1 ACPI ACPI support is enabled. 2 AGP AGP (Accelerated Graphics Port) is enabled. 3 ALSA ALSA sound support is enabled. 4 APIC APIC support is enabled. 5 APM Advanced Power Management support is enabled. 6 APPARMOR AppArmor support is enabled. 7 ARM ARM architecture is enabled. 8 ARM64 ARM64 architecture is enabled. 9 CLK Common clock infrastructure is enabled. 10 CMA Contiguous Memory Area support is enabled. 11 DRM Direct Rendering Management support is enabled. 12 DYNAMIC_DEBUG Build in debug messages and enable them at runtime 13 EARLY Parameter processed too early to be embedded in initrd. 14 EDD BIOS Enhanced Disk Drive Services (EDD) is enabled 15 EFI EFI Partitioning (GPT) is enabled 16 EVM Extended Verification Module 17 FB The frame buffer device is enabled. 18 FTRACE Function tracing enabled. 19 GCOV GCOV profiling is enabled. 20 HIBERNATION HIBERNATION is enabled. 21 HW Appropriate hardware is enabled. 22 HYPER_V HYPERV support is enabled. 23 IMA Integrity measurement architecture is enabled. 24 IP_PNP IP DHCP, BOOTP, or RARP is enabled. 25 IPV6 IPv6 support is enabled. 26 ISAPNP ISA PnP code is enabled. 27 ISOL CPU Isolation is enabled. 28 JOY Appropriate joystick support is enabled. 29 KGDB Kernel debugger support is enabled. 30 KVM Kernel Virtual Machine support is enabled. 31 LIBATA Libata driver is enabled 32 LOONGARCH LoongArch architecture is enabled. 33 LOOP Loopback device support is enabled. 34 LP Printer support is enabled. 35 M68k M68k architecture is enabled. 36 These options have more detailed description inside of 37 Documentation/arch/m68k/kernel-options.rst. 38 MIPS MIPS architecture is enabled. 39 MOUSE Appropriate mouse support is enabled. 40 MSI Message Signaled Interrupts (PCI). 41 MTD MTD (Memory Technology Device) support is enabled. 42 NET Appropriate network support is enabled. 43 NFS Appropriate NFS support is enabled. 44 NUMA NUMA support is enabled. 45 OF Devicetree is enabled. 46 PARISC The PA-RISC architecture is enabled. 47 PCI PCI bus support is enabled. 48 PCIE PCI Express support is enabled. 49 PCMCIA The PCMCIA subsystem is enabled. 50 PNP Plug & Play support is enabled. 51 PPC PowerPC architecture is enabled. 52 PPT Parallel port support is enabled. 53 PS2 Appropriate PS/2 support is enabled. 54 PV_OPS A paravirtualized kernel is enabled. 55 RAM RAM disk support is enabled. 56 RDT Intel Resource Director Technology. 57 RISCV RISCV architecture is enabled. 58 S390 S390 architecture is enabled. 59 SCSI Appropriate SCSI support is enabled. 60 A lot of drivers have their options described inside 61 the Documentation/scsi/ sub-directory. 62 SDW SoundWire support is enabled. 63 SECURITY Different security models are enabled. 64 SELINUX SELinux support is enabled. 65 SERIAL Serial support is enabled. 66 SH SuperH architecture is enabled. 67 SMP The kernel is an SMP kernel. 68 SPARC Sparc architecture is enabled. 69 SUSPEND System suspend states are enabled. 70 SWSUSP Software suspend (hibernation) is enabled. 71 TPM TPM drivers are enabled. 72 UMS USB Mass Storage support is enabled. 73 USB USB support is enabled. 74 NVME NVMe support is enabled 75 USBHID USB Human Interface Device support is enabled. 76 V4L Video For Linux support is enabled. 77 VGA The VGA console has been enabled. 78 VMMIO Driver for memory mapped virtio devices is enabled. 79 VT Virtual terminal support is enabled. 80 WDT Watchdog support is enabled. 81 X86-32 X86-32, aka i386 architecture is enabled. 82 X86-64 X86-64 architecture is enabled. 83 X86 Either 32-bit or 64-bit x86 (same as X86-32+X86-64) 84 X86_UV SGI UV support is enabled. 85 XEN Xen support is enabled 86 XTENSA xtensa architecture is enabled. 87 88In addition, the following text indicates that the option 89 90 BOOT Is a boot loader parameter. 91 BUGS= Relates to possible processor bugs on the said processor. 92 KNL Is a kernel start-up parameter. 93 94 95Kernel parameters 96 97 accept_memory= [MM] 98 Format: { eager | lazy } 99 default: lazy 100 By default, unaccepted memory is accepted lazily to 101 avoid prolonged boot times. The lazy option will add 102 some runtime overhead until all memory is eventually 103 accepted. In most cases the overhead is negligible. 104 For some workloads or for debugging purposes 105 accept_memory=eager can be used to accept all memory 106 at once during boot. 107 108 acpi= [HW,ACPI,X86,ARM64,RISCV64,EARLY] 109 Advanced Configuration and Power Interface 110 Format: { force | on | off | strict | noirq | rsdt | 111 copy_dsdt | nospcr } 112 force -- enable ACPI if default was off 113 on -- enable ACPI but allow fallback to DT [arm64,riscv64] 114 off -- disable ACPI if default was on 115 noirq -- do not use ACPI for IRQ routing 116 strict -- Be less tolerant of platforms that are not 117 strictly ACPI specification compliant. 118 rsdt -- prefer RSDT over (default) XSDT 119 copy_dsdt -- copy DSDT to memory 120 nocmcff -- Disable firmware first mode for corrected 121 errors. This disables parsing the HEST CMC error 122 source to check if firmware has set the FF flag. This 123 may result in duplicate corrected error reports. 124 nospcr -- disable console in ACPI SPCR table as 125 default _serial_ console on ARM64 126 spcr -- enable console in ACPI SPCR table as 127 default _serial_ console on x86 128 For ARM64, ONLY "acpi=off", "acpi=on", "acpi=force" or 129 "acpi=nospcr" are available 130 For RISCV64, ONLY "acpi=off", "acpi=on" or "acpi=force" 131 are available 132 133 See also Documentation/power/runtime_pm.rst, pci=noacpi 134 135 acpi_apic_instance= [ACPI,IOAPIC,EARLY] 136 Format: <int> 137 2: use 2nd APIC table, if available 138 1,0: use 1st APIC table 139 default: 0 140 141 acpi_backlight= [HW,ACPI] 142 { vendor | video | native | none } 143 If set to vendor, prefer vendor-specific driver 144 (e.g. thinkpad_acpi, sony_acpi, etc.) instead 145 of the ACPI video.ko driver. 146 If set to video, use the ACPI video.ko driver. 147 If set to native, use the device's native backlight mode. 148 If set to none, disable the ACPI backlight interface. 149 150 acpi_force_32bit_fadt_addr [ACPI,EARLY] 151 force FADT to use 32 bit addresses rather than the 152 64 bit X_* addresses. Some firmware have broken 64 153 bit addresses for force ACPI ignore these and use 154 the older legacy 32 bit addresses. 155 156 acpica_no_return_repair [HW, ACPI] 157 Disable AML predefined validation mechanism 158 This mechanism can repair the evaluation result to make 159 the return objects more ACPI specification compliant. 160 This option is useful for developers to identify the 161 root cause of an AML interpreter issue when the issue 162 has something to do with the repair mechanism. 163 164 acpi.debug_layer= [HW,ACPI,ACPI_DEBUG] 165 acpi.debug_level= [HW,ACPI,ACPI_DEBUG] 166 Format: <int> 167 CONFIG_ACPI_DEBUG must be enabled to produce any ACPI 168 debug output. Bits in debug_layer correspond to a 169 _COMPONENT in an ACPI source file, e.g., 170 #define _COMPONENT ACPI_EVENTS 171 Bits in debug_level correspond to a level in 172 ACPI_DEBUG_PRINT statements, e.g., 173 ACPI_DEBUG_PRINT((ACPI_DB_INFO, ... 174 The debug_level mask defaults to "info". See 175 Documentation/firmware-guide/acpi/debug.rst for more information about 176 debug layers and levels. 177 178 Enable processor driver info messages: 179 acpi.debug_layer=0x20000000 180 Enable AML "Debug" output, i.e., stores to the Debug 181 object while interpreting AML: 182 acpi.debug_layer=0xffffffff acpi.debug_level=0x2 183 Enable all messages related to ACPI hardware: 184 acpi.debug_layer=0x2 acpi.debug_level=0xffffffff 185 186 Some values produce so much output that the system is 187 unusable. The "log_buf_len" parameter may be useful 188 if you need to capture more output. 189 190 acpi.poweroff_on_fatal= [ACPI] 191 {0 | 1} 192 Causes the system to poweroff when the ACPI bytecode signals 193 a fatal error. The default value of this setting is 1. 194 Overriding this value should only be done for diagnosing 195 ACPI firmware problems, as the system might behave erratically 196 after having encountered a fatal ACPI error. 197 198 acpi_enforce_resources= [ACPI] 199 { strict | lax | no } 200 Check for resource conflicts between native drivers 201 and ACPI OperationRegions (SystemIO and SystemMemory 202 only). IO ports and memory declared in ACPI might be 203 used by the ACPI subsystem in arbitrary AML code and 204 can interfere with legacy drivers. 205 strict (default): access to resources claimed by ACPI 206 is denied; legacy drivers trying to access reserved 207 resources will fail to bind to device using them. 208 lax: access to resources claimed by ACPI is allowed; 209 legacy drivers trying to access reserved resources 210 will bind successfully but a warning message is logged. 211 no: ACPI OperationRegions are not marked as reserved, 212 no further checks are performed. 213 214 acpi_force_table_verification [HW,ACPI,EARLY] 215 Enable table checksum verification during early stage. 216 By default, this is disabled due to x86 early mapping 217 size limitation. 218 219 acpi_irq_balance [HW,ACPI] 220 ACPI will balance active IRQs 221 default in APIC mode 222 223 acpi_irq_nobalance [HW,ACPI] 224 ACPI will not move active IRQs (default) 225 default in PIC mode 226 227 acpi_irq_isa= [HW,ACPI] If irq_balance, mark listed IRQs used by ISA 228 Format: <irq>,<irq>... 229 230 acpi_irq_pci= [HW,ACPI] If irq_balance, clear listed IRQs for 231 use by PCI 232 Format: <irq>,<irq>... 233 234 acpi_mask_gpe= [HW,ACPI] 235 Due to the existence of _Lxx/_Exx, some GPEs triggered 236 by unsupported hardware/firmware features can result in 237 GPE floodings that cannot be automatically disabled by 238 the GPE dispatcher. 239 This facility can be used to prevent such uncontrolled 240 GPE floodings. 241 Format: <byte> or <bitmap-list> 242 243 acpi_no_auto_serialize [HW,ACPI] 244 Disable auto-serialization of AML methods 245 AML control methods that contain the opcodes to create 246 named objects will be marked as "Serialized" by the 247 auto-serialization feature. 248 This feature is enabled by default. 249 This option allows to turn off the feature. 250 251 acpi_no_memhotplug [ACPI] Disable memory hotplug. Useful for kdump 252 kernels. 253 254 acpi_no_static_ssdt [HW,ACPI,EARLY] 255 Disable installation of static SSDTs at early boot time 256 By default, SSDTs contained in the RSDT/XSDT will be 257 installed automatically and they will appear under 258 /sys/firmware/acpi/tables. 259 This option turns off this feature. 260 Note that specifying this option does not affect 261 dynamic table installation which will install SSDT 262 tables to /sys/firmware/acpi/tables/dynamic. 263 264 acpi_no_watchdog [HW,ACPI,WDT] 265 Ignore the ACPI-based watchdog interface (WDAT) and let 266 a native driver control the watchdog device instead. 267 268 acpi_rsdp= [ACPI,EFI,KEXEC,EARLY] 269 Pass the RSDP address to the kernel, mostly used 270 on machines running EFI runtime service to boot the 271 second kernel for kdump. 272 273 acpi_os_name= [HW,ACPI] Tell ACPI BIOS the name of the OS 274 Format: To spoof as Windows 98: ="Microsoft Windows" 275 276 acpi_rev_override [ACPI] Override the _REV object to return 5 (instead 277 of 2 which is mandated by ACPI 6) as the supported ACPI 278 specification revision (when using this switch, it may 279 be necessary to carry out a cold reboot _twice_ in a 280 row to make it take effect on the platform firmware). 281 282 acpi_osi= [HW,ACPI] Modify list of supported OS interface strings 283 acpi_osi="string1" # add string1 284 acpi_osi="!string2" # remove string2 285 acpi_osi=!* # remove all strings 286 acpi_osi=! # disable all built-in OS vendor 287 strings 288 acpi_osi=!! # enable all built-in OS vendor 289 strings 290 acpi_osi= # disable all strings 291 292 'acpi_osi=!' can be used in combination with single or 293 multiple 'acpi_osi="string1"' to support specific OS 294 vendor string(s). Note that such command can only 295 affect the default state of the OS vendor strings, thus 296 it cannot affect the default state of the feature group 297 strings and the current state of the OS vendor strings, 298 specifying it multiple times through kernel command line 299 is meaningless. This command is useful when one do not 300 care about the state of the feature group strings which 301 should be controlled by the OSPM. 302 Examples: 303 1. 'acpi_osi=! acpi_osi="Windows 2000"' is equivalent 304 to 'acpi_osi="Windows 2000" acpi_osi=!', they all 305 can make '_OSI("Windows 2000")' TRUE. 306 307 'acpi_osi=' cannot be used in combination with other 308 'acpi_osi=' command lines, the _OSI method will not 309 exist in the ACPI namespace. NOTE that such command can 310 only affect the _OSI support state, thus specifying it 311 multiple times through kernel command line is also 312 meaningless. 313 Examples: 314 1. 'acpi_osi=' can make 'CondRefOf(_OSI, Local1)' 315 FALSE. 316 317 'acpi_osi=!*' can be used in combination with single or 318 multiple 'acpi_osi="string1"' to support specific 319 string(s). Note that such command can affect the 320 current state of both the OS vendor strings and the 321 feature group strings, thus specifying it multiple times 322 through kernel command line is meaningful. But it may 323 still not able to affect the final state of a string if 324 there are quirks related to this string. This command 325 is useful when one want to control the state of the 326 feature group strings to debug BIOS issues related to 327 the OSPM features. 328 Examples: 329 1. 'acpi_osi="Module Device" acpi_osi=!*' can make 330 '_OSI("Module Device")' FALSE. 331 2. 'acpi_osi=!* acpi_osi="Module Device"' can make 332 '_OSI("Module Device")' TRUE. 333 3. 'acpi_osi=! acpi_osi=!* acpi_osi="Windows 2000"' is 334 equivalent to 335 'acpi_osi=!* acpi_osi=! acpi_osi="Windows 2000"' 336 and 337 'acpi_osi=!* acpi_osi="Windows 2000" acpi_osi=!', 338 they all will make '_OSI("Windows 2000")' TRUE. 339 340 acpi_pm_good [X86] 341 Override the pmtimer bug detection: force the kernel 342 to assume that this machine's pmtimer latches its value 343 and always returns good values. 344 345 acpi_sci= [HW,ACPI,EARLY] ACPI System Control Interrupt trigger mode 346 Format: { level | edge | high | low } 347 348 acpi_skip_timer_override [HW,ACPI,EARLY] 349 Recognize and ignore IRQ0/pin2 Interrupt Override. 350 For broken nForce2 BIOS resulting in XT-PIC timer. 351 352 acpi_sleep= [HW,ACPI] Sleep options 353 Format: { s3_bios, s3_mode, s3_beep, s4_hwsig, 354 s4_nohwsig, old_ordering, nonvs, 355 sci_force_enable, nobl } 356 See Documentation/power/video.rst for information on 357 s3_bios and s3_mode. 358 s3_beep is for debugging; it makes the PC's speaker beep 359 as soon as the kernel's real-mode entry point is called. 360 s4_hwsig causes the kernel to check the ACPI hardware 361 signature during resume from hibernation, and gracefully 362 refuse to resume if it has changed. This complies with 363 the ACPI specification but not with reality, since 364 Windows does not do this and many laptops do change it 365 on docking. So the default behaviour is to allow resume 366 and simply warn when the signature changes, unless the 367 s4_hwsig option is enabled. 368 s4_nohwsig prevents ACPI hardware signature from being 369 used (or even warned about) during resume. 370 old_ordering causes the ACPI 1.0 ordering of the _PTS 371 control method, with respect to putting devices into 372 low power states, to be enforced (the ACPI 2.0 ordering 373 of _PTS is used by default). 374 nonvs prevents the kernel from saving/restoring the 375 ACPI NVS memory during suspend/hibernation and resume. 376 sci_force_enable causes the kernel to set SCI_EN directly 377 on resume from S1/S3 (which is against the ACPI spec, 378 but some broken systems don't work without it). 379 nobl causes the internal blacklist of systems known to 380 behave incorrectly in some ways with respect to system 381 suspend and resume to be ignored (use wisely). 382 383 acpi_use_timer_override [HW,ACPI,EARLY] 384 Use timer override. For some broken Nvidia NF5 boards 385 that require a timer override, but don't have HPET 386 387 add_efi_memmap [EFI,X86,EARLY] Include EFI memory map in 388 kernel's map of available physical RAM. 389 390 agp= [AGP] 391 { off | try_unsupported } 392 off: disable AGP support 393 try_unsupported: try to drive unsupported chipsets 394 (may crash computer or cause data corruption) 395 396 ALSA [HW,ALSA] 397 See Documentation/sound/alsa-configuration.rst 398 399 alignment= [KNL,ARM] 400 Allow the default userspace alignment fault handler 401 behaviour to be specified. Bit 0 enables warnings, 402 bit 1 enables fixups, and bit 2 sends a segfault. 403 404 align_va_addr= [X86-64] 405 Align virtual addresses by clearing slice [14:12] when 406 allocating a VMA at process creation time. This option 407 gives you up to 3% performance improvement on AMD F15h 408 machines (where it is enabled by default) for a 409 CPU-intensive style benchmark, and it can vary highly in 410 a microbenchmark depending on workload and compiler. 411 412 32: only for 32-bit processes 413 64: only for 64-bit processes 414 on: enable for both 32- and 64-bit processes 415 off: disable for both 32- and 64-bit processes 416 417 alloc_snapshot [FTRACE] 418 Allocate the ftrace snapshot buffer on boot up when the 419 main buffer is allocated. This is handy if debugging 420 and you need to use tracing_snapshot() on boot up, and 421 do not want to use tracing_snapshot_alloc() as it needs 422 to be done where GFP_KERNEL allocations are allowed. 423 424 allow_mismatched_32bit_el0 [ARM64,EARLY] 425 Allow execve() of 32-bit applications and setting of the 426 PER_LINUX32 personality on systems where only a strict 427 subset of the CPUs support 32-bit EL0. When this 428 parameter is present, the set of CPUs supporting 32-bit 429 EL0 is indicated by /sys/devices/system/cpu/aarch32_el0 430 and hot-unplug operations may be restricted. 431 432 See Documentation/arch/arm64/asymmetric-32bit.rst for more 433 information. 434 435 amd_iommu= [HW,X86-64] 436 Pass parameters to the AMD IOMMU driver in the system. 437 Possible values are: 438 fullflush - Deprecated, equivalent to iommu.strict=1 439 off - do not initialize any AMD IOMMU found in 440 the system 441 force_isolation - Force device isolation for all 442 devices. The IOMMU driver is not 443 allowed anymore to lift isolation 444 requirements as needed. This option 445 does not override iommu=pt 446 force_enable - Force enable the IOMMU on platforms known 447 to be buggy with IOMMU enabled. Use this 448 option with care. 449 pgtbl_v1 - Use v1 page table for DMA-API (Default). 450 pgtbl_v2 - Use v2 page table for DMA-API. 451 irtcachedis - Disable Interrupt Remapping Table (IRT) caching. 452 nohugepages - Limit page-sizes used for v1 page-tables 453 to 4 KiB. 454 v2_pgsizes_only - Limit page-sizes used for v1 page-tables 455 to 4KiB/2Mib/1GiB. 456 457 458 amd_iommu_dump= [HW,X86-64] 459 Enable AMD IOMMU driver option to dump the ACPI table 460 for AMD IOMMU. With this option enabled, AMD IOMMU 461 driver will print ACPI tables for AMD IOMMU during 462 IOMMU initialization. 463 464 amd_iommu_intr= [HW,X86-64] 465 Specifies one of the following AMD IOMMU interrupt 466 remapping modes: 467 legacy - Use legacy interrupt remapping mode. 468 vapic - Use virtual APIC mode, which allows IOMMU 469 to inject interrupts directly into guest. 470 This mode requires kvm-amd.avic=1. 471 (Default when IOMMU HW support is present.) 472 473 amd_pstate= [X86,EARLY] 474 disable 475 Do not enable amd_pstate as the default 476 scaling driver for the supported processors 477 passive 478 Use amd_pstate with passive mode as a scaling driver. 479 In this mode autonomous selection is disabled. 480 Driver requests a desired performance level and platform 481 tries to match the same performance level if it is 482 satisfied by guaranteed performance level. 483 active 484 Use amd_pstate_epp driver instance as the scaling driver, 485 driver provides a hint to the hardware if software wants 486 to bias toward performance (0x0) or energy efficiency (0xff) 487 to the CPPC firmware. then CPPC power algorithm will 488 calculate the runtime workload and adjust the realtime cores 489 frequency. 490 guided 491 Activate guided autonomous mode. Driver requests minimum and 492 maximum performance level and the platform autonomously 493 selects a performance level in this range and appropriate 494 to the current workload. 495 496 amd_prefcore= 497 [X86] 498 disable 499 Disable amd-pstate preferred core. 500 501 amd_dynamic_epp= 502 [X86] 503 disable 504 Disable amd-pstate dynamic EPP. 505 enable 506 Enable amd-pstate dynamic EPP. 507 508 amijoy.map= [HW,JOY] Amiga joystick support 509 Map of devices attached to JOY0DAT and JOY1DAT 510 Format: <a>,<b> 511 See also Documentation/input/joydev/joystick.rst 512 513 analog.map= [HW,JOY] Analog joystick and gamepad support 514 Specifies type or capabilities of an analog joystick 515 connected to one of 16 gameports 516 Format: <type1>,<type2>,..<type16> 517 518 apc= [HW,SPARC] 519 Power management functions (SPARCstation-4/5 + deriv.) 520 Format: noidle 521 Disable APC CPU standby support. SPARCstation-Fox does 522 not play well with APC CPU idle - disable it if you have 523 APC and your system crashes randomly. 524 525 apic [APIC,X86-64] Use IO-APIC. Default. 526 527 apic= [APIC,X86,EARLY] Advanced Programmable Interrupt Controller 528 Change the output verbosity while booting 529 Format: { quiet (default) | verbose | debug } 530 Change the amount of debugging information output 531 when initialising the APIC and IO-APIC components. 532 533 apic_extnmi= [APIC,X86,EARLY] External NMI delivery setting 534 Format: { bsp (default) | all | none } 535 bsp: External NMI is delivered only to CPU 0 536 all: External NMIs are broadcast to all CPUs as a 537 backup of CPU 0 538 none: External NMI is masked for all CPUs. This is 539 useful so that a dump capture kernel won't be 540 shot down by NMI 541 542 apicpmtimer Do APIC timer calibration using the pmtimer. Implies 543 apicmaintimer. Useful when your PIT timer is totally 544 broken. 545 546 autoconf= [IPV6] 547 See Documentation/networking/ipv6.rst. 548 549 apm= [APM] Advanced Power Management 550 See header of arch/x86/kernel/apm_32.c. 551 552 apparmor= [APPARMOR] Disable or enable AppArmor at boot time 553 Format: { "0" | "1" } 554 See security/apparmor/Kconfig help text 555 0 -- disable. 556 1 -- enable. 557 Default value is set via kernel config option. 558 559 arcrimi= [HW,NET] ARCnet - "RIM I" (entirely mem-mapped) cards 560 Format: <io>,<irq>,<nodeID> 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.nomops [ARM64] Unconditionally disable Memory Copy and Memory 572 Set instructions support 573 574 arm64.nompam [ARM64] Unconditionally disable Memory Partitioning And 575 Monitoring support 576 577 arm64.nomte [ARM64] Unconditionally disable Memory Tagging Extension 578 support 579 580 arm64.nopauth [ARM64] Unconditionally disable Pointer Authentication 581 support 582 583 arm64.nosme [ARM64] Unconditionally disable Scalable Matrix 584 Extension support 585 586 arm64.nosve [ARM64] Unconditionally disable Scalable Vector 587 Extension support 588 589 ataflop= [HW,M68k] 590 591 atarimouse= [HW,MOUSE] Atari Mouse 592 593 atkbd.extra= [HW] Enable extra LEDs and keys on IBM RapidAccess, 594 EzKey and similar keyboards 595 596 atkbd.reset= [HW] Reset keyboard during initialization 597 598 atkbd.set= [HW] Select keyboard code set 599 Format: <int> (2 = AT (default), 3 = PS/2) 600 601 atkbd.scroll= [HW] Enable scroll wheel on MS Office and similar 602 keyboards 603 604 atkbd.softraw= [HW] Choose between synthetic and real raw mode 605 Format: <bool> (0 = real, 1 = synthetic (default)) 606 607 atkbd.softrepeat= [HW] 608 Use software keyboard repeat 609 610 audit= [KNL] Enable the audit sub-system 611 Format: { "0" | "1" | "off" | "on" } 612 0 | off - kernel audit is disabled and can not be 613 enabled until the next reboot 614 unset - kernel audit is initialized but disabled and 615 will be fully enabled by the userspace auditd. 616 1 | on - kernel audit is initialized and partially 617 enabled, storing at most audit_backlog_limit 618 messages in RAM until it is fully enabled by the 619 userspace auditd. 620 Default: unset 621 622 audit_backlog_limit= [KNL] Set the audit queue size limit. 623 Format: <int> (must be >=0) 624 Default: 64 625 626 bau= [X86_UV] Enable the BAU on SGI UV. The default 627 behavior is to disable the BAU (i.e. bau=0). 628 Format: { "0" | "1" } 629 0 - Disable the BAU. 630 1 - Enable the BAU. 631 unset - Disable the BAU. 632 633 bdev_allow_write_mounted= 634 Format: <bool> 635 Control the ability to open a mounted block device 636 for writing, i.e., allow / disallow writes that bypass 637 the FS. This was implemented as a means to prevent 638 fuzzers from crashing the kernel by overwriting the 639 metadata underneath a mounted FS without its awareness. 640 This also prevents destructive formatting of mounted 641 filesystems by naive storage tooling that don't use 642 O_EXCL. Default is Y and can be changed through the 643 Kconfig option CONFIG_BLK_DEV_WRITE_MOUNTED. 644 645 bert_disable [ACPI] 646 Disable BERT OS support on buggy BIOSes. 647 648 bgrt_disable [ACPI,X86,EARLY] 649 Disable BGRT to avoid flickering OEM logo. 650 651 blkdevparts= Manual partition parsing of block device(s) for 652 embedded devices based on command line input. 653 See Documentation/block/cmdline-partition.rst 654 655 boot_delay= [KNL,EARLY] 656 Milliseconds to delay each printk during boot. 657 Only works if CONFIG_BOOT_PRINTK_DELAY is enabled, 658 and you may also have to specify "lpj=". Boot_delay 659 values larger than 10 seconds (10000) are assumed 660 erroneous and ignored. 661 Format: integer 662 663 bootconfig [KNL,EARLY] 664 Extended command line options can be added to an initrd 665 and this will cause the kernel to look for it. 666 667 See Documentation/admin-guide/bootconfig.rst 668 669 bttv.card= [HW,V4L] bttv (bt848 + bt878 based grabber cards) 670 bttv.radio= Most important insmod options are available as 671 kernel args too. 672 bttv.pll= See Documentation/admin-guide/media/bttv.rst 673 bttv.tuner= 674 675 bulk_remove=off [PPC] This parameter disables the use of the pSeries 676 firmware feature for flushing multiple hpte entries 677 at a time. 678 679 c101= [NET] Moxa C101 synchronous serial card 680 681 cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection. 682 Sometimes CPU hardware bugs make them report the cache 683 size incorrectly. The kernel will attempt work arounds 684 to fix known problems, but for some CPUs it is not 685 possible to determine what the correct size should be. 686 This option provides an override for these situations. 687 688 carrier_timeout= 689 [NET] Specifies amount of time (in seconds) that 690 the kernel should wait for a network carrier. By default 691 it waits 120 seconds. 692 693 ca_keys= [KEYS] This parameter identifies a specific key(s) on 694 the system trusted keyring to be used for certificate 695 trust validation. 696 format: { id:<keyid> | builtin } 697 698 cca= [MIPS,EARLY] Override the kernel pages' cache coherency 699 algorithm. Accepted values range from 0 to 7 700 inclusive. See arch/mips/include/asm/pgtable-bits.h 701 for platform specific values (SB1, Loongson3 and 702 others). 703 704 ccw_timeout_log [S390] 705 See Documentation/arch/s390/common_io.rst for details. 706 707 cfi= [X86-64] Set Control Flow Integrity checking features 708 when CONFIG_FINEIBT is enabled. 709 Format: feature[,feature...] 710 Default: auto 711 712 auto: Use FineIBT if IBT available, otherwise kCFI. 713 Under FineIBT, enable "paranoid" mode when 714 FRED is not available. 715 off: Turn off CFI checking. 716 kcfi: Use kCFI (disable FineIBT). 717 fineibt: Use FineIBT (even if IBT not available). 718 norand: Do not re-randomize CFI hashes. 719 paranoid: Add caller hash checking under FineIBT. 720 bhi: Enable register poisoning to stop speculation 721 across FineIBT. (Disabled by default.) 722 warn: Do not enforce CFI checking: warn only. 723 debug: Report CFI initialization details. 724 725 cgroup_disable= [KNL] Disable a particular controller or optional feature 726 Format: {name of the controller(s) or feature(s) to disable} 727 The effects of cgroup_disable=foo are: 728 - foo isn't auto-mounted if you mount all cgroups in 729 a single hierarchy 730 - foo isn't visible as an individually mountable 731 subsystem 732 - if foo is an optional feature then the feature is 733 disabled and corresponding cgroup files are not 734 created 735 {Currently only "memory" controller deal with this and 736 cut the overhead, others just disable the usage. So 737 only cgroup_disable=memory is actually worthy} 738 Specifying "pressure" disables per-cgroup pressure 739 stall information accounting feature 740 741 cgroup_no_v1= [KNL] Disable cgroup controllers and named hierarchies in v1 742 Format: { { controller | "all" | "named" } 743 [,{ controller | "all" | "named" }...] } 744 Like cgroup_disable, but only applies to cgroup v1; 745 the blacklisted controllers remain available in cgroup2. 746 "all" blacklists all controllers and "named" disables 747 named mounts. Specifying both "all" and "named" disables 748 all v1 hierarchies. 749 750 cgroup_v1_proc= [KNL] Show also missing controllers in /proc/cgroups 751 Format: { "true" | "false" } 752 /proc/cgroups lists only v1 controllers by default. 753 This compatibility option enables listing also v2 754 controllers (whose v1 code is not compiled!), so that 755 semi-legacy software can check this file to decide 756 about usage of v2 (sic) controllers. 757 758 cgroup_favordynmods= [KNL] Enable or Disable favordynmods. 759 Format: { "true" | "false" } 760 Defaults to the value of CONFIG_CGROUP_FAVOR_DYNMODS. 761 762 cgroup.memory= [KNL] Pass options to the cgroup memory controller. 763 Format: <string> 764 nosocket -- Disable socket memory accounting. 765 nokmem -- Disable kernel memory accounting. 766 nobpf -- Disable BPF memory accounting. 767 768 check_pages= [MM,EARLY] Enable sanity checking of pages after 769 allocations / before freeing. This adds checks to catch 770 double-frees, use-after-frees, and other sources of 771 page corruption by inspecting page internals (flags, 772 mapcount/refcount, memcg_data, etc.). 773 Format: { "0" | "1" } 774 Default: 0 (1 if CONFIG_DEBUG_VM is set) 775 776 checkreqprot= [SELINUX] Set initial checkreqprot flag value. 777 Format: { "0" | "1" } 778 See security/selinux/Kconfig help text. 779 0 -- check protection applied by kernel (includes 780 any implied execute protection). 781 1 -- check protection requested by application. 782 Default value is set via a kernel config option. 783 Value can be changed at runtime via 784 /sys/fs/selinux/checkreqprot. 785 Setting checkreqprot to 1 is deprecated. 786 787 cio_ignore= [S390] 788 See Documentation/arch/s390/common_io.rst for details. 789 790 clk_ignore_unused 791 [CLK] 792 Prevents the clock framework from automatically gating 793 clocks that have not been explicitly enabled by a Linux 794 device driver but are enabled in hardware at reset or 795 by the bootloader/firmware. Note that this does not 796 force such clocks to be always-on nor does it reserve 797 those clocks in any way. This parameter is useful for 798 debug and development, but should not be needed on a 799 platform with proper driver support. For more 800 information, see Documentation/driver-api/clk.rst. 801 802 clock= [BUGS=X86-32, HW] gettimeofday clocksource override. 803 [Deprecated] 804 Forces specified clocksource (if available) to be used 805 when calculating gettimeofday(). If specified 806 clocksource is not available, it defaults to PIT. 807 Format: { pit | tsc | cyclone | pmtmr } 808 809 clocksource= Override the default clocksource 810 Format: <string> 811 Override the default clocksource and use the clocksource 812 with the name specified. 813 Some clocksource names to choose from, depending on 814 the platform: 815 [all] jiffies (this is the base, fallback clocksource) 816 [ACPI] acpi_pm 817 [ARM] imx_timer1,OSTS,netx_timer,mpu_timer2, 818 pxa_timer,timer3,32k_counter,timer0_1 819 [X86-32] pit,hpet,tsc; 820 scx200_hrt on Geode; cyclone on IBM x440 821 [MIPS] MIPS 822 [PARISC] cr16 823 [S390] tod 824 [SH] SuperH 825 [SPARC64] tick 826 [X86-64] hpet,tsc 827 828 clocksource.arm_arch_timer.evtstrm= 829 [ARM,ARM64,EARLY] 830 Format: <bool> 831 Enable/disable the eventstream feature of the ARM 832 architected timer so that code using WFE-based polling 833 loops can be debugged more effectively on production 834 systems. 835 836 clocksource.verify_n_cpus= [KNL] 837 Limit the number of CPUs checked for clocksources 838 marked with CLOCK_SOURCE_VERIFY_PERCPU that 839 are marked unstable due to excessive skew. 840 A negative value says to check all CPUs, while 841 zero says not to check any. Values larger than 842 nr_cpu_ids are silently truncated to nr_cpu_ids. 843 The actual CPUs are chosen randomly, with 844 no replacement if the same CPU is chosen twice. 845 846 clocksource-wdtest.holdoff= [KNL] 847 Set the time in seconds that the clocksource 848 watchdog test waits before commencing its tests. 849 Defaults to zero when built as a module and to 850 10 seconds when built into the kernel. 851 852 cma=nn[MG]@[start[MG][-end[MG]]] 853 [KNL,CMA,EARLY] 854 Sets the size of kernel global memory area for 855 contiguous memory allocations and optionally the 856 placement constraint by the physical address range of 857 memory allocations. A value of 0 disables CMA 858 altogether. For more information, see 859 kernel/dma/contiguous.c 860 861 cma_pernuma=nn[MG] 862 [KNL,CMA,EARLY] 863 Sets the size of kernel per-numa memory area for 864 contiguous memory allocations. A value of 0 disables 865 per-numa CMA altogether. And If this option is not 866 specified, the default value is 0. 867 With per-numa CMA enabled, DMA users on node nid will 868 first try to allocate buffer from the pernuma area 869 which is located in node nid, if the allocation fails, 870 they will fallback to the global default memory area. 871 872 numa_cma=<node>:nn[MG][,<node>:nn[MG]] 873 [KNL,CMA,EARLY] 874 Sets the size of kernel numa memory area for 875 contiguous memory allocations. It will reserve CMA 876 area for the specified node. 877 878 With numa CMA enabled, DMA users on node nid will 879 first try to allocate buffer from the numa area 880 which is located in node nid, if the allocation fails, 881 they will fallback to the global default memory area. 882 883 cmo_free_hint= [PPC] Format: { yes | no } 884 Specify whether pages are marked as being inactive 885 when they are freed. This is used in CMO environments 886 to determine OS memory pressure for page stealing by 887 a hypervisor. 888 Default: yes 889 890 coherent_pool=nn[KMG] [ARM,KNL,EARLY] 891 Sets the size of memory pool for coherent, atomic dma 892 allocations, by default set to 256K. 893 894 com20020= [HW,NET] ARCnet - COM20020 chipset 895 Format: 896 <io>[,<irq>[,<nodeID>[,<backplane>[,<ckp>[,<timeout>]]]]] 897 898 com90io= [HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers) 899 Format: <io>[,<irq>] 900 901 com90xx= [HW,NET] 902 ARCnet - COM90xx chipset (memory-mapped buffers) 903 Format: <io>[,<irq>[,<memstart>]] 904 905 condev= [HW,S390] console device 906 conmode= 907 908 con3215_drop= [S390,EARLY] 3215 console drop mode. 909 Format: y|n|Y|N|1|0 910 When set to true, drop data on the 3215 console when 911 the console buffer is full. In this case the 912 operator using a 3270 terminal emulator (for example 913 x3270) does not have to enter the clear key for the 914 console output to advance and the kernel to continue. 915 This leads to a much faster boot time when a 3270 916 terminal emulator is active. If no 3270 terminal 917 emulator is used, this parameter has no effect. 918 919 console= [KNL] Output console device and options. 920 921 tty<n> Use the virtual console device <n>. 922 923 ttyS<n>[,options] 924 ttyUSB0[,options] 925 Use the specified serial port. The options are of 926 the form "bbbbpnf", where "bbbb" is the baud rate, 927 "p" is parity ("n", "o", or "e"), "n" is number of 928 bits, and "f" is flow control ("r" for RTS or 929 omit it). Default is "9600n8". 930 931 See Documentation/admin-guide/serial-console.rst for more 932 information. See 933 Documentation/networking/netconsole.rst for an 934 alternative. 935 936 <DEVNAME>:<n>.<n>[,options] 937 Use the specified serial port on the serial core bus. 938 The addressing uses DEVNAME of the physical serial port 939 device, followed by the serial core controller instance, 940 and the serial port instance. The options are the same 941 as documented for the ttyS addressing above. 942 943 The mapping of the serial ports to the tty instances 944 can be viewed with: 945 946 $ ls -d /sys/bus/serial-base/devices/*:*.*/tty/* 947 /sys/bus/serial-base/devices/00:04:0.0/tty/ttyS0 948 949 In the above example, the console can be addressed with 950 console=00:04:0.0. Note that a console addressed this 951 way will only get added when the related device driver 952 is ready. The use of an earlycon parameter in addition to 953 the console may be desired for console output early on. 954 955 uart[8250],io,<addr>[,options] 956 uart[8250],mmio,<addr>[,options] 957 uart[8250],mmio16,<addr>[,options] 958 uart[8250],mmio32,<addr>[,options] 959 uart[8250],0x<addr>[,options] 960 Start an early, polled-mode console on the 8250/16550 961 UART at the specified I/O port or MMIO address, 962 switching to the matching ttyS device later. 963 MMIO inter-register address stride is either 8-bit 964 (mmio), 16-bit (mmio16), or 32-bit (mmio32). 965 If none of [io|mmio|mmio16|mmio32], <addr> is assumed 966 to be equivalent to 'mmio'. 'options' are specified in 967 the same format described for ttyS above; if unspecified, 968 the h/w is not re-initialized. 969 970 hvc<n> Use the hypervisor console device <n>. This is for 971 both Xen and PowerPC hypervisors. 972 973 { null | "" } 974 Use to disable console output, i.e., to have kernel 975 console messages discarded. 976 This must be the only console= parameter used on the 977 kernel command line. 978 979 If the device connected to the port is not a TTY but a braille 980 device, prepend "brl," before the device type, for instance 981 console=brl,ttyS0 982 For now, only VisioBraille is supported. 983 984 console_msg_format= 985 [KNL] Change console messages format 986 default 987 By default we print messages on consoles in 988 "[time stamp] text\n" format (time stamp may not be 989 printed, depending on CONFIG_PRINTK_TIME or 990 `printk_time' param). 991 syslog 992 Switch to syslog format: "<%u>[time stamp] text\n" 993 IOW, each message will have a facility and loglevel 994 prefix. The format is similar to one used by syslog() 995 syscall, or to executing "dmesg -S --raw" or to reading 996 from /proc/kmsg. 997 998 consoleblank= [KNL] The console blank (screen saver) timeout in 999 seconds. A value of 0 disables the blank timer. 1000 Defaults to 0. 1001 1002 coredump_filter= 1003 [KNL] Change the default value for 1004 /proc/<pid>/coredump_filter. 1005 See also Documentation/filesystems/proc.rst. 1006 1007 coresight_cpu_debug.enable 1008 [ARM,ARM64] 1009 Format: <bool> 1010 Enable/disable the CPU sampling based debugging. 1011 0: default value, disable debugging 1012 1: enable debugging at boot time 1013 1014 cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver 1015 Format: 1016 <first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>] 1017 1018 cpuidle.off=1 [CPU_IDLE] 1019 disable the cpuidle sub-system 1020 1021 cpuidle.governor= 1022 [CPU_IDLE] Name of the cpuidle governor to use. 1023 1024 cpufreq.off=1 [CPU_FREQ] 1025 disable the cpufreq sub-system 1026 1027 cpufreq.default_governor= 1028 [CPU_FREQ] Name of the default cpufreq governor or 1029 policy to use. This governor must be registered in the 1030 kernel before the cpufreq driver probes. 1031 1032 cpu_init_udelay=N 1033 [X86,EARLY] Delay for N microsec between assert and de-assert 1034 of APIC INIT to start processors. This delay occurs 1035 on every CPU online, such as boot, and resume from suspend. 1036 Default: 10000 1037 1038 cpuhp.parallel= 1039 [SMP] Enable/disable parallel bringup of secondary CPUs 1040 Format: <bool> 1041 Default is enabled if CONFIG_HOTPLUG_PARALLEL=y. Otherwise 1042 the parameter has no effect. 1043 1044 crash_kexec_post_notifiers 1045 Only jump to kdump kernel after running the panic 1046 notifiers and dumping kmsg. This option increases 1047 the risks of a kdump failure, since some panic 1048 notifiers can make the crashed kernel more unstable. 1049 In configurations where kdump may not be reliable, 1050 running the panic notifiers could allow collecting 1051 more data on dmesg, like stack traces from other CPUS 1052 or extra data dumped by panic_print. Note that some 1053 configurations enable this option unconditionally, 1054 like Hyper-V, PowerPC (fadump) and AMD SEV-SNP. 1055 1056 crashkernel=size[KMG][@offset[KMG]] 1057 [KNL,EARLY] Using kexec, Linux can switch to a 'crash kernel' 1058 upon panic. This parameter reserves the physical 1059 memory region [offset, offset + size] for that kernel 1060 image. If '@offset' is omitted, then a suitable offset 1061 is selected automatically. 1062 [KNL, X86-64, ARM64, RISCV, LoongArch] Select a region 1063 under 4G first, and fall back to reserve region above 1064 4G when '@offset' hasn't been specified. 1065 See Documentation/admin-guide/kdump/kdump.rst for further details. 1066 1067 crashkernel=range1:size1[,range2:size2,...][@offset] 1068 [KNL] Same as above, but depends on the memory 1069 in the running system. The syntax of range is 1070 start-[end] where start and end are both 1071 a memory unit (amount[KMG]). See also 1072 Documentation/admin-guide/kdump/kdump.rst for an example. 1073 1074 crashkernel=size[KMG],high 1075 [KNL, X86-64, ARM64, RISCV, LoongArch] range could be 1076 above 4G. 1077 Allow kernel to allocate physical memory region from top, 1078 so could be above 4G if system have more than 4G ram 1079 installed. Otherwise memory region will be allocated 1080 below 4G, if available. 1081 It will be ignored if crashkernel=X is specified. 1082 crashkernel=size[KMG],low 1083 [KNL, X86-64, ARM64, RISCV, LoongArch] range under 4G. 1084 When crashkernel=X,high is passed, kernel could allocate 1085 physical memory region above 4G, that cause second kernel 1086 crash on system that require some amount of low memory, 1087 e.g. swiotlb requires at least 64M+32K low memory, also 1088 enough extra low memory is needed to make sure DMA buffers 1089 for 32-bit devices won't run out. Kernel would try to allocate 1090 default size of memory below 4G automatically. The default 1091 size is platform dependent. 1092 --> x86: max(swiotlb_size_or_default() + 8MiB, 256MiB) 1093 --> arm64: 128MiB 1094 --> riscv: 128MiB 1095 --> loongarch: 128MiB 1096 This one lets the user specify own low range under 4G 1097 for second kernel instead. 1098 0: to disable low allocation. 1099 It will be ignored when crashkernel=X,high is not used 1100 or memory reserved is below 4G. 1101 crashkernel=size[KMG],cma 1102 [KNL, X86, ppc] Reserve additional crash kernel memory from 1103 CMA. This reservation is usable by the first system's 1104 userspace memory and kernel movable allocations (memory 1105 balloon, zswap). Pages allocated from this memory range 1106 will not be included in the vmcore so this should not 1107 be used if dumping of userspace memory is intended and 1108 it has to be expected that some movable kernel pages 1109 may be missing from the dump. 1110 1111 A standard crashkernel reservation, as described above, 1112 is still needed to hold the crash kernel and initrd. 1113 1114 This option increases the risk of a kdump failure: DMA 1115 transfers configured by the first kernel may end up 1116 corrupting the second kernel's memory. 1117 1118 This reservation method is intended for systems that 1119 can't afford to sacrifice enough memory for standard 1120 crashkernel reservation and where less reliable and 1121 possibly incomplete kdump is preferable to no kdump at 1122 all. 1123 1124 cryptomgr.notests 1125 [KNL] Disable crypto self-tests 1126 1127 cs89x0_dma= [HW,NET] 1128 Format: <dma> 1129 1130 cs89x0_media= [HW,NET] 1131 Format: { rj45 | aui | bnc } 1132 1133 csdlock_debug= [KNL] Enable or disable debug add-ons of cross-CPU 1134 function call handling. When switched on, 1135 additional debug data is printed to the console 1136 in case a hanging CPU is detected, and that 1137 CPU is pinged again in order to try to resolve 1138 the hang situation. The default value of this 1139 option depends on the CSD_LOCK_WAIT_DEBUG_DEFAULT 1140 Kconfig option. 1141 1142 dasd= [HW,NET] 1143 See header of drivers/s390/block/dasd_devmap.c. 1144 1145 db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port 1146 (one device per port) 1147 Format: <port#>,<type> 1148 See also Documentation/input/devices/joystick-parport.rst 1149 1150 debug [KNL,EARLY] Enable kernel debugging (events log level). 1151 1152 debug_boot_weak_hash 1153 [KNL,EARLY] Enable printing [hashed] pointers early in the 1154 boot sequence. If enabled, we use a weak hash instead 1155 of siphash to hash pointers. Use this option if you are 1156 seeing instances of '(___ptrval___)') and need to see a 1157 value (hashed pointer) instead. Cryptographically 1158 insecure, please do not use on production kernels. 1159 1160 debug_locks_verbose= 1161 [KNL] verbose locking self-tests 1162 Format: <int> 1163 Print debugging info while doing the locking API 1164 self-tests. 1165 Bitmask for the various LOCKTYPE_ tests. Defaults to 0 1166 (no extra messages), setting it to -1 (all bits set) 1167 will print _a_lot_ more information - normally only 1168 useful to lockdep developers. 1169 1170 debug_objects [KNL,EARLY] Enable object debugging 1171 1172 debug_guardpage_minorder= 1173 [KNL,EARLY] When CONFIG_DEBUG_PAGEALLOC is set, this 1174 parameter allows control of the order of pages that will 1175 be intentionally kept free (and hence protected) by the 1176 buddy allocator. Bigger value increase the probability 1177 of catching random memory corruption, but reduce the 1178 amount of memory for normal system use. The maximum 1179 possible value is MAX_PAGE_ORDER/2. Setting this 1180 parameter to 1 or 2 should be enough to identify most 1181 random memory corruption problems caused by bugs in 1182 kernel or driver code when a CPU writes to (or reads 1183 from) a random memory location. Note that there exists 1184 a class of memory corruptions problems caused by buggy 1185 H/W or F/W or by drivers badly programming DMA 1186 (basically when memory is written at bus level and the 1187 CPU MMU is bypassed) which are not detectable by 1188 CONFIG_DEBUG_PAGEALLOC, hence this option will not 1189 help tracking down these problems. 1190 1191 debug_pagealloc= 1192 [KNL,EARLY] When CONFIG_DEBUG_PAGEALLOC is set, this parameter 1193 enables the feature at boot time. By default, it is 1194 disabled and the system will work mostly the same as a 1195 kernel built without CONFIG_DEBUG_PAGEALLOC. 1196 Note: to get most of debug_pagealloc error reports, it's 1197 useful to also enable the page_owner functionality. 1198 on: enable the feature 1199 1200 debugfs= [KNL,EARLY] This parameter enables what is exposed to 1201 userspace and debugfs internal clients. 1202 Format: { on, off } 1203 on: All functions are enabled. 1204 off: Filesystem is not registered and clients 1205 get a -EPERM as result when trying to register files 1206 or directories within debugfs. 1207 This is equivalent of the runtime functionality if 1208 debugfs was not enabled in the kernel at all. 1209 Default value is set in build-time with a kernel configuration. 1210 1211 debugpat [X86] Enable PAT debugging 1212 1213 default_hugepagesz= 1214 [HW] The size of the default HugeTLB page. This is 1215 the size represented by the legacy /proc/ hugepages 1216 APIs. In addition, this is the default hugetlb size 1217 used for shmget(), mmap() and mounting hugetlbfs 1218 filesystems. If not specified, defaults to the 1219 architecture's default huge page size. Huge page 1220 sizes are architecture dependent. See also 1221 Documentation/admin-guide/mm/hugetlbpage.rst. 1222 Format: size[KMG] 1223 1224 deferred_probe_timeout= 1225 [KNL] Debugging option to set a timeout in seconds for 1226 deferred probe to give up waiting on dependencies to 1227 probe. Only specific dependencies (subsystems or 1228 drivers) that have opted in will be ignored. A timeout 1229 of 0 will timeout at the end of initcalls; a negative 1230 value is treated as an infinite timeout value. If the 1231 timeout hasn't expired, it'll be restarted by each 1232 successful driver registration. This option will also 1233 dump out devices still on the deferred probe list after 1234 retrying. 1235 1236 delayacct [KNL] Enable per-task delay accounting 1237 1238 dell_smm_hwmon.ignore_dmi= 1239 [HW] Continue probing hardware even if DMI data 1240 indicates that the driver is running on unsupported 1241 hardware. 1242 1243 dell_smm_hwmon.force= 1244 [HW] Activate driver even if SMM BIOS signature does 1245 not match list of supported models and enable otherwise 1246 blacklisted features. 1247 1248 dell_smm_hwmon.power_status= 1249 [HW] Report power status in /proc/i8k 1250 (disabled by default). 1251 1252 dell_smm_hwmon.restricted= 1253 [HW] Allow controlling fans only if SYS_ADMIN 1254 capability is set. 1255 1256 dell_smm_hwmon.fan_mult= 1257 [HW] Factor to multiply fan speed with. 1258 1259 dell_smm_hwmon.fan_max= 1260 [HW] Maximum configurable fan speed. 1261 1262 dfltcc= [HW,S390] 1263 Format: { on | off | def_only | inf_only | always } 1264 on: s390 zlib hardware support for compression on 1265 level 1 and decompression (default) 1266 off: No s390 zlib hardware support 1267 def_only: s390 zlib hardware support for deflate 1268 only (compression on level 1) 1269 inf_only: s390 zlib hardware support for inflate 1270 only (decompression) 1271 always: Same as 'on' but ignores the selected compression 1272 level always using hardware support (used for debugging) 1273 1274 dhash_entries= [KNL] 1275 Set number of hash buckets for dentry cache. 1276 1277 disable_1tb_segments [PPC,EARLY] 1278 Disables the use of 1TB hash page table segments. This 1279 causes the kernel to fall back to 256MB segments which 1280 can be useful when debugging issues that require an SLB 1281 miss to occur. 1282 1283 disable= [IPV6] 1284 See Documentation/networking/ipv6.rst. 1285 1286 disable_radix [PPC,EARLY] 1287 Disable RADIX MMU mode on POWER9 1288 1289 disable_tlbie [PPC] 1290 Disable TLBIE instruction. Currently does not work 1291 with KVM, with HASH MMU, or with coherent accelerators. 1292 1293 disable_ddw [PPC/PSERIES,EARLY] 1294 Disable Dynamic DMA Window support. Use this 1295 to workaround buggy firmware. 1296 1297 disable_ipv6= [IPV6] 1298 See Documentation/networking/ipv6.rst. 1299 1300 disable_mtrr_cleanup [X86,EARLY] 1301 The kernel tries to adjust MTRR layout from continuous 1302 to discrete, to make X server driver able to add WB 1303 entry later. This parameter disables that. 1304 1305 disable_mtrr_trim [X86, Intel and AMD only,EARLY] 1306 By default the kernel will trim any uncacheable 1307 memory out of your available memory pool based on 1308 MTRR settings. This parameter disables that behavior, 1309 possibly causing your machine to run very slowly. 1310 1311 disable_timer_pin_1 [X86,EARLY] 1312 Disable PIN 1 of APIC timer 1313 Can be useful to work around chipset bugs. 1314 1315 dis_ucode_ldr [X86] Disable the microcode loader. 1316 1317 dma_debug=off If the kernel is compiled with DMA_API_DEBUG support, 1318 this option disables the debugging code at boot. 1319 1320 dma_debug_entries=<number> 1321 This option allows to tune the number of preallocated 1322 entries for DMA-API debugging code. One entry is 1323 required per DMA-API allocation. Use this if the 1324 DMA-API debugging code disables itself because the 1325 architectural default is too low. 1326 1327 dma_debug_driver=<driver_name> 1328 With this option the DMA-API debugging driver 1329 filter feature can be enabled at boot time. Just 1330 pass the driver to filter for as the parameter. 1331 The filter can be disabled or changed to another 1332 driver later using sysfs. 1333 1334 reg_file_data_sampling= 1335 [X86] Controls mitigation for Register File Data 1336 Sampling (RFDS) vulnerability. RFDS is a CPU 1337 vulnerability which may allow userspace to infer 1338 kernel data values previously stored in floating point 1339 registers, vector registers, or integer registers. 1340 RFDS only affects Intel Atom processors. 1341 1342 on: Turns ON the mitigation. 1343 off: Turns OFF the mitigation. 1344 1345 This parameter overrides the compile time default set 1346 by CONFIG_MITIGATION_RFDS. Mitigation cannot be 1347 disabled when other VERW based mitigations (like MDS) 1348 are enabled. In order to disable RFDS mitigation all 1349 VERW based mitigations need to be disabled. 1350 1351 For details see: 1352 Documentation/admin-guide/hw-vuln/reg-file-data-sampling.rst 1353 1354 dm_verity.keyring_unsealed= 1355 [KNL] When set to 1, leave the dm-verity keyring 1356 unsealed after initialization so userspace can 1357 provision keys. Once the keyring is restricted 1358 it becomes active and is searched during signature 1359 verification. 1360 1361 driver_async_probe= [KNL] 1362 List of driver names to be probed asynchronously. * 1363 matches with all driver names. If * is specified, the 1364 rest of the listed driver names are those that will NOT 1365 match the *. 1366 Format: <driver_name1>,<driver_name2>... 1367 1368 drm.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>] 1369 Broken monitors, graphic adapters, KVMs and EDIDless 1370 panels may send no or incorrect EDID data sets. 1371 This parameter allows to specify an EDID data sets 1372 in the /lib/firmware directory that are used instead. 1373 An EDID data set will only be used for a particular 1374 connector, if its name and a colon are prepended to 1375 the EDID name. Each connector may use a unique EDID 1376 data set by separating the files with a comma. An EDID 1377 data set with no connector name will be used for 1378 any connectors not explicitly specified. 1379 1380 dscc4.setup= [NET] 1381 1382 dt_cpu_ftrs= [PPC,EARLY] 1383 Format: {"off" | "known"} 1384 Control how the dt_cpu_ftrs device-tree binding is 1385 used for CPU feature discovery and setup (if it 1386 exists). 1387 off: Do not use it, fall back to legacy cpu table. 1388 known: Do not pass through unknown features to guests 1389 or userspace, only those that the kernel is aware of. 1390 1391 dump_apple_properties [X86] 1392 Dump name and content of EFI device properties on 1393 x86 Macs. Useful for driver authors to determine 1394 what data is available or for reverse-engineering. 1395 1396 dyndbg[="val"] [KNL,DYNAMIC_DEBUG] 1397 <module>.dyndbg[="val"] 1398 Enable debug messages at boot time. See 1399 Documentation/admin-guide/dynamic-debug-howto.rst 1400 for details. 1401 1402 early_ioremap_debug [KNL,EARLY] 1403 Enable debug messages in early_ioremap support. This 1404 is useful for tracking down temporary early mappings 1405 which are not unmapped. 1406 1407 earlycon= [KNL,EARLY] Output early console device and options. 1408 1409 When used with no options, the early console is 1410 determined by stdout-path property in device tree's 1411 chosen node or the ACPI SPCR table if supported by 1412 the platform. 1413 1414 cdns,<addr>[,options] 1415 Start an early, polled-mode console on a Cadence 1416 (xuartps) serial port at the specified address. Only 1417 supported option is baud rate. If baud rate is not 1418 specified, the serial port must already be setup and 1419 configured. 1420 1421 uart[8250],io,<addr>[,options[,uartclk]] 1422 uart[8250],mmio,<addr>[,options[,uartclk]] 1423 uart[8250],mmio32,<addr>[,options[,uartclk]] 1424 uart[8250],mmio32be,<addr>[,options[,uartclk]] 1425 uart[8250],0x<addr>[,options] 1426 Start an early, polled-mode console on the 8250/16550 1427 UART at the specified I/O port or MMIO address. 1428 MMIO inter-register address stride is either 8-bit 1429 (mmio) or 32-bit (mmio32 or mmio32be). 1430 If none of [io|mmio|mmio32|mmio32be], <addr> is assumed 1431 to be equivalent to 'mmio'. 'options' are specified 1432 in the same format described for "console=ttyS<n>"; if 1433 unspecified, the h/w is not initialized. 'uartclk' is 1434 the uart clock frequency; if unspecified, it is set 1435 to 'BASE_BAUD' * 16. 1436 1437 pl011,<addr> 1438 pl011,mmio32,<addr> 1439 Start an early, polled-mode console on a pl011 serial 1440 port at the specified address. The pl011 serial port 1441 must already be setup and configured. Options are not 1442 yet supported. If 'mmio32' is specified, then only 1443 the driver will use only 32-bit accessors to read/write 1444 the device registers. 1445 1446 liteuart,<addr> 1447 Start an early console on a litex serial port at the 1448 specified address. The serial port must already be 1449 setup and configured. Options are not yet supported. 1450 1451 meson,<addr> 1452 Start an early, polled-mode console on a meson serial 1453 port at the specified address. The serial port must 1454 already be setup and configured. Options are not yet 1455 supported. 1456 1457 msm_serial,<addr> 1458 Start an early, polled-mode console on an msm serial 1459 port at the specified address. The serial port 1460 must already be setup and configured. Options are not 1461 yet supported. 1462 1463 msm_serial_dm,<addr> 1464 Start an early, polled-mode console on an msm serial 1465 dm port at the specified address. The serial port 1466 must already be setup and configured. Options are not 1467 yet supported. 1468 1469 owl,<addr> 1470 Start an early, polled-mode console on a serial port 1471 of an Actions Semi SoC, such as S500 or S900, at the 1472 specified address. The serial port must already be 1473 setup and configured. Options are not yet supported. 1474 1475 rda,<addr> 1476 Start an early, polled-mode console on a serial port 1477 of an RDA Micro SoC, such as RDA8810PL, at the 1478 specified address. The serial port must already be 1479 setup and configured. Options are not yet supported. 1480 1481 sbi 1482 Use RISC-V SBI (Supervisor Binary Interface) for early 1483 console. 1484 1485 smh Use ARM semihosting calls for early console. 1486 1487 s3c2410,<addr> 1488 s3c2412,<addr> 1489 s3c2440,<addr> 1490 s3c6400,<addr> 1491 s5pv210,<addr> 1492 exynos4210,<addr> 1493 Use early console provided by serial driver available 1494 on Samsung SoCs, requires selecting proper type and 1495 a correct base address of the selected UART port. The 1496 serial port must already be setup and configured. 1497 Options are not yet supported. 1498 1499 lantiq,<addr> 1500 Start an early, polled-mode console on a lantiq serial 1501 (lqasc) port at the specified address. The serial port 1502 must already be setup and configured. Options are not 1503 yet supported. 1504 1505 lpuart,<addr> 1506 lpuart32,<addr> 1507 Use early console provided by Freescale LP UART driver 1508 found on Freescale Vybrid and QorIQ LS1021A processors. 1509 A valid base address must be provided, and the serial 1510 port must already be setup and configured. 1511 1512 ec_imx21,<addr> 1513 ec_imx6q,<addr> 1514 Start an early, polled-mode, output-only console on the 1515 Freescale i.MX UART at the specified address. The UART 1516 must already be setup and configured. 1517 1518 ar3700_uart,<addr> 1519 Start an early, polled-mode console on the 1520 Armada 3700 serial port at the specified 1521 address. The serial port must already be setup 1522 and configured. Options are not yet supported. 1523 1524 qcom_geni,<addr> 1525 Start an early, polled-mode console on a Qualcomm 1526 Generic Interface (GENI) based serial port at the 1527 specified address. The serial port must already be 1528 setup and configured. Options are not yet supported. 1529 1530 efifb,[options] 1531 Start an early, unaccelerated console on the EFI 1532 memory mapped framebuffer (if available). On cache 1533 coherent non-x86 systems that use system memory for 1534 the framebuffer, pass the 'ram' option so that it is 1535 mapped with the correct attributes. 1536 1537 linflex,<addr> 1538 Use early console provided by Freescale LINFlexD UART 1539 serial driver for NXP S32V234 SoCs. A valid base 1540 address must be provided, and the serial port must 1541 already be setup and configured. 1542 1543 earlyprintk= [X86,SH,ARM,M68k,S390,UM,EARLY] 1544 earlyprintk=vga 1545 earlyprintk=sclp 1546 earlyprintk=xen 1547 earlyprintk=serial[,ttySn[,baudrate]] 1548 earlyprintk=serial[,0x...[,baudrate]] 1549 earlyprintk=ttySn[,baudrate] 1550 earlyprintk=dbgp[debugController#] 1551 earlyprintk=mmio32,membase[,{nocfg|baudrate}] 1552 earlyprintk=pciserial[,force],bus:device.function[,{nocfg|baudrate}] 1553 earlyprintk=xdbc[xhciController#] 1554 earlyprintk=bios 1555 1556 earlyprintk is useful when the kernel crashes before 1557 the normal console is initialized. It is not enabled by 1558 default because it has some cosmetic problems. 1559 1560 Use "nocfg" to skip UART configuration, assume 1561 BIOS/firmware has configured UART correctly. 1562 1563 Append ",keep" to not disable it when the real console 1564 takes over. 1565 1566 Only one of vga, serial, or usb debug port can 1567 be used at a time. 1568 1569 Currently only ttyS0 and ttyS1 may be specified by 1570 name. Other I/O ports may be explicitly specified 1571 on some architectures (x86 and arm at least) by 1572 replacing ttySn with an I/O port address, like this: 1573 earlyprintk=serial,0x1008,115200 1574 You can find the port for a given device in 1575 /proc/tty/driver/serial: 1576 2: uart:ST16650V2 port:00001008 irq:18 ... 1577 1578 Interaction with the standard serial driver is not 1579 very good. 1580 1581 The VGA output is eventually overwritten by 1582 the real console. 1583 1584 The xen option can only be used in Xen domains. 1585 1586 The sclp output can only be used on s390. 1587 1588 The bios output can only be used on SuperH. 1589 1590 The optional "force" to "pciserial" enables use of a 1591 PCI device even when its classcode is not of the 1592 UART class. 1593 1594 edac_report= [HW,EDAC] Control how to report EDAC event 1595 Format: {"on" | "off" | "force"} 1596 on: enable EDAC to report H/W event. May be overridden 1597 by other higher priority error reporting module. 1598 off: disable H/W event reporting through EDAC. 1599 force: enforce the use of EDAC to report H/W event. 1600 default: on. 1601 1602 edd= [EDD] 1603 Format: {"off" | "on" | "skip[mbr]"} 1604 1605 efi= [EFI,EARLY] 1606 Format: { "debug", "disable_early_pci_dma", 1607 "nochunk", "noruntime", "nosoftreserve", 1608 "novamap", "no_disable_early_pci_dma" } 1609 debug: enable misc debug output. 1610 disable_early_pci_dma: disable the busmaster bit on all 1611 PCI bridges while in the EFI boot stub. 1612 nochunk: disable reading files in "chunks" in the EFI 1613 boot stub, as chunking can cause problems with some 1614 firmware implementations. 1615 noruntime : disable EFI runtime services support 1616 nosoftreserve: The EFI_MEMORY_SP (Specific Purpose) 1617 attribute may cause the kernel to reserve the 1618 memory range for a memory mapping driver to 1619 claim. Specify efi=nosoftreserve to disable this 1620 reservation and treat the memory by its base type 1621 (i.e. EFI_CONVENTIONAL_MEMORY / "System RAM"). 1622 novamap: do not call SetVirtualAddressMap(). 1623 no_disable_early_pci_dma: Leave the busmaster bit set 1624 on all PCI bridges while in the EFI boot stub 1625 1626 efi_no_storage_paranoia [EFI,X86,EARLY] 1627 Using this parameter you can use more than 50% of 1628 your efi variable storage. Use this parameter only if 1629 you are really sure that your UEFI does sane gc and 1630 fulfills the spec otherwise your board may brick. 1631 1632 efivar_ssdt= [EFI; X86] Name of an EFI variable that contains an SSDT 1633 that is to be dynamically loaded by Linux. If there are 1634 multiple variables with the same name but with different 1635 vendor GUIDs, all of them will be loaded. See 1636 Documentation/admin-guide/acpi/ssdt-overlays.rst for details. 1637 1638 1639 eisa_irq_edge= [PARISC,HW] 1640 See header of drivers/parisc/eisa.c. 1641 1642 ekgdboc= [X86,KGDB,EARLY] Allow early kernel console debugging 1643 Format: ekgdboc=kbd 1644 1645 This is designed to be used in conjunction with 1646 the boot argument: earlyprintk=vga 1647 1648 This parameter works in place of the kgdboc parameter 1649 but can only be used if the backing tty is available 1650 very early in the boot process. For early debugging 1651 via a serial port see kgdboc_earlycon instead. 1652 1653 elfcorehdr=[size[KMG]@]offset[KMG] [PPC,SH,X86,S390,EARLY] 1654 Specifies physical address of start of kernel core 1655 image elf header and optionally the size. Generally 1656 kexec loader will pass this option to capture kernel. 1657 See Documentation/admin-guide/kdump/kdump.rst for details. 1658 1659 enable_mtrr_cleanup [X86,EARLY] 1660 The kernel tries to adjust MTRR layout from continuous 1661 to discrete, to make X server driver able to add WB 1662 entry later. This parameter enables that. 1663 1664 enable_timer_pin_1 [X86] 1665 Enable PIN 1 of APIC timer 1666 Can be useful to work around chipset bugs 1667 (in particular on some ATI chipsets). 1668 The kernel tries to set a reasonable default. 1669 1670 enforcing= [SELINUX] Set initial enforcing status. 1671 Format: {"0" | "1"} 1672 See security/selinux/Kconfig help text. 1673 0 -- permissive (log only, no denials). 1674 1 -- enforcing (deny and log). 1675 Default value is 0. 1676 Value can be changed at runtime via 1677 /sys/fs/selinux/enforce. 1678 1679 erst_disable [ACPI] 1680 Disable Error Record Serialization Table (ERST) 1681 support. 1682 1683 ether= [HW,NET] Ethernet cards parameters 1684 This option is obsoleted by the "netdev=" option, which 1685 has equivalent usage. See its documentation for details. 1686 1687 evm= [EVM] 1688 Format: { "fix" } 1689 Permit 'security.evm' to be updated regardless of 1690 current integrity status. 1691 1692 early_page_ext [KNL,EARLY] Enforces page_ext initialization to earlier 1693 stages so cover more early boot allocations. 1694 Please note that as side effect some optimizations 1695 might be disabled to achieve that (e.g. parallelized 1696 memory initialization is disabled) so the boot process 1697 might take longer, especially on systems with a lot of 1698 memory. Available with CONFIG_PAGE_EXTENSION=y. 1699 1700 failslab= 1701 fail_usercopy= 1702 fail_page_alloc= 1703 fail_skb_realloc= 1704 fail_make_request=[KNL] 1705 General fault injection mechanism. 1706 Format: <interval>,<probability>,<space>,<times> 1707 See also Documentation/fault-injection/. 1708 1709 fb_tunnels= [NET] 1710 Format: { initns | none } 1711 See Documentation/admin-guide/sysctl/net.rst for 1712 fb_tunnels_only_for_init_ns 1713 1714 floppy= [HW] 1715 See Documentation/admin-guide/blockdev/floppy.rst. 1716 1717 forcepae [X86-32] 1718 Forcefully enable Physical Address Extension (PAE). 1719 Many Pentium M systems disable PAE but may have a 1720 functionally usable PAE implementation. 1721 Warning: use of this parameter will taint the kernel 1722 and may cause unknown problems. 1723 1724 fred= [X86-64] 1725 Enable/disable Flexible Return and Event Delivery. 1726 Format: { on | off } 1727 on: enable FRED when it's present, the default setting. 1728 off: disable FRED. 1729 1730 ftrace=[tracer] 1731 [FTRACE] will set and start the specified tracer 1732 as early as possible in order to facilitate early 1733 boot debugging. 1734 1735 ftrace_boot_snapshot 1736 [FTRACE] On boot up, a snapshot will be taken of the 1737 ftrace ring buffer that can be read at: 1738 /sys/kernel/tracing/snapshot. 1739 This is useful if you need tracing information from kernel 1740 boot up that is likely to be overridden by user space 1741 start up functionality. 1742 1743 Optionally, the snapshot can also be defined for a tracing 1744 instance that was created by the trace_instance= command 1745 line parameter. 1746 1747 trace_instance=foo,sched_switch ftrace_boot_snapshot=foo 1748 1749 The above will cause the "foo" tracing instance to trigger 1750 a snapshot at the end of boot up. 1751 1752 ftrace_dump_on_oops[=2(orig_cpu) | =<instance>][,<instance> | 1753 ,<instance>=2(orig_cpu)] 1754 [FTRACE] will dump the trace buffers on oops. 1755 If no parameter is passed, ftrace will dump global 1756 buffers of all CPUs, if you pass 2 or orig_cpu, it 1757 will dump only the buffer of the CPU that triggered 1758 the oops, or the specific instance will be dumped if 1759 its name is passed. Multiple instance dump is also 1760 supported, and instances are separated by commas. Each 1761 instance supports only dump on CPU that triggered the 1762 oops by passing 2 or orig_cpu to it. 1763 1764 ftrace_dump_on_oops=foo=orig_cpu 1765 1766 The above will dump only the buffer of "foo" instance 1767 on CPU that triggered the oops. 1768 1769 ftrace_dump_on_oops,foo,bar=orig_cpu 1770 1771 The above will dump global buffer on all CPUs, the 1772 buffer of "foo" instance on all CPUs and the buffer 1773 of "bar" instance on CPU that triggered the oops. 1774 1775 ftrace_filter=[function-list] 1776 [FTRACE] Limit the functions traced by the function 1777 tracer at boot up. function-list is a comma-separated 1778 list of functions. This list can be changed at run 1779 time by the set_ftrace_filter file in the debugfs 1780 tracing directory. 1781 1782 ftrace_notrace=[function-list] 1783 [FTRACE] Do not trace the functions specified in 1784 function-list. This list can be changed at run time 1785 by the set_ftrace_notrace file in the debugfs 1786 tracing directory. 1787 1788 ftrace_graph_filter=[function-list] 1789 [FTRACE] Limit the top level callers functions traced 1790 by the function graph tracer at boot up. 1791 function-list is a comma-separated list of functions 1792 that can be changed at run time by the 1793 set_graph_function file in the debugfs tracing directory. 1794 1795 ftrace_graph_notrace=[function-list] 1796 [FTRACE] Do not trace from the functions specified in 1797 function-list. This list is a comma-separated list of 1798 functions that can be changed at run time by the 1799 set_graph_notrace file in the debugfs tracing directory. 1800 1801 ftrace_graph_max_depth=<uint> 1802 [FTRACE] Used with the function graph tracer. This is 1803 the max depth it will trace into a function. This value 1804 can be changed at run time by the max_graph_depth file 1805 in the tracefs tracing directory. default: 0 (no limit) 1806 1807 fw_devlink= [KNL,EARLY] Create device links between consumer and supplier 1808 devices by scanning the firmware to infer the 1809 consumer/supplier relationships. This feature is 1810 especially useful when drivers are loaded as modules as 1811 it ensures proper ordering of tasks like device probing 1812 (suppliers first, then consumers), supplier boot state 1813 clean up (only after all consumers have probed), 1814 suspend/resume & runtime PM (consumers first, then 1815 suppliers). 1816 Format: { off | permissive | on | rpm } 1817 off -- Don't create device links from firmware info. 1818 permissive -- Create device links from firmware info 1819 but use it only for ordering boot state clean 1820 up (sync_state() calls). 1821 on -- Create device links from firmware info and use it 1822 to enforce probe and suspend/resume ordering. 1823 rpm -- Like "on", but also use to order runtime PM. 1824 1825 fw_devlink.strict=<bool> 1826 [KNL,EARLY] Treat all inferred dependencies as mandatory 1827 dependencies. This only applies for fw_devlink=on|rpm. 1828 Format: <bool> 1829 1830 fw_devlink.sync_state = 1831 [KNL,EARLY] When all devices that could probe have finished 1832 probing, this parameter controls what to do with 1833 devices that haven't yet received their sync_state() 1834 calls. 1835 Format: { strict | timeout } 1836 strict -- Default. Continue waiting on consumers to 1837 probe successfully. 1838 timeout -- Give up waiting on consumers and call 1839 sync_state() on any devices that haven't yet 1840 received their sync_state() calls after 1841 deferred_probe_timeout has expired or by 1842 late_initcall() if !CONFIG_MODULES. 1843 1844 gamecon.map[2|3]= 1845 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad 1846 support via parallel port (up to 5 devices per port) 1847 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5> 1848 See also Documentation/input/devices/joystick-parport.rst 1849 1850 gamma= [HW,DRM] 1851 1852 gart_fix_e820= [X86-64,EARLY] disable the fix e820 for K8 GART 1853 Format: off | on 1854 default: on 1855 1856 gather_data_sampling= 1857 [X86,INTEL,EARLY] Control the Gather Data Sampling (GDS) 1858 mitigation. 1859 1860 Gather Data Sampling is a hardware vulnerability which 1861 allows unprivileged speculative access to data which was 1862 previously stored in vector registers. 1863 1864 This issue is mitigated by default in updated microcode. 1865 The mitigation may have a performance impact but can be 1866 disabled. On systems without the microcode mitigation 1867 disabling AVX serves as a mitigation. 1868 1869 force: Disable AVX to mitigate systems without 1870 microcode mitigation. No effect if the microcode 1871 mitigation is present. Known to cause crashes in 1872 userspace with buggy AVX enumeration. 1873 1874 off: Disable GDS mitigation. 1875 1876 gbpages [X86] Use GB pages for kernel direct mappings. 1877 1878 gcov_persist= [GCOV] When non-zero (default), profiling data for 1879 kernel modules is saved and remains accessible via 1880 debugfs, even when the module is unloaded/reloaded. 1881 When zero, profiling data is discarded and associated 1882 debugfs files are removed at module unload time. 1883 1884 goldfish [X86] Enable the goldfish android emulator platform. 1885 Don't use this when you are not running on the 1886 android emulator 1887 1888 gpio-mockup.gpio_mockup_ranges 1889 [HW] Sets the ranges of gpiochip of for this device. 1890 Format: <start1>,<end1>,<start2>,<end2>... 1891 gpio-mockup.gpio_mockup_named_lines 1892 [HW] Let the driver know GPIO lines should be named. 1893 1894 gpt [EFI] Forces disk with valid GPT signature but 1895 invalid Protective MBR to be treated as GPT. If the 1896 primary GPT is corrupted, it enables the backup/alternate 1897 GPT to be used instead. 1898 1899 grcan.enable0= [HW] Configuration of physical interface 0. Determines 1900 the "Enable 0" bit of the configuration register. 1901 Format: 0 | 1 1902 Default: 0 1903 grcan.enable1= [HW] Configuration of physical interface 1. Determines 1904 the "Enable 0" bit of the configuration register. 1905 Format: 0 | 1 1906 Default: 0 1907 grcan.select= [HW] Select which physical interface to use. 1908 Format: 0 | 1 1909 Default: 0 1910 grcan.txsize= [HW] Sets the size of the tx buffer. 1911 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0. 1912 Default: 1024 1913 grcan.rxsize= [HW] Sets the size of the rx buffer. 1914 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0. 1915 Default: 1024 1916 1917 hardened_usercopy= 1918 [KNL] Under CONFIG_HARDENED_USERCOPY, whether 1919 hardening is enabled for this boot. Hardened 1920 usercopy checking is used to protect the kernel 1921 from reading or writing beyond known memory 1922 allocation boundaries as a proactive defense 1923 against bounds-checking flaws in the kernel's 1924 copy_to_user()/copy_from_user() interface. 1925 The default is determined by 1926 CONFIG_HARDENED_USERCOPY_DEFAULT_ON. 1927 on Perform hardened usercopy checks. 1928 off Disable hardened usercopy checks. 1929 1930 hardlockup_all_cpu_backtrace= 1931 [KNL] Should the hard-lockup detector generate 1932 backtraces on all cpus. 1933 Format: 0 | 1 1934 1935 hash_pointers= 1936 [KNL,EARLY] 1937 By default, when pointers are printed to the console 1938 or buffers via the %p format string, that pointer is 1939 "hashed", i.e. obscured by hashing the pointer value. 1940 This is a security feature that hides actual kernel 1941 addresses from unprivileged users, but it also makes 1942 debugging the kernel more difficult since unequal 1943 pointers can no longer be compared. The choices are: 1944 Format: { auto | always | never } 1945 Default: auto 1946 1947 auto - Hash pointers unless slab_debug is enabled. 1948 always - Always hash pointers (even if slab_debug is 1949 enabled). 1950 never - Never hash pointers. This option should only 1951 be specified when debugging the kernel. Do 1952 not use on production kernels. The boot 1953 param "no_hash_pointers" is an alias for 1954 this mode. 1955 1956 For controlling hashing dynamically at runtime, 1957 use the "kernel.kptr_restrict" sysctl instead. 1958 1959 hashdist= [KNL,NUMA] Large hashes allocated during boot 1960 are distributed across NUMA nodes. Defaults on 1961 for 64-bit NUMA, off otherwise. 1962 Format: 0 | 1 (for off | on) 1963 1964 hd= [EIDE] (E)IDE hard drive subsystem geometry 1965 Format: <cyl>,<head>,<sect> 1966 1967 hest_disable [ACPI] 1968 Disable Hardware Error Source Table (HEST) support; 1969 corresponding firmware-first mode error processing 1970 logic will be disabled. 1971 1972 hibernate= [HIBERNATION] 1973 noresume Don't check if there's a hibernation image 1974 present during boot. 1975 nocompress Don't compress/decompress hibernation images. 1976 no Disable hibernation and resume. 1977 protect_image Turn on image protection during restoration 1978 (that will set all pages holding image data 1979 during restoration read-only). 1980 1981 hibernate.compressor= [HIBERNATION] Compression algorithm to be 1982 used with hibernation. 1983 Format: { lzo | lz4 } 1984 Default: lzo 1985 1986 lzo: Select LZO compression algorithm to 1987 compress/decompress hibernation image. 1988 1989 lz4: Select LZ4 compression algorithm to 1990 compress/decompress hibernation image. 1991 1992 hibernate.pm_test_delay= 1993 [HIBERNATION] 1994 Sets the number of seconds to remain in a hibernation test 1995 mode before resuming the system (see 1996 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG 1997 is set. Default value is 5. 1998 1999 hibernate_compression_threads= 2000 [HIBERNATION] 2001 Set the number of threads used for compressing or decompressing 2002 hibernation images. 2003 2004 Format: <integer> 2005 Default: 3 2006 Minimum: 1 2007 Example: hibernate_compression_threads=4 2008 2009 highmem=nn[KMG] [KNL,BOOT,EARLY] forces the highmem zone to have an exact 2010 size of <nn>. This works even on boxes that have no 2011 highmem otherwise. This also works to reduce highmem 2012 size on bigger boxes. 2013 2014 highres= [KNL] Enable/disable high resolution timer mode. 2015 Valid parameters: "on", "off" 2016 Default: "on" 2017 2018 hlt [BUGS=ARM,SH] 2019 2020 hostname= [KNL,EARLY] Set the hostname (aka UTS nodename). 2021 Format: <string> 2022 This allows setting the system's hostname during early 2023 startup. This sets the name returned by gethostname. 2024 Using this parameter to set the hostname makes it 2025 possible to ensure the hostname is correctly set before 2026 any userspace processes run, avoiding the possibility 2027 that a process may call gethostname before the hostname 2028 has been explicitly set, resulting in the calling 2029 process getting an incorrect result. The string must 2030 not exceed the maximum allowed hostname length (usually 2031 64 characters) and will be truncated otherwise. 2032 2033 hpet= [X86-32,HPET] option to control HPET usage 2034 Format: { enable (default) | disable | force | 2035 verbose } 2036 disable: disable HPET and use PIT instead 2037 force: allow force enabled of undocumented chips (ICH4, 2038 VIA, nVidia) 2039 verbose: show contents of HPET registers during setup 2040 2041 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET 2042 registers. Default set by CONFIG_HPET_MMAP_DEFAULT. 2043 2044 hugepages= [HW,EARLY] Number of HugeTLB pages to allocate at boot. 2045 If this follows hugepagesz (below), it specifies 2046 the number of pages of hugepagesz to be allocated. 2047 If this is the first HugeTLB parameter on the command 2048 line, it specifies the number of pages to allocate for 2049 the default huge page size. If using node format, the 2050 number of pages to allocate per-node can be specified. 2051 See also Documentation/admin-guide/mm/hugetlbpage.rst. 2052 Format: <integer> or (node format) 2053 <node>:<integer>[,<node>:<integer>] 2054 2055 hugepagesz= 2056 [HW,EARLY] The size of the HugeTLB pages. This is 2057 used in conjunction with hugepages (above) to 2058 allocate huge pages of a specific size at boot. The 2059 pair hugepagesz=X hugepages=Y can be specified once 2060 for each supported huge page size. Huge page sizes 2061 are architecture dependent. See also 2062 Documentation/admin-guide/mm/hugetlbpage.rst. 2063 Format: size[KMG] 2064 2065 hugepage_alloc_threads= 2066 [HW] The number of threads that should be used to 2067 allocate hugepages during boot. This option can be 2068 used to improve system bootup time when allocating 2069 a large amount of huge pages. 2070 The default value is 25% of the available hardware threads. 2071 2072 Note that this parameter only applies to non-gigantic huge pages. 2073 2074 hugetlb_cma= [HW,CMA,EARLY] The size of a CMA area used for allocation 2075 of gigantic hugepages. Or using node format, the size 2076 of a CMA area per node can be specified. 2077 Format: nn[KMGTPE] or (node format) 2078 <node>:nn[KMGTPE][,<node>:nn[KMGTPE]] 2079 2080 Reserve a CMA area of given size and allocate gigantic 2081 hugepages using the CMA allocator. If enabled, the 2082 boot-time allocation of gigantic hugepages is skipped. 2083 2084 hugetlb_cma_only= 2085 [HW,CMA,EARLY] When allocating new HugeTLB pages, only 2086 try to allocate from the CMA areas. 2087 2088 This option does nothing if hugetlb_cma= is not also 2089 specified. 2090 2091 hugetlb_free_vmemmap= 2092 [KNL] Requires CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP 2093 enabled. 2094 Control if HugeTLB Vmemmap Optimization (HVO) is enabled. 2095 Allows heavy hugetlb users to free up some more 2096 memory (7 * PAGE_SIZE for each 2MB hugetlb page). 2097 Format: { on | off (default) } 2098 2099 on: enable HVO 2100 off: disable HVO 2101 2102 Built with CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP_DEFAULT_ON=y, 2103 the default is on. 2104 2105 Note that the vmemmap pages may be allocated from the added 2106 memory block itself when memory_hotplug.memmap_on_memory is 2107 enabled, those vmemmap pages cannot be optimized even if this 2108 feature is enabled. Other vmemmap pages not allocated from 2109 the added memory block itself do not be affected. 2110 2111 hung_task_panic= 2112 [KNL] Number of hung tasks to trigger kernel panic. 2113 Format: <int> 2114 2115 When set to a non-zero value, a kernel panic will be triggered if 2116 the number of detected hung tasks reaches this value. 2117 2118 0: don't panic 2119 1: panic immediately on first hung task 2120 N: panic after N hung tasks are detected in a single scan 2121 2122 The default value is controlled by the 2123 CONFIG_BOOTPARAM_HUNG_TASK_PANIC build-time option. The value 2124 selected by this boot parameter can be changed later by the 2125 kernel.hung_task_panic sysctl. 2126 2127 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC) 2128 terminal devices. Valid values: 0..8 2129 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs. 2130 If specified, z/VM IUCV HVC accepts connections 2131 from listed z/VM user IDs only. 2132 2133 hv_nopvspin [X86,HYPER_V,EARLY] 2134 Disables the paravirt spinlock optimizations 2135 which allow the hypervisor to 'idle' the guest 2136 on lock contention. 2137 2138 hw_protection= [HW] 2139 Format: reboot | shutdown 2140 2141 Hardware protection action taken on critical events like 2142 overtemperature or imminent voltage loss. 2143 2144 i2c_bus= [HW] Override the default board specific I2C bus speed 2145 or register an additional I2C bus that is not 2146 registered from board initialization code. 2147 Format: 2148 <bus_id>,<clkrate> 2149 2150 i2c_touchscreen_props= [HW,ACPI,X86] 2151 Set device-properties for ACPI-enumerated I2C-attached 2152 touchscreen, to e.g. fix coordinates of upside-down 2153 mounted touchscreens. If you need this option please 2154 submit a drivers/platform/x86/touchscreen_dmi.c patch 2155 adding a DMI quirk for this. 2156 2157 Format: 2158 <ACPI_HW_ID>:<prop_name>=<val>[:prop_name=val][:...] 2159 Where <val> is one of: 2160 Omit "=<val>" entirely Set a boolean device-property 2161 Unsigned number Set a u32 device-property 2162 Anything else Set a string device-property 2163 2164 Examples (split over multiple lines): 2165 i2c_touchscreen_props=GDIX1001:touchscreen-inverted-x: 2166 touchscreen-inverted-y 2167 2168 i2c_touchscreen_props=MSSL1680:touchscreen-size-x=1920: 2169 touchscreen-size-y=1080:touchscreen-inverted-y: 2170 firmware-name=gsl1680-vendor-model.fw:silead,home-button 2171 2172 i8042.debug [HW] Toggle i8042 debug mode 2173 i8042.unmask_kbd_data 2174 [HW] Enable printing of interrupt data from the KBD port 2175 (disabled by default, and as a pre-condition 2176 requires that i8042.debug=1 be enabled) 2177 i8042.direct [HW] Put keyboard port into non-translated mode 2178 i8042.dumbkbd [HW] Pretend that controller can only read data from 2179 keyboard and cannot control its state 2180 (Don't attempt to blink the leds) 2181 i8042.noaux [HW] Don't check for auxiliary (== mouse) port 2182 i8042.nokbd [HW] Don't check/create keyboard port 2183 i8042.noloop [HW] Disable the AUX Loopback command while probing 2184 for the AUX port 2185 i8042.nomux [HW] Don't check presence of an active multiplexing 2186 controller 2187 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX 2188 controllers 2189 i8042.notimeout [HW] Ignore timeout condition signalled by controller 2190 i8042.reset [HW] Reset the controller during init, cleanup and 2191 suspend-to-ram transitions, only during s2r 2192 transitions, or never reset 2193 Format: { 1 | Y | y | 0 | N | n } 2194 1, Y, y: always reset controller 2195 0, N, n: don't ever reset controller 2196 Default: only on s2r transitions on x86; most other 2197 architectures force reset to be always executed 2198 i8042.unlock [HW] Unlock (ignore) the keylock 2199 i8042.kbdreset [HW] Reset device connected to KBD port 2200 i8042.probe_defer 2201 [HW] Allow deferred probing upon i8042 probe errors 2202 2203 i810= [HW,DRM] 2204 2205 i915.invert_brightness= 2206 [DRM] Invert the sense of the variable that is used to 2207 set the brightness of the panel backlight. Normally a 2208 brightness value of 0 indicates backlight switched off, 2209 and the maximum of the brightness value sets the backlight 2210 to maximum brightness. If this parameter is set to 0 2211 (default) and the machine requires it, or this parameter 2212 is set to 1, a brightness value of 0 sets the backlight 2213 to maximum brightness, and the maximum of the brightness 2214 value switches the backlight off. 2215 -1 -- never invert brightness 2216 0 -- machine default 2217 1 -- force brightness inversion 2218 2219 ia32_emulation= [X86-64] 2220 Format: <bool> 2221 When true, allows loading 32-bit programs and executing 32-bit 2222 syscalls, essentially overriding IA32_EMULATION_DEFAULT_DISABLED at 2223 boot time. When false, unconditionally disables IA32 emulation. 2224 2225 2226 idle= [X86,EARLY] 2227 Format: idle=poll, idle=halt, idle=nomwait 2228 2229 idle=poll: Don't do power saving in the idle loop 2230 using HLT, but poll for rescheduling event. This will 2231 make the CPUs eat a lot more power, but may be useful 2232 to get slightly better performance in multiprocessor 2233 benchmarks. It also makes some profiling using 2234 performance counters more accurate. Please note that 2235 on systems with MONITOR/MWAIT support (like Intel 2236 EM64T CPUs) this option has no performance advantage 2237 over the normal idle loop. It may also interact badly 2238 with hyperthreading. 2239 2240 idle=halt: Halt is forced to be used for CPU idle. 2241 In such case C2/C3 won't be used again. 2242 2243 idle=nomwait: Disable mwait for CPU C-states 2244 2245 idxd.sva= [HW] 2246 Format: <bool> 2247 Allow force disabling of Shared Virtual Memory (SVA) 2248 support for the idxd driver. By default it is set to 2249 true (1). 2250 2251 idxd.tc_override= [HW] 2252 Format: <bool> 2253 Allow override of default traffic class configuration 2254 for the device. By default it is set to false (0). 2255 2256 ieee754= [MIPS] Select IEEE Std 754 conformance mode 2257 Format: { strict | legacy | 2008 | relaxed | emulated } 2258 Default: strict 2259 2260 Choose which programs will be accepted for execution 2261 based on the IEEE 754 NaN encoding(s) supported by 2262 the FPU and the NaN encoding requested with the value 2263 of an ELF file header flag individually set by each 2264 binary. Hardware implementations are permitted to 2265 support either or both of the legacy and the 2008 NaN 2266 encoding mode. 2267 2268 Available settings are as follows: 2269 strict accept binaries that request a NaN encoding 2270 supported by the FPU 2271 legacy only accept legacy-NaN binaries, if supported 2272 by the FPU 2273 2008 only accept 2008-NaN binaries, if supported 2274 by the FPU 2275 relaxed accept any binaries regardless of whether 2276 supported by the FPU 2277 emulated accept any binaries but enable FPU emulator 2278 if binary mode is unsupported by the FPU. 2279 2280 The FPU emulator is always able to support both NaN 2281 encodings, so if no FPU hardware is present or it has 2282 been disabled with 'nofpu', then the settings of 2283 'legacy' and '2008' strap the emulator accordingly, 2284 'relaxed' straps the emulator for both legacy-NaN and 2285 2008-NaN, whereas 'strict' enables legacy-NaN only on 2286 legacy processors and both NaN encodings on MIPS32 or 2287 MIPS64 CPUs. 2288 2289 The setting for ABS.fmt/NEG.fmt instruction execution 2290 mode generally follows that for the NaN encoding, 2291 except where unsupported by hardware. 2292 2293 ignore_loglevel [KNL,EARLY] 2294 Ignore loglevel setting - this will print /all/ 2295 kernel messages to the console. Useful for debugging. 2296 We also add it as printk module parameter, so users 2297 could change it dynamically, usually by 2298 /sys/module/printk/parameters/ignore_loglevel. 2299 2300 ignore_rlimit_data 2301 Ignore RLIMIT_DATA setting for data mappings, 2302 print warning at first misuse. Can be changed via 2303 /sys/module/kernel/parameters/ignore_rlimit_data. 2304 2305 ihash_entries= [KNL] 2306 Set number of hash buckets for inode cache. 2307 2308 ima_appraise= [IMA] appraise integrity measurements 2309 Format: { "off" | "enforce" | "fix" | "log" } 2310 default: "enforce" 2311 2312 ima_appraise_tcb [IMA] Deprecated. Use ima_policy= instead. 2313 The builtin appraise policy appraises all files 2314 owned by uid=0. 2315 2316 ima_canonical_fmt [IMA] 2317 Use the canonical format for the binary runtime 2318 measurements, instead of host native format. 2319 2320 ima_hash= [IMA] 2321 Format: { md5 | sha1 | rmd160 | sha256 | sha384 2322 | sha512 | ... } 2323 default: "sha1" 2324 2325 The list of supported hash algorithms is defined 2326 in crypto/hash_info.h. 2327 2328 ima_policy= [IMA] 2329 The builtin policies to load during IMA setup. 2330 Format: "tcb | appraise_tcb | secure_boot | 2331 fail_securely | critical_data" 2332 2333 The "tcb" policy measures all programs exec'd, files 2334 mmap'd for exec, and all files opened with the read 2335 mode bit set by either the effective uid (euid=0) or 2336 uid=0. 2337 2338 The "appraise_tcb" policy appraises the integrity of 2339 all files owned by root. 2340 2341 The "secure_boot" policy appraises the integrity 2342 of files (eg. kexec kernel image, kernel modules, 2343 firmware, policy, etc) based on file signatures. 2344 2345 The "fail_securely" policy forces file signature 2346 verification failure also on privileged mounted 2347 filesystems with the SB_I_UNVERIFIABLE_SIGNATURE 2348 flag. 2349 2350 The "critical_data" policy measures kernel integrity 2351 critical data. 2352 2353 ima_tcb [IMA] Deprecated. Use ima_policy= instead. 2354 Load a policy which meets the needs of the Trusted 2355 Computing Base. This means IMA will measure all 2356 programs exec'd, files mmap'd for exec, and all files 2357 opened for read by uid=0. 2358 2359 ima_template= [IMA] 2360 Select one of defined IMA measurements template formats. 2361 Formats: { "ima" | "ima-ng" | "ima-ngv2" | "ima-sig" | 2362 "ima-sigv2" } 2363 Default: "ima-ng" 2364 2365 ima_template_fmt= 2366 [IMA] Define a custom template format. 2367 Format: { "field1|...|fieldN" } 2368 2369 ima= [IMA] Enable or disable IMA 2370 Format: { "off" | "on" } 2371 Default: "on" 2372 Note that disabling IMA is limited to kdump kernel. 2373 2374 indirect_target_selection= [X86,Intel] Mitigation control for Indirect 2375 Target Selection(ITS) bug in Intel CPUs. Updated 2376 microcode is also required for a fix in IBPB. 2377 2378 on: Enable mitigation (default). 2379 off: Disable mitigation. 2380 force: Force the ITS bug and deploy default 2381 mitigation. 2382 vmexit: Only deploy mitigation if CPU is affected by 2383 guest/host isolation part of ITS. 2384 stuff: Deploy RSB-fill mitigation when retpoline is 2385 also deployed. Otherwise, deploy the default 2386 mitigation. 2387 2388 For details see: 2389 Documentation/admin-guide/hw-vuln/indirect-target-selection.rst 2390 2391 init= [KNL] 2392 Format: <full_path> 2393 Run specified binary instead of /sbin/init as init 2394 process. 2395 2396 initcall_debug [KNL] Trace initcalls as they are executed. Useful 2397 for working out where the kernel is dying during 2398 startup. 2399 2400 initcall_blacklist= [KNL] Do not execute a comma-separated list of 2401 initcall functions. Useful for debugging built-in 2402 modules and initcalls. 2403 2404 initramfs_async= [KNL] 2405 Format: <bool> 2406 Default: 1 2407 This parameter controls whether the initramfs 2408 image is unpacked asynchronously, concurrently 2409 with devices being probed and 2410 initialized. This should normally just work, 2411 but as a debugging aid, one can get the 2412 historical behaviour of the initramfs 2413 unpacking being completed before device_ and 2414 late_ initcalls. 2415 2416 initrd= [BOOT,EARLY] Specify the location of the initial ramdisk 2417 2418 initrdmem= [KNL,EARLY] Specify a physical address and size from which to 2419 load the initrd. If an initrd is compiled in or 2420 specified in the bootparams, it takes priority over this 2421 setting. 2422 Format: ss[KMG],nn[KMG] 2423 Default is 0, 0 2424 2425 init_on_alloc= [MM,EARLY] Fill newly allocated pages and heap objects with 2426 zeroes. 2427 Format: 0 | 1 2428 Default set by CONFIG_INIT_ON_ALLOC_DEFAULT_ON. 2429 2430 init_on_free= [MM,EARLY] Fill freed pages and heap objects with zeroes. 2431 Format: 0 | 1 2432 Default set by CONFIG_INIT_ON_FREE_DEFAULT_ON. 2433 2434 init_pkru= [X86] Specify the default memory protection keys rights 2435 register contents for all processes. 0x55555554 by 2436 default (disallow access to all but pkey 0). Can 2437 override in debugfs after boot. 2438 2439 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver 2440 Format: <irq> 2441 2442 int_pln_enable [X86] Enable power limit notification interrupt 2443 2444 integrity_audit=[IMA] 2445 Format: { "0" | "1" } 2446 0 -- basic integrity auditing messages. (Default) 2447 1 -- additional integrity auditing messages. 2448 2449 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option 2450 on 2451 Enable intel iommu driver. 2452 off 2453 Disable intel iommu driver. 2454 igfx_off [Default Off] 2455 By default, gfx is mapped as normal device. If a gfx 2456 device has a dedicated DMAR unit, the DMAR unit is 2457 bypassed by not enabling DMAR with this option. In 2458 this case, gfx device will use physical address for 2459 DMA. 2460 strict [Default Off] 2461 Deprecated, equivalent to iommu.strict=1. 2462 sp_off [Default Off] 2463 By default, super page will be supported if Intel IOMMU 2464 has the capability. With this option, super page will 2465 not be supported. 2466 sm_on 2467 Enable the Intel IOMMU scalable mode if the hardware 2468 advertises that it has support for the scalable mode 2469 translation. 2470 sm_off 2471 Disallow use of the Intel IOMMU scalable mode. 2472 tboot_noforce [Default Off] 2473 Do not force the Intel IOMMU enabled under tboot. 2474 By default, tboot will force Intel IOMMU on, which 2475 could harm performance of some high-throughput 2476 devices like 40GBit network cards, even if identity 2477 mapping is enabled. 2478 Note that using this option lowers the security 2479 provided by tboot because it makes the system 2480 vulnerable to DMA attacks. 2481 2482 intel_idle.max_cstate= [KNL,HW,ACPI,X86] 2483 0 disables intel_idle and fall back on acpi_idle. 2484 1 to 9 specify maximum depth of C-state. 2485 2486 intel_pstate= [X86,EARLY] 2487 disable 2488 Do not enable intel_pstate as the default 2489 scaling driver for the supported processors 2490 active 2491 Use intel_pstate driver to bypass the scaling 2492 governors layer of cpufreq and provides it own 2493 algorithms for p-state selection. There are two 2494 P-state selection algorithms provided by 2495 intel_pstate in the active mode: powersave and 2496 performance. The way they both operate depends 2497 on whether or not the hardware managed P-states 2498 (HWP) feature has been enabled in the processor 2499 and possibly on the processor model. 2500 passive 2501 Use intel_pstate as a scaling driver, but configure it 2502 to work with generic cpufreq governors (instead of 2503 enabling its internal governor). This mode cannot be 2504 used along with the hardware-managed P-states (HWP) 2505 feature. 2506 force 2507 Enable intel_pstate on systems that prohibit it by default 2508 in favor of acpi-cpufreq. Forcing the intel_pstate driver 2509 instead of acpi-cpufreq may disable platform features, such 2510 as thermal controls and power capping, that rely on ACPI 2511 P-States information being indicated to OSPM and therefore 2512 should be used with caution. This option does not work with 2513 processors that aren't supported by the intel_pstate driver 2514 or on platforms that use pcc-cpufreq instead of acpi-cpufreq. 2515 no_hwp 2516 Do not enable hardware P state control (HWP) 2517 if available. 2518 hwp_only 2519 Only load intel_pstate on systems which support 2520 hardware P state control (HWP) if available. 2521 support_acpi_ppc 2522 Enforce ACPI _PPC performance limits. If the Fixed ACPI 2523 Description Table, specifies preferred power management 2524 profile as "Enterprise Server" or "Performance Server", 2525 then this feature is turned on by default. 2526 per_cpu_perf_limits 2527 Allow per-logical-CPU P-State performance control limits using 2528 cpufreq sysfs interface 2529 no_cas 2530 Do not enable capacity-aware scheduling (CAS) on 2531 hybrid systems 2532 2533 intremap= [X86-64,Intel-IOMMU,EARLY] 2534 on enable Interrupt Remapping (default) 2535 off disable Interrupt Remapping 2536 nosid disable Source ID checking 2537 no_x2apic_optout 2538 BIOS x2APIC opt-out request will be ignored 2539 nopost disable Interrupt Posting 2540 posted_msi 2541 enable MSIs delivered as posted interrupts 2542 2543 iomem= Disable strict checking of access to MMIO memory 2544 strict regions from userspace. 2545 relaxed 2546 2547 iommu= [X86,EARLY] 2548 2549 off 2550 Don't initialize and use any kind of IOMMU. 2551 2552 force 2553 Force the use of the hardware IOMMU even when 2554 it is not actually needed (e.g. because < 3 GB 2555 memory). 2556 2557 noforce 2558 Don't force hardware IOMMU usage when it is not 2559 needed. (default). 2560 2561 merge 2562 Do scatter-gather (SG) merging. Implies "force" 2563 (experimental). 2564 2565 nomerge 2566 Don't do scatter-gather (SG) merging. 2567 2568 biomerge 2569 Do scatter-gather (SG) merging. Implies "force" 2570 (experimental). [same as "merge"] 2571 2572 panic 2573 Always panic when IOMMU overflows. 2574 2575 nopanic 2576 Don't panic on IOMMU overflows. 2577 2578 pt 2579 Use passththrough mode by default 2580 (Equivalent to iommu.passthrough=1) 2581 2582 nopt 2583 Use translated mode for DMA by default 2584 (Equivalent to iommu.passthrough=0) 2585 2586 soft 2587 Use software bounce buffering (SWIOTLB) (default for 2588 Intel machines). This can be used to prevent the usage 2589 of an available hardware IOMMU. 2590 2591 usedac 2592 Use the DAC on VIA PCI bridge 2593 (default: disable the VIA PCI bridge DAC) 2594 2595 AMD Gart HW IOMMU-specific options: (CONFIG_GART_IOMMU) 2596 2597 <size> 2598 Set the size of the remapping area in bytes. 2599 2600 allowed 2601 Overwrite iommu off workarounds for specific chipsets 2602 2603 force 2604 Overwrite iommu off workarounds for specific chipsets 2605 2606 fullflush 2607 Flush IOMMU on each allocation (default). 2608 2609 nofullflush 2610 Don't use IOMMU fullflush. 2611 2612 memaper[=<order>] 2613 Allocate an own aperture over RAM with size 2614 32MB<<order. (default: order=1, i.e. 64MB) 2615 2616 noaperture 2617 Ask the IOMMU not to touch the aperture for AGP. 2618 2619 noagp 2620 Don't initialize the AGP driver and use full aperture. 2621 2622 iommu= [PPC/POWERNV] 2623 2624 nobypass 2625 Disable IOMMU bypass, using IOMMU for PCI devices. 2626 2627 iommu.forcedac= [ARM64,X86,EARLY] Control IOVA allocation for PCI devices. 2628 Format: { "0" | "1" } 2629 0 - Try to allocate a 32-bit DMA address first, before 2630 falling back to the full range if needed. 2631 1 - Allocate directly from the full usable range, 2632 forcing Dual Address Cycle for PCI cards supporting 2633 greater than 32-bit addressing. 2634 2635 iommu.strict= [ARM64,X86,S390,EARLY] Configure TLB invalidation behaviour 2636 Format: { "0" | "1" } 2637 0 - Lazy mode. 2638 Request that DMA unmap operations use deferred 2639 invalidation of hardware TLBs, for increased 2640 throughput at the cost of reduced device isolation. 2641 Will fall back to strict mode if not supported by 2642 the relevant IOMMU driver. 2643 1 - Strict mode. 2644 DMA unmap operations invalidate IOMMU hardware TLBs 2645 synchronously. 2646 unset - Use value of CONFIG_IOMMU_DEFAULT_DMA_{LAZY,STRICT}. 2647 Note: on x86, strict mode specified via one of the 2648 legacy driver-specific options takes precedence. 2649 2650 iommu.passthrough= 2651 [ARM64,X86,EARLY] Configure DMA to bypass the IOMMU by default. 2652 Format: { "0" | "1" } 2653 0 - Use IOMMU translation for DMA. 2654 1 - Bypass the IOMMU for DMA. 2655 unset - Use value of CONFIG_IOMMU_DEFAULT_PASSTHROUGH. 2656 2657 iommu.debug_pagealloc= 2658 [KNL,EARLY] When CONFIG_IOMMU_DEBUG_PAGEALLOC is set, this 2659 parameter enables the feature at boot time. By default, it 2660 is disabled and the system behaves the same way as a kernel 2661 built without CONFIG_IOMMU_DEBUG_PAGEALLOC. 2662 Format: { "0" | "1" } 2663 0 - Sanitizer disabled. 2664 1 - Sanitizer enabled, expect runtime overhead. 2665 2666 io7= [HW] IO7 for Marvel-based Alpha systems 2667 See comment before marvel_specify_io7 in 2668 arch/alpha/kernel/core_marvel.c. 2669 2670 io_delay= [X86,EARLY] I/O delay method 2671 0x80 2672 Standard port 0x80 based delay 2673 0xed 2674 Alternate port 0xed based delay (needed on some systems) 2675 udelay 2676 Simple two microseconds delay 2677 none 2678 No delay 2679 2680 ip= [IP_PNP] 2681 See Documentation/admin-guide/nfs/nfsroot.rst. 2682 2683 ipcmni_extend [KNL,EARLY] Extend the maximum number of unique System V 2684 IPC identifiers from 32,768 to 16,777,216. 2685 2686 ipe.enforce= [IPE] 2687 Format: <bool> 2688 Determine whether IPE starts in permissive (0) or 2689 enforce (1) mode. The default is enforce. 2690 2691 ipe.success_audit= 2692 [IPE] 2693 Format: <bool> 2694 Start IPE with success auditing enabled, emitting 2695 an audit event when a binary is allowed. The default 2696 is 0. 2697 2698 irqaffinity= [SMP] Set the default irq affinity mask 2699 The argument is a cpu list, as described above. 2700 2701 irqchip.gicv2_force_probe= 2702 [ARM,ARM64,EARLY] 2703 Format: <bool> 2704 Force the kernel to look for the second 4kB page 2705 of a GICv2 controller even if the memory range 2706 exposed by the device tree is too small. 2707 2708 irqchip.gicv3_nolpi= 2709 [ARM,ARM64,EARLY] 2710 Force the kernel to ignore the availability of 2711 LPIs (and by consequence ITSs). Intended for system 2712 that use the kernel as a bootloader, and thus want 2713 to let secondary kernels in charge of setting up 2714 LPIs. 2715 2716 irqchip.gicv3_pseudo_nmi= [ARM64,EARLY] 2717 Enables support for pseudo-NMIs in the kernel. This 2718 requires the kernel to be built with 2719 CONFIG_ARM64_PSEUDO_NMI. 2720 2721 irqchip.riscv_imsic_noipi 2722 [RISC-V,EARLY] 2723 Force the kernel to not use IMSIC software injected MSIs 2724 as IPIs. Intended for system where IMSIC is trap-n-emulated, 2725 and thus want to reduce MMIO traps when triggering IPIs 2726 to multiple harts. 2727 2728 irqfixup [HW] 2729 When an interrupt is not handled search all handlers 2730 for it. Intended to get systems with badly broken 2731 firmware running. 2732 2733 irqhandler.duration_warn_us= [KNL] 2734 Warn if an IRQ handler exceeds the specified duration 2735 threshold in microseconds. Useful for identifying 2736 long-running IRQs in the system. 2737 2738 irqpoll [HW] 2739 When an interrupt is not handled search all handlers 2740 for it. Also check all handlers each timer 2741 interrupt. Intended to get systems with badly broken 2742 firmware running. 2743 2744 isapnp= [ISAPNP] 2745 Format: <RDP>,<reset>,<pci_scan>,<verbosity> 2746 2747 isolcpus= [KNL,SMP,ISOL] Isolate a given set of CPUs from disturbance. 2748 Format: [flag-list,]<cpu-list> 2749 2750 Specify one or more CPUs to isolate from disturbances 2751 specified in the flag list (default: domain): 2752 2753 nohz 2754 Disable the tick when a single task runs as well as 2755 disabling other kernel noises like having RCU callbacks 2756 offloaded. This is equivalent to the nohz_full parameter. 2757 2758 A residual 1Hz tick is offloaded to workqueues, which you 2759 need to affine to housekeeping through the global 2760 workqueue's affinity configured via the 2761 /sys/devices/virtual/workqueue/cpumask sysfs file, or 2762 by using the 'domain' flag described below. 2763 2764 NOTE: by default the global workqueue runs on all CPUs, 2765 so to protect individual CPUs the 'cpumask' file has to 2766 be configured manually after bootup. 2767 2768 domain 2769 Isolate from the general SMP balancing and scheduling 2770 algorithms. Note that performing domain isolation this way 2771 is irreversible: it's not possible to bring back a CPU to 2772 the domains once isolated through this boot time 2773 configuration. Use cpusets for a dynamic configuration 2774 which can be altered at runtime. For details see 2775 Documentation/admin-guide/cpu-isolation.rst. 2776 2777 You can move a process onto or off an "isolated" CPU via 2778 the CPU affinity syscalls or cpuset. 2779 <cpu number> begins at 0 and the maximum value is 2780 "number of CPUs in system - 1". 2781 2782 managed_irq 2783 2784 Isolate from being targeted by managed interrupts 2785 which have an interrupt mask containing isolated 2786 CPUs. The affinity of managed interrupts is 2787 handled by the kernel and cannot be changed via 2788 the /proc/irq/* interfaces. 2789 2790 This isolation is best effort and only effective 2791 if the automatically assigned interrupt mask of a 2792 device queue contains isolated and housekeeping 2793 CPUs. If housekeeping CPUs are online then such 2794 interrupts are directed to the housekeeping CPU 2795 so that IO submitted on the housekeeping CPU 2796 cannot disturb the isolated CPU. 2797 2798 If a queue's affinity mask contains only isolated 2799 CPUs then this parameter has no effect on the 2800 interrupt routing decision, though interrupts are 2801 only delivered when tasks running on those 2802 isolated CPUs submit IO. IO submitted on 2803 housekeeping CPUs has no influence on those 2804 queues. 2805 2806 The format of <cpu-list> is described above. 2807 2808 iucv= [HW,NET] 2809 2810 ivrs_ioapic [HW,X86-64] 2811 Provide an override to the IOAPIC-ID<->DEVICE-ID 2812 mapping provided in the IVRS ACPI table. 2813 By default, PCI segment is 0, and can be omitted. 2814 2815 For example, to map IOAPIC-ID decimal 10 to 2816 PCI segment 0x1 and PCI device 00:14.0, 2817 write the parameter as: 2818 ivrs_ioapic=10@0001:00:14.0 2819 2820 Deprecated formats: 2821 * To map IOAPIC-ID decimal 10 to PCI device 00:14.0 2822 write the parameter as: 2823 ivrs_ioapic[10]=00:14.0 2824 * To map IOAPIC-ID decimal 10 to PCI segment 0x1 and 2825 PCI device 00:14.0 write the parameter as: 2826 ivrs_ioapic[10]=0001:00:14.0 2827 2828 ivrs_hpet [HW,X86-64] 2829 Provide an override to the HPET-ID<->DEVICE-ID 2830 mapping provided in the IVRS ACPI table. 2831 By default, PCI segment is 0, and can be omitted. 2832 2833 For example, to map HPET-ID decimal 10 to 2834 PCI segment 0x1 and PCI device 00:14.0, 2835 write the parameter as: 2836 ivrs_hpet=10@0001:00:14.0 2837 2838 Deprecated formats: 2839 * To map HPET-ID decimal 0 to PCI device 00:14.0 2840 write the parameter as: 2841 ivrs_hpet[0]=00:14.0 2842 * To map HPET-ID decimal 10 to PCI segment 0x1 and 2843 PCI device 00:14.0 write the parameter as: 2844 ivrs_ioapic[10]=0001:00:14.0 2845 2846 ivrs_acpihid [HW,X86-64] 2847 Provide an override to the ACPI-HID:UID<->DEVICE-ID 2848 mapping provided in the IVRS ACPI table. 2849 By default, PCI segment is 0, and can be omitted. 2850 2851 For example, to map UART-HID:UID AMD0020:0 to 2852 PCI segment 0x1 and PCI device ID 00:14.5, 2853 write the parameter as: 2854 ivrs_acpihid=AMD0020:0@0001:00:14.5 2855 2856 Deprecated formats: 2857 * To map UART-HID:UID AMD0020:0 to PCI segment is 0, 2858 PCI device ID 00:14.5, write the parameter as: 2859 ivrs_acpihid[00:14.5]=AMD0020:0 2860 * To map UART-HID:UID AMD0020:0 to PCI segment 0x1 and 2861 PCI device ID 00:14.5, write the parameter as: 2862 ivrs_acpihid[0001:00:14.5]=AMD0020:0 2863 2864 js= [HW,JOY] Analog joystick 2865 See Documentation/input/joydev/joystick.rst. 2866 2867 kasan_multi_shot 2868 [KNL] Enforce KASAN (Kernel Address Sanitizer) to print 2869 report on every invalid memory access. Without this 2870 parameter KASAN will print report only for the first 2871 invalid access. 2872 2873 keep_bootcon [KNL,EARLY] 2874 Do not unregister boot console at start. This is only 2875 useful for debugging when something happens in the window 2876 between unregistering the boot console and initializing 2877 the real console. 2878 2879 keepinitrd [HW,ARM] See retain_initrd. 2880 2881 kernelcore= [KNL,X86,PPC,EARLY] 2882 Format: nn[KMGTPE] | nn% | "mirror" 2883 This parameter specifies the amount of memory usable by 2884 the kernel for non-movable allocations. The requested 2885 amount is spread evenly throughout all nodes in the 2886 system as ZONE_NORMAL. The remaining memory is used for 2887 movable memory in its own zone, ZONE_MOVABLE. In the 2888 event, a node is too small to have both ZONE_NORMAL and 2889 ZONE_MOVABLE, kernelcore memory will take priority and 2890 other nodes will have a larger ZONE_MOVABLE. 2891 2892 ZONE_MOVABLE is used for the allocation of pages that 2893 may be reclaimed or moved by the page migration 2894 subsystem. Note that allocations like PTEs-from-HighMem 2895 still use the HighMem zone if it exists, and the Normal 2896 zone if it does not. 2897 2898 It is possible to specify the exact amount of memory in 2899 the form of "nn[KMGTPE]", a percentage of total system 2900 memory in the form of "nn%", or "mirror". If "mirror" 2901 option is specified, mirrored (reliable) memory is used 2902 for non-movable allocations and remaining memory is used 2903 for Movable pages. "nn[KMGTPE]", "nn%", and "mirror" 2904 are exclusive, so you cannot specify multiple forms. 2905 2906 kfence.burst= [MM,KFENCE] The number of additional successive 2907 allocations to be attempted through KFENCE for each 2908 sample interval. 2909 Format: <unsigned integer> 2910 Default: 0 2911 2912 kfence.check_on_panic= 2913 [MM,KFENCE] Whether to check all KFENCE-managed objects' 2914 canaries on panic. 2915 Format: <bool> 2916 Default: false 2917 2918 kfence.deferrable= 2919 [MM,KFENCE] Whether to use a deferrable timer to trigger 2920 allocations. This avoids forcing CPU wake-ups if the 2921 system is idle, at the risk of a less predictable 2922 sample interval. 2923 Format: <bool> 2924 Default: CONFIG_KFENCE_DEFERRABLE 2925 2926 kfence.fault= [MM,KFENCE] Controls the behavior when a KFENCE 2927 error is detected. 2928 report - print the error report and continue (default). 2929 oops - print the error report and oops. 2930 panic - print the error report and panic. 2931 2932 kfence.sample_interval= 2933 [MM,KFENCE] KFENCE's sample interval in milliseconds. 2934 Format: <unsigned integer> 2935 0 - Disable KFENCE. 2936 >0 - Enabled KFENCE with given sample interval. 2937 Default: CONFIG_KFENCE_SAMPLE_INTERVAL 2938 2939 kfence.skip_covered_thresh= 2940 [MM,KFENCE] If pool utilization reaches this threshold 2941 (pool usage%), KFENCE limits currently covered 2942 allocations of the same source from further filling 2943 up the pool. 2944 Format: <unsigned integer> 2945 Default: 75 2946 2947 kgdbdbgp= [KGDB,HW,EARLY] kgdb over EHCI usb debug port. 2948 Format: <Controller#>[,poll interval] 2949 The controller # is the number of the ehci usb debug 2950 port as it is probed via PCI. The poll interval is 2951 optional and is the number seconds in between 2952 each poll cycle to the debug port in case you need 2953 the functionality for interrupting the kernel with 2954 gdb or control-c on the dbgp connection. When 2955 not using this parameter you use sysrq-g to break into 2956 the kernel debugger. 2957 2958 kgdboc= [KGDB,HW] kgdb over consoles. 2959 Requires a tty driver that supports console polling, 2960 or a supported polling keyboard driver (non-usb). 2961 Serial only format: <serial_device>[,baud] 2962 keyboard only format: kbd 2963 keyboard and serial format: kbd,<serial_device>[,baud] 2964 Optional Kernel mode setting: 2965 kms, kbd format: kms,kbd 2966 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud] 2967 2968 kgdboc_earlycon= [KGDB,HW,EARLY] 2969 If the boot console provides the ability to read 2970 characters and can work in polling mode, you can use 2971 this parameter to tell kgdb to use it as a backend 2972 until the normal console is registered. Intended to 2973 be used together with the kgdboc parameter which 2974 specifies the normal console to transition to. 2975 2976 The name of the early console should be specified 2977 as the value of this parameter. Note that the name of 2978 the early console might be different than the tty 2979 name passed to kgdboc. It's OK to leave the value 2980 blank and the first boot console that implements 2981 read() will be picked. 2982 2983 kgdbwait [KGDB,EARLY] Stop kernel execution and enter the 2984 kernel debugger at the earliest opportunity. 2985 2986 kho= [KEXEC,EARLY] 2987 Format: { "0" | "1" | "off" | "on" | "y" | "n" } 2988 Enables or disables Kexec HandOver. 2989 "0" | "off" | "n" - kexec handover is disabled 2990 "1" | "on" | "y" - kexec handover is enabled 2991 2992 kho_scratch= [KEXEC,EARLY] 2993 Format: ll[KMG],mm[KMG],nn[KMG] | nn% 2994 Defines the size of the KHO scratch region. The KHO 2995 scratch regions are physically contiguous memory 2996 ranges that can only be used for non-kernel 2997 allocations. That way, even when memory is heavily 2998 fragmented with handed over memory, the kexeced 2999 kernel will always have enough contiguous ranges to 3000 bootstrap itself. 3001 3002 It is possible to specify the exact amount of 3003 memory in the form of "ll[KMG],mm[KMG],nn[KMG]" 3004 where the first parameter defines the size of a low 3005 memory scratch area, the second parameter defines 3006 the size of a global scratch area and the third 3007 parameter defines the size of additional per-node 3008 scratch areas. The form "nn%" defines scale factor 3009 (in percents) of memory that was used during boot. 3010 3011 kmac= [MIPS] Korina ethernet MAC address. 3012 Configure the RouterBoard 532 series on-chip 3013 Ethernet adapter MAC address. 3014 3015 kmemleak= [KNL,EARLY] Boot-time kmemleak enable/disable 3016 Valid arguments: on, off 3017 Default: on 3018 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y, 3019 the default is off. 3020 3021 kprobe_event=[probe-list] 3022 [FTRACE] Add kprobe events and enable at boot time. 3023 The probe-list is a semicolon delimited list of probe 3024 definitions. Each definition is same as kprobe_events 3025 interface, but the parameters are comma delimited. 3026 For example, to add a kprobe event on vfs_read with 3027 arg1 and arg2, add to the command line; 3028 3029 kprobe_event=p,vfs_read,$arg1,$arg2 3030 3031 See also Documentation/trace/kprobetrace.rst "Kernel 3032 Boot Parameter" section. 3033 3034 kpti= [ARM64,EARLY] Control page table isolation of 3035 user and kernel address spaces. 3036 Default: enabled on cores which need mitigation. 3037 0: force disabled 3038 1: force enabled 3039 3040 kunit.enable= [KUNIT] Enable executing KUnit tests. Requires 3041 CONFIG_KUNIT to be set to be fully enabled. The 3042 default value can be overridden via 3043 KUNIT_DEFAULT_ENABLED. 3044 Default is 1 (enabled) 3045 3046 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs. 3047 Default is 0 (don't ignore, but inject #GP) 3048 3049 kvm.eager_page_split= 3050 [KVM,X86] Controls whether or not KVM will try to 3051 proactively split all huge pages during dirty logging. 3052 Eager page splitting reduces interruptions to vCPU 3053 execution by eliminating the write-protection faults 3054 and MMU lock contention that would otherwise be 3055 required to split huge pages lazily. 3056 3057 VM workloads that rarely perform writes or that write 3058 only to a small region of VM memory may benefit from 3059 disabling eager page splitting to allow huge pages to 3060 still be used for reads. 3061 3062 The behavior of eager page splitting depends on whether 3063 KVM_DIRTY_LOG_INITIALLY_SET is enabled or disabled. If 3064 disabled, all huge pages in a memslot will be eagerly 3065 split when dirty logging is enabled on that memslot. If 3066 enabled, eager page splitting will be performed during 3067 the KVM_CLEAR_DIRTY ioctl, and only for the pages being 3068 cleared. 3069 3070 Eager page splitting is only supported when kvm.tdp_mmu=Y. 3071 3072 Default is Y (on). 3073 3074 kvm.enable_pmu=[KVM,X86] 3075 If enabled, KVM will virtualize PMU functionality based 3076 on the virtual CPU model defined by userspace. This 3077 can be overridden on a per-VM basis via 3078 KVM_CAP_PMU_CAPABILITY. 3079 3080 If disabled, KVM will not virtualize PMU functionality, 3081 e.g. MSRs, PMCs, PMIs, etc., even if userspace defines 3082 a virtual CPU model that contains PMU assets. 3083 3084 Note, KVM's vPMU support implicitly requires running 3085 with an in-kernel local APIC, e.g. to deliver PMIs to 3086 the guest. Running without an in-kernel local APIC is 3087 not supported, though KVM will allow such a combination 3088 (with severely degraded functionality). 3089 3090 See also enable_mediated_pmu. 3091 3092 Default is Y (on). 3093 3094 kvm.enable_virt_at_load=[KVM,ARM64,LOONGARCH,MIPS,RISCV,X86] 3095 If enabled, KVM will enable virtualization in hardware 3096 when KVM is loaded, and disable virtualization when KVM 3097 is unloaded (if KVM is built as a module). 3098 3099 If disabled, KVM will dynamically enable and disable 3100 virtualization on-demand when creating and destroying 3101 VMs, i.e. on the 0=>1 and 1=>0 transitions of the 3102 number of VMs. 3103 3104 Enabling virtualization at module load avoids potential 3105 latency for creation of the 0=>1 VM, as KVM serializes 3106 virtualization enabling across all online CPUs. The 3107 "cost" of enabling virtualization when KVM is loaded, 3108 is that doing so may interfere with using out-of-tree 3109 hypervisors that want to "own" virtualization hardware. 3110 3111 kvm.enable_vmware_backdoor=[KVM] Support VMware backdoor PV interface. 3112 Default is false (don't support). 3113 3114 kvm.nx_huge_pages= 3115 [KVM] Controls the software workaround for the 3116 X86_BUG_ITLB_MULTIHIT bug. 3117 force : Always deploy workaround. 3118 off : Never deploy workaround. 3119 auto : Deploy workaround based on the presence of 3120 X86_BUG_ITLB_MULTIHIT. 3121 3122 Default is 'auto'. 3123 3124 If the software workaround is enabled for the host, 3125 guests do need not to enable it for nested guests. 3126 3127 kvm.nx_huge_pages_recovery_ratio= 3128 [KVM] Controls how many 4KiB pages are periodically zapped 3129 back to huge pages. 0 disables the recovery, otherwise if 3130 the value is N KVM will zap 1/Nth of the 4KiB pages every 3131 period (see below). The default is 60. 3132 3133 kvm.nx_huge_pages_recovery_period_ms= 3134 [KVM] Controls the time period at which KVM zaps 4KiB pages 3135 back to huge pages. If the value is a non-zero N, KVM will 3136 zap a portion (see ratio above) of the pages every N msecs. 3137 If the value is 0 (the default), KVM will pick a period based 3138 on the ratio, such that a page is zapped after 1 hour on average. 3139 3140 kvm-{amd,intel}.enable_mediated_pmu=[KVM,AMD,INTEL] 3141 If enabled, KVM will provide a mediated virtual PMU, 3142 instead of the default perf-based virtual PMU (if 3143 kvm.enable_pmu is true and PMU is enumerated via the 3144 virtual CPU model). 3145 3146 With a perf-based vPMU, KVM operates as a user of perf, 3147 i.e. emulates guest PMU counters using perf events. 3148 KVM-created perf events are managed by perf as regular 3149 (guest-only) events, e.g. are scheduled in/out, contend 3150 for hardware resources, etc. Using a perf-based vPMU 3151 allows guest and host usage of the PMU to co-exist, but 3152 incurs non-trivial overhead and can result in silently 3153 dropped guest events (due to resource contention). 3154 3155 With a mediated vPMU, hardware PMU state is context 3156 switched around the world switch to/from the guest. 3157 KVM mediates which events the guest can utilize, but 3158 gives the guest direct access to all other PMU assets 3159 when possible (KVM may intercept some accesses if the 3160 virtual CPU model provides a subset of hardware PMU 3161 functionality). Using a mediated vPMU significantly 3162 reduces PMU virtualization overhead and eliminates lost 3163 guest events, but is mutually exclusive with using perf 3164 to profile KVM guests and adds latency to most VM-Exits 3165 (to context switch PMU state). 3166 3167 Default is N (off). 3168 3169 kvm-amd.nested= [KVM,AMD] Control nested virtualization feature in 3170 KVM/SVM. Default is 1 (enabled). 3171 3172 kvm-amd.npt= [KVM,AMD] Control KVM's use of Nested Page Tables, 3173 a.k.a. Two-Dimensional Page Tables. Default is 1 3174 (enabled). Disable by KVM if hardware lacks support 3175 for NPT. 3176 3177 kvm-amd.ciphertext_hiding_asids= 3178 [KVM,AMD] Ciphertext hiding prevents disallowed accesses 3179 to SNP private memory from reading ciphertext. Instead, 3180 reads will see constant default values (0xff). 3181 3182 If ciphertext hiding is enabled, the joint SEV-ES and 3183 SEV-SNP ASID space is partitioned into separate SEV-ES 3184 and SEV-SNP ASID ranges, with the SEV-SNP range being 3185 [1..max_snp_asid] and the SEV-ES range being 3186 (max_snp_asid..min_sev_asid), where min_sev_asid is 3187 enumerated by CPUID.0x.8000_001F[EDX]. 3188 3189 A non-zero value enables SEV-SNP ciphertext hiding and 3190 adjusts the ASID ranges for SEV-ES and SEV-SNP guests. 3191 KVM caps the number of SEV-SNP ASIDs at the maximum 3192 possible value, e.g. specifying -1u will assign all 3193 joint SEV-ES and SEV-SNP ASIDs to SEV-SNP. Note, 3194 assigning all joint ASIDs to SEV-SNP, i.e. configuring 3195 max_snp_asid == min_sev_asid-1, will effectively make 3196 SEV-ES unusable. 3197 3198 kvm-arm.mode= 3199 [KVM,ARM,EARLY] Select one of KVM/arm64's modes of 3200 operation. 3201 3202 none: Forcefully disable KVM. 3203 3204 nvhe: Standard nVHE-based mode, without support for 3205 protected guests. 3206 3207 protected: Mode with support for guests whose state is 3208 kept private from the host, using VHE or 3209 nVHE depending on HW support. 3210 3211 nested: VHE-based mode with support for nested 3212 virtualization. Requires at least ARMv8.4 3213 hardware (with FEAT_NV2). 3214 3215 Defaults to VHE/nVHE based on hardware support. Setting 3216 mode to "protected" will disable kexec and hibernation 3217 for the host. To force nVHE on VHE hardware, add 3218 "arm64_sw.hvhe=0 id_aa64mmfr1.vh=0" to the 3219 command-line. 3220 "nested" and "protected" are experimental and should be 3221 used with extreme caution. 3222 3223 kvm-arm.vgic_v3_group0_trap= 3224 [KVM,ARM,EARLY] Trap guest accesses to GICv3 group-0 3225 system registers 3226 3227 kvm-arm.vgic_v3_group1_trap= 3228 [KVM,ARM,EARLY] Trap guest accesses to GICv3 group-1 3229 system registers 3230 3231 kvm-arm.vgic_v3_common_trap= 3232 [KVM,ARM,EARLY] Trap guest accesses to GICv3 common 3233 system registers 3234 3235 kvm-arm.vgic_v4_enable= 3236 [KVM,ARM,EARLY] Allow use of GICv4 for direct 3237 injection of LPIs. 3238 3239 kvm-arm.wfe_trap_policy= 3240 [KVM,ARM] Control when to set WFE instruction trap for 3241 KVM VMs. Traps are allowed but not guaranteed by the 3242 CPU architecture. 3243 3244 trap: set WFE instruction trap 3245 3246 notrap: clear WFE instruction trap 3247 3248 kvm-arm.wfi_trap_policy= 3249 [KVM,ARM] Control when to set WFI instruction trap for 3250 KVM VMs. Traps are allowed but not guaranteed by the 3251 CPU architecture. 3252 3253 trap: set WFI instruction trap 3254 3255 notrap: clear WFI instruction trap 3256 3257 kvm_cma_resv_ratio=n [PPC,EARLY] 3258 Reserves given percentage from system memory area for 3259 contiguous memory allocation for KVM hash pagetable 3260 allocation. 3261 By default it reserves 5% of total system memory. 3262 Format: <integer> 3263 Default: 5 3264 3265 kvm-intel.ept= [KVM,Intel] Control KVM's use of Extended Page Tables, 3266 a.k.a. Two-Dimensional Page Tables. Default is 1 3267 (enabled). Disable by KVM if hardware lacks support 3268 for EPT. 3269 3270 kvm-intel.emulate_invalid_guest_state= 3271 [KVM,Intel] Control whether to emulate invalid guest 3272 state. Ignored if kvm-intel.enable_unrestricted_guest=1, 3273 as guest state is never invalid for unrestricted 3274 guests. This param doesn't apply to nested guests (L2), 3275 as KVM never emulates invalid L2 guest state. 3276 Default is 1 (enabled). 3277 3278 kvm-intel.flexpriority= 3279 [KVM,Intel] Control KVM's use of FlexPriority feature 3280 (TPR shadow). Default is 1 (enabled). Disable by KVM if 3281 hardware lacks support for it. 3282 3283 kvm-intel.nested= 3284 [KVM,Intel] Control nested virtualization feature in 3285 KVM/VMX. Default is 1 (enabled). 3286 3287 kvm-intel.unrestricted_guest= 3288 [KVM,Intel] Control KVM's use of unrestricted guest 3289 feature (virtualized real and unpaged mode). Default 3290 is 1 (enabled). Disable by KVM if EPT is disabled or 3291 hardware lacks support for it. 3292 3293 kvm-intel.vmentry_l1d_flush=[KVM,Intel] Mitigation for L1 Terminal Fault 3294 CVE-2018-3620. 3295 3296 Valid arguments: never, cond, always 3297 3298 always: L1D cache flush on every VMENTER. 3299 cond: Flush L1D on VMENTER only when the code between 3300 VMEXIT and VMENTER can leak host memory. 3301 never: Disables the mitigation 3302 3303 Default is cond (do L1 cache flush in specific instances) 3304 3305 kvm-intel.vpid= [KVM,Intel] Control KVM's use of Virtual Processor 3306 Identification feature (tagged TLBs). Default is 1 3307 (enabled). Disable by KVM if hardware lacks support 3308 for it. 3309 3310 l1d_flush= [X86,INTEL,EARLY] 3311 Control mitigation for L1D based snooping vulnerability. 3312 3313 Certain CPUs are vulnerable to an exploit against CPU 3314 internal buffers which can forward information to a 3315 disclosure gadget under certain conditions. 3316 3317 In vulnerable processors, the speculatively 3318 forwarded data can be used in a cache side channel 3319 attack, to access data to which the attacker does 3320 not have direct access. 3321 3322 This parameter controls the mitigation. The 3323 options are: 3324 3325 on - enable the interface for the mitigation 3326 3327 l1tf= [X86,EARLY] Control mitigation of the L1TF vulnerability on 3328 affected CPUs 3329 3330 The kernel PTE inversion protection is unconditionally 3331 enabled and cannot be disabled. 3332 3333 full 3334 Provides all available mitigations for the 3335 L1TF vulnerability. Disables SMT and 3336 enables all mitigations in the 3337 hypervisors, i.e. unconditional L1D flush. 3338 3339 SMT control and L1D flush control via the 3340 sysfs interface is still possible after 3341 boot. Hypervisors will issue a warning 3342 when the first VM is started in a 3343 potentially insecure configuration, 3344 i.e. SMT enabled or L1D flush disabled. 3345 3346 full,force 3347 Same as 'full', but disables SMT and L1D 3348 flush runtime control. Implies the 3349 'nosmt=force' command line option. 3350 (i.e. sysfs control of SMT is disabled.) 3351 3352 flush 3353 Leaves SMT enabled and enables the default 3354 hypervisor mitigation, i.e. conditional 3355 L1D flush. 3356 3357 SMT control and L1D flush control via the 3358 sysfs interface is still possible after 3359 boot. Hypervisors will issue a warning 3360 when the first VM is started in a 3361 potentially insecure configuration, 3362 i.e. SMT enabled or L1D flush disabled. 3363 3364 flush,nosmt 3365 3366 Disables SMT and enables the default 3367 hypervisor mitigation. 3368 3369 SMT control and L1D flush control via the 3370 sysfs interface is still possible after 3371 boot. Hypervisors will issue a warning 3372 when the first VM is started in a 3373 potentially insecure configuration, 3374 i.e. SMT enabled or L1D flush disabled. 3375 3376 flush,nowarn 3377 Same as 'flush', but hypervisors will not 3378 warn when a VM is started in a potentially 3379 insecure configuration. 3380 3381 off 3382 Disables hypervisor mitigations and doesn't 3383 emit any warnings. 3384 It also drops the swap size and available 3385 RAM limit restriction on both hypervisor and 3386 bare metal. 3387 3388 Default is 'flush'. 3389 3390 For details see: Documentation/admin-guide/hw-vuln/l1tf.rst 3391 3392 l2cr= [PPC] 3393 3394 l3cr= [PPC] 3395 3396 lapic [X86-32,APIC,EARLY] Enable the local APIC even if BIOS 3397 disabled it. 3398 3399 lapic= [X86,APIC] Do not use TSC deadline 3400 value for LAPIC timer one-shot implementation. Default 3401 back to the programmable timer unit in the LAPIC. 3402 Format: notscdeadline 3403 3404 lapic_timer_c2_ok [X86,APIC,EARLY] trust the local apic timer 3405 in C2 power state. 3406 3407 libata.dma= [LIBATA] DMA control 3408 libata.dma=0 Disable all PATA and SATA DMA 3409 libata.dma=1 PATA and SATA Disk DMA only 3410 libata.dma=2 ATAPI (CDROM) DMA only 3411 libata.dma=4 Compact Flash DMA only 3412 Combinations also work, so libata.dma=3 enables DMA 3413 for disks and CDROMs, but not CFs. 3414 3415 libata.ignore_hpa= [LIBATA] Ignore HPA limit 3416 libata.ignore_hpa=0 keep BIOS limits (default) 3417 libata.ignore_hpa=1 ignore limits, using full disk 3418 3419 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume 3420 when set. 3421 Format: <int> 3422 3423 libata.force= [LIBATA] Force configurations. The format is a comma- 3424 separated list of "[ID:]VAL" where ID is PORT[.DEVICE]. 3425 PORT and DEVICE are decimal numbers matching port, link 3426 or device. Basically, it matches the ATA ID string 3427 printed on console by libata. If the whole ID part is 3428 omitted, the last PORT and DEVICE values are used. If 3429 ID hasn't been specified yet, the configuration applies 3430 to all ports, links and devices. 3431 3432 If only DEVICE is omitted, the parameter applies to 3433 the port and all links and devices behind it. DEVICE 3434 number of 0 either selects the first device or the 3435 first fan-out link behind PMP device. It does not 3436 select the host link. DEVICE number of 15 selects the 3437 host link and device attached to it. 3438 3439 The VAL specifies the configuration to force. As long 3440 as there is no ambiguity, shortcut notation is allowed. 3441 For example, both 1.5 and 1.5G would work for 1.5Gbps. 3442 The following configurations can be forced. 3443 3444 * Cable type: 40c, 80c, short40c, unk, ign or sata. 3445 Any ID with matching PORT is used. 3446 3447 * SATA link speed limit: 1.5Gbps or 3.0Gbps. 3448 3449 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7]. 3450 udma[/][16,25,33,44,66,100,133] notation is also 3451 allowed. 3452 3453 * nohrst, nosrst, norst: suppress hard, soft and both 3454 resets. 3455 3456 * rstonce: only attempt one reset during hot-unplug 3457 link recovery. 3458 3459 * [no]dbdelay: Enable or disable the extra 200ms delay 3460 before debouncing a link PHY and device presence 3461 detection. 3462 3463 * [no]ncq: Turn on or off NCQ. 3464 3465 * [no]ncqtrim: Enable or disable queued DSM TRIM. 3466 3467 * [no]ncqati: Enable or disable NCQ trim on ATI chipset. 3468 3469 * [no]trim: Enable or disable (unqueued) TRIM. 3470 3471 * trim_zero: Indicate that TRIM command zeroes data. 3472 3473 * max_trim_128m: Set 128M maximum trim size limit. 3474 3475 * [no]dma: Turn on or off DMA transfers. 3476 3477 * atapi_dmadir: Enable ATAPI DMADIR bridge support. 3478 3479 * atapi_mod16_dma: Enable the use of ATAPI DMA for 3480 commands that are not a multiple of 16 bytes. 3481 3482 * [no]dmalog: Enable or disable the use of the 3483 READ LOG DMA EXT command to access logs. 3484 3485 * [no]iddevlog: Enable or disable access to the 3486 identify device data log. 3487 3488 * [no]logdir: Enable or disable access to the general 3489 purpose log directory. 3490 3491 * max_sec=<sectors>: Set the transfer size limit, in 3492 number of 512-byte sectors, to the value specified in 3493 <sectors>. The value specified in <sectors> has to be 3494 a non-zero positive integer. 3495 3496 * max_sec_128: Set transfer size limit to 128 sectors. 3497 3498 * max_sec_1024: Set or clear transfer size limit to 3499 1024 sectors. 3500 3501 * max_sec_lba48: Set or clear transfer size limit to 3502 65535 sectors. 3503 3504 * external: Mark port as external (hotplug-capable). 3505 3506 * [no]lpm: Enable or disable link power management. 3507 3508 * [no]setxfer: Indicate if transfer speed mode setting 3509 should be skipped. 3510 3511 * [no]fua: Disable or enable FUA (Force Unit Access) 3512 support for devices supporting this feature. 3513 3514 * dump_id: Dump IDENTIFY data. 3515 3516 * disable: Disable this device. 3517 3518 If there are multiple matching configurations changing 3519 the same attribute, the last one is used. 3520 3521 liveupdate= [KNL,EARLY] 3522 Format: <bool> 3523 Enable Live Update Orchestrator (LUO). 3524 Default: off. 3525 3526 lockd.nlm_grace_period=P [NFS] Assign grace period. 3527 Format: <integer> 3528 3529 lockd.nlm_tcpport=N [NFS] Assign TCP port. 3530 Format: <integer> 3531 3532 lockd.nlm_timeout=T [NFS] Assign timeout value. 3533 Format: <integer> 3534 3535 lockd.nlm_udpport=M [NFS] Assign UDP port. 3536 Format: <integer> 3537 3538 lockdown= [SECURITY,EARLY] 3539 { integrity | confidentiality } 3540 Enable the kernel lockdown feature. If set to 3541 integrity, kernel features that allow userland to 3542 modify the running kernel are disabled. If set to 3543 confidentiality, kernel features that allow userland 3544 to extract confidential information from the kernel 3545 are also disabled. 3546 3547 locktorture.acq_writer_lim= [KNL] 3548 Set the time limit in jiffies for a lock 3549 acquisition. Acquisitions exceeding this limit 3550 will result in a splat once they do complete. 3551 3552 locktorture.bind_readers= [KNL] 3553 Specify the list of CPUs to which the readers are 3554 to be bound. 3555 3556 locktorture.bind_writers= [KNL] 3557 Specify the list of CPUs to which the writers are 3558 to be bound. 3559 3560 locktorture.call_rcu_chains= [KNL] 3561 Specify the number of self-propagating call_rcu() 3562 chains to set up. These are used to ensure that 3563 there is a high probability of an RCU grace period 3564 in progress at any given time. Defaults to 0, 3565 which disables these call_rcu() chains. 3566 3567 locktorture.long_hold= [KNL] 3568 Specify the duration in milliseconds for the 3569 occasional long-duration lock hold time. Defaults 3570 to 100 milliseconds. Select 0 to disable. 3571 3572 locktorture.nested_locks= [KNL] 3573 Specify the maximum lock nesting depth that 3574 locktorture is to exercise, up to a limit of 8 3575 (MAX_NESTED_LOCKS). Specify zero to disable. 3576 Note that this parameter is ineffective on types 3577 of locks that do not support nested acquisition. 3578 3579 locktorture.nreaders_stress= [KNL] 3580 Set the number of locking read-acquisition kthreads. 3581 Defaults to being automatically set based on the 3582 number of online CPUs. 3583 3584 locktorture.nwriters_stress= [KNL] 3585 Set the number of locking write-acquisition kthreads. 3586 3587 locktorture.onoff_holdoff= [KNL] 3588 Set time (s) after boot for CPU-hotplug testing. 3589 3590 locktorture.onoff_interval= [KNL] 3591 Set time (s) between CPU-hotplug operations, or 3592 zero to disable CPU-hotplug testing. 3593 3594 locktorture.rt_boost= [KNL] 3595 Do periodic testing of real-time lock priority 3596 boosting. Select 0 to disable, 1 to boost 3597 only rt_mutex, and 2 to boost unconditionally. 3598 Defaults to 2, which might seem to be an 3599 odd choice, but which should be harmless for 3600 non-real-time spinlocks, due to their disabling 3601 of preemption. Note that non-realtime mutexes 3602 disable boosting. 3603 3604 locktorture.rt_boost_factor= [KNL] 3605 Number that determines how often and for how 3606 long priority boosting is exercised. This is 3607 scaled down by the number of writers, so that the 3608 number of boosts per unit time remains roughly 3609 constant as the number of writers increases. 3610 On the other hand, the duration of each boost 3611 increases with the number of writers. 3612 3613 locktorture.shuffle_interval= [KNL] 3614 Set task-shuffle interval (jiffies). Shuffling 3615 tasks allows some CPUs to go into dyntick-idle 3616 mode during the locktorture test. 3617 3618 locktorture.shutdown_secs= [KNL] 3619 Set time (s) after boot system shutdown. This 3620 is useful for hands-off automated testing. 3621 3622 locktorture.stat_interval= [KNL] 3623 Time (s) between statistics printk()s. 3624 3625 locktorture.stutter= [KNL] 3626 Time (s) to stutter testing, for example, 3627 specifying five seconds causes the test to run for 3628 five seconds, wait for five seconds, and so on. 3629 This tests the locking primitive's ability to 3630 transition abruptly to and from idle. 3631 3632 locktorture.torture_type= [KNL] 3633 Specify the locking implementation to test. 3634 3635 locktorture.verbose= [KNL] 3636 Enable additional printk() statements. 3637 3638 locktorture.writer_fifo= [KNL] 3639 Run the write-side locktorture kthreads at 3640 sched_set_fifo() real-time priority. 3641 3642 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver 3643 Format: <irq> 3644 3645 loglevel= [KNL,EARLY] 3646 All Kernel Messages with a loglevel smaller than the 3647 console loglevel will be printed to the console. It can 3648 also be changed with klogd or other programs. The 3649 loglevels are defined as follows: 3650 3651 0 (KERN_EMERG) system is unusable 3652 1 (KERN_ALERT) action must be taken immediately 3653 2 (KERN_CRIT) critical conditions 3654 3 (KERN_ERR) error conditions 3655 4 (KERN_WARNING) warning conditions 3656 5 (KERN_NOTICE) normal but significant condition 3657 6 (KERN_INFO) informational 3658 7 (KERN_DEBUG) debug-level messages 3659 3660 log_buf_len=n[KMG] [KNL,EARLY] 3661 Sets the size of the printk ring buffer, in bytes. 3662 n must be a power of two and greater than the 3663 minimal size. The minimal size is defined by 3664 LOG_BUF_SHIFT kernel config parameter. There 3665 is also CONFIG_LOG_CPU_MAX_BUF_SHIFT config 3666 parameter that allows to increase the default size 3667 depending on the number of CPUs. See init/Kconfig 3668 for more details. 3669 3670 logo.nologo [FB] Disables display of the built-in Linux logo. 3671 This may be used to provide more screen space for 3672 kernel log messages and is useful when debugging 3673 kernel boot problems. 3674 3675 lp=0 [LP] Specify parallel ports to use, e.g, 3676 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses 3677 lp=reset first parallel port). 'lp=0' disables the 3678 lp=auto printer driver. 'lp=reset' (which can be 3679 specified in addition to the ports) causes 3680 attached printers to be reset. Using 3681 lp=port1,port2,... specifies the parallel ports 3682 to associate lp devices with, starting with 3683 lp0. A port specification may be 'none' to skip 3684 that lp device, or a parport name such as 3685 'parport0'. Specifying 'lp=auto' instead of a 3686 port specification list means that device IDs 3687 from each port should be examined, to see if 3688 an IEEE 1284-compliant printer is attached; if 3689 so, the driver will manage that printer. 3690 See also header of drivers/char/lp.c. 3691 3692 lpj=n [KNL] 3693 Sets loops_per_jiffy to given constant, thus avoiding 3694 time-consuming boot-time autodetection (up to 250 ms per 3695 CPU). 0 enables autodetection (default). To determine 3696 the correct value for your kernel, boot with normal 3697 autodetection and see what value is printed. Note that 3698 on SMP systems the preset will be applied to all CPUs, 3699 which is likely to cause problems if your CPUs need 3700 significantly divergent settings. An incorrect value 3701 will cause delays in the kernel to be wrong, leading to 3702 unpredictable I/O errors and other breakage. Although 3703 unlikely, in the extreme case this might damage your 3704 hardware. 3705 3706 lsm.debug [SECURITY] Enable LSM initialization debugging output. 3707 3708 lsm=lsm1,...,lsmN 3709 [SECURITY] Choose order of LSM initialization. This 3710 overrides CONFIG_LSM, and the "security=" parameter. 3711 3712 machtype= [Loongson] Share the same kernel image file between 3713 different yeeloong laptops. 3714 Example: machtype=lemote-yeeloong-2f-7inch 3715 3716 maxcpus= [SMP,EARLY] Maximum number of processors that an SMP kernel 3717 will bring up during bootup. maxcpus=n : n >= 0 limits 3718 the kernel to bring up 'n' processors. Surely after 3719 bootup you can bring up the other plugged cpu by executing 3720 "echo 1 > /sys/devices/system/cpu/cpuX/online". So maxcpus 3721 only takes effect during system bootup. 3722 While n=0 is a special case, it is equivalent to "nosmp", 3723 which also disables the IO APIC. 3724 3725 max_loop= [LOOP] The number of loop block devices that get 3726 (loop.max_loop) unconditionally pre-created at init time. The default 3727 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead 3728 of statically allocating a predefined number, loop 3729 devices can be requested on-demand with the 3730 /dev/loop-control interface. 3731 3732 mce= [X86-{32,64}] 3733 3734 Please see Documentation/arch/x86/x86_64/machinecheck.rst for sysfs runtime tunables. 3735 3736 off 3737 disable machine check 3738 3739 no_cmci 3740 disable CMCI(Corrected Machine Check Interrupt) that 3741 Intel processor supports. Usually this disablement is 3742 not recommended, but it might be handy if your 3743 hardware is misbehaving. 3744 3745 Note that you'll get more problems without CMCI than 3746 with due to the shared banks, i.e. you might get 3747 duplicated error logs. 3748 3749 dont_log_ce 3750 don't make logs for corrected errors. All events 3751 reported as corrected are silently cleared by OS. This 3752 option will be useful if you have no interest in any 3753 of corrected errors. 3754 3755 ignore_ce 3756 disable features for corrected errors, e.g. 3757 polling timer and CMCI. All events reported as 3758 corrected are not cleared by OS and remained in its 3759 error banks. 3760 3761 Usually this disablement is not recommended, however 3762 if there is an agent checking/clearing corrected 3763 errors (e.g. BIOS or hardware monitoring 3764 applications), conflicting with OS's error handling, 3765 and you cannot deactivate the agent, then this option 3766 will be a help. 3767 3768 no_lmce 3769 do not opt-in to Local MCE delivery. Use legacy method 3770 to broadcast MCEs. 3771 3772 bootlog 3773 enable logging of machine checks left over from 3774 booting. Disabled by default on AMD Fam10h and older 3775 because some BIOS leave bogus ones. 3776 3777 If your BIOS doesn't do that it's a good idea to 3778 enable though to make sure you log even machine check 3779 events that result in a reboot. On Intel systems it is 3780 enabled by default. 3781 3782 nobootlog 3783 disable boot machine check logging. 3784 3785 monarchtimeout (number) 3786 sets the time in us to wait for other CPUs on machine 3787 checks. 0 to disable. 3788 3789 bios_cmci_threshold 3790 don't overwrite the bios-set CMCI threshold. This boot 3791 option prevents Linux from overwriting the CMCI 3792 threshold set by the bios. Without this option, Linux 3793 always sets the CMCI threshold to 1. Enabling this may 3794 make memory predictive failure analysis less effective 3795 if the bios sets thresholds for memory errors since we 3796 will not see details for all errors. 3797 3798 recovery 3799 force-enable recoverable machine check code paths 3800 3801 Everything else is in sysfs now. 3802 3803 3804 md= [HW] RAID subsystems devices and level 3805 See Documentation/admin-guide/md.rst. 3806 3807 mds= [X86,INTEL,EARLY] 3808 Control mitigation for the Micro-architectural Data 3809 Sampling (MDS) vulnerability. 3810 3811 Certain CPUs are vulnerable to an exploit against CPU 3812 internal buffers which can forward information to a 3813 disclosure gadget under certain conditions. 3814 3815 In vulnerable processors, the speculatively 3816 forwarded data can be used in a cache side channel 3817 attack, to access data to which the attacker does 3818 not have direct access. 3819 3820 This parameter controls the MDS mitigation. The 3821 options are: 3822 3823 full - Enable MDS mitigation on vulnerable CPUs 3824 full,nosmt - Enable MDS mitigation and disable 3825 SMT on vulnerable CPUs 3826 off - Unconditionally disable MDS mitigation 3827 3828 On TAA-affected machines, mds=off can be prevented by 3829 an active TAA mitigation as both vulnerabilities are 3830 mitigated with the same mechanism so in order to disable 3831 this mitigation, you need to specify tsx_async_abort=off 3832 too. 3833 3834 Not specifying this option is equivalent to 3835 mds=full. 3836 3837 For details see: Documentation/admin-guide/hw-vuln/mds.rst 3838 3839 mem=nn[KMG] [HEXAGON,EARLY] Set the memory size. 3840 Must be specified, otherwise memory size will be 0. 3841 3842 mem=nn[KMG] [KNL,BOOT,EARLY] Force usage of a specific amount 3843 of memory Amount of memory to be used in cases 3844 as follows: 3845 3846 1 for test; 3847 2 when the kernel is not able to see the whole system memory; 3848 3 memory that lies after 'mem=' boundary is excluded from 3849 the hypervisor, then assigned to KVM guests. 3850 4 to limit the memory available for kdump kernel. 3851 3852 [ARC,MICROBLAZE] - the limit applies only to low memory, 3853 high memory is not affected. 3854 3855 [ARM64] - only limits memory covered by the linear 3856 mapping. The NOMAP regions are not affected. 3857 3858 [X86] Work as limiting max address. Use together 3859 with memmap= to avoid physical address space collisions. 3860 Without memmap= PCI devices could be placed at addresses 3861 belonging to unused RAM. 3862 3863 Note that this only takes effects during boot time since 3864 in above case 3, memory may need be hot added after boot 3865 if system memory of hypervisor is not sufficient. 3866 3867 mem=nn[KMG]@ss[KMG] 3868 [ARM,MIPS,EARLY] - override the memory layout 3869 reported by firmware. 3870 Define a memory region of size nn[KMG] starting at 3871 ss[KMG]. 3872 Multiple different regions can be specified with 3873 multiple mem= parameters on the command line. 3874 3875 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel 3876 memory. 3877 3878 memblock=debug [KNL,EARLY] Enable memblock debug messages. 3879 3880 memchunk=nn[KMG] 3881 [KNL,SH] Allow user to override the default size for 3882 per-device physically contiguous DMA buffers. 3883 3884 memhp_default_state=online/offline/online_kernel/online_movable 3885 [KNL] Set the initial state for the memory hotplug 3886 onlining policy. If not specified, the default value is 3887 set according to the 3888 CONFIG_MHP_DEFAULT_ONLINE_TYPE kernel config 3889 options. 3890 See Documentation/admin-guide/mm/memory-hotplug.rst. 3891 3892 memmap=exactmap [KNL,X86,EARLY] Enable setting of an exact 3893 E820 memory map, as specified by the user. 3894 Such memmap=exactmap lines can be constructed based on 3895 BIOS output or other requirements. See the memmap=nn@ss 3896 option description. 3897 3898 memmap=nn[KMG]@ss[KMG] 3899 [KNL, X86,MIPS,XTENSA,EARLY] Force usage of a specific region of memory. 3900 Region of memory to be used is from ss to ss+nn. 3901 If @ss[KMG] is omitted, it is equivalent to mem=nn[KMG], 3902 which limits max address to nn[KMG]. 3903 Multiple different regions can be specified, 3904 comma delimited. 3905 Example: 3906 memmap=100M@2G,100M#3G,1G!1024G 3907 3908 memmap=nn[KMG]#ss[KMG] 3909 [KNL,ACPI,EARLY] Mark specific memory as ACPI data. 3910 Region of memory to be marked is from ss to ss+nn. 3911 3912 memmap=nn[KMG]$ss[KMG] 3913 [KNL,ACPI,EARLY] Mark specific memory as reserved. 3914 Region of memory to be reserved is from ss to ss+nn. 3915 Example: Exclude memory from 0x18690000-0x1869ffff 3916 memmap=64K$0x18690000 3917 or 3918 memmap=0x10000$0x18690000 3919 Some bootloaders may need an escape character before '$', 3920 like Grub2, otherwise '$' and the following number 3921 will be eaten. 3922 3923 memmap=nn[KMG]!ss[KMG,EARLY] 3924 [KNL,X86] Mark specific memory as protected. 3925 Region of memory to be used, from ss to ss+nn. 3926 The memory region may be marked as e820 type 12 (0xc) 3927 and is NVDIMM or ADR memory. 3928 3929 memmap=<size>%<offset>-<oldtype>+<newtype> 3930 [KNL,ACPI,EARLY] Convert memory within the specified region 3931 from <oldtype> to <newtype>. If "-<oldtype>" is left 3932 out, the whole region will be marked as <newtype>, 3933 even if previously unavailable. If "+<newtype>" is left 3934 out, matching memory will be removed. Types are 3935 specified as e820 types, e.g., 1 = RAM, 2 = reserved, 3936 3 = ACPI, 12 = PRAM. 3937 3938 memory_corruption_check=0/1 [X86,EARLY] 3939 Some BIOSes seem to corrupt the first 64k of 3940 memory when doing things like suspend/resume. 3941 Setting this option will scan the memory 3942 looking for corruption. Enabling this will 3943 both detect corruption and prevent the kernel 3944 from using the memory being corrupted. 3945 However, it's intended as a diagnostic tool; if 3946 repeatable BIOS-originated corruption always 3947 affects the same memory, you can use memmap= 3948 to prevent the kernel from using that memory. 3949 3950 memory_corruption_check_size=size [X86,EARLY] 3951 By default it checks for corruption in the low 3952 64k, making this memory unavailable for normal 3953 use. Use this parameter to scan for 3954 corruption in more or less memory. 3955 3956 memory_corruption_check_period=seconds [X86,EARLY] 3957 By default it checks for corruption every 60 3958 seconds. Use this parameter to check at some 3959 other rate. 0 disables periodic checking. 3960 3961 memory_hotplug.memmap_on_memory 3962 [KNL,X86,ARM] Boolean flag to enable this feature. 3963 Format: {on | off (default)} 3964 When enabled, runtime hotplugged memory will 3965 allocate its internal metadata (struct pages, 3966 those vmemmap pages cannot be optimized even 3967 if hugetlb_free_vmemmap is enabled) from the 3968 hotadded memory which will allow to hotadd a 3969 lot of memory without requiring additional 3970 memory to do so. 3971 This feature is disabled by default because it 3972 has some implication on large (e.g. GB) 3973 allocations in some configurations (e.g. small 3974 memory blocks). 3975 The state of the flag can be read in 3976 /sys/module/memory_hotplug/parameters/memmap_on_memory. 3977 Note that even when enabled, there are a few cases where 3978 the feature is not effective. 3979 3980 memtest= [KNL,X86,ARM,M68K,PPC,RISCV,EARLY] Enable memtest 3981 Format: <integer> 3982 default : 0 <disable> 3983 Specifies the number of memtest passes to be 3984 performed. Each pass selects another test 3985 pattern from a given set of patterns. Memtest 3986 fills the memory with this pattern, validates 3987 memory contents and reserves bad memory 3988 regions that are detected. 3989 3990 mem_encrypt= [X86-64] AMD Secure Memory Encryption (SME) control 3991 Valid arguments: on, off 3992 Default: off 3993 mem_encrypt=on: Activate SME 3994 mem_encrypt=off: Do not activate SME 3995 3996 Refer to Documentation/virt/kvm/x86/amd-memory-encryption.rst 3997 for details on when memory encryption can be activated. 3998 3999 mem_sleep_default= [SUSPEND] Default system suspend mode: 4000 s2idle - Suspend-To-Idle 4001 shallow - Power-On Suspend or equivalent (if supported) 4002 deep - Suspend-To-RAM or equivalent (if supported) 4003 See Documentation/admin-guide/pm/sleep-states.rst. 4004 4005 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when 4006 the BIOS has incorrectly applied a workaround. TinyBIOS 4007 version 0.98 is known to be affected, 0.99 fixes the 4008 problem by letting the user disable the workaround. 4009 4010 mga= [HW,DRM] 4011 4012 microcode= [X86] Control the behavior of the microcode loader. 4013 Available options, comma separated: 4014 4015 base_rev=X - with <X> with format: <u32> 4016 Set the base microcode revision of each thread when in 4017 debug mode. 4018 4019 dis_ucode_ldr: disable the microcode loader 4020 4021 force_minrev: 4022 Enable or disable the microcode minimal revision 4023 enforcement for the runtime microcode loader. 4024 4025 mini2440= [ARM,HW,KNL] 4026 Format:[0..2][b][c][t] 4027 Default: "0tb" 4028 MINI2440 configuration specification: 4029 0 - The attached screen is the 3.5" TFT 4030 1 - The attached screen is the 7" TFT 4031 2 - The VGA Shield is attached (1024x768) 4032 Leaving out the screen size parameter will not load 4033 the TFT driver, and the framebuffer will be left 4034 unconfigured. 4035 b - Enable backlight. The TFT backlight pin will be 4036 linked to the kernel VESA blanking code and a GPIO 4037 LED. This parameter is not necessary when using the 4038 VGA shield. 4039 c - Enable the s3c camera interface. 4040 t - Reserved for enabling touchscreen support. The 4041 touchscreen support is not enabled in the mainstream 4042 kernel as of 2.6.30, a preliminary port can be found 4043 in the "bleeding edge" mini2440 support kernel at 4044 https://repo.or.cz/w/linux-2.6/mini2440.git 4045 4046 mitigations= 4047 [X86,PPC,S390,ARM64,EARLY] Control optional mitigations for 4048 CPU vulnerabilities. This is a set of curated, 4049 arch-independent options, each of which is an 4050 aggregation of existing arch-specific options. 4051 4052 Note, "mitigations" is supported if and only if the 4053 kernel was built with CPU_MITIGATIONS=y. 4054 4055 off 4056 Disable all optional CPU mitigations. This 4057 improves system performance, but it may also 4058 expose users to several CPU vulnerabilities. 4059 Equivalent to: if nokaslr then kpti=0 [ARM64] 4060 gather_data_sampling=off [X86] 4061 indirect_target_selection=off [X86] 4062 kvm.nx_huge_pages=off [X86] 4063 l1tf=off [X86] 4064 mds=off [X86] 4065 mmio_stale_data=off [X86] 4066 no_entry_flush [PPC] 4067 no_uaccess_flush [PPC] 4068 nobp=0 [S390] 4069 nopti [X86,PPC] 4070 nospectre_bhb [ARM64] 4071 nospectre_v1 [X86,PPC] 4072 nospectre_v2 [X86,PPC,S390,ARM64] 4073 reg_file_data_sampling=off [X86] 4074 retbleed=off [X86] 4075 spec_rstack_overflow=off [X86] 4076 spec_store_bypass_disable=off [X86,PPC] 4077 spectre_bhi=off [X86] 4078 spectre_v2_user=off [X86] 4079 srbds=off [X86,INTEL] 4080 ssbd=force-off [ARM64] 4081 tsa=off [X86,AMD] 4082 tsx_async_abort=off [X86] 4083 vmscape=off [X86] 4084 4085 Exceptions: 4086 This does not have any effect on 4087 kvm.nx_huge_pages when 4088 kvm.nx_huge_pages=force. 4089 4090 auto (default) 4091 Mitigate all CPU vulnerabilities, but leave SMT 4092 enabled, even if it's vulnerable. This is for 4093 users who don't want to be surprised by SMT 4094 getting disabled across kernel upgrades, or who 4095 have other ways of avoiding SMT-based attacks. 4096 Equivalent to: (default behavior) 4097 4098 auto,nosmt 4099 Mitigate all CPU vulnerabilities, disabling SMT 4100 if needed. This is for users who always want to 4101 be fully mitigated, even if it means losing SMT. 4102 Equivalent to: l1tf=flush,nosmt [X86] 4103 mds=full,nosmt [X86] 4104 tsx_async_abort=full,nosmt [X86] 4105 mmio_stale_data=full,nosmt [X86] 4106 retbleed=auto,nosmt [X86] 4107 4108 [X86] After one of the above options, additionally 4109 supports attack-vector based controls as documented in 4110 Documentation/admin-guide/hw-vuln/attack_vector_controls.rst 4111 4112 mminit_loglevel= 4113 [KNL,EARLY] When CONFIG_DEBUG_MEMORY_INIT is set, this 4114 parameter allows control of the logging verbosity for 4115 the additional memory initialisation checks. A value 4116 of 0 disables mminit logging and a level of 4 will 4117 log everything. Information is printed at KERN_DEBUG 4118 so loglevel=8 may also need to be specified. 4119 4120 mmio_stale_data= 4121 [X86,INTEL,EARLY] Control mitigation for the Processor 4122 MMIO Stale Data vulnerabilities. 4123 4124 Processor MMIO Stale Data is a class of 4125 vulnerabilities that may expose data after an MMIO 4126 operation. Exposed data could originate or end in 4127 the same CPU buffers as affected by MDS and TAA. 4128 Therefore, similar to MDS and TAA, the mitigation 4129 is to clear the affected CPU buffers. 4130 4131 This parameter controls the mitigation. The 4132 options are: 4133 4134 full - Enable mitigation on vulnerable CPUs 4135 4136 full,nosmt - Enable mitigation and disable SMT on 4137 vulnerable CPUs. 4138 4139 off - Unconditionally disable mitigation 4140 4141 On MDS or TAA affected machines, 4142 mmio_stale_data=off can be prevented by an active 4143 MDS or TAA mitigation as these vulnerabilities are 4144 mitigated with the same mechanism so in order to 4145 disable this mitigation, you need to specify 4146 mds=off and tsx_async_abort=off too. 4147 4148 Not specifying this option is equivalent to 4149 mmio_stale_data=full. 4150 4151 For details see: 4152 Documentation/admin-guide/hw-vuln/processor_mmio_stale_data.rst 4153 4154 <module>.async_probe[=<bool>] [KNL] 4155 If no <bool> value is specified or if the value 4156 specified is not a valid <bool>, enable asynchronous 4157 probe on this module. Otherwise, enable/disable 4158 asynchronous probe on this module as indicated by the 4159 <bool> value. See also: module.async_probe 4160 4161 module.async_probe=<bool> 4162 [KNL] When set to true, modules will use async probing 4163 by default. To enable/disable async probing for a 4164 specific module, use the module specific control that 4165 is documented under <module>.async_probe. When both 4166 module.async_probe and <module>.async_probe are 4167 specified, <module>.async_probe takes precedence for 4168 the specific module. 4169 4170 module.enable_dups_trace 4171 [KNL] When CONFIG_MODULE_DEBUG_AUTOLOAD_DUPS is set, 4172 this means that duplicate request_module() calls will 4173 trigger a WARN_ON() instead of a pr_warn(). Note that 4174 if MODULE_DEBUG_AUTOLOAD_DUPS_TRACE is set, WARN_ON()s 4175 will always be issued and this option does nothing. 4176 module.sig_enforce 4177 [KNL] When CONFIG_MODULE_SIG is set, this means that 4178 modules without (valid) signatures will fail to load. 4179 Note that if CONFIG_MODULE_SIG_FORCE is set, that 4180 is always true, so this option does nothing. 4181 4182 module_blacklist= [KNL] Do not load a comma-separated list of 4183 modules. Useful for debugging problem modules. 4184 4185 mousedev.tap_time= 4186 [MOUSE] Maximum time between finger touching and 4187 leaving touchpad surface for touch to be considered 4188 a tap and be reported as a left button click (for 4189 touchpads working in absolute mode only). 4190 Format: <msecs> 4191 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices 4192 reporting absolute coordinates, such as tablets 4193 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices 4194 reporting absolute coordinates, such as tablets 4195 4196 movablecore= [KNL,X86,PPC,EARLY] 4197 Format: nn[KMGTPE] | nn% 4198 This parameter is the complement to kernelcore=, it 4199 specifies the amount of memory used for migratable 4200 allocations. If both kernelcore and movablecore is 4201 specified, then kernelcore will be at *least* the 4202 specified value but may be more. If movablecore on its 4203 own is specified, the administrator must be careful 4204 that the amount of memory usable for all allocations 4205 is not too small. 4206 4207 movable_node [KNL,EARLY] Boot-time switch to make hotplugable memory 4208 NUMA nodes to be movable. This means that the memory 4209 of such nodes will be usable only for movable 4210 allocations which rules out almost all kernel 4211 allocations. Use with caution! 4212 4213 MTD_Partition= [MTD] 4214 Format: <name>,<region-number>,<size>,<offset> 4215 4216 MTD_Region= [MTD] Format: 4217 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>] 4218 4219 mtdparts= [MTD] 4220 See drivers/mtd/parsers/cmdlinepart.c 4221 4222 mtouchusb.raw_coordinates= 4223 [HW] Make the MicroTouch USB driver use raw coordinates 4224 ('y', default) or cooked coordinates ('n') 4225 4226 mtrr=debug [X86,EARLY] 4227 Enable printing debug information related to MTRR 4228 registers at boot time. 4229 4230 mtrr_chunk_size=nn[KMG,X86,EARLY] 4231 used for mtrr cleanup. It is largest continuous chunk 4232 that could hold holes aka. UC entries. 4233 4234 mtrr_gran_size=nn[KMG,X86,EARLY] 4235 Used for mtrr cleanup. It is granularity of mtrr block. 4236 Default is 1. 4237 Large value could prevent small alignment from 4238 using up MTRRs. 4239 4240 mtrr_spare_reg_nr=n [X86,EARLY] 4241 Format: <integer> 4242 Range: 0,7 : spare reg number 4243 Default : 1 4244 Used for mtrr cleanup. It is spare mtrr entries number. 4245 Set to 2 or more if your graphical card needs more. 4246 4247 multitce=off [PPC] This parameter disables the use of the pSeries 4248 firmware feature for updating multiple TCE entries 4249 at a time. 4250 4251 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card 4252 4253 netdev= [NET] Network devices parameters 4254 Format: <irq>,<io>,<mem_start>,<mem_end>,<name> 4255 Note that mem_start is often overloaded to mean 4256 something different and driver-specific. 4257 This usage is only documented in each driver source 4258 file if at all. 4259 4260 netpoll.carrier_timeout= 4261 [NET] Specifies amount of time (in seconds) that 4262 netpoll should wait for a carrier. By default netpoll 4263 waits 4 seconds. 4264 4265 nf_conntrack.acct= 4266 [NETFILTER] Enable connection tracking flow accounting 4267 0 to disable accounting 4268 1 to enable accounting 4269 Default value is 0. 4270 4271 nfs.cache_getent= 4272 [NFS] sets the pathname to the program which is used 4273 to update the NFS client cache entries. 4274 4275 nfs.cache_getent_timeout= 4276 [NFS] sets the timeout after which an attempt to 4277 update a cache entry is deemed to have failed. 4278 4279 nfs.callback_nr_threads= 4280 [NFSv4] set the total number of threads that the 4281 NFS client will assign to service NFSv4 callback 4282 requests. 4283 4284 nfs.callback_tcpport= 4285 [NFS] set the TCP port on which the NFSv4 callback 4286 channel should listen. 4287 4288 nfs.delay_retrans= 4289 [NFS] specifies the number of times the NFSv4 client 4290 retries the request before returning an EAGAIN error, 4291 after a reply of NFS4ERR_DELAY from the server. 4292 Only applies if the softerr mount option is enabled, 4293 and the specified value is >= 0. 4294 4295 nfs.enable_ino64= 4296 [NFS] enable 64-bit inode numbers. 4297 If zero, the NFS client will fake up a 32-bit inode 4298 number for the readdir() and stat() syscalls instead 4299 of returning the full 64-bit number. 4300 The default is to return 64-bit inode numbers. 4301 4302 nfs.idmap_cache_timeout= 4303 [NFS] set the maximum lifetime for idmapper cache 4304 entries. 4305 4306 nfs.max_session_cb_slots= 4307 [NFSv4.1] Sets the maximum number of session 4308 slots the client will assign to the callback 4309 channel. This determines the maximum number of 4310 callbacks the client will process in parallel for 4311 a particular server. 4312 4313 nfs.max_session_slots= 4314 [NFSv4.1] Sets the maximum number of session slots 4315 the client will attempt to negotiate with the server. 4316 This limits the number of simultaneous RPC requests 4317 that the client can send to the NFSv4.1 server. 4318 Note that there is little point in setting this 4319 value higher than the max_tcp_slot_table_limit. 4320 4321 nfs.nfs4_disable_idmapping= 4322 [NFSv4] When set to the default of '1', this option 4323 ensures that both the RPC level authentication 4324 scheme and the NFS level operations agree to use 4325 numeric uids/gids if the mount is using the 4326 'sec=sys' security flavour. In effect it is 4327 disabling idmapping, which can make migration from 4328 legacy NFSv2/v3 systems to NFSv4 easier. 4329 Servers that do not support this mode of operation 4330 will be autodetected by the client, and it will fall 4331 back to using the idmapper. 4332 To turn off this behaviour, set the value to '0'. 4333 4334 nfs.nfs4_unique_id= 4335 [NFS4] Specify an additional fixed unique ident- 4336 ification string that NFSv4 clients can insert into 4337 their nfs_client_id4 string. This is typically a 4338 UUID that is generated at system install time. 4339 4340 nfs.recover_lost_locks= 4341 [NFSv4] Attempt to recover locks that were lost due 4342 to a lease timeout on the server. Please note that 4343 doing this risks data corruption, since there are 4344 no guarantees that the file will remain unchanged 4345 after the locks are lost. 4346 If you want to enable the kernel legacy behaviour of 4347 attempting to recover these locks, then set this 4348 parameter to '1'. 4349 The default parameter value of '0' causes the kernel 4350 not to attempt recovery of lost locks. 4351 4352 nfs.send_implementation_id= 4353 [NFSv4.1] Send client implementation identification 4354 information in exchange_id requests. 4355 If zero, no implementation identification information 4356 will be sent. 4357 The default is to send the implementation identification 4358 information. 4359 4360 nfs4.layoutstats_timer= 4361 [NFSv4.2] Change the rate at which the kernel sends 4362 layoutstats to the pNFS metadata server. 4363 4364 Setting this to value to 0 causes the kernel to use 4365 whatever value is the default set by the layout 4366 driver. A non-zero value sets the minimum interval 4367 in seconds between layoutstats transmissions. 4368 4369 nfsd.inter_copy_offload_enable= 4370 [NFSv4.2] When set to 1, the server will support 4371 server-to-server copies for which this server is 4372 the destination of the copy. 4373 4374 nfsd.nfs4_disable_idmapping= 4375 [NFSv4] When set to the default of '1', the NFSv4 4376 server will return only numeric uids and gids to 4377 clients using auth_sys, and will accept numeric uids 4378 and gids from such clients. This is intended to ease 4379 migration from NFSv2/v3. 4380 4381 nfsd.nfsd4_ssc_umount_timeout= 4382 [NFSv4.2] When used as the destination of a 4383 server-to-server copy, knfsd temporarily mounts 4384 the source server. It caches the mount in case 4385 it will be needed again, and discards it if not 4386 used for the number of milliseconds specified by 4387 this parameter. 4388 4389 nfsaddrs= [NFS] Deprecated. Use ip= instead. 4390 See Documentation/admin-guide/nfs/nfsroot.rst. 4391 4392 nfsroot= [NFS] nfs root filesystem for disk-less boxes. 4393 See Documentation/admin-guide/nfs/nfsroot.rst. 4394 4395 nfsrootdebug [NFS] enable nfsroot debugging messages. 4396 See Documentation/admin-guide/nfs/nfsroot.rst. 4397 4398 nmi_backtrace.backtrace_idle [KNL] 4399 Dump stacks even of idle CPUs in response to an 4400 NMI stack-backtrace request. 4401 4402 nmi_debug= [KNL,SH] Specify one or more actions to take 4403 when a NMI is triggered. 4404 Format: [state][,regs][,debounce][,die] 4405 4406 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels 4407 Format: [panic,][nopanic,][rNNN,][num] 4408 Valid num: 0 or 1 4409 0 - turn hardlockup detector in nmi_watchdog off 4410 1 - turn hardlockup detector in nmi_watchdog on 4411 rNNN - configure the watchdog with raw perf event 0xNNN 4412 4413 When panic is specified, panic when an NMI watchdog 4414 timeout occurs (or 'nopanic' to not panic on an NMI 4415 watchdog, if CONFIG_BOOTPARAM_HARDLOCKUP_PANIC is set) 4416 To disable both hard and soft lockup detectors, 4417 please see 'nowatchdog'. 4418 This is useful when you use a panic=... timeout and 4419 need the box quickly up again. 4420 4421 These settings can be accessed at runtime via 4422 the nmi_watchdog and hardlockup_panic sysctls. 4423 4424 no4lvl [RISCV,EARLY] Disable 4-level and 5-level paging modes. 4425 Forces kernel to use 3-level paging instead. 4426 4427 no5lvl [X86-64,RISCV,EARLY] Disable 5-level paging mode. Forces 4428 kernel to use 4-level paging instead. 4429 4430 noalign [KNL,ARM] 4431 4432 noapic [SMP,APIC,EARLY] Tells the kernel to not make use of any 4433 IOAPICs that may be present in the system. 4434 4435 noapictimer [APIC,X86] Don't set up the APIC timer 4436 4437 noautogroup Disable scheduler automatic task group creation. 4438 4439 nocache [ARM,EARLY] 4440 4441 no_console_suspend 4442 [HW] Never suspend the console 4443 Disable suspending of consoles during suspend and 4444 hibernate operations. Once disabled, debugging 4445 messages can reach various consoles while the rest 4446 of the system is being put to sleep (ie, while 4447 debugging driver suspend/resume hooks). This may 4448 not work reliably with all consoles, but is known 4449 to work with serial and VGA consoles. 4450 To facilitate more flexible debugging, we also add 4451 console_suspend, a printk module parameter to control 4452 it. Users could use console_suspend (usually 4453 /sys/module/printk/parameters/console_suspend) to 4454 turn on/off it dynamically. 4455 4456 no_debug_objects 4457 [KNL,EARLY] Disable object debugging 4458 4459 nodsp [SH] Disable hardware DSP at boot time. 4460 4461 noefi [EFI,EARLY] Disable EFI runtime services support. 4462 4463 no_entry_flush [PPC,EARLY] Don't flush the L1-D cache when entering the kernel. 4464 4465 noexec32 [X86-64] 4466 This affects only 32-bit executables. 4467 noexec32=on: enable non-executable mappings (default) 4468 read doesn't imply executable mappings 4469 noexec32=off: disable non-executable mappings 4470 read implies executable mappings 4471 4472 no_file_caps Tells the kernel not to honor file capabilities. The 4473 only way then for a file to be executed with privilege 4474 is to be setuid root or executed by root. 4475 4476 nofpu [MIPS,SH] Disable hardware FPU at boot time. 4477 4478 nofsgsbase [X86] Disables FSGSBASE instructions. 4479 4480 nofxsr [BUGS=X86-32] Disables x86 floating point extended 4481 register save and restore. The kernel will only save 4482 legacy floating-point registers on task switch. 4483 4484 nogbpages [X86] Do not use GB pages for kernel direct mappings. 4485 4486 no_hash_pointers 4487 [KNL,EARLY] 4488 Alias for "hash_pointers=never". 4489 4490 nohibernate [HIBERNATION] Disable hibernation and resume. 4491 4492 nohlt [ARM,ARM64,MICROBLAZE,MIPS,PPC,RISCV,SH] Forces the kernel to 4493 busy wait in do_idle() and not use the arch_cpu_idle() 4494 implementation; requires CONFIG_GENERIC_IDLE_POLL_SETUP 4495 to be effective. This is useful on platforms where the 4496 sleep(SH) or wfi(ARM,ARM64) instructions do not work 4497 correctly or when doing power measurements to evaluate 4498 the impact of the sleep instructions. This is also 4499 useful when using JTAG debugger. 4500 4501 nohpet [X86] Don't use the HPET timer. 4502 4503 nohugeiomap [KNL,X86,PPC,ARM64,EARLY] Disable kernel huge I/O mappings. 4504 4505 nohugevmalloc [KNL,X86,PPC,ARM64,EARLY] Disable kernel huge vmalloc mappings. 4506 4507 nohz= [KNL] Boottime enable/disable dynamic ticks 4508 Valid arguments: on, off 4509 Default: on 4510 4511 nohz_full= [KNL,BOOT,SMP,ISOL] 4512 The argument is a cpu list, as described above. 4513 In kernels built with CONFIG_NO_HZ_FULL=y, set 4514 the specified list of CPUs whose tick will be stopped 4515 whenever possible. The boot CPU will be forced outside 4516 the range to maintain the timekeeping. Any CPUs 4517 in this list will have their RCU callbacks offloaded, 4518 just as if they had also been called out in the 4519 rcu_nocbs= boot parameter. 4520 4521 Note that this argument takes precedence over 4522 the CONFIG_RCU_NOCB_CPU_DEFAULT_ALL option. 4523 4524 noinitrd [Deprecated,RAM] Tells the kernel not to load any configured 4525 initial RAM disk. Currently this parameter applies to 4526 initrd only, not to initramfs. But it applies to both 4527 in EFI mode. 4528 4529 nointremap [X86-64,Intel-IOMMU,EARLY] Do not enable interrupt 4530 remapping. 4531 [Deprecated - use intremap=off] 4532 4533 noinvpcid [X86,EARLY] Disable the INVPCID cpu feature. 4534 4535 noiotrap [SH] Disables trapped I/O port accesses. 4536 4537 noirqdebug [X86-32] Disables the code which attempts to detect and 4538 disable unhandled interrupt sources. 4539 4540 noisapnp [ISAPNP] Disables ISA PnP code. 4541 4542 nokaslr [KNL,EARLY] 4543 When CONFIG_RANDOMIZE_BASE is set, this disables 4544 kernel and module base offset ASLR (Address Space 4545 Layout Randomization). 4546 4547 no-kvmapf [X86,KVM,EARLY] Disable paravirtualized asynchronous page 4548 fault handling. 4549 4550 no-kvmclock [X86,KVM,EARLY] Disable paravirtualized KVM clock driver 4551 4552 nolapic [X86-32,APIC,EARLY] Do not enable or use the local APIC. 4553 4554 nolapic_timer [X86-32,APIC,EARLY] Do not use the local APIC timer. 4555 4556 nomce [X86-32] Disable Machine Check Exception 4557 4558 nomfgpt [X86-32] Disable Multi-Function General Purpose 4559 Timer usage (for AMD Geode machines). 4560 4561 nomodeset Disable kernel modesetting. Most systems' firmware 4562 sets up a display mode and provides framebuffer memory 4563 for output. With nomodeset, DRM and fbdev drivers will 4564 not load if they could possibly displace the pre- 4565 initialized output. Only the system framebuffer will 4566 be available for use. The respective drivers will not 4567 perform display-mode changes or accelerated rendering. 4568 4569 Useful as error fallback, or for testing and debugging. 4570 4571 nomodule Disable module load 4572 4573 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to 4574 shutdown the other cpus. Instead use the REBOOT_VECTOR 4575 irq. 4576 4577 nopat [X86,EARLY] Disable PAT (page attribute table extension of 4578 pagetables) support. 4579 4580 nopcid [X86-64,EARLY] Disable the PCID cpu feature. 4581 4582 nopku [X86] Disable Memory Protection Keys CPU feature found 4583 in some Intel CPUs. 4584 4585 nopti [X86-64,EARLY] 4586 Equivalent to pti=off 4587 4588 nopv= [X86,XEN,KVM,HYPER_V,VMWARE,EARLY] 4589 Disables the PV optimizations forcing the guest to run 4590 as generic guest with no PV drivers. Currently support 4591 XEN HVM, KVM, HYPER_V and VMWARE guest. 4592 4593 nopvspin [X86,XEN,KVM,EARLY] 4594 Disables the qspinlock slow path using PV optimizations 4595 which allow the hypervisor to 'idle' the guest on lock 4596 contention. 4597 4598 norandmaps Don't use address space randomization. Equivalent to 4599 echo 0 > /proc/sys/kernel/randomize_va_space 4600 4601 noreplace-smp [X86-32,SMP] Don't replace SMP instructions 4602 with UP alternatives 4603 4604 noresume [SWSUSP] Disables resume and restores original swap 4605 space. 4606 4607 no-scroll [VGA] Disables scrollback. 4608 This is required for the Braillex ib80-piezo Braille 4609 reader made by F.H. Papenmeier (Germany). 4610 4611 nosgx [X86-64,SGX,EARLY] Disables Intel SGX kernel support. 4612 4613 nosmap [PPC,EARLY] 4614 Disable SMAP (Supervisor Mode Access Prevention) 4615 even if it is supported by processor. 4616 4617 nosmep [PPC64s,EARLY] 4618 Disable SMEP (Supervisor Mode Execution Prevention) 4619 even if it is supported by processor. 4620 4621 nosmp [SMP,EARLY] Tells an SMP kernel to act as a UP kernel, 4622 and disable the IO APIC. legacy for "maxcpus=0". 4623 4624 nosmt [KNL,MIPS,PPC,EARLY] Disable symmetric multithreading (SMT). 4625 Equivalent to smt=1. 4626 4627 [KNL,LOONGARCH,X86,ARM64,PPC,S390] Disable symmetric multithreading (SMT). 4628 nosmt=force: Force disable SMT, cannot be undone 4629 via the sysfs control file. 4630 4631 nosoftlockup [KNL] Disable the soft-lockup detector. 4632 4633 nospec_store_bypass_disable 4634 [HW,EARLY] Disable all mitigations for the Speculative 4635 Store Bypass vulnerability 4636 4637 nospectre_bhb [ARM64,EARLY] Disable all mitigations for Spectre-BHB (branch 4638 history injection) vulnerability. System may allow data leaks 4639 with this option. 4640 4641 nospectre_v1 [X86,PPC,EARLY] Disable mitigations for Spectre Variant 1 4642 (bounds check bypass). With this option data leaks are 4643 possible in the system. 4644 4645 nospectre_v2 [X86,PPC_E500,ARM64,EARLY] Disable all mitigations 4646 for the Spectre variant 2 (indirect branch 4647 prediction) vulnerability. System may allow data 4648 leaks with this option. 4649 4650 no-steal-acc [X86,PV_OPS,ARM64,PPC/PSERIES,RISCV,LOONGARCH,EARLY] 4651 Disable paravirtualized steal time accounting. steal time 4652 is computed, but won't influence scheduler behaviour 4653 4654 nosync [HW,M68K] Disables sync negotiation for all devices. 4655 4656 no_timer_check [X86,APIC] Disables the code which tests for broken 4657 timer IRQ sources, i.e., the IO-APIC timer. This can 4658 work around problems with incorrect timer 4659 initialization on some boards. 4660 4661 no_uaccess_flush 4662 [PPC,EARLY] Don't flush the L1-D cache after accessing user data. 4663 4664 novmcoredd [KNL,KDUMP] 4665 Disable device dump. Device dump allows drivers to 4666 append dump data to vmcore so you can collect driver 4667 specified debug info. Drivers can append the data 4668 without any limit and this data is stored in memory, 4669 so this may cause significant memory stress. Disabling 4670 device dump can help save memory but the driver debug 4671 data will be no longer available. This parameter 4672 is only available when CONFIG_PROC_VMCORE_DEVICE_DUMP 4673 is set. 4674 4675 no-vmw-sched-clock 4676 [X86,PV_OPS,EARLY] Disable paravirtualized VMware 4677 scheduler clock and use the default one. 4678 4679 nowatchdog [KNL] Disable both lockup detectors, i.e. 4680 soft-lockup and NMI watchdog (hard-lockup). 4681 4682 nowb [ARM,EARLY] 4683 4684 nox2apic [X86-64,APIC,EARLY] Do not enable x2APIC mode. 4685 4686 NOTE: this parameter will be ignored on systems with the 4687 LEGACY_XAPIC_DISABLED bit set in the 4688 IA32_XAPIC_DISABLE_STATUS MSR. 4689 4690 noxsave [BUGS=X86] Disables x86 extended register state save 4691 and restore using xsave. The kernel will fallback to 4692 enabling legacy floating-point and sse state. 4693 4694 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended 4695 register states. The kernel will fall back to use 4696 xsave to save the states. By using this parameter, 4697 performance of saving the states is degraded because 4698 xsave doesn't support modified optimization while 4699 xsaveopt supports it on xsaveopt enabled systems. 4700 4701 noxsaves [X86] Disables xsaves and xrstors used in saving and 4702 restoring x86 extended register state in compacted 4703 form of xsave area. The kernel will fall back to use 4704 xsaveopt and xrstor to save and restore the states 4705 in standard form of xsave area. By using this 4706 parameter, xsave area per process might occupy more 4707 memory on xsaves enabled systems. 4708 4709 nr_cpus= [SMP,EARLY] Maximum number of processors that an SMP kernel 4710 could support. nr_cpus=n : n >= 1 limits the kernel to 4711 support 'n' processors. It could be larger than the 4712 number of already plugged CPU during bootup, later in 4713 runtime you can physically add extra cpu until it reaches 4714 n. So during boot up some boot time memory for per-cpu 4715 variables need be pre-allocated for later physical cpu 4716 hot plugging. 4717 4718 nr_uarts= [SERIAL] maximum number of UARTs to be registered. 4719 4720 numa=off [KNL, ARM64, PPC, RISCV, SPARC, X86, EARLY] 4721 Disable NUMA, Only set up a single NUMA node 4722 spanning all memory. 4723 4724 numa=fake=<size>[MG] 4725 [KNL, ARM64, RISCV, X86, EARLY] 4726 If given as a memory unit, fills all system RAM with 4727 nodes of size interleaved over physical nodes. 4728 4729 numa=fake=<N> 4730 [KNL, ARM64, RISCV, X86, EARLY] 4731 If given as an integer, fills all system RAM with N 4732 fake nodes interleaved over physical nodes. 4733 4734 numa=fake=<N>U 4735 [KNL, ARM64, RISCV, X86, EARLY] 4736 If given as an integer followed by 'U', it will 4737 divide each physical node into N emulated nodes. 4738 4739 numa=noacpi [X86] Don't parse the SRAT table for NUMA setup 4740 4741 numa=nohmat [X86] Don't parse the HMAT table for NUMA setup, or 4742 soft-reserved memory partitioning. 4743 4744 numa_balancing= [KNL,ARM64,PPC,RISCV,S390,X86] Enable or disable automatic 4745 NUMA balancing. 4746 Allowed values are enable and disable 4747 4748 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA. 4749 'node', 'default' can be specified 4750 This can be set from sysctl after boot. 4751 See Documentation/admin-guide/sysctl/vm.rst for details. 4752 4753 nvme.quirks= [NVME] A list of quirk entries to augment the built-in 4754 nvme quirk list. List entries are separated by a 4755 '-' character. 4756 Each entry has the form VendorID:ProductID:quirk_names. 4757 The IDs are 4-digits hex numbers and quirk_names is a 4758 list of quirk names separated by commas. A quirk name 4759 can be prefixed by '^', meaning that the specified 4760 quirk must be disabled. 4761 4762 Example: 4763 nvme.quirks=7710:2267:bogus_nid,^identify_cns-9900:7711:broken_msi 4764 4765 ohci1394_dma=early [HW,EARLY] enable debugging via the ohci1394 driver. 4766 See Documentation/core-api/debugging-via-ohci1394.rst for more 4767 info. 4768 4769 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands 4770 Rather than timing out after 20 ms if an EC 4771 command is not properly ACKed, override the length 4772 of the timeout. We have interrupts disabled while 4773 waiting for the ACK, so if this is set too high 4774 interrupts *may* be lost! 4775 4776 omap_mux= [OMAP] Override bootloader pin multiplexing. 4777 Format: <mux_mode0.mode_name=value>... 4778 For example, to override I2C bus2: 4779 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100 4780 4781 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration 4782 4783 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock] 4784 4785 boundary - index of last SLC block on Flex-OneNAND. 4786 The remaining blocks are configured as MLC blocks. 4787 lock - Configure if Flex-OneNAND boundary should be locked. 4788 Once locked, the boundary cannot be changed. 4789 1 indicates lock status, 0 indicates unlock status. 4790 4791 oops=panic [KNL,EARLY] 4792 Always panic on oopses. Default is to just kill the 4793 process, but there is a small probability of 4794 deadlocking the machine. 4795 This will also cause panics on machine check exceptions. 4796 Useful together with panic=30 to trigger a reboot. 4797 4798 page_alloc.shuffle= 4799 [KNL] Boolean flag to control whether the page allocator 4800 should randomize its free lists. This parameter can be 4801 used to enable/disable page randomization. The state of 4802 the flag can be read from sysfs at: 4803 /sys/module/page_alloc/parameters/shuffle. 4804 This parameter is only available if CONFIG_SHUFFLE_PAGE_ALLOCATOR=y. 4805 4806 page_owner= [KNL,EARLY] Boot-time page_owner enabling option. 4807 Storage of the information about who allocated 4808 each page is disabled in default. With this switch, 4809 we can turn it on. 4810 on: enable the feature 4811 4812 page_poison= [KNL,EARLY] Boot-time parameter changing the state of 4813 poisoning on the buddy allocator, available with 4814 CONFIG_PAGE_POISONING=y. 4815 off: turn off poisoning (default) 4816 on: turn on poisoning 4817 4818 page_reporting.page_reporting_order= 4819 [KNL] Minimal page reporting order 4820 Format: <integer> 4821 Adjust the minimal page reporting order. The page 4822 reporting is disabled when it exceeds MAX_PAGE_ORDER. 4823 4824 panic= [KNL] Kernel behaviour on panic: delay <timeout> 4825 timeout > 0: seconds before rebooting 4826 timeout = 0: wait forever 4827 timeout < 0: reboot immediately 4828 Format: <timeout> 4829 4830 panic_on_taint= [KNL,EARLY] 4831 Bitmask for conditionally calling panic() in add_taint() 4832 Format: <hex>[,nousertaint] 4833 Hexadecimal bitmask representing the set of TAINT flags 4834 that will cause the kernel to panic when add_taint() is 4835 called with any of the flags in this set. 4836 The optional switch "nousertaint" can be utilized to 4837 prevent userspace forced crashes by writing to sysctl 4838 /proc/sys/kernel/tainted any flagset matching with the 4839 bitmask set on panic_on_taint. 4840 See Documentation/admin-guide/tainted-kernels.rst for 4841 extra details on the taint flags that users can pick 4842 to compose the bitmask to assign to panic_on_taint. 4843 4844 panic_on_warn=1 panic() instead of WARN(). Useful to cause kdump 4845 on a WARN(). 4846 4847 panic_force_cpu= 4848 [KNL,SMP] Force panic handling to execute on a specific CPU. 4849 Format: <cpu number> 4850 Some platforms require panic handling to occur on a 4851 specific CPU for the crash kernel to function correctly. 4852 This can be due to firmware limitations, interrupt routing 4853 constraints, or platform-specific requirements where only 4854 a particular CPU can safely enter the crash kernel. 4855 When set, panic() will redirect execution to the specified 4856 CPU before proceeding with the normal panic and kexec flow. 4857 If the target CPU is offline or unavailable, panic proceeds 4858 on the current CPU. 4859 This option should only be used for systems with the above 4860 constraints as it might cause the panic operation to be less reliable. 4861 4862 panic_print= Bitmask for printing system info when panic happens. 4863 User can chose combination of the following bits: 4864 bit 0: print all tasks info 4865 bit 1: print system memory info 4866 bit 2: print timer info 4867 bit 3: print locks info if CONFIG_LOCKDEP is on 4868 bit 4: print ftrace buffer 4869 bit 5: replay all kernel messages on consoles at the end of panic 4870 bit 6: print all CPUs backtrace (if available in the arch) 4871 bit 7: print only tasks in uninterruptible (blocked) state 4872 *Be aware* that this option may print a _lot_ of lines, 4873 so there are risks of losing older messages in the log. 4874 Use this option carefully, maybe worth to setup a 4875 bigger log buffer with "log_buf_len" along with this. 4876 4877 panic_sys_info= A comma separated list of extra information to be dumped 4878 on panic. 4879 Format: val[,val...] 4880 Where @val can be any of the following: 4881 4882 tasks: print all tasks info 4883 mem: print system memory info 4884 timers: print timers info 4885 locks: print locks info if CONFIG_LOCKDEP is on 4886 ftrace: print ftrace buffer 4887 all_bt: print all CPUs backtrace (if available in the arch) 4888 blocked_tasks: print only tasks in uninterruptible (blocked) state 4889 4890 This is a human readable alternative to the 'panic_print' option. 4891 4892 panic_console_replay 4893 When panic happens, replay all kernel messages on 4894 consoles at the end of panic. 4895 4896 parkbd.port= [HW] Parallel port number the keyboard adapter is 4897 connected to, default is 0. 4898 Format: <parport#> 4899 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation, 4900 0 for XT, 1 for AT (default is AT). 4901 Format: <mode> 4902 4903 parport= [HW,PPT] Specify parallel ports. 0 disables. 4904 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] } 4905 Use 'auto' to force the driver to use any 4906 IRQ/DMA settings detected (the default is to 4907 ignore detected IRQ/DMA settings because of 4908 possible conflicts). You can specify the base 4909 address, IRQ, and DMA settings; IRQ and DMA 4910 should be numbers, or 'auto' (for using detected 4911 settings on that particular port), or 'nofifo' 4912 (to avoid using a FIFO even if it is detected). 4913 Parallel ports are assigned in the order they 4914 are specified on the command line, starting 4915 with parport0. 4916 4917 parport_init_mode= [HW,PPT] 4918 Configure VIA parallel port to operate in 4919 a specific mode. This is necessary on Pegasos 4920 computer where firmware has no options for setting 4921 up parallel port mode and sets it to spp. 4922 Currently this function knows 686a and 8231 chips. 4923 Format: [spp|ps2|epp|ecp|ecpepp] 4924 4925 pata_legacy.all= [HW,LIBATA] 4926 Format: <int> 4927 Set to non-zero to probe primary and secondary ISA 4928 port ranges on PCI systems where no PCI PATA device 4929 has been found at either range. Disabled by default. 4930 4931 pata_legacy.autospeed= [HW,LIBATA] 4932 Format: <int> 4933 Set to non-zero if a chip is present that snoops speed 4934 changes. Disabled by default. 4935 4936 pata_legacy.ht6560a= [HW,LIBATA] 4937 Format: <int> 4938 Set to 1, 2, or 3 for HT 6560A on the primary channel, 4939 the secondary channel, or both channels respectively. 4940 Disabled by default. 4941 4942 pata_legacy.ht6560b= [HW,LIBATA] 4943 Format: <int> 4944 Set to 1, 2, or 3 for HT 6560B on the primary channel, 4945 the secondary channel, or both channels respectively. 4946 Disabled by default. 4947 4948 pata_legacy.iordy_mask= [HW,LIBATA] 4949 Format: <int> 4950 IORDY enable mask. Set individual bits to allow IORDY 4951 for the respective channel. Bit 0 is for the first 4952 legacy channel handled by this driver, bit 1 is for 4953 the second channel, and so on. The sequence will often 4954 correspond to the primary legacy channel, the secondary 4955 legacy channel, and so on, but the handling of a PCI 4956 bus and the use of other driver options may interfere 4957 with the sequence. By default IORDY is allowed across 4958 all channels. 4959 4960 pata_legacy.opti82c46x= [HW,LIBATA] 4961 Format: <int> 4962 Set to 1, 2, or 3 for Opti 82c611A on the primary 4963 channel, the secondary channel, or both channels 4964 respectively. Disabled by default. 4965 4966 pata_legacy.opti82c611a= [HW,LIBATA] 4967 Format: <int> 4968 Set to 1, 2, or 3 for Opti 82c465MV on the primary 4969 channel, the secondary channel, or both channels 4970 respectively. Disabled by default. 4971 4972 pata_legacy.pio_mask= [HW,LIBATA] 4973 Format: <int> 4974 PIO mode mask for autospeed devices. Set individual 4975 bits to allow the use of the respective PIO modes. 4976 Bit 0 is for mode 0, bit 1 is for mode 1, and so on. 4977 All modes allowed by default. 4978 4979 pata_legacy.probe_all= [HW,LIBATA] 4980 Format: <int> 4981 Set to non-zero to probe tertiary and further ISA 4982 port ranges on PCI systems. Disabled by default. 4983 4984 pata_legacy.probe_mask= [HW,LIBATA] 4985 Format: <int> 4986 Probe mask for legacy ISA PATA ports. Depending on 4987 platform configuration and the use of other driver 4988 options up to 6 legacy ports are supported: 0x1f0, 4989 0x170, 0x1e8, 0x168, 0x1e0, 0x160, however probing 4990 of individual ports can be disabled by setting the 4991 corresponding bits in the mask to 1. Bit 0 is for 4992 the first port in the list above (0x1f0), and so on. 4993 By default all supported ports are probed. 4994 4995 pata_legacy.qdi= [HW,LIBATA] 4996 Format: <int> 4997 Set to non-zero to probe QDI controllers. By default 4998 set to 1 if CONFIG_PATA_QDI_MODULE, 0 otherwise. 4999 5000 pata_legacy.winbond= [HW,LIBATA] 5001 Format: <int> 5002 Set to non-zero to probe Winbond controllers. Use 5003 the standard I/O port (0x130) if 1, otherwise the 5004 value given is the I/O port to use (typically 0x1b0). 5005 By default set to 1 if CONFIG_PATA_WINBOND_VLB_MODULE, 5006 0 otherwise. 5007 5008 pata_platform.pio_mask= [HW,LIBATA] 5009 Format: <int> 5010 Supported PIO mode mask. Set individual bits to allow 5011 the use of the respective PIO modes. Bit 0 is for 5012 mode 0, bit 1 is for mode 1, and so on. Mode 0 only 5013 allowed by default. 5014 5015 pause_on_oops=<int> 5016 Halt all CPUs after the first oops has been printed for 5017 the specified number of seconds. This is to be used if 5018 your oopses keep scrolling off the screen. 5019 5020 pci=option[,option...] [PCI,EARLY] various PCI subsystem options. 5021 5022 Some options herein operate on a specific device 5023 or a set of devices (<pci_dev>). These are 5024 specified in one of the following formats: 5025 5026 [<domain>:]<bus>:<dev>.<func>[/<dev>.<func>]* 5027 pci:<vendor>:<device>[:<subvendor>:<subdevice>] 5028 5029 Note: the first format specifies a PCI 5030 bus/device/function address which may change 5031 if new hardware is inserted, if motherboard 5032 firmware changes, or due to changes caused 5033 by other kernel parameters. If the 5034 domain is left unspecified, it is 5035 taken to be zero. Optionally, a path 5036 to a device through multiple device/function 5037 addresses can be specified after the base 5038 address (this is more robust against 5039 renumbering issues). The second format 5040 selects devices using IDs from the 5041 configuration space which may match multiple 5042 devices in the system. 5043 5044 earlydump dump PCI config space before the kernel 5045 changes anything 5046 off [X86] don't probe for the PCI bus 5047 bios [X86-32] force use of PCI BIOS, don't access 5048 the hardware directly. Use this if your machine 5049 has a non-standard PCI host bridge. 5050 nobios [X86-32] disallow use of PCI BIOS, only direct 5051 hardware access methods are allowed. Use this 5052 if you experience crashes upon bootup and you 5053 suspect they are caused by the BIOS. 5054 conf1 [X86] Force use of PCI Configuration Access 5055 Mechanism 1 (config address in IO port 0xCF8, 5056 data in IO port 0xCFC, both 32-bit). 5057 conf2 [X86] Force use of PCI Configuration Access 5058 Mechanism 2 (IO port 0xCF8 is an 8-bit port for 5059 the function, IO port 0xCFA, also 8-bit, sets 5060 bus number. The config space is then accessed 5061 through ports 0xC000-0xCFFF). 5062 See http://wiki.osdev.org/PCI for more info 5063 on the configuration access mechanisms. 5064 noaer [PCIE] If the PCIEAER kernel config parameter is 5065 enabled, this kernel boot option can be used to 5066 disable the use of PCIE advanced error reporting. 5067 nodomains [PCI] Disable support for multiple PCI 5068 root domains (aka PCI segments, in ACPI-speak). 5069 nommconf [X86] Disable use of MMCONFIG for PCI 5070 Configuration 5071 check_enable_amd_mmconf [X86] check for and enable 5072 properly configured MMIO access to PCI 5073 config space on AMD family 10h CPU 5074 nomsi [MSI] If the PCI_MSI kernel config parameter is 5075 enabled, this kernel boot option can be used to 5076 disable the use of MSI interrupts system-wide. 5077 noioapicquirk [APIC] Disable all boot interrupt quirks. 5078 Safety option to keep boot IRQs enabled. This 5079 should never be necessary. 5080 ioapicreroute [APIC] Enable rerouting of boot IRQs to the 5081 primary IO-APIC for bridges that cannot disable 5082 boot IRQs. This fixes a source of spurious IRQs 5083 when the system masks IRQs. 5084 noioapicreroute [APIC] Disable workaround that uses the 5085 boot IRQ equivalent of an IRQ that connects to 5086 a chipset where boot IRQs cannot be disabled. 5087 The opposite of ioapicreroute. 5088 biosirq [X86-32] Use PCI BIOS calls to get the interrupt 5089 routing table. These calls are known to be buggy 5090 on several machines and they hang the machine 5091 when used, but on other computers it's the only 5092 way to get the interrupt routing table. Try 5093 this option if the kernel is unable to allocate 5094 IRQs or discover secondary PCI buses on your 5095 motherboard. 5096 rom [X86] Assign address space to expansion ROMs. 5097 Use with caution as certain devices share 5098 address decoders between ROMs and other 5099 resources. 5100 norom [X86] Do not assign address space to 5101 expansion ROMs that do not already have 5102 BIOS assigned address ranges. 5103 nobar [X86] Do not assign address space to the 5104 BARs that weren't assigned by the BIOS. 5105 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be 5106 assigned automatically to PCI devices. You can 5107 make the kernel exclude IRQs of your ISA cards 5108 this way. 5109 pirqaddr=0xAAAAA [X86] Specify the physical address 5110 of the PIRQ table (normally generated 5111 by the BIOS) if it is outside the 5112 F0000h-100000h range. 5113 lastbus=N [X86] Scan all buses thru bus #N. Can be 5114 useful if the kernel is unable to find your 5115 secondary buses and you want to tell it 5116 explicitly which ones they are. 5117 assign-busses [X86] Always assign all PCI bus 5118 numbers ourselves, overriding 5119 whatever the firmware may have done. 5120 usepirqmask [X86] Honor the possible IRQ mask stored 5121 in the BIOS $PIR table. This is needed on 5122 some systems with broken BIOSes, notably 5123 some HP Pavilion N5400 and Omnibook XE3 5124 notebooks. This will have no effect if ACPI 5125 IRQ routing is enabled. 5126 noacpi [X86] Do not use ACPI for IRQ routing 5127 or for PCI scanning. 5128 use_crs [X86] Use PCI host bridge window information 5129 from ACPI. On BIOSes from 2008 or later, this 5130 is enabled by default. If you need to use this, 5131 please report a bug. 5132 nocrs [X86] Ignore PCI host bridge windows from ACPI. 5133 If you need to use this, please report a bug. 5134 use_e820 [X86] Use E820 reservations to exclude parts of 5135 PCI host bridge windows. This is a workaround 5136 for BIOS defects in host bridge _CRS methods. 5137 If you need to use this, please report a bug to 5138 <linux-pci@vger.kernel.org>. 5139 no_e820 [X86] Ignore E820 reservations for PCI host 5140 bridge windows. This is the default on modern 5141 hardware. If you need to use this, please report 5142 a bug to <linux-pci@vger.kernel.org>. 5143 routeirq Do IRQ routing for all PCI devices. 5144 This is normally done in pci_enable_device(), 5145 so this option is a temporary workaround 5146 for broken drivers that don't call it. 5147 skip_isa_align [X86] do not align io start addr, so can 5148 handle more pci cards 5149 noearly [X86] Don't do any early type 1 scanning. 5150 This might help on some broken boards which 5151 machine check when some devices' config space 5152 is read. But various workarounds are disabled 5153 and some IOMMU drivers will not work. 5154 bfsort Sort PCI devices into breadth-first order. 5155 This sorting is done to get a device 5156 order compatible with older (<= 2.4) kernels. 5157 nobfsort Don't sort PCI devices into breadth-first order. 5158 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size) 5159 tuning and use the BIOS-configured MPS defaults. 5160 pcie_bus_safe Set every device's MPS to the largest value 5161 supported by all devices below the root complex. 5162 pcie_bus_perf Set device MPS to the largest allowable MPS 5163 based on its parent bus. Also set MRRS (Max 5164 Read Request Size) to the largest supported 5165 value (no larger than the MPS that the device 5166 or bus can support) for best performance. 5167 pcie_bus_peer2peer Set every device's MPS to 128B, which 5168 every device is guaranteed to support. This 5169 configuration allows peer-to-peer DMA between 5170 any pair of devices, possibly at the cost of 5171 reduced performance. This also guarantees 5172 that hot-added devices will work. 5173 cbiosize=nn[KMG] The fixed amount of bus space which is 5174 reserved for the CardBus bridge's IO window. 5175 The default value is 256 bytes. 5176 cbmemsize=nn[KMG] The fixed amount of bus space which is 5177 reserved for the CardBus bridge's memory 5178 window. The default value is 64 megabytes. 5179 resource_alignment= 5180 Format: 5181 [<order of align>@]<pci_dev>[; ...] 5182 Specifies alignment and device to reassign 5183 aligned memory resources. How to 5184 specify the device is described above. 5185 If <order of align> is not specified, 5186 PAGE_SIZE is used as alignment. 5187 A PCI-PCI bridge can be specified if resource 5188 windows need to be expanded. 5189 To specify the alignment for several 5190 instances of a device, the PCI vendor, 5191 device, subvendor, and subdevice may be 5192 specified, e.g., 12@pci:8086:9c22:103c:198f 5193 for 4096-byte alignment. 5194 ecrc= Enable/disable PCIe ECRC (transaction layer 5195 end-to-end CRC checking). Only effective if 5196 OS has native AER control (either granted by 5197 ACPI _OSC or forced via "pcie_ports=native") 5198 bios: Use BIOS/firmware settings. This is the 5199 the default. 5200 off: Turn ECRC off 5201 on: Turn ECRC on. 5202 hpiosize=nn[KMG] The fixed amount of bus space which is 5203 reserved for hotplug bridge's IO window. 5204 Default size is 256 bytes. 5205 hpmmiosize=nn[KMG] The fixed amount of bus space which is 5206 reserved for hotplug bridge's MMIO window. 5207 Default size is 2 megabytes. 5208 hpmmioprefsize=nn[KMG] The fixed amount of bus space which is 5209 reserved for hotplug bridge's MMIO_PREF window. 5210 Default size is 2 megabytes. 5211 hpmemsize=nn[KMG] The fixed amount of bus space which is 5212 reserved for hotplug bridge's MMIO and 5213 MMIO_PREF window. 5214 Default size is 2 megabytes. 5215 hpbussize=nn The minimum amount of additional bus numbers 5216 reserved for buses below a hotplug bridge. 5217 Default is 1. 5218 realloc= Enable/disable reallocating PCI bridge resources 5219 if allocations done by BIOS are too small to 5220 accommodate resources required by all child 5221 devices. 5222 off: Turn realloc off 5223 on: Turn realloc on 5224 realloc same as realloc=on 5225 noari do not use PCIe ARI. 5226 noats [PCIE, Intel-IOMMU, AMD-IOMMU] 5227 do not use PCIe ATS (and IOMMU device IOTLB). 5228 pcie_scan_all Scan all possible PCIe devices. Otherwise we 5229 only look for one device below a PCIe downstream 5230 port. 5231 big_root_window Try to add a big 64bit memory window to the PCIe 5232 root complex on AMD CPUs. Some GFX hardware 5233 can resize a BAR to allow access to all VRAM. 5234 Adding the window is slightly risky (it may 5235 conflict with unreported devices), so this 5236 taints the kernel. 5237 disable_acs_redir=<pci_dev>[; ...] 5238 Specify one or more PCI devices (in the format 5239 specified above) separated by semicolons. 5240 Each device specified will have the PCI ACS 5241 redirect capabilities forced off which will 5242 allow P2P traffic between devices through 5243 bridges without forcing it upstream. Note: 5244 this removes isolation between devices and 5245 may put more devices in an IOMMU group. 5246 config_acs= 5247 Format: 5248 <ACS flags>@<pci_dev>[; ...] 5249 Specify one or more PCI devices (in the format 5250 specified above) optionally prepended with flags 5251 and separated by semicolons. The respective 5252 capabilities will be enabled, disabled or 5253 unchanged based on what is specified in 5254 flags. 5255 5256 ACS Flags is defined as follows: 5257 bit-0 : ACS Source Validation 5258 bit-1 : ACS Translation Blocking 5259 bit-2 : ACS P2P Request Redirect 5260 bit-3 : ACS P2P Completion Redirect 5261 bit-4 : ACS Upstream Forwarding 5262 bit-5 : ACS P2P Egress Control 5263 bit-6 : ACS Direct Translated P2P 5264 Each bit can be marked as: 5265 '0' – force disabled 5266 '1' – force enabled 5267 'x' – unchanged 5268 For example, 5269 pci=config_acs=10x@pci:0:0 5270 would configure all devices that support 5271 ACS to enable P2P Request Redirect, disable 5272 Translation Blocking, and leave Source 5273 Validation unchanged from whatever power-up 5274 or firmware set it to. 5275 5276 Note: this may remove isolation between devices 5277 and may put more devices in an IOMMU group. 5278 force_floating [S390] Force usage of floating interrupts. 5279 nomio [S390] Do not use MIO instructions. 5280 norid [S390] ignore the RID field and force use of 5281 one PCI domain per PCI function 5282 notph [PCIE] If the PCIE_TPH kernel config parameter 5283 is enabled, this kernel boot option can be used 5284 to disable PCIe TLP Processing Hints support 5285 system-wide. 5286 5287 pcie_aspm= [PCIE] Forcibly enable or ignore PCIe Active State Power 5288 Management. 5289 off Don't touch ASPM configuration at all. Leave any 5290 configuration done by firmware unchanged. 5291 force Enable ASPM even on devices that claim not to support it. 5292 WARNING: Forcing ASPM on may cause system lockups. 5293 5294 pcie_ports= [PCIE] PCIe port services handling: 5295 native Use native PCIe services (PME, AER, DPC, PCIe hotplug) 5296 even if the platform doesn't give the OS permission to 5297 use them. This may cause conflicts if the platform 5298 also tries to use these services. 5299 dpc-native Use native PCIe service for DPC only. May 5300 cause conflicts if firmware uses AER or DPC. 5301 compat Disable native PCIe services (PME, AER, DPC, PCIe 5302 hotplug). 5303 5304 pcie_port_pm= [PCIE] PCIe port power management handling: 5305 off Disable power management of all PCIe ports 5306 force Forcibly enable power management of all PCIe ports 5307 5308 pcie_pme= [PCIE,PM] Native PCIe PME signaling options: 5309 nomsi Do not use MSI for native PCIe PME signaling (this makes 5310 all PCIe root ports use INTx for all services). 5311 5312 pcmv= [HW,PCMCIA] BadgePAD 4 5313 5314 pd_ignore_unused 5315 [PM] 5316 Keep all power-domains already enabled by bootloader on, 5317 even if no driver has claimed them. This is useful 5318 for debug and development, but should not be 5319 needed on a platform with proper driver support. 5320 5321 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at 5322 boot time. 5323 Format: { 0 | 1 } 5324 See arch/parisc/kernel/pdc_chassis.c 5325 5326 percpu_alloc= [MM,EARLY] 5327 Select which percpu first chunk allocator to use. 5328 Currently supported values are "embed" and "page". 5329 Archs may support subset or none of the selections. 5330 See comments in mm/percpu.c for details on each 5331 allocator. This parameter is primarily for debugging 5332 and performance comparison. 5333 5334 pirq= [SMP,APIC] Manual mp-table setup 5335 See Documentation/arch/x86/i386/IO-APIC.rst. 5336 5337 plip= [PPT,NET] Parallel port network link 5338 Format: { parport<nr> | timid | 0 } 5339 See also Documentation/admin-guide/parport.rst. 5340 5341 pmtmr= [X86] Manual setup of pmtmr I/O Port. 5342 Override pmtimer IOPort with a hex value. 5343 e.g. pmtmr=0x508 5344 5345 pmu_override= [PPC] Override the PMU. 5346 This option takes over the PMU facility, so it is no 5347 longer usable by perf. Setting this option starts the 5348 PMU counters by setting MMCR0 to 0 (the FC bit is 5349 cleared). If a number is given, then MMCR1 is set to 5350 that number, otherwise (e.g., 'pmu_override=on'), MMCR1 5351 remains 0. 5352 5353 pm_async= [PM] 5354 Format: off 5355 This parameter sets the initial value of the 5356 /sys/power/pm_async sysfs knob at boot time. 5357 If set to "off", disables asynchronous suspend and 5358 resume of devices during system-wide power transitions. 5359 This can be useful on platforms where device 5360 dependencies are not well-defined, or for debugging 5361 power management issues. Asynchronous operations are 5362 enabled by default. 5363 5364 5365 pm_debug_messages [SUSPEND,KNL] 5366 Enable suspend/resume debug messages during boot up. 5367 5368 pnp.debug=1 [PNP] 5369 Enable PNP debug messages (depends on the 5370 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time 5371 via /sys/module/pnp/parameters/debug. We always show 5372 current resource usage; turning this on also shows 5373 possible settings and some assignment information. 5374 5375 pnpacpi= [ACPI] 5376 { off } 5377 5378 pnpbios= [ISAPNP] 5379 { on | off | curr | res | no-curr | no-res } 5380 5381 pnp_reserve_irq= 5382 [ISAPNP] Exclude IRQs for the autoconfiguration 5383 5384 pnp_reserve_dma= 5385 [ISAPNP] Exclude DMAs for the autoconfiguration 5386 5387 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration 5388 Ranges are in pairs (I/O port base and size). 5389 5390 pnp_reserve_mem= 5391 [ISAPNP] Exclude memory regions for the 5392 autoconfiguration. 5393 Ranges are in pairs (memory base and size). 5394 5395 ports= [IP_VS_FTP] IPVS ftp helper module 5396 Default is 21. 5397 Up to 8 (IP_VS_APP_MAX_PORTS) ports 5398 may be specified. 5399 Format: <port>,<port>.... 5400 5401 possible_cpus= [SMP,S390,X86] 5402 Format: <unsigned int> 5403 Set the number of possible CPUs, overriding the 5404 regular discovery mechanisms (such as ACPI/FW, etc). 5405 5406 powersave=off [PPC] This option disables power saving features. 5407 It specifically disables cpuidle and sets the 5408 platform machine description specific power_save 5409 function to NULL. On Idle the CPU just reduces 5410 execution priority. 5411 5412 ppc_strict_facility_enable 5413 [PPC,ENABLE] This option catches any kernel floating point, 5414 Altivec, VSX and SPE outside of regions specifically 5415 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()). 5416 There is some performance impact when enabling this. 5417 5418 ppc_tm= [PPC,EARLY] 5419 Format: {"off"} 5420 Disable Hardware Transactional Memory 5421 5422 preempt= [KNL] 5423 Select preemption mode if you have CONFIG_PREEMPT_DYNAMIC 5424 none - Limited to cond_resched() calls 5425 voluntary - Limited to cond_resched() and might_sleep() calls 5426 full - Any section that isn't explicitly preempt disabled 5427 can be preempted anytime. Tasks will also yield 5428 contended spinlocks (if the critical section isn't 5429 explicitly preempt disabled beyond the lock itself). 5430 lazy - Scheduler controlled. Similar to full but instead 5431 of preempting the task immediately, the task gets 5432 one HZ tick time to yield itself before the 5433 preemption will be forced. One preemption is when the 5434 task returns to user space. 5435 5436 print-fatal-signals= 5437 [KNL] debug: print fatal signals 5438 5439 If enabled, warn about various signal handling 5440 related application anomalies: too many signals, 5441 too many POSIX.1 timers, fatal signals causing a 5442 coredump - etc. 5443 5444 If you hit the warning due to signal overflow, 5445 you might want to try "ulimit -i unlimited". 5446 5447 default: off. 5448 5449 printk.always_kmsg_dump= 5450 Trigger kmsg_dump for cases other than kernel oops or 5451 panics 5452 Format: <bool> (1/Y/y=enable, 0/N/n=disable) 5453 default: disabled 5454 5455 printk.console_no_auto_verbose= 5456 Disable console loglevel raise on oops, panic 5457 or lockdep-detected issues (only if lock debug is on). 5458 With an exception to setups with low baudrate on 5459 serial console, keeping this 0 is a good choice 5460 in order to provide more debug information. 5461 Format: <bool> 5462 default: 0 (auto_verbose is enabled) 5463 5464 printk.debug_non_panic_cpus= 5465 Allows storing messages from non-panic CPUs into 5466 the printk log buffer during panic(). They are 5467 flushed to consoles by the panic-CPU on 5468 a best-effort basis. 5469 Format: <bool> (1/Y/y=enable, 0/N/n=disable) 5470 Default: disabled 5471 5472 printk.devkmsg={on,off,ratelimit} 5473 Control writing to /dev/kmsg. 5474 on - unlimited logging to /dev/kmsg from userspace 5475 off - logging to /dev/kmsg disabled 5476 ratelimit - ratelimit the logging 5477 Default: ratelimit 5478 5479 printk.time= Show timing data prefixed to each printk message line 5480 Format: <bool> (1/Y/y=enable, 0/N/n=disable) 5481 5482 proc_mem.force_override= [KNL] 5483 Format: {always | ptrace | never} 5484 Traditionally /proc/pid/mem allows memory permissions to be 5485 overridden without restrictions. This option may be set to 5486 restrict that. Can be one of: 5487 - 'always': traditional behavior always allows mem overrides. 5488 - 'ptrace': only allow mem overrides for active ptracers. 5489 - 'never': never allow mem overrides. 5490 If not specified, default is the CONFIG_PROC_MEM_* choice. 5491 5492 processor.max_cstate= [HW,ACPI] 5493 Limit processor to maximum C-state 5494 max_cstate=9 overrides any DMI blacklist limit. 5495 5496 processor.nocst [HW,ACPI] 5497 Ignore the _CST method to determine C-states, 5498 instead using the legacy FADT method 5499 5500 profile= [KNL] Enable kernel profiling via /proc/profile 5501 Format: [<profiletype>,]<number> 5502 Param: <profiletype>: "schedule" or "kvm" 5503 [defaults to kernel profiling] 5504 Param: "schedule" - profile schedule points. 5505 Param: "kvm" - profile VM exits. 5506 Param: <number> - step/bucket size as a power of 2 for 5507 statistical time based profiling. 5508 5509 prot_virt= [S390] enable hosting protected virtual machines 5510 isolated from the hypervisor (if hardware supports 5511 that). If enabled, the default kernel base address 5512 might be overridden even when Kernel Address Space 5513 Layout Randomization is disabled. 5514 Format: <bool> 5515 5516 psi= [KNL] Enable or disable pressure stall information 5517 tracking. 5518 Format: <bool> 5519 5520 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to 5521 probe for; one of (bare|imps|exps|lifebook|any). 5522 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports 5523 per second. 5524 psmouse.resetafter= [HW,MOUSE] 5525 Try to reset the device after so many bad packets 5526 (0 = never). 5527 psmouse.resolution= 5528 [HW,MOUSE] Set desired mouse resolution, in dpi. 5529 psmouse.smartscroll= 5530 [HW,MOUSE] Controls Logitech smartscroll autorepeat. 5531 0 = disabled, 1 = enabled (default). 5532 5533 pstore.backend= Specify the name of the pstore backend to use 5534 5535 pti= [X86-64] Control Page Table Isolation of user and 5536 kernel address spaces. Disabling this feature 5537 removes hardening, but improves performance of 5538 system calls and interrupts. 5539 5540 on - unconditionally enable 5541 off - unconditionally disable 5542 auto - kernel detects whether your CPU model is 5543 vulnerable to issues that PTI mitigates 5544 5545 Not specifying this option is equivalent to pti=auto. 5546 5547 pty.legacy_count= 5548 [KNL] Number of legacy pty's. Overwrites compiled-in 5549 default number. 5550 5551 quiet [KNL,EARLY] Disable most log messages 5552 5553 r128= [HW,DRM] 5554 5555 radix_hcall_invalidate=on [PPC/PSERIES] 5556 Disable RADIX GTSE feature and use hcall for TLB 5557 invalidate. 5558 5559 raid= [HW,RAID] 5560 See Documentation/admin-guide/md.rst. 5561 5562 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes 5563 See Documentation/admin-guide/blockdev/ramdisk.rst. 5564 5565 ramdisk_start= [Deprecated,RAM] RAM disk image start address 5566 5567 random.trust_cpu=off 5568 [KNL,EARLY] Disable trusting the use of the CPU's 5569 random number generator (if available) to 5570 initialize the kernel's RNG. 5571 5572 random.trust_bootloader=off 5573 [KNL,EARLY] Disable trusting the use of the a seed 5574 passed by the bootloader (if available) to 5575 initialize the kernel's RNG. 5576 5577 randomize_kstack_offset= 5578 [KNL,EARLY] Enable or disable kernel stack offset 5579 randomization, which provides roughly 5 bits of 5580 entropy, frustrating memory corruption attacks 5581 that depend on stack address determinism or 5582 cross-syscall address exposures. This is only 5583 available on architectures that have defined 5584 CONFIG_HAVE_ARCH_RANDOMIZE_KSTACK_OFFSET. 5585 Format: <bool> (1/Y/y=enable, 0/N/n=disable) 5586 Default is CONFIG_RANDOMIZE_KSTACK_OFFSET_DEFAULT. 5587 5588 ras=option[,option,...] [KNL] RAS-specific options 5589 5590 cec_disable [X86] 5591 Disable the Correctable Errors Collector, 5592 see CONFIG_RAS_CEC help text. 5593 5594 rcu_nocbs[=cpu-list] 5595 [KNL] The optional argument is a cpu list, 5596 as described above. 5597 5598 In kernels built with CONFIG_RCU_NOCB_CPU=y, 5599 enable the no-callback CPU mode, which prevents 5600 such CPUs' callbacks from being invoked in 5601 softirq context. Invocation of such CPUs' RCU 5602 callbacks will instead be offloaded to "rcuox/N" 5603 kthreads created for that purpose, where "x" is 5604 "p" for RCU-preempt, "s" for RCU-sched, and "g" 5605 for the kthreads that mediate grace periods; and 5606 "N" is the CPU number. This reduces OS jitter on 5607 the offloaded CPUs, which can be useful for HPC 5608 and real-time workloads. It can also improve 5609 energy efficiency for asymmetric multiprocessors. 5610 5611 If a cpulist is passed as an argument, the specified 5612 list of CPUs is set to no-callback mode from boot. 5613 5614 Otherwise, if the '=' sign and the cpulist 5615 arguments are omitted, no CPU will be set to 5616 no-callback mode from boot but the mode may be 5617 toggled at runtime via cpusets. 5618 5619 Note that this argument takes precedence over 5620 the CONFIG_RCU_NOCB_CPU_DEFAULT_ALL option. 5621 5622 rcu_nocb_poll [KNL] 5623 Rather than requiring that offloaded CPUs 5624 (specified by rcu_nocbs= above) explicitly 5625 awaken the corresponding "rcuoN" kthreads, 5626 make these kthreads poll for callbacks. 5627 This improves the real-time response for the 5628 offloaded CPUs by relieving them of the need to 5629 wake up the corresponding kthread, but degrades 5630 energy efficiency by requiring that the kthreads 5631 periodically wake up to do the polling. 5632 5633 rcutree.blimit= [KNL] 5634 Set maximum number of finished RCU callbacks to 5635 process in one batch. 5636 5637 rcutree.csd_lock_suppress_rcu_stall= [KNL] 5638 Do only a one-line RCU CPU stall warning when 5639 there is an ongoing too-long CSD-lock wait. 5640 5641 rcutree.do_rcu_barrier= [KNL] 5642 Request a call to rcu_barrier(). This is 5643 throttled so that userspace tests can safely 5644 hammer on the sysfs variable if they so choose. 5645 If triggered before the RCU grace-period machinery 5646 is fully active, this will error out with EAGAIN. 5647 5648 rcutree.dump_tree= [KNL] 5649 Dump the structure of the rcu_node combining tree 5650 out at early boot. This is used for diagnostic 5651 purposes, to verify correct tree setup. 5652 5653 rcutree.gp_cleanup_delay= [KNL] 5654 Set the number of jiffies to delay each step of 5655 RCU grace-period cleanup. 5656 5657 rcutree.gp_init_delay= [KNL] 5658 Set the number of jiffies to delay each step of 5659 RCU grace-period initialization. 5660 5661 rcutree.gp_preinit_delay= [KNL] 5662 Set the number of jiffies to delay each step of 5663 RCU grace-period pre-initialization, that is, 5664 the propagation of recent CPU-hotplug changes up 5665 the rcu_node combining tree. 5666 5667 rcutree.jiffies_till_first_fqs= [KNL] 5668 Set delay from grace-period initialization to 5669 first attempt to force quiescent states. 5670 Units are jiffies, minimum value is zero, 5671 and maximum value is HZ. 5672 5673 rcutree.jiffies_till_next_fqs= [KNL] 5674 Set delay between subsequent attempts to force 5675 quiescent states. Units are jiffies, minimum 5676 value is one, and maximum value is HZ. 5677 5678 rcutree.jiffies_till_sched_qs= [KNL] 5679 Set required age in jiffies for a 5680 given grace period before RCU starts 5681 soliciting quiescent-state help from 5682 rcu_note_context_switch() and cond_resched(). 5683 If not specified, the kernel will calculate 5684 a value based on the most recent settings 5685 of rcutree.jiffies_till_first_fqs 5686 and rcutree.jiffies_till_next_fqs. 5687 This calculated value may be viewed in 5688 rcutree.jiffies_to_sched_qs. Any attempt to set 5689 rcutree.jiffies_to_sched_qs will be cheerfully 5690 overwritten. 5691 5692 rcutree.kthread_prio= [KNL,BOOT] 5693 Set the SCHED_FIFO priority of the RCU per-CPU 5694 kthreads (rcuc/N). This value is also used for 5695 the priority of the RCU boost threads (rcub/N) 5696 and for the RCU grace-period kthreads (rcu_bh, 5697 rcu_preempt, and rcu_sched). If RCU_BOOST is 5698 set, valid values are 1-99 and the default is 1 5699 (the least-favored priority). Otherwise, when 5700 RCU_BOOST is not set, valid values are 0-99 and 5701 the default is zero (non-realtime operation). 5702 When RCU_NOCB_CPU is set, also adjust the 5703 priority of NOCB callback kthreads. 5704 5705 rcutree.nocb_nobypass_lim_per_jiffy= [KNL] 5706 On callback-offloaded (rcu_nocbs) CPUs, 5707 RCU reduces the lock contention that would 5708 otherwise be caused by callback floods through 5709 use of the ->nocb_bypass list. However, in the 5710 common non-flooded case, RCU queues directly to 5711 the main ->cblist in order to avoid the extra 5712 overhead of the ->nocb_bypass list and its lock. 5713 But if there are too many callbacks queued during 5714 a single jiffy, RCU pre-queues the callbacks into 5715 the ->nocb_bypass queue. The definition of "too 5716 many" is supplied by this kernel boot parameter. 5717 5718 rcutree.nohz_full_patience_delay= [KNL] 5719 On callback-offloaded (rcu_nocbs) CPUs, avoid 5720 disturbing RCU unless the grace period has 5721 reached the specified age in milliseconds. 5722 Defaults to zero. Large values will be capped 5723 at five seconds. All values will be rounded down 5724 to the nearest value representable by jiffies. 5725 5726 rcutree.qhimark= [KNL] 5727 Set threshold of queued RCU callbacks beyond which 5728 batch limiting is disabled. 5729 5730 rcutree.qlowmark= [KNL] 5731 Set threshold of queued RCU callbacks below which 5732 batch limiting is re-enabled. 5733 5734 rcutree.qovld= [KNL] 5735 Set threshold of queued RCU callbacks beyond which 5736 RCU's force-quiescent-state scan will aggressively 5737 enlist help from cond_resched() and sched IPIs to 5738 help CPUs more quickly reach quiescent states. 5739 Set to less than zero to make this be set based 5740 on rcutree.qhimark at boot time and to zero to 5741 disable more aggressive help enlistment. 5742 5743 rcutree.rcu_delay_page_cache_fill_msec= [KNL] 5744 Set the page-cache refill delay (in milliseconds) 5745 in response to low-memory conditions. The range 5746 of permitted values is in the range 0:100000. 5747 5748 rcutree.rcu_divisor= [KNL] 5749 Set the shift-right count to use to compute 5750 the callback-invocation batch limit bl from 5751 the number of callbacks queued on this CPU. 5752 The result will be bounded below by the value of 5753 the rcutree.blimit kernel parameter. Every bl 5754 callbacks, the softirq handler will exit in 5755 order to allow the CPU to do other work. 5756 5757 Please note that this callback-invocation batch 5758 limit applies only to non-offloaded callback 5759 invocation. Offloaded callbacks are instead 5760 invoked in the context of an rcuoc kthread, which 5761 scheduler will preempt as it does any other task. 5762 5763 rcutree.rcu_fanout_exact= [KNL] 5764 Disable autobalancing of the rcu_node combining 5765 tree. This is used by rcutorture, and might 5766 possibly be useful for architectures having high 5767 cache-to-cache transfer latencies. 5768 5769 rcutree.rcu_fanout_leaf= [KNL] 5770 Change the number of CPUs assigned to each 5771 leaf rcu_node structure. Useful for very 5772 large systems, which will choose the value 64, 5773 and for NUMA systems with large remote-access 5774 latencies, which will choose a value aligned 5775 with the appropriate hardware boundaries. 5776 5777 rcutree.rcu_min_cached_objs= [KNL] 5778 Minimum number of objects which are cached and 5779 maintained per one CPU. Object size is equal 5780 to PAGE_SIZE. The cache allows to reduce the 5781 pressure to page allocator, also it makes the 5782 whole algorithm to behave better in low memory 5783 condition. 5784 5785 rcutree.rcu_nocb_gp_stride= [KNL] 5786 Set the number of NOCB callback kthreads in 5787 each group, which defaults to the square root 5788 of the number of CPUs. Larger numbers reduce 5789 the wakeup overhead on the global grace-period 5790 kthread, but increases that same overhead on 5791 each group's NOCB grace-period kthread. 5792 5793 rcutree.rcu_kick_kthreads= [KNL] 5794 Cause the grace-period kthread to get an extra 5795 wake_up() if it sleeps three times longer than 5796 it should at force-quiescent-state time. 5797 This wake_up() will be accompanied by a 5798 WARN_ONCE() splat and an ftrace_dump(). 5799 5800 rcutree.rcu_resched_ns= [KNL] 5801 Limit the time spend invoking a batch of RCU 5802 callbacks to the specified number of nanoseconds. 5803 By default, this limit is checked only once 5804 every 32 callbacks in order to limit the pain 5805 inflicted by local_clock() overhead. 5806 5807 rcutree.rcu_unlock_delay= [KNL] 5808 In CONFIG_RCU_STRICT_GRACE_PERIOD=y kernels, 5809 this specifies an rcu_read_unlock()-time delay 5810 in microseconds. This defaults to zero. 5811 Larger delays increase the probability of 5812 catching RCU pointer leaks, that is, buggy use 5813 of RCU-protected pointers after the relevant 5814 rcu_read_unlock() has completed. 5815 5816 rcutree.sysrq_rcu= [KNL] 5817 Commandeer a sysrq key to dump out Tree RCU's 5818 rcu_node tree with an eye towards determining 5819 why a new grace period has not yet started. 5820 5821 rcutree.use_softirq= [KNL] 5822 If set to zero, move all RCU_SOFTIRQ processing to 5823 per-CPU rcuc kthreads. Defaults to a non-zero 5824 value, meaning that RCU_SOFTIRQ is used by default. 5825 Specify rcutree.use_softirq=0 to use rcuc kthreads. 5826 5827 But note that CONFIG_PREEMPT_RT=y kernels disable 5828 this kernel boot parameter, forcibly setting it 5829 to zero. 5830 5831 rcutree.enable_rcu_lazy= [KNL] 5832 To save power, batch RCU callbacks and flush after 5833 delay, memory pressure or callback list growing too 5834 big. 5835 5836 rcutree.rcu_normal_wake_from_gp= [KNL] 5837 Reduces a latency of synchronize_rcu() call. This approach 5838 maintains its own track of synchronize_rcu() callers, so it 5839 does not interact with regular callbacks because it does not 5840 use a call_rcu[_hurry]() path. Please note, this is for a 5841 normal grace period. 5842 5843 How to disable it: 5844 5845 echo 0 > /sys/module/rcutree/parameters/rcu_normal_wake_from_gp 5846 or pass a boot parameter "rcutree.rcu_normal_wake_from_gp=0" 5847 5848 Default is 1 if it is not explicitly disabled by the boot parameter 5849 passing 0. 5850 5851 rcuscale.gp_async= [KNL] 5852 Measure performance of asynchronous 5853 grace-period primitives such as call_rcu(). 5854 5855 rcuscale.gp_async_max= [KNL] 5856 Specify the maximum number of outstanding 5857 callbacks per writer thread. When a writer 5858 thread exceeds this limit, it invokes the 5859 corresponding flavor of rcu_barrier() to allow 5860 previously posted callbacks to drain. 5861 5862 rcuscale.gp_exp= [KNL] 5863 Measure performance of expedited synchronous 5864 grace-period primitives. 5865 5866 rcuscale.holdoff= [KNL] 5867 Set test-start holdoff period. The purpose of 5868 this parameter is to delay the start of the 5869 test until boot completes in order to avoid 5870 interference. 5871 5872 rcuscale.kfree_by_call_rcu= [KNL] 5873 In kernels built with CONFIG_RCU_LAZY=y, test 5874 call_rcu() instead of kfree_rcu(). 5875 5876 rcuscale.kfree_mult= [KNL] 5877 Instead of allocating an object of size kfree_obj, 5878 allocate one of kfree_mult * sizeof(kfree_obj). 5879 Defaults to 1. 5880 5881 rcuscale.kfree_rcu_test= [KNL] 5882 Set to measure performance of kfree_rcu() flooding. 5883 5884 rcuscale.kfree_rcu_test_double= [KNL] 5885 Test the double-argument variant of kfree_rcu(). 5886 If this parameter has the same value as 5887 rcuscale.kfree_rcu_test_single, both the single- 5888 and double-argument variants are tested. 5889 5890 rcuscale.kfree_rcu_test_single= [KNL] 5891 Test the single-argument variant of kfree_rcu(). 5892 If this parameter has the same value as 5893 rcuscale.kfree_rcu_test_double, both the single- 5894 and double-argument variants are tested. 5895 5896 rcuscale.kfree_nthreads= [KNL] 5897 The number of threads running loops of kfree_rcu(). 5898 5899 rcuscale.kfree_alloc_num= [KNL] 5900 Number of allocations and frees done in an iteration. 5901 5902 rcuscale.kfree_loops= [KNL] 5903 Number of loops doing rcuscale.kfree_alloc_num number 5904 of allocations and frees. 5905 5906 rcuscale.minruntime= [KNL] 5907 Set the minimum test run time in seconds. This 5908 does not affect the data-collection interval, 5909 but instead allows better measurement of things 5910 like CPU consumption. 5911 5912 rcuscale.nreaders= [KNL] 5913 Set number of RCU readers. The value -1 selects 5914 N, where N is the number of CPUs. A value 5915 "n" less than -1 selects N-n+1, where N is again 5916 the number of CPUs. For example, -2 selects N 5917 (the number of CPUs), -3 selects N+1, and so on. 5918 A value of "n" less than or equal to -N selects 5919 a single reader. 5920 5921 rcuscale.nwriters= [KNL] 5922 Set number of RCU writers. The values operate 5923 the same as for rcuscale.nreaders. 5924 N, where N is the number of CPUs 5925 5926 rcuscale.scale_type= [KNL] 5927 Specify the RCU implementation to test. 5928 5929 rcuscale.shutdown= [KNL] 5930 Shut the system down after performance tests 5931 complete. This is useful for hands-off automated 5932 testing. 5933 5934 rcuscale.verbose= [KNL] 5935 Enable additional printk() statements. 5936 5937 rcuscale.writer_holdoff= [KNL] 5938 Write-side holdoff between grace periods, 5939 in microseconds. The default of zero says 5940 no holdoff. 5941 5942 rcuscale.writer_holdoff_jiffies= [KNL] 5943 Additional write-side holdoff between grace 5944 periods, but in jiffies. The default of zero 5945 says no holdoff. 5946 5947 rcutorture.fqs_duration= [KNL] 5948 Set duration of force_quiescent_state bursts 5949 in microseconds. 5950 5951 rcutorture.fqs_holdoff= [KNL] 5952 Set holdoff time within force_quiescent_state bursts 5953 in microseconds. 5954 5955 rcutorture.fqs_stutter= [KNL] 5956 Set wait time between force_quiescent_state bursts 5957 in seconds. 5958 5959 rcutorture.fwd_progress= [KNL] 5960 Specifies the number of kthreads to be used 5961 for RCU grace-period forward-progress testing 5962 for the types of RCU supporting this notion. 5963 Defaults to 1 kthread, values less than zero or 5964 greater than the number of CPUs cause the number 5965 of CPUs to be used. 5966 5967 rcutorture.fwd_progress_div= [KNL] 5968 Specify the fraction of a CPU-stall-warning 5969 period to do tight-loop forward-progress testing. 5970 5971 rcutorture.fwd_progress_holdoff= [KNL] 5972 Number of seconds to wait between successive 5973 forward-progress tests. 5974 5975 rcutorture.fwd_progress_need_resched= [KNL] 5976 Enclose cond_resched() calls within checks for 5977 need_resched() during tight-loop forward-progress 5978 testing. 5979 5980 rcutorture.gp_cond= [KNL] 5981 Use conditional/asynchronous update-side 5982 normal-grace-period primitives, if available. 5983 5984 rcutorture.gp_cond_exp= [KNL] 5985 Use conditional/asynchronous update-side 5986 expedited-grace-period primitives, if available. 5987 5988 rcutorture.gp_cond_full= [KNL] 5989 Use conditional/asynchronous update-side 5990 normal-grace-period primitives that also take 5991 concurrent expedited grace periods into account, 5992 if available. 5993 5994 rcutorture.gp_cond_exp_full= [KNL] 5995 Use conditional/asynchronous update-side 5996 expedited-grace-period primitives that also take 5997 concurrent normal grace periods into account, 5998 if available. 5999 6000 rcutorture.gp_cond_wi= [KNL] 6001 Nominal wait interval for normal conditional 6002 grace periods (specified by rcutorture's 6003 gp_cond and gp_cond_full module parameters), 6004 in microseconds. The actual wait interval will 6005 be randomly selected to nanosecond granularity up 6006 to this wait interval. Defaults to 16 jiffies, 6007 for example, 16,000 microseconds on a system 6008 with HZ=1000. 6009 6010 rcutorture.gp_cond_wi_exp= [KNL] 6011 Nominal wait interval for expedited conditional 6012 grace periods (specified by rcutorture's 6013 gp_cond_exp and gp_cond_exp_full module 6014 parameters), in microseconds. The actual wait 6015 interval will be randomly selected to nanosecond 6016 granularity up to this wait interval. Defaults to 6017 128 microseconds. 6018 6019 rcutorture.gp_exp= [KNL] 6020 Use expedited update-side primitives, if available. 6021 6022 rcutorture.gp_normal= [KNL] 6023 Use normal (non-expedited) asynchronous 6024 update-side primitives, if available. 6025 6026 rcutorture.gp_poll= [KNL] 6027 Use polled update-side normal-grace-period 6028 primitives, if available. 6029 6030 rcutorture.gp_poll_exp= [KNL] 6031 Use polled update-side expedited-grace-period 6032 primitives, if available. 6033 6034 rcutorture.gp_poll_full= [KNL] 6035 Use polled update-side normal-grace-period 6036 primitives that also take concurrent expedited 6037 grace periods into account, if available. 6038 6039 rcutorture.gp_poll_exp_full= [KNL] 6040 Use polled update-side expedited-grace-period 6041 primitives that also take concurrent normal 6042 grace periods into account, if available. 6043 6044 rcutorture.gp_poll_wi= [KNL] 6045 Nominal wait interval for normal conditional 6046 grace periods (specified by rcutorture's 6047 gp_poll and gp_poll_full module parameters), 6048 in microseconds. The actual wait interval will 6049 be randomly selected to nanosecond granularity up 6050 to this wait interval. Defaults to 16 jiffies, 6051 for example, 16,000 microseconds on a system 6052 with HZ=1000. 6053 6054 rcutorture.gp_poll_wi_exp= [KNL] 6055 Nominal wait interval for expedited conditional 6056 grace periods (specified by rcutorture's 6057 gp_poll_exp and gp_poll_exp_full module 6058 parameters), in microseconds. The actual wait 6059 interval will be randomly selected to nanosecond 6060 granularity up to this wait interval. Defaults to 6061 128 microseconds. 6062 6063 rcutorture.gp_sync= [KNL] 6064 Use normal (non-expedited) synchronous 6065 update-side primitives, if available. If all 6066 of rcutorture.gp_cond=, rcutorture.gp_exp=, 6067 rcutorture.gp_normal=, and rcutorture.gp_sync= 6068 are zero, rcutorture acts as if is interpreted 6069 they are all non-zero. 6070 6071 rcutorture.gpwrap_lag= [KNL] 6072 Enable grace-period wrap lag testing. Setting 6073 to false prevents the gpwrap lag test from 6074 running. Default is true. 6075 6076 rcutorture.gpwrap_lag_gps= [KNL] 6077 Set the value for grace-period wrap lag during 6078 active lag testing periods. This controls how many 6079 grace periods differences we tolerate between 6080 rdp and rnp's gp_seq before setting overflow flag. 6081 The default is always set to 8. 6082 6083 rcutorture.gpwrap_lag_cycle_mins= [KNL] 6084 Set the total cycle duration for gpwrap lag 6085 testing in minutes. This is the total time for 6086 one complete cycle of active and inactive 6087 testing periods. Default is 30 minutes. 6088 6089 rcutorture.gpwrap_lag_active_mins= [KNL] 6090 Set the duration for which gpwrap lag is active 6091 within each cycle, in minutes. During this time, 6092 the grace-period wrap lag will be set to the 6093 value specified by gpwrap_lag_gps. Default is 6094 5 minutes. 6095 6096 rcutorture.irqreader= [KNL] 6097 Run RCU readers from irq handlers, or, more 6098 accurately, from a timer handler. Not all RCU 6099 flavors take kindly to this sort of thing. 6100 6101 rcutorture.leakpointer= [KNL] 6102 Leak an RCU-protected pointer out of the reader. 6103 This can of course result in splats, and is 6104 intended to test the ability of things like 6105 CONFIG_RCU_STRICT_GRACE_PERIOD=y to detect 6106 such leaks. 6107 6108 rcutorture.n_barrier_cbs= [KNL] 6109 Set callbacks/threads for rcu_barrier() testing. 6110 6111 rcutorture.nfakewriters= [KNL] 6112 Set number of concurrent RCU writers. These just 6113 stress RCU, they don't participate in the actual 6114 test, hence the "fake". 6115 6116 rcutorture.nocbs_nthreads= [KNL] 6117 Set number of RCU callback-offload togglers. 6118 Zero (the default) disables toggling. 6119 6120 rcutorture.nocbs_toggle= [KNL] 6121 Set the delay in milliseconds between successive 6122 callback-offload toggling attempts. 6123 6124 rcutorture.nreaders= [KNL] 6125 Set number of RCU readers. The value -1 selects 6126 N-1, where N is the number of CPUs. A value 6127 "n" less than -1 selects N-n-2, where N is again 6128 the number of CPUs. For example, -2 selects N 6129 (the number of CPUs), -3 selects N+1, and so on. 6130 6131 rcutorture.object_debug= [KNL] 6132 Enable debug-object double-call_rcu() testing. 6133 6134 rcutorture.onoff_holdoff= [KNL] 6135 Set time (s) after boot for CPU-hotplug testing. 6136 6137 rcutorture.onoff_interval= [KNL] 6138 Set time (jiffies) between CPU-hotplug operations, 6139 or zero to disable CPU-hotplug testing. 6140 6141 rcutorture.preempt_duration= [KNL] 6142 Set duration (in milliseconds) of preemptions 6143 by a high-priority FIFO real-time task. Set to 6144 zero (the default) to disable. The CPUs to 6145 preempt are selected randomly from the set that 6146 are online at a given point in time. Races with 6147 CPUs going offline are ignored, with that attempt 6148 at preemption skipped. 6149 6150 rcutorture.preempt_interval= [KNL] 6151 Set interval (in milliseconds, defaulting to one 6152 second) between preemptions by a high-priority 6153 FIFO real-time task. This delay is mediated 6154 by an hrtimer and is further fuzzed to avoid 6155 inadvertent synchronizations. 6156 6157 rcutorture.read_exit_burst= [KNL] 6158 The number of times in a given read-then-exit 6159 episode that a set of read-then-exit kthreads 6160 is spawned. 6161 6162 rcutorture.read_exit_delay= [KNL] 6163 The delay, in seconds, between successive 6164 read-then-exit testing episodes. 6165 6166 rcutorture.reader_flavor= [KNL] 6167 A bit mask indicating which readers to use. 6168 If there is more than one bit set, the readers 6169 are entered from low-order bit up, and are 6170 exited in the opposite order. For SRCU, the 6171 0x1 bit is normal readers, 0x2 NMI-safe readers, 6172 and 0x4 light-weight readers. 6173 6174 rcutorture.shuffle_interval= [KNL] 6175 Set task-shuffle interval (s). Shuffling tasks 6176 allows some CPUs to go into dyntick-idle mode 6177 during the rcutorture test. 6178 6179 rcutorture.shutdown_secs= [KNL] 6180 Set time (s) after boot system shutdown. This 6181 is useful for hands-off automated testing. 6182 6183 rcutorture.stall_cpu= [KNL] 6184 Duration of CPU stall (s) to test RCU CPU stall 6185 warnings, zero to disable. 6186 6187 rcutorture.stall_cpu_block= [KNL] 6188 Sleep while stalling if set. This will result 6189 in warnings from preemptible RCU in addition to 6190 any other stall-related activity. Note that 6191 in kernels built with CONFIG_PREEMPTION=n and 6192 CONFIG_PREEMPT_COUNT=y, this parameter will 6193 cause the CPU to pass through a quiescent state. 6194 Given CONFIG_PREEMPTION=n, this will suppress 6195 RCU CPU stall warnings, but will instead result 6196 in scheduling-while-atomic splats. 6197 6198 Use of this module parameter results in splats. 6199 6200 6201 rcutorture.stall_cpu_holdoff= [KNL] 6202 Time to wait (s) after boot before inducing stall. 6203 6204 rcutorture.stall_cpu_irqsoff= [KNL] 6205 Disable interrupts while stalling if set, but only 6206 on the first stall in the set. 6207 6208 rcutorture.stall_cpu_repeat= [KNL] 6209 Number of times to repeat the stall sequence, 6210 so that rcutorture.stall_cpu_repeat=3 will result 6211 in four stall sequences. 6212 6213 rcutorture.stall_gp_kthread= [KNL] 6214 Duration (s) of forced sleep within RCU 6215 grace-period kthread to test RCU CPU stall 6216 warnings, zero to disable. If both stall_cpu 6217 and stall_gp_kthread are specified, the 6218 kthread is starved first, then the CPU. 6219 6220 rcutorture.stat_interval= [KNL] 6221 Time (s) between statistics printk()s. 6222 6223 rcutorture.stutter= [KNL] 6224 Time (s) to stutter testing, for example, specifying 6225 five seconds causes the test to run for five seconds, 6226 wait for five seconds, and so on. This tests RCU's 6227 ability to transition abruptly to and from idle. 6228 6229 rcutorture.test_boost= [KNL] 6230 Test RCU priority boosting? 0=no, 1=maybe, 2=yes. 6231 "Maybe" means test if the RCU implementation 6232 under test support RCU priority boosting. 6233 6234 rcutorture.test_boost_duration= [KNL] 6235 Duration (s) of each individual boost test. 6236 6237 rcutorture.test_boost_holdoff= [KNL] 6238 Holdoff time (s) from start of test to the start 6239 of RCU priority-boost testing. Defaults to zero, 6240 that is, no holdoff. 6241 6242 rcutorture.test_boost_interval= [KNL] 6243 Interval (s) between each boost test. 6244 6245 rcutorture.test_no_idle_hz= [KNL] 6246 Test RCU's dyntick-idle handling. See also the 6247 rcutorture.shuffle_interval parameter. 6248 6249 rcutorture.torture_type= [KNL] 6250 Specify the RCU implementation to test. 6251 6252 rcutorture.verbose= [KNL] 6253 Enable additional printk() statements. 6254 6255 rcupdate.rcu_cpu_stall_ftrace_dump= [KNL] 6256 Dump ftrace buffer after reporting RCU CPU 6257 stall warning. 6258 6259 rcupdate.rcu_cpu_stall_notifiers= [KNL] 6260 Provide RCU CPU stall notifiers, but see the 6261 warnings in the RCU_CPU_STALL_NOTIFIER Kconfig 6262 option's help text. TL;DR: You almost certainly 6263 do not want rcupdate.rcu_cpu_stall_notifiers. 6264 6265 rcupdate.rcu_cpu_stall_suppress= [KNL] 6266 Suppress RCU CPU stall warning messages. 6267 6268 rcupdate.rcu_cpu_stall_suppress_at_boot= [KNL] 6269 Suppress RCU CPU stall warning messages and 6270 rcutorture writer stall warnings that occur 6271 during early boot, that is, during the time 6272 before the init task is spawned. 6273 6274 rcupdate.rcu_cpu_stall_timeout= [KNL] 6275 Set timeout for RCU CPU stall warning messages. 6276 The value is in seconds and the maximum allowed 6277 value is 300 seconds. 6278 6279 rcupdate.rcu_exp_cpu_stall_timeout= [KNL] 6280 Set timeout for expedited RCU CPU stall warning 6281 messages. The value is in milliseconds 6282 and the maximum allowed value is 21000 6283 milliseconds. Please note that this value is 6284 adjusted to an arch timer tick resolution. 6285 Setting this to zero causes the value from 6286 rcupdate.rcu_cpu_stall_timeout to be used (after 6287 conversion from seconds to milliseconds). 6288 6289 rcupdate.rcu_cpu_stall_cputime= [KNL] 6290 Provide statistics on the cputime and count of 6291 interrupts and tasks during the sampling period. For 6292 multiple continuous RCU stalls, all sampling periods 6293 begin at half of the first RCU stall timeout. 6294 6295 rcupdate.rcu_exp_stall_task_details= [KNL] 6296 Print stack dumps of any tasks blocking the 6297 current expedited RCU grace period during an 6298 expedited RCU CPU stall warning. 6299 6300 rcupdate.rcu_expedited= [KNL] 6301 Use expedited grace-period primitives, for 6302 example, synchronize_rcu_expedited() instead 6303 of synchronize_rcu(). This reduces latency, 6304 but can increase CPU utilization, degrade 6305 real-time latency, and degrade energy efficiency. 6306 No effect on CONFIG_TINY_RCU kernels. 6307 6308 rcupdate.rcu_normal= [KNL] 6309 Use only normal grace-period primitives, 6310 for example, synchronize_rcu() instead of 6311 synchronize_rcu_expedited(). This improves 6312 real-time latency, CPU utilization, and 6313 energy efficiency, but can expose users to 6314 increased grace-period latency. This parameter 6315 overrides rcupdate.rcu_expedited. No effect on 6316 CONFIG_TINY_RCU kernels. 6317 6318 rcupdate.rcu_normal_after_boot= [KNL] 6319 Once boot has completed (that is, after 6320 rcu_end_inkernel_boot() has been invoked), use 6321 only normal grace-period primitives. No effect 6322 on CONFIG_TINY_RCU kernels. 6323 6324 But note that CONFIG_PREEMPT_RT=y kernels enables 6325 this kernel boot parameter, forcibly setting 6326 it to the value one, that is, converting any 6327 post-boot attempt at an expedited RCU grace 6328 period to instead use normal non-expedited 6329 grace-period processing. 6330 6331 rcupdate.rcu_task_collapse_lim= [KNL] 6332 Set the maximum number of callbacks present 6333 at the beginning of a grace period that allows 6334 the RCU Tasks flavors to collapse back to using 6335 a single callback queue. This switching only 6336 occurs when rcupdate.rcu_task_enqueue_lim is 6337 set to the default value of -1. 6338 6339 rcupdate.rcu_task_contend_lim= [KNL] 6340 Set the minimum number of callback-queuing-time 6341 lock-contention events per jiffy required to 6342 cause the RCU Tasks flavors to switch to per-CPU 6343 callback queuing. This switching only occurs 6344 when rcupdate.rcu_task_enqueue_lim is set to 6345 the default value of -1. 6346 6347 rcupdate.rcu_task_enqueue_lim= [KNL] 6348 Set the number of callback queues to use for the 6349 RCU Tasks family of RCU flavors. The default 6350 of -1 allows this to be automatically (and 6351 dynamically) adjusted. This parameter is intended 6352 for use in testing. 6353 6354 rcupdate.rcu_task_lazy_lim= [KNL] 6355 Number of callbacks on a given CPU that will 6356 cancel laziness on that CPU. Use -1 to disable 6357 cancellation of laziness, but be advised that 6358 doing so increases the danger of OOM due to 6359 callback flooding. 6360 6361 rcupdate.rcu_task_stall_info= [KNL] 6362 Set initial timeout in jiffies for RCU task stall 6363 informational messages, which give some indication 6364 of the problem for those not patient enough to 6365 wait for ten minutes. Informational messages are 6366 only printed prior to the stall-warning message 6367 for a given grace period. Disable with a value 6368 less than or equal to zero. Defaults to ten 6369 seconds. A change in value does not take effect 6370 until the beginning of the next grace period. 6371 6372 rcupdate.rcu_task_stall_info_mult= [KNL] 6373 Multiplier for time interval between successive 6374 RCU task stall informational messages for a given 6375 RCU tasks grace period. This value is clamped 6376 to one through ten, inclusive. It defaults to 6377 the value three, so that the first informational 6378 message is printed 10 seconds into the grace 6379 period, the second at 40 seconds, the third at 6380 160 seconds, and then the stall warning at 600 6381 seconds would prevent a fourth at 640 seconds. 6382 6383 rcupdate.rcu_task_stall_timeout= [KNL] 6384 Set timeout in jiffies for RCU task stall 6385 warning messages. Disable with a value less 6386 than or equal to zero. Defaults to ten minutes. 6387 A change in value does not take effect until 6388 the beginning of the next grace period. 6389 6390 rcupdate.rcu_tasks_lazy_ms= [KNL] 6391 Set timeout in milliseconds RCU Tasks asynchronous 6392 callback batching for call_rcu_tasks(). 6393 A negative value will take the default. A value 6394 of zero will disable batching. Batching is 6395 always disabled for synchronize_rcu_tasks(). 6396 6397 rcupdate.rcu_self_test= [KNL] 6398 Run the RCU early boot self tests 6399 6400 rdinit= [KNL] 6401 Format: <full_path> 6402 Run specified binary instead of /init from the ramdisk, 6403 used for early userspace startup. See initrd. 6404 6405 rdrand= [X86,EARLY] 6406 force - Override the decision by the kernel to hide the 6407 advertisement of RDRAND support (this affects 6408 certain AMD processors because of buggy BIOS 6409 support, specifically around the suspend/resume 6410 path). 6411 6412 rdt= [HW,X86,RDT] 6413 Turn on/off individual RDT features. List is: 6414 cmt, mbmtotal, mbmlocal, l3cat, l3cdp, l2cat, l2cdp, 6415 mba, smba, bmec, abmc, sdciae, energy[:guid], 6416 perf[:guid]. 6417 E.g. to turn on cmt and turn off mba use: 6418 rdt=cmt,!mba 6419 To turn off all energy telemetry monitoring and ensure that 6420 perf telemetry monitoring associated with guid 0x12345 6421 is enabled use: 6422 rdt=!energy,perf:0x12345 6423 6424 reboot= [KNL] 6425 Format (x86 or x86_64): 6426 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] | d[efault] \ 6427 [[,]s[mp]#### \ 6428 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \ 6429 [[,]f[orce] 6430 Where reboot_mode is one of warm (soft) or cold (hard) or gpio 6431 (prefix with 'panic_' to set mode for panic 6432 reboot only), 6433 reboot_type is one of bios, acpi, kbd, triple, efi, or pci, 6434 reboot_force is either force or not specified, 6435 reboot_cpu is s[mp]#### with #### being the processor 6436 to be used for rebooting. 6437 6438 acpi 6439 Use the ACPI RESET_REG in the FADT. If ACPI is not 6440 configured or the ACPI reset does not work, the reboot 6441 path attempts the reset using the keyboard controller. 6442 6443 bios 6444 Use the CPU reboot vector for warm reset 6445 6446 cold 6447 Set the cold reboot flag 6448 6449 default 6450 There are some built-in platform specific "quirks" 6451 - you may see: "reboot: <name> series board detected. 6452 Selecting <type> for reboots." In the case where you 6453 think the quirk is in error (e.g. you have newer BIOS, 6454 or newer board) using this option will ignore the 6455 built-in quirk table, and use the generic default 6456 reboot actions. 6457 6458 efi 6459 Use efi reset_system runtime service. If EFI is not 6460 configured or the EFI reset does not work, the reboot 6461 path attempts the reset using the keyboard controller. 6462 6463 force 6464 Don't stop other CPUs on reboot. This can make reboot 6465 more reliable in some cases. 6466 6467 kbd 6468 Use the keyboard controller. cold reset (default) 6469 6470 pci 6471 Use a write to the PCI config space register 0xcf9 to 6472 trigger reboot. 6473 6474 triple 6475 Force a triple fault (init) 6476 6477 warm 6478 Don't set the cold reboot flag 6479 6480 Using warm reset will be much faster especially on big 6481 memory systems because the BIOS will not go through 6482 the memory check. Disadvantage is that not all 6483 hardware will be completely reinitialized on reboot so 6484 there may be boot problems on some systems. 6485 6486 6487 refscale.holdoff= [KNL] 6488 Set test-start holdoff period. The purpose of 6489 this parameter is to delay the start of the 6490 test until boot completes in order to avoid 6491 interference. 6492 6493 refscale.lookup_instances= [KNL] 6494 Number of data elements to use for the forms of 6495 SLAB_TYPESAFE_BY_RCU testing. A negative number 6496 is negated and multiplied by nr_cpu_ids, while 6497 zero specifies nr_cpu_ids. 6498 6499 refscale.loops= [KNL] 6500 Set the number of loops over the synchronization 6501 primitive under test. Increasing this number 6502 reduces noise due to loop start/end overhead, 6503 but the default has already reduced the per-pass 6504 noise to a handful of picoseconds on ca. 2020 6505 x86 laptops. 6506 6507 refscale.nreaders= [KNL] 6508 Set number of readers. The default value of -1 6509 selects N, where N is roughly 75% of the number 6510 of CPUs. A value of zero is an interesting choice. 6511 6512 refscale.nruns= [KNL] 6513 Set number of runs, each of which is dumped onto 6514 the console log. 6515 6516 refscale.readdelay= [KNL] 6517 Set the read-side critical-section duration, 6518 measured in microseconds. 6519 6520 refscale.scale_type= [KNL] 6521 Specify the read-protection implementation to test. 6522 6523 refscale.shutdown= [KNL] 6524 Shut down the system at the end of the performance 6525 test. This defaults to 1 (shut it down) when 6526 refscale is built into the kernel and to 0 (leave 6527 it running) when refscale is built as a module. 6528 6529 refscale.verbose= [KNL] 6530 Enable additional printk() statements. 6531 6532 refscale.verbose_batched= [KNL] 6533 Batch the additional printk() statements. If zero 6534 (the default) or negative, print everything. Otherwise, 6535 print every Nth verbose statement, where N is the value 6536 specified. 6537 6538 regulator_ignore_unused 6539 [REGULATOR] 6540 Prevents regulator framework from disabling regulators 6541 that are unused, due no driver claiming them. This may 6542 be useful for debug and development, but should not be 6543 needed on a platform with proper driver support. 6544 6545 relax_domain_level= 6546 [KNL, SMP] Set scheduler's default relax_domain_level. 6547 See Documentation/admin-guide/cgroup-v1/cpusets.rst. 6548 6549 reserve= [KNL,BUGS] Force kernel to ignore I/O ports or memory 6550 Format: <base1>,<size1>[,<base2>,<size2>,...] 6551 Reserve I/O ports or memory so the kernel won't use 6552 them. If <base> is less than 0x10000, the region 6553 is assumed to be I/O ports; otherwise it is memory. 6554 6555 reserve_mem= [RAM] 6556 Format: nn[KMG]:<align>:<label> 6557 Reserve physical memory and label it with a name that 6558 other subsystems can use to access it. This is typically 6559 used for systems that do not wipe the RAM, and this command 6560 line will try to reserve the same physical memory on 6561 soft reboots. Note, it is not guaranteed to be the same 6562 location. For example, if anything about the system changes 6563 or if booting a different kernel. It can also fail if KASLR 6564 places the kernel at the location of where the RAM reservation 6565 was from a previous boot, the new reservation will be at a 6566 different location. 6567 Any subsystem using this feature must add a way to verify 6568 that the contents of the physical memory is from a previous 6569 boot, as there may be cases where the memory will not be 6570 located at the same location. 6571 6572 The format is size:align:label for example, to request 6573 12 megabytes of 4096 alignment for ramoops: 6574 6575 reserve_mem=12M:4096:oops ramoops.mem_name=oops 6576 6577 reservetop= [X86-32,EARLY] 6578 Format: nn[KMG] 6579 Reserves a hole at the top of the kernel virtual 6580 address space. 6581 6582 reset_devices [KNL] Force drivers to reset the underlying device 6583 during initialization. 6584 6585 resume= [SWSUSP] 6586 Specify the partition device for software suspend 6587 Format: 6588 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>} 6589 6590 resume_offset= [SWSUSP] 6591 Specify the offset from the beginning of the partition 6592 given by "resume=" at which the swap header is located, 6593 in <PAGE_SIZE> units (needed only for swap files). 6594 See Documentation/power/swsusp-and-swap-files.rst 6595 6596 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to 6597 read the resume files 6598 6599 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up. 6600 Useful for devices that are detected asynchronously 6601 (e.g. USB and MMC devices). 6602 6603 retain_initrd [RAM] Keep initrd memory after extraction. After boot, it will 6604 be accessible via /sys/firmware/initrd. 6605 6606 retbleed= [X86] Control mitigation of RETBleed (Arbitrary 6607 Speculative Code Execution with Return Instructions) 6608 vulnerability. 6609 6610 AMD-based UNRET and IBPB mitigations alone do not stop 6611 sibling threads from influencing the predictions of other 6612 sibling threads. For that reason, STIBP is used on pro- 6613 cessors that support it, and mitigate SMT on processors 6614 that don't. 6615 6616 off - no mitigation 6617 auto - automatically select a mitigation 6618 auto,nosmt - automatically select a mitigation, 6619 disabling SMT if necessary for 6620 the full mitigation (only on Zen1 6621 and older without STIBP). 6622 ibpb - On AMD, mitigate short speculation 6623 windows on basic block boundaries too. 6624 Safe, highest perf impact. It also 6625 enables STIBP if present. Not suitable 6626 on Intel. 6627 ibpb,nosmt - Like "ibpb" above but will disable SMT 6628 when STIBP is not available. This is 6629 the alternative for systems which do not 6630 have STIBP. 6631 unret - Force enable untrained return thunks, 6632 only effective on AMD f15h-f17h based 6633 systems. 6634 unret,nosmt - Like unret, but will disable SMT when STIBP 6635 is not available. This is the alternative for 6636 systems which do not have STIBP. 6637 6638 Selecting 'auto' will choose a mitigation method at run 6639 time according to the CPU. 6640 6641 Not specifying this option is equivalent to retbleed=auto. 6642 6643 rfkill.default_state= 6644 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm, 6645 etc. communication is blocked by default. 6646 1 Unblocked. 6647 6648 rfkill.master_switch_mode= 6649 0 The "airplane mode" button does nothing. 6650 1 The "airplane mode" button toggles between everything 6651 blocked and the previous configuration. 6652 2 The "airplane mode" button toggles between everything 6653 blocked and everything unblocked. 6654 6655 ring3mwait=disable 6656 [KNL] Disable ring 3 MONITOR/MWAIT feature on supported 6657 CPUs. 6658 6659 riscv_isa_fallback [RISCV,EARLY] 6660 When CONFIG_RISCV_ISA_FALLBACK is not enabled, permit 6661 falling back to detecting extension support by parsing 6662 "riscv,isa" property on devicetree systems when the 6663 replacement properties are not found. See the Kconfig 6664 entry for RISCV_ISA_FALLBACK. 6665 6666 riscv_nousercfi= 6667 all Disable user CFI ABI to userspace even if cpu extension 6668 are available. 6669 bcfi Disable user backward CFI ABI to userspace even if 6670 the shadow stack extension is available. 6671 fcfi Disable user forward CFI ABI to userspace even if the 6672 landing pad extension is available. 6673 6674 ro [KNL] Mount root device read-only on boot 6675 6676 rodata= [KNL,EARLY] 6677 on Mark read-only kernel memory as read-only (default). 6678 off Leave read-only kernel memory writable for debugging. 6679 noalias Mark read-only kernel memory as read-only but retain 6680 writable aliases in the direct map for regions outside 6681 of the kernel image. [arm64] 6682 6683 rockchip.usb_uart 6684 [EARLY] 6685 Enable the uart passthrough on the designated usb port 6686 on Rockchip SoCs. When active, the signals of the 6687 debug-uart get routed to the D+ and D- pins of the usb 6688 port and the regular usb controller gets disabled. 6689 6690 root= [KNL] Root filesystem 6691 Usually this is a block device specifier of some kind, 6692 see the early_lookup_bdev comment in 6693 block/early-lookup.c for details. 6694 Alternatively this can be "ram" for the legacy initial 6695 ramdisk, "nfs" and "cifs" for root on a network file 6696 system, or "mtd" and "ubi" for mounting from raw flash. 6697 6698 rootdelay= [KNL] Delay (in seconds) to pause before attempting to 6699 mount the root filesystem 6700 6701 rootflags= [KNL] Set root filesystem mount option string 6702 6703 rseq_slice_ext= [KNL] RSEQ based time slice extension 6704 Format: boolean 6705 Control enablement of RSEQ based time slice extension. 6706 Default is 'on'. 6707 6708 initramfs_options= [KNL] 6709 Specify mount options for the initramfs mount. 6710 6711 rootfstype= [KNL] Set root filesystem type 6712 6713 rootwait [KNL] Wait (indefinitely) for root device to show up. 6714 Useful for devices that are detected asynchronously 6715 (e.g. USB and MMC devices). 6716 6717 rootwait= [KNL] Maximum time (in seconds) to wait for root device 6718 to show up before attempting to mount the root 6719 filesystem. 6720 6721 rproc_mem=nn[KMG][@address] 6722 [KNL,ARM,CMA] Remoteproc physical memory block. 6723 Memory area to be used by remote processor image, 6724 managed by CMA. 6725 6726 rseq_debug= [KNL] Enable or disable restartable sequence 6727 debug mode. Defaults to CONFIG_RSEQ_DEBUG_DEFAULT_ENABLE. 6728 Format: <bool> 6729 6730 rt_group_sched= [KNL] Enable or disable SCHED_RR/FIFO group scheduling 6731 when CONFIG_RT_GROUP_SCHED=y. Defaults to 6732 !CONFIG_RT_GROUP_SCHED_DEFAULT_DISABLED. 6733 Format: <bool> 6734 6735 rw [KNL] Mount root device read-write on boot 6736 6737 S [KNL] Run init in single mode 6738 6739 s390_iommu= [HW,S390] 6740 Set s390 IOTLB flushing mode 6741 strict 6742 With strict flushing every unmap operation will result 6743 in an IOTLB flush. Default is lazy flushing before 6744 reuse, which is faster. Deprecated, equivalent to 6745 iommu.strict=1. 6746 6747 s390_iommu_aperture= [KNL,S390] 6748 Specifies the size of the per device DMA address space 6749 accessible through the DMA and IOMMU APIs as a decimal 6750 factor of the size of main memory. 6751 The default is 1 meaning that one can concurrently use 6752 as many DMA addresses as physical memory is installed, 6753 if supported by hardware, and thus map all of memory 6754 once. With a value of 2 one can map all of memory twice 6755 and so on. As a special case a factor of 0 imposes no 6756 restrictions other than those given by hardware at the 6757 cost of significant additional memory use for tables. 6758 6759 6760 sched_proxy_exec= [KNL] 6761 Enables or disables "proxy execution" style 6762 solution to mutex-based priority inversion. 6763 Format: <bool> 6764 6765 sched_verbose [KNL,EARLY] Enables verbose scheduler debug messages. 6766 6767 schedstats= [KNL,X86] Enable or disable scheduled statistics. 6768 Allowed values are enable and disable. This feature 6769 incurs a small amount of overhead in the scheduler 6770 but is useful for debugging and performance tuning. 6771 6772 sched_thermal_decay_shift= 6773 [Deprecated] 6774 [KNL, SMP] Set a decay shift for scheduler thermal 6775 pressure signal. Thermal pressure signal follows the 6776 default decay period of other scheduler pelt 6777 signals(usually 32 ms but configurable). Setting 6778 sched_thermal_decay_shift will left shift the decay 6779 period for the thermal pressure signal by the shift 6780 value. 6781 i.e. with the default pelt decay period of 32 ms 6782 sched_thermal_decay_shift thermal pressure decay pr 6783 1 64 ms 6784 2 128 ms 6785 and so on. 6786 Format: integer between 0 and 10 6787 Default is 0. 6788 6789 scftorture.holdoff= [KNL] 6790 Number of seconds to hold off before starting 6791 test. Defaults to zero for module insertion and 6792 to 10 seconds for built-in smp_call_function() 6793 tests. 6794 6795 scftorture.longwait= [KNL] 6796 Request ridiculously long waits randomly selected 6797 up to the chosen limit in seconds. Zero (the 6798 default) disables this feature. Please note 6799 that requesting even small non-zero numbers of 6800 seconds can result in RCU CPU stall warnings, 6801 softlockup complaints, and so on. 6802 6803 scftorture.nthreads= [KNL] 6804 Number of kthreads to spawn to invoke the 6805 smp_call_function() family of functions. 6806 The default of -1 specifies a number of kthreads 6807 equal to the number of CPUs. 6808 6809 scftorture.onoff_holdoff= [KNL] 6810 Number seconds to wait after the start of the 6811 test before initiating CPU-hotplug operations. 6812 6813 scftorture.onoff_interval= [KNL] 6814 Number seconds to wait between successive 6815 CPU-hotplug operations. Specifying zero (which 6816 is the default) disables CPU-hotplug operations. 6817 6818 scftorture.shutdown_secs= [KNL] 6819 The number of seconds following the start of the 6820 test after which to shut down the system. The 6821 default of zero avoids shutting down the system. 6822 Non-zero values are useful for automated tests. 6823 6824 scftorture.stat_interval= [KNL] 6825 The number of seconds between outputting the 6826 current test statistics to the console. A value 6827 of zero disables statistics output. 6828 6829 scftorture.stutter_cpus= [KNL] 6830 The number of jiffies to wait between each change 6831 to the set of CPUs under test. 6832 6833 scftorture.use_cpus_read_lock= [KNL] 6834 Use use_cpus_read_lock() instead of the default 6835 preempt_disable() to disable CPU hotplug 6836 while invoking one of the smp_call_function*() 6837 functions. 6838 6839 scftorture.verbose= [KNL] 6840 Enable additional printk() statements. 6841 6842 scftorture.weight_single= [KNL] 6843 The probability weighting to use for the 6844 smp_call_function_single() function with a zero 6845 "wait" parameter. A value of -1 selects the 6846 default if all other weights are -1. However, 6847 if at least one weight has some other value, a 6848 value of -1 will instead select a weight of zero. 6849 6850 scftorture.weight_single_wait= [KNL] 6851 The probability weighting to use for the 6852 smp_call_function_single() function with a 6853 non-zero "wait" parameter. See weight_single. 6854 6855 scftorture.weight_many= [KNL] 6856 The probability weighting to use for the 6857 smp_call_function_many() function with a zero 6858 "wait" parameter. See weight_single. 6859 Note well that setting a high probability for 6860 this weighting can place serious IPI load 6861 on the system. 6862 6863 scftorture.weight_many_wait= [KNL] 6864 The probability weighting to use for the 6865 smp_call_function_many() function with a 6866 non-zero "wait" parameter. See weight_single 6867 and weight_many. 6868 6869 scftorture.weight_all= [KNL] 6870 The probability weighting to use for the 6871 smp_call_function_all() function with a zero 6872 "wait" parameter. See weight_single and 6873 weight_many. 6874 6875 scftorture.weight_all_wait= [KNL] 6876 The probability weighting to use for the 6877 smp_call_function_all() function with a 6878 non-zero "wait" parameter. See weight_single 6879 and weight_many. 6880 6881 sdw_mclk_divider=[SDW] 6882 Specify the MCLK divider for Intel SoundWire buses in 6883 case the BIOS does not provide the clock rate properly. 6884 6885 skew_tick= [KNL,EARLY] Offset the periodic timer tick per cpu to mitigate 6886 xtime_lock contention on larger systems, and/or RCU lock 6887 contention on all systems with CONFIG_MAXSMP set. 6888 Format: { "0" | "1" } 6889 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1" 6890 1 -- enable. 6891 Note: increases power consumption, thus should only be 6892 enabled if running jitter sensitive (HPC/RT) workloads. 6893 6894 security= [SECURITY] Choose a legacy "major" security module to 6895 enable at boot. This has been deprecated by the 6896 "lsm=" parameter. 6897 6898 selinux= [SELINUX] Disable or enable SELinux at boot time. 6899 Format: { "0" | "1" } 6900 See security/selinux/Kconfig help text. 6901 0 -- disable. 6902 1 -- enable. 6903 Default value is 1. 6904 6905 serialnumber [BUGS=X86-32] 6906 6907 sev=option[,option...] [X86-64] 6908 6909 debug 6910 Enable debug messages. 6911 6912 nosnp 6913 Do not enable SEV-SNP (applies to host/hypervisor 6914 only). Setting 'nosnp' avoids the RMP check overhead 6915 in memory accesses when users do not want to run 6916 SEV-SNP guests. 6917 6918 shapers= [NET] 6919 Maximal number of shapers. 6920 6921 show_lapic= [APIC,X86] Advanced Programmable Interrupt Controller 6922 Limit apic dumping. The parameter defines the maximal 6923 number of local apics being dumped. Also it is possible 6924 to set it to "all" by meaning -- no limit here. 6925 Format: { 1 (default) | 2 | ... | all }. 6926 The parameter valid if only apic=debug or 6927 apic=verbose is specified. 6928 Example: apic=debug show_lapic=all 6929 6930 slab_debug[=options[,slabs][;[options[,slabs]]...] [MM] 6931 Enabling slab_debug allows one to determine the 6932 culprit if slab objects become corrupted. Enabling 6933 slab_debug can create guard zones around objects and 6934 may poison objects when not in use. Also tracks the 6935 last alloc / free. For more information see 6936 Documentation/admin-guide/mm/slab.rst. 6937 (slub_debug legacy name also accepted for now) 6938 6939 Using this option implies the "no_hash_pointers" 6940 option which can be undone by adding the 6941 "hash_pointers=always" option. 6942 6943 slab_max_order= [MM] 6944 Determines the maximum allowed order for slabs. 6945 A high setting may cause OOMs due to memory 6946 fragmentation. For more information see 6947 Documentation/admin-guide/mm/slab.rst. 6948 (slub_max_order legacy name also accepted for now) 6949 6950 slab_merge [MM] 6951 Enable merging of slabs with similar size when the 6952 kernel is built without CONFIG_SLAB_MERGE_DEFAULT. 6953 (slub_merge legacy name also accepted for now) 6954 6955 slab_min_objects= [MM] 6956 The minimum number of objects per slab. SLUB will 6957 increase the slab order up to slab_max_order to 6958 generate a sufficiently large slab able to contain 6959 the number of objects indicated. The higher the number 6960 of objects the smaller the overhead of tracking slabs 6961 and the less frequently locks need to be acquired. 6962 For more information see 6963 Documentation/admin-guide/mm/slab.rst. 6964 (slub_min_objects legacy name also accepted for now) 6965 6966 slab_min_order= [MM] 6967 Determines the minimum page order for slabs. Must be 6968 lower or equal to slab_max_order. For more information see 6969 Documentation/admin-guide/mm/slab.rst. 6970 (slub_min_order legacy name also accepted for now) 6971 6972 slab_nomerge [MM] 6973 Disable merging of slabs with similar size. May be 6974 necessary if there is some reason to distinguish 6975 allocs to different slabs, especially in hardened 6976 environments where the risk of heap overflows and 6977 layout control by attackers can usually be 6978 frustrated by disabling merging. This will reduce 6979 most of the exposure of a heap attack to a single 6980 cache (risks via metadata attacks are mostly 6981 unchanged). Debug options disable merging on their 6982 own. 6983 For more information see 6984 Documentation/admin-guide/mm/slab.rst. 6985 (slub_nomerge legacy name also accepted for now) 6986 6987 slab_strict_numa [MM] 6988 Support memory policies on a per object level 6989 in the slab allocator. The default is for memory 6990 policies to be applied at the folio level when 6991 a new folio is needed or a partial folio is 6992 retrieved from the lists. Increases overhead 6993 in the slab fastpaths but gains more accurate 6994 NUMA kernel object placement which helps with slow 6995 interconnects in NUMA systems. 6996 6997 slram= [HW,MTD] 6998 6999 smart2= [HW] 7000 Format: <io1>[,<io2>[,...,<io8>]] 7001 7002 smp.csd_lock_timeout= [KNL] 7003 Specify the period of time in milliseconds 7004 that smp_call_function() and friends will wait 7005 for a CPU to release the CSD lock. This is 7006 useful when diagnosing bugs involving CPUs 7007 disabling interrupts for extended periods 7008 of time. Defaults to 5,000 milliseconds, and 7009 setting a value of zero disables this feature. 7010 This feature may be more efficiently disabled 7011 using the csdlock_debug- kernel parameter. 7012 7013 smp.panic_on_ipistall= [KNL] 7014 If a csd_lock_timeout extends for more than 7015 the specified number of milliseconds, panic the 7016 system. By default, let CSD-lock acquisition 7017 take as long as they take. Specifying 300,000 7018 for this value provides a 5-minute timeout. 7019 7020 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices 7021 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port 7022 smsc-ircc2.ircc_sir= [HW] SIR base I/O port 7023 smsc-ircc2.ircc_fir= [HW] FIR base I/O port 7024 smsc-ircc2.ircc_irq= [HW] IRQ line 7025 smsc-ircc2.ircc_dma= [HW] DMA channel 7026 smsc-ircc2.ircc_transceiver= [HW] Transceiver type: 7027 0: Toshiba Satellite 1800 (GP data pin select) 7028 1: Fast pin select (default) 7029 2: ATC IRMode 7030 7031 smt= [KNL,MIPS,S390,EARLY] Set the maximum number of threads 7032 (logical CPUs) to use per physical CPU on systems 7033 capable of symmetric multithreading (SMT). Will 7034 be capped to the actual hardware limit. 7035 Format: <integer> 7036 Default: -1 (no limit) 7037 7038 softlockup_panic= 7039 [KNL] Should the soft-lockup detector generate panics. 7040 Format: <int> 7041 7042 A value of non-zero instructs the soft-lockup detector 7043 to panic the machine when a soft-lockup duration exceeds 7044 N thresholds. It is also controlled by the kernel.softlockup_panic 7045 sysctl and CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC, which is the 7046 respective build-time switch to that functionality. 7047 7048 softlockup_all_cpu_backtrace= 7049 [KNL] Should the soft-lockup detector generate 7050 backtraces on all cpus. 7051 Format: 0 | 1 7052 7053 sonypi.*= [HW] Sony Programmable I/O Control Device driver 7054 See Documentation/admin-guide/laptops/sonypi.rst 7055 7056 spectre_bhi= [X86] Control mitigation of Branch History Injection 7057 (BHI) vulnerability. This setting affects the 7058 deployment of the HW BHI control and the SW BHB 7059 clearing sequence. 7060 7061 on - (default) Enable the HW or SW mitigation as 7062 needed. This protects the kernel from 7063 both syscalls and VMs. 7064 vmexit - On systems which don't have the HW mitigation 7065 available, enable the SW mitigation on vmexit 7066 ONLY. On such systems, the host kernel is 7067 protected from VM-originated BHI attacks, but 7068 may still be vulnerable to syscall attacks. 7069 off - Disable the mitigation. 7070 7071 spectre_v2= [X86,EARLY] Control mitigation of Spectre variant 2 7072 (indirect branch speculation) vulnerability. 7073 The default operation protects the kernel from 7074 user space attacks. 7075 7076 on - unconditionally enable, implies 7077 spectre_v2_user=on 7078 off - unconditionally disable, implies 7079 spectre_v2_user=off 7080 auto - kernel detects whether your CPU model is 7081 vulnerable 7082 7083 Selecting 'on' will, and 'auto' may, choose a 7084 mitigation method at run time according to the 7085 CPU, the available microcode, the setting of the 7086 CONFIG_MITIGATION_RETPOLINE configuration option, 7087 and the compiler with which the kernel was built. 7088 7089 Selecting 'on' will also enable the mitigation 7090 against user space to user space task attacks. 7091 Selecting specific mitigation does not force enable 7092 user mitigations. 7093 7094 Selecting 'off' will disable both the kernel and 7095 the user space protections. 7096 7097 Specific mitigations can also be selected manually: 7098 7099 retpoline - replace indirect branches 7100 retpoline,generic - Retpolines 7101 retpoline,lfence - LFENCE; indirect branch 7102 retpoline,amd - alias for retpoline,lfence 7103 eibrs - Enhanced/Auto IBRS 7104 eibrs,retpoline - Enhanced/Auto IBRS + Retpolines 7105 eibrs,lfence - Enhanced/Auto IBRS + LFENCE 7106 ibrs - use IBRS to protect kernel 7107 7108 Not specifying this option is equivalent to 7109 spectre_v2=auto. 7110 7111 spectre_v2_user= 7112 [X86] Control mitigation of Spectre variant 2 7113 (indirect branch speculation) vulnerability between 7114 user space tasks 7115 7116 on - Unconditionally enable mitigations. Is 7117 enforced by spectre_v2=on 7118 7119 off - Unconditionally disable mitigations. Is 7120 enforced by spectre_v2=off 7121 7122 prctl - Indirect branch speculation is enabled, 7123 but mitigation can be enabled via prctl 7124 per thread. The mitigation control state 7125 is inherited on fork. 7126 7127 prctl,ibpb 7128 - Like "prctl" above, but only STIBP is 7129 controlled per thread. IBPB is issued 7130 always when switching between different user 7131 space processes. 7132 7133 seccomp 7134 - Same as "prctl" above, but all seccomp 7135 threads will enable the mitigation unless 7136 they explicitly opt out. 7137 7138 seccomp,ibpb 7139 - Like "seccomp" above, but only STIBP is 7140 controlled per thread. IBPB is issued 7141 always when switching between different 7142 user space processes. 7143 7144 auto - Kernel selects the mitigation depending on 7145 the available CPU features and vulnerability. 7146 7147 Default mitigation: "prctl" 7148 7149 Not specifying this option is equivalent to 7150 spectre_v2_user=auto. 7151 7152 spec_rstack_overflow= 7153 [X86,EARLY] Control RAS overflow mitigation on AMD Zen CPUs 7154 7155 off - Disable mitigation 7156 microcode - Enable microcode mitigation only 7157 safe-ret - Enable sw-only safe RET mitigation (default) 7158 ibpb - Enable mitigation by issuing IBPB on 7159 kernel entry 7160 ibpb-vmexit - Issue IBPB only on VMEXIT 7161 (cloud-specific mitigation) 7162 7163 spec_store_bypass_disable= 7164 [HW,EARLY] Control Speculative Store Bypass (SSB) Disable mitigation 7165 (Speculative Store Bypass vulnerability) 7166 7167 Certain CPUs are vulnerable to an exploit against a 7168 a common industry wide performance optimization known 7169 as "Speculative Store Bypass" in which recent stores 7170 to the same memory location may not be observed by 7171 later loads during speculative execution. The idea 7172 is that such stores are unlikely and that they can 7173 be detected prior to instruction retirement at the 7174 end of a particular speculation execution window. 7175 7176 In vulnerable processors, the speculatively forwarded 7177 store can be used in a cache side channel attack, for 7178 example to read memory to which the attacker does not 7179 directly have access (e.g. inside sandboxed code). 7180 7181 This parameter controls whether the Speculative Store 7182 Bypass optimization is used. 7183 7184 On x86 the options are: 7185 7186 on - Unconditionally disable Speculative Store Bypass 7187 off - Unconditionally enable Speculative Store Bypass 7188 auto - Kernel detects whether the CPU model contains an 7189 implementation of Speculative Store Bypass and 7190 picks the most appropriate mitigation. If the 7191 CPU is not vulnerable, "off" is selected. If the 7192 CPU is vulnerable the default mitigation is 7193 architecture and Kconfig dependent. See below. 7194 prctl - Control Speculative Store Bypass per thread 7195 via prctl. Speculative Store Bypass is enabled 7196 for a process by default. The state of the control 7197 is inherited on fork. 7198 seccomp - Same as "prctl" above, but all seccomp threads 7199 will disable SSB unless they explicitly opt out. 7200 7201 Default mitigations: 7202 X86: "prctl" 7203 7204 On powerpc the options are: 7205 7206 on,auto - On Power8 and Power9 insert a store-forwarding 7207 barrier on kernel entry and exit. On Power7 7208 perform a software flush on kernel entry and 7209 exit. 7210 off - No action. 7211 7212 Not specifying this option is equivalent to 7213 spec_store_bypass_disable=auto. 7214 7215 split_llc= 7216 [X86,EARLY] Split the LLC N-ways 7217 7218 When set, the LLC is split this many ways by matching 7219 'core_id % n'. This is setup before SMP bringup and 7220 used during SMP bringup before it knows the full 7221 topology. If your core count doesn't nicely divide by 7222 the number given, you get to keep the pieces. 7223 7224 This is mostly a debug feature to emulate multiple LLCs 7225 on hardware that only have a single LLC. 7226 7227 split_lock_detect= 7228 [X86] Enable split lock detection or bus lock detection 7229 7230 When enabled (and if hardware support is present), atomic 7231 instructions that access data across cache line 7232 boundaries will result in an alignment check exception 7233 for split lock detection or a debug exception for 7234 bus lock detection. 7235 7236 off - not enabled 7237 7238 warn - the kernel will emit rate-limited warnings 7239 about applications triggering the #AC 7240 exception or the #DB exception. This mode is 7241 the default on CPUs that support split lock 7242 detection or bus lock detection. Default 7243 behavior is by #AC if both features are 7244 enabled in hardware. 7245 7246 fatal - the kernel will send SIGBUS to applications 7247 that trigger the #AC exception or the #DB 7248 exception. Default behavior is by #AC if 7249 both features are enabled in hardware. 7250 7251 ratelimit:N - 7252 Set system wide rate limit to N bus locks 7253 per second for bus lock detection. 7254 0 < N <= 1000. 7255 7256 N/A for split lock detection. 7257 7258 7259 If an #AC exception is hit in the kernel or in 7260 firmware (i.e. not while executing in user mode) 7261 the kernel will oops in either "warn" or "fatal" 7262 mode. 7263 7264 #DB exception for bus lock is triggered only when 7265 CPL > 0. 7266 7267 srbds= [X86,INTEL,EARLY] 7268 Control the Special Register Buffer Data Sampling 7269 (SRBDS) mitigation. 7270 7271 Certain CPUs are vulnerable to an MDS-like 7272 exploit which can leak bits from the random 7273 number generator. 7274 7275 By default, this issue is mitigated by 7276 microcode. However, the microcode fix can cause 7277 the RDRAND and RDSEED instructions to become 7278 much slower. Among other effects, this will 7279 result in reduced throughput from /dev/urandom. 7280 7281 The microcode mitigation can be disabled with 7282 the following option: 7283 7284 off: Disable mitigation and remove 7285 performance impact to RDRAND and RDSEED 7286 7287 srcutree.big_cpu_lim [KNL] 7288 Specifies the number of CPUs constituting a 7289 large system, such that srcu_struct structures 7290 should immediately allocate an srcu_node array. 7291 This kernel-boot parameter defaults to 128, 7292 but takes effect only when the low-order four 7293 bits of srcutree.convert_to_big is equal to 3 7294 (decide at boot). 7295 7296 srcutree.convert_to_big [KNL] 7297 Specifies under what conditions an SRCU tree 7298 srcu_struct structure will be converted to big 7299 form, that is, with an rcu_node tree: 7300 7301 0: Never. 7302 1: At init_srcu_struct() time. 7303 2: When rcutorture decides to. 7304 3: Decide at boot time (default). 7305 0x1X: Above plus if high contention. 7306 7307 Either way, the srcu_node tree will be sized based 7308 on the actual runtime number of CPUs (nr_cpu_ids) 7309 instead of the compile-time CONFIG_NR_CPUS. 7310 7311 srcutree.counter_wrap_check [KNL] 7312 Specifies how frequently to check for 7313 grace-period sequence counter wrap for the 7314 srcu_data structure's ->srcu_gp_seq_needed field. 7315 The greater the number of bits set in this kernel 7316 parameter, the less frequently counter wrap will 7317 be checked for. Note that the bottom two bits 7318 are ignored. 7319 7320 srcutree.exp_holdoff [KNL] 7321 Specifies how many nanoseconds must elapse 7322 since the end of the last SRCU grace period for 7323 a given srcu_struct until the next normal SRCU 7324 grace period will be considered for automatic 7325 expediting. Set to zero to disable automatic 7326 expediting. 7327 7328 srcutree.srcu_max_nodelay [KNL] 7329 Specifies the number of no-delay instances 7330 per jiffy for which the SRCU grace period 7331 worker thread will be rescheduled with zero 7332 delay. Beyond this limit, worker thread will 7333 be rescheduled with a sleep delay of one jiffy. 7334 7335 srcutree.srcu_max_nodelay_phase [KNL] 7336 Specifies the per-grace-period phase, number of 7337 non-sleeping polls of readers. Beyond this limit, 7338 grace period worker thread will be rescheduled 7339 with a sleep delay of one jiffy, between each 7340 rescan of the readers, for a grace period phase. 7341 7342 srcutree.srcu_retry_check_delay [KNL] 7343 Specifies number of microseconds of non-sleeping 7344 delay between each non-sleeping poll of readers. 7345 7346 srcutree.small_contention_lim [KNL] 7347 Specifies the number of update-side contention 7348 events per jiffy will be tolerated before 7349 initiating a conversion of an srcu_struct 7350 structure to big form. Note that the value of 7351 srcutree.convert_to_big must have the 0x10 bit 7352 set for contention-based conversions to occur. 7353 7354 ssbd= [ARM64,HW,EARLY] 7355 Speculative Store Bypass Disable control 7356 7357 On CPUs that are vulnerable to the Speculative 7358 Store Bypass vulnerability and offer a 7359 firmware based mitigation, this parameter 7360 indicates how the mitigation should be used: 7361 7362 force-on: Unconditionally enable mitigation for 7363 for both kernel and userspace 7364 force-off: Unconditionally disable mitigation for 7365 for both kernel and userspace 7366 kernel: Always enable mitigation in the 7367 kernel, and offer a prctl interface 7368 to allow userspace to register its 7369 interest in being mitigated too. 7370 7371 stack_guard_gap= [MM] 7372 override the default stack gap protection. The value 7373 is in page units and it defines how many pages prior 7374 to (for stacks growing down) resp. after (for stacks 7375 growing up) the main stack are reserved for no other 7376 mapping. Default value is 256 pages. 7377 7378 stack_depot_disable= [KNL,EARLY] 7379 Setting this to true through kernel command line will 7380 disable the stack depot thereby saving the static memory 7381 consumed by the stack hash table. By default this is set 7382 to false. 7383 7384 stack_depot_max_pools= [KNL,EARLY] 7385 Specify the maximum number of pools to use for storing 7386 stack traces. Pools are allocated on-demand up to this 7387 limit. Default value is 8191 pools. 7388 7389 stacktrace [FTRACE] 7390 Enable the stack tracer on boot up. 7391 7392 stacktrace_filter=[function-list] 7393 [FTRACE] Limit the functions that the stack tracer 7394 will trace at boot up. function-list is a comma-separated 7395 list of functions. This list can be changed at run 7396 time by the stack_trace_filter file in the debugfs 7397 tracing directory. Note, this enables stack tracing 7398 and the stacktrace above is not needed. 7399 7400 sti= [PARISC,HW] 7401 Format: <num> 7402 Set the STI (builtin display/keyboard on the HP-PARISC 7403 machines) console (graphic card) which should be used 7404 as the initial boot-console. 7405 See also comment in drivers/video/sticore.c. 7406 7407 sti_font= [HW] 7408 See comment in drivers/video/sticore.c. 7409 7410 stifb= [HW] 7411 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]] 7412 7413 strict_sas_size= 7414 [X86] 7415 Format: <bool> 7416 Enable or disable strict sigaltstack size checks 7417 against the required signal frame size which 7418 depends on the supported FPU features. This can 7419 be used to filter out binaries which have 7420 not yet been made aware of AT_MINSIGSTKSZ. 7421 7422 stress_hpt [PPC,EARLY] 7423 Limits the number of kernel HPT entries in the hash 7424 page table to increase the rate of hash page table 7425 faults on kernel addresses. 7426 7427 stress_slb [PPC,EARLY] 7428 Limits the number of kernel SLB entries, and flushes 7429 them frequently to increase the rate of SLB faults 7430 on kernel addresses. 7431 7432 no_slb_preload [PPC,EARLY] 7433 Disables slb preloading for userspace. 7434 7435 sunrpc.min_resvport= 7436 sunrpc.max_resvport= 7437 [NFS,SUNRPC] 7438 SunRPC servers often require that client requests 7439 originate from a privileged port (i.e. a port in the 7440 range 0 < portnr < 1024). 7441 An administrator who wishes to reserve some of these 7442 ports for other uses may adjust the range that the 7443 kernel's sunrpc client considers to be privileged 7444 using these two parameters to set the minimum and 7445 maximum port values. 7446 7447 sunrpc.svc_rpc_per_connection_limit= 7448 [NFS,SUNRPC] 7449 Limit the number of requests that the server will 7450 process in parallel from a single connection. 7451 The default value is 0 (no limit). 7452 7453 sunrpc.pool_mode= 7454 [NFS] 7455 Control how the NFS server code allocates CPUs to 7456 service thread pools. Depending on how many NICs 7457 you have and where their interrupts are bound, this 7458 option will affect which CPUs will do NFS serving. 7459 Note: this parameter cannot be changed while the 7460 NFS server is running. 7461 7462 auto the server chooses an appropriate mode 7463 automatically using heuristics 7464 global a single global pool contains all CPUs 7465 percpu one pool for each CPU 7466 pernode one pool for each NUMA node (equivalent 7467 to global on non-NUMA machines) 7468 7469 sunrpc.tcp_slot_table_entries= 7470 sunrpc.udp_slot_table_entries= 7471 [NFS,SUNRPC] 7472 Sets the upper limit on the number of simultaneous 7473 RPC calls that can be sent from the client to a 7474 server. Increasing these values may allow you to 7475 improve throughput, but will also increase the 7476 amount of memory reserved for use by the client. 7477 7478 suspend.pm_test_delay= 7479 [SUSPEND] 7480 Sets the number of seconds to remain in a suspend test 7481 mode before resuming the system (see 7482 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG 7483 is set. Default value is 5. 7484 7485 svm= [PPC] 7486 Format: { on | off | y | n | 1 | 0 } 7487 This parameter controls use of the Protected 7488 Execution Facility on pSeries. 7489 7490 swiotlb= [ARM,PPC,MIPS,X86,S390,EARLY] 7491 Format: { <int> [,<int>] | force | noforce } 7492 <int> -- Number of I/O TLB slabs 7493 <int> -- Second integer after comma. Number of swiotlb 7494 areas with their own lock. Will be rounded up 7495 to a power of 2. 7496 force -- force using of bounce buffers even if they 7497 wouldn't be automatically used by the kernel 7498 noforce -- Never use bounce buffers (for debugging) 7499 7500 switches= [HW,M68k,EARLY] 7501 7502 sysctl.*= [KNL] 7503 Set a sysctl parameter, right before loading the init 7504 process, as if the value was written to the respective 7505 /proc/sys/... file. Both '.' and '/' are recognized as 7506 separators. Unrecognized parameters and invalid values 7507 are reported in the kernel log. Sysctls registered 7508 later by a loaded module cannot be set this way. 7509 Example: sysctl.vm.swappiness=40 7510 7511 sysrq_always_enabled 7512 [KNL] 7513 Ignore sysrq setting - this boot parameter will 7514 neutralize any effect of /proc/sys/kernel/sysrq. 7515 Useful for debugging. 7516 7517 tcpmhash_entries= [KNL,NET] 7518 Set the number of tcp_metrics_hash slots. 7519 Default value is 8192 or 16384 depending on total 7520 ram pages. This is used to specify the TCP metrics 7521 cache size. See Documentation/networking/ip-sysctl.rst 7522 "tcp_no_metrics_save" section for more details. 7523 7524 tdfx= [HW,DRM] 7525 7526 test_suspend= [SUSPEND] 7527 Format: { "mem" | "standby" | "freeze" }[,N] 7528 Specify "mem" (for Suspend-to-RAM) or "standby" (for 7529 standby suspend) or "freeze" (for suspend type freeze) 7530 as the system sleep state during system startup with 7531 the optional capability to repeat N number of times. 7532 The system is woken from this state using a 7533 wakeup-capable RTC alarm. 7534 7535 thash_entries= [KNL,NET] 7536 Set number of hash buckets for TCP connection 7537 7538 thermal.act= [HW,ACPI] 7539 -1: disable all active trip points in all thermal zones 7540 <degrees C>: override all lowest active trip points 7541 7542 thermal.crt= [HW,ACPI] 7543 -1: disable all critical trip points in all thermal zones 7544 <degrees C>: override all critical trip points 7545 7546 thermal.off= [HW,ACPI] 7547 1: disable ACPI thermal control 7548 7549 thermal.psv= [HW,ACPI] 7550 -1: disable all passive trip points 7551 <degrees C>: override all passive trip points to this 7552 value 7553 7554 thermal.tzp= [HW,ACPI] 7555 Specify global default ACPI thermal zone polling rate 7556 <deci-seconds>: poll all this frequency 7557 0: no polling (default) 7558 7559 thp_anon= [KNL] 7560 Format: <size>[KMG],<size>[KMG]:<state>;<size>[KMG]-<size>[KMG]:<state> 7561 state is one of "always", "madvise", "never" or "inherit". 7562 Control the default behavior of the system with respect 7563 to anonymous transparent hugepages. 7564 Can be used multiple times for multiple anon THP sizes. 7565 See Documentation/admin-guide/mm/transhuge.rst for more 7566 details. 7567 7568 threadirqs [KNL,EARLY] 7569 Force threading of all interrupt handlers except those 7570 marked explicitly IRQF_NO_THREAD. 7571 7572 thp_shmem= [KNL] 7573 Format: <size>[KMG],<size>[KMG]:<policy>;<size>[KMG]-<size>[KMG]:<policy> 7574 Control the default policy of each hugepage size for the 7575 internal shmem mount. <policy> is one of policies available 7576 for the shmem mount ("always", "inherit", "never", "within_size", 7577 and "advise"). 7578 It can be used multiple times for multiple shmem THP sizes. 7579 See Documentation/admin-guide/mm/transhuge.rst for more 7580 details. 7581 7582 topology= [S390,EARLY] 7583 Format: {off | on} 7584 Specify if the kernel should make use of the cpu 7585 topology information if the hardware supports this. 7586 The scheduler will make use of this information and 7587 e.g. base its process migration decisions on it. 7588 Default is on. 7589 7590 torture.disable_onoff_at_boot= [KNL] 7591 Prevent the CPU-hotplug component of torturing 7592 until after init has spawned. 7593 7594 torture.ftrace_dump_at_shutdown= [KNL] 7595 Dump the ftrace buffer at torture-test shutdown, 7596 even if there were no errors. This can be a 7597 very costly operation when many torture tests 7598 are running concurrently, especially on systems 7599 with rotating-rust storage. 7600 7601 torture.verbose_sleep_frequency= [KNL] 7602 Specifies how many verbose printk()s should be 7603 emitted between each sleep. The default of zero 7604 disables verbose-printk() sleeping. 7605 7606 torture.verbose_sleep_duration= [KNL] 7607 Duration of each verbose-printk() sleep in jiffies. 7608 7609 tpm.disable_pcr_integrity= [HW,TPM] 7610 Do not protect PCR registers from unintended physical 7611 access, or interposers in the bus by the means of 7612 having an integrity protected session wrapped around 7613 TPM2_PCR_Extend command. Consider this in a situation 7614 where TPM is heavily utilized by IMA, thus protection 7615 causing a major performance hit, and the space where 7616 machines are deployed is by other means guarded. 7617 7618 tpm_crb_ffa.busy_timeout_ms= [ARM64,TPM] 7619 Maximum time in milliseconds to retry sending a message 7620 to the TPM service before giving up. This parameter controls 7621 how long the system will continue retrying when the TPM 7622 service is busy. 7623 Format: <unsigned int> 7624 Default: 2000 (2 seconds) 7625 7626 tpm_suspend_pcr=[HW,TPM] 7627 Format: integer pcr id 7628 Specify that at suspend time, the tpm driver 7629 should extend the specified pcr with zeros, 7630 as a workaround for some chips which fail to 7631 flush the last written pcr on TPM_SaveState. 7632 This will guarantee that all the other pcrs 7633 are saved. 7634 7635 tpm_tis.interrupts= [HW,TPM] 7636 Enable interrupts for the MMIO based physical layer 7637 for the FIFO interface. By default it is set to false 7638 (0). For more information about TPM hardware interfaces 7639 defined by Trusted Computing Group (TCG) see 7640 https://trustedcomputinggroup.org/resource/pc-client-platform-tpm-profile-ptp-specification/ 7641 7642 tp_printk [FTRACE] 7643 Have the tracepoints sent to printk as well as the 7644 tracing ring buffer. This is useful for early boot up 7645 where the system hangs or reboots and does not give the 7646 option for reading the tracing buffer or performing a 7647 ftrace_dump_on_oops. 7648 7649 To turn off having tracepoints sent to printk, 7650 echo 0 > /proc/sys/kernel/tracepoint_printk 7651 Note, echoing 1 into this file without the 7652 tp_printk kernel cmdline option has no effect. 7653 7654 The tp_printk_stop_on_boot (see below) can also be used 7655 to stop the printing of events to console at 7656 late_initcall_sync. 7657 7658 ** CAUTION ** 7659 7660 Having tracepoints sent to printk() and activating high 7661 frequency tracepoints such as irq or sched, can cause 7662 the system to live lock. 7663 7664 tp_printk_stop_on_boot [FTRACE] 7665 When tp_printk (above) is set, it can cause a lot of noise 7666 on the console. It may be useful to only include the 7667 printing of events during boot up, as user space may 7668 make the system inoperable. 7669 7670 This command line option will stop the printing of events 7671 to console at the late_initcall_sync() time frame. 7672 7673 trace_buf_size=nn[KMG] 7674 [FTRACE] will set tracing buffer size on each cpu. 7675 7676 trace_clock= [FTRACE] Set the clock used for tracing events 7677 at boot up. 7678 local - Use the per CPU time stamp counter 7679 (converted into nanoseconds). Fast, but 7680 depending on the architecture, may not be 7681 in sync between CPUs. 7682 global - Event time stamps are synchronized across 7683 CPUs. May be slower than the local clock, 7684 but better for some race conditions. 7685 counter - Simple counting of events (1, 2, ..) 7686 note, some counts may be skipped due to the 7687 infrastructure grabbing the clock more than 7688 once per event. 7689 uptime - Use jiffies as the time stamp. 7690 perf - Use the same clock that perf uses. 7691 mono - Use ktime_get_mono_fast_ns() for time stamps. 7692 mono_raw - Use ktime_get_raw_fast_ns() for time 7693 stamps. 7694 boot - Use ktime_get_boot_fast_ns() for time stamps. 7695 Architectures may add more clocks. See 7696 Documentation/trace/ftrace.rst for more details. 7697 7698 trace_event=[event-list] 7699 [FTRACE] Set and start specified trace events in order 7700 to facilitate early boot debugging. The event-list is a 7701 comma-separated list of trace events to enable. See 7702 also Documentation/trace/events.rst 7703 7704 To enable modules, use :mod: keyword: 7705 7706 trace_event=:mod:<module> 7707 7708 The value before :mod: will only enable specific events 7709 that are part of the module. See the above mentioned 7710 document for more information. 7711 7712 trace_instance=[instance-info] 7713 [FTRACE] Create a ring buffer instance early in boot up. 7714 This will be listed in: 7715 7716 /sys/kernel/tracing/instances 7717 7718 Events can be enabled at the time the instance is created 7719 via: 7720 7721 trace_instance=<name>,<system1>:<event1>,<system2>:<event2> 7722 7723 Note, the "<system*>:" portion is optional if the event is 7724 unique. 7725 7726 trace_instance=foo,sched:sched_switch,irq_handler_entry,initcall 7727 7728 will enable the "sched_switch" event (note, the "sched:" is optional, and 7729 the same thing would happen if it was left off). The irq_handler_entry 7730 event, and all events under the "initcall" system. 7731 7732 Flags can be added to the instance to modify its behavior when it is 7733 created. The flags are separated by '^'. 7734 7735 The available flags are: 7736 7737 traceoff - Have the tracing instance tracing disabled after it is created. 7738 traceprintk - Have trace_printk() write into this trace instance 7739 (note, "printk" and "trace_printk" can also be used) 7740 7741 trace_instance=foo^traceoff^traceprintk,sched,irq 7742 7743 The flags must come before the defined events. 7744 7745 If memory has been reserved (see memmap for x86), the instance 7746 can use that memory: 7747 7748 memmap=12M$0x284500000 trace_instance=boot_map@0x284500000:12M 7749 7750 The above will create a "boot_map" instance that uses the physical 7751 memory at 0x284500000 that is 12Megs. The per CPU buffers of that 7752 instance will be split up accordingly. 7753 7754 Alternatively, the memory can be reserved by the reserve_mem option: 7755 7756 reserve_mem=12M:4096:trace trace_instance=boot_map@trace 7757 7758 This will reserve 12 megabytes at boot up with a 4096 byte alignment 7759 and place the ring buffer in this memory. Note that due to KASLR, the 7760 memory may not be the same location each time, which will not preserve 7761 the buffer content. 7762 7763 Also note that the layout of the ring buffer data may change between 7764 kernel versions where the validator will fail and reset the ring buffer 7765 if the layout is not the same as the previous kernel. 7766 7767 If the ring buffer is used for persistent bootups and has events enabled, 7768 it is recommend to disable tracing so that events from a previous boot do not 7769 mix with events of the current boot (unless you are debugging a random crash 7770 at boot up). 7771 7772 reserve_mem=12M:4096:trace trace_instance=boot_map^traceoff^traceprintk@trace,sched,irq 7773 7774 Note, saving the trace buffer across reboots does require that the system 7775 is set up to not wipe memory. For instance, CONFIG_RESET_ATTACK_MITIGATION 7776 can force a memory reset on boot which will clear any trace that was stored. 7777 This is just one of many ways that can clear memory. Make sure your system 7778 keeps the content of memory across reboots before relying on this option. 7779 7780 NB: Both the mapped address and size must be page aligned for the architecture. 7781 7782 See also Documentation/trace/debugging.rst 7783 7784 7785 trace_options=[option-list] 7786 [FTRACE] Enable or disable tracer options at boot. 7787 The option-list is a comma delimited list of options 7788 that can be enabled or disabled just as if you were 7789 to echo the option name into 7790 7791 /sys/kernel/tracing/trace_options 7792 7793 For example, to enable stacktrace option (to dump the 7794 stack trace of each event), add to the command line: 7795 7796 trace_options=stacktrace 7797 7798 See also Documentation/trace/ftrace.rst "trace options" 7799 section. 7800 7801 trace_trigger=[trigger-list] 7802 [FTRACE] Add an event trigger on specific events. 7803 Set a trigger on top of a specific event, with an optional 7804 filter. 7805 7806 The format is "trace_trigger=<event>.<trigger>[ if <filter>],..." 7807 Where more than one trigger may be specified that are comma delimited. 7808 7809 For example: 7810 7811 trace_trigger="sched_switch.stacktrace if prev_state == 2" 7812 7813 The above will enable the "stacktrace" trigger on the "sched_switch" 7814 event but only trigger it if the "prev_state" of the "sched_switch" 7815 event is "2" (TASK_UNINTERRUPTIBLE). 7816 7817 See also "Event triggers" in Documentation/trace/events.rst 7818 7819 7820 traceoff_after_boot 7821 [FTRACE] Sometimes tracing is used to debug issues 7822 during the boot process. Since the trace buffer has a 7823 limited amount of storage, it may be prudent to 7824 disable tracing after the boot is finished, otherwise 7825 the critical information may be overwritten. With this 7826 option, the main tracing buffer will be turned off at 7827 the end of the boot process. 7828 7829 traceoff_on_warning 7830 [FTRACE] enable this option to disable tracing when a 7831 warning is hit. This turns off "tracing_on". Tracing can 7832 be enabled again by echoing '1' into the "tracing_on" 7833 file located in /sys/kernel/tracing/ 7834 7835 This option is useful, as it disables the trace before 7836 the WARNING dump is called, which prevents the trace to 7837 be filled with content caused by the warning output. 7838 7839 This option can also be set at run time via the sysctl 7840 option: kernel/traceoff_on_warning 7841 7842 transparent_hugepage= 7843 [KNL] 7844 Format: [always|madvise|never] 7845 Can be used to control the default behavior of the system 7846 with respect to transparent hugepages. 7847 See Documentation/admin-guide/mm/transhuge.rst 7848 for more details. 7849 7850 transparent_hugepage_shmem= [KNL] 7851 Format: [always|within_size|advise|never|deny|force] 7852 Can be used to control the hugepage allocation policy for 7853 the internal shmem mount. 7854 See Documentation/admin-guide/mm/transhuge.rst 7855 for more details. 7856 7857 transparent_hugepage_tmpfs= [KNL] 7858 Format: [always|within_size|advise|never] 7859 Can be used to control the default hugepage allocation policy 7860 for the tmpfs mount. 7861 See Documentation/admin-guide/mm/transhuge.rst 7862 for more details. 7863 7864 trusted.source= [KEYS] 7865 Format: <string> 7866 This parameter identifies the trust source as a backend 7867 for trusted keys implementation. Supported trust 7868 sources: 7869 - "tpm" 7870 - "tee" 7871 - "caam" 7872 - "dcp" 7873 - "pkwm" 7874 If not specified then it defaults to iterating through 7875 the trust source list starting with TPM and assigns the 7876 first trust source as a backend which is initialized 7877 successfully during iteration. 7878 7879 trusted.rng= [KEYS] 7880 Format: <string> 7881 The RNG used to generate key material for trusted keys. 7882 Can be one of: 7883 - "kernel" 7884 - the same value as trusted.source: "tpm" or "tee" 7885 - "default" 7886 If not specified, "default" is used. In this case, 7887 the RNG's choice is left to each individual trust source. 7888 7889 trusted.dcp_use_otp_key 7890 This is intended to be used in combination with 7891 trusted.source=dcp and will select the DCP OTP key 7892 instead of the DCP UNIQUE key blob encryption. 7893 7894 trusted.dcp_skip_zk_test 7895 This is intended to be used in combination with 7896 trusted.source=dcp and will disable the check if the 7897 blob key is all zeros. This is helpful for situations where 7898 having this key zero'ed is acceptable. E.g. in testing 7899 scenarios. 7900 7901 tsa= [X86] Control mitigation for Transient Scheduler 7902 Attacks on AMD CPUs. Search the following in your 7903 favourite search engine for more details: 7904 7905 "Technical guidance for mitigating transient scheduler 7906 attacks". 7907 7908 off - disable the mitigation 7909 on - enable the mitigation (default) 7910 user - mitigate only user/kernel transitions 7911 vm - mitigate only guest/host transitions 7912 7913 7914 tsc= Disable clocksource stability checks for TSC. 7915 Format: <string> 7916 [x86] reliable: mark tsc clocksource as reliable, this 7917 disables clocksource verification at runtime, as well 7918 as the stability checks done at bootup. Used to enable 7919 high-resolution timer mode on older hardware, and in 7920 virtualized environment. 7921 [x86] noirqtime: Do not use TSC to do irq accounting. 7922 Used to run time disable IRQ_TIME_ACCOUNTING on any 7923 platforms where RDTSC is slow and this accounting 7924 can add overhead. 7925 [x86] unstable: mark the TSC clocksource as unstable, this 7926 marks the TSC unconditionally unstable at bootup and 7927 avoids any further wobbles once the TSC watchdog notices. 7928 [x86] nowatchdog: disable clocksource watchdog. Used 7929 in situations with strict latency requirements (where 7930 interruptions from clocksource watchdog are not 7931 acceptable). 7932 [x86] recalibrate: force recalibration against a HW timer 7933 (HPET or PM timer) on systems whose TSC frequency was 7934 obtained from HW or FW using either an MSR or CPUID(0x15). 7935 Warn if the difference is more than 500 ppm. 7936 [x86] watchdog: Enforce the clocksource watchdog on TSC 7937 7938 tsc_early_khz= [X86,EARLY] Skip early TSC calibration and use the given 7939 value instead. Useful when the early TSC frequency discovery 7940 procedure is not reliable, such as on overclocked systems 7941 with CPUID.16h support and partial CPUID.15h support. 7942 Format: <unsigned int> 7943 7944 tsx= [X86] Control Transactional Synchronization 7945 Extensions (TSX) feature in Intel processors that 7946 support TSX control. 7947 7948 This parameter controls the TSX feature. The options are: 7949 7950 on - Enable TSX on the system. Although there are 7951 mitigations for all known security vulnerabilities, 7952 TSX has been known to be an accelerator for 7953 several previous speculation-related CVEs, and 7954 so there may be unknown security risks associated 7955 with leaving it enabled. 7956 7957 off - Disable TSX on the system. (Note that this 7958 option takes effect only on newer CPUs which are 7959 not vulnerable to MDS, i.e., have 7960 MSR_IA32_ARCH_CAPABILITIES.MDS_NO=1 and which get 7961 the new IA32_TSX_CTRL MSR through a microcode 7962 update. This new MSR allows for the reliable 7963 deactivation of the TSX functionality.) 7964 7965 auto - Disable TSX if X86_BUG_TAA is present, 7966 otherwise enable TSX on the system. 7967 7968 Not specifying this option is equivalent to tsx=off. 7969 7970 See Documentation/admin-guide/hw-vuln/tsx_async_abort.rst 7971 for more details. 7972 7973 tsx_async_abort= [X86,INTEL,EARLY] Control mitigation for the TSX Async 7974 Abort (TAA) vulnerability. 7975 7976 Similar to Micro-architectural Data Sampling (MDS) 7977 certain CPUs that support Transactional 7978 Synchronization Extensions (TSX) are vulnerable to an 7979 exploit against CPU internal buffers which can forward 7980 information to a disclosure gadget under certain 7981 conditions. 7982 7983 In vulnerable processors, the speculatively forwarded 7984 data can be used in a cache side channel attack, to 7985 access data to which the attacker does not have direct 7986 access. 7987 7988 This parameter controls the TAA mitigation. The 7989 options are: 7990 7991 full - Enable TAA mitigation on vulnerable CPUs 7992 if TSX is enabled. 7993 7994 full,nosmt - Enable TAA mitigation and disable SMT on 7995 vulnerable CPUs. If TSX is disabled, SMT 7996 is not disabled because CPU is not 7997 vulnerable to cross-thread TAA attacks. 7998 off - Unconditionally disable TAA mitigation 7999 8000 On MDS-affected machines, tsx_async_abort=off can be 8001 prevented by an active MDS mitigation as both vulnerabilities 8002 are mitigated with the same mechanism so in order to disable 8003 this mitigation, you need to specify mds=off too. 8004 8005 Not specifying this option is equivalent to 8006 tsx_async_abort=full. On CPUs which are MDS affected 8007 and deploy MDS mitigation, TAA mitigation is not 8008 required and doesn't provide any additional 8009 mitigation. 8010 8011 For details see: 8012 Documentation/admin-guide/hw-vuln/tsx_async_abort.rst 8013 8014 turbografx.map[2|3]= [HW,JOY] 8015 TurboGraFX parallel port interface 8016 Format: 8017 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7> 8018 See also Documentation/input/devices/joystick-parport.rst 8019 8020 udbg-immortal [PPC] When debugging early kernel crashes that 8021 happen after console_init() and before a proper 8022 console driver takes over, this boot options might 8023 help "seeing" what's going on. 8024 8025 uhash_entries= [KNL,NET] 8026 Set number of hash buckets for UDP/UDP-Lite connections 8027 8028 uhci-hcd.ignore_oc= 8029 [USB] Ignore overcurrent events (default N). 8030 Some badly-designed motherboards generate lots of 8031 bogus events, for ports that aren't wired to 8032 anything. Set this parameter to avoid log spamming. 8033 Note that genuine overcurrent events won't be 8034 reported either. 8035 8036 unaligned_scalar_speed= 8037 [RISCV] 8038 Format: {slow | fast | unsupported} 8039 Allow skipping scalar unaligned access speed tests. This 8040 is useful for testing alternative code paths and to skip 8041 the tests in environments where they run too slowly. All 8042 CPUs must have the same scalar unaligned access speed. 8043 8044 unaligned_vector_speed= 8045 [RISCV] 8046 Format: {slow | fast | unsupported} 8047 Allow skipping vector unaligned access speed tests. This 8048 is useful for testing alternative code paths and to skip 8049 the tests in environments where they run too slowly. All 8050 CPUs must have the same vector unaligned access speed. 8051 8052 unknown_nmi_panic 8053 [X86] Cause panic on unknown NMI. 8054 8055 unwind_debug [X86-64,EARLY] 8056 Enable unwinder debug output. This can be 8057 useful for debugging certain unwinder error 8058 conditions, including corrupt stacks and 8059 bad/missing unwinder metadata. 8060 8061 usbcore.authorized_default= 8062 [USB] Default USB device authorization: 8063 (default -1 = authorized (same as 1), 8064 0 = not authorized, 1 = authorized, 2 = authorized 8065 if device connected to internal port) 8066 8067 usbcore.autosuspend= 8068 [USB] The autosuspend time delay (in seconds) used 8069 for newly-detected USB devices (default 2). This 8070 is the time required before an idle device will be 8071 autosuspended. Devices for which the delay is set 8072 to a negative value won't be autosuspended at all. 8073 8074 usbcore.usbfs_snoop= 8075 [USB] Set to log all usbfs traffic (default 0 = off). 8076 8077 usbcore.usbfs_snoop_max= 8078 [USB] Maximum number of bytes to snoop in each URB 8079 (default = 65536). 8080 8081 usbcore.blinkenlights= 8082 [USB] Set to cycle leds on hubs (default 0 = off). 8083 8084 usbcore.old_scheme_first= 8085 [USB] Start with the old device initialization 8086 scheme (default 0 = off). 8087 8088 usbcore.usbfs_memory_mb= 8089 [USB] Memory limit (in MB) for buffers allocated by 8090 usbfs (default = 16, 0 = max = 2047). 8091 8092 usbcore.use_both_schemes= 8093 [USB] Try the other device initialization scheme 8094 if the first one fails (default 1 = enabled). 8095 8096 usbcore.initial_descriptor_timeout= 8097 [USB] Specifies timeout for the initial 64-byte 8098 USB_REQ_GET_DESCRIPTOR request in milliseconds 8099 (default 5000 = 5.0 seconds). 8100 8101 usbcore.nousb [USB] Disable the USB subsystem 8102 8103 usbcore.quirks= 8104 [USB] A list of quirk entries to augment the built-in 8105 usb core quirk list. List entries are separated by 8106 commas. Each entry has the form 8107 VendorID:ProductID:Flags. The IDs are 4-digit hex 8108 numbers and Flags is a set of letters. Each letter 8109 will change the built-in quirk; setting it if it is 8110 clear and clearing it if it is set. The letters have 8111 the following meanings: 8112 a = USB_QUIRK_STRING_FETCH_255 (string 8113 descriptors must not be fetched using 8114 a 255-byte read); 8115 b = USB_QUIRK_RESET_RESUME (device can't resume 8116 correctly so reset it instead); 8117 c = USB_QUIRK_NO_SET_INTF (device can't handle 8118 Set-Interface requests); 8119 d = USB_QUIRK_CONFIG_INTF_STRINGS (device can't 8120 handle its Configuration or Interface 8121 strings); 8122 e = USB_QUIRK_RESET (device can't be reset 8123 (e.g morph devices), don't use reset); 8124 f = USB_QUIRK_HONOR_BNUMINTERFACES (device has 8125 more interface descriptions than the 8126 bNumInterfaces count, and can't handle 8127 talking to these interfaces); 8128 g = USB_QUIRK_DELAY_INIT (device needs a pause 8129 during initialization, after we read 8130 the device descriptor); 8131 h = USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL (For 8132 high speed and super speed interrupt 8133 endpoints, the USB 2.0 and USB 3.0 spec 8134 require the interval in microframes (1 8135 microframe = 125 microseconds) to be 8136 calculated as interval = 2 ^ 8137 (bInterval-1). 8138 Devices with this quirk report their 8139 bInterval as the result of this 8140 calculation instead of the exponent 8141 variable used in the calculation); 8142 i = USB_QUIRK_DEVICE_QUALIFIER (device can't 8143 handle device_qualifier descriptor 8144 requests); 8145 j = USB_QUIRK_IGNORE_REMOTE_WAKEUP (device 8146 generates spurious wakeup, ignore 8147 remote wakeup capability); 8148 k = USB_QUIRK_NO_LPM (device can't handle Link 8149 Power Management); 8150 l = USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL 8151 (Device reports its bInterval as linear 8152 frames instead of the USB 2.0 8153 calculation); 8154 m = USB_QUIRK_DISCONNECT_SUSPEND (Device needs 8155 to be disconnected before suspend to 8156 prevent spurious wakeup); 8157 n = USB_QUIRK_DELAY_CTRL_MSG (Device needs a 8158 pause after every control message); 8159 o = USB_QUIRK_HUB_SLOW_RESET (Hub needs extra 8160 delay after resetting its port); 8161 p = USB_QUIRK_SHORT_SET_ADDRESS_REQ_TIMEOUT 8162 (Reduce timeout of the SET_ADDRESS 8163 request from 5000 ms to 500 ms); 8164 q = USB_QUIRK_FORCE_ONE_CONFIG (Device 8165 claims zero configurations, 8166 forcing to 1); 8167 Example: quirks=0781:5580:bk,0a5c:5834:gij 8168 8169 usbhid.mousepoll= 8170 [USBHID] The interval which mice are to be polled at. 8171 8172 usbhid.jspoll= 8173 [USBHID] The interval which joysticks are to be polled at. 8174 8175 usbhid.kbpoll= 8176 [USBHID] The interval which keyboards are to be polled at. 8177 8178 usb-storage.delay_use= 8179 [UMS] The delay in seconds before a new device is 8180 scanned for Logical Units (default 1). 8181 Optionally the delay in milliseconds if the value has 8182 suffix with "ms". 8183 Example: delay_use=2567ms 8184 8185 usb-storage.quirks= 8186 [UMS] A list of quirks entries to supplement or 8187 override the built-in unusual_devs list. List 8188 entries are separated by commas. Each entry has 8189 the form VID:PID:Flags where VID and PID are Vendor 8190 and Product ID values (4-digit hex numbers) and 8191 Flags is a set of characters, each corresponding 8192 to a common usb-storage quirk flag as follows: 8193 a = SANE_SENSE (collect more than 18 bytes 8194 of sense data, not on uas); 8195 b = BAD_SENSE (don't collect more than 18 8196 bytes of sense data, not on uas); 8197 c = FIX_CAPACITY (decrease the reported 8198 device capacity by one sector); 8199 d = NO_READ_DISC_INFO (don't use 8200 READ_DISC_INFO command, not on uas); 8201 e = NO_READ_CAPACITY_16 (don't use 8202 READ_CAPACITY_16 command); 8203 f = NO_REPORT_OPCODES (don't use report opcodes 8204 command, uas only); 8205 g = MAX_SECTORS_240 (don't transfer more than 8206 240 sectors at a time, uas only); 8207 h = CAPACITY_HEURISTICS (decrease the 8208 reported device capacity by one 8209 sector if the number is odd); 8210 i = IGNORE_DEVICE (don't bind to this 8211 device); 8212 j = NO_REPORT_LUNS (don't use report luns 8213 command, uas only); 8214 k = NO_SAME (do not use WRITE_SAME, uas only) 8215 l = NOT_LOCKABLE (don't try to lock and 8216 unlock ejectable media, not on uas); 8217 m = MAX_SECTORS_64 (don't transfer more 8218 than 64 sectors = 32 KB at a time, 8219 not on uas); 8220 n = INITIAL_READ10 (force a retry of the 8221 initial READ(10) command, not on uas); 8222 o = CAPACITY_OK (accept the capacity 8223 reported by the device, not on uas); 8224 p = WRITE_CACHE (the device cache is ON 8225 by default, not on uas); 8226 r = IGNORE_RESIDUE (the device reports 8227 bogus residue values, not on uas); 8228 s = SINGLE_LUN (the device has only one 8229 Logical Unit); 8230 t = NO_ATA_1X (don't allow ATA(12) and ATA(16) 8231 commands, uas only); 8232 u = IGNORE_UAS (don't bind to the uas driver); 8233 w = NO_WP_DETECT (don't test whether the 8234 medium is write-protected). 8235 y = ALWAYS_SYNC (issue a SYNCHRONIZE_CACHE 8236 even if the device claims no cache, 8237 not on uas) 8238 Example: quirks=0419:aaf5:rl,0421:0433:rc 8239 8240 user_debug= [KNL,ARM] 8241 Format: <int> 8242 See arch/arm/Kconfig.debug help text. 8243 1 - undefined instruction events 8244 2 - system calls 8245 4 - invalid data aborts 8246 8 - SIGSEGV faults 8247 16 - SIGBUS faults 8248 Example: user_debug=31 8249 8250 vdso= [X86,SH,SPARC] 8251 On X86_32, this is an alias for vdso32=. Otherwise: 8252 8253 vdso=1: enable VDSO (the default) 8254 vdso=0: disable VDSO mapping 8255 8256 vdso32= [X86] Control the 32-bit vDSO 8257 vdso32=1: enable 32-bit VDSO 8258 vdso32=0 or vdso32=2: disable 32-bit VDSO 8259 8260 See the help text for CONFIG_COMPAT_VDSO for more 8261 details. If CONFIG_COMPAT_VDSO is set, the default is 8262 vdso32=0; otherwise, the default is vdso32=1. 8263 8264 For compatibility with older kernels, vdso32=2 is an 8265 alias for vdso32=0. 8266 8267 Try vdso32=0 if you encounter an error that says: 8268 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed! 8269 8270 video= [FB,EARLY] Frame buffer configuration 8271 See Documentation/fb/modedb.rst. 8272 8273 video.brightness_switch_enabled= [ACPI] 8274 Format: [0|1] 8275 If set to 1, on receiving an ACPI notify event 8276 generated by hotkey, video driver will adjust brightness 8277 level and then send out the event to user space through 8278 the allocated input device. If set to 0, video driver 8279 will only send out the event without touching backlight 8280 brightness level. 8281 default: 1 8282 8283 virtio_mmio.device= 8284 [VMMIO] Memory mapped virtio (platform) device. 8285 8286 <size>@<baseaddr>:<irq>[:<id>] 8287 where: 8288 <size> := size (can use standard suffixes 8289 like K, M and G) 8290 <baseaddr> := physical base address 8291 <irq> := interrupt number (as passed to 8292 request_irq()) 8293 <id> := (optional) platform device id 8294 example: 8295 virtio_mmio.device=1K@0x100b0000:48:7 8296 8297 Can be used multiple times for multiple devices. 8298 8299 vga= [BOOT,X86-32] Select a particular video mode 8300 See Documentation/arch/x86/boot.rst and 8301 Documentation/admin-guide/svga.rst. 8302 Use vga=ask for menu. 8303 This is actually a boot loader parameter; the value is 8304 passed to the kernel using a special protocol. 8305 8306 vm_debug[=options] [KNL] Available with CONFIG_DEBUG_VM=y. 8307 May slow down system boot speed, especially when 8308 enabled on systems with a large amount of memory. 8309 All options are enabled by default, and this 8310 interface is meant to allow for selectively 8311 enabling or disabling specific virtual memory 8312 debugging features. 8313 8314 Available options are: 8315 P Enable page structure init time poisoning 8316 - Disable all of the above options 8317 8318 vmalloc=nn[KMG] [KNL,BOOT,EARLY] Forces the vmalloc area to have an 8319 exact size of <nn>. This can be used to increase 8320 the minimum size (128MB on x86, arm32 platforms). 8321 It can also be used to decrease the size and leave more room 8322 for directly mapped kernel RAM. Note that this parameter does 8323 not exist on many other platforms (including arm64, alpha, 8324 loongarch, arc, csky, hexagon, microblaze, mips, nios2, openrisc, 8325 parisc, m64k, powerpc, riscv, sh, um, xtensa, s390, sparc). 8326 8327 vmcp_cma=nn[MG] [KNL,S390,EARLY] 8328 Sets the memory size reserved for contiguous memory 8329 allocations for the vmcp device driver. 8330 8331 vmhalt= [KNL,S390] Perform z/VM CP command after system halt. 8332 Format: <command> 8333 8334 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic. 8335 Format: <command> 8336 8337 vmpoff= [KNL,S390] Perform z/VM CP command after power off. 8338 Format: <command> 8339 8340 vmscape= [X86] Controls mitigation for VMscape attacks. 8341 VMscape attacks can leak information from a userspace 8342 hypervisor to a guest via speculative side-channels. 8343 8344 off - disable the mitigation 8345 ibpb - use Indirect Branch Prediction Barrier 8346 (IBPB) mitigation (default) 8347 force - force vulnerability detection even on 8348 unaffected processors 8349 8350 vsyscall= [X86-64,EARLY] 8351 Controls the behavior of vsyscalls (i.e. calls to 8352 fixed addresses of 0xffffffffff600x00 from legacy 8353 code). Most statically-linked binaries and older 8354 versions of glibc use these calls. Because these 8355 functions are at fixed addresses, they make nice 8356 targets for exploits that can control RIP. 8357 8358 emulate Vsyscalls turn into traps and are emulated 8359 reasonably safely. The vsyscall page is 8360 readable. This disables the Linear 8361 Address Space Separation (LASS) security 8362 feature and makes the system less secure. 8363 8364 xonly [default] Vsyscalls turn into traps and are 8365 emulated reasonably safely. The vsyscall 8366 page is not readable. 8367 8368 none Vsyscalls don't work at all. This makes 8369 them quite hard to use for exploits but 8370 might break your system. 8371 8372 vt.color= [VT] Default text color. 8373 Format: 0xYX, X = foreground, Y = background. 8374 Default: 0x07 = light gray on black. 8375 8376 vt.cur_default= [VT] Default cursor shape. 8377 Format: 0xCCBBAA, where AA, BB, and CC are the same as 8378 the parameters of the <Esc>[?A;B;Cc escape sequence; 8379 see vga-softcursor.rst. Default: 2 = underline. 8380 8381 vt.default_blu= [VT] 8382 Format: <blue0>,<blue1>,<blue2>,...,<blue15> 8383 Change the default blue palette of the console. 8384 This is a 16-member array composed of values 8385 ranging from 0-255. 8386 8387 vt.default_grn= [VT] 8388 Format: <green0>,<green1>,<green2>,...,<green15> 8389 Change the default green palette of the console. 8390 This is a 16-member array composed of values 8391 ranging from 0-255. 8392 8393 vt.default_red= [VT] 8394 Format: <red0>,<red1>,<red2>,...,<red15> 8395 Change the default red palette of the console. 8396 This is a 16-member array composed of values 8397 ranging from 0-255. 8398 8399 vt.default_utf8= 8400 [VT] 8401 Format=<0|1> 8402 Set system-wide default UTF-8 mode for all tty's. 8403 Default is 1, i.e. UTF-8 mode is enabled for all 8404 newly opened terminals. 8405 8406 vt.global_cursor_default= 8407 [VT] 8408 Format=<-1|0|1> 8409 Set system-wide default for whether a cursor 8410 is shown on new VTs. Default is -1, 8411 i.e. cursors will be created by default unless 8412 overridden by individual drivers. 0 will hide 8413 cursors, 1 will display them. 8414 8415 vt.italic= [VT] Default color for italic text; 0-15. 8416 Default: 2 = green. 8417 8418 vt.underline= [VT] Default color for underlined text; 0-15. 8419 Default: 3 = cyan. 8420 8421 watchdog timers [HW,WDT] For information on watchdog timers, 8422 see Documentation/watchdog/watchdog-parameters.rst 8423 or other driver-specific files in the 8424 Documentation/watchdog/ directory. 8425 8426 watchdog_thresh= 8427 [KNL] 8428 Set the hard lockup detector stall duration 8429 threshold in seconds. The soft lockup detector 8430 threshold is set to twice the value. A value of 0 8431 disables both lockup detectors. Default is 10 8432 seconds. 8433 8434 workqueue.unbound_cpus= 8435 [KNL,SMP] Specify to constrain one or some CPUs 8436 to use in unbound workqueues. 8437 Format: <cpu-list> 8438 By default, all online CPUs are available for 8439 unbound workqueues. 8440 8441 workqueue.watchdog_thresh= 8442 If CONFIG_WQ_WATCHDOG is configured, workqueue can 8443 warn stall conditions and dump internal state to 8444 help debugging. 0 disables workqueue stall 8445 detection; otherwise, it's the stall threshold 8446 duration in seconds. The default value is 30 and 8447 it can be updated at runtime by writing to the 8448 corresponding sysfs file. 8449 8450 workqueue.panic_on_stall=<uint> 8451 Panic when workqueue stall is detected by 8452 CONFIG_WQ_WATCHDOG. It sets the number times of the 8453 stall to trigger panic. 8454 8455 The default is set by CONFIG_BOOTPARAM_WQ_STALL_PANIC, 8456 which is 0 (disabled) if not configured. 8457 8458 workqueue.panic_on_stall_time=<uint> 8459 Panic when a workqueue stall has been continuous for 8460 the specified number of seconds. Unlike panic_on_stall 8461 which counts accumulated stall events, this triggers 8462 based on the duration of a single continuous stall. 8463 8464 The default is 0, which disables the time-based panic. 8465 8466 workqueue.cpu_intensive_thresh_us= 8467 Per-cpu work items which run for longer than this 8468 threshold are automatically considered CPU intensive 8469 and excluded from concurrency management to prevent 8470 them from noticeably delaying other per-cpu work 8471 items. Default is 10000 (10ms). 8472 8473 If CONFIG_WQ_CPU_INTENSIVE_REPORT is set, the kernel 8474 will report the work functions which violate this 8475 threshold repeatedly. They are likely good 8476 candidates for using WQ_UNBOUND workqueues instead. 8477 8478 workqueue.cpu_intensive_warning_thresh=<uint> 8479 If CONFIG_WQ_CPU_INTENSIVE_REPORT is set, the kernel 8480 will report the work functions which violate the 8481 intensive_threshold_us repeatedly. In order to prevent 8482 spurious warnings, start printing only after a work 8483 function has violated this threshold number of times. 8484 8485 The default is 4 times. 0 disables the warning. 8486 8487 workqueue.power_efficient 8488 Per-cpu workqueues are generally preferred because 8489 they show better performance thanks to cache 8490 locality; unfortunately, per-cpu workqueues tend to 8491 be more power hungry than unbound workqueues. 8492 8493 Enabling this makes the per-cpu workqueues which 8494 were observed to contribute significantly to power 8495 consumption unbound, leading to measurably lower 8496 power usage at the cost of small performance 8497 overhead. 8498 8499 The default value of this parameter is determined by 8500 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT. 8501 8502 workqueue.default_affinity_scope= 8503 Select the default affinity scope to use for unbound 8504 workqueues. Can be one of "cpu", "smt", "cache", 8505 "cache_shard", "numa" and "system". Default is 8506 "cache_shard". For more 8507 information, see the Affinity Scopes section in 8508 Documentation/core-api/workqueue.rst. 8509 8510 This can be changed after boot by writing to the 8511 matching /sys/module/workqueue/parameters file. All 8512 workqueues with the "default" affinity scope will be 8513 updated accordingly. 8514 8515 workqueue.debug_force_rr_cpu 8516 Workqueue used to implicitly guarantee that work 8517 items queued without explicit CPU specified are put 8518 on the local CPU. This guarantee is no longer true 8519 and while local CPU is still preferred work items 8520 may be put on foreign CPUs. This debug option 8521 forces round-robin CPU selection to flush out 8522 usages which depend on the now broken guarantee. 8523 When enabled, memory and cache locality will be 8524 impacted. 8525 8526 writecombine= [LOONGARCH,EARLY] Control the MAT (Memory Access 8527 Type) of ioremap_wc(). 8528 8529 on - Enable writecombine, use WUC for ioremap_wc() 8530 off - Disable writecombine, use SUC for ioremap_wc() 8531 8532 x2apic_phys [X86-64,APIC,EARLY] Use x2apic physical mode instead of 8533 default x2apic cluster mode on platforms 8534 supporting x2apic. 8535 8536 xen_512gb_limit [KNL,X86-64,XEN] 8537 Restricts the kernel running paravirtualized under Xen 8538 to use only up to 512 GB of RAM. The reason to do so is 8539 crash analysis tools and Xen tools for doing domain 8540 save/restore/migration must be enabled to handle larger 8541 domains. 8542 8543 xen_console_io [XEN,EARLY] 8544 Boolean option to enable/disable the usage of the Xen 8545 console_io hypercalls to read and write to the console. 8546 Mostly useful for debugging and development. 8547 8548 xen_emul_unplug= [HW,X86,XEN,EARLY] 8549 Unplug Xen emulated devices 8550 Format: [unplug0,][unplug1] 8551 ide-disks -- unplug primary master IDE devices 8552 aux-ide-disks -- unplug non-primary-master IDE devices 8553 nics -- unplug network devices 8554 all -- unplug all emulated devices (NICs and IDE disks) 8555 unnecessary -- unplugging emulated devices is 8556 unnecessary even if the host did not respond to 8557 the unplug protocol 8558 never -- do not unplug even if version check succeeds 8559 8560 xen_legacy_crash [X86,XEN,EARLY] 8561 Crash from Xen panic notifier, without executing late 8562 panic() code such as dumping handler. 8563 8564 xen_mc_debug [X86,XEN,EARLY] 8565 Enable multicall debugging when running as a Xen PV guest. 8566 Enabling this feature will reduce performance a little 8567 bit, so it should only be enabled for obtaining extended 8568 debug data in case of multicall errors. 8569 8570 xen_msr_safe= [X86,XEN,EARLY] 8571 Format: <bool> 8572 Select whether to always use non-faulting (safe) MSR 8573 access functions when running as Xen PV guest. The 8574 default value is controlled by CONFIG_XEN_PV_MSR_SAFE. 8575 8576 xen_nopv [X86] 8577 Disables the PV optimizations forcing the HVM guest to 8578 run as generic HVM guest with no PV drivers. 8579 This option is obsoleted by the "nopv" option, which 8580 has equivalent effect for XEN platform. 8581 8582 xen_no_vector_callback 8583 [KNL,X86,XEN,EARLY] Disable the vector callback for Xen 8584 event channel interrupts. 8585 8586 xen_scrub_pages= [XEN] 8587 Boolean option to control scrubbing pages before giving them back 8588 to Xen, for use by other domains. Can be also changed at runtime 8589 with /sys/devices/system/xen_memory/xen_memory0/scrub_pages. 8590 Default value controlled with CONFIG_XEN_SCRUB_PAGES_DEFAULT. 8591 8592 xen_timer_slop= [X86-64,XEN,EARLY] 8593 Set the timer slop (in nanoseconds) for the virtual Xen 8594 timers (default is 100000). This adjusts the minimum 8595 delta of virtualized Xen timers, where lower values 8596 improve timer resolution at the expense of processing 8597 more timer interrupts. 8598 8599 xen.balloon_boot_timeout= [XEN] 8600 The time (in seconds) to wait before giving up to boot 8601 in case initial ballooning fails to free enough memory. 8602 Applies only when running as HVM or PVH guest and 8603 started with less memory configured than allowed at 8604 max. Default is 180. 8605 8606 xen.event_eoi_delay= [XEN] 8607 How long to delay EOI handling in case of event 8608 storms (jiffies). Default is 10. 8609 8610 xen.event_loop_timeout= [XEN] 8611 After which time (jiffies) the event handling loop 8612 should start to delay EOI handling. Default is 2. 8613 8614 xen.fifo_events= [XEN] 8615 Boolean parameter to disable using fifo event handling 8616 even if available. Normally fifo event handling is 8617 preferred over the 2-level event handling, as it is 8618 fairer and the number of possible event channels is 8619 much higher. Default is on (use fifo events). 8620 8621 xirc2ps_cs= [NET,PCMCIA] 8622 Format: 8623 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]] 8624 8625 xive= [PPC] 8626 By default on POWER9 and above, the kernel will 8627 natively use the XIVE interrupt controller. This option 8628 allows the fallback firmware mode to be used: 8629 8630 off Fallback to firmware control of XIVE interrupt 8631 controller on both pseries and powernv 8632 platforms. Only useful on POWER9 and above. 8633 8634 xive.store-eoi=off [PPC] 8635 By default on POWER10 and above, the kernel will use 8636 stores for EOI handling when the XIVE interrupt mode 8637 is active. This option allows the XIVE driver to use 8638 loads instead, as on POWER9. 8639 8640 xhci-hcd.quirks [USB,KNL] 8641 A hex value specifying bitmask with supplemental xhci 8642 host controller quirks. Meaning of each bit can be 8643 consulted in header drivers/usb/host/xhci.h. 8644 8645 xmon [PPC,EARLY] 8646 Format: { early | on | rw | ro | off } 8647 Controls if xmon debugger is enabled. Default is off. 8648 Passing only "xmon" is equivalent to "xmon=early". 8649 early Call xmon as early as possible on boot; xmon 8650 debugger is called from setup_arch(). 8651 on xmon debugger hooks will be installed so xmon 8652 is only called on a kernel crash. Default mode, 8653 i.e. either "ro" or "rw" mode, is controlled 8654 with CONFIG_XMON_DEFAULT_RO_MODE. 8655 rw xmon debugger hooks will be installed so xmon 8656 is called only on a kernel crash, mode is write, 8657 meaning SPR registers, memory and, other data 8658 can be written using xmon commands. 8659 ro same as "rw" option above but SPR registers, 8660 memory, and other data can't be written using 8661 xmon commands. 8662 off xmon is disabled. 8663