1 ACPI ACPI support is enabled. 2 AGP AGP (Accelerated Graphics Port) is enabled. 3 ALSA ALSA sound support is enabled. 4 APIC APIC support is enabled. 5 APM Advanced Power Management support is enabled. 6 APPARMOR AppArmor support is enabled. 7 ARM ARM architecture is enabled. 8 ARM64 ARM64 architecture is enabled. 9 CLK Common clock infrastructure is enabled. 10 CMA Contiguous Memory Area support is enabled. 11 DRM Direct Rendering Management support is enabled. 12 DYNAMIC_DEBUG Build in debug messages and enable them at runtime 13 EARLY Parameter processed too early to be embedded in initrd. 14 EDD BIOS Enhanced Disk Drive Services (EDD) is enabled 15 EFI EFI Partitioning (GPT) is enabled 16 EVM Extended Verification Module 17 FB The frame buffer device is enabled. 18 FTRACE Function tracing enabled. 19 GCOV GCOV profiling is enabled. 20 HIBERNATION HIBERNATION is enabled. 21 HW Appropriate hardware is enabled. 22 HYPER_V HYPERV support is enabled. 23 IMA Integrity measurement architecture is enabled. 24 IP_PNP IP DHCP, BOOTP, or RARP is enabled. 25 IPV6 IPv6 support is enabled. 26 ISAPNP ISA PnP code is enabled. 27 ISOL CPU Isolation is enabled. 28 JOY Appropriate joystick support is enabled. 29 KGDB Kernel debugger support is enabled. 30 KVM Kernel Virtual Machine support is enabled. 31 LIBATA Libata driver is enabled 32 LOONGARCH LoongArch architecture is enabled. 33 LOOP Loopback device support is enabled. 34 LP Printer support is enabled. 35 M68k M68k architecture is enabled. 36 These options have more detailed description inside of 37 Documentation/arch/m68k/kernel-options.rst. 38 MIPS MIPS architecture is enabled. 39 MOUSE Appropriate mouse support is enabled. 40 MSI Message Signaled Interrupts (PCI). 41 MTD MTD (Memory Technology Device) support is enabled. 42 NET Appropriate network support is enabled. 43 NFS Appropriate NFS support is enabled. 44 NUMA NUMA support is enabled. 45 OF Devicetree is enabled. 46 PARISC The PA-RISC architecture is enabled. 47 PCI PCI bus support is enabled. 48 PCIE PCI Express support is enabled. 49 PCMCIA The PCMCIA subsystem is enabled. 50 PNP Plug & Play support is enabled. 51 PPC PowerPC architecture is enabled. 52 PPT Parallel port support is enabled. 53 PS2 Appropriate PS/2 support is enabled. 54 PV_OPS A paravirtualized kernel is enabled. 55 RAM RAM disk support is enabled. 56 RDT Intel Resource Director Technology. 57 RISCV RISCV architecture is enabled. 58 S390 S390 architecture is enabled. 59 SCSI Appropriate SCSI support is enabled. 60 A lot of drivers have their options described inside 61 the Documentation/scsi/ sub-directory. 62 SDW SoundWire support is enabled. 63 SECURITY Different security models are enabled. 64 SELINUX SELinux support is enabled. 65 SERIAL Serial support is enabled. 66 SH SuperH architecture is enabled. 67 SMP The kernel is an SMP kernel. 68 SPARC Sparc architecture is enabled. 69 SUSPEND System suspend states are enabled. 70 SWSUSP Software suspend (hibernation) is enabled. 71 TPM TPM drivers are enabled. 72 UMS USB Mass Storage support is enabled. 73 USB USB support is enabled. 74 NVME NVMe support is enabled 75 USBHID USB Human Interface Device support is enabled. 76 V4L Video For Linux support is enabled. 77 VGA The VGA console has been enabled. 78 VMMIO Driver for memory mapped virtio devices is enabled. 79 VT Virtual terminal support is enabled. 80 WDT Watchdog support is enabled. 81 X86-32 X86-32, aka i386 architecture is enabled. 82 X86-64 X86-64 architecture is enabled. 83 X86 Either 32-bit or 64-bit x86 (same as X86-32+X86-64) 84 X86_UV SGI UV support is enabled. 85 XEN Xen support is enabled 86 XTENSA xtensa architecture is enabled. 87 88In addition, the following text indicates that the option 89 90 BOOT Is a boot loader parameter. 91 BUGS= Relates to possible processor bugs on the said processor. 92 KNL Is a kernel start-up parameter. 93 94 95Kernel parameters 96 97 accept_memory= [MM] 98 Format: { eager | lazy } 99 default: lazy 100 By default, unaccepted memory is accepted lazily to 101 avoid prolonged boot times. The lazy option will add 102 some runtime overhead until all memory is eventually 103 accepted. In most cases the overhead is negligible. 104 For some workloads or for debugging purposes 105 accept_memory=eager can be used to accept all memory 106 at once during boot. 107 108 acpi= [HW,ACPI,X86,ARM64,RISCV64,EARLY] 109 Advanced Configuration and Power Interface 110 Format: { force | on | off | strict | noirq | rsdt | 111 copy_dsdt | nospcr } 112 force -- enable ACPI if default was off 113 on -- enable ACPI but allow fallback to DT [arm64,riscv64] 114 off -- disable ACPI if default was on 115 noirq -- do not use ACPI for IRQ routing 116 strict -- Be less tolerant of platforms that are not 117 strictly ACPI specification compliant. 118 rsdt -- prefer RSDT over (default) XSDT 119 copy_dsdt -- copy DSDT to memory 120 nocmcff -- Disable firmware first mode for corrected 121 errors. This disables parsing the HEST CMC error 122 source to check if firmware has set the FF flag. This 123 may result in duplicate corrected error reports. 124 nospcr -- disable console in ACPI SPCR table as 125 default _serial_ console on ARM64 126 spcr -- enable console in ACPI SPCR table as 127 default _serial_ console on x86 128 For ARM64, ONLY "acpi=off", "acpi=on", "acpi=force" or 129 "acpi=nospcr" are available 130 For RISCV64, ONLY "acpi=off", "acpi=on" or "acpi=force" 131 are available 132 133 See also Documentation/power/runtime_pm.rst, pci=noacpi 134 135 acpi_apic_instance= [ACPI,IOAPIC,EARLY] 136 Format: <int> 137 2: use 2nd APIC table, if available 138 1,0: use 1st APIC table 139 default: 0 140 141 acpi_backlight= [HW,ACPI] 142 { vendor | video | native | none } 143 If set to vendor, prefer vendor-specific driver 144 (e.g. thinkpad_acpi, sony_acpi, etc.) instead 145 of the ACPI video.ko driver. 146 If set to video, use the ACPI video.ko driver. 147 If set to native, use the device's native backlight mode. 148 If set to none, disable the ACPI backlight interface. 149 150 acpi_force_32bit_fadt_addr [ACPI,EARLY] 151 force FADT to use 32 bit addresses rather than the 152 64 bit X_* addresses. Some firmware have broken 64 153 bit addresses for force ACPI ignore these and use 154 the older legacy 32 bit addresses. 155 156 acpica_no_return_repair [HW, ACPI] 157 Disable AML predefined validation mechanism 158 This mechanism can repair the evaluation result to make 159 the return objects more ACPI specification compliant. 160 This option is useful for developers to identify the 161 root cause of an AML interpreter issue when the issue 162 has something to do with the repair mechanism. 163 164 acpi.debug_layer= [HW,ACPI,ACPI_DEBUG] 165 acpi.debug_level= [HW,ACPI,ACPI_DEBUG] 166 Format: <int> 167 CONFIG_ACPI_DEBUG must be enabled to produce any ACPI 168 debug output. Bits in debug_layer correspond to a 169 _COMPONENT in an ACPI source file, e.g., 170 #define _COMPONENT ACPI_EVENTS 171 Bits in debug_level correspond to a level in 172 ACPI_DEBUG_PRINT statements, e.g., 173 ACPI_DEBUG_PRINT((ACPI_DB_INFO, ... 174 The debug_level mask defaults to "info". See 175 Documentation/firmware-guide/acpi/debug.rst for more information about 176 debug layers and levels. 177 178 Enable processor driver info messages: 179 acpi.debug_layer=0x20000000 180 Enable AML "Debug" output, i.e., stores to the Debug 181 object while interpreting AML: 182 acpi.debug_layer=0xffffffff acpi.debug_level=0x2 183 Enable all messages related to ACPI hardware: 184 acpi.debug_layer=0x2 acpi.debug_level=0xffffffff 185 186 Some values produce so much output that the system is 187 unusable. The "log_buf_len" parameter may be useful 188 if you need to capture more output. 189 190 acpi.poweroff_on_fatal= [ACPI] 191 {0 | 1} 192 Causes the system to poweroff when the ACPI bytecode signals 193 a fatal error. The default value of this setting is 1. 194 Overriding this value should only be done for diagnosing 195 ACPI firmware problems, as the system might behave erratically 196 after having encountered a fatal ACPI error. 197 198 acpi_enforce_resources= [ACPI] 199 { strict | lax | no } 200 Check for resource conflicts between native drivers 201 and ACPI OperationRegions (SystemIO and SystemMemory 202 only). IO ports and memory declared in ACPI might be 203 used by the ACPI subsystem in arbitrary AML code and 204 can interfere with legacy drivers. 205 strict (default): access to resources claimed by ACPI 206 is denied; legacy drivers trying to access reserved 207 resources will fail to bind to device using them. 208 lax: access to resources claimed by ACPI is allowed; 209 legacy drivers trying to access reserved resources 210 will bind successfully but a warning message is logged. 211 no: ACPI OperationRegions are not marked as reserved, 212 no further checks are performed. 213 214 acpi_force_table_verification [HW,ACPI,EARLY] 215 Enable table checksum verification during early stage. 216 By default, this is disabled due to x86 early mapping 217 size limitation. 218 219 acpi_irq_balance [HW,ACPI] 220 ACPI will balance active IRQs 221 default in APIC mode 222 223 acpi_irq_nobalance [HW,ACPI] 224 ACPI will not move active IRQs (default) 225 default in PIC mode 226 227 acpi_irq_isa= [HW,ACPI] If irq_balance, mark listed IRQs used by ISA 228 Format: <irq>,<irq>... 229 230 acpi_irq_pci= [HW,ACPI] If irq_balance, clear listed IRQs for 231 use by PCI 232 Format: <irq>,<irq>... 233 234 acpi_mask_gpe= [HW,ACPI] 235 Due to the existence of _Lxx/_Exx, some GPEs triggered 236 by unsupported hardware/firmware features can result in 237 GPE floodings that cannot be automatically disabled by 238 the GPE dispatcher. 239 This facility can be used to prevent such uncontrolled 240 GPE floodings. 241 Format: <byte> or <bitmap-list> 242 243 acpi_no_auto_serialize [HW,ACPI] 244 Disable auto-serialization of AML methods 245 AML control methods that contain the opcodes to create 246 named objects will be marked as "Serialized" by the 247 auto-serialization feature. 248 This feature is enabled by default. 249 This option allows to turn off the feature. 250 251 acpi_no_memhotplug [ACPI] Disable memory hotplug. Useful for kdump 252 kernels. 253 254 acpi_no_static_ssdt [HW,ACPI,EARLY] 255 Disable installation of static SSDTs at early boot time 256 By default, SSDTs contained in the RSDT/XSDT will be 257 installed automatically and they will appear under 258 /sys/firmware/acpi/tables. 259 This option turns off this feature. 260 Note that specifying this option does not affect 261 dynamic table installation which will install SSDT 262 tables to /sys/firmware/acpi/tables/dynamic. 263 264 acpi_no_watchdog [HW,ACPI,WDT] 265 Ignore the ACPI-based watchdog interface (WDAT) and let 266 a native driver control the watchdog device instead. 267 268 acpi_rsdp= [ACPI,EFI,KEXEC,EARLY] 269 Pass the RSDP address to the kernel, mostly used 270 on machines running EFI runtime service to boot the 271 second kernel for kdump. 272 273 acpi_os_name= [HW,ACPI] Tell ACPI BIOS the name of the OS 274 Format: To spoof as Windows 98: ="Microsoft Windows" 275 276 acpi_rev_override [ACPI] Override the _REV object to return 5 (instead 277 of 2 which is mandated by ACPI 6) as the supported ACPI 278 specification revision (when using this switch, it may 279 be necessary to carry out a cold reboot _twice_ in a 280 row to make it take effect on the platform firmware). 281 282 acpi_osi= [HW,ACPI] Modify list of supported OS interface strings 283 acpi_osi="string1" # add string1 284 acpi_osi="!string2" # remove string2 285 acpi_osi=!* # remove all strings 286 acpi_osi=! # disable all built-in OS vendor 287 strings 288 acpi_osi=!! # enable all built-in OS vendor 289 strings 290 acpi_osi= # disable all strings 291 292 'acpi_osi=!' can be used in combination with single or 293 multiple 'acpi_osi="string1"' to support specific OS 294 vendor string(s). Note that such command can only 295 affect the default state of the OS vendor strings, thus 296 it cannot affect the default state of the feature group 297 strings and the current state of the OS vendor strings, 298 specifying it multiple times through kernel command line 299 is meaningless. This command is useful when one do not 300 care about the state of the feature group strings which 301 should be controlled by the OSPM. 302 Examples: 303 1. 'acpi_osi=! acpi_osi="Windows 2000"' is equivalent 304 to 'acpi_osi="Windows 2000" acpi_osi=!', they all 305 can make '_OSI("Windows 2000")' TRUE. 306 307 'acpi_osi=' cannot be used in combination with other 308 'acpi_osi=' command lines, the _OSI method will not 309 exist in the ACPI namespace. NOTE that such command can 310 only affect the _OSI support state, thus specifying it 311 multiple times through kernel command line is also 312 meaningless. 313 Examples: 314 1. 'acpi_osi=' can make 'CondRefOf(_OSI, Local1)' 315 FALSE. 316 317 'acpi_osi=!*' can be used in combination with single or 318 multiple 'acpi_osi="string1"' to support specific 319 string(s). Note that such command can affect the 320 current state of both the OS vendor strings and the 321 feature group strings, thus specifying it multiple times 322 through kernel command line is meaningful. But it may 323 still not able to affect the final state of a string if 324 there are quirks related to this string. This command 325 is useful when one want to control the state of the 326 feature group strings to debug BIOS issues related to 327 the OSPM features. 328 Examples: 329 1. 'acpi_osi="Module Device" acpi_osi=!*' can make 330 '_OSI("Module Device")' FALSE. 331 2. 'acpi_osi=!* acpi_osi="Module Device"' can make 332 '_OSI("Module Device")' TRUE. 333 3. 'acpi_osi=! acpi_osi=!* acpi_osi="Windows 2000"' is 334 equivalent to 335 'acpi_osi=!* acpi_osi=! acpi_osi="Windows 2000"' 336 and 337 'acpi_osi=!* acpi_osi="Windows 2000" acpi_osi=!', 338 they all will make '_OSI("Windows 2000")' TRUE. 339 340 acpi_pm_good [X86] 341 Override the pmtimer bug detection: force the kernel 342 to assume that this machine's pmtimer latches its value 343 and always returns good values. 344 345 acpi_sci= [HW,ACPI,EARLY] ACPI System Control Interrupt trigger mode 346 Format: { level | edge | high | low } 347 348 acpi_skip_timer_override [HW,ACPI,EARLY] 349 Recognize and ignore IRQ0/pin2 Interrupt Override. 350 For broken nForce2 BIOS resulting in XT-PIC timer. 351 352 acpi_sleep= [HW,ACPI] Sleep options 353 Format: { s3_bios, s3_mode, s3_beep, s4_hwsig, 354 s4_nohwsig, old_ordering, nonvs, 355 sci_force_enable, nobl } 356 See Documentation/power/video.rst for information on 357 s3_bios and s3_mode. 358 s3_beep is for debugging; it makes the PC's speaker beep 359 as soon as the kernel's real-mode entry point is called. 360 s4_hwsig causes the kernel to check the ACPI hardware 361 signature during resume from hibernation, and gracefully 362 refuse to resume if it has changed. This complies with 363 the ACPI specification but not with reality, since 364 Windows does not do this and many laptops do change it 365 on docking. So the default behaviour is to allow resume 366 and simply warn when the signature changes, unless the 367 s4_hwsig option is enabled. 368 s4_nohwsig prevents ACPI hardware signature from being 369 used (or even warned about) during resume. 370 old_ordering causes the ACPI 1.0 ordering of the _PTS 371 control method, with respect to putting devices into 372 low power states, to be enforced (the ACPI 2.0 ordering 373 of _PTS is used by default). 374 nonvs prevents the kernel from saving/restoring the 375 ACPI NVS memory during suspend/hibernation and resume. 376 sci_force_enable causes the kernel to set SCI_EN directly 377 on resume from S1/S3 (which is against the ACPI spec, 378 but some broken systems don't work without it). 379 nobl causes the internal blacklist of systems known to 380 behave incorrectly in some ways with respect to system 381 suspend and resume to be ignored (use wisely). 382 383 acpi_use_timer_override [HW,ACPI,EARLY] 384 Use timer override. For some broken Nvidia NF5 boards 385 that require a timer override, but don't have HPET 386 387 add_efi_memmap [EFI,X86,EARLY] Include EFI memory map in 388 kernel's map of available physical RAM. 389 390 agp= [AGP] 391 { off | try_unsupported } 392 off: disable AGP support 393 try_unsupported: try to drive unsupported chipsets 394 (may crash computer or cause data corruption) 395 396 ALSA [HW,ALSA] 397 See Documentation/sound/alsa-configuration.rst 398 399 alignment= [KNL,ARM] 400 Allow the default userspace alignment fault handler 401 behaviour to be specified. Bit 0 enables warnings, 402 bit 1 enables fixups, and bit 2 sends a segfault. 403 404 align_va_addr= [X86-64] 405 Align virtual addresses by clearing slice [14:12] when 406 allocating a VMA at process creation time. This option 407 gives you up to 3% performance improvement on AMD F15h 408 machines (where it is enabled by default) for a 409 CPU-intensive style benchmark, and it can vary highly in 410 a microbenchmark depending on workload and compiler. 411 412 32: only for 32-bit processes 413 64: only for 64-bit processes 414 on: enable for both 32- and 64-bit processes 415 off: disable for both 32- and 64-bit processes 416 417 alloc_snapshot [FTRACE] 418 Allocate the ftrace snapshot buffer on boot up when the 419 main buffer is allocated. This is handy if debugging 420 and you need to use tracing_snapshot() on boot up, and 421 do not want to use tracing_snapshot_alloc() as it needs 422 to be done where GFP_KERNEL allocations are allowed. 423 424 allow_mismatched_32bit_el0 [ARM64,EARLY] 425 Allow execve() of 32-bit applications and setting of the 426 PER_LINUX32 personality on systems where only a strict 427 subset of the CPUs support 32-bit EL0. When this 428 parameter is present, the set of CPUs supporting 32-bit 429 EL0 is indicated by /sys/devices/system/cpu/aarch32_el0 430 and hot-unplug operations may be restricted. 431 432 See Documentation/arch/arm64/asymmetric-32bit.rst for more 433 information. 434 435 amd_iommu= [HW,X86-64] 436 Pass parameters to the AMD IOMMU driver in the system. 437 Possible values are: 438 fullflush - Deprecated, equivalent to iommu.strict=1 439 off - do not initialize any AMD IOMMU found in 440 the system 441 force_isolation - Force device isolation for all 442 devices. The IOMMU driver is not 443 allowed anymore to lift isolation 444 requirements as needed. This option 445 does not override iommu=pt 446 force_enable - Force enable the IOMMU on platforms known 447 to be buggy with IOMMU enabled. Use this 448 option with care. 449 pgtbl_v1 - Use v1 page table for DMA-API (Default). 450 pgtbl_v2 - Use v2 page table for DMA-API. 451 irtcachedis - Disable Interrupt Remapping Table (IRT) caching. 452 nohugepages - Limit page-sizes used for v1 page-tables 453 to 4 KiB. 454 v2_pgsizes_only - Limit page-sizes used for v1 page-tables 455 to 4KiB/2Mib/1GiB. 456 457 458 amd_iommu_dump= [HW,X86-64] 459 Enable AMD IOMMU driver option to dump the ACPI table 460 for AMD IOMMU. With this option enabled, AMD IOMMU 461 driver will print ACPI tables for AMD IOMMU during 462 IOMMU initialization. 463 464 amd_iommu_intr= [HW,X86-64] 465 Specifies one of the following AMD IOMMU interrupt 466 remapping modes: 467 legacy - Use legacy interrupt remapping mode. 468 vapic - Use virtual APIC mode, which allows IOMMU 469 to inject interrupts directly into guest. 470 This mode requires kvm-amd.avic=1. 471 (Default when IOMMU HW support is present.) 472 473 amd_pstate= [X86,EARLY] 474 disable 475 Do not enable amd_pstate as the default 476 scaling driver for the supported processors 477 passive 478 Use amd_pstate with passive mode as a scaling driver. 479 In this mode autonomous selection is disabled. 480 Driver requests a desired performance level and platform 481 tries to match the same performance level if it is 482 satisfied by guaranteed performance level. 483 active 484 Use amd_pstate_epp driver instance as the scaling driver, 485 driver provides a hint to the hardware if software wants 486 to bias toward performance (0x0) or energy efficiency (0xff) 487 to the CPPC firmware. then CPPC power algorithm will 488 calculate the runtime workload and adjust the realtime cores 489 frequency. 490 guided 491 Activate guided autonomous mode. Driver requests minimum and 492 maximum performance level and the platform autonomously 493 selects a performance level in this range and appropriate 494 to the current workload. 495 496 amd_prefcore= 497 [X86] 498 disable 499 Disable amd-pstate preferred core. 500 501 amd_dynamic_epp= 502 [X86] 503 disable 504 Disable amd-pstate dynamic EPP. 505 enable 506 Enable amd-pstate dynamic EPP. 507 508 amijoy.map= [HW,JOY] Amiga joystick support 509 Map of devices attached to JOY0DAT and JOY1DAT 510 Format: <a>,<b> 511 See also Documentation/input/joydev/joystick.rst 512 513 analog.map= [HW,JOY] Analog joystick and gamepad support 514 Specifies type or capabilities of an analog joystick 515 connected to one of 16 gameports 516 Format: <type1>,<type2>,..<type16> 517 518 apc= [HW,SPARC] 519 Power management functions (SPARCstation-4/5 + deriv.) 520 Format: noidle 521 Disable APC CPU standby support. SPARCstation-Fox does 522 not play well with APC CPU idle - disable it if you have 523 APC and your system crashes randomly. 524 525 apic [APIC,X86-64] Use IO-APIC. Default. 526 527 apic= [APIC,X86,EARLY] Advanced Programmable Interrupt Controller 528 Change the output verbosity while booting 529 Format: { quiet (default) | verbose | debug } 530 Change the amount of debugging information output 531 when initialising the APIC and IO-APIC components. 532 533 apic_extnmi= [APIC,X86,EARLY] External NMI delivery setting 534 Format: { bsp (default) | all | none } 535 bsp: External NMI is delivered only to CPU 0 536 all: External NMIs are broadcast to all CPUs as a 537 backup of CPU 0 538 none: External NMI is masked for all CPUs. This is 539 useful so that a dump capture kernel won't be 540 shot down by NMI 541 542 apicpmtimer Do APIC timer calibration using the pmtimer. Implies 543 apicmaintimer. Useful when your PIT timer is totally 544 broken. 545 546 autoconf= [IPV6] 547 See Documentation/networking/ipv6.rst. 548 549 apm= [APM] Advanced Power Management 550 See header of arch/x86/kernel/apm_32.c. 551 552 apparmor= [APPARMOR] Disable or enable AppArmor at boot time 553 Format: { "0" | "1" } 554 See security/apparmor/Kconfig help text 555 0 -- disable. 556 1 -- enable. 557 Default value is set via kernel config option. 558 559 arm64.no32bit_el0 [ARM64] Unconditionally disable the execution of 560 32 bit applications. 561 562 arm64.nobti [ARM64] Unconditionally disable Branch Target 563 Identification support 564 565 arm64.nogcs [ARM64] Unconditionally disable Guarded Control Stack 566 support 567 568 arm64.nomops [ARM64] Unconditionally disable Memory Copy and Memory 569 Set instructions support 570 571 arm64.nompam [ARM64] Unconditionally disable Memory Partitioning And 572 Monitoring support 573 574 arm64.nomte [ARM64] Unconditionally disable Memory Tagging Extension 575 support 576 577 arm64.nopauth [ARM64] Unconditionally disable Pointer Authentication 578 support 579 580 arm64.nosme [ARM64] Unconditionally disable Scalable Matrix 581 Extension support 582 583 arm64.nosve [ARM64] Unconditionally disable Scalable Vector 584 Extension support 585 586 ataflop= [HW,M68k] 587 588 atarimouse= [HW,MOUSE] Atari Mouse 589 590 atkbd.extra= [HW] Enable extra LEDs and keys on IBM RapidAccess, 591 EzKey and similar keyboards 592 593 atkbd.reset= [HW] Reset keyboard during initialization 594 595 atkbd.set= [HW] Select keyboard code set 596 Format: <int> (2 = AT (default), 3 = PS/2) 597 598 atkbd.scroll= [HW] Enable scroll wheel on MS Office and similar 599 keyboards 600 601 atkbd.softraw= [HW] Choose between synthetic and real raw mode 602 Format: <bool> (0 = real, 1 = synthetic (default)) 603 604 atkbd.softrepeat= [HW] 605 Use software keyboard repeat 606 607 audit= [KNL] Enable the audit sub-system 608 Format: { "0" | "1" | "off" | "on" } 609 0 | off - kernel audit is disabled and can not be 610 enabled until the next reboot 611 unset - kernel audit is initialized but disabled and 612 will be fully enabled by the userspace auditd. 613 1 | on - kernel audit is initialized and partially 614 enabled, storing at most audit_backlog_limit 615 messages in RAM until it is fully enabled by the 616 userspace auditd. 617 Default: unset 618 619 audit_backlog_limit= [KNL] Set the audit queue size limit. 620 Format: <int> (must be >=0) 621 Default: 64 622 623 bau= [X86_UV] Enable the BAU on SGI UV. The default 624 behavior is to disable the BAU (i.e. bau=0). 625 Format: { "0" | "1" } 626 0 - Disable the BAU. 627 1 - Enable the BAU. 628 unset - Disable the BAU. 629 630 bdev_allow_write_mounted= 631 Format: <bool> 632 Control the ability to open a mounted block device 633 for writing, i.e., allow / disallow writes that bypass 634 the FS. This was implemented as a means to prevent 635 fuzzers from crashing the kernel by overwriting the 636 metadata underneath a mounted FS without its awareness. 637 This also prevents destructive formatting of mounted 638 filesystems by naive storage tooling that don't use 639 O_EXCL. Default is Y and can be changed through the 640 Kconfig option CONFIG_BLK_DEV_WRITE_MOUNTED. 641 642 bert_disable [ACPI] 643 Disable BERT OS support on buggy BIOSes. 644 645 bgrt_disable [ACPI,X86,EARLY] 646 Disable BGRT to avoid flickering OEM logo. 647 648 blkdevparts= Manual partition parsing of block device(s) for 649 embedded devices based on command line input. 650 See Documentation/block/cmdline-partition.rst 651 652 boot_delay= [KNL,EARLY] 653 Milliseconds to delay each printk during boot. 654 Only works if CONFIG_BOOT_PRINTK_DELAY is enabled, 655 and you may also have to specify "lpj=". Boot_delay 656 values larger than 10 seconds (10000) are assumed 657 erroneous and ignored. 658 Format: integer 659 660 bootconfig [KNL,EARLY] 661 Extended command line options can be added to an initrd 662 and this will cause the kernel to look for it. 663 664 See Documentation/admin-guide/bootconfig.rst 665 666 bttv.card= [HW,V4L] bttv (bt848 + bt878 based grabber cards) 667 bttv.radio= Most important insmod options are available as 668 kernel args too. 669 bttv.pll= See Documentation/admin-guide/media/bttv.rst 670 bttv.tuner= 671 672 bulk_remove=off [PPC] This parameter disables the use of the pSeries 673 firmware feature for flushing multiple hpte entries 674 at a time. 675 676 c101= [NET] Moxa C101 synchronous serial card 677 678 cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection. 679 Sometimes CPU hardware bugs make them report the cache 680 size incorrectly. The kernel will attempt work arounds 681 to fix known problems, but for some CPUs it is not 682 possible to determine what the correct size should be. 683 This option provides an override for these situations. 684 685 carrier_timeout= 686 [NET] Specifies amount of time (in seconds) that 687 the kernel should wait for a network carrier. By default 688 it waits 120 seconds. 689 690 ca_keys= [KEYS] This parameter identifies a specific key(s) on 691 the system trusted keyring to be used for certificate 692 trust validation. 693 format: { id:<keyid> | builtin } 694 695 cca= [MIPS,EARLY] Override the kernel pages' cache coherency 696 algorithm. Accepted values range from 0 to 7 697 inclusive. See arch/mips/include/asm/pgtable-bits.h 698 for platform specific values (SB1, Loongson3 and 699 others). 700 701 ccw_timeout_log [S390] 702 See Documentation/arch/s390/common_io.rst for details. 703 704 cfi= [X86-64] Set Control Flow Integrity checking features 705 when CONFIG_FINEIBT is enabled. 706 Format: feature[,feature...] 707 Default: auto 708 709 auto: Use FineIBT if IBT available, otherwise kCFI. 710 Under FineIBT, enable "paranoid" mode when 711 FRED is not available. 712 off: Turn off CFI checking. 713 kcfi: Use kCFI (disable FineIBT). 714 fineibt: Use FineIBT (even if IBT not available). 715 norand: Do not re-randomize CFI hashes. 716 paranoid: Add caller hash checking under FineIBT. 717 bhi: Enable register poisoning to stop speculation 718 across FineIBT. (Disabled by default.) 719 warn: Do not enforce CFI checking: warn only. 720 debug: Report CFI initialization details. 721 722 cgroup_disable= [KNL] Disable a particular controller or optional feature 723 Format: {name of the controller(s) or feature(s) to disable} 724 The effects of cgroup_disable=foo are: 725 - foo isn't auto-mounted if you mount all cgroups in 726 a single hierarchy 727 - foo isn't visible as an individually mountable 728 subsystem 729 - if foo is an optional feature then the feature is 730 disabled and corresponding cgroup files are not 731 created 732 {Currently only "memory" controller deal with this and 733 cut the overhead, others just disable the usage. So 734 only cgroup_disable=memory is actually worthy} 735 Specifying "pressure" disables per-cgroup pressure 736 stall information accounting feature 737 738 cgroup_no_v1= [KNL] Disable cgroup controllers and named hierarchies in v1 739 Format: { { controller | "all" | "named" } 740 [,{ controller | "all" | "named" }...] } 741 Like cgroup_disable, but only applies to cgroup v1; 742 the blacklisted controllers remain available in cgroup2. 743 "all" blacklists all controllers and "named" disables 744 named mounts. Specifying both "all" and "named" disables 745 all v1 hierarchies. 746 747 cgroup_v1_proc= [KNL] Show also missing controllers in /proc/cgroups 748 Format: { "true" | "false" } 749 /proc/cgroups lists only v1 controllers by default. 750 This compatibility option enables listing also v2 751 controllers (whose v1 code is not compiled!), so that 752 semi-legacy software can check this file to decide 753 about usage of v2 (sic) controllers. 754 755 cgroup_favordynmods= [KNL] Enable or Disable favordynmods. 756 Format: { "true" | "false" } 757 Defaults to the value of CONFIG_CGROUP_FAVOR_DYNMODS. 758 759 cgroup.memory= [KNL] Pass options to the cgroup memory controller. 760 Format: <string> 761 nosocket -- Disable socket memory accounting. 762 nokmem -- Disable kernel memory accounting. 763 nobpf -- Disable BPF memory accounting. 764 765 check_pages= [MM,EARLY] Enable sanity checking of pages after 766 allocations / before freeing. This adds checks to catch 767 double-frees, use-after-frees, and other sources of 768 page corruption by inspecting page internals (flags, 769 mapcount/refcount, memcg_data, etc.). 770 Format: { "0" | "1" } 771 Default: 0 (1 if CONFIG_DEBUG_VM is set) 772 773 checkreqprot= [SELINUX] Set initial checkreqprot flag value. 774 Format: { "0" | "1" } 775 See security/selinux/Kconfig help text. 776 0 -- check protection applied by kernel (includes 777 any implied execute protection). 778 1 -- check protection requested by application. 779 Default value is set via a kernel config option. 780 Value can be changed at runtime via 781 /sys/fs/selinux/checkreqprot. 782 Setting checkreqprot to 1 is deprecated. 783 784 cio_ignore= [S390] 785 See Documentation/arch/s390/common_io.rst for details. 786 787 clk_ignore_unused 788 [CLK] 789 Prevents the clock framework from automatically gating 790 clocks that have not been explicitly enabled by a Linux 791 device driver but are enabled in hardware at reset or 792 by the bootloader/firmware. Note that this does not 793 force such clocks to be always-on nor does it reserve 794 those clocks in any way. This parameter is useful for 795 debug and development, but should not be needed on a 796 platform with proper driver support. For more 797 information, see Documentation/driver-api/clk.rst. 798 799 clock= [BUGS=X86-32, HW] gettimeofday clocksource override. 800 [Deprecated] 801 Forces specified clocksource (if available) to be used 802 when calculating gettimeofday(). If specified 803 clocksource is not available, it defaults to PIT. 804 Format: { pit | tsc | cyclone | pmtmr } 805 806 clocksource= Override the default clocksource 807 Format: <string> 808 Override the default clocksource and use the clocksource 809 with the name specified. 810 Some clocksource names to choose from, depending on 811 the platform: 812 [all] jiffies (this is the base, fallback clocksource) 813 [ACPI] acpi_pm 814 [ARM] imx_timer1,OSTS,netx_timer,mpu_timer2, 815 pxa_timer,timer3,32k_counter,timer0_1 816 [X86-32] pit,hpet,tsc; 817 scx200_hrt on Geode; cyclone on IBM x440 818 [MIPS] MIPS 819 [PARISC] cr16 820 [S390] tod 821 [SH] SuperH 822 [SPARC64] tick 823 [X86-64] hpet,tsc 824 825 clocksource.arm_arch_timer.evtstrm= 826 [ARM,ARM64,EARLY] 827 Format: <bool> 828 Enable/disable the eventstream feature of the ARM 829 architected timer so that code using WFE-based polling 830 loops can be debugged more effectively on production 831 systems. 832 833 clocksource.verify_n_cpus= [KNL] 834 Limit the number of CPUs checked for clocksources 835 marked with CLOCK_SOURCE_VERIFY_PERCPU that 836 are marked unstable due to excessive skew. 837 A negative value says to check all CPUs, while 838 zero says not to check any. Values larger than 839 nr_cpu_ids are silently truncated to nr_cpu_ids. 840 The actual CPUs are chosen randomly, with 841 no replacement if the same CPU is chosen twice. 842 843 clocksource-wdtest.holdoff= [KNL] 844 Set the time in seconds that the clocksource 845 watchdog test waits before commencing its tests. 846 Defaults to zero when built as a module and to 847 10 seconds when built into the kernel. 848 849 cma=nn[MG]@[start[MG][-end[MG]]] 850 [KNL,CMA,EARLY] 851 Sets the size of kernel global memory area for 852 contiguous memory allocations and optionally the 853 placement constraint by the physical address range of 854 memory allocations. A value of 0 disables CMA 855 altogether. For more information, see 856 kernel/dma/contiguous.c 857 858 cma_pernuma=nn[MG] 859 [KNL,CMA,EARLY] 860 Sets the size of kernel per-numa memory area for 861 contiguous memory allocations. A value of 0 disables 862 per-numa CMA altogether. And If this option is not 863 specified, the default value is 0. 864 With per-numa CMA enabled, DMA users on node nid will 865 first try to allocate buffer from the pernuma area 866 which is located in node nid, if the allocation fails, 867 they will fallback to the global default memory area. 868 869 numa_cma=<node>:nn[MG][,<node>:nn[MG]] 870 [KNL,CMA,EARLY] 871 Sets the size of kernel numa memory area for 872 contiguous memory allocations. It will reserve CMA 873 area for the specified node. 874 875 If it is setup together with upper 'cmd_pernuma=' 876 (unlikely), its size setting takes priority for the 877 specified numa nodes. 878 879 With numa CMA enabled, DMA users on node nid will 880 first try to allocate buffer from the numa area 881 which is located in node nid, if the allocation fails, 882 they will fallback to the global default memory area. 883 884 cmo_free_hint= [PPC] Format: { yes | no } 885 Specify whether pages are marked as being inactive 886 when they are freed. This is used in CMO environments 887 to determine OS memory pressure for page stealing by 888 a hypervisor. 889 Default: yes 890 891 coherent_pool=nn[KMG] [ARM,KNL,EARLY] 892 Sets the size of memory pool for coherent, atomic dma 893 allocations, by default set to 256K. 894 895 condev= [HW,S390] console device 896 conmode= 897 898 con3215_drop= [S390,EARLY] 3215 console drop mode. 899 Format: y|n|Y|N|1|0 900 When set to true, drop data on the 3215 console when 901 the console buffer is full. In this case the 902 operator using a 3270 terminal emulator (for example 903 x3270) does not have to enter the clear key for the 904 console output to advance and the kernel to continue. 905 This leads to a much faster boot time when a 3270 906 terminal emulator is active. If no 3270 terminal 907 emulator is used, this parameter has no effect. 908 909 console= [KNL] Output console device and options. 910 911 tty<n> Use the virtual console device <n>. 912 913 ttyS<n>[,options] 914 ttyUSB0[,options] 915 Use the specified serial port. The options are of 916 the form "bbbbpnf", where "bbbb" is the baud rate, 917 "p" is parity ("n", "o", or "e"), "n" is number of 918 bits, and "f" is flow control ("r" for RTS or 919 omit it). Default is "9600n8". 920 921 See Documentation/admin-guide/serial-console.rst for more 922 information. See 923 Documentation/networking/netconsole.rst for an 924 alternative. 925 926 <DEVNAME>:<n>.<n>[,options] 927 Use the specified serial port on the serial core bus. 928 The addressing uses DEVNAME of the physical serial port 929 device, followed by the serial core controller instance, 930 and the serial port instance. The options are the same 931 as documented for the ttyS addressing above. 932 933 The mapping of the serial ports to the tty instances 934 can be viewed with: 935 936 $ ls -d /sys/bus/serial-base/devices/*:*.*/tty/* 937 /sys/bus/serial-base/devices/00:04:0.0/tty/ttyS0 938 939 In the above example, the console can be addressed with 940 console=00:04:0.0. Note that a console addressed this 941 way will only get added when the related device driver 942 is ready. The use of an earlycon parameter in addition to 943 the console may be desired for console output early on. 944 945 uart[8250],io,<addr>[,options] 946 uart[8250],mmio,<addr>[,options] 947 uart[8250],mmio16,<addr>[,options] 948 uart[8250],mmio32,<addr>[,options] 949 uart[8250],0x<addr>[,options] 950 Start an early, polled-mode console on the 8250/16550 951 UART at the specified I/O port or MMIO address, 952 switching to the matching ttyS device later. 953 MMIO inter-register address stride is either 8-bit 954 (mmio), 16-bit (mmio16), or 32-bit (mmio32). 955 If none of [io|mmio|mmio16|mmio32], <addr> is assumed 956 to be equivalent to 'mmio'. 'options' are specified in 957 the same format described for ttyS above; if unspecified, 958 the h/w is not re-initialized. 959 960 hvc<n> Use the hypervisor console device <n>. This is for 961 both Xen and PowerPC hypervisors. 962 963 { null | "" } 964 Use to disable console output, i.e., to have kernel 965 console messages discarded. 966 This must be the only console= parameter used on the 967 kernel command line. 968 969 If the device connected to the port is not a TTY but a braille 970 device, prepend "brl," before the device type, for instance 971 console=brl,ttyS0 972 For now, only VisioBraille is supported. 973 974 console_msg_format= 975 [KNL] Change console messages format 976 default 977 By default we print messages on consoles in 978 "[time stamp] text\n" format (time stamp may not be 979 printed, depending on CONFIG_PRINTK_TIME or 980 `printk_time' param). 981 syslog 982 Switch to syslog format: "<%u>[time stamp] text\n" 983 IOW, each message will have a facility and loglevel 984 prefix. The format is similar to one used by syslog() 985 syscall, or to executing "dmesg -S --raw" or to reading 986 from /proc/kmsg. 987 988 consoleblank= [KNL] The console blank (screen saver) timeout in 989 seconds. A value of 0 disables the blank timer. 990 Defaults to 0. 991 992 coredump_filter= 993 [KNL] Change the default value for 994 /proc/<pid>/coredump_filter. 995 See also Documentation/filesystems/proc.rst. 996 997 coresight_cpu_debug.enable 998 [ARM,ARM64] 999 Format: <bool> 1000 Enable/disable the CPU sampling based debugging. 1001 0: default value, disable debugging 1002 1: enable debugging at boot time 1003 1004 cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver 1005 Format: 1006 <first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>] 1007 1008 cpuidle.off=1 [CPU_IDLE] 1009 disable the cpuidle sub-system 1010 1011 cpuidle.governor= 1012 [CPU_IDLE] Name of the cpuidle governor to use. 1013 1014 cpufreq.off=1 [CPU_FREQ] 1015 disable the cpufreq sub-system 1016 1017 cpufreq.default_governor= 1018 [CPU_FREQ] Name of the default cpufreq governor or 1019 policy to use. This governor must be registered in the 1020 kernel before the cpufreq driver probes. 1021 1022 cpu_init_udelay=N 1023 [X86,EARLY] Delay for N microsec between assert and de-assert 1024 of APIC INIT to start processors. This delay occurs 1025 on every CPU online, such as boot, and resume from suspend. 1026 Default: 10000 1027 1028 cpuhp.parallel= 1029 [SMP] Enable/disable parallel bringup of secondary CPUs 1030 Format: <bool> 1031 Default is enabled if CONFIG_HOTPLUG_PARALLEL=y. Otherwise 1032 the parameter has no effect. 1033 1034 crash_kexec_post_notifiers 1035 Only jump to kdump kernel after running the panic 1036 notifiers and dumping kmsg. This option increases 1037 the risks of a kdump failure, since some panic 1038 notifiers can make the crashed kernel more unstable. 1039 In configurations where kdump may not be reliable, 1040 running the panic notifiers could allow collecting 1041 more data on dmesg, like stack traces from other CPUS 1042 or extra data dumped by panic_print. Note that some 1043 configurations enable this option unconditionally, 1044 like Hyper-V, PowerPC (fadump) and AMD SEV-SNP. 1045 1046 crashkernel=size[KMG][@offset[KMG]] 1047 [KNL,EARLY] Using kexec, Linux can switch to a 'crash kernel' 1048 upon panic. This parameter reserves the physical 1049 memory region [offset, offset + size] for that kernel 1050 image. If '@offset' is omitted, then a suitable offset 1051 is selected automatically. 1052 [KNL, X86-64, ARM64, RISCV, LoongArch] Select a region 1053 under 4G first, and fall back to reserve region above 1054 4G when '@offset' hasn't been specified. 1055 See Documentation/admin-guide/kdump/kdump.rst for further details. 1056 1057 crashkernel=range1:size1[,range2:size2,...][@offset] 1058 [KNL] Same as above, but depends on the memory 1059 in the running system. The syntax of range is 1060 start-[end] where start and end are both 1061 a memory unit (amount[KMG]). See also 1062 Documentation/admin-guide/kdump/kdump.rst for an example. 1063 1064 crashkernel=size[KMG],high 1065 [KNL, X86-64, ARM64, RISCV, LoongArch] range could be 1066 above 4G. 1067 Allow kernel to allocate physical memory region from top, 1068 so could be above 4G if system have more than 4G ram 1069 installed. Otherwise memory region will be allocated 1070 below 4G, if available. 1071 It will be ignored if crashkernel=X is specified. 1072 crashkernel=size[KMG],low 1073 [KNL, X86-64, ARM64, RISCV, LoongArch] range under 4G. 1074 When crashkernel=X,high is passed, kernel could allocate 1075 physical memory region above 4G, that cause second kernel 1076 crash on system that require some amount of low memory, 1077 e.g. swiotlb requires at least 64M+32K low memory, also 1078 enough extra low memory is needed to make sure DMA buffers 1079 for 32-bit devices won't run out. Kernel would try to allocate 1080 default size of memory below 4G automatically. The default 1081 size is platform dependent. 1082 --> x86: max(swiotlb_size_or_default() + 8MiB, 256MiB) 1083 --> arm64: 128MiB 1084 --> riscv: 128MiB 1085 --> loongarch: 128MiB 1086 This one lets the user specify own low range under 4G 1087 for second kernel instead. 1088 0: to disable low allocation. 1089 It will be ignored when crashkernel=X,high is not used 1090 or memory reserved is below 4G. 1091 crashkernel=size[KMG],cma 1092 [KNL, X86, ppc] Reserve additional crash kernel memory from 1093 CMA. This reservation is usable by the first system's 1094 userspace memory and kernel movable allocations (memory 1095 balloon, zswap). Pages allocated from this memory range 1096 will not be included in the vmcore so this should not 1097 be used if dumping of userspace memory is intended and 1098 it has to be expected that some movable kernel pages 1099 may be missing from the dump. 1100 1101 A standard crashkernel reservation, as described above, 1102 is still needed to hold the crash kernel and initrd. 1103 1104 This option increases the risk of a kdump failure: DMA 1105 transfers configured by the first kernel may end up 1106 corrupting the second kernel's memory. 1107 1108 This reservation method is intended for systems that 1109 can't afford to sacrifice enough memory for standard 1110 crashkernel reservation and where less reliable and 1111 possibly incomplete kdump is preferable to no kdump at 1112 all. 1113 1114 cryptomgr.notests 1115 [KNL] Disable crypto self-tests 1116 1117 cs89x0_dma= [HW,NET] 1118 Format: <dma> 1119 1120 cs89x0_media= [HW,NET] 1121 Format: { rj45 | aui | bnc } 1122 1123 csdlock_debug= [KNL] Enable or disable debug add-ons of cross-CPU 1124 function call handling. When switched on, 1125 additional debug data is printed to the console 1126 in case a hanging CPU is detected, and that 1127 CPU is pinged again in order to try to resolve 1128 the hang situation. The default value of this 1129 option depends on the CSD_LOCK_WAIT_DEBUG_DEFAULT 1130 Kconfig option. 1131 1132 dasd= [HW,NET] 1133 See header of drivers/s390/block/dasd_devmap.c. 1134 1135 db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port 1136 (one device per port) 1137 Format: <port#>,<type> 1138 See also Documentation/input/devices/joystick-parport.rst 1139 1140 debug [KNL,EARLY] Enable kernel debugging (events log level). 1141 1142 debug_boot_weak_hash 1143 [KNL,EARLY] Enable printing [hashed] pointers early in the 1144 boot sequence. If enabled, we use a weak hash instead 1145 of siphash to hash pointers. Use this option if you are 1146 seeing instances of '(___ptrval___)') and need to see a 1147 value (hashed pointer) instead. Cryptographically 1148 insecure, please do not use on production kernels. 1149 1150 debug_locks_verbose= 1151 [KNL] verbose locking self-tests 1152 Format: <int> 1153 Print debugging info while doing the locking API 1154 self-tests. 1155 Bitmask for the various LOCKTYPE_ tests. Defaults to 0 1156 (no extra messages), setting it to -1 (all bits set) 1157 will print _a_lot_ more information - normally only 1158 useful to lockdep developers. 1159 1160 debug_objects [KNL,EARLY] Enable object debugging 1161 1162 debug_guardpage_minorder= 1163 [KNL,EARLY] When CONFIG_DEBUG_PAGEALLOC is set, this 1164 parameter allows control of the order of pages that will 1165 be intentionally kept free (and hence protected) by the 1166 buddy allocator. Bigger value increase the probability 1167 of catching random memory corruption, but reduce the 1168 amount of memory for normal system use. The maximum 1169 possible value is MAX_PAGE_ORDER/2. Setting this 1170 parameter to 1 or 2 should be enough to identify most 1171 random memory corruption problems caused by bugs in 1172 kernel or driver code when a CPU writes to (or reads 1173 from) a random memory location. Note that there exists 1174 a class of memory corruptions problems caused by buggy 1175 H/W or F/W or by drivers badly programming DMA 1176 (basically when memory is written at bus level and the 1177 CPU MMU is bypassed) which are not detectable by 1178 CONFIG_DEBUG_PAGEALLOC, hence this option will not 1179 help tracking down these problems. 1180 1181 debug_pagealloc= 1182 [KNL,EARLY] When CONFIG_DEBUG_PAGEALLOC is set, this parameter 1183 enables the feature at boot time. By default, it is 1184 disabled and the system will work mostly the same as a 1185 kernel built without CONFIG_DEBUG_PAGEALLOC. 1186 Note: to get most of debug_pagealloc error reports, it's 1187 useful to also enable the page_owner functionality. 1188 on: enable the feature 1189 1190 debugfs= [KNL,EARLY] This parameter enables what is exposed to 1191 userspace and debugfs internal clients. 1192 Format: { on, off } 1193 on: All functions are enabled. 1194 off: Filesystem is not registered and clients 1195 get a -EPERM as result when trying to register files 1196 or directories within debugfs. 1197 This is equivalent of the runtime functionality if 1198 debugfs was not enabled in the kernel at all. 1199 Default value is set in build-time with a kernel configuration. 1200 1201 debugpat [X86] Enable PAT debugging 1202 1203 default_hugepagesz= 1204 [HW] The size of the default HugeTLB page. This is 1205 the size represented by the legacy /proc/ hugepages 1206 APIs. In addition, this is the default hugetlb size 1207 used for shmget(), mmap() and mounting hugetlbfs 1208 filesystems. If not specified, defaults to the 1209 architecture's default huge page size. Huge page 1210 sizes are architecture dependent. See also 1211 Documentation/admin-guide/mm/hugetlbpage.rst. 1212 Format: size[KMG] 1213 1214 deferred_probe_timeout= 1215 [KNL] Debugging option to set a timeout in seconds for 1216 deferred probe to give up waiting on dependencies to 1217 probe. Only specific dependencies (subsystems or 1218 drivers) that have opted in will be ignored. A timeout 1219 of 0 will timeout at the end of initcalls; a negative 1220 value is treated as an infinite timeout value. If the 1221 timeout hasn't expired, it'll be restarted by each 1222 successful driver registration. This option will also 1223 dump out devices still on the deferred probe list after 1224 retrying. 1225 1226 delayacct [KNL] Enable per-task delay accounting 1227 1228 dell_smm_hwmon.ignore_dmi= 1229 [HW] Continue probing hardware even if DMI data 1230 indicates that the driver is running on unsupported 1231 hardware. 1232 1233 dell_smm_hwmon.force= 1234 [HW] Activate driver even if SMM BIOS signature does 1235 not match list of supported models and enable otherwise 1236 blacklisted features. 1237 1238 dell_smm_hwmon.power_status= 1239 [HW] Report power status in /proc/i8k 1240 (disabled by default). 1241 1242 dell_smm_hwmon.restricted= 1243 [HW] Allow controlling fans only if SYS_ADMIN 1244 capability is set. 1245 1246 dell_smm_hwmon.fan_mult= 1247 [HW] Factor to multiply fan speed with. 1248 1249 dell_smm_hwmon.fan_max= 1250 [HW] Maximum configurable fan speed. 1251 1252 dfltcc= [HW,S390] 1253 Format: { on | off | def_only | inf_only | always } 1254 on: s390 zlib hardware support for compression on 1255 level 1 and decompression (default) 1256 off: No s390 zlib hardware support 1257 def_only: s390 zlib hardware support for deflate 1258 only (compression on level 1) 1259 inf_only: s390 zlib hardware support for inflate 1260 only (decompression) 1261 always: Same as 'on' but ignores the selected compression 1262 level always using hardware support (used for debugging) 1263 1264 dhash_entries= [KNL] 1265 Set number of hash buckets for dentry cache. 1266 1267 disable_1tb_segments [PPC,EARLY] 1268 Disables the use of 1TB hash page table segments. This 1269 causes the kernel to fall back to 256MB segments which 1270 can be useful when debugging issues that require an SLB 1271 miss to occur. 1272 1273 disable= [IPV6] 1274 See Documentation/networking/ipv6.rst. 1275 1276 disable_radix [PPC,EARLY] 1277 Disable RADIX MMU mode on POWER9 1278 1279 disable_tlbie [PPC] 1280 Disable TLBIE instruction. Currently does not work 1281 with KVM, with HASH MMU, or with coherent accelerators. 1282 1283 disable_ddw [PPC/PSERIES,EARLY] 1284 Disable Dynamic DMA Window support. Use this 1285 to workaround buggy firmware. 1286 1287 disable_ipv6= [IPV6] 1288 See Documentation/networking/ipv6.rst. 1289 1290 disable_mtrr_cleanup [X86,EARLY] 1291 The kernel tries to adjust MTRR layout from continuous 1292 to discrete, to make X server driver able to add WB 1293 entry later. This parameter disables that. 1294 1295 disable_mtrr_trim [X86, Intel and AMD only,EARLY] 1296 By default the kernel will trim any uncacheable 1297 memory out of your available memory pool based on 1298 MTRR settings. This parameter disables that behavior, 1299 possibly causing your machine to run very slowly. 1300 1301 disable_timer_pin_1 [X86,EARLY] 1302 Disable PIN 1 of APIC timer 1303 Can be useful to work around chipset bugs. 1304 1305 dis_ucode_ldr [X86] Disable the microcode loader. 1306 1307 dma_debug=off If the kernel is compiled with DMA_API_DEBUG support, 1308 this option disables the debugging code at boot. 1309 1310 dma_debug_entries=<number> 1311 This option allows to tune the number of preallocated 1312 entries for DMA-API debugging code. One entry is 1313 required per DMA-API allocation. Use this if the 1314 DMA-API debugging code disables itself because the 1315 architectural default is too low. 1316 1317 dma_debug_driver=<driver_name> 1318 With this option the DMA-API debugging driver 1319 filter feature can be enabled at boot time. Just 1320 pass the driver to filter for as the parameter. 1321 The filter can be disabled or changed to another 1322 driver later using sysfs. 1323 1324 reg_file_data_sampling= 1325 [X86] Controls mitigation for Register File Data 1326 Sampling (RFDS) vulnerability. RFDS is a CPU 1327 vulnerability which may allow userspace to infer 1328 kernel data values previously stored in floating point 1329 registers, vector registers, or integer registers. 1330 RFDS only affects Intel Atom processors. 1331 1332 on: Turns ON the mitigation. 1333 off: Turns OFF the mitigation. 1334 1335 This parameter overrides the compile time default set 1336 by CONFIG_MITIGATION_RFDS. Mitigation cannot be 1337 disabled when other VERW based mitigations (like MDS) 1338 are enabled. In order to disable RFDS mitigation all 1339 VERW based mitigations need to be disabled. 1340 1341 For details see: 1342 Documentation/admin-guide/hw-vuln/reg-file-data-sampling.rst 1343 1344 dm_verity.keyring_unsealed= 1345 [KNL] When set to 1, leave the dm-verity keyring 1346 unsealed after initialization so userspace can 1347 provision keys. Once the keyring is restricted 1348 it becomes active and is searched during signature 1349 verification. 1350 1351 driver_async_probe= [KNL] 1352 List of driver names to be probed asynchronously. * 1353 matches with all driver names. If * is specified, the 1354 rest of the listed driver names are those that will NOT 1355 match the *. 1356 Format: <driver_name1>,<driver_name2>... 1357 1358 drm.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>] 1359 Broken monitors, graphic adapters, KVMs and EDIDless 1360 panels may send no or incorrect EDID data sets. 1361 This parameter allows to specify an EDID data sets 1362 in the /lib/firmware directory that are used instead. 1363 An EDID data set will only be used for a particular 1364 connector, if its name and a colon are prepended to 1365 the EDID name. Each connector may use a unique EDID 1366 data set by separating the files with a comma. An EDID 1367 data set with no connector name will be used for 1368 any connectors not explicitly specified. 1369 1370 dscc4.setup= [NET] 1371 1372 dt_cpu_ftrs= [PPC,EARLY] 1373 Format: {"off" | "known"} 1374 Control how the dt_cpu_ftrs device-tree binding is 1375 used for CPU feature discovery and setup (if it 1376 exists). 1377 off: Do not use it, fall back to legacy cpu table. 1378 known: Do not pass through unknown features to guests 1379 or userspace, only those that the kernel is aware of. 1380 1381 dump_apple_properties [X86] 1382 Dump name and content of EFI device properties on 1383 x86 Macs. Useful for driver authors to determine 1384 what data is available or for reverse-engineering. 1385 1386 dyndbg[="val"] [KNL,DYNAMIC_DEBUG] 1387 <module>.dyndbg[="val"] 1388 Enable debug messages at boot time. See 1389 Documentation/admin-guide/dynamic-debug-howto.rst 1390 for details. 1391 1392 early_ioremap_debug [KNL,EARLY] 1393 Enable debug messages in early_ioremap support. This 1394 is useful for tracking down temporary early mappings 1395 which are not unmapped. 1396 1397 earlycon= [KNL,EARLY] Output early console device and options. 1398 1399 When used with no options, the early console is 1400 determined by stdout-path property in device tree's 1401 chosen node or the ACPI SPCR table if supported by 1402 the platform. 1403 1404 cdns,<addr>[,options] 1405 Start an early, polled-mode console on a Cadence 1406 (xuartps) serial port at the specified address. Only 1407 supported option is baud rate. If baud rate is not 1408 specified, the serial port must already be setup and 1409 configured. 1410 1411 uart[8250],io,<addr>[,options[,uartclk]] 1412 uart[8250],mmio,<addr>[,options[,uartclk]] 1413 uart[8250],mmio32,<addr>[,options[,uartclk]] 1414 uart[8250],mmio32be,<addr>[,options[,uartclk]] 1415 uart[8250],0x<addr>[,options] 1416 Start an early, polled-mode console on the 8250/16550 1417 UART at the specified I/O port or MMIO address. 1418 MMIO inter-register address stride is either 8-bit 1419 (mmio) or 32-bit (mmio32 or mmio32be). 1420 If none of [io|mmio|mmio32|mmio32be], <addr> is assumed 1421 to be equivalent to 'mmio'. 'options' are specified 1422 in the same format described for "console=ttyS<n>"; if 1423 unspecified, the h/w is not initialized. 'uartclk' is 1424 the uart clock frequency; if unspecified, it is set 1425 to 'BASE_BAUD' * 16. 1426 1427 pl011,<addr> 1428 pl011,mmio32,<addr> 1429 Start an early, polled-mode console on a pl011 serial 1430 port at the specified address. The pl011 serial port 1431 must already be setup and configured. Options are not 1432 yet supported. If 'mmio32' is specified, then only 1433 the driver will use only 32-bit accessors to read/write 1434 the device registers. 1435 1436 liteuart,<addr> 1437 Start an early console on a litex serial port at the 1438 specified address. The serial port must already be 1439 setup and configured. Options are not yet supported. 1440 1441 meson,<addr> 1442 Start an early, polled-mode console on a meson serial 1443 port at the specified address. The serial port must 1444 already be setup and configured. Options are not yet 1445 supported. 1446 1447 msm_serial,<addr> 1448 Start an early, polled-mode console on an msm serial 1449 port at the specified address. The serial port 1450 must already be setup and configured. Options are not 1451 yet supported. 1452 1453 msm_serial_dm,<addr> 1454 Start an early, polled-mode console on an msm serial 1455 dm port at the specified address. The serial port 1456 must already be setup and configured. Options are not 1457 yet supported. 1458 1459 owl,<addr> 1460 Start an early, polled-mode console on a serial port 1461 of an Actions Semi SoC, such as S500 or S900, at the 1462 specified address. The serial port must already be 1463 setup and configured. Options are not yet supported. 1464 1465 rda,<addr> 1466 Start an early, polled-mode console on a serial port 1467 of an RDA Micro SoC, such as RDA8810PL, at the 1468 specified address. The serial port must already be 1469 setup and configured. Options are not yet supported. 1470 1471 sbi 1472 Use RISC-V SBI (Supervisor Binary Interface) for early 1473 console. 1474 1475 smh Use ARM semihosting calls for early console. 1476 1477 s3c2410,<addr> 1478 s3c2412,<addr> 1479 s3c2440,<addr> 1480 s3c6400,<addr> 1481 s5pv210,<addr> 1482 exynos4210,<addr> 1483 Use early console provided by serial driver available 1484 on Samsung SoCs, requires selecting proper type and 1485 a correct base address of the selected UART port. The 1486 serial port must already be setup and configured. 1487 Options are not yet supported. 1488 1489 lantiq,<addr> 1490 Start an early, polled-mode console on a lantiq serial 1491 (lqasc) port at the specified address. The serial port 1492 must already be setup and configured. Options are not 1493 yet supported. 1494 1495 lpuart,<addr> 1496 lpuart32,<addr> 1497 Use early console provided by Freescale LP UART driver 1498 found on Freescale Vybrid and QorIQ LS1021A processors. 1499 A valid base address must be provided, and the serial 1500 port must already be setup and configured. 1501 1502 ec_imx21,<addr> 1503 ec_imx6q,<addr> 1504 Start an early, polled-mode, output-only console on the 1505 Freescale i.MX UART at the specified address. The UART 1506 must already be setup and configured. 1507 1508 ar3700_uart,<addr> 1509 Start an early, polled-mode console on the 1510 Armada 3700 serial port at the specified 1511 address. The serial port must already be setup 1512 and configured. Options are not yet supported. 1513 1514 qcom_geni,<addr> 1515 Start an early, polled-mode console on a Qualcomm 1516 Generic Interface (GENI) based serial port at the 1517 specified address. The serial port must already be 1518 setup and configured. Options are not yet supported. 1519 1520 efifb,[options] 1521 Start an early, unaccelerated console on the EFI 1522 memory mapped framebuffer (if available). On cache 1523 coherent non-x86 systems that use system memory for 1524 the framebuffer, pass the 'ram' option so that it is 1525 mapped with the correct attributes. 1526 1527 linflex,<addr> 1528 Use early console provided by Freescale LINFlexD UART 1529 serial driver for NXP S32V234 SoCs. A valid base 1530 address must be provided, and the serial port must 1531 already be setup and configured. 1532 1533 earlyprintk= [X86,SH,ARM,M68k,S390,UM,EARLY] 1534 earlyprintk=vga 1535 earlyprintk=sclp 1536 earlyprintk=xen 1537 earlyprintk=serial[,ttySn[,baudrate]] 1538 earlyprintk=serial[,0x...[,baudrate]] 1539 earlyprintk=ttySn[,baudrate] 1540 earlyprintk=dbgp[debugController#] 1541 earlyprintk=mmio32,membase[,{nocfg|baudrate}] 1542 earlyprintk=pciserial[,force],bus:device.function[,{nocfg|baudrate}] 1543 earlyprintk=xdbc[xhciController#] 1544 earlyprintk=bios 1545 1546 earlyprintk is useful when the kernel crashes before 1547 the normal console is initialized. It is not enabled by 1548 default because it has some cosmetic problems. 1549 1550 Use "nocfg" to skip UART configuration, assume 1551 BIOS/firmware has configured UART correctly. 1552 1553 Append ",keep" to not disable it when the real console 1554 takes over. 1555 1556 Only one of vga, serial, or usb debug port can 1557 be used at a time. 1558 1559 Currently only ttyS0 and ttyS1 may be specified by 1560 name. Other I/O ports may be explicitly specified 1561 on some architectures (x86 and arm at least) by 1562 replacing ttySn with an I/O port address, like this: 1563 earlyprintk=serial,0x1008,115200 1564 You can find the port for a given device in 1565 /proc/tty/driver/serial: 1566 2: uart:ST16650V2 port:00001008 irq:18 ... 1567 1568 Interaction with the standard serial driver is not 1569 very good. 1570 1571 The VGA output is eventually overwritten by 1572 the real console. 1573 1574 The xen option can only be used in Xen domains. 1575 1576 The sclp output can only be used on s390. 1577 1578 The bios output can only be used on SuperH. 1579 1580 The optional "force" to "pciserial" enables use of a 1581 PCI device even when its classcode is not of the 1582 UART class. 1583 1584 edac_report= [HW,EDAC] Control how to report EDAC event 1585 Format: {"on" | "off" | "force"} 1586 on: enable EDAC to report H/W event. May be overridden 1587 by other higher priority error reporting module. 1588 off: disable H/W event reporting through EDAC. 1589 force: enforce the use of EDAC to report H/W event. 1590 default: on. 1591 1592 edd= [EDD] 1593 Format: {"off" | "on" | "skip[mbr]"} 1594 1595 efi= [EFI,EARLY] 1596 Format: { "debug", "disable_early_pci_dma", 1597 "nochunk", "noruntime", "nosoftreserve", 1598 "novamap", "no_disable_early_pci_dma" } 1599 debug: enable misc debug output. 1600 disable_early_pci_dma: disable the busmaster bit on all 1601 PCI bridges while in the EFI boot stub. 1602 nochunk: disable reading files in "chunks" in the EFI 1603 boot stub, as chunking can cause problems with some 1604 firmware implementations. 1605 noruntime : disable EFI runtime services support 1606 nosoftreserve: The EFI_MEMORY_SP (Specific Purpose) 1607 attribute may cause the kernel to reserve the 1608 memory range for a memory mapping driver to 1609 claim. Specify efi=nosoftreserve to disable this 1610 reservation and treat the memory by its base type 1611 (i.e. EFI_CONVENTIONAL_MEMORY / "System RAM"). 1612 novamap: do not call SetVirtualAddressMap(). 1613 no_disable_early_pci_dma: Leave the busmaster bit set 1614 on all PCI bridges while in the EFI boot stub 1615 1616 efi_no_storage_paranoia [EFI,X86,EARLY] 1617 Using this parameter you can use more than 50% of 1618 your efi variable storage. Use this parameter only if 1619 you are really sure that your UEFI does sane gc and 1620 fulfills the spec otherwise your board may brick. 1621 1622 efivar_ssdt= [EFI; X86] Name of an EFI variable that contains an SSDT 1623 that is to be dynamically loaded by Linux. If there are 1624 multiple variables with the same name but with different 1625 vendor GUIDs, all of them will be loaded. See 1626 Documentation/admin-guide/acpi/ssdt-overlays.rst for details. 1627 1628 1629 eisa_irq_edge= [PARISC,HW] 1630 See header of drivers/parisc/eisa.c. 1631 1632 ekgdboc= [X86,KGDB,EARLY] Allow early kernel console debugging 1633 Format: ekgdboc=kbd 1634 1635 This is designed to be used in conjunction with 1636 the boot argument: earlyprintk=vga 1637 1638 This parameter works in place of the kgdboc parameter 1639 but can only be used if the backing tty is available 1640 very early in the boot process. For early debugging 1641 via a serial port see kgdboc_earlycon instead. 1642 1643 elfcorehdr=[size[KMG]@]offset[KMG] [PPC,SH,X86,S390,EARLY] 1644 Specifies physical address of start of kernel core 1645 image elf header and optionally the size. Generally 1646 kexec loader will pass this option to capture kernel. 1647 See Documentation/admin-guide/kdump/kdump.rst for details. 1648 1649 enable_mtrr_cleanup [X86,EARLY] 1650 The kernel tries to adjust MTRR layout from continuous 1651 to discrete, to make X server driver able to add WB 1652 entry later. This parameter enables that. 1653 1654 enable_timer_pin_1 [X86] 1655 Enable PIN 1 of APIC timer 1656 Can be useful to work around chipset bugs 1657 (in particular on some ATI chipsets). 1658 The kernel tries to set a reasonable default. 1659 1660 enforcing= [SELINUX] Set initial enforcing status. 1661 Format: {"0" | "1"} 1662 See security/selinux/Kconfig help text. 1663 0 -- permissive (log only, no denials). 1664 1 -- enforcing (deny and log). 1665 Default value is 0. 1666 Value can be changed at runtime via 1667 /sys/fs/selinux/enforce. 1668 1669 erst_disable [ACPI] 1670 Disable Error Record Serialization Table (ERST) 1671 support. 1672 1673 ether= [HW,NET] Ethernet cards parameters 1674 This option is obsoleted by the "netdev=" option, which 1675 has equivalent usage. See its documentation for details. 1676 1677 evm= [EVM] 1678 Format: { "fix" } 1679 Permit 'security.evm' to be updated regardless of 1680 current integrity status. 1681 1682 early_page_ext [KNL,EARLY] Enforces page_ext initialization to earlier 1683 stages so cover more early boot allocations. 1684 Please note that as side effect some optimizations 1685 might be disabled to achieve that (e.g. parallelized 1686 memory initialization is disabled) so the boot process 1687 might take longer, especially on systems with a lot of 1688 memory. Available with CONFIG_PAGE_EXTENSION=y. 1689 1690 failslab= 1691 fail_usercopy= 1692 fail_page_alloc= 1693 fail_skb_realloc= 1694 fail_make_request=[KNL] 1695 General fault injection mechanism. 1696 Format: <interval>,<probability>,<space>,<times> 1697 See also Documentation/fault-injection/. 1698 1699 fb_tunnels= [NET] 1700 Format: { initns | none } 1701 See Documentation/admin-guide/sysctl/net.rst for 1702 fb_tunnels_only_for_init_ns 1703 1704 floppy= [HW] 1705 See Documentation/admin-guide/blockdev/floppy.rst. 1706 1707 forcepae [X86-32] 1708 Forcefully enable Physical Address Extension (PAE). 1709 Many Pentium M systems disable PAE but may have a 1710 functionally usable PAE implementation. 1711 Warning: use of this parameter will taint the kernel 1712 and may cause unknown problems. 1713 1714 fred= [X86-64] 1715 Enable/disable Flexible Return and Event Delivery. 1716 Format: { on | off } 1717 on: enable FRED when it's present, the default setting. 1718 off: disable FRED. 1719 1720 ftrace=[tracer] 1721 [FTRACE] will set and start the specified tracer 1722 as early as possible in order to facilitate early 1723 boot debugging. 1724 1725 ftrace_boot_snapshot 1726 [FTRACE] On boot up, a snapshot will be taken of the 1727 ftrace ring buffer that can be read at: 1728 /sys/kernel/tracing/snapshot. 1729 This is useful if you need tracing information from kernel 1730 boot up that is likely to be overridden by user space 1731 start up functionality. 1732 1733 Optionally, the snapshot can also be defined for a tracing 1734 instance that was created by the trace_instance= command 1735 line parameter. 1736 1737 trace_instance=foo,sched_switch ftrace_boot_snapshot=foo 1738 1739 The above will cause the "foo" tracing instance to trigger 1740 a snapshot at the end of boot up. 1741 1742 ftrace_dump_on_oops[=2(orig_cpu) | =<instance>][,<instance> | 1743 ,<instance>=2(orig_cpu)] 1744 [FTRACE] will dump the trace buffers on oops. 1745 If no parameter is passed, ftrace will dump global 1746 buffers of all CPUs, if you pass 2 or orig_cpu, it 1747 will dump only the buffer of the CPU that triggered 1748 the oops, or the specific instance will be dumped if 1749 its name is passed. Multiple instance dump is also 1750 supported, and instances are separated by commas. Each 1751 instance supports only dump on CPU that triggered the 1752 oops by passing 2 or orig_cpu to it. 1753 1754 ftrace_dump_on_oops=foo=orig_cpu 1755 1756 The above will dump only the buffer of "foo" instance 1757 on CPU that triggered the oops. 1758 1759 ftrace_dump_on_oops,foo,bar=orig_cpu 1760 1761 The above will dump global buffer on all CPUs, the 1762 buffer of "foo" instance on all CPUs and the buffer 1763 of "bar" instance on CPU that triggered the oops. 1764 1765 ftrace_filter=[function-list] 1766 [FTRACE] Limit the functions traced by the function 1767 tracer at boot up. function-list is a comma-separated 1768 list of functions. This list can be changed at run 1769 time by the set_ftrace_filter file in the debugfs 1770 tracing directory. 1771 1772 ftrace_notrace=[function-list] 1773 [FTRACE] Do not trace the functions specified in 1774 function-list. This list can be changed at run time 1775 by the set_ftrace_notrace file in the debugfs 1776 tracing directory. 1777 1778 ftrace_graph_filter=[function-list] 1779 [FTRACE] Limit the top level callers functions traced 1780 by the function graph tracer at boot up. 1781 function-list is a comma-separated list of functions 1782 that can be changed at run time by the 1783 set_graph_function file in the debugfs tracing directory. 1784 1785 ftrace_graph_notrace=[function-list] 1786 [FTRACE] Do not trace from the functions specified in 1787 function-list. This list is a comma-separated list of 1788 functions that can be changed at run time by the 1789 set_graph_notrace file in the debugfs tracing directory. 1790 1791 ftrace_graph_max_depth=<uint> 1792 [FTRACE] Used with the function graph tracer. This is 1793 the max depth it will trace into a function. This value 1794 can be changed at run time by the max_graph_depth file 1795 in the tracefs tracing directory. default: 0 (no limit) 1796 1797 fw_devlink= [KNL,EARLY] Create device links between consumer and supplier 1798 devices by scanning the firmware to infer the 1799 consumer/supplier relationships. This feature is 1800 especially useful when drivers are loaded as modules as 1801 it ensures proper ordering of tasks like device probing 1802 (suppliers first, then consumers), supplier boot state 1803 clean up (only after all consumers have probed), 1804 suspend/resume & runtime PM (consumers first, then 1805 suppliers). 1806 Format: { off | permissive | on | rpm } 1807 off -- Don't create device links from firmware info. 1808 permissive -- Create device links from firmware info 1809 but use it only for ordering boot state clean 1810 up (sync_state() calls). 1811 on -- Create device links from firmware info and use it 1812 to enforce probe and suspend/resume ordering. 1813 rpm -- Like "on", but also use to order runtime PM. 1814 1815 fw_devlink.strict=<bool> 1816 [KNL,EARLY] Treat all inferred dependencies as mandatory 1817 dependencies. This only applies for fw_devlink=on|rpm. 1818 Format: <bool> 1819 1820 fw_devlink.sync_state = 1821 [KNL,EARLY] When all devices that could probe have finished 1822 probing, this parameter controls what to do with 1823 devices that haven't yet received their sync_state() 1824 calls. 1825 Format: { strict | timeout } 1826 strict -- Default. Continue waiting on consumers to 1827 probe successfully. 1828 timeout -- Give up waiting on consumers and call 1829 sync_state() on any devices that haven't yet 1830 received their sync_state() calls after 1831 deferred_probe_timeout has expired or by 1832 late_initcall() if !CONFIG_MODULES. 1833 1834 gamecon.map[2|3]= 1835 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad 1836 support via parallel port (up to 5 devices per port) 1837 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5> 1838 See also Documentation/input/devices/joystick-parport.rst 1839 1840 gamma= [HW,DRM] 1841 1842 gart_fix_e820= [X86-64,EARLY] disable the fix e820 for K8 GART 1843 Format: off | on 1844 default: on 1845 1846 gather_data_sampling= 1847 [X86,INTEL,EARLY] Control the Gather Data Sampling (GDS) 1848 mitigation. 1849 1850 Gather Data Sampling is a hardware vulnerability which 1851 allows unprivileged speculative access to data which was 1852 previously stored in vector registers. 1853 1854 This issue is mitigated by default in updated microcode. 1855 The mitigation may have a performance impact but can be 1856 disabled. On systems without the microcode mitigation 1857 disabling AVX serves as a mitigation. 1858 1859 force: Disable AVX to mitigate systems without 1860 microcode mitigation. No effect if the microcode 1861 mitigation is present. Known to cause crashes in 1862 userspace with buggy AVX enumeration. 1863 1864 off: Disable GDS mitigation. 1865 1866 gbpages [X86] Use GB pages for kernel direct mappings. 1867 1868 gcov_persist= [GCOV] When non-zero (default), profiling data for 1869 kernel modules is saved and remains accessible via 1870 debugfs, even when the module is unloaded/reloaded. 1871 When zero, profiling data is discarded and associated 1872 debugfs files are removed at module unload time. 1873 1874 goldfish [X86] Enable the goldfish android emulator platform. 1875 Don't use this when you are not running on the 1876 android emulator 1877 1878 gpio-mockup.gpio_mockup_ranges 1879 [HW] Sets the ranges of gpiochip of for this device. 1880 Format: <start1>,<end1>,<start2>,<end2>... 1881 gpio-mockup.gpio_mockup_named_lines 1882 [HW] Let the driver know GPIO lines should be named. 1883 1884 gpt [EFI] Forces disk with valid GPT signature but 1885 invalid Protective MBR to be treated as GPT. If the 1886 primary GPT is corrupted, it enables the backup/alternate 1887 GPT to be used instead. 1888 1889 grcan.enable0= [HW] Configuration of physical interface 0. Determines 1890 the "Enable 0" bit of the configuration register. 1891 Format: 0 | 1 1892 Default: 0 1893 grcan.enable1= [HW] Configuration of physical interface 1. Determines 1894 the "Enable 0" bit of the configuration register. 1895 Format: 0 | 1 1896 Default: 0 1897 grcan.select= [HW] Select which physical interface to use. 1898 Format: 0 | 1 1899 Default: 0 1900 grcan.txsize= [HW] Sets the size of the tx buffer. 1901 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0. 1902 Default: 1024 1903 grcan.rxsize= [HW] Sets the size of the rx buffer. 1904 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0. 1905 Default: 1024 1906 1907 hardened_usercopy= 1908 [KNL] Under CONFIG_HARDENED_USERCOPY, whether 1909 hardening is enabled for this boot. Hardened 1910 usercopy checking is used to protect the kernel 1911 from reading or writing beyond known memory 1912 allocation boundaries as a proactive defense 1913 against bounds-checking flaws in the kernel's 1914 copy_to_user()/copy_from_user() interface. 1915 The default is determined by 1916 CONFIG_HARDENED_USERCOPY_DEFAULT_ON. 1917 on Perform hardened usercopy checks. 1918 off Disable hardened usercopy checks. 1919 1920 hardlockup_all_cpu_backtrace= 1921 [KNL] Should the hard-lockup detector generate 1922 backtraces on all cpus. 1923 Format: 0 | 1 1924 1925 hash_pointers= 1926 [KNL,EARLY] 1927 By default, when pointers are printed to the console 1928 or buffers via the %p format string, that pointer is 1929 "hashed", i.e. obscured by hashing the pointer value. 1930 This is a security feature that hides actual kernel 1931 addresses from unprivileged users, but it also makes 1932 debugging the kernel more difficult since unequal 1933 pointers can no longer be compared. The choices are: 1934 Format: { auto | always | never } 1935 Default: auto 1936 1937 auto - Hash pointers unless slab_debug is enabled. 1938 always - Always hash pointers (even if slab_debug is 1939 enabled). 1940 never - Never hash pointers. This option should only 1941 be specified when debugging the kernel. Do 1942 not use on production kernels. The boot 1943 param "no_hash_pointers" is an alias for 1944 this mode. 1945 1946 For controlling hashing dynamically at runtime, 1947 use the "kernel.kptr_restrict" sysctl instead. 1948 1949 hashdist= [KNL,NUMA] Large hashes allocated during boot 1950 are distributed across NUMA nodes. Defaults on 1951 for 64-bit NUMA, off otherwise. 1952 Format: 0 | 1 (for off | on) 1953 1954 hd= [EIDE] (E)IDE hard drive subsystem geometry 1955 Format: <cyl>,<head>,<sect> 1956 1957 hest_disable [ACPI] 1958 Disable Hardware Error Source Table (HEST) support; 1959 corresponding firmware-first mode error processing 1960 logic will be disabled. 1961 1962 hibernate= [HIBERNATION] 1963 noresume Don't check if there's a hibernation image 1964 present during boot. 1965 nocompress Don't compress/decompress hibernation images. 1966 no Disable hibernation and resume. 1967 protect_image Turn on image protection during restoration 1968 (that will set all pages holding image data 1969 during restoration read-only). 1970 1971 hibernate.compressor= [HIBERNATION] Compression algorithm to be 1972 used with hibernation. 1973 Format: { lzo | lz4 } 1974 Default: lzo 1975 1976 lzo: Select LZO compression algorithm to 1977 compress/decompress hibernation image. 1978 1979 lz4: Select LZ4 compression algorithm to 1980 compress/decompress hibernation image. 1981 1982 hibernate.pm_test_delay= 1983 [HIBERNATION] 1984 Sets the number of seconds to remain in a hibernation test 1985 mode before resuming the system (see 1986 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG 1987 is set. Default value is 5. 1988 1989 hibernate_compression_threads= 1990 [HIBERNATION] 1991 Set the number of threads used for compressing or decompressing 1992 hibernation images. 1993 1994 Format: <integer> 1995 Default: 3 1996 Minimum: 1 1997 Example: hibernate_compression_threads=4 1998 1999 highmem=nn[KMG] [KNL,BOOT,EARLY] forces the highmem zone to have an exact 2000 size of <nn>. This works even on boxes that have no 2001 highmem otherwise. This also works to reduce highmem 2002 size on bigger boxes. 2003 2004 highres= [KNL] Enable/disable high resolution timer mode. 2005 Valid parameters: "on", "off" 2006 Default: "on" 2007 2008 hlt [BUGS=ARM,SH] 2009 2010 hostname= [KNL,EARLY] Set the hostname (aka UTS nodename). 2011 Format: <string> 2012 This allows setting the system's hostname during early 2013 startup. This sets the name returned by gethostname. 2014 Using this parameter to set the hostname makes it 2015 possible to ensure the hostname is correctly set before 2016 any userspace processes run, avoiding the possibility 2017 that a process may call gethostname before the hostname 2018 has been explicitly set, resulting in the calling 2019 process getting an incorrect result. The string must 2020 not exceed the maximum allowed hostname length (usually 2021 64 characters) and will be truncated otherwise. 2022 2023 hpet= [X86-32,HPET] option to control HPET usage 2024 Format: { enable (default) | disable | force | 2025 verbose } 2026 disable: disable HPET and use PIT instead 2027 force: allow force enabled of undocumented chips (ICH4, 2028 VIA, nVidia) 2029 verbose: show contents of HPET registers during setup 2030 2031 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET 2032 registers. Default set by CONFIG_HPET_MMAP_DEFAULT. 2033 2034 hugepages= [HW,EARLY] Number of HugeTLB pages to allocate at boot. 2035 If this follows hugepagesz (below), it specifies 2036 the number of pages of hugepagesz to be allocated. 2037 If this is the first HugeTLB parameter on the command 2038 line, it specifies the number of pages to allocate for 2039 the default huge page size. If using node format, the 2040 number of pages to allocate per-node can be specified. 2041 See also Documentation/admin-guide/mm/hugetlbpage.rst. 2042 Format: <integer> or (node format) 2043 <node>:<integer>[,<node>:<integer>] 2044 2045 hugepagesz= 2046 [HW,EARLY] The size of the HugeTLB pages. This is 2047 used in conjunction with hugepages (above) to 2048 allocate huge pages of a specific size at boot. The 2049 pair hugepagesz=X hugepages=Y can be specified once 2050 for each supported huge page size. Huge page sizes 2051 are architecture dependent. See also 2052 Documentation/admin-guide/mm/hugetlbpage.rst. 2053 Format: size[KMG] 2054 2055 hugepage_alloc_threads= 2056 [HW] The number of threads that should be used to 2057 allocate hugepages during boot. This option can be 2058 used to improve system bootup time when allocating 2059 a large amount of huge pages. 2060 The default value is 25% of the available hardware threads. 2061 2062 Note that this parameter only applies to non-gigantic huge pages. 2063 2064 hugetlb_cma= [HW,CMA,EARLY] The size of a CMA area used for allocation 2065 of gigantic hugepages. Or using node format, the size 2066 of a CMA area per node can be specified. 2067 The size can be an absolute value (e.g., 2G) or a 2068 percentage of the total memory or node memory (e.g., 20%). 2069 Percentage-derived sizes are rounded down to a multiple of 2070 the architecture's gigantic hugepage size and may become 2071 zero. 2072 Format: nn[KMGTPE] or nn% or (node format) 2073 <node>:nn[KMGTPE][,<node>:nn[KMGTPE]] or 2074 <node>:nn%[,<node>:nn%] 2075 2076 The size must be a multiple of the gigantic page size. 2077 When using node format, this applies to each per-node size. 2078 Missaligned values are dropped with a warning. 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_flush_htable [IMA] 2321 Flush the IMA hash table when deleting all the 2322 staged measurement records, to achieve maximum 2323 memory saving at the cost of having duplicate 2324 records across the staged measurement lists. 2325 2326 ima_hash= [IMA] 2327 Format: { md5 | sha1 | rmd160 | sha256 | sha384 2328 | sha512 | ... } 2329 default: "sha1" 2330 2331 The list of supported hash algorithms is defined 2332 in crypto/hash_info.h. 2333 2334 ima_policy= [IMA] 2335 The builtin policies to load during IMA setup. 2336 Format: "tcb | appraise_tcb | secure_boot | 2337 fail_securely | critical_data" 2338 2339 The "tcb" policy measures all programs exec'd, files 2340 mmap'd for exec, and all files opened with the read 2341 mode bit set by either the effective uid (euid=0) or 2342 uid=0. 2343 2344 The "appraise_tcb" policy appraises the integrity of 2345 all files owned by root. 2346 2347 The "secure_boot" policy appraises the integrity 2348 of files (eg. kexec kernel image, kernel modules, 2349 firmware, policy, etc) based on file signatures. 2350 2351 The "fail_securely" policy forces file signature 2352 verification failure also on privileged mounted 2353 filesystems with the SB_I_UNVERIFIABLE_SIGNATURE 2354 flag. 2355 2356 The "critical_data" policy measures kernel integrity 2357 critical data. 2358 2359 ima_tcb [IMA] Deprecated. Use ima_policy= instead. 2360 Load a policy which meets the needs of the Trusted 2361 Computing Base. This means IMA will measure all 2362 programs exec'd, files mmap'd for exec, and all files 2363 opened for read by uid=0. 2364 2365 ima_template= [IMA] 2366 Select one of defined IMA measurements template formats. 2367 Formats: { "ima" | "ima-ng" | "ima-ngv2" | "ima-sig" | 2368 "ima-sigv2" } 2369 Default: "ima-ng" 2370 2371 ima_template_fmt= 2372 [IMA] Define a custom template format. 2373 Format: { "field1|...|fieldN" } 2374 2375 ima= [IMA] Enable or disable IMA 2376 Format: { "off" | "on" } 2377 Default: "on" 2378 Note that disabling IMA is limited to kdump kernel. 2379 2380 indirect_target_selection= [X86,Intel] Mitigation control for Indirect 2381 Target Selection(ITS) bug in Intel CPUs. Updated 2382 microcode is also required for a fix in IBPB. 2383 2384 on: Enable mitigation (default). 2385 off: Disable mitigation. 2386 force: Force the ITS bug and deploy default 2387 mitigation. 2388 vmexit: Only deploy mitigation if CPU is affected by 2389 guest/host isolation part of ITS. 2390 stuff: Deploy RSB-fill mitigation when retpoline is 2391 also deployed. Otherwise, deploy the default 2392 mitigation. 2393 2394 For details see: 2395 Documentation/admin-guide/hw-vuln/indirect-target-selection.rst 2396 2397 init= [KNL] 2398 Format: <full_path> 2399 Run specified binary instead of /sbin/init as init 2400 process. 2401 2402 initcall_debug [KNL] Trace initcalls as they are executed. Useful 2403 for working out where the kernel is dying during 2404 startup. 2405 2406 initcall_blacklist= [KNL] Do not execute a comma-separated list of 2407 initcall functions. Useful for debugging built-in 2408 modules and initcalls. 2409 2410 initramfs_async= [KNL] 2411 Format: <bool> 2412 Default: 1 2413 This parameter controls whether the initramfs 2414 image is unpacked asynchronously, concurrently 2415 with devices being probed and 2416 initialized. This should normally just work, 2417 but as a debugging aid, one can get the 2418 historical behaviour of the initramfs 2419 unpacking being completed before device_ and 2420 late_ initcalls. 2421 2422 initrd= [BOOT,EARLY] Specify the location of the initial ramdisk 2423 2424 initrdmem= [KNL,EARLY] Specify a physical address and size from which to 2425 load the initrd. If an initrd is compiled in or 2426 specified in the bootparams, it takes priority over this 2427 setting. 2428 Format: ss[KMG],nn[KMG] 2429 Default is 0, 0 2430 2431 init_on_alloc= [MM,EARLY] Fill newly allocated pages and heap objects with 2432 zeroes. 2433 Format: 0 | 1 2434 Default set by CONFIG_INIT_ON_ALLOC_DEFAULT_ON. 2435 2436 init_on_free= [MM,EARLY] Fill freed pages and heap objects with zeroes. 2437 Format: 0 | 1 2438 Default set by CONFIG_INIT_ON_FREE_DEFAULT_ON. 2439 2440 init_pkru= [X86] Specify the default memory protection keys rights 2441 register contents for all processes. 0x55555554 by 2442 default (disallow access to all but pkey 0). Can 2443 override in debugfs after boot. 2444 2445 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver 2446 Format: <irq> 2447 2448 int_pln_enable [X86] Enable power limit notification interrupt 2449 2450 integrity_audit=[IMA] 2451 Format: { "0" | "1" } 2452 0 -- basic integrity auditing messages. (Default) 2453 1 -- additional integrity auditing messages. 2454 2455 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option 2456 on 2457 Enable intel iommu driver. 2458 off 2459 Disable intel iommu driver. 2460 igfx_off [Default Off] 2461 By default, gfx is mapped as normal device. If a gfx 2462 device has a dedicated DMAR unit, the DMAR unit is 2463 bypassed by not enabling DMAR with this option. In 2464 this case, gfx device will use physical address for 2465 DMA. 2466 strict [Default Off] 2467 Deprecated, equivalent to iommu.strict=1. 2468 sp_off [Default Off] 2469 By default, super page will be supported if Intel IOMMU 2470 has the capability. With this option, super page will 2471 not be supported. 2472 sm_on 2473 Enable the Intel IOMMU scalable mode if the hardware 2474 advertises that it has support for the scalable mode 2475 translation. 2476 sm_off 2477 Disallow use of the Intel IOMMU scalable mode. 2478 tboot_noforce [Default Off] 2479 Do not force the Intel IOMMU enabled under tboot. 2480 By default, tboot will force Intel IOMMU on, which 2481 could harm performance of some high-throughput 2482 devices like 40GBit network cards, even if identity 2483 mapping is enabled. 2484 Note that using this option lowers the security 2485 provided by tboot because it makes the system 2486 vulnerable to DMA attacks. 2487 2488 intel_idle.max_cstate= [KNL,HW,ACPI,X86] 2489 0 disables intel_idle and fall back on acpi_idle. 2490 1 to 9 specify maximum depth of C-state. 2491 2492 intel_pstate= [X86,EARLY] 2493 disable 2494 Do not enable intel_pstate as the default 2495 scaling driver for the supported processors 2496 active 2497 Use intel_pstate driver to bypass the scaling 2498 governors layer of cpufreq and provides it own 2499 algorithms for p-state selection. There are two 2500 P-state selection algorithms provided by 2501 intel_pstate in the active mode: powersave and 2502 performance. The way they both operate depends 2503 on whether or not the hardware managed P-states 2504 (HWP) feature has been enabled in the processor 2505 and possibly on the processor model. 2506 passive 2507 Use intel_pstate as a scaling driver, but configure it 2508 to work with generic cpufreq governors (instead of 2509 enabling its internal governor). This mode cannot be 2510 used along with the hardware-managed P-states (HWP) 2511 feature. 2512 force 2513 Enable intel_pstate on systems that prohibit it by default 2514 in favor of acpi-cpufreq. Forcing the intel_pstate driver 2515 instead of acpi-cpufreq may disable platform features, such 2516 as thermal controls and power capping, that rely on ACPI 2517 P-States information being indicated to OSPM and therefore 2518 should be used with caution. This option does not work with 2519 processors that aren't supported by the intel_pstate driver 2520 or on platforms that use pcc-cpufreq instead of acpi-cpufreq. 2521 no_hwp 2522 Do not enable hardware P state control (HWP) 2523 if available. 2524 hwp_only 2525 Only load intel_pstate on systems which support 2526 hardware P state control (HWP) if available. 2527 support_acpi_ppc 2528 Enforce ACPI _PPC performance limits. If the Fixed ACPI 2529 Description Table, specifies preferred power management 2530 profile as "Enterprise Server" or "Performance Server", 2531 then this feature is turned on by default. 2532 per_cpu_perf_limits 2533 Allow per-logical-CPU P-State performance control limits using 2534 cpufreq sysfs interface 2535 no_cas 2536 Do not enable capacity-aware scheduling (CAS) on 2537 hybrid systems 2538 2539 intremap= [X86-64,Intel-IOMMU,EARLY] 2540 on enable Interrupt Remapping (default) 2541 off disable Interrupt Remapping 2542 nosid disable Source ID checking 2543 no_x2apic_optout 2544 BIOS x2APIC opt-out request will be ignored 2545 nopost disable Interrupt Posting 2546 posted_msi 2547 enable MSIs delivered as posted interrupts 2548 2549 iomem= Disable strict checking of access to MMIO memory 2550 strict regions from userspace. 2551 relaxed 2552 2553 iommu= [X86,EARLY] 2554 2555 off 2556 Don't initialize and use any kind of IOMMU. 2557 2558 force 2559 Force the use of the hardware IOMMU even when 2560 it is not actually needed (e.g. because < 3 GB 2561 memory). 2562 2563 noforce 2564 Don't force hardware IOMMU usage when it is not 2565 needed. (default). 2566 2567 merge 2568 Do scatter-gather (SG) merging. Implies "force" 2569 (experimental). 2570 2571 nomerge 2572 Don't do scatter-gather (SG) merging. 2573 2574 biomerge 2575 Do scatter-gather (SG) merging. Implies "force" 2576 (experimental). [same as "merge"] 2577 2578 panic 2579 Always panic when IOMMU overflows. 2580 2581 nopanic 2582 Don't panic on IOMMU overflows. 2583 2584 pt 2585 Use passththrough mode by default 2586 (Equivalent to iommu.passthrough=1) 2587 2588 nopt 2589 Use translated mode for DMA by default 2590 (Equivalent to iommu.passthrough=0) 2591 2592 soft 2593 Use software bounce buffering (SWIOTLB) (default for 2594 Intel machines). This can be used to prevent the usage 2595 of an available hardware IOMMU. 2596 2597 usedac 2598 Use the DAC on VIA PCI bridge 2599 (default: disable the VIA PCI bridge DAC) 2600 2601 AMD Gart HW IOMMU-specific options: (CONFIG_GART_IOMMU) 2602 2603 <size> 2604 Set the size of the remapping area in bytes. 2605 2606 allowed 2607 Overwrite iommu off workarounds for specific chipsets 2608 2609 force 2610 Overwrite iommu off workarounds for specific chipsets 2611 2612 fullflush 2613 Flush IOMMU on each allocation (default). 2614 2615 nofullflush 2616 Don't use IOMMU fullflush. 2617 2618 memaper[=<order>] 2619 Allocate an own aperture over RAM with size 2620 32MB<<order. (default: order=1, i.e. 64MB) 2621 2622 noaperture 2623 Ask the IOMMU not to touch the aperture for AGP. 2624 2625 noagp 2626 Don't initialize the AGP driver and use full aperture. 2627 2628 iommu= [PPC/POWERNV] 2629 2630 nobypass 2631 Disable IOMMU bypass, using IOMMU for PCI devices. 2632 2633 iommu.forcedac= [ARM64,X86,EARLY] Control IOVA allocation for PCI devices. 2634 Format: { "0" | "1" } 2635 0 - Try to allocate a 32-bit DMA address first, before 2636 falling back to the full range if needed. 2637 1 - Allocate directly from the full usable range, 2638 forcing Dual Address Cycle for PCI cards supporting 2639 greater than 32-bit addressing. 2640 2641 iommu.strict= [ARM64,X86,S390,EARLY] Configure TLB invalidation behaviour 2642 Format: { "0" | "1" } 2643 0 - Lazy mode. 2644 Request that DMA unmap operations use deferred 2645 invalidation of hardware TLBs, for increased 2646 throughput at the cost of reduced device isolation. 2647 Will fall back to strict mode if not supported by 2648 the relevant IOMMU driver. 2649 1 - Strict mode. 2650 DMA unmap operations invalidate IOMMU hardware TLBs 2651 synchronously. 2652 unset - Use value of CONFIG_IOMMU_DEFAULT_DMA_{LAZY,STRICT}. 2653 Note: on x86, strict mode specified via one of the 2654 legacy driver-specific options takes precedence. 2655 2656 iommu.passthrough= 2657 [ARM64,X86,EARLY] Configure DMA to bypass the IOMMU by default. 2658 Format: { "0" | "1" } 2659 0 - Use IOMMU translation for DMA. 2660 1 - Bypass the IOMMU for DMA. 2661 unset - Use value of CONFIG_IOMMU_DEFAULT_PASSTHROUGH. 2662 2663 iommu.debug_pagealloc= 2664 [KNL,EARLY] When CONFIG_IOMMU_DEBUG_PAGEALLOC is set, this 2665 parameter enables the feature at boot time. By default, it 2666 is disabled and the system behaves the same way as a kernel 2667 built without CONFIG_IOMMU_DEBUG_PAGEALLOC. 2668 Format: { "0" | "1" } 2669 0 - Sanitizer disabled. 2670 1 - Sanitizer enabled, expect runtime overhead. 2671 2672 io7= [HW] IO7 for Marvel-based Alpha systems 2673 See comment before marvel_specify_io7 in 2674 arch/alpha/kernel/core_marvel.c. 2675 2676 io_delay= [X86,EARLY] I/O delay method 2677 0x80 2678 Standard port 0x80 based delay 2679 0xed 2680 Alternate port 0xed based delay (needed on some systems) 2681 udelay 2682 Simple two microseconds delay 2683 none 2684 No delay 2685 2686 ip= [IP_PNP] 2687 See Documentation/admin-guide/nfs/nfsroot.rst. 2688 2689 ipcmni_extend [KNL,EARLY] Extend the maximum number of unique System V 2690 IPC identifiers from 32,768 to 16,777,216. 2691 2692 ipe.enforce= [IPE] 2693 Format: <bool> 2694 Determine whether IPE starts in permissive (0) or 2695 enforce (1) mode. The default is enforce. 2696 2697 ipe.success_audit= 2698 [IPE] 2699 Format: <bool> 2700 Start IPE with success auditing enabled, emitting 2701 an audit event when a binary is allowed. The default 2702 is 0. 2703 2704 irqaffinity= [SMP] Set the default irq affinity mask 2705 The argument is a cpu list, as described above. 2706 2707 irqchip.gicv2_force_probe= 2708 [ARM,ARM64,EARLY] 2709 Format: <bool> 2710 Force the kernel to look for the second 4kB page 2711 of a GICv2 controller even if the memory range 2712 exposed by the device tree is too small. 2713 2714 irqchip.gicv3_nolpi= 2715 [ARM,ARM64,EARLY] 2716 Force the kernel to ignore the availability of 2717 LPIs (and by consequence ITSs). Intended for system 2718 that use the kernel as a bootloader, and thus want 2719 to let secondary kernels in charge of setting up 2720 LPIs. 2721 2722 irqchip.gicv3_pseudo_nmi= [ARM64,EARLY] 2723 Enables support for pseudo-NMIs in the kernel. This 2724 requires the kernel to be built with 2725 CONFIG_ARM64_PSEUDO_NMI. 2726 2727 irqchip.riscv_imsic_noipi 2728 [RISC-V,EARLY] 2729 Force the kernel to not use IMSIC software injected MSIs 2730 as IPIs. Intended for system where IMSIC is trap-n-emulated, 2731 and thus want to reduce MMIO traps when triggering IPIs 2732 to multiple harts. 2733 2734 irqfixup [HW] 2735 When an interrupt is not handled search all handlers 2736 for it. Intended to get systems with badly broken 2737 firmware running. 2738 2739 irqhandler.duration_warn_us= [KNL] 2740 Warn if an IRQ handler exceeds the specified duration 2741 threshold in microseconds. Useful for identifying 2742 long-running IRQs in the system. 2743 2744 irqpoll [HW] 2745 When an interrupt is not handled search all handlers 2746 for it. Also check all handlers each timer 2747 interrupt. Intended to get systems with badly broken 2748 firmware running. 2749 2750 isapnp= [ISAPNP] 2751 Format: <RDP>,<reset>,<pci_scan>,<verbosity> 2752 2753 isolcpus= [KNL,SMP,ISOL] Isolate a given set of CPUs from disturbance. 2754 Format: [flag-list,]<cpu-list> 2755 2756 Specify one or more CPUs to isolate from disturbances 2757 specified in the flag list (default: domain): 2758 2759 nohz 2760 Disable the tick when a single task runs as well as 2761 disabling other kernel noises like having RCU callbacks 2762 offloaded. This is equivalent to the nohz_full parameter. 2763 2764 A residual 1Hz tick is offloaded to workqueues, which you 2765 need to affine to housekeeping through the global 2766 workqueue's affinity configured via the 2767 /sys/devices/virtual/workqueue/cpumask sysfs file, or 2768 by using the 'domain' flag described below. 2769 2770 NOTE: by default the global workqueue runs on all CPUs, 2771 so to protect individual CPUs the 'cpumask' file has to 2772 be configured manually after bootup. 2773 2774 domain 2775 Isolate from the general SMP balancing and scheduling 2776 algorithms. Note that performing domain isolation this way 2777 is irreversible: it's not possible to bring back a CPU to 2778 the domains once isolated through this boot time 2779 configuration. Use cpusets for a dynamic configuration 2780 which can be altered at runtime. For details see 2781 Documentation/admin-guide/cpu-isolation.rst. 2782 2783 You can move a process onto or off an "isolated" CPU via 2784 the CPU affinity syscalls or cpuset. 2785 <cpu number> begins at 0 and the maximum value is 2786 "number of CPUs in system - 1". 2787 2788 managed_irq 2789 2790 Isolate from being targeted by managed interrupts 2791 which have an interrupt mask containing isolated 2792 CPUs. The affinity of managed interrupts is 2793 handled by the kernel and cannot be changed via 2794 the /proc/irq/* interfaces. 2795 2796 This isolation is best effort and only effective 2797 if the automatically assigned interrupt mask of a 2798 device queue contains isolated and housekeeping 2799 CPUs. If housekeeping CPUs are online then such 2800 interrupts are directed to the housekeeping CPU 2801 so that IO submitted on the housekeeping CPU 2802 cannot disturb the isolated CPU. 2803 2804 If a queue's affinity mask contains only isolated 2805 CPUs then this parameter has no effect on the 2806 interrupt routing decision, though interrupts are 2807 only delivered when tasks running on those 2808 isolated CPUs submit IO. IO submitted on 2809 housekeeping CPUs has no influence on those 2810 queues. 2811 2812 The format of <cpu-list> is described above. 2813 2814 iucv= [HW,NET] 2815 2816 ivrs_ioapic [HW,X86-64] 2817 Provide an override to the IOAPIC-ID<->DEVICE-ID 2818 mapping provided in the IVRS ACPI table. 2819 By default, PCI segment is 0, and can be omitted. 2820 2821 For example, to map IOAPIC-ID decimal 10 to 2822 PCI segment 0x1 and PCI device 00:14.0, 2823 write the parameter as: 2824 ivrs_ioapic=10@0001:00:14.0 2825 2826 Deprecated formats: 2827 * To map IOAPIC-ID decimal 10 to PCI device 00:14.0 2828 write the parameter as: 2829 ivrs_ioapic[10]=00:14.0 2830 * To map IOAPIC-ID decimal 10 to PCI segment 0x1 and 2831 PCI device 00:14.0 write the parameter as: 2832 ivrs_ioapic[10]=0001:00:14.0 2833 2834 ivrs_hpet [HW,X86-64] 2835 Provide an override to the HPET-ID<->DEVICE-ID 2836 mapping provided in the IVRS ACPI table. 2837 By default, PCI segment is 0, and can be omitted. 2838 2839 For example, to map HPET-ID decimal 10 to 2840 PCI segment 0x1 and PCI device 00:14.0, 2841 write the parameter as: 2842 ivrs_hpet=10@0001:00:14.0 2843 2844 Deprecated formats: 2845 * To map HPET-ID decimal 0 to PCI device 00:14.0 2846 write the parameter as: 2847 ivrs_hpet[0]=00:14.0 2848 * To map HPET-ID decimal 10 to PCI segment 0x1 and 2849 PCI device 00:14.0 write the parameter as: 2850 ivrs_ioapic[10]=0001:00:14.0 2851 2852 ivrs_acpihid [HW,X86-64] 2853 Provide an override to the ACPI-HID:UID<->DEVICE-ID 2854 mapping provided in the IVRS ACPI table. 2855 By default, PCI segment is 0, and can be omitted. 2856 2857 For example, to map UART-HID:UID AMD0020:0 to 2858 PCI segment 0x1 and PCI device ID 00:14.5, 2859 write the parameter as: 2860 ivrs_acpihid=AMD0020:0@0001:00:14.5 2861 2862 Deprecated formats: 2863 * To map UART-HID:UID AMD0020:0 to PCI segment is 0, 2864 PCI device ID 00:14.5, write the parameter as: 2865 ivrs_acpihid[00:14.5]=AMD0020:0 2866 * To map UART-HID:UID AMD0020:0 to PCI segment 0x1 and 2867 PCI device ID 00:14.5, write the parameter as: 2868 ivrs_acpihid[0001:00:14.5]=AMD0020:0 2869 2870 js= [HW,JOY] Analog joystick 2871 See Documentation/input/joydev/joystick.rst. 2872 2873 kasan_multi_shot 2874 [KNL] Enforce KASAN (Kernel Address Sanitizer) to print 2875 report on every invalid memory access. Without this 2876 parameter KASAN will print report only for the first 2877 invalid access. 2878 2879 keep_bootcon [KNL,EARLY] 2880 Do not unregister boot console at start. This is only 2881 useful for debugging when something happens in the window 2882 between unregistering the boot console and initializing 2883 the real console. 2884 2885 keepinitrd [HW,ARM] See retain_initrd. 2886 2887 kernelcore= [KNL,X86,PPC,EARLY] 2888 Format: nn[KMGTPE] | nn% | "mirror" 2889 This parameter specifies the amount of memory usable by 2890 the kernel for non-movable allocations. The requested 2891 amount is spread evenly throughout all nodes in the 2892 system as ZONE_NORMAL. The remaining memory is used for 2893 movable memory in its own zone, ZONE_MOVABLE. In the 2894 event, a node is too small to have both ZONE_NORMAL and 2895 ZONE_MOVABLE, kernelcore memory will take priority and 2896 other nodes will have a larger ZONE_MOVABLE. 2897 2898 ZONE_MOVABLE is used for the allocation of pages that 2899 may be reclaimed or moved by the page migration 2900 subsystem. Note that allocations like PTEs-from-HighMem 2901 still use the HighMem zone if it exists, and the Normal 2902 zone if it does not. 2903 2904 It is possible to specify the exact amount of memory in 2905 the form of "nn[KMGTPE]", a percentage of total system 2906 memory in the form of "nn%", or "mirror". If "mirror" 2907 option is specified, mirrored (reliable) memory is used 2908 for non-movable allocations and remaining memory is used 2909 for Movable pages. "nn[KMGTPE]", "nn%", and "mirror" 2910 are exclusive, so you cannot specify multiple forms. 2911 2912 kfence.burst= [MM,KFENCE] The number of additional successive 2913 allocations to be attempted through KFENCE for each 2914 sample interval. 2915 Format: <unsigned integer> 2916 Default: 0 2917 2918 kfence.check_on_panic= 2919 [MM,KFENCE] Whether to check all KFENCE-managed objects' 2920 canaries on panic. 2921 Format: <bool> 2922 Default: false 2923 2924 kfence.deferrable= 2925 [MM,KFENCE] Whether to use a deferrable timer to trigger 2926 allocations. This avoids forcing CPU wake-ups if the 2927 system is idle, at the risk of a less predictable 2928 sample interval. 2929 Format: <bool> 2930 Default: CONFIG_KFENCE_DEFERRABLE 2931 2932 kfence.fault= [MM,KFENCE] Controls the behavior when a KFENCE 2933 error is detected. 2934 report - print the error report and continue (default). 2935 oops - print the error report and oops. 2936 panic - print the error report and panic. 2937 2938 kfence.sample_interval= 2939 [MM,KFENCE] KFENCE's sample interval in milliseconds. 2940 Format: <unsigned integer> 2941 0 - Disable KFENCE. 2942 >0 - Enabled KFENCE with given sample interval. 2943 Default: CONFIG_KFENCE_SAMPLE_INTERVAL 2944 2945 kfence.skip_covered_thresh= 2946 [MM,KFENCE] If pool utilization reaches this threshold 2947 (pool usage%), KFENCE limits currently covered 2948 allocations of the same source from further filling 2949 up the pool. 2950 Format: <unsigned integer> 2951 Default: 75 2952 2953 kgdbdbgp= [KGDB,HW,EARLY] kgdb over EHCI usb debug port. 2954 Format: <Controller#>[,poll interval] 2955 The controller # is the number of the ehci usb debug 2956 port as it is probed via PCI. The poll interval is 2957 optional and is the number seconds in between 2958 each poll cycle to the debug port in case you need 2959 the functionality for interrupting the kernel with 2960 gdb or control-c on the dbgp connection. When 2961 not using this parameter you use sysrq-g to break into 2962 the kernel debugger. 2963 2964 kgdboc= [KGDB,HW] kgdb over consoles. 2965 Requires a tty driver that supports console polling, 2966 or a supported polling keyboard driver (non-usb). 2967 Serial only format: <serial_device>[,baud] 2968 keyboard only format: kbd 2969 keyboard and serial format: kbd,<serial_device>[,baud] 2970 Optional Kernel mode setting: 2971 kms, kbd format: kms,kbd 2972 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud] 2973 2974 kgdboc_earlycon= [KGDB,HW,EARLY] 2975 If the boot console provides the ability to read 2976 characters and can work in polling mode, you can use 2977 this parameter to tell kgdb to use it as a backend 2978 until the normal console is registered. Intended to 2979 be used together with the kgdboc parameter which 2980 specifies the normal console to transition to. 2981 2982 The name of the early console should be specified 2983 as the value of this parameter. Note that the name of 2984 the early console might be different than the tty 2985 name passed to kgdboc. It's OK to leave the value 2986 blank and the first boot console that implements 2987 read() will be picked. 2988 2989 kgdbwait [KGDB,EARLY] Stop kernel execution and enter the 2990 kernel debugger at the earliest opportunity. 2991 2992 kho= [KEXEC,EARLY] 2993 Format: { "0" | "1" | "off" | "on" | "y" | "n" } 2994 Enables or disables Kexec HandOver. 2995 "0" | "off" | "n" - kexec handover is disabled 2996 "1" | "on" | "y" - kexec handover is enabled 2997 2998 kho_scratch= [KEXEC,EARLY] 2999 Format: ll[KMG],mm[KMG],nn[KMG] | nn% 3000 Defines the size of the KHO scratch region. The KHO 3001 scratch regions are physically contiguous memory 3002 ranges that can only be used for non-kernel 3003 allocations. That way, even when memory is heavily 3004 fragmented with handed over memory, the kexeced 3005 kernel will always have enough contiguous ranges to 3006 bootstrap itself. 3007 3008 It is possible to specify the exact amount of 3009 memory in the form of "ll[KMG],mm[KMG],nn[KMG]" 3010 where the first parameter defines the size of a low 3011 memory scratch area, the second parameter defines 3012 the size of a global scratch area and the third 3013 parameter defines the size of additional per-node 3014 scratch areas. The form "nn%" defines scale factor 3015 (in percents) of memory that was used during boot. 3016 3017 kmac= [MIPS] Korina ethernet MAC address. 3018 Configure the RouterBoard 532 series on-chip 3019 Ethernet adapter MAC address. 3020 3021 kmemleak= [KNL,EARLY] Boot-time kmemleak enable/disable 3022 Valid arguments: on, off 3023 Default: on 3024 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y, 3025 the default is off. 3026 3027 kprobe_event=[probe-list] 3028 [FTRACE] Add kprobe events and enable at boot time. 3029 The probe-list is a semicolon delimited list of probe 3030 definitions. Each definition is same as kprobe_events 3031 interface, but the parameters are comma delimited. 3032 For example, to add a kprobe event on vfs_read with 3033 arg1 and arg2, add to the command line; 3034 3035 kprobe_event=p,vfs_read,$arg1,$arg2 3036 3037 See also Documentation/trace/kprobetrace.rst "Kernel 3038 Boot Parameter" section. 3039 3040 kpti= [ARM64,EARLY] Control page table isolation of 3041 user and kernel address spaces. 3042 Default: enabled on cores which need mitigation. 3043 0: force disabled 3044 1: force enabled 3045 3046 kunit.enable= [KUNIT] Enable executing KUnit tests. Requires 3047 CONFIG_KUNIT to be set to be fully enabled. The 3048 default value can be overridden via 3049 KUNIT_DEFAULT_ENABLED. 3050 Default is 1 (enabled) 3051 3052 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs. 3053 Default is 0 (don't ignore, but inject #GP) 3054 3055 kvm.eager_page_split= 3056 [KVM,X86] Controls whether or not KVM will try to 3057 proactively split all huge pages during dirty logging. 3058 Eager page splitting reduces interruptions to vCPU 3059 execution by eliminating the write-protection faults 3060 and MMU lock contention that would otherwise be 3061 required to split huge pages lazily. 3062 3063 VM workloads that rarely perform writes or that write 3064 only to a small region of VM memory may benefit from 3065 disabling eager page splitting to allow huge pages to 3066 still be used for reads. 3067 3068 The behavior of eager page splitting depends on whether 3069 KVM_DIRTY_LOG_INITIALLY_SET is enabled or disabled. If 3070 disabled, all huge pages in a memslot will be eagerly 3071 split when dirty logging is enabled on that memslot. If 3072 enabled, eager page splitting will be performed during 3073 the KVM_CLEAR_DIRTY ioctl, and only for the pages being 3074 cleared. 3075 3076 Eager page splitting is only supported when kvm.tdp_mmu=Y. 3077 3078 Default is Y (on). 3079 3080 kvm.enable_pmu=[KVM,X86] 3081 If enabled, KVM will virtualize PMU functionality based 3082 on the virtual CPU model defined by userspace. This 3083 can be overridden on a per-VM basis via 3084 KVM_CAP_PMU_CAPABILITY. 3085 3086 If disabled, KVM will not virtualize PMU functionality, 3087 e.g. MSRs, PMCs, PMIs, etc., even if userspace defines 3088 a virtual CPU model that contains PMU assets. 3089 3090 Note, KVM's vPMU support implicitly requires running 3091 with an in-kernel local APIC, e.g. to deliver PMIs to 3092 the guest. Running without an in-kernel local APIC is 3093 not supported, though KVM will allow such a combination 3094 (with severely degraded functionality). 3095 3096 See also enable_mediated_pmu. 3097 3098 Default is Y (on). 3099 3100 kvm.enable_virt_at_load=[KVM,ARM64,LOONGARCH,MIPS,RISCV,X86] 3101 If enabled, KVM will enable virtualization in hardware 3102 when KVM is loaded, and disable virtualization when KVM 3103 is unloaded (if KVM is built as a module). 3104 3105 If disabled, KVM will dynamically enable and disable 3106 virtualization on-demand when creating and destroying 3107 VMs, i.e. on the 0=>1 and 1=>0 transitions of the 3108 number of VMs. 3109 3110 Enabling virtualization at module load avoids potential 3111 latency for creation of the 0=>1 VM, as KVM serializes 3112 virtualization enabling across all online CPUs. The 3113 "cost" of enabling virtualization when KVM is loaded, 3114 is that doing so may interfere with using out-of-tree 3115 hypervisors that want to "own" virtualization hardware. 3116 3117 kvm.enable_vmware_backdoor=[KVM] Support VMware backdoor PV interface. 3118 Default is false (don't support). 3119 3120 kvm.nx_huge_pages= 3121 [KVM] Controls the software workaround for the 3122 X86_BUG_ITLB_MULTIHIT bug. 3123 force : Always deploy workaround. 3124 off : Never deploy workaround. 3125 auto : Deploy workaround based on the presence of 3126 X86_BUG_ITLB_MULTIHIT. 3127 3128 Default is 'auto'. 3129 3130 If the software workaround is enabled for the host, 3131 guests do need not to enable it for nested guests. 3132 3133 kvm.nx_huge_pages_recovery_ratio= 3134 [KVM] Controls how many 4KiB pages are periodically zapped 3135 back to huge pages. 0 disables the recovery, otherwise if 3136 the value is N KVM will zap 1/Nth of the 4KiB pages every 3137 period (see below). The default is 60. 3138 3139 kvm.nx_huge_pages_recovery_period_ms= 3140 [KVM] Controls the time period at which KVM zaps 4KiB pages 3141 back to huge pages. If the value is a non-zero N, KVM will 3142 zap a portion (see ratio above) of the pages every N msecs. 3143 If the value is 0 (the default), KVM will pick a period based 3144 on the ratio, such that a page is zapped after 1 hour on average. 3145 3146 kvm-{amd,intel}.enable_mediated_pmu=[KVM,AMD,INTEL] 3147 If enabled, KVM will provide a mediated virtual PMU, 3148 instead of the default perf-based virtual PMU (if 3149 kvm.enable_pmu is true and PMU is enumerated via the 3150 virtual CPU model). 3151 3152 With a perf-based vPMU, KVM operates as a user of perf, 3153 i.e. emulates guest PMU counters using perf events. 3154 KVM-created perf events are managed by perf as regular 3155 (guest-only) events, e.g. are scheduled in/out, contend 3156 for hardware resources, etc. Using a perf-based vPMU 3157 allows guest and host usage of the PMU to co-exist, but 3158 incurs non-trivial overhead and can result in silently 3159 dropped guest events (due to resource contention). 3160 3161 With a mediated vPMU, hardware PMU state is context 3162 switched around the world switch to/from the guest. 3163 KVM mediates which events the guest can utilize, but 3164 gives the guest direct access to all other PMU assets 3165 when possible (KVM may intercept some accesses if the 3166 virtual CPU model provides a subset of hardware PMU 3167 functionality). Using a mediated vPMU significantly 3168 reduces PMU virtualization overhead and eliminates lost 3169 guest events, but is mutually exclusive with using perf 3170 to profile KVM guests and adds latency to most VM-Exits 3171 (to context switch PMU state). 3172 3173 Default is N (off). 3174 3175 kvm-amd.nested= [KVM,AMD] Control nested virtualization feature in 3176 KVM/SVM. Default is 1 (enabled). 3177 3178 kvm-amd.npt= [KVM,AMD] Control KVM's use of Nested Page Tables, 3179 a.k.a. Two-Dimensional Page Tables. Default is 1 3180 (enabled). Disable by KVM if hardware lacks support 3181 for NPT. 3182 3183 kvm-amd.ciphertext_hiding_asids= 3184 [KVM,AMD] Ciphertext hiding prevents disallowed accesses 3185 to SNP private memory from reading ciphertext. Instead, 3186 reads will see constant default values (0xff). 3187 3188 If ciphertext hiding is enabled, the joint SEV-ES and 3189 SEV-SNP ASID space is partitioned into separate SEV-ES 3190 and SEV-SNP ASID ranges, with the SEV-SNP range being 3191 [1..max_snp_asid] and the SEV-ES range being 3192 (max_snp_asid..min_sev_asid), where min_sev_asid is 3193 enumerated by CPUID.0x.8000_001F[EDX]. 3194 3195 A non-zero value enables SEV-SNP ciphertext hiding and 3196 adjusts the ASID ranges for SEV-ES and SEV-SNP guests. 3197 KVM caps the number of SEV-SNP ASIDs at the maximum 3198 possible value, e.g. specifying -1u will assign all 3199 joint SEV-ES and SEV-SNP ASIDs to SEV-SNP. Note, 3200 assigning all joint ASIDs to SEV-SNP, i.e. configuring 3201 max_snp_asid == min_sev_asid-1, will effectively make 3202 SEV-ES unusable. 3203 3204 kvm-arm.mode= 3205 [KVM,ARM,EARLY] Select one of KVM/arm64's modes of 3206 operation. 3207 3208 none: Forcefully disable KVM. 3209 3210 nvhe: Standard nVHE-based mode, without support for 3211 protected guests. 3212 3213 protected: Mode with support for guests whose state is 3214 kept private from the host, using VHE or 3215 nVHE depending on HW support. 3216 3217 nested: VHE-based mode with support for nested 3218 virtualization. Requires at least ARMv8.4 3219 hardware (with FEAT_NV2). 3220 3221 Defaults to VHE/nVHE based on hardware support. Setting 3222 mode to "protected" will disable kexec and hibernation 3223 for the host. To force nVHE on VHE hardware, add 3224 "arm64_sw.hvhe=0 id_aa64mmfr1.vh=0" to the 3225 command-line. 3226 "nested" and "protected" are experimental and should be 3227 used with extreme caution. 3228 3229 kvm-arm.vgic_v3_group0_trap= 3230 [KVM,ARM,EARLY] Trap guest accesses to GICv3 group-0 3231 system registers 3232 3233 kvm-arm.vgic_v3_group1_trap= 3234 [KVM,ARM,EARLY] Trap guest accesses to GICv3 group-1 3235 system registers 3236 3237 kvm-arm.vgic_v3_common_trap= 3238 [KVM,ARM,EARLY] Trap guest accesses to GICv3 common 3239 system registers 3240 3241 kvm-arm.vgic_v4_enable= 3242 [KVM,ARM,EARLY] Allow use of GICv4 for direct 3243 injection of LPIs. 3244 3245 kvm-arm.wfe_trap_policy= 3246 [KVM,ARM] Control when to set WFE instruction trap for 3247 KVM VMs. Traps are allowed but not guaranteed by the 3248 CPU architecture. 3249 3250 trap: set WFE instruction trap 3251 3252 notrap: clear WFE instruction trap 3253 3254 kvm-arm.wfi_trap_policy= 3255 [KVM,ARM] Control when to set WFI instruction trap for 3256 KVM VMs. Traps are allowed but not guaranteed by the 3257 CPU architecture. 3258 3259 trap: set WFI instruction trap 3260 3261 notrap: clear WFI instruction trap 3262 3263 kvm_cma_resv_ratio=n [PPC,EARLY] 3264 Reserves given percentage from system memory area for 3265 contiguous memory allocation for KVM hash pagetable 3266 allocation. 3267 By default it reserves 5% of total system memory. 3268 Format: <integer> 3269 Default: 5 3270 3271 kvm-intel.ept= [KVM,Intel] Control KVM's use of Extended Page Tables, 3272 a.k.a. Two-Dimensional Page Tables. Default is 1 3273 (enabled). Disable by KVM if hardware lacks support 3274 for EPT. 3275 3276 kvm-intel.emulate_invalid_guest_state= 3277 [KVM,Intel] Control whether to emulate invalid guest 3278 state. Ignored if kvm-intel.enable_unrestricted_guest=1, 3279 as guest state is never invalid for unrestricted 3280 guests. This param doesn't apply to nested guests (L2), 3281 as KVM never emulates invalid L2 guest state. 3282 Default is 1 (enabled). 3283 3284 kvm-intel.flexpriority= 3285 [KVM,Intel] Control KVM's use of FlexPriority feature 3286 (TPR shadow). Default is 1 (enabled). Disable by KVM if 3287 hardware lacks support for it. 3288 3289 kvm-intel.nested= 3290 [KVM,Intel] Control nested virtualization feature in 3291 KVM/VMX. Default is 1 (enabled). 3292 3293 kvm-intel.unrestricted_guest= 3294 [KVM,Intel] Control KVM's use of unrestricted guest 3295 feature (virtualized real and unpaged mode). Default 3296 is 1 (enabled). Disable by KVM if EPT is disabled or 3297 hardware lacks support for it. 3298 3299 kvm-intel.vmentry_l1d_flush=[KVM,Intel] Mitigation for L1 Terminal Fault 3300 CVE-2018-3620. 3301 3302 Valid arguments: never, cond, always 3303 3304 always: L1D cache flush on every VMENTER. 3305 cond: Flush L1D on VMENTER only when the code between 3306 VMEXIT and VMENTER can leak host memory. 3307 never: Disables the mitigation 3308 3309 Default is cond (do L1 cache flush in specific instances) 3310 3311 kvm-intel.vpid= [KVM,Intel] Control KVM's use of Virtual Processor 3312 Identification feature (tagged TLBs). Default is 1 3313 (enabled). Disable by KVM if hardware lacks support 3314 for it. 3315 3316 l1d_flush= [X86,INTEL,EARLY] 3317 Control mitigation for L1D based snooping vulnerability. 3318 3319 Certain CPUs are vulnerable to an exploit against CPU 3320 internal buffers which can forward information to a 3321 disclosure gadget under certain conditions. 3322 3323 In vulnerable processors, the speculatively 3324 forwarded data can be used in a cache side channel 3325 attack, to access data to which the attacker does 3326 not have direct access. 3327 3328 This parameter controls the mitigation. The 3329 options are: 3330 3331 on - enable the interface for the mitigation 3332 3333 l1tf= [X86,EARLY] Control mitigation of the L1TF vulnerability on 3334 affected CPUs 3335 3336 The kernel PTE inversion protection is unconditionally 3337 enabled and cannot be disabled. 3338 3339 full 3340 Provides all available mitigations for the 3341 L1TF vulnerability. Disables SMT and 3342 enables all mitigations in the 3343 hypervisors, i.e. unconditional L1D flush. 3344 3345 SMT control and L1D flush control via the 3346 sysfs interface is still possible after 3347 boot. Hypervisors will issue a warning 3348 when the first VM is started in a 3349 potentially insecure configuration, 3350 i.e. SMT enabled or L1D flush disabled. 3351 3352 full,force 3353 Same as 'full', but disables SMT and L1D 3354 flush runtime control. Implies the 3355 'nosmt=force' command line option. 3356 (i.e. sysfs control of SMT is disabled.) 3357 3358 flush 3359 Leaves SMT enabled and enables the default 3360 hypervisor mitigation, i.e. conditional 3361 L1D flush. 3362 3363 SMT control and L1D flush control via the 3364 sysfs interface is still possible after 3365 boot. Hypervisors will issue a warning 3366 when the first VM is started in a 3367 potentially insecure configuration, 3368 i.e. SMT enabled or L1D flush disabled. 3369 3370 flush,nosmt 3371 3372 Disables SMT and enables the default 3373 hypervisor mitigation. 3374 3375 SMT control and L1D flush control via the 3376 sysfs interface is still possible after 3377 boot. Hypervisors will issue a warning 3378 when the first VM is started in a 3379 potentially insecure configuration, 3380 i.e. SMT enabled or L1D flush disabled. 3381 3382 flush,nowarn 3383 Same as 'flush', but hypervisors will not 3384 warn when a VM is started in a potentially 3385 insecure configuration. 3386 3387 off 3388 Disables hypervisor mitigations and doesn't 3389 emit any warnings. 3390 It also drops the swap size and available 3391 RAM limit restriction on both hypervisor and 3392 bare metal. 3393 3394 Default is 'flush'. 3395 3396 For details see: Documentation/admin-guide/hw-vuln/l1tf.rst 3397 3398 l2cr= [PPC] 3399 3400 l3cr= [PPC] 3401 3402 lapic [X86-32,APIC,EARLY] Enable the local APIC even if BIOS 3403 disabled it. 3404 3405 lapic= [X86,APIC] Do not use TSC deadline 3406 value for LAPIC timer one-shot implementation. Default 3407 back to the programmable timer unit in the LAPIC. 3408 Format: notscdeadline 3409 3410 lapic_timer_c2_ok [X86,APIC,EARLY] trust the local apic timer 3411 in C2 power state. 3412 3413 libata.dma= [LIBATA] DMA control 3414 libata.dma=0 Disable all PATA and SATA DMA 3415 libata.dma=1 PATA and SATA Disk DMA only 3416 libata.dma=2 ATAPI (CDROM) DMA only 3417 libata.dma=4 Compact Flash DMA only 3418 Combinations also work, so libata.dma=3 enables DMA 3419 for disks and CDROMs, but not CFs. 3420 3421 libata.ignore_hpa= [LIBATA] Ignore HPA limit 3422 libata.ignore_hpa=0 keep BIOS limits (default) 3423 libata.ignore_hpa=1 ignore limits, using full disk 3424 3425 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume 3426 when set. 3427 Format: <int> 3428 3429 libata.force= [LIBATA] Force configurations. The format is a comma- 3430 separated list of "[ID:]VAL" where ID is PORT[.DEVICE]. 3431 PORT and DEVICE are decimal numbers matching port, link 3432 or device. Basically, it matches the ATA ID string 3433 printed on console by libata. If the whole ID part is 3434 omitted, the last PORT and DEVICE values are used. If 3435 ID hasn't been specified yet, the configuration applies 3436 to all ports, links and devices. 3437 3438 If only DEVICE is omitted, the parameter applies to 3439 the port and all links and devices behind it. DEVICE 3440 number of 0 either selects the first device or the 3441 first fan-out link behind PMP device. It does not 3442 select the host link. DEVICE number of 15 selects the 3443 host link and device attached to it. 3444 3445 The VAL specifies the configuration to force. As long 3446 as there is no ambiguity, shortcut notation is allowed. 3447 For example, both 1.5 and 1.5G would work for 1.5Gbps. 3448 The following configurations can be forced. 3449 3450 * Cable type: 40c, 80c, short40c, unk, ign or sata. 3451 Any ID with matching PORT is used. 3452 3453 * SATA link speed limit: 1.5Gbps or 3.0Gbps. 3454 3455 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7]. 3456 udma[/][16,25,33,44,66,100,133] notation is also 3457 allowed. 3458 3459 * nohrst, nosrst, norst: suppress hard, soft and both 3460 resets. 3461 3462 * rstonce: only attempt one reset during hot-unplug 3463 link recovery. 3464 3465 * [no]dbdelay: Enable or disable the extra 200ms delay 3466 before debouncing a link PHY and device presence 3467 detection. 3468 3469 * [no]ncq: Turn on or off NCQ. 3470 3471 * [no]ncqtrim: Enable or disable queued DSM TRIM. 3472 3473 * [no]ncqati: Enable or disable NCQ trim on ATI chipset. 3474 3475 * [no]trim: Enable or disable (unqueued) TRIM. 3476 3477 * trim_zero: Indicate that TRIM command zeroes data. 3478 3479 * max_trim_128m: Set 128M maximum trim size limit. 3480 3481 * [no]dma: Turn on or off DMA transfers. 3482 3483 * atapi_dmadir: Enable ATAPI DMADIR bridge support. 3484 3485 * atapi_mod16_dma: Enable the use of ATAPI DMA for 3486 commands that are not a multiple of 16 bytes. 3487 3488 * [no]dmalog: Enable or disable the use of the 3489 READ LOG DMA EXT command to access logs. 3490 3491 * [no]iddevlog: Enable or disable access to the 3492 identify device data log. 3493 3494 * [no]logdir: Enable or disable access to the general 3495 purpose log directory. 3496 3497 * max_sec=<sectors>: Set the transfer size limit, in 3498 number of 512-byte sectors, to the value specified in 3499 <sectors>. The value specified in <sectors> has to be 3500 a non-zero positive integer. 3501 3502 * max_sec_128: Set transfer size limit to 128 sectors. 3503 3504 * max_sec_1024: Set or clear transfer size limit to 3505 1024 sectors. 3506 3507 * max_sec_lba48: Set or clear transfer size limit to 3508 65535 sectors. 3509 3510 * external: Mark port as external (hotplug-capable). 3511 3512 * [no]lpm: Enable or disable link power management. 3513 3514 * [no]setxfer: Indicate if transfer speed mode setting 3515 should be skipped. 3516 3517 * [no]fua: Disable or enable FUA (Force Unit Access) 3518 support for devices supporting this feature. 3519 3520 * dump_id: Dump IDENTIFY data. 3521 3522 * disable: Disable this device. 3523 3524 If there are multiple matching configurations changing 3525 the same attribute, the last one is used. 3526 3527 liveupdate= [KNL,EARLY] 3528 Format: <bool> 3529 Enable Live Update Orchestrator (LUO). 3530 Default: off. 3531 3532 lockd.nlm_grace_period=P [NFS] Assign grace period. 3533 Format: <integer> 3534 3535 lockd.nlm_tcpport=N [NFS] Assign TCP port. 3536 Format: <integer> 3537 3538 lockd.nlm_timeout=T [NFS] Assign timeout value. 3539 Format: <integer> 3540 3541 lockd.nlm_udpport=M [NFS] Assign UDP port. 3542 Format: <integer> 3543 3544 lockdown= [SECURITY,EARLY] 3545 { integrity | confidentiality } 3546 Enable the kernel lockdown feature. If set to 3547 integrity, kernel features that allow userland to 3548 modify the running kernel are disabled. If set to 3549 confidentiality, kernel features that allow userland 3550 to extract confidential information from the kernel 3551 are also disabled. 3552 3553 locktorture.acq_writer_lim= [KNL] 3554 Set the time limit in jiffies for a lock 3555 acquisition. Acquisitions exceeding this limit 3556 will result in a splat once they do complete. 3557 3558 locktorture.bind_readers= [KNL] 3559 Specify the list of CPUs to which the readers are 3560 to be bound. 3561 3562 locktorture.bind_writers= [KNL] 3563 Specify the list of CPUs to which the writers are 3564 to be bound. 3565 3566 locktorture.call_rcu_chains= [KNL] 3567 Specify the number of self-propagating call_rcu() 3568 chains to set up. These are used to ensure that 3569 there is a high probability of an RCU grace period 3570 in progress at any given time. Defaults to 0, 3571 which disables these call_rcu() chains. 3572 3573 locktorture.long_hold= [KNL] 3574 Specify the duration in milliseconds for the 3575 occasional long-duration lock hold time. Defaults 3576 to 100 milliseconds. Select 0 to disable. 3577 3578 locktorture.nested_locks= [KNL] 3579 Specify the maximum lock nesting depth that 3580 locktorture is to exercise, up to a limit of 8 3581 (MAX_NESTED_LOCKS). Specify zero to disable. 3582 Note that this parameter is ineffective on types 3583 of locks that do not support nested acquisition. 3584 3585 locktorture.nreaders_stress= [KNL] 3586 Set the number of locking read-acquisition kthreads. 3587 Defaults to being automatically set based on the 3588 number of online CPUs. 3589 3590 locktorture.nwriters_stress= [KNL] 3591 Set the number of locking write-acquisition kthreads. 3592 3593 locktorture.onoff_holdoff= [KNL] 3594 Set time (s) after boot for CPU-hotplug testing. 3595 3596 locktorture.onoff_interval= [KNL] 3597 Set time (s) between CPU-hotplug operations, or 3598 zero to disable CPU-hotplug testing. 3599 3600 locktorture.rt_boost= [KNL] 3601 Do periodic testing of real-time lock priority 3602 boosting. Select 0 to disable, 1 to boost 3603 only rt_mutex, and 2 to boost unconditionally. 3604 Defaults to 2, which might seem to be an 3605 odd choice, but which should be harmless for 3606 non-real-time spinlocks, due to their disabling 3607 of preemption. Note that non-realtime mutexes 3608 disable boosting. 3609 3610 locktorture.rt_boost_factor= [KNL] 3611 Number that determines how often and for how 3612 long priority boosting is exercised. This is 3613 scaled down by the number of writers, so that the 3614 number of boosts per unit time remains roughly 3615 constant as the number of writers increases. 3616 On the other hand, the duration of each boost 3617 increases with the number of writers. 3618 3619 locktorture.shuffle_interval= [KNL] 3620 Set task-shuffle interval (jiffies). Shuffling 3621 tasks allows some CPUs to go into dyntick-idle 3622 mode during the locktorture test. 3623 3624 locktorture.shutdown_secs= [KNL] 3625 Set time (s) after boot system shutdown. This 3626 is useful for hands-off automated testing. 3627 3628 locktorture.stat_interval= [KNL] 3629 Time (s) between statistics printk()s. 3630 3631 locktorture.stutter= [KNL] 3632 Time (s) to stutter testing, for example, 3633 specifying five seconds causes the test to run for 3634 five seconds, wait for five seconds, and so on. 3635 This tests the locking primitive's ability to 3636 transition abruptly to and from idle. 3637 3638 locktorture.torture_type= [KNL] 3639 Specify the locking implementation to test. 3640 3641 locktorture.verbose= [KNL] 3642 Enable additional printk() statements. 3643 3644 locktorture.writer_fifo= [KNL] 3645 Run the write-side locktorture kthreads at 3646 sched_set_fifo() real-time priority. 3647 3648 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver 3649 Format: <irq> 3650 3651 loglevel= [KNL,EARLY] 3652 All Kernel Messages with a loglevel smaller than the 3653 console loglevel will be printed to the console. It can 3654 also be changed with klogd or other programs. The 3655 loglevels are defined as follows: 3656 3657 0 (KERN_EMERG) system is unusable 3658 1 (KERN_ALERT) action must be taken immediately 3659 2 (KERN_CRIT) critical conditions 3660 3 (KERN_ERR) error conditions 3661 4 (KERN_WARNING) warning conditions 3662 5 (KERN_NOTICE) normal but significant condition 3663 6 (KERN_INFO) informational 3664 7 (KERN_DEBUG) debug-level messages 3665 3666 log_buf_len=n[KMG] [KNL,EARLY] 3667 Sets the size of the printk ring buffer, in bytes. 3668 n must be a power of two and greater than the 3669 minimal size. The minimal size is defined by 3670 LOG_BUF_SHIFT kernel config parameter. There 3671 is also CONFIG_LOG_CPU_MAX_BUF_SHIFT config 3672 parameter that allows to increase the default size 3673 depending on the number of CPUs. See init/Kconfig 3674 for more details. 3675 3676 logo.nologo [FB] Disables display of the built-in Linux logo. 3677 This may be used to provide more screen space for 3678 kernel log messages and is useful when debugging 3679 kernel boot problems. 3680 3681 lp=0 [LP] Specify parallel ports to use, e.g, 3682 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses 3683 lp=reset first parallel port). 'lp=0' disables the 3684 lp=auto printer driver. 'lp=reset' (which can be 3685 specified in addition to the ports) causes 3686 attached printers to be reset. Using 3687 lp=port1,port2,... specifies the parallel ports 3688 to associate lp devices with, starting with 3689 lp0. A port specification may be 'none' to skip 3690 that lp device, or a parport name such as 3691 'parport0'. Specifying 'lp=auto' instead of a 3692 port specification list means that device IDs 3693 from each port should be examined, to see if 3694 an IEEE 1284-compliant printer is attached; if 3695 so, the driver will manage that printer. 3696 See also header of drivers/char/lp.c. 3697 3698 lpj=n [KNL] 3699 Sets loops_per_jiffy to given constant, thus avoiding 3700 time-consuming boot-time autodetection (up to 250 ms per 3701 CPU). 0 enables autodetection (default). To determine 3702 the correct value for your kernel, boot with normal 3703 autodetection and see what value is printed. Note that 3704 on SMP systems the preset will be applied to all CPUs, 3705 which is likely to cause problems if your CPUs need 3706 significantly divergent settings. An incorrect value 3707 will cause delays in the kernel to be wrong, leading to 3708 unpredictable I/O errors and other breakage. Although 3709 unlikely, in the extreme case this might damage your 3710 hardware. 3711 3712 lsm.debug [SECURITY] Enable LSM initialization debugging output. 3713 3714 lsm=lsm1,...,lsmN 3715 [SECURITY] Choose order of LSM initialization. This 3716 overrides CONFIG_LSM, and the "security=" parameter. 3717 3718 machtype= [Loongson] Share the same kernel image file between 3719 different yeeloong laptops. 3720 Example: machtype=lemote-yeeloong-2f-7inch 3721 3722 maxcpus= [SMP,EARLY] Maximum number of processors that an SMP kernel 3723 will bring up during bootup. maxcpus=n : n >= 0 limits 3724 the kernel to bring up 'n' processors. Surely after 3725 bootup you can bring up the other plugged cpu by executing 3726 "echo 1 > /sys/devices/system/cpu/cpuX/online". So maxcpus 3727 only takes effect during system bootup. 3728 While n=0 is a special case, it is equivalent to "nosmp", 3729 which also disables the IO APIC. 3730 3731 max_loop= [LOOP] The number of loop block devices that get 3732 (loop.max_loop) unconditionally pre-created at init time. The default 3733 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead 3734 of statically allocating a predefined number, loop 3735 devices can be requested on-demand with the 3736 /dev/loop-control interface. 3737 3738 mce= [X86-{32,64}] 3739 3740 Please see Documentation/arch/x86/x86_64/machinecheck.rst for sysfs runtime tunables. 3741 3742 off 3743 disable machine check 3744 3745 no_cmci 3746 disable CMCI(Corrected Machine Check Interrupt) that 3747 Intel processor supports. Usually this disablement is 3748 not recommended, but it might be handy if your 3749 hardware is misbehaving. 3750 3751 Note that you'll get more problems without CMCI than 3752 with due to the shared banks, i.e. you might get 3753 duplicated error logs. 3754 3755 dont_log_ce 3756 don't make logs for corrected errors. All events 3757 reported as corrected are silently cleared by OS. This 3758 option will be useful if you have no interest in any 3759 of corrected errors. 3760 3761 ignore_ce 3762 disable features for corrected errors, e.g. 3763 polling timer and CMCI. All events reported as 3764 corrected are not cleared by OS and remained in its 3765 error banks. 3766 3767 Usually this disablement is not recommended, however 3768 if there is an agent checking/clearing corrected 3769 errors (e.g. BIOS or hardware monitoring 3770 applications), conflicting with OS's error handling, 3771 and you cannot deactivate the agent, then this option 3772 will be a help. 3773 3774 no_lmce 3775 do not opt-in to Local MCE delivery. Use legacy method 3776 to broadcast MCEs. 3777 3778 bootlog 3779 enable logging of machine checks left over from 3780 booting. Disabled by default on AMD Fam10h and older 3781 because some BIOS leave bogus ones. 3782 3783 If your BIOS doesn't do that it's a good idea to 3784 enable though to make sure you log even machine check 3785 events that result in a reboot. On Intel systems it is 3786 enabled by default. 3787 3788 nobootlog 3789 disable boot machine check logging. 3790 3791 monarchtimeout (number) 3792 sets the time in us to wait for other CPUs on machine 3793 checks. 0 to disable. 3794 3795 bios_cmci_threshold 3796 don't overwrite the bios-set CMCI threshold. This boot 3797 option prevents Linux from overwriting the CMCI 3798 threshold set by the bios. Without this option, Linux 3799 always sets the CMCI threshold to 1. Enabling this may 3800 make memory predictive failure analysis less effective 3801 if the bios sets thresholds for memory errors since we 3802 will not see details for all errors. 3803 3804 recovery 3805 force-enable recoverable machine check code paths 3806 3807 Everything else is in sysfs now. 3808 3809 3810 md= [HW] RAID subsystems devices and level 3811 See Documentation/admin-guide/md.rst. 3812 3813 mds= [X86,INTEL,EARLY] 3814 Control mitigation for the Micro-architectural Data 3815 Sampling (MDS) vulnerability. 3816 3817 Certain CPUs are vulnerable to an exploit against CPU 3818 internal buffers which can forward information to a 3819 disclosure gadget under certain conditions. 3820 3821 In vulnerable processors, the speculatively 3822 forwarded data can be used in a cache side channel 3823 attack, to access data to which the attacker does 3824 not have direct access. 3825 3826 This parameter controls the MDS mitigation. The 3827 options are: 3828 3829 full - Enable MDS mitigation on vulnerable CPUs 3830 full,nosmt - Enable MDS mitigation and disable 3831 SMT on vulnerable CPUs 3832 off - Unconditionally disable MDS mitigation 3833 3834 On TAA-affected machines, mds=off can be prevented by 3835 an active TAA mitigation as both vulnerabilities are 3836 mitigated with the same mechanism so in order to disable 3837 this mitigation, you need to specify tsx_async_abort=off 3838 too. 3839 3840 Not specifying this option is equivalent to 3841 mds=full. 3842 3843 For details see: Documentation/admin-guide/hw-vuln/mds.rst 3844 3845 mem=nn[KMG] [HEXAGON,EARLY] Set the memory size. 3846 Must be specified, otherwise memory size will be 0. 3847 3848 mem=nn[KMG] [KNL,BOOT,EARLY] Force usage of a specific amount 3849 of memory Amount of memory to be used in cases 3850 as follows: 3851 3852 1 for test; 3853 2 when the kernel is not able to see the whole system memory; 3854 3 memory that lies after 'mem=' boundary is excluded from 3855 the hypervisor, then assigned to KVM guests. 3856 4 to limit the memory available for kdump kernel. 3857 3858 [ARC,MICROBLAZE] - the limit applies only to low memory, 3859 high memory is not affected. 3860 3861 [ARM64] - only limits memory covered by the linear 3862 mapping. The NOMAP regions are not affected. 3863 3864 [X86] Work as limiting max address. Use together 3865 with memmap= to avoid physical address space collisions. 3866 Without memmap= PCI devices could be placed at addresses 3867 belonging to unused RAM. 3868 3869 Note that this only takes effects during boot time since 3870 in above case 3, memory may need be hot added after boot 3871 if system memory of hypervisor is not sufficient. 3872 3873 mem=nn[KMG]@ss[KMG] 3874 [ARM,MIPS,EARLY] - override the memory layout 3875 reported by firmware. 3876 Define a memory region of size nn[KMG] starting at 3877 ss[KMG]. 3878 Multiple different regions can be specified with 3879 multiple mem= parameters on the command line. 3880 3881 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel 3882 memory. 3883 3884 memblock=debug [KNL,EARLY] Enable memblock debug messages. 3885 3886 memchunk=nn[KMG] 3887 [KNL,SH] Allow user to override the default size for 3888 per-device physically contiguous DMA buffers. 3889 3890 memhp_default_state=online/offline/online_kernel/online_movable 3891 [KNL] Set the initial state for the memory hotplug 3892 onlining policy. If not specified, the default value is 3893 set according to the 3894 CONFIG_MHP_DEFAULT_ONLINE_TYPE kernel config 3895 options. 3896 See Documentation/admin-guide/mm/memory-hotplug.rst. 3897 3898 memmap=exactmap [KNL,X86,EARLY] Enable setting of an exact 3899 E820 memory map, as specified by the user. 3900 Such memmap=exactmap lines can be constructed based on 3901 BIOS output or other requirements. See the memmap=nn@ss 3902 option description. 3903 3904 memmap=nn[KMG]@ss[KMG] 3905 [KNL, X86,MIPS,XTENSA,EARLY] Force usage of a specific region of memory. 3906 Region of memory to be used is from ss to ss+nn. 3907 If @ss[KMG] is omitted, it is equivalent to mem=nn[KMG], 3908 which limits max address to nn[KMG]. 3909 Multiple different regions can be specified, 3910 comma delimited. 3911 Example: 3912 memmap=100M@2G,100M#3G,1G!1024G 3913 3914 memmap=nn[KMG]#ss[KMG] 3915 [KNL,ACPI,EARLY] Mark specific memory as ACPI data. 3916 Region of memory to be marked is from ss to ss+nn. 3917 3918 memmap=nn[KMG]$ss[KMG] 3919 [KNL,ACPI,EARLY] Mark specific memory as reserved. 3920 Region of memory to be reserved is from ss to ss+nn. 3921 Example: Exclude memory from 0x18690000-0x1869ffff 3922 memmap=64K$0x18690000 3923 or 3924 memmap=0x10000$0x18690000 3925 Some bootloaders may need an escape character before '$', 3926 like Grub2, otherwise '$' and the following number 3927 will be eaten. 3928 3929 memmap=nn[KMG]!ss[KMG,EARLY] 3930 [KNL,X86] Mark specific memory as protected. 3931 Region of memory to be used, from ss to ss+nn. 3932 The memory region may be marked as e820 type 12 (0xc) 3933 and is NVDIMM or ADR memory. 3934 3935 memmap=<size>%<offset>-<oldtype>+<newtype> 3936 [KNL,ACPI,EARLY] Convert memory within the specified region 3937 from <oldtype> to <newtype>. If "-<oldtype>" is left 3938 out, the whole region will be marked as <newtype>, 3939 even if previously unavailable. If "+<newtype>" is left 3940 out, matching memory will be removed. Types are 3941 specified as e820 types, e.g., 1 = RAM, 2 = reserved, 3942 3 = ACPI, 12 = PRAM. 3943 3944 memory_corruption_check=0/1 [X86,EARLY] 3945 Some BIOSes seem to corrupt the first 64k of 3946 memory when doing things like suspend/resume. 3947 Setting this option will scan the memory 3948 looking for corruption. Enabling this will 3949 both detect corruption and prevent the kernel 3950 from using the memory being corrupted. 3951 However, it's intended as a diagnostic tool; if 3952 repeatable BIOS-originated corruption always 3953 affects the same memory, you can use memmap= 3954 to prevent the kernel from using that memory. 3955 3956 memory_corruption_check_size=size [X86,EARLY] 3957 By default it checks for corruption in the low 3958 64k, making this memory unavailable for normal 3959 use. Use this parameter to scan for 3960 corruption in more or less memory. 3961 3962 memory_corruption_check_period=seconds [X86,EARLY] 3963 By default it checks for corruption every 60 3964 seconds. Use this parameter to check at some 3965 other rate. 0 disables periodic checking. 3966 3967 memory_hotplug.memmap_on_memory 3968 [KNL,X86,ARM] Boolean flag to enable this feature. 3969 Format: {on | off (default)} 3970 When enabled, runtime hotplugged memory will 3971 allocate its internal metadata (struct pages, 3972 those vmemmap pages cannot be optimized even 3973 if hugetlb_free_vmemmap is enabled) from the 3974 hotadded memory which will allow to hotadd a 3975 lot of memory without requiring additional 3976 memory to do so. 3977 This feature is disabled by default because it 3978 has some implication on large (e.g. GB) 3979 allocations in some configurations (e.g. small 3980 memory blocks). 3981 The state of the flag can be read in 3982 /sys/module/memory_hotplug/parameters/memmap_on_memory. 3983 Note that even when enabled, there are a few cases where 3984 the feature is not effective. 3985 3986 memtest= [KNL,X86,ARM,M68K,PPC,RISCV,EARLY] Enable memtest 3987 Format: <integer> 3988 default : 0 <disable> 3989 Specifies the number of memtest passes to be 3990 performed. Each pass selects another test 3991 pattern from a given set of patterns. Memtest 3992 fills the memory with this pattern, validates 3993 memory contents and reserves bad memory 3994 regions that are detected. 3995 3996 mem_encrypt= [X86-64] AMD Secure Memory Encryption (SME) control 3997 Valid arguments: on, off 3998 Default: off 3999 mem_encrypt=on: Activate SME 4000 mem_encrypt=off: Do not activate SME 4001 4002 Refer to Documentation/virt/kvm/x86/amd-memory-encryption.rst 4003 for details on when memory encryption can be activated. 4004 4005 mem_sleep_default= [SUSPEND] Default system suspend mode: 4006 s2idle - Suspend-To-Idle 4007 shallow - Power-On Suspend or equivalent (if supported) 4008 deep - Suspend-To-RAM or equivalent (if supported) 4009 See Documentation/admin-guide/pm/sleep-states.rst. 4010 4011 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when 4012 the BIOS has incorrectly applied a workaround. TinyBIOS 4013 version 0.98 is known to be affected, 0.99 fixes the 4014 problem by letting the user disable the workaround. 4015 4016 mga= [HW,DRM] 4017 4018 microcode= [X86] Control the behavior of the microcode loader. 4019 Available options, comma separated: 4020 4021 base_rev=X - with <X> with format: <u32> 4022 Set the base microcode revision of each thread when in 4023 debug mode. 4024 4025 dis_ucode_ldr: disable the microcode loader 4026 4027 force_minrev: 4028 Enable or disable the microcode minimal revision 4029 enforcement for the runtime microcode loader. 4030 4031 mini2440= [ARM,HW,KNL] 4032 Format:[0..2][b][c][t] 4033 Default: "0tb" 4034 MINI2440 configuration specification: 4035 0 - The attached screen is the 3.5" TFT 4036 1 - The attached screen is the 7" TFT 4037 2 - The VGA Shield is attached (1024x768) 4038 Leaving out the screen size parameter will not load 4039 the TFT driver, and the framebuffer will be left 4040 unconfigured. 4041 b - Enable backlight. The TFT backlight pin will be 4042 linked to the kernel VESA blanking code and a GPIO 4043 LED. This parameter is not necessary when using the 4044 VGA shield. 4045 c - Enable the s3c camera interface. 4046 t - Reserved for enabling touchscreen support. The 4047 touchscreen support is not enabled in the mainstream 4048 kernel as of 2.6.30, a preliminary port can be found 4049 in the "bleeding edge" mini2440 support kernel at 4050 https://repo.or.cz/w/linux-2.6/mini2440.git 4051 4052 mitigations= 4053 [X86,PPC,S390,ARM64,EARLY] Control optional mitigations for 4054 CPU vulnerabilities. This is a set of curated, 4055 arch-independent options, each of which is an 4056 aggregation of existing arch-specific options. 4057 4058 Note, "mitigations" is supported if and only if the 4059 kernel was built with CPU_MITIGATIONS=y. 4060 4061 off 4062 Disable all optional CPU mitigations. This 4063 improves system performance, but it may also 4064 expose users to several CPU vulnerabilities. 4065 Equivalent to: if nokaslr then kpti=0 [ARM64] 4066 gather_data_sampling=off [X86] 4067 indirect_target_selection=off [X86] 4068 kvm.nx_huge_pages=off [X86] 4069 l1tf=off [X86] 4070 mds=off [X86] 4071 mmio_stale_data=off [X86] 4072 no_entry_flush [PPC] 4073 no_uaccess_flush [PPC] 4074 nobp=0 [S390] 4075 nopti [X86,PPC] 4076 nospectre_bhb [ARM64] 4077 nospectre_v1 [X86,PPC] 4078 nospectre_v2 [X86,PPC,S390,ARM64] 4079 reg_file_data_sampling=off [X86] 4080 retbleed=off [X86] 4081 spec_rstack_overflow=off [X86] 4082 spec_store_bypass_disable=off [X86,PPC] 4083 spectre_bhi=off [X86] 4084 spectre_v2_user=off [X86] 4085 srbds=off [X86,INTEL] 4086 ssbd=force-off [ARM64] 4087 tsa=off [X86,AMD] 4088 tsx_async_abort=off [X86] 4089 vmscape=off [X86] 4090 4091 Exceptions: 4092 This does not have any effect on 4093 kvm.nx_huge_pages when 4094 kvm.nx_huge_pages=force. 4095 4096 auto (default) 4097 Mitigate all CPU vulnerabilities, but leave SMT 4098 enabled, even if it's vulnerable. This is for 4099 users who don't want to be surprised by SMT 4100 getting disabled across kernel upgrades, or who 4101 have other ways of avoiding SMT-based attacks. 4102 Equivalent to: (default behavior) 4103 4104 auto,nosmt 4105 Mitigate all CPU vulnerabilities, disabling SMT 4106 if needed. This is for users who always want to 4107 be fully mitigated, even if it means losing SMT. 4108 Equivalent to: l1tf=flush,nosmt [X86] 4109 mds=full,nosmt [X86] 4110 tsx_async_abort=full,nosmt [X86] 4111 mmio_stale_data=full,nosmt [X86] 4112 retbleed=auto,nosmt [X86] 4113 4114 [X86] After one of the above options, additionally 4115 supports attack-vector based controls as documented in 4116 Documentation/admin-guide/hw-vuln/attack_vector_controls.rst 4117 4118 mminit_loglevel= 4119 [KNL,EARLY] When CONFIG_DEBUG_MEMORY_INIT is set, this 4120 parameter allows control of the logging verbosity for 4121 the additional memory initialisation checks. A value 4122 of 0 disables mminit logging and a level of 4 will 4123 log everything. Information is printed at KERN_DEBUG 4124 so loglevel=8 may also need to be specified. 4125 4126 mmio_stale_data= 4127 [X86,INTEL,EARLY] Control mitigation for the Processor 4128 MMIO Stale Data vulnerabilities. 4129 4130 Processor MMIO Stale Data is a class of 4131 vulnerabilities that may expose data after an MMIO 4132 operation. Exposed data could originate or end in 4133 the same CPU buffers as affected by MDS and TAA. 4134 Therefore, similar to MDS and TAA, the mitigation 4135 is to clear the affected CPU buffers. 4136 4137 This parameter controls the mitigation. The 4138 options are: 4139 4140 full - Enable mitigation on vulnerable CPUs 4141 4142 full,nosmt - Enable mitigation and disable SMT on 4143 vulnerable CPUs. 4144 4145 off - Unconditionally disable mitigation 4146 4147 On MDS or TAA affected machines, 4148 mmio_stale_data=off can be prevented by an active 4149 MDS or TAA mitigation as these vulnerabilities are 4150 mitigated with the same mechanism so in order to 4151 disable this mitigation, you need to specify 4152 mds=off and tsx_async_abort=off too. 4153 4154 Not specifying this option is equivalent to 4155 mmio_stale_data=full. 4156 4157 For details see: 4158 Documentation/admin-guide/hw-vuln/processor_mmio_stale_data.rst 4159 4160 <module>.async_probe[=<bool>] [KNL] 4161 If no <bool> value is specified or if the value 4162 specified is not a valid <bool>, enable asynchronous 4163 probe on this module. Otherwise, enable/disable 4164 asynchronous probe on this module as indicated by the 4165 <bool> value. See also: module.async_probe 4166 4167 module.async_probe=<bool> 4168 [KNL] When set to true, modules will use async probing 4169 by default. To enable/disable async probing for a 4170 specific module, use the module specific control that 4171 is documented under <module>.async_probe. When both 4172 module.async_probe and <module>.async_probe are 4173 specified, <module>.async_probe takes precedence for 4174 the specific module. 4175 4176 module.enable_dups_trace 4177 [KNL] When CONFIG_MODULE_DEBUG_AUTOLOAD_DUPS is set, 4178 this means that duplicate request_module() calls will 4179 trigger a WARN_ON() instead of a pr_warn(). Note that 4180 if MODULE_DEBUG_AUTOLOAD_DUPS_TRACE is set, WARN_ON()s 4181 will always be issued and this option does nothing. 4182 module.sig_enforce 4183 [KNL] When CONFIG_MODULE_SIG is set, this means that 4184 modules without (valid) signatures will fail to load. 4185 Note that if CONFIG_MODULE_SIG_FORCE is set, that 4186 is always true, so this option does nothing. 4187 4188 module_blacklist= [KNL] Do not load a comma-separated list of 4189 modules. Useful for debugging problem modules. 4190 4191 mousedev.tap_time= 4192 [MOUSE] Maximum time between finger touching and 4193 leaving touchpad surface for touch to be considered 4194 a tap and be reported as a left button click (for 4195 touchpads working in absolute mode only). 4196 Format: <msecs> 4197 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices 4198 reporting absolute coordinates, such as tablets 4199 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices 4200 reporting absolute coordinates, such as tablets 4201 4202 movablecore= [KNL,X86,PPC,EARLY] 4203 Format: nn[KMGTPE] | nn% 4204 This parameter is the complement to kernelcore=, it 4205 specifies the amount of memory used for migratable 4206 allocations. If both kernelcore and movablecore is 4207 specified, then kernelcore will be at *least* the 4208 specified value but may be more. If movablecore on its 4209 own is specified, the administrator must be careful 4210 that the amount of memory usable for all allocations 4211 is not too small. 4212 4213 movable_node [KNL,EARLY] Boot-time switch to make hotplugable memory 4214 NUMA nodes to be movable. This means that the memory 4215 of such nodes will be usable only for movable 4216 allocations which rules out almost all kernel 4217 allocations. Use with caution! 4218 4219 MTD_Partition= [MTD] 4220 Format: <name>,<region-number>,<size>,<offset> 4221 4222 MTD_Region= [MTD] Format: 4223 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>] 4224 4225 mtdparts= [MTD] 4226 See drivers/mtd/parsers/cmdlinepart.c 4227 4228 mtouchusb.raw_coordinates= 4229 [HW] Make the MicroTouch USB driver use raw coordinates 4230 ('y', default) or cooked coordinates ('n') 4231 4232 mtrr=debug [X86,EARLY] 4233 Enable printing debug information related to MTRR 4234 registers at boot time. 4235 4236 mtrr_chunk_size=nn[KMG,X86,EARLY] 4237 used for mtrr cleanup. It is largest continuous chunk 4238 that could hold holes aka. UC entries. 4239 4240 mtrr_gran_size=nn[KMG,X86,EARLY] 4241 Used for mtrr cleanup. It is granularity of mtrr block. 4242 Default is 1. 4243 Large value could prevent small alignment from 4244 using up MTRRs. 4245 4246 mtrr_spare_reg_nr=n [X86,EARLY] 4247 Format: <integer> 4248 Range: 0,7 : spare reg number 4249 Default : 1 4250 Used for mtrr cleanup. It is spare mtrr entries number. 4251 Set to 2 or more if your graphical card needs more. 4252 4253 multitce=off [PPC] This parameter disables the use of the pSeries 4254 firmware feature for updating multiple TCE entries 4255 at a time. 4256 4257 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card 4258 4259 netdev= [NET] NE2000 ISA network devices parameters 4260 Format: <irq>,<io>,<mem_start>,<mem_end>,<name> 4261 4262 netpoll.carrier_timeout= 4263 [NET] Specifies amount of time (in seconds) that 4264 netpoll should wait for a carrier. By default netpoll 4265 waits 4 seconds. 4266 4267 nf_conntrack.acct= 4268 [NETFILTER] Enable connection tracking flow accounting 4269 0 to disable accounting 4270 1 to enable accounting 4271 Default value is 0. 4272 4273 nfs.cache_getent= 4274 [NFS] sets the pathname to the program which is used 4275 to update the NFS client cache entries. 4276 4277 nfs.cache_getent_timeout= 4278 [NFS] sets the timeout after which an attempt to 4279 update a cache entry is deemed to have failed. 4280 4281 nfs.callback_nr_threads= 4282 [NFSv4] set the total number of threads that the 4283 NFS client will assign to service NFSv4 callback 4284 requests. 4285 4286 nfs.callback_tcpport= 4287 [NFS] set the TCP port on which the NFSv4 callback 4288 channel should listen. 4289 4290 nfs.delay_retrans= 4291 [NFS] specifies the number of times the NFSv4 client 4292 retries the request before returning an EAGAIN error, 4293 after a reply of NFS4ERR_DELAY from the server. 4294 Only applies if the softerr mount option is enabled, 4295 and the specified value is >= 0. 4296 4297 nfs.idmap_cache_timeout= 4298 [NFS] set the maximum lifetime for idmapper cache 4299 entries. 4300 4301 nfs.max_session_cb_slots= 4302 [NFSv4.1] Sets the maximum number of session 4303 slots the client will assign to the callback 4304 channel. This determines the maximum number of 4305 callbacks the client will process in parallel for 4306 a particular server. 4307 4308 nfs.max_session_slots= 4309 [NFSv4.1] Sets the maximum number of session slots 4310 the client will attempt to negotiate with the server. 4311 This limits the number of simultaneous RPC requests 4312 that the client can send to the NFSv4.1 server. 4313 Note that there is little point in setting this 4314 value higher than the max_tcp_slot_table_limit. 4315 4316 nfs.nfs4_disable_idmapping= 4317 [NFSv4] When set to the default of '1', this option 4318 ensures that both the RPC level authentication 4319 scheme and the NFS level operations agree to use 4320 numeric uids/gids if the mount is using the 4321 'sec=sys' security flavour. In effect it is 4322 disabling idmapping, which can make migration from 4323 legacy NFSv2/v3 systems to NFSv4 easier. 4324 Servers that do not support this mode of operation 4325 will be autodetected by the client, and it will fall 4326 back to using the idmapper. 4327 To turn off this behaviour, set the value to '0'. 4328 4329 nfs.nfs4_unique_id= 4330 [NFS4] Specify an additional fixed unique ident- 4331 ification string that NFSv4 clients can insert into 4332 their nfs_client_id4 string. This is typically a 4333 UUID that is generated at system install time. 4334 4335 nfs.recover_lost_locks= 4336 [NFSv4] Attempt to recover locks that were lost due 4337 to a lease timeout on the server. Please note that 4338 doing this risks data corruption, since there are 4339 no guarantees that the file will remain unchanged 4340 after the locks are lost. 4341 If you want to enable the kernel legacy behaviour of 4342 attempting to recover these locks, then set this 4343 parameter to '1'. 4344 The default parameter value of '0' causes the kernel 4345 not to attempt recovery of lost locks. 4346 4347 nfs.send_implementation_id= 4348 [NFSv4.1] Send client implementation identification 4349 information in exchange_id requests. 4350 If zero, no implementation identification information 4351 will be sent. 4352 The default is to send the implementation identification 4353 information. 4354 4355 nfs4.layoutstats_timer= 4356 [NFSv4.2] Change the rate at which the kernel sends 4357 layoutstats to the pNFS metadata server. 4358 4359 Setting this to value to 0 causes the kernel to use 4360 whatever value is the default set by the layout 4361 driver. A non-zero value sets the minimum interval 4362 in seconds between layoutstats transmissions. 4363 4364 nfsd.inter_copy_offload_enable= 4365 [NFSv4.2] When set to 1, the server will support 4366 server-to-server copies for which this server is 4367 the destination of the copy. 4368 4369 nfsd.nfs4_disable_idmapping= 4370 [NFSv4] When set to the default of '1', the NFSv4 4371 server will return only numeric uids and gids to 4372 clients using auth_sys, and will accept numeric uids 4373 and gids from such clients. This is intended to ease 4374 migration from NFSv2/v3. 4375 4376 nfsd.nfsd4_ssc_umount_timeout= 4377 [NFSv4.2] When used as the destination of a 4378 server-to-server copy, knfsd temporarily mounts 4379 the source server. It caches the mount in case 4380 it will be needed again, and discards it if not 4381 used for the number of milliseconds specified by 4382 this parameter. 4383 4384 nfsaddrs= [NFS] Deprecated. Use ip= instead. 4385 See Documentation/admin-guide/nfs/nfsroot.rst. 4386 4387 nfsroot= [NFS] nfs root filesystem for disk-less boxes. 4388 See Documentation/admin-guide/nfs/nfsroot.rst. 4389 4390 nfsrootdebug [NFS] enable nfsroot debugging messages. 4391 See Documentation/admin-guide/nfs/nfsroot.rst. 4392 4393 nmi_backtrace.backtrace_idle [KNL] 4394 Dump stacks even of idle CPUs in response to an 4395 NMI stack-backtrace request. 4396 4397 nmi_debug= [KNL,SH] Specify one or more actions to take 4398 when a NMI is triggered. 4399 Format: [state][,regs][,debounce][,die] 4400 4401 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels 4402 Format: [panic,][nopanic,][rNNN,][num] 4403 Valid num: 0 or 1 4404 0 - turn hardlockup detector in nmi_watchdog off 4405 1 - turn hardlockup detector in nmi_watchdog on 4406 rNNN - configure the watchdog with raw perf event 0xNNN 4407 4408 When panic is specified, panic when an NMI watchdog 4409 timeout occurs (or 'nopanic' to not panic on an NMI 4410 watchdog, if CONFIG_BOOTPARAM_HARDLOCKUP_PANIC is set) 4411 To disable both hard and soft lockup detectors, 4412 please see 'nowatchdog'. 4413 This is useful when you use a panic=... timeout and 4414 need the box quickly up again. 4415 4416 These settings can be accessed at runtime via 4417 the nmi_watchdog and hardlockup_panic sysctls. 4418 4419 no4lvl [RISCV,EARLY] Disable 4-level and 5-level paging modes. 4420 Forces kernel to use 3-level paging instead. 4421 4422 no5lvl [X86-64,RISCV,EARLY] Disable 5-level paging mode. Forces 4423 kernel to use 4-level paging instead. 4424 4425 noalign [KNL,ARM] 4426 4427 noapic [SMP,APIC,EARLY] Tells the kernel to not make use of any 4428 IOAPICs that may be present in the system. 4429 4430 noapictimer [APIC,X86] Don't set up the APIC timer 4431 4432 noautogroup Disable scheduler automatic task group creation. 4433 4434 nocache [ARM,EARLY] 4435 4436 no_console_suspend 4437 [HW] Never suspend the console 4438 Disable suspending of consoles during suspend and 4439 hibernate operations. Once disabled, debugging 4440 messages can reach various consoles while the rest 4441 of the system is being put to sleep (ie, while 4442 debugging driver suspend/resume hooks). This may 4443 not work reliably with all consoles, but is known 4444 to work with serial and VGA consoles. 4445 To facilitate more flexible debugging, we also add 4446 console_suspend, a printk module parameter to control 4447 it. Users could use console_suspend (usually 4448 /sys/module/printk/parameters/console_suspend) to 4449 turn on/off it dynamically. 4450 4451 no_debug_objects 4452 [KNL,EARLY] Disable object debugging 4453 4454 nodsp [SH] Disable hardware DSP at boot time. 4455 4456 noefi [EFI,EARLY] Disable EFI runtime services support. 4457 4458 no_entry_flush [PPC,EARLY] Don't flush the L1-D cache when entering the kernel. 4459 4460 noexec32 [X86-64] 4461 This affects only 32-bit executables. 4462 noexec32=on: enable non-executable mappings (default) 4463 read doesn't imply executable mappings 4464 noexec32=off: disable non-executable mappings 4465 read implies executable mappings 4466 4467 no_file_caps Tells the kernel not to honor file capabilities. The 4468 only way then for a file to be executed with privilege 4469 is to be setuid root or executed by root. 4470 4471 nofpu [MIPS,SH] Disable hardware FPU at boot time. 4472 4473 nofsgsbase [X86] Disables FSGSBASE instructions. 4474 4475 nofxsr [BUGS=X86-32] Disables x86 floating point extended 4476 register save and restore. The kernel will only save 4477 legacy floating-point registers on task switch. 4478 4479 nogbpages [X86] Do not use GB pages for kernel direct mappings. 4480 4481 no_hash_pointers 4482 [KNL,EARLY] 4483 Alias for "hash_pointers=never". 4484 4485 nohibernate [HIBERNATION] Disable hibernation and resume. 4486 4487 nohlt [ARM,ARM64,MICROBLAZE,MIPS,PPC,RISCV,SH] Forces the kernel to 4488 busy wait in do_idle() and not use the arch_cpu_idle() 4489 implementation; requires CONFIG_GENERIC_IDLE_POLL_SETUP 4490 to be effective. This is useful on platforms where the 4491 sleep(SH) or wfi(ARM,ARM64) instructions do not work 4492 correctly or when doing power measurements to evaluate 4493 the impact of the sleep instructions. This is also 4494 useful when using JTAG debugger. 4495 4496 nohpet [X86] Don't use the HPET timer. 4497 4498 nohugeiomap [KNL,X86,PPC,ARM64,EARLY] Disable kernel huge I/O mappings. 4499 4500 nohugevmalloc [KNL,X86,PPC,ARM64,EARLY] Disable kernel huge vmalloc mappings. 4501 4502 nohz= [KNL] Boottime enable/disable dynamic ticks 4503 Valid arguments: on, off 4504 Default: on 4505 4506 nohz_full= [KNL,BOOT,SMP,ISOL] 4507 The argument is a cpu list, as described above. 4508 In kernels built with CONFIG_NO_HZ_FULL=y, set 4509 the specified list of CPUs whose tick will be stopped 4510 whenever possible. The boot CPU will be forced outside 4511 the range to maintain the timekeeping. Any CPUs 4512 in this list will have their RCU callbacks offloaded, 4513 just as if they had also been called out in the 4514 rcu_nocbs= boot parameter. 4515 4516 Note that this argument takes precedence over 4517 the CONFIG_RCU_NOCB_CPU_DEFAULT_ALL option. 4518 4519 noinitrd [Deprecated,RAM] Tells the kernel not to load any configured 4520 initial RAM disk. Currently this parameter applies to 4521 initrd only, not to initramfs. But it applies to both 4522 in EFI mode. 4523 4524 nointremap [X86-64,Intel-IOMMU,EARLY] Do not enable interrupt 4525 remapping. 4526 [Deprecated - use intremap=off] 4527 4528 noinvpcid [X86,EARLY] Disable the INVPCID cpu feature. 4529 4530 noiotrap [SH] Disables trapped I/O port accesses. 4531 4532 noirqdebug [X86-32] Disables the code which attempts to detect and 4533 disable unhandled interrupt sources. 4534 4535 noisapnp [ISAPNP] Disables ISA PnP code. 4536 4537 nokaslr [KNL,EARLY] 4538 When CONFIG_RANDOMIZE_BASE is set, this disables 4539 kernel and module base offset ASLR (Address Space 4540 Layout Randomization). 4541 4542 no-kvmapf [X86,KVM,EARLY] Disable paravirtualized asynchronous page 4543 fault handling. 4544 4545 no-kvmclock [X86,KVM,EARLY] Disable paravirtualized KVM clock driver 4546 4547 nolapic [X86-32,APIC,EARLY] Do not enable or use the local APIC. 4548 4549 nolapic_timer [X86-32,APIC,EARLY] Do not use the local APIC timer. 4550 4551 nomce [X86-32] Disable Machine Check Exception 4552 4553 nomfgpt [X86-32] Disable Multi-Function General Purpose 4554 Timer usage (for AMD Geode machines). 4555 4556 nomodeset Disable kernel modesetting. Most systems' firmware 4557 sets up a display mode and provides framebuffer memory 4558 for output. With nomodeset, DRM and fbdev drivers will 4559 not load if they could possibly displace the pre- 4560 initialized output. Only the system framebuffer will 4561 be available for use. The respective drivers will not 4562 perform display-mode changes or accelerated rendering. 4563 4564 Useful as error fallback, or for testing and debugging. 4565 4566 nomodule Disable module load 4567 4568 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to 4569 shutdown the other cpus. Instead use the REBOOT_VECTOR 4570 irq. 4571 4572 nopat [X86,EARLY] Disable PAT (page attribute table extension of 4573 pagetables) support. 4574 4575 nopcid [X86-64,EARLY] Disable the PCID cpu feature. 4576 4577 nopku [X86] Disable Memory Protection Keys CPU feature found 4578 in some Intel CPUs. 4579 4580 nopti [X86-64,EARLY] 4581 Equivalent to pti=off 4582 4583 nopv= [X86,XEN,KVM,HYPER_V,VMWARE,EARLY] 4584 Disables the PV optimizations forcing the guest to run 4585 as generic guest with no PV drivers. Currently support 4586 XEN HVM, KVM, HYPER_V and VMWARE guest. 4587 4588 nopvspin [X86,XEN,KVM,EARLY] 4589 Disables the qspinlock slow path using PV optimizations 4590 which allow the hypervisor to 'idle' the guest on lock 4591 contention. 4592 4593 norandmaps Don't use address space randomization. Equivalent to 4594 echo 0 > /proc/sys/kernel/randomize_va_space 4595 4596 noreplace-smp [X86-32,SMP] Don't replace SMP instructions 4597 with UP alternatives 4598 4599 noresume [SWSUSP] Disables resume and restores original swap 4600 space. 4601 4602 no-scroll [VGA] Disables scrollback. 4603 This is required for the Braillex ib80-piezo Braille 4604 reader made by F.H. Papenmeier (Germany). 4605 4606 nosgx [X86-64,SGX,EARLY] Disables Intel SGX kernel support. 4607 4608 nosmap [PPC,EARLY] 4609 Disable SMAP (Supervisor Mode Access Prevention) 4610 even if it is supported by processor. 4611 4612 nosmep [PPC64s,EARLY] 4613 Disable SMEP (Supervisor Mode Execution Prevention) 4614 even if it is supported by processor. 4615 4616 nosmp [SMP,EARLY] Tells an SMP kernel to act as a UP kernel, 4617 and disable the IO APIC. legacy for "maxcpus=0". 4618 4619 nosmt [KNL,MIPS,PPC,EARLY] Disable symmetric multithreading (SMT). 4620 Equivalent to smt=1. 4621 4622 [KNL,LOONGARCH,X86,ARM64,PPC,S390] Disable symmetric multithreading (SMT). 4623 nosmt=force: Force disable SMT, cannot be undone 4624 via the sysfs control file. 4625 4626 nosoftlockup [KNL] Disable the soft-lockup detector. 4627 4628 nospec_store_bypass_disable 4629 [HW,EARLY] Disable all mitigations for the Speculative 4630 Store Bypass vulnerability 4631 4632 nospectre_bhb [ARM64,EARLY] Disable all mitigations for Spectre-BHB (branch 4633 history injection) vulnerability. System may allow data leaks 4634 with this option. 4635 4636 nospectre_v1 [X86,PPC,EARLY] Disable mitigations for Spectre Variant 1 4637 (bounds check bypass). With this option data leaks are 4638 possible in the system. 4639 4640 nospectre_v2 [X86,PPC_E500,ARM64,EARLY] Disable all mitigations 4641 for the Spectre variant 2 (indirect branch 4642 prediction) vulnerability. System may allow data 4643 leaks with this option. 4644 4645 no-steal-acc [X86,PV_OPS,ARM64,PPC/PSERIES,RISCV,LOONGARCH,EARLY] 4646 Disable paravirtualized steal time accounting. steal time 4647 is computed, but won't influence scheduler behaviour 4648 4649 nosync [HW,M68K] Disables sync negotiation for all devices. 4650 4651 no_timer_check [X86,APIC] Disables the code which tests for broken 4652 timer IRQ sources, i.e., the IO-APIC timer. This can 4653 work around problems with incorrect timer 4654 initialization on some boards. 4655 4656 no_uaccess_flush 4657 [PPC,EARLY] Don't flush the L1-D cache after accessing user data. 4658 4659 novmcoredd [KNL,KDUMP] 4660 Disable device dump. Device dump allows drivers to 4661 append dump data to vmcore so you can collect driver 4662 specified debug info. Drivers can append the data 4663 without any limit and this data is stored in memory, 4664 so this may cause significant memory stress. Disabling 4665 device dump can help save memory but the driver debug 4666 data will be no longer available. This parameter 4667 is only available when CONFIG_PROC_VMCORE_DEVICE_DUMP 4668 is set. 4669 4670 no-vmw-sched-clock 4671 [X86,PV_OPS,EARLY] Disable paravirtualized VMware 4672 scheduler clock and use the default one. 4673 4674 nowatchdog [KNL] Disable both lockup detectors, i.e. 4675 soft-lockup and NMI watchdog (hard-lockup). 4676 4677 nowb [ARM,EARLY] 4678 4679 nox2apic [X86-64,APIC,EARLY] Do not enable x2APIC mode. 4680 4681 NOTE: this parameter will be ignored on systems with the 4682 LEGACY_XAPIC_DISABLED bit set in the 4683 IA32_XAPIC_DISABLE_STATUS MSR. 4684 4685 noxsave [BUGS=X86] Disables x86 extended register state save 4686 and restore using xsave. The kernel will fallback to 4687 enabling legacy floating-point and sse state. 4688 4689 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended 4690 register states. The kernel will fall back to use 4691 xsave to save the states. By using this parameter, 4692 performance of saving the states is degraded because 4693 xsave doesn't support modified optimization while 4694 xsaveopt supports it on xsaveopt enabled systems. 4695 4696 noxsaves [X86] Disables xsaves and xrstors used in saving and 4697 restoring x86 extended register state in compacted 4698 form of xsave area. The kernel will fall back to use 4699 xsaveopt and xrstor to save and restore the states 4700 in standard form of xsave area. By using this 4701 parameter, xsave area per process might occupy more 4702 memory on xsaves enabled systems. 4703 4704 nr_cpus= [SMP,EARLY] Maximum number of processors that an SMP kernel 4705 could support. nr_cpus=n : n >= 1 limits the kernel to 4706 support 'n' processors. It could be larger than the 4707 number of already plugged CPU during bootup, later in 4708 runtime you can physically add extra cpu until it reaches 4709 n. So during boot up some boot time memory for per-cpu 4710 variables need be pre-allocated for later physical cpu 4711 hot plugging. 4712 4713 nr_uarts= [SERIAL] maximum number of UARTs to be registered. 4714 4715 numa=off [KNL, ARM64, PPC, RISCV, SPARC, X86, EARLY] 4716 Disable NUMA, Only set up a single NUMA node 4717 spanning all memory. 4718 4719 numa=fake=<size>[MG] 4720 [KNL, ARM64, RISCV, X86, EARLY] 4721 If given as a memory unit, fills all system RAM with 4722 nodes of size interleaved over physical nodes. 4723 4724 numa=fake=<N> 4725 [KNL, ARM64, RISCV, X86, EARLY] 4726 If given as an integer, fills all system RAM with N 4727 fake nodes interleaved over physical nodes. 4728 4729 numa=fake=<N>U 4730 [KNL, ARM64, RISCV, X86, EARLY] 4731 If given as an integer followed by 'U', it will 4732 divide each physical node into N emulated nodes. 4733 4734 numa=noacpi [X86] Don't parse the SRAT table for NUMA setup 4735 4736 numa=nohmat [X86] Don't parse the HMAT table for NUMA setup, or 4737 soft-reserved memory partitioning. 4738 4739 numa_balancing= [KNL,ARM64,PPC,RISCV,S390,X86] Enable or disable automatic 4740 NUMA balancing. 4741 Allowed values are enable and disable 4742 4743 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA. 4744 'node', 'default' can be specified 4745 This can be set from sysctl after boot. 4746 See Documentation/admin-guide/sysctl/vm.rst for details. 4747 4748 nvme.quirks= [NVME] A list of quirk entries to augment the built-in 4749 nvme quirk list. List entries are separated by a 4750 '-' character. 4751 Each entry has the form VendorID:ProductID:quirk_names. 4752 The IDs are 4-digits hex numbers and quirk_names is a 4753 list of quirk names separated by commas. A quirk name 4754 can be prefixed by '^', meaning that the specified 4755 quirk must be disabled. 4756 4757 Example: 4758 nvme.quirks=7710:2267:bogus_nid,^identify_cns-9900:7711:broken_msi 4759 4760 ohci1394_dma=early [HW,EARLY] enable debugging via the ohci1394 driver. 4761 See Documentation/core-api/debugging-via-ohci1394.rst for more 4762 info. 4763 4764 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands 4765 Rather than timing out after 20 ms if an EC 4766 command is not properly ACKed, override the length 4767 of the timeout. We have interrupts disabled while 4768 waiting for the ACK, so if this is set too high 4769 interrupts *may* be lost! 4770 4771 omap_mux= [OMAP] Override bootloader pin multiplexing. 4772 Format: <mux_mode0.mode_name=value>... 4773 For example, to override I2C bus2: 4774 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100 4775 4776 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration 4777 4778 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock] 4779 4780 boundary - index of last SLC block on Flex-OneNAND. 4781 The remaining blocks are configured as MLC blocks. 4782 lock - Configure if Flex-OneNAND boundary should be locked. 4783 Once locked, the boundary cannot be changed. 4784 1 indicates lock status, 0 indicates unlock status. 4785 4786 oops=panic [KNL,EARLY] 4787 Always panic on oopses. Default is to just kill the 4788 process, but there is a small probability of 4789 deadlocking the machine. 4790 This will also cause panics on machine check exceptions. 4791 Useful together with panic=30 to trigger a reboot. 4792 4793 page_alloc.shuffle= 4794 [KNL] Boolean flag to control whether the page allocator 4795 should randomize its free lists. This parameter can be 4796 used to enable/disable page randomization. The state of 4797 the flag can be read from sysfs at: 4798 /sys/module/page_alloc/parameters/shuffle. 4799 This parameter is only available if CONFIG_SHUFFLE_PAGE_ALLOCATOR=y. 4800 4801 page_owner= [KNL,EARLY] Boot-time page_owner enabling option. 4802 Storage of the information about who allocated 4803 each page is disabled in default. With this switch, 4804 we can turn it on. 4805 on: enable the feature 4806 4807 page_poison= [KNL,EARLY] Boot-time parameter changing the state of 4808 poisoning on the buddy allocator, available with 4809 CONFIG_PAGE_POISONING=y. 4810 off: turn off poisoning (default) 4811 on: turn on poisoning 4812 4813 page_reporting.page_reporting_order= 4814 [KNL] Minimal page reporting order 4815 Format: <integer> 4816 Adjust the minimal page reporting order. The page 4817 reporting is disabled when it exceeds MAX_PAGE_ORDER. 4818 4819 page_reporting.page_reporting_delay_ms= 4820 [KNL] Free page reporting delay in milliseconds 4821 Format: <unsigned integer> 4822 Adjust the delay in milliseconds between free page 4823 reporting intervals. Default is 2000 (2 seconds). 4824 4825 panic= [KNL] Kernel behaviour on panic: delay <timeout> 4826 timeout > 0: seconds before rebooting 4827 timeout = 0: wait forever 4828 timeout < 0: reboot immediately 4829 Format: <timeout> 4830 4831 panic_on_taint= [KNL,EARLY] 4832 Bitmask for conditionally calling panic() in add_taint() 4833 Format: <hex>[,nousertaint] 4834 Hexadecimal bitmask representing the set of TAINT flags 4835 that will cause the kernel to panic when add_taint() is 4836 called with any of the flags in this set. 4837 The optional switch "nousertaint" can be utilized to 4838 prevent userspace forced crashes by writing to sysctl 4839 /proc/sys/kernel/tainted any flagset matching with the 4840 bitmask set on panic_on_taint. 4841 See Documentation/admin-guide/tainted-kernels.rst for 4842 extra details on the taint flags that users can pick 4843 to compose the bitmask to assign to panic_on_taint. 4844 4845 panic_on_warn=1 panic() instead of WARN(). Useful to cause kdump 4846 on a WARN(). 4847 4848 panic_force_cpu= 4849 [KNL,SMP] Force panic handling to execute on a specific CPU. 4850 Format: <cpu number> 4851 Some platforms require panic handling to occur on a 4852 specific CPU for the crash kernel to function correctly. 4853 This can be due to firmware limitations, interrupt routing 4854 constraints, or platform-specific requirements where only 4855 a particular CPU can safely enter the crash kernel. 4856 When set, panic() will redirect execution to the specified 4857 CPU before proceeding with the normal panic and kexec flow. 4858 If the target CPU is offline or unavailable, panic proceeds 4859 on the current CPU. 4860 This option should only be used for systems with the above 4861 constraints as it might cause the panic operation to be less reliable. 4862 4863 panic_print= Bitmask for printing system info when panic happens. 4864 User can chose combination of the following bits: 4865 bit 0: print all tasks info 4866 bit 1: print system memory info 4867 bit 2: print timer info 4868 bit 3: print locks info if CONFIG_LOCKDEP is on 4869 bit 4: print ftrace buffer 4870 bit 5: replay all kernel messages on consoles at the end of panic 4871 bit 6: print all CPUs backtrace (if available in the arch) 4872 bit 7: print only tasks in uninterruptible (blocked) state 4873 *Be aware* that this option may print a _lot_ of lines, 4874 so there are risks of losing older messages in the log. 4875 Use this option carefully, maybe worth to setup a 4876 bigger log buffer with "log_buf_len" along with this. 4877 4878 panic_sys_info= A comma separated list of extra information to be dumped 4879 on panic. 4880 Format: val[,val...] 4881 Where @val can be any of the following: 4882 4883 tasks: print all tasks info 4884 mem: print system memory info 4885 timers: print timers info 4886 locks: print locks info if CONFIG_LOCKDEP is on 4887 ftrace: print ftrace buffer 4888 all_bt: print all CPUs backtrace (if available in the arch) 4889 blocked_tasks: print only tasks in uninterruptible (blocked) state 4890 4891 This is a human readable alternative to the 'panic_print' option. 4892 4893 panic_console_replay 4894 When panic happens, replay all kernel messages on 4895 consoles at the end of panic. 4896 4897 parkbd.port= [HW] Parallel port number the keyboard adapter is 4898 connected to, default is 0. 4899 Format: <parport#> 4900 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation, 4901 0 for XT, 1 for AT (default is AT). 4902 Format: <mode> 4903 4904 parport= [HW,PPT] Specify parallel ports. 0 disables. 4905 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] } 4906 Use 'auto' to force the driver to use any 4907 IRQ/DMA settings detected (the default is to 4908 ignore detected IRQ/DMA settings because of 4909 possible conflicts). You can specify the base 4910 address, IRQ, and DMA settings; IRQ and DMA 4911 should be numbers, or 'auto' (for using detected 4912 settings on that particular port), or 'nofifo' 4913 (to avoid using a FIFO even if it is detected). 4914 Parallel ports are assigned in the order they 4915 are specified on the command line, starting 4916 with parport0. 4917 4918 parport_init_mode= [HW,PPT] 4919 Configure VIA parallel port to operate in 4920 a specific mode. This is necessary on Pegasos 4921 computer where firmware has no options for setting 4922 up parallel port mode and sets it to spp. 4923 Currently this function knows 686a and 8231 chips. 4924 Format: [spp|ps2|epp|ecp|ecpepp] 4925 4926 pata_legacy.all= [HW,LIBATA] 4927 Format: <int> 4928 Set to non-zero to probe primary and secondary ISA 4929 port ranges on PCI systems where no PCI PATA device 4930 has been found at either range. Disabled by default. 4931 4932 pata_legacy.autospeed= [HW,LIBATA] 4933 Format: <int> 4934 Set to non-zero if a chip is present that snoops speed 4935 changes. Disabled by default. 4936 4937 pata_legacy.ht6560a= [HW,LIBATA] 4938 Format: <int> 4939 Set to 1, 2, or 3 for HT 6560A on the primary channel, 4940 the secondary channel, or both channels respectively. 4941 Disabled by default. 4942 4943 pata_legacy.ht6560b= [HW,LIBATA] 4944 Format: <int> 4945 Set to 1, 2, or 3 for HT 6560B on the primary channel, 4946 the secondary channel, or both channels respectively. 4947 Disabled by default. 4948 4949 pata_legacy.iordy_mask= [HW,LIBATA] 4950 Format: <int> 4951 IORDY enable mask. Set individual bits to allow IORDY 4952 for the respective channel. Bit 0 is for the first 4953 legacy channel handled by this driver, bit 1 is for 4954 the second channel, and so on. The sequence will often 4955 correspond to the primary legacy channel, the secondary 4956 legacy channel, and so on, but the handling of a PCI 4957 bus and the use of other driver options may interfere 4958 with the sequence. By default IORDY is allowed across 4959 all channels. 4960 4961 pata_legacy.opti82c46x= [HW,LIBATA] 4962 Format: <int> 4963 Set to 1, 2, or 3 for Opti 82c611A on the primary 4964 channel, the secondary channel, or both channels 4965 respectively. Disabled by default. 4966 4967 pata_legacy.opti82c611a= [HW,LIBATA] 4968 Format: <int> 4969 Set to 1, 2, or 3 for Opti 82c465MV on the primary 4970 channel, the secondary channel, or both channels 4971 respectively. Disabled by default. 4972 4973 pata_legacy.pio_mask= [HW,LIBATA] 4974 Format: <int> 4975 PIO mode mask for autospeed devices. Set individual 4976 bits to allow the use of the respective PIO modes. 4977 Bit 0 is for mode 0, bit 1 is for mode 1, and so on. 4978 All modes allowed by default. 4979 4980 pata_legacy.probe_all= [HW,LIBATA] 4981 Format: <int> 4982 Set to non-zero to probe tertiary and further ISA 4983 port ranges on PCI systems. Disabled by default. 4984 4985 pata_legacy.probe_mask= [HW,LIBATA] 4986 Format: <int> 4987 Probe mask for legacy ISA PATA ports. Depending on 4988 platform configuration and the use of other driver 4989 options up to 6 legacy ports are supported: 0x1f0, 4990 0x170, 0x1e8, 0x168, 0x1e0, 0x160, however probing 4991 of individual ports can be disabled by setting the 4992 corresponding bits in the mask to 1. Bit 0 is for 4993 the first port in the list above (0x1f0), and so on. 4994 By default all supported ports are probed. 4995 4996 pata_legacy.qdi= [HW,LIBATA] 4997 Format: <int> 4998 Set to non-zero to probe QDI controllers. By default 4999 set to 1 if CONFIG_PATA_QDI_MODULE, 0 otherwise. 5000 5001 pata_legacy.winbond= [HW,LIBATA] 5002 Format: <int> 5003 Set to non-zero to probe Winbond controllers. Use 5004 the standard I/O port (0x130) if 1, otherwise the 5005 value given is the I/O port to use (typically 0x1b0). 5006 By default set to 1 if CONFIG_PATA_WINBOND_VLB_MODULE, 5007 0 otherwise. 5008 5009 pata_platform.pio_mask= [HW,LIBATA] 5010 Format: <int> 5011 Supported PIO mode mask. Set individual bits to allow 5012 the use of the respective PIO modes. Bit 0 is for 5013 mode 0, bit 1 is for mode 1, and so on. Mode 0 only 5014 allowed by default. 5015 5016 pause_on_oops=<int> 5017 Halt all CPUs after the first oops has been printed for 5018 the specified number of seconds. This is to be used if 5019 your oopses keep scrolling off the screen. 5020 5021 pci=option[,option...] [PCI,EARLY] various PCI subsystem options. 5022 5023 Some options herein operate on a specific device 5024 or a set of devices (<pci_dev>). These are 5025 specified in one of the following formats: 5026 5027 [<domain>:]<bus>:<dev>.<func>[/<dev>.<func>]* 5028 pci:<vendor>:<device>[:<subvendor>:<subdevice>] 5029 5030 Note: the first format specifies a PCI 5031 bus/device/function address which may change 5032 if new hardware is inserted, if motherboard 5033 firmware changes, or due to changes caused 5034 by other kernel parameters. If the 5035 domain is left unspecified, it is 5036 taken to be zero. Optionally, a path 5037 to a device through multiple device/function 5038 addresses can be specified after the base 5039 address (this is more robust against 5040 renumbering issues). The second format 5041 selects devices using IDs from the 5042 configuration space which may match multiple 5043 devices in the system. 5044 5045 earlydump dump PCI config space before the kernel 5046 changes anything 5047 off [X86] don't probe for the PCI bus 5048 bios [X86-32] force use of PCI BIOS, don't access 5049 the hardware directly. Use this if your machine 5050 has a non-standard PCI host bridge. 5051 nobios [X86-32] disallow use of PCI BIOS, only direct 5052 hardware access methods are allowed. Use this 5053 if you experience crashes upon bootup and you 5054 suspect they are caused by the BIOS. 5055 conf1 [X86] Force use of PCI Configuration Access 5056 Mechanism 1 (config address in IO port 0xCF8, 5057 data in IO port 0xCFC, both 32-bit). 5058 conf2 [X86] Force use of PCI Configuration Access 5059 Mechanism 2 (IO port 0xCF8 is an 8-bit port for 5060 the function, IO port 0xCFA, also 8-bit, sets 5061 bus number. The config space is then accessed 5062 through ports 0xC000-0xCFFF). 5063 See http://wiki.osdev.org/PCI for more info 5064 on the configuration access mechanisms. 5065 noaer [PCIE] If the PCIEAER kernel config parameter is 5066 enabled, this kernel boot option can be used to 5067 disable the use of PCIE advanced error reporting. 5068 nodomains [PCI] Disable support for multiple PCI 5069 root domains (aka PCI segments, in ACPI-speak). 5070 nommconf [X86] Disable use of MMCONFIG for PCI 5071 Configuration 5072 check_enable_amd_mmconf [X86] check for and enable 5073 properly configured MMIO access to PCI 5074 config space on AMD family 10h CPU 5075 nomsi [MSI] If the PCI_MSI kernel config parameter is 5076 enabled, this kernel boot option can be used to 5077 disable the use of MSI interrupts system-wide. 5078 noioapicquirk [APIC] Disable all boot interrupt quirks. 5079 Safety option to keep boot IRQs enabled. This 5080 should never be necessary. 5081 ioapicreroute [APIC] Enable rerouting of boot IRQs to the 5082 primary IO-APIC for bridges that cannot disable 5083 boot IRQs. This fixes a source of spurious IRQs 5084 when the system masks IRQs. 5085 noioapicreroute [APIC] Disable workaround that uses the 5086 boot IRQ equivalent of an IRQ that connects to 5087 a chipset where boot IRQs cannot be disabled. 5088 The opposite of ioapicreroute. 5089 biosirq [X86-32] Use PCI BIOS calls to get the interrupt 5090 routing table. These calls are known to be buggy 5091 on several machines and they hang the machine 5092 when used, but on other computers it's the only 5093 way to get the interrupt routing table. Try 5094 this option if the kernel is unable to allocate 5095 IRQs or discover secondary PCI buses on your 5096 motherboard. 5097 rom [X86] Assign address space to expansion ROMs. 5098 Use with caution as certain devices share 5099 address decoders between ROMs and other 5100 resources. 5101 norom [X86] Do not assign address space to 5102 expansion ROMs that do not already have 5103 BIOS assigned address ranges. 5104 nobar [X86] Do not assign address space to the 5105 BARs that weren't assigned by the BIOS. 5106 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be 5107 assigned automatically to PCI devices. You can 5108 make the kernel exclude IRQs of your ISA cards 5109 this way. 5110 pirqaddr=0xAAAAA [X86] Specify the physical address 5111 of the PIRQ table (normally generated 5112 by the BIOS) if it is outside the 5113 F0000h-100000h range. 5114 lastbus=N [X86] Scan all buses thru bus #N. Can be 5115 useful if the kernel is unable to find your 5116 secondary buses and you want to tell it 5117 explicitly which ones they are. 5118 assign-busses [X86] Always assign all PCI bus 5119 numbers ourselves, overriding 5120 whatever the firmware may have done. 5121 usepirqmask [X86] Honor the possible IRQ mask stored 5122 in the BIOS $PIR table. This is needed on 5123 some systems with broken BIOSes, notably 5124 some HP Pavilion N5400 and Omnibook XE3 5125 notebooks. This will have no effect if ACPI 5126 IRQ routing is enabled. 5127 noacpi [X86] Do not use ACPI for IRQ routing 5128 or for PCI scanning. 5129 use_crs [X86] Use PCI host bridge window information 5130 from ACPI. On BIOSes from 2008 or later, this 5131 is enabled by default. If you need to use this, 5132 please report a bug. 5133 nocrs [X86] Ignore PCI host bridge windows from ACPI. 5134 If you need to use this, please report a bug. 5135 use_e820 [X86] Use E820 reservations to exclude parts of 5136 PCI host bridge windows. This is a workaround 5137 for BIOS defects in host bridge _CRS methods. 5138 If you need to use this, please report a bug to 5139 <linux-pci@vger.kernel.org>. 5140 no_e820 [X86] Ignore E820 reservations for PCI host 5141 bridge windows. This is the default on modern 5142 hardware. If you need to use this, please report 5143 a bug to <linux-pci@vger.kernel.org>. 5144 routeirq Do IRQ routing for all PCI devices. 5145 This is normally done in pci_enable_device(), 5146 so this option is a temporary workaround 5147 for broken drivers that don't call it. 5148 skip_isa_align [X86] do not align io start addr, so can 5149 handle more pci cards 5150 noearly [X86] Don't do any early type 1 scanning. 5151 This might help on some broken boards which 5152 machine check when some devices' config space 5153 is read. But various workarounds are disabled 5154 and some IOMMU drivers will not work. 5155 bfsort Sort PCI devices into breadth-first order. 5156 This sorting is done to get a device 5157 order compatible with older (<= 2.4) kernels. 5158 nobfsort Don't sort PCI devices into breadth-first order. 5159 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size) 5160 tuning and use the BIOS-configured MPS defaults. 5161 pcie_bus_safe Set every device's MPS to the largest value 5162 supported by all devices below the root complex. 5163 pcie_bus_perf Set device MPS to the largest allowable MPS 5164 based on its parent bus. Also set MRRS (Max 5165 Read Request Size) to the largest supported 5166 value (no larger than the MPS that the device 5167 or bus can support) for best performance. 5168 pcie_bus_peer2peer Set every device's MPS to 128B, which 5169 every device is guaranteed to support. This 5170 configuration allows peer-to-peer DMA between 5171 any pair of devices, possibly at the cost of 5172 reduced performance. This also guarantees 5173 that hot-added devices will work. 5174 cbiosize=nn[KMG] The fixed amount of bus space which is 5175 reserved for the CardBus bridge's IO window. 5176 The default value is 256 bytes. 5177 cbmemsize=nn[KMG] The fixed amount of bus space which is 5178 reserved for the CardBus bridge's memory 5179 window. The default value is 64 megabytes. 5180 resource_alignment= 5181 Format: 5182 [<order of align>@]<pci_dev>[; ...] 5183 Specifies alignment and device to reassign 5184 aligned memory resources. How to 5185 specify the device is described above. 5186 If <order of align> is not specified, 5187 PAGE_SIZE is used as alignment. 5188 A PCI-PCI bridge can be specified if resource 5189 windows need to be expanded. 5190 To specify the alignment for several 5191 instances of a device, the PCI vendor, 5192 device, subvendor, and subdevice may be 5193 specified, e.g., 12@pci:8086:9c22:103c:198f 5194 for 4096-byte alignment. 5195 ecrc= Enable/disable PCIe ECRC (transaction layer 5196 end-to-end CRC checking). Only effective if 5197 OS has native AER control (either granted by 5198 ACPI _OSC or forced via "pcie_ports=native") 5199 bios: Use BIOS/firmware settings. This is the 5200 the default. 5201 off: Turn ECRC off 5202 on: Turn ECRC on. 5203 hpiosize=nn[KMG] The fixed amount of bus space which is 5204 reserved for hotplug bridge's IO window. 5205 Default size is 256 bytes. 5206 hpmmiosize=nn[KMG] The fixed amount of bus space which is 5207 reserved for hotplug bridge's MMIO window. 5208 Default size is 2 megabytes. 5209 hpmmioprefsize=nn[KMG] The fixed amount of bus space which is 5210 reserved for hotplug bridge's MMIO_PREF window. 5211 Default size is 2 megabytes. 5212 hpmemsize=nn[KMG] The fixed amount of bus space which is 5213 reserved for hotplug bridge's MMIO and 5214 MMIO_PREF window. 5215 Default size is 2 megabytes. 5216 hpbussize=nn The minimum amount of additional bus numbers 5217 reserved for buses below a hotplug bridge. 5218 Default is 1. 5219 realloc= Enable/disable reallocating PCI bridge resources 5220 if allocations done by BIOS are too small to 5221 accommodate resources required by all child 5222 devices. 5223 off: Turn realloc off 5224 on: Turn realloc on 5225 realloc same as realloc=on 5226 noari do not use PCIe ARI. 5227 noats [PCIE, Intel-IOMMU, AMD-IOMMU] 5228 do not use PCIe ATS (and IOMMU device IOTLB). 5229 pcie_scan_all Scan all possible PCIe devices. Otherwise we 5230 only look for one device below a PCIe downstream 5231 port. 5232 big_root_window Try to add a big 64bit memory window to the PCIe 5233 root complex on AMD CPUs. Some GFX hardware 5234 can resize a BAR to allow access to all VRAM. 5235 Adding the window is slightly risky (it may 5236 conflict with unreported devices), so this 5237 taints the kernel. 5238 disable_acs_redir=<pci_dev>[; ...] 5239 Specify one or more PCI devices (in the format 5240 specified above) separated by semicolons. 5241 Each device specified will have the PCI ACS 5242 redirect capabilities forced off which will 5243 allow P2P traffic between devices through 5244 bridges without forcing it upstream. Note: 5245 this removes isolation between devices and 5246 may put more devices in an IOMMU group. 5247 config_acs= 5248 Format: 5249 <ACS flags>@<pci_dev>[; ...] 5250 Specify one or more PCI devices (in the format 5251 specified above) optionally prepended with flags 5252 and separated by semicolons. The respective 5253 capabilities will be enabled, disabled or 5254 unchanged based on what is specified in 5255 flags. 5256 5257 ACS Flags is defined as follows: 5258 bit-0 : ACS Source Validation 5259 bit-1 : ACS Translation Blocking 5260 bit-2 : ACS P2P Request Redirect 5261 bit-3 : ACS P2P Completion Redirect 5262 bit-4 : ACS Upstream Forwarding 5263 bit-5 : ACS P2P Egress Control 5264 bit-6 : ACS Direct Translated P2P 5265 Each bit can be marked as: 5266 '0' – force disabled 5267 '1' – force enabled 5268 'x' – unchanged 5269 For example, 5270 pci=config_acs=10x@pci:0:0 5271 would configure all devices that support 5272 ACS to enable P2P Request Redirect, disable 5273 Translation Blocking, and leave Source 5274 Validation unchanged from whatever power-up 5275 or firmware set it to. 5276 5277 Note: this may remove isolation between devices 5278 and may put more devices in an IOMMU group. 5279 force_floating [S390] Force usage of floating interrupts. 5280 nomio [S390] Do not use MIO instructions. 5281 norid [S390] ignore the RID field and force use of 5282 one PCI domain per PCI function 5283 notph [PCIE] If the PCIE_TPH kernel config parameter 5284 is enabled, this kernel boot option can be used 5285 to disable PCIe TLP Processing Hints support 5286 system-wide. 5287 5288 pcie_aspm= [PCIE] Forcibly enable or ignore PCIe Active State Power 5289 Management. 5290 off Don't touch ASPM configuration at all. Leave any 5291 configuration done by firmware unchanged. 5292 force Enable ASPM even on devices that claim not to support it. 5293 WARNING: Forcing ASPM on may cause system lockups. 5294 5295 pcie_ports= [PCIE] PCIe port services handling: 5296 native Use native PCIe services (PME, AER, DPC, PCIe hotplug) 5297 even if the platform doesn't give the OS permission to 5298 use them. This may cause conflicts if the platform 5299 also tries to use these services. 5300 dpc-native Use native PCIe service for DPC only. May 5301 cause conflicts if firmware uses AER or DPC. 5302 compat Disable native PCIe services (PME, AER, DPC, PCIe 5303 hotplug). 5304 5305 pcie_port_pm= [PCIE] PCIe port power management handling: 5306 off Disable power management of all PCIe ports 5307 force Forcibly enable power management of all PCIe ports 5308 5309 pcie_pme= [PCIE,PM] Native PCIe PME signaling options: 5310 nomsi Do not use MSI for native PCIe PME signaling (this makes 5311 all PCIe root ports use INTx for all services). 5312 5313 pcmv= [HW,PCMCIA] BadgePAD 4 5314 5315 pd_ignore_unused 5316 [PM] 5317 Keep all power-domains already enabled by bootloader on, 5318 even if no driver has claimed them. This is useful 5319 for debug and development, but should not be 5320 needed on a platform with proper driver support. 5321 5322 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at 5323 boot time. 5324 Format: { 0 | 1 } 5325 See arch/parisc/kernel/pdc_chassis.c 5326 5327 percpu_alloc= [MM,EARLY] 5328 Select which percpu first chunk allocator to use. 5329 Currently supported values are "embed" and "page". 5330 Archs may support subset or none of the selections. 5331 See comments in mm/percpu.c for details on each 5332 allocator. This parameter is primarily for debugging 5333 and performance comparison. 5334 5335 pirq= [SMP,APIC] Manual mp-table setup 5336 See Documentation/arch/x86/i386/IO-APIC.rst. 5337 5338 plip= [PPT,NET] Parallel port network link 5339 Format: { parport<nr> | timid | 0 } 5340 See also Documentation/admin-guide/parport.rst. 5341 5342 pmtmr= [X86] Manual setup of pmtmr I/O Port. 5343 Override pmtimer IOPort with a hex value. 5344 e.g. pmtmr=0x508 5345 5346 pmu_override= [PPC] Override the PMU. 5347 This option takes over the PMU facility, so it is no 5348 longer usable by perf. Setting this option starts the 5349 PMU counters by setting MMCR0 to 0 (the FC bit is 5350 cleared). If a number is given, then MMCR1 is set to 5351 that number, otherwise (e.g., 'pmu_override=on'), MMCR1 5352 remains 0. 5353 5354 pm_async= [PM] 5355 Format: off 5356 This parameter sets the initial value of the 5357 /sys/power/pm_async sysfs knob at boot time. 5358 If set to "off", disables asynchronous suspend and 5359 resume of devices during system-wide power transitions. 5360 This can be useful on platforms where device 5361 dependencies are not well-defined, or for debugging 5362 power management issues. Asynchronous operations are 5363 enabled by default. 5364 5365 5366 pm_debug_messages [SUSPEND,KNL] 5367 Enable suspend/resume debug messages during boot up. 5368 5369 pnp.debug=1 [PNP] 5370 Enable PNP debug messages (depends on the 5371 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time 5372 via /sys/module/pnp/parameters/debug. We always show 5373 current resource usage; turning this on also shows 5374 possible settings and some assignment information. 5375 5376 pnpacpi= [ACPI] 5377 { off } 5378 5379 pnpbios= [ISAPNP] 5380 { on | off | curr | res | no-curr | no-res } 5381 5382 pnp_reserve_irq= 5383 [ISAPNP] Exclude IRQs for the autoconfiguration 5384 5385 pnp_reserve_dma= 5386 [ISAPNP] Exclude DMAs for the autoconfiguration 5387 5388 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration 5389 Ranges are in pairs (I/O port base and size). 5390 5391 pnp_reserve_mem= 5392 [ISAPNP] Exclude memory regions for the 5393 autoconfiguration. 5394 Ranges are in pairs (memory base and size). 5395 5396 ports= [IP_VS_FTP] IPVS ftp helper module 5397 Default is 21. 5398 Up to 8 (IP_VS_APP_MAX_PORTS) ports 5399 may be specified. 5400 Format: <port>,<port>.... 5401 5402 possible_cpus= [SMP,S390,X86] 5403 Format: <unsigned int> 5404 Set the number of possible CPUs, overriding the 5405 regular discovery mechanisms (such as ACPI/FW, etc). 5406 5407 powersave=off [PPC] This option disables power saving features. 5408 It specifically disables cpuidle and sets the 5409 platform machine description specific power_save 5410 function to NULL. On Idle the CPU just reduces 5411 execution priority. 5412 5413 ppc_strict_facility_enable 5414 [PPC,ENABLE] This option catches any kernel floating point, 5415 Altivec, VSX and SPE outside of regions specifically 5416 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()). 5417 There is some performance impact when enabling this. 5418 5419 ppc_tm= [PPC,EARLY] 5420 Format: {"off"} 5421 Disable Hardware Transactional Memory 5422 5423 preempt= [KNL] 5424 Select preemption mode if you have CONFIG_PREEMPT_DYNAMIC 5425 none - Limited to cond_resched() calls 5426 voluntary - Limited to cond_resched() and might_sleep() calls 5427 full - Any section that isn't explicitly preempt disabled 5428 can be preempted anytime. Tasks will also yield 5429 contended spinlocks (if the critical section isn't 5430 explicitly preempt disabled beyond the lock itself). 5431 lazy - Scheduler controlled. Similar to full but instead 5432 of preempting the task immediately, the task gets 5433 one HZ tick time to yield itself before the 5434 preemption will be forced. One preemption is when the 5435 task returns to user space. 5436 5437 print-fatal-signals= 5438 [KNL] debug: print fatal signals 5439 5440 If enabled, warn about various signal handling 5441 related application anomalies: too many signals, 5442 too many POSIX.1 timers, fatal signals causing a 5443 coredump - etc. 5444 5445 If you hit the warning due to signal overflow, 5446 you might want to try "ulimit -i unlimited". 5447 5448 default: off. 5449 5450 printk.always_kmsg_dump= 5451 Trigger kmsg_dump for cases other than kernel oops or 5452 panics 5453 Format: <bool> (1/Y/y=enable, 0/N/n=disable) 5454 default: disabled 5455 5456 printk.console_no_auto_verbose= 5457 Disable console loglevel raise on oops, panic 5458 or lockdep-detected issues (only if lock debug is on). 5459 With an exception to setups with low baudrate on 5460 serial console, keeping this 0 is a good choice 5461 in order to provide more debug information. 5462 Format: <bool> 5463 default: 0 (auto_verbose is enabled) 5464 5465 printk.debug_non_panic_cpus= 5466 Allows storing messages from non-panic CPUs into 5467 the printk log buffer during panic(). They are 5468 flushed to consoles by the panic-CPU on 5469 a best-effort basis. 5470 Format: <bool> (1/Y/y=enable, 0/N/n=disable) 5471 Default: disabled 5472 5473 printk.devkmsg={on,off,ratelimit} 5474 Control writing to /dev/kmsg. 5475 on - unlimited logging to /dev/kmsg from userspace 5476 off - logging to /dev/kmsg disabled 5477 ratelimit - ratelimit the logging 5478 Default: ratelimit 5479 5480 printk.time= Show timing data prefixed to each printk message line 5481 Format: <bool> (1/Y/y=enable, 0/N/n=disable) 5482 5483 proc_mem.force_override= [KNL] 5484 Format: {always | ptrace | never} 5485 Traditionally /proc/pid/mem allows memory permissions to be 5486 overridden without restrictions. This option may be set to 5487 restrict that. Can be one of: 5488 - 'always': traditional behavior always allows mem overrides. 5489 - 'ptrace': only allow mem overrides for active ptracers. 5490 - 'never': never allow mem overrides. 5491 If not specified, default is the CONFIG_PROC_MEM_* choice. 5492 5493 processor.max_cstate= [HW,ACPI] 5494 Limit processor to maximum C-state 5495 max_cstate=9 overrides any DMI blacklist limit. 5496 5497 processor.nocst [HW,ACPI] 5498 Ignore the _CST method to determine C-states, 5499 instead using the legacy FADT method 5500 5501 profile= [KNL] Enable kernel profiling via /proc/profile 5502 Format: [<profiletype>,]<number> 5503 Param: <profiletype>: "schedule" or "kvm" 5504 [defaults to kernel profiling] 5505 Param: "schedule" - profile schedule points. 5506 Param: "kvm" - profile VM exits. 5507 Param: <number> - step/bucket size as a power of 2 for 5508 statistical time based profiling. 5509 5510 prot_virt= [S390] enable hosting protected virtual machines 5511 isolated from the hypervisor (if hardware supports 5512 that). If enabled, the default kernel base address 5513 might be overridden even when Kernel Address Space 5514 Layout Randomization is disabled. 5515 Format: <bool> 5516 5517 psi= [KNL] Enable or disable pressure stall information 5518 tracking. 5519 Format: <bool> 5520 5521 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to 5522 probe for; one of (bare|imps|exps|lifebook|any). 5523 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports 5524 per second. 5525 psmouse.resetafter= [HW,MOUSE] 5526 Try to reset the device after so many bad packets 5527 (0 = never). 5528 psmouse.resolution= 5529 [HW,MOUSE] Set desired mouse resolution, in dpi. 5530 psmouse.smartscroll= 5531 [HW,MOUSE] Controls Logitech smartscroll autorepeat. 5532 0 = disabled, 1 = enabled (default). 5533 5534 pstore.backend= Specify the name of the pstore backend to use 5535 5536 pti= [X86-64] Control Page Table Isolation of user and 5537 kernel address spaces. Disabling this feature 5538 removes hardening, but improves performance of 5539 system calls and interrupts. 5540 5541 on - unconditionally enable 5542 off - unconditionally disable 5543 auto - kernel detects whether your CPU model is 5544 vulnerable to issues that PTI mitigates 5545 5546 Not specifying this option is equivalent to pti=auto. 5547 5548 pty.legacy_count= 5549 [KNL] Number of legacy pty's. Overwrites compiled-in 5550 default number. 5551 5552 quiet [KNL,EARLY] Disable most log messages 5553 5554 r128= [HW,DRM] 5555 5556 radix_hcall_invalidate=on [PPC/PSERIES] 5557 Disable RADIX GTSE feature and use hcall for TLB 5558 invalidate. 5559 5560 raid= [HW,RAID] 5561 See Documentation/admin-guide/md.rst. 5562 5563 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes 5564 See Documentation/admin-guide/blockdev/ramdisk.rst. 5565 5566 ramdisk_start= [Deprecated,RAM] RAM disk image start address 5567 5568 random.trust_cpu=off 5569 [KNL,EARLY] Disable trusting the use of the CPU's 5570 random number generator (if available) to 5571 initialize the kernel's RNG. 5572 5573 random.trust_bootloader=off 5574 [KNL,EARLY] Disable trusting the use of the a seed 5575 passed by the bootloader (if available) to 5576 initialize the kernel's RNG. 5577 5578 randomize_kstack_offset= 5579 [KNL,EARLY] Enable or disable kernel stack offset 5580 randomization, which provides roughly 5 bits of 5581 entropy, frustrating memory corruption attacks 5582 that depend on stack address determinism or 5583 cross-syscall address exposures. This is only 5584 available on architectures that have defined 5585 CONFIG_HAVE_ARCH_RANDOMIZE_KSTACK_OFFSET. 5586 Format: <bool> (1/Y/y=enable, 0/N/n=disable) 5587 Default is CONFIG_RANDOMIZE_KSTACK_OFFSET_DEFAULT. 5588 5589 ras=option[,option,...] [KNL] RAS-specific options 5590 5591 cec_disable [X86] 5592 Disable the Correctable Errors Collector, 5593 see CONFIG_RAS_CEC help text. 5594 5595 rcu_nocbs[=cpu-list] 5596 [KNL] The optional argument is a cpu list, 5597 as described above. 5598 5599 In kernels built with CONFIG_RCU_NOCB_CPU=y, 5600 enable the no-callback CPU mode, which prevents 5601 such CPUs' callbacks from being invoked in 5602 softirq context. Invocation of such CPUs' RCU 5603 callbacks will instead be offloaded to "rcuox/N" 5604 kthreads created for that purpose, where "x" is 5605 "p" for RCU-preempt, "s" for RCU-sched, and "g" 5606 for the kthreads that mediate grace periods; and 5607 "N" is the CPU number. This reduces OS jitter on 5608 the offloaded CPUs, which can be useful for HPC 5609 and real-time workloads. It can also improve 5610 energy efficiency for asymmetric multiprocessors. 5611 5612 If a cpulist is passed as an argument, the specified 5613 list of CPUs is set to no-callback mode from boot. 5614 5615 Otherwise, if the '=' sign and the cpulist 5616 arguments are omitted, no CPU will be set to 5617 no-callback mode from boot but the mode may be 5618 toggled at runtime via cpusets. 5619 5620 Note that this argument takes precedence over 5621 the CONFIG_RCU_NOCB_CPU_DEFAULT_ALL option. 5622 5623 rcu_nocb_poll [KNL] 5624 Rather than requiring that offloaded CPUs 5625 (specified by rcu_nocbs= above) explicitly 5626 awaken the corresponding "rcuoN" kthreads, 5627 make these kthreads poll for callbacks. 5628 This improves the real-time response for the 5629 offloaded CPUs by relieving them of the need to 5630 wake up the corresponding kthread, but degrades 5631 energy efficiency by requiring that the kthreads 5632 periodically wake up to do the polling. 5633 5634 rcutree.blimit= [KNL] 5635 Set maximum number of finished RCU callbacks to 5636 process in one batch. 5637 5638 rcutree.csd_lock_suppress_rcu_stall= [KNL] 5639 Do only a one-line RCU CPU stall warning when 5640 there is an ongoing too-long CSD-lock wait. 5641 5642 rcutree.do_rcu_barrier= [KNL] 5643 Request a call to rcu_barrier(). This is 5644 throttled so that userspace tests can safely 5645 hammer on the sysfs variable if they so choose. 5646 If triggered before the RCU grace-period machinery 5647 is fully active, this will error out with EAGAIN. 5648 5649 rcutree.dump_tree= [KNL] 5650 Dump the structure of the rcu_node combining tree 5651 out at early boot. This is used for diagnostic 5652 purposes, to verify correct tree setup. 5653 5654 rcutree.gp_cleanup_delay= [KNL] 5655 Set the number of jiffies to delay each step of 5656 RCU grace-period cleanup. 5657 5658 rcutree.gp_init_delay= [KNL] 5659 Set the number of jiffies to delay each step of 5660 RCU grace-period initialization. 5661 5662 rcutree.gp_preinit_delay= [KNL] 5663 Set the number of jiffies to delay each step of 5664 RCU grace-period pre-initialization, that is, 5665 the propagation of recent CPU-hotplug changes up 5666 the rcu_node combining tree. 5667 5668 rcutree.jiffies_till_first_fqs= [KNL] 5669 Set delay from grace-period initialization to 5670 first attempt to force quiescent states. 5671 Units are jiffies, minimum value is zero, 5672 and maximum value is HZ. 5673 5674 rcutree.jiffies_till_next_fqs= [KNL] 5675 Set delay between subsequent attempts to force 5676 quiescent states. Units are jiffies, minimum 5677 value is one, and maximum value is HZ. 5678 5679 rcutree.jiffies_till_sched_qs= [KNL] 5680 Set required age in jiffies for a 5681 given grace period before RCU starts 5682 soliciting quiescent-state help from 5683 rcu_note_context_switch() and cond_resched(). 5684 If not specified, the kernel will calculate 5685 a value based on the most recent settings 5686 of rcutree.jiffies_till_first_fqs 5687 and rcutree.jiffies_till_next_fqs. 5688 This calculated value may be viewed in 5689 rcutree.jiffies_to_sched_qs. Any attempt to set 5690 rcutree.jiffies_to_sched_qs will be cheerfully 5691 overwritten. 5692 5693 rcutree.kthread_prio= [KNL,BOOT] 5694 Set the SCHED_FIFO priority of the RCU per-CPU 5695 kthreads (rcuc/N). This value is also used for 5696 the priority of the RCU boost threads (rcub/N) 5697 and for the RCU grace-period kthreads (rcu_bh, 5698 rcu_preempt, and rcu_sched). If RCU_BOOST is 5699 set, valid values are 1-99 and the default is 1 5700 (the least-favored priority). Otherwise, when 5701 RCU_BOOST is not set, valid values are 0-99 and 5702 the default is zero (non-realtime operation). 5703 When RCU_NOCB_CPU is set, also adjust the 5704 priority of NOCB callback kthreads. 5705 5706 rcutree.nocb_nobypass_lim_per_jiffy= [KNL] 5707 On callback-offloaded (rcu_nocbs) CPUs, 5708 RCU reduces the lock contention that would 5709 otherwise be caused by callback floods through 5710 use of the ->nocb_bypass list. However, in the 5711 common non-flooded case, RCU queues directly to 5712 the main ->cblist in order to avoid the extra 5713 overhead of the ->nocb_bypass list and its lock. 5714 But if there are too many callbacks queued during 5715 a single jiffy, RCU pre-queues the callbacks into 5716 the ->nocb_bypass queue. The definition of "too 5717 many" is supplied by this kernel boot parameter. 5718 5719 rcutree.nohz_full_patience_delay= [KNL] 5720 On callback-offloaded (rcu_nocbs) CPUs, avoid 5721 disturbing RCU unless the grace period has 5722 reached the specified age in milliseconds. 5723 Defaults to zero. Large values will be capped 5724 at five seconds. All values will be rounded down 5725 to the nearest value representable by jiffies. 5726 5727 rcutree.qhimark= [KNL] 5728 Set threshold of queued RCU callbacks beyond which 5729 batch limiting is disabled. 5730 5731 rcutree.qlowmark= [KNL] 5732 Set threshold of queued RCU callbacks below which 5733 batch limiting is re-enabled. 5734 5735 rcutree.qovld= [KNL] 5736 Set threshold of queued RCU callbacks beyond which 5737 RCU's force-quiescent-state scan will aggressively 5738 enlist help from cond_resched() and sched IPIs to 5739 help CPUs more quickly reach quiescent states. 5740 Set to less than zero to make this be set based 5741 on rcutree.qhimark at boot time and to zero to 5742 disable more aggressive help enlistment. 5743 5744 rcutree.rcu_delay_page_cache_fill_msec= [KNL] 5745 Set the page-cache refill delay (in milliseconds) 5746 in response to low-memory conditions. The range 5747 of permitted values is in the range 0:100000. 5748 5749 rcutree.rcu_divisor= [KNL] 5750 Set the shift-right count to use to compute 5751 the callback-invocation batch limit bl from 5752 the number of callbacks queued on this CPU. 5753 The result will be bounded below by the value of 5754 the rcutree.blimit kernel parameter. Every bl 5755 callbacks, the softirq handler will exit in 5756 order to allow the CPU to do other work. 5757 5758 Please note that this callback-invocation batch 5759 limit applies only to non-offloaded callback 5760 invocation. Offloaded callbacks are instead 5761 invoked in the context of an rcuoc kthread, which 5762 scheduler will preempt as it does any other task. 5763 5764 rcutree.rcu_fanout_exact= [KNL] 5765 Disable autobalancing of the rcu_node combining 5766 tree. This is used by rcutorture, and might 5767 possibly be useful for architectures having high 5768 cache-to-cache transfer latencies. 5769 5770 rcutree.rcu_fanout_leaf= [KNL] 5771 Change the number of CPUs assigned to each 5772 leaf rcu_node structure. Useful for very 5773 large systems, which will choose the value 64, 5774 and for NUMA systems with large remote-access 5775 latencies, which will choose a value aligned 5776 with the appropriate hardware boundaries. 5777 5778 rcutree.rcu_min_cached_objs= [KNL] 5779 Minimum number of objects which are cached and 5780 maintained per one CPU. Object size is equal 5781 to PAGE_SIZE. The cache allows to reduce the 5782 pressure to page allocator, also it makes the 5783 whole algorithm to behave better in low memory 5784 condition. 5785 5786 rcutree.rcu_nocb_gp_stride= [KNL] 5787 Set the number of NOCB callback kthreads in 5788 each group, which defaults to the square root 5789 of the number of CPUs. Larger numbers reduce 5790 the wakeup overhead on the global grace-period 5791 kthread, but increases that same overhead on 5792 each group's NOCB grace-period kthread. 5793 5794 rcutree.rcu_kick_kthreads= [KNL] 5795 Cause the grace-period kthread to get an extra 5796 wake_up() if it sleeps three times longer than 5797 it should at force-quiescent-state time. 5798 This wake_up() will be accompanied by a 5799 WARN_ONCE() splat and an ftrace_dump(). 5800 5801 rcutree.rcu_resched_ns= [KNL] 5802 Limit the time spend invoking a batch of RCU 5803 callbacks to the specified number of nanoseconds. 5804 By default, this limit is checked only once 5805 every 32 callbacks in order to limit the pain 5806 inflicted by local_clock() overhead. 5807 5808 rcutree.rcu_unlock_delay= [KNL] 5809 In CONFIG_RCU_STRICT_GRACE_PERIOD=y kernels, 5810 this specifies an rcu_read_unlock()-time delay 5811 in microseconds. This defaults to zero. 5812 Larger delays increase the probability of 5813 catching RCU pointer leaks, that is, buggy use 5814 of RCU-protected pointers after the relevant 5815 rcu_read_unlock() has completed. 5816 5817 rcutree.sysrq_rcu= [KNL] 5818 Commandeer a sysrq key to dump out Tree RCU's 5819 rcu_node tree with an eye towards determining 5820 why a new grace period has not yet started. 5821 5822 rcutree.use_softirq= [KNL] 5823 If set to zero, move all RCU_SOFTIRQ processing to 5824 per-CPU rcuc kthreads. Defaults to a non-zero 5825 value, meaning that RCU_SOFTIRQ is used by default. 5826 Specify rcutree.use_softirq=0 to use rcuc kthreads. 5827 5828 But note that CONFIG_PREEMPT_RT=y kernels disable 5829 this kernel boot parameter, forcibly setting it 5830 to zero. 5831 5832 rcutree.enable_rcu_lazy= [KNL] 5833 To save power, batch RCU callbacks and flush after 5834 delay, memory pressure or callback list growing too 5835 big. 5836 5837 rcutree.rcu_normal_wake_from_gp= [KNL] 5838 Reduces a latency of synchronize_rcu() call. This approach 5839 maintains its own track of synchronize_rcu() callers, so it 5840 does not interact with regular callbacks because it does not 5841 use a call_rcu[_hurry]() path. Please note, this is for a 5842 normal grace period. 5843 5844 How to disable it: 5845 5846 echo 0 > /sys/module/rcutree/parameters/rcu_normal_wake_from_gp 5847 or pass a boot parameter "rcutree.rcu_normal_wake_from_gp=0" 5848 5849 Default is 1 if it is not explicitly disabled by the boot parameter 5850 passing 0. 5851 5852 rcuscale.gp_async= [KNL] 5853 Measure performance of asynchronous 5854 grace-period primitives such as call_rcu(). 5855 5856 rcuscale.gp_async_max= [KNL] 5857 Specify the maximum number of outstanding 5858 callbacks per writer thread. When a writer 5859 thread exceeds this limit, it invokes the 5860 corresponding flavor of rcu_barrier() to allow 5861 previously posted callbacks to drain. 5862 5863 rcuscale.gp_exp= [KNL] 5864 Measure performance of expedited synchronous 5865 grace-period primitives. 5866 5867 rcuscale.holdoff= [KNL] 5868 Set test-start holdoff period. The purpose of 5869 this parameter is to delay the start of the 5870 test until boot completes in order to avoid 5871 interference. 5872 5873 rcuscale.kfree_by_call_rcu= [KNL] 5874 In kernels built with CONFIG_RCU_LAZY=y, test 5875 call_rcu() instead of kfree_rcu(). 5876 5877 rcuscale.kfree_mult= [KNL] 5878 Instead of allocating an object of size kfree_obj, 5879 allocate one of kfree_mult * sizeof(kfree_obj). 5880 Defaults to 1. 5881 5882 rcuscale.kfree_rcu_test= [KNL] 5883 Set to measure performance of kfree_rcu() flooding. 5884 5885 rcuscale.kfree_rcu_test_double= [KNL] 5886 Test the double-argument variant of kfree_rcu(). 5887 If this parameter has the same value as 5888 rcuscale.kfree_rcu_test_single, both the single- 5889 and double-argument variants are tested. 5890 5891 rcuscale.kfree_rcu_test_single= [KNL] 5892 Test the single-argument variant of kfree_rcu(). 5893 If this parameter has the same value as 5894 rcuscale.kfree_rcu_test_double, both the single- 5895 and double-argument variants are tested. 5896 5897 rcuscale.kfree_nthreads= [KNL] 5898 The number of threads running loops of kfree_rcu(). 5899 5900 rcuscale.kfree_alloc_num= [KNL] 5901 Number of allocations and frees done in an iteration. 5902 5903 rcuscale.kfree_loops= [KNL] 5904 Number of loops doing rcuscale.kfree_alloc_num number 5905 of allocations and frees. 5906 5907 rcuscale.minruntime= [KNL] 5908 Set the minimum test run time in seconds. This 5909 does not affect the data-collection interval, 5910 but instead allows better measurement of things 5911 like CPU consumption. 5912 5913 rcuscale.nreaders= [KNL] 5914 Set number of RCU readers. The value -1 selects 5915 N, where N is the number of CPUs. A value 5916 "n" less than -1 selects N-n+1, where N is again 5917 the number of CPUs. For example, -2 selects N 5918 (the number of CPUs), -3 selects N+1, and so on. 5919 A value of "n" less than or equal to -N selects 5920 a single reader. 5921 5922 rcuscale.nwriters= [KNL] 5923 Set number of RCU writers. The values operate 5924 the same as for rcuscale.nreaders. 5925 N, where N is the number of CPUs 5926 5927 rcuscale.scale_type= [KNL] 5928 Specify the RCU implementation to test. 5929 5930 rcuscale.shutdown= [KNL] 5931 Shut the system down after performance tests 5932 complete. This is useful for hands-off automated 5933 testing. 5934 5935 rcuscale.verbose= [KNL] 5936 Enable additional printk() statements. 5937 5938 rcuscale.writer_holdoff= [KNL] 5939 Write-side holdoff between grace periods, 5940 in microseconds. The default of zero says 5941 no holdoff. 5942 5943 rcuscale.writer_holdoff_jiffies= [KNL] 5944 Additional write-side holdoff between grace 5945 periods, but in jiffies. The default of zero 5946 says no holdoff. 5947 5948 rcutorture.fqs_duration= [KNL] 5949 Set duration of force_quiescent_state bursts 5950 in microseconds. 5951 5952 rcutorture.fqs_holdoff= [KNL] 5953 Set holdoff time within force_quiescent_state bursts 5954 in microseconds. 5955 5956 rcutorture.fqs_stutter= [KNL] 5957 Set wait time between force_quiescent_state bursts 5958 in seconds. 5959 5960 rcutorture.fwd_progress= [KNL] 5961 Specifies the number of kthreads to be used 5962 for RCU grace-period forward-progress testing 5963 for the types of RCU supporting this notion. 5964 Defaults to 1 kthread, values less than zero or 5965 greater than the number of CPUs cause the number 5966 of CPUs to be used. 5967 5968 rcutorture.fwd_progress_div= [KNL] 5969 Specify the fraction of a CPU-stall-warning 5970 period to do tight-loop forward-progress testing. 5971 5972 rcutorture.fwd_progress_holdoff= [KNL] 5973 Number of seconds to wait between successive 5974 forward-progress tests. 5975 5976 rcutorture.fwd_progress_need_resched= [KNL] 5977 Enclose cond_resched() calls within checks for 5978 need_resched() during tight-loop forward-progress 5979 testing. 5980 5981 rcutorture.gp_cond= [KNL] 5982 Use conditional/asynchronous update-side 5983 normal-grace-period primitives, if available. 5984 5985 rcutorture.gp_cond_exp= [KNL] 5986 Use conditional/asynchronous update-side 5987 expedited-grace-period primitives, if available. 5988 5989 rcutorture.gp_cond_full= [KNL] 5990 Use conditional/asynchronous update-side 5991 normal-grace-period primitives that also take 5992 concurrent expedited grace periods into account, 5993 if available. 5994 5995 rcutorture.gp_cond_exp_full= [KNL] 5996 Use conditional/asynchronous update-side 5997 expedited-grace-period primitives that also take 5998 concurrent normal grace periods into account, 5999 if available. 6000 6001 rcutorture.gp_cond_wi= [KNL] 6002 Nominal wait interval for normal conditional 6003 grace periods (specified by rcutorture's 6004 gp_cond and gp_cond_full module parameters), 6005 in microseconds. The actual wait interval will 6006 be randomly selected to nanosecond granularity up 6007 to this wait interval. Defaults to 16 jiffies, 6008 for example, 16,000 microseconds on a system 6009 with HZ=1000. 6010 6011 rcutorture.gp_cond_wi_exp= [KNL] 6012 Nominal wait interval for expedited conditional 6013 grace periods (specified by rcutorture's 6014 gp_cond_exp and gp_cond_exp_full module 6015 parameters), in microseconds. The actual wait 6016 interval will be randomly selected to nanosecond 6017 granularity up to this wait interval. Defaults to 6018 128 microseconds. 6019 6020 rcutorture.gp_exp= [KNL] 6021 Use expedited update-side primitives, if available. 6022 6023 rcutorture.gp_normal= [KNL] 6024 Use normal (non-expedited) asynchronous 6025 update-side primitives, if available. 6026 6027 rcutorture.gp_poll= [KNL] 6028 Use polled update-side normal-grace-period 6029 primitives, if available. 6030 6031 rcutorture.gp_poll_exp= [KNL] 6032 Use polled update-side expedited-grace-period 6033 primitives, if available. 6034 6035 rcutorture.gp_poll_full= [KNL] 6036 Use polled update-side normal-grace-period 6037 primitives that also take concurrent expedited 6038 grace periods into account, if available. 6039 6040 rcutorture.gp_poll_exp_full= [KNL] 6041 Use polled update-side expedited-grace-period 6042 primitives that also take concurrent normal 6043 grace periods into account, if available. 6044 6045 rcutorture.gp_poll_wi= [KNL] 6046 Nominal wait interval for normal conditional 6047 grace periods (specified by rcutorture's 6048 gp_poll and gp_poll_full module parameters), 6049 in microseconds. The actual wait interval will 6050 be randomly selected to nanosecond granularity up 6051 to this wait interval. Defaults to 16 jiffies, 6052 for example, 16,000 microseconds on a system 6053 with HZ=1000. 6054 6055 rcutorture.gp_poll_wi_exp= [KNL] 6056 Nominal wait interval for expedited conditional 6057 grace periods (specified by rcutorture's 6058 gp_poll_exp and gp_poll_exp_full module 6059 parameters), in microseconds. The actual wait 6060 interval will be randomly selected to nanosecond 6061 granularity up to this wait interval. Defaults to 6062 128 microseconds. 6063 6064 rcutorture.gp_sync= [KNL] 6065 Use normal (non-expedited) synchronous 6066 update-side primitives, if available. If all 6067 of rcutorture.gp_cond=, rcutorture.gp_exp=, 6068 rcutorture.gp_normal=, and rcutorture.gp_sync= 6069 are zero, rcutorture acts as if is interpreted 6070 they are all non-zero. 6071 6072 rcutorture.gpwrap_lag= [KNL] 6073 Enable grace-period wrap lag testing. Setting 6074 to false prevents the gpwrap lag test from 6075 running. Default is true. 6076 6077 rcutorture.gpwrap_lag_gps= [KNL] 6078 Set the value for grace-period wrap lag during 6079 active lag testing periods. This controls how many 6080 grace periods differences we tolerate between 6081 rdp and rnp's gp_seq before setting overflow flag. 6082 The default is always set to 8. 6083 6084 rcutorture.gpwrap_lag_cycle_mins= [KNL] 6085 Set the total cycle duration for gpwrap lag 6086 testing in minutes. This is the total time for 6087 one complete cycle of active and inactive 6088 testing periods. Default is 30 minutes. 6089 6090 rcutorture.gpwrap_lag_active_mins= [KNL] 6091 Set the duration for which gpwrap lag is active 6092 within each cycle, in minutes. During this time, 6093 the grace-period wrap lag will be set to the 6094 value specified by gpwrap_lag_gps. Default is 6095 5 minutes. 6096 6097 rcutorture.irqreader= [KNL] 6098 Run RCU readers from irq handlers, or, more 6099 accurately, from a timer handler. Not all RCU 6100 flavors take kindly to this sort of thing. 6101 6102 rcutorture.leakpointer= [KNL] 6103 Leak an RCU-protected pointer out of the reader. 6104 This can of course result in splats, and is 6105 intended to test the ability of things like 6106 CONFIG_RCU_STRICT_GRACE_PERIOD=y to detect 6107 such leaks. 6108 6109 rcutorture.n_barrier_cbs= [KNL] 6110 Set callbacks/threads for rcu_barrier() testing. 6111 6112 rcutorture.nfakewriters= [KNL] 6113 Set number of concurrent RCU writers. These just 6114 stress RCU, they don't participate in the actual 6115 test, hence the "fake". 6116 6117 rcutorture.nocbs_nthreads= [KNL] 6118 Set number of RCU callback-offload togglers. 6119 Zero (the default) disables toggling. 6120 6121 rcutorture.nocbs_toggle= [KNL] 6122 Set the delay in milliseconds between successive 6123 callback-offload toggling attempts. 6124 6125 rcutorture.nreaders= [KNL] 6126 Set number of RCU readers. The value -1 selects 6127 N-1, where N is the number of CPUs. A value 6128 "n" less than -1 selects N-n-2, where N is again 6129 the number of CPUs. For example, -2 selects N 6130 (the number of CPUs), -3 selects N+1, and so on. 6131 6132 rcutorture.object_debug= [KNL] 6133 Enable debug-object double-call_rcu() testing. 6134 6135 rcutorture.onoff_holdoff= [KNL] 6136 Set time (s) after boot for CPU-hotplug testing. 6137 6138 rcutorture.onoff_interval= [KNL] 6139 Set time (jiffies) between CPU-hotplug operations, 6140 or zero to disable CPU-hotplug testing. 6141 6142 rcutorture.preempt_duration= [KNL] 6143 Set duration (in milliseconds) of preemptions 6144 by a high-priority FIFO real-time task. Set to 6145 zero (the default) to disable. The CPUs to 6146 preempt are selected randomly from the set that 6147 are online at a given point in time. Races with 6148 CPUs going offline are ignored, with that attempt 6149 at preemption skipped. 6150 6151 rcutorture.preempt_interval= [KNL] 6152 Set interval (in milliseconds, defaulting to one 6153 second) between preemptions by a high-priority 6154 FIFO real-time task. This delay is mediated 6155 by an hrtimer and is further fuzzed to avoid 6156 inadvertent synchronizations. 6157 6158 rcutorture.read_exit_burst= [KNL] 6159 The number of times in a given read-then-exit 6160 episode that a set of read-then-exit kthreads 6161 is spawned. 6162 6163 rcutorture.read_exit_delay= [KNL] 6164 The delay, in seconds, between successive 6165 read-then-exit testing episodes. 6166 6167 rcutorture.reader_flavor= [KNL] 6168 A bit mask indicating which readers to use. 6169 If there is more than one bit set, the readers 6170 are entered from low-order bit up, and are 6171 exited in the opposite order. For SRCU, the 6172 0x1 bit is normal readers, 0x2 NMI-safe readers, 6173 and 0x4 light-weight readers. 6174 6175 rcutorture.shuffle_interval= [KNL] 6176 Set task-shuffle interval (s). Shuffling tasks 6177 allows some CPUs to go into dyntick-idle mode 6178 during the rcutorture test. 6179 6180 rcutorture.shutdown_secs= [KNL] 6181 Set time (s) after boot system shutdown. This 6182 is useful for hands-off automated testing. 6183 6184 rcutorture.stall_cpu= [KNL] 6185 Duration of CPU stall (s) to test RCU CPU stall 6186 warnings, zero to disable. 6187 6188 rcutorture.stall_cpu_block= [KNL] 6189 Sleep while stalling if set. This will result 6190 in warnings from preemptible RCU in addition to 6191 any other stall-related activity. Note that 6192 in kernels built with CONFIG_PREEMPTION=n and 6193 CONFIG_PREEMPT_COUNT=y, this parameter will 6194 cause the CPU to pass through a quiescent state. 6195 Given CONFIG_PREEMPTION=n, this will suppress 6196 RCU CPU stall warnings, but will instead result 6197 in scheduling-while-atomic splats. 6198 6199 Use of this module parameter results in splats. 6200 6201 6202 rcutorture.stall_cpu_holdoff= [KNL] 6203 Time to wait (s) after boot before inducing stall. 6204 6205 rcutorture.stall_cpu_irqsoff= [KNL] 6206 Disable interrupts while stalling if set, but only 6207 on the first stall in the set. 6208 6209 rcutorture.stall_cpu_repeat= [KNL] 6210 Number of times to repeat the stall sequence, 6211 so that rcutorture.stall_cpu_repeat=3 will result 6212 in four stall sequences. 6213 6214 rcutorture.stall_gp_kthread= [KNL] 6215 Duration (s) of forced sleep within RCU 6216 grace-period kthread to test RCU CPU stall 6217 warnings, zero to disable. If both stall_cpu 6218 and stall_gp_kthread are specified, the 6219 kthread is starved first, then the CPU. 6220 6221 rcutorture.stat_interval= [KNL] 6222 Time (s) between statistics printk()s. 6223 6224 rcutorture.stutter= [KNL] 6225 Time (s) to stutter testing, for example, specifying 6226 five seconds causes the test to run for five seconds, 6227 wait for five seconds, and so on. This tests RCU's 6228 ability to transition abruptly to and from idle. 6229 6230 rcutorture.test_boost= [KNL] 6231 Test RCU priority boosting? 0=no, 1=maybe, 2=yes. 6232 "Maybe" means test if the RCU implementation 6233 under test support RCU priority boosting. 6234 6235 rcutorture.test_boost_duration= [KNL] 6236 Duration (s) of each individual boost test. 6237 6238 rcutorture.test_boost_holdoff= [KNL] 6239 Holdoff time (s) from start of test to the start 6240 of RCU priority-boost testing. Defaults to zero, 6241 that is, no holdoff. 6242 6243 rcutorture.test_boost_interval= [KNL] 6244 Interval (s) between each boost test. 6245 6246 rcutorture.test_no_idle_hz= [KNL] 6247 Test RCU's dyntick-idle handling. See also the 6248 rcutorture.shuffle_interval parameter. 6249 6250 rcutorture.torture_type= [KNL] 6251 Specify the RCU implementation to test. 6252 6253 rcutorture.verbose= [KNL] 6254 Enable additional printk() statements. 6255 6256 rcupdate.rcu_cpu_stall_ftrace_dump= [KNL] 6257 Dump ftrace buffer after reporting RCU CPU 6258 stall warning. 6259 6260 rcupdate.rcu_cpu_stall_notifiers= [KNL] 6261 Provide RCU CPU stall notifiers, but see the 6262 warnings in the RCU_CPU_STALL_NOTIFIER Kconfig 6263 option's help text. TL;DR: You almost certainly 6264 do not want rcupdate.rcu_cpu_stall_notifiers. 6265 6266 rcupdate.rcu_cpu_stall_suppress= [KNL] 6267 Suppress RCU CPU stall warning messages. 6268 6269 rcupdate.rcu_cpu_stall_suppress_at_boot= [KNL] 6270 Suppress RCU CPU stall warning messages and 6271 rcutorture writer stall warnings that occur 6272 during early boot, that is, during the time 6273 before the init task is spawned. 6274 6275 rcupdate.rcu_cpu_stall_timeout= [KNL] 6276 Set timeout for RCU CPU stall warning messages. 6277 The value is in seconds and the maximum allowed 6278 value is 300 seconds. 6279 6280 rcupdate.rcu_exp_cpu_stall_timeout= [KNL] 6281 Set timeout for expedited RCU CPU stall warning 6282 messages. The value is in milliseconds 6283 and the maximum allowed value is 21000 6284 milliseconds. Please note that this value is 6285 adjusted to an arch timer tick resolution. 6286 Setting this to zero causes the value from 6287 rcupdate.rcu_cpu_stall_timeout to be used (after 6288 conversion from seconds to milliseconds). 6289 6290 rcupdate.rcu_cpu_stall_cputime= [KNL] 6291 Provide statistics on the cputime and count of 6292 interrupts and tasks during the sampling period. For 6293 multiple continuous RCU stalls, all sampling periods 6294 begin at half of the first RCU stall timeout. 6295 6296 rcupdate.rcu_exp_stall_task_details= [KNL] 6297 Print stack dumps of any tasks blocking the 6298 current expedited RCU grace period during an 6299 expedited RCU CPU stall warning. 6300 6301 rcupdate.rcu_expedited= [KNL] 6302 Use expedited grace-period primitives, for 6303 example, synchronize_rcu_expedited() instead 6304 of synchronize_rcu(). This reduces latency, 6305 but can increase CPU utilization, degrade 6306 real-time latency, and degrade energy efficiency. 6307 No effect on CONFIG_TINY_RCU kernels. 6308 6309 rcupdate.rcu_normal= [KNL] 6310 Use only normal grace-period primitives, 6311 for example, synchronize_rcu() instead of 6312 synchronize_rcu_expedited(). This improves 6313 real-time latency, CPU utilization, and 6314 energy efficiency, but can expose users to 6315 increased grace-period latency. This parameter 6316 overrides rcupdate.rcu_expedited. No effect on 6317 CONFIG_TINY_RCU kernels. 6318 6319 rcupdate.rcu_normal_after_boot= [KNL] 6320 Once boot has completed (that is, after 6321 rcu_end_inkernel_boot() has been invoked), use 6322 only normal grace-period primitives. No effect 6323 on CONFIG_TINY_RCU kernels. 6324 6325 But note that CONFIG_PREEMPT_RT=y kernels enables 6326 this kernel boot parameter, forcibly setting 6327 it to the value one, that is, converting any 6328 post-boot attempt at an expedited RCU grace 6329 period to instead use normal non-expedited 6330 grace-period processing. 6331 6332 rcupdate.rcu_task_collapse_lim= [KNL] 6333 Set the maximum number of callbacks present 6334 at the beginning of a grace period that allows 6335 the RCU Tasks flavors to collapse back to using 6336 a single callback queue. This switching only 6337 occurs when rcupdate.rcu_task_enqueue_lim is 6338 set to the default value of -1. 6339 6340 rcupdate.rcu_task_contend_lim= [KNL] 6341 Set the minimum number of callback-queuing-time 6342 lock-contention events per jiffy required to 6343 cause the RCU Tasks flavors to switch to per-CPU 6344 callback queuing. This switching only occurs 6345 when rcupdate.rcu_task_enqueue_lim is set to 6346 the default value of -1. 6347 6348 rcupdate.rcu_task_enqueue_lim= [KNL] 6349 Set the number of callback queues to use for the 6350 RCU Tasks family of RCU flavors. The default 6351 of -1 allows this to be automatically (and 6352 dynamically) adjusted. This parameter is intended 6353 for use in testing. 6354 6355 rcupdate.rcu_task_lazy_lim= [KNL] 6356 Number of callbacks on a given CPU that will 6357 cancel laziness on that CPU. Use -1 to disable 6358 cancellation of laziness, but be advised that 6359 doing so increases the danger of OOM due to 6360 callback flooding. 6361 6362 rcupdate.rcu_task_stall_info= [KNL] 6363 Set initial timeout in jiffies for RCU task stall 6364 informational messages, which give some indication 6365 of the problem for those not patient enough to 6366 wait for ten minutes. Informational messages are 6367 only printed prior to the stall-warning message 6368 for a given grace period. Disable with a value 6369 less than or equal to zero. Defaults to ten 6370 seconds. A change in value does not take effect 6371 until the beginning of the next grace period. 6372 6373 rcupdate.rcu_task_stall_info_mult= [KNL] 6374 Multiplier for time interval between successive 6375 RCU task stall informational messages for a given 6376 RCU tasks grace period. This value is clamped 6377 to one through ten, inclusive. It defaults to 6378 the value three, so that the first informational 6379 message is printed 10 seconds into the grace 6380 period, the second at 40 seconds, the third at 6381 160 seconds, and then the stall warning at 600 6382 seconds would prevent a fourth at 640 seconds. 6383 6384 rcupdate.rcu_task_stall_timeout= [KNL] 6385 Set timeout in jiffies for RCU task stall 6386 warning messages. Disable with a value less 6387 than or equal to zero. Defaults to ten minutes. 6388 A change in value does not take effect until 6389 the beginning of the next grace period. 6390 6391 rcupdate.rcu_tasks_lazy_ms= [KNL] 6392 Set timeout in milliseconds RCU Tasks asynchronous 6393 callback batching for call_rcu_tasks(). 6394 A negative value will take the default. A value 6395 of zero will disable batching. Batching is 6396 always disabled for synchronize_rcu_tasks(). 6397 6398 rcupdate.rcu_self_test= [KNL] 6399 Run the RCU early boot self tests 6400 6401 rdinit= [KNL] 6402 Format: <full_path> 6403 Run specified binary instead of /init from the ramdisk, 6404 used for early userspace startup. See initrd. 6405 6406 rdrand= [X86,EARLY] 6407 force - Override the decision by the kernel to hide the 6408 advertisement of RDRAND support (this affects 6409 certain AMD processors because of buggy BIOS 6410 support, specifically around the suspend/resume 6411 path). 6412 6413 rdt= [HW,X86,RDT] 6414 Turn on/off individual RDT features. List is: 6415 cmt, mbmtotal, mbmlocal, l3cat, l3cdp, l2cat, l2cdp, 6416 mba, smba, bmec, abmc, sdciae, energy[:guid], 6417 perf[:guid]. 6418 E.g. to turn on cmt and turn off mba use: 6419 rdt=cmt,!mba 6420 To turn off all energy telemetry monitoring and ensure that 6421 perf telemetry monitoring associated with guid 0x12345 6422 is enabled use: 6423 rdt=!energy,perf:0x12345 6424 6425 reboot= [KNL] 6426 Format (x86 or x86_64): 6427 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] | d[efault] \ 6428 [[,]s[mp]#### \ 6429 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \ 6430 [[,]f[orce] 6431 Where reboot_mode is one of warm (soft) or cold (hard) or gpio 6432 (prefix with 'panic_' to set mode for panic 6433 reboot only), 6434 reboot_type is one of bios, acpi, kbd, triple, efi, or pci, 6435 reboot_force is either force or not specified, 6436 reboot_cpu is s[mp]#### with #### being the processor 6437 to be used for rebooting. 6438 6439 acpi 6440 Use the ACPI RESET_REG in the FADT. If ACPI is not 6441 configured or the ACPI reset does not work, the reboot 6442 path attempts the reset using the keyboard controller. 6443 6444 bios 6445 Use the CPU reboot vector for warm reset 6446 6447 cold 6448 Set the cold reboot flag 6449 6450 default 6451 There are some built-in platform specific "quirks" 6452 - you may see: "reboot: <name> series board detected. 6453 Selecting <type> for reboots." In the case where you 6454 think the quirk is in error (e.g. you have newer BIOS, 6455 or newer board) using this option will ignore the 6456 built-in quirk table, and use the generic default 6457 reboot actions. 6458 6459 efi 6460 Use efi reset_system runtime service. If EFI is not 6461 configured or the EFI reset does not work, the reboot 6462 path attempts the reset using the keyboard controller. 6463 6464 force 6465 Don't stop other CPUs on reboot. This can make reboot 6466 more reliable in some cases. 6467 6468 kbd 6469 Use the keyboard controller. cold reset (default) 6470 6471 pci 6472 Use a write to the PCI config space register 0xcf9 to 6473 trigger reboot. 6474 6475 triple 6476 Force a triple fault (init) 6477 6478 warm 6479 Don't set the cold reboot flag 6480 6481 Using warm reset will be much faster especially on big 6482 memory systems because the BIOS will not go through 6483 the memory check. Disadvantage is that not all 6484 hardware will be completely reinitialized on reboot so 6485 there may be boot problems on some systems. 6486 6487 6488 refscale.holdoff= [KNL] 6489 Set test-start holdoff period. The purpose of 6490 this parameter is to delay the start of the 6491 test until boot completes in order to avoid 6492 interference. 6493 6494 refscale.lookup_instances= [KNL] 6495 Number of data elements to use for the forms of 6496 SLAB_TYPESAFE_BY_RCU testing. A negative number 6497 is negated and multiplied by nr_cpu_ids, while 6498 zero specifies nr_cpu_ids. 6499 6500 refscale.loops= [KNL] 6501 Set the number of loops over the synchronization 6502 primitive under test. Increasing this number 6503 reduces noise due to loop start/end overhead, 6504 but the default has already reduced the per-pass 6505 noise to a handful of picoseconds on ca. 2020 6506 x86 laptops. 6507 6508 refscale.nreaders= [KNL] 6509 Set number of readers. The default value of -1 6510 selects N, where N is roughly 75% of the number 6511 of CPUs. A value of zero is an interesting choice. 6512 6513 refscale.nruns= [KNL] 6514 Set number of runs, each of which is dumped onto 6515 the console log. 6516 6517 refscale.readdelay= [KNL] 6518 Set the read-side critical-section duration, 6519 measured in microseconds. 6520 6521 refscale.scale_type= [KNL] 6522 Specify the read-protection implementation to test. 6523 6524 refscale.shutdown= [KNL] 6525 Shut down the system at the end of the performance 6526 test. This defaults to 1 (shut it down) when 6527 refscale is built into the kernel and to 0 (leave 6528 it running) when refscale is built as a module. 6529 6530 refscale.verbose= [KNL] 6531 Enable additional printk() statements. 6532 6533 refscale.verbose_batched= [KNL] 6534 Batch the additional printk() statements. If zero 6535 (the default) or negative, print everything. Otherwise, 6536 print every Nth verbose statement, where N is the value 6537 specified. 6538 6539 regulator_ignore_unused 6540 [REGULATOR] 6541 Prevents regulator framework from disabling regulators 6542 that are unused, due no driver claiming them. This may 6543 be useful for debug and development, but should not be 6544 needed on a platform with proper driver support. 6545 6546 relax_domain_level= 6547 [KNL, SMP] Set scheduler's default relax_domain_level. 6548 See Documentation/admin-guide/cgroup-v1/cpusets.rst. 6549 6550 reserve= [KNL,BUGS] Force kernel to ignore I/O ports or memory 6551 Format: <base1>,<size1>[,<base2>,<size2>,...] 6552 Reserve I/O ports or memory so the kernel won't use 6553 them. If <base> is less than 0x10000, the region 6554 is assumed to be I/O ports; otherwise it is memory. 6555 6556 reserve_mem= [RAM] 6557 Format: nn[KMG]:<align>:<label> 6558 Reserve physical memory and label it with a name that 6559 other subsystems can use to access it. This is typically 6560 used for systems that do not wipe the RAM, and this command 6561 line will try to reserve the same physical memory on 6562 soft reboots. Note, it is not guaranteed to be the same 6563 location. For example, if anything about the system changes 6564 or if booting a different kernel. It can also fail if KASLR 6565 places the kernel at the location of where the RAM reservation 6566 was from a previous boot, the new reservation will be at a 6567 different location. 6568 Any subsystem using this feature must add a way to verify 6569 that the contents of the physical memory is from a previous 6570 boot, as there may be cases where the memory will not be 6571 located at the same location. 6572 6573 The format is size:align:label for example, to request 6574 12 megabytes of 4096 alignment for ramoops: 6575 6576 reserve_mem=12M:4096:oops ramoops.mem_name=oops 6577 6578 reservetop= [X86-32,EARLY] 6579 Format: nn[KMG] 6580 Reserves a hole at the top of the kernel virtual 6581 address space. 6582 6583 reset_devices [KNL] Force drivers to reset the underlying device 6584 during initialization. 6585 6586 resume= [SWSUSP] 6587 Specify the partition device for software suspend 6588 Format: 6589 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>} 6590 6591 resume_offset= [SWSUSP] 6592 Specify the offset from the beginning of the partition 6593 given by "resume=" at which the swap header is located, 6594 in <PAGE_SIZE> units (needed only for swap files). 6595 See Documentation/power/swsusp-and-swap-files.rst 6596 6597 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to 6598 read the resume files 6599 6600 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up. 6601 Useful for devices that are detected asynchronously 6602 (e.g. USB and MMC devices). 6603 6604 retain_initrd [RAM] Keep initrd memory after extraction. After boot, it will 6605 be accessible via /sys/firmware/initrd. 6606 6607 retbleed= [X86] Control mitigation of RETBleed (Arbitrary 6608 Speculative Code Execution with Return Instructions) 6609 vulnerability. 6610 6611 AMD-based UNRET and IBPB mitigations alone do not stop 6612 sibling threads from influencing the predictions of other 6613 sibling threads. For that reason, STIBP is used on pro- 6614 cessors that support it, and mitigate SMT on processors 6615 that don't. 6616 6617 off - no mitigation 6618 auto - automatically select a mitigation 6619 auto,nosmt - automatically select a mitigation, 6620 disabling SMT if necessary for 6621 the full mitigation (only on Zen1 6622 and older without STIBP). 6623 ibpb - On AMD, mitigate short speculation 6624 windows on basic block boundaries too. 6625 Safe, highest perf impact. It also 6626 enables STIBP if present. Not suitable 6627 on Intel. 6628 ibpb,nosmt - Like "ibpb" above but will disable SMT 6629 when STIBP is not available. This is 6630 the alternative for systems which do not 6631 have STIBP. 6632 unret - Force enable untrained return thunks, 6633 only effective on AMD f15h-f17h based 6634 systems. 6635 unret,nosmt - Like unret, but will disable SMT when STIBP 6636 is not available. This is the alternative for 6637 systems which do not have STIBP. 6638 6639 Selecting 'auto' will choose a mitigation method at run 6640 time according to the CPU. 6641 6642 Not specifying this option is equivalent to retbleed=auto. 6643 6644 rfkill.default_state= 6645 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm, 6646 etc. communication is blocked by default. 6647 1 Unblocked. 6648 6649 rfkill.master_switch_mode= 6650 0 The "airplane mode" button does nothing. 6651 1 The "airplane mode" button toggles between everything 6652 blocked and the previous configuration. 6653 2 The "airplane mode" button toggles between everything 6654 blocked and everything unblocked. 6655 6656 ring3mwait=disable 6657 [KNL] Disable ring 3 MONITOR/MWAIT feature on supported 6658 CPUs. 6659 6660 riscv_isa_fallback [RISCV,EARLY] 6661 When CONFIG_RISCV_ISA_FALLBACK is not enabled, permit 6662 falling back to detecting extension support by parsing 6663 "riscv,isa" property on devicetree systems when the 6664 replacement properties are not found. See the Kconfig 6665 entry for RISCV_ISA_FALLBACK. 6666 6667 riscv_nousercfi= 6668 all Disable user CFI ABI to userspace even if cpu extension 6669 are available. 6670 bcfi Disable user backward CFI ABI to userspace even if 6671 the shadow stack extension is available. 6672 fcfi Disable user forward CFI ABI to userspace even if the 6673 landing pad extension is available. 6674 6675 ro [KNL] Mount root device read-only on boot 6676 6677 rodata= [KNL,EARLY] 6678 on Mark read-only kernel memory as read-only (default). 6679 off Leave read-only kernel memory writable for debugging. 6680 noalias Mark read-only kernel memory as read-only but retain 6681 writable aliases in the direct map for regions outside 6682 of the kernel image. [arm64] 6683 6684 rockchip.usb_uart 6685 [EARLY] 6686 Enable the uart passthrough on the designated usb port 6687 on Rockchip SoCs. When active, the signals of the 6688 debug-uart get routed to the D+ and D- pins of the usb 6689 port and the regular usb controller gets disabled. 6690 6691 root= [KNL] Root filesystem 6692 Usually this is a block device specifier of some kind, 6693 see the early_lookup_bdev comment in 6694 block/early-lookup.c for details. 6695 Alternatively this can be "ram" for the legacy initial 6696 ramdisk, "nfs" and "cifs" for root on a network file 6697 system, or "mtd" and "ubi" for mounting from raw flash. 6698 6699 rootdelay= [KNL] Delay (in seconds) to pause before attempting to 6700 mount the root filesystem 6701 6702 rootflags= [KNL] Set root filesystem mount option string 6703 6704 rseq_slice_ext= [KNL] RSEQ based time slice extension 6705 Format: boolean 6706 Control enablement of RSEQ based time slice extension. 6707 Default is 'on'. 6708 6709 initramfs_options= [KNL] 6710 Specify mount options for the initramfs mount. 6711 6712 rootfstype= [KNL] Set root filesystem type 6713 6714 rootwait [KNL] Wait (indefinitely) for root device to show up. 6715 Useful for devices that are detected asynchronously 6716 (e.g. USB and MMC devices). 6717 6718 rootwait= [KNL] Maximum time (in seconds) to wait for root device 6719 to show up before attempting to mount the root 6720 filesystem. 6721 6722 rproc_mem=nn[KMG][@address] 6723 [KNL,ARM,CMA] Remoteproc physical memory block. 6724 Memory area to be used by remote processor image, 6725 managed by CMA. 6726 6727 rseq_debug= [KNL] Enable or disable restartable sequence 6728 debug mode. Defaults to CONFIG_RSEQ_DEBUG_DEFAULT_ENABLE. 6729 Format: <bool> 6730 6731 rt_group_sched= [KNL] Enable or disable SCHED_RR/FIFO group scheduling 6732 when CONFIG_RT_GROUP_SCHED=y. Defaults to 6733 !CONFIG_RT_GROUP_SCHED_DEFAULT_DISABLED. 6734 Format: <bool> 6735 6736 rw [KNL] Mount root device read-write on boot 6737 6738 S [KNL] Run init in single mode 6739 6740 s390_iommu= [HW,S390] 6741 Set s390 IOTLB flushing mode 6742 strict 6743 With strict flushing every unmap operation will result 6744 in an IOTLB flush. Default is lazy flushing before 6745 reuse, which is faster. Deprecated, equivalent to 6746 iommu.strict=1. 6747 6748 s390_iommu_aperture= [KNL,S390] 6749 Specifies the size of the per device DMA address space 6750 accessible through the DMA and IOMMU APIs as a decimal 6751 factor of the size of main memory. 6752 The default is 1 meaning that one can concurrently use 6753 as many DMA addresses as physical memory is installed, 6754 if supported by hardware, and thus map all of memory 6755 once. With a value of 2 one can map all of memory twice 6756 and so on. As a special case a factor of 0 imposes no 6757 restrictions other than those given by hardware at the 6758 cost of significant additional memory use for tables. 6759 6760 6761 sched_proxy_exec= [KNL] 6762 Enables or disables "proxy execution" style 6763 solution to mutex-based priority inversion. 6764 Format: <bool> 6765 6766 sched_verbose [KNL,EARLY] Enables verbose scheduler debug messages. 6767 6768 schedstats= [KNL,X86] Enable or disable scheduled statistics. 6769 Allowed values are enable and disable. This feature 6770 incurs a small amount of overhead in the scheduler 6771 but is useful for debugging and performance tuning. 6772 6773 sched_thermal_decay_shift= 6774 [Deprecated] 6775 [KNL, SMP] Set a decay shift for scheduler thermal 6776 pressure signal. Thermal pressure signal follows the 6777 default decay period of other scheduler pelt 6778 signals(usually 32 ms but configurable). Setting 6779 sched_thermal_decay_shift will left shift the decay 6780 period for the thermal pressure signal by the shift 6781 value. 6782 i.e. with the default pelt decay period of 32 ms 6783 sched_thermal_decay_shift thermal pressure decay pr 6784 1 64 ms 6785 2 128 ms 6786 and so on. 6787 Format: integer between 0 and 10 6788 Default is 0. 6789 6790 scftorture.holdoff= [KNL] 6791 Number of seconds to hold off before starting 6792 test. Defaults to zero for module insertion and 6793 to 10 seconds for built-in smp_call_function() 6794 tests. 6795 6796 scftorture.longwait= [KNL] 6797 Request ridiculously long waits randomly selected 6798 up to the chosen limit in seconds. Zero (the 6799 default) disables this feature. Please note 6800 that requesting even small non-zero numbers of 6801 seconds can result in RCU CPU stall warnings, 6802 softlockup complaints, and so on. 6803 6804 scftorture.nthreads= [KNL] 6805 Number of kthreads to spawn to invoke the 6806 smp_call_function() family of functions. 6807 The default of -1 specifies a number of kthreads 6808 equal to the number of CPUs. 6809 6810 scftorture.onoff_holdoff= [KNL] 6811 Number seconds to wait after the start of the 6812 test before initiating CPU-hotplug operations. 6813 6814 scftorture.onoff_interval= [KNL] 6815 Number seconds to wait between successive 6816 CPU-hotplug operations. Specifying zero (which 6817 is the default) disables CPU-hotplug operations. 6818 6819 scftorture.shutdown_secs= [KNL] 6820 The number of seconds following the start of the 6821 test after which to shut down the system. The 6822 default of zero avoids shutting down the system. 6823 Non-zero values are useful for automated tests. 6824 6825 scftorture.stat_interval= [KNL] 6826 The number of seconds between outputting the 6827 current test statistics to the console. A value 6828 of zero disables statistics output. 6829 6830 scftorture.stutter_cpus= [KNL] 6831 The number of jiffies to wait between each change 6832 to the set of CPUs under test. 6833 6834 scftorture.use_cpus_read_lock= [KNL] 6835 Use use_cpus_read_lock() instead of the default 6836 preempt_disable() to disable CPU hotplug 6837 while invoking one of the smp_call_function*() 6838 functions. 6839 6840 scftorture.verbose= [KNL] 6841 Enable additional printk() statements. 6842 6843 scftorture.weight_single= [KNL] 6844 The probability weighting to use for the 6845 smp_call_function_single() function with a zero 6846 "wait" parameter. A value of -1 selects the 6847 default if all other weights are -1. However, 6848 if at least one weight has some other value, a 6849 value of -1 will instead select a weight of zero. 6850 6851 scftorture.weight_single_wait= [KNL] 6852 The probability weighting to use for the 6853 smp_call_function_single() function with a 6854 non-zero "wait" parameter. See weight_single. 6855 6856 scftorture.weight_many= [KNL] 6857 The probability weighting to use for the 6858 smp_call_function_many() function with a zero 6859 "wait" parameter. See weight_single. 6860 Note well that setting a high probability for 6861 this weighting can place serious IPI load 6862 on the system. 6863 6864 scftorture.weight_many_wait= [KNL] 6865 The probability weighting to use for the 6866 smp_call_function_many() function with a 6867 non-zero "wait" parameter. See weight_single 6868 and weight_many. 6869 6870 scftorture.weight_all= [KNL] 6871 The probability weighting to use for the 6872 smp_call_function_all() function with a zero 6873 "wait" parameter. See weight_single and 6874 weight_many. 6875 6876 scftorture.weight_all_wait= [KNL] 6877 The probability weighting to use for the 6878 smp_call_function_all() function with a 6879 non-zero "wait" parameter. See weight_single 6880 and weight_many. 6881 6882 sdw_mclk_divider=[SDW] 6883 Specify the MCLK divider for Intel SoundWire buses in 6884 case the BIOS does not provide the clock rate properly. 6885 6886 skew_tick= [KNL,EARLY] Offset the periodic timer tick per cpu to mitigate 6887 xtime_lock contention on larger systems, and/or RCU lock 6888 contention on all systems with CONFIG_MAXSMP set. 6889 Format: { "0" | "1" } 6890 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1" 6891 1 -- enable. 6892 Note: increases power consumption, thus should only be 6893 enabled if running jitter sensitive (HPC/RT) workloads. 6894 6895 security= [SECURITY] Choose a legacy "major" security module to 6896 enable at boot. This has been deprecated by the 6897 "lsm=" parameter. 6898 6899 selinux= [SELINUX] Disable or enable SELinux at boot time. 6900 Format: { "0" | "1" } 6901 See security/selinux/Kconfig help text. 6902 0 -- disable. 6903 1 -- enable. 6904 Default value is 1. 6905 6906 serialnumber [BUGS=X86-32] 6907 6908 sev=option[,option...] [X86-64] 6909 6910 debug 6911 Enable debug messages. 6912 6913 nosnp 6914 Do not enable SEV-SNP (applies to host/hypervisor 6915 only). Setting 'nosnp' avoids the RMP check overhead 6916 in memory accesses when users do not want to run 6917 SEV-SNP guests. 6918 6919 shapers= [NET] 6920 Maximal number of shapers. 6921 6922 show_lapic= [APIC,X86] Advanced Programmable Interrupt Controller 6923 Limit apic dumping. The parameter defines the maximal 6924 number of local apics being dumped. Also it is possible 6925 to set it to "all" by meaning -- no limit here. 6926 Format: { 1 (default) | 2 | ... | all }. 6927 The parameter valid if only apic=debug or 6928 apic=verbose is specified. 6929 Example: apic=debug show_lapic=all 6930 6931 slab_debug[=options[,slabs][;[options[,slabs]]...] [MM] 6932 Enabling slab_debug allows one to determine the 6933 culprit if slab objects become corrupted. Enabling 6934 slab_debug can create guard zones around objects and 6935 may poison objects when not in use. Also tracks the 6936 last alloc / free. For more information see 6937 Documentation/admin-guide/mm/slab.rst. 6938 (slub_debug legacy name also accepted for now) 6939 6940 Using this option implies the "no_hash_pointers" 6941 option which can be undone by adding the 6942 "hash_pointers=always" option. 6943 6944 slab_max_order= [MM] 6945 Determines the maximum allowed order for slabs. 6946 A high setting may cause OOMs due to memory 6947 fragmentation. For more information see 6948 Documentation/admin-guide/mm/slab.rst. 6949 (slub_max_order legacy name also accepted for now) 6950 6951 slab_merge [MM] 6952 Enable merging of slabs with similar size when the 6953 kernel is built without CONFIG_SLAB_MERGE_DEFAULT. 6954 (slub_merge legacy name also accepted for now) 6955 6956 slab_min_objects= [MM] 6957 The minimum number of objects per slab. SLUB will 6958 increase the slab order up to slab_max_order to 6959 generate a sufficiently large slab able to contain 6960 the number of objects indicated. The higher the number 6961 of objects the smaller the overhead of tracking slabs 6962 and the less frequently locks need to be acquired. 6963 For more information see 6964 Documentation/admin-guide/mm/slab.rst. 6965 (slub_min_objects legacy name also accepted for now) 6966 6967 slab_min_order= [MM] 6968 Determines the minimum page order for slabs. Must be 6969 lower or equal to slab_max_order. For more information see 6970 Documentation/admin-guide/mm/slab.rst. 6971 (slub_min_order legacy name also accepted for now) 6972 6973 slab_nomerge [MM] 6974 Disable merging of slabs with similar size. May be 6975 necessary if there is some reason to distinguish 6976 allocs to different slabs, especially in hardened 6977 environments where the risk of heap overflows and 6978 layout control by attackers can usually be 6979 frustrated by disabling merging. This will reduce 6980 most of the exposure of a heap attack to a single 6981 cache (risks via metadata attacks are mostly 6982 unchanged). Debug options disable merging on their 6983 own. 6984 For more information see 6985 Documentation/admin-guide/mm/slab.rst. 6986 (slub_nomerge legacy name also accepted for now) 6987 6988 slab_strict_numa [MM] 6989 Support memory policies on a per object level 6990 in the slab allocator. The default is for memory 6991 policies to be applied at the folio level when 6992 a new folio is needed or a partial folio is 6993 retrieved from the lists. Increases overhead 6994 in the slab fastpaths but gains more accurate 6995 NUMA kernel object placement which helps with slow 6996 interconnects in NUMA systems. 6997 6998 slram= [HW,MTD] 6999 7000 smart2= [HW] 7001 Format: <io1>[,<io2>[,...,<io8>]] 7002 7003 smp.csd_lock_timeout= [KNL] 7004 Specify the period of time in milliseconds 7005 that smp_call_function() and friends will wait 7006 for a CPU to release the CSD lock. This is 7007 useful when diagnosing bugs involving CPUs 7008 disabling interrupts for extended periods 7009 of time. Defaults to 5,000 milliseconds, and 7010 setting a value of zero disables this feature. 7011 This feature may be more efficiently disabled 7012 using the csdlock_debug- kernel parameter. 7013 7014 smp.panic_on_ipistall= [KNL] 7015 If a csd_lock_timeout extends for more than 7016 the specified number of milliseconds, panic the 7017 system. By default, let CSD-lock acquisition 7018 take as long as they take. Specifying 300,000 7019 for this value provides a 5-minute timeout. 7020 7021 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices 7022 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port 7023 smsc-ircc2.ircc_sir= [HW] SIR base I/O port 7024 smsc-ircc2.ircc_fir= [HW] FIR base I/O port 7025 smsc-ircc2.ircc_irq= [HW] IRQ line 7026 smsc-ircc2.ircc_dma= [HW] DMA channel 7027 smsc-ircc2.ircc_transceiver= [HW] Transceiver type: 7028 0: Toshiba Satellite 1800 (GP data pin select) 7029 1: Fast pin select (default) 7030 2: ATC IRMode 7031 7032 smt= [KNL,MIPS,S390,EARLY] Set the maximum number of threads 7033 (logical CPUs) to use per physical CPU on systems 7034 capable of symmetric multithreading (SMT). Will 7035 be capped to the actual hardware limit. 7036 Format: <integer> 7037 Default: -1 (no limit) 7038 7039 softlockup_panic= 7040 [KNL] Should the soft-lockup detector generate panics. 7041 Format: <int> 7042 7043 A value of non-zero instructs the soft-lockup detector 7044 to panic the machine when a soft-lockup duration exceeds 7045 N thresholds. It is also controlled by the kernel.softlockup_panic 7046 sysctl and CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC, which is the 7047 respective build-time switch to that functionality. 7048 7049 softlockup_all_cpu_backtrace= 7050 [KNL] Should the soft-lockup detector generate 7051 backtraces on all cpus. 7052 Format: 0 | 1 7053 7054 sonypi.*= [HW] Sony Programmable I/O Control Device driver 7055 See Documentation/admin-guide/laptops/sonypi.rst 7056 7057 spectre_bhi= [X86] Control mitigation of Branch History Injection 7058 (BHI) vulnerability. This setting affects the 7059 deployment of the HW BHI control and the SW BHB 7060 clearing sequence. 7061 7062 on - (default) Enable the HW or SW mitigation as 7063 needed. This protects the kernel from 7064 both syscalls and VMs. 7065 vmexit - On systems which don't have the HW mitigation 7066 available, enable the SW mitigation on vmexit 7067 ONLY. On such systems, the host kernel is 7068 protected from VM-originated BHI attacks, but 7069 may still be vulnerable to syscall attacks. 7070 off - Disable the mitigation. 7071 7072 spectre_v2= [X86,EARLY] Control mitigation of Spectre variant 2 7073 (indirect branch speculation) vulnerability. 7074 The default operation protects the kernel from 7075 user space attacks. 7076 7077 on - unconditionally enable, implies 7078 spectre_v2_user=on 7079 off - unconditionally disable, implies 7080 spectre_v2_user=off 7081 auto - kernel detects whether your CPU model is 7082 vulnerable 7083 7084 Selecting 'on' will, and 'auto' may, choose a 7085 mitigation method at run time according to the 7086 CPU, the available microcode, the setting of the 7087 CONFIG_MITIGATION_RETPOLINE configuration option, 7088 and the compiler with which the kernel was built. 7089 7090 Selecting 'on' will also enable the mitigation 7091 against user space to user space task attacks. 7092 Selecting specific mitigation does not force enable 7093 user mitigations. 7094 7095 Selecting 'off' will disable both the kernel and 7096 the user space protections. 7097 7098 Specific mitigations can also be selected manually: 7099 7100 retpoline - replace indirect branches 7101 retpoline,generic - Retpolines 7102 retpoline,lfence - LFENCE; indirect branch 7103 retpoline,amd - alias for retpoline,lfence 7104 eibrs - Enhanced/Auto IBRS 7105 eibrs,retpoline - Enhanced/Auto IBRS + Retpolines 7106 eibrs,lfence - Enhanced/Auto IBRS + LFENCE 7107 ibrs - use IBRS to protect kernel 7108 7109 Not specifying this option is equivalent to 7110 spectre_v2=auto. 7111 7112 spectre_v2_user= 7113 [X86] Control mitigation of Spectre variant 2 7114 (indirect branch speculation) vulnerability between 7115 user space tasks 7116 7117 on - Unconditionally enable mitigations. Is 7118 enforced by spectre_v2=on 7119 7120 off - Unconditionally disable mitigations. Is 7121 enforced by spectre_v2=off 7122 7123 prctl - Indirect branch speculation is enabled, 7124 but mitigation can be enabled via prctl 7125 per thread. The mitigation control state 7126 is inherited on fork. 7127 7128 prctl,ibpb 7129 - Like "prctl" above, but only STIBP is 7130 controlled per thread. IBPB is issued 7131 always when switching between different user 7132 space processes. 7133 7134 seccomp 7135 - Same as "prctl" above, but all seccomp 7136 threads will enable the mitigation unless 7137 they explicitly opt out. 7138 7139 seccomp,ibpb 7140 - Like "seccomp" above, but only STIBP is 7141 controlled per thread. IBPB is issued 7142 always when switching between different 7143 user space processes. 7144 7145 auto - Kernel selects the mitigation depending on 7146 the available CPU features and vulnerability. 7147 7148 Default mitigation: "prctl" 7149 7150 Not specifying this option is equivalent to 7151 spectre_v2_user=auto. 7152 7153 spec_rstack_overflow= 7154 [X86,EARLY] Control RAS overflow mitigation on AMD Zen CPUs 7155 7156 off - Disable mitigation 7157 microcode - Enable microcode mitigation only 7158 safe-ret - Enable sw-only safe RET mitigation (default) 7159 ibpb - Enable mitigation by issuing IBPB on 7160 kernel entry 7161 ibpb-vmexit - Issue IBPB only on VMEXIT 7162 (cloud-specific mitigation) 7163 7164 spec_store_bypass_disable= 7165 [HW,EARLY] Control Speculative Store Bypass (SSB) Disable mitigation 7166 (Speculative Store Bypass vulnerability) 7167 7168 Certain CPUs are vulnerable to an exploit against a 7169 a common industry wide performance optimization known 7170 as "Speculative Store Bypass" in which recent stores 7171 to the same memory location may not be observed by 7172 later loads during speculative execution. The idea 7173 is that such stores are unlikely and that they can 7174 be detected prior to instruction retirement at the 7175 end of a particular speculation execution window. 7176 7177 In vulnerable processors, the speculatively forwarded 7178 store can be used in a cache side channel attack, for 7179 example to read memory to which the attacker does not 7180 directly have access (e.g. inside sandboxed code). 7181 7182 This parameter controls whether the Speculative Store 7183 Bypass optimization is used. 7184 7185 On x86 the options are: 7186 7187 on - Unconditionally disable Speculative Store Bypass 7188 off - Unconditionally enable Speculative Store Bypass 7189 auto - Kernel detects whether the CPU model contains an 7190 implementation of Speculative Store Bypass and 7191 picks the most appropriate mitigation. If the 7192 CPU is not vulnerable, "off" is selected. If the 7193 CPU is vulnerable the default mitigation is 7194 architecture and Kconfig dependent. See below. 7195 prctl - Control Speculative Store Bypass per thread 7196 via prctl. Speculative Store Bypass is enabled 7197 for a process by default. The state of the control 7198 is inherited on fork. 7199 seccomp - Same as "prctl" above, but all seccomp threads 7200 will disable SSB unless they explicitly opt out. 7201 7202 Default mitigations: 7203 X86: "prctl" 7204 7205 On powerpc the options are: 7206 7207 on,auto - On Power8 and Power9 insert a store-forwarding 7208 barrier on kernel entry and exit. On Power7 7209 perform a software flush on kernel entry and 7210 exit. 7211 off - No action. 7212 7213 Not specifying this option is equivalent to 7214 spec_store_bypass_disable=auto. 7215 7216 split_llc= 7217 [X86,EARLY] Split the LLC N-ways 7218 7219 When set, the LLC is split this many ways by matching 7220 'core_id % n'. This is setup before SMP bringup and 7221 used during SMP bringup before it knows the full 7222 topology. If your core count doesn't nicely divide by 7223 the number given, you get to keep the pieces. 7224 7225 This is mostly a debug feature to emulate multiple LLCs 7226 on hardware that only have a single LLC. 7227 7228 split_lock_detect= 7229 [X86] Enable split lock detection or bus lock detection 7230 7231 When enabled (and if hardware support is present), atomic 7232 instructions that access data across cache line 7233 boundaries will result in an alignment check exception 7234 for split lock detection or a debug exception for 7235 bus lock detection. 7236 7237 off - not enabled 7238 7239 warn - the kernel will emit rate-limited warnings 7240 about applications triggering the #AC 7241 exception or the #DB exception. This mode is 7242 the default on CPUs that support split lock 7243 detection or bus lock detection. Default 7244 behavior is by #AC if both features are 7245 enabled in hardware. 7246 7247 fatal - the kernel will send SIGBUS to applications 7248 that trigger the #AC exception or the #DB 7249 exception. Default behavior is by #AC if 7250 both features are enabled in hardware. 7251 7252 ratelimit:N - 7253 Set system wide rate limit to N bus locks 7254 per second for bus lock detection. 7255 0 < N <= 1000. 7256 7257 N/A for split lock detection. 7258 7259 7260 If an #AC exception is hit in the kernel or in 7261 firmware (i.e. not while executing in user mode) 7262 the kernel will oops in either "warn" or "fatal" 7263 mode. 7264 7265 #DB exception for bus lock is triggered only when 7266 CPL > 0. 7267 7268 srbds= [X86,INTEL,EARLY] 7269 Control the Special Register Buffer Data Sampling 7270 (SRBDS) mitigation. 7271 7272 Certain CPUs are vulnerable to an MDS-like 7273 exploit which can leak bits from the random 7274 number generator. 7275 7276 By default, this issue is mitigated by 7277 microcode. However, the microcode fix can cause 7278 the RDRAND and RDSEED instructions to become 7279 much slower. Among other effects, this will 7280 result in reduced throughput from /dev/urandom. 7281 7282 The microcode mitigation can be disabled with 7283 the following option: 7284 7285 off: Disable mitigation and remove 7286 performance impact to RDRAND and RDSEED 7287 7288 srcutree.big_cpu_lim [KNL] 7289 Specifies the number of CPUs constituting a 7290 large system, such that srcu_struct structures 7291 should immediately allocate an srcu_node array. 7292 This kernel-boot parameter defaults to 128, 7293 but takes effect only when the low-order four 7294 bits of srcutree.convert_to_big is equal to 3 7295 (decide at boot). 7296 7297 srcutree.convert_to_big [KNL] 7298 Specifies under what conditions an SRCU tree 7299 srcu_struct structure will be converted to big 7300 form, that is, with an rcu_node tree: 7301 7302 0: Never. 7303 1: At init_srcu_struct() time. 7304 2: When rcutorture decides to. 7305 3: Decide at boot time (default). 7306 0x1X: Above plus if high contention. 7307 7308 Either way, the srcu_node tree will be sized based 7309 on the actual runtime number of CPUs (nr_cpu_ids) 7310 instead of the compile-time CONFIG_NR_CPUS. 7311 7312 srcutree.counter_wrap_check [KNL] 7313 Specifies how frequently to check for 7314 grace-period sequence counter wrap for the 7315 srcu_data structure's ->srcu_gp_seq_needed field. 7316 The greater the number of bits set in this kernel 7317 parameter, the less frequently counter wrap will 7318 be checked for. Note that the bottom two bits 7319 are ignored. 7320 7321 srcutree.exp_holdoff [KNL] 7322 Specifies how many nanoseconds must elapse 7323 since the end of the last SRCU grace period for 7324 a given srcu_struct until the next normal SRCU 7325 grace period will be considered for automatic 7326 expediting. Set to zero to disable automatic 7327 expediting. 7328 7329 srcutree.srcu_max_nodelay [KNL] 7330 Specifies the number of no-delay instances 7331 per jiffy for which the SRCU grace period 7332 worker thread will be rescheduled with zero 7333 delay. Beyond this limit, worker thread will 7334 be rescheduled with a sleep delay of one jiffy. 7335 7336 srcutree.srcu_max_nodelay_phase [KNL] 7337 Specifies the per-grace-period phase, number of 7338 non-sleeping polls of readers. Beyond this limit, 7339 grace period worker thread will be rescheduled 7340 with a sleep delay of one jiffy, between each 7341 rescan of the readers, for a grace period phase. 7342 7343 srcutree.srcu_retry_check_delay [KNL] 7344 Specifies number of microseconds of non-sleeping 7345 delay between each non-sleeping poll of readers. 7346 7347 srcutree.small_contention_lim [KNL] 7348 Specifies the number of update-side contention 7349 events per jiffy will be tolerated before 7350 initiating a conversion of an srcu_struct 7351 structure to big form. Note that the value of 7352 srcutree.convert_to_big must have the 0x10 bit 7353 set for contention-based conversions to occur. 7354 7355 ssbd= [ARM64,HW,EARLY] 7356 Speculative Store Bypass Disable control 7357 7358 On CPUs that are vulnerable to the Speculative 7359 Store Bypass vulnerability and offer a 7360 firmware based mitigation, this parameter 7361 indicates how the mitigation should be used: 7362 7363 force-on: Unconditionally enable mitigation for 7364 for both kernel and userspace 7365 force-off: Unconditionally disable mitigation for 7366 for both kernel and userspace 7367 kernel: Always enable mitigation in the 7368 kernel, and offer a prctl interface 7369 to allow userspace to register its 7370 interest in being mitigated too. 7371 7372 stack_guard_gap= [MM] 7373 override the default stack gap protection. The value 7374 is in page units and it defines how many pages prior 7375 to (for stacks growing down) resp. after (for stacks 7376 growing up) the main stack are reserved for no other 7377 mapping. Default value is 256 pages. 7378 7379 stack_depot_disable= [KNL,EARLY] 7380 Setting this to true through kernel command line will 7381 disable the stack depot thereby saving the static memory 7382 consumed by the stack hash table. By default this is set 7383 to false. 7384 7385 stack_depot_max_pools= [KNL,EARLY] 7386 Specify the maximum number of pools to use for storing 7387 stack traces. Pools are allocated on-demand up to this 7388 limit. Default value is 8191 pools. 7389 7390 stacktrace [FTRACE] 7391 Enable the stack tracer on boot up. 7392 7393 stacktrace_filter=[function-list] 7394 [FTRACE] Limit the functions that the stack tracer 7395 will trace at boot up. function-list is a comma-separated 7396 list of functions. This list can be changed at run 7397 time by the stack_trace_filter file in the debugfs 7398 tracing directory. Note, this enables stack tracing 7399 and the stacktrace above is not needed. 7400 7401 sti= [PARISC,HW] 7402 Format: <num> 7403 Set the STI (builtin display/keyboard on the HP-PARISC 7404 machines) console (graphic card) which should be used 7405 as the initial boot-console. 7406 See also comment in drivers/video/sticore.c. 7407 7408 sti_font= [HW] 7409 See comment in drivers/video/sticore.c. 7410 7411 stifb= [HW] 7412 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]] 7413 7414 strict_sas_size= 7415 [X86] 7416 Format: <bool> 7417 Enable or disable strict sigaltstack size checks 7418 against the required signal frame size which 7419 depends on the supported FPU features. This can 7420 be used to filter out binaries which have 7421 not yet been made aware of AT_MINSIGSTKSZ. 7422 7423 stress_hpt [PPC,EARLY] 7424 Limits the number of kernel HPT entries in the hash 7425 page table to increase the rate of hash page table 7426 faults on kernel addresses. 7427 7428 stress_slb [PPC,EARLY] 7429 Limits the number of kernel SLB entries, and flushes 7430 them frequently to increase the rate of SLB faults 7431 on kernel addresses. 7432 7433 no_slb_preload [PPC,EARLY] 7434 Disables slb preloading for userspace. 7435 7436 sunrpc.min_resvport= 7437 sunrpc.max_resvport= 7438 [NFS,SUNRPC] 7439 SunRPC servers often require that client requests 7440 originate from a privileged port (i.e. a port in the 7441 range 0 < portnr < 1024). 7442 An administrator who wishes to reserve some of these 7443 ports for other uses may adjust the range that the 7444 kernel's sunrpc client considers to be privileged 7445 using these two parameters to set the minimum and 7446 maximum port values. 7447 7448 sunrpc.svc_rpc_per_connection_limit= 7449 [NFS,SUNRPC] 7450 Limit the number of requests that the server will 7451 process in parallel from a single connection. 7452 The default value is 0 (no limit). 7453 7454 sunrpc.pool_mode= 7455 [NFS] 7456 Control how the NFS server code allocates CPUs to 7457 service thread pools. Depending on how many NICs 7458 you have and where their interrupts are bound, this 7459 option will affect which CPUs will do NFS serving. 7460 Note: this parameter cannot be changed while the 7461 NFS server is running. 7462 7463 auto the server chooses an appropriate mode 7464 automatically using heuristics 7465 global a single global pool contains all CPUs 7466 percpu one pool for each CPU 7467 pernode one pool for each NUMA node (equivalent 7468 to global on non-NUMA machines) 7469 7470 sunrpc.tcp_slot_table_entries= 7471 sunrpc.udp_slot_table_entries= 7472 [NFS,SUNRPC] 7473 Sets the upper limit on the number of simultaneous 7474 RPC calls that can be sent from the client to a 7475 server. Increasing these values may allow you to 7476 improve throughput, but will also increase the 7477 amount of memory reserved for use by the client. 7478 7479 suspend.pm_test_delay= 7480 [SUSPEND] 7481 Sets the number of seconds to remain in a suspend test 7482 mode before resuming the system (see 7483 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG 7484 is set. Default value is 5. 7485 7486 svm= [PPC] 7487 Format: { on | off | y | n | 1 | 0 } 7488 This parameter controls use of the Protected 7489 Execution Facility on pSeries. 7490 7491 swiotlb= [ARM,PPC,MIPS,X86,S390,EARLY] 7492 Format: { <int> [,<int>] | force | noforce } 7493 <int> -- Number of I/O TLB slabs 7494 <int> -- Second integer after comma. Number of swiotlb 7495 areas with their own lock. Will be rounded up 7496 to a power of 2. 7497 force -- force using of bounce buffers even if they 7498 wouldn't be automatically used by the kernel 7499 noforce -- Never use bounce buffers (for debugging) 7500 7501 switches= [HW,M68k,EARLY] 7502 7503 sysctl.*= [KNL] 7504 Set a sysctl parameter, right before loading the init 7505 process, as if the value was written to the respective 7506 /proc/sys/... file. Both '.' and '/' are recognized as 7507 separators. Unrecognized parameters and invalid values 7508 are reported in the kernel log. Sysctls registered 7509 later by a loaded module cannot be set this way. 7510 Example: sysctl.vm.swappiness=40 7511 7512 sysrq_always_enabled 7513 [KNL] 7514 Ignore sysrq setting - this boot parameter will 7515 neutralize any effect of /proc/sys/kernel/sysrq. 7516 Useful for debugging. 7517 7518 tcpmhash_entries= [KNL,NET] 7519 Set the number of tcp_metrics_hash slots. 7520 Default value is 8192 or 16384 depending on total 7521 ram pages. This is used to specify the TCP metrics 7522 cache size. See Documentation/networking/ip-sysctl.rst 7523 "tcp_no_metrics_save" section for more details. 7524 7525 tdfx= [HW,DRM] 7526 7527 test_suspend= [SUSPEND] 7528 Format: { "mem" | "standby" | "freeze" }[,N] 7529 Specify "mem" (for Suspend-to-RAM) or "standby" (for 7530 standby suspend) or "freeze" (for suspend type freeze) 7531 as the system sleep state during system startup with 7532 the optional capability to repeat N number of times. 7533 The system is woken from this state using a 7534 wakeup-capable RTC alarm. 7535 7536 thash_entries= [KNL,NET] 7537 Set number of hash buckets for TCP connection 7538 7539 thermal.act= [HW,ACPI] 7540 -1: disable all active trip points in all thermal zones 7541 <degrees C>: override all lowest active trip points 7542 7543 thermal.crt= [HW,ACPI] 7544 -1: disable all critical trip points in all thermal zones 7545 <degrees C>: override all critical trip points 7546 7547 thermal.off= [HW,ACPI] 7548 1: disable ACPI thermal control 7549 7550 thermal.psv= [HW,ACPI] 7551 -1: disable all passive trip points 7552 <degrees C>: override all passive trip points to this 7553 value 7554 7555 thermal.tzp= [HW,ACPI] 7556 Specify global default ACPI thermal zone polling rate 7557 <deci-seconds>: poll all this frequency 7558 0: no polling (default) 7559 7560 thp_anon= [KNL] 7561 Format: <size>[KMG],<size>[KMG]:<state>;<size>[KMG]-<size>[KMG]:<state> 7562 state is one of "always", "madvise", "never" or "inherit". 7563 Control the default behavior of the system with respect 7564 to anonymous transparent hugepages. 7565 Can be used multiple times for multiple anon THP sizes. 7566 See Documentation/admin-guide/mm/transhuge.rst for more 7567 details. 7568 7569 threadirqs [KNL,EARLY] 7570 Force threading of all interrupt handlers except those 7571 marked explicitly IRQF_NO_THREAD. 7572 7573 thp_shmem= [KNL] 7574 Format: <size>[KMG],<size>[KMG]:<policy>;<size>[KMG]-<size>[KMG]:<policy> 7575 Control the default policy of each hugepage size for the 7576 internal shmem mount. <policy> is one of policies available 7577 for the shmem mount ("always", "inherit", "never", "within_size", 7578 and "advise"). 7579 It can be used multiple times for multiple shmem THP sizes. 7580 See Documentation/admin-guide/mm/transhuge.rst for more 7581 details. 7582 7583 topology= [S390,EARLY] 7584 Format: {off | on} 7585 Specify if the kernel should make use of the cpu 7586 topology information if the hardware supports this. 7587 The scheduler will make use of this information and 7588 e.g. base its process migration decisions on it. 7589 Default is on. 7590 7591 torture.disable_onoff_at_boot= [KNL] 7592 Prevent the CPU-hotplug component of torturing 7593 until after init has spawned. 7594 7595 torture.ftrace_dump_at_shutdown= [KNL] 7596 Dump the ftrace buffer at torture-test shutdown, 7597 even if there were no errors. This can be a 7598 very costly operation when many torture tests 7599 are running concurrently, especially on systems 7600 with rotating-rust storage. 7601 7602 torture.verbose_sleep_frequency= [KNL] 7603 Specifies how many verbose printk()s should be 7604 emitted between each sleep. The default of zero 7605 disables verbose-printk() sleeping. 7606 7607 torture.verbose_sleep_duration= [KNL] 7608 Duration of each verbose-printk() sleep in jiffies. 7609 7610 tpm.disable_pcr_integrity= [HW,TPM] 7611 Do not protect PCR registers from unintended physical 7612 access, or interposers in the bus by the means of 7613 having an integrity protected session wrapped around 7614 TPM2_PCR_Extend command. Consider this in a situation 7615 where TPM is heavily utilized by IMA, thus protection 7616 causing a major performance hit, and the space where 7617 machines are deployed is by other means guarded. 7618 7619 tpm_crb_ffa.busy_timeout_ms= [ARM64,TPM] 7620 Maximum time in milliseconds to retry sending a message 7621 to the TPM service before giving up. This parameter controls 7622 how long the system will continue retrying when the TPM 7623 service is busy. 7624 Format: <unsigned int> 7625 Default: 2000 (2 seconds) 7626 7627 tpm_suspend_pcr=[HW,TPM] 7628 Format: integer pcr id 7629 Specify that at suspend time, the tpm driver 7630 should extend the specified pcr with zeros, 7631 as a workaround for some chips which fail to 7632 flush the last written pcr on TPM_SaveState. 7633 This will guarantee that all the other pcrs 7634 are saved. 7635 7636 tpm_tis.interrupts= [HW,TPM] 7637 Enable interrupts for the MMIO based physical layer 7638 for the FIFO interface. By default it is set to false 7639 (0). For more information about TPM hardware interfaces 7640 defined by Trusted Computing Group (TCG) see 7641 https://trustedcomputinggroup.org/resource/pc-client-platform-tpm-profile-ptp-specification/ 7642 7643 tp_printk [FTRACE] 7644 Have the tracepoints sent to printk as well as the 7645 tracing ring buffer. This is useful for early boot up 7646 where the system hangs or reboots and does not give the 7647 option for reading the tracing buffer or performing a 7648 ftrace_dump_on_oops. 7649 7650 To turn off having tracepoints sent to printk, 7651 echo 0 > /proc/sys/kernel/tracepoint_printk 7652 Note, echoing 1 into this file without the 7653 tp_printk kernel cmdline option has no effect. 7654 7655 The tp_printk_stop_on_boot (see below) can also be used 7656 to stop the printing of events to console at 7657 late_initcall_sync. 7658 7659 ** CAUTION ** 7660 7661 Having tracepoints sent to printk() and activating high 7662 frequency tracepoints such as irq or sched, can cause 7663 the system to live lock. 7664 7665 tp_printk_stop_on_boot [FTRACE] 7666 When tp_printk (above) is set, it can cause a lot of noise 7667 on the console. It may be useful to only include the 7668 printing of events during boot up, as user space may 7669 make the system inoperable. 7670 7671 This command line option will stop the printing of events 7672 to console at the late_initcall_sync() time frame. 7673 7674 trace_buf_size=nn[KMG] 7675 [FTRACE] will set tracing buffer size on each cpu. 7676 7677 trace_clock= [FTRACE] Set the clock used for tracing events 7678 at boot up. 7679 local - Use the per CPU time stamp counter 7680 (converted into nanoseconds). Fast, but 7681 depending on the architecture, may not be 7682 in sync between CPUs. 7683 global - Event time stamps are synchronized across 7684 CPUs. May be slower than the local clock, 7685 but better for some race conditions. 7686 counter - Simple counting of events (1, 2, ..) 7687 note, some counts may be skipped due to the 7688 infrastructure grabbing the clock more than 7689 once per event. 7690 uptime - Use jiffies as the time stamp. 7691 perf - Use the same clock that perf uses. 7692 mono - Use ktime_get_mono_fast_ns() for time stamps. 7693 mono_raw - Use ktime_get_raw_fast_ns() for time 7694 stamps. 7695 boot - Use ktime_get_boot_fast_ns() for time stamps. 7696 Architectures may add more clocks. See 7697 Documentation/trace/ftrace.rst for more details. 7698 7699 trace_event=[event-list] 7700 [FTRACE] Set and start specified trace events in order 7701 to facilitate early boot debugging. The event-list is a 7702 comma-separated list of trace events to enable. See 7703 also Documentation/trace/events.rst 7704 7705 To enable modules, use :mod: keyword: 7706 7707 trace_event=:mod:<module> 7708 7709 The value before :mod: will only enable specific events 7710 that are part of the module. See the above mentioned 7711 document for more information. 7712 7713 trace_instance=[instance-info] 7714 [FTRACE] Create a ring buffer instance early in boot up. 7715 This will be listed in: 7716 7717 /sys/kernel/tracing/instances 7718 7719 Events can be enabled at the time the instance is created 7720 via: 7721 7722 trace_instance=<name>,<system1>:<event1>,<system2>:<event2> 7723 7724 Note, the "<system*>:" portion is optional if the event is 7725 unique. 7726 7727 trace_instance=foo,sched:sched_switch,irq_handler_entry,initcall 7728 7729 will enable the "sched_switch" event (note, the "sched:" is optional, and 7730 the same thing would happen if it was left off). The irq_handler_entry 7731 event, and all events under the "initcall" system. 7732 7733 Flags can be added to the instance to modify its behavior when it is 7734 created. The flags are separated by '^'. 7735 7736 The available flags are: 7737 7738 traceoff - Have the tracing instance tracing disabled after it is created. 7739 traceprintk - Have trace_printk() write into this trace instance 7740 (note, "printk" and "trace_printk" can also be used) 7741 7742 trace_instance=foo^traceoff^traceprintk,sched,irq 7743 7744 The flags must come before the defined events. 7745 7746 If memory has been reserved (see memmap for x86), the instance 7747 can use that memory: 7748 7749 memmap=12M$0x284500000 trace_instance=boot_map@0x284500000:12M 7750 7751 The above will create a "boot_map" instance that uses the physical 7752 memory at 0x284500000 that is 12Megs. The per CPU buffers of that 7753 instance will be split up accordingly. 7754 7755 Alternatively, the memory can be reserved by the reserve_mem option: 7756 7757 reserve_mem=12M:4096:trace trace_instance=boot_map@trace 7758 7759 This will reserve 12 megabytes at boot up with a 4096 byte alignment 7760 and place the ring buffer in this memory. Note that due to KASLR, the 7761 memory may not be the same location each time, which will not preserve 7762 the buffer content. 7763 7764 Also note that the layout of the ring buffer data may change between 7765 kernel versions where the validator will fail and reset the ring buffer 7766 if the layout is not the same as the previous kernel. 7767 7768 If the ring buffer is used for persistent bootups and has events enabled, 7769 it is recommend to disable tracing so that events from a previous boot do not 7770 mix with events of the current boot (unless you are debugging a random crash 7771 at boot up). 7772 7773 reserve_mem=12M:4096:trace trace_instance=boot_map^traceoff^traceprintk@trace,sched,irq 7774 7775 Note, saving the trace buffer across reboots does require that the system 7776 is set up to not wipe memory. For instance, CONFIG_RESET_ATTACK_MITIGATION 7777 can force a memory reset on boot which will clear any trace that was stored. 7778 This is just one of many ways that can clear memory. Make sure your system 7779 keeps the content of memory across reboots before relying on this option. 7780 7781 NB: Both the mapped address and size must be page aligned for the architecture. 7782 7783 See also Documentation/trace/debugging.rst 7784 7785 7786 trace_options=[option-list] 7787 [FTRACE] Enable or disable tracer options at boot. 7788 The option-list is a comma delimited list of options 7789 that can be enabled or disabled just as if you were 7790 to echo the option name into 7791 7792 /sys/kernel/tracing/trace_options 7793 7794 For example, to enable stacktrace option (to dump the 7795 stack trace of each event), add to the command line: 7796 7797 trace_options=stacktrace 7798 7799 See also Documentation/trace/ftrace.rst "trace options" 7800 section. 7801 7802 trace_trigger=[trigger-list] 7803 [FTRACE] Add an event trigger on specific events. 7804 Set a trigger on top of a specific event, with an optional 7805 filter. 7806 7807 The format is "trace_trigger=<event>.<trigger>[ if <filter>],..." 7808 Where more than one trigger may be specified that are comma delimited. 7809 7810 For example: 7811 7812 trace_trigger="sched_switch.stacktrace if prev_state == 2" 7813 7814 The above will enable the "stacktrace" trigger on the "sched_switch" 7815 event but only trigger it if the "prev_state" of the "sched_switch" 7816 event is "2" (TASK_UNINTERRUPTIBLE). 7817 7818 See also "Event triggers" in Documentation/trace/events.rst 7819 7820 7821 traceoff_after_boot 7822 [FTRACE] Sometimes tracing is used to debug issues 7823 during the boot process. Since the trace buffer has a 7824 limited amount of storage, it may be prudent to 7825 disable tracing after the boot is finished, otherwise 7826 the critical information may be overwritten. With this 7827 option, the main tracing buffer will be turned off at 7828 the end of the boot process. 7829 7830 traceoff_on_warning 7831 [FTRACE] enable this option to disable tracing when a 7832 warning is hit. This turns off "tracing_on". Tracing can 7833 be enabled again by echoing '1' into the "tracing_on" 7834 file located in /sys/kernel/tracing/ 7835 7836 This option is useful, as it disables the trace before 7837 the WARNING dump is called, which prevents the trace to 7838 be filled with content caused by the warning output. 7839 7840 This option can also be set at run time via the sysctl 7841 option: kernel/traceoff_on_warning 7842 7843 transparent_hugepage= 7844 [KNL] 7845 Format: [always|madvise|never] 7846 Can be used to control the default behavior of the system 7847 with respect to transparent hugepages. 7848 See Documentation/admin-guide/mm/transhuge.rst 7849 for more details. 7850 7851 transparent_hugepage_shmem= [KNL] 7852 Format: [always|within_size|advise|never|deny|force] 7853 Can be used to control the hugepage allocation policy for 7854 the internal shmem mount. 7855 See Documentation/admin-guide/mm/transhuge.rst 7856 for more details. 7857 7858 transparent_hugepage_tmpfs= [KNL] 7859 Format: [always|within_size|advise|never] 7860 Can be used to control the default hugepage allocation policy 7861 for the tmpfs mount. 7862 See Documentation/admin-guide/mm/transhuge.rst 7863 for more details. 7864 7865 trusted.source= [KEYS] 7866 Format: <string> 7867 This parameter identifies the trust source as a backend 7868 for trusted keys implementation. Supported trust 7869 sources: 7870 - "tpm" 7871 - "tee" 7872 - "caam" 7873 - "dcp" 7874 - "pkwm" 7875 If not specified then it defaults to iterating through 7876 the trust source list starting with TPM and assigns the 7877 first trust source as a backend which is initialized 7878 successfully during iteration. 7879 7880 trusted.debug= [KEYS] 7881 Format: <bool> 7882 Enable trusted keys debug traces at runtime when 7883 CONFIG_TRUSTED_KEYS_DEBUG=y. 7884 7885 To make the traces visible after enabling the option, 7886 use trusted.dyndbg='+p' as needed. By convention, 7887 the subsystem uses pr_debug() for these traces. 7888 7889 SAFETY: The traces can leak sensitive data, so be 7890 cautious before enabling this. They remain inactive 7891 unless this parameter is set this option to a true 7892 value. 7893 7894 Default: false 7895 7896 trusted.rng= [KEYS] 7897 Format: <string> 7898 The RNG used to generate key material for trusted keys. 7899 Can be one of: 7900 - "kernel" 7901 - the same value as trusted.source: "tpm" or "tee" 7902 - "default" 7903 If not specified, "default" is used. In this case, 7904 the RNG's choice is left to each individual trust source. 7905 7906 trusted.dcp_use_otp_key 7907 This is intended to be used in combination with 7908 trusted.source=dcp and will select the DCP OTP key 7909 instead of the DCP UNIQUE key blob encryption. 7910 7911 trusted.dcp_skip_zk_test 7912 This is intended to be used in combination with 7913 trusted.source=dcp and will disable the check if the 7914 blob key is all zeros. This is helpful for situations where 7915 having this key zero'ed is acceptable. E.g. in testing 7916 scenarios. 7917 7918 tsa= [X86] Control mitigation for Transient Scheduler 7919 Attacks on AMD CPUs. Search the following in your 7920 favourite search engine for more details: 7921 7922 "Technical guidance for mitigating transient scheduler 7923 attacks". 7924 7925 off - disable the mitigation 7926 on - enable the mitigation (default) 7927 user - mitigate only user/kernel transitions 7928 vm - mitigate only guest/host transitions 7929 7930 7931 tsc= Disable clocksource stability checks for TSC. 7932 Format: <string> 7933 [x86] reliable: mark tsc clocksource as reliable, this 7934 disables clocksource verification at runtime, as well 7935 as the stability checks done at bootup. Used to enable 7936 high-resolution timer mode on older hardware, and in 7937 virtualized environment. 7938 [x86] noirqtime: Do not use TSC to do irq accounting. 7939 Used to run time disable IRQ_TIME_ACCOUNTING on any 7940 platforms where RDTSC is slow and this accounting 7941 can add overhead. 7942 [x86] unstable: mark the TSC clocksource as unstable, this 7943 marks the TSC unconditionally unstable at bootup and 7944 avoids any further wobbles once the TSC watchdog notices. 7945 [x86] nowatchdog: disable clocksource watchdog. Used 7946 in situations with strict latency requirements (where 7947 interruptions from clocksource watchdog are not 7948 acceptable). 7949 [x86] recalibrate: force recalibration against a HW timer 7950 (HPET or PM timer) on systems whose TSC frequency was 7951 obtained from HW or FW using either an MSR or CPUID(0x15). 7952 Warn if the difference is more than 500 ppm. 7953 [x86] watchdog: Enforce the clocksource watchdog on TSC 7954 7955 tsc_early_khz= [X86,EARLY] Skip early TSC calibration and use the given 7956 value instead. Useful when the early TSC frequency discovery 7957 procedure is not reliable, such as on overclocked systems 7958 with CPUID.16h support and partial CPUID.15h support. 7959 Format: <unsigned int> 7960 7961 tsx= [X86] Control Transactional Synchronization 7962 Extensions (TSX) feature in Intel processors that 7963 support TSX control. 7964 7965 This parameter controls the TSX feature. The options are: 7966 7967 on - Enable TSX on the system. Although there are 7968 mitigations for all known security vulnerabilities, 7969 TSX has been known to be an accelerator for 7970 several previous speculation-related CVEs, and 7971 so there may be unknown security risks associated 7972 with leaving it enabled. 7973 7974 off - Disable TSX on the system. (Note that this 7975 option takes effect only on newer CPUs which are 7976 not vulnerable to MDS, i.e., have 7977 MSR_IA32_ARCH_CAPABILITIES.MDS_NO=1 and which get 7978 the new IA32_TSX_CTRL MSR through a microcode 7979 update. This new MSR allows for the reliable 7980 deactivation of the TSX functionality.) 7981 7982 auto - Disable TSX if X86_BUG_TAA is present, 7983 otherwise enable TSX on the system. 7984 7985 Not specifying this option is equivalent to tsx=off. 7986 7987 See Documentation/admin-guide/hw-vuln/tsx_async_abort.rst 7988 for more details. 7989 7990 tsx_async_abort= [X86,INTEL,EARLY] Control mitigation for the TSX Async 7991 Abort (TAA) vulnerability. 7992 7993 Similar to Micro-architectural Data Sampling (MDS) 7994 certain CPUs that support Transactional 7995 Synchronization Extensions (TSX) are vulnerable to an 7996 exploit against CPU internal buffers which can forward 7997 information to a disclosure gadget under certain 7998 conditions. 7999 8000 In vulnerable processors, the speculatively forwarded 8001 data can be used in a cache side channel attack, to 8002 access data to which the attacker does not have direct 8003 access. 8004 8005 This parameter controls the TAA mitigation. The 8006 options are: 8007 8008 full - Enable TAA mitigation on vulnerable CPUs 8009 if TSX is enabled. 8010 8011 full,nosmt - Enable TAA mitigation and disable SMT on 8012 vulnerable CPUs. If TSX is disabled, SMT 8013 is not disabled because CPU is not 8014 vulnerable to cross-thread TAA attacks. 8015 off - Unconditionally disable TAA mitigation 8016 8017 On MDS-affected machines, tsx_async_abort=off can be 8018 prevented by an active MDS mitigation as both vulnerabilities 8019 are mitigated with the same mechanism so in order to disable 8020 this mitigation, you need to specify mds=off too. 8021 8022 Not specifying this option is equivalent to 8023 tsx_async_abort=full. On CPUs which are MDS affected 8024 and deploy MDS mitigation, TAA mitigation is not 8025 required and doesn't provide any additional 8026 mitigation. 8027 8028 For details see: 8029 Documentation/admin-guide/hw-vuln/tsx_async_abort.rst 8030 8031 turbografx.map[2|3]= [HW,JOY] 8032 TurboGraFX parallel port interface 8033 Format: 8034 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7> 8035 See also Documentation/input/devices/joystick-parport.rst 8036 8037 udbg-immortal [PPC] When debugging early kernel crashes that 8038 happen after console_init() and before a proper 8039 console driver takes over, this boot options might 8040 help "seeing" what's going on. 8041 8042 uhash_entries= [KNL,NET] 8043 Set number of hash buckets for UDP/UDP-Lite connections 8044 8045 uhci-hcd.ignore_oc= 8046 [USB] Ignore overcurrent events (default N). 8047 Some badly-designed motherboards generate lots of 8048 bogus events, for ports that aren't wired to 8049 anything. Set this parameter to avoid log spamming. 8050 Note that genuine overcurrent events won't be 8051 reported either. 8052 8053 unaligned_scalar_speed= 8054 [RISCV] 8055 Format: {slow | fast | unsupported} 8056 Allow skipping scalar unaligned access speed tests. This 8057 is useful for testing alternative code paths and to skip 8058 the tests in environments where they run too slowly. All 8059 CPUs must have the same scalar unaligned access speed. 8060 8061 unaligned_vector_speed= 8062 [RISCV] 8063 Format: {slow | fast | unsupported} 8064 Allow skipping vector unaligned access speed tests. This 8065 is useful for testing alternative code paths and to skip 8066 the tests in environments where they run too slowly. All 8067 CPUs must have the same vector unaligned access speed. 8068 8069 unknown_nmi_panic 8070 [X86] Cause panic on unknown NMI. 8071 8072 unwind_debug [X86-64,EARLY] 8073 Enable unwinder debug output. This can be 8074 useful for debugging certain unwinder error 8075 conditions, including corrupt stacks and 8076 bad/missing unwinder metadata. 8077 8078 usbcore.authorized_default= 8079 [USB] Default USB device authorization: 8080 (default -1 = authorized (same as 1), 8081 0 = not authorized, 1 = authorized, 2 = authorized 8082 if device connected to internal port) 8083 8084 usbcore.autosuspend= 8085 [USB] The autosuspend time delay (in seconds) used 8086 for newly-detected USB devices (default 2). This 8087 is the time required before an idle device will be 8088 autosuspended. Devices for which the delay is set 8089 to a negative value won't be autosuspended at all. 8090 8091 usbcore.usbfs_snoop= 8092 [USB] Set to log all usbfs traffic (default 0 = off). 8093 8094 usbcore.usbfs_snoop_max= 8095 [USB] Maximum number of bytes to snoop in each URB 8096 (default = 65536). 8097 8098 usbcore.blinkenlights= 8099 [USB] Set to cycle leds on hubs (default 0 = off). 8100 8101 usbcore.old_scheme_first= 8102 [USB] Start with the old device initialization 8103 scheme (default 0 = off). 8104 8105 usbcore.usbfs_memory_mb= 8106 [USB] Memory limit (in MB) for buffers allocated by 8107 usbfs (default = 16, 0 = max = 2047). 8108 8109 usbcore.use_both_schemes= 8110 [USB] Try the other device initialization scheme 8111 if the first one fails (default 1 = enabled). 8112 8113 usbcore.initial_descriptor_timeout= 8114 [USB] Specifies timeout for the initial 64-byte 8115 USB_REQ_GET_DESCRIPTOR request in milliseconds 8116 (default 5000 = 5.0 seconds). 8117 8118 usbcore.nousb [USB] Disable the USB subsystem 8119 8120 usbcore.quirks= 8121 [USB] A list of quirk entries to augment the built-in 8122 usb core quirk list. List entries are separated by 8123 commas. Each entry has the form 8124 VendorID:ProductID:Flags. The IDs are 4-digit hex 8125 numbers and Flags is a set of letters. Each letter 8126 will change the built-in quirk; setting it if it is 8127 clear and clearing it if it is set. The letters have 8128 the following meanings: 8129 a = USB_QUIRK_STRING_FETCH_255 (string 8130 descriptors must not be fetched using 8131 a 255-byte read); 8132 b = USB_QUIRK_RESET_RESUME (device can't resume 8133 correctly so reset it instead); 8134 c = USB_QUIRK_NO_SET_INTF (device can't handle 8135 Set-Interface requests); 8136 d = USB_QUIRK_CONFIG_INTF_STRINGS (device can't 8137 handle its Configuration or Interface 8138 strings); 8139 e = USB_QUIRK_RESET (device can't be reset 8140 (e.g morph devices), don't use reset); 8141 f = USB_QUIRK_HONOR_BNUMINTERFACES (device has 8142 more interface descriptions than the 8143 bNumInterfaces count, and can't handle 8144 talking to these interfaces); 8145 g = USB_QUIRK_DELAY_INIT (device needs a pause 8146 during initialization, after we read 8147 the device descriptor); 8148 h = USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL (For 8149 high speed and super speed interrupt 8150 endpoints, the USB 2.0 and USB 3.0 spec 8151 require the interval in microframes (1 8152 microframe = 125 microseconds) to be 8153 calculated as interval = 2 ^ 8154 (bInterval-1). 8155 Devices with this quirk report their 8156 bInterval as the result of this 8157 calculation instead of the exponent 8158 variable used in the calculation); 8159 i = USB_QUIRK_DEVICE_QUALIFIER (device can't 8160 handle device_qualifier descriptor 8161 requests); 8162 j = USB_QUIRK_IGNORE_REMOTE_WAKEUP (device 8163 generates spurious wakeup, ignore 8164 remote wakeup capability); 8165 k = USB_QUIRK_NO_LPM (device can't handle Link 8166 Power Management); 8167 l = USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL 8168 (Device reports its bInterval as linear 8169 frames instead of the USB 2.0 8170 calculation); 8171 m = USB_QUIRK_DISCONNECT_SUSPEND (Device needs 8172 to be disconnected before suspend to 8173 prevent spurious wakeup); 8174 n = USB_QUIRK_DELAY_CTRL_MSG (Device needs a 8175 pause after every control message); 8176 o = USB_QUIRK_HUB_SLOW_RESET (Hub needs extra 8177 delay after resetting its port); 8178 p = USB_QUIRK_SHORT_SET_ADDRESS_REQ_TIMEOUT 8179 (Reduce timeout of the SET_ADDRESS 8180 request from 5000 ms to 500 ms); 8181 q = USB_QUIRK_FORCE_ONE_CONFIG (Device 8182 claims zero configurations, 8183 forcing to 1); 8184 Example: quirks=0781:5580:bk,0a5c:5834:gij 8185 8186 usbhid.mousepoll= 8187 [USBHID] The interval which mice are to be polled at. 8188 8189 usbhid.jspoll= 8190 [USBHID] The interval which joysticks are to be polled at. 8191 8192 usbhid.kbpoll= 8193 [USBHID] The interval which keyboards are to be polled at. 8194 8195 usb-storage.delay_use= 8196 [UMS] The delay in seconds before a new device is 8197 scanned for Logical Units (default 1). 8198 Optionally the delay in milliseconds if the value has 8199 suffix with "ms". 8200 Example: delay_use=2567ms 8201 8202 usb-storage.quirks= 8203 [UMS] A list of quirks entries to supplement or 8204 override the built-in unusual_devs list. List 8205 entries are separated by commas. Each entry has 8206 the form VID:PID:Flags where VID and PID are Vendor 8207 and Product ID values (4-digit hex numbers) and 8208 Flags is a set of characters, each corresponding 8209 to a common usb-storage quirk flag as follows: 8210 a = SANE_SENSE (collect more than 18 bytes 8211 of sense data, not on uas); 8212 b = BAD_SENSE (don't collect more than 18 8213 bytes of sense data, not on uas); 8214 c = FIX_CAPACITY (decrease the reported 8215 device capacity by one sector); 8216 d = NO_READ_DISC_INFO (don't use 8217 READ_DISC_INFO command, not on uas); 8218 e = NO_READ_CAPACITY_16 (don't use 8219 READ_CAPACITY_16 command); 8220 f = NO_REPORT_OPCODES (don't use report opcodes 8221 command, uas only); 8222 g = MAX_SECTORS_240 (don't transfer more than 8223 240 sectors at a time, uas only); 8224 h = CAPACITY_HEURISTICS (decrease the 8225 reported device capacity by one 8226 sector if the number is odd); 8227 i = IGNORE_DEVICE (don't bind to this 8228 device); 8229 j = NO_REPORT_LUNS (don't use report luns 8230 command, uas only); 8231 k = NO_SAME (do not use WRITE_SAME, uas only) 8232 l = NOT_LOCKABLE (don't try to lock and 8233 unlock ejectable media, not on uas); 8234 m = MAX_SECTORS_64 (don't transfer more 8235 than 64 sectors = 32 KB at a time, 8236 not on uas); 8237 n = INITIAL_READ10 (force a retry of the 8238 initial READ(10) command, not on uas); 8239 o = CAPACITY_OK (accept the capacity 8240 reported by the device, not on uas); 8241 p = WRITE_CACHE (the device cache is ON 8242 by default, not on uas); 8243 r = IGNORE_RESIDUE (the device reports 8244 bogus residue values, not on uas); 8245 s = SINGLE_LUN (the device has only one 8246 Logical Unit); 8247 t = NO_ATA_1X (don't allow ATA(12) and ATA(16) 8248 commands, uas only); 8249 u = IGNORE_UAS (don't bind to the uas driver); 8250 w = NO_WP_DETECT (don't test whether the 8251 medium is write-protected). 8252 y = ALWAYS_SYNC (issue a SYNCHRONIZE_CACHE 8253 even if the device claims no cache, 8254 not on uas) 8255 Example: quirks=0419:aaf5:rl,0421:0433:rc 8256 8257 user_debug= [KNL,ARM] 8258 Format: <int> 8259 See arch/arm/Kconfig.debug help text. 8260 1 - undefined instruction events 8261 2 - system calls 8262 4 - invalid data aborts 8263 8 - SIGSEGV faults 8264 16 - SIGBUS faults 8265 Example: user_debug=31 8266 8267 vdso= [X86,SH,SPARC] 8268 On X86_32, this is an alias for vdso32=. Otherwise: 8269 8270 vdso=1: enable VDSO (the default) 8271 vdso=0: disable VDSO mapping 8272 8273 vdso32= [X86] Control the 32-bit vDSO 8274 vdso32=1: enable 32-bit VDSO 8275 vdso32=0 or vdso32=2: disable 32-bit VDSO 8276 8277 See the help text for CONFIG_COMPAT_VDSO for more 8278 details. If CONFIG_COMPAT_VDSO is set, the default is 8279 vdso32=0; otherwise, the default is vdso32=1. 8280 8281 For compatibility with older kernels, vdso32=2 is an 8282 alias for vdso32=0. 8283 8284 Try vdso32=0 if you encounter an error that says: 8285 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed! 8286 8287 video= [FB,EARLY] Frame buffer configuration 8288 See Documentation/fb/modedb.rst. 8289 8290 video.brightness_switch_enabled= [ACPI] 8291 Format: [0|1] 8292 If set to 1, on receiving an ACPI notify event 8293 generated by hotkey, video driver will adjust brightness 8294 level and then send out the event to user space through 8295 the allocated input device. If set to 0, video driver 8296 will only send out the event without touching backlight 8297 brightness level. 8298 default: 1 8299 8300 virtio_mmio.device= 8301 [VMMIO] Memory mapped virtio (platform) device. 8302 8303 <size>@<baseaddr>:<irq>[:<id>] 8304 where: 8305 <size> := size (can use standard suffixes 8306 like K, M and G) 8307 <baseaddr> := physical base address 8308 <irq> := interrupt number (as passed to 8309 request_irq()) 8310 <id> := (optional) platform device id 8311 example: 8312 virtio_mmio.device=1K@0x100b0000:48:7 8313 8314 Can be used multiple times for multiple devices. 8315 8316 vga= [BOOT,X86-32] Select a particular video mode 8317 See Documentation/arch/x86/boot.rst and 8318 Documentation/admin-guide/svga.rst. 8319 Use vga=ask for menu. 8320 This is actually a boot loader parameter; the value is 8321 passed to the kernel using a special protocol. 8322 8323 vm_debug[=options] [KNL] Available with CONFIG_DEBUG_VM=y. 8324 May slow down system boot speed, especially when 8325 enabled on systems with a large amount of memory. 8326 All options are enabled by default, and this 8327 interface is meant to allow for selectively 8328 enabling or disabling specific virtual memory 8329 debugging features. 8330 8331 Available options are: 8332 P Enable page structure init time poisoning 8333 - Disable all of the above options 8334 8335 vmalloc=nn[KMG] [KNL,BOOT,EARLY] Forces the vmalloc area to have an 8336 exact size of <nn>. This can be used to increase 8337 the minimum size (128MB on x86, arm32 platforms). 8338 It can also be used to decrease the size and leave more room 8339 for directly mapped kernel RAM. Note that this parameter does 8340 not exist on many other platforms (including arm64, alpha, 8341 loongarch, arc, csky, hexagon, microblaze, mips, nios2, openrisc, 8342 parisc, m64k, powerpc, riscv, sh, um, xtensa, s390, sparc). 8343 8344 vmcp_cma=nn[MG] [KNL,S390,EARLY] 8345 Sets the memory size reserved for contiguous memory 8346 allocations for the vmcp device driver. 8347 8348 vmhalt= [KNL,S390] Perform z/VM CP command after system halt. 8349 Format: <command> 8350 8351 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic. 8352 Format: <command> 8353 8354 vmpoff= [KNL,S390] Perform z/VM CP command after power off. 8355 Format: <command> 8356 8357 vmscape= [X86] Controls mitigation for VMscape attacks. 8358 VMscape attacks can leak information from a userspace 8359 hypervisor to a guest via speculative side-channels. 8360 8361 off - disable the mitigation 8362 ibpb - use Indirect Branch Prediction Barrier 8363 (IBPB) mitigation (default) 8364 force - force vulnerability detection even on 8365 unaffected processors 8366 8367 vsyscall= [X86-64,EARLY] 8368 Controls the behavior of vsyscalls (i.e. calls to 8369 fixed addresses of 0xffffffffff600x00 from legacy 8370 code). Most statically-linked binaries and older 8371 versions of glibc use these calls. Because these 8372 functions are at fixed addresses, they make nice 8373 targets for exploits that can control RIP. 8374 8375 emulate Vsyscalls turn into traps and are emulated 8376 reasonably safely. The vsyscall page is 8377 readable. This disables the Linear 8378 Address Space Separation (LASS) security 8379 feature and makes the system less secure. 8380 8381 xonly [default] Vsyscalls turn into traps and are 8382 emulated reasonably safely. The vsyscall 8383 page is not readable. 8384 8385 none Vsyscalls don't work at all. This makes 8386 them quite hard to use for exploits but 8387 might break your system. 8388 8389 vt.color= [VT] Default text color. 8390 Format: 0xYX, X = foreground, Y = background. 8391 Default: 0x07 = light gray on black. 8392 8393 vt.cur_default= [VT] Default cursor shape. 8394 Format: 0xCCBBAA, where AA, BB, and CC are the same as 8395 the parameters of the <Esc>[?A;B;Cc escape sequence; 8396 see vga-softcursor.rst. Default: 2 = underline. 8397 8398 vt.default_blu= [VT] 8399 Format: <blue0>,<blue1>,<blue2>,...,<blue15> 8400 Change the default blue palette of the console. 8401 This is a 16-member array composed of values 8402 ranging from 0-255. 8403 8404 vt.default_grn= [VT] 8405 Format: <green0>,<green1>,<green2>,...,<green15> 8406 Change the default green palette of the console. 8407 This is a 16-member array composed of values 8408 ranging from 0-255. 8409 8410 vt.default_red= [VT] 8411 Format: <red0>,<red1>,<red2>,...,<red15> 8412 Change the default red palette of the console. 8413 This is a 16-member array composed of values 8414 ranging from 0-255. 8415 8416 vt.default_utf8= 8417 [VT] 8418 Format=<0|1> 8419 Set system-wide default UTF-8 mode for all tty's. 8420 Default is 1, i.e. UTF-8 mode is enabled for all 8421 newly opened terminals. 8422 8423 vt.global_cursor_default= 8424 [VT] 8425 Format=<-1|0|1> 8426 Set system-wide default for whether a cursor 8427 is shown on new VTs. Default is -1, 8428 i.e. cursors will be created by default unless 8429 overridden by individual drivers. 0 will hide 8430 cursors, 1 will display them. 8431 8432 vt.italic= [VT] Default color for italic text; 0-15. 8433 Default: 2 = green. 8434 8435 vt.underline= [VT] Default color for underlined text; 0-15. 8436 Default: 3 = cyan. 8437 8438 watchdog timers [HW,WDT] For information on watchdog timers, 8439 see Documentation/watchdog/watchdog-parameters.rst 8440 or other driver-specific files in the 8441 Documentation/watchdog/ directory. 8442 8443 watchdog_thresh= 8444 [KNL] 8445 Set the hard lockup detector stall duration 8446 threshold in seconds. The soft lockup detector 8447 threshold is set to twice the value. A value of 0 8448 disables both lockup detectors. Default is 10 8449 seconds. 8450 8451 workqueue.unbound_cpus= 8452 [KNL,SMP] Specify to constrain one or some CPUs 8453 to use in unbound workqueues. 8454 Format: <cpu-list> 8455 By default, all online CPUs are available for 8456 unbound workqueues. 8457 8458 workqueue.watchdog_thresh= 8459 If CONFIG_WQ_WATCHDOG is configured, workqueue can 8460 warn stall conditions and dump internal state to 8461 help debugging. 0 disables workqueue stall 8462 detection; otherwise, it's the stall threshold 8463 duration in seconds. The default value is 30 and 8464 it can be updated at runtime by writing to the 8465 corresponding sysfs file. 8466 8467 workqueue.panic_on_stall=<uint> 8468 Panic when workqueue stall is detected by 8469 CONFIG_WQ_WATCHDOG. It sets the number times of the 8470 stall to trigger panic. 8471 8472 The default is set by CONFIG_BOOTPARAM_WQ_STALL_PANIC, 8473 which is 0 (disabled) if not configured. 8474 8475 workqueue.panic_on_stall_time=<uint> 8476 Panic when a workqueue stall has been continuous for 8477 the specified number of seconds. Unlike panic_on_stall 8478 which counts accumulated stall events, this triggers 8479 based on the duration of a single continuous stall. 8480 8481 The default is 0, which disables the time-based panic. 8482 8483 workqueue.cpu_intensive_thresh_us= 8484 Per-cpu work items which run for longer than this 8485 threshold are automatically considered CPU intensive 8486 and excluded from concurrency management to prevent 8487 them from noticeably delaying other per-cpu work 8488 items. Default is 10000 (10ms). 8489 8490 If CONFIG_WQ_CPU_INTENSIVE_REPORT is set, the kernel 8491 will report the work functions which violate this 8492 threshold repeatedly. They are likely good 8493 candidates for using WQ_UNBOUND workqueues instead. 8494 8495 workqueue.cpu_intensive_warning_thresh=<uint> 8496 If CONFIG_WQ_CPU_INTENSIVE_REPORT is set, the kernel 8497 will report the work functions which violate the 8498 intensive_threshold_us repeatedly. In order to prevent 8499 spurious warnings, start printing only after a work 8500 function has violated this threshold number of times. 8501 8502 The default is 4 times. 0 disables the warning. 8503 8504 workqueue.power_efficient 8505 Per-cpu workqueues are generally preferred because 8506 they show better performance thanks to cache 8507 locality; unfortunately, per-cpu workqueues tend to 8508 be more power hungry than unbound workqueues. 8509 8510 Enabling this makes the per-cpu workqueues which 8511 were observed to contribute significantly to power 8512 consumption unbound, leading to measurably lower 8513 power usage at the cost of small performance 8514 overhead. 8515 8516 The default value of this parameter is determined by 8517 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT. 8518 8519 workqueue.default_affinity_scope= 8520 Select the default affinity scope to use for unbound 8521 workqueues. Can be one of "cpu", "smt", "cache", 8522 "cache_shard", "numa" and "system". Default is 8523 "cache_shard". For more 8524 information, see the Affinity Scopes section in 8525 Documentation/core-api/workqueue.rst. 8526 8527 This can be changed after boot by writing to the 8528 matching /sys/module/workqueue/parameters file. All 8529 workqueues with the "default" affinity scope will be 8530 updated accordingly. 8531 8532 workqueue.debug_force_rr_cpu 8533 Workqueue used to implicitly guarantee that work 8534 items queued without explicit CPU specified are put 8535 on the local CPU. This guarantee is no longer true 8536 and while local CPU is still preferred work items 8537 may be put on foreign CPUs. This debug option 8538 forces round-robin CPU selection to flush out 8539 usages which depend on the now broken guarantee. 8540 When enabled, memory and cache locality will be 8541 impacted. 8542 8543 writecombine= [LOONGARCH,EARLY] Control the MAT (Memory Access 8544 Type) of ioremap_wc(). 8545 8546 on - Enable writecombine, use WUC for ioremap_wc() 8547 off - Disable writecombine, use SUC for ioremap_wc() 8548 8549 x2apic_phys [X86-64,APIC,EARLY] Use x2apic physical mode instead of 8550 default x2apic cluster mode on platforms 8551 supporting x2apic. 8552 8553 xen_512gb_limit [KNL,X86-64,XEN] 8554 Restricts the kernel running paravirtualized under Xen 8555 to use only up to 512 GB of RAM. The reason to do so is 8556 crash analysis tools and Xen tools for doing domain 8557 save/restore/migration must be enabled to handle larger 8558 domains. 8559 8560 xen_console_io [XEN,EARLY] 8561 Boolean option to enable/disable the usage of the Xen 8562 console_io hypercalls to read and write to the console. 8563 Mostly useful for debugging and development. 8564 8565 xen_emul_unplug= [HW,X86,XEN,EARLY] 8566 Unplug Xen emulated devices 8567 Format: [unplug0,][unplug1] 8568 ide-disks -- unplug primary master IDE devices 8569 aux-ide-disks -- unplug non-primary-master IDE devices 8570 nics -- unplug network devices 8571 all -- unplug all emulated devices (NICs and IDE disks) 8572 unnecessary -- unplugging emulated devices is 8573 unnecessary even if the host did not respond to 8574 the unplug protocol 8575 never -- do not unplug even if version check succeeds 8576 8577 xen_legacy_crash [X86,XEN,EARLY] 8578 Crash from Xen panic notifier, without executing late 8579 panic() code such as dumping handler. 8580 8581 xen_mc_debug [X86,XEN,EARLY] 8582 Enable multicall debugging when running as a Xen PV guest. 8583 Enabling this feature will reduce performance a little 8584 bit, so it should only be enabled for obtaining extended 8585 debug data in case of multicall errors. 8586 8587 xen_msr_safe= [X86,XEN,EARLY] 8588 Format: <bool> 8589 Select whether to always use non-faulting (safe) MSR 8590 access functions when running as Xen PV guest. The 8591 default value is controlled by CONFIG_XEN_PV_MSR_SAFE. 8592 8593 xen_nopv [X86] 8594 Disables the PV optimizations forcing the HVM guest to 8595 run as generic HVM guest with no PV drivers. 8596 This option is obsoleted by the "nopv" option, which 8597 has equivalent effect for XEN platform. 8598 8599 xen_no_vector_callback 8600 [KNL,X86,XEN,EARLY] Disable the vector callback for Xen 8601 event channel interrupts. 8602 8603 xen_scrub_pages= [XEN] 8604 Boolean option to control scrubbing pages before giving them back 8605 to Xen, for use by other domains. Can be also changed at runtime 8606 with /sys/devices/system/xen_memory/xen_memory0/scrub_pages. 8607 Default value controlled with CONFIG_XEN_SCRUB_PAGES_DEFAULT. 8608 8609 xen_timer_slop= [X86-64,XEN,EARLY] 8610 Set the timer slop (in nanoseconds) for the virtual Xen 8611 timers (default is 100000). This adjusts the minimum 8612 delta of virtualized Xen timers, where lower values 8613 improve timer resolution at the expense of processing 8614 more timer interrupts. 8615 8616 xen.balloon_boot_timeout= [XEN] 8617 The time (in seconds) to wait before giving up to boot 8618 in case initial ballooning fails to free enough memory. 8619 Applies only when running as HVM or PVH guest and 8620 started with less memory configured than allowed at 8621 max. Default is 180. 8622 8623 xen.event_eoi_delay= [XEN] 8624 How long to delay EOI handling in case of event 8625 storms (jiffies). Default is 10. 8626 8627 xen.event_loop_timeout= [XEN] 8628 After which time (jiffies) the event handling loop 8629 should start to delay EOI handling. Default is 2. 8630 8631 xen.fifo_events= [XEN] 8632 Boolean parameter to disable using fifo event handling 8633 even if available. Normally fifo event handling is 8634 preferred over the 2-level event handling, as it is 8635 fairer and the number of possible event channels is 8636 much higher. Default is on (use fifo events). 8637 8638 xirc2ps_cs= [NET,PCMCIA] 8639 Format: 8640 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]] 8641 8642 xive= [PPC] 8643 By default on POWER9 and above, the kernel will 8644 natively use the XIVE interrupt controller. This option 8645 allows the fallback firmware mode to be used: 8646 8647 off Fallback to firmware control of XIVE interrupt 8648 controller on both pseries and powernv 8649 platforms. Only useful on POWER9 and above. 8650 8651 xive.store-eoi=off [PPC] 8652 By default on POWER10 and above, the kernel will use 8653 stores for EOI handling when the XIVE interrupt mode 8654 is active. This option allows the XIVE driver to use 8655 loads instead, as on POWER9. 8656 8657 xhci-hcd.quirks [USB,KNL] 8658 A hex value specifying bitmask with supplemental xhci 8659 host controller quirks. Meaning of each bit can be 8660 consulted in header drivers/usb/host/xhci.h. 8661 8662 xmon [PPC,EARLY] 8663 Format: { early | on | rw | ro | off } 8664 Controls if xmon debugger is enabled. Default is off. 8665 Passing only "xmon" is equivalent to "xmon=early". 8666 early Call xmon as early as possible on boot; xmon 8667 debugger is called from setup_arch(). 8668 on xmon debugger hooks will be installed so xmon 8669 is only called on a kernel crash. Default mode, 8670 i.e. either "ro" or "rw" mode, is controlled 8671 with CONFIG_XMON_DEFAULT_RO_MODE. 8672 rw xmon debugger hooks will be installed so xmon 8673 is called only on a kernel crash, mode is write, 8674 meaning SPR registers, memory and, other data 8675 can be written using xmon commands. 8676 ro same as "rw" option above but SPR registers, 8677 memory, and other data can't be written using 8678 xmon commands. 8679 off xmon is disabled. 8680