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 CPU_FREQ CPU frequency scaling is enabled. 12 CPU_IDLE CPU idle time management is enabled. 13 DRM Direct Rendering Management support is enabled. 14 DYNAMIC_DEBUG Build in debug messages and enable them at runtime 15 EARLY Parameter processed too early to be embedded in initrd. 16 EDD BIOS Enhanced Disk Drive Services (EDD) is enabled 17 EFI EFI Partitioning (GPT) is enabled 18 EVM Extended Verification Module 19 FB The frame buffer device is enabled. 20 FTRACE Function tracing enabled. 21 GCOV GCOV profiling is enabled. 22 HIBERNATION HIBERNATION is enabled. 23 HW Appropriate hardware is enabled. 24 HYPER_V HYPERV support is enabled. 25 IMA Integrity measurement architecture is enabled. 26 IP_PNP IP DHCP, BOOTP, or RARP is enabled. 27 IPV6 IPv6 support is enabled. 28 ISAPNP ISA PnP code is enabled. 29 ISOL CPU Isolation is enabled. 30 JOY Appropriate joystick support is enabled. 31 KGDB Kernel debugger support is enabled. 32 KVM Kernel Virtual Machine support is enabled. 33 LIBATA Libata driver is enabled 34 LOONGARCH LoongArch architecture is enabled. 35 LOOP Loopback device support is enabled. 36 LP Printer support is enabled. 37 M68k M68k architecture is enabled. 38 These options have more detailed description inside of 39 Documentation/arch/m68k/kernel-options.rst. 40 MIPS MIPS architecture is enabled. 41 MOUSE Appropriate mouse support is enabled. 42 MSI Message Signaled Interrupts (PCI). 43 MTD MTD (Memory Technology Device) support is enabled. 44 NET Appropriate network support is enabled. 45 NFS Appropriate NFS support is enabled. 46 NUMA NUMA support is enabled. 47 OF Devicetree is enabled. 48 PARISC The PA-RISC architecture is enabled. 49 PCI PCI bus support is enabled. 50 PCIE PCI Express support is enabled. 51 PCMCIA The PCMCIA subsystem is enabled. 52 PM Power Management support is enabled. 53 PNP Plug & Play support is enabled. 54 PPC PowerPC architecture is enabled. 55 PPT Parallel port support is enabled. 56 PS2 Appropriate PS/2 support is enabled. 57 PV_OPS A paravirtualized kernel is enabled. 58 RAM RAM disk support is enabled. 59 RDT Intel Resource Director Technology. 60 RISCV RISCV architecture is enabled. 61 S390 S390 architecture is enabled. 62 SCSI Appropriate SCSI support is enabled. 63 A lot of drivers have their options described inside 64 the Documentation/scsi/ sub-directory. 65 SDW SoundWire support is enabled. 66 SECURITY Different security models are enabled. 67 SELINUX SELinux support is enabled. 68 SERIAL Serial support is enabled. 69 SH SuperH architecture is enabled. 70 SMP The kernel is an SMP kernel. 71 SPARC Sparc architecture is enabled. 72 SUSPEND System suspend states are enabled. 73 SWSUSP Software suspend (hibernation) is enabled. 74 TPM TPM drivers are enabled. 75 UMS USB Mass Storage support is enabled. 76 USB USB support is enabled. 77 NVME NVMe support is enabled 78 USBHID USB Human Interface Device support is enabled. 79 V4L Video For Linux support is enabled. 80 VGA The VGA console has been enabled. 81 VMMIO Driver for memory mapped virtio devices is enabled. 82 VT Virtual terminal support is enabled. 83 WDT Watchdog support is enabled. 84 X86-32 X86-32, aka i386 architecture is enabled. 85 X86-64 X86-64 architecture is enabled. 86 X86 Either 32-bit or 64-bit x86 (same as X86-32+X86-64) 87 X86_UV SGI UV support is enabled. 88 XEN Xen support is enabled 89 XTENSA xtensa architecture is enabled. 90 91In addition, the following text indicates that the option 92 93 BOOT Is a boot loader parameter. 94 BUGS= Relates to possible processor bugs on the said processor. 95 KNL Is a kernel start-up parameter. 96 97 98Kernel parameters 99 100 accept_memory= [MM] 101 Format: { eager | lazy } 102 default: lazy 103 By default, unaccepted memory is accepted lazily to 104 avoid prolonged boot times. The lazy option will add 105 some runtime overhead until all memory is eventually 106 accepted. In most cases the overhead is negligible. 107 For some workloads or for debugging purposes 108 accept_memory=eager can be used to accept all memory 109 at once during boot. 110 111 acpi= [HW,ACPI,X86,ARM64,RISCV64,EARLY] 112 Advanced Configuration and Power Interface 113 Format: { force | on | off | strict | noirq | rsdt | 114 copy_dsdt | nospcr } 115 force -- enable ACPI if default was off 116 on -- enable ACPI but allow fallback to DT [arm64,riscv64] 117 off -- disable ACPI if default was on 118 noirq -- do not use ACPI for IRQ routing 119 strict -- Be less tolerant of platforms that are not 120 strictly ACPI specification compliant. 121 rsdt -- prefer RSDT over (default) XSDT 122 copy_dsdt -- copy DSDT to memory 123 nocmcff -- Disable firmware first mode for corrected 124 errors. This disables parsing the HEST CMC error 125 source to check if firmware has set the FF flag. This 126 may result in duplicate corrected error reports. 127 nospcr -- disable console in ACPI SPCR table as 128 default _serial_ console on ARM64 129 spcr -- enable console in ACPI SPCR table as 130 default _serial_ console on x86 131 For ARM64, ONLY "acpi=off", "acpi=on", "acpi=force" or 132 "acpi=nospcr" are available 133 For RISCV64, ONLY "acpi=off", "acpi=on" or "acpi=force" 134 are available 135 136 See also Documentation/power/runtime_pm.rst, pci=noacpi 137 138 acpi_apic_instance= [ACPI,IOAPIC,EARLY] 139 Format: <int> 140 2: use 2nd APIC table, if available 141 1,0: use 1st APIC table 142 default: 0 143 144 acpi_backlight= [HW,ACPI] 145 { vendor | video | native | none } 146 If set to vendor, prefer vendor-specific driver 147 (e.g. thinkpad_acpi, sony_acpi, etc.) instead 148 of the ACPI video.ko driver. 149 If set to video, use the ACPI video.ko driver. 150 If set to native, use the device's native backlight mode. 151 If set to none, disable the ACPI backlight interface. 152 153 acpi_force_32bit_fadt_addr [ACPI,EARLY] 154 force FADT to use 32 bit addresses rather than the 155 64 bit X_* addresses. Some firmware have broken 64 156 bit addresses for force ACPI ignore these and use 157 the older legacy 32 bit addresses. 158 159 acpica_no_return_repair [HW, ACPI] 160 Disable AML predefined validation mechanism 161 This mechanism can repair the evaluation result to make 162 the return objects more ACPI specification compliant. 163 This option is useful for developers to identify the 164 root cause of an AML interpreter issue when the issue 165 has something to do with the repair mechanism. 166 167 acpi.debug_layer= [HW,ACPI,ACPI_DEBUG] 168 acpi.debug_level= [HW,ACPI,ACPI_DEBUG] 169 Format: <int> 170 CONFIG_ACPI_DEBUG must be enabled to produce any ACPI 171 debug output. Bits in debug_layer correspond to a 172 _COMPONENT in an ACPI source file, e.g., 173 #define _COMPONENT ACPI_EVENTS 174 Bits in debug_level correspond to a level in 175 ACPI_DEBUG_PRINT statements, e.g., 176 ACPI_DEBUG_PRINT((ACPI_DB_INFO, ... 177 The debug_level mask defaults to "info". See 178 Documentation/firmware-guide/acpi/debug.rst for more information about 179 debug layers and levels. 180 181 Enable processor driver info messages: 182 acpi.debug_layer=0x20000000 183 Enable AML "Debug" output, i.e., stores to the Debug 184 object while interpreting AML: 185 acpi.debug_layer=0xffffffff acpi.debug_level=0x2 186 Enable all messages related to ACPI hardware: 187 acpi.debug_layer=0x2 acpi.debug_level=0xffffffff 188 189 Some values produce so much output that the system is 190 unusable. The "log_buf_len" parameter may be useful 191 if you need to capture more output. 192 193 acpi.poweroff_on_fatal= [ACPI] 194 {0 | 1} 195 Causes the system to poweroff when the ACPI bytecode signals 196 a fatal error. The default value of this setting is 1. 197 Overriding this value should only be done for diagnosing 198 ACPI firmware problems, as the system might behave erratically 199 after having encountered a fatal ACPI error. 200 201 acpi_enforce_resources= [ACPI] 202 { strict | lax | no } 203 Check for resource conflicts between native drivers 204 and ACPI OperationRegions (SystemIO and SystemMemory 205 only). IO ports and memory declared in ACPI might be 206 used by the ACPI subsystem in arbitrary AML code and 207 can interfere with legacy drivers. 208 strict (default): access to resources claimed by ACPI 209 is denied; legacy drivers trying to access reserved 210 resources will fail to bind to device using them. 211 lax: access to resources claimed by ACPI is allowed; 212 legacy drivers trying to access reserved resources 213 will bind successfully but a warning message is logged. 214 no: ACPI OperationRegions are not marked as reserved, 215 no further checks are performed. 216 217 acpi_force_table_verification [HW,ACPI,EARLY] 218 Enable table checksum verification during early stage. 219 By default, this is disabled due to x86 early mapping 220 size limitation. 221 222 acpi_irq_balance [HW,ACPI] 223 ACPI will balance active IRQs 224 default in APIC mode 225 226 acpi_irq_nobalance [HW,ACPI] 227 ACPI will not move active IRQs (default) 228 default in PIC mode 229 230 acpi_irq_isa= [HW,ACPI] If irq_balance, mark listed IRQs used by ISA 231 Format: <irq>,<irq>... 232 233 acpi_irq_pci= [HW,ACPI] If irq_balance, clear listed IRQs for 234 use by PCI 235 Format: <irq>,<irq>... 236 237 acpi_mask_gpe= [HW,ACPI] 238 Due to the existence of _Lxx/_Exx, some GPEs triggered 239 by unsupported hardware/firmware features can result in 240 GPE floodings that cannot be automatically disabled by 241 the GPE dispatcher. 242 This facility can be used to prevent such uncontrolled 243 GPE floodings. 244 Format: <byte> or <bitmap-list> 245 246 acpi_no_auto_serialize [HW,ACPI] 247 Disable auto-serialization of AML methods 248 AML control methods that contain the opcodes to create 249 named objects will be marked as "Serialized" by the 250 auto-serialization feature. 251 This feature is enabled by default. 252 This option allows to turn off the feature. 253 254 acpi_no_memhotplug [ACPI] Disable memory hotplug. Useful for kdump 255 kernels. 256 257 acpi_no_static_ssdt [HW,ACPI,EARLY] 258 Disable installation of static SSDTs at early boot time 259 By default, SSDTs contained in the RSDT/XSDT will be 260 installed automatically and they will appear under 261 /sys/firmware/acpi/tables. 262 This option turns off this feature. 263 Note that specifying this option does not affect 264 dynamic table installation which will install SSDT 265 tables to /sys/firmware/acpi/tables/dynamic. 266 267 acpi_no_watchdog [HW,ACPI,WDT] 268 Ignore the ACPI-based watchdog interface (WDAT) and let 269 a native driver control the watchdog device instead. 270 271 acpi_rsdp= [ACPI,EFI,KEXEC,EARLY] 272 Pass the RSDP address to the kernel, mostly used 273 on machines running EFI runtime service to boot the 274 second kernel for kdump. 275 276 acpi_os_name= [HW,ACPI] Tell ACPI BIOS the name of the OS 277 Format: To spoof as Windows 98: ="Microsoft Windows" 278 279 acpi_rev_override [ACPI] Override the _REV object to return 5 (instead 280 of 2 which is mandated by ACPI 6) as the supported ACPI 281 specification revision (when using this switch, it may 282 be necessary to carry out a cold reboot _twice_ in a 283 row to make it take effect on the platform firmware). 284 285 acpi_osi= [HW,ACPI] Modify list of supported OS interface strings 286 acpi_osi="string1" # add string1 287 acpi_osi="!string2" # remove string2 288 acpi_osi=!* # remove all strings 289 acpi_osi=! # disable all built-in OS vendor 290 strings 291 acpi_osi=!! # enable all built-in OS vendor 292 strings 293 acpi_osi= # disable all strings 294 295 'acpi_osi=!' can be used in combination with single or 296 multiple 'acpi_osi="string1"' to support specific OS 297 vendor string(s). Note that such command can only 298 affect the default state of the OS vendor strings, thus 299 it cannot affect the default state of the feature group 300 strings and the current state of the OS vendor strings, 301 specifying it multiple times through kernel command line 302 is meaningless. This command is useful when one do not 303 care about the state of the feature group strings which 304 should be controlled by the OSPM. 305 Examples: 306 1. 'acpi_osi=! acpi_osi="Windows 2000"' is equivalent 307 to 'acpi_osi="Windows 2000" acpi_osi=!', they all 308 can make '_OSI("Windows 2000")' TRUE. 309 310 'acpi_osi=' cannot be used in combination with other 311 'acpi_osi=' command lines, the _OSI method will not 312 exist in the ACPI namespace. NOTE that such command can 313 only affect the _OSI support state, thus specifying it 314 multiple times through kernel command line is also 315 meaningless. 316 Examples: 317 1. 'acpi_osi=' can make 'CondRefOf(_OSI, Local1)' 318 FALSE. 319 320 'acpi_osi=!*' can be used in combination with single or 321 multiple 'acpi_osi="string1"' to support specific 322 string(s). Note that such command can affect the 323 current state of both the OS vendor strings and the 324 feature group strings, thus specifying it multiple times 325 through kernel command line is meaningful. But it may 326 still not able to affect the final state of a string if 327 there are quirks related to this string. This command 328 is useful when one want to control the state of the 329 feature group strings to debug BIOS issues related to 330 the OSPM features. 331 Examples: 332 1. 'acpi_osi="Module Device" acpi_osi=!*' can make 333 '_OSI("Module Device")' FALSE. 334 2. 'acpi_osi=!* acpi_osi="Module Device"' can make 335 '_OSI("Module Device")' TRUE. 336 3. 'acpi_osi=! acpi_osi=!* acpi_osi="Windows 2000"' is 337 equivalent to 338 'acpi_osi=!* acpi_osi=! acpi_osi="Windows 2000"' 339 and 340 'acpi_osi=!* acpi_osi="Windows 2000" acpi_osi=!', 341 they all will make '_OSI("Windows 2000")' TRUE. 342 343 acpi_pm_good [X86] 344 Override the pmtimer bug detection: force the kernel 345 to assume that this machine's pmtimer latches its value 346 and always returns good values. 347 348 acpi_sci= [HW,ACPI,EARLY] ACPI System Control Interrupt trigger mode 349 Format: { level | edge | high | low } 350 351 acpi_skip_timer_override [HW,ACPI,EARLY] 352 Recognize and ignore IRQ0/pin2 Interrupt Override. 353 For broken nForce2 BIOS resulting in XT-PIC timer. 354 355 acpi_sleep= [HW,ACPI] Sleep options 356 Format: { s3_bios, s3_mode, s3_beep, s4_hwsig, 357 s4_nohwsig, old_ordering, nonvs, 358 sci_force_enable, nobl } 359 See Documentation/power/video.rst for information on 360 s3_bios and s3_mode. 361 s3_beep is for debugging; it makes the PC's speaker beep 362 as soon as the kernel's real-mode entry point is called. 363 s4_hwsig causes the kernel to check the ACPI hardware 364 signature during resume from hibernation, and gracefully 365 refuse to resume if it has changed. This complies with 366 the ACPI specification but not with reality, since 367 Windows does not do this and many laptops do change it 368 on docking. So the default behaviour is to allow resume 369 and simply warn when the signature changes, unless the 370 s4_hwsig option is enabled. 371 s4_nohwsig prevents ACPI hardware signature from being 372 used (or even warned about) during resume. 373 old_ordering causes the ACPI 1.0 ordering of the _PTS 374 control method, with respect to putting devices into 375 low power states, to be enforced (the ACPI 2.0 ordering 376 of _PTS is used by default). 377 nonvs prevents the kernel from saving/restoring the 378 ACPI NVS memory during suspend/hibernation and resume. 379 sci_force_enable causes the kernel to set SCI_EN directly 380 on resume from S1/S3 (which is against the ACPI spec, 381 but some broken systems don't work without it). 382 nobl causes the internal blacklist of systems known to 383 behave incorrectly in some ways with respect to system 384 suspend and resume to be ignored (use wisely). 385 386 acpi_use_timer_override [HW,ACPI,EARLY] 387 Use timer override. For some broken Nvidia NF5 boards 388 that require a timer override, but don't have HPET 389 390 add_efi_memmap [EFI,X86,EARLY] Include EFI memory map in 391 kernel's map of available physical RAM. 392 393 agp= [AGP] 394 { off | try_unsupported } 395 off: disable AGP support 396 try_unsupported: try to drive unsupported chipsets 397 (may crash computer or cause data corruption) 398 399 ALSA [HW,ALSA] 400 See Documentation/sound/alsa-configuration.rst 401 402 alignment= [KNL,ARM] 403 Allow the default userspace alignment fault handler 404 behaviour to be specified. Bit 0 enables warnings, 405 bit 1 enables fixups, and bit 2 sends a segfault. 406 407 align_va_addr= [X86-64] 408 Align virtual addresses by clearing slice [14:12] when 409 allocating a VMA at process creation time. This option 410 gives you up to 3% performance improvement on AMD F15h 411 machines (where it is enabled by default) for a 412 CPU-intensive style benchmark, and it can vary highly in 413 a microbenchmark depending on workload and compiler. 414 415 32: only for 32-bit processes 416 64: only for 64-bit processes 417 on: enable for both 32- and 64-bit processes 418 off: disable for both 32- and 64-bit processes 419 420 alloc_snapshot [FTRACE] 421 Allocate the ftrace snapshot buffer on boot up when the 422 main buffer is allocated. This is handy if debugging 423 and you need to use tracing_snapshot() on boot up, and 424 do not want to use tracing_snapshot_alloc() as it needs 425 to be done where GFP_KERNEL allocations are allowed. 426 427 allow_mismatched_32bit_el0 [ARM64,EARLY] 428 Allow execve() of 32-bit applications and setting of the 429 PER_LINUX32 personality on systems where only a strict 430 subset of the CPUs support 32-bit EL0. When this 431 parameter is present, the set of CPUs supporting 32-bit 432 EL0 is indicated by /sys/devices/system/cpu/aarch32_el0 433 and hot-unplug operations may be restricted. 434 435 See Documentation/arch/arm64/asymmetric-32bit.rst for more 436 information. 437 438 amd_iommu= [HW,X86-64] 439 Pass parameters to the AMD IOMMU driver in the system. 440 Possible values are: 441 fullflush - Deprecated, equivalent to iommu.strict=1 442 off - do not initialize any AMD IOMMU found in 443 the system 444 force_isolation - Force device isolation for all 445 devices. The IOMMU driver is not 446 allowed anymore to lift isolation 447 requirements as needed. This option 448 does not override iommu=pt 449 force_enable - Force enable the IOMMU on platforms known 450 to be buggy with IOMMU enabled. Use this 451 option with care. 452 pgtbl_v1 - Use v1 page table for DMA-API (Default). 453 pgtbl_v2 - Use v2 page table for DMA-API. 454 irtcachedis - Disable Interrupt Remapping Table (IRT) caching. 455 nohugepages - Limit page-sizes used for v1 page-tables 456 to 4 KiB. 457 v2_pgsizes_only - Limit page-sizes used for v1 page-tables 458 to 4KiB/2Mib/1GiB. 459 460 461 amd_iommu_dump= [HW,X86-64] 462 Enable AMD IOMMU driver option to dump the ACPI table 463 for AMD IOMMU. With this option enabled, AMD IOMMU 464 driver will print ACPI tables for AMD IOMMU during 465 IOMMU initialization. 466 467 amd_iommu_intr= [HW,X86-64] 468 Specifies one of the following AMD IOMMU interrupt 469 remapping modes: 470 legacy - Use legacy interrupt remapping mode. 471 vapic - Use virtual APIC mode, which allows IOMMU 472 to inject interrupts directly into guest. 473 This mode requires kvm-amd.avic=1. 474 (Default when IOMMU HW support is present.) 475 476 amd_pstate= [X86,EARLY] 477 disable 478 Do not enable amd_pstate as the default 479 scaling driver for the supported processors 480 passive 481 Use amd_pstate with passive mode as a scaling driver. 482 In this mode autonomous selection is disabled. 483 Driver requests a desired performance level and platform 484 tries to match the same performance level if it is 485 satisfied by guaranteed performance level. 486 active 487 Use amd_pstate_epp driver instance as the scaling driver, 488 driver provides a hint to the hardware if software wants 489 to bias toward performance (0x0) or energy efficiency (0xff) 490 to the CPPC firmware. then CPPC power algorithm will 491 calculate the runtime workload and adjust the realtime cores 492 frequency. 493 guided 494 Activate guided autonomous mode. Driver requests minimum and 495 maximum performance level and the platform autonomously 496 selects a performance level in this range and appropriate 497 to the current workload. 498 499 amd_prefcore= 500 [X86] 501 disable 502 Disable amd-pstate preferred core. 503 504 amd_dynamic_epp= 505 [X86] 506 disable 507 Disable amd-pstate dynamic EPP. 508 enable 509 Enable amd-pstate dynamic EPP. 510 511 amijoy.map= [HW,JOY] Amiga joystick support 512 Map of devices attached to JOY0DAT and JOY1DAT 513 Format: <a>,<b> 514 See also Documentation/input/joydev/joystick.rst 515 516 analog.map= [HW,JOY] Analog joystick and gamepad support 517 Specifies type or capabilities of an analog joystick 518 connected to one of 16 gameports 519 Format: <type1>,<type2>,..<type16> 520 521 apc= [HW,SPARC] 522 Power management functions (SPARCstation-4/5 + deriv.) 523 Format: noidle 524 Disable APC CPU standby support. SPARCstation-Fox does 525 not play well with APC CPU idle - disable it if you have 526 APC and your system crashes randomly. 527 528 apic [APIC,X86-64] Use IO-APIC. Default. 529 530 apic= [APIC,X86,EARLY] Advanced Programmable Interrupt Controller 531 Change the output verbosity while booting 532 Format: { quiet (default) | verbose | debug } 533 Change the amount of debugging information output 534 when initialising the APIC and IO-APIC components. 535 536 apic_extnmi= [APIC,X86,EARLY] External NMI delivery setting 537 Format: { bsp (default) | all | none } 538 bsp: External NMI is delivered only to CPU 0 539 all: External NMIs are broadcast to all CPUs as a 540 backup of CPU 0 541 none: External NMI is masked for all CPUs. This is 542 useful so that a dump capture kernel won't be 543 shot down by NMI 544 545 apicpmtimer Do APIC timer calibration using the pmtimer. Implies 546 apicmaintimer. Useful when your PIT timer is totally 547 broken. 548 549 autoconf= [IPV6] 550 See Documentation/networking/ipv6.rst. 551 552 apm= [APM] Advanced Power Management 553 See header of arch/x86/kernel/apm_32.c. 554 555 apparmor= [APPARMOR] Disable or enable AppArmor at boot time 556 Format: { "0" | "1" } 557 See security/apparmor/Kconfig help text 558 0 -- disable. 559 1 -- enable. 560 Default value is set via kernel config option. 561 562 arm64.no32bit_el0 [ARM64] Unconditionally disable the execution of 563 32 bit applications. 564 565 arm64.nobti [ARM64] Unconditionally disable Branch Target 566 Identification support 567 568 arm64.nogcs [ARM64] Unconditionally disable Guarded Control Stack 569 support 570 571 arm64.nohaft [ARM64] Unconditionally disable Hardware managed Access 572 Flag for Table descriptors support 573 574 arm64.nomops [ARM64] Unconditionally disable Memory Copy and Memory 575 Set instructions support 576 577 arm64.nompam [ARM64] Unconditionally disable Memory Partitioning And 578 Monitoring support 579 580 arm64.nomte [ARM64] Unconditionally disable Memory Tagging Extension 581 support 582 583 arm64.nopauth [ARM64] Unconditionally disable Pointer Authentication 584 support 585 586 arm64.nosme [ARM64] Unconditionally disable Scalable Matrix 587 Extension support 588 589 arm64.nosve [ARM64] Unconditionally disable Scalable Vector 590 Extension support 591 592 ataflop= [HW,M68k] 593 594 atarimouse= [HW,MOUSE] Atari Mouse 595 596 atkbd.extra= [HW] Enable extra LEDs and keys on IBM RapidAccess, 597 EzKey and similar keyboards 598 599 atkbd.reset= [HW] Reset keyboard during initialization 600 601 atkbd.set= [HW] Select keyboard code set 602 Format: <int> (2 = AT (default), 3 = PS/2) 603 604 atkbd.scroll= [HW] Enable scroll wheel on MS Office and similar 605 keyboards 606 607 atkbd.softraw= [HW] Choose between synthetic and real raw mode 608 Format: <bool> (0 = real, 1 = synthetic (default)) 609 610 atkbd.softrepeat= [HW] 611 Use software keyboard repeat 612 613 audit= [KNL] Enable the audit sub-system 614 Format: { "0" | "1" | "off" | "on" } 615 0 | off - kernel audit is disabled and can not be 616 enabled until the next reboot 617 unset - kernel audit is initialized but disabled and 618 will be fully enabled by the userspace auditd. 619 1 | on - kernel audit is initialized and partially 620 enabled, storing at most audit_backlog_limit 621 messages in RAM until it is fully enabled by the 622 userspace auditd. 623 Default: unset 624 625 audit_backlog_limit= [KNL] Set the audit queue size limit. 626 Format: <int> (must be >=0) 627 Default: 64 628 629 bau= [X86_UV] Enable the BAU on SGI UV. The default 630 behavior is to disable the BAU (i.e. bau=0). 631 Format: { "0" | "1" } 632 0 - Disable the BAU. 633 1 - Enable the BAU. 634 unset - Disable the BAU. 635 636 bdev_allow_write_mounted= 637 Format: <bool> 638 Control the ability to open a mounted block device 639 for writing, i.e., allow / disallow writes that bypass 640 the FS. This was implemented as a means to prevent 641 fuzzers from crashing the kernel by overwriting the 642 metadata underneath a mounted FS without its awareness. 643 This also prevents destructive formatting of mounted 644 filesystems by naive storage tooling that don't use 645 O_EXCL. Default is Y and can be changed through the 646 Kconfig option CONFIG_BLK_DEV_WRITE_MOUNTED. 647 648 bert_disable [ACPI] 649 Disable BERT OS support on buggy BIOSes. 650 651 bgrt_disable [ACPI,X86,EARLY] 652 Disable BGRT to avoid flickering OEM logo. 653 654 blkdevparts= Manual partition parsing of block device(s) for 655 embedded devices based on command line input. 656 See Documentation/block/cmdline-partition.rst 657 658 boot_delay= [KNL,EARLY] 659 Milliseconds to delay each printk during boot. 660 Only works if CONFIG_BOOT_PRINTK_DELAY is enabled, 661 and you may also have to specify "lpj=". Boot_delay 662 values larger than 10 seconds (10000) are assumed 663 erroneous and ignored. 664 Format: integer 665 666 bootconfig [KNL,EARLY] 667 Extended command line options can be added to an initrd 668 and this will cause the kernel to look for it. 669 670 See Documentation/admin-guide/bootconfig.rst 671 672 bttv.card= [HW,V4L] bttv (bt848 + bt878 based grabber cards) 673 bttv.radio= Most important insmod options are available as 674 kernel args too. 675 bttv.pll= See Documentation/admin-guide/media/bttv.rst 676 bttv.tuner= 677 678 bulk_remove=off [PPC] This parameter disables the use of the pSeries 679 firmware feature for flushing multiple hpte entries 680 at a time. 681 682 c101= [NET] Moxa C101 synchronous serial card 683 684 cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection. 685 Sometimes CPU hardware bugs make them report the cache 686 size incorrectly. The kernel will attempt work arounds 687 to fix known problems, but for some CPUs it is not 688 possible to determine what the correct size should be. 689 This option provides an override for these situations. 690 691 carrier_timeout= 692 [NET] Specifies amount of time (in seconds) that 693 the kernel should wait for a network carrier. By default 694 it waits 120 seconds. 695 696 ca_keys= [KEYS] This parameter identifies a specific key(s) on 697 the system trusted keyring to be used for certificate 698 trust validation. 699 format: { id:<keyid> | builtin } 700 701 cca= [MIPS,EARLY] Override the kernel pages' cache coherency 702 algorithm. Accepted values range from 0 to 7 703 inclusive. See arch/mips/include/asm/pgtable-bits.h 704 for platform specific values (SB1, Loongson3 and 705 others). 706 707 ccw_timeout_log [S390] 708 See Documentation/arch/s390/common_io.rst for details. 709 710 cfi= [X86-64] Set Control Flow Integrity checking features 711 when CONFIG_FINEIBT is enabled. 712 Format: feature[,feature...] 713 Default: auto 714 715 auto: Use FineIBT if IBT available, otherwise kCFI. 716 Under FineIBT, enable "paranoid" mode when 717 FRED is not available. 718 off: Turn off CFI checking. 719 kcfi: Use kCFI (disable FineIBT). 720 fineibt: Use FineIBT (even if IBT not available). 721 norand: Do not re-randomize CFI hashes. 722 paranoid: Add caller hash checking under FineIBT. 723 bhi: Enable register poisoning to stop speculation 724 across FineIBT. (Disabled by default.) 725 warn: Do not enforce CFI checking: warn only. 726 debug: Report CFI initialization details. 727 728 cgroup_disable= [KNL] Disable a particular controller or optional feature 729 Format: {name of the controller(s) or feature(s) to disable} 730 The effects of cgroup_disable=foo are: 731 - foo isn't auto-mounted if you mount all cgroups in 732 a single hierarchy 733 - foo isn't visible as an individually mountable 734 subsystem 735 - if foo is an optional feature then the feature is 736 disabled and corresponding cgroup files are not 737 created 738 {Currently only "memory" controller deal with this and 739 cut the overhead, others just disable the usage. So 740 only cgroup_disable=memory is actually worthy} 741 Specifying "pressure" disables per-cgroup pressure 742 stall information accounting feature 743 744 cgroup_no_v1= [KNL] Disable cgroup controllers and named hierarchies in v1 745 Format: { { controller | "all" | "named" } 746 [,{ controller | "all" | "named" }...] } 747 Like cgroup_disable, but only applies to cgroup v1; 748 the blacklisted controllers remain available in cgroup2. 749 "all" blacklists all controllers and "named" disables 750 named mounts. Specifying both "all" and "named" disables 751 all v1 hierarchies. 752 753 cgroup_v1_proc= [KNL] Show also missing controllers in /proc/cgroups 754 Format: { "true" | "false" } 755 /proc/cgroups lists only v1 controllers by default. 756 This compatibility option enables listing also v2 757 controllers (whose v1 code is not compiled!), so that 758 semi-legacy software can check this file to decide 759 about usage of v2 (sic) controllers. 760 761 cgroup_favordynmods= [KNL] Enable or Disable favordynmods. 762 Format: { "true" | "false" } 763 Defaults to the value of CONFIG_CGROUP_FAVOR_DYNMODS. 764 765 cgroup.memory= [KNL] Pass options to the cgroup memory controller. 766 Format: <string> 767 nosocket -- Disable socket memory accounting. 768 nokmem -- Disable kernel memory accounting. 769 nobpf -- Disable BPF memory accounting. 770 771 check_pages= [MM,EARLY] Enable sanity checking of pages after 772 allocations / before freeing. This adds checks to catch 773 double-frees, use-after-frees, and other sources of 774 page corruption by inspecting page internals (flags, 775 mapcount/refcount, memcg_data, etc.). 776 Format: { "0" | "1" } 777 Default: 0 (1 if CONFIG_DEBUG_VM is set) 778 779 checkreqprot= [SELINUX] Set initial checkreqprot flag value. 780 Format: { "0" | "1" } 781 See security/selinux/Kconfig help text. 782 0 -- check protection applied by kernel (includes 783 any implied execute protection). 784 1 -- check protection requested by application. 785 Default value is set via a kernel config option. 786 Value can be changed at runtime via 787 /sys/fs/selinux/checkreqprot. 788 Setting checkreqprot to 1 is deprecated. 789 790 cio_ignore= [S390] 791 See Documentation/arch/s390/common_io.rst for details. 792 793 clk_ignore_unused 794 [CLK] 795 Prevents the clock framework from automatically gating 796 clocks that have not been explicitly enabled by a Linux 797 device driver but are enabled in hardware at reset or 798 by the bootloader/firmware. Note that this does not 799 force such clocks to be always-on nor does it reserve 800 those clocks in any way. This parameter is useful for 801 debug and development, but should not be needed on a 802 platform with proper driver support. For more 803 information, see Documentation/driver-api/clk.rst. 804 805 clock= [BUGS=X86-32, HW] gettimeofday clocksource override. 806 [Deprecated] 807 Forces specified clocksource (if available) to be used 808 when calculating gettimeofday(). If specified 809 clocksource is not available, it defaults to PIT. 810 Format: { pit | tsc | cyclone | pmtmr } 811 812 clocksource= Override the default clocksource 813 Format: <string> 814 Override the default clocksource and use the clocksource 815 with the name specified. 816 Some clocksource names to choose from, depending on 817 the platform: 818 [all] jiffies (this is the base, fallback clocksource) 819 [ACPI] acpi_pm 820 [ARM] imx_timer1,OSTS,netx_timer,mpu_timer2, 821 pxa_timer,timer3,32k_counter,timer0_1 822 [X86-32] pit,hpet,tsc; 823 scx200_hrt on Geode; cyclone on IBM x440 824 [MIPS] MIPS 825 [PARISC] cr16 826 [S390] tod 827 [SH] SuperH 828 [SPARC64] tick 829 [X86-64] hpet,tsc 830 831 clocksource.arm_arch_timer.evtstrm= 832 [ARM,ARM64,EARLY] 833 Format: <bool> 834 Enable/disable the eventstream feature of the ARM 835 architected timer so that code using WFE-based polling 836 loops can be debugged more effectively on production 837 systems. 838 839 clocksource.verify_n_cpus= [KNL] 840 Limit the number of CPUs checked for clocksources 841 marked with CLOCK_SOURCE_VERIFY_PERCPU that 842 are marked unstable due to excessive skew. 843 A negative value says to check all CPUs, while 844 zero says not to check any. Values larger than 845 nr_cpu_ids are silently truncated to nr_cpu_ids. 846 The actual CPUs are chosen randomly, with 847 no replacement if the same CPU is chosen twice. 848 849 clocksource-wdtest.holdoff= [KNL] 850 Set the time in seconds that the clocksource 851 watchdog test waits before commencing its tests. 852 Defaults to zero when built as a module and to 853 10 seconds when built into the kernel. 854 855 cma=nn[MG]@[start[MG][-end[MG]]] 856 [KNL,CMA,EARLY] 857 Sets the size of kernel global memory area for 858 contiguous memory allocations and optionally the 859 placement constraint by the physical address range of 860 memory allocations. A value of 0 disables CMA 861 altogether. For more information, see 862 kernel/dma/contiguous.c 863 864 cma_pernuma=nn[MG] 865 [KNL,CMA,EARLY] 866 Sets the size of kernel per-numa memory area for 867 contiguous memory allocations. A value of 0 disables 868 per-numa CMA altogether. And If this option is not 869 specified, the default value is 0. 870 With per-numa CMA enabled, DMA users on node nid will 871 first try to allocate buffer from the pernuma area 872 which is located in node nid, if the allocation fails, 873 they will fallback to the global default memory area. 874 875 numa_cma=<node>:nn[MG][,<node>:nn[MG]] 876 [KNL,CMA,EARLY] 877 Sets the size of kernel numa memory area for 878 contiguous memory allocations. It will reserve CMA 879 area for the specified node. 880 881 If it is setup together with upper 'cmd_pernuma=' 882 (unlikely), its size setting takes priority for the 883 specified numa nodes. 884 885 With numa CMA enabled, DMA users on node nid will 886 first try to allocate buffer from the numa area 887 which is located in node nid, if the allocation fails, 888 they will fallback to the global default memory area. 889 890 cmo_free_hint= [PPC] Format: { yes | no } 891 Specify whether pages are marked as being inactive 892 when they are freed. This is used in CMO environments 893 to determine OS memory pressure for page stealing by 894 a hypervisor. 895 Default: yes 896 897 coherent_pool=nn[KMG] [ARM,KNL,EARLY] 898 Sets the size of memory pool for coherent, atomic dma 899 allocations, by default set to 256K. 900 901 condev= [HW,S390] console device 902 conmode= 903 904 con3215_drop= [S390,EARLY] 3215 console drop mode. 905 Format: y|n|Y|N|1|0 906 When set to true, drop data on the 3215 console when 907 the console buffer is full. In this case the 908 operator using a 3270 terminal emulator (for example 909 x3270) does not have to enter the clear key for the 910 console output to advance and the kernel to continue. 911 This leads to a much faster boot time when a 3270 912 terminal emulator is active. If no 3270 terminal 913 emulator is used, this parameter has no effect. 914 915 console= [KNL] Output console device and options. 916 917 tty<n> Use the virtual console device <n>. 918 919 ttyS<n>[,options] 920 ttyUSB0[,options] 921 Use the specified serial port. The options are of 922 the form "bbbbpnf", where "bbbb" is the baud rate, 923 "p" is parity ("n", "o", or "e"), "n" is number of 924 bits, and "f" is flow control ("r" for RTS or 925 omit it). Default is "9600n8". 926 927 See Documentation/admin-guide/serial-console.rst for more 928 information. See 929 Documentation/networking/netconsole.rst for an 930 alternative. 931 932 <DEVNAME>:<n>.<n>[,options] 933 Use the specified serial port on the serial core bus. 934 The addressing uses DEVNAME of the physical serial port 935 device, followed by the serial core controller instance, 936 and the serial port instance. The options are the same 937 as documented for the ttyS addressing above. 938 939 The mapping of the serial ports to the tty instances 940 can be viewed with: 941 942 $ ls -d /sys/bus/serial-base/devices/*:*.*/tty/* 943 /sys/bus/serial-base/devices/00:04:0.0/tty/ttyS0 944 945 In the above example, the console can be addressed with 946 console=00:04:0.0. Note that a console addressed this 947 way will only get added when the related device driver 948 is ready. The use of an earlycon parameter in addition to 949 the console may be desired for console output early on. 950 951 uart[8250],io,<addr>[,options] 952 uart[8250],mmio,<addr>[,options] 953 uart[8250],mmio16,<addr>[,options] 954 uart[8250],mmio32,<addr>[,options] 955 uart[8250],0x<addr>[,options] 956 Start an early, polled-mode console on the 8250/16550 957 UART at the specified I/O port or MMIO address, 958 switching to the matching ttyS device later. 959 MMIO inter-register address stride is either 8-bit 960 (mmio), 16-bit (mmio16), or 32-bit (mmio32). 961 If none of [io|mmio|mmio16|mmio32], <addr> is assumed 962 to be equivalent to 'mmio'. 'options' are specified in 963 the same format described for ttyS above; if unspecified, 964 the h/w is not re-initialized. 965 966 hvc<n> Use the hypervisor console device <n>. This is for 967 both Xen and PowerPC hypervisors. 968 969 { null | "" } 970 Use to disable console output, i.e., to have kernel 971 console messages discarded. 972 This must be the only console= parameter used on the 973 kernel command line. 974 975 If the device connected to the port is not a TTY but a braille 976 device, prepend "brl," before the device type, for instance 977 console=brl,ttyS0 978 For now, only VisioBraille is supported. 979 980 console_msg_format= 981 [KNL] Change console messages format 982 default 983 By default we print messages on consoles in 984 "[time stamp] text\n" format (time stamp may not be 985 printed, depending on CONFIG_PRINTK_TIME or 986 `printk_time' param). 987 syslog 988 Switch to syslog format: "<%u>[time stamp] text\n" 989 IOW, each message will have a facility and loglevel 990 prefix. The format is similar to one used by syslog() 991 syscall, or to executing "dmesg -S --raw" or to reading 992 from /proc/kmsg. 993 994 consoleblank= [KNL] The console blank (screen saver) timeout in 995 seconds. A value of 0 disables the blank timer. 996 Defaults to 0. 997 998 coredump_filter= 999 [KNL] Change the default value for 1000 /proc/<pid>/coredump_filter. 1001 See also Documentation/filesystems/proc.rst. 1002 1003 coresight_cpu_debug.enable 1004 [ARM,ARM64] 1005 Format: <bool> 1006 Enable/disable the CPU sampling based debugging. 1007 0: default value, disable debugging 1008 1: enable debugging at boot time 1009 1010 cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver 1011 Format: 1012 <first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>] 1013 1014 cpuidle.off=1 [CPU_IDLE] 1015 disable the cpuidle sub-system 1016 1017 cpuidle.governor= 1018 [CPU_IDLE] Name of the cpuidle governor to use. 1019 1020 cpufreq.off=1 [CPU_FREQ] 1021 disable the cpufreq sub-system 1022 1023 cpufreq.default_governor= 1024 [CPU_FREQ] Name of the default cpufreq governor or 1025 policy to use. This governor must be registered in the 1026 kernel before the cpufreq driver probes. 1027 1028 cpu_init_udelay=N 1029 [X86,EARLY] Delay for N microsec between assert and de-assert 1030 of APIC INIT to start processors. This delay occurs 1031 on every CPU online, such as boot, and resume from suspend. 1032 Default: 10000 1033 1034 cpuhp.parallel= 1035 [SMP] Enable/disable parallel bringup of secondary CPUs 1036 Format: <bool> 1037 Default is enabled if CONFIG_HOTPLUG_PARALLEL=y. Otherwise 1038 the parameter has no effect. 1039 1040 crash_kexec_post_notifiers 1041 Only jump to kdump kernel after running the panic 1042 notifiers and dumping kmsg. This option increases 1043 the risks of a kdump failure, since some panic 1044 notifiers can make the crashed kernel more unstable. 1045 In configurations where kdump may not be reliable, 1046 running the panic notifiers could allow collecting 1047 more data on dmesg, like stack traces from other CPUS 1048 or extra data dumped by panic_print. Note that some 1049 configurations enable this option unconditionally, 1050 like Hyper-V, PowerPC (fadump) and AMD SEV-SNP. 1051 1052 crashkernel=size[KMG][@offset[KMG]] 1053 [KNL,EARLY] Using kexec, Linux can switch to a 'crash kernel' 1054 upon panic. This parameter reserves the physical 1055 memory region [offset, offset + size] for that kernel 1056 image. If '@offset' is omitted, then a suitable offset 1057 is selected automatically. 1058 [KNL, X86-64, ARM64, RISCV, LoongArch] Select a region 1059 under 4G first, and fall back to reserve region above 1060 4G when '@offset' hasn't been specified. 1061 See Documentation/admin-guide/kdump/kdump.rst for further details. 1062 1063 crashkernel=range1:size1[,range2:size2,...][@offset] 1064 [KNL] Same as above, but depends on the memory 1065 in the running system. The syntax of range is 1066 start-[end] where start and end are both 1067 a memory unit (amount[KMG]). See also 1068 Documentation/admin-guide/kdump/kdump.rst for an example. 1069 1070 crashkernel=size[KMG],high 1071 [KNL, X86-64, ARM64, RISCV, LoongArch] range could be 1072 above 4G. 1073 Allow kernel to allocate physical memory region from top, 1074 so could be above 4G if system have more than 4G ram 1075 installed. Otherwise memory region will be allocated 1076 below 4G, if available. 1077 It will be ignored if crashkernel=X is specified. 1078 crashkernel=size[KMG],low 1079 [KNL, X86-64, ARM64, RISCV, LoongArch] range under 4G. 1080 When crashkernel=X,high is passed, kernel could allocate 1081 physical memory region above 4G, that cause second kernel 1082 crash on system that require some amount of low memory, 1083 e.g. swiotlb requires at least 64M+32K low memory, also 1084 enough extra low memory is needed to make sure DMA buffers 1085 for 32-bit devices won't run out. Kernel would try to allocate 1086 default size of memory below 4G automatically. The default 1087 size is platform dependent. 1088 --> x86: max(swiotlb_size_or_default() + 8MiB, 256MiB) 1089 --> arm64: 128MiB 1090 --> riscv: 128MiB 1091 --> loongarch: 128MiB 1092 This one lets the user specify own low range under 4G 1093 for second kernel instead. 1094 0: to disable low allocation. 1095 It will be ignored when crashkernel=X,high is not used 1096 or memory reserved is below 4G. 1097 crashkernel=size[KMG],cma 1098 [KNL, X86, ARM64, RISCV, PPC] Reserve additional crash 1099 kernel memory from CMA. This reservation is usable by 1100 the first system's userspace memory and kernel movable 1101 allocations (memory balloon, zswap). Pages allocated 1102 from this memory range will not be included in the vmcore 1103 so this should not be used if dumping of userspace memory 1104 is intended and it has to be expected that some movable 1105 kernel pages may be missing from the dump. 1106 1107 A standard crashkernel reservation, as described above, 1108 is still needed to hold the crash kernel and initrd. 1109 1110 This option increases the risk of a kdump failure: DMA 1111 transfers configured by the first kernel may end up 1112 corrupting the second kernel's memory. 1113 1114 This reservation method is intended for systems that 1115 can't afford to sacrifice enough memory for standard 1116 crashkernel reservation and where less reliable and 1117 possibly incomplete kdump is preferable to no kdump at 1118 all. 1119 1120 cryptomgr.notests 1121 [KNL] Disable crypto self-tests 1122 1123 cs89x0_dma= [HW,NET] 1124 Format: <dma> 1125 1126 cs89x0_media= [HW,NET] 1127 Format: { rj45 | aui | bnc } 1128 1129 csdlock_debug= [KNL] Enable or disable debug add-ons of cross-CPU 1130 function call handling. When switched on, 1131 additional debug data is printed to the console 1132 in case a hanging CPU is detected, and that 1133 CPU is pinged again in order to try to resolve 1134 the hang situation. The default value of this 1135 option depends on the CSD_LOCK_WAIT_DEBUG_DEFAULT 1136 Kconfig option. 1137 1138 dasd= [HW,NET] 1139 See header of drivers/s390/block/dasd_devmap.c. 1140 1141 db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port 1142 (one device per port) 1143 Format: <port#>,<type> 1144 See also Documentation/input/devices/joystick-parport.rst 1145 1146 debug [KNL,EARLY] Enable kernel debugging (events log level). 1147 1148 debug_boot_weak_hash 1149 [KNL,EARLY] Enable printing [hashed] pointers early in the 1150 boot sequence. If enabled, we use a weak hash instead 1151 of siphash to hash pointers. Use this option if you are 1152 seeing instances of '(___ptrval___)') and need to see a 1153 value (hashed pointer) instead. Cryptographically 1154 insecure, please do not use on production kernels. 1155 1156 debug_locks_verbose= 1157 [KNL] verbose locking self-tests 1158 Format: <int> 1159 Print debugging info while doing the locking API 1160 self-tests. 1161 Bitmask for the various LOCKTYPE_ tests. Defaults to 0 1162 (no extra messages), setting it to -1 (all bits set) 1163 will print _a_lot_ more information - normally only 1164 useful to lockdep developers. 1165 1166 debug_objects [KNL,EARLY] Enable object debugging 1167 1168 debug_guardpage_minorder= 1169 [KNL,EARLY] When CONFIG_DEBUG_PAGEALLOC is set, this 1170 parameter allows control of the order of pages that will 1171 be intentionally kept free (and hence protected) by the 1172 buddy allocator. Bigger value increase the probability 1173 of catching random memory corruption, but reduce the 1174 amount of memory for normal system use. The maximum 1175 possible value is MAX_PAGE_ORDER/2. Setting this 1176 parameter to 1 or 2 should be enough to identify most 1177 random memory corruption problems caused by bugs in 1178 kernel or driver code when a CPU writes to (or reads 1179 from) a random memory location. Note that there exists 1180 a class of memory corruptions problems caused by buggy 1181 H/W or F/W or by drivers badly programming DMA 1182 (basically when memory is written at bus level and the 1183 CPU MMU is bypassed) which are not detectable by 1184 CONFIG_DEBUG_PAGEALLOC, hence this option will not 1185 help tracking down these problems. 1186 1187 debug_pagealloc= 1188 [KNL,EARLY] When CONFIG_DEBUG_PAGEALLOC is set, this parameter 1189 enables the feature at boot time. By default, it is 1190 disabled and the system will work mostly the same as a 1191 kernel built without CONFIG_DEBUG_PAGEALLOC. 1192 Note: to get most of debug_pagealloc error reports, it's 1193 useful to also enable the page_owner functionality. 1194 on: enable the feature 1195 1196 debugfs= [KNL,EARLY] This parameter enables what is exposed to 1197 userspace and debugfs internal clients. 1198 Format: { on, off } 1199 on: All functions are enabled. 1200 off: Filesystem is not registered and clients 1201 get a -EPERM as result when trying to register files 1202 or directories within debugfs. 1203 This is equivalent of the runtime functionality if 1204 debugfs was not enabled in the kernel at all. 1205 Default value is set in build-time with a kernel configuration. 1206 1207 debugpat [X86] Enable PAT debugging 1208 1209 default_hugepagesz= 1210 [HW] The size of the default HugeTLB page. This is 1211 the size represented by the legacy /proc/ hugepages 1212 APIs. In addition, this is the default hugetlb size 1213 used for shmget(), mmap() and mounting hugetlbfs 1214 filesystems. If not specified, defaults to the 1215 architecture's default huge page size. Huge page 1216 sizes are architecture dependent. See also 1217 Documentation/admin-guide/mm/hugetlbpage.rst. 1218 Format: size[KMG] 1219 1220 deferred_probe_timeout= 1221 [KNL] Debugging option to set a timeout in seconds for 1222 deferred probe to give up waiting on dependencies to 1223 probe. Only specific dependencies (subsystems or 1224 drivers) that have opted in will be ignored. A timeout 1225 of 0 will timeout at the end of initcalls; a negative 1226 value is treated as an infinite timeout value. If the 1227 timeout hasn't expired, it'll be restarted by each 1228 successful driver registration. This option will also 1229 dump out devices still on the deferred probe list after 1230 retrying. 1231 1232 delayacct [KNL] Enable per-task delay accounting 1233 1234 dell_smm_hwmon.ignore_dmi= 1235 [HW] Continue probing hardware even if DMI data 1236 indicates that the driver is running on unsupported 1237 hardware. 1238 1239 dell_smm_hwmon.force= 1240 [HW] Activate driver even if SMM BIOS signature does 1241 not match list of supported models and enable otherwise 1242 blacklisted features. 1243 1244 dell_smm_hwmon.power_status= 1245 [HW] Report power status in /proc/i8k 1246 (disabled by default). 1247 1248 dell_smm_hwmon.restricted= 1249 [HW] Allow controlling fans only if SYS_ADMIN 1250 capability is set. 1251 1252 dell_smm_hwmon.fan_mult= 1253 [HW] Factor to multiply fan speed with. 1254 1255 dell_smm_hwmon.fan_max= 1256 [HW] Maximum configurable fan speed. 1257 1258 dfltcc= [HW,S390] 1259 Format: { on | off | def_only | inf_only | always } 1260 on: s390 zlib hardware support for compression on 1261 level 1 and decompression (default) 1262 off: No s390 zlib hardware support 1263 def_only: s390 zlib hardware support for deflate 1264 only (compression on level 1) 1265 inf_only: s390 zlib hardware support for inflate 1266 only (decompression) 1267 always: Same as 'on' but ignores the selected compression 1268 level always using hardware support (used for debugging) 1269 1270 dhash_entries= [KNL] 1271 Set number of hash buckets for dentry cache. 1272 1273 disable_1tb_segments [PPC,EARLY] 1274 Disables the use of 1TB hash page table segments. This 1275 causes the kernel to fall back to 256MB segments which 1276 can be useful when debugging issues that require an SLB 1277 miss to occur. 1278 1279 disable= [IPV6] 1280 See Documentation/networking/ipv6.rst. 1281 1282 disable_radix [PPC,EARLY] 1283 Disable RADIX MMU mode on POWER9 1284 1285 disable_tlbie [PPC] 1286 Disable TLBIE instruction. Currently does not work 1287 with KVM, with HASH MMU, or with coherent accelerators. 1288 1289 disable_ddw [PPC/PSERIES,EARLY] 1290 Disable Dynamic DMA Window support. Use this 1291 to workaround buggy firmware. 1292 1293 disable_ipv6= [IPV6] 1294 See Documentation/networking/ipv6.rst. 1295 1296 disable_mtrr_cleanup [X86,EARLY] 1297 The kernel tries to adjust MTRR layout from continuous 1298 to discrete, to make X server driver able to add WB 1299 entry later. This parameter disables that. 1300 1301 disable_mtrr_trim [X86, Intel and AMD only,EARLY] 1302 By default the kernel will trim any uncacheable 1303 memory out of your available memory pool based on 1304 MTRR settings. This parameter disables that behavior, 1305 possibly causing your machine to run very slowly. 1306 1307 disable_timer_pin_1 [X86,EARLY] 1308 Disable PIN 1 of APIC timer 1309 Can be useful to work around chipset bugs. 1310 1311 dis_ucode_ldr [X86] Disable the microcode loader. 1312 1313 dma_debug=off If the kernel is compiled with DMA_API_DEBUG support, 1314 this option disables the debugging code at boot. 1315 1316 dma_debug_entries=<number> 1317 This option allows to tune the number of preallocated 1318 entries for DMA-API debugging code. One entry is 1319 required per DMA-API allocation. Use this if the 1320 DMA-API debugging code disables itself because the 1321 architectural default is too low. 1322 1323 dma_debug_driver=<driver_name> 1324 With this option the DMA-API debugging driver 1325 filter feature can be enabled at boot time. Just 1326 pass the driver to filter for as the parameter. 1327 The filter can be disabled or changed to another 1328 driver later using sysfs. 1329 1330 reg_file_data_sampling= 1331 [X86] Controls mitigation for Register File Data 1332 Sampling (RFDS) vulnerability. RFDS is a CPU 1333 vulnerability which may allow userspace to infer 1334 kernel data values previously stored in floating point 1335 registers, vector registers, or integer registers. 1336 RFDS only affects Intel Atom processors. 1337 1338 on: Turns ON the mitigation. 1339 off: Turns OFF the mitigation. 1340 1341 This parameter overrides the compile time default set 1342 by CONFIG_MITIGATION_RFDS. Mitigation cannot be 1343 disabled when other VERW based mitigations (like MDS) 1344 are enabled. In order to disable RFDS mitigation all 1345 VERW based mitigations need to be disabled. 1346 1347 For details see: 1348 Documentation/admin-guide/hw-vuln/reg-file-data-sampling.rst 1349 1350 dm_verity.keyring_unsealed= 1351 [KNL] When set to 1, leave the dm-verity keyring 1352 unsealed after initialization so userspace can 1353 provision keys. Once the keyring is restricted 1354 it becomes active and is searched during signature 1355 verification. 1356 1357 driver_async_probe= [KNL] 1358 List of driver names to be probed asynchronously. * 1359 matches with all driver names. If * is specified, the 1360 rest of the listed driver names are those that will NOT 1361 match the *. 1362 Format: <driver_name1>,<driver_name2>... 1363 1364 drm.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>] 1365 Broken monitors, graphic adapters, KVMs and EDIDless 1366 panels may send no or incorrect EDID data sets. 1367 This parameter allows to specify an EDID data sets 1368 in the /lib/firmware directory that are used instead. 1369 An EDID data set will only be used for a particular 1370 connector, if its name and a colon are prepended to 1371 the EDID name. Each connector may use a unique EDID 1372 data set by separating the files with a comma. An EDID 1373 data set with no connector name will be used for 1374 any connectors not explicitly specified. 1375 1376 dscc4.setup= [NET] 1377 1378 dt_cpu_ftrs= [PPC,EARLY] 1379 Format: {"off" | "known"} 1380 Control how the dt_cpu_ftrs device-tree binding is 1381 used for CPU feature discovery and setup (if it 1382 exists). 1383 off: Do not use it, fall back to legacy cpu table. 1384 known: Do not pass through unknown features to guests 1385 or userspace, only those that the kernel is aware of. 1386 1387 dump_apple_properties [X86] 1388 Dump name and content of EFI device properties on 1389 x86 Macs. Useful for driver authors to determine 1390 what data is available or for reverse-engineering. 1391 1392 dyndbg[="val"] [KNL,DYNAMIC_DEBUG] 1393 <module>.dyndbg[="val"] 1394 Enable debug messages at boot time. See 1395 Documentation/admin-guide/dynamic-debug-howto.rst 1396 for details. 1397 1398 early_ioremap_debug [KNL,EARLY] 1399 Enable debug messages in early_ioremap support. This 1400 is useful for tracking down temporary early mappings 1401 which are not unmapped. 1402 1403 earlycon= [KNL,EARLY] Output early console device and options. 1404 1405 When used with no options, the early console is 1406 determined by stdout-path property in device tree's 1407 chosen node or the ACPI SPCR table if supported by 1408 the platform. 1409 1410 cdns,<addr>[,options] 1411 Start an early, polled-mode console on a Cadence 1412 (xuartps) serial port at the specified address. Only 1413 supported option is baud rate. If baud rate is not 1414 specified, the serial port must already be setup and 1415 configured. 1416 1417 uart[8250],io,<addr>[,options[,uartclk]] 1418 uart[8250],mmio,<addr>[,options[,uartclk]] 1419 uart[8250],mmio32,<addr>[,options[,uartclk]] 1420 uart[8250],mmio32be,<addr>[,options[,uartclk]] 1421 uart[8250],0x<addr>[,options] 1422 Start an early, polled-mode console on the 8250/16550 1423 UART at the specified I/O port or MMIO address. 1424 MMIO inter-register address stride is either 8-bit 1425 (mmio) or 32-bit (mmio32 or mmio32be). 1426 If none of [io|mmio|mmio32|mmio32be], <addr> is assumed 1427 to be equivalent to 'mmio'. 'options' are specified 1428 in the same format described for "console=ttyS<n>"; if 1429 unspecified, the h/w is not initialized. 'uartclk' is 1430 the uart clock frequency; if unspecified, it is set 1431 to 'BASE_BAUD' * 16. 1432 1433 pl011,<addr> 1434 pl011,mmio32,<addr> 1435 Start an early, polled-mode console on a pl011 serial 1436 port at the specified address. The pl011 serial port 1437 must already be setup and configured. Options are not 1438 yet supported. If 'mmio32' is specified, then only 1439 the driver will use only 32-bit accessors to read/write 1440 the device registers. 1441 1442 liteuart,<addr> 1443 Start an early console on a litex serial port at the 1444 specified address. The serial port must already be 1445 setup and configured. Options are not yet supported. 1446 1447 meson,<addr> 1448 Start an early, polled-mode console on a meson serial 1449 port at the specified address. The serial port must 1450 already be setup and configured. Options are not yet 1451 supported. 1452 1453 msm_serial,<addr> 1454 Start an early, polled-mode console on an msm serial 1455 port at the specified address. The serial port 1456 must already be setup and configured. Options are not 1457 yet supported. 1458 1459 msm_serial_dm,<addr> 1460 Start an early, polled-mode console on an msm serial 1461 dm port at the specified address. The serial port 1462 must already be setup and configured. Options are not 1463 yet supported. 1464 1465 owl,<addr> 1466 Start an early, polled-mode console on a serial port 1467 of an Actions Semi SoC, such as S500 or S900, at the 1468 specified address. The serial port must already be 1469 setup and configured. Options are not yet supported. 1470 1471 rda,<addr> 1472 Start an early, polled-mode console on a serial port 1473 of an RDA Micro SoC, such as RDA8810PL, at the 1474 specified address. The serial port must already be 1475 setup and configured. Options are not yet supported. 1476 1477 sbi 1478 Use RISC-V SBI (Supervisor Binary Interface) for early 1479 console. 1480 1481 smh Use ARM semihosting calls for early console. 1482 1483 s3c2410,<addr> 1484 s3c2412,<addr> 1485 s3c2440,<addr> 1486 s3c6400,<addr> 1487 s5pv210,<addr> 1488 exynos4210,<addr> 1489 Use early console provided by serial driver available 1490 on Samsung SoCs, requires selecting proper type and 1491 a correct base address of the selected UART port. The 1492 serial port must already be setup and configured. 1493 Options are not yet supported. 1494 1495 lantiq,<addr> 1496 Start an early, polled-mode console on a lantiq serial 1497 (lqasc) port at the specified address. The serial port 1498 must already be setup and configured. Options are not 1499 yet supported. 1500 1501 lpuart,<addr> 1502 lpuart32,<addr> 1503 Use early console provided by Freescale LP UART driver 1504 found on Freescale Vybrid and QorIQ LS1021A processors. 1505 A valid base address must be provided, and the serial 1506 port must already be setup and configured. 1507 1508 ec_imx21,<addr> 1509 ec_imx6q,<addr> 1510 Start an early, polled-mode, output-only console on the 1511 Freescale i.MX UART at the specified address. The UART 1512 must already be setup and configured. 1513 1514 ar3700_uart,<addr> 1515 Start an early, polled-mode console on the 1516 Armada 3700 serial port at the specified 1517 address. The serial port must already be setup 1518 and configured. Options are not yet supported. 1519 1520 qcom_geni,<addr> 1521 Start an early, polled-mode console on a Qualcomm 1522 Generic Interface (GENI) based serial port at the 1523 specified address. The serial port must already be 1524 setup and configured. Options are not yet supported. 1525 1526 efifb,[options] 1527 Start an early, unaccelerated console on the EFI 1528 memory mapped framebuffer (if available). On cache 1529 coherent non-x86 systems that use system memory for 1530 the framebuffer, pass the 'ram' option so that it is 1531 mapped with the correct attributes. 1532 1533 linflex,<addr> 1534 Use early console provided by Freescale LINFlexD UART 1535 serial driver for NXP S32V234 SoCs. A valid base 1536 address must be provided, and the serial port must 1537 already be setup and configured. 1538 1539 earlyprintk= [X86,SH,ARM,M68k,S390,UM,EARLY] 1540 earlyprintk=vga 1541 earlyprintk=sclp 1542 earlyprintk=xen 1543 earlyprintk=serial[,ttySn[,baudrate]] 1544 earlyprintk=serial[,0x...[,baudrate]] 1545 earlyprintk=ttySn[,baudrate] 1546 earlyprintk=dbgp[debugController#] 1547 earlyprintk=mmio32,membase[,{nocfg|baudrate}] 1548 earlyprintk=pciserial[,force],bus:device.function[,{nocfg|baudrate}] 1549 earlyprintk=xdbc[xhciController#] 1550 earlyprintk=bios 1551 1552 earlyprintk is useful when the kernel crashes before 1553 the normal console is initialized. It is not enabled by 1554 default because it has some cosmetic problems. 1555 1556 Use "nocfg" to skip UART configuration, assume 1557 BIOS/firmware has configured UART correctly. 1558 1559 Append ",keep" to not disable it when the real console 1560 takes over. 1561 1562 Only one of vga, serial, or usb debug port can 1563 be used at a time. 1564 1565 Currently only ttyS0 and ttyS1 may be specified by 1566 name. Other I/O ports may be explicitly specified 1567 on some architectures (x86 and arm at least) by 1568 replacing ttySn with an I/O port address, like this: 1569 earlyprintk=serial,0x1008,115200 1570 You can find the port for a given device in 1571 /proc/tty/driver/serial: 1572 2: uart:ST16650V2 port:00001008 irq:18 ... 1573 1574 Interaction with the standard serial driver is not 1575 very good. 1576 1577 The VGA output is eventually overwritten by 1578 the real console. 1579 1580 The xen option can only be used in Xen domains. 1581 1582 The sclp output can only be used on s390. 1583 1584 The bios output can only be used on SuperH. 1585 1586 The optional "force" to "pciserial" enables use of a 1587 PCI device even when its classcode is not of the 1588 UART class. 1589 1590 edac_report= [HW,EDAC] Control how to report EDAC event 1591 Format: {"on" | "off" | "force"} 1592 on: enable EDAC to report H/W event. May be overridden 1593 by other higher priority error reporting module. 1594 off: disable H/W event reporting through EDAC. 1595 force: enforce the use of EDAC to report H/W event. 1596 default: on. 1597 1598 edd= [EDD] 1599 Format: {"off" | "on" | "skip[mbr]"} 1600 1601 efi= [EFI,EARLY] 1602 Format: { "debug", "disable_early_pci_dma", 1603 "nochunk", "noruntime", "nosoftreserve", 1604 "novamap", "no_disable_early_pci_dma" } 1605 debug: enable misc debug output. 1606 disable_early_pci_dma: disable the busmaster bit on all 1607 PCI bridges while in the EFI boot stub. 1608 nochunk: disable reading files in "chunks" in the EFI 1609 boot stub, as chunking can cause problems with some 1610 firmware implementations. 1611 noruntime : disable EFI runtime services support 1612 nosoftreserve: The EFI_MEMORY_SP (Specific Purpose) 1613 attribute may cause the kernel to reserve the 1614 memory range for a memory mapping driver to 1615 claim. Specify efi=nosoftreserve to disable this 1616 reservation and treat the memory by its base type 1617 (i.e. EFI_CONVENTIONAL_MEMORY / "System RAM"). 1618 novamap: do not call SetVirtualAddressMap(). 1619 no_disable_early_pci_dma: Leave the busmaster bit set 1620 on all PCI bridges while in the EFI boot stub 1621 1622 efi_no_storage_paranoia [EFI,X86,EARLY] 1623 The kernel reserves 5KB of EFI variable storage for 1624 safety, because some UEFI implementation may fail to 1625 boot if there's insufficient space in the EFI variable 1626 storage. 1627 1628 Using this parameter, you can use the 5KB reservation 1629 in the EFI variable storage. 1630 1631 However, Use this parameter only if you are really 1632 sure that your UEFI does sane gc and fulfills the spec 1633 otherwise your board may brick. 1634 1635 efivar_ssdt= [EFI; X86] Name of an EFI variable that contains an SSDT 1636 that is to be dynamically loaded by Linux. If there are 1637 multiple variables with the same name but with different 1638 vendor GUIDs, all of them will be loaded. See 1639 Documentation/admin-guide/acpi/ssdt-overlays.rst for details. 1640 1641 1642 eisa_irq_edge= [PARISC,HW] 1643 See header of drivers/parisc/eisa.c. 1644 1645 ekgdboc= [X86,KGDB,EARLY] Allow early kernel console debugging 1646 Format: ekgdboc=kbd 1647 1648 This is designed to be used in conjunction with 1649 the boot argument: earlyprintk=vga 1650 1651 This parameter works in place of the kgdboc parameter 1652 but can only be used if the backing tty is available 1653 very early in the boot process. For early debugging 1654 via a serial port see kgdboc_earlycon instead. 1655 1656 elfcorehdr=[size[KMG]@]offset[KMG] [PPC,SH,X86,S390,EARLY] 1657 Specifies physical address of start of kernel core 1658 image elf header and optionally the size. Generally 1659 kexec loader will pass this option to capture kernel. 1660 See Documentation/admin-guide/kdump/kdump.rst for details. 1661 1662 enable_mtrr_cleanup [X86,EARLY] 1663 The kernel tries to adjust MTRR layout from continuous 1664 to discrete, to make X server driver able to add WB 1665 entry later. This parameter enables that. 1666 1667 enable_timer_pin_1 [X86] 1668 Enable PIN 1 of APIC timer 1669 Can be useful to work around chipset bugs 1670 (in particular on some ATI chipsets). 1671 The kernel tries to set a reasonable default. 1672 1673 enforcing= [SELINUX] Set initial enforcing status. 1674 Format: {"0" | "1"} 1675 See security/selinux/Kconfig help text. 1676 0 -- permissive (log only, no denials). 1677 1 -- enforcing (deny and log). 1678 Default value is 0. 1679 Value can be changed at runtime via 1680 /sys/fs/selinux/enforce. 1681 1682 erst_disable [ACPI] 1683 Disable Error Record Serialization Table (ERST) 1684 support. 1685 1686 ether= [HW,NET] Ethernet cards parameters 1687 This option is obsoleted by the "netdev=" option, which 1688 has equivalent usage. See its documentation for details. 1689 1690 evm= [EVM] 1691 Format: { "fix" } 1692 Permit 'security.evm' to be updated regardless of 1693 current integrity status. 1694 1695 early_page_ext [KNL,EARLY] Enforces page_ext initialization to earlier 1696 stages so cover more early boot allocations. 1697 Please note that as side effect some optimizations 1698 might be disabled to achieve that (e.g. parallelized 1699 memory initialization is disabled) so the boot process 1700 might take longer, especially on systems with a lot of 1701 memory. Available with CONFIG_PAGE_EXTENSION=y. 1702 1703 failslab= 1704 fail_usercopy= 1705 fail_page_alloc= 1706 fail_skb_realloc= 1707 fail_make_request=[KNL] 1708 General fault injection mechanism. 1709 Format: <interval>,<probability>,<space>,<times> 1710 See also Documentation/fault-injection/. 1711 1712 fb_tunnels= [NET] 1713 Format: { initns | none } 1714 See Documentation/admin-guide/sysctl/net.rst for 1715 fb_tunnels_only_for_init_ns 1716 1717 floppy= [HW] 1718 See Documentation/admin-guide/blockdev/floppy.rst. 1719 1720 forcepae [X86-32] 1721 Forcefully enable Physical Address Extension (PAE). 1722 Many Pentium M systems disable PAE but may have a 1723 functionally usable PAE implementation. 1724 Warning: use of this parameter will taint the kernel 1725 and may cause unknown problems. 1726 1727 fred= [X86-64] 1728 Enable/disable Flexible Return and Event Delivery. 1729 Format: { on | off } 1730 on: enable FRED when it's present, the default setting. 1731 off: disable FRED. 1732 1733 ftrace=[tracer] 1734 [FTRACE] will set and start the specified tracer 1735 as early as possible in order to facilitate early 1736 boot debugging. 1737 1738 ftrace_boot_snapshot 1739 [FTRACE] On boot up, a snapshot will be taken of the 1740 ftrace ring buffer that can be read at: 1741 /sys/kernel/tracing/snapshot. 1742 This is useful if you need tracing information from kernel 1743 boot up that is likely to be overridden by user space 1744 start up functionality. 1745 1746 Optionally, the snapshot can also be defined for a tracing 1747 instance that was created by the trace_instance= command 1748 line parameter. 1749 1750 trace_instance=foo,sched_switch ftrace_boot_snapshot=foo 1751 1752 The above will cause the "foo" tracing instance to trigger 1753 a snapshot at the end of boot up. 1754 1755 ftrace_dump_on_oops[=2(orig_cpu) | =<instance>][,<instance> | 1756 ,<instance>=2(orig_cpu)] 1757 [FTRACE] will dump the trace buffers on oops. 1758 If no parameter is passed, ftrace will dump global 1759 buffers of all CPUs, if you pass 2 or orig_cpu, it 1760 will dump only the buffer of the CPU that triggered 1761 the oops, or the specific instance will be dumped if 1762 its name is passed. Multiple instance dump is also 1763 supported, and instances are separated by commas. Each 1764 instance supports only dump on CPU that triggered the 1765 oops by passing 2 or orig_cpu to it. 1766 1767 ftrace_dump_on_oops=foo=orig_cpu 1768 1769 The above will dump only the buffer of "foo" instance 1770 on CPU that triggered the oops. 1771 1772 ftrace_dump_on_oops,foo,bar=orig_cpu 1773 1774 The above will dump global buffer on all CPUs, the 1775 buffer of "foo" instance on all CPUs and the buffer 1776 of "bar" instance on CPU that triggered the oops. 1777 1778 ftrace_filter=[function-list] 1779 [FTRACE] Limit the functions traced by the function 1780 tracer at boot up. function-list is a comma-separated 1781 list of functions. This list can be changed at run 1782 time by the set_ftrace_filter file in the debugfs 1783 tracing directory. 1784 1785 ftrace_notrace=[function-list] 1786 [FTRACE] Do not trace the functions specified in 1787 function-list. This list can be changed at run time 1788 by the set_ftrace_notrace file in the debugfs 1789 tracing directory. 1790 1791 ftrace_graph_filter=[function-list] 1792 [FTRACE] Limit the top level callers functions traced 1793 by the function graph tracer at boot up. 1794 function-list is a comma-separated list of functions 1795 that can be changed at run time by the 1796 set_graph_function file in the debugfs tracing directory. 1797 1798 ftrace_graph_notrace=[function-list] 1799 [FTRACE] Do not trace from the functions specified in 1800 function-list. This list is a comma-separated list of 1801 functions that can be changed at run time by the 1802 set_graph_notrace file in the debugfs tracing directory. 1803 1804 ftrace_graph_max_depth=<uint> 1805 [FTRACE] Used with the function graph tracer. This is 1806 the max depth it will trace into a function. This value 1807 can be changed at run time by the max_graph_depth file 1808 in the tracefs tracing directory. default: 0 (no limit) 1809 1810 fw_devlink= [KNL,EARLY] Create device links between consumer and supplier 1811 devices by scanning the firmware to infer the 1812 consumer/supplier relationships. This feature is 1813 especially useful when drivers are loaded as modules as 1814 it ensures proper ordering of tasks like device probing 1815 (suppliers first, then consumers), supplier boot state 1816 clean up (only after all consumers have probed), 1817 suspend/resume & runtime PM (consumers first, then 1818 suppliers). 1819 Format: { off | permissive | on | rpm } 1820 off -- Don't create device links from firmware info. 1821 permissive -- Create device links from firmware info 1822 but use it only for ordering boot state clean 1823 up (sync_state() calls). 1824 on -- Create device links from firmware info and use it 1825 to enforce probe and suspend/resume ordering. 1826 rpm -- Like "on", but also use to order runtime PM. 1827 1828 fw_devlink.strict=<bool> 1829 [KNL,EARLY] Treat all inferred dependencies as mandatory 1830 dependencies. This only applies for fw_devlink=on|rpm. 1831 Format: <bool> 1832 1833 fw_devlink.sync_state = 1834 [KNL,EARLY] When all devices that could probe have finished 1835 probing, this parameter controls what to do with 1836 devices that haven't yet received their sync_state() 1837 calls. 1838 Format: { strict | timeout } 1839 strict -- Default. Continue waiting on consumers to 1840 probe successfully. 1841 timeout -- Give up waiting on consumers and call 1842 sync_state() on any devices that haven't yet 1843 received their sync_state() calls after 1844 deferred_probe_timeout has expired or by 1845 late_initcall() if !CONFIG_MODULES. 1846 1847 gamecon.map[2|3]= 1848 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad 1849 support via parallel port (up to 5 devices per port) 1850 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5> 1851 See also Documentation/input/devices/joystick-parport.rst 1852 1853 gamma= [HW,DRM] 1854 1855 gart_fix_e820= [X86-64,EARLY] disable the fix e820 for K8 GART 1856 Format: off | on 1857 default: on 1858 1859 gather_data_sampling= 1860 [X86,INTEL,EARLY] Control the Gather Data Sampling (GDS) 1861 mitigation. 1862 1863 Gather Data Sampling is a hardware vulnerability which 1864 allows unprivileged speculative access to data which was 1865 previously stored in vector registers. 1866 1867 This issue is mitigated by default in updated microcode. 1868 The mitigation may have a performance impact but can be 1869 disabled. On systems without the microcode mitigation 1870 disabling AVX serves as a mitigation. 1871 1872 force: Disable AVX to mitigate systems without 1873 microcode mitigation. No effect if the microcode 1874 mitigation is present. Known to cause crashes in 1875 userspace with buggy AVX enumeration. 1876 1877 off: Disable GDS mitigation. 1878 1879 gbpages [X86] Use GB pages for kernel direct mappings. 1880 1881 gcov_persist= [GCOV] When non-zero (default), profiling data for 1882 kernel modules is saved and remains accessible via 1883 debugfs, even when the module is unloaded/reloaded. 1884 When zero, profiling data is discarded and associated 1885 debugfs files are removed at module unload time. 1886 1887 goldfish [X86] Enable the goldfish android emulator platform. 1888 Don't use this when you are not running on the 1889 android emulator 1890 1891 gpio-mockup.gpio_mockup_ranges 1892 [HW] Sets the ranges of gpiochip of for this device. 1893 Format: <start1>,<end1>,<start2>,<end2>... 1894 gpio-mockup.gpio_mockup_named_lines 1895 [HW] Let the driver know GPIO lines should be named. 1896 1897 gpt [EFI] Forces disk with valid GPT signature but 1898 invalid Protective MBR to be treated as GPT. If the 1899 primary GPT is corrupted, it enables the backup/alternate 1900 GPT to be used instead. 1901 1902 grcan.enable0= [HW] Configuration of physical interface 0. Determines 1903 the "Enable 0" bit of the configuration register. 1904 Format: 0 | 1 1905 Default: 0 1906 grcan.enable1= [HW] Configuration of physical interface 1. Determines 1907 the "Enable 0" bit of the configuration register. 1908 Format: 0 | 1 1909 Default: 0 1910 grcan.select= [HW] Select which physical interface to use. 1911 Format: 0 | 1 1912 Default: 0 1913 grcan.txsize= [HW] Sets the size of the tx buffer. 1914 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0. 1915 Default: 1024 1916 grcan.rxsize= [HW] Sets the size of the rx buffer. 1917 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0. 1918 Default: 1024 1919 1920 hardened_usercopy= 1921 [KNL] Under CONFIG_HARDENED_USERCOPY, whether 1922 hardening is enabled for this boot. Hardened 1923 usercopy checking is used to protect the kernel 1924 from reading or writing beyond known memory 1925 allocation boundaries as a proactive defense 1926 against bounds-checking flaws in the kernel's 1927 copy_to_user()/copy_from_user() interface. 1928 The default is determined by 1929 CONFIG_HARDENED_USERCOPY_DEFAULT_ON. 1930 on Perform hardened usercopy checks. 1931 off Disable hardened usercopy checks. 1932 1933 hardlockup_all_cpu_backtrace= 1934 [KNL] Should the hard-lockup detector generate 1935 backtraces on all cpus. 1936 Format: 0 | 1 1937 1938 hash_pointers= 1939 [KNL,EARLY] 1940 By default, when pointers are printed to the console 1941 or buffers via the %p format string, that pointer is 1942 "hashed", i.e. obscured by hashing the pointer value. 1943 This is a security feature that hides actual kernel 1944 addresses from unprivileged users, but it also makes 1945 debugging the kernel more difficult since unequal 1946 pointers can no longer be compared. The choices are: 1947 Format: { auto | always | never } 1948 Default: auto 1949 1950 auto - Hash pointers unless slab_debug is enabled. 1951 always - Always hash pointers (even if slab_debug is 1952 enabled). 1953 never - Never hash pointers. This option should only 1954 be specified when debugging the kernel. Do 1955 not use on production kernels. The boot 1956 param "no_hash_pointers" is an alias for 1957 this mode. 1958 1959 For controlling hashing dynamically at runtime, 1960 use the "kernel.kptr_restrict" sysctl instead. 1961 1962 hashdist= [KNL,NUMA] Large hashes allocated during boot 1963 are distributed across NUMA nodes. Defaults on 1964 for 64-bit NUMA, off otherwise. 1965 Format: 0 | 1 (for off | on) 1966 1967 hd= [EIDE] (E)IDE hard drive subsystem geometry 1968 Format: <cyl>,<head>,<sect> 1969 1970 hest_disable [ACPI] 1971 Disable Hardware Error Source Table (HEST) support; 1972 corresponding firmware-first mode error processing 1973 logic will be disabled. 1974 1975 hibernate= [HIBERNATION] 1976 noresume Don't check if there's a hibernation image 1977 present during boot. 1978 nocompress Don't compress/decompress hibernation images. 1979 no Disable hibernation and resume. 1980 protect_image Turn on image protection during restoration 1981 (that will set all pages holding image data 1982 during restoration read-only). 1983 1984 hibernate.compressor= [HIBERNATION] Compression algorithm to be 1985 used with hibernation. 1986 Format: { lzo | lz4 } 1987 Default: lzo 1988 1989 lzo: Select LZO compression algorithm to 1990 compress/decompress hibernation image. 1991 1992 lz4: Select LZ4 compression algorithm to 1993 compress/decompress hibernation image. 1994 1995 hibernate.pm_test_delay= 1996 [HIBERNATION] 1997 Sets the number of seconds to remain in a hibernation test 1998 mode before resuming the system (see 1999 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG 2000 is set. Default value is 5. 2001 2002 hibernate_compression_threads= 2003 [HIBERNATION] 2004 Set the number of threads used for compressing or decompressing 2005 hibernation images. 2006 2007 Format: <integer> 2008 Default: 3 2009 Minimum: 1 2010 Example: hibernate_compression_threads=4 2011 2012 highmem=nn[KMG] [KNL,BOOT,EARLY] forces the highmem zone to have an exact 2013 size of <nn>. This works even on boxes that have no 2014 highmem otherwise. This also works to reduce highmem 2015 size on bigger boxes. 2016 2017 highres= [KNL] Enable/disable high resolution timer mode. 2018 Valid parameters: "on", "off" 2019 Default: "on" 2020 2021 hlt [BUGS=ARM,SH] 2022 2023 hostname= [KNL,EARLY] Set the hostname (aka UTS nodename). 2024 Format: <string> 2025 This allows setting the system's hostname during early 2026 startup. This sets the name returned by gethostname. 2027 Using this parameter to set the hostname makes it 2028 possible to ensure the hostname is correctly set before 2029 any userspace processes run, avoiding the possibility 2030 that a process may call gethostname before the hostname 2031 has been explicitly set, resulting in the calling 2032 process getting an incorrect result. The string must 2033 not exceed the maximum allowed hostname length (usually 2034 64 characters) and will be truncated otherwise. 2035 2036 hpet= [X86-32,HPET] option to control HPET usage 2037 Format: { enable (default) | disable | force | 2038 verbose } 2039 disable: disable HPET and use PIT instead 2040 force: allow force enabled of undocumented chips (ICH4, 2041 VIA, nVidia) 2042 verbose: show contents of HPET registers during setup 2043 2044 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET 2045 registers. Default set by CONFIG_HPET_MMAP_DEFAULT. 2046 2047 hugepages= [HW,EARLY] Number of HugeTLB pages to allocate at boot. 2048 If this follows hugepagesz (below), it specifies 2049 the number of pages of hugepagesz to be allocated. 2050 If this is the first HugeTLB parameter on the command 2051 line, it specifies the number of pages to allocate for 2052 the default huge page size. If using node format, the 2053 number of pages to allocate per-node can be specified. 2054 See also Documentation/admin-guide/mm/hugetlbpage.rst. 2055 Format: <integer> or (node format) 2056 <node>:<integer>[,<node>:<integer>] 2057 2058 hugepagesz= 2059 [HW,EARLY] The size of the HugeTLB pages. This is 2060 used in conjunction with hugepages (above) to 2061 allocate huge pages of a specific size at boot. The 2062 pair hugepagesz=X hugepages=Y can be specified once 2063 for each supported huge page size. Huge page sizes 2064 are architecture dependent. See also 2065 Documentation/admin-guide/mm/hugetlbpage.rst. 2066 Format: size[KMG] 2067 2068 hugepage_alloc_threads= 2069 [HW] The number of threads that should be used to 2070 allocate hugepages during boot. This option can be 2071 used to improve system bootup time when allocating 2072 a large amount of huge pages. 2073 The default value is 25% of the available hardware threads. 2074 2075 Note that this parameter only applies to non-gigantic huge pages. 2076 2077 hugetlb_cma= [HW,CMA,EARLY] The size of a CMA area used for allocation 2078 of gigantic hugepages. Or using node format, the size 2079 of a CMA area per node can be specified. 2080 The size can be an absolute value (e.g., 2G) or a 2081 percentage of the total memory or node memory (e.g., 20%). 2082 Percentage-derived sizes are rounded down to a multiple of 2083 the architecture's gigantic hugepage size and may become 2084 zero. 2085 Format: nn[KMGTPE] or nn% or (node format) 2086 <node>:nn[KMGTPE][,<node>:nn[KMGTPE]] or 2087 <node>:nn%[,<node>:nn%] 2088 2089 The size must be a multiple of the gigantic page size. 2090 When using node format, this applies to each per-node size. 2091 Missaligned values are dropped with a warning. 2092 2093 Reserve a CMA area of given size and allocate gigantic 2094 hugepages using the CMA allocator. If enabled, the 2095 boot-time allocation of gigantic hugepages is skipped. 2096 2097 hugetlb_cma_only= 2098 [HW,CMA,EARLY] When allocating new HugeTLB pages, only 2099 try to allocate from the CMA areas. 2100 2101 This option does nothing if hugetlb_cma= is not also 2102 specified. 2103 2104 hugetlb_free_vmemmap= 2105 [KNL] Requires CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP 2106 enabled. 2107 Control if HugeTLB Vmemmap Optimization (HVO) is enabled. 2108 Allows heavy hugetlb users to free up some more 2109 memory (7 * PAGE_SIZE for each 2MB hugetlb page). 2110 Format: { on | off (default) } 2111 2112 on: enable HVO 2113 off: disable HVO 2114 2115 Built with CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP_DEFAULT_ON=y, 2116 the default is on. 2117 2118 Note that the vmemmap pages may be allocated from the added 2119 memory block itself when memory_hotplug.memmap_on_memory is 2120 enabled, those vmemmap pages cannot be optimized even if this 2121 feature is enabled. Other vmemmap pages not allocated from 2122 the added memory block itself do not be affected. 2123 2124 hung_task_panic= 2125 [KNL] Number of hung tasks to trigger kernel panic. 2126 Format: <int> 2127 2128 When set to a non-zero value, a kernel panic will be triggered if 2129 the number of detected hung tasks reaches this value. 2130 2131 0: don't panic 2132 1: panic immediately on first hung task 2133 N: panic after N hung tasks are detected in a single scan 2134 2135 The default value is controlled by the 2136 CONFIG_BOOTPARAM_HUNG_TASK_PANIC build-time option. The value 2137 selected by this boot parameter can be changed later by the 2138 kernel.hung_task_panic sysctl. 2139 2140 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC) 2141 terminal devices. Valid values: 0..8 2142 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs. 2143 If specified, z/VM IUCV HVC accepts connections 2144 from listed z/VM user IDs only. 2145 2146 hv_nopvspin [X86,HYPER_V,EARLY] 2147 Disables the paravirt spinlock optimizations 2148 which allow the hypervisor to 'idle' the guest 2149 on lock contention. 2150 2151 hw_protection= [HW] 2152 Format: reboot | shutdown 2153 2154 Hardware protection action taken on critical events like 2155 overtemperature or imminent voltage loss. 2156 2157 i2c_bus= [HW] Override the default board specific I2C bus speed 2158 or register an additional I2C bus that is not 2159 registered from board initialization code. 2160 Format: 2161 <bus_id>,<clkrate> 2162 2163 i2c_touchscreen_props= [HW,ACPI,X86] 2164 Set device-properties for ACPI-enumerated I2C-attached 2165 touchscreen, to e.g. fix coordinates of upside-down 2166 mounted touchscreens. If you need this option please 2167 submit a drivers/platform/x86/touchscreen_dmi.c patch 2168 adding a DMI quirk for this. 2169 2170 Format: 2171 <ACPI_HW_ID>:<prop_name>=<val>[:prop_name=val][:...] 2172 Where <val> is one of: 2173 Omit "=<val>" entirely Set a boolean device-property 2174 Unsigned number Set a u32 device-property 2175 Anything else Set a string device-property 2176 2177 Examples (split over multiple lines): 2178 i2c_touchscreen_props=GDIX1001:touchscreen-inverted-x: 2179 touchscreen-inverted-y 2180 2181 i2c_touchscreen_props=MSSL1680:touchscreen-size-x=1920: 2182 touchscreen-size-y=1080:touchscreen-inverted-y: 2183 firmware-name=gsl1680-vendor-model.fw:silead,home-button 2184 2185 i8042.debug [HW] Toggle i8042 debug mode 2186 i8042.unmask_kbd_data 2187 [HW] Enable printing of interrupt data from the KBD port 2188 (disabled by default, and as a pre-condition 2189 requires that i8042.debug=1 be enabled) 2190 i8042.direct [HW] Put keyboard port into non-translated mode 2191 i8042.dumbkbd [HW] Pretend that controller can only read data from 2192 keyboard and cannot control its state 2193 (Don't attempt to blink the leds) 2194 i8042.noaux [HW] Don't check for auxiliary (== mouse) port 2195 i8042.nokbd [HW] Don't check/create keyboard port 2196 i8042.noloop [HW] Disable the AUX Loopback command while probing 2197 for the AUX port 2198 i8042.nomux [HW] Don't check presence of an active multiplexing 2199 controller 2200 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX 2201 controllers 2202 i8042.notimeout [HW] Ignore timeout condition signalled by controller 2203 i8042.reset [HW] Reset the controller during init, cleanup and 2204 suspend-to-ram transitions, only during s2r 2205 transitions, or never reset 2206 Format: { 1 | Y | y | 0 | N | n } 2207 1, Y, y: always reset controller 2208 0, N, n: don't ever reset controller 2209 Default: only on s2r transitions on x86; most other 2210 architectures force reset to be always executed 2211 i8042.unlock [HW] Unlock (ignore) the keylock 2212 i8042.kbdreset [HW] Reset device connected to KBD port 2213 i8042.probe_defer 2214 [HW] Allow deferred probing upon i8042 probe errors 2215 2216 i810= [HW,DRM] 2217 2218 i915.invert_brightness= 2219 [DRM] Invert the sense of the variable that is used to 2220 set the brightness of the panel backlight. Normally a 2221 brightness value of 0 indicates backlight switched off, 2222 and the maximum of the brightness value sets the backlight 2223 to maximum brightness. If this parameter is set to 0 2224 (default) and the machine requires it, or this parameter 2225 is set to 1, a brightness value of 0 sets the backlight 2226 to maximum brightness, and the maximum of the brightness 2227 value switches the backlight off. 2228 -1 -- never invert brightness 2229 0 -- machine default 2230 1 -- force brightness inversion 2231 2232 ia32_emulation= [X86-64] 2233 Format: <bool> 2234 When true, allows loading 32-bit programs and executing 32-bit 2235 syscalls, essentially overriding IA32_EMULATION_DEFAULT_DISABLED at 2236 boot time. When false, unconditionally disables IA32 emulation. 2237 2238 2239 idle= [X86,EARLY] 2240 Format: idle=poll, idle=halt, idle=nomwait 2241 2242 idle=poll: Don't do power saving in the idle loop 2243 using HLT, but poll for rescheduling event. This will 2244 make the CPUs eat a lot more power, but may be useful 2245 to get slightly better performance in multiprocessor 2246 benchmarks. It also makes some profiling using 2247 performance counters more accurate. Please note that 2248 on systems with MONITOR/MWAIT support (like Intel 2249 EM64T CPUs) this option has no performance advantage 2250 over the normal idle loop. It may also interact badly 2251 with hyperthreading. 2252 2253 idle=halt: Halt is forced to be used for CPU idle. 2254 In such case C2/C3 won't be used again. 2255 2256 idle=nomwait: Disable mwait for CPU C-states 2257 2258 [ARM64,EARLY] 2259 Format: idle=wfi, idle=yield, idle=nop 2260 2261 idle=wfi: Use the WFI (Wait For Interrupt) hint 2262 instruction in the idle loop. This is the default and 2263 allows the CPU to enter a low-power state until an 2264 interrupt arrives. 2265 2266 idle=yield: Use the YIELD hint instruction instead of 2267 WFI. CPUs supporting simultaneous multi-threading (SMT), 2268 can continue executing another thread when the current 2269 thread reaches the idle loop. This will make the CPUs 2270 eat more power, but may be useful to get slightly better 2271 performance in some applications, since the CPUs will 2272 not enter a low-power state. 2273 2274 idle=nop: Do not execute any idle instruction in the 2275 idle loop. This is useful on platforms where WFI 2276 misbehaves, leading to system instability or loss of CPU 2277 state. This will make the CPUs eat more power, but may 2278 give slightly better performance in some applications, 2279 since the CPUs will not enter a low-power state. 2280 2281 idxd.sva= [HW] 2282 Format: <bool> 2283 Allow force disabling of Shared Virtual Memory (SVA) 2284 support for the idxd driver. By default it is set to 2285 true (1). 2286 2287 idxd.tc_override= [HW] 2288 Format: <bool> 2289 Allow override of default traffic class configuration 2290 for the device. By default it is set to false (0). 2291 2292 ieee754= [MIPS] Select IEEE Std 754 conformance mode 2293 Format: { strict | legacy | 2008 | relaxed | emulated } 2294 Default: strict 2295 2296 Choose which programs will be accepted for execution 2297 based on the IEEE 754 NaN encoding(s) supported by 2298 the FPU and the NaN encoding requested with the value 2299 of an ELF file header flag individually set by each 2300 binary. Hardware implementations are permitted to 2301 support either or both of the legacy and the 2008 NaN 2302 encoding mode. 2303 2304 Available settings are as follows: 2305 strict accept binaries that request a NaN encoding 2306 supported by the FPU 2307 legacy only accept legacy-NaN binaries, if supported 2308 by the FPU 2309 2008 only accept 2008-NaN binaries, if supported 2310 by the FPU 2311 relaxed accept any binaries regardless of whether 2312 supported by the FPU 2313 emulated accept any binaries but enable FPU emulator 2314 if binary mode is unsupported by the FPU. 2315 2316 The FPU emulator is always able to support both NaN 2317 encodings, so if no FPU hardware is present or it has 2318 been disabled with 'nofpu', then the settings of 2319 'legacy' and '2008' strap the emulator accordingly, 2320 'relaxed' straps the emulator for both legacy-NaN and 2321 2008-NaN, whereas 'strict' enables legacy-NaN only on 2322 legacy processors and both NaN encodings on MIPS32 or 2323 MIPS64 CPUs. 2324 2325 The setting for ABS.fmt/NEG.fmt instruction execution 2326 mode generally follows that for the NaN encoding, 2327 except where unsupported by hardware. 2328 2329 ignore_loglevel [KNL,EARLY] 2330 Ignore loglevel setting - this will print /all/ 2331 kernel messages to the console. Useful for debugging. 2332 We also add it as printk module parameter, so users 2333 could change it dynamically, usually by 2334 /sys/module/printk/parameters/ignore_loglevel. 2335 2336 ignore_rlimit_data 2337 Ignore RLIMIT_DATA setting for data mappings, 2338 print warning at first misuse. Can be changed via 2339 /sys/module/kernel/parameters/ignore_rlimit_data. 2340 2341 ihash_entries= [KNL] 2342 Set number of hash buckets for inode cache. 2343 2344 ima_appraise= [IMA] appraise integrity measurements 2345 Format: { "off" | "enforce" | "fix" | "log" } 2346 default: "enforce" 2347 2348 ima_appraise_tcb [IMA] Deprecated. Use ima_policy= instead. 2349 The builtin appraise policy appraises all files 2350 owned by uid=0. 2351 2352 ima_canonical_fmt [IMA] 2353 Use the canonical format for the binary runtime 2354 measurements, instead of host native format. 2355 2356 ima_flush_htable [IMA] 2357 Flush the IMA hash table when deleting all the 2358 staged measurement records, to achieve maximum 2359 memory saving at the cost of having duplicate 2360 records across the staged measurement lists. 2361 2362 ima_hash= [IMA] 2363 Format: { md5 | sha1 | rmd160 | sha256 | sha384 2364 | sha512 | ... } 2365 default: "sha1" 2366 2367 The list of supported hash algorithms is defined 2368 in crypto/hash_info.h. 2369 2370 ima_policy= [IMA] 2371 The builtin policies to load during IMA setup. 2372 Format: "tcb | appraise_tcb | secure_boot | 2373 fail_securely | critical_data" 2374 2375 The "tcb" policy measures all programs exec'd, files 2376 mmap'd for exec, and all files opened with the read 2377 mode bit set by either the effective uid (euid=0) or 2378 uid=0. 2379 2380 The "appraise_tcb" policy appraises the integrity of 2381 all files owned by root. 2382 2383 The "secure_boot" policy appraises the integrity 2384 of files (eg. kexec kernel image, kernel modules, 2385 firmware, policy, etc) based on file signatures. 2386 2387 The "fail_securely" policy forces file signature 2388 verification failure also on privileged mounted 2389 filesystems with the SB_I_UNVERIFIABLE_SIGNATURE 2390 flag. 2391 2392 The "critical_data" policy measures kernel integrity 2393 critical data. 2394 2395 ima_tcb [IMA] Deprecated. Use ima_policy= instead. 2396 Load a policy which meets the needs of the Trusted 2397 Computing Base. This means IMA will measure all 2398 programs exec'd, files mmap'd for exec, and all files 2399 opened for read by uid=0. 2400 2401 ima_template= [IMA] 2402 Select one of defined IMA measurements template formats. 2403 Formats: { "ima" | "ima-ng" | "ima-ngv2" | "ima-sig" | 2404 "ima-sigv2" } 2405 Default: "ima-ng" 2406 2407 ima_template_fmt= 2408 [IMA] Define a custom template format. 2409 Format: { "field1|...|fieldN" } 2410 2411 ima= [IMA] Enable or disable IMA 2412 Format: { "off" | "on" } 2413 Default: "on" 2414 Note that disabling IMA is limited to kdump kernel. 2415 2416 indirect_target_selection= [X86,Intel] Mitigation control for Indirect 2417 Target Selection(ITS) bug in Intel CPUs. Updated 2418 microcode is also required for a fix in IBPB. 2419 2420 on: Enable mitigation (default). 2421 off: Disable mitigation. 2422 force: Force the ITS bug and deploy default 2423 mitigation. 2424 vmexit: Only deploy mitigation if CPU is affected by 2425 guest/host isolation part of ITS. 2426 stuff: Deploy RSB-fill mitigation when retpoline is 2427 also deployed. Otherwise, deploy the default 2428 mitigation. 2429 2430 For details see: 2431 Documentation/admin-guide/hw-vuln/indirect-target-selection.rst 2432 2433 init= [KNL] 2434 Format: <full_path> 2435 Run specified binary instead of /sbin/init as init 2436 process. 2437 2438 initcall_debug [KNL] Trace initcalls as they are executed. Useful 2439 for working out where the kernel is dying during 2440 startup. 2441 2442 initcall_blacklist= [KNL] Do not execute a comma-separated list of 2443 initcall functions. Useful for debugging built-in 2444 modules and initcalls. 2445 2446 initramfs_async= [KNL] 2447 Format: <bool> 2448 Default: 1 2449 This parameter controls whether the initramfs 2450 image is unpacked asynchronously, concurrently 2451 with devices being probed and 2452 initialized. This should normally just work, 2453 but as a debugging aid, one can get the 2454 historical behaviour of the initramfs 2455 unpacking being completed before device_ and 2456 late_ initcalls. 2457 2458 initrd= [BOOT,EARLY] Specify the location of the initial ramdisk 2459 2460 initrdmem= [KNL,EARLY] Specify a physical address and size from which to 2461 load the initrd. If an initrd is compiled in or 2462 specified in the bootparams, it takes priority over this 2463 setting. 2464 Format: ss[KMG],nn[KMG] 2465 Default is 0, 0 2466 2467 init_on_alloc= [MM,EARLY] Fill newly allocated pages and heap objects with 2468 zeroes. 2469 Format: 0 | 1 2470 Default set by CONFIG_INIT_ON_ALLOC_DEFAULT_ON. 2471 2472 init_on_free= [MM,EARLY] Fill freed pages and heap objects with zeroes. 2473 Format: 0 | 1 2474 Default set by CONFIG_INIT_ON_FREE_DEFAULT_ON. 2475 2476 init_pkru= [X86] Specify the default memory protection keys rights 2477 register contents for all processes. 0x55555554 by 2478 default (disallow access to all but pkey 0). Can 2479 override in debugfs after boot. 2480 2481 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver 2482 Format: <irq> 2483 2484 int_pln_enable [X86] Enable power limit notification interrupt 2485 2486 integrity_audit=[IMA] 2487 Format: { "0" | "1" } 2488 0 -- basic integrity auditing messages. (Default) 2489 1 -- additional integrity auditing messages. 2490 2491 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option 2492 on 2493 Enable intel iommu driver. 2494 off 2495 Disable intel iommu driver. 2496 igfx_off [Default Off] 2497 By default, gfx is mapped as normal device. If a gfx 2498 device has a dedicated DMAR unit, the DMAR unit is 2499 bypassed by not enabling DMAR with this option. In 2500 this case, gfx device will use physical address for 2501 DMA. 2502 strict [Default Off] 2503 Deprecated, equivalent to iommu.strict=1. 2504 sp_off [Default Off] 2505 By default, super page will be supported if Intel IOMMU 2506 has the capability. With this option, super page will 2507 not be supported. 2508 sm_on 2509 Enable the Intel IOMMU scalable mode if the hardware 2510 advertises that it has support for the scalable mode 2511 translation. 2512 sm_off 2513 Disallow use of the Intel IOMMU scalable mode. 2514 tboot_noforce [Default Off] 2515 Do not force the Intel IOMMU enabled under tboot. 2516 By default, tboot will force Intel IOMMU on, which 2517 could harm performance of some high-throughput 2518 devices like 40GBit network cards, even if identity 2519 mapping is enabled. 2520 Note that using this option lowers the security 2521 provided by tboot because it makes the system 2522 vulnerable to DMA attacks. 2523 2524 intel_idle.max_cstate= [KNL,HW,ACPI,X86] 2525 0 disables intel_idle and fall back on acpi_idle. 2526 1 to 9 specify maximum depth of C-state. 2527 2528 intel_pstate= [X86,EARLY] 2529 disable 2530 Do not enable intel_pstate as the default 2531 scaling driver for the supported processors 2532 active 2533 Use intel_pstate driver to bypass the scaling 2534 governors layer of cpufreq and provides it own 2535 algorithms for p-state selection. There are two 2536 P-state selection algorithms provided by 2537 intel_pstate in the active mode: powersave and 2538 performance. The way they both operate depends 2539 on whether or not the hardware managed P-states 2540 (HWP) feature has been enabled in the processor 2541 and possibly on the processor model. 2542 passive 2543 Use intel_pstate as a scaling driver, but configure it 2544 to work with generic cpufreq governors (instead of 2545 enabling its internal governor). This mode cannot be 2546 used along with the hardware-managed P-states (HWP) 2547 feature. 2548 force 2549 Enable intel_pstate on systems that prohibit it by default 2550 in favor of acpi-cpufreq. Forcing the intel_pstate driver 2551 instead of acpi-cpufreq may disable platform features, such 2552 as thermal controls and power capping, that rely on ACPI 2553 P-States information being indicated to OSPM and therefore 2554 should be used with caution. This option does not work with 2555 processors that aren't supported by the intel_pstate driver 2556 or on platforms that use pcc-cpufreq instead of acpi-cpufreq. 2557 no_hwp 2558 Do not enable hardware P state control (HWP) 2559 if available. 2560 hwp_only 2561 Only load intel_pstate on systems which support 2562 hardware P state control (HWP) if available. 2563 support_acpi_ppc 2564 Enforce ACPI _PPC performance limits. If the Fixed ACPI 2565 Description Table, specifies preferred power management 2566 profile as "Enterprise Server" or "Performance Server", 2567 then this feature is turned on by default. 2568 per_cpu_perf_limits 2569 Allow per-logical-CPU P-State performance control limits using 2570 cpufreq sysfs interface 2571 no_cas 2572 Do not enable capacity-aware scheduling (CAS) on 2573 hybrid systems 2574 2575 intremap= [X86-64,Intel-IOMMU,EARLY] 2576 on enable Interrupt Remapping (default) 2577 off disable Interrupt Remapping 2578 nosid disable Source ID checking 2579 no_x2apic_optout 2580 BIOS x2APIC opt-out request will be ignored 2581 nopost disable Interrupt Posting 2582 posted_msi 2583 enable MSIs delivered as posted interrupts 2584 2585 iomem= Disable strict checking of access to MMIO memory 2586 strict regions from userspace. 2587 relaxed 2588 2589 iommu= [X86,EARLY] 2590 2591 off 2592 Don't initialize and use any kind of IOMMU. 2593 2594 force 2595 Force the use of the hardware IOMMU even when 2596 it is not actually needed (e.g. because < 3 GB 2597 memory). 2598 2599 noforce 2600 Don't force hardware IOMMU usage when it is not 2601 needed. (default). 2602 2603 merge 2604 Do scatter-gather (SG) merging. Implies "force" 2605 (experimental). 2606 2607 nomerge 2608 Don't do scatter-gather (SG) merging. 2609 2610 biomerge 2611 Do scatter-gather (SG) merging. Implies "force" 2612 (experimental). [same as "merge"] 2613 2614 panic 2615 Always panic when IOMMU overflows. 2616 2617 nopanic 2618 Don't panic on IOMMU overflows. 2619 2620 pt 2621 Use passththrough mode by default 2622 (Equivalent to iommu.passthrough=1) 2623 2624 nopt 2625 Use translated mode for DMA by default 2626 (Equivalent to iommu.passthrough=0) 2627 2628 soft 2629 Use software bounce buffering (SWIOTLB) (default for 2630 Intel machines). This can be used to prevent the usage 2631 of an available hardware IOMMU. 2632 2633 usedac 2634 Use the DAC on VIA PCI bridge 2635 (default: disable the VIA PCI bridge DAC) 2636 2637 AMD Gart HW IOMMU-specific options: (CONFIG_GART_IOMMU) 2638 2639 <size> 2640 Set the size of the remapping area in bytes. 2641 2642 allowed 2643 Overwrite iommu off workarounds for specific chipsets 2644 2645 force 2646 Overwrite iommu off workarounds for specific chipsets 2647 2648 fullflush 2649 Flush IOMMU on each allocation (default). 2650 2651 nofullflush 2652 Don't use IOMMU fullflush. 2653 2654 memaper[=<order>] 2655 Allocate an own aperture over RAM with size 2656 32MB<<order. (default: order=1, i.e. 64MB) 2657 2658 noaperture 2659 Ask the IOMMU not to touch the aperture for AGP. 2660 2661 noagp 2662 Don't initialize the AGP driver and use full aperture. 2663 2664 iommu= [PPC/POWERNV] 2665 2666 nobypass 2667 Disable IOMMU bypass, using IOMMU for PCI devices. 2668 2669 iommu.forcedac= [ARM64,X86,EARLY] Control IOVA allocation for PCI devices. 2670 Format: { "0" | "1" } 2671 0 - Try to allocate a 32-bit DMA address first, before 2672 falling back to the full range if needed. 2673 1 - Allocate directly from the full usable range, 2674 forcing Dual Address Cycle for PCI cards supporting 2675 greater than 32-bit addressing. 2676 2677 iommu.strict= [ARM64,X86,S390,EARLY] Configure TLB invalidation behaviour 2678 Format: { "0" | "1" } 2679 0 - Lazy mode. 2680 Request that DMA unmap operations use deferred 2681 invalidation of hardware TLBs, for increased 2682 throughput at the cost of reduced device isolation. 2683 Will fall back to strict mode if not supported by 2684 the relevant IOMMU driver. 2685 1 - Strict mode. 2686 DMA unmap operations invalidate IOMMU hardware TLBs 2687 synchronously. 2688 unset - Use value of CONFIG_IOMMU_DEFAULT_DMA_{LAZY,STRICT}. 2689 Note: on x86, strict mode specified via one of the 2690 legacy driver-specific options takes precedence. 2691 2692 iommu.passthrough= 2693 [ARM64,X86,EARLY] Configure DMA to bypass the IOMMU by default. 2694 Format: { "0" | "1" } 2695 0 - Use IOMMU translation for DMA. 2696 1 - Bypass the IOMMU for DMA. 2697 unset - Use value of CONFIG_IOMMU_DEFAULT_PASSTHROUGH. 2698 2699 iommu.debug_pagealloc= 2700 [KNL,EARLY] When CONFIG_IOMMU_DEBUG_PAGEALLOC is set, this 2701 parameter enables the feature at boot time. By default, it 2702 is disabled and the system behaves the same way as a kernel 2703 built without CONFIG_IOMMU_DEBUG_PAGEALLOC. 2704 Format: { "0" | "1" } 2705 0 - Sanitizer disabled. 2706 1 - Sanitizer enabled, expect runtime overhead. 2707 2708 io7= [HW] IO7 for Marvel-based Alpha systems 2709 See comment before marvel_specify_io7 in 2710 arch/alpha/kernel/core_marvel.c. 2711 2712 io_delay= [X86,EARLY] I/O delay method 2713 0x80 2714 Standard port 0x80 based delay 2715 0xed 2716 Alternate port 0xed based delay (needed on some systems) 2717 udelay 2718 Simple two microseconds delay 2719 none 2720 No delay 2721 2722 ip= [IP_PNP] 2723 See Documentation/admin-guide/nfs/nfsroot.rst. 2724 2725 ipcmni_extend [KNL,EARLY] Extend the maximum number of unique System V 2726 IPC identifiers from 32,768 to 16,777,216. 2727 2728 ipe.enforce= [IPE] 2729 Format: <bool> 2730 Determine whether IPE starts in permissive (0) or 2731 enforce (1) mode. The default is enforce. 2732 2733 ipe.success_audit= 2734 [IPE] 2735 Format: <bool> 2736 Start IPE with success auditing enabled, emitting 2737 an audit event when a binary is allowed. The default 2738 is 0. 2739 2740 irqaffinity= [SMP] Set the default irq affinity mask 2741 The argument is a cpu list, as described above. 2742 2743 irqchip.gicv2_force_probe= 2744 [ARM,ARM64,EARLY] 2745 Format: <bool> 2746 Force the kernel to look for the second 4kB page 2747 of a GICv2 controller even if the memory range 2748 exposed by the device tree is too small. 2749 2750 irqchip.gicv3_nolpi= 2751 [ARM,ARM64,EARLY] 2752 Force the kernel to ignore the availability of 2753 LPIs (and by consequence ITSs). Intended for system 2754 that use the kernel as a bootloader, and thus want 2755 to let secondary kernels in charge of setting up 2756 LPIs. 2757 2758 irqchip.gicv3_pseudo_nmi= [ARM64,EARLY] 2759 Enables support for pseudo-NMIs in the kernel. This 2760 requires the kernel to be built with 2761 CONFIG_ARM64_PSEUDO_NMI. 2762 2763 irqchip.riscv_imsic_noipi 2764 [RISC-V,EARLY] 2765 Force the kernel to not use IMSIC software injected MSIs 2766 as IPIs. Intended for system where IMSIC is trap-n-emulated, 2767 and thus want to reduce MMIO traps when triggering IPIs 2768 to multiple harts. 2769 2770 irqfixup [HW] 2771 When an interrupt is not handled search all handlers 2772 for it. Intended to get systems with badly broken 2773 firmware running. 2774 2775 irqhandler.duration_warn_us= [KNL] 2776 Warn if an IRQ handler exceeds the specified duration 2777 threshold in microseconds. Useful for identifying 2778 long-running IRQs in the system. 2779 2780 irqpoll [HW] 2781 When an interrupt is not handled search all handlers 2782 for it. Also check all handlers each timer 2783 interrupt. Intended to get systems with badly broken 2784 firmware running. 2785 2786 isapnp= [ISAPNP] 2787 Format: <RDP>,<reset>,<pci_scan>,<verbosity> 2788 2789 isolcpus= [KNL,SMP,ISOL] Isolate a given set of CPUs from disturbance. 2790 Format: [flag-list,]<cpu-list> 2791 2792 Specify one or more CPUs to isolate from disturbances 2793 specified in the flag list (default: domain): 2794 2795 nohz 2796 Disable the tick when a single task runs as well as 2797 disabling other kernel noises like having RCU callbacks 2798 offloaded. This is equivalent to the nohz_full parameter. 2799 2800 A residual 1Hz tick is offloaded to workqueues, which you 2801 need to affine to housekeeping through the global 2802 workqueue's affinity configured via the 2803 /sys/devices/virtual/workqueue/cpumask sysfs file, or 2804 by using the 'domain' flag described below. 2805 2806 NOTE: by default the global workqueue runs on all CPUs, 2807 so to protect individual CPUs the 'cpumask' file has to 2808 be configured manually after bootup. 2809 2810 domain 2811 Isolate from the general SMP balancing and scheduling 2812 algorithms. Note that performing domain isolation this way 2813 is irreversible: it's not possible to bring back a CPU to 2814 the domains once isolated through this boot time 2815 configuration. Use cpusets for a dynamic configuration 2816 which can be altered at runtime. For details see 2817 Documentation/admin-guide/cpu-isolation.rst. 2818 2819 You can move a process onto or off an "isolated" CPU via 2820 the CPU affinity syscalls or cpuset. 2821 <cpu number> begins at 0 and the maximum value is 2822 "number of CPUs in system - 1". 2823 2824 managed_irq 2825 2826 Isolate from being targeted by managed interrupts 2827 which have an interrupt mask containing isolated 2828 CPUs. The affinity of managed interrupts is 2829 handled by the kernel and cannot be changed via 2830 the /proc/irq/* interfaces. 2831 2832 This isolation is best effort and only effective 2833 if the automatically assigned interrupt mask of a 2834 device queue contains isolated and housekeeping 2835 CPUs. If housekeeping CPUs are online then such 2836 interrupts are directed to the housekeeping CPU 2837 so that IO submitted on the housekeeping CPU 2838 cannot disturb the isolated CPU. 2839 2840 If a queue's affinity mask contains only isolated 2841 CPUs then this parameter has no effect on the 2842 interrupt routing decision, though interrupts are 2843 only delivered when tasks running on those 2844 isolated CPUs submit IO. IO submitted on 2845 housekeeping CPUs has no influence on those 2846 queues. 2847 2848 The format of <cpu-list> is described above. 2849 2850 iucv= [HW,NET] 2851 2852 ivrs_ioapic [HW,X86-64] 2853 Provide an override to the IOAPIC-ID<->DEVICE-ID 2854 mapping provided in the IVRS ACPI table. 2855 By default, PCI segment is 0, and can be omitted. 2856 2857 For example, to map IOAPIC-ID decimal 10 to 2858 PCI segment 0x1 and PCI device 00:14.0, 2859 write the parameter as: 2860 ivrs_ioapic=10@0001:00:14.0 2861 2862 Deprecated formats: 2863 * To map IOAPIC-ID decimal 10 to PCI device 00:14.0 2864 write the parameter as: 2865 ivrs_ioapic[10]=00:14.0 2866 * To map IOAPIC-ID decimal 10 to PCI segment 0x1 and 2867 PCI device 00:14.0 write the parameter as: 2868 ivrs_ioapic[10]=0001:00:14.0 2869 2870 ivrs_hpet [HW,X86-64] 2871 Provide an override to the HPET-ID<->DEVICE-ID 2872 mapping provided in the IVRS ACPI table. 2873 By default, PCI segment is 0, and can be omitted. 2874 2875 For example, to map HPET-ID decimal 10 to 2876 PCI segment 0x1 and PCI device 00:14.0, 2877 write the parameter as: 2878 ivrs_hpet=10@0001:00:14.0 2879 2880 Deprecated formats: 2881 * To map HPET-ID decimal 0 to PCI device 00:14.0 2882 write the parameter as: 2883 ivrs_hpet[0]=00:14.0 2884 * To map HPET-ID decimal 10 to PCI segment 0x1 and 2885 PCI device 00:14.0 write the parameter as: 2886 ivrs_ioapic[10]=0001:00:14.0 2887 2888 ivrs_acpihid [HW,X86-64] 2889 Provide an override to the ACPI-HID:UID<->DEVICE-ID 2890 mapping provided in the IVRS ACPI table. 2891 By default, PCI segment is 0, and can be omitted. 2892 2893 For example, to map UART-HID:UID AMD0020:0 to 2894 PCI segment 0x1 and PCI device ID 00:14.5, 2895 write the parameter as: 2896 ivrs_acpihid=AMD0020:0@0001:00:14.5 2897 2898 Deprecated formats: 2899 * To map UART-HID:UID AMD0020:0 to PCI segment is 0, 2900 PCI device ID 00:14.5, write the parameter as: 2901 ivrs_acpihid[00:14.5]=AMD0020:0 2902 * To map UART-HID:UID AMD0020:0 to PCI segment 0x1 and 2903 PCI device ID 00:14.5, write the parameter as: 2904 ivrs_acpihid[0001:00:14.5]=AMD0020:0 2905 2906 js= [HW,JOY] Analog joystick 2907 See Documentation/input/joydev/joystick.rst. 2908 2909 kasan_multi_shot 2910 [KNL] Enforce KASAN (Kernel Address Sanitizer) to print 2911 report on every invalid memory access. Without this 2912 parameter KASAN will print report only for the first 2913 invalid access. 2914 2915 keep_bootcon [KNL,EARLY] 2916 Do not unregister boot console at start. This is only 2917 useful for debugging when something happens in the window 2918 between unregistering the boot console and initializing 2919 the real console. 2920 2921 keepinitrd [HW,ARM] See retain_initrd. 2922 2923 kernelcore= [KNL,X86,PPC,EARLY] 2924 Format: nn[KMGTPE] | nn% | "mirror" 2925 This parameter specifies the amount of memory usable by 2926 the kernel for non-movable allocations. The requested 2927 amount is spread evenly throughout all nodes in the 2928 system as ZONE_NORMAL. The remaining memory is used for 2929 movable memory in its own zone, ZONE_MOVABLE. In the 2930 event, a node is too small to have both ZONE_NORMAL and 2931 ZONE_MOVABLE, kernelcore memory will take priority and 2932 other nodes will have a larger ZONE_MOVABLE. 2933 2934 ZONE_MOVABLE is used for the allocation of pages that 2935 may be reclaimed or moved by the page migration 2936 subsystem. Note that allocations like PTEs-from-HighMem 2937 still use the HighMem zone if it exists, and the Normal 2938 zone if it does not. 2939 2940 It is possible to specify the exact amount of memory in 2941 the form of "nn[KMGTPE]", a percentage of total system 2942 memory in the form of "nn%", or "mirror". If "mirror" 2943 option is specified, mirrored (reliable) memory is used 2944 for non-movable allocations and remaining memory is used 2945 for Movable pages. "nn[KMGTPE]", "nn%", and "mirror" 2946 are exclusive, so you cannot specify multiple forms. 2947 2948 kfence.burst= [MM,KFENCE] The number of additional successive 2949 allocations to be attempted through KFENCE for each 2950 sample interval. 2951 Format: <unsigned integer> 2952 Default: 0 2953 2954 kfence.check_on_panic= 2955 [MM,KFENCE] Whether to check all KFENCE-managed objects' 2956 canaries on panic. 2957 Format: <bool> 2958 Default: false 2959 2960 kfence.deferrable= 2961 [MM,KFENCE] Whether to use a deferrable timer to trigger 2962 allocations. This avoids forcing CPU wake-ups if the 2963 system is idle, at the risk of a less predictable 2964 sample interval. 2965 Format: <bool> 2966 Default: CONFIG_KFENCE_DEFERRABLE 2967 2968 kfence.fault= [MM,KFENCE] Controls the behavior when a KFENCE 2969 error is detected. 2970 report - print the error report and continue (default). 2971 oops - print the error report and oops. 2972 panic - print the error report and panic. 2973 2974 kfence.sample_interval= 2975 [MM,KFENCE] KFENCE's sample interval in milliseconds. 2976 Format: <unsigned integer> 2977 0 - Disable KFENCE. 2978 >0 - Enabled KFENCE with given sample interval. 2979 Default: CONFIG_KFENCE_SAMPLE_INTERVAL 2980 2981 kfence.skip_covered_thresh= 2982 [MM,KFENCE] If pool utilization reaches this threshold 2983 (pool usage%), KFENCE limits currently covered 2984 allocations of the same source from further filling 2985 up the pool. 2986 Format: <unsigned integer> 2987 Default: 75 2988 2989 kgdbdbgp= [KGDB,HW,EARLY] kgdb over EHCI usb debug port. 2990 Format: <Controller#>[,poll interval] 2991 The controller # is the number of the ehci usb debug 2992 port as it is probed via PCI. The poll interval is 2993 optional and is the number seconds in between 2994 each poll cycle to the debug port in case you need 2995 the functionality for interrupting the kernel with 2996 gdb or control-c on the dbgp connection. When 2997 not using this parameter you use sysrq-g to break into 2998 the kernel debugger. 2999 3000 kgdboc= [KGDB,HW] kgdb over consoles. 3001 Requires a tty driver that supports console polling, 3002 or a supported polling keyboard driver (non-usb). 3003 Serial only format: <serial_device>[,baud] 3004 keyboard only format: kbd 3005 keyboard and serial format: kbd,<serial_device>[,baud] 3006 Optional Kernel mode setting: 3007 kms, kbd format: kms,kbd 3008 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud] 3009 3010 kgdboc_earlycon= [KGDB,HW,EARLY] 3011 If the boot console provides the ability to read 3012 characters and can work in polling mode, you can use 3013 this parameter to tell kgdb to use it as a backend 3014 until the normal console is registered. Intended to 3015 be used together with the kgdboc parameter which 3016 specifies the normal console to transition to. 3017 3018 The name of the early console should be specified 3019 as the value of this parameter. Note that the name of 3020 the early console might be different than the tty 3021 name passed to kgdboc. It's OK to leave the value 3022 blank and the first boot console that implements 3023 read() will be picked. 3024 3025 kgdbwait [KGDB,EARLY] Stop kernel execution and enter the 3026 kernel debugger at the earliest opportunity. 3027 3028 kho= [KEXEC,EARLY] 3029 Format: { "0" | "1" | "off" | "on" | "y" | "n" } 3030 Enables or disables Kexec HandOver. 3031 "0" | "off" | "n" - kexec handover is disabled 3032 "1" | "on" | "y" - kexec handover is enabled 3033 3034 kho_scratch= [KEXEC,EARLY] 3035 Format: ll[KMG],mm[KMG],nn[KMG] | nn% 3036 Defines the size of the KHO scratch region. The KHO 3037 scratch regions are physically contiguous memory 3038 ranges that can only be used for non-kernel 3039 allocations. That way, even when memory is heavily 3040 fragmented with handed over memory, the kexeced 3041 kernel will always have enough contiguous ranges to 3042 bootstrap itself. 3043 3044 It is possible to specify the exact amount of 3045 memory in the form of "ll[KMG],mm[KMG],nn[KMG]" 3046 where the first parameter defines the size of a low 3047 memory scratch area, the second parameter defines 3048 the size of a global scratch area and the third 3049 parameter defines the size of additional per-node 3050 scratch areas. The form "nn%" defines scale factor 3051 (in percents) of memory that was used during boot. 3052 3053 kmac= [MIPS] Korina ethernet MAC address. 3054 Configure the RouterBoard 532 series on-chip 3055 Ethernet adapter MAC address. 3056 3057 kmemleak= [KNL,EARLY] Boot-time kmemleak enable/disable 3058 Valid arguments: on, off 3059 Default: on 3060 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y, 3061 the default is off. 3062 3063 kprobe_event=[probe-list] 3064 [FTRACE] Add kprobe events and enable at boot time. 3065 The probe-list is a semicolon delimited list of probe 3066 definitions. Each definition is same as kprobe_events 3067 interface, but the parameters are comma delimited. 3068 For example, to add a kprobe event on vfs_read with 3069 arg1 and arg2, add to the command line; 3070 3071 kprobe_event=p,vfs_read,$arg1,$arg2 3072 3073 See also Documentation/trace/kprobetrace.rst "Kernel 3074 Boot Parameter" section. 3075 3076 kpti= [ARM64,EARLY] Control page table isolation of 3077 user and kernel address spaces. 3078 Default: enabled on cores which need mitigation. 3079 0: force disabled 3080 1: force enabled 3081 3082 kunit.enable= [KUNIT] Enable executing KUnit tests. Requires 3083 CONFIG_KUNIT to be set to be fully enabled. The 3084 default value can be overridden via 3085 KUNIT_DEFAULT_ENABLED. 3086 Default is 1 (enabled) 3087 3088 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs. 3089 Default is 0 (don't ignore, but inject #GP) 3090 3091 kvm.eager_page_split= 3092 [KVM,X86,RISCV] Controls whether or not KVM will try to 3093 proactively split all huge pages during dirty logging. 3094 Eager page splitting reduces interruptions to vCPU 3095 execution by eliminating the write-protection faults 3096 and MMU lock contention that would otherwise be 3097 required to split huge pages lazily. 3098 3099 VM workloads that rarely perform writes or that write 3100 only to a small region of VM memory may benefit from 3101 disabling eager page splitting to allow huge pages to 3102 still be used for reads. 3103 3104 The behavior of eager page splitting depends on whether 3105 KVM_DIRTY_LOG_INITIALLY_SET is enabled or disabled. If 3106 disabled, all huge pages in a memslot will be eagerly 3107 split when dirty logging is enabled on that memslot. If 3108 enabled, eager page splitting will be performed during 3109 the KVM_CLEAR_DIRTY ioctl, and only for the pages being 3110 cleared. 3111 3112 On x86, eager page splitting is only supported when 3113 kvm.tdp_mmu=Y. 3114 3115 On RISCV, eager page splitting is supported by default. 3116 3117 Default is Y (on). 3118 3119 kvm.enable_pmu=[KVM,X86] 3120 If enabled, KVM will virtualize PMU functionality based 3121 on the virtual CPU model defined by userspace. This 3122 can be overridden on a per-VM basis via 3123 KVM_CAP_PMU_CAPABILITY. 3124 3125 If disabled, KVM will not virtualize PMU functionality, 3126 e.g. MSRs, PMCs, PMIs, etc., even if userspace defines 3127 a virtual CPU model that contains PMU assets. 3128 3129 Note, KVM's vPMU support implicitly requires running 3130 with an in-kernel local APIC, e.g. to deliver PMIs to 3131 the guest. Running without an in-kernel local APIC is 3132 not supported, though KVM will allow such a combination 3133 (with severely degraded functionality). 3134 3135 See also enable_mediated_pmu. 3136 3137 Default is Y (on). 3138 3139 kvm.enable_virt_at_load=[KVM,ARM64,LOONGARCH,MIPS,RISCV,X86] 3140 If enabled, KVM will enable virtualization in hardware 3141 when KVM is loaded, and disable virtualization when KVM 3142 is unloaded (if KVM is built as a module). 3143 3144 If disabled, KVM will dynamically enable and disable 3145 virtualization on-demand when creating and destroying 3146 VMs, i.e. on the 0=>1 and 1=>0 transitions of the 3147 number of VMs. 3148 3149 Enabling virtualization at module load avoids potential 3150 latency for creation of the 0=>1 VM, as KVM serializes 3151 virtualization enabling across all online CPUs. The 3152 "cost" of enabling virtualization when KVM is loaded, 3153 is that doing so may interfere with using out-of-tree 3154 hypervisors that want to "own" virtualization hardware. 3155 3156 kvm.enable_vmware_backdoor=[KVM] Support VMware backdoor PV interface. 3157 Default is false (don't support). 3158 3159 kvm.nx_huge_pages= 3160 [KVM] Controls the software workaround for the 3161 X86_BUG_ITLB_MULTIHIT bug. 3162 force : Always deploy workaround. 3163 off : Never deploy workaround. 3164 auto : Deploy workaround based on the presence of 3165 X86_BUG_ITLB_MULTIHIT. 3166 3167 Default is 'auto'. 3168 3169 If the software workaround is enabled for the host, 3170 guests do need not to enable it for nested guests. 3171 3172 kvm.nx_huge_pages_recovery_ratio= 3173 [KVM] Controls how many 4KiB pages are periodically zapped 3174 back to huge pages. 0 disables the recovery, otherwise if 3175 the value is N KVM will zap 1/Nth of the 4KiB pages every 3176 period (see below). The default is 60. 3177 3178 kvm.nx_huge_pages_recovery_period_ms= 3179 [KVM] Controls the time period at which KVM zaps 4KiB pages 3180 back to huge pages. If the value is a non-zero N, KVM will 3181 zap a portion (see ratio above) of the pages every N msecs. 3182 If the value is 0 (the default), KVM will pick a period based 3183 on the ratio, such that a page is zapped after 1 hour on average. 3184 3185 kvm-{amd,intel}.enable_mediated_pmu=[KVM,AMD,INTEL] 3186 If enabled, KVM will provide a mediated virtual PMU, 3187 instead of the default perf-based virtual PMU (if 3188 kvm.enable_pmu is true and PMU is enumerated via the 3189 virtual CPU model). 3190 3191 With a perf-based vPMU, KVM operates as a user of perf, 3192 i.e. emulates guest PMU counters using perf events. 3193 KVM-created perf events are managed by perf as regular 3194 (guest-only) events, e.g. are scheduled in/out, contend 3195 for hardware resources, etc. Using a perf-based vPMU 3196 allows guest and host usage of the PMU to co-exist, but 3197 incurs non-trivial overhead and can result in silently 3198 dropped guest events (due to resource contention). 3199 3200 With a mediated vPMU, hardware PMU state is context 3201 switched around the world switch to/from the guest. 3202 KVM mediates which events the guest can utilize, but 3203 gives the guest direct access to all other PMU assets 3204 when possible (KVM may intercept some accesses if the 3205 virtual CPU model provides a subset of hardware PMU 3206 functionality). Using a mediated vPMU significantly 3207 reduces PMU virtualization overhead and eliminates lost 3208 guest events, but is mutually exclusive with using perf 3209 to profile KVM guests and adds latency to most VM-Exits 3210 (to context switch PMU state). 3211 3212 Default is N (off). 3213 3214 kvm-amd.nested= [KVM,AMD] Control nested virtualization feature in 3215 KVM/SVM. Default is 1 (enabled). 3216 3217 kvm-amd.npt= [KVM,AMD] Control KVM's use of Nested Page Tables, 3218 a.k.a. Two-Dimensional Page Tables. Default is 1 3219 (enabled). Disable by KVM if hardware lacks support 3220 for NPT. 3221 3222 kvm-amd.ciphertext_hiding_asids= 3223 [KVM,AMD] Ciphertext hiding prevents disallowed accesses 3224 to SNP private memory from reading ciphertext. Instead, 3225 reads will see constant default values (0xff). 3226 3227 If ciphertext hiding is enabled, the joint SEV-ES and 3228 SEV-SNP ASID space is partitioned into separate SEV-ES 3229 and SEV-SNP ASID ranges, with the SEV-SNP range being 3230 [1..max_snp_asid] and the SEV-ES range being 3231 (max_snp_asid..min_sev_asid), where min_sev_asid is 3232 enumerated by CPUID.0x.8000_001F[EDX]. 3233 3234 A non-zero value enables SEV-SNP ciphertext hiding and 3235 adjusts the ASID ranges for SEV-ES and SEV-SNP guests. 3236 KVM caps the number of SEV-SNP ASIDs at the maximum 3237 possible value, e.g. specifying -1u will assign all 3238 joint SEV-ES and SEV-SNP ASIDs to SEV-SNP. Note, 3239 assigning all joint ASIDs to SEV-SNP, i.e. configuring 3240 max_snp_asid == min_sev_asid-1, will effectively make 3241 SEV-ES unusable. 3242 3243 kvm-arm.mode= 3244 [KVM,ARM,EARLY] Select one of KVM/arm64's modes of 3245 operation. 3246 3247 none: Forcefully disable KVM. 3248 3249 nvhe: Standard nVHE-based mode, without support for 3250 protected guests. 3251 3252 protected: Mode with support for guests whose state is 3253 kept private from the host, using VHE or 3254 nVHE depending on HW support. 3255 3256 nested: VHE-based mode with support for nested 3257 virtualization. Requires at least ARMv8.4 3258 hardware (with FEAT_NV2). 3259 3260 Defaults to VHE/nVHE based on hardware support. Setting 3261 mode to "protected" will disable kexec and hibernation 3262 for the host. To force nVHE on VHE hardware, add 3263 "arm64_sw.hvhe=0 id_aa64mmfr1.vh=0" to the 3264 command-line. 3265 "nested" and "protected" are experimental and should be 3266 used with extreme caution. 3267 3268 kvm-arm.vgic_v3_group0_trap= 3269 [KVM,ARM,EARLY] Trap guest accesses to GICv3 group-0 3270 system registers 3271 3272 kvm-arm.vgic_v3_group1_trap= 3273 [KVM,ARM,EARLY] Trap guest accesses to GICv3 group-1 3274 system registers 3275 3276 kvm-arm.vgic_v3_common_trap= 3277 [KVM,ARM,EARLY] Trap guest accesses to GICv3 common 3278 system registers 3279 3280 kvm-arm.vgic_v4_enable= 3281 [KVM,ARM,EARLY] Allow use of GICv4 for direct 3282 injection of LPIs. 3283 3284 kvm-arm.wfe_trap_policy= 3285 [KVM,ARM] Control when to set WFE instruction trap for 3286 KVM VMs. Traps are allowed but not guaranteed by the 3287 CPU architecture. 3288 3289 trap: set WFE instruction trap 3290 3291 notrap: clear WFE instruction trap 3292 3293 kvm-arm.wfi_trap_policy= 3294 [KVM,ARM] Control when to set WFI instruction trap for 3295 KVM VMs. Traps are allowed but not guaranteed by the 3296 CPU architecture. 3297 3298 trap: set WFI instruction trap 3299 3300 notrap: clear WFI instruction trap 3301 3302 kvm_cma_resv_ratio=n [PPC,EARLY] 3303 Reserves given percentage from system memory area for 3304 contiguous memory allocation for KVM hash pagetable 3305 allocation. 3306 By default it reserves 5% of total system memory. 3307 Format: <integer> 3308 Default: 5 3309 3310 kvm-intel.ept= [KVM,Intel] Control KVM's use of Extended Page Tables, 3311 a.k.a. Two-Dimensional Page Tables. Default is 1 3312 (enabled). Disable by KVM if hardware lacks support 3313 for EPT. 3314 3315 kvm-intel.emulate_invalid_guest_state= 3316 [KVM,Intel] Control whether to emulate invalid guest 3317 state. Ignored if kvm-intel.enable_unrestricted_guest=1, 3318 as guest state is never invalid for unrestricted 3319 guests. This param doesn't apply to nested guests (L2), 3320 as KVM never emulates invalid L2 guest state. 3321 Default is 1 (enabled). 3322 3323 kvm-intel.flexpriority= 3324 [KVM,Intel] Control KVM's use of FlexPriority feature 3325 (TPR shadow). Default is 1 (enabled). Disable by KVM if 3326 hardware lacks support for it. 3327 3328 kvm-intel.nested= 3329 [KVM,Intel] Control nested virtualization feature in 3330 KVM/VMX. Default is 1 (enabled). 3331 3332 kvm-intel.unrestricted_guest= 3333 [KVM,Intel] Control KVM's use of unrestricted guest 3334 feature (virtualized real and unpaged mode). Default 3335 is 1 (enabled). Disable by KVM if EPT is disabled or 3336 hardware lacks support for it. 3337 3338 kvm-intel.vmentry_l1d_flush=[KVM,Intel] Mitigation for L1 Terminal Fault 3339 CVE-2018-3620. 3340 3341 Valid arguments: never, cond, always 3342 3343 always: L1D cache flush on every VMENTER. 3344 cond: Flush L1D on VMENTER only when the code between 3345 VMEXIT and VMENTER can leak host memory. 3346 never: Disables the mitigation 3347 3348 Default is cond (do L1 cache flush in specific instances) 3349 3350 kvm-intel.vpid= [KVM,Intel] Control KVM's use of Virtual Processor 3351 Identification feature (tagged TLBs). Default is 1 3352 (enabled). Disable by KVM if hardware lacks support 3353 for it. 3354 3355 l1d_flush= [X86,INTEL,EARLY] 3356 Control mitigation for L1D based snooping vulnerability. 3357 3358 Certain CPUs are vulnerable to an exploit against CPU 3359 internal buffers which can forward information to a 3360 disclosure gadget under certain conditions. 3361 3362 In vulnerable processors, the speculatively 3363 forwarded data can be used in a cache side channel 3364 attack, to access data to which the attacker does 3365 not have direct access. 3366 3367 This parameter controls the mitigation. The 3368 options are: 3369 3370 on - enable the interface for the mitigation 3371 3372 l1tf= [X86,EARLY] Control mitigation of the L1TF vulnerability on 3373 affected CPUs 3374 3375 The kernel PTE inversion protection is unconditionally 3376 enabled and cannot be disabled. 3377 3378 full 3379 Provides all available mitigations for the 3380 L1TF vulnerability. Disables SMT and 3381 enables all mitigations in the 3382 hypervisors, i.e. unconditional L1D flush. 3383 3384 SMT control and L1D flush control via the 3385 sysfs interface is still possible after 3386 boot. Hypervisors will issue a warning 3387 when the first VM is started in a 3388 potentially insecure configuration, 3389 i.e. SMT enabled or L1D flush disabled. 3390 3391 full,force 3392 Same as 'full', but disables SMT and L1D 3393 flush runtime control. Implies the 3394 'nosmt=force' command line option. 3395 (i.e. sysfs control of SMT is disabled.) 3396 3397 flush 3398 Leaves SMT enabled and enables the default 3399 hypervisor mitigation, i.e. conditional 3400 L1D flush. 3401 3402 SMT control and L1D flush control via the 3403 sysfs interface is still possible after 3404 boot. Hypervisors will issue a warning 3405 when the first VM is started in a 3406 potentially insecure configuration, 3407 i.e. SMT enabled or L1D flush disabled. 3408 3409 flush,nosmt 3410 3411 Disables SMT and enables the default 3412 hypervisor mitigation. 3413 3414 SMT control and L1D flush control via the 3415 sysfs interface is still possible after 3416 boot. Hypervisors will issue a warning 3417 when the first VM is started in a 3418 potentially insecure configuration, 3419 i.e. SMT enabled or L1D flush disabled. 3420 3421 flush,nowarn 3422 Same as 'flush', but hypervisors will not 3423 warn when a VM is started in a potentially 3424 insecure configuration. 3425 3426 off 3427 Disables hypervisor mitigations and doesn't 3428 emit any warnings. 3429 It also drops the swap size and available 3430 RAM limit restriction on both hypervisor and 3431 bare metal. 3432 3433 Default is 'flush'. 3434 3435 For details see: Documentation/admin-guide/hw-vuln/l1tf.rst 3436 3437 l2cr= [PPC] 3438 3439 l3cr= [PPC] 3440 3441 lapic [X86-32,APIC,EARLY] Enable the local APIC even if BIOS 3442 disabled it. 3443 3444 lapic= [X86,APIC] Do not use TSC deadline 3445 value for LAPIC timer one-shot implementation. Default 3446 back to the programmable timer unit in the LAPIC. 3447 Format: notscdeadline 3448 3449 lapic_timer_c2_ok [X86,APIC,EARLY] trust the local apic timer 3450 in C2 power state. 3451 3452 libata.dma= [LIBATA] DMA control 3453 libata.dma=0 Disable all PATA and SATA DMA 3454 libata.dma=1 PATA and SATA Disk DMA only 3455 libata.dma=2 ATAPI (CDROM) DMA only 3456 libata.dma=4 Compact Flash DMA only 3457 Combinations also work, so libata.dma=3 enables DMA 3458 for disks and CDROMs, but not CFs. 3459 3460 libata.ignore_hpa= [LIBATA] Ignore HPA limit 3461 libata.ignore_hpa=0 keep BIOS limits (default) 3462 libata.ignore_hpa=1 ignore limits, using full disk 3463 3464 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume 3465 when set. 3466 Format: <int> 3467 3468 libata.force= [LIBATA] Force configurations. The format is a comma- 3469 separated list of "[ID:]VAL" where ID is PORT[.DEVICE]. 3470 PORT and DEVICE are decimal numbers matching port, link 3471 or device. Basically, it matches the ATA ID string 3472 printed on console by libata. If the whole ID part is 3473 omitted, the last PORT and DEVICE values are used. If 3474 ID hasn't been specified yet, the configuration applies 3475 to all ports, links and devices. 3476 3477 If only DEVICE is omitted, the parameter applies to 3478 the port and all links and devices behind it. DEVICE 3479 number of 0 either selects the first device or the 3480 first fan-out link behind PMP device. It does not 3481 select the host link. DEVICE number of 15 selects the 3482 host link and device attached to it. 3483 3484 The VAL specifies the configuration to force. As long 3485 as there is no ambiguity, shortcut notation is allowed. 3486 For example, both 1.5 and 1.5G would work for 1.5Gbps. 3487 The following configurations can be forced. 3488 3489 * Cable type: 40c, 80c, short40c, unk, ign or sata. 3490 Any ID with matching PORT is used. 3491 3492 * SATA link speed limit: 1.5Gbps or 3.0Gbps. 3493 3494 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7]. 3495 udma[/][16,25,33,44,66,100,133] notation is also 3496 allowed. 3497 3498 * nohrst, nosrst, norst: suppress hard, soft and both 3499 resets. 3500 3501 * rstonce: only attempt one reset during hot-unplug 3502 link recovery. 3503 3504 * [no]dbdelay: Enable or disable the extra 200ms delay 3505 before debouncing a link PHY and device presence 3506 detection. 3507 3508 * [no]ncq: Turn on or off NCQ. 3509 3510 * [no]ncqtrim: Enable or disable queued DSM TRIM. 3511 3512 * [no]ncqati: Enable or disable NCQ trim on ATI chipset. 3513 3514 * [no]trim: Enable or disable (unqueued) TRIM. 3515 3516 * trim_zero: Indicate that TRIM command zeroes data. 3517 3518 * max_trim_128m: Set 128M maximum trim size limit. 3519 3520 * [no]dma: Turn on or off DMA transfers. 3521 3522 * atapi_dmadir: Enable ATAPI DMADIR bridge support. 3523 3524 * atapi_mod16_dma: Enable the use of ATAPI DMA for 3525 commands that are not a multiple of 16 bytes. 3526 3527 * [no]dmalog: Enable or disable the use of the 3528 READ LOG DMA EXT command to access logs. 3529 3530 * [no]iddevlog: Enable or disable access to the 3531 identify device data log. 3532 3533 * [no]logdir: Enable or disable access to the general 3534 purpose log directory. 3535 3536 * max_sec=<sectors>: Set the transfer size limit, in 3537 number of 512-byte sectors, to the value specified in 3538 <sectors>. The value specified in <sectors> has to be 3539 a non-zero positive integer. 3540 3541 * max_sec_128: Set transfer size limit to 128 sectors. 3542 3543 * max_sec_1024: Set or clear transfer size limit to 3544 1024 sectors. 3545 3546 * max_sec_lba48: Set or clear transfer size limit to 3547 65535 sectors. 3548 3549 * external: Mark port as external (hotplug-capable). 3550 3551 * [no]lpm: Enable or disable link power management. 3552 3553 * [no]setxfer: Indicate if transfer speed mode setting 3554 should be skipped. 3555 3556 * [no]fua: Disable or enable FUA (Force Unit Access) 3557 support for devices supporting this feature. 3558 3559 * dump_id: Dump IDENTIFY data. 3560 3561 * disable: Disable this device. 3562 3563 If there are multiple matching configurations changing 3564 the same attribute, the last one is used. 3565 3566 liveupdate= [KNL,EARLY] 3567 Format: <bool> 3568 Enable Live Update Orchestrator (LUO). 3569 Default: off. 3570 3571 lockd.nlm_grace_period=P [NFS] Assign grace period. 3572 Format: <integer> 3573 3574 lockd.nlm_tcpport=N [NFS] Assign TCP port. 3575 Format: <integer> 3576 3577 lockd.nlm_timeout=T [NFS] Assign timeout value. 3578 Format: <integer> 3579 3580 lockd.nlm_udpport=M [NFS] Assign UDP port. 3581 Format: <integer> 3582 3583 lockdown= [SECURITY,EARLY] 3584 { integrity | confidentiality } 3585 Enable the kernel lockdown feature. If set to 3586 integrity, kernel features that allow userland to 3587 modify the running kernel are disabled. If set to 3588 confidentiality, kernel features that allow userland 3589 to extract confidential information from the kernel 3590 are also disabled. 3591 3592 locktorture.acq_writer_lim= [KNL] 3593 Set the time limit in jiffies for a lock 3594 acquisition. Acquisitions exceeding this limit 3595 will result in a splat once they do complete. 3596 3597 locktorture.bind_readers= [KNL] 3598 Specify the list of CPUs to which the readers are 3599 to be bound. 3600 3601 locktorture.bind_writers= [KNL] 3602 Specify the list of CPUs to which the writers are 3603 to be bound. 3604 3605 locktorture.call_rcu_chains= [KNL] 3606 Specify the number of self-propagating call_rcu() 3607 chains to set up. These are used to ensure that 3608 there is a high probability of an RCU grace period 3609 in progress at any given time. Defaults to 0, 3610 which disables these call_rcu() chains. 3611 3612 locktorture.long_hold= [KNL] 3613 Specify the duration in milliseconds for the 3614 occasional long-duration lock hold time. Defaults 3615 to 100 milliseconds. Select 0 to disable. 3616 3617 locktorture.nested_locks= [KNL] 3618 Specify the maximum lock nesting depth that 3619 locktorture is to exercise, up to a limit of 8 3620 (MAX_NESTED_LOCKS). Specify zero to disable. 3621 Note that this parameter is ineffective on types 3622 of locks that do not support nested acquisition. 3623 3624 locktorture.nreaders_stress= [KNL] 3625 Set the number of locking read-acquisition kthreads. 3626 Defaults to being automatically set based on the 3627 number of online CPUs. 3628 3629 locktorture.nwriters_stress= [KNL] 3630 Set the number of locking write-acquisition kthreads. 3631 3632 locktorture.onoff_holdoff= [KNL] 3633 Set time (s) after boot for CPU-hotplug testing. 3634 3635 locktorture.onoff_interval= [KNL] 3636 Set time (s) between CPU-hotplug operations, or 3637 zero to disable CPU-hotplug testing. 3638 3639 locktorture.rt_boost= [KNL] 3640 Do periodic testing of real-time lock priority 3641 boosting. Select 0 to disable, 1 to boost 3642 only rt_mutex, and 2 to boost unconditionally. 3643 Defaults to 2, which might seem to be an 3644 odd choice, but which should be harmless for 3645 non-real-time spinlocks, due to their disabling 3646 of preemption. Note that non-realtime mutexes 3647 disable boosting. 3648 3649 locktorture.rt_boost_factor= [KNL] 3650 Number that determines how often and for how 3651 long priority boosting is exercised. This is 3652 scaled down by the number of writers, so that the 3653 number of boosts per unit time remains roughly 3654 constant as the number of writers increases. 3655 On the other hand, the duration of each boost 3656 increases with the number of writers. 3657 3658 locktorture.shuffle_interval= [KNL] 3659 Set task-shuffle interval (jiffies). Shuffling 3660 tasks allows some CPUs to go into dyntick-idle 3661 mode during the locktorture test. 3662 3663 locktorture.shutdown_secs= [KNL] 3664 Set time (s) after boot system shutdown. This 3665 is useful for hands-off automated testing. 3666 3667 locktorture.stat_interval= [KNL] 3668 Time (s) between statistics printk()s. 3669 3670 locktorture.stutter= [KNL] 3671 Time (s) to stutter testing, for example, 3672 specifying five seconds causes the test to run for 3673 five seconds, wait for five seconds, and so on. 3674 This tests the locking primitive's ability to 3675 transition abruptly to and from idle. 3676 3677 locktorture.torture_type= [KNL] 3678 Specify the locking implementation to test. 3679 3680 locktorture.verbose= [KNL] 3681 Enable additional printk() statements. 3682 3683 locktorture.writer_fifo= [KNL] 3684 Run the write-side locktorture kthreads at 3685 sched_set_fifo() real-time priority. 3686 3687 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver 3688 Format: <irq> 3689 3690 loglevel= [KNL,EARLY] 3691 All Kernel Messages with a loglevel smaller than the 3692 console loglevel will be printed to the console. It can 3693 also be changed with klogd or other programs. The 3694 loglevels are defined as follows: 3695 3696 0 (KERN_EMERG) system is unusable 3697 1 (KERN_ALERT) action must be taken immediately 3698 2 (KERN_CRIT) critical conditions 3699 3 (KERN_ERR) error conditions 3700 4 (KERN_WARNING) warning conditions 3701 5 (KERN_NOTICE) normal but significant condition 3702 6 (KERN_INFO) informational 3703 7 (KERN_DEBUG) debug-level messages 3704 3705 log_buf_len=n[KMG] [KNL,EARLY] 3706 Sets the size of the printk ring buffer, in bytes. 3707 n must be a power of two and greater than the 3708 minimal size. The minimal size is defined by 3709 LOG_BUF_SHIFT kernel config parameter. There 3710 is also CONFIG_LOG_CPU_MAX_BUF_SHIFT config 3711 parameter that allows to increase the default size 3712 depending on the number of CPUs. See init/Kconfig 3713 for more details. 3714 3715 logo.nologo [FB] Disables display of the built-in Linux logo. 3716 This may be used to provide more screen space for 3717 kernel log messages and is useful when debugging 3718 kernel boot problems. 3719 3720 lp=0 [LP] Specify parallel ports to use, e.g, 3721 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses 3722 lp=reset first parallel port). 'lp=0' disables the 3723 lp=auto printer driver. 'lp=reset' (which can be 3724 specified in addition to the ports) causes 3725 attached printers to be reset. Using 3726 lp=port1,port2,... specifies the parallel ports 3727 to associate lp devices with, starting with 3728 lp0. A port specification may be 'none' to skip 3729 that lp device, or a parport name such as 3730 'parport0'. Specifying 'lp=auto' instead of a 3731 port specification list means that device IDs 3732 from each port should be examined, to see if 3733 an IEEE 1284-compliant printer is attached; if 3734 so, the driver will manage that printer. 3735 See also header of drivers/char/lp.c. 3736 3737 lpj=n [KNL] 3738 Sets loops_per_jiffy to given constant, thus avoiding 3739 time-consuming boot-time autodetection (up to 250 ms per 3740 CPU). 0 enables autodetection (default). To determine 3741 the correct value for your kernel, boot with normal 3742 autodetection and see what value is printed. Note that 3743 on SMP systems the preset will be applied to all CPUs, 3744 which is likely to cause problems if your CPUs need 3745 significantly divergent settings. An incorrect value 3746 will cause delays in the kernel to be wrong, leading to 3747 unpredictable I/O errors and other breakage. Although 3748 unlikely, in the extreme case this might damage your 3749 hardware. 3750 3751 lsm.debug [SECURITY] Enable LSM initialization debugging output. 3752 3753 lsm=lsm1,...,lsmN 3754 [SECURITY] Choose order of LSM initialization. This 3755 overrides CONFIG_LSM, and the "security=" parameter. 3756 3757 machtype= [Loongson] Share the same kernel image file between 3758 different yeeloong laptops. 3759 Example: machtype=lemote-yeeloong-2f-7inch 3760 3761 maxcpus= [SMP,EARLY] Maximum number of processors that an SMP kernel 3762 will bring up during bootup. maxcpus=n : n >= 0 limits 3763 the kernel to bring up 'n' processors. Surely after 3764 bootup you can bring up the other plugged cpu by executing 3765 "echo 1 > /sys/devices/system/cpu/cpuX/online". So maxcpus 3766 only takes effect during system bootup. 3767 While n=0 is a special case, it is equivalent to "nosmp", 3768 which also disables the IO APIC. 3769 3770 max_loop= [LOOP] The number of loop block devices that get 3771 (loop.max_loop) unconditionally pre-created at init time. The default 3772 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead 3773 of statically allocating a predefined number, loop 3774 devices can be requested on-demand with the 3775 /dev/loop-control interface. 3776 3777 mce= [X86-{32,64}] 3778 3779 Please see Documentation/arch/x86/x86_64/machinecheck.rst for sysfs runtime tunables. 3780 3781 off 3782 disable machine check 3783 3784 no_cmci 3785 disable CMCI(Corrected Machine Check Interrupt) that 3786 Intel processor supports. Usually this disablement is 3787 not recommended, but it might be handy if your 3788 hardware is misbehaving. 3789 3790 Note that you'll get more problems without CMCI than 3791 with due to the shared banks, i.e. you might get 3792 duplicated error logs. 3793 3794 dont_log_ce 3795 don't make logs for corrected errors. All events 3796 reported as corrected are silently cleared by OS. This 3797 option will be useful if you have no interest in any 3798 of corrected errors. 3799 3800 ignore_ce 3801 disable features for corrected errors, e.g. 3802 polling timer and CMCI. All events reported as 3803 corrected are not cleared by OS and remained in its 3804 error banks. 3805 3806 Usually this disablement is not recommended, however 3807 if there is an agent checking/clearing corrected 3808 errors (e.g. BIOS or hardware monitoring 3809 applications), conflicting with OS's error handling, 3810 and you cannot deactivate the agent, then this option 3811 will be a help. 3812 3813 no_lmce 3814 do not opt-in to Local MCE delivery. Use legacy method 3815 to broadcast MCEs. 3816 3817 bootlog 3818 enable logging of machine checks left over from 3819 booting. Disabled by default on AMD Fam10h and older 3820 because some BIOS leave bogus ones. 3821 3822 If your BIOS doesn't do that it's a good idea to 3823 enable though to make sure you log even machine check 3824 events that result in a reboot. On Intel systems it is 3825 enabled by default. 3826 3827 nobootlog 3828 disable boot machine check logging. 3829 3830 print_all 3831 print all machine check logs to the console. 3832 3833 monarchtimeout (number) 3834 sets the time in us to wait for other CPUs on machine 3835 checks. 0 to disable. 3836 3837 bios_cmci_threshold 3838 don't overwrite the bios-set CMCI threshold. This boot 3839 option prevents Linux from overwriting the CMCI 3840 threshold set by the bios. Without this option, Linux 3841 always sets the CMCI threshold to 1. Enabling this may 3842 make memory predictive failure analysis less effective 3843 if the bios sets thresholds for memory errors since we 3844 will not see details for all errors. 3845 3846 recovery 3847 force-enable recoverable machine check code paths 3848 3849 Everything else is in sysfs now. 3850 3851 3852 md= [HW] RAID subsystems devices and level 3853 See Documentation/admin-guide/md.rst. 3854 3855 mds= [X86,INTEL,EARLY] 3856 Control mitigation for the Micro-architectural Data 3857 Sampling (MDS) vulnerability. 3858 3859 Certain CPUs are vulnerable to an exploit against CPU 3860 internal buffers which can forward information to a 3861 disclosure gadget under certain conditions. 3862 3863 In vulnerable processors, the speculatively 3864 forwarded data can be used in a cache side channel 3865 attack, to access data to which the attacker does 3866 not have direct access. 3867 3868 This parameter controls the MDS mitigation. The 3869 options are: 3870 3871 full - Enable MDS mitigation on vulnerable CPUs 3872 full,nosmt - Enable MDS mitigation and disable 3873 SMT on vulnerable CPUs 3874 off - Unconditionally disable MDS mitigation 3875 3876 On TAA-affected machines, mds=off can be prevented by 3877 an active TAA mitigation as both vulnerabilities are 3878 mitigated with the same mechanism so in order to disable 3879 this mitigation, you need to specify tsx_async_abort=off 3880 too. 3881 3882 Not specifying this option is equivalent to 3883 mds=full. 3884 3885 For details see: Documentation/admin-guide/hw-vuln/mds.rst 3886 3887 mem=nn[KMG] [HEXAGON,EARLY] Set the memory size. 3888 Must be specified, otherwise memory size will be 0. 3889 3890 mem=nn[KMG] [KNL,BOOT,EARLY] Force usage of a specific amount 3891 of memory Amount of memory to be used in cases 3892 as follows: 3893 3894 1 for test; 3895 2 when the kernel is not able to see the whole system memory; 3896 3 memory that lies after 'mem=' boundary is excluded from 3897 the hypervisor, then assigned to KVM guests. 3898 4 to limit the memory available for kdump kernel. 3899 3900 [ARC,MICROBLAZE] - the limit applies only to low memory, 3901 high memory is not affected. 3902 3903 [ARM64] - only limits memory covered by the linear 3904 mapping. The NOMAP regions are not affected. 3905 3906 [X86] Work as limiting max address. Use together 3907 with memmap= to avoid physical address space collisions. 3908 Without memmap= PCI devices could be placed at addresses 3909 belonging to unused RAM. 3910 3911 Note that this only takes effects during boot time since 3912 in above case 3, memory may need be hot added after boot 3913 if system memory of hypervisor is not sufficient. 3914 3915 mem=nn[KMG]@ss[KMG] 3916 [ARM,MIPS,EARLY] - override the memory layout 3917 reported by firmware. 3918 Define a memory region of size nn[KMG] starting at 3919 ss[KMG]. 3920 Multiple different regions can be specified with 3921 multiple mem= parameters on the command line. 3922 3923 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel 3924 memory. 3925 3926 memblock=debug [KNL,EARLY] Enable memblock debug messages. 3927 3928 memchunk=nn[KMG] 3929 [KNL,SH] Allow user to override the default size for 3930 per-device physically contiguous DMA buffers. 3931 3932 memhp_default_state=online/offline/online_kernel/online_movable 3933 [KNL] Set the initial state for the memory hotplug 3934 onlining policy. If not specified, the default value is 3935 set according to the 3936 CONFIG_MHP_DEFAULT_ONLINE_TYPE kernel config 3937 options. 3938 See Documentation/admin-guide/mm/memory-hotplug.rst. 3939 3940 memmap=exactmap [KNL,X86,EARLY] Enable setting of an exact 3941 E820 memory map, as specified by the user. 3942 Such memmap=exactmap lines can be constructed based on 3943 BIOS output or other requirements. See the memmap=nn@ss 3944 option description. 3945 3946 memmap=nn[KMG]@ss[KMG] 3947 [KNL, X86,MIPS,XTENSA,EARLY] Force usage of a specific region of memory. 3948 Region of memory to be used is from ss to ss+nn. 3949 If @ss[KMG] is omitted, it is equivalent to mem=nn[KMG], 3950 which limits max address to nn[KMG]. 3951 Multiple different regions can be specified, 3952 comma delimited. 3953 Example: 3954 memmap=100M@2G,100M#3G,1G!1024G 3955 3956 memmap=nn[KMG]#ss[KMG] 3957 [KNL,ACPI,EARLY] Mark specific memory as ACPI data. 3958 Region of memory to be marked is from ss to ss+nn. 3959 3960 memmap=nn[KMG]$ss[KMG] 3961 [KNL,ACPI,EARLY] Mark specific memory as reserved. 3962 Region of memory to be reserved is from ss to ss+nn. 3963 Example: Exclude memory from 0x18690000-0x1869ffff 3964 memmap=64K$0x18690000 3965 or 3966 memmap=0x10000$0x18690000 3967 Some bootloaders may need an escape character before '$', 3968 like Grub2, otherwise '$' and the following number 3969 will be eaten. 3970 3971 memmap=nn[KMG]!ss[KMG,EARLY] 3972 [KNL,X86] Mark specific memory as protected. 3973 Region of memory to be used, from ss to ss+nn. 3974 The memory region may be marked as e820 type 12 (0xc) 3975 and is NVDIMM or ADR memory. 3976 3977 memmap=<size>%<offset>-<oldtype>+<newtype> 3978 [KNL,ACPI,EARLY] Convert memory within the specified region 3979 from <oldtype> to <newtype>. If "-<oldtype>" is left 3980 out, the whole region will be marked as <newtype>, 3981 even if previously unavailable. If "+<newtype>" is left 3982 out, matching memory will be removed. Types are 3983 specified as e820 types, e.g., 1 = RAM, 2 = reserved, 3984 3 = ACPI, 12 = PRAM. 3985 3986 memory_corruption_check=0/1 [X86,EARLY] 3987 Some BIOSes seem to corrupt the first 64k of 3988 memory when doing things like suspend/resume. 3989 Setting this option will scan the memory 3990 looking for corruption. Enabling this will 3991 both detect corruption and prevent the kernel 3992 from using the memory being corrupted. 3993 However, it's intended as a diagnostic tool; if 3994 repeatable BIOS-originated corruption always 3995 affects the same memory, you can use memmap= 3996 to prevent the kernel from using that memory. 3997 3998 memory_corruption_check_size=size [X86,EARLY] 3999 By default it checks for corruption in the low 4000 64k, making this memory unavailable for normal 4001 use. Use this parameter to scan for 4002 corruption in more or less memory. 4003 4004 memory_corruption_check_period=seconds [X86,EARLY] 4005 By default it checks for corruption every 60 4006 seconds. Use this parameter to check at some 4007 other rate. 0 disables periodic checking. 4008 4009 memory_hotplug.memmap_on_memory 4010 [KNL,X86,ARM] Boolean flag to enable this feature. 4011 Format: {on | off (default)} 4012 When enabled, runtime hotplugged memory will 4013 allocate its internal metadata (struct pages, 4014 those vmemmap pages cannot be optimized even 4015 if hugetlb_free_vmemmap is enabled) from the 4016 hotadded memory which will allow to hotadd a 4017 lot of memory without requiring additional 4018 memory to do so. 4019 This feature is disabled by default because it 4020 has some implication on large (e.g. GB) 4021 allocations in some configurations (e.g. small 4022 memory blocks). 4023 The state of the flag can be read in 4024 /sys/module/memory_hotplug/parameters/memmap_on_memory. 4025 Note that even when enabled, there are a few cases where 4026 the feature is not effective. 4027 4028 mempool_debug [MM] 4029 Enable mempool debugging. This enables element 4030 poison checking when freeing elements back to the 4031 pool. Useful for debugging mempool corruption. 4032 4033 memtest= [KNL,X86,ARM,M68K,PPC,RISCV,EARLY] Enable memtest 4034 Format: <integer> 4035 default : 0 <disable> 4036 Specifies the number of memtest passes to be 4037 performed. Each pass selects another test 4038 pattern from a given set of patterns. Memtest 4039 fills the memory with this pattern, validates 4040 memory contents and reserves bad memory 4041 regions that are detected. 4042 4043 mem_encrypt= [X86-64] AMD Secure Memory Encryption (SME) control 4044 Valid arguments: on, off 4045 Default: off 4046 mem_encrypt=on: Activate SME 4047 mem_encrypt=off: Do not activate SME 4048 4049 Refer to Documentation/virt/kvm/x86/amd-memory-encryption.rst 4050 for details on when memory encryption can be activated. 4051 4052 mem_sleep_default= [SUSPEND] Default system suspend mode: 4053 s2idle - Suspend-To-Idle 4054 shallow - Power-On Suspend or equivalent (if supported) 4055 deep - Suspend-To-RAM or equivalent (if supported) 4056 See Documentation/admin-guide/pm/sleep-states.rst. 4057 4058 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when 4059 the BIOS has incorrectly applied a workaround. TinyBIOS 4060 version 0.98 is known to be affected, 0.99 fixes the 4061 problem by letting the user disable the workaround. 4062 4063 mga= [HW,DRM] 4064 4065 microcode= [X86] Control the behavior of the microcode loader. 4066 Available options, comma separated: 4067 4068 base_rev=X - with <X> with format: <u32> 4069 Set the base microcode revision of each thread when in 4070 debug mode. 4071 4072 dis_ucode_ldr: disable the microcode loader 4073 4074 force_minrev: 4075 Enable or disable the microcode minimal revision 4076 enforcement for the runtime microcode loader. 4077 4078 mini2440= [ARM,HW,KNL] 4079 Format:[0..2][b][c][t] 4080 Default: "0tb" 4081 MINI2440 configuration specification: 4082 0 - The attached screen is the 3.5" TFT 4083 1 - The attached screen is the 7" TFT 4084 2 - The VGA Shield is attached (1024x768) 4085 Leaving out the screen size parameter will not load 4086 the TFT driver, and the framebuffer will be left 4087 unconfigured. 4088 b - Enable backlight. The TFT backlight pin will be 4089 linked to the kernel VESA blanking code and a GPIO 4090 LED. This parameter is not necessary when using the 4091 VGA shield. 4092 c - Enable the s3c camera interface. 4093 t - Reserved for enabling touchscreen support. The 4094 touchscreen support is not enabled in the mainstream 4095 kernel as of 2.6.30, a preliminary port can be found 4096 in the "bleeding edge" mini2440 support kernel at 4097 https://repo.or.cz/w/linux-2.6/mini2440.git 4098 4099 mitigations= 4100 [X86,PPC,S390,ARM64,EARLY] Control optional mitigations for 4101 CPU vulnerabilities. This is a set of curated, 4102 arch-independent options, each of which is an 4103 aggregation of existing arch-specific options. 4104 4105 Note, "mitigations" is supported if and only if the 4106 kernel was built with CPU_MITIGATIONS=y. 4107 4108 off 4109 Disable all optional CPU mitigations. This 4110 improves system performance, but it may also 4111 expose users to several CPU vulnerabilities. 4112 Equivalent to: if nokaslr then kpti=0 [ARM64] 4113 gather_data_sampling=off [X86] 4114 indirect_target_selection=off [X86] 4115 kvm.nx_huge_pages=off [X86] 4116 l1tf=off [X86] 4117 mds=off [X86] 4118 mmio_stale_data=off [X86] 4119 no_entry_flush [PPC] 4120 no_uaccess_flush [PPC] 4121 nobp=0 [S390] 4122 nopti [X86,PPC] 4123 nospectre_bhb [ARM64] 4124 nospectre_v1 [X86,PPC] 4125 nospectre_v2 [X86,PPC,S390,ARM64] 4126 reg_file_data_sampling=off [X86] 4127 retbleed=off [X86] 4128 spec_rstack_overflow=off [X86] 4129 spec_store_bypass_disable=off [X86,PPC] 4130 spectre_bhi=off [X86] 4131 spectre_v2_user=off [X86] 4132 srbds=off [X86,INTEL] 4133 ssbd=force-off [ARM64] 4134 tsa=off [X86,AMD] 4135 tsx_async_abort=off [X86] 4136 vmscape=off [X86] 4137 4138 Exceptions: 4139 This does not have any effect on 4140 kvm.nx_huge_pages when 4141 kvm.nx_huge_pages=force. 4142 4143 auto (default) 4144 Mitigate all CPU vulnerabilities, but leave SMT 4145 enabled, even if it's vulnerable. This is for 4146 users who don't want to be surprised by SMT 4147 getting disabled across kernel upgrades, or who 4148 have other ways of avoiding SMT-based attacks. 4149 Equivalent to: (default behavior) 4150 4151 auto,nosmt 4152 Mitigate all CPU vulnerabilities, disabling SMT 4153 if needed. This is for users who always want to 4154 be fully mitigated, even if it means losing SMT. 4155 Equivalent to: l1tf=flush,nosmt [X86] 4156 mds=full,nosmt [X86] 4157 tsx_async_abort=full,nosmt [X86] 4158 mmio_stale_data=full,nosmt [X86] 4159 retbleed=auto,nosmt [X86] 4160 4161 [X86] After one of the above options, additionally 4162 supports attack-vector based controls as documented in 4163 Documentation/admin-guide/hw-vuln/attack_vector_controls.rst 4164 4165 mminit_loglevel= 4166 [KNL,EARLY] When CONFIG_DEBUG_MEMORY_INIT is set, this 4167 parameter allows control of the logging verbosity for 4168 the additional memory initialisation checks. A value 4169 of 0 disables mminit logging and a level of 4 will 4170 log everything. Information is printed at KERN_DEBUG 4171 so loglevel=8 may also need to be specified. 4172 4173 mmio_stale_data= 4174 [X86,INTEL,EARLY] Control mitigation for the Processor 4175 MMIO Stale Data vulnerabilities. 4176 4177 Processor MMIO Stale Data is a class of 4178 vulnerabilities that may expose data after an MMIO 4179 operation. Exposed data could originate or end in 4180 the same CPU buffers as affected by MDS and TAA. 4181 Therefore, similar to MDS and TAA, the mitigation 4182 is to clear the affected CPU buffers. 4183 4184 This parameter controls the mitigation. The 4185 options are: 4186 4187 full - Enable mitigation on vulnerable CPUs 4188 4189 full,nosmt - Enable mitigation and disable SMT on 4190 vulnerable CPUs. 4191 4192 off - Unconditionally disable mitigation 4193 4194 On MDS or TAA affected machines, 4195 mmio_stale_data=off can be prevented by an active 4196 MDS or TAA mitigation as these vulnerabilities are 4197 mitigated with the same mechanism so in order to 4198 disable this mitigation, you need to specify 4199 mds=off and tsx_async_abort=off too. 4200 4201 Not specifying this option is equivalent to 4202 mmio_stale_data=full. 4203 4204 For details see: 4205 Documentation/admin-guide/hw-vuln/processor_mmio_stale_data.rst 4206 4207 <module>.async_probe[=<bool>] [KNL] 4208 If no <bool> value is specified or if the value 4209 specified is not a valid <bool>, enable asynchronous 4210 probe on this module. Otherwise, enable/disable 4211 asynchronous probe on this module as indicated by the 4212 <bool> value. See also: module.async_probe 4213 4214 module.async_probe=<bool> 4215 [KNL] When set to true, modules will use async probing 4216 by default. To enable/disable async probing for a 4217 specific module, use the module specific control that 4218 is documented under <module>.async_probe. When both 4219 module.async_probe and <module>.async_probe are 4220 specified, <module>.async_probe takes precedence for 4221 the specific module. 4222 4223 module.enable_dups_trace 4224 [KNL] When CONFIG_MODULE_DEBUG_AUTOLOAD_DUPS is set, 4225 this means that duplicate request_module() calls will 4226 trigger a WARN_ON() instead of a pr_warn(). Note that 4227 if MODULE_DEBUG_AUTOLOAD_DUPS_TRACE is set, WARN_ON()s 4228 will always be issued and this option does nothing. 4229 module.sig_enforce 4230 [KNL] When CONFIG_MODULE_SIG is set, this means that 4231 modules without (valid) signatures will fail to load. 4232 Note that if CONFIG_MODULE_SIG_FORCE is set, that 4233 is always true, so this option does nothing. 4234 4235 module_blacklist= [KNL] Do not load a comma-separated list of 4236 modules. Useful for debugging problem modules. 4237 4238 mousedev.tap_time= 4239 [MOUSE] Maximum time between finger touching and 4240 leaving touchpad surface for touch to be considered 4241 a tap and be reported as a left button click (for 4242 touchpads working in absolute mode only). 4243 Format: <msecs> 4244 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices 4245 reporting absolute coordinates, such as tablets 4246 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices 4247 reporting absolute coordinates, such as tablets 4248 4249 movablecore= [KNL,X86,PPC,EARLY] 4250 Format: nn[KMGTPE] | nn% 4251 This parameter is the complement to kernelcore=, it 4252 specifies the amount of memory used for migratable 4253 allocations. If both kernelcore and movablecore is 4254 specified, then kernelcore will be at *least* the 4255 specified value but may be more. If movablecore on its 4256 own is specified, the administrator must be careful 4257 that the amount of memory usable for all allocations 4258 is not too small. 4259 4260 movable_node [KNL,EARLY] Boot-time switch to make hotplugable memory 4261 NUMA nodes to be movable. This means that the memory 4262 of such nodes will be usable only for movable 4263 allocations which rules out almost all kernel 4264 allocations. Use with caution! 4265 4266 MTD_Partition= [MTD] 4267 Format: <name>,<region-number>,<size>,<offset> 4268 4269 MTD_Region= [MTD] Format: 4270 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>] 4271 4272 mtdparts= [MTD] 4273 See drivers/mtd/parsers/cmdlinepart.c 4274 4275 mtouchusb.raw_coordinates= 4276 [HW] Make the MicroTouch USB driver use raw coordinates 4277 ('y', default) or cooked coordinates ('n') 4278 4279 mtrr=debug [X86,EARLY] 4280 Enable printing debug information related to MTRR 4281 registers at boot time. 4282 4283 mtrr_chunk_size=nn[KMG,X86,EARLY] 4284 used for mtrr cleanup. It is largest continuous chunk 4285 that could hold holes aka. UC entries. 4286 4287 mtrr_gran_size=nn[KMG,X86,EARLY] 4288 Used for mtrr cleanup. It is granularity of mtrr block. 4289 Default is 1. 4290 Large value could prevent small alignment from 4291 using up MTRRs. 4292 4293 mtrr_spare_reg_nr=n [X86,EARLY] 4294 Format: <integer> 4295 Range: 0,7 : spare reg number 4296 Default : 1 4297 Used for mtrr cleanup. It is spare mtrr entries number. 4298 Set to 2 or more if your graphical card needs more. 4299 4300 multitce=off [PPC] This parameter disables the use of the pSeries 4301 firmware feature for updating multiple TCE entries 4302 at a time. 4303 4304 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card 4305 4306 netdev= [NET] NE2000 ISA network devices parameters 4307 Format: <irq>,<io>,<mem_start>,<mem_end>,<name> 4308 4309 netpoll.carrier_timeout= 4310 [NET] Specifies amount of time (in seconds) that 4311 netpoll should wait for a carrier. By default netpoll 4312 waits 4 seconds. 4313 4314 nf_conntrack.acct= 4315 [NETFILTER] Enable connection tracking flow accounting 4316 0 to disable accounting 4317 1 to enable accounting 4318 Default value is 0. 4319 4320 nfs.cache_getent= 4321 [NFS] sets the pathname to the program which is used 4322 to update the NFS client cache entries. 4323 4324 nfs.cache_getent_timeout= 4325 [NFS] sets the timeout after which an attempt to 4326 update a cache entry is deemed to have failed. 4327 4328 nfs.callback_nr_threads= 4329 [NFSv4] set the total number of threads that the 4330 NFS client will assign to service NFSv4 callback 4331 requests. 4332 4333 nfs.callback_tcpport= 4334 [NFS] set the TCP port on which the NFSv4 callback 4335 channel should listen. 4336 4337 nfs.delay_retrans= 4338 [NFS] specifies the number of times the NFSv4 client 4339 retries the request before returning an EAGAIN error, 4340 after a reply of NFS4ERR_DELAY from the server. 4341 Only applies if the softerr mount option is enabled, 4342 and the specified value is >= 0. 4343 4344 nfs.idmap_cache_timeout= 4345 [NFS] set the maximum lifetime for idmapper cache 4346 entries. 4347 4348 nfs.max_session_cb_slots= 4349 [NFSv4.1] Sets the maximum number of session 4350 slots the client will assign to the callback 4351 channel. This determines the maximum number of 4352 callbacks the client will process in parallel for 4353 a particular server. 4354 4355 nfs.max_session_slots= 4356 [NFSv4.1] Sets the maximum number of session slots 4357 the client will attempt to negotiate with the server. 4358 This limits the number of simultaneous RPC requests 4359 that the client can send to the NFSv4.1 server. 4360 Note that there is little point in setting this 4361 value higher than the max_tcp_slot_table_limit. 4362 4363 nfs.nfs4_disable_idmapping= 4364 [NFSv4] When set to the default of '1', this option 4365 ensures that both the RPC level authentication 4366 scheme and the NFS level operations agree to use 4367 numeric uids/gids if the mount is using the 4368 'sec=sys' security flavour. In effect it is 4369 disabling idmapping, which can make migration from 4370 legacy NFSv2/v3 systems to NFSv4 easier. 4371 Servers that do not support this mode of operation 4372 will be autodetected by the client, and it will fall 4373 back to using the idmapper. 4374 To turn off this behaviour, set the value to '0'. 4375 4376 nfs.nfs4_unique_id= 4377 [NFS4] Specify an additional fixed unique ident- 4378 ification string that NFSv4 clients can insert into 4379 their nfs_client_id4 string. This is typically a 4380 UUID that is generated at system install time. 4381 4382 nfs.recover_lost_locks= 4383 [NFSv4] Attempt to recover locks that were lost due 4384 to a lease timeout on the server. Please note that 4385 doing this risks data corruption, since there are 4386 no guarantees that the file will remain unchanged 4387 after the locks are lost. 4388 If you want to enable the kernel legacy behaviour of 4389 attempting to recover these locks, then set this 4390 parameter to '1'. 4391 The default parameter value of '0' causes the kernel 4392 not to attempt recovery of lost locks. 4393 4394 nfs.send_implementation_id= 4395 [NFSv4.1] Send client implementation identification 4396 information in exchange_id requests. 4397 If zero, no implementation identification information 4398 will be sent. 4399 The default is to send the implementation identification 4400 information. 4401 4402 nfs4.layoutstats_timer= 4403 [NFSv4.2] Change the rate at which the kernel sends 4404 layoutstats to the pNFS metadata server. 4405 4406 Setting this to value to 0 causes the kernel to use 4407 whatever value is the default set by the layout 4408 driver. A non-zero value sets the minimum interval 4409 in seconds between layoutstats transmissions. 4410 4411 nfsd.inter_copy_offload_enable= 4412 [NFSv4.2] When set to 1, the server will support 4413 server-to-server copies for which this server is 4414 the destination of the copy. 4415 4416 nfsd.nfs4_disable_idmapping= 4417 [NFSv4] When set to the default of '1', the NFSv4 4418 server will return only numeric uids and gids to 4419 clients using auth_sys, and will accept numeric uids 4420 and gids from such clients. This is intended to ease 4421 migration from NFSv2/v3. 4422 4423 nfsd.nfsd4_ssc_umount_timeout= 4424 [NFSv4.2] When used as the destination of a 4425 server-to-server copy, knfsd temporarily mounts 4426 the source server. It caches the mount in case 4427 it will be needed again, and discards it if not 4428 used for the number of milliseconds specified by 4429 this parameter. 4430 4431 nfsaddrs= [NFS] Deprecated. Use ip= instead. 4432 See Documentation/admin-guide/nfs/nfsroot.rst. 4433 4434 nfsroot= [NFS] nfs root filesystem for disk-less boxes. 4435 See Documentation/admin-guide/nfs/nfsroot.rst. 4436 4437 nfsrootdebug [NFS] enable nfsroot debugging messages. 4438 See Documentation/admin-guide/nfs/nfsroot.rst. 4439 4440 nmi_backtrace.backtrace_idle [KNL] 4441 Dump stacks even of idle CPUs in response to an 4442 NMI stack-backtrace request. 4443 4444 nmi_debug= [KNL,SH] Specify one or more actions to take 4445 when a NMI is triggered. 4446 Format: [state][,regs][,debounce][,die] 4447 4448 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels 4449 Format: [panic,][nopanic,][rNNN,][num] 4450 Valid num: 0 or 1 4451 0 - turn hardlockup detector in nmi_watchdog off 4452 1 - turn hardlockup detector in nmi_watchdog on 4453 rNNN - configure the watchdog with raw perf event 0xNNN 4454 4455 When panic is specified, panic when an NMI watchdog 4456 timeout occurs (or 'nopanic' to not panic on an NMI 4457 watchdog, if CONFIG_BOOTPARAM_HARDLOCKUP_PANIC is set) 4458 To disable both hard and soft lockup detectors, 4459 please see 'nowatchdog'. 4460 This is useful when you use a panic=... timeout and 4461 need the box quickly up again. 4462 4463 These settings can be accessed at runtime via 4464 the nmi_watchdog and hardlockup_panic sysctls. 4465 4466 no4lvl [RISCV,EARLY] Disable 4-level and 5-level paging modes. 4467 Forces kernel to use 3-level paging instead. 4468 4469 no5lvl [X86-64,RISCV,EARLY] Disable 5-level paging mode. Forces 4470 kernel to use 4-level paging instead. 4471 4472 noalign [KNL,ARM] 4473 4474 noapic [SMP,APIC,EARLY] Tells the kernel to not make use of any 4475 IOAPICs that may be present in the system. 4476 4477 noapictimer [APIC,X86] Don't set up the APIC timer 4478 4479 noautogroup Disable scheduler automatic task group creation. 4480 4481 nocache [ARM,EARLY] 4482 4483 no_console_suspend 4484 [HW] Never suspend the console 4485 Disable suspending of consoles during suspend and 4486 hibernate operations. Once disabled, debugging 4487 messages can reach various consoles while the rest 4488 of the system is being put to sleep (ie, while 4489 debugging driver suspend/resume hooks). This may 4490 not work reliably with all consoles, but is known 4491 to work with serial and VGA consoles. 4492 To facilitate more flexible debugging, we also add 4493 console_suspend, a printk module parameter to control 4494 it. Users could use console_suspend (usually 4495 /sys/module/printk/parameters/console_suspend) to 4496 turn on/off it dynamically. 4497 4498 no_debug_objects 4499 [KNL,EARLY] Disable object debugging 4500 4501 nodsp [SH] Disable hardware DSP at boot time. 4502 4503 noefi [EFI,EARLY] Disable EFI runtime services support. 4504 4505 no_entry_flush [PPC,EARLY] Don't flush the L1-D cache when entering the kernel. 4506 4507 noexec32 [X86-64] 4508 This affects only 32-bit executables. 4509 noexec32=on: enable non-executable mappings (default) 4510 read doesn't imply executable mappings 4511 noexec32=off: disable non-executable mappings 4512 read implies executable mappings 4513 4514 no_file_caps Tells the kernel not to honor file capabilities. The 4515 only way then for a file to be executed with privilege 4516 is to be setuid root or executed by root. 4517 4518 nofpu [MIPS,SH] Disable hardware FPU at boot time. 4519 4520 nofsgsbase [X86] Disables FSGSBASE instructions. 4521 4522 nofxsr [BUGS=X86-32] Disables x86 floating point extended 4523 register save and restore. The kernel will only save 4524 legacy floating-point registers on task switch. 4525 4526 nogbpages [X86] Do not use GB pages for kernel direct mappings. 4527 4528 no_hash_pointers 4529 [KNL,EARLY] 4530 Alias for "hash_pointers=never". 4531 4532 nohibernate [HIBERNATION] Disable hibernation and resume. 4533 4534 nohlt [ARM,ARM64,MICROBLAZE,MIPS,PPC,RISCV,SH] Forces the kernel to 4535 busy wait in do_idle() and not use the arch_cpu_idle() 4536 implementation; requires CONFIG_GENERIC_IDLE_POLL_SETUP 4537 to be effective. This is useful on platforms where the 4538 sleep(SH) or wfi(ARM,ARM64) instructions do not work 4539 correctly or when doing power measurements to evaluate 4540 the impact of the sleep instructions. This is also 4541 useful when using JTAG debugger. 4542 4543 nohpet [X86] Don't use the HPET timer. 4544 4545 nohugeiomap [KNL,X86,PPC,ARM64,EARLY] Disable kernel huge I/O mappings. 4546 4547 nohugevmalloc [KNL,X86,PPC,ARM64,EARLY] Disable kernel huge vmalloc mappings. 4548 4549 nohz= [KNL] Boottime enable/disable dynamic ticks 4550 Valid arguments: on, off 4551 Default: on 4552 4553 nohz_full= [KNL,BOOT,SMP,ISOL] 4554 The argument is a cpu list, as described above. 4555 In kernels built with CONFIG_NO_HZ_FULL=y, set 4556 the specified list of CPUs whose tick will be stopped 4557 whenever possible. The boot CPU will be forced outside 4558 the range to maintain the timekeeping. Any CPUs 4559 in this list will have their RCU callbacks offloaded, 4560 just as if they had also been called out in the 4561 rcu_nocbs= boot parameter. 4562 4563 Note that this argument takes precedence over 4564 the CONFIG_RCU_NOCB_CPU_DEFAULT_ALL option. 4565 4566 noinitrd [Deprecated,RAM] Tells the kernel not to load any configured 4567 initial RAM disk. Currently this parameter applies to 4568 initrd only, not to initramfs. But it applies to both 4569 in EFI mode. 4570 4571 nointremap [X86-64,Intel-IOMMU,EARLY] Do not enable interrupt 4572 remapping. 4573 [Deprecated - use intremap=off] 4574 4575 noinvpcid [X86,EARLY] Disable the INVPCID cpu feature. 4576 4577 noiotrap [SH] Disables trapped I/O port accesses. 4578 4579 noirqdebug [X86-32] Disables the code which attempts to detect and 4580 disable unhandled interrupt sources. 4581 4582 noisapnp [ISAPNP] Disables ISA PnP code. 4583 4584 nokaslr [KNL,EARLY] 4585 When CONFIG_RANDOMIZE_BASE is set, this disables 4586 kernel and module base offset ASLR (Address Space 4587 Layout Randomization). 4588 4589 no-kvmapf [X86,KVM,EARLY] Disable paravirtualized asynchronous page 4590 fault handling. 4591 4592 no-kvmclock [X86,KVM,EARLY] Disable paravirtualized KVM clock driver 4593 4594 nolapic [X86-32,APIC,EARLY] Do not enable or use the local APIC. 4595 4596 nolapic_timer [X86-32,APIC,EARLY] Do not use the local APIC timer. 4597 4598 nomce [X86-32] Disable Machine Check Exception 4599 4600 nomfgpt [X86-32] Disable Multi-Function General Purpose 4601 Timer usage (for AMD Geode machines). 4602 4603 nomodeset Disable kernel modesetting. Most systems' firmware 4604 sets up a display mode and provides framebuffer memory 4605 for output. With nomodeset, DRM and fbdev drivers will 4606 not load if they could possibly displace the pre- 4607 initialized output. Only the system framebuffer will 4608 be available for use. The respective drivers will not 4609 perform display-mode changes or accelerated rendering. 4610 4611 Useful as error fallback, or for testing and debugging. 4612 4613 nomodule Disable module load 4614 4615 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to 4616 shutdown the other cpus. Instead use the REBOOT_VECTOR 4617 irq. 4618 4619 nopat [X86,EARLY] Disable PAT (page attribute table extension of 4620 pagetables) support. 4621 4622 nopcid [X86-64,EARLY] Disable the PCID cpu feature. 4623 4624 nopku [X86] Disable Memory Protection Keys CPU feature found 4625 in some Intel CPUs. 4626 4627 nopti [X86-64,EARLY] 4628 Equivalent to pti=off 4629 4630 nopv= [X86,XEN,KVM,HYPER_V,VMWARE,EARLY] 4631 Disables the PV optimizations forcing the guest to run 4632 as generic guest with no PV drivers. Currently support 4633 XEN HVM, KVM, HYPER_V and VMWARE guest. 4634 4635 nopvspin [X86,XEN,KVM,EARLY] 4636 Disables the qspinlock slow path using PV optimizations 4637 which allow the hypervisor to 'idle' the guest on lock 4638 contention. 4639 4640 norandmaps Don't use address space randomization. Equivalent to 4641 echo 0 > /proc/sys/kernel/randomize_va_space 4642 4643 noresume [SWSUSP] Disables resume and restores original swap 4644 space. 4645 4646 no-scroll [VGA] Disables scrollback. 4647 This is required for the Braillex ib80-piezo Braille 4648 reader made by F.H. Papenmeier (Germany). 4649 4650 nosgx [X86-64,SGX,EARLY] Disables Intel SGX kernel support. 4651 4652 nosmap [PPC,EARLY] 4653 Disable SMAP (Supervisor Mode Access Prevention) 4654 even if it is supported by processor. 4655 4656 nosmep [PPC64s,EARLY] 4657 Disable SMEP (Supervisor Mode Execution Prevention) 4658 even if it is supported by processor. 4659 4660 nosmp [SMP,EARLY] Tells an SMP kernel to act as a UP kernel, 4661 and disable the IO APIC. legacy for "maxcpus=0". 4662 4663 nosmt [KNL,MIPS,PPC,EARLY] Disable symmetric multithreading (SMT). 4664 Equivalent to smt=1. 4665 4666 [KNL,LOONGARCH,X86,ARM64,PPC,S390] Disable symmetric multithreading (SMT). 4667 nosmt=force: Force disable SMT, cannot be undone 4668 via the sysfs control file. 4669 4670 nosoftlockup [KNL] Disable the soft-lockup detector. 4671 4672 nospec_store_bypass_disable 4673 [HW,EARLY] Disable all mitigations for the Speculative 4674 Store Bypass vulnerability 4675 4676 nospectre_bhb [ARM64,EARLY] Disable all mitigations for Spectre-BHB (branch 4677 history injection) vulnerability. System may allow data leaks 4678 with this option. 4679 4680 nospectre_v1 [X86,PPC,EARLY] Disable mitigations for Spectre Variant 1 4681 (bounds check bypass). With this option data leaks are 4682 possible in the system. 4683 4684 nospectre_v2 [X86,PPC_E500,ARM64,EARLY] Disable all mitigations 4685 for the Spectre variant 2 (indirect branch 4686 prediction) vulnerability. System may allow data 4687 leaks with this option. 4688 4689 no-steal-acc [X86,PV_OPS,ARM64,PPC/PSERIES,RISCV,LOONGARCH,EARLY] 4690 Disable paravirtualized steal time accounting. steal time 4691 is computed, but won't influence scheduler behaviour 4692 4693 nosync [HW,M68K] Disables sync negotiation for all devices. 4694 4695 no_timer_check [X86,APIC] Disables the code which tests for broken 4696 timer IRQ sources, i.e., the IO-APIC timer. This can 4697 work around problems with incorrect timer 4698 initialization on some boards. 4699 4700 no_uaccess_flush 4701 [PPC,EARLY] Don't flush the L1-D cache after accessing user data. 4702 4703 novmcoredd [KNL,KDUMP] 4704 Disable device dump. Device dump allows drivers to 4705 append dump data to vmcore so you can collect driver 4706 specified debug info. Drivers can append the data 4707 without any limit and this data is stored in memory, 4708 so this may cause significant memory stress. Disabling 4709 device dump can help save memory but the driver debug 4710 data will be no longer available. This parameter 4711 is only available when CONFIG_PROC_VMCORE_DEVICE_DUMP 4712 is set. 4713 4714 no-vmw-sched-clock 4715 [X86,PV_OPS,EARLY] Disable paravirtualized VMware 4716 scheduler clock and use the default one. 4717 4718 nowatchdog [KNL] Disable both lockup detectors, i.e. 4719 soft-lockup and NMI watchdog (hard-lockup). 4720 4721 nowb [ARM,EARLY] 4722 4723 nox2apic [X86-64,APIC,EARLY] Do not enable x2APIC mode. 4724 4725 NOTE: this parameter will be ignored on systems with the 4726 LEGACY_XAPIC_DISABLED bit set in the 4727 IA32_XAPIC_DISABLE_STATUS MSR. 4728 4729 noxsave [BUGS=X86] Disables x86 extended register state save 4730 and restore using xsave. The kernel will fallback to 4731 enabling legacy floating-point and sse state. 4732 4733 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended 4734 register states. The kernel will fall back to use 4735 xsave to save the states. By using this parameter, 4736 performance of saving the states is degraded because 4737 xsave doesn't support modified optimization while 4738 xsaveopt supports it on xsaveopt enabled systems. 4739 4740 noxsaves [X86] Disables xsaves and xrstors used in saving and 4741 restoring x86 extended register state in compacted 4742 form of xsave area. The kernel will fall back to use 4743 xsaveopt and xrstor to save and restore the states 4744 in standard form of xsave area. By using this 4745 parameter, xsave area per process might occupy more 4746 memory on xsaves enabled systems. 4747 4748 nr_cpus= [SMP,EARLY] Maximum number of processors that an SMP kernel 4749 could support. nr_cpus=n : n >= 1 limits the kernel to 4750 support 'n' processors. It could be larger than the 4751 number of already plugged CPU during bootup, later in 4752 runtime you can physically add extra cpu until it reaches 4753 n. So during boot up some boot time memory for per-cpu 4754 variables need be pre-allocated for later physical cpu 4755 hot plugging. 4756 4757 nr_uarts= [SERIAL] maximum number of UARTs to be registered. 4758 4759 numa=off [KNL, ARM64, PPC, RISCV, SPARC, X86, EARLY] 4760 Disable NUMA, Only set up a single NUMA node 4761 spanning all memory. 4762 4763 numa=fake=<size>[MG] 4764 [KNL, ARM64, RISCV, X86, EARLY] 4765 If given as a memory unit, fills all system RAM with 4766 nodes of size interleaved over physical nodes. 4767 4768 numa=fake=<N> 4769 [KNL, ARM64, RISCV, X86, EARLY] 4770 If given as an integer, fills all system RAM with N 4771 fake nodes interleaved over physical nodes. 4772 4773 numa=fake=<N>U 4774 [KNL, ARM64, RISCV, X86, EARLY] 4775 If given as an integer followed by 'U', it will 4776 divide each physical node into N emulated nodes. 4777 4778 numa=noacpi [X86] Don't parse the SRAT table for NUMA setup 4779 4780 numa=nohmat [X86] Don't parse the HMAT table for NUMA setup, or 4781 soft-reserved memory partitioning. 4782 4783 numa_balancing= [KNL,ARM64,PPC,RISCV,S390,X86] Enable or disable automatic 4784 NUMA balancing. 4785 Allowed values are enable and disable 4786 4787 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA. 4788 'node', 'default' can be specified 4789 This can be set from sysctl after boot. 4790 See Documentation/admin-guide/sysctl/vm.rst for details. 4791 4792 nvme.quirks= [NVME] A list of quirk entries to augment the built-in 4793 nvme quirk list. List entries are separated by a 4794 '-' character. 4795 Each entry has the form VendorID:ProductID:quirk_names. 4796 The IDs are 4-digits hex numbers and quirk_names is a 4797 list of quirk names separated by commas. A quirk name 4798 can be prefixed by '^', meaning that the specified 4799 quirk must be disabled. 4800 4801 Example: 4802 nvme.quirks=7710:2267:bogus_nid,^identify_cns-9900:7711:broken_msi 4803 4804 ohci1394_dma=early [HW,EARLY] enable debugging via the ohci1394 driver. 4805 See Documentation/core-api/debugging-via-ohci1394.rst for more 4806 info. 4807 4808 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands 4809 Rather than timing out after 20 ms if an EC 4810 command is not properly ACKed, override the length 4811 of the timeout. We have interrupts disabled while 4812 waiting for the ACK, so if this is set too high 4813 interrupts *may* be lost! 4814 4815 omap_mux= [OMAP] Override bootloader pin multiplexing. 4816 Format: <mux_mode0.mode_name=value>... 4817 For example, to override I2C bus2: 4818 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100 4819 4820 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration 4821 4822 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock] 4823 4824 boundary - index of last SLC block on Flex-OneNAND. 4825 The remaining blocks are configured as MLC blocks. 4826 lock - Configure if Flex-OneNAND boundary should be locked. 4827 Once locked, the boundary cannot be changed. 4828 1 indicates lock status, 0 indicates unlock status. 4829 4830 oops=panic [KNL,EARLY] 4831 Always panic on oopses. Default is to just kill the 4832 process, but there is a small probability of 4833 deadlocking the machine. 4834 This will also cause panics on machine check exceptions. 4835 Useful together with panic=30 to trigger a reboot. 4836 4837 page_alloc.shuffle= 4838 [KNL] Boolean flag to control whether the page allocator 4839 should randomize its free lists. This parameter can be 4840 used to enable/disable page randomization. The state of 4841 the flag can be read from sysfs at: 4842 /sys/module/page_alloc/parameters/shuffle. 4843 This parameter is only available if CONFIG_SHUFFLE_PAGE_ALLOCATOR=y. 4844 4845 page_owner= [KNL,EARLY] Boot-time page_owner enabling option. 4846 Storage of the information about who allocated 4847 each page is disabled in default. With this switch, 4848 we can turn it on. 4849 on: enable the feature 4850 4851 page_poison= [KNL,EARLY] Boot-time parameter changing the state of 4852 poisoning on the buddy allocator, available with 4853 CONFIG_PAGE_POISONING=y. 4854 off: turn off poisoning (default) 4855 on: turn on poisoning 4856 4857 page_reporting.page_reporting_order= 4858 [KNL] Minimal page reporting order 4859 Format: <integer> 4860 Adjust the minimal page reporting order. The page 4861 reporting is disabled when it exceeds MAX_PAGE_ORDER. 4862 4863 page_reporting.page_reporting_delay_ms= 4864 [KNL] Free page reporting delay in milliseconds 4865 Format: <unsigned integer> 4866 Adjust the delay in milliseconds between free page 4867 reporting intervals. Default is 2000 (2 seconds). 4868 4869 panic= [KNL] Kernel behaviour on panic: delay <timeout> 4870 timeout > 0: seconds before rebooting 4871 timeout = 0: wait forever 4872 timeout < 0: reboot immediately 4873 Format: <timeout> 4874 4875 panic_on_taint= [KNL,EARLY] 4876 Bitmask for conditionally calling panic() in add_taint() 4877 Format: <hex>[,nousertaint] 4878 Hexadecimal bitmask representing the set of TAINT flags 4879 that will cause the kernel to panic when add_taint() is 4880 called with any of the flags in this set. 4881 The optional switch "nousertaint" can be utilized to 4882 prevent userspace forced crashes by writing to sysctl 4883 /proc/sys/kernel/tainted any flagset matching with the 4884 bitmask set on panic_on_taint. 4885 See Documentation/admin-guide/tainted-kernels.rst for 4886 extra details on the taint flags that users can pick 4887 to compose the bitmask to assign to panic_on_taint. 4888 4889 panic_on_warn=1 panic() instead of WARN(). Useful to cause kdump 4890 on a WARN(). 4891 4892 panic_force_cpu= 4893 [KNL,SMP] Force panic handling to execute on a specific CPU. 4894 Format: <cpu number> 4895 Some platforms require panic handling to occur on a 4896 specific CPU for the crash kernel to function correctly. 4897 This can be due to firmware limitations, interrupt routing 4898 constraints, or platform-specific requirements where only 4899 a particular CPU can safely enter the crash kernel. 4900 When set, panic() will redirect execution to the specified 4901 CPU before proceeding with the normal panic and kexec flow. 4902 If the target CPU is offline or unavailable, panic proceeds 4903 on the current CPU. 4904 This option should only be used for systems with the above 4905 constraints as it might cause the panic operation to be less reliable. 4906 4907 panic_print= Bitmask for printing system info when panic happens. 4908 User can chose combination of the following bits: 4909 4910 bit 0: print all tasks info 4911 bit 1: print system memory info 4912 bit 2: print timer info 4913 bit 3: print locks info if CONFIG_LOCKDEP is on 4914 bit 4: print ftrace buffer 4915 bit 5: replay all kernel messages on consoles at the end of panic 4916 bit 6: print all CPUs backtrace (if available in the arch) 4917 bit 7: print only tasks in uninterruptible (blocked) state 4918 4919 *Be aware* that this option may print a _lot_ of lines. 4920 4921 There is an increased risk of losing older messages in 4922 the log. Maybe worth to setup a bigger log buffer with 4923 "log_buf_len" along with this. 4924 4925 Also consider using "printk.console_no_auto_verbose=Y" 4926 if using this along with pstore, to avoid extra delays 4927 due to increased console verbosity during panic. 4928 4929 panic_sys_info= A comma separated list of extra information to be dumped 4930 on panic. 4931 Format: val[,val...] 4932 Where @val can be any of the following: 4933 4934 tasks: print all tasks info 4935 mem: print system memory info 4936 timers: print timers info 4937 locks: print locks info if CONFIG_LOCKDEP is on 4938 ftrace: print ftrace buffer 4939 all_bt: print all CPUs backtrace (if available in the arch) 4940 blocked_tasks: print only tasks in uninterruptible (blocked) state 4941 4942 This is a human readable alternative to the 'panic_print' 4943 option above. Please be aware of the caveats using this, 4944 check the 'panic_print' parameter for more details. 4945 4946 panic_console_replay 4947 When panic happens, replay all kernel messages on 4948 consoles at the end of panic. 4949 4950 parkbd.port= [HW] Parallel port number the keyboard adapter is 4951 connected to, default is 0. 4952 Format: <parport#> 4953 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation, 4954 0 for XT, 1 for AT (default is AT). 4955 Format: <mode> 4956 4957 parport= [HW,PPT] Specify parallel ports. 0 disables. 4958 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] } 4959 Use 'auto' to force the driver to use any 4960 IRQ/DMA settings detected (the default is to 4961 ignore detected IRQ/DMA settings because of 4962 possible conflicts). You can specify the base 4963 address, IRQ, and DMA settings; IRQ and DMA 4964 should be numbers, or 'auto' (for using detected 4965 settings on that particular port), or 'nofifo' 4966 (to avoid using a FIFO even if it is detected). 4967 Parallel ports are assigned in the order they 4968 are specified on the command line, starting 4969 with parport0. 4970 4971 parport_init_mode= [HW,PPT] 4972 Configure VIA parallel port to operate in 4973 a specific mode. This is necessary on Pegasos 4974 computer where firmware has no options for setting 4975 up parallel port mode and sets it to spp. 4976 Currently this function knows 686a and 8231 chips. 4977 Format: [spp|ps2|epp|ecp|ecpepp] 4978 4979 pata_legacy.all= [HW,LIBATA] 4980 Format: <int> 4981 Set to non-zero to probe primary and secondary ISA 4982 port ranges on PCI systems where no PCI PATA device 4983 has been found at either range. Disabled by default. 4984 4985 pata_legacy.autospeed= [HW,LIBATA] 4986 Format: <int> 4987 Set to non-zero if a chip is present that snoops speed 4988 changes. Disabled by default. 4989 4990 pata_legacy.ht6560a= [HW,LIBATA] 4991 Format: <int> 4992 Set to 1, 2, or 3 for HT 6560A on the primary channel, 4993 the secondary channel, or both channels respectively. 4994 Disabled by default. 4995 4996 pata_legacy.ht6560b= [HW,LIBATA] 4997 Format: <int> 4998 Set to 1, 2, or 3 for HT 6560B on the primary channel, 4999 the secondary channel, or both channels respectively. 5000 Disabled by default. 5001 5002 pata_legacy.iordy_mask= [HW,LIBATA] 5003 Format: <int> 5004 IORDY enable mask. Set individual bits to allow IORDY 5005 for the respective channel. Bit 0 is for the first 5006 legacy channel handled by this driver, bit 1 is for 5007 the second channel, and so on. The sequence will often 5008 correspond to the primary legacy channel, the secondary 5009 legacy channel, and so on, but the handling of a PCI 5010 bus and the use of other driver options may interfere 5011 with the sequence. By default IORDY is allowed across 5012 all channels. 5013 5014 pata_legacy.opti82c46x= [HW,LIBATA] 5015 Format: <int> 5016 Set to 1, 2, or 3 for Opti 82c611A on the primary 5017 channel, the secondary channel, or both channels 5018 respectively. Disabled by default. 5019 5020 pata_legacy.opti82c611a= [HW,LIBATA] 5021 Format: <int> 5022 Set to 1, 2, or 3 for Opti 82c465MV on the primary 5023 channel, the secondary channel, or both channels 5024 respectively. Disabled by default. 5025 5026 pata_legacy.pio_mask= [HW,LIBATA] 5027 Format: <int> 5028 PIO mode mask for autospeed devices. Set individual 5029 bits to allow the use of the respective PIO modes. 5030 Bit 0 is for mode 0, bit 1 is for mode 1, and so on. 5031 All modes allowed by default. 5032 5033 pata_legacy.probe_all= [HW,LIBATA] 5034 Format: <int> 5035 Set to non-zero to probe tertiary and further ISA 5036 port ranges on PCI systems. Disabled by default. 5037 5038 pata_legacy.probe_mask= [HW,LIBATA] 5039 Format: <int> 5040 Probe mask for legacy ISA PATA ports. Depending on 5041 platform configuration and the use of other driver 5042 options up to 6 legacy ports are supported: 0x1f0, 5043 0x170, 0x1e8, 0x168, 0x1e0, 0x160, however probing 5044 of individual ports can be disabled by setting the 5045 corresponding bits in the mask to 1. Bit 0 is for 5046 the first port in the list above (0x1f0), and so on. 5047 By default all supported ports are probed. 5048 5049 pata_legacy.qdi= [HW,LIBATA] 5050 Format: <int> 5051 Set to non-zero to probe QDI controllers. By default 5052 set to 1 if CONFIG_PATA_QDI_MODULE, 0 otherwise. 5053 5054 pata_legacy.winbond= [HW,LIBATA] 5055 Format: <int> 5056 Set to non-zero to probe Winbond controllers. Use 5057 the standard I/O port (0x130) if 1, otherwise the 5058 value given is the I/O port to use (typically 0x1b0). 5059 By default set to 1 if CONFIG_PATA_WINBOND_VLB_MODULE, 5060 0 otherwise. 5061 5062 pata_platform.pio_mask= [HW,LIBATA] 5063 Format: <int> 5064 Supported PIO mode mask. Set individual bits to allow 5065 the use of the respective PIO modes. Bit 0 is for 5066 mode 0, bit 1 is for mode 1, and so on. Mode 0 only 5067 allowed by default. 5068 5069 pause_on_oops=<int> 5070 Halt all CPUs after the first oops has been printed for 5071 the specified number of seconds. This is to be used if 5072 your oopses keep scrolling off the screen. 5073 5074 pci=option[,option...] [PCI,EARLY] various PCI subsystem options. 5075 5076 Some options herein operate on a specific device 5077 or a set of devices (<pci_dev>). These are 5078 specified in one of the following formats: 5079 5080 [<domain>:]<bus>:<dev>.<func>[/<dev>.<func>]* 5081 pci:<vendor>:<device>[:<subvendor>:<subdevice>] 5082 5083 Note: the first format specifies a PCI 5084 bus/device/function address which may change 5085 if new hardware is inserted, if motherboard 5086 firmware changes, or due to changes caused 5087 by other kernel parameters. If the 5088 domain is left unspecified, it is 5089 taken to be zero. Optionally, a path 5090 to a device through multiple device/function 5091 addresses can be specified after the base 5092 address (this is more robust against 5093 renumbering issues). The second format 5094 selects devices using IDs from the 5095 configuration space which may match multiple 5096 devices in the system. 5097 5098 earlydump dump PCI config space before the kernel 5099 changes anything 5100 off [X86] don't probe for the PCI bus 5101 bios [X86-32] force use of PCI BIOS, don't access 5102 the hardware directly. Use this if your machine 5103 has a non-standard PCI host bridge. 5104 nobios [X86-32] disallow use of PCI BIOS, only direct 5105 hardware access methods are allowed. Use this 5106 if you experience crashes upon bootup and you 5107 suspect they are caused by the BIOS. 5108 conf1 [X86] Force use of PCI Configuration Access 5109 Mechanism 1 (config address in IO port 0xCF8, 5110 data in IO port 0xCFC, both 32-bit). 5111 conf2 [X86] Force use of PCI Configuration Access 5112 Mechanism 2 (IO port 0xCF8 is an 8-bit port for 5113 the function, IO port 0xCFA, also 8-bit, sets 5114 bus number. The config space is then accessed 5115 through ports 0xC000-0xCFFF). 5116 See http://wiki.osdev.org/PCI for more info 5117 on the configuration access mechanisms. 5118 noaer [PCIE] If the PCIEAER kernel config parameter is 5119 enabled, this kernel boot option can be used to 5120 disable the use of PCIE advanced error reporting. 5121 nodomains [PCI] Disable support for multiple PCI 5122 root domains (aka PCI segments, in ACPI-speak). 5123 nommconf [X86] Disable use of MMCONFIG for PCI 5124 Configuration 5125 check_enable_amd_mmconf [X86] check for and enable 5126 properly configured MMIO access to PCI 5127 config space on AMD family 10h CPU 5128 nomsi [MSI] If the PCI_MSI kernel config parameter is 5129 enabled, this kernel boot option can be used to 5130 disable the use of MSI interrupts system-wide. 5131 noioapicquirk [APIC] Disable all boot interrupt quirks. 5132 Safety option to keep boot IRQs enabled. This 5133 should never be necessary. 5134 ioapicreroute [APIC] Enable rerouting of boot IRQs to the 5135 primary IO-APIC for bridges that cannot disable 5136 boot IRQs. This fixes a source of spurious IRQs 5137 when the system masks IRQs. 5138 noioapicreroute [APIC] Disable workaround that uses the 5139 boot IRQ equivalent of an IRQ that connects to 5140 a chipset where boot IRQs cannot be disabled. 5141 The opposite of ioapicreroute. 5142 biosirq [X86-32] Use PCI BIOS calls to get the interrupt 5143 routing table. These calls are known to be buggy 5144 on several machines and they hang the machine 5145 when used, but on other computers it's the only 5146 way to get the interrupt routing table. Try 5147 this option if the kernel is unable to allocate 5148 IRQs or discover secondary PCI buses on your 5149 motherboard. 5150 rom [X86] Assign address space to expansion ROMs. 5151 Use with caution as certain devices share 5152 address decoders between ROMs and other 5153 resources. 5154 norom [X86] Do not assign address space to 5155 expansion ROMs that do not already have 5156 BIOS assigned address ranges. 5157 nobar [X86] Do not assign address space to the 5158 BARs that weren't assigned by the BIOS. 5159 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be 5160 assigned automatically to PCI devices. You can 5161 make the kernel exclude IRQs of your ISA cards 5162 this way. 5163 pirqaddr=0xAAAAA [X86] Specify the physical address 5164 of the PIRQ table (normally generated 5165 by the BIOS) if it is outside the 5166 F0000h-100000h range. 5167 lastbus=N [X86] Scan all buses thru bus #N. Can be 5168 useful if the kernel is unable to find your 5169 secondary buses and you want to tell it 5170 explicitly which ones they are. 5171 assign-busses [X86] Always assign all PCI bus 5172 numbers ourselves, overriding 5173 whatever the firmware may have done. 5174 usepirqmask [X86] Honor the possible IRQ mask stored 5175 in the BIOS $PIR table. This is needed on 5176 some systems with broken BIOSes, notably 5177 some HP Pavilion N5400 and Omnibook XE3 5178 notebooks. This will have no effect if ACPI 5179 IRQ routing is enabled. 5180 noacpi [X86] Do not use ACPI for IRQ routing 5181 or for PCI scanning. 5182 use_crs [X86] Use PCI host bridge window information 5183 from ACPI. On BIOSes from 2008 or later, this 5184 is enabled by default. If you need to use this, 5185 please report a bug. 5186 nocrs [X86] Ignore PCI host bridge windows from ACPI. 5187 If you need to use this, please report a bug. 5188 use_e820 [X86] Use E820 reservations to exclude parts of 5189 PCI host bridge windows. This is a workaround 5190 for BIOS defects in host bridge _CRS methods. 5191 If you need to use this, please report a bug to 5192 <linux-pci@vger.kernel.org>. 5193 no_e820 [X86] Ignore E820 reservations for PCI host 5194 bridge windows. This is the default on modern 5195 hardware. If you need to use this, please report 5196 a bug to <linux-pci@vger.kernel.org>. 5197 routeirq Do IRQ routing for all PCI devices. 5198 This is normally done in pci_enable_device(), 5199 so this option is a temporary workaround 5200 for broken drivers that don't call it. 5201 skip_isa_align [X86] do not align io start addr, so can 5202 handle more pci cards 5203 noearly [X86] Don't do any early type 1 scanning. 5204 This might help on some broken boards which 5205 machine check when some devices' config space 5206 is read. But various workarounds are disabled 5207 and some IOMMU drivers will not work. 5208 bfsort Sort PCI devices into breadth-first order. 5209 This sorting is done to get a device 5210 order compatible with older (<= 2.4) kernels. 5211 nobfsort Don't sort PCI devices into breadth-first order. 5212 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size) 5213 tuning and use the BIOS-configured MPS defaults. 5214 pcie_bus_safe Set every device's MPS to the largest value 5215 supported by all devices below the root complex. 5216 pcie_bus_perf Set device MPS to the largest allowable MPS 5217 based on its parent bus. Also set MRRS (Max 5218 Read Request Size) to the largest supported 5219 value (no larger than the MPS that the device 5220 or bus can support) for best performance. 5221 pcie_bus_peer2peer Set every device's MPS to 128B, which 5222 every device is guaranteed to support. This 5223 configuration allows peer-to-peer DMA between 5224 any pair of devices, possibly at the cost of 5225 reduced performance. This also guarantees 5226 that hot-added devices will work. 5227 cbiosize=nn[KMG] The fixed amount of bus space which is 5228 reserved for the CardBus bridge's IO window. 5229 The default value is 256 bytes. 5230 cbmemsize=nn[KMG] The fixed amount of bus space which is 5231 reserved for the CardBus bridge's memory 5232 window. The default value is 64 megabytes. 5233 resource_alignment= 5234 Format: 5235 [<order of align>@]<pci_dev>[; ...] 5236 Specifies alignment and device to reassign 5237 aligned memory resources. How to 5238 specify the device is described above. 5239 If <order of align> is not specified, 5240 PAGE_SIZE is used as alignment. 5241 A PCI-PCI bridge can be specified if resource 5242 windows need to be expanded. 5243 To specify the alignment for several 5244 instances of a device, the PCI vendor, 5245 device, subvendor, and subdevice may be 5246 specified, e.g., 12@pci:8086:9c22:103c:198f 5247 for 4096-byte alignment. 5248 ecrc= Enable/disable PCIe ECRC (transaction layer 5249 end-to-end CRC checking). Only effective if 5250 OS has native AER control (either granted by 5251 ACPI _OSC or forced via "pcie_ports=native") 5252 bios: Use BIOS/firmware settings. This is the 5253 the default. 5254 off: Turn ECRC off 5255 on: Turn ECRC on. 5256 hpiosize=nn[KMG] The fixed amount of bus space which is 5257 reserved for hotplug bridge's IO window. 5258 Default size is 256 bytes. 5259 hpmmiosize=nn[KMG] The fixed amount of bus space which is 5260 reserved for hotplug bridge's MMIO window. 5261 Default size is 2 megabytes. 5262 hpmmioprefsize=nn[KMG] The fixed amount of bus space which is 5263 reserved for hotplug bridge's MMIO_PREF window. 5264 Default size is 2 megabytes. 5265 hpmemsize=nn[KMG] The fixed amount of bus space which is 5266 reserved for hotplug bridge's MMIO and 5267 MMIO_PREF window. 5268 Default size is 2 megabytes. 5269 hpbussize=nn The minimum amount of additional bus numbers 5270 reserved for buses below a hotplug bridge. 5271 Default is 1. 5272 realloc= Enable/disable reallocating PCI bridge resources 5273 if allocations done by BIOS are too small to 5274 accommodate resources required by all child 5275 devices. 5276 off: Turn realloc off 5277 on: Turn realloc on 5278 realloc same as realloc=on 5279 noari do not use PCIe ARI. 5280 noats [PCIE, Intel-IOMMU, AMD-IOMMU] 5281 do not use PCIe ATS (and IOMMU device IOTLB). 5282 pcie_scan_all Scan all possible PCIe devices. Otherwise we 5283 only look for one device below a PCIe downstream 5284 port. 5285 big_root_window Try to add a big 64bit memory window to the PCIe 5286 root complex on AMD CPUs. Some GFX hardware 5287 can resize a BAR to allow access to all VRAM. 5288 Adding the window is slightly risky (it may 5289 conflict with unreported devices), so this 5290 taints the kernel. 5291 disable_acs_redir=<pci_dev>[; ...] 5292 Specify one or more PCI devices (in the format 5293 specified above) separated by semicolons. 5294 Each device specified will have the PCI ACS 5295 redirect capabilities forced off which will 5296 allow P2P traffic between devices through 5297 bridges without forcing it upstream. Note: 5298 this removes isolation between devices and 5299 may put more devices in an IOMMU group. 5300 config_acs= 5301 Format: 5302 <ACS flags>@<pci_dev>[; ...] 5303 Specify one or more PCI devices (in the format 5304 specified above) optionally prepended with flags 5305 and separated by semicolons. The respective 5306 capabilities will be enabled, disabled or 5307 unchanged based on what is specified in 5308 flags. 5309 5310 ACS Flags is defined as follows: 5311 bit-0 : ACS Source Validation 5312 bit-1 : ACS Translation Blocking 5313 bit-2 : ACS P2P Request Redirect 5314 bit-3 : ACS P2P Completion Redirect 5315 bit-4 : ACS Upstream Forwarding 5316 bit-5 : ACS P2P Egress Control 5317 bit-6 : ACS Direct Translated P2P 5318 Each bit can be marked as: 5319 '0' – force disabled 5320 '1' – force enabled 5321 'x' – unchanged 5322 For example, 5323 pci=config_acs=10x@pci:0:0 5324 would configure all devices that support 5325 ACS to enable P2P Request Redirect, disable 5326 Translation Blocking, and leave Source 5327 Validation unchanged from whatever power-up 5328 or firmware set it to. 5329 5330 Note: this may remove isolation between devices 5331 and may put more devices in an IOMMU group. 5332 force_floating [S390] Force usage of floating interrupts. 5333 nomio [S390] Do not use MIO instructions. 5334 norid [S390] ignore the RID field and force use of 5335 one PCI domain per PCI function 5336 notph [PCIE] If the PCIE_TPH kernel config parameter 5337 is enabled, this kernel boot option can be used 5338 to disable PCIe TLP Processing Hints support 5339 system-wide. 5340 5341 pcie_aspm= [PCIE] Forcibly enable or ignore PCIe Active State Power 5342 Management. 5343 off Don't touch ASPM configuration at all. Leave any 5344 configuration done by firmware unchanged. 5345 force Enable ASPM even on devices that claim not to support it. 5346 WARNING: Forcing ASPM on may cause system lockups. 5347 5348 pcie_ports= [PCIE] PCIe port services handling: 5349 native Use native PCIe services (PME, AER, DPC, PCIe hotplug) 5350 even if the platform doesn't give the OS permission to 5351 use them. This may cause conflicts if the platform 5352 also tries to use these services. 5353 dpc-native Use native PCIe service for DPC only. May 5354 cause conflicts if firmware uses AER or DPC. 5355 compat Disable native PCIe services (PME, AER, DPC, PCIe 5356 hotplug). 5357 5358 pcie_port_pm= [PCIE] PCIe port power management handling: 5359 off Disable power management of all PCIe ports 5360 force Forcibly enable power management of all PCIe ports 5361 5362 pcie_pme= [PCIE,PM] Native PCIe PME signaling options: 5363 nomsi Do not use MSI for native PCIe PME signaling (this makes 5364 all PCIe root ports use INTx for all services). 5365 5366 pcmv= [HW,PCMCIA] BadgePAD 4 5367 5368 pd_ignore_unused 5369 [PM] 5370 Keep all power-domains already enabled by bootloader on, 5371 even if no driver has claimed them. This is useful 5372 for debug and development, but should not be 5373 needed on a platform with proper driver support. 5374 5375 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at 5376 boot time. 5377 Format: { 0 | 1 } 5378 See arch/parisc/kernel/pdc_chassis.c 5379 5380 percpu_alloc= [MM,EARLY] 5381 Select which percpu first chunk allocator to use. 5382 Currently supported values are "embed" and "page". 5383 Archs may support subset or none of the selections. 5384 See comments in mm/percpu.c for details on each 5385 allocator. This parameter is primarily for debugging 5386 and performance comparison. 5387 5388 pirq= [SMP,APIC] Manual mp-table setup 5389 See Documentation/arch/x86/i386/IO-APIC.rst. 5390 5391 plip= [PPT,NET] Parallel port network link 5392 Format: { parport<nr> | timid | 0 } 5393 See also Documentation/admin-guide/parport.rst. 5394 5395 pmtmr= [X86] Manual setup of pmtmr I/O Port. 5396 Override pmtimer IOPort with a hex value. 5397 e.g. pmtmr=0x508 5398 5399 pmu_override= [PPC] Override the PMU. 5400 This option takes over the PMU facility, so it is no 5401 longer usable by perf. Setting this option starts the 5402 PMU counters by setting MMCR0 to 0 (the FC bit is 5403 cleared). If a number is given, then MMCR1 is set to 5404 that number, otherwise (e.g., 'pmu_override=on'), MMCR1 5405 remains 0. 5406 5407 pm_async= [PM] 5408 Format: off 5409 This parameter sets the initial value of the 5410 /sys/power/pm_async sysfs knob at boot time. 5411 If set to "off", disables asynchronous suspend and 5412 resume of devices during system-wide power transitions. 5413 This can be useful on platforms where device 5414 dependencies are not well-defined, or for debugging 5415 power management issues. Asynchronous operations are 5416 enabled by default. 5417 5418 5419 pm_debug_messages [SUSPEND,KNL] 5420 Enable suspend/resume debug messages during boot up. 5421 5422 pm.dpm_watchdog_enabled= 5423 [PM] Enable or disable the DPM watchdog. Requires 5424 CONFIG_PM_SLEEP and CONFIG_DPM_WATCHDOG enabled. 5425 Format: <bool> 5426 Default value is set by CONFIG_DPM_WATCHDOG_ENABLED. 5427 5428 pnp.debug=1 [PNP] 5429 Enable PNP debug messages (depends on the 5430 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time 5431 via /sys/module/pnp/parameters/debug. We always show 5432 current resource usage; turning this on also shows 5433 possible settings and some assignment information. 5434 5435 pnpacpi= [ACPI] 5436 { off } 5437 5438 pnpbios= [ISAPNP] 5439 { on | off | curr | res | no-curr | no-res } 5440 5441 pnp_reserve_irq= 5442 [ISAPNP] Exclude IRQs for the autoconfiguration 5443 5444 pnp_reserve_dma= 5445 [ISAPNP] Exclude DMAs for the autoconfiguration 5446 5447 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration 5448 Ranges are in pairs (I/O port base and size). 5449 5450 pnp_reserve_mem= 5451 [ISAPNP] Exclude memory regions for the 5452 autoconfiguration. 5453 Ranges are in pairs (memory base and size). 5454 5455 ports= [IP_VS_FTP] IPVS ftp helper module 5456 Default is 21. 5457 Up to 8 (IP_VS_APP_MAX_PORTS) ports 5458 may be specified. 5459 Format: <port>,<port>.... 5460 5461 possible_cpus= [SMP,S390,X86] 5462 Format: <unsigned int> 5463 Set the number of possible CPUs, overriding the 5464 regular discovery mechanisms (such as ACPI/FW, etc). 5465 5466 powersave=off [PPC] This option disables power saving features. 5467 It specifically disables cpuidle and sets the 5468 platform machine description specific power_save 5469 function to NULL. On Idle the CPU just reduces 5470 execution priority. 5471 5472 ppc_strict_facility_enable 5473 [PPC,ENABLE] This option catches any kernel floating point, 5474 Altivec, VSX and SPE outside of regions specifically 5475 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()). 5476 There is some performance impact when enabling this. 5477 5478 ppc_tm= [PPC,EARLY] 5479 Format: {"off"} 5480 Disable Hardware Transactional Memory 5481 5482 preempt= [KNL] 5483 Select preemption mode if you have CONFIG_PREEMPT_DYNAMIC 5484 none - Limited to cond_resched() calls 5485 voluntary - Limited to cond_resched() and might_sleep() calls 5486 full - Any section that isn't explicitly preempt disabled 5487 can be preempted anytime. Tasks will also yield 5488 contended spinlocks (if the critical section isn't 5489 explicitly preempt disabled beyond the lock itself). 5490 lazy - Scheduler controlled. Similar to full but instead 5491 of preempting the task immediately, the task gets 5492 one HZ tick time to yield itself before the 5493 preemption will be forced. One preemption is when the 5494 task returns to user space. 5495 5496 print-fatal-signals= 5497 [KNL] debug: print fatal signals 5498 5499 If enabled, warn about various signal handling 5500 related application anomalies: too many signals, 5501 too many POSIX.1 timers, fatal signals causing a 5502 coredump - etc. 5503 5504 If you hit the warning due to signal overflow, 5505 you might want to try "ulimit -i unlimited". 5506 5507 default: off. 5508 5509 printk.always_kmsg_dump= 5510 Trigger kmsg_dump for cases other than kernel oops or 5511 panics 5512 Format: <bool> (1/Y/y=enable, 0/N/n=disable) 5513 default: disabled 5514 5515 printk.console_no_auto_verbose= 5516 Disable console loglevel raise on oops, panic 5517 or lockdep-detected issues (only if lock debug is on). 5518 With an exception to setups with low baudrate on 5519 serial console, keeping this 0 is a good choice 5520 in order to provide more debug information. 5521 Format: <bool> 5522 default: 0 (auto_verbose is enabled) 5523 5524 printk.debug_non_panic_cpus= 5525 Allows storing messages from non-panic CPUs into 5526 the printk log buffer during panic(). They are 5527 flushed to consoles by the panic-CPU on 5528 a best-effort basis. 5529 Format: <bool> (1/Y/y=enable, 0/N/n=disable) 5530 Default: disabled 5531 5532 printk.devkmsg={on,off,ratelimit} 5533 Control writing to /dev/kmsg. 5534 on - unlimited logging to /dev/kmsg from userspace 5535 off - logging to /dev/kmsg disabled 5536 ratelimit - ratelimit the logging 5537 Default: ratelimit 5538 5539 printk.time= Show timing data prefixed to each printk message line 5540 Format: <bool> (1/Y/y=enable, 0/N/n=disable) 5541 5542 proc_mem.force_override= [KNL] 5543 Format: {always | ptrace | never} 5544 Traditionally /proc/pid/mem allows memory permissions to be 5545 overridden without restrictions. This option may be set to 5546 restrict that. Can be one of: 5547 - 'always': traditional behavior always allows mem overrides. 5548 - 'ptrace': only allow mem overrides for active ptracers. 5549 - 'never': never allow mem overrides. 5550 If not specified, default is the CONFIG_PROC_MEM_* choice. 5551 5552 processor.max_cstate= [HW,ACPI] 5553 Limit processor to maximum C-state 5554 max_cstate=9 overrides any DMI blacklist limit. 5555 5556 processor.nocst [HW,ACPI] 5557 Ignore the _CST method to determine C-states, 5558 instead using the legacy FADT method 5559 5560 profile= [KNL] Enable kernel profiling via /proc/profile 5561 Format: [<profiletype>,]<number> 5562 Param: <profiletype>: "schedule" or "kvm" 5563 [defaults to kernel profiling] 5564 Param: "schedule" - profile schedule points. 5565 Param: "kvm" - profile VM exits. 5566 Param: <number> - step/bucket size as a power of 2 for 5567 statistical time based profiling. 5568 5569 prot_virt= [S390] enable hosting protected virtual machines 5570 isolated from the hypervisor (if hardware supports 5571 that). If enabled, the default kernel base address 5572 might be overridden even when Kernel Address Space 5573 Layout Randomization is disabled. 5574 Format: <bool> 5575 5576 psi= [KNL] Enable or disable pressure stall information 5577 tracking. 5578 Format: <bool> 5579 5580 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to 5581 probe for; one of (bare|imps|exps|lifebook|any). 5582 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports 5583 per second. 5584 psmouse.resetafter= [HW,MOUSE] 5585 Try to reset the device after so many bad packets 5586 (0 = never). 5587 psmouse.resolution= 5588 [HW,MOUSE] Set desired mouse resolution, in dpi. 5589 psmouse.smartscroll= 5590 [HW,MOUSE] Controls Logitech smartscroll autorepeat. 5591 0 = disabled, 1 = enabled (default). 5592 5593 pstore.backend= Specify the name of the pstore backend to use 5594 5595 pti= [X86-64] Control Page Table Isolation of user and 5596 kernel address spaces. Disabling this feature 5597 removes hardening, but improves performance of 5598 system calls and interrupts. 5599 5600 on - unconditionally enable 5601 off - unconditionally disable 5602 auto - kernel detects whether your CPU model is 5603 vulnerable to issues that PTI mitigates 5604 5605 Not specifying this option is equivalent to pti=auto. 5606 5607 pty.legacy_count= 5608 [KNL] Number of legacy pty's. Overwrites compiled-in 5609 default number. 5610 5611 quiet [KNL,EARLY] Disable most log messages 5612 5613 r128= [HW,DRM] 5614 5615 radix_hcall_invalidate=on [PPC/PSERIES] 5616 Disable RADIX GTSE feature and use hcall for TLB 5617 invalidate. 5618 5619 raid= [HW,RAID] 5620 See Documentation/admin-guide/md.rst. 5621 5622 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes 5623 See Documentation/admin-guide/blockdev/ramdisk.rst. 5624 5625 ramdisk_start= [Deprecated,RAM] RAM disk image start address 5626 5627 random.trust_cpu=off 5628 [KNL,EARLY] Disable trusting the use of the CPU's 5629 random number generator (if available) to 5630 initialize the kernel's RNG. 5631 5632 random.trust_bootloader=off 5633 [KNL,EARLY] Disable trusting the use of the a seed 5634 passed by the bootloader (if available) to 5635 initialize the kernel's RNG. 5636 5637 randomize_kstack_offset= 5638 [KNL,EARLY] Enable or disable kernel stack offset 5639 randomization, which provides roughly 5 bits of 5640 entropy, frustrating memory corruption attacks 5641 that depend on stack address determinism or 5642 cross-syscall address exposures. This is only 5643 available on architectures that have defined 5644 CONFIG_HAVE_ARCH_RANDOMIZE_KSTACK_OFFSET. 5645 Format: <bool> (1/Y/y=enable, 0/N/n=disable) 5646 Default is CONFIG_RANDOMIZE_KSTACK_OFFSET_DEFAULT. 5647 5648 ras=option[,option,...] [KNL] RAS-specific options 5649 5650 cec_disable [X86] 5651 Disable the Correctable Errors Collector, 5652 see CONFIG_RAS_CEC help text. 5653 5654 rcu_nocbs[=cpu-list] 5655 [KNL] The optional argument is a cpu list, 5656 as described above. 5657 5658 In kernels built with CONFIG_RCU_NOCB_CPU=y, 5659 enable the no-callback CPU mode, which prevents 5660 such CPUs' callbacks from being invoked in 5661 softirq context. Invocation of such CPUs' RCU 5662 callbacks will instead be offloaded to "rcuox/N" 5663 kthreads created for that purpose, where "x" is 5664 "p" for RCU-preempt, "s" for RCU-sched, and "g" 5665 for the kthreads that mediate grace periods; and 5666 "N" is the CPU number. This reduces OS jitter on 5667 the offloaded CPUs, which can be useful for HPC 5668 and real-time workloads. It can also improve 5669 energy efficiency for asymmetric multiprocessors. 5670 5671 If a cpulist is passed as an argument, the specified 5672 list of CPUs is set to no-callback mode from boot. 5673 5674 Otherwise, if the '=' sign and the cpulist 5675 arguments are omitted, no CPU will be set to 5676 no-callback mode from boot but the mode may be 5677 toggled at runtime via cpusets. 5678 5679 Note that this argument takes precedence over 5680 the CONFIG_RCU_NOCB_CPU_DEFAULT_ALL option. 5681 5682 rcu_nocb_poll [KNL] 5683 Rather than requiring that offloaded CPUs 5684 (specified by rcu_nocbs= above) explicitly 5685 awaken the corresponding "rcuoN" kthreads, 5686 make these kthreads poll for callbacks. 5687 This improves the real-time response for the 5688 offloaded CPUs by relieving them of the need to 5689 wake up the corresponding kthread, but degrades 5690 energy efficiency by requiring that the kthreads 5691 periodically wake up to do the polling. 5692 5693 rcutree.blimit= [KNL] 5694 Set maximum number of finished RCU callbacks to 5695 process in one batch. 5696 5697 rcutree.csd_lock_suppress_rcu_stall= [KNL] 5698 Do only a one-line RCU CPU stall warning when 5699 there is an ongoing too-long CSD-lock wait. 5700 5701 rcutree.do_rcu_barrier= [KNL] 5702 Request a call to rcu_barrier(). This is 5703 throttled so that userspace tests can safely 5704 hammer on the sysfs variable if they so choose. 5705 If triggered before the RCU grace-period machinery 5706 is fully active, this will error out with EAGAIN. 5707 5708 rcutree.dump_tree= [KNL] 5709 Dump the structure of the rcu_node combining tree 5710 out at early boot. This is used for diagnostic 5711 purposes, to verify correct tree setup. 5712 5713 rcutree.gp_cleanup_delay= [KNL] 5714 Set the number of jiffies to delay each step of 5715 RCU grace-period cleanup. 5716 5717 rcutree.gp_init_delay= [KNL] 5718 Set the number of jiffies to delay each step of 5719 RCU grace-period initialization. 5720 5721 rcutree.gp_preinit_delay= [KNL] 5722 Set the number of jiffies to delay each step of 5723 RCU grace-period pre-initialization, that is, 5724 the propagation of recent CPU-hotplug changes up 5725 the rcu_node combining tree. 5726 5727 rcutree.jiffies_till_first_fqs= [KNL] 5728 Set delay from grace-period initialization to 5729 first attempt to force quiescent states. 5730 Units are jiffies, minimum value is zero, 5731 and maximum value is HZ. 5732 5733 rcutree.jiffies_till_next_fqs= [KNL] 5734 Set delay between subsequent attempts to force 5735 quiescent states. Units are jiffies, minimum 5736 value is one, and maximum value is HZ. 5737 5738 rcutree.jiffies_till_sched_qs= [KNL] 5739 Set required age in jiffies for a 5740 given grace period before RCU starts 5741 soliciting quiescent-state help from 5742 rcu_note_context_switch() and cond_resched(). 5743 If not specified, the kernel will calculate 5744 a value based on the most recent settings 5745 of rcutree.jiffies_till_first_fqs 5746 and rcutree.jiffies_till_next_fqs. 5747 This calculated value may be viewed in 5748 rcutree.jiffies_to_sched_qs. Any attempt to set 5749 rcutree.jiffies_to_sched_qs will be cheerfully 5750 overwritten. 5751 5752 rcutree.kthread_prio= [KNL,BOOT] 5753 Set the SCHED_FIFO priority of the RCU per-CPU 5754 kthreads (rcuc/N). This value is also used for 5755 the priority of the RCU boost threads (rcub/N) 5756 and for the RCU grace-period kthreads (rcu_bh, 5757 rcu_preempt, and rcu_sched). If RCU_BOOST is 5758 set, valid values are 1-99 and the default is 1 5759 (the least-favored priority). Otherwise, when 5760 RCU_BOOST is not set, valid values are 0-99 and 5761 the default is zero (non-realtime operation). 5762 When RCU_NOCB_CPU is set, also adjust the 5763 priority of NOCB callback kthreads. 5764 5765 rcutree.nocb_nobypass_lim_per_jiffy= [KNL] 5766 On callback-offloaded (rcu_nocbs) CPUs, 5767 RCU reduces the lock contention that would 5768 otherwise be caused by callback floods through 5769 use of the ->nocb_bypass list. However, in the 5770 common non-flooded case, RCU queues directly to 5771 the main ->cblist in order to avoid the extra 5772 overhead of the ->nocb_bypass list and its lock. 5773 But if there are too many callbacks queued during 5774 a single jiffy, RCU pre-queues the callbacks into 5775 the ->nocb_bypass queue. The definition of "too 5776 many" is supplied by this kernel boot parameter. 5777 5778 rcutree.nohz_full_patience_delay= [KNL] 5779 On callback-offloaded (rcu_nocbs) CPUs, avoid 5780 disturbing RCU unless the grace period has 5781 reached the specified age in milliseconds. 5782 Defaults to zero. Large values will be capped 5783 at five seconds. All values will be rounded down 5784 to the nearest value representable by jiffies. 5785 5786 rcutree.qhimark= [KNL] 5787 Set threshold of queued RCU callbacks beyond which 5788 batch limiting is disabled. 5789 5790 rcutree.qlowmark= [KNL] 5791 Set threshold of queued RCU callbacks below which 5792 batch limiting is re-enabled. 5793 5794 rcutree.qovld= [KNL] 5795 Set threshold of queued RCU callbacks beyond which 5796 RCU's force-quiescent-state scan will aggressively 5797 enlist help from cond_resched() and sched IPIs to 5798 help CPUs more quickly reach quiescent states. 5799 Set to less than zero to make this be set based 5800 on rcutree.qhimark at boot time and to zero to 5801 disable more aggressive help enlistment. 5802 5803 rcutree.rcu_delay_page_cache_fill_msec= [KNL] 5804 Set the page-cache refill delay (in milliseconds) 5805 in response to low-memory conditions. The range 5806 of permitted values is in the range 0:100000. 5807 5808 rcutree.rcu_divisor= [KNL] 5809 Set the shift-right count to use to compute 5810 the callback-invocation batch limit bl from 5811 the number of callbacks queued on this CPU. 5812 The result will be bounded below by the value of 5813 the rcutree.blimit kernel parameter. Every bl 5814 callbacks, the softirq handler will exit in 5815 order to allow the CPU to do other work. 5816 5817 Please note that this callback-invocation batch 5818 limit applies only to non-offloaded callback 5819 invocation. Offloaded callbacks are instead 5820 invoked in the context of an rcuoc kthread, which 5821 scheduler will preempt as it does any other task. 5822 5823 rcutree.rcu_fanout_exact= [KNL] 5824 Disable autobalancing of the rcu_node combining 5825 tree. This is used by rcutorture, and might 5826 possibly be useful for architectures having high 5827 cache-to-cache transfer latencies. 5828 5829 rcutree.rcu_fanout_leaf= [KNL] 5830 Change the number of CPUs assigned to each 5831 leaf rcu_node structure. Useful for very 5832 large systems, which will choose the value 64, 5833 and for NUMA systems with large remote-access 5834 latencies, which will choose a value aligned 5835 with the appropriate hardware boundaries. 5836 5837 rcutree.rcu_min_cached_objs= [KNL] 5838 Minimum number of objects which are cached and 5839 maintained per one CPU. Object size is equal 5840 to PAGE_SIZE. The cache allows to reduce the 5841 pressure to page allocator, also it makes the 5842 whole algorithm to behave better in low memory 5843 condition. 5844 5845 rcutree.rcu_nocb_gp_stride= [KNL] 5846 Set the number of NOCB callback kthreads in 5847 each group, which defaults to the square root 5848 of the number of CPUs. Larger numbers reduce 5849 the wakeup overhead on the global grace-period 5850 kthread, but increases that same overhead on 5851 each group's NOCB grace-period kthread. 5852 5853 rcutree.rcu_kick_kthreads= [KNL] 5854 Cause the grace-period kthread to get an extra 5855 wake_up() if it sleeps three times longer than 5856 it should at force-quiescent-state time. 5857 This wake_up() will be accompanied by a 5858 WARN_ONCE() splat and an ftrace_dump(). 5859 5860 rcutree.rcu_resched_ns= [KNL] 5861 Limit the time spend invoking a batch of RCU 5862 callbacks to the specified number of nanoseconds. 5863 By default, this limit is checked only once 5864 every 32 callbacks in order to limit the pain 5865 inflicted by local_clock() overhead. 5866 5867 rcutree.rcu_unlock_delay= [KNL] 5868 In CONFIG_RCU_STRICT_GRACE_PERIOD=y kernels, 5869 this specifies an rcu_read_unlock()-time delay 5870 in microseconds. This defaults to zero. 5871 Larger delays increase the probability of 5872 catching RCU pointer leaks, that is, buggy use 5873 of RCU-protected pointers after the relevant 5874 rcu_read_unlock() has completed. 5875 5876 rcutree.sysrq_rcu= [KNL] 5877 Commandeer a sysrq key to dump out Tree RCU's 5878 rcu_node tree with an eye towards determining 5879 why a new grace period has not yet started. 5880 5881 rcutree.use_softirq= [KNL] 5882 If set to zero, move all RCU_SOFTIRQ processing to 5883 per-CPU rcuc kthreads. Defaults to a non-zero 5884 value, meaning that RCU_SOFTIRQ is used by default. 5885 Specify rcutree.use_softirq=0 to use rcuc kthreads. 5886 5887 But note that CONFIG_PREEMPT_RT=y kernels disable 5888 this kernel boot parameter, forcibly setting it 5889 to zero. 5890 5891 rcutree.enable_rcu_lazy= [KNL] 5892 To save power, batch RCU callbacks and flush after 5893 delay, memory pressure or callback list growing too 5894 big. 5895 5896 rcutree.rcu_normal_wake_from_gp= [KNL] 5897 Reduces a latency of synchronize_rcu() call. This approach 5898 maintains its own track of synchronize_rcu() callers, so it 5899 does not interact with regular callbacks because it does not 5900 use a call_rcu[_hurry]() path. Please note, this is for a 5901 normal grace period. 5902 5903 How to disable it: 5904 5905 echo 0 > /sys/module/rcutree/parameters/rcu_normal_wake_from_gp 5906 or pass a boot parameter "rcutree.rcu_normal_wake_from_gp=0" 5907 5908 Default is 1 if it is not explicitly disabled by the boot parameter 5909 passing 0. 5910 5911 rcuscale.gp_async= [KNL] 5912 Measure performance of asynchronous 5913 grace-period primitives such as call_rcu(). 5914 5915 rcuscale.gp_async_max= [KNL] 5916 Specify the maximum number of outstanding 5917 callbacks per writer thread. When a writer 5918 thread exceeds this limit, it invokes the 5919 corresponding flavor of rcu_barrier() to allow 5920 previously posted callbacks to drain. 5921 5922 rcuscale.gp_exp= [KNL] 5923 Measure performance of expedited synchronous 5924 grace-period primitives. 5925 5926 rcuscale.holdoff= [KNL] 5927 Set test-start holdoff period. The purpose of 5928 this parameter is to delay the start of the 5929 test until boot completes in order to avoid 5930 interference. 5931 5932 rcuscale.kfree_by_call_rcu= [KNL] 5933 In kernels built with CONFIG_RCU_LAZY=y, test 5934 call_rcu() instead of kfree_rcu(). 5935 5936 rcuscale.kfree_mult= [KNL] 5937 Instead of allocating an object of size kfree_obj, 5938 allocate one of kfree_mult * sizeof(kfree_obj). 5939 Defaults to 1. 5940 5941 rcuscale.kfree_rcu_test= [KNL] 5942 Set to measure performance of kfree_rcu() flooding. 5943 5944 rcuscale.kfree_rcu_test_double= [KNL] 5945 Test the double-argument variant of kfree_rcu(). 5946 If this parameter has the same value as 5947 rcuscale.kfree_rcu_test_single, both the single- 5948 and double-argument variants are tested. 5949 5950 rcuscale.kfree_rcu_test_single= [KNL] 5951 Test the single-argument variant of kfree_rcu(). 5952 If this parameter has the same value as 5953 rcuscale.kfree_rcu_test_double, both the single- 5954 and double-argument variants are tested. 5955 5956 rcuscale.kfree_nthreads= [KNL] 5957 The number of threads running loops of kfree_rcu(). 5958 5959 rcuscale.kfree_alloc_num= [KNL] 5960 Number of allocations and frees done in an iteration. 5961 5962 rcuscale.kfree_loops= [KNL] 5963 Number of loops doing rcuscale.kfree_alloc_num number 5964 of allocations and frees. 5965 5966 rcuscale.minruntime= [KNL] 5967 Set the minimum test run time in seconds. This 5968 does not affect the data-collection interval, 5969 but instead allows better measurement of things 5970 like CPU consumption. 5971 5972 rcuscale.nreaders= [KNL] 5973 Set number of RCU readers. The value -1 selects 5974 N, where N is the number of CPUs. A value 5975 "n" less than -1 selects N-n+1, where N is again 5976 the number of CPUs. For example, -2 selects N 5977 (the number of CPUs), -3 selects N+1, and so on. 5978 A value of "n" less than or equal to -N selects 5979 a single reader. 5980 5981 rcuscale.nwriters= [KNL] 5982 Set number of RCU writers. The values operate 5983 the same as for rcuscale.nreaders. 5984 N, where N is the number of CPUs 5985 5986 rcuscale.scale_type= [KNL] 5987 Specify the RCU implementation to test. 5988 5989 rcuscale.shutdown= [KNL] 5990 Shut the system down after performance tests 5991 complete. This is useful for hands-off automated 5992 testing. 5993 5994 rcuscale.verbose= [KNL] 5995 Enable additional printk() statements. 5996 5997 rcuscale.writer_holdoff= [KNL] 5998 Write-side holdoff between grace periods, 5999 in microseconds. The default of zero says 6000 no holdoff. 6001 6002 rcuscale.writer_holdoff_jiffies= [KNL] 6003 Additional write-side holdoff between grace 6004 periods, but in jiffies. The default of zero 6005 says no holdoff. 6006 6007 rcutorture.fqs_duration= [KNL] 6008 Set duration of force_quiescent_state bursts 6009 in microseconds. 6010 6011 rcutorture.fqs_holdoff= [KNL] 6012 Set holdoff time within force_quiescent_state bursts 6013 in microseconds. 6014 6015 rcutorture.fqs_stutter= [KNL] 6016 Set wait time between force_quiescent_state bursts 6017 in seconds. 6018 6019 rcutorture.fwd_progress= [KNL] 6020 Specifies the number of kthreads to be used 6021 for RCU grace-period forward-progress testing 6022 for the types of RCU supporting this notion. 6023 Defaults to 1 kthread, values less than zero or 6024 greater than the number of CPUs cause the number 6025 of CPUs to be used. 6026 6027 rcutorture.fwd_progress_div= [KNL] 6028 Specify the fraction of a CPU-stall-warning 6029 period to do tight-loop forward-progress testing. 6030 6031 rcutorture.fwd_progress_holdoff= [KNL] 6032 Number of seconds to wait between successive 6033 forward-progress tests. 6034 6035 rcutorture.fwd_progress_need_resched= [KNL] 6036 Enclose cond_resched() calls within checks for 6037 need_resched() during tight-loop forward-progress 6038 testing. 6039 6040 rcutorture.gp_cond= [KNL] 6041 Use conditional/asynchronous update-side 6042 normal-grace-period primitives, if available. 6043 6044 rcutorture.gp_cond_exp= [KNL] 6045 Use conditional/asynchronous update-side 6046 expedited-grace-period primitives, if available. 6047 6048 rcutorture.gp_cond_full= [KNL] 6049 Use conditional/asynchronous update-side 6050 normal-grace-period primitives that also take 6051 concurrent expedited grace periods into account, 6052 if available. 6053 6054 rcutorture.gp_cond_exp_full= [KNL] 6055 Use conditional/asynchronous update-side 6056 expedited-grace-period primitives that also take 6057 concurrent normal grace periods into account, 6058 if available. 6059 6060 rcutorture.gp_cond_wi= [KNL] 6061 Nominal wait interval for normal conditional 6062 grace periods (specified by rcutorture's 6063 gp_cond and gp_cond_full module parameters), 6064 in microseconds. The actual wait interval will 6065 be randomly selected to nanosecond granularity up 6066 to this wait interval. Defaults to 16 jiffies, 6067 for example, 16,000 microseconds on a system 6068 with HZ=1000. 6069 6070 rcutorture.gp_cond_wi_exp= [KNL] 6071 Nominal wait interval for expedited conditional 6072 grace periods (specified by rcutorture's 6073 gp_cond_exp and gp_cond_exp_full module 6074 parameters), in microseconds. The actual wait 6075 interval will be randomly selected to nanosecond 6076 granularity up to this wait interval. Defaults to 6077 128 microseconds. 6078 6079 rcutorture.gp_exp= [KNL] 6080 Use expedited update-side primitives, if available. 6081 6082 rcutorture.gp_normal= [KNL] 6083 Use normal (non-expedited) asynchronous 6084 update-side primitives, if available. 6085 6086 rcutorture.gp_poll= [KNL] 6087 Use polled update-side normal-grace-period 6088 primitives, if available. 6089 6090 rcutorture.gp_poll_exp= [KNL] 6091 Use polled update-side expedited-grace-period 6092 primitives, if available. 6093 6094 rcutorture.gp_poll_full= [KNL] 6095 Use polled update-side normal-grace-period 6096 primitives that also take concurrent expedited 6097 grace periods into account, if available. 6098 6099 rcutorture.gp_poll_exp_full= [KNL] 6100 Use polled update-side expedited-grace-period 6101 primitives that also take concurrent normal 6102 grace periods into account, if available. 6103 6104 rcutorture.gp_poll_wi= [KNL] 6105 Nominal wait interval for normal conditional 6106 grace periods (specified by rcutorture's 6107 gp_poll and gp_poll_full module parameters), 6108 in microseconds. The actual wait interval will 6109 be randomly selected to nanosecond granularity up 6110 to this wait interval. Defaults to 16 jiffies, 6111 for example, 16,000 microseconds on a system 6112 with HZ=1000. 6113 6114 rcutorture.gp_poll_wi_exp= [KNL] 6115 Nominal wait interval for expedited conditional 6116 grace periods (specified by rcutorture's 6117 gp_poll_exp and gp_poll_exp_full module 6118 parameters), in microseconds. The actual wait 6119 interval will be randomly selected to nanosecond 6120 granularity up to this wait interval. Defaults to 6121 128 microseconds. 6122 6123 rcutorture.gp_sync= [KNL] 6124 Use normal (non-expedited) synchronous 6125 update-side primitives, if available. If all 6126 of rcutorture.gp_cond=, rcutorture.gp_exp=, 6127 rcutorture.gp_normal=, and rcutorture.gp_sync= 6128 are zero, rcutorture acts as if is interpreted 6129 they are all non-zero. 6130 6131 rcutorture.gpwrap_lag= [KNL] 6132 Enable grace-period wrap lag testing. Setting 6133 to false prevents the gpwrap lag test from 6134 running. Default is true. 6135 6136 rcutorture.gpwrap_lag_gps= [KNL] 6137 Set the value for grace-period wrap lag during 6138 active lag testing periods. This controls how many 6139 grace periods differences we tolerate between 6140 rdp and rnp's gp_seq before setting overflow flag. 6141 The default is always set to 8. 6142 6143 rcutorture.gpwrap_lag_cycle_mins= [KNL] 6144 Set the total cycle duration for gpwrap lag 6145 testing in minutes. This is the total time for 6146 one complete cycle of active and inactive 6147 testing periods. Default is 30 minutes. 6148 6149 rcutorture.gpwrap_lag_active_mins= [KNL] 6150 Set the duration for which gpwrap lag is active 6151 within each cycle, in minutes. During this time, 6152 the grace-period wrap lag will be set to the 6153 value specified by gpwrap_lag_gps. Default is 6154 5 minutes. 6155 6156 rcutorture.irqreader= [KNL] 6157 Run RCU readers from irq handlers, or, more 6158 accurately, from a timer handler. Not all RCU 6159 flavors take kindly to this sort of thing. 6160 6161 rcutorture.leakpointer= [KNL] 6162 Leak an RCU-protected pointer out of the reader. 6163 This can of course result in splats, and is 6164 intended to test the ability of things like 6165 CONFIG_RCU_STRICT_GRACE_PERIOD=y to detect 6166 such leaks. 6167 6168 rcutorture.n_barrier_cbs= [KNL] 6169 Set callbacks/threads for rcu_barrier() testing. 6170 6171 rcutorture.nfakewriters= [KNL] 6172 Set number of concurrent RCU writers. These just 6173 stress RCU, they don't participate in the actual 6174 test, hence the "fake". 6175 6176 rcutorture.nocbs_nthreads= [KNL] 6177 Set number of RCU callback-offload togglers. 6178 Zero (the default) disables toggling. 6179 6180 rcutorture.nocbs_toggle= [KNL] 6181 Set the delay in milliseconds between successive 6182 callback-offload toggling attempts. 6183 6184 rcutorture.nreaders= [KNL] 6185 Set number of RCU readers. The value -1 selects 6186 N-1, where N is the number of CPUs. A value 6187 "n" less than -1 selects N-n-2, where N is again 6188 the number of CPUs. For example, -2 selects N 6189 (the number of CPUs), -3 selects N+1, and so on. 6190 6191 rcutorture.nwriters= [KNL] 6192 Set number of RCU writers, which must be either 6193 zero or one. For additional writers, use instead 6194 the rcutorture.nfakewriters parameter. 6195 6196 rcutorture.object_debug= [KNL] 6197 Enable debug-object double-call_rcu() testing. 6198 6199 rcutorture.onoff_holdoff= [KNL] 6200 Set time (s) after boot for CPU-hotplug testing. 6201 6202 rcutorture.onoff_interval= [KNL] 6203 Set time (jiffies) between CPU-hotplug operations, 6204 or zero to disable CPU-hotplug testing. 6205 6206 rcutorture.preempt_duration= [KNL] 6207 Set duration (in milliseconds) of preemptions 6208 by a high-priority FIFO real-time task. Set to 6209 zero (the default) to disable. The CPUs to 6210 preempt are selected randomly from the set that 6211 are online at a given point in time. Races with 6212 CPUs going offline are ignored, with that attempt 6213 at preemption skipped. 6214 6215 rcutorture.preempt_interval= [KNL] 6216 Set interval (in milliseconds, defaulting to one 6217 second) between preemptions by a high-priority 6218 FIFO real-time task. This delay is mediated 6219 by an hrtimer and is further fuzzed to avoid 6220 inadvertent synchronizations. 6221 6222 rcutorture.read_exit_burst= [KNL] 6223 The number of times in a given read-then-exit 6224 episode that a set of read-then-exit kthreads 6225 is spawned. 6226 6227 rcutorture.read_exit_delay= [KNL] 6228 The delay, in seconds, between successive 6229 read-then-exit testing episodes. 6230 6231 rcutorture.reader_flavor= [KNL] 6232 A bit mask indicating which readers to use. 6233 If there is more than one bit set, the readers 6234 are entered from low-order bit up, and are 6235 exited in the opposite order. For SRCU, the 6236 0x1 bit is normal readers, 0x2 NMI-safe readers, 6237 and 0x4 light-weight readers. 6238 6239 rcutorture.shuffle_interval= [KNL] 6240 Set task-shuffle interval (s). Shuffling tasks 6241 allows some CPUs to go into dyntick-idle mode 6242 during the rcutorture test. 6243 6244 rcutorture.shutdown_secs= [KNL] 6245 Set time (s) after boot system shutdown. This 6246 is useful for hands-off automated testing. 6247 6248 rcutorture.stall_cpu= [KNL] 6249 Duration of CPU stall (s) to test RCU CPU stall 6250 warnings, zero to disable. 6251 6252 rcutorture.stall_cpu_block= [KNL] 6253 Sleep while stalling if set. This will result 6254 in warnings from preemptible RCU in addition to 6255 any other stall-related activity. Note that 6256 in kernels built with CONFIG_PREEMPTION=n and 6257 CONFIG_PREEMPT_COUNT=y, this parameter will 6258 cause the CPU to pass through a quiescent state. 6259 Given CONFIG_PREEMPTION=n, this will suppress 6260 RCU CPU stall warnings, but will instead result 6261 in scheduling-while-atomic splats. 6262 6263 Use of this module parameter results in splats. 6264 6265 6266 rcutorture.stall_cpu_holdoff= [KNL] 6267 Time to wait (s) after boot before inducing stall. 6268 6269 rcutorture.stall_cpu_irqsoff= [KNL] 6270 Disable interrupts while stalling if set, but only 6271 on the first stall in the set. 6272 6273 rcutorture.stall_cpu_repeat= [KNL] 6274 Number of times to repeat the stall sequence, 6275 so that rcutorture.stall_cpu_repeat=3 will result 6276 in four stall sequences. 6277 6278 rcutorture.stall_gp_kthread= [KNL] 6279 Duration (s) of forced sleep within RCU 6280 grace-period kthread to test RCU CPU stall 6281 warnings, zero to disable. If both stall_cpu 6282 and stall_gp_kthread are specified, the 6283 kthread is starved first, then the CPU. 6284 6285 rcutorture.stall_only= [KNL] 6286 Shut off all rcutorture kthreads other than the 6287 RCU CPU stall-warning test kthreads. The purpose 6288 of this is to test production applictions' 6289 reactions to CPU stalls, and with minimal 6290 additional overhead. Or you can omit the 6291 stall-warning tests as well and get a heavy 6292 no-op, your choice! 6293 6294 rcutorture.stat_interval= [KNL] 6295 Time (s) between statistics printk()s. 6296 6297 rcutorture.stutter= [KNL] 6298 Time (s) to stutter testing, for example, specifying 6299 five seconds causes the test to run for five seconds, 6300 wait for five seconds, and so on. This tests RCU's 6301 ability to transition abruptly to and from idle. 6302 6303 rcutorture.test_boost= [KNL] 6304 Test RCU priority boosting? 0=no, 1=maybe, 2=yes. 6305 "Maybe" means test if the RCU implementation 6306 under test support RCU priority boosting. 6307 6308 rcutorture.test_boost_duration= [KNL] 6309 Duration (s) of each individual boost test. 6310 6311 rcutorture.test_boost_holdoff= [KNL] 6312 Holdoff time (s) from start of test to the start 6313 of RCU priority-boost testing. Defaults to zero, 6314 that is, no holdoff. 6315 6316 rcutorture.test_boost_interval= [KNL] 6317 Interval (s) between each boost test. 6318 6319 rcutorture.test_no_idle_hz= [KNL] 6320 Test RCU's dyntick-idle handling. See also the 6321 rcutorture.shuffle_interval parameter. 6322 6323 rcutorture.torture_type= [KNL] 6324 Specify the RCU implementation to test. 6325 6326 rcutorture.verbose= [KNL] 6327 Enable additional printk() statements. 6328 6329 rcupdate.rcu_cpu_stall_ftrace_dump= [KNL] 6330 Dump ftrace buffer after reporting RCU CPU 6331 stall warning. 6332 6333 rcupdate.rcu_cpu_stall_notifiers= [KNL] 6334 Provide RCU CPU stall notifiers, but see the 6335 warnings in the RCU_CPU_STALL_NOTIFIER Kconfig 6336 option's help text. TL;DR: You almost certainly 6337 do not want rcupdate.rcu_cpu_stall_notifiers. 6338 6339 rcupdate.rcu_cpu_stall_suppress= [KNL] 6340 Suppress RCU CPU stall warning messages. 6341 6342 rcupdate.rcu_cpu_stall_suppress_at_boot= [KNL] 6343 Suppress RCU CPU stall warning messages and 6344 rcutorture writer stall warnings that occur 6345 during early boot, that is, during the time 6346 before the init task is spawned. 6347 6348 rcupdate.rcu_cpu_stall_timeout= [KNL] 6349 Set timeout for RCU CPU stall warning messages. 6350 The value is in seconds and the maximum allowed 6351 value is 300 seconds. 6352 6353 rcupdate.rcu_exp_cpu_stall_timeout= [KNL] 6354 Set timeout for expedited RCU CPU stall warning 6355 messages. The value is in milliseconds 6356 and the maximum allowed value is 21000 6357 milliseconds. Please note that this value is 6358 adjusted to an arch timer tick resolution. 6359 Setting this to zero causes the value from 6360 rcupdate.rcu_cpu_stall_timeout to be used (after 6361 conversion from seconds to milliseconds). 6362 6363 rcupdate.rcu_cpu_stall_cputime= [KNL] 6364 Provide statistics on the cputime and count of 6365 interrupts and tasks during the sampling period. For 6366 multiple continuous RCU stalls, all sampling periods 6367 begin at half of the first RCU stall timeout. 6368 6369 rcupdate.rcu_exp_stall_task_details= [KNL] 6370 Print stack dumps of any tasks blocking the 6371 current expedited RCU grace period during an 6372 expedited RCU CPU stall warning. 6373 6374 rcupdate.rcu_expedited= [KNL] 6375 Use expedited grace-period primitives, for 6376 example, synchronize_rcu_expedited() instead 6377 of synchronize_rcu(). This reduces latency, 6378 but can increase CPU utilization, degrade 6379 real-time latency, and degrade energy efficiency. 6380 No effect on CONFIG_TINY_RCU kernels. 6381 6382 rcupdate.rcu_normal= [KNL] 6383 Use only normal grace-period primitives, 6384 for example, synchronize_rcu() instead of 6385 synchronize_rcu_expedited(). This improves 6386 real-time latency, CPU utilization, and 6387 energy efficiency, but can expose users to 6388 increased grace-period latency. This parameter 6389 overrides rcupdate.rcu_expedited. No effect on 6390 CONFIG_TINY_RCU kernels. 6391 6392 rcupdate.rcu_normal_after_boot= [KNL] 6393 Once boot has completed (that is, after 6394 rcu_end_inkernel_boot() has been invoked), use 6395 only normal grace-period primitives. No effect 6396 on CONFIG_TINY_RCU kernels. 6397 6398 But note that CONFIG_PREEMPT_RT=y kernels enables 6399 this kernel boot parameter, forcibly setting 6400 it to the value one, that is, converting any 6401 post-boot attempt at an expedited RCU grace 6402 period to instead use normal non-expedited 6403 grace-period processing. 6404 6405 rcupdate.rcu_task_collapse_lim= [KNL] 6406 Set the maximum number of callbacks present 6407 at the beginning of a grace period that allows 6408 the RCU Tasks flavors to collapse back to using 6409 a single callback queue. This switching only 6410 occurs when rcupdate.rcu_task_enqueue_lim is 6411 set to the default value of -1. 6412 6413 rcupdate.rcu_task_contend_lim= [KNL] 6414 Set the minimum number of callback-queuing-time 6415 lock-contention events per jiffy required to 6416 cause the RCU Tasks flavors to switch to per-CPU 6417 callback queuing. This switching only occurs 6418 when rcupdate.rcu_task_enqueue_lim is set to 6419 the default value of -1. 6420 6421 rcupdate.rcu_task_enqueue_lim= [KNL] 6422 Set the number of callback queues to use for the 6423 RCU Tasks family of RCU flavors. The default 6424 of -1 allows this to be automatically (and 6425 dynamically) adjusted. This parameter is intended 6426 for use in testing. 6427 6428 rcupdate.rcu_task_lazy_lim= [KNL] 6429 Number of callbacks on a given CPU that will 6430 cancel laziness on that CPU. Use -1 to disable 6431 cancellation of laziness, but be advised that 6432 doing so increases the danger of OOM due to 6433 callback flooding. 6434 6435 rcupdate.rcu_task_stall_info= [KNL] 6436 Set initial timeout in jiffies for RCU task stall 6437 informational messages, which give some indication 6438 of the problem for those not patient enough to 6439 wait for ten minutes. Informational messages are 6440 only printed prior to the stall-warning message 6441 for a given grace period. Disable with a value 6442 less than or equal to zero. Defaults to ten 6443 seconds. A change in value does not take effect 6444 until the beginning of the next grace period. 6445 6446 rcupdate.rcu_task_stall_info_mult= [KNL] 6447 Multiplier for time interval between successive 6448 RCU task stall informational messages for a given 6449 RCU tasks grace period. This value is clamped 6450 to one through ten, inclusive. It defaults to 6451 the value three, so that the first informational 6452 message is printed 10 seconds into the grace 6453 period, the second at 40 seconds, the third at 6454 160 seconds, and then the stall warning at 600 6455 seconds would prevent a fourth at 640 seconds. 6456 6457 rcupdate.rcu_task_stall_timeout= [KNL] 6458 Set timeout in jiffies for RCU task stall 6459 warning messages. Disable with a value less 6460 than or equal to zero. Defaults to ten minutes. 6461 A change in value does not take effect until 6462 the beginning of the next grace period. 6463 6464 rcupdate.rcu_tasks_lazy_ms= [KNL] 6465 Set timeout in milliseconds RCU Tasks asynchronous 6466 callback batching for call_rcu_tasks(). 6467 A negative value will take the default. A value 6468 of zero will disable batching. Batching is 6469 always disabled for synchronize_rcu_tasks(). 6470 6471 rcupdate.rcu_self_test= [KNL] 6472 Run the RCU early boot self tests 6473 6474 rdinit= [KNL] 6475 Format: <full_path> 6476 Run specified binary instead of /init from the ramdisk, 6477 used for early userspace startup. See initrd. 6478 6479 rdrand= [X86,EARLY] 6480 force - Override the decision by the kernel to hide the 6481 advertisement of RDRAND support (this affects 6482 certain AMD processors because of buggy BIOS 6483 support, specifically around the suspend/resume 6484 path). 6485 6486 rdt= [HW,X86,RDT] 6487 Turn on/off individual RDT features. List is: 6488 cmt, mbmtotal, mbmlocal, l3cat, l3cdp, l2cat, l2cdp, 6489 mba, smba, bmec, abmc, sdciae, energy[:guid], 6490 perf[:guid]. 6491 E.g. to turn on cmt and turn off mba use: 6492 rdt=cmt,!mba 6493 To turn off all energy telemetry monitoring and ensure that 6494 perf telemetry monitoring associated with guid 0x12345 6495 is enabled use: 6496 rdt=!energy,perf:0x12345 6497 6498 reboot= [KNL] 6499 Format (x86 or x86_64): 6500 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] | d[efault] \ 6501 [[,]s[mp]####] \ 6502 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \ 6503 [[,]f[orce]] 6504 Where reboot_mode is one of warm (soft) or cold (hard) or gpio 6505 (prefix with 'panic_' to set mode for panic 6506 reboot only), 6507 reboot_type is one of bios, acpi, kbd, triple, efi, or pci, 6508 reboot_force is either force or not specified, 6509 reboot_cpu is s[mp]#### with #### being the processor 6510 to be used for rebooting. 6511 6512 acpi 6513 Use the ACPI RESET_REG in the FADT. If ACPI is not 6514 configured or the ACPI reset does not work, the reboot 6515 path attempts the reset using the keyboard controller. 6516 6517 bios 6518 Use the CPU reboot vector for warm reset 6519 6520 cold 6521 Set the cold reboot flag 6522 6523 default 6524 There are some built-in platform specific "quirks" 6525 - you may see: "reboot: <name> series board detected. 6526 Selecting <type> for reboots." In the case where you 6527 think the quirk is in error (e.g. you have newer BIOS, 6528 or newer board) using this option will ignore the 6529 built-in quirk table, and use the generic default 6530 reboot actions. 6531 6532 efi 6533 Use efi reset_system runtime service. If EFI is not 6534 configured or the EFI reset does not work, the reboot 6535 path attempts the reset using the keyboard controller. 6536 6537 force 6538 Don't stop other CPUs on reboot. This can make reboot 6539 more reliable in some cases. 6540 6541 kbd 6542 Use the keyboard controller. cold reset (default) 6543 6544 pci 6545 Use a write to the PCI config space register 0xcf9 to 6546 trigger reboot. 6547 6548 triple 6549 Force a triple fault (init) 6550 6551 warm 6552 Don't set the cold reboot flag 6553 6554 Using warm reset will be much faster especially on big 6555 memory systems because the BIOS will not go through 6556 the memory check. Disadvantage is that not all 6557 hardware will be completely reinitialized on reboot so 6558 there may be boot problems on some systems. 6559 6560 6561 refscale.holdoff= [KNL] 6562 Set test-start holdoff period. The purpose of 6563 this parameter is to delay the start of the 6564 test until boot completes in order to avoid 6565 interference. 6566 6567 refscale.lookup_instances= [KNL] 6568 Number of data elements to use for the forms of 6569 SLAB_TYPESAFE_BY_RCU testing. A negative number 6570 is negated and multiplied by nr_cpu_ids, while 6571 zero specifies nr_cpu_ids. 6572 6573 refscale.loops= [KNL] 6574 Set the number of loops over the synchronization 6575 primitive under test. Increasing this number 6576 reduces noise due to loop start/end overhead, 6577 but the default has already reduced the per-pass 6578 noise to a handful of picoseconds on ca. 2020 6579 x86 laptops. 6580 6581 refscale.nreaders= [KNL] 6582 Set number of readers. The default value of -1 6583 selects N, where N is roughly 75% of the number 6584 of CPUs. A value of zero is an interesting choice. 6585 6586 refscale.nruns= [KNL] 6587 Set number of runs, each of which is dumped onto 6588 the console log. 6589 6590 refscale.readdelay= [KNL] 6591 Set the read-side critical-section duration, 6592 measured in microseconds. 6593 6594 refscale.scale_type= [KNL] 6595 Specify the read-protection implementation to test. 6596 6597 refscale.shutdown= [KNL] 6598 Shut down the system at the end of the performance 6599 test. This defaults to 1 (shut it down) when 6600 refscale is built into the kernel and to 0 (leave 6601 it running) when refscale is built as a module. 6602 6603 refscale.verbose= [KNL] 6604 Enable additional printk() statements. 6605 6606 refscale.verbose_batched= [KNL] 6607 Batch the additional printk() statements. If zero 6608 (the default) or negative, print everything. Otherwise, 6609 print every Nth verbose statement, where N is the value 6610 specified. 6611 6612 regulator_ignore_unused 6613 [REGULATOR] 6614 Prevents regulator framework from disabling regulators 6615 that are unused, due no driver claiming them. This may 6616 be useful for debug and development, but should not be 6617 needed on a platform with proper driver support. 6618 6619 relax_domain_level= 6620 [KNL, SMP] Set scheduler's default relax_domain_level. 6621 See Documentation/admin-guide/cgroup-v1/cpusets.rst. 6622 6623 reserve= [KNL,BUGS] Force kernel to ignore I/O ports or memory 6624 Format: <base1>,<size1>[,<base2>,<size2>,...] 6625 Reserve I/O ports or memory so the kernel won't use 6626 them. If <base> is less than 0x10000, the region 6627 is assumed to be I/O ports; otherwise it is memory. 6628 6629 reserve_mem= [RAM] 6630 Format: nn[KMG]:<align>:<label> 6631 Reserve physical memory and label it with a name that 6632 other subsystems can use to access it. This is typically 6633 used for systems that do not wipe the RAM, and this command 6634 line will try to reserve the same physical memory on 6635 soft reboots. Note, it is not guaranteed to be the same 6636 location. For example, if anything about the system changes 6637 or if booting a different kernel. It can also fail if KASLR 6638 places the kernel at the location of where the RAM reservation 6639 was from a previous boot, the new reservation will be at a 6640 different location. 6641 Any subsystem using this feature must add a way to verify 6642 that the contents of the physical memory is from a previous 6643 boot, as there may be cases where the memory will not be 6644 located at the same location. 6645 6646 The format is size:align:label for example, to request 6647 12 megabytes of 4096 alignment for ramoops: 6648 6649 reserve_mem=12M:4096:oops ramoops.mem_name=oops 6650 6651 reservetop= [X86-32,EARLY] 6652 Format: nn[KMG] 6653 Reserves a hole at the top of the kernel virtual 6654 address space. 6655 6656 reset_devices [KNL] Force drivers to reset the underlying device 6657 during initialization. 6658 6659 resume= [SWSUSP] 6660 Specify the partition device for software suspend 6661 Format: 6662 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>} 6663 6664 resume_offset= [SWSUSP] 6665 Specify the offset from the beginning of the partition 6666 given by "resume=" at which the swap header is located, 6667 in <PAGE_SIZE> units (needed only for swap files). 6668 See Documentation/power/swsusp-and-swap-files.rst 6669 6670 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to 6671 read the resume files 6672 6673 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up. 6674 Useful for devices that are detected asynchronously 6675 (e.g. USB and MMC devices). 6676 6677 retain_initrd [RAM] Keep initrd memory after extraction. After boot, it will 6678 be accessible via /sys/firmware/initrd. 6679 6680 retbleed= [X86] Control mitigation of RETBleed (Arbitrary 6681 Speculative Code Execution with Return Instructions) 6682 vulnerability. 6683 6684 AMD-based UNRET and IBPB mitigations alone do not stop 6685 sibling threads from influencing the predictions of other 6686 sibling threads. For that reason, STIBP is used on pro- 6687 cessors that support it, and mitigate SMT on processors 6688 that don't. 6689 6690 off - no mitigation 6691 auto - automatically select a mitigation 6692 auto,nosmt - automatically select a mitigation, 6693 disabling SMT if necessary for 6694 the full mitigation (only on Zen1 6695 and older without STIBP). 6696 ibpb - On AMD, mitigate short speculation 6697 windows on basic block boundaries too. 6698 Safe, highest perf impact. It also 6699 enables STIBP if present. Not suitable 6700 on Intel. 6701 ibpb,nosmt - Like "ibpb" above but will disable SMT 6702 when STIBP is not available. This is 6703 the alternative for systems which do not 6704 have STIBP. 6705 unret - Force enable untrained return thunks, 6706 only effective on AMD f15h-f17h based 6707 systems. 6708 unret,nosmt - Like unret, but will disable SMT when STIBP 6709 is not available. This is the alternative for 6710 systems which do not have STIBP. 6711 6712 Selecting 'auto' will choose a mitigation method at run 6713 time according to the CPU. 6714 6715 Not specifying this option is equivalent to retbleed=auto. 6716 6717 rfkill.default_state= 6718 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm, 6719 etc. communication is blocked by default. 6720 1 Unblocked. 6721 6722 rfkill.master_switch_mode= 6723 0 The "airplane mode" button does nothing. 6724 1 The "airplane mode" button toggles between everything 6725 blocked and the previous configuration. 6726 2 The "airplane mode" button toggles between everything 6727 blocked and everything unblocked. 6728 6729 ring3mwait=disable 6730 [KNL] Disable ring 3 MONITOR/MWAIT feature on supported 6731 CPUs. 6732 6733 riscv_isa_fallback [RISCV,EARLY] 6734 When CONFIG_RISCV_ISA_FALLBACK is not enabled, permit 6735 falling back to detecting extension support by parsing 6736 "riscv,isa" property on devicetree systems when the 6737 replacement properties are not found. See the Kconfig 6738 entry for RISCV_ISA_FALLBACK. 6739 6740 riscv_nousercfi= 6741 all Disable user CFI ABI to userspace even if cpu extension 6742 are available. 6743 bcfi Disable user backward CFI ABI to userspace even if 6744 the shadow stack extension is available. 6745 fcfi Disable user forward CFI ABI to userspace even if the 6746 landing pad extension is available. 6747 6748 ro [KNL] Mount root device read-only on boot 6749 6750 rodata= [KNL,EARLY] 6751 on Mark read-only kernel memory as read-only (default). 6752 off Leave read-only kernel memory writable for debugging. 6753 noalias Mark read-only kernel memory as read-only but retain 6754 writable aliases in the direct map for regions outside 6755 of the kernel image. [arm64] 6756 6757 rockchip.usb_uart 6758 [EARLY] 6759 Enable the uart passthrough on the designated usb port 6760 on Rockchip SoCs. When active, the signals of the 6761 debug-uart get routed to the D+ and D- pins of the usb 6762 port and the regular usb controller gets disabled. 6763 6764 root= [KNL] Root filesystem 6765 Usually this is a block device specifier of some kind, 6766 see the early_lookup_bdev comment in 6767 block/early-lookup.c for details. 6768 Alternatively this can be "ram" for the legacy initial 6769 ramdisk, "nfs" and "cifs" for root on a network file 6770 system, or "mtd" and "ubi" for mounting from raw flash. 6771 6772 rootdelay= [KNL] Delay (in seconds) to pause before attempting to 6773 mount the root filesystem 6774 6775 rootflags= [KNL] Set root filesystem mount option string 6776 6777 rseq_slice_ext= [KNL] RSEQ based time slice extension 6778 Format: boolean 6779 Control enablement of RSEQ based time slice extension. 6780 Default is 'on'. 6781 6782 initramfs_options= [KNL] 6783 Specify mount options for the initramfs mount. 6784 6785 rootfstype= [KNL] Set root filesystem type 6786 6787 rootwait [KNL] Wait (indefinitely) for root device to show up. 6788 Useful for devices that are detected asynchronously 6789 (e.g. USB and MMC devices). 6790 6791 rootwait= [KNL] Maximum time (in seconds) to wait for root device 6792 to show up before attempting to mount the root 6793 filesystem. 6794 6795 rproc_mem=nn[KMG][@address] 6796 [KNL,ARM,CMA] Remoteproc physical memory block. 6797 Memory area to be used by remote processor image, 6798 managed by CMA. 6799 6800 rseq_debug= [KNL] Enable or disable restartable sequence 6801 debug mode. Defaults to CONFIG_RSEQ_DEBUG_DEFAULT_ENABLE. 6802 Format: <bool> 6803 6804 rt_group_sched= [KNL] Enable or disable SCHED_RR/FIFO group scheduling 6805 when CONFIG_RT_GROUP_SCHED=y. Defaults to 6806 !CONFIG_RT_GROUP_SCHED_DEFAULT_DISABLED. 6807 Format: <bool> 6808 6809 rw [KNL] Mount root device read-write on boot 6810 6811 S [KNL] Run init in single mode 6812 6813 s390_iommu= [HW,S390] 6814 Set s390 IOTLB flushing mode 6815 strict 6816 With strict flushing every unmap operation will result 6817 in an IOTLB flush. Default is lazy flushing before 6818 reuse, which is faster. Deprecated, equivalent to 6819 iommu.strict=1. 6820 6821 s390_iommu_aperture= [KNL,S390] 6822 Specifies the size of the per device DMA address space 6823 accessible through the DMA and IOMMU APIs as a decimal 6824 factor of the size of main memory. 6825 The default is 1 meaning that one can concurrently use 6826 as many DMA addresses as physical memory is installed, 6827 if supported by hardware, and thus map all of memory 6828 once. With a value of 2 one can map all of memory twice 6829 and so on. As a special case a factor of 0 imposes no 6830 restrictions other than those given by hardware at the 6831 cost of significant additional memory use for tables. 6832 6833 6834 sched_proxy_exec= [KNL] 6835 Enables or disables "proxy execution" style 6836 solution to mutex-based priority inversion. 6837 Format: <bool> 6838 6839 sched_verbose [KNL,EARLY] Enables verbose scheduler debug messages. 6840 6841 schedstats= [KNL,X86] Enable or disable scheduled statistics. 6842 Allowed values are enable and disable. This feature 6843 incurs a small amount of overhead in the scheduler 6844 but is useful for debugging and performance tuning. 6845 6846 sched_thermal_decay_shift= 6847 [Deprecated] 6848 [KNL, SMP] Set a decay shift for scheduler thermal 6849 pressure signal. Thermal pressure signal follows the 6850 default decay period of other scheduler pelt 6851 signals(usually 32 ms but configurable). Setting 6852 sched_thermal_decay_shift will left shift the decay 6853 period for the thermal pressure signal by the shift 6854 value. 6855 i.e. with the default pelt decay period of 32 ms 6856 sched_thermal_decay_shift thermal pressure decay pr 6857 1 64 ms 6858 2 128 ms 6859 and so on. 6860 Format: integer between 0 and 10 6861 Default is 0. 6862 6863 scftorture.holdoff= [KNL] 6864 Number of seconds to hold off before starting 6865 test. Defaults to zero for module insertion and 6866 to 10 seconds for built-in smp_call_function() 6867 tests. 6868 6869 scftorture.longwait= [KNL] 6870 Request ridiculously long waits randomly selected 6871 up to the chosen limit in seconds. Zero (the 6872 default) disables this feature. Please note 6873 that requesting even small non-zero numbers of 6874 seconds can result in RCU CPU stall warnings, 6875 softlockup complaints, and so on. 6876 6877 scftorture.nthreads= [KNL] 6878 Number of kthreads to spawn to invoke the 6879 smp_call_function() family of functions. 6880 The default of -1 specifies a number of kthreads 6881 equal to the number of CPUs. 6882 6883 scftorture.onoff_holdoff= [KNL] 6884 Number seconds to wait after the start of the 6885 test before initiating CPU-hotplug operations. 6886 6887 scftorture.onoff_interval= [KNL] 6888 Number seconds to wait between successive 6889 CPU-hotplug operations. Specifying zero (which 6890 is the default) disables CPU-hotplug operations. 6891 6892 scftorture.shutdown_secs= [KNL] 6893 The number of seconds following the start of the 6894 test after which to shut down the system. The 6895 default of zero avoids shutting down the system. 6896 Non-zero values are useful for automated tests. 6897 6898 scftorture.stat_interval= [KNL] 6899 The number of seconds between outputting the 6900 current test statistics to the console. A value 6901 of zero disables statistics output. 6902 6903 scftorture.stutter_cpus= [KNL] 6904 The number of jiffies to wait between each change 6905 to the set of CPUs under test. 6906 6907 scftorture.use_cpus_read_lock= [KNL] 6908 Use use_cpus_read_lock() instead of the default 6909 preempt_disable() to disable CPU hotplug 6910 while invoking one of the smp_call_function*() 6911 functions. 6912 6913 scftorture.verbose= [KNL] 6914 Enable additional printk() statements. 6915 6916 scftorture.weight_single= [KNL] 6917 The probability weighting to use for the 6918 smp_call_function_single() function with a zero 6919 "wait" parameter. A value of -1 selects the 6920 default if all other weights are -1. However, 6921 if at least one weight has some other value, a 6922 value of -1 will instead select a weight of zero. 6923 6924 scftorture.weight_single_wait= [KNL] 6925 The probability weighting to use for the 6926 smp_call_function_single() function with a 6927 non-zero "wait" parameter. See weight_single. 6928 6929 scftorture.weight_many= [KNL] 6930 The probability weighting to use for the 6931 smp_call_function_many() function with a zero 6932 "wait" parameter. See weight_single. 6933 Note well that setting a high probability for 6934 this weighting can place serious IPI load 6935 on the system. 6936 6937 scftorture.weight_many_wait= [KNL] 6938 The probability weighting to use for the 6939 smp_call_function_many() function with a 6940 non-zero "wait" parameter. See weight_single 6941 and weight_many. 6942 6943 scftorture.weight_all= [KNL] 6944 The probability weighting to use for the 6945 smp_call_function_all() function with a zero 6946 "wait" parameter. See weight_single and 6947 weight_many. 6948 6949 scftorture.weight_all_wait= [KNL] 6950 The probability weighting to use for the 6951 smp_call_function_all() function with a 6952 non-zero "wait" parameter. See weight_single 6953 and weight_many. 6954 6955 sdw_mclk_divider=[SDW] 6956 Specify the MCLK divider for Intel SoundWire buses in 6957 case the BIOS does not provide the clock rate properly. 6958 6959 skew_tick= [KNL,EARLY] Offset the periodic timer tick per cpu to mitigate 6960 xtime_lock contention on larger systems, and/or RCU lock 6961 contention on all systems with CONFIG_MAXSMP set. 6962 Format: { "0" | "1" } 6963 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1") 6964 1 -- enable. 6965 Note: increases power consumption, thus should only be 6966 enabled if running jitter sensitive (HPC/RT) workloads. 6967 6968 security= [SECURITY] Choose a legacy "major" security module to 6969 enable at boot. This has been deprecated by the 6970 "lsm=" parameter. 6971 6972 selinux= [SELINUX] Disable or enable SELinux at boot time. 6973 Format: { "0" | "1" } 6974 See security/selinux/Kconfig help text. 6975 0 -- disable. 6976 1 -- enable. 6977 Default value is 1. 6978 6979 serialnumber [BUGS=X86-32] 6980 6981 sev=option[,option...] [X86-64] 6982 6983 debug 6984 Enable debug messages. 6985 6986 nosnp 6987 Do not enable SEV-SNP (applies to host/hypervisor 6988 only). Setting 'nosnp' avoids the RMP check overhead 6989 in memory accesses when users do not want to run 6990 SEV-SNP guests. 6991 6992 shapers= [NET] 6993 Maximal number of shapers. 6994 6995 show_lapic= [APIC,X86] Advanced Programmable Interrupt Controller 6996 Limit apic dumping. The parameter defines the maximal 6997 number of local apics being dumped. Also it is possible 6998 to set it to "all" by meaning -- no limit here. 6999 Format: { 1 (default) | 2 | ... | all }. 7000 The parameter valid if only apic=debug or 7001 apic=verbose is specified. 7002 Example: apic=debug show_lapic=all 7003 7004 slab_debug[=options[,slabs][;[options[,slabs]]...]] [MM] 7005 Enabling slab_debug allows one to determine the 7006 culprit if slab objects become corrupted. Enabling 7007 slab_debug can create guard zones around objects and 7008 may poison objects when not in use. Also tracks the 7009 last alloc / free. For more information see 7010 Documentation/admin-guide/mm/slab.rst. 7011 (slub_debug legacy name also accepted for now) 7012 7013 Using this option implies the "no_hash_pointers" 7014 option which can be undone by adding the 7015 "hash_pointers=always" option. 7016 7017 slab_max_order= [MM] 7018 Determines the maximum allowed order for slabs. 7019 A high setting may cause OOMs due to memory 7020 fragmentation. For more information see 7021 Documentation/admin-guide/mm/slab.rst. 7022 (slub_max_order legacy name also accepted for now) 7023 7024 slab_merge [MM] 7025 Enable merging of slabs with similar size when the 7026 kernel is built without CONFIG_SLAB_MERGE_DEFAULT. 7027 (slub_merge legacy name also accepted for now) 7028 7029 slab_min_objects= [MM] 7030 The minimum number of objects per slab. SLUB will 7031 increase the slab order up to slab_max_order to 7032 generate a sufficiently large slab able to contain 7033 the number of objects indicated. The higher the number 7034 of objects the smaller the overhead of tracking slabs 7035 and the less frequently locks need to be acquired. 7036 For more information see 7037 Documentation/admin-guide/mm/slab.rst. 7038 (slub_min_objects legacy name also accepted for now) 7039 7040 slab_min_order= [MM] 7041 Determines the minimum page order for slabs. Must be 7042 lower or equal to slab_max_order. For more information see 7043 Documentation/admin-guide/mm/slab.rst. 7044 (slub_min_order legacy name also accepted for now) 7045 7046 slab_nomerge [MM] 7047 Disable merging of slabs with similar size. May be 7048 necessary if there is some reason to distinguish 7049 allocs to different slabs, especially in hardened 7050 environments where the risk of heap overflows and 7051 layout control by attackers can usually be 7052 frustrated by disabling merging. This will reduce 7053 most of the exposure of a heap attack to a single 7054 cache (risks via metadata attacks are mostly 7055 unchanged). Debug options disable merging on their 7056 own. 7057 For more information see 7058 Documentation/admin-guide/mm/slab.rst. 7059 (slub_nomerge legacy name also accepted for now) 7060 7061 slab_strict_numa [MM] 7062 Support memory policies on a per object level 7063 in the slab allocator. The default is for memory 7064 policies to be applied at the folio level when 7065 a new folio is needed or a partial folio is 7066 retrieved from the lists. Increases overhead 7067 in the slab fastpaths but gains more accurate 7068 NUMA kernel object placement which helps with slow 7069 interconnects in NUMA systems. 7070 7071 slram= [HW,MTD] 7072 7073 smart2= [HW] 7074 Format: <io1>[,<io2>[,...,<io8>]] 7075 7076 smp.csd_lock_timeout= [KNL] 7077 Specify the period of time in milliseconds 7078 that smp_call_function() and friends will wait 7079 for a CPU to release the CSD lock. This is 7080 useful when diagnosing bugs involving CPUs 7081 disabling interrupts for extended periods 7082 of time. Defaults to 5,000 milliseconds, and 7083 setting a value of zero disables this feature. 7084 This feature may be more efficiently disabled 7085 using the csdlock_debug- kernel parameter. 7086 7087 smp.panic_on_ipistall= [KNL] 7088 If a csd_lock_timeout extends for more than 7089 the specified number of milliseconds, panic the 7090 system. By default, let CSD-lock acquisition 7091 take as long as they take. Specifying 300,000 7092 for this value provides a 5-minute timeout. 7093 7094 smt= [KNL,MIPS,S390,EARLY] Set the maximum number of threads 7095 (logical CPUs) to use per physical CPU on systems 7096 capable of symmetric multithreading (SMT). Will 7097 be capped to the actual hardware limit. 7098 Format: <integer> 7099 Default: -1 (no limit) 7100 7101 softlockup_panic= 7102 [KNL] Should the soft-lockup detector generate panics. 7103 Format: <int> 7104 7105 A value of non-zero instructs the soft-lockup detector 7106 to panic the machine when a soft-lockup duration exceeds 7107 N thresholds. It is also controlled by the kernel.softlockup_panic 7108 sysctl and CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC, which is the 7109 respective build-time switch to that functionality. 7110 7111 softlockup_all_cpu_backtrace= 7112 [KNL] Should the soft-lockup detector generate 7113 backtraces on all cpus. 7114 Format: 0 | 1 7115 7116 sonypi.*= [HW] Sony Programmable I/O Control Device driver 7117 See Documentation/admin-guide/laptops/sonypi.rst 7118 7119 spectre_bhi= [X86] Control mitigation of Branch History Injection 7120 (BHI) vulnerability. This setting affects the 7121 deployment of the HW BHI control and the SW BHB 7122 clearing sequence. 7123 7124 on - (default) Enable the HW or SW mitigation as 7125 needed. This protects the kernel from 7126 both syscalls and VMs. 7127 vmexit - On systems which don't have the HW mitigation 7128 available, enable the SW mitigation on vmexit 7129 ONLY. On such systems, the host kernel is 7130 protected from VM-originated BHI attacks, but 7131 may still be vulnerable to syscall attacks. 7132 off - Disable the mitigation. 7133 7134 spectre_v2= [X86,EARLY] Control mitigation of Spectre variant 2 7135 (indirect branch speculation) vulnerability. 7136 The default operation protects the kernel from 7137 user space attacks. 7138 7139 on - unconditionally enable, implies 7140 spectre_v2_user=on 7141 off - unconditionally disable, implies 7142 spectre_v2_user=off 7143 auto - kernel detects whether your CPU model is 7144 vulnerable 7145 7146 Selecting 'on' will, and 'auto' may, choose a 7147 mitigation method at run time according to the 7148 CPU, the available microcode, the setting of the 7149 CONFIG_MITIGATION_RETPOLINE configuration option, 7150 and the compiler with which the kernel was built. 7151 7152 Selecting 'on' will also enable the mitigation 7153 against user space to user space task attacks. 7154 Selecting specific mitigation does not force enable 7155 user mitigations. 7156 7157 Selecting 'off' will disable both the kernel and 7158 the user space protections. 7159 7160 Specific mitigations can also be selected manually: 7161 7162 retpoline - replace indirect branches 7163 retpoline,generic - Retpolines 7164 retpoline,lfence - LFENCE; indirect branch 7165 retpoline,amd - alias for retpoline,lfence 7166 eibrs - Enhanced/Auto IBRS 7167 eibrs,retpoline - Enhanced/Auto IBRS + Retpolines 7168 eibrs,lfence - Enhanced/Auto IBRS + LFENCE 7169 ibrs - use IBRS to protect kernel 7170 7171 Not specifying this option is equivalent to 7172 spectre_v2=auto. 7173 7174 spectre_v2_user= 7175 [X86] Control mitigation of Spectre variant 2 7176 (indirect branch speculation) vulnerability between 7177 user space tasks 7178 7179 on - Unconditionally enable mitigations. Is 7180 enforced by spectre_v2=on 7181 7182 off - Unconditionally disable mitigations. Is 7183 enforced by spectre_v2=off 7184 7185 prctl - Indirect branch speculation is enabled, 7186 but mitigation can be enabled via prctl 7187 per thread. The mitigation control state 7188 is inherited on fork. 7189 7190 prctl,ibpb 7191 - Like "prctl" above, but only STIBP is 7192 controlled per thread. IBPB is issued 7193 always when switching between different user 7194 space processes. 7195 7196 seccomp 7197 - Same as "prctl" above, but all seccomp 7198 threads will enable the mitigation unless 7199 they explicitly opt out. 7200 7201 seccomp,ibpb 7202 - Like "seccomp" above, but only STIBP is 7203 controlled per thread. IBPB is issued 7204 always when switching between different 7205 user space processes. 7206 7207 auto - Kernel selects the mitigation depending on 7208 the available CPU features and vulnerability. 7209 7210 Default mitigation: "prctl" 7211 7212 Not specifying this option is equivalent to 7213 spectre_v2_user=auto. 7214 7215 spec_rstack_overflow= 7216 [X86,EARLY] Control RAS overflow mitigation on AMD Zen CPUs 7217 7218 off - Disable mitigation 7219 microcode - Enable microcode mitigation only 7220 safe-ret - Enable sw-only safe RET mitigation (default) 7221 ibpb - Enable mitigation by issuing IBPB on 7222 kernel entry 7223 ibpb-vmexit - Issue IBPB only on VMEXIT 7224 (cloud-specific mitigation) 7225 7226 spec_store_bypass_disable= 7227 [HW,EARLY] Control Speculative Store Bypass (SSB) Disable mitigation 7228 (Speculative Store Bypass vulnerability) 7229 7230 Certain CPUs are vulnerable to an exploit against a 7231 a common industry wide performance optimization known 7232 as "Speculative Store Bypass" in which recent stores 7233 to the same memory location may not be observed by 7234 later loads during speculative execution. The idea 7235 is that such stores are unlikely and that they can 7236 be detected prior to instruction retirement at the 7237 end of a particular speculation execution window. 7238 7239 In vulnerable processors, the speculatively forwarded 7240 store can be used in a cache side channel attack, for 7241 example to read memory to which the attacker does not 7242 directly have access (e.g. inside sandboxed code). 7243 7244 This parameter controls whether the Speculative Store 7245 Bypass optimization is used. 7246 7247 On x86 the options are: 7248 7249 on - Unconditionally disable Speculative Store Bypass 7250 off - Unconditionally enable Speculative Store Bypass 7251 auto - Kernel detects whether the CPU model contains an 7252 implementation of Speculative Store Bypass and 7253 picks the most appropriate mitigation. If the 7254 CPU is not vulnerable, "off" is selected. If the 7255 CPU is vulnerable the default mitigation is 7256 architecture and Kconfig dependent. See below. 7257 prctl - Control Speculative Store Bypass per thread 7258 via prctl. Speculative Store Bypass is enabled 7259 for a process by default. The state of the control 7260 is inherited on fork. 7261 seccomp - Same as "prctl" above, but all seccomp threads 7262 will disable SSB unless they explicitly opt out. 7263 7264 Default mitigations: 7265 X86: "prctl" 7266 7267 On powerpc the options are: 7268 7269 on,auto - On Power8 and Power9 insert a store-forwarding 7270 barrier on kernel entry and exit. On Power7 7271 perform a software flush on kernel entry and 7272 exit. 7273 off - No action. 7274 7275 Not specifying this option is equivalent to 7276 spec_store_bypass_disable=auto. 7277 7278 split_llc= 7279 [X86,EARLY] Split the LLC N-ways 7280 7281 When set, the LLC is split this many ways by matching 7282 'core_id % n'. This is setup before SMP bringup and 7283 used during SMP bringup before it knows the full 7284 topology. If your core count doesn't nicely divide by 7285 the number given, you get to keep the pieces. 7286 7287 This is mostly a debug feature to emulate multiple LLCs 7288 on hardware that only have a single LLC. 7289 7290 split_lock_detect= 7291 [X86] Enable split lock detection or bus lock detection 7292 7293 When enabled (and if hardware support is present), atomic 7294 instructions that access data across cache line 7295 boundaries will result in an alignment check exception 7296 for split lock detection or a debug exception for 7297 bus lock detection. 7298 7299 off - not enabled 7300 7301 warn - the kernel will emit rate-limited warnings 7302 about applications triggering the #AC 7303 exception or the #DB exception. This mode is 7304 the default on CPUs that support split lock 7305 detection or bus lock detection. Default 7306 behavior is by #AC if both features are 7307 enabled in hardware. 7308 7309 fatal - the kernel will send SIGBUS to applications 7310 that trigger the #AC exception or the #DB 7311 exception. Default behavior is by #AC if 7312 both features are enabled in hardware. 7313 7314 ratelimit:N - 7315 Set system wide rate limit to N bus locks 7316 per second for bus lock detection. 7317 0 < N <= 1000. 7318 7319 N/A for split lock detection. 7320 7321 7322 If an #AC exception is hit in the kernel or in 7323 firmware (i.e. not while executing in user mode) 7324 the kernel will oops in either "warn" or "fatal" 7325 mode. 7326 7327 #DB exception for bus lock is triggered only when 7328 CPL > 0. 7329 7330 srbds= [X86,INTEL,EARLY] 7331 Control the Special Register Buffer Data Sampling 7332 (SRBDS) mitigation. 7333 7334 Certain CPUs are vulnerable to an MDS-like 7335 exploit which can leak bits from the random 7336 number generator. 7337 7338 By default, this issue is mitigated by 7339 microcode. However, the microcode fix can cause 7340 the RDRAND and RDSEED instructions to become 7341 much slower. Among other effects, this will 7342 result in reduced throughput from /dev/urandom. 7343 7344 The microcode mitigation can be disabled with 7345 the following option: 7346 7347 off: Disable mitigation and remove 7348 performance impact to RDRAND and RDSEED 7349 7350 srcutree.big_cpu_lim [KNL] 7351 Specifies the number of CPUs constituting a 7352 large system, such that srcu_struct structures 7353 should immediately allocate an srcu_node array. 7354 This kernel-boot parameter defaults to 128, 7355 but takes effect only when the low-order four 7356 bits of srcutree.convert_to_big is equal to 3 7357 (decide at boot). 7358 7359 srcutree.convert_to_big [KNL] 7360 Specifies under what conditions an SRCU tree 7361 srcu_struct structure will be converted to big 7362 form, that is, with an rcu_node tree: 7363 7364 0: Never. 7365 1: At init_srcu_struct() time. 7366 2: When rcutorture decides to. 7367 3: Decide at boot time (default). 7368 0x1X: Above plus if high contention. 7369 7370 Either way, the srcu_node tree will be sized based 7371 on the actual runtime number of CPUs (nr_cpu_ids) 7372 instead of the compile-time CONFIG_NR_CPUS. 7373 7374 srcutree.counter_wrap_check [KNL] 7375 Specifies how frequently to check for 7376 grace-period sequence counter wrap for the 7377 srcu_data structure's ->srcu_gp_seq_needed field. 7378 The greater the number of bits set in this kernel 7379 parameter, the less frequently counter wrap will 7380 be checked for. Note that the bottom two bits 7381 are ignored. 7382 7383 srcutree.exp_holdoff [KNL] 7384 Specifies how many nanoseconds must elapse 7385 since the end of the last SRCU grace period for 7386 a given srcu_struct until the next normal SRCU 7387 grace period will be considered for automatic 7388 expediting. Set to zero to disable automatic 7389 expediting. 7390 7391 srcutree.srcu_max_nodelay [KNL] 7392 Specifies the number of no-delay instances 7393 per jiffy for which the SRCU grace period 7394 worker thread will be rescheduled with zero 7395 delay. Beyond this limit, worker thread will 7396 be rescheduled with a sleep delay of one jiffy. 7397 7398 srcutree.srcu_max_nodelay_phase [KNL] 7399 Specifies the per-grace-period phase, number of 7400 non-sleeping polls of readers. Beyond this limit, 7401 grace period worker thread will be rescheduled 7402 with a sleep delay of one jiffy, between each 7403 rescan of the readers, for a grace period phase. 7404 7405 srcutree.srcu_retry_check_delay [KNL] 7406 Specifies number of microseconds of non-sleeping 7407 delay between each non-sleeping poll of readers. 7408 7409 srcutree.small_contention_lim [KNL] 7410 Specifies the number of update-side contention 7411 events per jiffy will be tolerated before 7412 initiating a conversion of an srcu_struct 7413 structure to big form. Note that the value of 7414 srcutree.convert_to_big must have the 0x10 bit 7415 set for contention-based conversions to occur. 7416 7417 ssbd= [ARM64,HW,EARLY] 7418 Speculative Store Bypass Disable control 7419 7420 On CPUs that are vulnerable to the Speculative 7421 Store Bypass vulnerability and offer a 7422 firmware based mitigation, this parameter 7423 indicates how the mitigation should be used: 7424 7425 force-on: Unconditionally enable mitigation for 7426 for both kernel and userspace 7427 force-off: Unconditionally disable mitigation for 7428 for both kernel and userspace 7429 kernel: Always enable mitigation in the 7430 kernel, and offer a prctl interface 7431 to allow userspace to register its 7432 interest in being mitigated too. 7433 7434 stack_guard_gap= [MM] 7435 override the default stack gap protection. The value 7436 is in page units and it defines how many pages prior 7437 to (for stacks growing down) resp. after (for stacks 7438 growing up) the main stack are reserved for no other 7439 mapping. Default value is 256 pages. 7440 7441 stack_depot_disable= [KNL,EARLY] 7442 Setting this to true through kernel command line will 7443 disable the stack depot thereby saving the static memory 7444 consumed by the stack hash table. By default this is set 7445 to false. 7446 7447 stack_depot_max_pools= [KNL,EARLY] 7448 Specify the maximum number of pools to use for storing 7449 stack traces. Pools are allocated on-demand up to this 7450 limit. Default value is 8191 pools. 7451 7452 stacktrace [FTRACE] 7453 Enable the stack tracer on boot up. 7454 7455 stacktrace_filter=[function-list] 7456 [FTRACE] Limit the functions that the stack tracer 7457 will trace at boot up. function-list is a comma-separated 7458 list of functions. This list can be changed at run 7459 time by the stack_trace_filter file in the debugfs 7460 tracing directory. Note, this enables stack tracing 7461 and the stacktrace above is not needed. 7462 7463 sti= [PARISC,HW] 7464 Format: <num> 7465 Set the STI (builtin display/keyboard on the HP-PARISC 7466 machines) console (graphic card) which should be used 7467 as the initial boot-console. 7468 See also comment in drivers/video/sticore.c. 7469 7470 sti_font= [HW] 7471 See comment in drivers/video/sticore.c. 7472 7473 stifb= [HW] 7474 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]] 7475 7476 strict_sas_size= 7477 [X86] 7478 Format: <bool> 7479 Enable or disable strict sigaltstack size checks 7480 against the required signal frame size which 7481 depends on the supported FPU features. This can 7482 be used to filter out binaries which have 7483 not yet been made aware of AT_MINSIGSTKSZ. 7484 7485 stress_hpt [PPC,EARLY] 7486 Limits the number of kernel HPT entries in the hash 7487 page table to increase the rate of hash page table 7488 faults on kernel addresses. 7489 7490 stress_slb [PPC,EARLY] 7491 Limits the number of kernel SLB entries, and flushes 7492 them frequently to increase the rate of SLB faults 7493 on kernel addresses. 7494 7495 no_slb_preload [PPC,EARLY] 7496 Disables slb preloading for userspace. 7497 7498 sunrpc.min_resvport= 7499 sunrpc.max_resvport= 7500 [NFS,SUNRPC] 7501 SunRPC servers often require that client requests 7502 originate from a privileged port (i.e. a port in the 7503 range 0 < portnr < 1024). 7504 An administrator who wishes to reserve some of these 7505 ports for other uses may adjust the range that the 7506 kernel's sunrpc client considers to be privileged 7507 using these two parameters to set the minimum and 7508 maximum port values. 7509 7510 sunrpc.svc_rpc_per_connection_limit= 7511 [NFS,SUNRPC] 7512 Limit the number of requests that the server will 7513 process in parallel from a single connection. 7514 The default value is 0 (no limit). 7515 7516 sunrpc.pool_mode= 7517 [NFS] 7518 Deprecated. The NFS server now always uses one 7519 service thread pool per NUMA node (equivalent to a 7520 single global pool on non-NUMA machines). All of 7521 the previously accepted values (auto, global, 7522 percpu, pernode) are still accepted for backward 7523 compatibility but are ignored: the mode is always 7524 pernode, and reads always return "pernode". 7525 7526 sunrpc.tcp_slot_table_entries= 7527 sunrpc.udp_slot_table_entries= 7528 [NFS,SUNRPC] 7529 Sets the upper limit on the number of simultaneous 7530 RPC calls that can be sent from the client to a 7531 server. Increasing these values may allow you to 7532 improve throughput, but will also increase the 7533 amount of memory reserved for use by the client. 7534 7535 suspend.pm_test_delay= 7536 [SUSPEND] 7537 Sets the number of seconds to remain in a suspend test 7538 mode before resuming the system (see 7539 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG 7540 is set. Default value is 5. 7541 7542 svm= [PPC] 7543 Format: { on | off | y | n | 1 | 0 } 7544 This parameter controls use of the Protected 7545 Execution Facility on pSeries. 7546 7547 swiotlb= [ARM,PPC,MIPS,X86,S390,EARLY] 7548 Format: { <int> [,<int>] | force | noforce | track_hiwater} 7549 <int> -- Number of I/O TLB slabs 7550 <int> -- Second integer after comma. Number of swiotlb 7551 areas with their own lock. Will be rounded up 7552 to a power of 2. 7553 force -- force using of bounce buffers even if they 7554 wouldn't be automatically used by the kernel 7555 noforce -- Never use bounce buffers (for debugging) 7556 track_hiwater -- Track high watermark of swiotlb buffers. 7557 Only available when CONFIG_DEBUG_FS is set. 7558 7559 switches= [HW,M68k,EARLY] 7560 7561 sysctl.*= [KNL] 7562 Set a sysctl parameter, right before loading the init 7563 process, as if the value was written to the respective 7564 /proc/sys/... file. Both '.' and '/' are recognized as 7565 separators. Unrecognized parameters and invalid values 7566 are reported in the kernel log. Sysctls registered 7567 later by a loaded module cannot be set this way. 7568 Example: sysctl.vm.swappiness=40 7569 7570 sysrq_always_enabled 7571 [KNL] 7572 Ignore sysrq setting - this boot parameter will 7573 neutralize any effect of /proc/sys/kernel/sysrq. 7574 Useful for debugging. 7575 7576 tcpmhash_entries= [KNL,NET] 7577 Set the number of tcp_metrics_hash slots. 7578 Default value is 8192 or 16384 depending on total 7579 ram pages. This is used to specify the TCP metrics 7580 cache size. See Documentation/networking/ip-sysctl.rst 7581 "tcp_no_metrics_save" section for more details. 7582 7583 tdfx= [HW,DRM] 7584 7585 test_suspend= [SUSPEND] 7586 Format: { "mem" | "standby" | "freeze" }[,N] 7587 Specify "mem" (for Suspend-to-RAM) or "standby" (for 7588 standby suspend) or "freeze" (for suspend type freeze) 7589 as the system sleep state during system startup with 7590 the optional capability to repeat N number of times. 7591 The system is woken from this state using a 7592 wakeup-capable RTC alarm. 7593 7594 thash_entries= [KNL,NET] 7595 Set number of hash buckets for TCP connection 7596 7597 thermal.act= [HW,ACPI] 7598 -1: disable all active trip points in all thermal zones 7599 <degrees C>: override all lowest active trip points 7600 7601 thermal.crt= [HW,ACPI] 7602 -1: disable all critical trip points in all thermal zones 7603 <degrees C>: override all critical trip points 7604 7605 thermal.off= [HW,ACPI] 7606 1: disable ACPI thermal control 7607 7608 thermal.psv= [HW,ACPI] 7609 -1: disable all passive trip points 7610 <degrees C>: override all passive trip points to this 7611 value 7612 7613 thermal.tzp= [HW,ACPI] 7614 Specify global default ACPI thermal zone polling rate 7615 <deci-seconds>: poll all this frequency 7616 0: no polling (default) 7617 7618 thp_anon= [KNL] 7619 Format: <size>[KMG],<size>[KMG]:<state>;<size>[KMG]-<size>[KMG]:<state> 7620 state is one of "always", "madvise", "never" or "inherit". 7621 Control the default behavior of the system with respect 7622 to anonymous transparent hugepages. 7623 Can be used multiple times for multiple anon THP sizes. 7624 See Documentation/admin-guide/mm/transhuge.rst for more 7625 details. 7626 7627 threadirqs [KNL,EARLY] 7628 Force threading of all interrupt handlers except those 7629 marked explicitly IRQF_NO_THREAD. 7630 7631 thp_shmem= [KNL] 7632 Format: <size>[KMG],<size>[KMG]:<policy>;<size>[KMG]-<size>[KMG]:<policy> 7633 Control the default policy of each hugepage size for the 7634 internal shmem mount. <policy> is one of policies available 7635 for the shmem mount ("always", "inherit", "never", "within_size", 7636 and "advise"). 7637 It can be used multiple times for multiple shmem THP sizes. 7638 See Documentation/admin-guide/mm/transhuge.rst for more 7639 details. 7640 7641 tlbi= [X86-64] 7642 Format: {ipi} 7643 ipi: switch to IPI-based TLB flushing 7644 7645 topology= [S390,EARLY] 7646 Format: {off | on} 7647 Specify if the kernel should make use of the cpu 7648 topology information if the hardware supports this. 7649 The scheduler will make use of this information and 7650 e.g. base its process migration decisions on it. 7651 Default is on. 7652 7653 torture.disable_onoff_at_boot= [KNL] 7654 Prevent the CPU-hotplug component of torturing 7655 until after init has spawned. 7656 7657 torture.ftrace_dump_at_shutdown= [KNL] 7658 Dump the ftrace buffer at torture-test shutdown, 7659 even if there were no errors. This can be a 7660 very costly operation when many torture tests 7661 are running concurrently, especially on systems 7662 with rotating-rust storage. 7663 7664 torture.verbose_sleep_frequency= [KNL] 7665 Specifies how many verbose printk()s should be 7666 emitted between each sleep. The default of zero 7667 disables verbose-printk() sleeping. 7668 7669 torture.verbose_sleep_duration= [KNL] 7670 Duration of each verbose-printk() sleep in jiffies. 7671 7672 tpm.disable_pcr_integrity= [HW,TPM] 7673 Do not protect PCR registers from unintended physical 7674 access, or interposers in the bus by the means of 7675 having an integrity protected session wrapped around 7676 TPM2_PCR_Extend command. Consider this in a situation 7677 where TPM is heavily utilized by IMA, thus protection 7678 causing a major performance hit, and the space where 7679 machines are deployed is by other means guarded. 7680 7681 tpm_crb_ffa.busy_timeout_ms= [ARM64,TPM] 7682 Maximum time in milliseconds to retry sending a message 7683 to the TPM service before giving up. This parameter controls 7684 how long the system will continue retrying when the TPM 7685 service is busy. 7686 Format: <unsigned int> 7687 Default: 2000 (2 seconds) 7688 7689 tpm_suspend_pcr=[HW,TPM] 7690 Format: integer pcr id 7691 Specify that at suspend time, the tpm driver 7692 should extend the specified pcr with zeros, 7693 as a workaround for some chips which fail to 7694 flush the last written pcr on TPM_SaveState. 7695 This will guarantee that all the other pcrs 7696 are saved. 7697 7698 tpm_tis.interrupts= [HW,TPM] 7699 Enable interrupts for the MMIO based physical layer 7700 for the FIFO interface. By default it is set to false 7701 (0). For more information about TPM hardware interfaces 7702 defined by Trusted Computing Group (TCG) see 7703 https://trustedcomputinggroup.org/resource/pc-client-platform-tpm-profile-ptp-specification/ 7704 7705 tp_printk [FTRACE] 7706 Have the tracepoints sent to printk as well as the 7707 tracing ring buffer. This is useful for early boot up 7708 where the system hangs or reboots and does not give the 7709 option for reading the tracing buffer or performing a 7710 ftrace_dump_on_oops. 7711 7712 To turn off having tracepoints sent to printk, 7713 echo 0 > /proc/sys/kernel/tracepoint_printk 7714 Note, echoing 1 into this file without the 7715 tp_printk kernel cmdline option has no effect. 7716 7717 The tp_printk_stop_on_boot (see below) can also be used 7718 to stop the printing of events to console at 7719 late_initcall_sync. 7720 7721 ** CAUTION ** 7722 7723 Having tracepoints sent to printk() and activating high 7724 frequency tracepoints such as irq or sched, can cause 7725 the system to live lock. 7726 7727 tp_printk_stop_on_boot [FTRACE] 7728 When tp_printk (above) is set, it can cause a lot of noise 7729 on the console. It may be useful to only include the 7730 printing of events during boot up, as user space may 7731 make the system inoperable. 7732 7733 This command line option will stop the printing of events 7734 to console at the late_initcall_sync() time frame. 7735 7736 trace_buf_size=nn[KMG] 7737 [FTRACE] will set tracing buffer size on each cpu. 7738 7739 trace_clock= [FTRACE] Set the clock used for tracing events 7740 at boot up. 7741 local - Use the per CPU time stamp counter 7742 (converted into nanoseconds). Fast, but 7743 depending on the architecture, may not be 7744 in sync between CPUs. 7745 global - Event time stamps are synchronized across 7746 CPUs. May be slower than the local clock, 7747 but better for some race conditions. 7748 counter - Simple counting of events (1, 2, ..) 7749 note, some counts may be skipped due to the 7750 infrastructure grabbing the clock more than 7751 once per event. 7752 uptime - Use jiffies as the time stamp. 7753 perf - Use the same clock that perf uses. 7754 mono - Use ktime_get_mono_fast_ns() for time stamps. 7755 mono_raw - Use ktime_get_raw_fast_ns() for time 7756 stamps. 7757 boot - Use ktime_get_boot_fast_ns() for time stamps. 7758 Architectures may add more clocks. See 7759 Documentation/trace/ftrace.rst for more details. 7760 7761 trace_event=[event-list] 7762 [FTRACE] Set and start specified trace events in order 7763 to facilitate early boot debugging. The event-list is a 7764 comma-separated list of trace events to enable. See 7765 also Documentation/trace/events.rst 7766 7767 To enable modules, use :mod: keyword: 7768 7769 trace_event=:mod:<module> 7770 7771 The value before :mod: will only enable specific events 7772 that are part of the module. See the above mentioned 7773 document for more information. 7774 7775 trace_instance=[instance-info] 7776 [FTRACE] Create a ring buffer instance early in boot up. 7777 This will be listed in: 7778 7779 /sys/kernel/tracing/instances 7780 7781 Events can be enabled at the time the instance is created 7782 via: 7783 7784 trace_instance=<name>,<system1>:<event1>,<system2>:<event2> 7785 7786 Note, the "<system*>:" portion is optional if the event is 7787 unique. 7788 7789 trace_instance=foo,sched:sched_switch,irq_handler_entry,initcall 7790 7791 will enable the "sched_switch" event (note, the "sched:" is optional, and 7792 the same thing would happen if it was left off). The irq_handler_entry 7793 event, and all events under the "initcall" system. 7794 7795 Flags can be added to the instance to modify its behavior when it is 7796 created. The flags are separated by '^'. 7797 7798 The available flags are: 7799 7800 traceoff - Have the tracing instance tracing disabled after it is created. 7801 traceprintk - Have trace_printk() write into this trace instance 7802 (note, "printk" and "trace_printk" can also be used) 7803 7804 trace_instance=foo^traceoff^traceprintk,sched,irq 7805 7806 The flags must come before the defined events. 7807 7808 If memory has been reserved (see memmap for x86), the instance 7809 can use that memory: 7810 7811 memmap=12M$0x284500000 trace_instance=boot_map@0x284500000:12M 7812 7813 The above will create a "boot_map" instance that uses the physical 7814 memory at 0x284500000 that is 12Megs. The per CPU buffers of that 7815 instance will be split up accordingly. 7816 7817 Alternatively, the memory can be reserved by the reserve_mem option: 7818 7819 reserve_mem=12M:4096:trace trace_instance=boot_map@trace 7820 7821 This will reserve 12 megabytes at boot up with a 4096 byte alignment 7822 and place the ring buffer in this memory. Note that due to KASLR, the 7823 memory may not be the same location each time, which will not preserve 7824 the buffer content. 7825 7826 Also note that the layout of the ring buffer data may change between 7827 kernel versions where the validator will fail and reset the ring buffer 7828 if the layout is not the same as the previous kernel. 7829 7830 If the ring buffer is used for persistent bootups and has events enabled, 7831 it is recommend to disable tracing so that events from a previous boot do not 7832 mix with events of the current boot (unless you are debugging a random crash 7833 at boot up). 7834 7835 reserve_mem=12M:4096:trace trace_instance=boot_map^traceoff^traceprintk@trace,sched,irq 7836 7837 Note, saving the trace buffer across reboots does require that the system 7838 is set up to not wipe memory. For instance, CONFIG_RESET_ATTACK_MITIGATION 7839 can force a memory reset on boot which will clear any trace that was stored. 7840 This is just one of many ways that can clear memory. Make sure your system 7841 keeps the content of memory across reboots before relying on this option. 7842 7843 NB: Both the mapped address and size must be page aligned for the architecture. 7844 7845 See also Documentation/trace/debugging.rst 7846 7847 7848 trace_options=[option-list] 7849 [FTRACE] Enable or disable tracer options at boot. 7850 The option-list is a comma delimited list of options 7851 that can be enabled or disabled just as if you were 7852 to echo the option name into 7853 7854 /sys/kernel/tracing/trace_options 7855 7856 For example, to enable stacktrace option (to dump the 7857 stack trace of each event), add to the command line: 7858 7859 trace_options=stacktrace 7860 7861 See also Documentation/trace/ftrace.rst "trace options" 7862 section. 7863 7864 trace_trigger=[trigger-list] 7865 [FTRACE] Add an event trigger on specific events. 7866 Set a trigger on top of a specific event, with an optional 7867 filter. 7868 7869 The format is "trace_trigger=<event>.<trigger>[ if <filter>],..." 7870 Where more than one trigger may be specified that are comma delimited. 7871 7872 For example: 7873 7874 trace_trigger="sched_switch.stacktrace if prev_state == 2" 7875 7876 The above will enable the "stacktrace" trigger on the "sched_switch" 7877 event but only trigger it if the "prev_state" of the "sched_switch" 7878 event is "2" (TASK_UNINTERRUPTIBLE). 7879 7880 See also "Event triggers" in Documentation/trace/events.rst 7881 7882 7883 traceoff_after_boot 7884 [FTRACE] Sometimes tracing is used to debug issues 7885 during the boot process. Since the trace buffer has a 7886 limited amount of storage, it may be prudent to 7887 disable tracing after the boot is finished, otherwise 7888 the critical information may be overwritten. With this 7889 option, the main tracing buffer will be turned off at 7890 the end of the boot process. 7891 7892 traceoff_on_warning 7893 [FTRACE] enable this option to disable tracing when a 7894 warning is hit. This turns off "tracing_on". Tracing can 7895 be enabled again by echoing '1' into the "tracing_on" 7896 file located in /sys/kernel/tracing/ 7897 7898 This option is useful, as it disables the trace before 7899 the WARNING dump is called, which prevents the trace to 7900 be filled with content caused by the warning output. 7901 7902 This option can also be set at run time via the sysctl 7903 option: kernel/traceoff_on_warning 7904 7905 transparent_hugepage= 7906 [KNL] 7907 Format: [always|madvise|never] 7908 Can be used to control the default behavior of the system 7909 with respect to transparent hugepages. 7910 See Documentation/admin-guide/mm/transhuge.rst 7911 for more details. 7912 7913 transparent_hugepage_shmem= [KNL] 7914 Format: [always|within_size|advise|never|deny|force] 7915 Can be used to control the hugepage allocation policy for 7916 the internal shmem mount. 7917 See Documentation/admin-guide/mm/transhuge.rst 7918 for more details. 7919 7920 transparent_hugepage_tmpfs= [KNL] 7921 Format: [always|within_size|advise|never] 7922 Can be used to control the default hugepage allocation policy 7923 for the tmpfs mount. 7924 See Documentation/admin-guide/mm/transhuge.rst 7925 for more details. 7926 7927 trusted.source= [KEYS] 7928 Format: <string> 7929 This parameter identifies the trust source as a backend 7930 for trusted keys implementation. Supported trust 7931 sources: 7932 - "tpm" 7933 - "tee" 7934 - "caam" 7935 - "dcp" 7936 - "pkwm" 7937 If not specified then it defaults to iterating through 7938 the trust source list starting with TPM and assigns the 7939 first trust source as a backend which is initialized 7940 successfully during iteration. 7941 7942 trusted.debug= [KEYS] 7943 Format: <bool> 7944 Enable trusted keys debug traces at runtime when 7945 CONFIG_TRUSTED_KEYS_DEBUG=y. 7946 7947 To make the traces visible after enabling the option, 7948 use trusted.dyndbg='+p' as needed. By convention, 7949 the subsystem uses pr_debug() for these traces. 7950 7951 SAFETY: The traces can leak sensitive data, so be 7952 cautious before enabling this. They remain inactive 7953 unless this parameter is set this option to a true 7954 value. 7955 7956 Default: false 7957 7958 trusted.rng= [KEYS] 7959 Format: <string> 7960 The RNG used to generate key material for trusted keys. 7961 Can be one of: 7962 - "kernel" 7963 - the same value as trusted.source: "tpm" or "tee" 7964 - "default" 7965 If not specified, "default" is used. In this case, 7966 the RNG's choice is left to each individual trust source. 7967 7968 trusted.dcp_use_otp_key 7969 This is intended to be used in combination with 7970 trusted.source=dcp and will select the DCP OTP key 7971 instead of the DCP UNIQUE key blob encryption. 7972 7973 trusted.dcp_skip_zk_test 7974 This is intended to be used in combination with 7975 trusted.source=dcp and will disable the check if the 7976 blob key is all zeros. This is helpful for situations where 7977 having this key zero'ed is acceptable. E.g. in testing 7978 scenarios. 7979 7980 tsa= [X86] Control mitigation for Transient Scheduler 7981 Attacks on AMD CPUs. Search the following in your 7982 favourite search engine for more details: 7983 7984 "Technical guidance for mitigating transient scheduler 7985 attacks". 7986 7987 off - disable the mitigation 7988 on - enable the mitigation (default) 7989 user - mitigate only user/kernel transitions 7990 vm - mitigate only guest/host transitions 7991 7992 7993 tsc= Disable clocksource stability checks for TSC. 7994 Format: <string> 7995 [x86] reliable: mark tsc clocksource as reliable, this 7996 disables clocksource verification at runtime, as well 7997 as the stability checks done at bootup. Used to enable 7998 high-resolution timer mode on older hardware, and in 7999 virtualized environment. 8000 [x86] noirqtime: Do not use TSC to do irq accounting. 8001 Used to run time disable IRQ_TIME_ACCOUNTING on any 8002 platforms where RDTSC is slow and this accounting 8003 can add overhead. 8004 [x86] unstable: mark the TSC clocksource as unstable, this 8005 marks the TSC unconditionally unstable at bootup and 8006 avoids any further wobbles once the TSC watchdog notices. 8007 [x86] nowatchdog: disable clocksource watchdog. Used 8008 in situations with strict latency requirements (where 8009 interruptions from clocksource watchdog are not 8010 acceptable). 8011 [x86] recalibrate: force recalibration against a HW timer 8012 (HPET or PM timer) on systems whose TSC frequency was 8013 obtained from HW or FW using either an MSR or CPUID(0x15). 8014 Warn if the difference is more than 500 ppm. 8015 [x86] watchdog: Enforce the clocksource watchdog on TSC 8016 8017 tsc_early_khz= [X86,EARLY] Skip early TSC calibration and use the given 8018 value instead. Useful when the early TSC frequency discovery 8019 procedure is not reliable, such as on overclocked systems 8020 with CPUID.16h support and partial CPUID.15h support. 8021 Format: <unsigned int> 8022 8023 tsx= [X86] Control Transactional Synchronization 8024 Extensions (TSX) feature in Intel processors that 8025 support TSX control. 8026 8027 This parameter controls the TSX feature. The options are: 8028 8029 on - Enable TSX on the system. Although there are 8030 mitigations for all known security vulnerabilities, 8031 TSX has been known to be an accelerator for 8032 several previous speculation-related CVEs, and 8033 so there may be unknown security risks associated 8034 with leaving it enabled. 8035 8036 off - Disable TSX on the system. (Note that this 8037 option takes effect only on newer CPUs which are 8038 not vulnerable to MDS, i.e., have 8039 MSR_IA32_ARCH_CAPABILITIES.MDS_NO=1 and which get 8040 the new IA32_TSX_CTRL MSR through a microcode 8041 update. This new MSR allows for the reliable 8042 deactivation of the TSX functionality.) 8043 8044 auto - Disable TSX if X86_BUG_TAA is present, 8045 otherwise enable TSX on the system. 8046 8047 Not specifying this option is equivalent to tsx=off. 8048 8049 See Documentation/admin-guide/hw-vuln/tsx_async_abort.rst 8050 for more details. 8051 8052 tsx_async_abort= [X86,INTEL,EARLY] Control mitigation for the TSX Async 8053 Abort (TAA) vulnerability. 8054 8055 Similar to Micro-architectural Data Sampling (MDS) 8056 certain CPUs that support Transactional 8057 Synchronization Extensions (TSX) are vulnerable to an 8058 exploit against CPU internal buffers which can forward 8059 information to a disclosure gadget under certain 8060 conditions. 8061 8062 In vulnerable processors, the speculatively forwarded 8063 data can be used in a cache side channel attack, to 8064 access data to which the attacker does not have direct 8065 access. 8066 8067 This parameter controls the TAA mitigation. The 8068 options are: 8069 8070 full - Enable TAA mitigation on vulnerable CPUs 8071 if TSX is enabled. 8072 8073 full,nosmt - Enable TAA mitigation and disable SMT on 8074 vulnerable CPUs. If TSX is disabled, SMT 8075 is not disabled because CPU is not 8076 vulnerable to cross-thread TAA attacks. 8077 off - Unconditionally disable TAA mitigation 8078 8079 On MDS-affected machines, tsx_async_abort=off can be 8080 prevented by an active MDS mitigation as both vulnerabilities 8081 are mitigated with the same mechanism so in order to disable 8082 this mitigation, you need to specify mds=off too. 8083 8084 Not specifying this option is equivalent to 8085 tsx_async_abort=full. On CPUs which are MDS affected 8086 and deploy MDS mitigation, TAA mitigation is not 8087 required and doesn't provide any additional 8088 mitigation. 8089 8090 For details see: 8091 Documentation/admin-guide/hw-vuln/tsx_async_abort.rst 8092 8093 turbografx.map[2|3]= [HW,JOY] 8094 TurboGraFX parallel port interface 8095 Format: 8096 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7> 8097 See also Documentation/input/devices/joystick-parport.rst 8098 8099 udbg-immortal [PPC] When debugging early kernel crashes that 8100 happen after console_init() and before a proper 8101 console driver takes over, this boot options might 8102 help "seeing" what's going on. 8103 8104 uhash_entries= [KNL,NET] 8105 Set number of hash buckets for UDP/UDP-Lite connections 8106 8107 uhci-hcd.ignore_oc= 8108 [USB] Ignore overcurrent events (default N). 8109 Some badly-designed motherboards generate lots of 8110 bogus events, for ports that aren't wired to 8111 anything. Set this parameter to avoid log spamming. 8112 Note that genuine overcurrent events won't be 8113 reported either. 8114 8115 unaligned_scalar_speed= 8116 [RISCV] 8117 Format: {slow | fast | unsupported} 8118 Allow skipping scalar unaligned access speed tests. This 8119 is useful for testing alternative code paths and to skip 8120 the tests in environments where they run too slowly. All 8121 CPUs must have the same scalar unaligned access speed. 8122 8123 unaligned_vector_speed= 8124 [RISCV] 8125 Format: {slow | fast | unsupported} 8126 Allow skipping vector unaligned access speed tests. This 8127 is useful for testing alternative code paths and to skip 8128 the tests in environments where they run too slowly. All 8129 CPUs must have the same vector unaligned access speed. 8130 8131 unknown_nmi_panic 8132 [X86] Cause panic on unknown NMI. 8133 8134 unwind_debug [X86-64,EARLY] 8135 Enable unwinder debug output. This can be 8136 useful for debugging certain unwinder error 8137 conditions, including corrupt stacks and 8138 bad/missing unwinder metadata. 8139 8140 usbcore.authorized_default= 8141 [USB] Default USB device authorization: 8142 (default -1 = authorized (same as 1), 8143 0 = not authorized, 1 = authorized, 2 = authorized 8144 if device connected to internal port) 8145 8146 usbcore.autosuspend= 8147 [USB] The autosuspend time delay (in seconds) used 8148 for newly-detected USB devices (default 2). This 8149 is the time required before an idle device will be 8150 autosuspended. Devices for which the delay is set 8151 to a negative value won't be autosuspended at all. 8152 8153 usbcore.usbfs_snoop= 8154 [USB] Set to log all usbfs traffic (default 0 = off). 8155 8156 usbcore.usbfs_snoop_max= 8157 [USB] Maximum number of bytes to snoop in each URB 8158 (default = 65536). 8159 8160 usbcore.blinkenlights= 8161 [USB] Set to cycle leds on hubs (default 0 = off). 8162 8163 usbcore.old_scheme_first= 8164 [USB] Start with the old device initialization 8165 scheme (default 0 = off). 8166 8167 usbcore.usbfs_memory_mb= 8168 [USB] Memory limit (in MB) for buffers allocated by 8169 usbfs (default = 16, 0 = max = 2047). 8170 8171 usbcore.use_both_schemes= 8172 [USB] Try the other device initialization scheme 8173 if the first one fails (default 1 = enabled). 8174 8175 usbcore.initial_descriptor_timeout= 8176 [USB] Specifies timeout for the initial 64-byte 8177 USB_REQ_GET_DESCRIPTOR request in milliseconds 8178 (default 5000 = 5.0 seconds). 8179 8180 usbcore.nousb [USB] Disable the USB subsystem 8181 8182 usbcore.quirks= 8183 [USB] A list of quirk entries to augment the built-in 8184 usb core quirk list. List entries are separated by 8185 commas. Each entry has the form 8186 VendorID:ProductID:Flags. The IDs are 4-digit hex 8187 numbers and Flags is a set of letters. Each letter 8188 will change the built-in quirk; setting it if it is 8189 clear and clearing it if it is set. The letters have 8190 the following meanings: 8191 a = USB_QUIRK_STRING_FETCH_255 (string 8192 descriptors must not be fetched using 8193 a 255-byte read); 8194 b = USB_QUIRK_RESET_RESUME (device can't resume 8195 correctly so reset it instead); 8196 c = USB_QUIRK_NO_SET_INTF (device can't handle 8197 Set-Interface requests); 8198 d = USB_QUIRK_CONFIG_INTF_STRINGS (device can't 8199 handle its Configuration or Interface 8200 strings); 8201 e = USB_QUIRK_RESET (device can't be reset 8202 (e.g morph devices), don't use reset); 8203 f = USB_QUIRK_HONOR_BNUMINTERFACES (device has 8204 more interface descriptions than the 8205 bNumInterfaces count, and can't handle 8206 talking to these interfaces); 8207 g = USB_QUIRK_DELAY_INIT (device needs a pause 8208 during initialization, after we read 8209 the device descriptor); 8210 h = USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL (For 8211 high speed and super speed interrupt 8212 endpoints, the USB 2.0 and USB 3.0 spec 8213 require the interval in microframes (1 8214 microframe = 125 microseconds) to be 8215 calculated as interval = 2 ^ 8216 (bInterval-1). 8217 Devices with this quirk report their 8218 bInterval as the result of this 8219 calculation instead of the exponent 8220 variable used in the calculation); 8221 i = USB_QUIRK_DEVICE_QUALIFIER (device can't 8222 handle device_qualifier descriptor 8223 requests); 8224 j = USB_QUIRK_IGNORE_REMOTE_WAKEUP (device 8225 generates spurious wakeup, ignore 8226 remote wakeup capability); 8227 k = USB_QUIRK_NO_LPM (device can't handle Link 8228 Power Management); 8229 l = USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL 8230 (Device reports its bInterval as linear 8231 frames instead of the USB 2.0 8232 calculation); 8233 m = USB_QUIRK_DISCONNECT_SUSPEND (Device needs 8234 to be disconnected before suspend to 8235 prevent spurious wakeup); 8236 n = USB_QUIRK_DELAY_CTRL_MSG (Device needs a 8237 pause after every control message); 8238 o = USB_QUIRK_HUB_SLOW_RESET (Hub needs extra 8239 delay after resetting its port); 8240 p = USB_QUIRK_SHORT_SET_ADDRESS_REQ_TIMEOUT 8241 (Reduce timeout of the SET_ADDRESS 8242 request from 5000 ms to 500 ms); 8243 q = USB_QUIRK_FORCE_ONE_CONFIG (Device 8244 claims zero configurations, 8245 forcing to 1); 8246 r = USB_QUIRK_WINDOWS_CONFIG_REQ_SIZE (Device 8247 fails during initialization when asked for 8248 9-bytes configuration descriptor request. 8249 Ask for 255-bytes request instead to mirror 8250 Windows' behavior); 8251 Example: quirks=0781:5580:bk,0a5c:5834:gij 8252 8253 usbhid.mousepoll= 8254 [USBHID] The interval which mice are to be polled at. 8255 8256 usbhid.jspoll= 8257 [USBHID] The interval which joysticks are to be polled at. 8258 8259 usbhid.kbpoll= 8260 [USBHID] The interval which keyboards are to be polled at. 8261 8262 usb-storage.delay_use= 8263 [UMS] The delay in seconds before a new device is 8264 scanned for Logical Units (default 1). 8265 Optionally the delay in milliseconds if the value has 8266 suffix with "ms". 8267 Example: delay_use=2567ms 8268 8269 usb-storage.quirks= 8270 [UMS] A list of quirks entries to supplement or 8271 override the built-in unusual_devs list. List 8272 entries are separated by commas. Each entry has 8273 the form VID:PID:Flags where VID and PID are Vendor 8274 and Product ID values (4-digit hex numbers) and 8275 Flags is a set of characters, each corresponding 8276 to a common usb-storage quirk flag as follows: 8277 a = SANE_SENSE (collect more than 18 bytes 8278 of sense data, not on uas); 8279 b = BAD_SENSE (don't collect more than 18 8280 bytes of sense data, not on uas); 8281 c = FIX_CAPACITY (decrease the reported 8282 device capacity by one sector); 8283 d = NO_READ_DISC_INFO (don't use 8284 READ_DISC_INFO command, not on uas); 8285 e = NO_READ_CAPACITY_16 (don't use 8286 READ_CAPACITY_16 command); 8287 f = NO_REPORT_OPCODES (don't use report opcodes 8288 command, uas only); 8289 g = MAX_SECTORS_240 (don't transfer more than 8290 240 sectors at a time, uas only); 8291 h = CAPACITY_HEURISTICS (decrease the 8292 reported device capacity by one 8293 sector if the number is odd); 8294 i = IGNORE_DEVICE (don't bind to this 8295 device); 8296 j = NO_REPORT_LUNS (don't use report luns 8297 command, uas only); 8298 k = NO_SAME (do not use WRITE_SAME, uas only) 8299 l = NOT_LOCKABLE (don't try to lock and 8300 unlock ejectable media, not on uas); 8301 m = MAX_SECTORS_64 (don't transfer more 8302 than 64 sectors = 32 KB at a time, 8303 not on uas); 8304 n = INITIAL_READ10 (force a retry of the 8305 initial READ(10) command, not on uas); 8306 o = CAPACITY_OK (accept the capacity 8307 reported by the device, not on uas); 8308 p = WRITE_CACHE (the device cache is ON 8309 by default, not on uas); 8310 r = IGNORE_RESIDUE (the device reports 8311 bogus residue values, not on uas); 8312 s = SINGLE_LUN (the device has only one 8313 Logical Unit); 8314 t = NO_ATA_1X (don't allow ATA(12) and ATA(16) 8315 commands, uas only); 8316 u = IGNORE_UAS (don't bind to the uas driver); 8317 w = NO_WP_DETECT (don't test whether the 8318 medium is write-protected). 8319 y = ALWAYS_SYNC (issue a SYNCHRONIZE_CACHE 8320 even if the device claims no cache, 8321 not on uas) 8322 Example: quirks=0419:aaf5:rl,0421:0433:rc 8323 8324 user_debug= [KNL,ARM] 8325 Format: <int> 8326 See arch/arm/Kconfig.debug help text. 8327 1 - undefined instruction events 8328 2 - system calls 8329 4 - invalid data aborts 8330 8 - SIGSEGV faults 8331 16 - SIGBUS faults 8332 Example: user_debug=31 8333 8334 vdso= [X86,SH,SPARC] 8335 On X86_32, this is an alias for vdso32=. Otherwise: 8336 8337 vdso=1: enable VDSO (the default) 8338 vdso=0: disable VDSO mapping 8339 8340 vdso32= [X86] Control the 32-bit vDSO 8341 vdso32=1: enable 32-bit VDSO 8342 vdso32=0 or vdso32=2: disable 32-bit VDSO 8343 8344 See the help text for CONFIG_COMPAT_VDSO for more 8345 details. If CONFIG_COMPAT_VDSO is set, the default is 8346 vdso32=0; otherwise, the default is vdso32=1. 8347 8348 For compatibility with older kernels, vdso32=2 is an 8349 alias for vdso32=0. 8350 8351 Try vdso32=0 if you encounter an error that says: 8352 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed! 8353 8354 video= [FB,EARLY] Frame buffer configuration 8355 See Documentation/fb/modedb.rst. 8356 8357 video.brightness_switch_enabled= [ACPI] 8358 Format: [0|1] 8359 If set to 1, on receiving an ACPI notify event 8360 generated by hotkey, video driver will adjust brightness 8361 level and then send out the event to user space through 8362 the allocated input device. If set to 0, video driver 8363 will only send out the event without touching backlight 8364 brightness level. 8365 default: 1 8366 8367 virtio_mmio.device= 8368 [VMMIO] Memory mapped virtio (platform) device. 8369 8370 <size>@<baseaddr>:<irq>[:<id>] 8371 where: 8372 <size> := size (can use standard suffixes 8373 like K, M and G) 8374 <baseaddr> := physical base address 8375 <irq> := interrupt number (as passed to 8376 request_irq()) 8377 <id> := (optional) platform device id 8378 example: 8379 virtio_mmio.device=1K@0x100b0000:48:7 8380 8381 Can be used multiple times for multiple devices. 8382 8383 vga= [BOOT,X86-32] Select a particular video mode 8384 See Documentation/arch/x86/boot.rst and 8385 Documentation/admin-guide/svga.rst. 8386 Use vga=ask for menu. 8387 This is actually a boot loader parameter; the value is 8388 passed to the kernel using a special protocol. 8389 8390 vm_debug[=options] [KNL] Available with CONFIG_DEBUG_VM=y. 8391 May slow down system boot speed, especially when 8392 enabled on systems with a large amount of memory. 8393 All options are enabled by default, and this 8394 interface is meant to allow for selectively 8395 enabling or disabling specific virtual memory 8396 debugging features. 8397 8398 Available options are: 8399 P Enable page structure init time poisoning 8400 - Disable all of the above options 8401 8402 vmalloc=nn[KMG] [KNL,BOOT,EARLY] Forces the vmalloc area to have an 8403 exact size of <nn>. This can be used to increase 8404 the minimum size (128MB on x86, arm32 platforms). 8405 It can also be used to decrease the size and leave more room 8406 for directly mapped kernel RAM. Note that this parameter does 8407 not exist on many other platforms (including arm64, alpha, 8408 loongarch, arc, csky, hexagon, microblaze, mips, nios2, openrisc, 8409 parisc, m64k, powerpc, riscv, sh, um, xtensa, s390, sparc). 8410 8411 vmcp_cma=nn[MG] [KNL,S390,EARLY] 8412 Sets the memory size reserved for contiguous memory 8413 allocations for the vmcp device driver. 8414 8415 vmhalt= [KNL,S390] Perform z/VM CP command after system halt. 8416 Format: <command> 8417 8418 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic. 8419 Format: <command> 8420 8421 vmpoff= [KNL,S390] Perform z/VM CP command after power off. 8422 Format: <command> 8423 8424 vmscape= [X86] Controls mitigation for VMscape attacks. 8425 VMscape attacks can leak information from a userspace 8426 hypervisor to a guest via speculative side-channels. 8427 8428 off - disable the mitigation 8429 ibpb - use Indirect Branch Prediction Barrier 8430 (IBPB) mitigation (default) 8431 force - force vulnerability detection even on 8432 unaffected processors 8433 8434 vsyscall= [X86-64,EARLY] 8435 Controls the behavior of vsyscalls (i.e. calls to 8436 fixed addresses of 0xffffffffff600x00 from legacy 8437 code). Most statically-linked binaries and older 8438 versions of glibc use these calls. Because these 8439 functions are at fixed addresses, they make nice 8440 targets for exploits that can control RIP. 8441 8442 emulate Vsyscalls turn into traps and are emulated 8443 reasonably safely. The vsyscall page is 8444 readable. This disables the Linear 8445 Address Space Separation (LASS) security 8446 feature and makes the system less secure. 8447 8448 xonly [default] Vsyscalls turn into traps and are 8449 emulated reasonably safely. The vsyscall 8450 page is not readable. 8451 8452 none Vsyscalls don't work at all. This makes 8453 them quite hard to use for exploits but 8454 might break your system. 8455 8456 vt.color= [VT] Default text color. 8457 Format: 0xYX, X = foreground, Y = background. 8458 Default: 0x07 = light gray on black. 8459 8460 vt.cur_default= [VT] Default cursor shape. 8461 Format: 0xCCBBAA, where AA, BB, and CC are the same as 8462 the parameters of the <Esc>[?A;B;Cc escape sequence; 8463 see vga-softcursor.rst. Default: 2 = underline. 8464 8465 vt.default_blu= [VT] 8466 Format: <blue0>,<blue1>,<blue2>,...,<blue15> 8467 Change the default blue palette of the console. 8468 This is a 16-member array composed of values 8469 ranging from 0-255. 8470 8471 vt.default_grn= [VT] 8472 Format: <green0>,<green1>,<green2>,...,<green15> 8473 Change the default green palette of the console. 8474 This is a 16-member array composed of values 8475 ranging from 0-255. 8476 8477 vt.default_red= [VT] 8478 Format: <red0>,<red1>,<red2>,...,<red15> 8479 Change the default red palette of the console. 8480 This is a 16-member array composed of values 8481 ranging from 0-255. 8482 8483 vt.default_utf8= 8484 [VT] 8485 Format=<0|1> 8486 Set system-wide default UTF-8 mode for all tty's. 8487 Default is 1, i.e. UTF-8 mode is enabled for all 8488 newly opened terminals. 8489 8490 vt.global_cursor_default= 8491 [VT] 8492 Format=<-1|0|1> 8493 Set system-wide default for whether a cursor 8494 is shown on new VTs. Default is -1, 8495 i.e. cursors will be created by default unless 8496 overridden by individual drivers. 0 will hide 8497 cursors, 1 will display them. 8498 8499 vt.italic= [VT] Default color for italic text; 0-15. 8500 Default: 2 = green. 8501 8502 vt.underline= [VT] Default color for underlined text; 0-15. 8503 Default: 3 = cyan. 8504 8505 watchdog timers [HW,WDT] For information on watchdog timers, 8506 see Documentation/watchdog/watchdog-parameters.rst 8507 or other driver-specific files in the 8508 Documentation/watchdog/ directory. 8509 8510 watchdog_thresh= 8511 [KNL] 8512 Set the hard lockup detector stall duration 8513 threshold in seconds. The soft lockup detector 8514 threshold is set to twice the value. A value of 0 8515 disables both lockup detectors. Default is 10 8516 seconds. 8517 8518 workqueue.unbound_cpus= 8519 [KNL,SMP] Specify to constrain one or some CPUs 8520 to use in unbound workqueues. 8521 Format: <cpu-list> 8522 By default, all online CPUs are available for 8523 unbound workqueues. 8524 8525 workqueue.watchdog_thresh= 8526 If CONFIG_WQ_WATCHDOG is configured, workqueue can 8527 warn stall conditions and dump internal state to 8528 help debugging. 0 disables workqueue stall 8529 detection; otherwise, it's the stall threshold 8530 duration in seconds. The default value is 30 and 8531 it can be updated at runtime by writing to the 8532 corresponding sysfs file. 8533 8534 workqueue.panic_on_stall=<uint> 8535 Panic when workqueue stall is detected by 8536 CONFIG_WQ_WATCHDOG. It sets the number times of the 8537 stall to trigger panic. 8538 8539 The default is set by CONFIG_BOOTPARAM_WQ_STALL_PANIC, 8540 which is 0 (disabled) if not configured. 8541 8542 workqueue.panic_on_stall_time=<uint> 8543 Panic when a workqueue stall has been continuous for 8544 the specified number of seconds. Unlike panic_on_stall 8545 which counts accumulated stall events, this triggers 8546 based on the duration of a single continuous stall. 8547 8548 The default is 0, which disables the time-based panic. 8549 8550 workqueue.cpu_intensive_thresh_us= 8551 Per-cpu work items which run for longer than this 8552 threshold are automatically considered CPU intensive 8553 and excluded from concurrency management to prevent 8554 them from noticeably delaying other per-cpu work 8555 items. Default is 10000 (10ms). 8556 8557 If CONFIG_WQ_CPU_INTENSIVE_REPORT is set, the kernel 8558 will report the work functions which violate this 8559 threshold repeatedly. They are likely good 8560 candidates for using WQ_UNBOUND workqueues instead. 8561 8562 workqueue.cpu_intensive_warning_thresh=<uint> 8563 If CONFIG_WQ_CPU_INTENSIVE_REPORT is set, the kernel 8564 will report the work functions which violate the 8565 intensive_threshold_us repeatedly. In order to prevent 8566 spurious warnings, start printing only after a work 8567 function has violated this threshold number of times. 8568 8569 The default is 4 times. 0 disables the warning. 8570 8571 workqueue.power_efficient 8572 Per-cpu workqueues are generally preferred because 8573 they show better performance thanks to cache 8574 locality; unfortunately, per-cpu workqueues tend to 8575 be more power hungry than unbound workqueues. 8576 8577 Enabling this makes the per-cpu workqueues which 8578 were observed to contribute significantly to power 8579 consumption unbound, leading to measurably lower 8580 power usage at the cost of small performance 8581 overhead. 8582 8583 The default value of this parameter is determined by 8584 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT. 8585 8586 workqueue.default_affinity_scope= 8587 Select the default affinity scope to use for unbound 8588 workqueues. Can be one of "cpu", "smt", "cache", 8589 "cache_shard", "numa" and "system". Default is 8590 "cache_shard". For more 8591 information, see the Affinity Scopes section in 8592 Documentation/core-api/workqueue.rst. 8593 8594 This can be changed after boot by writing to the 8595 matching /sys/module/workqueue/parameters file. All 8596 workqueues with the "default" affinity scope will be 8597 updated accordingly. 8598 8599 workqueue.debug_force_rr_cpu 8600 Workqueue used to implicitly guarantee that work 8601 items queued without explicit CPU specified are put 8602 on the local CPU. This guarantee is no longer true 8603 and while local CPU is still preferred work items 8604 may be put on foreign CPUs. This debug option 8605 forces round-robin CPU selection to flush out 8606 usages which depend on the now broken guarantee. 8607 When enabled, memory and cache locality will be 8608 impacted. 8609 8610 writecombine= [LOONGARCH,EARLY] Control the MAT (Memory Access 8611 Type) of ioremap_wc(). 8612 8613 on - Enable writecombine, use WUC for ioremap_wc() 8614 off - Disable writecombine, use SUC for ioremap_wc() 8615 8616 x2apic_phys [X86-64,APIC,EARLY] Use x2apic physical mode instead of 8617 default x2apic cluster mode on platforms 8618 supporting x2apic. 8619 8620 xen_512gb_limit [KNL,X86-64,XEN] 8621 Restricts the kernel running paravirtualized under Xen 8622 to use only up to 512 GB of RAM. The reason to do so is 8623 crash analysis tools and Xen tools for doing domain 8624 save/restore/migration must be enabled to handle larger 8625 domains. 8626 8627 xen_console_io [XEN,EARLY] 8628 Boolean option to enable/disable the usage of the Xen 8629 console_io hypercalls to read and write to the console. 8630 Mostly useful for debugging and development. 8631 8632 xen_emul_unplug= [HW,X86,XEN,EARLY] 8633 Unplug Xen emulated devices 8634 Format: [unplug0,][unplug1] 8635 ide-disks -- unplug primary master IDE devices 8636 aux-ide-disks -- unplug non-primary-master IDE devices 8637 nics -- unplug network devices 8638 all -- unplug all emulated devices (NICs and IDE disks) 8639 unnecessary -- unplugging emulated devices is 8640 unnecessary even if the host did not respond to 8641 the unplug protocol 8642 never -- do not unplug even if version check succeeds 8643 8644 xen_legacy_crash [X86,XEN,EARLY] 8645 Crash from Xen panic notifier, without executing late 8646 panic() code such as dumping handler. 8647 8648 xen_mc_debug [X86,XEN,EARLY] 8649 Enable multicall debugging when running as a Xen PV guest. 8650 Enabling this feature will reduce performance a little 8651 bit, so it should only be enabled for obtaining extended 8652 debug data in case of multicall errors. 8653 8654 xen_msr_safe= [X86,XEN,EARLY] 8655 Format: <bool> 8656 Select whether to always use non-faulting (safe) MSR 8657 access functions when running as Xen PV guest. The 8658 default value is controlled by CONFIG_XEN_PV_MSR_SAFE. 8659 8660 xen_nopv [X86] 8661 Disables the PV optimizations forcing the HVM guest to 8662 run as generic HVM guest with no PV drivers. 8663 This option is obsoleted by the "nopv" option, which 8664 has equivalent effect for XEN platform. 8665 8666 xen_no_vector_callback 8667 [KNL,X86,XEN,EARLY] Disable the vector callback for Xen 8668 event channel interrupts. 8669 8670 xen_scrub_pages= [XEN] 8671 Boolean option to control scrubbing pages before giving them back 8672 to Xen, for use by other domains. Can be also changed at runtime 8673 with /sys/devices/system/xen_memory/xen_memory0/scrub_pages. 8674 Default value controlled with CONFIG_XEN_SCRUB_PAGES_DEFAULT. 8675 8676 xen_timer_slop= [X86-64,XEN,EARLY] 8677 Set the timer slop (in nanoseconds) for the virtual Xen 8678 timers (default is 100000). This adjusts the minimum 8679 delta of virtualized Xen timers, where lower values 8680 improve timer resolution at the expense of processing 8681 more timer interrupts. 8682 8683 xen.balloon_boot_timeout= [XEN] 8684 The time (in seconds) to wait before giving up to boot 8685 in case initial ballooning fails to free enough memory. 8686 Applies only when running as HVM or PVH guest and 8687 started with less memory configured than allowed at 8688 max. Default is 180. 8689 8690 xen.event_eoi_delay= [XEN] 8691 How long to delay EOI handling in case of event 8692 storms (jiffies). Default is 10. 8693 8694 xen.event_loop_timeout= [XEN] 8695 After which time (jiffies) the event handling loop 8696 should start to delay EOI handling. Default is 2. 8697 8698 xen.fifo_events= [XEN] 8699 Boolean parameter to disable using fifo event handling 8700 even if available. Normally fifo event handling is 8701 preferred over the 2-level event handling, as it is 8702 fairer and the number of possible event channels is 8703 much higher. Default is on (use fifo events). 8704 8705 xirc2ps_cs= [NET,PCMCIA] 8706 Format: 8707 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]] 8708 8709 xive= [PPC] 8710 By default on POWER9 and above, the kernel will 8711 natively use the XIVE interrupt controller. This option 8712 allows the fallback firmware mode to be used: 8713 8714 off Fallback to firmware control of XIVE interrupt 8715 controller on both pseries and powernv 8716 platforms. Only useful on POWER9 and above. 8717 8718 xive.store-eoi=off [PPC] 8719 By default on POWER10 and above, the kernel will use 8720 stores for EOI handling when the XIVE interrupt mode 8721 is active. This option allows the XIVE driver to use 8722 loads instead, as on POWER9. 8723 8724 xhci-hcd.quirks [USB,KNL] 8725 A hex value specifying bitmask with supplemental xhci 8726 host controller quirks. Meaning of each bit can be 8727 consulted in header drivers/usb/host/xhci.h. 8728 8729 xmon [PPC,EARLY] 8730 Format: { early | on | rw | ro | off } 8731 Controls if xmon debugger is enabled. Default is off. 8732 Passing only "xmon" is equivalent to "xmon=early". 8733 early Call xmon as early as possible on boot; xmon 8734 debugger is called from setup_arch(). 8735 on xmon debugger hooks will be installed so xmon 8736 is only called on a kernel crash. Default mode, 8737 i.e. either "ro" or "rw" mode, is controlled 8738 with CONFIG_XMON_DEFAULT_RO_MODE. 8739 rw xmon debugger hooks will be installed so xmon 8740 is called only on a kernel crash, mode is write, 8741 meaning SPR registers, memory and, other data 8742 can be written using xmon commands. 8743 ro same as "rw" option above but SPR registers, 8744 memory, and other data can't be written using 8745 xmon commands. 8746 off xmon is disabled. 8747