PIO modes. 4968 Bit 0 is for mode 0, bit 1 is for mode 1, and so on. 4969 All modes allowed by default. 4970 4971 pata_legacy.probe_all= [HW,LIBATA] 4972 Format: <int> 4973 Set to non-zero to probe tertiary and further ISA 4974 port ranges on PCI systems. Disabled by default. 4975 4976 pata_legacy.probe_mask= [HW,LIBATA] 4977 Format: <int> 4978 Probe mask for legacy ISA PATA ports. Depending on 4979 platform configuration and the use of other driver 4980 options up to 6 legacy ports are supported: 0x1f0, 4981 0x170, 0x1e8, 0x168, 0x1e0, 0x160, however probing 4982 of individual ports can be disabled by setting the 4983 corresponding bits in the mask to 1. Bit 0 is for 4984 the first port in the list above (0x1f0), and so on. 4985 By default all supported ports are probed. 4986 4987 pata_legacy.qdi= [HW,LIBATA] 4988 Format: <int> 4989 Set to non-zero to probe QDI controllers. By default 4990 set to 1 if CONFIG_PATA_QDI_MODULE, 0 otherwise. 4991 4992 pata_legacy.winbond= [HW,LIBATA] 4993 Format: <int> 4994 Set to non-zero to probe Winbond controllers. Use 4995 the standard I/O port (0x130) if 1, otherwise the 4996 value given is the I/O port to use (typically 0x1b0). 4997 By default set to 1 if CONFIG_PATA_WINBOND_VLB_MODULE, 4998 0 otherwise. 4999 5000 pata_platform.pio_mask= [HW,LIBATA] 5001 Format: <int> 5002 Supported PIO mode mask. Set individual bits to allow 5003 the use of the respective PIO modes. Bit 0 is for 5004 mode 0, bit 1 is for mode 1, and so on. Mode 0 only 5005 allowed by default. 5006 5007 pause_on_oops=<int> 5008 Halt all CPUs after the first oops has been printed for 5009 the specified number of seconds. This is to be used if 5010 your oopses keep scrolling off the screen. 5011 5012 pci=option[,option...] [PCI,EARLY] various PCI subsystem options. 5013 5014 Some options herein operate on a specific device 5015 or a set of devices (<pci_dev>). These are 5016 specified in one of the following formats: 5017 5018 [<domain>:]<bus>:<dev>.<func>[/<dev>.<func>]* 5019 pci:<vendor>:<device>[:<subvendor>:<subdevice>] 5020 5021 Note: the first format specifies a PCI 5022 bus/device/function address which may change 5023 if new hardware is inserted, if motherboard 5024 firmware changes, or due to changes caused 5025 by other kernel parameters. If the 5026 domain is left unspecified, it is 5027 taken to be zero. Optionally, a path 5028 to a device through multiple device/function 5029 addresses can be specified after the base 5030 address (this is more robust against 5031 renumbering issues). The second format 5032 selects devices using IDs from the 5033 configuration space which may match multiple 5034 devices in the system. 5035 5036 earlydump dump PCI config space before the kernel 5037 changes anything 5038 off [X86] don't probe for the PCI bus 5039 bios [X86-32] force use of PCI BIOS, don't access 5040 the hardware directly. Use this if your machine 5041 has a non-standard PCI host bridge. 5042 nobios [X86-32] disallow use of PCI BIOS, only direct 5043 hardware access methods are allowed. Use this 5044 if you experience crashes upon bootup and you 5045 suspect they are caused by the BIOS. 5046 conf1 [X86] Force use of PCI Configuration Access 5047 Mechanism 1 (config address in IO port 0xCF8, 5048 data in IO port 0xCFC, both 32-bit). 5049 conf2 [X86] Force use of PCI Configuration Access 5050 Mechanism 2 (IO port 0xCF8 is an 8-bit port for 5051 the function, IO port 0xCFA, also 8-bit, sets 5052 bus number. The config space is then accessed 5053 through ports 0xC000-0xCFFF). 5054 See http://wiki.osdev.org/PCI for more info 5055 on the configuration access mechanisms. 5056 noaer [PCIE] If the PCIEAER kernel config parameter is 5057 enabled, this kernel boot option can be used to 5058 disable the use of PCIE advanced error reporting. 5059 nodomains [PCI] Disable support for multiple PCI 5060 root domains (aka PCI segments, in ACPI-speak). 5061 nommconf [X86] Disable use of MMCONFIG for PCI 5062 Configuration 5063 check_enable_amd_mmconf [X86] check for and enable 5064 properly configured MMIO access to PCI 5065 config space on AMD family 10h CPU 5066 nomsi [MSI] If the PCI_MSI kernel config parameter is 5067 enabled, this kernel boot option can be used to 5068 disable the use of MSI interrupts system-wide. 5069 noioapicquirk [APIC] Disable all boot interrupt quirks. 5070 Safety option to keep boot IRQs enabled. This 5071 should never be necessary. 5072 ioapicreroute [APIC] Enable rerouting of boot IRQs to the 5073 primary IO-APIC for bridges that cannot disable 5074 boot IRQs. This fixes a source of spurious IRQs 5075 when the system masks IRQs. 5076 noioapicreroute [APIC] Disable workaround that uses the 5077 boot IRQ equivalent of an IRQ that connects to 5078 a chipset where boot IRQs cannot be disabled. 5079 The opposite of ioapicreroute. 5080 biosirq [X86-32] Use PCI BIOS calls to get the interrupt 5081 routing table. These calls are known to be buggy 5082 on several machines and they hang the machine 5083 when used, but on other computers it's the only 5084 way to get the interrupt routing table. Try 5085 this option if the kernel is unable to allocate 5086 IRQs or discover secondary PCI buses on your 5087 motherboard. 5088 rom [X86] Assign address space to expansion ROMs. 5089 Use with caution as certain devices share 5090 address decoders between ROMs and other 5091 resources. 5092 norom [X86] Do not assign address space to 5093 expansion ROMs that do not already have 5094 BIOS assigned address ranges. 5095 nobar [X86] Do not assign address space to the 5096 BARs that weren't assigned by the BIOS. 5097 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be 5098 assigned automatically to PCI devices. You can 5099 make the kernel exclude IRQs of your ISA cards 5100 this way. 5101 pirqaddr=0xAAAAA [X86] Specify the physical address 5102 of the PIRQ table (normally generated 5103 by the BIOS) if it is outside the 5104 F0000h-100000h range. 5105 lastbus=N [X86] Scan all buses thru bus #N. Can be 5106 useful if the kernel is unable to find your 5107 secondary buses and you want to tell it 5108 explicitly which ones they are. 5109 assign-busses [X86] Always assign all PCI bus 5110 numbers ourselves, overriding 5111 whatever the firmware may have done. 5112 usepirqmask [X86] Honor the possible IRQ mask stored 5113 in the BIOS $PIR table. This is needed on 5114 some systems with broken BIOSes, notably 5115 some HP Pavilion N5400 and Omnibook XE3 5116 notebooks. This will have no effect if ACPI 5117 IRQ routing is enabled. 5118 noacpi [X86] Do not use ACPI for IRQ routing 5119 or for PCI scanning. 5120 use_crs [X86] Use PCI host bridge window information 5121 from ACPI. On BIOSes from 2008 or later, this 5122 is enabled by default. If you need to use this, 5123 please report a bug. 5124 nocrs [X86] Ignore PCI host bridge windows from ACPI. 5125 If you need to use this, please report a bug. 5126 use_e820 [X86] Use E820 reservations to exclude parts of 5127 PCI host bridge windows. This is a workaround 5128 for BIOS defects in host bridge _CRS methods. 5129 If you need to use this, please report a bug to 5130 <linux-pci@vger.kernel.org>. 5131 no_e820 [X86] Ignore E820 reservations for PCI host 5132 bridge windows. This is the default on modern 5133 hardware. If you need to use this, please report 5134 a bug to <linux-pci@vger.kernel.org>. 5135 routeirq Do IRQ routing for all PCI devices. 5136 This is normally done in pci_enable_device(), 5137 so this option is a temporary workaround 5138 for broken drivers that don't call it. 5139 skip_isa_align [X86] do not align io start addr, so can 5140 handle more pci cards 5141 noearly [X86] Don't do any early type 1 scanning. 5142 This might help on some broken boards which 5143 machine check when some devices' config space 5144 is read. But various workarounds are disabled 5145 and some IOMMU drivers will not work. 5146 bfsort Sort PCI devices into breadth-first order. 5147 This sorting is done to get a device 5148 order compatible with older (<= 2.4) kernels. 5149 nobfsort Don't sort PCI devices into breadth-first order. 5150 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size) 5151 tuning and use the BIOS-configured MPS defaults. 5152 pcie_bus_safe Set every device's MPS to the largest value 5153 supported by all devices below the root complex. 5154 pcie_bus_perf Set device MPS to the largest allowable MPS 5155 based on its parent bus. Also set MRRS (Max 5156 Read Request Size) to the largest supported 5157 value (no larger than the MPS that the device 5158 or bus can support) for best performance. 5159 pcie_bus_peer2peer Set every device's MPS to 128B, which 5160 every device is guaranteed to support. This 5161 configuration allows peer-to-peer DMA between 5162 any pair of devices, possibly at the cost of 5163 reduced performance. This also guarantees 5164 that hot-added devices will work. 5165 cbiosize=nn[KMG] The fixed amount of bus space which is 5166 reserved for the CardBus bridge's IO window. 5167 The default value is 256 bytes. 5168 cbmemsize=nn[KMG] The fixed amount of bus space which is 5169 reserved for the CardBus bridge's memory 5170 window. The default value is 64 megabytes. 5171 resource_alignment= 5172 Format: 5173 [<order of align>@]<pci_dev>[; ...] 5174 Specifies alignment and device to reassign 5175 aligned memory resources. How to 5176 specify the device is described above. 5177 If <order of align> is not specified, 5178 PAGE_SIZE is used as alignment. 5179 A PCI-PCI bridge can be specified if resource 5180 windows need to be expanded. 5181 To specify the alignment for several 5182 instances of a device, the PCI vendor, 5183 device, subvendor, and subdevice may be 5184 specified, e.g., 12@pci:8086:9c22:103c:198f 5185 for 4096-byte alignment. 5186 ecrc= Enable/disable PCIe ECRC (transaction layer 5187 end-to-end CRC checking). Only effective if 5188 OS has native AER control (either granted by 5189 ACPI _OSC or forced via "pcie_ports=native") 5190 bios: Use BIOS/firmware settings. This is the 5191 the default. 5192 off: Turn ECRC off 5193 on: Turn ECRC on. 5194 hpiosize=nn[KMG] The fixed amount of bus space which is 5195 reserved for hotplug bridge's IO window. 5196 Default size is 256 bytes. 5197 hpmmiosize=nn[KMG] The fixed amount of bus space which is 5198 reserved for hotplug bridge's MMIO window. 5199 Default size is 2 megabytes. 5200 hpmmioprefsize=nn[KMG] The fixed amount of bus space which is 5201 reserved for hotplug bridge's MMIO_PREF window. 5202 Default size is 2 megabytes. 5203 hpmemsize=nn[KMG] The fixed amount of bus space which is 5204 reserved for hotplug bridge's MMIO and 5205 MMIO_PREF window. 5206 Default size is 2 megabytes. 5207 hpbussize=nn The minimum amount of additional bus numbers 5208 reserved for buses below a hotplug bridge. 5209 Default is 1. 5210 realloc= Enable/disable reallocating PCI bridge resources 5211 if allocations done by BIOS are too small to 5212 accommodate resources required by all child 5213 devices. 5214 off: Turn realloc off 5215 on: Turn realloc on 5216 realloc same as realloc=on 5217 noari do not use PCIe ARI. 5218 noats [PCIE, Intel-IOMMU, AMD-IOMMU] 5219 do not use PCIe ATS (and IOMMU device IOTLB). 5220 pcie_scan_all Scan all possible PCIe devices. Otherwise we 5221 only look for one device below a PCIe downstream 5222 port. 5223 big_root_window Try to add a big 64bit memory window to the PCIe 5224 root complex on AMD CPUs. Some GFX hardware 5225 can resize a BAR to allow access to all VRAM. 5226 Adding the window is slightly risky (it may 5227 conflict with unreported devices), so this 5228 taints the kernel. 5229 disable_acs_redir=<pci_dev>[; ...] 5230 Specify one or more PCI devices (in the format 5231 specified above) separated by semicolons. 5232 Each device specified will have the PCI ACS 5233 redirect capabilities forced off which will 5234 allow P2P traffic between devices through 5235 bridges without forcing it upstream. Note: 5236 this removes isolation between devices and 5237 may put more devices in an IOMMU group. 5238 config_acs= 5239 Format: 5240 <ACS flags>@<pci_dev>[; ...] 5241 Specify one or more PCI devices (in the format 5242 specified above) optionally prepended with flags 5243 and separated by semicolons. The respective 5244 capabilities will be enabled, disabled or 5245 unchanged based on what is specified in 5246 flags. 5247 5248 ACS Flags is defined as follows: 5249 bit-0 : ACS Source Validation 5250 bit-1 : ACS Translation Blocking 5251 bit-2 : ACS P2P Request Redirect 5252 bit-3 : ACS P2P Completion Redirect 5253 bit-4 : ACS Upstream Forwarding 5254 bit-5 : ACS P2P Egress Control 5255 bit-6 : ACS Direct Translated P2P 5256 Each bit can be marked as: 5257 '0' – force disabled 5258 '1' – force enabled 5259 'x' – unchanged 5260 For example, 5261 pci=config_acs=10x@pci:0:0 5262 would configure all devices that support 5263 ACS to enable P2P Request Redirect, disable 5264 Translation Blocking, and leave Source 5265 Validation unchanged from whatever power-up 5266 or firmware set it to. 5267 5268 Note: this may remove isolation between devices 5269 and may put more devices in an IOMMU group. 5270 force_floating [S390] Force usage of floating interrupts. 5271 nomio [S390] Do not use MIO instructions. 5272 norid [S390] ignore the RID field and force use of 5273 one PCI domain per PCI function 5274 notph [PCIE] If the PCIE_TPH kernel config parameter 5275 is enabled, this kernel boot option can be used 5276 to disable PCIe TLP Processing Hints support 5277 system-wide. 5278 5279 pcie_aspm= [PCIE] Forcibly enable or ignore PCIe Active State Power 5280 Management. 5281 off Don't touch ASPM configuration at all. Leave any 5282 configuration done by firmware unchanged. 5283 force Enable ASPM even on devices that claim not to support it. 5284 WARNING: Forcing ASPM on may cause system lockups. 5285 5286 pcie_ports= [PCIE] PCIe port services handling: 5287 native Use native PCIe services (PME, AER, DPC, PCIe hotplug) 5288 even if the platform doesn't give the OS permission to 5289 use them. This may cause conflicts if the platform 5290 also tries to use these services. 5291 dpc-native Use native PCIe service for DPC only. May 5292 cause conflicts if firmware uses AER or DPC. 5293 compat Disable native PCIe services (PME, AER, DPC, PCIe 5294 hotplug). 5295 5296 pcie_port_pm= [PCIE] PCIe port power management handling: 5297 off Disable power management of all PCIe ports 5298 force Forcibly enable power management of all PCIe ports 5299 5300 pcie_pme= [PCIE,PM] Native PCIe PME signaling options: 5301 nomsi Do not use MSI for native PCIe PME signaling (this makes 5302 all PCIe root ports use INTx for all services). 5303 5304 pcmv= [HW,PCMCIA] BadgePAD 4 5305 5306 pd_ignore_unused 5307 [PM] 5308 Keep all power-domains already enabled by bootloader on, 5309 even if no driver has claimed them. This is useful 5310 for debug and development, but should not be 5311 needed on a platform with proper driver support. 5312 5313 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at 5314 boot time. 5315 Format: { 0 | 1 } 5316 See arch/parisc/kernel/pdc_chassis.c 5317 5318 percpu_alloc= [MM,EARLY] 5319 Select which percpu first chunk allocator to use. 5320 Currently supported values are "embed" and "page". 5321 Archs may support subset or none of the selections. 5322 See comments in mm/percpu.c for details on each 5323 allocator. This parameter is primarily for debugging 5324 and performance comparison. 5325 5326 pirq= [SMP,APIC] Manual mp-table setup 5327 See Documentation/arch/x86/i386/IO-APIC.rst. 5328 5329 plip= [PPT,NET] Parallel port network link 5330 Format: { parport<nr> | timid | 0 } 5331 See also Documentation/admin-guide/parport.rst. 5332 5333 pmtmr= [X86] Manual setup of pmtmr I/O Port. 5334 Override pmtimer IOPort with a hex value. 5335 e.g. pmtmr=0x508 5336 5337 pmu_override= [PPC] Override the PMU. 5338 This option takes over the PMU facility, so it is no 5339 longer usable by perf. Setting this option starts the 5340 PMU counters by setting MMCR0 to 0 (the FC bit is 5341 cleared). If a number is given, then MMCR1 is set to 5342 that number, otherwise (e.g., 'pmu_override=on'), MMCR1 5343 remains 0. 5344 5345 pm_async= [PM] 5346 Format: off 5347 This parameter sets the initial value of the 5348 /sys/power/pm_async sysfs knob at boot time. 5349 If set to "off", disables asynchronous suspend and 5350 resume of devices during system-wide power transitions. 5351 This can be useful on platforms where device 5352 dependencies are not well-defined, or for debugging 5353 power management issues. Asynchronous operations are 5354 enabled by default. 5355 5356 5357 pm_debug_messages [SUSPEND,KNL] 5358 Enable suspend/resume debug messages during boot up. 5359 5360 pnp.debug=1 [PNP] 5361 Enable PNP debug messages (depends on the 5362 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time 5363 via /sys/module/pnp/parameters/debug. We always show 5364 current resource usage; turning this on also shows 5365 possible settings and some assignment information. 5366 5367 pnpacpi= [ACPI] 5368 { off } 5369 5370 pnpbios= [ISAPNP] 5371 { on | off | curr | res | no-curr | no-res } 5372 5373 pnp_reserve_irq= 5374 [ISAPNP] Exclude IRQs for the autoconfiguration 5375 5376 pnp_reserve_dma= 5377 [ISAPNP] Exclude DMAs for the autoconfiguration 5378 5379 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration 5380 Ranges are in pairs (I/O port base and size). 5381 5382 pnp_reserve_mem= 5383 [ISAPNP] Exclude memory regions for the 5384 autoconfiguration. 5385 Ranges are in pairs (memory base and size). 5386 5387 ports= [IP_VS_FTP] IPVS ftp helper module 5388 Default is 21. 5389 Up to 8 (IP_VS_APP_MAX_PORTS) ports 5390 may be specified. 5391 Format: <port>,<port>.... 5392 5393 possible_cpus= [SMP,S390,X86] 5394 Format: <unsigned int> 5395 Set the number of possible CPUs, overriding the 5396 regular discovery mechanisms (such as ACPI/FW, etc). 5397 5398 powersave=off [PPC] This option disables power saving features. 5399 It specifically disables cpuidle and sets the 5400 platform machine description specific power_save 5401 function to NULL. On Idle the CPU just reduces 5402 execution priority. 5403 5404 ppc_strict_facility_enable 5405 [PPC,ENABLE] This option catches any kernel floating point, 5406 Altivec, VSX and SPE outside of regions specifically 5407 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()). 5408 There is some performance impact when enabling this. 5409 5410 ppc_tm= [PPC,EARLY] 5411 Format: {"off"} 5412 Disable Hardware Transactional Memory 5413 5414 preempt= [KNL] 5415 Select preemption mode if you have CONFIG_PREEMPT_DYNAMIC 5416 none - Limited to cond_resched() calls 5417 voluntary - Limited to cond_resched() and might_sleep() calls 5418 full - Any section that isn't explicitly preempt disabled 5419 can be preempted anytime. Tasks will also yield 5420 contended spinlocks (if the critical section isn't 5421 explicitly preempt disabled beyond the lock itself). 5422 lazy - Scheduler controlled. Similar to full but instead 5423 of preempting the task immediately, the task gets 5424 one HZ tick time to yield itself before the 5425 preemption will be forced. One preemption is when the 5426 task returns to user space. 5427 5428 print-fatal-signals= 5429 [KNL] debug: print fatal signals 5430 5431 If enabled, warn about various signal handling 5432 related application anomalies: too many signals, 5433 too many POSIX.1 timers, fatal signals causing a 5434 coredump - etc. 5435 5436 If you hit the warning due to signal overflow, 5437 you might want to try "ulimit -i unlimited". 5438 5439 default: off. 5440 5441 printk.always_kmsg_dump= 5442 Trigger kmsg_dump for cases other than kernel oops or 5443 panics 5444 Format: <bool> (1/Y/y=enable, 0/N/n=disable) 5445 default: disabled 5446 5447 printk.console_no_auto_verbose= 5448 Disable console loglevel raise on oops, panic 5449 or lockdep-detected issues (only if lock debug is on). 5450 With an exception to setups with low baudrate on 5451 serial console, keeping this 0 is a good choice 5452 in order to provide more debug information. 5453 Format: <bool> 5454 default: 0 (auto_verbose is enabled) 5455 5456 printk.debug_non_panic_cpus= 5457 Allows storing messages from non-panic CPUs into 5458 the printk log buffer during panic(). They are 5459 flushed to consoles by the panic-CPU on 5460 a best-effort basis. 5461 Format: <bool> (1/Y/y=enable, 0/N/n=disable) 5462 Default: disabled 5463 5464 printk.devkmsg={on,off,ratelimit} 5465 Control writing to /dev/kmsg. 5466 on - unlimited logging to /dev/kmsg from userspace 5467 off - logging to /dev/kmsg disabled 5468 ratelimit - ratelimit the logging 5469 Default: ratelimit 5470 5471 printk.time= Show timing data prefixed to each printk message line 5472 Format: <bool> (1/Y/y=enable, 0/N/n=disable) 5473 5474 proc_mem.force_override= [KNL] 5475 Format: {always | ptrace | never} 5476 Traditionally /proc/pid/mem allows memory permissions to be 5477 overridden without restrictions. This option may be set to 5478 restrict that. Can be one of: 5479 - 'always': traditional behavior always allows mem overrides. 5480 - 'ptrace': only allow mem overrides for active ptracers. 5481 - 'never': never allow mem overrides. 5482 If not specified, default is the CONFIG_PROC_MEM_* choice. 5483 5484 processor.max_cstate= [HW,ACPI] 5485 Limit processor to maximum C-state 5486 max_cstate=9 overrides any DMI blacklist limit. 5487 5488 processor.nocst [HW,ACPI] 5489 Ignore the _CST method to determine C-states, 5490 instead using the legacy FADT method 5491 5492 profile= [KNL] Enable kernel profiling via /proc/profile 5493 Format: [<profiletype>,]<number> 5494 Param: <profiletype>: "schedule" or "kvm" 5495 [defaults to kernel profiling] 5496 Param: "schedule" - profile schedule points. 5497 Param: "kvm" - profile VM exits. 5498 Param: <number> - step/bucket size as a power of 2 for 5499 statistical time based profiling. 5500 5501 prot_virt= [S390] enable hosting protected virtual machines 5502 isolated from the hypervisor (if hardware supports 5503 that). If enabled, the default kernel base address 5504 might be overridden even when Kernel Address Space 5505 Layout Randomization is disabled. 5506 Format: <bool> 5507 5508 psi= [KNL] Enable or disable pressure stall information 5509 tracking. 5510 Format: <bool> 5511 5512 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to 5513 probe for; one of (bare|imps|exps|lifebook|any). 5514 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports 5515 per second. 5516 psmouse.resetafter= [HW,MOUSE] 5517 Try to reset the device after so many bad packets 5518 (0 = never). 5519 psmouse.resolution= 5520 [HW,MOUSE] Set desired mouse resolution, in dpi. 5521 psmouse.smartscroll= 5522 [HW,MOUSE] Controls Logitech smartscroll autorepeat. 5523 0 = disabled, 1 = enabled (default). 5524 5525 pstore.backend= Specify the name of the pstore backend to use 5526 5527 pti= [X86-64] Control Page Table Isolation of user and 5528 kernel address spaces. Disabling this feature 5529 removes hardening, but improves performance of 5530 system calls and interrupts. 5531 5532 on - unconditionally enable 5533 off - unconditionally disable 5534 auto - kernel detects whether your CPU model is 5535 vulnerable to issues that PTI mitigates 5536 5537 Not specifying this option is equivalent to pti=auto. 5538 5539 pty.legacy_count= 5540 [KNL] Number of legacy pty's. Overwrites compiled-in 5541 default number. 5542 5543 quiet [KNL,EARLY] Disable most log messages 5544 5545 r128= [HW,DRM] 5546 5547 radix_hcall_invalidate=on [PPC/PSERIES] 5548 Disable RADIX GTSE feature and use hcall for TLB 5549 invalidate. 5550 5551 raid= [HW,RAID] 5552 See Documentation/admin-guide/md.rst. 5553 5554 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes 5555 See Documentation/admin-guide/blockdev/ramdisk.rst. 5556 5557 ramdisk_start= [Deprecated,RAM] RAM disk image start address 5558 5559 random.trust_cpu=off 5560 [KNL,EARLY] Disable trusting the use of the CPU's 5561 random number generator (if available) to 5562 initialize the kernel's RNG. 5563 5564 random.trust_bootloader=off 5565 [KNL,EARLY] Disable trusting the use of the a seed 5566 passed by the bootloader (if available) to 5567 initialize the kernel's RNG. 5568 5569 randomize_kstack_offset= 5570 [KNL,EARLY] Enable or disable kernel stack offset 5571 randomization, which provides roughly 5 bits of 5572 entropy, frustrating memory corruption attacks 5573 that depend on stack address determinism or 5574 cross-syscall address exposures. This is only 5575 available on architectures that have defined 5576 CONFIG_HAVE_ARCH_RANDOMIZE_KSTACK_OFFSET. 5577 Format: <bool> (1/Y/y=enable, 0/N/n=disable) 5578 Default is CONFIG_RANDOMIZE_KSTACK_OFFSET_DEFAULT. 5579 5580 ras=option[,option,...] [KNL] RAS-specific options 5581 5582 cec_disable [X86] 5583 Disable the Correctable Errors Collector, 5584 see CONFIG_RAS_CEC help text. 5585 5586 rcu_nocbs[=cpu-list] 5587 [KNL] The optional argument is a cpu list, 5588 as described above. 5589 5590 In kernels built with CONFIG_RCU_NOCB_CPU=y, 5591 enable the no-callback CPU mode, which prevents 5592 such CPUs' callbacks from being invoked in 5593 softirq context. Invocation of such CPUs' RCU 5594 callbacks will instead be offloaded to "rcuox/N" 5595 kthreads created for that purpose, where "x" is 5596 "p" for RCU-preempt, "s" for RCU-sched, and "g" 5597 for the kthreads that mediate grace periods; and 5598 "N" is the CPU number. This reduces OS jitter on 5599 the offloaded CPUs, which can be useful for HPC 5600 and real-time workloads. It can also improve 5601 energy efficiency for asymmetric multiprocessors. 5602 5603 If a cpulist is passed as an argument, the specified 5604 list of CPUs is set to no-callback mode from boot. 5605 5606 Otherwise, if the '=' sign and the cpulist 5607 arguments are omitted, no CPU will be set to 5608 no-callback mode from boot but the mode may be 5609 toggled at runtime via cpusets. 5610 5611 Note that this argument takes precedence over 5612 the CONFIG_RCU_NOCB_CPU_DEFAULT_ALL option. 5613 5614 rcu_nocb_poll [KNL] 5615 Rather than requiring that offloaded CPUs 5616 (specified by rcu_nocbs= above) explicitly 5617 awaken the corresponding "rcuoN" kthreads, 5618 make these kthreads poll for callbacks. 5619 This improves the real-time response for the 5620 offloaded CPUs by relieving them of the need to 5621 wake up the corresponding kthread, but degrades 5622 energy efficiency by requiring that the kthreads 5623 periodically wake up to do the polling. 5624 5625 rcutree.blimit= [KNL] 5626 Set maximum number of finished RCU callbacks to 5627 process in one batch. 5628 5629 rcutree.csd_lock_suppress_rcu_stall= [KNL] 5630 Do only a one-line RCU CPU stall warning when 5631 there is an ongoing too-long CSD-lock wait. 5632 5633 rcutree.do_rcu_barrier= [KNL] 5634 Request a call to rcu_barrier(). This is 5635 throttled so that userspace tests can safely 5636 hammer on the sysfs variable if they so choose. 5637 If triggered before the RCU grace-period machinery 5638 is fully active, this will error out with EAGAIN. 5639 5640 rcutree.dump_tree= [KNL] 5641 Dump the structure of the rcu_node combining tree 5642 out at early boot. This is used for diagnostic 5643 purposes, to verify correct tree setup. 5644 5645 rcutree.gp_cleanup_delay= [KNL] 5646 Set the number of jiffies to delay each step of 5647 RCU grace-period cleanup. 5648 5649 rcutree.gp_init_delay= [KNL] 5650 Set the number of jiffies to delay each step of 5651 RCU grace-period initialization. 5652 5653 rcutree.gp_preinit_delay= [KNL] 5654 Set the number of jiffies to delay each step of 5655 RCU grace-period pre-initialization, that is, 5656 the propagation of recent CPU-hotplug changes up 5657 the rcu_node combining tree. 5658 5659 rcutree.jiffies_till_first_fqs= [KNL] 5660 Set delay from grace-period initialization to 5661 first attempt to force quiescent states. 5662 Units are jiffies, minimum value is zero, 5663 and maximum value is HZ. 5664 5665 rcutree.jiffies_till_next_fqs= [KNL] 5666 Set delay between subsequent attempts to force 5667 quiescent states. Units are jiffies, minimum 5668 value is one, and maximum value is HZ. 5669 5670 rcutree.jiffies_till_sched_qs= [KNL] 5671 Set required age in jiffies for a 5672 given grace period before RCU starts 5673 soliciting quiescent-state help from 5674 rcu_note_context_switch() and cond_resched(). 5675 If not specified, the kernel will calculate 5676 a value based on the most recent settings 5677 of rcutree.jiffies_till_first_fqs 5678 and rcutree.jiffies_till_next_fqs. 5679 This calculated value may be viewed in 5680 rcutree.jiffies_to_sched_qs. Any attempt to set 5681 rcutree.jiffies_to_sched_qs will be cheerfully 5682 overwritten. 5683 5684 rcutree.kthread_prio= [KNL,BOOT] 5685 Set the SCHED_FIFO priority of the RCU per-CPU 5686 kthreads (rcuc/N). This value is also used for 5687 the priority of the RCU boost threads (rcub/N) 5688 and for the RCU grace-period kthreads (rcu_bh, 5689 rcu_preempt, and rcu_sched). If RCU_BOOST is 5690 set, valid values are 1-99 and the default is 1 5691 (the least-favored priority). Otherwise, when 5692 RCU_BOOST is not set, valid values are 0-99 and 5693 the default is zero (non-realtime operation). 5694 When RCU_NOCB_CPU is set, also adjust the 5695 priority of NOCB callback kthreads. 5696 5697 rcutree.nocb_nobypass_lim_per_jiffy= [KNL] 5698 On callback-offloaded (rcu_nocbs) CPUs, 5699 RCU reduces the lock contention that would 5700 otherwise be caused by callback floods through 5701 use of the ->nocb_bypass list. However, in the 5702 common non-flooded case, RCU queues directly to 5703 the main ->cblist in order to avoid the extra 5704 overhead of the ->nocb_bypass list and its lock. 5705 But if there are too many callbacks queued during 5706 a single jiffy, RCU pre-queues the callbacks into 5707 the ->nocb_bypass queue. The definition of "too 5708 many" is supplied by this kernel boot parameter. 5709 5710 rcutree.nohz_full_patience_delay= [KNL] 5711 On callback-offloaded (rcu_nocbs) CPUs, avoid 5712 disturbing RCU unless the grace period has 5713 reached the specified age in milliseconds. 5714 Defaults to zero. Large values will be capped 5715 at five seconds. All values will be rounded down 5716 to the nearest value representable by jiffies. 5717 5718 rcutree.qhimark= [KNL] 5719 Set threshold of queued RCU callbacks beyond which 5720 batch limiting is disabled. 5721 5722 rcutree.qlowmark= [KNL] 5723 Set threshold of queued RCU callbacks below which 5724 batch limiting is re-enabled. 5725 5726 rcutree.qovld= [KNL] 5727 Set threshold of queued RCU callbacks beyond which 5728 RCU's force-quiescent-state scan will aggressively 5729 enlist help from cond_resched() and sched IPIs to 5730 help CPUs more quickly reach quiescent states. 5731 Set to less than zero to make this be set based 5732 on rcutree.qhimark at boot time and to zero to 5733 disable more aggressive help enlistment. 5734 5735 rcutree.rcu_delay_page_cache_fill_msec= [KNL] 5736 Set the page-cache refill delay (in milliseconds) 5737 in response to low-memory conditions. The range 5738 of permitted values is in the range 0:100000. 5739 5740 rcutree.rcu_divisor= [KNL] 5741 Set the shift-right count to use to compute 5742 the callback-invocation batch limit bl from 5743 the number of callbacks queued on this CPU. 5744 The result will be bounded below by the value of 5745 the rcutree.blimit kernel parameter. Every bl 5746 callbacks, the softirq handler will exit in 5747 order to allow the CPU to do other work. 5748 5749 Please note that this callback-invocation batch 5750 limit applies only to non-offloaded callback 5751 invocation. Offloaded callbacks are instead 5752 invoked in the context of an rcuoc kthread, which 5753 scheduler will preempt as it does any other task. 5754 5755 rcutree.rcu_fanout_exact= [KNL] 5756 Disable autobalancing of the rcu_node combining 5757 tree. This is used by rcutorture, and might 5758 possibly be useful for architectures having high 5759 cache-to-cache transfer latencies. 5760 5761 rcutree.rcu_fanout_leaf= [KNL] 5762 Change the number of CPUs assigned to each 5763 leaf rcu_node structure. Useful for very 5764 large systems, which will choose the value 64, 5765 and for NUMA systems with large remote-access 5766 latencies, which will choose a value aligned 5767 with the appropriate hardware boundaries. 5768 5769 rcutree.rcu_min_cached_objs= [KNL] 5770 Minimum number of objects which are cached and 5771 maintained per one CPU. Object size is equal 5772 to PAGE_SIZE. The cache allows to reduce the 5773 pressure to page allocator, also it makes the 5774 whole algorithm to behave better in low memory 5775 condition. 5776 5777 rcutree.rcu_nocb_gp_stride= [KNL] 5778 Set the number of NOCB callback kthreads in 5779 each group, which defaults to the square root 5780 of the number of CPUs. Larger numbers reduce 5781 the wakeup overhead on the global grace-period 5782 kthread, but increases that same overhead on 5783 each group's NOCB grace-period kthread. 5784 5785 rcutree.rcu_kick_kthreads= [KNL] 5786 Cause the grace-period kthread to get an extra 5787 wake_up() if it sleeps three times longer than 5788 it should at force-quiescent-state time. 5789 This wake_up() will be accompanied by a 5790 WARN_ONCE() splat and an ftrace_dump(). 5791 5792 rcutree.rcu_resched_ns= [KNL] 5793 Limit the time spend invoking a batch of RCU 5794 callbacks to the specified number of nanoseconds. 5795 By default, this limit is checked only once 5796 every 32 callbacks in order to limit the pain 5797 inflicted by local_clock() overhead. 5798 5799 rcutree.rcu_unlock_delay= [KNL] 5800 In CONFIG_RCU_STRICT_GRACE_PERIOD=y kernels, 5801 this specifies an rcu_read_unlock()-time delay 5802 in microseconds. This defaults to zero. 5803 Larger delays increase the probability of 5804 catching RCU pointer leaks, that is, buggy use 5805 of RCU-protected pointers after the relevant 5806 rcu_read_unlock() has completed. 5807 5808 rcutree.sysrq_rcu= [KNL] 5809 Commandeer a sysrq key to dump out Tree RCU's 5810 rcu_node tree with an eye towards determining 5811 why a new grace period has not yet started. 5812 5813 rcutree.use_softirq= [KNL] 5814 If set to zero, move all RCU_SOFTIRQ processing to 5815 per-CPU rcuc kthreads. Defaults to a non-zero 5816 value, meaning that RCU_SOFTIRQ is used by default. 5817 Specify rcutree.use_softirq=0 to use rcuc kthreads. 5818 5819 But note that CONFIG_PREEMPT_RT=y kernels disable 5820 this kernel boot parameter, forcibly setting it 5821 to zero. 5822 5823 rcutree.enable_rcu_lazy= [KNL] 5824 To save power, batch RCU callbacks and flush after 5825 delay, memory pressure or callback list growing too 5826 big. 5827 5828 rcutree.rcu_normal_wake_from_gp= [KNL] 5829 Reduces a latency of synchronize_rcu() call. This approach 5830 maintains its own track of synchronize_rcu() callers, so it 5831 does not interact with regular callbacks because it does not 5832 use a call_rcu[_hurry]() path. Please note, this is for a 5833 normal grace period. 5834 5835 How to disable it: 5836 5837 echo 0 > /sys/module/rcutree/parameters/rcu_normal_wake_from_gp 5838 or pass a boot parameter "rcutree.rcu_normal_wake_from_gp=0" 5839 5840 Default is 1 if it is not explicitly disabled by the boot parameter 5841 passing 0. 5842 5843 rcuscale.gp_async= [KNL] 5844 Measure performance of asynchronous 5845 grace-period primitives such as call_rcu(). 5846 5847 rcuscale.gp_async_max= [KNL] 5848 Specify the maximum number of outstanding 5849 callbacks per writer thread. When a writer 5850 thread exceeds this limit, it invokes the 5851 corresponding flavor of rcu_barrier() to allow 5852 previously posted callbacks to drain. 5853 5854 rcuscale.gp_exp= [KNL] 5855 Measure performance of expedited synchronous 5856 grace-period primitives. 5857 5858 rcuscale.holdoff= [KNL] 5859 Set test-start holdoff period. The purpose of 5860 this parameter is to delay the start of the 5861 test until boot completes in order to avoid 5862 interference. 5863 5864 rcuscale.kfree_by_call_rcu= [KNL] 5865 In kernels built with CONFIG_RCU_LAZY=y, test 5866 call_rcu() instead of kfree_rcu(). 5867 5868 rcuscale.kfree_mult= [KNL] 5869 Instead of allocating an object of size kfree_obj, 5870 allocate one of kfree_mult * sizeof(kfree_obj). 5871 Defaults to 1. 5872 5873 rcuscale.kfree_rcu_test= [KNL] 5874 Set to measure performance of kfree_rcu() flooding. 5875 5876 rcuscale.kfree_rcu_test_double= [KNL] 5877 Test the double-argument variant of kfree_rcu(). 5878 If this parameter has the same value as 5879 rcuscale.kfree_rcu_test_single, both the single- 5880 and double-argument variants are tested. 5881 5882 rcuscale.kfree_rcu_test_single= [KNL] 5883 Test the single-argument variant of kfree_rcu(). 5884 If this parameter has the same value as 5885 rcuscale.kfree_rcu_test_double, both the single- 5886 and double-argument variants are tested. 5887 5888 rcuscale.kfree_nthreads= [KNL] 5889 The number of threads running loops of kfree_rcu(). 5890 5891 rcuscale.kfree_alloc_num= [KNL] 5892 Number of allocations and frees done in an iteration. 5893 5894 rcuscale.kfree_loops= [KNL] 5895 Number of loops doing rcuscale.kfree_alloc_num number 5896 of allocations and frees. 5897 5898 rcuscale.minruntime= [KNL] 5899 Set the minimum test run time in seconds. This 5900 does not affect the data-collection interval, 5901 but instead allows better measurement of things 5902 like CPU consumption. 5903 5904 rcuscale.nreaders= [KNL] 5905 Set number of RCU readers. The value -1 selects 5906 N, where N is the number of CPUs. A value 5907 "n" less than -1 selects N-n+1, where N is again 5908 the number of CPUs. For example, -2 selects N 5909 (the number of CPUs), -3 selects N+1, and so on. 5910 A value of "n" less than or equal to -N selects 5911 a single reader. 5912 5913 rcuscale.nwriters= [KNL] 5914 Set number of RCU writers. The values operate 5915 the same as for rcuscale.nreaders. 5916 N, where N is the number of CPUs 5917 5918 rcuscale.scale_type= [KNL] 5919 Specify the RCU implementation to test. 5920 5921 rcuscale.shutdown= [KNL] 5922 Shut the system down after performance tests 5923 complete. This is useful for hands-off automated 5924 testing. 5925 5926 rcuscale.verbose= [KNL] 5927 Enable additional printk() statements. 5928 5929 rcuscale.writer_holdoff= [KNL] 5930 Write-side holdoff between grace periods, 5931 in microseconds. The default of zero says 5932 no holdoff. 5933 5934 rcuscale.writer_holdoff_jiffies= [KNL] 5935 Additional write-side holdoff between grace 5936 periods, but in jiffies. The default of zero 5937 says no holdoff. 5938 5939 rcutorture.fqs_duration= [KNL] 5940 Set duration of force_quiescent_state bursts 5941 in microseconds. 5942 5943 rcutorture.fqs_holdoff= [KNL] 5944 Set holdoff time within force_quiescent_state bursts 5945 in microseconds. 5946 5947 rcutorture.fqs_stutter= [KNL] 5948 Set wait time between force_quiescent_state bursts 5949 in seconds. 5950 5951 rcutorture.fwd_progress= [KNL] 5952 Specifies the number of kthreads to be used 5953 for RCU grace-period forward-progress testing 5954 for the types of RCU supporting this notion. 5955 Defaults to 1 kthread, values less than zero or 5956 greater than the number of CPUs cause the number 5957 of CPUs to be used. 5958 5959 rcutorture.fwd_progress_div= [KNL] 5960 Specify the fraction of a CPU-stall-warning 5961 period to do tight-loop forward-progress testing. 5962 5963 rcutorture.fwd_progress_holdoff= [KNL] 5964 Number of seconds to wait between successive 5965 forward-progress tests. 5966 5967 rcutorture.fwd_progress_need_resched= [KNL] 5968 Enclose cond_resched() calls within checks for 5969 need_resched() during tight-loop forward-progress 5970 testing. 5971 5972 rcutorture.gp_cond= [KNL] 5973 Use conditional/asynchronous update-side 5974 normal-grace-period primitives, if available. 5975 5976 rcutorture.gp_cond_exp= [KNL] 5977 Use conditional/asynchronous update-side 5978 expedited-grace-period primitives, if available. 5979 5980 rcutorture.gp_cond_full= [KNL] 5981 Use conditional/asynchronous update-side 5982 normal-grace-period primitives that also take 5983 concurrent expedited grace periods into account, 5984 if available. 5985 5986 rcutorture.gp_cond_exp_full= [KNL] 5987 Use conditional/asynchronous update-side 5988 expedited-grace-period primitives that also take 5989 concurrent normal grace periods into account, 5990 if available. 5991 5992 rcutorture.gp_cond_wi= [KNL] 5993 Nominal wait interval for normal conditional 5994 grace periods (specified by rcutorture's 5995 gp_cond and gp_cond_full module parameters), 5996 in microseconds. The actual wait interval will 5997 be randomly selected to nanosecond granularity up 5998 to this wait interval. Defaults to 16 jiffies, 5999 for example, 16,000 microseconds on a system 6000 with HZ=1000. 6001 6002 rcutorture.gp_cond_wi_exp= [KNL] 6003 Nominal wait interval for expedited conditional 6004 grace periods (specified by rcutorture's 6005 gp_cond_exp and gp_cond_exp_full module 6006 parameters), in microseconds. The actual wait 6007 interval will be randomly selected to nanosecond 6008 granularity up to this wait interval. Defaults to 6009 128 microseconds. 6010 6011 rcutorture.gp_exp= [KNL] 6012 Use expedited update-side primitives, if available. 6013 6014 rcutorture.gp_normal= [KNL] 6015 Use normal (non-expedited) asynchronous 6016 update-side primitives, if available. 6017 6018 rcutorture.gp_poll= [KNL] 6019 Use polled update-side normal-grace-period 6020 primitives, if available. 6021 6022 rcutorture.gp_poll_exp= [KNL] 6023 Use polled update-side expedited-grace-period 6024 primitives, if available. 6025 6026 rcutorture.gp_poll_full= [KNL] 6027 Use polled update-side normal-grace-period 6028 primitives that also take concurrent expedited 6029 grace periods into account, if available. 6030 6031 rcutorture.gp_poll_exp_full= [KNL] 6032 Use polled update-side expedited-grace-period 6033 primitives that also take concurrent normal 6034 grace periods into account, if available. 6035 6036 rcutorture.gp_poll_wi= [KNL] 6037 Nominal wait interval for normal conditional 6038 grace periods (specified by rcutorture's 6039 gp_poll and gp_poll_full module parameters), 6040 in microseconds. The actual wait interval will 6041 be randomly selected to nanosecond granularity up 6042 to this wait interval. Defaults to 16 jiffies, 6043 for example, 16,000 microseconds on a system 6044 with HZ=1000. 6045 6046 rcutorture.gp_poll_wi_exp= [KNL] 6047 Nominal wait interval for expedited conditional 6048 grace periods (specified by rcutorture's 6049 gp_poll_exp and gp_poll_exp_full module 6050 parameters), in microseconds. The actual wait 6051 interval will be randomly selected to nanosecond 6052 granularity up to this wait interval. Defaults to 6053 128 microseconds. 6054 6055 rcutorture.gp_sync= [KNL] 6056 Use normal (non-expedited) synchronous 6057 update-side primitives, if available. If all 6058 of rcutorture.gp_cond=, rcutorture.gp_exp=, 6059 rcutorture.gp_normal=, and rcutorture.gp_sync= 6060 are zero, rcutorture acts as if is interpreted 6061 they are all non-zero. 6062 6063 rcutorture.gpwrap_lag= [KNL] 6064 Enable grace-period wrap lag testing. Setting 6065 to false prevents the gpwrap lag test from 6066 running. Default is true. 6067 6068 rcutorture.gpwrap_lag_gps= [KNL] 6069 Set the value for grace-period wrap lag during 6070 active lag testing periods. This controls how many 6071 grace periods differences we tolerate between 6072 rdp and rnp's gp_seq before setting overflow flag. 6073 The default is always set to 8. 6074 6075 rcutorture.gpwrap_lag_cycle_mins= [KNL] 6076 Set the total cycle duration for gpwrap lag 6077 testing in minutes. This is the total time for 6078 one complete cycle of active and inactive 6079 testing periods. Default is 30 minutes. 6080 6081 rcutorture.gpwrap_lag_active_mins= [KNL] 6082 Set the duration for which gpwrap lag is active 6083 within each cycle, in minutes. During this time, 6084 the grace-period wrap lag will be set to the 6085 value specified by gpwrap_lag_gps. Default is 6086 5 minutes. 6087 6088 rcutorture.irqreader= [KNL] 6089 Run RCU readers from irq handlers, or, more 6090 accurately, from a timer handler. Not all RCU 6091 flavors take kindly to this sort of thing. 6092 6093 rcutorture.leakpointer= [KNL] 6094 Leak an RCU-protected pointer out of the reader. 6095 This can of course result in splats, and is 6096 intended to test the ability of things like 6097 CONFIG_RCU_STRICT_GRACE_PERIOD=y to detect 6098 such leaks. 6099 6100 rcutorture.n_barrier_cbs= [KNL] 6101 Set callbacks/threads for rcu_barrier() testing. 6102 6103 rcutorture.nfakewriters= [KNL] 6104 Set number of concurrent RCU writers. These just 6105 stress RCU, they don't participate in the actual 6106 test, hence the "fake". 6107 6108 rcutorture.nocbs_nthreads= [KNL] 6109 Set number of RCU callback-offload togglers. 6110 Zero (the default) disables toggling. 6111 6112 rcutorture.nocbs_toggle= [KNL] 6113 Set the delay in milliseconds between successive 6114 callback-offload toggling attempts. 6115 6116 rcutorture.nreaders= [KNL] 6117 Set number of RCU readers. The value -1 selects 6118 N-1, where N is the number of CPUs. A value 6119 "n" less than -1 selects N-n-2, where N is again 6120 the number of CPUs. For example, -2 selects N 6121 (the number of CPUs), -3 selects N+1, and so on. 6122 6123 rcutorture.object_debug= [KNL] 6124 Enable debug-object double-call_rcu() testing. 6125 6126 rcutorture.onoff_holdoff= [KNL] 6127 Set time (s) after boot for CPU-hotplug testing. 6128 6129 rcutorture.onoff_interval= [KNL] 6130 Set time (jiffies) between CPU-hotplug operations, 6131 or zero to disable CPU-hotplug testing. 6132 6133 rcutorture.preempt_duration= [KNL] 6134 Set duration (in milliseconds) of preemptions 6135 by a high-priority FIFO real-time task. Set to 6136 zero (the default) to disable. The CPUs to 6137 preempt are selected randomly from the set that 6138 are online at a given point in time. Races with 6139 CPUs going offline are ignored, with that attempt 6140 at preemption skipped. 6141 6142 rcutorture.preempt_interval= [KNL] 6143 Set interval (in milliseconds, defaulting to one 6144 second) between preemptions by a high-priority 6145 FIFO real-time task. This delay is mediated 6146 by an hrtimer and is further fuzzed to avoid 6147 inadvertent synchronizations. 6148 6149 rcutorture.read_exit_burst= [KNL] 6150 The number of times in a given read-then-exit 6151 episode that a set of read-then-exit kthreads 6152 is spawned. 6153 6154 rcutorture.read_exit_delay= [KNL] 6155 The delay, in seconds, between successive 6156 read-then-exit testing episodes. 6157 6158 rcutorture.reader_flavor= [KNL] 6159 A bit mask indicating which readers to use. 6160 If there is more than one bit set, the readers 6161 are entered from low-order bit up, and are 6162 exited in the opposite order. For SRCU, the 6163 0x1 bit is normal readers, 0x2 NMI-safe readers, 6164 and 0x4 light-weight readers. 6165 6166 rcutorture.shuffle_interval= [KNL] 6167 Set task-shuffle interval (s). Shuffling tasks 6168 allows some CPUs to go into dyntick-idle mode 6169 during the rcutorture test. 6170 6171 rcutorture.shutdown_secs= [KNL] 6172 Set time (s) after boot system shutdown. This 6173 is useful for hands-off automated testing. 6174 6175 rcutorture.stall_cpu= [KNL] 6176 Duration of CPU stall (s) to test RCU CPU stall 6177 warnings, zero to disable. 6178 6179 rcutorture.stall_cpu_block= [KNL] 6180 Sleep while stalling if set. This will result 6181 in warnings from preemptible RCU in addition to 6182 any other stall-related activity. Note that 6183 in kernels built with CONFIG_PREEMPTION=n and 6184 CONFIG_PREEMPT_COUNT=y, this parameter will 6185 cause the CPU to pass through a quiescent state. 6186 Given CONFIG_PREEMPTION=n, this will suppress 6187 RCU CPU stall warnings, but will instead result 6188 in scheduling-while-atomic splats. 6189 6190 Use of this module parameter results in splats. 6191 6192 6193 rcutorture.stall_cpu_holdoff= [KNL] 6194 Time to wait (s) after boot before inducing stall. 6195 6196 rcutorture.stall_cpu_irqsoff= [KNL] 6197 Disable interrupts while stalling if set, but only 6198 on the first stall in the set. 6199 6200 rcutorture.stall_cpu_repeat= [KNL] 6201 Number of times to repeat the stall sequence, 6202 so that rcutorture.stall_cpu_repeat=3 will result 6203 in four stall sequences. 6204 6205 rcutorture.stall_gp_kthread= [KNL] 6206 Duration (s) of forced sleep within RCU 6207 grace-period kthread to test RCU CPU stall 6208 warnings, zero to disable. If both stall_cpu 6209 and stall_gp_kthread are specified, the 6210 kthread is starved first, then the CPU. 6211 6212 rcutorture.stat_interval= [KNL] 6213 Time (s) between statistics printk()s. 6214 6215 rcutorture.stutter= [KNL] 6216 Time (s) to stutter testing, for example, specifying 6217 five seconds causes the test to run for five seconds, 6218 wait for five seconds, and so on. This tests RCU's 6219 ability to transition abruptly to and from idle. 6220 6221 rcutorture.test_boost= [KNL] 6222 Test RCU priority boosting? 0=no, 1=maybe, 2=yes. 6223 "Maybe" means test if the RCU implementation 6224 under test support RCU priority boosting. 6225 6226 rcutorture.test_boost_duration= [KNL] 6227 Duration (s) of each individual boost test. 6228 6229 rcutorture.test_boost_holdoff= [KNL] 6230 Holdoff time (s) from start of test to the start 6231 of RCU priority-boost testing. Defaults to zero, 6232 that is, no holdoff. 6233 6234 rcutorture.test_boost_interval= [KNL] 6235 Interval (s) between each boost test. 6236 6237 rcutorture.test_no_idle_hz= [KNL] 6238 Test RCU's dyntick-idle handling. See also the 6239 rcutorture.shuffle_interval parameter. 6240 6241 rcutorture.torture_type= [KNL] 6242 Specify the RCU implementation to test. 6243 6244 rcutorture.verbose= [KNL] 6245 Enable additional printk() statements. 6246 6247 rcupdate.rcu_cpu_stall_ftrace_dump= [KNL] 6248 Dump ftrace buffer after reporting RCU CPU 6249 stall warning. 6250 6251 rcupdate.rcu_cpu_stall_notifiers= [KNL] 6252 Provide RCU CPU stall notifiers, but see the 6253 warnings in the RCU_CPU_STALL_NOTIFIER Kconfig 6254 option's help text. TL;DR: You almost certainly 6255 do not want rcupdate.rcu_cpu_stall_notifiers. 6256 6257 rcupdate.rcu_cpu_stall_suppress= [KNL] 6258 Suppress RCU CPU stall warning messages. 6259 6260 rcupdate.rcu_cpu_stall_suppress_at_boot= [KNL] 6261 Suppress RCU CPU stall warning messages and 6262 rcutorture writer stall warnings that occur 6263 during early boot, that is, during the time 6264 before the init task is spawned. 6265 6266 rcupdate.rcu_cpu_stall_timeout= [KNL] 6267 Set timeout for RCU CPU stall warning messages. 6268 The value is in seconds and the maximum allowed 6269 value is 300 seconds. 6270 6271 rcupdate.rcu_exp_cpu_stall_timeout= [KNL] 6272 Set timeout for expedited RCU CPU stall warning 6273 messages. The value is in milliseconds 6274 and the maximum allowed value is 21000 6275 milliseconds. Please note that this value is 6276 adjusted to an arch timer tick resolution. 6277 Setting this to zero causes the value from 6278 rcupdate.rcu_cpu_stall_timeout to be used (after 6279 conversion from seconds to milliseconds). 6280 6281 rcupdate.rcu_cpu_stall_cputime= [KNL] 6282 Provide statistics on the cputime and count of 6283 interrupts and tasks during the sampling period. For 6284 multiple continuous RCU stalls, all sampling periods 6285 begin at half of the first RCU stall timeout. 6286 6287 rcupdate.rcu_exp_stall_task_details= [KNL] 6288 Print stack dumps of any tasks blocking the 6289 current expedited RCU grace period during an 6290 expedited RCU CPU stall warning. 6291 6292 rcupdate.rcu_expedited= [KNL] 6293 Use expedited grace-period primitives, for 6294 example, synchronize_rcu_expedited() instead 6295 of synchronize_rcu(). This reduces latency, 6296 but can increase CPU utilization, degrade 6297 real-time latency, and degrade energy efficiency. 6298 No effect on CONFIG_TINY_RCU kernels. 6299 6300 rcupdate.rcu_normal= [KNL] 6301 Use only normal grace-period primitives, 6302 for example, synchronize_rcu() instead of 6303 synchronize_rcu_expedited(). This improves 6304 real-time latency, CPU utilization, and 6305 energy efficiency, but can expose users to 6306 increased grace-period latency. This parameter 6307 overrides rcupdate.rcu_expedited. No effect on 6308 CONFIG_TINY_RCU kernels. 6309 6310 rcupdate.rcu_normal_after_boot= [KNL] 6311 Once boot has completed (that is, after 6312 rcu_end_inkernel_boot() has been invoked), use 6313 only normal grace-period primitives. No effect 6314 on CONFIG_TINY_RCU kernels. 6315 6316 But note that CONFIG_PREEMPT_RT=y kernels enables 6317 this kernel boot parameter, forcibly setting 6318 it to the value one, that is, converting any 6319 post-boot attempt at an expedited RCU grace 6320 period to instead use normal non-expedited 6321 grace-period processing. 6322 6323 rcupdate.rcu_task_collapse_lim= [KNL] 6324 Set the maximum number of callbacks present 6325 at the beginning of a grace period that allows 6326 the RCU Tasks flavors to collapse back to using 6327 a single callback queue. This switching only 6328 occurs when rcupdate.rcu_task_enqueue_lim is 6329 set to the default value of -1. 6330 6331 rcupdate.rcu_task_contend_lim= [KNL] 6332 Set the minimum number of callback-queuing-time 6333 lock-contention events per jiffy required to 6334 cause the RCU Tasks flavors to switch to per-CPU 6335 callback queuing. This switching only occurs 6336 when rcupdate.rcu_task_enqueue_lim is set to 6337 the default value of -1. 6338 6339 rcupdate.rcu_task_enqueue_lim= [KNL] 6340 Set the number of callback queues to use for the 6341 RCU Tasks family of RCU flavors. The default 6342 of -1 allows this to be automatically (and 6343 dynamically) adjusted. This parameter is intended 6344 for use in testing. 6345 6346 rcupdate.rcu_task_lazy_lim= [KNL] 6347 Number of callbacks on a given CPU that will 6348 cancel laziness on that CPU. Use -1 to disable 6349 cancellation of laziness, but be advised that 6350 doing so increases the danger of OOM due to 6351 callback flooding. 6352 6353 rcupdate.rcu_task_stall_info= [KNL] 6354 Set initial timeout in jiffies for RCU task stall 6355 informational messages, which give some indication 6356 of the problem for those not patient enough to 6357 wait for ten minutes. Informational messages are 6358 only printed prior to the stall-warning message 6359 for a given grace period. Disable with a value 6360 less than or equal to zero. Defaults to ten 6361 seconds. A change in value does not take effect 6362 until the beginning of the next grace period. 6363 6364 rcupdate.rcu_task_stall_info_mult= [KNL] 6365 Multiplier for time interval between successive 6366 RCU task stall informational messages for a given 6367 RCU tasks grace period. This value is clamped 6368 to one through ten, inclusive. It defaults to 6369 the value three, so that the first informational 6370 message is printed 10 seconds into the grace 6371 period, the second at 40 seconds, the third at 6372 160 seconds, and then the stall warning at 600 6373 seconds would prevent a fourth at 640 seconds. 6374 6375 rcupdate.rcu_task_stall_timeout= [KNL] 6376 Set timeout in jiffies for RCU task stall 6377 warning messages. Disable with a value less 6378 than or equal to zero. Defaults to ten minutes. 6379 A change in value does not take effect until 6380 the beginning of the next grace period. 6381 6382 rcupdate.rcu_tasks_lazy_ms= [KNL] 6383 Set timeout in milliseconds RCU Tasks asynchronous 6384 callback batching for call_rcu_tasks(). 6385 A negative value will take the default. A value 6386 of zero will disable batching. Batching is 6387 always disabled for synchronize_rcu_tasks(). 6388 6389 rcupdate.rcu_self_test= [KNL] 6390 Run the RCU early boot self tests 6391 6392 rdinit= [KNL] 6393 Format: <full_path> 6394 Run specified binary instead of /init from the ramdisk, 6395 used for early userspace startup. See initrd. 6396 6397 rdrand= [X86,EARLY] 6398 force - Override the decision by the kernel to hide the 6399 advertisement of RDRAND support (this affects 6400 certain AMD processors because of buggy BIOS 6401 support, specifically around the suspend/resume 6402 path). 6403 6404 rdt= [HW,X86,RDT] 6405 Turn on/off individual RDT features. List is: 6406 cmt, mbmtotal, mbmlocal, l3cat, l3cdp, l2cat, l2cdp, 6407 mba, smba, bmec, abmc, sdciae, energy[:guid], 6408 perf[:guid]. 6409 E.g. to turn on cmt and turn off mba use: 6410 rdt=cmt,!mba 6411 To turn off all energy telemetry monitoring and ensure that 6412 perf telemetry monitoring associated with guid 0x12345 6413 is enabled use: 6414 rdt=!energy,perf:0x12345 6415 6416 reboot= [KNL] 6417 Format (x86 or x86_64): 6418 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] | d[efault] \ 6419 [[,]s[mp]#### \ 6420 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \ 6421 [[,]f[orce] 6422 Where reboot_mode is one of warm (soft) or cold (hard) or gpio 6423 (prefix with 'panic_' to set mode for panic 6424 reboot only), 6425 reboot_type is one of bios, acpi, kbd, triple, efi, or pci, 6426 reboot_force is either force or not specified, 6427 reboot_cpu is s[mp]#### with #### being the processor 6428 to be used for rebooting. 6429 6430 acpi 6431 Use the ACPI RESET_REG in the FADT. If ACPI is not 6432 configured or the ACPI reset does not work, the reboot 6433 path attempts the reset using the keyboard controller. 6434 6435 bios 6436 Use the CPU reboot vector for warm reset 6437 6438 cold 6439 Set the cold reboot flag 6440 6441 default 6442 There are some built-in platform specific "quirks" 6443 - you may see: "reboot: <name> series board detected. 6444 Selecting <type> for reboots." In the case where you 6445 think the quirk is in error (e.g. you have newer BIOS, 6446 or newer board) using this option will ignore the 6447 built-in quirk table, and use the generic default 6448 reboot actions. 6449 6450 efi 6451 Use efi reset_system runtime service. If EFI is not 6452 configured or the EFI reset does not work, the reboot 6453 path attempts the reset using the keyboard controller. 6454 6455 force 6456 Don't stop other CPUs on reboot. This can make reboot 6457 more reliable in some cases. 6458 6459 kbd 6460 Use the keyboard controller. cold reset (default) 6461 6462 pci 6463 Use a write to the PCI config space register 0xcf9 to 6464 trigger reboot. 6465 6466 triple 6467 Force a triple fault (init) 6468 6469 warm 6470 Don't set the cold reboot flag 6471 6472 Using warm reset will be much faster especially on big 6473 memory systems because the BIOS will not go through 6474 the memory check. Disadvantage is that not all 6475 hardware will be completely reinitialized on reboot so 6476 there may be boot problems on some systems. 6477 6478 6479 refscale.holdoff= [KNL] 6480 Set test-start holdoff period. The purpose of 6481 this parameter is to delay the start of the 6482 test until boot completes in order to avoid 6483 interference. 6484 6485 refscale.lookup_instances= [KNL] 6486 Number of data elements to use for the forms of 6487 SLAB_TYPESAFE_BY_RCU testing. A negative number 6488 is negated and multiplied by nr_cpu_ids, while 6489 zero specifies nr_cpu_ids. 6490 6491 refscale.loops= [KNL] 6492 Set the number of loops over the synchronization 6493 primitive under test. Increasing this number 6494 reduces noise due to loop start/end overhead, 6495 but the default has already reduced the per-pass 6496 noise to a handful of picoseconds on ca. 2020 6497 x86 laptops. 6498 6499 refscale.nreaders= [KNL] 6500 Set number of readers. The default value of -1 6501 selects N, where N is roughly 75% of the number 6502 of CPUs. A value of zero is an interesting choice. 6503 6504 refscale.nruns= [KNL] 6505 Set number of runs, each of which is dumped onto 6506 the console log. 6507 6508 refscale.readdelay= [KNL] 6509 Set the read-side critical-section duration, 6510 measured in microseconds. 6511 6512 refscale.scale_type= [KNL] 6513 Specify the read-protection implementation to test. 6514 6515 refscale.shutdown= [KNL] 6516 Shut down the system at the end of the performance 6517 test. This defaults to 1 (shut it down) when 6518 refscale is built into the kernel and to 0 (leave 6519 it running) when refscale is built as a module. 6520 6521 refscale.verbose= [KNL] 6522 Enable additional printk() statements. 6523 6524 refscale.verbose_batched= [KNL] 6525 Batch the additional printk() statements. If zero 6526 (the default) or negative, print everything. Otherwise, 6527 print every Nth verbose statement, where N is the value 6528 specified. 6529 6530 regulator_ignore_unused 6531 [REGULATOR] 6532 Prevents regulator framework from disabling regulators 6533 that are unused, due no driver claiming them. This may 6534 be useful for debug and development, but should not be 6535 needed on a platform with proper driver support. 6536 6537 relax_domain_level= 6538 [KNL, SMP] Set scheduler's default relax_domain_level. 6539 See Documentation/admin-guide/cgroup-v1/cpusets.rst. 6540 6541 reserve= [KNL,BUGS] Force kernel to ignore I/O ports or memory 6542 Format: <base1>,<size1>[,<base2>,<size2>,...] 6543 Reserve I/O ports or memory so the kernel won't use 6544 them. If <base> is less than 0x10000, the region 6545 is assumed to be I/O ports; otherwise it is memory. 6546 6547 reserve_mem= [RAM] 6548 Format: nn[KMG]:<align>:<label> 6549 Reserve physical memory and label it with a name that 6550 other subsystems can use to access it. This is typically 6551 used for systems that do not wipe the RAM, and this command 6552 line will try to reserve the same physical memory on 6553 soft reboots. Note, it is not guaranteed to be the same 6554 location. For example, if anything about the system changes 6555 or if booting a different kernel. It can also fail if KASLR 6556 places the kernel at the location of where the RAM reservation 6557 was from a previous boot, the new reservation will be at a 6558 different location. 6559 Any subsystem using this feature must add a way to verify 6560 that the contents of the physical memory is from a previous 6561 boot, as there may be cases where the memory will not be 6562 located at the same location. 6563 6564 The format is size:align:label for example, to request 6565 12 megabytes of 4096 alignment for ramoops: 6566 6567 reserve_mem=12M:4096:oops ramoops.mem_name=oops 6568 6569 reservetop= [X86-32,EARLY] 6570 Format: nn[KMG] 6571 Reserves a hole at the top of the kernel virtual 6572 address space. 6573 6574 reset_devices [KNL] Force drivers to reset the underlying device 6575 during initialization. 6576 6577 resume= [SWSUSP] 6578 Specify the partition device for software suspend 6579 Format: 6580 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>} 6581 6582 resume_offset= [SWSUSP] 6583 Specify the offset from the beginning of the partition 6584 given by "resume=" at which the swap header is located, 6585 in <PAGE_SIZE> units (needed only for swap files). 6586 See Documentation/power/swsusp-and-swap-files.rst 6587 6588 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to 6589 read the resume files 6590 6591 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up. 6592 Useful for devices that are detected asynchronously 6593 (e.g. USB and MMC devices). 6594 6595 retain_initrd [RAM] Keep initrd memory after extraction. After boot, it will 6596 be accessible via /sys/firmware/initrd. 6597 6598 retbleed= [X86] Control mitigation of RETBleed (Arbitrary 6599 Speculative Code Execution with Return Instructions) 6600 vulnerability. 6601 6602 AMD-based UNRET and IBPB mitigations alone do not stop 6603 sibling threads from influencing the predictions of other 6604 sibling threads. For that reason, STIBP is used on pro- 6605 cessors that support it, and mitigate SMT on processors 6606 that don't. 6607 6608 off - no mitigation 6609 auto - automatically select a mitigation 6610 auto,nosmt - automatically select a mitigation, 6611 disabling SMT if necessary for 6612 the full mitigation (only on Zen1 6613 and older without STIBP). 6614 ibpb - On AMD, mitigate short speculation 6615 windows on basic block boundaries too. 6616 Safe, highest perf impact. It also 6617 enables STIBP if present. Not suitable 6618 on Intel. 6619 ibpb,nosmt - Like "ibpb" above but will disable SMT 6620 when STIBP is not available. This is 6621 the alternative for systems which do not 6622 have STIBP. 6623 unret - Force enable untrained return thunks, 6624 only effective on AMD f15h-f17h based 6625 systems. 6626 unret,nosmt - Like unret, but will disable SMT when STIBP 6627 is not available. This is the alternative for 6628 systems which do not have STIBP. 6629 6630 Selecting 'auto' will choose a mitigation method at run 6631 time according to the CPU. 6632 6633 Not specifying this option is equivalent to retbleed=auto. 6634 6635 rfkill.default_state= 6636 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm, 6637 etc. communication is blocked by default. 6638 1 Unblocked. 6639 6640 rfkill.master_switch_mode= 6641 0 The "airplane mode" button does nothing. 6642 1 The "airplane mode" button toggles between everything 6643 blocked and the previous configuration. 6644 2 The "airplane mode" button toggles between everything 6645 blocked and everything unblocked. 6646 6647 ring3mwait=disable 6648 [KNL] Disable ring 3 MONITOR/MWAIT feature on supported 6649 CPUs. 6650 6651 riscv_isa_fallback [RISCV,EARLY] 6652 When CONFIG_RISCV_ISA_FALLBACK is not enabled, permit 6653 falling back to detecting extension support by parsing 6654 "riscv,isa" property on devicetree systems when the 6655 replacement properties are not found. See the Kconfig 6656 entry for RISCV_ISA_FALLBACK. 6657 6658 riscv_nousercfi= 6659 all Disable user CFI ABI to userspace even if cpu extension 6660 are available. 6661 bcfi Disable user backward CFI ABI to userspace even if 6662 the shadow stack extension is available. 6663 fcfi Disable user forward CFI ABI to userspace even if the 6664 landing pad extension is available. 6665 6666 ro [KNL] Mount root device read-only on boot 6667 6668 rodata= [KNL,EARLY] 6669 on Mark read-only kernel memory as read-only (default). 6670 off Leave read-only kernel memory writable for debugging. 6671 noalias Mark read-only kernel memory as read-only but retain 6672 writable aliases in the direct map for regions outside 6673 of the kernel image. [arm64] 6674 6675 rockchip.usb_uart 6676 [EARLY] 6677 Enable the uart passthrough on the designated usb port 6678 on Rockchip SoCs. When active, the signals of the 6679 debug-uart get routed to the D+ and D- pins of the usb 6680 port and the regular usb controller gets disabled. 6681 6682 root= [KNL] Root filesystem 6683 Usually this is a block device specifier of some kind, 6684 see the early_lookup_bdev comment in 6685 block/early-lookup.c for details. 6686 Alternatively this can be "ram" for the legacy initial 6687 ramdisk, "nfs" and "cifs" for root on a network file 6688 system, or "mtd" and "ubi" for mounting from raw flash. 6689 6690 rootdelay= [KNL] Delay (in seconds) to pause before attempting to 6691 mount the root filesystem 6692 6693 rootflags= [KNL] Set root filesystem mount option string 6694 6695 rseq_slice_ext= [KNL] RSEQ based time slice extension 6696 Format: boolean 6697 Control enablement of RSEQ based time slice extension. 6698 Default is 'on'. 6699 6700 initramfs_options= [KNL] 6701 Specify mount options for the initramfs mount. 6702 6703 rootfstype= [KNL] Set root filesystem type 6704 6705 rootwait [KNL] Wait (indefinitely) for root device to show up. 6706 Useful for devices that are detected asynchronously 6707 (e.g. USB and MMC devices). 6708 6709 rootwait= [KNL] Maximum time (in seconds) to wait for root device 6710 to show up before attempting to mount the root 6711 filesystem. 6712 6713 rproc_mem=nn[KMG][@address] 6714 [KNL,ARM,CMA] Remoteproc physical memory block. 6715 Memory area to be used by remote processor image, 6716 managed by CMA. 6717 6718 rseq_debug= [KNL] Enable or disable restartable sequence 6719 debug mode. Defaults to CONFIG_RSEQ_DEBUG_DEFAULT_ENABLE. 6720 Format: <bool> 6721 6722 rt_group_sched= [KNL] Enable or disable SCHED_RR/FIFO group scheduling 6723 when CONFIG_RT_GROUP_SCHED=y. Defaults to 6724 !CONFIG_RT_GROUP_SCHED_DEFAULT_DISABLED. 6725 Format: <bool> 6726 6727 rw [KNL] Mount root device read-write on boot 6728 6729 S [KNL] Run init in single mode 6730 6731 s390_iommu= [HW,S390] 6732 Set s390 IOTLB flushing mode 6733 strict 6734 With strict flushing every unmap operation will result 6735 in an IOTLB flush. Default is lazy flushing before 6736 reuse, which is faster. Deprecated, equivalent to 6737 iommu.strict=1. 6738 6739 s390_iommu_aperture= [KNL,S390] 6740 Specifies the size of the per device DMA address space 6741 accessible through the DMA and IOMMU APIs as a decimal 6742 factor of the size of main memory. 6743 The default is 1 meaning that one can concurrently use 6744 as many DMA addresses as physical memory is installed, 6745 if supported by hardware, and thus map all of memory 6746 once. With a value of 2 one can map all of memory twice 6747 and so on. As a special case a factor of 0 imposes no 6748 restrictions other than those given by hardware at the 6749 cost of significant additional memory use for tables. 6750 6751 6752 sched_proxy_exec= [KNL] 6753 Enables or disables "proxy execution" style 6754 solution to mutex-based priority inversion. 6755 Format: <bool> 6756 6757 sched_verbose [KNL,EARLY] Enables verbose scheduler debug messages. 6758 6759 schedstats= [KNL,X86] Enable or disable scheduled statistics. 6760 Allowed values are enable and disable. This feature 6761 incurs a small amount of overhead in the scheduler 6762 but is useful for debugging and performance tuning. 6763 6764 sched_thermal_decay_shift= 6765 [Deprecated] 6766 [KNL, SMP] Set a decay shift for scheduler thermal 6767 pressure signal. Thermal pressure signal follows the 6768 default decay period of other scheduler pelt 6769 signals(usually 32 ms but configurable). Setting 6770 sched_thermal_decay_shift will left shift the decay 6771 period for the thermal pressure signal by the shift 6772 value. 6773 i.e. with the default pelt decay period of 32 ms 6774 sched_thermal_decay_shift thermal pressure decay pr 6775 1 64 ms 6776 2 128 ms 6777 and so on. 6778 Format: integer between 0 and 10 6779 Default is 0. 6780 6781 scftorture.holdoff= [KNL] 6782 Number of seconds to hold off before starting 6783 test. Defaults to zero for module insertion and 6784 to 10 seconds for built-in smp_call_function() 6785 tests. 6786 6787 scftorture.longwait= [KNL] 6788 Request ridiculously long waits randomly selected 6789 up to the chosen limit in seconds. Zero (the 6790 default) disables this feature. Please note 6791 that requesting even small non-zero numbers of 6792 seconds can result in RCU CPU stall warnings, 6793 softlockup complaints, and so on. 6794 6795 scftorture.nthreads= [KNL] 6796 Number of kthreads to spawn to invoke the 6797 smp_call_function() family of functions. 6798 The default of -1 specifies a number of kthreads 6799 equal to the number of CPUs. 6800 6801 scftorture.onoff_holdoff= [KNL] 6802 Number seconds to wait after the start of the 6803 test before initiating CPU-hotplug operations. 6804 6805 scftorture.onoff_interval= [KNL] 6806 Number seconds to wait between successive 6807 CPU-hotplug operations. Specifying zero (which 6808 is the default) disables CPU-hotplug operations. 6809 6810 scftorture.shutdown_secs= [KNL] 6811 The number of seconds following the start of the 6812 test after which to shut down the system. The 6813 default of zero avoids shutting down the system. 6814 Non-zero values are useful for automated tests. 6815 6816 scftorture.stat_interval= [KNL] 6817 The number of seconds between outputting the 6818 current test statistics to the console. A value 6819 of zero disables statistics output. 6820 6821 scftorture.stutter_cpus= [KNL] 6822 The number of jiffies to wait between each change 6823 to the set of CPUs under test. 6824 6825 scftorture.use_cpus_read_lock= [KNL] 6826 Use use_cpus_read_lock() instead of the default 6827 preempt_disable() to disable CPU hotplug 6828 while invoking one of the smp_call_function*() 6829 functions. 6830 6831 scftorture.verbose= [KNL] 6832 Enable additional printk() statements. 6833 6834 scftorture.weight_single= [KNL] 6835 The probability weighting to use for the 6836 smp_call_function_single() function with a zero 6837 "wait" parameter. A value of -1 selects the 6838 default if all other weights are -1. However, 6839 if at least one weight has some other value, a 6840 value of -1 will instead select a weight of zero. 6841 6842 scftorture.weight_single_wait= [KNL] 6843 The probability weighting to use for the 6844 smp_call_function_single() function with a 6845 non-zero "wait" parameter. See weight_single. 6846 6847 scftorture.weight_many= [KNL] 6848 The probability weighting to use for the 6849 smp_call_function_many() function with a zero 6850 "wait" parameter. See weight_single. 6851 Note well that setting a high probability for 6852 this weighting can place serious IPI load 6853 on the system. 6854 6855 scftorture.weight_many_wait= [KNL] 6856 The probability weighting to use for the 6857 smp_call_function_many() function with a 6858 non-zero "wait" parameter. See weight_single 6859 and weight_many. 6860 6861 scftorture.weight_all= [KNL] 6862 The probability weighting to use for the 6863 smp_call_function_all() function with a zero 6864 "wait" parameter. See weight_single and 6865 weight_many. 6866 6867 scftorture.weight_all_wait= [KNL] 6868 The probability weighting to use for the 6869 smp_call_function_all() function with a 6870 non-zero "wait" parameter. See weight_single 6871 and weight_many. 6872 6873 sdw_mclk_divider=[SDW] 6874 Specify the MCLK divider for Intel SoundWire buses in 6875 case the BIOS does not provide the clock rate properly. 6876 6877 skew_tick= [KNL,EARLY] Offset the periodic timer tick per cpu to mitigate 6878 xtime_lock contention on larger systems, and/or RCU lock 6879 contention on all systems with CONFIG_MAXSMP set. 6880 Format: { "0" | "1" } 6881 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1" 6882 1 -- enable. 6883 Note: increases power consumption, thus should only be 6884 enabled if running jitter sensitive (HPC/RT) workloads. 6885 6886 security= [SECURITY] Choose a legacy "major" security module to 6887 enable at boot. This has been deprecated by the 6888 "lsm=" parameter. 6889 6890 selinux= [SELINUX] Disable or enable SELinux at boot time. 6891 Format: { "0" | "1" } 6892 See security/selinux/Kconfig help text. 6893 0 -- disable. 6894 1 -- enable. 6895 Default value is 1. 6896 6897 serialnumber [BUGS=X86-32] 6898 6899 sev=option[,option...] [X86-64] 6900 6901 debug 6902 Enable debug messages. 6903 6904 nosnp 6905 Do not enable SEV-SNP (applies to host/hypervisor 6906 only). Setting 'nosnp' avoids the RMP check overhead 6907 in memory accesses when users do not want to run 6908 SEV-SNP guests. 6909 6910 shapers= [NET] 6911 Maximal number of shapers. 6912 6913 show_lapic= [APIC,X86] Advanced Programmable Interrupt Controller 6914 Limit apic dumping. The parameter defines the maximal 6915 number of local apics being dumped. Also it is possible 6916 to set it to "all" by meaning -- no limit here. 6917 Format: { 1 (default) | 2 | ... | all }. 6918 The parameter valid if only apic=debug or 6919 apic=verbose is specified. 6920 Example: apic=debug show_lapic=all 6921 6922 slab_debug[=options[,slabs][;[options[,slabs]]...] [MM] 6923 Enabling slab_debug allows one to determine the 6924 culprit if slab objects become corrupted. Enabling 6925 slab_debug can create guard zones around objects and 6926 may poison objects when not in use. Also tracks the 6927 last alloc / free. For more information see 6928 Documentation/admin-guide/mm/slab.rst. 6929 (slub_debug legacy name also accepted for now) 6930 6931 Using this option implies the "no_hash_pointers" 6932 option which can be undone by adding the 6933 "hash_pointers=always" option. 6934 6935 slab_max_order= [MM] 6936 Determines the maximum allowed order for slabs. 6937 A high setting may cause OOMs due to memory 6938 fragmentation. For more information see 6939 Documentation/admin-guide/mm/slab.rst. 6940 (slub_max_order legacy name also accepted for now) 6941 6942 slab_merge [MM] 6943 Enable merging of slabs with similar size when the 6944 kernel is built without CONFIG_SLAB_MERGE_DEFAULT. 6945 (slub_merge legacy name also accepted for now) 6946 6947 slab_min_objects= [MM] 6948 The minimum number of objects per slab. SLUB will 6949 increase the slab order up to slab_max_order to 6950 generate a sufficiently large slab able to contain 6951 the number of objects indicated. The higher the number 6952 of objects the smaller the overhead of tracking slabs 6953 and the less frequently locks need to be acquired. 6954 For more information see 6955 Documentation/admin-guide/mm/slab.rst. 6956 (slub_min_objects legacy name also accepted for now) 6957 6958 slab_min_order= [MM] 6959 Determines the minimum page order for slabs. Must be 6960 lower or equal to slab_max_order. For more information see 6961 Documentation/admin-guide/mm/slab.rst. 6962 (slub_min_order legacy name also accepted for now) 6963 6964 slab_nomerge [MM] 6965 Disable merging of slabs with similar size. May be 6966 necessary if there is some reason to distinguish 6967 allocs to different slabs, especially in hardened 6968 environments where the risk of heap overflows and 6969 layout control by attackers can usually be 6970 frustrated by disabling merging. This will reduce 6971 most of the exposure of a heap attack to a single 6972 cache (risks via metadata attacks are mostly 6973 unchanged). Debug options disable merging on their 6974 own. 6975 For more information see 6976 Documentation/admin-guide/mm/slab.rst. 6977 (slub_nomerge legacy name also accepted for now) 6978 6979 slab_strict_numa [MM] 6980 Support memory policies on a per object level 6981 in the slab allocator. The default is for memory 6982 policies to be applied at the folio level when 6983 a new folio is needed or a partial folio is 6984 retrieved from the lists. Increases overhead 6985 in the slab fastpaths but gains more accurate 6986 NUMA kernel object placement which helps with slow 6987 interconnects in NUMA systems. 6988 6989 slram= [HW,MTD] 6990 6991 smart2= [HW] 6992 Format: <io1>[,<io2>[,...,<io8>]] 6993 6994 smp.csd_lock_timeout= [KNL] 6995 Specify the period of time in milliseconds 6996 that smp_call_function() and friends will wait 6997 for a CPU to release the CSD lock. This is 6998 useful when diagnosing bugs involving CPUs 6999 disabling interrupts for extended periods 7000 of time. Defaults to 5,000 milliseconds, and 7001 setting a value of zero disables this feature. 7002 This feature may be more efficiently disabled 7003 using the csdlock_debug- kernel parameter. 7004 7005 smp.panic_on_ipistall= [KNL] 7006 If a csd_lock_timeout extends for more than 7007 the specified number of milliseconds, panic the 7008 system. By default, let CSD-lock acquisition 7009 take as long as they take. Specifying 300,000 7010 for this value provides a 5-minute timeout. 7011 7012 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices 7013 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port 7014 smsc-ircc2.ircc_sir= [HW] SIR base I/O port 7015 smsc-ircc2.ircc_fir= [HW] FIR base I/O port 7016 smsc-ircc2.ircc_irq= [HW] IRQ line 7017 smsc-ircc2.ircc_dma= [HW] DMA channel 7018 smsc-ircc2.ircc_transceiver= [HW] Transceiver type: 7019 0: Toshiba Satellite 1800 (GP data pin select) 7020 1: Fast pin select (default) 7021 2: ATC IRMode 7022 7023 smt= [KNL,MIPS,S390,EARLY] Set the maximum number of threads 7024 (logical CPUs) to use per physical CPU on systems 7025 capable of symmetric multithreading (SMT). Will 7026 be capped to the actual hardware limit. 7027 Format: <integer> 7028 Default: -1 (no limit) 7029 7030 softlockup_panic= 7031 [KNL] Should the soft-lockup detector generate panics. 7032 Format: <int> 7033 7034 A value of non-zero instructs the soft-lockup detector 7035 to panic the machine when a soft-lockup duration exceeds 7036 N thresholds. It is also controlled by the kernel.softlockup_panic 7037 sysctl and CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC, which is the 7038 respective build-time switch to that functionality. 7039 7040 softlockup_all_cpu_backtrace= 7041 [KNL] Should the soft-lockup detector generate 7042 backtraces on all cpus. 7043 Format: 0 | 1 7044 7045 sonypi.*= [HW] Sony Programmable I/O Control Device driver 7046 See Documentation/admin-guide/laptops/sonypi.rst 7047 7048 spectre_bhi= [X86] Control mitigation of Branch History Injection 7049 (BHI) vulnerability. This setting affects the 7050 deployment of the HW BHI control and the SW BHB 7051 clearing sequence. 7052 7053 on - (default) Enable the HW or SW mitigation as 7054 needed. This protects the kernel from 7055 both syscalls and VMs. 7056 vmexit - On systems which don't have the HW mitigation 7057 available, enable the SW mitigation on vmexit 7058 ONLY. On such systems, the host kernel is 7059 protected from VM-originated BHI attacks, but 7060 may still be vulnerable to syscall attacks. 7061 off - Disable the mitigation. 7062 7063 spectre_v2= [X86,EARLY] Control mitigation of Spectre variant 2 7064 (indirect branch speculation) vulnerability. 7065 The default operation protects the kernel from 7066 user space attacks. 7067 7068 on - unconditionally enable, implies 7069 spectre_v2_user=on 7070 off - unconditionally disable, implies 7071 spectre_v2_user=off 7072 auto - kernel detects whether your CPU model is 7073 vulnerable 7074 7075 Selecting 'on' will, and 'auto' may, choose a 7076 mitigation method at run time according to the 7077 CPU, the available microcode, the setting of the 7078 CONFIG_MITIGATION_RETPOLINE configuration option, 7079 and the compiler with which the kernel was built. 7080 7081 Selecting 'on' will also enable the mitigation 7082 against user space to user space task attacks. 7083 Selecting specific mitigation does not force enable 7084 user mitigations. 7085 7086 Selecting 'off' will disable both the kernel and 7087 the user space protections. 7088 7089 Specific mitigations can also be selected manually: 7090 7091 retpoline - replace indirect branches 7092 retpoline,generic - Retpolines 7093 retpoline,lfence - LFENCE; indirect branch 7094 retpoline,amd - alias for retpoline,lfence 7095 eibrs - Enhanced/Auto IBRS 7096 eibrs,retpoline - Enhanced/Auto IBRS + Retpolines 7097 eibrs,lfence - Enhanced/Auto IBRS + LFENCE 7098 ibrs - use IBRS to protect kernel 7099 7100 Not specifying this option is equivalent to 7101 spectre_v2=auto. 7102 7103 spectre_v2_user= 7104 [X86] Control mitigation of Spectre variant 2 7105 (indirect branch speculation) vulnerability between 7106 user space tasks 7107 7108 on - Unconditionally enable mitigations. Is 7109 enforced by spectre_v2=on 7110 7111 off - Unconditionally disable mitigations. Is 7112 enforced by spectre_v2=off 7113 7114 prctl - Indirect branch speculation is enabled, 7115 but mitigation can be enabled via prctl 7116 per thread. The mitigation control state 7117 is inherited on fork. 7118 7119 prctl,ibpb 7120 - Like "prctl" above, but only STIBP is 7121 controlled per thread. IBPB is issued 7122 always when switching between different user 7123 space processes. 7124 7125 seccomp 7126 - Same as "prctl" above, but all seccomp 7127 threads will enable the mitigation unless 7128 they explicitly opt out. 7129 7130 seccomp,ibpb 7131 - Like "seccomp" above, but only STIBP is 7132 controlled per thread. IBPB is issued 7133 always when switching between different 7134 user space processes. 7135 7136 auto - Kernel selects the mitigation depending on 7137 the available CPU features and vulnerability. 7138 7139 Default mitigation: "prctl" 7140 7141 Not specifying this option is equivalent to 7142 spectre_v2_user=auto. 7143 7144 spec_rstack_overflow= 7145 [X86,EARLY] Control RAS overflow mitigation on AMD Zen CPUs 7146 7147 off - Disable mitigation 7148 microcode - Enable microcode mitigation only 7149 safe-ret - Enable sw-only safe RET mitigation (default) 7150 ibpb - Enable mitigation by issuing IBPB on 7151 kernel entry 7152 ibpb-vmexit - Issue IBPB only on VMEXIT 7153 (cloud-specific mitigation) 7154 7155 spec_store_bypass_disable= 7156 [HW,EARLY] Control Speculative Store Bypass (SSB) Disable mitigation 7157 (Speculative Store Bypass vulnerability) 7158 7159 Certain CPUs are vulnerable to an exploit against a 7160 a common industry wide performance optimization known 7161 as "Speculative Store Bypass" in which recent stores 7162 to the same memory location may not be observed by 7163 later loads during speculative execution. The idea 7164 is that such stores are unlikely and that they can 7165 be detected prior to instruction retirement at the 7166 end of a particular speculation execution window. 7167 7168 In vulnerable processors, the speculatively forwarded 7169 store can be used in a cache side channel attack, for 7170 example to read memory to which the attacker does not 7171 directly have access (e.g. inside sandboxed code). 7172 7173 This parameter controls whether the Speculative Store 7174 Bypass optimization is used. 7175 7176 On x86 the options are: 7177 7178 on - Unconditionally disable Speculative Store Bypass 7179 off - Unconditionally enable Speculative Store Bypass 7180 auto - Kernel detects whether the CPU model contains an 7181 implementation of Speculative Store Bypass and 7182 picks the most appropriate mitigation. If the 7183 CPU is not vulnerable, "off" is selected. If the 7184 CPU is vulnerable the default mitigation is 7185 architecture and Kconfig dependent. See below. 7186 prctl - Control Speculative Store Bypass per thread 7187 via prctl. Speculative Store Bypass is enabled 7188 for a process by default. The state of the control 7189 is inherited on fork. 7190 seccomp - Same as "prctl" above, but all seccomp threads 7191 will disable SSB unless they explicitly opt out. 7192 7193 Default mitigations: 7194 X86: "prctl" 7195 7196 On powerpc the options are: 7197 7198 on,auto - On Power8 and Power9 insert a store-forwarding 7199 barrier on kernel entry and exit. On Power7 7200 perform a software flush on kernel entry and 7201 exit. 7202 off - No action. 7203 7204 Not specifying this option is equivalent to 7205 spec_store_bypass_disable=auto. 7206 7207 split_llc= 7208 [X86,EARLY] Split the LLC N-ways 7209 7210 When set, the LLC is split this many ways by matching 7211 'core_id % n'. This is setup before SMP bringup and 7212 used during SMP bringup before it knows the full 7213 topology. If your core count doesn't nicely divide by 7214 the number given, you get to keep the pieces. 7215 7216 This is mostly a debug feature to emulate multiple LLCs 7217 on hardware that only have a single LLC. 7218 7219 split_lock_detect= 7220 [X86] Enable split lock detection or bus lock detection 7221 7222 When enabled (and if hardware support is present), atomic 7223 instructions that access data across cache line 7224 boundaries will result in an alignment check exception 7225 for split lock detection or a debug exception for 7226 bus lock detection. 7227 7228 off - not enabled 7229 7230 warn - the kernel will emit rate-limited warnings 7231 about applications triggering the #AC 7232 exception or the #DB exception. This mode is 7233 the default on CPUs that support split lock 7234 detection or bus lock detection. Default 7235 behavior is by #AC if both features are 7236 enabled in hardware. 7237 7238 fatal - the kernel will send SIGBUS to applications 7239 that trigger the #AC exception or the #DB 7240 exception. Default behavior is by #AC if 7241 both features are enabled in hardware. 7242 7243 ratelimit:N - 7244 Set system wide rate limit to N bus locks 7245 per second for bus lock detection. 7246 0 < N <= 1000. 7247 7248 N/A for split lock detection. 7249 7250 7251 If an #AC exception is hit in the kernel or in 7252 firmware (i.e. not while executing in user mode) 7253 the kernel will oops in either "warn" or "fatal" 7254 mode. 7255 7256 #DB exception for bus lock is triggered only when 7257 CPL > 0. 7258 7259 srbds= [X86,INTEL,EARLY] 7260 Control the Special Register Buffer Data Sampling 7261 (SRBDS) mitigation. 7262 7263 Certain CPUs are vulnerable to an MDS-like 7264 exploit which can leak bits from the random 7265 number generator. 7266 7267 By default, this issue is mitigated by 7268 microcode. However, the microcode fix can cause 7269 the RDRAND and RDSEED instructions to become 7270 much slower. Among other effects, this will 7271 result in reduced throughput from /dev/urandom. 7272 7273 The microcode mitigation can be disabled with 7274 the following option: 7275 7276 off: Disable mitigation and remove 7277 performance impact to RDRAND and RDSEED 7278 7279 srcutree.big_cpu_lim [KNL] 7280 Specifies the number of CPUs constituting a 7281 large system, such that srcu_struct structures 7282 should immediately allocate an srcu_node array. 7283 This kernel-boot parameter defaults to 128, 7284 but takes effect only when the low-order four 7285 bits of srcutree.convert_to_big is equal to 3 7286 (decide at boot). 7287 7288 srcutree.convert_to_big [KNL] 7289 Specifies under what conditions an SRCU tree 7290 srcu_struct structure will be converted to big 7291 form, that is, with an rcu_node tree: 7292 7293 0: Never. 7294 1: At init_srcu_struct() time. 7295 2: When rcutorture decides to. 7296 3: Decide at boot time (default). 7297 0x1X: Above plus if high contention. 7298 7299 Either way, the srcu_node tree will be sized based 7300 on the actual runtime number of CPUs (nr_cpu_ids) 7301 instead of the compile-time CONFIG_NR_CPUS. 7302 7303 srcutree.counter_wrap_check [KNL] 7304 Specifies how frequently to check for 7305 grace-period sequence counter wrap for the 7306 srcu_data structure's ->srcu_gp_seq_needed field. 7307 The greater the number of bits set in this kernel 7308 parameter, the less frequently counter wrap will 7309 be checked for. Note that the bottom two bits 7310 are ignored. 7311 7312 srcutree.exp_holdoff [KNL] 7313 Specifies how many nanoseconds must elapse 7314 since the end of the last SRCU grace period for 7315 a given srcu_struct until the next normal SRCU 7316 grace period will be considered for automatic 7317 expediting. Set to zero to disable automatic 7318 expediting. 7319 7320 srcutree.srcu_max_nodelay [KNL] 7321 Specifies the number of no-delay instances 7322 per jiffy for which the SRCU grace period 7323 worker thread will be rescheduled with zero 7324 delay. Beyond this limit, worker thread will 7325 be rescheduled with a sleep delay of one jiffy. 7326 7327 srcutree.srcu_max_nodelay_phase [KNL] 7328 Specifies the per-grace-period phase, number of 7329 non-sleeping polls of readers. Beyond this limit, 7330 grace period worker thread will be rescheduled 7331 with a sleep delay of one jiffy, between each 7332 rescan of the readers, for a grace period phase. 7333 7334 srcutree.srcu_retry_check_delay [KNL] 7335 Specifies number of microseconds of non-sleeping 7336 delay between each non-sleeping poll of readers. 7337 7338 srcutree.small_contention_lim [KNL] 7339 Specifies the number of update-side contention 7340 events per jiffy will be tolerated before 7341 initiating a conversion of an srcu_struct 7342 structure to big form. Note that the value of 7343 srcutree.convert_to_big must have the 0x10 bit 7344 set for contention-based conversions to occur. 7345 7346 ssbd= [ARM64,HW,EARLY] 7347 Speculative Store Bypass Disable control 7348 7349 On CPUs that are vulnerable to the Speculative 7350 Store Bypass vulnerability and offer a 7351 firmware based mitigation, this parameter 7352 indicates how the mitigation should be used: 7353 7354 force-on: Unconditionally enable mitigation for 7355 for both kernel and userspace 7356 force-off: Unconditionally disable mitigation for 7357 for both kernel and userspace 7358 kernel: Always enable mitigation in the 7359 kernel, and offer a prctl interface 7360 to allow userspace to register its 7361 interest in being mitigated too. 7362 7363 stack_guard_gap= [MM] 7364 override the default stack gap protection. The value 7365 is in page units and it defines how many pages prior 7366 to (for stacks growing down) resp. after (for stacks 7367 growing up) the main stack are reserved for no other 7368 mapping. Default value is 256 pages. 7369 7370 stack_depot_disable= [KNL,EARLY] 7371 Setting this to true through kernel command line will 7372 disable the stack depot thereby saving the static memory 7373 consumed by the stack hash table. By default this is set 7374 to false. 7375 7376 stack_depot_max_pools= [KNL,EARLY] 7377 Specify the maximum number of pools to use for storing 7378 stack traces. Pools are allocated on-demand up to this 7379 limit. Default value is 8191 pools. 7380 7381 stacktrace [FTRACE] 7382 Enable the stack tracer on boot up. 7383 7384 stacktrace_filter=[function-list] 7385 [FTRACE] Limit the functions that the stack tracer 7386 will trace at boot up. function-list is a comma-separated 7387 list of functions. This list can be changed at run 7388 time by the stack_trace_filter file in the debugfs 7389 tracing directory. Note, this enables stack tracing 7390 and the stacktrace above is not needed. 7391 7392 sti= [PARISC,HW] 7393 Format: <num> 7394 Set the STI (builtin display/keyboard on the HP-PARISC 7395 machines) console (graphic card) which should be used 7396 as the initial boot-console. 7397 See also comment in drivers/video/sticore.c. 7398 7399 sti_font= [HW] 7400 See comment in drivers/video/sticore.c. 7401 7402 stifb= [HW] 7403 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]] 7404 7405 strict_sas_size= 7406 [X86] 7407 Format: <bool> 7408 Enable or disable strict sigaltstack size checks 7409 against the required signal frame size which 7410 depends on the supported FPU features. This can 7411 be used to filter out binaries which have 7412 not yet been made aware of AT_MINSIGSTKSZ. 7413 7414 stress_hpt [PPC,EARLY] 7415 Limits the number of kernel HPT entries in the hash 7416 page table to increase the rate of hash page table 7417 faults on kernel addresses. 7418 7419 stress_slb [PPC,EARLY] 7420 Limits the number of kernel SLB entries, and flushes 7421 them frequently to increase the rate of SLB faults 7422 on kernel addresses. 7423 7424 no_slb_preload [PPC,EARLY] 7425 Disables slb preloading for userspace. 7426 7427 sunrpc.min_resvport= 7428 sunrpc.max_resvport= 7429 [NFS,SUNRPC] 7430 SunRPC servers often require that client requests 7431 originate from a privileged port (i.e. a port in the 7432 range 0 < portnr < 1024). 7433 An administrator who wishes to reserve some of these 7434 ports for other uses may adjust the range that the 7435 kernel's sunrpc client considers to be privileged 7436 using these two parameters to set the minimum and 7437 maximum port values. 7438 7439 sunrpc.svc_rpc_per_connection_limit= 7440 [NFS,SUNRPC] 7441 Limit the number of requests that the server will 7442 process in parallel from a single connection. 7443 The default value is 0 (no limit). 7444 7445 sunrpc.pool_mode= 7446 [NFS] 7447 Control how the NFS server code allocates CPUs to 7448 service thread pools. Depending on how many NICs 7449 you have and where their interrupts are bound, this 7450 option will affect which CPUs will do NFS serving. 7451 Note: this parameter cannot be changed while the 7452 NFS server is running. 7453 7454 auto the server chooses an appropriate mode 7455 automatically using heuristics 7456 global a single global pool contains all CPUs 7457 percpu one pool for each CPU 7458 pernode one pool for each NUMA node (equivalent 7459 to global on non-NUMA machines) 7460 7461 sunrpc.tcp_slot_table_entries= 7462 sunrpc.udp_slot_table_entries= 7463 [NFS,SUNRPC] 7464 Sets the upper limit on the number of simultaneous 7465 RPC calls that can be sent from the client to a 7466 server. Increasing these values may allow you to 7467 improve throughput, but will also increase the 7468 amount of memory reserved for use by the client. 7469 7470 suspend.pm_test_delay= 7471 [SUSPEND] 7472 Sets the number of seconds to remain in a suspend test 7473 mode before resuming the system (see 7474 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG 7475 is set. Default value is 5. 7476 7477 svm= [PPC] 7478 Format: { on | off | y | n | 1 | 0 } 7479 This parameter controls use of the Protected 7480 Execution Facility on pSeries. 7481 7482 swiotlb= [ARM,PPC,MIPS,X86,S390,EARLY] 7483 Format: { <int> [,<int>] | force | noforce } 7484 <int> -- Number of I/O TLB slabs 7485 <int> -- Second integer after comma. Number of swiotlb 7486 areas with their own lock. Will be rounded up 7487 to a power of 2. 7488 force -- force using of bounce buffers even if they 7489 wouldn't be automatically used by the kernel 7490 noforce -- Never use bounce buffers (for debugging) 7491 7492 switches= [HW,M68k,EARLY] 7493 7494 sysctl.*= [KNL] 7495 Set a sysctl parameter, right before loading the init 7496 process, as if the value was written to the respective 7497 /proc/sys/... file. Both '.' and '/' are recognized as 7498 separators. Unrecognized parameters and invalid values 7499 are reported in the kernel log. Sysctls registered 7500 later by a loaded module cannot be set this way. 7501 Example: sysctl.vm.swappiness=40 7502 7503 sysrq_always_enabled 7504 [KNL] 7505 Ignore sysrq setting - this boot parameter will 7506 neutralize any effect of /proc/sys/kernel/sysrq. 7507 Useful for debugging. 7508 7509 tcpmhash_entries= [KNL,NET] 7510 Set the number of tcp_metrics_hash slots. 7511 Default value is 8192 or 16384 depending on total 7512 ram pages. This is used to specify the TCP metrics 7513 cache size. See Documentation/networking/ip-sysctl.rst 7514 "tcp_no_metrics_save" section for more details. 7515 7516 tdfx= [HW,DRM] 7517 7518 test_suspend= [SUSPEND] 7519 Format: { "mem" | "standby" | "freeze" }[,N] 7520 Specify "mem" (for Suspend-to-RAM) or "standby" (for 7521 standby suspend) or "freeze" (for suspend type freeze) 7522 as the system sleep state during system startup with 7523 the optional capability to repeat N number of times. 7524 The system is woken from this state using a 7525 wakeup-capable RTC alarm. 7526 7527 thash_entries= [KNL,NET] 7528 Set number of hash buckets for TCP connection 7529 7530 thermal.act= [HW,ACPI] 7531 -1: disable all active trip points in all thermal zones 7532 <degrees C>: override all lowest active trip points 7533 7534 thermal.crt= [HW,ACPI] 7535 -1: disable all critical trip points in all thermal zones 7536 <degrees C>: override all critical trip points 7537 7538 thermal.off= [HW,ACPI] 7539 1: disable ACPI thermal control 7540 7541 thermal.psv= [HW,ACPI] 7542 -1: disable all passive trip points 7543 <degrees C>: override all passive trip points to this 7544 value 7545 7546 thermal.tzp= [HW,ACPI] 7547 Specify global default ACPI thermal zone polling rate 7548 <deci-seconds>: poll all this frequency 7549 0: no polling (default) 7550 7551 thp_anon= [KNL] 7552 Format: <size>[KMG],<size>[KMG]:<state>;<size>[KMG]-<size>[KMG]:<state> 7553 state is one of "always", "madvise", "never" or "inherit". 7554 Control the default behavior of the system with respect 7555 to anonymous transparent hugepages. 7556 Can be used multiple times for multiple anon THP sizes. 7557 See Documentation/admin-guide/mm/transhuge.rst for more 7558 details. 7559 7560 threadirqs [KNL,EARLY] 7561 Force threading of all interrupt handlers except those 7562 marked explicitly IRQF_NO_THREAD. 7563 7564 thp_shmem= [KNL] 7565 Format: <size>[KMG],<size>[KMG]:<policy>;<size>[KMG]-<size>[KMG]:<policy> 7566 Control the default policy of each hugepage size for the 7567 internal shmem mount. <policy> is one of policies available 7568 for the shmem mount ("always", "inherit", "never", "within_size", 7569 and "advise"). 7570 It can be used multiple times for multiple shmem THP sizes. 7571 See Documentation/admin-guide/mm/transhuge.rst for more 7572 details. 7573 7574 topology= [S390,EARLY] 7575 Format: {off | on} 7576 Specify if the kernel should make use of the cpu 7577 topology information if the hardware supports this. 7578 The scheduler will make use of this information and 7579 e.g. base its process migration decisions on it. 7580 Default is on. 7581 7582 torture.disable_onoff_at_boot= [KNL] 7583 Prevent the CPU-hotplug component of torturing 7584 until after init has spawned. 7585 7586 torture.ftrace_dump_at_shutdown= [KNL] 7587 Dump the ftrace buffer at torture-test shutdown, 7588 even if there were no errors. This can be a 7589 very costly operation when many torture tests 7590 are running concurrently, especially on systems 7591 with rotating-rust storage. 7592 7593 torture.verbose_sleep_frequency= [KNL] 7594 Specifies how many verbose printk()s should be 7595 emitted between each sleep. The default of zero 7596 disables verbose-printk() sleeping. 7597 7598 torture.verbose_sleep_duration= [KNL] 7599 Duration of each verbose-printk() sleep in jiffies. 7600 7601 tpm.disable_pcr_integrity= [HW,TPM] 7602 Do not protect PCR registers from unintended physical 7603 access, or interposers in the bus by the means of 7604 having an integrity protected session wrapped around 7605 TPM2_PCR_Extend command. Consider this in a situation 7606 where TPM is heavily utilized by IMA, thus protection 7607 causing a major performance hit, and the space where 7608 machines are deployed is by other means guarded. 7609 7610 tpm_crb_ffa.busy_timeout_ms= [ARM64,TPM] 7611 Maximum time in milliseconds to retry sending a message 7612 to the TPM service before giving up. This parameter controls 7613 how long the system will continue retrying when the TPM 7614 service is busy. 7615 Format: <unsigned int> 7616 Default: 2000 (2 seconds) 7617 7618 tpm_suspend_pcr=[HW,TPM] 7619 Format: integer pcr id 7620 Specify that at suspend time, the tpm driver 7621 should extend the specified pcr with zeros, 7622 as a workaround for some chips which fail to 7623 flush the last written pcr on TPM_SaveState. 7624 This will guarantee that all the other pcrs 7625 are saved. 7626 7627 tpm_tis.interrupts= [HW,TPM] 7628 Enable interrupts for the MMIO based physical layer 7629 for the FIFO interface. By default it is set to false 7630 (0). For more information about TPM hardware interfaces 7631 defined by Trusted Computing Group (TCG) see 7632 https://trustedcomputinggroup.org/resource/pc-client-platform-tpm-profile-ptp-specification/ 7633 7634 tp_printk [FTRACE] 7635 Have the tracepoints sent to printk as well as the 7636 tracing ring buffer. This is useful for early boot up 7637 where the system hangs or reboots and does not give the 7638 option for reading the tracing buffer or performing a 7639 ftrace_dump_on_oops. 7640 7641 To turn off having tracepoints sent to printk, 7642 echo 0 > /proc/sys/kernel/tracepoint_printk 7643 Note, echoing 1 into this file without the 7644 tp_printk kernel cmdline option has no effect. 7645 7646 The tp_printk_stop_on_boot (see below) can also be used 7647 to stop the printing of events to console at 7648 late_initcall_sync. 7649 7650 ** CAUTION ** 7651 7652 Having tracepoints sent to printk() and activating high 7653 frequency tracepoints such as irq or sched, can cause 7654 the system to live lock. 7655 7656 tp_printk_stop_on_boot [FTRACE] 7657 When tp_printk (above) is set, it can cause a lot of noise 7658 on the console. It may be useful to only include the 7659 printing of events during boot up, as user space may 7660 make the system inoperable. 7661 7662 This command line option will stop the printing of events 7663 to console at the late_initcall_sync() time frame. 7664 7665 trace_buf_size=nn[KMG] 7666 [FTRACE] will set tracing buffer size on each cpu. 7667 7668 trace_clock= [FTRACE] Set the clock used for tracing events 7669 at boot up. 7670 local - Use the per CPU time stamp counter 7671 (converted into nanoseconds). Fast, but 7672 depending on the architecture, may not be 7673 in sync between CPUs. 7674 global - Event time stamps are synchronized across 7675 CPUs. May be slower than the local clock, 7676 but better for some race conditions. 7677 counter - Simple counting of events (1, 2, ..) 7678 note, some counts may be skipped due to the 7679 infrastructure grabbing the clock more than 7680 once per event. 7681 uptime - Use jiffies as the time stamp. 7682 perf - Use the same clock that perf uses. 7683 mono - Use ktime_get_mono_fast_ns() for time stamps. 7684 mono_raw - Use ktime_get_raw_fast_ns() for time 7685 stamps. 7686 boot - Use ktime_get_boot_fast_ns() for time stamps. 7687 Architectures may add more clocks. See 7688 Documentation/trace/ftrace.rst for more details. 7689 7690 trace_event=[event-list] 7691 [FTRACE] Set and start specified trace events in order 7692 to facilitate early boot debugging. The event-list is a 7693 comma-separated list of trace events to enable. See 7694 also Documentation/trace/events.rst 7695 7696 To enable modules, use :mod: keyword: 7697 7698 trace_event=:mod:<module> 7699 7700 The value before :mod: will only enable specific events 7701 that are part of the module. See the above mentioned 7702 document for more information. 7703 7704 trace_instance=[instance-info] 7705 [FTRACE] Create a ring buffer instance early in boot up. 7706 This will be listed in: 7707 7708 /sys/kernel/tracing/instances 7709 7710 Events can be enabled at the time the instance is created 7711 via: 7712 7713 trace_instance=<name>,<system1>:<event1>,<system2>:<event2> 7714 7715 Note, the "<system*>:" portion is optional if the event is 7716 unique. 7717 7718 trace_instance=foo,sched:sched_switch,irq_handler_entry,initcall 7719 7720 will enable the "sched_switch" event (note, the "sched:" is optional, and 7721 the same thing would happen if it was left off). The irq_handler_entry 7722 event, and all events under the "initcall" system. 7723 7724 Flags can be added to the instance to modify its behavior when it is 7725 created. The flags are separated by '^'. 7726 7727 The available flags are: 7728 7729 traceoff - Have the tracing instance tracing disabled after it is created. 7730 traceprintk - Have trace_printk() write into this trace instance 7731 (note, "printk" and "trace_printk" can also be used) 7732 7733 trace_instance=foo^traceoff^traceprintk,sched,irq 7734 7735 The flags must come before the defined events. 7736 7737 If memory has been reserved (see memmap for x86), the instance 7738 can use that memory: 7739 7740 memmap=12M$0x284500000 trace_instance=boot_map@0x284500000:12M 7741 7742 The above will create a "boot_map" instance that uses the physical 7743 memory at 0x284500000 that is 12Megs. The per CPU buffers of that 7744 instance will be split up accordingly. 7745 7746 Alternatively, the memory can be reserved by the reserve_mem option: 7747 7748 reserve_mem=12M:4096:trace trace_instance=boot_map@trace 7749 7750 This will reserve 12 megabytes at boot up with a 4096 byte alignment 7751 and place the ring buffer in this memory. Note that due to KASLR, the 7752 memory may not be the same location each time, which will not preserve 7753 the buffer content. 7754 7755 Also note that the layout of the ring buffer data may change between 7756 kernel versions where the validator will fail and reset the ring buffer 7757 if the layout is not the same as the previous kernel. 7758 7759 If the ring buffer is used for persistent bootups and has events enabled, 7760 it is recommend to disable tracing so that events from a previous boot do not 7761 mix with events of the current boot (unless you are debugging a random crash 7762 at boot up). 7763 7764 reserve_mem=12M:4096:trace trace_instance=boot_map^traceoff^traceprintk@trace,sched,irq 7765 7766 Note, saving the trace buffer across reboots does require that the system 7767 is set up to not wipe memory. For instance, CONFIG_RESET_ATTACK_MITIGATION 7768 can force a memory reset on boot which will clear any trace that was stored. 7769 This is just one of many ways that can clear memory. Make sure your system 7770 keeps the content of memory across reboots before relying on this option. 7771 7772 NB: Both the mapped address and size must be page aligned for the architecture. 7773 7774 See also Documentation/trace/debugging.rst 7775 7776 7777 trace_options=[option-list] 7778 [FTRACE] Enable or disable tracer options at boot. 7779 The option-list is a comma delimited list of options 7780 that can be enabled or disabled just as if you were 7781 to echo the option name into 7782 7783 /sys/kernel/tracing/trace_options 7784 7785 For example, to enable stacktrace option (to dump the 7786 stack trace of each event), add to the command line: 7787 7788 trace_options=stacktrace 7789 7790 See also Documentation/trace/ftrace.rst "trace options" 7791 section. 7792 7793 trace_trigger=[trigger-list] 7794 [FTRACE] Add an event trigger on specific events. 7795 Set a trigger on top of a specific event, with an optional 7796 filter. 7797 7798 The format is "trace_trigger=<event>.<trigger>[ if <filter>],..." 7799 Where more than one trigger may be specified that are comma delimited. 7800 7801 For example: 7802 7803 trace_trigger="sched_switch.stacktrace if prev_state == 2" 7804 7805 The above will enable the "stacktrace" trigger on the "sched_switch" 7806 event but only trigger it if the "prev_state" of the "sched_switch" 7807 event is "2" (TASK_UNINTERRUPTIBLE). 7808 7809 See also "Event triggers" in Documentation/trace/events.rst 7810 7811 7812 traceoff_after_boot 7813 [FTRACE] Sometimes tracing is used to debug issues 7814 during the boot process. Since the trace buffer has a 7815 limited amount of storage, it may be prudent to 7816 disable tracing after the boot is finished, otherwise 7817 the critical information may be overwritten. With this 7818 option, the main tracing buffer will be turned off at 7819 the end of the boot process. 7820 7821 traceoff_on_warning 7822 [FTRACE] enable this option to disable tracing when a 7823 warning is hit. This turns off "tracing_on". Tracing can 7824 be enabled again by echoing '1' into the "tracing_on" 7825 file located in /sys/kernel/tracing/ 7826 7827 This option is useful, as it disables the trace before 7828 the WARNING dump is called, which prevents the trace to 7829 be filled with content caused by the warning output. 7830 7831 This option can also be set at run time via the sysctl 7832 option: kernel/traceoff_on_warning 7833 7834 transparent_hugepage= 7835 [KNL] 7836 Format: [always|madvise|never] 7837 Can be used to control the default behavior of the system 7838 with respect to transparent hugepages. 7839 See Documentation/admin-guide/mm/transhuge.rst 7840 for more details. 7841 7842 transparent_hugepage_shmem= [KNL] 7843 Format: [always|within_size|advise|never|deny|force] 7844 Can be used to control the hugepage allocation policy for 7845 the internal shmem mount. 7846 See Documentation/admin-guide/mm/transhuge.rst 7847 for more details. 7848 7849 transparent_hugepage_tmpfs= [KNL] 7850 Format: [always|within_size|advise|never] 7851 Can be used to control the default hugepage allocation policy 7852 for the tmpfs mount. 7853 See Documentation/admin-guide/mm/transhuge.rst 7854 for more details. 7855 7856 trusted.source= [KEYS] 7857 Format: <string> 7858 This parameter identifies the trust source as a backend 7859 for trusted keys implementation. Supported trust 7860 sources: 7861 - "tpm" 7862 - "tee" 7863 - "caam" 7864 - "dcp" 7865 - "pkwm" 7866 If not specified then it defaults to iterating through 7867 the trust source list starting with TPM and assigns the 7868 first trust source as a backend which is initialized 7869 successfully during iteration. 7870 7871 trusted.debug= [KEYS] 7872 Format: <bool> 7873 Enable trusted keys debug traces at runtime when 7874 CONFIG_TRUSTED_KEYS_DEBUG=y. 7875 7876 To make the traces visible after enabling the option, 7877 use trusted.dyndbg='+p' as needed. By convention, 7878 the subsystem uses pr_debug() for these traces. 7879 7880 SAFETY: The traces can leak sensitive data, so be 7881 cautious before enabling this. They remain inactive 7882 unless this parameter is set this option to a true 7883 value. 7884 7885 Default: false 7886 7887 trusted.rng= [KEYS] 7888 Format: <string> 7889 The RNG used to generate key material for trusted keys. 7890 Can be one of: 7891 - "kernel" 7892 - the same value as trusted.source: "tpm" or "tee" 7893 - "default" 7894 If not specified, "default" is used. In this case, 7895 the RNG's choice is left to each individual trust source. 7896 7897 trusted.dcp_use_otp_key 7898 This is intended to be used in combination with 7899 trusted.source=dcp and will select the DCP OTP key 7900 instead of the DCP UNIQUE key blob encryption. 7901 7902 trusted.dcp_skip_zk_test 7903 This is intended to be used in combination with 7904 trusted.source=dcp and will disable the check if the 7905 blob key is all zeros. This is helpful for situations where 7906 having this key zero'ed is acceptable. E.g. in testing 7907 scenarios. 7908 7909 tsa= [X86] Control mitigation for Transient Scheduler 7910 Attacks on AMD CPUs. Search the following in your 7911 favourite search engine for more details: 7912 7913 "Technical guidance for mitigating transient scheduler 7914 attacks". 7915 7916 off - disable the mitigation 7917 on - enable the mitigation (default) 7918 user - mitigate only user/kernel transitions 7919 vm - mitigate only guest/host transitions 7920 7921 7922 tsc= Disable clocksource stability checks for TSC. 7923 Format: <string> 7924 [x86] reliable: mark tsc clocksource as reliable, this 7925 disables clocksource verification at runtime, as well 7926 as the stability checks done at bootup. Used to enable 7927 high-resolution timer mode on older hardware, and in 7928 virtualized environment. 7929 [x86] noirqtime: Do not use TSC to do irq accounting. 7930 Used to run time disable IRQ_TIME_ACCOUNTING on any 7931 platforms where RDTSC is slow and this accounting 7932 can add overhead. 7933 [x86] unstable: mark the TSC clocksource as unstable, this 7934 marks the TSC unconditionally unstable at bootup and 7935 avoids any further wobbles once the TSC watchdog notices. 7936 [x86] nowatchdog: disable clocksource watchdog. Used 7937 in situations with strict latency requirements (where 7938 interruptions from clocksource watchdog are not 7939 acceptable). 7940 [x86] recalibrate: force recalibration against a HW timer 7941 (HPET or PM timer) on systems whose TSC frequency was 7942 obtained from HW or FW using either an MSR or CPUID(0x15). 7943 Warn if the difference is more than 500 ppm. 7944 [x86] watchdog: Enforce the clocksource watchdog on TSC 7945 7946 tsc_early_khz= [X86,EARLY] Skip early TSC calibration and use the given 7947 value instead. Useful when the early TSC frequency discovery 7948 procedure is not reliable, such as on overclocked systems 7949 with CPUID.16h support and partial CPUID.15h support. 7950 Format: <unsigned int> 7951 7952 tsx= [X86] Control Transactional Synchronization 7953 Extensions (TSX) feature in Intel processors that 7954 support TSX control. 7955 7956 This parameter controls the TSX feature. The options are: 7957 7958 on - Enable TSX on the system. Although there are 7959 mitigations for all known security vulnerabilities, 7960 TSX has been known to be an accelerator for 7961 several previous speculation-related CVEs, and 7962 so there may be unknown security risks associated 7963 with leaving it enabled. 7964 7965 off - Disable TSX on the system. (Note that this 7966 option takes effect only on newer CPUs which are 7967 not vulnerable to MDS, i.e., have 7968 MSR_IA32_ARCH_CAPABILITIES.MDS_NO=1 and which get 7969 the new IA32_TSX_CTRL MSR through a microcode 7970 update. This new MSR allows for the reliable 7971 deactivation of the TSX functionality.) 7972 7973 auto - Disable TSX if X86_BUG_TAA is present, 7974 otherwise enable TSX on the system. 7975 7976 Not specifying this option is equivalent to tsx=off. 7977 7978 See Documentation/admin-guide/hw-vuln/tsx_async_abort.rst 7979 for more details. 7980 7981 tsx_async_abort= [X86,INTEL,EARLY] Control mitigation for the TSX Async 7982 Abort (TAA) vulnerability. 7983 7984 Similar to Micro-architectural Data Sampling (MDS) 7985 certain CPUs that support Transactional 7986 Synchronization Extensions (TSX) are vulnerable to an 7987 exploit against CPU internal buffers which can forward 7988 information to a disclosure gadget under certain 7989 conditions. 7990 7991 In vulnerable processors, the speculatively forwarded 7992 data can be used in a cache side channel attack, to 7993 access data to which the attacker does not have direct 7994 access. 7995 7996 This parameter controls the TAA mitigation. The 7997 options are: 7998 7999 full - Enable TAA mitigation on vulnerable CPUs 8000 if TSX is enabled. 8001 8002 full,nosmt - Enable TAA mitigation and disable SMT on 8003 vulnerable CPUs. If TSX is disabled, SMT 8004 is not disabled because CPU is not 8005 vulnerable to cross-thread TAA attacks. 8006 off - Unconditionally disable TAA mitigation 8007 8008 On MDS-affected machines, tsx_async_abort=off can be 8009 prevented by an active MDS mitigation as both vulnerabilities 8010 are mitigated with the same mechanism so in order to disable 8011 this mitigation, you need to specify mds=off too. 8012 8013 Not specifying this option is equivalent to 8014 tsx_async_abort=full. On CPUs which are MDS affected 8015 and deploy MDS mitigation, TAA mitigation is not 8016 required and doesn't provide any additional 8017 mitigation. 8018 8019 For details see: 8020 Documentation/admin-guide/hw-vuln/tsx_async_abort.rst 8021 8022 turbografx.map[2|3]= [HW,JOY] 8023 TurboGraFX parallel port interface 8024 Format: 8025 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7> 8026 See also Documentation/input/devices/joystick-parport.rst 8027 8028 udbg-immortal [PPC] When debugging early kernel crashes that 8029 happen after console_init() and before a proper 8030 console driver takes over, this boot options might 8031 help "seeing" what's going on. 8032 8033 uhash_entries= [KNL,NET] 8034 Set number of hash buckets for UDP/UDP-Lite connections 8035 8036 uhci-hcd.ignore_oc= 8037 [USB] Ignore overcurrent events (default N). 8038 Some badly-designed motherboards generate lots of 8039 bogus events, for ports that aren't wired to 8040 anything. Set this parameter to avoid log spamming. 8041 Note that genuine overcurrent events won't be 8042 reported either. 8043 8044 unaligned_scalar_speed= 8045 [RISCV] 8046 Format: {slow | fast | unsupported} 8047 Allow skipping scalar unaligned access speed tests. This 8048 is useful for testing alternative code paths and to skip 8049 the tests in environments where they run too slowly. All 8050 CPUs must have the same scalar unaligned access speed. 8051 8052 unaligned_vector_speed= 8053 [RISCV] 8054 Format: {slow | fast | unsupported} 8055 Allow skipping vector unaligned access speed tests. This 8056 is useful for testing alternative code paths and to skip 8057 the tests in environments where they run too slowly. All 8058 CPUs must have the same vector unaligned access speed. 8059 8060 unknown_nmi_panic 8061 [X86] Cause panic on unknown NMI. 8062 8063 unwind_debug [X86-64,EARLY] 8064 Enable unwinder debug output. This can be 8065 useful for debugging certain unwinder error 8066 conditions, including corrupt stacks and 8067 bad/missing unwinder metadata. 8068 8069 usbcore.authorized_default= 8070 [USB] Default USB device authorization: 8071 (default -1 = authorized (same as 1), 8072 0 = not authorized, 1 = authorized, 2 = authorized 8073 if device connected to internal port) 8074 8075 usbcore.autosuspend= 8076 [USB] The autosuspend time delay (in seconds) used 8077 for newly-detected USB devices (default 2). This 8078 is the time required before an idle device will be 8079 autosuspended. Devices for which the delay is set 8080 to a negative value won't be autosuspended at all. 8081 8082 usbcore.usbfs_snoop= 8083 [USB] Set to log all usbfs traffic (default 0 = off). 8084 8085 usbcore.usbfs_snoop_max= 8086 [USB] Maximum number of bytes to snoop in each URB 8087 (default = 65536). 8088 8089 usbcore.blinkenlights= 8090 [USB] Set to cycle leds on hubs (default 0 = off). 8091 8092 usbcore.old_scheme_first= 8093 [USB] Start with the old device initialization 8094 scheme (default 0 = off). 8095 8096 usbcore.usbfs_memory_mb= 8097 [USB] Memory limit (in MB) for buffers allocated by 8098 usbfs (default = 16, 0 = max = 2047). 8099 8100 usbcore.use_both_schemes= 8101 [USB] Try the other device initialization scheme 8102 if the first one fails (default 1 = enabled). 8103 8104 usbcore.initial_descriptor_timeout= 8105 [USB] Specifies timeout for the initial 64-byte 8106 USB_REQ_GET_DESCRIPTOR request in milliseconds 8107 (default 5000 = 5.0 seconds). 8108 8109 usbcore.nousb [USB] Disable the USB subsystem 8110 8111 usbcore.quirks= 8112 [USB] A list of quirk entries to augment the built-in 8113 usb core quirk list. List entries are separated by 8114 commas. Each entry has the form 8115 VendorID:ProductID:Flags. The IDs are 4-digit hex 8116 numbers and Flags is a set of letters. Each letter 8117 will change the built-in quirk; setting it if it is 8118 clear and clearing it if it is set. The letters have 8119 the following meanings: 8120 a = USB_QUIRK_STRING_FETCH_255 (string 8121 descriptors must not be fetched using 8122 a 255-byte read); 8123 b = USB_QUIRK_RESET_RESUME (device can't resume 8124 correctly so reset it instead); 8125 c = USB_QUIRK_NO_SET_INTF (device can't handle 8126 Set-Interface requests); 8127 d = USB_QUIRK_CONFIG_INTF_STRINGS (device can't 8128 handle its Configuration or Interface 8129 strings); 8130 e = USB_QUIRK_RESET (device can't be reset 8131 (e.g morph devices), don't use reset); 8132 f = USB_QUIRK_HONOR_BNUMINTERFACES (device has 8133 more interface descriptions than the 8134 bNumInterfaces count, and can't handle 8135 talking to these interfaces); 8136 g = USB_QUIRK_DELAY_INIT (device needs a pause 8137 during initialization, after we read 8138 the device descriptor); 8139 h = USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL (For 8140 high speed and super speed interrupt 8141 endpoints, the USB 2.0 and USB 3.0 spec 8142 require the interval in microframes (1 8143 microframe = 125 microseconds) to be 8144 calculated as interval = 2 ^ 8145 (bInterval-1). 8146 Devices with this quirk report their 8147 bInterval as the result of this 8148 calculation instead of the exponent 8149 variable used in the calculation); 8150 i = USB_QUIRK_DEVICE_QUALIFIER (device can't 8151 handle device_qualifier descriptor 8152 requests); 8153 j = USB_QUIRK_IGNORE_REMOTE_WAKEUP (device 8154 generates spurious wakeup, ignore 8155 remote wakeup capability); 8156 k = USB_QUIRK_NO_LPM (device can't handle Link 8157 Power Management); 8158 l = USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL 8159 (Device reports its bInterval as linear 8160 frames instead of the USB 2.0 8161 calculation); 8162 m = USB_QUIRK_DISCONNECT_SUSPEND (Device needs 8163 to be disconnected before suspend to 8164 prevent spurious wakeup); 8165 n = USB_QUIRK_DELAY_CTRL_MSG (Device needs a 8166 pause after every control message); 8167 o = USB_QUIRK_HUB_SLOW_RESET (Hub needs extra 8168 delay after resetting its port); 8169 p = USB_QUIRK_SHORT_SET_ADDRESS_REQ_TIMEOUT 8170 (Reduce timeout of the SET_ADDRESS 8171 request from 5000 ms to 500 ms); 8172 q = USB_QUIRK_FORCE_ONE_CONFIG (Device 8173 claims zero configurations, 8174 forcing to 1); 8175 r = USB_QUIRK_WINDOWS_CONFIG_REQ_SIZE (Device 8176 fails during initialization when asked for 8177 9-bytes configuration descriptor request. 8178 Ask for 255-bytes request instead to mirror 8179 Windows' behavior); 8180 Example: quirks=0781:5580:bk,0a5c:5834:gij 8181 8182 usbhid.mousepoll= 8183 [USBHID] The interval which mice are to be polled at. 8184 8185 usbhid.jspoll= 8186 [USBHID] The interval which joysticks are to be polled at. 8187 8188 usbhid.kbpoll= 8189 [USBHID] The interval which keyboards are to be polled at. 8190 8191 usb-storage.delay_use= 8192 [UMS] The delay in seconds before a new device is 8193 scanned for Logical Units (default 1). 8194 Optionally the delay in milliseconds if the value has 8195 suffix with "ms". 8196 Example: delay_use=2567ms 8197 8198 usb-storage.quirks= 8199 [UMS] A list of quirks entries to supplement or 8200 override the built-in unusual_devs list. List 8201 entries are separated by commas. Each entry has 8202 the form VID:PID:Flags where VID and PID are Vendor 8203 and Product ID values (4-digit hex numbers) and 8204 Flags is a set of characters, each corresponding 8205 to a common usb-storage quirk flag as follows: 8206 a = SANE_SENSE (collect more than 18 bytes 8207 of sense data, not on uas); 8208 b = BAD_SENSE (don't collect more than 18 8209 bytes of sense data, not on uas); 8210 c = FIX_CAPACITY (decrease the reported 8211 device capacity by one sector); 8212 d = NO_READ_DISC_INFO (don't use 8213 READ_DISC_INFO command, not on uas); 8214 e = NO_READ_CAPACITY_16 (don't use 8215 READ_CAPACITY_16 command); 8216 f = NO_REPORT_OPCODES (don't use report opcodes 8217 command, uas only); 8218 g = MAX_SECTORS_240 (don't transfer more than 8219 240 sectors at a time, uas only); 8220 h = CAPACITY_HEURISTICS (decrease the 8221 reported device capacity by one 8222 sector if the number is odd); 8223 i = IGNORE_DEVICE (don't bind to this 8224 device); 8225 j = NO_REPORT_LUNS (don't use report luns 8226 command, uas only); 8227 k = NO_SAME (do not use WRITE_SAME, uas only) 8228 l = NOT_LOCKABLE (don't try to lock and 8229 unlock ejectable media, not on uas); 8230 m = MAX_SECTORS_64 (don't transfer more 8231 than 64 sectors = 32 KB at a time, 8232 not on uas); 8233 n = INITIAL_READ10 (force a retry of the 8234 initial READ(10) command, not on uas); 8235 o = CAPACITY_OK (accept the capacity 8236 reported by the device, not on uas); 8237 p = WRITE_CACHE (the device cache is ON 8238 by default, not on uas); 8239 r = IGNORE_RESIDUE (the device reports 8240 bogus residue values, not on uas); 8241 s = SINGLE_LUN (the device has only one 8242 Logical Unit); 8243 t = NO_ATA_1X (don't allow ATA(12) and ATA(16) 8244 commands, uas only); 8245 u = IGNORE_UAS (don't bind to the uas driver); 8246 w = NO_WP_DETECT (don't test whether the 8247 medium is write-protected). 8248 y = ALWAYS_SYNC (issue a SYNCHRONIZE_CACHE 8249 even if the device claims no cache, 8250 not on uas) 8251 Example: quirks=0419:aaf5:rl,0421:0433:rc 8252 8253 user_debug= [KNL,ARM] 8254 Format: <int> 8255 See arch/arm/Kconfig.debug help text. 8256 1 - undefined instruction events 8257 2 - system calls 8258 4 - invalid data aborts 8259 8 - SIGSEGV faults 8260 16 - SIGBUS faults 8261 Example: user_debug=31 8262 8263 vdso= [X86,SH,SPARC] 8264 On X86_32, this is an alias for vdso32=. Otherwise: 8265 8266 vdso=1: enable VDSO (the default) 8267 vdso=0: disable VDSO mapping 8268 8269 vdso32= [X86] Control the 32-bit vDSO 8270 vdso32=1: enable 32-bit VDSO 8271 vdso32=0 or vdso32=2: disable 32-bit VDSO 8272 8273 See the help text for CONFIG_COMPAT_VDSO for more 8274 details. If CONFIG_COMPAT_VDSO is set, the default is 8275 vdso32=0; otherwise, the default is vdso32=1. 8276 8277 For compatibility with older kernels, vdso32=2 is an 8278 alias for vdso32=0. 8279 8280 Try vdso32=0 if you encounter an error that says: 8281 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed! 8282 8283 video= [FB,EARLY] Frame buffer configuration 8284 See Documentation/fb/modedb.rst. 8285 8286 video.brightness_switch_enabled= [ACPI] 8287 Format: [0|1] 8288 If set to 1, on receiving an ACPI notify event 8289 generated by hotkey, video driver will adjust brightness 8290 level and then send out the event to user space through 8291 the allocated input device. If set to 0, video driver 8292 will only send out the event without touching backlight 8293 brightness level. 8294 default: 1 8295 8296 virtio_mmio.device= 8297 [VMMIO] Memory mapped virtio (platform) device. 8298 8299 <size>@<baseaddr>:<irq>[:<id>] 8300 where: 8301 <size> := size (can use standard suffixes 8302 like K, M and G) 8303 <baseaddr> := physical base address 8304 <irq> := interrupt number (as passed to 8305 request_irq()) 8306 <id> := (optional) platform device id 8307 example: 8308 virtio_mmio.device=1K@0x100b0000:48:7 8309 8310 Can be used multiple times for multiple devices. 8311 8312 vga= [BOOT,X86-32] Select a particular video mode 8313 See Documentation/arch/x86/boot.rst and 8314 Documentation/admin-guide/svga.rst. 8315 Use vga=ask for menu. 8316 This is actually a boot loader parameter; the value is 8317 passed to the kernel using a special protocol. 8318 8319 vm_debug[=options] [KNL] Available with CONFIG_DEBUG_VM=y. 8320 May slow down system boot speed, especially when 8321 enabled on systems with a large amount of memory. 8322 All options are enabled by default, and this 8323 interface is meant to allow for selectively 8324 enabling or disabling specific virtual memory 8325 debugging features. 8326 8327 Available options are: 8328 P Enable page structure init time poisoning 8329 - Disable all of the above options 8330 8331 vmalloc=nn[KMG] [KNL,BOOT,EARLY] Forces the vmalloc area to have an 8332 exact size of <nn>. This can be used to increase 8333 the minimum size (128MB on x86, arm32 platforms). 8334 It can also be used to decrease the size and leave more room 8335 for directly mapped kernel RAM. Note that this parameter does 8336 not exist on many other platforms (including arm64, alpha, 8337 loongarch, arc, csky, hexagon, microblaze, mips, nios2, openrisc, 8338 parisc, m64k, powerpc, riscv, sh, um, xtensa, s390, sparc). 8339 8340 vmcp_cma=nn[MG] [KNL,S390,EARLY] 8341 Sets the memory size reserved for contiguous memory 8342 allocations for the vmcp device driver. 8343 8344 vmhalt= [KNL,S390] Perform z/VM CP command after system halt. 8345 Format: <command> 8346 8347 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic. 8348 Format: <command> 8349 8350 vmpoff= [KNL,S390] Perform z/VM CP command after power off. 8351 Format: <command> 8352 8353 vmscape= [X86] Controls mitigation for VMscape attacks. 8354 VMscape attacks can leak information from a userspace 8355 hypervisor to a guest via speculative side-channels. 8356 8357 off - disable the mitigation 8358 ibpb - use Indirect Branch Prediction Barrier 8359 (IBPB) mitigation (default) 8360 force - force vulnerability detection even on 8361 unaffected processors 8362 8363 vsyscall= [X86-64,EARLY] 8364 Controls the behavior of vsyscalls (i.e. calls to 8365 fixed addresses of 0xffffffffff600x00 from legacy 8366 code). Most statically-linked binaries and older 8367 versions of glibc use these calls. Because these 8368 functions are at fixed addresses, they make nice 8369 targets for exploits that can control RIP. 8370 8371 emulate Vsyscalls turn into traps and are emulated 8372 reasonably safely. The vsyscall page is 8373 readable. This disables the Linear 8374 Address Space Separation (LASS) security 8375 feature and makes the system less secure. 8376 8377 xonly [default] Vsyscalls turn into traps and are 8378 emulated reasonably safely. The vsyscall 8379 page is not readable. 8380 8381 none Vsyscalls don't work at all. This makes 8382 them quite hard to use for exploits but 8383 might break your system. 8384 8385 vt.color= [VT] Default text color. 8386 Format: 0xYX, X = foreground, Y = background. 8387 Default: 0x07 = light gray on black. 8388 8389 vt.cur_default= [VT] Default cursor shape. 8390 Format: 0xCCBBAA, where AA, BB, and CC are the same as 8391 the parameters of the <Esc>[?A;B;Cc escape sequence; 8392 see vga-softcursor.rst. Default: 2 = underline. 8393 8394 vt.default_blu= [VT] 8395 Format: <blue0>,<blue1>,<blue2>,...,<blue15> 8396 Change the default blue palette of the console. 8397 This is a 16-member array composed of values 8398 ranging from 0-255. 8399 8400 vt.default_grn= [VT] 8401 Format: <green0>,<green1>,<green2>,...,<green15> 8402 Change the default green palette of the console. 8403 This is a 16-member array composed of values 8404 ranging from 0-255. 8405 8406 vt.default_red= [VT] 8407 Format: <red0>,<red1>,<red2>,...,<red15> 8408 Change the default red palette of the console. 8409 This is a 16-member array composed of values 8410 ranging from 0-255. 8411 8412 vt.default_utf8= 8413 [VT] 8414 Format=<0|1> 8415 Set system-wide default UTF-8 mode for all tty's. 8416 Default is 1, i.e. UTF-8 mode is enabled for all 8417 newly opened terminals. 8418 8419 vt.global_cursor_default= 8420 [VT] 8421 Format=<-1|0|1> 8422 Set system-wide default for whether a cursor 8423 is shown on new VTs. Default is -1, 8424 i.e. cursors will be created by default unless 8425 overridden by individual drivers. 0 will hide 8426 cursors, 1 will display them. 8427 8428 vt.italic= [VT] Default color for italic text; 0-15. 8429 Default: 2 = green. 8430 8431 vt.underline= [VT] Default color for underlined text; 0-15. 8432 Default: 3 = cyan. 8433 8434 watchdog timers [HW,WDT] For information on watchdog timers, 8435 see Documentation/watchdog/watchdog-parameters.rst 8436 or other driver-specific files in the 8437 Documentation/watchdog/ directory. 8438 8439 watchdog_thresh= 8440 [KNL] 8441 Set the hard lockup detector stall duration 8442 threshold in seconds. The soft lockup detector 8443 threshold is set to twice the value. A value of 0 8444 disables both lockup detectors. Default is 10 8445 seconds. 8446 8447 workqueue.unbound_cpus= 8448 [KNL,SMP] Specify to constrain one or some CPUs 8449 to use in unbound workqueues. 8450 Format: <cpu-list> 8451 By default, all online CPUs are available for 8452 unbound workqueues. 8453 8454 workqueue.watchdog_thresh= 8455 If CONFIG_WQ_WATCHDOG is configured, workqueue can 8456 warn stall conditions and dump internal state to 8457 help debugging. 0 disables workqueue stall 8458 detection; otherwise, it's the stall threshold 8459 duration in seconds. The default value is 30 and 8460 it can be updated at runtime by writing to the 8461 corresponding sysfs file. 8462 8463 workqueue.panic_on_stall=<uint> 8464 Panic when workqueue stall is detected by 8465 CONFIG_WQ_WATCHDOG. It sets the number times of the 8466 stall to trigger panic. 8467 8468 The default is set by CONFIG_BOOTPARAM_WQ_STALL_PANIC, 8469 which is 0 (disabled) if not configured. 8470 8471 workqueue.panic_on_stall_time=<uint> 8472 Panic when a workqueue stall has been continuous for 8473 the specified number of seconds. Unlike panic_on_stall 8474 which counts accumulated stall events, this triggers 8475 based on the duration of a single continuous stall. 8476 8477 The default is 0, which disables the time-based panic. 8478 8479 workqueue.cpu_intensive_thresh_us= 8480 Per-cpu work items which run for longer than this 8481 threshold are automatically considered CPU intensive 8482 and excluded from concurrency management to prevent 8483 them from noticeably delaying other per-cpu work 8484 items. Default is 10000 (10ms). 8485 8486 If CONFIG_WQ_CPU_INTENSIVE_REPORT is set, the kernel 8487 will report the work functions which violate this 8488 threshold repeatedly. They are likely good 8489 candidates for using WQ_UNBOUND workqueues instead. 8490 8491 workqueue.cpu_intensive_warning_thresh=<uint> 8492 If CONFIG_WQ_CPU_INTENSIVE_REPORT is set, the kernel 8493 will report the work functions which violate the 8494 intensive_threshold_us repeatedly. In order to prevent 8495 spurious warnings, start printing only after a work 8496 function has violated this threshold number of times. 8497 8498 The default is 4 times. 0 disables the warning. 8499 8500 workqueue.power_efficient 8501 Per-cpu workqueues are generally preferred because 8502 they show better performance thanks to cache 8503 locality; unfortunately, per-cpu workqueues tend to 8504 be more power hungry than unbound workqueues. 8505 8506 Enabling this makes the per-cpu workqueues which 8507 were observed to contribute significantly to power 8508 consumption unbound, leading to measurably lower 8509 power usage at the cost of small performance 8510 overhead. 8511 8512 The default value of this parameter is determined by 8513 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT. 8514 8515 workqueue.default_affinity_scope= 8516 Select the default affinity scope to use for unbound 8517 workqueues. Can be one of "cpu", "smt", "cache", 8518 "cache_shard", "numa" and "system". Default is 8519 "cache_shard". For more 8520 information, see the Affinity Scopes section in 8521 Documentation/core-api/workqueue.rst. 8522 8523 This can be changed after boot by writing to the 8524 matching /sys/module/workqueue/parameters file. All 8525 workqueues with the "default" affinity scope will be 8526 updated accordingly. 8527 8528 workqueue.debug_force_rr_cpu 8529 Workqueue used to implicitly guarantee that work 8530 items queued without explicit CPU specified are put 8531 on the local CPU. This guarantee is no longer true 8532 and while local CPU is still preferred work items 8533 may be put on foreign CPUs. This debug option 8534 forces round-robin CPU selection to flush out 8535 usages which depend on the now broken guarantee. 8536 When enabled, memory and cache locality will be 8537 impacted. 8538 8539 writecombine= [LOONGARCH,EARLY] Control the MAT (Memory Access 8540 Type) of ioremap_wc(). 8541 8542 on - Enable writecombine, use WUC for ioremap_wc() 8543 off - Disable writecombine, use SUC for ioremap_wc() 8544 8545 x2apic_phys [X86-64,APIC,EARLY] Use x2apic physical mode instead of 8546 default x2apic cluster mode on platforms 8547 supporting x2apic. 8548 8549 xen_512gb_limit [KNL,X86-64,XEN] 8550 Restricts the kernel running paravirtualized under Xen 8551 to use only up to 512 GB of RAM. The reason to do so is 8552 crash analysis tools and Xen tools for doing domain 8553 save/restore/migration must be enabled to handle larger 8554 domains. 8555 8556 xen_console_io [XEN,EARLY] 8557 Boolean option to enable/disable the usage of the Xen 8558 console_io hypercalls to read and write to the console. 8559 Mostly useful for debugging and development. 8560 8561 xen_emul_unplug= [HW,X86,XEN,EARLY] 8562 Unplug Xen emulated devices 8563 Format: [unplug0,][unplug1] 8564 ide-disks -- unplug primary master IDE devices 8565 aux-ide-disks -- unplug non-primary-master IDE devices 8566 nics -- unplug network devices 8567 all -- unplug all emulated devices (NICs and IDE disks) 8568 unnecessary -- unplugging emulated devices is 8569 unnecessary even if the host did not respond to 8570 the unplug protocol 8571 never -- do not unplug even if version check succeeds 8572 8573 xen_legacy_crash [X86,XEN,EARLY] 8574 Crash from Xen panic notifier, without executing late 8575 panic() code such as dumping handler. 8576 8577 xen_mc_debug [X86,XEN,EARLY] 8578 Enable multicall debugging when running as a Xen PV guest. 8579 Enabling this feature will reduce performance a little 8580 bit, so it should only be enabled for obtaining extended 8581 debug data in case of multicall errors. 8582 8583 xen_msr_safe= [X86,XEN,EARLY] 8584 Format: <bool> 8585 Select whether to always use non-faulting (safe) MSR 8586 access functions when running as Xen PV guest. The 8587 default value is controlled by CONFIG_XEN_PV_MSR_SAFE. 8588 8589 xen_nopv [X86] 8590 Disables the PV optimizations forcing the HVM guest to 8591 run as generic HVM guest with no PV drivers. 8592 This option is obsoleted by the "nopv" option, which 8593 has equivalent effect for XEN platform. 8594 8595 xen_no_vector_callback 8596 [KNL,X86,XEN,EARLY] Disable the vector callback for Xen 8597 event channel interrupts. 8598 8599 xen_scrub_pages= [XEN] 8600 Boolean option to control scrubbing pages before giving them back 8601 to Xen, for use by other domains. Can be also changed at runtime 8602 with /sys/devices/system/xen_memory/xen_memory0/scrub_pages. 8603 Default value controlled with CONFIG_XEN_SCRUB_PAGES_DEFAULT. 8604 8605 xen_timer_slop= [X86-64,XEN,EARLY] 8606 Set the timer slop (in nanoseconds) for the virtual Xen 8607 timers (default is 100000). This adjusts the minimum 8608 delta of virtualized Xen timers, where lower values 8609 improve timer resolution at the expense of processing 8610 more timer interrupts. 8611 8612 xen.balloon_boot_timeout= [XEN] 8613 The time (in seconds) to wait before giving up to boot 8614 in case initial ballooning fails to free enough memory. 8615 Applies only when running as HVM or PVH guest and 8616 started with less memory configured than allowed at 8617 max. Default is 180. 8618 8619 xen.event_eoi_delay= [XEN] 8620 How long to delay EOI handling in case of event 8621 storms (jiffies). Default is 10. 8622 8623 xen.event_loop_timeout= [XEN] 8624 After which time (jiffies) the event handling loop 8625 should start to delay EOI handling. Default is 2. 8626 8627 xen.fifo_events= [XEN] 8628 Boolean parameter to disable using fifo event handling 8629 even if available. Normally fifo event handling is 8630 preferred over the 2-level event handling, as it is 8631 fairer and the number of possible event channels is 8632 much higher. Default is on (use fifo events). 8633 8634 xirc2ps_cs= [NET,PCMCIA] 8635 Format: 8636 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]] 8637 8638 xive= [PPC] 8639 By default on POWER9 and above, the kernel will 8640 natively use the XIVE interrupt controller. This option 8641 allows the fallback firmware mode to be used: 8642 8643 off Fallback to firmware control of XIVE interrupt 8644 controller on both pseries and powernv 8645 platforms. Only useful on POWER9 and above. 8646 8647 xive.store-eoi=off [PPC] 8648 By default on POWER10 and above, the kernel will use 8649 stores for EOI handling when the XIVE interrupt mode 8650 is active. This option allows the XIVE driver to use 8651 loads instead, as on POWER9. 8652 8653 xhci-hcd.quirks [USB,KNL] 8654 A hex value specifying bitmask with supplemental xhci 8655 host controller quirks. Meaning of each bit can be 8656 consulted in header drivers/usb/host/xhci.h. 8657 8658 xmon [PPC,EARLY] 8659 Format: { early | on | rw | ro | off } 8660 Controls if xmon debugger is enabled. Default is off. 8661 Passing only "xmon" is equivalent to "xmon=early". 8662 early Call xmon as early as possible on boot; xmon 8663 debugger is called from setup_arch(). 8664 on xmon debugger hooks will be installed so xmon 8665 is only called on a kernel crash. Default mode, 8666 i.e. either "ro" or "rw" mode, is controlled 8667 with CONFIG_XMON_DEFAULT_RO_MODE. 8668 rw xmon debugger hooks will be installed so xmon 8669 is called only on a kernel crash, mode is write, 8670 meaning SPR registers, memory and, other data 8671 can be written using xmon commands. 8672 ro same as "rw" option above but SPR registers, 8673 memory, and other data can't be written using 8674 xmon commands. 8675 off xmon is disabled. 8676