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