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 AX25 Appropriate AX.25 support is enabled. 10 CLK Common clock infrastructure is enabled. 11 CMA Contiguous Memory Area support is enabled. 12 DRM Direct Rendering Management support is enabled. 13 DYNAMIC_DEBUG Build in debug messages and enable them at runtime 14 EARLY Parameter processed too early to be embedded in initrd. 15 EDD BIOS Enhanced Disk Drive Services (EDD) is enabled 16 EFI EFI Partitioning (GPT) is enabled 17 EVM Extended Verification Module 18 FB The frame buffer device is enabled. 19 FTRACE Function tracing enabled. 20 GCOV GCOV profiling is enabled. 21 HIBERNATION HIBERNATION is enabled. 22 HW Appropriate hardware is enabled. 23 HYPER_V HYPERV support is enabled. 24 IMA Integrity measurement architecture is enabled. 25 IP_PNP IP DHCP, BOOTP, or RARP is enabled. 26 IPV6 IPv6 support is enabled. 27 ISAPNP ISA PnP code is enabled. 28 ISDN Appropriate ISDN support is enabled. 29 ISOL CPU Isolation is enabled. 30 JOY Appropriate joystick support is enabled. 31 KGDB Kernel debugger support is enabled. 32 KVM Kernel Virtual Machine support is enabled. 33 LIBATA Libata driver is enabled 34 LOONGARCH LoongArch architecture is enabled. 35 LOOP Loopback device support is enabled. 36 LP Printer support is enabled. 37 M68k M68k architecture is enabled. 38 These options have more detailed description inside of 39 Documentation/arch/m68k/kernel-options.rst. 40 MDA MDA console support is enabled. 41 MIPS MIPS architecture is enabled. 42 MOUSE Appropriate mouse support is enabled. 43 MSI Message Signaled Interrupts (PCI). 44 MTD MTD (Memory Technology Device) support is enabled. 45 NET Appropriate network support is enabled. 46 NFS Appropriate NFS support is enabled. 47 NUMA NUMA support is enabled. 48 OF Devicetree is enabled. 49 PARISC The PA-RISC architecture is enabled. 50 PCI PCI bus support is enabled. 51 PCIE PCI Express support is enabled. 52 PCMCIA The PCMCIA subsystem is enabled. 53 PNP Plug & Play support is enabled. 54 PPC PowerPC architecture is enabled. 55 PPT Parallel port support is enabled. 56 PS2 Appropriate PS/2 support is enabled. 57 PV_OPS A paravirtualized kernel is enabled. 58 RAM RAM disk support is enabled. 59 RDT Intel Resource Director Technology. 60 RISCV RISCV architecture is enabled. 61 S390 S390 architecture is enabled. 62 SCSI Appropriate SCSI support is enabled. 63 A lot of drivers have their options described inside 64 the Documentation/scsi/ sub-directory. 65 SDW SoundWire support is enabled. 66 SECURITY Different security models are enabled. 67 SELINUX SELinux support is enabled. 68 SERIAL Serial support is enabled. 69 SH SuperH architecture is enabled. 70 SMP The kernel is an SMP kernel. 71 SPARC Sparc architecture is enabled. 72 SUSPEND System suspend states are enabled. 73 SWSUSP Software suspend (hibernation) is enabled. 74 TPM TPM drivers are enabled. 75 UMS USB Mass Storage support is enabled. 76 USB USB support is enabled. 77 USBHID USB Human Interface Device support is enabled. 78 V4L Video For Linux support is enabled. 79 VGA The VGA console has been enabled. 80 VMMIO Driver for memory mapped virtio devices is enabled. 81 VT Virtual terminal support is enabled. 82 WDT Watchdog support is enabled. 83 X86-32 X86-32, aka i386 architecture is enabled. 84 X86-64 X86-64 architecture is enabled. 85 X86 Either 32-bit or 64-bit x86 (same as X86-32+X86-64) 86 X86_UV SGI UV support is enabled. 87 XEN Xen support is enabled 88 XTENSA xtensa architecture is enabled. 89 90In addition, the following text indicates that the option 91 92 BOOT Is a boot loader parameter. 93 BUGS= Relates to possible processor bugs on the said processor. 94 KNL Is a kernel start-up parameter. 95 96 97Kernel parameters 98 99 accept_memory= [MM] 100 Format: { eager | lazy } 101 default: lazy 102 By default, unaccepted memory is accepted lazily to 103 avoid prolonged boot times. The lazy option will add 104 some runtime overhead until all memory is eventually 105 accepted. In most cases the overhead is negligible. 106 For some workloads or for debugging purposes 107 accept_memory=eager can be used to accept all memory 108 at once during boot. 109 110 acpi= [HW,ACPI,X86,ARM64,RISCV64,EARLY] 111 Advanced Configuration and Power Interface 112 Format: { force | on | off | strict | noirq | rsdt | 113 copy_dsdt | nospcr } 114 force -- enable ACPI if default was off 115 on -- enable ACPI but allow fallback to DT [arm64,riscv64] 116 off -- disable ACPI if default was on 117 noirq -- do not use ACPI for IRQ routing 118 strict -- Be less tolerant of platforms that are not 119 strictly ACPI specification compliant. 120 rsdt -- prefer RSDT over (default) XSDT 121 copy_dsdt -- copy DSDT to memory 122 nocmcff -- Disable firmware first mode for corrected 123 errors. This disables parsing the HEST CMC error 124 source to check if firmware has set the FF flag. This 125 may result in duplicate corrected error reports. 126 nospcr -- disable console in ACPI SPCR table as 127 default _serial_ console on ARM64 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_enforce_resources= [ACPI] 191 { strict | lax | no } 192 Check for resource conflicts between native drivers 193 and ACPI OperationRegions (SystemIO and SystemMemory 194 only). IO ports and memory declared in ACPI might be 195 used by the ACPI subsystem in arbitrary AML code and 196 can interfere with legacy drivers. 197 strict (default): access to resources claimed by ACPI 198 is denied; legacy drivers trying to access reserved 199 resources will fail to bind to device using them. 200 lax: access to resources claimed by ACPI is allowed; 201 legacy drivers trying to access reserved resources 202 will bind successfully but a warning message is logged. 203 no: ACPI OperationRegions are not marked as reserved, 204 no further checks are performed. 205 206 acpi_force_table_verification [HW,ACPI,EARLY] 207 Enable table checksum verification during early stage. 208 By default, this is disabled due to x86 early mapping 209 size limitation. 210 211 acpi_irq_balance [HW,ACPI] 212 ACPI will balance active IRQs 213 default in APIC mode 214 215 acpi_irq_nobalance [HW,ACPI] 216 ACPI will not move active IRQs (default) 217 default in PIC mode 218 219 acpi_irq_isa= [HW,ACPI] If irq_balance, mark listed IRQs used by ISA 220 Format: <irq>,<irq>... 221 222 acpi_irq_pci= [HW,ACPI] If irq_balance, clear listed IRQs for 223 use by PCI 224 Format: <irq>,<irq>... 225 226 acpi_mask_gpe= [HW,ACPI] 227 Due to the existence of _Lxx/_Exx, some GPEs triggered 228 by unsupported hardware/firmware features can result in 229 GPE floodings that cannot be automatically disabled by 230 the GPE dispatcher. 231 This facility can be used to prevent such uncontrolled 232 GPE floodings. 233 Format: <byte> or <bitmap-list> 234 235 acpi_no_auto_serialize [HW,ACPI] 236 Disable auto-serialization of AML methods 237 AML control methods that contain the opcodes to create 238 named objects will be marked as "Serialized" by the 239 auto-serialization feature. 240 This feature is enabled by default. 241 This option allows to turn off the feature. 242 243 acpi_no_memhotplug [ACPI] Disable memory hotplug. Useful for kdump 244 kernels. 245 246 acpi_no_static_ssdt [HW,ACPI,EARLY] 247 Disable installation of static SSDTs at early boot time 248 By default, SSDTs contained in the RSDT/XSDT will be 249 installed automatically and they will appear under 250 /sys/firmware/acpi/tables. 251 This option turns off this feature. 252 Note that specifying this option does not affect 253 dynamic table installation which will install SSDT 254 tables to /sys/firmware/acpi/tables/dynamic. 255 256 acpi_no_watchdog [HW,ACPI,WDT] 257 Ignore the ACPI-based watchdog interface (WDAT) and let 258 a native driver control the watchdog device instead. 259 260 acpi_rsdp= [ACPI,EFI,KEXEC,EARLY] 261 Pass the RSDP address to the kernel, mostly used 262 on machines running EFI runtime service to boot the 263 second kernel for kdump. 264 265 acpi_os_name= [HW,ACPI] Tell ACPI BIOS the name of the OS 266 Format: To spoof as Windows 98: ="Microsoft Windows" 267 268 acpi_rev_override [ACPI] Override the _REV object to return 5 (instead 269 of 2 which is mandated by ACPI 6) as the supported ACPI 270 specification revision (when using this switch, it may 271 be necessary to carry out a cold reboot _twice_ in a 272 row to make it take effect on the platform firmware). 273 274 acpi_osi= [HW,ACPI] Modify list of supported OS interface strings 275 acpi_osi="string1" # add string1 276 acpi_osi="!string2" # remove string2 277 acpi_osi=!* # remove all strings 278 acpi_osi=! # disable all built-in OS vendor 279 strings 280 acpi_osi=!! # enable all built-in OS vendor 281 strings 282 acpi_osi= # disable all strings 283 284 'acpi_osi=!' can be used in combination with single or 285 multiple 'acpi_osi="string1"' to support specific OS 286 vendor string(s). Note that such command can only 287 affect the default state of the OS vendor strings, thus 288 it cannot affect the default state of the feature group 289 strings and the current state of the OS vendor strings, 290 specifying it multiple times through kernel command line 291 is meaningless. This command is useful when one do not 292 care about the state of the feature group strings which 293 should be controlled by the OSPM. 294 Examples: 295 1. 'acpi_osi=! acpi_osi="Windows 2000"' is equivalent 296 to 'acpi_osi="Windows 2000" acpi_osi=!', they all 297 can make '_OSI("Windows 2000")' TRUE. 298 299 'acpi_osi=' cannot be used in combination with other 300 'acpi_osi=' command lines, the _OSI method will not 301 exist in the ACPI namespace. NOTE that such command can 302 only affect the _OSI support state, thus specifying it 303 multiple times through kernel command line is also 304 meaningless. 305 Examples: 306 1. 'acpi_osi=' can make 'CondRefOf(_OSI, Local1)' 307 FALSE. 308 309 'acpi_osi=!*' can be used in combination with single or 310 multiple 'acpi_osi="string1"' to support specific 311 string(s). Note that such command can affect the 312 current state of both the OS vendor strings and the 313 feature group strings, thus specifying it multiple times 314 through kernel command line is meaningful. But it may 315 still not able to affect the final state of a string if 316 there are quirks related to this string. This command 317 is useful when one want to control the state of the 318 feature group strings to debug BIOS issues related to 319 the OSPM features. 320 Examples: 321 1. 'acpi_osi="Module Device" acpi_osi=!*' can make 322 '_OSI("Module Device")' FALSE. 323 2. 'acpi_osi=!* acpi_osi="Module Device"' can make 324 '_OSI("Module Device")' TRUE. 325 3. 'acpi_osi=! acpi_osi=!* acpi_osi="Windows 2000"' is 326 equivalent to 327 'acpi_osi=!* acpi_osi=! acpi_osi="Windows 2000"' 328 and 329 'acpi_osi=!* acpi_osi="Windows 2000" acpi_osi=!', 330 they all will make '_OSI("Windows 2000")' TRUE. 331 332 acpi_pm_good [X86] 333 Override the pmtimer bug detection: force the kernel 334 to assume that this machine's pmtimer latches its value 335 and always returns good values. 336 337 acpi_sci= [HW,ACPI,EARLY] ACPI System Control Interrupt trigger mode 338 Format: { level | edge | high | low } 339 340 acpi_skip_timer_override [HW,ACPI,EARLY] 341 Recognize and ignore IRQ0/pin2 Interrupt Override. 342 For broken nForce2 BIOS resulting in XT-PIC timer. 343 344 acpi_sleep= [HW,ACPI] Sleep options 345 Format: { s3_bios, s3_mode, s3_beep, s4_hwsig, 346 s4_nohwsig, old_ordering, nonvs, 347 sci_force_enable, nobl } 348 See Documentation/power/video.rst for information on 349 s3_bios and s3_mode. 350 s3_beep is for debugging; it makes the PC's speaker beep 351 as soon as the kernel's real-mode entry point is called. 352 s4_hwsig causes the kernel to check the ACPI hardware 353 signature during resume from hibernation, and gracefully 354 refuse to resume if it has changed. This complies with 355 the ACPI specification but not with reality, since 356 Windows does not do this and many laptops do change it 357 on docking. So the default behaviour is to allow resume 358 and simply warn when the signature changes, unless the 359 s4_hwsig option is enabled. 360 s4_nohwsig prevents ACPI hardware signature from being 361 used (or even warned about) during resume. 362 old_ordering causes the ACPI 1.0 ordering of the _PTS 363 control method, with respect to putting devices into 364 low power states, to be enforced (the ACPI 2.0 ordering 365 of _PTS is used by default). 366 nonvs prevents the kernel from saving/restoring the 367 ACPI NVS memory during suspend/hibernation and resume. 368 sci_force_enable causes the kernel to set SCI_EN directly 369 on resume from S1/S3 (which is against the ACPI spec, 370 but some broken systems don't work without it). 371 nobl causes the internal blacklist of systems known to 372 behave incorrectly in some ways with respect to system 373 suspend and resume to be ignored (use wisely). 374 375 acpi_use_timer_override [HW,ACPI,EARLY] 376 Use timer override. For some broken Nvidia NF5 boards 377 that require a timer override, but don't have HPET 378 379 add_efi_memmap [EFI,X86,EARLY] Include EFI memory map in 380 kernel's map of available physical RAM. 381 382 agp= [AGP] 383 { off | try_unsupported } 384 off: disable AGP support 385 try_unsupported: try to drive unsupported chipsets 386 (may crash computer or cause data corruption) 387 388 ALSA [HW,ALSA] 389 See Documentation/sound/alsa-configuration.rst 390 391 alignment= [KNL,ARM] 392 Allow the default userspace alignment fault handler 393 behaviour to be specified. Bit 0 enables warnings, 394 bit 1 enables fixups, and bit 2 sends a segfault. 395 396 align_va_addr= [X86-64] 397 Align virtual addresses by clearing slice [14:12] when 398 allocating a VMA at process creation time. This option 399 gives you up to 3% performance improvement on AMD F15h 400 machines (where it is enabled by default) for a 401 CPU-intensive style benchmark, and it can vary highly in 402 a microbenchmark depending on workload and compiler. 403 404 32: only for 32-bit processes 405 64: only for 64-bit processes 406 on: enable for both 32- and 64-bit processes 407 off: disable for both 32- and 64-bit processes 408 409 alloc_snapshot [FTRACE] 410 Allocate the ftrace snapshot buffer on boot up when the 411 main buffer is allocated. This is handy if debugging 412 and you need to use tracing_snapshot() on boot up, and 413 do not want to use tracing_snapshot_alloc() as it needs 414 to be done where GFP_KERNEL allocations are allowed. 415 416 allow_mismatched_32bit_el0 [ARM64,EARLY] 417 Allow execve() of 32-bit applications and setting of the 418 PER_LINUX32 personality on systems where only a strict 419 subset of the CPUs support 32-bit EL0. When this 420 parameter is present, the set of CPUs supporting 32-bit 421 EL0 is indicated by /sys/devices/system/cpu/aarch32_el0 422 and hot-unplug operations may be restricted. 423 424 See Documentation/arch/arm64/asymmetric-32bit.rst for more 425 information. 426 427 amd_iommu= [HW,X86-64] 428 Pass parameters to the AMD IOMMU driver in the system. 429 Possible values are: 430 fullflush - Deprecated, equivalent to iommu.strict=1 431 off - do not initialize any AMD IOMMU found in 432 the system 433 force_isolation - Force device isolation for all 434 devices. The IOMMU driver is not 435 allowed anymore to lift isolation 436 requirements as needed. This option 437 does not override iommu=pt 438 force_enable - Force enable the IOMMU on platforms known 439 to be buggy with IOMMU enabled. Use this 440 option with care. 441 pgtbl_v1 - Use v1 page table for DMA-API (Default). 442 pgtbl_v2 - Use v2 page table for DMA-API. 443 irtcachedis - Disable Interrupt Remapping Table (IRT) caching. 444 nohugepages - Limit page-sizes used for v1 page-tables 445 to 4 KiB. 446 v2_pgsizes_only - Limit page-sizes used for v1 page-tables 447 to 4KiB/2Mib/1GiB. 448 449 450 amd_iommu_dump= [HW,X86-64] 451 Enable AMD IOMMU driver option to dump the ACPI table 452 for AMD IOMMU. With this option enabled, AMD IOMMU 453 driver will print ACPI tables for AMD IOMMU during 454 IOMMU initialization. 455 456 amd_iommu_intr= [HW,X86-64] 457 Specifies one of the following AMD IOMMU interrupt 458 remapping modes: 459 legacy - Use legacy interrupt remapping mode. 460 vapic - Use virtual APIC mode, which allows IOMMU 461 to inject interrupts directly into guest. 462 This mode requires kvm-amd.avic=1. 463 (Default when IOMMU HW support is present.) 464 465 amd_pstate= [X86,EARLY] 466 disable 467 Do not enable amd_pstate as the default 468 scaling driver for the supported processors 469 passive 470 Use amd_pstate with passive mode as a scaling driver. 471 In this mode autonomous selection is disabled. 472 Driver requests a desired performance level and platform 473 tries to match the same performance level if it is 474 satisfied by guaranteed performance level. 475 active 476 Use amd_pstate_epp driver instance as the scaling driver, 477 driver provides a hint to the hardware if software wants 478 to bias toward performance (0x0) or energy efficiency (0xff) 479 to the CPPC firmware. then CPPC power algorithm will 480 calculate the runtime workload and adjust the realtime cores 481 frequency. 482 guided 483 Activate guided autonomous mode. Driver requests minimum and 484 maximum performance level and the platform autonomously 485 selects a performance level in this range and appropriate 486 to the current workload. 487 488 amd_prefcore= 489 [X86] 490 disable 491 Disable amd-pstate preferred core. 492 493 amijoy.map= [HW,JOY] Amiga joystick support 494 Map of devices attached to JOY0DAT and JOY1DAT 495 Format: <a>,<b> 496 See also Documentation/input/joydev/joystick.rst 497 498 analog.map= [HW,JOY] Analog joystick and gamepad support 499 Specifies type or capabilities of an analog joystick 500 connected to one of 16 gameports 501 Format: <type1>,<type2>,..<type16> 502 503 apc= [HW,SPARC] 504 Power management functions (SPARCstation-4/5 + deriv.) 505 Format: noidle 506 Disable APC CPU standby support. SPARCstation-Fox does 507 not play well with APC CPU idle - disable it if you have 508 APC and your system crashes randomly. 509 510 apic [APIC,X86-64] Use IO-APIC. Default. 511 512 apic= [APIC,X86,EARLY] Advanced Programmable Interrupt Controller 513 Change the output verbosity while booting 514 Format: { quiet (default) | verbose | debug } 515 Change the amount of debugging information output 516 when initialising the APIC and IO-APIC components. 517 518 apic_extnmi= [APIC,X86,EARLY] External NMI delivery setting 519 Format: { bsp (default) | all | none } 520 bsp: External NMI is delivered only to CPU 0 521 all: External NMIs are broadcast to all CPUs as a 522 backup of CPU 0 523 none: External NMI is masked for all CPUs. This is 524 useful so that a dump capture kernel won't be 525 shot down by NMI 526 527 apicpmtimer Do APIC timer calibration using the pmtimer. Implies 528 apicmaintimer. Useful when your PIT timer is totally 529 broken. 530 531 autoconf= [IPV6] 532 See Documentation/networking/ipv6.rst. 533 534 apm= [APM] Advanced Power Management 535 See header of arch/x86/kernel/apm_32.c. 536 537 apparmor= [APPARMOR] Disable or enable AppArmor at boot time 538 Format: { "0" | "1" } 539 See security/apparmor/Kconfig help text 540 0 -- disable. 541 1 -- enable. 542 Default value is set via kernel config option. 543 544 arcrimi= [HW,NET] ARCnet - "RIM I" (entirely mem-mapped) cards 545 Format: <io>,<irq>,<nodeID> 546 547 arm64.no32bit_el0 [ARM64] Unconditionally disable the execution of 548 32 bit applications. 549 550 arm64.nobti [ARM64] Unconditionally disable Branch Target 551 Identification support 552 553 arm64.nogcs [ARM64] Unconditionally disable Guarded Control Stack 554 support 555 556 arm64.nomops [ARM64] Unconditionally disable Memory Copy and Memory 557 Set instructions support 558 559 arm64.nompam [ARM64] Unconditionally disable Memory Partitioning And 560 Monitoring support 561 562 arm64.nomte [ARM64] Unconditionally disable Memory Tagging Extension 563 support 564 565 arm64.nopauth [ARM64] Unconditionally disable Pointer Authentication 566 support 567 568 arm64.nosme [ARM64] Unconditionally disable Scalable Matrix 569 Extension support 570 571 arm64.nosve [ARM64] Unconditionally disable Scalable Vector 572 Extension support 573 574 ataflop= [HW,M68k] 575 576 atarimouse= [HW,MOUSE] Atari Mouse 577 578 atkbd.extra= [HW] Enable extra LEDs and keys on IBM RapidAccess, 579 EzKey and similar keyboards 580 581 atkbd.reset= [HW] Reset keyboard during initialization 582 583 atkbd.set= [HW] Select keyboard code set 584 Format: <int> (2 = AT (default), 3 = PS/2) 585 586 atkbd.scroll= [HW] Enable scroll wheel on MS Office and similar 587 keyboards 588 589 atkbd.softraw= [HW] Choose between synthetic and real raw mode 590 Format: <bool> (0 = real, 1 = synthetic (default)) 591 592 atkbd.softrepeat= [HW] 593 Use software keyboard repeat 594 595 audit= [KNL] Enable the audit sub-system 596 Format: { "0" | "1" | "off" | "on" } 597 0 | off - kernel audit is disabled and can not be 598 enabled until the next reboot 599 unset - kernel audit is initialized but disabled and 600 will be fully enabled by the userspace auditd. 601 1 | on - kernel audit is initialized and partially 602 enabled, storing at most audit_backlog_limit 603 messages in RAM until it is fully enabled by the 604 userspace auditd. 605 Default: unset 606 607 audit_backlog_limit= [KNL] Set the audit queue size limit. 608 Format: <int> (must be >=0) 609 Default: 64 610 611 bau= [X86_UV] Enable the BAU on SGI UV. The default 612 behavior is to disable the BAU (i.e. bau=0). 613 Format: { "0" | "1" } 614 0 - Disable the BAU. 615 1 - Enable the BAU. 616 unset - Disable the BAU. 617 618 baycom_epp= [HW,AX25] 619 Format: <io>,<mode> 620 621 baycom_par= [HW,AX25] BayCom Parallel Port AX.25 Modem 622 Format: <io>,<mode> 623 See header of drivers/net/hamradio/baycom_par.c. 624 625 baycom_ser_fdx= [HW,AX25] 626 BayCom Serial Port AX.25 Modem (Full Duplex Mode) 627 Format: <io>,<irq>,<mode>[,<baud>] 628 See header of drivers/net/hamradio/baycom_ser_fdx.c. 629 630 baycom_ser_hdx= [HW,AX25] 631 BayCom Serial Port AX.25 Modem (Half Duplex Mode) 632 Format: <io>,<irq>,<mode> 633 See header of drivers/net/hamradio/baycom_ser_hdx.c. 634 635 bdev_allow_write_mounted= 636 Format: <bool> 637 Control the ability to open a mounted block device 638 for writing, i.e., allow / disallow writes that bypass 639 the FS. This was implemented as a means to prevent 640 fuzzers from crashing the kernel by overwriting the 641 metadata underneath a mounted FS without its awareness. 642 This also prevents destructive formatting of mounted 643 filesystems by naive storage tooling that don't use 644 O_EXCL. Default is Y and can be changed through the 645 Kconfig option CONFIG_BLK_DEV_WRITE_MOUNTED. 646 647 bert_disable [ACPI] 648 Disable BERT OS support on buggy BIOSes. 649 650 bgrt_disable [ACPI,X86,EARLY] 651 Disable BGRT to avoid flickering OEM logo. 652 653 blkdevparts= Manual partition parsing of block device(s) for 654 embedded devices based on command line input. 655 See Documentation/block/cmdline-partition.rst 656 657 boot_delay= [KNL,EARLY] 658 Milliseconds to delay each printk during boot. 659 Only works if CONFIG_BOOT_PRINTK_DELAY is enabled, 660 and you may also have to specify "lpj=". Boot_delay 661 values larger than 10 seconds (10000) are assumed 662 erroneous and ignored. 663 Format: integer 664 665 bootconfig [KNL,EARLY] 666 Extended command line options can be added to an initrd 667 and this will cause the kernel to look for it. 668 669 See Documentation/admin-guide/bootconfig.rst 670 671 bttv.card= [HW,V4L] bttv (bt848 + bt878 based grabber cards) 672 bttv.radio= Most important insmod options are available as 673 kernel args too. 674 bttv.pll= See Documentation/admin-guide/media/bttv.rst 675 bttv.tuner= 676 677 bulk_remove=off [PPC] This parameter disables the use of the pSeries 678 firmware feature for flushing multiple hpte entries 679 at a time. 680 681 c101= [NET] Moxa C101 synchronous serial card 682 683 cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection. 684 Sometimes CPU hardware bugs make them report the cache 685 size incorrectly. The kernel will attempt work arounds 686 to fix known problems, but for some CPUs it is not 687 possible to determine what the correct size should be. 688 This option provides an override for these situations. 689 690 carrier_timeout= 691 [NET] Specifies amount of time (in seconds) that 692 the kernel should wait for a network carrier. By default 693 it waits 120 seconds. 694 695 ca_keys= [KEYS] This parameter identifies a specific key(s) on 696 the system trusted keyring to be used for certificate 697 trust validation. 698 format: { id:<keyid> | builtin } 699 700 cca= [MIPS,EARLY] Override the kernel pages' cache coherency 701 algorithm. Accepted values range from 0 to 7 702 inclusive. See arch/mips/include/asm/pgtable-bits.h 703 for platform specific values (SB1, Loongson3 and 704 others). 705 706 ccw_timeout_log [S390] 707 See Documentation/arch/s390/common_io.rst for details. 708 709 cfi= [X86-64] Set Control Flow Integrity checking features 710 when CONFIG_FINEIBT is enabled. 711 Format: feature[,feature...] 712 Default: auto 713 714 auto: Use FineIBT if IBT available, otherwise kCFI. 715 Under FineIBT, enable "paranoid" mode when 716 FRED is not available. 717 off: Turn off CFI checking. 718 kcfi: Use kCFI (disable FineIBT). 719 fineibt: Use FineIBT (even if IBT not available). 720 norand: Do not re-randomize CFI hashes. 721 paranoid: Add caller hash checking under FineIBT. 722 bhi: Enable register poisoning to stop speculation 723 across FineIBT. (Disabled by default.) 724 warn: Do not enforce CFI checking: warn only. 725 debug: Report CFI initialization details. 726 727 cgroup_disable= [KNL] Disable a particular controller or optional feature 728 Format: {name of the controller(s) or feature(s) to disable} 729 The effects of cgroup_disable=foo are: 730 - foo isn't auto-mounted if you mount all cgroups in 731 a single hierarchy 732 - foo isn't visible as an individually mountable 733 subsystem 734 - if foo is an optional feature then the feature is 735 disabled and corresponding cgroup files are not 736 created 737 {Currently only "memory" controller deal with this and 738 cut the overhead, others just disable the usage. So 739 only cgroup_disable=memory is actually worthy} 740 Specifying "pressure" disables per-cgroup pressure 741 stall information accounting feature 742 743 cgroup_no_v1= [KNL] Disable cgroup controllers and named hierarchies in v1 744 Format: { { controller | "all" | "named" } 745 [,{ controller | "all" | "named" }...] } 746 Like cgroup_disable, but only applies to cgroup v1; 747 the blacklisted controllers remain available in cgroup2. 748 "all" blacklists all controllers and "named" disables 749 named mounts. Specifying both "all" and "named" disables 750 all v1 hierarchies. 751 752 cgroup_v1_proc= [KNL] Show also missing controllers in /proc/cgroups 753 Format: { "true" | "false" } 754 /proc/cgroups lists only v1 controllers by default. 755 This compatibility option enables listing also v2 756 controllers (whose v1 code is not compiled!), so that 757 semi-legacy software can check this file to decide 758 about usage of v2 (sic) controllers. 759 760 cgroup_favordynmods= [KNL] Enable or Disable favordynmods. 761 Format: { "true" | "false" } 762 Defaults to the value of CONFIG_CGROUP_FAVOR_DYNMODS. 763 764 cgroup.memory= [KNL] Pass options to the cgroup memory controller. 765 Format: <string> 766 nosocket -- Disable socket memory accounting. 767 nokmem -- Disable kernel memory accounting. 768 nobpf -- Disable BPF memory accounting. 769 770 check_pages= [MM,EARLY] Enable sanity checking of pages after 771 allocations / before freeing. This adds checks to catch 772 double-frees, use-after-frees, and other sources of 773 page corruption by inspecting page internals (flags, 774 mapcount/refcount, memcg_data, etc.). 775 Format: { "0" | "1" } 776 Default: 0 (1 if CONFIG_DEBUG_VM is set) 777 778 checkreqprot= [SELINUX] Set initial checkreqprot flag value. 779 Format: { "0" | "1" } 780 See security/selinux/Kconfig help text. 781 0 -- check protection applied by kernel (includes 782 any implied execute protection). 783 1 -- check protection requested by application. 784 Default value is set via a kernel config option. 785 Value can be changed at runtime via 786 /sys/fs/selinux/checkreqprot. 787 Setting checkreqprot to 1 is deprecated. 788 789 cio_ignore= [S390] 790 See Documentation/arch/s390/common_io.rst for details. 791 792 clearcpuid=X[,X...] [X86] 793 Disable CPUID feature X for the kernel. See 794 arch/x86/include/asm/cpufeatures.h for the valid bit 795 numbers X. Note the Linux-specific bits are not necessarily 796 stable over kernel options, but the vendor-specific 797 ones should be. 798 X can also be a string as appearing in the flags: line 799 in /proc/cpuinfo which does not have the above 800 instability issue. However, not all features have names 801 in /proc/cpuinfo. 802 Note that using this option will taint your kernel. 803 Also note that user programs calling CPUID directly 804 or using the feature without checking anything 805 will still see it. This just prevents it from 806 being used by the kernel or shown in /proc/cpuinfo. 807 Also note the kernel might malfunction if you disable 808 some critical bits. 809 810 clk_ignore_unused 811 [CLK] 812 Prevents the clock framework from automatically gating 813 clocks that have not been explicitly enabled by a Linux 814 device driver but are enabled in hardware at reset or 815 by the bootloader/firmware. Note that this does not 816 force such clocks to be always-on nor does it reserve 817 those clocks in any way. This parameter is useful for 818 debug and development, but should not be needed on a 819 platform with proper driver support. For more 820 information, see Documentation/driver-api/clk.rst. 821 822 clock= [BUGS=X86-32, HW] gettimeofday clocksource override. 823 [Deprecated] 824 Forces specified clocksource (if available) to be used 825 when calculating gettimeofday(). If specified 826 clocksource is not available, it defaults to PIT. 827 Format: { pit | tsc | cyclone | pmtmr } 828 829 clocksource= Override the default clocksource 830 Format: <string> 831 Override the default clocksource and use the clocksource 832 with the name specified. 833 Some clocksource names to choose from, depending on 834 the platform: 835 [all] jiffies (this is the base, fallback clocksource) 836 [ACPI] acpi_pm 837 [ARM] imx_timer1,OSTS,netx_timer,mpu_timer2, 838 pxa_timer,timer3,32k_counter,timer0_1 839 [X86-32] pit,hpet,tsc; 840 scx200_hrt on Geode; cyclone on IBM x440 841 [MIPS] MIPS 842 [PARISC] cr16 843 [S390] tod 844 [SH] SuperH 845 [SPARC64] tick 846 [X86-64] hpet,tsc 847 848 clocksource.arm_arch_timer.evtstrm= 849 [ARM,ARM64,EARLY] 850 Format: <bool> 851 Enable/disable the eventstream feature of the ARM 852 architected timer so that code using WFE-based polling 853 loops can be debugged more effectively on production 854 systems. 855 856 clocksource.verify_n_cpus= [KNL] 857 Limit the number of CPUs checked for clocksources 858 marked with CLOCK_SOURCE_VERIFY_PERCPU that 859 are marked unstable due to excessive skew. 860 A negative value says to check all CPUs, while 861 zero says not to check any. Values larger than 862 nr_cpu_ids are silently truncated to nr_cpu_ids. 863 The actual CPUs are chosen randomly, with 864 no replacement if the same CPU is chosen twice. 865 866 clocksource-wdtest.holdoff= [KNL] 867 Set the time in seconds that the clocksource 868 watchdog test waits before commencing its tests. 869 Defaults to zero when built as a module and to 870 10 seconds when built into the kernel. 871 872 cma=nn[MG]@[start[MG][-end[MG]]] 873 [KNL,CMA,EARLY] 874 Sets the size of kernel global memory area for 875 contiguous memory allocations and optionally the 876 placement constraint by the physical address range of 877 memory allocations. A value of 0 disables CMA 878 altogether. For more information, see 879 kernel/dma/contiguous.c 880 881 cma_pernuma=nn[MG] 882 [KNL,CMA,EARLY] 883 Sets the size of kernel per-numa memory area for 884 contiguous memory allocations. A value of 0 disables 885 per-numa CMA altogether. And If this option is not 886 specified, the default value is 0. 887 With per-numa CMA enabled, DMA users on node nid will 888 first try to allocate buffer from the pernuma area 889 which is located in node nid, if the allocation fails, 890 they will fallback to the global default memory area. 891 892 numa_cma=<node>:nn[MG][,<node>:nn[MG]] 893 [KNL,CMA,EARLY] 894 Sets the size of kernel numa memory area for 895 contiguous memory allocations. It will reserve CMA 896 area for the specified node. 897 898 With numa CMA enabled, DMA users on node nid will 899 first try to allocate buffer from the numa area 900 which is located in node nid, if the allocation fails, 901 they will fallback to the global default memory area. 902 903 cmo_free_hint= [PPC] Format: { yes | no } 904 Specify whether pages are marked as being inactive 905 when they are freed. This is used in CMO environments 906 to determine OS memory pressure for page stealing by 907 a hypervisor. 908 Default: yes 909 910 coherent_pool=nn[KMG] [ARM,KNL,EARLY] 911 Sets the size of memory pool for coherent, atomic dma 912 allocations, by default set to 256K. 913 914 com20020= [HW,NET] ARCnet - COM20020 chipset 915 Format: 916 <io>[,<irq>[,<nodeID>[,<backplane>[,<ckp>[,<timeout>]]]]] 917 918 com90io= [HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers) 919 Format: <io>[,<irq>] 920 921 com90xx= [HW,NET] 922 ARCnet - COM90xx chipset (memory-mapped buffers) 923 Format: <io>[,<irq>[,<memstart>]] 924 925 condev= [HW,S390] console device 926 conmode= 927 928 con3215_drop= [S390,EARLY] 3215 console drop mode. 929 Format: y|n|Y|N|1|0 930 When set to true, drop data on the 3215 console when 931 the console buffer is full. In this case the 932 operator using a 3270 terminal emulator (for example 933 x3270) does not have to enter the clear key for the 934 console output to advance and the kernel to continue. 935 This leads to a much faster boot time when a 3270 936 terminal emulator is active. If no 3270 terminal 937 emulator is used, this parameter has no effect. 938 939 console= [KNL] Output console device and options. 940 941 tty<n> Use the virtual console device <n>. 942 943 ttyS<n>[,options] 944 ttyUSB0[,options] 945 Use the specified serial port. The options are of 946 the form "bbbbpnf", where "bbbb" is the baud rate, 947 "p" is parity ("n", "o", or "e"), "n" is number of 948 bits, and "f" is flow control ("r" for RTS or 949 omit it). Default is "9600n8". 950 951 See Documentation/admin-guide/serial-console.rst for more 952 information. See 953 Documentation/networking/netconsole.rst for an 954 alternative. 955 956 <DEVNAME>:<n>.<n>[,options] 957 Use the specified serial port on the serial core bus. 958 The addressing uses DEVNAME of the physical serial port 959 device, followed by the serial core controller instance, 960 and the serial port instance. The options are the same 961 as documented for the ttyS addressing above. 962 963 The mapping of the serial ports to the tty instances 964 can be viewed with: 965 966 $ ls -d /sys/bus/serial-base/devices/*:*.*/tty/* 967 /sys/bus/serial-base/devices/00:04:0.0/tty/ttyS0 968 969 In the above example, the console can be addressed with 970 console=00:04:0.0. Note that a console addressed this 971 way will only get added when the related device driver 972 is ready. The use of an earlycon parameter in addition to 973 the console may be desired for console output early on. 974 975 uart[8250],io,<addr>[,options] 976 uart[8250],mmio,<addr>[,options] 977 uart[8250],mmio16,<addr>[,options] 978 uart[8250],mmio32,<addr>[,options] 979 uart[8250],0x<addr>[,options] 980 Start an early, polled-mode console on the 8250/16550 981 UART at the specified I/O port or MMIO address, 982 switching to the matching ttyS device later. 983 MMIO inter-register address stride is either 8-bit 984 (mmio), 16-bit (mmio16), or 32-bit (mmio32). 985 If none of [io|mmio|mmio16|mmio32], <addr> is assumed 986 to be equivalent to 'mmio'. 'options' are specified in 987 the same format described for ttyS above; if unspecified, 988 the h/w is not re-initialized. 989 990 hvc<n> Use the hypervisor console device <n>. This is for 991 both Xen and PowerPC hypervisors. 992 993 { null | "" } 994 Use to disable console output, i.e., to have kernel 995 console messages discarded. 996 This must be the only console= parameter used on the 997 kernel command line. 998 999 If the device connected to the port is not a TTY but a braille 1000 device, prepend "brl," before the device type, for instance 1001 console=brl,ttyS0 1002 For now, only VisioBraille is supported. 1003 1004 console_msg_format= 1005 [KNL] Change console messages format 1006 default 1007 By default we print messages on consoles in 1008 "[time stamp] text\n" format (time stamp may not be 1009 printed, depending on CONFIG_PRINTK_TIME or 1010 `printk_time' param). 1011 syslog 1012 Switch to syslog format: "<%u>[time stamp] text\n" 1013 IOW, each message will have a facility and loglevel 1014 prefix. The format is similar to one used by syslog() 1015 syscall, or to executing "dmesg -S --raw" or to reading 1016 from /proc/kmsg. 1017 1018 consoleblank= [KNL] The console blank (screen saver) timeout in 1019 seconds. A value of 0 disables the blank timer. 1020 Defaults to 0. 1021 1022 coredump_filter= 1023 [KNL] Change the default value for 1024 /proc/<pid>/coredump_filter. 1025 See also Documentation/filesystems/proc.rst. 1026 1027 coresight_cpu_debug.enable 1028 [ARM,ARM64] 1029 Format: <bool> 1030 Enable/disable the CPU sampling based debugging. 1031 0: default value, disable debugging 1032 1: enable debugging at boot time 1033 1034 cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver 1035 Format: 1036 <first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>] 1037 1038 cpuidle.off=1 [CPU_IDLE] 1039 disable the cpuidle sub-system 1040 1041 cpuidle.governor= 1042 [CPU_IDLE] Name of the cpuidle governor to use. 1043 1044 cpufreq.off=1 [CPU_FREQ] 1045 disable the cpufreq sub-system 1046 1047 cpufreq.default_governor= 1048 [CPU_FREQ] Name of the default cpufreq governor or 1049 policy to use. This governor must be registered in the 1050 kernel before the cpufreq driver probes. 1051 1052 cpu_init_udelay=N 1053 [X86,EARLY] Delay for N microsec between assert and de-assert 1054 of APIC INIT to start processors. This delay occurs 1055 on every CPU online, such as boot, and resume from suspend. 1056 Default: 10000 1057 1058 cpuhp.parallel= 1059 [SMP] Enable/disable parallel bringup of secondary CPUs 1060 Format: <bool> 1061 Default is enabled if CONFIG_HOTPLUG_PARALLEL=y. Otherwise 1062 the parameter has no effect. 1063 1064 crash_kexec_post_notifiers 1065 Only jump to kdump kernel after running the panic 1066 notifiers and dumping kmsg. This option increases 1067 the risks of a kdump failure, since some panic 1068 notifiers can make the crashed kernel more unstable. 1069 In configurations where kdump may not be reliable, 1070 running the panic notifiers could allow collecting 1071 more data on dmesg, like stack traces from other CPUS 1072 or extra data dumped by panic_print. Note that some 1073 configurations enable this option unconditionally, 1074 like Hyper-V, PowerPC (fadump) and AMD SEV-SNP. 1075 1076 crashkernel=size[KMG][@offset[KMG]] 1077 [KNL,EARLY] Using kexec, Linux can switch to a 'crash kernel' 1078 upon panic. This parameter reserves the physical 1079 memory region [offset, offset + size] for that kernel 1080 image. If '@offset' is omitted, then a suitable offset 1081 is selected automatically. 1082 [KNL, X86-64, ARM64, RISCV, LoongArch] Select a region 1083 under 4G first, and fall back to reserve region above 1084 4G when '@offset' hasn't been specified. 1085 See Documentation/admin-guide/kdump/kdump.rst for further details. 1086 1087 crashkernel=range1:size1[,range2:size2,...][@offset] 1088 [KNL] Same as above, but depends on the memory 1089 in the running system. The syntax of range is 1090 start-[end] where start and end are both 1091 a memory unit (amount[KMG]). See also 1092 Documentation/admin-guide/kdump/kdump.rst for an example. 1093 1094 crashkernel=size[KMG],high 1095 [KNL, X86-64, ARM64, RISCV, LoongArch] range could be 1096 above 4G. 1097 Allow kernel to allocate physical memory region from top, 1098 so could be above 4G if system have more than 4G ram 1099 installed. Otherwise memory region will be allocated 1100 below 4G, if available. 1101 It will be ignored if crashkernel=X is specified. 1102 crashkernel=size[KMG],low 1103 [KNL, X86-64, ARM64, RISCV, LoongArch] range under 4G. 1104 When crashkernel=X,high is passed, kernel could allocate 1105 physical memory region above 4G, that cause second kernel 1106 crash on system that require some amount of low memory, 1107 e.g. swiotlb requires at least 64M+32K low memory, also 1108 enough extra low memory is needed to make sure DMA buffers 1109 for 32-bit devices won't run out. Kernel would try to allocate 1110 default size of memory below 4G automatically. The default 1111 size is platform dependent. 1112 --> x86: max(swiotlb_size_or_default() + 8MiB, 256MiB) 1113 --> arm64: 128MiB 1114 --> riscv: 128MiB 1115 --> loongarch: 128MiB 1116 This one lets the user specify own low range under 4G 1117 for second kernel instead. 1118 0: to disable low allocation. 1119 It will be ignored when crashkernel=X,high is not used 1120 or memory reserved is below 4G. 1121 crashkernel=size[KMG],cma 1122 [KNL, X86, ppc] Reserve additional crash kernel memory from 1123 CMA. This reservation is usable by the first system's 1124 userspace memory and kernel movable allocations (memory 1125 balloon, zswap). Pages allocated from this memory range 1126 will not be included in the vmcore so this should not 1127 be used if dumping of userspace memory is intended and 1128 it has to be expected that some movable kernel pages 1129 may be missing from the dump. 1130 1131 A standard crashkernel reservation, as described above, 1132 is still needed to hold the crash kernel and initrd. 1133 1134 This option increases the risk of a kdump failure: DMA 1135 transfers configured by the first kernel may end up 1136 corrupting the second kernel's memory. 1137 1138 This reservation method is intended for systems that 1139 can't afford to sacrifice enough memory for standard 1140 crashkernel reservation and where less reliable and 1141 possibly incomplete kdump is preferable to no kdump at 1142 all. 1143 1144 cryptomgr.notests 1145 [KNL] Disable crypto self-tests 1146 1147 cs89x0_dma= [HW,NET] 1148 Format: <dma> 1149 1150 cs89x0_media= [HW,NET] 1151 Format: { rj45 | aui | bnc } 1152 1153 csdlock_debug= [KNL] Enable or disable debug add-ons of cross-CPU 1154 function call handling. When switched on, 1155 additional debug data is printed to the console 1156 in case a hanging CPU is detected, and that 1157 CPU is pinged again in order to try to resolve 1158 the hang situation. The default value of this 1159 option depends on the CSD_LOCK_WAIT_DEBUG_DEFAULT 1160 Kconfig option. 1161 1162 dasd= [HW,NET] 1163 See header of drivers/s390/block/dasd_devmap.c. 1164 1165 db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port 1166 (one device per port) 1167 Format: <port#>,<type> 1168 See also Documentation/input/devices/joystick-parport.rst 1169 1170 debug [KNL,EARLY] Enable kernel debugging (events log level). 1171 1172 debug_boot_weak_hash 1173 [KNL,EARLY] Enable printing [hashed] pointers early in the 1174 boot sequence. If enabled, we use a weak hash instead 1175 of siphash to hash pointers. Use this option if you are 1176 seeing instances of '(___ptrval___)') and need to see a 1177 value (hashed pointer) instead. Cryptographically 1178 insecure, please do not use on production kernels. 1179 1180 debug_locks_verbose= 1181 [KNL] verbose locking self-tests 1182 Format: <int> 1183 Print debugging info while doing the locking API 1184 self-tests. 1185 Bitmask for the various LOCKTYPE_ tests. Defaults to 0 1186 (no extra messages), setting it to -1 (all bits set) 1187 will print _a_lot_ more information - normally only 1188 useful to lockdep developers. 1189 1190 debug_objects [KNL,EARLY] Enable object debugging 1191 1192 debug_guardpage_minorder= 1193 [KNL,EARLY] When CONFIG_DEBUG_PAGEALLOC is set, this 1194 parameter allows control of the order of pages that will 1195 be intentionally kept free (and hence protected) by the 1196 buddy allocator. Bigger value increase the probability 1197 of catching random memory corruption, but reduce the 1198 amount of memory for normal system use. The maximum 1199 possible value is MAX_PAGE_ORDER/2. Setting this 1200 parameter to 1 or 2 should be enough to identify most 1201 random memory corruption problems caused by bugs in 1202 kernel or driver code when a CPU writes to (or reads 1203 from) a random memory location. Note that there exists 1204 a class of memory corruptions problems caused by buggy 1205 H/W or F/W or by drivers badly programming DMA 1206 (basically when memory is written at bus level and the 1207 CPU MMU is bypassed) which are not detectable by 1208 CONFIG_DEBUG_PAGEALLOC, hence this option will not 1209 help tracking down these problems. 1210 1211 debug_pagealloc= 1212 [KNL,EARLY] When CONFIG_DEBUG_PAGEALLOC is set, this parameter 1213 enables the feature at boot time. By default, it is 1214 disabled and the system will work mostly the same as a 1215 kernel built without CONFIG_DEBUG_PAGEALLOC. 1216 Note: to get most of debug_pagealloc error reports, it's 1217 useful to also enable the page_owner functionality. 1218 on: enable the feature 1219 1220 debugfs= [KNL,EARLY] This parameter enables what is exposed to 1221 userspace and debugfs internal clients. 1222 Format: { on, off } 1223 on: All functions are enabled. 1224 off: Filesystem is not registered and clients 1225 get a -EPERM as result when trying to register files 1226 or directories within debugfs. 1227 This is equivalent of the runtime functionality if 1228 debugfs was not enabled in the kernel at all. 1229 Default value is set in build-time with a kernel configuration. 1230 1231 debugpat [X86] Enable PAT debugging 1232 1233 default_hugepagesz= 1234 [HW] The size of the default HugeTLB page. This is 1235 the size represented by the legacy /proc/ hugepages 1236 APIs. In addition, this is the default hugetlb size 1237 used for shmget(), mmap() and mounting hugetlbfs 1238 filesystems. If not specified, defaults to the 1239 architecture's default huge page size. Huge page 1240 sizes are architecture dependent. See also 1241 Documentation/admin-guide/mm/hugetlbpage.rst. 1242 Format: size[KMG] 1243 1244 deferred_probe_timeout= 1245 [KNL] Debugging option to set a timeout in seconds for 1246 deferred probe to give up waiting on dependencies to 1247 probe. Only specific dependencies (subsystems or 1248 drivers) that have opted in will be ignored. A timeout 1249 of 0 will timeout at the end of initcalls. If the time 1250 out hasn't expired, it'll be restarted by each 1251 successful driver registration. This option will also 1252 dump out devices still on the deferred probe list after 1253 retrying. 1254 1255 delayacct [KNL] Enable per-task delay accounting 1256 1257 dell_smm_hwmon.ignore_dmi= 1258 [HW] Continue probing hardware even if DMI data 1259 indicates that the driver is running on unsupported 1260 hardware. 1261 1262 dell_smm_hwmon.force= 1263 [HW] Activate driver even if SMM BIOS signature does 1264 not match list of supported models and enable otherwise 1265 blacklisted features. 1266 1267 dell_smm_hwmon.power_status= 1268 [HW] Report power status in /proc/i8k 1269 (disabled by default). 1270 1271 dell_smm_hwmon.restricted= 1272 [HW] Allow controlling fans only if SYS_ADMIN 1273 capability is set. 1274 1275 dell_smm_hwmon.fan_mult= 1276 [HW] Factor to multiply fan speed with. 1277 1278 dell_smm_hwmon.fan_max= 1279 [HW] Maximum configurable fan speed. 1280 1281 dfltcc= [HW,S390] 1282 Format: { on | off | def_only | inf_only | always } 1283 on: s390 zlib hardware support for compression on 1284 level 1 and decompression (default) 1285 off: No s390 zlib hardware support 1286 def_only: s390 zlib hardware support for deflate 1287 only (compression on level 1) 1288 inf_only: s390 zlib hardware support for inflate 1289 only (decompression) 1290 always: Same as 'on' but ignores the selected compression 1291 level always using hardware support (used for debugging) 1292 1293 dhash_entries= [KNL] 1294 Set number of hash buckets for dentry cache. 1295 1296 disable_1tb_segments [PPC,EARLY] 1297 Disables the use of 1TB hash page table segments. This 1298 causes the kernel to fall back to 256MB segments which 1299 can be useful when debugging issues that require an SLB 1300 miss to occur. 1301 1302 disable= [IPV6] 1303 See Documentation/networking/ipv6.rst. 1304 1305 disable_radix [PPC,EARLY] 1306 Disable RADIX MMU mode on POWER9 1307 1308 disable_tlbie [PPC] 1309 Disable TLBIE instruction. Currently does not work 1310 with KVM, with HASH MMU, or with coherent accelerators. 1311 1312 disable_ddw [PPC/PSERIES,EARLY] 1313 Disable Dynamic DMA Window support. Use this 1314 to workaround buggy firmware. 1315 1316 disable_ipv6= [IPV6] 1317 See Documentation/networking/ipv6.rst. 1318 1319 disable_mtrr_cleanup [X86,EARLY] 1320 The kernel tries to adjust MTRR layout from continuous 1321 to discrete, to make X server driver able to add WB 1322 entry later. This parameter disables that. 1323 1324 disable_mtrr_trim [X86, Intel and AMD only,EARLY] 1325 By default the kernel will trim any uncacheable 1326 memory out of your available memory pool based on 1327 MTRR settings. This parameter disables that behavior, 1328 possibly causing your machine to run very slowly. 1329 1330 disable_timer_pin_1 [X86,EARLY] 1331 Disable PIN 1 of APIC timer 1332 Can be useful to work around chipset bugs. 1333 1334 dis_ucode_ldr [X86] Disable the microcode loader. 1335 1336 dma_debug=off If the kernel is compiled with DMA_API_DEBUG support, 1337 this option disables the debugging code at boot. 1338 1339 dma_debug_entries=<number> 1340 This option allows to tune the number of preallocated 1341 entries for DMA-API debugging code. One entry is 1342 required per DMA-API allocation. Use this if the 1343 DMA-API debugging code disables itself because the 1344 architectural default is too low. 1345 1346 dma_debug_driver=<driver_name> 1347 With this option the DMA-API debugging driver 1348 filter feature can be enabled at boot time. Just 1349 pass the driver to filter for as the parameter. 1350 The filter can be disabled or changed to another 1351 driver later using sysfs. 1352 1353 reg_file_data_sampling= 1354 [X86] Controls mitigation for Register File Data 1355 Sampling (RFDS) vulnerability. RFDS is a CPU 1356 vulnerability which may allow userspace to infer 1357 kernel data values previously stored in floating point 1358 registers, vector registers, or integer registers. 1359 RFDS only affects Intel Atom processors. 1360 1361 on: Turns ON the mitigation. 1362 off: Turns OFF the mitigation. 1363 1364 This parameter overrides the compile time default set 1365 by CONFIG_MITIGATION_RFDS. Mitigation cannot be 1366 disabled when other VERW based mitigations (like MDS) 1367 are enabled. In order to disable RFDS mitigation all 1368 VERW based mitigations need to be disabled. 1369 1370 For details see: 1371 Documentation/admin-guide/hw-vuln/reg-file-data-sampling.rst 1372 1373 dm_verity.keyring_unsealed= 1374 [KNL] When set to 1, leave the dm-verity keyring 1375 unsealed after initialization so userspace can 1376 provision keys. Once the keyring is restricted 1377 it becomes active and is searched during signature 1378 verification. 1379 1380 driver_async_probe= [KNL] 1381 List of driver names to be probed asynchronously. * 1382 matches with all driver names. If * is specified, the 1383 rest of the listed driver names are those that will NOT 1384 match the *. 1385 Format: <driver_name1>,<driver_name2>... 1386 1387 drm.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>] 1388 Broken monitors, graphic adapters, KVMs and EDIDless 1389 panels may send no or incorrect EDID data sets. 1390 This parameter allows to specify an EDID data sets 1391 in the /lib/firmware directory that are used instead. 1392 An EDID data set will only be used for a particular 1393 connector, if its name and a colon are prepended to 1394 the EDID name. Each connector may use a unique EDID 1395 data set by separating the files with a comma. An EDID 1396 data set with no connector name will be used for 1397 any connectors not explicitly specified. 1398 1399 dscc4.setup= [NET] 1400 1401 dt_cpu_ftrs= [PPC,EARLY] 1402 Format: {"off" | "known"} 1403 Control how the dt_cpu_ftrs device-tree binding is 1404 used for CPU feature discovery and setup (if it 1405 exists). 1406 off: Do not use it, fall back to legacy cpu table. 1407 known: Do not pass through unknown features to guests 1408 or userspace, only those that the kernel is aware of. 1409 1410 dump_apple_properties [X86] 1411 Dump name and content of EFI device properties on 1412 x86 Macs. Useful for driver authors to determine 1413 what data is available or for reverse-engineering. 1414 1415 dyndbg[="val"] [KNL,DYNAMIC_DEBUG] 1416 <module>.dyndbg[="val"] 1417 Enable debug messages at boot time. See 1418 Documentation/admin-guide/dynamic-debug-howto.rst 1419 for details. 1420 1421 early_ioremap_debug [KNL,EARLY] 1422 Enable debug messages in early_ioremap support. This 1423 is useful for tracking down temporary early mappings 1424 which are not unmapped. 1425 1426 earlycon= [KNL,EARLY] Output early console device and options. 1427 1428 When used with no options, the early console is 1429 determined by stdout-path property in device tree's 1430 chosen node or the ACPI SPCR table if supported by 1431 the platform. 1432 1433 cdns,<addr>[,options] 1434 Start an early, polled-mode console on a Cadence 1435 (xuartps) serial port at the specified address. Only 1436 supported option is baud rate. If baud rate is not 1437 specified, the serial port must already be setup and 1438 configured. 1439 1440 uart[8250],io,<addr>[,options[,uartclk]] 1441 uart[8250],mmio,<addr>[,options[,uartclk]] 1442 uart[8250],mmio32,<addr>[,options[,uartclk]] 1443 uart[8250],mmio32be,<addr>[,options[,uartclk]] 1444 uart[8250],0x<addr>[,options] 1445 Start an early, polled-mode console on the 8250/16550 1446 UART at the specified I/O port or MMIO address. 1447 MMIO inter-register address stride is either 8-bit 1448 (mmio) or 32-bit (mmio32 or mmio32be). 1449 If none of [io|mmio|mmio32|mmio32be], <addr> is assumed 1450 to be equivalent to 'mmio'. 'options' are specified 1451 in the same format described for "console=ttyS<n>"; if 1452 unspecified, the h/w is not initialized. 'uartclk' is 1453 the uart clock frequency; if unspecified, it is set 1454 to 'BASE_BAUD' * 16. 1455 1456 pl011,<addr> 1457 pl011,mmio32,<addr> 1458 Start an early, polled-mode console on a pl011 serial 1459 port at the specified address. The pl011 serial port 1460 must already be setup and configured. Options are not 1461 yet supported. If 'mmio32' is specified, then only 1462 the driver will use only 32-bit accessors to read/write 1463 the device registers. 1464 1465 liteuart,<addr> 1466 Start an early console on a litex serial port at the 1467 specified address. The serial port must already be 1468 setup and configured. Options are not yet supported. 1469 1470 meson,<addr> 1471 Start an early, polled-mode console on a meson serial 1472 port at the specified address. The serial port must 1473 already be setup and configured. Options are not yet 1474 supported. 1475 1476 msm_serial,<addr> 1477 Start an early, polled-mode console on an msm serial 1478 port at the specified address. The serial port 1479 must already be setup and configured. Options are not 1480 yet supported. 1481 1482 msm_serial_dm,<addr> 1483 Start an early, polled-mode console on an msm serial 1484 dm port at the specified address. The serial port 1485 must already be setup and configured. Options are not 1486 yet supported. 1487 1488 owl,<addr> 1489 Start an early, polled-mode console on a serial port 1490 of an Actions Semi SoC, such as S500 or S900, at the 1491 specified address. The serial port must already be 1492 setup and configured. Options are not yet supported. 1493 1494 rda,<addr> 1495 Start an early, polled-mode console on a serial port 1496 of an RDA Micro SoC, such as RDA8810PL, at the 1497 specified address. The serial port must already be 1498 setup and configured. Options are not yet supported. 1499 1500 sbi 1501 Use RISC-V SBI (Supervisor Binary Interface) for early 1502 console. 1503 1504 smh Use ARM semihosting calls for early console. 1505 1506 s3c2410,<addr> 1507 s3c2412,<addr> 1508 s3c2440,<addr> 1509 s3c6400,<addr> 1510 s5pv210,<addr> 1511 exynos4210,<addr> 1512 Use early console provided by serial driver available 1513 on Samsung SoCs, requires selecting proper type and 1514 a correct base address of the selected UART port. The 1515 serial port must already be setup and configured. 1516 Options are not yet supported. 1517 1518 lantiq,<addr> 1519 Start an early, polled-mode console on a lantiq serial 1520 (lqasc) port at the specified address. The serial port 1521 must already be setup and configured. Options are not 1522 yet supported. 1523 1524 lpuart,<addr> 1525 lpuart32,<addr> 1526 Use early console provided by Freescale LP UART driver 1527 found on Freescale Vybrid and QorIQ LS1021A processors. 1528 A valid base address must be provided, and the serial 1529 port must already be setup and configured. 1530 1531 ec_imx21,<addr> 1532 ec_imx6q,<addr> 1533 Start an early, polled-mode, output-only console on the 1534 Freescale i.MX UART at the specified address. The UART 1535 must already be setup and configured. 1536 1537 ar3700_uart,<addr> 1538 Start an early, polled-mode console on the 1539 Armada 3700 serial port at the specified 1540 address. The serial port must already be setup 1541 and configured. Options are not yet supported. 1542 1543 qcom_geni,<addr> 1544 Start an early, polled-mode console on a Qualcomm 1545 Generic Interface (GENI) based serial port at the 1546 specified address. The serial port must already be 1547 setup and configured. Options are not yet supported. 1548 1549 efifb,[options] 1550 Start an early, unaccelerated console on the EFI 1551 memory mapped framebuffer (if available). On cache 1552 coherent non-x86 systems that use system memory for 1553 the framebuffer, pass the 'ram' option so that it is 1554 mapped with the correct attributes. 1555 1556 linflex,<addr> 1557 Use early console provided by Freescale LINFlexD UART 1558 serial driver for NXP S32V234 SoCs. A valid base 1559 address must be provided, and the serial port must 1560 already be setup and configured. 1561 1562 earlyprintk= [X86,SH,ARM,M68k,S390,UM,EARLY] 1563 earlyprintk=vga 1564 earlyprintk=sclp 1565 earlyprintk=xen 1566 earlyprintk=serial[,ttySn[,baudrate]] 1567 earlyprintk=serial[,0x...[,baudrate]] 1568 earlyprintk=ttySn[,baudrate] 1569 earlyprintk=dbgp[debugController#] 1570 earlyprintk=mmio32,membase[,{nocfg|baudrate}] 1571 earlyprintk=pciserial[,force],bus:device.function[,{nocfg|baudrate}] 1572 earlyprintk=xdbc[xhciController#] 1573 earlyprintk=bios 1574 1575 earlyprintk is useful when the kernel crashes before 1576 the normal console is initialized. It is not enabled by 1577 default because it has some cosmetic problems. 1578 1579 Use "nocfg" to skip UART configuration, assume 1580 BIOS/firmware has configured UART correctly. 1581 1582 Append ",keep" to not disable it when the real console 1583 takes over. 1584 1585 Only one of vga, serial, or usb debug port can 1586 be used at a time. 1587 1588 Currently only ttyS0 and ttyS1 may be specified by 1589 name. Other I/O ports may be explicitly specified 1590 on some architectures (x86 and arm at least) by 1591 replacing ttySn with an I/O port address, like this: 1592 earlyprintk=serial,0x1008,115200 1593 You can find the port for a given device in 1594 /proc/tty/driver/serial: 1595 2: uart:ST16650V2 port:00001008 irq:18 ... 1596 1597 Interaction with the standard serial driver is not 1598 very good. 1599 1600 The VGA output is eventually overwritten by 1601 the real console. 1602 1603 The xen option can only be used in Xen domains. 1604 1605 The sclp output can only be used on s390. 1606 1607 The bios output can only be used on SuperH. 1608 1609 The optional "force" to "pciserial" enables use of a 1610 PCI device even when its classcode is not of the 1611 UART class. 1612 1613 edac_report= [HW,EDAC] Control how to report EDAC event 1614 Format: {"on" | "off" | "force"} 1615 on: enable EDAC to report H/W event. May be overridden 1616 by other higher priority error reporting module. 1617 off: disable H/W event reporting through EDAC. 1618 force: enforce the use of EDAC to report H/W event. 1619 default: on. 1620 1621 edd= [EDD] 1622 Format: {"off" | "on" | "skip[mbr]"} 1623 1624 efi= [EFI,EARLY] 1625 Format: { "debug", "disable_early_pci_dma", 1626 "nochunk", "noruntime", "nosoftreserve", 1627 "novamap", "no_disable_early_pci_dma" } 1628 debug: enable misc debug output. 1629 disable_early_pci_dma: disable the busmaster bit on all 1630 PCI bridges while in the EFI boot stub. 1631 nochunk: disable reading files in "chunks" in the EFI 1632 boot stub, as chunking can cause problems with some 1633 firmware implementations. 1634 noruntime : disable EFI runtime services support 1635 nosoftreserve: The EFI_MEMORY_SP (Specific Purpose) 1636 attribute may cause the kernel to reserve the 1637 memory range for a memory mapping driver to 1638 claim. Specify efi=nosoftreserve to disable this 1639 reservation and treat the memory by its base type 1640 (i.e. EFI_CONVENTIONAL_MEMORY / "System RAM"). 1641 novamap: do not call SetVirtualAddressMap(). 1642 no_disable_early_pci_dma: Leave the busmaster bit set 1643 on all PCI bridges while in the EFI boot stub 1644 1645 efi_no_storage_paranoia [EFI,X86,EARLY] 1646 Using this parameter you can use more than 50% of 1647 your efi variable storage. Use this parameter only if 1648 you are really sure that your UEFI does sane gc and 1649 fulfills the spec otherwise your board may brick. 1650 1651 efivar_ssdt= [EFI; X86] Name of an EFI variable that contains an SSDT 1652 that is to be dynamically loaded by Linux. If there are 1653 multiple variables with the same name but with different 1654 vendor GUIDs, all of them will be loaded. See 1655 Documentation/admin-guide/acpi/ssdt-overlays.rst for details. 1656 1657 1658 eisa_irq_edge= [PARISC,HW] 1659 See header of drivers/parisc/eisa.c. 1660 1661 ekgdboc= [X86,KGDB,EARLY] Allow early kernel console debugging 1662 Format: ekgdboc=kbd 1663 1664 This is designed to be used in conjunction with 1665 the boot argument: earlyprintk=vga 1666 1667 This parameter works in place of the kgdboc parameter 1668 but can only be used if the backing tty is available 1669 very early in the boot process. For early debugging 1670 via a serial port see kgdboc_earlycon instead. 1671 1672 elanfreq= [X86-32] 1673 See comment before function elanfreq_setup() in 1674 arch/x86/kernel/cpu/cpufreq/elanfreq.c. 1675 1676 elfcorehdr=[size[KMG]@]offset[KMG] [PPC,SH,X86,S390,EARLY] 1677 Specifies physical address of start of kernel core 1678 image elf header and optionally the size. Generally 1679 kexec loader will pass this option to capture kernel. 1680 See Documentation/admin-guide/kdump/kdump.rst for details. 1681 1682 enable_mtrr_cleanup [X86,EARLY] 1683 The kernel tries to adjust MTRR layout from continuous 1684 to discrete, to make X server driver able to add WB 1685 entry later. This parameter enables that. 1686 1687 enable_timer_pin_1 [X86] 1688 Enable PIN 1 of APIC timer 1689 Can be useful to work around chipset bugs 1690 (in particular on some ATI chipsets). 1691 The kernel tries to set a reasonable default. 1692 1693 enforcing= [SELINUX] Set initial enforcing status. 1694 Format: {"0" | "1"} 1695 See security/selinux/Kconfig help text. 1696 0 -- permissive (log only, no denials). 1697 1 -- enforcing (deny and log). 1698 Default value is 0. 1699 Value can be changed at runtime via 1700 /sys/fs/selinux/enforce. 1701 1702 erst_disable [ACPI] 1703 Disable Error Record Serialization Table (ERST) 1704 support. 1705 1706 ether= [HW,NET] Ethernet cards parameters 1707 This option is obsoleted by the "netdev=" option, which 1708 has equivalent usage. See its documentation for details. 1709 1710 evm= [EVM] 1711 Format: { "fix" } 1712 Permit 'security.evm' to be updated regardless of 1713 current integrity status. 1714 1715 early_page_ext [KNL,EARLY] Enforces page_ext initialization to earlier 1716 stages so cover more early boot allocations. 1717 Please note that as side effect some optimizations 1718 might be disabled to achieve that (e.g. parallelized 1719 memory initialization is disabled) so the boot process 1720 might take longer, especially on systems with a lot of 1721 memory. Available with CONFIG_PAGE_EXTENSION=y. 1722 1723 failslab= 1724 fail_usercopy= 1725 fail_page_alloc= 1726 fail_skb_realloc= 1727 fail_make_request=[KNL] 1728 General fault injection mechanism. 1729 Format: <interval>,<probability>,<space>,<times> 1730 See also Documentation/fault-injection/. 1731 1732 fb_tunnels= [NET] 1733 Format: { initns | none } 1734 See Documentation/admin-guide/sysctl/net.rst for 1735 fb_tunnels_only_for_init_ns 1736 1737 floppy= [HW] 1738 See Documentation/admin-guide/blockdev/floppy.rst. 1739 1740 forcepae [X86-32] 1741 Forcefully enable Physical Address Extension (PAE). 1742 Many Pentium M systems disable PAE but may have a 1743 functionally usable PAE implementation. 1744 Warning: use of this parameter will taint the kernel 1745 and may cause unknown problems. 1746 1747 fred= [X86-64] 1748 Enable/disable Flexible Return and Event Delivery. 1749 Format: { on | off } 1750 on: enable FRED when it's present. 1751 off: disable FRED, the default setting. 1752 1753 ftrace=[tracer] 1754 [FTRACE] will set and start the specified tracer 1755 as early as possible in order to facilitate early 1756 boot debugging. 1757 1758 ftrace_boot_snapshot 1759 [FTRACE] On boot up, a snapshot will be taken of the 1760 ftrace ring buffer that can be read at: 1761 /sys/kernel/tracing/snapshot. 1762 This is useful if you need tracing information from kernel 1763 boot up that is likely to be overridden by user space 1764 start up functionality. 1765 1766 Optionally, the snapshot can also be defined for a tracing 1767 instance that was created by the trace_instance= command 1768 line parameter. 1769 1770 trace_instance=foo,sched_switch ftrace_boot_snapshot=foo 1771 1772 The above will cause the "foo" tracing instance to trigger 1773 a snapshot at the end of boot up. 1774 1775 ftrace_dump_on_oops[=2(orig_cpu) | =<instance>][,<instance> | 1776 ,<instance>=2(orig_cpu)] 1777 [FTRACE] will dump the trace buffers on oops. 1778 If no parameter is passed, ftrace will dump global 1779 buffers of all CPUs, if you pass 2 or orig_cpu, it 1780 will dump only the buffer of the CPU that triggered 1781 the oops, or the specific instance will be dumped if 1782 its name is passed. Multiple instance dump is also 1783 supported, and instances are separated by commas. Each 1784 instance supports only dump on CPU that triggered the 1785 oops by passing 2 or orig_cpu to it. 1786 1787 ftrace_dump_on_oops=foo=orig_cpu 1788 1789 The above will dump only the buffer of "foo" instance 1790 on CPU that triggered the oops. 1791 1792 ftrace_dump_on_oops,foo,bar=orig_cpu 1793 1794 The above will dump global buffer on all CPUs, the 1795 buffer of "foo" instance on all CPUs and the buffer 1796 of "bar" instance on CPU that triggered the oops. 1797 1798 ftrace_filter=[function-list] 1799 [FTRACE] Limit the functions traced by the function 1800 tracer at boot up. function-list is a comma-separated 1801 list of functions. This list can be changed at run 1802 time by the set_ftrace_filter file in the debugfs 1803 tracing directory. 1804 1805 ftrace_notrace=[function-list] 1806 [FTRACE] Do not trace the functions specified in 1807 function-list. This list can be changed at run time 1808 by the set_ftrace_notrace file in the debugfs 1809 tracing directory. 1810 1811 ftrace_graph_filter=[function-list] 1812 [FTRACE] Limit the top level callers functions traced 1813 by the function graph tracer at boot up. 1814 function-list is a comma-separated list of functions 1815 that can be changed at run time by the 1816 set_graph_function file in the debugfs tracing directory. 1817 1818 ftrace_graph_notrace=[function-list] 1819 [FTRACE] Do not trace from the functions specified in 1820 function-list. This list is a comma-separated list of 1821 functions that can be changed at run time by the 1822 set_graph_notrace file in the debugfs tracing directory. 1823 1824 ftrace_graph_max_depth=<uint> 1825 [FTRACE] Used with the function graph tracer. This is 1826 the max depth it will trace into a function. This value 1827 can be changed at run time by the max_graph_depth file 1828 in the tracefs tracing directory. default: 0 (no limit) 1829 1830 fw_devlink= [KNL,EARLY] Create device links between consumer and supplier 1831 devices by scanning the firmware to infer the 1832 consumer/supplier relationships. This feature is 1833 especially useful when drivers are loaded as modules as 1834 it ensures proper ordering of tasks like device probing 1835 (suppliers first, then consumers), supplier boot state 1836 clean up (only after all consumers have probed), 1837 suspend/resume & runtime PM (consumers first, then 1838 suppliers). 1839 Format: { off | permissive | on | rpm } 1840 off -- Don't create device links from firmware info. 1841 permissive -- Create device links from firmware info 1842 but use it only for ordering boot state clean 1843 up (sync_state() calls). 1844 on -- Create device links from firmware info and use it 1845 to enforce probe and suspend/resume ordering. 1846 rpm -- Like "on", but also use to order runtime PM. 1847 1848 fw_devlink.strict=<bool> 1849 [KNL,EARLY] Treat all inferred dependencies as mandatory 1850 dependencies. This only applies for fw_devlink=on|rpm. 1851 Format: <bool> 1852 1853 fw_devlink.sync_state = 1854 [KNL,EARLY] When all devices that could probe have finished 1855 probing, this parameter controls what to do with 1856 devices that haven't yet received their sync_state() 1857 calls. 1858 Format: { strict | timeout } 1859 strict -- Default. Continue waiting on consumers to 1860 probe successfully. 1861 timeout -- Give up waiting on consumers and call 1862 sync_state() on any devices that haven't yet 1863 received their sync_state() calls after 1864 deferred_probe_timeout has expired or by 1865 late_initcall() if !CONFIG_MODULES. 1866 1867 gamecon.map[2|3]= 1868 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad 1869 support via parallel port (up to 5 devices per port) 1870 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5> 1871 See also Documentation/input/devices/joystick-parport.rst 1872 1873 gamma= [HW,DRM] 1874 1875 gart_fix_e820= [X86-64,EARLY] disable the fix e820 for K8 GART 1876 Format: off | on 1877 default: on 1878 1879 gather_data_sampling= 1880 [X86,INTEL,EARLY] Control the Gather Data Sampling (GDS) 1881 mitigation. 1882 1883 Gather Data Sampling is a hardware vulnerability which 1884 allows unprivileged speculative access to data which was 1885 previously stored in vector registers. 1886 1887 This issue is mitigated by default in updated microcode. 1888 The mitigation may have a performance impact but can be 1889 disabled. On systems without the microcode mitigation 1890 disabling AVX serves as a mitigation. 1891 1892 force: Disable AVX to mitigate systems without 1893 microcode mitigation. No effect if the microcode 1894 mitigation is present. Known to cause crashes in 1895 userspace with buggy AVX enumeration. 1896 1897 off: Disable GDS mitigation. 1898 1899 gbpages [X86] Use GB pages for kernel direct mappings. 1900 1901 gcov_persist= [GCOV] When non-zero (default), profiling data for 1902 kernel modules is saved and remains accessible via 1903 debugfs, even when the module is unloaded/reloaded. 1904 When zero, profiling data is discarded and associated 1905 debugfs files are removed at module unload time. 1906 1907 goldfish [X86] Enable the goldfish android emulator platform. 1908 Don't use this when you are not running on the 1909 android emulator 1910 1911 gpio-mockup.gpio_mockup_ranges 1912 [HW] Sets the ranges of gpiochip of for this device. 1913 Format: <start1>,<end1>,<start2>,<end2>... 1914 gpio-mockup.gpio_mockup_named_lines 1915 [HW] Let the driver know GPIO lines should be named. 1916 1917 gpt [EFI] Forces disk with valid GPT signature but 1918 invalid Protective MBR to be treated as GPT. If the 1919 primary GPT is corrupted, it enables the backup/alternate 1920 GPT to be used instead. 1921 1922 grcan.enable0= [HW] Configuration of physical interface 0. Determines 1923 the "Enable 0" bit of the configuration register. 1924 Format: 0 | 1 1925 Default: 0 1926 grcan.enable1= [HW] Configuration of physical interface 1. Determines 1927 the "Enable 0" bit of the configuration register. 1928 Format: 0 | 1 1929 Default: 0 1930 grcan.select= [HW] Select which physical interface to use. 1931 Format: 0 | 1 1932 Default: 0 1933 grcan.txsize= [HW] Sets the size of the tx buffer. 1934 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0. 1935 Default: 1024 1936 grcan.rxsize= [HW] Sets the size of the rx buffer. 1937 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0. 1938 Default: 1024 1939 1940 hardened_usercopy= 1941 [KNL] Under CONFIG_HARDENED_USERCOPY, whether 1942 hardening is enabled for this boot. Hardened 1943 usercopy checking is used to protect the kernel 1944 from reading or writing beyond known memory 1945 allocation boundaries as a proactive defense 1946 against bounds-checking flaws in the kernel's 1947 copy_to_user()/copy_from_user() interface. 1948 The default is determined by 1949 CONFIG_HARDENED_USERCOPY_DEFAULT_ON. 1950 on Perform hardened usercopy checks. 1951 off Disable hardened usercopy checks. 1952 1953 hardlockup_all_cpu_backtrace= 1954 [KNL] Should the hard-lockup detector generate 1955 backtraces on all cpus. 1956 Format: 0 | 1 1957 1958 hash_pointers= 1959 [KNL,EARLY] 1960 By default, when pointers are printed to the console 1961 or buffers via the %p format string, that pointer is 1962 "hashed", i.e. obscured by hashing the pointer value. 1963 This is a security feature that hides actual kernel 1964 addresses from unprivileged users, but it also makes 1965 debugging the kernel more difficult since unequal 1966 pointers can no longer be compared. The choices are: 1967 Format: { auto | always | never } 1968 Default: auto 1969 1970 auto - Hash pointers unless slab_debug is enabled. 1971 always - Always hash pointers (even if slab_debug is 1972 enabled). 1973 never - Never hash pointers. This option should only 1974 be specified when debugging the kernel. Do 1975 not use on production kernels. The boot 1976 param "no_hash_pointers" is an alias for 1977 this mode. 1978 1979 hashdist= [KNL,NUMA] Large hashes allocated during boot 1980 are distributed across NUMA nodes. Defaults on 1981 for 64-bit NUMA, off otherwise. 1982 Format: 0 | 1 (for off | on) 1983 1984 hd= [EIDE] (E)IDE hard drive subsystem geometry 1985 Format: <cyl>,<head>,<sect> 1986 1987 hest_disable [ACPI] 1988 Disable Hardware Error Source Table (HEST) support; 1989 corresponding firmware-first mode error processing 1990 logic will be disabled. 1991 1992 hibernate= [HIBERNATION] 1993 noresume Don't check if there's a hibernation image 1994 present during boot. 1995 nocompress Don't compress/decompress hibernation images. 1996 no Disable hibernation and resume. 1997 protect_image Turn on image protection during restoration 1998 (that will set all pages holding image data 1999 during restoration read-only). 2000 2001 hibernate.compressor= [HIBERNATION] Compression algorithm to be 2002 used with hibernation. 2003 Format: { lzo | lz4 } 2004 Default: lzo 2005 2006 lzo: Select LZO compression algorithm to 2007 compress/decompress hibernation image. 2008 2009 lz4: Select LZ4 compression algorithm to 2010 compress/decompress hibernation image. 2011 2012 hibernate.pm_test_delay= 2013 [HIBERNATION] 2014 Sets the number of seconds to remain in a hibernation test 2015 mode before resuming the system (see 2016 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG 2017 is set. Default value is 5. 2018 2019 hibernate_compression_threads= 2020 [HIBERNATION] 2021 Set the number of threads used for compressing or decompressing 2022 hibernation images. 2023 2024 Format: <integer> 2025 Default: 3 2026 Minimum: 1 2027 Example: hibernate_compression_threads=4 2028 2029 highmem=nn[KMG] [KNL,BOOT,EARLY] forces the highmem zone to have an exact 2030 size of <nn>. This works even on boxes that have no 2031 highmem otherwise. This also works to reduce highmem 2032 size on bigger boxes. 2033 2034 highres= [KNL] Enable/disable high resolution timer mode. 2035 Valid parameters: "on", "off" 2036 Default: "on" 2037 2038 hlt [BUGS=ARM,SH] 2039 2040 hostname= [KNL,EARLY] Set the hostname (aka UTS nodename). 2041 Format: <string> 2042 This allows setting the system's hostname during early 2043 startup. This sets the name returned by gethostname. 2044 Using this parameter to set the hostname makes it 2045 possible to ensure the hostname is correctly set before 2046 any userspace processes run, avoiding the possibility 2047 that a process may call gethostname before the hostname 2048 has been explicitly set, resulting in the calling 2049 process getting an incorrect result. The string must 2050 not exceed the maximum allowed hostname length (usually 2051 64 characters) and will be truncated otherwise. 2052 2053 hpet= [X86-32,HPET] option to control HPET usage 2054 Format: { enable (default) | disable | force | 2055 verbose } 2056 disable: disable HPET and use PIT instead 2057 force: allow force enabled of undocumented chips (ICH4, 2058 VIA, nVidia) 2059 verbose: show contents of HPET registers during setup 2060 2061 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET 2062 registers. Default set by CONFIG_HPET_MMAP_DEFAULT. 2063 2064 hugepages= [HW,EARLY] Number of HugeTLB pages to allocate at boot. 2065 If this follows hugepagesz (below), it specifies 2066 the number of pages of hugepagesz to be allocated. 2067 If this is the first HugeTLB parameter on the command 2068 line, it specifies the number of pages to allocate for 2069 the default huge page size. If using node format, the 2070 number of pages to allocate per-node can be specified. 2071 See also Documentation/admin-guide/mm/hugetlbpage.rst. 2072 Format: <integer> or (node format) 2073 <node>:<integer>[,<node>:<integer>] 2074 2075 hugepagesz= 2076 [HW,EARLY] The size of the HugeTLB pages. This is 2077 used in conjunction with hugepages (above) to 2078 allocate huge pages of a specific size at boot. The 2079 pair hugepagesz=X hugepages=Y can be specified once 2080 for each supported huge page size. Huge page sizes 2081 are architecture dependent. See also 2082 Documentation/admin-guide/mm/hugetlbpage.rst. 2083 Format: size[KMG] 2084 2085 hugepage_alloc_threads= 2086 [HW] The number of threads that should be used to 2087 allocate hugepages during boot. This option can be 2088 used to improve system bootup time when allocating 2089 a large amount of huge pages. 2090 The default value is 25% of the available hardware threads. 2091 2092 Note that this parameter only applies to non-gigantic huge pages. 2093 2094 hugetlb_cma= [HW,CMA,EARLY] The size of a CMA area used for allocation 2095 of gigantic hugepages. Or using node format, the size 2096 of a CMA area per node can be specified. 2097 Format: nn[KMGTPE] or (node format) 2098 <node>:nn[KMGTPE][,<node>:nn[KMGTPE]] 2099 2100 Reserve a CMA area of given size and allocate gigantic 2101 hugepages using the CMA allocator. If enabled, the 2102 boot-time allocation of gigantic hugepages is skipped. 2103 2104 hugetlb_cma_only= 2105 [HW,CMA,EARLY] When allocating new HugeTLB pages, only 2106 try to allocate from the CMA areas. 2107 2108 This option does nothing if hugetlb_cma= is not also 2109 specified. 2110 2111 hugetlb_free_vmemmap= 2112 [KNL] Requires CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP 2113 enabled. 2114 Control if HugeTLB Vmemmap Optimization (HVO) is enabled. 2115 Allows heavy hugetlb users to free up some more 2116 memory (7 * PAGE_SIZE for each 2MB hugetlb page). 2117 Format: { on | off (default) } 2118 2119 on: enable HVO 2120 off: disable HVO 2121 2122 Built with CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP_DEFAULT_ON=y, 2123 the default is on. 2124 2125 Note that the vmemmap pages may be allocated from the added 2126 memory block itself when memory_hotplug.memmap_on_memory is 2127 enabled, those vmemmap pages cannot be optimized even if this 2128 feature is enabled. Other vmemmap pages not allocated from 2129 the added memory block itself do not be affected. 2130 2131 hung_task_panic= 2132 [KNL] Number of hung tasks to trigger kernel panic. 2133 Format: <int> 2134 2135 When set to a non-zero value, a kernel panic will be triggered if 2136 the number of detected hung tasks reaches this value. 2137 2138 0: don't panic 2139 1: panic immediately on first hung task 2140 N: panic after N hung tasks are detected in a single scan 2141 2142 The default value is controlled by the 2143 CONFIG_BOOTPARAM_HUNG_TASK_PANIC build-time option. The value 2144 selected by this boot parameter can be changed later by the 2145 kernel.hung_task_panic sysctl. 2146 2147 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC) 2148 terminal devices. Valid values: 0..8 2149 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs. 2150 If specified, z/VM IUCV HVC accepts connections 2151 from listed z/VM user IDs only. 2152 2153 hv_nopvspin [X86,HYPER_V,EARLY] 2154 Disables the paravirt spinlock optimizations 2155 which allow the hypervisor to 'idle' the guest 2156 on lock contention. 2157 2158 hw_protection= [HW] 2159 Format: reboot | shutdown 2160 2161 Hardware protection action taken on critical events like 2162 overtemperature or imminent voltage loss. 2163 2164 i2c_bus= [HW] Override the default board specific I2C bus speed 2165 or register an additional I2C bus that is not 2166 registered from board initialization code. 2167 Format: 2168 <bus_id>,<clkrate> 2169 2170 i2c_touchscreen_props= [HW,ACPI,X86] 2171 Set device-properties for ACPI-enumerated I2C-attached 2172 touchscreen, to e.g. fix coordinates of upside-down 2173 mounted touchscreens. If you need this option please 2174 submit a drivers/platform/x86/touchscreen_dmi.c patch 2175 adding a DMI quirk for this. 2176 2177 Format: 2178 <ACPI_HW_ID>:<prop_name>=<val>[:prop_name=val][:...] 2179 Where <val> is one of: 2180 Omit "=<val>" entirely Set a boolean device-property 2181 Unsigned number Set a u32 device-property 2182 Anything else Set a string device-property 2183 2184 Examples (split over multiple lines): 2185 i2c_touchscreen_props=GDIX1001:touchscreen-inverted-x: 2186 touchscreen-inverted-y 2187 2188 i2c_touchscreen_props=MSSL1680:touchscreen-size-x=1920: 2189 touchscreen-size-y=1080:touchscreen-inverted-y: 2190 firmware-name=gsl1680-vendor-model.fw:silead,home-button 2191 2192 i8042.debug [HW] Toggle i8042 debug mode 2193 i8042.unmask_kbd_data 2194 [HW] Enable printing of interrupt data from the KBD port 2195 (disabled by default, and as a pre-condition 2196 requires that i8042.debug=1 be enabled) 2197 i8042.direct [HW] Put keyboard port into non-translated mode 2198 i8042.dumbkbd [HW] Pretend that controller can only read data from 2199 keyboard and cannot control its state 2200 (Don't attempt to blink the leds) 2201 i8042.noaux [HW] Don't check for auxiliary (== mouse) port 2202 i8042.nokbd [HW] Don't check/create keyboard port 2203 i8042.noloop [HW] Disable the AUX Loopback command while probing 2204 for the AUX port 2205 i8042.nomux [HW] Don't check presence of an active multiplexing 2206 controller 2207 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX 2208 controllers 2209 i8042.notimeout [HW] Ignore timeout condition signalled by controller 2210 i8042.reset [HW] Reset the controller during init, cleanup and 2211 suspend-to-ram transitions, only during s2r 2212 transitions, or never reset 2213 Format: { 1 | Y | y | 0 | N | n } 2214 1, Y, y: always reset controller 2215 0, N, n: don't ever reset controller 2216 Default: only on s2r transitions on x86; most other 2217 architectures force reset to be always executed 2218 i8042.unlock [HW] Unlock (ignore) the keylock 2219 i8042.kbdreset [HW] Reset device connected to KBD port 2220 i8042.probe_defer 2221 [HW] Allow deferred probing upon i8042 probe errors 2222 2223 i810= [HW,DRM] 2224 2225 i915.invert_brightness= 2226 [DRM] Invert the sense of the variable that is used to 2227 set the brightness of the panel backlight. Normally a 2228 brightness value of 0 indicates backlight switched off, 2229 and the maximum of the brightness value sets the backlight 2230 to maximum brightness. If this parameter is set to 0 2231 (default) and the machine requires it, or this parameter 2232 is set to 1, a brightness value of 0 sets the backlight 2233 to maximum brightness, and the maximum of the brightness 2234 value switches the backlight off. 2235 -1 -- never invert brightness 2236 0 -- machine default 2237 1 -- force brightness inversion 2238 2239 ia32_emulation= [X86-64] 2240 Format: <bool> 2241 When true, allows loading 32-bit programs and executing 32-bit 2242 syscalls, essentially overriding IA32_EMULATION_DEFAULT_DISABLED at 2243 boot time. When false, unconditionally disables IA32 emulation. 2244 2245 icn= [HW,ISDN] 2246 Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]] 2247 2248 2249 idle= [X86,EARLY] 2250 Format: idle=poll, idle=halt, idle=nomwait 2251 2252 idle=poll: Don't do power saving in the idle loop 2253 using HLT, but poll for rescheduling event. This will 2254 make the CPUs eat a lot more power, but may be useful 2255 to get slightly better performance in multiprocessor 2256 benchmarks. It also makes some profiling using 2257 performance counters more accurate. Please note that 2258 on systems with MONITOR/MWAIT support (like Intel 2259 EM64T CPUs) this option has no performance advantage 2260 over the normal idle loop. It may also interact badly 2261 with hyperthreading. 2262 2263 idle=halt: Halt is forced to be used for CPU idle. 2264 In such case C2/C3 won't be used again. 2265 2266 idle=nomwait: Disable mwait for CPU C-states 2267 2268 idxd.sva= [HW] 2269 Format: <bool> 2270 Allow force disabling of Shared Virtual Memory (SVA) 2271 support for the idxd driver. By default it is set to 2272 true (1). 2273 2274 idxd.tc_override= [HW] 2275 Format: <bool> 2276 Allow override of default traffic class configuration 2277 for the device. By default it is set to false (0). 2278 2279 ieee754= [MIPS] Select IEEE Std 754 conformance mode 2280 Format: { strict | legacy | 2008 | relaxed | emulated } 2281 Default: strict 2282 2283 Choose which programs will be accepted for execution 2284 based on the IEEE 754 NaN encoding(s) supported by 2285 the FPU and the NaN encoding requested with the value 2286 of an ELF file header flag individually set by each 2287 binary. Hardware implementations are permitted to 2288 support either or both of the legacy and the 2008 NaN 2289 encoding mode. 2290 2291 Available settings are as follows: 2292 strict accept binaries that request a NaN encoding 2293 supported by the FPU 2294 legacy only accept legacy-NaN binaries, if supported 2295 by the FPU 2296 2008 only accept 2008-NaN binaries, if supported 2297 by the FPU 2298 relaxed accept any binaries regardless of whether 2299 supported by the FPU 2300 emulated accept any binaries but enable FPU emulator 2301 if binary mode is unsupported by the FPU. 2302 2303 The FPU emulator is always able to support both NaN 2304 encodings, so if no FPU hardware is present or it has 2305 been disabled with 'nofpu', then the settings of 2306 'legacy' and '2008' strap the emulator accordingly, 2307 'relaxed' straps the emulator for both legacy-NaN and 2308 2008-NaN, whereas 'strict' enables legacy-NaN only on 2309 legacy processors and both NaN encodings on MIPS32 or 2310 MIPS64 CPUs. 2311 2312 The setting for ABS.fmt/NEG.fmt instruction execution 2313 mode generally follows that for the NaN encoding, 2314 except where unsupported by hardware. 2315 2316 ignore_loglevel [KNL,EARLY] 2317 Ignore loglevel setting - this will print /all/ 2318 kernel messages to the console. Useful for debugging. 2319 We also add it as printk module parameter, so users 2320 could change it dynamically, usually by 2321 /sys/module/printk/parameters/ignore_loglevel. 2322 2323 ignore_rlimit_data 2324 Ignore RLIMIT_DATA setting for data mappings, 2325 print warning at first misuse. Can be changed via 2326 /sys/module/kernel/parameters/ignore_rlimit_data. 2327 2328 ihash_entries= [KNL] 2329 Set number of hash buckets for inode cache. 2330 2331 ima_appraise= [IMA] appraise integrity measurements 2332 Format: { "off" | "enforce" | "fix" | "log" } 2333 default: "enforce" 2334 2335 ima_appraise_tcb [IMA] Deprecated. Use ima_policy= instead. 2336 The builtin appraise policy appraises all files 2337 owned by uid=0. 2338 2339 ima_canonical_fmt [IMA] 2340 Use the canonical format for the binary runtime 2341 measurements, instead of host native format. 2342 2343 ima_hash= [IMA] 2344 Format: { md5 | sha1 | rmd160 | sha256 | sha384 2345 | sha512 | ... } 2346 default: "sha1" 2347 2348 The list of supported hash algorithms is defined 2349 in crypto/hash_info.h. 2350 2351 ima_policy= [IMA] 2352 The builtin policies to load during IMA setup. 2353 Format: "tcb | appraise_tcb | secure_boot | 2354 fail_securely | critical_data" 2355 2356 The "tcb" policy measures all programs exec'd, files 2357 mmap'd for exec, and all files opened with the read 2358 mode bit set by either the effective uid (euid=0) or 2359 uid=0. 2360 2361 The "appraise_tcb" policy appraises the integrity of 2362 all files owned by root. 2363 2364 The "secure_boot" policy appraises the integrity 2365 of files (eg. kexec kernel image, kernel modules, 2366 firmware, policy, etc) based on file signatures. 2367 2368 The "fail_securely" policy forces file signature 2369 verification failure also on privileged mounted 2370 filesystems with the SB_I_UNVERIFIABLE_SIGNATURE 2371 flag. 2372 2373 The "critical_data" policy measures kernel integrity 2374 critical data. 2375 2376 ima_tcb [IMA] Deprecated. Use ima_policy= instead. 2377 Load a policy which meets the needs of the Trusted 2378 Computing Base. This means IMA will measure all 2379 programs exec'd, files mmap'd for exec, and all files 2380 opened for read by uid=0. 2381 2382 ima_template= [IMA] 2383 Select one of defined IMA measurements template formats. 2384 Formats: { "ima" | "ima-ng" | "ima-ngv2" | "ima-sig" | 2385 "ima-sigv2" } 2386 Default: "ima-ng" 2387 2388 ima_template_fmt= 2389 [IMA] Define a custom template format. 2390 Format: { "field1|...|fieldN" } 2391 2392 ima.ahash_minsize= [IMA] Minimum file size for asynchronous hash usage 2393 Format: <min_file_size> 2394 Set the minimal file size for using asynchronous hash. 2395 If left unspecified, ahash usage is disabled. 2396 2397 ahash performance varies for different data sizes on 2398 different crypto accelerators. This option can be used 2399 to achieve the best performance for a particular HW. 2400 2401 ima.ahash_bufsize= [IMA] Asynchronous hash buffer size 2402 Format: <bufsize> 2403 Set hashing buffer size. Default: 4k. 2404 2405 ahash performance varies for different chunk sizes on 2406 different crypto accelerators. This option can be used 2407 to achieve best performance for particular HW. 2408 2409 ima= [IMA] Enable or disable IMA 2410 Format: { "off" | "on" } 2411 Default: "on" 2412 Note that disabling IMA is limited to kdump kernel. 2413 2414 indirect_target_selection= [X86,Intel] Mitigation control for Indirect 2415 Target Selection(ITS) bug in Intel CPUs. Updated 2416 microcode is also required for a fix in IBPB. 2417 2418 on: Enable mitigation (default). 2419 off: Disable mitigation. 2420 force: Force the ITS bug and deploy default 2421 mitigation. 2422 vmexit: Only deploy mitigation if CPU is affected by 2423 guest/host isolation part of ITS. 2424 stuff: Deploy RSB-fill mitigation when retpoline is 2425 also deployed. Otherwise, deploy the default 2426 mitigation. 2427 2428 For details see: 2429 Documentation/admin-guide/hw-vuln/indirect-target-selection.rst 2430 2431 init= [KNL] 2432 Format: <full_path> 2433 Run specified binary instead of /sbin/init as init 2434 process. 2435 2436 initcall_debug [KNL] Trace initcalls as they are executed. Useful 2437 for working out where the kernel is dying during 2438 startup. 2439 2440 initcall_blacklist= [KNL] Do not execute a comma-separated list of 2441 initcall functions. Useful for debugging built-in 2442 modules and initcalls. 2443 2444 initramfs_async= [KNL] 2445 Format: <bool> 2446 Default: 1 2447 This parameter controls whether the initramfs 2448 image is unpacked asynchronously, concurrently 2449 with devices being probed and 2450 initialized. This should normally just work, 2451 but as a debugging aid, one can get the 2452 historical behaviour of the initramfs 2453 unpacking being completed before device_ and 2454 late_ initcalls. 2455 2456 initrd= [BOOT,EARLY] Specify the location of the initial ramdisk 2457 2458 initrdmem= [KNL,EARLY] Specify a physical address and size from which to 2459 load the initrd. If an initrd is compiled in or 2460 specified in the bootparams, it takes priority over this 2461 setting. 2462 Format: ss[KMG],nn[KMG] 2463 Default is 0, 0 2464 2465 init_on_alloc= [MM,EARLY] Fill newly allocated pages and heap objects with 2466 zeroes. 2467 Format: 0 | 1 2468 Default set by CONFIG_INIT_ON_ALLOC_DEFAULT_ON. 2469 2470 init_on_free= [MM,EARLY] Fill freed pages and heap objects with zeroes. 2471 Format: 0 | 1 2472 Default set by CONFIG_INIT_ON_FREE_DEFAULT_ON. 2473 2474 init_pkru= [X86] Specify the default memory protection keys rights 2475 register contents for all processes. 0x55555554 by 2476 default (disallow access to all but pkey 0). Can 2477 override in debugfs after boot. 2478 2479 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver 2480 Format: <irq> 2481 2482 int_pln_enable [X86] Enable power limit notification interrupt 2483 2484 integrity_audit=[IMA] 2485 Format: { "0" | "1" } 2486 0 -- basic integrity auditing messages. (Default) 2487 1 -- additional integrity auditing messages. 2488 2489 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option 2490 on 2491 Enable intel iommu driver. 2492 off 2493 Disable intel iommu driver. 2494 igfx_off [Default Off] 2495 By default, gfx is mapped as normal device. If a gfx 2496 device has a dedicated DMAR unit, the DMAR unit is 2497 bypassed by not enabling DMAR with this option. In 2498 this case, gfx device will use physical address for 2499 DMA. 2500 strict [Default Off] 2501 Deprecated, equivalent to iommu.strict=1. 2502 sp_off [Default Off] 2503 By default, super page will be supported if Intel IOMMU 2504 has the capability. With this option, super page will 2505 not be supported. 2506 sm_on 2507 Enable the Intel IOMMU scalable mode if the hardware 2508 advertises that it has support for the scalable mode 2509 translation. 2510 sm_off 2511 Disallow use of the Intel IOMMU scalable mode. 2512 tboot_noforce [Default Off] 2513 Do not force the Intel IOMMU enabled under tboot. 2514 By default, tboot will force Intel IOMMU on, which 2515 could harm performance of some high-throughput 2516 devices like 40GBit network cards, even if identity 2517 mapping is enabled. 2518 Note that using this option lowers the security 2519 provided by tboot because it makes the system 2520 vulnerable to DMA attacks. 2521 2522 intel_idle.max_cstate= [KNL,HW,ACPI,X86] 2523 0 disables intel_idle and fall back on acpi_idle. 2524 1 to 9 specify maximum depth of C-state. 2525 2526 intel_pstate= [X86,EARLY] 2527 disable 2528 Do not enable intel_pstate as the default 2529 scaling driver for the supported processors 2530 active 2531 Use intel_pstate driver to bypass the scaling 2532 governors layer of cpufreq and provides it own 2533 algorithms for p-state selection. There are two 2534 P-state selection algorithms provided by 2535 intel_pstate in the active mode: powersave and 2536 performance. The way they both operate depends 2537 on whether or not the hardware managed P-states 2538 (HWP) feature has been enabled in the processor 2539 and possibly on the processor model. 2540 passive 2541 Use intel_pstate as a scaling driver, but configure it 2542 to work with generic cpufreq governors (instead of 2543 enabling its internal governor). This mode cannot be 2544 used along with the hardware-managed P-states (HWP) 2545 feature. 2546 force 2547 Enable intel_pstate on systems that prohibit it by default 2548 in favor of acpi-cpufreq. Forcing the intel_pstate driver 2549 instead of acpi-cpufreq may disable platform features, such 2550 as thermal controls and power capping, that rely on ACPI 2551 P-States information being indicated to OSPM and therefore 2552 should be used with caution. This option does not work with 2553 processors that aren't supported by the intel_pstate driver 2554 or on platforms that use pcc-cpufreq instead of acpi-cpufreq. 2555 no_hwp 2556 Do not enable hardware P state control (HWP) 2557 if available. 2558 hwp_only 2559 Only load intel_pstate on systems which support 2560 hardware P state control (HWP) if available. 2561 support_acpi_ppc 2562 Enforce ACPI _PPC performance limits. If the Fixed ACPI 2563 Description Table, specifies preferred power management 2564 profile as "Enterprise Server" or "Performance Server", 2565 then this feature is turned on by default. 2566 per_cpu_perf_limits 2567 Allow per-logical-CPU P-State performance control limits using 2568 cpufreq sysfs interface 2569 no_cas 2570 Do not enable capacity-aware scheduling (CAS) on 2571 hybrid systems 2572 2573 intremap= [X86-64,Intel-IOMMU,EARLY] 2574 on enable Interrupt Remapping (default) 2575 off disable Interrupt Remapping 2576 nosid disable Source ID checking 2577 no_x2apic_optout 2578 BIOS x2APIC opt-out request will be ignored 2579 nopost disable Interrupt Posting 2580 posted_msi 2581 enable MSIs delivered as posted interrupts 2582 2583 iomem= Disable strict checking of access to MMIO memory 2584 strict regions from userspace. 2585 relaxed 2586 2587 iommu= [X86,EARLY] 2588 2589 off 2590 Don't initialize and use any kind of IOMMU. 2591 2592 force 2593 Force the use of the hardware IOMMU even when 2594 it is not actually needed (e.g. because < 3 GB 2595 memory). 2596 2597 noforce 2598 Don't force hardware IOMMU usage when it is not 2599 needed. (default). 2600 2601 biomerge 2602 panic 2603 nopanic 2604 merge 2605 nomerge 2606 2607 soft 2608 Use software bounce buffering (SWIOTLB) (default for 2609 Intel machines). This can be used to prevent the usage 2610 of an available hardware IOMMU. 2611 2612 [X86] 2613 pt 2614 [X86] 2615 nopt 2616 [PPC/POWERNV] 2617 nobypass 2618 Disable IOMMU bypass, using IOMMU for PCI devices. 2619 2620 [X86] 2621 AMD Gart HW IOMMU-specific options: 2622 2623 <size> 2624 Set the size of the remapping area in bytes. 2625 2626 allowed 2627 Overwrite iommu off workarounds for specific chipsets 2628 2629 fullflush 2630 Flush IOMMU on each allocation (default). 2631 2632 nofullflush 2633 Don't use IOMMU fullflush. 2634 2635 memaper[=<order>] 2636 Allocate an own aperture over RAM with size 2637 32MB<<order. (default: order=1, i.e. 64MB) 2638 2639 merge 2640 Do scatter-gather (SG) merging. Implies "force" 2641 (experimental). 2642 2643 nomerge 2644 Don't do scatter-gather (SG) merging. 2645 2646 noaperture 2647 Ask the IOMMU not to touch the aperture for AGP. 2648 2649 noagp 2650 Don't initialize the AGP driver and use full aperture. 2651 2652 panic 2653 Always panic when IOMMU overflows. 2654 2655 iommu.forcedac= [ARM64,X86,EARLY] Control IOVA allocation for PCI devices. 2656 Format: { "0" | "1" } 2657 0 - Try to allocate a 32-bit DMA address first, before 2658 falling back to the full range if needed. 2659 1 - Allocate directly from the full usable range, 2660 forcing Dual Address Cycle for PCI cards supporting 2661 greater than 32-bit addressing. 2662 2663 iommu.strict= [ARM64,X86,S390,EARLY] Configure TLB invalidation behaviour 2664 Format: { "0" | "1" } 2665 0 - Lazy mode. 2666 Request that DMA unmap operations use deferred 2667 invalidation of hardware TLBs, for increased 2668 throughput at the cost of reduced device isolation. 2669 Will fall back to strict mode if not supported by 2670 the relevant IOMMU driver. 2671 1 - Strict mode. 2672 DMA unmap operations invalidate IOMMU hardware TLBs 2673 synchronously. 2674 unset - Use value of CONFIG_IOMMU_DEFAULT_DMA_{LAZY,STRICT}. 2675 Note: on x86, strict mode specified via one of the 2676 legacy driver-specific options takes precedence. 2677 2678 iommu.passthrough= 2679 [ARM64,X86,EARLY] Configure DMA to bypass the IOMMU by default. 2680 Format: { "0" | "1" } 2681 0 - Use IOMMU translation for DMA. 2682 1 - Bypass the IOMMU for DMA. 2683 unset - Use value of CONFIG_IOMMU_DEFAULT_PASSTHROUGH. 2684 2685 io7= [HW] IO7 for Marvel-based Alpha systems 2686 See comment before marvel_specify_io7 in 2687 arch/alpha/kernel/core_marvel.c. 2688 2689 io_delay= [X86,EARLY] I/O delay method 2690 0x80 2691 Standard port 0x80 based delay 2692 0xed 2693 Alternate port 0xed based delay (needed on some systems) 2694 udelay 2695 Simple two microseconds delay 2696 none 2697 No delay 2698 2699 ip= [IP_PNP] 2700 See Documentation/admin-guide/nfs/nfsroot.rst. 2701 2702 ipcmni_extend [KNL,EARLY] Extend the maximum number of unique System V 2703 IPC identifiers from 32,768 to 16,777,216. 2704 2705 ipe.enforce= [IPE] 2706 Format: <bool> 2707 Determine whether IPE starts in permissive (0) or 2708 enforce (1) mode. The default is enforce. 2709 2710 ipe.success_audit= 2711 [IPE] 2712 Format: <bool> 2713 Start IPE with success auditing enabled, emitting 2714 an audit event when a binary is allowed. The default 2715 is 0. 2716 2717 irqaffinity= [SMP] Set the default irq affinity mask 2718 The argument is a cpu list, as described above. 2719 2720 irqchip.gicv2_force_probe= 2721 [ARM,ARM64,EARLY] 2722 Format: <bool> 2723 Force the kernel to look for the second 4kB page 2724 of a GICv2 controller even if the memory range 2725 exposed by the device tree is too small. 2726 2727 irqchip.gicv3_nolpi= 2728 [ARM,ARM64,EARLY] 2729 Force the kernel to ignore the availability of 2730 LPIs (and by consequence ITSs). Intended for system 2731 that use the kernel as a bootloader, and thus want 2732 to let secondary kernels in charge of setting up 2733 LPIs. 2734 2735 irqchip.gicv3_pseudo_nmi= [ARM64,EARLY] 2736 Enables support for pseudo-NMIs in the kernel. This 2737 requires the kernel to be built with 2738 CONFIG_ARM64_PSEUDO_NMI. 2739 2740 irqchip.riscv_imsic_noipi 2741 [RISC-V,EARLY] 2742 Force the kernel to not use IMSIC software injected MSIs 2743 as IPIs. Intended for system where IMSIC is trap-n-emulated, 2744 and thus want to reduce MMIO traps when triggering IPIs 2745 to multiple harts. 2746 2747 irqfixup [HW] 2748 When an interrupt is not handled search all handlers 2749 for it. Intended to get systems with badly broken 2750 firmware running. 2751 2752 irqhandler.duration_warn_us= [KNL] 2753 Warn if an IRQ handler exceeds the specified duration 2754 threshold in microseconds. Useful for identifying 2755 long-running IRQs in the system. 2756 2757 irqpoll [HW] 2758 When an interrupt is not handled search all handlers 2759 for it. Also check all handlers each timer 2760 interrupt. Intended to get systems with badly broken 2761 firmware running. 2762 2763 isapnp= [ISAPNP] 2764 Format: <RDP>,<reset>,<pci_scan>,<verbosity> 2765 2766 isolcpus= [KNL,SMP,ISOL] Isolate a given set of CPUs from disturbance. 2767 [Deprecated - use cpusets instead] 2768 Format: [flag-list,]<cpu-list> 2769 2770 Specify one or more CPUs to isolate from disturbances 2771 specified in the flag list (default: domain): 2772 2773 nohz 2774 Disable the tick when a single task runs as well as 2775 disabling other kernel noises like having RCU callbacks 2776 offloaded. This is equivalent to the nohz_full parameter. 2777 2778 A residual 1Hz tick is offloaded to workqueues, which you 2779 need to affine to housekeeping through the global 2780 workqueue's affinity configured via the 2781 /sys/devices/virtual/workqueue/cpumask sysfs file, or 2782 by using the 'domain' flag described below. 2783 2784 NOTE: by default the global workqueue runs on all CPUs, 2785 so to protect individual CPUs the 'cpumask' file has to 2786 be configured manually after bootup. 2787 2788 domain 2789 Isolate from the general SMP balancing and scheduling 2790 algorithms. Note that performing domain isolation this way 2791 is irreversible: it's not possible to bring back a CPU to 2792 the domains once isolated through isolcpus. It's strongly 2793 advised to use cpusets instead to disable scheduler load 2794 balancing through the "cpuset.sched_load_balance" file. 2795 It offers a much more flexible interface where CPUs can 2796 move in and out of an isolated set anytime. 2797 2798 You can move a process onto or off an "isolated" CPU via 2799 the CPU affinity syscalls or cpuset. 2800 <cpu number> begins at 0 and the maximum value is 2801 "number of CPUs in system - 1". 2802 2803 managed_irq 2804 2805 Isolate from being targeted by managed interrupts 2806 which have an interrupt mask containing isolated 2807 CPUs. The affinity of managed interrupts is 2808 handled by the kernel and cannot be changed via 2809 the /proc/irq/* interfaces. 2810 2811 This isolation is best effort and only effective 2812 if the automatically assigned interrupt mask of a 2813 device queue contains isolated and housekeeping 2814 CPUs. If housekeeping CPUs are online then such 2815 interrupts are directed to the housekeeping CPU 2816 so that IO submitted on the housekeeping CPU 2817 cannot disturb the isolated CPU. 2818 2819 If a queue's affinity mask contains only isolated 2820 CPUs then this parameter has no effect on the 2821 interrupt routing decision, though interrupts are 2822 only delivered when tasks running on those 2823 isolated CPUs submit IO. IO submitted on 2824 housekeeping CPUs has no influence on those 2825 queues. 2826 2827 The format of <cpu-list> is described above. 2828 2829 iucv= [HW,NET] 2830 2831 ivrs_ioapic [HW,X86-64] 2832 Provide an override to the IOAPIC-ID<->DEVICE-ID 2833 mapping provided in the IVRS ACPI table. 2834 By default, PCI segment is 0, and can be omitted. 2835 2836 For example, to map IOAPIC-ID decimal 10 to 2837 PCI segment 0x1 and PCI device 00:14.0, 2838 write the parameter as: 2839 ivrs_ioapic=10@0001:00:14.0 2840 2841 Deprecated formats: 2842 * To map IOAPIC-ID decimal 10 to PCI device 00:14.0 2843 write the parameter as: 2844 ivrs_ioapic[10]=00:14.0 2845 * To map IOAPIC-ID decimal 10 to PCI segment 0x1 and 2846 PCI device 00:14.0 write the parameter as: 2847 ivrs_ioapic[10]=0001:00:14.0 2848 2849 ivrs_hpet [HW,X86-64] 2850 Provide an override to the HPET-ID<->DEVICE-ID 2851 mapping provided in the IVRS ACPI table. 2852 By default, PCI segment is 0, and can be omitted. 2853 2854 For example, to map HPET-ID decimal 10 to 2855 PCI segment 0x1 and PCI device 00:14.0, 2856 write the parameter as: 2857 ivrs_hpet=10@0001:00:14.0 2858 2859 Deprecated formats: 2860 * To map HPET-ID decimal 0 to PCI device 00:14.0 2861 write the parameter as: 2862 ivrs_hpet[0]=00:14.0 2863 * To map HPET-ID decimal 10 to PCI segment 0x1 and 2864 PCI device 00:14.0 write the parameter as: 2865 ivrs_ioapic[10]=0001:00:14.0 2866 2867 ivrs_acpihid [HW,X86-64] 2868 Provide an override to the ACPI-HID:UID<->DEVICE-ID 2869 mapping provided in the IVRS ACPI table. 2870 By default, PCI segment is 0, and can be omitted. 2871 2872 For example, to map UART-HID:UID AMD0020:0 to 2873 PCI segment 0x1 and PCI device ID 00:14.5, 2874 write the parameter as: 2875 ivrs_acpihid=AMD0020:0@0001:00:14.5 2876 2877 Deprecated formats: 2878 * To map UART-HID:UID AMD0020:0 to PCI segment is 0, 2879 PCI device ID 00:14.5, write the parameter as: 2880 ivrs_acpihid[00:14.5]=AMD0020:0 2881 * To map UART-HID:UID AMD0020:0 to PCI segment 0x1 and 2882 PCI device ID 00:14.5, write the parameter as: 2883 ivrs_acpihid[0001:00:14.5]=AMD0020:0 2884 2885 js= [HW,JOY] Analog joystick 2886 See Documentation/input/joydev/joystick.rst. 2887 2888 kasan_multi_shot 2889 [KNL] Enforce KASAN (Kernel Address Sanitizer) to print 2890 report on every invalid memory access. Without this 2891 parameter KASAN will print report only for the first 2892 invalid access. 2893 2894 keep_bootcon [KNL,EARLY] 2895 Do not unregister boot console at start. This is only 2896 useful for debugging when something happens in the window 2897 between unregistering the boot console and initializing 2898 the real console. 2899 2900 keepinitrd [HW,ARM] See retain_initrd. 2901 2902 kernelcore= [KNL,X86,PPC,EARLY] 2903 Format: nn[KMGTPE] | nn% | "mirror" 2904 This parameter specifies the amount of memory usable by 2905 the kernel for non-movable allocations. The requested 2906 amount is spread evenly throughout all nodes in the 2907 system as ZONE_NORMAL. The remaining memory is used for 2908 movable memory in its own zone, ZONE_MOVABLE. In the 2909 event, a node is too small to have both ZONE_NORMAL and 2910 ZONE_MOVABLE, kernelcore memory will take priority and 2911 other nodes will have a larger ZONE_MOVABLE. 2912 2913 ZONE_MOVABLE is used for the allocation of pages that 2914 may be reclaimed or moved by the page migration 2915 subsystem. Note that allocations like PTEs-from-HighMem 2916 still use the HighMem zone if it exists, and the Normal 2917 zone if it does not. 2918 2919 It is possible to specify the exact amount of memory in 2920 the form of "nn[KMGTPE]", a percentage of total system 2921 memory in the form of "nn%", or "mirror". If "mirror" 2922 option is specified, mirrored (reliable) memory is used 2923 for non-movable allocations and remaining memory is used 2924 for Movable pages. "nn[KMGTPE]", "nn%", and "mirror" 2925 are exclusive, so you cannot specify multiple forms. 2926 2927 kgdbdbgp= [KGDB,HW,EARLY] kgdb over EHCI usb debug port. 2928 Format: <Controller#>[,poll interval] 2929 The controller # is the number of the ehci usb debug 2930 port as it is probed via PCI. The poll interval is 2931 optional and is the number seconds in between 2932 each poll cycle to the debug port in case you need 2933 the functionality for interrupting the kernel with 2934 gdb or control-c on the dbgp connection. When 2935 not using this parameter you use sysrq-g to break into 2936 the kernel debugger. 2937 2938 kgdboc= [KGDB,HW] kgdb over consoles. 2939 Requires a tty driver that supports console polling, 2940 or a supported polling keyboard driver (non-usb). 2941 Serial only format: <serial_device>[,baud] 2942 keyboard only format: kbd 2943 keyboard and serial format: kbd,<serial_device>[,baud] 2944 Optional Kernel mode setting: 2945 kms, kbd format: kms,kbd 2946 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud] 2947 2948 kgdboc_earlycon= [KGDB,HW,EARLY] 2949 If the boot console provides the ability to read 2950 characters and can work in polling mode, you can use 2951 this parameter to tell kgdb to use it as a backend 2952 until the normal console is registered. Intended to 2953 be used together with the kgdboc parameter which 2954 specifies the normal console to transition to. 2955 2956 The name of the early console should be specified 2957 as the value of this parameter. Note that the name of 2958 the early console might be different than the tty 2959 name passed to kgdboc. It's OK to leave the value 2960 blank and the first boot console that implements 2961 read() will be picked. 2962 2963 kgdbwait [KGDB,EARLY] Stop kernel execution and enter the 2964 kernel debugger at the earliest opportunity. 2965 2966 kho= [KEXEC,EARLY] 2967 Format: { "0" | "1" | "off" | "on" | "y" | "n" } 2968 Enables or disables Kexec HandOver. 2969 "0" | "off" | "n" - kexec handover is disabled 2970 "1" | "on" | "y" - kexec handover is enabled 2971 2972 kho_scratch= [KEXEC,EARLY] 2973 Format: ll[KMG],mm[KMG],nn[KMG] | nn% 2974 Defines the size of the KHO scratch region. The KHO 2975 scratch regions are physically contiguous memory 2976 ranges that can only be used for non-kernel 2977 allocations. That way, even when memory is heavily 2978 fragmented with handed over memory, the kexeced 2979 kernel will always have enough contiguous ranges to 2980 bootstrap itself. 2981 2982 It is possible to specify the exact amount of 2983 memory in the form of "ll[KMG],mm[KMG],nn[KMG]" 2984 where the first parameter defines the size of a low 2985 memory scratch area, the second parameter defines 2986 the size of a global scratch area and the third 2987 parameter defines the size of additional per-node 2988 scratch areas. The form "nn%" defines scale factor 2989 (in percents) of memory that was used during boot. 2990 2991 kmac= [MIPS] Korina ethernet MAC address. 2992 Configure the RouterBoard 532 series on-chip 2993 Ethernet adapter MAC address. 2994 2995 kmemleak= [KNL,EARLY] Boot-time kmemleak enable/disable 2996 Valid arguments: on, off 2997 Default: on 2998 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y, 2999 the default is off. 3000 3001 kprobe_event=[probe-list] 3002 [FTRACE] Add kprobe events and enable at boot time. 3003 The probe-list is a semicolon delimited list of probe 3004 definitions. Each definition is same as kprobe_events 3005 interface, but the parameters are comma delimited. 3006 For example, to add a kprobe event on vfs_read with 3007 arg1 and arg2, add to the command line; 3008 3009 kprobe_event=p,vfs_read,$arg1,$arg2 3010 3011 See also Documentation/trace/kprobetrace.rst "Kernel 3012 Boot Parameter" section. 3013 3014 kpti= [ARM64,EARLY] Control page table isolation of 3015 user and kernel address spaces. 3016 Default: enabled on cores which need mitigation. 3017 0: force disabled 3018 1: force enabled 3019 3020 kunit.enable= [KUNIT] Enable executing KUnit tests. Requires 3021 CONFIG_KUNIT to be set to be fully enabled. The 3022 default value can be overridden via 3023 KUNIT_DEFAULT_ENABLED. 3024 Default is 1 (enabled) 3025 3026 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs. 3027 Default is 0 (don't ignore, but inject #GP) 3028 3029 kvm.eager_page_split= 3030 [KVM,X86] Controls whether or not KVM will try to 3031 proactively split all huge pages during dirty logging. 3032 Eager page splitting reduces interruptions to vCPU 3033 execution by eliminating the write-protection faults 3034 and MMU lock contention that would otherwise be 3035 required to split huge pages lazily. 3036 3037 VM workloads that rarely perform writes or that write 3038 only to a small region of VM memory may benefit from 3039 disabling eager page splitting to allow huge pages to 3040 still be used for reads. 3041 3042 The behavior of eager page splitting depends on whether 3043 KVM_DIRTY_LOG_INITIALLY_SET is enabled or disabled. If 3044 disabled, all huge pages in a memslot will be eagerly 3045 split when dirty logging is enabled on that memslot. If 3046 enabled, eager page splitting will be performed during 3047 the KVM_CLEAR_DIRTY ioctl, and only for the pages being 3048 cleared. 3049 3050 Eager page splitting is only supported when kvm.tdp_mmu=Y. 3051 3052 Default is Y (on). 3053 3054 kvm.enable_virt_at_load=[KVM,ARM64,LOONGARCH,MIPS,RISCV,X86] 3055 If enabled, KVM will enable virtualization in hardware 3056 when KVM is loaded, and disable virtualization when KVM 3057 is unloaded (if KVM is built as a module). 3058 3059 If disabled, KVM will dynamically enable and disable 3060 virtualization on-demand when creating and destroying 3061 VMs, i.e. on the 0=>1 and 1=>0 transitions of the 3062 number of VMs. 3063 3064 Enabling virtualization at module load avoids potential 3065 latency for creation of the 0=>1 VM, as KVM serializes 3066 virtualization enabling across all online CPUs. The 3067 "cost" of enabling virtualization when KVM is loaded, 3068 is that doing so may interfere with using out-of-tree 3069 hypervisors that want to "own" virtualization hardware. 3070 3071 kvm.enable_vmware_backdoor=[KVM] Support VMware backdoor PV interface. 3072 Default is false (don't support). 3073 3074 kvm.nx_huge_pages= 3075 [KVM] Controls the software workaround for the 3076 X86_BUG_ITLB_MULTIHIT bug. 3077 force : Always deploy workaround. 3078 off : Never deploy workaround. 3079 auto : Deploy workaround based on the presence of 3080 X86_BUG_ITLB_MULTIHIT. 3081 3082 Default is 'auto'. 3083 3084 If the software workaround is enabled for the host, 3085 guests do need not to enable it for nested guests. 3086 3087 kvm.nx_huge_pages_recovery_ratio= 3088 [KVM] Controls how many 4KiB pages are periodically zapped 3089 back to huge pages. 0 disables the recovery, otherwise if 3090 the value is N KVM will zap 1/Nth of the 4KiB pages every 3091 period (see below). The default is 60. 3092 3093 kvm.nx_huge_pages_recovery_period_ms= 3094 [KVM] Controls the time period at which KVM zaps 4KiB pages 3095 back to huge pages. If the value is a non-zero N, KVM will 3096 zap a portion (see ratio above) of the pages every N msecs. 3097 If the value is 0 (the default), KVM will pick a period based 3098 on the ratio, such that a page is zapped after 1 hour on average. 3099 3100 kvm-amd.nested= [KVM,AMD] Control nested virtualization feature in 3101 KVM/SVM. Default is 1 (enabled). 3102 3103 kvm-amd.npt= [KVM,AMD] Control KVM's use of Nested Page Tables, 3104 a.k.a. Two-Dimensional Page Tables. Default is 1 3105 (enabled). Disable by KVM if hardware lacks support 3106 for NPT. 3107 3108 kvm-amd.ciphertext_hiding_asids= 3109 [KVM,AMD] Ciphertext hiding prevents disallowed accesses 3110 to SNP private memory from reading ciphertext. Instead, 3111 reads will see constant default values (0xff). 3112 3113 If ciphertext hiding is enabled, the joint SEV-ES and 3114 SEV-SNP ASID space is partitioned into separate SEV-ES 3115 and SEV-SNP ASID ranges, with the SEV-SNP range being 3116 [1..max_snp_asid] and the SEV-ES range being 3117 (max_snp_asid..min_sev_asid), where min_sev_asid is 3118 enumerated by CPUID.0x.8000_001F[EDX]. 3119 3120 A non-zero value enables SEV-SNP ciphertext hiding and 3121 adjusts the ASID ranges for SEV-ES and SEV-SNP guests. 3122 KVM caps the number of SEV-SNP ASIDs at the maximum 3123 possible value, e.g. specifying -1u will assign all 3124 joint SEV-ES and SEV-SNP ASIDs to SEV-SNP. Note, 3125 assigning all joint ASIDs to SEV-SNP, i.e. configuring 3126 max_snp_asid == min_sev_asid-1, will effectively make 3127 SEV-ES unusable. 3128 3129 kvm-arm.mode= 3130 [KVM,ARM,EARLY] Select one of KVM/arm64's modes of 3131 operation. 3132 3133 none: Forcefully disable KVM. 3134 3135 nvhe: Standard nVHE-based mode, without support for 3136 protected guests. 3137 3138 protected: Mode with support for guests whose state is 3139 kept private from the host, using VHE or 3140 nVHE depending on HW support. 3141 3142 nested: VHE-based mode with support for nested 3143 virtualization. Requires at least ARMv8.4 3144 hardware (with FEAT_NV2). 3145 3146 Defaults to VHE/nVHE based on hardware support. Setting 3147 mode to "protected" will disable kexec and hibernation 3148 for the host. To force nVHE on VHE hardware, add 3149 "arm64_sw.hvhe=0 id_aa64mmfr1.vh=0" to the 3150 command-line. 3151 "nested" is experimental and should be used with 3152 extreme caution. 3153 3154 kvm-arm.vgic_v3_group0_trap= 3155 [KVM,ARM,EARLY] Trap guest accesses to GICv3 group-0 3156 system registers 3157 3158 kvm-arm.vgic_v3_group1_trap= 3159 [KVM,ARM,EARLY] Trap guest accesses to GICv3 group-1 3160 system registers 3161 3162 kvm-arm.vgic_v3_common_trap= 3163 [KVM,ARM,EARLY] Trap guest accesses to GICv3 common 3164 system registers 3165 3166 kvm-arm.vgic_v4_enable= 3167 [KVM,ARM,EARLY] Allow use of GICv4 for direct 3168 injection of LPIs. 3169 3170 kvm-arm.wfe_trap_policy= 3171 [KVM,ARM] Control when to set WFE instruction trap for 3172 KVM VMs. Traps are allowed but not guaranteed by the 3173 CPU architecture. 3174 3175 trap: set WFE instruction trap 3176 3177 notrap: clear WFE instruction trap 3178 3179 kvm-arm.wfi_trap_policy= 3180 [KVM,ARM] Control when to set WFI instruction trap for 3181 KVM VMs. Traps are allowed but not guaranteed by the 3182 CPU architecture. 3183 3184 trap: set WFI instruction trap 3185 3186 notrap: clear WFI instruction trap 3187 3188 kvm_cma_resv_ratio=n [PPC,EARLY] 3189 Reserves given percentage from system memory area for 3190 contiguous memory allocation for KVM hash pagetable 3191 allocation. 3192 By default it reserves 5% of total system memory. 3193 Format: <integer> 3194 Default: 5 3195 3196 kvm-intel.ept= [KVM,Intel] Control KVM's use of Extended Page Tables, 3197 a.k.a. Two-Dimensional Page Tables. Default is 1 3198 (enabled). Disable by KVM if hardware lacks support 3199 for EPT. 3200 3201 kvm-intel.emulate_invalid_guest_state= 3202 [KVM,Intel] Control whether to emulate invalid guest 3203 state. Ignored if kvm-intel.enable_unrestricted_guest=1, 3204 as guest state is never invalid for unrestricted 3205 guests. This param doesn't apply to nested guests (L2), 3206 as KVM never emulates invalid L2 guest state. 3207 Default is 1 (enabled). 3208 3209 kvm-intel.flexpriority= 3210 [KVM,Intel] Control KVM's use of FlexPriority feature 3211 (TPR shadow). Default is 1 (enabled). Disable by KVM if 3212 hardware lacks support for it. 3213 3214 kvm-intel.nested= 3215 [KVM,Intel] Control nested virtualization feature in 3216 KVM/VMX. Default is 1 (enabled). 3217 3218 kvm-intel.unrestricted_guest= 3219 [KVM,Intel] Control KVM's use of unrestricted guest 3220 feature (virtualized real and unpaged mode). Default 3221 is 1 (enabled). Disable by KVM if EPT is disabled or 3222 hardware lacks support for it. 3223 3224 kvm-intel.vmentry_l1d_flush=[KVM,Intel] Mitigation for L1 Terminal Fault 3225 CVE-2018-3620. 3226 3227 Valid arguments: never, cond, always 3228 3229 always: L1D cache flush on every VMENTER. 3230 cond: Flush L1D on VMENTER only when the code between 3231 VMEXIT and VMENTER can leak host memory. 3232 never: Disables the mitigation 3233 3234 Default is cond (do L1 cache flush in specific instances) 3235 3236 kvm-intel.vpid= [KVM,Intel] Control KVM's use of Virtual Processor 3237 Identification feature (tagged TLBs). Default is 1 3238 (enabled). Disable by KVM if hardware lacks support 3239 for it. 3240 3241 l1d_flush= [X86,INTEL,EARLY] 3242 Control mitigation for L1D based snooping vulnerability. 3243 3244 Certain CPUs are vulnerable to an exploit against CPU 3245 internal buffers which can forward information to a 3246 disclosure gadget under certain conditions. 3247 3248 In vulnerable processors, the speculatively 3249 forwarded data can be used in a cache side channel 3250 attack, to access data to which the attacker does 3251 not have direct access. 3252 3253 This parameter controls the mitigation. The 3254 options are: 3255 3256 on - enable the interface for the mitigation 3257 3258 l1tf= [X86,EARLY] Control mitigation of the L1TF vulnerability on 3259 affected CPUs 3260 3261 The kernel PTE inversion protection is unconditionally 3262 enabled and cannot be disabled. 3263 3264 full 3265 Provides all available mitigations for the 3266 L1TF vulnerability. Disables SMT and 3267 enables all mitigations in the 3268 hypervisors, i.e. unconditional L1D flush. 3269 3270 SMT control and L1D flush control via the 3271 sysfs interface is still possible after 3272 boot. Hypervisors will issue a warning 3273 when the first VM is started in a 3274 potentially insecure configuration, 3275 i.e. SMT enabled or L1D flush disabled. 3276 3277 full,force 3278 Same as 'full', but disables SMT and L1D 3279 flush runtime control. Implies the 3280 'nosmt=force' command line option. 3281 (i.e. sysfs control of SMT is disabled.) 3282 3283 flush 3284 Leaves SMT enabled and enables the default 3285 hypervisor mitigation, i.e. conditional 3286 L1D flush. 3287 3288 SMT control and L1D flush control via the 3289 sysfs interface is still possible after 3290 boot. Hypervisors will issue a warning 3291 when the first VM is started in a 3292 potentially insecure configuration, 3293 i.e. SMT enabled or L1D flush disabled. 3294 3295 flush,nosmt 3296 3297 Disables SMT and enables the default 3298 hypervisor mitigation. 3299 3300 SMT control and L1D flush control via the 3301 sysfs interface is still possible after 3302 boot. Hypervisors will issue a warning 3303 when the first VM is started in a 3304 potentially insecure configuration, 3305 i.e. SMT enabled or L1D flush disabled. 3306 3307 flush,nowarn 3308 Same as 'flush', but hypervisors will not 3309 warn when a VM is started in a potentially 3310 insecure configuration. 3311 3312 off 3313 Disables hypervisor mitigations and doesn't 3314 emit any warnings. 3315 It also drops the swap size and available 3316 RAM limit restriction on both hypervisor and 3317 bare metal. 3318 3319 Default is 'flush'. 3320 3321 For details see: Documentation/admin-guide/hw-vuln/l1tf.rst 3322 3323 l2cr= [PPC] 3324 3325 l3cr= [PPC] 3326 3327 lapic [X86-32,APIC,EARLY] Enable the local APIC even if BIOS 3328 disabled it. 3329 3330 lapic= [X86,APIC] Do not use TSC deadline 3331 value for LAPIC timer one-shot implementation. Default 3332 back to the programmable timer unit in the LAPIC. 3333 Format: notscdeadline 3334 3335 lapic_timer_c2_ok [X86,APIC,EARLY] trust the local apic timer 3336 in C2 power state. 3337 3338 libata.dma= [LIBATA] DMA control 3339 libata.dma=0 Disable all PATA and SATA DMA 3340 libata.dma=1 PATA and SATA Disk DMA only 3341 libata.dma=2 ATAPI (CDROM) DMA only 3342 libata.dma=4 Compact Flash DMA only 3343 Combinations also work, so libata.dma=3 enables DMA 3344 for disks and CDROMs, but not CFs. 3345 3346 libata.ignore_hpa= [LIBATA] Ignore HPA limit 3347 libata.ignore_hpa=0 keep BIOS limits (default) 3348 libata.ignore_hpa=1 ignore limits, using full disk 3349 3350 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume 3351 when set. 3352 Format: <int> 3353 3354 libata.force= [LIBATA] Force configurations. The format is a comma- 3355 separated list of "[ID:]VAL" where ID is PORT[.DEVICE]. 3356 PORT and DEVICE are decimal numbers matching port, link 3357 or device. Basically, it matches the ATA ID string 3358 printed on console by libata. If the whole ID part is 3359 omitted, the last PORT and DEVICE values are used. If 3360 ID hasn't been specified yet, the configuration applies 3361 to all ports, links and devices. 3362 3363 If only DEVICE is omitted, the parameter applies to 3364 the port and all links and devices behind it. DEVICE 3365 number of 0 either selects the first device or the 3366 first fan-out link behind PMP device. It does not 3367 select the host link. DEVICE number of 15 selects the 3368 host link and device attached to it. 3369 3370 The VAL specifies the configuration to force. As long 3371 as there is no ambiguity, shortcut notation is allowed. 3372 For example, both 1.5 and 1.5G would work for 1.5Gbps. 3373 The following configurations can be forced. 3374 3375 * Cable type: 40c, 80c, short40c, unk, ign or sata. 3376 Any ID with matching PORT is used. 3377 3378 * SATA link speed limit: 1.5Gbps or 3.0Gbps. 3379 3380 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7]. 3381 udma[/][16,25,33,44,66,100,133] notation is also 3382 allowed. 3383 3384 * nohrst, nosrst, norst: suppress hard, soft and both 3385 resets. 3386 3387 * rstonce: only attempt one reset during hot-unplug 3388 link recovery. 3389 3390 * [no]dbdelay: Enable or disable the extra 200ms delay 3391 before debouncing a link PHY and device presence 3392 detection. 3393 3394 * [no]ncq: Turn on or off NCQ. 3395 3396 * [no]ncqtrim: Enable or disable queued DSM TRIM. 3397 3398 * [no]ncqati: Enable or disable NCQ trim on ATI chipset. 3399 3400 * [no]trim: Enable or disable (unqueued) TRIM. 3401 3402 * trim_zero: Indicate that TRIM command zeroes data. 3403 3404 * max_trim_128m: Set 128M maximum trim size limit. 3405 3406 * [no]dma: Turn on or off DMA transfers. 3407 3408 * atapi_dmadir: Enable ATAPI DMADIR bridge support. 3409 3410 * atapi_mod16_dma: Enable the use of ATAPI DMA for 3411 commands that are not a multiple of 16 bytes. 3412 3413 * [no]dmalog: Enable or disable the use of the 3414 READ LOG DMA EXT command to access logs. 3415 3416 * [no]iddevlog: Enable or disable access to the 3417 identify device data log. 3418 3419 * [no]logdir: Enable or disable access to the general 3420 purpose log directory. 3421 3422 * max_sec_128: Set transfer size limit to 128 sectors. 3423 3424 * max_sec_1024: Set or clear transfer size limit to 3425 1024 sectors. 3426 3427 * max_sec_lba48: Set or clear transfer size limit to 3428 65535 sectors. 3429 3430 * external: Mark port as external (hotplug-capable). 3431 3432 * [no]lpm: Enable or disable link power management. 3433 3434 * [no]setxfer: Indicate if transfer speed mode setting 3435 should be skipped. 3436 3437 * [no]fua: Disable or enable FUA (Force Unit Access) 3438 support for devices supporting this feature. 3439 3440 * dump_id: Dump IDENTIFY data. 3441 3442 * disable: Disable this device. 3443 3444 If there are multiple matching configurations changing 3445 the same attribute, the last one is used. 3446 3447 load_ramdisk= [RAM] [Deprecated] 3448 3449 lockd.nlm_grace_period=P [NFS] Assign grace period. 3450 Format: <integer> 3451 3452 lockd.nlm_tcpport=N [NFS] Assign TCP port. 3453 Format: <integer> 3454 3455 lockd.nlm_timeout=T [NFS] Assign timeout value. 3456 Format: <integer> 3457 3458 lockd.nlm_udpport=M [NFS] Assign UDP port. 3459 Format: <integer> 3460 3461 lockdown= [SECURITY,EARLY] 3462 { integrity | confidentiality } 3463 Enable the kernel lockdown feature. If set to 3464 integrity, kernel features that allow userland to 3465 modify the running kernel are disabled. If set to 3466 confidentiality, kernel features that allow userland 3467 to extract confidential information from the kernel 3468 are also disabled. 3469 3470 locktorture.acq_writer_lim= [KNL] 3471 Set the time limit in jiffies for a lock 3472 acquisition. Acquisitions exceeding this limit 3473 will result in a splat once they do complete. 3474 3475 locktorture.bind_readers= [KNL] 3476 Specify the list of CPUs to which the readers are 3477 to be bound. 3478 3479 locktorture.bind_writers= [KNL] 3480 Specify the list of CPUs to which the writers are 3481 to be bound. 3482 3483 locktorture.call_rcu_chains= [KNL] 3484 Specify the number of self-propagating call_rcu() 3485 chains to set up. These are used to ensure that 3486 there is a high probability of an RCU grace period 3487 in progress at any given time. Defaults to 0, 3488 which disables these call_rcu() chains. 3489 3490 locktorture.long_hold= [KNL] 3491 Specify the duration in milliseconds for the 3492 occasional long-duration lock hold time. Defaults 3493 to 100 milliseconds. Select 0 to disable. 3494 3495 locktorture.nested_locks= [KNL] 3496 Specify the maximum lock nesting depth that 3497 locktorture is to exercise, up to a limit of 8 3498 (MAX_NESTED_LOCKS). Specify zero to disable. 3499 Note that this parameter is ineffective on types 3500 of locks that do not support nested acquisition. 3501 3502 locktorture.nreaders_stress= [KNL] 3503 Set the number of locking read-acquisition kthreads. 3504 Defaults to being automatically set based on the 3505 number of online CPUs. 3506 3507 locktorture.nwriters_stress= [KNL] 3508 Set the number of locking write-acquisition kthreads. 3509 3510 locktorture.onoff_holdoff= [KNL] 3511 Set time (s) after boot for CPU-hotplug testing. 3512 3513 locktorture.onoff_interval= [KNL] 3514 Set time (s) between CPU-hotplug operations, or 3515 zero to disable CPU-hotplug testing. 3516 3517 locktorture.rt_boost= [KNL] 3518 Do periodic testing of real-time lock priority 3519 boosting. Select 0 to disable, 1 to boost 3520 only rt_mutex, and 2 to boost unconditionally. 3521 Defaults to 2, which might seem to be an 3522 odd choice, but which should be harmless for 3523 non-real-time spinlocks, due to their disabling 3524 of preemption. Note that non-realtime mutexes 3525 disable boosting. 3526 3527 locktorture.rt_boost_factor= [KNL] 3528 Number that determines how often and for how 3529 long priority boosting is exercised. This is 3530 scaled down by the number of writers, so that the 3531 number of boosts per unit time remains roughly 3532 constant as the number of writers increases. 3533 On the other hand, the duration of each boost 3534 increases with the number of writers. 3535 3536 locktorture.shuffle_interval= [KNL] 3537 Set task-shuffle interval (jiffies). Shuffling 3538 tasks allows some CPUs to go into dyntick-idle 3539 mode during the locktorture test. 3540 3541 locktorture.shutdown_secs= [KNL] 3542 Set time (s) after boot system shutdown. This 3543 is useful for hands-off automated testing. 3544 3545 locktorture.stat_interval= [KNL] 3546 Time (s) between statistics printk()s. 3547 3548 locktorture.stutter= [KNL] 3549 Time (s) to stutter testing, for example, 3550 specifying five seconds causes the test to run for 3551 five seconds, wait for five seconds, and so on. 3552 This tests the locking primitive's ability to 3553 transition abruptly to and from idle. 3554 3555 locktorture.torture_type= [KNL] 3556 Specify the locking implementation to test. 3557 3558 locktorture.verbose= [KNL] 3559 Enable additional printk() statements. 3560 3561 locktorture.writer_fifo= [KNL] 3562 Run the write-side locktorture kthreads at 3563 sched_set_fifo() real-time priority. 3564 3565 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver 3566 Format: <irq> 3567 3568 loglevel= [KNL,EARLY] 3569 All Kernel Messages with a loglevel smaller than the 3570 console loglevel will be printed to the console. It can 3571 also be changed with klogd or other programs. The 3572 loglevels are defined as follows: 3573 3574 0 (KERN_EMERG) system is unusable 3575 1 (KERN_ALERT) action must be taken immediately 3576 2 (KERN_CRIT) critical conditions 3577 3 (KERN_ERR) error conditions 3578 4 (KERN_WARNING) warning conditions 3579 5 (KERN_NOTICE) normal but significant condition 3580 6 (KERN_INFO) informational 3581 7 (KERN_DEBUG) debug-level messages 3582 3583 log_buf_len=n[KMG] [KNL,EARLY] 3584 Sets the size of the printk ring buffer, in bytes. 3585 n must be a power of two and greater than the 3586 minimal size. The minimal size is defined by 3587 LOG_BUF_SHIFT kernel config parameter. There 3588 is also CONFIG_LOG_CPU_MAX_BUF_SHIFT config 3589 parameter that allows to increase the default size 3590 depending on the number of CPUs. See init/Kconfig 3591 for more details. 3592 3593 logo.nologo [FB] Disables display of the built-in Linux logo. 3594 This may be used to provide more screen space for 3595 kernel log messages and is useful when debugging 3596 kernel boot problems. 3597 3598 lp=0 [LP] Specify parallel ports to use, e.g, 3599 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses 3600 lp=reset first parallel port). 'lp=0' disables the 3601 lp=auto printer driver. 'lp=reset' (which can be 3602 specified in addition to the ports) causes 3603 attached printers to be reset. Using 3604 lp=port1,port2,... specifies the parallel ports 3605 to associate lp devices with, starting with 3606 lp0. A port specification may be 'none' to skip 3607 that lp device, or a parport name such as 3608 'parport0'. Specifying 'lp=auto' instead of a 3609 port specification list means that device IDs 3610 from each port should be examined, to see if 3611 an IEEE 1284-compliant printer is attached; if 3612 so, the driver will manage that printer. 3613 See also header of drivers/char/lp.c. 3614 3615 lpj=n [KNL] 3616 Sets loops_per_jiffy to given constant, thus avoiding 3617 time-consuming boot-time autodetection (up to 250 ms per 3618 CPU). 0 enables autodetection (default). To determine 3619 the correct value for your kernel, boot with normal 3620 autodetection and see what value is printed. Note that 3621 on SMP systems the preset will be applied to all CPUs, 3622 which is likely to cause problems if your CPUs need 3623 significantly divergent settings. An incorrect value 3624 will cause delays in the kernel to be wrong, leading to 3625 unpredictable I/O errors and other breakage. Although 3626 unlikely, in the extreme case this might damage your 3627 hardware. 3628 3629 lsm.debug [SECURITY] Enable LSM initialization debugging output. 3630 3631 lsm=lsm1,...,lsmN 3632 [SECURITY] Choose order of LSM initialization. This 3633 overrides CONFIG_LSM, and the "security=" parameter. 3634 3635 machtype= [Loongson] Share the same kernel image file between 3636 different yeeloong laptops. 3637 Example: machtype=lemote-yeeloong-2f-7inch 3638 3639 maxcpus= [SMP,EARLY] Maximum number of processors that an SMP kernel 3640 will bring up during bootup. maxcpus=n : n >= 0 limits 3641 the kernel to bring up 'n' processors. Surely after 3642 bootup you can bring up the other plugged cpu by executing 3643 "echo 1 > /sys/devices/system/cpu/cpuX/online". So maxcpus 3644 only takes effect during system bootup. 3645 While n=0 is a special case, it is equivalent to "nosmp", 3646 which also disables the IO APIC. 3647 3648 max_loop= [LOOP] The number of loop block devices that get 3649 (loop.max_loop) unconditionally pre-created at init time. The default 3650 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead 3651 of statically allocating a predefined number, loop 3652 devices can be requested on-demand with the 3653 /dev/loop-control interface. 3654 3655 mce= [X86-{32,64}] 3656 3657 Please see Documentation/arch/x86/x86_64/machinecheck.rst for sysfs runtime tunables. 3658 3659 off 3660 disable machine check 3661 3662 no_cmci 3663 disable CMCI(Corrected Machine Check Interrupt) that 3664 Intel processor supports. Usually this disablement is 3665 not recommended, but it might be handy if your 3666 hardware is misbehaving. 3667 3668 Note that you'll get more problems without CMCI than 3669 with due to the shared banks, i.e. you might get 3670 duplicated error logs. 3671 3672 dont_log_ce 3673 don't make logs for corrected errors. All events 3674 reported as corrected are silently cleared by OS. This 3675 option will be useful if you have no interest in any 3676 of corrected errors. 3677 3678 ignore_ce 3679 disable features for corrected errors, e.g. 3680 polling timer and CMCI. All events reported as 3681 corrected are not cleared by OS and remained in its 3682 error banks. 3683 3684 Usually this disablement is not recommended, however 3685 if there is an agent checking/clearing corrected 3686 errors (e.g. BIOS or hardware monitoring 3687 applications), conflicting with OS's error handling, 3688 and you cannot deactivate the agent, then this option 3689 will be a help. 3690 3691 no_lmce 3692 do not opt-in to Local MCE delivery. Use legacy method 3693 to broadcast MCEs. 3694 3695 bootlog 3696 enable logging of machine checks left over from 3697 booting. Disabled by default on AMD Fam10h and older 3698 because some BIOS leave bogus ones. 3699 3700 If your BIOS doesn't do that it's a good idea to 3701 enable though to make sure you log even machine check 3702 events that result in a reboot. On Intel systems it is 3703 enabled by default. 3704 3705 nobootlog 3706 disable boot machine check logging. 3707 3708 monarchtimeout (number) 3709 sets the time in us to wait for other CPUs on machine 3710 checks. 0 to disable. 3711 3712 bios_cmci_threshold 3713 don't overwrite the bios-set CMCI threshold. This boot 3714 option prevents Linux from overwriting the CMCI 3715 threshold set by the bios. Without this option, Linux 3716 always sets the CMCI threshold to 1. Enabling this may 3717 make memory predictive failure analysis less effective 3718 if the bios sets thresholds for memory errors since we 3719 will not see details for all errors. 3720 3721 recovery 3722 force-enable recoverable machine check code paths 3723 3724 Everything else is in sysfs now. 3725 3726 3727 md= [HW] RAID subsystems devices and level 3728 See Documentation/admin-guide/md.rst. 3729 3730 mdacon= [MDA] 3731 Format: <first>,<last> 3732 Specifies range of consoles to be captured by the MDA. 3733 3734 mds= [X86,INTEL,EARLY] 3735 Control mitigation for the Micro-architectural Data 3736 Sampling (MDS) vulnerability. 3737 3738 Certain CPUs are vulnerable to an exploit against CPU 3739 internal buffers which can forward information to a 3740 disclosure gadget under certain conditions. 3741 3742 In vulnerable processors, the speculatively 3743 forwarded data can be used in a cache side channel 3744 attack, to access data to which the attacker does 3745 not have direct access. 3746 3747 This parameter controls the MDS mitigation. The 3748 options are: 3749 3750 full - Enable MDS mitigation on vulnerable CPUs 3751 full,nosmt - Enable MDS mitigation and disable 3752 SMT on vulnerable CPUs 3753 off - Unconditionally disable MDS mitigation 3754 3755 On TAA-affected machines, mds=off can be prevented by 3756 an active TAA mitigation as both vulnerabilities are 3757 mitigated with the same mechanism so in order to disable 3758 this mitigation, you need to specify tsx_async_abort=off 3759 too. 3760 3761 Not specifying this option is equivalent to 3762 mds=full. 3763 3764 For details see: Documentation/admin-guide/hw-vuln/mds.rst 3765 3766 mem=nn[KMG] [HEXAGON,EARLY] Set the memory size. 3767 Must be specified, otherwise memory size will be 0. 3768 3769 mem=nn[KMG] [KNL,BOOT,EARLY] Force usage of a specific amount 3770 of memory Amount of memory to be used in cases 3771 as follows: 3772 3773 1 for test; 3774 2 when the kernel is not able to see the whole system memory; 3775 3 memory that lies after 'mem=' boundary is excluded from 3776 the hypervisor, then assigned to KVM guests. 3777 4 to limit the memory available for kdump kernel. 3778 3779 [ARC,MICROBLAZE] - the limit applies only to low memory, 3780 high memory is not affected. 3781 3782 [ARM64] - only limits memory covered by the linear 3783 mapping. The NOMAP regions are not affected. 3784 3785 [X86] Work as limiting max address. Use together 3786 with memmap= to avoid physical address space collisions. 3787 Without memmap= PCI devices could be placed at addresses 3788 belonging to unused RAM. 3789 3790 Note that this only takes effects during boot time since 3791 in above case 3, memory may need be hot added after boot 3792 if system memory of hypervisor is not sufficient. 3793 3794 mem=nn[KMG]@ss[KMG] 3795 [ARM,MIPS,EARLY] - override the memory layout 3796 reported by firmware. 3797 Define a memory region of size nn[KMG] starting at 3798 ss[KMG]. 3799 Multiple different regions can be specified with 3800 multiple mem= parameters on the command line. 3801 3802 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel 3803 memory. 3804 3805 memblock=debug [KNL,EARLY] Enable memblock debug messages. 3806 3807 memchunk=nn[KMG] 3808 [KNL,SH] Allow user to override the default size for 3809 per-device physically contiguous DMA buffers. 3810 3811 memhp_default_state=online/offline/online_kernel/online_movable 3812 [KNL] Set the initial state for the memory hotplug 3813 onlining policy. If not specified, the default value is 3814 set according to the 3815 CONFIG_MHP_DEFAULT_ONLINE_TYPE kernel config 3816 options. 3817 See Documentation/admin-guide/mm/memory-hotplug.rst. 3818 3819 memmap=exactmap [KNL,X86,EARLY] Enable setting of an exact 3820 E820 memory map, as specified by the user. 3821 Such memmap=exactmap lines can be constructed based on 3822 BIOS output or other requirements. See the memmap=nn@ss 3823 option description. 3824 3825 memmap=nn[KMG]@ss[KMG] 3826 [KNL, X86,MIPS,XTENSA,EARLY] Force usage of a specific region of memory. 3827 Region of memory to be used is from ss to ss+nn. 3828 If @ss[KMG] is omitted, it is equivalent to mem=nn[KMG], 3829 which limits max address to nn[KMG]. 3830 Multiple different regions can be specified, 3831 comma delimited. 3832 Example: 3833 memmap=100M@2G,100M#3G,1G!1024G 3834 3835 memmap=nn[KMG]#ss[KMG] 3836 [KNL,ACPI,EARLY] Mark specific memory as ACPI data. 3837 Region of memory to be marked is from ss to ss+nn. 3838 3839 memmap=nn[KMG]$ss[KMG] 3840 [KNL,ACPI,EARLY] Mark specific memory as reserved. 3841 Region of memory to be reserved is from ss to ss+nn. 3842 Example: Exclude memory from 0x18690000-0x1869ffff 3843 memmap=64K$0x18690000 3844 or 3845 memmap=0x10000$0x18690000 3846 Some bootloaders may need an escape character before '$', 3847 like Grub2, otherwise '$' and the following number 3848 will be eaten. 3849 3850 memmap=nn[KMG]!ss[KMG,EARLY] 3851 [KNL,X86] Mark specific memory as protected. 3852 Region of memory to be used, from ss to ss+nn. 3853 The memory region may be marked as e820 type 12 (0xc) 3854 and is NVDIMM or ADR memory. 3855 3856 memmap=<size>%<offset>-<oldtype>+<newtype> 3857 [KNL,ACPI,EARLY] Convert memory within the specified region 3858 from <oldtype> to <newtype>. If "-<oldtype>" is left 3859 out, the whole region will be marked as <newtype>, 3860 even if previously unavailable. If "+<newtype>" is left 3861 out, matching memory will be removed. Types are 3862 specified as e820 types, e.g., 1 = RAM, 2 = reserved, 3863 3 = ACPI, 12 = PRAM. 3864 3865 memory_corruption_check=0/1 [X86,EARLY] 3866 Some BIOSes seem to corrupt the first 64k of 3867 memory when doing things like suspend/resume. 3868 Setting this option will scan the memory 3869 looking for corruption. Enabling this will 3870 both detect corruption and prevent the kernel 3871 from using the memory being corrupted. 3872 However, it's intended as a diagnostic tool; if 3873 repeatable BIOS-originated corruption always 3874 affects the same memory, you can use memmap= 3875 to prevent the kernel from using that memory. 3876 3877 memory_corruption_check_size=size [X86,EARLY] 3878 By default it checks for corruption in the low 3879 64k, making this memory unavailable for normal 3880 use. Use this parameter to scan for 3881 corruption in more or less memory. 3882 3883 memory_corruption_check_period=seconds [X86,EARLY] 3884 By default it checks for corruption every 60 3885 seconds. Use this parameter to check at some 3886 other rate. 0 disables periodic checking. 3887 3888 memory_hotplug.memmap_on_memory 3889 [KNL,X86,ARM] Boolean flag to enable this feature. 3890 Format: {on | off (default)} 3891 When enabled, runtime hotplugged memory will 3892 allocate its internal metadata (struct pages, 3893 those vmemmap pages cannot be optimized even 3894 if hugetlb_free_vmemmap is enabled) from the 3895 hotadded memory which will allow to hotadd a 3896 lot of memory without requiring additional 3897 memory to do so. 3898 This feature is disabled by default because it 3899 has some implication on large (e.g. GB) 3900 allocations in some configurations (e.g. small 3901 memory blocks). 3902 The state of the flag can be read in 3903 /sys/module/memory_hotplug/parameters/memmap_on_memory. 3904 Note that even when enabled, there are a few cases where 3905 the feature is not effective. 3906 3907 memtest= [KNL,X86,ARM,M68K,PPC,RISCV,EARLY] Enable memtest 3908 Format: <integer> 3909 default : 0 <disable> 3910 Specifies the number of memtest passes to be 3911 performed. Each pass selects another test 3912 pattern from a given set of patterns. Memtest 3913 fills the memory with this pattern, validates 3914 memory contents and reserves bad memory 3915 regions that are detected. 3916 3917 mem_encrypt= [X86-64] AMD Secure Memory Encryption (SME) control 3918 Valid arguments: on, off 3919 Default: off 3920 mem_encrypt=on: Activate SME 3921 mem_encrypt=off: Do not activate SME 3922 3923 Refer to Documentation/virt/kvm/x86/amd-memory-encryption.rst 3924 for details on when memory encryption can be activated. 3925 3926 mem_sleep_default= [SUSPEND] Default system suspend mode: 3927 s2idle - Suspend-To-Idle 3928 shallow - Power-On Suspend or equivalent (if supported) 3929 deep - Suspend-To-RAM or equivalent (if supported) 3930 See Documentation/admin-guide/pm/sleep-states.rst. 3931 3932 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when 3933 the BIOS has incorrectly applied a workaround. TinyBIOS 3934 version 0.98 is known to be affected, 0.99 fixes the 3935 problem by letting the user disable the workaround. 3936 3937 mga= [HW,DRM] 3938 3939 microcode= [X86] Control the behavior of the microcode loader. 3940 Available options, comma separated: 3941 3942 base_rev=X - with <X> with format: <u32> 3943 Set the base microcode revision of each thread when in 3944 debug mode. 3945 3946 dis_ucode_ldr: disable the microcode loader 3947 3948 force_minrev: 3949 Enable or disable the microcode minimal revision 3950 enforcement for the runtime microcode loader. 3951 3952 mini2440= [ARM,HW,KNL] 3953 Format:[0..2][b][c][t] 3954 Default: "0tb" 3955 MINI2440 configuration specification: 3956 0 - The attached screen is the 3.5" TFT 3957 1 - The attached screen is the 7" TFT 3958 2 - The VGA Shield is attached (1024x768) 3959 Leaving out the screen size parameter will not load 3960 the TFT driver, and the framebuffer will be left 3961 unconfigured. 3962 b - Enable backlight. The TFT backlight pin will be 3963 linked to the kernel VESA blanking code and a GPIO 3964 LED. This parameter is not necessary when using the 3965 VGA shield. 3966 c - Enable the s3c camera interface. 3967 t - Reserved for enabling touchscreen support. The 3968 touchscreen support is not enabled in the mainstream 3969 kernel as of 2.6.30, a preliminary port can be found 3970 in the "bleeding edge" mini2440 support kernel at 3971 https://repo.or.cz/w/linux-2.6/mini2440.git 3972 3973 mitigations= 3974 [X86,PPC,S390,ARM64,EARLY] Control optional mitigations for 3975 CPU vulnerabilities. This is a set of curated, 3976 arch-independent options, each of which is an 3977 aggregation of existing arch-specific options. 3978 3979 Note, "mitigations" is supported if and only if the 3980 kernel was built with CPU_MITIGATIONS=y. 3981 3982 off 3983 Disable all optional CPU mitigations. This 3984 improves system performance, but it may also 3985 expose users to several CPU vulnerabilities. 3986 Equivalent to: if nokaslr then kpti=0 [ARM64] 3987 gather_data_sampling=off [X86] 3988 indirect_target_selection=off [X86] 3989 kvm.nx_huge_pages=off [X86] 3990 l1tf=off [X86] 3991 mds=off [X86] 3992 mmio_stale_data=off [X86] 3993 no_entry_flush [PPC] 3994 no_uaccess_flush [PPC] 3995 nobp=0 [S390] 3996 nopti [X86,PPC] 3997 nospectre_bhb [ARM64] 3998 nospectre_v1 [X86,PPC] 3999 nospectre_v2 [X86,PPC,S390,ARM64] 4000 reg_file_data_sampling=off [X86] 4001 retbleed=off [X86] 4002 spec_rstack_overflow=off [X86] 4003 spec_store_bypass_disable=off [X86,PPC] 4004 spectre_bhi=off [X86] 4005 spectre_v2_user=off [X86] 4006 srbds=off [X86,INTEL] 4007 ssbd=force-off [ARM64] 4008 tsx_async_abort=off [X86] 4009 vmscape=off [X86] 4010 4011 Exceptions: 4012 This does not have any effect on 4013 kvm.nx_huge_pages when 4014 kvm.nx_huge_pages=force. 4015 4016 auto (default) 4017 Mitigate all CPU vulnerabilities, but leave SMT 4018 enabled, even if it's vulnerable. This is for 4019 users who don't want to be surprised by SMT 4020 getting disabled across kernel upgrades, or who 4021 have other ways of avoiding SMT-based attacks. 4022 Equivalent to: (default behavior) 4023 4024 auto,nosmt 4025 Mitigate all CPU vulnerabilities, disabling SMT 4026 if needed. This is for users who always want to 4027 be fully mitigated, even if it means losing SMT. 4028 Equivalent to: l1tf=flush,nosmt [X86] 4029 mds=full,nosmt [X86] 4030 tsx_async_abort=full,nosmt [X86] 4031 mmio_stale_data=full,nosmt [X86] 4032 retbleed=auto,nosmt [X86] 4033 4034 [X86] After one of the above options, additionally 4035 supports attack-vector based controls as documented in 4036 Documentation/admin-guide/hw-vuln/attack_vector_controls.rst 4037 4038 mminit_loglevel= 4039 [KNL,EARLY] When CONFIG_DEBUG_MEMORY_INIT is set, this 4040 parameter allows control of the logging verbosity for 4041 the additional memory initialisation checks. A value 4042 of 0 disables mminit logging and a level of 4 will 4043 log everything. Information is printed at KERN_DEBUG 4044 so loglevel=8 may also need to be specified. 4045 4046 mmio_stale_data= 4047 [X86,INTEL,EARLY] Control mitigation for the Processor 4048 MMIO Stale Data vulnerabilities. 4049 4050 Processor MMIO Stale Data is a class of 4051 vulnerabilities that may expose data after an MMIO 4052 operation. Exposed data could originate or end in 4053 the same CPU buffers as affected by MDS and TAA. 4054 Therefore, similar to MDS and TAA, the mitigation 4055 is to clear the affected CPU buffers. 4056 4057 This parameter controls the mitigation. The 4058 options are: 4059 4060 full - Enable mitigation on vulnerable CPUs 4061 4062 full,nosmt - Enable mitigation and disable SMT on 4063 vulnerable CPUs. 4064 4065 off - Unconditionally disable mitigation 4066 4067 On MDS or TAA affected machines, 4068 mmio_stale_data=off can be prevented by an active 4069 MDS or TAA mitigation as these vulnerabilities are 4070 mitigated with the same mechanism so in order to 4071 disable this mitigation, you need to specify 4072 mds=off and tsx_async_abort=off too. 4073 4074 Not specifying this option is equivalent to 4075 mmio_stale_data=full. 4076 4077 For details see: 4078 Documentation/admin-guide/hw-vuln/processor_mmio_stale_data.rst 4079 4080 <module>.async_probe[=<bool>] [KNL] 4081 If no <bool> value is specified or if the value 4082 specified is not a valid <bool>, enable asynchronous 4083 probe on this module. Otherwise, enable/disable 4084 asynchronous probe on this module as indicated by the 4085 <bool> value. See also: module.async_probe 4086 4087 module.async_probe=<bool> 4088 [KNL] When set to true, modules will use async probing 4089 by default. To enable/disable async probing for a 4090 specific module, use the module specific control that 4091 is documented under <module>.async_probe. When both 4092 module.async_probe and <module>.async_probe are 4093 specified, <module>.async_probe takes precedence for 4094 the specific module. 4095 4096 module.enable_dups_trace 4097 [KNL] When CONFIG_MODULE_DEBUG_AUTOLOAD_DUPS is set, 4098 this means that duplicate request_module() calls will 4099 trigger a WARN_ON() instead of a pr_warn(). Note that 4100 if MODULE_DEBUG_AUTOLOAD_DUPS_TRACE is set, WARN_ON()s 4101 will always be issued and this option does nothing. 4102 module.sig_enforce 4103 [KNL] When CONFIG_MODULE_SIG is set, this means that 4104 modules without (valid) signatures will fail to load. 4105 Note that if CONFIG_MODULE_SIG_FORCE is set, that 4106 is always true, so this option does nothing. 4107 4108 module_blacklist= [KNL] Do not load a comma-separated list of 4109 modules. Useful for debugging problem modules. 4110 4111 mousedev.tap_time= 4112 [MOUSE] Maximum time between finger touching and 4113 leaving touchpad surface for touch to be considered 4114 a tap and be reported as a left button click (for 4115 touchpads working in absolute mode only). 4116 Format: <msecs> 4117 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices 4118 reporting absolute coordinates, such as tablets 4119 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices 4120 reporting absolute coordinates, such as tablets 4121 4122 movablecore= [KNL,X86,PPC,EARLY] 4123 Format: nn[KMGTPE] | nn% 4124 This parameter is the complement to kernelcore=, it 4125 specifies the amount of memory used for migratable 4126 allocations. If both kernelcore and movablecore is 4127 specified, then kernelcore will be at *least* the 4128 specified value but may be more. If movablecore on its 4129 own is specified, the administrator must be careful 4130 that the amount of memory usable for all allocations 4131 is not too small. 4132 4133 movable_node [KNL,EARLY] Boot-time switch to make hotplugable memory 4134 NUMA nodes to be movable. This means that the memory 4135 of such nodes will be usable only for movable 4136 allocations which rules out almost all kernel 4137 allocations. Use with caution! 4138 4139 MTD_Partition= [MTD] 4140 Format: <name>,<region-number>,<size>,<offset> 4141 4142 MTD_Region= [MTD] Format: 4143 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>] 4144 4145 mtdparts= [MTD] 4146 See drivers/mtd/parsers/cmdlinepart.c 4147 4148 mtouchusb.raw_coordinates= 4149 [HW] Make the MicroTouch USB driver use raw coordinates 4150 ('y', default) or cooked coordinates ('n') 4151 4152 mtrr=debug [X86,EARLY] 4153 Enable printing debug information related to MTRR 4154 registers at boot time. 4155 4156 mtrr_chunk_size=nn[KMG,X86,EARLY] 4157 used for mtrr cleanup. It is largest continuous chunk 4158 that could hold holes aka. UC entries. 4159 4160 mtrr_gran_size=nn[KMG,X86,EARLY] 4161 Used for mtrr cleanup. It is granularity of mtrr block. 4162 Default is 1. 4163 Large value could prevent small alignment from 4164 using up MTRRs. 4165 4166 mtrr_spare_reg_nr=n [X86,EARLY] 4167 Format: <integer> 4168 Range: 0,7 : spare reg number 4169 Default : 1 4170 Used for mtrr cleanup. It is spare mtrr entries number. 4171 Set to 2 or more if your graphical card needs more. 4172 4173 multitce=off [PPC] This parameter disables the use of the pSeries 4174 firmware feature for updating multiple TCE entries 4175 at a time. 4176 4177 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card 4178 4179 netdev= [NET] Network devices parameters 4180 Format: <irq>,<io>,<mem_start>,<mem_end>,<name> 4181 Note that mem_start is often overloaded to mean 4182 something different and driver-specific. 4183 This usage is only documented in each driver source 4184 file if at all. 4185 4186 netpoll.carrier_timeout= 4187 [NET] Specifies amount of time (in seconds) that 4188 netpoll should wait for a carrier. By default netpoll 4189 waits 4 seconds. 4190 4191 nf_conntrack.acct= 4192 [NETFILTER] Enable connection tracking flow accounting 4193 0 to disable accounting 4194 1 to enable accounting 4195 Default value is 0. 4196 4197 nfs.cache_getent= 4198 [NFS] sets the pathname to the program which is used 4199 to update the NFS client cache entries. 4200 4201 nfs.cache_getent_timeout= 4202 [NFS] sets the timeout after which an attempt to 4203 update a cache entry is deemed to have failed. 4204 4205 nfs.callback_nr_threads= 4206 [NFSv4] set the total number of threads that the 4207 NFS client will assign to service NFSv4 callback 4208 requests. 4209 4210 nfs.callback_tcpport= 4211 [NFS] set the TCP port on which the NFSv4 callback 4212 channel should listen. 4213 4214 nfs.delay_retrans= 4215 [NFS] specifies the number of times the NFSv4 client 4216 retries the request before returning an EAGAIN error, 4217 after a reply of NFS4ERR_DELAY from the server. 4218 Only applies if the softerr mount option is enabled, 4219 and the specified value is >= 0. 4220 4221 nfs.enable_ino64= 4222 [NFS] enable 64-bit inode numbers. 4223 If zero, the NFS client will fake up a 32-bit inode 4224 number for the readdir() and stat() syscalls instead 4225 of returning the full 64-bit number. 4226 The default is to return 64-bit inode numbers. 4227 4228 nfs.idmap_cache_timeout= 4229 [NFS] set the maximum lifetime for idmapper cache 4230 entries. 4231 4232 nfs.max_session_cb_slots= 4233 [NFSv4.1] Sets the maximum number of session 4234 slots the client will assign to the callback 4235 channel. This determines the maximum number of 4236 callbacks the client will process in parallel for 4237 a particular server. 4238 4239 nfs.max_session_slots= 4240 [NFSv4.1] Sets the maximum number of session slots 4241 the client will attempt to negotiate with the server. 4242 This limits the number of simultaneous RPC requests 4243 that the client can send to the NFSv4.1 server. 4244 Note that there is little point in setting this 4245 value higher than the max_tcp_slot_table_limit. 4246 4247 nfs.nfs4_disable_idmapping= 4248 [NFSv4] When set to the default of '1', this option 4249 ensures that both the RPC level authentication 4250 scheme and the NFS level operations agree to use 4251 numeric uids/gids if the mount is using the 4252 'sec=sys' security flavour. In effect it is 4253 disabling idmapping, which can make migration from 4254 legacy NFSv2/v3 systems to NFSv4 easier. 4255 Servers that do not support this mode of operation 4256 will be autodetected by the client, and it will fall 4257 back to using the idmapper. 4258 To turn off this behaviour, set the value to '0'. 4259 4260 nfs.nfs4_unique_id= 4261 [NFS4] Specify an additional fixed unique ident- 4262 ification string that NFSv4 clients can insert into 4263 their nfs_client_id4 string. This is typically a 4264 UUID that is generated at system install time. 4265 4266 nfs.recover_lost_locks= 4267 [NFSv4] Attempt to recover locks that were lost due 4268 to a lease timeout on the server. Please note that 4269 doing this risks data corruption, since there are 4270 no guarantees that the file will remain unchanged 4271 after the locks are lost. 4272 If you want to enable the kernel legacy behaviour of 4273 attempting to recover these locks, then set this 4274 parameter to '1'. 4275 The default parameter value of '0' causes the kernel 4276 not to attempt recovery of lost locks. 4277 4278 nfs.send_implementation_id= 4279 [NFSv4.1] Send client implementation identification 4280 information in exchange_id requests. 4281 If zero, no implementation identification information 4282 will be sent. 4283 The default is to send the implementation identification 4284 information. 4285 4286 nfs4.layoutstats_timer= 4287 [NFSv4.2] Change the rate at which the kernel sends 4288 layoutstats to the pNFS metadata server. 4289 4290 Setting this to value to 0 causes the kernel to use 4291 whatever value is the default set by the layout 4292 driver. A non-zero value sets the minimum interval 4293 in seconds between layoutstats transmissions. 4294 4295 nfsd.inter_copy_offload_enable= 4296 [NFSv4.2] When set to 1, the server will support 4297 server-to-server copies for which this server is 4298 the destination of the copy. 4299 4300 nfsd.nfs4_disable_idmapping= 4301 [NFSv4] When set to the default of '1', the NFSv4 4302 server will return only numeric uids and gids to 4303 clients using auth_sys, and will accept numeric uids 4304 and gids from such clients. This is intended to ease 4305 migration from NFSv2/v3. 4306 4307 nfsd.nfsd4_ssc_umount_timeout= 4308 [NFSv4.2] When used as the destination of a 4309 server-to-server copy, knfsd temporarily mounts 4310 the source server. It caches the mount in case 4311 it will be needed again, and discards it if not 4312 used for the number of milliseconds specified by 4313 this parameter. 4314 4315 nfsaddrs= [NFS] Deprecated. Use ip= instead. 4316 See Documentation/admin-guide/nfs/nfsroot.rst. 4317 4318 nfsroot= [NFS] nfs root filesystem for disk-less boxes. 4319 See Documentation/admin-guide/nfs/nfsroot.rst. 4320 4321 nfsrootdebug [NFS] enable nfsroot debugging messages. 4322 See Documentation/admin-guide/nfs/nfsroot.rst. 4323 4324 nmi_backtrace.backtrace_idle [KNL] 4325 Dump stacks even of idle CPUs in response to an 4326 NMI stack-backtrace request. 4327 4328 nmi_debug= [KNL,SH] Specify one or more actions to take 4329 when a NMI is triggered. 4330 Format: [state][,regs][,debounce][,die] 4331 4332 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels 4333 Format: [panic,][nopanic,][rNNN,][num] 4334 Valid num: 0 or 1 4335 0 - turn hardlockup detector in nmi_watchdog off 4336 1 - turn hardlockup detector in nmi_watchdog on 4337 rNNN - configure the watchdog with raw perf event 0xNNN 4338 4339 When panic is specified, panic when an NMI watchdog 4340 timeout occurs (or 'nopanic' to not panic on an NMI 4341 watchdog, if CONFIG_BOOTPARAM_HARDLOCKUP_PANIC is set) 4342 To disable both hard and soft lockup detectors, 4343 please see 'nowatchdog'. 4344 This is useful when you use a panic=... timeout and 4345 need the box quickly up again. 4346 4347 These settings can be accessed at runtime via 4348 the nmi_watchdog and hardlockup_panic sysctls. 4349 4350 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths 4351 emulation library even if a 387 maths coprocessor 4352 is present. 4353 4354 no4lvl [RISCV,EARLY] Disable 4-level and 5-level paging modes. 4355 Forces kernel to use 3-level paging instead. 4356 4357 no5lvl [X86-64,RISCV,EARLY] Disable 5-level paging mode. Forces 4358 kernel to use 4-level paging instead. 4359 4360 noalign [KNL,ARM] 4361 4362 noapic [SMP,APIC,EARLY] Tells the kernel to not make use of any 4363 IOAPICs that may be present in the system. 4364 4365 noapictimer [APIC,X86] Don't set up the APIC timer 4366 4367 noautogroup Disable scheduler automatic task group creation. 4368 4369 nocache [ARM,EARLY] 4370 4371 no_console_suspend 4372 [HW] Never suspend the console 4373 Disable suspending of consoles during suspend and 4374 hibernate operations. Once disabled, debugging 4375 messages can reach various consoles while the rest 4376 of the system is being put to sleep (ie, while 4377 debugging driver suspend/resume hooks). This may 4378 not work reliably with all consoles, but is known 4379 to work with serial and VGA consoles. 4380 To facilitate more flexible debugging, we also add 4381 console_suspend, a printk module parameter to control 4382 it. Users could use console_suspend (usually 4383 /sys/module/printk/parameters/console_suspend) to 4384 turn on/off it dynamically. 4385 4386 no_debug_objects 4387 [KNL,EARLY] Disable object debugging 4388 4389 nodsp [SH] Disable hardware DSP at boot time. 4390 4391 noefi [EFI,EARLY] Disable EFI runtime services support. 4392 4393 no_entry_flush [PPC,EARLY] Don't flush the L1-D cache when entering the kernel. 4394 4395 noexec32 [X86-64] 4396 This affects only 32-bit executables. 4397 noexec32=on: enable non-executable mappings (default) 4398 read doesn't imply executable mappings 4399 noexec32=off: disable non-executable mappings 4400 read implies executable mappings 4401 4402 no_file_caps Tells the kernel not to honor file capabilities. The 4403 only way then for a file to be executed with privilege 4404 is to be setuid root or executed by root. 4405 4406 nofpu [MIPS,SH] Disable hardware FPU at boot time. 4407 4408 nofsgsbase [X86] Disables FSGSBASE instructions. 4409 4410 nofxsr [BUGS=X86-32] Disables x86 floating point extended 4411 register save and restore. The kernel will only save 4412 legacy floating-point registers on task switch. 4413 4414 nogbpages [X86] Do not use GB pages for kernel direct mappings. 4415 4416 no_hash_pointers 4417 [KNL,EARLY] 4418 Alias for "hash_pointers=never". 4419 4420 nohibernate [HIBERNATION] Disable hibernation and resume. 4421 4422 nohlt [ARM,ARM64,MICROBLAZE,MIPS,PPC,RISCV,SH] Forces the kernel to 4423 busy wait in do_idle() and not use the arch_cpu_idle() 4424 implementation; requires CONFIG_GENERIC_IDLE_POLL_SETUP 4425 to be effective. This is useful on platforms where the 4426 sleep(SH) or wfi(ARM,ARM64) instructions do not work 4427 correctly or when doing power measurements to evaluate 4428 the impact of the sleep instructions. This is also 4429 useful when using JTAG debugger. 4430 4431 nohpet [X86] Don't use the HPET timer. 4432 4433 nohugeiomap [KNL,X86,PPC,ARM64,EARLY] Disable kernel huge I/O mappings. 4434 4435 nohugevmalloc [KNL,X86,PPC,ARM64,EARLY] Disable kernel huge vmalloc mappings. 4436 4437 nohz= [KNL] Boottime enable/disable dynamic ticks 4438 Valid arguments: on, off 4439 Default: on 4440 4441 nohz_full= [KNL,BOOT,SMP,ISOL] 4442 The argument is a cpu list, as described above. 4443 In kernels built with CONFIG_NO_HZ_FULL=y, set 4444 the specified list of CPUs whose tick will be stopped 4445 whenever possible. The boot CPU will be forced outside 4446 the range to maintain the timekeeping. Any CPUs 4447 in this list will have their RCU callbacks offloaded, 4448 just as if they had also been called out in the 4449 rcu_nocbs= boot parameter. 4450 4451 Note that this argument takes precedence over 4452 the CONFIG_RCU_NOCB_CPU_DEFAULT_ALL option. 4453 4454 noinitrd [RAM] Tells the kernel not to load any configured 4455 initial RAM disk. 4456 4457 nointremap [X86-64,Intel-IOMMU,EARLY] Do not enable interrupt 4458 remapping. 4459 [Deprecated - use intremap=off] 4460 4461 noinvpcid [X86,EARLY] Disable the INVPCID cpu feature. 4462 4463 noiotrap [SH] Disables trapped I/O port accesses. 4464 4465 noirqdebug [X86-32] Disables the code which attempts to detect and 4466 disable unhandled interrupt sources. 4467 4468 noisapnp [ISAPNP] Disables ISA PnP code. 4469 4470 nokaslr [KNL,EARLY] 4471 When CONFIG_RANDOMIZE_BASE is set, this disables 4472 kernel and module base offset ASLR (Address Space 4473 Layout Randomization). 4474 4475 no-kvmapf [X86,KVM,EARLY] Disable paravirtualized asynchronous page 4476 fault handling. 4477 4478 no-kvmclock [X86,KVM,EARLY] Disable paravirtualized KVM clock driver 4479 4480 nolapic [X86-32,APIC,EARLY] Do not enable or use the local APIC. 4481 4482 nolapic_timer [X86-32,APIC,EARLY] Do not use the local APIC timer. 4483 4484 nomce [X86-32] Disable Machine Check Exception 4485 4486 nomfgpt [X86-32] Disable Multi-Function General Purpose 4487 Timer usage (for AMD Geode machines). 4488 4489 nomodeset Disable kernel modesetting. Most systems' firmware 4490 sets up a display mode and provides framebuffer memory 4491 for output. With nomodeset, DRM and fbdev drivers will 4492 not load if they could possibly displace the pre- 4493 initialized output. Only the system framebuffer will 4494 be available for use. The respective drivers will not 4495 perform display-mode changes or accelerated rendering. 4496 4497 Useful as error fallback, or for testing and debugging. 4498 4499 nomodule Disable module load 4500 4501 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to 4502 shutdown the other cpus. Instead use the REBOOT_VECTOR 4503 irq. 4504 4505 nopat [X86,EARLY] Disable PAT (page attribute table extension of 4506 pagetables) support. 4507 4508 nopcid [X86-64,EARLY] Disable the PCID cpu feature. 4509 4510 nopku [X86] Disable Memory Protection Keys CPU feature found 4511 in some Intel CPUs. 4512 4513 nopti [X86-64,EARLY] 4514 Equivalent to pti=off 4515 4516 nopv= [X86,XEN,KVM,HYPER_V,VMWARE,EARLY] 4517 Disables the PV optimizations forcing the guest to run 4518 as generic guest with no PV drivers. Currently support 4519 XEN HVM, KVM, HYPER_V and VMWARE guest. 4520 4521 nopvspin [X86,XEN,KVM,EARLY] 4522 Disables the qspinlock slow path using PV optimizations 4523 which allow the hypervisor to 'idle' the guest on lock 4524 contention. 4525 4526 norandmaps Don't use address space randomization. Equivalent to 4527 echo 0 > /proc/sys/kernel/randomize_va_space 4528 4529 noreplace-smp [X86-32,SMP] Don't replace SMP instructions 4530 with UP alternatives 4531 4532 noresume [SWSUSP] Disables resume and restores original swap 4533 space. 4534 4535 no-scroll [VGA] Disables scrollback. 4536 This is required for the Braillex ib80-piezo Braille 4537 reader made by F.H. Papenmeier (Germany). 4538 4539 nosgx [X86-64,SGX,EARLY] Disables Intel SGX kernel support. 4540 4541 nosmap [PPC,EARLY] 4542 Disable SMAP (Supervisor Mode Access Prevention) 4543 even if it is supported by processor. 4544 4545 nosmep [PPC64s,EARLY] 4546 Disable SMEP (Supervisor Mode Execution Prevention) 4547 even if it is supported by processor. 4548 4549 nosmp [SMP,EARLY] Tells an SMP kernel to act as a UP kernel, 4550 and disable the IO APIC. legacy for "maxcpus=0". 4551 4552 nosmt [KNL,MIPS,PPC,EARLY] Disable symmetric multithreading (SMT). 4553 Equivalent to smt=1. 4554 4555 [KNL,X86,PPC,S390] Disable symmetric multithreading (SMT). 4556 nosmt=force: Force disable SMT, cannot be undone 4557 via the sysfs control file. 4558 4559 nosoftlockup [KNL] Disable the soft-lockup detector. 4560 4561 nospec_store_bypass_disable 4562 [HW,EARLY] Disable all mitigations for the Speculative 4563 Store Bypass vulnerability 4564 4565 nospectre_bhb [ARM64,EARLY] Disable all mitigations for Spectre-BHB (branch 4566 history injection) vulnerability. System may allow data leaks 4567 with this option. 4568 4569 nospectre_v1 [X86,PPC,EARLY] Disable mitigations for Spectre Variant 1 4570 (bounds check bypass). With this option data leaks are 4571 possible in the system. 4572 4573 nospectre_v2 [X86,PPC_E500,ARM64,EARLY] Disable all mitigations 4574 for the Spectre variant 2 (indirect branch 4575 prediction) vulnerability. System may allow data 4576 leaks with this option. 4577 4578 no-steal-acc [X86,PV_OPS,ARM64,PPC/PSERIES,RISCV,LOONGARCH,EARLY] 4579 Disable paravirtualized steal time accounting. steal time 4580 is computed, but won't influence scheduler behaviour 4581 4582 nosync [HW,M68K] Disables sync negotiation for all devices. 4583 4584 no_timer_check [X86,APIC] Disables the code which tests for broken 4585 timer IRQ sources, i.e., the IO-APIC timer. This can 4586 work around problems with incorrect timer 4587 initialization on some boards. 4588 4589 no_uaccess_flush 4590 [PPC,EARLY] Don't flush the L1-D cache after accessing user data. 4591 4592 novmcoredd [KNL,KDUMP] 4593 Disable device dump. Device dump allows drivers to 4594 append dump data to vmcore so you can collect driver 4595 specified debug info. Drivers can append the data 4596 without any limit and this data is stored in memory, 4597 so this may cause significant memory stress. Disabling 4598 device dump can help save memory but the driver debug 4599 data will be no longer available. This parameter 4600 is only available when CONFIG_PROC_VMCORE_DEVICE_DUMP 4601 is set. 4602 4603 no-vmw-sched-clock 4604 [X86,PV_OPS,EARLY] Disable paravirtualized VMware 4605 scheduler clock and use the default one. 4606 4607 nowatchdog [KNL] Disable both lockup detectors, i.e. 4608 soft-lockup and NMI watchdog (hard-lockup). 4609 4610 nowb [ARM,EARLY] 4611 4612 nox2apic [X86-64,APIC,EARLY] Do not enable x2APIC mode. 4613 4614 NOTE: this parameter will be ignored on systems with the 4615 LEGACY_XAPIC_DISABLED bit set in the 4616 IA32_XAPIC_DISABLE_STATUS MSR. 4617 4618 noxsave [BUGS=X86] Disables x86 extended register state save 4619 and restore using xsave. The kernel will fallback to 4620 enabling legacy floating-point and sse state. 4621 4622 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended 4623 register states. The kernel will fall back to use 4624 xsave to save the states. By using this parameter, 4625 performance of saving the states is degraded because 4626 xsave doesn't support modified optimization while 4627 xsaveopt supports it on xsaveopt enabled systems. 4628 4629 noxsaves [X86] Disables xsaves and xrstors used in saving and 4630 restoring x86 extended register state in compacted 4631 form of xsave area. The kernel will fall back to use 4632 xsaveopt and xrstor to save and restore the states 4633 in standard form of xsave area. By using this 4634 parameter, xsave area per process might occupy more 4635 memory on xsaves enabled systems. 4636 4637 nr_cpus= [SMP,EARLY] Maximum number of processors that an SMP kernel 4638 could support. nr_cpus=n : n >= 1 limits the kernel to 4639 support 'n' processors. It could be larger than the 4640 number of already plugged CPU during bootup, later in 4641 runtime you can physically add extra cpu until it reaches 4642 n. So during boot up some boot time memory for per-cpu 4643 variables need be pre-allocated for later physical cpu 4644 hot plugging. 4645 4646 nr_uarts= [SERIAL] maximum number of UARTs to be registered. 4647 4648 numa=off [KNL, ARM64, PPC, RISCV, SPARC, X86, EARLY] 4649 Disable NUMA, Only set up a single NUMA node 4650 spanning all memory. 4651 4652 numa=fake=<size>[MG] 4653 [KNL, ARM64, RISCV, X86, EARLY] 4654 If given as a memory unit, fills all system RAM with 4655 nodes of size interleaved over physical nodes. 4656 4657 numa=fake=<N> 4658 [KNL, ARM64, RISCV, X86, EARLY] 4659 If given as an integer, fills all system RAM with N 4660 fake nodes interleaved over physical nodes. 4661 4662 numa=fake=<N>U 4663 [KNL, ARM64, RISCV, X86, EARLY] 4664 If given as an integer followed by 'U', it will 4665 divide each physical node into N emulated nodes. 4666 4667 numa=noacpi [X86] Don't parse the SRAT table for NUMA setup 4668 4669 numa=nohmat [X86] Don't parse the HMAT table for NUMA setup, or 4670 soft-reserved memory partitioning. 4671 4672 numa_balancing= [KNL,ARM64,PPC,RISCV,S390,X86] Enable or disable automatic 4673 NUMA balancing. 4674 Allowed values are enable and disable 4675 4676 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA. 4677 'node', 'default' can be specified 4678 This can be set from sysctl after boot. 4679 See Documentation/admin-guide/sysctl/vm.rst for details. 4680 4681 ohci1394_dma=early [HW,EARLY] enable debugging via the ohci1394 driver. 4682 See Documentation/core-api/debugging-via-ohci1394.rst for more 4683 info. 4684 4685 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands 4686 Rather than timing out after 20 ms if an EC 4687 command is not properly ACKed, override the length 4688 of the timeout. We have interrupts disabled while 4689 waiting for the ACK, so if this is set too high 4690 interrupts *may* be lost! 4691 4692 omap_mux= [OMAP] Override bootloader pin multiplexing. 4693 Format: <mux_mode0.mode_name=value>... 4694 For example, to override I2C bus2: 4695 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100 4696 4697 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration 4698 4699 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock] 4700 4701 boundary - index of last SLC block on Flex-OneNAND. 4702 The remaining blocks are configured as MLC blocks. 4703 lock - Configure if Flex-OneNAND boundary should be locked. 4704 Once locked, the boundary cannot be changed. 4705 1 indicates lock status, 0 indicates unlock status. 4706 4707 oops=panic [KNL,EARLY] 4708 Always panic on oopses. Default is to just kill the 4709 process, but there is a small probability of 4710 deadlocking the machine. 4711 This will also cause panics on machine check exceptions. 4712 Useful together with panic=30 to trigger a reboot. 4713 4714 page_alloc.shuffle= 4715 [KNL] Boolean flag to control whether the page allocator 4716 should randomize its free lists. This parameter can be 4717 used to enable/disable page randomization. The state of 4718 the flag can be read from sysfs at: 4719 /sys/module/page_alloc/parameters/shuffle. 4720 This parameter is only available if CONFIG_SHUFFLE_PAGE_ALLOCATOR=y. 4721 4722 page_owner= [KNL,EARLY] Boot-time page_owner enabling option. 4723 Storage of the information about who allocated 4724 each page is disabled in default. With this switch, 4725 we can turn it on. 4726 on: enable the feature 4727 4728 page_poison= [KNL,EARLY] Boot-time parameter changing the state of 4729 poisoning on the buddy allocator, available with 4730 CONFIG_PAGE_POISONING=y. 4731 off: turn off poisoning (default) 4732 on: turn on poisoning 4733 4734 page_reporting.page_reporting_order= 4735 [KNL] Minimal page reporting order 4736 Format: <integer> 4737 Adjust the minimal page reporting order. The page 4738 reporting is disabled when it exceeds MAX_PAGE_ORDER. 4739 4740 panic= [KNL] Kernel behaviour on panic: delay <timeout> 4741 timeout > 0: seconds before rebooting 4742 timeout = 0: wait forever 4743 timeout < 0: reboot immediately 4744 Format: <timeout> 4745 4746 panic_on_taint= [KNL,EARLY] 4747 Bitmask for conditionally calling panic() in add_taint() 4748 Format: <hex>[,nousertaint] 4749 Hexadecimal bitmask representing the set of TAINT flags 4750 that will cause the kernel to panic when add_taint() is 4751 called with any of the flags in this set. 4752 The optional switch "nousertaint" can be utilized to 4753 prevent userspace forced crashes by writing to sysctl 4754 /proc/sys/kernel/tainted any flagset matching with the 4755 bitmask set on panic_on_taint. 4756 See Documentation/admin-guide/tainted-kernels.rst for 4757 extra details on the taint flags that users can pick 4758 to compose the bitmask to assign to panic_on_taint. 4759 4760 panic_on_warn=1 panic() instead of WARN(). Useful to cause kdump 4761 on a WARN(). 4762 4763 panic_print= Bitmask for printing system info when panic happens. 4764 User can chose combination of the following bits: 4765 bit 0: print all tasks info 4766 bit 1: print system memory info 4767 bit 2: print timer info 4768 bit 3: print locks info if CONFIG_LOCKDEP is on 4769 bit 4: print ftrace buffer 4770 bit 5: replay all kernel messages on consoles at the end of panic 4771 bit 6: print all CPUs backtrace (if available in the arch) 4772 bit 7: print only tasks in uninterruptible (blocked) state 4773 *Be aware* that this option may print a _lot_ of lines, 4774 so there are risks of losing older messages in the log. 4775 Use this option carefully, maybe worth to setup a 4776 bigger log buffer with "log_buf_len" along with this. 4777 4778 panic_sys_info= A comma separated list of extra information to be dumped 4779 on panic. 4780 Format: val[,val...] 4781 Where @val can be any of the following: 4782 4783 tasks: print all tasks info 4784 mem: print system memory info 4785 timers: print timers info 4786 locks: print locks info if CONFIG_LOCKDEP is on 4787 ftrace: print ftrace buffer 4788 all_bt: print all CPUs backtrace (if available in the arch) 4789 blocked_tasks: print only tasks in uninterruptible (blocked) state 4790 4791 This is a human readable alternative to the 'panic_print' option. 4792 4793 panic_console_replay 4794 When panic happens, replay all kernel messages on 4795 consoles at the end of panic. 4796 4797 parkbd.port= [HW] Parallel port number the keyboard adapter is 4798 connected to, default is 0. 4799 Format: <parport#> 4800 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation, 4801 0 for XT, 1 for AT (default is AT). 4802 Format: <mode> 4803 4804 parport= [HW,PPT] Specify parallel ports. 0 disables. 4805 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] } 4806 Use 'auto' to force the driver to use any 4807 IRQ/DMA settings detected (the default is to 4808 ignore detected IRQ/DMA settings because of 4809 possible conflicts). You can specify the base 4810 address, IRQ, and DMA settings; IRQ and DMA 4811 should be numbers, or 'auto' (for using detected 4812 settings on that particular port), or 'nofifo' 4813 (to avoid using a FIFO even if it is detected). 4814 Parallel ports are assigned in the order they 4815 are specified on the command line, starting 4816 with parport0. 4817 4818 parport_init_mode= [HW,PPT] 4819 Configure VIA parallel port to operate in 4820 a specific mode. This is necessary on Pegasos 4821 computer where firmware has no options for setting 4822 up parallel port mode and sets it to spp. 4823 Currently this function knows 686a and 8231 chips. 4824 Format: [spp|ps2|epp|ecp|ecpepp] 4825 4826 pata_legacy.all= [HW,LIBATA] 4827 Format: <int> 4828 Set to non-zero to probe primary and secondary ISA 4829 port ranges on PCI systems where no PCI PATA device 4830 has been found at either range. Disabled by default. 4831 4832 pata_legacy.autospeed= [HW,LIBATA] 4833 Format: <int> 4834 Set to non-zero if a chip is present that snoops speed 4835 changes. Disabled by default. 4836 4837 pata_legacy.ht6560a= [HW,LIBATA] 4838 Format: <int> 4839 Set to 1, 2, or 3 for HT 6560A on the primary channel, 4840 the secondary channel, or both channels respectively. 4841 Disabled by default. 4842 4843 pata_legacy.ht6560b= [HW,LIBATA] 4844 Format: <int> 4845 Set to 1, 2, or 3 for HT 6560B on the primary channel, 4846 the secondary channel, or both channels respectively. 4847 Disabled by default. 4848 4849 pata_legacy.iordy_mask= [HW,LIBATA] 4850 Format: <int> 4851 IORDY enable mask. Set individual bits to allow IORDY 4852 for the respective channel. Bit 0 is for the first 4853 legacy channel handled by this driver, bit 1 is for 4854 the second channel, and so on. The sequence will often 4855 correspond to the primary legacy channel, the secondary 4856 legacy channel, and so on, but the handling of a PCI 4857 bus and the use of other driver options may interfere 4858 with the sequence. By default IORDY is allowed across 4859 all channels. 4860 4861 pata_legacy.opti82c46x= [HW,LIBATA] 4862 Format: <int> 4863 Set to 1, 2, or 3 for Opti 82c611A on the primary 4864 channel, the secondary channel, or both channels 4865 respectively. Disabled by default. 4866 4867 pata_legacy.opti82c611a= [HW,LIBATA] 4868 Format: <int> 4869 Set to 1, 2, or 3 for Opti 82c465MV on the primary 4870 channel, the secondary channel, or both channels 4871 respectively. Disabled by default. 4872 4873 pata_legacy.pio_mask= [HW,LIBATA] 4874 Format: <int> 4875 PIO mode mask for autospeed devices. Set individual 4876 bits to allow the use of the respective PIO modes. 4877 Bit 0 is for mode 0, bit 1 is for mode 1, and so on. 4878 All modes allowed by default. 4879 4880 pata_legacy.probe_all= [HW,LIBATA] 4881 Format: <int> 4882 Set to non-zero to probe tertiary and further ISA 4883 port ranges on PCI systems. Disabled by default. 4884 4885 pata_legacy.probe_mask= [HW,LIBATA] 4886 Format: <int> 4887 Probe mask for legacy ISA PATA ports. Depending on 4888 platform configuration and the use of other driver 4889 options up to 6 legacy ports are supported: 0x1f0, 4890 0x170, 0x1e8, 0x168, 0x1e0, 0x160, however probing 4891 of individual ports can be disabled by setting the 4892 corresponding bits in the mask to 1. Bit 0 is for 4893 the first port in the list above (0x1f0), and so on. 4894 By default all supported ports are probed. 4895 4896 pata_legacy.qdi= [HW,LIBATA] 4897 Format: <int> 4898 Set to non-zero to probe QDI controllers. By default 4899 set to 1 if CONFIG_PATA_QDI_MODULE, 0 otherwise. 4900 4901 pata_legacy.winbond= [HW,LIBATA] 4902 Format: <int> 4903 Set to non-zero to probe Winbond controllers. Use 4904 the standard I/O port (0x130) if 1, otherwise the 4905 value given is the I/O port to use (typically 0x1b0). 4906 By default set to 1 if CONFIG_PATA_WINBOND_VLB_MODULE, 4907 0 otherwise. 4908 4909 pata_platform.pio_mask= [HW,LIBATA] 4910 Format: <int> 4911 Supported PIO mode mask. Set individual bits to allow 4912 the use of the respective PIO modes. Bit 0 is for 4913 mode 0, bit 1 is for mode 1, and so on. Mode 0 only 4914 allowed by default. 4915 4916 pause_on_oops=<int> 4917 Halt all CPUs after the first oops has been printed for 4918 the specified number of seconds. This is to be used if 4919 your oopses keep scrolling off the screen. 4920 4921 pcbit= [HW,ISDN] 4922 4923 pci=option[,option...] [PCI,EARLY] various PCI subsystem options. 4924 4925 Some options herein operate on a specific device 4926 or a set of devices (<pci_dev>). These are 4927 specified in one of the following formats: 4928 4929 [<domain>:]<bus>:<dev>.<func>[/<dev>.<func>]* 4930 pci:<vendor>:<device>[:<subvendor>:<subdevice>] 4931 4932 Note: the first format specifies a PCI 4933 bus/device/function address which may change 4934 if new hardware is inserted, if motherboard 4935 firmware changes, or due to changes caused 4936 by other kernel parameters. If the 4937 domain is left unspecified, it is 4938 taken to be zero. Optionally, a path 4939 to a device through multiple device/function 4940 addresses can be specified after the base 4941 address (this is more robust against 4942 renumbering issues). The second format 4943 selects devices using IDs from the 4944 configuration space which may match multiple 4945 devices in the system. 4946 4947 earlydump dump PCI config space before the kernel 4948 changes anything 4949 off [X86] don't probe for the PCI bus 4950 bios [X86-32] force use of PCI BIOS, don't access 4951 the hardware directly. Use this if your machine 4952 has a non-standard PCI host bridge. 4953 nobios [X86-32] disallow use of PCI BIOS, only direct 4954 hardware access methods are allowed. Use this 4955 if you experience crashes upon bootup and you 4956 suspect they are caused by the BIOS. 4957 conf1 [X86] Force use of PCI Configuration Access 4958 Mechanism 1 (config address in IO port 0xCF8, 4959 data in IO port 0xCFC, both 32-bit). 4960 conf2 [X86] Force use of PCI Configuration Access 4961 Mechanism 2 (IO port 0xCF8 is an 8-bit port for 4962 the function, IO port 0xCFA, also 8-bit, sets 4963 bus number. The config space is then accessed 4964 through ports 0xC000-0xCFFF). 4965 See http://wiki.osdev.org/PCI for more info 4966 on the configuration access mechanisms. 4967 noaer [PCIE] If the PCIEAER kernel config parameter is 4968 enabled, this kernel boot option can be used to 4969 disable the use of PCIE advanced error reporting. 4970 nodomains [PCI] Disable support for multiple PCI 4971 root domains (aka PCI segments, in ACPI-speak). 4972 nommconf [X86] Disable use of MMCONFIG for PCI 4973 Configuration 4974 check_enable_amd_mmconf [X86] check for and enable 4975 properly configured MMIO access to PCI 4976 config space on AMD family 10h CPU 4977 nomsi [MSI] If the PCI_MSI kernel config parameter is 4978 enabled, this kernel boot option can be used to 4979 disable the use of MSI interrupts system-wide. 4980 noioapicquirk [APIC] Disable all boot interrupt quirks. 4981 Safety option to keep boot IRQs enabled. This 4982 should never be necessary. 4983 ioapicreroute [APIC] Enable rerouting of boot IRQs to the 4984 primary IO-APIC for bridges that cannot disable 4985 boot IRQs. This fixes a source of spurious IRQs 4986 when the system masks IRQs. 4987 noioapicreroute [APIC] Disable workaround that uses the 4988 boot IRQ equivalent of an IRQ that connects to 4989 a chipset where boot IRQs cannot be disabled. 4990 The opposite of ioapicreroute. 4991 biosirq [X86-32] Use PCI BIOS calls to get the interrupt 4992 routing table. These calls are known to be buggy 4993 on several machines and they hang the machine 4994 when used, but on other computers it's the only 4995 way to get the interrupt routing table. Try 4996 this option if the kernel is unable to allocate 4997 IRQs or discover secondary PCI buses on your 4998 motherboard. 4999 rom [X86] Assign address space to expansion ROMs. 5000 Use with caution as certain devices share 5001 address decoders between ROMs and other 5002 resources. 5003 norom [X86] Do not assign address space to 5004 expansion ROMs that do not already have 5005 BIOS assigned address ranges. 5006 nobar [X86] Do not assign address space to the 5007 BARs that weren't assigned by the BIOS. 5008 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be 5009 assigned automatically to PCI devices. You can 5010 make the kernel exclude IRQs of your ISA cards 5011 this way. 5012 pirqaddr=0xAAAAA [X86] Specify the physical address 5013 of the PIRQ table (normally generated 5014 by the BIOS) if it is outside the 5015 F0000h-100000h range. 5016 lastbus=N [X86] Scan all buses thru bus #N. Can be 5017 useful if the kernel is unable to find your 5018 secondary buses and you want to tell it 5019 explicitly which ones they are. 5020 assign-busses [X86] Always assign all PCI bus 5021 numbers ourselves, overriding 5022 whatever the firmware may have done. 5023 usepirqmask [X86] Honor the possible IRQ mask stored 5024 in the BIOS $PIR table. This is needed on 5025 some systems with broken BIOSes, notably 5026 some HP Pavilion N5400 and Omnibook XE3 5027 notebooks. This will have no effect if ACPI 5028 IRQ routing is enabled. 5029 noacpi [X86] Do not use ACPI for IRQ routing 5030 or for PCI scanning. 5031 use_crs [X86] Use PCI host bridge window information 5032 from ACPI. On BIOSes from 2008 or later, this 5033 is enabled by default. If you need to use this, 5034 please report a bug. 5035 nocrs [X86] Ignore PCI host bridge windows from ACPI. 5036 If you need to use this, please report a bug. 5037 use_e820 [X86] Use E820 reservations to exclude parts of 5038 PCI host bridge windows. This is a workaround 5039 for BIOS defects in host bridge _CRS methods. 5040 If you need to use this, please report a bug to 5041 <linux-pci@vger.kernel.org>. 5042 no_e820 [X86] Ignore E820 reservations for PCI host 5043 bridge windows. This is the default on modern 5044 hardware. If you need to use this, please report 5045 a bug to <linux-pci@vger.kernel.org>. 5046 routeirq Do IRQ routing for all PCI devices. 5047 This is normally done in pci_enable_device(), 5048 so this option is a temporary workaround 5049 for broken drivers that don't call it. 5050 skip_isa_align [X86] do not align io start addr, so can 5051 handle more pci cards 5052 noearly [X86] Don't do any early type 1 scanning. 5053 This might help on some broken boards which 5054 machine check when some devices' config space 5055 is read. But various workarounds are disabled 5056 and some IOMMU drivers will not work. 5057 bfsort Sort PCI devices into breadth-first order. 5058 This sorting is done to get a device 5059 order compatible with older (<= 2.4) kernels. 5060 nobfsort Don't sort PCI devices into breadth-first order. 5061 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size) 5062 tuning and use the BIOS-configured MPS defaults. 5063 pcie_bus_safe Set every device's MPS to the largest value 5064 supported by all devices below the root complex. 5065 pcie_bus_perf Set device MPS to the largest allowable MPS 5066 based on its parent bus. Also set MRRS (Max 5067 Read Request Size) to the largest supported 5068 value (no larger than the MPS that the device 5069 or bus can support) for best performance. 5070 pcie_bus_peer2peer Set every device's MPS to 128B, which 5071 every device is guaranteed to support. This 5072 configuration allows peer-to-peer DMA between 5073 any pair of devices, possibly at the cost of 5074 reduced performance. This also guarantees 5075 that hot-added devices will work. 5076 cbiosize=nn[KMG] The fixed amount of bus space which is 5077 reserved for the CardBus bridge's IO window. 5078 The default value is 256 bytes. 5079 cbmemsize=nn[KMG] The fixed amount of bus space which is 5080 reserved for the CardBus bridge's memory 5081 window. The default value is 64 megabytes. 5082 resource_alignment= 5083 Format: 5084 [<order of align>@]<pci_dev>[; ...] 5085 Specifies alignment and device to reassign 5086 aligned memory resources. How to 5087 specify the device is described above. 5088 If <order of align> is not specified, 5089 PAGE_SIZE is used as alignment. 5090 A PCI-PCI bridge can be specified if resource 5091 windows need to be expanded. 5092 To specify the alignment for several 5093 instances of a device, the PCI vendor, 5094 device, subvendor, and subdevice may be 5095 specified, e.g., 12@pci:8086:9c22:103c:198f 5096 for 4096-byte alignment. 5097 ecrc= Enable/disable PCIe ECRC (transaction layer 5098 end-to-end CRC checking). Only effective if 5099 OS has native AER control (either granted by 5100 ACPI _OSC or forced via "pcie_ports=native") 5101 bios: Use BIOS/firmware settings. This is the 5102 the default. 5103 off: Turn ECRC off 5104 on: Turn ECRC on. 5105 hpiosize=nn[KMG] The fixed amount of bus space which is 5106 reserved for hotplug bridge's IO window. 5107 Default size is 256 bytes. 5108 hpmmiosize=nn[KMG] The fixed amount of bus space which is 5109 reserved for hotplug bridge's MMIO window. 5110 Default size is 2 megabytes. 5111 hpmmioprefsize=nn[KMG] The fixed amount of bus space which is 5112 reserved for hotplug bridge's MMIO_PREF window. 5113 Default size is 2 megabytes. 5114 hpmemsize=nn[KMG] The fixed amount of bus space which is 5115 reserved for hotplug bridge's MMIO and 5116 MMIO_PREF window. 5117 Default size is 2 megabytes. 5118 hpbussize=nn The minimum amount of additional bus numbers 5119 reserved for buses below a hotplug bridge. 5120 Default is 1. 5121 realloc= Enable/disable reallocating PCI bridge resources 5122 if allocations done by BIOS are too small to 5123 accommodate resources required by all child 5124 devices. 5125 off: Turn realloc off 5126 on: Turn realloc on 5127 realloc same as realloc=on 5128 noari do not use PCIe ARI. 5129 noats [PCIE, Intel-IOMMU, AMD-IOMMU] 5130 do not use PCIe ATS (and IOMMU device IOTLB). 5131 pcie_scan_all Scan all possible PCIe devices. Otherwise we 5132 only look for one device below a PCIe downstream 5133 port. 5134 big_root_window Try to add a big 64bit memory window to the PCIe 5135 root complex on AMD CPUs. Some GFX hardware 5136 can resize a BAR to allow access to all VRAM. 5137 Adding the window is slightly risky (it may 5138 conflict with unreported devices), so this 5139 taints the kernel. 5140 disable_acs_redir=<pci_dev>[; ...] 5141 Specify one or more PCI devices (in the format 5142 specified above) separated by semicolons. 5143 Each device specified will have the PCI ACS 5144 redirect capabilities forced off which will 5145 allow P2P traffic between devices through 5146 bridges without forcing it upstream. Note: 5147 this removes isolation between devices and 5148 may put more devices in an IOMMU group. 5149 config_acs= 5150 Format: 5151 <ACS flags>@<pci_dev>[; ...] 5152 Specify one or more PCI devices (in the format 5153 specified above) optionally prepended with flags 5154 and separated by semicolons. The respective 5155 capabilities will be enabled, disabled or 5156 unchanged based on what is specified in 5157 flags. 5158 5159 ACS Flags is defined as follows: 5160 bit-0 : ACS Source Validation 5161 bit-1 : ACS Translation Blocking 5162 bit-2 : ACS P2P Request Redirect 5163 bit-3 : ACS P2P Completion Redirect 5164 bit-4 : ACS Upstream Forwarding 5165 bit-5 : ACS P2P Egress Control 5166 bit-6 : ACS Direct Translated P2P 5167 Each bit can be marked as: 5168 '0' – force disabled 5169 '1' – force enabled 5170 'x' – unchanged 5171 For example, 5172 pci=config_acs=10x@pci:0:0 5173 would configure all devices that support 5174 ACS to enable P2P Request Redirect, disable 5175 Translation Blocking, and leave Source 5176 Validation unchanged from whatever power-up 5177 or firmware set it to. 5178 5179 Note: this may remove isolation between devices 5180 and may put more devices in an IOMMU group. 5181 force_floating [S390] Force usage of floating interrupts. 5182 nomio [S390] Do not use MIO instructions. 5183 norid [S390] ignore the RID field and force use of 5184 one PCI domain per PCI function 5185 notph [PCIE] If the PCIE_TPH kernel config parameter 5186 is enabled, this kernel boot option can be used 5187 to disable PCIe TLP Processing Hints support 5188 system-wide. 5189 5190 pcie_aspm= [PCIE] Forcibly enable or ignore PCIe Active State Power 5191 Management. 5192 off Don't touch ASPM configuration at all. Leave any 5193 configuration done by firmware unchanged. 5194 force Enable ASPM even on devices that claim not to support it. 5195 WARNING: Forcing ASPM on may cause system lockups. 5196 5197 pcie_ports= [PCIE] PCIe port services handling: 5198 native Use native PCIe services (PME, AER, DPC, PCIe hotplug) 5199 even if the platform doesn't give the OS permission to 5200 use them. This may cause conflicts if the platform 5201 also tries to use these services. 5202 dpc-native Use native PCIe service for DPC only. May 5203 cause conflicts if firmware uses AER or DPC. 5204 compat Disable native PCIe services (PME, AER, DPC, PCIe 5205 hotplug). 5206 5207 pcie_port_pm= [PCIE] PCIe port power management handling: 5208 off Disable power management of all PCIe ports 5209 force Forcibly enable power management of all PCIe ports 5210 5211 pcie_pme= [PCIE,PM] Native PCIe PME signaling options: 5212 nomsi Do not use MSI for native PCIe PME signaling (this makes 5213 all PCIe root ports use INTx for all services). 5214 5215 pcmv= [HW,PCMCIA] BadgePAD 4 5216 5217 pd_ignore_unused 5218 [PM] 5219 Keep all power-domains already enabled by bootloader on, 5220 even if no driver has claimed them. This is useful 5221 for debug and development, but should not be 5222 needed on a platform with proper driver support. 5223 5224 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at 5225 boot time. 5226 Format: { 0 | 1 } 5227 See arch/parisc/kernel/pdc_chassis.c 5228 5229 percpu_alloc= [MM,EARLY] 5230 Select which percpu first chunk allocator to use. 5231 Currently supported values are "embed" and "page". 5232 Archs may support subset or none of the selections. 5233 See comments in mm/percpu.c for details on each 5234 allocator. This parameter is primarily for debugging 5235 and performance comparison. 5236 5237 pirq= [SMP,APIC] Manual mp-table setup 5238 See Documentation/arch/x86/i386/IO-APIC.rst. 5239 5240 plip= [PPT,NET] Parallel port network link 5241 Format: { parport<nr> | timid | 0 } 5242 See also Documentation/admin-guide/parport.rst. 5243 5244 pmtmr= [X86] Manual setup of pmtmr I/O Port. 5245 Override pmtimer IOPort with a hex value. 5246 e.g. pmtmr=0x508 5247 5248 pmu_override= [PPC] Override the PMU. 5249 This option takes over the PMU facility, so it is no 5250 longer usable by perf. Setting this option starts the 5251 PMU counters by setting MMCR0 to 0 (the FC bit is 5252 cleared). If a number is given, then MMCR1 is set to 5253 that number, otherwise (e.g., 'pmu_override=on'), MMCR1 5254 remains 0. 5255 5256 pm_async= [PM] 5257 Format: off 5258 This parameter sets the initial value of the 5259 /sys/power/pm_async sysfs knob at boot time. 5260 If set to "off", disables asynchronous suspend and 5261 resume of devices during system-wide power transitions. 5262 This can be useful on platforms where device 5263 dependencies are not well-defined, or for debugging 5264 power management issues. Asynchronous operations are 5265 enabled by default. 5266 5267 5268 pm_debug_messages [SUSPEND,KNL] 5269 Enable suspend/resume debug messages during boot up. 5270 5271 pnp.debug=1 [PNP] 5272 Enable PNP debug messages (depends on the 5273 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time 5274 via /sys/module/pnp/parameters/debug. We always show 5275 current resource usage; turning this on also shows 5276 possible settings and some assignment information. 5277 5278 pnpacpi= [ACPI] 5279 { off } 5280 5281 pnpbios= [ISAPNP] 5282 { on | off | curr | res | no-curr | no-res } 5283 5284 pnp_reserve_irq= 5285 [ISAPNP] Exclude IRQs for the autoconfiguration 5286 5287 pnp_reserve_dma= 5288 [ISAPNP] Exclude DMAs for the autoconfiguration 5289 5290 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration 5291 Ranges are in pairs (I/O port base and size). 5292 5293 pnp_reserve_mem= 5294 [ISAPNP] Exclude memory regions for the 5295 autoconfiguration. 5296 Ranges are in pairs (memory base and size). 5297 5298 ports= [IP_VS_FTP] IPVS ftp helper module 5299 Default is 21. 5300 Up to 8 (IP_VS_APP_MAX_PORTS) ports 5301 may be specified. 5302 Format: <port>,<port>.... 5303 5304 possible_cpus= [SMP,S390,X86] 5305 Format: <unsigned int> 5306 Set the number of possible CPUs, overriding the 5307 regular discovery mechanisms (such as ACPI/FW, etc). 5308 5309 powersave=off [PPC] This option disables power saving features. 5310 It specifically disables cpuidle and sets the 5311 platform machine description specific power_save 5312 function to NULL. On Idle the CPU just reduces 5313 execution priority. 5314 5315 ppc_strict_facility_enable 5316 [PPC,ENABLE] This option catches any kernel floating point, 5317 Altivec, VSX and SPE outside of regions specifically 5318 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()). 5319 There is some performance impact when enabling this. 5320 5321 ppc_tm= [PPC,EARLY] 5322 Format: {"off"} 5323 Disable Hardware Transactional Memory 5324 5325 preempt= [KNL] 5326 Select preemption mode if you have CONFIG_PREEMPT_DYNAMIC 5327 none - Limited to cond_resched() calls 5328 voluntary - Limited to cond_resched() and might_sleep() calls 5329 full - Any section that isn't explicitly preempt disabled 5330 can be preempted anytime. Tasks will also yield 5331 contended spinlocks (if the critical section isn't 5332 explicitly preempt disabled beyond the lock itself). 5333 lazy - Scheduler controlled. Similar to full but instead 5334 of preempting the task immediately, the task gets 5335 one HZ tick time to yield itself before the 5336 preemption will be forced. One preemption is when the 5337 task returns to user space. 5338 5339 print-fatal-signals= 5340 [KNL] debug: print fatal signals 5341 5342 If enabled, warn about various signal handling 5343 related application anomalies: too many signals, 5344 too many POSIX.1 timers, fatal signals causing a 5345 coredump - etc. 5346 5347 If you hit the warning due to signal overflow, 5348 you might want to try "ulimit -i unlimited". 5349 5350 default: off. 5351 5352 printk.always_kmsg_dump= 5353 Trigger kmsg_dump for cases other than kernel oops or 5354 panics 5355 Format: <bool> (1/Y/y=enable, 0/N/n=disable) 5356 default: disabled 5357 5358 printk.console_no_auto_verbose= 5359 Disable console loglevel raise on oops, panic 5360 or lockdep-detected issues (only if lock debug is on). 5361 With an exception to setups with low baudrate on 5362 serial console, keeping this 0 is a good choice 5363 in order to provide more debug information. 5364 Format: <bool> 5365 default: 0 (auto_verbose is enabled) 5366 5367 printk.debug_non_panic_cpus= 5368 Allows storing messages from non-panic CPUs into 5369 the printk log buffer during panic(). They are 5370 flushed to consoles by the panic-CPU on 5371 a best-effort basis. 5372 Format: <bool> (1/Y/y=enable, 0/N/n=disable) 5373 Default: disabled 5374 5375 printk.devkmsg={on,off,ratelimit} 5376 Control writing to /dev/kmsg. 5377 on - unlimited logging to /dev/kmsg from userspace 5378 off - logging to /dev/kmsg disabled 5379 ratelimit - ratelimit the logging 5380 Default: ratelimit 5381 5382 printk.time= Show timing data prefixed to each printk message line 5383 Format: <bool> (1/Y/y=enable, 0/N/n=disable) 5384 5385 proc_mem.force_override= [KNL] 5386 Format: {always | ptrace | never} 5387 Traditionally /proc/pid/mem allows memory permissions to be 5388 overridden without restrictions. This option may be set to 5389 restrict that. Can be one of: 5390 - 'always': traditional behavior always allows mem overrides. 5391 - 'ptrace': only allow mem overrides for active ptracers. 5392 - 'never': never allow mem overrides. 5393 If not specified, default is the CONFIG_PROC_MEM_* choice. 5394 5395 processor.max_cstate= [HW,ACPI] 5396 Limit processor to maximum C-state 5397 max_cstate=9 overrides any DMI blacklist limit. 5398 5399 processor.nocst [HW,ACPI] 5400 Ignore the _CST method to determine C-states, 5401 instead using the legacy FADT method 5402 5403 profile= [KNL] Enable kernel profiling via /proc/profile 5404 Format: [<profiletype>,]<number> 5405 Param: <profiletype>: "schedule" or "kvm" 5406 [defaults to kernel profiling] 5407 Param: "schedule" - profile schedule points. 5408 Param: "kvm" - profile VM exits. 5409 Param: <number> - step/bucket size as a power of 2 for 5410 statistical time based profiling. 5411 5412 prompt_ramdisk= [RAM] [Deprecated] 5413 5414 prot_virt= [S390] enable hosting protected virtual machines 5415 isolated from the hypervisor (if hardware supports 5416 that). If enabled, the default kernel base address 5417 might be overridden even when Kernel Address Space 5418 Layout Randomization is disabled. 5419 Format: <bool> 5420 5421 psi= [KNL] Enable or disable pressure stall information 5422 tracking. 5423 Format: <bool> 5424 5425 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to 5426 probe for; one of (bare|imps|exps|lifebook|any). 5427 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports 5428 per second. 5429 psmouse.resetafter= [HW,MOUSE] 5430 Try to reset the device after so many bad packets 5431 (0 = never). 5432 psmouse.resolution= 5433 [HW,MOUSE] Set desired mouse resolution, in dpi. 5434 psmouse.smartscroll= 5435 [HW,MOUSE] Controls Logitech smartscroll autorepeat. 5436 0 = disabled, 1 = enabled (default). 5437 5438 pstore.backend= Specify the name of the pstore backend to use 5439 5440 pti= [X86-64] Control Page Table Isolation of user and 5441 kernel address spaces. Disabling this feature 5442 removes hardening, but improves performance of 5443 system calls and interrupts. 5444 5445 on - unconditionally enable 5446 off - unconditionally disable 5447 auto - kernel detects whether your CPU model is 5448 vulnerable to issues that PTI mitigates 5449 5450 Not specifying this option is equivalent to pti=auto. 5451 5452 pty.legacy_count= 5453 [KNL] Number of legacy pty's. Overwrites compiled-in 5454 default number. 5455 5456 quiet [KNL,EARLY] Disable most log messages 5457 5458 r128= [HW,DRM] 5459 5460 radix_hcall_invalidate=on [PPC/PSERIES] 5461 Disable RADIX GTSE feature and use hcall for TLB 5462 invalidate. 5463 5464 raid= [HW,RAID] 5465 See Documentation/admin-guide/md.rst. 5466 5467 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes 5468 See Documentation/admin-guide/blockdev/ramdisk.rst. 5469 5470 ramdisk_start= [RAM] RAM disk image start address 5471 5472 random.trust_cpu=off 5473 [KNL,EARLY] Disable trusting the use of the CPU's 5474 random number generator (if available) to 5475 initialize the kernel's RNG. 5476 5477 random.trust_bootloader=off 5478 [KNL,EARLY] Disable trusting the use of the a seed 5479 passed by the bootloader (if available) to 5480 initialize the kernel's RNG. 5481 5482 randomize_kstack_offset= 5483 [KNL,EARLY] Enable or disable kernel stack offset 5484 randomization, which provides roughly 5 bits of 5485 entropy, frustrating memory corruption attacks 5486 that depend on stack address determinism or 5487 cross-syscall address exposures. This is only 5488 available on architectures that have defined 5489 CONFIG_HAVE_ARCH_RANDOMIZE_KSTACK_OFFSET. 5490 Format: <bool> (1/Y/y=enable, 0/N/n=disable) 5491 Default is CONFIG_RANDOMIZE_KSTACK_OFFSET_DEFAULT. 5492 5493 ras=option[,option,...] [KNL] RAS-specific options 5494 5495 cec_disable [X86] 5496 Disable the Correctable Errors Collector, 5497 see CONFIG_RAS_CEC help text. 5498 5499 rcu_nocbs[=cpu-list] 5500 [KNL] The optional argument is a cpu list, 5501 as described above. 5502 5503 In kernels built with CONFIG_RCU_NOCB_CPU=y, 5504 enable the no-callback CPU mode, which prevents 5505 such CPUs' callbacks from being invoked in 5506 softirq context. Invocation of such CPUs' RCU 5507 callbacks will instead be offloaded to "rcuox/N" 5508 kthreads created for that purpose, where "x" is 5509 "p" for RCU-preempt, "s" for RCU-sched, and "g" 5510 for the kthreads that mediate grace periods; and 5511 "N" is the CPU number. This reduces OS jitter on 5512 the offloaded CPUs, which can be useful for HPC 5513 and real-time workloads. It can also improve 5514 energy efficiency for asymmetric multiprocessors. 5515 5516 If a cpulist is passed as an argument, the specified 5517 list of CPUs is set to no-callback mode from boot. 5518 5519 Otherwise, if the '=' sign and the cpulist 5520 arguments are omitted, no CPU will be set to 5521 no-callback mode from boot but the mode may be 5522 toggled at runtime via cpusets. 5523 5524 Note that this argument takes precedence over 5525 the CONFIG_RCU_NOCB_CPU_DEFAULT_ALL option. 5526 5527 rcu_nocb_poll [KNL] 5528 Rather than requiring that offloaded CPUs 5529 (specified by rcu_nocbs= above) explicitly 5530 awaken the corresponding "rcuoN" kthreads, 5531 make these kthreads poll for callbacks. 5532 This improves the real-time response for the 5533 offloaded CPUs by relieving them of the need to 5534 wake up the corresponding kthread, but degrades 5535 energy efficiency by requiring that the kthreads 5536 periodically wake up to do the polling. 5537 5538 rcutree.blimit= [KNL] 5539 Set maximum number of finished RCU callbacks to 5540 process in one batch. 5541 5542 rcutree.csd_lock_suppress_rcu_stall= [KNL] 5543 Do only a one-line RCU CPU stall warning when 5544 there is an ongoing too-long CSD-lock wait. 5545 5546 rcutree.do_rcu_barrier= [KNL] 5547 Request a call to rcu_barrier(). This is 5548 throttled so that userspace tests can safely 5549 hammer on the sysfs variable if they so choose. 5550 If triggered before the RCU grace-period machinery 5551 is fully active, this will error out with EAGAIN. 5552 5553 rcutree.dump_tree= [KNL] 5554 Dump the structure of the rcu_node combining tree 5555 out at early boot. This is used for diagnostic 5556 purposes, to verify correct tree setup. 5557 5558 rcutree.gp_cleanup_delay= [KNL] 5559 Set the number of jiffies to delay each step of 5560 RCU grace-period cleanup. 5561 5562 rcutree.gp_init_delay= [KNL] 5563 Set the number of jiffies to delay each step of 5564 RCU grace-period initialization. 5565 5566 rcutree.gp_preinit_delay= [KNL] 5567 Set the number of jiffies to delay each step of 5568 RCU grace-period pre-initialization, that is, 5569 the propagation of recent CPU-hotplug changes up 5570 the rcu_node combining tree. 5571 5572 rcutree.jiffies_till_first_fqs= [KNL] 5573 Set delay from grace-period initialization to 5574 first attempt to force quiescent states. 5575 Units are jiffies, minimum value is zero, 5576 and maximum value is HZ. 5577 5578 rcutree.jiffies_till_next_fqs= [KNL] 5579 Set delay between subsequent attempts to force 5580 quiescent states. Units are jiffies, minimum 5581 value is one, and maximum value is HZ. 5582 5583 rcutree.jiffies_till_sched_qs= [KNL] 5584 Set required age in jiffies for a 5585 given grace period before RCU starts 5586 soliciting quiescent-state help from 5587 rcu_note_context_switch() and cond_resched(). 5588 If not specified, the kernel will calculate 5589 a value based on the most recent settings 5590 of rcutree.jiffies_till_first_fqs 5591 and rcutree.jiffies_till_next_fqs. 5592 This calculated value may be viewed in 5593 rcutree.jiffies_to_sched_qs. Any attempt to set 5594 rcutree.jiffies_to_sched_qs will be cheerfully 5595 overwritten. 5596 5597 rcutree.kthread_prio= [KNL,BOOT] 5598 Set the SCHED_FIFO priority of the RCU per-CPU 5599 kthreads (rcuc/N). This value is also used for 5600 the priority of the RCU boost threads (rcub/N) 5601 and for the RCU grace-period kthreads (rcu_bh, 5602 rcu_preempt, and rcu_sched). If RCU_BOOST is 5603 set, valid values are 1-99 and the default is 1 5604 (the least-favored priority). Otherwise, when 5605 RCU_BOOST is not set, valid values are 0-99 and 5606 the default is zero (non-realtime operation). 5607 When RCU_NOCB_CPU is set, also adjust the 5608 priority of NOCB callback kthreads. 5609 5610 rcutree.nocb_nobypass_lim_per_jiffy= [KNL] 5611 On callback-offloaded (rcu_nocbs) CPUs, 5612 RCU reduces the lock contention that would 5613 otherwise be caused by callback floods through 5614 use of the ->nocb_bypass list. However, in the 5615 common non-flooded case, RCU queues directly to 5616 the main ->cblist in order to avoid the extra 5617 overhead of the ->nocb_bypass list and its lock. 5618 But if there are too many callbacks queued during 5619 a single jiffy, RCU pre-queues the callbacks into 5620 the ->nocb_bypass queue. The definition of "too 5621 many" is supplied by this kernel boot parameter. 5622 5623 rcutree.nohz_full_patience_delay= [KNL] 5624 On callback-offloaded (rcu_nocbs) CPUs, avoid 5625 disturbing RCU unless the grace period has 5626 reached the specified age in milliseconds. 5627 Defaults to zero. Large values will be capped 5628 at five seconds. All values will be rounded down 5629 to the nearest value representable by jiffies. 5630 5631 rcutree.qhimark= [KNL] 5632 Set threshold of queued RCU callbacks beyond which 5633 batch limiting is disabled. 5634 5635 rcutree.qlowmark= [KNL] 5636 Set threshold of queued RCU callbacks below which 5637 batch limiting is re-enabled. 5638 5639 rcutree.qovld= [KNL] 5640 Set threshold of queued RCU callbacks beyond which 5641 RCU's force-quiescent-state scan will aggressively 5642 enlist help from cond_resched() and sched IPIs to 5643 help CPUs more quickly reach quiescent states. 5644 Set to less than zero to make this be set based 5645 on rcutree.qhimark at boot time and to zero to 5646 disable more aggressive help enlistment. 5647 5648 rcutree.rcu_delay_page_cache_fill_msec= [KNL] 5649 Set the page-cache refill delay (in milliseconds) 5650 in response to low-memory conditions. The range 5651 of permitted values is in the range 0:100000. 5652 5653 rcutree.rcu_divisor= [KNL] 5654 Set the shift-right count to use to compute 5655 the callback-invocation batch limit bl from 5656 the number of callbacks queued on this CPU. 5657 The result will be bounded below by the value of 5658 the rcutree.blimit kernel parameter. Every bl 5659 callbacks, the softirq handler will exit in 5660 order to allow the CPU to do other work. 5661 5662 Please note that this callback-invocation batch 5663 limit applies only to non-offloaded callback 5664 invocation. Offloaded callbacks are instead 5665 invoked in the context of an rcuoc kthread, which 5666 scheduler will preempt as it does any other task. 5667 5668 rcutree.rcu_fanout_exact= [KNL] 5669 Disable autobalancing of the rcu_node combining 5670 tree. This is used by rcutorture, and might 5671 possibly be useful for architectures having high 5672 cache-to-cache transfer latencies. 5673 5674 rcutree.rcu_fanout_leaf= [KNL] 5675 Change the number of CPUs assigned to each 5676 leaf rcu_node structure. Useful for very 5677 large systems, which will choose the value 64, 5678 and for NUMA systems with large remote-access 5679 latencies, which will choose a value aligned 5680 with the appropriate hardware boundaries. 5681 5682 rcutree.rcu_min_cached_objs= [KNL] 5683 Minimum number of objects which are cached and 5684 maintained per one CPU. Object size is equal 5685 to PAGE_SIZE. The cache allows to reduce the 5686 pressure to page allocator, also it makes the 5687 whole algorithm to behave better in low memory 5688 condition. 5689 5690 rcutree.rcu_nocb_gp_stride= [KNL] 5691 Set the number of NOCB callback kthreads in 5692 each group, which defaults to the square root 5693 of the number of CPUs. Larger numbers reduce 5694 the wakeup overhead on the global grace-period 5695 kthread, but increases that same overhead on 5696 each group's NOCB grace-period kthread. 5697 5698 rcutree.rcu_kick_kthreads= [KNL] 5699 Cause the grace-period kthread to get an extra 5700 wake_up() if it sleeps three times longer than 5701 it should at force-quiescent-state time. 5702 This wake_up() will be accompanied by a 5703 WARN_ONCE() splat and an ftrace_dump(). 5704 5705 rcutree.rcu_resched_ns= [KNL] 5706 Limit the time spend invoking a batch of RCU 5707 callbacks to the specified number of nanoseconds. 5708 By default, this limit is checked only once 5709 every 32 callbacks in order to limit the pain 5710 inflicted by local_clock() overhead. 5711 5712 rcutree.rcu_unlock_delay= [KNL] 5713 In CONFIG_RCU_STRICT_GRACE_PERIOD=y kernels, 5714 this specifies an rcu_read_unlock()-time delay 5715 in microseconds. This defaults to zero. 5716 Larger delays increase the probability of 5717 catching RCU pointer leaks, that is, buggy use 5718 of RCU-protected pointers after the relevant 5719 rcu_read_unlock() has completed. 5720 5721 rcutree.sysrq_rcu= [KNL] 5722 Commandeer a sysrq key to dump out Tree RCU's 5723 rcu_node tree with an eye towards determining 5724 why a new grace period has not yet started. 5725 5726 rcutree.use_softirq= [KNL] 5727 If set to zero, move all RCU_SOFTIRQ processing to 5728 per-CPU rcuc kthreads. Defaults to a non-zero 5729 value, meaning that RCU_SOFTIRQ is used by default. 5730 Specify rcutree.use_softirq=0 to use rcuc kthreads. 5731 5732 But note that CONFIG_PREEMPT_RT=y kernels disable 5733 this kernel boot parameter, forcibly setting it 5734 to zero. 5735 5736 rcutree.enable_rcu_lazy= [KNL] 5737 To save power, batch RCU callbacks and flush after 5738 delay, memory pressure or callback list growing too 5739 big. 5740 5741 rcutree.rcu_normal_wake_from_gp= [KNL] 5742 Reduces a latency of synchronize_rcu() call. This approach 5743 maintains its own track of synchronize_rcu() callers, so it 5744 does not interact with regular callbacks because it does not 5745 use a call_rcu[_hurry]() path. Please note, this is for a 5746 normal grace period. 5747 5748 How to enable it: 5749 5750 echo 1 > /sys/module/rcutree/parameters/rcu_normal_wake_from_gp 5751 or pass a boot parameter "rcutree.rcu_normal_wake_from_gp=1" 5752 5753 Default is 1 if num_possible_cpus() <= 16 and it is not explicitly 5754 disabled by the boot parameter passing 0. 5755 5756 rcuscale.gp_async= [KNL] 5757 Measure performance of asynchronous 5758 grace-period primitives such as call_rcu(). 5759 5760 rcuscale.gp_async_max= [KNL] 5761 Specify the maximum number of outstanding 5762 callbacks per writer thread. When a writer 5763 thread exceeds this limit, it invokes the 5764 corresponding flavor of rcu_barrier() to allow 5765 previously posted callbacks to drain. 5766 5767 rcuscale.gp_exp= [KNL] 5768 Measure performance of expedited synchronous 5769 grace-period primitives. 5770 5771 rcuscale.holdoff= [KNL] 5772 Set test-start holdoff period. The purpose of 5773 this parameter is to delay the start of the 5774 test until boot completes in order to avoid 5775 interference. 5776 5777 rcuscale.kfree_by_call_rcu= [KNL] 5778 In kernels built with CONFIG_RCU_LAZY=y, test 5779 call_rcu() instead of kfree_rcu(). 5780 5781 rcuscale.kfree_mult= [KNL] 5782 Instead of allocating an object of size kfree_obj, 5783 allocate one of kfree_mult * sizeof(kfree_obj). 5784 Defaults to 1. 5785 5786 rcuscale.kfree_rcu_test= [KNL] 5787 Set to measure performance of kfree_rcu() flooding. 5788 5789 rcuscale.kfree_rcu_test_double= [KNL] 5790 Test the double-argument variant of kfree_rcu(). 5791 If this parameter has the same value as 5792 rcuscale.kfree_rcu_test_single, both the single- 5793 and double-argument variants are tested. 5794 5795 rcuscale.kfree_rcu_test_single= [KNL] 5796 Test the single-argument variant of kfree_rcu(). 5797 If this parameter has the same value as 5798 rcuscale.kfree_rcu_test_double, both the single- 5799 and double-argument variants are tested. 5800 5801 rcuscale.kfree_nthreads= [KNL] 5802 The number of threads running loops of kfree_rcu(). 5803 5804 rcuscale.kfree_alloc_num= [KNL] 5805 Number of allocations and frees done in an iteration. 5806 5807 rcuscale.kfree_loops= [KNL] 5808 Number of loops doing rcuscale.kfree_alloc_num number 5809 of allocations and frees. 5810 5811 rcuscale.minruntime= [KNL] 5812 Set the minimum test run time in seconds. This 5813 does not affect the data-collection interval, 5814 but instead allows better measurement of things 5815 like CPU consumption. 5816 5817 rcuscale.nreaders= [KNL] 5818 Set number of RCU readers. The value -1 selects 5819 N, where N is the number of CPUs. A value 5820 "n" less than -1 selects N-n+1, where N is again 5821 the number of CPUs. For example, -2 selects N 5822 (the number of CPUs), -3 selects N+1, and so on. 5823 A value of "n" less than or equal to -N selects 5824 a single reader. 5825 5826 rcuscale.nwriters= [KNL] 5827 Set number of RCU writers. The values operate 5828 the same as for rcuscale.nreaders. 5829 N, where N is the number of CPUs 5830 5831 rcuscale.scale_type= [KNL] 5832 Specify the RCU implementation to test. 5833 5834 rcuscale.shutdown= [KNL] 5835 Shut the system down after performance tests 5836 complete. This is useful for hands-off automated 5837 testing. 5838 5839 rcuscale.verbose= [KNL] 5840 Enable additional printk() statements. 5841 5842 rcuscale.writer_holdoff= [KNL] 5843 Write-side holdoff between grace periods, 5844 in microseconds. The default of zero says 5845 no holdoff. 5846 5847 rcuscale.writer_holdoff_jiffies= [KNL] 5848 Additional write-side holdoff between grace 5849 periods, but in jiffies. The default of zero 5850 says no holdoff. 5851 5852 rcutorture.fqs_duration= [KNL] 5853 Set duration of force_quiescent_state bursts 5854 in microseconds. 5855 5856 rcutorture.fqs_holdoff= [KNL] 5857 Set holdoff time within force_quiescent_state bursts 5858 in microseconds. 5859 5860 rcutorture.fqs_stutter= [KNL] 5861 Set wait time between force_quiescent_state bursts 5862 in seconds. 5863 5864 rcutorture.fwd_progress= [KNL] 5865 Specifies the number of kthreads to be used 5866 for RCU grace-period forward-progress testing 5867 for the types of RCU supporting this notion. 5868 Defaults to 1 kthread, values less than zero or 5869 greater than the number of CPUs cause the number 5870 of CPUs to be used. 5871 5872 rcutorture.fwd_progress_div= [KNL] 5873 Specify the fraction of a CPU-stall-warning 5874 period to do tight-loop forward-progress testing. 5875 5876 rcutorture.fwd_progress_holdoff= [KNL] 5877 Number of seconds to wait between successive 5878 forward-progress tests. 5879 5880 rcutorture.fwd_progress_need_resched= [KNL] 5881 Enclose cond_resched() calls within checks for 5882 need_resched() during tight-loop forward-progress 5883 testing. 5884 5885 rcutorture.gp_cond= [KNL] 5886 Use conditional/asynchronous update-side 5887 normal-grace-period primitives, if available. 5888 5889 rcutorture.gp_cond_exp= [KNL] 5890 Use conditional/asynchronous update-side 5891 expedited-grace-period primitives, if available. 5892 5893 rcutorture.gp_cond_full= [KNL] 5894 Use conditional/asynchronous update-side 5895 normal-grace-period primitives that also take 5896 concurrent expedited grace periods into account, 5897 if available. 5898 5899 rcutorture.gp_cond_exp_full= [KNL] 5900 Use conditional/asynchronous update-side 5901 expedited-grace-period primitives that also take 5902 concurrent normal grace periods into account, 5903 if available. 5904 5905 rcutorture.gp_cond_wi= [KNL] 5906 Nominal wait interval for normal conditional 5907 grace periods (specified by rcutorture's 5908 gp_cond and gp_cond_full module parameters), 5909 in microseconds. The actual wait interval will 5910 be randomly selected to nanosecond granularity up 5911 to this wait interval. Defaults to 16 jiffies, 5912 for example, 16,000 microseconds on a system 5913 with HZ=1000. 5914 5915 rcutorture.gp_cond_wi_exp= [KNL] 5916 Nominal wait interval for expedited conditional 5917 grace periods (specified by rcutorture's 5918 gp_cond_exp and gp_cond_exp_full module 5919 parameters), in microseconds. The actual wait 5920 interval will be randomly selected to nanosecond 5921 granularity up to this wait interval. Defaults to 5922 128 microseconds. 5923 5924 rcutorture.gp_exp= [KNL] 5925 Use expedited update-side primitives, if available. 5926 5927 rcutorture.gp_normal= [KNL] 5928 Use normal (non-expedited) asynchronous 5929 update-side primitives, if available. 5930 5931 rcutorture.gp_poll= [KNL] 5932 Use polled update-side normal-grace-period 5933 primitives, if available. 5934 5935 rcutorture.gp_poll_exp= [KNL] 5936 Use polled update-side expedited-grace-period 5937 primitives, if available. 5938 5939 rcutorture.gp_poll_full= [KNL] 5940 Use polled update-side normal-grace-period 5941 primitives that also take concurrent expedited 5942 grace periods into account, if available. 5943 5944 rcutorture.gp_poll_exp_full= [KNL] 5945 Use polled update-side expedited-grace-period 5946 primitives that also take concurrent normal 5947 grace periods into account, if available. 5948 5949 rcutorture.gp_poll_wi= [KNL] 5950 Nominal wait interval for normal conditional 5951 grace periods (specified by rcutorture's 5952 gp_poll and gp_poll_full module parameters), 5953 in microseconds. The actual wait interval will 5954 be randomly selected to nanosecond granularity up 5955 to this wait interval. Defaults to 16 jiffies, 5956 for example, 16,000 microseconds on a system 5957 with HZ=1000. 5958 5959 rcutorture.gp_poll_wi_exp= [KNL] 5960 Nominal wait interval for expedited conditional 5961 grace periods (specified by rcutorture's 5962 gp_poll_exp and gp_poll_exp_full module 5963 parameters), in microseconds. The actual wait 5964 interval will be randomly selected to nanosecond 5965 granularity up to this wait interval. Defaults to 5966 128 microseconds. 5967 5968 rcutorture.gp_sync= [KNL] 5969 Use normal (non-expedited) synchronous 5970 update-side primitives, if available. If all 5971 of rcutorture.gp_cond=, rcutorture.gp_exp=, 5972 rcutorture.gp_normal=, and rcutorture.gp_sync= 5973 are zero, rcutorture acts as if is interpreted 5974 they are all non-zero. 5975 5976 rcutorture.gpwrap_lag= [KNL] 5977 Enable grace-period wrap lag testing. Setting 5978 to false prevents the gpwrap lag test from 5979 running. Default is true. 5980 5981 rcutorture.gpwrap_lag_gps= [KNL] 5982 Set the value for grace-period wrap lag during 5983 active lag testing periods. This controls how many 5984 grace periods differences we tolerate between 5985 rdp and rnp's gp_seq before setting overflow flag. 5986 The default is always set to 8. 5987 5988 rcutorture.gpwrap_lag_cycle_mins= [KNL] 5989 Set the total cycle duration for gpwrap lag 5990 testing in minutes. This is the total time for 5991 one complete cycle of active and inactive 5992 testing periods. Default is 30 minutes. 5993 5994 rcutorture.gpwrap_lag_active_mins= [KNL] 5995 Set the duration for which gpwrap lag is active 5996 within each cycle, in minutes. During this time, 5997 the grace-period wrap lag will be set to the 5998 value specified by gpwrap_lag_gps. Default is 5999 5 minutes. 6000 6001 rcutorture.irqreader= [KNL] 6002 Run RCU readers from irq handlers, or, more 6003 accurately, from a timer handler. Not all RCU 6004 flavors take kindly to this sort of thing. 6005 6006 rcutorture.leakpointer= [KNL] 6007 Leak an RCU-protected pointer out of the reader. 6008 This can of course result in splats, and is 6009 intended to test the ability of things like 6010 CONFIG_RCU_STRICT_GRACE_PERIOD=y to detect 6011 such leaks. 6012 6013 rcutorture.n_barrier_cbs= [KNL] 6014 Set callbacks/threads for rcu_barrier() testing. 6015 6016 rcutorture.nfakewriters= [KNL] 6017 Set number of concurrent RCU writers. These just 6018 stress RCU, they don't participate in the actual 6019 test, hence the "fake". 6020 6021 rcutorture.nocbs_nthreads= [KNL] 6022 Set number of RCU callback-offload togglers. 6023 Zero (the default) disables toggling. 6024 6025 rcutorture.nocbs_toggle= [KNL] 6026 Set the delay in milliseconds between successive 6027 callback-offload toggling attempts. 6028 6029 rcutorture.nreaders= [KNL] 6030 Set number of RCU readers. The value -1 selects 6031 N-1, where N is the number of CPUs. A value 6032 "n" less than -1 selects N-n-2, where N is again 6033 the number of CPUs. For example, -2 selects N 6034 (the number of CPUs), -3 selects N+1, and so on. 6035 6036 rcutorture.object_debug= [KNL] 6037 Enable debug-object double-call_rcu() testing. 6038 6039 rcutorture.onoff_holdoff= [KNL] 6040 Set time (s) after boot for CPU-hotplug testing. 6041 6042 rcutorture.onoff_interval= [KNL] 6043 Set time (jiffies) between CPU-hotplug operations, 6044 or zero to disable CPU-hotplug testing. 6045 6046 rcutorture.preempt_duration= [KNL] 6047 Set duration (in milliseconds) of preemptions 6048 by a high-priority FIFO real-time task. Set to 6049 zero (the default) to disable. The CPUs to 6050 preempt are selected randomly from the set that 6051 are online at a given point in time. Races with 6052 CPUs going offline are ignored, with that attempt 6053 at preemption skipped. 6054 6055 rcutorture.preempt_interval= [KNL] 6056 Set interval (in milliseconds, defaulting to one 6057 second) between preemptions by a high-priority 6058 FIFO real-time task. This delay is mediated 6059 by an hrtimer and is further fuzzed to avoid 6060 inadvertent synchronizations. 6061 6062 rcutorture.read_exit_burst= [KNL] 6063 The number of times in a given read-then-exit 6064 episode that a set of read-then-exit kthreads 6065 is spawned. 6066 6067 rcutorture.read_exit_delay= [KNL] 6068 The delay, in seconds, between successive 6069 read-then-exit testing episodes. 6070 6071 rcutorture.reader_flavor= [KNL] 6072 A bit mask indicating which readers to use. 6073 If there is more than one bit set, the readers 6074 are entered from low-order bit up, and are 6075 exited in the opposite order. For SRCU, the 6076 0x1 bit is normal readers, 0x2 NMI-safe readers, 6077 and 0x4 light-weight readers. 6078 6079 rcutorture.shuffle_interval= [KNL] 6080 Set task-shuffle interval (s). Shuffling tasks 6081 allows some CPUs to go into dyntick-idle mode 6082 during the rcutorture test. 6083 6084 rcutorture.shutdown_secs= [KNL] 6085 Set time (s) after boot system shutdown. This 6086 is useful for hands-off automated testing. 6087 6088 rcutorture.stall_cpu= [KNL] 6089 Duration of CPU stall (s) to test RCU CPU stall 6090 warnings, zero to disable. 6091 6092 rcutorture.stall_cpu_block= [KNL] 6093 Sleep while stalling if set. This will result 6094 in warnings from preemptible RCU in addition to 6095 any other stall-related activity. Note that 6096 in kernels built with CONFIG_PREEMPTION=n and 6097 CONFIG_PREEMPT_COUNT=y, this parameter will 6098 cause the CPU to pass through a quiescent state. 6099 Given CONFIG_PREEMPTION=n, this will suppress 6100 RCU CPU stall warnings, but will instead result 6101 in scheduling-while-atomic splats. 6102 6103 Use of this module parameter results in splats. 6104 6105 6106 rcutorture.stall_cpu_holdoff= [KNL] 6107 Time to wait (s) after boot before inducing stall. 6108 6109 rcutorture.stall_cpu_irqsoff= [KNL] 6110 Disable interrupts while stalling if set, but only 6111 on the first stall in the set. 6112 6113 rcutorture.stall_cpu_repeat= [KNL] 6114 Number of times to repeat the stall sequence, 6115 so that rcutorture.stall_cpu_repeat=3 will result 6116 in four stall sequences. 6117 6118 rcutorture.stall_gp_kthread= [KNL] 6119 Duration (s) of forced sleep within RCU 6120 grace-period kthread to test RCU CPU stall 6121 warnings, zero to disable. If both stall_cpu 6122 and stall_gp_kthread are specified, the 6123 kthread is starved first, then the CPU. 6124 6125 rcutorture.stat_interval= [KNL] 6126 Time (s) between statistics printk()s. 6127 6128 rcutorture.stutter= [KNL] 6129 Time (s) to stutter testing, for example, specifying 6130 five seconds causes the test to run for five seconds, 6131 wait for five seconds, and so on. This tests RCU's 6132 ability to transition abruptly to and from idle. 6133 6134 rcutorture.test_boost= [KNL] 6135 Test RCU priority boosting? 0=no, 1=maybe, 2=yes. 6136 "Maybe" means test if the RCU implementation 6137 under test support RCU priority boosting. 6138 6139 rcutorture.test_boost_duration= [KNL] 6140 Duration (s) of each individual boost test. 6141 6142 rcutorture.test_boost_holdoff= [KNL] 6143 Holdoff time (s) from start of test to the start 6144 of RCU priority-boost testing. Defaults to zero, 6145 that is, no holdoff. 6146 6147 rcutorture.test_boost_interval= [KNL] 6148 Interval (s) between each boost test. 6149 6150 rcutorture.test_no_idle_hz= [KNL] 6151 Test RCU's dyntick-idle handling. See also the 6152 rcutorture.shuffle_interval parameter. 6153 6154 rcutorture.torture_type= [KNL] 6155 Specify the RCU implementation to test. 6156 6157 rcutorture.verbose= [KNL] 6158 Enable additional printk() statements. 6159 6160 rcupdate.rcu_cpu_stall_ftrace_dump= [KNL] 6161 Dump ftrace buffer after reporting RCU CPU 6162 stall warning. 6163 6164 rcupdate.rcu_cpu_stall_notifiers= [KNL] 6165 Provide RCU CPU stall notifiers, but see the 6166 warnings in the RCU_CPU_STALL_NOTIFIER Kconfig 6167 option's help text. TL;DR: You almost certainly 6168 do not want rcupdate.rcu_cpu_stall_notifiers. 6169 6170 rcupdate.rcu_cpu_stall_suppress= [KNL] 6171 Suppress RCU CPU stall warning messages. 6172 6173 rcupdate.rcu_cpu_stall_suppress_at_boot= [KNL] 6174 Suppress RCU CPU stall warning messages and 6175 rcutorture writer stall warnings that occur 6176 during early boot, that is, during the time 6177 before the init task is spawned. 6178 6179 rcupdate.rcu_cpu_stall_timeout= [KNL] 6180 Set timeout for RCU CPU stall warning messages. 6181 The value is in seconds and the maximum allowed 6182 value is 300 seconds. 6183 6184 rcupdate.rcu_exp_cpu_stall_timeout= [KNL] 6185 Set timeout for expedited RCU CPU stall warning 6186 messages. The value is in milliseconds 6187 and the maximum allowed value is 21000 6188 milliseconds. Please note that this value is 6189 adjusted to an arch timer tick resolution. 6190 Setting this to zero causes the value from 6191 rcupdate.rcu_cpu_stall_timeout to be used (after 6192 conversion from seconds to milliseconds). 6193 6194 rcupdate.rcu_cpu_stall_cputime= [KNL] 6195 Provide statistics on the cputime and count of 6196 interrupts and tasks during the sampling period. For 6197 multiple continuous RCU stalls, all sampling periods 6198 begin at half of the first RCU stall timeout. 6199 6200 rcupdate.rcu_exp_stall_task_details= [KNL] 6201 Print stack dumps of any tasks blocking the 6202 current expedited RCU grace period during an 6203 expedited RCU CPU stall warning. 6204 6205 rcupdate.rcu_expedited= [KNL] 6206 Use expedited grace-period primitives, for 6207 example, synchronize_rcu_expedited() instead 6208 of synchronize_rcu(). This reduces latency, 6209 but can increase CPU utilization, degrade 6210 real-time latency, and degrade energy efficiency. 6211 No effect on CONFIG_TINY_RCU kernels. 6212 6213 rcupdate.rcu_normal= [KNL] 6214 Use only normal grace-period primitives, 6215 for example, synchronize_rcu() instead of 6216 synchronize_rcu_expedited(). This improves 6217 real-time latency, CPU utilization, and 6218 energy efficiency, but can expose users to 6219 increased grace-period latency. This parameter 6220 overrides rcupdate.rcu_expedited. No effect on 6221 CONFIG_TINY_RCU kernels. 6222 6223 rcupdate.rcu_normal_after_boot= [KNL] 6224 Once boot has completed (that is, after 6225 rcu_end_inkernel_boot() has been invoked), use 6226 only normal grace-period primitives. No effect 6227 on CONFIG_TINY_RCU kernels. 6228 6229 But note that CONFIG_PREEMPT_RT=y kernels enables 6230 this kernel boot parameter, forcibly setting 6231 it to the value one, that is, converting any 6232 post-boot attempt at an expedited RCU grace 6233 period to instead use normal non-expedited 6234 grace-period processing. 6235 6236 rcupdate.rcu_task_collapse_lim= [KNL] 6237 Set the maximum number of callbacks present 6238 at the beginning of a grace period that allows 6239 the RCU Tasks flavors to collapse back to using 6240 a single callback queue. This switching only 6241 occurs when rcupdate.rcu_task_enqueue_lim is 6242 set to the default value of -1. 6243 6244 rcupdate.rcu_task_contend_lim= [KNL] 6245 Set the minimum number of callback-queuing-time 6246 lock-contention events per jiffy required to 6247 cause the RCU Tasks flavors to switch to per-CPU 6248 callback queuing. This switching only occurs 6249 when rcupdate.rcu_task_enqueue_lim is set to 6250 the default value of -1. 6251 6252 rcupdate.rcu_task_enqueue_lim= [KNL] 6253 Set the number of callback queues to use for the 6254 RCU Tasks family of RCU flavors. The default 6255 of -1 allows this to be automatically (and 6256 dynamically) adjusted. This parameter is intended 6257 for use in testing. 6258 6259 rcupdate.rcu_task_ipi_delay= [KNL] 6260 Set time in jiffies during which RCU tasks will 6261 avoid sending IPIs, starting with the beginning 6262 of a given grace period. Setting a large 6263 number avoids disturbing real-time workloads, 6264 but lengthens grace periods. 6265 6266 rcupdate.rcu_task_lazy_lim= [KNL] 6267 Number of callbacks on a given CPU that will 6268 cancel laziness on that CPU. Use -1 to disable 6269 cancellation of laziness, but be advised that 6270 doing so increases the danger of OOM due to 6271 callback flooding. 6272 6273 rcupdate.rcu_task_stall_info= [KNL] 6274 Set initial timeout in jiffies for RCU task stall 6275 informational messages, which give some indication 6276 of the problem for those not patient enough to 6277 wait for ten minutes. Informational messages are 6278 only printed prior to the stall-warning message 6279 for a given grace period. Disable with a value 6280 less than or equal to zero. Defaults to ten 6281 seconds. A change in value does not take effect 6282 until the beginning of the next grace period. 6283 6284 rcupdate.rcu_task_stall_info_mult= [KNL] 6285 Multiplier for time interval between successive 6286 RCU task stall informational messages for a given 6287 RCU tasks grace period. This value is clamped 6288 to one through ten, inclusive. It defaults to 6289 the value three, so that the first informational 6290 message is printed 10 seconds into the grace 6291 period, the second at 40 seconds, the third at 6292 160 seconds, and then the stall warning at 600 6293 seconds would prevent a fourth at 640 seconds. 6294 6295 rcupdate.rcu_task_stall_timeout= [KNL] 6296 Set timeout in jiffies for RCU task stall 6297 warning messages. Disable with a value less 6298 than or equal to zero. Defaults to ten minutes. 6299 A change in value does not take effect until 6300 the beginning of the next grace period. 6301 6302 rcupdate.rcu_tasks_lazy_ms= [KNL] 6303 Set timeout in milliseconds RCU Tasks asynchronous 6304 callback batching for call_rcu_tasks(). 6305 A negative value will take the default. A value 6306 of zero will disable batching. Batching is 6307 always disabled for synchronize_rcu_tasks(). 6308 6309 rcupdate.rcu_tasks_trace_lazy_ms= [KNL] 6310 Set timeout in milliseconds RCU Tasks 6311 Trace asynchronous callback batching for 6312 call_rcu_tasks_trace(). A negative value 6313 will take the default. A value of zero will 6314 disable batching. Batching is always disabled 6315 for synchronize_rcu_tasks_trace(). 6316 6317 rcupdate.rcu_self_test= [KNL] 6318 Run the RCU early boot self tests 6319 6320 rdinit= [KNL] 6321 Format: <full_path> 6322 Run specified binary instead of /init from the ramdisk, 6323 used for early userspace startup. See initrd. 6324 6325 rdrand= [X86,EARLY] 6326 force - Override the decision by the kernel to hide the 6327 advertisement of RDRAND support (this affects 6328 certain AMD processors because of buggy BIOS 6329 support, specifically around the suspend/resume 6330 path). 6331 6332 rdt= [HW,X86,RDT] 6333 Turn on/off individual RDT features. List is: 6334 cmt, mbmtotal, mbmlocal, l3cat, l3cdp, l2cat, l2cdp, 6335 mba, smba, bmec, abmc, sdciae. 6336 E.g. to turn on cmt and turn off mba use: 6337 rdt=cmt,!mba 6338 6339 reboot= [KNL] 6340 Format (x86 or x86_64): 6341 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] | d[efault] \ 6342 [[,]s[mp]#### \ 6343 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \ 6344 [[,]f[orce] 6345 Where reboot_mode is one of warm (soft) or cold (hard) or gpio 6346 (prefix with 'panic_' to set mode for panic 6347 reboot only), 6348 reboot_type is one of bios, acpi, kbd, triple, efi, or pci, 6349 reboot_force is either force or not specified, 6350 reboot_cpu is s[mp]#### with #### being the processor 6351 to be used for rebooting. 6352 6353 acpi 6354 Use the ACPI RESET_REG in the FADT. If ACPI is not 6355 configured or the ACPI reset does not work, the reboot 6356 path attempts the reset using the keyboard controller. 6357 6358 bios 6359 Use the CPU reboot vector for warm reset 6360 6361 cold 6362 Set the cold reboot flag 6363 6364 default 6365 There are some built-in platform specific "quirks" 6366 - you may see: "reboot: <name> series board detected. 6367 Selecting <type> for reboots." In the case where you 6368 think the quirk is in error (e.g. you have newer BIOS, 6369 or newer board) using this option will ignore the 6370 built-in quirk table, and use the generic default 6371 reboot actions. 6372 6373 efi 6374 Use efi reset_system runtime service. If EFI is not 6375 configured or the EFI reset does not work, the reboot 6376 path attempts the reset using the keyboard controller. 6377 6378 force 6379 Don't stop other CPUs on reboot. This can make reboot 6380 more reliable in some cases. 6381 6382 kbd 6383 Use the keyboard controller. cold reset (default) 6384 6385 pci 6386 Use a write to the PCI config space register 0xcf9 to 6387 trigger reboot. 6388 6389 triple 6390 Force a triple fault (init) 6391 6392 warm 6393 Don't set the cold reboot flag 6394 6395 Using warm reset will be much faster especially on big 6396 memory systems because the BIOS will not go through 6397 the memory check. Disadvantage is that not all 6398 hardware will be completely reinitialized on reboot so 6399 there may be boot problems on some systems. 6400 6401 6402 refscale.holdoff= [KNL] 6403 Set test-start holdoff period. The purpose of 6404 this parameter is to delay the start of the 6405 test until boot completes in order to avoid 6406 interference. 6407 6408 refscale.lookup_instances= [KNL] 6409 Number of data elements to use for the forms of 6410 SLAB_TYPESAFE_BY_RCU testing. A negative number 6411 is negated and multiplied by nr_cpu_ids, while 6412 zero specifies nr_cpu_ids. 6413 6414 refscale.loops= [KNL] 6415 Set the number of loops over the synchronization 6416 primitive under test. Increasing this number 6417 reduces noise due to loop start/end overhead, 6418 but the default has already reduced the per-pass 6419 noise to a handful of picoseconds on ca. 2020 6420 x86 laptops. 6421 6422 refscale.nreaders= [KNL] 6423 Set number of readers. The default value of -1 6424 selects N, where N is roughly 75% of the number 6425 of CPUs. A value of zero is an interesting choice. 6426 6427 refscale.nruns= [KNL] 6428 Set number of runs, each of which is dumped onto 6429 the console log. 6430 6431 refscale.readdelay= [KNL] 6432 Set the read-side critical-section duration, 6433 measured in microseconds. 6434 6435 refscale.scale_type= [KNL] 6436 Specify the read-protection implementation to test. 6437 6438 refscale.shutdown= [KNL] 6439 Shut down the system at the end of the performance 6440 test. This defaults to 1 (shut it down) when 6441 refscale is built into the kernel and to 0 (leave 6442 it running) when refscale is built as a module. 6443 6444 refscale.verbose= [KNL] 6445 Enable additional printk() statements. 6446 6447 refscale.verbose_batched= [KNL] 6448 Batch the additional printk() statements. If zero 6449 (the default) or negative, print everything. Otherwise, 6450 print every Nth verbose statement, where N is the value 6451 specified. 6452 6453 regulator_ignore_unused 6454 [REGULATOR] 6455 Prevents regulator framework from disabling regulators 6456 that are unused, due no driver claiming them. This may 6457 be useful for debug and development, but should not be 6458 needed on a platform with proper driver support. 6459 6460 relax_domain_level= 6461 [KNL, SMP] Set scheduler's default relax_domain_level. 6462 See Documentation/admin-guide/cgroup-v1/cpusets.rst. 6463 6464 reserve= [KNL,BUGS] Force kernel to ignore I/O ports or memory 6465 Format: <base1>,<size1>[,<base2>,<size2>,...] 6466 Reserve I/O ports or memory so the kernel won't use 6467 them. If <base> is less than 0x10000, the region 6468 is assumed to be I/O ports; otherwise it is memory. 6469 6470 reserve_mem= [RAM] 6471 Format: nn[KMG]:<align>:<label> 6472 Reserve physical memory and label it with a name that 6473 other subsystems can use to access it. This is typically 6474 used for systems that do not wipe the RAM, and this command 6475 line will try to reserve the same physical memory on 6476 soft reboots. Note, it is not guaranteed to be the same 6477 location. For example, if anything about the system changes 6478 or if booting a different kernel. It can also fail if KASLR 6479 places the kernel at the location of where the RAM reservation 6480 was from a previous boot, the new reservation will be at a 6481 different location. 6482 Any subsystem using this feature must add a way to verify 6483 that the contents of the physical memory is from a previous 6484 boot, as there may be cases where the memory will not be 6485 located at the same location. 6486 6487 The format is size:align:label for example, to request 6488 12 megabytes of 4096 alignment for ramoops: 6489 6490 reserve_mem=12M:4096:oops ramoops.mem_name=oops 6491 6492 reservetop= [X86-32,EARLY] 6493 Format: nn[KMG] 6494 Reserves a hole at the top of the kernel virtual 6495 address space. 6496 6497 reset_devices [KNL] Force drivers to reset the underlying device 6498 during initialization. 6499 6500 resume= [SWSUSP] 6501 Specify the partition device for software suspend 6502 Format: 6503 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>} 6504 6505 resume_offset= [SWSUSP] 6506 Specify the offset from the beginning of the partition 6507 given by "resume=" at which the swap header is located, 6508 in <PAGE_SIZE> units (needed only for swap files). 6509 See Documentation/power/swsusp-and-swap-files.rst 6510 6511 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to 6512 read the resume files 6513 6514 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up. 6515 Useful for devices that are detected asynchronously 6516 (e.g. USB and MMC devices). 6517 6518 retain_initrd [RAM] Keep initrd memory after extraction. After boot, it will 6519 be accessible via /sys/firmware/initrd. 6520 6521 retbleed= [X86] Control mitigation of RETBleed (Arbitrary 6522 Speculative Code Execution with Return Instructions) 6523 vulnerability. 6524 6525 AMD-based UNRET and IBPB mitigations alone do not stop 6526 sibling threads from influencing the predictions of other 6527 sibling threads. For that reason, STIBP is used on pro- 6528 cessors that support it, and mitigate SMT on processors 6529 that don't. 6530 6531 off - no mitigation 6532 auto - automatically select a mitigation 6533 auto,nosmt - automatically select a mitigation, 6534 disabling SMT if necessary for 6535 the full mitigation (only on Zen1 6536 and older without STIBP). 6537 ibpb - On AMD, mitigate short speculation 6538 windows on basic block boundaries too. 6539 Safe, highest perf impact. It also 6540 enables STIBP if present. Not suitable 6541 on Intel. 6542 ibpb,nosmt - Like "ibpb" above but will disable SMT 6543 when STIBP is not available. This is 6544 the alternative for systems which do not 6545 have STIBP. 6546 unret - Force enable untrained return thunks, 6547 only effective on AMD f15h-f17h based 6548 systems. 6549 unret,nosmt - Like unret, but will disable SMT when STIBP 6550 is not available. This is the alternative for 6551 systems which do not have STIBP. 6552 6553 Selecting 'auto' will choose a mitigation method at run 6554 time according to the CPU. 6555 6556 Not specifying this option is equivalent to retbleed=auto. 6557 6558 rfkill.default_state= 6559 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm, 6560 etc. communication is blocked by default. 6561 1 Unblocked. 6562 6563 rfkill.master_switch_mode= 6564 0 The "airplane mode" button does nothing. 6565 1 The "airplane mode" button toggles between everything 6566 blocked and the previous configuration. 6567 2 The "airplane mode" button toggles between everything 6568 blocked and everything unblocked. 6569 6570 ring3mwait=disable 6571 [KNL] Disable ring 3 MONITOR/MWAIT feature on supported 6572 CPUs. 6573 6574 riscv_isa_fallback [RISCV,EARLY] 6575 When CONFIG_RISCV_ISA_FALLBACK is not enabled, permit 6576 falling back to detecting extension support by parsing 6577 "riscv,isa" property on devicetree systems when the 6578 replacement properties are not found. See the Kconfig 6579 entry for RISCV_ISA_FALLBACK. 6580 6581 ro [KNL] Mount root device read-only on boot 6582 6583 rodata= [KNL,EARLY] 6584 on Mark read-only kernel memory as read-only (default). 6585 off Leave read-only kernel memory writable for debugging. 6586 noalias Mark read-only kernel memory as read-only but retain 6587 writable aliases in the direct map for regions outside 6588 of the kernel image. [arm64] 6589 6590 rockchip.usb_uart 6591 [EARLY] 6592 Enable the uart passthrough on the designated usb port 6593 on Rockchip SoCs. When active, the signals of the 6594 debug-uart get routed to the D+ and D- pins of the usb 6595 port and the regular usb controller gets disabled. 6596 6597 root= [KNL] Root filesystem 6598 Usually this is a block device specifier of some kind, 6599 see the early_lookup_bdev comment in 6600 block/early-lookup.c for details. 6601 Alternatively this can be "ram" for the legacy initial 6602 ramdisk, "nfs" and "cifs" for root on a network file 6603 system, or "mtd" and "ubi" for mounting from raw flash. 6604 6605 rootdelay= [KNL] Delay (in seconds) to pause before attempting to 6606 mount the root filesystem 6607 6608 rootflags= [KNL] Set root filesystem mount option string 6609 6610 initramfs_options= [KNL] 6611 Specify mount options for for the initramfs mount. 6612 6613 rootfstype= [KNL] Set root filesystem type 6614 6615 rootwait [KNL] Wait (indefinitely) for root device to show up. 6616 Useful for devices that are detected asynchronously 6617 (e.g. USB and MMC devices). 6618 6619 rootwait= [KNL] Maximum time (in seconds) to wait for root device 6620 to show up before attempting to mount the root 6621 filesystem. 6622 6623 rproc_mem=nn[KMG][@address] 6624 [KNL,ARM,CMA] Remoteproc physical memory block. 6625 Memory area to be used by remote processor image, 6626 managed by CMA. 6627 6628 rseq_debug= [KNL] Enable or disable restartable sequence 6629 debug mode. Defaults to CONFIG_RSEQ_DEBUG_DEFAULT_ENABLE. 6630 Format: <bool> 6631 6632 rt_group_sched= [KNL] Enable or disable SCHED_RR/FIFO group scheduling 6633 when CONFIG_RT_GROUP_SCHED=y. Defaults to 6634 !CONFIG_RT_GROUP_SCHED_DEFAULT_DISABLED. 6635 Format: <bool> 6636 6637 rw [KNL] Mount root device read-write on boot 6638 6639 S [KNL] Run init in single mode 6640 6641 s390_iommu= [HW,S390] 6642 Set s390 IOTLB flushing mode 6643 strict 6644 With strict flushing every unmap operation will result 6645 in an IOTLB flush. Default is lazy flushing before 6646 reuse, which is faster. Deprecated, equivalent to 6647 iommu.strict=1. 6648 6649 s390_iommu_aperture= [KNL,S390] 6650 Specifies the size of the per device DMA address space 6651 accessible through the DMA and IOMMU APIs as a decimal 6652 factor of the size of main memory. 6653 The default is 1 meaning that one can concurrently use 6654 as many DMA addresses as physical memory is installed, 6655 if supported by hardware, and thus map all of memory 6656 once. With a value of 2 one can map all of memory twice 6657 and so on. As a special case a factor of 0 imposes no 6658 restrictions other than those given by hardware at the 6659 cost of significant additional memory use for tables. 6660 6661 sa1100ir [NET] 6662 See drivers/net/irda/sa1100_ir.c. 6663 6664 sched_proxy_exec= [KNL] 6665 Enables or disables "proxy execution" style 6666 solution to mutex-based priority inversion. 6667 Format: <bool> 6668 6669 sched_verbose [KNL,EARLY] Enables verbose scheduler debug messages. 6670 6671 schedstats= [KNL,X86] Enable or disable scheduled statistics. 6672 Allowed values are enable and disable. This feature 6673 incurs a small amount of overhead in the scheduler 6674 but is useful for debugging and performance tuning. 6675 6676 sched_thermal_decay_shift= 6677 [Deprecated] 6678 [KNL, SMP] Set a decay shift for scheduler thermal 6679 pressure signal. Thermal pressure signal follows the 6680 default decay period of other scheduler pelt 6681 signals(usually 32 ms but configurable). Setting 6682 sched_thermal_decay_shift will left shift the decay 6683 period for the thermal pressure signal by the shift 6684 value. 6685 i.e. with the default pelt decay period of 32 ms 6686 sched_thermal_decay_shift thermal pressure decay pr 6687 1 64 ms 6688 2 128 ms 6689 and so on. 6690 Format: integer between 0 and 10 6691 Default is 0. 6692 6693 scftorture.holdoff= [KNL] 6694 Number of seconds to hold off before starting 6695 test. Defaults to zero for module insertion and 6696 to 10 seconds for built-in smp_call_function() 6697 tests. 6698 6699 scftorture.longwait= [KNL] 6700 Request ridiculously long waits randomly selected 6701 up to the chosen limit in seconds. Zero (the 6702 default) disables this feature. Please note 6703 that requesting even small non-zero numbers of 6704 seconds can result in RCU CPU stall warnings, 6705 softlockup complaints, and so on. 6706 6707 scftorture.nthreads= [KNL] 6708 Number of kthreads to spawn to invoke the 6709 smp_call_function() family of functions. 6710 The default of -1 specifies a number of kthreads 6711 equal to the number of CPUs. 6712 6713 scftorture.onoff_holdoff= [KNL] 6714 Number seconds to wait after the start of the 6715 test before initiating CPU-hotplug operations. 6716 6717 scftorture.onoff_interval= [KNL] 6718 Number seconds to wait between successive 6719 CPU-hotplug operations. Specifying zero (which 6720 is the default) disables CPU-hotplug operations. 6721 6722 scftorture.shutdown_secs= [KNL] 6723 The number of seconds following the start of the 6724 test after which to shut down the system. The 6725 default of zero avoids shutting down the system. 6726 Non-zero values are useful for automated tests. 6727 6728 scftorture.stat_interval= [KNL] 6729 The number of seconds between outputting the 6730 current test statistics to the console. A value 6731 of zero disables statistics output. 6732 6733 scftorture.stutter_cpus= [KNL] 6734 The number of jiffies to wait between each change 6735 to the set of CPUs under test. 6736 6737 scftorture.use_cpus_read_lock= [KNL] 6738 Use use_cpus_read_lock() instead of the default 6739 preempt_disable() to disable CPU hotplug 6740 while invoking one of the smp_call_function*() 6741 functions. 6742 6743 scftorture.verbose= [KNL] 6744 Enable additional printk() statements. 6745 6746 scftorture.weight_single= [KNL] 6747 The probability weighting to use for the 6748 smp_call_function_single() function with a zero 6749 "wait" parameter. A value of -1 selects the 6750 default if all other weights are -1. However, 6751 if at least one weight has some other value, a 6752 value of -1 will instead select a weight of zero. 6753 6754 scftorture.weight_single_wait= [KNL] 6755 The probability weighting to use for the 6756 smp_call_function_single() function with a 6757 non-zero "wait" parameter. See weight_single. 6758 6759 scftorture.weight_many= [KNL] 6760 The probability weighting to use for the 6761 smp_call_function_many() function with a zero 6762 "wait" parameter. See weight_single. 6763 Note well that setting a high probability for 6764 this weighting can place serious IPI load 6765 on the system. 6766 6767 scftorture.weight_many_wait= [KNL] 6768 The probability weighting to use for the 6769 smp_call_function_many() function with a 6770 non-zero "wait" parameter. See weight_single 6771 and weight_many. 6772 6773 scftorture.weight_all= [KNL] 6774 The probability weighting to use for the 6775 smp_call_function_all() function with a zero 6776 "wait" parameter. See weight_single and 6777 weight_many. 6778 6779 scftorture.weight_all_wait= [KNL] 6780 The probability weighting to use for the 6781 smp_call_function_all() function with a 6782 non-zero "wait" parameter. See weight_single 6783 and weight_many. 6784 6785 sdw_mclk_divider=[SDW] 6786 Specify the MCLK divider for Intel SoundWire buses in 6787 case the BIOS does not provide the clock rate properly. 6788 6789 skew_tick= [KNL,EARLY] Offset the periodic timer tick per cpu to mitigate 6790 xtime_lock contention on larger systems, and/or RCU lock 6791 contention on all systems with CONFIG_MAXSMP set. 6792 Format: { "0" | "1" } 6793 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1" 6794 1 -- enable. 6795 Note: increases power consumption, thus should only be 6796 enabled if running jitter sensitive (HPC/RT) workloads. 6797 6798 security= [SECURITY] Choose a legacy "major" security module to 6799 enable at boot. This has been deprecated by the 6800 "lsm=" parameter. 6801 6802 selinux= [SELINUX] Disable or enable SELinux at boot time. 6803 Format: { "0" | "1" } 6804 See security/selinux/Kconfig help text. 6805 0 -- disable. 6806 1 -- enable. 6807 Default value is 1. 6808 6809 serialnumber [BUGS=X86-32] 6810 6811 sev=option[,option...] [X86-64] 6812 6813 debug 6814 Enable debug messages. 6815 6816 nosnp 6817 Do not enable SEV-SNP (applies to host/hypervisor 6818 only). Setting 'nosnp' avoids the RMP check overhead 6819 in memory accesses when users do not want to run 6820 SEV-SNP guests. 6821 6822 shapers= [NET] 6823 Maximal number of shapers. 6824 6825 show_lapic= [APIC,X86] Advanced Programmable Interrupt Controller 6826 Limit apic dumping. The parameter defines the maximal 6827 number of local apics being dumped. Also it is possible 6828 to set it to "all" by meaning -- no limit here. 6829 Format: { 1 (default) | 2 | ... | all }. 6830 The parameter valid if only apic=debug or 6831 apic=verbose is specified. 6832 Example: apic=debug show_lapic=all 6833 6834 slab_debug[=options[,slabs][;[options[,slabs]]...] [MM] 6835 Enabling slab_debug allows one to determine the 6836 culprit if slab objects become corrupted. Enabling 6837 slab_debug can create guard zones around objects and 6838 may poison objects when not in use. Also tracks the 6839 last alloc / free. For more information see 6840 Documentation/admin-guide/mm/slab.rst. 6841 (slub_debug legacy name also accepted for now) 6842 6843 Using this option implies the "no_hash_pointers" 6844 option which can be undone by adding the 6845 "hash_pointers=always" option. 6846 6847 slab_max_order= [MM] 6848 Determines the maximum allowed order for slabs. 6849 A high setting may cause OOMs due to memory 6850 fragmentation. For more information see 6851 Documentation/admin-guide/mm/slab.rst. 6852 (slub_max_order legacy name also accepted for now) 6853 6854 slab_merge [MM] 6855 Enable merging of slabs with similar size when the 6856 kernel is built without CONFIG_SLAB_MERGE_DEFAULT. 6857 (slub_merge legacy name also accepted for now) 6858 6859 slab_min_objects= [MM] 6860 The minimum number of objects per slab. SLUB will 6861 increase the slab order up to slab_max_order to 6862 generate a sufficiently large slab able to contain 6863 the number of objects indicated. The higher the number 6864 of objects the smaller the overhead of tracking slabs 6865 and the less frequently locks need to be acquired. 6866 For more information see 6867 Documentation/admin-guide/mm/slab.rst. 6868 (slub_min_objects legacy name also accepted for now) 6869 6870 slab_min_order= [MM] 6871 Determines the minimum page order for slabs. Must be 6872 lower or equal to slab_max_order. For more information see 6873 Documentation/admin-guide/mm/slab.rst. 6874 (slub_min_order legacy name also accepted for now) 6875 6876 slab_nomerge [MM] 6877 Disable merging of slabs with similar size. May be 6878 necessary if there is some reason to distinguish 6879 allocs to different slabs, especially in hardened 6880 environments where the risk of heap overflows and 6881 layout control by attackers can usually be 6882 frustrated by disabling merging. This will reduce 6883 most of the exposure of a heap attack to a single 6884 cache (risks via metadata attacks are mostly 6885 unchanged). Debug options disable merging on their 6886 own. 6887 For more information see 6888 Documentation/admin-guide/mm/slab.rst. 6889 (slub_nomerge legacy name also accepted for now) 6890 6891 slab_strict_numa [MM] 6892 Support memory policies on a per object level 6893 in the slab allocator. The default is for memory 6894 policies to be applied at the folio level when 6895 a new folio is needed or a partial folio is 6896 retrieved from the lists. Increases overhead 6897 in the slab fastpaths but gains more accurate 6898 NUMA kernel object placement which helps with slow 6899 interconnects in NUMA systems. 6900 6901 slram= [HW,MTD] 6902 6903 smart2= [HW] 6904 Format: <io1>[,<io2>[,...,<io8>]] 6905 6906 smp.csd_lock_timeout= [KNL] 6907 Specify the period of time in milliseconds 6908 that smp_call_function() and friends will wait 6909 for a CPU to release the CSD lock. This is 6910 useful when diagnosing bugs involving CPUs 6911 disabling interrupts for extended periods 6912 of time. Defaults to 5,000 milliseconds, and 6913 setting a value of zero disables this feature. 6914 This feature may be more efficiently disabled 6915 using the csdlock_debug- kernel parameter. 6916 6917 smp.panic_on_ipistall= [KNL] 6918 If a csd_lock_timeout extends for more than 6919 the specified number of milliseconds, panic the 6920 system. By default, let CSD-lock acquisition 6921 take as long as they take. Specifying 300,000 6922 for this value provides a 5-minute timeout. 6923 6924 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices 6925 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port 6926 smsc-ircc2.ircc_sir= [HW] SIR base I/O port 6927 smsc-ircc2.ircc_fir= [HW] FIR base I/O port 6928 smsc-ircc2.ircc_irq= [HW] IRQ line 6929 smsc-ircc2.ircc_dma= [HW] DMA channel 6930 smsc-ircc2.ircc_transceiver= [HW] Transceiver type: 6931 0: Toshiba Satellite 1800 (GP data pin select) 6932 1: Fast pin select (default) 6933 2: ATC IRMode 6934 6935 smt= [KNL,MIPS,S390,EARLY] Set the maximum number of threads 6936 (logical CPUs) to use per physical CPU on systems 6937 capable of symmetric multithreading (SMT). Will 6938 be capped to the actual hardware limit. 6939 Format: <integer> 6940 Default: -1 (no limit) 6941 6942 softlockup_panic= 6943 [KNL] Should the soft-lockup detector generate panics. 6944 Format: 0 | 1 6945 6946 A value of 1 instructs the soft-lockup detector 6947 to panic the machine when a soft-lockup occurs. It is 6948 also controlled by the kernel.softlockup_panic sysctl 6949 and CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC, which is the 6950 respective build-time switch to that functionality. 6951 6952 softlockup_all_cpu_backtrace= 6953 [KNL] Should the soft-lockup detector generate 6954 backtraces on all cpus. 6955 Format: 0 | 1 6956 6957 sonypi.*= [HW] Sony Programmable I/O Control Device driver 6958 See Documentation/admin-guide/laptops/sonypi.rst 6959 6960 spectre_bhi= [X86] Control mitigation of Branch History Injection 6961 (BHI) vulnerability. This setting affects the 6962 deployment of the HW BHI control and the SW BHB 6963 clearing sequence. 6964 6965 on - (default) Enable the HW or SW mitigation as 6966 needed. This protects the kernel from 6967 both syscalls and VMs. 6968 vmexit - On systems which don't have the HW mitigation 6969 available, enable the SW mitigation on vmexit 6970 ONLY. On such systems, the host kernel is 6971 protected from VM-originated BHI attacks, but 6972 may still be vulnerable to syscall attacks. 6973 off - Disable the mitigation. 6974 6975 spectre_v2= [X86,EARLY] Control mitigation of Spectre variant 2 6976 (indirect branch speculation) vulnerability. 6977 The default operation protects the kernel from 6978 user space attacks. 6979 6980 on - unconditionally enable, implies 6981 spectre_v2_user=on 6982 off - unconditionally disable, implies 6983 spectre_v2_user=off 6984 auto - kernel detects whether your CPU model is 6985 vulnerable 6986 6987 Selecting 'on' will, and 'auto' may, choose a 6988 mitigation method at run time according to the 6989 CPU, the available microcode, the setting of the 6990 CONFIG_MITIGATION_RETPOLINE configuration option, 6991 and the compiler with which the kernel was built. 6992 6993 Selecting 'on' will also enable the mitigation 6994 against user space to user space task attacks. 6995 Selecting specific mitigation does not force enable 6996 user mitigations. 6997 6998 Selecting 'off' will disable both the kernel and 6999 the user space protections. 7000 7001 Specific mitigations can also be selected manually: 7002 7003 retpoline - replace indirect branches 7004 retpoline,generic - Retpolines 7005 retpoline,lfence - LFENCE; indirect branch 7006 retpoline,amd - alias for retpoline,lfence 7007 eibrs - Enhanced/Auto IBRS 7008 eibrs,retpoline - Enhanced/Auto IBRS + Retpolines 7009 eibrs,lfence - Enhanced/Auto IBRS + LFENCE 7010 ibrs - use IBRS to protect kernel 7011 7012 Not specifying this option is equivalent to 7013 spectre_v2=auto. 7014 7015 spectre_v2_user= 7016 [X86] Control mitigation of Spectre variant 2 7017 (indirect branch speculation) vulnerability between 7018 user space tasks 7019 7020 on - Unconditionally enable mitigations. Is 7021 enforced by spectre_v2=on 7022 7023 off - Unconditionally disable mitigations. Is 7024 enforced by spectre_v2=off 7025 7026 prctl - Indirect branch speculation is enabled, 7027 but mitigation can be enabled via prctl 7028 per thread. The mitigation control state 7029 is inherited on fork. 7030 7031 prctl,ibpb 7032 - Like "prctl" above, but only STIBP is 7033 controlled per thread. IBPB is issued 7034 always when switching between different user 7035 space processes. 7036 7037 seccomp 7038 - Same as "prctl" above, but all seccomp 7039 threads will enable the mitigation unless 7040 they explicitly opt out. 7041 7042 seccomp,ibpb 7043 - Like "seccomp" above, but only STIBP is 7044 controlled per thread. IBPB is issued 7045 always when switching between different 7046 user space processes. 7047 7048 auto - Kernel selects the mitigation depending on 7049 the available CPU features and vulnerability. 7050 7051 Default mitigation: "prctl" 7052 7053 Not specifying this option is equivalent to 7054 spectre_v2_user=auto. 7055 7056 spec_rstack_overflow= 7057 [X86,EARLY] Control RAS overflow mitigation on AMD Zen CPUs 7058 7059 off - Disable mitigation 7060 microcode - Enable microcode mitigation only 7061 safe-ret - Enable sw-only safe RET mitigation (default) 7062 ibpb - Enable mitigation by issuing IBPB on 7063 kernel entry 7064 ibpb-vmexit - Issue IBPB only on VMEXIT 7065 (cloud-specific mitigation) 7066 7067 spec_store_bypass_disable= 7068 [HW,EARLY] Control Speculative Store Bypass (SSB) Disable mitigation 7069 (Speculative Store Bypass vulnerability) 7070 7071 Certain CPUs are vulnerable to an exploit against a 7072 a common industry wide performance optimization known 7073 as "Speculative Store Bypass" in which recent stores 7074 to the same memory location may not be observed by 7075 later loads during speculative execution. The idea 7076 is that such stores are unlikely and that they can 7077 be detected prior to instruction retirement at the 7078 end of a particular speculation execution window. 7079 7080 In vulnerable processors, the speculatively forwarded 7081 store can be used in a cache side channel attack, for 7082 example to read memory to which the attacker does not 7083 directly have access (e.g. inside sandboxed code). 7084 7085 This parameter controls whether the Speculative Store 7086 Bypass optimization is used. 7087 7088 On x86 the options are: 7089 7090 on - Unconditionally disable Speculative Store Bypass 7091 off - Unconditionally enable Speculative Store Bypass 7092 auto - Kernel detects whether the CPU model contains an 7093 implementation of Speculative Store Bypass and 7094 picks the most appropriate mitigation. If the 7095 CPU is not vulnerable, "off" is selected. If the 7096 CPU is vulnerable the default mitigation is 7097 architecture and Kconfig dependent. See below. 7098 prctl - Control Speculative Store Bypass per thread 7099 via prctl. Speculative Store Bypass is enabled 7100 for a process by default. The state of the control 7101 is inherited on fork. 7102 seccomp - Same as "prctl" above, but all seccomp threads 7103 will disable SSB unless they explicitly opt out. 7104 7105 Default mitigations: 7106 X86: "prctl" 7107 7108 On powerpc the options are: 7109 7110 on,auto - On Power8 and Power9 insert a store-forwarding 7111 barrier on kernel entry and exit. On Power7 7112 perform a software flush on kernel entry and 7113 exit. 7114 off - No action. 7115 7116 Not specifying this option is equivalent to 7117 spec_store_bypass_disable=auto. 7118 7119 split_lock_detect= 7120 [X86] Enable split lock detection or bus lock detection 7121 7122 When enabled (and if hardware support is present), atomic 7123 instructions that access data across cache line 7124 boundaries will result in an alignment check exception 7125 for split lock detection or a debug exception for 7126 bus lock detection. 7127 7128 off - not enabled 7129 7130 warn - the kernel will emit rate-limited warnings 7131 about applications triggering the #AC 7132 exception or the #DB exception. This mode is 7133 the default on CPUs that support split lock 7134 detection or bus lock detection. Default 7135 behavior is by #AC if both features are 7136 enabled in hardware. 7137 7138 fatal - the kernel will send SIGBUS to applications 7139 that trigger the #AC exception or the #DB 7140 exception. Default behavior is by #AC if 7141 both features are enabled in hardware. 7142 7143 ratelimit:N - 7144 Set system wide rate limit to N bus locks 7145 per second for bus lock detection. 7146 0 < N <= 1000. 7147 7148 N/A for split lock detection. 7149 7150 7151 If an #AC exception is hit in the kernel or in 7152 firmware (i.e. not while executing in user mode) 7153 the kernel will oops in either "warn" or "fatal" 7154 mode. 7155 7156 #DB exception for bus lock is triggered only when 7157 CPL > 0. 7158 7159 srbds= [X86,INTEL,EARLY] 7160 Control the Special Register Buffer Data Sampling 7161 (SRBDS) mitigation. 7162 7163 Certain CPUs are vulnerable to an MDS-like 7164 exploit which can leak bits from the random 7165 number generator. 7166 7167 By default, this issue is mitigated by 7168 microcode. However, the microcode fix can cause 7169 the RDRAND and RDSEED instructions to become 7170 much slower. Among other effects, this will 7171 result in reduced throughput from /dev/urandom. 7172 7173 The microcode mitigation can be disabled with 7174 the following option: 7175 7176 off: Disable mitigation and remove 7177 performance impact to RDRAND and RDSEED 7178 7179 srcutree.big_cpu_lim [KNL] 7180 Specifies the number of CPUs constituting a 7181 large system, such that srcu_struct structures 7182 should immediately allocate an srcu_node array. 7183 This kernel-boot parameter defaults to 128, 7184 but takes effect only when the low-order four 7185 bits of srcutree.convert_to_big is equal to 3 7186 (decide at boot). 7187 7188 srcutree.convert_to_big [KNL] 7189 Specifies under what conditions an SRCU tree 7190 srcu_struct structure will be converted to big 7191 form, that is, with an rcu_node tree: 7192 7193 0: Never. 7194 1: At init_srcu_struct() time. 7195 2: When rcutorture decides to. 7196 3: Decide at boot time (default). 7197 0x1X: Above plus if high contention. 7198 7199 Either way, the srcu_node tree will be sized based 7200 on the actual runtime number of CPUs (nr_cpu_ids) 7201 instead of the compile-time CONFIG_NR_CPUS. 7202 7203 srcutree.counter_wrap_check [KNL] 7204 Specifies how frequently to check for 7205 grace-period sequence counter wrap for the 7206 srcu_data structure's ->srcu_gp_seq_needed field. 7207 The greater the number of bits set in this kernel 7208 parameter, the less frequently counter wrap will 7209 be checked for. Note that the bottom two bits 7210 are ignored. 7211 7212 srcutree.exp_holdoff [KNL] 7213 Specifies how many nanoseconds must elapse 7214 since the end of the last SRCU grace period for 7215 a given srcu_struct until the next normal SRCU 7216 grace period will be considered for automatic 7217 expediting. Set to zero to disable automatic 7218 expediting. 7219 7220 srcutree.srcu_max_nodelay [KNL] 7221 Specifies the number of no-delay instances 7222 per jiffy for which the SRCU grace period 7223 worker thread will be rescheduled with zero 7224 delay. Beyond this limit, worker thread will 7225 be rescheduled with a sleep delay of one jiffy. 7226 7227 srcutree.srcu_max_nodelay_phase [KNL] 7228 Specifies the per-grace-period phase, number of 7229 non-sleeping polls of readers. Beyond this limit, 7230 grace period worker thread will be rescheduled 7231 with a sleep delay of one jiffy, between each 7232 rescan of the readers, for a grace period phase. 7233 7234 srcutree.srcu_retry_check_delay [KNL] 7235 Specifies number of microseconds of non-sleeping 7236 delay between each non-sleeping poll of readers. 7237 7238 srcutree.small_contention_lim [KNL] 7239 Specifies the number of update-side contention 7240 events per jiffy will be tolerated before 7241 initiating a conversion of an srcu_struct 7242 structure to big form. Note that the value of 7243 srcutree.convert_to_big must have the 0x10 bit 7244 set for contention-based conversions to occur. 7245 7246 ssbd= [ARM64,HW,EARLY] 7247 Speculative Store Bypass Disable control 7248 7249 On CPUs that are vulnerable to the Speculative 7250 Store Bypass vulnerability and offer a 7251 firmware based mitigation, this parameter 7252 indicates how the mitigation should be used: 7253 7254 force-on: Unconditionally enable mitigation for 7255 for both kernel and userspace 7256 force-off: Unconditionally disable mitigation for 7257 for both kernel and userspace 7258 kernel: Always enable mitigation in the 7259 kernel, and offer a prctl interface 7260 to allow userspace to register its 7261 interest in being mitigated too. 7262 7263 stack_guard_gap= [MM] 7264 override the default stack gap protection. The value 7265 is in page units and it defines how many pages prior 7266 to (for stacks growing down) resp. after (for stacks 7267 growing up) the main stack are reserved for no other 7268 mapping. Default value is 256 pages. 7269 7270 stack_depot_disable= [KNL,EARLY] 7271 Setting this to true through kernel command line will 7272 disable the stack depot thereby saving the static memory 7273 consumed by the stack hash table. By default this is set 7274 to false. 7275 7276 stack_depot_max_pools= [KNL,EARLY] 7277 Specify the maximum number of pools to use for storing 7278 stack traces. Pools are allocated on-demand up to this 7279 limit. Default value is 8191 pools. 7280 7281 stacktrace [FTRACE] 7282 Enable the stack tracer on boot up. 7283 7284 stacktrace_filter=[function-list] 7285 [FTRACE] Limit the functions that the stack tracer 7286 will trace at boot up. function-list is a comma-separated 7287 list of functions. This list can be changed at run 7288 time by the stack_trace_filter file in the debugfs 7289 tracing directory. Note, this enables stack tracing 7290 and the stacktrace above is not needed. 7291 7292 sti= [PARISC,HW] 7293 Format: <num> 7294 Set the STI (builtin display/keyboard on the HP-PARISC 7295 machines) console (graphic card) which should be used 7296 as the initial boot-console. 7297 See also comment in drivers/video/console/sticore.c. 7298 7299 sti_font= [HW] 7300 See comment in drivers/video/console/sticore.c. 7301 7302 stifb= [HW] 7303 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]] 7304 7305 strict_sas_size= 7306 [X86] 7307 Format: <bool> 7308 Enable or disable strict sigaltstack size checks 7309 against the required signal frame size which 7310 depends on the supported FPU features. This can 7311 be used to filter out binaries which have 7312 not yet been made aware of AT_MINSIGSTKSZ. 7313 7314 stress_hpt [PPC,EARLY] 7315 Limits the number of kernel HPT entries in the hash 7316 page table to increase the rate of hash page table 7317 faults on kernel addresses. 7318 7319 stress_slb [PPC,EARLY] 7320 Limits the number of kernel SLB entries, and flushes 7321 them frequently to increase the rate of SLB faults 7322 on kernel addresses. 7323 7324 no_slb_preload [PPC,EARLY] 7325 Disables slb preloading for userspace. 7326 7327 sunrpc.min_resvport= 7328 sunrpc.max_resvport= 7329 [NFS,SUNRPC] 7330 SunRPC servers often require that client requests 7331 originate from a privileged port (i.e. a port in the 7332 range 0 < portnr < 1024). 7333 An administrator who wishes to reserve some of these 7334 ports for other uses may adjust the range that the 7335 kernel's sunrpc client considers to be privileged 7336 using these two parameters to set the minimum and 7337 maximum port values. 7338 7339 sunrpc.svc_rpc_per_connection_limit= 7340 [NFS,SUNRPC] 7341 Limit the number of requests that the server will 7342 process in parallel from a single connection. 7343 The default value is 0 (no limit). 7344 7345 sunrpc.pool_mode= 7346 [NFS] 7347 Control how the NFS server code allocates CPUs to 7348 service thread pools. Depending on how many NICs 7349 you have and where their interrupts are bound, this 7350 option will affect which CPUs will do NFS serving. 7351 Note: this parameter cannot be changed while the 7352 NFS server is running. 7353 7354 auto the server chooses an appropriate mode 7355 automatically using heuristics 7356 global a single global pool contains all CPUs 7357 percpu one pool for each CPU 7358 pernode one pool for each NUMA node (equivalent 7359 to global on non-NUMA machines) 7360 7361 sunrpc.tcp_slot_table_entries= 7362 sunrpc.udp_slot_table_entries= 7363 [NFS,SUNRPC] 7364 Sets the upper limit on the number of simultaneous 7365 RPC calls that can be sent from the client to a 7366 server. Increasing these values may allow you to 7367 improve throughput, but will also increase the 7368 amount of memory reserved for use by the client. 7369 7370 suspend.pm_test_delay= 7371 [SUSPEND] 7372 Sets the number of seconds to remain in a suspend test 7373 mode before resuming the system (see 7374 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG 7375 is set. Default value is 5. 7376 7377 svm= [PPC] 7378 Format: { on | off | y | n | 1 | 0 } 7379 This parameter controls use of the Protected 7380 Execution Facility on pSeries. 7381 7382 swiotlb= [ARM,PPC,MIPS,X86,S390,EARLY] 7383 Format: { <int> [,<int>] | force | noforce } 7384 <int> -- Number of I/O TLB slabs 7385 <int> -- Second integer after comma. Number of swiotlb 7386 areas with their own lock. Will be rounded up 7387 to a power of 2. 7388 force -- force using of bounce buffers even if they 7389 wouldn't be automatically used by the kernel 7390 noforce -- Never use bounce buffers (for debugging) 7391 7392 switches= [HW,M68k,EARLY] 7393 7394 sysctl.*= [KNL] 7395 Set a sysctl parameter, right before loading the init 7396 process, as if the value was written to the respective 7397 /proc/sys/... file. Both '.' and '/' are recognized as 7398 separators. Unrecognized parameters and invalid values 7399 are reported in the kernel log. Sysctls registered 7400 later by a loaded module cannot be set this way. 7401 Example: sysctl.vm.swappiness=40 7402 7403 sysrq_always_enabled 7404 [KNL] 7405 Ignore sysrq setting - this boot parameter will 7406 neutralize any effect of /proc/sys/kernel/sysrq. 7407 Useful for debugging. 7408 7409 tcpmhash_entries= [KNL,NET] 7410 Set the number of tcp_metrics_hash slots. 7411 Default value is 8192 or 16384 depending on total 7412 ram pages. This is used to specify the TCP metrics 7413 cache size. See Documentation/networking/ip-sysctl.rst 7414 "tcp_no_metrics_save" section for more details. 7415 7416 tdfx= [HW,DRM] 7417 7418 test_suspend= [SUSPEND] 7419 Format: { "mem" | "standby" | "freeze" }[,N] 7420 Specify "mem" (for Suspend-to-RAM) or "standby" (for 7421 standby suspend) or "freeze" (for suspend type freeze) 7422 as the system sleep state during system startup with 7423 the optional capability to repeat N number of times. 7424 The system is woken from this state using a 7425 wakeup-capable RTC alarm. 7426 7427 thash_entries= [KNL,NET] 7428 Set number of hash buckets for TCP connection 7429 7430 thermal.act= [HW,ACPI] 7431 -1: disable all active trip points in all thermal zones 7432 <degrees C>: override all lowest active trip points 7433 7434 thermal.crt= [HW,ACPI] 7435 -1: disable all critical trip points in all thermal zones 7436 <degrees C>: override all critical trip points 7437 7438 thermal.off= [HW,ACPI] 7439 1: disable ACPI thermal control 7440 7441 thermal.psv= [HW,ACPI] 7442 -1: disable all passive trip points 7443 <degrees C>: override all passive trip points to this 7444 value 7445 7446 thermal.tzp= [HW,ACPI] 7447 Specify global default ACPI thermal zone polling rate 7448 <deci-seconds>: poll all this frequency 7449 0: no polling (default) 7450 7451 thp_anon= [KNL] 7452 Format: <size>[KMG],<size>[KMG]:<state>;<size>[KMG]-<size>[KMG]:<state> 7453 state is one of "always", "madvise", "never" or "inherit". 7454 Control the default behavior of the system with respect 7455 to anonymous transparent hugepages. 7456 Can be used multiple times for multiple anon THP sizes. 7457 See Documentation/admin-guide/mm/transhuge.rst for more 7458 details. 7459 7460 threadirqs [KNL,EARLY] 7461 Force threading of all interrupt handlers except those 7462 marked explicitly IRQF_NO_THREAD. 7463 7464 thp_shmem= [KNL] 7465 Format: <size>[KMG],<size>[KMG]:<policy>;<size>[KMG]-<size>[KMG]:<policy> 7466 Control the default policy of each hugepage size for the 7467 internal shmem mount. <policy> is one of policies available 7468 for the shmem mount ("always", "inherit", "never", "within_size", 7469 and "advise"). 7470 It can be used multiple times for multiple shmem THP sizes. 7471 See Documentation/admin-guide/mm/transhuge.rst for more 7472 details. 7473 7474 topology= [S390,EARLY] 7475 Format: {off | on} 7476 Specify if the kernel should make use of the cpu 7477 topology information if the hardware supports this. 7478 The scheduler will make use of this information and 7479 e.g. base its process migration decisions on it. 7480 Default is on. 7481 7482 torture.disable_onoff_at_boot= [KNL] 7483 Prevent the CPU-hotplug component of torturing 7484 until after init has spawned. 7485 7486 torture.ftrace_dump_at_shutdown= [KNL] 7487 Dump the ftrace buffer at torture-test shutdown, 7488 even if there were no errors. This can be a 7489 very costly operation when many torture tests 7490 are running concurrently, especially on systems 7491 with rotating-rust storage. 7492 7493 torture.verbose_sleep_frequency= [KNL] 7494 Specifies how many verbose printk()s should be 7495 emitted between each sleep. The default of zero 7496 disables verbose-printk() sleeping. 7497 7498 torture.verbose_sleep_duration= [KNL] 7499 Duration of each verbose-printk() sleep in jiffies. 7500 7501 tpm.disable_pcr_integrity= [HW,TPM] 7502 Do not protect PCR registers from unintended physical 7503 access, or interposers in the bus by the means of 7504 having an integrity protected session wrapped around 7505 TPM2_PCR_Extend command. Consider this in a situation 7506 where TPM is heavily utilized by IMA, thus protection 7507 causing a major performance hit, and the space where 7508 machines are deployed is by other means guarded. 7509 7510 tpm_crb_ffa.busy_timeout_ms= [ARM64,TPM] 7511 Maximum time in milliseconds to retry sending a message 7512 to the TPM service before giving up. This parameter controls 7513 how long the system will continue retrying when the TPM 7514 service is busy. 7515 Format: <unsigned int> 7516 Default: 2000 (2 seconds) 7517 7518 tpm_suspend_pcr=[HW,TPM] 7519 Format: integer pcr id 7520 Specify that at suspend time, the tpm driver 7521 should extend the specified pcr with zeros, 7522 as a workaround for some chips which fail to 7523 flush the last written pcr on TPM_SaveState. 7524 This will guarantee that all the other pcrs 7525 are saved. 7526 7527 tpm_tis.interrupts= [HW,TPM] 7528 Enable interrupts for the MMIO based physical layer 7529 for the FIFO interface. By default it is set to false 7530 (0). For more information about TPM hardware interfaces 7531 defined by Trusted Computing Group (TCG) see 7532 https://trustedcomputinggroup.org/resource/pc-client-platform-tpm-profile-ptp-specification/ 7533 7534 tp_printk [FTRACE] 7535 Have the tracepoints sent to printk as well as the 7536 tracing ring buffer. This is useful for early boot up 7537 where the system hangs or reboots and does not give the 7538 option for reading the tracing buffer or performing a 7539 ftrace_dump_on_oops. 7540 7541 To turn off having tracepoints sent to printk, 7542 echo 0 > /proc/sys/kernel/tracepoint_printk 7543 Note, echoing 1 into this file without the 7544 tp_printk kernel cmdline option has no effect. 7545 7546 The tp_printk_stop_on_boot (see below) can also be used 7547 to stop the printing of events to console at 7548 late_initcall_sync. 7549 7550 ** CAUTION ** 7551 7552 Having tracepoints sent to printk() and activating high 7553 frequency tracepoints such as irq or sched, can cause 7554 the system to live lock. 7555 7556 tp_printk_stop_on_boot [FTRACE] 7557 When tp_printk (above) is set, it can cause a lot of noise 7558 on the console. It may be useful to only include the 7559 printing of events during boot up, as user space may 7560 make the system inoperable. 7561 7562 This command line option will stop the printing of events 7563 to console at the late_initcall_sync() time frame. 7564 7565 trace_buf_size=nn[KMG] 7566 [FTRACE] will set tracing buffer size on each cpu. 7567 7568 trace_clock= [FTRACE] Set the clock used for tracing events 7569 at boot up. 7570 local - Use the per CPU time stamp counter 7571 (converted into nanoseconds). Fast, but 7572 depending on the architecture, may not be 7573 in sync between CPUs. 7574 global - Event time stamps are synchronized across 7575 CPUs. May be slower than the local clock, 7576 but better for some race conditions. 7577 counter - Simple counting of events (1, 2, ..) 7578 note, some counts may be skipped due to the 7579 infrastructure grabbing the clock more than 7580 once per event. 7581 uptime - Use jiffies as the time stamp. 7582 perf - Use the same clock that perf uses. 7583 mono - Use ktime_get_mono_fast_ns() for time stamps. 7584 mono_raw - Use ktime_get_raw_fast_ns() for time 7585 stamps. 7586 boot - Use ktime_get_boot_fast_ns() for time stamps. 7587 Architectures may add more clocks. See 7588 Documentation/trace/ftrace.rst for more details. 7589 7590 trace_event=[event-list] 7591 [FTRACE] Set and start specified trace events in order 7592 to facilitate early boot debugging. The event-list is a 7593 comma-separated list of trace events to enable. See 7594 also Documentation/trace/events.rst 7595 7596 To enable modules, use :mod: keyword: 7597 7598 trace_event=:mod:<module> 7599 7600 The value before :mod: will only enable specific events 7601 that are part of the module. See the above mentioned 7602 document for more information. 7603 7604 trace_instance=[instance-info] 7605 [FTRACE] Create a ring buffer instance early in boot up. 7606 This will be listed in: 7607 7608 /sys/kernel/tracing/instances 7609 7610 Events can be enabled at the time the instance is created 7611 via: 7612 7613 trace_instance=<name>,<system1>:<event1>,<system2>:<event2> 7614 7615 Note, the "<system*>:" portion is optional if the event is 7616 unique. 7617 7618 trace_instance=foo,sched:sched_switch,irq_handler_entry,initcall 7619 7620 will enable the "sched_switch" event (note, the "sched:" is optional, and 7621 the same thing would happen if it was left off). The irq_handler_entry 7622 event, and all events under the "initcall" system. 7623 7624 Flags can be added to the instance to modify its behavior when it is 7625 created. The flags are separated by '^'. 7626 7627 The available flags are: 7628 7629 traceoff - Have the tracing instance tracing disabled after it is created. 7630 traceprintk - Have trace_printk() write into this trace instance 7631 (note, "printk" and "trace_printk" can also be used) 7632 7633 trace_instance=foo^traceoff^traceprintk,sched,irq 7634 7635 The flags must come before the defined events. 7636 7637 If memory has been reserved (see memmap for x86), the instance 7638 can use that memory: 7639 7640 memmap=12M$0x284500000 trace_instance=boot_map@0x284500000:12M 7641 7642 The above will create a "boot_map" instance that uses the physical 7643 memory at 0x284500000 that is 12Megs. The per CPU buffers of that 7644 instance will be split up accordingly. 7645 7646 Alternatively, the memory can be reserved by the reserve_mem option: 7647 7648 reserve_mem=12M:4096:trace trace_instance=boot_map@trace 7649 7650 This will reserve 12 megabytes at boot up with a 4096 byte alignment 7651 and place the ring buffer in this memory. Note that due to KASLR, the 7652 memory may not be the same location each time, which will not preserve 7653 the buffer content. 7654 7655 Also note that the layout of the ring buffer data may change between 7656 kernel versions where the validator will fail and reset the ring buffer 7657 if the layout is not the same as the previous kernel. 7658 7659 If the ring buffer is used for persistent bootups and has events enabled, 7660 it is recommend to disable tracing so that events from a previous boot do not 7661 mix with events of the current boot (unless you are debugging a random crash 7662 at boot up). 7663 7664 reserve_mem=12M:4096:trace trace_instance=boot_map^traceoff^traceprintk@trace,sched,irq 7665 7666 Note, saving the trace buffer across reboots does require that the system 7667 is set up to not wipe memory. For instance, CONFIG_RESET_ATTACK_MITIGATION 7668 can force a memory reset on boot which will clear any trace that was stored. 7669 This is just one of many ways that can clear memory. Make sure your system 7670 keeps the content of memory across reboots before relying on this option. 7671 7672 NB: Both the mapped address and size must be page aligned for the architecture. 7673 7674 See also Documentation/trace/debugging.rst 7675 7676 7677 trace_options=[option-list] 7678 [FTRACE] Enable or disable tracer options at boot. 7679 The option-list is a comma delimited list of options 7680 that can be enabled or disabled just as if you were 7681 to echo the option name into 7682 7683 /sys/kernel/tracing/trace_options 7684 7685 For example, to enable stacktrace option (to dump the 7686 stack trace of each event), add to the command line: 7687 7688 trace_options=stacktrace 7689 7690 See also Documentation/trace/ftrace.rst "trace options" 7691 section. 7692 7693 trace_trigger=[trigger-list] 7694 [FTRACE] Add an event trigger on specific events. 7695 Set a trigger on top of a specific event, with an optional 7696 filter. 7697 7698 The format is "trace_trigger=<event>.<trigger>[ if <filter>],..." 7699 Where more than one trigger may be specified that are comma delimited. 7700 7701 For example: 7702 7703 trace_trigger="sched_switch.stacktrace if prev_state == 2" 7704 7705 The above will enable the "stacktrace" trigger on the "sched_switch" 7706 event but only trigger it if the "prev_state" of the "sched_switch" 7707 event is "2" (TASK_UNINTERRUPTIBLE). 7708 7709 See also "Event triggers" in Documentation/trace/events.rst 7710 7711 7712 traceoff_after_boot 7713 [FTRACE] Sometimes tracing is used to debug issues 7714 during the boot process. Since the trace buffer has a 7715 limited amount of storage, it may be prudent to 7716 disable tracing after the boot is finished, otherwise 7717 the critical information may be overwritten. With this 7718 option, the main tracing buffer will be turned off at 7719 the end of the boot process. 7720 7721 traceoff_on_warning 7722 [FTRACE] enable this option to disable tracing when a 7723 warning is hit. This turns off "tracing_on". Tracing can 7724 be enabled again by echoing '1' into the "tracing_on" 7725 file located in /sys/kernel/tracing/ 7726 7727 This option is useful, as it disables the trace before 7728 the WARNING dump is called, which prevents the trace to 7729 be filled with content caused by the warning output. 7730 7731 This option can also be set at run time via the sysctl 7732 option: kernel/traceoff_on_warning 7733 7734 transparent_hugepage= 7735 [KNL] 7736 Format: [always|madvise|never] 7737 Can be used to control the default behavior of the system 7738 with respect to transparent hugepages. 7739 See Documentation/admin-guide/mm/transhuge.rst 7740 for more details. 7741 7742 transparent_hugepage_shmem= [KNL] 7743 Format: [always|within_size|advise|never|deny|force] 7744 Can be used to control the hugepage allocation policy for 7745 the internal shmem mount. 7746 See Documentation/admin-guide/mm/transhuge.rst 7747 for more details. 7748 7749 transparent_hugepage_tmpfs= [KNL] 7750 Format: [always|within_size|advise|never] 7751 Can be used to control the default hugepage allocation policy 7752 for the tmpfs mount. 7753 See Documentation/admin-guide/mm/transhuge.rst 7754 for more details. 7755 7756 trusted.source= [KEYS] 7757 Format: <string> 7758 This parameter identifies the trust source as a backend 7759 for trusted keys implementation. Supported trust 7760 sources: 7761 - "tpm" 7762 - "tee" 7763 - "caam" 7764 - "dcp" 7765 If not specified then it defaults to iterating through 7766 the trust source list starting with TPM and assigns the 7767 first trust source as a backend which is initialized 7768 successfully during iteration. 7769 7770 trusted.rng= [KEYS] 7771 Format: <string> 7772 The RNG used to generate key material for trusted keys. 7773 Can be one of: 7774 - "kernel" 7775 - the same value as trusted.source: "tpm" or "tee" 7776 - "default" 7777 If not specified, "default" is used. In this case, 7778 the RNG's choice is left to each individual trust source. 7779 7780 trusted.dcp_use_otp_key 7781 This is intended to be used in combination with 7782 trusted.source=dcp and will select the DCP OTP key 7783 instead of the DCP UNIQUE key blob encryption. 7784 7785 trusted.dcp_skip_zk_test 7786 This is intended to be used in combination with 7787 trusted.source=dcp and will disable the check if the 7788 blob key is all zeros. This is helpful for situations where 7789 having this key zero'ed is acceptable. E.g. in testing 7790 scenarios. 7791 7792 tsa= [X86] Control mitigation for Transient Scheduler 7793 Attacks on AMD CPUs. Search the following in your 7794 favourite search engine for more details: 7795 7796 "Technical guidance for mitigating transient scheduler 7797 attacks". 7798 7799 off - disable the mitigation 7800 on - enable the mitigation (default) 7801 user - mitigate only user/kernel transitions 7802 vm - mitigate only guest/host transitions 7803 7804 7805 tsc= Disable clocksource stability checks for TSC. 7806 Format: <string> 7807 [x86] reliable: mark tsc clocksource as reliable, this 7808 disables clocksource verification at runtime, as well 7809 as the stability checks done at bootup. Used to enable 7810 high-resolution timer mode on older hardware, and in 7811 virtualized environment. 7812 [x86] noirqtime: Do not use TSC to do irq accounting. 7813 Used to run time disable IRQ_TIME_ACCOUNTING on any 7814 platforms where RDTSC is slow and this accounting 7815 can add overhead. 7816 [x86] unstable: mark the TSC clocksource as unstable, this 7817 marks the TSC unconditionally unstable at bootup and 7818 avoids any further wobbles once the TSC watchdog notices. 7819 [x86] nowatchdog: disable clocksource watchdog. Used 7820 in situations with strict latency requirements (where 7821 interruptions from clocksource watchdog are not 7822 acceptable). 7823 [x86] recalibrate: force recalibration against a HW timer 7824 (HPET or PM timer) on systems whose TSC frequency was 7825 obtained from HW or FW using either an MSR or CPUID(0x15). 7826 Warn if the difference is more than 500 ppm. 7827 [x86] watchdog: Use TSC as the watchdog clocksource with 7828 which to check other HW timers (HPET or PM timer), but 7829 only on systems where TSC has been deemed trustworthy. 7830 This will be suppressed by an earlier tsc=nowatchdog and 7831 can be overridden by a later tsc=nowatchdog. A console 7832 message will flag any such suppression or overriding. 7833 7834 tsc_early_khz= [X86,EARLY] Skip early TSC calibration and use the given 7835 value instead. Useful when the early TSC frequency discovery 7836 procedure is not reliable, such as on overclocked systems 7837 with CPUID.16h support and partial CPUID.15h support. 7838 Format: <unsigned int> 7839 7840 tsx= [X86] Control Transactional Synchronization 7841 Extensions (TSX) feature in Intel processors that 7842 support TSX control. 7843 7844 This parameter controls the TSX feature. The options are: 7845 7846 on - Enable TSX on the system. Although there are 7847 mitigations for all known security vulnerabilities, 7848 TSX has been known to be an accelerator for 7849 several previous speculation-related CVEs, and 7850 so there may be unknown security risks associated 7851 with leaving it enabled. 7852 7853 off - Disable TSX on the system. (Note that this 7854 option takes effect only on newer CPUs which are 7855 not vulnerable to MDS, i.e., have 7856 MSR_IA32_ARCH_CAPABILITIES.MDS_NO=1 and which get 7857 the new IA32_TSX_CTRL MSR through a microcode 7858 update. This new MSR allows for the reliable 7859 deactivation of the TSX functionality.) 7860 7861 auto - Disable TSX if X86_BUG_TAA is present, 7862 otherwise enable TSX on the system. 7863 7864 Not specifying this option is equivalent to tsx=off. 7865 7866 See Documentation/admin-guide/hw-vuln/tsx_async_abort.rst 7867 for more details. 7868 7869 tsx_async_abort= [X86,INTEL,EARLY] Control mitigation for the TSX Async 7870 Abort (TAA) vulnerability. 7871 7872 Similar to Micro-architectural Data Sampling (MDS) 7873 certain CPUs that support Transactional 7874 Synchronization Extensions (TSX) are vulnerable to an 7875 exploit against CPU internal buffers which can forward 7876 information to a disclosure gadget under certain 7877 conditions. 7878 7879 In vulnerable processors, the speculatively forwarded 7880 data can be used in a cache side channel attack, to 7881 access data to which the attacker does not have direct 7882 access. 7883 7884 This parameter controls the TAA mitigation. The 7885 options are: 7886 7887 full - Enable TAA mitigation on vulnerable CPUs 7888 if TSX is enabled. 7889 7890 full,nosmt - Enable TAA mitigation and disable SMT on 7891 vulnerable CPUs. If TSX is disabled, SMT 7892 is not disabled because CPU is not 7893 vulnerable to cross-thread TAA attacks. 7894 off - Unconditionally disable TAA mitigation 7895 7896 On MDS-affected machines, tsx_async_abort=off can be 7897 prevented by an active MDS mitigation as both vulnerabilities 7898 are mitigated with the same mechanism so in order to disable 7899 this mitigation, you need to specify mds=off too. 7900 7901 Not specifying this option is equivalent to 7902 tsx_async_abort=full. On CPUs which are MDS affected 7903 and deploy MDS mitigation, TAA mitigation is not 7904 required and doesn't provide any additional 7905 mitigation. 7906 7907 For details see: 7908 Documentation/admin-guide/hw-vuln/tsx_async_abort.rst 7909 7910 turbografx.map[2|3]= [HW,JOY] 7911 TurboGraFX parallel port interface 7912 Format: 7913 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7> 7914 See also Documentation/input/devices/joystick-parport.rst 7915 7916 udbg-immortal [PPC] When debugging early kernel crashes that 7917 happen after console_init() and before a proper 7918 console driver takes over, this boot options might 7919 help "seeing" what's going on. 7920 7921 uhash_entries= [KNL,NET] 7922 Set number of hash buckets for UDP/UDP-Lite connections 7923 7924 uhci-hcd.ignore_oc= 7925 [USB] Ignore overcurrent events (default N). 7926 Some badly-designed motherboards generate lots of 7927 bogus events, for ports that aren't wired to 7928 anything. Set this parameter to avoid log spamming. 7929 Note that genuine overcurrent events won't be 7930 reported either. 7931 7932 unaligned_scalar_speed= 7933 [RISCV] 7934 Format: {slow | fast | unsupported} 7935 Allow skipping scalar unaligned access speed tests. This 7936 is useful for testing alternative code paths and to skip 7937 the tests in environments where they run too slowly. All 7938 CPUs must have the same scalar unaligned access speed. 7939 7940 unaligned_vector_speed= 7941 [RISCV] 7942 Format: {slow | fast | unsupported} 7943 Allow skipping vector unaligned access speed tests. This 7944 is useful for testing alternative code paths and to skip 7945 the tests in environments where they run too slowly. All 7946 CPUs must have the same vector unaligned access speed. 7947 7948 unknown_nmi_panic 7949 [X86] Cause panic on unknown NMI. 7950 7951 unwind_debug [X86-64,EARLY] 7952 Enable unwinder debug output. This can be 7953 useful for debugging certain unwinder error 7954 conditions, including corrupt stacks and 7955 bad/missing unwinder metadata. 7956 7957 usbcore.authorized_default= 7958 [USB] Default USB device authorization: 7959 (default -1 = authorized (same as 1), 7960 0 = not authorized, 1 = authorized, 2 = authorized 7961 if device connected to internal port) 7962 7963 usbcore.autosuspend= 7964 [USB] The autosuspend time delay (in seconds) used 7965 for newly-detected USB devices (default 2). This 7966 is the time required before an idle device will be 7967 autosuspended. Devices for which the delay is set 7968 to a negative value won't be autosuspended at all. 7969 7970 usbcore.usbfs_snoop= 7971 [USB] Set to log all usbfs traffic (default 0 = off). 7972 7973 usbcore.usbfs_snoop_max= 7974 [USB] Maximum number of bytes to snoop in each URB 7975 (default = 65536). 7976 7977 usbcore.blinkenlights= 7978 [USB] Set to cycle leds on hubs (default 0 = off). 7979 7980 usbcore.old_scheme_first= 7981 [USB] Start with the old device initialization 7982 scheme (default 0 = off). 7983 7984 usbcore.usbfs_memory_mb= 7985 [USB] Memory limit (in MB) for buffers allocated by 7986 usbfs (default = 16, 0 = max = 2047). 7987 7988 usbcore.use_both_schemes= 7989 [USB] Try the other device initialization scheme 7990 if the first one fails (default 1 = enabled). 7991 7992 usbcore.initial_descriptor_timeout= 7993 [USB] Specifies timeout for the initial 64-byte 7994 USB_REQ_GET_DESCRIPTOR request in milliseconds 7995 (default 5000 = 5.0 seconds). 7996 7997 usbcore.nousb [USB] Disable the USB subsystem 7998 7999 usbcore.quirks= 8000 [USB] A list of quirk entries to augment the built-in 8001 usb core quirk list. List entries are separated by 8002 commas. Each entry has the form 8003 VendorID:ProductID:Flags. The IDs are 4-digit hex 8004 numbers and Flags is a set of letters. Each letter 8005 will change the built-in quirk; setting it if it is 8006 clear and clearing it if it is set. The letters have 8007 the following meanings: 8008 a = USB_QUIRK_STRING_FETCH_255 (string 8009 descriptors must not be fetched using 8010 a 255-byte read); 8011 b = USB_QUIRK_RESET_RESUME (device can't resume 8012 correctly so reset it instead); 8013 c = USB_QUIRK_NO_SET_INTF (device can't handle 8014 Set-Interface requests); 8015 d = USB_QUIRK_CONFIG_INTF_STRINGS (device can't 8016 handle its Configuration or Interface 8017 strings); 8018 e = USB_QUIRK_RESET (device can't be reset 8019 (e.g morph devices), don't use reset); 8020 f = USB_QUIRK_HONOR_BNUMINTERFACES (device has 8021 more interface descriptions than the 8022 bNumInterfaces count, and can't handle 8023 talking to these interfaces); 8024 g = USB_QUIRK_DELAY_INIT (device needs a pause 8025 during initialization, after we read 8026 the device descriptor); 8027 h = USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL (For 8028 high speed and super speed interrupt 8029 endpoints, the USB 2.0 and USB 3.0 spec 8030 require the interval in microframes (1 8031 microframe = 125 microseconds) to be 8032 calculated as interval = 2 ^ 8033 (bInterval-1). 8034 Devices with this quirk report their 8035 bInterval as the result of this 8036 calculation instead of the exponent 8037 variable used in the calculation); 8038 i = USB_QUIRK_DEVICE_QUALIFIER (device can't 8039 handle device_qualifier descriptor 8040 requests); 8041 j = USB_QUIRK_IGNORE_REMOTE_WAKEUP (device 8042 generates spurious wakeup, ignore 8043 remote wakeup capability); 8044 k = USB_QUIRK_NO_LPM (device can't handle Link 8045 Power Management); 8046 l = USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL 8047 (Device reports its bInterval as linear 8048 frames instead of the USB 2.0 8049 calculation); 8050 m = USB_QUIRK_DISCONNECT_SUSPEND (Device needs 8051 to be disconnected before suspend to 8052 prevent spurious wakeup); 8053 n = USB_QUIRK_DELAY_CTRL_MSG (Device needs a 8054 pause after every control message); 8055 o = USB_QUIRK_HUB_SLOW_RESET (Hub needs extra 8056 delay after resetting its port); 8057 p = USB_QUIRK_SHORT_SET_ADDRESS_REQ_TIMEOUT 8058 (Reduce timeout of the SET_ADDRESS 8059 request from 5000 ms to 500 ms); 8060 Example: quirks=0781:5580:bk,0a5c:5834:gij 8061 8062 usbhid.mousepoll= 8063 [USBHID] The interval which mice are to be polled at. 8064 8065 usbhid.jspoll= 8066 [USBHID] The interval which joysticks are to be polled at. 8067 8068 usbhid.kbpoll= 8069 [USBHID] The interval which keyboards are to be polled at. 8070 8071 usb-storage.delay_use= 8072 [UMS] The delay in seconds before a new device is 8073 scanned for Logical Units (default 1). 8074 Optionally the delay in milliseconds if the value has 8075 suffix with "ms". 8076 Example: delay_use=2567ms 8077 8078 usb-storage.quirks= 8079 [UMS] A list of quirks entries to supplement or 8080 override the built-in unusual_devs list. List 8081 entries are separated by commas. Each entry has 8082 the form VID:PID:Flags where VID and PID are Vendor 8083 and Product ID values (4-digit hex numbers) and 8084 Flags is a set of characters, each corresponding 8085 to a common usb-storage quirk flag as follows: 8086 a = SANE_SENSE (collect more than 18 bytes 8087 of sense data, not on uas); 8088 b = BAD_SENSE (don't collect more than 18 8089 bytes of sense data, not on uas); 8090 c = FIX_CAPACITY (decrease the reported 8091 device capacity by one sector); 8092 d = NO_READ_DISC_INFO (don't use 8093 READ_DISC_INFO command, not on uas); 8094 e = NO_READ_CAPACITY_16 (don't use 8095 READ_CAPACITY_16 command); 8096 f = NO_REPORT_OPCODES (don't use report opcodes 8097 command, uas only); 8098 g = MAX_SECTORS_240 (don't transfer more than 8099 240 sectors at a time, uas only); 8100 h = CAPACITY_HEURISTICS (decrease the 8101 reported device capacity by one 8102 sector if the number is odd); 8103 i = IGNORE_DEVICE (don't bind to this 8104 device); 8105 j = NO_REPORT_LUNS (don't use report luns 8106 command, uas only); 8107 k = NO_SAME (do not use WRITE_SAME, uas only) 8108 l = NOT_LOCKABLE (don't try to lock and 8109 unlock ejectable media, not on uas); 8110 m = MAX_SECTORS_64 (don't transfer more 8111 than 64 sectors = 32 KB at a time, 8112 not on uas); 8113 n = INITIAL_READ10 (force a retry of the 8114 initial READ(10) command, not on uas); 8115 o = CAPACITY_OK (accept the capacity 8116 reported by the device, not on uas); 8117 p = WRITE_CACHE (the device cache is ON 8118 by default, not on uas); 8119 r = IGNORE_RESIDUE (the device reports 8120 bogus residue values, not on uas); 8121 s = SINGLE_LUN (the device has only one 8122 Logical Unit); 8123 t = NO_ATA_1X (don't allow ATA(12) and ATA(16) 8124 commands, uas only); 8125 u = IGNORE_UAS (don't bind to the uas driver); 8126 w = NO_WP_DETECT (don't test whether the 8127 medium is write-protected). 8128 y = ALWAYS_SYNC (issue a SYNCHRONIZE_CACHE 8129 even if the device claims no cache, 8130 not on uas) 8131 Example: quirks=0419:aaf5:rl,0421:0433:rc 8132 8133 user_debug= [KNL,ARM] 8134 Format: <int> 8135 See arch/arm/Kconfig.debug help text. 8136 1 - undefined instruction events 8137 2 - system calls 8138 4 - invalid data aborts 8139 8 - SIGSEGV faults 8140 16 - SIGBUS faults 8141 Example: user_debug=31 8142 8143 vdso= [X86,SH,SPARC] 8144 On X86_32, this is an alias for vdso32=. Otherwise: 8145 8146 vdso=1: enable VDSO (the default) 8147 vdso=0: disable VDSO mapping 8148 8149 vdso32= [X86] Control the 32-bit vDSO 8150 vdso32=1: enable 32-bit VDSO 8151 vdso32=0 or vdso32=2: disable 32-bit VDSO 8152 8153 See the help text for CONFIG_COMPAT_VDSO for more 8154 details. If CONFIG_COMPAT_VDSO is set, the default is 8155 vdso32=0; otherwise, the default is vdso32=1. 8156 8157 For compatibility with older kernels, vdso32=2 is an 8158 alias for vdso32=0. 8159 8160 Try vdso32=0 if you encounter an error that says: 8161 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed! 8162 8163 video= [FB,EARLY] Frame buffer configuration 8164 See Documentation/fb/modedb.rst. 8165 8166 video.brightness_switch_enabled= [ACPI] 8167 Format: [0|1] 8168 If set to 1, on receiving an ACPI notify event 8169 generated by hotkey, video driver will adjust brightness 8170 level and then send out the event to user space through 8171 the allocated input device. If set to 0, video driver 8172 will only send out the event without touching backlight 8173 brightness level. 8174 default: 1 8175 8176 virtio_mmio.device= 8177 [VMMIO] Memory mapped virtio (platform) device. 8178 8179 <size>@<baseaddr>:<irq>[:<id>] 8180 where: 8181 <size> := size (can use standard suffixes 8182 like K, M and G) 8183 <baseaddr> := physical base address 8184 <irq> := interrupt number (as passed to 8185 request_irq()) 8186 <id> := (optional) platform device id 8187 example: 8188 virtio_mmio.device=1K@0x100b0000:48:7 8189 8190 Can be used multiple times for multiple devices. 8191 8192 vga= [BOOT,X86-32] Select a particular video mode 8193 See Documentation/arch/x86/boot.rst and 8194 Documentation/admin-guide/svga.rst. 8195 Use vga=ask for menu. 8196 This is actually a boot loader parameter; the value is 8197 passed to the kernel using a special protocol. 8198 8199 vm_debug[=options] [KNL] Available with CONFIG_DEBUG_VM=y. 8200 May slow down system boot speed, especially when 8201 enabled on systems with a large amount of memory. 8202 All options are enabled by default, and this 8203 interface is meant to allow for selectively 8204 enabling or disabling specific virtual memory 8205 debugging features. 8206 8207 Available options are: 8208 P Enable page structure init time poisoning 8209 - Disable all of the above options 8210 8211 vmalloc=nn[KMG] [KNL,BOOT,EARLY] Forces the vmalloc area to have an 8212 exact size of <nn>. This can be used to increase 8213 the minimum size (128MB on x86, arm32 platforms). 8214 It can also be used to decrease the size and leave more room 8215 for directly mapped kernel RAM. Note that this parameter does 8216 not exist on many other platforms (including arm64, alpha, 8217 loongarch, arc, csky, hexagon, microblaze, mips, nios2, openrisc, 8218 parisc, m64k, powerpc, riscv, sh, um, xtensa, s390, sparc). 8219 8220 vmcp_cma=nn[MG] [KNL,S390,EARLY] 8221 Sets the memory size reserved for contiguous memory 8222 allocations for the vmcp device driver. 8223 8224 vmhalt= [KNL,S390] Perform z/VM CP command after system halt. 8225 Format: <command> 8226 8227 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic. 8228 Format: <command> 8229 8230 vmpoff= [KNL,S390] Perform z/VM CP command after power off. 8231 Format: <command> 8232 8233 vmscape= [X86] Controls mitigation for VMscape attacks. 8234 VMscape attacks can leak information from a userspace 8235 hypervisor to a guest via speculative side-channels. 8236 8237 off - disable the mitigation 8238 ibpb - use Indirect Branch Prediction Barrier 8239 (IBPB) mitigation (default) 8240 force - force vulnerability detection even on 8241 unaffected processors 8242 8243 vsyscall= [X86-64,EARLY] 8244 Controls the behavior of vsyscalls (i.e. calls to 8245 fixed addresses of 0xffffffffff600x00 from legacy 8246 code). Most statically-linked binaries and older 8247 versions of glibc use these calls. Because these 8248 functions are at fixed addresses, they make nice 8249 targets for exploits that can control RIP. 8250 8251 emulate Vsyscalls turn into traps and are emulated 8252 reasonably safely. The vsyscall page is 8253 readable. 8254 8255 xonly [default] Vsyscalls turn into traps and are 8256 emulated reasonably safely. The vsyscall 8257 page is not readable. 8258 8259 none Vsyscalls don't work at all. This makes 8260 them quite hard to use for exploits but 8261 might break your system. 8262 8263 vt.color= [VT] Default text color. 8264 Format: 0xYX, X = foreground, Y = background. 8265 Default: 0x07 = light gray on black. 8266 8267 vt.cur_default= [VT] Default cursor shape. 8268 Format: 0xCCBBAA, where AA, BB, and CC are the same as 8269 the parameters of the <Esc>[?A;B;Cc escape sequence; 8270 see vga-softcursor.rst. Default: 2 = underline. 8271 8272 vt.default_blu= [VT] 8273 Format: <blue0>,<blue1>,<blue2>,...,<blue15> 8274 Change the default blue palette of the console. 8275 This is a 16-member array composed of values 8276 ranging from 0-255. 8277 8278 vt.default_grn= [VT] 8279 Format: <green0>,<green1>,<green2>,...,<green15> 8280 Change the default green palette of the console. 8281 This is a 16-member array composed of values 8282 ranging from 0-255. 8283 8284 vt.default_red= [VT] 8285 Format: <red0>,<red1>,<red2>,...,<red15> 8286 Change the default red palette of the console. 8287 This is a 16-member array composed of values 8288 ranging from 0-255. 8289 8290 vt.default_utf8= 8291 [VT] 8292 Format=<0|1> 8293 Set system-wide default UTF-8 mode for all tty's. 8294 Default is 1, i.e. UTF-8 mode is enabled for all 8295 newly opened terminals. 8296 8297 vt.global_cursor_default= 8298 [VT] 8299 Format=<-1|0|1> 8300 Set system-wide default for whether a cursor 8301 is shown on new VTs. Default is -1, 8302 i.e. cursors will be created by default unless 8303 overridden by individual drivers. 0 will hide 8304 cursors, 1 will display them. 8305 8306 vt.italic= [VT] Default color for italic text; 0-15. 8307 Default: 2 = green. 8308 8309 vt.underline= [VT] Default color for underlined text; 0-15. 8310 Default: 3 = cyan. 8311 8312 watchdog timers [HW,WDT] For information on watchdog timers, 8313 see Documentation/watchdog/watchdog-parameters.rst 8314 or other driver-specific files in the 8315 Documentation/watchdog/ directory. 8316 8317 watchdog_thresh= 8318 [KNL] 8319 Set the hard lockup detector stall duration 8320 threshold in seconds. The soft lockup detector 8321 threshold is set to twice the value. A value of 0 8322 disables both lockup detectors. Default is 10 8323 seconds. 8324 8325 workqueue.unbound_cpus= 8326 [KNL,SMP] Specify to constrain one or some CPUs 8327 to use in unbound workqueues. 8328 Format: <cpu-list> 8329 By default, all online CPUs are available for 8330 unbound workqueues. 8331 8332 workqueue.watchdog_thresh= 8333 If CONFIG_WQ_WATCHDOG is configured, workqueue can 8334 warn stall conditions and dump internal state to 8335 help debugging. 0 disables workqueue stall 8336 detection; otherwise, it's the stall threshold 8337 duration in seconds. The default value is 30 and 8338 it can be updated at runtime by writing to the 8339 corresponding sysfs file. 8340 8341 workqueue.panic_on_stall=<uint> 8342 Panic when workqueue stall is detected by 8343 CONFIG_WQ_WATCHDOG. It sets the number times of the 8344 stall to trigger panic. 8345 8346 The default is 0, which disables the panic on stall. 8347 8348 workqueue.cpu_intensive_thresh_us= 8349 Per-cpu work items which run for longer than this 8350 threshold are automatically considered CPU intensive 8351 and excluded from concurrency management to prevent 8352 them from noticeably delaying other per-cpu work 8353 items. Default is 10000 (10ms). 8354 8355 If CONFIG_WQ_CPU_INTENSIVE_REPORT is set, the kernel 8356 will report the work functions which violate this 8357 threshold repeatedly. They are likely good 8358 candidates for using WQ_UNBOUND workqueues instead. 8359 8360 workqueue.cpu_intensive_warning_thresh=<uint> 8361 If CONFIG_WQ_CPU_INTENSIVE_REPORT is set, the kernel 8362 will report the work functions which violate the 8363 intensive_threshold_us repeatedly. In order to prevent 8364 spurious warnings, start printing only after a work 8365 function has violated this threshold number of times. 8366 8367 The default is 4 times. 0 disables the warning. 8368 8369 workqueue.power_efficient 8370 Per-cpu workqueues are generally preferred because 8371 they show better performance thanks to cache 8372 locality; unfortunately, per-cpu workqueues tend to 8373 be more power hungry than unbound workqueues. 8374 8375 Enabling this makes the per-cpu workqueues which 8376 were observed to contribute significantly to power 8377 consumption unbound, leading to measurably lower 8378 power usage at the cost of small performance 8379 overhead. 8380 8381 The default value of this parameter is determined by 8382 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT. 8383 8384 workqueue.default_affinity_scope= 8385 Select the default affinity scope to use for unbound 8386 workqueues. Can be one of "cpu", "smt", "cache", 8387 "numa" and "system". Default is "cache". For more 8388 information, see the Affinity Scopes section in 8389 Documentation/core-api/workqueue.rst. 8390 8391 This can be changed after boot by writing to the 8392 matching /sys/module/workqueue/parameters file. All 8393 workqueues with the "default" affinity scope will be 8394 updated accordingly. 8395 8396 workqueue.debug_force_rr_cpu 8397 Workqueue used to implicitly guarantee that work 8398 items queued without explicit CPU specified are put 8399 on the local CPU. This guarantee is no longer true 8400 and while local CPU is still preferred work items 8401 may be put on foreign CPUs. This debug option 8402 forces round-robin CPU selection to flush out 8403 usages which depend on the now broken guarantee. 8404 When enabled, memory and cache locality will be 8405 impacted. 8406 8407 writecombine= [LOONGARCH,EARLY] Control the MAT (Memory Access 8408 Type) of ioremap_wc(). 8409 8410 on - Enable writecombine, use WUC for ioremap_wc() 8411 off - Disable writecombine, use SUC for ioremap_wc() 8412 8413 x2apic_phys [X86-64,APIC,EARLY] Use x2apic physical mode instead of 8414 default x2apic cluster mode on platforms 8415 supporting x2apic. 8416 8417 xen_512gb_limit [KNL,X86-64,XEN] 8418 Restricts the kernel running paravirtualized under Xen 8419 to use only up to 512 GB of RAM. The reason to do so is 8420 crash analysis tools and Xen tools for doing domain 8421 save/restore/migration must be enabled to handle larger 8422 domains. 8423 8424 xen_emul_unplug= [HW,X86,XEN,EARLY] 8425 Unplug Xen emulated devices 8426 Format: [unplug0,][unplug1] 8427 ide-disks -- unplug primary master IDE devices 8428 aux-ide-disks -- unplug non-primary-master IDE devices 8429 nics -- unplug network devices 8430 all -- unplug all emulated devices (NICs and IDE disks) 8431 unnecessary -- unplugging emulated devices is 8432 unnecessary even if the host did not respond to 8433 the unplug protocol 8434 never -- do not unplug even if version check succeeds 8435 8436 xen_legacy_crash [X86,XEN,EARLY] 8437 Crash from Xen panic notifier, without executing late 8438 panic() code such as dumping handler. 8439 8440 xen_mc_debug [X86,XEN,EARLY] 8441 Enable multicall debugging when running as a Xen PV guest. 8442 Enabling this feature will reduce performance a little 8443 bit, so it should only be enabled for obtaining extended 8444 debug data in case of multicall errors. 8445 8446 xen_msr_safe= [X86,XEN,EARLY] 8447 Format: <bool> 8448 Select whether to always use non-faulting (safe) MSR 8449 access functions when running as Xen PV guest. The 8450 default value is controlled by CONFIG_XEN_PV_MSR_SAFE. 8451 8452 xen_nopv [X86] 8453 Disables the PV optimizations forcing the HVM guest to 8454 run as generic HVM guest with no PV drivers. 8455 This option is obsoleted by the "nopv" option, which 8456 has equivalent effect for XEN platform. 8457 8458 xen_no_vector_callback 8459 [KNL,X86,XEN,EARLY] Disable the vector callback for Xen 8460 event channel interrupts. 8461 8462 xen_scrub_pages= [XEN] 8463 Boolean option to control scrubbing pages before giving them back 8464 to Xen, for use by other domains. Can be also changed at runtime 8465 with /sys/devices/system/xen_memory/xen_memory0/scrub_pages. 8466 Default value controlled with CONFIG_XEN_SCRUB_PAGES_DEFAULT. 8467 8468 xen_timer_slop= [X86-64,XEN,EARLY] 8469 Set the timer slop (in nanoseconds) for the virtual Xen 8470 timers (default is 100000). This adjusts the minimum 8471 delta of virtualized Xen timers, where lower values 8472 improve timer resolution at the expense of processing 8473 more timer interrupts. 8474 8475 xen.balloon_boot_timeout= [XEN] 8476 The time (in seconds) to wait before giving up to boot 8477 in case initial ballooning fails to free enough memory. 8478 Applies only when running as HVM or PVH guest and 8479 started with less memory configured than allowed at 8480 max. Default is 180. 8481 8482 xen.event_eoi_delay= [XEN] 8483 How long to delay EOI handling in case of event 8484 storms (jiffies). Default is 10. 8485 8486 xen.event_loop_timeout= [XEN] 8487 After which time (jiffies) the event handling loop 8488 should start to delay EOI handling. Default is 2. 8489 8490 xen.fifo_events= [XEN] 8491 Boolean parameter to disable using fifo event handling 8492 even if available. Normally fifo event handling is 8493 preferred over the 2-level event handling, as it is 8494 fairer and the number of possible event channels is 8495 much higher. Default is on (use fifo events). 8496 8497 xirc2ps_cs= [NET,PCMCIA] 8498 Format: 8499 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]] 8500 8501 xive= [PPC] 8502 By default on POWER9 and above, the kernel will 8503 natively use the XIVE interrupt controller. This option 8504 allows the fallback firmware mode to be used: 8505 8506 off Fallback to firmware control of XIVE interrupt 8507 controller on both pseries and powernv 8508 platforms. Only useful on POWER9 and above. 8509 8510 xive.store-eoi=off [PPC] 8511 By default on POWER10 and above, the kernel will use 8512 stores for EOI handling when the XIVE interrupt mode 8513 is active. This option allows the XIVE driver to use 8514 loads instead, as on POWER9. 8515 8516 xhci-hcd.quirks [USB,KNL] 8517 A hex value specifying bitmask with supplemental xhci 8518 host controller quirks. Meaning of each bit can be 8519 consulted in header drivers/usb/host/xhci.h. 8520 8521 xmon [PPC,EARLY] 8522 Format: { early | on | rw | ro | off } 8523 Controls if xmon debugger is enabled. Default is off. 8524 Passing only "xmon" is equivalent to "xmon=early". 8525 early Call xmon as early as possible on boot; xmon 8526 debugger is called from setup_arch(). 8527 on xmon debugger hooks will be installed so xmon 8528 is only called on a kernel crash. Default mode, 8529 i.e. either "ro" or "rw" mode, is controlled 8530 with CONFIG_XMON_DEFAULT_RO_MODE. 8531 rw xmon debugger hooks will be installed so xmon 8532 is called only on a kernel crash, mode is write, 8533 meaning SPR registers, memory and, other data 8534 can be written using xmon commands. 8535 ro same as "rw" option above but SPR registers, 8536 memory, and other data can't be written using 8537 xmon commands. 8538 off xmon is disabled. 8539